The large-scale distribution of snow cover over Northern hemisphere lands has been a topic of increasing interest in recent years. This interest is in part driven by pronounced deficits in snow extent over northern hemisphere lands since 1987 compared to the 1972 to 1986 period. Earlier hemispheric charts from the beginning of the satellite era in 1966 through 1971 suggest snow extents closer to those of more recent years, however they are known to be inaccurate. A major reason for this problem was the early inexperience of analysts in distinguishing snow cover from clouds or even from snow-free ground during this period, compounded by the lower resolution of imagery during this era compared to the post 1972 period. An effort to reanalyze the 1966-71 imagery and produce a new set of snow charts is underway at Rutgers. This paper will describe the effort, present results from the period, and merge them with post 1972 observations to examine variations in snow extent over the past 32 years (fall 1966 - spring 1998).
Snow cover between 1966 and 1971 was charted at Rutgers using daily imagery that is available as gridded eastern and western hemisphere composites in visible and infrared wavelengths. Surface resolution of the imagery is about 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. Daily surface weather charts also provided information on cloud cover, precipitation and temperature.
Charting was done by a manual interpretation of the visible satellite imagery. The infrared imagery and other ancillary information were employed in many areas to confirm interpretations made from visible data. Snow was identified in the visible imagery by recognizing characteristic textured surface features and brightness. The weekly charts show snow boundaries on the last day that the surface in a given region is observed. The ancillary data alone were not used to chart cover where clouds or low solar illumination prohibited visible interpretation. This would have greatly differed from charting methodologies employed since 1972, sacrificing the consistency sought between the two periods. Once completed, the charts were digitized using the National Meteorological Center Primitive Equation grid. This is an 89 x 89 cell northern hemisphere grid having a polar stereographic projection. Cell resolution ranges from 16,000 km2 to 42,000 km2. Only cells interpreted to be at least fifty percent snow covered are considered snow covered.
Charting is presently completed and digitization about finished. Regional to hemispheric analyses are taking place, and will be available for the October preprint modification.