ABSTRACTS OF PRESENTATIONS FROM PACLIM2007 (May 13-16, 2007)

CHARACTERIZING REGIONAL MODES OF CLIMATE VARIABILITY IN CALIFORNIA

JOHN T. ABATZOGLOU, KELLY T. REDMOND, AND LAURA M. EDWARDS

Western Regional Climate Center, Desert Research Institute, Reno NV 89512

John.Abatzoglou@dri.edu A novel approach is presented to identify regional patterns of climate variability in California. Coherent regions of climate variability across the state are identified using principal component analysis of monthly precipitation and temperature data from a wide network of climate stations and the gridded PRISM dataset. This multivariate analysis results in the division of California into 11 homogeneous regions with respect to variations in climate. Although large-scale modes of climate variability dominate patterns statewide, the complex interplay between circulation patterns and maritime and topographic influences leads to a number of important regional modes of variability. These climate regions are useful in exploring mechanisms for regional climate variability as well as for tracking variations in climate from 1895 to the present.

THE COLD EVENT 8,200 YEARS AGO: COSMOPOLITAN CONSEQUENCES OF A PROVINCIAL PERTURBATION

DON BARBER

Geology Department, Bryn Mawr College, Bryn Mawr, PA 19010 dbarber@brynmawr.edu

NO ABSTRACT SUBMITTED

1

SANTA BARBARA BASIN DIATOM RECORD SUGGESTS COINCIDENCE OF COOLER SST WITH WIDESPREAD OCCURENCE OF DROUGHT IN THE WEST

DURING THE PAST 2,200 YEARS

JOHN A. BARRON U.S. Geological Survey, MS 910, 345 Middlefield Road, Menlo Park, CA 94025

jbarron@usgs.gov Diatom assemblages from the upper 4.50 m of sediment from Ocean Drilling Program Hole 893A in the Santa Barbara Basin were studied every 5 cm. The varve chronology of Arndt Schimmelman (written comm., 2006) suggests that this interval covers the period from ca. 2,240 cal yr B.P. to A.D. 1880 with sampling intervals ranging between 12 and 43 years. The diatom assemblages in these 1-cm-thick samples are assumed to represent a composite of between 2.5 and 8 years of deposition. The ratio of the subtropical diatom Fragilariopsis doliolus to the more cosmopolitan diatom Thalassionema nitzschioides (Fd/Tn) has been used successfully as a sea surface temperature (SST) proxy in Holocene paleoclimatic studies off northern California. Cooler intervals probably represent times when spring upwelling conditions were dominant over summer conditions during the interval sampled. In the ODP 843A record Fd/Tn generally varies between 0.0 and 0.5 with cooler intervals (< 0.20) more common than warmer (> 0.30) intervals. Throughout the record of the past 2,200 years, a strong correspondence exists between lower Fd/Tn (< 0.20) values and heavier oxygen isotope values in the Globigerina bulloides record of Kennett and Kennett (2000), suggesting that diatoms and planktonic foraminifers record periods of enhanced upwelling. These cooler intervals, including the Medieval Warm Period between ca. A.D. 1000 and 1250, correspond with intervals of wider geographic occurrence of long-term drought in the West, according to the tree ring studies of Cook and others (2004). The relationship between cooler SST in the Santa Barbara Basin and more widespread drought in the West is consistent with the hypothesis that long-term drought in the West is associated with La Niña-like conditions in the Pacific Ocean. Cook, E.R., Woodhouse, C.A., Eakin, C.M., Meko, D.M., and Stahle, D.W., 2004, Long-term aridity changes in the western United States: Science, v. 306, p. 1015-1018. Kennett, D.J., and Kennett, J.P., 2000, Competitive and cooperative responses to climatic instability in coastal southern California: American Antiquity , v. 65, no. 2, p. 379-395. 2

THE RELATION OF GREAT BASIN LATE QUATERNARY HYDROLOGIC AND CRYOLOGIC VARIABILITY TO NORTH ATLANTIC CLIMATE OSCILLATIONS

L.V. BENSON (1), R.J. SPENCER (2), D. RHODE (3), L. LOUDERBACK (4),

AND R. RYE (5) (1) U. S. Geological Survey, 3215 Marine Street, Boulder, CO 80303

(2) Department of Geology and Geophysics, University of Calgary, Alberta T2N 1N4, Canada

(3) Division of Earth and Ecosystem Sciences, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512

(4) Anthropology Department, University of Nevada,1664 North Virginia Street, MS 096, Reno, NV 89557

(5) U.S. Geological Survey, MS 963, Denver Federal Center, Lakewood, CO 80225 lbenson@usgs.gov

New data from the Lake Bonneville basin and previously published data from four western Great Basin lakes indicates that some Heinrich events were associated with periods of relative aridity in Great Basin lake basins and that Dansgaard-Oeschger stades were associated with periods of alpine glacier advance in the Sierra Nevada. The climatically induced highstands of Lake Bonneville, Mono Lake, and Pyramid Lake terminate shortly after Heinrich event 1. Glacial recession in the Sierra Nevada and Wasatch Mountains appears to have occurred during the climate warming that began at the end of Heinrich event 1 and that culminated in the Bølling-Allerød warm period. Lake level oscillations occur in the Bonneville and Owens Lake basins during the Preboreal oscillation and Younger Dryas interval; however, the wet period does not appear to occupy the entire Younger Dryas interval. In general, a clear correlation between the hydrologic balances of Great Basin lakes and Dansgaard-Oeschger (DO) oscillations can not be made; however, the δ18O records for Lake Bonneville and Pyramid Lake, in a smoothed sense, resemble the GISP2 δ18O record. This suggests the possibility, that whatever the nature of the climate mechanism involved in producing DO oscillations, it also affected the hydrologic balances of some, if not all, Great Basin lakes. The lack of obvious coherence between the GISP2 and Great Basin hydrologic records suggests that the DO signal was shifted in time and(or) space when it reached the Great Basin. We offer the hypothesis that the change in size and shape of the Laurentide ice sheet associated with a Heinrich event caused a shift in the mean position of the polar jetstream away from the catchment areas of some of the Great Basin lakes. The coherence of Sierra Nevada glacial oscillations with DO events also suggests the existence of a teleconnection between the western Great Basin and the North Atlantic.

ON TELECONNECTIONS OF THE TREE RING RECORD IN THE TORREY PINE: 3

ENSO, NAO, AND DROUGHT IN THE SOUTHWEST

W. H. BERGER Scripps Institution of Oceanography, La Jolla, CA 92093-0244

wberger@ucsd.edu The Torrey pine (Pinus torreyana) in northern San Diego County is restricted to a small area along the coast, near Del Mar, with one other population existing on Santa Rosa Island. The reasons for the restricted distribution is not known. Likewise, the history of the connection between the two populations is obscure. What is known is that the tree-ring record of the San Diego population largely reflects winter precipitation in southern California, which is in turn informed by southwestern precipitation patterns that are linked to sea surface temperatures in the tropical eastern North Pacific through (El Niño – Southern Oscillation) ENSO variations (Biondi and others, 1997). Also, and rather surprisingly, southwestern patterns of precipitation are linked to sea surface temperatures in the northern North Atlantic Ocean (Cayan and others, 1998); that is, to the dynamics of the North Atlantic Oscillation (NAO). The dual linkage implies that the Torrey pine record – and indeed all such records in the southwest – should bear a combination signal from ENSO and (North Atlantic Oscillation) sources. If the ENSO and NAO contain periodicities that are well defined (e.g., through linkage to astronomical forcing) the tree-ring records should show the appropriate interference patterns in addition to the primary stimulation periods. Thus, it should be possible to obtain some notion of the relative importance of the ENSO and NAO linkage through history, as concerns the precipitation patterns in the southwest. Clearly, such changes in linkage patterns greatly influence shifts in the distribution of floras and faunas; that is, the ‘migration’ of plants and animals. Biondi, F., Cayan, D.R., and Berger, W.H., 1997, Dendroclimatology of Torrey Pine (Pinus torreyana Parry ex Carr.): American Midland Naturalist, v. 138, p. 237-251. Cayan, D.R., Dettinger, M.D., Diaz, H.F., Graham, N.E., 1998, Decadal variability of precipitation over western North America: Journal of Climate, v. 11, p. 3148-3166.

AN EXTREME TURN-OF-THE-CENTURY HYDROLOGIC EVENT RECORDED IN δ18O FROM BRISTLECONE PINE TREE-RING CELLULOSE

4

MAX B. BERKELHAMMER and LOWELL D. STOTT

Department of Earth Sciences, University of Southern California, Los Angeles 90089-0740 berkelha@usc.edu

The bristlecone pine chronology from the White Mountains of California has provided important insight into the climate of the last two millennia. Ring-width chronologies from the species have been shown to record subtle changes in climate, showing a strong correlation with the Palmer Drought Severity Index, an integrated climate index incorporating both changes in temperature and humidity. Previous studies have shown that the D/H and δ13C in the tree-ring cellulose of bristlecone pines respond to low frequency changes in temperature (Feng and Epstein, 1994) and soil moisture (Leavitt, 1994), respectively. This demonstrates the potential to reconstruct climate characteristics otherwise hidden in the integrated ring-width data. Due perhaps to the diminutive size of its rings, no previous study has attempted to create an annually resolved stable oxygen isotope chronology for this species. Some studies suggest that at mid-latitude locations tree cellulose δ18O can record changes in atmospheric circulation patterns that influence the source of precipitation. Such a reconstruction from the bristlecone pines would be a useful tracer of environmental history in California. Here we present preliminary δ18O (cellulose) results from bristlecone pines from the White Mountains. The δ18O chronology follows a distinctive bidecadal oscillation throughout the 20th century with peak excursions being on the order of 4‰. This trend gives way to a dramatic enrichment of approximately 10‰ that began in the latter part of the nineteenth century and continued in a stepwise fashion until it stabilized at approximately 1860. The δ18O values were stable throughout the rest of the 18th and 19th centuries. Previous studies of climate patterns in the western United States have also purported to see a bidecadal hydrologic cycle. The phasing of the δ18O oscillation indicates a relationship with Pacific-wide ocean-atmosphere systems that modulate the hydrologic cycle in the western United States. Geochemical modeling of the cellulosic δ18O indicates that the only likely candidate to explain the magnitude of the excursions could be changes in source water. This provides further credence to the conclusion that the isotopic pattern is controlled by regional hydrologic changes. The magnitude of the extreme event at the beginning of the 20th century is unequaled in the record and suggests a major regime change that is consistent with the results of other proxy records that show major environmental changes occurred during this time interval. Feng X, and Epstein, S., 1994, Climatic implications of an 8,000-year hydrogen isotope time series form bristlecone pine trees: Science, v. 265, no. 5175, p. 1079-1081. Leavitt, S.W., 1994, Major wet interval in White Mountains Medieval Warm Period evidenced in δ13C of bristlecone-pine tree-rings: Climatic Change, v. 26, no. 2-3, p. 299-307. THE MAAR LAKES OF GUANAJUATO, MEXICO: HIGH RESOLUTION ARCHIVES

OF HOLOCENE CLIMATE CHANGE

ROGER BYRNE (1), JUNGJAE PARK (1), JAMES JOHNSTON (1), HARALD BOHNEL (2) 5

LYNN INGRAM (3), WENBOW PARK (3), AND ULRIKE KIENEL (4) (1) Department of Geography, University of California, Berkeley, CA 94720

(2) Centro de Geociencias, Campus UNAM Juriquilla, 76230 Queretero, Mexico (3) Department of Earth and Planetary Science, University of California,

Berkeley, CA 94720 (4) GeoForschungsZentrum Potsdam, Section 3.3 Climate Dynamics and Sediments,

Telegrafenberg Haus C, D-14473 Potsdam, Germany arbyrne@berkeley.edu

There are seven maar craters in the Valle de Santiago area of Guanajuato, Mexico. Two of them, Rincon de Parangueo and La Alberca, had deep (50 m+) lakes until recent (post-1970) lowering of the regional water table caused them to desiccate. The desiccation, which was due to groundwater exploitation for irrigation, has made it possible to recover core samples directly from the exposed lake floors. Since 2001 we have recovered numerous piston cores from both Rincon and Alberca; the longest core from Rincon has a basal date of 9,000 cal years B.P., and the longest core from Alberca a date of ca. 6,000 cal years B.P. Here we report on the pollen, magnetic susceptibility, sediment chemistry, and stable isotopic composition of the Rincon cores. The pollen record is rather complacent and is dominated by pine and oak pollen, both of which reach the lake via long-distance dispersal. Changes in sediment chemistry reflect alternating periods of drier and wetter climate, which in part are also evident in the pollen record. The stable isotope records (oxygen and carbon) also indicate changes in precipitation and evaporation throughout the Holocene. Of particular interest is a high resolution (sub-decadal) isotope record that covers the last 2,000 years. This record provides new evidence of solar forcing of climate in Central Mexico during the late Holocene. RAPID BUT VARIABLE RESPONSES OF SIERRA NEVADA GLACIERS TO ABRUPT

CLIMATE CHANGE

DOUGLAS H. CLARK (1) AND NICOLE D. BOWERMAN (2) (1) Geology Department, Western Washington University, Bellingham, WA 98225

(2) North Cascades National Park, Marblemount, WA 98267 Doug.Clark@wwu.edu

Detailed analyses of lake sediments below glaciers in the Sierra Nevada demonstrate that the small glaciers are highly sensitive to changes in climate, but that their responses may not be simple and monotonic (i.e., advances during cooling). Whereas moraines offer “snapshots” of maximum glacier positions, proglacial lakes provide more complete records of glacier ebb and flow. The lakes act as efficient sediment traps for rock flour emanating from the glaciers and provide continuous, datable, high-resolution proxy records of extent and timing of glacier growth and degradation. The moraines then provide a means to quantify the climate changes driving these fluctuations. The archetypal extreme climate event, the Younger Dryas, has spawned abundant research efforts, and even a Hollywood disaster film. Although the event had a remarkable effect on climate and glaciation in the circum-North Atlantic region, its effects elsewhere around the globe were more subtle. In the Sierra Nevada, the Younger Dryas chronozone (YD) does not

