10. The Little Ice Age: Megafloods and Climate Swings

TEN

The Little Ice Age

MEGAFLOODS AND CLIMATE SWINGS

The lessons of the Little Ice Age are twofold. First, climate change does not come in gentle, easy stages. It comes in sudden shifts from one regime to another—shifts whose causes are unknown to us and whose direction is beyond our control. Second, climate will have its sway in human events. Its influence may be profound, occasionally even decisive. The Little Ice Age is a chronicle of human vulnerability in the face of sudden climate change. In our own ways, despite our air-conditioned cars and computer-controlled irrigation systems, we are no less vulnerable today. There is no doubt that we will adapt again, or that the price, as always, will be high.

BRIAN FAGAN, The Little Ice Age

SAN FRANCISCO BAY

THE DEEP DROUGHTS OF THE Medieval Climate Anomaly eventually drew to a close around AD 1400. For much of the 150 years prior to this date, extreme flooding was relatively rare. In California, people who lived in the broad Central Valley and around the San Francisco Bay would have seen irregular rainfall for hundreds of years, with the winters often failing to deliver the big storms that filled the lakes and fed the mountain snowpack. Around the bay, the mounded villages were empty most years; since the drought had begun, the creeks flowing to the marshlands dried early in the summer, forcing the people to make long treks for freshwater or, if their camps were inland near the larger creeks, long foraging expeditions to the mudflats in the bay to collect mussels and other shellfish. The native populations throughout the region suffered, some perished, and some managed to adapt to the dry conditions.

The pattern of weather began to change, however, and those who paid attention would have noticed the series of cold storms that now began the wet season, dumping rain on the coastal settlements and throughout the Central Valley and the lower mountain slopes. Winters were bringing more storms that were colder, and in the Sierra Nevada the snowpack grew thicker and lower on the mountain slopes than it had been in human memory. This period was known as the Little Ice Age, and, elsewhere around the globe, rivers and canals froze throughout Europe, and valley glaciers grew larger in northern and central Europe, New Zealand, and Alaska. The Thames River in London froze at least eleven times in the seventeenth century. Lower summer temperatures throughout Europe decreased the growing season by several weeks, causing crop failures in Scotland, Norway, and Switzerland and leading to widespread famine. In Norway, there is evidence for more frequent landslides, avalanches, and floods, and, in the Alps, expanding glaciers occasionally crossed valley floors, damming streams and forming glacial lakes. These lakes would on occasion break through the ice dams holding them back, flooding the valleys below.

In California, some of the storms originated in tropical regions of the Pacific, delivering copious amounts of warm rain that led to flooding. These storms would have swept over the Coast Ranges, across the Central Valley, and up into the Sierra Nevada, delivering heavy rain. The rain and the rapidly melting snow would have caused small creeks and rivers to swell into raging torrents, ripping up vegetation and soils in the mountains, and turning California’s Central Valley into a vast inland sea.

Many proxy climate records from throughout California contain evidence that such extreme winter floods occurred between 900 and 150 years ago. Tidal marshes around the San Francisco Bay contain buried evidence of these events in their sediments. Normally, the inflowing river waters spread across the marshes, depositing only a small amount of the finest sediments—clays and silts. Floodwaters, however, carry larger particles in their higher energy flows, depositing a layer of sand and silt on the marshes. Marsh sediment cores reveal that one such layer was deposited on the marshes around AD 1100, and two others around AD 1400 and AD 1650.

Sediment cores taken from beneath the San Francisco Bay itself provide more evidence of the flooding: a gap or hiatus in the sedimentary sequence suggests erosion of hundreds of years’ worth of accumulated sediments from the bay floor. The date for this hiatus matches the earliest of the sand layers in the marsh cores. That such an event is recorded in San Francisco Bay estuarine and marsh sediments suggests a regionally significant flood, affecting almost half of the state—the drainage area of the bay.

Though considered “young” in geological terms, these large floods were not documented by humans living at the time and left no evidence in the archaeological record. To understand them from our standpoint in the twenty-first century requires the same kind of detective work described in earlier chapters to uncover older climate patterns. Below, we describe several key examples of evidence for megafloods throughout the state of California.

SOUTHERN CALIFORNIA FLOOD EVIDENCE

Some of the most compelling evidence of large floods in California has been found off the southern coast, once again in the favorable environment of the Santa Barbara Basin. The location and conditions of that basin have combined to produce one of the most detailed and complete records of climate and environmental change anywhere in the world’s coastal oceans, including a record of runoff into this coastal environment during extreme storms.

