Drought is a ubiquitous feature of the U.S. Great Plains. Droughts can be short (seasonal), prolonged (occur over several years) or, according to the prehistoric record of so-called “megadroughts,” even persist for decades or longer. The definition of drought depends on context; that is, a short-term drought may be highly detrimental to agriculture, while stream flow may only be affected over prolonged time periods.
Whatever the timescale or affect, drought is due to insufficient rain and snow, possibly in combination with excess evaporation. Drought is likely initiated by remote factors such as cyclical changes in ocean-surface temperature patterns that affect the large-scale atmospheric circulation. These remote factors work in conjunction with local feedbacks due to soil moisture, snow cover and vegetation that can enhance and/or prolong the drought. In this article I will describe how global and regional models of climate have been used to evaluate the relative importance of these local and remote factors. Of course, what everyone interested in the Great Plains wants to know is what will happen to droughts in the future, especially under scenarios of global warming. I will summarize the current understanding of likely climate change for the Great Plains for the remainder of the 21st century, including implications for future drought.
Past History of Drought in the Great Plains
I need not convince anyone that the approximately 150-year historic record for the Great Plains clearly indicates major episodes of drought—the Dust Bowls of the 1930s and the even more extreme droughts of the 1950s serve as ample reminder. Though less well-remembered, episodes of drought also occurred in earlier decades as well. Nonetheless, the historic record also indicates that much of the time has been relatively wet. Indeed, a series of wet years in the 1870s helped induce large-scale settlement and farming in the Great Plains (the fallacious “rainfall follows the plow”). Similarly, a series of wet years preceded the Republican River Compact that since the 1920s has legally apportioned the Republican’s water between Colorado, Nebraska and Kansas. Indeed, the original motivation for the Compact was flood control! Today this controversial compact is used to divvy up scarce water resources
The proxy record over the past 2,000 years clearly indicates the prevalence of drought; indeed, it reveals so-called megadroughts that affected much of North America, including the Great Plains. (The term “megadroughts” refers to periods of drought much more prolonged than what has occurred during the historic record; these periods usually last several decades, with occasional wet years interspersed.) The megadroughts were especially prevalent during the medieval time period that lasted from approximately A.D. 900–1300. Tree-ring records in particular show that droughts were particularly frequent and persistent throughout much of the western and central U.S. during this time. As shown in Figure 1, these droughts usually lasted for decades, indeed, sometimes for most of a given century. In the Great Plains, the grassland cover of the Nebraska Sandhills was destroyed, and the dunes became mobilized, indicating drought conditions much more severe than any of the 20th century. In summary, multiple lines of evidence suggest that during medieval times, drought was the dominant feature of the Great Plains rather than the exception, as at present.
So, it seems clear that the past century and a half has been in a relatively wet period, in an environment that much of the time is considerably drier. So what are the implications for the future when the predicted global warming does indeed occur? To understand potential consequences, we must first better understand the cause of drought.
Causes of Drought
Mid-latitude drought, such as afflicts much of the U.S. on a recurring basis, is thought to have both remote and local influences (called “forcings” by climate scientists). The remote forcings, typically associated with cyclical changes in various Atlantic and Pacific Ocean sea-surface temperature patterns, influence the large-scale atmospheric flow, affecting moisture transport as well as the lift necessary to cause the moisture to condense and precipitate out. The local forcings are associated with land surface-atmosphere interactions, including roles played by vegetation, soil moisture and snow cover. These can affect the moisture fluxes between the surface and the atmosphere, but importantly also the surface energy budget, with consequences for atmospheric stability.
Key remote forcings of precipitation on seasonal and through decadal timescales are changing patterns of sea-surface temperatures, such as El Nino/La Nina (ENSO), the Pacific Decadal Oscillation (PDO) and the Atlantic Multidecadal Oscillation (AMO). ENSO operates on a several-year timescale, the PDO on a decadal timescale and the AMO on a multidecadal, approximately 30- to 60-year timescale. Singly and, importantly, in combination they influence the atmospheric circulation in ways that are conducive to prolonged precipitation anomalies over the U.S.
