Prediction and Modeling of Seasonal Concentrations of Air Pollutants in Semi-Urban Region Employing Artificial Neural Network Ensembles (Published)
This study utilizes Artificial Neural Network (ANN) ensembles to predict seasonal variation of air pollutants in semi-urban region of Eleme, Rivers state, Nigeria. A ten year monthly concentrations of SO2, NO2, CO and CH4 in the region was obtained for dry and rainy seasons. Air pollutant concentrations in semi urban area of Eleme can be attributed mainly to industrial activities, vehicular emissions and some local background concentrations influenced by meteorological and geographical conditions of the area. Training of the network models was achieved using Neural NetTime Series feature of MATLAB software. Observed concentrations of pollutants and meteorological parameters were used as input variables for the prognostic models. The developed ANN prognostic models accurately captured the dynamic relationships between pollutant concentrations and meteorological predictor variables. The relationships between predicted and observed values were highly significant at 95% of confidence level for all models as dry and rainy seasons models gave R2 greater than 0.99 (indicating close relationships between observed and predicted values). CH4 showed R2 of 0.9946 and 0.9974 for dry and rainy seasons respectively; CO showed R2 of 0.9918 and 0.9972 for dry and rainy seasons respectively; NO2 showed R2 of 0.9998 and 0.9982 for dry and rainy seasons respectively; SO2 showed R2 of 0.9921 and 0.9991 for dry and rainy seasons respectively. The trend in predicted pollutants indicated that the study area is a major receptor of air pollutants emanating mainly from industrial activities and vehicular exhaust emissions. Further research study is needed to compare ANN model with other modeling approaches such as with multiple linear regression models for the prediction of air pollutants.