INVENTORY AND MAPPING OF PROBABLY CONTAMI- NATED … · INVENTORY AND MAPPING OF PROBABLY CONTAMI-NATED ... Basic data from standardized site screening visit ... INVENTORY AND MAPPING

JULY 2013

MINISTRY OF ENVIRONMENT AND FORESTS

INVENTORY AND MAPPING

OF PROBABLY CONTAMI-

NATED SITES IN INDIA DRAFT REPORT OF TASK 5: PRIORITIZATION OF SITES

JULY 2013

MINISTRY OF ENVIRONMENT AND FORESTS

INVENTORY AND MAPPING

OF PROBABLY CONTAMI-

NATED SITES IN INDIA DRAFT REPORT OF TASK 5: PRIORITIZATION OF SITES

ADDRESS COWI A/S

Parallelvej 2

2800 Kongens Lyngby

Denmark

TEL +45 56 40 00 00

FAX +45 56 40 99 99

WWW cowi.com

PROJECT NO. A019251

DOCUMENT NO. 011

VERSION 001

DATE OF ISSUE 10 July 2013

PREPARED DH, TJR

CHECKED SV

APPROVED SV

http://projects.cowiportal.com/ps/A019251/Documents/3 Project documents/Reports/Task 5 Proritization system/Draft Task 5 report Prioritization system (2).docx

CONTENTS

1 Introduction 7

2 Review of existing approaches 9

3 Selection of appropriate prioritization /ranking

system 12

4 Level of information 14

5 Approach proposed 16

5.1 Introduction 16

5.2 Step 1 Prioritization 17

5.3 Step 2 Prioritization 19

5.4 Calibration and refinement 22

5.5 Testing and review against initial site data 22

5.6 Application to full database 22

6 Plan for completing Task 5 23

APPENDICES

Appendix A Comparison of Prioritization Systems

Appendix B Parameters in Prioritization Systems

Appendix C Methodologies examined

INVENTORY AND MAPPING OF PROBABLY CONTAMI-NATED SITES IN INDIA


7

1 Introduction

The current Report provides a status of the Task 5 Prioritization of Sites for the

project Inventory and Mapping of Probably Contaminated Sites In India, which

was awarded by the Ministry of Environment and Forest to a Consortium

consisting of COWI as lead partner in association with KADAM, Witteveen+Bos

and Tauw as sub-consultants. The Project is funded by the World Bank (WB). Our

work is coordinated with the other two assignments of the National Program for

Rehabilitation of Polluted Sites (NPRPS); Assignment 2: the Development of

Methodologies for NPRPS and Assignment 3: Development of National Program

for Rehabilitation of Polluted Sites

Task 5 is one out of the five tasks to be completed in this project, see the

conceptual model shown on next page.

Objectives The objective of the total assignment is to prepare an inventory of contaminated

sites across India, which will provide information to Central and State governments

when preparing and implementing the National Program for Rehabilitation of

Polluted Sites (NPRPS). The Inventory must be a dynamic tool where data are

adjusted and supplemented on a regular basis.

8 INVENTORY AND MAPPING OF PROBABLY CONTAMI-NATED SITES IN INDIA


The objective of Task 5 is to provide a simple and robust ranking system for

prioritization of sites for further investigation and for remediation actions. The

system will establish an objective ranking, based on the identified contamination

and on the possible impacts on human health and the environment. As such it will

inform decisions on specific remediation activities to be carried out under NPRPS

or by individual agencies.

Output The required output of task 5 is as follows:

A simple and robust system for ranking different sites in terms of priority for

further review and/or action, which can be applied using the information typically

available in the database.

The aim of Task 5 is to establish priorities for further investigation and/or

remediation of sites that have been identified and recorded in the database as

probably contaminated sites. Information collected in the database will often be

incomplete or limited. The system must therefore be easy to apply, be robust to a

wide range of sites and provide consistent relative rankings. The requirements of

the prioritization process were part of the discussions underlying the design of the

Site Inspection approach and the database, so that the systems are consistent.

The task includes The task includes:

› Review of international approaches to objective ranking and priority setting

systems for contaminated or hazardous sites, including the USEPA’s Hazard

Ranking System. The Comprehensive Environmental Pollution Index applied

by CPCB to industrial clusters will also be reviewed as a possible model.

› Discussions of the findings of the initial review with a wide range of

stakeholders (e.g. CPCB and some State Boards) to identify system(s) which

can be the applied easily using the data typically available in the database.

› Testing of the preferred system against a sample of sites from the database and

comparison of the rankings from the systems with the opinions of

knowledgeable specialists who are familiar with typical sites.

› Presentation of the model with the best overall performance to a representative

panel of experts, for further refinement on the basis of discussion and

feedback.

› Practical trialing of the refined model with the two SPCBs in the project to see

how its application supports the objective of identification of contaminated

sites and how it can guide state and local government officials in setting their

priorities and approaches to dealing with the list of sites.

› As the final step in development of the model, it will be applied to the whole

database and the broad rankings which are produced will be reviewed for

consistency and realism by MOEF.

This Draft Report is set out in the structure defined in the Terms of Reference and

covers in detail the first two Sub-Tasks.

Aim of Task



9

2 Review of existing approaches

There are a range of different prioritization systems that have been developed over

the years, mainly drawing on initial work in the 1990s in the USA to assess

identified sites in terms of their importance for inclusion into the National Priorities

List under Superfund.

