Widespread summer droughts across North America are particularly damaging and costly. The last of these droughts was in 2012 and cost over an estimated $30 billion. In an effort to understand these droughts better, Hun Baek, a graduate student in the PaleoDynamics lab, has published a paper in Journal of Geophysical Research - Atmospheres exploring the oceanic and atmospheric causes of these droughts over the last ~150 years. He finds that the tropical Pacific Ocean has played an important role in forcing these droughts, but an equal contribution has come from internal atmospheric variability. He also finds a passive if ambiguous role for the Atlantic, which is in contrast to previous thinking about the influence of the Atlantic Ocean on the occurrence of widespread summer droughts in the United States. The abstract of Hun's paper is given below, while his paper can be found here.
Abstract: The contributions of oceanic and atmospheric variability to spatially widespread summer droughts in the contiguous United States (hereafter, pan‐CONUS droughts) are investigated using 16‐member ensembles of the Community Climate Model version 3 (CCM3) forced with observed sea surface temperatures (SSTs) from 1856–2012. The employed SST forcing fields are either (i) global or restricted to the (ii) tropical Pacific or (iii) tropical Atlantic to isolate the impacts of these two ocean regions on pan‐CONUS droughts. Model results show that SST forcing of pan‐CONUS droughts originates almost entirely from the tropical Pacific because of atmospheric highs from the northern Pacific to eastern North America established by La Niña conditions, with little contribution from the tropical Atlantic. Notably, in all three model configurations, internal atmospheric variability influences pan‐CONUS drought occurrence by as much or more than the ocean forcing and can alone cause pan‐CONUS droughts by establishing a dominant high centered over the U.S. montane west. Similar results are found for the Community Atmosphere Model version 5 (CAM5). Model results are compared to the observational record, which supports model‐inferred contributions to pan‐CONUS droughts from La Niñas and internal atmospheric variability. While there may be an additional association with warm Atlantic SSTs in the observational record, this association is ambiguous due to the limited number of observed pan‐CONUS droughts. The ambiguity thus opens the possibility that the observational results are limited by sampling over the twentieth century and not at odds with the suggested dominance of Pacific Ocean forcing in the model ensembles.