Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
NUTRIENTS BUILD-UP AND WASH-OFF
PROCESSES IN URBAN LAND USES
Nandika Prasadani Miguntanna
BSc. (Civil Engineering, Honours)
A THESIS SUBMITTED
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF
THE DEGREE OF DOCTOR OF PHILOSOPHY
FACULTY OF BUILT ENVIRONMENT AND ENGINEERING
QUEENSLAND UNIVERSITY OF TECHNOLOGY
October- 2009
i
KEYWORDS
Nutrients build-up process, Nutrients wash-off process, Urban stormwater quality,
Urban water quality, Urban water pollution
ii
iii
ABSTRACT
This thesis describes outcomes of a research study conducted to investigate the
nutrient build-up and wash-off processes on urban impervious surfaces. The data
needed for the study was generated through a series of field investigations and
laboratory test procedures. The study sites were selected in urbanised catchments to
represent typical characteristics of residential, industrial and commercial land uses.
The build-up and wash-off samples were collected from road surfaces in the selected
study sites. A specially designed vacuum collection system and a rainfall simulator
were used for sample collection.
According to the data analysis, the solids build-up on road surfaces was significantly
finer with more than 80% of the particles below 150 µm for all the land uses.
Nutrients were mostly associated with the particle size range below 150 µm in both
build-up and wash-off samples irrespective of type of land use. Therefore, the finer
fraction of solids was the most important for the nutrient build-up and particulate
nutrient wash-off processes. Consequently, the design of stormwater quality
mitigation measures should target particles less than 150 µm for the removal of
nutrients irrespective of type of land use. Total kjeldahl nitrogen (TKN) was the
most dominant form of nitrogen species in build-up on road surfaces. Phosphorus
build-up on road surfaces was mainly in inorganic form and phosphate (PO43-) was
the most dominant form.
The nutrient wash-off process was found to be dependent on rainfall intensity and
duration. Concentration of both total nitrogen and phosphorus was higher at the
beginning of the rain event and decreased with the increase in rainfall duration.
Consequently, in the design of stormwater quality mitigation strategies for nutrients
removal, it is important to target the initial period of rain events. The variability of
wash-off of nitrogen with rainfall intensity was significantly different to phosphorus
wash-off. The concentration of nitrogen was higher in the wash-off for low intensity
rain events compared to the wash-off for high intensity rain events. On the other
iv
hand, the concentration of phosphorus in the wash-off was high for high intensity
rain events compared to low intensity rain events. Consequently, the nitrogen wash-
off can be defined as a source limiting process and phosphorus wash-off as a
transport limiting process. This highlights the importance of taking into
consideration the wash-off of low intensity rain events in the design of stormwater
quality mitigation strategies targeting the nitrogen removal.
All the nitrogen species in wash-off are primarily in dissolved form whereas
phosphorus is in particulate form. The differences in the nitrogen and phosphorus
wash-off processes is principally due to the degree of solubility, attachment to
particulates, composition of total nitrogen and total phosphorus and the degree of
adherence of the solids particles to the surface to which nutrients are attached. The
particulate nitrogen available for wash-off is removed readily as these are mobilised
as free solids particles on the surface. Phosphorus is washed-off mostly with the
solids particles which are strongly adhered to the surface or as the fixed solids load.
Investigation of the nitrogen wash-off process using bulk wash-off samples was in
close agreement with the investigation of dissolved fraction of wash-off solids. This
was primarily due to the predominant nature of dissolved nitrogen. However, the
investigation of the processes which underpin phosphorus wash-off using bulk wash-
off samples could lead to loss of information. This is due to the composition of total
phosphorus in wash-off solids and the inherent variability of the wash-off process for
the different particle size ranges. This variability should preferably be taken into
consideration as phosphorus wash-off is predominantly in particulate form.
Therefore, care needs to be taken in the investigation of the phosphorus wash-off
process using bulk wash-off samples to ensure that there is no loss of information
and hence result in misleading outcomes. The investigation of different particle size
ranges of wash-off solids is preferable in the interest of designing effective
stormwater quality management strategies targeting phosphorus removal.
v
TABLE OF CONTENTS
Chapter 1 Introduction..............................................................................................1
1.1 Background ..................................................................................................1
1.2 Aims and objectives.....................................................................................2
1.3 Research hypotheses ....................................................................................3
1.4 Scope ...........................................................................................................3
1.5 Significance and justification for the research.............................................4
1.6 Methodology for the study...........................................................................5
1.7 Outline of the thesis .....................................................................................6
Chapter 2 Catchment Urbanisation .........................................................................9
2.1 Background ..................................................................................................9
2.2 Impacts of urbanisation..............................................................................10
2.2.1 Hydrologic impacts ................................................................................10
2.2.2 Water quality impacts ............................................................................14
2.3 Primary water pollutants in an urban environment....................................37
2.3.1 Suspended solids....................................................................................37
2.3.2 Organic carbon.......................................................................................40
2.3.3 Nutrients.................................................................................................40
2.3.4 Heavy metals..........................................................................................42
2.3.5 Hydrocarbons.........................................................................................45
2.3.6 Litter.......................................................................................................46
2.3.7 Pathogens ...............................................................................................47
2.4 Nutrient build-up and wash-off in urban catchments.................................47
2.5 Conclusions................................................................................................54
vi
Chapter 3 Field Investigation Apparatus ..............................................................57
3.1 Background................................................................................................57
3.2 Vacuum collection system .........................................................................58
3.2.1 Selection of vacuum cleaner ..................................................................59
3.2.2 Sampling efficiency ...............................................................................61
3.3 Rainfall simulator.......................................................................................63
3.3.1 Calibration of rainfall simulator.............................................................65
3.3.2 Calibration for rainfall intensity and uniformity of rainfall...................66
3.3.3 Drop size distribution and kinetic energy of rainfall .............................68
3.4 Summary....................................................................................................73
Chapter 4 Study Site Selection and Sample Collection ........................................75
4.1 Background................................................................................................75
4.2 Study area...................................................................................................76
4.3 Study site selection ....................................................................................77
4.4 Collection of pollutant build-up samples...................................................83
4.5 Collection of pollutant wash-off ................................................................84
4.6 Sample handling.........................................................................................87
4.7 Summary....................................................................................................87
Chapter 5 Analytical Procedures ...........................................................................89
5.1 Background................................................................................................89
5.2 Sub Sampling.............................................................................................90
5.3 Laboratory Testing.....................................................................................92
5.3.1 Particle size distribution.........................................................................95
5.3.2 pH /EC ...................................................................................................96
5.3.3 Total suspended solids (TSS), Total dissolved solids (TDS) ................97
5.3.4 Total organic carbon (TOC), Dissolved organic carbon (DOC)............98
vii
5.3.5 Nutrients.................................................................................................99
5.4 Data analysis techniques ..........................................................................102
5.4.1 Mean and Standard deviation...............................................................103
5.4.2 Principal Component Analysis ............................................................103
5.4.3 Multi Criteria Decision Making Methods (MCDM)............................106
5.5 Summary ..................................................................................................110
Chapter 6 Analysis of Nutrient Build-up.............................................................113
6.1 Background ..............................................................................................113
6.2 Investigation of primary characteristics of solids build-up......................114
6.2.1 Total solid (TS) ....................................................................................114
6.2.2 Particle size distribution.......................................................................115
6.3 Investigation of total organic carbon (TOC) in solids build-up...............117
6.4 Investigation of nutrients build-up process..............................................118
6.4.1 Nutrient build-up process.....................................................................119
6.5 Conclusions..............................................................................................124
Chapter 7 Understanding Nutrient Wash-off .....................................................127
7.1 Background ..............................................................................................127
7.2 Selected wash-off data and pre-treatment................................................128
7.3 Investigation of total solids (TS) wash-off ..............................................129
7.3.1 Variation of total solids concentration with rainfall intensity and
duration .........................................................................................129
7.3.2 Particle size distribution.......................................................................132
7.4 Investigation of nutrient wash-off process...............................................138
7.4.1 Variability of nutrient wash-off process with rainfall intensity and
duration .........................................................................................138
7.4.2 Nutrients in different particle size ranges of wash-off solids ..............145
7.5 Conclusions..............................................................................................148
viii
Chapter 8 Analysis of Nutrient Wash-off ............................................................151
8.1 Background..............................................................................................151
8.1.1 Selection of wash-off data for analysis................................................152
8.1.2 Data pre-processing .............................................................................154
8.2 Pattern recognition of nutrient wash-off ..................................................155
8.2.1 Analysis of nutrient wash-off in different particle size ranges ............155
8.2.2 Analysis of nutrient wash-off process in all the particle size ranges...169
8.3 Conclusions..............................................................................................174
Chapter 9 Conclusions and Recommendations for further research ...............177
9.1 Conclusions..............................................................................................177
9.1.1 Detailed knowledge of the nutrient build-up process ..........................178
9.1.2 Detailed knowledge of the nutrient wash-off process..........................179
9.1.3 Concluding remarks .............................................................................181
9.2 Recommendations for future research .....................................................182
Chapter 10 References...........................................................................................185
ix
LIST OF FIGURES
Figure 2.1 - Natural water cycle ............................................................................10 Figure 2.2 - Changes in runoff hydrograph after urbanisation ..............................12 Figure 2.3 - Impacts of urbanisation on stormwater runoff...................................13 Figure 2.4 - Hypothetical representations of surface pollutant load over time......31 Figure 2.5 - Hyperbolic function for estimation of the available solids................32 Figure 2.6 - Power function estimation of the available sediment load ................32 Figure 3.1 - Collection efficiencies of dry and wet vacuum sampling on a same
floor for various concentrations of pollutants (Adopted from Bris et
al. 1999) .............................................................................................59
Figure 3.2 - The design of the water filter system of Delonghi Aqualand model .61 Figure 3.3a - Section of sample road surface ……………………………………. 61 Figure 3.3b - Section of road surface ……………………………………………. 61 Figure 3.4 - Comparison of particle size distribution of original sample and
recovered sample................................................................................63 Figure 3.5 - Rainfall Simulator (Adapted from Herngren et al. 2005) ..................65 Figure 3.6 - Calibration of the simulator for intensity...........................................67 Figure 3.7 - Pellets separated into each size ranges...............................................70 Figure 3.8 - Experimental setup for drop size calibration .....................................71 Figure 3.9 - Calibration curve for flour pellets......................................................72 Figure 4.1 - Map of Gold Coast Region ................................................................76 Figure 4.2 - Residential research site (Armstrong Way) .......................................79 Figure 4.3 - Industrial research site (Stevens Street) .............................................80 Figure 4.4 - Commercial research site (Lawrence Drive) .....................................82
x
Figure 4.5 - Collection of build-up sample............................................................84 Figure 4.6 - Wash-off collection test plot..............................................................85 Figure 4.7 - Arrangement of rainfall simulator in the field ...................................86 Figure 4.8 - Collection of wash-off samples into polyethylene containers ...........86 Figure 4.9 - Labelling of the samples collected in the filed ..................................87 Figure 5.1 - Wet sieving of samples ......................................................................91 Figure 5.2 - Analytical schemata of build-up and wash-off samples ....................94 Figure 5.3 - Malvern Mastersizer S .......................................................................96 Figure 5.4 - Shimadzu TOC- VCSH Total Organic Carbon Analyzer..................98 Figure 5.5a- SmartChem 140 ...............................................................................100
Figure 5.5b - Seal Discrete Analyser ……………………………………………100
Figure 5.6- DR 4000 spectrophotometer ............................................................101 Figure 5.7 - Digestion block ................................................................................102 Figure 6.1 - Cumulative particle size distribution of solids build-up at each road
surface..............................................................................................116 Figure 6.2- GAIA analysis for all the road surfaces (∆ = 72.69%) .................121 Figure 6.3a - Amount of nutrients in different particle size ranges of solids for the
residential road surface ....................................................................122
Figure 6.3b - Amount of nutrients in in different particle size ranges of solids for
the industrial road surface …………………………………….......122
Figure 6.3c - Amount of nutrients in in different particle size ranges of solids for
the commercial road surface ………………………………………122
Figure 7.1a - Variation of TS concentration with rainfall intensity and duration for
the residential road surface ..............................................................130
Figure 7.1b - Variation of TS concentration with rainfall intensity and duration for
the industrial road surface ………………………………………...130
xi
Figure 7.1c - Variation of TS concentration with rainfall intensity and duration for
the commercial road surface …………………………………….. 131
Figure 7.2a - Variation of particle size distribution of solids with rainfall intensity
for the residential road surface.........................................................134
Figure 7.2b - Variation of particle size distribution of solids with rainfall intensity
for the industrial road surface ……………………………………. 134
Figure 7.2c - Variation of particle size distribution of solids with rainfall intensity
for the commercial road surface …………………………………. 135
Figure 7.3 - Variation of particle size distribution of wash-off solids with rainfall
duration -20mm/hr rainfall intensity for the industrial road surface136
Figure 7.4a - Variation of TN concentration with rainfall intensity and duration for
the residential road surface ..............................................................139
Figure 7.4b - Variation of TP concentration with rainfall intensity and duration for
the residential road surface………………………………………...139
Figure 7.4c - Variation of TN concentration with rainfall intensity and duration for
the industrial road surface…………………………………………140
Figure 7.4d - Variation of TP concentration with rainfall intensity and duration for
the industrial road surface…………………………………………140
Figure 7.4e - Variation of TN concentration with rainfall intensity and duration for
the commercial road surface ………………………………………141
Figure 7.4f - Variation of TP concentration with rainfall intensity and duration for
the commercial road surface……………………………………. ...141
Figure 7.5 - Variation of average TN concentration with rainfall intensity and
duration for all the road surfaces .....................................................143
xii
Figure 7.6 - Variation of average TP concentration with rainfall intensity and
duration for all the road surfaces .....................................................144
Figure 7.7a - Nutrients concentration in different particle size ranges of wash-off
solids for the residential road surface ..............................................146
Figure 7.7b - Nutrients concentration in different particle size ranges of wash-off
solids for the industrial road surface………………………………146
Figure 7.7c - Nutrients concentration in different particle size ranges of wash-off
solids for the commercial road surface…………………………….147
Figure 8.1a - PC1 vs PC2 biplot obtained from PCA in the dissolved fraction of
wash-off ...........................................................................................156
Figure 8.1b - PC1 vs PC3 biplot obtained from PCA in the dissolved fraction of
wash-off……………………………………………………………157
Figure 8.2 - Variation of TN concentration with TDS ........................................159 Figure 8.3 - Variation of average concentration of TN with DOC......................160 Figure 8.4 - Principal component biplot obtained from PCA on the particle size
range 1-150 µm of wash-off solids ..................................................162
Figure 8.5 - Principal component biplot obtained from PCA on the particle size
range >150 µm of wash-off solids ...................................................166
Figure 8.6 - Principal component biplot obtained from PCA on all the particle size
ranges of wash-off solids .................................................................170
xiii
LIST OF TABLES
Table 2.1 - Potential hydrologic impacts of urbanisation.....................................11
Table 2.2 - Typical road runoff contaminants and their sources..........................17
Table 2.3 - Fraction of pollutants associated with different particle size ranges- 39
Table 2.4 - Sources of heavy metals from traffic related activities......................43
Table 2.5 - Summary of observed particulate quality for P and TKN in different
land uses (means for <125 µm particles) (mg constituent/kg solids) 52
Table 3.1 - Sample recovering efficiencies ..........................................................63
Table 4.1 - Description of possible residential sites.............................................78
Table 4.2 - Description of possible industrial sites ..............................................80
Table 4.3 - Description of possible commercial sites...........................................82
Table 4.4 - Number of antecedent dry days .........................................................83
Table 4.5 - Rainfall intensities and durations simulated during the study ...........85
Table 5.1 - Test parameters and methods used.....................................................92
Table 5.2 - List of preference functions .............................................................109
Table 6.1 - Amount of total solid load at each road surface and respective
antecedent dry period.......................................................................114
Table 6.2 - Amount of solids as a percentage of total solids load......................117
Table 6.3 - TOC in the solids build-up at each road surface..............................118
Table 6.4 - PROMETHEE 2 ranking .................................................................120
Table 7.1 - Solids concentration of wash-off samples .......................................137
Table 8.1 - Correlation matrix for the dissolved fraction...................................158
Table 8.2 - Correlation matrix for the particle size range 1-150 µm..................163
Table 8.3 - Correlation matrix for the particle size range >150 µm...................166
Table 8.4 - Correlation matrix for the PCA analysis of all the particle size ranges
of wash-off solids.............................................................................171
xiv
xv
LIST OF APPENDICES
Appendix 1 Calibration of Rainfall Simulator………………………………….211
Appendix 2 Test results…………………………………………………………227
Appendix 3 Analysis of nutrient build-up ……………………………………...289
Appendix 4 Analysis of nutrient wash-off ……………………………………..293
Appendix 5 Analysis of nutrient wash-off using PCA………………………….305
xvi
xvii
ABBREVIATIONS
Al - Aluminium
ARI - Average recurrence interval
As - Arsenic
BOD - Biochemical oxygen demand
C - Commercial
Cd - Cadmium
COD - Chemical Oxygen demand
Cr - Chromium
Cu - Copper
D50 - Median drop size
DIN - Dissolved inorganic nitrogen
DOC - Dissolved organic carbon
EC - Electrical conductivity
Fe - Iron
Hg - Mercury
HPO42- - Hydrogen phosphate
H2PO4- - Dihydrogen phosphate
I - Industrial
ISC - Impervious surface cover
MCDM - Multi Criteria Decision Making Methods
Nd - Not detected
NH4+ - Ammonium
Ni - Nickel
NO2- -Nitrite-nitrogen
NO3- - Nitrate-nitrogen
PAHs - Polycyclic aromatic hydrocarbons
Pb - Lead
PCs - Principal components
PCA - Principal Component Analysis
PO43- - Phosphate
QA - Quality Assurance
xviii
QC - Quality Control
R - Residential
SRP - Soluble reactive phosphorus
Std - Standard deviation
TC - Total carbon
TDS - Total dissolved solids
TKN - Total kjeldahl nitrogen
TN - Total nitrogen
TOC - Total organic carbon
TP - Total phosphorus
TS - Total solids
TSS - Total suspended solids
TPH - Total petroleum hydrocarbons
Zn - Zinc
xix
STATEMENT OF ORIGINAL AUTHORSHIP
I herby declare that the work contained in this thesis has not been previously
submitted for a degree or diploma or any other higher degree to the best of my
knowledge and belief. The thesis contains no material previously published or
written by another person except where due reference is made.
Nandika Miguntanna Date
xx
xxi
ACKNOWLEDGEMENTS
Completion of this doctoral research would not have been possible without the
guidance and support of numerous people throughout the research project. I am
extremely grateful to my principal supervisor Prof. Ashantha Goonetilleke for his
invaluable support, guidance and constructive criticism provided to me throughout
the completion of this research. My appreciation is further extended to my associate
supervisors Dr. Prasanna Egodawatta and Dr. Serge Kokot for always answering my
many questions and for their dedicated guidance despite their busy schedule.
I am deeply indebted to the Faculty of Built Environment and Engineering,
Queensland University of Technology (QUT), for the financial aid provided in
pursuing this study. A very special thank you is also due to Adjunct Prof. Evan
Thomas for helping me to get permission from Gold Coast City Council to conduct
field data collection in the Gold Coast.
I would also wish to acknowledge the members of the technical staff at QUT for
their generous support given to me during my field work and laboratory testing. In
particular, I wish to thank to Mr. Terry Beach, Mr. Wayne Moore, Mr. Brian Pelin
and Mr. Jim Hazelman, Mr. Scott Allbery and Mr. Colin Phipps for their patience
and endless support in carrying out the field work. My thanks are also extended to
Mr. Bill Kwiecien, Mrs. Wathsala Kumar and Mr. Shane Russell for providing me
necessary laboratory facilities and invaluable technical support in my laboratory
testings. I am also grateful to A/Prof. Malcolm Cox for giving me permission to use
the cold room when I was faced with difficulties in storing water samples.
I would also like to acknowledge Mrs. Diane Kolomeitz and Mr. Peter Nelson for
their time and support given to me for improving my writing skills. My
acknowledgement is further extended to my fellow researchers and friends Ms.
Nadeeka Miguntanna, Ms. Chandima Gunawardena, Mr. Jang Won, Mr. Isri
Manganka, Mr. Parvez Mahbub, Mr. Manjula Dewadasa, Mr. Indika Thilakarathne ,
Mrs. Thanuja Ranawaka, Mr. Chanaka Madushan, Mr. Kanchana Rathnayaka and
Mr. Rakkitha Thilakarathne for their valuable support in my research. Finally, I
xxii
would like to thank all who have bestowed love and encouragement during this
work.
xxiii
DEDICATION
I would like to dedicate this thesis to my beloved parents, Mr. Piyasena Miguntanna
and Mrs. Jayanthi Wijesekara and to my two sisters, Nadeeka and Poshitha and my
best friends Aruni and Chandi. Their love, morale support and motivation throughout
the completion of this doctoral research are gratefully appreciated.
xxiv
1
Chapter 1 Introduction
1.1 Background
Urbanisation around the world is transforming large unpaved areas such as forest
and pastureland to urban land use with high percentage of impervious areas such as
road surfaces, sidewalks and parking lots. This significantly influences the water
environments with increased runoff and the degradation of stormwater quality
(Arnold and Gibbons 1996; Goonetilleke et al. 2005; Ren et al. 2008). The increased
percentage of impervious areas results in an increased volume of surface runoff.
Furthermore, due to the increased anthropogenic activities, a variety of pollutants
such as nutrients, suspended solids, heavy metals and hydrocarbons accumulate on
catchment surfaces. These pollutants are washed-off during storm events
contributing a significant amount of pollutant loads to receiving water bodies
(Navotny et al. 1985; Bannerman et al. 1993). Thus, urban stormwater runoff has
been identified as one of the major causes of the deterioration of receiving water
quality and is currently of great environmental concern.
With the increasing attention on stormwater pollution as a critical problem,
regulatory authorities have responded by placing greater emphasis to mitigate its
adverse impacts by planning and implementing stormwater quality mitigation
strategies in urban catchments. In this context, accurate knowledge on stormwater
runoff pollution is of crucial importance to develop appropriate mitigation strategies.
Consequently, numerous research studies have been undertaken in order to develop
the knowledge which will support the decision making process. However, the
transferability of current knowledge on runoff pollution between different
geographical areas is still limited and it has significantly constrained the
effectiveness of mitigation strategies.
Generation and transport of pollutants is complex and depends on multiple variables
and processes. The significance of these variables and processes is often hard to
assess and can be highly variable in the urban environment. This has resulted in
2
limited transferability of past research outcomes. Therefore, the detailed
understanding of the important variables and processes is imperative to provide
successful mitigation of urban stormwater quality impacts on receiving waters.
Nutrients have been recognised as one of the important stormwater pollutants
(Bannerman et al. 1993; Wit and Bendoricchio et al. 2001; Taylor et al. 2005; Filik
et al. 2008). Nutrients build-up and wash-off processes are complex and can vary
with a range of parameters such as the gradation of solids on the surface to which
nutrients are attached, rainfall characteristics and anthropogenic activities in the
surrounding area. The current knowledge relating to these parameters, which
influence nutrients build-up and wash-off processes is still limited. Therefore, it is
important to develop an in-depth understanding of the processes and variables which
underpin nutrients build-up and wash-off in order to strengthen effective stormwater
quality mitigation strategies.
1.2 Aims and objectives
The primary objective of the research undertaken was to identify governing
parameters of nutrient build-up and wash-off processes in urban environments.
Consequently, the main aims of the research project were:
• Define nutrient speciation in the road surface build-up of different land uses
in an urbanised catchment;
• Define total nitrogen and phosphorus concentration in runoff under different
rainfall intensities and duration in an urbanised catchment;
• Identify nutrient speciation in different particle size ranges of solids in build-
up and wash-off in an urbanised catchment;
• To study the relationships between nutrient species in build-up and wash-off
with key physio-chemical parameters associated with urban stormwater
pollution.
3
1.3 Research hypotheses
• Parameters such as suspended solids and organic carbon exert a strong
influence on nutrient build-up and wash-off processes.
• Underlying physico-chemical processes for nutrient build-up and wash-off
are independent of pollutant loading and therefore, the type of land use.
1.4 Scope
The primary focus of the research project was to investigate the processes and
parameters involved in the distribution of nutrients as a stormwater pollutant in
urban catchments. The research developed an in-depth understanding of nutrient
build-up and wash-off processes. The scope of the research project was as follows:
• The study was confined to the Gold Coast area. However, the knowledge
developed relating to nutrient build-up and wash-off processes are generic
and applicable outside of the study area irrespective of regional and climatic
variations.
• The research was confined to road surfaces in different urban land uses such
as residential, industrial and commercial. The selection of road surfaces was
due to its significance as a major contributor of pollutants to stormwater
runoff.
• Only one site per land use was chosen based on the hypothesis that pollutant
processes are independent of the pollutant load.
• The road surface conditions such as slope and texture depth were not
measured based on the assumption that pollutant processes are independent
of road surface condition but these conditions were only considered to
determine the suitability of the study site to conduct the field investigations.
• The influence of seasonal variability and traffic volume on pollutant build-up
and wash-off processes was not considered.
• The research was confined to wash-off generated using simulated rain events
but not the natural storm events.
4
• The research focused only on physical and chemical water quality parameters
and not biological parameters.
1.5 Significance and justification for the research
To eliminate the adverse impacts of polluted stormwater runoff on receiving water
quality, it is essential to have appropriate stormwater quality mitigation measures
and efficient treatment designs. The effectiveness of these mitigation strategies is
strongly dependent on the understanding of pollutant processes and various
influential parameters. In this context, limited knowledge on pollutant build-up and
wash-off processes severely impedes the effectiveness of the mitigation actions.
Consequently, it is imperative to understand nutrients build-up and wash-off
processes as nutrients have been identified as a primary stormwater pollutant.
In the past, efforts adopted to investigate urban water quality have been mostly based
on water quantity research such as investigation of pollutant loadings. As a result,
past research outcomes exhibit strong reliance on physical factors and can be highly
location specific, which limits the transferability of research outcomes between
geographical areas. Limited attention has been given to identify the processes
underlying the build-up and wash-off of nutrients. These processes are the basis
which determines the quality characteristics of urban stormwater runoff. This
underlines the need for the in-depth understanding of the processes and influential
parameters which are inherent to nutrients build-up and wash-off in urban areas and
hence, to enhance effective stormwater quality mitigation strategies.
The research undertaken generated comprehensive knowledge on nutrient build-up
and wash-off processes on urban surfaces. In turn, the knowledge on nutrient build-
up and wash-off processes was extended to understand the physico-chemical
parameters that influence these processes. This will enhance the current knowledge
about nutrients as a stormwater pollutant and hence contribute to the knowledge base
on effective design of stormwater quality mitigation strategies. Furthermore, the
detailed understanding gained about nutrient build-up and wash-off processes in
5
urban catchments will extend the accuracy and reliability of water quality modelling
approaches.
1.6 Methodology for the study
The research methodology was primarily an integration of a series of field
investigations and laboratory experimental procedures. The research methodology
consisted of four main activities as follows:
1. Literature review
2. Study site selection and selection of simulated rainfall events
3. Collection and testing of pollutant build-up and wash-off samples
4. Data analysis
Literature review
A comprehensive literature review was carried out to gain an in-depth knowledge of
urban stormwater pollution and its impacts. The literature review provided
knowledge primarily in the following areas:
• Adverse impacts of urbanisation in terms of hydrologic and stormwater
quality;
• Primary urban stormwater pollutants and their pathways;
• Current state of knowledge in relation to urban pollutant build-up and wash-
off processes and in particular nitrogen and phosphorus;
Study site selection and selection of rainfall events Three study sites were selected in urbanised catchments representing typical
residential, industrial and commercial land uses. Road surfaces were selected for
field investigations as roads have been recognised as a major contributor of
pollutants to urban stormwater runoff (Fulcher 1994; Ball et al. 1998; Han et al.
2006). One road surface from each land use was selected and a number of test plots
from each road surface were chosen based on the number of rainfall events to be
simulated.
6
Six different rainfall intensities were selected to be simulated at each study site.
These rainfall intensities were identified as typical rainfall intensities in the study
region based on long-term records of regional rainfall data. The durations of the rain
events were selected based on the 1 to 10 year average recurrence intervals.
Collection and testing of build-up and wash-off samples Field activities were conducted to collect pollutant build-up and wash-off samples
from the selected study sites. Initially, pollutant build-up samples were collected
from the selected study sites. Pollutant wash-off samples were obtained using
simulated rainfall events of different durations at the same study sites used for
pollutant build-up investigations. Both build-up and wash-off investigations were
conducted using small homogeneous plot areas which provides better control of
physical factors which can limit the transferability of research outcomes. Based on
the findings of the literature review, the build-up and wash-off samples were tested
for nutrients and range of physico-chemical parameters.
Data analysis
Data analysis primarily focused on investigating the variability of nutrient build-up
and wash-off processes with influential parameters and the identification of linkages
between nutrients and physico-chemical parameters. Firstly, data analysis was
carried out to understand the nutrient build-up process and the underpinning physico-
chemical parameters. Secondly, data analysis was carried out to understand the
primary physical and chemical processes underpinning the nutrient wash-off process.
In this context, both univariate and multivariate chemometric data analysis
techniques were used. The selection of data analysis techniques was primarily based
on the type of analysis to be performed and capability of the analytical technique.
1.7 Outline of the thesis
The thesis has ten chapters. Chapter 1 provides background that the thesis addresses,
aims and objectives and overview for the research. Chapter 2 presents the outcomes
of the extensive literature review undertaken. The selection of field investigation
7
apparatus and their calibration methods are described in Chapter 3. Chapter 4
discusses the details about study site selection and the procedures for the collection
of pollutant build-up and wash-off samples. A detailed discussion on laboratory
testing is presented in Chapter 5. This Chapter also discusses the theory and the
application of the different data analysis techniques used in the study. The outcomes
of the data analysis are presented in Chapter 6, Chapter 7 and Chapter 8. Chapter 6
focuses on the analysis of nutrient build-up process. Chapter 7 describes the primary
physical process in nutrient wash-off and Chapter 8 has focused on identification of
physico-chemical parameters which govern the nutrient wash-off process. The
conclusions made in this study together with suggested recommendations for future
research are presented in Chapter 9. The final Chapter provides a list of references
used throughout the thesis.
8
9
Chapter 2 Catchment Urbanisation
2.1 Background The concentration of human population near cities has increased markedly in recent
years. According to recent population statistics, nearly one-half of the world’s
population live in urban areas and the trend of increasing urban population is
continuing (UN 2004). The improvements in transportation, the expansion of
manufacturing industries and increased employment opportunities have resulted in
increased density of population in cities. The concentration of population and
consequent anthropogenic activities near cities can be defined as urbanisation.
The growing urbanisation all over the world gives rise to a number of detrimental
impacts on the environment such as land, air and water pollution. Among all these
impacts, the adverse effects of urbanisation on the water environment are crucially
important because it is one of the most essential resources for human existence and
settlement. Natural catchment characteristics disappear with the various
anthropogenic activities that primarily include construction of roads and buildings
and transportation activities. The impacts of these activities on the water
environment can be discussed under two main categories. This includes quantity
impacts in terms of floods, erosion and degradation of stream habitats and
deterioration of water quality (ASCE 1975; Brabec et al. 2002; Hatt et al. 2004;
Goonetilleke et al. 2005). These impacts are discussed in detail under Section 2.2.
The impact of urban stormwater runoff on receiving water bodies is one of the key
environmental concerns at the present time. Numerous researchers have found that
urban stormwater is highly polluting and it is a threat to the quality of receiving
water bodies (for example Balmer et al. 1984; Hall and Ellis 1985; Fulcher 1994;
Ellison and Brett 2006). It is important to have a clear understanding of the quality
impacts of urban stormwater runoff in order to achieve best management practices to
safeguard the urban water quality.
10
2.2 Impacts of urbanisation
2.2.1 Hydrologic impacts
According to the hydrologic cycle shown in Figure 2.1, water is balanced in the
atmosphere, surface and ocean and groundwater in the form of moisture. Water is
circulating through these sources of moisture storage by various water transport
processes such as precipitation, surface water inflow and outflow, groundwater
exchange and evapotranspiration. The hydrologic cycle in the urban areas is
dramatically affected by urbanisation (Hall and Ellis 1985; Codner et al. 1988;
Arnold and Gibbons 1996; Nelson and Booth 2002). The potential hydrologic
impacts of urbanisation are summarised in the Table 2.1.
Figure 2.1- Natural water cycle (Adapted from www.crwa.org/alert/wmapermits/CRWAWhitePaper )
In a hydrologic perspective, the major changes due to urbanisation are based on two
primary physical changes on catchment surfaces. Firstly, the presence of impervious
surfaces such as roofs, streets, sidewalks and parking lots results in reducing the
infiltration capacity and the depression storage while increasing more uniform
surface slopes. The urban areas not covered by impervious material are usually re-
landscaped, covered with grass and vegetation. Frequently, these landscape
11
modifications increase the overland flow, which in turn enhances pollutant wash-off.
Secondly, in the urban runoff process there is an increase in the ‘hydraulic
conveyance’ in the flow channels. The primary causes for this are the introduction of
uniform slopes and lined channels in place of the catchment’s natural drainage
network (Hall and Ellis 1985; Booth 1991).
Table 2.1- Potential hydrologic impacts of urbanisation (Adapted from American Society of Civil Engineers 1975)
Urbanising influence Potential hydrologic Response
Removal of trees and
vegetation
Decrease in evapotranspiration and
interception; increase in stream sedimentation
Initial construction of
houses, streets and culverts
Decrease in infiltration and lowered
groundwater table; increased storm flows and
decreased base flows during dry periods
Complete development of
residential, commercial, and
industrial areas
Increased imperviousness reduces time of
runoff concentration thereby increasing peak
discharges and compressing the time
distribution of flow; runoff volume and flood
damage potential is greatly increased
Construction of storm drains
and channel improvements
Local relief from flooding; concentration of
flood waters may aggravate flood problems
downstream
The changes in impervious fraction and hydraulic conveyance efficiency result in
major changes to the runoff hydrograph (Kibler 1982). These changes are illustrated
in Figure 2.2.
12
0
20
40
60
80
100
120
140
160
180
200
0 50 100 150 200 250 300 350Time
Dis
char
ge
After Urbanisation Before urbanisation
Figure 2.2- Changes in runoff hydrograph after urbanisation
(Adapted from Kibler 1982)
� Increase in runoff peak
Most urbanised areas have larger peak flows than undeveloped areas (Hall and Ellis
1985; Booth 1991; Schueler 1994). When the catchment is urbanised, the percentage
of impervious area within the catchment increases. This leads to increases in the
volume and rate of runoff whilst decreasing groundwater recharge and increasing the
hydraulic conveyance in the flow channels. The increase in hydrograph peaks is the
most important impact of urbanisation (Rao and Delleur 1974; Ferguson and
Suckling 1990). Increased peak runoff may cause flooding both within and
downstream of the urban area. Studies by Espey et al. (1969) and Seaburn (1969)
have shown that the peak flow in urbanised catchments may be as much as 2-4 times
higher than its natural state. The magnitude of the increase in runoff peak is
dependent on the fraction of impervious surfaces in the catchment, type of drainage
channel improvements and amount of vegetation present (Rao and Delleur 1974).
� Increase in runoff volume
Urbanisation leads to an increase in runoff volume from an individual storm as well
as annual yield (Waananen et al. 1961; Seaburn 1969; ASCE 1975; Arnold and
Qup
Qnp
Vu
Vn
tup tnp
Qup -Urban peak discharge
Qnp - Natural peak discharge
tup - Urban time to peak discharge tnp - Natural time to peak discharge Vu -Urban runoff volume Vn - Natural runoff volume
13
Gibbons 1996; Zhao et al. 2007). Changes of runoff volume for different stages of
urbanisation are illustrated in the Figure 2.3.
Figure 2.3- Impacts of urbanisation on stormwater runoff
(Adapted from Connecticut Stormwater Quality Manual 2004)
The main cause of the increase in runoff volume is the increase in impervious area.
As the percentage of catchment impervious surface cover (ISC) increases to 10-20%,
runoff can increase two fold; 35-50% ISC can increase runoff three fold and 75-
100% ISC can increase surface runoff more than five fold over natural catchments
(Figure 2.3). Seaburn (1969) has shown that the volume of storm runoff from urban
catchments can increase from 1.1 to 4.6 times greater than the corresponding runoff
in the pre-urban period depending on the ARI of the individual storm. Furthermore,
Cech and Assaf (1976) noted that even 100% runoff is possible for some urban
catchments under certain rainfall conditions. For example, an urbanised catchment
with saturated impervious surfaces by previous storms may produce 100% runoff.
40% evapotranspiration 38% evapotranspiration
35% evapotranspiration 30% evapotranspiration
10%
runoff
20%
runoff
30%
runoff 55%
runoff
25% shallow
infiltration 21% shallow
infiltration
20% shallow
infiltration
10% shallow
infiltration
25%deep
infiltration
21%deep
infiltration
15%deep
infiltration 5%deep
infiltration
Natural Ground 10%-20% Impervious
35%-50% Impervious 75%-100% Impervious
14
� Reduced time to peak
As illustrated in Figure 2.2, urbanisation reduces the time to peak. As noted by
Espey et al. (1969), time to peak flow is reduced considerably depending on the
amount of impervious surfaces and the type of channel improvements in urban
catchments. Furthermore, Espey et al. (1969) observed a 30% to 40% reduction in
time to peak flow in urban catchments. However, this is not a unique characteristic
for a particular catchment. It can vary from storm to storm (Rao and Delleur 1974).
Waananen (1969) observed that the catchment lag may reduce by as much as 70% in
an urban catchment.
2.2.2 Water quality impacts
The quality of stormwater runoff is an important issue same as quantity because the
polluted stormwater runoff significantly impacts on the quality of receiving water
bodies. Stormwater runoff can contribute up to a third of the total annual pollution
load discharged to the receiving water and is primarily responsible for deterioration
of water quality (Shuttleworth 1986). The physical, chemical and microbiological
quality of urban storm runoff is significantly affected by urbanisation (Qureshi and
Dutka 1979; Gnecco et al. 2005; Ren et al. 2008). Major water quality problems in
urban stormwater are produced by salinity, temperature, sedimentation, dissolved
oxygen depletion, toxic substances and biological effects (Zoppou 2001). Sartor and
Boyd (1972) and Cordery (1977) noted that the pollution load in urban stormwater
runoff can be significantly higher than that from secondary treated sewage effluent.
Anthropogenic activities associated with urbanisation create a number of sources of
pollutants for incorporation into surface runoff, which dramatically degrade the
stormwater quality (Barrios 2000; Line et al. 2002; Goonetilleke et al. 2005, 2009).
These pollutants accumulate on both paved and unpaved surfaces. The energy
associated with rainfall and runoff will dislodge particles from these surfaces. Many
pollutants are adsorbed to these particles and are transported along with soluble
pollutants by the stormwater runoff to receiving water bodies (Zoppou 2001; Jartun
et al. 2008).
15
The quality changes in the stormwater are mainly related to the type of land use and
by implication the anthropogenic activities and vary with the duration and intensity
of the rainfall event (Sonzogni et al. 1980; Sekhar and Raj 1995; Line et al. 2002;
Kim et al. 2007; Goonetilleke et al. 2009). Sonzogni et al. (1980) investigating the
pollutant loads contributed by different land uses noted 10 to 100 times greater loads
of suspended solids, nitrogen and phosphorus from urban areas compared to forested
lands. Many complex chemical and biological processes influence the incorporation
of pollutants into stormwater runoff (Mikkelsen et al. 1994).
The concentration of pollutants in urban runoff varies, both during the storm event
and from event to event in a given catchment, from catchment to catchment within a
given urban region to another across the country (US EPA 1993). Cordery (1977)
noted that the concentration of pollutants tends to increase rapidly at the beginning
of the flood flow and then decrease gradually as the flood progresses. Furthermore,
according to Brezonik and Stadelmann (2002) long duration storms generate more
diluted runoff. The impact of urbanisation on water quality can be broadly discussed
under the following headings.
A. Pollutant sources
Urban stormwater runoff has been recognised as the most important non point source
of pollution to receiving water bodies (Cordery 1977; Brinkmann 1985;
Charbonneau and Kondolf 1993; Goonetilleke et al. 2005). In non-point source
pollution, water flows on the surface and dissolves and washes away pollutants and
soil sediments along its path and finally discharges into receiving water bodies
(Stevenson and Wyman 1991). A variety of sources are responsible for the
accumulation of these pollutants on urban catchment surfaces.
Researchers have identified a number of major sources of stormwater quality
degradation in urban areas (for example Sartor and Boyd 1972; Brinkmann 1985;
Pitt et al. 1995; Kim et al. 2007). The sources of pollutants in an urban area include:
• Transportation activities
• Industrial/commercial activities
16
• Construction activities
• Vegetation
• Soil erosion
• Corrosion
• Atmospheric fallout
• Spills
• Transportation activities
Transportation activities can be considered as one of the major contributing sources
of stormwater pollution. This mainly includes pollutants generated from street
surfaces and motor vehicles. Hoffman et al. (1985) investigating stormwater runoff
from Interstate Highway 95 in Rhode Island, USA, recognised street runoff as a
considerable source of pollution to receiving water bodies. They found that over
50% of the total pollutant inputs of suspended solids, polycyclic aromatic
hydrocarbons and heavy metals into the Pawtuext River which is adjacent to the
highway originated from street runoff. A summary of the detailed investigation into
road runoff contaminants and their sources given by Ball et al. (1998) is presented in
Table 2.2.
17
Table 2.2- Typical road runoff contaminants and their sources (Adapted from Ball et al. 1998)
Contaminant Primary source
Particulates Pavement wear, vehicles, maintenance activities
Nitrogen/Phosphorous Roadside fertiliser applications, atmosphere
Lead
Auto exhaust, tyre wear, lubricating oil and
grease, bearing wear
Zinc Tyre wear, motor oil, grease
Iron
Automobile rust, highway structures (eg guard
rails), engine parts
Copper
Metal plating, bearing and brush wear, engine
parts, brake lining wear, fungicides, insecticides,
pesticides
Cadmium Tyre wear, insecticide application
Chromium Metal plating, moving parts, brake lining wear
Nickel
Diesel fuel and petrol exhaust, lubricating oil,
metal plating, brush wear, brake lining wear,
asphalt paving
Manganese Engine parts, automobile exhaust
Sulfate Roadways surfaces, fuels
Petroleum Hydrocarbons
Spills, leakages of motor lubricants, anti-freeze
and hydraulic fluids, asphalt surface leachate
Polychlorinated
biphenyls (PCB)
Background atmospheric deposition, PCB
catalysts in synthetic tyres, spraying of rights-of-
way
Polycyclic aromatic hydrocarbons
Asphalt surface leachate
Streets have a profound impact on stormwater quality as the pollutants generated by
vehicles are mostly confined to the street surface. The availability of pollutant
constituents on road surfaces is relatively high compared to other impervious
surfaces (Sartor and Boyd 1972; Shaheen et al. 1975; Ellis and Revitt 1982; Fulcher
1994; Ball et al. 1998). Furthermore, street surfaces represent a significant
18
proportion of impervious surfaces in a catchment, which accelerates the pollutant
accumulation process. Even though the pollutants present on street surfaces are
mainly traffic related, other factors such as atmospheric deposition and erosion of
street surfaces also make a significant contribution to pollutant generation. As
Bannerman et al. (1993) pointed out, the pollutants present on street surfaces are
mainly generated from:
o Vehicle exhaust emissions
o Vehicle lubrication system losses
o Degradation of vehicle tyres and brake lining
o Degradation of road surfaces
o Load losses from vehicles
o Atmospheric deposition and soil inputs.
The generation of pollutants on street surfaces varies widely depending on a range of
factors. Sartor and Boyd (1972) conducting a comprehensive study of street surface
contaminants found that the quantity of pollutants varied widely according to:
o The time gap since the street was last cleaned either by rainfall or street
sweeping
o Surrounding land use
o Traffic volume and other traffic characteristics
o Street surface characteristics
o Maintenance practices.
Additionally, pavement material can have significant impacts on pollutant loading.
Several research studies have been undertaken to investigate the pollutant loading
from different paved surfaces such as asphalt and concrete (for example, Sartor and
Boyd 1972; Shaheen 1975). Sartor and Boyd (1972) found that the asphalt
pavements contribute 80% more pollutant loading than concrete surfaces.
Furthermore, road characteristics such as the location of traffic lights, road layout,
surface roughness and driver habits also influence the accumulation of pollutants on
road surfaces (Brinkmann 1985; Bohemen and De Laak 2003).
The contribution by motor vehicles to the accumulation of pollutants on street
surfaces is also of vital importance. The role of motor vehicles in generating
19
pollutants on street surfaces can be direct or indirect. The pollutants that are directly
sourced from vehicles are those related to their operation through frictional wear and
combustion by-products. Sources of pollutants that directly relate to motor vehicles
are leakage of fuels, lubricants, coolants, fine particles from worn-off tyres, clutch
and brake linings, exhaust emissions, decomposing coatings dropped from the under-
carriages and vehicle components broken by vibration or impact (Sartor and Boyd
1972; Bohemen and De Laak 2003).
Drapper et al. (2000) investigating road runoff quality in Brisbane, Australia noted
that higher concentration of heavy metals in the runoff from the exit lanes. This
could be attributed to the brake pad and tyre wear caused by rapid deceleration of
vehicles. On the other hand, the indirectly sourced pollutants are those that are
transported to roads through vehicular activity. An example of this is the
transportation of solids from parking lots, construction sites and poorly maintained
roads. The importance of indirect source of pollutants from vehicles can be assessed
from the outcomes of the study by Shaheen (1975) who found that more than 95% of
the solids on a given highway originate from sources other than the vehicles
themselves.
• Industrial /Commercial activities
Industrial/Commercial activities contribute a range of pollutants to an urban area.
Bannerman et al. (1993) noted that streets and parking lots as critical source areas for
the generation of pollutants in industrial and commercial land uses. The pollutants
resulting from various industrial/commercial activities can contain high amounts of
pollutants such as nutrients, heavy metals and chemical toxins (Line et al. 1997; Lee
and Bang 2000). Numerous researchers have noted that there is a relatively high
pollutant concentration in stormwater runoff from industrial areas when compared to
that from other land uses (For example, Weeks 1981; Kelly et al. 1996; Lee and
Bang 2000; Ghafouri and Swain 2005; Lau and Stenstrom 2005).
Wanielista et al. (1977) have stated that industrial areas have higher loadings of
street surface pollutants than others, probably because they are not swept as
frequently as residential and commercial areas. Pucket, (1995) found that 53% of
20
atmospheric deposited nitrogen in northeastern states in US comes from industries
such as coal and oil burning and electric utilities. The industrial sources of pollutants
are exposed storage, loading and unloading, equipment, spills and leaks, industrial
materials and waste products. The concentration and types of pollutants that
accumulate in industrial areas depends on the nature of the industry and the
management practices that have been taken to safeguard the environment. Ellis
(1989b) noted that the specific pollutants and concentrations depend solely on the
particular industrial process employed.
The significant sources of pollutants in commercial areas are mainly motor fluids
from parked cars, large parking lots, auto service stations, gas stations, shopping
centres and restaurants. These sources produce high hydrocarbon loadings and metal
concentrations that are twice those found in an average urban area (US EPA 2005).
These pollutants can be attributed to heavy traffic volume and large areas of
impervious surfaces where vehicular related pollutants accumulate.
• Construction activities
Construction activities are the most hazardous of urban land activities in terms of
generating solids (Sonzogni et al. 1980; Brinkmann 1985). Consequently,
construction activities have a significant impact on stormwater quality. Solids export
rates for construction sites are significantly higher than that for other land uses (US
EPA 1993; Line et al. 2002). According to the US EPA (1993), solids runoff rates
from construction sites are typically 10 to 20 times greater than those of agricultural
lands and 1000 to 2000 times greater than those of forest lands. Furthermore, during
a relatively short period, construction sites can contribute more solids to receiving
water bodies than that can be deposited naturally during several decades. This is
primarily attributed to higher soil erosion from construction sites. The resulting
siltation and the contribution of pollutants can cause physical, chemical and
biological harm to receiving water bodies (US EPA 1993). However, the pollutant
load can vary considerably with the amount of construction, catchment area,
management of the site and maintenance activities.
21
• Vegetation
Waste vegetative matter from tree leaves and other plant materials such as pollen,
bark, twigs and grass is an important source of both organic pollutants and nutrients
in urban stormwater. Phosphorus, nitrogen and organic matter are increased in runoff
passing through fallen leaves and crop residues (Cordery 1977; Dorney 1986). Line
et al. (2002) investigating pollutant exports from various land uses in the Upper
Neuse river basin in the U.S. found a considerable increase in the concentration of
total kjeldahl nitrogen during periods of pollen deposition. The concentration of
pollutants in stormwater runoff due to vegetation varies significantly with the density
and type of vegetation, catchment characteristics and seasonal changes (OECD
1986).
However, Allison et al. (1998) have questioned the importance of leaf litter as a
nutrient source in urban stormwater. Based on the outcomes of a study in an urban
area in Melbourne, Australia they found that the contribution from leaf litter was
about two magnitudes smaller than the total nutrient load measured.
• Soil erosion
Soil Erosion is a major contributor of suspended solids to stormwater runoff and
occurs not only within new construction areas but also in all unpaved areas.
Commonly in most construction sites, protective vegetative cover is removed and
unprotected soil is left exposed to rainfall. Loss of vegetative cover on the ground
surface greatly increases soil erosion and leads to increases in the suspended solids
loading to stormwater runoff. Factors such as soil type, topography, vegetation and
climatic conditions have a significant influence on soil erosion.
Liu et al. (1997) noted that, while erosion from lawns and other open surfaces are
generally low, erosion from construction areas represent the largest source of solids
in urban runoff. Additionally, the erosion of stream banks leads to an increase in
annual solids load in urbanising catchments (Nelson and Booth 2002).
22
• Corrosion
Corrosion of roofs and other metal surfaces due to rainfall and in particular acid rain
and aggressive gases is a common phenomenon in urban areas. The corroded
particles accumulate on the ground and on roof surfaces and are eventually washed-
off with rainfall and runoff (OECD 1986). Metallic roofs are a significant contributor
to stormwater pollution. Bannerman et al. (1993) noted that the concentration of
heavy metals in runoff from galvanised roofs is higher when compared to runoff
from streets. Furthermore, Pitt et al. (1995) found significantly higher concentrations
of zinc and aluminium in roof runoff than in street runoff. This is mainly due to the
acidity of rainfall leading to corrosion of metallic components in roofs.
According to Brinkmann (1985), corrosion rates will depend on the following
factors:
o Availability of corrodible materials
o The frequency and intensity of exposure to an aggressive environment
o The drying - wetting frequency of the exposed surfaces
o The structure of the material
o Maintenance practices
• Atmospheric fallout
Pollutants from atmospheric fallout include dust and dirt and sulphur and nitrous
oxides from industrial emissions. These are combined with precipitation which falls
on the ground surface to form acid rain. Atmospheric fallout compared to other
sources is clearly one of the significant contributors to stormwater pollution (OECD
1986).
Emissions from vehicles will initially contribute to the pollution of the atmosphere,
but will return to the surface due to atmospheric deposition and thereby contribute to
the pollutant loading in urban stormwater. Ball et al. (2000) noted a significant
contribution from atmospheric sources to the mass of pollutants available on the
ground surface for transport by surface runoff. This deposition may occur during a
storm event or as dustfall during dry periods. Furthermore, roof surfaces have been
23
recognised as efficient collectors of particle fallout from the atmosphere (Davis et al.
2001).
• Spills
This includes spillage or release of substances such as oil, pesticides, chemicals or
raw sewage and spills at construction sites. Spills can degrade stormwater quality
physically, chemically and biologically. Sartor and Boyd (1972) noted that vehicular
transport, building construction and industrial activities are considerable sources of
spills. The adverse impacts on water quality due to spills can be reduced through
good maintenance and management practices.
B. Pollutant pathways
Pollutant pathways are the processes available for transporting pollutants both during
the build-up stage prior to a rainfall event and during the runoff event. The two main
processes available for transporting pollutants in an urban environment are
atmospheric deposition and wash-off of pollutants deposited on ground surfaces
(Hall and Ellis 1985; Bohemen and De Laak 2003; Pitt et al. 2004; Huston et al.
2009)
Atmospheric deposition
The atmosphere itself is considered as a major pathway for transporting dust for
accumulation on the ground surface. According to Brinkmann (1985), the
composition and concentration pattern of atmospheric deposition vary widely in both
time and space. The two forms of atmospheric deposition are wet and dry deposition.
Wet deposition is brought about by rain, fog or mist. In wet deposition, pollutants are
washed out during storm events. Rainfall acts as a scrubbing mechanism to the
atmosphere and washes gaseous and particulate material and transports to the ground
surface (OECD 1986). Randall et al. (1981) found that a significant proportion of
some pollutant loadings in urban areas could be attributed to wet deposition. This is
especially true for nitrogen which urban precipitation is a major source. Wet
deposition is controlled by the concentration and size distribution of atmospheric
24
particles, solubility and reactivity of gases and by the meteorological factors that
control precipitation (Lovett 1994).
The main sources of dry deposition from the atmosphere are dust emissions and
transport. Dustfall (particles <60 µm are considered as dust in most urban pollutant
studies) originates mostly from paved and unpaved roads, parking lots, railroads,
uncovered material storage sites, construction and demolition sites, urban refuse
disposal landfill operations and industrial emissions (Novotny et al. 1985; Huston et
al. 2009). In dry deposition, removal by stormwater runoff is not the sole mechanism
responsible for depletion of the pollutant load accumulated on the catchment surface.
Removal of pollutants can also occur through street sweeping, local turbulence
arising from the movement of vehicles on roads and wind events where the wind has
sufficient capacity to entrain the pollutant particles (Ball et al. 2000; Bohemen and
De Laak 2003).
Atmospheric deposition represents an important source of pollutants such as
nutrients and trace metals in stormwater (Jassby et al. 1994; Sabin et al. 2005; Wu
and Wang 2007). Jassby et al. (1994) noted that most of the of the dissolved
inorganic nitrogen (DIN) and total nitrogen in the annual nutrient load to Lake
Tahoe, California-Nevada could be attributed to atmospheric deposition. Davis et al.
(2001) found that atmospheric deposition is an important source of heavy metals
such as Cadmium, Copper and Lead originating from rooftops. The results of the
study by Davis et al. (2001) were supported in a study by Van Meter and Mahler
(2003). Atmospheric pollutants are generally present in three different forms as
liquid, solid and gaseous (Brinkmann 1985).
Common substances in atmospheric deposition are carbon monoxides, nitrogen
oxides, hydrocarbons and dust. These pollutants are washed out during the early
stages of rainfall or deposited on ground surfaces as solids and later washed-off (Hall
and Ellis 1985). They are brought into the urban atmosphere from long distances or
emitted from various sources on a regional or even local scale. Additionally,
physical and chemical processes that occur in the atmosphere change the phases of
some atmospheric pollutants (Brinkmann 1985; Novotny et al. 1985).
25
The concentration of pollutants in the atmosphere is influenced by meteorological,
topographical and land use factors which can vary widely. Lewis (1981) has shown
that in regions with a well defined seasonability of rainfall, a significant proportion
of the total annual atmospheric loading may be flushed out in the first few days after
the rain. Consequently, subsequent rainfall events would be relatively less polluted.
Wash-off of pollutants deposited on ground surfaces
The wash-off of pollutants build-up on the ground and other surfaces is an important
pathway for contaminated stormwater runoff in the urban environment. In addition
to flushing dissolved materials from impervious surfaces, rainfall-runoff events also
detach and transport significant amounts of particulate matter (Sartor et al. 1974;
Characklis and Wiesner 1997; Sansalone et al. 1998; Vaze and Chiew 2002).
The rate of removal of pollutants in particulate form is affected by the density of
particles, particle size distribution and rainfall and runoff characteristics such as the
intensity of rainfall and velocity of the runoff (Characklis and Wiesner 1997;
Fujiwara et al. 2005; Shaw et al. 2006; Shigaki et al. 2007). Furthermore, pervious
surfaces also contribute pollutant loadings to surface runoff during relatively high
intensity rainfall events. Sartor and Boyd (1972) noted that management practices
such as street cleaning also influence pollutant wash-off. The pollutant wash-off
process is discussed in detail in Section 2.2.2E.
C. Significance of impervious surfaces
The presence of impervious surfaces is an important issue for both the quantity and
the quality of urban runoff as they can produce considerable runoff even during
minor rain events (Pitt 1987; Gilbert and Clausen 2006; Kim et al. 2007). The
impervious surfaces are almost always hydrologically active because their
infiltration capacity is low. On the other hand, the natural pollutant removal
mechanisms provided by on site vegetation and soils have reduced opportunity to
remove pollutants from stormwater runoff (Connecticut 2004).
Consequently, increase in imperviousness in urban catchments results in more runoff
and hence higher pollutants loads. Examples of impervious surfaces include paved
26
parking lots, streets, driveways, roofs and sidewalks. Solids, nutrients and various
other pollutants accumulate on these impervious areas between storm events and are
eventually washed-off by runoff causing degradation of receiving water quality. The
study of Jartun et al. (2008), clearly showed that particle bound pollutants
accumulated on the impervious surfaces are easily picked up by stormwater runoff
and transported through the urban environment.
According to numerous researchers (for example, Sartor and Boyd 1972; Shaheen et
al. 1975; Bannerman et al. 1993; Ball et al. 1998; Kim et al. 2007) road surfaces and
parking lots are the most common types of impervious surfaces which contribute
pollutant loads to stormwater runoff. Bannerman et al. (1993) investigated the
pollutant loads in stormwater runoff from different impervious surfaces in
residential, industrial and commercial land uses. They noted that the majority of the
pollutants are generated from street surfaces and parking lots in all land use types.
The main reasons for this are the high imperviousness of these surfaces and high
pollutant mass emissions from direct and continuous anthropogenic activities such as
vehicular traffic (Bannerman et al. 1993; Kim et al. 2007).
However, road surfaces may not always be a substantial fraction of urban
catchments. According to Ball et al. (2000) in a typical residential area, road surfaces
will comprise about 10% to 15% of the total area. Therefore, other impervious
surfaces such as roofs, sidewalks, compacted soil surfaces and driveways too can
make a significant contribution to the pollution of stormwater runoff (Bannerman et
al. 1993; Van Meter and Mahler 2003; Chang et al. 2004; Huang et al. 2007).
Pollutant generation from impervious surfaces in a particular land use depends on
several factors such as percentage of impervious area, fraction of road and roof
surfaces within the impervious area and surface characteristics. Lee and Bang (2000)
observing the hydrographs and pollutographs of nine catchments in Korea found that
the pollution concentration peak occurred before the flow peak in catchments with
areas smaller than 100 ha where the impervious area was more than 80%. In
addition, they noted that the pollutant concentration peak is followed by the flow
peak in the catchments with areas larger than 100 ha in which the impervious area
was less than 50%.
27
Konrad et al. (1978) noted that for individual rainfall events, nutrient loading rates
generally increase with an increasing percentage of impervious surface area.
Goonetilleke et al. (2005) investigating six different land uses in Queensland state,
Australia, noted that even for the same fraction of impervious area, two catchments
can produce considerably different pollutant loads. As they noted, a number of
reasons could be attributed to this phenomenon. Firstly, it could be due to the
difference in the spatial distribution of impervious areas in a catchment. This would
have a significant influence on the time and velocity of travel of surface runoff and
hence the pollutant load. Secondly, it could be due to the difference in the fraction of
road surfaces within the impervious areas in the catchment they studied. Thirdly, it
could be due to the extent of the anthropogenic activities carried on the catchments.
D. Pollutant build-up
In general terms, pollutant build-up is the accumulation of pollutants on surfaces.
This is a dynamic process where there is equilibrium between pollutant deposition
and removal and between pollutant sources and sinks at any given time (Duncan
1995). Build-up can be measured directly by sweeping and washing of the
impervious area after an antecedent build-up period.
Numerous researchers have recognised that pollutant build-up load varies
significantly according to factors such as:
• Land use
• Antecedent dry period
• Fraction of impervious area
• Population density
• Traffic characteristics
• Fraction of road surfaces
• Street cleaning practices
• Pavement material and condition
• Climate
(Ball et al. 1998; Egodawatta and Goonetilleke 2006; Bian and Zhu 2008; Jartun et
al. 2008).
28
Road surfaces are a major contributor to pollutant build-up (Ball et al. 1998; Brodie
and Porter 2006; Sabin et al. 2006). Cleaning events such as wind, rain and road
sweeping have a significant impact on pollutant build-up on road surfaces. Data
collected by Sartor and Boyd (1972), Pitt (1979) and Ball et al. (1998) suggest that
the rate of accumulation of pollutants is high for several days after a street cleaning
or rainfall event. Then the rate decreases gradually and approaches to almost zero.
The investigation of accumulation of pollutants on road surfaces in different land
uses is an important issue in water quality research. The composition of pollutant
build-up and the loading of pollutants vary with the different land uses. This is
primarily due to the difference in nature of anthropogenic activities and the type of
pollutant sources. Sartor and Boyd (1972) from their detailed study of pollutant
build-up on road surfaces for different land use types such as residential, commercial
and industrial areas found that the major pollutants of street surfaces was inorganic,
mineral-like matter similar to common sand and silt. From their study, they noted
that industrial areas had the highest loads and accumulation rates due to less
sweeping, more unpaved areas and spillage from trucks and other vehicles. The
intermediate and the lowest pollutant loads belonged to residential areas and
commercial areas respectively. This was attributed to better road surfaces and more
frequent street sweeping.
Bian and Zhu (2008) investigating the pollutant build-up on road surfaces in
different urban land uses noted that the commercial area road surface they
investigated contributes the highest amount of organic matter, nitrogen and
phosphorus. The contribution of high amount of pollutant loads from the commercial
area road surface was attributed to sources such as high amount of anthropogenic
activities, higher vehicular traffic, tyre wear and plant debris in the commercial area
road surface.
The concept that pollutant build-up is influenced by the antecedent dry period is
fundamental to most water quality studies (Yamada et al. 1993; Fulcher 1994; Vaze
et al. 2000; Egodawatta and Goonetilleke 2006). Deletic (1998) and Vaze et al.
(2000) noted that pollutant build-up increases with the antecedent dry period.
According to Egodawatta and Goonetilleke (2006) the build-up rate reduces and
29
build-up load asymptote to a constant value as antecedent dry days increases.
Furthermore, Yamada et al. (1993) showed that the amount of pollutants that
accumulate on an urban paved surface is a function of antecedent dry days and the
frequency of storm events or street sweeping.
The amount of pollutant build-up varies across the road. According to Novotny et al.
(1985), the pollution profile along a road cross section can be considered as an
exponential decay towards the centreline of the road. On road surfaces, the pollutants
accumulated are redistributed by wind and vehicular induced turbulence. The
resuspended pollutants are more likely to be deposited in less turbulent areas of the
road. Deletic and Orr (2005) investigating an urban road in Aberdeen, Scotland
observed that 66% of the total solids load was within a 0.5 m strip next to the kerb.
