Evaluación del modelo hidrológico Cemeneige-GR6J para la cuenca del Río Aysén
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
La cuenca del río Aysén, ubicada en una zona montañosa del sur de Chile, presenta un régimen pluvio-nival influenciado por procesos de acumulación y derretimiento de nieve. Su difícil acceso, topografía abrupta y limitada cobertura de estaciones dificultan el monitoreo directo y hacen de esta cuenca un sistema desafiante para modelar. Este estudio evalúa el modelo conceptual CemaNeige-GR6J como herramienta simplificada para representar su respuesta hidrológica.
Se utilizaron datos grillados de temperatura, precipitación y evapotranspiración potencial del producto CR2MET (1979–2020), junto con caudales observados de la DGA (1996–2020). La caracterización de la cuenca consideró la curva hipsométrica para aplicar la rutina nival. La calibración se realizó bajo tres escenarios temporales mediante el algoritmo de Michel, utilizando el coeficiente de eficiencia de Nash–Sutcliffe (NSE) como función objetivo.
El modelo mostró mejor desempeño en años fríos y húmedos, con valores de NSE y KGE superiores a 0,75 y un RMSE promedio de 1,15 mm/día. Se identificó una subestimación sistemática de eventos extremos. El análisis de sensibilidad destacó a X2 (escorrentía directa) y X4 (tiempo de retardo) como los parámetros más influyentes.
Pese a su estructura conceptual simple y al uso de datos no calibrados localmente, CemaNeige-GR6J demostró ser una herramienta confiable para representar la dinámica hidrológica de una cuenca compleja. Su desempeño depende del tipo de año y de una calibración adecuada. Se concluye que su aplicación en contextos con datos limitados aporta valor a la gestión hídrica en zonas de montaña sujetas a alta variabilidad climática.
The Aysén River basin, located in a mountainous region of southern Chile, has a pluvio-nival regime influenced by snow accumulation and melt. Its steep topography, limited accessibility, and scarce monitoring make it particularly difficult to model. This study evaluates the conceptual model CemaNeige-GR6J as a simplified yet effective tool for representing the basin’s hydrological response. Inputs included gridded data from CR2MET (1979–2020): temperature, precipitation, and potential evapotranspiration, along with observed streamflow from DGA (1996–2020). Hypsometric analysis supported the elevation-based application of the snow routine. Calibration was conducted using the Michel algorithm under three temporal scenarios, with NSE as the objective function The model showed better performance in cold and wet years, with NSE and KGE above 0.75 and average RMSE around 1.15 mm/day. X2 (runoff coefficient) and X4 (routing time) were identified as the most sensitive parameters. A correction factor was proposed to address underestimation of peak flows. Despite its conceptual simplicity, CemaNeige-GR6J proved to be a reliable tool for simulating hydrological behavior in complex mountainous basins with limited data availability. Its capacity to represent daily streamflow depended on the type of year, with wet years having the best performance. This study highlights the need for more meteorological and snow stations in the region to assess the snowmelt and accumulation dynamics simulated by the model.
The Aysén River basin, located in a mountainous region of southern Chile, has a pluvio-nival regime influenced by snow accumulation and melt. Its steep topography, limited accessibility, and scarce monitoring make it particularly difficult to model. This study evaluates the conceptual model CemaNeige-GR6J as a simplified yet effective tool for representing the basin’s hydrological response. Inputs included gridded data from CR2MET (1979–2020): temperature, precipitation, and potential evapotranspiration, along with observed streamflow from DGA (1996–2020). Hypsometric analysis supported the elevation-based application of the snow routine. Calibration was conducted using the Michel algorithm under three temporal scenarios, with NSE as the objective function The model showed better performance in cold and wet years, with NSE and KGE above 0.75 and average RMSE around 1.15 mm/day. X2 (runoff coefficient) and X4 (routing time) were identified as the most sensitive parameters. A correction factor was proposed to address underestimation of peak flows. Despite its conceptual simplicity, CemaNeige-GR6J proved to be a reliable tool for simulating hydrological behavior in complex mountainous basins with limited data availability. Its capacity to represent daily streamflow depended on the type of year, with wet years having the best performance. This study highlights the need for more meteorological and snow stations in the region to assess the snowmelt and accumulation dynamics simulated by the model.
Description
Tesis presentada para optar al título de Ingeniero Civil Agrícola
Keywords
Evapotranspiración, Río Aysén, Cuencas hidrográficas