Snow Water Equivalent Data Research Articles

Characterizing parameter sensitivity and uncertainty for a snow model across hydroclimatic regimes

The National Weather Service (NWS) uses the SNOW17 model to forecast snow accumulation and ablation processes in snow-dominated watersheds nationwide. Successful application of the SNOW17 relies heavily on site-specific estimation of model parameters. The current study undertakes a comprehensive sensitivity and uncertainty analysis of SNOW17 model parameters using forcing and snow water equivalent (SWE) data from 12 sites with differing meteorological and geographic characteristics. The Generalized Sensitivity Analysis and the recently developed Differential Evolution Adaptive Metropolis (DREAM) algorithm are utilized to explore the parameter space and assess model parametric and predictive uncertainty. Results indicate that SNOW17 parameter sensitivity and uncertainty generally varies between sites. Of the six hydroclimatic characteristics studied, only air temperature shows strong correlation with the sensitivity and uncertainty ranges of two parameters, while precipitation is highly correlated with the uncertainty of one parameter. Posterior marginal distributions of two parameters are also shown to be site-dependent in terms of distribution type. The SNOW17 prediction ensembles generated by the DREAM-derived posterior parameter sets contain most of the observed SWE. The proposed uncertainty analysis provides posterior parameter information on parameter uncertainty and distribution types that can serve as a foundation for a data assimilation framework for hydrologic models.

Noah LSM Snow Model Diagnostics and Enhancements

Abstract A negative snow water equivalent (SWE) bias in the snow model of the Noah land surface scheme used in the NCEP suite of numerical weather and climate prediction models has been noted by several investigators. This bias motivated a series of offline tests of model extensions and improvements intended to reduce or eliminate the bias. These improvements consist of changes to the model’s albedo formulation that include a parameterization for snowpack aging, changes to how pack temperature is computed, and inclusion of a provision for refreeze of liquid water in the pack. Less extensive testing was done on the performance of model extensions with alternate areal depletion parameterizations. Model improvements were evaluated through comparisons of point simulations with National Resources Conservation Service (NRCS) Snowpack Telemetry (SNOTEL) SWE data for deep-mountain snowpacks at selected stations in the western United States, as well as simulations of snow areal extent over the conterminous United States (CONUS) domain, compared with observational data from the NOAA Interactive Multisensor Snow and Ice Mapping System (IMS). The combination of snow-albedo decay and liquid-water refreeze results in substantial improvements in the magnitude and timing of peak SWE, as well as increased snow-covered extent at large scales. Modifications to areal snow depletion thresholds yielded more realistic snow-covered albedos at large scales.

Defining similar regions of snow in the Colorado River Basin using self‐organizing maps

We used self‐organizing maps (SOMs) to define regions of homogeneity in the Colorado River Basin using snow telemetry (SNOTEL) snow water equivalent (SWE) data. SOMs are a specific application of artificial neural networks. Daily data for 216 stations using 15 years (1991–2005) of data from 1 October through 30 June were used. To identify areas of similar snow accumulation, persistence, and ablation patterns, data were transformed by dividing by the 15 year average peak SWE. Three experiments were performed to determine how the regions of homogeneous snowpack characteristics changed. The number of groups was increased from 4 to 6 to 9 to 16. By increasing this resolution, more subtle variations were defined. The temporal resolution of the data was decreased from daily to weekly to monthly to yearly. The accumulation and ablation of the snowpack over time represents a plot called a niveograph, which was summarized for yearly data by three variables. These were peak SWE, length of season, and date of peak SWE. Very similar results were produced using daily, weekly, and monthly time steps. However, using peak SWE produced only 50% of the same groupings, while using the other annual summary variables, even together, produced less than 25% of the same groupings. Using 18 physiographic variables to represent the SNOTEL stations yielded groups that were similar to those from using peak SWE but more evenly distributed in space. Using Ward's method of cluster analysis could only be performed with the individual annual summary variables. It produced groupings similar to the comparable SOM application but slightly less representative of the daily data groupings.

