HMS Investigator Underwater Archaeology Survey

Ryan Harris and Henry Cary (NWT Archaeologist’s Permit 2010-001)

After a brief search using towed side-scan sonar the wreck of HMS Investigator, one of the first two ships sent by the Royal Navy in search of the ill-fated Franklin Expedition and whose crew was ultimately credited with identifying the final missing link in the elusive Northwest Passage, was successfully located in close proximity to its reported position of abandonment in Mercy Bay (Banks Island) in 1853. The wreck was found on July 24, 2010 by Parks Canada’s Underwater Archaeology Service (UAS) during a field project spanning from July 22 to August 4.

While initial detection was hindered by surrounding ice floes that masked the target in the side-scan sonar record, these gradually drifted off site, affording an unimpeded acoustic view of the vessel. Archaeological study of the wreck of HMS Investigator continued for the balance of the two-week project, despite persistent ice cover and strong northerly and southerly winds. Towards the end of the survey, and with a blanket of ice steadily advancing back into Mercy Bay, a large area of the seabed surrounding the wreck was systematically scanned with side-scan sonar to search for detached wreck sections, rigging, and the ship’s anchors and to determine the area where debris may be found. However, the wreckage is largely concentrated in one location, with the vast majority of detached timbers either immediately alongside or atop of the intact hull. This survey also served to identify the considerable number of ice drags that scar the ocean floor in the general vicinity of the site; damage from ice representing the most obvious long term threat to the site’s integrity.

A series of remotely operated vehicle (ROV) dives were completed over the course of two days covering the entire wreck from stem to stern and capturing upwards of three hours of diagnostic video imagery. Of particular note are images of the stern showing draught marks, rudder attachments and copper hull sheathing as well as images of the wooden ice chocks and iron plating protecting the bow. Comparison of the ROV imagery with copies of 1848 colour plans of the Investigator held by the National Maritime Museum in England, allowed us to do a preliminary integrity evaluation.

The wreck lies in 11 m of water and appears to be substantially intact to the level of the upper deck, with a considerable portion of the lower hull down to the keel lying deeply buried in sediment. The three masts and bowsprit have all been carried away. The entire bulwark and transom structure from bow to stern, and all manner of protruding upper deck fittings — including the helm, pin rails, davits, capstan, and windlass — have been completely flattened, shorn away or otherwise displaced by icebergs which have periodically impacted the site.

While the UAS team focused on the preliminary evaluation of the shipwreck, a terrestrial team concentrated on mapping ‘McClure’s Cache’, a depot of ship’s stores, provisions, boats, rigging, and other supplies that was landed ashore in 1853 and 1854 on a sloping terrace on the western side of Mercy Bay. Located within the boundary of Aulavik National Park, this work was conducted under a Parks Canada Research permit held by Henry Cary of Parks Canada’s Western Arctic Field Unit. The land crew also included Parks Canada members John Lucas and Letitia Pokiak, and University of Western Ontario archaeologist Edward Eastaugh. The land team also visited Mottley Island, situated in the middle of Mercy Bay (outside the Park), with an aim to relocating the site of reported ‘Esquimaux Remains’ as they are listed on an 1853 hydrographic chart of Mercy Bay drafted by Stephen Court, Second Master of the Investigator.

Mottley Island is a limestone and sand islet near the centre of Mercy Bay. The ship’s Second Master Stephen Court clearly indicates it on his 1853 hydrographic map of Mercy Bay, and above it inscribed the words ‘Esquimaux Remains.’ This is a reference to the ‘evidence’ of Inuit ‘migration’ found in June 1852 by the ship’s surgeon and naturalist Alexander Armstrong (2010:527), and the ‘two Esquimaux huts’ and ‘houses of stone’ mentioned by Captain Robert McClure (Great Britain Parliament 1854:51) and Inuit translator Johann Miertsching (1967:165) in July the same year. All three men clearly state that their findings were on an island near the centre of Mercy Bay, yet despite these seemingly clear descriptions archaeological surveys since the early 1950s have failed to uncover any substantial evidence of Inuit or earlier occupation (Webster 1996:11). A single site was found by Parks Canada archaeologist Stephen Toews (1998:42) on the island’s south-west sand spit in 1997, but this was simply a scatter of wood fragments surrounding a possible stone feature, and may actually post-date 1854, the year Investigator was abandoned.

Although the 1997 survey was thorough, we hoped reinvestigation of the island would offer new perspectives on the historic artefacts reported there, and might result in discovering previously unknown archaeological remains. We completed a foot survey of the entire island and, apart from re-locating the site found by Toews, were similarly unsuccessful in finding sites of the magnitude visited by Investigator’s officers. The site is still much as Toews describes, being a scatter of wood fragments spread over an 21.5 x 14.5 m area and including a roughly 3 m diameter semi-circular arrangement of large angular limestone slabs measuring up to 50 x 20 cm.