Sartor and Boyd (1972) found that about 90% of the street dust was within about 30
cm of the kerb face that is typically within the gutter area.
However, Pitt (1979) found contradictory results from that of Sartor and Boyd
(1972), Ball et al. (1998) and Deletic and Orr (2005) . Pitt (1979) investigated the
dust distributions across a street and found that most of the street dust was actually in
the driving lanes, trapped by the texture of the rough surface. Where on-street
parking is common, much of the dust was found several distance out into the street
where most of the re-suspended (in air) street dirt blew against the parked cars and
settled on the pavement. He argued that the distributions found in the study by Sartor
and Boyd (1972) is valid only on smooth surfaced streets with moderate to heavy
traffic and with no on-street parking.
Furthermore, the loading of pollutants vary considerably between areas even having
similar land use type but with a different degree of use. This can be further explained
that different roof materials and different traffic densities within the same land use
type have different influence on pollutant distribution and concentration (Gobel et al.
2006).
Particle size distribution of solids accumulated on road surfaces is an important
parameter as it determines the mobility of the particles during wash-off and their
association with other pollutants (Deletic and Orr 2005; Lau and Stenstrom 2005;
30
Bian and Zhu 2008;). The particle size distribution of solids build-up varies with
factors such as type of land use, surrounding soil conditions and road surface
conditions such as texture depth (De Miguel et al. 1997; Herngren et al. 2006; Bian
and Zhu 2008).
Researchers have noted that build-up on road surfaces have a significant fraction of
fine particulates (for example Walker and Wong 1999; Herngren et al. 2006; Bian
and Zhu 2008). Herngren et al. (2006) found up to 90% of particles are below 150
µm in the road surface build-up on the residential, industrial and commercial road
surfaces investigated. Walker and Wong (1999) noted that 70% of the particles
found on Australian road surfaces investigated by them were smaller than 125 µm.
Similar observations have been noted by Bian and Zhu (2008) and Andral et al.
(1999) for the road surfaces they investigated.
According to several research findings, the finer fraction of the build-up is the most
important as a high amount of pollutants are also attached (Deletic and Orr 2005;
Herngren et al. 2006; Bian and Zhu 2008). This is attributed to relatively higher
adsorption capacity of fine particles due to their larger surface area (Sansalone et al.
1998; Li et al. 2008).
Sartor and Boyd (1972) modelled the build-up as starting from zero after street
cleaning or a heavy rain event. Vaze and Chiew (2002) questioned the assumption
that build-up starts from zero after a rain event. For example, Malmquist (1978)
through an experimental study and Chiew et al. (1997) through a modelling study
showed that storm events typically remove only a fraction of the total pollutant load.
According to Vaze and Chiew (2002), the two alternative views, which describe the
pollutant accumulation process are:
• Surface pollutant load builds up from zero over the antecedent dry days and
most of the available pollutant load is then washed-off during a storm event
(Figure 2.4 a).
• Storm events remove only a fraction of the surface pollutant load and build-up
occurs relatively quickly to return the pollutant load back to the level before the
storm. This phenomenon and the redistribution of pollutants would result in a
31
catchment surface having a similar amount of pollutant load most of the time
(Figure 2.4 b).
Figure 2.4- Hypothetical representations of surface pollutant load over time
(Adapted from Vaze and Chiew 2002)
Different mathematical relationships and various models have been developed to
replicate pollutant build-up and to predict the load and concentration of pollutants.
Pollutant build-up has typically been treated as a linear, exponential, power or log
normal function (Ball et al. 1998; Charbeneau and Barrett 1998; Vaze and Chiew
2002;).
Ball et al. (1998) deriving regression relationships for estimating build-up of
pollutants on road surfaces found that the power and hyperbolic functions produced
the most significant relationship for the build-up of pollutants on road surfaces. Even
though, both of these functions result in similar predictions during the build-up
period, there is one significant difference between them. This difference arises from
the hyperbolic function tending to reach an asymptotic value whereas the power
function does not. In addition, the hyperbolic function shows the increase of
pollutant load on the road surface at a decreasing rate. These tendencies are
illustrated in Figure 2.5 and Figure 2.6.
32
Figure 2.5- Hyperbolic function for estimation of the available solids
load
(Adapted from Ball et al. 1998)
Figure 2.6- Power function estimation of the available sediment load
(Adapted from Ball et al. 1998)
y=axb
33
According to the study by Ball et al. (1998), the best generic form of relationship in
estimating pollutants was found to be a power function, although the hyperbolic
function was also found to provide satisfactory relationships for the estimation of
many pollutants. However, limited data sets and the large variations have made it
hard to determine the exact form of the relationships with regards to pollutant build-
up (Duncan 1995).
E. Pollutant wash-off
Pollutant wash-off is the process of removing the accumulated pollutants from
catchment surfaces during rainfall events and incorporating into stormwater runoff.
Urban runoff originates both from impervious and pervious surfaces. The amount of
pollutants washed-off by rainfall from the surface depends on the amount of
pollutants that accumulate during the preceding dry period and the capacity of the
runoff to transport the loosened material (Novotny et al. 1985; Pitt 1987; Vaze and
Chiew 2002). Bujon et al. (1992) have noted that pollutant wash-off incorporates
two simultaneously occurring phenomena. Firstly, the catchment surface gets wet
with the rainfall and dissolves the pollutants accumulated on it. Secondly, due to the
impact energy of the raindrops, the pollutants detach from the surface and are
incorporated into the runoff.
Wash-off significantly varies with the primary characteristics of rainfall, runoff and
the surface. Deletic et al. (1997) have shown that rainfall and overland flow
characteristics would govern the wash-off process. Hartigan et al. (1978) and
Characklis et al. (1979a) noted that wash-off varies with runoff volume while
Aalderink et al. (1990) and Lager et al. (1971) noted that it is closely related to the
runoff rate. Sartor and Boyd (1972) and Egodawatta et al. (2006) confirmed that
wash-off is most strongly associated with rainfall intensity rather than other rainfall
and runoff parameters.
It is commonly accepted that the total amount of pollutant wash-off load will be
larger for storms with higher rainfall intensity. That is due to the particle detachment
by raindrop impact energy and the suspension and transport of overland flow due to
high flow energy and the transport capacity of the flow (Shivalingaiah and James
1984(a); Coleman 1993; Duncan 1995; Vaze and Chiew 2002).
34
Chui (1997) showed that event mean concentrations of total suspended solids
increases with increasing rainfall intensity. Vaze and Chiew (2002) further
strengthened this concept having noted that common storms only remove a small
proportion of the total surface pollutant load. However, rainfall runoff relationships
are well known and the variability between rainfall volume and its contribution to
wash-off pollutant loads is significant (Ebbert and Wangner 1987). Additionally,
when the overland flow is disturbed by physical factors such as vehicles and wind, it
also increases the wash-off. Shivalingaiah and James (1984a) found that wash-off
from roads increases with the increase in vehicle movements on a wet road.
The effect of antecedent dry period on wash-off is less clear than its effect on build-
up. While some researchers (for example Mance and Harman 1978; Yamada et al.
1993; Ren et al. 2008) have noted a relationship between antecedent dry period and
wash-off, some have found (Weeks 1981; Hoffman et al. 1984) no significant
relationship between these factors. However, most importantly pollutant wash-off is
influenced by the preceding build-up process.
Ren et al. (2008) found that the peak concentrations of total nitrogen and total
phosphorus in roof runoff increased with the increase in the antecedent dry period.
According to Duncan (1995), there is a notable correlation between pollutant build-
up and wash-off. However, according to Coleman (1993), build-up is not the limiting
factor in determining wash-off loads. Coleman (1993) further suggested particle
detachment as the critical process. This conclusion was further strengthened by
Egodwatta and Goonetilleke (2008) who noted that a rainfall event has a specific
capacity to mobilise pollutants and it always removes only a fraction of the build-up
load.
The first flush effect is a distinctive feature of wash-off. First flush produces higher
concentrations of pollutants early in the runoff event. Kim et al. (2007) found that
pollutant concentrations sharply decreased in the initial 30–40 minutes of the storm
period. However, the duration of the first flush on a given catchment depends on the
time of concentration and generally increases with the catchment size (Weeks 1981).
It has often been noted that the peak of the pollutant concentration in the first flush
precedes the peak in the runoff (Deletic 1998; Duncan 1995).
35
Hoffman et al. (1985) evaluating highway runoff, found that the concentration of
monitored pollutants such as heavy metals, hydrocarbons and suspended solids is
considerably higher during the first flush. Kayhanian et al. (2008), investigating
runoff from highly urbanised highway sites in Los Angeles, California found that
more than 40% of the toxicity was associated with the first 20% of the discharged
runoff volume. Furthermore, on average, 90% of the toxicity was observed during
the first 30% of storm duration. Consequently, they suggested that stormwater
quality mitigation actions focusing on treating the first portion of stormwater runoff
volume might be more beneficial in protecting the water environment.
The effect of road sweeping in the case of pollutant wash-off is of crucial
importance. Most of the pollutants are associated with the finer fraction of solids
(Sartor and Boyd 1972; Vaze and Chiew 2004; Herngren et al. 2005, 2006). When
sweeping, the sweeper releases the fine particles but may not have enough suction to
remove them completely from the surface. This makes the fine particles readily
available for wash-off by the next storm (Vaze and Chiew 2002).
A number of research studies have confirmed that road sweeping is generally
ineffective in improving stormwater runoff quality (Sartor and Boyd 1972; Pitt 1979;
Walker and Wong 1999). Walker and Wong (1999) investigated the efficiency of
Australian road sweeping practices in the removal of pollutants from street surfaces.
They concluded that current practices are generally ineffective as an at-source
stormwater pollution control measure. They suggested that road sweeping should be
accompanied by structural pollutant treatment measures to effectively reduce the
discharge of pollutants into stormwater.
Pollutant wash-off is commonly modelled as an exponential decay of the available
surface pollutant load with rainfall intensity, rainfall volume and runoff rate or
runoff volume used as the explanatory variables to describe the decay (Sartor and
Boyd 1972; Bujon et al.1992; Egodawatta and Goonetilleke 2008). In the detailed
investigation of street dirt wash-off by Sartor et al. (1974), they used an exponential
equation as described below, by assuming that the rate of removal of a given particle
size is proportional to the street dirt loading and the constant rain intensity.
36
W =Wo (1-e-kIt) Equation 2.1
Where:
W - Weight of the material mobilized after time t
Wo - the initial weight of the material on the surface
I - rainfall intensity
K - Wash-off coefficient
Egodwatta et al. (2007) showed that a storm event has the capacity to wash-off only
a fraction of pollutants available and this fraction varies primarily with rainfall
intensity, kinetic energy of rainfall and characteristics of the pollutants. Therefore,
they suggested a modification for the equation proposed by Sartor et al. (1974) in
order to incorporate the wash-off capacity of rainfall. The format of the equation
they developed was:
Where:
Fw =W/Wo=CF (1-e-kIt) Equation 2.2
Fw - Fraction of wash-off
CF - Capacity factor
According to Egodawatta et al. (2007) CF varies between 0 to 1 depending on the
rainfall intensity.
However, Duncan (1995) has pointed out the ineffectiveness of using an exponential
form for pollutant wash-off. His argument was based on the fact that an exponential
wash-off function cannot simulate an increase in concentration at any time during a
storm. He noted four explanatory variables that relate to wash-off. They are rainfall
intensity and volume and runoff rate and volume and the two main processes; shear
stress generated by surface flow and energy input from raindrops. The problem with
four explanatory variables is their correlation with each other. Duncan (1995) further
argued that it is quite possible that different processes dominate under different
conditions or at different scales. Asante and Stephenson (2006) found that pollutant
wash-off is better estimated as a power function of rainfall and runoff volume and
rate as independent variables.
37
2.3 Primary water pollutants in an urban environment
As urban stormwater runoff has been recognised as one of the major sources for the
deterioration of water quality in receiving water bodies, it is critically important to
identify the types of urban stormwater pollutants and their specific characteristics.
The primary stormwater pollutants which are present on urban catchments are:
• Suspended solids
• Organic carbon
• Nutrients
• Heavy metals
• Hydrocarbons
• Litter
• Pathogens
2.3.1 Suspended solids
Suspended solids has additional significance as a stormwater pollutant because as
highlighted in several research studies, other pollutants such as heavy metals,
nutrients, pathogens and hydrocarbons are associated with the solid particles (For
example, Sartor and Boyd 1972; Bodo 1989; Hvitved- Jacobsen et al. 1994; Ball et
al. 2000; Vaze and Chiew 2004; Jartun et al. 2008). Suspended solids in stormwater
runoff originate from both impervious and pervious areas in the catchment. Potential
sources of suspended solids are erosion of landscaped areas, construction and
demolition activities, open areas that drain to the site, floodwaters, irrigation
activities, road surface erosion, vehicles and maintenance activities (Ellis 1989;
Nelson and Booth 2002; Jartun et al. 2008).
The particle size of suspended solids is an important issue because it determines the
mobility of the particles and their associated pollutant concentrations (Dong et al.
1983; Herngren et al. 2005; Zafra et al. 2008). Coarse particles may not be carried in
suspension even in the fastest of flows. As Dong et al. (1983) has identified, finer
material stays in suspension for a longer period of time due to its relatively larger
surface area and electrostatic charge and is therefore transported a greater distance
38
by runoff. Consequently, the finer fraction of suspended solids has been recognised
as the most significant transporter of pollutants (Sartor and Boyd 1972; Vaze and
Chiew 2004; Zafra et al. 2008).
According to Robert and Kondolf (1993), suspended solids is the most signficant
non point source of pollution on a volumetric basis. During the early part of
stormwater runoff, the concentration of suspended solids is usually higher than that
for secondary treated sewage effluent (Cordery 1977; Taebi et al. 2004).
Most importantly, anthropogenic sources contribute a high amount of fine particles
in urban environments than natural sources (Fergusson and Ryan 1984). Road
surfaces are a major contributor of fine particles into stormwater runoff (Drapper et
al. 2000; Nelson and Booth 2002; Egodawatta and Goonetilleke 2008; Zafra et al.
2008). Drapper et al. (2000) who investigated runoff from a number of roads with
different traffic volume, surrounding land uses and pavement surface types noted
that a significant proportion of the suspended solids present in the urban runoff is
smaller than 100 µm.
According to Andral et al. (1999), it is particles smaller than 100 µm in diameter,
which remain in suspension, while particles larger than 100 µm are easily separated.
Therefore, they concluded that when treating stormwater runoff, particles smaller
than 100 µm in diameter, which can represent up to 90% of the weight of solids
remaining in suspension in runoff, should be removed. According to Robert and
Kondolf (1993), finer particles have a higher capacity per unit of mass to adsorb
pollutants and tend to have a higher proportion of organic matter. Sartor and Boyd
(1972) also noted significant fraction of pollutants is attached to the finer fraction of
solids from their detailed investigation of road surface solids build-up in residential,
industrial and commercial land uses in USA. They found significantly high
percentage of nutrients and organic material in particles smaller than 43 µm although
this fraction constituted only 5.9% of the total mass of suspended solids in the
samples analysed. They also noted a significantly high percentage of heavy metals in
the fraction of solids which was smaller than 104 µm (See Table 2.3).
39
Furthermore, the deposition of suspended solids at the bottom of the channels
accelerates the silting up of water bodies and increases the potential for flooding.
Additionally, the accumulation of suspended solids on the bed of the receiving water
body is detrimental to aquatic species due to the biochemical reactions which occur
within the deposited solids (Hall and Ellis 1985; House et al. 1993). Additionally,
turbidity associated with fine suspended solids reduces light penetration.
Suspended solids concentration in stormwater runoff can vary with several factors
such as rainfall intensity and duration (Deletic 1998; Egodawatta et al. 2006).
Egodawatta et al. (2006) analysing water quality and rainfall-runoff data for a mixed
urban catchment in Gold Coast, Queensland, Australia found that average rainfall
intensity strongly correlates with the total suspended solids concentration and the
load. Tiefenthaler and Schiff (2001) noted that the highest suspended solids
concentration occurred within the first four to six minutes of each rainfall event and
decreased to a relatively consistent level with the increase in time duration.
Table 2.3- Fraction of pollutants associated with different particle size ranges-
(Adapted from Sartor and Boyd 1972)
Average percentage by weight Parameter
>2000 µm
840-
2000
µm
246-
840
µm
104-
246
µm
43-
104
µm
<43
µm
Total Solids 24.4 7.6 24.6 27.8 9.7 5.9
Volatile Solids 11.0 17.4 12.0 16.1 17.9 25.6
BOD 7.4 20.1 15.7 15.2 17.3 24.3
COD 2.4 4.5 13.0 12.4 45.0 22.7
Kjeldahl Nitrogen 9.9 11.6 20.0 20.2 19.6 18.7
Nitrates 8.6 6.5 7.9 16.7 28.4 31.9
Phosphates 0.0 0.9 6.9 6.4 29.6 56.2
Total Heavy metals 16.3 17.5 14.9 23.5 27.8
Total pesticides 0 16.0 26.5 25.8 31.7
40
2.3.2 Organic carbon
Organic carbon or oxygen demanding wastes are primarily organic materials that are
oxidised by bacteria to carbon dioxide and water. The decomposition of organic
carbon in water bodies can deplete dissolved oxygen present, which is one of the
most important indicators of water quality. In addition, it also leads to undesirable
odours, endanger water supplies and decrease the recreational value of waterways
(Ellis 1989). Common sources of organic carbon are, vegetation debris, animal
waste, street litter, vehicular activities and tyre wear (Rogee et al. 1993; Strynchuk et
al. 1999).
Sartor and Boyd (1972) found road surfaces as a major contributor of oxygen
demanding materials to receiving water bodies. The concentration of organic carbon
in runoff from road surfaces was found to depend on the characteristics of the urban
area and the frequency of the road sweeping. They also noted that the accumulation
of organic material on road surfaces is much faster than that of inorganic materials.
Sansalone and Tittlebaum (2001) investigating runoff from a major road overpass in
Louisiana, USA and found that organic matter represented 29% on average of the
TSS concentration.
Gromaire-Mertz et al. (1999), found the organic fraction of suspended solids
collected in road runoff varied from 40 to 70%. According to Sartor and Boyd
(1972), finer particles of suspended solids contain more organic matter than the
coarser particles. This further confirms the need to implement stormwater treatment
facilities aimed at removing finer particles in the stormwater runoff. Furthermore,
organic carbon adsorbed on suspended solids increases their absorption capacity for
combining with hydrophobic organic chemicals and some heavy metals such as Pb
and Zn (Parks and Baker 1997).
2.3.3 Nutrients
Nutrients are chemical compounds such as nitrogen, phosphorus, carbon, calcium,
potassium, iron and manganese. From a water quality perspective, nutrients that play
41
the most vital role in the deterioration of water quality are nitrogen and phosphorous
(Carpenter et al. 1998; Wit and Bendoricchio et al. 2001; Taylor et al. 2005; Ballo et
al. 2009). Urban stormwater runoff typically contains elevated concentrations of
nitrogen and phosphorous that are most commonly derived from lawn fertilizer,
detergents, animal waste, atmospheric deposition, organic matter, vehicle exhausts
and improperly installed or failing on-site wastewater treatment systems (Graves et
al. 2004; Ellis 1989; Puckett 1995; Wong et al. 2000). Cordery (1977) investigating
the quality characteristics of urban stormwater in Sydney, Australia found that the
nutrient concentrations in stormwater runoff is in the range of 5-40% of secondary
treated sewage effluent.
Excess nutrients can be considered as pollutants when their concentrations are
sufficient to allow excessive growth of aquatic plants such as algae. Nutrient
enrichment which is referred to as eutrophication can lead to blooms of algae which
eventually die and decompose resulting in oxygen depletion. Many studies have
demonstrated the impact of excessive nitrogen and phosphorous loads on receiving
water bodies. A study of Neuse river catchment in North Carolina, USA found that
high nitrogen loading in the river can lead to excessive algal blooms, low level of
dissolved oxygen and large fish kills (Artola et al.1995; Paerl et al. 2006). According
to Sonzogni et al. (1980), phosphorous is the key nutrient affecting productivity of
the Great Lakes, USA and hence the most important pollutant in runoff water from
the contributing catchments.
A variety of impairments can result from excessive plant growth associated with
nutrient loadings. These impairments result primarily when dead plant matter settles
to the bottom of a water body, stimulating microbial breakdown processes that
require oxygen. This decomposition removes oxygen from the water which leads to a
reduction in the dissolved oxygen. This in turn leads to changes in the benthic
community structure from aerobic to anaerobic organisms. Oxygen depletion might
also occur nightly throughout the water body because of plant respiration. Extreme
oxygen depletion can adversely affect aquatic life and toxins can be released from
solids when dissolved oxygen and pH are lowered (Brick and Moore 1996; Wong et
al. 2000). Furthermore, the breakdown of dead organic matter in water can produce
un-ionised ammonia that can adversely affect aquatic life.
42
Algae can clog water treatment plant filters and reduce the time between
backwashing. The greater concern is the contamination of drinking water supplies
with nitrates. Excessive nitrates in drinking water can cause Methemoglobinemia in
infants and stomach cancers. In addition, water supplies containing more than 10
mg/L of nitrate can taste bitter and can cause physiological distress (Straub1989).
Additionally, an increased rate of production and breakdown of plant matter can also
deteriorate the water quality in terms of taste and odour (US EPA 1999).
The excessive plant growth in a eutrophic water body can affect recreational water
use. Extensive growth of algae and other plants can interfere with swimming,
boating and fishing activities while appearance and odour of decaying plant matter
impair the aesthetic uses of the water body (US EPA 1999).
Nutrients are transported into stormwater runoff in both particulate and dissolved
forms. As phosphorus has a tendency to adsorb to soil particles and organic matter, it
is primarily transported in surface runoff in particulate form with eroded soil
particles (Uusitalo et al. 2000; Quinton et al. 2001; Zhao et al. 2007). Phosphorus
associated with fine grained particulate matter also exists in the atmosphere. The
sorbed phosphorus can enter surface runoff by both dry fallout and rainfall (US EPA
1999). On the other hand, nitrogen does not adsorb strongly and can be transported
in both particulate and dissolved phases in surface runoff (Wong et al. 2000; Vaze
and Chiew 2004). Additionally, the gaseous phase of nitrogen can be transported to
surface water via atmospheric deposition.
2.3.4 Heavy metals
According to research literature, urban stormwater runoff can contain significant
amounts of heavy metals (for example, Pitt 1979; Davis et al. 2001; Brezonik and
Stadelmann 2002; Herngren et al. 2006). The presence of heavy metals in urban
stormwater runoff is of concern because of their potential toxicity. Additionally,
unlike most other stormwater pollutants, heavy metals do not degrade in the
environment. Heavy metals can be defined as metals with specific gravity greater
than about 4 or 5. The common heavy metals found in stormwater runoff are
43
Mercury (Hg), Lead (Pb), Cadmium (Cd), Arsenic (As), Nickel (Ni), Chromium
(Cr), Zinc (Zn), Iron (Fe), Copper (Cu) and Aluminium (Al) (Sartor and Boyd 1972;
Davis et al. 2001; Gobel et al. 2006; Herngren et al. 2006). For most metals, it is the
cumulative load that is important due to their accumulation in the environment rather
than the instantaneous flux (Ball et al. 2000).
The primary sources of heavy metals in urban stormwater runoff are vehicular
traffic, fossil fuel combustion, corrosion of galvanised and chrome plated products,
roof runoff and industrial activities. Sartor and Boyd (1972) found that the highest
metal concentrations in road sweepings are from industrial areas. According to
Sartor and Boyd (1972) and Sansalone et al. (1996), vehicular related sources of
heavy metals are vehicular component wear, fuel leakages and fuel combustion.
Table 2.4 shows the sources of heavy metals from traffic related activities in an
urban environment.
Table 2.4- Sources of heavy metals from traffic related activities
(Adapted from Sansalone et al. 1996)
Source
Tyre wear Brake wear Engine wear Fuel leakage
Al
Cd
Cr
Cu
Fe
Mn
Ni
Pb
Zn
As several researchers have identified that roofs also make a significant contribution
of heavy metals to stormwater runoff (Bannerman et al. 1993; Forster 1996; Gnecco
et al. 2005; Gobel et al. 2006). This is mainly due to the corrosion of metallic
44
components on roof surfaces. Different metal elements are used for roof surfaces; for
example, Lead (Pb), Copper (Cu) and Zinc (Zn) are used as roof covering, gutters
and down pipes. Furthermore, Van Meter and Mahler (2003) estimated that the
contribution of roof wash-off to catchment loading to range from 6% for Cr and As
to 55% for Zn. The corrosion processes are enhanced because of the low pH of
rainwater. Several other sources of heavy metals into urban runoff are asphalt
paving, metallurgical industries, medical uses and car washing liquids.
The concentration of metal fluxes that are released from various sources depend on:
• surrounding land use
• type of metal
• extent of use of metallic products in roofs and other building parts
• antecedent dry days
• street cleaning practices
(Thomas and Greene 1993; Davis et al. 2001; Vermette et al. 1991).
According to Davis et al. (2001), the high metal concentrations in roof runoff
originate from the roof material and not from the atmospheric depositions collected
on the roof. The contribution of heavy metals from roof surfaces varies with the
roofing material. Van Meter and Mahler (2003) in their study of two different types
of roofs; asphalt shingle and galvanised metal roofs found that, while metal roofing
was a source for Cd and Zn whilst asphalt shingle was a source of Pb. Gobel et al.
(2006) noted significant differences in the concentration of heavy metals in the
runoff generated from roof surfaces made by different metals such as Cu, Zn and Al.
According to them, this difference can be attributed to the impact of the pH value on
the material.
Davis et al. (2001) noted that building construction with metallic components is
more common in non-residential buildings. For example, Brezonik and Stadelmann
(2002) found that commercial and industrial land uses in Minnesota, USA
contributed a higher amount of heavy metals than a residential site. Herngren et al.
(2006) also confirmed similar results after investigating three different land uses
(residential, industrial and commercial) in Queensland State, Australia. According to
45
Herngren et al. (2006), although the highest concentration of most of the heavy
metals investigated was found at the industrial site, the highest heavy metal loading
coincided with the highest sediment load at the commercial site.
Most of the heavy metals have a strong affinity to attach to smaller particle sizes of
solids due to higher adsorption capacity of fine solids resulting from a relatively
larger surface area (Deletic and Orr 2005). On the other hand, conventional pollutant
removal processes such as road sweeping is capable of only removing large particles.
Therefore, road sweeping may have little impact in reducing the concentration of
heavy metals in urban runoff as fine suspended solids are readily transported in
stormwater.
However, Dong et al. (1984) found contradictory results to the concept that fine
particles have a higher content of sorbed metals. They found higher contents of Cr,
Cu, Fe and Ni in the coarse fractions of urban road dust than in the fine fraction.
They suggested that the higher contents of Cr, Cu, Fe and Ni in coarser fraction are
attributed to the contribution of these metals from eroded metal surfaces.
2.3.5 Hydrocarbons
Urban stormwater runoff transports a wide array of hydrocarbon compounds into
receiving waters (Hoffman et al. 1984,1985; Larkin and Hall 1998). Increase in the
concentration of hydrocarbons leads to an increase in the toxicity of receiving water
bodies (Polkowska. et al. 2001). In general, hydrocarbons refer to organic
compounds, which contain carbon and hydrogen. According to Ball et al. (2000),
hydrocarbons in stormwater runoff show a strong affinity to suspended solids.
Hydrocarbons can originate from both natural and anthropogenic sources (Van;
Meter et al. 2000; Barbara et al. 2009). Anaerobic degradation of organic materials
and forest fires are the natural sources of hydrocarbons. Anthropogenic sources
which provide significant concentration of hydrocarbons into stormwater runoff
include roads, parking lots, vehicle service stations and bulk petroleum storage
facilities. Gobel et al. (2006) noted that stormwater runoff from road surfaces
46
contain more hydrocarbons than the stormwater runoff from roof surfaces.
Additionally, asphalt particles also have been recognised as a significant source of
hydrocarbons in urban runoff (Hoffman et al. 1984).
The main types of hydrocarbons that degrade water quality are total petroleum
hydrocarbons (TPH) and particularly polycyclic aromatic hydrocarbons (PAHs)
which forms a subgroup within TPH. Kayhanian et al. (2006) noted that there can be
an appreciable concentration of petroleum hydrocarbons from highway runoff.
Latimer et al. (2004) noted that used crankcase oil as the primary source of
petroleum hydrocarbons in all land use types they investigated.
PAHs contain varying number of Benzene rings. Compounds with more than three
rings such as Benzopyrene is considered to be carcinogenic (Gobel et al. 2006). Due
to their high hydrophobic and stable chemical structure, PAHs are not very soluble
in water (Marsalek et al. 1997). The amount of PAHs in stormwater runoff increases
considerably with increasing automobile use (Van Meter et al. 2000). The specific
sources of PAHs in relation to motor vehicles are tyre wear, crankcase oil and fuel
exhausts (Walker et al. 1999). Hydrocarbons in urban runoff are mostly associated
with particulate matter. Hoffman et al. (1984) found that the distribution of total
hydrocarbons associated with particulate material in runoff to range from 79% to
96%.
2.3.6 Litter
Litter can be categorised as artificial litter and natural litter. Artificial litter includes
materials such as glass, metal and plastics and mainly originate from building or
demolition activities. Generally, natural litter originates from vegetative materials
such as plant debris. Shaheen (1975) noted that 20% of the weight of pollutants
accumulated on road surfaces is litter. Litter not only affects the aesthetic appearance
of the water surface, but can also clog the drainage system. Litter can be effectively
removed by regular road sweeping (Sartor and Boyd 1972).
47
2.3.7 Pathogens
Pathogens are disease causing organisms that grow and multiply within the host.
Common examples of pathogens associated with water include, bacteria, viruses and
parasitic worms. These pathogens can cause epidemics of significant proportions.
Bacteria and viruses are responsible for a number of waterborne diseases such as
cholera and hepatitis. Sources of pathogens in stormwater runoff include animal
waste from pets, wildlife, combined sewers, failing decentralised wastewater
treatment systems and illegal sanitary sewer cross-connections.
The concentration of pathogens in stormwater runoff varies with the type of land
use. Wanielista et al. (1977) found that total coliforms were higher in industrial areas
than in commercial areas. Furthermore, they noted that the lowest total coliform
loading is in residential areas.
2.4 Nutrient build-up and wash-off in urban catchments
The investigation of the role of nutrient build-up and wash-off in urban catchment
surfaces is crucially important as it is a significant source of urban stormwater
pollution. Past researchers have recognised stormwater runoff as a considerable
nonpoint source of nutrient loads to receiving water bodies (for example, Brezonik
and Stadelmann 2002; Hranova et al. 2002; Tomasko et al. 2005; Edwatds and
Withers 2008).
According to Novotny and Chesters (1981), on average about 50% of phosphorus
and an even greater proportion of nitrogen, originate from uncontrolled urban runoff.
There is considerable variability in nutrient export data from urban catchments. Line
et al. (2002) quoting data from other studies noted that the annual average export
values from urban areas have ranged from 1.6 to 38.5 kg/ha for nitrogen and from
0.03 to 6.23 kg/ha for phosphorus.
Nitrogen and phosphorus in stormwater runoff are present in different forms. The
different forms of nitrogen that impair stormwater quality are dissolved inorganic
48
nitrogen which includes ammonia (NH4+ and NH3), nitrite (NO2
-) and nitrate (NO3-)
and organic nitrogen which is in particulate or dissolved form (Heathwaite et al.
1993; Lee and Bang 2000; Taylor et al. 2005; Ballo et al. 2009). The most important
forms of nitrogen in terms of their immediate impact on water quality are ammonium
and nitrates (US EPA 1999). Both ammonium and nitrate ions are readily available
algal nutrients which can lead to eutophication of water bodies. Organic nitrogen is
convertible to ammonium and nitrate by bacterial processes and thereby made
available for eutrophication of receiving water bodies (Hart and Grace 2000). Total
Kjeldahl nitrogen represents the organic form of nitrogen in stormwater (Lee and
Woong 2000; Kayhanian et al. 2007). Total nitrogen (TN) includes all forms of the
nitrogen in both dissolved and particulate phases.
Phosphorus in water exists in either in organic or inorganic form and available in
particulate or dissolved phases. Organic particulate phosphorus includes living and
dead particulate matter such as plankton and detritus. Dissolved organic phosphorus
is derived from organisms and colloidal phosphorus compounds. On the other hand
inorganic particulate phosphorus includes phosphorus precipitates, phosphorus
adsorbed to particulates and amorphous phosphorus. H2PO4-, HPO4
2- and PO43- are
the forms of soluble inorganic phosphorus which are also known as soluble reactive
phosphorus (SRP). According to USEPA (1999), PO43- measures the phosphorus that
is most immediately bioavailable. Most of the soluble phosphorus in stormwater is
usually present in the form of PO43- . Total phosphorus (TP) is the measurement of
all forms phosphorus.
Atmospheric deposition is one of the important sources of nutrients to catchment
surfaces. The potential contribution of atmospheric deposition to nutrient loading on
urban catchment surfaces has been the focus of several research studies (for example
Randall et al.1981; Michael and Timpe 1999; Hope et al. 2004; Gonzalez Benitez
2009). Nitrates have relatively high atmospheric lifetime compared to ammonia.
Ammonia is very reactive and has much shorter atmospheric lifetime. Therefore,
ammonia will be present in high concentrations around its sources and much lower
concentration away from the source. Hope et al. (2004) noted high concentrations of
ammonia at the commercial and residential sites that they studied, as they were
closer to agricultural areas that are likely to emit ammonia.
49
Michael and Timpe (1999) concluded that about 28% of the stormwater runoff
nitrogen loading is directly attributable to wet atmospheric deposition alone, with the
remainder originating from the catchment. They further found that 15 -20% of the
atmospheric wet deposition of nitrogen loading discharges from the catchment
immediately as runoff whilst the remaining 80-85% of the atmospheric wet
deposition nitrogen input is assumed to be retained at least temporarily within the
catchment area, entering the normal nutrient cycle.
As common to other types of pollutants, the contribution of anthropogenic activities
to nutrients build-up is significant (Puckett 1995; Line et al. 2002; Bian and Zhu
2008; Ballo et al. 2009). Researchers have noted that the concentration of nutrients
in stormwater is strongly associated with residential activities (Brezonik and
Stadelmann 2002; Xian et al. 2006). Schoonover and Lockaby (2006) and Kolasa
(1999) found that residential fertilisers, which mainly include lawn and garden
fertiliser is the largest nutrient contributor to stormwater. Pitt et al. (2004) found that
lawns could contribute as much as 50% of the annual total phosphorus load in a
residential area.
However, Cordery (1977) has shown that fertilisers make an insignificant
contribution of phosphorus and that a major source is probably vegetable material
such as leaves and grass cuttings. Strynchuk et al. (1999) too, has identified grass
clippings and leaves as a significant source of nutrients to stormwater. The amount
of nutrients that leaches from grass clippings and leaves vary with time that has
elapsed after deposition. For example, they noted that most of the phosphorous was
released within the first day of deposition on the catchment surface.
Loading of nutrients from these sources and their concentration in stormwater wash-
off varies widely, making the determination of nutrients export rates for a given
catchment difficult. A large number of factors are responsible for influencing the
amount of nutrient build- up and wash-off in urban catchments. These include
factors such as:
• Type of land use
• Fraction of impervious area
50
• Cleaning and maintenance activities
• Soil texture
• Rainfall and runoff characteristics
• Catchment characteristics
• Population density
• Seasonal variations
(Panuska and Lillie 1995; Line et al. 2002; Tomasko et al. 2005; Xian et al. 2006;
Bian and Zhu 2008)
Line et al. (2002) investigating seven different land uses noted that the highest total
nitrogen export is from construction sites during the building phase. The main reason
which could be attributed for this is the high runoff rate produced by the construction
site due to the compacted soil surfaces compared to other sites.
Atasoy et al. (2006) estimating the effects of urban residential development on water
quality in upper Neuse river catchment, North Carolina showed that both the extent
of urban residential land use and the amount of conversion of undeveloped land to
residential use significantly increased nutrient loadings. They further noted that the
magnitude of the impact of residential use are similar for total nitrogen (TN) and
total phosphorous (TP) and the impact of land conversion is larger than the impact of
the amount of land in urban residential use, particularly for TP levels.
Sartor and Boyd (1972) in their comprehensive study of street surface pollutants in a
number of cities in the United States found significantly high nutrient loadings in
industrial sites compared to residential and commercial sites. However, Kayhanian et
al. (2006) found significantly high concentrations of TP and TKN (Total kjeldahl
nitrogen) in highway runoff from commercial areas compared to other land uses
such as residential, transportation and mixed land use. On the other hand, they could
not observe any significant variation for NO3--N with the surrounding land use.
Further strengthening the findings of Kayhanian et al. (2006), Bain and Zhu (2008)
also found that there was a higher amount of both TN and TP in the commercial land
use compared to the residential area and the intense traffic area they investigated.
51
This further underlines the uncertainty associated with estimating nutrient loadings
from urban areas.
The nutrient loadings increase with increasing impervious surface area. This is most
likely due to the ease of wash-off and transport in kerb and gutter systems and on
other hard surfaces (Schueler 1994; Brezonik et al. 2002). Xian et al. (2006)
analysing the urban development and its environmental impacts on Tampa Bay
catchment in West-Central Florida noted that nutrient build-up is significantly
correlated with the extent of impervious surface area. Furthermore, Schoonover and
Lockaby (2006) noted that nutrient concentrations within catchments with
impervious surface area >24% were often higher than in non-urban catchments (i.e.,
<5% impervious area) during both base flow (1.64 mg/L versus 0.61 mg/L) and
storm runoff (1.93 mg/L versus 0.36 mg/L). According to them the main reason for
this difference was the increase of impervious surface area.
The type of source areas where pollutants accumulate, affects the concentration of
nutrients in stormwater runoff. Hope et al. (2004) noted comparatively higher
loading of nitrate nitrogen (151.2 mg m-2) in the runoff from parking lots than the
results (4.6 mg m-2) obtained by Sartor and Boyd (1972) for road surface runoff.
According to Hope et al. (2004), the difference is that traffic turbulence on road
surfaces limits the accumulation of nutrients compared to parking lots, where
turbulence is likely to be lower due to much slower vehicle speeds. Concentration of
P and TKN from different source area categories in two different land uses is shown
in Table 2.5.
52
Table 2.5- Summary of observed particulate quality for P and TKN in different
land uses (means for <125 µm particles) (mg constituent/kg solids)
(Adapted from Pitt and McLean 1986)
Concentration (mg/kg solids) Residential/Commercial Land Use P TKN
Roofs
Paved parking
Unpaved driveways
Paved driveways
Dirt footpath
Paved sidewalk
Garden soil
Road shoulder
1500
600
400
550
360
1100
1300
870
5700
790
850
2750
760
3620
1950
720
Industrial Land Use
Paved parking
Unpaved parking/storage
Paved footpath
Bare ground
770
620
890
700
1060
700
1900
1700
The type of paving material on the surface also has a considerable influence on
nutrient build-up and wash-off (Ellison and Brett 2006). Hope et al. (2004)
recognised asphalt parking lot surfaces as a significant source area for nutrients.
Investigating the stormwater runoff from asphalt parking lot surfaces in different
land uses, they noted higher soluble inorganic nitrogen loads than in the wash-off
from natural desert soil surfaces. The reason that can be attributed to this is the
enhanced atmospheric nitrogen deposition and accumulation rates on urban asphalt
surfaces. On the other hand, plant or microbial uptake may result in relatively low
nitrogen loads on soil surfaces.
Dixon and Murray (1999) investigating the bulk atmospheric deposition of nutrients
in Tampa Bay catchment in Florida noted that seasonal variations are important in
atmospheric deposition of nutrients. They found that the highest nitrogen, loading is
during the summer season. McCorquodale et al. (2002) investigating the
53
composition of urban stormwater in New Orleans noted that the average
concentrations of NO3--N and TP have clear seasonal variations with higher levels in
spring and summer and lower levels in winter. The effect of wind and lightening
may cause the nitrogen concentrations to be higher in the summer as lightening
converts large amount of atmospheric nitrogen directly to nitrates (McCorquodale et
al. 2002). Furthermore, the wind would carry over long distances nutrients resulting
from sources such as industrial discharges and vehicular emissions.
Additionally, several researchers have noted that there is considerable correlation
between nutrient load in the wash-off with the rainfall duration and runoff volume
(Brezonik and Stadelmann 2002; Tomasko et al. 2005). Brezonik and Stadelmann
(2002) noted that all forms of nitrogen and phosphorus pollutant loads correlate with
rainfall duration. Downs (2003) noted strong correlation between annual phosphorus
loading and the runoff volume.
There is no doubt that stormwater runoff from urban areas is a significant source of
nutrients to receiving water bodies. However, the studies relating to the investigation
of nutrient build-up and wash-off process are limited. It is crucially important to
have a clear understanding of processes which lead to transport of nutrients in
stormwater runoff to ensure that better management strategies and treatment
facilities can be designed targeting the effective removal of nutrients from the
stormwater.
According to research literature, phosphorus is mostly attached to particulates while
most nitrogen is in dissolved form (Hvitved- Jacobsen et al. 1994; Sakai et al. 1996;
Vaze and Chiew 2004; Kato et al. 2009). According to Nelson and Booth (2002), the
finer fraction of suspended solids in stormwater runoff potentially contributes to
long-standing eutrophication problems because of associated phosphorus.
Vaze and Chiew (2004) investigated nutrient loads associated with different particle
size ranges of suspended solids in urban stormwater runoff and solids build-up on
road surfaces. They found that although more than half of the pollutant load was
coarser than 300 µm, less than 15% of the total phosphorus (TP) and total nitrogen
(TN) loads were attached to particle sizes greater than 300 µm. Furthermore, they
54
noted that all the particulate TP and TN were attached to particles between 11 µm
and 150 µm. They also noted that dissolved TP is low compared to particulates and a
large proportion of TN (20-50%) in stormwater is dissolved. Furthermore, Shaver
(1996) noted that less than 10% of particulates in stormwater runoff are in the silt
and clay soil size but they contain over half the phosphorous.
However, contrary to the findings that nutrients are highly correlated with suspended
solids, Goonetilleke et al. (2005) found that TN and TP are not always correlated
with suspended solids. They found that a significant proportion of nitrogen and
phosphorus in surface runoff was present in dissolved form for the catchments they
studied. They further argued that the common management technique based on
treating suspended solids as a primary water quality parameter in urban stormwater
treatment facilities to be ineffective due to these observations.
Taylor et al. (2005) conducted a study to characterise the composition of nitrogen in
urban stormwater from fourteen different urban catchments in Melbourne, Australia.
They found that nitrogen was predominantly dissolved (80%) with ammonia being
the least-abundant form (~11%). They also concluded that while the composition of
nitrogen in the study was broadly consistent with international data, the level of
dissolved inorganic nitrogen was higher than in the international research literature.
This research also confirmed that dissolved nitrogen forms dominate urban runoff,
during both dry and wet weather.
2.5 Conclusions
The above discussion summarises the important conclusions derived from reviewing
research literature relating to the impacts of urbanisation on the water environment.
This includes the conclusions drawn regarding the current state of knowledge on the
hydrologic and water quality impacts of urbanisation, key pollutant processes and
current knowledge related to nutrients as a stormwater pollutant in urban catchments.
Urbanisation dramatically alters both the quality and quantity of urban stormwater
runoff. Water quantity impacts are well recognised in research literature. Increase in
55
impervious area such as buildings, roads and parking lots due to urbanisation lead to
an increase in the peak discharge, volume and rate of stormwater runoff, whilst
decreasing stormwater infiltration. On the other hand, urban stormwater runoff has
been identified as a major non point source of pollution to receiving water bodies.
Urbanisation in an area creates pollutants, which are carried by stormwater to
receiving water bodies, such as rivers and lakes, and deteriorate their quality and
endanger their ecosystems. Two main pollutant processes which affect water quality
are pollutant build-up and wash-off. The pollutants accumulated during dry periods
are washed-off during a storm event.
The main types of pollutants that deteriorate the quality of stormwater are total
suspended solids, organic carbon, nutrients, heavy metals, hydrocarbons, litter and
pathogens. Activities associated with urbanisation create a variety of sources for
these pollutants. The pollutants generated from these sources are transported to the
catchment surface by different pathways. Pollutant sources and pathways have been
clearly identified in research literature. The accumulation and concentration of
pollutants on the catchment surfaces vary with factors such as type of land use,
percentage of impervious surface area, rainfall characteristics, surface
characteristics, seasonal variations, traffic volume and population density.
In particular, nutrients build-up and wash-off processes in urban land uses are
important in the context of urban stormwater quality. The excess nutrients in
stormwater can lead to eutrophication of receiving water bodies and hence create
adverse impacts. This includes depletion of oxygen, increased incidence of fish kills,
decrease in perceived aesthetic value of the water body, deterioration of taste and
odour and water treatment problems. Nutrients build-up and wash-off processes are
complex and the nutrient loads vary widely with parameters such as land use, type of
impervious surfaces and rainfall intensity.
The most important nutrients in relation to water quality are nitrogen and
phosphorus. Phosphorus is mostly attached to particulates and most of the nitrogen
in stormwater is dissolved. There is no doubt that an in-depth knowledge of nutrients
build-up and wash-off processes is crucially important to implement best
56
management practices to safeguard stormwater quality such as the design of
pollutant control structures.
57
Chapter 3 Field Investigation Apparatus
3.1 Background The research project was primarily based on a series of field investigations which
were conducted to collect build-up and wash-off samples from selected road surfaces
in different urban land uses. With the variable nature of the pollutant build-up and
wash-off processes, it was important to select efficient, reliable and convenient
techniques for these investigations. The data generated from the research undertaken
was used to define the nutrient build-up and wash-off processes on urban road
surfaces.
According to the research literature, a number of techniques have been used to
investigate pollutant build-up and wash-off from urban impervious surfaces. Sample
collection for pollutant build-up studies have been carried out by vacuuming,
brushing and sweeping or by using a combination of these techniques (Bris et al.
1999; Vaze and Chiew 2002; Robertson et al. 2003). The selection of a suitable
technique depends on factors such as ease of use, degree of discrimination between
collecting finer and coarser particles and collection efficiency (Bris et al. 1999). For
example, Vaze and Chiew (2002) collected dry pollutant samples from the street
surface they studied by using both brushing and vacuuming. The purpose of brushing
the surface lightly before vacuuming was to release most of the fine pollutants which
are attached to the surface.
Pollutant wash-off studies are commonly based on two investigation methods. These
are the use of natural rainfall wash-off data and artificial rainfall wash-off data
(Herngren et al. 2005; Egodawatta et al. 2006; Goonetilleke et al. 2009). The natural
rainfall wash-off data varies considerably with the rainfall characteristics, such as
rainfall intensity and kinetic energy. It constrains the transferability of research
outcomes to develop fundamental concepts and relationships relating to the pollutant
wash-off process (Herngren et al. 2005). Furthermore, the random nature of
58
occurrence of natural rainfall makes the investigations difficult (Thomas et al. 1989).
Hence, the use of artificial rainfall has become a reliable investigation method in
water quality research as it provides better control of physical factors which limit the
transferability of research outcomes (Herngren et al. 2005; Egodawatta et al. 2006;
Shigaki et al. 2007; Goonetilleke et al. 2009).
This chapter discusses the investigation techniques and apparatus used in the
research project undertaken. The selection of investigation techniques and apparatus,
which can be used to achieve particular research objectives, depends on factors such
as efficiency in operation and portability to the study sites. The selected techniques
and apparatus for field investigation used are as follows:
• build-up sampling using a vacuum collection system
• wash-off sampling using a rainfall simulator
3.2 Vacuum collection system
According to research literature, most of the stormwater pollutants are attached to
the finer fraction of solids (Sartor and Boyd 1972; Deletic and Orr 2005; Bian and
Zhu 2008; Zafra et al. 2008). In this context, the investigation of the finer fraction of
solids has become crucially important in most stormwater quality studies. Hence, the
build-up sampling technique selected was to enhance the finer particle collection
after a careful consideration of techniques which have been used in previous
research studies.
According to Robertson et al. (2003), sweeping is a preferable technique only when
coarser particulates are of interest. However, Bris et al. (1999) identified that wet
vacuuming is considerably efficient in collecting finer solids particles compared to
other methods such as sweeping, brushing or dry vacuuming. They performed a
model experiment on collecting solids deposited on the laboratory floor by using
both dry and wet vacuum collection systems and noted that wet vacuuming is more
efficient in collecting whatever the particle size fraction, as shown in Figure 3.1.
This is due to the high collection and retention efficiency of particles in wet
vacuuming.
59
Figure 3.1- Collection efficiencies of dry and wet vacuum sampling on a same floor for various concentrations of pollutants (Adopted from Bris et al. 1999)
3.2.1 Selection of vacuum cleaner
The selection of an appropriate and well designed vacuum cleaner is vitally
important in pollutant sampling. Both industrial and domestic type vacuum cleaners
have been used in stormwater quality research studies to collect pollutant samples
from roads and other impervious surfaces (Vaze et al. 2000; Tai 1991). Scientific
studies conducted by Ewers et al. (1994) have shown that different vacuum cleaners
remove and retain different amounts of particles even from the same surface. This is
primarily due to factors such as the efficiency of the filtration system and the power
of the system.
The industrial vacuum cleaners are more powerful in collecting particles (Vaze et al.
2000). However, according to Tai (1991), the collection efficiency is not the only
criterion which makes a vacuum system preferable to use. Tai (1991) noted a 96.4%
retaining efficiency for the domestic vacuum cleaner he used. As he noted, higher
retaining efficiency was attributed to the effective filtration system used in domestic
vacuum systems. Therefore, the selection of the vacuum system in the research was
based on the combination of criteria including power and the efficiency of the
filtration system.
60
The vacuum cleaner used for the research was a Delonghi Aqualand model, which
incorporates an efficient filtration system. The model also provides 1500W suction
power with adjustable suction control. The same vacuum system was successfully
used by Herngren (2005) and Egodawatta (2007) to investigate pollutant build-up on
road surfaces. Additionally, other features that resulted in the selection of the
vacuum cleaner were ease of portability and convenient use of the system in the
field.
As most of the primary stormwater pollutants are attached to the finer fraction of
solids (Particles smaller than 150 µm), the ability of the vacuum cleaner to collect
finer particles was an important consideration in this research study. Therefore, the
efficiency of the vacuum cleaner in collecting finer particles from the surfaces was
improved by the arrangement of a small vacuum foot with a brush that was attached
to the end of the hose of the vacuum cleaner. The small foot was able to concentrate
the airflow into a smaller area enabling the use of the system power effectively to
collect both finer and coarser particles adhering to the surface. The attached brush
enhanced the collection efficiency of the vacuum cleaner by dislodging the finer
particles from the surface effectively (Herngren 2005; Egodawatta 2007). Therefore,
the combination of brushing and vacuuming enhanced the collection of
representative samples including both finer and coarser fractions of solids.
The High Efficiency Particulate Air (HEPA) filter in the selected vacuum cleaner
provided efficient filtration with minimal escape of finer particles from the exhaust
system. The water filter system of the vacuum cleaner is illustrated in the Figure 3.2.
As stated in the manufacturer’s specifications, the HEPA filter has 99.97%
efficiency in filtration. The filter mechanism acts to direct the air intake through a
column of water so that the particulate pollutants are retained in the water. The
pollutant sample retained in the water can be easily extracted to sample containers
for further analysis.
61
Figure 3.2- The design of the water filter system of Delonghi Aqualand model
3.2.2 Sampling efficiency
The sampling efficiency of the vacuum cleaner was tested under simulated field
conditions before commencing the field investigation. A 400 mm x 400 mm test plot
from a road surface was selected as the sample surface. Figure 3.3a, 3.3b show the
section of the sample road surface where the test was carried out and a section of the
actual road surface where field investigations were carried out. As shown in Figure
3.3a, 3.3b both road surfaces were asphalt paved with average surface condition
reflecting the similar characteristics of sample road surface to actual road surfaces
investigated.
Figure 3.3a- Section of sample Figure 3.3b- Section of road surface road surface at Industrial site The test was performed by distributing a known weight of graded pollutant sample
on the selected surface and then analysing the collected amount of pollutant after
vacuuming under similar test conditions that apply in the field. A 100 g soil sample
62
was selected as the pollutant sample and analysed for particle size distribution. The
pollutant sample represented a particle size range of 1 to 1000 µm which is the
generally expected particle size distribution of pollutants from road surfaces (Sartor
and Boyd 1972; Ball et al. 1998; Zafra et al. 2008).
The selected sample surface was cleaned properly prior to the test by repeated
vacuuming and washing with deionised water and allowed to dry by applying a flow
of air. Then the graded soil sample was distributed uniformly throughout the test plot
by using a straight edge and a brush without spilling over the edges of the test plot.
The vacuum cleaner compartment, hoses and foot were cleaned thoroughly, using
deionised water to ensure that no pollutants remained from previous use. The filter
compartment was filled with 3 L of deionised water and the surface was vacuumed
three times in a perpendicular direction simulating the field test conditions. After
vacuuming the surface properly, the vacuum cleaner compartment was emptied into
a clean container and washed thoroughly to ensure all the pollutants collected were
transferred into the container. Additionally, all the hoses, foot and brush pieces were
cleaned thoroughly using deionised water and the water was poured into the same
container to further ensure all the particulates were collected.
The collected sample was oven dried and the recovered weight was measured. Table
3.1 shows the particle recovery efficiency for each size range and Figure 3.4 shows
the particle size distribution of the original sample and recovered sample. As evident
in Table 3.1, the total sample recovery efficiency was found to be 95%, which was
considered satisfactory for investigations (Herngren 2005; Egodawatta 2007). The
particle recovery efficiency for each particle size class was greater than 88%. The
loss of particles in each particle size class of recovered sample could be attributed to
the entrapment of particles in the vacuum cleaner compartment and hoses. As
evident in Figure 3.4, a relatively higher loss occurs for particles which are greater
than 300 µm. This confirms the high efficiency of the selected vacuum system in
collecting fine particles.
63
Table 3.1- Sample recovering efficiencies
Sieve class (µm)
Original sample Weight (g)
Vacuumed sample
weight (g)
Recovering efficiency
(%) <75 0.1 0.088 88.0
75-150 0.2 0.175 87.5 150-180 0.3 0.286 95.3 180-210 0.6 0.592 98.7 210-250 1.8 1.744 98.6 250-300 7.2 7.069 98.2 300-425 42.5 41.435 97.5 425-500 22.6 21.570 95.4 500-600 12.7 11.939 94.0 600-850 8.6 8.083 94.0
>850 2.0 1.930 96.5
0
5
10
15
20
25
30
35
40
45
0 200 400 600 800 1000 1200 1400Particle size (µm)
Wei
ght r
etai
ned
(g)
Original sample
Vaccumed sample
Figure 3.4- Comparison of particle size distribution of original sample and recovered sample
3.3 Rainfall simulator
As discussed in Chapter 2, pollutant wash-off is a complex process which varies
with a range of parameters such as rainfall, runoff and catchment characteristics.
Investigation of such a complex process under natural rainfall conditions is difficult
64
due to the high spatial and temporal variability of natural rainfall characteristics such
as rainfall intensity, uniformity of rainfall over the area and kinetic energy. In this
context, as mentioned in Section 3.1, it is worth considering the use of simulated
rainfall events for investigating the pollutant wash-off process. This expedites data
collection under controlled conditions and has the capacity to repeat experiments
over a short time period to produce a large amount of data (Exeter et al. 1990).
Additionally, the use of rainfall simulation has been recognised as a time and cost
efficient method which can be used to overcome the dependency on natural rainfall
(Herngren et al. 2005).
The rainfall simulator has been successfully used as a tool in urban water quality
research to simulate and reproduce rainfall with its natural characteristics
(Tiefenthaler and Schiff 2001; Herngren et al. 2005; Egodawatta 2007; Goonetilleke
et al. 2009). The rainfall simulator used in the research undertaken was designed and
fabricated by Herngren (2005). The main features of the simulator are:
• ease of portability and easy assembly and operation;
• the ability to reproduce rainfall with drop size distribution, kinetic energy and
terminal velocity similar to natural rainfall;
• the ability to create the rainfall intensities required to replicate for the research
undertaken;
• a collection system for sampling runoff from impervious surfaces.
As shown in Figure 3.5, the rainfall simulator consists of an A frame structure made
out of Aluminium tubing of 40 mm diameter. Three Veejet 80100 nozzles are
mounted with equal spacing on a swinging nozzle boom. The motor connected to the
nozzle boom makes it swing in either direction to spray water uniformly. Two catch
trays which control the water return system are connected to two aluminium tubes
that are located under the nozzle boom. The speed of swinging and the delay
between the raindrops are controlled by an electronic control box. The control box
settings are calibrated for different rainfall intensities before use so that it can be
adjusted to simulate rainfall with known intensity. The water was pumped by using
an externally located tank such that the water pressure in the nozzle boom was
adjusted to achieve the required drop size distribution and velocity.
65
Figure 3.5- Rainfall Simulator (Adapted from Herngren et al. 2005) The runoff collection system was designed for a 2 m x 1.5 m plot area which is
connected to a collection trough made from sheet metal ensuring no leakage of
runoff from the plot. The runoff plot area of 2 m x 1.5 m was selected to ensure
uniformity in rainfall generated by the simulator (Herngren et al. 2005).
3.3.1 Calibration of rainfall simulator
The use of the rainfall simulator to replicate natural rainfall events is complex and
involves many parameters such as intensity, drop size and kinetic energy. Lack of
adequate replication of these characteristics of natural rainfall can lead to incorrect
data. Therefore, it was needed to ensure that the rainfall simulator is capable of
replicating these natural rainfall characteristics accurately, prior to use in the field.
66
The experimental set up and procedure adopted to calibrate the simulator for the
required intensities and verified for kinetic energy and drop size distribution is well
documented in Herngren (2005). For this study, the simulator was re-calibrated for
six rainfall intensities adopted for the investigations and verified for drop size
distribution and kinetic energy. The intensities selected are typical to the study
region.
3.3.2 Calibration for rainfall intensity and uniformity of rainfall
The calibration of the rainfall simulator for rainfall intensity and uniformity of
rainfall was carried out in the laboratory. This was done by measuring the average
depth of water collected for a known duration in using an array of containers placed
in a grid pattern exposed to the simulated rainfall. Twenty containers were placed
under the plot area of 2 m x 1.5 m in a grid pattern as shown in Figure 3.6. Different
rainfall intensities were simulated using the control system incorporated with the
simulator.
The control system of the simulator is composed of a control box that includes the
controller circuit board. The control box contains two control knobs, with one to
control the speed of oscillation (demarcated 1 to 5) and the other to control the delay
time (demarcated from A to M). Both of these control knobs were set to a known
setting before starting the operation. Simulated rainfall was generated for a duration
of 5 minutes. Then the amount of water collected in all the containers was measured
and the rainfall intensity in terms of depth of water per unit time (mm/hr) was
calculated. The experiment was repeated for different settings of the control box.
The complete set of data generated is tabulated in Table A in Appendix 1.
It was noted that the simulator was capable of simulating rainfall intensities ranging
from 20 mm/hr to 160 mm/hr. Therefore, the use of the rainfall simulator to replicate
the selected rainfall intensities was considered satisfactory. More details on rainfall
intensities and durations can be found in Section 4.5.
67
Figure 3.6- Calibration of the simulator for intensity The uniformity of rainfall over the area is expressed in terms of the spatial variation
of the rainfall intensity. This is evaluated by the use of a coefficient called a
uniformity coefficient (Cu) (Rickson 2001). The uniformity coefficient was
calculated according to the equation developed by Christiansen (1942) (Equation
3.1). In this context, the collected data during the calibration of the simulator for
rainfall intensity was used.
Cu = 100 (1-∑X) Equation 3.1
Where:
Cu - Coefficient of uniformity (%)
X - Absolute deviation of individual observation from mean value
m - Mean value
n - Number of observations
A high Cu value indicates small deviations from the mean intensity. The more
uniform the rainfall intensity is throughout the plot, the closer the uniformity
coefficient approaches 100%. The uniformity coefficients obtained for the different
rainfall intensities tested in this study were around 70%. This was considered
sufficient for a successful rainfall simulation (Loch et al. 2001; Herngren 2005).
m.n
68
3.3.3 Drop size distribution and kinetic energy of rainfall
Drop size distribution and kinetic energy are two important parameters which
indicate the capability of the rainfall simulator for reproducing the characteristics of
natural rainfall events (Bubenzer et al. 1984; Kinnell 1987; Herngren 2005). The
mass of the raindrop and terminal velocity are greatly influenced by the drop size. As
the drop size varies with rainfall intensity, the kinetic energy of rainfall is in turn
affected by the rainfall intensity (Roswell 1986).
In general, rainfall simulators should be able to generate drop size up to 5 mm in
diameter. The large drops in natural rainfall tend to be unstable, as they oscillate,
vibrate and spin when they are falling. Consequently, they break up into smaller
drops when reaching the ground (Rickson 2001). The median drop size (D50) is a
parameter which can be used to compare the different rainfall events conveniently.
This is defined as the drop size where 50% of drops generated in the storm event are
larger and 50% are smaller.
According to Carter et al. (1974) the median drop size of rainfall increases when the
rainfall intensity increases up to 63.5 mm/hr. They noted that there is a decrease in
median drop size for further increase in the intensity. The decrease in median drop
size after the 63.5 mm/hr intensity was attributed to the breaking of raindrops due to
their instability after reaching to a size of approximately 6 mm in diameter.
However, even though D50 varies throughout a natural storm event as intensity and
thus drop size varies, for simulated work it can be assumed that D50 is constant as the
intensity is constant for a particular rainfall event (Rickson 2001).
The rainfall simulator used in this research project was originally designed to
simulate a median drop size of 2.1 mm and kinetic energy of 25.44 J/m2/mm where
rainfall intensities are greater than 40 mm/hr. As recommended by Herngren (2005),
the pressure at the nozzle boom should be adjusted to 41 kPa in order to maintain the
drop size and kinetic energy for rainfall intensities greater than 40 mm/hr. These
values are in close agreement with the drop size and kinetic energy calculated by
several other researchers for natural rainfall at intensities greater than 40 mm/hr
(Hudson 1963; Roswell 1986; Loch et al. 2001).
69
According to the mechanical arrangement of the simulator, different rainfall
intensities are achieved by varying the nozzle boom movement but not the simulator
hydraulics such as height of fall, pump and nozzle specifications. Therefore, it was
expected to have the same kinetic energy per unit depth of rain for all the rainfall
intensities. Consequently, a verification test was performed to check the capability of
the simulator in producing the initially calibrated drop size and kinetic energy under
a pressure of 41 kPa.
Two most widely used methods for measurement of drop size are the stain method
and flour pellet method (Bubenzer et al. 1984; Erpul et al. 1998; Herngren 2005). In
the stain method, drops are allowed to fall on a uniform absorbent surface such as
blotting paper and then the diameter of the stain which is proportional to the
diameter of the drop is measured. In the flour pellet method, drops are allowed to fall
into a pan of sifted flour and the resulting flour pellets are oven dried and
subsequently sized by passing them through a variously graded set of sieves. In both
methods, the drop size is obtained by comparing the size of the stains or pellets with
those produced by the drops of known diameter (Hudson 1963; Bubenzer et al. 1984;
Assouline et al. 1997; Herngren 2005). The method adopted to measure the drop size
distribution in this research project was the flour pellet method developed by Hudson
(1963). Selection of this method over the stain method was primarily due to the easy
preparation of experimental setup and less technical difficulties to be incurred
compared to stain method.
