C34A-01
Changing Snowcover, Temperature, and Streamflow Patterns in a Pacific Northwest Watershed
The Oregon Cascade range receives over 2500 mm of precipitation annually making it one of the wettest places in the US, despite a pattern of seasonal drought and summer low flows. Because snow in much of the Cascades accumulates close to the melting point, future warming would mean that large areas could shift from a snow- dominated to a rain-dominated winter precipitation regime (Nolin and Daly, 2006), with the potential for higher winter peak flows and lower summer low flows. In this work, we examine long-term records of snow depth, surface air temperature and streamflow at the H. J. Andrews Long Term Ecological Research (LTER) site. Situated in Lookout Creek watershed on the western side of the Oregon Cascades, this LTER is characterized by rugged topography. It spans a precipitation gradient from the snow-dominant zone, the rain-snow transition zone, to the rain-dominant zone. Analyses of snow depth, temperature, and streamflow show that the expected patterns of climate change are not straightforward. Other factors such as temperature inversions (controlled by atmospheric circulation) and geology affect snowmelt and streamflow dynamics.
C34A-02
Sensitivity of the seasonal snowcover to warming climate trends in a snow dominated semi- arid mountain basin
Temperature of the western United States has increased by 1-2 C since 1950's. Over the last 45 years of record at the Reynolds Creek Experimental Watershed (RCEW), annual precipitation is unchanged, but these climate trends have altered the snowmelt dominated hydrologic cycle. More precipitation falls as rain, causing earlier snowmelt, earlier peak streamflow, reduced peak snow water equivalent, and reduced summer soil moisture and stream flow. Natural variation in weather and precipitation (wet-dry precipitation and warm-cold weather cycles) make it difficult to quantify the impacts of warming climate on hydrology. This study aims at understanding the differences in sensitivity of wet and dry snow seasons to the warming climate. For this purpose simulation of four snow seasons, representing range of precipitation (wet and dry) and temperature (warm and cold) conditions will be done using a spatially distributed snow energy and mass balance model (Isnobal). The four water years selected represent a range of conditions from early in the RCEW record to recent. The1984 water year (WY) was cold and very wet, the 1987 WY was cold and dry, the 2001 WY was warm and dry, and the 2006 WY was warm and wet. The sensitivity of the development and ablation of the seasonal snowcover to warming climate trends will be evaluated by altering temperature and humidity during the simulations. The 1984 adjusted simulation will use temperature and humidity conditions from 2006, 1987 will use temperature and humidity from 2001, 2001 will use temperature and humidity from 1987, and 2006 will use temperature and humidity from 1984. Comparison of the simulation results for the selected snow seasons under actual and adjusted forcing data would be helpful in understanding the impacts of climate warming on snow hydrology and the differences in sensitivity of wet and dry snow seasons to the warming climate.
C34A-03
Nitrous oxide fluxes from seasonally snow covered subalpine soils
The release to the atmosphere of nitrogen oxides (N2O and NO) produced by soil microorganisms is an important source of greenhouse gases and affects air quality. Recent evidence suggests that considerable microbial activity occurs beneath the snow in seasonally snow-covered soils, but relatively few studies have addressed nitrogen oxide emissions. The soils in the Colorado subalpine may be particularly prone to N2O emissions during the winter. The snow-covered season typically lasts at least six months, and during this time the soils are prevented from freezing by insulation of the snow. Competition between denitrifiers and plants for nitrate is likely lower in the winter. Plant senescence in the fall has recently provided considerable labile carbon. Finally, the amount of N deposition is growing, likely increasing soil nitrate concentrations. In the winter of 2005- 2006 we measured N2O concentration gradients through the snowpack hourly at a subalpine meadow site in the Front Range of the Colorado Rocky Mountains. We believe these measurements represent the first continuous record of winter N2O fluxes from snow covered soils. The snowpack concentrations of N2O were related to snow depth with concentrations peaking at 1000 ppbv at the time of maximum snow depth. On a daily basis the calculated N2O fluxes averaged 1 ng N cm-2 h-1·. The flux also peaked at the time of maximum snow depth. The simultaneous measurements of NO and CO2 in the winter indicated similar seasonal patterns in their concentration profiles. NO fluxes were measurable, but were typically an order of magnitude less than N2O fluxes. Based on weekly chamber flux measurements, the summer fluxes of N2O were, on average, 4-fold greater than the winter fluxes but exhibited a rapid decline as the soils dried out from June to September. Although the fluxes were lower in winter, the cumulative winter flux contributed approximately 20% of the annual flux in 2005-6. A second year of flux measurements is underway which will allow comparisons of interannual variability. The pattern of the winter fluxes may be quite different because the snowpack development was significantly impacted by an unusual snow melt event in March 2006. This event caused the maximum snow depth to occur two months earlier than average. Overall, it appears that the rates of N2O emissions from these seasonally snow-covered subalpine soils are significant, and a sizeable proportion of the annual flux occurred in the winter.