6

show as a significant glacial advance. Instead, lake records indicate that it was a cold, dry period following closely on the heels of a small but regionally significant late glacial advance, the Recess Peak event (~12,200-11,200 14C yr B.P.; 14,200-12,200 cal yr B.P.) Glaciers during the YD, if extent, were smaller than those related to Neoglaciation because no YD moraines have been identified in the range. This finding is consistent with sediment records from the east (playas) and west (Santa Barbara Basin), which indicate cold and dry conditions elsewhere in California during the YD. Glacial deposits from farther a-field (Cascade Range, Rocky Mountains) indicate minor local glacier advances during the YD, but at a scale on par with those related to Neoglaciation (~ 1-2�C). Most mountains of the western US appear to have been glacier-free following the YD until the late Holocene. Beginning about 3,200 14C yr B.P. (~ 3400 cal yr B.P.), cirque glaciers began to reform in the Sierra Nevada and the CascadeRange, culminating with maximum advances late in the Little Ice Age (LIA). In the Sierra Nevada, the Palisade Glacier appears to have reached progressively greater maxima at ~ 2,800, ~ 2,200, ~ 1,600, ~ 700 and 250-170 cal yr B.P. Since the last maximum, the lakes below the Palisade Glacier have experienced a substantial decline in rock flour flux, despite the fact that the glacier was not actually retreating for most of this period. Instead, this decreased rock flour flux appears to record the thinning of the glacier. With such small glaciers, even minor thinning or shrinking related to ice loss (negative mass balance) will reduce the flow velocity, which in turn reduces the production and delivery of rock flour to the downstream lakes. The glaciers and adjoining lakes are thus highly sensitive recorders of net snowfall in the mountains. The combined records of moraines and alpine lake sediments provide clear constraints on abrupt climate events in the western U.S. During the YD, glaciers throughout the region were only locally slightly larger than during the height of the LIA, indicating a substantially diminished climate event compared to that in the North Atlantic. Conversely, during the late Holocene, glaciers in the Cordillera repeatedly expanded and contracted after 3,400 cal yr B.P., reaching their Holocene maxima during the past 150-200 years. Since then, glaciers throughout the west have retreated significantly.

POST-YOUNGER DRYAS WARMING: AN ANALOG FOR THE NEXT 100 YEARS?

KENNETH L. COLE U.S. Geological Survey, Southwest Biological Science Center, P.O. Box 5614,

Northern Arizona University, Flagstaff, AZ 86011 ken_cole@usgs.gov

Recent paleotemperature data from northern Arizona indicating a rapid temperature increase at end of the Younger Dryas correlate well with global records . The δ13C values and presence of Utah Agave in fossil packrat middens from the Grand Canyon suggest that winter temperatures increased by at least 4 oC between ca. 11,700 and 11,500 years ago. The timing of this rapid warming at the end of the Younger Dryas is indistinguishable from similar increases observed in the better-dated records from Greenland ice cores and eastern Pacific Ocean cores. The rate and magnitude of this warming is similar to that projected for the next century. Previous studies have noted unusual, anomalous plant communities typical of the very early Holocene immediately follow this rapid temperature increase. Packrat midden records 7

suggest that species diversity was lower during this time and that plant communities were dominated by mid-seral species typical of substrates that are several hundred years in age, such as Acacia greggi at low elevations and Fraxinus anomala at middle elevations. Higher-elevation pollen records from this time are often poorly constrained by radiocarbon dates, yet some contain higher levels of some early successional species such as Artemisia spp., Poaceae, and Asteraceae. Earlier interpretations of these anomalous assemblages have focused on such things as the high levels of summer insolation during this period. More precise chronological control suggests that because of the rapid temperature warming just prior to this period, a successional perspective on this vegetation change may allow a more robust understanding. If plant successional dynamics did cause these unique early Holocene assemblages, then similar changes should be expected in the future due to the continuing effects of global warming.

DYNAMIC SIMULATION MODELING OF MOSQUITO POPULATIONS WITH CLIMATE DATA

ANDREW C. COMRIE AND CORY MORIN

Department of Geography and CLIMAS Project, University of Arizona, Tucson, AZ 85721 comrie@arizona.edu

Climate is an important control on populations of mosquito disease vectors such as Aedes aegypti. Improved understanding and modeling of climate effects on mosquito populations is needed to help predict disease epidemics, and to understand the impacts of climate on mosquitoes as disease vectors. There is a surprisingly limited number of mosquito models with a climate component reported in the literature. The majority of these are empirical and statistical, and in particular, there are no dynamic simulation models. In this study we report on the development of a dynamic mosquito population model driven in part by climatic data. Model inputs include temperature, precipitation and relative humidity that influence mosquito population development through maturity and mortality in relation to temperature, population density, and water availability. These factors are considered through all phases of mosquito life cycle development from mosquito ovulation though each larval stage to pupae and finally adult. Temporal resolution is daily, and the model can be calibrated for specific mosquito species and locations. We use the model to evaluate the role of climate variability in mosquito populations, using Tucson, Arizona as a case study. Using different versions of the model we examine the relative behavior of the species of mosquito most likely to dominate the mosquito population given a specific area and climate. We also provide model results on the effects of different types of climate change and variability, including abrupt change and ENSO impacts.

RAIN-FED FLOOD RISKS IN A WARMING WEST

MICHAEL DETTINGER (1), DANIEL CAYAN (1), AND JESSICA LUNDQUIST (2), (1) U.S. Geological Survey, Scripps Institution of Oceanography, La Jolla, CA 92093

(2) University of Washington, Seattle, WA 98195 mddettin@usgs.gov

8

In recent decades, western North America has warmed significantly with far-reaching hydrological and landscape consequences. For example, precipitation falls more often now as rain rather than snow, springtime snowpacks are smaller, the annual hydrographs of snow-fed streams now rise earlier in the year, plants are greening and blooming earlier, and wildfire risks have increased. One important aspect of the western landscape that does not appear to have trended in statistically significant ways is the occurrence of floods. Documentation of flood-risk responses to recent warming, to the extent that such responses may exist, is difficult because floods are extreme and infrequent events and thus many decades of warming may be required before the changes can be confidently detected, even under the best conditions. Another factor that makes detection of flood-risk changes difficult is that we do not yet know what kinds of changes we should be looking for: Will western floods become more common or less, more severe or less, different in origins from historical floods or more of the same? A well-documented flood in Yosemite National Park on 16 May 2005 provides an example of one way that floods may be expected to change in a warmer world, because the flood generation was driven directly by unusually warm storm-time conditions, rather than by extreme precipitation amounts (only ~ 2.5 cm of precipitation fell) and without deriving much of its flow from melting snowpacks (which may not be as large in a warmer world). A Pacific storm drew warm, wet subtropical air into the Sierra Nevada, causing moderate rains and major flooding (ninth largest flood peak in 90 years of record) in Yosemite Valley. The flood derived its runoff mostly from high-altitude rainfall on soils already wet due to the onset of spring snowmelt a few days earlier. No significant snow-water content changes were measured. Flooding occurred because, although the average storm in Yosemite provides rain—rather than snow—to only about 15% of the park’s river-basin areas, this storm was warm enough to blanket 85% of the park with rain, vastly increasing the area and runoff volumes contributing to the flood. This flood is an example of one way that flood generation and risks are likely to change in a warmer world: The flood was caused by warmer, rainier than normal conditions rather than by intense precipitation. The flood did not result from rain that melted snow, which will presumably be less common in a warmer world. The flood was made possible because the soils were already wet from earlier snowmelt. To the extent that snowpacks still blanket western mountains in a warmer climate, they are likely to form from warmer, wetter snows and to remain warmer and wetter until they melt (earlier). Thus soils are likely to receive more frequent replenishments during the winters and springs of a warmer world, so that flood facilitation by wetter soils may become more common in the warmer world. From our growing cache of temperature observations in the high country of Yosemite National Park, we have learned that storm-time temperature lapse rates tend to conform to moist-adiabatic conditions, cooling fairly regularly by about -6.5 ºC for every kilometer of elevation rise. Thus the 90 year historical record of wet-day temperatures measured in Yosemite Valley can be extrapolated over the topography to estimate, on a storm-by-storm basis, what fractions of the river basins received rain rather than snow. This exercise shows that the May 2005 storm was quite unusual in terms of the large basin area that received rain rather than snow. Such rain-contributing areas have been encountered during only about 0.6% of all historical storms yielding more than 2.5 cm of precipitation, or four times in 95 years. If a modest +3 ºC warming is imposed on the historical record — to represent the simplest form that continued warming trends could take--such runoff-contributing areas are encountered much more often, occurring in 9

about 5% of storms or in about one year in three, thereby increasing the opportunities for this type of flooding dramatically, even if the intensities or frequencies of precipitation extremes do not change. Notably, a comparison of how often the day of each year’s maximum flow in the Merced River were wet or dry historically indicates that annual-flood peaks on wet days became considerably more common as the 20th century progressed. Decadal counts of the numbers of wet- vs dry-day flood peaks rose from no wet-day floods, or 100% dry-day snowmelt-fed flood peaks, in the 1920s to about 60% wet-day, rainfed flood peaks in the 1990s. These results from Yosemite rivers can be extrapolated to other rivers in the Sierra Nevada and across the conterminous U.S., and when this is done, a sobering picture emerges: The Sierra Nevada, together with the Cascade Range, are by far the most susceptible to increased flood risks due to increases of rainfed basin areas with warming. With a +3 ºC warming, the number of opportunities for rain-fed flood generation to occur might be expected to increase by as much as 5-10 additional flood opportunities per year in broad swathes of the Sierra Nevada and Cascade Range; while in most of the rest of the west (as in the rest of the country), opportunities for such rain-area driven floods do not increase by more than a couple opportunities per decade. Thus, whether or not storms become larger or more frequent, and with or without future snowpacks, the Pacific Coast states should expect their flood risks to increase significantly if recent warming trends continue.

AIR TEMPERATURE, COLD AIR DRAINAGE, AND TOPOCLIMATIC EFFECTS IN THE LAKE TAHOE BASIN

SOLOMON Z. DOBROWSKI (1), JONATHAN GREENBERG (2), GEOFF SCHLADOW (1)

(1) Tahoe Environmental Research Center, University of California, Davis, CA 95616 (2) Center for Spatial Technologies and Remote Sensing, University of California,

Davis, CA 95616 szdobrowski@ucdavis.edu

Air temperature exhibits complex spatial and temporal patterns in mountain environments. Spatial estimates of air temperature in these environments are often estimated at coarse spatial grains while more fine-grained estimates often rely solely on elevation-based lapse rates. We provide a case study of the development and interpretation of fine-scale and physiographically informed spatial temperature estimates within the Lake Tahoe Basin. As a reference, we characterize daily elevation-based lapse rates for the study area over a 10 year period for minimum, maximum, and average temperatures using data from 14 SNOTEL stations. We then employ a regional regression modeling approach that utilizes a single, domain wide, multivariate regression function between temperature and elevation, solar insolation, topographic convergence, latitude, and longitude. To drive this model, we utilize GIS terrain analysis techniques to model topographic convergence as a proxy for cold air drainage, and a mechanistic solar radiation model. Our results demonstrate that for minimum temperature, approximately 40% of the daily observations exhibited positive lapse rates, the majority of which occurred through the summer months (May- September). In contrast, 4% of the maximum and average temperature lapse rates were positive, the majority of which occurred during the winter months (November-March). This latter group showed concomitant positive minimum temperature lapse rates suggesting temperature inversions that persisted over the course of a day or over multiple 10

days. Results from the regional regression model show that elevation has little explanatory power when characterizing minimum temperatures. In contrast, topographic convergence (a measure of cold air drainage) has substantial explanatory power. Average temperatures show sensitivity to elevation, topographic convergence, and solar insolation. Our results demonstrate that spatial modeling of air temperature estimates in mountain environments greatly benefit from physiographically informed modeling techniques. Moreover, our results have direct implications on the use of spatial temperature estimates for climate change impact assessments.

ANALYSIS AND SUMMARY OF JULY 2006 RECORD-BREAKING HEAT WAVE IN CALIFORNIA

LAURA M. EDWARDS (1) AND DANIEL R. KOZLOWSKI (2)

(1) Western Regional Climate Center, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512-1095

(2) California-Nevada River Forecast Center, 3310 El Camino Avenue, Sacramento, CA 95821-6373

laura.edwards@dri.edu

A record-breaking heat wave affected much of the state of California during the period from Sunday, July 16 through Wednesday, July 26, 2006. Although numerous daily maximum temperature records were set, the aspect that made this event unique was the elevated overnight minimum temperatures at several reporting stations. This was especially true in the southern Sacramento Valley and much of the San Joaquin Valley. Along with the intensity of this heat wave, the duration of abnormally high maximum and minimum temperatures was particularly noteworthy. Previous studies have shown that the duration of a heat episode is an important factor to consider when determining its impact. These factors, along with the meteorological synoptic situation, local effects and impacts, were addressed in this study of the July 2006 California event. CHANGES IN RAINFALL AND WATER AVAILABILITY ON ISLA ISABELA, GULF

OF CALIFORNIA

AMY ENGLEBRECHT (1), LYNN INGRAM (1), ROGER BYRNE (2), ULRIKE KIENEL (3), HARALD BOEHNEL (4), AND GERALD HAUG (3)

(1) Department of Earth and Planetary Science, University of California, Berkeley, CA 94720 (2) Department of Geography, University of California, Berkeley, CA 94720

(3) Department 3.3 Climate Dynamics and Sediments, GeoForschungsZentrum Potsdam, Potsdam Germany

11

(4) Centro de Geociencias, Universidad Nacional Autonoma de Mexico Campus Juriquilla, Querétaro Mexico

aenglebrecht@berkeley.edu Here we use the stable isotopic composition (oxygen and carbon) and pollen content of seasonally laminated lake sediments from Isla Isabela (21�52’ N, 105�54’ W) to reconstruct the history of recent climate change in northwestern Mexico. Isabela crater lake is located approximately 30 km offshore the state of Nayarit, in the precipitation region of the Mexican Monsoon (also called North American Monsoon). Work on a short core indicates coherent periodicities in δ13C and δ18O values from bulk inorganic carbonate throughout the past two centuries. Most notably, strong excursions in δ13C and observable sedimentological changes occur at depths in the core corresponding to the 1780s, 1810s, 1860s, 1950s, and 1973-74 — recognized periods of historical drought in Mexico. Further downcore work reveals additional periods of reduced water availability that may correspond to variations in the Mexican Monsoon.

