Analysis of Flash Flood Routing by Means of 1D - Hydraulic ... Thesis... · iv Analysis of Flash Flood Routing by Means of 1D - Hydraulic Modeling 2010 Abstract This study was conducted

Analysis of Flash Flood Routing by Means of 1D - Hydraulic Modelling

By

Zerisenay Tesfay Abraha

(Mtr.–No.: 3537870)

A Master thesis submitted in Partial Fulfillment of the Requirements

for the MSc Degree of

Hydro Science and Engineering

Supervised by:

Dipl. Hydrol. Mr. Alexander Gerner

Dr. Franz Lennartz

Professor in Charge: Prof. Dr.-Ing. habil. Gerd H. Schmitz

Institute of Hydrology and Meteorology

Faculty of Forest, Geo and Hydro sciences

Dresden University of Technology

Dresden, Germany

September 17, 2010

Analysis of Flash Flood Routing by Means of 1D - Hydraulic Modeling 2010

Signature Page

The Thesis committee for TECHNICHE UNIVERSITÄT DRESDEN Certifies that this is

the approved version of the following Master thesis:

'Analysis of Flash Flood Routing by Means of 1D – Hydraulic Modeling'

APPROVED BY SUPERVISING COMMITTEE:

Prof. Dr.-Ing. habil. Gerd H. Schmitz

_________________________________________

Supervisors:

Dipl. Hydrol. Mr. Alexander Gerner

Dr. Franz Lennartz

________________________________________


Declaration

Although, I do believe that nothing is new on earth under the sun may be not heard to

many people; I hereby declare that this thesis report titled 'Analysis of Flash Flood

Routing by Means of 1D – Hydraulic Modeling' is my own work and that where any

material used as the work of others, it is fully cited and/or referenced.

Eidesstattliche Erklärung

Hiermit versichere ich, daß ich die vorliegende Diplomarbeit selbständig angefertigt,

anderweitig nicht für Prüfungszwecke vorgelegt, alle zitierten Quellen und

benutzten Hilfsmittel angegeben sowie wörtliche und sinngemäße Zitate

gekennzeichnet habe.

Signed: _____________ Date: September 18, 2010

Zerisenay Tesfay Abraha

Dresden, Germany.

i


Acknowledgments

First and for most, acknowledgement to the Almighty God because on His mercy, finally

I am here, writing the acknowledgement for this important document of my academic life

career with an unprecedented dedication and determination successfully on time.

I would like to acknowledge my first advisor Mr. Alexander Gerner, who owns a friendly,

patient, quick understanding character that makes him easy to approach. I thank him for

his consistent assistance, advice, critics provided to me throughout the study period; as

well as to my second supervisor, Dr. Franz Lennartz, for his willingness and the overall

guidance on which this thesis work was enabled and sustained by their vision and

ideas. Both of them shared me new and challenging ideas that made the study work

more interesting as well as they have created for me a very conducive and convenient

working atmosphere within the chair of hydrology office which was a very significant

asset during my study work. Furthermore, I am very grateful to all staff members of the

chair especially to Prof. Dr.-Ing. habil. Gerd H. Schmitz, Head of the chair of Hydrology,

for being willing full to undertake the thesis work under his chair as well as for his

approval.

Very special thanks to Deutscher Akadamischer Austausch Dienst / German Academic

Exchange Service (DAAD) for providing me a full scholarship during my study period

which indeed played a key role towards my successful accomplishment of my study.

Furthermore, many thanks to Mr. Klaus Stark, DAAD’s staff in charge, for his on time

and valuable support he offered me whenever I needed him.

I am very also grateful to Danish Hydraulic Institute (DHI), and/or DHI WASY GmbH for

providing me the MIKE software package (student version) for free so as to conduct my

thesis work; and also to Engineer Lucie Legay and her colleagues for being willing-full

to share their support and constructive advices during the software installation.

ii


Last but not least, no one can ever truly value the day to day support of one’s own

family. My family swallowed hard when I chose this career; and with them behind me, I

knew that I could not fail. Hence I am as ever, especially indebted to the whole family

member of mine for their love and support throughout my life. I am also most grateful to

have a very understanding family, friends, and colleagues throughout my life career.

iii


I would like to dedicate this work to my beloved family who swallowed hard when I

chose this study career.

iv


Abstract

This study was conducted at the mountainous catchment part of Batinah Region of the

Sultanate of Oman called Al-Awabi watershed which is about 260km2 in area and with

about 40 Km long Wadi main channel. The study paper presents a proposed modeling

approach and possible scenario analysis which uses 1D - hydraulic modeling for flood

routing analysis; and the main tasks of this study work are (1) Model setup for Al-Awabi

watershed area, (2) Sensitivity Analysis, and (3) Scenario Analysis on impacts of rainfall

characteristics and transmission losses.

The model was set for the lower 24 Km long of Al-Awabi main channel (Figure 13).

Channel cross-sections were the main input to the 1D-Hydraulic Model used for the

analysis of flash flood routing of the Al-Awabi watershed. As field measurements of the

Wadi channel cross-sections are labor intensive and expensive activities, availability of

measured channel cross-sections is barely found in this study area region of Batinah,

Oman; thereby making it difficult to simulate the flood water level and discharge using

MIKE 11 HD. Hence, a methodology for extracting the channel cross-sections from

ASTER DEM (27mX27m) and Google Earth map were used in this study area.

The performance of the model setup was assessed so as to simulate the flash flood

routing analysis at different cross-sections of the modeled reach. And from this study,

although there were major gap and problems in data as well as in the prevailing

topography, slope and other HD parameters, it was concluded that the 1D-Hydraulic

Modelling utilized for flood routing analysis work can be applied for the Al-Awabi

watershed. And from the simulated model results, it was observed that the model was

sensitive to the type BC chosen and taken, channel cross sections and its roughness

coefficient utilized throughout the model reach.

Key words: Flash Flood Routing, 1D-Hydraulic Model (MIKE 11 HD), Sensitivity

Analysis, Scenario Analysis, Al-Awabi watershed, Oman.

v


Table of Contents

Acknowledgments .............................................................................................................................. i

Abstract ............................................................................................................................................. iv

List of Figures .................................................................................................................................. vii

List of Tables .................................................................................................................................... ix

Acronyms ........................................................................................................................................... x

1. Introduction ................................................................................ 1

1.1 Problem Statement ..................................................................................................................... 1

1.2 Research Objectives .................................................................................................................. 2

1.3 Research Potentiality ................................................................................................................. 2

1.4 Overview of Relevant Literature Review .................................................................................. 3

1.4.1 Overview of GIS................................................................................................................... 3

1.4.2 Digital Elevation Models...................................................................................................... 3

1.4.3 Applications of GIS in Hydrology........................................................................................ 4

1.4.4 Flood Routing....................................................................................................................... 4

1.4.5 MIKE 11 Hydrodynamic Models ......................................................................................... 5

2. Study Area and data ................................................................. 6

2.1 Study area ................................................................................................................................... 6

2.1.1 Target Watershed Area ....................................................................................................... 7

2.2 Data ............................................................................................................................................. 8

3. Research Methodology .......................................................... 13

3.1 Overall Scheme ........................................................................................................................ 13

3.2 Task-1: Model setup for Al-Awabi watershed area ................................................................ 13

3.3 Task-2: Sensitivity Analysis ..................................................................................................... 14

3.4 Task-3: Scenario analysis ........................................................................................................ 14

3.5 Task-4: Discussion of Results ................................................................................................. 14

4. ASTER DEM and Hydro Dynamic Model Set up ............... 15

4.1 ASTER DEM and Digitization of MIKE GIS ............................................................................ 15

4.2 Channel Geometry Realization from Google Earth Map....................................................... 16

4.3 Model Setup of MIKE 11 HD ................................................................................................... 18

5. Results and Discussions ....................................................... 22

vi


5.1 Flood Routing using 1D-Hydraulic Model ............................................................................... 22

5.1.1 Model Cross-section - Hydraulic Parameters ................................................................. 22

5.1.2 Calibration of MIKE 11 HD Model .................................................................................... 23

5.1.3 Simulation of Water Surface Profile ................................................................................. 24

5.1.4 Model Sensitivity ................................................................................................................ 26

5.1.5 Visualization of Simulated Model Results using MIKE View ......................................... 28

5.1.6 Comparison of Modeled Results using MIKE GIS .......................................................... 31

5.2 Sensitivity Analysis ................................................................................................................... 32

5.2.1 Impacts of uncertainties in channel geometry................................................................. 32

5.2.2 Impacts of uncertainties in channel roughness ............................................................... 34

5.2.3 Impacts of numerical flow descriptions ............................................................................ 37

5.3 Scenario Analysis ..................................................................................................................... 39

5.3.1 Scenario analysis considering partial area coverage of rainfall .................................... 39

5.3.2 Scenario analysis considering transmission losses ....................................................... 42

6. Conclusion ................................................................................ 45

7. Limitation of study and Recommendation ........................ 47

Theses ................................................................................................ 49

Bibliography ..................................................................................... 51

Appendix .......................................................................................................................................... 52

Appendix 1 ................................................................................................................................... 52

Annex ............................................................................................................................................... 55

Annex 1 ........................................................................................................................................ 55

Annex 2 ........................................................................................................................................ 57

Annex 3 ........................................................................................................................................ 59

vii


List of Figures

Figure 2.1: Al-Awabi watershed area location with respect to Oman (Google Earth map)

........................................................................................................................................ 6

Figure 2.2: Al-Awabi watershed area .............................................................................. 7

Figure 2.3: Al-Awabi Watershed Hill-shade (left) and Slope (right) ................................. 8

Figure 2.4: Al-Awabi stream Network and Sub-watersheds .......................................... 10

Figure 2.5: Plan view of channel cross-section and profiles.......................................... 10

Figure 2.6: Al-Awabi Runoff Gauge (left) and typical geological structure of study area

(right) (TU Dresden – Chair of Hydrology) .................................................................... 12

Figure 2.7: Digitized Al-Awabi channel network and its sub-watersheds ...................... 12

Figure 3.1: Schematization of methodology to assess the flash flood routing analysis . 14

Figure 4.1: Model network of main channel and few tributaries, and their respective

longitudinal views. ......................................................................................................... 16

Figure 4.2: MIKE GIS digitized network and its typical Google Earth cross-sections .... 17

Figure 4.3: Comparison of typical cross-section profiles from Google Earth and DEM . 18

Figure 4.4: MIKE 11 HD Model input and output........................................................... 18

Figure 4.5: Modeled channel network ........................................................................... 19

Figure 4.6: Model input UBC – Inflow in m^3/s .............................................................. 20

Figure 5.1: Comparison of modeled result of discharge at outlet of main channel with

DELTA=0.70 and 0.85 .................................................................................................. 24

Figure 5.2: Simulated water surface profile of Al-Awabi Wadi....................................... 24

Figure 5.3: Typical settlements along Al-Awabi Wadi channel ...................................... 26

Figure 5.4: Variation of simulated water depth with different Manning’s-M roughness

coefficient ...................................................................................................................... 27

Figure 5.5: Typical discharge and Q - H relationship of the Modeled results ................ 30

Figure 5.6: Comparisons of delta discharge in m^3/s at outlet chainage ....................... 31

Figure 5.7: Simulated water depth using cross-section from DEM & Google Earth map.

