Application of geostatistical methods for estimation of Tehran Air pollutants spatial distribution

Ambient air quality is a major concern in highly urbanized and industrialized regions such as Tehran.
In this paper, the spatial distribution of 4 air pollutants in Tehran atmosphere was analyzed. The analyzed air pullutants were Carbon monoxide (CO), Nitrogen Dioxide (NO2), Ozone (O3) and atmospheric particulate matters less than 10 micrometers in diameter (PM10). For this purpose, 4 common geostatistical interpolation methods namely: Ordinary Kriging (OK), Universal Kriging (UK), Sample Kriging (SK), and Ordinary Cokriging (COK) with Gaussian modeled semivariogram, were used to estimate the continuous surface for the 4 mentioned air pollutants. The data were collected from 21 air quality monitoring stations located in different districts of Tehran during a 2-year period from 2012 to 2013. The Kriging interpolation schemes are stochastic, local, gradual and exact interpolators. After preprocessing the collected data, they were imported to GIS by using metric coordinate system (UTM Zone 39). Finally, the Kriging predicted map was evaluate using 3 statistical indices of validation namely: Mean Absolute Error (MBE), Mean Bias Error (MAE), and Root Mean Square Error (RMSE) that can be divide into systematic and unsystematic errors (RMSEs, RMSEu).
Findings: The results indicated that, using 2 auxiliary variables, the OCK is the optimum schema for spatial estimation of CO and NO2 pollutants in Tehran. Moreover, SK is found out as the best model for spatial estimation of NO2 and CO. According to optimal model, the highest concentrations of ozone (O3) and particulate matters greater than 10 microns (PM10) are observed in the marginal areas of Tehran, while the highest concentrations of CO, NO2 are observed in the central and northern districts of Tehran such as districts 1 to 4.
The developed optimized model can be used for real time estimation of any pollutants in Tehran atmosphere by updating the observed data.