[1] In the polar atmosphere, significant chemical and ionization changes occur during solar proton events (SPEs). The access of solar protons to this region is limited by the dynamically changing geomagnetic field. In this study, we have used riometer absorption observations to investigate the accuracy of a model to predict K p -dependent geomagnetic rigidity cutoffs, and hence the changing proton fluxes. The imaging riometer at Halley, Antarctica is ideally situated for such a study, as the rigidity cutoff sweeps back and forth across the instrument's field of view, providing a severe test of the rigidity cutoff model. Using observations from this riometer during five solar proton events, we have confirmed the basic accuracy of this rigidity model. However, we find that the model can be improved by setting a lower K p limit (i.e., K p = 5 instead of 6) at which the rigidity modeling saturates. We also find that, for L > 4.5, the apparent L-shell of the beam moves equatorward. In addition, the Sodankylä Ion and Neutral Chemistry (SIC) model is used to determine an empirical relationship between integral proton precipitation fluxes and nighttime ionosphere riometer absorption, in order to allow consideration of wintertime SPEs. We find that during the nighttime, the proton flux energy threshold is lowered to include protons with energies of >5 MeV in comparison with >10 MeV for the daytime empirical relationships. In addition, we provide an indication of the southern and northern geographic regions inside which SPEs play a role in modifying the neutral chemistry of the stratosphere and mesosphere.Citation: Clilverd, M. A., C. J. Rodger, T. Moffat-Griffin, and P. T. Verronen (2007), Improved dynamic geomagnetic rigidity cutoff modeling: Testing predictive accuracy,