Global Environmental Change [GC]

GC32A  MW:3002   Wednesday
Climate Change in High-Elevation Mountain Environments II
Presiding: C Woodhouse, University of Arizona; C Millar, USDA Forest Service; H Diaz, NOAA Earth System Research Laboratory

GC32A-01 

Climate response in the western United States to dust-shortened snow cover duration since late 1800s soil disturbance

* Painter, T H (painter@geog.utah.edu), Department of Geography, University of Utah, Salt Lake City, UT 84112, United States Mahowald, N (nmm63@cornell.edu), Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, NY 14853, United States Neff, J C (Jason.C.Neff@Colorado.EDU), Department of Geological Sciences, University of Colorado, Boulder, CO 80309, United States

Dust emitted from the deserts of the western US is currently shortening snow cover duration in the Rocky Mountains by 20-35 days and affecting a Spring radiative forcing of 25 - 50 W m-2 through snow albedo reduction. With the injection of railroads and infrastructure into the interior of the western US in the mid 1800s came substantial increases grazing, farming, and mining, and in turn a disturbance of 70 percent of natural ecosystems resulting in losses of soil stability and increased dust emission. Recent geochemical analysis of alpine lake sediments in southwest Colorado shows that since the 1800s settlement of the western US, dust loading to mountain environments abruptly increased by a factor of ~5 above average rates of the last 5000 years. Given present dust loading to the Rocky Mountains and its associated forcing of snowmelt and reduction of snow cover duration, it is evident that snow cover endured significantly longer prior to the mid 1800s disturbance. In this work, we assess the climate response in the western US to the perturbed duration of mountain snow cover affected by the dramatic increase in dust loading using the NCAR Community Atmospheric Model (CAM). We drive the CAM with snow cover duration scenarios consistent with the snow albedo reductions associated with pre-1850 and post-1850 dust loadings. The 2nd indirect effect of dust loading to snow is given by the enhanced absorption of shortwave radiation by a darker substrate emerging from the early ablation of snow cover. In this region, 30 days of reduced snow cover results in mean daily 2nd indirect effect of ~ 150 W m- 2. Therefore, we hypothesize that reduced snow cover duration results in significant regional tropospheric warming, a perturbed monsoon in the southwest US, and reduced cloud cover. In turn, the historical record of hydrologic observations in these regions has likely been entirely disturbed by anthropogenic forcing.

GC32A-02 

Changes in the timing of snowmelt and associated runoff in the Colorado Rocky Mountains

* Clow, D W (dwclow@usgs.gov), U.S. Geological Survey, MS 415 Federal Center Box 25046, Denver, CO 80225, United States

Previous studies have documented changes in the timing of springtime runoff in the Western United States, with runoff occurring up to two weeks earlier in 1950 than in the 2000. Changes were most pronounced in the Sierra Nevada, Cascades, and northern Rocky Mountains. Few statistically significant changes were identified in Colorado, despite local observations of earlier than normal snowmelt during many recent spring seasons. To further elucidate recent trends in snowmelt timing and associated runoff in Colorado, data from 72 SNOTEL sites and 40 high-elevation streamflow-gaging stations with minimal upstream diversions were tested for trends during 1978 to 2004. Trends were tested using linear regression, as was done in the previous studies, and using the Regional Kendall test (RKT), which is a relatively new test derived from the Seasonal Kendall test. The RKT provides increased power of trend detection in short records with substantial interannual variability. Results indicated that although few sites exhibited statistically-significant trends using linear regression, the RKT identified pervasive earlier snowmelt and runoff throughout the State, with an average change of 0.5 days per year. The RKT revealed important regional variations in the snowmelt- and runoff-timing trends. The strongest trends were in the western and southern parts of Colorado; trends in the north-central part of the State were relatively weak, perhaps because of an increase in upslope storms. Changes in snowmelt timing were strongly correlated with increasing springtime air temperatures, which showed strong positive (warmer) trends during the study period. In contrast with previous studies, this study identified significant shifts in the timing of snowmelt and associated runoff towards earlier in the year in Colorado, and the shift is related to springtime warming.