CAN HYDROGEN ISOTOPE RATIOS IN PLANT LIPIDS PROVIDE A QUANTITATIVE PROXY FOR ARIDITY?

SARAH J. FEAKINS AND ALEX L. SESSIONS

California Institute of Technology, Mail Stop 100-23, Pasadena, CA 91125 feakins@gps.caltech.edu

Hydrogen isotopic fractionation within the hydrological cycle respond strongly to climatic variables, including evaporation. We are investigating whether the D/H ratios of leaf-wax lipids can provide a quantitative proxy for aridity. Under low precipitation, high evaporation climate regimes, differences in water uptake, conservation, and biosynthesis between plant species may exert a significant control on the hydrogen isotopic composition of plant leaf waxes. We measure the D/H ratios of environmental waters, xylem water, leaf water, and leaf wax lipids in order to quantify the isotopic enrichment within individual plants. We survey a diverse range of species and floristic regions across an aridity gradient in southern California in order to assess the range of isotopic compositions of plant leaf waxes and enrichment factors associated with ecological and climatic differences. Preliminary data from this sampling program will be presented at the meeting.

SEA SURFACE TEMPERATURE, DROUGHT, AND DROUGHT PREDICTION IN CALIFORNIA

IAN FERGUSON (1,2) AND JOHN DRACUP (1)

(1) Department of Civil and Environmental Engineering, University of California, Berkeley, CA 94720

(2) Lawrence Livermore National Laboratory, Livermore, CA 94550 iferguson@berkeley.edu

Droughts are complex ocean-atmosphere phenomena, with impacts on virtually all 12

human and environmental systems at timescales ranging from seasonal to inter-decadal. Since 1863, eleven droughts have affected California, with lengths ranging from two to six years. The last major drought in California occurred from 1987 to 1992; water managers are anxious that the low 2006-07 Sierra Nevada snowpack is the beginning of California’s next drought. While several studies have shown that ocean-atmosphere interactions influence seasonal and longer scale climate variability, the physical mechanisms that lead to drought events are not well understood. Here we evaluate seasonal-to-interannual drought characteristics in the Sacramento and Colorado River basins, which together supply most of California’s municipal, industrial, and agricultural water. Using a global sea surface temperature (SST) dataset, we evaluate the relationship between SST anomalies and drought in both basins. While a simple correlation analysis shows that 20th century streamflow variability is significantly correlated with SST anomalies over several ocean regions—most notably the tropical Pacific—a composite analysis of individual drought events shows only a weak relationship between SST anomalies and droughts. Results are discussed with respect to the physical mechanisms of drought and the potential for long-lead drought prediction.

FLOOD VARIABILITY IN RESPONSE TO CLIMATE VARIABILITY, SACRAMENTO-SAN JOAQUIN RIVER SYSTEM, CALIFORNIA

JOAN L. FLORSHEIM (1) AND MICHAEL DETTINGER (2)

(1) Geology Department, University of California, Davis, CA 95616 (2) U.S. Geological Survey, Scripps Institution of Oceanography, La Jolla, CA 92093

florsheim@geology.ucdavis.edu Climate variability governs river discharges and flood variability and thus modulates a river’s ability to accomplish geomorphic work. We explore the propensity of climate phenomena such as El Niño and La Niña phases of the ENSO cycle, positive and negative phases of the PDO, and seasonal “pineapple-express” storms to initiate river discharges and floods capable of causing geomorphic change in California’s Sacramento-San Joaquin River system. Historical levee breaks serve as a metric of the potential for geomorphic change and dangerous flood damage, representing an important way that the anthropogenically altered system currently responds to floods. Time series analysis of post-dam discharge and flood records from three representative stations shows a signal with a period of ~10-15 years with slight differences from north to south. In this decadal context, strong relations exist between levee breaks and annual peak discharge normalized by the mean of the peaks over the period of record (Qpeak/Qmean(pk)). Qualitatively, it appears that high flows caused by pineapple-express storm conditions generate levee breaks. As suggested by the non-standard 10-15 year time scale noted above, on multi-year time scale, relatively poor relations exist between annual numbers of levee breaks and phases of cyclic climate phenomena such as ENSO and PDO. The history of levee breaks through the 20th century does not indicate that the presence of flood control infrastructure such as dams and levees has reduced the frequency of breaks. A better understanding of the influence of climate variability on floods and their geomorphic consequences in the modern fluvial system will aid in improving hazard-risk assessments.

CLIMATE CHANGE’S INFLUENCE ON SUDDEN OAK DEATH 13

SUSAN J. FRANKEL

U.S.D.A. Forest Service, Pacific Southwest Research Station, Albany, CA 94710 sfrankel@fs.fed.us

Sudden Oak Death, caused by the exotic invasive plant pathogen Phytophthora ramorum, was first recognized in California in the mid-1990s. Over the past decade, millions of tanoak (Lithocarpus densiflorus) and coast live oak (Quercus agrifolia) have been killed from Big Sur (Monterey County) north to Curry County, in southwest Oregon. The pathogen also causes non-lethal, sporulating infections on camellia, rhododendron, and other popular ornamental nursery plants which can serve as vehicles for long-distance spread. Phytophthora ramorum was new to science upon its discovery in 2000. Researchers are just now unraveling the biology and ecology of this pathogen in coastal redwood/tanoak and mixed evergreen forest ecosystems. Extremes in climate appear to be increasing tree mortality and inoculum levels in California’s infested forests. The global risk of pathogen establishment is also being influenced by changes in climate.

LAKE TAHOE AND THE WORLD WATER CRISIS

CHARLES R. GOLDMAN Distinguished Professor of Limnology, Environmental Science and Policy, University of

California, Davis, CA 95616 crgoldman@ucdavis.edu

The World Water and Climate Network (WWCN) was established to assemble the best possible information on inland water problems associated with climate change. The effects of climate change and variability on the thermal structure of Lake Tahoe were studied over a 32-year period. The changes in thermal structure were related to an upward trend in air temperature since 1970 and a four-year decline, probably due to volcanic activity, followed by recovery in solar radiation. Thermal structure changes in Lake Tahoe included a long-term warming trend, the expected increase in resistance to mixing, and a trend toward decreasing depth of the October thermocline. The implications of these and other changes studied over almost a half-century at Lake Tahoe and Castle Lake in northern California have direct application for anticipating the impact of climatic change and global warming on the world’s surface water supplies. HOLOCENE FIRE HISTORY OF CONIFER FORESTS IN THE SIERRA NEVADA OF CALIFORNIA, SOUTHERN BRITISH COLUMBIA, AND COASTAL ALASKA, AND

ITS CONNECTION TO ABRUPT CLIMATE CHANGE

DOUGLAS J. HALLETT (1), R. SCOTT ANDERSON (2), AND DANIEL G. GAVIN (3)

(1) Department of Geography and School of Environmental Studies, Queen's University, Kingston, Ontario, K7L 3N6 Canada

(2) Center for Environmental Sciences & Quaternary Sciences Program, Box 5694, Northern Arizona University, Flagstaff, AZ 86011

(3) Department of Geography, University of Oregon, Eugene, OR 97403-1251 14

hallettd@post.queensu.ca Synchrony of forest fire activity between independent lake sites can be tested using well-dated millennial-scale records of sedimentary charcoal. Paleo-fire records have the potential to demonstrate large-scale climatic control of crown fire response across forest ecosystems. Fire episodes recorded at three sites in the Sierra Nevada show significant synchrony across 100 to 1,000-year windows for the overlapping portions of the record (8,500 cal yr B.P. to present). Frequent fire episodes tend to cluster during portions of the middle and late Holocene and may indicate a response to the onset of enhanced ENSO activity. The 13,500-year-long charcoal record from Swamp Lake, Yosemite National Park displays frequent fire activity during the early Holocene and an increasing, but periodic, trend in fire episode frequency from 8,000 cal yr B.P. to present. This trend in fire episode frequency and the raw charcoal data corroborate well with other proxy records that reconstruct century-scale variability of ENSO frequency during the Holocene. Synchronous fire response occurs during the Mediaeval Climatic Anomaly (1,000-700 cal yr B.P.) at all the Sierra Nevada sites and supports other paleoenvironmental records in the region. Long-term fire history records from southern British Columbia and Alaska show a response to Neoglacial climate change and the advance of regional glaciers in the late Holocene. Significant synchrony between fire episodes at five British Columbian lake sites (7,000 cal yr B.P. to present) also occurs across 100 to 800-year windows. Synchronous forest fire response at these independent sites is another important test linking large-scale circulation features to ecosystem disturbance. We suggest that century-scale climate variability is an important control on forest fire response across 50° N and future changes in climate will produce abrupt shifts in disturbance regimes across Pacific states and provinces.

CHANGING PERIODIC BEHAVIOR OF THE HOLOCENE NORTH AMERICAN MONSOON FROM VARVED SEDIMENTS IN THE

GULF OF CALIFORNIA

ANN MARIE HARRIS AND ROBERT E. KARLIN Department of Geological Sciences, University of Nevada, Reno, NV 89519

thelite@ix.netcom.com Laminated sediments from DSDP Site 480 in the central Gulf of California yield a virtually continuous sedimentation record of monsoonal circulation down to a sub-annual scale for the interval from 11,500 to ~1,800 years ago. Site 480 lies under the influence of the North America Monsoon (NAM) with seasonally reversing winds resulting in winter upwelling with high diatom productivity and summer deposition of aeolian terrigenous material. An annually- 15

resolved paleoclimate history of the Holocene was extracted from digital images of radiographs of the Site 480 Cores by identifying and counting annual and sub-annual units and measuring their thicknesses. The varve chronology agrees very closely with the independent chronology constructed from available radiocarbon and isotopic ages. This suggests that the varve record is complete and that seasonal monsoonal circulation has persisted over the Gulf throughout the Holocene. Holocene sedimentation on annual and seasonal scales is somewhat chaotic, but has ~50- to 500-year long intervals with characteristic patterns showing quasi-periodic behavior. Both the annual and seasonal proxies have periodicities which wax and wane throughout the Holocene. The most significant periodicities falling within the decadal (10-11 year) and bi-decadal (18-22 years) solar cycles and the Pacific Decadal Oscillation (PDO) (20-50 years). The fewest and shortest annual and seasonal periodicities are in the bi-decadal to ENSO range and occur in the early Holocene from 11,500 to 10,400 and 9,400 to 8,900 cal yr B.P. Predominant annual periodicities shift back and forth between bi-decadal to PDO range to the decadal to bi-decadal range on a millennial-scale until 4,100 cal yr B.P. Annual patterns are less persistent from 4,100 cal yr B.P. to ~1,800 cal yr B.P. Seasonal patterns are similar to the annual ones in about half of the intervals. Dissimilarities between the varve and both seasonal records are primarily due to the weakness of one or both of the seasonal signals. Persistent decadal and bi-decadal signals in summer sedimentation from 6,500 to 7,700 cal yr B.P. and a coincident summer insolation maximum suggest a solar influence on the strength of the summer monsoon. There is little to no periodicity the summer sedimentation from 4,200 to 2,400 cal yr B.P. associated with overall lower and highly variable sedimentation. Periodicities in winter sedimentation are not as persistent and undergo abrupt shifts on a centennial to millennial scale. In general, fewer periodicities and lower frequencies tend to be associated with intervals of lower winter sedimentation. Low overall sedimentation at 9,300-9,000, 8,850-8,700, and 8,100-8,000 cal yrs B.P. appear to indicate weak monsoonal circulation. DROUGHT AND FAILURE OF PACIFIC-NORTH AMERICAN MONSOON DURING THE HOLOCENE FROM VARVED SEDIMENTS AT DSDP SITE 480 IN THE GULF

OF CALIFORNIA

ANN MARIE HARRIS AND ROBERT E. KARLIN Department of Geological Sciences, University of Nevada, Reno, NV 89519

thelite@ix.netcom.com Laminated sediments from DSDP Site 480 in the central Gulf of California contain a virtually continuous sedimentation record of monsoonal circulation down to a sub-annual scale and multidecadal- to centennial- scale summer droughts for the interval 11,500 to ~1,850 cal yr B.P. Site 480 lies under the influence of the Pacific-North America Monsoon (PNAM) with seasonally reversing winds, resulting in winter upwelling with high diatom productivity and 16

summer deposition of aeolian terrigenous material. Varve counts and laminae width measurements were made using densiometry of x-radiographs of sliced and half core sections to create a varve chronology and sedimentation history. An independent chronology was constructed from available radiocarbon and isotopic ages. The age model was refined by comparing paleosecular variation records at Site 480 with well-dated inclination records from Saanich Inlet, British Columbia. The varve chronology agrees very closely with the age model, suggesting that the varve record is complete and that seasonal monsoonal circulation has persisted over the Gulf throughout the Holocene. Both monsoonal intensity and summer precipitation generally have increased since the early Holocene as evidenced by increasing annual and seasonal sedimentation toward the present. The long-term trend does not appear to be due to compaction. Superimposed on the long-term linear trend are several large-scale, episodic fluctuations 500-1,000 years in length. Weaker winter upwelling and lower summer terrigenous sedimentation rates occur during the intervals from 11,500-10,400, 9,400-9,000, and 3,850-3,100 cal yr B.P. The interval from 8,200-6,700 cal yr B.P. also averages less than the long-term linear trend and includes two abrupt, multi-centennial-scale drops in sedimentation, suggesting drought conditions. The 8,200-8,000 yr B.P. drought correlates with a high latitude 8,200 yr B.P. cold event observed elsewhere. A similar drought present at ~7,200 yr B.P. may be more regional in nature. Most multi-decadal to centennial-scale declines in sedimentation at DSDP Site 480 correspond to periods of intense drought along the northern border of the PNAM region.

