Assessing dependence between land use/land cover and water quality: a comparison at a small and a large watershed in Uruguay
Evaluación de la dependencia entre el uso/cobertura del suelo y la calidad del agua: comparación entre una cuenca pequeña y una grande en Uruguay
Resumen:
Changes in land use/land cover (LULC) directly or indirectly affect water quality in watercourses and impoundments. Sustainable management strategies aimed to enhance ecosystem health and community well-being require an accurate water-quality evaluation. This study looks into the correlation between temporal changes in LULC, represented by selected landscape variables (land cover area and proportion, patch density, Euclidean nearest-neighbor distance, mean shape index, and Shannon index), and water quality variables (nitrate, total phosphorus, and total suspended solids) at catchment scale. To compare the watershed-size influence, this analysis was performed at two different spatial scales represented by two Uruguayan basins of different sizes, San Salvador (3,118 km2) and Del Tala (160 km2). Partial Least Squares and Random Forest unsupervised machine-learning models were employed for this analysis. By exploiting a non-model-biased method based on game theory (SHAP), the LULC characteristics were quantified and ranked based on their level of importance in the water-quality evaluation. The main outcomes of this study proved that patch density is one of the most influencing metrics in both watersheds and for both models. Agricultural land use is the most critical one at both catchments and agricultural with a forage crop land uses are the most important ones for both algorithms. Furthermore, it is possible to state that the adopted techniques are valuable tools that can provide an adequate overview of the water‐quality behavior in space and time and the correlations between water-quality variables and LULC.
Los cambios en el uso del suelo y la cobertura del suelo (LULC) afectan directa o indirectamente la calidad del agua en cursos de agua y embalses. Las estrategias de gestión sostenible destinadas a mejorar la salud del ecosistema y el bienestar de la comunidad requieren una evaluación precisa de la calidad del agua. Este estudio analiza la correlación entre los cambios temporales en LULC, representados por variables de paisaje seleccionadas (área y proporción de cobertura del suelo, densidad de parches, distancia euclidiana al vecino más cercano, índice de forma promedio e índice de Shannon), y las variables de calidad del agua (nitrato, fósforo total y sólidos suspendidos totales) a nivel de cuenca. Para comparar la influencia del tamaño de la cuenca, este análisis se realizó a dos escalas espaciales diferentes representadas por dos cuencas uruguayas de diferentes tamaños, San Salvador (3118 km2) y Del Tala (160 km2). Se emplearon modelos de aprendizaje automático no supervisados de Mínimos Cuadrados Parciales y Bosque Aleatorio para este análisis. Al aprovechar un método no sesgado basado en teoría de juegos (SHAP), las características de LULC se cuantificaron y clasificaron según su nivel de importancia en la evaluación de la calidad del agua. Los principales resultados de este estudio demostraron que la densidad de parches es una de las métricas más influyentes en ambas cuencas y para ambos modelos. El uso agrícola del suelo es crítico en ambas cuencas, y los usos agrícolas con cultivos forrajeros son los más importantes para ambos algoritmos. Además, es posible afirmar que las técnicas adoptadas son herramientas valiosas que pueden proporcionar una visión adecuada del comportamiento de la calidad del agua en el espacio y el tiempo, así como las correlaciones entre las variables de calidad del agua y LULC.
2023 | |
ANII: FSA_PI_2018_1_147713 ANII: SA_PI_2018_1_148628 ANII: FSA_PP_2018_1_147701 |
|
Water quality Land use/land cover Unsupervised learning Feature importance Calidad del agua Uso/cobertura del suelo Aprendizaje no supervisado Características relevantes |
|
Inglés | |
Universidad de la República | |
COLIBRI | |
https://hdl.handle.net/20.500.12008/43370 | |
Acceso abierto | |
Licencia Creative Commons Atribución (CC - By 4.0) |
Sumario: | Changes in land use/land cover (LULC) directly or indirectly affect water quality in watercourses and impoundments. Sustainable management strategies aimed to enhance ecosystem health and community well-being require an accurate water-quality evaluation. This study looks into the correlation between temporal changes in LULC, represented by selected landscape variables (land cover area and proportion, patch density, Euclidean nearest-neighbor distance, mean shape index, and Shannon index), and water quality variables (nitrate, total phosphorus, and total suspended solids) at catchment scale. To compare the watershed-size influence, this analysis was performed at two different spatial scales represented by two Uruguayan basins of different sizes, San Salvador (3,118 km2) and Del Tala (160 km2). Partial Least Squares and Random Forest unsupervised machine-learning models were employed for this analysis. By exploiting a non-model-biased method based on game theory (SHAP), the LULC characteristics were quantified and ranked based on their level of importance in the water-quality evaluation. The main outcomes of this study proved that patch density is one of the most influencing metrics in both watersheds and for both models. Agricultural land use is the most critical one at both catchments and agricultural with a forage crop land uses are the most important ones for both algorithms. Furthermore, it is possible to state that the adopted techniques are valuable tools that can provide an adequate overview of the water‐quality behavior in space and time and the correlations between water-quality variables and LULC. |
---|