资源环境科技发展态势分析平台

In recent years, the city of Tehran, Iran's capital, has encountered numerous dust events so that the dust concentration of PM₁₀ has reached even more than 800 μg m⁻³. This emphasizes the importance of the statistical study of dust in Tehran and the development of correlations for estimating dust concentration of PM₁₀. In the present study, by evaluating the data measured during dust observations over the years 2013–2016 in Tehran, new statistical models are established for estimating PM₁₀ concentration in terms of horizontal visibility and MODIS AOD. Firstly, simple nonlinear regression models between dust concentration of PM₁₀ and horizontal visibility as well as MODIS AOD are developed. Afterward, by combining the mentioned parameters with meteorological measurements, including wind speed, relative humidity, and temperature, multivariate nonlinear regression models are obtained. The descriptive statistical parameters and Taylor diagram concept are used in order to compare the performance of various models. Analyzing the developed models indicates that the simple model based on horizontal visibility has the best performance (R = 0.90) and considering meteorological parameters does not have a significant positive impact on the model performance. Furthermore, the estimated dust concentration of PM₁₀ using simple models developed in this study has a better correlation with the measurements in Tehran in comparison with the results obtained by models of other researchers. This demonstrates the importance of developing such models according to the climatic and geological conditions of the investigated region.