A Hybrid Deterministic-Statistical Model Integrating Economic, Meteorological and Environmental Variables to Air Pollution

Abstract

The following study is based on a hybrid statistical-deterministic model designed for the assessment of the daily concentration of sulfur dioxide, carbon monoxide and particulate matter (PM10) as major pollutants in the Greater Tehran Area (GTA): the capital of Iran. The model uses three available or assessable variables including economic, meteorological and environmental in the GTA for the year 2003. Economic sectors which are examined in this study are firstly traffic, secondly residential-commercial heating and thirdly industry. The model determines to what degree each of the aforementioned sectors, in accordance to their associated fuel consumption, is responsible for air pollution. The model also relates emission data from the three sectors whilst taking into consideration meteorological parameters. Thereafter, economic and meteorological parameters as independent explanatory variables opposed to the concentration of pollutants measured at the monitoring network stations which are dependent variables. All data is given in the form of time series for the year 2003 in specified areas discussed. The method adopted for the calculation of the regression coefficients of the model, is based on nonlinear least squares multiple regression analysis. The model has been tested on the available monitoring network stations for aforementioned pollutants in the GTA. Model verification has been carried out spatially in the year 2003 and temporally for the year 2005. Results show that the concentration of pollutants in the GTA can be estimated using this model. Areas of further research are outlined which indicate possible enhancement of this approach and relevant application extensions.