Droughts in the United States
From Climate Change Reconsidered, a work of the Nongovernmental International Panel on Climate Change
Andreadis and Lettenmaier (2006) examined twentieth century trends in soil moisture, runoff, and drought over the conterminous United States with a hydro-climatological model forced by real-world measurements of precipitation, air temperature, and wind speed over the period 1915-2003. This work revealed, in their words, that “droughts have, for the most part, become shorter, less frequent, less severe, and cover a smaller portion of the country over the last century.”
Using the self-calibrating Palmer Drought Severity Index (SCPDSI), as described by Wells et al. (2004), Van der Schrier et al. (2006) constructed maps of summer moisture availability across a large portion of North America (20-50°N, 130-60°W) for the period 1901-2002 with a spatial latitude/longitude resolution of 0.5° x 0.5°. This operation revealed, in their words, that over the area as a whole, “the 1930s and 1950s stand out as times of persistent and exceptionally dry conditions, whereas the 1970s and the 1990s were generally wet.” However, they say that “no statistically significant trend was found in the mean summer SCPDSI over the 1901-2002 period, nor in the area percentage with moderate or severe moisture excess or deficit.” In fact, they could not find a single coherent area within the SCPDSI maps that “showed a statistically significant trend over the 1901-2002 period.”
Going back considerably further in time, Fye et al. (2003) developed gridded reconstructions of the summer (June-August) basic Palmer Drought Severity Index over the continental United States, based on “annual proxies of drought and wetness provided by 426 climatically sensitive tree-ring chronologies.” This work revealed that the greatest twentieth century moisture anomalies across the United States were the 13-year pluvial over the West in the early part of the century, and the epic droughts of the 1930s (the Dust Bowl years) and 1950s, which lasted 12 and 11 years, respectively.
The researchers found the 13-year pluvial from 1905 to 1917 had three earlier analogs: an extended 16-year pluvial from 1825 to 1840, a prolonged 21-year wet period from 1602 to 1622, and a 10-year pluvial from 1549 to 1558. The 11-year drought from 1946 to 1956, on the other hand, had at least 12 earlier analogs in terms of location, intensity, and duration; but the Dust Bowl drought was greater than all of them, except for a sixteenth century “megadrought” which lasted some 18 years and was, in the words of Fye et al., “the most severe sustained drought to impact North America in the past 500 to perhaps 1000 years.”
In another long-term study, Stahle et al. (2000) developed a long-term history of North American drought from reconstructions of the Palmer Drought Severity Index based on analyses of many lengthy tree-ring records. This history also revealed that the 1930s Dust Bowl drought in the United States was eclipsed in all three of these categories by the sixteenth century megadrought. This incredible period of dryness, as they describe it, persisted “from the 1540s to 1580s in Mexico, from the 1550s to 1590s over the [U.S.] Southwest, and from the 1570s to 1600s over Wyoming and Montana.” In addition, it “extended across most of the continental United States during the 1560s,” and it recurred with greater intensity over the Southeast during the 1580s to 1590s. So horrendous were its myriad impacts, Stahle et al. unequivocally state that “the ‘megadrought’ of the sixteenth century far exceeded any drought of the 20th century.” They state that a “precipitation reconstruction for western New Mexico suggests that the sixteenth century drought was the most extreme prolonged drought in the past 2000 years.”
Last, we come to the intriguing study of Herweijer et al. (2006), who begin the report of their work by noting that “drought is a recurring major natural hazard that has dogged civilizations through time and remains the ‘world’s costliest natural disaster’.” With respect to the twentieth century, they report that the “major long-lasting droughts of the 1930s and 1950s covered large areas of the interior and southern states and have long served as paradigms for the social and economic cost of sustained drought in the USA.” However, they add that “these events are not unique to the twentieth century,” and they go on to describe three periods of widespread and persistent drought in the latter half of the nineteenth century—1856-1865 (the “Civil War” drought), 1870-1877 and 1890-1896—based on evidence obtained from proxy, historical, and instrumental data.
With respect to the first of these impressive mid- to late-nineteenth century droughts, Herweijer et al. say it “is likely to have had a profound ecological and cultural impact on the interior USA, with the persistence and severity of drought conditions in the Plains surpassing those of the infamous 1930s Dust Bowl drought.” In addition, they report that “drought conditions during the Civil War, 1870s and 1890s droughts were not restricted to the summer months, but existed year round, with a large signal in the winter and spring months.”
Taking a still longer look back in time, the three researchers cite the work of Cook and Krusic (2004), who constructed a North American Drought Atlas using hundreds of tree-ring records. This atlas reveals what Herweijer et al. describe as “a ‘Mediaeval Megadrought’ that occurred from AD 900 to AD 1300,” along with “an abrupt shift to wetter conditions after AD 1300, coinciding with the ‘Little Ice Age’, a time of globally cooler temperatures” that ultimately gave way to “a return to more drought-prone conditions beginning in the nineteenth century.”
