A risk model for predicting the effects of lake-side development on wildfowl populations

✍ Scribed by Graham J. Floater

• Publisher: Elsevier Science
• Year: 2002
• Tongue: English
• Weight: 169 KB
• Volume: 66
• Category: Article
• ISSN: 0301-4797
• DOI: 10.1006/jema.2002.0586

✦ Synopsis

The lack of quantitative analysis and general scientific rigour in environmental impact assessment (EIA) is well documented. While reasons for this may include political and economic factors, the lack of high-level statistical knowledge and skills in environmental consultancies probably contributes to the problem, particularly with regard to ecological studies. This paper develops a simple model for wintering wildfowl populations that predicts different levels of risk from lake-side development. The aim was to create a model that could be used easily and quickly by consultants, is readily understandable for developers and various groups associated with the planning process, with explicit assumptions that can be criticised, and predictions that can be tested with post-development audits. The model is used in a case study. The basic parameters of the model are (i) average wildfowl abundance on lakes before development, (ii) maximum potential density of wildfowl across lakes (Kmax) before development, and (iii) reduction in lake area suitable for wildfowl after development. The model includes abundance-area relationships that are useful for highlighting the importance of particular lakes at a site. In the case study, abundance-area relationships focused attention on two lakes with thick charophyte beds which supported higher than expected numbers of pochard and coot given their size. As well as being robust (confidence limits are calculated), the model's predictions are quantitative and testable, making it possible to compare the predictions with on-going post-development monitoring of wildfowl population levels. The predictions rely on the effectiveness of path screening, and post-development monitoring can suggest where screening should be strengthened if the model's predictions are not met. Similarly, if other assumptions in the model are not met by the development, appropriate action can be implemented.