Snow Processes Research Articles

A New Formula for Estimating the Threshold Wind Speed for Snow Movement

AbstractThreshold wind speed for snow movement is one of the most important parameters describing the wind‐transport snow process. A majority of previous studies used empirical and constant threshold wind speeds while the variations of atmosphere condition and snow age seem to greatly affect the snow settlement process. This study tested the hypothesis that the threshold wind speed for snow transport increases as deposition time passed since last snowfall by introducing a new formula of the threshold wind speed for snow movement. It was theoretically derived based on the sintering process modeling and the moment balance of a snow particle. Through this formula, the influences of snow particle size, temperature, and deposition time on the threshold wind speed were explicitly taken into consideration. However, two empirical parameters in the sintering process modeling remained in the new formula, and they were determined by a calibration‐validation procedure using the field observed snow flux data. The snow flux and meteorological data collected at the ISAW stations (http://www.iav.ch) including large number of snow transport events during five winter seasons were used to test this formula. It was shown that the new formula qualitatively described the threshold wind speed required for the incipient motion of snow under various conditions in the natural environment of the Alps.

Evaluation and enhancement of permafrost modeling with the NASA Catchment Land Surface Model.

Besides soil hydrology and snow processes, the NASA Catchment Land Surface Model (CLSM) simulates soil temperature in six layers from the surface down to 13m depth. In this study, to examine CLSM's treatment of subsurface thermodynamics, a baseline simulation produced subsurface temperatures for 1980-2014 across Alaska at 9-km resolution. The results were evaluated using in situ observations from permafrost sites across Alaska. The baseline simulation was found to capture the broad features of inter- and intra-annual variations in soil temperature. Additional model experiments revealed that: (i) the representativeness of local meteorological forcing limits the model's ability to accurately reproduce soil temperature, and (ii) vegetation heterogeneity has a profound influence on subsurface thermodynamics via impacts on the snow physics and energy exchange at surface. Specifically, the profile-average RMSE for soil temperature was reduced from 2.96°C to 2.10°C at one site and from 2.38°C to 2.25°C at another by using local forcing and land cover, respectively. Moreover, accounting for the influence of soil organic carbon on the soil thermal properties in CLSM leads to further improvements in profile-average soil temperature RMSE, with reductions of 16% to 56% across the different study sites. The mean bias of climatological ALT is reduced by 36% to 89%, and the RMSE is reduced by 11% to 47%. Finally, results reveal that at some sites it may be essential to include a purely organic soil layer to obtain, in conjunction with vegetation and snow effects, a realistic "buffer zone" between the atmospheric forcing and soil thermal processes.

Improving the Representation of Polar Snow and Firn in the Community Earth System Model

AbstractIn Earth system models, terrestrial snow is usually modeled by the land surface component. In most cases, these snow models have been developed with an emphasis on seasonal snow. Questions about future sea level rise, however, prompt the need for a realistic representation of perennial snow, as snow processes play a key role in the mass balance of glaciers and ice sheets. Here we enhance realism of modeled polar snow in the Community Land Model (CLM), the land component of the Community Earth System Model (CESM), by implementing (1) new parametrizations for fresh snow density, destructive metamorphism, and compaction by overburden pressure, (2) by allowing for deeper snow packs, and (3) by introducing drifting snow compaction, with a focus on the ice sheet interior. Comparison with Greenlandic and Antarctic snow density observations show that the new physics improve model skill in predicting firn and near‐surface density in the absence of melt. Moreover, compensating biases are removed and spurious subsurface melt rates at ice sheets are eliminated. The deeper snow pack enhances refreezing and allows for deeper percolation, raising ice temperatures up to 15°C above the skin temperature.

Disentangling the mechanisms behind winter snow impact on vegetation activity in northern ecosystems.

