Understanding the interaction mechanism between electromagnetic waves and ground targets is a primary task for intelligent processing and quantitative inversion of multi-source, multi-band remote sensing data, and generating the remote sensing data products. The applications of remote sensing in geosciences are essentially based on this understanding.

(1) In polarimetric remote sensing, the polarimetric spectra of typical ground objects such as rocks and soils are measured, and the inversion theory of ground objects using polarimetric spectral data is developed. Based on the observed neutral points of atmospheric polarization, the ground and atmospheric parameters are decoupled, and therefore a clearer image of ground targets is obtained, and the stereo tomography of the atmosphere image is available.

(2) In microwave remote sensing, the processing algorithm of multi-mode InSAR, and the atmospheric correction method of InSAR based on the numerical forecast model and data assimilation method are studied. The interferometric processing of the multi-mode SAR images is realized, and the resulting open-source PUMSIP software has been used by many researchers from institutes around the world.

(3) In quantitative remote sensing, the unified inversion models of LAI and fPAR are developed, which are suitable for both discrete and continuous vegetation and has been widely used in the production of regional and global remote sensing products. In combination with the optical and microwave remote sensor datasets, a robust co-inversion model for the retrieval of land cover parameters such as soil moisture and snow depth is created, which has been utilized in the applications of the ecological and environmental assessment and hazard monitoring over the arid regions. A cropland monitoring model based on the multi-dimensional spectral space technique, a drought monitoring model applicable to the whole growth period of crops, and an evapotranspiration model based on the hybrid linear dual-source datasets were proposed, and they have been applied to achieve agricultural drought monitoring in some arid regions in China and other countries. A backpack mobile terrestrial laser scanning system is developed, which is applied in precision forestry and canopy structure ecology.

(4) In planetary remote sensing, a radar scattering model from lunar regolith layer with rough surface and buried rocks is developed, a thermal emission model from lunar surface is proposed, then microwave brightness temperature of the Moon and lunar radar images are analyzed, and typical characteristic parameters of the lunar surface are inverted. One of the applications of these models is to provide a new way in the search of the potential water-ice deposits in the lunar polar regions.

So far, we have published 233 papers in leading international journals, including Remote Sensing of Environment, IEEE Transactions on Geoscience and Remote Sensing, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal of Geophysical Research and Geophysical Research Letters, and the citations is more than 1100. In addition, the faculty members published two monographs, got nine National Invention Patents, and six Software Copyrights. Thanks to the above scientific achievements, the faculties received the first prize of Beijing Natural Science Awards, one first and one second prizes for scientific and technological progress in survey and mapping, and the gold medal of Geneva International Invention Awards.