eft \ and 5. Another significant pointe 1s that west of tre £. somsidie almesphere occurs at Maca mobers SDereeen Last 3.0 if is zricuriiy ffl isne cachnigue. swever, because ‘a tte accuragcs of Chis approximation wsicn Jctermines Che eccersecs oF tte tices F gta moter ani walinegrtts cf tte wariablas snich determine trajectory, mo sisple step exists for soins trea Cte known accuracy of tte soove approxisaticn to an estiratien of tre accuracy of the tecnnicce in teras of erfece on tae Crajectom. The saajer effort of tre trajectory analysis was devoted co estisacing this over-all accuracy, The thrust phases of a trajectory are included bs considering cnis tho c they change the horizontal and vertical velocity components, ant chat the magnitude of these chanses is independezt of the value of the velocity cons penencs. altncugh not in reality correct, this statesenc aay convenientls be assuzed to be valid by megleceting drag uring thrusting and using 2 pocuto thrust tnuset will yield casired vaiccity changes. analysis only, 6.3.5 This approach is used for not for trajectory conputations. Trajectory Coapitaticn Uncil this tise, the effort had been devored only to the untracaed wehicles, since trajectorics of Che teo tracked vehicles sere assunsl fo be known. However, agptication of the time-vt-rlight technique to the tev tracked vehicles indicated thac these trajectories were serfcusly in error. AC the time of the saove to Johnston Island, it was realized chat raphy did not perait installacion of a satisfactory MDOT sssten. local geo: Althoush seme doubts were raised at the rine about che performance of MIDUT under th e restricted geometry, Chere was litcle chvice but to use it. The »3.fe7 was used co track tne aain carrier so that ic couid be correlated with DOWA? "Doppler determination of velocit; am! position), which was the privary tracking ssveten for che win carrier. favorably with DWAP. In this correlation, MOOT cocpured ven The MIDOT trajectories cbtained on fifteen of tne Deorkneb venicles also looked vers sant wnen compared with tnevretical trajecturies. (There was a discrepancs en the sixtcenth reend anich Aas »ub- aeguentl. been ressived.) le was prinaril. Never been guestioned on Adi-lo. for tnese reasons tmut YD. aod However, there eas a basic difference Detecen tne dielo venicles amd the other roevets: the veher ucnies aod a much bette c antenna pattera chan did the di-Lo crits. The por antensa pattern cotbined