prediction of temporal and spatial variations of the nddi drought index in the khorramabad watershed using remote sensing

Remote sensing-based indices are effective tools for monitoring drought and wet conditions. in this study, the nddi, derived from ndvi and ndwi data obtained from sentinel-2 satellite imagery, was employed to assess drought conditions. time series of these indices were generated using coding in the gee platform, and the nddi was subsequently calculated in excel. additionally, future predictions of the nddi were conducted using time series modeling techniques. the results indicate that the nddi is a reliable indicator for representing droughts caused by water scarcity and reduced vegetation cover. analysis of nddi values from 2016 to 2023 in the khorramabad watershed revealed a range between -3.20 and -11.21, suggesting that the region generally experienced very low drought intensity during this period. furthermore, drought prediction results based on nddi, using time series modeling, identified the ma2 model as the most accurate, with a high coefficient of determination (r² = 0.92) and an akaike information criterion (aic) value of less than 50. the findings indicate that the decline in nddi during the spring (-5.3) and winter (-5.4) of 2024 reflects improvements in relative vegetation cover, precipitation levels, and water reserves. however, an increase in this index during the summer and autumn (approximately -3) of 2024 suggests worsening drought conditions and reduced rainfall. this trend is projected to persist across different seasons in 2025 and 2026. in conclusion, the nddi is recommended as a valuable tool for analyzing vegetation cover status and water fluctuations, enabling optimal watershed management strategies.