...................................................................................................................................... 33

Figure 5.8: Comparison of model results of discharge at Al-Awabi watershed outlet of

main channel with different Manning - M values ........................................................... 34

Figure 5.9: Simulated discharge hydrograph at outlet chainage ................................... 35

viii


Figure 5.10: Comparison of simulated discharge at outlet of main channel with different

Manning – M values ...................................................................................................... 36

Figure 5.11: Simulated outlet discharge using Fully Dynamic wave approximation ...... 38

Figure 5.12: Simulated outlet discharge using Diffusive wave approximation ............... 38

Figure 5.13: Comparison of simulated discharge at outlet using fully dynamic versus

Diffusive wave approximation ....................................................................................... 39

Figure 5.14: Relative locations of lateral inflows at three tributaries to main channel ... 40

Figure 5.15: Comparison of simulated discharge at outlet using lateral inflows along

longitudinal main channel.............................................................................................. 41

Figure 5.16: Simulated outlet discharge with GWL of 0.00001 ..................................... 43

Figure 5.17: Comparison of simulated outlet discharge using 0.00001 GWL along

longitudinal main channel.............................................................................................. 43

Figure 5.18: Simulated water depth time series at chainage 31365 .............................. 44

ix


List of Tables

Table 2.1: Location of Study Area-Runoff Gauge ........................................................... 8

Table 2.2: Annual runoff at gauge Awabi from 1984 - 2007 ........................................... 9

Table 5.1: Simulated water-surface elevations at cross sections for the Al-Awabi main

channel ......................................................................................................................... 25

Table 5.2: Variation of simulated discharge in m^3/s on March 01, 2005 with different

Manning’s-M roughness coefficient ............................................................................... 28

Table 5.3: Comparison of partial rainfall coverage on simulated water-depth ............... 41

x


Acronyms

a.m.s.l Above mean sea level

Arc GIS Software for Geographic Information Systems

ASTER - DEM Advanced Spaceborne Thermal Emission and

Reflection Radiometer - DEM

BC Boundary Condition

DBC Downstream Boundary Condition

DEM Digital Elevation Model

DHI Danish Hydraulic Institute

DHI WASY GmbH DHI WASY GmbH represents the group in Germany,

Austria and German-speaking Switzerland and is a

center of excellence in ground water for the entire DHI

group.

DS Down Stream

ESRI Environmental Systems Research Institute

FRM Flood Risk Management

GIS Geographic Information System

GDEM Global Digital Elevation Model

Google Earth It is a virtual globe, map, and geographic information

computer model by Google company.

GWL Ground Water Leakage; and it defines the leakage

coefficient to the loss of water from the Wadi channel

flow.

HD Hydrodynamic

i.e. That is

METI Ministry of Economy, Trade, and Industry

MIKE 11 Professional software package for 1D simulation of

flows in rivers and channels.

MIKE GIS / MIKE 11 GIS MIKE software with an extension to Arc GIS

MIKE NAM or MIKE RR Hydrological model to simulate runoff from the rainfall &

xi


NAM stands for Nedbor Afstromnings Model in Danish

language; whereas RR stands for rainfall-Runoff.

MIKE View Viewing software used to view MIKE 11 model results

MIKE Zero MIKE Zero is a software informer developed by DHI

which gives access to DHI’s modeling system.

NASA United States National Aeronautics and Space

Administration

RASTER DEM Data structure where the geographic area is divided into

cells i.e. row and column.

UBC Upstream Boundary Condition

US Up Stream

1D-Hydraulic Model One Dimensional Hydraulic Model

3D Three Dimensional

1


1. Introduction

This chapter introduces a distributed meso-scale catchment modelling of flood routing

and addresses the research challenges, objectives and potentiality of the distributed

catchment model of Al-Awabi watershed area located at the mountainous area of

Batinah Region of the Sultanate of Oman. It also provides a brief relevant literature

background typically utilized for executing and analyzing the overall tasks within the

scope and motivation of this study work.

1.1 Problem Statement

Arid areas such as Oman are typically characterized with both extremes of hydrological

events that are drought and flood. Due to this extreme nature of rainfall – runoff events,

and an inadequate or no existing coping mechanism measures and structures

implemented; a natural disaster such as flash flood has been a common phenomenon

in the Batinah region of the sultanate of Oman. Thus, in this study area, flash flooding is

the major problem.

Channel cross-sections are one of the main inputs to the 1D-Hydraulic Model used for

the analysis of flash flood routing. Whereas, field measurements of the Wadi channel

cross-sections are labor intensive and expensive activities. Thus, availability of

measured channel cross-sections is barely found in this study area; thereby making it

difficult to simulate the flood water level and discharge using MIKE 11 HD. Moreover,

uncertainty in channel geometry, roughness, and a flow description are highly

anticipated to impact on the sensitivity analysis of the model.

For these reasons, flash flood routing technique should be developed and analyzed

continuously so as to prevent any of such effects. However, since there are a lot of

uncertainties on the estimated data, roughness and geometry of channel, transmission

losses and vegetation covers, analysis is not an easy process. Data necessary for

carrying out routing analysis is very limited in the study area. Thus, to bridge these data

gaps, a methodology of gathering information from the ASTER DEM (27mX27m),

Google Earth maps, and Russian topographic maps is to be used in order to assess the

2


topographic nature of the watershed surface and the network of the Wadi channel

networks.

Therefore, the above reputed problems will be overcome from this research work when

carefully studied over the key and controlling model parameters.

1.2 Research Objectives

In view of the above stated shortcomings, derivation and processing of these flood

routing parameters by means of 1D-Hydraulic Modeling i.e. MIKE 11 HD is expected to

contribute to an overall comprehension of the sensitivity and scenario analysis of the

flash flood routing in Al-Awabi watershed area, and reliable estimation of hydrological

and hydraulic characteristics. With the aid of GIS-supported assessment, analysis

capability will be enhanced with regard to sub-watersheds topology, querying, display

and mapping of results, in other words making the analysis process at the finger tips of

the end user.

The ultimate goal of this study work will serve as a catchment modeling tool that

investigate both the uncertainties in flash flood routing analysis based on limited data

and distinct characteristics of rainfall-runoff-processes in the study area; for instance,

partial area coverage of rainfall and transmission losses. And this goal will be achieved

by performing and executing the following tasks accordingly i.e. the 1D-Hydraulic model

setup for the target watershed of Al-Awabi i.e. MIKE 11 HD; sensitivity analysis;

scenario analysis on impacts of rainfall characteristics; and finally working on the overall

discussion of results.

1.3 Research Potentiality

This study work is an essential part of the distributed meso-scale catchment modelling,

and the setup of a 1D – Hydraulic Model based on available data shall support the

development of the distributed catchment model of Al-Awabi watershed. It will further

have its contribution to the ultimate goal of the water balance assessment work in that

region via replication of the process both temporally and spatially. Furthermore, as

3


partnership using the watershed approach, this study work will serve as a corner

foundation stone towards the complete catchment-scale model which consists of hill-

slope surface and channel flow sub-models coupled together in one comprehensive

model to account for catchment rainfall-runoff production and flood routing applicable to

flood transmission losses in the mountainous area of the Batinah region, Oman.

1.4 Overview of Relevant Literature Review

1.4.1 Overview of GIS

Networks can be easily analyzed and assessed if there is an interaction between

databases and maps; and this is why the Geographic Information System (GIS) comes

to an application. GIS is an information system that answers questions from a database

of spatially distributed features and procedures to collect, store, retrieve, analyze and

display geographic data (Shamsi, 2005). GIS helps professionals in mapping,

monitoring, modeling and maintenance of water related systems. Thus, by doing so, a

considerable time and money is saved. GIS users benefit from the GIS’s easier and

quicker results in analyzing problems and recommending solutions within a fraction of

the time otherwise that would be required with a tedious manual working means.

1.4.2 Digital Elevation Models

Digital Elevation Models (DEM) also called digital terrain models provide a 3D

representation of the real-world topography. DEM creation requires data collection and

processing procedures. Data collection step depends on the areal extent and

importance of the study. They can be constructed by ordinary ground survey when the

study area is relatively small or of minor importance. On the other hand if the study area

is large, satellites can be used for mapping the topography. These maps then have to

be processed by remote sensing imagery to give topographic elevation information. This

is how the DEMs covering the entire globe, for example, the Advanced Spaceborne

Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model

(GDEM) which was developed jointly by the Ministry of Economy, Trade, and Industry

(METI) of Japan and the United States National Aeronautics and Space Administration

(NASA) are formed.

4


DEMs have vital role in hydrological analysis especially in delineating watersheds,

obtaining stream network, and related analysis. The accuracy of the stream network

obtained from DEMs is highly dependent on the resolution of the DEM. For instance, the

ASTER GDEM tiles (1◦×1◦tiles) of resolution 27m×27m were downloaded from the

website of ASTER GDEM (GDEM, 2009), and the DEM tiles were made mosaic to get

the complete DEM of the Al-Awabi watershed area.

1.4.3 Applications of GIS in Hydrology

Hydrological applications of GIS are extremely varied. Whilst hydrological scientists

have progressed in their representations of hydrological processes from lumped through

semi-distributed to distributed hydrological models, water resource managers have

followed a parallel route in the increasing spatial resolution with which assets,

particularly infrastructure, have been represented, interrelated and managed

(Garbrecht, 2000). With the increasing availability of high-resolution DEMs such as

ASTER DEM (27m×27m), the most widespread application of GIS in hydrology is the

identification of drainage pathways and runoff contributing areas based on topographic

form, and their coupling with hydrological and hydraulic models, for example MIKE NAM

and MIKE 11 HD.

Although catchment-scale hydrological modeling represents an important GIS

application within hydrology, GIS has relevance to the solution of many other

hydrological problems at local, catchment and regional scales. And the same is also

true with this study work’s GIS utilization.

1.4.4 Flood Routing

Also called streamflow routing and channel routing, is one of the classical problems in

applied hydrology. The word routing refers in general to the mathematical procedure of

tracking or following water movement from one place to another; as such, the word also

includes the description of the conversion of precipitation into various subsurface and

surface runoff phenomena. However, flood routing refers specifically to the description

5


of the behavior of a flood wave as it moves along in a well defined open channel

(Brutsaert, 2005). The wave, to be dealt in this study area, is typically the result of

inflows into the Wadi channel following heavy rainfall. For the routing of a flood wave,

numerical methods that solve the complete continuity and momentum equations may be

used.

Fully Dynamic: It is an HD Module which provides fully dynamic solution to the

complete nonlinear Saint Venant equations for an open channel flow.

Kinematic Models: These models are based on the solution of the continuity

equation and the steady-uniform equation for the dynamic equation. The waves

propagated using these models are called kinematic waves, and routing is called

kinematic routing.

Diffusion Routing: This is formulated based on the simplified versions of

momentum equation.

Muskingum-Cunge Method: This method is actually a particular finite-difference

approximation of the kinematic wave equations and present expressions for the

Muskingum coefficients in terms of the physical properties of the channel. And

the coefficients for the method were determined from the observed flood records.

