RegCM4.6两种积云参数化方案在东亚模拟结果的评估

Evaluation of simulation results from two cumulus parameterization schemes in RegCM4.6 in East Asia

新一代区域气候模式RegCM4.6引进了Mix积云参数化方案,可以将之前版本中的Emanuel和Grell方案结合在一起,以弥补单个参数化方案的不足。利用2016年MODIS(Moderate-resolution imaging spectroradiometer)数据对RegCM4.6中Mix和Emanuel积云参数化方案模拟的东亚云量(Cloud fraction,CF)、冰水柱含量(Ice water path,IWP)和液水柱含量(Liquid water path,LWP)进行初步评估,计算了相关系数(r)、平均绝对误差(Mean absolute error,MAE)、平均偏差(Mean bias error,MBE)和均方根误差(Root mean square error,RMSE),以便为相关研究选取积云参数化方案提供参考依据。结果表明:(1)模拟的CF的MBE大致以胡焕庸线为界,西北部为轻微高估,东南部通常为低估。2种方案在夏季的模拟效果最好,冬季最差。Mix方案的MAE、MBE和RMSE的绝对值在四季普遍小于Emanuel方案。(2)模式明显低估了东亚的IWP,除夏季外,2种方案模拟的IWP与MODIS的都呈显著负相关,表明模式难以准确模拟出云中冰晶相关的物理过程。(3)2种方案模拟的LWP在青藏高原和东部海域均为低估,在中国南部、中部和北部为高估,但Mix方案的偏差更接近于0。冬季,2种方案的评估参数相近,其他季节Mix方案的MAE、MBE和RMSE的绝对值均小于Emanuel方案,其中MAE相差21~39 g·m-2。因此,Mix方案更适用于在东亚进行云水资源方面的模拟研究。

Previous studies have shown that the Emanuel scheme performs relatively well in simulating temperature and precipitation in East Asia.However,the user’s guide of RegCM4.6 points out that the Emanuel scheme tends to produce excessive precipitation over lands,especially in some intense individual precipitation events.In contrast,the Grell scheme tends to produce weak precipitation over tropical oceans.Therefore,the new version of the regional climate model RegCM4.6 has incorporated the Mix cumulus convective parameterization scheme,which means that the Emanuel scheme can be used over oceans and the Grell scheme over land,to compensate for the deficiencies of a single scheme.Previous validation studies have mainly focused on temperature and precipitation,and few studies have been conducted on the Mix scheme.The MODIS product from January 1st,2016,to December 31st,2016,was used as a reference to evaluate the simulation results of cloud fraction(CF),ice water path(IWP),and liquid water path(LWP)in East Asia from the Emanuel and Mix schemes in RegCM4.6 at various time scales.Some statistical parameters were calculated,such as the correlation coefficient(r),mean absolute error(MAE),mean bias error(MBE),and root mean square error(RMSE).The results were as follows.(1)The simulated CF were slightly overestimated in the northwest and mainly underestimated in the southeast roughly bounded by the Hu Huanyong line.The performance of the two schemes in simulating CF was the best in summer and the worst in winter.In the four seasons,the absolute values of MAE,MBE,and RMSE of the Mix scheme were generally lower than those of the Emanuel scheme.(2)The systematic deviations of IWP were negative in the whole of East Asia.Except in summer,the IWP from the two simulations and MODIS was significantly negatively correlated in the other three seasons,indicating that it was a challenge to accurately simulate physical processes related to ice particles in the cloud.(3)The LWP was underestimated by the two schemes in the Qinghai Tibet Plateau and Eastern Ocean and was overestimated in southern,central,and northern China,but the annual MBE of the Mix scheme were closer to 0.The performances of the two schemes were similar in winter.In the other three seasons,the absolute values of MAE,MBE,and RMSE of the Mix scheme were less than those of the Emanuel scheme,and the differences in MAE for the two schemes were 21-39 g·m-2.In conclusion,the Mix scheme is more suitable to simulate cloud water resources in East Asia.This study will contribute to the exploitation of cloud water resources in East Asia and provide a reference for the selection and improvement of the cumulus convection parameterization scheme in a regional climate model.