IMIS | Flanders Marine Institute
 

Flanders Marine Institute

Platform for marine research

IMIS

Publications | Institutes | Persons | Datasets | Projects | Maps
[ report an error in this record ]basket (0): add | show Printer-friendly version

Evaluation of the European Fish Index: false-positive and false-negative error rate to detect disturbance and consistency with alternative fish indices
Quataert, P.; Breine, J.; Simoens, I. (2007). Evaluation of the European Fish Index: false-positive and false-negative error rate to detect disturbance and consistency with alternative fish indices. Fish. Manage. Ecol. 14(6): 465-472. dx.doi.org/10.1111/j.1365-2400.2007.00573.x
In: Fisheries Management and Ecology. Blackwel Science Ltd.: Oxford. ISSN 0969-997X, more
Peer reviewed article  

Available in Authors 

Keywords
    Errors; False negative results; False positive results; Fish; Indexes
Author keywords
    error curve; false-negative error; false-positive error; fish index; precision

Authors  Top 

Abstract
    An important requirement for meeting obligations under the European Water Framework Directive is the development of a fish-based index that is able to predict the ecological status of surface waters, and particularly be able to distinguish between (nearly) pristine and disturbed conditions. The European Fish Index (EFI), based on the concept of the Index of Biological Integrity, was developed alongside alternative models such as the Spatially Based Method on a European level (SBM-EU), for this purpose. A critical issue about these models is that they are simple to use but are able to predict whether a site is disturbed with a high degree of precision. From this perspective, two prediction errors need to be small: falsely declaring a site disturbed when it is not (falsepositive error; FP) and wrongly classifying a disturbed site as undisturbed (false-negative error, FN). For the EFI, the overall FP rate was 22% and the FN rate was 19%. The performance was better for the SBM-EU method with a smaller FP rate of 7% and an FN rate of 20%, but the EFI is preferred because, with only marginal loss of precision, it is far less complex. The EFI consists of a single model based on 10 fish metrics, while the SBM-EU comprises 12 models covering 49 metrics. Comparison of the EFI with existing national or regional fish-based assessment methods found major discrepancies that make intercalibration between these methods impractical.

All data in IMIS is subject to the VLIZ privacy policy Top | Authors