Ursula Gaedke

• Large (472 km2) and deep (zmean=101 m) Lake Constance is undergoing re-oligotrophication. Total phosphorus during winter mixing (TPmix) decreased from >80 during 1975-1981 to 22 ;g/l in 1996. Average summer values of secchi and euphotic depth increased significantly from 4.5 to 6.5 m and from 10.5 to 13 m, respectively. The algal species composition changed and, during summer, total algal biomass decreased by 50 % and primary production by 25 %. Standing stocks of well-edible algae, rotifers, and herbivorous and carnivorous crustaceans did not exhibit a trend with TPmix, whereas their species compositions or egg-ratios were partially altered. The age-at-capture of planktivorous whitefish increased slightly. I tested the hypotheses that (1) changes should first be observed at the level of individuals or within species (altering e. g. C:P or egg-ratios) prior to changes within communities (affecting e. g. the taxonomic composition) and at the community level (affecting e. g. total biomass or production). This would imply that it is moreLarge (472 km2) and deep (zmean=101 m) Lake Constance is undergoing re-oligotrophication. Total phosphorus during winter mixing (TPmix) decreased from >80 during 1975-1981 to 22 ;g/l in 1996. Average summer values of secchi and euphotic depth increased significantly from 4.5 to 6.5 m and from 10.5 to 13 m, respectively. The algal species composition changed and, during summer, total algal biomass decreased by 50 % and primary production by 25 %. Standing stocks of well-edible algae, rotifers, and herbivorous and carnivorous crustaceans did not exhibit a trend with TPmix, whereas their species compositions or egg-ratios were partially altered. The age-at-capture of planktivorous whitefish increased slightly. I tested the hypotheses that (1) changes should first be observed at the level of individuals or within species (altering e. g. C:P or egg-ratios) prior to changes within communities (affecting e. g. the taxonomic composition) and at the community level (affecting e. g. total biomass or production). This would imply that it is more appropriate to conceptualize step-wise responses along a hierarchical gradient of increasing aggregation as suggested by hierarchy theory, rather than simultaneous changes at all hierarchical levels. (2) Responses become dampened along the food chain and with increasing body size, i. e. bottom-up control is most important for autotrophs. All communities studied (phytoplankton, crustaceans, fish) reacted at the individual level (e. g. by changes of (re)production rates), and/or within the community (e. g. altered taxonomic composition) whereas changes of bulk parameters of the entire community were restricted to phytoplankton. Hence, the first hypothesis is partially supported by the observed reactions and demands further testing. The second hypothesis is clearly supported by our data when comparing autotrophs and consumers, but not when comparing crustaceans and fish. The testing of these hypotheses is complicated by the large differences in size and, consequently, in reaction times of pelagic organisms on the one hand and the rather fixed time scale of limnological research on the other hand. The different time scales imply a selective perception of the various potential responses of the differently sized organisms as the time scales of the responses depend on body size and the level of aggregation. For example, we are more likely to establish physiological or behaviourial changes of fish, and taxonomical or biomass changes of phytoplankton. Acknowledging the scale dependence and level of aggregation is also crucial for cross-system comparisons.…