Freshwater bivalves tell of past climates: But how clearly do shells from polluted rivers speak?

Freshwater bivalves, Margaritifera margaritifera (Linnaeus) and Unio crassus (Philipsson), from rivers in Sweden (79 specimens) and England (one specimen) were used to study the effects of human-induced pollution on shell growth (Table 1). We analyzed variations in annual and daily shell growth rates of 80 specimens from unpolluted and polluted (pH < 5, oxygen depletion and eutrophication) localities. 35% of the variability in annual growth of shells from unpolluted rivers is explained by ambient temperature during June through August. Daily shell growth also co-varies with the temperature during the growth season (approximately April-October). Long-term trends in temperature and growth compare well to each other. A weak correlation was also found for shell growth and the summer North Atlantic Oscillation (NAO) index. However, all of these environmental signals are obscured in specimens from polluted settings. In settings with high human impact, shell growth does not co-vary with summer temperatures or the NAO. Results of our study suggest a judicious sampling strategy when shells are used for climate reconstructions.