The broad picture that emerges from these observations is one where the most severe North American droughts of the past millennium were associated with the globally warmer temperatures of the Medieval Warm Period plus the initial stage of the globally warmer Current Warm Period. Superimposed upon this low-frequency behavior, however, Herweijer et al. find evidence for a “linkage between a colder eastern equatorial Pacific and persistent North American drought over the last 1000 years,” noting further that “Rosby wave propagation from the cooler equatorial Pacific amplifies dry conditions over the USA.” In addition, they report that after using “published coral data for the last millennium to reconstruct a NINO 3.4 history,” they applied “the modern-day relationship between NINO 3.4 and North American drought … to recreate two of the severest Mediaeval ‘drought epochs’ in the western USA.”
But how is it that simultaneous global-scale warmth and regional-scale cold combine to produce the most severe North American droughts? One possible answer is variable solar activity. When solar activity is in an ascending mode, the globe as a whole warms; but at the same time, to quote from Herweijer et al.’s concluding sentence, increased irradiance typically “corresponds to a colder eastern equatorial Pacific and, by extension, increased drought occurrence in North America and other mid-latitude continental regions.”
An important implication of these observations is that the most severe North American droughts should occur during major multi-centennial global warm periods, as has in fact been observed to be the case. Since the greatest such droughts of the Current Warm Period have not yet approached the severity of those that occurred during the Medieval Warm Period, it seems a good bet that the global temperature of the Current Warm Period is not yet as high as the global temperature that prevailed throughout the Medieval Warm Period.
Central United States
Starting at the U.S.-Canadian border and working our way south, we begin with the study of Fritz et al. (2000), who utilized data derived from sediment cores retrieved from three North Dakota lakes to reconstruct a 2,000-year history of drought in this portion of the Northern Great Plains. This work suggested, in their words, “that droughts equal or greater in magnitude to those of the Dust Bowl period were a common occurrence during the last 2000 years.”
Also working in the Northern Great Plains, but extending down into South Dakota, Shapley et al. (2005) developed a 1,000-year hydroclimate reconstruction from local bur oak tree-ring records and various lake sediment cores. Based on this record, they determined that prior to 1800, “droughts tended towards greater persistence than during the past two centuries,” suggesting that droughts of the region became shorter-lived as opposed to longer-lasting as the earth gradually recovered from the cold temperatures of the Little Ice Age.
The above observations are significant because the United States’ Northern Great Plains is an important agricultural region, providing a significant source of grain for both local and international consumption. However, the region is susceptible to periodic extreme droughts that tend to persist longer than those in any other part of the country (Karl et al., 1987; Soule, 1992). Because of this fact, Laird et al. (1998) examined the region’s historical record of drought in an attempt to establish a baseline of natural drought variability that could help in attempts to determine if current and future droughts might be anthropogenically influenced.
Working with a high-resolution sediment core obtained from Moon Lake, North Dakota, which provided a sub-decadal record of salinity (drought) over the past 2,300 years, Laird et al. discovered that the U.S. Northern Great Plains were relatively wet during the final 750 years of this period. Throughout the 1,550 prior years, they determined that “recurring severe droughts were more the norm,” and that they were “of much greater intensity and duration than any in the 20th century,” including the great Dust Bowl event of the 1930s. There were, as they put it, “no modern equivalents” to Northern Great Plains droughts experienced prior to AD 1200, which means the human presence has not led to unusual drought conditions in this part of the world.
Continuing our southward trek, we encounter the work of Forman et al. (2005), who note that “periods of dune reactivation reflect sustained moisture deficits for years to decades and reflect broader environmental change with diminished surface- and ground-water resources.” This observation prompted them to focus on “the largest dune system in North America, the Nebraska Sand Hills,” where they utilized “recent advances in optically stimulated luminescence dating (Murray and Wintle, 2000) to improve chronologic control on the timing of dune reactivation.” They also linked landscape response to drought over the past 1,500 years to tree-ring records of aridity.
Forman et al. identified six major aeolian depositional events in the past 1,500 years, all but one of which (the 1930s “Dust Bowl” drought) occurred prior to the twentieth century. Moving backwards in time from the Dust Bowl, the next three major events occurred during the depths of the Little Ice Age, the next one near the Little Ice Age’s inception, and the earliest one near the end of the Dark Ages Cold Period. As for how the earlier droughts compare with those of the past century, the researchers say the 1930s drought (the twentieth century’s worst depositional event) was less severe than the others, especially the one that has come to be known as the sixteenth century megadrought. Forman et al. thus conclude that the aeolian landforms they studied “are clear indicators of climate variability beyond twentieth century norms, and signify droughts of greater severity and persistence than thus far instrumentally recorded.”
In a study that covered the entirety of the U.S. Great Plains, Daniels and Knox (2005) analyzed the alluvial stratigraphic evidence for an episode of major channel incision in tributaries of the upper Republican River that occurred between 1,100 and 800 years ago, after which they compared their findings with proxy drought records from 28 other locations throughout the Great Plains and surrounding regions. This work revealed that channel incision in the Republican River between about AD 900 and 1200 was well correlated with a multi-centennial episode of widespread drought, which in the words of Daniels and Knox, “coincides with the globally recognized Medieval Warm Period.” Of great interest, however, is the fact that modern twentieth century warming has not led to a repeat of those widespread drought conditions.
