Observed And Projected Changes In Drought Risks Over South Asia Under Global Warming Scenarios

In the recent past,South Asia(SA)has faced detrimental droughts due to changing climate that affected water availability,agriculture production and disrupted the lives of more than 2-billion people.Generally,SA droughts are primarily driven by the failure of the summer monsoon(June-September)providing 78% of the total annual rainfall,which is a lifeline for the millions of people across the region.The interannual variations of the summer monsoon precipitation in the region are associated with the large-scale atmospheric and oceanic circulations,particularly El-Ni(?)o Southern Oscillation(ENSO),Pacific Decadal Oscillation(PDO),and the Indian Ocean Dipole(IOD).The weakening of the summer monsoon precipitation can cause meteorological droughts,which,if prolonged,could propagate hydrological to agriculture droughts.Drought risk in SA has been changing under the observed and projected climate,however,it is still unclear how the drought frequency and severity in the observed climate could affect the water resources,agriculture activities,and socioeconomic growth with a warmer climate over SA.Nevertheless,major droughts have posed insightful challenges in all the spectrums of lives across SA,little is known about their features and drivers in a future warming climate.Moreover,there might be substantial uncertainty in the characteristic of the worst droughts in distinct reanalysis datasets,which has not been carefully evaluated over SA.The present study characterizes the observed and future changes in amplified drought risks across SA and its sub-regions under changing climate.The in-situ observational and reanalysis datasets of the Climatic Research Unit(CRU TS),National Centers for Environmental Prediction version II(NCEP-2),European Centre for Medium-Range Weather Forecasts Version-5(ERA-5),Modern-Era Retrospective Analysis for Research and Applications version II(MERRA-2),and projections from the phase six of the Coupled Model Inter-comparison Project(CMIP6),models have been analyzed.Different statistical metrics,anomaly analysis,Bayesian Dynamic Linear(BDL)model,climate indices,run theory,and multivariate return period approaches were used to assess the observed and future changes in meteorological drought.This study employed two Shared Socioeconomic Pathways(SSPs)and warming targets,i.e.,(SSP2-4.5 and SSP5-8.5)and(1.5 and 2 oC),respectively.Firstly,this study sought to evaluate the performance of reanalysis products against in-situ observations over diverse regions of Pakistan as a representative of the South Asian region.Secondly,the study provides detailed assessments of observed changes in seasonal drought characteristics and their potential drivers over Pakistan.Moreover,the study assesses recent changes in drought episodes and their possible linkage to atmospheric circulation and human influences over SA.Finally,the research investigated the future changes in increased drought risks over SA and its sub-regions under 1.5 and 2 oC warming levels.The main conclusions of the study are as follows:1)Drought indices and drought areas assessed from reanalysis products are more representatives of historical droughts in southern Pakistan,and overestimation is evident for drought severity in western than eastern Pakistan.Statistically significant increasing trends(1984-1998 and 2000-2010)in monthly drought areas and occurrence are evident by CRU TS and MERRA-2 in dominant arid and semi-arid regions.Climate variables and drought features of southern Pakistan are best represented by CRU TS and MERRA-2,while ERA-5 best represents that of southwestern and western parts.The Nash–Sutcliffe efficiency(NSE)result ranges from-2 to 0.9,where the NSE of SPEI values(-1.0)show relatively weaker than SPI values(0.5)in most parts of the regions,specifically in southern Pakistan.A strong positive linear relationship on a monthly scale is evident in CRU TS,MERRA-2,and ERA-5 exhibiting relatively high correlation coefficient(0.84),except for NCEP-2;2)An increase in drought severity was revealed mostly over arid and semi-arid regions for both cropping seasons.While temperature played a significant role in defining droughts over dry and hot seasons,rainfall is influential over the western disturbances-influenced region.The analysis of atmospheric circulation patterns revealed that large-scale changes in wind speed,air temperature,relative humidity,and geopotential height anomalies are the likely drivers of droughts in the region.The Sea Surface Temperature(SST)and multivariate ENSO Index are the most influential factors for seasonal droughts across Pakistan;3)A significant drying trend was revealed towards the southwestern and northeastern parts of SA during winter.Drought durations show dipolar patterns;however,prolonged drought duration was noticed in the southwestern parts of SA compared to northern parts.Notably,arid and semi-arid regions in SA have shown increasing and significant drought trends in terms of area,frequency,and severity,implying that droughts are relatively more severe in those regions.The relationship fluctuates over time between drought variability and climate anomalies.The drought variability in SA is strongly correlated with IOD and ENSO but relatively less associated with the Himalayan(Tibetan Plateau)thermal forcing.The Dipole Mode Index(DMI)impacts drought episodes over southwest SA relative to northern parts and is mainly influenced by the SST;4)The frequency of 50-year historical droughts(i.e.,under bivariate framework)might double across majority part of SA under the 1.5 oC warming,and additional 20% land would be exposed to extreme drought events under the 2 oC warming climate.The frequency of severe drought episodes is also expected to increase under 1.5 oC(40%-75%)and 2 oC(60%-90%)relative to the recent climate.The largest increase in exposure is projected in R2 and R4(e.g.,southern parts of Pakistan and India),followed by R1(e.g.,south-western part of Afghanistan).Besides,about 75%(65%)SA population(gross domestic products: GDP)could suffer from increased drought risks under the 1.5 oC warmer climate,whereas the additional 0.5 oC warmings will further lead to an unbearable situation in the region.Limiting global warming to 1.5 oC compared with 2 oC levels can significantly reduce the influence of drought risk over most areas of the SA.Future projection results for the surface air temperature and relative humidity further clarify the dry conditions and elevated air temperature due to less precipitation across subtropical and arid regions of SA.The continental landmass of SA is under a strong anticyclonic pattern,which(northern hemisphere)is associated with descending winds and less precipitation.Thus,future teleconnection changes could possibly enhance the drought prediction challenges over most parts of the SA in near future.This study uses the newly released global climate models(GCMs)from the CMIP6 ensemble mean,yearly population,and GDP datasets,which has been widely used in many previous studies,omitted the role of dynamic socioeconomic influences.This is because the global population and GDP projections have coarse resolution and uncertainty,which may lead to biased findings.Future studies should focus on the dynamic characteristics,which will improve our understanding of socioeconomic exposure in a warming climate.These are some of the points that need to be thoroughly investigated in future studies,including the climate sensitivity analysis,human-induced influences,and their role as potential drivers of regional drought variability.