GC32A-03 

Toward Standardization in Methods and Techniques for Measuring and Monitoring Snowcover Albedo.

* Landry, C C (clandry@snowstudies.org), Center for Snow and Avalanche Studies, PO Box 190 1315 Snowden - Rm 8, Silverton, CO 81433, United States Painter, T H (painter@geog.utah.edu), University of Utah, Department of Geography 260 S. Central Campus Dr. Rm. 270, Salt Lake City, UT 84112-9155, United States Barrett, A P (apbarret@kryos.colorado.edu), National Snow and Ice Data Center, University of Colorado CIRES - 449 UCB, Boulder, CO 80309-0449, United States Cassidy, M (Maureen.Cassidy@colorado.edu), University of Colorado, Department of Geography Guggenheim 110 - 260 UCB, Boulder, CO 80309-0260, United States

Global climate change portends increasing uncertainty regarding the reliability of mountain snow and ice fields as a source of fresh water for one-sixth of the world's population. Standardization of system measurements is required to enhance our understanding of cryospheric responses and forcings. Irrespective of projected temperature trends, we have shown that interactions between deserts and down-wind mountain ranges can and do result in significant advancements of snowmelt timing as well as increased snowmelt intensity, substantially altering regional hydrographs. We have developed and refined methods for monitoring enhanced radiative forcing of snowmelt caused by dust induced reductions in snowcover albedo. In- situ, continuous measurements of snowcover albedo and energy budget parameters are obtained in the Senator Beck Basin Study Area with two arrays of up- and down-looking pyranometers, pyrgeometers, and infrared snow surface temperature sensors. Air temperature, relative humidity, and wind speed are monitored at two heights above the snowcover. Measurements of short wave radiation reflected by the snowpack are corrected for surface geometry by monitoring an array of snow stakes referenced to a level plane. The efficiency of enhanced energy absorption by exposed and near-surface dust layers is monitored using a volumetric sampling design whereby ten snow samples are collected to a depth of 30 cm, near the limit of significant light penetration. Those samples are then processed to quantify the mass of absorbing material per unit of area at a given depth, enabling the estimation of enhanced absorption throughout the near-surface and surface of the snowcover. These methods, in conjunction with traditional snowpack profiling techniques, have proven to be a reliable, practical, and repeatable approach to monitoring the influence of desert dust on mountain hydrology. The enhanced rigor with which these measurements are performed presents a platform upon which to build consensus protocols for adoption in other system monitoring applications and locales.

GC32A-04 

Runoff Efficiency of Sierra Snowmelt: Evaporative Water Losses in Wet vs. Dry Years

* Lundquist, J (jdlund@u.washington.edu), University of Washington, Civil and Env. Engineering Box 352700, Seattle, WA 98195, United States

High altitude Sierra basins have negligible summer precipitation and very little groundwater storage, which makes them ideal laboratories for indirectly monitoring changes in evaporative losses between wet and dry years. Dry years have greater potential evapotranspiration, due to warmer June and July air temperatures, warmer summer water temperatures, greater solar radiation exposure, and longer growing seasons. However, dry years also have limited saturated surface areas as compared to wetter years, and thus actual evapotranspiration is much less than the potential in dry years. Assessing the balance of these factors is important in estimating the effect of warming temperatures and shrinking snowpacks on Sierra ecosystems. When spring and summer rain events are excluded from the analysis, the annual sum of basin snowmelt (calculated from 119 CA DWR snow pillows) minus the sum of March to October streamflow (calculated from USGS records at 10 high-elevation California river basins) indicates water losses from the basin. Assuming negligible groundwater storage from one year to the next, these water losses are a measure of evaporation and evapotranspiration (ET). Records from 1968 to 2005 show that the least amount of water is lost to ET in the wettest years, but the story for dry to normal years is more complicated. Conceptual models are used to test the sensitivity of annual ET to snow cover extent, length of summer season, moisture availability, basin elevation distribution, and air and water temperatures. Results are compared with observations. http://faculty.washington.edu/jdlund/home/

GC32A-05 

Secondary Effects of Climate Change on Streamflow Through Wildfire: Compensating or Exacerbating?

