Ecological impact of urban stormwater runoff studied in experimental flumes: Population loss by drift and availability of refugial space

Urban stormwater runoff discharged through sewer systems into streams causes flush spills of water and pollutants in the receiving water. To make the right decisions in future plannings of the very costly rehabilitation of sewer systems, a solid ecological data base on the critical parameters of sewer overflows is badly needed. Therefore, we designed a laboratory flume which was operated in circular flow mode (to ensure adaptation of the test organisms) and in flow-through mode during the simulation of sewer overflows (to allow a proper evaluation of population loss by drift). Examples on the behaviour during the adaptation phase and the population loss during the exposure to flush spills of water and/or a mixture of sewage and clean water of a benthic invertebrate (Gammarus pulex) demonstrate the potential of the flume to identify critical parameters of sewer overflows at quasireal-world-conditions. We found clear evidence for synergetic effects since the exposure to high flow and sewage caused higher population loss ofGammarus than the sum of population loss at exposure to only high flow or only sewage. Population loss considerably depended on the availability of refugial space: if the interstices of the gravel in the flume were silted, this loss was higher than at open interstices. Only ten minutes of movement of the material forming the flume bottom reduced the population ofGammarus to about 60 or 50% of its initial size. Hence, our data strongly suggest that the characteristics of the receiving stream (refugial space, bed stability) play an important role for the potential ecological impact of a sewer overflow. Changes of stream morphology and/or creation of refugial space plus an appropriate technical solution for overflow treatment may be less costly and more effective than a large-scale technical project. Thus, the stream itself should be a major element in future management decisions.