W. Kent Tobiska, Space Environment Technologies; Bruce Bowman, Space Environment Technologies; Marcin Pilinski, Space Environment Technologies; Piyush Mehta, West Virginia University; Richard Licata, West Virginia University
Keywords: HASDM, space weather, machine-learning, atmosphere density, forecasting
Abstract:
The number of Low Earth Orbit (LEO) objects will TRIPLE in the next 2 years and collisional hazards will increase. The Machine learning Enabled Thermosphere Advanced by HASDM (META-HASDM) system is a collaborative project between Space Environment Technologies and West Virginia University. It will significantly reduce uncertainty in thermospheric density specification and will improve conjunction assessment as well as operational global space traffic management. META-HASDM, in particular, will aid with Space Weather Forecasting Technologies and Techniques by providing:
information for scientific and operational use via new machine learning (ML) algorithms;
absolute atmosphere density at HASDM’s current 210% uncertainty algorithmically;
predicted values for outside HASDM’s historic time period;
improved LEO ballistic coefficients above 500 km;
dynamic uncertainties for HASDM, JB2008, and forecast drivers; and
improved forecasts for solar and geomagnetic indices.
In addition, META-HASDM already provides a new Space Weather Benchmark with:
the two solar cycle SET HASDM density database that has been released publicly at https://spacewx.com/hasdm/and
accuracy, time resolution, and global scale, where no comparable dataset currently exists.
We will report on the progress of META-HASDM in this presentation.
Date of Conference: September 14-17, 2021
Track: Atmospherics/Space Weather