A review for the European Environment Agency identified 27 different

methodologies worldwide which were, or had been, used in preliminary risk

assessment (see appendix C). The coverage and level of detail of these varies

considerably. After examination of the available information on the system, six of

these are judged to be relevant to the present task. To these six, are added the

Blacksmith Index (from the Blacksmith Institute, currently applied to sites

worldwide) and the Comprehensive Environmental Pollution Index (CEPI)

developed for the CPCB and used in assessment of industrial clusters in India.

The eight selected systems have been examined in some detail for essential

characteristics and for similarities. This range of approaches is believed to cover

the plausible range of practical systems that are available as models for the present

task.

The essential characteristics of the eight systems are summarized immediately

below and the systems are compared in Annex A.

Objective is to assess the relative potential of sites to pose a threat to human health.

Use to register “uncontrolled waste sites” onto National Priorities List. Does not

determine prioritize for funding.

Uses “complex and structured analytical approach to scoring risks”. Main

indicators are in three categories: likelihood of release of hazardous material from

site; toxicity and quantity of the waste; people or sensitive areas as “targets”. Each

pathway scored on sources; likelihood of release; waste characteristics; targets.

Algorithm: Each pathway scored based on estimation of indicators, factors and

parameters. Multiplicative relations, then normalized to 100 point scale. HRS

overall score is based on root-mean square of pathway scores. Source hazard is

based on toxicity/carcinogenicity. Pathways evaluated are Groundwater; surface

water; soil exposure; air. For each of these specific receptors are used.

USEPA Hazard

Ranking System

(1990)



Objective is to establish rational and consistent basis for comparative assessment of

contaminated sites, as input to national remediation program.

Additive numerical method based on sum of indicator scores for Contaminants

Characteristics plus Exposure Pathways plus Receptors.

Ranking puts sites in one of four classes: high; medium; medium-low; and low

risk.

Objective is to provide rapid hazard assessment, sufficient to prioritise sites for

further investigation.

Based on a multiplicative Source –Pathway - Receptor approach, scaled to 0 to1.

Highest score of three pathways used. Factors based on mix of quantitative and

qualitative values.

Objective is to define the “remediation urgency” of sites affected by serious soil

contamination.

Compares site contamination levels against generic intervention values. If above,

then calculation to show whether expected exposure is above human toxicological

intervention values. Separate procedure for ecological risk.

Concentrations in soil can be used to calculate expected concentrations in contact

media and thus exposures, unless actual data exist for contact media, which then

take precedence. Different pathways considered for human health risk, including

direct contact, food, fishing, drinking water.

Output is determination of timescale over which that responsible authority must

start remediation.

Objective is to determine whether further investigations are needed, based on

ranking to three classes of risk.

Focus is on direct contact and water pathways, with calculations using basic data

on contamination levels at source. Scores given on 0 to 3 scale for each element

and combined in “simple algorithm” for each pathway/land use. Highest score

applies.

Objective is to provide local authorities with prioritization system for sites about

which little is known. The prioritization system uses the Source-Pathway-Receptor

concept to assess risks. It is split into two steps. The Step I assessment involves

hazard ranking sites based on their historical industrial uses and the receptor’s

sensitivity. The Step II procedure involves refining the assessment from Step I by

carrying out an exposure assessment. Default data on industry types and potential

receptors used in Step 1. For Step 2 source contaminant data used. Land,

Groundwater and Surface water pathways/receptors assessed separately. Additive

algorithm (matrix) for Contaminant and Pathway/Receptor.

Final risk score is max of the three pathways.

Canada National

Classification

System (NCS) for

contaminated sites

(1992)

New Zealand Risk

Screening System

(2003)

The Netherlands

Remediation

Urgency Method

(1995)

France - Simplified

Risk Assessment

(2001)

GeoEnviron System

for the Prioritization

of Point Sources

(2003)



11

Objective is comparative ranking of contaminated sites worldwide in terms of

health risk.

Source contaminant levels and population at risk are main factors, with pathway

adjustment. The ranking is additive algorithm, using logarithms of key values.

Basic data from standardized site screening visit methodology.

Objective is to help categorizing areas in terms of priority of planning for

intervention in relation to environmental parameters. The system has been applied

to 88 industrial clusters across India, using data from local agencies. Use additive

approach across pollution pathway and receptor. Penalty values calculated for each

factor in each pathway, and then cluster ranked by total score, with 70/100

regarded as “critically polluted”.

The eight systems listed here have been reviewed in detail and a comparative table

prepared (below) setting out key features of each, as the basis for developing a

system appropriate to the situation in India.

Blacksmith Index

(2006)

India

Comprehensive

Environmental

Pollution Index

(CEPI) (2009)



3 Selection of appropriate prioritization

/ranking system

Based on the analysis carried out of the candidate systems, there are common

features in several that can be taken and adapted for use but none is directly

applicable to the current situation in India. The US system is very detailed and

requires considerable information and then draws on specific US data. The

Canadian and New Zealand approaches were developed for specific programs and

are not readily adaptable. The Netherlands approach requires detailed information,

which is typically already available for sites in that country. The French system is

based on specific values for contaminants which are not readily available and

comparable. The UK/Denmark model from Geokon is broadly applicable. A GIS

system is incorporated and the system also includes a 2 step prioritization system

which is suitable for India condition. The Blacksmith Index is applicable but may

not be detailed enough for the requirements of the Ministry. The CEP'I Index

developed for the CPCB is also broadly appropriate but has been developed for

ongoing industrial activity where monitoring data is available and so cannot be

applied directly.