Arndt Schimmelmann, a marine geochemist at Indiana University, has analyzed the environmental history contained within the Santa Barbara Basin’s deep sediments. He first looked at X-ray images of sediment cores, which elucidate differences in the density of the sediments; annual sediment layers show clearly as alternating light and dark stripes. Schimmelmann found that some layers were much thicker than others. When the core was cut open, he found that the thicker layers were olive-grey, and, under the microscope, he could see that these olive-grey grains included tiny angular lithic fragments. These particular sediments were probably eroded off the hill slopes in Southern California and subsequently washed into local rivers, which transported them to the ocean floor. Schimmelmann concluded that these sediment deposits must have resulted from enormous flood events—“megafloods,” as he called them.

Schimmelmann was intrigued by what he saw in the Santa Barbara Basin sediments, and he began looking for evidence of flood events in other records of past environmental and climate change that occurred at the same time. He found that other researchers in the coastal region had described evidence of large floods contained in sediments from lakebeds, floodplains, and submarine basins along the coastline. These are typical settings for finding traces of ancient floods. As floodwaters rage down slopes and across the landscape, they scour the hills, picking up clay, silt, sand, and even gravel and carrying them entrained in the swollen current. Eventually the velocity of the rivers decreases upon reaching sea level, and the rivers release their burden: first the larger gravels, then the sands, and finally the silts and clays. Nature rebuilds after these events, and in time the flood deposits are themselves buried beneath other sediments.

FIGURE 28. Flood sediment layers for the past 2,000 years from Santa Barbara Basin cores. (Figure courtesy of Arndt Schimmelmann, University of Indiana, redrawn by B. Lynn Ingram.)

Ultimately, Schimmelmann found evidence for six megafloods over the past 2,000 years in Santa Barbara Basin sediments that would have affected the entire region. The flood layers could be precisely dated because the sediments are composed of annual layers, analogous to the growth rings in trees. Based on the ages of these megafloods, they appear to have recurred on average every 200 years: AD 212, 440, 603, 1029, 1418, and 1605. During the period straddling the Medieval Climate Anomaly, two cycles were skipped, with 400 years after the floods that occurred in AD 603 and AD 1029.

Schimmelmann reasoned that the thickness of the flood layers is proportional to the size of the event. The thickest layers were deposited before the Medieval Climate Anomaly (AD 212, 440, and 603) and during the Little Ice Age (AD 1418 and 1605). The thinnest flood layer occurred during the Medieval Climate Anomaly, in AD 1029 (see figure 28). These results suggest that periodic megafloods may be a normal part of the larger cycle of climate in this region, and the Medieval Climate Anomaly may have been anomalous in part for its absence of such floods.

The largest flood to hit the Santa Barbara Basin over the past two millennia occurred in AD 1605, leaving a sediment layer two inches thick. Almost as large, the AD 440 and 1418 floods each had a thickness of one and a quarter inches. For perspective, consider that the two major floods that struck the region during the twentieth century—the floods of 1958 and 1969—left sediment layers that were only 0.24 and 0.08 inches thick, respectively. In its entire history as a state, California has yet to experience one of the megafloods that Schimmelmann found to be a repeating part of the region’s natural climate. Schimmelmann points out that the periodicity reflected in his record suggests the region is due for another one soon; it has been more than four hundred years since the last megaflood, which occurred in AD 1605.

THE MOJAVE DESERT UNDER WATER

For most of the Holocene, the Mojave Desert of southeastern California has remained one of the driest places in the West, with annual rainfall less than four inches per year in recent times. Only during the wettest years of the twentieth century (every fifteen to thirty years or so) did small ephemeral lakes form in the region, lasting just two to eighteen months before drying up. However, paleoclimate researchers have examined sediment cores extracted from a playa located in the Mojave and found evidence of past periods that were wet enough to form much larger and long-lived lakes. This playa, called Silver Lake, lies at the terminus of the Mojave River. During the late Pleistocene, Silver Lake was much larger but has remained mostly dry since about 8,000 years ago. Only during two periods since the late Pleistocene were conditions wet enough to create a lake that lasted several decades.

