THE MARITIME ATLASThe Tunnels of Halifax Data method

DATA METHOD

The Tunnels of Halifax

How the film, the live map and the record on this page were made, the limits of each, and where every fact comes from.

Snapshot: 9 October 2026. Nothing on the page comes from a database; every moment, line and source sits in a small data file beside it (data/), built by a script from the research files of the reel the page grew from.

The film

The film is the 82-second reel "The Tunnels of Halifax", published on Instagram. It is drawn frame by frame from the same map data as the live map, so the two cannot disagree. Its script was written by Joel Hirtle, checked line by line against the fact table below by a separate editor pass, and approved word for word before it was spoken. The narration is a computer voice (Microsoft neural text-to-speech). The archive photographs along its edge are listed under Pictures.

The record and its clock

Twenty-two moments, 1749 to 2023, each citing its facts (data/facts.v1.json) and each fact its source (data/sources.v1.json).

  • A solid dot means two independent sources agree, or the moment rests on a primary record (the 1866 report, Piers's history, the federal heritage statement).
  • A hollow dot, "one source", means only one account gives it. It is used, and marked.
  • Years are the record's. Where a thing was built over years (the Citadel's gallery, the island's tunnels, the Ocean Terminals) the map draws it across those years.
  • Quotations are no longer than a phrase.

The live map's clock runs 40 seconds and is not even: 1749 to 1860 takes a fifth of it, the 1860s (the Halifax Club, the island's tunnels, the 1866 report) take another sixth, and 1910 to 2023 takes the rest, where most of the finds are. The tick marks under the slider show the real years.

The map

Today's Halifax waterfront: the shore from the Nova Scotia Topographic DataBase at 1:10,000, which runs behind today's wharves and keeps today's filled land; streets and building footprints from Halifax Regional Municipality's open data. Positions are metres east and north of the Citadel's official name point (GeoNAMES, 44.64722 N, 63.58056 W), drawn north up. At this size, about 3 km across, that flat frame differs from the atlas's Lambert projection by far less than a street's width; the atlas's other pages use Lambert.

The straight line from the Citadel to Georges Island, 1.81 km (1.12 miles), is the great-circle distance between the two GeoNAMES points.

The original shore and the filled land

The original shore comes from the Lunenburg Rebellion page of this atlas, which drew Halifax as it stood in 1753. Its makers took Statistics Canada's 2021 shoreline, removed today's piers, and cut the Halifax waterfront back: from Point Pleasant to the Narrows to the shore on J. F. W. Des Barres's chart "Halifax Harbour" (The Atlantic Neptune, 1779), traced by eye and fitted to today's shore to about 50 m; along the town, to the beach on Moses Harris's plan of 1749, fitted to today's streets, which keep the 1749 grid. North of Buckingham Street the shore turns inland as on a French plan of Halifax of 1755 (Nova Scotia Archives, Beamish collection; a private-study copy, not reproduced).

The filled land is today's land that was water on that shore, on the Halifax side only: Georges Island and Dartmouth are left out, and slivers under 500 m² (5,400 sq ft) are dropped. On this map it comes to about 76 ha (188 acres). Today's piers are not counted: the shoreline runs behind them. The line is an estimate, good to about 50 m along the harbour and about a block along the town; it is not a survey.

Two written sources agree that the 1749 shore ran about where Water Street runs today (Tulloch; Buggey). No source gives one figure for how far the waterfront moved, so the page gives none.

Where the finds are

The library's list gives a year and a street for most finds, not a place. Each find is drawn on its street at a point chosen for the drawing, with a dashed ring that says it is placed, not surveyed. The 1973 find is drawn near the foot of George Street, where the road caved in; the 1976 find on Prince Street; the Halifax Club passage at the club on Hollis Street; the 2017 manhole on Argyle Street.

The tunnel plans

The red plans of the Citadel's gallery and countermines and of the Georges Island tunnels are drawn to show the idea, not surveyed. The gallery is drawn as a ring a little outside the Citadel's star, along the counterscarp, with T-shaped countermines pointing out under the glacis; the island's tunnels as a spine with branches. Neither is to scale. The 1866 sewers are drawn along today's centrelines of the four streets the finds are on; Kelly names a fifth, Buckingham Street, which is not drawn. The legend's route is dashed because no one has found it; its line is the story's, not a survey's.

Pictures

  • The Citadel, 1870 (NSA no. 6804), the Citadel's ditch, about 1879 (NSA no. 6818), and Fort Charlotte, about 1876 (NSA no. 6889): Royal Engineers photographs, Nova Scotia Archives. Shown with the Archives' credit, as the atlas does for its old maps and photographs.
  • A page of John Kelly's Report on Halifax Sewerage (1866): Internet Archive, CIHM 63443, public domain.
  • No picture on the page or in the film is made by AI.

Where the record disagrees

  1. When Piers answered the legend. Harry Piers died in 1940; his history was printed in 1947. The page dates his answer 1947, the year it could be read.
  2. Who found no tunnel under the harbour, and how. Parks Canada's Hal Thompson told CTV Atlantic in 2013 that a survey of the harbour floor showed no evidence of a tunnel. The Geological Survey of Canada's 2023 poster names the tunnel as a myth; the poster itself has not yet been read, so the page does not name a survey method.
  3. Finds and sewers. The library's list and Kelly's report match by street name only.
  4. The countermines. That fifteen have collapsed, and that one engineer cut half of those planned in the 1830s, rest on the federal heritage statement alone.
  5. The Cameron story. A family story, given by the library, that a Pictou County mason was sworn to secrecy in the 1860s; the archives record him only on the St. Peter's Canal, 1876 to 1881.

Every source

Every source the page and the film draw on, each linked to the document. One is held only in an archive; it is marked.

The record

The waterfront

Maps and the map's data

Research notes, with every request, licence and quotation, are kept with the project and are not published. Built by Joel Hirtle, October 2026, in the shape of the Lunenburg Rebellion page.