Thursday, October 11, 2018

Deep Dive on Current and Future Drought in Utah

A recent article in the Salt Lake Tribune (Utah just experienced its driest year since scientists have kept records) strongly suggests that our recent drought is due to climate change.  While global warming is a contributor, drought has many drivers and it is essential that we recognize that fact as we plan for the future.

Drought is nothing new to the American southwest.  Historical records and proxies based on tree ring records show long-term fluctuations from drought (dry periods) to pluvial (wet periods) are characteristic to the region even before human influence on climate.

Tree-ring reconstructed streamflow for the Bear River. Drought and pluvial (wet) periods illustrated by black and grey fill, respectively.  Note that this graph is smoothed to illustrate slower climate variability. Source: DeRose et al. (2015).

That human influence on climate is, however, growing and expected to play an increasing role in drought over the American southwest.  It is, however, very difficult to disentangle the contribution of long-term global warming from weather and climate variability that cause variations in temperature and precipitation over periods of years or even decades.

Drought has many definitions.  Although it is frequently defined as a prolonged period of abnormally low rainfall it can also be defined based on water availability with drought being a period of natural water scarcity.  It is possible, for example, to have near average precipitation but below average soil moisture, snowpack, and runoff depending on the atmospheric conditions.  For example, as temperature increases, declines in the fraction of precipitation that falls as snow and the amount of precipitation retained in the end-of-winter snowpack, combined with increased evapotranspiration (drying of soils due to evaporation and transpiration by plants), can yield water scarcity even if precipitation does not change.

Drought conditions over the American southwest in recent years have been the result of both decreased precipitation and anomalous warmth.  At issue is how much each of these is due to long-term global warming vs. atmospheric circulations due to the slowly evolving climate system.  The bulk of the evidence today indicates that our precipitation drought at least partially if not primarily reflects the latter and is related to slowly evolving sea surface temperatures in the Pacific Ocean.  It is also possible that it partially reflects what is known as Hadley cell expansion, a northward expansion of an atmospheric circulation characterized by mean ascent near the equator and descent near 30ยบ latitude, at least in the extreme southwest.

In addition, it is important to recognize that climate models project a complex picture of the change in annual precipitation with global warming.  For example, the median projection by 16 downscaled climate models calls for either very little change or a small increase in precipitation in the mountains of northern Utah depending on future greenhouse gas emissions.  Trends across the southwest as a whole, however, vary geographically.  They also vary by model, indicating a lack of a clear consensus and this leads to only low to medium confidence in these trends.

Source: Cayan et al. (2013)
Thus, I believe we should be very cautious about attributing the recent precipitation deficit over northern Utah to global warming and I have avoided making such a statement in public talks and discussions with the media.

On the other hand, it is clear that recent warmth over the American Southwest has enhanced the drought and increased water scarcity through increased evapotranspiration.  This warmth does appear to be largely attributed to global warming.  Thus, global warming is a drought enhancer.

Many of these issues are discussed in the report we prepared for Governor Jon Huntsman Jr's Blue Ribbon Advisory Council on Climate Change in 2007.  Below is an excerpt. I have preserved the italics, underlines, and bullets.
Ongoing greenhouse gas emissions at or above current levels will likely result in a decline in Utah’s mountain snowpack and associated changes to spring runoff. Year-to-year variations in snowfall will continue to dominate mountain snowpack, streamflow, and water supply during the next couple of decades. As temperature increases through the century, it is likely that a greater fraction of precipitation will fall as rain instead of snow, the length of the snow accumulation season will decrease, and snowpack loss due to evaporation will increase. Trends that are likely to emerge as the climate warms during the 21st century include:

  • A reduction in natural snowpack and snowfall in the early and late winter for the winter recreation industry, particularly in lower-to-mid elevation mountain areas (trends in high elevation areas are unclear).

  • An earlier and less intense average spring runoff for reservoir recharge.

  • Increased demand for agricultural and residential irrigation due to more rapid drying of soils.

  • Warming of lakes and rivers with associated changes on aquatic life, including increased algal abundance and upstream shifts of fish habitat.

Future water supplies are strongly dependent on long-term trends in precipitation. If average precipitation remains similar to that of the 20th century, the changes noted above will result in a declining water supply. This decline will be exacerbated if the region becomes more arid. An increase in precipitation is required to offset the changes noted above. Current climate models project a decline in summer precipitation across all of Utah. During the winter, projections indicate a decrease in precipitation over the southwest United States and an increase in the northwest. Utah is located in the transition zone between these regimes where there is low confidence in future precipitation trends. Although a shift to a wetter climate cannot be ruled out, it is more likely than not that water supplies in Utah and the Colorado River Basin will decline during the 21st century. In addition, since precipitation will continue to fluctuate from year to year, the threat of severe and prolonged episodic drought is real and ongoing.
Although we don't have high confidence in long-term trends in annual precipitation, we do have high confidence that we are going to see fluctuations in precipitation from year to year and decade to decade and that our future is going to be a warmer one.  That warmth will load the dice toward water scarcity (i.e., drier soils, declines in runoff) even if mean annual precipitation trends are flat or slightly upwards.  Droughts will be worse.  Pluvials less "productive."

What concerns me about attributing the recent precipitation drought to global warming is that we are going to see years and periods with above average precipitation in the future.  When that happens, we shouldn't be discussing if this is a long-term trend, if it suggests that we might have a wetter future with global warming, or whether or not projections of future change were wrong [Note: climate reports for the southwest are very cautious in projecting future precipitation trends given the low confidence and uncertainty that exists (e.g. Cayan et al. 2013)].  Instead, should be discussing how to take advantage of that bounty given a future in which the dice are loaded for more sustained and severe drought than we have experienced in the past.  The future of Utah is hotter and drier, even if we have a string of good precipitation years in the future and even if the long-term average of mean annual precipitation is unchanged or increases.


  1. Great post, Jim. Coincident to this was a small piece on the Trib about the 1922 Colorado River Compact and that regarding Lake Mead (was just there!) "The Bureau of Reclamation says the chances of a shortfall are 57 percent by 2020."
    Drew Hardesty

  2. Jim, flying into SLC these days, I see a lot of dirt where there used to be lake. While human activity, climate change, and short term fluctuations probably all play a role in this, has anyone looked to see what the effect of a smaller lake may be on our snow pack? If storms cross 20% less open water before slamming into LCC, does that mean our lake effect loses 20% of its punch? And is this a self-reinforcing cycle - less lake = less lake effect = less snow = less runoff = less lake ....?