Joel Creswell studied mercury in New Haven area streams to test two hypotheses; 1) that mercury concentrations are positively correlated with the percentage of watershed impervious cover under storm water conditions, and 2) that mercury concentrations and impervious surfaces would show no correlation under base flow conditions.  His study of four different streams found no correlation between mercury concentrations and impervious surfaces under baseflow conditions.  The study of mercury concentrations under stormflow conditions was inconclusive due to a small data set.