Streamflows Prediction Models for the Colombian Generation System Considering El Niño Effect

by Oscar J. Mesa,
Ricardo A. Smith,
Pedro J. Restrepo,
José E. Salazar,
Luis F. Carvajal,
Juan D. Velásquez,

Document Type: Proceeding Paper

Part of: North American Water and Environment Congress & Destructive Water


A monthly and a daily prediction models for some Colombian rivers with hydropower developments were developed and implemented. Several monthly streamfiow prediction models are presented. These models include the ENSO effect on Colombian hydrology. This effect could be considered by using some macro climatic variables that represents the ENSO effect, such as sea surface temperatures, humidities, wind velocities and other derived indexes in some regions of the Pacific Ocean. The developed models are: a multiple linear regression model, a models based on singular spectral analysis, a model based on neural networks and a regime dependent autoregressive model (RAR). The developed daily models were based on the extended streamfiow prediction procedure (ESP). In the ESP weighing procedure a climatic variable related with the El Niño—Southern Occilation (ENSO) phenomena was used. Two rainfall runoff models were used: a multiple linear regression model that relates the streamflow daily discharges with daily precipitation and streamflow discharges at previous time periods and the Sacramento watershed model. Some results, conclusions and recommendations are presented.

Subject Headings: Streamflow | Autoregressive models | Hydrologic models | Regression analysis | Hydro power | Hydrology | Climate change | Linear analysis | Pacific Ocean | California | United States | Colombia | South America

Services: Buy this book/Buy this article


Return to search