| Title: |
Comparison of the European Fish Index with the Standardised European Model, the Spatially Based Models (eco-regional and European), and Existing Methods.
|
| Resource Type: |
document --> technical publication --> report
|
| Country: |
EU Projects
|
| Year: |
2004 |
| Availability: |
Quataert, P., Breine, J. and Simoens, I. (2004) Comparison of the European Fish Index with the Standardised European Model, the Spatially Based Models (eco-regional and European), and Existing Methods. Institute for Forestry and Game Management, Groenendaal-Hoeilaart, Belgium.
First author: Paul Quataert
Other authors: Jan Breine and Ilse Simoens
Year: 2004
Title: Comparison of the European Fish Index with the Standardised European Model, the Spatially Based Models (eco-regional and European), and Existing Methods.
Source details: A Contribution to the Water Framework Directive by the Fame Consortium: http://www.boku.ac.at/fame/downloads/D16_17_MethodComparison.pdf
Publisher name: Fame Consortium
Publisher details: Belgium
|
| Author 1/Producer: |
Fame Consortium
|
| Author / Producer Type: |
EC Project
|
| Format (e.g. PDF): |
PDF
|
|
EUGRIS Keyword(s): |
Groundwater protection-->Groundwater processes-->Ecotoxicology
|
| Long description: |
When developing a decision tool as a fish index it is crucial to document its performance in
relation to its goals. An important requirement of the WFD is that the newly developed
European Fish Index (EFI) can distinguish between a (nearly) pristine and disturbed status.
Also the position of EFI with respect to existing national or regional fish indexes should be
clear as some of them offer a long time series. Finally, comparison of the EFI with respect to
other possible approaches gives insight in its relative merits and shortcomings.
To realize this evaluation, the central idea of this chapter is to think the fish index as a
laboratory test to detect whether or not a site is disturbed. This analogy allows expressing the
performance in terms of the detection capacity (sensitivity and specificity, impacted and nonimpacted
predictive value) and consistency. The cumulative distribution function of the new
metric conditional on other indices turned out to be a powerful graphical tool to detect
anomalies and to gain insight on a more profound level. These cumulative curves can be
estimated easily from the data by the empirical distribution function.
|
|
Submitted By:
|
Dr Stefan Gödeke WhoDoesWhat?
Last update: 20/07/2009
|
|
|