There is enough broad consistency between these different systems to be able to

identify some common features which are taken as the basis for the prioritization

system for this project.

Broad findings On the basis of detailed review of the models discussed above, the following

findings can be set out, with their implications for this project.

› There is no analytical model which is common to most of the models but the

Source-Pathway-Receptor approach is basis for all the systems.

› The key parameters can be values (concentrations, populations) or can be

marks (typically on a scale of 1 to 10).

› Some systems are tailored/biased to particular problems (such as petroleum

contamination) while others are more general but less detailed

› Most systems address impacts on humans, groundwater and surface water. In

some cases they are combined into one overall score: in others they are



13

reported separately. Ground- or surface water can, in some cases, be both

pathway and receptor.

› Data requirements vary. In some systems, default data on sources is provided

or can be used, avoiding or reducing the need for site investigations (but

requiring information on past use and industrial types).

› There are a number of definitions of the receptor, ranging from any human

exposure to estimating the populations affected, usually by distance from the

source.

› The sensitivity of a system to small differences can be increased, at the cost of

additional complexity, which means more variables and data required. (See

table comparing parameters in appendix A)

› The basic algorithm is based either on adding scores for element (additive) or

on multiplying them together (multiplicative). There is no fundamental

reason to prefer one approach over the other. Some systems are a mixture of

both approaches.

› Systems produce either a score or assign a site to a class (typically high,

medium, low, or unclassified). Some systems use the score to produce a

classification.

The approach adopted for this project is based on the Source – Pathway – Receptor

conceptual model, which underlies all the systems reviewed. The logic for this

approach is simple and universally accepted.

It draws on data from Task 1 and 4, which includes site specific information and

sampling data to the extent possible.

The model is to be robust to uncertainty in the data (a small change in the

information available should not make a significant difference in the priority

ranking or classification). Given the constrained coverage and quality of the

information from Task 1 and 4, the precision expected of the output is to be no

more than two significant figures – in other words, it will rank sites on a scale of 1

to 100.

The model uses professional experience and judgment for the details of

calculations and the calibration of the parameters. The approach used is clear and

the relevance and importance of each factor can easily be seen in the outputs.

The model contains the three major components of source, pathways and receptor.

The table in Annex B sets out the main parameters in the five of the current

systems examined that are most accessible and are the best models for the system

to be used in this project.

Basics of proposed

approach

Detailed options for

prioritization process



4 Level of information

Based on the information from the visited institutions and organizations, we have

in Task 1 identified a total of approx. 600 sites all around the country. Table 4.1

gives an overall status of the information level of obtained data in Task 1 and

expected outcome of information obtained in Task 4 (Site Inspection).The level of

information of each site varies widely.

In Task 1 (desktop study) we have obtained reasonable good information about the

general site information and information about source and waste characteristics,

but very little information about pathways. At some sites, there is detailed

information including primary data and knowledge about source, pathways and

receptors. For a few sites, the knowledge is very limited e.g. with virtually only

information about the name and location of the site.

In Task 4 we will obtain supplementary information about the sites based on the

Site Inspections at 100 sites with special information about pathways, receptors and

primary data. In general we expect that the sites can be classified in two categories:

› Site with high level of information: This includes the 100 sites where Site

Inspection will be conducted and probably some additional sites with data

from previous site investigations (e.g. sites investigated by SPCBs, MoEF and

CPCB). Due to combined desk top study and Site Inspection most data in table

4.1 will be available.

› Site with medium to low level of information. This include the remaining

number of sites identified in Task 1 (in total 400-500 sites)



15

Table 4-1 Typical level of information based on data collection in task 1

Topic Explanation Level of information

Task 1

(~ 600 sites)

Task 4

(~100 sites)

Ge

ne

ral

site

in

form

ati

on

Approximate area of site 3-4 4-5

Type of Site 1 = "Point" site (Single industrial site/dump site), 2 =

"Area" site Industrial area or estate (cluster) 3 = Other

4 5

Type of contamination Effluent, Air, Municipal Solid Waste, Bio-medical Waste,

Hazardous Waste, Ship Break Waste, Other

4 5

Type of Industry Types of industrial uses the site has been subjected to 3-4 5

Industrial processes, Type

of hazardous waste

Activity which caused the contamination: According to

schedule 1 – HW Rules 2008

3-4 4-5

Period of operation Beginning year and end year e.g. 1988 - 1995 2 3-4

Operational status Active/ongoing, Closed, Abandoned, Other 3-4 5

Previous or ongoing

remediation activities

Remediation activities 2-3 5

So

urc

e

Primary Contaminant(s) The most important Chemical names is specified 3-4 5

Source characteristic Area and volume of contaminated soil, Physically state of

waste as deposited, Visual contamination

2-3 4-5

Origin of the deposit 2-3 4-5

Existing primary data

Analysis results from e.g. soil, groundwater, surface water,

air

2-3 5

Pa

thw

ay

s

Geology at the site Broad description of the typical stratigraphical sequences 2-3 5

Hydrogeology at the site Overall description of aquifers 2-3 5

Any drainage system

on/outside the site

General description of (drain, trenches, streams) or

streams at the site

1-2 5

Any flooding If any flooding frequency and type should be described 1-2 5

Surface water body Type of Surface water Body , distance from site, 3-4 5

Sensitive use of surface water

Prevailing wind direction 1 4

Re

cep

tors

Use of groundwater aquifer Current and future planned use of any aquifer 3-4 5

Drinking water intakes Distances to nearest well (private and public) 1-2 5

Surface water use Sensitive use of surface water 1-2 5

Ecological areas (m) E.g. Reserves, wetland 1 5

Land use (on site) Agriculture land, Waste land, Water bodies, Forests,

Habitation settlement, Commercial, Industrial, Mixed ,

Other

Previous = 2-3

Current = 3-4

Future = 1-2

Previous = 2-3

Current = 5

Future = 2-3

Land use (vicinity of site) 3-4 5

Activity in the vicinity of the

site?

