Validation of Auroral Oval Models Using DMSP SSUSI

A study was performed to determine the accuracy of current operationally-ready auroral oval boundary models to assist in the determination of the quality of various approaches to auroral oval boundary specification. Four models were evaluated in this study: OVATION Prime, OVATION, Hardy and an ensemble based on the combination of these models. These auroral oval boundary models were compared to the Defense Meteorological Satellite Program (DMSP) Special Sensor Ultraviolet Spectrographic Imager (SSUSI) Auroral Boundary Environmental Data Record (EDR). Results of the study showed the solar wind-driven OVATION Prime model produced the most consistent results compared to the SSUSI data. The magnetometer-driven Hardy model produced the next most consistent results. The in-situ measurement/climatology/look-up table model, OVATION, produced the least consistent results. In terms of magnitude of difference, the Hardy model had the lowest mean error but the variability of the error was quite large (1 to 5 degrees). The OVATION Prime model had the next lowest mean error with a much more consistent peak in error near 5 degrees in geomagnetic latitude. The OVATION model had a wide spread of error (1 to 8 degrees). The ensemble model developed for this study performed very well. Most of the correlation coefficients for the ensemble model equatorward boundary were near .8 with mean absolute errors reduced to 1 degree in geomagnetic latitude.