Recopilación y análisis de información hidroclimatológica para el diagnóstico y modelación hidrológica para la formulación del plan de ordenamiento del recurso hídrico en la cuenca del río Guavio

Abstract

This internship was carried out at the Regional Autonomous Corporation of Cundinamarca (CAR) with the objective of contributing to the formulation of the Water Resource Management Plan (PORH) for the Guavio River basin, through the collection, refinement, analysis, and modeling of hydroclimatological information. The need for this study arises from conflicts regarding water use and availability in the basin, associated with the pressure of human activities and climate variability, which demands robust technical tools to understand the hydrological dynamics of the system. The methodology involved several stages, including data collection from hydrometeorological stations of precipitation and temperature time series, and the implementation of a hydrological model. A combined modeling approach was adopted using the GR4J conceptual rainfall-runoff model and the RS MINERVE platform, allowing the integration of a parsimonious model structure with a semi-distributed spatial representation. This approach enhanced the capacity to simulate runoff generation and flow propagation processes within the basin. Model calibration was performed by comparing observed and simulated streamflow data, using performance indicators such as the Nash-Sutcliffe efficiency, Pearson correlation coefficient, and Kling-Gupta efficiency. The results showed an acceptable model performance, with a good representation of the temporal dynamics of streamflow, although peak flows were generally underestimated. Overall, the model proved to be a useful tool for representing the hydrological behavior of the basin and for generating relevant inputs for water availability assessment. These findings contribute to water resource planning by identifying critical periods and supporting decision-making processes. However, the results should be interpreted considering uncertainties related to data quality and model simplifications.