The Neural Density Model: An Open 3D Thermospheric Density Reanalysis for 2003--2026 From Tracked Space Debris
Abstract
Thermospheric mass density controls satellite drag but is observed directly by few missions, and the leading operational product (HASDM) relies on non-public radar data. We present the Neural Density Model (NDM), a 3D density reanalysis for October 2003--October 2026 at 250--800~km. Previous approaches either convert the orbital decay of tracked objects into point densities or correct empirical models with a few global scale factors. NDM instead fits a spatially resolved neural correction to NRLMSISE-00 directly to the observed decay of 600--800 objects per year, through a differentiable drag operator, jointly with their ballistic coefficients, calibration spheres and accelerometer densities. On withheld satellites the median error is 0.078 (natural-log units), against 0.101 for HASDM and 0.310 for NRLMSISE-00; NDM has the smaller bias, HASDM slightly smaller scatter. On Spire GNSS orbits, never used in training, NDM and HASDM perform comparably. NDM uses only public observations.
Related
- Impact of geomagnetic storm intensity on thermospheric density and orbital decay of LEO satellites: a G1–G5 per-event analysis
- Thermospheric response to solar EUV variability probed with orbital-energy decay from NinjaSat CubeSat GNSS telemetry
- HD 152384: A 5-megayear-old debris disk system showing a gaseous ring of refractory elements at the dust sublimation radius
- Production of White Dwarf Debris Disks from Planetary Collisions and Poynting-Robertson Drag
- Efficient Identification and Analysis of Space debris Trails in Time-Domain Astronomy
BibTeX
@misc{camporeale2026neural,
title = {{The Neural Density Model: An Open 3D Thermospheric Density Reanalysis for 2003--2026 From Tracked Space Debris}},
author = {Enrico Camporeale},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.02103},
eprint = {2611.02103},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.02103},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - The Neural Density Model: An Open 3D Thermospheric Density Reanalysis for 2003--2026 From Tracked Space Debris
AU - Enrico Camporeale
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.02103
UR - https://arxiv.org/abs/2611.02103
AB - Thermospheric mass density controls satellite drag but is observed directly by few missions, and the leading operational product (HASDM) relies on non-public radar data. We present the Neural Density Model (NDM), a 3D density reanalysis for October 2003--October 2026 at 250--800~km. Previous approaches either convert the orbital decay of tracked objects into point densities or correct empirical models with a few global scale factors. NDM instead fits a spatially resolved neural correction to NRLMSISE-00 directly to the observed decay of 600--800 objects per year, through a differentiable drag operator, jointly with their ballistic coefficients, calibration spheres and accelerometer densities. On withheld satellites the median error is 0.078 (natural-log units), against 0.101 for HASDM and 0.310 for NRLMSISE-00; NDM has the smaller bias, HASDM slightly smaller scatter. On Spire GNSS orbits, never used in training, NDM and HASDM perform comparably. NDM uses only public observations.
ER -