All-sky auroral images acquired at Ny-¿lesund were used in conjunction with observations of a Polar overflight on November 30, 1997, to determine where prenoon, 0900--1000 magnetic local time (MLT) auroral structures map to the outer magnetosphere. Polar observations at midaltitudes are used to constrain the mapping between the aurora and the magnetosphere. The Tsyganenko 96 magnetic field model (T96), driven by interplanetary conditions and DST, is used for that mapping. When the T96 model is driven by conditions on this day, the open/closed field line boundary maps 2--3 degrees lower in latitude than observed for this day. By making an ad hoc adjustment to match the location of the model open/closed field line boundary to observations, we find that the agreement with other aspects of the observations is also improved. These include (1) Polar observations of the dayside extension of the boundary plasma sheet (BPS) (structured low-energy electrons, detected in a region of sunward convection and region 1 field-aligned currents) mapped to the ionosphere in a region of discrete aurora, (2) Polar observations of the central plasma sheet (high-energy electron precipitation) maps to diffuse green auroral emissions equatorward of the discrete aurora, and (3) Polar electric field observations are consistent with convective motions of the auroral forms and changing interplanetary magnetic field (IMF) conditions. When the all-sky images are mapped into the magnetosphere, we find that the discrete aurora identified with the BPS mapped to the outer dawnside edge of the magnetosphere. This mapping indicates that earlier Geotail reports of a sunward flowing mixing region in the equatorial magnetosphere [Fujimoto et al., 1998> is really within the dayside extension of the BPS, compatible with both the low-altitude BPS observations of Newell and Meng [1992> and type 4 dayside aurora of Sandholt et al. [1998>. Since the Polar observations place the sunward flowing mixing region as a source of region 1 currents for this IMF BY positive case [Farrugia et al., this issue>, these results extend the source of the region 1 currents from the antisunward flowing low-latitude boundary layer into the sunward flowing BPS, commensurate with Yamauchi et al. [1998> and Sonnerup [1980>. ¿ 2000 American Geophysical Union