2.2 Data sources
This study used three different spatial scale survey methods to monitor ephemeral gully: satellite remote sensing monitoring, unmanned aerial vehicle (UAV) monitoring, and field investigations (Fig. 2). The remote sensing image data included 0.50 m high-resolution satellite images (digital orthophoto map, DOM) in 2009, 2012, and 2018 and 0.50 m high-resolution DOM images by UAV in 2021 (11,720 images). A DEM with 0.15 m resolution and DOM with 0.50 m resolution were created using Agisoft Metashape (Windows1.7.3, Agisoft, Russia). The 1:10,000 DEM data used in this study were obtained from the Shaanxi Geomatics Center of the Ministry of Natural Resources of China, and the 0.15 m resolution DEM data were obtained by the authors using UAV (Table 1). In order to avoid the influence of different DEM resolutions on the extraction accuracy of ephemeral gully morphological parameters, we changed the resolution of 0.15 m to 5 m by resampling in ArcGIS software (ArcGIS 10.2, Esri, U.S.A). The remote sensing and UAV images were obtained from April to June each year when the vegetation was not yet lush and thus had no significant impact on ephemeral gully recognition.
Table 1 Basic information on the DOM and DEM data used in this study.