SUNDAY O. UDO, MFON DAVID UMOH, IGWE O. EWONA AND CLETUS NZAN AGBOR
Abstract
Ozone characterization has been carried out for Calabar, Nigeria. The study was conducted between December 2015 and November 2016, a period of one (1) year. The station lies at the heart of the city of Calabar (lat. 5o 1607" N, Long. 8o23 34. 56E). The data for the study was obtained from a stationary AQM 65 equipment programmed to measure and upload pollutants automatically. The pollutants measured include ozone (O3), nitrogen oxides (NOx), carbonmonoxide (CO), with some meteorological parameters (temperature, wind speed and relative humidity). Ozone characterization has been carried out for Calabar, Nigeria. The result shows that ozone correlates strongly though negatively with wind speed R = -0.678. Also, ozone correlates very poorly with temperature R = -0.279, relative humidity R = -0.082. The coefficient of determination R2 = 0.46 shows that wind speed has the greatest effect on the concentration of ozone in Calabar. The ANN models developed for the estimation of ozone shows that the model with 20 hidden neurons was found to be the best model because it had the least value of mean squared error (MSE). North Easterly winds were the prevalent wind throughout the year of study. The Republic of Cameroon has contributed to ozone pollution in Calabar. The ANN and linear regression (LRM) models were tested using mean absolute error MAE, mean percentage error MPE, mean squared error MSE, and root mean squared error RMSE. The result showed that the linear regression model had the smallest values of MAE, MAPE, MSE and RMSE. This shows that the LRM proves to be the better model in predicting ozone in this study.