Remote sensing assessment of multi-year drought vulnerability of agriculture in Kangavar, Kermanshah Province, west of Iran

Abstract

Abstract The current study investigated the spatiotemporal distribution and detection of drought severity using ground indices in R-Studio software and applying indices on the images of Landsat 8 and Sentinel 2 satellites in the growing season and annually. The outcomes showed that typical months had the largest range in terrestrial indices results, and the NDVI index had the greatest compatibility with SPEI. By comparing these two times, during the growing season, plant health was evident in the northwest and in annual images in the northeast. Moreover, a negative correlation was found between LST and NDVI in elevation, which was more moderate at northern heights. High mean annual temperatures during growing seasons were common in the east and south. The eastern side of the study area had high annual vegetation density, but soil moisture was higher in the southwest. Floods negatively affected plant health; with the 2019 flood, the canola yield was very low. In summary, Sentinel 2 images showed drought better in the growing season and Landsat 8 images showed drought better in the dry season or at the end of plant growth.