SpillMap.ca · TMX Dilbit Spill — Impact Scenario Explorer

1 · Where

Spill source

Accident type

2 · How big

Spill size

Tanker class

ⓘ Where could a VLCC actually load? (Burrard Inlet vs Roberts Bank)

Burrard Inlet stays Aframax-only. The Second Narrows dredging (from Sept 2026) lets today's Aframax tankers load more fully (past the current ~80%) — the port authority is explicit it "will not change the number or size of the largest vessel type… Aframax-class tankers will remain the largest vessels." Bridge air-draft (Second Narrows twice per trip, plus Lions Gate), channel depth and the Aframax-built Westridge berths rule out Suezmax/VLCC there; a fully laden VLCC draws ~20–22 m.

But VLCCs on the South Coast are now a formal federal proposal. On July 2, 2026 the Government of Alberta submitted the West Coast Oil Pipeline (Bruderheim → Delta, ~1 million bpd) to Canada's Major Projects Office, and Ottawa began considering it for national-interest listing under the Building Canada Act — listing decision targeted October 1, 2026 (once listed, the Act shifts the question from whether to how). The federal information package confirms two marine berths designed for Very Large Crude Carriers — "these vessels typically have a capacity of 2 million barrels" — at a Delta-area terminal beside Deltaport/RBT2, open-strait berths with no bridge or narrows constraint.

And ships already hit infrastructure at this exact site. In December 2012 the 289-m bulk carrier Cape Apricot, under the conduct of a BC coast pilot, missed her turn at Roberts Bank and cut through the Westshore coal causeway — ~128 m of it collapsed into the sea, the berth was out for two months (TSB M12W0207; find it highlighted on the incidents map). That night the structure alongside carried coal. Select the “Roberts Bank / Deltaport approach” spill source and the VLCC ~2M class to model what the same miss looks like with crude. Sources in the research brief §2.

ⓘ Doesn't the double hull protect us?

It's already the law — and already in these numbers. Since Exxon Valdez, every crude tanker must be double-hulled (MARPOL Annex I Reg. 19 / US OPA-90; single hulls fully phased out). Every tanker in these scenarios has one. It genuinely helps: US National Research Council assessments found double hulls cut expected spillage by roughly two-thirds — almost all of it by eliminating small, low-speed incidents (harbour bumps, soft groundings) where the ~2–3 m outer hull absorbs the damage. The spill probabilities shown here are the risk that remains after double hulls, escort tugs and pilotage are counted.

What it doesn't change is the worst case. A powered collision or a grounding on rock at transit speed penetrates both skins — analyses found a double hull would have reduced the Exxon Valdez outflow by only ~25–60%: a smaller catastrophe, not none. The modern proof is the Sanchi (Jan. 2018): a modern double-hulled tanker carrying ~960,000 bbl of natural-gas condensate — the same ultra-light, volatile hydrocarbon used to dilute bitumen — was struck in a powered collision, ignited at once, burned and drifted for eight days while the vapours kept rescuers away, then exploded and sank. All 32 crew died; the entire cargo was lost. The double hull changed nothing that mattered that night — and no hull design touches the benzene plume of the first 48 hours, or dilbit's ability to submerge beyond skimmers. Sources in the research brief §2b.

ⓘ What if the Big One hits the terminal?

The site question nobody prices. The proposed Roberts Bank crude terminal would sit on Fraser-delta soils — loose, water-saturated sands whose liquefaction was the principal engineering concern in the neighbouring RBT2 assessment, directly upslope of a submarine feature geologists have mapped and named the Roberts Bank Failure Complex. The RBT2 answer was a commitment: ground improvement, to be finalized in detailed design. A commitment is not a frequency — no published QRA prices an earthquake-triggered terminal release here.

It has happened before, on exactly this soil. In the 1964 Alaska earthquake, the waterfront tank farms at Valdez and Seward — built on the same kind of saturated delta deposits — liquefied and slumped into the sea; the tanks ruptured and put burning oil on the water. The City of Vancouver's own planning scenario is a M7.2 Strait of Georgia quake (2024 council report: ~6,100 heavily damaged buildings, 1,350+ severe injuries/fatalities, >$17B) — meaning the same event that breaches the terminal also consumes every responder, bridge and hospital the spill plan assumes. And don't look south: the same shaking hits Washington State (Point Roberts is ~5 km away), the Canada–US spill plan (CANUS JCP 2022) is explicitly non-binding — “does not give rise to any rights or obligations” — and Washington's own full-scale quake exercise (Cascadia Rising, 2016) judged its plan “grossly inadequate.” Load the “M7.2 earthquake → terminal failure” preset to model it. Sources in the research brief §2c.

Pipeline throughput

3 · Conditions

Season

Time of day

Wind regime Wind speed

Sea state

4 · Water & tide

Tide state

Tidal phase

Storm surge Fraser River flow

Pipeline corridor Tanker route Oil on water Benzene / VOC plume Wind → plume drift Tidal current (flood/ebb) Net slick drift

◔ How likely? — spill probability & return period

—return period for a spill ≥ this size

—annual probability

—over 50 yr of operation

Project lifetime 50 yr

🛢 Has this actually happened? — five tanker disasters since 2015

Every one of these spills was a small fraction of one Aframax cargo (~600,000 bbl) — let alone a ~2M-bbl VLCC — and each was still a regional catastrophe.

And the double hull? A double hull protects a floating ship from puncture. It is irrelevant when the ship itself founders — sinks, capsizes, or breaks apart — because the cargo goes down with the vessel. That is four of the five incidents below. The fifth, the Sanchi, was a modern double-hulled tanker — and lost everything anyway.

Sanchi, East China Sea 2018 — double-hulled tanker, ~960,000 bbl of condensate (dilbit's diluent family); collision, cargo ignited, burned 8 days while vapour kept rescuers away; all 32 crew died, total cargo loss. Volatile cargo makes the casualty unapproachable.

Agia Zoni II, Athens 2017 — sank at anchor, in good weather, in the enclosed gulf beside the capital; ~3,700 bbl escaped (<1% of an Aframax) and oiled ~29 km of Athens-area beaches for months. Calm urban waters are where small spills do maximal damage.

Princess Empress, Philippines 2023 — ~5,700 bbl (~0.3% of a VLCC) sank into the Verde Island Passage, the global epicentre of shore-fish biodiversity; months of fishing bans, 33,649 fisher compensation claims. Biodiversity hotspot + fishing livelihoods = the Salish Sea's profile.

Terra Nova, Manila Bay 2024 — capsized in typhoon seas inside a sheltered urban bay (~9,400 bbl); ~14,000 fishing-industry workers affected. "Sheltered water" isn't a safety property when weather arrives — our winter southeasterlies are this coast's typhoon.

Volgoneft-212 & -239, Kerch Strait 2024 — two (aged, river-class) tankers broke apart in a winter storm; the heavy fuel oil submerged and kept re-oiling ~60 km of beaches for months, defeating booms and skimmers. The closest recent analog to weathered dilbit's submerge-and-sink behaviour.

Full accounts and citations in the research brief §5c.1.

Seen enough? Ottawa confirmed in July 2026 that the North Coast tanker ban stays — but the ban stops at Vancouver Island's northern tip. No ban protects these waters, and the national-interest listing decision on the VLCC terminal is targeted for October 1, 2026.

📞 Call your MP — it takes 2 minutes Strong Coast — the campaign

Outbound links to StrongCoast.org — SpillMap.ca uses no cookies and collects no personal information (anonymous, aggregate visit counts only). · How to cite these numbers ↗