By Sheri Fritz
The major droughts of the 20th century, such as those of the Dust Bowl period and 1950s, had profound environmental, economic and social impacts in the Great Plains and are viewed by many as extreme events. Yet the 20th century provides a relatively short-term view of climate variability, and it is useful to extend our perspective to include longer periods of time. A longer-term perspective gives us a better understanding of both the natural recurrence of drought for planning purposes and of whether recent trends may be a product of human impact on climate or are simply a manifestation of long-term natural variation.
In most parts of North America, the instrumental record of climate only extends back about a hundred years or less. This instrumental record can be augmented by the written accounts of early explorers and naturalists, such as Lewis and Clark or Fremont and Nicollet in the Great Plains. However, the diaries and notes of early explorers describe only short windows of time, may reflect subjective judgment, and still only extend the record of climate variation by a hundred years or so. To put recent climate variation in the context of hundreds or thousands of years, we can turn to geological or biological records of the history of climate change.
One of the best tools for reconstructing recent climate history is measuring growth rings of living, recently dead or fossil trees. Annual rings of growth can be distinguished in trees that grow in a seasonal climate, and the width of each ring reflects how favorable the climate conditions were for growth. Because a ring is added each year, the age of each ring can be determined simply by counting from the newest growth back through the sequence of rings. Ring-width measurements from living trees can be compared with instrumental measurements of climate from those same years and used to derive a quantitative and predictive relationship between ring width and a specific climate parameter (such as growing-season temperature or precipitation). This relationship can then be applied to tree-ring sequences that formed in years prior to the instrumental record to reconstruct the climate of the past. The length of time covered by tree-ring sequences depends on the longevity of individual tree species and how well preserved fossil wood might be. Regardless, tree-ring records commonly extend back in time only a few hundred years.
Trees are infrequent in the grasslands of the Great Plains, but they do occur in river valleys, and these trees provide regional tree-ring records that give us an insight into the climate of the last few hundred years. Tree-ring sequences from Long Pine Creek and the Niobrara River Valley in north-central Nebraska (see Figure 1) show that major droughts equivalent in severity and length to those of the 20th century occurred in the late 19th century, whereas fewer major droughts occurred during the 18th century. The ring measurements also suggest that the first two decades of the 20th century were quite wet relative to the 300-year record, a pattern that is apparent in many trees from western North America, as well. In parts of western United States, water policy was codified in these decades when the climate was unusually wet, thus the perception that rainfall was abundant likely biased policy formulations that were based on so-called average or “normal” water availability.
One limitation of tree-ring records is that they only extend back a few hundred years in most cases, which may not provide an adequate window into the full range of natural climate variability. This is particularly true, because the interval from approximately A.D. 1300–1850 was known as the Little Ice Age, and many parts of Europe and North America experienced cool conditions, which in turn may have affected regional water availability. Thus, it’s possible that conditions during the Little Ice Age interval may not be characteristic of variability during slightly warmer time periods, such as in the 20th century and times prior to A.D. 1300.
Lakes, wetlands, dunes and other landscape features often persist for centuries, or even millennia, and can be used to learn about climate even further back in time. My own research uses the bottom sediment of lakes as recorders of drought. Lake sediments are a product of material that forms in the lake itself, that is washed in from the watershed, and that falls in from the atmosphere. These materials enter the lake, fall to the lake bottom, and accumulate over time in a layer-cake fashion. As you go deeper into the mud, you are going back in time. Fossils can be extracted from individual layers of the sediment and dated, and these fossils can be used to tell the history of the lake, watershed and atmosphere back through time.
Precipitation history can be reconstructed from lake sediments, because lakes change in depth in response to increases and decreases in moisture. In some situations, they also change in salinity, because increased evaporation in a dry climate removes water and leaves behind dissolved salts, whereas increased precipitation dilutes the lake water and produces fresher conditions. These changes in lake depth and salinity influence the species composition of various organisms that live in the lake. Some of these species are resistant to degradation and thus are preserved in the sediments at the bottom of the lake after they die. Various chemical compounds also reflect changes in lake depth and salinity and can be analyzed in the sediments to reconstruct past moisture conditions.
Records from lakes in the northern Great Plains show that droughts the severity of the Dust Bowl period are a recurring part of natural climate variability (see Figure 2). At times, such as in the 16th century, major droughts occurred more frequently, whereas at other times, such as the early 1800s, major drought was rare. What is most striking about moisture records from the Great Plains is that drought was prolonged and persisted for multiple decades during some time periods within the last few thousand years. The most recent of these periods occurred between ~800 and 1,000 years ago—also referred to as the Medieval Period, because it correlates with the European cultural period by the same name (much of northern Europe experienced unusually warm conditions during Medieval times). Some people refer to the Medieval drought in the Great Plains as a “megadrought,” because it was unusually severe and persistent.
Evidence for major prolonged drought in the Great Plains during the Medieval Period is also present in the dunes and wetlands of the Nebraska Sand Hills. Some of the modern wetlands surrounded by grass-stabilized dunes have thin layers of sand at depth in their sediments (see Figure 3). These thin layers of sand suggest intervals of time when the grass cover of the dunes was destroyed by drought, and sand moved out across the dried wetland surface. Later, when precipitation increased again, these lenses of sand were buried by peat formed in the wetland. In Jumbo Valley Fen, a site in north-central Nebraska studied by University of Nebraska - Lincoln geologists, the most recent of these buried sand lenses dates from the Medieval Period. Dating of the dune sand near the surface of the modern dunes also shows that, throughout many areas of the Sand Hills, the youngest sections of the dunes were deposited in Medieval times as winds moved sand across the land surface during a period of severe drought. Thus, there is widespread evidence from lakes, dunes and wetlands in the northern and central Great Plains for major and persistent drought 800 to 1,000 years ago that was much more prolonged than anything in human- recorded climate history.
Scientists commonly use the last one to two thousand years as a yardstick for evaluating natural patterns of climate variation, because the configuration of the Earth’s climate system over this interval is very similar to what it is today. Yet drought even more severe than that of Medieval times was common in the Great Plains during other periods of the last 10,000 years, particularly within the so-called mid-Holocene, between approximately 9,000 and 5,000 years ago. During this interval, the Earth received more summer solar radiation (insolation) than it does presently, because of cyclic changes in the Earth’s orbit around the sun. Because the sun-Earth configuration was different, this interval is not a good model for natural drought variation under current conditions. Nonetheless the dominance of drought, as severe or more severe than observed in the Dust Bowl period, serves to reinforce the notion that major drought is a common and natural part of the climate variability of the Great Plains.
The geological record of drought recorded in tree rings, lake and wetland sediments, and sand dunes does not really allow us to predict the future occurrence of drought and whether or not drought severity will increase as a result of human impacts on climate. Yet it does tell us that, even without human-induced climate change, major drought is recurrent and should be considered a natural part of the climate variability of the Great Plains. So, instead of adopting the common viewpoint that droughts are anomalies, we should manage our water resources to plan for severe droughts, such as those of the Dust Bowl period, the 1950s, and early decades of the 21st century. It may not be realistic to plan for the much less common but more persistent “megadroughts,” such as during Medieval times—such a prolonged interval of drought certainly would have devastating social and economic impacts. Yet we clearly can do a better job of water conservation, management and planning to create a society and an economy that are less vulnerable to the natural fluctuations between wet and dry that characterize the semi-arid climate of the Great Plains.