THE RESPONSE OF THE CALIFORNIA MARGIN TO RAPID CLIMATE CHANGE: INFLUENCES OF CHANGING ATMOSPHERE AND OCEAN CIRCULATION ON

THE MARINE ENVIRONMENT

INGRID L. HENDY Department of Geological Sciences, University of Michigan, Ann Arbor, MI 48109-1063

ihendy@umich.edu The discovery of large abrupt submillennial scale climate change in ice cores fundamentally altered our perception of climate change. Paleoclimate research objectives following this discovery have focused on identifying the forcing mechanisms behind these profound changes. ODP Site 893, Santa Barbara Basin (SBB) produced the first rapid climate change record outside the Atlantic. The impact of this discovery was twofold. First, it demonstrated a possible global influence of these events, and second, amplification of forcing 17

mechanisms had to have occurred to produce the magnitude of changes identified in this key record. In recent years, research along the California Margin has focused on further understanding aspects of local atmosphere and ocean circulation and their response to submillennial climate variability. Our understanding of the California Margin marine environmental response to submillennial scale climate variability during Marine Isotope Stage (MIS) 3, the deglaciation and the late Holocene is summarized. The warm climatic events of MIS 3 in SBB were characterized by sea surface temperatures (SSTs) warmer than 12 �C and a planktonic foraminiferal assemblage dominated by dextral (right coiling) Neogloboquadrina pachyderma and Globigerina bulloides. Along with these foraminiferal assemblage changes, the isotopic differences suggest a stable, strong thermocline. The oxygen minimum zone (OMZ; 400 and 1000 m water depth) was present and at ~ 400 m water depth subsurface water was warm (~ 6 �C). The Point Conception upwelling cell was active, and the high nutrient content of upwelled water stimulated G. bulloides blooms and marine snow production. Cool intervals of MIS 3 in comparison were < 8�C and associated with an isothermal water column (no apparent thermocline) demonstrated by a sinistral (left coiling) N. pachyderma, Globigerina quinqueloba, and Globigernita glutinata assemblage and the isotopic similarity between G. bulloides and N. pachyderma. Subsurface water was cool (~ 3 �C) and well-oxygenated, suggesting the production of an intermediate water mass in the North Pacific. The transition between warm and cool intervals was rapid (< 50 years), but more abrupt during cooling. The sequence during warming was as follows: Gradual warming of intermediate waters occurred 60 to 200 years prior to abrupt surface waters warming and reduced ventilation. A further reduction in subsurface ventilation and increased productivity then followed surface warming. Deglaciation was for the most part similar to MIS 3 with the rapid warming of the Bølling-Ållerød resembling the warm intervals of MIS 3, and the Younger Dryas akin to the cool intervals. Several distinct differences occur however: First, productivity during the Bølling was more pronounced than any other interval of the last 60,000 years. Second, productivity dropped during the Ållerød and Younger Dryas, however, oxygen concentrations increased only during the Younger Dryas. This is the strongest evidence yet for changes in North Pacific Intermediate Water (NPIW) production ventilating the California Margin OMZ. Climate change during the last two millennia follow some of the patterns that occurred during the last glacial. During the Medieval Warm Period (A.D. 800 to 1200) surface waters were warmer than 10�C and well stratified as indicated by an assemblage dominated by dextral N. pachyderma and isotopic differences between G. bulloides and N. pachyderma. In contrast, during the Little Ice Age (LIA; A.D. 1300 to 1700) SSTs rapidly cooled (> 8 �C for a portion of the year) and the water column became isothermal. Differing to MIS 3, these warming and cooling trends are not apparent in the planktonic foraminiferal isotope record and upwelling increased during the LIA. The bi-modal shift between dextral N. pachyderma-G. bulloides and sinistral N. pachyderma-G. quinqueloba assemblages has been associated with atmospheric pressure system changes in the Pacific. A strong North Pacific high pressure system during warm intervals intensified upwelling and strengthened the California Current System (CCS) such that the California Countercurrent advected warm water into SBB. During cool intervals, a strong Aleutian Low resulted in subpolar characteristics of water advected into the CCS. Cooling of the North Pacific also led to greater NPIW production in the Sea of Okhotsk, and subsequent ventilation of the OMZ along the North American margin. In contrast to the last glacial, cooling 18

during the LIA may be related to strengthening of global atmospheric circulation rather than relative changes in the North Pacific High and Aleutian Low.

A HYPOTHESIS FOR THE 1976-1977 WESTERN DROUGHT

JAMES JOHNSTONE Department of Geography, University of California, Berkeley, CA 94720

jajstone@berkeley,edu During the winter of 1976-77, the western U.S. experienced the most severe meteorological drought in over a century of instrumental records. The anomalous circulation pattern surrounding this drought is discussed, and a hypothesis is presented for its occurrence, based on the connection of related biennial and decadal global oscillatory waves. The nature of the cycle coupling tended to inhibit severe drought winters during most of the 20th century; the anomalous conditions of 1976-77 are attributed to a temporary alteration of the long-term phase relationship.

THE 20,000 YEAR LAKE RECORD OF THE SALTON BASIN: APPLICATIONS OF AMS 14C, δ 18O, δ13C AND δ87Sr ON THE LAKE TUFA DEPOSITS

HONG-CHUN LI (1,2)

(1) Department of Earth Sciences, National Cheng-Kung University, Tainan, Taiwan 701 (2) Department of Earth Sciences, University of Southern California, Los Angeles,

CA 90089-0740 hli@usc.edu

Investigation of δ18O and salt content of surface waters in the Salton Basin, California, shows that there is a strongly positive correlation between δ18O values and salt contents and that the δ18O of Salton Sea is much heavier than that of its input waters due to strong evaporation. The 87Sr/86Sr ratio of Salton Sea is very close to that of the Colorado River input, but much lower than that of Whitewater River which is another major tributary. During the past four years, we have made 37 AMS 14C measurements on samples including lake tufa, mollusk shell and charcoal from the ancient Lake Cahuilla, Salton Basin. Three tufa slabs, LC-1 (64-cm thick), SST-1 (19-cm thick) and SST-5 (38-cm thick), reveal 14C ages of 1,7840±80 to 1,310±30 yr B.P. (15 dates), 4,790±25 to 1,365±20 yr B.P. (5 dates) and 7,080±20 to 1,265±15 yr B.P. (9 dates), respectively. All age results show minor influence of reservoir effect and are in stratigraphic order. These dates indicate that Lake Cahuilla in the Salton Basin had a continuous history during 20,500-800 yr B.P. when the tufas grew continuously in the lake. Between 18,000 and 1,300 yr B.P., Lake Cahuilla level was above -24m a.s.l. which was at least 50 m higher than modern Salton Sea. Between 7,000 and 1,300 yr B.P., the lake stayed at the highest level — 12 m a.s.l. which was 87 m higher than Salton Sea. During these periods, Colorado River was connected with the Salton Basin. After 1,300 yr B.P., the lake dropped its level rapidly, probably due to both arid climates and a change in course of Colorado River. Salton Basin became a part of Sonora Desert after 1,300 yr B.P. Human occupation on the lake shore below 12m a.s.l. should be before 7,000 yr B.P. or after 1,300 yr 19

B.P. The Sr isotope measurements made on LC-1 and SST-1 exhibit relatively constant 87Sr/86Sr ratios throughout the past 20,000 years and the average is very close to that of the Colorado River input. This means that the Colorado River input was the dominant water source for Lake Cahuilla at least during 20,500-800 yr B.P. Hence, the paleoclimatic proxies retrieved from the Lake Cahuilla tufas in the Salton Basin can decipher the discharge and flood history of the Colorado River under the influence of climate variability in the Colorado River drainage basin. Theδ18O and δ13C values of LC-1 and SST-1 indicate that the lake experienced closed and/or overflowed conditions between 20,500 and 800 yr B.P. The variations in δ18O and δ13C values reflect changes in the relative humidity in Salton Basin and variations in the Colorado River inflow and hence climatic conditions in the Colorado River drainage basin. The tufa records exhibits palaeoclimates of Salton Basin as follows: (1) Relatively dry climate between 20,000 and 18,000 yr B.P. was followed by a wet climate between 18,000 and 17,000 yr B.P.. (2) Decreasing effective moisture (precipitation − evaporation) from 17,000 to 16,000 yr B.P. corresponded to Heinrich 1 event, whereas effective moisture increased at ca.15,500 yr B.P. during the Trans-US Wet Period. (3) From 15,200 to 14,500 yr B.P., the effective moisture decreased, and followed by a strongly increased effective moisture until 14,000 yr B.P. probably corresponding to the Bølling-Ållerød interstadial. (4) Climate in Salton Basin during the Younger Dryas was wet/cold. (5) Wet/warm early Holocene corresponding to the Maximum Effective Moisture Period perhaps brought about by the strengthening of North American monsoon. (6) Toward 2,000 yr B.P., the climate became dryer, apparently resulting from weakening of the monsoons, especially around 6,200 and 2,500 yr B.P. (7) A wet regime occurred in the basin during the Medieval Warm Period. CHANGES IN SEDIMENT SOURCE AND CLIMATE IN THE NORTHERN REACH OF

SAN FRANCISCO BAY OVER THE PAST MILLENNIUM

F. MALAMUD-ROAM (1) AND B. LYNN INGRAM (2)

(1) Department of Geography, University of California, Berkeley, CA 94720 (2) Department of Earth and Planetary Sciences, University of California, Berkeley, CA 94720

fmalamud@eps.berkeley,edu The geochemistry of inorganic sediments (silts and clays) from a sediment core from Petaluma marsh (the northern reach of San Francisco Bay) were analyzed in order to assess the changes in sediment source over time. These sediments were carried in suspension and deposited on the marsh surface. The main sources for these sediments are local rivers (Petaluma), the Sacramento River, draining the northern central valley watershed, and the San Joaquin River, 20

draining the southern central valley watershed. Changes in the relative proportion of sediments from each watershed reflects overall climatic variability over the entire watershed region of the San Francisco Bay estuary. Sediment flux history of Petaluma marsh in the northern reach of the San Francisco Bay indicates considerable variability over the last ca. 900 years, and that variability has increased over the last 200 years. The sedimentary record from Petaluma indicates that detrital sediment influx was generally high from about 450 to 150 cal yr B.P., corresponding to the Little Ice Age, a period shown in other California climate records to have been cool and wet. The San Joaquin River appears to account for the majority of sediments for the early part of this period of high sediment influx (450 – 300 cal yr B.P. The Sacramento River became increasingly dominant as a source of sediments after 300 cal yr B.P. During the Medieval Climate Anomaly (1,000 – 600 cal yr B.P., a period shown to be warm and dry by many proxy climate records in California, including several records in San Francisco Bay, sediment flux from the Sacramento River to Petaluma marsh was reduced during most of this time, while the San Joaquin was generally moderately high. Local sediment flux was variable, though generally more often high during this time. The recent 200 years of sediment flux to Petaluma marsh shows considerably more variability than the prior 700-year period. In general, the Sacramento River’s contribution to the sediment load was greater, possibly reflecting human disturbance in the watershed, particularly hydraulic gold mining activity that flushed huge volumes of sediment through the Sacramento River system.

JOINT VARIABILITY OF GLOBAL RUNOFF AND GLOBAL SEA-SURFACE TEMPERATURES

GREGORY J. MCCABE (1) and DAVID M. WOLOCK (2)

(1) U.S. Geological Survey, P.O. Box 25046, Denver, CO 80225 (2) U.S. Geological Survey, 4821 Quail Crest Place, Lawrence, KS 66049

gmccabe@usgs.gov A monthly water-balance model is used with global monthly temperature and precipitation data (at a 0.5 grid resolution) to compute time series of annual runoff for 1905-2002. The global runoff is aggregated to a 5o grid resolution and is analyzed jointly with global SST data (at a near10o grid resolution) using principal components analysis (PCA) with varimax rotation to identify the primary modes of variability in these data sets. The first three components from the rotated PCA explain almost 30% of the total variability in the combined runoff/SST data. The first component explains 13% of the total variance and primarily represents long-term 21

trends in the data. The long-term trends in SSTs are most evident as warming in the Atlantic and Indian oceans. The associated long-term trends in runoff suggest increasing flows for much of North America and Australia, and in parts of South America and Eurasia; decreasing runoff is most notable in Africa and southern Asia. The second component explains 11% of the total variance and reflects variability of the El Niño-Southern Oscillation (ENSO) and its associated influence on global annual runoff patterns. The third component explains 6% of the total variance and indicates a response of global annual runoff to variability in tropical Atlantic SSTs. The association between runoff and tropical Atlantic SSTs appears to explain an apparent step-like change in runoff that occurred around 1970 for a number of continental regions.

