The estimation and evaluation of a satellite-based drought index using rainfall and evapotranspiration.
Year: 2017
Type: dissertation
Abstract: South Africa is naturally a water-scarce country and the climate is semi-arid hence, become particularly prone to droughts, which impact all facets of society and the environment.Climate change projections predict that droughts will worsen with the higher temperatures and fluctuating rainfall patterns.It is therefore vital that the country prepares for this natural phenomenon.In this study, drought estimation will be undertaken by using drought indices that have been utilised to analyse drought events in the past and by using the currently available sources of climatic data and satellite earth observation data.Since Evapotranspiration (ET) and rainfall drives the hydrological cycle, drought indices based on these variables will be investigated.The overall aim of this research study was to investigate, apply and evaluate a satellite-derived evaporative drought index in South Africa.The spatio-temporal evolution of droughts were analysed in the Upper Thukela and Umgeni Catchments, within the province of KwaZulu-Natal, South Africa for the study period 2011-2016.Drought analysis was carried out by using the Evapotranspiration Deficit Index (ETDI), the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI), at a monthly temporal scale.The LSA-SAF DMET product was used to obtain actual ET (ETa) estimates for the study.A validation of the product against the surface renewal (SR) and eddy covariance (EC) systems within the Umgeni Catchment produced a correlation coefficient of 0.90 and a R 2 of 0.81.The satellite-derived Hargreaves-LST method was followed to obtain reference ET (ETo) estimates; the approach was calibrated using the FAO Penman Monteith ETo estimates and produced correlation coefficients of ≥0.86 for the selected quaternary catchment's (QC's).The SPI was calculated using in-situ data and the Famine Early Warning Systems Network African Rainfall Climatology (FEWS ARC 2.0) rainfall product for the Umgeni Catchment.This produced a R 2 , ranging from 0.80-0.96,whilst between the FEWS ARC 2.0 and in-situ data for the Upper Thukela produced an R 2 that ranged from 0.55-0.72.The years, 2015 and 2016, was revealed as major drought years by the SPI, SPEI and ETDI, with prolonged dryness being detected for 2010 until 2016.The ETDI was compared using cross correlation analyses, which produced R 2 values that ranged from 0.31-0.73for the Umgeni Catchment.Evaporative indices are suggested for use in semi-arid to arid regions as they contribute significantly to the water cycle.In addition, by utilizing satellite-derived drought indices, the analysis of the spatiotemporal patterns of droughts can contribute to an understanding of the mechanism of drought propagation and planning required for drought hazard mitigation.This Masters project, entitled "The Estimation and Evaluation of a Satellite Based Drought Index using Rainfall and Evapotranspiration", has been funded by the NRF (National Research Foundation), in collaboration with ACCESS (Alliance for Collaboration on Climate and Earth Systems Science) and the DST (Department of Science and Technology).The financial assistance of the aforementioned institutions towards this research is hereby acknowledged.The opinions expressed and the conclusions arrived at, are those of the author and are not necessarily attributed to the NRF.I also wish to acknowledge and express my sincere gratitude towards the following people and institutions: I would like to thank my supervisor, Ms
Author: Maqsooda Mahomed
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2947800942