Implementación y calibración de una estación hidrométrica en el río Zamora

Abstract

Abstract: In the estimation and prediction of streamflow, a range of tools and techniques is available, encompassing both automatic and manual hydrometric stations, flow measurement methods, statistical techniques, and predictive models such as machine learning through neural networks. The focus of this study lies in the compilation of data related to the Zamora River's streamflow through flow measurements. Additionally, a statistical analysis of the obtained streamflow data was conducted using regression techniques and artificial intelligence approaches to predict streamflow based on observed and extrapolated water levels. The research includes an assessment of the effectiveness of prediction models to identify which exhibits better agreement between observed and predicted streamflow. The results indicate that the linear regression model, applying least squares, demonstrates a significantly higher correlation compared to neural networks. Despite this distinction, both models showcase a substantial capacity for streamflow prediction, with linear regression standing out as the preferred option in terms of correlation. This analysis contributes to the field of river streamflow estimation and underscores the importance of appropriate model selection based on required precision.