Eric Sutton, University of Colorado / SWx TREC; Marcin Pilinski, University of Colorado at Boulder / Laboratory for Atmospheric and Space Physics; Shaylah Mutschler, University of Colorado Boulder; Jeffrey Thayer, University of Colorado; Thomas Berger, University of Colorado / Space Weather Technology, Research, and Education Center (SWx TREC); Vu Nguyen, Spire Global, Inc.; Dallas Masters, Spire Global, Inc.
Keywords: Thermosphere models, Satellite drag, Space Weather, Physics-based, STM
Abstract:
Within low-Earth orbit (LEO), a region spanning roughly 100 to 1000 km in altitude, interactions between satellites and the ambient atmosphere cause large uncertainties in the orbit determination and prediction processes. During episodic periods of severe space weather activity, such atmospheric drag uncertainties can amplify by a factor of 2-5 in a matter of minutes to hours. These uncertainties, when combined with the steadily growing launch rate for small satellites and CubeSats and our advancing abilities to track smaller and smaller objects, are poised to overwhelm the DOD infrastructure currently executing the Detect—Track—Catalog mission. Products of this mission are pervasive across SSA, forming the foundation of nearly all space-based activities. Thus, knowledge and prediction of the space environment, particularly the neutral mass density of the thermosphere and lower exosphere, are essential for operating satellites within LEO.
One of the obstacles in predicting orbit trajectories hours to days in advance and in correlating consecutive or irregular object tracks comes from the legacy framework used to model the upper atmosphere’s state and its interaction with satellites and debris. The current model employed by the Combined Space Operations Center (CSpOC) is the High Accuracy Satellite Drag Model (HASDM), an empirical model that self-calibrates by ingesting ground-based tracking data of a select set of orbiting “calibration objects”—i.e., operational and defunct satellites passing through LEO with reasonably stable ballistic coefficients. While this method provides a global-average picture of the upper atmosphere, its ability to forecast or even capture realistic spatial structure is limited. Physics-based upper atmosphere modeling approaches offer a vast potential improvement. Models in this category solve a set of Navier-Stokes fluid equations appropriately tailored to the upper atmosphere, and therefore are inherently better equipped for simulating a dynamic system response to impulsive energy input from the solar wind. The primary reason such physics-based methods have not been adopted to date is the lack of robust data assimilation schemes capable of self-calibrating at a level equal to or better than empirical models.
Fortunately, significant strides have been made over the last 2 years toward supplanting empirical methods with physics-based data assimilation models of the upper atmosphere. This has been accomplished by accounting for the complex driver/response characteristics of the upper atmosphere in a new least-squares filter, similar in function to an unscented Kalman filter (UKF). Using this new technique, much better spatial accuracy has already been demonstrated, which can help to lower uncertainty across the LEO catalog and increase the efficiency of Space Traffic Management (STM) activities. In addition, the emergence of large constellations of commercial and academic CubeSats in the past 3 to 5 years also brings with it an excellent opportunity. Because many CubeSats are equipped with GNSS devices, they are excellent sources of Precision Orbit Determination (POD) data that can be used to initialize and constrain these upper atmosphere models. When compared with the conventional observations from ground-based radar tracks of known objects, satellite-based GNSS observations can describe the space environment at a much higher spatial resolution and temporal cadence. This talk will summarize our efforts to drive the new physics-based data assimilation technique with information from CubeSats.
Date of Conference: September 15-18, 2020
Track: Atmospherics/Space Weather