Working in pretty much the same area some seven years earlier, Woodhouse and Overpeck (1998) reviewed what we know about the frequency and severity of drought in the central United States over the past two thousand years based upon empirical evidence of drought from various proxy indicators. Their study indicated the presence of numerous “multidecadal- to century-scale droughts,” leading them to conclude that “twentieth-century droughts are not representative of the full range of drought variability that has occurred over the last 2000 years.” In addition, they noted that the twentieth century was characterized by droughts of “moderate severity and comparatively short duration, relative to the full range of past drought variability.”
With respect to the causes of drought, Woodhouse and Overpeck suggest a number of different possibilities that either directly or indirectly induce changes in atmospheric circulation and moisture transport. However, they caution that “the causes of droughts with durations of years (i.e., the 1930s) to decades or centuries (i.e., paleodroughts) are not well understood.” They conclude that “the full range of past natural drought variability, deduced from a comprehensive review of the paleoclimatic literature, suggests that droughts more severe than those of the 1930s and 1950s are likely to occur in the future,” whatever the air’s CO2 concentration or temperature might.
Mauget (2004) looked for what he called “initial clues” to the commencement of the great drying of the U.S. Heartland that had been predicted to occur in response to CO2-induced global warming by Manabe and Wetherald (1987), Rind et al. (1990), Rosenzweig and Hillel (1993), and Manabe et al. (2004), which Mauget reasoned would be apparent in the observational streamflow record of the region. In this endeavor, he employed data obtained from the archives of the U.S. Geological Survey’s Hydro-Climatic Data Network, which come from 42 stations covering the central third of the United States that stretch from the Canadian border on the north to the Gulf of Mexico on the south, with the most dense coverage being found within the U.S. Corn Belt.
Mauget reports finding “an overall pattern of low flow periods before 1972, and high flow periods occurring over time windows beginning after 1969.” Of the 42 stations’ high flow periods, he says that “34 occur during 1969-1998, with 25 of those periods ending in either 1997 or 1998,” and that “of those 25 stations 21 are situated in the key agricultural region known as the Corn Belt.” He also reports that “among most of the stations in the western portions of the Corn Belt during the 1980s and 1990s there is an unprecedented tendency toward extended periods of daily high flow conditions, which lead to marked increases in the mean annual frequency of hydrological surplus conditions relative to previous years.” What is more, he notes that “in 15 of the 18 Corn Belt gage stations considered here at daily resolution, a more than 50 percent reduction in the mean annual incidence of hydrological drought conditions is evident during those periods.” Last, Mauget reports that “the gage station associated with the largest watershed area—the Mississippi at Vicksburg—shows more than a doubling of the mean annual frequency of hydrological surplus days during its 1973-1998 high flow period relative to previous years, and more than a 50% reduction in the mean annual incidence of hydrological drought condition.”
In summarizing his findings, Mauget states that the overall pattern of climate variation “is that of a reduced tendency to hydrological drought and an increased incidence of hydrological surplus over the Corn Belt and most of the Mississippi River basin during the closing decades of the 20th century,” noting further that “some of the most striking evidence of a transition to wetter conditions in the streamflow analyses is found among streams and rivers situated along the Corn Belt’s climatologically drier western edge.”
Mauget states that the streamflow data “suggest a fundamental climate shift, as the most significant incidence of high ranked annual flow was found over relatively long time scales at the end of the data record.” In other words, the shift is away from the droughty conditions predicted by the IPCC to result from CO2-induced global warming in this important agricultural region of the United States.
Eastern United States
Cronin et al. (2000) studied the salinity gradient across sediment cores from Chesapeake Bay, the largest estuary in the United Sates, in an effort to examine precipitation variability in the surrounding watershed over the past millennium. Their work revealed the existence of a high degree of decadal and multidecadal variability between wet and dry conditions throughout the 1,000-year record, where regional precipitation totals fluctuated by 25 to 30 percent, often in “extremely rapid [shifts] occurring over about a decade.” In addition, precipitation over the past two centuries of the record was found to be generally greater than it was during the previous eight centuries, with the exception of the Medieval Warm Period (AD 1250-1350) when the [local] climate was found to be “extremely wet.” Equally significant was the 10 researchers’ finding that the region had experienced several “mega-droughts” that had lasted for 60 to 70 years, several of which they judged to have been “more severe than twentieth century droughts.”
Building upon the work of Cronin et al. were Willard et al. (2003), who examined the last 2,300 years of the Holocene record of Chesapeake Bay and the adjacent terrestrial ecosystem “through the study of fossil dinoflagellate cysts and pollen from sediment cores.” In doing so, they found that “several dry periods ranging from decades to centuries in duration are evident in Chesapeake Bay records.” The first of these periods of lower-than-average precipitation (200 BC-AD 300) occurred during the latter part of the Roman Warm Period, while the next such period (~AD 800-1200), according to Willard et al., “corresponds to the ‘Medieval Warm Period’.” In addition, they identified several decadal-scale dry intervals that spanned the years AD 1320-1400 and 1525-1650.
