338 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1964 



Thereafter a number of simple, but relatively important refinements 

 were made in the program as the computations group of the Observa- 

 tory became more sophisticated in their approach. 



The second program was the subsatellite procedure, developed by 

 Dr. Luigi Jacchia, which provided a quick analysis of incoming ob- 

 servations. From each observation, a subsatellite point was computed 

 from a given set of orbital elements derived from the initial deter- 

 mination program. From the subsatellite points, the position and time 

 of the crossing of the ascending node were computed, as well as the 

 nearest perigee crossing. A plot of these quantities was sufficient to 

 tell whether the observation was good or bad. 



The subsatellite program could be used to predict all the modifica- 

 tions of the orbit. One had only to follow the position of the satellite ; 

 therefore, air drag could be determined as a byproduct of the program. 

 It w^as an empirical approach, but the modifications of the orbit were 

 observed; from these one could deduce theoretically the changes of 

 the orbit. Again, during the months that followed, the staff was to 

 make various improvements to this program. 



Usmg input from the subsatellite program, the ephemeris gave 

 the time of crossing of a satellite at various parallels — 10°, 20°, 30°, 

 etc. — with height, correction for time, angle of trajectory, and so forth, 

 so that an observer with a minimum amount of calculation could work 

 out fairly accurately the appearance of a satellite transit from his 

 particular position. This program was started shortly after Sputnik I 

 was launched, and became the basic prediction procedure for Moon- 

 watch teams and for people interested in making their own observa- 

 tions of the satellite. 



For the Baker-Nunn camera stations, however, a somewhat more 

 complex ephemeris was required. By early 1958, the basic program- 

 ing of the detailed station ephemeris was completed and debugging 

 was in process. Not until a year later, however, was the program fully 

 operational. By February of 1959, it had proved itself to be com- 

 pletely satisfactory and thereafter only minor refinements were made. 



Meanwhile, during the latter part of 1957 and continuing for sev- 

 eral years. Dr. Cunningham's major project was to develop a very 

 precise method of deriving, from the details of the equations of 

 motion, the position of a satellite as a function of time. This ap- 

 proach meant starting with an initial position of the satellite in terms 

 of its velocity and time. Then, by numerical integration, which 

 simply means step-by-step calculations using intervals of perhaps 

 one minute or less in time, the position of the satellite is computed. 

 This numerical integration program ref)resents a difficult procedure 

 if one wishes to carry out the calculations for, let us say, 10 days in 