EPISODIC 20TH CENTURY RECRUITMENT OF LIMBER AND BRISTLECONE PINES IN THE WHITE MOUNTAINS, CALIFORNIA

CONNIE MILLAR, BOB WESTFALL, AND DIANE DELANY

U.S.D.A. Forest Service, Sierra Nevada Research Center, Albany, CA 94710 cmillar@fs.fed.us

Subalpine forests are assumed to respond to warming temperatures by shifting upslope relative to their current elevational zones. Given that minimum temperatures in the Sierra Nevada have increased by > 1�C over the 20th century, we might expect a gradual recruitment upslope of conifer forests paralleling temperature trends. In the White Mountains of California, subalpine forests comprise bristlecone pine (Pinus longaeva, PiLo) and limber pine (P. flexilis, PiFl); current PiLo treeline is higher than PiFl by ~ 200 m, especially on dolomitic soils, which are thought to competitively exclude PiFl. Lower treeline elevation of both species is similar, although PiLo extends lower than PiFl on calcareous soils. We aged recruitment (trees < 100 years) along transects in three elevational zones: above upper treeline, middle elevations, and below lower treeline. PiFl exceeded PiLo in abundance of recruits at all sites. At lower elevations, recruitment was limited to narrow, west- and north-facing ravines; at middle elevations, recruitment occurred within the general forest zone and also into sagebrush-dominated depressions and basins of the mid-elevation plateaus. Greatest abundance of recruitment was at high elevations above current treeline. Surprisingly PiFl abundance exceeded PiLo by > 400%. On dolomite substrates, PiFl was often the only recruiting species at the highest elevations -- as much as 100 m above current PiLo treeline and 300 m above current PiFl treeline. While some recruitment occurred early in the 20th century, over 80% occurred as an episodic pulse between 1965 and 1992; a smaller pulse occurred in 1995-2001. The negative phase of the AMO correlates strongly with the larger recruitment pulse. Highest positive monthly correlations between climate and recruitment are with September precipitation, winter maximum temperature, and summer minimum temperature. Recruitment is also greater under high annual minimum temperature and higher precipitation. Combinations of these conditions may contribute to successful episodic seedling establishment above treeline at these very high elevations. While we are unclear why the species differ in response, PiFl appears to be advancing episodically in dominance over PiLo in the White Mountains, and particularly in leap-frog fashion above current PiLo treeline.

HOLOCENE LANDSCAPE RESPONSE TO SEASONALITY OF STORMS IN THE 22

MOJAVE DESERT

DAVID M. MILLER (1), SHANNON A. MAHAN (2), JOHN P. MCGEEHIN (3), JOHN A. BARRON (1) AND LEWIS OWEN (4)

(1) U.S. Geological Survey, 345 Middlefield Road, Menlo Park CA 94025

(2) U.S. Geological Survey, P.O. 25046, Denver Federal Center, Denver, CO 80225 (3) U.S. Geological Survey, 12201 Sunrise Valley Drive, Reston, VA 20192 (4) Department of Geology, University of Cincinnati, Cincinnati, OH 45221

dmiller@usgs.gov Ongoing research in Quaternary landscape evolution of the Mojave Desert is advancing the understanding of geology-ecosystem links, climate history, and earth-surface processes. High-resolution geochronology allows us to map deposits at time scales that connect the history of landscape response to distinctive climate periods. Widespread alluvial fan deposits present a significant challenge for dating due to lack of suitable organic material for radiocarbon dating. To overcome this problem, we use luminescence and U/Th series methods to date these deposits. We find that the Mojave Desert depositional record during the Holocene is similar to previously determined climate-landscape records of the region. Evidence from ephemeral rivers and lakes indicates periods of sustained stream flow during the Little Ice Age (ca. A.D. 1300-1800) and during a brief cool interval within the Medieval Warm Period at about 1050 A.D. Periods of alluvial fan aggradation across the desert are approximately 15,000-11,000 and 6,000-3,000 years ago, corresponding to nearby Gulf of California marine records of dominance by tropical diatoms and silicoflagellates. We suggest that periods of increased sea surface temperatures in the Gulf of California experience more frequent and more intense summer and fall monsoons and hurricanes. Sustained alluvial fan aggradation may be driven by these periods of intense storms. The position of the Mojave Desert near the Pacific coast may promote a partitioning of landscape process responses to climate forcings that vary with seasonality of the dominant storms. Winter Pacific frontal storms create river flow and ephemeral lakes, whereas periods of intense summer monsoons cause alluvial fan aggradation. Ecosystem effects of the two proposed landscape processes are poorly known; however, wood rat midden studies could target periods of climatic transition and possibly isolate vegetation response to different landscape processes. River and lake expansion may promote integration of formerly separate riparian and aquatic habitats. Alluvial fan aggradation may greatly increase the landscape occupied by early successional species and decrease connections among older parts of the landscape. HISTORICAL TRENDS IN MOSQUITO POPULATIONS IN RESPONSE TO CLIMATE

DATA IN THE WESTERN UNITED STATES

CORY MORIN AND ANDREW C. COMRIE Department of Geography, University of Arizona, Tucson, AZ 85719

coryWM417@aol.com Climate is an important control on populations of disease vectors such as Aedes aegypti and Culex quinquefasciatus mosquitoes. Improved understanding and modeling of climate 23

effects on mosquito populations is needed to help predict and mitigate disease epidemics. This study reports on the development of a dynamic mosquito population model driven in part by climatic data. Model inputs include daily temperature, precipitation, and evaporation rates that influence mosquito population development and mortality in relation to temperature, population density, and water availability. These factors are considered through all phases of the mosquito life cycle from ovulation, though the larvae and pupae stages, and finally emergence into an adult mosquito. In this study, the model is utilized in order to construct historical predictions of A. aegypti and C. quinquefasciatus populations in select areas throughout the western United States. In addition, data of future climate projections are input into the model in order to assess the affect that climate change may play on the distribution of these vectors. Using this method produces both a geographic and historical analysis of the ecological factors affecting population patterns of these mosquitoes. Ultimately, this study seeks to evaluate the role climate variability has had and will have on the distribution of important disease vectors in the western United States.

TIMING AND SPATIAL EXTENT OF MIDDLE HOLOCENE ABRUPT CLIMATE CHANGE

CARRIE MORRILL

National Climatic Data Center, Paleoclimatology Branch, Boulder, CO 80305 carrie.morrill@noaa.gov

A number of recently-published paleoclimate records suggest that the middle Holocene (about 6,000 to 4,000 cal yr B.P.) was a period of particularly widespread abrupt climate change. These changes punctuated long-term trends in temperature and precipitation through the Holocene and could have been a rapid response of the climate system to steadily decreasing summer insolation. Detecting abrupt climate changes in individual records is difficult, however, because the time-series also reflects thresholds and non-linearities that are inherent to the proxies. To assess the evidence for abrupt climate change during the middle Holocene, a compilation of more than a hundred previously published proxy records from around the globe is used in this research. The records include all types of proxies (i.e., pollen, ice cores, lake sediments, marine sediments, loess, peat, speleothems). All records have a resolution of 150 years or better, well-defined age models, and clear proxy interpretations. Abrupt climate changes are identified objectively through the use of several statistical tests. Some coherent regional 24

abrupt changes are apparent, but evidence for globally-synchronous changes is not currently strong. These results will be used to discuss hypotheses for the causes of middle Holocene abrupt climate change.

CALIFORNIA FISHERIES AND ECOLOGICAL RESPONSES TO ABRUPT OCEAN CLIMATE CHANGES

JERROLD G. NORTON (1), JANET E. MASON (1), AND SAMUEL F. HERRICK (2)

(1) Environmental Research Division, S.W.F.S.C., N.O.A.A. 1352 Lighthouse Avenue, Pacific Grove, CA 93950

(2) Fisheries Research Division, S.W.F.S.C., N.O.A.A. 8604 La Jolla Shores Drive, La Jolla, CA 92037

jerrold.g.norton@noaa.gov The responses of California fish populations and their ecosystems processes to abrupt ocean climate changes can be rapid, delayed, or both. The delay in response depends on the mobility, life cycle and abundance of each species when climate change occurs. Rapid responses may be observed within a year in mobile short lived dolphinfish (Coryphaena hippursus). Less rapid responses may not be apparent for 5 to 10 years in sedentary long-lived rockfish (Sebastes sp.) populations. The impact of the responses of commercially important species to abrupt climate changes on the fisher's income and influence on local economies depends on the value of the fisheries. The revenue produced by a fishery comes from the abundance of an available species and from the revenue generated per unit harvested. California fisheries have produced the greatest revenue when both abundant lower value and less abundant higher value fisheries have been available. Market squid (Loligo opalescens), accounted for 44% of the total commercial weight landed and 25% of the total exvessel revenue in 1999; this squid is a high abundance and relatively low value species that has a short life cycle and a conspicuous and rapid fisheries response to climate changes. Swordfish (Xiphias gladius) have contributed less than 5% of the total landings weight, but have high unit value and contributed more than 10% of the total California landings revenue in 1986. Swordfish have a longer generation time and prolonged response to environmental changes, but their mobility and their ability to feed over a wide range of temperatures influence the response of the swordfish fishery to environmental changes. When physical environmental changes are multi-decadal, ecological adjustments accumulate and species that are commercially abundant in one phase may be absent in commercial quantities in the cycle's reverse phase (e.g. Sardinops sagax, Thunnus alalunga). The 20th century has seen alternating cool and warm periods, with cool periods before 1920 and between 1960 and 1975. Although the California fisheries were not fully developed before 1940, it appears from our analysis of commercial species harvested throughout the 20th century that the warm periods and the cool periods both present characteristic ecosystems associated with characteristic fisheries opportunities.

SIERRA NEVADA SPELEOTHEMS: POTENTIAL FOR HIGH-RESOLUTION ARCHIVES OF CHANGES IN ATMOSPHERIC CIRCULATION OVER

WESTERN NORTH AMERICA

25

JESSICA L. OSTER (1), ISABEL P. MONTAÑEZ (1), AND WARREN D SHARP (2) (1) Geology Department, University of California, Davis, CA 95616

(2) Berkeley Geochronology Center, Berkeley, CA 94709 oster@geology.ucdavis.edu

Speleothems from caves developed in limestone and marble terranes of the western Sierra Nevada from 37o to ~41° N allow reconstruction of high-resolution paleoprecipitation records for western North America near the boundary between Hadley and Ferrel atmospheric circulation cells. Global circulation models and paleoclimate proxy records suggest that the polar cell expanded during northern hemisphere cold periods of the last glacial cycle and the Holocene, pushing the polar jet stream southward over the southwestern United States. We plan to test this long-standing hypothesis by using these speleothems to track changes in temperature and precipitation amounts that are expected to accompany latitudinal movement of the polar jet stream since the last glacial maximum (LGM). Existing paleoclimate records from Sierra Nevada lake sediments provide insight into the past hydrologic balance of the region, but do not reveal the specific causes of effective moisture change or their link to modified ocean-atmosphere circulation. An actively forming stalagmite from Moaning Cave (38° N) precipitated without visible discontinuity for the last 18,000 years based on 230Th/U dating. Analyses of O and C isotopes indicate that modern speleothem and drip water are in isotopic equilibrium. The O and C isotopes display large variations (~2‰ for O and ~5‰ for C) along the speleothem’s growth axis. These variations correlate with substantial changes in Mg and Sr concentrations. Comprehensive evaluation of modern cave conditions and water chemistry will delineate seasonal and inter-annual variations in source water and fluid-rock interaction, and further test for kinetic isotopic fractionation in the modern speleothem. Preliminary results suggest that carefully calibrated, high-resolution records from Sierra Nevada speleothems can help test the hypothesis of polar jet stream migration since the LGM and clarify the environmental parameters recorded by geochemical proxies in speleothems.

A POSSIBLE COSMOCLIMATOLOGICAL DRIVER OF OBSERVED CYCLIC HOLOCENE CLIMATE CHANGE AND MARINE

PRODUCTIVITY IN THE NORTHEASTERN PACIFIC

R. TIMOTHY PATTERSON (1), ANDREAS PROKOPH (2), ALICE CHANG (3), HELEN M. ROE (4), NATALIA VAZQUEZ RIVEIROS (1), AND ANDREW WIGSTON (1)

(1) Department of Earth Sciences, Carleton University, Ottawa, ON, K1S 5B6 Canada (2) SPEEDSTAT, Ottawa, ON, Canada

(3) School of Earth and Ocean Sciences, University of Victoria, Victoria, BC, V8W 3P6 Canada (4) School of Geography, Archaeology and Palaeoecology, Queen’s University of Belfast,

Belfast, BT7 1NN United Kingdom tpatters@earthsci.carleton.ca

Annually deposited laminated sediment couplets deposited under anoxic conditions in fjords along the British Columbia coast provide a high-resolution archive of Holocene climate variability and marine productivity. In the Seymour-Belize Inlet Complex (SBIC) on the 26

mainland coast and in Effingham Inlet, Vancouver Island, diatom-dominated laminae are deposited when marine productivity is higher in the spring to fall, and darker clay-dominated laminae are laid down under higher precipitation conditions during winter. Wavelet transform and other time series analysis methods were applied to sediment color (i.e. gray-scale values) line scans obtained from x-ray images of cores and compared with global records of cosmogenic nuclides (14C and 10Be), as well as the Ice Drift Index (hematite-stained grains) record to detect cycles, trends and non-stationarities in the sedimentary record. We carried out similar analyses on foraminifera, diatom, dinoflagellate, and fish scale records recovered from these same cores. The results indicate that the marine productivity and sedimentary record of the northeast Pacific responded to abrupt changes and long-term variability in climate that can be linked to external forcing (e.g. variations in cosmic ray influx due to solar magnetic activity). For example, a strong cooling in the NE Pacific at ~ 3,550 yr B.P. that is correlated with widespread neoglaciation in the region can be correlated to a weakening of high-frequency (50-150) year pulses at the Gleissberg solar cycle band. Episodes of low sun activity are characterized in the sedimentary record (e.g. 3,350, 2,750, and 2,350 yr B.P.) of the cores by intervals of clay-rich and thick laminae that were deposited under unusually wet conditions. These intervals of higher precipitation may have been related to a regional intensification of the Aleutian Low (AL) caused by an eastward migration of the center of action (COA) of the AL, which occurs during intervals of solar minima, similar to what occurred during the “Little Ice Age” (e.g. Spörer, Maunder, and Dalton minima). Dryer conditions prevail in the region when the COA of the AL migrates westward and the COA of the North Pacific High (NPH) migrates northward during intervals of solar maxima. These NPH and AL COA changes greatly impact the influence of open ocean upwelling in Effingham Inlet and estuarine circulation in the SBIC.

