Gully Erosion Monitoring At Multiplescales Based On Multi-source Remote Sensing Data Of The Typical Black Soil Area,Northeast China

The black soil area of Northeast is an important granary in China,and it is also one of the most serious areas of soil erosion.Gully erosion is the region's most important types of soil erosion,which is closely related to spatial and temporal scales.Multi-source and multi-scale remote sensing data can help to fully reflect the characteristics of gully erosion on different time and space scales.Based on the multi-source data including field investigation,CORS,UAV,high-resolution satellite and geographical census results,this paper takes Jiutai District of Jilin Province as the research area and analyzes the distribution and dynamic changes of gully erosion from different spatial and temporal scales such as headcut-area,ground-low-space,several months-several decades,and so on.Then,the multi-sources and multi-scales gully erosion remote sensing monitoring system is constructed.The research results will provide technical support for the comprehensive and effective monitoring of land degradation and its impact on the environment and social economy,and provide a scientific basis for land managers to make sustainable use of land sustainable planning policies.The main conclusions are as follows:(1)Remote sensing integrated with ground survey,high resolution satellite and UAV can comprehensively monitor the present situation of gullies in the study area including the location,quantity,length,area,volume and 3D information of gully erosion.UAV remote sensing can provide centimeter-level orthoimage and DSM of erosion gullies in different spatial ranges.These results can show details of the effects of human activities and soil erosion on farmland areas.In comparison,UAV remote sensing data is more detailed than the satellite data,more efficient than the traditional field work,and more complete on space.UAV remote sensing monitoring scale can effectively fill gaps in field measurement and satellite monitoring scales.(2)Compared with the traditional ground measurement method to obtain the 2D erosion groove data,the photo-based three-dimensional terrain reconstruction technology can take the photos taken by the consumption level camera to obtain the dense point cloud and the 3D model of the gully quickly,and use the more detailed 3D data to show the morphological characteristics of gullies.Due to the limitation of the ground condition,the three-dimensional reconstruction based on the ground photo does not completely show head of the structure of a typical erosion gully,but it can show the basic form of the erosion gully and clearly reflect the real details of the gully and the 3D model based on remote sensing of UAVs can show the 3D structure of the erosion gulls in a complete way,but some parts(such as steep gully walls)still appear cloudy,and the point cloud generated by the ground photo can fill the data.Based on 3D-PR technology,the combination of UAV images and terrestrial images can provide a more comprehensive high-resolution erosion gully model.(3)In 2013,the erosion gully is widely distributed in the Sancha River Basin,especially in the northeastern region,and the whole belongs to moderate erosion.The soil erosion on the slope of the Sancha River Basin is mainly due to mild erosion and micro-erosion,but from mild erosion to extremely strong erosion are the main types of soil erosion in the study area.The results of high-precision geography census improve the operation of the whole soil erosion model and improve the accuracy of some model factors.The slope-trench erosion is not exactly the same: the slope-trench erosion occurs mainly on the slopes in the middle elevation,but the peak of the gully peak is lower than that of the elevation of the erosion;the slope-trench erosion occurs mainly on the sunny slope.(4)Based on the Corona image of 1965 and the Pleiades image of 2013,the dynamic changes of the erosion gully in the past 50 years have been monitored that the number,length and area of erosion gullies increased significantly.UAV remote sensing can monitor the area and volume change of the erosion gully in the short term.From the time of 2015/10/31 to 2016/11/04,the retreat distance of the gully head reaches 1.6 m,while the area and volume of the gully head increased by 7.4 m2 and 3.36 m3 respectively.(5)The improvement of the resolution of the remote sensing image will make the image contour features and internal details of the erosion gully more clear,but will also increase the interference noise of the remote sensing identification and information extraction of the erosion trench to a certain extent;The image with more than 2m resolution can be effectively interpreted under the visual interpretation of large-scale branch-like gully system and cut gully,and the effective identification of the shallow trench requires no less than 1m resolution of the image;development gully,especially the remote sensing recognition of shallow strench is more sensitive to the change of image spatial resolution;sub-meter image compared to the frequently-used 2m image in the development of gully,especially shallow gully,has a more significant improvement for extraction accuracy(6)Gully erosion monitoring system at multiplescales based on multi-source remote sensing data can realize the effective identification,dynamic change monitoring and risk early warning of gully erosion at different time and space scales.