1.4.5 MIKE 11 Hydrodynamic Models

MIKE 11HD model is a one-dimensional hydraulic modeling software package,

developed at Danish Hydraulic Institute in 1987. The model has been widely used to

simulate water levels and flow in the river systems. It has an interface to GIS allowing

for preparation of model input and presentation of model output in a GIS environment, in

which this study work was conducted using this package called MIKE11 GIS. It merges

the technologies of hydraulic modeling, MIKE 11, and Arc GIS developed by ESRI. This

is designed to run into Arc GIS environment which can automatically integrate water

level resulting from MIKE11-HD into digital elevation map (DEM) for which we can

determine the inundated area at the rainfall-runoff events within the study area.

6


2. Study Area and data

2.1 Study area

This study was conducted at the mountainous catchment part of Batinah Region of the

Sultanate of Oman called Al-Awabi watershed which is about 260km2 in area (Figure

2.1). Oman is known as one of the world’s arid areas. It is located in the southeast of

the Arabian Peninsula. It is bordered by Saudi Arabia to the west, United Arab Emirates

to the northwest, Yemen to the south east and Arabian Sea to the east.

Figure 2.1: Al-Awabi watershed area location with respect to Oman (Google Earth map)

Among the flood events which took place in Oman, almost all are categorized as flash

floods. Oman is characterized by arid and/or semi arid climatic conditions, with many

periods of drought followed by few periods of convective or advective rainstorms

typically intense and erratic rainfalls produced in short time. Flash flooding can cause

severe damage to buildings and infrastructures and pose a high risk to life and

properties. And it is naturally very difficult to model and forecast bearing in mind the lack

7


of enough data and sustainable working system in the target area, Al-Awabi watershed,

Batinah region, Oman.

2.1.1 Target Watershed Area

The study area i.e. Al-Awabi watershed is located at the northern escarpment part of

Oman, south of Batinah region and is one of the main tributaries to Wadi Bani-Kharus. It

is situated at the south of the whole Bani-Kharus catchment area. Furthermore, the

DEM and channel network within the Al-Awabi watershed as depicted in Figure 2.2 was

generated from ASTER DEM using Arc GIS; and the given watershed was described as

third order stream that is a tributary formed by two or more second order streams as

well as streams of lower order.

Figure 2.2: Al-Awabi watershed area

The watershed lies between latitude 2555187 to 2576337 and longitude 545997 to

573706. The total area is 254 km2 with a major portion of it is mountainous area and

lays in the south of Batinah Region with a surface area and volume of about 300km2

and 235km3 respectively as shown in Figure 2.3. Within the watershed, there is one

8


runoff gauge which is located at latitude of 23o17’47” and longitude of 57o31’43” (Table

2.1).

Table 2.1: Location of Study Area-Runoff Gauge

Station Name Latitude Longitude Watershed Tributary Area Region Elevation

Awabi near Awabi 23°17’47” 57°31’43” Al-Awabi Bani Kharus 254Km^2 South Batinah 500m

The average surface elevation of the watershed area ranges from 496 m to 2483 m

a.m.s.l. with a mean slope of about 25o. The area being almost ragged mountainous,

there was severe flash flood occurrence at downstream of the watershed and its vicinity

during heavy rainfall events. The hill-shade and slope of the Al-Awabi watershed were

generated and shown in Figure 2.3 for better visualization of the ground surface.

Figure 2.3: Al-Awabi Watershed Hill-shade (left) and Slope (right)

2.2 Data

Daily runoff data at Awabi gauging site for a period of 23 years (1984–2007) was

recorded and analyzed as shown in Table 2.2. For the hydraulic model set-up, water

level and channel cross-section data are required; hence, based on the available runoff

data an estimated inflow hydrograph was tried to be generated using MIKE NAM in

9


order to be used for model calibration and validation respectively. But, due to lack of

data and other required model parameters an intuitive estimate of inflow hydrographs

were utilized using the available run-off data as a basis as shown on Appendix 1.

Hence, this study deals with Al-Awabi Wadi channel rainfall - runoff estimation using an

estimated synthetic hydrographs till a satisfactory estimation was obtained.

Table 2.2: Annual runoff at gauge Awabi from 1984 - 2007

Year Annual Runoff Measured at Gauge Awabi in Cubic Meter per month Total

Jan Feb March April May June July August Sep Oct Nov Dec (m^3/yr)

1984 0 0 0 0 0 0 0 0 0 0 0 0 0

1985 0 0 0 0 0 0 4722 0 0 0 0 0 4722

1986 0 0 11840 8385 0 135740 46760 35850 0 0 0 0 238575

1987 0 99020 737610 263000 0 0 0 98860 0 0 0 0 1198490

1988 0 13925 0 0 0 0 953435 31900 0 0 0 20290 1019550

1989 0 0 4014 4903 0 0 83220 0 142070 0 50930 121360 406497

1990 0 47517 0 4770 76650 326000 205480 81380 12140 7266 0 0 761203

1991 0 0 14640 34410 0 0 0 0 100110 0 18130 26470 193760

1992 0 0 0 9823400 924500 0 80400 2224300 38500 0 0 0 13091100

1993 0 0 40003 0 0 0 152064 8640 0 119664 0 39312 359683

1994 0 0 0 0 0 0 693014 524448 0 270518 0 0 1487980

1995 0 0 0 0 124762 0 432086 918950 97632 0 0 0 1573430

1996 0 0 0 0 0 400032 84326 105408 0 0 0 0 589766

1997 49334 0 1390176 575424 0 0 38707 102038 1642 0 11578 0 2168899

1998 0 60307 0 27476 0 25402 0 0 431136 87610 0 0 631931

1999 0 132365 361238 0 0 0 241315 152755 320112 0 0 0 1207785

2000 0 0 0 0 0 19094 0 88042 147398 69725 0 0 324259

2001 46915 0 196992 0 0 324950 1122336 1642 0 215741 0 36893 1945469

2002 0 0 0 151200 120614 0 0 0 0 0 0 0 271814

2003 0 0 0 586051 0 0 0 0 0 0 0 0 586051

2004 0 0 0 69206 0 0 214358 0 41990 0 0 144720 470274

2005 0 2419200 3257366 0 0 73958 0 0 0 0 0 0 5750524

2006 0 482890 0 165802 0 0 0 0 27389 50371 0 30758 757210

2007 0 0 3191962 0 0 3682196 0 0 0 0 0 0 6874158

10


Within the whole watershed area,

four sub-watersheds were created,

and an adequate number of cross-

sections at different locations of the

main tributary channel of each sub-

watershed were extracted from the

DEM. Flow accumulation was used

to generate a drainage network; for

instance, in this study work cells

with accumulated flow greater than

5000 cells was used (Figure 2.4) as

the value of 5000 looks reasonable.

Figure 2.4: Al-Awabi stream Network and Sub-watersheds

Figure 2.5: Plan view of channel cross-section and profiles

11


Six cross-sections at different channel locations of the selected sub-watersheds were

identified and extracted from the DEM of Al-Awabi watershed as shown in Figure 2.5

using Arc GIS for demonstration purpose otherwise numerous number of cross-sections

had been extracted using MIKE GIS and were modified during model setup as

described in section 4.

In this study, analysis of flood routing was conducted using the 1D-Hydraulic Model i.e.

MIKE 11 HD and further geometrical and hydraulically uncertainty analysis and possible

scenario analysis were assessed particularly in regard to transmission losses due to the

prevailing soil and geological conditions as shown in Figure 2.6; which is generally bare

and porous rocky ragged mountainous area. Most of the wade channel networks are

also used as access earth roads for the inhabitants within Al-Awabi watershed and its

vicinity as depicted in figure 2.6 (left). Thus, the execution of this study work would be

helpful in describing the consequence i.e. disruptions and damages that might occur as

a result of flash flood to the ongoing activities of the receptors in that area.

The total length of the Al-Awabi Wadi main channel is about 40km. It has about 40

branches or tributaries with 70 points, 29 cross sections along the mainstream. Each

channel branch was executed by digitizing the Al-Awabi DEM, flow direction, and flow

accumulation along the path of the Wades using MIKE GIS. Furthermore, the digitized

reaches were smoothened using the Arc-map smooth tool that is

BEZIER_INTERPOLATION smoothing algorithm so as to avoid sharp bends (see

Figure 2.7).

It was observed that, the drainage pattern of Al-Awabi watershed can be categorized as

Dendritic pattern where tributary branch and erode headwater in random fashion which

results in slopes with no predominant direction or orientation; as well as Rectangular

pattern mainly occurring in this case along and near to the main channel network as

shown in Figure 2.7. Furthermore, in order to have detailed information of the whole

watershed topography, three hundred fifteen (315) sub-watersheds had been

12


delineated; although finally only the main channel was modelled as discussed in section

four and five.

Figure 2.6: Al-Awabi Runoff Gauge (left) and typical geological structure of study area

(right) (TU Dresden – Chair of Hydrology)

Figure 2.7: Digitized Al-Awabi channel network and its sub-watersheds

13


3. Research Methodology

3.1 Overall Scheme

A good starting point for a quantitative assessment of runoff is to consider the physical

processes occurring in the hydrological cycle of the study area. And from this, a set of

influencing factors could be proposed which determine the response of the watershed to

rainfall; namely, watershed area, soil type and depth, channel and surface slopes, rock

type and area, vegetation cover, reservoirs, sealed areas if available and so on of the

study area. Therefore, in this study work, we are dealing with Wadi channel routing

problem i.e. to find the outflow hydrograph from the Wadi river reach from the inflow

hydrograph. And MIKE 11 HD was used for this study work.

Furthermore, as in many practical situations, no historical time series of inflow-outflow

data were available for the Al-Awabi watershed area. Hence, the model must be

synthesized from physical information on the system available from the topographical

map, Google Earth map, and/or from RASTER DEMs. Therefore, it would be highly

desirable to find a linkage between physically sound hydrodynamic models and

hydrological conceptual models.

3.2 Task-1: Model setup for Al-Awabi watershed area

This first task comprises the relevant data assimilation and estimation, and setup of 1D-

Hydroulic modeling i.e. MIKE 11 HD, the model applied to assess the flash flood routing

analysis within the Al-Awabi watershed area. Hence, this task comprises the following

sub-tasks:

Derivation of sub watersheds and/or channel network using Arc GIS or MIKE GIS

for the Al-Awabi watershed based on ASTER-DEM (27X27m).

Digitization of cross-section at decisive stations of the longitudinal sections based

on ASTER-DEM (27x27m), and Google Earth map.

Setup of the 1D-Hydraulic model that is MIKE 11HD.

14


3.3 Task-2: Sensitivity Analysis

It was the task of this paper to make a sensitivity analysis of as to how and to what

degree the variation of model input affects the output uncertainty particularly in regard to

the following factors:

Preparation of upper boundaries (inflow) –based on reasonable assumptions.

Impacts of uncertainties in channel geometry.

Impacts of uncertainties in channel roughness.

Impacts of numerical flow descriptions.

3.4 Task-3: Scenario analysis

Scenario analysis is defined as a process of analyzing possible future events by

considering alternative possible outcomes. Thus, this study work focuses on two main

Scenarios analyzed on the target watershed.

Scenario analysis considering the partial area coverage of rainfall.

Scenario analysis considering transmission losses.