EVIDENCE OF THE DIFFERENTIAL RESPONSE OF LAKES IN THE SIERRA NEVADA, CALIFORNIA DURING THE YOUNGER DRYAS CHRONOZONE

DAVID F. PORINCHU (1), GLEN M. MACDONALD (2), KATRINA A. MOSER (3),

AARON P. POTITO (4), AND AMY M. BLOOM (5) (1) Department of Geography, The Ohio State University, Columbus, OH 43210 (2) Department of Geography, University of California, Los Angeles, CA 90095

(3) Department of Geography, University of Western Ontario, London, ON, N6A 5B7 Canada (4) Department of Geography, NUIG, Galway, Ireland

(5) Department of Geography-Geology, Illinois State University, Normal, IL 61790-4400 porinchu.1@osu.edu

Sub-fossil midge stratigraphies were developed for five small lakes in the eastern Sierra Nevada, California. Application of a midge-based inference model for surface water temperature provided quantitative estimates of the thermal conditions that existed during the Younger Dryas chronozone (12,900 – 11,600 cal yr B.P.; YD). The sediment sequences were recovered from lakes spanning an elevation gradient of approximately 800 m, with the low elevation site surrounded by Jeffrey pine woodland, mid-elevation sites situated in upper Montane forest and the high elevation sites characterized by sub-alpine vegetation. The chironomid-based reconstructions suggest that post-glacial climate amelioration was spatially and temporally 27

complex within the Sierra Nevada. All sites experienced a warming immediately following de-glaciation, with water temperatures increasing approximately 2-3 oC between 14,000 and 13,000 cal yr B.P. Surface water temperatures at mid-elevation sites were depressed ~ 3 oC during the YD. However, it appears that surface water temperatures at high elevation sites increased slightly or remained stable during this interval; whereas, the low elevation site experienced greater variability in surface water temperature. Temperatures at all lakes stabilized during the early Holocene. Paleohydrological changes inferred from diatom and stable isotope records developed for the mid-elevation sites will also be discussed.

THE WEATHER AND CLIMATE OF PACLIM YEAR 2006-2007

KELLY REDMOND

Western Regional Climate Center, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512-1095

kelly.redmond@dri.edu After an auspicious and even exceedingly soggy or white start in parts of the West, the main feature of the 2006-07 winter turned out to be a major deficiency in mountain snowpack, and an extremely dry winter in the Southwest. In southern California this driest winter on record follows only two years after the wettest on record. In the Sierra Nevada, the effects of two immediately prior significantly wet winters greatly alleviated water angst, at least for summer 2007. The summer heat wave of 2006 was outstanding in several aspects, and more mundane in others, with some very dramatic differentials along the coast. The snowfall deficit encompassed much of the drainage of the upper Colorado River, and the longstanding and major drought there once again intensified after a brief increase two years ago. Reservoir storage on Lake Powell and Lake Mead now stands just about half full. Reminiscent of 2004, a warm and dry March in the interior West produced a rapid loss of snowpack, and the inflow forecast to Lake Powell fell steadily to 50 percent of average. Spring months have been warming over the past two decades, but the last few years have brought cool episodes along the West Coast. Arizona remained dry all winter, but parts of New Mexico experienced significant precipitation later in the winter. Western and Colorado Basin temperatures in 2006 continued their rising trend, and the period 2000-2006 shows decidedly different temperature anomalies for the eastern and western United States. Recent papers have fueled speculation as to the nature of the Southwest drought: Is this "just another typical large drought" or is this "a new type of drought"? Hemispheric patterns will be discussed briefly. The national fire acreage in 2006 was the greatest on record, and expectations for 2007 are for another active year. The wet north / dry south precipitation pattern in the West occurred despite a mild El Niño that many widely presumed would produce the opposite effect. This later gave way to a nascent La Niña whose prospects for winter 2007-08 are not fully clear, but would portend yet another dry winter in the southern West.

IMPACTS OF CLIMATE CHANGE AND LAND USE ON THE NAVAJO NATION IN THE SOUTHWESTERN UNITED STATES

MARGARET HIZA REDSTEER

U.S. Geological Survey, Flagstaff Science Center, Flagstaff, AZ 86011 28

mhiza@usgs.gov Native Americans of the southwest, including the people of the Navajo Nation, live in an ecologically sensitive arid to semi-arid region with limited resources. In this region, as on other Native lands, traditional people live a subsistence lifestyle that is closely tied to, and dependent upon landscape conditions. Global change, with regard to climate, includes significant changes in long-term average temperature and precipitation. Many studies have shown that in the past couple of decades, average temperatures have risen (compared to historic averages) in many parts of the world, including the southwestern United States. During this past decade, the southwestern United States, including the Navajo Nation, has also been experiencing a major drought. Global warming may already be contributing to drought severity on the Navajo Nation and adjacent lands. The dual effects of increased temperatures and changing patterns of precipitation are likely to have profound effects (both geologic and biologic) on this semiarid landscape. For the region of the Navajo Nation and those lands at a similar latitude, the amount of effective precipitation, the moisture not lost to evaporation, decreases by approximately 5 cm for every 1 oC increase in temperature. These calculations highlight the clear connection between climate change and drought severity for the western United States. Annual rainfall in many parts of the Western Navajo Nation averages between 12.7 and 18 cm in “normal” years, but has been as low as 2.5 to 5 cm in many areas during recent drought years. In addition, average annual snowfall for the Navajo Nation has decreased significantly since the early part of the 20th century. Snowfall and snow pack are important for their contribution to recharge of aquifers and water storage in reservoirs. Our work has found that some land use practices, such as importing weedy hay during periods of drought, and off-highway vehicle use, are compounding the sobering effects of drought and climate change to mobilize surface sediment and degrade rangeland. Sand dunes that cover approximately two-thirds of rangeland on the Navajo Nation have become partly active to active in many areas in recent years. These changes can be attributed to changing climatic factors.

EXAMINATION OF DROUGHT TOLERANCE IN PINYON PINE USING ANCIENT DNA

BETHANY RIGGINS (1), AMY WHIPPLE (1), KRIS HASKINS (1), AND

KENNETH COLE (2) (1) College of Engineering and Natural Science, Northern Arizona University, AZ 86011

(2) U. S. Geological Survey, Southwest Biological Science Center, P.O. Box 5614, Northern Arizona University, AZ 86011

bmr34@nau.edu Due to a recent drought in Arizona, thriving pinyon forests have turned into sparse forests. The purpose of this study is to discover why many pinyon trees lived, while many others died. This study specifically focuses on the aspect of genetic adaptations to drought by examining drought tolerance genes through time. Comparing ancient and recent DNA will allow us to look for patterns between genetic variation and climate. It is expected that genes specific to drought tolerance have changed as surrounding climate has changed. Packrat middens were the source of needle tissue from trees which have been dead for hundreds of years. Twenty-two 29

drought specific genes will be sequenced to study how and if pinyon drought genes have significantly changed over time. The genes have been selected but not yet. We expect to see variation between genes, but the cause of that variation is unknown at this time. In addition to genetic change detection, a second goal is to test if the genetic change is the result of neutral variation or natural selection.

POSTGLACIAL RELATIVE SEA-LEVEL AND CLIMATIC HISTORY OF THE SEYMOUR-BELIZE INLET COMPLEX, BRITISH COLUMBIA: BIOSTRATIGRAPHIC EVIDENCE FROM ISOLATION BASINS

HELEN M. ROE (1), CHRISTINE T. DOHERTY (1), R. TIMOTHY PATTERSON (2), AND

GLENN A. MILNE (3) (1) School of Geography, Archaeology and Palaeoecology, Queen’s University of Belfast,

Belfast, BT7 1NN, United Kingdom (2) Department of Earth Sciences, College of Natural Sciences, Carleton University, Ottawa,

ON, K1S 5B6, Canada (3) Department of Earth Sciences, Durham University, Durham, DH1 3LE,

United Kingdom h.roe@qub.ac.uk

The late glacial and Holocene relative sea-level (RSL) history of the sparsely populated central mainland coast of British Columbia is poorly understood, yet this area remains critical for testing models of post-glacial rebound following deglaciation of the Cordilleran ice sheet and for assessing residual crustal movements. This paper presents the results of a recent study of the late glacial and Holocene RSL history of the Seymour-Belize Inlet Complex (SBIC) of central British Columbia, approximately 50 km north of Vancouver Island. Detailed sedimentological and biostratigraphic analyses of sediment cores from three coastal isolation basins (Woods Lake, Two Frog Lake, and Tiny Lake), ranging in sill elevation from 2.13-3.59 m MTL, provide new data from an area previously devoid of RSL information. Diatom analysis reveals that the lake basins were inundated by the sea and subsequently isolated during the late glacial at ca. 11,800 14C yr B.P. The RSL rose during the early Holocene and breached the 2.13 m sill at Woods Lake at ca. 8,000 14C yr B.P. Brackish conditions existed in the basin between 7,000-2,400 14C yr B.P. The RSL rose to 1.49 m MTL at ca. 2,400 14C yr B.P. Subsequently, RSL fell, with the final isolation of Woods Lake at ca. 1,900 14C yr B.P. The results have been used to produce a RSL curve for the SBIC for the late glacial and Holocene and draw comparisons with RSL estimates derived from geophysical models. The diatom record also includes evidence of regional cooling during Neoglacial times. The late glacial and Holocene RSL history recorded in this region compares well to that recorded in eastern Vancouver Island.

WY 2007: ARE THE GOOD YEARS OVER?

MAURICE ROOS AND MICHAEL ANDERSON California Department of Water Resources, P.O. Box 219000, Sacramento, CA 95821-9000

mroos@water.ca.gov 30

Both water year 2005 and 2006 were wet in northern California, mainly due to wet springs. Last year was sixth wettest of record (101 years) for the Sacramento River basin and ninth wettest on the San Joaquin River system. Statewide runoff was estimated to be 170% of average, quite a contrast to the 67% for the Colorado River inflow to Lake Powell. So far (as of mid-March), water year 2007 seems to be heading for a dry designation with projected runoff in the 60% range. Northwestern California seasonal precipitation has been near average, sharing somewhat in the bounty of the Pacific Northwest. But southern California has been extremely dry in spite of the El Niño year, with brush fires even in the winter. So far, February was the only month with above average in precipitation. December was fairly good, mostly in the north. January, which is our wettest month on average, was extremely dry. Downtown Sacramento had only 0.18 cm, compared to an average of about 10.7 cm, less than the previous record monthly low of 0.38 cm in 1889, more than a century ago. 1994 was the last “critically” dry year, although 2001 and 2002 were “dry”, the next category up. In spite of rainfall deficits, we don’t expect 2007 to be “critically” dry although it may well be dry. Carryover reservoir storage on October 1 from the last water year was 20% above average and that will save most of us this year from water shortages this year. Next year could be another story. Temperature was also an important factor this winter with an extended cold spell in January. The freeze caused major damage to Central Valley citrus crops and to winter vegetables in the southland desert areas, with losses estimated to be over $ 1 billion. Some San Joaquin Valley stations saw lows under 20 º F, especially on the west side with lows of 15 º F at Lemoore and Kettleman City. Looking at past years which had extended cold spells in winter, a wet or normal December followed by a dry January, there is a suggestion that the following year will also be quite a bit below average. This happened in 1932, 1948, 1949, and 1991. I am not sure if this is my selective imagination or whether it would be an explainable response to a strong mid-winter perturbation of the atmosphere. If so, it does not bode well for next year’s water supply. I leave it to you research folks to see if there is any dependable long range forecast potential in this in northern California for next year.