In discussing their findings, Willard et al. note that “mid-Atlantic dry periods generally correspond to central and southwestern USA ‘megadroughts’, which are described by Woodhouse and Overpeck (1998) as major droughts of decadal or more duration that probably exceeded twentieth-century droughts in severity.” Emphasizing this important point, they additionally indicate that “droughts in the late sixteenth century that lasted several decades, and those in the ‘Medieval Warm Period’ and between ~AD 50 and AD 350 spanning a century or more have been indicated by Great Plains tree-ring (Stahle et al., 1985; Stahle and Cleaveland, 1994), lacustrine diatom and ostracode (Fritz et al., 2000; Laird et al., 1996a, 1996b) and detrital clastic records (Dean, 1997).” Their work in the eastern United States, together with the work of other researchers in still other parts of the country, demonstrates that twentieth century global warming has not led to the occurrence of unusually strong wet or dry periods.
Quiring (2004) introduced his study of the subject by describing the drought of 2001-2002, which had produced anomalously dry conditions along most of the east coast of the U.S., including severe drought conditions from New Jersey to northern Florida that forced 13 states to ration water. Shortly after the drought began to subside in October 2002, however, moist conditions returned and persisted for about a year, producing the wettest growing-season of the instrumental record. These observations, in Quiring’s words, “raise some interesting questions,” including the one we are considering here. As he phrased the call to inquiry, “are moisture conditions in this region becoming more variable?”
Using an 800-year tree-ring-based reconstruction of the Palmer Hydrological Drought Index to address this question, Quiring documented the frequency, severity, and duration of growing-season moisture anomalies in the southern mid-Atlantic region of the United States. Among other things, this work revealed, in Quiring’s words, that “conditions during the 18th century were much wetter than they are today, and the droughts that occurred during the sixteenth century tended to be both longer and more severe.” He concluded that “the recent growing-season moisture anomalies that occurred during 2002 and 2003 can only be considered rare events if they are evaluated with respect to the relatively short instrumental record (1895-2003),” for when compared to the 800-year reconstructed record, he notes that “neither of these events is particularly unusual.” In addition, Quiring reports that “although climate models predict decreases in summer precipitation and significant increases in the frequency and duration of extreme droughts, the data indicate that growing-season moisture conditions during the 20th century (and even the last 19 years) appear to be near normal (well within the range of natural climate variability) when compared to the 800-year record.”
Western United States
Is there evidence of more severe and longer-lasting droughts in the western United States? We begin our journey of inquiry just below Canada, in the U.S. Pacific Northwest, from whence we gradually wend our way to the U.S./Mexico border.
Knapp et al. (2002) created a 500-year history of severe single-year Pacific Northwest droughts from a study of 18 western juniper tree-ring chronologies that they used to identify what they call extreme Climatic Pointer Years or CPYs, which are indicative of severe single-year droughts. As they describe it, this procedure revealed that “widespread and extreme CPYs were concentrated in the 16th and early part of the 17th centuries,” while “both the 18th and 19th centuries were largely characterized by a paucity of drought events that were severe and widespread.” Thereafter, however, they say that “CPYs became more numerous during the 20th century,” although the number of twentieth century extreme CPYs (26) was still substantially less than the mean of the number of sixteenth and seventeenth century extreme CPYs (38), when the planet was colder. The data of this study fail to support the IPCC’s claim that global warming increases the frequency of severe droughts.
Gedalof et al. (2004) used a network of 32 drought-sensitive tree-ring chronologies to reconstruct mean water-year flow on the Columbia River at The Dales in Oregon since 1750. This study of the second-largest drainage basin in the United States is stated by them to have been done “for the purpose of assessing the representativeness of recent observations, especially with respect to low frequency changes and extreme events.” When finished, it revealed, in their words, that “persistent low flows during the 1840s were probably the most severe of the past 250 years,” and that “the drought of the 1930s is probably the second most severe.”
More recent droughts, in the words of the researchers, “have led to conflicts among uses (e.g., hydroelectric production versus protecting salmon runs), increased costs to end users (notably municipal power users), and in some cases the total loss of access to water (in particular junior water rights holders in the agricultural sector).” Nevertheless, they say that “these recent droughts were not exceptional in the context of the last 250 years and were of shorter duration than many past events.” In fact, they say, “the period from 1950 to 1987 is anomalous in the context of this record for having no notable multiyear drought events.”
Working in the Bighorn Basin of north-central Wyoming and south-central Montana, Gray et al. (2004a) used cores and cross sections from 79 Douglas fir and limber pine trees at four different sites to develop a proxy for annual precipitation spanning the period AD 1260-1998. This reconstruction, in their words, “exhibits considerable nonstationarity, and the instrumental era (post-1900) in particular fails to capture the full range of precipitation variability experienced in the past ~750 years.” More specifically, they say that “both single-year and decadal-scale dry events were more severe before 1900,” and that “dry spells in the late thirteenth and sixteenth centuries surpass both [the] magnitude and duration of any droughts in the Bighorn Basin after 1900.” They say that “single- and multi-year droughts regularly surpassed the severity and magnitude of the ‘worst-case scenarios’ presented by the 1930s and 1950s droughts.” If twentieth century global warming had any effect at all on Bighorn Basin precipitation, it was to make it less extreme rather than more extreme.