3.5 Task-4: Discussion of Results

Last but not least, the overall discussion of all the findings of the above reputed tasks

were reported in detail leading to the study’s conclusion and recommendation; see

Figure 3.1, for the schematic arrangement of the methodologies developed to

accomplish all those tasks. It comprises and discusses the input data utilized, model

set-up and run, and its output of the 1D-Hydraulic Model, MIKE 11 HD.

Figure 3.1: Schematization of methodology to assess the flash flood routing analysis

Data

Base,

DEM

&

Google

Earth

Flood Routing

Analysis

Map

Graph

Table

Report

Sensitivity

Analysis

Routing

Analysis Uncertainty

Analysis

Scenario

Analysis

MIKE GIS

MIKE11-HD

MIKE GIS /

MIKE View

15


4. ASTER DEM and Hydro Dynamic Model Set up

4.1 ASTER DEM and Digitization of MIKE GIS

As mentioned in section one, an ASTER DEM for Al-Awabi watershed was extracted

using Arc GIS 9.3 and digitization of the respected channel network and cross-section

were executed using the MIKE 11 GIS as shown in Figure 4.1. Furthermore, data error

analysis and modification of extracted channel cross-sections had been dealt by doing a

comparison work for the processed DEM that was digitized in MIKE GIS with that of

cross-sections derived from Google Earth.

Putting into consideration the MIKE 11 simulation requirement, and the limited data

availability, this study work focuses at analyzing for a well selected and quite

representative four sub-watersheds which are distributed spatially within Al-Awabi

watershed so as to meet the study’s overall task. Furthermore, although the whole

watershed is a mountainous ragged area, relatively speaking, the whole main channel

bed slope can be categorized into three divisions such as: chainage 0 to 10,000: flat

bed slope; chainage 10,000 to 25,000: gentle bed slope; and chainage 25,000 to

37,690: steep bed slope. (See the longitudinal profile section of the Al-Awabi main

channel in Figure 4.1).

16


Figure 4.1: Model network of main channel and few tributaries, and their respective

longitudinal views.

4.2 Channel Geometry Realization from Google Earth Map

Further realization of detailed channel geometry of the study area under investigation

was analyzed based on the free version Google earth map downloaded from

500

700

900

1100

1300

1500

1700

1900

2100

2300

0 5000 10000 15000 20000 25000 30000 35000

Al-

Aw

abi

bed

ele

vat

ion

in

met

er

Longitudinal distance in meter

Longitudinal Al-Awabi Channel

550

650

750

850

950

1050

0 2000 4000 6000 8000 10000 12000

Ele

vat

ion

in

met

er

Horizontal distance in meter

Longitudinal section of Tributary 1

600

800

1000

1200

1400

1600

0 2500 5000

Ele

vat

ion

in

met

er

Horizontal distance in meterLongitudinal section of Tributary 2

750

1000

1250

1500

1750

2000

0 1000 2000 3000 4000

Ele

vat

ion

in

met

er

Horizontal distance in meterLongitudinal section of Tributary 3

17


http://earth.google.com/index.html. Hence, through thorough observation on the

digitized Al-Awabi watershed in Google Earth, the channel network cross-sections were

taken and used as one input parameter in setting up the model, MIKE 11 HD. Although

there was minor contradiction in the channel network extracted using Arc GIS with that

of Google Earth map, channel geometry measurement estimations were taken from that

of Google Earth map as shown in Figure 4.2 and Annex 2.

Figure 4.2: MIKE GIS digitized network and its typical Google Earth cross-sections

As depicted in Figure 4.2, the delineated watershed border line, channel network,

reaches, cross-sections, and other necessary geographic information of the whole Al-

Awabi watershed area processed by Arc GIS 9.3 and/or MIKE GIS were linked to that of

Google Earth map so as to come up with a fine and reasonable estimation of the

channel geometry dimensions which could be used as an input for the MIKE 11 HD

model set-up.

965

966

967

968

969

0 10 20 30

Ele

vat

ion

(m

)

Width (m)

Chainage 8365

782

783

784

785

786

0 10 20 30 40 50

Ele

vat

ion

(m

)

Width (m)

Chainage 15459 (JT3)

640

645

650

0 50 100

Ele

vat

ion

(m

)

Width (m)

Chainage 25785 (TJ2)

583

588

593

598

603

0 100 200

Ele

vat

ion

(m

)

Width (m)


506

511

516

0 20 40 60 80

Ele

vat

ion

(m

)

Width (m)

Chainage 36630 (DBC)

http://earth.google.com/index.html

18


Furthermore, it was noticed that there were minor contradictions between the DEM

delineated channel bed elevations with that of the Google Earth’s elevations on which

an intuitive guess was applied during the model setup particularly at the channel

junctions. Resolution of DEMs plays a big role in the quality of estimation that comes

from it and definitely higher resolution DEM data will give more accurate results. Thus,

the necessity of executing a detailed survey measurement is worthwhile to mention here

as a recommended task for better accuracy as there was significant difference

especially in elevations as depicted in Figure 4.3 (see also Annex 1 2 and 3 for more

comparison).

Figure 4.3: Comparison of typical cross-section profiles from Google Earth and DEM

4.3 Model Setup of MIKE 11 HD

MIKE 11 HD model set up was prepared and run for the a specific flash flood period of

Figure 4.4: MIKE 11 HD Model input and output

760

762

764

766

768

770

772

774

776

778

780

0 25 50 75 100 125 150

Bed

Ele

vat

ion

in

met

er

Channel Width in meter

Chainage 16365 Elevation from Google Earth in Meter

Chainage 16365 Elevation from DEM in Meter

682

684

686

688

690

692

694

696

698

0 20 40 60 80

Bed

Ele

vat

ion

in

met

er

Channel Width in meter

Chainage 22365 Elevation from Google Earth in Meter

Chainage 22365 Elevation from DEM in Meter

Flood Routing Analysis

MIKE 11 HD

Output MIKE 11 model results

Flood maps

Flood prone area & X-Sections Graphs

Input DEM

MIKE GIS (network &

cross-sections)

Synthetic Hydrographs Google Earth map

Russian map

19


March in 2005 to simulate water level and discharge at different cross-sections of the

channel reaches within the Al-Awabi watershed area. The input and output of the MIKE

11 HD conducted and executed in this study work is depicted in Figure 4.4.

A step by step MIKE 11 model generation was executed from the MIKE Zero base

screen for the whole Al-Awabi watershed channel network as reputed below:

Channel Network: The total length of the modeled Al-Awabi Wadi main channel is

about 24km; and it has three main tributaries, 23 cross sections along the mainstream

which are automatically generated from the DEM using MIKE GIS as shown in Figure

4.5.

Figure 4.5: Modeled channel network

Channel Cross-sections: The executed cross-sections were all oriented in

perpendicular direction to the respected Wadi/reach orientation. They were represented

20


in two dimensional coordinates, namely, the transverse distance from a fixed point

represented in the abscissa (X - coordinate) and the channel bed elevation represented

in the Ordinate (Z - coordinate); and were automatically generated using MIKE GIS

which in turn further compared and checked for channel geometry dimensions from that

of Google Earth map as shown in Figure 4.5.

Boundary conditions: Boundary conditions (BC) were defined both on the upstream

and downstream side of the Al-Awabi Wadi channel. The choice of the boundary

conditions depends upon the availability of the data; but, in this case, estimated values

were taken based on the available measured hydrographs at the Awabi runoff gauge as

depicted in Appendix 1. For the upstream boundary condition (UBC), estimation of

inflows i.e. discharges for each channel branch at a specific point was used (Figure

4.6). Whereas, at the outflow of the model, a rating curve can be defined; therefore, a

Q/h BC was used for the downstream boundary condition (DBC).

Figure 4.6: Model input UBC – Inflow in m^3/s

21


Hydrodynamic parameters: Global initial conditions in the Wadi flow, Bed Resistance

and wave approximation were set for the HD calculations. The fully dynamic wave

approximation method was used for the simulation of the Wade flow in order to

conserve both momentum and continuity in the calculation as it might not be possible to

use the diffusive wave approach because the whole Awabi channel bed was rather

steep in most of the branches and flow may become super-critical at several points.

And, the initial bed resistance value based on the Manning co-efficient, M (M = 1/n,

where n is Manning’s coefficient) was set to 20, 30 and 40. Whereas, initial water depth

of 0.003m was specified; and discharge was also set to 15m3/s so as to avoid the drying

out of the flow channel.

Simulation: Last but not least, model simulation was done using the unsteady

simulation mode of the HD model for the flash flood Wade flow within Al-Awabi

watershed. It was simulated for the period of time on March 1st at 00:00 to 2nd at 18:00

in 2005 for a simulation time step of 30 seconds. Furthermore, a Hotstart file type of

condition that is the initial conditions were loaded from an existing result file that was

executed using the quasi steady simulation mode.

22


5. Results and Discussions

5.1 Flood Routing using 1D-Hydraulic Model

The model set up of MIKE 11 HD for the flash flood routing of Al-Awabi Wadi was one of

the most important application of free surface water flows in ephemeral rivers. Hence,

the modules applied to analyze the flash flood routing for the selected watershed

includes the module HD in MIKE11 which was responsible to simulate the hydraulic

regimes including water level, and discharge along the channel. In this study paper,

MIKE NAM was not used due to unavailability of required data; instead, an estimated

inflow hydrograph was used that was derived based on the recorded output hydrograph.

Therefore, the model set here basically was MIKE 11 HD; where the channel network

and cross-sections were generated using the MIKE GIS and exported to MIKE 11

network and cross-section editors; while BC, HD parameters, and simulation were

performed in MIKE 11. Finally, the executed simulation results were viewed via MIKE

View and/or MIKE GIS.

5.1.1 Model Cross-section - Hydraulic Parameters

The cross-section hydraulic parameters were computed automatically at different stages

for the estimated and/or DEM extracted cross-sections along the digitized channel

network of Al-Awabi watershed. An open section type and Resistance Radius were

chosen for the channel cross-section settings. In addition, a bed slope was computed

automatically from the cross-section data executed by setting the datum function. The

bed resistance of the cross-sections in this study work was described using the three

transversal distribution options given in MIKE 11 that is Uniform, High/Low flow zones,

and Distributed where the uniform one was used during the model set up.

Furthermore, the Manning’s - n runoff coefficient of the Al-Awabi Wadi was estimated to

be within the range of 0.025 to 0.05; and the corresponding values for M are from 40 to

20. The Chezy coefficient, C, is related to Manning's n in terms of hydraulic radius, R,

as: and was determined during model calibration accordingly.

23


5.1.2 Calibration of MIKE 11 HD Model

MIKE 11 HD model set up was prepared and run for the period of time on March 1st at

00:00 to March 2nd at 18:00 in 2005 to simulate water level and discharge at different

cross-sections of the channel reach within the Al-Awabi watershed area. For the UBC,

an estimated inflow was used; whereas, a rating curve Q/h BC was used for the DBC.

Manning’s roughness coefficient was taken as a model calibration parameter and its

value at different locations along the main channel was estimated while the GWL

coefficient was set to be zero. The simulation time step was set to 30 seconds and the

interval between consecutive computational grids was kept as 450 meter.

