SEVEN
The “Long Drought” of the Mid-Holocene
There are only a limited number of ways societies can respond to accumulated climatic stress: movement or social collaboration; muddling their way from crisis to crisis; decisive, centralized leadership on the part of a few individuals; or developing innovations that increase the carrying capacity of the land. The alternative to all these options is collapse.
BRIAN FAGAN, Floods, Famines, and Emperors
DURING THE MID-HOLOCENE, CLIMATE IN THE WEST shifted toward warmer, drier conditions. This climatic upheaval and prolonged drought forced many humans to leave their home terrain in search of water and food, especially in the Great Basin and what is known today as southeastern California. The inland areas were the hardest hit, and, in response, the archaeological evidence suggests that populations migrated to the coast, which offered a cooler climate, more moisture, and plentiful food.
Along the coast, the steady rise of sea level following the last ice age slowed after the massive inflow of glacial melt waters began to taper off. Coastal habitats began to expand, and coastal ecosystems flourished. Coastal groups took advantage of these abundant marine resources, establishing settlements along the Pacific Northwest coast, around present-day San Francisco Bay, and as far south as California’s Santa Barbara Basin and Channel Islands.
Coastal conditions also became more favorable for salmon migration, spawning, and population growth. Salmon soon became the single most important Native American food resource along the coast, attracting still more humans to the area.
TAHOE’S SUBMERGED TREES
Some of the most vivid evidence of the mid-Holocene drought comes from Lake Tahoe, the largest alpine lake in the United States. Appropriately enough, evidence of the prolonged mid-Holocene drought was discovered during a modern drought—the Great Dust Bowl drought of the twentieth century, which was California’s worst multiyear drought in recent times. Four years after that drought began in California in 1928, the water level of Lake Tahoe in the Sierra Nevada had dropped by fourteen inches, exposing a mysterious clustering of tree stumps sticking up from the water’s surface along the lake’s southern shore (see figure 17). These trees attracted the attention of Samuel Harding, a scientist at the University of California, Berkeley, who discovered that the trees were large, with trunks upwards of three feet in diameter, and that they appeared to be firmly rooted in the lake bottom.
FIGURE 17. Ancient tree stumps exposed in southern Lake Tahoe in the Sierra Nevada, California. (Photo courtesy of Derrick Kelly.)
Wanting to know more about these trees, Harding collected eleven core samples from their trunks. He reasoned that the trees had grown in this location for a long time to attain such sizes and, since they were now submerged in over twelve feet of water, that at some time in the past the lake level had been much lower. He counted 150 rings in one of the cores, which indicated that south Lake Tahoe had been dry and below the shoreline for well over a century. The site must then have been abruptly inundated with water, drowning the trees fast enough that they were left standing in their current upright positions.
Harding considered several possible explanations for the location (and elevation) of these trees. A large tectonic event might have lowered the southern shoreline where the trees were found or raised the outlet of the lake. Alternatively, an extreme climate event involving an extended dry spell might have caused the lake level to drop, allowing the trees to grow along the former shoreline. If the drought was followed by a return to wetter conditions, the lake would have risen, submerging and drowning the trees. The tree stumps were minimally decomposed, suggesting that, whatever the cause, the drowning had been rapid.
Harding found little evidence to support a sudden tectonic event; no visible signs of this could be seen in the sediments or landscape. Moreover, submerged tree stumps had been found in other nearby lakes, suggesting a more regional cause and leading Harding to ultimately favor the drought hypothesis. But, in the mid-1930s, he did not yet have the tools to date the trees, and he was forced to put aside the mystery of these tree stumps for a later time. It would be three decades before he was able to piece together the story.
In the years following Harding’s original discoveries, scientists found more tree stumps along the southern shore of Lake Tahoe, submerged at depths ranging from two to twelve feet below the modern water surface. Since Lake Tahoe is the largest and second-deepest alpine lake in North America, only a very prolonged and severe drought could have lowered its surface level enough to allow trees to establish and survive for centuries. But when did this devastating drought occur? Fortunately, a method of dating fossils and artifacts containing carbon (organic carbon or carbonate) by measuring their radioactive content was invented in the 1950s. Using this new technique, Harding himself had the gratification of finally dating the outer-most rings of the tree stumps to determine the time of their death. Sampling three of them, he found that the trees had been flooded and killed approximately 4,800 years ago, in the mid-Holocene.