Although seasonal snow is recognized as an important component in the global climate system, the ability of snow to affect plant production remains an important unknown for assessing climate change impacts on vegetation dynamics at high-latitude ecosystems. Here, we compile data on satellite observation of vegetation greenness and spring onset date, satellite-based soil moisture, passive microwave snow water equivalent (SWE) and climate data to show that winter SWE can significantly influence vegetation greenness during the early growing season (the period between spring onset date and peak photosynthesis timing) over nearly one-fifth of the land surface in the region north of 30 degrees, but the magnitude and sign of correlation exhibits large spatial heterogeneity. We then apply an assembled path model to disentangle the two main processes (via changing early growing-season soil moisture, and via changing the growth period) in controlling the impact of winter SWE on vegetation greenness, and suggest that the "moisture" and "growth period" effect, to a larger extent, result in positive and negative snow-productivity associations, respectively. The magnitude and sign of snow-productivity association is then dependent upon the relative dominance of these two processes, with the "moisture" effect and positive association predominating in Central, western North America and Greater Himalaya, and the "growth period" effect and negative association in Central Europe. We also indicate that current state-of-the-art models in general reproduce satellite-based snow-productivity relationship in the region north of 30 degrees, and do a relatively better job of capturing the "moisture" effect than the "growth period" effect. Our results therefore work towards an improved understanding of winter snow impact on vegetation greenness in northern ecosystems, and provide a mechanistic basis for more realistic terrestrial carbon cycle models that consider the impacts of winter snow processes.

Snow and ice melting properties of self-healing asphalt mixtures with induction heating and microwave heating

In this paper, the snow and ice melting properties of self-healing asphalt mixtures with induction heating and microwave heating were investigated. The self-healing performance of the mixtures during ice melting process was measured with cyclic three-point bending test as well. The results indicated that the snow and ice melting performance of self-healing asphalt mixtures with appropriate heating techniques was remarkable. The average melting velocity of ice melting was improved from less than 1 g/min (previous researches) to more than 10 g/min, while the average snow melting velocity of steel fiber modified asphalt mixture and steel slag asphalt mixture with microwave heating could reach 53.9 g/min and 48.5 g/min respectively. The water from melted ice/snow played a major role in the melting process of snow with microwave heating. However, the moisture from melted ice/snow on the surfaces of crack prevented the thermal healing of asphalt mixtures. It is recommended that another heating process after the melting process should be applied to promote the healing of cracks as well as to prevent the melted snow and ice water on the pavement from freezing.

Improvement of a snow albedo parameterization in the Snow–Atmosphere–Soil Transfer model: evaluation of impacts of aerosol on seasonal snow cover

The presence of light-absorbing aerosols (LAA) in snow profoundly influence the surface energy balance and water budget. However, most snow-process schemes in land-surface and climate models currently do not take this into consideration. To better represent the snow process and to evaluate the impacts of LAA on snow, this study presents an improved snow albedo parameterization in the Snow–Atmosphere–Soil Transfer (SAST) model, which includes the impacts of LAA on snow. Specifically, the Snow, Ice and Aerosol Radiation (SNICAR) model is incorporated into the SAST model with an LAA mass stratigraphy scheme. The new coupled model is validated against in-situ measurements at the Swamp Angel Study Plot (SASP), Colorado, USA. Results show that the snow albedo and snow depth are better reproduced than those in the original SAST, particularly during the period of snow ablation. Furthermore, the impacts of LAA on snow are estimated in the coupled model through case comparisons of the snowpack, with or without LAA. The LAA particles directly absorb extra solar radiation, which accelerates the growth rate of the snow grain size. Meanwhile, these larger snow particles favor more radiative absorption. The average total radiative forcing of the LAA at the SASP is 47.5 W m−2. This extra radiative absorption enhances the snowmelt rate. As a result, the peak runoff time and “snow all gone” day have shifted 18 and 19.5 days earlier, respectively, which could further impose substantial impacts on the hydrologic cycle and atmospheric processes.

Temporal and Spatial Variations of Hydrological Processes on the Landscape Zone Scale in an Alpine Cold Region (Mafengou River Basin, China): An Update

This study investigates precipitation, snow, groundwater, glaciers and frozen soil in different landscape zones using isotopic and hydrogeochemical tracers. The aim of this study is to identify temporal and spatial variations, as well as hydrological processes in the alpine cold region. The results show that there was no significant difference in water chemical characteristics of various waterbodies, and no obvious temporal variation, but exhibited spatial variation. In the wet season, various waterbodies are enriched in oxygen δ18O and deuterium δD due to a temperature effect. Precipitation and the temperature decrease during the dry season, which cannot easily be affected by secondary evaporation. The d-excess (deuterium excess) of various waterbodies was greater than 10‰. There are no altitude effects during wet and dry seasons because the recharged water resources are different in the wet and dry seasons. It is influenced by the freezing-thawing process of glacier snow and frozen soil. The river water is recharged by thawed frozen soil water and precipitation in the wet season, but glacier snow meltwater with negative δ18O and δD is less (14–18%). In the dry season, glacier snow meltwater and groundwater are the dominant source of the river water, and thawed frozen soil water is less (10–15%).