A tray with an uncompacted layer of flour was exposed to simulated rainfall for a
few seconds, allowing a number of pellets to be formed. Then the flour was oven
dried for 12 hrs at 105 0C and the pellets formed were passed through a set of sieves
and separated into different particle size ranges as shown in Figure 3.7. The size
ranges selected were:
1) <4.75 mm
2) 4.75 mm- 3.35 mm
3) 3.35 mm- 2.36 mm
4) 2.36 mm -1.68 mm
5) 1.68 mm-1.18 mm
6) 1.18 mm - 0.85 mm
70
7) >0.85 mm
Figure 3.7- Pellets separated into each size ranges
The average weight of pellets in each size range was determined by dividing the
weight by the number of pellets in each size class. The next step was to calculate the
drop mass which had formed the pellet. Even though Hudson (1963) provides a
calibration curve indicating the relationship between pellet mass vs the ratio of drop
mass with pellet mass, the direct use of that curve in this study had to be validated.
The main reasons for that were the difference in the type of flour used, degree of
compaction and drop falling height compared to the study by Hudson (1963).
Therefore, a pilot experiment was conducted to determine the possibility of using the
curve provided by Hudson (1963) to determine the drop mass of the simulated rain
event.
The experimental set up used for the pilot experiment was similar to that used by
Egodawatta (2007). A medical needle with a known diameter was connected to a
large reservoir which had been placed at a 3 m height as shown in Figure 3.8 and a
known number of drops was collected to a pre-weighed beaker in order to calculate
an average drop weight. The beaker was lined with cotton wool to avoid splashing
and evaporation.
71
Figure 3.8- Experimental setup for drop size calibration A known number of flour pellets were made by replacing the beaker with the tray
containing uncompacted flour and oven dried. Then the separated flour pellets were
weighed to determine the average weight. The experiment was repeated for ten
different sizes of needles. The pellet mass and the ratio of drop mass to pellet mass
were plotted as shown in Figure 3.9. As it was not possible to make smaller size
pellets with available needles in the laboratory, the pilot experiment was able to
verify only a range within the calibration curve. However, as shown in Figure 3.9,
the results obtained from the pilot experiment were in close agreement with the
calibration curve developed by Hudson (1963). Therefore, it was decided to use the
Hudson’s curve to convert the pellet mass to rain drop mass.
Reservoir
Needle
Cotton wool
Collection beaker
3m
72
0.5
0.7
0.9
1.1
1.3
1.5
0.1 1 10 100Pellet Mass (mg)
Mas
s R
atio
(Dro
p M
ass/
Pel
let M
ass)
Hudson, (1963)
Experiment
Figure 3.9- Calibration curve for flour pellets
The drop mass was then converted to the drop diameter and D50 was calculated. The
calculated D50 was 2.45 mm. According to Hudson (1963), the D50 for a natural
rainfall event is 2 mm to 2.5 mm. Therefore, the calculated drop size was accepted
for the research undertaken. Table B in Appendix 1 shows the calculation of drop
size in detail.
In order to calculate the kinetic energy, it was necessary to find the terminal velocity
of each drop in each class. Terminal velocity depends on the drop size with the
larger drops having higher terminal velocities than smaller drops (Laws 1941).
According to Herngren et al. (2005), the simulator height of 2.4 m was adequate for
creating terminal velocities similar to natural rainfall for all drop sizes. Therefore, it
was assumed that all the drops have reached terminal velocity. The terminal velocity
was obtained for each drop diameter based on Laws (1941) data. The kinetic energy
of rainfall was calculated by the sum of the kinetic energy of the individual drops
which is 25.63 J/m2/mm. The simulated kinetic energy was in close agreement with
the kinetic energy obtained by Herngren et al. (2005) for the same simulator. The
values obtained for drop size and kinetic energy in this study were considered to be a
satisfactory replication of natural rainfall kinetic energy.
73
As explained in Chapter 4, it was necessary to simulate 20 mm/hr rainfall in order to
cover the natural rainfall intensity range in the study area. According to Roswell
(1986), the kinetic energy for 20 mm/hr rainfall intensity is in the range of 16 to 18
J/m2/mm. However, as discussed above, the simulator was able to simulate a
constant kinetic energy of 25.63 J/m2/mm. Therefore, in order to simulate 20 mm/hr
rainfall intensity satisfactorily, it was necessary to reduce the kinetic energy
accordingly. Egodawatta (2007) incorporated an energy dissipater to reduce the
kinetic energy. According to that method, a fly screen mesh frame is placed just
below the nozzles in order to reduce the drop size and hence the kinetic energy.
Egodawatta (2007) confirmed the suitability of this method in reducing the kinetic
energy to simulate 20 mm/hr rainfall intensity by re-calibrating the same simulator
used in this study.
3.4 Summary Two main apparatus used for field investigations in the research project were;
• Vacuum collection system
• Rainfall simulator
The vacuum collection system was used to collect pollutant build-up and wash-off
samples from impervious surfaces. The selection of the vacuum system was
primarily based on the collection efficiency of fine particles from the road surface.
The system was tested for particle retention efficiency and was found to be
satisfactory for use in sampling.
A rainfall simulator was used to create wash-off data from impervious surfaces. The
rainfall simulator was calibrated for six rainfall intensities and verified for drop size
and kinetic energy. In order to reduce the kinetic energy of the rainfall for low
intensities, a kinetic energy dissipater was introduced.
74
75
Chapter 4 Study Site Selection and Sample Collection
4.1 Background Field investigations were focused on collecting pollutant build-up and wash-off data
from impervious surfaces. As discussed in Section 2.2.2C road surfaces have been
identified as among the impervious surfaces which exert a strong influence on
stormwater quality. Three road surfaces were selected to collect pollutant build-up
and wash-off samples. It was hypothesised that the processes of pollutant build-up
and wash-off is independent of land use. In order to confirm this hypothesis, the road
surfaces were selected from three different urban land uses; residential, industrial
and commercial from the selected study area.
Furthermore, in the context of understanding the fundamental pollutant processes,
the use of small plots has been identified as the more suitable approach, rather than
catchment scale studies. Use of small plots will ensure the homogeneity of collection
of build-up and wash-off samples from the surfaces, reduce a number of variables
and lessen the location specific nature of outcomes inherent in catchment scale
studies (Egodawatta and Goonetilleke 2008; Goonetilleke et al. 2009). Therefore,
both build-up and wash-off sample collection was confined to small plot areas which
enabled the acquisition of more reliable and detailed knowledge on nutrient build-up
and wash-off processes.
The focus of this chapter is to present a detailed description of the study sites and the
site selection procedures. In addition, it describes the sample collection procedures
used for build-up and wash-off sampling. Moreover, the chapter also includes the
methods adopted for transport and treatment of the samples collected from the study
sites.
76
4.2 Study area The field investigations were undertaken in the Gold Coast area. The Gold Coast is
located just south of Brisbane, which is the capital city of Queensland, Australia.
The Gold Coast is Australia’s sixth largest city. The current population in the city is
approximately 500,000 and expected to increase to over 600,000 by the year 2021
(GCCC 2008). The Gold Coast encompasses a network of water bodies, which is
made up of five main rivers, numerous creeks, many of which connect to lakes and
canals. The city’s major drainage basins are shown in Figure 4.1.
Figure 4.1- Map of Gold Coast Region (Adapted from GCCC 2008)
High density urban areas are located on the east coast of the region providing
waterside living for residents. However, rapid urban development closer to water
bodies has influenced a change in quality of water due to a variety of pollutants
generated from different urban activities (GCCC 2008). Consequently, a number of
comprehensive urban catchment monitoring programs are already being undertaken
77
in the area in order to provide successful stormwater quality mitigation measures.
The study sites were selected after careful consideration of the suitability of the
roads to carryout the rainfall simulation.
4.3 Study site selection
The sites considered were in Nerang, which is a highly urbanised suburb in Gold
Coast. As discussed in Section 4.1, three road surfaces were selected for field
investigations from residential, industrial and commercial land uses. The study sites
were selected to cause minimum disturbance to residents and traffic. Additionally,
the following criteria were considered in site selection:
• Fair to good surface condition;
• Convenient access to the site and space for the use of a rainfall simulator;
• Sufficient slope for gravity flow of runoff;
• For industrial sites, there needed to be a mix of light industrial activities; and
• In commercial sites typical commercial activities needed to be represented.
Following is a description of the selected study sites.
Residential site A number of possible research sites in a typical suburban residential area were
considered prior to the selection of the study area. The descriptions of possible sites
considered are listed in Table 4.1.
78
Table 4.1- Description of possible residential sites Name of the site Characteristics
Bernadette Place
Access road, variable slope, poor
surface condition, number of bends,
uniform build-up cannot be assumed
Issel Place Close to highway, good surface
condition, slope not sufficient for the
gravity flow of runoff
Leesa Court
Steep slope, surface condition is not
uniform, construction and demolition
activities
Kilmuir Street
Steep to mild slope, good surface
condition, space for simulation poor
Renfrew Drive
Good surface slope for runoff, space for
simulation is good, high traffic flows,
good surface condition
Armstrong Way
Access road, good surface condition,
good slope for gravity flow of runoff,
space for simulations
After a careful investigation of the characteristics of each road, two possible research
sites were selected (Renfrew Drive, Armstrong Way). Both ends of Renfrew Drive
are connected to a main road and three access roads are connected in-between the
ends. Therefore, Renfrew Drive was a much busier road compared to Armstrong
Way. Consequently, Armstrong Way (Figure 4.2) was selected for field
investigations, considering the minimum disturbance to residents and traffic.
The surrounding area of Armstrong Way included detached family houses with small
gardens. The road is used by residents for access and the surface condition was
considered satisfactory. The gardens were well maintained which could have been
the result of the frequent use of garden fertilisers. Consequently, these could also be
expected to contribute nutrients to the road surface (Ellis 1989; Pitt et al. 2004).
79
Figure 4.2- Residential research site (Armstrong Way) Industrial site Similar to the procedure for selecting the residential site, several possible sites were
considered prior to the selection of the industrial site. The identified research sites
and their characteristics are shown in Table 4.2.
According to the criteria developed for the site selection procedure, two possible
research sites (Indy Street, Stevens Street) were identified. The width of Stevens
Street was approximately 8 m and therefore the fieldwork could be carried out by
closing one lane of the road easily without disturbing traffic. This was in addition to
the fact that this site met the pre-determined criteria. Therefore, Stevens Street
(Figure 4.3) was selected for the investigations.
80
Table 4.2- Description of possible industrial sites Name of site Characteristics
Indy Street
Tyre, mattresses and steel industries,
vehicle service station, good surface
slope, good surface condition, high
traffic Jaygee Court
Tile, soap and paint industries, high
traffic flow, slope is not good for
simulations
Patrick Road
Timber and truss industries, dust from
timber industry high, disturbance to
traffic, poor space for simulations
Palings Court
Furniture and welding industries, close
to highway, medium slope, two large
bends, high traffic flow
Stevens Street
Paint, furniture, welding and cement
mixing industries, steep slope, large
space for simulations, large road width,
disturbance to traffic minimal
Hope Street
Electrical appliances industry, space and
slope is not good for simulations
Figure 4.3- Industrial research site (Stevens Street)
81
The Stevens Street site was used as an access road to the industrial enterprises in the
area. The road surface was in a poorer condition when compared to the road in the
residential land use area. It was degraded and had been subjected to oil leakages and
spills due to the regular movement of heavy vehicles. The surface was relatively
coarse textured, suggesting that large numbers of particles could be embedded within
the voids.
Commercial site Similar to the selection of residential and industrial road sites several possible
research sites were considered prior to the selection of the commercial site. Table 4.3
lists the road sites considered and their characteristics.
Lawrence Drive was selected for field investigations (Figure 4.4). The primary
reason for selecting Lawrence Drive over others was that there would be minimal
disturbance to traffic when conducting the field work. Lawrence Drive was located
very close to a motorway and one end is connected to a busy highway. This
suggested that the pollutants generated from vehicular activities could significantly
contribute to the pollutant concentration at the site due to the effect of wind and
frequent vehicle movement (Brinkmann et al. 1985; Han et al. 2006). Additionally,
due to relatively higher traffic volume in Lawrence Drive compared to the selected
residential and industrial area roads, the field investigation in this site was conducted
at night. This ensured minimum disturbance to traffic.
82
Table 4.3- Description of possible commercial sites
Name of road surface
Characteristics
Recreation Drive
Surface condition is not uniform, light
traffic, space sufficient for simulations,
convenience shops
Lawrence Drive
Good surface condition, vehicle service
station, car parking area, motorcycle
sales centre, food stores, slope and
space is good for simulations, high
traffic flow
Spencer Road
Fair slope, mechanical workshops, food
stores, good surface condition, high
traffic flow
O’Shea Drive
Good surface condition, space for simulation is not sufficient
Figure 4.4- Commercial research site (Lawrence Drive)
83
4.4 Collection of pollutant build-up samples Pollutant build-up is the accumulation of dry pollutants on surfaces. According to
several researchers, pollutant build-up varies with a range of parameters such as land
use, antecedent dry period and traffic characteristics (Sartor and Boyd 1972;
Bradford 1977; Pitt 1979; Ball et al. 1998). As the selected study sites belonged to
different land uses, traffic characteristics were not uniform. Consequently, a
considerable variation in pollutant build-up characteristics could be expected on the
selected road surfaces. However, Egodwatta and Goonetilleke (2006) noted that the
build-up load approaches a constant value with the increase in antecedent dry days
and that after about seven days it asymptotes to an almost constant value.
Consequently, a minimum of seven days of antecedent dry period was allowed prior
to sample collection. The antecedent dry period allowed for each study site is shown
in Table 4.4.
Table 4.4- Number of antecedent dry days Study site
Number of antecedent dry days
Residential road- Renfrew Drive
8
Industrial road- Stevens Street
9
Commercial road- Lawrence Drive
11
Build-up samples were collected from all the selected surfaces by using the vacuum
cleaner described in Section 3.2. According to past researchers, the composition of
build-up can vary across the road surface (Deletic and Orr 2005; Zafra et al. 2008).
Therefore, a test plot of size 2 m x 1.5 m was selected from each road surface at the
middle of one side of the road in order to maintain the consistency of the build-up
sampling. The boundary of the test plot was demarcated by using a wooden frame as
shown in Figure 4.5. Prior to use, the vacuum cleaner and all the connected
components were thoroughly cleaned. 3 L of deionised water was added to the
vacuum cleaner compartment as the filtration agent.
84
Figure 4.5- Collection of build-up sample The selected study plot was vacuumed three times in perpendicular directions and
build-up was collected into the filtration compartment. After vacuuming, the
collected solids were transferred to a polyethylene container which was pre washed
according to standard methods (APHA 2005). The vacuum cleaner compartment and
hoses were thoroughly washed using deionised water ensuring that all the particles
collected were added to the container.
4.5 Collection of pollutant wash-off Pollutant wash-off samples were collected from the same road surfaces where
pollutant build-up samples were collected. Pollutant wash-off samples were
collected for six rainfall intensities for different durations (Table 4.5) in five minute
time steps using the rainfall simulator as discussed in Section 3.3. These intensities
were identified as typical rainfall intensities in the region based on long-term records
of regional rainfall data for the Gold Coast area. The durations of the rain events
were selected based on the 1 year to 10 year average recurrence intervals (ARI).
85
Table 4.5- Rainfall intensities and durations simulated during the study
(Adapted from Egodawatta 2007)
Rainfall Duration (min) Rainfall Intensity
(mm/hr) Event 1 Event 2 Event 3 Event 4
20
40
65
86
115
135
10
10
10
10
5
5
20
15
15
15
10
10
30
25
20
20
15
15
40
35
30
25
20
20
Wash-off samples were also collected from a 3 m2 plot area. As shown in Figure 4.6,
the plot area was enclosed by a plastic frame. A rubber flap was attached to the
inward side of the frame and sealed to the road surface by gutter tape and waterproof
sealant, thus ensuring that no water would enter or leave the demarcated plot.
Figure 4.6- Wash-off collection test plot The wash-off collected in the trough was vacuumed throughout the design rainfall
event using the vacuum system especially designed for picking up fine particles as
Gutter tape Collection trough
Plot boundary
86
discussed in Section 3.3. The wash-off samples were directed into clean 25 L
polyethylene containers along with the vacuuming. The arrangement of the rainfall
simulator in the field is shown in Figure 4.7. Figure 4.8 shows the collection of
wash-off into polyethylene containers.
Figure 4.7- Arrangement of rainfall simulator in the field
Figure 4.8- Collection of wash-off samples into polyethylene containers There was a minor reduction in runoff volume compared to the simulated rain
volume. The reduction in volume was primarily attributed to losses such as
87
infiltration, errors in simulating rainfall intensities and leakage from the plot area.
Therefore, a great effort was made to reduce these losses by sealing the plot area
carefully, continuous monitoring of intensities and continuous vacuuming of the
wash-off into sample containers.
4.6 Sample handling
A complete record of all the samples collected in the field was maintained
throughout the sampling period. Each wash-off sample was properly labelled by
representing the rainfall intensity and time duration at the time of collection (Figure
4.9). Furthermore, deionised water blanks and field water blanks were also included
to maintain standard quality control procedures as specified in Australia / New
Zealand Standards, Water Quality - Sampling (AS/NZS 5667.1: 1998). The collected
samples were transported to the QUT laboratory within the same day. Resampling
was performed in the laboratory as early as possible and the samples were preserved
and refrigerated under 40C as specified in Standard Methods for the Examination of
Water and Waste Water (APHA 2005) for the analysis of physico-chemical
parameters.
Figure 4.9- Labelling of the samples collected in the filed
4.7 Summary The field investigations were undertaken in the Gold Coast region. Three road
surfaces from three different urban land uses, namely, residential, industrial and
88
commercial were selected for field investigations which focused on collecting
pollutant build-up and wash-off samples.
Build-up samples were collected using a modified vacuum cleaner. A minimum
antecedent dry period of seven days was ensured prior to the sampling. Wash-off
samples were collected for six simulated rainfall intensities for different durations at
the same road surface where build-up investigations were conducted. Special
attention was given to collecting the wash-off samples such that almost all the runoff
generated from the rain event was collected. Collected samples were transported to
the laboratory, resampled and preserved under standard methods for further testing.
89
Chapter 5 Analytical Procedures
5.1 Background
As discussed in Chapter 2, nutrients that play the most vital role in the degradation
of water quality are nitrogen and phosphorous (Carpenter et al. 1998; Bian and Zhu
2008; Drewry et al. 2009). The analytical procedures undertaken focused on the
different species of nitrogen and phosphorous in pollutant build-up and wash-off
samples. Furthermore, primary physico-chemical parameters such as pH, EC,
particle size distribution, total suspended solids, total dissolved solids and organic
carbon content were also investigated in order to understand the influence of these
parameters on nutrient build-up and wash-off processes (Vaze and Chiew 2004;
Mesner and Geiger 2005; Brilly et al. 2006). The complete set of parameters which
were investigated included:
1) pH/ EC;
2) Particle size distribution;
3) Total suspended solids (TSS), Total dissolved solids (TDS);
4) Total organic carbon (TOC), Dissolved organic carbon (DOC);
5) Nitrite-nitrogen (NO2-), Nitrate-nitrogen (NO3
-), Total kjeldahl nitrogen (TKN)
and Total nitrogen (TN); and
6) Phosphates (PO4-3), Total phosphorus (TP).
A series of laboratory experiments was conducted to measure these parameters
accurately. This chapter discusses the laboratory analytical methods used.
Additionally, it outlines the data analysis techniques adopted to analyse the raw data
obtained.
90
5.2 Sub Sampling Prior to the laboratory analysis of physico-chemcial parameters described in Section
5.1, it was needed to obtain representative samples of all the collected build-up and
wash-off samples. This was achieved by an extensive sub sampling procedure. Prior
to sub sampling, the weight of build-up and wash-off samples was determined.
Consequently, the volume of each sample was calculated. The volume of the build-
up and wash-off samples were used to determine the pollutant loads when required
for the data analysis discussed in Chapter 6 and Chapter 7.
The sub sampling procedure was conducted primarily in three stages. Firstly, a 4 L
representative sample from each of the build-up and wash-off samples was prepared.
Prior to this, the original sample was well stirred to ensure the homogeneity of the
sub sample collected. Secondly, from each 4 L sample, 2 L were collected in two 1 L
sample bottles. One 1 L sample was used for the analysis of pH, EC and particle size
distribution and the other 1 L was used for the analysis of the remaining physico-
chemical parameters described in Section 5.1. Furthermore, as noted by past
researchers, the amount of pollutants in build-up and wash-off varies considerably
with the particle size range of solids to which they are attached (Vaze and Chiew
2004; Herngren et al. 2005; Goonetilleke et al. 2009). Therefore, the investigation of
nutrients and influential physico-chemical parameters for different particle size
ranges in the collected build-up and wash-off samples was an important aspect of the
research undertaken.
The remaining 2 L was separated into five particles size ranges in order to analyse
the physico-chemical parameters for individual size range of solids. This was done in
two steps. Firstly, the original sample was separated into four particle size ranges
(>300 µm, 300 µm -150 µm, 150 µm -75 µm, <75 µm) by wet sieving (Figure 5.1).
The remaining pollutants on each sieve was washed with deionised water and diluted
to 2 L using deionised water in order to maintain the consistency of the
concentration of parameters measured in total samples and wet sieved samples.
Following wet sieving, a 1 L representative sample from each wet sieved samples
was prepared for the analysis of physico-chemical parameters. Secondly, as it was
91
difficult to obtain the dissolved fraction of pollutants from the wet sieving, 1 L of
wet sieved sample passing through the 75 µm sieve was filtered using a 1 µm glass
fiber filter paper to obtain the dissolved portion (<1 µm) of the sample. Finally, the
different particle size ranges and the dissolved fraction were subjected to physio-
chemical analysis.
All wet sieved samples were subjected to the physico-chemical parameters listed in
Section 5.1, except for pH, EC and particle size distribution. The selection of these
particle size ranges is supported by the findings of past researchers who identified
the critical particle size ranges of solids in relation to most of the stormwater
pollutants (Sartor and Boyd 1972; Herngren et al. 2005; Jartun et al. 2008; Zafra et
al. 2008).
Figure 5.1-Wet sieving of samples
At the end of the sub sampling programme, the prepared samples were primarily
divided into two categories as follows:
• Total of build-up and wash-off samples (samples prepared without wet sieving)
• Wet sieved build-up and wash-off samples (including filtrate)
92
5.3 Laboratory Testing Each sample was analysed for the relevant physico- chemical parameters under
standard laboratory conditions and according to standard methods as shown in Table
5.1.
Table 5.1-Test parameters and methods used Parameter Test Method
pH Measured with combined pH/EC meter according to method 4500H (APHA 2005)
Electrical conductivity (EC)
Measured with combine pH/EC meter according to method 2520B (APHA 2005)
Total suspended solids (TSS)
Total dissolved solids (TDS)
Method 2540C (APHA 2005)
Method 2540D (APHA 2005)
Total organic carbon (TOC) Dissolved organic carbon (DOC)
Measured using Shimadzu TOC-5000A Total Organic Carbon Analyzer according to the 5310C (APHA 2005)
Particle size distribution
Using a Malvern Mastersizer S Particle Size Analyzer
Nitrite nitrogen (NO2-) SmartChem 140 Discreet Analyser according to the
method 4500 -NO2- -B (APHA 2005)
Nitrate nitrogen (NO3-) Measured using SmartChem 140 Discreet Analyser
according to the method 4500 –NO3- -F (APHA
2005)
Total kjeldahl nitrogen (TKN)
Measured using SEAL Discrete Analyser according to method 351.2 (US EPA 1993)
Total nitrogen (TN) Addition of NO2-, NO3
- and TKN
Phosphate (PO43-) Measured using SEAL Discrete Analyser according
to method 4500-P-F (APHA 2005)
Total phosphorus (TP) Measured using SEAL Discrete Analyser according to method 365.4 (US EPA 1983)
Furthermore, in order to ensure the accuracy of the test data, standard quality control
procedures were followed according to methods specified in Australia / New
93
Zealand Standards, Water Quality - Sampling (AS/NZS 5667.1: 1998). The samples
tested for quality control purposes were included as laboratory blanks, field blanks
and solutions of known concentrations of the analyte. Figure 5.2 shows the analytical
schemata of the build-up and wash-off sub samples prepared. The complete set of
test results obtained for the physico-chemical parameters measured in build-up and
wash-off samples are given in Table A, Table B and Table C in Appendix 2.
94
Figure 5.2- Analytical schemata of build-up and wash-off samples
Sub sampled build-up and wash-off samples 4L
Sample 2L
Total Sample-2 1L
Particle size distribution
pH and EC (Only in wash-off samples)
Wet Sieving
150 µm -300 µm >300 µm 75 µm -150 µm Smaller than 75 µm
Filtering 250mL (through 1 µm glass fibber filter paper
Residue Filtrate Total sample
• TSS
• TDS • DOC • nutrient
parameters
• TOC • nutrient
parameters
Filtering 250mL (through 1 µm glass fibber filter paper
Filtering 250mL (through 1 µm glass fibber filter paper
Filtering 250mL (through 1 µm glass fibber filter paper
Filtering 250mL (through 1 µm glass fibber filter paper
Residue Total sample
• TSS • TSS • TOC • Nutrient
parameters
Residue Total sample
• TSS • TSS • TOC • Nutrient
parameters
Residue Total sample
• TSS • TSS • TOC • Nutrient
parameters
Residue Filtrate Total sample
• TSS • TSS • TOC • Nutrient
parameters
• TDS • DOC • Nutrient
parameters
Total Sample-1 1L
95
5.3.1 Particle size distribution
The particle size distribution of solids is an important characteristic as it determines
the mobility and the concentration of associated pollutants (Hvitved- Jacobsen et al.
1994; Sakai et al. 1996; Grant et al. 2003; Vaze and Chiew 2004). The particle size
distribution of build-up and wash-off samples was measured using a Malvern
Mastersizer S instrument (See Figure 5.3).
This instrument consists of a sample dispersion unit connected by two flow cells to
the optical unit. The Malvern Mastersizer S uses a laser beam to record the scatter
pattern from a field of particles assuming that the particles are spherical. An
analytical procedure is then used to determine the size and distribution of particles
that created the scatter pattern. The instrument uses specialised software supplied by
the manufacturer to analyse results obtained from the optical unit.
The analyser which was used in the research study had a reverse fourier lens of 300
mm diameter. It was able to analyse particles in the range of 0.05-900 µm. The
specified reading accuracy for this range was ±2% of the volume median diameter
(Malvern- Instrument- Ltd 1997). The particle size distribution, which is interpreted
by the instrument is volume based. The instrument analyses the volume of the
particles initially and the particle size is then determined by equating that volume to
an equivalent sphere. This technique has been widely applied in water quality
research studies to determine the particle size distribution in build-up and wash-off
solids (Grant et al. 2003; Egodawatta 2007; Badin et al. 2008; Goonetilleke et al.
2009). More details of the Malvern Mastersizer S can be found in the instrument
instruction manual (Malvern- Instrument- Ltd 1997).
96
Figure 5.3- Malvern Mastersizer S
All the samples were well mixed before inserting into the machine by gently rolling
the sample containers. A sample of water which was prepared for rainfall simulation
in the field (field blank) was taken as the blank to obtain the background
measurement. The analysis of wash-off and build-up samples was then carried out by
measuring the scatter pattern of each sample and then comparing it to the
background profile generated by the blank.
5.3.2 pH /EC pH is a measure of the acidity or alkalinity of the water. Changes to pH can cause a
range of potential water quality problems. One of the most significant impacts of pH
is the effect that it has on the solubility and thus the bioavailability of chemical
constituents such as nutrients, heavy metals and hydrocarbons. Extremes of pH can
be toxic to aquatic organisms. For example, ammonia is relatively harmless to fish in
water that is neutral or acidic but if the pH increases, ammonia becomes increasingly
toxic (Mesner and Geiger 2005).
Electrical conductivity (EC) is a measure of the amount of dissolved salts in the
water and therefore an indicator of salinity. EC value can be used as an indicator of
the overall ionic content or concentration of dissolved salts (Yusop et al. 2006).
97
As shown in Figure 5.2, pH and EC were measured only in wash-off samples using
the combined pH/EC meter. It was calibrated using fresh buffer solutions and
standard salinity solution prior to use. pH and EC were measured in all the samples
immediately after they reached the laboratory. As shown in Table 5.1, the test
methods used were 4500H and 2520B in the Standards Methods for the Examination
of Water and Waste Water (APHA 2005).
5.3.3 Total suspended solids (TSS), Total dissolved solids (TDS)
TSS and TDS represent the concentration of solids in suspended and dissolved forms
respectively. TSS and TDS are reliable indicators of the stormwater quality as most
of the primary stormwater pollutants are attached to solids (Randall et al. 1998;
Sansalone and Glenn 2000; Herngren et al. 2005; Zafra et al. 2008). Therefore,
measurement of TSS and TDS was imperative to fully investigate nutrients build-up
and wash-off processes.
As shown in Figure 5.1, the concentration of TSS was measured in all of the total
samples and wet sieved samples. The analysis was conducted by filtering a known
volume of sample through a weighted 1µm glass fibre filter paper and measuring the
weight of the residue retained on it. Filter papers were pre-washed by using
deionised water and oven dried before filtering. Samples were well mixed and a 250
mL representative portion from each sample was filtered through the pre-weighed
filter papers and the residue retained on the filter paper was oven dried at 1030C -
1050C. Finally, the increase in weight of the filter paper was determined to obtain the
weight of TSS in the volume filtered.
TDS was analysed by measuring the dry weight of the filtrate. A known volume of
filtrate was poured into a clean pre-weighted petri dish and oven dried. The volume
was selected so as to achieve a noticeable increase in weight of the petri dish. The
test methods used were 2540C and 2540D in the Standards Methods for Water and
Waste Water (APHA 2005).
98
5.3.4 Total organic carbon (TOC), Dissolved organic carbon (DOC) Organic carbon in water is important because of its interaction with other pollutants.
As noted by several researchers, a large fraction of nitrogen and phosphorus is
contained within the organic fraction (Hendrickson 2007; Graves et al. 2004).
Therefore, the analysis of organic carbon content in the collected build-up and wash-
off samples was essential.
Total organic carbon (TOC) and dissolved organic carbon (DOC) are good indicators
of organic pollutants present in water (Tao and Lin 2000; Herngren et al. 2005;
Kayhanian 2007). TOC was measured in all the build-up and wash-off samples
according to test method 5310C (APHA 2005) (See Table 5.1). The organic carbon
measured in the filtrates represented DOC (See Figure 5.1).
Shimadzu TOC-VCSH Total Organic Carbon Analyzer (Figure 5.4) was used to
measure the total and dissolved organic carbon content in build-up and wash-off
samples. TOC-VCSH unit measures TOC using a automatic sample injection system
for a extremely wide range from 4 µg/L to 25,000 mg/L. The instrument is
programmed using manufacturer designed software. It includes a program to
automatically conduct the blank check by creating and analyzing ultra pure water
inside the system. High-concentration samples are analysed by diluting to 25,000
mg/L with the built-in automatic dilution function.
Figure 5.4- Shimadzu TOC- VCSH Total Organic Carbon Analyzer
99
5.3.5 Nutrients Table 5.1 shows the test methods used for the testing of nitrogen and phosphorus
parameters discussed in Section 5.1. As shown in Table 5.1, the main laboratory
instruments which were used for nutrient analysis included SmartChem 140 Discrete
Analyser, Seal Discrete Analyser and HACH Spectrophotometer. All these
instruments are colorimetric instruments. Additionally, a block digestion system was
used for digestion of samples as outlined in the test procedures. Following is a short
description about the instruments used for testing of nutrients.
SmartChem 140/ Seal Discrete Analyser
Both these instruments are colorimetric, highly automated and works with minimal
operator interaction. They are computer controlled and operate on manufacturer
designed software. The main components of each instrument are sample probe, high
performance wash station, reaction cuvette system and sample and reagent position
trays. The high performance wash station washes the sample probe at every liquid
contact to prevent cross contamination. Once the samples and required reagents are
loaded to the instrument, the test sample is prepared inside the reaction cuvettes
automatically. The instrument measures absorbances directly in the reaction
cuvettes. This eliminates the need to transfer reaction mixtures to a common flow
cell and thereby prevent any possibility of contamination.
A number of test methods can be conducted in a single run by both instruments. The
calibration of each test method can be inspected separately while the instrument is
running. When the concentration of the analyte in the sample is above the method
detection range, the samples are diluted automatically. Furthermore, sample blanking
and quality control can be included according to the preference of the user. More
details on these instruments can be found in instrument manuals (Westoc Scientific
Instruments, Inc; AQ2 Discrete Analyser operator manual 2006). Figure 5.5a, 5.5b
show the SmartChem 140 and Seal Discrete Analyser respectively.
100
Figure 5.5a- SmartChem 140
Figure 5.5b- Seal Discrete Analyser
HACH DR/4000 Spectrophotometer
HACH DR/4000 spectrophotometer is a direct reading instrument that is
programmed with calibrations for a variety of tests (see Figure 5.6). The instrument
model DR/4000U that was used employs both ultra violet and visible wavelengths.
The instrument provides digital readouts in direct concentration units, absorbance or
percent transmittance. The relevant HACH programmed method is selected
101
according to the parameter to be tested. When the HACH programmed method is
selected, the on screen menus and prompts appear on the screen and directs through
the test.
The HACH DR/4000U is equipped with two modules namely single cell module and
carousel module. More details about the instrument can be found in the DR/4000
spectrophotometer instrument manual (DR/4000 spectrophotometer instrument
manual 1999).
Figure 5.6- DR 4000 spectrophotometer
Block digester
A block digester was used for the digestion of samples for TKN and TP testing. The
primary objective of using a block digester was to perform precise and reliable acid
digestions on samples. The AIM600 block digester was used.
The main components of the AIM600 block digester are digestion block,
programmable controller, set of digestion tubes and tube rack and cooling stand. The
digestion block is heated with a flat plate heater in order to ensure an even
temperature across the block. The digestion block has 50 wells to place 100 mL
tubes (Figure 5.7).
102
Figure 5.7- Digestion block The digestion block is controlled by a programmable controller. The controller
allows the temperature of the digestion block to be controlled to within ± 2 0C. More
details on this digestion system can be found in the instrument manual (AIM600
block digestion system -user manual 2005).
5.4 Data analysis techniques The focus of the research undertaken was to develop knowledge on nutrient build-up
and wash-off processes. In this context, different data analysis techniques which are
capable of identification of patterns of data variability, relationships among variables
and relationships between variables and objects were selected.
The data analysis techniques used can be catergorised into univariate statistical
analysis techniques and multivariate analysis techniques. Univariate statistical
analysis techniques were used to explore the primary variability of data in terms of
mean and standard deviation. However, the application of univariate data analysis
techniques was limited because of the constraints associated with manipulating and
investigating multiple measurements on many samples. This was overcome by the
application of multivariate data analysis techniques. Multivariate data analysis
techniques such as principal component analysis for pattern recognition and multi
103
criteria decision making methods for ranking objects such as particle size ranges and
to visually display the relationships between variables and objects were employed.
Section 5.4.1, 5.4.2 and 5.4.3 describe the different data analysis techniques used.
The results of data analysis are discussed in Chapter 6, Chapter 7 and Chapter 8.
5.4.1 Mean and Standard deviation Mean and standard deviation (SD) are two widely used univariate statistical
measurements to describe the characteristics of a single variable data set (Bahar and
Ohmori 2007, Adams 1995). Mean is the arithmetic average of the data set. Standard
deviation (SD) measures the dispersion of data about the mean value and is
calculated in the same units as the data. The wider the spread of the data, the higher
the standard deviation (Adams 1995).
5.4.2 Principal Component Analysis Principal Component Analysis (PCA) is one of the most powerful and common
techniques capable of reducing the dimensionality of large sets of data without loss
of information in the original data set. PCA has been extensively used as a pattern
recognition technique in numerous water quality research studies to analyse
multivariate statistical data (Librando et al. 1995; Alberto et al. 2001; Goonetilleke et
al. 2005; Huang et al. 2007). For example, Librando et al. (1995) used PCA to
analyse multivariate micro pollutants in marine waters and Goonetilleke et al. (2005)
used PCA to understand the relationship between water quality and urban form.
Herngren et al. (2005) used PCA to correlate the distribution of heavy metals on
different particle size ranges of suspended solids with a number of chemical
parameters such as pH and dissolved organic carbon in urban wash-off samples.
Huang et al. (2007) used PCA to identify the relationships of pollutant parameters
such as nutrients and heavy metals with total suspended solids and total organic
carbon in the stormwater runoff generated from different urban surface types. In the
research study undertaken, PCA was employed to explore the relationships among
the measured nutrient parameters with influential physico-chemical parameters
104
described in Section 5.3 and thereby to investigate the nutrient build-up and wash-off
processes on urban impervious surfaces.
In general, PCA seeks to establish a combination of variables capable of describing
principal tendencies in the data set of interest. In mathematical terms, PCA
transforms multivariate data into orthogonal components called principal
components (PCs) which are uncorrelated to each other and are linear combinations
of the original variables. In PCA, PCs are calculated with the assumption of
normally distributed data set and therefore needs a large number of samples (objects)
for effective modelling (Adams 1995; Shlens 2005). PCs retain the most variance
within the original data so that transformation is achieved without loss of
information in the data set. The first principal component describes most of the data
variance while the second PC, the next largest amount and so on until as many PCs
are generated depending on the number of variables.
However, though PCA produces a number of PCs, the first few PCs are selected for
interpretation as they represent most of the variance in the data set. The number of
PCs which are selected for interpretation is typically decided by using the scree plot
method (Jackson 1991). Scree plot shows the variation of the eigen values in
descending order with corresponding principal components. According to the point
where the graph first twist, the number of principal components that should be taken
into consideration is decided (Adams 1995).
Before proceeding with PCA, the original data is arranged into a matrix representing
variables by columns and objects by rows. In this research, nutrient parameters and
other physio-chemical parameters measured were taken as variables and relevant
samples were considered as objects. In order to avoid any influence resulting from
different scales of variables, the data were subjected to standard pre-treatment
techniques. The most common pre-treatment techniques that are commonly used are
standardisation, mean centering and normalisation (Kokot and Yang 1995;
Libarando et al. 1995; Nguyen et al. 1999; Tyler et al. 2007).
Standardisation is achieved by dividing the individual value in each cell from the
standard deviation of that column. Mean centering consists of subtracting the mean
105
value of each variable from each element in their respective column. In PCA, mean
centered data tend to describe the first PC in the direction of the largest variance in
the data. Normalisation is performed individually on the objects (rows) and not on
the variables. In this, the sum of all variables in each object is computed and each
variable value is then divided by the object sum. Normalisation is typically done to
eliminate the effects on the raw data set arising from stability and sensitivity
differences of instruments. However, in this research, due to the large differences in
the magnitudes between variables in the original data, it was pre treated by both
mean centering and standardisation which is called auto scaling, in order to ensure
that all the variables have equal weights (Purcell et al. 2005; Settle et al. 2007).
Auto scaling is achieved by subtracting the mean value of each column from each
individual value in each cell and dividing by the standard deviation of the relevant
column. This results in new values that have a mean of zero and unit standard
deviation. Additionally, since PCA is sensitive for atypical objects, they are removed
from the data matrices prior to analysis by performing Hotelling T2 test (Lim et al.
2006). According to this method, all the objects that lie within 95% confidence level
are enclosed in Hotelling T2 ellipse and the objects that lie outside the ellipse are
considered as atypical objects.
The pre treated data matrix is then subjected to PCA. Each PC generates scores and
loadings. The scores are projections of the objects on to a particular PC and the
loadings are a measure of the contribution of original variables to a particular PC.
Therefore, the data can be conveniently displayed as a PC scores–scores plot and as
a PC loadings plot. These plots provide guidance for the recognition of important
variables and objects on a PC. Most importantly, valuable information can be
obtained from a biplot, which overlays the loadings over a scores plot.
A biplot represents the loading of each variable as a vector and the score of each
object in the form of a data point. Consequently, a biplot provides an effective
method for studying object–object, variable–variable and object–variable
relationships (Kokot and Phuong 1999; Lim et al. 2006; Huang et al. 2007). The
degree of correlation between variables is decided depending on the angle between
variable vectors. An acute angle between two vectors indicates strong correlation
106
between the respective variables whereas obtuse angle indicates weak correlation.
Right-angled vectors indicate no correlation. Objects with similar characteristics
make clusters. The PCA in this study was carried out using StatistiXL 2004 software
(Roberts and Withers 2004).
5.4.3 Multi Criteria Decision Making Methods (MCDM) Multi criteria decision making methods facilitate decisions making when dealing
with multivariate problems. Most common decision making methods which can be
found in the literature are ELECTRE, SMART, PROMETHEE and GAIA
(Moshkovich et al. 1998; Lahdelma et al. 2003; Martin et al. 2007). In general, all of
these methods are designed to provide a decision by comparing the performance or
preference of one object to another.
However, PROMETHEE (Preference Ranking Organization Method for Enrichment
Evaluation) method that is coupled with the principal component analysis approach
of the GAIA (Graphical Analysis for Interactive Assistance) visualisation technique
has been recognised as a more sophisticated technique in multi criteria decision
making compared to the other MCDM methods (Brans et al. 1986; Keller et
al.1991). PROMETHEE methods are more favourable than ELCTRE due to its
simplicity, clearness and stability in application (Brans et al. 1986).
Furthermore, unlike the conventional PCA method, which generally needs a large
number of samples for effective modelling, PROMETHEE provides ranking even for
as few as two samples and GAIA visually presents PROMETHEE results.
Consequently, PROMETHEE and GAIA have been increasingly employed to handle
multivariate data in environment decision making (Khalil et al. 2004; Herngren et al.
2006; Ayoko et al. 2007). PROMETHEE and GAIA methods were selected in this
study to analyse the nutrient build-up process. The selection of the method was
primarily due to the lower number of build-up samples investigated and hence the
limited possibility of application of the conventional PCA method.
107
PROMETHEE and GAIA are out ranking methods based on the principle of pair
wise comparisons of the objects and variables. Khalil et al. (2004) used
PROMETHEE and GAIA for site selection for sustainable on-site sewage effluent
disposal based on the physico-chemical characteristics of the different types of soils.
In their study, these methods were applied to assist in understanding the relationships
between different physico- chemical characteristics and important soil properties in
the context of selection of soil for effluent renovation capacity. In addition, these
methods were used to rank the different sites according to their ability to renovate
effluent.
Most importantly, the study by Herngren et. al (2006), extended the use of
PROMETHEE and GAIA methods to investigate the stormwater pollutants
deposited on impervious surfaces. In their study, solids build-up on road surfaces in
different land uses were analysed for heavy metal concentrations. PROMETHEE
was applied to identify the most polluted site and particle size of the solids which is
most highly polluted in terms of heavy metals. GAIA was applied to determine the
correlations between heavy metals and the particle size ranges of solids and to assess
possible relationships between total organic carbon and heavy metals. As such, the
application of PROMETHEE and GAIA was considered to be quite appropriate to
investigate the nutrients build-up process on urban road surfaces in different land
uses.
The use of PROMETHEE and GAIA methods to investigate the nutrient build-up
processes and results of the data analysis are described in Chapter 6. In this work,
PROMETHEE and GAIA were used with the aid of DecisionLab software (Visual
Decision Inc. 2000). The theory relating to these methods have been discussed in
detail in the literature (Keller et al. 1991; Khalil et al. 2004). However, a brief
description of PROMETHEE and GAIA methods are given below.
A. PROMETHEE
PROMETHEE is a non parametric method which ranks a number of objects or
actions (in this study, build-up samples) based on a range of variables or criteria
(concentrations of pollutants) in a data matrix. For each criterion, ranking order,
108
weighting condition, a specific preference function, a threshold value must be
defined.
Ranking order
In the PROMETHEE method, ranking order can be modelled according to the
preference of the user. Minimised (lower value of a variable) or maximised (higher
values of a variable) conditions are allocated to each criterion to declare the
preference ranking order. Consequently, actions are ranked top-down (maximised) or
bottom-up (minimised).
Weighting
Weight reflects the importance of one criterion over another. The criterion weight is
a positive value, independent from the scale of the criterion. The larger the value, the
more important the criterion. Unless the alternative scenarios are required in the
investigation, a weighting 1 (default) is assigned for all criteria (Visual Decision Inc.
2000).
Preference function
The preference function is a mathematical function P(a, b) which defines how one
object is to be ranked relative to another and translates the deviation between the
evaluations of two actions (samples) on a single criterion (parameter) into a
preference degree. The preference degree is an increasing function of the deviation
where smaller deviations will contribute weaker degrees of preference while larger
ones to stronger degrees of preference (Visual Decision Inc. 2000). Within the
Decision Lab software used, six preference functions with six specific shapes are
available as shown in Table 5.2.
Threshold
Each shape of the preference function depends on up to two thresholds which should
be provided by the user.
• Q is the indifference threshold that represents the largest deviation that the
decision maker considers negligible when comparing two actions on a single
criterion.
109
• P is preference threshold that represents the smallest deviation that is
considered as decisive in comparison of two actions. P cannot be smaller than
Q.
• S is the Gaussian threshold which is a middle value that is only used with the
Gaussian preference function.
(Keller et al. 1991; Kokot and Phuong 1999; Visual Decision Inc. 2000; Carmody et
al. 2005)
Table 5.2- List of preference functions Function Shape
Threshold
Usual
No threshold
U-shape
Q threshold
V-shape
P threshold
Level
Q and P
thresholds
Linear
Q and P
thresholds
Gaussian
S threshold
B. GAIA
GAIA is a visualisation method which displays PROMETHEE results visually as a
principal component biplot (PC1 vs PC2 plot). Here, actions are regarded as objects
and the criteria as variables which provide a suitable form for the application of PCA
algorithm. A typical GAIA biplot shows visually how objects relate to one another
and to the variables, as well as how variables relate to each other. In addition, it
displays the decision axis, π, which represents the weights of the criteria. The
decision axis shows the kind of compromise solution that is proposed by
PROMETHEE. The orientation of the decision axis emphasizes which criteria are
110
predominant and which are possibly neglected (Keller at al.1991; Visual Decision
Inc. 2000; Lim et al. 2005).
In this study, the GAIA results were interpreted according to the guidelines
presented by Espinasse et al. (1997) for the interpretation of GAIA plots and
summarised as follows.
(i) The longer a projected vector for a variable, the stronger the deviations while
shorter the vector the weaker the deviations.
(ii) Independent variables have almost orthogonal vectors while equivalent
variables have close vectors and conflicting variables have opposite vectors.
(iii) Variable vectors oriented in the same direction are correlated while those
oriented in opposite directions are conflicting.
(iv) Objects projected in the direction of a particular variable are strongly related to
that variable while the opposite objects are weakly related to that variable.
(v) Dissimilar objects have significantly different PC coordinates while similar
objects appear as clusters.
(vi) If the decision vector, π, is long, the best objects are those found in that
direction and vice versa.
5.5 Summary
Detailed analysis of nutrients parameters and other-physico chemical parameters
were imperative to fully understand the nutrients build-up and wash-off processes on
urban impervious surfaces. Consequently, a series of laboratory tests were
undertaken to measure the concentration of different species of nutrients and
influential physio-chemical parameters in build-up and wash-off samples. The build-
up and wash-off samples tested included total samples, particulate samples and
dissolved samples.
Both univariate and multivariate data analysis techniques were selected for data
analysis. Univariate data analysis techniques were applied to explore the data set in
terms of mean and standard deviation. However, due to the multiple number of
variables involved in the data set, the use of univariate data analysis only was
111
inadequate to understand the relative importance of each variable in nutrient build-up
and wash-off processes. This problem was overcome by the application of
multivariate data analysis techniques. Consequently, principle component analysis
and multi criteria decision making methods, PROMETHEE and GAIA were used for
pattern recognition, to identify the relationships among variables and to identify the
relationship between variables and objects.
112
113
Chapter 6 Analysis of Nutrient Build-up
6.1 Background
The focus of this chapter is to understand the nutrient build-up process on urban
impervious surfaces. The build-up samples collected from the three road surfaces,
namely, residential, industrial and commercial were tested for solids, organic carbon,
nutrients and particle size distribution as described in Section 5.3. Table A in
Appendix 2 shows the test results of all the physico-chemical parameters measured.
According to past researchers, most of the primary stormwater pollutants such as
organic matter and nutrients are attached to solids (Ball et al. 2000; Vaze and Chiew
2004; Deletic and Orr 2005). Therefore, in order to understand the nutrient build-up
process an understanding of the primary characteristics of solids build-up was
important. The amount of build-up on surfaces is typically expressed in terms of total
solids load (Sartor and Boyd 1972; Herngren 2005; Egodawatta 2007). Furthermore,
the gradation of solids is defined by the particle size distribution (Zafra et al. 2008;
Egodawatta 2007). Consequently, the amount of total solids load and particle size
distribution of collected build-up samples were analysed to understand the primary
characteristics of solids build-up on road surfaces. Moreover, as organic matter has
been identified as a major source of nutrients, test results obtained for total organic
carbon in the collected build-up samples were also analysed (Makepeace et al. 1995;
Graves et al. 2004; Hendrickson 2007; Bian and Zhu 2008).
Fundamental knowledge on the nutrient build-up process was developed based on
the understanding generated on solids build-up. In turn, the knowledge on nutrient
build-up process was extended to understand the physico-chemical parameters that
influence this process. The analysis undertaken was underpinned by the laboratory
test results obtained for the various nutrient parameters.
114
6.2 Investigation of primary characteristics of solids build-up
6.2.1 Total solid (TS) As discussed in Section 4.4, the build-up samples were collected into a water
filtration system where the sample was retained in a water column. Consequently,
laboratory test results obtained for build-up samples included both TSS and TDS.
TDS represented the potential dissolved fraction of solids build-up when wash-off
occurs. The TS load was obtained for each road surface using the laboratory test
results of TSS and TDS. In order to standardise the TS load for each road surface,
the final outcome of TS load was obtained as TS load per unit area of the road
surface. This was done by dividing the TS load by 3 m2, which was the build-up
sampling plot area used for the field investigations (see Section 4.4). Consequently,
TS load was obtained in the form of g/m2. Table 6.1 shows the total solids load for
each road surface and the respective antecedent dry period.
Table 6.1- Amount of total solid load at each road surface and respective antecedent dry period
Road surface TS load (g/m2)
Antecedent dry days
Residential 2.25 8
Industrial 3.44 9
Commercial 4.06 11 The variation of TS load at each road surface is attributed to the nature of
anthropogenic activities and number of antecedent dry days (Egodawatta 2007; Ball
et al. 1998; Bian and Zhu 2008; Zafra et al. 2008; Vaze and Chiew 2002). This is
confirmed by the largest TS load at the commercial road surface as shown in Table
6.1. The relatively high traffic volume, high anthropogenic activities and greater
antecedent dry days could primarily contribute to the highest TS load being present
at the commercial road surface.
115
6.2.2 Particle size distribution Particle size distribution of solids is an important parameter to characterise the
mobility of the particles during wash-off and their association with other pollutants
(Deletic and Orr 2005; Lau and Stenstrom 2005; Bian and Zhu 2008). It is well
understood that the particle size distribution of solids build-up can vary with factors
such as type of land use, surrounding soil conditions and road surface conditions
such as texture depth (De Miguel et al. 1997; Herngren et al. 2006; Bian and Zhu
2008).
As described in Section 5.3, the total build-up samples collected from each road
surface was analysed for the particle size distribution using a Malvern Mastersizer S
instrument. The analysis size range was 0.06 µm to 900 µm. The results of the
particle size distribution measurements were obtained as volumetric particle size
percentages. More details on particle size distribution measurements can be found in
Section 5.3. Figure 6.1 shows the cumulative particle volume distribution curves
obtained for the three build-up samples collected from each road surface. Several
researchers have considered 150 µm as a cut-off range to differentiate the finer and
coarser fractions of solids build-up (for example Herngren et al. 2006; Goonetilleke
et al. 2009). As evident in Figure 6.1, more than 80% of the particles are below 150
µm for all the road surfaces. This indicates a significant amount of fine particles in
the road surface solids build-up. This is in agreement with the findings of Herngren
et al. (2006), who found that up to 90% of particles were below 150 µm. They
investigated residential, industrial and commercial road surfaces in the Gold Coast
which is the same study region where this research study was undertaken. Moreover,
Walker and Wong (1999) noted that 70% of the particles found on Australian road
surfaces are less than 125 µm. Similar observations have been noted by Bian and
Zhu (2008) and Andral et al. (1999) for the road surfaces they investigated.
116
0
10
20
30
40
50
60
70
80
90
100
0.1 1 10 100 1000
Particle size(µm)
Cum
ulat
ive
perc
enta
ge(%
)
Residential Industrial Commercial
Figure 6.1- Cumulative particle size distribution of solids build-up at each road surface
As seen in Figure 6.1, particle size distribution of solids for each road surface is
different to each other. This is attributed to the different land use characteristics
(Herngren et al. 2006; Bian and Zhu 2008). For example, the industrial road surface
shows a relatively higher amount of fine particles compared to the other two road
surfaces. It is hypothesised that this is due to the industrial enterprises such as
cement industries in the surrounding area as discussed in Section 4.3.
In order to understand the composition of solids build-up for each road surface in
detail, the solids load in different particle size ranges was analysed. This was done
by wet sieving the collected samples into five particle size ranges and testing them
separately for solids as described in Section 5.2. Table 6.2, shows the solids load in
each particle size range as a percentage of total solids load for each road surface.
117
Table 6.2- Amount of solids as a percentage of total solids load
Solids load as a percentage (%) Road surface
<1 µm 1-75 µm 75-150 µm 150-300 µm >300 µm
Residential 29.33 14.18 42.50 11.82 2.17
Industrial 33.01 8.27 48.62 6.60 3.50
Commercial 26.41 7.36 38.10 19.33 8.81
As shown in Table 6.2, the solids load is significantly different between the particle
size ranges. For all the road surfaces, the particle size range 75 - 150 µm contains the
highest percentage of total solids load. Furthermore, particle size less than 150 µm
contains more than 70% of the total solids load for all road surfaces. Therefore,
according to both Figure 6.1 and Table 6.2, it can be surmised that a higher amount
of solids build-up on road surfaces is finer than 150 µm irrespective of the type of
land use.
6.3 Investigation of total organic carbon (TOC) in solids build-up
According to Section 6.2, the amount of build-up varies considerably with the
particle size range. Consequently, in order to understand the variability of total
organic carbon content in the different particle size ranges of solids build-up, the
amount of TOC in wet sieved build-up samples was analysed. TOC content in the
particle size range <1 µm was assumed as the fraction that has potential to be in
dissolved form during wash-off. The organic carbon load per unit weight of total
solids is given in Table 6.3.
As shown in Table 6.3, the particle size class 75 -150 µm shows the highest amount
of TOC and the particle size range below 150 µm contains the predominant amount
of TOC for all road surfaces. These results confirm the finding of past researchers
who noted significantly higher amounts of TOC in finer particles than in coarser
particles (Sartor and Boyd 1972; Andral et al. 1999; Bian and Zhu 2008). According
to Sartor and Boyd (1972), this is attributed to the low structural strength of organic
matter which results in being easily ground into fine particles. Furthermore, higher
118
amounts of TOC in fine particles could also be attributed to the relatively larger
surface area of fine particles (Sansalone et al. 1998; Li et al. 2008).
Table 6.3- TOC in the solids build-up at each road surface Road surface
Particle size range
<1 µm (mg/g)
1-75 µm (mg/g)
75-150 µm (mg/g)
150-300 µm (mg/g)
>300 µm (mg/g)
Residential 16.32 11.43 23.50 5.63 4.25
Industrial 3.08 2.55 5.03 0.78 1.14
Commercial 4.93 2.31 7.39 2.47 3.47
Additionally, the difference in the TOC content in each particle size class at the three
road surfaces can be attributed mainly to the different land use characteristics (Bian
and Zhu 2008). This is supported by the highest organic carbon load in residential
road surface for all the particle size ranges. This could be attributed to increased
presence of vegetation in the surrounding area of the residential road surface
investigated, as discussed in Section 4.2.
6.4 Investigation of nutrients build-up process
A detailed analysis of underlying physico-chemical parameters was needed to
understand the nutrient build-up process. In this context, firstly, the linkage between
different nutrient parameters and influential physico-chemical parameters was
investigated. Secondly, the identified linkages between parameters were further
explored to derive a detailed understanding of the nutrient build-up process.
The laboratory test results obtained for nutrient parameters in different particle size
ranges was used for the analysis. As described in Section 5.1, nutrient parameters
investigated were nitrite-nitrogen (NO2-), nitrate- nitrogen (NO3
-), total kjeldahl
nitrogen (TKN), phosphate (PO43-) and total phosphorus (TP). The total nitrogen
(TN) concentration was obtained by adding concentrations of NO2-, NO3
- and TKN.
119
Furthermore, as noted in Section 6.3, TOC showed considerable variability with the
particle size range of solids. Therefore, it was also included in the analysis to
investigate the influence of TOC on the nutrient build-up process.
Principal component analysis (PCA) was selected as the primary analytical tool to
investigate the linkages between different nutrient parameters and influential
physico-chemical parameters. However, as discussed in Section 5.2, only five wet
sieved build-up samples were subjected to laboratory analysis from each road
surface (a total of only fifteen samples for all three road surfaces). Hence, the direct
application of PCA to build-up data analysis in this research could not be
recommended due to the relative scarcity of the number of objects (build-up
samples).
In order to overcome this problem, PROMETHEE and GAIA analysis was used to
investigate the linkage between different nutrient parameters and influential physico-
chemical parameters in solids build-up. As discussed in Section 5.4.3,
PROMETHEE provides a ranking even for two samples while GAIA displays
PROMETHEE results visually as a principal component biplot. More details on
PROMETHEE and GAIA can be found in Section 5.4.3.
6.4.1 Nutrient build-up process
PROMETHEE and GAIA analysis was undertaken considering all the wet sieved
build-up samples for all the road surfaces together. This was based on the hypothesis
that the underlying chemical process of pollutant build-up would not be influenced
by the type of land use. The criterion (variables) used for the PROMETHEE and
GAIA analysis were total solids (TS), total organic carbon (TOC), nitrite-nitrogen
(NO2), nitrate-nitrogen (NO3), total kjeldahl nitrogen (TKN), total nitrogen (TN),
phosphates (PO4) and total phosphorus (TP).
As discussed in Section 5.4.3, the PROMETHEE method required each criterion set
as ‘maximise’ or ‘minimise’ so that higher values or lower values of variables are
ranked first. In this study, all variables were set to maximise such that the most
120
polluted particle size class was ranked first in the PROMETHEE analysis. All
parameters were given the same weighting, and hence no variable was favoured over
others. The preference function selected for all variables was the V-shaped function
which required a threshold value to be applied to each variable. The preference
threshold P was set to the maximum concentration of each variable (Herngren et al.
2005; Ayoko et al. 2007). All concentrations recorded below the lower detection
limit of the analytical instrument used for laboratory testing were set to a value
equivalent to half of the lower detection limit (Harrison et al. 1996; Herngren 2005).
Outcomes of the PROMETHEE analysis are shown in Table 6.4.
Table 6.4- PROMETHEE 2 ranking
Figure 6.2 shows the principal component biplot obtained from GAIA analysis.
Triangular shapes and solid circles display the different size ranges of wet sieved
build-up samples (actions) whereas quadratic shapes represent the variables
(criterion) in the GAIA plane. The letters R, I and C indicate the objects of
Sample Net Φ Ranking order
75-150C 0.42 1
75-150I 0.19 2
<1C 0.10 3
1-75I 0.08 4
75-150R 0.06 5
<1I 0.04 6
<1R 0.05 7
150-300C -0.01 8
1-75C -0.04 9
>300C -0.05 10
1-75R -0.10 11
>300I -0.17 12
150-300I -0.18 13
150-300R -0.19 14
>300R -0.21 15
121
residential, industrial and commercial road surfaces respectively. The total data
variance of 72.69 % explained by the GAIA biplot indicates that the majority of the
information is included in the analysis.
Figure 6.2- GAIA analysis for all the road surfaces (∆ = 72.69%) It is recommended to evaluate findings from PROMETHEE and GAIA analysis with
confirmation from the raw data. In this context, the amount of nutrients available per
unit weight of solids in each particles size range was obtained for each road surface.
Figure 6.3a, 6.3b, 6.3c show the graphical representation of outcomes and Table A in
Appendix 3 presents the numerical values.
PC1
PC2 PC2 PC2 PC2 PC2 PC2
PC1
PC2
PC1
122
Figure 6.3a- Amount of nutrients in Figure 6.3b- Amount of nutrients in
different particle size ranges of solids different particle size ranges of solids
for the residential road surface for the industrial road surface
Figure 6.3c- Amount of nutrients in different particle size
ranges of solids for the commercial road surface
123
As seen in Table 6.4, all the particle size ranges ranked first are below 150 µm for all
the road surfaces. These are the worst polluted objects. This suggests that finer
fraction of solids build-up is the most polluted. Furthermore, Figure 6.2 shows all the
particle size ranges with positive scores on PC1 are below 150 µm. On the other
hand, the variables NO2-, TKN, TN, PO4
3- and TP show positive loadings on PC1.
This indicates higher affinity of nutrients to the particle size range below 150 µm.
Therefore, the finer fraction of solids is the most important for the nutrient build-up
process. This can be confirmed by the findings of several researchers who noted
higher amounts of nutrients adsorbed to the finer fraction of build-up (Shi 1990;
Vaze and Chiew 2002; Bian and Zhu 2008). For example, Vaze and Chiew (2002)
found more than 60% of TN and TP are attached to particles below 150 µm.
Table 6.4 indicates the particle size range 75-150 µm for all road surfaces ranks
highest, which highlights the highly polluted nature of this particle size range. On the
other hand, the GAIA biplot (Figure 6.2) shows particle size range 75-150 µm for all
road surfaces with high PC1 scores. Furthermore, TN and TP vectors show high
positive loadings on PC1. This suggests that TN and TP are mostly associated with
this particle size range. Notably, the decision axis π vector points towards the
particle size range 75-150 µm confirming the significance of this particle size range
in the nutrient build-up process. According to Figure 6.3a, 6.3b, 6.3c, the highest TN
load is in the particle size range 75 -150 µm for all the road surfaces. The particle
size range 75 -150 µm shows the highest TP load for both industrial and commercial
road surfaces. The residential road surface shows the second highest amount of TP
for the same particle size range. Therefore, it can be argued that the particle size
range 75-150 µm exerts the strongest influence on the nutrient build-up process.
According to Figure 6.2, TN is strongly correlated to TOC and TKN which is the
organic form of nitrogen. Furthermore, as evident in Figure 6.3a, 6.3b, 6.3c, more
than half of the TN amount is attributed to TKN in most of the particle size ranges.
