Spatial and temporal variability in the annual cycle of Northern Hemisphere snow extent, using a newly completed climate data record

Motivated by recent monthly extremes in Northern Hemisphere (NH) winter snow seasons, both spatial and temporal variability in extent are explored using a newly completed snow cover extent (SCE) climate data record (CDR). The recent winter season of 2009/10 displayed swings from most to least extensive SCEs occurring in adjacent months. Recent snow literature has heralded reports ranging from earlier spring melt, to an overall seasonal decrease in SCE. From a hydrologic standpoint, such shifts have huge implications to the overall climatology of any region with snow present in its seasonal cycle.

The SCE CDR utilized in this study is the result of a reanalysis of NOAA satellite-derived maps of NH continental SCE and begins in 1966. The earlier 33 years of data are drawn from coarse-scale weekly NOAA maps, whilst recent years are based on a downgraded version of the IMS 24 km NOAA product, currently produced by the National Ice Center. The Rutgers Global Snow Lab undertook reanalysis and merging of the different resolution datasets, as addressed in a companion paper at this meeting.

Principal component (PC) analysis was performed on the annual cycle of NH, and individual continental snow extent totals. Close to 75% of variance was explained by the first three PCs, in the case of NH and Eurasia snow extents. Snow composite analysis based on positive PC scores time series years displays greater snow extent, in excess of 4 x 10⁶ km², in the case of the NH alone. Preliminary assessment of associations with large-scale atmospheric patterns (e.g. surface air temperature, mid-tropospheric geopotential height), yielded patterns indicative of know teleconnection patterns such as the AO and NAO indices. Results from additional PC analysis performed spatially on the SCE CDR will be discussed.