Pakistan is among the most vulnerable countries due to climate change. Historical data shows an increase in both rainfall and temperature over the entire country in the last 70 years. In addition, the severity and intensity of extreme events (floods, droughts, and heat waves) have also increased. Furthermore, as predicted by the climate models, future global warming could alter the water cycle and cause more extreme events, such as droughts and floods, in this region (Dars et al., 2020). Drought is one of the most complex natural hazards that affects water availability for agriculture and various socio-economic activities in a particular region (Adnan and Ullah, 2020). Quantifying drought characteristics, such as duration, severity, intensity, and peak is critical for policymakers and water managers.
Numerous drought indices have been used for this purpose, including standardized precipitation index (SPI) (McKee et al., 1993), decile index (DI) (Gibbs Maher, 1967), Palmer drought severity index (PDSI) (Palmer, 1965), and standardized precipitation evapotranspiration index (SPEI) (Vicente-Serrano et al., 2010). Several studies have adopted these drought indices to model droughts. For example, Edossa et al. (2016) used the self-calibrated Palmer Drought Severity Index (PDSI) to analyze droughts in central South Africa. Zou et al. (2005) used PDSI derived from monthly air temperature and precipitation data from 1951 to 2003. Bazrkar et al. (2020)developed a hydro-climatic aggregate drought index (HADI) to categorize drought in colder regions. In addition, Haroon et al. (2016) applied the drought severity index (DSI) and showed that the DSI could be efficiently applied only in agricultural and plain areas. Another research study by Wang et al. (2015) used SPI and SPEI to investigate drought severity in China by using monthly observed climatic data from 633 stations. In addition, Tirivarombo et al. (2018) used the SPEI and SPI to monitor droughts over the Kafue basin Zambia. They also compared both indices and found that SPEI efficiently calculated the drought duration and severity. Moreover, Jamro et al. (2019), (Jamro et al., 2020) conducted two studies on Pakistan and Balochistan Province to analyze the spatiotemporal variability of droughts using SPEI. Xie et al. (2013), (Ashraf and Routray, 2015, Jamro et al., 2020) applied the Standardized Precipitation Index (SPI) over Pakistan and Balochistan Province to analyze drought trends (Adnan et al., 2015). However, under the current scenarios of increasing temperatures, the effects of climate change cannot be ignored, and the temperature cannot be assumed to be stationary in the future. The SPEI integrates the effects of climate change on temperature as well as precipitation and is found to produce more reliable results than indices such as SPI that only use precipitation amount (Jamro et al., 2019).
To our knowledge, no study has been conducted in the Sindh Province to characterize droughts and their trends using SPEI. This research aims to analyze the spatial variability of drought characteristics (drought duration, severity, intensity, and peak) using 3- and 12- month SPEI across Sindh Province, Pakistan, from 1902–2015. The studied period is also divided into two parts, pre-1960 and post-1960, to highlight the effect climate change has had on the drought characteristics of the region.