Watershed Classification in the Great Lakes Basin: Implications for Water Quality and Agricultural Management Practices

Abstract

In recent years, the Great Lakes have faced a resurgence of cyanobacterial harmful algae blooms (cHAB), primarily attributed to non-point sources, notably agricultural activities. While significant efforts have been directed toward implementing conservation practices to mitigate nutrient losses, existing literature often examines the efficacy of best management practices (BMPs) and spatiotemporal drivers of nutrient loss separately, neglecting their interconnectedness. Recent studies suggest that conservation practices' effectiveness may vary spatially, necessitating targeted interventions to avoid trade-offs. This study aims to delineate distinct ecoregions based on known spatiotemporal drivers of nutrient loss and analyze their implications for water quality across different land use-land cover (LULC) types. Using Google Earth Engine (GEE), two Cascade K-means clustering analyses were conducted separately on climate and geophysical variables, resulting in three distinct ecoregions for each domain. These findings were integrated with data from the Provincial Water Quality Monitoring Network (PWQMN) and HYDAT stations to assess patterns in water quality degradation and nutrient loss mechanisms across ecoregions. Additionally, statistically downscaled climate change datasets from Environment and Climate Change Canada (ECCC) were utilized to determine shifts in climate conditions across established climate ecoregions. Furthermore, climatic ecoregions displayed a latitude-dependent pattern in water quality degradation. Under projected climate changes, the coolest regions are anticipated to resemble current conditions in the warmest regions, leading to a northward shift in agricultural suitability. These findings underscore the necessity of adopting a context-dependent approach to agricultural management practices, especially in light of projected climate shifts. A one-size-fits-all approach to BMP recommendations and implementation falls short, highlighting the importance of tailored strategies to address the unique challenges posed by each ecoregion.