C33A-01
Analysis of the Sierra Nevada Snowpack in the 21st Century
Models of California's future climate postulate greater precipitation variability, warmer temperatures, less snow, earlier runoff, and greater likelihood of droughts and floods; statistical analyses covering the last half-century show that some trends have already started. Accompanying these changes will be greater demand from downstream economic development and population growth. Today's measurements at snow courses and pillows support empirical methods of analyzing the snowpack and forecasting runoff, but they do not cover the highest elevations well and they do not represent snow's topographic distribution. These historical data document that climate is changing, but the changes mean that the empirical relations are becoming less reliable. Blending strategically placed ground measurements with broad-coverage satellite and aircraft data offers the opportunity for continual, accurate estimates of snow and other hydrologic variables. Augmented ground observations include small transmitting sensors along elevation gradients, enhanced sites with energy-balance variables, and improved communications and cyberinfrastructure for making remote observations available in real-time. Some satellite and aircraft observations can be used immediately, while others require further research and implementation. Snow-covered area and albedo products are available now from both research and operational satellites, and these data can be used to forecast the rate of snowmelt when combined with energy- balance measurements, ground-based snow water equivalent measurements, and modeling. Airborne lidar surveys can measure snow depth in small watersheds and along transects, and can be scaled to larger basins. Satellite measurements of time-variable gravity can monitor seasonal changes in total water storage over large regions such as the Sacramento-San Joaquin valley. A decade away, a future hybrid active-passive microwave system could measure the spatial distribution of snow water equivalent. The final critical component of a distributed measurement system is the cyberinfrastructure that links measurements, data processing, models, and users.
C33A-02
Implications of climate change on Waters in the Pacific Northwest: a stochastic approach
We evaluate the impact of climate change on the water balance in the Pacific Northwest. A stochastic weather generator is used to generate meteorological data with precipitation and air temperature both modeled for wet and dry periods. The meteorological forcing is used to estimate snow state and streamflow through a simplified hydrologic model developed at Pacific Northwest National Laboratory (PNNL). The model itself is calibrated on the historical data before running for the synthetically generated meteorological data. A number of climate scenarios based on precipitation and temperature are designed and are used to study the impact of those scenarios on the streamflow and snow pack volumes. The findings can have direct relevance for the operation and management of water resources in the Pacific Northwest.
C33A-03
Snowmelt, Summer Drought and Fires: Using Stream Gage Data to Extend Snowmelt Records in Fire-prone Areas of Central Idaho
In Idaho, 2007 and 2006 were the warmest and third warmest (respectively) summers recorded since 1904. In 2007, high summer temperatures combined with severe drought conditions have led to one of the most severe fire seasons since the Yellowstone fires of 1988, and by September 1, severe fires continue to burn > 2 million acres. In snowmelt-dominated watersheds of Idaho, earlier snowmelt, the earlier onset of spring and subsequent dry fuels promote severe fire seasons. Data on the timing of snowmelt from National Resources Conservation Service (NRCS) Snow Telemetry (SNOTEL) sites, however, extend back <30 years, and placing recent droughts and fires within a historic context requires examination of a longer record. In order to examine relationships between timing of final snowmelt and summer drought and fire conditions since the early 1900s, we have developed techniques to infer the timing of final snowmelt from stream flow records. Long-term stream gage stations in Idaho were selected from the Hydro-Climatic Data Network (HCDN) and matched to nearby SNOTEL stations. We selected 15 HCDN gage sites within Idaho that had an average of 75 years of gage data, and were located in close proximity to the drainages of SNOTEL sites with at least 10 years of continuous snow water equivalent data through 2006. We used only gage data from the years that were complete 97% of the time or greater, >98% of all yearly data was usable. Missing data within years that met the 97% or greater level were estimated using linearly extended values between known values. Less than 0.05% of the data required estimation. Using a Short Term Fourier Transform (STFT) we computed the historic final snowmelt dates and compared the results with actual snowmelt dates from SNOTEL data. Computed vs. actual snowmelt dates were within +/- 4 days ~93% of the time and ~98.5% of the time within +/- 6 days for all years tested for the 15 selected gauge sites. This provided us with over 1100 yearly final snowmelt dates extended throughout Idaho. The extended final snowmelt dates where compared to nearby NRCS Snow Course data, extended dates corresponded well with the existing snow course values for all 15 sites. Ongoing research compares the extended snowmelt data with historical fire seasons while focusing on the development of drying curves to be used with current final snowmelt dates as a predictor to the onset of fire conditions in Idaho. Future work will extend additional final snowmelt records in Idaho and other western US states and will examine relationships between final snowmelt dates and the onset of fire conditions in the western US.
