BES Facebook pageBES Annual Meeting 2010
7 - 9 September
University of Leeds, UK
Managing modelling and managing model managers
Wednesday 8 September
Organiser: Graham Smith, Food and Environment Research Agency
The BES has for many years taken a lead in linking ecological research to government policy, and also promotes the advancement of applied modelling. The recent launch of the Policy Newsletter aims to inform members of relevant details on environmental policy changes and consultations. Meanwhile, the government has promoted the importance of peer-reviewed evidence to inform policy decisions. Many governmental and non-governmental bodies are now more frequently turning to modelling to investigate scenarios and ask "what-if" questions.
Defra's Science Advisory Council has recently established a sub-group on "Modelling, uncertainty and prediction" to examine the way modelling is done and used across the department. Defra Agencies, such as Fera, have always supplied evidence and advice to government, and have had an applied focus to all their research. As such, the experience of staff in these agencies can be utilised to inform and educate both the Defra customer, and the contractor, on how to improve the level of interaction in these areas. I have been involved for many years in supplying modelling evidence and advice to Defra, and have organised and run workshops in this area previously. I am running one for Defra economists and policy officials in June.
The overall objective is to improve the way that individuals approach modelling projects, both for commissioning and contracting, and to avoid the final pitfall that the customer finds the model does not answer their requirements. The individual objectives of the session are:
- To define what a model is, and what it can and cannot do
- To introduce the process of model commissioning and defining the question to be answered
- To give experience in producing a simple model and providing output (all on paper)
- To give expectations on what a report on modelling needs to contain (i.e. defining model structure, assumptions, verification, validation, sensitivity analysis and uncertainty)
- To provide information on ways to improve model commissioning and model reporting
- To introduce approaches to how competing models may be considered.
