Page 174 - Green - Maritime Archaeology: A Technical Handbook. 2nd ed

Chapter 4: Conventional Survey 153
made prior to erecting the tower (thus avoiding tower interference with the tape measurements). The tower was then erected and measurements were made to each of the targets from the five different control points. This involved 15 intercontrol point measurements and 80 control point mea- surements (total of 95 measurements). The underwater recording took 58 minutes with about 15 minutes used to set up the control stakes. Tran- scribing and processing data took about 30 minutes. The resulting adjust- ment showed AR = 0.008 m and RMS = 0.018 m.
6. PhotoModeler Land
A standard, previously calibrated Olympus Camedia 4050 digital camera was used for the photography. The use of the PhotoModeler for land and underwater survey work has been described by Green et al. (2002). The only difficulty on land was the problem of achieving a high aspect view of the tower. Eight low oblique photographs were taken from ground level at 45° intervals around the tower. Using a tall step ladder a similar set of eight high oblique photographs was obtained. The whole recording operation took about 10 minutes. Photographs which gave good coverage of all the targets, were as far orthogonal as possible, and were both high and low oblique were then selected. Downloading data and processing took about 25 minutes. Results showed an average tightness (AT) of 0.009m with a standard deviation (SD) of 0.005m.
7. PhotoModeler Under Water
The same digital camera was mounted in an underwater housing with a water-corrected lens which had previously been calibrated (this system has been described by Green et al., 2002). Essentially the same set of pho- tographs that was taken under water (see Figure 4.28) was taken on land, but with the inclusion of a vertical view of the tower from above (not pos- sible to easily obtain on land). The whole recording operation took about 10 minutes. The only difficulty experienced under water was the problem with weed gathering around the base of the tower and obscuring the targets. Downloading data and processing took about 25 minutes. It was noted that target points close to the edge of the frame of the photograph tended to give large residuals. This is thought to be due to the calibration system. Results showed AT = 0.020 m and SD = 0.014 m.
8. HPASS
The application of HPASS has been described in Section VI above and by Green and Souter (2002). In these tests, the transponders were set up on