Research On Retrieval Of Cloudy Atmospheric Parameters By Ground-Based Multi-Channel Radiometers

Ground-based meteorological multi-channel microwave radiometer has been widely used in atmospheric remote sensing.It can retrieve real-time atmospheric parameters such as temperature and humidity profile of atmosphere and liquid water path in clouds.It has the advantages of full-time,high temporal resolution and unattended.In order to obtain the atmospheric parameters accurately in various weather conditions,higher requirements are imposed on the inversion algorithm of cloud-sky atmospheric parameters.The inversion of cloud-sky atmospheric parameters based on ground-based microwave radiometer is studied in this thesis.Firstly,the theory of cloud-sky atmospheric microwave radiation is introduced and the theoretical brightness temperature is calculated.The scattering of hydrometers in clouds is considered in the microwave radiation transmission in cloudy atmosphere.Then,based on the forward modeling of the brightness temperature of atmospheric microwave radiation,the temperature and humidity profiles and liquid water paths of cloudy atmosphere are simulated and inverted by using LMBP neural network algorithm.The results show that the inversion error of cloud temperature profile is less than 1K at low altitude and about 3.5K at high altitude,while the inversion error of humidity is very small at high altitude and large at low altitude,and the maximum error of humidity at low altitude is 1.5g/m~3.The correlation coefficient between the inversion value of liquid water path and the actual value from sounding is 0.96.The influence of scattering on retrieval accuracy of cloud-sky atmospheric parameters is further analyzed in this thesis.The simulation shows that scattering has great influence on retrieval accuracy of humidity profile and liquid water path in clouds.Considering scattering in inversion algorithm can improve retrieval accuracy.Then,based on the LMBP neural network inversion algorithm,the temperature and humidity profiles of cloudy and sunny atmosphere are retrieved from the atmospheric brightness temperature measurement data of RPG-HATPRO ground-based microwave radiometer,and the inversion results are close to the real values of sounding balloon measurements.Finally,to improve the inversion effect,we proposed a method which adds oblique observation on the basis of radiometer zenith measurement.When angular observation information is added into the inversion algorithm,the inversion error of high altitude temperature decreases from 3.5K to 2K,and humidity inversion error at low altitude decreases from 1.5g/m~3 to 1.2g/m~3,which also has better inversion effect in clouds.