论文部分内容阅读
Eurasian snow cover fraction (SCF) prediction is analyzed using the recently developed NCEP Climate Forecast System, version 2 (CFSv2) monthly retrospective forecasts for 1983-2009.Generally, the CFSv2 is capable of reproducing the observed Eurasian SCF seasonal cycle and climatology except some systematic biases.This study focuses on the prediction skill and predictability of Eurasian SCF in snowmelt and snowfall seasons since the most robust variability occurs in these months.The CFSv2 reasonably predicts the interannual variations, long-term trend and leading EOF pattern in snowmelt season several months ahead.Compared with the prediction skill in snowmelt season, although the CFSv2 shows a slightly better prediction skill in climatological values, it underestimates the observed interannual variability in snowfall season.Additionally, The CFSv2 predicts a downward trend whereas the observations exhibit an upward trend in snowfall season.The systematic biases, the larger climatology and smaller interannual variations in both snowmelt and snowfall seasons, might be the result that the CFSv2 predicts the cooler atmosphere in the low to middle troposphere and more precipitation and weaker variability in them compared with the observations.It is interesting that the prediction skill is higher and more stable after late-1990s than before in the two seasons.It is possibly caused by the changesbothin the initial conditionin the CFSv2 and the observed SCF in late-1990s.