An Attempt to Forecast Seasonal Precipitation in the Comahue River Basins (Argentina) to Increase Productivity Performance in the Region

Abstract

Precipitation decreased in the last century in the southern Argentinian Andes mountains, especially in winter, and this trend is expected to continue in the future, generating a significant negative impact on the local and national economy. This fact is relevant since the Comahue river basin, compressed by the sub-basins of the Negro, Neuquén and Limay rivers, is characterized by fruit-horticultural production and by the presence of hydroelectric dams. As the river flows are largely influenced by the interannual variability of autumn and winter rainfall, the availability of a good seasonal forecast provides an efficient tool to minimize risks. The study addresses the analysis of the main climatic forcing of precipitation to design some statistical models to forecast autumn rainfall. Using monthly precipitation data for the period 1981–2017 from several national institutions and data of atmospheric variables from the NCEP/NCAR reanalysis, correlation maps have been used to define the best set of predictors. The multiple linear regression methodology was used to build predictive models. The main conclusion is that the South Atlantic Ocean High, the sea surface temperature of the Pacific and Indian Oceans and the polar jet are the most relevant predictors and the designed models have good skill.