Inversion Of Lunar FeO And Numerical Simulation Of The Detached Dust Layers On Mars

As the development of China deep space exploration, Moon and Mars have been the important targets for Chinese space missions. According to the recent released "the 13th five-year plan for the country's scientific and technological progress", the studies on planetary geology, building up of Martian and Lunar testing fields, and comparative studies of Earth-Moon-Mars are encouraged. The geological map compilation of Moon and Mars will also be organized. These governmental policies will further promote the Chinese deep space explorations and planetary sciences.The Interference Imaging Spectrometer (?M) aboard the Chinese first Moon orbiter Chang'E-1 obtained large amount of hyperspectral data. ?M data processing has been implemented by many studies, but some imperfections still exist. The bad pixels widely scatter in the ?M raw data. The ?M reflectance data calibrated through the Apollo sample does not make agreement with the real lunar surface reflectance. The telescopic data using in ?M cross-calibration procedure are inadequate to represent the various compositions of lunar surface. The boundaries of different ?M swath images are visible in the mosaic ?M image even after photometric correction.Dust is one of the important aerosol in Martian atmosphere. Its distribution has an impact on the thermal balance of the atmosphere. The observation from Mars Climate Sounder (MCS) aboard the Mars Reconnaissance Orbiter (MRO) revealed the existence of detached dust layers which is above the planetary boundary layer on Mars. In spring and summer, the altitude of observed detached dust layers is?20 km. In autumn and winter, the altitudes of the detached dust layers vary significantly. The variation range can reach to 20 km. The detached dust layers observed by MCS have never been noticed in former studies. And their origins are still debated.In this thesis, studies related to the aforementioned topics have been performed. The major results and conclusions can be summarized as follows:(1) First, in order to remove the bad pixels in ?M data, a new bad pixel identification and correction method using spectral angle and Euclidean distance has been proposed. To verify the accuracy of this new method, several bad pixel identification methods used by other studies are also implented for comparison purpose. The results show that the bad pixels in ?M data are removed properly by the new method. Second, to improve the reflectance calibration of ?M data, a new in-flight calibration has been conducted. In the new calibration procedure, the lunar surface is divided into four classes based on the lunar FeO content. For each class, several parameters for Lommel-Seeliger model is set to perform the photometric correction. The derived reflectance data after photometric correction show significant improvement. The boundaries between adjacent orbits are invisible, indicating good performance of photometric condistions. Meanwhile, the overestimate of reflectance originating from calibrating ?M data to Apollo sample spectra has been avoided. Third, the cross-calibration process has also been improved by importing more telescopic data to better represent various composition of the lunar surface.(2) Before using the preprocessed ?M data to inverse FeO abundance, we firstly performed a band selection study to choose the optimal near infrared (NIR) band to inverse FeO. By setting tour experiments, we found that the cross-calibrated 891 nm band is the best one to be used in FeO inversion. After confirmation of the optimal NIR band, global mapping of FeO has been achieved. Results show that the FeO abundance in lunar Mare derived from ?M is lower than that from Clementine-derived FeO. While, on the highlands, the abundance of ?M-derived FeO is slightly higher than that of Clementine-derived FeO.(3) In order to simulate the MCS-observed detached dust layers in GCM, a parameterization for rocket dust storms has been designed and implemented in the LMD Mars GCM. The model simulations show that in Martian autumn and winter, the GCM with rocket dust storm parameterization is capable of reproducing the detached dust layers. The formation and evolution of GCM-simulated detached dust layers are in agreement with that of MCS observation. Meanwhile, the simulation also suggests that the large altitude variation of detached dust layers in autumn and winter are contributed by the deep convection induced by rocket dust storms. Therefore, we conclude that rocket dust storm is the source of the MCS-observed detached dust layers. However, only with the rocket dust storm parameterization, GCM cannot simulate the detached dust layers occurred in spring and summer.(4) To further identify the origin of the detached dust layers in spring and summer on Mars, a parameterization for daytime slope winds has been designed and implemented in the LMD Mars GCM. The simulation of the GCM with daytime slope winds parameterization shows that with the help of daytime slope winds, the GCM can reproduce the detached dust layers in spring and summer, which cannot be simulated by the rocket dust storm process. The altitude and altitude variation of the simulated detached dust layers are comparable to those of MCS observations. Therefore, slope winds are the source of detached dust layers in Martian spring and summer. The altitude of GCM-simulated detached dust layer doesn't vary a lot, indicating that the convection induced by slope winds is shallower than that induced by rocket dust storm.