RE-EMERGING TREE STUMPS
The late 1980s were marked by another historic drought: California’s second most severe drought on record. Lake Tahoe’s surface level dropped again to expose more tree stumps. This time, it was an archaeologist, Susan Lindstrom, who noticed the tops of trees sticking out of the water along Tahoe’s southern shore. Donning scuba gear, Lindstrom was able to find fifteen submerged tree stumps that had escaped Harding’s attention, some measuring up to three and a half feet in diameter and seven feet tall. The core samples and radiocarbon dates from this much larger population of trees refined and extended the boundaries of the mid-Holocene drought, moving its beginning to approximately 6,290 years ago and its ending to 4,840 years ago. These stumps, located deeper in the lake, showed that the lake level had dropped by even more than Harding originally thought—by more than twenty feet. Lindstrom has since located tree stumps in more places around the shores of Lake Tahoe, including Emerald Bay, Rubicon Point, and Baldwin Beach. Stumps of similar age have also been discovered in other nearby Sierra lakes, including Donner, Independence, and Fallen Leaf.
The lowered shoreline of Lake Tahoe in the late 1980s revealed other interesting traces of the past. Archaeologists discovered bedrock mortars containing grinding cups and milling slicks on the west shore, close to Tahoe City, providing evidence that humans lived there during the mid-Holocene. These artifacts and others from archaeological sites around the Tahoe Basin confirm that humans were able to survive in the region despite the prolonged drought.
PYRAMID LAKE
More evidence of the mid-Holocene drought has been found in Pyramid Lake, located in northwestern Nevada. This lake is hydrologically connected to Lake Tahoe by the Truckee River, the only outlet for Lake Tahoe. After leaving Lake Tahoe, the river flows down the steep eastern flank of the Sierra Nevada, through Reno, and forty miles northeast into Pyramid Lake. Under natural conditions, therefore, the size of Pyramid Lake is determined in large part by the level of Lake Tahoe.
Pyramid Lake is a remnant of the ancient Lake Lahontan, which occupied a large part of northern Nevada at the end of the last ice age (see figure 18). Pyramid is a relatively deep lake, at 890 feet, and has no outlet. This makes it ideal for paleoclimate research. Since 1970, Larry Benson, a geochemist and hydrologist at the U.S. Geological Survey, has been studying what the lake can tell us about the West’s climatic history, including the relationship between past climate change and human culture.
One of Benson’s achievements has been the development of methods for assessing past lake levels by measuring the oxygen isotopes in calcium carbonate minerals in lake sediments. These minerals precipitate in the surface waters and eventually settle to the bottom, where they accumulate over time, preserving information about the climate. During dry periods, more water evaporates from the lake surface than is replenished by river inflow, lowering the lake level. Benson reasoned that, since the waters evaporating from the lake are more enriched in oxygen’s lighter isotope (oxygen-16), the remaining lake waters should be enriched in oxygen’s heavier isotope (oxygen-18). The enriched lake water is then incorporated into the carbonate minerals precipitating in the lake. Therefore, carbonate minerals that precipitate in the lake during dry spells should have a relatively high amount of the heavier oxygen isotope. Conversely, during wetter periods, rain and runoff enter the lake (freshwater that is enriched in the lighter oxygen-16 isotope) is greater than evaporation off the lake’s surface; the lake waters and carbonates forming in them are more enriched in oxygen-16 than they are during droughts. Thus, these isotopes in precipitated carbonates provide Benson and other researchers with an archive of past lake chemistry and a record of climate change in the region.
FIGURE 18. Map of the distribution of enormous late Pleistocene lakes, including Lake Bonneville in Utah and Lake Lahontan in western Nevada. Modern lakes discussed in the text are also shown, as well as the Cordilleran ice sheet over western Canada that extends into northern Washington, Idaho, and Montana. Arrows show the paths of ancient “megafloods” from Lake Missoula in Montana through the Columbia River into Washington, and from Lake Bonneville into southern Idaho. (Map drawn by B. Lynn Ingram and Wenbo Yang.)
