Relationship between rainfall and streamflow in the La Plata Basin: Annual cycles, interdecadal and multidecadal variability

Main Article Content

Carla Gulizia
Inés Camilloni

Abstract

The aim of this study is to understand the interaction between rainfall and streamflow variability in the La Plata basin (LPB) along a wide range of timescales. The LPB is divided in six sub-basins associated to the main regional rivers (Paraguay, Paraná, Uruguay and Iguazú). The amplification of the streamflow response is addressed in order to evaluate to what extent river discharges variability can be explained by precipitation fluctuations. Mean annual cycles corresponding to the period 1931-2010 and to each decade of this interval are analyzed. Streamflow interdecadal changes are observed in most of the gauging stations. In addition, an 11-year moving-average filter is applied to the normalized annual time series. Results exhibit a considerable higher percentage of explained variance in the streamflow filtered series, highlighting the predominance of medium and low frequencies variability present in these compared to those of precipitation. Consistently, river discharges show higher spectral density in the interdecadal/multidecadal frequencies compared to precipitation analysis. A simple statistical approach to advance in the understanding of the complex rainfall-streamflow physical relationship is addressed with promising results: streamflow spectrums are derived directly from the precipitation spectrum, transformed by a “basin” operator, characteristic of the basin itself. It is assumed that watersheds act on precipitation as spatiotemporal integrators operating as low-pass filters, like a moving average. Streamflow power spectrums are simulated assuming that the underlying process is an autoregressive moving average. Considering the sub-basin areal-averaged precipitation time series as the only input, results show that simulated streamflow spectrums fit effectively the observations at the sub-basin scale. 

Downloads

Download data is not yet available.

Article Details

Sharing on:

PLUMX metrics