SACOL站冰云微物理特性的反演

Retrieval of ice cloud microphysical properties at the SACOL

冰云是全球分布范围最广的云类型之一,对地气系统的辐射收支有重要作用.本文利用兰州大学半干旱气候与环境监测站(SACOL)Ka波段毫米波云雷达(KAZR)2013年8月至2014年7月的观测数据,结合微脉冲激光雷达(MPL)观测数据,分别使用传统经验算法、考虑温度订正的反演算法和KAZR-MPL联合反演算法,反演了冰云的冰水含量和有效粒径两个微物理量,形成了云微物理量的反演数据产品.在此基础上对SACOL站上空云微观特性进行了统计分析,并进一步结合Fu-Liou辐射传输模式,模拟了冰云在大气层顶处对长短波辐射通量的影响,与CERES卫星观测的辐射通量值进行比较,用以评估不同冰云微物理量的差异对辐射所造成的影响.结果发现:SACOL站冰云主要出现在2~14 km的范围内,在7 km处出现频率最高,达到13.02%;3种云微物理方法反演的冰水含量相差不大,均呈现单峰型分布,概率密度最大处的冰水含量集中在10-3g/m^3左右,而有效粒径的谱分布呈现出较大差异,KAZR-MPL反演算法模拟的冰晶粒径普遍偏大,也正是由于这种差异,造成了对大气层顶的短波辐射通量模拟的偏差.而各种算法对于大气层顶出射的长波辐射模拟效果远好于对短波的模拟,平均误差为5.04%,相关系数均大于0.6,而模拟效果较差的部分,主要存在于光学厚度较薄的个例当中.

Ice clouds are one of the most widely distributed cloud types in the world and play an important role in the radiation budget.The microphysical parameters of ice clouds,such as ice water content,effective radius and particle concentration,have a significant effect on the artificial influences on weather,the calculation of the radiation effect and climate feedback of ice clouds.Because of their unique physical and optical properties,ice clouds play an important and unique role in the radiation budget.Understanding and recognizing the microphysical properties of ice clouds are of great significance for studying the effects of ice clouds on climate change.Ka-band zenith radar(KAZR),which has been deployed at the Semi-Arid Climate and Environment Observatory of Lanzhou University(SACOL)since July 2013,has continuously operated for many years.By using the observation data from KAZR and micropulse lidar(MPL),we compare the microphysical properties of ice clouds,including ice effective particle size(Dge)and ice water content(IWC),which were derived from three algorithms:a traditional algorithm(Microbase),the Z-IWC-T relationship algorithm(Hogan)and the KAZR-MPL retrieval algorithm(Combined).Then,we compute and examine the radiative fluxes using each set of cloud properties as inputs to a radiative transfer model and compare the top-of-atmosphere(TOA)radiative fluxes to the satellite measurements to assess the impact of the differences in the microphysical quantities of different ice clouds on radiation.Our analysis shows that ice clouds mainly appear from 2 to 14 km and peak by approximately 13.02%at 7 km.The ice water contents retrieved from the three different inversion methods show a unimodal distribution.The ice water content is mainly concentrated in the range of 0.01-10 mg/m^3,and the corresponding maximum probability of ice water content is concentrated in the vicinity of 0.1 mg/m^3.However,there are some differences between the degrees of data concentration and the peak probabilities of the different algorithms.The effective particle size distributions obtained by the three methods are quite different.The effective particle size of the ice crystal inversion by the Microbase algorithm is unimodal,and the data are mainly concentrated in the small particle radius region of 0-100μm.The KAZR-MPL algorithm has a wider range of particle distributions,with two peaks at 40 and 96μm.However,the distribution of the ice effective particle radius shows significant discrepancies,which sometimes translate into large differences in the cloud shortwave radiative effect.Using the results of three different microphysical properties,the longwave and shortwave radiation fluxes at the top of the atmosphere are calculated through the Fu-Liou radiation transfer model.The results show that the longwave radiation fluxes calculated by using different cloud microphysical characteristics are in good agreement with the satellite observation data.The average error is 5.04%,and the correlation coefficients are all greater than 0.6.Compared with the observed longwave radiation,the Hogan algorithm that considers the temperature effect has a better simulation effect,the correlation coefficient is larger,and the root mean square error is smaller.The shortwave radiation is smaller than the observed values in the case of clouds,and the maximum error reaches 24.61%.The shortwave radiation flux simulated by the traditional empirical algorithm is closest to the satellite observations,and its root mean square error is significantly reduced compared with that of the other methods.