This indicates that TKN is the most dominant nitrogen species in the solids build-up
on road surfaces. Consequently, TOC could exert a strong influence on the nitrogen
build-up process. This is further confirmed by the highest amount of both TOC and
TN in the particle size range 75 -150 µm (Table 6.3 and Figure 6.3a, 6.3b, 6.3c) for
124
all road surfaces. Bian and Zhu (2008) investigating build-up on several road
surfaces, found the road surface with the highest amount of organic matter coincided
with the road surface which showed the highest amount of nutrients. Additionally,
the GAIA biplot (Figure 6.2) indicates NO2- and NO3
- are associated mostly with the
particle size range <1 µm which confirms their high degree of solubility.
As shown in Figure 6.2, TP is strongly correlated to PO43-. As evident in Figure 6.3a,
6.3b, 6.3c, more than half of the TP content is attributed to PO43- in a majority of the
different particle size ranges for all the road surfaces. This indicates the dominant
contribution of PO43- to TP in the solids build-up on road surfaces. Additionally,
Figure 6.3a, 6.3b, 6.3c show that the particle size range 1-75 µm has the highest
amount of PO43-. Sartor and Boyd (1972) noted that PO4
3- is mainly associated with
the particle size of less than 43 µm.
As shown in Figure 6.3a, 6.3b, 6.3c, PO43- contribution to the TP amount in the
particle size ranges <1 µm, 75-150 µm and >300 µm at the residential road surface
and particle size range 75-150 µm at the commercial road surface is limited. This
suggests that some other phosphorus forms such as organic phosphorus could also
contribute to the TP amount in solids build-up. However, this relationship is not very
clear as TP shows no correlation to TOC in the biplot shown in Figure 6.2.
Furthermore, as shown in Figure 6.2, all objects in the GAIA biplot are discriminated
based on the particle size range of the solids rather than the type of land use. This
suggests that the nutrients build-up process is solely dependent on the particle size of
solids. Additionally, it confirms that the nutrient build-up process is independent of
land use.
6.5 Conclusions The chapter has discussed the analytical outcomes of the investigation on nutrient
build-up process on three urban road surfaces. The analysis undertaken was
conducted using the laboratory test results obtained for the various nutrient
parameters, solids, organic carbon and particle size distribution. Fundamental
125
knowledge on the nutrient build-up process was developed based on the
understanding generated on solids build-up. Both univariate and multivariate data
analysis techniques were used for the analysis. Following conclusions can be made
regarding the nutrient build-up process.
• Prior to chemical analysis, particle size distribution of solids build-up was
investigated. Particle size distribution of solids for each road surface is
different to each other. This could be primarily due to the different land use
characteristics. However, more than 80% of the particles in solids build-up
on all the road surfaces are below 150 µm. This indicates the dominant nature
of fine particles in the road surface solids build-up.
• Particle size range 75-150 µm shows the highest amount of both TN and TP
for all road surfaces. Therefore, the finer fraction of solids (particles below
150 µm) exerts the strongest influence on the nutrient build-up process. This
is of serious concern as conventional street sweeping practices are generally
ineffective in removing fine particulates. Consequently, best management
practices targeting particles less than 150 µm is important for the removal of
nutrients from road surface solids build-up.
• TKN which is the organic form of nitrogen is the most dominant form of
nitrogen species in solids build-up on road surfaces. This indicates the strong
influence of organic material on the nutrient build-up process. This was
confirmed by the strong correlation of TN with TOC.
• NO2- and NO3
- are mostly associated with the particle size class below 1 µm
which confirms their high degree of solubility.
• PO43- is the most dominant form of phosphorus in solids build-up on road
surfaces. PO43- is mainly associated with the particle size class 1-75 µm.
• All the objects in the GAIA biplot were discriminated based on the particle
size range of the solids but not on the type of land use. This indicates that the
nutrient build-up process is solely dependent on the particle size of solids.
This further confirms the hypothesis that underlying chemical process of
nutrient build-up is independent of the type of land use.
126
127
Chapter 7 Understanding Nutrient Wash-off
7.1 Background As discussed in Section 2.2.2E, nutrient wash-off is one of the key processes
responsible for the pollution of stormwater runoff. It is the process of removal of
pollutants from surfaces during rainfall events and incorporation into stormwater
runoff. Pollutant wash-off is a complex process. Wash-off varies significantly with
the amount of pollutants that accumulate during the preceding dry period, types of
pollutants, characteristics of the surface and rainfall and runoff characteristics
(Duncan 1995; Deletic et al. 1997; Fujiwara et al. 2005; Egodawatta et al. 2006).
Moreover, the wash-off carries pollutants in both dissolved and particulate forms
making it more difficult to understand the wash-off process for each pollutant
(Robien et al. 1997; Herngren et al. 2005). Therefore, extensive investigations are
needed to understand the pollutant wash-off process.
The focus of this chapter is to define the primary physical process of nutrient wash-
off from urban impervious surfaces during rain events. Based on the implicit
assumption that a significant fraction of stormwater pollutants are transported as
solids bound contaminants, firstly, the primary understanding of wash-off behaviour
of total solids (TS) was developed (Thomson et al. 1997; Egodawatta and
Goonetilleke 2008; Mallin et al. 2008; Zafra et al. 2008). Secondly, the composition
of solids in wash-off was investigated by analysing the particle size distribution. As
discussed in Section 2.2.2D particle size is an important characteristic of solids as it
determines the mobility of the particles during the wash-off and their association
with other pollutants (Deletic and Orr 2005; Lau and Stenstrom 2005; Zafra et al.
2008). Therefore, the analysis of variability of particle size distribution of wash-off
solids during wash-off is important to understand the underlying physical process of
solids wash-off.
128
Thirdly, the primary physical process of nutrient wash-off was investigated in
comparison to the knowledge derived for the TS wash-off process. The investigation
of nutrient wash-off process focused on analysing its variability with physical factors
such as rainfall intensity, duration and particle size distribution of wash-off solids.
The investigation was conducted using the data obtained from laboratory analysis
described in Section 5.3. The variability of nutrient concentration with rainfall
intensity and duration was investigated to understand the overall physical process of
nutrient wash-off. Furthermore, in order to understand the influence of the
composition of nutrients in the wash-off process, the concentration of different
nutrient species in the different particle size ranges of wash-off solids was also
investigated.
7.2 Selected wash-off data and pre-treatment The wash-off samples were collected for six different rainfall intensities, each
simulated in five minute duration compartments. However, due to technical
problems which arose during rainfall simulations, the simulation patterns used for
three road sites contained minor differences. A total of 32, 31 and 34 total wash-off
samples were collected from residential, industrial and commercial road surfaces
respectively. More details on the wash-of sample collection procedure can be found
in Section 4.5. The collected samples were analysed for physico-chemical
parameters as discussed in Section 5.3.
Test results obtained for particle size distribution (only in total samples), total
suspended solids (TSS), total dissolved solids (TDS), nitrite nitrogen (NO2--N),
nitrate nitrogen (NO3--N), total kjeldahl nitrogen (TKN), total nitrogen (TN),
phosphate (PO43-) and total phosphorus (TP) in both total and wet sieved wash-off
samples were used for the analysis (See Section 5.2 for more details). The analysis of
chemical parameters such as total organic carbon (TOC) and electrical conductivity
(EC) are not discussed as this chapter is only focuses on the investigating of the
physical process of nutrient wash-off. The influence of chemical parameters on
nutrient wash-off process is separately discussed in Chapter 8.
129
The data generated from laboratory analyses were in the form of pollutant
concentrations for the wash-off of each five minute duration compartments.
Therefore, prior to data analysis, event mean concentration of each parameter was
calculated by assuming the event durations were 0-5 min, 0-10 min, 0-15 min and so
on depending on the total duration of each rainfall intensity simulated. For example,
the 135 mm/hr rainfall intensity was simulated for total duration of 20 min for the
residential road surface. Therefore, the event mean concentrations of parameters for
the wash-off of 135 mm/hr rainfall intensity for residential road surface was obtained
for 0-5 min, 0-10 min, 0-15 min and 0-20 min duration components. The event mean
concentrations of all the parameters measured are shown in Table B and Table C in
Appendix 2.
7.3 Investigation of total solids (TS) wash-off According to research literature, the primary rainfall and runoff variables which
influence the pollutant wash-off process are rainfall intensity, rainfall duration and
runoff volume (Chui 1997; Shaw et al. 2006; Egodawatta et al. 2006, 2007).
However, a number of research studies have focused only on rainfall intensity and
rainfall duration as primary variables to describe the wash-off process, due to the
interdependency of these variables with runoff volume (Yaziz et al. 1989; Gupta and
Saul 1996; Han et al. 2006; Egodawatta et al. 2006, 2007; Brodie and Rosewell
2007). Egodawatta et al. (2006) confirmed that wash-off process is more strongly
associated with rainfall intensity rather than the other rainfall and runoff parameters.
Therefore, the analysis of total solids wash-off process was conducted using rainfall
intensity and duration as explanatory rainfall runoff variables.
7.3.1 Variation of total solids concentration with rainfall intensity and duration
The variation of TS concentration with rainfall intensity and duration was analysed
in order to understand the primary nature of the solids wash-off process. The
concentration of TS was obtained by the addition of TSS and TDS concentrations
obtained for total wash-off samples (See Section 5.3 for more details). Figure 7.1a,
130
7.1b, 7.1c show the variation of TS concentration with the rainfall intensity and
duration for the three road surfaces investigated.
100
150
200
250
300
350
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TS
con
cent
ratio
n m
g/L
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.1a -Variation of TS concentration with rainfall intensity and duration for the residential road surface
300
500
700
900
1100
1300
1500
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TS
con
cent
ratio
n m
g/L
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.1b- Variation of TS concentration with rainfall intensity and duration for the industrial road surface
131
300
500
700
900
1100
1300
1500
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TS
con
cent
ratio
n m
g/L
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.1c- Variation of TS concentration with rainfall intensity and duration for the commercial road surface As evident in Figure 7.1a, 7.1b, 7.1c, the wash-off behaviour of TS for the three road
surfaces exhibits a similar pattern. The wash-off for the first five minute duration of
each intensity shows the highest concentration of TS for all the road surfaces. The
concentration decreases with the increase in duration. As noted by several past
researchers this is attributed to the first flush effect of the pollutant wash-off process
(for example, Tiefenthaler and Schiff 2001; Kim et al. 2007; Ren et al. 2008). Kim et
al. (2007) noted that the solids concentrations sharply decreased in the initial 30–40
minutes. This is due to the wash-off of easily removable pollutants from surfaces
during the initial period of a rainfall event. Tiefenthaler and Schiff (2001) noted that
the highest solids concentration occurred within the first four to six minutes of each
simulated rainfall event and decreasing to a relatively consistent level with the
increase in duration.
Furthermore, as shown in Figure 7.1a, 7.1b, 7.1c, TS concentration shows an
increasing trend with the increase in rainfall intensity. This is in conformity with the
findings of Egodawatta (2007) who noted an increase in the TS load with the
increase in rainfall intensity. The reason for this can be explained by the increased
kinetic energy of the rainfall and the increased flow energy in the wash-off for high
132
rainfall intensities (Shivalingaiah and James 1984(b); Duncan 1995; Van Gijk et al.
2002).
According to research literature, kinetic energy of rainfall is expressed in terms of
energy per unit area per unit time (Jm-2s-1) or as energy per unit area per unit depth
of rain (Jm-2mm-1) (Roswell 1986; Kinnell 1987; Salles et al. 2002). As discussed in
Section 3.3.3, the rainfall simulator was assumed to generate a constant kinetic
energy for all the intensities above 40 mm/hr. This was based on the concept that the
kinetic energy per unit area per unit depth of the rainfall reaches a constant with the
increase in rainfall intensity (Brodie and Roswell 2007).
However, even though the kinetic energy per unit area per unit depth of rainfall is
constant for all the intensities, the kinetic energy per unit area per unit time of
rainfall is not a constant. The kinetic energy per unit area per unit time of rainfall
increases with the increase of rainfall intensity. Consequently, the kinetic energy per
unit time of the rain event is higher for high intensity rain events compared to low
intensity rain events. This could result in the wash-off of relatively higher solids load
with runoff for high intensity rain events.
Furthermore, when the rainfall intensity increases more solids can be held in
suspension due to the higher flow energy which results in increased transport
capacity of the runoff (Shivalingaiah and James 1984(b); Duncan 1995; Vaze and
Chiew 2002). Consequently, when the rainfall intensity increases, suspended solids
concentrations in the runoff become higher, which in turn increases the TS
concentration. Additionally, it can be surmised that the simulated wash-off process
undertaken is in close agreement with the current knowledge on the wash-off
process.
7.3.2 Particle size distribution The composition of solids in the wash-off is determined by the particle size
distribution. It is an important characteristic of solids which influences the pollutant
wash-off process. According to several researchers, the amount of pollutants such as
133
nutrients attached to solids and the degree of attachment varies with the particle size
(for example; Vaze and Chiew 2004; Li et al. 2005; Zafra et al. 2008). In this
context, two types of analysis were performed to understand the particle size
distribution of solids in the wash-off. Firstly, the variation of particle size
distribution of wash-off solids with rainfall intensity and duration was analysed.
Secondly, since the amount of pollutants attached to wash-off solids varies
considerably between the different particle size ranges, the concentration of TS in
different particle size ranges was analysed.
A. Variation of particle size distribution of total solids with rainfall
intensity This analysis was conducted to understand the variability of particle size distribution
of wash-off solids with the rainfall intensity. Figure 7.2a, 7.2b, 7.2c show the
cumulative particle size distribution curves plotted for the total wash-off sample per
each intensity for each road surface. As shown in Figure 7.2a, 7.2b, 7.2c, the wash-
off samples for 65 mm/hr, 86 mm/hr, 115 mm/hr and 135 mm/hr intensities contain
relatively higher amount of fine particles (particles below 150 µm) compared to the
wash-off of 20 mm/hr and 40 mm/hr rain events for both residential and industrial
road surfaces. This indicates that runoff from high intensity rain events carries a
relatively higher amount of fine particles, thus confirming the observations of past
researchers (Deletic 1998; Sansalone et al. 1998; Vaze and Chiew 2002; Brodie
2007).
There is a fraction of fine particles which is not mobilised during the low intensity
rain events (Vaze and Chiew 2002; Egodawatta and Goonetilleke 2008; Zafra et al.
2008). This is due to the inability of the runoff to remove a fraction of fine particles
which are strongly adhered to the surface (Sansalone et al. 1998; Vaze and Chiew
2002; Egodawatta and Goonetilleke 2008).
According to Vaze and Chiew (2002), it is mainly the free solids load (load which is
not strongly adhered to the surface) on the surface that readily detaches and mobilise
during low intensity rain events. On the other hand, high intensity rain events can
also detach a fraction of the fixed solids load (solids which is more strongly adhered
to the surface) which contains more fine particles (Vaze and Chiew 2002; Brodie
134
2007; Zafra et al. 2008). Egodawatta and Goonetilleke (2008) hypothesised that the
fraction of fine particles which remains on the road surface without mobilising by
the runoff from low intensity rain events contains particles with relatively high
density.
0
10
20
30
40
50
60
70
80
90
100
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge(%
)
20mm/hr 40mm/hr
65mm/hr 86mm/hr
115mm/hr 135mm/hr
Figure 7.2a- Variation of particle size distribution of solids with rainfall intensity for the residential road surface
0
10
20
30
40
50
60
70
80
90
100
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge (%
)
20mm/hr 40mm/hr
65mm/hr 86mm/hr
115mm/hr 135mm/hr
Figure 7.2b- Variation of particle size distribution of solids with rainfall
intensity for the industrial road surface
135
0
10
20
30
40
50
60
70
80
90
100
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge (%
)
20mm/hr 40mm/hr
65mm/hr 86mm/hr
115mm/hr 135mm/hr
Figure 7.2c- Variation of particle size distribution of solids with rainfall intensity for the commercial road surface
However, unlike for the residential and industrial road surfaces, the wash-off
generated from even 20 mm/hr and 40 mm/hr rainfall events for the commercial road
surface exhibited a relatively higher amount of fine particles. The reason for this
occurrence is not clear, but it could be possibly due to road texture. For example,
Egodawatta and Goonetilleke (2008) noted reduced pollutant wash-off during low
intensity rain events from comparatively high textured road surfaces due to the
inability of the runoff to mobilise solids attached to the surface.
Consequently, it can be argued that a higher amount of fine particles can be
mobilised by runoff from even low intensity rain events from comparatively smooth
textured road surfaces. However, since the texture depth of the road surfaces was not
measured in this research, the reasons for the higher amount of fine particles in the
wash-off for 20 mm/hr and 40 mm/hr intensity events cannot be well explained in
terms of the texture depth of the commercial road surface. Nevertheless, it can be
clearly stated that irrespective of the land use high intensity rainfall will invariably
mobilise a relatively higher fraction of fine particles (particles < 150µm).
136
B. Variation of particle size distribution of solids with rainfall duration In order to understand the variation of particle size distribution with rainfall duration,
the particle size distribution of wash-off solids collected for each duration was
analysed separately. Figure 7.3, shows the variation of particle size distribution with
rainfall duration for the 20 mm/hr rainfall intensity simulated on the industrial road
surface. Figure A.1, Figure A.2 and Figure A.3 in Appendix 4 show the variation of
particle size distribution with rainfall duration for each rainfall intensity simulated
for residential, industrial and commercial road surfaces respectively.
As evident in Figure 7.3 and Figure A.1, Figure A.2 and Figure A.3 in Appendix 4,
the majority of the wash-off samples show a trend of decreasing amount of fine
particles with increasing rainfall duration for all the road surfaces. This suggests that
pollutant characteristics can vary considerably with the rainfall duration even for a
constant intensity rain event. Furthermore, during long duration storm events, the
coarser particles are able to be detached from the surface easily and transported with
the runoff. This is supported by Li et al. (2005) who noted the increase in coarse
particles in the runoff as a storm progressed.
20mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00Particle size(µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
20-25min 0-30min
0-35min
Figure 7.3- Variation of particle size distribution of wash-off solids with rainfall duration -20mm/hr rainfall intensity for the industrial road surface
137
C. Total Solids in different particle size ranges of wash-off solids The particle size distribution analysis provided primary understanding of the
composition of solids in the wash-off. Particle size of wash-off solids is a major
concern as the amount of pollutants attached to solids considerably vary with the
particle size (Vaze and Chiew 2002; Herngren et al. 2005; Goonetilleke et al. 2009).
For this analysis, mean concentrations of solids measured in the wet sieved wash-off
samples of all the rainfall intensities and durations were used.
The solids concentrations measured in the particle size ranges 1-75 µm, 75-150 µm,
150-300 µm and >300 µm represented the TSS concentration. The solids
concentration measured in the particle size <1 µm was assumed to represent the TDS
concentration. More details on the measurement of TSS and TDS concentrations can
be found in Section 5.3. Table 7.1 shows the mean and standard deviation of solids
concentration for each particle size range for the wash-off samples analysed.
Table 7.1- Solids concentration of wash-off samples
Mean concentration (mg/L) Standard Deviation Particle size range (µm) Res Ind Com Res Ind Com
<1 130.61 311.08 778.23 36.85 86.04 233.53
1-75 9.98 253.76 117.71 4.07 100.54 56.12
75-150 11.72 118.65 54.18 7.51 61.72 24.89
150-300 7.76 49.98 22.44 3.61 19.35 6.62
>300 6.73 30.85 20.06 2.92 11.62 7.87
Note: Res- Residential; Ind- Industrial; Com-Commercial
Table 7.1, shows relatively high standard deviation values for solids concentration
for the majority of the particle size ranges below 150 µm. The solids concentrations
in Table 7.1 shows the average values for the wash-off samples collected for all the
intensities and durations. The higher standard deviation values confirm the high
variability of the solids concentration in the fine particle size ranges. This suggests
138
the wash-off process in fine particle size ranges significantly varies with the rainfall
intensity and duration in comparison to the coarser particle size ranges.
According to Table 7.1, the fine particle size range (<150 µm) shows a significantly
higher amount of solids compared to the coarser particle size ranges (>150 µm). This
is primarily attributed to higher amount of fine particles in the build-up as noted in
Section 6.2.2. The relatively high amount of wash-off solids in the particle size
ranges <150 µm indicates that a correspondingly high amount of pollutants can be
washed-off with the fine particles. This is mainly due to the relatively larger surface
area of fine particles (Dong et al. 1984; Grout et al. 1999; Li et al. 2008).
Furthermore, the particle size range <1 µm which is the dissolved fraction of wash-
off shows the highest concentration for all the road surfaces. This indicates the
dominant nature of dissolved solids in road surface runoff.
7.4 Investigation of nutrient wash-off process As noted in Section 7.3, physical parameters such as rainfall intensity, duration and
particle size of solids considerably influence the solids wash-off process.
Consequently, the investigation of the physical process of nutrient wash-off was
undertaken based on the physical parameters noted above. The nutrient wash-off
process in terms of these variables is described below.
7.4.1 Variability of nutrient wash-off process with rainfall intensity and duration
In order to understand the overall physical process of nutrient wash-off, the variation
in the concentration of nutrient parameters with rainfall intensity and duration was
investigated. Only TN and TP were selected for this analysis as the total amounts of
all nitrogen and phosphorus species are represented by these two parameters. Figure
7.4a, 7.4b, 7.4c, 7.4d, 7.4e, 7.4f show the variation of concentration of TN and TP
with rainfall intensity and duration for each road surface investigated.
As evident in Figure 7.4a, 7.4b, 7.4c, 7.4d, 7.4e, 7.4f, the highest concentrations of
TN and TP belongs to the first five minute duration of each rainfall intensity. Then
139
the concentration decreases with the increase in duration. This is very similar to the
variation observed for TS with rainfall duration. As discussed in Section 7.3.1, TS
exhibits a strong first flush effect in pollutant wash-off. Consequently, it can be
surmised that the nutrient wash-off process is strongly influenced by the first flush
effect of solids wash-off.
TN
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TN
con
cent
ratio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr86mm/hr 115mm/hr 135mm/hr
Figure 7.4a- Variation of TN concentration with rainfall intensity and duration for the residential road surface
TP
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TP
con
cent
ratio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.4b- Variation of TP concentration with rainfall intensity and duration
for the residential road surface
140
TN
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TN
con
cent
ratio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.4c- Variation of TN concentration with rainfall intensity and duration
for the industrial road surface
TP
0
2
4
6
8
10
12
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TP
con
cent
ratio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.4d- Variation of TP concentration with rainfall intensity and duration
for the industrial road surface
141
TN
0
2
4
6
8
10
12
14
16
18
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TN
con
cent
ratio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.4e- Variation of TN concentration with rainfall intensity and duration
for the commercial road surface
TP
0
2
4
6
8
10
12
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TP
conc
entr
atio
n (
mg/
L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.4f- Variation of TP concentration with rainfall intensity and duration
for the commercial road surface
142
As shown in Figure 7.4a, 7.4b, 7.4c, 7.4d, 7.4e, 7.4f, the decrease in concentration is
rapid for the first 20 minutes for most of the rainfall intensities simulated for each
road surface. This indicates that most of the nitrogen and phosphorus could be
washed-off from the surface during the initial period of a rain event. Therefore, the
runoff from short duration rain events would contain a higher concentration of
nutrients compared to long duration events even for the same rainfall intensity.
Consequently, in the design of stormwater quality mitigation strategies, targeting the
nutrients in the wash-off during the initial period of rain events is of importance.
As evident in Figure 7.4a, 7.4b, 7.4c, 7.4d, 7.4e, 7.4f, the variation in concentrations
of TN and TP with rainfall intensities have shown notable differences. The TN
concentration is higher in the wash-off of low intensity rain events (20 mm/hr and 40
mm/hr) and it is considerably low in the wash-of high intensity rain events (115
mm/hr and 135 mm/hr). On the other hand, TP concentration is higher in the wash-
off of high intensity rain events and it is low in the wash-off of low intensity rain
events. For example, the wash-off for 135 mm/hr rainfall intensity has a relatively
low TN and high TP concentrations compared to the wash-off for the 20 mm/hr
rainfall intensity for each road surface. It is hypothesised that this variability of TN
and TP concentrations is due to differences in TN and TP wash-off processes. It is
also important to take into consideration the dilution with surface runoff with high
intensity events having higher flows. In order to develop a detailed understanding
regarding these observations, further analysis was conducted.
A. Understanding the nitrogen wash-off process
Figure 7.5 shows the variation of average concentration of TN with rainfall duration
for all the road surfaces for each intensity. According to Figure 7.5, the concentration
of TN in the wash-off of 20 mm/hr and 40 mm/hr rainfall events are considerably
higher than the concentration of TN in the wash-off of 115 mm/hr and 135 mm/hr
rainfall events. The higher concentrations of TN in the wash-off in 20 mm/hr and 40
mm/hr rainfall events indicate the readily washable nature of nitrogen compounds.
This is attributed to the high degree of solubility of nitrogen compounds
(Goonetilleke et al. 2005; Taylor et al. 2005). On the other hand, the high runoff
volumes generated by the 115 mm/hr and 135 mm/hr rainfall intensities lead to
143
lower concentrations due to the limited nitrogen availability for wash-off. This is
supported by the findings of past researchers who noted that the amount of pollutants
washed-off is primarily influenced by the amount of initially available pollutants on
the surface (Duncan 1995; Pitt et al. 2004). Therefore, it can be hypothesised that the
TN wash-off process is a source limiting process.
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TN
con
cent
ratio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.5-Variation of average TN concentration with rainfall intensity and duration for all the road surfaces As noted in Section 7.3.2, lower rainfall intensities can remove only a fraction of
fine particles. This fraction belongs to the free solids load on the surface and
contains relatively lower density particles. Consequently, these particles can be
easily mobilised in the wash-off by even low intensity rain events. Furthermore, as
noted in Section 6.4.1 nitrogen was mostly associated with the finer fraction of solids
build-up. Vaze and Chiew (2004) also found that a high TN load is attached to the
finer fraction of wash-off solids. Consequently, it can be hypothesised that
particulate nitrogen is primarily attached to the fine particles in the free solids load.
This is further strengthened by the lower concentration of TN in the wash-off of 115
mm/hr and 135 mm/hr rainfall intensities even with relatively high amount of fine
particles. As discussed in Section 7.3.2, high intensity rain events can also detach the
fine particles which are strongly adhered to the surface (fixed solids load).
20 mm/hr and 40 mm/hr
115 mm/hr and 135 mm/hr
144
Additionally, the attachment of higher amount of TN to free solids load could be
attributed to atmospheric deposition of nitrogen. This can be further supported by the
findings of several researchers who noted that atmospheric deposition as one of the
major sources of nitrogen to catchment surfaces (for example Michael and Timpe
1999; Hope et al. 2004; Gonzalez Benitez 2009) as discussed in Section 2.4.
B. Understanding the phosphorus wash-off process
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0 5 10 15 20 25 30 35 40
rainfall duration (min)
TP
conc
entr
atio
n (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 7.6- Variation of average TP concentration with rainfall intensity and duration for all the road surfaces
As evident in Figure 7.6, TP shows a similar behaviour to the wash-off of TS with
rainfall intensity as described in Section 7.3.1. The wash-off for the 115 mm/hr and
135 mm/hr rainfall intensities shows a considerably higher concentration of TP
compared to the wash-off for the 20 mm/hr and 40 mm/hr rainfall intensities. This
indicates that a relatively high amount of phosphorus is washed-off with the runoff
of high intensity rain events. Consequently, it can be argued that a high transport
capacity of flow is needed to wash-off phosphorus. As noted by past researchers,
phosphorus in stormwater runoff is mainly attached to particulates (Sharpley and
Smith 1990; Quinton et al. 2001; Vaze and Chiew 2004; Zhao et al. 2007).
Therefore, the high intensity rain events could wash-off a relatively high amount of
20 mm/hr and 40 mm/hr
115 mm/hr and 135 mm/hr
145
particulates due to higher kinetic energy and transport capacity of the flow.
Consequently, it can be hypothesised that the phosphorus wash-off process is a
transport limiting process.
Similar to nitrogen, the highest amount of TP is also in the finer fraction of solids
build-up. On the other hand, as described in Section 7.3.2, high intensity rainfall
events wash-off relatively higher amount of fine particles which also includes the
fixed solids load on the surface. Therefore, it is also possible to hypothesise that
phosphorus could be attached to the finer fraction of solids which is strongly bound
to the surface (fixed solids load) where only high intensity rain events can detach
them due to higher kinetic energy of the rainfall. This is further supported by the
findings of Vaze and Chiew (2004) who noted all the particulate TP is attached to
the finer fraction of solids in the wash-off samples they investigated. Moreover, the
fixed solids load on the surface contains relatively higher density particles. This
further confirms that the need for higher transport capacity of the flow to wash-off
phosphorus and hence the transport limiting behaviour of the phosphorus wash-off
process.
7.4.2 Nutrients in different particle size ranges of wash-off solids
As discussed in Section 7.4.1, nitrogen and phosphorus show two distinct wash-off
behaviours. It is concluded that this is primarily due to the degree of solubility,
attachment to particulates and the degree of adherence of the particles to the surface.
Furthermore, these differences could also be attributed to the composition of TN and
TP. As discussed in Section 7.3.2, the amount of solids washed-off significantly
varies with the particle size. Therefore, the composition of TN and TP was analysed
based on the different particle size ranges of solids wash-off.
Figure 7.7 provides a graphical representation of the mean concentrations of the
nitrogen and phosphorus species for the different particle size ranges. Table A in
Appendix 4 gives the numerical values of mean concentrations and the standard
deviation of each parameter for each particle size range. As shown in Table A in
Appendix 4, the relatively higher standard deviation values indicates the highly
146
variable nature of the concentration of different nutrient species for each particle size
range. This confirms that the highly variable nature of the nutrient wash-off process
with the rainfall intensity and duration. This is similar to the variability of solids
concentration for different particle size ranges as described in Section 7.3.2.
Figure 7.7a- Nutrients concentration in different particle size ranges of wash-off solids for the residential road surface
Figure 7.7b- Nutrients concentration in different particle size ranges of wash-off solids for the industrial road surface
147
Figure 7.7c- Nutrients concentration in different particle size ranges of wash-off solids for the commercial road surface According to Figure 7.7a, 7.7b, 7.7c and Table A in Appendix 4, particle size range
<1 µm shows the highest concentration for NO2-, NO3
- and TKN for all the road
surfaces. This confirms that nitrogen in the stormwater runoff is predominantly in
dissolved form. Consequently, due to higher solubility of nitrogen species they could
easily wash-off even during low intensity rain events thus confirming the source
limiting behaviour of nitrogen wash-off process as noted in Section 7.4.1. Even
though a higher runoff volume is generated by high intensity rain events, the
concentration of total nitrogen would be low due to the limited amount of nitrogen
available for wash-off on the surface as evident in Figure 7.4.
Furthermore, as shown in Figure 7.7a, 7.7b, 7.7c, particle size ranges 1-75 µm and
75-150 µm contain a higher concentration of particulate total nitrogen compared to
other size ranges. For example, for residential road surface, the TN concentration in
particle size ranges 1-75 µm and 75-150 µm is 0.210 mg/L and 0.172 mg/L
respectively whereas it is 0.03 mg/L for both particle size range 150-300 µm and
>300 µm. This indicates that fine particles are the most important in the particulate
nitrogen wash-off. This is similar to the observations in relation to nitrogen build-up
(see Section 6.4). This observation is further supported by the findings of researchers
who noted higher amounts of particulate nitrogen in the finer fraction of solids in
148
stormwater runoff (for example, Hvitved- Jacobsen et al. 1994;Vaze and Chiew
2004).
As evident in Figure 7.7a, 7.7b, 7.7c, particulate phosphorus is considerably higher
compared to the dissolved phosphorus. Therefore, a higher amount of particulate
phosphorus could detach from the surface and wash-off during high intensity rain
events due to the higher kinetic energy of the rainfall and the transport capacity of
the flow as discussed in Section 7.4.1. This further strengthens the conclusion that
phosphorus wash-off process is a transport limiting process.
Furthermore, as evident in Figure 7.7a, 7.7b, 7.7c, particle size range 1-75 µm shows
the highest concentration of PO43- for all the road surfaces. Particle size range 1-75
µm for the residential road surface and the 75-150 µm for the industrial and
commercial road surfaces show the highest concentration of TP. This indicates the
importance of the fine particle size ranges of solids in the phosphorus wash-off
process.
On the other hand, unlike nitrogen, particles greater than 150 µm also contain an
appreciable amount of particulate phosphorus. Therefore, the particle size range
>150 µm would also exert an appreciable influence on the phosphorus wash-off
process. This is further supported by the relatively higher amount of phosphorus in
the wash-off of high intensity rain events which can easily dislodge the coarser
particles. However, it can be argued that the influence of coarser particles on
phosphorus wash-off process is limited due to the transport limiting behaviour of the
phosphorus wash-off process.
7.5 Conclusions The chapter has discussed the outcomes of the data analysis which focused on
understanding the primary physical process of nutrient wash-off. Rainfall intensity,
duration and particle size of wash-off solids were selected as explanatory variables.
The analysis was conducted using the concentrations of nutrient parameters
149
measured in both total and wet sieved wash-off samples. The conclusions from the
analyses are as follows:
• The nutrient wash-off process is strongly dependent on the rainfall intensity
and duration. However, the variability of the wash-off of nitrogen with
rainfall intensity is significantly different to the phosphorus wash-off. A
higher amount of nitrogen is readily washed-off with low intensity rainfall
events, whereas high intensity rainfall events are needed to wash-off a higher
amount of phosphorus.
• The highest concentrations of both TN and TP were present during the first
five minute duration of the rainfall event. This confirms the first flush effect
on the nutrient wash-off process. Therefore, the runoff from short duration
rain events would contain high concentrations of nutrients compared to long
duration events even for the same intensity. Consequently, in the design of
stormwater quality mitigation strategies for nutrients removal, it is important
to target the initial period of rain events.
• The nitrogen wash-off process can be defined as a source limiting process
and the phosphorus wash-off process as a transport limiting process.
• The nutrient wash-off process is also influenced by the solids particles
adhering to the surface. Nitrogen is washed-off with the readily mobilised
free solids particles on the surface. Phosphorus is washed-off mostly with the
solids particles in the fixed solids load which is strongly adhered to the
surface.
• The composition of solids wash-off influences the nutrient wash-off process.
All the nitrogen species are mostly washed-off in dissolved form whereas
phosphorus is in particulate form. Finer fraction (particles <150 µm) of the
solids wash-off is the most important for both particulate nitrogen and
phosphorus wash-off. This strengthens the importance of removing finer
fraction of solids in the context of urban stormwater quality mitigation.
• An appreciable amount of phosphorus is also present in the particles >150
µm. Therefore, coarser particles could also exert an appreciable influence on
the phosphorus wash-off process. However, this influence is limited due to
the transport limiting behaviour of the phosphorus wash-off process.
150
• The nutrient wash-off process exhibited similar behaviour for all three road
surfaces investigated, thus confirming the hypothesis that the type of land use
is not an influential parameter in the wash-off process on impervious
surfaces.
151
Chapter 8 Analysis of Nutrient wash-off
8.1 Background Chapter 7 discussed the outcomes of the data analysis conducted to understand the
primary physical process of nutrient wash-off. In Chapter 7, the difference in wash-
off processes for total nitrogen and total phosphorus were identified. However, the
analysis in Chapter 7 only is not adequate to develop a comprehensive knowledge on
nutrient wash-off processes. This is due to several reasons. Firstly, complexities in
nutrient wash-off processes can arise due to the presence of range of different
species of nutrients. For example, TN is a combination of TKN, NO3- and NO2
- and
the wash-off process identified for TN may not be common for individual different
species. Differences in the wash-off processes for different species can lead to
differences in the TN wash-off process for a catchment with different composition of
nutrient pollutants. Secondly, wash-off processes identified in Chapter 7 could be
influenced by the presence of substances such as organic carbon. For example, it is
well understood that properties such as solubility of pollutants can be strongly
influenced by the interaction with substances such as organic carbon. Warren et al.
(2003) found that dissolved organic carbon acts as a solubility enhancer for
pollutants such as heavy metals and polycyclic aromatic hydrocarbons.
In order to develop a comprehensive knowledge on the nutrient wash-off process,
analysis of the different species of nutrient parameters and other chemical
parameters is important. Furthermore, the relationships of nutrients with total
suspended solids and dissolved solids which are the key indicators of solids in the
wash-off was investigated. This was done since the nitrogen and phosphorus were
identified as being associated with the dissolved and particulate fractions of solids in
Chapter 7.
Knowledge regarding the influence of physico-chemical parameters on the nutrient
wash-off process is the most important aspect which influences the transferability of
152
the research outcomes between geographical areas. The transferability of past
research outcomes is commonly constrained by their strong reliance on physical
factors such as type of land use and rainfall characteristics and the limited
recognition of physico-chemical parameters which influence the wash-off process.
This chapter has investigated the influence of the different species of nutrient
parameters and other physico-chemical parameters in the nutrient wash-off process.
The data analysis was primarily conducted using multivariate data analytical
techniques. This was to avoid the difficulties that arise in interpreting outcomes from
univaraite data analysis involving many variables. Principal component analysis
(PCA) which has been proven to be a viable tool for water quality data analysis was
selected for the analyses described in this chapter (Goonetilleke et al. 2005;
Herngren et al. 2005; Huang et al. 2007). A detailed discussion on the theory and
application of PCA can be found in Section 5.4.2.
The PCA was mainly conducted to identify the linkages between the different
species of nutrient parameters with TN and TP and linkages between TN and TP
with other physico-chemical parameters and thereby to obtain comprehensive
knowledge on the nutrient wash-off process. In this context, two types of analyses
were conducted. Firstly, PCA was conducted on the wash-off data obtained for
different particle size ranges of solids. This was to identify the influential physico-
chemical parameters in nutrient wash-off for each size range. Secondly, a combined
data matrix which included wash-off data of all the particle size ranges was
subjected to PCA. This analysis was conducted to investigate whether the influence
of physico-chemical parameters on nutrient wash-off could be explained irrespective
of the particle size of wash-off solids.
8.1.1 Selection of wash-off data for analysis
As discussed in Section 5.3, five particle size ranges of wash-off solids (<1 µm, 1-75
µm, 75-150 µm, 150-300 µm, >300 µm) were subjected to laboratory analysis. Prior
to the data analysis, the wash-off data for these five particle size ranges for each road
surface were re-arranged into three particle size categories. This was done since
153
particle size ranges within each group demonstrate common wash-off characteristics
as discussed in Section 7.3.2. The three categories were;
• dissolved fraction (<1 µm),
• particle size ranges 1- 150 µm (which includes 1-75 µm and 75-150 µm
particle size ranges)
• particle size ranges >150 µm (which includes150-300 µm and >300 µm
particle size ranges).
Then, the wash-off data for all the road surfaces were combined for each of the three
particle size ranges. This was based on the hypothesis that the nutrient wash-off
process is independent of the type of land use. This resulted in three matrices with
97, 194 and 194 objects (wash-off samples) representing particle size ranges <1 µm,
1- 150 µm and >150 µm respectively.
Similar to the analysis discussed in Chapter 7, the variables (nutrient parameters and
other physico-chemical parameters) of the three matrices represented the
concentrations of all the water quality parameters analysed. The water quality
parameters used for the analysis were nutrients, namely, nitrite-nitrogen (NO2--N),
nitrate-nitrogen (NO3--N), total kjeldahl nitrogen (TKN), total nitrogen (TN),
phosphate (PO43-), total phosphorus (TP) and other physico-chemical parameters,
namely, total suspended solids (TSS), total dissolved solids (TDS), electrical
conductivity (EC), total organic carbon (TOC) and dissolved organic carbon (DOC).
pH was not included in the analysis as it showed only minor variations as seen in the
Table B in Appendix 2.
Both nitrogen and phosphorus parameters were included in the same data matrix
even though these two nutrients show significantly different wash-off behaviour as
noted in Chapter 7. This was due to the capabilities of multivariate data analytical
approaches to distinguish between processes and influential variables in the presence
of a multiple number of variables. However, NO2- was not included in the two data
matrices developed for particulate wash-off. This was done since NO2- was not
detected in most of the particulate fractions of wash-off solids (See Table C in
Appendix 2). The data matrices prepared were of size 97x 9, 194x8 and 194x8 for
154
particle size range <1 µm, 1- 150 µm and > 150 µm respectively. The mean
concentrations of all the parameters used for the analysis is shown in Table A in
Appendix 5.
8.1.2 Data pre-processing
As discussed in Section 5.4.2, pre-processing of data was done prior to the PCA in
order to eliminate possible biased outcomes. The data pre-processing undertaken was
in number of steps. Firstly, all the concentrations which were below the lower
detection limit of the analytical instrument were set to half the lower detection limit
value (Herngren et al. 2005). This was done to maintain the consistency of the data
matrix. Secondly, all the concentrations were converted to concentrations per unit
weight of build-up load (mg/L/g). This was to eliminate any bias resulting from data
from three different sites and the consequential different build-up loads (see Section
6.2.1).
Thirdly, all the concentration data was auto scaled in order to ensure that all the
variables had equal weight in the analysis (Purcell et al. 2005; Settle et al. 2007).
This was done by subtracting the column mean from each element in the respective
columns and dividing by the standard deviation for that column. Finally, Hotelling
T2 test was performed to identify the atypical objects. The identified objects were
removed from the data matrices in order to improve the sensitivity of the PCA
analysis. Details of Hotelling T2 test are available in Section 5.4.2. After removing
atypical samples, the data matrices were subjected to PCA.
Due to the presence of a large number of objects in a data matrix, a specific
convention was adopted to identify each object. The first character in the label, R, I
or C, corresponded to the samples from residential, industrial or commercial road
surfaces respectively. The second character was a number representing the rainfall
intensity. The numbers 2, 4, 6, 8, 1 and 3 represent the rainfall intensities of 20
mm/hr, 40 mm/hr, 65 mm/hr, 86 mm/hr, 115 mm/hr and 135 mm/hr respectively.
The third character was again a number representing the duration of each rainfall
intensity such that 1 represents 0-5 min duration, 2 represents 0-10 min duration and
155
so on. The final character was a Roman numerical, i, ii, iii, iv or v such that v
represents particle size range <1 µm, iii, iv represent particle size ranges belonged to
1-150 µm size and i, ii represent particle size ranges belonged to >150 µm size.
8.2 Pattern recognition of nutrient wash-off The principal component biplots obtained from PCA analysis were used to
understand the linkage between different species of nutrients with TN and TP and
the linkage between TN and TP with physico-chemical parameters which influence
the nutrient wash-off process. Nutrients and other physico-chemical parameters were
considered as variables and relevant wash-off samples were considered as objects.
As discussed in Section 5.4.2, a principal component biplot represents the loading of
each variable as a vector and the score of each object in the form of a data point. The
degree of correlation between variables is decided depending on the angle between
variable vectors. A small acute angle between two vectors indicates a strong
correlation between the respective variables whereas a large acute angle indicates a
weak correlation between variables. Vectors at right-angles are not correlated.
Furthermore, the correlation matrix which shows the extent of interaction between
variables was also consulted to verify the outcomes of the PCA biplot. A correlation
coefficient greater than 0.7 was considered as a strong correlation between variables,
0.5-0.7 as a good correlation and below 0.5 as a weak correlation between variables.
Additionally, the raw data matrix was also explored to confirm the observations from
the PCA biplot and the correlation matrix.
8.2.1 Analysis of nutrient wash-off in different particle size ranges The nutrient wash-off process was initially investigated for the three different
particle size ranges of wash-off solids as discussed in Section 8.1.1. The outcomes of
the analysis in each particle size range are discussed below.
156
Particle size range <1 µm (Dissolved fraction)
The pre treated data matrix (92x9) was subjected to the PCA. Figure 8.1 shows the
resulting principal component biplots. Figure 8.1 consists of two biplots where
Figure 8.1a shows the PC1 vs PC2 variation and Figure 8.1b shows the PC1 vs PC3
variation. The number of PCs needed for the analysis was determined based on the
variance associated in each PC as indicated in the Scree plot (Jackson 1991) (Figure
A in Appendix 5). Details of the selection of number of PCs and Scree plots are
discussed in Section 5.4.2. As shown in Figure 8.1a, 8.1b, the first three PCs contain
38.8%, 26.2% and 18.6% of data variance with total data variance being about 84%.
This indicates that the inclusion of PC1, PC2 and PC3 for the analysis explains most
of the information.
C34vC33vC32v
C14vC13vC12v
C11v
C85vC84vC83vC82v
C81v
C66vC65vC64vC63vC62vC61vC47vC46v
C45vC44vC43v
C42v
C41v
C28vC27vC26vC25vC24vC23v
C22v
I34vI33vI32v
I14vI13v
I12v
I11v
I85vI84vI83vI82vI81v
I66vI65vI64vI63vI62vI61v
I46vI45vI44v
I42v
I27vI26vI25vI24vI23vI22vI21v
R34vR33vR32v
R13vR12v
R11v
R84v
R83vR82v
R81v
R66vR65vR64vR63vR62vR61v
R47vR46vR45vR44vR43v
R42vR41v
R28vR27vR26vR25vR24v
R23v
R22v
R21v
TP
PO4
TN
TKN
NO3
NO2
DOC
TDS
EC
-5
-4
-3
-2
-1
0
1
2
3
4
-5 0 5 10
PCA 1 (38.8%)
PC
A 2
(26
.2%
)
Figure 8.1a- PC1 vs PC2 biplot obtained from PCA in the dissolved fraction of wash-off Note: R- Residential; I- Industrial; C-Commercial
Residential
Industrial
Commercial
157
R21vR22vR23vR24vR25vR26vR27vR28v
R41vR42vR43vR44vR45vR46vR47vR61vR62vR63vR64vR65vR66v
R81vR82v
R83vR84v
R11v
R12vR13v
R32vR33vR34v
I21vI22vI23vI24vI25vI26vI27v
I42v
I44vI45vI46v
I61v
I62vI63vI64vI65vI66v
I81v
I82vI83vI84vI85v
I11v
I12vI13v
I14vI32v
I33vI34v
C22vC23v
C24vC25vC26vC27vC28v
C41v
C42vC43vC44vC45vC46vC47vC61vC62vC63vC64vC65vC66v
C81vC82vC83vC84vC85v
C11v
C12vC13v
C14v
C32v
C33vC34v
EC
TDS
DOC
NO2
NO3
TKN
TN
PO4
TP
-3
-1
1
3
5
7
9
11
-5 0 5 10
PCA 1 (38.8%)
PC
A 3
(18
.6%
)
Figure 8.1b- PC1 vs PC3 biplot obtained from PCA in the dissolved fraction of
wash-off
Note: R- Residential; I- Industrial; C-Commercial
As shown in Figure 8.1a, the three road surfaces are clearly discriminated on the
biplot. This discrimination could be primarily attributed to the differences in
concentration of physico-chemical parameters associated with each road surface. For
example, as seen in Figure 8.1a, majority of the wash-off samples belonging to the
commercial road surface show positive PC1 scores. At the same time, EC, TDS,
NO2- and DOC show relatively higher positive loading on PC1. This indicates that
the positive PC1 scores for the samples from commercial road surface are due to the
association of EC, TDS, NO2- and DOC.
Therefore, it can be argued that these physico-chemical parameters are principally
responsible for the discrimination of samples from the commercial road surface with
the other two road surfaces. As evident in Table A in Appendix 5, the wash-off
samples for the commercial road surface exhibit significantly higher values for EC,
TDS, NO2- and DOC concentrations in the dissolved fraction of wash-off compared
to other two road surfaces. This confirms that the discrimination of the three road
158
surfaces can be attributed to the difference in the concentration of parameters and not
due to the type of land use. Consequently, it confirms the hypothesis that the nutrient
wash-off process is independent of the type of land use.
As shown in Figure 8.1a, 8.1b and the correlation matrix (Table 8.1), TN shows a
very weak correlation with TDS. However, as noted in Section 7.4.2, TN is mostly
washed-off in the dissolved form. Therefore, in order to understand the relationship
between TN and TDS the raw data was explored. This was done by plotting the
average concentration of TN with TDS concentration for all the road surfaces for
each rainfall intensity (See Figure 8.2).
Table 8.1 - Correlation matrix for the dissolved fraction
EC TDS DOC NO2- NO3
- TKN TN PO43- TP
EC 1.000 TDS 0.638 1.000 DOC 0.713 0.668 1.000 NO2
- 0.835 0.580 0.873 1.000 NO3
- -0.090 -0.149 0.201 0.063 1.000 TKN -0.157 0.195 0.432 0.109 0.344 1.000 TN -0.129 0.208 0.562 0.142 0.400 0.997 1.000 PO4
3- -0.276 -0.220 -0.186 -0.195 -0.267 -0.098 -0.125 1.000
TP 0.003 -0.063 -0.013 0.043 -0.179 -0.227 -0.238 0.866 1.000
As shown in Figure 8.2, TN shows a decreasing trend with the decrease in TDS
concentration for all the intensities. This confirms that TN is mostly washed-off in
the dissolved form. Consequently, it can be said that dissolved nitrogen wash-off
process could be replicated by the TDS wash-off process. However, the reason for
not presenting a good correlation between TN and TDS in the biplots shown in
Figure 8.1 could be due to the lower variance explained by PC1.
159
0
1
2
3
4
5
6
100 200 300 400 500 600 700
TDS (mg/L)
TN (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 8.2-Variation of TN concentration with TDS According to Figure 8.1a, 8.1b and the correlation matrix (Table 8.1), TN is strongly
correlated to TKN. TKN is the organic form of nitrogen species present in
stormwater (Williamson 1985; Gilbert and Clausen 2006; Han et al. 2006; Gioda et
al. 2008). As seen in Table A in Appendix 5, the mean concentrations of TN and
TKN are 1.975 and 1.862 mg/L for the residential road surface, 1.315 and 1.164
mg/L for the industrial road surface and 4.639 and 4.362 mg/L for the commercial
road surface in the dissolved fraction of wash-off solids. This indicates a high
amount (more than 88%) of TN can be attributed to TKN. Therefore, organic
nitrogen is the dominant form of nitrogen in the dissolved fraction of wash-off.
Furthermore, as can be seen in Figure 8.1a, 8.1b and the correlation matrix (Table
8.1), TN shows very weak correlation to both NO2- and NO3
-. As evident in Table A
in Appendix 5, NO2- and NO3
- show very minor contribution to TN compared to
TKN. Therefore, it can be said that the identified TN wash-off process in Chapter 7
could mostly replicate the wash-off process of TKN. Due to the minor contribution
of NO2- or NO3
- to TN, the influence of the wash-off process of these two different
species of nitrogen to TN wash-off process could be considered to be not significant.
160
As seen in Figure 8.1a, 8.1b, the correlations between TN with DOC are
contradictory. Figure 8.1a, shows weak correlation of TN with DOC whereas Figure
8.1b shows good correlation of TN with DOC. The reason for this is attributed to the
relatively lower variance explained by the PC1. In order to clarify the extent of
correlation of TN with DOC the correlation matrix which shows the correlations in
numerical format was consulted. As shown in Table 8.1, the correlation coefficient
between TN and DOC is 0.562. This indicates that TN has a good correlation with
DOC. Therefore, DOC can be regarded as an influential parameter for the dissolved
nitrogen wash-off process.
In order to understand the relationship between TN and DOC, the raw data was also
explored. For this purpose the variation of average concentration of TN with DOC
for all the road surfaces was analysed for each rainfall intensity (See Figure 8.3). As
seen in Figure 8.3, TN concentration shows a decreasing trend with the decrease in
DOC. This confirms that DOC is an influential parameter for the dissolved nitrogen
wash-off process. The influence of DOC on the dissolved nitrogen wash-off process
could be attributed to the presence of the high amount of TKN.
0
1
2
3
4
5
6
0 5 10 15 20 25 30
DOC (mg/L)
TN (m
g/L)
20mm/hr 40mm/hr 65mm/hr
86mm/hr 115mm/hr 135mm/hr
Figure 8.3- Variation of average concentration of TN with DOC
161
Additionally, as evident in Figure 8.1a, 8.1b and the correlation matrix (Table 8.1),
both TN and TKN are not correlated to EC. This indicates that EC is not an
influential parameter for the dissolved nitrogen wash-off process. However, as
shown in Figures 8.1a, 8.1b and the correlation matrix (Table 8.1), NO2- is strongly
correlated to EC. This is attributed to the ionic nature of the nitrite-nitrogen.
However, this influence is not significant due to the limited amount of NO2-
available in the dissolved fraction. Therefore, it can be argued that the composition
of nitrogen strongly influences the underlying chemical process of nitrogen wash-
off.
According to Figure 8.1a, 8.1b, Table 8.1, TP is not correlated to TDS. This
indicates that only a limited amount of phosphorous is washed-off in the dissolved
phase. This confirms the wash-off of process of phosphorus is mostly in particulate
form as discussed in Section 7.4.2.
According to Figure 8.1a, 8.1b, TP is strongly correlated to PO43-. As shown in Table
A in Appendix 5, the mean concentration of TP and PO43- in the dissolved fraction is
0.017, 0.012 mg/L for the residential road surface, 0.590, 0.479 mg/L for the
industrial road surface and 0.429, 0.356 mg/L for the commercial road surface
respectively. This indicates that a high amount of TP is attributed to PO43-.
Therefore, PO43- is the most dominant form of phosphorus in the dissolved fraction
of wash-off. As noted in US EPA (1999), PO43- is the dominant form of soluble
phosphorus in the stormwater and it is readily bioavailable due to its solubility.
Dissolved phosphorus in stormwater runoff could have a major implication in the
eutrophication of receiving water bodies (Hatch et al. 1999, Pacini and Gachter
1999; Reynolds and Davies 2001).
As seen in Figures 8.1a and 8.1b, TP shows no correlation to DOC. This indicates
that DOC is not an influential parameter in the dissolved phosphorus wash-off
process. This could be primarily attributed to the dominant availability of PO43-
which is the inorganic form of phosphorus. Therefore, it can be hypothesised that the
dissolved phosphorus wash-off process could also be influenced by the composition
of phosphorus similar to the nitrogen wash-off process. Additionally, TP is not
correlated to EC (see Figure 8.1 and Table 8.1). This indicates that EC is not an
162
influential parameter for the phosphorus wash-off process similar to the observation
made in relation to the nitrogen wash-off process.
Particle size range 1-150 µm
The matrix (183x8) of the wash-off data for the particle size range 1-150 µm was
used for the analysis. NO2- was not included in the analysis as it was not detected in
most of the wash-off samples belonging to this particle size range. Figure 8.4 shows
the principal component biplot obtained. This size range is regarded as the finer
fraction of the wash-off. The first two principal components were selected for the
analysis based on the Scree plot (Figure B in Appendix 5). As shown in Figure 8.4,
the first two principal components explain 37.6 % and 19.5% of the total data
variance respectively. Therefore, the collective representation of data variance,
which is 57% is relatively low. This highlights the uncertainty of the observations
generated from Figure 8.4. Therefore, the observations derived from Figure 8.4 were
also confirmed by the use of the correlation matrix as given in Table 8.2.
R21iv
R22iv
R23ivR24ivR25ivR26ivR27iv
R28iv
R41ivR42iv
R43ivR44ivR45ivR46ivR47iv
R61ivR62iv
R63ivR64ivR65ivR66iv
R82iv
R83ivR84iv
R11iv
R12ivR13iv
R32ivR33iv
R34iv
R21iii
R23iii
R24iiiR25iii
R26iiiR27iiiR28iii
R41iiiR42iiiR43iiiR44iiiR45iiiR46iiiR47iii
R62iii
R63iii
R64iiiR65iiiR66iiiR81iiiR82iii
R83iiiR84iii
R11iii
R12iiiR13iii
R31iiiR32iiiR33iii
R34iiiI21iv I22ivI23iv I24ivI25ivI26ivI27iv
I42iv
I44iv
I45ivI46iv
I64ivI65iv
I66iv
I82iv
I83ivI84iv
I85iv
I12ivI13iv
I14iv
I32iv
I33iv
I34iv
I21iiiI22iiiI23iiiI24iii
I25iiiI26iiiI27iii
I41iii
I42iii I44iiiI45iiiI46iii
I62iiiI63iiiI64iii
I65iiiI66iii
I81iii
I83iiiI84iiiI85iii
I11iii
I12iiiI13iii
I14iii
I32iii I33iiiI34iii
C23iv
C24ivC25ivC26ivC27ivC28iv
C41iv
C42iv
C43ivC44iv
C45ivC46ivC47iv
C61iv
C62ivC63iv
C64ivC65ivC66iv
C81iv
C82iv
C83ivC84ivC85iv
C12iv
C13ivC14iv
C31ivC32ivC33ivC34iv
C21iiiC22iiiC23iiiC24iiiC25iiiC26iiiC27iiiC28iiiC42iiiC43iiiC44iiiC45iiiC46iiiC47iii
C61iii
C62iiiC63iiiC64iiiC65iii
C66iii
C81iii
C82iiiC83iiiC84iii
C85iii
C31iiiC32iiiC33iiiC34iii
EC
TSS
TOC
NO3
TKN
TN
PO4
TP
-4
-3
-2
-1
0
1
2
3
4
-10 -5 0 5
PCA 1 (37.6%)
PC
A 2
(19
.5%
)
Figure 8.4- Principal component biplot obtained from PCA on the particle size range 1-150 µm of wash-off solids Note: R- Residential; I- Industrial; C-Commercial
163
Table 8.2- Correlation matrix for the particle size range 1-150 µm
EC TSS TOC NO3- TKN TN PO4
3- TP EC 1.000 TSS -0.094 1.000
TOC 0.166 0.529 1.000 NO3
- 0.137 -0.395 -0.233 1.000
TKN 0.045 0.434 0.346 -0.230 1.000 TN 0.028 0.419 0.580 -0.101 0.977 1.000 PO4
3- -0.438 0.388 0.371 -0.086 0.133 0.144 1.000
TP 0.011 0.551 0.260 -0.397 0.156 0.114 0.082 1.000
As shown in Figure 8.4, the three road surfaces are clearly discriminated on the PC1
vs PC2 biplot. This once again can be attributed to the difference in the
concentration of parameters responsible for the discrimination of each group of
objects. For example PO43- shows the highest negative loading on PC2 whereas the
objects belonging to the industrial road surface show negative PC2 scores. As can be
seen in Table A in Appendix 5, the industrial road surface exhibits high PO43-
concentration in the fine particle size ranges compared to that of the other two road
surfaces. Therefore, it can be said that the discrimination of the three road surfaces
on the biplot is not due to the differences in the nutrient wash-off process but rather
due to the differences in the parameters concentrations.
As shown in Figure 8.4, TN shows almost no correlation with TSS. However, the
correlation matrix shows a weak correlation between the TN and TSS with a
correlation coefficient of 0.419. This indicates that only a limited amount of TN is in
this particle size range in the wash-off. Therefore, the TN wash-off process does not
closely follow the TSS wash-off process for this particle size range. Furthermore, as
discussed in Section 7.4.1, nitrogen is washed-off with the readily mobilised free
solids particles on the surface. Additionally, it was also understood that the amount
of fine particles in the free solids load represents only a fraction of fine particles on
the surface (See Section 7.3.2). Therefore, the weak correlation of TN with TSS
further confirms the TN wash-off process identified in Chapter 7.
According to Figure 8.4 and the correlation matrix in Table 8.2, TN is strongly
correlated with TKN and shows very weak correlation to NO3-. As shown in Table A
in Appendix 5, a high amount of nitrogen in the fine particle size range in wash-off
164
is also attributed to TKN and the contribution of NO3- to TN is considerably low.
This would mean that TKN is the dominant form of nitrogen species present in the
fine particles similar to the dissolved fraction of wash-off. This further indicates that
the identified TN wash-off process in Chapter 7 mostly replicates the TKN wash-off
process. Additionally, according to Chapra (1996), as the organic nitrogen derived
from sources such as dead plant matter is easily solubilised, the particulate nitrogen
is important when there is a significant amount of organic nitrogen present.
According to Figure 8.4 and the correlation matrix (Table 8.2), TN shows a good
correlation with TOC with a correlation coefficient of 0.580. Therefore, TOC would
be an influential parameter for the TN wash-off process in this particle size range.
Furthermore, the influence of TOC in the TN wash-off process in the fine particle
size range could also be attributed to the dominant nature of TKN in this size range.
On the other hand, as evident in Figure 8.4 and the correlation matrix (Table 8.2),
NO3- is very weakly correlated to EC and TN shows no correlation to EC. This
indicates that EC is not an influential parameter for the nitrogen wash-off process in
this particle size range. Consequently, it can be postulated that the influence of
chemical parameters in the nitrogen wash-off process in the fine particle size ranges
is also dependent on the composition of nitrogen.
According to Figure 8.4 and the correlation matrix (Table 8.2), TP shows good
correlation to TSS. This would mean that phosphorus in this particle size range is
mostly washed-off with the suspended solids. Consequently, TP wash-off process
could be closely replicated by the TSS wash-off process in this particle size range.
This is further confirmed by the findings of Chapter 7 where the presence of a high
amount of phosphorus in the fine particle size ranges and the similar wash-off
behaviour of phosphorus to total solids was noted. Furthermore, past researchers
have also noted that phosphorus is often closely associated with suspended solids in
stormwater runoff (Williamson 1985; Haygarth and Jarvis 1997; Uusitalo et al. 2000;
Ng Kee Kwong 2002). Rimer et al. (1978) noted that there is increasing phosphorous
concentration with the increase in suspended solids concentration in the runoff.
Figure 8.4, shows a good correlation of TP with PO43-. However, the correlation
matrix shows 0.082 correlation coefficient which indicates that there is no
165
correlation between TP and PO43-. In order to understand the relationship between
TP and PO43-, the raw data was explored. As discussed in Section 8.1.1, the particle
size range 1-150 µm included two primary particle size ranges, namely, 1-75 µm and
75-150 µm.
As shown in Table A in Appendix 5, even though PO43- shows a significant
contribution to TP in the particle size range 1-75 µm for industrial and commercial
road surfaces, but not for the residential road surface and also for the particle size
range 75-150 µm for all the road surfaces. This indicates that PO43- is not the only
form of phosphorus presents in the fine particles in the wash-off. This suggests the
wash-off of other species of particulate phosphorus such as organic phosphorus in
this particle size range. This can be further supported by the findings of the build-up
analysis discussed in Section 6.4.1 where the presence of other forms of phosphorus
was noted.
Notably, unlike the dissolved fraction, TP shows a weak correlation to TOC. This
indicates some contribution of organic phosphorus to TP in this particle size range.
However, as organic phosphorus was not measured, the contribution of organic
phosphorus to TP in this particle size range could not be determined. Arguably, it
can be said that the composition of phosphorous in the fine particle size ranges could
influence the underlying chemical process of phosphorus wash-off. Additionally, as
evident in Figure 8.3, TP is not correlated to EC. Consequently, EC would not be an
influential parameter in the phosphorus wash-off process in this particle size range,
similar to the dissolved fraction.