Cores taken from the Silver Lake playa reveal two layers of clay sediments deposited within the lake. Within those cores, the lake sediment layers lie between sediments deposited by the Mojave River on the desert floor. Radiocarbon dating of the lake sediments reveals two extremely wet periods in the past: the Neoglacial (about 3,600 years ago), and the Little Ice Age (about 400 years ago). A period of cool summer temperatures, with a greater than average number of storms per year—for decades to up to a century—would have been required to form these lakes in the Mojave.

NORTHERN CALIFORNIA FLOODS

The ominous history of ancient megafloods identified in Southern California is not unique. Researchers in the northern half of the state have found deposits suggesting floods of similar magnitude and frequency—from river valleys to the higher mountain ranges. One of these studies was located on the floodplain of the Sacramento River, the largest river in California and the one that drains much of the northern half of the state. Geographers Roger Byrne and Don Sullivan from the University of California, Berkeley, have found evidence for large floods in sediments from beneath oxbow (or arc-shaped) lakes on the Sacramento River floodplain of California’s Central Valley. They were not initially looking for evidence of megafloods; rather, they were evaluating how the burgeoning population in California after the Gold Rush had influenced the natural vegetation in the region. But as they examined their sediment cores, they found an unexpected legacy of major flood events.

To understand what Sullivan and Byrne discovered, it helps to visualize the setting of their research. Floodplains, by their nature, contain the most accurate picture of past flooding events. During a flood, the waters swell out of their channels and spread across the surrounding plain, depositing layers of sediment. These layers build on each other, flood after flood, over hundreds to thousands of years. The thickness of the layers yields information about the size of the floods. The lakes within the floodplain, though some are short-lived, contain records that can be interpreted both for past flood events and for the climate conditions under which the lakes formed.

The lake Sullivan and Byrne chose to core, Little Packer, began forming 800 years ago. Once a bend in the Sacramento River, Little Packer was abandoned by the meandering main channel and was left behind as an oxbow lake. This occasionally occurs when rivers meander in wide and gently sloping floodplains, as the Sacramento does in the Central Valley. Such abandoned lakes become recorders of large flood events that temporarily reunite them with their rivers. Sediment-laden floodwaters spill into them, and the sediment settles to the bottom, forming a distinct layer within the normal lake sediments.

Sullivan and Byrne took a series of sediment cores from beneath Little Packer Lake and brought them back to Byrne’s laboratory. During the routine X-ray analyses, they noticed some unusual bright layers in the cores that turned out to be rich in river-borne sand and silt. These denser sand layers shone white next to the darker, more organic layers produced by algae growing in the lake during normal (non-flood) years. Sullivan and Byrne reasoned that, since the lake had no direct hydrologic connection to the river except during major floods, these deposits must have occurred when the Sacramento River overtopped its banks and carried suspended sediments into Little Packer Lake. Because larger floods transport more sediment, the thickness of the sand layers provides an approximate measure of the relative size of the flooding event. Sullivan and Byrne also compared the flood layer thicknesses to floods of known magnitude, such as the catastrophic 1861–62 event.

Eager to understand more about how these floods fit into California’s natural history, Byrne and Sullivan used radiocarbon analysis to date the cores and estimate the ages of the flood layers. Although these ages lack the precision of ages that are determined by counting annual varves in the Santa Barbara Basin, they can still provide an approximate measure of the flood frequencies. The researchers calculated that, over the past 800 years, floods at least the size of the 1861–62 event struck Northern California on average every 100 to 120 years.

This is bad news to state water planners, who have estimated that the 1861–62 flood was a “500- to 1,000-year event,” which was determined by plotting known flood magnitudes against their recurrence intervals from California’s Central Valley over the past century. This so-called flood frequency curve then allows hydrologists to extrapolate the recurrence intervals of much larger floods (like the 1861–62 event) using their estimated magnitudes. Because paleoflood records provide actual data about the magnitudes and recurrence intervals, they should be considered along with those estimated from the flood frequency curves. For instance, the research of Byrne and Sullivan revealed evidence of four even larger floods that dwarfed the 1861–62 event over the past 800 years. Like Schimmelmann with his results from the Santa Barbara Basin sediments, Byrne and Sullivan found evidence of megafloods recurring in the region approximately every 200 years.

