STK's Simplified General Perturbations No. 4 (SGP4) propagator is used with two-line mean element (TLE) sets to propagate a satellite's orbit over time. The TLE sets used are maintained and updated in the space catalog, which has been maintained by various military organizations at the Cheyenne Mountain Complex in Colorado Springs, CO, since 1965.
The main product of the space catalog is the maintenance of the orbital element sets. The original analytic theory, Simplified General Perturbations (SGP), was developed by Aeronutronic-Ford. In 1965, Max Lane began developing a slightly different analytic theory. His work, along with contributions by Ken Cranford, resulted in the Simplified General Perturbations Theory No. 4 (SGP4). In the early 1970s, the original SGP analytic theory was replaced by a version of the Air Force General Perturbations Theory No. 4.
The main objective of Lane, Cranford and other scientists of the 14th Aerospace Force was to develop an analytic orbit theory that provided better orbit determinations and predictions for high drag satellites, yet did not significantly increase computer time requirements. While the new theory was implemented into the Cheyenne Mountain Complex, operational requirements prevented it from being implemented at that time at any of the space surveillance/missile warning sensor sites due to operational issues. To resolve the astrodynamic compatibility issue, pseudo SGP elements were developed to maintain compatibility with the SGP orbital theory.
In the mid- to late 1970s, the SGP4 was modified to address deep space requirements. The incorporation of deep space algorithms into the SGP4 was developed primarily by Air Force Captain Bruce Bowman and Richard Hujsak of the 14th Aerospace Force/Air Defense Command/NORAD. The current SGP4 propagator, and then, is really an SGP4/DP4 propagator.
While sensor sites and most military users have changed to SGP4, the practice of providing element set data that can be used in either SGP or SGP4 has been preserved. The primary difference between the two element sets is the formulation of mean motion and the atmospheric drag representation, i.e., the SGP is a Kozai-based theory while SGP4 is a Brouwer-based theory. Due to the operational requirements, the Kozai mean motion is the standard for TLE orbital products. You will normally find a zero (0) for ephemeris type on the first line of the TLE set product but if a two (2) should ever appear, that indicates the mean motion is the Brouwer formulation and does not need to be converted for use in SGP4.
Beginning with STK 9, the SGP4 algorithm used internally by STK has been switched to use the algorithm provided by CSSI, available at www.centerforspace.com. In most instances, the differences in positions obtained using different SGP4 algorithms are small and have no information content; the advantage of the CSSI routine is that the algorithm descriptions, comparisons to other routines, validation tests, and even the source code are freely available.