Research On Sand Dune Distribution And Mobility Monitoring Method Based On Multi-source Remote Sensing Data

Desertification has a certain impact on the natural environment and human survival and development.Hunshandake Sandy Land is located in the central desert area of Inner Mongolia Autonomous Region,which is a typical grassland desertification area.Because of its special geographical location,it has become the key management area of the Three North Shelterbelt Project and the Beijing-Tianjin Sand Source Control Project.Sand dune is a common aeolian landform in desertification area.The dunes in Hunshandake Sandy Land are mainly aeolian parabolic dunes.Dunes move after being driven by wind,which is the main manifestation of wind erosion desertification.The remote sensing quantitative identification and extraction of dune distribution pattern and its dynamic migration characteristics can provide important indication information for regional desertification evolution and provide decision-making method reference for the optimal management of desertification ecological environment.In this paper,multi-source remote sensing data,including airborne lidar data,Sentinel-1 satellite data and high-resolution multispectral data,are used to monitor the distribution and mobility of dunes in Hunshandake Sandy Land.The main research contents include the following two aspects:1.Aiming at the problem that the traditional dune feature line method is not applicable to the extraction of compound parabolic dunes,a dune height and distribution extraction method based on dune reference surface is proposed.The digital elevation model of the study area is generated by lidar technology.This method first determines non-dune points in a small area digital elevation model.The Kriging interpolation is used to interpolate the non-dune points to obtain the dune reference surface.Then the height of the dune is obtained by subtracting the digital elevation model from the elevation of the reference surface of the dune.Finally,multi-spectral data is used to adjust the threshold to obtain accurate dune distribution and height.Morphological parameters are the basis of aeolian geomorphology research.After obtaining the height and distribution information of dunes,the morphological parameters of dunes in the study area are calculated.The results show that the slope of large-scale parabolic dunes in the study area is mainly concentrated in 3-21°,and the projected area of dunes is 663km~2,accounting for about 42.7%of the study area.The two typical dunes in the study area are affected by the two main wind directions of west and northwest.The west and northwest are the opening directions,and the vegetation mainly grows on the leeward slope of the dune.2.Aiming at the interference of vegetation coverage area and lake area on the monitoring of dune synthetic aperture radar(SAR)intensity offset tracking algorithm.A coherence index assisted SAR intensity offset tracking method is proposed.This method first uses the SAR intensity offset tracking algorithm to obtain the overall movement of the sample area,and then combines two coherence indicators to exclude the interference of vegetation growth and lake areas,and obtains the main movement direction and cumulative displacement of the dunes in the three sample areas.The feasibility of the experimental results was verified by combining the wind condition data of the meteorological stations in the study area.The results show that from June 4,2021 to September 20,2021,the dune group of sample area a moved 0.47 meters to the north-east38°,the dune group of sample area b moved 0.36 meters to the north-east 66°,and the dune group of sample area c moved 0.38 meters to the south-east 18°.The amount of dune movement is positively correlated with wind speed,and the main direction of dune movement is consistent with the wind direction.