HR: 11:15h
AN: SM11D-04 INVITED     [PDF]
TI: A Coupled Model Using Solar Observations to Make Probabilistic Forecasts of Magnetospheric Activity
AU: * McPherron, R L
EM: rmcpherron@igpp.ucla.edu
AF: Institute of Geophysics and Planetary Physics, University of California, Los Angeles, CA 90095-1567 United States
AU: Siscoe, G
EM: siscoe@bu.edu
AF: Center for Space Physics, Boston University 725 Commonwealth Ave., Boston, MA 02215 United States
AU: Crooker, N N
EM: crooker@bu.edu
AF: Center for Space Physics, Boston University 725 Commonwealth Ave., Boston, MA 02215 United States
AU: Arge, C N
EM: Nick.Arge@noaa.gov
AF: Institute for Research in Environmental Sciences, University of Colorado, Boulder, CO 80309
AB: It is well established that the driver of magnetospheric activity is the rectified dawn-dusk component of the interplanetary electric field, VBz. Accurate forecasts of the temporal development of magnetospheric activity require prediction of the Bz waveform that will arrive at the Earth. This can be accomplished relatively well using observations from L1, but only with about 30-60 minutes lead time. Longer term forecasts are unlikely to be successful because Bz is a turbulent variable produced by a superposition of evolving structures and waves in the solar wind. For forecasts with 1-4 days lead time we advocate the use of probabilistic forecasting, e.g the probability that Dst $<$ -50 nT is 10%. In this report we describe a technique we call probabilistic forecasting by air mass climatology. The basic premise underlying this method is that structures in the solar wind such as CIRs and CMEs organize the statistical properties of the solar wind and hence magnetospheric activity (climatology). We have found that the stream interface between high and low speed solar wind is extremely effective in organizing the statistical properties of geomagnetic indices during the declining phase of a solar cycle. We have used the time of many interfaces in 1994-1996 to determine the cumulative probability distributions (cdfs) for 3-hr ap and 1-minute Dst indices. We couple these statistical results to the Wang-Sheeley (W-S) model. The W-S model utilizes telescopic observations of the Sun's photospheric magnetic field to predict the speed of the solar wind, the polarity of the IMF, and the magnitude of the IMF at the Earth 3-4 days in advance. By detecting stream interfaces in the predicted velocity we obtain 2-3 days advance warning of a stream interface. We then use the climatology of an index relative to the predicted interface (the cdfs) to forecast the probability of enhanced activity. We will evaluate the quality of the predictions for the years 1995-1996 and apply it to current observations as we progress into the declining phase of the current cycle.
DE: 2102 Corotating streams
DE: 2722 Forecasting
DE: 2778 Ring current
DE: 2784 Solar wind/magnetosphere interactions
DE: 2788 Storms and substorms
SC: SPA - Magnetospheric Physics [SM]
MN: 2003 Fall Meeting