Introduction to the updated CLEA framework reports Ian Martin, Land - - PowerPoint PPT Presentation

Introduction to the updated CLEA framework reports

Ian Martin, Land Contamination Policy Advisor

Next 60 minutes

• Background
• TOX guidance report (updated CLR9)
• CLEA report (updated CLR10)
• Focus on providing an overview and key changes

Land contamination in England

• Land may be contaminated by previous use, by diffuse

pollution, or by its natural state

• Agency (2005) estimated that as many as 325000 sites

may be ‘potentially affected’

• Managing land contamination:

4 Planning regime 4 Part 2A 4 Other such as ‘due diligence’

Estimating risk to human health

• What level of exposure constitutes a minimal or tolerable

risk to health?

• How likely is a person living or working on land

contamination to be exposed to chemicals from soil?

Adopting a tiered approach

Problem Formulation Risk Prioritisation Hazard Identification Exposure Assessment Risk Estimation Risk Characterisation * Stages within each tier of Risk Assessment Economics Technology Social Issues Management Risk Management Collect data, iterate processes & monitor Tier 1 Risk Screening * Tier 2 Generic Quantitative Risk Assessment * Tier 3 Detailed Quantitative Risk Assessment * Options Appraisal

CLEA project objectives

• To develop tools that provide a Government supported

methodology that help estimate chronic health risks to people from soil contamination

• To provide generic assessment levels of contamination

in soil below which these risks are considered minimal

• To provide a starting point to help assess risks, for

instance under Part 2A of the Environmental Protection Act 1990

CLEA project and tools (caveats)

h No legal requirement to use in assessing land

contamination

h Do not cover other types of risk to humans, such as fire,

suffocation or explosion, drinking water, or short-term and acute exposures

h Do not cover risks to the environment, such as

groundwater, ecosystems or buildings

h Do not provide a definitive test for telling when human

health risks are significant

CLEA project outputs

• CLEA framework reports

Human health toxicological assessment of

contaminants in soil (August 2008)

Updated technical background to the CLEA model

(August 2008)

CLEA software version 1.03 beta and handbook

• TOX reports and SGV technical notes (in preparation)
• Supporting work

Compilation of chemical data and estimation

methods (in preparation)

Vapour Handbook (in collaboration with CIRIA)

TOX guidance report

• Sets out toxicological basis and approaches to deriving

health criteria values

• Updates and replaces CLR9

Comprehensively re-written to improve clarity Expanded explanation on reviewing tox data and

guidance on multiple exposures / dealing with mixtures

Supported by Defra, HPA and FSA

• Defines health criteria values as levels of minimal or

tolerable risk to health from long-term human exposure to chemicals in soil

Deriving health based guidance values

1000 100

Dose

10 1 0.1

Response

Toxic effect LOAEL NOAEL Control UF Sensitive man Average man UF Test species

Framework to derive HCV

• Collection of data
• Evaluation of data
• Reporting of data

Physical-chemical characteristics Toxicokinetics Toxicity Background intakes

• Differences between threshold / non-threshold toxicity

Threshold and non-threshold toxicity

Response Dose A B

A: Toxicant has no

• threshold. There is some

risk at any level of exposure.

B: Toxicant has a

• threshold. There is a

finite dose below which adverse effects are not discernible.

Threshold HCV

h Tolerable daily intake (TDI) is an estimate of the daily intake of a

chemical that can be experienced over a lifetime without appreciable health risk

h TDI equals point of departure (e.g. NOAEL) divided by uncertainty factor

and where possible derived for each exposure route

h Existing HCVs should not be adopted naively (there is no hierarchy of

sources or simple pick list)

h TDSI is the TDI minus non-soil background exposure MDI is mean daily intake by adults from non-soil sources Child MDI corrected from adult MDI Minimum of 50 per cent from soil sources If no data or information on background exposure are available,

assume background is negligible unless qualitative evidence suggests contrary

Non-threshold HCV

h Index Dose (ID) is the estimate of the daily intake of a chemical

that can be experienced over a lifetime with minimal cancer risk (ALARP principle applies)

h Main derivation methods Apply large UF to Bench Mark Dose (BMDL10) derived from

animal study (Defra, UF = 10000)

Quantitative dose-response modelling of human cancer data

(Defra, 1 in 100 000 excess lifetime risk)

h Threshold effects must also be considered and may drive risk h Comparison with other UK guidelines, it may be disproportionate

to enforce a stricter limit for contaminated land than other media

h Exceeding an HCV will result in increased risk to health

CLEA report

h Replaces CLR10 and provides improved clarity, internal

consistency and practical usability

h Consolidates previous briefing notes and other

guidance into a single report

h Reconsiders the generic land-use scenarios and default

assumptions used in the CLEA model to derive SGV

h Updates the data sets and model algorithms used in the

CLEA model based on recent scientific literature

CLEA Model

FATE AND TRANSPORT OF CONTAMINANTS IN SOIL ENVIRONMENT

Understanding and predicting how contaminants move within each environmental compartment - air, water, and soil