Spatial‐temporal patterns of snow cover in western Canada

Snow cover is often measured as snow‐water equivalent (SWE), which refers to the amount of water stored in a snow pack that would be available upon melting. Snow cover and SWE represent a source of local snow‐melt release, and are sensitive to regional and global atmospheric circulation, and changes in climate. Monitoring SWE using satellite‐based passive microwave radiometry has provided nearly three decades of continuous data for North America.The availability of spatially and temporally extensive SWE data enables a better understanding of the nature of space‐time trends in snow cover, changes in these trends and linking these trends to underlying landscape and terrain characteristics. To address these interests, we quantify the spatial pattern of SWE by applying a local measure of spatial autocorrelation to 25 years of mean February SWE derived from passive microwave retrievals. Using a method for characterizing the temporal trends in the spatial pattern of SWE, temporal trends and variability in spatial autocorrelation are quantified. Results indicate that within the Canadian Prairies, extreme values of SWE are becoming more spatially coherent, with potential impacts on water availability, and hazards such as flooding. These results also highlight the need for Canadian ecological management units that consider winter conditions.

Improved Simulation of the East Asian Summer Monsoon Rainfall with Satellite-Derived Snow Water Equivalent Data*

Abstract The effect of Eurasian spring snow amount on the summer monsoon rainfall over East Asia has been studied both observationally and numerically. The results indicate that the Eurasian spring snow amount could be important for seasonal prediction of East Asian summer monsoon (EASM) rainfall. Therefore, accurately initializing snow could be critical to improving seasonal prediction of EASM rainfall by numerical models. An attempt has been made in this study to initialize snow in a regional climate model using snow water equivalent (SWE) data derived from a microwave imager. Results from an ensemble seasonal prediction experiment for the 2005 EASM show that the satellite-derived SWE data can be effectively used to initialize a dynamical seasonal prediction model, which leads to improved seasonal prediction of EASM rainfall. Possible effects of snow anomalies over the Tibetan Plateau on EASM rainfall were also studied through a comparative ensemble simulation in which snow was initialized by spinning up the same model from the previous winter. It is found that the anomalous snow amount over the Tibetan Plateau could lead to cooling of the surface and lower troposphere not only over the Tibetan Plateau but also in the surrounding areas because of the reduced net surface shortwave radiation associated with the high snow albedo. This would result in positive anomalies in geopotential height and weaken the cyclonic monsoon circulation in the lower troposphere in East Asia, causing a rainfall increase in South China but a reduction in the Yangtze River Valley in early summer (May–June). The difference in rainfall in midsummer (July–August) was not significant when compared with that in early summer. The surface heat budget indicates that the reduced net surface shortwave radiation is largely balanced by the reduced surface sensible heat flux.

Comparison of different automatic methods for estimating snow water equivalent

Manual measurements of snow water equivalent (SWE) require a considerable effort, which is why several automatic methods for single-point SWE estimates have been developed. This study provides a first comprehensive review and comparison of seven different automatic ways to determine SWE and/or the daily new snow water equivalent (HNW). We therefore deployed the methods tested over 4 subsequent winter seasons at a test site in the Swiss Alps. As reference we manually measured SWE on a biweekly schedule and HNW on a daily schedule over the same four winter periods. We tested 4 functional categories of methods: (1) direct recording of SWE ( SNOWPILLOW , SNOWPOWER ), (2) direct recording of HNW ( PARSIVEL , GAUGE ), (3) complex numerical models driven by meteorological data ( SNOWPACK , COSMO-7 ), and (4) a simple stochastic model based on snow depth (HS) data ( SIMPLE ). As at our site melting/sublimation was insignificant for the mass balance during the snow accumulation period, differential SWE data could be converted into HNW, while HNW data could be cumulated to SWE. In general, our assessment showed that most of the methods performed reasonably well with respect to SWE estimations, but featured clear deficiencies with respect to HNW estimations. SNOWPILLOW , SNOWPACK , and SIMPLE showed a reasonable performance with respect to both, SWE and HNW estimates, and may hence be attractive for general-purpose applications. However, method-specific limitations have to be considered. This study is intended to facilitate finding the optimum instrumentation for specific applications, different purposes, and given boundary conditions.

‘Quantifying the effects of forest canopy cover on net snow accumulation at a continental, mid‐latitude site’

