Mapping abandoned land is very important for accurate agricultural management.However,in karst mountainous areas,continuous high-resolution optical images are difficult to obtain in rainy weather,and the land is fragm...Mapping abandoned land is very important for accurate agricultural management.However,in karst mountainous areas,continuous high-resolution optical images are difficult to obtain in rainy weather,and the land is fragmented,which poses a great challenge for remote sensing monitoring of agriculture activities.In this study,a new method for identifying abandoned land is proposed:firstly,a few Google Earth images are used to transform arable land into accurate vectorized geo-parcels;secondly,a time-series data set was constructed using Sentinel-1A Alpha parameters for 2020 on each farmland geoparcel;thirdly,the semi-variation function(SVF)was used to analyze the spatial-temporal characteristics,then identify abandoned land.The results show:(1)On the basis of accurate spatial information and boundary of farmland land,the SAR time-series dataset reflects the structure and time-series response.abandoned land with an accuracy of 80.25%.The problem of remote sensing monitoring in rainy regions and complex surface areas is well-resolved.(2)The spatial heterogeneity of abandoned land is more obvious than that of cultivated land within geoparcels.The step size for significant changes in the SVF of abandoned land is shorter than that of cultivated land.(3)The SVF time sequence curve presented a strong peak feature when farmland was abandoned.This reveals that the internal spatial structure of abandoned land is more disordered and complex.It showed that time-series variations of spatial structure within cultivated land have broader applications in remote sensing monitoring of agriculture in complex imaging environments.展开更多
基金supported by the Guizhou Provincial Science and Technology Foundation(Qiankehe ZK[2022]-302)the National Natural Science Foundation of China,(Grant NO.41661088,41631179 and 42071316)+2 种基金the National Key Research and Development Program of China(Grant NO.2017YFB0503600)the Key Laboratory of Natural Resources Monitoring in Tropical and Subtropical Area of South China,Ministry of Natural Resources(No.2022NRM0004)Excellent Youth Project of Hunan Provincial Education Department(22B0725)。
文摘Mapping abandoned land is very important for accurate agricultural management.However,in karst mountainous areas,continuous high-resolution optical images are difficult to obtain in rainy weather,and the land is fragmented,which poses a great challenge for remote sensing monitoring of agriculture activities.In this study,a new method for identifying abandoned land is proposed:firstly,a few Google Earth images are used to transform arable land into accurate vectorized geo-parcels;secondly,a time-series data set was constructed using Sentinel-1A Alpha parameters for 2020 on each farmland geoparcel;thirdly,the semi-variation function(SVF)was used to analyze the spatial-temporal characteristics,then identify abandoned land.The results show:(1)On the basis of accurate spatial information and boundary of farmland land,the SAR time-series dataset reflects the structure and time-series response.abandoned land with an accuracy of 80.25%.The problem of remote sensing monitoring in rainy regions and complex surface areas is well-resolved.(2)The spatial heterogeneity of abandoned land is more obvious than that of cultivated land within geoparcels.The step size for significant changes in the SVF of abandoned land is shorter than that of cultivated land.(3)The SVF time sequence curve presented a strong peak feature when farmland was abandoned.This reveals that the internal spatial structure of abandoned land is more disordered and complex.It showed that time-series variations of spatial structure within cultivated land have broader applications in remote sensing monitoring of agriculture in complex imaging environments.