Digital Soil Mapping Based on Fine Temporal Resolution Landsat Data Produced by Spatiotemporal Fusion

Multi-temporal Landsat-8 satellite images with fine spatial resolution (i.e., 30 m) are crucial for modern digital soil mapping (DSM). Generally, cloud-free covering bare topsoil common choices DSM. However, the number of effective data is greatly limited due to cloud contamination coupled coarse temporal resolution, and interference material in most months, hindering development accurate To address this issue, temporally dense Landsat were predicted using a spatio-temporal fusion method improve Specifically, recently developed virtual image pair-based (VIPSTF) was adopted produce simulated time-series, by fusing 500 m Moderate Resolution Imaging Spectroradiometer (MODIS) time-series frequent observations. Subsequently, used distinguishing different classes via random forest (RF) model. Training validation samples collected from legacy data. Our results indicate that beneficial improving DSM owing increase class separability. More precisely, after combining observed data, overall accuracy (OA) kappa coefficient (Kappa) increased 3.099% 0.047, respectively. This research explored potential DSM, providing new solution remote sensing-based