AbstractAlthough many studies have investigated the effects of forest cover on streamflow and runoff, and several have examined the effects of canopy density on snowpack accumulation, the impacts of forest canopy density on spatial patterns of snowmelt input to catchments remain relatively underquantified. We performed an intensive snow depth and density survey during maximum accumulation in a mid‐latitude montane environment in northern New Mexico, taking 900 snow depth measurements and excavating six snow pits across a continuum of canopy densities. Snow water equivalent (SWE) data are correlated with forest canopy density (R2 = 0·21, p < 0·0001), with maximum snow accumulation in forests with density between 25 and 40%. Forest edges are shown to be highly influential on patterns of snow depth, with unforested areas shaded by forest to their immediate south holding approximately 25% deeper snow than either large open areas or densely forested areas. This indicates that the combination of canopy influences on throughfall and snowpack shading are key processes underlying snow distribution in the high solar load environments typical of mountainous, mid‐latitude areas. We further show that statistical models of snow distribution are improved with the addition of remotely sensed forest canopy information (R2 increased in 10 of 11 cases, deviance lowered in 9 of 11 cases), making these findings broadly relevant for improving estimation of water resources, predicting the ecohydrological implications of vegetation and climate change, and informing integrated forest and water resources management. Copyright © 2009 John Wiley & Sons, Ltd.

Comparison of Four Instruments for Measuring Solid Precipitation Below the Freezing Point Condition

A comparative observation of solid precipitation was carried out at conditions below freezing point during the winter of 2005-2006 by using four instruments, including non-collecting measurement instruments. Based on a comparison with the measured increment of snow water equivalent data, data corrected for the catch ratio from a Japanese standard precipitation gauge (RT-4), measured data from a non-collecting sensor using a laser beam (PWD11) and a totalizer using a load sensor (T-200B) accurately determined the amount of solid precipitation. The measured T-200B data fluctuated during the period of non-precipitation. Positive results of consecutive time were obtained from the measured PWD11 and T-200B data. Data collected in this study from a non-collecting sensor using microwaves (R2S) were unstable or underestimated the amount of precipitation.

Snow cover characteristics in the Aral Sea Basin from different data sources and their relation with river runoff

The snow water equivalent (SWE) and snow depth (SD) in the Aral Sea Basin from the NCEP/DOE, ERA-40 and JRA-25 reanalyses were examined against the measured data. The SWE data from the ERA-40 and SD from the JRA-25 reanalyses demonstrated good agreement with the measured data in the plain part of the Aral Sea Basin in the terms of the mean values (1979–2000) and the correlation coefficients. For the mountain part of the basin, however, the agreement was less consistent, and all the reanalyses tended to underestimate the SWE and SD. The interannual variability of the snow cover area (SCA) from satellite data in the Aral Sea Basin (1971–2006) was also analyzed. The biggest fluctuations of the SCA were observed between 1971 and 1982, and between 1995 and 2005, and the lowest ones during 1983 through 1993. The analysis of the long-term variability of the SD revealed a positive trend during the period from 1930 through 1965 in the upper and middle reaches of the Amudarya and Syrdarya rivers. After 1965, a tendency towards a decrease of the SD in the southern part of the Aral Sea Basin was identified. The analysis of the long-term trends of the SD and the river runoffs in the Amudarya and Syrdarya basins showed that the runoffs of Amudarya and, especially, Syrdarya rivers were increasing, while the SD was mostly decreasing. The SCA in the Aral Sea Basin also exhibited a statistically significant decreasing trend.

Toward improved daily snow cover mapping with advanced combination of MODIS and AMSR-E measurements

Taking three snow seasons from November 1 to March 31 of year 2002 to 2005 in northern Xinjiang, China as an example, this study develops a new daily snow cover product (500 m) through combining MODIS daily snow cover data and AMSR-E daily snow water equivalent (SWE) data. By taking advantage of both high spatial resolution of optical data and cloud transparency of passive microwave data, the new daily snow cover product greatly complements the deficiency of MODIS product when cloud cover is present especially for snow cover product on a daily basis and effectively improves daily snow detection accuracy. In our example, the daily snow agreement of the new product with the in situ measurements at 20 stations is 75.4%, which is much higher than the 33.7% of the MODIS daily product in all weather conditions, even a little higher than the 71% of the MODIS 8-day product (cloud cover of ~ 5%). Our results also indicate that i) AMSR-E daily SWE imagery generally agrees with MOD10A1 data in detecting snow cover, with overall agreement of 93.4% and snow agreement of 96.6% in the study area; ii) AMSR-E daily SWE imagery underestimates the snow covered area (SCA) due to its coarse spatial resolution; iii) The new snow cover product can better and effectively capture daily SCA dynamics during the snow seasons, which plays a significant role in reduction, mitigation, and prevention of snow-caused disasters in pastoral areas.

