The aim of the 1999 project was to produce a scale drawing of the structure remaining in Alum Bay.  It was decided that the plan should be:

Students from University College London (UCL) had started drawing the site to scale earlier in the year, whilst this was continued during the project it was not completed. During the following winter, an attempt was made to trace the completed drawing from the assembled sketches and a reasonable drawing was produced. However, problems were encountered in positioning some of the individually drawn metre squares.

Different groups of divers over time have used markers ranging from cow tags to garden seed markers and even gaffer tape during successive years work on this site. It was decided early on that nothing should be removed or replaced, and therefore the team inherited a number of datum points which were imprecise as to the reason for their location and clustered in certain areas. A difficulty encountered when measuring from such points was that it was not always possible to determine the "point" to which measurements should be made. Had the survey been started from new at this time, a more coherent system could have been set up, using nails, crayon crosses or white tack blobs to mark the points to measure to.  For the purpose of the drawing, temporary datum points consisting of garden tags were positioned to allow orientation of isolated structures.

Equipment consisted of fibreglass tape measures, plastic slates, pencils, permatrace paper and metre planning frames. Several types of frame were used during the project, made out of a variety of materials ranging from plastic to brass piping. Problems that were encountered such as difficulty positioning the frames over raised structures and positive buoyancy were solved by use of frames made out of brass piping with adjustable screw legs. Thin nylon cord was used to double string the frames so that the diver could identify a position immediately above the object that was being drawn eliminating parallax.

Records were made on permatrace sheets showing two 100 mm square grids or 10mm square graph paper that facilitated recording of the structure on a scale of 1 to 10. The larger areas of paper allowed greater areas to be recorded during each dive.

Planning

Two tape measures were run the length of the site between control points W99 and X99 and Y99 and Z99. Temporary tape measures were positioned between these tapes to provide a reference grid. Planning frames were positioned along the fixed tape measures and in line within the temporary tapes and moved after drawing had been completed to ensure an overlap of at least 100 mm. The position of wreck detail points was recorded on the drawings.

Measurement and data input

Distance measurements between points were recorded and entered into the Site Surveyor program. Depth measurements were recorded, adjusted for the effects of rise and fall of tide and were also entered into the program.

Initially, measurements from wreck detail points to other datum points and to control points, as well as control points to control points were entered in the order in which they were measured. This produced a complicated plot that computer programs found difficult to resolve.  Ultimately, the data was entered systematically.  The control network was measured first to itself, this was then ‘fixed’ when a minimal error plot had been achieved, then the measurements from each of the wreck detail points to the fixed control points were added later. The detail points could be isolated from the control points both in terms of calculation and graphically.

This is possibly the most significant departure from techniques used previously, in that the control points were fixed on the computer. Thereafter, all detail points were measured (and so fixed) to the control points, not to other detail points.

Plotting

Co-ordinates from Site Surveyor were plotted onto graph paper, initially by hand, but ultimately by transfer of the information to the drawing program AutoCAD.  Once the position of the wreck detail points had been fixed in this way, the planned squares could be positioned relatively accurately by aligning the detail points marked by divers on the drawing to the detail point plotted on the graph paper.

The time spent surveying this site over the years has been considerable. Due to the repositioning and enlarging of the control network, only limited use could be made of the data collected in previous years. It is estimated that approximately 75% of the drawing has been completed this year and all the measurements have been taken or checked. A total of 30 hours has been spent drawing underwater and 40 hours measuring underwater. The first attempt at completing the tracing took 72 hours but the final drawing and data processing on land took only 32 hours, primarily due to the use of the techniques stated and the use of technology.

This exercise has demonstrated that surveying without a control network has inherent problems. While this may seem obvious, it is not something that volunteer or avocational divers will necessarily be aware of. The fact that a control network must surround the site and that it must be positioned before the artifacts are positioned is not well understood by divers who have limited archaeology or survey experience. Nor are the disadvantages of measuring between artifacts themselves really understood.