An Empirical Density Model for Predicting Low-Altitude Satellite Ephemerides. Part 1. Data Analysis and Model Formulation.

An empirical density model based on drag analyses of 25 low-perigee satellites is developed for use in predicting low-altitude satellite ephemerides. A stepwise multiple regression analysis is performed with density at 145 km as the dependent variable, and a number of independent variables chosen to represent variations with solar cycle, geomagnetic activity, geomagnetic latitude, season, day of the year, and time of day. Densities between 120 and 500 km are determined from the hydrostatic law in a simple (but physically realistic) analytic form. Comparison of our model with the 1971 Jacchia model shows them both to give very similar variations, differing mostly with respect to solar activity and geomagnetic latitude. The main virtue of our model is that it is represented in an extremely simple analytic form, and therefore ideally suited for implementation in an operational system. (Author)