C34A-04
Interactions Between Snow Cover, Frozen Soils, and the Carbon Cycle
We evaluate the complex interactions between frozen soil, snow cover, and the carbon cycle to understand how the soil thermal regime and freeze-thaw processes determine seasonal and inter-annual variability in terrestrial biomass, photosynthesis, respiration, and net CO2 fluxes over continental North America. We use the Simple Biosphere Carnegie-Ames-Stanford Approach (SiBCASA) model driven by the NCEP North American Regional Reanalysis (NARR), the GIMMS NDVI dataset, and in situ snow depth. To improve simulated soil temperatures, we integrated the Sturm et al. [1995] global snow classification system into the snow parameterization from the Community Climate Model and added simple representations of depth hoar and wind slab development. We also added the effects of peat development in tundra and boreal ecosystems. These additions give more realistic soil and snowpack thermodynamic properties, resulting in significantly improved simulated soil temperatures, freeze and thaw depths, and permafrost. Here, we compare simulated and observed soil temperatures, biomass, and carbon fluxes at eddy covariance flux towers representing a broad range of biome types and snow conditions, from permafrost and boreal forests at high latitudes, to deep snow in the northeast US, to wind blown snow of the Great Plains.
C34A-05
The Impact of Heterogeneous Snow Cover On Simulated Northern Hemisphere Carbon, Water and Heat Exchange
We present the impact of a new sub-grid heterogeneous snow scheme in the Joint UK Land Environment Simulator (JULES). Snow tends to accumulate on vegetation and in topographic hollows. These snow drifts melt out later, extending the snow lie period and reducing the available growing season for buried plants. The scheme represents the interaction between vegetation and topography on the landscape scale, and the effect of partial burying of vegetation on gross primary productivity. We present the results from global 1 degree simulations from 1982-1995 for the new scheme and base line simulations. Results suggest that the representation of northern hemisphere snow cover play an important role in the seasonal carbon, water and heat cycles. The improved scheme delays the onset of snow-free conditions by representing the formation of deep drifts which take longer to melt out. In turn, this decreases the volume of surface runoff resulting from the rate of snow melt exceeding the surface infiltration capacity. Snow is maintained longer into the growing season which delays the seasonal drawdown in atmospheric carbon dioxide. Results suggest that this process may improve the representation of biogeochemical cycling and biogeophysics in the seasonally frozen high-latitudes. This represents a step towards improved coupled GCM simulations in these regions.
