GRAPES不同参数化方案对新疆“4.18”寒潮降雪的敏感试验

Sensitive Experiment of Various Parameterization Schemes in GRAPES Model on "4.18" Snowstorm Weather in Xinjiang

利用我国新一代数值预报模式GRAPES,选择Kessler、NCEP3-class simple ice和simple ice微物理过程参数化方案和Kain-Fritsch、Betts-Miller-Janjic积云参数化方案形成6个组合,对新疆"4.18"寒潮天气的降雪过程进行预报试验,并与实况资料进行比较、检验。结果表明,就降雪个例而言,预报效果相对好的参数化方案存在不确定性,积云参数化方案对降雪的影响比微物理过程参数化方案对降雪的影响大。NCEP 3-class simple ice与Kain-Fritsch方案组合对"4.18"寒潮天气降雪量的预报较好。

Based on the new generation numerical prediction model GRAPES, with three kinds of physics option paramcterization schemes （Kessler, NCEP3-class simple ice and simple ice） and two kinds of cumulus parameterization schemes （Kain-Fritsch and Betts-Mincr-Janjic）, six combinational schemes were formed,experiments were made for the snowstorm case on ＂4. 18＂ in Xinjiang. The comparison among the experiments and analyses of the observations indicated that it was uncertain which scheme was better for a certain ease and the predicting effect of cumulus parameterization schemes was better than that of physics option parameterlzation schemes. The predicting ability of ＂4.18＂ snowstorm of combinational schemes with NCEP 3-class simple ice physics option parumeterization schemes and Kain-Fritseh cumulus parametcrization schemes was superior to the others.