Comparison Of Atmospheric Correction Models Of Lunar Ground-based Observation Data

The moon is the only natural satellite of the earth,and the nearest celestial body.Because of its special space environment and huge strategic value,it has an extremely important status in the process of deep space exploration.The radiation characteristics of the moon are stable,which can not only serve as the natural radiation calibration source of the orbit sensors,but also can play an important role in the quantitative observation of the foundation atmosphere at night.In addition,the study of the radiation nature of the lunar surface understanding contributes to the formation and evolution of the moon.Ground-based observation is playing a more and more important role with its unique advantages.However,the existence of the surface atmosphere will affect the results of the ground-based observation.Therefore,the atmospheric correction of ground-based observed data is particularly important.In numerous atmospheric correction methods,the radiative transfer model has clear physical meaning and high accuracy,and plays an important role in atmospheric radiation properties,atmospheric parameter inversion and atmospheric correction of remote sensing images.This paper uses LBLRTM and MODTRAN radiative transfer model to simulate atmospheric transmittance,upwelling and downwelling radiation,studies the radiative transfer properties of the atmosphere,summarize the atmospheric radiative transfer rules,explores the effects of the observation zenith angle and the height of observation points on the atmospheric radiative transport properties,and compare the difference between LBLRTM and MODTRAN simulation results.The change trend and characteristic position of LBLRTM and MODRAN for the simulation results of atmospheric radiation transmission are the same.However,there are some deviations between LBLRTM and MODTRAN,and the variation of the atmospheric radiation transmission properties is intense,the deviations are obviously increased.The simulation results of atmospheric transmittance,upwelling and downwelling radiation are all affected by the zenith angle and the height of observation points,but the trend and feature position will not change accordingly.To carry out the ground-based observation experiment using hyperspectral imaging spectrometer and TEL-1000 LW infrared camera.In order to ensure the success of the experiment and the accuracy of subsequent atmospheric correction,the azimuth and zenith angle of the moon were simulated by STK software before the experiment,the mobile weather station was used to monitor the weather conditions during the experiment.After the experiment,the center reflection spectrum data of the moon are obtained by the radiation calibration,and lunar brightness temperature data are obtained by decomposing video files.Based on the study of the atmospheric radiative transfer properties,we use the LBLRTM and MODTRAN radiative transfer models to make atmospheric correction for ground-based observation reflectance spectra and thermal infrared brightness temperature data.The M3 data obtained by Chandrayaan-1 satellite and Racca surface temperature modelis are selected to verify lunar atmospheric correction results.The overall trend of reflection spectrum consistent with the M3 data,but the correction results of the accuracy of LBLRTM is better.The lunar brightness temperature has been significantly improved after atmospheric correction,the brightness temperature correction result of the LBLRTM is between 330-380 K,the results of MODTRAN is between 320-360 K.The correction results and Racca surface temperature model calculation results are similar,the relative error below 11%.The results showed that LBLRTM and MODTRAN can well complete the ground-based data of lunar atmospheric correction.