C34A-06
Observations and Processes Near the Snow-Air Interface: Insights Gained from New and Comparative Sensor Systems in View of Snow Surface Energy Balance Closure
Global warming drastically affects the seasonal snow cover in high altitude regions. The thermodynamic evolution of the snow pack is mainly controlled by the surface energy balance, however, most studies to date fail to close this budget on short time scales when using measurements of all its components. Also dynamic processes such as air movement in the snow pack associated with air exchange and the snow-atmosphere interface have to be taken into account. To investigate snow-atmosphere interaction, measurements of radiative and turbulent heat fluxes, and other meteorological quantities were obtained over a snow-covered glacier in the Swiss Alps during winter 2007. Humidity, air, surface, and snow temperature – quantities required to calculate energy fluxes for the surface energy budget – were measured with different sensors and techniques. Data revealed significant discrepancies between individual measurements at a location and time mainly due to solar heating of the sensors. We show that even shielded sensors overestimate air temperature during the day when compared to a radiation-independent reference sensor (sonic anemometer). Subsurface heat flux was determined from snow internal temperature and density data. High resolution temperature profiles were measured in the snow using traditional (thermocouple) and novel fiber optic distributed temperature instrumentation. To better understand the rate of gas exchange with the atmosphere controlling latent heat transport in the snow associated to phase changes (sublimation/deposition), air movement in the snow was investigated with using a new in-situ carbon monoxide trace gas measurement system providing high-resolution observation of snow transport process without gas extraction. http://eflum.epfl.ch
C34A-07
Evaluation of Measured and Simulated Turbulent Components of a Snow Cover Energy Balance Model in Order to Refine the Turbulent Transfer Algorithm.
Energy balance models use physically based principles to simulate snow cover accumulation and melt. Snobal, a snow cover energy balance model, uses a flux-profile approach to calculating the turbulent flux (sensible and latent heat flux) components of the energy balance. Historically, validation data for turbulent flux simulations have been difficult to obtain at snow dominated sites characterized by complex terrain and heterogeneous vegetation. Currently, eddy covariance (EC) is the most defensible method available to measure turbulent flux and hence to validate this component of an energy balance model. EC was used to measure sensible and latent heat flux at two sites over three winter seasons (2004, 2005, and 2006). Both sites are located in Reynolds Creek Experimental Watershed in southwestern Idaho, USA and are characterized as semi-arid rangeland. One site is on a wind-exposed ridge with small shrubs and the other is in a wind-protected area in a small aspen stand. EC data were post processed from 10 Hz measurements. The first objective of this work was to compare EC- measured sensible and latent heat flux and sublimation/condensation to Snobal-simulated values. Comparisons were made on several temporal scales, including inter-annual, seasonal and diurnal. The flux- profile method used in Snobal assumes equal roughness lengths for moisture and temperature, and roughness lengths are constant and not a function of stability. Furthermore, there has been extensive work on improving profile function constants that is not considered in the current version of Snobal. Therefore, the second objective of this work was to modify the turbulent flux algorithm in Snobal. Modifications were made to calculate roughness lengths as a function of stability and separately for moisture and temperature. Also, more recent formulations of the profile function constants were incorporated. The third objective was to compare EC-measured sensible and latent heat flux and sublimation/condensation to the modified Snobal simulated values. The final objective was to determine if the modified turbulent flux algorithm in Snobal results in hydrologically significant improvements to simulations.
C34A-08
Eddy correlations measurements of turbulent fluxes over a high altitude glacier in the Bolivian Andes
The turbulent fluxes remain poorly known on tropical glaciers characterized by very dry air at high altitude and low thermal seasonality. Different studies based on the bulk method have shown that sublimation can be important during the dry season, reducing the energy available for melting. However, uncertainties on the bulk method are large, especially when the catabatic wind causes a wind speed maximum at low height. Eddy-correlations measurements have been conducted at 5050 m elevation on the Zongo Glacier in Bolivia (16°S) from July to September 2007 as part of the ‘TAG' project (Turbulence et Ablation Glaciaire). Concomitant measurements of all radiation components, snow surface temperature and of vertical gradients over 6 meters of air temperature and wind speed were carried out. The site was approximately level within several hundred meters, with drainage winds prevailing at night and most of the day. The results show that in this dry and thin atmosphere, surface temperatures below freezing are maintained with significant cooling of the surface due to sublimation and very low long-wave radiation input. The resulting latent heat flux rates exceeded 100 W m2 at times and were generally matched by opposing sensible heat fluxes. A comparisons with eddy correlation measurement conducted in summer 2006 on the St Sorlin glacier, French Alps, was also investigated.