The Use of Radar Altimetry in the Mapping of a Reservoir
by P. G. Mott
In July 1959 Hunting Surveys Ltd. were commissioned to produce a 1:50, 000
scale map with 20 ft, contours to cover a projected reservoir area of about
1,500 sq. miles on the Niger above Jebba; the urgency of the scheme called for the
maps to be delivered to the consulting engineers by February 1960. Photography
of the area was precluded by the rains until September, and levelling for height
control by normal ground methods was likely to extend the field programme until
February, the date when the maps were due for delivery. It was decided that the
only way to produce the maps on time was to base the height control for photogram-
metric mapping on data supplied by the Airborne Profile Recorder.
A plan was adopted whereby each strip of photographs was supplied with three
APR profiles; the central line flown simultaneously with the survey photography at
20,000 ft., while purely APR lines were flown along the lateral overlap of the
photo cover at heights between 5,000 and 10,000 ft., to suit prevailing weather.
The APR pattern was 'held' by tie strips, controlled by ground levelling.
Subsequent analysis of the APR records showed that the "error spread" from
the central lines flown at 20, 000 ft. was much greater than that from the lateral
lines flown below 10,000 ft., and it was therefore decided to discard the high level
record, and set the overlap in the plotting machines solely from the low level APR
records in the lateral overlaps of the survey photography. No aerotriangulation was
required and the work of plotting topographical detail and contours was divided among
a number of Wild A. 8 plotters.
The following facts are worth noting -
1. Ground levelling to control the APR profiles took 38 man-weeks compared with
the estimated 170 man-weeks need to provide normal height control for
photogrammetric mapping.
2. The contoured maps of the whole reservoir area were delivered within three
months of completion of photography and APR flying.
3. The cost of the survey with APR control showed a saving of about 23% over the
estimated cost of the job without APR.
4, 80% of the 192 APR points checked against ground levelling showed no error
greater than 5 ft., and 100% were within 10 ft.
5. The main lesson shown by the survey is that in practice the accuracy of the APR
falls off rapidly with increasing altitude.
6. Amajoreconomy is effected, both intime and expense during the photogrammetric
stage, by flying the APR control along the lateraloverlaps instead of downthe centre
of the runs, thereby eliminating the requirements for aerotriangulation,