Particle size range >150µm
Figure 8.5 shows the principal component biplot obtained for the particle size range
>150 µm which is the coarser fraction of wash-off solids. Similar to the fine particle
size range, NO2- was not included in the analysis as it was not detected in most of the
wash-off samples belonging to this size range. The first two principal components
were suggested by the Scree plot (Figure C in Appendix 5) to visually explore the
wash-off data in the particle size range. As shown in Figure 8.5, the first two
principal components contain 31.7% and 27.3% of the total data variance
166
respectively. As the total data variance (around 59%) explained by the first two
principal components is relatively low it was needed to interpret the biplot with a
great care similar to the fine particle size ranges. This was done by analysing the
correlation matrix (Table 8.3) at the same time for the interpretation of the
observations derived from the biplot.
C34i
C33i
C32iC31i
C14iC13i
C12i
C11iC85i
C84i
C83iC82i
C81i
C47iC46iC45i
C44i
C43i
C42iC41i
C28iC27iC26iC25iC24iC23iC22i
C21i
C34iiC33ii
C32ii
C31ii
C14iiC13ii
C12ii
C11ii
C85iiC84iiC83ii
C82ii
C66iiC65iiC64iiC63ii
C62iiC61ii
C47iiC46iiC45iiC44iiC43iiC42ii
C41ii
C28iiC27iiC26iiC25iiC24iiC23iiC22iiC21ii
I34iI33iI32iI31iI14i
I13iI12iI11iI85i
I84i
I83i
I82i
I66iI65i
I64iI63i
I62iI61i
I46iI45i
I44i I42i
I27iI26iI25iI24iI23iI22iI21i
I34iiI33ii
I32iiI31iiI14iiI13iiI12ii
I11iiI85ii
I84ii
I83iiI82ii
I66iiI65iiI64ii
I63iiI62iiI61ii
I46iiI45ii
I44ii I42ii
I41ii
I27iiI26iiI25iiI24ii
I23ii
I22ii
I21ii
R34iR33iR32i
R31i
R13iR12i
R11i
R84i
R83i
R82iR81i
R66iR65iR64i
R63i
R62i
R47iR46iR45iR44iR43iR42i
R41i
R28iR27iR26iR25iR24iR23i
R22iR21i
R34iiR33iiR32iiR31ii
R13iiR12ii
R11ii
R84ii
R83iiR82ii
R81ii
R66iiR65iiR64ii
R63iiR62ii
R61ii
R47iiR46iiR45iiR44iiR43iiR42ii
R41ii
R28iiR27iiR26iiR25iiR24iiR23ii
R22ii
TP
PO4
TN
TKN
NO3
TOC
TSS
EC
-3
-2
-1
0
1
2
3
4
5
-5 0 5 10
PCA 1 (31.7%)
PC
A 2
(27
.3%
)
Figure 8.5- Principal component biplot obtained from PCA on the particle size range >150 µm of wash-off solids Note: R- Residential; I- Industrial; C-Commercial
Table 8.3- Correlation matrix for the particle size range >150 µm
EC TSS TOC NO3- TKN TN PO4
3- TP EC 1.000 TSS -0.342 1.000
TOC 0.207 0.224 1.000
NO3- 0.619 -0.436 0.028 1.000
TKN -0.199 0.245 0.151 -0.253 1.000
TN 0.238 0.080 0.245 0.055 0.761 1.000 PO4
3- -0.603 0.195 0.024 -0.032 -0.012 -0.267 1.000
TP -0.017 0.362 0.649 -0.306 0.364 0.401 -0.180 1.000
167
As shown in Figure 8.5, the three road surfaces are once again clearly discriminated
from each other. This is also attributed to the difference in the concentration of
parameters similar to the observation made in the analysis of the dissolved fraction
and fine particle size range. For example, the objects belonging to commercial road
surface show positive scores on PC2. On the other hand, TN, EC and NO3- show the
highest positive loading on PC2. This shows that TN, EC and NO3- are principally
responsible for the discrimination of the commercial road surface from the other two
road surfaces.
As can be seen in Table A in Appendix 5, the wash-off samples from the commercial
road surface exhibit higher EC value and higher concentration of NO3- and TN in
this particle size range compared to that for the other two road surfaces. This
confirms the discrimination of the three road surfaces is governed by the
concentration of the parameters but not the type of land use. Furthermore, this again
supports the hypothesis that the nutrient wash-off process is not dependent on the
type of land use.
As shown in Figure 8.5 and Table 8.3, TN is not correlated to TSS. This once again
suggests that very limited amount of nitrogen is washed-off in particulate form. This
agrees with the observed TN wash-off process as discussed in Chapter 7. As noted in
Chapter 7, the amount of TN in the majority of the coarser particle size ranges was
considerably low. Consequently, the particulate nitrogen wash-off process could not
be replicated by the TSS wash-off process in the particle size range >150 µm, similar
to the fine particle size range.
According to Figure 8.4 and the correlation matrix (Table 8.3), TN is strongly
correlated to TKN. However, it should be noted that the degree of correlation of TN
with TKN is relatively low in this particle size range compared to that in the
dissolved fraction and the fine particle size range. According to the correlation
matrices (Table 8.1, Table 8.2 and Table 8.3) the correlation coefficients between
TN and TKN are 0.997, 0.977, 0.761 for the dissolved fraction, particle size range 1-
150 µm and the particle size range >150 µm respectively. As seen in Table A in
Appendix 5, TKN makes a very limited contribution to TN for the particle size
ranges 150-300 µm and >300 µm of the wash-off from the residential road surface
168
and the particle size range 150-300 µm for the commercial road surface. On the other
hand, NO3- shows relatively higher contribution to TN in these particle size ranges.
This indicates that TKN is not the dominant nitrogen species for half of the coarser
particle size ranges investigated.
Even though Figure 8.5 visually presents a strong correlation of TN with TOC, the
correlation matrix (Table 8.3) indicates a very weak correlation between TN and
TOC with a correlation coefficient of 0.245 between TN and TOC. Therefore, TOC
would not exert a strong influence on the TN wash-off process in the coarser particle
size range unlike the dissolved fraction and the fine particle size range. This could be
attributed to the presence of relatively low amount of TKN. Furthermore, according
to Figure 8.5 and the Table 8.3, even though NO3- is strongly correlated to EC, TN is
very weakly correlated to EC. This indicates that EC is not an influential parameter
for the nitrogen wash-off process in this particle size range. This could be due to the
relatively lower amount of NO3- in this particle size range similar to the fine particle
size range. Consequently, it can be surmised that the composition of nitrogen could
influence the underlying chemical process of nutrient wash-off in this particle size
range similar to the observations made in the analysis of the dissolved fraction and
the fine particle size range.
As shown in Figure 8.5 and Table 8.3, unlike in the fine particle size range, TP is
weakly correlated to TSS. This indicates that very limited amount of phosphorus
could wash-off in the coarse particle size ranges. This agrees well with the transport
limiting behaviour of the phosphorus wash-off process as noted in Chapter 7.
According to Figure 8.5 and Table 8.3, TP is weakly correlated to PO43- in this
particle size range. As can be seen in Table A in Appendix 5, it is clear that PO43- is
not the dominant phosphorus species in the particle size range >150 µm. This
indicates the presence of other forms of phosphorus in this particle size range similar
to the particle size range 1-150 µm.
As evident in Figure 8.5 and Table 8.3, TP shows strong correlation to TOC unlike
the particle size range 1-150 µm. This indicates that TOC is an influential parameter
for the phosphorus wash-off process in this particle size range. It also indicates a
169
considerable contribution of organic phosphorus to TP in this particle size range.
This is further confirmed by the presence of a lower amount of PO43- in this particle
size range. This points to the influence of the composition of phosphorus for the
underlying chemical process of phosphorus wash-off. Additionally, it can be
hypothesised that coarser particles in the wash-off could contain an appreciable
amount of organic phosphorus. Furthermore, as shown in Figure 8.3, EC is not
correlated to TP. Therefore, it can be surmised that EC is not an influential parameter
for the nutrient wash-off process in this particle size range similar to the dissolved
fraction and the fine particle size range.
8.2.2 Analysis of nutrient wash-off process in all the particle size ranges Section 8.2.1 described the physico-chemical parameters which underpin the nutrient
wash-off process for the different particle size ranges of wash-off solids. However,
currently, design of most of the stormwater quality mitigation strategies are based on
the outcomes of the wash-off investigations conducted on bulk wash-off samples
where both particulate and dissolved fractions are included. Therefore, it was
important to understand if the underlying physico-chemical processes of nutrient
wash-off could be explained independent of the particle size range of wash-off
solids. Consequently, it could lead to a more robust approach for the design of
stromwater quality mitigation strategies targeting the removal of nutrients from the
stormwater runoff. Consequently, PCA was conducted on a single data matrix which
included the wash-off data for all the particle size ranges of wash-off samples.
The number of objects included in the data matrix was 457 which represented, 92
dissolved wash-off samples, 183 samples in the particle size range 1- 150 µm and
182 samples in the particle size range >150 µm. The same set of parameters except
EC which was included in the analysis discussed in Section 8.2.1 was used as EC
was identified as a non influential parameter in the nutrient wash-off process.
Figure 8.6 shows the principal component analysis biplot obtained for the
interpretation of nutrient wash-off process for all the particle size ranges together.
Similar to the PCA analysis described in Section 8.2.1, the number of PCs for the
170
interpretation of wash-off data was determined by the Scree plot method (Figure D
in Appendix 5) and the first two PCs were selected to explore the nutrients wash-off
process.
As can be seen in Figure 8.6, the first two principal components contain 52.7% and
19.4% data variance respectively. The relatively high variance explained by PC1
compared to the biplots described in Section 8.2.1 together with the total data
variance (about 72%) indicates that the majority of the information relating to the
nutrient wash-off process for all the particle size ranges of wash-off solids is
included in the analysis. Furthermore, similar to the analysis discussed in Section
8.2.1, the correlation matrix (Table 8.4) was also used to verify the outcomes of the
PCA biplot.
C34-iC33-iC32-iC31-i
C14-iC13-iC12-iC11-iC85-iC84-iC83-i
C82-iC81-iC66-iC65-iC64-i
C63-i
C47-iC46-iC45-iC44-iC43-iC42-iC41-i
C28-iC27-iC26-iC25-iC24-iC23-iC22-iC21-i
C34-iiC33-iiC32-iiC31-ii
C14-iiC13-iiC12-iiC11-iiC85-iiC84-iiC83-iiC82-iiC66-iiC65-iiC64-iiC63-iiC62-iiC61-iiC47-iiC46-iiC45-iiC44-iiC43-iiC42-iiC41-iiC28-iiC27-iiC26-iiC25-iiC24-iiC23-iiC22-iiC21-iiC34-iiiC33-iii
C32-iiiC31-iii
C14-iii
C12-iii
C85-iiiC84-iiiC83-iiiC82-iiiC81-iiiC66-iiiC65-iiiC64-iiiC63-iiiC62-iiiC61-iii
C47-iiiC46-iiiC45-iiiC44-iiiC43-iiiC42-iii
C28-iiiC27-iiiC26-iiiC25-iiiC24-iiiC23-iiiC22-iiiC21-iii
C34-ivC33-ivC32-iv
C31-iv
C14-ivC13-iv
C12-iv
C11-iv
C85-ivC84-ivC83-ivC82-ivC81-iv
C66-ivC65-ivC64-ivC63-iv
C62-iv
C61-iv
C47-ivC46-ivC45-ivC44-ivC43-iv
C42-iv
C41-iv
C28-ivC27-ivC26-ivC25-ivC24-ivC23-ivC22-iv
C34-vC33-v
C32-vC14-v
C13-vC12-v
C11-v
C85-vC84-vC83-vC82-v
C81-v
C66-vC65-vC64-vC63-vC62-v
C61-vC47-vC46-v
C45-vC44-v
C43-v
C42-v
C41-v
C28-vC27-vC26-v
C25-vC24-v
C23-v
C22-v
I34-iI33-iI32-iI31-iI14-i
I13-iI12-iI11-i
I85-iI84-iI83-i
I82-i
I66-iI65-iI64-iI63-iI62-iI61-iI46-iI45-iI44-i
I42-i
I27-iI26-iI25-iI24-iI23-iI22-iI21-iI34-iiI33-iiI32-ii
I31-ii
I14-iiI13-iiI12-ii
I11-ii
I85-iiI84-iiI83-iiI82-ii
I66-iiI65-iiI64-iiI63-iiI62-iiI61-iiI46-iiI45-iiI44-ii
I42-ii
I41-ii
I27-iiI26-iiI25-iiI24-iiI23-iiI22-iiI21-ii
I34-iii
I33-iii
I32-iii
I14-iiiI13-iii
I12-iii
I11-iii
I85-iiiI84-iiiI83-iii
I82-iii
I81-iii
I66-iiiI65-iii
I64-iiiI63-iii
I62-iii
I61-iii
I46-iiiI45-iiiI44-iii
I42-iiiI41-iii
I27-iiiI26-iiiI25-iiiI24-iiiI23-iiiI22-iiiI21-iii
I34-iv
I33-iv
I32-iv
I31-iv
I14-iv
I13-iv
I12-iv
I11-iv
I85-ivI84-iv
I83-iv
I82-iv
I66-iv
I65-iv
I64-iv
I63-iv
I62-iv
I46-ivI45-ivI44-iv
I42-iv
I41-iv
I27-ivI26-iv
I25-iv
I24-iv
I23-ivI22-iv
I21-iv
I34-vI33-v
I32-v
I14-vI13-v
I12-v
I11-v
I85-vI84-vI83-vI82-v
I81-v
I66-vI65-vI64-v
I63-vI62-v
I61-v
I46-vI45-vI44-v
I42-v
I27-vI26-vI25-vI24-vI23-v
I22-vI21-v
R34-iR33-iR32-iR31-iR13-iR12-iR11-iR84-iR83-iR82-iR81-iR66-iR65-iR64-i
R63-i
R62-i
R47-iR46-iR45-iR44-iR43-iR42-iR41-iR28-iR27-iR26-iR25-iR24-iR23-iR22-iR21-i
R34-iiR33-iiR32-iiR31-ii
R13-iiR12-iiR11-iiR84-iiR83-iiR82-iiR81-iiR66-iiR65-iiR64-iiR63-iiR62-ii
R61-ii
R47-iiR46-iiR45-iiR44-iiR43-iiR42-iiR41-iiR28-iiR27-iiR26-iiR25-iiR24-iiR23-iiR22-ii
R34-iii
R33-iiiR32-iiiR31-iiiR13-iiiR12-iiiR11-iiiR84-iiiR83-iiiR82-iiiR81-iii
R66-iii
R65-iii
R64-iii
R63-iii
R62-iii
R47-iiiR46-iiiR45-iiiR44-iiiR43-iiiR42-iiiR41-iiiR28-iii
R27-iii
R26-iiiR25-iii
R24-iiiR23-iiiR21-iiiR34-ivR33-iv
R32-iv
R13-ivR12-ivR11-iv
R84-ivR83-iv
R82-iv
R66-iv
R65-ivR64-iv
R63-ivR62-iv
R61-iv
R47-ivR46-ivR45-iv
R44-iv
R43-iv
R42-iv
R41-iv
R28-iv
R27-iv
R26-ivR25-iv
R24-ivR23-iv
R22-ivR21-iv
R34-vR33-v
R32-vR13-v
R12-v
R11-v
R84-vR83-v
R82-vR81-v
R66-vR65-vR64-vR63-vR62-vR61-vR47-vR46-vR45-vR44-vR43-v
R42-vR41-v
R28-vR27-vR26-vR25-vR24-vR23-v
R22-v
R21-v
TP
PO4
TN
TKN
NO3
NO2
DOC
TOC
TDS
TSS
-2
0
2
4
6
8
10
-15 -10 -5 0 5 10
PCA 1 (52.7%)
PC
A 2
(19
.4%
)
Figure 8.6- Principal component biplot obtained from PCA on all the particle size ranges of wash-off solids Note: Yellow- Dissolved fraction (v), Pink- Particulate fractions (i, ii, iii, iv)
Dissolved fraction
Particulate fraction
171
Table 8.4- Correlation matrix for the PCA analysis of all the particle size ranges of wash-off solids
Figure 8.6, shows two distinct groups of objects with positive and negative scores on
PC1. The former group consists of all the particulate fraction of wash-off samples (i,
ii, iii, iv). This group is discriminated from the other group by the relatively higher
positive loading of TP and PO43- on PC1 and PC2 respectively. This confirms that
the presence of phosphorus is mostly in particulate form. The latter group includes
the dissolved fraction of wash-off samples (v type) and associated with all the
nitrogen species. This indicates a clear separation between the nitrogen and
phosphorus wash-off processes. As described in Chapter 7, nitrogen and phosphorus
wash-off processes are significantly different to each other.
The nitrogen wash-off process was defined as a source limiting process and
phosphorus wash-off process as a transport limiting process. This is further
strengthened by Figure 8.6 which shows N and P principally responsible for the
separation of dissolved and particulate fractions of wash-off respectively.
Furthermore, the strong discrimination of particulate and dissolved fractions of
wash-off illustrates the potential errors that can arise from the replication of the
nutrient wash-off process based on the wash-off data of bulk wash-off samples.
Consequently, it could lead to misleading information in the modelling approaches
when designing stormwater quality mitigation strategies. However, in order to
understand the validity of this argument in relation to identifying physico-chemical
TSS TDS TOC DOC NO2- NO3
- TKN TN PO43- TP
TSS 1.000
TDS -0.314 1.000
TOC 0.696 -0.478 1.000
DOC -0.275 0.887 -0.419 1.000
NO2- -0.168 0.737 -0.295 0.898 1.000
NO3- -0.352 0.616 -0.424 0.633 0.428 1.000
TKN -0.218 0.810 -0.384 0.795 0.552 0.698 1.000
TN -0.220 0.817 -0.380 0.805 0.566 0.726 0.998 1.000
PO43- 0.563 -0.046 0.458 -0.048 -0.033 -0.072 0.013 0.010 1.000
TP 0.408 -0.313 0.409 -0.271 -0.144 -0.415 -0.283 -0.287 0.120 1.000
172
parameters which underpin the nutrient wash-off process, the correlations between
nutrient parameters and other physico-chemical parameters were investigated.
As shown in Figure 8.6, TN shows strong correlation to TDS. This is confirmed by
high correlation coefficient of 0.817 between TN and TDS (See Table 8.4). This
further confirms that the dominant nature of dissolved nitrogen in the wash-off as
discussed in Chapter 7 and Section 8.2.1. Consequently, it can be said that the
outcomes of the analysis of nitrogen wash-off process based on the wash-off data of
bulk wash-off samples agrees well with the outcomes of the analysis of the nitrogen
wash-off process based on the wash-off data for the dissolved fraction.
Notably, TN shows a strong correlation with all the different species of nitrogen
namely TKN, NO3- and NO2
-. However, as noted in the analysis of the dissolved
fraction of wash-off discussed in Section 8.2.1, the contribution of NO3- and NO2
- to
TN is very low. Nevertheless, this indicates the strong influence which could be
exerted on the nitrogen wash-off process if these two nitrogen species are dominant
in the wash-off. This further confirms that differences in wash-off processes of the
different species of nitrogen could lead to differences in the TN wash-off process
observed in Chapter 7.
Furthermore, as shown in Figure 8.6 and Table 8.4, TN is strongly correlated to
DOC. This confirms the strong influence of DOC to dissolved nitrogen wash-off
process as noted in the analysis of wash-off in the dissolved fraction (See Section
8.2.1). This further confirms the agreement between the outcomes of the analysis of
bulk wash-off samples and the dissolved fraction which were undertaken separately.
Therefore, the outcomes of analysis of bulk wash-off samples can be recommended
for the investigation of dissolved nitrogen wash-off process.
As evident in Figure 8.6, TP shows a weak correlation to TSS. This is confirmed by
the correlation matrix which indicates a correlation coefficient of 0.408 between TP
and TSS. This indicates that the TP wash-off process is not completely replicated by
the TSS wash-off process. However, as noted in Section 8.2.1, TP did show good
correlation to TSS in the fine particle size ranges suggesting that TP wash-off
process can be closely replicated by the TSS wash-off process. Consequently, it can
173
be concluded that when analysing wash-off data in bulk wash-off samples, it could
lead to the loss of some information in the wash-off processes.
As noted in Chapter 7, the composition of phosphorus varies with the particle size of
wash-off solids. Furthermore, as noted in Section 8.2.1, the wash-off process is
strongly influenced by the composition of nutrients. Hence, TP wash-off process can
vary with the particle size of solids to which they are attached. This is primarily due
to the difference in the composition of phosphorus in the different particle size
ranges of wash-off solids but not specifically due to the particle size of wash-off
solids to which the phosphorus are attached. This issue is noteworthy especially in
relation to the investigation of phosphorus wash-off process as it is predominantly
available in the particulate form unlike to nitrogen as noted in Section 7.4.2.
Consequently, care needs to be taken in the analysis of the phosphorous wash-off
process using the bulk wash-off samples to ensure that there is no significant loss of
information. This observation is particularly relevant when designing the stormwater
quality mitigation strategies targeting the phosphorus removal from the runoff.
Furthermore, as noted in Section 8.2.1, the correlation of TP with TSS in fine
particle size ranges has a significant implication to the design of stormwater
management strategies as fine particles contain high amount of phosphorus. This
underlines the importance of the investigation of processes inherent to phosphorus
wash-off in different particle size ranges as it could generate enhanced knowledge.
In turn, this knowledge could lead to the design of more effective stormwater quality
mitigation strategies.
This conclusion can be further supported by the extent of correlation between TP and
TOC described in Figure 8.6 and the correlation matrix (Table 8.4). The correlation
matrix (Table 8.4) shows a correlation coefficient of 0.409 between TP and TOC
indicating a weak correlation. On the other hand, the analysis of different particle
size ranges discussed in Section 8.2.1, revealed a strong correlation between TP to
TOC in the coarser fraction of wash-off solids. This further confirms the possible
loss of some information relating to the phosphorus wash-off process in the analysis
of bulk wash-off samples.
174
8.3 Conclusions The focus of this chapter was to identify the influence of wash-off of different
species of nitrogen and phosphorus in the nutrient wash-off process and to identify
underpinning physico-chemical parameters. For this purpose, the linkage between
different species of nitrogen and phosphorous with TN and TP and the linkage of TN
and TP with physico-chemical parameters namely total suspended solids (TSS), total
dissolved solids (TDS), electrical conductivity (EC), total organic carbon (TOC) and
dissolved organic carbon (DOC) was investigated.
Firstly, the nutrient wash-off process was investigated for the different particle size
ranges of wash-off solids separately to identify the processes inherent to each
particle size range. Secondly, nutrient wash-off process for all the particle size
ranges together was investigated to understand whether the process can be explained
independent of the particle size range. The conclusions from the analysis are as
follows:
• The nutrient wash-off process varies between the different particle size
ranges of wash-off solids. This is primarily due to the difference in the
composition of total nitrogen and total phosphorus for different particle size
ranges and the differences in the wash-off processes of different species of
nutrients.
• TKN is the dominant form of nitrogen species in the dissolved fraction of
wash-off. Both TDS and DOC can be used to replicate the dissolved nitrogen
wash-off process.
• PO43- is the dominant form of phosphorus present in the dissolved fraction.
The amount of phosphorus in the dissolved fraction is very limited and
therefore dissolved phosphorus wash-off process cannot be replicated by the
TDS wash-off process. DOC is also not an influential parameter for dissolved
phosphorous wash-off process.
• TKN is the most dominant form of nitrogen in the fine particle size range.
TOC is an influential parameter for the TN wash-off process in this particle
size range. TN wash-off process cannot be replicated by the TSS wash-off
process in this particle size range.
175
• TP wash-off process strongly follows the TSS wash-off process in the fine
particle size range.
• The contribution of TKN to TN in coarser particle size range is relatively
low. Consequently, the extent of correlation between TN and TOC is
relatively low in comparison to the dissolved fraction and fine particle size
range. Furthermore, TN wash-off process cannot be replicated by the TSS
wash-off process similar to the fine particle size range. This is attributed to
the limited amount of nitrogen present in particulate form.
• TP wash-off process cannot be replicated by the TSS wash-off process in the
coarser particle size range. TOC is an influential parameter for the
phosphorus wash-off process in this particle size range. This could be
attributed to the presence of an appreciable amount of organic phosphorous
in the coarser particle size range.
• EC is not an influential parameter for either nitrogen or phosphorus wash-off
processes.
• Care needs to be taken in the investigation of the phosphorus wash-off
process using bulk wash-off samples to ensure that there is no loss of
information and hence lead to misleading outcomes. Therefore, the
investigation of different particle size ranges of wash-off solids is preferable
in the interest of designing effective stormwater quality management
strategies targeting phosphorus removal.
• The analysis also confirmed the hypothesis that the nutrient wash-off process
is not dependent on the type of land use.
176
177
Chapter 9 Conclusions and Recommendations for further Research
9.1 Conclusions The research project developed in-depth knowledge of the nutrient-build-up and
wash-off processes on urban road surfaces. The research study investigated both the
physical and chemical processes which are inherent to nutrient build-up and wash-
off. The main findings of this research study can be outlined as follows:
Nutrient build-up process • Nutrient build-up is solely dependent on the particle size of the solids build-up
on road surfaces and it is not dependent on the type of land use.
• The finer fraction (particles <150 µm) of the solids is the most important for
the nutrient build-up process.
• Nitrogen build-up is mostly in organic form and total kjeldahl nitrogen is the
most dominant form of nitrogen.
• Phosphorus build-up is mostly in inorganic form and phosphate is the most
dominant form of phosphorus.
Nutrient wash-off process
• The variability of the nitrogen and phosphorous wash-off processes with
rainfall duration is similar.
• The variability of the nitrogen wash-off process with rainfall intensity is
significantly different to the phosphorus wash-off process.
• Nitrogen wash-off process is a source limiting process whereas phosphorus
wash-off process is a transport limiting process.
178
• All the nitrogen species are primarily washed-off in dissolved form whereas
phosphorus is in particulate form. Phosphorus wash-off process closely follows
the total suspended solids wash-off process in fine particle size ranges.
• The nutrient wash-off process varies with the variability of composition of total
nitrogen and total phosphorus. The difference in the wash-off processes of
different species of nitrogen and phosphorous could lead to differences in the
wash-off process of total nitrogen and phosphorus.
• Most of the particulate nitrogen and phosphorus are attached to the finer
fraction (particles <150 µm) of the solids in wash-off.
• Total kjeldahl nitrogen is the dominant form of nitrogen species in the wash-
off. Phosphate is the dominant form of phosphorus present in the dissolved
fraction.
9.1.1 Detailed knowledge of the nutrient build-up process
Fundamental knowledge on nutrient build-up was developed based on the
understanding generated from solids build-up on road surfaces. In turn, the
knowledge on solids build-up process was extended to investigate the physico-
chemical parameters that influence the nutrient build-up process. The analysis of
solids build-up revealed that the total solids load and particle size distribution for
each road surface was different to each other. It is hypothesised that this is primarily
due to the different land use and road surface characteristics and anthropogenic
activities inherent to urban areas.
However, the dominant nature of fine particles (particles <150 µm) in road surface
solids build-up was noted. More than 80% of the particles on all the road surfaces
were below 150 µm. Even though the total solids loads observed for the three road
surfaces were different, the pattern of nutrient build-up was similar irrespective of
the type of land use.
The finer fraction of solids is the most important for the nutrient build-up process.
Nutrients were mostly associated with the particle size range below 150 µm.
Furthermore, the particle size range 75-150 µm exerts the strongest influence on the
179
nutrient build-up process. It was noted that this size range shows the highest amount
of both total nitrogen and total phosphrous for all road surfaces. Therefore, removal
of particles below 150 µm from road surfaces is of importance for the removal of
nitrogen and phosphorus from road surface solids build-up.
Total kjeldahl nitrogen which is the organic form of nitrogen is the most dominant
form of nitrogen species in solids build-up on road surfaces. This indicates that the
nitrogen build-up on road surfaces is primarily in organic form. Phosphate is the
most dominant form of phosphorus in solids build-up on road surfaces which
suggests that phosphorus build-up on road surfaces is primarily in inorganic form.
Phosphate is mainly associated with the particle size range 1-75 µm.
9.1.2 Detailed knowledge of the nutrient wash-off process
The knowledge on the nutrient wash-off process was developed by investigating
both the underpinning physical and chemical processes. In this context, firstly,
knowledge on the primary physical process was developed by analysing the
variability of nutrient wash-off process with physical factors such as rainfall
intensity, duration and particle size distribution of wash-off solids. Secondly, the
influence of the different species of nutrients and physico-chemical parameters to the
overall nutrient wash-off process was investigated.
The variability of the nitrogen and phosphorous wash-off processes with rainfall
duration is similar. The concentration of both nitrogen and phosphorus is higher at
the beginning of a rain event. Consequently, in the design of stormwater quality
mitigation strategies for nutrients removal, it is important to target the initial period
of rain events. However, the variability of the wash-off of nitrogen with rainfall
intensity is significantly different to phosphorus wash-off. A relatively higher
concentration of nitrogen was observed in the wash-off for low intensity rain events
compared to the wash-off for high intensity rain events. This is primarily due to the
limited amount of nitrogen available on the surface to wash-off with the higher
runoff volume of high intensity rain events. Consequently, the nitrogen wash-off
process was defined as a source limiting process. On the other hand, the
180
concentration of phosphorus in the wash-off was high for high intensity rain events
compared to low intensity rain events. The phosphorus wash-off is limited due to the
limited transport capacity of the runoff of even high intensity rain events.
Consequently, phosphorus wash-off process was defined as a transport limiting
process.
The difference in the nitrogen and phosphorus wash-off processes are primarily due
to the difference in the degree of solubility, attachment to particulates, the difference
in the degree of adherence of the solids particles to the surface to which nutrients are
attached and the composition of total nitrogen and phosphorus. All the nitrogen
species are primarily washed-off in dissolved form whereas phosphorus is in
particulate form. The amount of particulate nitrogen available for wash-off is
removed readily as these are mobilised as free solids particles on the surface.
Phosphorus is washed-off mostly with the solids particles which are strongly adhered
to the surface or the fixed solids load. The influence of total organic carbon on the
nutrient wash-off process varies with the particle size of solids. This is primarily due
to the variability of the organic nitrogen content with the particle size. The total
nitrogen and total phosphorus wash-off processes also vary due to the differences in
the wash-off processes of the different nutrients species.
Total nitrogen was not correlated to total suspended solids for both fine (particles
<150 µm) and coarse (particles >150 µm) particle size ranges. Therefore, the total
nitrogen wash-off process cannot be replicated by the total suspended solids wash-
off process for both fine and coarse particle size ranges. The total phosphorus wash-
off process can be closely replicated by the total suspended solids wash-off process
in the case of fine particle size ranges. However, total phosphorus wash-off process
cannot be replicated by the total suspended solids in the case of coarser particle size
ranges.
Most of the particulate nitrogen and phosphorus are washed-off with the finer
fraction (particles <150 µm) of the solids in wash-off. This strengthens the
importance of removing the finer fraction of solids in the context of urban
stormwater quality mitigation.
181
Total kjeldahl nitrogen is the dominant form of nitrogen species in the dissolved
fraction of wash-off and fine particle size ranges of wash-off solids. Phosphate is the
dominant form of phosphorus present in the dissolved fraction. Due to the high
degree of bioavailability of phosphate, the dissolved fraction of wash-off could
strongly influence the water quality of receiving water bodies. The particulate
phosphorus in the wash-off could contain an appreciable amount of organic
phosphorus.
Investigation of the processes which underpin phosphorus wash-off using bulk wash-
off samples could lead to loss of information. This is due to the differences in the
composition of total phosphorus and the inherent variability of the wash-off process
for different particle size ranges. This variability should preferably be taken into
consideration as phosphorus wash-off is predominantly in particulate form.
Therefore, in the investigation of the phosphorus wash-off process using bulk wash-
off samples, the outcomes should be considered with care from the perspective of
designing effective stormwater quality management strategies.
9.1.3 Concluding remarks
Implications of the findings of this research study to urban stormwater quality
mitigation strategies are:
• Nutrients on road surfaces are mostly attached to finer fraction (particles <150
µm) of solids build-up irrespective of type of land use. Consequently, they can
be easily washed-off with the stormwater runoff and impose a significant threat
to receiving waters. Hence the use of conventional street sweeping methods
which are not capable of removing particles below 150 µm could not be
effective in removing nutrients from road surfaces.
• Nitrogen is primarily washed-off in dissolved form hence can be easily
adsorbed by plants and also cause the eutorphication in receiving water bodies.
• Particulate nitrogen and phosphorus are attached to the finer fraction (particles
<150 µm) of solids in the wash-off irrespective of type of land use. Hence
182
stormwater quality mitigation strategies targeting the removal of particles
below 150 µm of solids from urban stormwater runoff of importance.
• Relatively higher amounts of nitrogen and phosphorus wash-off at the
beginning of a rain event. Consequently, it is important to target the initial
period of rain events from the perspective of providing stormwater quality
mitigation strategies for nutrients removal.
• A higher amount of nitrogen is washed-off even with low intensity rain events.
This highlights the importance of taking into consideration the wash-off of low
intensity rain events in the design of stormwater quality mitigation strategies
targeting nitrogen removal.
• The investigation of different particle size ranges of wash-off solids is
preferable in the interest of designing effective stormwater quality mitigation
strategies targeting phosphorus removal.
9.2 Recommendations for future research
This study has covered a broad range of work in relation to understanding nutrient
build-up and wash-off processes on urban impervious surfaces. However, there are a
number of critical areas that have not been addressed in the scope of this study.
Therefore, it is recommended that the following studies are undertaken to further
enhance the in-depth knowledge on nutrients build-up and wash-off processes:
• This study investigated pollutant build-up using samples which included both
free solids particles and fixed solids particles. From the study outcomes, it
was hypothesised that particulate nitrogen is washed-off mostly with the free
solids particles whereas phosphorus is washed-off with the fixed solids
particles. Therefore, it is important to investigate the free solids and fixed
solids loads separately to confirm this hypothesis.
• It is recommended that build-up samples should be collected so as to
investigate if there is any influence of the antecedent dry period and the
climatic seasons on the nutrient build-up process.
183
• It is also recommended wash-off samples should be collected so as to
investigate if there is any influence of the antecedent dry period and the
climatic seasons on the nutrient wash-off process.
• Validation of relationships and processes derived by this research using
natural rainfall runoff data is recommended. This will enhance the
transferability of the outcomes of this study.
• The study undertaken investigated only the build-up and wash-off processes
of phosphate as a different species of phosphorus. However, from the build-
up and wash-off investigations it was understood that an appreciable amount
of organic phosphorus could also be present. Therefore, the investigation of
organic phosphorus wash-off process is recommended to obtain more
detailed understanding of the phosphorus wash-off process.
• It was understood that the total nitrogen and total phosphorus wash-off
processes could vary based on the differences in the wash-off processes of
different species of nutrients. However, the investigation of this was limited
in the research undertaken as total kjeldahl nitrogen and phosphate were
found to have significant contribution and a limited contribution of nitrate
nitrogen and nitrite nitrogen for most of the particle size ranges. Therefore,
the investigation of nutrient wash-off process having variable contributions
of different species of nutrients is recommended. This would enhance the
understanding of the influence of the differences in the wash-off processes
for different nutrient species.
• Biological parameters were not investigated in this research. However, the
investigation of biological parameters is recommended so as to understand if
these parameters exert any influence on the nutrient wash-off processes.
184
185
Chapter 10 References
1. Aalderink, R.H., Van Duin, E.H.S., Peels, C.E. and Scholten, M.J.M. (1990).
Some characteristics of runoff quality from a separated sewer system in
Lelystad, The Netherlands. Proceedings of the Fifth International Conference
on Urban Storm Drainage, Osaka, Japan, Vol. 1, pp. 427-432.
2. Adams, M.J. (1995). Chemometrics in analytical spectroscopy. Cambridge
[England], Royal Society of Chemistry.
3. AIM600 User Manual (2005), AIM600 Block Digestion System-User Manual.
4. Alberto, W.D., Del Pilar, D.M., Valeria, A.M., Fabiana, P.S., Cecilia, H.A. and
Angeles, B.M.D.L. (2001). Pattern recognition techniques for the evaluation of
spatial and temporal variations in water quality. A Case Study: Suquia River
Basin (Cordoba-Argentina). Water Research, Vol. 35 (No. 12), pp. 2881-2894.
5. Allison, R.A., Chiew, F.H.S. and McMahon, T.A. (1998). Nutrient contribution
of leaf litter in urban stormwater. Journal of Environmental Management, Vol.
54, pp. 269-272.
6. Ayoko, G.A., Singh, K., Balera, S. and Kokot, S. (2007). Exploratory
multivariate modelling and prediction of the physico-chemical properties of
surface water and groundwater. Journal of Hydrology, Vol. 336 (No 1-2), pp.
115-124.
7. Andral, M.C., Roger, S., Montrejaud-Vignoles, M. and Herremans, L. (1999).
Particle size distribution and hydrodynamic characteristics of solid matter
carried by runoff from motorways. Water Environment Research, Vol. 71 (No.
4), pp. 398-407.
8. Angino, E.E., Magnuson, L.M. and Stewart, G.F. (1972). Effects of
urbanisation on stormwater runoff quality: A limited experiment, Naismith
ditch, Lawrence, Kansas. Water Resources Research, Vol. 8 (No 1), pp. 135-
140.
9. APHA (2005). Standard Methods for the Examination of Water and Waste
Water. Washington, D.C: American Public Health Association.
186
10. AQ 2 Discrete Analyser Operator Manual (2006). Seal Analytical Operator
Hand Book.
11. Arnold, C.L.Jr. and Gibbons, C.J. (1996). Impervious surface coverage:
Emergence of a key environmental factor. Journal of the American Planning
Association, Vol. 62 (No. 2), pp. 243-258.
12. Artola, C.G., Pareja, B.L. and Garcia, G. (1995). Impact on hydrology and
nutrient movements of developments in river basins draining into reservoirs.
Water Research, Vol. 29 (No. 2), pp. 601-609.
13. Asante, Y.O. and Stephenson, D. (2006). Estimation of storm runoff loads
based on rainfall-related variables and power law models: Case study in
Alexandra, Water South Africa, Vol. 32 (No. 2), pp. 137-144.
14. AS/NZS., (1998). 5667.1 - Australian New Zealand Standards, Water quality -
Sampling. Unpublished manuscript.
15. ASCE (1975). Aspects of hydrological effects of urbanisation. Journal of
Hydraulics Division (ASCE), Vol. HY5, pp. 449-468.
16. Assouline, S., El Idrissi, A. and Persoons, E. (1997). Modelling the physical
characteristics of simulated rainfall: A comparison with natural rainfall. Journal
of Hydrology, Vol. 196 (No. 1-4), pp. 336-347.
17. Atasoy, M., Palmquist, R.B. and Phaneuf, D.J. (2006). Estimating the effects of
urban residential development on water quality using microdata. Journal of
Environmental Management, Vol. 79 (No 4), pp. 399-408.
18. Bahar, Md. M., and Ohmori, H. (2007). Relationship between river water
quality and land use in a small river basin running through the urbanizing area
of Central Japan. The Japanese Society of Limnology.
19. Ball, J.E., Jenks, R. and Aubourg, D. (1998). An assessment of the availability
of pollutant concentrations on road surfaces. Science of the Total Environment,
Vol. 209 (No 2-3), pp. 243-254.
20. Ball, J.E., Wojcik, A. and Tilley, J. (2000). Stormwater quality from road
surfaces: Monitoring of the Hume Highway at South Strathfield, The
University of New South Wales, School of Civil and Environmental
Engineering, Water Research Laboratory, Research Report 204.
21. Balmer, P., Malmquist, M.A. and Sjoberg, A. (1984). Urban storm drainage.
Proceedings of the Third International Conference, Goteborg. Chalmers
University of Technology, Sweden.
187
22. Ballo, S., Liu, M., Hou, L. and Chang, J. (2009). Pollutants in stormwater
runoff in Shanghai (China): Implications for management of urban runoff
pollution. Progress in Natural Science, Vol. 19 (No 7), pp. 873-880.
23. Bannerman, R.T., Owens, D.W., Dodds, R.B. and Hornewer, N.J. (1993).
Sources of pollutants in Wisconsin stormwater. Water Science and
Technology, Vol. 28 (No. 3-5), pp. 241-259.
24. Barbara, M.S., Ahif, S., Felipe, V. and Rullkotter, J. (2009). Natural vs.
anthropogenic sources of hydrocarbons as revealed through biomarker
analysis: A case study in the southern Gulf of Mexico.
25. Barrios, A. (2000). Urbanisation and Water Quality. CAE Working paper
series, WP00-1. American Farmland Trust’s Center for Agriculture in the
Environment, DeKalb.
26. Bella, A., Duckstein, L. and Szidarovszky, F. (1996). A multicriterion analysis
of the water allocation conflict in the Upper Rio Grande basin. Applied
Mathematics and Computation, Vol. 77 (No. 2-3), pp. 245-265.
27. Bian, B. and Zhu, W. (2008). Particle size distribution and pollutants in road-
deposited sediments in different areas of Zhenjiang, China. Environmental
Geochem Health, DOI 10.1007.
28. Bodo, G.M. (1989). Heavy metals in water and suspended particulates from an
urban basin impacting Lake Ontario. Science of the Total Environment, Vol.
87/88, pp. 329-344.
29. Bohemen, H.D. and De Laak, W.H.J.V. (2003). The influence of road
infrastructure and traffic on soil, water and air quality. Environmental
Management, Vol. 31 (No. 1), pp. 50-68.
30. Booth, D.B. (1991). Urbanization and the natural drainage system-impacts,
solutions and progress. The Northwest Environmental Journal, Vol. 7, pp. 93-
118.
31. Brabec, E., Schulte, S. and Richards, P.L. (2002). Impervious surfaces and
water quality: A review of current literature and its implications for watershed
planning. Journal of Planning Literature, Vol. 16 (No. 4), pp. 499-514.
32. Bradford, W.L. (1977). Urban stormwater pollutant loadings: A stastical
summary through 1972. Journal of Water Pollution Control Federation, Vol.
49, pp. 613-657.
188
33. Brans, J.P., Vincke, Ph., Mareschal, B. (1986). How to select and how to rank
projects, the PROMETHEE method. European Journal of Operational
Research, Vol. 24, pp. 228-238.
34. Brezonik, P.L. and Stadelmann, T.H. (2002). Analysis and predictive models of
stormwater runoff volumes, loads and pollutant concentrations from
watersheds in the twin cities Metropolitan Area, Minnesota, USA. Journal of
Water Research, Vol. 36, pp. 1743-1757.
35. Brezonik, P.L., Osgood, R.A., Olmanson, L., Day, E., Hatch, L., Doyle, J.,
Perry, J.A., Bauer, M., MacBeth, E. and Anderle, T. (2002). Cumulative
impacts of development on lakes in the north central hardwood forest eco
region of Minnesota: An exploratory study. University of Minnesota, Water
Resources Centre, Technical Report 144.
36. Brick, C. and Moore, J. (1996). Diel variation of trace metals in the upper
Clark Fork River Montana. Environmental Science and Technology, Vol. 30
(No. 6), pp.1953-1960.
37. Brilly, M., Rusjan,S. and Vidmar, A. (2006). Monitoring the impact of
urbanisation on the Glinscica stream. Physics and Chemistry of the Earth, Vol.
31, pp. 1089-1096.
38. Brinkmann, W.L.F. (1985). Urban storm water pollutants: Sources and
loadings. Geo Journal, Vol. 11 (No. 3), pp. 277-283.
39. Bris, F.J., Garnaud, S., Apperry, N., Gonzalez, A., Mouchel, J.M. and Chebbo,
G. and Thevenot, D.R. (1999). A street deposit sampling method for metal and
hydrocarbon contamination assessment. The Science of the Total Environment,
Vol. 121, pp. 229-237.
40. Brodie, I.M. (2007). Investigation of stormwater particles generated from
common urban surfaces. University of Southern Queensland-Thesis, Australia.
41. Brodie, I.M. and Porter, M.A. (2006). Stormwater particle characteristics of
five different urban surfaces. Faculty of Engineering and Surveying, University
of Southern Queensland, Toowoomba, Australia.
42. Brodie, I. and Rosewell, C. (2007). Theoretical relationships between rainfall
intensity and kinetic energy variants associated with stormwater particle wash-
off. Journal of Hydrology, Vol. 340, pp. 40- 47.
43. Bubenzer, G.D., Molnau, M. and McCool, D.K. (1984). Low intensity rainfall
with a rotating disk simulator. Transactions of the ASAE, pp. 1230-1232.
189
44. Bujon, G., Herremans, L. and Phan, L. (1992). Flupol: A forecasting model for
flow and pollutant discharge from sewerage systems during rainfall events.
Water Science and Technology, Vol. 25 (No. 8), pp. 207-215.
45. Carmody, O., Kristof, J., Frost, R.L., Mako, E., Kloprogge, J.T. and Kokot, S.
(2005). A spectroscopic study of mechanochemically activated kaolinite with
the aid of chemometrics. Journal of Colloid and Interface Science, Vol. 287,
pp. 43-56.
46. Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W., Sharpley, A.N.
and Smith, V.H. (1998). Nonpoint pollution of surface waters with phosphorus
and nitrogen. Issues in Ecology, Vol. 8 (No 3), pp. 559 - 568.
47. Carter, C.E., Greer, J.D., Braud, H.J. and Floyd, J.M. (1974). Raindrop
characteristics in South Central United States, Transactions of the ASAE
Vol.17, pp.1033-1037.
48. Cech, I. and Assaf, K. (1976). Quantitative assessment of changes in urban
runoff. Journal of Irrigation and Drainage Division, Vol. 102 (No 1), pp. 119-
125.
49. Chang, M., McBroom, M.W. and Scott B.R. (2004). Roofing as a source of
nonpoint water pollution. Journal of Environmental Management, Vol. 73 (No.
4), pp. 307-315.
50. Chapra, S.C. (1997). Surface water quality modelling. McGraw-Hill, New
York, USA, pp. 844.
51. Characklis, W.G., Gaudet, F.J., Roe, F.L. and Bedient, P.B. (1979a). Maximum
utilization of water resources in a planned community-stormwater runoff
quality: Data collection, reduction and analysis. United States Environmental
Protection Agency, Report No. EPA- 600/2-79-050b.
52. Characklis, G.W. and Wiesner, M.R. (1997). Particles, metals and water
quality in runoff from large urban watershed. Journal of Environmental
Engineering, Vol 123 (No. 8), pp. 753-759.
53. Charbeneau, R.J. and Barrett, M.E. (1998). Evaluation of methods for
estimating stormwater pollutant loads. Water Environment Research, Vol. 70
(No. 7), pp. 1295-1302.
54. Charbonneau, R. and Kondolf, G.M. (1993). Land use change in California,
USA: Nonpoint source water quality impacts. Environmental Management,
Vol. 17, pp. 453-460.
190
55. Chiew, F.H.S., Duncan, H.P. and Smith, W. (1997). Modelling pollutant build-
up and wash-off: Keep it simple. Proceedings of the 24th Hydrology and Water
Resources Symposium, Auckland, New Zealand, pp.131-136.
56. Christiansen, J. P. (1942). Irrigation by sprinkling. Bulletin No 670, University
of California Agricultural Experiment Station.
57. Chui, P.C. (1997). Characteristics of stormwater quality from two urban
catchments in Singapore. Environmental Monitoring Assessment, Vol. 44, pp.
173-181.
58. Codner, G.P., Laurenson, E.M. and Mein, R.G. (1988). Hydrologic effects of
urbanisation: A case study. Proceedings of the Hydrology and Water Resources
Symposium, pp. 201-205.
59. Coleman, T.J. (1993). A comparison of the modelling of suspended solids
using SWMM3 quality prediction algorithms with a model based on sediment
transport theory. Water Systems Research Group, University of the
Witwatersrand, South Africa.
60. Connecticut Stormwater Quality Manual. (2004). The Connecticut Department
of Environmental Protection.
61. Cordery, I. (1977). Quality characteristics of urban storm-runoff water in
Sydney, Australia. Water Resources Research, Vol. 13 (No. 1), pp. 197-202.
62. Cowen, W.F. and Lee, G.F. (1973). Leaves as source of phosphorus.
Environmental Science and Technology, Vol.7 (No. 9), pp. 853-854.
63. D’Arcy, J.B., Ellis, J.B., Ferrier, R.C., Jenkins, A. and Dils, R. (2000). Diffuse
pollution impacts: The environmental and economic effects of diffuse pollution
in the UK. Terence Dalton Publ. Ltd. (CIWEM), Lavenham.
64. Davis, A.P., Shokouhian, M. and Ni, S. (2001). Loading estimates of Lead,
Copper, Cadmium and Zinc in urban runoff from specific sources.
Chemosphere, Vol. 44 (No. 5), pp. 997-1009.
65. Decision Lab 2000, Executive Edition, Getting Started Guide, Visual Decision
Inc. 2000.
66. Deletic, A. (1998). The first flush load of urban surface runoff. Water
Research, Vol. 32 (No. 8), pp. 2462-2470.
67. Deletic, A., Maksimovic, C. and Ivetic, M. (1997). Modelling of storm wash-
off of suspended solids from impervious surfaces. Journal of Hydraulic
Research, Vol. 35 (No. 1), pp. 99-118.
191
68. Deletic, A. and Orr, D.W. (2005). Pollution build-up on road surfaces. Journal
of Environmental Engineering, Vol. 131, pp. 49-59.
69. De Miguel, E., Llamas, J.F., Chacon, E., Berg, T., Larssen, S., Royset, O. and
Vadset, M. (1997). Origin and patterns of distribution of trace elements in
street dust: Unleaded petrol and urban lead. Atmospheric Environment, Vol. 31
(No. 17), pp. 2733-2740.
70. Dixon, L.K. and Murray, S. (1999). Bulk atmospheric deposition of nutrients
and metals in the Tampa Bay region of Florida. Proceedings of sixth Biennial
Stormwater Research and Watershed Management Conference, Tampa,
Florida, pp.1-11.
71. Dong, A., Simsiman, G.V. and Chesters, G. (1983). Particle size distribution
and phosphorous levels in soil, sediment and urban dust and dirt samples from
the Menomonee River watershed, Wisconsin, USA. Water Research, Vol. 17
(No. 5), pp. 569-577.
72. Dong, A., Chesters, G. and Simsiman, G.V. (1984). Metal composition of soil,
sediments and urban dust and dirt samples from the Menomonee river
watershed, Wisconsin, USA. Water, Air and Soil Pollution, Vol. 22, pp. 257-
275.
73. Dorney, J.R. (1986). Leachable and total phosphorus in urban street tree leaves.
Water, Air and Soil Pollution, Vol. 28, pp. 439-443.
74. Downs, J. (2003). An estimation of the annual loading of phosphorus and
sediment - Hillsdale Lake, 2002, Hillsdale Water Quality Project.
75. DR/4000, Spectrophotometer instrument manual (1999).
76. Drapper, D., Tomlinson, R.M. and Williams, P. (2000). Pollutant
concentrations in road runoff: Southeast Queensland case study. Journal of
Environmental Engineering, Vol. 126 (No. 4), pp. 313-320.
77. Drewry, J.J., Newhama, L.T.H. and Croke, B.F.W. (2009). Suspended
sediment, nitrogen and phosphorus concentrations and exports during storm-
events to the Tuross estuary, Australia. Journal of Environmental Management,
Vol. 90, pp. 879-887.
78. Droste, R.L. and Hartt, J.P. (1975). Quality and variation of pollutant loads in
urban stormwater runoff. Canadian Journal of Civil Engineering, Vol. 2 (No.
4), pp. 418-429.
192
79. Duke, L.D., Buffleben, M. and Bauersachs, L.A (1998). Pollutants in
stormwater runoff from metal plating facilities. Los Angeles, California Waste
Management, Vol. 18, pp. 25-38.
80. Duncan, H.P. (1995). A review of urban stormwater quality processes. Report
No. 95/9, Cooperative Research Centre for Catchment Hydrology.
81. Ebbert, J.C. and Wanger, R.J. (1987). Contributions of rainfall to constituent
loads in storm runoff from urban catchments. Water Resources Bulletin, Vol.
23 (No. 5), pp. 867-871.
82. Edwards, A.C. and Withers, P.J.A. (2008). Transport and delivery of
suspended solids, nitrogen and phosphorus from various sources to freshwaters.
Journal of Hydrology, Vol. 350, pp. 144- 153.
83. Egodawatta, P. (2007). Translation of small plot pollution mobilisation and
transport measurements from impermeable urban surfaces to urban catchment
scale. PhD Thesis, Queensland University of Technology, Brisbane Australia.
84. Egodawatta, P., Goonetilleke, A., Ayoko, G.A. and Thomas, E. (2006).
Understanding the interrelationship between stormwater quality and rainfall
and runoff factors in residential catchments. Proceedings of the Seventh
International Conference on Urban Drainage Modeling and the Fourth
International Conference on Water Sensitive Urban Design, Melbourne,
Australia, pp. 43-50.
85. Egodawatta, P. and Goonetilleke, A. (2006). Characteristics of pollutants build-
up on residential road surfaces. Proceedings of the 7th International Conference
on HydroScience and Engineering Philadelphia, USA September 10-13.
86. Egodawatta, P., Thomas, E. and Goonetilleke, A. (2007). Mathematical
interpretation of pollutant wash-off from urban road surfaces using simulated
rainfall. Water Research, Vol. 41 (No 13), pp. 3025-3031.
87. Egodawatta, P. and Goonetilleke, A. (2008). Understanding road surface
pollutant wash-off and underling physical processes using simulated rainfall.
Water Science and Technology, Vol. 57 (No. 8), pp. 1241-1246.
88. Ellis, K.V. (1989). Types and sources of contamination, Surface water
pollution and its control, Basingstoke, Macmillan, London, pp. 99-109.
89. Ellison, M.E. and Brett, M.T. (2006). Particulate phosphorus bioavailability as
a function of stream flow and land cover. Water Research, Vol. 40, pp. 1258-
1268.
193
90. Ellis, J.B. and Revitt, D.M. (1982). Incidence of heavy metals in street surface
sediments-solubility and grain size studies. Water, Air and Soil Pollution, Vol.
17, pp. 87-100.
91. Erpul, G., Gabriels, D. and Janssens, D. (1998). Short communication,
assessing the drop size distribution of simulated rainfall in a wind tunnel. Soil
and Tillage Research, Vol. 45 (No. 3-4), pp. 455-463.
92. Espey, W.H., Winslow, D.E. and Moore, W.L. (1969). Urban effects on the
unit hydrograph. Proceedings of the Second Water Resources Symposium,
Effects of Watershed Changes on Stream flow, pp. 215-228.
93. Espinasse, B., Picolet, G. and Chouraqui, E. (1997). Negotiation support
systems: A multi-criteria and multi-agent approach. European Journal of
Operational Research, Vol. 103, pp. 389-409.
94. Ewers. L., Clark. S., Nenrath. W., Succop. P. and Bornschein, R. (1994).
Clean-up of lead in household carpet and floor dust. American Industrial
Hygiene Association Journal, Vol. 55 (No. 7), pp. 650-657.
95. Exeter, A.N., Alberto, F., Machin, J., Galan, A. and Zaragoza, G. (1990).
Design and operation of a rainfall simulator for field studies of runoff and soil
erosion. Soil Technology, Vol. 3, pp. 385-397.
96. Fergusson, J.E. and Ryan, D. (1984). The elemental composition of street dust
from large and small urban areas related to city type, source and particle size.
Science of the Total Environment, Vol. 34, pp.101-116.
97. Fergusson, B.K. and Suckling, P.W. (1999). Changing rainfall runoff
relationships in the urbanizing Peachtree Creek Watershed Atlanta, Gerogia.
Journal of the American Water Resources Association, Vol. 26 (No. 2), pp.
313-322.
98. Filik, I.C., Emiroglu, O, Ilhan,S., Arslan, N., Yilmaz, V. and Ahiska,S. (2008).
Application of multivariate statistical techniques in the assessment of surface
water quality in Uluabat Lake, Turkey. Environment Monitoring Assess, Vol.
144, pp. 269-276.
99. Forster, J. (1996). Patterns of roof runoff contamination and their potential
implications on practice and regulation of treatment and local infiltration.
Water Science and Technology, Vol. 33 (No. 6), pp. 39-48.
100. Fulcher, G.A. (1994). Urban stormwater quality from a residential catchment.
Science of the Total Environment, Vol. 146/147, pp. 535-542.
194
101. Fujiwara, T., Yoshida, A. and Ohtoshi, K. (2005). Temporal variations of
pollutant loads during storm events in a small river basin. Journal of Water and
Environment Technology, Vol. 3 (No. 2), pp. 183-190.
102. GCCC-Web. (2008). Gold Coast City Council’s web page-
www.goldcoast.qld.gov.au
103. Ghafouri, M. and Swain, C.E. (2005). Spatial analysis of urban stormwater
quality. Journal of Spatial Hydrology, Vol. 5 (No. 1), pp. 33-46.
104. Gilbert, J. K. and Clausen, J.C. (2006). Stormwater runoff quality and quantity
from asphalt, paver and crushed stone driveways in Connecticut. Water
Research, Vol. 40, pp. 826- 832.
105. Gnecco, I., Berretta, C., Lanza, L.G. and La Barbera, P. (2005). Stormwater
pollution in the urban environment of Genoa, Italy. Atmospheric Research,
Vol. 77, pp. 60- 73.
106. Gobel, P., Dierkes, C. and Coldewey, W.G. (2006). Stormwater runoff
concentration matrix for urban areas. Journal of Contaminant Hydrology, In
press, Corrected Proof.
107. Gonzalez Bentez, J.M., Cape, J.N., Heal, M.R., Dijk, N.V. and Diez, A.V.
(2009). Atmospheric nitrogen deposition in South-East Scotland:
Quantification of the organic nitrogen fraction in wet, dry and bulk deposition.
Atmospheric Environment, Article in press.
108. Goonetilleke, A., Egodawatta, P. and Kitchen, B. (2009). Evaluation of
pollutant build-up and wash-off from selected land uses at the Port of Brisbane,
Australia. Marine Pollution Bulletin, Vol. 58 (No. 2), pp. 213-221.
109. Goonetilleke, A., Thomas, E., Ginn, S. and Gilbert, D. (2005). Understanding
the role of land use in urban stormwater quality management. Journal of
Environmental Management, Vol. 74 (No. 1), pp. 31-42.
110. Grant, S.B., Rekhi, N.V., Pise, N.R. and Reeves, R.L. (2003). A review of the
contaminants and toxicity associated with particles in stormwater runoff.
Prepared for California Department of Transportation, CTSW-RT-03-
059.73.15.
111. Graves G.A., Wan, Y. and Fike D.L. (2004). Water quality characteristics of
stormwater from land uses in South Florida. Journal of the American Water
Resources Association, Vol. 40 (No. 6), pp. 1405-1419.
195
112. Gromaire-Mertz, M.C., Garnaud, S., Gonzalez, A. and Chebbo, G. (1999).
Characterisation of urban runoff pollution in Paris. Water Science and
Technology, Vol. 39 (No. 2), pp. 1-8.
113. Grout, H., Wiesner, M. and Bottero, J. (1999). Analysis of colloidal phases in
urban stormwater runoff. Environmental Science and Technology, Vol. 33 (No.
6), pp. 831-839.
114. Gupta, K. and Saul, A.J. (1996). Specific relationships for the first flush load in
combined sewer flows, Water. Research, Vol. 30 (No. 5), pp. 1244-1252.
115. HACH DR/4000 Spectrophotometer Hand Book.
116. Haith, D.A. (1976). Land use and water quality in New York rivers. Journal of
the Environmental Engineering Division, Vol. 102 (No. EE1), pp. 1-15.
117. Hall, K.J. and Ellis, J.B. (1985). Water quality problems of urban areas. Geo
Journal, Vol. 11 (No. 3), pp. 265-275.
118. Hall, M.J. (1970). A critique of methods of simulating rainfall. Water
Resources Research, Vol. 6 (No. 4), pp. 1104-1114.
119. Han, Y., Lau, S., Kayhanian, M. and Stenstorm, M.K. (2006). Characteristics
of highway stormwater runoff. Water Environment Research, Vol. 78 (No. 12),
pp. 2377-2388.
120. Hartigan, J.P., Lumb, A.M., Smullen, J.T., Grizzard, T.J. and Alcivar, P.A.
(1978). Calibration of urban non point pollution loading models: Verification
of mathematical and physical models in hydraulic engineering. Proceedings of
the 26 th Annual Hydraulics Division Specialty Conference, University of
Maryland, Hydraulics Division, ASCE, New York, pp. 630-650.
121. Harrison, R.M.M., Smith, D.J.T. and Luhana, A. (1996). Source apportionment
of atmospheric Polycyclic Aromatic Hydrocarbons collected from an urban
location in Birmingham, U.K. Environment. Science and Technology, Vol. 30,
pp. 825-832.
122. Hart, B.T. and Grace, M.R. (2000). Sources, transformations, effects and
management of nitrogen in freshwater ecosystems, Nitrogen Workshop 2000.
123. Hatt, B.E., Fletcher, T.D., Walsh, C.J. and Taylor, S.L. (2004). The influence
of urban density and drainage infrastructure on the concentrations and loads of
pollutants in small streams. Environmental Management, Vol. 34 (No.1), pp.
112-124.
196
124. Heathwaite, A.L., Burt, T.P. and Trudgill, S.T. (1993). Overview – the nitrate
issue, Nitrate: Processes, patterns and management, Vol. 3, pp. 3-21.
125. Hendrickson, J., Trahan, N., Gordon, E. and Ouyang, Y. (2007). Estimating
relevance of organic carbon, nitrogen and phosphorus loads to a black water
river estuary. Journal of the American Water Resources Association, Vol. 43
(No. 1), pp. 264-279.
126. Herngren, L. (2005). Build-up and wash-off process kinetics of PAHs and
heavy metals on paved surfaces using simulated rainfall. PhD Thesis,
Queensland University of Technology, Brisbane, Australia.
127. Herngren, L., Goonetilleke, A. and Ayoko, G.A. (2006). Analysis of heavy
metals in road-deposited sediments. Analytica Chimica Acta, Vol. 571 (No. 2),
pp. 270-278.
128. Herngren, L., Goonetilleke, A. and Ayoko, G.A. (2005). Understanding heavy
metal and suspended solids relationships in urban stormwater using simulated
rainfall. Journal of Environmental Management, Vol. 76 (No. 2), pp. 149-158.
129. Hewitt, C.N. and Rashed, M.B. (1991). The deposition of selected pollutants
adjacent to a rural highway. Atmospheric Environment, Vol. 25, pp. 979-983.
130. Hoffman, E.J., Latimer, J.S., Hunt, C.D., Mills, G.L. and Quinn, J.G. (1985).
Stormwater runoff from highways. Water, Air and Soil Pollution, Vol. 25, pp.
349-364.
131. Hoffman, E.J., Mills, G.L., Latimer, J.S. and Quinn, J.G. (1984). Urban runoff
as a source of polycyclic aromatic hydrocarbons to coastal waters.
Environmental Science and Technology, Vol. 18, pp. 580-587.
132. Hope, D., Naegeli, M.W., Chan, A.H. and Grimm, N.B. (2004). Nutrients on
asphalt parking surfaces in an urban environment. Water, Air and Soil
pollution, Vol. 4, pp. 371-390.