Moving further south, Benson et al. (2002) developed continuous high-resolution δ18O records from cored sediments of Pyramid Lake, Nevada, which they used to help construct a 7,600-year history of droughts throughout the surrounding region. Oscillations in the hydrologic balance that were evident in this record occurred, on average, about every 150 years, but with significant variability. Over the most recent 2,740 years, for example, intervals between droughts ranged from 80 to 230 years; while drought durations ranged from 20 to 100 years, with some of the larger ones forcing mass migrations of indigenous peoples from lands that could no longer support them. In contrast, historical droughts typically have lasted less than a decade.
In another study based on sediment cores extracted from Pyramid Lake, Nevada, Mensing et al. (2004) analyzed pollen and algal microfossils deposited there over the prior 7,630 years that allowed them to infer the hydrologic history of the area over that time period. Their results indicated that “sometime after 3430 but before 2750 cal yr B.P., climate became cool and wet,” but, paradoxically, that “the past 2500 yr have been marked by recurring persistent droughts.” The longest of these droughts, according to them, “occurred between 2500 and 2000 cal yr B.P.,” while others occurred “between 1500 and 1250, 800 and 725, and 600 and 450 cal yr B.P,” with none recorded in more recent warmer times.
The researchers also note that “the timing and magnitude of droughts identified in the pollen record compares favorably with previously published δ18O data from Pyramid Lake” and with “the ages of submerged rooted stumps in the Eastern Sierra Nevada and woodrat midden data from central Nevada.” Noting that Bond et al. (2001) “found that over the past 12,000 yr, decreases in [North Atlantic] drift ice abundance corresponded to increased solar output,” they report that when they “compared the pollen record of droughts from Pyramid Lake with the stacked petrologic record of North Atlantic drift ice … nearly every occurrence of a shift from ice maxima (reduced solar output) to ice minima (increased solar output) corresponded with a period of prolonged drought in the Pyramid Lake record.” As a result, Mensing et al. concluded that “changes in solar irradiance may be a possible mechanism influencing century-scale drought in the western Great Basin [of the United States].”
Only a state away, Gray et al. (2004b) used samples from 107 piñon pines at four different sites to develop a proxy record of annual precipitation spanning the AD 1226- 2001 interval for the Uinta Basin watershed of northeastern Utah. This effort revealed, in their words, that “single-year dry events before the instrumental period tended to be more severe than those after 1900,” and that decadal-scale dry events were longer and more severe prior to 1900 as well. In particular, they found that “dry events in the late 13th, 16th, and 18th Centuries surpass the magnitude and duration of droughts seen in the Uinta Basin after 1900.”
At the other end of the moisture spectrum, Gray et al. report that the twentieth century was host to two of the strongest wet intervals (1938-1952 and 1965-1987), although these two periods were only the seventh and second most intense wet regimes, respectively, of the entire record. Hence, it would appear that in conjunction with twentieth century global warming, precipitation extremes (both high and low) within northeastern Utah’s Uinta Basin have become more attenuated as opposed to more amplified.
Working in the central and southern Rocky Mountains, Gray et al. (2003) examined 15 tree ring-width chronologies that had been used in previous reconstructions of drought for evidence of low-frequency variations in five regional composite precipitation histories. In doing so, they found that “strong multidecadal phasing of moisture variation was present in all regions during the late 16th-century megadrought,” and that “oscillatory modes in the 30-70 year domain persisted until the mid-19th century in two regions, and wet-dry cycles were apparently synchronous at some sites until the 1950s drought.” They thus speculate that “severe drought conditions across consecutive seasons and years in the central and southern Rockies may ensue from coupling of the cold phase Pacific Decadal Oscillation with the warm phase Atlantic Multidecadal Oscillation,” which is something they envision as having happened in both the severe 1950s drought and the late sixteenth century megadrought. Hence, there is reason to believe that episodes of extreme dryness in this part of the country may be driven in part by naturally recurring climate “regime shifts” in the Pacific and Atlantic Oceans.
Hidalgo et al. (2000) used a new form of principal components analysis to reconstruct a history of streamflow in the Upper Colorado River Basin based on information obtained from tree-ring data, after which they compared their results to those of Stockton and Jacoby (1976). In doing so, they found the two approaches to yield similar results, except that Hidalgo et al.’s approach responded with more intensity to periods of below-average streamflow or regional drought. Hence, it was easier for them to determine there has been “a near-centennial return period of extreme drought events in this region,” going all the way back to the early 1500s. It is reasonable to assume that if such an extreme drought were to commence today, it would not be related to either the air’s CO2 content or its temperature.