HIGH RESOLUTION STUDY OF PYRITE FRAMBOID DISTRIBUTION IN SANTA BARBARA BASIN SEDIMENTS AND IMPLICATIONS FOR WATER-COLUMN

OXYGENATION

JUERGEN SCHIEBER AND ARNDT SCHIMMELMANN Department of Geological Sciences, Indiana University, Bloomington, IN 47405-1405

aschimme@indiana.edu Monitoring of the concentration of dissolved elemental oxygen (i.e., ‘oxygenation’) of sub-sill waters in the Santa Barbara Basin (SBB) over the last few decades has consistently demonstrated sub-oxic bottom water conditions. However, anoxic conditions and hydrogen sulfide (H2S) are present below a few millimeters depth in the sedimentary column. The absence of euxinic conditions (i.e., anoxic-sulfidic: no free O2 but free H2S) in any part of the modern SBB water column makes this basin geochemically distinct from the Black Sea and other more oxygen-depleted basins. 31

The varved sediment record from the central SBB has been used extensively for high-resolution paleoceanographic reconstructions. The SBB and other oxygen-depleted basins are considered modern analogs for black shales. Geochemical proxies for paleo-redox conditions in black shales need to be tested against modern, well-constrained depositional systems like the SBB. Pyrite framboids are a common component in many anoxic marine sediments and are deemed to carry paleoceanographic information. Observations from modern Black Sea sediment yielded the hypotheses that (i) abundant small framboids below 5 µm diameter indicate euxinic (anoxic-sulfidic) bottom waters, and (ii) mean framboid sizes above 5 µm indicate suboxic or normally oxygenated bottom waters. We are using scanning electron microscopy to compile a detailed inventory of framboid occurrence in A.D. 1983-2004 SBB varves and compare framboid size distribution against the known history of SBB water column oxygenation. Our ground-truthing effort has shown that framboids in modern SBB sediment dominantly are in the 2-4 µm range and thus show closely similar size distributions to those measured in Black Sea sediments underlying an euxinic water column. Of course, the lower SBB water column is suboxic, and therefore we conclude that the framboid size distribution in modern sediment and in black shales has no diagnostic value for paleo-water column oxygenation. Our data further indicate significant variability in mean framboid size between successive varves that do not correlate with historic changes in SBB water oxygenation. We are evaluating other potential factors that could affect framboid size, for example (i) the availability of iron from clastic terrestrial input from the continent, (ii) the availability of freshly deposited organic matter nourishing the microbial community and influencing the redox profile near the sediment-water interface, and (iii) the depth of burial within the near-surface sediment that fosters the precipitation of framboids. The generally lower abundance of framboids in the youngest, uppermost varves suggests that it takes several years of mineralization before framboid formation goes to completion in a given sediment layer. After about five years framboid abundance appears to stabilize, providing further support of the view that framboids do not form in the water column and should not be considered recorders of water column oxygenation.

VEGETATION RESPONSES TO CLIMATE CHANGE: 11,000 YEARS AGO, 6,000 YEARS AGO, AND 21ST CENTURY VEGETATION SIMULATIONS FOR BERINGIA

SARAH L. SHAFER (1), PATRICK J. BARTLEIN (2), MARY E. EDWARDS (3),

AND STEVEN W. HOSTETLER (1) (1) U.S. Geological Survey, Corvallis, OR 97333

(2) Department of Geography, University of Oregon, Eugene, OR 97403 (3) School of Geography, University of Southampton, UK and

Alaska Quaternary Center, University of Alaska-Fairbanks, AK 99775 sshafer@usgs.gov

Future climate changes are projected to be particularly large for northern hemisphere high latitudes. Coupled atmosphere-ocean general circulation model (AOGCM) simulations produced for the recent Intergovernmental Panel on Climate Change Fourth Assessment Report (IPCC AR4) indicate warmer and wetter conditions for Beringia (eastern Siberia, Alaska, and northwestern Canada) at the end of the 21st century. The magnitude of these climate changes would significantly affect the vegetation of this region. To understand the climate-forced 32

dynamics of vegetation in Beringia, we compare paleovegetation simulations for 11,000 and 6,000 years ago with those for the future. We examine a variety of variables, including effective moisture and atmospheric CO2 concentrations, to explain the simulated differences in vegetation response for the three time periods. Paleoclimatic data for the vegetation simulations are derived from model experiments conducted with a regional climate model, RegCM2, which was run with time-varying boundary conditions from Genesis 2.0, an atmospheric general circulation model with a mixed-layer ocean. Future climate data are derived from AOGCM simulations for the 21st century (2001-2100) downscaled to a 5-minute grid over Beringia. Vegetation is simulated with BIOME4, an equilibrium vegetation model, and LPJ-GUESS, a dynamic ecosystem model. For both 11,000 years ago and 6,000 years ago, areas of Beringia are simulated to have had warmer summer temperatures and slightly more summer precipitation than present. These paleoclimate anomalies are of the same magnitude and direction as those simulated for Beringia in the future. The simulated paleovegetation data are compared with paleovegetation proxy data (e.g., pollen from lake sediments) for 11,000 and 6,000 years ago from the region.

PROJECTING CLIMATE CHANGES AND ECOLOGICAL RESPONSES

GARY D. SHARP (1), LEONID KLYASHTORIN (2), AND DOUGLAS McLAIN (1)

(1) Center for Climate/Ocean Resources Study, Salinas, CA 93907 (2) V.N.I.R.O., Moscow, Russia

gsharp@redshift.com Climate change has long been documented in many locations as being responsible for major ecological responses in regional ocean ecosystems. Unfortunately, due to reluctance by many agency policy makers to fund the necessary monitoring of both physical and ecological components, and the dominant presumption that underlying all ecology there are fundamental equilibrium processes that determine the System State – ‘surprises’ keep occurring. In recent decades there have been many gross simplifications and analyses of particularly undependable data sets, from limited time series of species landed catch data, various local temperature series, and some in situ sampling of plankton, etc., and thus the potential for any dependable projections have been minimal, and those that have been made are for the most part, quite gloomy. However, in the 1970-80 period when biological oceanography was the basis for many powerful efforts to understand the fundamental requirements for successful year class production in many ocean species, it became obvious that wind speed and direction, incident sunlight and temperature were the fundamental factors that determine which species would thrive, and by tracking these and the subsequent production patterns, useful forecasts of likely species-specific survival changes could be made. Given these insights, the long-term insights from old cultures about atmospheric cycles, and consequent species dynamics, it is now possible to focus on when and where to monitor various properties that will enhance the benefits of both system monitoring and resource management. We provide examples for future reference. 33

ELEVATIONAL AND TEMPORAL TRENDS IN SALIX-FEEDING BEETLES AND

ASSOCIATED INSECTS ALONG THREE SIERRA NEVADA, CALIFORNIA DRAINAGES

JOHN T. SMILEY (1), NATHAN E. RANK (2), ELIZABETH P. DAHLHOFF (3) (1) University of California White Mountain Research Station, 3000 E. Line Street,

Bishop, CA 93514 (2) Department of Biology, Sonoma State University, 1801 East Cotati Avenue,

Rohnert Park, CA 94928 (3) Department of Biology, Santa Clara University, 500 El Camino Real,

Santa Clara, CA 95053 jsmiley@wmrs.edu

We have been studying the ecology, evolution, and physiology of the willow leaf beetle, Chrysomela aeneicollis, in the eastern Sierra Nevada, California, since the early 1980's. By examining changes along three elevation gradients, Big Pine (BP), Bishop Creek (BC), and Rock Creek (RC) we hoped to discover asymmetrical or unidirectional shifts that might be predicted in a warming climate regime. Foliage air temperature measurements since 2000 revealed that daytime temperatures experienced by the beetles were reduced at higher elevations with a lapse rate of approximately 10 oC/1,000 m of elevation. In contrast, nighttime low temperatures were often coldest at lower elevations, with reduced (BP) or no elevation effects (BC and RC). Beginning in 1982 in BP, C. aeneicollis beetle populations expanded and contracted their elevational range, but also shifted upwards by about 300 m so that in 2006 the highest elevation populations are at timberline (3,550 m). A concomitant 300 m upward shift also occurred in populations of one of the beetles’ principal predators, the wasp Symmorphus cristatus. Beginning in 1998 we sampled C. aeneicollis beetles in BC and RC. They did not appear to shift their ranges upward until the summer of 2006, at which time beetles colonized several timberline sites. The pre-2006 data are consistent with measurements of temperature physiology which suggest that nighttime low temperatures constrain beetle distributions. However, the 2006 upward shift, along with some new physiology and genetics findings, suggest a more complex relationship between foliage air temperatures and beetle survival. The overall results suggest that beetle populations may be moving upward in elevation, and we hope to continue to characterize the underlying ecology, physiology, and genetics of this process.

CORRELATING LATE GLACIAL AND HOLOCENE MARINE AND LACUSTRINE CLIMATE RECORDS IN NORTHERN CALIFORNIA

SCOTT W. STARRATT AND JOHN A. BARRON

U.S. Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025 sstarrat@usgs.gov

Evidence of long-distance connections between the marine and terrestrial realms along the northern California margin is affected by both latitude and elevation. In this study, the late glacial and Holocene history of the California Current derived from a marine core is compared 34

with the records from two lakes. The 16,000-year-long marine record (ODP Site 1019) is based on the analysis of the diatom and pollen assemblages, alkenones, sediment geochemistry, and physical properties (density, biogenic silica, and environmental magnetism). The terrestrial record at Medicine Lake, located on the Modoc Plateau in northeastern California at an elevation of more than 2,000 meters, contains an 11,400-year-long record derived from diatom and pollen assemblages, sediment geochemistry, and physical properties. The 19,000-year-long record from Swamp Lake, located near the northern boundary of Yosemite National Park at an elevation of about 1500 meters, includes pollen (Smith and Anderson, 1992) and diatom assemblages, biogenic silica, loss-on-ignition, and magnetic susceptibility. The extent of the record in both of the lakes is limited by late Wisconsin glaciation. The marine and Swamp Lake records indicate cooler temperatures and decreased productivity during the transition from the Bølling-Allerød to the Younger Dryas, followed by warmer conditions prior to the end of the Younger Dryas. This interval was not covered at Medicine Lake. The early Holocene (11,500-8,000 cal yr B.P.) is characterized by a transition from warmer and wetter to cooler and drier conditions along the coast and in inland areas, indicating declining July insolation. This interval was characterized by warm winter sea surface temperatures and a California Current of moderate strength. Warmer winters led to earlier snowmelt and an increase in the volume of both lakes. Biogenic silica levels in Swamp Lake increase during this time as does the abundance of oligotrophic planktonic taxa. Clastic input decreased in both lakes during this period, suggesting increases vegetation in the watershed. The pollen record indicates that along the coast, the warmest temperatures occurred early in this interval, followed by cooling and increased moisture. At Medicine Lake, conditions were relatively stable whereas the warmest temperatures were reached toward the end of the early Holocene at Swamp Lake. The early part of the middle Holocene (8,000-3,000 cal yr B.P.) was characterized by weaker, less regular El Niño events, and more sustained La Niña-like conditions resulting in lower coastal sea surface temperatures and drier conditions inland. Gradual strengthening of the California Current due to a stronger thermal gradient led to increased seasonal (spring-summer) coastal upwelling and associated fog along the coast during the latter half of this interval. At Swamp Lake, biogenic silica values leveled off after reaching their highest values in the record, and pollen suggests an increase in effective moisture. However, the record is equivocal, because it suggests this increase in moisture may have been seasonal; there appears to be a hiatus of several thousand years in the shallower part of the lake. The pollen record from Medicine Lake also indicates an increase in effective moisture. The late Holocene (3,000-0 cal yr B.P.) is distinguished by increasing winter precipitation and enhanced ENSO cycles. The marine record suggests warmer winter sea surface temperatures and a stronger seasonal gradient. The pollen records at both lakes show an increase in effective moisture. However, the productivity in Medicine Lake decreases to levels recorded in the earliest Holocene, while the productivity in Swamp Lake remains at middle Holocene values. This may reflect not only the difference in latitude and elevation, but also the characteristics of the watershed. During the late Holocene, marine and lacustrine records show several extended periods of lower precipitation, alternating with wetter periods. Smith, S.J., and Anderson, R.S., 1992, Late Wisconsin paleoecologic record from Swamp Lake, Yosemite National

35

Park, California: Quaternary Research, v. 38, p. 91-102.

TOO CUNNING TO BE UNDERSTOOD: THE RECORD OF LATE HOLOCENE CENTRAL CALIFORNIA CLIMATE FROM SAN FRANCISCO

BAY MARSH SEDIMENTS

SCOTT W. STARRATT U.S. Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025

sstarrat@usgs.gov Sediment cores collected from three marshes on the northern margin of San Francisco Bay provide differing records of late Holocene climate variation in central California. The timing of changes in the diatom floras illustrates the complex interaction between local and regional climatic processes. Results from Rush Ranch and Petaluma marsh suggest that conditions along the central coast became drier prior to the Medieval Climatic Anomaly and that fresh water became increasingly abundant during the transition from the Medieval Climate Anomaly (A.D. 800-1300) to the Little Ice Age (A.D. 1300-1900). In contrast, the Benicia State Park site is dominated by a freshwater flora during the Medieval Climate Anomaly, and conditions become brackish at the beginning of the Little Ice Age. The Rush Ranch site shows periods of increased salinity between 3,000 to 2,700 cal yr B.P., 1,750 to 750 cal yr B.P., and from about A.D. 1930 to the present. The 3,700 year-long Benicia State Park record shows a rapid decrease in salinity around 3,200 cal yr B.P. Fresher conditions continue until 500 cal yr B.P., after which salinity increases. During this later period, sediment deposition occurred in the summer and fall when river flow rates were lower, resulting in a more brackish diatom flora. Differences in the timing and duration of the fresher and more saline intervals at Rush Ranch and Benicia State Park are largely controlled by proximity to the main channel of the Sacramento-San Joaquin River system. The short (about 15 km) distance from the mouth of the tidal channel to Rush Ranch appears to have a dampening effect on the signal of climate variation. For example, the transition from brackish to fresher conditions takes place between 3,200 and 3,100 cal yr B.P. at Benicia State Park, but does not occur until several hundred years later at Rush Ranch. The subsequent transition from fresher to more brackish conditions occurs 150 to 200 years earlier at Rush Ranch. The record at Petaluma marsh is strongly controlled by precipitation in the Coast Ranges. Beginning about 1,550 cal yr B.P., conditions become more saline, and, with the exception of a 200-year period from ~700 to 500 cal yr B.P., continued to increase in salinity to the present. This shift in the diatom flora may be influenced by marsh accretion, resulting in longer periods of exposure during the summer and fall.