Natural and human-induced terrestrial water storage change: A global analysis using hydrological models and GRACE

Hydrological models and the data derived from the Gravity Recovery and Climate Experiment (GRACE) satellite mission have been widely used to study the variations in terrestrial water storage (TWS) over large regions. However, both GRACE products and model results suffer from inherent uncertainties, calling for the need to make a combined use of GRACE and models to examine the variations in total TWS and their individual components, especially in relation to natural and human-induced changes in the terrestrial water cycle. In this study, we use the results from two state-of-the-art hydrological models and different GRACE spherical harmonic products to examine the variations in TWS and its individual components, and to attribute the changes to natural and human-induced factors over large global river basins. Analysis of the spatial patterns of the long-term trend in TWS from the two models and GRACE suggests that both models capture the GRACE-measured direction of change, but differ from GRACE as well as each other in terms of the magnitude over different regions. A detailed analysis of the seasonal cycle of TWS variations over 30 river basins shows notable differences not only between models and GRACE but also among different GRACE products and between the two models. Further, it is found that while one model performs well in highly-managed river basins, it fails to reproduce the GRACE-observed signal in snow-dominated regions, and vice versa. The isolation of natural and human-induced changes in TWS in some of the managed basins reveals a consistently declining TWS trend during 2002–2010, however; significant differences are again obvious both between GRACE and models and among different GRACE products and models. Results from the decomposition of the TWS signal into the general trend and seasonality indicate that both models do not adequately capture both the trend and seasonality in the managed or snow-dominated basins implying that the TWS variations from a single model cannot be reliably used for all global regions. It is also found that the uncertainties arising from climate forcing datasets can introduce significant additional uncertainties, making direct comparison of model results and GRACE products even more difficult. Our results highlight the need to further improve the representation of human land-water management and snow processes in large-scale models to enable a reliable use of models and GRACE to study the changes in freshwater systems in all global regions.

Evaluation of Radial Pressure Generated at Cylindrical Pressure Vessel Wall during High Speed Compression Formation of Ice Pieces.

In snowy country, a compaction process of snow has been demanded for performing efficient snow removal and disposal work. For achieving the work, a design of large pressure vessel is demanded in order to perform high speed compression of the snow and ice. This report aims to investigate the radial pressure generated during the high speed compression of ice pieces for obtaining a fundamental data to design a large pressure vessel. In the investigation, ice pieces were compressed by an axial formation pressure pz for pz = 0.0 to 5.0 MPa, and the radial pressure pr during the compression test was evaluated. It was found that the ice column having D = 80 mm diameter, H = 80 mm height and density of ρ = 0.84 g·cm-3 was obtained at pz = 5.0 MPa. And, the maximum radial pressure pr_max reached about 60 % of the axial formation pressure pz.

Snow water equivalent in the Alps as seen by gridded data sets, CMIP5 and CORDEX climate models

Abstract. The estimate of the current and future conditions of snow resources in mountain areas would require reliable, kilometre-resolution, regional-observation-based gridded data sets and climate models capable of properly representing snow processes and snow–climate interactions. At the moment, the development of such tools is hampered by the sparseness of station-based reference observations. In past decades passive microwave remote sensing and reanalysis products have mainly been used to infer information on the snow water equivalent distribution. However, the investigation has usually been limited to flat terrains as the reliability of these products in mountain areas is poorly characterized.This work considers the available snow water equivalent data sets from remote sensing and from reanalyses for the greater Alpine region (GAR), and explores their ability to provide a coherent view of the snow water equivalent distribution and climatology in this area. Further we analyse the simulations from the latest-generation regional and global climate models (RCMs, GCMs), participating in the Coordinated Regional Climate Downscaling Experiment over the European domain (EURO-CORDEX) and in the Fifth Coupled Model Intercomparison Project (CMIP5) respectively. We evaluate their reliability in reproducing the main drivers of snow processes – near-surface air temperature and precipitation – against the observational data set EOBS, and compare the snow water equivalent climatology with the remote sensing and reanalysis data sets previously considered. We critically discuss the model limitations in the historical period and we explore their potential in providing reliable future projections.The results of the analysis show that the time-averaged spatial distribution of snow water equivalent and the amplitude of its annual cycle are reproduced quite differently by the different remote sensing and reanalysis data sets, which in fact exhibit a large spread around the ensemble mean. We find that GCMs at spatial resolutions equal to or finer than 1.25° longitude are in closer agreement with the ensemble mean of satellite and reanalysis products in terms of root mean square error and standard deviation than lower-resolution GCMs. The set of regional climate models from the EURO-CORDEX ensemble provides estimates of snow water equivalent at 0.11° resolution that are locally much larger than those indicated by the gridded data sets, and only in a few cases are these differences smoothed out when snow water equivalent is spatially averaged over the entire Alpine domain. ERA-Interim-driven RCM simulations show an annual snow cycle that is comparable in amplitude to those provided by the reference data sets, while GCM-driven RCMs present a large positive bias. RCMs and higher-resolution GCM simulations are used to provide an estimate of the snow reduction expected by the mid-21st century (RCP 8.5 scenario) compared to the historical climatology, with the main purpose of highlighting the limits of our current knowledge and the need for developing more reliable snow simulations.

