Modeling and Spatio-Temporal Analysis of the Distribution of O3 in Tehran City Based on Neural Network and Spatial Analysis in GIS Environment

Air pollution is one of the most problems that people are facing today in metropolitan areas. Suspended particulates, carbon monoxide, sulfur dioxide, ozone and nitrogen dioxide are the five major pollutants of air that pose many problems to human health. The goal of this study is to propose a spatial approach for estimation and analyzing the spatial and temporal distribution of ozone based on GIS analysis and multi perceptron neural network. In the first step, by considering the accuracy of different interpolation methods, IDW method was selected as the best interpolation method for mapping the concentration of ozone in Tehran. according to the daily data of these pollutants, the daily, monthly, and annual mean concentrations maps were prepared for years 2015, 2016 and 2017. According to the results, it can be said that the highest concentrations of ozone are found in the southwest and parts of the central part of the city. Finally, a neural network was developed to predict the amount of ozone pollutants according to meteorological parameters. According to the data of ozone pollutants in year 2018, the accuracy of neural network for hot and cold days of year were about 68% and 77% respectively. Therefore, it can be said that the meteorological parameters of temperature, wind speed and direction, and precipitation are significantly related to the concentration of O3 pollutant.