A method for computing accurate daily atmospheric density values from satellite drag data

A method has been developed for computing accurate daily density values based on satellite drag data. A differential orbit correction program using special perturbations orbit integration is used to fit radar and optical observational data to obtain the standard 6-element state vector plus a ballistic coefficient (B). The atmospheric density model used in the integration is a modified Jacchia 1970 model that was developed for incorporation into the USAF High Accuracy Satellite Drag Model (HASDM). Energy dissipation rates (EDR) are computed over the observation span (3 to 8 days) using the modified Jacchia model density values and the fitted B value. Overlapping EDR values are obtained by moving the start of each succeeding orbit fit by one day. These overlapping fits are then used to compute accurate daily EDR values. Daily temperature and density values are then computed from the observed EDR values using the "true" 30-year B value of each satellite plus using partial derivatives of temperature and EDR changes computed by the HASDM density model. The daily temperature values were validated by computing the daily values for satellites during year 2001, and comparing the results obtained using the HASDM Dynamic Calibration Atmosphere (DCA) program. The comparison is excellent between the current method described herein and the HASDM DCA method. Finally, the daily density computation was validated by comparing historical daily density values computed for the last 30 years for over 25 satellites. The accuracy of the density values was determined from comparisons of geographically overlapping perigee location data, with over 8500 pairs of density values used in the comparisons. The density errors are less than 4% overall, with errors on the order of 2% for values covering the latest solar maximum.