Benson uncovered this long climate history at Pyramid Lake by coring sediments from its bottom. After bringing these cores back to his laboratory in Denver, he separated the carbonate minerals from the lake sediments and dated the cores, finding that the sediments spanned the past 6,500 years. This record revealed that, between 6,500 and 3,800 years ago (the mid-Holocene period), the oxygen isotopes in the lake water were enriched in the heavier isotope, oxygen-18, relative to the lighter isotope, oxygen-16, by an amount indicating that the climate during that time was about 30 percent drier than today, and warmer by 3–5°C (5.5–9°F).
DISAPPEARING LAKES
Benson and colleagues also studied lake sediments from other lakes in the Great Basin, including Owens Lake, to assess the impacts of past climate changes in the Owens Valley east of the Sierra Nevada in California. Like many Great Basin lakes, Owens Lake is a closed basin, receiving inflow only from the Owens River, with no outlet. For many thousands of years, this large lake, twelve miles long and eight miles wide, persisted in the Owens Valley (see figure 18), its size fluctuating in response to changes in the amount of precipitation and evaporation. The lake is now dry—not because of a drier climate but because water managers, starting in 1913, diverted the Owens River south through the Los Angeles Aqueduct to Southern California. By 1924, the lake was completely desiccated, leaving in its place an enormous salt flat composed of clay, silt, and evaporite minerals—the salts that precipitate from the briny waters during evaporation. Beneath this surface, however, lie thousands of years of subsurface sediment layers documenting the lake’s response to climate change.
Using the methods he developed for Pyramid Lake, Benson found that Owens Lake had also diminished during the mid-Holocene—between 7,700 and 3,200 years ago. This finding was based partly on oxygen isotope measurements, showing an increase in oxygen-18 in carbonate sediments, and on an assessment of the carbonate mineral types in the lake sediments. Sedimentary carbonates precipitating in lakes generally have two origins: either the carbonate is formed organically during the formation of the shells of organisms that live in the water, or it is formed inorganically, as carbonate precipitates out of the water through chemical reactions. Benson’s research at Owens Lake showed that the amount of carbonate in the sediments that formed inorganically (the total inorganic carbon, or TIC), compared with the carbonate that formed organically, was significantly higher during this mid-Holocene period. A higher level of TIC usually indicates that the prevailing climate conditions were drier, resulting in more evaporation off the lake surface, thereby leaving a greater concentration of dissolved salts in the lake waters that precipitate as carbonate minerals. Moreover, the time period between 6,480 and 3,930 years ago was devoid of sediments, and Benson hypothesized that this “hiatus” in sediment deposition resulted from desiccation of the lake during that period, suggesting an extremely prolonged period of drought during the mid-Holocene.
Another large and ancient lake, Tulare, offers its own evidence of a mid-Holocene drought. Located in the southern end of California’s Central Valley, Tulare Lake was once the largest freshwater lake west of the Great Lakes until its source of freshwater was diverted for agriculture in the early twentieth century. As recently as 1879, the surface area of the lake was 690 square miles. However, lake sediments and ancient shorelines show that Tulare Lake completely evaporated during the mid-Holocene, again suggesting an extremely dry climate during that period. Researchers have identified imprints of buried mud cracks that developed in the lake bottom as the water evaporated away, and they have studied the fossil remains of land plants that grew on the newly exposed lake bottom, all dating to 5,500 years ago. The lake remained relatively small until 3,600 years ago, when moister climate conditions returned to the region.
Clearly, a large area in the American West was affected by a hotter, drier climate during the mid-Holocene. This climate also altered the vegetation: fossilized pollen preserved in marsh and lake sediments at many locations throughout southern Oregon and extending into the Great Basin and California show shifting plant assemblages. In the mountains, drought-tolerant species, including grasses and shrubs, replaced forests and other densely vegetated areas, and oaks moved to elevations once dominated by pines.
Vegetation change was also seen in coastal regions. In the vast Sacramento–San Joaquin Delta, a freshwater marsh plant community gave way to more salt-tolerant species (saltgrass, pickleweed, and certain types of rushes) as river flows diminished. Along the Southern California coast, including Santa Barbara and the Santa Barbara Channel Islands, drought-adapted marsh plants came to dominate the tidal wetlands.
