HR: 1340h
AN: SH53B-0317 [Abstracts]
TI: Predicting Intensity of Major Geomagnetic Storms from Solar and Interplanetary Properties of CMEs: A
Statistical Model
AU: * Srivastava, N
EM: nandita@prl.ernet.in
AF: Udaipur Solar Observatory, Physical Research Laboratory, P.O. Box 198, Udaipur, 313001
India
AB:
This paper describes a statistical model based on logistic regression for predicting the occurrence of intense geomagnetic
storms. Previous studies show that the strength of the resulting geomagnetic storms depend upon various solar and
interplanetary properties. During 1996-2002, 64 geo-effective events were recorded which were used for determining the
parameters influencing the strength or the magnitude of the resulting geomagnetic storms. 60% of the events recorded were
used for determining the model parameters while the rest 40% were used for validating the model. Independent variables which
include a number of solar and interplanetary factors were provided as input to the model and regressed against a binary
dependent variable viz. occurrence of the intense geomagnetic storms. The present model predicts 90% of the training events
correctly and 90% of the validation events correctly. The results indicate that the logistic regression model can be
effectively used for space weather prediction i.e. the strength of the geomagnetic storms.
DE: 7513 Coronal mass ejections
DE: 7519 Flares
DE: 2134 Interplanetary magnetic fields
DE: 2164 Solar wind plasma
DE: 2788 Storms and substorms
SC: SPA-Solar and Heliospheric Physics [SH]
MN: 2004 AGU Fall Meeting