The dust deposition model (DDM): An empirical model for monitoring dust deposition using meteorological data over the Isfahan province in central Iran

Abstract

An understanding of the relationship between climatic or meteorological data and dust deposition is of great utility for aeolian dust studies. Since dust sampling is a time-consuming and costly, and on the other side, climatic data are easily available from meteorological stations worldwide, such relationships allow climatic data to be converted into dust deposition. In this line, there have been several attempts but few desirable results have been found to present an accurate, simple and applied model to predict dust deposition from meteorological data. In this research a new model is presented for deriving monthly values of dust deposition. Dust samples were collected from 20 locations over the sampling period of one month during each season in 2017 and 2018, in Isfahan province, central Iran. Then, dust deposition was modeled using multi-linear regression (MLR) and artificial neural networks (ANN) approaches and employed metrological parameters. Results demonstrated that the obtained models can perfectly predict dust deposition (r2 = 0.95 to 0.97) from visibility, only. Since there was no significant difference between MLR and ANN approaches, a linear model with an r2 = 0.95 was chosen as the best fit between visibility (m) and dust deposition rate (gm-2month−1). As the employed climatic data were collected according toWorld Meteorological Organization(WMO) standards, the DDM can be applied to other regions of the world, especially in areas similar to the study area, but this doesn’t guarantee it can accurately estimate the dust deposition all over the world.