Historical gridded snow water equivalent over the Northern Hemisphere from remote sensing and land surface models
ECCC has produced observation-based estimates of snow water equivalent (SWE) over the Northern Hemisphere (excluding Greenland). The data includes daily and monthly versions of SWE and covers the time period 1981-2020. Technical notes on the methods are described below.
On this page
- Download datasets
- Technical notes
- Source datasets
- Preprocessing
- Spatial interpolation
- Daily snow water equivalent
- Monthly snow water equivalent
- Use limitation
- Contact information
- References
Download datasets
Technical notes
ECCC has produced datasets of daily and monthly SWE over the Northern Hemisphere (1981-2020) using a multi-dataset approach. Users can also find datasets for Canada (1981-2016) based on an older generation of source datasets. These datasets include monthly snow cover fraction (SCF), monthly SWE, and annual maximum SWE (SWEmax).
The general method to create these datasets follows that of Mudryk et al. (2015).Reference4
Table 1. Key features of the datasets.
- Variables
-
Daily SWE (millimetre water equivalent)
Monthly SWE (millimetre water equivalent)
- Spatial resolution and geographical coverage
0.5- by 0.5-degree grid across the Northern Hemisphere (excluding Greenland)
- Time period
1981-2020
- Temporal resolution
Daily, monthly
Source Datasets
Daily SWE data is based on four sources for the 1981 to 2020 period.
Four sources of daily SWE data:
- Modern-Era Retrospective Analysis for Research and Applications Version 2 (MERRA-2).Reference2 MERRA-2 is a National Aeronautics and Space Administration (NASA) atmospheric reanalysis product. It was generated with the Goddard Earth Observing System Model, Version 5.2.0 (GEOS-5), atmospheric general circulation model, and atmospheric data assimilation system (ADAS)
- Fifth Generation European Centre for Medium-Range Weather Forecasts (ECMWF) Land reanalysis (ERA5-Land)Reference5
- The physical snowpack model Crocus-ERA5 simulates daily SWE using meteorology from the standard ERA5 reanalysisReference1
- The Snow CCI + SWE Version 2 (CCIv2) data productReference3, estimates terrestrial SWE using a combination of passive-microwave brightness temperatures and in situ snow depth measurements
Preprocessing
The CCIv2 product contains missing days during the snow season, particularly prior to 1988. ECCC linearly interpolated missing days from surrounding days with data outside of the summer season.
Each year, the CCIv2 product also lacks data for some interval between May and September. To avoid creating issues in the blended product, ECCC used exponential interpolation. This method mimics spring melts and fall snow buildup from the last spring, and first fall, dates with data.
Spatial interpolation
ECCC interpolated daily SWE data over the Northern Hemisphere from the four products to a common 0.5° x 0.5° grid. The process used the nearest neighbor remapping function (‘remapnn’) from Climate Data Operators (CDO) (version 2.0.3).
Daily snow water equivalent
ECCC removed SWE values greater than 2000 mm from the component products to focus on seasonal snow.
The four source datasets were weighted equally in the daily SWE mean in almost all cases with two exceptions.
ECCC partially masked the CCIv2 product over mountainous regions. Mountainous regions were defined as areas with slopes of 2 degrees or larger. A blend of CCIv2 SWE data (if any) and the mean from the other three data sources replaced grid cells containing mountains.
The blend’s weighting depended on the fraction of the grid cell area that was mountainous. For grid cells with no mountainous terrain, the original CCIv2 data were used. As the fraction of mountainous terrain increased, the weight applied to the CCIv2 data was linearly reduced. It reached zero for grid cells containing only mountainous terrain.
The CCIv2 product contains no data past May 24th, 2020. As such, the remainder of 2020 is a mean of the three remaining data sources.
Monthly snow water equivalent
ECCC produced monthly SWE by averaging the regridded daily SWE fields from each source over the given month.
Use limitation
Open Government Licence - Canada (http://open.canada.ca/en/open-government-licence-canada).
Individual model datasets and all related derived products are subject to the terms of use of the source organization.