The Environmental Kuznets Curve (EKC) postulates that environmental quality is initially degraded with increasing economic prosperity, until reaching some turning point where environmental quality improves with increases in wealth. Tests using environmental indicators beyond those that affect human health have been less supportive of the EKC idea. We hypothesize that environmental changes impacting human health are more likely to show evidence of an EKC than variables less directly related to human health. Furthermore, the EKC hypothesis assumes that ecological damage is reversible, and we argue that this assumption is not always valid. We test for evidence of an EKC in Dane County, Wisconsin, using non‐point‐source pollution time series data for Lake Mendota throughout the 20th century. We examine metals deposited in lake sediments (cadmium, chromium, copper, and lead), other non‐point‐source pollutants such as sulfur and dissolved reactive phosphorus (DRP), and water clarity (measured by Secchi disk depth). We relate changes in ecological variables to changes in real wealth per capita (RWPC) in Dane County over time. The EKC did not describe the relationship between all ecological and economic indicators tested; however, several variables were related to RWPC. Our strongest results (for Secchi depth, DRP, and copper) show increasing pollution with increasing wealth. Secchi depth and DRP are related to water quality and clarity, which have value to society but less direct, immediate health consequences. P pollution may also be fairly irreversible over short time scales. The best models and plots for cadmium, chromium, and lead suggest improvements in environmental quality with increases in RWPC, although these trends were not statistically significant. Results for sulfur were inconclusive. Overall, wealth did not explain much of the variability in any of the ecological variables examined here, suggesting that consideration of additional factors are necessary to explain their dynamics. Economic prosperity cannot be expected to improve all aspects of the environment, but may be biased toward aspects that are ecologically reversible phenomena or of concern to human health.