Combining Artificial Neural Network Models, Geostatistics, and Passive Microwave Data for Snow Water Equivalent Retrieval and Mapping

A new modeling framework combining neural-network-based models, passive microwave data, and geostatistics is proposed for snow water equivalent (SWE) retrieval and mapping. Brightness temperature data from the seven-channel special sensor microwave/imager and the interpolated minimum temperature are the inputs of a multilayer feedforward neural network (MFF). Kriging with an external drift algorithm is applied to ground-based SWE data to produce gridded SWE data that are used as the target of the neural network. An optimal division of the sample of available pixels is achieved by a self-organizing feature map. Prediction error is used for model selection and is assessed by bootstrap. It is shown that a committee of a network containing neural networks with different architectures can provide consistent SWE retrievals. This modeling framework is applied for SWE retrieval and mapping over La Grande River basin in north eastern Quebec (Canada). The results are very promising for operational purposes particularly for SWE mapping during periods with no ground measurements and operational streamflow forecasting.

Enhancing the estimation of continental‐scale snow water equivalent by assimilating MODIS snow cover with the ensemble Kalman filter

High‐quality continental‐scale snow water equivalent (SWE) data sets are generally not available, although they are important for climate research and water resources management. This study investigates the feasibility of a framework for developing such needed data sets over North America, through the ensemble Kalman filter (EnKF) approach, which assimilates the snow cover fraction observed by the Moderate Resolution Imaging Spectroradiometer (MODIS) into the Community Land Model (CLM). We use meteorological forcing from the Global Land Data Assimilation System (GLDAS) to drive the CLM and apply a snow density‐based observation operator. This new operator is able to fit the observed seasonally varying relationship between the snow cover fraction and the snow depth. Surface measurements from Canada and the Advanced Microwave Scanning Radiometer‐Earth Observing System (AMSR‐E) estimates (in particular regions) are used to evaluate the assimilation results. The filter performance, including its ensemble statistics in different landscapes and climatic zones, is interpreted. Compared to the open loop, the EnKF method more accurately simulates the seasonal variability of SWE and reduces the uncertainties in the ensemble spread. Different simulations are also compared with spatially distributed climatological statistics from a re‐gridded data set, which shows that the SWE estimates from the EnKF are most improved in the mountainous west, the northern Great Plains, and the west and east coast regions. Limitations of the assimilation system are analyzed, and the domain‐wide innovation mean and normalized innovation variance are assessed, yielding valuable insights (e.g., about the misrepresentation of filter parameters) as to implementing the EnKF method for large‐scale snow properties estimation.

Streamflow response to seasonal snow cover mass changes over large Siberian watersheds

We used remotely sensed weekly snow water equivalent (SWE) data (1988–2000) to investigate streamflow response to seasonal snow cover change in the large Siberian watersheds (the Ob, Yenisei, and Lena basins). We quantified the seasonal cycles and variations of snow cover mass and river streamflow and identified a clear correspondence of river discharge to seasonal snow cover mass change. We also examined and compared the weekly mean streamflow with the weekly basin SWE for the study period. The results revealed a strong relation between the streamflow and snow cover mass change during the spring melt season over the large Siberian watersheds. This relationship provides a practical procedure of using remotely sensed snow cover information for snowmelt runoff estimation over the large northern watersheds. Analyses of extreme (high/low) SWE cases (years) and the associated streamflow conditions indicate an association of high (low) flood peak with high (low) maximum SWE in the Ob and Yenisei basins. Comparative analyses of weekly basin SWE data versus snow cover extent (SCE), peak snowmelt floods, and climatic variables (temperature and winter precipitation) indicate consistency among basin SWE, SCE, and temperature but incompatibility between basin SWE and winter precipitation, particularly for the Lena watershed. The inconsistency suggests uncertainties in determination of basin winter snowfall amounts and limitations in applications of the SWE retrieval algorithm over large watersheds/regions with very different physical characteristics. Overall, the results of this study clearly demonstrate that the weekly SWE data/products derived from microwave remote sensing technology are useful in understanding seasonal streamflow changes in the arctic regions.

Snow cover variability across central Canada (1978–2002) derived from satellite passive microwave data

Twenty-four winter seasons (1978-2002) of mean February snow water equivalent (SWE) values were analyzed in an exploration of the spatial pattern of temporal variability in snow cover across the non-mountainous interior of Canada. The SWE data were derived from space-borne passive microwave brightness temperatures processed with a land cover- sensitive suite of algorithms. Spatial patterns in the frequency and amount of variability were investigated on an annual basis through comparisons with average trends over all 24 years. Changes in temporal variability through time were also investigated by comparing three eight year time periods to general trends. Analyses were synthesized at the ecozone scale in order to link results both to potential land cover influences on algorithm performance and climatological variability in SWE. Prairie and northern ecozones were typically found to be the most variable in terms of SWE magnitude. Analyses indicate that non-treed land cover classes are generally more variable than treed classes. The results also indicate that extreme weather events appear to be occurring with in- creasing consistency in the Prairie and Arctic regions. Discerning climatologically significant variability in the time series, compared to algorithm-related issues can be a challenge, but in an era of eroding surface observing networks the passive microwave time series represents an important resource for monitoring and detecting trends and variability in terrestrial snow cover.