HUMAN EXPOSURE TO CONTAMINANTS IN THE SOIL ENVIRONMENT

Predicting how we behave on a contaminated site according to land-use and the extent to which this may lead to exposure to soil contamination for critical receptors (e.g. small children)

EVALUATION OF THE RISK TO HUMAN HEALTH ARISING FROM EXPOSURE

Judging the potential risks to health from exposure to contaminants in soil

Contaminated Land Exposure Assessment model

Exposure pathways

Forwards and backwards

Define conceptual model Collect exposure characteristics and chemical fate and transport data Determine contaminant concentration in soil from review of investigation data Estimate exposure

Mouth Nose Skin

Compare with health criteria value (TDSI or Index Dose) Decision made after comparison of exposure and health criteria

FORWARD

Define conceptual model Collect exposure characteristics and chemical fate and transport data Vary soil concentration until decision criteria are met Estimate exposure

Mouth Nose Skin

Compare with health criteria value (TDSI or Index Dose) Repeat until exposure equals health criteria value

BACKWARD

Managing the uncertainty

h Types of uncertainty include parameter, model and

scenario uncertainty plus variability

h Deterministic and stochastic approaches

Simple generic screening uses deterministic Detailed assessment may benefit from stochastic

h Moving away from maximal exposed individuals to

reasonable or more realistic worst-case

Identify common activities or pathways likely to result in

exposure and use reasonable worst-case estimates

Identify an unusual (but not unlikely) worst-case activity and

use typical case assumptions

Generic land-uses – key changes

h Residential

No residential with/without plant uptake Homegrown produce based on general population Default building type – small terraced house

h Allotment

Children are still critical receptor Homegrown produce based on high-end grower No tracked back component

h Commercial

Default building type – three storey office (pre-1970)

How much space do you need?

Category Annual Household Consumption (kg fw) Typical Yields (kg fw m-2) Required Area (m2) Green vegetables 16.8 2.8 7.7 Root vegetables 5.7 4.7 1.7 Tuber vegetables 3.0 4.4 0.7 Herbaceous fruit 8.2 5.1 4.0 Shrub fruit 1.5 0.8 1.9 Tree fruit 7.8 1.8 3.9 19.9 h Growing area required for average population consumption rates

used in the CLEA model for a family of four is:

h 85 per cent of residential gardens greater than 100 m2 and 34 per

cent greater than 450 m2 (MAFF 1999)

Generic data sets – extensively revised

h Chemical

Simplified (although property adjustments still required)

h Soil

Eight soil types defined from UK soil LANDIS database

h Human characteristics

Revised height and weight data from Health Survey 2003 Revised skin area and inhalation rates based on Exposure

Factor Handbook

Revised produce consumption rates in line with produce

categories used by FSA’s PRISM model

h Building

Nine building types based on BRE review

Soil ingestion and dermal contact

h Soil ingestion no longer probabilistic distribution

Daily intake for children 100 mg day-1 Daily intake for adults 50 mg day-1 Preparation factor added to estimate soil attached to produce

h Dermal contact

Briefing Note 1 incorporated US EPA approach based on empirical absorption factor

Consumption of produce

h Produce categories revised and expanded h Generic models are a back-stop to be used only in

absence of real data

Inorganic chemicals based on PRISM Organic chemicals based on suite following review of generic

model performance

Comparison of generic / specific models

• 2.0
• 1.5
• 1.0
• 0.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0 1 2 3 4 5 6 7 log Kow log CR T + M Trapp Comparison of predicted versus observed uptake of various chemicals by carrot using Trapp and Matthies (1995) and Trapp (2002). Log CR is the ratio

• f observed to predicted plant concentrations.

Inhalation of dust

h Revised air dispersion factors provided by Agency’s Air

Quality Modelling and Assessment Unit much higher than original values

h Cowherd et al. erosion parameters tweaked for UK

climate characteristics

h Indoor dust loading include a fixed factor plus an

ambient contribution to reflect role as a reservoir

Inhalation of vapour

h Indoor vapour intrusion guidance includes revised

caveats on use and interpretation

h Large uncertainties and contentious technical area h Further work supported by CIRIA Vapour Handbook h Johnson and Ettinger approach as described in Briefing

Note 2 largely adopted

h Outdoor vapour model uses consistent dispersion

parameters with dust model

Changes and SGV

h SGV based on minimal risk h Revised approach to deriving HCV Higher SGV - threshold toxicity and high non-soil

background exposure

Variable - non-threshold toxicity h Revised approach to exposure assessment Higher SGV - soil ingestion and outdoor inhalation Lower SGV - dermal contact and indoor inhalation Variable - consumption of homegrown produce h Actually very difficult to predict (and don’t take into account

chemical-specific issues)

CLEA software 1.03 beta

h Includes approach

and parameters from the CLEA report

h Includes additional

features such as limited source model / bioaccessibility

h Highly customisable

We want feedback

h Framework reports and software until end of

November 2008

h How readable and user friendly are these reports? h Have you found problems or technical inaccuracies

with the way that the CLEA software performs during site-specific assessment?

h How easy is the software to use? h cleacomment@environment-agency.gov.uk