Distance to nearest habitation or other sensitive activities

(e.g. Kindergarden)

3-4 5

Socio economic conditions 2-3 5

Potential/observed

pathways for spreading

Groundwater pathway, Surface Water pathway, Soil

exposure pathway, Air pathway

2-3 5

Potential/observed

exposure

Direct human contact, Ingestion, Groundwater/surface

water use, Inhalation, Sensitive environments

2-3 5

Observed impacts E.g. observed impacts on humans, animals, flora, fauna 1-2 5

Total population at risk Population at risk 3-4 5

Note: * Score: 1 = Obtained for nearly no sites (< 10%), 2 = Obtained for around 10-40 % of sites, 3 = Obtained for

half of sites (40-60 %), 4 = Obtained for most sites (60-90%), 5 = Obtained for nearly all sites (>90 %)



5 Approach proposed

5.1 Introduction

Based on the detailed review of the different systems in use worldwide, an

approach has been adopted for the project which is simple and robust. In the Indian

context, where there are a very wide range of potentially contaminated sites and

where the population numbers to be considered can be high, reliability over a range

of situations is important.

The approach uses an algorithm (a formula) which links the Source, Pathway and

Receptor elements, using values for each parameter based on actual measurements

from the site, as recorded in the database. The algorithm produces a priority value

for each site, which is compared with the equivalent values for other sites to

provide the overall ranking of all the sites in the database.

The practical construction of the algorithm comprises two parts: the STRUCTURE,

which links the different parameters, and the individual PARAMETERS, which are

based on the values in the database.

Flexibility must be allowed for refinement of the model after the initial period of

use so that additional factors can be included if these are demonstrated to be

important. For this reason, the model has to be conceptually clear and logical.

Overall structure The proposed overall structure for the algorithm, based on review of the other

systems and initial testing of the robustness of the approach, is as follows. This can

be simplified for a first step analysis with limited data.

SCORE = (source term)(modified by pathway factor) PLUS (terms for receptors:

population; groundwater, sensitivity)

This is a simple and flexible structure which is consistent with the models

reviewed. The effectiveness of the model comes from the selection of the specific

terms and of the values to be used for the parameters.



17

Two step approach As described in section 4 the level of information for the individual sites in general

can be divided in 2 groups: 1) Sites with information based on Desk top studies and

b) Sites with information from both Desk top study and Site inspection). After

discussions with MoEF at the meeting in Delhi 16 May 2013, it was agreed that

the prioritization system should include a two-step approach:

› Step 1: Initial Prioritization. This ranking of sites are based on basic

information which typically will be obtained in a desktop study and will cover

all the sites identified in Task 1. The Step I assessment involves hazard

ranking sites based on their historical industrial uses and the receptor’s

sensitivity. The step I assessment can be carried out very rapidly, providing

that source and receptor information is available

› Step 2: Detailed Prioritization: This ranking of sites is based on more detailed

information about the sites including primary data and a site reconnaissance

(e.g. a Site Inspection). This ranking will cover the inspected sites in Task 4,

and also other sites where previous investigations has been performed. The

Step II assessment involves refining the priority listing obtained from step I,

by carrying out a Site Inspection with assessment of pathway and exposure to

determine whether or not a potential pollutant linkage exists. The step II also

include obtaining primary data of contaminants.

5.2 Step 1 Prioritization

Objective The objective of Step 1 prioritization is to establish a prioritization system for

probably contaminated sites about which little is known. The system should be able

to include all sites which have been identified as a “Probably contaminated site”.

The system should prioritize sites based on their potential risk to humans.

Step 1 prioritization applies when the information available consists only of basic

information (or even assumptions) about the site and its surroundings and it does

not require a Site Inspection. It is applied at an early phase to a large number of

sites, in order to identify those which appear to present a serious risk and for which

further data should be obtained. In the current project, Step 1 can be applied for all

sites in the database.

Formula The formula is simplified to reflect the basics of the situation, which are the scale

of the source and the size and sensitivity of the receptor. The Step 1 score could be

determined as:

Step I Priority Score = [Industry; source] + [Land use + Population +

Groundwater + Surface water; receptors]

The first step of the site prioritization is based solely on the types of industrial uses

the site has been subjected to and the sensitivity of the potential receptors. The

issue of pathways is considered in Step 2.

Parameters Proposed parameters for the Step 1 Prioritization is shown in table 5.1.



Table 5.1 Input parameter for Step 1 Prioritization

Para-

meter

S-P-R* Description Basis Range Data source Comment

I Source Industry type Scoring Low,

Med, High risk

1 - 3 Information is obtained in

Task 1 (desktop study) or in

Task 4 (Site Inspection).

List of “Industry types” are

presented in Task 1 report .

The Geoenviron include a

list of 50 Industry types

which can be used initially

L Recepto

r

Land Use Scoring Low,

Med, High Risk

1 - 3 Information is obtained in

Task 1 (desktop study) or in

Task 4 (Site Inspection).

