We have just published a new paper on near-term future hydrological change in the American Southwest that adds to the growing evidence for declining surface water availability. But first some background.
Back in 2007 we published a paper 'Model Projections of an Imminent Transition to a More Arid Climate in Southwestern North America'. That paper showed that the climate models assessed for Intergovernmental Panel on Climate Change Assessment Report 4 projected that rising greenhouse gases would cause southwestern North America to become more arid and that this transition should already be underway. It occurs as part of a general drying and expansion of the subtropical dry zones with roots in an intensification of the hydrological cycle ('wet getting wetter and dry getting drier') and poleward shifts of the jet stream and storm track and expansion of the Hadley Cell.
At the time this work led to a lot of questions such as whether the climate model projections are reliable, whether drying of the region is already occurring and what it means for things that mattered like soil moisture and river flows? A whole new set of climate model projections have now been done for IPCC AR5 and many of these models also have higher spatial resolution that can better represent the complex topography of the West (though still not as well as we would like) and other improvements. So we have now published another study (Projections of Declining Surface-water Availability for the Southwestern United States) that goes beyond the broad brush treatment in the 2007 paper to look at how precipitation minus evapotranspiration (P-E, the net flux of water at the Earth's land surface), runoff and soil moisture change by season in the coming decades for three water resource regions: 1) California and Nevada, 2) the Colorado River headwaters and 3) Texas.
Like the last generation models, the new model simulations indicate that southwestern North America will become progressively more arid as the century advances with important changes appearing in the immediate future. Despite the fact that precipitation might increase in some regions and seasons (e.g. winter in northern California) while decrease in others, rising temperatures mean that a large majority of the model simulations project that spring and annual mean runoff will decrease. Soil moisture is also projected to decrease throughout the year, especially in Texas. The mean of the multiple climate models analyzed projects that annual mean runoff in the Colorado River headwaters in 2021-2040 will be 10% less than in the decades at the end of the 20th Century.
The path towards a drier climate with less soil moisture and runoff will not be smooth since the natural variability of hydroclimate, on seasonal to multidecadal timescales, is immense. The range of results amongst the model simulations illustrates this well and it is always possible to find some individual model simulations that, because of their particular history of internally generated variability, have wet conditions in the coming decades. However, the best bet would be that, even as natural variability continues, dry spells will get progressively drier and wet spells progressively less wet.
Colorado River flow, as well as water resources in California and Texas, are already severely taxed and any reduction of supply due to climate change is worrying. Long tree ring records allow estimations of past variations in Colorado River flow and suggest a 15% reduction of flow during the 12th Century megadrought. Therefore the new paper concludes (in typical dry academic style):
Hence, anthropogenic climate change is projected to lead to a potential reduction of Colorado River flow comparable to the most severe, but temporary, long-term decreases in flow recorded. These projected declines in surface-water availability for the coming two decades are probably of sufficient amplitude to place additional stress on regional water resources given the pressure of meeting agricultural demands as well as those of a growing population while needing to preserve riparian ecosystems.REFERENCES