Dust source dynamics in arid Iran: Examining the relationship between MODIS AOD and land surface characteristics in a dried catchment

Abstract

Dust sources are expanding unprecedentedly across the arid parts of Iran. To quantify this trend in the Kaftar Catchment in SW Iran (5,600 km²), the MODIS AOD product “Optical_Depth_047″, accessible through Google Earth Engine was utilized, as one of the moderate-resolution (1 km) aerosol data sources spanning from 2002 to 2023. The layers were preprocessed and averaged to derive annual mean data (AM-AOD), and its accuracy was verified using reference dust data collected from 24 stations in 2022 (correlation coefficient = 0.565; p-value < 0.01). The Mann-Kendall trend test confirmed a statistically significant upward trend in AOD values. A positive Sen’s slope predominated for the majority of the AM-AOD pixels. The Generalized Additive Model (mean R² = 0.753) successfully predicted AM-AOD values at each pixel for the year 2035. According to the Landsat 8-derived LULC map of the region (Kappa = 83.56), the wetland area is projected to experience the highest increase in AM-AOD in 2035, reaching 0.43 ± 0.06, while rangelands are expected to observe the lowest increment at 0.17 ± 0.07. Our study underscores the urgency of rehabilitating the Kaftar Wetland, promoting low-water crops, and improving ecosystem resilience in rangelands.