List of land use categories

proposed in Task 1 report

will be used

P Population at risk Estimated within

1 km. Log(pop) or

Low, Med, High

1 - 3 Information can be

obtained in Task 1 (desktop

study) as a simple spatial

query performed in a digital

map (e.g. Google map) or in

a GIS system. Information

can also be obtained in

Task 4 (Site Inspection)

This will be a rough

estimate. Accuracy of

population estimates only

order of magnitude

G Groundwater

system at risk

Use/importance

of aquifer; Low,

Med, High

0 – 3 Information can be

obtained from “Central

Ground Water Board”,

http://cgwb.gov.in/District_

Profile/AP_districtProfiles.h

tml

Classification system must

be performed. Should be

discussed with Central

Ground Water Board.

S Surface water

system at risk

Use/importance

of water source;

Low, Med, High

0 - 3 To be specified but should

be obtained from Central

Government Board

*: Source – Pathway – Receptor

Source The source scale can be taken from a table prepared for common types of Industry

types (e.g. industrial dumps, petrochemical plants, tanneries, chemical

manufacturing etc.), where each Industry type is given a hazard score based on the

contaminants likely to be present on the site. E.g. Tannery operation will give a

high hazard score due to the use of several toxic compounds (e.g. Chromium VI

which is carcinogenic and highly mobile in groundwater and surface water). Also

chemical manufacturing of e.g. herbicides will have a high score due to the highly

toxic compounds. The scoring for each Industry type can be divided in hazard

scores in relation to land use, ground and surface water receptors. For example lead

contamination from “Metal smelting operation” can cause a high risk for humans

(direct contact) but only a low risk for groundwater due to the low mobility of lead

in the subsurface.

Information about Industry types will be available for most sites.

The population at risk and the presence of an important groundwater/surface water

resource can be determined from readily available maps (whether hard copy or

GIS).



19

Pathways Because of the lack of information for many sites the pathway term is not included

in the Step I prioritization.

Receptor The fundamental receptor is the population potentially exposed (P). The current

“Land use” at the site (L) is also included as a parameter although it is our

experience from Task 1 (data collection), that information about land use can be

difficult to obtain in a desktop study. The relevant population has to be identified in

order to characterize P. This will be estimated based on distance (e.g. population

within a certain radius – typically 1 km).

Groundwater (G) and Surface water (S) at risk are included in the receptor

parameters. These parameters are added to the human receptor score. Information

can be obtained from Central Government Board.

Where the data required is not available, implementation of the method can be

based on assumptions. In such cases, we propose to assume a worst case scenario

for each situation. Further information should then be collected in order to verify

assumptions made and further refine the priority listing.

5.3 Step 2 Prioritization

For the sites where a Preliminary Investigation has been carried out and where

more details are provided in the database, a step 2 Prioritization system can be

applied.

Objective The objective is to establish a more robust prioritization system. The system will

cover sites where primary data is available and where a Site Inspection has been

performed. The system should prioritize sites based on their potential risk to

humans and the environment.

The Step 2 assessment involves refining the priority listing obtained from step I, by

carrying out a Site Inspection to obtain primary data and to assess pathway and

exposure routes to determine whether or not a potential pollutant linkage exists.

Prosed scoring formula The proposed algorithm is as follows:

Priority score =

{[Concentration + Quantity + Toxicity + Mobility]*Pathway} +

{Land use + Population + Sensitivity + Groundwater + Surface water +

Ecosystem}

The proposed algorithm is basically additive, except for the pathway factor.

Alternative structures for the algorithm and possible ranges of values for each of

the parameters will be considered during the testing and the refinement of this

prioritization process.

The priority score, as outlined in the box above, can be written as:



Step 2 Priority Score = [C +Q +T +M](F) + [L + P +S +G +S + E]

Parameters The starting parameters and numerical ranges used in the first iteration of

calibration and testing of the model are shown in table 5.2.

Table 5.2: Initial Parameter values for testing system.

Para-

meter

S-P-R * Description Basis Range Data Source

C Source

Source concentration Marks as Low, Med, High or

Ratio to Screening Level

1 - 4 To be obtained from Site Inspection in

Task 4 and/or from previous

investigation performed at the site

Q Quantity of source Volume, or Low, Med, High 1 - 3 To be obtained from desktop study

and/or Site Inspection in Task 4

T Toxicity Factor A list of Chemicals 0 - 2 Can be obtained from CEPI or the

Geoenviron Data Base

M Mobility List – chemical

characteristics

0 - 2 Can be obtained from Geoenviron Data

Base

F

Pathway Pathway factor Conn x Att (see description

below)

1 – 0.01 To be calculated based on Conn and Att

Conn (Containment) Site Report 1 – 0.1 To be specified, Technical judgement

Att (Attenuation) Pathway, tables 1 – 0.1 Reflecting directness of path to receptor.

Can be obtained from US values

L Recep-

tors

Landuse at the site Scoring Low, Med, High

Risk

1 - 3 Information is obtained during Site

Inspection

P Population at risk Estimated within 1km.

Log(pop) or Low, Med, High

1 – 3 Can be obtained from Task 1 (desktop

study) or from Task 4 (Site Inspection)

S

Sensitivity of receptor Guidance to be developed 0 – 1 Based on observation from Site

Inspection e.g. part of the population is

particularly at risk or disadvantaged

G Groundwater system

at risk

Use/importance of aquifer;

Low, Med, High

0 – 3 Classification system must be performed.