133. House, M.A., Ellis, J.B., Herricks, E.E., Hvitved-Jacobsen, T., Seagar, J.,
Lijklema, L., Aalderink, H. and Clifforde, I.T. (1993). Urban drainage- impacts
on receiving water quality. Water Science and Technology, Vol. 27 (No. 12),
pp. 117-158.
134. Hranova, R., Gumbo, B., Klein, J. and Van der Zaag, P. (2002). Aspects of the
water resources management practice with emphasis on nutrients control in the
Chivero Basin, Zimbabwe. Physics and Chemistry of the Earth, Vol. 27 (No.
11-22), pp. 875-885.
197
135. Huang, J., Du, P., Ao, C., Ho, M., Lei, M., Zhao, D. and Wang, Z. (2007).
Multivariate analysis for stormwater quality characteristics identification from
different urban surface types in Macau. Bulletin of Environmental.
Contamination and Toxicology, Vol. 79 (No.6), pp. 650-654.
136. Hudson, N.W. (1963). Raindrop size distribution in high intensity storms.
Rhodesian Journal of Agricultural Research, Vol. 1, pp. 6-11.
137. Huston, R., Chan, Y.C., Gardner, T., Shaw, G. and Chapman, H. (2009).
Characterisation of urban atmospheric deposition as a source of contaminants
in urban rainwater tanks. Water Research, Vol. 43, pp. 1630-1640.
138. Hvitved-Jacobsen, T., Johansen, N.B. and Yousef, Y.A., (1994). Treatment
systems for urban and highway runoff in Denmark. Science of the Total
Environment, Vol. 146/147, pp. 499-506.
139. Jackson, D.A. (1991). Stopping rule in principal component analysis: A
comparison of heuristical and statistical approach. Ecology, Vol. 74 (No. 8),
pp. 2204-2214.
140. Jassby, A. D., Reuter, J.E., Axler, R.P., Goldman, C.R. and Hackley S.H.
(1994). Atmospheric deposition of nitrogen and phosphorus in the annual
nutrient load of lake Tahoe (California-Nevada), Water Resources Research,
Vol. 30 (No. 7), pp. 2207-2216.
141. Jartun, M., Ottesen, R.T., Steinnes, E. and Voldena, T. (2008). Runoff of
particle bound pollutants from urban impervious surfaces studied by analysis of
sediments from stormwater traps. Science of the Total Environment, Vol. 396,
pp. 147-163.
142. Kato, T., Kuroda, H. and Nakasone, H. (2009). Runoff characteristics of
nutrients from an agricultural watershed with intensive livelock production.
Journal of Hydrology, Vol. 368 (No. 1- 4), pp. 79-87.
143. Kayhanian, M., Suverkropp, C., Ruby, A. and Tsay, K. (2006).Characterization
and prediction of highway runoff constituent event mean concentration. Journal
of Environmental Management, Vol. 85 (No. 2), pp. 279-295.
144. Kayhanian, M., Suverkropp, C., Ruby, A. and Tsay, K. (2007).
Characterization and prediction of highway runoff constituent event mean
concentration. Journal of Environment Management. Journal of Environmental
Management, Vol. 85, pp. 279-295.
198
145. Kayhanian, M., Stransky, C., Bay, S., Lau, S.L. and Stenstrom, M.K. (2008).
Toxicity of urban highway runoff with respect to storm duration. Science of
The Total Environment, Vol. 389 (No. 2-3), pp. 386-406.
146. Khalil, W.A., Goonetilleke, A., Kokot, S. and Carroll, S. (2004). Use of
chemometrics methods and multicriteria decision-making for site selection for
sustainable on-site sewage effluent disposal. Analytica Chimica Acta, Vol. 506,
pp. 41-56.
147. Keller, H.R., Massart, D.L. and Brans, J.P. (1991). Multicriteria decision
making: a case study. Chemometrics and Intelligent laboratory systems, Vol.
11, pp. 175-189.
148. Kelly, J., Thornton, I. and Simpson, P.R. (1996). Urban geochemistry: A study
of the influence of anthropogenic activity on the heavy metal content of soils in
traditionally industrial and non- industrial areas of Britain. Journal of Water
Science and Technology, Vol. 22, pp. 363-370.
149. Kibler, D.F. (1982). Urban hydrology and the stormwater problem.
Washington, DC. American Geophysical Union.
150. Kinnell, P.I. (1987). Rainfall Energy in Eastern Australia: Intensity kinetic
energy relationships for Canberra, A.C.T. Australian Journal of Soil Research,
Vol. 25, pp. 547-553.
151. Kim, L., Ko, S., Jeong, S. and Yoon, J. (2007). Characteristics of washed-off
pollutants and dynamic EMCs in parking lots and bridges during a storm.
Science of The Total Environment, Vol. 376 (No. 1-3), pp. 178-184.
152. Kolasa, K.V. (1999). Little lake Jackson stormwater runoff analysis.
Proceedings of Sixth Bienniul Stormwater Research and Watershed
Management Conference, pp. 41-59.
153. Kokot, S. and Yang, P. (1995). Comparision of thermogravimetric and
differential scanning calorimetric results for cellulosic fabrics by
chemometrics. Analytica Chemica Acta, Vol. 304, pp. 297-305.
154. Kokot, S. and Phuong, T.D. (1999). Elemental content of Vietnamese rice, Part
2. Multivariate data analysis. Analyst, Vol. 124, pp. 561-569.
155. Konrad, J.G., Chesters, G. and Bauer, K.W. (1978). Memomonee River basin,
Wisconsin: Summary pilot watershed report. PLUARG Technical Report
Series.
199
156. Lager, J.A., Shubinski, R.P. and Russell, L.W. (1971). Development of
simulation model for stormwater management. Journal of Water Pollution
Control Federation, Vol. 43 (No. 12), pp. 2424-2435.
157. Lahdelma, R., Miettinen, K. and Salminen. (2003). Ordinal criteria in
stochastic multicriteria acceptability analysis (SMMA). European Journal of
Operational Research, Vol. 147, pp. 117-127.
158. Larkin, G.A. and Hall, K.J., (1998). Hydrocarbon pollution in the Brunette
River Watershed. Water Quality Research, Journal of Canada, Vol. 33, pp. 73-
94.
159. Latimer, J.S., Hoffman, E.J., Hoffman, G., Fasching, J.L. and Quinn, J.G.
(2004). Sources of petroleum hydrocarbons in urban runoff. Water, Air and
Soil Pollution, Vol. 52 (No. 1-2), pp. 1-21.
160. Lau, S. and Stenstrom, M. K. (2005). Metals and PAHs adsorbed to street
particles. Water Research, Vol. 39 (No. 17), pp. 4083-4092.
161. Laws, J.O. (1941). Measurements of the fall velocity of water drops and
raindrops. Transactions of The American Geophysical Union, Vol. 22, pp. 709-
721.
162. Lee, J.G. and Bang, K.W. (2000). Characterization of urban stormwater runoff.
Water Research, Vol. 34 (No. 6), pp. 1773-1780.
163. Lewis, W.M. (1981). Precipitation chemistry and nutrient loadings in the
runoff waters from a major rural highway. The Science of the Total
Environment, Vol. 235, pp. 143-150.
164. Li, Y., Lau, S., Kayhanian, M. and Stenstrom, M.K. (2005). Particle size
distribution in highway runoff. Journal of Environmental Engineering, Vol.
131 (No. 9), pp 1267- 1276.
165. Li, Y., Xiang, L., Wang, C., Tian, P., Song, K., Liu, J. and Yang, Z. (2008).
Characteristics of particles on urban impervious surfaces: Heavy metals, TOC
and specific surface area. Proceedings of 12th International Conference on
Intergrated Diffuse Pollution Management (IWA DIPCON 2008).
166. Librando, V., Magazzu, G. and Puglisi, A. (1995). Multivariate micropollutants
analysis in marine waters. Water Science Technology, Vol. 32 (No. 9-10), pp.
341-348.
167. Lim, M.C.H., Ayoko, G.A., Morawska, L., Ristovski, Z.D., Jayaratne, E.R. and
Kokot, S. (2006). A comparative study of the elemental composition of the
200
exhaust emissions of cars powered by liquefied petroleum gas and unleaded
petrol. Atmospheric Environment, Vol. 40, pp. 3111-3122.
168. Lim, M.C.H., Ayoko, G.A. and Morawska, L. (2005).Characterization of
elemental and polycyclic aromatic hydrocarbon compositions of urban air in
Brisbane. Atmospheric Environment, Vol. 39 (No. 3), pp. 463-476.
169. Line, D.E., White, N.M., Osmond, D.L., Jennings, G.D. and Mojonnier, C.B.
(2002). Pollutant export from various land uses in the upper neuse river basin,
Water Environment Research, Vol. 74, pp. 100-108.
170. Liu, D.H.F., Liptak, B.G. and Bouis, P.A. (1997). Non point source pollution.
Environmental Engineers Hand Book, D. H. F. Liu. (ed.) Lewis Publishers,
New York, pp. 1115-1119.
171. Line, D.E., White, N.M., Osmond, D.L., Jennings, G.D. and Mojonnier, C.B.
(2002). Pollutant export from various land uses in the Upper Neuse river basin.
Water Environment Research, Vol. 74 (No. 1), pp. 100-108.
172. Line, D.E., Wu, J., Arnold, J.A., Jennings, G.D. and Rubin, A.R. (1997). Water
quality of first flush runoff from 20 industrial sites. Water Environment
Research, Vol. 69, pp. 305-310.
173. Loch, R.J., Robotham, B.G., Zeller, L., Masterman, N., Orange, D.N., Bridge,
B.J., Sheridan, G. and Bourke, J. J. (2001). A multi-purpose rainfall simulator
for field infiltration and erosion studies. Australian Journal of Soil Research,
Vol. 39, pp. 599-610.
174. Lovett, G.M. (1994). Atmospheric deposition of nutrients and pollutants in
North America: An Ecological Perspective. Ecological Applications, Vol. 4
(No. 4), pp. 629-650.
175. Makepeace, D.K., Smith, D.W. and Stanley, S.J. (1995). Urban stormwater
quality: Summary of contaminant data. Critical Reviews in Environmental
Science and Technology, Vol. 25 (No. 2), pp. 93-139.
176. Malmquist, P., (1978). Atmospheric fallout and effects of urban stormwater
and snow. Progress in Water Technology, Vol. 10, pp. 495-505.
177. Malvern Instrument Ltd. (1997). Sample dispersion and refractive index guide.
MAN 0079, U.K.
178. Mallin, M.A., Johnson, V.L. and Ensign, S.H. (2008). Comparative impacts of
stormwater runoff on water quality of an urban, a suburban and a rural stream,
Environ Monitoring Assess.
201
179. Mance, G., Harman, M.M.I. (1978). The quality of urban storm water runoff.
Urban Storm Drainage, Proceedings of the International Conference held at the
University of Southamption, Southamption, (ed.) Helliwell, P.R., pp. 603-617.
180. Marsalek, J., Brownlee, B., Mayer, T., Lawal, S. and Larkin, G.A. (1997).
Heavy metals and PAHs in stormwater runoff from the Skyway Bridge,
Burlington, Ontario. Water Quality Research Journal of Canada, Vol. 32, (No.
4), pp. 815-827.
181. Martin, C., Ruperd, Y. and Legret, M. (2007). Urban stormwater drainage
management: The development of a multicriteria decision aid approach for best
management practices. European Journal of Operational Research, Vol. 181
(No. 1), pp. 338-349.
182. McCorquodale, J.A., Barbe, D. and Wang, Y. (2002). Urban rainwater
composition study. Best Modeling Practices for Urban Water Systems, James,
W. (ed.), Vol. 10 in the Monograph Series, pp. 57-70.
183. Mesner, N. and Geiger, J. (2005). Understanding your watershed, Water
Quality, NR/WQ/2005-19.
184. Michael, P. and Timpe, P.E. (1999). Impacts of atmospheric deposition on
stormwater quality. Proceedings of the Sixth Biennial Stormwater Research
Watershed Management Conference, Tampa, Florida, pp. 31-40.
185. Mikkelsen, P.S., Weyer, G., Berry, C., Walden, Y., Colandini, V., Poulsen, S.,
Grotehusmann, D. and Rohlfing, R. (1994). Pollution from urban stormwater
infiltration. Water Science and Technology, Vol. 29 (No.1-2), pp.293-302.
186. Moshkovich, H.M., Schellenberger, R.E. and Olson, D.L. (1998). Data
influences the result more than preferences: Some lessons from implementation
of multiattribute techniques in a real decision task, Decision Support Systems,
Vol. 22, pp. 73-84.
187. Nelson, E.J. and Booth, D.B. (2002). Sediment sources in an urbanising, mixed
land-use watershed. Journal of Hydrology, Vol. 264, pp. 51-68.
188. Nguyen, T.A., Kokot, S., Ongarato, D.M. and Wallace, G.G. (1999). The use
of chronoamperometry and chemometrics for optimization of conducting
polymer sensor arrays. Electroanalysis, Vol. 11 (No. 8), pp. 1327-1332.
189. Novotny, V. and Chesters, B. (1981). Handbook on nonpoint pollution: sources
and management. Van Nostrand and Reinhold Company, New York.
202
190. Novotny, V., Sung, H.M., Bannerman, R. and Baum, K. (1985). Estimating
nonpoint pollution from small urban watersheds. Journal Water Pollution
Control Federation, Vol. 57, pp. 339-348.
191. Organisation for Economic Co-operation and Development (OECD), (1986).
Control of water pollution from urban runoff. Oxford University Press, New
York.
192. Paerl, H. W., Pinckney, J.L., Fear, J.M. and Peierls, B.L. (1998). Ecosystem
responses to internal and watershed organic matter loading: consequences for
hypoxia in the eutrophying Neuse river estuary, North Carolina, USA. Marine
Ecology-Progress Series, Vol. 166, pp 17-25.
193. Panuska, J.C. and Lillie, R.A. (1995). Phosphorus loadings from Wisconsin
watersheds: Recommended phosphorus export coefficients for agricultural and
forested watersheds. Research Management Findings, Wisconsin Department
of Natural Resources, Report No. 38.
194. Parks, S.J. and Baker, L.A. (1997). Sources and organic carbon in an Arizona
river- reservoir system. Water Research, Vol. 31 (No. 7), pp. 1751-1759.
195. Pitt, R. (1979). Demonstration of nonpoint pollution abatement through
improved street cleaning practices. U.S. Environmental Protection Agency,
Report No. EPA/600/2-79-161.
196. Pitt, R. (1987). Small storm urban flow and particulate washoff contributions to
outfall discharges. University of Wisconsin-Thesis, Madison.
197. Pitt, R., Bannerman, R., Clark, S. and Williamson, D. (2004). Sources of
Pollutants in Urban Areas (Part 2) – Recent Sheetflow Monitoring. Effective
Modeling of Urban Water Systems, James, W., Irvine, K.N., McBean, E.A. and
Pitt, R.E. (eds.), Monograph 13, pp. 485-506.
198. Pitt, R., Field, R., Lalor, M. and Brown, M. (1995). Urban stormwater toxic
pollutants: Assessment, sources and treatability. Water Environment Research,
Vol. 67 (No. 3), pp. 260-275.
199. Pitt, R. and McLean, J. (1986). Toronto area watershed management strategy
study. Humber River Pilot Watershed Project, Ontario Ministry of the
Environment, Toronto, Ontario.
200. Polkowska, Z., Grynkiewicz, M., Zabiegala, B. and Namiecenik, J. (2001).
Levels of pollutants in runoff water from roads with high traffic intensity in the
203
city of Gdansk. Poland, Polish Journal of Environmental Studies, Vol. 10 (No.
5), pp. 351-363.
201. Puckett, L.J. (1995). Identifying the major sources of nutrient water pollution.
Environmental Science and Technology, Vol. 29 (No. 9), pp. 408-414.
202. Purcell, D.E., Leonard, G.J., O’Shea, M. and Kokot, S. (2005). A
chemometrics investigation of sugarcane plant properties based on the
molecular composition of epicuticular wax. Chemometrics and Intelligent
Laboratory Systems, Vol. 76, pp. 135-147.
203. Quinton, J.N., Catt, J.A. and Hess, T. M. (2001). The selective removal of
phosphorus from soil: Is event size important?. Journal of Environ. Qual. Vol.
30, pp. 538-545.
204. Qureshi, A.A. and Dutka, B.J. (1979). Microbiological studies on the quality of
urban stormwater runoff in Southern Ontario. Canada. Water Research, Vol. 13
(No. 10), pp. 977-985.
205. Randall J.C, Michael E. and Barrett (1998). Evaluation of methods for
estimating stormwater pollutant loads. Water Environment Research, Vol. 70,
pp. 1295-1302.
206. Randoll, C.W., Grizzard, T.J., Helsel, D.R. and Griffin, D.M. (1981).
Comparison of pollutant mass loads in precipitation and runoff in urban areas.
Proceedings of the Second International Conference on Urban Storm Drainage,
Urbana, Illinois, USA, Yen, B.C. pp. 29-38.
207. Rao, R.A. and Delleur, J.W. (1974). Instantaneous unit hydrograph, peak
discharge and time lags in urban basins. Hydrological Science Bulletin, Vol.
19. pp. 185-198.
208. Ren,Y., Xiaoke, W., Zhiyun, O., Hua, Z., Xiaonan, D. and Hong, M. (2008).
Stormwater runoff quality from different surfaces in an urban catchment in
Beijing, China. Water Environment Research, Vol. 80 (No. 8), pp. 719-724.
209. Rickson, R.J. (2001). Experimental techniques for erosion studies: Rainfall
simulation. http://www.silsoe.cranfield.ac.uk/staff/cv/rainfall_simulation. pdf.
210. Robert, C. and Kondolf, G.M. (1993). Land use change in California, USA:
Nonpoint source water quality impacts. Journal of Environmental
Management, Vol. 17 (No. 4), pp. 453-460.
211. Roberts, A. R. and Withers, P. (2004). Statisti XL, Verison 1.4.
204
212. Robertson, D.J., Taylor K.G. and Hoon, S.R. (2003). Geochemical and mineral
magnetic characterisation of urban sediment particulates, Manchester, UK. .
Applied Geochemistry, Vol. 18 (No. 2), pp. 269-282.
213. Robien,A., Striebel, T. and Herrmann, R. (1997). Modeling of dissolved and
particle-bound pollutants in urban street runoff. Water Science and
Technology, Vol. 36. (No. 8-9), pp. 77-82.
214. Robson, M., Spence, K. and Beech, L. (2006). Stream quality in a small
urbanised catchment. Science of the Total Environment, Vol. 357, pp. 194-207.
215. Rogge, W.F., Hlldemann, L.M., Mazurek, M.A. and Cass, G.R. (1993).
Sources of fine organic aerosol-road dust, tire debris and organometallic brake
lining dust: Roads as sources and sinks. Environmental Science and
Technology, Vol. 27 (No. 9), pp. 1892-1904.
216. Rosewell, C.J. (1986). Rainfall kinetic energy in eastern Australia. Journal of
Climate and Applied Meteorology, Vol. 25, pp. 1695-1701.
217. Sabin, L.D., Lim, J.H., Stolzenbach, K.D. and Schiff, K.C. (2005).
Contribution of trace metals from atmospheric deposition to stormwater runoff
in a small impervious urban catchment. Water Research, Vol. 39 (No. 16), pp.
3929-3937.
218. Sabin, L.D., Lim, J.H., Venezia, M.T., Winer, A.M, Schiff, K.C. and
Stolzenbach, K.D. (2006). Dry deposition and resuspension of particle-
associated metals near a freeway in Los Angeles. Atmospheric Environment,
Vol. 40, pp. 7528-7538.
219. Sakai, A., Sumiyama, M. and Tanaka, K. (1996). Analysis of non-point source
pollutant runoff processes. Proceedings of urban wet weather pollution:
Controlling sewer overflows and stormwater runoff. Specialty Conf., Quebec
City, PQ, Canada, pp. 11-63.
220. Salles, C., Poesen, J. and Sempere-Torres, D. (2002). Kinetic energy of rain
and its functional relationship with intensity. Journal of Hydrology, Vol. 257,
pp 256-270.
221. Sansalone, J.J., Buchberger, S.G. and Al-Abed, S.R. (1996). Fractionation of
heavy metals in pavement runoff. Science of the Total Environment, Vol.
189/190, pp. 371-378.
205
222. Sansalone, J.J. and Glenn, D.W. (2000). Temporal variations in heavy metal
partitioning and loading in urban highway pavement sheet flow: Implications
for in situ treatment design, Transportation Research Record, pp. 100-111.
223. Sansalone, J.J., Koran, J.M., Smithson, J.A. and Buchberger, S.G. (1998).
Physical characteristics of urban roadway solids transported during rain events,
Journal of Environmental Engineering, Vol. 124 (No. 5), pp.427-440.
224. Sansalone J.J. and Tittlebaum, M.E. (2001). Storm water transport of
particulate matter from elevated urban transportation corridors into waterways
of Louisiana: The role of partitioning and implications for treatment, Lousiana.
Water Resources Research Institute, Annual Technical Report.
225. Sartor, J.D. and Boyd, B.G. (1972). Water pollution aspects of street surface
contaminants. U.S. Environmental Protection Agency, Washington, DC, USA,
Report No. EPA- R2-72/081.
226. Sartor, J.D., Boyd, B.G. and Agardy, F.J. (1974). Water pollution aspects of
street surface contaminants. U.S. Environmental Protection Agency,
Washington, DC, USA, Report No. EPA- R2-72/081.
227. Schoonover, J.E. and Lockaby, B.G. (2006). Land cover impacts on stream
nutrients and fecal coliform in the lower Piedmont of West Georgia. Journal of
Hydrology, Vol. 331, pp. 371-382.
228. Schueler, T.R. (1994). The importance of imperviousness. Watershed
Protection Techniques, Vol. 1 (No. 3), pp. 100-111.
229. Seaburn, G.E. (1969). Effect of urban development on direct runoff to East
Meadow Brook, Nassau Country, Long Island, New York, U.S. Geological
Survey Professional Paper 627-B, pp. B1-14.
230. Sekhar, M.C. and Raj, P.A. (1995). Land use- Water quality modelling: A case
study. Journal of Water Science and Technology, Vol. 31 (No. 8), pp. 383-386.
231. Settle, S., Goonetilleke, A. and Ayoko, G.A. (2007). Determination of
surrogate indicators for phosphorus and solids in urban stormwater:
Application of multivariate data analysis techniques. Water, Air and Soil
Pollution, Vol. 182 (No. 1-4), pp. 149-161.
232. Shaheen, D.O. (1975). Contributions of urban roadway usage to water
pollution. Municipal Pollution Control Division, Office of Research and
Development, U.S. Environmental Protection Agency, Washington, DC, USA,
Report No. EPA-600/ 2-75/004.
206
233. Sharpley, A.N. and Smith, S.J. (1990). Phosphorus transport in agricultural
runoff: The role of soil erosion. Soil Erosion on Agricultural Land, Boardman,
J., Foster, I.D.L. and Dearing J.A. (eds.), pp. 351-366.
234. Shaver, E. (1996). Sand filter design for stormwater treatment, Water-Fall, Vol.
4, pp. 31-36.
235. Shaw, S.B., Walter M.T. and Steenhuis, T.S. (2006). A physical model of
particulate wash-off from rough impervious surfaces. Journal of Hydrology,
Vol. 327, pp. 618-626.
236. Shi, W. G. (1990). Street surface sediment accumulation and pollution
characteristics: For example in Chengdu. Environmental Science, Vol. 3,
pp.18-23.
237. Shigaki, F., Sharpley, A. and Prochnow, L. (2007). Rainfall intensity and
phosphorus source effects on phosphorus transport in surface runoff from soil
trays. Science of the Total Environment, Vol. 373, pp. 334-343.
238. Shimadzu TOC-5000A with Autosampler, www.labx.com.
239. Shivalingaiah, B. and James, W. (1984a). Algorithms for build up, wash off
and routing pollutants in urban runoff. Proceedings of the Third International
Conference on Urban Storm Drainage, Goteborg, Sweden, ed. Balmer, P.,
Malmquist, P. and Sjoberg, A., Vol. 4, pp. 1445-1455.
240. Shivalingaiah, B. and James, W. (1984b). Applied algorithms for build up,
wash off and transport in stormwater: Water quality evolution within the
hydrological cycles of watersheds. Proceedings of the Canadian Hydrology
Symposium, Vol. 2 (No. 15), pp. 895-914.
241. Shlens, J. (2005). A Tutorial on principal component analysis. Systems
Neurobiology laboratory, University of California, San Diego.
242. Shuttlewroth, A.J (1986). State of rivers and sewers in Britain: Is there a
pollution problem?. Public Health Engineer. Vol. 14 (No. 3), pp. 49-50.
243. Stevenson, H. and Wyman, B. (1991). The facts on file dictionary of
environmental sciences. Facts On File Limited, Oxford, UK.
244. Straub, C.P. (1989). Practical Handbook of Environmental Control, CRC Press,
Inc., Boca Raton, FL.
245. Strynchuk, J., Royal, J. and England, G.P.E. (1999). Grass and leaf
decomposition and nutrient release study under wet conditions. Proceedings of
207
Sixth Biennial Stormwater Research and Watershed Management Conference,
Tampa, Florida, pp. 12-20.
246. Stutter, M.L., Langan, S.J. and Cooper, R.J. (2008). Spatial contribution of
diffuse inputs and within-channel processes to the form of stream water
phosphorus over storm events, Journal of Hydrology, Vol. 350 (No. 3-4), pp.
203-214.
247. Sonzogni, W.C., Chesters, G., Coote, D.R., Jeffs, D.N., Konrad, J.C., Ostry,
R.C. and Robinson, J.B. (1980). Pollution from land runoff. Environmental
Science and Technology, Vol.14 (No. 2), pp.148-153.
248. Taebi, A. and Droste. R.L. (2004). Pollution loads in urban runoff and sanitary
wastewater. Science of the Total Environment, Vol. 327, pp. 175 -184.
249. Tai, Y. L. (1991). Physical and chemical characterisation of street dust and dirt
from urban areas, Pennsylvania State University- Thesis. USA.
250. Tao, S. and Lin, B. (2000). Water soluble organic carbon and its measurement
in soil and sediment. Water. Research. Vol. 34 (No. 5), pp. 1751-1755.
251. Taylor, G.D., Fletcher, T.D., Wong, T.H.F., Breen, P.F. and Duncan, H.P.
(2005). Nitrogen composition in urban runoff: implications for stormwater
management. Water Research, Vol. 39, pp. 1982-1989.
252. Thomas, P.R. and Greene, G.R. (1993). Rainwater quality from different roof
catchments. Water Science and Technology, Vol. 28, pp. 291-299.
253. Thomos, N.P., Samir, A. and Swaify, E. L. (1989). Construction and
calibration of rainfall simulator. Journal of Agricultural Engineering Research,
Vol. 43, pp. 1-9.
254. Tiefenthaler, L.L. and Kenneth, C.S. (2001). Effects of rainfall intensity and
duration on first flush of stormwater pollutants. SCCWRP Annual Report
2001-2002.
255. Tomasko, D.A., Corbett, C.A., Greening, H.S. and Raulerson, G.E. (2005).
Spatial and temporal variation in seagrass coverage in Southwest Florida:
Assessing the relative effects of anthropogenic nutrient load reductions and
rainfall in four contiguous estuaries. Marine Pollution Bulletin, Vol. 50, pp.
797-805.
256. Tyler, B.J., Rayala, G. and Castnera, D.G. (2007). Multivariate analysis
strategies for processing ToF-SIMS images of biomaterials. Biomaterials Vol.
28, pp. 2412-2423.
208
257. UN. (2004). UN report says world urban population of 3 billion today expected
to reach 5 billion by 2030. Press release- Department of Economic and Social
Affairs, Population Division, POP/899.
258. US EPA, (1983). Method 365.4: Phosphorous, total (Colorimetric, automated,
block digester AA II). US EPA Environmental Protection Agency,
Washington, DC. EPA/600/4-79/020.
259. US EPA, (1993). Natural wetlands and urban stormwater: Potential impacts
and management. U.S. Environmental Protection Agency.
260. US EPA, (1999). Protocol for developing nutrient TMDLs. U.S. Environmental
Protection Agency, Report No. EPA /841/B-99/007.
261. US EPA, (2005). National management measures to control nonpoint source
pollution from urban areas. U.S. Environmental Protection Agency, Report No.
EPA /841/B-05/004.
262. Uusitalo, R., Yli-Halla, M. and Turtola, E. (2000). Suspended soil as a source
of potentially bioavailable phosphorus in surface runoff. Water Research, Vol.
34 (No. 9), pp. 2477-2482.
263. Van Dijk, A. I. J. M., Bruijnzeel, L. A. and Rosewell, C. J. (2002). Rainfall
Intensity- Kinetic energy relationships: A critical literature appraisal, Journal of
Hydrology, Vol. 261, pp. 1-23.
264. Van Meter, P.C., Mahler, B.J. and Furlong, E.T. (2000). Urban sprawl leaves
its PAH signature. Environmental Science and Technology, Vol. 34 (No. 19),
pp. 4064-4070.
265. Van Meter, P.C. and Mahler, B.J. (2003). The contribution of particles washed
from rooftops to contaminant loading to urban streams. Chemosphere, Vol. 52,
(No. 10), pp. 1727-1741.
266. Vaze, J., Chiew, F.H.S., Suryadi, L. and Khanal, K. (2000). Pollutant
accumulation on an urban road surface. Urban Water, Vol. 4, pp. 265-270.
267. Vaze, J. and Chiew, F.H.S. (2002). Experimental study of pollutant
accumulation on an urban road surface. Urban Water, Vol. 4 (No. 4), pp. 379-
389.
268. Vaze, J. and Chiew, F.H.S. (2004). Nutrient loads associated with different
sediment sizes in urban stormwater and surface pollutants. Journal of
Environmental Engineering, Vol. 130 (No. 4), pp. 391-396.
209
269. Vermette, S.J., Irvine, K.N. and Drake, J.J. (1991). Temporal variability of the
elemental composition in urban street dust. Environmental Monitoring and
Assessment, Vol. 18, pp. 69-77.
270. Vice, R.B., Guy, H.P. and Ferguson, G.E. (1969). Sediment movement in an
area of suburban highway construction, Sacott Run basin, Fairfax Country,
Virginia. Geological Survey Water Supply Paper 1591-EUSGS, Reston, VA.
271. Waananen, A.O. (1969).Urban effects of water yield. Proceedings of the
Second Water Resources Symposium, Effects of watershed changes on stream
flow, Austin.
272. Waananen, A.O., Park, M. and Calif. (1961). Hydrologic effects of urban
growth - Some characteristics of urban runoff. US Geological Survey
Professional Paper 424-C, pp. C353-356.
273. Walker, T.A. and Wong, T.H.F. (1999). Effectiveness of street sweeping for
stormwater pollution control, Technical Report No. 99/8, Cooperative Research
Centre for Catchment Hydrology.
274. Walker, W.J., McNutt, R.P. and Maslanka, C.A.K. (1999). The potential
contribution of urban runoff to surface sediments of the Passaic River: Sources
and chemical characteristics. Chemosphere, Vol.38 (No. 2), pp. 363-377.
275. Waneilista, M.P., Yousef, Y.A., McLellon, W.M. (1977). Non point source
effects on water quality. Journal of Water Pollution Control Federation, Vol. 49
(No. 3), pp. 441-451.
276. Weeks, C.R. (1981). Pollution in urban stormwater runoff. Second
International Conference on Urban Storm Drainage, pp. 39-47.
277. Westco Scientific Instruments, Inc. SmartChem 140 Method, Operator Hand
Book.
278. White, C.S. (1975). Factors influencing natural water quality and changes
resulting from land- use practices. Dept. of Biology, University of New
Mexico, U.S.A.
279. Wilkinson, R. (1954). The quality of rainfall runoff water from a housing
estate. Journal of Public Health Engineering.
280. Wit, M.D. and Bendoricchio, G. (2001). Nutrient fluxes in the Po basin. The
Science of Total Environment, Vol. 273 (No. 1-3), pp. 147-161.
210
281. Wong, T., Breen, P. and Lloyd, S. (2000). Water sensitive road design-Design
options for improve stormwater quality of road runoff. Technical Report
No.00/1, Cooperative research Centre, Catchment Hydrology.
282. Wu, M.L., Wang, Y.S. (2007). Using chemometrics to evaluate anthropogenic
effects in Daya Bay, China. Estuarine, Coastal and Shelf Science, Vol.72 (No.
4), pp. 732-742.
283. www.crwa.org/alert/wmapermits/CRWAWhitePaper
284. Xian, G., Crane, M. and Su, J. (2006). An analysis of urban development and
its environmental impact on the Tampa Bay watershed. Journal of
Environmental Management, In Press, Corrected Proof.
285. Yamada, K., Umehara, T. and Ichiki, A. (1993). Study on statistical
characteristics of nonpoint pollutants deposited in urban areas. Water Science
and Technology, Vol. 28 (No. 3-5), pp. 283-290.
286. Yaziz, M.I., Gunting, H., Sapari, N. and Ghazali, A.W. (1989). Variations in
rainwater quality from roof catchments. Water resources, Vol. 23 (No. 6), pp.
761-765.
287. Yen, B.C. (1981). Urban storm drainage. Proceedings of the Second
International Conference held in Urban-Champaign, Illinois, USA.
288. Yusop, Z., Douglas, I. and Nik, A.R. (2006). Export of dissolved and
undissolved nutrients from forested catchments in Peninsular Malaysia, Forest
Ecology and Management, Vol. 224, pp. 26-44.
289. Zafra, C.A., Temprano, J. and Tejero, I. (2008). Particle size distribution of
accumulated sediments on an urban road in rainy weather. Environmental
Technology, Vol. 29, pp. 571-582.
290. Zhao, J., Shan, B. and Yin, C. (2007). Pollutant loads of surface runoff in
Wuhan city Zoo, an urban tourist area. Journal of Environmental Sciences, Vol.
19, pp. 464-468.
291. Zoppou, C. (2001). Review of urban stormwater models. Environmental
Modelling and Software, Vol. 16 (No. 3), pp. 195-231.
211
APPENDIX 1
CALIBRATION OF RAINFALL SIMULATOR
212
213
Calibration of the rainfall simulator for the rainfall intensity and uniformity of rainfall
The calibration of the rainfall simulator for rainfall intensity was done by measuring
the average depth of water collected within a five minute duration in an array of
containers placed in a grid pattern exposed to the simulated rainfall. The containers
were positioned as shown in Figure A.
Figure A - The position of the containers
A simulated rainfall was generated with a known setting of the control box for a 5
minute of duration. Then the amount of water collected in all the containers was
measured and the rainfall intensity in terms of depth of water per unit time (mm/hr)
was calculated. Table A shows the volume of water collected and calculated average
rainfall intensity. A specimen calculation for control box setting A-2 and for the
position a 2 is shown below.
Specimen calculation for rainfall intensity
For container position a2
Volume of water collected = 12 ml
Area of the container opening = 6162.55mm2
Depth of water = 12x1000/6162.5
Depth of water per hour = 12x1000x60 /6162.5x5
= 23.4 mm/hr
214
Table A - Calculation of rainfall intensity from the measured water volume
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
A-1 1 10 15 12 4 19.5 29.2 23.4 7.8 2 12 16 16 6 23.4 31.2 31.2 11.7 3 14 20 14 10 27.3 38.9 27.3 19.5 4 10 16 14 10 19.5 31.2 27.3 19.5 5 8 12 14 8 15.6 23.4 27.3 15.6
A-2 1 10 16 22 10 19.5 31.2 42.8 19.5 2 12 20 20 10 23.4 38.9 38.9 19.5 3 12 22 28 14 23.4 42.8 54.5 27.3 4 10 20 20 14 19.5 38.9 38.9 27.3 5 8 16 28 12 15.6 31.2 54.5 23.4
A-3 1 14 40 30 10 27.3 77.9 58.4 19.5 2 14 26 24 12 27.3 50.6 46.7 23.4 3 16 30 30 20 31.2 58.4 58.4 38.9 4 12 26 22 18 23.4 50.6 42.8 35.1 5 10 20 28 16 19.5 38.9 54.5 31.2
A-4 1 18 32 38 12 35.1 62.3 74.0 23.4 2 16 26 26 12 31.2 50.6 50.6 23.4 3 22 34 38 20 42.8 66.2 74.0 38.9 4 14 26 24 18 27.3 50.6 46.7 35.1 5 10 20 32 14 19.5 38.9 62.3 27.3
A-5 1 16 30 38 12 31.2 58.4 74.0 23.4 2 16 28 28 14 31.2 54.5 54.5 27.3 3 20 32 34 24 38.9 62.3 66.2 46.7 4 14 28 26 20 27.3 54.5 50.6 38.9 5 10 22 30 20 19.5 42.8 58.4 38.9
A-6 1 12 18 14 8 23.4 35.1 27.3 15.6 2 14 24 22 12 27.3 46.7 42.8 23.4 3 16 24 24 20 31.2 46.7 46.7 38.9 4 10 20 22 16 19.5 38.9 42.8 31.2 5 8 18 24 16 15.6 35.1 46.7 31.2
B-1 1 12 18 14 8 23.4 35.1 27.3 15.6 2 10 20 18 8 19.5 38.9 35.1 15.6 3 14 30 36 14 27.3 58.4 70.1 27.3 4 10 18 18 10 19.5 35.1 35.1 19.5 5 8 16 22 10 15.6 31.2 42.8 19.5
B-2 1 12 20 16 6 23.4 38.9 31.2 11.7 2 12 20 20 10 23.4 38.9 38.9 19.5 3 16 28 26 12 31.2 54.5 50.6 23.4 4 12 20 20 10 23.4 38.9 38.9 19.5 5 10 16 24 10 19.5 31.2 46.7 19.5
215
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
B-3 1 10 18 16 10 19.5 35.1 31.2 19.5 2 12 20 18 12 23.4 38.9 35.1 23.4 3 16 30 24 14 31.2 58.4 46.7 27.3 4 10 20 18 10 19.5 38.9 35.1 19.5 5 8 16 22 5 15.6 31.2 42.8 9.7
B-4 1 14 38 18 8 27.3 74.0 35.1 15.6 2 14 26 24 12 27.3 50.6 46.7 23.4 3 16 30 26 20 31.2 58.4 50.6 38.9 4 12 24 24 16 23.4 46.7 46.7 31.2 5 10 22 32 14 19.5 42.8 62.3 27.3
B-5 1 14 30 26 10 27.3 58.4 50.6 19.5 2 16 28 24 14 31.2 54.5 46.7 27.3 3 18 38 32 22 35.1 74.0 62.3 42.8 4 14 26 26 18 27.3 50.6 50.6 35.1 5 10 22 28 18 19.5 42.8 54.5 35.1
B-6 1 16 68 28 10 31.2 132.4 54.5 19.5 2 20 28 26 14 38.9 54.5 50.6 27.3 3 22 32 30 24 42.8 62.3 58.4 46.7 4 16 30 28 16 31.2 58.4 54.5 31.2 5 12 22 28 18 23.4 42.8 54.5 35.1
C-1 1 12 18 16 8 23.4 35.1 31.2 15.6 2 14 20 16 8 27.3 38.9 31.2 15.6 3 14 30 22 12 27.3 58.4 42.8 23.4 4 12 18 18 10 23.4 35.1 35.1 19.5 5 8 14 18 10 15.6 27.3 35.1 19.5
C-2 1 12 28 18 5 23.4 54.5 35.1 9.7 2 12 18 16 5 23.4 35.1 31.2 9.7 3 16 28 22 14 31.2 54.5 42.8 27.3 4 12 20 16 12 23.4 38.9 31.2 23.4 5 5 16 18 12 9.7 31.2 35.1 23.4
C-3 1 10 30 22 12 19.5 58.4 42.8 23.4 2 14 22 20 14 27.3 42.8 38.9 27.3 3 16 24 22 16 31.2 46.7 42.8 31.2 4 10 20 20 10 19.5 38.9 38.9 19.5 5 10 20 18 8 19.5 38.9 35.1 15.6
C-4 1 14 24 18 8 27.3 46.7 35.1 15.6 2 16 26 24 10 31.2 50.6 46.7 19.5 3 20 32 30 20 38.9 62.3 58.4 38.9 4 12 24 24 16 23.4 46.7 46.7 31.2 5 10 22 28 14 19.5 42.8 54.5 27.3
216
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
C-5 1 14 22 18 10 27.3 42.8 35.1 19.5 2 16 26 24 12 31.2 50.6 46.7 23.4 3 16 30 26 22 31.2 58.4 50.6 42.8 4 12 26 24 16 23.4 50.6 46.7 31.2 5 10 20 26 16 19.5 38.9 50.6 31.2
C-6 1 14 32 26 12 27.3 62.3 50.6 23.4 2 18 32 30 14 35.1 62.3 58.4 27.3 3 20 34 34 26 38.9 66.2 66.2 50.6 4 14 30 30 20 27.3 58.4 58.4 38.9 5 12 24 28 20 23.4 46.7 54.5 38.9
D-1 1 10 12 14 6 19.5 23.4 27.3 11.7 2 12 28 16 8 23.4 54.5 31.2 15.6 3 14 24 22 14 27.3 46.7 42.8 27.3 4 10 18 18 12 19.5 35.1 35.1 23.4 5 8 16 22 12 15.6 31.2 42.8 23.4
D-2 1 12 18 14 8 23.4 35.1 27.3 15.6 2 14 20 20 10 27.3 38.9 38.9 19.5 3 16 26 20 16 31.2 50.6 38.9 31.2 4 10 20 20 12 19.5 38.9 38.9 23.4 5 10 18 20 12 19.5 35.1 38.9 23.4
D-3 1 12 26 20 6 23.4 50.6 38.9 11.7 2 14 20 18 8 27.3 38.9 35.1 15.6 3 18 24 24 12 35.1 46.7 46.7 23.4 4 10 20 20 10 19.5 38.9 38.9 19.5 5 8 14 18 12 15.6 27.3 35.1 23.4
D-4 1 14 24 20 8 27.3 46.7 38.9 15.6 2 16 24 20 10 31.2 46.7 38.9 19.5 3 16 28 28 14 31.2 54.5 54.5 27.3 4 12 22 20 12 23.4 42.8 38.9 23.4 5 10 20 26 12 19.5 38.9 50.6 23.4
D-5 1 18 66 26 10 35.1 128.5 50.6 19.5 2 18 26 28 12 35.1 50.6 54.5 23.4 3 22 40 28 18 42.8 77.9 54.5 35.1 4 14 26 24 20 27.3 50.6 46.7 38.9 5 12 22 24 16 23.4 42.8 46.7 31.2
D-6 1 20 60 24 12 38.9 116.8 46.7 23.4 2 22 30 28 14 42.8 58.4 54.5 27.3 3 20 38 32 24 38.9 74.0 62.3 46.7 4 16 30 28 20 31.2 58.4 54.5 38.9 5 10 24 28 18 19.5 46.7 54.5 35.1
217
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
E-1 1 12 56 22 8 23.4 109.0 42.8 15.6 2 12 20 18 8 23.4 38.9 35.1 15.6 3 16 20 22 14 31.2 38.9 42.8 27.3 4 12 20 20 12 23.4 38.9 38.9 23.4 5 10 20 18 12 19.5 38.9 35.1 23.4
E-2 1 14 26 18 8 27.3 50.6 35.1 15.6 2 16 22 22 8 31.2 42.8 42.8 15.6 3 16 28 24 16 31.2 54.5 46.7 31.2 4 12 20 22 12 23.4 38.9 42.8 23.4 5 8 18 22 12 15.6 35.1 42.8 23.4
E-3 1 14 50 16 8 27.3 97.4 31.2 15.6 2 16 22 20 8 31.2 42.8 38.9 15.6 3 18 30 22 18 35.1 58.4 42.8 35.1 4 12 22 22 12 23.4 42.8 42.8 23.4 5 8 24 30 12 15.6 46.7 58.4 23.4
E-4 1 16 34 18 10 31.2 66.2 35.1 19.5 2 20 26 24 10 38.9 50.6 46.7 19.5 3 20 32 28 18 38.9 62.3 54.5 35.1 4 14 24 26 14 27.3 46.7 50.6 27.3 5 10 20 24 14 19.5 38.9 46.7 27.3
E-5 1 18 76 14 10 35.1 148.0 27.3 19.5 2 20 28 26 12 38.9 54.5 50.6 23.4 3 26 38 32 20 50.6 74.0 62.3 38.9 4 16 28 26 18 31.2 54.5 50.6 35.1 5 12 22 26 16 23.4 42.8 50.6 31.2
E-6 1 16 70 28 10 31.2 136.3 54.5 19.5 2 22 32 32 12 42.8 62.3 62.3 23.4 3 22 36 32 28 42.8 70.1 62.3 54.5 4 16 32 32 20 31.2 62.3 62.3 38.9 5 12 28 34 20 23.4 54.5 66.2 38.9
F-1 1 14 40 20 10 27.3 77.9 38.9 19.5 2 14 24 22 12 27.3 46.7 42.8 23.4 3 16 30 24 16 31.2 58.4 46.7 31.2 4 12 22 22 14 23.4 42.8 42.8 27.3 5 8 16 40 12 15.6 31.2 77.9 23.4
F-2 1 12 18 20 5 23.4 35.1 38.9 9.7 2 14 20 20 8 27.3 38.9 38.9 15.6 3 16 22 22 16 31.2 42.8 42.8 31.2 4 10 22 20 12 19.5 42.8 38.9 23.4 5 8 24 24 12 15.6 46.7 46.7 23.4
218
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
F-3 1 18 32 22 10 35.1 62.3 42.8 19.5 2 22 30 26 12 42.8 58.4 50.6 23.4 3 22 36 30 20 42.8 70.1 58.4 38.9 4 14 30 26 16 27.3 58.4 50.6 31.2 5 12 26 32 18 23.4 50.6 62.3 35.1
F-4 1 16 34 22 14 31.2 66.2 42.8 27.3 2 16 26 24 16 31.2 50.6 46.7 31.2 3 18 38 28 18 35.1 74.0 54.5 35.1 4 12 24 24 10 23.4 46.7 46.7 19.5 5 10 22 26 8 19.5 42.8 50.6 15.6
F-5 1 20 60 32 18 38.9 116.8 62.3 35.1 2 20 32 30 20 38.9 62.3 58.4 38.9 3 24 46 32 24 46.7 89.6 62.3 46.7 4 16 30 30 14 31.2 58.4 58.4 27.3 5 12 26 36 10 23.4 50.6 70.1 19.5
F-6 1 20 72 32 10 38.9 140.2 62.3 19.5 2 24 36 34 16 46.7 70.1 66.2 31.2 3 28 40 40 26 54.5 77.9 77.9 50.6 4 16 34 34 22 31.2 66.2 66.2 42.8 5 12 30 64 20 23.4 58.4 124.6 38.9
G-1 1 14 48 24 8 27.3 93.5 46.7 15.6 2 16 24 22 10 31.2 46.7 42.8 19.5 3 20 28 26 16 38.9 54.5 50.6 31.2 4 12 22 22 14 23.4 42.8 42.8 27.3 5 10 20 24 12 19.5 38.9 46.7 23.4
G-2 1 14 60 20 8 27.3 116.8 38.9 15.6 2 16 24 22 10 31.2 46.7 42.8 19.5 3 14 30 20 12 27.3 58.4 38.9 23.4 4 12 24 22 12 23.4 46.7 42.8 23.4 5 10 24 24 12 19.5 46.7 46.7 23.4
G-3 1 16 24 26 8 31.2 46.7 50.6 15.6 2 20 28 26 10 38.9 54.5 50.6 19.5 3 22 34 28 18 42.8 66.2 54.5 35.1 4 14 28 26 16 27.3 54.5 50.6 31.2 5 12 24 22 26 23.4 46.7 42.8 50.6
G-4 1 16 34 26 10 31.2 66.2 50.6 19.5 2 20 30 28 12 38.9 58.4 54.5 23.4 3 22 36 30 22 42.8 70.1 58.4 42.8 4 14 26 26 18 27.3 50.6 50.6 35.1 5 10 22 26 16 19.5 42.8 50.6 31.2
219
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
G-5 1 22 40 28 10 42.8 77.9 54.5 19.5 2 22 32 32 14 42.8 62.3 62.3 27.3 3 26 38 36 24 50.6 74.0 70.1 46.7 4 18 34 32 20 35.1 66.2 62.3 38.9 5 12 22 38 18 23.4 42.8 74.0 35.1
G-6 1 22 98 36 14 42.8 190.8 70.1 27.3 2 30 44 42 18 58.4 85.7 81.8 35.1 3 32 46 44 34 62.3 89.6 85.7 66.2 4 20 42 38 28 38.9 81.8 74.0 54.5 5 16 32 42 28 31.2 62.3 81.8 54.5
H-1 1 14 28 22 10 27.3 54.5 42.8 19.5 2 14 26 24 10 27.3 50.6 46.7 19.5 3 18 28 30 25 35.1 54.5 58.4 48.7 4 12 26 24 24 23.4 50.6 46.7 46.7 5 10 20 30 20 19.5 38.9 58.4 38.9
H-2 1 16 44 24 10 31.2 85.7 46.7 19.5 2 18 30 26 12 35.1 58.4 50.6 23.4 3 22 34 30 22 42.8 66.2 58.4 42.8 4 14 28 28 16 27.3 54.5 54.5 31.2 5 10 24 30 16 19.5 46.7 58.4 31.2
H-3 1 16 58 24 12 31.2 112.9 46.7 23.4 2 18 30 28 12 35.1 58.4 54.5 23.4 3 24 40 32 24 46.7 77.9 62.3 46.7 4 16 28 28 18 31.2 54.5 54.5 35.1 5 10 26 34 16 19.5 50.6 66.2 31.2
H-4 1 18 72 32 10 35.1 140.2 62.3 19.5 2 20 30 30 14 38.9 58.4 58.4 27.3 3 24 40 34 22 46.7 77.9 66.2 42.8 4 10 28 30 18 19.5 54.5 58.4 35.1 5 12 26 28 18 23.4 50.6 54.5 35.1
H-5 1 22 70 36 14 42.8 136.3 70.1 27.3 2 24 40 38 18 46.7 77.9 74.0 35.1 3 30 40 40 30 58.4 77.9 77.9 58.4 4 20 40 38 22 38.9 77.9 74.0 42.8 5 16 32 36 22 31.2 62.3 70.1 42.8
H-6 1 24 52 30 14 46.7 101.3 58.4 27.3 2 30 46 46 20 58.4 89.6 89.6 38.9 3 36 48 48 36 70.1 93.5 93.5 70.1 4 22 42 42 28 42.8 81.8 81.8 54.5 5 18 36 58 26 35.1 70.1 112.9 50.6
220
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
I-1 1 14 36 24 10 27.3 70.1 46.7 19.5 2 16 28 26 12 31.2 54.5 50.6 23.4 3 20 32 30 20 38.9 62.3 58.4 38.9 4 12 26 26 16 23.4 50.6 50.6 31.2 5 10 22 24 16 19.5 42.8 46.7 31.2
I-2 1 20 80 40 10 38.9 155.8 77.9 19.5 2 22 32 32 12 42.8 62.3 62.3 23.4 3 28 42 36 20 54.5 81.8 70.1 38.9 4 18 32 30 18 35.1 62.3 58.4 35.1 5 14 30 32 18 27.3 58.4 62.3 35.1
I-3 1 20 86 36 12 38.9 167.5 70.1 23.4 2 22 36 32 14 42.8 70.1 62.3 27.3 3 28 40 34 26 54.5 77.9 66.2 50.6 4 18 24 30 20 35.1 46.7 58.4 38.9 5 14 30 32 20 27.3 58.4 62.3 38.9
I-4 1 22 76 40 10 42.8 148.0 77.9 19.5 2 26 36 32 16 50.6 70.1 62.3 31.2 3 30 44 40 26 58.4 85.7 77.9 50.6 4 18 34 34 20 35.1 66.2 66.2 38.9 5 12 30 42 20 23.4 58.4 81.8 38.9
I-5 1 30 110 48 16 58.4 214.2 93.5 31.2 2 32 46 42 20 62.3 89.6 81.8 38.9 3 36 50 48 34 70.1 97.4 93.5 66.2 4 24 44 44 26 46.7 85.7 85.7 50.6 5 16 38 42 26 31.2 74.0 81.8 50.6
I-6 1 30 100 58 16 58.4 194.7 112.9 31.2 2 34 50 50 24 66.2 97.4 97.4 46.7 3 42 60 52 40 81.8 116.8 101.3 77.9 4 26 50 48 30 50.6 97.4 93.5 58.4 5 18 40 46 30 35.1 77.9 89.6 58.4
J-1 1 20 70 32 10 38.9 136.3 62.3 19.5 2 20 34 32 14 38.9 66.2 62.3 27.3 3 26 38 34 26 50.6 74.0 66.2 50.6 4 18 34 30 20 35.1 66.2 58.4 38.9 5 12 28 30 16 23.4 54.5 58.4 31.2
J-2 1 24 92 44 12 46.7 179.1 85.7 23.4 2 24 38 36 14 46.7 74.0 70.1 27.3 3 32 46 42 26 62.3 89.6 81.8 50.6 4 22 38 36 20 42.8 74.0 70.1 38.9 5 16 36 36 22 31.2 70.1 70.1 42.8
221
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
J-3 1 24 86 48 24 46.7 167.5 93.5 46.7 2 26 40 40 24 50.6 77.9 77.9 46.7 3 32 46 46 30 62.3 89.6 89.6 58.4 4 22 40 40 18 42.8 77.9 77.9 35.1 5 18 34 40 16 35.1 66.2 77.9 31.2
J-4 1 24 88 44 18 46.7 171.4 85.7 35.1 2 28 42 40 20 54.5 81.8 77.9 38.9 3 34 44 44 32 66.2 85.7 85.7 62.3 4 22 40 40 26 42.8 77.9 77.9 50.6 5 18 32 30 24 35.1 62.3 58.4 46.7
J-5 1 24 112 50 12 46.7 218.1 97.4 23.4 2 24 44 36 14 46.7 85.7 70.1 27.3 3 20 50 44 16 38.9 97.4 85.7 31.2 4 20 42 40 22 38.9 81.8 77.9 42.8 5 14 56 66 20 27.3 109.0 128.5 38.9
J-6 1 38 92 58 22 74.0 179.1 112.9 42.8 2 42 64 60 28 81.8 124.6 116.8 54.5 3 52 70 66 50 101.3 136.3 128.5 97.4 4 32 64 60 40 62.3 124.6 116.8 77.9 5 22 52 60 38 42.8 101.3 116.8 74.0
K-1 1 30 76 56 16 58.4 148.0 109.0 31.2 2 34 54 50 22 66.2 105.2 97.4 42.8 3 42 54 56 38 81.8 105.2 109.0 74.0 4 26 54 50 30 50.6 105.2 97.4 58.4 5 20 46 50 30 38.9 89.6 97.4 58.4
K-2 1 26 96 44 14 50.6 186.9 85.7 27.3 2 30 48 44 20 58.4 93.5 85.7 38.9 3 36 50 46 34 70.1 97.4 89.6 66.2 4 24 44 44 26 46.7 85.7 85.7 50.6 5 18 34 38 26 35.1 66.2 74.0 50.6
K-3 1 30 128 52 16 58.4 249.2 101.3 31.2 2 36 50 46 20 70.1 97.4 89.6 38.9 3 42 56 52 36 81.8 109.0 101.3 70.1 4 28 50 46 28 54.5 97.4 89.6 54.5 5 22 44 50 26 42.8 85.7 97.4 50.6
K-4 1 30 96 48 16 58.4 186.9 93.5 31.2 2 32 50 46 20 62.3 97.4 89.6 38.9 3 40 48 52 36 77.9 93.5 101.3 70.1 4 24 50 46 30 46.7 97.4 89.6 58.4 5 18 40 44 26 35.1 77.9 85.7 50.6
222
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
K-5 1 42 78 80 24 81.8 151.9 155.8 46.7 2 48 78 70 30 93.5 151.9 136.3 58.4 3 60 76 78 50 116.8 148.0 151.9 97.4 4 36 72 70 44 70.1 140.2 136.3 85.7 5 26 60 66 42 50.6 116.8 128.5 81.8
K-6 1 50 118 94 28 97.4 229.8 183.0 54.5 2 56 84 78 34 109.0 163.6 151.9 66.2 3 60 84 90 62 116.8 163.6 175.3 120.7 4 40 80 78 50 77.9 155.8 151.9 97.4 5 30 66 76 48 58.4 128.5 148.0 93.5
L-1 1 33 121 66 17 64.3 235.6 128.5 33.1 2 39 61 50 21 75.9 118.8 97.4 40.9 3 47 68 61 41 91.5 132.4 118.8 79.8 4 31 59 57 33 60.4 114.9 111.0 64.3 5 21 51 52 27 40.9 99.3 101.3 52.6