Furthermore, due to the topography of the Al-Awabi watershed and its rainfall intensity

and coverage, a flash flood occurs which makes the resulted runoff to be governed by

dynamic waves rather than the kinematic waves. Thus, the fully dynamic option was

used in this model work out of the three flow description module options such as

dynamic wave, diffusive wave, and kinematic wave approaches.

Finally, a modification work was applied to the channel cross-section during the model

setup particularly at the junctions of the tributaries due to discrepancies on the derived

bed elevations from respective channel braches with that of the main channel. It was

also observed that a significant impact to the modeled results as well as among each of

the corresponding digitized cross-sections which in turn make the model setup task very

complex. Thus, in order to overcome this instability and complexity in the model set up,

a DELTA value was altered from its default value 0.50 to 0.70 during calibration. And it

was noticed that changing the DELTA value i.e. a coefficient in HD parameter used to

dampen potential instabilities had a very significant impact to the model set up; whereas

almost no effect to the modeled results of this study paper as shown on Figure 5.1.

24


Figure 5.1: Comparison of modeled result of discharge at outlet of main channel with

DELTA=0.70 and 0.85

5.1.3 Simulation of Water Surface Profile

The simulated water surface profile along the modeled main channel of Al-Awabi

watershed was depicted in Figure 5.2 and Table 5.1. Furthermore, the continuous water

surface profile for the study reach was determined by assuming a linear change

between computed cross-section water-surface elevations.

Figure 5.2: Simulated water surface profile of Al-Awabi Wadi

509

609

709

809

909

8365 13365 18365 23365 28365 33365

Wate

r L

evel

ele

vati

on

in

met

er

Longitudinal Chainages in meter

Water-surface elevation (meter a.m.s.l)

25


Table 5.1: Simulated water-surface elevations at cross sections for the Al-Awabi main

channel

Cross-sectional

Chainage

Water-surface elevation

(meter a.m.s.l)

Cross-sectional

Chainage


(meter a.m.s.l)

Cross-sectional

Chainage


(meter a.m.s.l)

8365 (UBC) 967

17698 755

27365 599

8698 945

18031 751

27698 598

9031 922

18365 746

28031 596

9365 901

18698 741

28365 594

9698 894

19031 736

28796 591

10031 887

19365 731

29228 588

10365 881

19698 727

29660 (JT3) 585

10698 872

20031 723

30012 583

11031 863

20365 719

30365 582

11365 855

20698 718

30698 581

11698 846

21031 717

31031 580

12031 838

21365 716

31365 579

12365 830

21698 708

31698 578

12698 824

22031 701

32031 577

13031 818

22365 695

32365 576

13365 813

22698 689

32698 567

13698 809

23031 683

33031 558

14031 806

23365 677

33365 549

14365 802

23698 671

33698 547

14698 797

24031 665

34031 546

15031 792

24365 659

34365 543

15365 786

24698 654

34698 540

15459 (JT1) 784

25031 650

35031 536

15761 780

25365 646

35365 533

16063 776

25786 (JT2) 644

35698 533

16365 772

26075 639

36031 532

16698 768

26365 634

36365 530

17031 764

26698 622

36630(DBC) 509

17365 760

27031 610

26


From the simulated results (Figure 5.2), it was observed that minimal or almost no flow

were identified at the locations where there exist inundated areas along the Wadi

channel banks; whereas, in contrast to this a higher flow depth was observed in the

places where there exist limited areas due to settlements and/or those areas without

inundation areas. Thus, flash flooding could be risky in case of higher flows; although

comparatively speaking, it might not be considered as risky by the people who are living

along the Wade’s desert route as they use the flood-recharged aquifer intensively and

the area’s entire ecology for their day to day life activities (Figure 5.3).

Figure 5.3: Typical settlements along Al-Awabi Wadi channel.

Therefore, it had been observed that the MIKE 11 system results can provide helpful

information about FRM and should be useful in assigning priority for the development of

risk area map for flood control plans and countermeasures for the settlements and

inhabitants located right along the main channel within the watershed, and for that of

Awabi town located just downstream of the Al-Awabi watershed outlet (Figure 5.3).

5.1.4 Model Sensitivity

Sensitivity of the model was carried out only for the successfully simulated models for

the main channel. The model was very sensitive to the DBC and UBC values and types

chosen. Furthermore, the sensitivity analysis was done based on the Manning –

27


Strickler Roughness coefficient on which a very slight variations was observed in the

output discharge hydrograph at the of Wadi and maximum water levels along the

channel’s downstream reach as depicted Figure 5.4 and Table 5.2.

Figure 5.4: Variation of simulated water depth with different Manning’s-M roughness

coefficient.

From Figure 5.4, it can be observed that the higher M – value (meaning lower

roughness of the channel) produces lower maximum water level at the downstream of

the Wadi channel flow; and this could be due to higher flow velocity which in turns

induces higher progression of the flood to the downstream. Whereas, as shown in Table

5.2, the higher roughness of the Wadi channel bed slightly decreases the discharge.

450,0

500,0

550,0

600,0

650,0

700,0

750,0

800,0

850,0

900,0

950,0

1000,0

8365 13365 18365 23365 28365 33365

Max. W

ate

r l

ev

el

in m

ete

r

Chainage in meter

M=20

M=25

M=30

M=35

M=40

708,0

710,0

712,0

714,0

716,0

718,0

720,0

722,0

20031 20531 21031 21531

Max. W

ate

r l

ev

el

in m

ete

r

Chainage in meter

M=20 M=25 M=30 M=35 M=40

708.9708.7708.6708.5708.4

532,0

534,0

536,0

538,0

540,0

542,0

544,0

546,0

34031 34531 35031 35531 36031

Max. W

ate

r l

ev

el

in m

ete

r

Chainage in meter

M=20 M=25 M=30 M=35 M=40

533.3533.0532.8532.6532.5

28


Although it was very insignificant, there were lower peaks with a time lag; and this was

due to the higher channel roughness on which generally slows down the progression of

the flood to the downstream.

Table 5.2: Variation of simulated discharge in m^3/s on March 01, 2005 with different

Manning’s-M roughness coefficient

Chainage M=20 M=25 M=30 M=35 M=40

Max. Q Time Max. Q Time Max. Q Time Max. Q Time Max. Q Time

8531 56.0 12:00 56.0 12:00 56.0 12:00 56.0 12:00 56.0 12:00

15412 55.8 12:37 55.8 12:33 55.8 12:29 55.8 12:26 55.8 12:24

15610 110.9 12:33 111.0 12:29 111.1 12:25 111.2 12:23 111.3 12:21

25575 110.7 13:29 110.9 13:15 111.0 13:06 111.1 12:59 111.2 12:54

25930 164.6 13:22 165.0 13:10 165.4 13:00 165.7 12:55 165.9 12:50

29444 164.5 13:42 165.0 13:28 165.4 13:16 165.6 13:08 165.8 13:03

29836 218.1 13:35 218.9 13:22 219.6 13:11 220.0 13:03 220.3 12:58

36497 217.9 14:08 218.8 13:51 219.4 13:37 219.9 13:26 220.2 13:18

5.1.5 Visualization of Simulated Model Results using MIKE View

The simulated model results were viewed finally using MIKE View and Figure 5.5

depicts typical modeled results for demonstration purpose.

From the simulated discharge hydrograph, it can be observed that the first 20minutes

were somehow wearied and this initial irregularity in a negative (decreasing) variable

trend lasts up to about three hours so as to absorb similar trend throughout the model

reach downstream up to the outlet of the watershed. And, the simulated peak discharge

at the outlet of the watershed was observed to be 55.4 m^3/s and occurs after 3 hours

and 15 minute relative to the 56 m^3/s peak at the UBC. Furthermore, it can be easily

observed that the simulated water depth time series trend was similar to that of

discharge hydrograph. Hence, discharge is proportional to water depth and this trend is

also almost linear with different slopes depending on the channel cross-section as

depicted on the Q – H relationship plot in Figure 5.5 throughout the modeled reach.

29


00:00:00

1-3-2005

03:00:00 06:00:00 09:00:00 12:00:00 15:00:00 18:00:00 21:00:00 00:00:00

2-3-2005

03:00:00 06:00:00 09:00:00 12:00:00 15:00:00 18:00:00

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

55.0

[m^3/s] Time Series Discharge

Inlet & outlet discharge hydrograph

30


Figure 5.5: Typical simulated discharge and Q - H relationship of the Modeled results

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0 46.0 48.0 50.0 52.0 54.0 56.0

[m^3/s]

508.1

508.2

508.2

508.3

508.3

508.4

508.4

508.5

508.5

508.6

508.6

508.7

508.7

508.8

508.8

508.9

508.9

509.0

509.0

509.1

509.1

509.2

[meter] Q - H MAIN CHANNEL 36497.14

Q – H relationship at chainage 36497 m

31


5.1.6 Comparison of Modeled Results using MIKE GIS

Comparison work for the simulated model results that comprise the same number of

item was conducted using the MIKE GIS tool as depicted here after with their respective

descriptions for demonstration purpose (Figure 5.6).

Figure 5.6: Comparisons of delta discharge in m^3/s at outlet chainage

Comparison of modeled result without and with a constant rainfall of 25 mm/day

Comparison of modeled result without and with a constant evaporation of 10 mm/day

32


These above figure shows the comparison analysis of consideration of evaporation and

rainfall rates during model simulation; and it can been concluded that they were not very

significant to consider them during the model set up as they were practically

insignificant bearing in mind the study area.

5.2 Sensitivity Analysis

Sensitivity analysis is a technique used to determine how different values of an

independent variable will impact a particular dependent variable under a given set of

assumptions. Therefore, this study work conducts the sensitivity analysis of the impact

of uncertainties in channel geometry, roughness and impacts of numerical flow

descriptions by creating a given set of scenarios within the study area based on the

prevailing conditions on ground.

5.2.1 Impacts of uncertainties in channel geometry

It is known that the spatial and temporal variations of rainfall and the concurrent

variation of the abstraction processes such as depressions define the runoff

characteristics resulted from the given rainfall. Thus, when runoff commences, the

geometry of the drainage channels have a large influence on the runoff characteristics

from the watershed; although it was very difficult to quantify the effect. In this study

report, geometry of the channel network of the watershed was analyzed which

comprises basically the shape of cross-section, length, and slope of the channel as

those have a significant impact on the resulted hydrograph within the study area.

Therefore, there was definitely an uncertainty created due to the assumed fixed channel

geometry considered while running the model which might be different with the

prevailing condition as there was difference as well among the DEM and Google Earth

derived model cross-sections as shown on Figure 5.7 as well as annex 1 and 2.

33


Figure 5.7: Simulated water depth using cross-section from DEM & Google Earth map.