Comparison of seasonal and interannual variability of snow cover in Russian watersheds according to observations and reanalyses

Seasonal and interannual variability of the water equivalent of snow over the period 1966–1996 is analyzed with the use of the ERA/ECMWF reanalysis data, two versions of the NCEP reanalysis, and ground-based instrumental observations. Regions have been determined where snow characteristics from various sources are most consistent. In general, the snow water equivalent data from the ERA reanalysis are closer to the snow survey data than the NCEP reanalysis data. The best consistency of the water equivalent of snow in the reanalyses with measured data is observed in the period when the snow cover begins to form, in November and December, and the largest discrepancies are found for the snowmelt period, April and May. The interannual variability of the water equivalent in Asian Russia in the 1990s is better reproduced by the NCEP/DOE-II reanalysis. The snow cover data from satellite observations are compared with ground-based instrumental observations. It is shown that large discrepancies between the snow surveys and satellite observations may be observed at the snow-cover boundary during periods of snow formation and snowmelt.

Lagged relationships between North American snow mass and atmospheric teleconnection indices

AbstractRelationships between North American winter (January, February, March or JFM) snow mass, or snow water equivalent (SWE), between 1980 and 1997, and four teleconnection indices are explored at different spatial and temporal scales, with teleconnection indices leading SWE by one to two seasons. Summer (July, August, September, or JAS) and fall (October, November, December, or OND) Pacific North American pattern (PNA), North Atlantic Oscillation (NAO), El‐Nino Southern Oscillation (ENSO), and the Pacific Decadal Oscillation (PDO) are included in this analysis. Principal components analysis of the SWE data set results in four components, explaining over 56% of the variance in the SWE signal which have significant relationships to ENSO, PDO, and NAO. Strong spatial components associated with these relationships emerge, with the first component (NAO, ENSO) located in the northcentral to northwestern regions of the United States and the southcentral to southwestern regions of Canada. A third component (PDO) stretches from the midwest to the east coast of the United States, New England, and the Atlantic Provinces. Ranked correlation analyses using the SWE data set, and additional analyses of station observations in order to extend the time domain, corroborate and elucidate the PCA results. Examinations of these relationships at different spatial scales, and over varying time domains, indicate that there may be some scale‐dependant predictive ability for North American snow mass. Copyright © 2006 Royal Meteorological Society.

Oceanic‐atmospheric variability and western U.S. snowfall

A study of the influences of interdecadal and interannual oceanic‐atmospheric influences on April 1 Snow‐Water Equivalent (SWE) in the western U.S. is presented. SWE data was identified at 323 Natural Resources Conservation Service (NRCS) SNOTEL (SNOwpack TELemetrysites) stations for the period of 1961 to 2004 and for 121 SNOTEL stations for the period 1941 to 2004. The phases (cold/negative or warm/positive) of Pacific Ocean [El Niño ‐Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO)] and Atlantic Ocean [Atlantic Multidecadal Oscillation (AMO) and North Atlantic Oscillation (NAO)] oceanic‐atmospheric influences were identified for the year prior to the SWE data set. Statistical significance testing of SWE data set, based on the interdecadal and interannual oceanic‐atmospheric phase (warm/positive or cold/negative) was performed by applying the nonparametric rank‐sum test. The results show that in addition to the well established ENSO signal in the northwest, the PDO and AMO influence SWE variability. Additionally, the coupled effects of the oceanic‐atmospheric influences were evaluated on the basis of the long‐term phase (cold/negative or warm/positive) of the interdecadal (PDO, AMO, NAO) influences and the interannual ENSO. Finally, the coupled effects of the oceanic‐atmospheric influences were evaluated on the basis of the long‐term phase (cold/negative or warm/positive) of the interdecadal (AMO, PDO, NAO) phenomena. Regions in the west were identified that responded to the interdecadal/decadal climatic coupling. By utilizing the April 1 SWE and the long lead‐time approach for the oceanic‐atmospheric variables, useful information can be provided to snow forecasters and water managers.