The presence of wood led Toews to believe the site dated to after Investigator’s abandonment, and was created by Copper Inuit travelling to Mercy Bay to salvage the depot, since wood, apart from occasional drift or fossilized sources, is virtually non-existent on Banks Island. The wood found on Mottley Island is similar to the remnants of ship’s boats and boxes found at the depot. Toews (1998:42) believed that the stone arrangement could date to an earlier period but without excavation this is difficult to confirm. Another clue suggesting a later date for QaPu-1 is Court’s map showing this part of the island as a shoal, and not until a sea level drop would the area be inhabitable.

Where then are the sites mentioned by Armstrong, McClure and Miertsching? It is possible they eroded into Mercy Bay but if so this natural destruction in an area of low tidal fluctuation and in just 160 years must have been considerably more pronounced that found elsewhere in the bay. A more likely conclusion is that the Investigator’s officers and translator were describing a boulder plateau on a low point extending into the southern portion of Mercy Bay (1998:43). This plateau is several metres higher than the surrounding land and in conditions of higher sea levels or river outflow it may have appeared as an island in June and July 1852. Further support for this hypothesis is that the plateau has a significant number of archaeological features strewn across its surface, including a structure surrounded by large whale bones, which matches Miertsching’s account of a stone ‘house’ with a ‘roof, made of whalebone, [that] had fallen in.’ Miertsching (1967:166) also describes ‘two circular receptacles of stone, which probably served to protect food from foxes and wolves,’ and these could be the large stone caches found on the plateau in 2010. This said, Court does not show an island off the southern Mercy Bay point, and marks no ‘Esquimaux Remains’ there as he had done for Mottley Island, and on the summit of a hill south of Providence Point. A possible explanation is that Court was at the same time both accurate and imprecise. Armstrong, Miertsching and McClure made their discoveries in 1852, but Court dates his map to the following year. In the period between the officers and translators travels and Court’s map, the water may have dropped to reveal that the plateau was not an island but instead part of the point. However, Court may have annotated his map using verbal descriptions that the ‘Esquimaux Remains’ were discovered on an island, and since the only island in Mercy Bay was Mottley, the remains must therefore be located there.

Excavation on Mottley Island may yield prehistoric features but these are unlikely to be as extensive as those described historically. Research of the island provides an interesting example of the problems encountered when attempting to reconcile historic accounts with the archaeological record.

Many thanks to Brian Healy of Canadian Helicopters for his safe and enthusiastic support during the 2010 HMS Investigator Rediscovery Project.

(Edited by Shelley Crouch, Prince of Wales Northern Heritage Centre)

References Cited

  • Armstrong, Alexander. 2010. Reprinted. A Personal Narrative of the Discovery of the North-West Passage with Numerous Incidents of Travel and Adventure During Nearly Five Years’ Continuous Service in the Arctic Regions While in Search of the Expedition Under Sir John Franklin. Milton Keynes UK: BiblioLife, Hurst and Blackett, Publishers. 1857.
  • Court, Stephen. 1853. Harbour of Mercy. Surveyed by Mr Stephen Court_Second Master H.M. Ship Investigator, Commanded by Robert McClure Esq’r Commander. Hydrographic Map. Taunton: United Kingdom Hydrographic Office. D1073 Oct 6 54.
  • Great Britain Parliament, Accounts and Papers, Navy. 1854. Proceedings of Captain McClure, of Her Majesty’s Discovery Ship “Investigator,” in Search of the Expedition Under Sir John Franklin, from August 1850 to April 1853, and Reporting the Discovery of the North-West Passage. Pp. 23-62. Section 5, 1. London: Great Britain Parliament.
  • Miertsching, Johann. 1967. Frozen Ships: The Arctic Diary of Johann Miertsching, 1850-1854. L.H. Neatby, trans. Toronto: Macmillan of Canada.
  • Toews [Sauvage], Stephen. 1998. “The Place Where People Travel”: The Archaeology of Aulavik National Park, Banks Island. Winnipeg: Parks Canada, Western Canada Service Centre. Manuscript on file, Parks Canada Western Service Centre, Winnipeg, Manitoba.
  • Webster, Deborah Kigjugalik. 1996. 1994-1995 Thomsen River Archaeology Survey Aulavik National Park. Winnipeg: Parks Canada, Archaeological Field Service, Professional and Technical Service Centre. Manuscript on file, Parks Canada Western Service Centre, Winnipeg, Manitoba.