A comparison of phytoplankton assemblages generated by two sampling protocols in a German lowland catchment

Department of Hydrology and Water Resources Management, Institute for the Conservation of Natural Resources, Kiel University, 24118 Kiel, Germany

^* Corresponding author: nwu@hydrology.uni-kiel.de; wunaicheng2003@yahoo.com.cn

Received: 7 December 2010
Accepted: 17 June 2011

Abstract

Research in the phytoplankton community has become an important part of the overall water-quality monitoring. However, to date studies in small rivers and streams were still scarce and sampling methods were also diverse and not as well developed as in lakes or large rivers. We investigated whether two sampling protocols collected different phytoplankton assemblages within a lowland catchment and, consequently, influenced the outcome of bio-assessment. Data collected from 77 sites by plankton net (PLNET) collection and sedimentation (SEDIM) protocols were analyzed. Median Bray–Curtis (BC) similarity between phytoplankton assemblages generated by the two protocols was 48.5% (range: 7.5–82.0%), and sites with the lowest BC similarities tended to have lower chlorophyll a (Chl a), water temperature (WT), total suspended solid (TSS) and volatile suspended solid (VSS), but higher channel width and water depth, than other sites with higher BC similarities. Reduced total algal density and biomass, but higher species richness, were observed by the PLNET protocol. However, overall phytoplankton assemblages generated by the two protocols were similar, as indicated by dominant species (paired t-test, P>0.05) and non-metric multidimensional scaling (NMDS) ordination. Nevertheless, from the phytoplankton-based bio-assessment point of view, PLNET protocol was a better method compared with SEDIM protocol because algal data collected by PLNET protocol had higher relationship with environmental variables as indicated by ‘Correlation Index’ (CoI), Cumulative_R² and canonical correspondence analysis (CCA).