Woodhouse et al. (2006) also generated updated proxy reconstructions of water-year streamflow for the Upper Colorado River Basin, based on four key gauges (Green River at Green River, Utah; Colorado near Cisco, Utah; San Juan near Bluff, Utah; and Colorado at Lees Ferry, Arizona) and using an expanded tree-ring network and longer calibration records than in previous efforts. The results of this program indicated that the major drought of 2000-2004, “as measured by 5-year running means of water-year total flow at Lees Ferry … is not without precedence in the tree ring record,” and that “average reconstructed annual flow for the period 1844-1848 was lower.” They also report that “two additional periods, in the early 1500s and early 1600s, have a 25% or greater chance of being as dry as 1999-2004,” and that six other periods “have a 10% or greater chance of being drier.” In addition, their work revealed that “longer duration droughts have occurred in the past,” and that “the Lees Ferry reconstruction contains one sequence each of six, eight, and eleven consecutive years with flows below the 1906-1995 average.”
“Overall,” in the words of the three researchers, “these analyses demonstrate that severe, sustained droughts are a defining feature of Upper Colorado River hydroclimate.” In fact, they conclude from their work that “droughts more severe than any 20th to 21st century event occurred in the past,” meaning the preceding few centuries.
Moving closer still to the U.S. border with Mexico, Ni et al. (2002) developed a 1,000-year history of cool-season (November-April) precipitation for each climate division of Arizona and New Mexico from a network of 19 tree-ring chronologies. They determined that “sustained dry periods comparable to the 1950s drought” occurred in “the late 1000s, the mid 1100s, 1570-97, 1664-70, the 1740s, the 1770s, and the late 1800s.” They also note that although the 1950s drought was large in both scale and severity, “it only lasted from approximately 1950 to 1956,” whereas the sixteenth century mega-drought lasted more than four times longer.
With respect to the opposite of drought, Ni et al. report that “several wet periods comparable to the wet conditions seen in the early 1900s and after 1976” occurred in “1108-20, 1195-1204, 1330-45, the 1610s, and the early 1800s,” and they add that “the most persistent and extreme wet interval occurred in the 1330s.” Consequently, for the particular part of the world covered by Ni et al.’s study, there appears to be nothing unusual about the extremes of both wetness and dryness experienced during the twentieth century.
Also working in New Mexico, Rasmussen et al. (2006) derived a record of regional relative moisture from variations in the annual band thickness and mineralogy of two columnar stalagmites collected from Carlsbad Cavern and Hidden Cave in the Guadalupe Mountains near the New Mexico/Texas border. From this work they discovered that both records “suggest periods of dramatic precipitation variability over the last 3000 years, exhibiting large shifts unlike anything seen in the modern record.”
We come now to two papers that deal with the western United States as a whole. In the first, Cook et al. (2004) developed a 1,200-year history of drought for the western half of the country and adjacent parts of Canada and Mexico (hereafter the “West”), based on annually resolved tree-ring records of summer-season Palmer Drought Severity Index that were derived for 103 points on a 2.5° x 2.5° grid, 68 of which grid points (66 percent of them) possessed data that extended back to AD 800. This reconstruction, in the words of Cook et al., revealed “some remarkable earlier increases in aridity that dwarf [our italics] the comparatively short-duration current drought in the ‘West’.” Interestingly, they report that “the four driest epochs, centered on AD 936, 1034, 1150 and 1253, all occur during a ~400 year interval of overall elevated aridity from AD 900 to 1300,” which they say is “broadly consistent with the Medieval Warm Period.”
Commenting on their findings, the five scientists say “the overall coincidence between our megadrought epoch and the Medieval Warm Period suggests that anomalously warm climate conditions during that time may have contributed to the development of more frequent and persistent droughts in the ‘West’,” as well as the megadrought that was discovered by Rein et al. (2004) to have occurred in Peru at about the same time (AD 800-1250); and after citing nine other studies that provide independent evidence of drought during this time period for various sub-regions of the West, they warn that “any trend toward warmer temperatures in the future could lead to a serious long-term increase in aridity over western North America,” noting that “future droughts in the ‘West’ of similar duration to those seen prior to AD 1300 would be disastrous.”
While we agree with Cook et al.’s conclusion, we cannot help but note that the droughts that occurred during the Medieval Warm Period were obviously not CO2-induced. If the association between global warmth and drought in the western United States is robust, it suggests that current world temperatures are still well below those experienced during large segments of the Medieval Warm Period.
The last of the two papers to cover the western United States as a whole is that of Woodhouse (2004), who reports what is known about natural hydroclimatic variability throughout the region via descriptions of several major droughts that occurred there over the past three millennia, all but the last century of which had atmospheric CO2 concentrations that never varied by more than about 10 ppm from a mean value of 280 ppm.
For comparative purposes, Woodhouse begins by noting that “the most extensive U.S. droughts in the 20th century were the 1930s Dust Bowl and the 1950s droughts.” The first of these lasted “most of the decade of the 1930s” and “occurred in several waves,” while the latter “also occurred in several waves over the years 1951-1956.” More severe than either of these two droughts was what has come to be known as the Sixteenth Century Megadrought, which lasted from 1580 to 1600 and included northwestern Mexico in addition to the southwestern United States and the western Great Plains. Then there was what is simply called The Great Drought, which spanned the last quarter of the thirteenth century and was actually the last in a series of three thirteenth century droughts, the first of which may have been even more severe than the last. In addition, Woodhouse notes there was a period of remarkably sustained drought in the second half of the twelfth century.