THE LONG (AND WARM) DROUGHT
This period of prolonged mid-Holocene dryness was coined the “Long Drought” in the late 1930s by Swedish sedimentologist Ernst Antevs, who documented that Lake Summer in south-central Oregon (see location in figure 18) had completely desiccated between 7,500 and 4,000 years ago. This drought in the American West occurred during what is now called the mid-Holocene “climatic optimum”—a period of unusual global warmth. In the early twentieth century, when Antevs was assembling paleoclimatic data, he named this period the “Altithermal.” Antevs’s research showed that relatively high temperatures, estimated to be 2–9°C (3.5–16°F) warmer than they are today, prevailed during this time in the high latitudes of the Northern Hemisphere. Later studies showed that temperatures at lower latitudes rose less dramatically, to perhaps 1°C (1.8°F) warmer than at present. Antevs’s research also suggested that this mid-Holocene period of warm and dry climate was followed by a cool, wet period from about 4,000 to 2,000 years ago, a period we now refer to as the Neoglacial (discussed in the next chapter).
More compelling evidence for increased warmth during the mid-Holocene comes from studies of bristlecone pines in California’s White Mountains at their upper limits in the mountains (the tree line, which represents the elevation above which trees cannot survive the cold, inhospitable conditions). During warmer climate conditions, bristlecone pines are able to survive at higher elevations, and the tree line moves upslope. The radiocarbon ages of ancient tree stumps along the upper elevations of the White Mountains indicate ages of between 5,700 to 4,100 years ago, suggesting a warmer climate. More evidence of warmth was observed in glaciers in the mountain ranges of the western United States that retreated during the mid-Holocene.
COASTAL CONDITIONS
Sediment cores retrieved from along the continental shelf of the Pacific Northwest coast contain the remains of marine organisms that provide clues to past oceanic conditions during the mid-Holocene. These clues include the minuscule fossilized remains of microscopic plants (phytoplankton) living in the sunlit surface waters of the oceans, which convert carbon dioxide and sunlight into energy (in a process known as photosynthesis), providing the primary fuel for the entire marine food web. Some phytoplankton secrete a silica cell wall that cannot be digested by the tiny marine animals (zooplankton) that feed on and then excrete them. As the excrement (or “fecal pellets,” filled with these hard, tiny cells) sinks slowly to the bottom, it eventually becomes invaluable data for the oceans’ sedimentary archive.
John Barron, a micropaleontologist at the U.S. Geological Survey in Menlo Park, California, has devoted his career to studying the fossil remains of phytoplankton. He is the world’s expert on diatoms, a type of phytoplankton that secretes silica cell walls shaped like minuscule hat boxes, footballs, or other fanciful forms—elegant ornamentation that helps them stay afloat in the sunlit surface waters of the ocean by increasing their surface area. Barron has analyzed the different diatom species found in this region and noted that each is adapted to specific ocean conditions, including water temperature, oxygen content, nutrient levels, and water depth.
Barron and his colleagues determined the relative abundances of a dozen or so species found in sediment cores recovered from the continental shelf in the eastern Pacific, from Baja California to Washington state. The oceanographic conditions of this region, just off the coast of western North America, play a major role in influencing climate along the coast, thereby determining its biological resources. Barron’s research findings reveal that cold, nutrient-rich waters were upwelling along the coast more often and more extensively between 6,300 and 5,800 years ago. His evidence consists of the increased presence of diatom species (including Thalassionema nitzschioides) that live in waters with enhanced upwelling and diatoms (Neodenticula seminae) that live in cooler, subarctic waters.
There is an increased concentration of biogenic silica and organic carbon in the sediments (from an increased flux of diatoms remains to the bottom), providing independent evidence for enhanced biological productivity as a result of the increased upwelling of cold, nutrient-rich coastal waters. Moreover, a decrease in the abundance of diatoms (Pseudoeunotia doliolus) that live in warm, low-nutrient waters also supports the idea that conditions were cooler and nutrient-rich as the result of increased upwelling. The abundance of nutrient-rich waters would have allowed all marine organisms along the coast to flourish, including those in the intertidal zone that provided food for humans.
