Study On The Retrievals Of Cloud Properties Based On Multi-Directional Polarimetric Data

Cloud is a visible aggregate of minute water droplets and/or ice particles suspended in the atmosphere which result from water vapor condensation.The radiative force of clouds exerts significant influence on radiative budget so it is important to acquire the optical and microphysical properties of clouds and temporal spatial variation correctly.Due to the extensive spatial coverage and continual observation,space-borne remote sensing serves as an effective way to observe clouds globally.The directional polarimetric camera(DPC),developed by Anhui Institute of Optics and Fine Mechanics,Chinese Academy of Sciences,launched in May 2008 onboard the GF-5 satellite subjected to Chinese high-resolution earth observation program,is the first domestic space-borne sensor with the operational capability of multi-spectral,multi-viewing and polarimetric observations.Studies on the inversion of cloud parameters makes great sense for a better understanding of the influence of cloud optical and microphysical properties on radiative budget,as well as the development of DPC operational cloud property products.Based on this purpose,several studies on how to retrieve cloud mask,cloud thermodynamic phase and cloud optical thickness correctly,which are essential parameters of cloud property characterization,are conducted.1.Study of cloud mask algorithm.The cloud mask algorithm serves to acquire the relative distribution of cloudy pixels and clear pixels,i.e.the cloud mask,which is the first step of further study.A cloud mask algorithm based on the reflectance at 670nm and 865nm and the polarized reflectance at 490nm is developed and validated.The cloud mask of the regions observed by DPC is obtained by applying the algorithm to DPC data.Then,the reliability of the cloud mask is validated through the comparison with MSG cloud mask product.2.Study of cloud thermodynamic phase algorithmCloud thermodynamic phase is an important property to influence the optical and microphysical properties of clouds which influence reginal radiative force and radiative budget moreover.A cloud thermodynamic phase algorithm based on the normalized polarized radiance at 865nm is developed and validated.The cloud thermodynamic phase of the regions observed by DPC is obtained by applying the algorithm to DPC data.Then,the reliability of the regional cloud thermodynamic phase is validated through the comparison with MODIS cloud thermodynamic phase product.3.Study of cloud optical thickness algorithmCloud optical thickness(COT),which indicates the extinction ability and depends on various intrinsic properties,is a crucial indicator of radiative force.The uncertainty of COT inversion is analyzed.A COT algorithm based on the Look up Table method including previous cloud mask result and thermodynamic phase result is developed and validated.The COT of the regions observed by DPC is obtained by applying the algorithm to DPC data.Then,the reliability of the COT is validated through the comparison with MODIS COT product.The studies made by this paper is expected to provide the development of the DPC operational cloud property products with algorithmic support.