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 110.0 120.0 130.0 140.0 150.0 [Meter]

531.0 532.0 533.0 534.0 535.0 536.0 537.0 538.0 539.0 540.0 541.0 542.0 543.0 544.0 545.0 546.0 547.0 548.0 549.0 550.0 551.0 552.0 553.0 554.0 555.0 556.0 557.0 558.0 559.0 560.0 561.0 562.0

563.0

Main Channel 35,365 on 01.03.2005 13:43:30 (Google Earth)

Water depth ranges 1.2 – 2.4 m

[Depth in meter]

400.0 420.0 440.0 460.0 480.0 500.0 520.0 540.0 560.0 580.0 600.0 620.0 640.0 660.0 [Meter]

515.5 516.0

516.5 517.0 517.5 518.0 518.5 519.0 519.5 520.0 520.5 521.0 521.5 522.0 522.5 523.0 523.5 524.0 524.5 525.0 525.5 526.0 526.5 527.0 527.5 528.0 528.5 529.0 529.5 530.0 530.5

531.0

[Depth in meter] Main Channel 35,365 on 01.03.2005 12:51:30 (DEM)

Water depth ranges 0.0 – 0.5 m

34


As shown on Figure 5.7, it can be concluded that there exist a very significant impact on

the simulated minimum and maximum water depth time series throughout the model

reach due to the uncertainty of the accuracy of the utilized cross-section inputs.

5.2.2 Impacts of uncertainties in channel roughness

In a similar fashion as mentioned above, the roughness of the channel also plays a

significant role towards the runoff characteristics resulted from the watershed. Thus, the

impacts of uncertainties in channel roughness executed in this study work alters the

shape of the hydrograph produced i.e. more roughness value results flatter hydrograph

with lesser peak discharge in comparison to that of less rough. And, from the

comparison analysis work of the modeled results depicted in Figure 5.8, it can be

concluded that there exist a slight uncertainty in the simulated model results due to the

uncertainty of the accuracy of the roughness coefficient utilized.

Figure 5.8: Comparison of model results of discharge at Al-Awabi watershed outlet of

main channel with different Manning - M values

Comparison of modeled result using Manning’s - M = 20 Versus M = 40

Comparison of modeled result using Manning’s - M = 20 Versus M = 30

35


It was also observed that there exist slight differences to the simulated model results

whether the distributed resistance number was chosen instead of uniform one along the

Transversal Distribution as shown in Figures 5.9. For instance, the simulated peak

discharge at the outlet using uniform Manning – M = 20 was 55.4 m^3/s compared to

that of triple zone of Manning – M = 20, 25 & 30 which was 55.5m^3/s and that of

55.6m^3/s using M = 30, 35, & 40 (Figure 5.9).

Figure 5.9: Simulated discharge hydrograph at outlet chainage

00:00 1-3-2005

03:00 06:00 09:00 12:00 15:00 18:000 21:00 00:00 2-3-2005

03:00 06:00 09:00 12:00 15:00 18:00 10 12

14

16 18

20

22 24

26

28 30

32

34 36

38

40 42

44

46

48 50

52

54 56

[M^3/s]

Time Series Discharge

Model result of main channel with Triple Zone of Manning - M 30, 35 & 40

00:00 1-3-2005

03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00 2-3-2005

03:00 06:00 09:00 12:00 15:00 18:0

5

10

15

20

25

30

35

40

45

50

55.0 [M^3/s]


Model result of main channel with Triple Zone of Manning - M 20, 25 & 30

36


In addition, from the comparison analysis work of the modeled results depicted in Figure

5.10, it can be concluded that there exist a significant uncertainty in the simulated model

results due to the uncertainty of the accuracy of the roughness coefficient utilized

particularly in comparison with that of Figure 5.8. And, this proves the degree of

uncertainty with the assumption of uniform roughness coefficient along the traverse

cross-section on which on ground might be practically variable.

Figure 5.10: Comparison of simulated discharge at outlet of main channel with different

Manning – M values

Comparison of modeled result using Uniform M = 20 Versus Triple M = 30, 35, 40

Comparison of modeled result using Uniform M = 20 Versus Triple M = 20, 25, 30

37


Therefore, from all reputed figures above in this section, it can be concluded that the

higher channel roughness coefficient of the Wadi bed decreases the resulted runoff at

the downstream of Al-Awabi Wadi; and of course due to the certain degree of

uncertainty level on the estimated roughness coefficient utilized a minor variations was

observed in the modeled results in all the model runs executed.

5.2.3 Impacts of numerical flow descriptions

The numerical flow descriptions are derived through the continuity equation in one

dimensional form and a simplified form of the momentum equation. Thus, there is

definitely an impact as the type of flow found in the prevailing natural watershed was not

absolutely the same with the model set up. For instance, it was realized that there exist

a significant difference with the simulated results obtained using the fully dynamic option

(Figure 5.11) versus the diffusive wave option (Figure 5.12) on which it was noticed a

huge error occurred at early time of simulation at the upstream of the model reach in the

second case.

Figure 5.13 compares the simulated results using the different options of wave

approximation provided at the HD parameter MIKE module. And from this comparison

figure, it can be easily observed that there exist a slight difference to the model results

as to which flow condition option was taken and the uncertainty level at deciding the

correct flow conditions during model simulation. For instance, it was noticed that the

kinematic wave option was producing an incomplete simulation; and this could be

because the kinematic wave has certain degree of uncertainty to that of the real one on

ground due to transmission losses, for instance, which leads to no hydrologic balance

between the rainfall events.

38


Figure 5.11: Simulated outlet discharge using Fully Dynamic wave approximation

Figure 5.12: Simulated outlet discharge using Diffusive wave approximation

Finally, relatively speaking the simulated model results using the fully dynamic option

was found to be better result in comparison to those of diffusive wave and kinematic

wave approaches model result.

00:00 1-3-2005

03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00 2-3-2005

03:00 06:00 09:00 12:00 15:00 18:00

0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

3200

3400

3600.0 [M^3/s]


00:00 1-3-2005

03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00 2-3-2005

03:00 06:00 09:00 12:00 15:00 18:00

0

5

10

15

20

25

30

35

40

45

50

55

[M^3/s]


39


Figure 5.13: Comparison of simulated discharge at outlet using fully dynamic versus

Diffusive wave approximation

5.3 Scenario Analysis

This is a process of analyzing possible future events by considering alternative possible

outcomes. There are many different ways to approach scenario analysis, but this study

work mainly focuses on developing a simple method scenario analysis on the impacts of

rainfall characteristics considering the partial area coverage of rainfall and transmission

losses realized within the Al-Awabi watershed area.

5.3.1 Scenario analysis considering partial area coverage of rainfall

From the literature and studies conducted in this paper, it was discovered that rainfall

events were composed of several rain-producing cells which appear and decay during

the duration of the rainfall event both singly and with several cells together. For

example, there could be a flooding hydrograph produced at an outlet of the sub-

watershed as a result of the localized convective type of rainfall, where the cells are

typically in the order of few square kilometers in spatial coverage for short durations of

time within the Al-Awabi watershed. Thus, in order to envisage the impact of partial

40


rainfall coverage on the generated output runoff, the model was set with different BC of

lateral inflows along the main channel as shown in Figure 5.14.

Furthermore, as mentioned in Section two, a daily runoff data at Awabi gauging site for

a period of 23 years (1984–2007) was recorded and analyzed (Table 2.2); and

accordingly an observed unit hydrograph was created. Thus, using the observed unit

hydrograph as a basis, a synthetic inflow hydrograph for each sub watershed was

derived using SCS Dimensionless Hydrograph as shown in Appendix 1; and was

utilized as an input inflow, UBC, for the 1D-Hydraulic Model Set-up. Furthermore,

consulting published literatures such as (Andrew Chadwick, 2004), runoff coefficient for

the Al-Awabi watershed was estimated to be in the range of 0.025 to 0.05 and this was

used for calculating the discharge while developing the synthetic inflow hydrograph as

shown the figures within Appendix 1.

Figure 5.14: Relative locations of lateral inflows at three tributaries to main channel.

JT1

JT2

JT3

UBC

DBC

41


The runoff hydrograph generated as well as the peak discharge at the Al-Awabi outlet

was found to be highly dependent and variable with the time variation of the rainfall

intensity, coverage size whether it is full or partial, and with the rainfall movement

direction as shown in Figure 5.15 putting into consideration its shape and geographical

location. The hydrograph could be flatter with less peak or vice-versa with different

peak values depending on the conditions taken.

Table 5.3: Comparison of partial rainfall coverage on simulated water-depth

Cross-

sectional

Chainage

Simulated Water Depth with lateral inflows at

UBC UBC & JT1 UBC & JT2 UBC & JT3 UBC & 36365

Min. Max. Min. Max. Min. Max. Min. Max. Min. Max.

8365 (UBC) 0.0 0.6 0.0 0.6 0.0 0.6 0.0 0.6 0.0 0.6

15459 (JT1) 0.5 1.2 0.7 1.6 0.5 1.2 0.5 1.2 0.5 1.2

25786 (JT2) 0.6 1.2 0.7 1.5 0.7 1.5 0.6 1.2 0.6 1.2

29660 (JT3) 0.6 1.3 0.7 1.7 0.7 1.7 0.7 1.6 0.6 1.3

31365 1.6 3.1 1.7 4.0 1.8 4.0 2.0 4.0 1.6 3.1

36365 0.0 1.2 0.0 1.6 0.0 1.6 0.0 1.6 0.0 1.6

36630(DBC) 1.1 2.2 1.1 2.9 1.1 2.8 1.1 2.8 1.1 2.8

Figure 5.15: Comparison of simulated discharge at outlet using lateral inflows along

longitudinal main channel

Comparison of modeled result including lateral inflows versus without of them

42


It was also observed that partial coverage has a significant impact to the simulated

discharge as well as to its peaks, and to the minimum and maximum water depth at any

relative location within the channel reach; and this was dependent not only on the

amount of rainfall occurred partially but also where it occurs in terms of spatial location

as shown in Table 5.3 and Figure 5.15. Therefore, there exist very significant impacts

on the modeled output results whether the occurred rainfall spatially and temporally

varies as well as whether it was fully or partially covered.

5.3.2 Scenario analysis considering transmission losses

Even though it was very difficult to quantify, it was known that transmission losses does

exist in the Wadi bed channel networks of this study area i.e. infiltration or leakages

through Wadi channel bed. Thus, this section focuses on conducting a scenario

analysis helpful in developing and representing a good tool for estimation of the

transmission loss within Al-Awabi watershed when coupled with a channel routing

model so as to see the effect. Hence, bearing in mind the complexity as well as the lack

of data of the geology and soil profiles in the study area, and the need to simulate

infiltration through these vertically heterogeneous layers, a more generalized and simple

method was used.

The model has an option of setting a ground water leakage (GWL) in the HD parameter

module so as to define a leakage coefficient for additional loss of water from the Wadi

flow to the groundwater which in turn could be used indirectly to estimate the amount of

transmission loss. Thus, for this study area, a GWL coefficient in the range of 0.001 to

0.00001 was estimated and used for the model simulation in order to visualize the

impact of transmission loss through the Wadi channel. For demonstration purpose the

model result obtained using GWL = 0.00001 is shown in Figure 5.16, and from this

figure, there was a minor impact on the simulated discharge or water depth; for instance

the peak discharge reduces by about 0.1m^3/s as shown on Figure 5.17. But, if

infiltration rate or GWL coefficient was taken to be higher, even a significant impact on

the simulated model result was observed.

