Investigating water quality sensitivity to climate variability and its influencing factors in four Lake Erie watersheds

From forest to urban: Assessing the impact of land cover on water quality

Land use changes driven by population growth and economic development lead to water pollution, making clean water access a strategic global issue. Water quality is not only compromised by factors such as land use, seasonality, and human activities, but it is also vulnerable to climate conditions. This study explored the influence of land cover in sub-catchment and riparian areas on surface water quality, considering temporal, seasonal, and climatic variations. Ten sampling points were selected in a Brazilian populous watershed. Water quality data and semi-automatic land cover classifications were collected, providing a combined remote sensing and statistical analysis. The results indicated that increased urban development in the watershed negatively impacts water quality. Anthropogenic areas have a strong negative correlation with water quality in both sub-catchment and riparian area, with urbanized (ρ = -0.69 and ρ = -0.60; p-value <0.001) and exposed soil (ρ = -0.45 and ρ = -0.42; p-value <0.001) areas having the greatest effect. Forest cover in riparian areas is crucial for minimizing pollutant transfer to water bodies (ρ = 0.41), even in dry or rainy periods (ρ = 0.45 and ρ = 0.37, respectively). The Redundancy Analysis (RDA) indicates that urban areas negatively affect water quality with key contaminants such as organic matter, solids and Escherichia coli. This study emphasizes the challenges posed to water quality by urbanization and the importance of adaptive analysis beyond seasonal patterns, underscoring the need for effective policies and targeted conservation strategies for safeguarding water resources and ecosystem health.

Pathways linking watershed development and riparian quality to stream water quality and fish communities: Insights from 233 subbasins of the Great Lakes region

Anthropogenic stressors such as urban development, agricultural runoff, and riparian zone degradation impair stream water quality and biodiversity. However, the intricate pathways that connect these stressors at watershed and riparian scales to stream ecosystems-and their interplay with climate and hydrology-remain understudied. In this study, we used Partial Least Squares (PLS) path modeling to examine these pathways and their collective impacts on stream water quality and fish community structures across 233 watersheds in the Great Lakes region. Our study suggests that moderate levels of watershed development enhance overall fish richness, potentially due to increased water temperature and nutrient availability, but reduces both the percentages and richness of cold water and intolerant taxa. Riparian quality exerts indirect effects on water quality with climate and stream order serving as key mediators. Complementing our SEM analysis, we also used Multiple Linear Regression (MLR) models and identified a significant positive relationship between the proportion of clay and agricultural land with TN concentrations. However, TP concentrations are influenced by a more complex set of interactions involving developed areas, soil, and slope. These findings emphasize the necessity of adopting integrated management strategies to preserve the health and integrity of freshwater ecosystems in the Great Lakes region. These strategies should integrate watershed and riparian protection measures while also taking into account the effects of climate change and specific local conditions.

The role of agricultural drainage, storm-events, and natural filtration on the biogeochemical cycling capacity of aquatic and sediment environments in Lake Erie's drainage basin

Lake Erie is the most at risk of the Great Lakes for degraded water quality due to non-point source pollution caused by agricultural activities in the lake's watershed. The extent and temporal patterns of nutrient loading from these agricultural activities is influenced by the timing of agronomic events, precipitation events, and water flow through areas of natural filtration within the watershed. Downstream impacts of these nutrient loading events may be moderated by the co-loading of functionally relevant biogeochemical cycling microbial communities from agricultural soils. This study quantified loading patterns of these communities from tile drain sources, assessed whether functional communities from agricultural sources influenced downstream microbial functionality, and investigated how distance from agricultural sources, storm events, and areas of natural filtration altered nutrient cycling and nutrient fluxes in aquatic and sediment environments. Water and sediment samples were collected in the Wigle Creek watershed in Ontario, from tile drains through to Lake Erie, from May to November 2021, and microbial nitrogen (N) and phosphorous (P) cycling capacity (quantitative PCR), and nutrient levels were evaluated. Results showed that N and P functional groups were co-loaded with nutrients, with increased loading occurring during storm events and during agricultural activities including fertilization and harvest. Overall functional capacity in the aquatic environment decreased with distance from the agricultural sources and as water transited through natural filtration areas. In contrast, the sediment environment was more resilient to both agricultural disturbances and abiotic factors. This study expands our understanding of when and where different stages of N and P cycling occurs in agriculturally impacted watersheds, and identifies both seasons and regions to target with nutrient mitigation strategies.

Investigating machine learning models in predicting lake water quality parameters as a 3-year moving average

Lake water quality plays a vital role in the lake ecosystem, including biotic (for living creatures, such as plants, animals, and micro-organisms) and abiotic interactions. In this research, various types of machine learning (ML) methodologies, such as classification and regression tree (CART), chi-squared automatic interaction detector (CHAID), C5 tree, quick, unbiased, and efficient statistical tree (QUEST), along with multilayer perceptron (MLP) neural network, and radial basis function (RBF) neural network, are employed to predict the concentration of water quality parameters (P, EC, TDS, pH, DO, NH3, SO4, and θ). Lake Erie is situated at the international border of the USA and Canada. The C5 tree and QUEST tree are used to classify data and predict the number of groups, while the other methods are used to predict the concentration of water quality parameters in the form of a 3-year moving average. The greater matching between the observed and predicted data of dissolved oxygen (NSE = 0.978, bias = 0.126) shows that the CART decision tree has higher accuracy in correctly detecting the concentration of this parameter. The C5 tree could identify 33 groups correctly out of 36 total groups, which shows better accuracy for the C5 tree in classifying the data for this parameter.