dc.description.abstract | It is well known that the dayside magnetopause moves closer to the Earth with increasing southward IMF Bz and solar wind dynamic pressure Dp. But the response of magnetopause location to the southward Bz and Dp is unclear when the magnetopause locates around the geosynchronous orbit. To understand the macroscopic interaction between solar wind and dayside magnetopause under such extreme conditions, the predictions of magnetopause models are compared with the magnetosheath encounters observed by GOES spacecraft during 1986-1992 and 1999-2000. This comparison is made for each 1-min data point. Three magnetopause models, Petrinec and Russell [1996], Shue et al. [1998], and modified Chao et al. [2002], are separately examined for their forecasting capability by using the following parameters: probability of prediction (PoP), probability of detection (PoD), and false alarm rate (FAR). Higher PoP and PoD with a lower FAR imply a better forecasting model. The corresponding solar wind conditions are obtained from the upstream monitors of ACE, Wind, IMP 8 or Geotail spacecraft.
On 6 April 2000, the predicted magnetopause is inside the geosynchronous orbit for corresponding Dp ~ 8 nPa and southward Bz > 25 nT, while the magnetopause is still outside 6.6 RE for such strong solar wind condition based on the magnetic field observations of GOES spacecraft. We propose this phenomenon comes from the saturated Bz-influence on magnetopause locations when southward Bz is large such that the earthward motion of dayside magnetopause stagnates. By means of the calculations of the modified Chao et al. [2002] model, possible magnetosheath encounters at geosynchronous orbit during 1999-2000 are used to obtain a relationship between and Dp, , such that the FAR is minimized and PoP is maximized, where is the threshold of IMF Bz for saturation occurring in nT and Dp is in unit of nPa. This obtained dependence is applied to the modified Chao et al. [2002] model to compare against the magnetosheath encounters observed by the LANL MPA instruments on 31 March 2001. Our results indicate that the model’s prediction capability is indeed improved when the saturation effect is considered. | en_US |