Modeling the isotopic evolution of snowpack and snowmelt: Testing a spatially distributed parsimonious approach.

Use of stable water isotopes has become increasingly popular in quantifying water flow paths and travel times in hydrological systems using tracer‐aided modeling. In snow‐influenced catchments, snowmelt produces a traceable isotopic signal, which differs from original snowfall isotopic composition because of isotopic fractionation in the snowpack. These fractionation processes in snow are relatively well understood, but representing their spatiotemporal variability in tracer‐aided studies remains a challenge. We present a novel, parsimonious modeling method to account for the snowpack isotope fractionation and estimate isotope ratios in snowmelt water in a fully spatially distributed manner. Our model introduces two calibration parameters that alone account for the isotopic fractionation caused by sublimation from interception and ground snow storage, and snowmelt fractionation progressively enriching the snowmelt runoff. The isotope routines are linked to a generic process‐based snow interception‐accumulation‐melt model facilitating simulation of spatially distributed snowmelt runoff. We use a synthetic modeling experiment to demonstrate the functionality of the model algorithms in different landscape locations and under different canopy characteristics. We also provide a proof‐of‐concept model test and successfully reproduce isotopic ratios in snowmelt runoff sampled with snowmelt lysimeters in two long‐term experimental catchment with contrasting winter conditions. To our knowledge, the method is the first such tool to allow estimation of the spatially distributed nature of isotopic fractionation in snowpacks and the resulting isotope ratios in snowmelt runoff. The method can thus provide a useful tool for tracer‐aided modeling to better understand the integrated nature of flow, mixing, and transport processes in snow‐influenced catchments.

Snow process monitoring in montane forests with time‐lapse photography

AbstractA camera network with hourly resolution was used to monitor the complex snow processes in montane forest environments. We developed a semi‐automatic procedure to interpret snow depths from the digital images, which exhibited high consistency with manual measurements and station‐based recordings. To extract snow interception dynamics, six binary classification methods were compared. The MaxEntropy classifier demonstrated better performance than the other methods under conditions of varying illumination and was therefore selected as the method used for quantifying snow in tree canopies. Snow accumulation and ablation on the ground, as well as snow loading and unloading in the forest canopies, were investigated using snow parameters derived from the time‐lapse photography monitoring. The influences of meteorologic conditions, forest cover, and elevation on the snow processes were also considered. Time‐lapse photography proved to be an effective and low‐cost approach for collecting useful information on snow processes and facilitating the set‐up of hydrological models.

Effect of spatial variability of wet snow on modeled and observed microwave emissions

Melting snow provides an essential source of water in many regions of the world and can also contribute to devastating, wide-scale flooding. Global datasets of recorded passive microwave emissions provide non-destructive, daily information on snow processes including the presence of liquid water in the snow, which can be an indicator of snowmelt. The objective of this research is to test the sensitivity of the emission signal as it relates to the spatial distribution of liquid water content in the snowpack. This signal response was evaluated over an area approximately the size of a microwave pixel to assess whether a relationship exists between the aerial extent of wet snow and the magnitude of the TB response. A sensitivity analysis was performed using a high-resolution, physically based snow-emission model to simulate microwave emissions. The signal response to wet snow was evaluated given a range of spatially distributed snowpack conditions. Daily snow states were simulated for a 9-year period using a high-resolution (50m) energy balance snow model over a 34×34km domain. These data were fed into a microwave emission model to simulate brightness temperatures. A near-linear relationship was found between the TB signal response over a spatially heterogeneous snowpack and the percent area with liquid water content (LWC) present. The results were confirmed by evaluating actual wet snow events over a 9-year period. The model output was also compared to AMSR-E passive microwave satellite data and discharge data at a basin outlet within the study area. The results are used to help understand the impact of spatially distributed snowmelt as detected by passive microwave data.

