Study on the Classification and Change Detection Methods of Drylands in Arid and Semi-Arid Regions

The aim of this study was to clarify the distribution irrigated drylands in arid and semi-arid areas, where complex terrain, diverse crops staggered cultivated lands exist. This paper studied classification methods based on temperature, precipitation, Normalized Difference Vegetation Index (NDVI) Water (NDWI) from Landsat data one-harvest area northern Loess Plateau China by using Google Earth Engine (GEE) platform. An extraction method proposed for regions northwest China. In addition, change types rainfed two periods were classified, a also put forward directly classify image differences between combined with results each period. It found that combining ratio NDVI NDWI accumulated values temperature precipitation 30 days before imaging could effectively improve accuracy. Moreover, classifying gaps remote sensing factors time dimension spatial clustering information yield more accurate type change, because accumulation errors maps obtained overlay analysis be avoided. accuracy improved introducing dynamic into historical periods. complements dryland, can existing products terms resolution, which fill shortage detailed region.