Defra - Department for Environment, Food and Rural Affairs.

Science Search

Science and Research Projects

Return to Science Search homepage   Return to Project List

Scientific approaches to support the use of WEBFRAM probabilistic risk estimates in decision making - PS2331

Description
The WEBFRAM projects (funded by Defra) have delivered a suite of probabilistic models for refined (higher tier) assessment of pesticide risks to birds, mammals and aquatic organisms in UK scenarios (www.webfram.com). In comparison with first tier assessments in current use, they provide a richer description of the potential risks to wildlife, offering quantitative estimates of how bad the consequences could be, how often they could occur and how sure we can be about our predictions.

Two of the key issues emerging from the peer review of these models, and from stakeholder meetings, are the reliability of probabilistic risk estimates and how they would be used in decision-making. The Pesticides Safety Directorate (PSD) will consider these issues as part of a wider evaluation of the WEBFRAM tools during the period up to September 2007, including consultation with the Environmental Panel in June and the Advisory Committee on Pesticides (ACP) in September. This project will explore some scientific approaches to assist PSD in addressing these issues.


Objective 1. Develop methods for comparing risks predicted from probabilistic modelling with established criteria for acceptable risk in first-tier assessments.

Probabilistic approaches produce a different type of risk estimate than current first tier procedures, e.g. an estimate of the level of mortality instead of a toxicity-exposure ratio. Decision criteria to determine whether risk is acceptable exist already for toxicity-exposure ratios, and are established in legislation. Currently there are no established decision-criteria for use with probabilistic risk estimates. This project will develop methods for expressing the existing decision criteria in probabilistic terms, so that they can be applied to probabilistic risk estimates. This would provide regulators with a way of deciding whether the refinements incorporated in a probabilistic assessment are sufficient to meet the established criteria for acceptable risk, or whether further refinement or evaluation is required.
Objective 2. Evaluate the reliability of higher tier probabilistic predictions of wildlife mortality by comparing them with data on actual impacts in the field.

To provide a sound basis for decision-making, probabilistic risk estimates should be reliable predictors of effects in the field. Data on effects in the field are available for birds and mammals. This project will use these data to carry out extensive comparisons between probabilistic risk estimates and appropriate measures of impact in the field, taking account of the uncertainties on both sides. The results will enable PSD and others to evaluate the realism of the WEBFRAM models and also, in effect, calibrate the probabilistic predictions against real-world impacts. If successful, this will enable confident use of WEBFRAM outputs in decision-making, especially in cases where the probabilistic risk estimates exceed the level associated with first tier decision criteria (based on the approach of objective 1).


Thus the results from these two objectives will provide complementary support for PSD to develop rational and defensible strategies for using outputs from the WEBFRAM models in decision-making.
Objective
Objective 1. Develop and evaluate methods for comparing WEBFRAM results to the level of probabilistic risk implied by current first-tier decision criteria.



Objective 2. Evaluate probabilistic risk estimates for birds and mammals by comparison with data on mortality in field studies.
Project Documents
• Final Report : Scientific approaches to support the use of WEBFRAM   (736k)
Time-Scale and Cost
From: 2007

To: 2007

Cost: £91,938
Contractor / Funded Organisations
Central Science Laboratory
Keywords
Animal              
Birds              
Decision Making              
Environment              
Environmental Effects              
Indicators              
Modelling              
Pest and Weed Control              
Pesticide use              
Pesticides              
Plants and Animals              
Risk assessment and management              
Water              
Fields of Study
Pesticide Safety