Bayesian prediction of geomagnetic storms: Wind data, 1996-2010

摘要: The feature-based Bayesian method previously developed by Chen et al. (1996, 1997) to predict the occurrence, severity, and duration of large geomagnetic storms has been run on a daily basis Wind/Magnetic Fields Investigation (MFI) data from January 1996 until March 2010. algorithm uses as input real-time solar wind magnetic field obtained at L1 Lagrange point, output is probability prediction structure upstream that yet arrive, its geoeffectiveness, where geoeffectiveness measured traditional Dst index. performance characteristics are evaluated using four-level contingency table: nonstorm disturbances (−80 nT < Dst ≤ − 50 nT), weak (−120 < Dst ≤ − 80 moderate (−160 < Dst ≤ − 120 strong (Dst ≤ − 160 nT). It found greater level disturbances, more accurate is. With combined, correctly predicted 30 out 37 (81%). false negatives caused structures with short durations (≲ 1–2 hrs) southward (say, Bz ≲ −30 which sparsely represented in distribution functions constructed prior (OMNI set 1973–1981). does not storm onset time, but results present previous tests show average warning time ranges few hours maximum 10–15 hours.