Deliverable D4.4-2, part 1: Testing the adaptability and behaviour of 6 fish indices on a common dataset composed of multiple gears samplings from 8 estuaries and lagoons

The Water Framework Directive (WFD) aims at achieving good ecological status for surface waterbodies throughout Europe, by 2015. Consequently European countries are currently developing and intercalibrating methods based on biological, hydromorphological and physicochemical quality elements for the assessment and the monitoring of their rivers, lakes, coastal and transitional waters. In this context, the FP7 WISER project aims to support the implementation of the WFD by contributing (i) to making the existing assessment methods more comparable, and (ii) to estimating the uncertainty along each step of the assessment. The present work focuses on fish indicators for estuaries and lagoons (transitional waters in the WFD). Six fish indices (AFI, EFAI, ELFI, TFCI, BHI, Z-EBI) were tested on a common dataset, covering eight estuaries and lagoons throughout Europe. Fish sampling was carried out using several gears in 2009 and 2010. The objectives were twofold: (i) to test the adaptability of fish indices to different gears and different types of transitional waters; and (ii) to compare the behaviour of the indices with regard to their level of agreement. Five out of the six tested indices were gear-specific and all were specific to some type(s) of transitional waters and have particular data needs. Therefore calculating the indices on a common dataset raised many difficulties, especially regarding the determination of appropriate reference condition values. However, taking some reasonable assumptions it was possible to calculate most of the indices on the majority of the available data, thus showing their relative adaptability. The indication of extreme values for quality (high or bad) was relatively rare and the results of the indices often differed: when used outside of their initial framework, geographical limits or with different sampling methods, fish indices’ results are highly uncertain. This shows that the assessment of quality using fish indicators highly depends on the assessment tool, the sampling methodology and the type of transitional water. Despite these results, some statistically significant correlations between pairs of indices’ results were found, indicating the possibility of intercalibration between some of the tested indices. This work corresponds to the first step of WISER WP4.4 Deliverable 2. In a second step, some fish metrics in the indices tested here will be selected for a more in depth study: the effect of several natural sources of variability on these metrics will be studied and uncertainty will be quantified along the assessment process, from the sampling to final EQS formulation using WiserBugs software..