L-2 1 34 100 70 20 66.2 194.7 136.3 38.9 2 40 60 56 24 77.9 116.8 109.0 46.7 3 46 64 62 42 89.6 124.6 120.7 81.8 4 30 60 56 34 58.4 116.8 109.0 66.2 5 20 48 60 32 38.9 93.5 116.8 62.3
L-3 1 36 112 56 20 70.1 218.1 109.0 38.9 2 40 64 60 24 77.9 124.6 116.8 46.7 3 50 64 64 48 97.4 124.6 124.6 93.5 4 32 60 58 34 62.3 116.8 112.9 66.2 5 22 50 66 36 42.8 97.4 128.5 70.1
L-4 1 44 88 76 22 85.7 171.4 148.0 42.8 2 50 72 66 30 97.4 140.2 128.5 58.4 3 56 78 72 50 109.0 151.9 140.2 97.4 4 36 70 68 40 70.1 136.3 132.4 77.9 5 26 58 60 38 50.6 112.9 116.8 74.0
L-5 1 40 80 90 24 77.9 155.8 175.3 46.7 2 50 72 66 30 97.4 140.2 128.5 58.4 3 58 78 74 54 112.9 151.9 144.1 105.2 4 38 70 68 42 74.0 136.3 132.4 81.8 5 24 58 66 40 46.7 112.9 128.5 77.9
L-6 1 44 120 80 26 85.7 233.7 155.8 50.6 2 50 76 70 32 97.4 148.0 136.3 62.3 3 56 78 80 50 109.0 151.9 155.8 97.4 4 38 72 70 44 74.0 140.2 136.3 85.7 5 26 60 66 40 50.6 116.8 128.5 77.9
223
Volume (mL) Intensity (mm/hr) Control box setting
Position a b c d a b c d
M-1 1 62 112 138 30 120.7 218.1 268.7 58.4 2 72 116 100 42 140.2 225.9 194.7 81.8 3 76 110 110 80 148.0 214.2 214.2 155.8 4 50 100 100 62 97.4 194.7 194.7 120.7 5 36 86 94 60 70.1 167.5 183.0 116.8
M-2 1 58 86 110 30 112.9 167.5 214.2 58.4 2 68 108 96 40 132.4 210.3 186.9 77.9 3 84 108 104 80 163.6 210.3 202.5 155.8 4 52 100 100 62 101.3 194.7 194.7 120.7 5 34 88 98 58 66.2 171.4 190.8 112.9
M-3 1 64 100 102 32 124.6 194.7 198.6 62.3 2 66 108 96 40 128.5 210.3 186.9 77.9 3 78 110 112 70 151.9 214.2 218.1 136.3 4 50 104 100 62 97.4 202.5 194.7 120.7 5 38 84 94 60 74.0 163.6 183.0 116.8
M-4 1 54 90 90 28 105.2 175.3 175.3 54.5 2 64 96 90 40 124.6 186.9 175.3 77.9 3 74 84 100 64 144.1 163.6 194.7 124.6 4 46 90 88 52 89.6 175.3 171.4 101.3 5 30 74 86 52 58.4 144.1 167.5 101.3
M-5 1 54 150 90 30 105.2 292.1 175.3 58.4 2 60 98 86 40 116.8 190.8 167.5 77.9 3 76 100 100 68 148.0 194.7 194.7 132.4 4 50 92 90 60 97.4 179.1 175.3 116.8 5 34 84 84 52 66.2 163.6 163.6 101.3
M-6 1 60 86 100 32 116.8 167.5 194.7 62.3 2 64 98 90 44 124.6 190.8 175.3 85.7 3 78 100 116 66 151.9 194.7 225.9 128.5 4 48 96 94 58 93.5 186.9 183.0 112.9 5 36 76 88 56 70.1 148.0 171.4 109.0
224
Table B - Calculation of median drop size using the flour pellet method
Sieve size
(mm) Weight of
pellets (g)
Weight of a single
pellet (mg)
Number of pellets
Calibration ratio
Mass of a water drop
(mg)
>4.75 1.338 57.50 23 1.25 71.95
4.75-3.35 3.183 21.85 146 1.28 27.90
3.35-2.36 4.361 13.57 321 1.27 17.29
2.36-1.68 3.799 4.60 826 1.21 5.58
1.68-1.18 2.007 1.87 1070 1.12 2.09
1.18-0.85 1.030 0.49 2098 0.79 0.39 <0.85 0.730 0.22 3266 0.66 0.15
Sieve size
(mm) Volume (cm3)
Average drop diameter(mm)
Total volume (mm3)
% of total volume
>4.75 0.072 5.16 1.67 8.61 4.75-3.35 0.028 3.76 4.07 20.91 3.35-2.36 0.017 3.21 5.56 28.59 2.36-1.68 0.006 2.20 4.61 23.71 1.68-1.18 0.002 1.59 2.24 11.52 1.18-0.85 0.000 0.90 0.81 4.17
<0.85 0.000 0.66 0.48 2.49
Average drop
diameter (mm)
Percentage of total volume
(%)
Cumulative volume (%)
0.66 2.49 2.49 0.90 4.17 6.66 1.59 11.52 18.18 2.20 23.71 41.89 3.21 28.59 70.48 3.76 20.91 91.39
5.16 8.61 100.00
225
0
10
20
30
40
50
60
70
80
90
100
110
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Average drop diameter(mm)
Per
cen
tage
of c
umul
ativ
e vo
lum
e
Figure B - Variation of drop diameter vs %of cumulative volume According to Figure B, Median drop size diameter = 2.45mm Table C - Calculation of kinetic energy
Average drop
diameter (mm)
Mass of a drop (mg)
Terminal velocity (m/s)
Kinetic energy per
drop (J)
Number of drops in
each class
Kinetic energy of all drops
(J)
5.16 71.95 8.71 0.00273 23 0.064 3.76 27.90 8.70 0.00106 146 0.154 3.21 17.29 8.26 0.00059 321 0.189 2.20 5.58 6.98 0.00014 826 0.112 1.59 2.09 5.97 0.00004 1070 0.040 0.90 0.39 4.23 0.00001 2098 0.007 0.66 0.15 3.71 0.00001 3266 0.003
Total kinetic energy of the rain drops = 0.57 J
Sample area = 34.5*45.5 cm2
Duration the sample was taken =3.2s
Intensity = 159.3mm/hr
According to Kinnell (1987)
Kinetic energy per unit depth of rain = (0.57x3600)/ [159.3x3.2x 0.345x0.455]
= 25.63 J/m2/mm
226
227
APPENDIX 2
TEST RESULTS
228
229
Table A1 - Build-up data for all the road surfaces (Total build-up samples)
Sample name Sample volume (L)
Antecedent dry days
TS load (g)
R-BU 8.42 8 6.76
I-BU 7.04 9 10.31
C-BU 9.76 11 12.18
Sample name Particle size distribution- volume %
<1 µm
1-75 µm
75-150 µm
150-300 µm
>300 µm
R-BU 1.22 50.30 29.81 12.41 6.26
I-BU 7.96 77.04 5.30 0.61 9.10
C-BU 2.64 59.02 19.79 11.07 7.48
Sample name TOC
(mg/L) NO2
-
(mg/L) NO3
-
(mg/L) TKN (mg/L
TN (mg/L)
R-BU 28.19 0.101 1.841 5.268 7.210
I-BU 19.22 0.018 0.639 3.576 4.233
C-BU 23.30 0.063 0.757 12.43 13.245
Sample name PO4
3-
(mg/L) TP
(mg/L)
R-BU 0.795 1.670
I-BU 5.756 7.005
C-BU 3.779 6.855 Note:
R- Residential
I- Industrial
C- Commercial
BU- Build-up
230
Table A2 - Build-up data for all the road surfaces (Wet sieved build-up samples)
Sample name TS (mg/L)
TOC (mg/L)
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
R-BU<1 225.0 7.29 0.009 1.269 0.194 1.472
R-BU-1-75 108.8 5.10 <0.001 0.206 1.576 1.782
R-BU-75-150 326.0 10.49 <0.001 0.176 3.118 3.294
R-BU-150-300 90.7 2.52 <0.001 0.168 0.497 0.665
R-BU>300 16.7 1.90 0.001 0.165 <0.035 0.166
I-BU<1 400.0 5.39 0.012 0.416 0.291 0.719
I-BU-1-75 100.2 4.46 <0.001 0.090 1.075 1.164
I-BU-75-150 589.2 3.56 <0.001 0.010 1.357 1.366
I-BU-150-300 80.0 1.37 <0.001 <0.003 0.149 0.149
I-BU>300 42.4 1.99 <0.001 0.009 0.257 0.266
C-BU<1 295.0 6.16 0.054 0.535 0.408 0.998
C-BU-1-75 82.2 3.36 <0.001 0.000 1.604 1.604
C-BU-75-150 425.6 9.74 0.021 0.144 8.335 8.500
C-BU-150-300 216.0 3.58 <0.001 0.155 0.647 0.802
C-BU>300 98.4 4.04 0.001 0.169 0.707 0.877
Sample name PO43-
(mg/L) TP
(mg/L) R-BU<1 0.093 0.310 R-BU-1-75 0.219 0.389 R-BU-75-150 0.125 0.439 R-BU-150-300 0.146 0.197 R-BU>300 0.168 0.528 I-BU<1 0.826 0.851 I-BU-1-75 2.052 2.172 I-BU-75-150 1.911 2.509 I-BU-150-300 0.552 0.865 I-BU>300 0.585 0.894 C-BU<1 0.256 0.284 C-BU-1-75 1.036 1.445 C-BU-75-150 0.679 1.955 C-BU-150-300 0.980 1.618 C-BU>300 0.978 1.086
Note:
R- Residential
I- Industrial
C- Commercial
BU- Build-up
231
Table B1 – Residential site wash-off data –total wash-off samples
Rainfall intensity mm/hr
Duration min Particle size distribution- volume in %
Identification
<1 µm
0.75-75 µm
75-150 µm
150-300 µm
>300 µm
R21 20 0-5 2.65 54.87 12.40 9.50 20.49
R22 20 0-10 1.91 37.06 9.21 5.69 46.09
R23 20 0-15 1.27 28.05 7.04 3.88 59.73
R24 20 0-20 1.41 25.28 6.88 3.49 62.92
R25 20 0-25 1.13 20.22 5.50 2.79 70.34
R26 20 0-30 0.94 28.44 9.59 2.32 58.62
R27 20 0-35 1.31 20.50 18.16 2.92 57.17
R28 20 0-40 1.14 28.28 8.34 4.36 58.59
R41 40 0-5 0.70 17.39 3.47 1.44 77.01
R42 40 0-10 0.71 12.17 2.25 0.80 84.04
R43 40 0-15 0.84 13.62 2.32 0.98 82.21
R44 40 0-20 1.30 19.08 4.23 2.56 72.89
R45 40 0-25 1.88 22.63 4.98 4.75 65.81
R46 40 0-30 1.57 20.14 4.22 3.96 70.16
R47 40 0-35 1.77 21.81 4.43 4.07 67.97
R61 65 0-5 1.44 40.62 16.56 10.05 31.33
R62 65 0-10 1.46 35.19 15.76 9.37 38.24
R63 65 0-15 0.97 27.96 12.04 8.22 50.82
R64 65 0-20 0.73 24.54 10.21 6.99 57.54
R65 65 0-25 0.58 23.38 9.44 7.66 58.93
R66 65 0-30 0.49 19.83 8.07 7.49 64.14
R81 86 0-5 1.85 36.30 13.25 12.65 35.95
R82 86 0-10 1.55 23.49 8.02 7.03 59.92
R83 86 0-15 2.29 26.33 7.11 4.75 59.52
R84 86 0-20 1.72 22.09 5.86 3.61 66.73
R11 115 0-5 3.43 36.12 17.78 16.94 25.74
R12 115 0-10 10.11 57.20 11.36 8.47 12.87
R13 115 0-15 6.74 39.17 9.08 10.03 35.00
R31 115 0-20 0.64 29.72 7.83 2.99 58.81
R32 135 0-5 2.50 35.17 11.55 5.18 45.58
R33 135 0-10 3.38 29.82 9.32 4.11 53.36
R34 135 0-15 4.00 30.10 7.08 3.08 55.74 Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-35 min
232
Table B1 – Residential site wash-off data –total wash-off samples (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min pH EC
(µS/cm) TSS
(mg/L) TOC
(mg/L)
R21 20 0-5 7.30 44.00 28.42 17.33
R22 20 0-10 7.09 42.10 27.54 13.81
R23 20 0-15 7.01 41.10 27.03 12.39
R24 20 0-20 7.00 39.85 23.77 11.38
R25 20 0-25 7.00 38.60 23.22 10.60
R26 20 0-30 7.01 37.65 22.85 9.98
R27 20 0-35 7.02 36.77 22.15 9.47
R28 20 0-40 6.94 35.70 21.01 9.03
R41 40 0-5 6.47 42.80 30.00 18.74
R42 40 0-10 6.49 39.75 29.50 16.71
R43 40 0-15 6.52 38.83 28.67 15.70
R44 40 0-20 6.53 38.18 27.25 15.00
R45 40 0-25 6.56 38.00 26.00 14.03
R46 40 0-30 6.58 37.62 23.33 13.48
R47 40 0-35 6.64 37.17 21.14 13.01
R61 65 0-5 6.66 36.80 51.80 14.65
R62 65 0-10 6.65 27.69 47.36 12.80
R63 65 0-15 6.64 25.12 44.89 11.82
R64 65 0-20 6.64 24.56 42.18 11.30
R65 65 0-25 6.63 24.22 39.96 10.86
R66 65 0-30 6.63 23.73 35.77 10.51
R81 86 0-5 6.21 45.90 58.90 23.84
R82 86 0-10 6.19 37.95 44.30 21.51
R83 86 0-15 6.17 35.60 36.73 19.90
R84 86 0-20 6.22 34.33 32.82 18.70
R11 115 0-5 6.58 34.40 69.44 24.49
R12 115 0-10 6.62 32.90 53.94 17.88
R13 115 0-15 6.73 32.40 40.51 14.61
R31 115 0-20 6.97 35.60 73.44 12.55
R32 135 0-5 6.71 35.55 43.47 9.95
R33 135 0-10 6.56 36.10 34.74 8.80
R34 135 0-15 6.61 36.05 30.11 8.12 Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-35 min
233
Table B1 – Residential site wash-off data –total wash-off samples (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
R21 20 0-5 0.005 0.150 3.581 3.736
R22 20 0-10 0.004 0.121 3.183 3.308
R23 20 0-15 0.004 0.111 2.577 2.692
R24 20 0-20 0.003 0.094 2.187 2.284
R25 20 0-25 0.003 0.086 1.901 1.990
R26 20 0-30 0.002 0.080 1.692 1.774
R27 20 0-35 0.002 0.073 1.533 1.608
R28 20 0-40 0.002 0.066 1.392 1.459
R41 40 0-5 0.007 0.188 3.486 3.681
R42 40 0-10 0.006 0.177 3.040 3.222
R43 40 0-15 0.005 0.168 2.572 2.744
R44 40 0-20 0.004 0.149 2.318 2.471
R45 40 0-25 0.003 0.132 2.146 2.281
R46 40 0-30 0.003 0.115 2.019 2.137
R47 40 0-35 0.002 0.102 1.914 2.018
R61 65 0-5 0.006 0.289 2.042 2.337
R62 65 0-10 0.004 0.256 1.568 1.828
R63 65 0-15 0.003 0.244 1.312 1.559
R64 65 0-20 0.002 0.236 1.164 1.401
R65 65 0-25 0.002 0.228 1.058 1.288
R66 65 0-30 0.001 0.223 0.981 1.205
R81 86 0-5 0.008 0.096 2.880 2.984
R82 86 0-10 0.008 0.090 2.836 2.934
R83 86 0-15 0.007 0.095 2.571 2.673
R84 86 0-20 0.005 0.073 2.308 2.385
R11 115 0-5 0.006 0.051 2.847 2.904
R12 115 0-10 0.006 0.042 2.172 2.219
R13 115 0-15 0.004 0.037 1.679 1.721
R31 115 0-20 0.007 0.073 2.212 2.292
R32 135 0-5 0.006 0.054 1.581 1.641
R33 135 0-10 0.004 0.043 1.262 1.310
R34 135 0-15 0.003 0.036 1.098 1.137 Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-35 min
234
Table B1 – Residential site wash-off data –total wash-off samples (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
R21 20 0-5 0.135 0.157
R22 20 0-10 0.078 0.146
R23 20 0-15 0.058 0.142
R24 20 0-20 0.047 0.126
R25 20 0-25 0.040 0.115
R26 20 0-30 0.035 0.107
R27 20 0-35 <0.035 0.100
R28 20 0-40 <0.035 0.093
R41 40 0-5 0.151 0.567
R42 40 0-10 0.104 0.408
R43 40 0-15 0.084 0.353
R44 40 0-20 0.064 0.291
R45 40 0-25 0.052 0.239
R46 40 0-30 0.044 0.223
R47 40 0-35 0.038 0.208
R61 65 0-5 0.126 0.679
R62 65 0-10 0.080 0.670
R63 65 0-15 0.062 0.657
R64 65 0-20 0.049 0.646
R65 65 0-25 0.042 0.619
R66 65 0-30 0.038 0.602
R81 86 0-5 0.129 0.862
R82 86 0-10 0.102 0.757
R83 86 0-15 0.090 0.719
R84 86 0-20 0.075 0.693
R11 115 0-5 0.096 0.911
R12 115 0-10 0.090 0.899
R13 115 0-15 0.063 0.623
R31 115 0-20 0.042 0.928
R32 135 0-5 0.042 0.925
R33 135 0-10 0.037 0.788
R34 135 0-15 0.043 0.664 Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-35 min
235
Table B2 – Industrial site wash-off data –total wash-off samples
Identification Rainfall intensity mm/hr
Duration min Particle size distribution- volume in %
<0.75 µm
0.75-75 µm
75-150 µm
150-300 µm
>300 µm
I21 20 0-5 5.06 62.59 2.73 0.18 29.42
I22 20 0-10 4.80 60.54 3.46 1.12 30.08
I23 20 0-15 3.54 47.36 2.93 0.88 45.29
I24 20 0-20 2.65 39.00 2.33 0.66 55.34
I25 20 0-25 2.42 33.73 1.95 0.53 61.36
I26 20 0-30 2.02 29.61 1.73 0.45 66.17
I27 20 0-35 1.73 25.38 1.49 0.38 71.00
I41 40 0-5 4.77 49.45 1.91 0.50 43.37
I42 40 0-10 2.85 35.82 3.85 1.00 56.50
I44 40 0-15 2.94 48.02 8.28 2.67 38.05
I45 40 0-20 2.39 42.62 10.34 3.51 41.11
I46 40 0-25 2.17 25.48 7.14 4.80 60.42
I61 65 0-5 19.53 75.31 2.55 1.30 1.32
I62 65 0-10 11.38 66.79 7.79 2.08 11.98
I63 65 0-15 7.93 51.99 6.98 1.70 31.42
I64 65 0-20 5.95 42.57 6.41 2.10 42.99
I65 65 0-25 5.00 39.22 6.44 1.99 47.35
I66 65 0-30 4.19 35.39 6.00 1.71 52.70
I81 86 0-5 9.21 74.76 3.57 0.77 11.33
I82 86 0-10 5.60 58.23 5.54 1.02 29.45
I83 86 0-15 4.15 48.98 6.32 1.01 39.44
I84 86 0-20 3.48 45.36 7.72 1.50 41.87
I85 86 0-25 2.78 36.82 6.22 1.20 52.92
I11 115 0-5 8.11 79.70 7.20 3.30 1.68
I12 115 0-10 4.84 57.40 10.80 3.40 23.56
I13 115 0-15 3.23 43.20 8.80 2.39 42.37
I14 115 0-20 2.47 36.87 7.89 1.93 50.84
I31 135 0-5 9.21 84.29 4.84 1.52 0.10
I32 135 0-10 6.35 73.39 14.75 5.27 0.23
I33 135 0-15 4.23 49.66 10.02 3.52 32.56
I34 135 0-20 3.18 37.25 7.52 2.64 49.42 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min
236
Table B2 – Industrial site wash-off data –total wash-off samples (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min pH EC
(µS/cm) TSS
(mg/L) TOC
(mg/L)
I21 20 0-5 7.28 37.50 498.80 11.75
I22 20 0-10 7.19 33.95 481.10 10.83
I23 20 0-15 7.13 30.60 400.91 10.18
I24 20 0-20 7.16 28.68 339.93 9.15
I25 20 0-25 7.15 27.68 297.02 8.38
I26 20 0-30 7.16 26.68 267.58 7.81
I27 20 0-35 7.15 25.76 243.81 7.39
I41 40 0-5 7.45 161.80 621.60 25.87
I42 40 0-10 7.44 99.70 618.60 19.11
I44 40 0-15 5.57 57.50 373.40 11.77
I45 40 0-20 5.92 51.48 349.07 10.90
I46 40 0-25 6.15 47.62 331.24 10.27
I61 65 0-5 7.92 132.60 692.00 28.80
I62 65 0-10 7.72 107.35 525.90 18.07
I63 65 0-15 7.76 78.77 469.00 14.00
I64 65 0-20 7.76 63.24 429.80 11.70
I65 65 0-25 7.75 54.75 379.74 10.18
I66 65 0-30 7.66 48.16 334.06 9.11
I81 86 0-5 7.53 65.80 778.80 20.32
I82 86 0-10 7.32 43.45 558.50 14.55
I83 86 0-15 7.34 33.25 460.91 11.91
I84 86 0-20 7.32 29.89 402.66 11.49
I85 86 0-25 7.31 25.87 347.69 10.89
I11 115 0-5 7.37 35.80 797.60 19.62
I12 115 0-10 7.24 22.30 582.80 12.82
I13 115 0-15 7.20 17.49 474.49 10.08
I14 115 0-20 7.18 14.47 388.19 8.23
I31 135 0-5 7.16 39.40 816.40 15.62
I32 135 0-10 7.14 30.26 610.60 10.78
I33 135 0-15 7.10 26.25 502.62 9.39
I34 135 0-20 6.89 23.84 416.22 8.10 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min
237
Table B2 – Industrial site wash-off data –total wash-off samples (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2 -(mg/L)
NO3-
(mg/L) TKN
(mg/L) TN
(mg/L)
I21 20 0-5 0.004 0.284 2.612 2.900
I22 20 0-10 0.004 0.221 2.004 2.228
I23 20 0-15 0.003 0.180 1.709 1.892
I24 20 0-20 0.002 0.154 1.481 1.637
I25 20 0-25 0.002 0.135 1.325 1.461
I26 20 0-30 0.001 0.120 1.206 1.328
I27 20 0-35 0.001 0.108 1.118 1.227
I41 40 0-5 0.011 0.274 2.587 2.872
I42 40 0-10 0.006 0.214 2.098 2.317
I44 40 0-15 0.003 0.137 1.440 1.580
I45 40 0-20 0.003 0.130 1.349 1.482
I46 40 0-25 0.002 0.125 1.266 1.393
I61 65 0-5 0.003 0.272 2.000 2.275
I62 65 0-10 0.003 0.265 1.395 1.662
I63 65 0-15 0.002 0.250 1.182 1.433
I64 65 0-20 0.002 0.237 1.068 1.307
I65 65 0-25 0.001 0.226 1.050 1.277
I66 65 0-30 0.001 0.215 0.977 1.194
I81 86 0-5 0.006 0.383 2.365 2.754
I82 86 0-10 0.004 0.330 1.903 2.236
I83 86 0-15 0.003 0.300 1.715 2.017
I84 86 0-20 0.002 0.279 1.879 2.160
I85 86 0-25 0.002 0.263 1.790 2.055
I11 115 0-5 0.010 0.392 1.576 1.978
I12 115 0-10 0.006 0.278 1.278 1.562
I13 115 0-15 0.004 0.221 1.100 1.325
I14 115 0-20 0.003 0.184 0.968 1.154
I31 135 0-5 0.013 0.268 1.427 1.708
I32 135 0-10 0.007 0.186 1.042 1.235
I33 135 0-15 0.005 0.143 0.871 1.019
I34 135 0-20 0.004 0.118 0.737 0.858 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min
238
Table B2 – Industrial site wash-off data –total wash-off samples (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
I21 20 0-5 1.385 4.857
I22 20 0-10 1.164 4.624
I23 20 0-15 1.020 4.498
I24 20 0-20 0.961 3.723
I25 20 0-25 0.906 3.256
I26 20 0-30 0.860 2.928
I27 20 0-35 0.826 2.692
I41 40 0-5 1.537 5.993
I42 40 0-10 1.629 5.187
I44 40 0-15 1.166 3.651
I45 40 0-20 1.181 3.770
I46 40 0-25 1.185 3.509
I61 65 0-5 1.616 7.711
I62 65 0-10 1.425 7.075
I63 65 0-15 1.243 6.146
I64 65 0-20 1.116 5.413
I65 65 0-25 0.923 4.494
I66 65 0-30 0.770 3.871
I81 86 0-5 1.648 8.904
I82 86 0-10 1.496 6.933
I83 86 0-15 1.397 6.158
I84 86 0-20 1.258 5.685
I85 86 0-25 1.177 4.786
I11 115 0-5 2.065 9.465
I12 115 0-10 1.671 7.296
I13 115 0-15 1.478 6.250
I14 115 0-20 1.272 5.148
I31 135 0-5 2.706 9.870
I32 135 0-10 1.872 7.219
I33 135 0-15 1.536 6.202
I34 135 0-20 1.293 5.686
Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min
239
Table B3 – Commercial site wash-off data –total wash-off samples
Identification Rainfall intensity mm/hr
Duration min Particle size distribution- volume in %
<1 µm
0.75-75 µm
75-150 µm
150-300 µm
>300 µm
C21 20 0-5 2.18 62.07 17.73 12.86 5.14 C22 20 0-10 2.23 69.54 15.66 8.50 4.06 C23 20 0-15 2.38 72.22 14.49 7.20 3.70 C24 20 0-20 2.32 71.66 14.74 7.10 4.18 C25 20 0-25 2.23 69.58 15.34 7.49 5.35 C26 20 0-30 2.15 68.33 15.84 7.41 6.28 C27 20 0-35 2.18 67.78 15.84 7.41 6.79 C28 20 0-40 2.18 67.63 16.07 7.33 6.80 C41 40 0-5 3.24 78.22 11.12 3.43 3.99 C42 40 0-10 2.86 74.36 13.16 4.32 5.31 C43 40 0-15 2.51 69.96 14.69 5.69 7.14 C44 40 0-20 2.61 67.46 14.69 6.91 8.35 C45 40 0-25 2.52 66.45 15.26 6.96 8.82 C46 40 0-30 2.41 64.25 15.27 7.17 10.91 C47 40 0-35 2.35 62.85 15.54 7.10 12.17 C61 65 0-5 3.01 70.54 11.54 7.71 7.20 C62 65 0-10 2.54 61.31 13.25 10.27 12.65 C63 65 0-15 2.22 57.06 14.76 11.54 14.43 C64 65 0-20 2.10 55.19 15.72 11.99 14.95 C65 65 0-25 1.98 53.13 15.91 12.37 16.54 C66 65 0-30 1.90 51.65 15.68 12.20 18.53 C81 86 0-5 2.40 65.59 11.78 7.68 12.58 C82 86 0-10 2.11 60.91 13.65 7.22 16.12 C83 86 0-15 2.12 56.61 14.90 9.34 17.04 C84 86 0-20 1.92 53.22 15.32 8.97 20.56 C85 86 0-25 1.76 48.94 15.38 8.85 25.06 C11 115 0-5 2.54 63.89 13.82 7.79 11.95 C12 115 0-10 2.05 59.23 17.18 8.27 13.27 C13 115 0-15 1.82 55.39 19.03 8.86 14.89 C14 115 0-20 1.75 51.45 20.27 10.89 15.65 C31 135 0-5 2.63 70.48 11.21 6.35 8.99 C32 135 0-10 2.09 72.04 8.88 6.83 11.06 C33 135 0-15 1.39 64.34 16.10 7.78 10.65 C34 135 0-20 1.43 61.91 18.82 10.87 12.43
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min
240
Table B3 – Commercial site wash-off data –total wash-off samples (parameters continued from previous page)
Identification
Rainfall intensit mm/hr y
Duration min pH EC
(µS/cm) TSS
(mg/L) TOC
(mg/L)
C21 20 0-5 7.97 502.00 295.60 67.84 C22 20 0-10 7.55 466.50 251.30 48.76 C23 20 0-15 7.39 457.00 231.53 40.46 C24 20 0-20 7.32 449.75 207.45 35.43 C25 20 0-25 7.28 442.40 192.84 31.97 C26 20 0-30 7.26 438.50 177.23 29.65 C27 20 0-35 7.25 434.14 164.94 27.89 C28 20 0-40 7.24 430.00 155.68 26.05 C41 40 0-5 7.56 482.00 307.60 57.02 C42 40 0-10 7.33 486.00 286.60 45.44 C43 40 0-15 7.52 474.33 250.67 38.91 C44 40 0-20 7.39 477.50 228.20 34.07 C45 40 0-25 7.49 470.60 199.12 30.64 C46 40 0-30 7.56 466.83 176.07 28.01 C47 40 0-35 7.61 456.43 159.77 25.99 C61 65 0-5 7.55 612.00 310.60 27.10 C62 65 0-10 7.45 541.50 280.50 21.74 C63 65 0-15 7.55 518.33 230.07 18.88 C64 65 0-20 7.57 503.75 191.85 16.72 C65 65 0-25 7.63 500.20 167.72 15.27 C66 65 0-30 7.69 493.67 148.00 14.14 C81 86 0-5 7.73 498.00 335.60 31.28 C82 86 0-10 7.51 486.00 232.20 23.23 C83 86 0-15 7.44 485.33 177.60 20.52 C84 86 0-20 7.47 482.50 147.50 18.10 C85 86 0-25 7.46 494.80 129.28 16.63 C11 115 0-5 7.12 448.00 374.00 40.20 C12 115 0-10 7.22 470.50 297.20 29.31 C13 115 0-15 7.19 464.33 278.80 24.25 C14 115 0-20 7.19 468.75 245.60 20.69 C31 135 0-5 7.94 387.00 428.00 14.66 C32 135 0-10 7.51 370.50 328.80 11.62 C33 135 0-15 7.37 357.67 239.73 10.24 C34 135 0-20 7.30 348.00 189.00 9.23
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min
241
Table B3 – Commercial site wash-off data –total wash-off samples (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
C21 20 0-5 0.138 0.905 15.532 16.575 C22 20 0-10 0.088 0.669 11.134 11.891 C23 20 0-15 0.070 0.578 9.121 9.770 C24 20 0-20 0.059 0.523 8.075 8.657 C25 20 0-25 0.052 0.490 7.141 7.683 C26 20 0-30 0.048 0.449 6.513 7.010 C27 20 0-35 0.045 0.420 6.054 6.519 C28 20 0-40 0.042 0.393 5.600 6.035 C41 40 0-5 0.075 0.903 12.499 13.477 C42 40 0-10 0.051 0.642 9.495 10.188 C43 40 0-15 0.042 0.553 7.996 8.592 C44 40 0-20 0.038 0.481 6.947 7.466 C45 40 0-25 0.034 0.431 6.133 6.598 C46 40 0-30 0.031 0.398 5.524 5.953 C47 40 0-35 0.028 0.371 5.011 5.410 C61 65 0-5 0.046 0.653 5.885 6.584 C62 65 0-10 0.036 0.535 4.380 4.950 C63 65 0-15 0.031 0.492 3.530 4.053 C64 65 0-20 0.028 0.469 3.084 3.580 C65 65 0-25 0.025 0.451 2.780 3.257 C66 65 0-30 0.023 0.410 2.465 2.897 C81 86 0-5 0.038 0.738 5.920 6.696 C82 86 0-10 0.028 0.697 4.097 4.822 C83 86 0-15 0.025 0.683 3.156 3.863 C84 86 0-20 0.023 0.581 2.655 3.259 C85 86 0-25 0.021 0.516 2.316 2.853 C11 115 0-5 0.039 0.280 3.180 3.499 C12 115 0-10 0.032 0.258 3.141 3.432 C13 115 0-15 0.029 0.249 2.790 3.067 C14 115 0-20 0.026 0.240 2.417 2.683 C31 135 0-5 0.021 0.285 2.066 2.372 C32 135 0-10 0.018 0.262 1.584 1.864 C33 135 0-15 0.016 0.252 1.229 1.497 C34 135 0-20 0.014 0.243 1.042 1.300
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min
242
Table B3 – Commercial site wash-off data –total wash-off samples (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
C21 20 0-5 1.289 4.453
C22 20 0-10 1.224 4.454
C23 20 0-15 1.194 4.453
C24 20 0-20 1.160 4.452
C25 20 0-25 1.082 4.256
C26 20 0-30 1.017 4.068
C27 20 0-35 0.968 3.860
C28 20 0-40 0.847 3.468
C41 40 0-5 0.901 4.462
C42 40 0-10 0.887 3.861
C43 40 0-15 0.803 3.651
C44 40 0-20 0.757 3.387
C45 40 0-25 0.715 3.061
C46 40 0-30 0.683 2.841
C47 40 0-35 0.658 2.678
C61 65 0-5 0.942 4.546
C62 65 0-10 0.897 4.353
C63 65 0-15 0.860 4.502
C64 65 0-20 0.825 4.497
C65 65 0-25 0.773 4.210
C66 65 0-30 0.644 3.570
C81 86 0-5 0.979 5.314
C82 86 0-10 0.919 4.820
C83 86 0-15 0.827 4.639
C84 86 0-20 0.755 4.519
C85 86 0-25 0.604 3.927
C11 115 0-5 1.408 5.808
C12 115 0-10 1.378 5.330
C13 115 0-15 1.336 4.737
C14 115 0-20 1.002 4.213
C31 135 0-5 2.621 10.393
C32 135 0-10 1.938 9.761
C33 135 0-15 1.621 8.214
C34 135 0-20 1.430 7.193 Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min
243
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range <1 µm
Identification Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TDS (mg/L)
DOC (mg/L)
R21-v 20 0-5 44.00 135.00 13.76 R22-v 20 0-10 42.10 133.75 11.02 R23-v 20 0-15 41.10 130.00 9.46 R24-v 20 0-20 39.85 120.00 8.33 R25-v 20 0-25 38.60 119.50 7.45 R26-v 20 0-30 37.65 118.75 6.87 R27-v 20 0-35 36.77 116.43 6.40 R28-v 20 0-40 35.70 114.37 6.03 R41-v 40 0-5 42.80 130.00 11.65 R42-v 40 0-10 39.75 121.25 11.63 R43-v 40 0-15 38.83 117.50 11.15 R44-v 40 0-20 38.18 115.62 10.78 R45-v 40 0-25 38.00 114.00 10.25 R46-v 40 0-30 37.62 104.58 9.97 R47-v 40 0-35 37.17 96.79 9.53 R61-v 65 0-5 36.80 75.00 10.13 R62-v 65 0-10 27.69 75.00 8.06 R63-v 65 0-15 25.12 68.33 7.30 R64-v 65 0-20 24.56 63.75 7.03 R65-v 65 0-25 24.22 60.00 6.83 R66-v 65 0-30 23.73 52.50 6.56 R81-v 86 0-5 45.90 187.50 15.52 R82-v 86 0-10 37.95 148.75 14.21 R83-v 86 0-15 35.60 133.33 13.01 R84-v 86 0-20 34.33 123.13 11.95 R11-v 115 0-5 34.40 175.25 12.87 R12-v 115 0-10 32.90 153.87 10.20 R13-v 115 0-15 32.40 140.08 8.51 R31-v 115 0-20 35.60 222.50 8.15 R32-v 135 0-5 35.55 160.00 6.13 R33-v 135 0-10 36.10 137.50 5.32 R34-v 135 0-15 36.05 126.25 4.82
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min v- Particle size range <1 µm
244
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range <1 µm (parameters continued from previous page)
Identification
Rainfall intensit mm/hr y
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
R21-v 20 0-5 0.004 0.137 3.067 3.208 R22-v 20 0-10 0.004 0.113 2.382 2.498 R23-v 20 0-15 0.003 0.103 2.030 2.136 R24-v 20 0-20 0.003 0.091 1.760 1.853 R25-v 20 0-25 0.002 0.082 1.512 1.596 R26-v 20 0-30 0.002 0.072 1.349 1.423 R27-v 20 0-35 0.001 0.065 1.206 1.273 R28-v 20 0-40 0.001 0.058 1.097 1.157 R41-v 40 0-5 0.005 0.187 2.933 3.125 R42-v 40 0-10 0.004 0.153 2.592 2.749 R43-v 40 0-15 0.003 0.136 2.291 2.430 R44-v 40 0-20 0.003 0.123 2.106 2.232 R45-v 40 0-25 0.002 0.108 1.960 2.070 R46-v 40 0-30 0.002 0.095 1.834 1.931 R47-v 40 0-35 0.002 0.084 1.708 1.794 R61-v 65 0-5 0.004 0.207 1.559 1.770 R62-v 65 0-10 0.003 0.215 1.325 1.542 R63-v 65 0-15 0.002 0.205 1.069 1.276 R64-v 65 0-20 0.001 0.204 0.960 1.165 R65-v 65 0-25 0.001 0.201 0.893 1.095 R66-v 65 0-30 0.001 0.201 0.833 1.034 R81-v 86 0-5 0.007 0.106 3.724 3.837 R82-v 86 0-10 0.007 0.091 3.238 3.336 R83-v 86 0-15 0.005 0.096 2.806 2.906 R84-v 86 0-20 0.004 0.072 2.442 2.518 R11-v 115 0-5 0.004 0.049 2.539 2.592 R12-v 115 0-10 0.004 0.037 1.894 1.934 R13-v 115 0-15 0.003 0.033 1.481 1.517 R31-v 115 0-20 0.006 0.063 1.677 1.746 R32-v 135 0-5 0.005 0.049 1.305 1.358 R33-v 135 0-10 0.003 0.042 1.075 1.120 R34-v 135 0-15 0.003 0.035 0.946 0.983
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min v- Particle size range <1 µm
245
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range <1 µm (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
R21-v 20 0-5 0.014 0.014 R22-v 20 0-10 0.012 0.012 R23-v 20 0-15 0.010 0.031 R24-v 20 0-20 0.009 0.025 R25-v 20 0-25 0.008 0.021 R26-v 20 0-30 0.007 0.018 R27-v 20 0-35 0.006 0.016 R28-v 20 0-40 0.006 0.015 R41-v 40 0-5 0.009 0.010 R42-v 40 0-10 0.008 0.009 R43-v 40 0-15 0.008 0.009 R44-v 40 0-20 0.007 <0.009 R45-v 40 0-25 0.006 <0.009 R46-v 40 0-30 0.006 <0.009 R47-v 40 0-35 0.006 <0.009 R61-v 65 0-5 0.006 0.020 R62-v 65 0-10 0.005 0.016 R63-v 65 0-15 <0.005 0.012 R64-v 65 0-20 <0.005 0.010 R65-v 65 0-25 <0.005 0.009 R66-v 65 0-30 <0.005 <0.009 R81-v 86 0-5 0.013 0.014 R82-v 86 0-10 0.012 0.014 R83-v 86 0-15 0.010 0.011 R84-v 86 0-20 0.009 0.009 R11-v 115 0-5 0.026 0.027 R12-v 115 0-10 0.020 0.021 R13-v 115 0-15 0.015 0.016 R31-v 115 0-20 0.039 0.049 R32-v 135 0-5 0.034 0.044 R33-v 135 0-10 0.031 0.038 R34-v 135 0-15 0.026 0.032
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min v- Particle size range <1 µm
246
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm
Identification Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
R21-iv 20 0-5 44.00 9.20 1.29 R22-iv 20 0-10 42.10 6.40 0.77 R23-iv 20 0-15 41.10 5.87 0.57 R24-iv 20 0-20 39.85 4.80 0.67 R25-iv 20 0-25 38.60 5.12 0.76 R26-iv 20 0-30 37.65 5.47 0.75 R27-iv 20 0-35 36.77 6.29 0.69 R28-iv 20 0-40 35.70 9.05 0.93 R41-iv 40 0-5 42.80 15.60 3.80 R42-iv 40 0-10 39.75 13.60 2.57 R43-iv 40 0-15 38.83 13.60 1.94 R44-iv 40 0-20 38.18 12.90 1.58 R45-iv 40 0-25 38.00 12.64 1.44 R46-iv 40 0-30 37.62 11.67 1.24 R47-iv 40 0-35 37.17 10.91 1.28 R61-iv 65 0-5 36.80 9.20 0.00 R62-iv 65 0-10 27.69 11.20 0.33 R63-iv 65 0-15 25.12 10.67 0.61 R64-iv 65 0-20 24.56 9.90 0.65 R65-iv 65 0-25 24.22 9.44 0.69 R66-iv 65 0-30 23.73 8.93 0.76 R81-iv 86 0-5 45.90 24.40 8.60 R82-iv 86 0-10 37.95 14.40 4.79 R83-iv 86 0-15 35.60 11.87 3.59 R84-iv 86 0-20 34.33 10.20 3.06 R11-iv 115 0-5 34.40 9.60 1.16 R12-iv 115 0-10 32.90 7.80 1.03 R13-iv 115 0-15 32.40 6.80 1.02 R31-iv 115 0-20 35.60 14.00 0.87 R32-iv 135 0-5 35.55 7.80 1.01 R33-iv 135 0-10 36.10 5.60 0.98 R34-iv 135 0-15 36.05 4.40 0.93
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min iv- Particle size range 1-75 µm
247
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
R21-iv 20 0-5 0.001 0.003 0.087 0.091 R22-iv 20 0-10 0.001 0.018 0.142 0.161 R23-iv 20 0-15 0.001 0.036 0.135 0.171 R24-iv 20 0-20 0.001 0.037 0.131 0.168 R25-iv 20 0-25 <0.001 0.040 0.107 0.148 R26-iv 20 0-30 <0.001 0.037 0.096 0.133 R27-iv 20 0-35 <0.001 0.034 0.083 0.117 R28-iv 20 0-40 <0.001 0.031 0.099 0.130 R41-iv 40 0-5 0.001 0.071 0.110 0.182 R42-iv 40 0-10 0.001 0.062 0.058 0.121 R43-iv 40 0-15 0.001 0.070 0.039 0.110 R44-iv 40 0-20 0.001 0.072 <0.035 0.072 R45-iv 40 0-25 <0.001 0.066 0.041 0.107 R46-iv 40 0-30 <0.001 0.059 0.036 0.095 R47-iv 40 0-35 <0.001 0.053 <0.035 0.053 R61-iv 65 0-5 <0.001 0.092 0.113 0.205 R62-iv 65 0-10 <0.001 0.077 0.069 0.146 R63-iv 65 0-15 <0.001 0.072 0.145 0.217 R64-iv 65 0-20 <0.001 0.064 0.116 0.179 R65-iv 65 0-25 <0.001 0.057 0.106 0.163 R66-iv 65 0-30 <0.001 0.051 0.103 0.154 R81-iv 86 0-5 0.001 0.046 <0.035 0.047 R82-iv 86 0-10 0.001 0.047 <0.035 0.047 R83-iv 86 0-15 <0.001 0.062 <0.035 0.062 R84-iv 86 0-20 <0.001 0.046 0.043 0.089 R11-iv 115 0-5 0.001 0.042 0.611 0.654 R12-iv 115 0-10 0.001 0.062 0.347 0.409 R13-iv 115 0-15 <0.001 0.047 0.336 0.383 R31-iv 115 0-20 0.001 0.057 0.702 0.760 R32-iv 135 0-5 0.001 0.057 0.361 0.418 R33-iv 135 0-10 <0.001 0.043 0.394 0.437 R34-iv 135 0-15 <0.001 0.034 0.315 0.348
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min iv- Particle size range 1-75 µm
248
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
R21-iv 20 0-5 0.119 0.128 R22-iv 20 0-10 0.059 0.071 R23-iv 20 0-15 0.040 0.048 R24-iv 20 0-20 0.030 0.037 R25-iv 20 0-25 <0.03 0.035 R26-iv 20 0-30 <0.03 0.034 R27-iv 20 0-35 <0.03 0.038 R28-iv 20 0-40 0.052 0.080 R41-iv 40 0-5 0.157 0.421 R42-iv 40 0-10 0.106 0.234 R43-iv 40 0-15 0.071 0.176 R44-iv 40 0-20 0.053 0.164 R45-iv 40 0-25 0.042 0.154 R46-iv 40 0-30 0.035 0.135 R47-iv 40 0-35 0.030 0.133 R61-iv 65 0-5 0.050 0.137 R62-iv 65 0-10 0.025 0.159 R63-iv 65 0-15 <0.03 0.171 R64-iv 65 0-20 <0.03 0.173 R65-iv 65 0-25 <0.03 0.165 R66-iv 65 0-30 <0.03 0.184 R81-iv 86 0-5 0.069 0.173 R82-iv 86 0-10 0.042 0.142 R83-iv 86 0-15 <0.03 0.171 R84-iv 86 0-20 0.024 0.183 R11-iv 115 0-5 <0.03 0.148 R12-iv 115 0-10 <0.03 0.103 R13-iv 115 0-15 <0.03 0.083 R31-iv 115 0-20 <0.03 0.036 R32-iv 135 0-5 <0.03 0.033 R33-iv 135 0-10 <0.03 0.032 R34-iv 135 0-15 <0.03 0.043
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min iv- Particle size range 1-75 µm
249
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
R21-iii 20 0-5 44.00 15.33 0.92 R22-iii 20 0-10 42.10 10.00 0.92 R23-iii 20 0-15 41.10 9.78 0.87 R24-iii 20 0-20 39.85 8.50 0.79 R25-iii 20 0-25 38.60 9.47 0.89 R26-iii 20 0-30 37.65 8.00 0.89 R27-iii 20 0-35 36.77 7.71 0.85 R28-iii 20 0-40 35.70 7.83 0.88 R41-iii 40 0-5 42.80 0.67 0.88 R42-iii 40 0-10 39.75 2.00 0.76 R43-iii 40 0-15 38.83 1.78 0.70 R44-iii 40 0-20 38.18 2.83 0.68 R45-iii 40 0-25 38.00 3.33 0.66 R46-iii 40 0-30 37.62 3.56 0.67 R47-iii 40 0-35 37.17 4.00 0.65 R61-iii 65 0-5 36.80 34.00 1.46 R62-iii 65 0-10 27.69 23.00 1.39 R63-iii 65 0-15 25.12 17.33 1.24 R64-iii 65 0-20 24.56 16.00 1.19 R65-iii 65 0-25 24.22 13.60 1.14 R66-iii 65 0-30 23.73 12.44 1.09 R81-iii 86 0-5 45.90 19.33 1.03 R82-iii 86 0-10 37.95 13.33 0.94 R83-iii 86 0-15 35.60 14.00 0.95 R84-iii 86 0-20 34.33 12.33 0.99 R11-iii 115 0-5 34.40 12.67 0.99 R12-iii 115 0-10 32.90 11.00 1.00 R13-iii 115 0-15 32.40 8.22 1.02 R31-iii 115 0-20 35.60 28.00 1.07 R32-iii 135 0-5 35.55 18.67 1.01 R33-iii 135 0-10 36.10 14.44 0.93 R34-iii 135 0-15 36.05 12.00 0.85
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min iii- Particle size range 75-150 µm
250
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
min NO3
-
(mg/L TKN (mg/L
TN (mg/L
R21-iii 20 0-5 0.003 <0.003 <0.035 0.003 R22-iii 20 0-10 0.003 0.023 0.860 0.885 R23-iii 20 0-15 0.002 0.037 0.573 0.612 R24-iii 20 0-20 0.002 0.030 0.430 0.461 R25-iii 20 0-25 0.001 0.033 0.344 0.378 R26-iii 20 0-30 0.001 0.030 0.287 0.318 R27-iii 20 0-35 0.001 0.029 0.246 0.275 R28-iii 20 0-40 0.001 0.026 0.215 0.242 R41-iii 40 0-5 <0.001 0.051 <0.035 0.051 R42-iii 40 0-10 <0.001 0.044 <0.035 0.044 R43-iii 40 0-15 <0.001 0.040 <0.035 0.040 R44-iii 40 0-20 <0.001 0.045 <0.035 0.045 R45-iii 40 0-25 <0.001 0.044 <0.035 0.044 R46-iii 40 0-30 <0.001 0.041 <0.035 0.041 R47-iii 40 0-35 <0.001 0.037 <0.035 0.037 R61-iii 65 0-5 <0.001 0.069 0.473 0.542 R62-iii 65 0-10 <0.001 0.064 0.236 0.300 R63-iii 65 0-15 <0.001 0.060 0.158 0.217 R64-iii 65 0-20 <0.001 0.054 0.118 0.172 R65-iii 65 0-25 <0.001 0.055 0.095 0.150 R66-iii 65 0-30 <0.001 0.053 0.079 0.132 R81-iii 86 0-5 <0.001 0.036 <0.035 0.036 R82-iii 86 0-10 0.004 0.030 <0.035 0.033 R83-iii 86 0-15 0.002 0.043 0.077 0.122 R84-iii 86 0-20 0.002 0.033 0.057 0.092 R11-iii 115 0-5 0.000 <0.003 <0.035 0.000 R12-iii 115 0-10 0.000 0.034 <0.035 0.034 R13-iii 115 0-15 0.000 0.025 <0.035 0.025 R31-iii 115 0-20 0.001 0.046 <0.035 0.047 R32-iii 135 0-5 0.002 0.047 <0.035 0.048 R33-iii 135 0-10 0.001 0.040 <0.035 0.041 R34-iii 135 0-15 0.002 0.031 <0.035 0.032
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min iii- Particle size range 75-150 µm
251
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
R21-iii 20 0-5 0.047 0.112 R22-iii 20 0-10 <0.03 0.079 R23-iii 20 0-15 <0.03 0.073 R24-iii 20 0-20 <0.03 0.055 R25-iii 20 0-25 <0.03 0.047 R26-iii 20 0-30 <0.03 0.041 R27-iii 20 0-35 <0.03 0.036 R28-iii 20 0-40 <0.03 0.065 R41-iii 40 0-5 <0.03 0.062 R42-iii 40 0-10 <0.03 0.066 R43-iii 40 0-15 <0.03 0.085 R44-iii 40 0-20 <0.03 0.085 R45-iii 40 0-25 <0.03 0.092 R46-iii 40 0-30 <0.03 0.078 R47-iii 40 0-35 <0.03 0.085 R61-iii 65 0-5 <0.03 0.032 R62-iii 65 0-10 <0.03 0.099 R63-iii 65 0-15 <0.03 0.127 R64-iii 65 0-20 <0.03 0.146 R65-iii 65 0-25 <0.03 0.151 R66-iii 65 0-30 <0.03 0.157 R81-iii 86 0-5 <0.03 0.179 R82-iii 86 0-10 <0.03 0.183 R83-iii 86 0-15 <0.03 0.148 R84-iii 86 0-20 <0.03 0.184 R11-iii 115 0-5 0.038 0.192 R12-iii 115 0-10 0.048 0.144 R13-iii 115 0-15 0.032 0.100 R31-iii 115 0-20 <0.03 0.002 R32-iii 135 0-5 <0.03 0.021 R33-iii 135 0-10 <0.03 0.016 R34-iii 135 0-15 <0.03 0.012
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min iii- Particle size range 75-150 µm
252
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
R21-ii 20 0-5 44.00 11.33 0.81 R22-ii 20 0-10 42.10 9.00 0.79 R23-ii 20 0-15 41.10 7.11 0.74 R24-ii 20 0-20 39.85 9.67 0.80 R25-ii 20 0-25 38.60 9.47 0.75 R26-ii 20 0-30 37.65 8.00 0.76 R27-ii 20 0-35 36.77 7.24 0.73 R28-ii 20 0-40 35.70 6.75 0.73 R41-ii 40 0-5 42.80 4.00 0.81 R42-ii 40 0-10 39.75 2.33 0.71 R43-ii 40 0-15 38.83 1.78 0.67 R44-ii 40 0-20 38.18 2.50 0.68 R45-ii 40 0-25 38.00 3.33 0.64 R46-ii 40 0-30 37.62 3.78 0.64 R47-ii 40 0-35 37.17 3.90 0.64 R61-ii 65 0-5 36.80 10.67 1.10 R62-ii 65 0-10 27.69 8.67 1.11 R63-ii 65 0-15 25.12 8.22 1.06 R64-ii 65 0-20 24.56 7.67 1.06 R65-ii 65 0-25 24.22 6.27 1.02 R66-ii 65 0-30 23.73 6.11 0.99 R81-ii 86 0-5 45.90 16.67 1.37 R82-ii 86 0-10 37.95 12.33 1.10 R83-ii 86 0-15 35.60 10.44 0.96 R84-ii 86 0-20 34.33 9.00 0.90 R11-ii 115 0-5 34.40 9.33 0.96 R12-ii 115 0-10 32.90 7.00 0.84 R13-ii 115 0-15 32.40 6.67 0.79 R31-ii 115 0-20 35.60 17.33 0.81 R32-ii 135 0-5 35.55 8.67 0.72 R33-ii 135 0-10 36.10 6.44 0.66 R34-ii 135 0-15 36.05 6.67 0.67
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min ii- Particle size range 150-300 µm
253
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
R21-ii 20 0-5 0.002 <0.003 0.100 0.003 R22-ii 20 0-10 0.002 0.009 0.050 0.011 R23-ii 20 0-15 <0.001 0.022 0.033 0.023 R24-ii 20 0-20 <0.001 0.018 0.025 0.020 R25-ii 20 0-25 <0.001 0.024 0.020 0.025 R26-ii 20 0-30 <0.001 0.022 0.017 0.023 R27-ii 20 0-35 <0.001 0.021 0.014 0.022 R28-ii 20 0-40 0.001 0.019 0.013 0.020 R41-ii 40 0-5 <0.001 0.033 <0.035 0.033 R42-ii 40 0-10 <0.001 0.039 <0.035 0.039 R43-ii 40 0-15 <0.001 0.035 <0.035 0.035 R44-ii 40 0-20 <0.001 0.040 <0.035 0.040 R45-ii 40 0-25 <0.001 0.040 <0.035 0.040 R46-ii 40 0-30 <0.001 0.037 <0.035 0.037 R47-ii 40 0-35 <0.001 0.034 <0.035 0.034 R61-ii 65 0-5 <0.001 0.046 <0.035 0.046 R62-ii 65 0-10 <0.001 0.049 <0.035 0.049 R63-ii 65 0-15 <0.001 0.049 <0.035 0.049 R64-ii 65 0-20 <0.001 0.053 <0.035 0.053 R65-ii 65 0-25 <0.001 0.055 <0.035 0.055 R66-ii 65 0-30 <0.001 0.054 <0.035 0.054 R81-ii 86 0-5 <0.001 0.024 <0.035 0.024 R82-ii 86 0-10 <0.001 0.025 <0.035 0.025 R83-ii 86 0-15 <0.001 0.016 <0.035 0.016 R84-ii 86 0-20 <0.001 0.012 <0.035 0.012 R11-ii 115 0-5 <0.001 <0.003 <0.035 0.000 R12-ii 115 0-10 <0.001 0.018 <0.035 0.018 R13-ii 115 0-15 <0.001 0.013 <0.035 0.013 R31-ii 115 0-20 <0.001 0.052 <0.035 0.053 R32-ii 135 0-5 0.002 0.038 <0.035 0.039 R33-ii 135 0-10 0.002 0.031 <0.035 0.033 R34-ii 135 0-15 0.002 0.024 <0.035 0.025
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min ii- Particle size range 150-300 µm
254
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
R21-ii 20 0-5 <0.03 0.098 R22-ii 20 0-10 <0.03 0.065 R23-ii 20 0-15 <0.03 0.073 R24-ii 20 0-20 <0.03 0.065 R25-ii 20 0-25 <0.03 0.060 R26-ii 20 0-30 <0.03 0.055 R27-ii 20 0-35 <0.03 0.050 R28-ii 20 0-40 <0.03 0.071 R41-ii 40 0-5 <0.03 0.049 R42-ii 40 0-10 <0.03 0.058 R43-ii 40 0-15 <0.03 0.052 R44-ii 40 0-20 <0.03 0.106 R45-ii 40 0-25 <0.03 0.107 R46-ii 40 0-30 <0.03 0.093 R47-ii 40 0-35 <0.03 0.087 R61-ii 65 0-5 0.055 0.111 R62-ii 65 0-10 <0.03 0.154 R63-ii 65 0-15 0.036 0.173 R64-ii 65 0-20 <0.03 0.182 R65-ii 65 0-25 <0.03 0.182 R66-ii 65 0-30 <0.03 0.189 R81-ii 86 0-5 <0.03 0.244 R82-ii 86 0-10 <0.03 0.218 R83-ii 86 0-15 <0.03 0.207 R84-ii 86 0-20 <0.03 0.198 R11-ii 115 0-5 <0.03 0.218 R12-ii 115 0-10 <0.03 0.147 R13-ii 115 0-15 <0.03 0.109 R31-ii 115 0-20 <0.03 0.010 R32-ii 135 0-5 <0.03 0.019 R33-ii 135 0-10 <0.03 0.014 R34-ii 135 0-15 <0.03 0.011
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min ii- Particle size range 150-300 µm
255
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range >300 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
R21-i 20 0-5 44.00 14.67 0.86 R22-i 20 0-10 42.10 9.33 0.76 R23-i 20 0-15 41.10 9.11 0.72 R24-i 20 0-20 39.85 8.67 0.70 R25-i 20 0-25 38.60 8.80 0.69 R26-i 20 0-30 37.65 7.44 0.66 R27-i 20 0-35 36.77 6.48 0.67 R28-i 20 0-40 35.70 7.58 0.77 R41-i 40 0-5 42.80 10.00 1.04 R42-i 40 0-10 39.75 11.00 0.89 R43-i 40 0-15 38.83 8.67 0.87 R44-i 40 0-20 38.18 6.83 0.81 R45-i 40 0-25 38.00 7.33 0.77 R46-i 40 0-30 37.62 7.22 0.75 R47-i 40 0-35 37.17 7.43 0.75 R61-i 65 0-5 36.80 8.67 1.66 R62-i 65 0-10 27.69 8.33 1.44 R63-i 65 0-15 25.12 7.11 1.28 R64-i 65 0-20 24.56 6.33 1.31 R65-i 65 0-25 24.22 5.33 1.30 R66-i 65 0-30 23.73 4.67 1.26 R81-i 86 0-5 45.90 0.67 1.34 R82-i 86 0-10 37.95 1.33 1.18 R83-i 86 0-15 35.60 3.33 1.02 R84-i 86 0-20 34.33 4.00 0.91 R11-i 115 0-5 34.40 8.67 0.93 R12-i 115 0-10 32.90 6.33 0.77 R13-i 115 0-15 32.40 5.56 0.71 R31-i 115 0-20 35.60 5.33 1.16 R32-i 135 0-5 35.55 3.00 1.00 R33-i 135 0-10 36.10 2.44 0.91 R34-i 135 0-15 36.05 3.67 0.83
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min i- Particle size range >300 µm
256
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range >300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
R21-i 20 0-5 <0.001 <0.003 <0.035 nd R22-i 20 0-10 <0.001 0.004 <0.035 0.004 R23-i 20 0-15 <0.001 0.021 <0.035 0.021 R24-i 20 0-20 <0.001 0.017 <0.035 0.017 R25-i 20 0-25 <0.001 0.022 <0.035 0.022 R26-i 20 0-30 <0.001 0.020 <0.035 0.020 R27-i 20 0-35 <0.001 0.018 <0.035 0.018 R28-i 20 0-40 <0.001 0.016 <0.035 0.016 R41-i 40 0-5 <0.001 0.031 <0.035 0.031 R42-i 40 0-10 <0.001 0.032 <0.035 0.032 R43-i 40 0-15 <0.001 0.031 <0.035 0.031 R44-i 40 0-20 <0.001 0.034 <0.035 0.034 R45-i 40 0-25 <0.001 0.036 <0.035 0.036 R46-i 40 0-30 <0.001 0.033 <0.035 0.033 R47-i 40 0-35 <0.001 0.030 <0.035 0.030 R61-i 65 0-5 <0.001 0.050 <0.035 0.050 R62-i 65 0-10 <0.001 0.052 <0.035 0.052 R63-i 65 0-15 <0.001 0.047 <0.035 0.047 R64-i 65 0-20 <0.001 0.050 <0.035 0.050 R65-i 65 0-25 <0.001 0.048 <0.035 0.048 R66-i 65 0-30 <0.001 0.049 <0.035 0.049 R81-i 86 0-5 <0.001 0.019 <0.035 0.019 R82-i 86 0-10 <0.001 0.017 <0.035 0.017 R83-i 86 0-15 <0.001 0.065 <0.035 0.065 R84-i 86 0-20 <0.001 0.049 <0.035 0.049 R11-i 115 0-5 <0.001 <0.003 <0.035 0.000 R12-i 115 0-10 <0.001 <0.003 <0.035 0.000 R13-i 115 0-15 <0.001 0.004 <0.035 0.004 R31-i 115 0-20 <0.001 0.054 <0.035 0.054 R32-i 135 0-5 <0.001 0.038 <0.035 0.038 R33-i 135 0-10 <0.001 0.031 <0.035 0.031 R34-i 135 0-15 <0.001 0.024 <0.035 0.024
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min i- Particle size range >300 µm
257
Table C1 – Residential site wash-off data –wet sieved wash-off samples Particle size range >300 µm (parameters continued from previous page)
Identification
Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
R21-i 20 0-5 <0.03 0.110 R22-i 20 0-10 <0.03 0.075 R23-i 20 0-15 <0.03 0.086 R24-i 20 0-20 <0.03 0.070 R25-i 20 0-25 <0.03 0.065 R26-i 20 0-30 <0.03 0.056 R27-i 20 0-35 <0.03 0.051 R28-i 20 0-40 <0.03 0.083 R41-i 40 0-5 <0.03 0.127 R42-i 40 0-10 <0.03 0.092 R43-i 40 0-15 <0.03 0.073 R44-i 40 0-20 <0.03 0.077 R45-i 40 0-25 <0.03 0.079 R46-i 40 0-30 <0.03 0.077 R47-i 40 0-35 <0.03 0.079 R61-i 65 0-5 0.112 0.155 R62-i 65 0-10 0.065 0.179 R63-i 65 0-15 0.046 0.184 R64-i 65 0-20 0.037 0.193 R65-i 65 0-25 0.033 0.191 R66-i 65 0-30 0.030 0.194 R81-i 86 0-5 0.036 0.203 R82-i 86 0-10 <0.03 0.207 R83-i 86 0-15 <0.03 0.233 R84-i 86 0-20 <0.03 0.228 R11-i 115 0-5 <0.03 0.196 R12-i 115 0-10 <0.03 0.198 R13-i 115 0-15 <0.03 0.134 R31-i 115 0-20 <0.03 <0.009 R32-i 135 0-5 <0.03 <0.009 R33-i 135 0-10 <0.03 <0.009 R34-i 135 0-15 <0.03 <0.009
Note: R- Residential 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8-0-40 min i- Particle size range >300 µm
258
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range <1 µm
Identification Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TDS (mg/L)
DOC (mg/L)
I21-v 20 0-5 37.50 340.00 10.38
I22-v 20 0-10 33.95 316.00 8.39
I23-v 20 0-15 30.60 301.33 7.20
I24-v 20 0-20 28.68 293.00 6.46
I25-v 20 0-25 27.68 276.40 5.88
I26-v 20 0-30 26.68 263.33 5.52
I27-v 20 0-35 25.76 251.71 5.29
I41-v 40 0-5 161.80 376.00 18.67
I42-v 40 0-10 99.70 357.00 13.59
I44-v 40 0-15 57.50 257.50 8.14
I45-v 40 0-20 51.48 260.40 7.48
I46-v 40 0-25 47.62 261.33 7.05
I61-v 65 0-5 132.60 378.00 11.33
I62-v 65 0-10 107.35 318.00 7.62
I63-v 65 0-15 78.77 262.00 5.95
I64-v 65 0-20 63.24 231.50 5.05
I65-v 65 0-25 54.75 209.60 4.47
I66-v 65 0-30 48.16 183.00 4.07
I81-v 86 0-5 65.80 388.00 15.55
I82-v 86 0-10 43.45 282.00 10.77
I83-v 86 0-15 33.25 230.67 8.64
I84-v 86 0-20 29.89 202.00 8.67
I85-v 86 0-25 25.87 182.00 8.12
I11-v 115 0-5 35.80 438.00 12.64
I12-v 115 0-10 22.30 400.00 8.43
I13-v 115 0-15 17.49 368.67 6.67
I14-v 115 0-20 14.47 326.00 5.67
I31-v 135 0-5 39.40 584.00 7.50
I32-v 135 0-10 30.26 427.00 5.13
I33-v 135 0-15 26.25 358.00 4.56
I34-v 135 0-20 23.84 321.00 4.17 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min v- Particle size range <1 µm
259
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range <1 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
I21-v 20 0-5 0.002 0.161 1.507 1.670
I22-v 20 0-10 0.002 0.158 1.322 1.481
I23-v 20 0-15 0.001 0.126 1.249 1.376
I24-v 20 0-20 0.001 0.109 1.118 1.228
I25-v 20 0-25 0.001 0.097 1.013 1.111
I26-v 20 0-30 0.001 0.089 0.935 1.025
I27-v 20 0-35 0.000 0.082 0.861 0.943
I41-v 40 0-5 <0.001 0.283 2.255 2.538
I42-v 40 0-10 0.001 0.237 1.940 2.177
I44-v 40 0-15 <0.001 0.140 1.234 1.375
I45-v 40 0-20 <0.001 0.135 1.184 1.319
I46-v 40 0-25 0.001 0.127 1.127 1.254
I61-v 65 0-5 0.002 0.137 1.084 1.223
I62-v 65 0-10 0.002 0.133 0.928 1.062
I63-v 65 0-15 0.001 0.119 0.821 0.941
I64-v 65 0-20 0.001 0.109 0.773 0.883
I65-v 65 0-25 0.001 0.102 0.732 0.835
I66-v 65 0-30 0.001 0.094 0.711 0.806
I81-v 86 0-5 0.001 0.222 1.579 1.802
I82-v 86 0-10 0.002 0.169 1.422 1.593
I83-v 86 0-15 0.002 0.143 1.267 1.412
I84-v 86 0-20 0.002 0.130 1.415 1.546
I85-v 86 0-25 0.001 0.117 1.417 1.535
I11-v 115 0-5 <0.001 0.272 1.707 1.979
I12-v 115 0-10 0.003 0.184 1.319 1.506
I13-v 115 0-15 0.002 0.142 1.120 1.264
I14-v 115 0-20 0.002 0.118 0.968 1.088
I31-v 135 0-5 0.008 0.261 0.965 1.234
I32-v 135 0-10 0.005 0.178 0.810 0.992
I33-v 135 0-15 0.003 0.136 0.705 0.845
I34-v 135 0-20 0.004 0.114 0.607 0.724 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min v- Particle size range <1 µm
260
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range <1 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
I21-v 20 0-5 0.499 0.520
I22-v 20 0-10 0.464 0.477
I23-v 20 0-15 0.425 0.459
I24-v 20 0-20 0.401 0.447
I25-v 20 0-25 0.382 0.435
I26-v 20 0-30 0.370 0.411
I27-v 20 0-35 0.360 0.391
I41-v 40 0-5 0.531 0.823
I42-v 40 0-10 0.482 0.663
I44-v 40 0-15 0.336 0.444
I45-v 40 0-20 0.341 0.443
I46-v 40 0-25 0.342 0.424
I61-v 65 0-5 0.895 0.930
I62-v 65 0-10 0.688 0.709
I63-v 65 0-15 0.587 0.635
I64-v 65 0-20 0.524 0.573
I65-v 65 0-25 0.425 0.492
I66-v 65 0-30 0.357 0.436
I81-v 86 0-5 0.688 0.843
I82-v 86 0-10 0.530 0.685
I83-v 86 0-15 0.460 0.581
I84-v 86 0-20 0.424 0.518
I85-v 86 0-25 0.401 0.478
