Colombian monthly energy inflows predictability

Abstract

Streamflow forecasting is essential for water resources management in several social and economic strategic sectors, involving space-temporal variability modeling of the hydrological processes and the influence of several climatic phenomena. Furthermore, high water-dependent sectors such as the Colombian electricity market, require not only the expected streamflow values but also the occurrence probability or reliability bands of such forecast inflows necessary in robust risk analyses. We propose a mathematical approach for monthly streamflow forecasting in Colombia and quantify its predictability, incorporating climate model outcomes as a time series of macroclimatic indexes and punctual hydro-climatological stations. The methodology integrates parametric and non-parametric models, exogenous variables analysis, and uncertainty estimation through stochastic modeling. This research will contribute to the Colombian hydrology understanding and provide elements for risk analysis, planning, and decision-making in social and economic sectors involved with water resources management.

Keywords

• streamflow forecasting
• uncertainty
• Colombian streamflow’s predictability