Testing a distributed snowpack simulation model against spatial observations

Summary A snowpack simulation model was developed for high-resolution spatial prediction of wildlife habitat over large areas (100–1000 km 2 ). The model simulates a single layer water and energy balance, based on the Utah Energy Balance (UEB) model. It is driven by daily precipitation and temperature data, and 28.5-m maps of mean annual precipitation, terrain, vegetation, and geothermal heat flux. Parameters were calibrated against daily snow water equivalent (SWE) data at SNOTEL sites. The model was tested against spatial SWE data collected throughout the landscape, as well as snowpack temperature data, and photographic data. Spatial SWE data were supplied as a set of 40 high-accuracy means for different landscape strata, based on 1058 snow cores. At the time of peak snowpack accumulation in 2002, differences between modeled and measured SWE at the pixel scale were apportioned into three potential components: field sampling error (∼22%), parameter mapping error (∼27%), and actual model functional error (∼51%). Since the present study controlled for sampling and mapping error, clear priorities could be identified for reducing model functional error, including improvement of sub-models for snow interception, and snowpack thermal dynamics. The utility and detectability of improvements along these lines is contingent on controlling and quantifying both sampling and mapping error. The model strikes a unique balance between physical detail and spatial resolution and extent that is well suited for use in wildlife studies.

Parameterisation and application of a hillslope hydrological model to assess impacts of a forest clear-cutting on runoff generation

This paper combines results of an empirical paired-catchment method and hydrological simulation techniques in quantifying and explaining the effect of a clear-cutting on runoff generation. The pair of catchments (56 and 24 ha) is located in the middle boreal zone in Eastern Finland. In the first part, measured runoff from the two catchments, of which one had been partly clear-cut (35%) and the other had not been treated (control), was analysed to detect differences in runoff preceding and following the cutting. In the next part, snow and canopy models were calibrated against snow water equivalent data in an open and forested environment, respectively. Output from the canopy and snow models provided an input to a hillslope hydrological model, which was parameterised for the two catchments and calibrated against runoff measurements. The hydrological model was applied to identify mechanisms that could explain the observed difference in the hydrological behaviour of the two catchments. According to the model results, both the decreased interception in the clear-cut catchment and the increased evapotranspiration in the young, growing forest in the control catchment contributed to the observed differences in spring flood volumes. During the growing season, most of the change in runoff generation between the catchments was attributed to the increased evapotranspiration in the young forest.

Evaluation of passive microwave snow water equivalent retrievals across the boreal forest/tundra transition of western Canada

Composite patterns of monthly averaged (1978–2002) satellite passive microwave derived snow water equivalent (SWE) data produced using the Meteorological Service of Canada land cover sensitive algorithm suite contain an interannually consistent and well-defined zone of high SWE retrievals (> 100 mm) across the northern boreal forest of western Canada. Consistently lower SWE retrievals are present across both the comparatively dense boreal forest to the south, and open tundra to the north. Because of the sparse conventional observing network across this region, a dedicated field sampling campaign was conducted during the 2003/2004 winter season to measure snow properties across these landscape zones in northern Manitoba, Canada. Using road and helicopter access, snow cover measurements were made along an approximately 500 km transect between the communities of Thompson (boreal forest), Gillam (sparse northern boreal forest), and Churchill (open tundra) during late November 2003, and early March 2004. The November 2003 snow survey revealed good agreement between Special Sensor Microwave/Imager (SSM/I) derived SWE retrievals and in situ measurements across the gradient of increasing SWE between Thompson and Gillam, and the well defined zone of high SWE values to the north and east of Gillam. Further to the north, SSM/I derived SWE retrievals over the boreal ecotone and open tundra were consistently lower than the in situ measurements. Results of the March survey (at which point mean snow depths had increased from approximately 30–60 cm) showed an increase in absolute SWE differences between the in situ measurements and passive microwave estimates, but relative agreement between the datasets remained strong. The same SWE gradient across the boreal forest was again captured by both datasets, although tundra retrievals remained consistently low. These findings, along with consideration of snowpack stratigraphy, density, and the potential influence of tundra lakes and ponds on microwave emission and scatter, suggest that development of a tundra-specific SWE retrieval algorithm is necessary. From a scaling perspective, a large range of SWE values were measured on the ground within each satellite footprint, but a comparison of frequency histograms showed that a high percentage of satellite retrievals (November: 93%; March: 63%) fell within ± 20 mm of the median ground measurement.