PDS_VERSION_ID        = PDS3                                                  
RECORD_TYPE           = STREAM                                                
OBJECT                = TEXT                                                  
  PUBLICATION_DATE      =  " "                                                
  NOTE                        = "Experiment description for the               
                                 Experiments at Earth1 and Earth2 Gravity     
                                 Assists conducted in 1990 (DOY 333 through   
                                 DOY 344 for Earth1), and in 1992 (DOY 229    
                                 through DOY 346 for Earth2). Formatted for   
                                 display or printing with up to 78 constant-  
                                 width characters per line."                  
END_OBJECT            = TEXT                                                  
END                                                                           
                                                                              
                                                                              
Coherent Doppler and ranging data were generated by the DSN for the two       
Galileo gravity assists during Earth flybys on 8 December 1990 (Earth1) and   
8 December 1992 (Earth2).  The original scientific objective of using these   
data was to improve the determination of Earth's mass ME (M sub E) in the     
form GME (GM sub E), where G is the laboratory determined gravitational       
constant.  However during the data analysis phase, it was discovered that it  
is not possible to fit the data at Earth1 without somehow accounting for the  
fact that the spacecraft picked up orbital energy during the flyby.  For      
example, a good fit can be obtained by introducing a fictitious instantaneous 
velocity increase of about 4 mm s^(-1) at perigee along the direction of      
motion.  Numerous data fits using various subsets of Doppler and ranging data,
sometimes both range and Doppler, sometimes Doppler only, and sometimes       
ranging only, all point to the same conclusion; the outgoing Galileo          
trajectory has more orbital energy than the incoming trajectory.  This is     
theoretically impossible in a conservative gravitational field, even taking   
into account the noninertial dynamical situation introduced by longitudinal   
gravity harmonics undergoing Earth rotation.  In an attempt to understand     
this anomalous behavior, the Tracking and Data Relay Satellite (TDRSS) was    
successfully used to track the Galileo spacecraft during Earth2.  This filled 
in an unaviodable gap in DSN coverage for about two hours surrounding         
perigee.  The similar gap at Earth1 amounted to about 1.2 hours.              
	Fits to a combination of DSN data and TDRSS data reveal that there    
are significant systematic effects in the post-fit TDRSS residuals.  However, 
the lower altiude for Earth2 (303 km as opposed to 960 km for Earth1), and    
the considerable uncertainty in atmospheric drag models at the lower altitude,
make it impossible to determine if an orbital energy increase indeed occurred 
during the Earth2 flyby.  Also, by leaving the TDRSS data out of the fit, and 
by including an atmospheric drag parameter, it is possible to fit the Earth2  
DSN data to the noise level.  Hence it is not known whether the difficulty in 
fitting the TDRSS data is caused by unmodeled physical effects or by flawed   
TDRSS data.  As expected, the inclusion of a drag paramenter in the fit to    
the Earth1 DSN data results in a non-physical positive drag acceleration along
the direction of motion.  Based on numerous numerical tests, the possibility  
of software bugs is deemed extremely unlikely for both the Earth1 and Earth2  
fits.