Phytoplankton, macrophytes and benthic diatoms in lake classification: Consistent, congruent, redundant? Lessons learnt from WFD-compliant monitoring in Poland

The lake monitoring programme compliant with the Water Framework Directive has been implemented in Poland since 2007. Currently, the methods for three biological quality elements (BQEs): phytoplankton (the Phytoplankton Multimetric for Polish Lakes, PMPL), macrophytes (the Ecological State Macrophyte Index, ESMI) and phytobenthos (the Diatom Index for Lakes, IOJ) are officially applied and internationally intercalibrated. Based on the monitoring data from 256 lakes surveyed in 2010-2013 and assessed for all the three BQEs, we tested whether the assessment results obtained by the three biological methods were consistent and we searched for the causes of inconsistencies which we found. The lake classifications obtained from the PMPL and ESMI were highly consistent and the relationship between these metrics was relatively strong (R = 0.66, p < 0.001). Both metrics correlated equally strongly with water quality indicators. However, the PMPL and ESMI indicated systematic dissimilarities in the sensitivity to eutrophication between shallow and deep lakes. In shallow lakes, the alarming symptoms of macrophyte community deterioration (lower values of ESMI) occurred at lower nutrient and Chla concentrations and were accompanied by a better status of phytoplankton (higher values of PMPL) than in deep lakes that can be explained by a synergistic effect of inorganic suspended solids and algal growth on water transparency. As a consequence, the positions of phytoplankton and macrophytes as early warning indicators in the eutrophication gradient in shallow lakes were inverted compared to those in deep lakes. Compared to the PMPL and ESMI, the IOJ method gave the least stringent assessment results, with 22% of lakes failing to meet the environmental objectives. The relationships between IOJ and PMPL, and ESMI were relatively weak (R = 0.17, p = 0.008 and R = 0.17, p = 0.007, respectively). Moreover, the phytobenthos index IOJ correlated significantly more weakly with all the water quality indicators than either PMPL or ESMI did. The poor performance of the phytobenthos method in this study may suggest a limited indicator value of this BQE for lake assessment or inappropriate sampling design.