A Study on the Difference of LULC Classification Results Based on Landsat 8 and Landsat 9 Data

Landsat 9 enhances the radiation resolution of operational land imager from 12 bits 8 to 14 bits. The higher improves sensitivity sensor detect many subtler differences, especially in case dense forests or water. However, it remains unclear whether difference between and actually affects classification results water tree species. Accordingly, spectral reflectance vegetation indices were extracted this study, based on images. Then, models use cover (LULC) species developed by using a gradient boosting algorithm. Subsequently, analyzed further investigate how differences affect LULC It is shown that are relatively favorable most cases. poor test areas with lower accuracy Further analysis, results, indicates there significant two datasets. In other words, produces better than areas. temperature close zero may lead inverse results. addition, forest datasets small, but superior those 8, an improvement overall 6.01%. demonstrate has little impact Nevertheless, some areas, suited for enhancing