An Improved Cloud Mask Algorithm For KAZR And Physical Properties At The SACOL

Clouds are one of the most important components in the climate system through their profound effects on the radiative energy budget.Cloud Radar is the powerful instrument to research cloud.A new method is proposed to distinguish radar echoes from noise based on Ka-Band Zenith Radar(KAZR)which contain noise reduction scheme in order to recognize more weak signal clouds.Macrophysical properties of clouds above the SACOL site also analyzed using KAZR observations.(1)Based on an image noise reduction technique,we propose a new method to detect hydrometeors from cloud radar return signals.The basic idea is to treat the SNR value of each range gate as a pixel brightness.The SNR distributions of noise are suppressed to a very narrow range by convolving with a 2-D Gaussian kernel.After the noise smoothing process,those range gates with the SNRs larger than a threshold value are accepted as hydrometeors signals.The false positive and false negative of this method is 0.12%,2.52%.(2)A new method is proposed to distinguish clouds and other hydrometeors from noise in cloud radar observations.A noise reduction scheme that can reduce the noise distribution to a narrow range is proposed in our method in order to recognize more weak signal clouds.A spatial filter with central weighting,which is used in current cloud radar hydrometeor detection algorithms,is also involved in our method to examine radar return for significant levels of signals.“Square clouds” were constructed to test the two schemes,the false positive for strong,moderate and weak signals are 0.048%,0.103%,0.007% and the false negative are 0.244%,0.229%,9.774%.This method has a significant improvement compared with operational algorithm of the U.S.Department of Energy(DOE)Atmospheric Radiation Measurements(ARM).(3)We analyze the macrophysical properties of clouds above the SACOL site using the KAZR observations through 2014.It is shown that the occurrence of cloud base peaks at both 1.5 and 5.5 km while the cloud top peaks at 2.5 and 8.5 km.About 67% of the cloud thickness is within 2 km.The monthly mean occurrence of all clouds varies between 44% and 76%.The occurrence of low,middle and high clouds is 13%,30%,34%,respectively.Clouds can also be classified based on the number of cloud layers.Clouds with layers less than three contribute to ~98% of the total clouds and multi-layer clouds occur frequently in summer and autumn.(4)The cloud base and top heights detected by KAZR,CloudSat and CALIPSO are also compared in this study.The reflectivities detected by CloudSat are comparable with those from KAZR.CloudSat cannot detect clouds with weak signal less than-30 dBZ due to the limitation of its sensitivity.This part of missing clouds can be captured by CALIPSO but CALIPSO lidar signals can be attenuated in thick clouds leading to an overestimation of cloud base.(5)The terminal velocity of cloud is an important factor in the model that can be calculated through cloud radar.The algorithm of the terminal velocity of cloud droplet as well as the vertical velocity through KAZR is introduced.