Geochemist Timothy Herbert at Brown University, working on the same cores as Barron, was able to make inferences about sea surface temperatures using the fossil remains of the cell walls in a particular species of phytoplankton, Emiliana huxleyii. These microscopic plants alter the rigidity of their cell walls in response to changing water temperature. They do this by changing the number of double bonds in their lipid molecules, compounds known as alkenones that behave much like butter, a saturated fat that is a solid at room temperature. (Its structure is “saturated” with hydrogen atoms, so the carbon atoms are held to each other with single bonds.) Unsaturated fat, like oil, is a liquid at room temperature because it has a greater number of double bonds between its carbon atoms. Phytoplankton have evolved the ability to change the number of double bonds in their cell walls to maintain somewhat fluid, or less rigid, cell walls as the water temperature changes. These changes in the ratio of double to single bonds can be measured in the laboratory.
The alkenone results show that coastal surface water temperatures during the mid-Holocene were 1–2°C (1.8–3.5°F) cooler than conditions immediately before or after. The upwelling of cooler, nutrient-rich waters would have led to more prolific phytoplankton blooms, providing more food for organisms higher on the food chain: mollusks, fish, seabirds, and, ultimately, humans.
COASTAL REDWOODS: UPWELLING AND FOG
Some of the most dramatic evidence for increased coastal upwelling during the mid-Holocene comes not from the ocean but from the coastal mountains, where magnificent coast redwood trees grow between southern Oregon and central California. These awe-inspiring trees are the world’s tallest living organisms, some towering higher than 360 feet. The coast redwoods require summer fog to survive, and, because of this, they are intimately linked with the oceanographic conditions just offshore in the eastern Pacific Ocean. The fog reduces temperatures during the warm, dry summers and provides moisture for the trees by a process known as “fog drip”—that is, when moisture in the fog condenses on the redwood’s fine needles, it drips to the ground where it can soak into the soil and travel to the roots.
Coastal fog is thickest along the Pacific coast during the warm, dry summers. The fog forms when moist air flows inland off the ocean and passes over the cold surface waters of the coast, which causes condensation of the water vapor, resulting in fog. Simultaneously, the warm land surface heats the overlying air, causing it to expand and rise, and, as it rises, that warm air is replaced by the cool marine air and fog flowing onshore. In regions of the coast with small mountain ranges, this onshore movement of fog can appear like a cloud waterfall pouring over the coastal mountains, bringing precious moisture to coastal ecosystems during the dry summer months.
The Pacific coastal redwoods form a geographic boundary, or ecotone, between the more arid oak woodland and open scrub of Southern California and the wetter western hemlock, spruce, and alder lowland coastal forests of Oregon and Washington. An analysis of pollen from cores taken along the California and Oregon coasts suggests that coastal upwelling and the development of summer fog were coincident with a great expansion of these majestic redwood forests that began about 5,150 years ago.
HUMAN MIGRATIONS
Against this backdrop of a warming and drying climate in the West during the mid-Holocene, it is clear that the early humans living in the region faced tough times. Archaeologists have found evidence that humans were likely suffering from the severe droughts, especially in southeastern California and the Great Basin. Prolonged and severe droughts appear to have forced large-scale migrations out of the driest inland regions as people went in search of food and water. Douglas Kennett, an archaeologist at Penn State University who has studied these migrations, reasons that the moderate climate and abundant marine resources along the coast would have attracted native human populations from the drier interior deserts. Kennett has collaborated with his father, paleoceanographer James Kennett (see figure 19), to reconstruct coastal sea surface temperatures in the Santa Barbara Basin during the Holocene. Climate conditions along the coast were not as severe as inland, and the high biological productivity along the coast would have provided marine resources not available in the desert interior. The wetlands of California’s Central Valley would also have provided more resources and favorable conditions than the interior desert regions. At a time when the Great Basin and southeastern California were severely dry, the coast would have offered increased food and water resources.
According to Douglas Kennett, archaeological evidence supporting a coastal migration during the mid-Holocene comes from a study of the distribution of language groups in western North America. In particular, the Uto-Aztecan language family is distributed between coastal regions of present-day Southern California, the Southern California desert, and much of the Great Basin. This distribution suggests a movement of people from the interior to the coast.