43


Figure 5.16: Simulated outlet discharge with GWL of 0.00001

Figure 5.17: Comparison of simulated outlet discharge using 0.00001 GWL along

longitudinal main channel

00:00 1-3-2005

03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00 2-3-2005

03:00 06:00 09:00 12:00 15:00 18:00

0

5

10

15

20

25

30

35

40

45

50

55

[M^3/s]


44


Generally, the model results obtained demonstrate that with increasing in infiltration rate

or leakage coefficient, channel length and/or active channel width increases the

transmission loss from the Al-Awabi Wadi flow. The infiltration process was also

detected through the water-content variations in the vadose zone; hence, the

irregularities of the plot depicted in Figure 5.17 above could be typically because of the

variations in water content, and the effective wetted perimeter for infiltration which in

turn depends on the variations of active channel width and water depth along the

modeled Awabi reach as depicted on Figure 5.18. Furthermore, it was observed that

large floods with higher peak water levels produce higher transmission losses.

Figure 5.18: Simulated water depth time series at chainage 31365

Therefore, there exist very significant impacts on the modeled output results whether

the transmission losses realized from the channel bed was considered or not

particularly with higher flash flood water.

00:00 1-3-2005

03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00 2-3-2005

03:00 06:00 09:00 12:00 15:00 18:00

2

2.2

2.4

2.6

2.8

3

3.2

3.4

3.6

3.8

4

4.2

4.4

4.6

4.8

5

5.2 [Meter]

Time Series Depth in meter

45


6. Conclusion

Analysis of flash flood routing within the Al-Awabi watershed channels were investigated

in this study, focusing on: (1) Preparation of inflow hydrographs and estimation of model

inflows BC; (2) flood routing model set up, (3) sensitivity analysis work on certain

uncertainties such as channel geometry and roughness, and (4) studying the scenario

analysis on impacts of rainfall characteristics and transmission loss. And, the main

conclusions inferred from the study are:

As it had been already reputed above, there was scarcity of data which was

relevant for the model setup; hence an estimated inflow hydrographs was applied

during the model simulation by utilizing the recorded runoff data at Awabi

gauging as a basis. Thus, it was very difficult to talk about quality of data, and it

was also not easy task to model and forecast inflow hydrographs bearing in mind

the lack of enough available data in the target area, Al-Awabi watershed.

Furthermore, Channel cross-sections were one of the main inputs to the 1D-

Hydraulic Model used for the analysis of flash flood routing of the Al-Awabi

watershed and it was observed that there were huge uncertainties with the

values utilized for the model run which contributes to poor model data quality.

The performance of the calibrated 1D-Hydraulic Model i.e. MIKE 11 HD was

assessed and validated to simulate the flash flood routing analysis at different

cross-sections of the main channel reach. And from this study, although there

were major gap and problems in data as well as in the prevailing topography,

slope and other HD parameters, it was concluded that the 1D-Hydraulic

Modelling utilized for flood routing analysis work can be applied for the Al-Awabi

watershed. Hence, the calibrated model can be used to simulate possible future

floods.

From the simulated model results, it was observed that the model was sensitive

to the type BC chosen and taken, channel cross sections and its roughness

46


coefficient utilized throughout the model reach. Furthermore, this study work

conducts the sensitivity analysis of the impact of uncertainties in channel

geometry, roughness and impacts of numerical flow descriptions; and all cases it

was observed that there was a little impact on the simulated model results as the

result of the reputed uncertainty as discussed earlier. Hence, it can be concluded

that the model was not robust.

Although further insight work is required in order to come up with a concrete

conclusion about its impact quantification due to partial rainfall coverage and

transmission losses via the Wade channel beds, it was noticed that there exist an

impact to the simulated downstream discharge of the study area.

47


7. Limitation of study and Recommendation

As it had been mentioned in the previous sections respectively, the main limitation and

drawback of this study work was lack and/or scarcity of data requirement as everything

was almost an estimate which makes the whole task executed and covered in this

thesis like a virtual work. For instance, the BC used were quite questionable for their

practicability as well as there was certain degree of uncertainty with the cross sections,

and the model parameters such as the Manning’s roughness coefficient used in model

setup.

Having said that here are some recommendations drawn out of this study work and

reputed below:

The use of modern surveying technologies such as Total Station survey

measurements are recommended as they could provide better results with higher

degree of accuracy about the topography of the study area than the utilized DEM

and Google Earth map information. And this could be practically executed in this

study area by taking field survey measurements for the whole Al-Awabi

watershed topography with special emphasis on and along the wade channel

networks. Therefore, it would be advisable to make a pre-model test for a smaller

sub-watershed with the entire data requirement full-filled on one of the tributary

channels rather than analyzing for the whole Al-Awabi watershed. As the HD

model requires continuous flow of water with adequate channel geometry

information; but, to the contrary, the prevailing condition in this study area is

Ephemeral River i.e. wades which has no flow of water in most of the time period

as well as the topography is really ragged and steep slope area.

In order to have better understanding about the dynamics of the flash flood that

could occur within the watershed, an adequate number of structures/stations

have to be set up along the main channel such as weirs, culverts, and control

structures. For example, there could be a flooding hydrograph produced at an

48


outlet of the sub-watershed as a result of the localized convective type of rainfall

for short durations of time within the Al-Awabi watershed.

Last but not least, further investigation work is recommended which embraces,

for example, comparison of the same tasks that could be obtained from the rest

of watersheds found at the mountainous part of the Batinah region, Oman; so as

to have a better understanding of the flood routing analysis work which couples

the hydrological water balance effect between the rainfall events. Therefore, this

paper can be used as a starting material for further detailed modeling work that

encompasses the whole channel network of the entire watershed area.

49


Theses

1. This study was conducted at the mountainous catchment part of Batinah Region

of the Sultanate of Oman called Al-Awabi watershed which is about 260km2 in

area and with about 40 Km long main channel.

2. It was observed that, the drainage pattern of Al-Awabi watershed can be

categorized as Dendritic pattern where tributary branch and erode headwater in

random fashion which results in slopes with no predominant direction or

orientation; as well as Rectangular pattern mainly occurring in this case along

and near to the main channel network as shown in Figure 2.7.

3. Because the Al-Awabi channel network comprises a Wadi main channel which

stretches for about 40 Km with its bed level ranging from about 500m at the

downstream end to 2200m at the upstream end a.m.s.l, and a numerous number

of channel braches on top of the uncertainty on channel geometry; it was

practically difficult task to set up a model for the whole watershed. Therefore, the

model set up was executed only for the lower about 24 Km long of the main

channel.

4. There was scarcity of rainfall – runoff data which was relevant for the model

setup; hence an estimated inflow hydrographs was applied during the model

simulation by utilizing the recorded runoff data at Awabi gauge as a basis.

5. Channel cross-sections were the main input to the 1D-Hydraulic Model used for

the analysis of flash flood routing of the Al-Awabi watershed. Hence, a

methodology for extracting the channel cross-sections from ASTER DEM

(27mX27m) and Google Earth map were used in this study area.

6. From this study, although there were major gap and problems in data as well as

in the prevailing topography, slope and other HD parameters, it was concluded

that the 1D-Hydraulic Modelling utilized for flood routing analysis work can be

applied for the Al-Awabi watershed.

7. From the simulated model results, it was observed that the calibrated model was

sensitive to the type BC chosen and taken, channel cross sections and its

roughness coefficient utilized throughout the model reach.

50


8. Although further insight work is required in order to come up with a concrete

conclusion about its impact quantification due to partial rainfall coverage and

transmission losses via the Wade channel beds, it was noticed that there exist an

impact to the simulated downstream discharge within the study area.

9. The main limitation and drawback of this study work was lack and/or scarcity of

model data requirement as everything was almost an estimate which makes the

whole task executed and covered in this thesis like a virtual work.

10. Finally, this paper can be used as a starting material for further detailed modeling

work that encompasses the whole channel network of the entire watershed area

so as to have a better understanding of the flood routing analysis work.

51


Bibliography

Andrew Chadwick, J. M. (2004). HYDRAULICS IN CIVIL AND ENVIRONMENTAL

ENGINEERING (Fourth ed.). London, EC4P 4EE, UK: Spon Press.

Brutsaert, W. (2005). Hydrology An Introduction (2006 ed.). Cambridge, UK: Cambridge

University Press.

DHI. (2009). MIKE 11 - A Modelling System for Rivers and Channels. Denmark: DHI.

Garbrecht, L. W. (2000). The Treatment of Flat Areas and Depressions in Automated

Drainage Analysis for Raster Digital Elevation Models. In A. G. Montgomery,

Hydrological Applications of GIS (pp. 23-35). Chichester, West Sussex PO19 1UD,

England: John Wiley & Sons Ltd.

GDEM, A. (2009, June 29). ASTER GDEM. Retrieved March 31, 2010, from ASTER

GDEM: http://www.gdem.aster.ersdac.or.jp/

John A. Roberson, J. J. (1988). Hydraulic Engineering (1988 ed.). Boston,

Massachusetts 02108, USA: Houghton Mifflin Company.

Mays, L. W. (2005). WATER RESOURCES ENGINEERING (2005 ed.). Tempe,

Arizona, USA: Wiley & Sons.

Shamsi, U. (2005). GIS Applications for Water, Wastewater, and Stormwater Systems.

Boca Raton, Florida 33431, USA: CRC Press.

52


Appendix

Appendix 1: SCS Dimensionless Hydrograph, and calculated model inflow hydrograph

“The SCS dimensionless hydrograph is an idealized shape that approximates the flow

from an intense storm from a small watershed. The dimensionless hydrograph

arbitrarily has units of 100 units of flow for the peak and 100 units of time for the

duration of flow. The area under a dimensionless hydrograph has 2,620 square units of

runoff. The SCS hydrograph has 19 constant ordinates that represent percentages of

flow and time as depicted in the figure below.”

(http://www.egr.msu.edu/~northco2/BE481/SCShydrograph.htm, 2000).