EVALUATING ISOTOPE TRACERS OF PALEOSALINITY, FRESHWATER INPUT, AND ENVIRONMENTAL CHANGE AT

CELESTUN LAGOON, YUCATAN

JOSEPH H. STREET (1), MEGAN B. YOUNG (2), MICHELLE GOMAN (3),

36

JORGE HERRERA-SILVERIA (4) AND ADINA PAYTAN (1)

(1) Department of Geological and Environmental Sciences, Stanford University 94305-2115 (2) Stable Isotope Tracers Project, U.S. Geological Survey, Menlo Park 94025

(3) Department of Earth and Atmospheric Sciences, Cornell University, Ithaca NY 14853-1504 (4) Departmento de Recursos del Mar, CINVESTAV, Merida, Yucatan, Mexico

jstreet@stanford.edu Several studies have attempted to reconstruct Quaternary climate variability in Meso-America and the Caribbean region (e.g., Hodell and others, 1995; Hughen and others, 2004). Interest in this region has been driven in part by evidence that changes in tropical ocean conditions may drive climate on a global scale and by the realization that the region is well-placed to track past shifts in the average position of the Intertropical Convergence Zone (ITCZ), which is a major control on precipitation and drought cycles in tropical and sub-tropical areas of the world (Hoerling and others, 2001). With this in mind, we seek to develop a new record of regional paleoprecipitation and environmental change based on the Sr, O, and C isotopic compositions of benthic foraminiferal tests recovered from estuarine sediment cores collected in Celestun Lagoon, on the northwestern Yucatan Peninsula. In addition, we have measured foraminiferal assemblages and the C and N isotope composition of sedimentary organic matter from these cores. The groundwater that dominates freshwater input into Celestun is recharged over a large area of the Yucatan, suggesting that a record of paleosalinity and groundwater discharge to Celestun will integrate a regional signal. Along a present-day, head-to-mouth transect, estuarine salinity is strongly correlated to 87Sr/86Sr of both water and core-top foraminifera. The relationship between salinity and δ18O is more complex, though foraminifera appear to faithfully record δ18O water, with little temperature effect, and may provide a means of tracing the relative importance regional and local groundwater sources to the lagoon in the past. Preliminary downcore results suggest that the estuary has experienced pronounced environmental changes over the course of the late Holocene. We will compare the utility of the various tracers for estuarine paleoclimate reconstruction and evaluate our results in the broader regional context. Hodell, D.A., Curtis, J.H., and Brenner, M., 1995, Possible role of climate in the collapse of Classic Maya civilization: Nature, v. 375, p. 391-394. Hoerling, M.P., Hurrell, J.W. and Xu, T., 2001, Tropical origins for recent North Atlantic climate change: Science, v. 292, p. 90-92. Hughen, K.A., Eglinton, T.I., Xu, L., and Makou, M., 2004, Abrupt tropical vegetation response to rapid climate changes: Science, v. 304, p. 1955-1959.

OCEANS AND CLIMATE, AND THEIR EFFECT ON FOREST-INSECT INTERACTIONS

ALAN J. THOMSON

Pacific Forestry Centre, Victoria, BC, V8Z 1M5, Canada athomson@pfc.cfs.nrcan.gc.ca

37

The earliest recorded outbreaks of western spruce budworm (Choristoneura occidentalis Freeman) on its principal host, Douglas fir (Pseudotsuga menziesii (Mirb.) Franco), in British Columbia occurred on southern Vancouver Island. Outbreaks have increased in extent in British Columbia since that time, yet there have been no outbreaks on southern Vancouver Island since the 1930s. When outbreaks occur, bands of defoliation occur at different positions along topographic gradients from year to year as the insects respond to changes in weather. However, this adaptive capacity was insufficient for outbreaks to reoccur on Vancouver Island. Using a web-based program for exploring budworm-host phenology, we have recently found that over the last 90 years for which records are available, warming winter temperatures on southern Vancouver Island, linked to a rise in sea temperature, promoted earlier larval emergence while the timing of bud flush has remained constant. This differential response to changed climate has resulted in a significant departure from optimal phenological synchronization that would explain the outbreak history. Changing phenological relationships along topographic gradients due to climate change will have significant consequences for the interactions between herbivorous insects and their host plants. Studies of such relationships have generally focused on changes in growing season of the plants. However, as the mode of action of the effect in our present study was through the insect in the off-season, rather than through the plant during the growing season, will alteration of inter-species relationships by climate change through this route result in more surprises than those resulting though changes in plant growing seasons?

ABRUPT CLIMATE CHANGE AND ECOSYSTEM RESPONSE: A 2,000 YEAR HISTORY OF SEVERE DROUGHTS, EXTREME TEMPERATURES, AND STAND-

REPLACING FIRES IN THE SIERRA NEVADA IN RESPONSE TO ABRUPT CLIMATE CHANGE

STEPHEN F. WATHEN

Department of Plant Sciences, University of California, Davis, CA 95616 sfwathen@ucdavis.edu

Climate change is increasingly seen as a serious threat to modern environments and yet little is known about the environmental effects of abrupt climate change during the Holocene. I investigated the environmental effects of abrupt climate change over the past 2,000 years at Coburn Lake, California to test the hypothesis that Coburn Lake charcoal peaks represent stand-replacing fires that occurred within the Coburn Lake watershed in response to abrupt climate change. I compared the timing of Coburn Lake charcoal peaks with both the timing of the onset of droughts and the presence of temperature extremes in the Sierra Nevada. Charcoal peaks, with associated proxies for erosion, occurred at the beginning of all ten droughts, and only then, in the northern Sierra Nevada over the last 1,900 years. The environmental stress caused by the onset of severe droughts was made all the more severe due to severe cold temperatures, or abrupt shifts in temperatures, at the beginning of most droughts in the Sierra Nevada over the last 1,900 years. These results suggest an “abrupt climate change-severe fire hypothesis,” that implies that abrupt climate change during the late Holocene caused vegetation and mountain slopes around Coburn Lake, and perhaps more widely, to be seriously out of balance with changing climates; resulting in forest die-off, stand-replacing fires, and severe soil erosion. Taken together, this 38

study provides a rare detailed history of abrupt climate and environmental change in the Sierra Nevada over the past 2,000 years.

CLIMATE CHANGE AND WILDFIRE

ANTHONY L. WESTERLING (1,2), HUGO HIDALGO (2), DANIEL R. CAYAN (2,3) AND THOMAS W. SWETNAM (4)

(1) Sierra Nevada Research Institute, University of California, Merced, CA 95344 (2) Climate Applications Program, Scripps Institution of Oceanography, La Jolla, CA 92093

(3) U.S. Geological Survey, La Jolla, CA, 92093 (4) Laboratory for Tree-Ring Research, University of Arizona, Tucson, AZ 85721-0058

awesterling@ucmerced.edu Western U.S. forest wildfire regimes shifted abruptly in the mid-1980s, with larger, more frequent, longer-burning fires occurring in many fire seasons since then. The greatest increases occurred in mid-elevation forests of the northern Rocky Mountains. Other forests of the west have seen increased wildfire, but the change has varied greatly with location. Wildfire regimes in other, non-forested ecosystems of the western U.S. have not changed dramatically in recent decades. The increase in forest wildfire is strongly associated with warming and earlier springs. Why have some ecosystems responded so dramatically to these changes in climate, while others have not? Climatic impacts on wildfire regimes are the result of complex, nonlinear responses of ecosystems and wildfire to changes in climate and other factors. The same climatic influences that limit the kinds of vegetation that can grow in a specific location also affect fire regime-ecosystem interactions and their responses to changes in climate. We use documentary wildfire histories and climate records, models and proxies from the western U.S. to demonstrate some important climate-vegetation-wildfire interactions, from which we infer causes of the observed variability in wildfire regime responses to recent climate change, and extrapolate impacts of future climate changes.

TREE RINGS INDICATE SUMMERTIME STRATUS CLOUDS AS IMPORTANT SOURCES OF PRE-DAWN WATER AND AFTERNOON SHADE FOR COASTAL

VEGETATION IN CALIFORNIA

A. PARK WILLIAMS (1), CHRISTOPHER J. STILL (1), DOUGLAS T. FISCHER (1) 39

AND STEVEN W. LEAVITT (2) (1) Department of Geography, University of California, Santa Barbara, CA 93101

(2) Laboratory of Tree-Ring Research, University of Arizona, Tucson, AZ 85721-0058 williams@geog.ucsb.edu

The coast of California is home to many rare, endemic conifers and other plants that are drought sensitive and generally not well adapted to the summer-dry Mediterranean climate that prevails across most of the state. Ecologists have long assumed that coastal pines survived the early-Pleistocene transition to a warmer and drier environment because they benefit from frequent fog and low stratus clouds that provide water and shade during the rainless summer. Here, we report evidence for the importance of summer cloud cover to a stand of Torrey pines (Pinus torreyana) on Santa Rosa Island in Channel Islands National Park. We developed a tree-ring width chronology (1920-2004) and quantified the effects of summer cloud cover by comparing ring widths to airport cloud-frequency records. Summertime cloud frequency correlated significantly and positively with ring widths after we removed the effect of wet-season precipitation. Correlation with cloud frequency at altitudes low enough for fog to inundate the entire pine stand was strongest in July early mornings (03:00 PST, R2 = 0.262). Correlation for clouds high enough to provide shade but not water was strongest in July late afternoons (16:00 – 18:00 PST, R2 = 0.148). These relationships were substantially stronger in years when the early part of the rainy season was relatively dry and/or the bulk of the rainy season was relatively wet, suggesting that these pines do not grow during every summer. This may be due to an adaptive growing season length with a termination date dictated by water availability, and possibly an upper limit on growing-season length. A decrease in fog activity along the California coastline, which is apparent in various local climate records, would likely impact relict stands of species that depend, in part, on frequent and predictable summertime stratus-cloud cover. Unfinished research focuses on comparing modeled vs. actual stable isotope (carbon and oxygen) ratios in Bishop pine tree rings from Santa Cruz Island to further evaluate the relationship between growth and cloud cover. We are also compiling a daily record (2000-2006) of spatially continuous cloud cover, as observed via satellite (MODIS). We will compare the spatial distribution of cloud-cover frequency to the spatial distributions of various endemic plant species on the U.S. west coast to further evaluate where and for which species regular cloudiness is crucial. We will also compare this spatial record of coastal cloud cover to other daily climate records such as spatial distributions of atmospheric pressure, sea surface temperature, and wind direction to improve our understanding of what causes low stratus and fog to develop in specific locations.

MAGNITUDE OF OCEAN WARMING DURING THE PALEOCENE-EOCENE THERMAL MAXIMUM

JAMES C. ZACHOS

Department of Earth and Planetary Science, University of California, Santa Cruz, CA 95064 jzachos@pmc.ucsc.edu

The Paleocene-Eocene Thermal Maximum (PETM) provides a unique opportunity to 40

observe the global climate sensitivity to a rapid and large increase in greenhouse gas levels. Although the approximate changes in atmospheric CO2 levels have yet to be determined, carbon isotope and deep-sea carbonate accumulation records confirm the rapid release of several thousand gigatons of carbon (<10,000 years). Such a massive release over such short time scales should have raised pCO2 by hundreds of ppm. In response, sea surface temperatures should have risen uniformily over most of the planet. Such a response is supported by more than a dozen SST anomaly records from pelagic and shallow marine sections which show a 5 to 9 �C increase. Most of these SST anomaly records are based on oxygen isotopes analyses of planktonic foraminifera, and a few are based on Mg/Ca and TEX86. In locations where multiple proxies are available, there is generally good agreement on the magnitude of SST change. The few exceptions are primarily in tropical pelagic cores where it appears the oxygen isotope based anomalies are damped. Reliable reconstructions of absolute SST, on the other hand, are far more limited. Nevertheless, the few available points consistently show 5-8 �C of warming with peak PETM temperatures well above modern, as high as 33 �C in the mid-latitudes and 22 �C in the Arctic. Along the Pacific Coast of North America, new data from the Toomey Hills in central California indicate that SST over the shelf peaked at 27 �C. The pattern of global warming is consistent with greenhouse forcing in the absence of ice-albedo feedbacks. However, the peak temperatures appear to be higher than can be achieved with just CO2 forcing alone (using current climate models) implying that other trace gasses may have contributed as well.

MONO BASIN CLIMATE CHANGES CORRELATIVE WITH NORTH ATLANTIC DANSGAARD-OESCHGER OSCILLATIONS

SUSAN ZIMMERMAN (1, 2), CORINNE HARTIN (3,4), CRYSTAL PEARL (3),

STEPHANIE SEARLE (1), SIDNEY HEMMING (1), AND N. GARY HEMMING (1,3) (1) Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY 10964

(2) now at Lawrence Livermore National Laboratory, Livermore, CA 94550 (3) Queens College, City University of New York, New York, NY 11367

(4) now at Rosentiel School of Marine and Atmospheric Sciences, University of Miami, Miami, FL 33149

herrzim@gmail.com Lakes are highly sensitive recorders of climate processes, but can be extremely difficult to correlate precisely to ice core and marine records, especially beyond the range of radiocarbon dating. Mono Lake, California (38�N, 119�W) preserves high-elevation strandlines and multiple exposures of deep-lake silts of the late Pleistocene Wilson Creek Formation, indicating that the lake level was high and variable in the past. Revision of the Wilson Creek Formation age model, based on relative paleointensity (RPI) of Earth’s magnetic field, allows independent correlation of paleoenvironmental changes in the Mono Basin to global climate records at millennial timescales. Comparison of relative lake level changes, based on mapping of facies in the Wilson Creek Formation, with high-resolution bulk-sediment carbonate records reveals correlation of high carbonate concentration with high lake level. Application of the new age model to the carbonate record shows that during times of high local spring insolation the Mono Basin was generally wetter, with high-amplitude, short-duration fluctuations in lake level correlative to the 41

42

Dansgaard-Oeschger (D-O) events of Greenland. Such regional expressions of the D-O events have now been found globally, which the currently-favored hypothesis suggests may be related to changes in the mean position of the Intertropical Convergence Zone.