Detailed streamflow information is essential for water resources management, hydraulic infrastructure design, flood-risk assessment, and related decision-making processes. However, in many developing countries, the limited availability of discharge gauging stations constrains hydrological analysis and increases the need for rainfallrunoff models to address data scarcity. In this context, this study proposes a hybrid morphometric–satellite-based methodology that integrates remote sensing products and basin morphometric descriptors. The approach was applied to the Caquetá River Basin in Colombia, where daily CHIRPS precipitation and ERA5 evapotranspiration data were used to simulate streamflow regimes at 12 streamflow-gauging stations with discharge measurements, using the GR4J rainfall–runoff model. The main objective was to establish mathematical relationships for estimating GR4J model parameters from morphometric variables derived for each sub-basin, thereby enabling the generation of daily streamflow time series at ungauged locations within the watershed. Model calibration was performed using the Differential Evolution Markov Chain (DE-MC) algorithm with 8,000 iterations, while validation was conducted using at least 20% of the time-series length for the 12 streamflowgauged stations assessed. The calibration and validation quality were assessed using the Nash-Sutcliffe efficiency (NSE) and the Kling-Gupta efficiency (KGE). The results showed that the GR4J parameters (X1–X4) can be estimated from morphometric characteristics, including basin perimeter, circularity ratio, main channel length, and channel slope, with coefficients of determination ranging from 0.80 to 0.98. These findings demonstrate the potential of the proposed methodology for regionalizing rainfallrunoff model parameters based on the physiographic and environmental characteristics of river basins