Evidence of megafloods has also emerged from the far northwest corner of California. In that mountainous region, floods ravage the steep, narrow canyons in the Klamath and northern Coast Ranges, where flood debris tends to form thicker, coarser layers than are found in the broader and more gently sloping floodplains of the Central Valley. In those mountains, only the larger flood deposits are preserved, in contrast to the sequence of flood deposits that built up over time in the Sacramento Valley. This is due to the erosive power of these floods in the narrow canyons that destroys any evidence of older, smaller floods. Researchers at the U.S. Geological Survey discovered two enormous flood deposits preserved in this region, estimated to be larger than one of the most catastrophic historical floods in Northern California, which occurred in 1964. These two floods were dated by counting the growth rings of the oldest trees rooted in the flood deposits, and they occurred in approximately AD 1600 and 1750. A flood layer dated between AD 1750 and 1770 was also found in the Sacramento Valley by Byrne and Sullivan, and the AD 1600 flood deposit was likely the same as the AD 1605 events found in both the Sacramento Valley and the Santa Barbara Basin. The evidence for a 200-year recurrence interval for floods as large or larger than the 1861–62 event in California is growing, bringing more urgency to flood policy and planning efforts in the state.

THE AD 1605 MEGAFLOOD

The evidence for a megaflood occurring in AD 1605 has been found throughout California, including in the northwest, central, and southern regions as well as in San Francisco Bay, making this a truly extraordinary event even by “megaflood” standards. In the Little Packer Lake record from the Sacramento River basin, this flood was estimated to be at least 50 percent greater than any of the four 200-year events that Sullivan and Byrne detected in their cores. It must have been a catastrophe of gargantuan proportions, clearing hill slopes, drowning the Central Valley, and inundating most of the state before ultimately flowing through the Sacramento–San Joaquin Delta, through the San Francisco Bay, and out to the Pacific Ocean.

Sediment cores taken from beneath the southern San Francisco Bay contain oxygen isotope evidence that this time period was marked by very low salinity in the estuary, presumably from the high river inflows from the watershed. In addition, cores taken from the northern reach of the bay, closer to where rivers flow in through the delta, contain a sedimentary hiatus, suggesting that the flood was large enough to erode and wash away the sediments out to the Pacific Ocean.

Intrigued by the phenomenon of cataclysmic floods and their possible causes, Schimmelmann combed through the scientific literature and compiled a list of other global events that occurred at the same time as these megafloods. He found that, around AD 1605, at least two major volcanic eruptions occurred: the Huaynaputina volcano in Peru in 1600, and other volcanic evidence revealed in an Antarctic ice core. He also discovered numerous paleoclimate records worldwide showing that the years between 1600 and 1610 were unusually cold, particularly during the summer, with the summer of 1601 being the coldest in the Northern Hemisphere of the past millennium. These records were largely based on tree-ring records from California, Idaho, the Canadian Rockies, Subarctic Quebec, the Gulf of Alaska, New Zealand, Mongolia, and Norway. Ice-core records from both Greenland and Antarctica also suggest unusually cold conditions. But was there a connection between the volcanic eruptions, Northern Hemisphere cooling, and the California megafloods? We will return to this question in the next chapter, where we discuss possible causes of past climate change.

THE LITTLE ICE AGE AND THE WEST

These megafloods occurred during the Little Ice Age, an event that, as described above, left traces throughout the Northern hemisphere between the fifteenth and nineteenth centuries. In the western United States, glaciers also advanced in the Sierra Nevada, Cascades, and Rockies. Conditions became wetter in the West, and floods were larger and more frequent, as discussed above. In the Sierra Nevada, abundant moisture allowed trees to grow faster, putting on wider rings. East of the Sierra Nevada, Mono Lake rose to its highest level in centuries. In the Great Basin, Owens and Pyramid lakes were also high.

Forest fires in California were reduced during the Little Ice Age. For instance, though the Giant Sequoias in the southern Sierra Nevada bear scars from fires over the past millennia, those scar records show a marked decline in fire frequency after AD 1600. The records also closely track the summer values of the Palmer Drought Severity Index (PDSI), a measure of the moisture content of soil in the root zone. Tree-ring records contain a history of the PDSI. A comparison between the PDSI and the fire-scar records shows that three-fourths of the largest fire years occurred during the driest years. Mean summer temperatures, the PDSI, and fire frequencies all peaked about AD 1200 during the Medieval period, followed by a steady decline in these factors (suggesting a cooler, wetter climate with fewer fires) during the Little Ice Age (see figure 27B).

