Understanding the effects of anthropogenic stress on the provisioning of ecosystem services in the Laurentian Great Lakes

Background/Question/Methods

Effectively managing coupled human-natural systems requires understanding the effects of anthropogenic stressors on ecosystem services provisioning. Environmental stressors represent human influences on ecosystem health, while ecosystem services (particularly cultural and provisioning ones) connect ecosystem health to human wellbeing and create subsequent feedbacks on stressors. Understanding these relationships is important in the Great Lakes, which are highly valued but impacted by many stressors, including invasive species, climate change, chemical pollution, and habitat alteration. In the Great Lakes Environmental Assessment and Mapping (GLEAM) Project, we previously developed maps of 34 environmental stressors across all five lakes. Here, we expand our evaluation of cumulative stress at locations providing human benefits. We mapped locations and approximated usage of beaches, parks, fishing (charter, private, and commercial), birding, and boating. We evaluated cumulative stress for each service, customizing stress based on the stressors most affecting that service. We expected services reliant on sensitive biota and having higher human travel tolerances (e.g., birding, some fishing) to be more concentrated in lower stress locations than services less fitting these criteria (e.g., beaches, boating). We also examined whether the distributions of stressors and services are each correlated with human populations, one hypothesis to explain service provisioning at high stress.

Results/Conclusions

We found that services occurred predominantly in places that had high levels of cumulative stress. High incidence of stressors and services were seen along coasts, in rivermouths, and in other nearshore areas where human populations are concentrated. However, population densities were a variable predictor of these distributions. Stressors were not always positively correlated with population density; half were moderately positively correlated (r>.3), while the other half had more neutral or negative correlations. Services were not necessarily correlated to human population density either. For example, highly valued birding usually occurred with <1000 people/km². The differences in customized cumulative stress among services were not fully explained by human willingness to travel or taxa affected. In the future, we will investigate whether services show greater persistence in lower stress, and whether adjustments in the weightings of stressors affect these conclusions. These results will help us understand ecosystem services provisioning in the face of multiple environmental stressors.