Empirical radiation belt models: Comparison with in-situ data and implications for environment definition

The empirical AP8/AE8 model has been the de-facto Earth\textquoterights radiation belts engineering reference for decades. The need from the community for a better model incubated the development of AP9/AE9/SPM, which addresses several shortcomings of the old model. We provide additional validation of AP9/AE9 by comparing in-situ electron and proton data from Jason-2, POES, and the Van Allen Probes spacecraft with the 5th, 50th, and 95th percentiles from AE9/AP9 and with the model outputs from AE8/AP8. The relatively short duration of Van Allen Probes and Jason-2 missions means that their measurements are most certainly the result of specific climatological conditions. In LEO, the Jason-2 proton flux is better reproduced by AP8 compared to AP9, while the POES electron data are well enveloped by AE9 5th and 95th percentiles. The shape of the SAA from Jason-2 data is better captured by AP9 compared to AP8, while the peak SAA flux is better reproduced by AP8. The <1.5 MeV inner belt electrons from MagEIS are well enveloped by AE9 5th and 95th percentiles while AE8 over-predicts the measurements. In the outer radiation belt, MagEIS and REPT electrons closely follow the median estimate from AE9, while AP9 5th and 95th percentiles generally envelope REPT proton measurements in the inner belt and slot regions. While AE9/AP9 offer the flexibility to specify the environment with different confidence levels, the dose and trapped proton peak flux for POES and Jason-2 trajectories from the AE9/AP9 50th percentile and above are larger than the estimates from the AE8/AP8 models.