Additional evidence for increased wetness during the Little Ice Age has been found downstream of the Sierra Nevada, in the Central Valley. Tulare Lake, filled with increased river inflows and floodwaters, reached a high stand during the Little Ice Age. San Francisco Bay became fresher as river inflows increased. Vegetation records from the marsh sediment cores surrounding San Francisco Bay reveal a shift to a dominance of freshwater species, reflecting greater inflows from rivers swollen with heavy rainfall in the watershed.

CLIMATE SWINGS, WILDFIRES, AND DROUGHTS

Many of the paleoclimate records show increased variability during the Little Ice Age, with a generally wetter and cooler climate punctuated by episodes of drought—sometimes severe. Droughts have been detected in tree rings from the Sierra Nevada, salinity levels in San Francisco Bay, and lake levels of Mono, Pyramid, and Owens lakes. At Mono Lake, for instance, Larry Benson and his colleagues examined sediment cores using the oxygen isotopic ratios of the sedimentary carbonates to determine changes in the lake’s surface level. As discussed in chapters 5 and 7, when the inflow of water entering a closed-basin lake exceeds the rate of evaporation, the lake expands and the ratio of oxygen-18 to oxygen-16 is lower. When evaporation is the dominant process, however, the lake shrinks, and the lighter oxygen-16 water molecules evaporate more readily, leaving behind more oxygen-18 molecules in the lake. Benson and his colleagues reconstructed the relative changes in lake level based on the oxygen isotopic ratios in carbonates, and they have shown that the water level in Mono Lake dropped five times over the past 300 years, implying severe droughts. These dry periods lasted for several years, causing a significant drop in lake levels, centered on the dates AD 1710, 1770, 1820, 1850, and 1930.

The San Francisco Bay records also contain evidence of these periodic swings in climate seen in the Sierra during the Little Ice Age. Oxygen isotope measurements of sediments from the northern part of the estuary have enabled the authors to reconstruct the bay’s salinity (as discussed in chapters 5, 8, and 9). The results reveal multiple fluctuations in salinity over the centuries, exhibiting 55-, 90-, and 200-year cycles.

In the American Southwest, climate during the Little Ice Age appears to have been highly variable. Tree-ring studies from the Colorado Plateau show that Colorado River flows experienced major fluctuations during that period. Looking at reconstructed 25-year average flow conditions, researchers have found that it was a time of generally cooler conditions, with at least six wet periods marked by significantly high flows. However, there were also periods of significantly low river flows, including AD 1564–1600, AD 1844–48, and AD 1868–92. Over the past 450 years, the long-term annual average flow for the Colorado River was 14.6 million acre-feet. Flows during the identified dry periods were as low as 9.6 million acre-feet, only 66 percent of the long-term average.

Although these flows were low, they were not as low as during the Medieval Climate Anomaly. In particular, from AD 1130 to 1154, Colorado River flows were 85 percent lower than the twentieth-century mean. We will discuss this variability and its origins further in the next chapter.

THE DRY-WET CLIMATE KNOCKOUT

Some of the paleoclimate records reveal repeated patterns of dry and wet climate extremes, such as severe droughts immediately followed by catastrophic floods—or the opposite, wet periods followed by droughts. These combinations can wreak havoc on ecosystems, especially when accompanied by unusually large wildfires, soil erosion, and disease.

One example of a “dry-wet knockout” has been documented by Scott Mensing, Roger Byrne, and Joel Michaelsen, who analyzed the flux of large charcoal particles and pollen deposited in sediments cored from the Santa Barbara Basin in Southern California. Their analysis reveals that the largest fires—those often associated with warm, dry Santa Ana winds blowing from the northeast to the Pacific during the fall months—have repeatedly occurred in the first year of a drought that follows an extended wet period. Moreover, Mensing and his colleagues showed that between twenty and thirty large wildfires occurred over the past 560 years, the majority following unusually wet periods.

Examples of this pattern in the twentieth century include the wildfires of 1955 and 1964, which followed two of the wettest winters of the century. Investigators reason that vegetation grew prolifically during years with greater precipitation. When followed by unusually warm and dry summer conditions, the additional vegetation provided extra fuel for unusually large forest fires.

Periods of deep drought that are followed by heavy rain and flooding can also cause problems. This phenomenon has been observed both in the past century and in the more distant past. Prolonged, multiyear droughts followed by severe storms and massive flooding are nature’s version of a one-two climate punch. University of Nevada, Reno, dendroclimatologist Franco Biondi and his colleagues have studied these patterns and dubbed them “dry-wet knockouts” because the floodwaters often wash away hillsides that have been denuded of vegetation during extended drought, transporting huge volumes of sediment to coastal waters. There are many examples of extended droughts—including those that occurred during Medieval times—terminated by extremely wet years, which often lead to a chain-reaction of impacts on local ecosystems.

