The Location of the Wrecksite
In brief the sequence of events leading to the location of relics that have since proved an indisputable pointer to the location of William Dampier’s ship are:
On the morning of the 15th March, 2001, the day after arrival on the Island, the Museum team proceeded in the RAF fishing/diving boat Ascension Frigate skippered by our liaison officer Flt Lt Richard Burke RAF for a familiarization dive of the waters in the region.
There, an indication was had of the sea conditions, of the underwater terrain, of the fish life and (very importantly) of the appearance of wreckage and other detritus in Ascension Island waters. This was essential, for initially (and quite incorrectly) the team had expected substantial coral growth over wreckage, as was common in the Indo-Pacific region.
Later in the day, the group then sailed to the position recorded by Wm. Dampier on 24th February 1701 (Julian Calendar), while his ship was in a sinking state. Winds were offshore, seas were slight, on a low swell.
Journals and the other accounts of the loss of the ship were read to those onboard while Ascension Frigate was stationed as close as possible at Roebuck’s first anchorage in 10 fathoms of water. An exact replication of the positions given by Dampier i.e. ‘… in ten fathom and half water, sandy ground about half a mile from the shoare, the S. point of the bay bore S.S.W. dist. one mile and a half and the northernmost point, N.E.1/2 N.dist. two mile…’ was not possible, it being occupied on the occasion of our visit by the Island’s Tanker, the Maersk Gannet.
While there pondering the situation and with an offshore (predominantly easterly) wind, a small ‘willy-willy’ (circular localised wind gust) was seen onshore heralding the temporary change of wind direction that is customary around midday. This had been predicted by our guide Richard Burke, as a regular, but short-lived daily phenomenon. On this occasion it translated into a rare sustained sea-breeze as per the events that befell Dampier 300 years ago.
With these fortuitous events, the team were able to deduce with an element of surety Dampier’s next course of action in the minutes that would have elapsed as he recognised the opportunity and ordered his men to the windlass and aloft. Then with time running out as the anchor was raised and the sails were unfurled and filled, there was just enough of the breeze remaining to allow him to run into seven fathoms where the breeze then died. He then sent an anchor ashore in order to warp the ship in close.
Thus we were able to settle on the probable grounding site of HM Ship Roebuck. When quizzed on their choice of action in the circumstances, for example, six of the seven on board Ascension Frigate chose a direct easterly route into the beach, with one selecting a south easterly course in towards the beach at what is now the pier. The winds certainly allowed for such a range of choices, even for a cumbersome and water-laden ‘square rigger’ such as the sinking Roebuck. Being closest and clear of any visible obstructions, the middle-to-northern end of Long Beach was deemed the best starting place for the search by the majority, though it needs be admitted that it was a near-unanimous choice that was partly dictated choice by the heavy swells which seemed lower in that area.
At the evening meeting, experience at wrecks driven ashore in storms or similar circumstances was drawn on to form a predictive model for the ensuing search. One very useful example was the VOC ship Zuytdorp (1712), a much larger vessel than Roebuck and one drawing c.4-5 metres, which came to rest lying in 2-3 metres of extremely rough water on the Western Australian coast. Another was the French exploration Frigate L’Uranie (1820) of slightly less draught, that the team had found and inspected in 3-4 metres of much less turbulent water at the Falkland Islands the previous week.
These two examples indicated that, if it had not been blown back out to sea, Dampier’s wreck most likely lay in much less than the 21 feet of water (3 and a half fathoms) described by Dampier as the grounding site. This depth and the area to seaward of it was the object of past sea-based searches in 1979 and 1985.
As a result, the Western Australian Museum search was to be focused in the shallows inshore of that depth, close by shore and right in the breakers and if unsuccessful, it was intended to commence a series of water-drive probes onshore on the beach itself. The first stages of the search were to be conducted from land through the waves utilising a common, very simple but quite accurate ‘transit’ survey method. This entailed swimming out from the beach along a line marked onshore with temporary ‘navigation leads’ similar to those used by vessels to navigate in and out of narrow channels. At the completion of that line the divers were to move down to the next line five metres south and then swim back to the shore.