* Luce, C (cluce@fs.fed.us), USDA Forest Service Rocky Mountain Research Station, 322 E Front St, Boise, ID 83702, United States

Changes to snow accumulation in mountains have yielded two noteworthy phenomena: A shift in the timing of snowmelt and streamflow and increased occurrence and severity of wildfires. Wildfire, in turn, produces feedbacks on the snow accumulation and melt process that further alters streamflow generation processes and future prospects for vegetation. An important question is whether increased wildfire will exacerbate or ameliorate climate related changes. Vegetation reduction experiments in small watersheds (O~1-5 km2) suggest we should expect greater snow accumulation and reduced evapotranspiration following wildfire, however they also show slight advances in the timing of snowmelt peaks. Recent large scale wildfires are providing an opportunity to see how the secondary effects of climate change through wildfire play out at scales comparable to those where direct changes have been assessed. In the Boise River basin, two mountain watersheds on the order of 2,000 km2 each form a paired watershed experiment with approximately 45% of one basin burned and nearly no fire in the other. With sixty years of calibration, and twelve years post fire, significant increases in annual water yield on the order of 5% were found. Increases primarily occurred in winter and early spring, however, exacerbating the shift in flow timing for this high elevation basin. The tradeoff in timing and yield produced dramatically lower flows in early summer, but a slight increase in late summer, compensating for some climate change effects. Ecohydrologic theory predicts that vegetation changes accompanying increased evaporative demands should partition the losses between vegetation consumption and streamflow. The longer summer produced by these combined climate and land cover changes may yield less consumptive vegetation for the next generation. http://www.fs.fed.us/rm/boise/research/watershed/bio/luce.shtml

GC32A-06 

A century of glacier change in the American West

* Fountain, A G (andrew@pdx.edu AF:

Over the past 100 years glaciers in the American West (exclusive of Alaska) have largely receded. The magnitude of the recession varies across the west, with the greatest loss in Montana and California (>50% area loss) and the least loss on the stratovolcanoes (>35%) of the Pacific Northwest. The variations can be broadly characterized by elevation. Our results suggest that increased mass loss caused by increased summer temperatures affect all glaciers, whereas increasing winter temperatures, that change the phase of precipitation from snow to rain adversely affect those glaciers less than 3000m in elevation. The high glaciers (>3000m) of California and Colorado appear to be immune to variations in snowfall making them sensitive to variations in temperature alone. We infer that these very small, steep glaciers can only hold a given amount of snow beyond which extra snow avalanches or is blown off. Conversely, during winters of little direct snowfall, additional snow may be added through win drift from the surrounding terrain. The relatively little glacier shrinkage on the stratovolcanoes is due to the high altitude of the glacier accumulation zones. An east to west decrease in glacier shrinkage from Montana through Washington is due to enhanced winter precipitation along the west coast that somewhat buffers ice loss due to summer temperatures and winter precipitation phase changes.

GC32A-07 

Response of Rwenzori (Uganda - DR Congo) Glaciers and Mountain Lake Ecosystems to Climate Change: Past, Present, Future

* Eggermont, H (Hilde.Eggermont@ugent.be), Limnology Unit, Department of Biology, Ghent University, K.L. Ledegancstraat 35, Ghent, 9000, Belgium Russell, J (James_Russell@brown.edu), Geological Sciences, Brown University, Box 1846, Providence, RI, 0912, United States Verschuren, D (Dirk.Verschuren@ugent.be), Limnology Unit, Department of Biology, Ghent University, K.L. Ledegancstraat 35, Ghent, 9000, Belgium