A framework to improve hyper-resolution hydrological simulation in snow-affected regions

Snow processes in mid- and north-latitude basins and their interaction with runoff generation at hyperresolution (<1km and <hourly) pose challenges in current state-of-the-art distributed hydrological models. These models run typically at macro to moderate scales (>5km), representing land surface processes based on simplified couplings of snow thermal physics and the water cycle in the soil-vegetation-atmosphere (SVA) layers. This paper evaluates a new hydrological model capable of simulating river flows for a range of basin scales (100km2 to >10,000km2), and a particular focus on mid- and north-latitude regions. The new model combines the runoff generation and fully distributed routing framework of the Coupled Routing and Excess STorage (CREST) model with a new land surface process model that strictly couples water and energy balances at the SVA layer, imposing closed energy balance solutions. The model is vectorized and parallelized to achieve long-term (>30years) high-resolution (30m to 500m and subhourly) simulations of large river basins utilizing high-performance computing. The model is tested in the Connecticut River basin (20,000km2), where flooding is frequently associated with interactions of snowmelt triggered by rainfall events. Model simulations of distributed evapotranspiration (ET) and snow water equivalence (SWE) at daily time step are shown to match accurately ET estimates from MODIS (average NSCE and bias are 0.77 and 6.79%) and SWE estimates from SNODAS (average correlation and normalized root mean square error are 0.94 and of 19%); the modeled daily river flow simulations exhibit an NSCE of 0.58 against USGS streamflow observations.

Regional distribution and variability of model-simulated Arctic snow on sea ice

Numerical models face the challenge of representing the present-day spatiotemporal distribution of snow on sea ice realistically. We present modeled Arctic-wide snow depths on sea ice (hs_mod) obtained with the MITgcm configured with a single snow layer that accumulates proportionally to the thickness of sea ice. When compared to snow depths derived from radar measurements (NASA Operation IceBridge, 2009–2013), the model snow depths are overestimated on first-year ice (2.5 ± 8.1 cm) and multiyear ice (0.8 ± 8.3 cm). The large variance between model and observations lies mainly in the limitations of the model snow scheme and the large uncertainties in the radar measurements. In a temporal analysis, during the peak of snowfall accumulation (April), hs_mod show a decline between 2000 and 2013 associated to long-term reduction of summer sea ice extent, surface melting and sublimation. With the aim of gaining knowledge on how to improve hs_mod, we investigate the contribution of the explicitly modeled snow processes to the resulting hs_mod. Our analysis reveals that this simple snow scheme offers a practical solution to general circulation models due to its ability to replicate robustly the distribution of the large-scale Arctic snow depths. However, benefit can be gained from the integration of explicit wind redistribution processes to potentially improve the model performance and to better understand the interaction between sources and sinks of contemporary Arctic snow.

Observations of atmospheric chemical deposition to high Arctic snow

Abstract. Rapidly rising temperatures and loss of snow and ice cover have demonstrated the unique vulnerability of the high Arctic to climate change. There are major uncertainties in modelling the chemical depositional and scavenging processes of Arctic snow. To that end, fresh snow samples collected on average every 4 days at Alert, Nunavut, from September 2014 to June 2015 were analyzed for black carbon, major ions, and metals, and their concentrations and fluxes were reported. Comparison with simultaneous measurements of atmospheric aerosol mass loadings yields effective deposition velocities that encompass all processes by which the atmospheric species are transferred to the snow. It is inferred from these values that dry deposition is the dominant removal mechanism for several compounds over the winter while wet deposition increased in importance in the fall and spring, possibly due to enhanced scavenging by mixed-phase clouds. Black carbon aerosol was the least efficiently deposited species to the snow.

