In 1966, NOAA began to map the snow and ice areas in the Northern Hemisphere on a weekly basis. That effort continues today, and remains the only such hemispheric product. NOAA maps are based on a visual interpretation of photographic copies of shortwave imagery by trained meteorologists. Up to 1972, the subpoint resolution of the meteorological satellites commonly used was around 4 km. Beginning in October 1972, the Very High Resolution Radiometer (VHRR) provided imagery with a spatial resolution of 1.0 km, which in November 1978, with the launching of the Advanced VHRR (AVHRR), was reduced slightly to 1.1 km. Maps show boundaries on the last day that the surface in a given region is seen. Since May 1982, dates when a region was last observed have been placed on the maps. An examination of these dates shows the maps to be most representative of the fifth day of the week.
It is recognized that in early years the snow extent was underestimated on the NOAA maps, especially during Fall. Mapping improved considerably in 1972 with the deployment of the VHRR sensor, and since then mapping accuracy is such that this product is considered suitable for continental-scale climate studies.
Despite the shortwave limitations mentioned earlier, the NOAA maps are quite reliable at many times and in many regions. These include regions where, 1) skies are frequently clear, commonly in Spring near the snowline, 2) solar zenith angles are relatively low and illumination is high, 3) the snow cover is reasonably stable or changes slowly, and 4) pronounced local and regional signatures are present owing to the distribution of vegetation, lakes and rivers. Under these conditions, the satellite-derived product will be superior to maps of snow extent gleaned from station data, particularly in mountainous and sparsely inhabited regions. Another advantage of the NOAA snow maps is their portrayal of regionally-representative snow extent, whereas maps based on ground station reports may be biased, due to the preferred position of weather stations in valleys and in places affected by urban heat islands, such as airports.
The NOAA maps were digitized on a weekly basis using an extension of the National Meteorological Center Limited Area Fine-Mesh grid. This is an 89 x 89 cell Northern Hemisphere grid, with cell resolution ranging from ~10,700 sq. km to ~41,800 sq. km. If a cell was interpreted to be at least fifty percent snow covered it was considered to be completely covered, otherwise it was considered to be snowfree.Data Preparation
The following section contains information about the processing applied to the NOAA visible satellite snow charts at the GSL.
During the development of a routine to improve the accuracy of monthly snow areas from the NOAA data, an inconsistency in the cells which NOAA considered to be land, thus potentially snow covered, was discovered. In 1981, NOAA changed their land mask, in the process eliminating 26 cells from consideration of being snow covered (categorizing them as water), while 27 others began to be examined. These grid cells encompass some 1.8 x 106 sq. km, which could amount to a several percent or greater inaccuracy in snow estimates. Neither of the NOAA masks is accurate; both fail to accurately identify all cells, and only those cells, at least half covered by land. Therefore, a definitive land mask has been developed using digital map files analyzed on a geographic information system. The percentage of land in each of the 7921 NMC grid cells was calculated using the National Geophysical Data Center's five minute resolution ETOPO5 file as the primary data source. As this file does not include large interior lakes, the Navy Fleet Numerical Oceanography Center's 10 minute resolution Primary Terrain Cover Types file was used to properly account for these water bodies. Some 48 cells polewards of approximately 30° N which had been considered land in the pre 1981 NOAA and/or the 1981 to present NOAA mask are actually predominantly water covered (< 50% land). Conversely, 54 land cells were found to have been considered water on one or both NOAA masks. Those cells falling under the latter required a first-time analysis to determine whether they might be snow covered. This was accomplished by selecting nearest representative land cells (cells which NOAA has continuously mapped as land) and assigning their snow status to the "new" land cells. Spot checks of a number of hard copy weekly maps proved this to be an adequate approach.
Snow cover between 1966 and 1971 was reanalyzed here at the Rutgers University Climate Lab using daily gridded composites of visible imagery for the eastern and western hemispheres of the Northern Hemisphere. Surface resolution of the imagery is approximately 25 km. The imagery was supplemented with daily reports of snow depth at several thousand stations in the U.S., Canada, China and the former Soviet Union, gridded to 1° x 1° grid cells using all reports from within a given cell. Daily surface weather charts also provided information on cloud cover, precipitation and temperature. Infrared imagery and the above ancillary information were employed in many areas to confirm interpretations made from visible data. The weekly maps were digitized using an extension of the National Meteorological Center Limited Area Fine-Mesh grid. This is an 89 x 89 cell Cartesian grid laid over a polar stereographic projection of the Northern Hemisphere. Cell resolution ranges from ~10,700 to ~41,800 square kilometers. Each grid cell in the digitized product has a binary value. Cells with at least 50% of their surface covered with snow were considered snow covered. All other cells were considered snow free.
In June 1999 NOAA ceased production of weekly snow maps. IMS daily snow charts are now produced at NIC and utilized at Rutgers to create a unique Northern Hemisphere snow cover product. Using Perl software created at the GSL, weekly and monthly 89 x 89 grid cell charts are generated. In this procedure, weekly areas are calculated from digitized snow files, and monthly values are calculated by weighting the weekly areas according to the number of days of a map week falling in the given month. During this process, the 24 km resolution IMS daily snow product is reduced and subjected to filtering through the corrected land mask created here at the GSL. The result is a Climate Data Record (CDR) that details Northern Hemisphere snow cover over the last 49 years.