HR: 1340h
AN: SM43A-1140    [Abstracts]
TI: How Much do Numerics Affect the Results of Global MHD Magnetsopheric Calculations?
AU: * Lyon, J
EM: lyon@tinman.dartmouth.edu
AF: Dept. of Physics and Astronomy, 6127 Wilder Lab Dartmouth College, Hanover, NH 03755 United States
AB: Global MHD simulations of the magnetosphere are increasingly used for interpretation of magnetospheric data and for providing a realistic picture of the state of the system. However, the major codes used often do not agree, sometimes at a level above details.In some limit, however, since the codes are all solving the same equations, the results should converge. The limits where this occurs (usually high spatial and temporal resolution) may be far beyond the capabilities of current computer systems. This paper is aimed at obtaining systematic insight into how numerical resolution, spatial order of differencing, and numerical shock capturing algorithms affect the ``accuracy'' of the resulting solution. We will do this by analyzing the results for a case with constant Northward IMF. Raeder (JGR, 104,17357, 1999) analyzed a similar case and concluded that numerical resistance was the determining factor in the length of the magnetotail. In our case, the LFM code was modified to examine a three dimensional parameter space in resolution, differe8663ncing order, and sharpness of shock capturing. We will present results for three quantities: general configuration, reconnection structure, and ionospheric field-aligned currents. Not surprisingly, the results can be quite different depending on the numerics, with the structure of the field-aligned currents perhaps being the most sensitive to calculation details.
DE: 2708 Current systems (2409)
DE: 2740 Magnetospheric configuration and dynamics
DE: 2753 Numerical modeling
SC: SPA-Magnetospheric Physics [SM]
MN: 2004 AGU Fall Meeting