Ice caps and glaciers in the mountains of tropical Africa are expected to disappear within two decades as a result of human-induced global warming. Loss of permanent ice from Africa's mountains will have profound effects on Afroalpine ecosystems, as well as the hydrology and temperature regime of Africa's unique, alpine cold-water lakes. Consequently, there is an urgent need to document the baseline climatic, environmental, and biological conditions in Africa's mountains against which to evaluate future changes. While climatic, limnological, and ecological monitoring can provide key insights into modern conditions, sediments accumulating on the bottom of alpine glacial lakes chronicle the history of central African climate and environmental dynamics, and can thus produce the historical perspective needed for resource conservation. In this context, over the course of three field expeditions (2005-2007) we surveyed virtually all lakes (19) on the Ugandan side of the Rwenzori Mountains, Uganda-Congo, to evaluate the sensitivity of these high-mountain lake ecosystems to climate change and glacier melting. We installed temperature and humidity loggers to monitor ongoing climate change, calibrated paleolimnological proxy indicators (sediment composition, organic carbon, biomarkers, aquatic biota) in surface sediments in relation to their present local lake environment, and analyzed gravity cores from selected lakes to document the history of recent glacier recession and ecosystem response. Here, we present an overview of the currently available climatic and environmental data sets, with a focus on sediment core data. Through comparison of the sedimentary history of glaciated and non-glaciated lake basins, we show that recent glacier recession started around 1880 AD, broadly coincident in timing with declining East African rainfall as documented by regional lake level datasets. However, our data do not suggest rapid glacial expansion coincident with the initiation of this wet phase in the early 19th century, highlighting the complexity of the relationship between tropical alpine glaciers and climate. Glacier recession starting at 1880 AD is accompanied by declining aquatic primary productivity, recorded in the isotopic composition of sedimentary organic matter. Together, our data show that Rwenzori's high-elevation lakes are responsive to alpine glaciation and constitute a unique laboratory to assess quantitatively the relationship between the extent of its glaciers, changes in central African climate, and ecosystem processes.

GC32A-08 

Hydrological implications of glacial recession in the Rwenzori Mountains of Uganda

* Taylor, R (r.taylor@geog.ucl.ac.uk), University College London, Department of Geography Gower Street, London, WC1E 6BT, United Kingdom Mileham, L (ucfaljm@ucl.ac.uk), University College London, Department of Geography Gower Street, London, WC1E 6BT, United Kingdom Tindimugaya, C (ctindi.wrmd@dwd.co.ug), Ministry Water & Environment, Water Res. Management Dir., Kampala, Box 20026, Uganda Leng, M (mjl@nigl.nerc.ac.uk), British Geological Survey, NERC Isotope Geosciences Lab. Keyworth, Nottingham, NG12 5GG, United Kingdom

The areal extent of tropical icefields in the Rwenzori Mountains of East Africa has reduced steadily over the last century from 7.5 km2 in 1906 to <1 km2 in 2003. Considerable debate persists, however, regarding the impact of observed deglaciation on alpine riverflow and the precise climate signals indicated by glacial recession in the Rwenzori Mountains and other areas of the East African Highlands including Kilimanjaro and Mount Kenya. We present the first measurements of alpine riverflow in the Rwenzori Mountains to show that the contribution of meltwater flows from dwindling icefields to local freshwater resources is negligible, <1% of the mean annual river discharge recorded at the base of the mountains. Stable isotope ratios (18O:16O, 2H:1H) in streamflow trace the contemporary origin of precipitation in the Rwenzori Mountains to the same source (Indian Ocean) as other alpine icefields in East Africa. Hydrometeorological trends, indicated by observational datasets (precipitation, lake levels) over the 20th century and palaeoenvironmental and anecdotal evidence over the 19th century, are inconsistent with an abrupt reduction in humidity that has been proposed to account for observed deglaciation in the East African Highlands. http://www.geog.ucl.ac.uk/~rtaylor/rwenzori.htm