Prediction of the geomagnetic storm associated D-st index using an artificial neural network algorithm

In order to enhance the reproduction of the recovery phase D-st index of a geomagnetic storm which has been shown by previous studies to be poorly rep roduced when compared with the initial and main phases, an artificial neura l network with one hidden layer and error back-propagation learning has bee n developed. Three hourly D-st values before the minimum D-st in the main p hase in addition to solar wind data of IMF southward-component B-s, the tot al strength B-t and the square root of the dynamic pressure, root nV(2), fo r the minimum D-st, i.e., information on the main phase was used to train t he network. Twenty carefully selected storms from 1972-1982 were used for t he training, and the performance of the trained network was then tested wit h three storms of different D-st strengths outside the training data set. E xtremely good agreement between the measured D-st and the modeled D-st has been obtained for the recovery phase. The correlation coefficient between t he predicted and observed D-st is more than 0.95. The average relative vari ance is 0.1 or less, which means that more than 90% of the observed D-st va riance is predictable in our model. Our neural network model suggests that the minimum D-st of a storm is significant in the storm recovery process.