Information can be obtained from

“Central Government Boards ” and/or

from Site Inspection

S Surface water at risk Use/importance of surface

water; Low, Med, High

0-3

E Sensitive ecosystems Designated reserves, etc 0 – 3

*: Source – Pathway – Receptor

Source term The source term is a measure of the scale and risk of the critical contaminant. This

is in common with most of the systems reviewed, the approach here uses one

“dominant” contaminant but it is possible to use a combination where there are

high values for several different contaminants.

The basic source term is the concentration (C) of the pollutant at the site in relation

to the relevant screening value. This number can be a simple ratio of actual to

action value or can be a score reflecting the extent of the over-standard. Where

there is a very wide range of high values across sites, the logarithm of the values

can be used. The current uses of the site are taken into account by selecting the

screening value appropriate for that land use. In cases where there is no

information on the concentration (or where the sampling data is very limited or

unreliable) it is possible to use estimates of the typical levels of contamination

using data from similar sites or informed judgments.



21

To characterize the source better, terms are added for the quantity (Q), the toxicity

(T) and the mobility of the contaminant (M). The quantity is scored in terms of an

estimate of size (small, medium, large). This factor can also be used to reflect a

source that is a cluster of industries, where the value of Q is increased to capture

the overall scale of the collective source. This is equivalent to the CEPI factor for

scale of industrial activity. Toxicity and mobility scores are based on specific

characteristics of the contaminant, as recorded in the scientific literature.

Pathway The pathway term is a measure of how direct is the path from the source to the

receptor. In this model, it is structured to be a modifier of the source term.

For sites where the critical receptor is actually on the site (for example people

living on polluted land) then the pathway is direct and the pathway factor (F) is

unity. A land use risk factor (L) is incorporated to reflect a generalized concept of

the pathway and receptor, in cases where there is little information on these.

Where the receptors are off-site then some reduction in the risk will occur, based

on two main factors. First, the existence or provision of containment (or restriction

of access) will reduce exposure to the source. Second, distance from the source

will normally reduce the exposure (simple geometry shows that the concentration

will drop with increasing radius from the site) unless there is a narrow and direct

pathway (such as prevailing winds, a defined groundwater plume or a waterway).

The factor F is therefore a function of (i) containment, and (ii) attenuation and is

assessed on a scale of 1 to zero. Containment can be estimated in terms of the

probable effectiveness of control measures. Attenuation depends on the pathway

and requires judgment to assess but there is some guidance from international

practice, which suggests the appropriate ranges.

Receptor The fundamental receptor is the population potentially exposed (P) and in the

present formulation, additional receptor parameters can be used for groundwater

(G), surface water (W) and for sensitive environments (E). These parameters are

added to the human receptor score. This means that a site which has many impacts

will get a high score, while a site that has serious ecological impacts but limited

human ones (for example) will be given a score to reflect this risk. Plausible

values for these parameters are being tested empirically.

The relevant population has to be identified in order to characterize P. This is

sometimes based on distance (e.g. population within a certain radius – typically 1

km) or may be estimated in terms of pathway assumptions, such as number of well

water users or the population directly downwind.

The present model includes a sensitivity factor (S) which would signify that the

exposed population (or environment) were particularly at risk or disadvantaged.

This could refer, for example, to children being particularly exposed or that

particular population having particularly constrained options to avoid or deal with

the problem.



5.4 Calibration and refinement

The prioritization process is a pragmatic and judgmental one. The next step is

calibration of the proposed form and the main alternatives against a representative

sample of sites, to ensure that consistent results are obtained, and that they are

reviewed and modified in the light of informed expert opinion in India.

The proposed algorithm can be readily tested and revised in a spreadsheet format.

Once the final structure is agreed, it will be integrated into the final database. This

initial application in a spreadsheet will allow the algorithm and alternative

formulations to be readily and transparently applied to a sample of the sites in the

database, as the basis for review, discussion and refinement.

5.5 Testing and review against initial site data

This prototype system will be calibrated against the available site data in the

database. This will test the application of the prioritization and ranking system. A

critical element will be to carry out sensitivity testing to ensure that changes in the

input data do not create unreasonable changes in priorities.

5.6 Application to full database

The prioritization system will be constructed so that it can be readily applied to the

whole database and will allow for any future adjustments easily to be implemented.

A process for interpretation, evaluation and upgrading of the results will also be set

out.



23

6 Plan for completing Task 5

We propose the following activities to complete the prioritization system in Task 5:

› Our proposed approach will be discussed with MoEF and World Bank

› Our approach will be modified based on comments from MoEF and World

Bank.

› The refined model will be tested against representative sites from the database

in a spreadsheet format

› The model will be presented to a representative panel of stakeholders (e.g.

CPCB and some State Boards) and experts and will be refined on the basis of

discussion and feedback.

Once the final structure is agreed, it will be integrated into the GeoEnviron

database.



Appendix A Comparison of Prioritization

Systems



25

Annex A: Comparison of Prioritisation systems

Topic US EPA: Hazard Ranking System

(HRS)

Canada: National Classification System for

Contaminated Sites

New Zealand: Risk Screening System Netherlands: Remediation Urgency Method (in

Dutch “SUS”)

Broad structure HRS addresses risks in terms of potential or

actual release of hazardous substances; waste

characteristics; and targets affected. Produces

score which is used to put sites on National

Priorities List

Assesses hazard of site on basis of limited site

specific data. Calculates Site classification score

based on indicator scores for sources, pathways

and receptors. Sites ranked into four classes,

based on score.

Aims to provide a nationally consistent ranking of

sites that are, or suspected, contaminated using

readily available information. The purpose of

ranking a site is to prioritize for further

investigation. A simplified version of earlier

system.

