Variabilidad regional de los componentes del balance hídrico

Abstract

Nowadays land planning and management requires a thorough knowledge of the available resources, and water is one of the most important natural resources. In order that water demands could be satisfied, there is a need to assess the allocation of theses resource in time and space. The water balance method allows determining the volumes of available water that potentially may be distributed seasonally. However, accurate application of these methods requires a rather dense network of hydrometerorological measurements, so that in general the available information is not enough to adequately describe properly the study system. Most commonly, therefore, the preliminary information required to perform a water balance is scarce or lacking, both in space and time. In this way the information required, which is not available from direct measurements or synoptic observations can only be estimated in the best case using statistical methods. It has been shown that in Argentina annual precipitation show diminution from east to west and from north to south, and this general rule gives an approximate idea spatial rainfall variation. Also according to the different geographical regions, rainy or dry periods indicate the seasonal variability in the annual cycle. Under conditions responsible for high variability of precipitations, networks of meteorological stations with, and with big proportion of short registers, like the existent in Argentina, the challenge is which is the most efficient method to consider the variability temporary space of the hydrological variables. Traditional methods used for estimating precipitation, relying on averages obtained by the Thiessen polygon approach have been used in the past to assess the areal average precipitation of a basin. Subsequently, the need to study, validate and disseminate different alternative methodologies has been arisen; these should allow estimating areal precipitations and evapotranspiration rates in a more reliable manner and therefore it is expected to obtain better and reliable results in the components of the water balances. The use of regionalisation techniques and kriging of point data allows to obtain fields of variables, which are associated to a given pattern of spatial distribution, and this pattern can be represented The challenge lies in associate these spatial fields with time variations, using the spatial data sets as input in the models of water balance assessment. In other words, how big can be the effect of of the spatial variability in the results provided by water balance models?. Therefore, the objective of this Ph. D. thesis was to incorporate the variability of the main components of the water balance for assessing the water availability status at the studied catchments. The study has been carried out using the information available in various subctachments of Rio de la Plata basin, using the CHAC model, which is based on the Témez method and the hydrometerorological information obtained from several national organisms of Argentina associated to management of water resources. Modelling has been performed on a monthly base and involved a total period of 40 years. Three different methods have been utilized to obtain precipitation and evapotranspiration data sets as input data of the models, i.e., Thiessen polygons, regionalization and kriging techniques. Stream flow estimated using the above mentioned three methods for 12 catchments located at the left bank of the Paraná river have been checked to analyse what is the method that provides most reliable data sets matching the measured hygrograms at the control section of the available stations, taking into account levels of error adapted. The obtained results also clearly showed that recording data sets of hydrometeorological variables, which are uniformly distributed along the space, is very important for assessing the water balance ant to minimise the errors of interpolation. Geostatistical techniques allow generation of precipitation and evapotranspiration from point data. Nevertheless the use of time series has been shown to be irreplaceable for the application of methods of balance. Using balance models at the monthly time scale has been shown to be adequate to quantify the availability of water and to assess its seasonal occurrence. The monthly water availability estimated by means of hydrological balances has been found to be a valuable tool for scheduling and managing the studied basins, located within Rio de la Plata basin. This catchment provides water for a variety of uses, including navigation, agriculture, and human consumption, and husbandry, fisheries, energetic, industrial, and recreational and also plays a role as a receptor of industrial and domestic effluents.