Perspectives on the Southwest Drought

Utah and southwestern North America are currently in a deep drought.  Although this year represents the greatest coverage of extreme and exceptional drought in Utah, the reality is that southwestern North America have been in the grips of an extended drought period, sometimes referred to as a megadrought, since about 2000.  As described in a paper in Science by Williams et al. (2020), the evidence for this megadrought is apparent in several climate indicators including reduced snowpack, reduced runoff and lake levels, declining ground water availability, drought stress, wildfire activity, etc.  

Weather and climate records for southwestern North America only extend back to the late 1800s.  However, it is possible to reconstruct past climate using other indicators, such as tree rings.  This typically requires the identification of a suitable tree in a representative region, coring of the tree, and measurement and dating of the tree rings.  Statistical models are then used to develop relationships between tree rings and climate variables, such as soil moisture, which are then used to "reconstruct" climate back into the past.

Matthew Bekker's Tree Ring lab at BYU does some great work with tree rings in Utah and environs.  If you are interested, below is a talk that he gave at the 2013 Restoring the West Conference at Utah State.

A key finding from tree-ring studies is that the climate of southwest North America has experienced extended periods of extreme drought and wetness in the past.  Periods of wetness are called pluvials.  The causes of these variations in climate are not completely understood, but are likely related to to sea surface temperatures in the Pacific Ocean, which in turn affect mid latitude flow patterns.  

The more recent paper by Williams et al. (2020) specifically examines the ongoing megadrought in the southwest United States.  Using tree rings, they produced a regional 19-year running-mean reconstruction of soil moisture anomalies in southwest North America.  Thus, the time series is designed to focus on long-term (in this case 19 years), wide spread (i.e., averaged over 30-45˚N and 105–125˚W.) climate variations in southwest North America.    

The key figure is below, which shows anomalies of 19-year soil-moisture anomalies from 800-2018.  In this plot, zero indicates average soil moisture conditions, positive values wetter than average, and negative values drier than average.  The red line is the reconstructed time series and the blue line recent observations.  Note that there are substantial variations throughout the record.  The five droughts and pluvials with the largest negative and positive anomalies are identified with red and green shading, respectively.  

Source: Williams et al. (2020)

A key finding to take from this figure (and other tree-ring studies), is that variations are essentially baked into the southwest North America hydroclimate cake.  These variations can be slow and extreme, as occurred from about 1550 to 1650, or more rapid and smaller amplitude, as occurred from about 1700-1900.  

In terms of the current drought, the 19-year running mean soil moisture as of 2018 was the 2nd lowest in the reconstruction, exceeded only by the late 1500s megadrought, although there are megadroughts prior to 900, around 1150, and around 1300 that are comparable.  From this longer range perspective, the current drought is one of the worst in the past 1200 years.  

Williams et al. (2000), however, take it a step further.  Using climate modeling and other techniques, they estimate the contribution of anthropogenic (human caused) climate change to the recent soil moisture anomalies, finding a 46% contribution.  Basically, recent global warming has taken would would have been a moderate drought and made it one of the worst on record.  The current drought is not caused by global warming.  Instead, it is exacerbated by global warming.  

This is an important distinction.  Global warming is shifting the characteristics of the southwest hydroclimate.  You could think of it as pulling down that red curve above so that drought tends to be more persistent and severe, whereas the pluvials become less common and less extreme.  

Thus, it's not all global warming.  What will happen next year or during the next decade is also going to depend on the climate variations described above.  At one extreme we are in a period where the climate variations favor drought.  Between now and 2030, we see maybe a couple of good snow years, but for the most part, hot and dry persists.  At the other extreme, we swing back a pluvial period.  Between now and 2030, we see more cool-season snowfall and some recovery of reservoirs and the Great Salt Lake, although that recovery won't be as robust as it might be in a cooler climate.  

This is why we need to be cautious about having what I'll call "global warming myopia."  Global warming is important, but it is not acting in isolation.  It is important to also consider how climate variations, which are cooked into the southwest climate cake, will affect our snow and water resources on yearly to decadal time scales.  This is a critical area of scientific research.  Reliable guidance on what sort of snow year we will have next year or over the next several years would pay huge dividends for water resource management.  Unfortunately, we are not there yet.  

Note: There is an Erratum (correction) for the Williams et al. (2020) paper as there was an error in the tree-ring chronology.  These errors are described here and have a negligible influence on the results.