I11-v 115 0-5 0.601 0.825
I12-v 115 0-10 0.465 0.699
I13-v 115 0-15 0.409 0.581
I14-v 115 0-20 0.377 0.525
I31-v 135 0-5 0.736 0.982
I32-v 135 0-10 0.511 0.714
I33-v 135 0-15 0.437 0.609
I34-v 135 0-20 0.394 0.553 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min v- Particle size range <1 µm
261
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
I21-iv 20 0-5 37.50 463.20 1.31
I22-iv 20 0-10 33.95 368.20 1.12
I23-iv 20 0-15 30.60 284.00 2.49
I24-iv 20 0-20 28.68 234.00 1.94
I25-iv 20 0-25 27.68 199.28 1.65
I26-iv 20 0-30 26.68 174.47 1.44
I27-iv 20 0-35 25.76 155.66 1.26
I41-iv 40 0-5 161.80 392.00 2.29
I42-iv 40 0-10 99.70 371.20 1.48
I44-iv 40 0-15 57.50 214.20 0.93
I45-iv 40 0-20 51.48 192.24 0.87
I46-iv 40 0-25 47.62 184.87 0.78
I61-iv 65 0-5 132.60 396.00 10.96
I62-iv 65 0-10 107.35 287.40 5.99
I63-iv 65 0-15 78.77 239.20 4.28
I64-iv 65 0-20 63.24 208.20 3.34
I65-iv 65 0-25 54.75 175.44 2.76
I66-iv 65 0-30 48.16 152.47 2.36
I81-iv 86 0-5 65.80 365.20 1.98
I82-iv 86 0-10 43.45 256.40 1.37
I83-iv 86 0-15 33.25 200.00 1.11
I84-iv 86 0-20 29.89 181.40 1.11
I85-iv 86 0-25 25.87 153.84 1.00
I11-iv 115 0-5 35.80 487.20 3.64
I12-iv 115 0-10 22.30 282.20 1.95
I13-iv 115 0-15 17.49 203.07 1.36
I14-iv 115 0-20 14.47 173.30 1.04
I31-iv 135 0-5 39.40 388.00 2.81
I32-iv 135 0-10 30.26 214.00 1.65
I33-iv 135 0-15 26.25 148.00 1.14
I34-iv 135 0-20 23.84 122.00 0.89 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min iv- Particle size range 1-75 µm
262
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
I21-iv 20 0-5 <0.001 0.067 0.287 0.355
I22-iv 20 0-10 <0.001 0.037 0.222 0.259
I23-iv 20 0-15 <0.001 0.034 0.160 0.195
I24-iv 20 0-20 <0.001 0.031 0.123 0.154
I25-iv 20 0-25 <0.001 0.027 0.108 0.135
I26-iv 20 0-30 <0.001 0.023 0.107 0.130
I27-iv 20 0-35 <0.001 0.020 0.114 0.133
I41-iv 40 0-5 <0.001 0.036 0.431 0.467
I42-iv 40 0-10 <0.001 0.021 0.238 0.259
I44-iv 40 0-15 <0.001 0.012 0.319 0.330
I45-iv 40 0-20 <0.001 0.010 0.257 0.266
I46-iv 40 0-25 <0.001 0.010 0.229 0.239
I61-iv 65 0-5 <0.001 0.030 0.585 0.667
I62-iv 65 0-10 <0.001 0.028 0.298 0.352
I63-iv 65 0-15 <0.001 0.036 0.210 0.264
I64-iv 65 0-20 <0.001 0.030 0.166 0.208
I65-iv 65 0-25 <0.001 0.031 0.151 0.192
I66-iv 65 0-30 <0.001 0.027 0.153 0.189
I81-iv 86 0-5 <0.001 0.034 1.172 1.217
I82-iv 86 0-10 <0.001 0.027 0.623 0.656
I83-iv 86 0-15 <0.001 0.037 0.493 0.533
I84-iv 86 0-20 <0.001 0.042 0.488 0.532
I85-iv 86 0-25 <0.001 0.036 0.422 0.461
I11-iv 115 0-5 <0.001 0.073 0.186 0.259
I12-iv 115 0-10 <0.001 0.049 0.117 0.166
I13-iv 115 0-15 <0.001 0.034 0.085 0.119
I14-iv 115 0-20 <0.001 0.026 0.086 0.113
I31-iv 135 0-5 <0.001 0.020 0.022 0.042
I32-iv 135 0-10 <0.001 0.014 0.061 0.076
I33-iv 135 0-15 <0.001 0.010 0.046 0.056
I34-iv 135 0-20 <0.001 0.008 0.067 0.075 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min iv- Particle size range 1-75 µm
263
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
I21-iv 20 0-5 0.984 0.974
I22-iv 20 0-10 0.674 0.943
I23-iv 20 0-15 0.707 0.898
I24-iv 20 0-20 0.546 0.675
I25-iv 20 0-25 0.458 0.570
I26-iv 20 0-30 0.383 0.489
I27-iv 20 0-35 0.336 0.424
I41-iv 40 0-5 2.041 2.060
I42-iv 40 0-10 1.266 1.435
I44-iv 40 0-15 0.644 0.903
I45-iv 40 0-20 0.667 0.912
I46-iv 40 0-25 0.571 0.786
I61-iv 65 0-5 2.219 2.197
I62-iv 65 0-10 1.756 2.032
I63-iv 65 0-15 1.173 1.607
I64-iv 65 0-20 0.988 1.334
I65-iv 65 0-25 0.801 1.110
I66-iv 65 0-30 0.667 0.977
I81-iv 86 0-5 1.663 2.016
I82-iv 86 0-10 0.908 1.372
I83-iv 86 0-15 0.904 1.198
I84-iv 86 0-20 0.687 1.122
I85-iv 86 0-25 0.714 1.070
I11-iv 115 0-5 1.367 1.778
I12-iv 115 0-10 1.056 1.185
I13-iv 115 0-15 0.832 0.923
I14-iv 115 0-20 0.653 0.707
I31-iv 135 0-5 1.951 1.763
I32-iv 135 0-10 1.172 1.026
I33-iv 135 0-15 0.908 0.799
I34-iv 135 0-20 0.726 0.778 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min iv- Particle size range 1-75 µm
264
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
I21-iii 20 0-5 37.50 64.80 0.57
I22-iii 20 0-10 33.95 64.20 0.70
I23-iii 20 0-15 30.60 66.00 0.59
I24-iii 20 0-20 28.68 54.40 0.53
I25-iii 20 0-25 27.68 50.56 0.49
I26-iii 20 0-30 26.68 53.47 0.48
I27-iii 20 0-35 25.76 49.71 0.48
I41-iii 40 0-5 161.80 202.80 1.39
I42-iii 40 0-10 99.70 219.60 1.46
I44-iii 40 0-15 57.50 138.00 1.08
I45-iii 40 0-20 51.48 131.04 1.04
I46-iii 40 0-25 47.62 145.33 1.10
I61-iii 65 0-5 132.60 194.00 3.87
I62-iii 65 0-10 107.35 173.00 2.35
I63-iii 65 0-15 78.77 172.00 1.83
I64-iii 65 0-20 63.24 148.90 1.51
I65-iii 65 0-25 54.75 121.92 1.32
I66-iii 65 0-30 48.16 105.53 1.17
I81-iii 86 0-5 65.80 274.00 1.09
I82-iii 86 0-10 43.45 191.80 0.90
I83-iii 86 0-15 33.25 154.67 0.80
I84-iii 86 0-20 29.89 147.50 0.77
I85-iii 86 0-25 25.87 136.64 0.78
I11-iii 115 0-5 35.80 154.40 1.30
I12-iii 115 0-10 22.30 113.80 1.05
I13-iii 115 0-15 17.49 87.20 0.91
I14-iii 115 0-20 14.47 69.80 0.81
I31-iii 135 0-5 39.40 63.00 3.55
I32-iii 135 0-10 30.26 49.70 2.09
I33-iii 135 0-15 26.25 40.73 1.96
I34-iii 135 0-20 23.84 39.65 1.60 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min iii- Particle size range 75-150 µm
265
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
I21-iii 20 0-5 <0.001 <0.003 0.108 0.108
I22-iii 20 0-10 <0.001 <0.003 0.054 0.054
I23-iii 20 0-15 <0.001 <0.003 0.036 0.036
I24-iii 20 0-20 <0.001 <0.003 0.027 0.027
I25-iii 20 0-25 <0.001 <0.003 0.224 0.224
I26-iii 20 0-30 <0.001 <0.003 0.186 0.186
I27-iii 20 0-35 <0.001 <0.003 0.160 0.160
I41-iii 40 0-5 <0.001 <0.003 0.193 0.193
I42-iii 40 0-10 <0.001 <0.003 0.096 0.096
I44-iii 40 0-15 <0.001 <0.003 0.048 0.048
I45-iii 40 0-20 <0.001 <0.003 0.039 0.039
I46-iii 40 0-25 <0.001 <0.003 <0.035 nd
I61-iii 65 0-5 <0.001 0.027 0.416 0.443
I62-iii 65 0-10 <0.001 0.027 0.208 0.235
I63-iii 65 0-15 <0.001 0.029 0.139 0.168
I64-iii 65 0-20 <0.001 0.029 0.133 0.162
I65-iii 65 0-25 <0.001 0.042 0.107 0.148
I66-iii 65 0-30 <0.001 0.052 0.089 0.141
I81-iii 86 0-5 <0.001 0.039 0.504 0.543
I82-iii 86 0-10 <0.001 0.031 0.252 0.282
I83-iii 86 0-15 <0.001 0.028 0.168 0.196
I84-iii 86 0-20 <0.001 0.028 0.126 0.154
I85-iii 86 0-25 <0.001 0.035 0.101 0.136
I11-iii 115 0-5 <0.001 0.016 <0.035 0.016
I12-iii 115 0-10 <0.001 0.008 <0.035 0.008
I13-iii 115 0-15 <0.001 0.005 <0.035 0.005
I14-iii 115 0-20 <0.001 0.004 <0.035 0.004
I31-iii 135 0-5 <0.001 <0.003 0.523 0.523
I32-iii 135 0-10 <0.001 <0.003 0.261 0.261
I33-iii 135 0-15 <0.001 <0.003 0.174 0.174
I34-iii 135 0-20 <0.001 <0.003 0.131 0.131 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min iii- Particle size range 75-150 µm
266
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
I21-iii 20 0-5 0.080 1.190
I22-iii 20 0-10 0.049 1.185
I23-iii 20 0-15 0.032 1.180
I24-iii 20 0-20 <0.03 0.978
I25-iii 20 0-25 <0.03 0.911
I26-iii 20 0-30 <0.03 0.820
I27-iii 20 0-35 <0.03 0.756
I41-iii 40 0-5 0.289 2.244
I42-iii 40 0-10 0.404 1.755
I44-iii 40 0-15 0.327 1.175
I45-iii 40 0-20 0.261 1.221
I46-iii 40 0-25 0.312 1.131
I61-iii 65 0-5 0.694 2.460
I62-iii 65 0-10 0.520 2.406
I63-iii 65 0-15 0.535 2.039
I64-iii 65 0-20 0.570 1.750
I65-iii 65 0-25 0.458 1.431
I66-iii 65 0-30 0.381 1.267
I81-iii 86 0-5 0.589 2.371
I82-iii 86 0-10 0.418 1.812
I83-iii 86 0-15 0.327 1.602
I84-iii 86 0-20 0.341 1.507
I85-iii 86 0-25 0.508 1.445
I11-iii 115 0-5 0.483 2.288
I12-iii 115 0-10 0.250 1.746
I13-iii 115 0-15 0.167 1.413
I14-iii 115 0-20 0.125 1.163
I31-iii 135 0-5 1.487 2.524
I32-iii 135 0-10 0.744 1.638
I33-iii 135 0-15 0.499 1.329
I34-iii 135 0-20 0.374 1.296 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min iii- Particle size range 75-150 µm
267
Table C2– Industrial site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
I21-ii 20 0-5 37.50 24.00 0.47
I22-ii 20 0-10 33.95 30.40 0.48
I23-ii 20 0-15 30.60 36.40 0.45
I24-ii 20 0-20 28.68 37.10 0.42
I25-ii 20 0-25 27.68 32.80 0.39
I26-ii 20 0-30 26.68 29.87 0.38
I27-ii 20 0-35 25.76 27.77 0.37
I41-ii 40 0-5 161.80 90.40 1.16
I42-ii 40 0-10 99.70 68.20 1.04
I44-ii 40 0-15 57.50 39.40 0.69
I45-ii 40 0-20 51.48 42.08 0.68
I46-ii 40 0-25 47.62 39.47 0.68
I61-ii 65 0-5 132.60 36.40 1.00
I62-ii 65 0-10 107.35 29.40 0.90
I63-ii 65 0-15 78.77 52.40 0.84
I64-ii 65 0-20 63.24 60.50 0.74
I65-ii 65 0-25 54.75 55.84 0.71
I66-ii 65 0-30 48.16 48.87 0.67
I81-ii 86 0-5 65.80 75.60 0.89
I82-ii 86 0-10 43.45 71.80 0.72
I83-ii 86 0-15 33.25 65.33 0.67
I84-ii 86 0-20 29.89 71.60 0.67
I85-ii 86 0-25 25.87 60.88 0.66
I11-ii 115 0-5 35.80 90.00 1.15
I12-ii 115 0-10 22.30 50.00 1.00
I13-ii 115 0-15 17.49 51.07 0.82
I14-ii 115 0-20 14.47 43.50 0.71
I31-ii 135 0-5 39.40 24.40 0.68
I32-ii 135 0-10 30.26 26.20 0.66
I33-ii 135 0-15 26.25 75.33 0.59
I34-ii 135 0-20 23.84 62.40 0.56 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min ii- Particle size range 150-300 µm
268
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
I21-ii 20 0-5 <0.001 <0.003 0.265 0.265
I22-ii 20 0-10 <0.001 <0.003 0.214 0.214
I23-ii 20 0-15 <0.001 <0.003 0.143 0.143
I24-ii 20 0-20 <0.001 <0.003 0.107 0.107
I25-ii 20 0-25 <0.001 <0.003 0.086 0.086
I26-ii 20 0-30 <0.001 <0.003 0.071 0.071
I27-ii 20 0-35 <0.001 <0.003 0.061 0.061
I41-ii 40 0-5 <0.001 <0.003 0.049 0.049
I42-ii 40 0-10 <0.001 <0.003 <0.035 nd
I44-ii 40 0-15 <0.001 <0.003 <0.035 nd
I45-ii 40 0-20 <0.001 <0.003 <0.035 nd
I46-ii 40 0-25 <0.001 <0.003 <0.035 nd
I61-ii 65 0-5 <0.001 0.005 <0.035 0.005
I62-ii 65 0-10 <0.001 0.020 <0.035 0.020
I63-ii 65 0-15 <0.001 0.024 0.070 0.095
I64-ii 65 0-20 <0.001 0.021 0.053 0.074
I65-ii 65 0-25 <0.001 0.025 0.042 0.067
I66-ii 65 0-30 <0.001 0.025 0.035 0.061
I81-ii 86 0-5 <0.001 0.034 0.236 0.270
I82-ii 86 0-10 <0.001 0.025 0.118 0.143
I83-ii 86 0-15 <0.001 0.025 0.079 0.104
I84-ii 86 0-20 <0.001 0.025 0.059 0.084
I85-ii 86 0-25 <0.001 0.023 0.047 0.070
I11-ii 115 0-5 <0.001 0.026 <0.035 0.026
I12-ii 115 0-10 <0.001 0.013 <0.035 0.013
I13-ii 115 0-15 <0.001 0.009 <0.035 0.009
I14-ii 115 0-20 <0.001 0.007 <0.035 0.007
I31-ii 135 0-5 <0.001 <0.003 <0.035 nd
I32-ii 135 0-10 <0.001 <0.003 <0.035 nd
I33-ii 135 0-15 <0.001 <0.003 <0.035 nd
I34-ii 135 0-20 <0.001 <0.003 <0.035 nd Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min ii- Particle size range 150-300 µm
269
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
I21-ii 20 0-5 <0.03 1.133
I22-ii 20 0-10 <0.03 1.133
I23-ii 20 0-15 <0.03 1.133
I24-ii 20 0-20 <0.03 0.941
I25-ii 20 0-25 <0.03 0.875
I26-ii 20 0-30 <0.03 0.785
I27-ii 20 0-35 <0.03 0.728
I41-ii 40 0-5 0.271 2.208
I42-ii 40 0-10 0.183 1.666
I44-ii 40 0-15 0.092 1.120
I45-ii 40 0-20 0.314 1.154
I46-ii 40 0-25 0.302 1.068
I61-ii 65 0-5 <0.03 2.186
I62-ii 65 0-10 0.099 2.203
I63-ii 65 0-15 0.137 1.873
I64-ii 65 0-20 0.251 1.600
I65-ii 65 0-25 0.201 1.302
I66-ii 65 0-30 0.167 1.157
I81-ii 86 0-5 0.159 2.258
I82-ii 86 0-10 0.150 1.723
I83-ii 86 0-15 0.118 1.531
I84-ii 86 0-20 0.155 1.453
I85-ii 86 0-25 0.345 1.397
I11-ii 115 0-5 0.342 2.239
I12-ii 115 0-10 0.171 1.680
I13-ii 115 0-15 0.114 1.357
I14-ii 115 0-20 0.086 1.123
I31-ii 135 0-5 0.291 2.333
I32-ii 135 0-10 0.146 1.533
I33-ii 135 0-15 0.097 1.252
I34-ii 135 0-20 0.102 1.237 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min ii- Particle size range 150-300 µm
270
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range >300 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
I21-i 20 0-5 37.50 31.60 0.63
I22-i 20 0-10 33.95 30.20 0.59
I23-i 20 0-15 30.60 32.00 0.59
I24-i 20 0-20 28.68 28.60 0.54
I25-i 20 0-25 27.68 24.88 0.52
I26-i 20 0-30 26.68 23.47 0.49
I27-i 20 0-35 25.76 20.86 0.49
I41-i 40 0-5 161.80 44.00 1.56
I42-i 40 0-10 99.70 42.40 1.18
I44-i 40 0-15 57.50 27.90 0.75
I45-i 40 0-20 51.48 25.36 0.73
I46-i 40 0-25 47.62 24.87 0.71
I61-i 65 0-5 132.60 53.60 1.16
I62-i 65 0-10 107.35 29.00 1.08
I63-i 65 0-15 78.77 27.07 0.98
I64-i 65 0-20 63.24 27.70 0.96
I65-i 65 0-25 54.75 31.92 0.92
I66-i 65 0-30 48.16 28.40 0.85
I81-i 86 0-5 65.80 61.60 1.26
I82-i 86 0-10 43.45 49.60 0.96
I83-i 86 0-15 33.25 43.07 0.82
I84-i 86 0-20 29.89 49.40 0.80
I85-i 86 0-25 25.87 40.88 0.76
I11-i 115 0-5 35.80 24.00 0.99
I12-i 115 0-10 22.30 19.20 0.84
I13-i 115 0-15 17.49 19.33 0.70
I14-i 115 0-20 14.47 16.80 1.23
I31-i 135 0-5 39.40 19.20 0.57
I32-i 135 0-10 30.26 16.40 0.69
I33-i 135 0-15 26.25 21.20 0.71
I34-i 135 0-20 23.84 21.88 0.64 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min i- Particle size range >300 µm
271
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range >300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
I21-i 20 0-5 <0.001 <0.003 0.097 0.097
I22-i 20 0-10 <0.001 <0.003 0.049 0.049
I23-i 20 0-15 <0.001 <0.003 0.032 0.032
I24-i 20 0-20 <0.001 <0.003 0.024 0.024
I25-i 20 0-25 <0.001 <0.003 0.019 0.019
I26-i 20 0-30 <0.001 <0.003 0.016 0.016
I27-i 20 0-35 <0.001 <0.003 0.043 0.043
I41-i 40 0-5 <0.001 <0.003 <0.035 nd
I42-i 40 0-10 <0.001 <0.003 <0.035 nd
I44-i 40 0-15 <0.001 <0.003 <0.035 nd
I45-i 40 0-20 <0.001 <0.003 <0.035 nd
I46-i 40 0-25 <0.001 <0.003 <0.035 nd
I61-i 65 0-5 <0.001 0.046 <0.035 0.063
I62-i 65 0-10 <0.001 0.029 <0.035 0.038
I63-i 65 0-15 <0.001 0.041 <0.035 0.047
I64-i 65 0-20 <0.001 0.040 0.042 0.082
I65-i 65 0-25 <0.001 0.049 0.120 0.169
I66-i 65 0-30 <0.001 0.053 0.112 0.152
I81-i 86 0-5 <0.001 0.065 0.572 0.638
I82-i 86 0-10 <0.001 0.044 0.304 0.348
I83-i 86 0-15 <0.001 0.042 0.203 0.246
I84-i 86 0-20 <0.001 0.041 0.153 0.193
I85-i 86 0-25 <0.001 0.035 0.122 0.157
I11-i 115 0-5 <0.001 0.004 <0.035 0.004
I12-i 115 0-10 <0.001 <0.003 <0.035 0.002
I13-i 115 0-15 <0.001 <0.003 <0.035 0.001
I14-i 115 0-20 <0.001 <0.003 <0.035 0.001
I31-i 135 0-5 <0.001 <0.003 <0.035 nd
I32-i 135 0-10 <0.001 <0.003 <0.035 nd
I33-i 135 0-15 <0.001 <0.003 <0.035 nd
I34-i 135 0-20 <0.001 <0.003 <0.035 nd Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min i- Particle size range >300 µm
272
Table C2 – Industrial site wash-off data –wet sieved wash-off samples Particle size range >300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
I21-i 20 0-5 <0.03 1.109
I22-i 20 0-10 <0.03 1.127
I23-i 20 0-15 <0.03 1.128
I24-i 20 0-20 <0.03 0.922
I25-i 20 0-25 <0.03 0.873
I26-i 20 0-30 <0.03 0.786
I27-i 20 0-35 <0.03 0.748
I41-i 40 0-5 0.731 2.317
I42-i 40 0-10 0.378 1.710
I44-i 40 0-15 0.189 1.133
I45-i 40 0-20 0.446 1.242
I46-i 40 0-25 0.398 1.127
I61-i 65 0-5 <0.03 2.193
I62-i 65 0-10 0.037 2.204
I63-i 65 0-15 0.188 1.871
I64-i 65 0-20 0.326 1.598
I65-i 65 0-25 0.260 1.309
I66-i 65 0-30 0.217 1.163
I81-i 86 0-5 0.483 2.374
I82-i 86 0-10 0.365 1.815
I83-i 86 0-15 0.243 1.604
I84-i 86 0-20 0.319 1.521
I85-i 86 0-25 0.476 1.447
I11-i 115 0-5 0.044 2.172
I12-i 115 0-10 <0.03 1.644
I13-i 115 0-15 <0.03 1.322
I14-i 115 0-20 <0.03 1.097
I31-i 135 0-5 <0.03 2.112
I32-i 135 0-10 <0.03 1.414
I33-i 135 0-15 <0.03 1.175
I34-i 135 0-20 <0.03 1.153 Note: I- Industrial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min i- Particle size range >300 µm
273
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range <1 µm
Identification Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TDS (mg/L)
DOC (mg/L)
C21-v 20 0-5 502.00 792.50 61.44 C22-v 20 0-10 466.50 771.25 43.82 C23-v 20 0-15 457.00 757.50 35.82 C24-v 20 0-20 449.75 767.50 31.02 C25-v 20 0-25 442.40 770.50 27.80 C26-v 20 0-30 438.50 759.58 25.73 C27-v 20 0-35 434.14 715.00 24.02 C28-v 20 0-40 430.00 647.81 22.38 C41-v 40 0-5 482.00 985.00 49.05 C42-v 40 0-10 486.00 913.75 38.93 C43-v 40 0-15 474.33 876.67 33.11 C44-v 40 0-20 477.50 863.13 29.02 C45-v 40 0-25 470.60 845.50 25.99 C46-v 40 0-30 466.83 828.33 23.61 C47-v 40 0-35 456.43 812.50 21.77 C61-v 65 0-5 612.00 340.00 18.18 C62-v 65 0-10 541.50 283.75 14.68 C63-v 65 0-15 518.33 261.67 12.97 C64-v 65 0-20 503.75 248.12 11.73 C65-v 65 0-25 500.20 239.50 10.83 C66-v 65 0-30 493.67 230.00 10.47 C81-v 86 0-5 498.00 872.50 26.28 C82-v 86 0-10 486.00 852.50 19.00 C83-v 86 0-15 485.33 847.50 15.89 C84-v 86 0-20 482.50 825.00 14.15 C85-v 86 0-25 494.80 796.00 12.93 C11-v 115 0-5 448.00 895.00 35.01 C12-v 115 0-10 470.50 883.75 24.48 C13-v 115 0-15 464.33 875.83 19.88 C14-v 115 0-20 468.75 869.38 17.03 C31-v 135 0-5 387.00 987.50 10.93 C32-v 135 0-10 370.50 970.00 8.31 C33-v 135 0-15 357.67 904.17 7.28 C34-v 135 0-20 348.00 856.88 6.69
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 v- Particle size range <1 µm
274
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range <1 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
C21-v 20 0-5 0.122 0.498 11.196 11.815 C22-v 20 0-10 0.080 0.370 8.094 8.544 C23-v 20 0-15 0.063 0.313 6.771 7.147 C24-v 20 0-20 0.054 0.281 5.941 6.275 C25-v 20 0-25 0.048 0.261 5.402 5.711 C26-v 20 0-30 0.044 0.240 4.995 5.278 C27-v 20 0-35 0.041 0.227 4.726 4.994 C28-v 20 0-40 0.038 0.214 4.385 4.638 C41-v 40 0-5 0.058 0.567 9.948 10.573 C42-v 40 0-10 0.043 0.369 7.577 7.989 C43-v 40 0-15 0.036 0.294 6.378 6.708 C44-v 40 0-20 0.032 0.254 5.606 5.893 C45-v 40 0-25 0.029 0.250 5.018 5.297 C46-v 40 0-30 0.027 0.229 4.573 4.829 C47-v 40 0-35 0.025 0.214 4.200 4.439 C61-v 65 0-5 0.045 0.301 3.624 3.970 C62-v 65 0-10 0.035 0.249 2.891 3.175 C63-v 65 0-15 0.030 0.230 2.474 2.733 C64-v 65 0-20 0.027 0.220 2.131 2.377 C65-v 65 0-25 0.024 0.210 1.899 2.134 C66-v 65 0-30 0.022 0.199 1.705 1.926 C81-v 86 0-5 0.035 0.206 4.963 5.204 C82-v 86 0-10 0.027 0.194 3.464 3.685 C83-v 86 0-15 0.023 0.187 2.715 2.925 C84-v 86 0-20 0.022 0.182 2.299 2.502 C85-v 86 0-25 0.020 0.177 2.011 2.208 C11-v 115 0-5 0.036 0.168 6.225 6.430 C12-v 115 0-10 0.029 0.153 4.514 4.695 C13-v 115 0-15 0.026 0.156 3.670 3.851 C14-v 115 0-20 0.024 0.146 3.064 3.233 C31-v 135 0-5 0.018 0.199 1.993 2.210 C32-v 135 0-10 0.015 0.162 1.545 1.722 C33-v 135 0-15 0.013 0.148 1.218 1.379 C34-v 135 0-20 0.012 0.141 1.089 1.242
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 v- Particle size range <1 µm
275
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range <1 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
C21-v 20 0-5 0.281 0.389 C22-v 20 0-10 0.278 0.361 C23-v 20 0-15 0.275 0.332 C24-v 20 0-20 0.273 0.318 C25-v 20 0-25 0.269 0.307 C26-v 20 0-30 0.268 0.300 C27-v 20 0-35 0.266 0.294 C28-v 20 0-40 0.235 0.278 C41-v 40 0-5 0.258 0.270 C42-v 40 0-10 0.257 0.267 C43-v 40 0-15 0.257 0.264 C44-v 40 0-20 0.257 0.265 C45-v 40 0-25 0.256 0.265 C46-v 40 0-30 0.256 0.265 C47-v 40 0-35 0.256 0.263 C61-v 65 0-5 0.405 0.412 C62-v 65 0-10 0.400 0.407 C63-v 65 0-15 0.356 0.387 C64-v 65 0-20 0.334 0.374 C65-v 65 0-25 0.320 0.363 C66-v 65 0-30 0.269 0.326 C81-v 86 0-5 0.266 0.379 C82-v 86 0-10 0.264 0.372 C83-v 86 0-15 0.264 0.362 C84-v 86 0-20 0.263 0.358 C85-v 86 0-25 0.214 0.320 C11-v 115 0-5 0.370 0.374 C12-v 115 0-10 0.315 0.368 C13-v 115 0-15 0.295 0.332 C14-v 115 0-20 0.225 0.287 C31-v 135 0-5 1.492 1.512 C32-v 135 0-10 0.875 1.345 C33-v 135 0-15 0.678 1.038 C34-v 135 0-20 0.571 0.845
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 v- Particle size range <1 µm
276
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
C21-iv 20 0-5 502.00 217.60 3.66 C22-iv 20 0-10 466.50 178.00 2.73 C23-iv 20 0-15 457.00 163.11 2.35 C24-iv 20 0-20 449.75 140.93 2.25 C25-iv 20 0-25 442.40 128.11 2.13 C26-iv 20 0-30 438.50 118.42 2.03 C27-iv 20 0-35 434.14 110.02 1.99 C28-iv 20 0-40 430.00 103.22 1.81 C41-iv 40 0-5 482.00 278.00 4.06 C42-iv 40 0-10 486.00 207.80 2.89 C43-iv 40 0-15 474.33 168.93 2.50 C44-iv 40 0-20 477.50 149.40 2.13 C45-iv 40 0-25 470.60 127.92 1.87 C46-iv 40 0-30 466.83 111.20 1.69 C47-iv 40 0-35 456.43 98.17 1.53 C61-iv 65 0-5 612.00 228.40 3.36 C62-iv 65 0-10 541.50 135.20 2.72 C63-iv 65 0-15 518.33 100.40 2.18 C64-iv 65 0-20 503.75 80.70 1.80 C65-iv 65 0-25 500.20 70.80 1.61 C66-iv 65 0-30 493.67 61.87 1.51 C81-iv 86 0-5 498.00 78.80 2.34 C82-iv 86 0-10 486.00 77.40 1.88 C83-iv 86 0-15 485.33 62.13 1.46 C84-iv 86 0-20 482.50 52.50 1.21 C85-iv 86 0-25 494.80 45.12 1.30 C11-iv 115 0-5 448.00 94.80 3.22 C12-iv 115 0-10 470.50 93.80 2.63 C13-iv 115 0-15 464.33 78.93 2.13 C14-iv 115 0-20 468.75 69.50 1.87 C31-iv 135 0-5 387.00 158.80 0.95 C32-iv 135 0-10 370.50 92.80 1.28 C33-iv 135 0-15 357.67 66.80 1.23 C34-iv 135 0-20 348.00 52.50 1.02
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 iv- Particle size range 1-75 µm
277
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
C21-iv 20 0-5 <0.001 0.095 3.245 3.339 C22-iv 20 0-10 <0.001 0.071 2.258 2.329 C23-iv 20 0-15 <0.001 0.062 1.707 1.769 C24-iv 20 0-20 <0.001 0.054 1.442 1.496 C25-iv 20 0-25 <0.001 0.052 1.187 1.239 C26-iv 20 0-30 <0.001 0.058 1.078 1.136 C27-iv 20 0-35 <0.001 0.060 0.960 1.020 C28-iv 20 0-40 <0.001 0.062 0.875 0.937 C41-iv 40 0-5 <0.001 0.105 1.679 1.784 C42-iv 40 0-10 <0.001 0.091 1.305 1.396 C43-iv 40 0-15 <0.001 0.082 0.968 1.050 C44-iv 40 0-20 <0.001 0.078 0.782 0.859 C45-iv 40 0-25 <0.001 0.066 0.644 0.711 C46-iv 40 0-30 <0.001 0.072 0.552 0.624 C47-iv 40 0-35 <0.001 0.074 0.474 0.548 C61-iv 65 0-5 <0.001 0.049 1.675 1.723 C62-iv 65 0-10 <0.001 0.058 0.961 1.019 C63-iv 65 0-15 <0.001 0.059 0.797 0.856 C64-iv 65 0-20 <0.001 0.051 0.713 0.764 C65-iv 65 0-25 <0.001 0.051 0.689 0.739 C66-iv 65 0-30 <0.001 0.059 0.594 0.653 C81-iv 86 0-5 <0.001 0.100 1.469 1.569 C82-iv 86 0-10 <0.001 0.080 0.937 1.017 C83-iv 86 0-15 <0.001 0.075 0.629 0.704 C84-iv 86 0-20 <0.001 0.068 0.593 0.661 C85-iv 86 0-25 <0.001 0.068 0.489 0.558 C11-iv 115 0-5 <0.001 0.085 2.800 2.885 C12-iv 115 0-10 <0.001 0.081 1.423 1.503 C13-iv 115 0-15 <0.001 0.075 1.003 1.078 C14-iv 115 0-20 <0.001 0.085 0.788 0.873 C31-iv 135 0-5 <0.001 0.053 0.559 0.613 C32-iv 135 0-10 <0.001 0.064 0.297 0.361 C33-iv 135 0-15 <0.001 0.069 0.207 0.276 C34-iv 135 0-20 <0.001 0.068 0.210 0.277
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 iv- Particle size range 1-75 µm
278
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 1-75 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
C21-iv 20 0-5 1.003 1.114 C22-iv 20 0-10 0.942 1.074 C23-iv 20 0-15 0.918 1.072 C24-iv 20 0-20 0.887 1.062 C25-iv 20 0-25 0.829 0.986 C26-iv 20 0-30 0.775 0.921 C27-iv 20 0-35 0.722 0.863 C28-iv 20 0-40 0.633 0.768 C41-iv 40 0-5 0.209 0.288 C42-iv 40 0-10 0.397 0.460 C43-iv 40 0-15 0.391 0.475 C44-iv 40 0-20 0.383 0.457 C45-iv 40 0-25 0.363 0.446 C46-iv 40 0-30 0.347 0.423 C47-iv 40 0-35 0.335 0.405 C61-iv 65 0-5 0.531 0.564 C62-iv 65 0-10 0.496 0.761 C63-iv 65 0-15 0.482 0.827 C64-iv 65 0-20 0.473 0.799 C65-iv 65 0-25 0.430 0.748 C66-iv 65 0-30 0.361 0.639 C81-iv 86 0-5 0.676 1.053 C82-iv 86 0-10 0.630 0.956 C83-iv 86 0-15 0.545 0.914 C84-iv 86 0-20 0.474 0.869 C85-iv 86 0-25 0.382 0.760 C11-iv 115 0-5 0.888 0.957 C12-iv 115 0-10 0.901 0.897 C13-iv 115 0-15 0.797 0.817 C14-iv 115 0-20 0.602 0.744 C31-iv 135 0-5 1.023 1.843 C32-iv 135 0-10 0.961 1.548 C33-iv 135 0-15 0.837 1.289 C34-iv 135 0-20 0.756 1.134
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 iv- Particle size range 1-75 µm
279
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
C21-iii 20 0-5 502.00 58.40 1.25 C22-iii 20 0-10 466.50 49.20 1.00 C23-iii 20 0-15 457.00 46.13 0.90 C24-iii 20 0-20 449.75 42.60 0.85 C25-iii 20 0-25 442.40 41.84 0.80 C26-iii 20 0-30 438.50 39.60 0.77 C27-iii 20 0-35 434.14 37.89 0.77 C28-iii 20 0-40 430.00 36.65 0.75 C41-iii 40 0-5 482.00 118.80 1.41 C42-iii 40 0-10 486.00 109.60 1.18 C43-iii 40 0-15 474.33 92.67 1.07 C44-iii 40 0-20 477.50 78.20 0.94 C45-iii 40 0-25 470.60 67.60 0.84 C46-iii 40 0-30 466.83 58.73 0.80 C47-iii 40 0-35 456.43 55.14 0.78 C61-iii 65 0-5 612.00 83.20 1.17 C62-iii 65 0-10 541.50 102.60 1.07 C63-iii 65 0-15 518.33 76.53 0.87 C64-iii 65 0-20 503.75 61.00 0.78 C65-iii 65 0-25 500.20 50.88 0.69 C66-iii 65 0-30 493.67 45.40 0.69 C81-iii 86 0-5 498.00 46.00 1.02 C82-iii 86 0-10 486.00 30.00 0.85 C83-iii 86 0-15 485.33 27.20 0.82 C84-iii 86 0-20 482.50 24.40 0.72 C85-iii 86 0-25 494.80 21.92 0.67 C11-iii 115 0-5 448.00 35.60 0.72 C12-iii 115 0-10 470.50 50.00 0.76 C13-iii 115 0-15 464.33 60.53 0.74 C14-iii 115 0-20 468.75 62.10 0.68 C31-iii 135 0-5 387.00 47.60 0.93 C32-iii 135 0-10 370.50 34.80 0.64 C33-iii 135 0-15 357.67 26.27 0.52 C34-iii 135 0-20 348.00 23.20 0.46
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 iii- Particle size range 75-150 µm
280
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
C21-iii 20 0-5 <0.001 0.102 0.214 0.315 C22-iii 20 0-10 <0.001 0.098 0.139 0.237 C23-iii 20 0-15 <0.001 0.110 0.207 0.317 C24-iii 20 0-20 <0.001 0.112 0.195 0.307 C25-iii 20 0-25 <0.001 0.110 0.156 0.266 C26-iii 20 0-30 <0.001 0.109 0.159 0.269 C27-iii 20 0-35 <0.001 0.110 0.150 0.260 C28-iii 20 0-40 <0.001 0.109 0.144 0.252 C41-iii 40 0-5 <0.001 0.015 0.647 0.662 C42-iii 40 0-10 <0.001 0.023 0.324 0.347 C43-iii 40 0-15 <0.001 0.042 0.216 0.258 C44-iii 40 0-20 <0.001 0.060 0.162 0.222 C45-iii 40 0-25 <0.001 0.068 0.129 0.197 C46-iii 40 0-30 <0.001 0.074 0.132 0.206 C47-iii 40 0-35 <0.001 0.079 0.113 0.192 C61-iii 65 0-5 <0.001 0.159 0.797 0.955 C62-iii 65 0-10 <0.001 0.161 0.511 0.671 C63-iii 65 0-15 <0.001 0.166 0.381 0.547 C64-iii 65 0-20 <0.001 0.168 0.286 0.454 C65-iii 65 0-25 <0.001 0.164 0.228 0.393 C66-iii 65 0-30 <0.001 0.137 0.190 0.328 C81-iii 86 0-5 <0.001 0.156 0.716 0.871 C82-iii 86 0-10 <0.001 0.156 0.358 0.514 C83-iii 86 0-15 <0.001 0.158 0.239 0.396 C84-iii 86 0-20 <0.001 0.147 0.179 0.326 C85-iii 86 0-25 <0.001 0.147 0.364 0.512 C11-iii 115 0-5 <0.001 0.143 9.401 9.544 C12-iii 115 0-10 <0.001 0.129 4.701 4.830 C13-iii 115 0-15 <0.001 0.125 3.144 3.269 C14-iii 115 0-20 <0.001 0.106 2.365 2.471 C31-iii 135 0-5 <0.001 0.108 0.237 0.345 C32-iii 135 0-10 <0.001 0.105 0.130 0.235 C33-iii 135 0-15 <0.001 0.107 0.087 0.194 C34-iii 135 0-20 <0.001 0.111 0.075 0.186
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 iii- Particle size range 75-150 µm
281
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 75-150 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
C21-iii 20 0-5 <0.03 1.158 C22-iii 20 0-10 <0.03 1.143 C23-iii 20 0-15 <0.03 1.139 C24-iii 20 0-20 <0.03 1.147 C25-iii 20 0-25 <0.03 1.089 C26-iii 20 0-30 <0.03 1.037 C27-iii 20 0-35 <0.03 0.983 C28-iii 20 0-40 <0.03 0.881 C41-iii 40 0-5 0.105 0.240 C42-iii 40 0-10 0.053 0.438 C43-iii 40 0-15 0.035 0.501 C44-iii 40 0-20 0.026 0.499 C45-iii 40 0-25 0.021 0.488 C46-iii 40 0-30 0.018 0.468 C47-iii 40 0-35 0.015 0.458 C61-iii 65 0-5 <0.03 0.652 C62-iii 65 0-10 <0.03 0.915 C63-iii 65 0-15 <0.03 0.995 C64-iii 65 0-20 <0.03 1.030 C65-iii 65 0-25 <0.03 0.973 C66-iii 65 0-30 <0.03 0.822 C81-iii 86 0-5 <0.03 1.112 C82-iii 86 0-10 <0.03 1.057 C83-iii 86 0-15 <0.03 1.067 C84-iii 86 0-20 <0.03 1.054 C85-iii 86 0-25 <0.03 0.919 C11-iii 115 0-5 <0.03 1.187 C12-iii 115 0-10 0.102 1.130 C13-iii 115 0-15 0.121 1.007 C14-iii 115 0-20 0.091 0.918 C31-iii 135 0-5 <0.03 3.296 C32-iii 135 0-10 <0.03 2.830 C33-iii 135 0-15 <0.03 2.292 C34-iii 135 0-20 <0.03 2.005
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 iii- Particle size range 75-150 µm
282
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
C21-ii 20 0-5 502.00 32.40 0.97 C22-ii 20 0-10 466.50 24.40 0.99 C23-ii 20 0-15 457.00 22.13 0.86 C24-ii 20 0-20 449.75 21.20 0.79 C25-ii 20 0-25 442.40 20.64 0.74 C26-ii 20 0-30 438.50 19.07 0.70 C27-ii 20 0-35 434.14 19.03 0.66 C28-ii 20 0-40 430.00 18.95 0.65 C41-ii 40 0-5 482.00 39.20 0.83 C42-ii 40 0-10 486.00 36.60 0.77 C43-ii 40 0-15 474.33 33.07 0.75 C44-ii 40 0-20 477.50 32.40 0.70 C45-ii 40 0-25 470.60 28.16 0.64 C46-ii 40 0-30 466.83 26.27 0.61 C47-ii 40 0-35 456.43 24.46 0.60 C61-ii 65 0-5 612.00 17.20 0.91 C62-ii 65 0-10 541.50 13.40 0.92 C63-ii 65 0-15 518.33 24.27 0.86 C64-ii 65 0-20 503.75 23.00 0.84 C65-ii 65 0-25 500.20 19.68 0.75 C66-ii 65 0-30 493.67 17.67 0.73 C81-ii 86 0-5 498.00 23.60 1.04 C82-ii 86 0-10 486.00 23.00 0.75 C83-ii 86 0-15 485.33 18.80 0.70 C84-ii 86 0-20 482.50 16.90 0.63 C85-ii 86 0-25 494.80 15.20 0.62 C11-ii 115 0-5 448.00 16.80 0.60 C12-ii 115 0-10 470.50 20.80 0.68 C13-ii 115 0-15 464.33 26.93 0.64 C14-ii 115 0-20 468.75 25.40 0.58 C31-ii 135 0-5 387.00 21.60 0.67 C32-ii 135 0-10 370.50 16.40 0.54 C33-ii 135 0-15 357.67 12.93 0.47 C34-ii 135 0-20 348.00 11.30 0.43
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 ii- Particle size range 150-300 µm
283
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm (parameters continued from previous page)
Identification Rainfall intensity
mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
C21-ii 20 0-5 <0.001 0.108 0.080 0.188 C22-ii 20 0-10 <0.001 0.100 0.040 0.140 C23-ii 20 0-15 <0.001 0.105 <0.035 0.105 C24-ii 20 0-20 <0.001 0.109 <0.035 0.109 C25-ii 20 0-25 <0.001 0.109 0.051 0.160 C26-ii 20 0-30 <0.001 0.113 0.043 0.156 C27-ii 20 0-35 <0.001 0.112 0.037 0.149 C28-ii 20 0-40 <0.001 0.113 0.037 0.150 C41-ii 40 0-5 <0.001 0.110 0.219 0.329 C42-ii 40 0-10 <0.001 0.112 0.131 0.243 C43-ii 40 0-15 <0.001 0.107 0.087 0.194 C44-ii 40 0-20 <0.001 0.106 0.065 0.172 C45-ii 40 0-25 <0.001 0.105 0.052 0.157 C46-ii 40 0-30 <0.001 0.104 0.044 0.147 C47-ii 40 0-35 <0.001 0.103 0.037 0.140 C61-ii 65 0-5 <0.001 0.128 0.145 0.272 C62-ii 65 0-10 <0.001 0.155 0.129 0.283 C63-ii 65 0-15 <0.001 0.133 0.086 0.219 C64-ii 65 0-20 <0.001 0.137 0.103 0.241 C65-ii 65 0-25 <0.001 0.141 0.083 0.223 C66-ii 65 0-30 <0.001 0.142 0.069 0.211 C81-ii 86 0-5 <0.001 0.145 1.053 1.198 C82-ii 86 0-10 <0.001 0.146 0.526 0.672 C83-ii 86 0-15 <0.001 0.153 0.351 0.503 C84-ii 86 0-20 <0.001 0.149 0.263 0.412 C85-ii 86 0-25 <0.001 0.149 0.211 0.360 C11-ii 115 0-5 <0.001 0.153 <0.035 0.153 C12-ii 115 0-10 <0.001 0.129 0.441 0.570 C13-ii 115 0-15 <0.001 0.125 0.294 0.419 C14-ii 115 0-20 <0.001 0.119 0.220 0.339 C31-ii 135 0-5 <0.001 0.113 0.286 0.399 C32-ii 135 0-10 <0.001 0.115 0.143 0.258 C33-ii 135 0-15 <0.001 0.116 0.095 0.212 C34-ii 135 0-20 <0.001 0.115 0.071 0.187
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 ii- Particle size range 150-300 µm
284
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range 150-300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
C21-ii 20 0-5 <0.03 1.126 C22-ii 20 0-10 <0.03 1.064 C23-ii 20 0-15 <0.03 1.078 C24-ii 20 0-20 <0.03 1.088 C25-ii 20 0-25 <0.03 1.044 C26-ii 20 0-30 <0.03 0.991 C27-ii 20 0-35 <0.03 0.940 C28-ii 20 0-40 <0.03 0.842 C41-ii 40 0-5 <0.03 0.150 C42-ii 40 0-10 <0.03 0.438 C43-ii 40 0-15 <0.03 0.488 C44-ii 40 0-20 <0.03 0.490 C45-ii 40 0-25 <0.03 0.485 C46-ii 40 0-30 <0.03 0.465 C47-ii 40 0-35 <0.03 0.456 C61-ii 65 0-5 <0.03 0.572 C62-ii 65 0-10 <0.03 0.873 C63-ii 65 0-15 <0.03 0.962 C64-ii 65 0-20 <0.03 1.013 C65-ii 65 0-25 <0.03 0.957 C66-ii 65 0-30 <0.03 0.803 C81-ii 86 0-5 <0.03 1.151 C82-ii 86 0-10 <0.03 1.088 C83-ii 86 0-15 <0.03 1.083 C84-ii 86 0-20 <0.03 1.092 C85-ii 86 0-25 <0.03 0.940 C11-ii 115 0-5 <0.03 1.062 C12-ii 115 0-10 0.075 1.107 C13-ii 115 0-15 0.050 0.971 C14-ii 115 0-20 0.037 0.887 C31-ii 135 0-5 <0.03 3.270 C32-ii 135 0-10 <0.03 2.743 C33-ii 135 0-15 <0.03 2.233 C34-ii 135 0-20 <0.03 1.926
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 ii- Particle size range 150-300 µm
285
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range >300 µm
Identification
Rainfall intensity mm/hr
Duration min
EC (µS/cm)
TSS (mg/L)
TOC (mg/L)
C21-i 20 0-5 502.00 29.60 1.95 C22-i 20 0-10 466.50 21.00 1.30 C23-i 20 0-15 457.00 18.80 1.14 C24-i 20 0-20 449.75 18.40 1.00 C25-i 20 0-25 442.40 17.20 0.92 C26-i 20 0-30 438.50 15.87 0.88 C27-i 20 0-35 434.14 14.86 0.81 C28-i 20 0-40 430.00 14.40 0.78 C41-i 40 0-5 482.00 33.20 2.00 C42-i 40 0-10 486.00 30.00 1.59 C43-i 40 0-15 474.33 27.47 1.27 C44-i 40 0-20 477.50 29.00 1.09 C45-i 40 0-25 470.60 26.16 1.00 C46-i 40 0-30 466.83 23.13 0.92 C47-i 40 0-35 456.43 22.17 0.88 C61-i 65 0-5 612.00 23.60 3.33 C62-i 65 0-10 541.50 21.60 2.28 C63-i 65 0-15 518.33 19.33 1.94 C64-i 65 0-20 503.75 17.50 1.68 C65-i 65 0-25 500.20 14.96 1.43 C66-i 65 0-30 493.67 13.27 1.28 C81-i 86 0-5 498.00 16.40 1.27 C82-i 86 0-10 486.00 13.60 1.00 C83-i 86 0-15 485.33 11.87 1.81 C84-i 86 0-20 482.50 12.40 1.53 C85-i 86 0-25 494.80 11.52 1.32 C11-i 115 0-5 448.00 26.40 1.03 C12-i 115 0-10 470.50 19.80 0.93 C13-i 115 0-15 464.33 39.20 0.98 C14-i 115 0-20 468.75 36.20 0.82 C31-i 135 0-5 387.00 16.40 0.92 C32-i 135 0-10 370.50 10.60 0.69 C33-i 135 0-15 357.67 8.40 0.62 C34-i 135 0-20 348.00 7.70 0.56
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 i- Particle size range >300 µm
286
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range >300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
NO2-
(mg/L) NO3
-
(mg/L) TKN
(mg/L) TN
(mg/L)
C21-i 20 0-5 <0.001 0.104 0.266 0.370 C22-i 20 0-10 <0.001 0.098 0.133 0.231 C23-i 20 0-15 <0.001 0.099 0.089 0.188 C24-i 20 0-20 <0.001 0.099 0.100 0.199 C25-i 20 0-25 <0.001 0.099 0.080 0.179 C26-i 20 0-30 <0.001 0.098 0.072 0.170 C27-i 20 0-35 <0.001 0.099 0.061 0.161 C28-i 20 0-40 <0.001 0.101 0.054 0.154 C41-i 40 0-5 <0.001 0.107 0.422 0.528 C42-i 40 0-10 <0.001 0.101 0.411 0.512 C43-i 40 0-15 <0.001 0.096 0.274 0.370 C44-i 40 0-20 <0.001 0.090 0.206 0.296 C45-i 40 0-25 <0.001 0.087 0.165 0.252 C46-i 40 0-30 <0.001 0.084 0.137 0.221 C47-i 40 0-35 <0.001 0.080 0.122 0.202 C61-i 65 0-5 <0.001 0.118 2.353 2.471 C62-i 65 0-10 <0.001 0.138 3.356 3.494 C63-i 65 0-15 <0.001 0.127 3.051 3.178 C64-i 65 0-20 <0.001 0.122 2.288 2.411 C65-i 65 0-25 <0.001 0.117 1.831 1.948 C66-i 65 0-30 <0.001 0.115 1.525 1.640 C81-i 86 0-5 <0.001 0.113 0.704 0.818 C82-i 86 0-10 <0.001 0.110 0.831 0.940 C83-i 86 0-15 <0.001 0.107 0.554 0.661 C84-i 86 0-20 <0.001 0.104 0.415 0.519 C85-i 86 0-25 <0.001 0.102 0.332 0.434 C11-i 115 0-5 <0.001 0.144 0.283 0.427 C12-i 115 0-10 <0.001 0.120 0.141 0.261 C13-i 115 0-15 <0.001 0.111 0.135 0.246 C14-i 115 0-20 <0.001 0.101 0.123 0.224 C31-i 135 0-5 <0.001 0.110 0.440 0.550 C32-i 135 0-10 <0.001 0.113 0.574 0.687 C33-i 135 0-15 <0.001 0.110 0.383 0.493 C34-i 135 0-20 <0.001 0.111 0.287 0.398
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 i- Particle size range >300 µm
287
Table C3 – Commercial site wash-off data –wet sieved wash-off samples Particle size range >300 µm (parameters continued from previous page)
Identification Rainfall intensity mm/hr
Duration min
PO43-
(mg/L) TP
(mg/L)
C21-i 20 0-5 <0.03 1.129 C22-i 20 0-10 <0.03 1.099 C23-i 20 0-15 <0.03 1.090 C24-i 20 0-20 <0.03 1.103 C25-i 20 0-25 <0.03 1.050 C26-i 20 0-30 <0.03 0.998 C27-i 20 0-35 <0.03 0.945 C28-i 20 0-40 <0.03 0.843 C41-i 40 0-5 0.027 0.152 C42-i 40 0-10 0.014 0.458 C43-i 40 0-15 <0.03 0.501 C44-i 40 0-20 <0.03 0.497 C45-i 40 0-25 <0.03 0.398 C46-i 40 0-30 <0.03 0.396 C47-i 40 0-35 <0.03 0.397 C61-i 65 0-5 <0.03 0.684 C62-i 65 0-10 <0.03 1.042 C63-i 65 0-15 <0.03 1.109 C64-i 65 0-20 <0.03 1.112 C65-i 65 0-25 <0.03 1.036 C66-i 65 0-30 <0.03 0.871 C81-i 86 0-5 <0.03 1.116 C82-i 86 0-10 <0.03 1.095 C83-i 86 0-15 <0.03 1.071 C84-i 86 0-20 <0.03 1.085 C85-i 86 0-25 <0.03 0.942 C11-i 115 0-5 <0.03 1.110 C12-i 115 0-10 <0.03 1.110 C13-i 115 0-15 0.081 1.021 C14-i 115 0-20 0.061 0.915 C31-i 135 0-5 <0.03 3.286 C32-i 135 0-10 <0.03 2.828 C33-i 135 0-15 <0.03 2.308 C34-i 135 0-20 <0.03 1.966
Note: C- Commercial 2- 20 mm/hr; 4-40 mm/hr; 6-65 mm/hr; 8-85 mm/hr; 1- 115 mm/hr; 3-135 mm/hr 1- 0-5 min; 2- 0-10 min; 3-0-15 min; 4-0-20 min; 5-0-25 min; 6-0-30 min; 7-0-35 min; 8- 0-40 i- Particle size range >300 µm
288
289
APPENDIX 3
ANALYSIS OF NUTRIENT BUILD-UP
290
291
Table A - Amount of nutrients in different particle size ranges of solids build-up
at each road surface
Road surface
Particle size range (µm)
NO2-
(mg/g) NO3
-
(mg/g) TKN (mg/g)
TN (mg/g)
PO43-
(mg/g) TP (mg/g)
<1 0.020 2.84 0.43 3.30 0.21 0.70
1-75 nd 0.46 3.53 3.99 0.49 0.87
75-150 nd 0.39 6.98 7.38 0.28 0.98
150-300 nd 0.38 1.11 1.49 0.33 0.44
Residential
>300 0.002 0.37 nd 0.37 0.38 1.18
<1 0.007 0.24 0.17 0.41 0.47 0.49
1-75 nd 0.05 0.61 0.67 1.17 1.24
75-150 nd 0.01 0.78 0.78 1.09 1.44
150-300 nd nd 0.09 0.09 0.32 0.49
Industrial
>300 nd 0.01 0.15 0.15 0.33 0.51
<1 0.037 0.37 0.28 0.69 0.18 0.20
1-75 nd 0.00 1.10 1.10 0.71 0.99
75-150 0.014 0.10 5.74 5.85 0.47 1.35
150-300 nd 0.11 0.45 0.55 0.67 1.11
Commercial
>300 0.001 0.12 0.49 0.60 0.67 0.75 nd- not detected
292
293
APPENDIX 4
ANALYSIS OF NUTRIENT WASH-OFF
294
295
20 mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
0-35min 0-40min
Figure A.1a - Variation of particle size distribution with rainfall duration at
residential road surface - 20mm/hr
40mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size(µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
0-35min
Figure A.1b - Variation of particle size distribution with rainfall duration at
residential road surface - 40 mm/hr
296
65mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
Figure A.1c - Variation of particle size distribution with rainfall duration at
residential road surface - 65 mm/hr
86mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
Figure A.1d - Variation of particle size distribution with rainfall duration at
residential road surface - 86 mm/hr
297
115 mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min
Figure A.1e - Variation of particle size distribution with rainfall duration at
residential road surface - 115 mm/hr
135 mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 5-10min
10-15min 0-20min
Figure A.1f - Variation of particle size distribution with rainfall duration at
residential road surface – 135 mm/hr
298
40mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-20min 0-25min
0-30min
Figure A.2a - Variation of particle size distribution with rainfall duration at
industrial road surface -40 mm/hr
65mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
Figure A.2b - Variation of particle size distribution with rainfall duration at
industrial road surface - 65 mm/hr
299
86mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
0-25min
Figure A.2c - Variation of particle size distribution with rainfall duration at
industrial road surface - 86 mm/hr
115 mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
Figure A.2d - Variation of particle size distribution with rainfall duration at
industrial road surface - 115 mm/hr
300
135 mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00Particle size(µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
Figure A.2e - Variation of particle size distribution with rainfall duration at
industrial road surface - 135 mm/hr
20mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size/ (µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
0-35min 0-40min
Figure A.3a - Variation of particle size distribution with rainfall duration at
commercial road surface -20 mm/hr
301
40mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
0-35min
Figure A.3b - Variation of particle size distribution with rainfall duration at
commercial road surface - 40 mm/hr
65mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(
%)
0-5min 0-10min
0-15min 0-20min
0-25min 0-30min
Figure A.3c - Variation of particle size distribution with rainfall duration at
commercial road surface - 65 mm/hr
302
86mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
0-25min
Figure A.3d - Variation of particle size distribution with rainfall duration at
commercial road surface - 86 mm/hr
115mm/hr
0.0000
10.0000
20.0000
30.0000
40.0000
50.0000
60.0000
70.0000
80.0000
90.0000
100.0000
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
Figure A.3e - Variation of particle size distribution with rainfall duration at
commercial road surface - 115 mm/hr
303
135mm/hr
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
100.000
0.10 1.00 10.00 100.00 1000.00
Particle size (µm)
Cum
ulat
ive
perc
enta
ge/(%
)
0-5min 0-10min
0-15min 0-20min
Figure A.3f - Variation of particle size distribution with rainfall duration at
commercial road surface - 135 mm/hr
304
Table A - Mean concentration of nutrient parameters in the wash-off samples Particle size range(µm) Mean/Std NO2
- NO3- TKN TN PO4
3- TP (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) Residential <1 Mean 0.003 0.110 1.862 1.975 0.012 0.017 Std 0.002 0.058 0.761 0.761 0.009 0.011 1-75 Mean 0.001 0.051 0.158 0.210 0.038 0.126 Std 0.001 0.018 0.169 0.168 0.036 0.080 75-150 Mean 0.001 0.038 0.133 0.172 0.016 0.092 Std 0.001 0.015 0.208 0.209 0.013 0.055 150-300 Mean 0.001 0.030 0.008 0.030 0.009 0.109 Std 0.001 0.015 0.020 0.015 0.014 0.069 >300 Mean nd 0.029 nd 0.030 0.014 0.116 Std - 0.018 - 0.017 0.025 0.071 Industrial <1 Mean 0.002 0.149 1.164 1.315 0.479 0.590 Std 0.002 0.054 0.377 0.417 0.131 0.161 1-75 Mean nd 0.030 0.259 0.294 0.981 1.163 Std - 0.015 0.234 0.241 0.492 0.486 75-150 Mean nd 0.013 0.146 0.159 0.364 1.549 Std - 0.016 0.135 0.140 0.296 0.509 150-300 Mean nd 0.010 0.058 0.068 0.138 1.464 Std - 0.012 0.072 0.074 0.111 0.470 >300 Mean nd 0.016 0.063 0.078 0.167 1.465 Std - 0.022 0.119 0.134 0.202 0.472 Commercial <1 Mean 0.036 0.241 4.362 4.639 0.356 0.429 Std 0.021 0.094 2.437 2.536 0.242 0.298 1-75 Mean nd 0.070 1.058 1.128 0.629 0.851 Std - 0.015 0.685 0.691 0.235 0.327 75-150 Mean nd 0.114 0.808 0.922 0.017 1.086 Std - 0.040 1.799 1.806 0.035 0.643 150-300 Mean nd 0.123 0.163 0.286 0.005 1.055 Std - 0.018 0.201 0.209 0.016 0.634 >300 Mean nd 0.107 0.653 0.760 0.006 1.081 Std - 0.014 0.885 0.894 0.018 0.649
Std-standard deviation, nd-not detected
305
APPENDIX 5
ANALYIS OF NUTRIENT WASH-OFF USING PCA
306
307
Table A - Mean and standard deviation of all the parameters measured in different particle size ranges of wash-off solids TS= TSS and TOC=TOC for particles size ranges 1-75 µm, 75-150 µm, 150-300 µm, >300 µm; TS=TDS and TOC=DOC in particle size range below 1 µm
Particle size range(µm) Mean/Std TS (mg/L)
TOC (mg/L)
NO2-
(mg/L) NO3
-
(mg/L) TKN (mg/L)
TN (mg/L) PO4
3-(mg/L) TP (mg/L)
EC (µS/cm)
Residential <1 Mean 121.57 9.40 0.003 0.110 1.862 1.975 0.012 0.017 Std 36.85 2.78 0.002 0.058 0.761 0.761 0.009 0.011 1-75 Mean 9.98 1.57 0.001 0.051 0.158 0.210 0.038 0.126 Std 4.07 1.67 0.001 0.018 0.169 0.168 0.036 0.080 75-150 Mean 11.72 1.15 0.001 0.038 0.133 0.172 0.016 0.092 Std 7.51 0.20 0.001 0.015 0.208 0.209 0.013 0.055 150-300 Mean 7.76 0.84 0.001 0.030 0.008 0.030 0.009 0.109 Std 3.61 0.18 0.001 0.015 0.020 0.015 0.014 0.069 >300 Mean 6.73 0.96 nd 0.029 nd 0.030 0.014 0.116 Std 2.92 0.26 - 0.018 - 0.017 0.025 0.071
35.84 5.60
Industrial <1 Mean 311.08 8.03 0.002 0.149 1.164 1.315 0.479 0.590 Std 86.04 3.44 0.002 0.054 0.377 0.417 0.131 0.161 1-75 Mean 253.76 2.20 nd 0.030 0.259 0.294 0.981 1.163 Std 100.54 1.99 - 0.015 0.234 0.241 0.492 0.486 75-150 Mean 118.65 1.28 nd 0.013 0.146 0.159 0.364 1.549 Std 61.72 0.81 - 0.016 0.135 0.140 0.296 0.509 150-300 Mean 49.98 0.70 nd 0.010 0.058 0.068 0.138 1.464 Std 19.35 0.22 - 0.012 0.072 0.074 0.111 0.470 >300 Mean 30.85 0.83 nd 0.016 0.063 0.078 0.167 1.465 Std 11.62 0.26 - 0.022 0.119 0.134 0.202 0.472
49.09 34.46
Commercial <1 Mean 739.58 23.24 0.036 0.241 4.362 4.639 0.356 0.429 Std 233.53 12.55 0.021 0.094 2.437 2.536 0.242 0.298 1-75 Mean 117.71 2.10 nd 0.070 1.058 1.128 0.629 0.851 Std 56.12 0.75 - 0.015 0.685 0.691 0.235 0.327 75-150 Mean 54.18 0.85 nd 0.114 0.808 0.922 0.017 1.086 Std 24.89 0.20 - 0.040 1.799 1.806 0.035 0.643 150-300 Mean 22.44 0.72 nd 0.123 0.163 0.286 0.005 1.055 Std 6.62 0.14 - 0.018 0.201 0.209 0.016 0.634 >300 Mean 20.06 1.16 nd 0.107 0.653 0.760 0.006 1.081 Std 7.87 0.56 - 0.014 0.885 0.894 0.018 0.649
466.61 50.34
308
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1 2 3 4 5 6 7 8 9
Component Number
Eig
enva
lue
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1 2 3 4 5 6 7 8 9
Component Number
Eig
enva
lue
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1 2 3 4 5 6 7 8 9
Component Number
Eig
enva
lue
Figure A - PCA scree plot for the dissolved fraction of wash-off
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
1 2 3 4 5 6 7 8
Component Number
Eig
enva
lue
Figure B - PCA scree plot for the particle size range 1-150 µm of wash-off solids
Point of sharp change
309
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1 2 3 4 5 6 7 8
Component Number
Eig
enva
lue
Figure C - PCA scree plot for the particle size range >150 µm of wash-off solids
0
1
2
3
4
5
6
1 2 3 4 5 6 7 8 9 10
Component Number
Eig
enva
lue
Figure D - PCA scree plot for the PCA analysis of all the particle size ranges of wash-off solids