Modelling the thermal conductivity of a melting snow layer on a heated pavement

A snow layer on a heated pavement strongly affects the energy balance at the pavement surface. Freshly fallen snow has a fairly low thermal conductivity and acts as a good insulator. The thermal properties can vary greatly during the melting process. This study looks into the change of the thermal conductivity during the melting process of dry uncompressed and compressed snow on heated pavements. The energy balance of the heated pavement system is described and the effective thermal conductivity of the melting snow layer during the melting process of snow on a heated pavement system is calculated based on the volume fractions and thermal conductivity of ice, water and air. The results show that the thermal conductivity of an uncompressed and compressed melting snow layer on a heated pavement can be best described as a combination of a parallel and series system. Ice and air are modelled as a series system and water and ice/air are modelled in parallel.

Modeling blowing snow accumulation downwind of an obstruction: The Ohara Eulerian particle distribution equation

AbstractAn equation was proposed to model the height of blowing snow accumulation downwind of an obstacle such as vegetation, a snow fence, a building, or a topographic feature. The equation does not require aerodynamic flow condition parameters such as wind speed, allowing for the spatial distribution of snow to be determined at locations where meteorological data is not available. However, snow particle diffusion, drift, and erosion coefficients must be estimated for application of the equation. These coefficients can be used to provide insight into the relative magnitude of blowing snow processes at a field location. Further research is required to determine efficient methods for coefficient estimation. The equation could be used with other models of wind‐transported snow to predict snow accumulation downwind of an obstacle without the need for wind speed adjustments or correction equations. Applications for this equation include the design of snow fences, and the use of this equation with other hydrological models to predict snow distribution, climate change, drought, flooding, and avalanches.

Snow disappearance timing is dominated by forest effects on snow accumulation in warm winter climates of the Pacific Northwest, United States

AbstractForests modify snow processes and affect snow water storage as well as snow disappearance timing. However, forest influences on snow accumulation and ablation vary with climate and topography and are therefore subject to temporal and spatial variability. We utilize multiple years of snow observations from across the Pacific Northwest, United States, to assess forest–snow interactions in the relatively warm winter conditions characteristic of maritime and transitional maritime–continental climates. We (a) quantify the difference in snow magnitude and disappearance timing between forests and open areas and (b) assess how forest modifications of snow accumulation and ablation combine to determine whether snow disappears later in the forest or in the open.We find that snow disappearance timing at 12 (out of 14) sites ranges from synchronous in the forest and open to snow persisting up to 13 weeks longer in the open relative to a forested area. By analyzing accumulation and ablation rates up to the day when snow first disappears from the forest, we find that the difference between accumulation rates in the open and forest is larger than the difference between ablation rates. Thus, canopy snow interception and subsequent loss, rather than ablation, set up longer snow duration in the open. However, at two relatively windy sites (hourly average wind speeds up to 8 and 17 m/s), differential snow disappearance timing is reversed: Snow persists 2–5 weeks longer in the forest. At the windiest sites, accumulation rates in the forest and open are similar. Ablation rates are higher in the open, but the difference between ablation rates in the forest and open at these sites is approximately equivalent to the difference at less windy sites. Thus, longer snow retention in the forest at the windiest sites is controlled by depositional differences rather than by reduced ablation rates. These findings suggest that improved quantification of forest effects on snow accumulation processes is needed to accurately predict the effect of forest management or natural disturbance on snow water resources.

Uncertainty analysis of impacts of climate change on snow processes: Case study of interactions of GCM uncertainty and an impact model

The impact of climate change on snow water equivalent (SWE) and its uncertainty were investigated in snowy areas of subarctic and temperate climate zones in Japan by using a snow process model and climate projections derived from general circulation models (GCMs). In particular, we examined how the uncertainty due to GCMs propagated through the snow model, which contained nonlinear processes defined by thresholds, as an example of the uncertainty caused by interactions among multiple sources of uncertainty. An assessment based on the climate projections in Coupled Model Intercomparison Project Phase 5 indicated that heavy-snowfall areas in the temperate zone (especially in low-elevation areas) were markedly vulnerable to temperature change, showing a large SWE reduction even under slight changes in winter temperature. The uncertainty analysis demonstrated that the uncertainty associated with snow processes (1) can be accounted for mainly by the interactions between GCM uncertainty (in particular, the differences of projected temperature changes between GCMs) and the nonlinear responses of the snow model and (2) depends on the balance between the magnitude of projected temperature changes and present climates dominated largely by climate zones and elevation. Specifically, when the peaks of the distributions of daily mean temperature projected by GCMs cross the key thresholds set in the model, the GCM uncertainty, even if tiny, can be amplified by the nonlinear propagation through the snow process model. This amplification results in large uncertainty in projections of CC impact on snow processes.