Integrating watershed and ecosystem service models to assess best management practice efficiency: guidelines for Lake Erie managers and watershed modellers

Lake Erie is the shallowest and most biologically productive system of the Great Lakes, surrounded by large urban, industrial, and agricultural areas. This combination prompted extensive efforts to promote best management practices (BMPs) to mitigate non-point source pollution in Lake Erie's watershed. Recent technical and conceptual advancements caution that significant variability exists in the BMP efficiency to reduce the severity of runoff and nutrient concentrations due to differences in implementation, the dependence of operational performance on local soil and climatic conditions, storm events and seasonality, and declining performance over time owing to imperfect maintenance. Given the uncertainty surrounding the design and efficiency of BMPs in abating non-point source pollution, our primary objective is to review the critical strengths and potential weaknesses of nine agricultural BMPs promoted for use in the Lake Erie watershed. We examine the capacity of the current generation of watershed process-based models to recreate possible BMP-mediated changes in the water and nutrient cycles. After reviewing modelling strategies (dynamic, external forcing, and empirical) to recreate non-linear watershed responses and feedback loops to BMP efficiency, our study recommends adopting dynamic representations of the interplay among key mechanisms, like soil moisture, water table, nutrient availability, plant uptake, and subsequent growth. Notwithstanding the increased sophistication of complex mathematical models, their learning capacity is usually compromised by the coarse resolution of environmental data and limited empirical knowledge to accurately parameterize environmental properties and partially understood biogeochemical processes. Moreover, BMPs may differentially affect the provisions of an ecosystem (e.g., a BMP may amplify one ecosystem service while dampening another). In this context, we highlight the expression of the value of ecosystem services in monetary and non-monetary terms as a critical information piece when considering trade-offs among costly and diverse policy decisions. Our study also examines the degree to which different types of valuation methods, socioeconomic models, and data have been operationalized in Lake Erie. Consistent with the Integrated Watershed Management framework, we advocate the adoption of a rigorous mass-balance approach to assess the impact of BMPs on nutrient cycles, as well as the integration of the projected environmental improvements with terrestrial ecosystem services, beneficial use impairments, and aquatic ecosystem services. The proposed strategy has the potential to improve the decision-making process by identifying cost-effective management actions and balancing different goods and services provided by the agroecosystems at different time scales.