Introduced in by government in 1994 to prioritise

the remediation of contaminated sites . Based

on human and ecological risk assessment and

risks of contaminant migration. Replaced by new

model “SANSCRIT” available only in Dutch

Where applied Used in US since 1990. Basis for listing of

Superfund sites. Still available for assessment of

new sites.

Developed in 2005 and updated 2008. One of

several tools used to ensure consistency in

implementation of national programme

Released 2001, updated 2007. To prioritize sites

for further investigation. Originally linked to

fund which is now closed.

Regularly used in Netherlands in context of

extensive available technical information and

pathway specific models.

Details of source Uses sample data showing values above human

health benchmark; or estimate using waste

quantity and toxicity; mobility; etc

characteristics. Also release/containment

characteristics.

Scores given for toxicity, area and state of

contaminants. If no data, then may be

“unclassified”.

Score 1 to 10 based on broad types of waste or

source.

Site contamination levels (normally as measured)

compared with generic intervention levels, both

human and ecological. Subsequently develop

site specific standards.

Treatment of

pathways

Distinguishes: groundwater; surface water; soil

or air. Considers physical characteristics and

potential for release

Provides score based on qualitative assessment

of groundwater; surface water and direct

contact. Known or potential score differently.

Addresses Surface water; groundwater; direct

contact. Allows for containment, distance and

other physical factors. Qualitative estimates.

Modeling of different pathways used with

toxicological intervention values to determining

exposure risks. Includes soil, water, air or food.

For environment, consider sensitivity of area

affected.

Receptors For each pathway, distance limit and population

affected are defined.

Scores given for known or potential impact on

humans and animals via each pathway. Also

score for impact on environment.

Scores estimated based on land use. No factor

for number of people etc.

Exposure to risk examined in terms of pathway

used. Dispersion modeling used if specific

targets.

Data

requirements

Site specific investigation data Some site specific data required, else site is not

unclassified.

Site specific data not required Data on actual site contamination levels

Link to No specific link. Relevant data entered for each No link No link Not for risk assessment



geographical data site.

Algorithm Complex and structured “black box“ using typical

US factors. Score for each pathway built up as

product of indicators and factor scores. For

multiple pathways applies Root Mean Square

calculation. Generates HRS score.

Site Classification score determined as sum of

scores for contaminant, exposure and receptor.

Based on hazard-pathway-receptor approach.

Scores multiplicative for eac pathway so any low

component score will eliminate site. Results

reported as High/Med/Low for each of three

pathways

Based on exposure models for human risks,

together with dispersion modelling if relevant.

Screening levels Hazard and toxicity calculated internally from

reference data.

Uses ratio of sample to CCME guideline value.

Scores on order of magnitude basis.

Not used as such Dutch Intervention levels, which are published.

Sensitivity of

outputs

Highly detailed inputs, very specific outputs.

Well demonstrated and reliable.

Generally simple model. Probably quite robust

to qualitative scores for individual elements

Simple model which ranks sites high, medium,

low. Robust rather than sensitive.

The approach is more advanced than a simple

risk assessment and produces detailed analytical

results.

Usability Designed for use in US physical and regulatory

conditions. Can be accessed via EPA website but

not simple to use. Not directly adaptable to

Indian conditions.

Paper and Excel versions available on website. Guidance and spreadsheet available on website.

Uses mainly US health data.

Models are publicly available but in Dutch and

calibrated for Dutch conditions. Approaches are

technical applicable in India but too advanced for

screening or initial risk assessment stage.



27

Annex A – continued

France: Simplified Risk Assessment Geoenviron (System for Prioritisation of Point

Sources)

India CPCB: Comprehensive Environmental

Pollution Index

Worldwide: Blacksmith Index

Broad structure Developed by BROM for use in preliminary

stages, to determine where more detailed

investigations are needed. Based on source,

transfer and target, for groundwater, surface

water and soil.

Developed by Geokon as commercial product.

GIS data used to identify where sources and

receptors overlap. Separate assessment for land,

groundwater and surface water receptors

Algorithm based on Source-Pathway-Receptor.

Draws on information about industry distribution

and also recorded pollutant levels

Uncomplicated Source-Pathway-Receptor model,

using limited data obtained from site surveys.

Where applied Supporting national database Used by local authorities to prioritise sites about

which little is known. Aimed at industrial sites

and landfills.

Used to support prioritisation of needs for

interventions at 88 active industrial clusters

across India

Applied to large database of varied sites

worldwide in order to identify those with highest

priority for further interventions

Details of source Hazard scores (0-3) given for material, quantity. Contaminant toxicity defined as high, medium,

low and given score. Different scores for contact

pathway contaminants and for volatiles.

Scores given for toxicity of key pollutant(s)

multiplied by scale of industrial activity

Uses ratio of composite sample data to

Screening Level. Additional score for

toxicity/persistence

Treatment of

pathways

Scores given for mobility, state, hydrology,

containment, etc. and transfer potential

Scores given for different pathway/receptor

combinations. Direct contact and inhalation

score for each are additive

Scored on reported ambient pollution and

evidence of impacts on people and environment

Dissipation factors applied to pathways: soil,

groundwater, surface water

Receptors Scores given for accessibility, population, water

system, water use,

Included in pathway/receptor analysis; can be

adjusted for “special conditions”

Numbers of potentially affected people within

2km, with additional score for level of exposure

and sensitivity. Additional score for (lack of)

control and containment

Number of people potentially affected by each

pathway. No ecological component.

