雨滴谱分布函数的选择:M-P和Gamma分布的对比研究

COMPARATIVE STUDY OF EXPONENTIAL AND GAMMA FUNCTIONAL FITS TO OBSERVED RAINDROP SIZE DISTRIBUTION

M-P分布和Gamma分布是常用的两种雨滴谱分布函数。利用PMS GBPP-100型雨滴谱仪2003年7-8月在沈阳观测的雨滴谱资料,采用阶矩法对两种分布函数进行拟合,对比分析两种分布函数对谱及数浓度、雨强和雷达反射率因子的拟合效果。结果表明,在降水较弱、小滴偏少时Gamma分布会低估小滴,而M-P分布会高估小滴;降水强时,两种分布均低估小滴。M-P和Gamma分布对数浓度、雨强和雷达反射率因子这些特征量的拟合效果,在降水较强时差异很小,在降水较弱时差异较大。Gamma分布的代表性更好。此外,还讨论了两种分布的参数和雨强的关系。

The Marshall-Palmer （M-P） exponential and the gamma functions are widely used to represent characteristics of observed raindrop size distributions. In order to compare the differences in the spectrum,number concentration, rain rate and radar reflectivity factor between the resulting M-P and gamma distribution functions ,we use the moment methods to determine parameters for the two fitted functions. Data used here were obtained from Shenyang city in summer in 2003 with PMS GBPP-100 probes. The analysis shows that if observations of raindrop size distributions are fit by using appropriate moments to determine parameters of the fitted functions,it makes a more perfect agreement between fits and observations than using the least squares methods. Differences among the concentrations,rain rate and radar reflectivity factor between the two fitted functions by using moment methods were insignificant for high rain rate, but significant for low rain rate with fewer small drops due to underestimate by the gamma functions and overestimate by the M-P functions.