Responses of aquatic organisms to metal pollution in a lowland river in Flanders: a comparison of diatoms and macroinvertebrates.

The role of macroinvertebrates and diatoms as indicator for metal pollution was investigated by assessing both biota along a metal gradient in the Belgian river the Dommel. Macroinvertebrates and diatoms were sampled in summer and winter and physical-chemical characteristics of the water were measured at four different sample periods and related to sediment characteristics. Although metal concentrations, except cadmium, in the water nowhere exceeded water quality standards, high metal concentrations were measured in the sediment, indicating historical contamination of the Dommel. At the sites that were situated downstream of the pollution source, high levels of conductivity and chloride were measured in the water. Redundancy Analysis (RDA) indicated pH, phosphate and zinc as the significant environmental variables explaining each respectively 7.7%, 11.6% and 22.6% of the macroinvertebrate community composition. Two clusters could be separated, with Gammarus pulex, Leptocerus interruptus, Baetis rhodani and Cloeon dipterum associated with low zinc concentrations and Tubificidae, Asellus aquaticus, Erpobdella sp. and Chironomus thummi-plumosus associated with higher zinc concentrations. Ammonium (10.6%), conductivity (16.5%), chloride (11.4%) and zinc (5.9%) turned out to be significant variables explaining the diatom community structure. Based on physical-chemical differences and species composition, three different groups could be separated. With this Tabellaria flocculosa and Fragilaria capucina var. rumpens were associated with low metal concentrations, Gomphonema parvulum and Nitzschia palea with elevated concentrations and Eolimna minima and Sellaphora seminulum with high zinc concentrations. In conclusion, the diatom community best reflected the metal gradient. With regard to water quality indices, those based on macroinvertebrates best followed the metal pollution gradient and were most strongly correlated with physical-chemical variables of water and sediment. This study indicated that to assess the effect of metal pollution in lowland rivers, the combined use of macroinvertebrates and diatoms is more appropriate than the use of both biota separately.