Other archaeological evidence suggests that populations from the dry desert interior increased their interactions and trade with peoples along the California coast and Central Valley during the mid-Holocene drought. This evidence is contained in shell mound deposits excavated along the central and southern coasts, including the Santa Barbara Channel Islands. Exchanges among the tribes may have buffered the impacts of the drought, lowering the risks of reduced resources that would have accompanied extended dry conditions. Evidence for increased trade comes from beads crafted from the marine shell Olivella biplicata, or “purple olive snail.” During the mid-Holocene, rectangular beads were cut from the body whirl of these snail shells and used for trade and exchange. Specialized tools (micro-drills and anvils) for fashioning these beads, as well as the debris from their manufacture, were recovered on the Santa Barbara Channel Islands. The beads have been found in sites along the Southern California coast and in the northern and western Great Basin. Based on the ages and contexts of these beads, their distribution suggests that new trade routes were established between 5,900 and 4,700 years ago—the driest period of the mid-Holocene.
EXPANDING COASTAL ENVIRONMENTS
The coastal environment was just beginning to stabilize starting about 6,000 years ago, in the midst of this mid-Holocene drought. The period that followed the end of the Last Glacial Maximum, or 20,000 years ago, was marked by the melting of the massive glaciers that had shrouded much of the Northern Hemisphere. The retreat of the glaciers started slowly, but, as more land was exposed, the bare ground and open waters absorbed more solar radiation than did the glacial ice sheets, warming the earth’s surface. This led to an increase in the rate of melting. Large volumes of the water from the melting ice sheets flowed into the world’s oceans, raising global sea level at a rapid pace and transforming coastal regions worldwide.
FIGURE 19. Archaeologist Douglas Kennett (on left ) with his father, paleoceanographer and paleoclimatologist James Kennett, resting after fieldwork on San Miguel Island in the Santa Barbara Channel off the Southern California coast. (Photo courtesy of Douglas Kennett, Penn State University.)
Along the California coast, San Francisco Bay began its transition from a river valley 18,000 years ago to a large estuary system 8,300 years ago as rising sea levels began to enter the bay through the Golden Gate. This rapid sea level rise continued until the mid-Holocene, about 6,000 years ago, when the rate slowed to only about an eighth of an inch per year and the shoreline stabilized. This allowed coastal habitats along the Pacific Ocean, including tidal marshes, the rocky intertidal zone, beaches, bays, and estuaries (including San Francisco Bay) to develop and expand. These newly created coastal habitats would have provided new sources of abundant marine resources. Douglas Kennett proposed that some of the Uto-Aztecan people who had migrated from the dry interior deserts to Southern California then migrated northward, through the Central Valley of California and into the wetland environments of the San Francisco Bay region. The oldest shell mound site along the shores of San Francisco Bay—the West Berkeley shell mound—has a basal age that dates to the mid-Holocene, or 5,200 years ago.
Along the Pacific Northwest coast, the basal dates of a number of shell mounds are mid-Holocene in age, suggesting that coastal populations expanded there as well. Food resources expanded in that region, as they did along the southerly California coast, when sea levels stabilized. The newly formed intertidal zone would have brought with it the expansion of resources, such as shellfish, which allowed human populations to grow and construct the numerous shell mounds. Conditions also improved for salmon migration, spawning, and development of juvenile salmon. Archaeologist Knut Fladmark at Simon Fraser University proposed that an increase in the abundance of sockeye and coho salmon, considered the single most important Native American food resource in that region, led to an increase in coastal settlements.
The climatic backdrop of these changes based on evidence from the Pacific Basin suggests that the mid-Holocene was a period of increased La Niña conditions, when the trade winds blew with more intensity. It is possible that the La Niña conditions over the Pacific were at least partly responsible for the extended period of drought over the American Southwest.
Following the mid-Holocene drought, climate in the American West turned cool and wet again during a period known as the Neoglacial. The archaeological record contains evidence for increased human sedentism, cultural complexity, and more advanced technology beginning about 4,900 years ago, becoming even more widespread after 3,800 years ago. As we will explore further in the next chapter, the Neoglacial brought with it increased climatic variability.