C33A-04
Detection of Severe Rain on Snow events using passive microwave remote sensing
Severe wintertime rain-on-snow (ROS) events create a strong ice layer or layers in the snow on arctic tundra that act as a barrier to ungulate grazing. These events are linked with large-scale ungulate herd declines via starvation and reduced calf production rate when the animals are unable to penetrate through the resulting ice layer. ROS events also produce considerable perturbation in the mean wintertime soil temperature beneath the snow pack. ROS is a sporadic but well-known and significant phenomenon that is currently very poorly documented. Characterization of the distribution and occurrence of severe rain-on-snow events is based only on anecdotal evidence, indirect observations of carcasses found adjacent to iced snow packs, and irregular detection by a sparse observational weather network. We have analyzed in detail a particular well-identified ROS event that took place on Banks Island in early October 2003 that resulted in the death of 20,000 musk oxen. We make use of multifrequency passive microwave imagery from the special sensing microwave imager satellite sensor suite (SSM/I) in conjunction with a strong-fluctuation-theory (SFT) emissivity model. We show that a combination of time series analysis and cluster analysis based on microwave spectral gradients and polarization ratios provides a means to detect the stages of the ROS event resulting from the modification of the vertical structure of the snow pack, specifically wetting the snow, the accumulation of liquid water at the base of the snow during the rain event, and the subsequent modification of the snowpack after refreezing. SFT model analysis provides quantitative confirmation of our interpretation of the evolution of the microwave properties of the snowpack as a result of the ROS event. In particular, in addition to the grain coarsening due to destructive metamorphism, we detect the presence of the internal water and ice layers, directly identifying the physical properties producing the hazardous conditions. This analysis offers the potential to characterize both the frequency and global distribution of ROS using satellite passive microwave imagery.
C33A-05
Seasonal Snow Cover in Tundra Environments:Patterns, Processes and Scaling.
Landscape-scale patterning of depth and duration of snow cover is the single most important meso-scale variable controlling biological systems in the Arctic, and is the primary determinant of landscape-scale ecosystem heterogeneity. Duration of snow-lie also determines length of the plant growing season and is a primary determinant of the dynamics of soil temperature, soil moisture, depth of freezing and heat flux. Duration and extent of these seasonal effects have major impacts upon both plant communities and the soils with which they are associated, generating landscape mosaics ranging from exposed snow-free ridges to depressions characterised by deep snow accumulation. Our primary objective has been, through a suite of measurements, models, and appropriate up-scaling methodologies, to provide the understanding of biogeochemical processes in tundra regions required to answer a number of basic questions about the impacts of global warming upon tundra ecosystems and the potential magnitudes of any positive feedbacks. Given that in the mosaic that forms the Tundra landscape depth and duration of snow cover are primary determinants of thermal and hydrological regimes, and hence indirectly of soil biogeochemical processes plus vegetation composition and cover, we hypothesised: (i) That a change in depth/duration of snow-lie will, as a result of direct effects upon soil thermal and hydrological regimes, alter the thermal balance of the soil and hence result indirectly in altered seasonal dynamics of soil biogeochemistry. (ii) That changes in seasonal soil biogeochemical dynamics and vegetation growth and development will, together with direct effects of changes in depth/duration of snow-lie, result in altered seasonal and annual ecosystem carbon and energy budgets. (iii) That the sign and magnitude of alterations in soil biogeochemical dynamics and trace-gas fluxes, resulting from climate change, will differ between components of the landscape mosaic that occupy contrasting hydro- topographic positions. (iv) That the relative contributions to trace-gas flux of individual elements of the landscape mosaic, and the differential impacts of altered climate upon them, can be integrated and up-scaled in a manner that will provide significant improvements over current spatially-non-explicit models. Here we present the key results of this 4 year programme of research related to the hypotheses posed above, carried out at Abisko, northern Sweden. In particular we will concentrate upon a series of manipulations of snow- lie across different facets of the tundra landscape upon plants (e.g. phenology, degree of herbivory, litter decomposition) and soils (soil moisture, freezing, nutrient mineralisation rates) and together, integrated impacts upon carbon exchange between landscape and atmosphere. The work will placed in the context of likely consequences of climatic change across this tundra landscape, in terms of a potentially altered snow-lie.
