Changes in the water balance is of increasing concern on a warmer planet. As a broad rule of thumb, areas that are traditionally dry are expected to become drier, and areas traditionally wet will likely become wetter. But that average pattern is also reflected in the interannual variability as higher temperatures enhance the feedback from more quickly drying soils, even if precipitation doesn't change. Therefore, there is a need to plan for more severe and more frequent drought years, almost anywhere. The standardized precipitation evapo-transpiration index (SPEI), computed over 12 month periods, captures the cumulative balance between gain and loss of water across the interannual time scale. The likelihood for severe drought analyzes the frequency at which prolonged dry conditions are expected, and shown in the map is the probability for change by 2050, using the most aggressive RCP8.5. Other scenarios will show similar direction of changes, albeit as somewhat reduced probabilities.
The map shows change in projected Annual Likelihood of Severe Drought by 2050 compared to the reference period (1986-2005) under RCP 8.5 of CIMP5 ensemble modeling. Brown/Yellow areas are more likely to experience severe drought compared to the baseline period. Meanwhile, Blue/Green areas are less likely to experience severe drought.
Drought projections are somewhat controversial because a large part of the outcome hinges on the evapo-transpiration (ET) feedback. The literature covers both projections of significant increases in global drought as well as some lesser trends. Here, an intermediate ET formulation is used that is less sensitive to warming than some, but also doesn't require many assumptions of surface conditions that are required for others because the models don't represent these well enough. Ultimately, because drought potential increases everywhere, the likelihood for severe drought offers insight if a region is in a broader domain where increases could be substantial or if the region might not be less affected by changes in precipitation and moisture.