Dust Source Prioritization with using Statistical models (Case Study: Khorasan Razavi Provience)

Due to the drought and land use changes in recent years, dust storm phenomenon in Iran has increased. The purpose of this research is to identify and prioritize dust source areas using bivariate models (probability model of event weight and frequency ratio model) in R software environment and determine the importance of each of the environmental factors affecting it in Khorasan Razavi province. To this porpuse, 65 dust extraction points were identified in the study area and a dust distribution map was prepared. Then maps of each of the factors influencing the occurrence of dust including soil maps, lithology, slope, vegetation index (NDVI), distance from the river, geomorphology and land use were prepared. Using the frequency and event weighting models, the weight of each effective factor and the relationship between each of the factors and the points of dust source were determined and, finally, priority maps of the dust source areas were prepared for the case study. Models were evaluated using the ROC curve. According to the results of both models, geomorphology units, land use and slope have the most effect on the occurrence of dust in the region, and both models have a frequency ratio and event weight with a sub-curve of 0.818 and 0.825, respectively. They accept.