This proposal outlines a study of the formation of auroral arcs caused by the large scale dynamics of ions and electrons in the magnetotail. The first step is to construct ion distribution functions moments (density, temperature, pressure, etc.) in a three-dimensional large scale magnetospheric model. This will be done using the technique of large scale kinetics, whereby thousands of ion trajectories are launched from a variety of sources into a magnetic and electric field model of the magnetosphere. The 3D auroral precipitation profile (with local time dependence) will be determined and compared with observations. The second step will be to perform a similar large scale kinetic study using electrons, after which the ion and electron results will be compared and in regions where there are charge imbalances, parallel electric fields will be imposed. The precipitated electron flux will then be calculated and theoretical auroral images will be produced. The third step is to introduce time dependence into the large scale kinetic model. This will be done by using different field model "snapshots" which represent different states (i.e., growth phase, recovery, etc.) of the magnetosphere and calculating the ion and electron precipitation profiles. Once a general understanding using these snapshots has been achieved, an attempt will be made to model the growth phase by temporally evolving the quiet time field model to a more disturbed time. The last step is to bring all of these results together to delineate the formation of auroral arcs. The physical mechanisms in the magnetotail that result in the large longitudinal extent of auroral arcs will be identified and the results will be compared to observations.