53


Conversion Factors: The first factor is u and each single unit has a value of u = Q

/2620. Where: Q is the total runoff volume in hectare-meter. The second factor is w and

each unit of flow has a value of w = q/100 where: q is the peak runoff. And, the third

factor is k and each unit of time on the dimensionless hydrograph represents in the

design hydrograph. Since w is equal to 1/100 of the design peak flow, k must be equal

to 1/100 of the design duration, and u is 1/2620 of the design flood volume; therefore

w*k = u and k = u/w. Thus, the coordinates of the design hydrograph are obtained by

multiplying the ordinates and abscissas of the dimensionless hydrograph by w and k

respectively. For instance, synthetic hydrograph calculations based on observed rainfall

and runoff event on March 01, 2005 is shown below:

k=98 w=0.0 k=95 w=0.0 k=96 w=0.1 k=96 w=0.1 k=92 w=0.9

Point Time

Axis

Flow

Axis

K*t in

Hr.

w*q in

m^3/s

K*t in

Hr.

w*q in

m^3/s

K*t in

Hr.

w*q in

m^3/s

K*t in

Hr.

w*q in

m^3/s

K*t in

Hr.

w*q in

m^3/s

a 0 0 0 0 0 0 0 0 0 0 0 0

b 2 3 3.3 0.1 3.2 0.1 3.2 0.2 3.2 0.3 3.1 2.6

c 6 19 9.8 0.3 9.5 0.7 9.6 1 9.6 1.6 9.2 16.7

d 8 31 13.1 0.5 12.7 1.1 12.8 1.6 12.8 2.7 12.3 27.3

e 12 66 19.6 1.1 19 2.3 19.2 3.4 19.2 5.7 18.4 58.1

f 14 82 22.9 1.4 22.2 2.9 22.4 4.3 22.4 7.1 21.5 72.2

g 16 93 26.1 1.6 25.3 3.3 25.6 4.8 25.6 8.1 24.5 81.8

h 18 99 29.4 1.7 28.5 3.5 28.8 5.1 28.8 8.6 27.6 87.1

i 20 100 32.7 1.7 31.7 3.5 32 5.2 32 8.7 30.7 88

j 22 99 35.9 1.7 34.8 3.5 35.2 5.1 35.2 8.6 33.7 87.2

k 24 93 39.2 1.6 38 3.3 38.4 4.8 38.4 8.1 36.8 81.8

l 26 86 42.5 1.5 41.2 3 41.6 4.5 41.6 7.5 39.9 75.7

m 30 68 49 1.2 47.5 2.4 48 3.5 48 5.9 46 59.8

n 34 46 55.5 0.8 53.8 1.6 54.4 2.4 54.4 4.0 52.1 40.5

o 38 33 62.1 0.6 60.2 1.2 60.8 1.7 60.8 2.9 58.3 29

p 44 21 71.9 0.4 69.7 0.7 70.4 1.1 70.4 1.8 67.5 18.5

q 52 11 84.9 0.2 82.3 0.4 83.2 0.6 83.2 1 79.7 9.7

r 64 4 104 0.1 101 0.1 102 0.2 102 0.3 98.1 3.5

s 100 0 163 0 158 0 160 0 160 0 153 0

54


Date Areal Rainfall (mm) Area( Km^2) Runoff Coefficient © Discharge (Q=CIA)

01.03.2005 50 5 0.025 1,7

01.03.2005 50 10 0.025 3,5

01.03.2005 50 15 0.025 5,2

01.03.2005 50 25 0.025 8,7

Total 50 254 88

00,5

11,5

0 2500 5000 7500 10000

Dis

char

ge

(m^3/s

)

Time (minute)

Synthetic Inflow Hydrograph - 1

0

2

4

0 2500 5000 7500 10000

Dis

char

ge

(m^3/s

)

Time (minute)


0

2

4

6

0 2500 5000 7500 10000Dis

char

ge

(m^3

/s)

Time (minute)


0

5

10

0 2000 4000 6000 8000 10000Dis

char

ge

(m^3

/s)

Time (minute)


0

20

40

60

80

100

0 2000 4000 6000 8000 10000

Dis

char

ge(

m^3

/s)

Time (minute)

Synthetic Outflow Hydrograph

Synthetic Outflow Hydrograph

55


Annex

Annex 1: DEM (27x27m) extracted cross-sections of Al-Awabi Wadi (where X-axis and

Y-axis represents channel width and elevation in meter respectively)

960

965

0 10 20 30

Chainage 8365

899,3

899,4

899,5

899,6

899,7

899,8

0 10 20 30

Chainage 9365

845,0

845,5

846,0

846,5

847,0

0 10 20 30

Chainage 11365

834,0

835,0

836,0

837,0

0 10 20 30 40 50

Chainage 12365

805,5

806,0

806,5

807,0

0 10 20 30 40 50

Chainage 13365

794,0

795,0

796,0

797,0

798,0

0 20 40 60 80

Chainage 14365

780,0

781,0

782,0

783,0

784,0

0 20 40 60

Chainage 15365

760,0

765,0

770,0

775,0

0 50 100 150

Chainage 16365

750,0

755,0

760,0

765,0

0 25 50 75 100

Chainage 17365

740,0

741,0

742,0

743,0

744,0

0 20 40 60 80

Chainage 18635

726,5

727,0

727,5

728,0

728,5

0 50 100 150

Chainage 19365

713,0

714,0

715,0

716,0

717,0

718,0

719,0

0 25 50 75 100

Chainage 20365

705,0

706,0

707,0

708,0

0 20 40

Chainage 21365

684,0

684,0

684,0

684,0

684,1

0 25 50

Chainage 22365

667,8

668,0

668,2

668,4

668,6

668,8

0 20 40

Chainage 23365

654,5

655,0

655,5

656,0

656,5

657,0

0 25 50 75 100

Chainage 24365

56


645,0

646,0

647,0

648,0

649,0

650,0

0 50 100 150

Chainage 25365

624,0

626,0

628,0

630,0

632,0

634,0

636,0

0 50 100 150

Chainage 26365

613,0

614,0

615,0

616,0

617,0

0 40 80 120

Chainage 27365

596,0

597,0

598,0

599,0

600,0

601,0

602,0

603,0

0 25 50 75 100

Chainage 28365

570,0

575,0

580,0

585,0

590,0

0 50 100 150 200

Chainage 30365

560,0

565,0

570,0

575,0

580,0

0 50 100 150

Chainage 31365

550,0

551,0

552,0

553,0

554,0

555,0

556,0

0 25 50 75 100

Chainage 32365

545,0

550,0

555,0

560,0

565,0

0 25 50 75

Chainage 33365

525,0

530,0

535,0

540,0

545,0

0 50 100 150

Chainage 34365

515,0

520,0

525,0

530,0

535,0

0 75 150 225

Chainage 35365

505,0

510,0

515,0

520,0

525,0

530,0

0 25 50 75 100

Chainage 36365

57


Annex 2: Cross-sections derived from Google Earth of Al-Awabi Wadi (where X-axis

and Y-axis represents channel width and elevation in meter respectively)

965

966

967

968

969

0 10 20 30

Chainage 8365

898

903

908

0 20 40 60

Chainage 9365

879,5

880

880,5

881

0 10 20 30 40

Chainage 10365

853,5

854

854,5

855

855,5

856

0 20 40 60

Chainage 11365

827

828

829

830

831

0 10 20 30 40

Chainage 12365

811,5

812

812,5

813

813,5

814

0 20 40 60

Chainage 13365

800

802

804

806

808

0 25 50 75

Chainage 14365

784

786

788

790

0 10 20 30 40 50

Chainage 15365

782

783

784

785

786

0 10 20 30 40 50


771,5

772,5

773,5

774,5

0 25 50 75 100 125 150 175

Chainage 16365

758

760

762

764

766

0 25 50 75 100

Chainage 17365

744

745

746

747

748

749

750

0 25 50 75 100 125 150

Chainage 18635

728

731

734

737

740

743

0 25 50 75 100 125 150 175

Chainage 19365

717

719

721

723

725

727

729

0 50 100 150 200

Chainage 20365

713,5

714,5

715,5

716,5

717,5

0 10 20 30 40

Chainage 21365

58


693,5

694,5

695,5

696,5

0 25 50 75

Chainage 22365

675

676

677

678

679

0 10 20 30 40 50

Chainage 23365

656

661

666

0 25 50 75

Chainage 24365

644

649

654

0 50 100 150 200

Chainage 25365

640

645

650

0 25 50 75 100 125

Chainage 25785 (TJ2)

632

635

638

641

644

647

0 50 100 150 200 250

Chainage 26365

595

605

615

0 50 100 150

Chainage 27365

590

595

600

605

610

0 50 100 150 200 250

Chainage 28365

583

588

593

598

603

0 50 100 150 200 250


578

588

598

0 50 100 150

Chainage 30365

574

579

584

589

0 25 50 75 100

Chainage 31365

570

575

580

585

590

0 25 50 75 100

Chainage 32365

547

549

551

553

0 20 40 60 80 100 120

Chainage 33365

540

545

550

555

560

0 25 50 75 100 125

Chainage 34365

525

535

545

555

0 25 50 75 100 125

Chainage 35365

528

533

538

543

0 25 50 75 100

Chainage 36365

879,5

880

880,5

881

0 10 20 30 40

Chainage 10365

506

511

516

0 25 50 75

Chainage 36630 (DBC)

59


Annex 3: Comparison of cross-sections derived from Google Earth (GE) and DEM of

Al-Awabi Wadi Channel.

960

962

964

966

968

0 10 20 30 40

Elev

atio

n (m

)

Width @ 8365 (m)GE DEM

895

900

905

910

915

0 10 20 30 40

Elev

atio

n (m

)

Width @ 9365 (m)

GE DEM

840

845

850

855

860

0 10 20 30 40

Elev

atio

n (m

)

Width @ 11365 (m)

GE DEM

826

828

830

832

834

836

838

0 10 20 30 40 50 60

Elev

atio

n (m

)

Width @ 12365 (m)

GE DEM

804

806

808

810

812

814

816

0 10 20 30 40 50 60

Elev

atio

n (m

)


790

795

800

805

810

0 20 40 60 80 100

Elev

atio

n (m

)

Width @ 14365 (m)

GE DEM

780

782

784

786

788

790

792

0 20 40 60

Elev

atio

n (m

)

Width @ 15365 (m)

GE DEM

760

765

770

775

780

0 50 100 150 200

Elev

atio

n (m

)

Width @ 16365 (m)

GE DEM

60


750

755

760

765

770

0 25 50 75 100 125

Ele

vati

on

(m)

Width @ 17365 (m)

GE DEM

740

742

744

746

748

750

752

0 20 40 60 80 100

Elev

atio

n (m

)

Width @ 18365 (m)

GE DEM

725

730

735

740

0 40 80 120 160 200

Ele

vati

on

(m)


710

720

730

740

0 25 50 75 100 125

Ele

vati

on

(m)

Width @ 20365 (m)

GE DEM

0

5000

10000

0 10 20 30 40 50 60

Ele

vati

on

(m)

Width @ 21365 (m)

GE DEM

680

685

690

695

700

0 20 40 60 80 100

Elev

atio

n (m

)

Width @ 22365 (m)

GE DEM

665

670

675

680

0 10 20 30 40 50 60

Elev

atio

n (m

)

Width @ 23365 (m)

GE DEM

650

655

660

665

670

0 25 50 75 100 125

Elev

atio

n (m

)


640

645

650

655

660

0 50 100 150 200

Elev

atio

n (m

)


620

630

640

650

0 50 100 150 200

Ele

vati

on

(m)


61


590

600

610

620

0 25 50 75 100 125 150

Elev

atio

n (m

)

Width @ 27365 (m)

GE DEM

595

600

605

610

615

0 25 50 75 100 125

Elev

atio

n (m

)

Width @ 28365 (m)

GE DEM

560

570

580

590

600

610

0 50 100 150 200

Ele

vati

on

(m)

Width @ 30365 (m)

GE DEM

560

570

580

590

600

0 25 50 75 100 125

Ele

vati

on

(m)


540550560570580590600

0 25 50 75 100 125

Elev

atio

n (m

)

Width @ 32365 (m)

GE DEM

545

550

555

560

565

0 25 50 75 100

Elev

atio

n (m

)

Width @ 33365 (m)

GE DEM

520530540550560570

0 50 100 150 200

Ele

vati

on

(m)


500

520

540

560

580

0 50 100 150 200

Elev

atio

n (m

)

Width @ 35365 (m)

GE DEM

500

510

520

530

540

550

0 25 50 75 100 125

Elev

atio

n (m

)

Width @ 36365 (m)

GE DEM

Documents

Analysis of Flash Flood Routing by Means of 1D - Hydraulic ... Thesis... · iv Analysis of Flash Flood Routing by Means of 1D - Hydraulic Modeling 2010 Abstract This study was conducted