It is evident from these observations, according to Woodhouse, that “the 20th century climate record contains only a subset of the range of natural climate variability in centuries-long and longer paleoclimatic records.” This subset, as it pertains to water shortage, does not approach the level of drought severity and duration experienced in prior centuries and millennia. A drought much more extreme than the most extreme droughts of the twentieth century would be required to propel the western United States and adjacent portions of Canada and Mexico into a truly unprecedented state of dryness.
Andreadis, K.M. and Lettenmaier, D.P. 2006. Trends in 20th century drought over the continental United States. Geophysical Research Letters 33: 10.1029/2006GL025711.
Benson, L., Kashgarian, M., Rye, R., Lund, S., Paillet, F., Smoot, J., Kester, C., Mensing, S., Meko, D. and Lindstrom, S. 2002. Holocene multidecadal and multicentennial droughts affecting Northern California and Nevada. Quaternary Science Reviews 21: 659-682.
Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffmann, S., Lotti-Bond, R., Hajdas, I. and Bonani, G. 2001. Persistent solar influence on North Atlantic climate during the Holocene. Science 294: 2130-2136.
Climate Change Reconsidered: Website of the Nongovernmental International Panel on Climate Change. http://www.nipccreport.org/archive/archive.html
Cook, E.R. and Krusic, P.J. 2004. North American Summer PDSI Reconstructions. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series # 2004-045. NOAA/NGDC Paleoclimatology Program.
Cook, E.R., Woodhouse, C., Eakin, C.M., Meko, D.M. and Stahle, D.W. 2004. Long-term aridity changes in the western United States. Sciencexpress.org / 7 October 2004.
Daniels, J.M. and Knox, J.C. 2005. Alluvial stratigraphic evidence for channel incision during the Mediaeval Warm Period on the central Great plains, USA. The Holocene 15: 736-747.
Dean, W.E. 1997. Rates, timing, and cyclicity of Holocene aeolian activity in north-central United States: evidence from varved lake sediments. Geology 25: 331-334.
Forman, S.L., Marin, L., Pierson, J., Gomez, J., Miller, G.H. and Webb, R.S. 2005. Aeolian sand depositional records from western Nebraska: landscape response to droughts in the past 1500 years. The Holocene 15: 973-981.
Fritz, S.C., Ito, E., Yu, Z., Laird, K.R. and Engstrom, D.R. 2000. Hydrologic variation in the Northern Great Plains during the last two millennia. Quaternary Research 53: 175-184.
Fye, F.K., Stahle, D.W. and Cook, E.R. 2003. Paleoclimatic analogs to twentieth-century moisture regimes across the United States. Bulletin of the American Meteorological Society 84: 901-909.
Gedalof, Z., Peterson, D.L. and Mantua, N.J. 2004. Columbia River flow and drought since 1750. Journal of the American Water Resources Association 40: 1579-1592.
Gray, S.T., Betancourt, J.L., Fastie, C.L. and Jackson, S.T. 2003. Patterns and sources of multidecadal oscillations in drought-sensitive tree-ring records from the central and southern Rocky Mountains. Geophysical Research Letters 30: 10.1029/2002GL016154.
Gray, S.T., Fastie, C.L., Jackson, S.T. and Betancourt, J.L. 2004a. Tree-ring-based reconstruction of precipitation in the Bighorn Basin, Wyoming, since 1260 A.D. Journal of Climate 17: 3855-3865.
Gray, S.T., Jackson, S.T. and Betancourt, J.L. 2004b. Tree-ring based reconstructions of interannual to decadal scale precipitation variability for northeastern Utah since 1226 A.D. Journal of the American Water Resources Association 40: 947-960.
Herweijer, C., Seager, R. and Cook, E.R. 2006. North American droughts of the mid to late nineteenth century: a history, simulation and implication for Mediaeval drought. The Holocene 16: 159-171.
Hidalgo, H.G., Piechota, T.C. and Dracup, J.A. 2000. Alternative principal components regression procedures for dendrohydrologic reconstructions. Water Resources Research 36: 3241-3249.
Karl, T., Quinlan, F. and Ezell, D.S. 1987. Drought termination and amelioration: its climatological probability. Journal of Climate and Applied Meteorology 26: 1198-1209.
Knapp, P.A., Grissino-Mayer, H.D. and Soule, P.T. 2002. Climatic regionalization and the spatio-temporal occurrence of extreme single-year drought events (1500-1998) in the interior Pacific Northwest, USA. Quaternary Research 58: 226-233.
Laird, K.R., Fritz, S.C., Grimm, E.C. and Mueller, P.G. 1996a. Century-scale paleoclimatic reconstruction from Moon Lake, a closed-basin lake in the northern Great Plains. Limnology and Oceanography 41: 890-902.
Laird, K.R., Fritz, S.C., Maasch, K.A. and Cumming, B.F. 1996b. Greater drought intensity and frequency before AD 1200 in the Northern Great Plains, USA. Nature 384: 552-554.
