Development And Application Of HECORA Cloud Retrieval Algorithm Based On Hyperspectral Satellite Data

With the acceleration of industrialization and urbanization process of China,the problem of atmospheric pollution has become increasingly prominent,it has gradually exceeded the atmospheric environment carrying capacity.Using satellite remote sensing technology to monitor the column density and spatial-temporal distribution of atmospheric trace gas,and analyze the cross-regional transmission characteristics of pollutants has become an important method of atmospheric environment monitoring.The next generation atmospheric sounding instrument TROPOMI can implement the monitoring of atmospheric trace gas transmission between cities with high spatial resolution of 7kmx3.5km,and provide reliable data for analyzing the contribution of local resource and external resource to atmospheric pollution.However,when using satellite remote sensing technology for atmospheric monitoring,the atmospheric cloud has an important influence on the retrieval of trace gas due to the albedo effect,shielding effect and in-cloud absorption effect.In the ultraviolet and visible wavelength range,in order to obtain accurate column density and spatial-temporal distribution of atmospheric trace gas,it is necessary to retrieval the atmospheric cloud information at first.In this paper,the HECORA cloud retrieval algorithm based on the O₂-O₂477nm absorption band is developed to retrieve the effective cloud fraction and cloud pressure parameters to correct the effect of cloud on the trace gas retrieval.The construction of HECORA firstly uses the VLIDORT radiative transfer model to simulate 460-490nm reflectance spectrum in a set of scenarios.The DOAS method is used to fit and retrieve O₂-O₂ slant column density and continuum reflectance,and adopt the linear radial basis function to generate the effective cloud fraction and cloud pressure lookup tables.Then use QDOAS software to fit the measured spectrum of observation instrument onboard satellite to retrieve the actual O₂-O₂ slant column density and continuum reflectance,and combine processed auxiliary data to complete the retrieval of effective cloud fraction and cloud pressure.HECORA is validated using VLIDORT simulation spectrum.When using the Lambertian cloud model in the simulation,HECORA retrieved cloud pressure is consistent with real cloud pressure.For single-layer scattering cloud,HECORA cloud pressure is inside the real scattering cloud boundary,and for double-layer scattering cloud,HECORA cloud pressure is between the double-layer scattering clouds,the retrieval results are agreement with the actual situation.The HECORA is applied to TROPOMI,and the retrieval results are compared with FRESCO+ product that based on O₂ A absorption band.When the surface type is ocean,the correlation coefficients R of effective cloud fraction and cloud pressure from both algorithms are greater than 0.95,which has good correlation.Comparisons with on-board lidar CALIOP cloud layer products show that HECORA cloud pressure retrieval results are better than FRESCO+ in low cloud condition,while FRESCO+ cloud pressure is closer to CALIOP cloud pressure in high cloud condition.In the TROPOMI NO₂ retrieval,the effective cloud fraction and cloud pressure retrieved by HECORA are used.The TROPOMI tropospheric NO₂ vertical column density is validated with the ground-based MAX-DOAS tropospheric NO₂ vertical column density,the correlation coefficients R of the 4 observation sites are greater than 0.94,the correlation is good,which proves that effective cloud fraction and cloud pressure from HECORA can be used in the retrieval of regional NO₂ vertical column density and correct the cloud influences of cloud on TROPOMI NO₂ retrieval.The successful application of HECORA on TROPOMI shows that HECORA has high accuracy and reliability in atmospheric cloud retrieval,and can provide effective cloud fraction and cloud pressure for cloud correction of atmospheric trace gas retrieval.The more import is that HECORA could provide reference for the development of cloud retrieval algorithm applied to the same type of satellite payload of China.