A 0～6 h Quantitative Snow(Rain) Forecast Technique and Its Application in Vancouver Winter Olympics

Aiming to reach the scientific goal of SNOWV10 project,a short-range(0～6 h) quantitative snow(including rain) forecast technique was developed by blending mesoscale numerical prediction with radar-based extrapolative forecast,in conjunction with radar estimated snowfall(and rainfall) approaches and automatic identification of precipitation phase(rain,snow or their mix).This technique was then integrated into the GRAPES-SWIFT nowcasting system so that venue forecast could be issued during the Vancouver Winter Olympics on a real-time basis.Results from Vancouver Olympics application in February 2010 show that this quantitative snow(rain) forecast technique,to a certain extent,could meet the requirement of winter snow(or rain) nowcast or warning,whereas the magnitude of rainfall forecast was much less than that of observation and the forecast timing for snow(or rain) initiation and ending was quite inconsistent with observations,which call for future improvement for both NWP models and extrapolation,as well as their blending technique.