Millimeter/Submillimeter-wave Ice Cloud Radiometer: Model And Retrieval Algorithm

Remoting sensing of ice clouds by millimeter/submillimeter-wave radiometer has received increasing attention in recent years,as ice clouds play a key role in the earth’s energy budget.This paper presents our study on the radiative transfer model and retrieval algorithm of the ice cloud radiometer.It consists of three parts,which are the radiative transfer model,the statistical retrieval algorithm,and the optimization retrieval algorithm.The first part of the paper sets up a radiative transfer model for the ice cloud radiometer.We first introduce the theoretical background of the radiative transfer,which includes the radiative transfer equation,the black-body radiation theory,and the absorption efficiency.Subsequently,we present the approach of calculating the cloud particles’ scattering properties.We introduce the Mie algorithm and the Discrete Dipole Approximation algorithm,and we use these two algorithms to simulate the scattering properties of several cloud particles.We also introduce the calculating method for the size distribution function.Finally,we set up a radiative transfer model based on ARTS.We show the detailed configuration and discuss the simulation results.The second part of the paper applied the Bayesian Monte Carlo statistical retrieval algorithm to retrieve the water path parameter.Bayesian Monte Carlo uses the prior information as the regularization to solve this ill-conditioned problem,so the retrieval results become more robust.The prior information is introduced by the way of the prior database.There are two steps to generate the prior database: creating the random atmosphere and cloud properties,and computing the simulated brightness temperatures.The atmospheric parameters are generated from the prior information.We introduce the Bayesian Monte Carlo theory,and we perform the retrieval simulation experiment to determine the accuracy of the retrieval results.The third part of the paper presents the ensemble optimization retrieval algorithm.This algorithm is proposed by the author of this paper to solve the hydrometeor profiles retrieval problem.The ensemble optimization algorithm incorporates the optimization into the Bayesian Monte Carlo framework.The algorithm doesn’t need the gradient information,so it avoids the complexity of the gradient radiative transfer calculations.The algorithm can effectively minimize the cost function,and it can derive the retrieval uncertainties automatically.The algorithm represents the unknown continuous posterior pdf by an ensemble of discrete cases,and it iteratively decreases the estimate uncertainties by generating a new ensemble from the learned distribution.The optimization algorithm includes two procedures: the estimation procedure numerically estimates the posterior pdf using the discrete cases in the last ensemble,and the sampling procedure samples this pdf to generate a new ensemble.These two procedures are carried out in eigenspace,due to the high interdependencies between variables in the parameter state vector.Once the termination criterion is met,retrieval results and uncertainties are derived by integration over the final ensemble.We introduce the theory of the ensemble algorithm,and we perform the retrieval simulation experiments.We present the retrieval results and conduct the statistical analysis.This paper presents our study on the radiative transfer model and the retrieval algorithm of the millimeter/submillimeter-wave ice cloud radiometer,and it also proposes a new retrieval algorithm for retrieving the hydrometeor profiles.The remote sensing of ice clouds is receiving increasing attention,and we wish this work could make some contributions to the ice clouds study.