Forecasting Total Ozone Column over Iraq using Artificial Intelligence Approach: a case study for Baghdad Province

Abstract

The present paper presents a new approach for total ozone column forecasting using an Artificial Neural Network technique for Baghdad, Iraq. Total ozone column data for the period (1979-2000) were used as training and the period (2009-2011) for testing and one year(2012) for forecasting also the combination of meteorological elements have been used as input parameters (stratospheric temperature, Geopotential height and zonal wind). at 50 and 70 mb respectively. The developed ANN models are being applied for aimed at providing forecasts in a near real time schedule, different network types were tested with different kinds of input information. Preliminary tests showed that a generalized feed forward ANN model using Gaussian activation function achieved the optimum forecasting of total ozone column. To calculate differences between measured and forecasted values of total ozone column, root mean square error (RMSE), mean absolute error (MAE), Relative error (RE) and determination coefficient (R 2 ) were calculated. According to four statistical indices were determined in order to examine the accuracy of the optimum ANN model, it was found that the more accurate model among the four considered statistical indices were RMSE, MAE, RE and R 2 was 0.05244, 0.0392, 0.4709 and 0.9974 respectively, these results show that ANN forecasts of total ozone column over