C33A-06
Black Carbon in Arctic Snow: Preliminary Results from Recent Field Measurements
Annual snowpacks act to amplify variations in regional solar heating of the surface due to positive feedback processes associated with areal melting and precipitation. Small amounts of black carbon (BC) in the snow can reduce the albedo and modulate shortwave absorption and transmission affecting the onset of melt and heating of the snow pack. The effect of black carbon on the albedo of snow in the Arctic is estimated to be up to a few percent. The only prior survey of arctic snow was that of Clarke and Noone in 1983-84. We have begun a wide- area survey of the BC content of arctic snow in order to update and expand the 1983/84 survey. Samples of snow have been collected in mid to late spring when the entire winter snowpack was accessible. The samples have been melted and filtered, and the filters analyzed for absorptive impurities. To date, sites in Alaska, Canada, Greenland, and in the Arctic Basin have been sampled. In March and April 2007 we also carried out a field program at four sites in northwestern Russia as part of the International Polar Year. Preliminary results based on visual comparison with the standard filters indicate that the snow cover in arctic North America and the Beaufort Sea have lower BC concentrations now than 20 years ago while levels in Greenland are about the same. Background levels of BC in Russia are approximately twice those in North America consistent with modeling predictions of Flanner et al., 2007. More accurate values of absorption will be obtained by measurement of spectral transmission of the filters, which will also allow the relative contributions of BC and soil dust to be determined. http://www.atmos.washington.edu/sootinsnow
C33A-07
Black Carbon Mass Concentration in California Mountain Snow
Recent modeling studies have shown that deposition of black carbon (BC) to snow and ice lowers the albedo of snow and exerts a positive forcing on the climate. This effect is also a likely contributor to the observed ice and snow retreat in glaciers, ice sheets, and mountain snow pack. Observational data of actual BC concentrations in snow, which would help to constrain and validate these results, are scarce. This study presents the concentration of BC in fresh snow measured at two mountain locations in Northern California (Lassen Volcano Natl. Park and Donner Summit), as well as that in coastal rainfall at Trinidad Head, CA. These measurements are the first of this kind made in California. Average BC concentration at Lassen Natl. Park and Donner summit were respectively 6.8 and 9.7 ng per gram of snow. When placed in context with the modeled effect of BC on fresh snow albedo, these concentrations indicate a lowering of the snow albedo by 0.5 to 0.8%. As the snow pack ages, models predict that this effect will be increasingly amplified. Measurements of ambient aerosols during the rain events, as well as HySPLIT back-trajectories, provide additional information regarding possible sources of BC in California mountain snow packs. For the samples collected in this study, most of the soot in the snow appears to be of local origin. The average BC concentration in the coastal rain was 6.0 ng per gram of water, with the highest concentrations (12 ng/g) corresponding to the only back-trajectories clearly indicative of long-range trans-Pacific transport.
C33A-08
Snow-mediated Ptarmigan Browsing Controls Shrub Expansion in Arctic Alaska
Large, late-winter ptarmigan migrations heavily impact the patch, shrub, and branch morphology of shrubs that remain above the snow surface. Ptarmigan browsing on arctic shrubs was assessed in the vicinity of Toolik Lake, on the Arctic Slope of Alaska, by counting the fraction of browsed versus total bud locations on Salix alaxensis, Salix pulchra, Salix lanata, and Betula nana. Data were collected in early May 2007, at maximum snow depth, after the bulk of the ptarmigan migration had passed through the area. In an area of large shrubs, half of the buds were browsed by ptarmigan. Three percent of the buds that were buried beneath the snow were browsed, 97 percent of the buds that were within 25 cm of the maximum snow level were browsed, and 46 percent of the buds above that height were browsed. In other areas, where all the shrubs were buried by snow at the height of the migration, 35 percent of the buds were browsed. These data were qualitatively extrapolated by observation westward to the Nanushuk River and eastward to the Jago River, about 250 km across a series of north-flowing rivers with headwaters in the Brooks Range. Ptarmigan hedging shrub patches, and shrub expansion under a warmer climate, are competing forces mediated by snow distribution. The outcome of these competing forces varies spatially, and can be inferred by examining the morphology of shrub patches, individuals, and branches.