Hydrological drought assessment in the tempisque-bebedero catchment system in Costa Rica

Abstract

The present thesis research was focused on the assessment of hydrological drought in the Tempisque-Bebedero catchment system in Costa Rica as part of the TropiSeca project framework. The study area is in the province of Guanacaste and has an extension of 5449.94 Km2, the region is characterized by a defined wet and dry season resulting in a marked seasonality in precipitation and streamflow regime which provokes frequent periods of water deficits considered as drought. The main objective of this research is to conduct an analysis on hydrological drought in the study area through the understanding of the behavior of hydrological cycle and its implications for the rice sector by applying different drought indices such as Standardized Precipitation Index (SPI) and Streamflow Drought Index (SDI). From the values obtained in the calculation of drought indices were studied the temporal distribution and spatial distribution based on the characterization of drought periods in terms of frequency, severity, duration, and seasonality. For the characterization of meteorological drought in the study area an approach based on area average precipitation was implemented to calculate a regional representative SPI for each sub-basin, in contrast, hydrological drought was assessed using only two streamflow gauges data provided by the ICE from 1973 until 2003. As result, Tempisque showed longer drought periods in comparison with Bebedero whose mean duration was lower but the number of drought events were more frequent. In terms of spatial distribution, it could be found that the upper basin experienced extreme meteorological drought periods at high time scales tied to a severe streamflow deficit probably justified by its low permeability due to geological characteristics that allow a slow movement of groundwater. Additionally, one of the aims of this thesis was to analyze the existence of correlation between precipitation and streamflow anomalies with rice yield and, to determine the influence of ENSO in climate variability using Sea Surface Temperature indices; in this phase of the research was found that climate patterns in the catchment system exhibited a significant influence by ENSO events with a significance level of 99% (r > 0.7) showing an important dependence of meteorological drought periods presented during the period 1980-2016. In terms of temporal behavior of rice yield anomalies was revealed moderate correlation coefficients (r < 0.4) in both watersheds due to in most of the cases the response of water deficit did not have significant impact in terms of magnitude as expected; in some periods in which drought period was present categorized as mild-drought, rice yield had a considerable decreasing compared with those in which was categorized as extreme event; these differences can be justified mostly because crop yield depends not only on weather, but also on variety of seed used and its coping capacity to periods of water scarcity, fertilizers, soil moisture, farming techniques, sowing date, temperature, irrigation, use of pesticides etc. The results of this thesis can be used to motivate future researches in the elaboration of crop models to predict yields based on physiological processes during plant development considering water requirement to take enough measures to mitigate the effects of drought periods. Furthermore, it should be considered to implement a drought monitor system in the area as an important tool of early warning system and as an indicator for the efficient water resources management.