The next morning on March 16th, the team then proceeded to the northern end of Clarence Bay by land to commence the search of Long Beach. The first two sets of ‘leads’ were erected and two divers (Geoff Kimpton and John Lashmar) commenced work. While they were occupied in this manner the others in the team were engaged erecting the next pair of ‘leads’ for the divers to use as their navigational guide. A search of the beach and of the rocks in the wave line was also conducted.
The Location of Wreckage
Again fortuitously, the beach was found to be heavily eroded and very steep just inshore of high water mark, with rocks, formerly covered with sand at the northern end completely exposed.
Many heavily-concreted iron fastenings were seen amongst the rocks at the north end of the bay. Though clearly of ancient origin given their form, these were interspersed with such large quantities of modern concreted, iron and steel detritus, that it precluded any definitive conclusions being made, for the area was clearly a natural wreckage trap.
In searching along the first 8 transit lines (0-45m) in a moderate swell and smooth seas from the shore to depths offshore of around 7-8 metres deep (the practical limit of the visibility) and no more than 200 metres from the beach, Kimpton and Lashmar saw numerous concreted objects and debris. It soon became apparent from an examination of the recently exposed beach that its entire northern end was littered above and below water with steel drums (appearing as iron hoops), jetty fittings, wreckage and rubbish. Significant or interesting items requiring further inspection were marked from shore, again using a series of intersecting transit markers.
Amongst the rocks at the northern end of the beach was a considerable amount of material, consisting of concreted iron work, modern detritus, ceramic fragments, and what appeared to be a broken ceramic amulet, similar to those seen by the author on the ‘Blackbirder’ Foam (1893 ) on the Great Barrier Reef. The find was inconclusive, however and in not being proceeded with, was left in-situ.
After just under an hour of searching with Geoff Kimpton along the transit lines set on the beach,John Lashmar descended and while swimming along the bottom he located a bronze bell on the 9th transit line. It was found lying almost totally uncovered but affixed to a cleft in rocks c. 90m from shore on a rock/sand seabed c. 4m deep. Indications were that it had only recently been exposed—with a distinct line on its surface delineating the high point of the latest sand movement around it. The search regime was then halted, with the team uniformly in disbelief.
After the find was examined by all present on snorkel, the search was suspended for the day while the bell was recorded with video and still cameras on SCUBA. While thus engaged numerous iron concretions, some lead sheeting, crumpled iron work and large quantities of clearly modern detritus were seen in the vicinity of the bell, including many 44 gallon drum ends, and a buried tractor tyre with only a small part of is side casing visible. This provided a very useful clue to the depth of the sand in the area surrounding the bell find.
For a while the ordered progression of the search was lost and a random swim of the immediate area was conducted. A short time later, a heavily-concreted longboat grapnel was found concreted to a rock on an exposed rocky seabed. Then a large clam was found exposed in a cleft in the reef on the seabed. It lay in the swell in shallower water c. 100 metres south of the bell to the south and c. 8 metres from shore. The grapnel lay closer to the bell in slightly deeper water than the clam.
Further south in shallow water close to the beach and in a very turbulent location, a slightly tapering iron object similar to the heavily eroded cannon found on the wreck of the VOC Ship Zuytdorp (1712), again in the wave line and firmly wedged amongst the rocks. Once a tapering cylinder, the object had its upper surface worn away with sandblasting in the surge. Trunnions or, as is often the case in turbulent exposed conditions, stubs of trunnions, were not visible, precluding any identification of it as a cannon.
It also had some of the characteristics of an iron bilge pump pipe, (whose discrete lengths are often incorrectly reported as cannon), again no identification was made. In this case measurements could not be taken that would have resolved these issues. Throughout there were also landing craft/airstrip tracks, 44 gallon drum ends, and other unidentified modern detritus, including various sizes and diameters of modern pipe, which though they had no discernible taper were ubiquitous enough to preclude any conclusive identification of the object seen.
The dive was then concluded and the evidence reviewed. This led to the decision to raise the Bell and Clam as shown below.