Home Table of Contents What's New Image Index Copyright ScienceViews Search


On Mars: Exploration of the Red Planet. 1958-1978

 
 
EVOLVING A CERTIFICATION PROCESS
 
 
 
[323] Along with radar and C site problems, site certification remained an open issue. In late May 1975, the Viking Project Office released one of the major products of the Landing Site Staff, a draft of the "Site A-1 Certification Procedure." This document described how the landing specialists would establish the acceptability of A-1 for landing and how they would....
 

 
During the 24 February 1975 landing site working group meeting, Len Tyler explains his complex radar studies of the Martian surface to (left to right) B. G. Lee, William Michael, Thomas Mutch, Don Anderson, Richard Shorthill, Gary Price, and Robert Hargraves.

 
[324]....recommended a target point that maximized the probability of a safe touchdown. Their recommendations would be based primarily on their analyses of low-altitude photographs taken during the first 10 orbiter revolutions. Key to the certification process would be the stereophotographic swaths of the A-1 site taken during the fourth and sixth revolutions - called P4 and P6 photos, as the revolutions were numbered from Viking's periapsides. After playback on video recorders, reconstruction, image processing, and enhancement, these photo frames, in the form of stereo pairs, would be analyzed at the U.S. Geological Survey's Flagstaff facility where ground elevation, slope angles. and surface roughness would be estimated. This information would be used with photo mosaics made from the orbiter frames to produce geologic maps of the proposed landing site region. Earth-based radar and telescopic observations, oblique and high-altitude photos, as well as Mars atmospheric-water detector and infrared thermal- mapping coverage of the landing site region, would provide supportive information about the nature of the surface. From this data base, landing site specialists would prepare a safety assessment report of the target zone.
 
During the site certification process, the Landing Site Staff would provide a series of recommendations. Just before the craft inserted into orbit of Mars, the team would decide either to execute the normal insertion maneuver and proceed with data acquisition or to modify the maneuver if a dust storm or some other anomaly were detected during approach. After playback, processing, and inspection of the imaging system frames from a pass over the A-1 site, it might be necessary to adjust the timing of the orbit or the pointing of the camera platform to obtain optimum coverage of the site on subsequent passes. Four such data-acquisition-adjustment opportunities were planned that would affect the camera sequences at or near P3, P4, P6, and P10. The Viking team would then have to answer the crucial question: would it land the craft or reject the site the team had selected? Recommendations would be made at three points before lander separation from the orbiter. A preliminary commitment to A-I would be made seven days before separation (at about P9), based on a preliminary assessment of available data. A firm commitment to land would be made three days later, and a precise target point would be established. A final commitment to land, made just before separation, would be determined after examining photos taken during the previous five days to confirm the absence of dust storms and high winds. 12
 
In the time that remained before the spacecraft reached Mars, the Landing Site Staff continued extensive preparations for completing site certification and lander release. In June and July, a functional test checked the ground-based hardware that would process photos from the orbiter and make the photomosaics and maps. The weakest link in the several-hundred-million-kilometer chain from Mars to the photo analysis labs in northern Arizona seemed to be the 850 kilometers the photographs traveled across the western U.S. Continental Trailways bus express, a leased army [325] aircraft and datafax were used to strengthen connection between Los Angeles and Flagstaff. The team did parts of the test a second time, verifying the readiness of the processing equipment and the personnel. 13
 
During the last months of 1975 and early 1976, the staff gave considerable attention to timing. Since so much depended on timely certification, scheduling became a paramount concern. The landing site specialists, working closely with the mission design team and the orbiter performance and analysis group, were ready by early February 1976 to test the timeline in what they called the ``SAMPD- l'' test, an exercise developed by B. Gentry Lee's Science Analysis and Mission Planning Directorate. 14