Satellite-based high latitude snow volume trend, variability and contribution to sea level over 1989/2006

Snow volume change over the 1989/2006 period has been derived from Special Sensor Microwave/Imager (SSM/I) radiometric measurements for all land surfaces above 50 degrees N, except Greenland. The mean annual snow volumes over the whole study domain, Eurasia and North America are respectively equal to 3713 km(3), 2272 km(3) and 1441 km(3), for the Pan Arctic regions, over this 18-year time period. While the snow volume exhibits a statistically significant negative trend (-9.7 +/- 3.8 km(3) yr(-1), p-value =0.02) over North America, it presents a positive, but not statistically significant trend (11.3 +/- 9.3 km(3) yr(-1), p-value = 0.25) over Eurasia. These opposite variations can be related to different regional climatic conditions over these two regions: over Eurasia, snow depth is mainly influenced by the Arctic Oscillation (AO) and the Atlantic Multidecadal Oscillation (AMO)-correlation coefficient = 0.68 between the SSM/I-derived snow volume and a linear combination of AO and AMO indices, whereas over North America snow depth is mainly influenced by the Pacific North American (PNA) pattern and the AMO-correlation coefficient = 0.75 for a linear combination of the PNA and AMO indices. This study confirms that snow volume is a key driver of the sea level seasonal cycle, but net snow volume trend for the Pan Arctic regions indicates a negligible and not statistically significant contribution to sea level rise (-0.004 +/- 0.009 mm yr(-1) p-value = 0.88 once converted into sea level). (C) 2010 Elsevier B.V. All rights reserved.