The state of the land surface, especially the amount of moisture in the soil, provides an important local forcing, particularly in semiarid regions such as the Great Plains where feedback between land cover and climate are particularly strong (Figure 2). Lowering soil moisture leads to reduced evaporation and warmer surface temperatures (as energy goes into heating the land surface instead of evaporating water). This means less water in the atmosphere locally available for rainfall, though this effect is fairly small. More important is the stabilization of the atmosphere, with the warmer surface conditions leading to rising air and high pressure aloft. This high pressure in the upper atmosphere strongly inhibits precipitation, especially the convective thunderstorms so important to the Great Plains from late spring through summer.
So what causes the soil to dry out? Lack of precipitation, obviously, along with continued or even increased evaporation. For example, envision a situation whereby a remote forcing (such as a La Nina episode) leads to dry conditions over the Great Plains. As these dry conditions persist (a La Nina event, for example, usually lasts 12–18 months), the ground eventually dries out. This in turn results in less precipitation, thereby making the drought event more severe than it already is. Further, even as the remote forcing diminishes (e.g., La Nina event terminates), the reduced soil moisture by itself causes the drought to persist even longer. A series of large rainfall events may be required to finally moisten the ground.
Climate Change in the Great Plains
So what is likely to happen to climate in the Great Plains over the remainder of the 21st century, and what are the implications for drought? The most obvious concern is the impact that global warming may have on the region. The most recent climate change assessment by the Intergovernmental Panel on Climate Change (2007 IPCC AR4)1 suggests a warming in the Great Plains, most likely on the order of 2–3 degrees Celsius (3.6–5.4 degrees Fahrenheit) between now and 2100. This warming is greater than that to the east but less than that to the west. Precipitation (rain and snowfall) is more complicated. Regions to the north and east (e.g., New England and the mid-Atlantic) tend to be wetter (along with somewhat less warming), while regions to the south and west are drier (and warmer). This means there must be a “zero” line of no change somewhere in between. All 21 models used for the AR4 “Multi-model Mean” (MMD) have that zero line running north to south somewhere in the Great Plains, though the exact location varies from model to model. For example, 11 of the models have wetter conditions for Lincoln, while ten have them drier. Does this mean the models are all over the place and cannot be trusted? Absolutely not—they are all saying the same basic thing; it’s just that at the 100-mile resolution, what will happen for Lincoln is too close to call. More worrisome is the consistent reduction in rain and snow (and warming) in the Rockies—this implies greatly reduced flow through the entire Platte River basin.
Global warming is hardly the only important consideration for climate change in the Great Plains. First off, one needs to consider the long-term perspective. We know that the past 2,000 years were relatively dry, with evidence of more wet conditions for several thousand years before then. So, absent any other significant influences or forcings, what would climate be doing naturally?
Almost certainly of greater consequence are land-use changes, e.g., the amount of land put into agricultural production or the even the nature of the crop. If, for example, a large-scale replacement of corn by switchgrass (tall prairie grass) occurs, this alone might lead to warmer and drier conditions. This is because the extensive irrigation required to grow corn is largely unnecessary for switchgrass. This effect is essentially the same as described above for reduced soil moisture. Put another way, this large-scale irrigation may have helped keep the Great Plains anomalously cool and moist since its advent in the 1960s.
Summary: Implications for Future Drought in the Great Plains
So what are the overall implications for future drought in the Great Plains? We must first realize that the “natural” system has been in an anomalous wet period, and while we cannot be certain, most likely the tendency would be to return to the dry conditions that have so dominated the past 2,000 years. Human activities are also likely to make things worse, as global warming reduces snowpack in the Rockies and at least the western Plains become drier. The impact of human agricultural activities is less clear and rests largely on the extent to which large-scale irrigation stays the same or changes to cover either a larger or smaller areal extent.
In conclusion, while hoping for the best, I also have to realize that drought is a ubiquitous feature of the Great Plains and is likely to get worse, not lessen, as the century progresses.
1. The new IPCC AR5 model simulations have just recently been released and are still being analyzed. Preliminary results suggest that they confirm the previous results, providing increased detail and understanding without significantly changing the interpretations based on IPCC AR4.