One of the long term goals of NSF's Geospace Environment Modeling (GEM) program is the construction of a Geospace General Circulation Model (GGCM), a comprehensive and predictively powerful numerical model of geospace. One popular approach to GGCM design consists of using a global magnetohydrodynamics (MHD) code as the computational "spine" of the model, including additional physics where needed (e.g., by coupling the global MHD code to a kinetic model of the ring current). Despite its successes, however, the MHD spine approach suffers from a number of well known deficiencies, the most serious of which is the inability of MHD to model the kinetic processes which are thought to play an essential role in magnetic reconnection. In order to achieve fast reconnection in global MHD codes, modelers usually resort to ad hoc localized, numerical or current dependent resistivities, and it is not clear how the reconnection dynamics depends on the resistivity or whether the results even converge as the numerical resistivity decreases with increasing grid resolution. Despite the fundamental role reconnection plays in driving magnetospheric dynamics, the reconnection time scale problem hasn't received much attention by global MHD modelers. The issue is often dismissed with appeals to the "Axford Conjecture" which states that the reconnection rate is determined by external boundary conditions, with the diffusion region adjusting to accommodate these conditions.
There is now compelling evidence that the Axford Conjecture does not apply at the dayside magnetopause and it has never been systematically tested in global MHD simulations of magnetotail reconnection. This research program will attack the problem of how the generation and dynamics of global MHD storms and substorms depend on the resistivity. It will address the following issues: 1) How does the physics of magnetopause reconnection in global MHD simulations depend on the resistivity model during strong, sustained southward IMF solar wind driving (i.e., conditions which typically produce strong storms)? 2) Can global MHD simulations produce storms and substorms in the high Lundquist number limit? 3) How does the storm-substorm relationship depend on the resistivity model?
The research will combine a straightforward parameter study with event study model/data comparisons: we will run the OpenGGCM global MHD code with coronal mass ejection solar wind boundary conditions, varying both the solar wind parameters as well as the resistivity model. The primary new result will be a definitive test of the Axford Conjecture in the context of magnetic storm and substorm dynamics.
The analysis of the storm/substorm relationship will require the development of a Vlasov test-particle code (which will solve the Vlasov equation in prescribed electric and magnetic fields output from OpenGGCM). This code will be developed primarily by a graduate student at the University of New Hampshire (UNH).
The Earth's magnetosphere is a system in which collisions between individual particles, and hence the resistivity of the tenuous plasma, are extremely small. However, computer simulation codes of the magnetopshere----which are global in nature, include realistic geometry, and are often used for space weather predictions----are often developed, for reasons of computational efficiency, with high values of the plasma resistivity. While such codes are routinely used and are often valuable as approximate predictors of large-scale magnetopsheric dynamics, the accuracy of their predictions in mesoscales or microscales are often inadequate. Furthermore, since large scales are inexorably coupled to small scales, the overall predictive power of these codes is compromised severely unless we are able to extend them to regimes of low resistivity. In this project, we have made substantial improvements in applying global codes to the regime of low resistivity. In the process, we have uncovered new plasma instabilities and dynamics that were heretofore not seen. These instabilities include the discovery of the so-called plasmoid instability of extended thin and intense current sheets. We have demonstrated here for the first time that these thin unstable current sheets spontaneously break up to produce many plasmoids, which can be thought of cat's eye-shaped magnetic bubbles as a result of the instability, and lead to a nonlinear regime of fast reconnection in which the reconnection rate becomes independent of the resistivity. We have explored some of the ramifications of these new findings in applications to fast magnetic reconnection in the magnetotail and flux transfer events at the magnetopause.