During multiyear droughts, mountain forests become dangerously dry, and summer lightning strikes spark wildfires that sweep through and engulf the forests. Subsequent heavy rains generate floods that can easily erode fire-charred slopes, washing massive amounts of sediment and charcoal downhill, eventually into streams and lakes. Once on the move, such sediments raise the bed-level of the rivers, further adding to flood potential. These sediment-laden waters can disrupt aquatic ecosystems downstream in ponds, lakes, estuaries, and even the coastal ocean.

Evidence of these past knockout floods can be found throughout the West. Tree stumps found in lakes in the southern and eastern Sierra Nevada are what remain of medieval trees that grew for over a century during the prolonged droughts of the Medieval Climate Anomaly and were suddenly drowned when the drought was terminated by heavy rains and huge floods. Downstream in the San Francisco Bay tidal marshes, sediment cores reveal evidence of large floods dating to around AD 1100, 1400, and 1650. These appear to have occurred during the interval between the Medieval megadroughts and immediately after the Medieval droughts. Similarly, in Yellowstone National Park, Wyoming, evidence of massive debris flows, caused by flash floods, has been dated to the end of the Medieval drought.

The greatest megaflood on record in the West over the past two millennia—the AD 1605 event described above—also followed on the heels of an extended dry period, which occurred at the end of the sixteenth century. And the deep drought during the mid-nineteenth century that peaked in 1860 was followed by the catastrophic 1861–62 flood. More recently, the drought of 1987–92 was followed by the unusually wet winter in 1993, leading to massive mudslides, landslides, and record flooding in some parts of the West.

THE COASTAL OCEAN DURING THE LITTLE ICE AGE

Because the climate in the West is strongly influenced by the temperature of the eastern Pacific Ocean, paleoceanographers—those who study past ocean conditions—have examined sediment cores from the coastal ocean for evidence of correlations with climate on land. One such researcher, John Barron of the U.S. Geological Survey, used the fossils of diatoms in the Santa Barbara Basin to reconstruct long records of ocean surface water conditions. Barron and his colleagues have shown that, whereas coastal surface water was generally cooler during the Medieval Climate Anomaly than today, the Little Ice Age coastal surface waters tended to be warmer, and more variable, than coastal waters during the Medieval period. Warm waters in the eastern Pacific are associated with El Niño, and, as we will discuss in the next chapter, a growing body of evidence, including coral records from the tropical Pacific Ocean, suggests that El Niño events were stronger and more frequent during the Little Ice Age, leading to wetter and more variable conditions in the West.

The droughts that punctuated this generally wetter period were also shown to be tied to coastal ocean conditions. Larry Benson and his research team compared the dry periods from Mono Lake in the eastern Sierra with sea surface temperature records from coastal California, and they were intrigued to find a close correlation between the droughts and the periods when coastal waters were relatively cool. As we discussed in chapter 4, one of the oceanic cycles that influences climate in the West is the Pacific Decadal Oscillation (PDO). This is an oscillation of water temperature in the North Pacific Ocean, with cooler surface water temperatures associated with the negative phase of the oscillation and warmer surface temperatures associated with the positive phase. Benson and his colleagues found that the drought periods in the Sierra that led to the lowering of the Mono Lake surface level occurred when the PDO was in a negative (cool) phase.

The evidence of past megafloods is a cause of concern for California. Even as we enter a period that is predicted to be warmer and drier, catastrophic flooding will continue to threaten the region. As more precipitation falls as rain instead of snow, flooding during the winter is predicted to inundate the region even more frequently. The deadly climate patterns that have occurred repeatedly in the past are likely to recur in the future.

Climate scientists are currently studying the causes of these past climate changes and weather extremes. Some of these causes are truly remote, such as variations in solar output, and others are close at hand, such as the coal-fired power plants that generate much of our electricity. Needless to say, the so-called anthropogenic, or human-caused, changers of climate are most readily modified—but are also controversial and incompletely understood. To comprehend the full range of climate variability and to better prepare ourselves for an uncertain climatic future, we need to use all the knowledge and tools at our disposal. In the next chapter, a review of some known cycles and oscillations of climate change in the past may suggest the scale on which we can expect such changes to recur.