Physics-Informed Multi-Task Surrogate Model for the Martian Nightside Thermosphere
Abstract
Modeling the Martian nightside thermosphere remains challenging due to sparse in situ sampling and strong coupling among transport, magnetic, and seasonal processes. Purely data-driven models can produce non-physical artifacts, such as density inversions, in poorly sampled altitude regimes. We present a multi-task physics-informed neural network that simultaneously predicts the base-10 logarithmic densities of four neutral species (O, CO2, N2, and Ar) using more than a decade of MAVEN/NGIMS observations (MY 32-38, 2014-2025). A shared backbone learns a common representation of the nightside thermospheric state and branches into species-specific output heads. A weak monotonicity prior is incorporated via automatic differentiation by penalizing positive vertical gradients in logarithmic density. Experiments using an orbit-disjoint train/validation/test split show that physics-informed regularization substantially reduces non-physical inversions while preserving predictive skill and slightly improving it in the best-performing configuration, as measured by RMSE, MAE, and R2. The resulting model provides a computationally efficient surrogate for nightside thermospheric reconstruction with improved vertical consistency.
Related
- Modeling the Nightside Martian Ionosphere: Structure, Composition, and Energetics
- Metaheuristic physics-informed extreme-learning machine for initial orbit determination of near-Earth asteroids from too-short arcs
- Energetic heavy ions in Jupiter's nightside inner and middle magnetosphere: Juno/JEDI statistics
- Reconstructing the generalized Barrow holographic dark energy with physics-informed neural networks
- Physics Informed Neural Network for Multi-dimensional Turbulent Self-gravitating Flows
BibTeX
@misc{nikiforov2026physicsinformed,
title = {{Physics-Informed Multi-Task Surrogate Model for the Martian Nightside Thermosphere}},
author = {Sergey Nikiforov},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.10077},
eprint = {2609.10077},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.10077},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Physics-Informed Multi-Task Surrogate Model for the Martian Nightside Thermosphere
AU - Sergey Nikiforov
PY - 2026
DA - 2026/09/09
PB - arXiv
DO - 10.48550/arxiv.2609.10077
UR - https://arxiv.org/abs/2609.10077
AB - Modeling the Martian nightside thermosphere remains challenging due to sparse in situ sampling and strong coupling among transport, magnetic, and seasonal processes. Purely data-driven models can produce non-physical artifacts, such as density inversions, in poorly sampled altitude regimes. We present a multi-task physics-informed neural network that simultaneously predicts the base-10 logarithmic densities of four neutral species (O, CO_2, N_2, and Ar) using more than a decade of MAVEN/NGIMS observations (MY 32-38, 2014-2025). A shared backbone learns a common representation of the nightside thermospheric state and branches into species-specific output heads. A weak monotonicity prior is incorporated via automatic differentiation by penalizing positive vertical gradients in logarithmic density. Experiments using an orbit-disjoint train/validation/test split show that physics-informed regularization substantially reduces non-physical inversions while preserving predictive skill and slightly improving it in the best-performing configuration, as measured by RMSE, MAE, and R^2. The resulting model provides a computationally efficient surrogate for nightside thermospheric reconstruction with improved vertical consistency.
ER -