Laird, K.R., Fritz, S.C. and Cumming, B.F. 1998. A diatom-based reconstruction of drought intensity, duration, and frequency from Moon Lake, North Dakota: a sub-decadal record of the last 2300 years. Journal of Paleolimnology 19: 161-179.
Manabe, S., Milly, P.C.D. and Wetherald, R. 2004. Simulated long-term changes in river discharge and soil moisture due to global warming. Hydrological Sciences Journal 49: 625-642.
Manabe, S. and Wetherald, R.T. 1987. Large-scale changes of soil wetness induced by an increase in atmospheric carbon dioxide. Journal of the Atmospheric Sciences 44: 1211-1235.
Mauget, S.A. 2004. Low frequency streamflow regimes over the central United States: 1939-1998. Climatic Change 63: 121-144.
Mensing, S.A., Benson, L.V., Kashgarian, M. and Lund, S. 2004. A Holocene pollen record of persistent droughts from Pyramid Lake, Nevada, USA. Quaternary Research 62: 29-38.
Murray, A.S. and Wintle, A.G. 2000. Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 32: 57-73.
Ni, F., Cavazos, T., Hughes, M.K., Comrie, A.C. and Funkhouser, G. 2002. Cool-season precipitation in the southwestern USA since AD 1000: Comparison of linear and nonlinear techniques for reconstruction. International Journal of Climatology 22: 1645-1662.
Palmer, W.C. 1965. Meteorological Drought. Office of Climatology Research Paper 45. U.S. Weather Bureau, Washington, DC, USA.
Quiring, S.M. 2004. Growing-season moisture variability in the eastern USA during the last 800 years. Climate Research 27: 9-17.
Rasmussen, J.B.T., Polyak, V.J. and Asmerom, Y. 2006. Evidence for Pacific-modulated precipitation variability during the late Holocene from the southwestern USA. Geophysical Research Letters 33: 10.1029/2006GL025714.
Rein, B., Luckge, A. and Sirocko, F. 2004. A major Holocene ENSO anomaly during the Medieval period. Geophysical Research Letters 31: 10.1029/2004GL020161. Rind, D., Goldberg, R., Hansen, J., Rosenzweig, C. and Ruedy, R. 1990. Potential evapotranspiration and the likelihood of future drought. Journal of Geophysical Research 95: 9983-10004.
Rosenzweig, C. and Hillel, D. 1993. The Dust Bowl of the 1930s: Analog of greenhouse effect in the Great Plains? Journal of Environmental Quality 22: 9-22.
Shapley, M.D., Johnson, W.C., Engstrom, D.R. and Osterkamp, W.R. 2005. Late-Holocene flooding and drought in the Northern Great Plains, USA, reconstructed from tree rings, lake sediments and ancient shorelines. The Holocene 15: 29-41.
Soule, P.T. 1992. Spatial patterns of drought frequency and duration in the contiguous USA based on multiple drought event definitions. International Journal of Climatology 12: 11-24.
Stahle, D.W. and Cleaveland, M.K. 1994. Tree-ring reconstructed rainfall over the southeastern U.S.A. during the Medieval Warm Period and Little Ice Age. Climatic Change 26: 199-212.
Stahle, D.W., Cleaveland, M.K. and Hehr, J.G. 1985. A 450-year drought reconstruction for Arkansas, United States. Nature 316: 530-532.
Stahle, D.W., Cook, E.R., Cleaveland, M.K, Therrell, M.D., Meko, D.M., Grissino-Mayer, H.D., Watson, E. and Luckman, B.H. 2000. Tree-ring data document 16th century megadrought over North America. EOS: Transactions, American Geophysical Union 81: 121, 125.
Stockton, C.W. and Jacoby Jr., G.C. 1976. Long-term surface-water supply and streamflow trends in the Upper Colorado River Basin based on tree-ring analysis. Lake Powell Research Project Bulletin 18, Institute of Geophysics and Planetary Physics, University of California, Los Angeles.
Van der Schrier, G., Briffa, K.R., Osborn, T.J. and Cook, E.R. 2006. Summer moisture availability across North America. Journal of Geophysical Research 111: 10.1029/2005JD006745.
Wells, N., Goddard, S. and Hayes, M.J. 2004. A self-calibrating Palmer drought severity index. Journal of Climate 17: 2335-2351.
Willard, D.A., Cronin, T.M. and Verardo, S. 2003. Late-Holocene climate and ecosystem history from Chesapeake Bay sediment cores, USA. The Holocene 13: 201-214.
Woodhouse, C.A. and Overpeck, J.T. 1998. 2000 years of drought variability in the central United States. Bulletin of the American Meteorological Society 79: 2693-2714.
Woodhouse, C.A. 2004. A paleo perspective on hydroclimatic variability in the western United States. Aquatic Sciences 66: 346-356.
Woodhouse, C.A., Gray, S.T. and Meko, D.M. 2006. Updated streamflow reconstructions for the Upper Colorado River Basin. Water Resources Research 42: 10.1029/2005WR004455.