PROSERPINA: An Automated Pipeline for Martian Ionospheric Radio-occultation Retrievals with Probabilistic Uncertainty and Spherical-symmetry Bias Quantification
- 1Boston University
- 2California Institute of Technology
- 3Johns Hopkins University
Abstract
Spacecraft-to-Earth radio occultation (RO) experiments are a powerful tool for probing planetary atmospheres and ionospheres. However, standard processing typically requires manual intervention and assumes spherical symmetry. This assumption is rarely tested, and its impact on retrieved electron density profiles is not often quantified. Here we present Processing Radio Occultations System for Examining and Reconstructing Planetary Ionised and Neutral Atmospheres (PROSERPINA), a fully automated Python pipeline designed to retrieve electron density profiles from one-way radio occultation observations at Mars. The pipeline also introduces Forward–Backward Ionospheric Density Estimation System (FIDES), a module that quantifies the bias introduced by the spherical-symmetry assumption in Abel inversion. FIDES generates synthetic two-dimensional ionospheres anchored to the retrieved Abel peak and analyzes the difference between FIDES and Abel results. This difference is decomposed into amplitude, peak-altitude, and width components, producing a structured model-error covariance for each profile. The Abel-retrieved profile remains the primary estimate. The reported 95% credible intervals incorporate contributions from thermal noise, uncertainty in the baseline slope, and the model uncertainty derived from FIDES. We validate PROSERPINA using Mars Global Surveyor (MGS) electron density profiles archived and processed independently in the NASA Planetary Data System and apply it to 100 previously unprocessed MGS profiles, of which 20 are fully shown here. Across these cases, the uncertainty associated with the spherical-symmetry assumption dominates over baseline-slope uncertainty and thermal noise for high solar zenith angle dayside M2 peaks, contributing a median uncertainty of about 30% of the peak density.
Related
- Interpretable human-label-free deep learning for real-bogus classification with uncertainty quantification
- Uncertainty-Aware Tidal Disruption Event Classification : A Host-Agnostic Probabilistic Random Forest Approach
- Uncertainty-Aware Tidal Disruption Event Classification : A Host-Agnostic Probabilistic Random Forest Approach
- Size, atmospheric, and ring constraints on the large Kuiper belt object (55565) Aya from a stellar occultation
- Occultation polarimetry as a probe of unresolved polarization
BibTeX
@article{felici2026proserpina,
title = {{PROSERPINA: An Automated Pipeline for Martian Ionospheric Radio-occultation Retrievals with Probabilistic Uncertainty and Spherical-symmetry Bias Quantification}},
author = {Marianna Felici and Yu-Ming (Oscar) Yang and Luke Moore and Jennifer Segale and Joshua Semeter and Prakash Ishwar and Paul Withers and Kamal Oudrhiri},
journal = {The Planetary Science Journal},
year = {2026},
month = sep,
doi = {10.3847/psj/aea408},
url = {https://iopscience.iop.org/article/10.3847/PSJ/aea408},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - PROSERPINA: An Automated Pipeline for Martian Ionospheric Radio-occultation Retrievals with Probabilistic Uncertainty and Spherical-symmetry Bias Quantification
AU - Marianna Felici
AU - Yu-Ming (Oscar) Yang
AU - Luke Moore
AU - Jennifer Segale
AU - Joshua Semeter
AU - Prakash Ishwar
AU - Paul Withers
AU - Kamal Oudrhiri
PY - 2026
DA - 2026/09/30
JO - The Planetary Science Journal
DO - 10.3847/psj/aea408
UR - https://iopscience.iop.org/article/10.3847/PSJ/aea408
AB - Spacecraft-to-Earth radio occultation (RO) experiments are a powerful tool for probing planetary atmospheres and ionospheres. However, standard processing typically requires manual intervention and assumes spherical symmetry. This assumption is rarely tested, and its impact on retrieved electron density profiles is not often quantified. Here we present Processing Radio Occultations System for Examining and Reconstructing Planetary Ionised and Neutral Atmospheres (PROSERPINA), a fully automated Python pipeline designed to retrieve electron density profiles from one-way radio occultation observations at Mars. The pipeline also introduces Forward–Backward Ionospheric Density Estimation System (FIDES), a module that quantifies the bias introduced by the spherical-symmetry assumption in Abel inversion. FIDES generates synthetic two-dimensional ionospheres anchored to the retrieved Abel peak and analyzes the difference between FIDES and Abel results. This difference is decomposed into amplitude, peak-altitude, and width components, producing a structured model-error covariance for each profile. The Abel-retrieved profile remains the primary estimate. The reported 95% credible intervals incorporate contributions from thermal noise, uncertainty in the baseline slope, and the model uncertainty derived from FIDES. We validate PROSERPINA using Mars Global Surveyor (MGS) electron density profiles archived and processed independently in the NASA Planetary Data System and apply it to 100 previously unprocessed MGS profiles, of which 20 are fully shown here. Across these cases, the uncertainty associated with the spherical-symmetry assumption dominates over baseline-slope uncertainty and thermal noise for high solar zenith angle dayside M2 peaks, contributing a median uncertainty of about 30% of the peak density.
SN - 2632-3338
ER -