Data

Requirements

Basic information on contamination levels Stage 1 can used generic industrial figures. Stage

2 applies site specific data

Uses monitored data for the industrial clusters

assessed

Based on (limited) actual site sampling data

Link to

geographical data

Not linked Allows link to GIS and identification of spatial

relationships between sources and potential

receptors

No specific link No specific link (although each site geo-

referenced)

Algorithm Weighted sum of scores determined for three

groundwater cases; three surface water and for

soil. Scores used to assign sites to 3 classes.

Contaminant and pathway/receptor score

additive for each of the three receptors

considered.

Total score is additive with highest (worst) value

of 100

Log of source ratio and log of population used in

additive overall Index value



Screening levels French levels, with German and Dutch as fall-

back

Uses appropriate soil quality standard values,

together with volatility data, where appropriate

Relevant Indian Standards Based on authoritative international systems

(WHO, USEPA, etc)

Sensitivity of

outputs

Straightforward approach, based on expert

defined scores and weights

Broadly robust as individual scores not sensitive

to small changes.

Quite robust to changes in individual factor

scores

Use of logs of key parameters makes system

robust to uncertainties in values

Usability Summary hard copy version available in English.

Scores and weights developed for French

conditions. Not clear how to translate to India.

Overall applicable in India. Would have to

ensure values used are consistent with local

regulations. Marking system could be tested for

robustness to changed assumptions.

Applicable in principle but uses discharge and

contamination data that may not be readily

available

Readily usable or adaptable



29

Appendix B Parameters in Prioritization

Systems



Annex B: Parameters in prioritisation systems

CANADA NEW ZEALAND

Surface water

NEW ZEALAND

Groundwater

NEW ZEALAND Direct

Contact

GEOKON Blacksmith CEPI (India)

SOURCE

Residency media Toxicity Toxicity Toxicity Industry type Concentration Contaminants

Hazard Quantity Quantity Quantity Chemical compound Persistence Scale

Exceedance Mobility Mobility Mobility Mobility

Quantity Toxicity

Modifiers Volatility

Degradation, aerobic

Evidence of impact

PATHWAY

G/w movement Containment Containment Containment Exposure to buildings Attenuation Concentration

Surface water Cover Cover Depth People impacts

Soil Soil permeability Drinking water intake Ecological impacts

Vapour Depth Aquifer protection

Sediment Distance, surface water

Modifiers

RECEPTOR

Land use Runoff Aquifer type Land use Landuse (current) Population Population

Accessibility Aquifer use Exceedance

Direct contact Quality goal, surface water Sensitivity

Inhalation Special risk element

Ingestion

Terrestrial

Aquatic

Modifiers

Other receptors



31

Appendix C Methodologies examined

The European Environment Agency prepared a listing in 2004 of different

methodologies that were identified worldwide for “preliminary risk assessment” of

soil contamination. The 27 sites listed (see table) were reviewed as candidates for

this project and 8 of these were then taken as possible models for this project.



Table: Systems identified by EEA

Acronym

Name Year Country

1 A.G.A.P.E A.G.A.P.E 1988 Germany

2 A.H.R.M Alaska Hazard Ranking Method 2003 USA

3 A.R.G.I.A. Analysis of Risk for the Prioritisation of

polluted sites of the register

2003

Italy (Emilia Romagna)

4 BWM Baden-Wurttemberg 1988 Germany (Baden-

Wurttemberg)

5 C.S.S.M. Contaminated Sites Screening Model 1993 Italy (Sicilia)

6 D.R.E.A.M. Dundee Risk Evaluator Assessment

Model

2001 Scotland

7 G.T.K. Geologian Tutkimus Keskus 2001 Finland

8 H.R.S. Hazard Ranking System 1990 USA

9 I.S.M. Indiana Scoring Model 1989 USA (Indiana)

10 L.R. Lombardia Risorse 1991 Italy (Lombardia)

11 M.I.F.O. Method for the Inventory of

Contaminated Sites

2002 Sweden

12 N.C.S.C.S. National Classification System for

Contaminated Sites

1992

Canada

13 N.R.S. Numerical Ranking System 2004 USA (Massachusetts)

14 R.P. Metodo Regione Piemonte 2000 Italy (Piemonte)

15 R.R.S.E. Relative Risk Site Evaluation 1996 USA

16 R.R.S.M. Receptor Source Proximity Relative Risk-

Screening Model

2001

England (Eden)

17 R.A.S.C.L. Risk Assessment for Small Closed

Landfills

2002 New Zealand

18 RI.SI.CO. RIsk of COntaminated SItes 2001 Italy

19 R.S.S. Risk Screening System for contaminated

sites

2003 New Zealand

20 R.U.M. Remediation Urgency Method 1995 Netherlands

21 S.R.A. Simplified Risk Assessment 2001 French

22 S.A.M. Site Assessment Model 1990 USA (Michigan)

23 S.A.P.S. Site Assessment Prioritisation System 2000 USA (Oregon)

24 S.P.C. Site Prioritisation Criteria USA (Colorado)

25 S.P. Snamprogetti 1990 Italy (Piemonte)

26 S.P.P.S. System for the Prioritisation of Point

Sources

2003 Denmark

27 W.A.R.M. Washington Ranking Methods 1992 USA (Washington)

Reference: “Towards an EEA Europe-wide assessment of areas under risk for soil

contamination – Volume II: Review and analysis of existing methodologies for preliminary

risk assessment.” Final Version. December 2004. European Environment Agency,

Copenhagen, Denmark.