The Depletion Playbook

Published EROI studies count only the energy that flows through the operator's own meters — the gas piped into a SAGD facility, the diesel through the mine fleet fueling station. Everything consumed by contractors, subcontractors, camp operators, transportation companies, and support services falls outside the measurement boundary. The table below reconstructs what a full-system EROI looks like when those invisible inputs are added back.

A peer reviewer — particularly one working inside the Hall / Lambert / Guay-Boutet tradition — will want to see the computation underlying the waterfall, not just the endpoints. Compressed tables invite the suspicion that the ratios were back-solved. Shown line by line, the math is defensible, but the range it yields is sensitive to a single interpretive choice that should be made visible.

The interpretive choice. Each line item above is expressed as "X% additional energy input." That phrase admits two readings. Under the conservative (multiplicative) reading, each percentage represents a fractional increase in the operator's currently-metered energy input — so "5% winter heating" means 5% more energy in than the operator counts. Under the aggressive (additive-to-output) reading, each percentage represents energy consumed as a share of gross energy output, added directly to the input side of the ratio. Both are used in the EROI literature depending on context; boundary analyses that track industry-wide invisible inputs more often use the second. The summary waterfall in this document implicitly uses the second. The two readings bracket the real answer.

What this shows. The summary waterfall's headline figure of 2–2.5:1 mining and 1.5–2:1 SAGD sits closer to the aggressive reading than the conservative one. Under the conservative reading alone, the full-system collapse is more modest — but still drops mining from 5:1 to roughly 3:1, and SAGD from 4:1 to roughly 2.5:1. Both of those conservative endpoints are already below the ~10–12:1 civilization threshold established by Hall, Lambert & Balogh. The qualitative conclusion — that honest boundary accounting places the oil sands below civilization-sustaining EROI — is robust to the interpretive choice. The specific numeric endpoints are not.

Further complications not captured above. First, all energy is treated as thermodynamically equivalent. In reality the oil sands convert premium natural gas into lower-grade bitumen; energy-quality weighting (the "exergy" correction in the biophysical economics literature) pushes every number further toward unity regardless of reading. Second, the estimates compound: contractor fuel is embedded in the diluent loop, which is embedded in refining, which is embedded in decommissioning. Double-counting is possible and has been kept modest here, but the direction of bias is toward overstating EROI, not understating it. Third, the percentage ranges in the section above are the author's estimates from 20 years in the field, cross-referenced with Guay-Boutet's published boundary analysis; they are ranges, not measurements. A formal boundary study with industry cooperation would tighten them — and it has never been conducted.

The point for a sceptical reviewer. Any attempt to dismiss the full-system EROI collapse requires demonstrating either that the contractor-energy line items don't exist, that they're already counted somewhere, or that the author's percentage ranges are implausibly high. None of these defences survive contact with a Fort McMurray site. The line items exist, they are not counted at the operator boundary, and the ranges are if anything conservative for peak-construction periods.

Everything documented above — the oil sands debt obfuscation, the forestry falldown, the fisheries data suppression — is a local expression of a global pattern. The same playbook of flattering metrics, deferred liabilities, and institutional denial is running simultaneously across every major resource system on Earth. Western Canada is not an outlier. It is a preview.

How the oceans reached 10%. The decline didn't begin with industrial fishing. Roman garum factories processed fish at industrial scale across the Mediterranean two thousand years ago. Medieval European fleets depleted the North Sea and Baltic centuries before anyone counted a stock. Caribbean sea turtles, once so abundant Columbus described the ocean as "thick with them," were hunted to functional extinction before Darwin was born. Chesapeake Bay oysters, which once filtered the entire bay every three days, were dredged to less than 1% of historical biomass by the early 1900s. Great whale populations were reduced by 70-95% by industrial whaling before the moratorium. Each generation inherited a diminished ocean and called it normal — the phenomenon Daniel Pauly named "shifting baseline syndrome" in 1995. By the time modern fisheries science began systematically counting in the mid-20th century, the ocean was already a shadow of what it had been. The published finding that large predatory fish are at roughly 10% of pre-industrial levels (Myers & Worm, Nature 2003) uses "pre-industrial" as the baseline — not pre-exploitation. The true pre-human-impact ocean is unrecoverable from the data because we destroyed it before we thought to measure it.

The mineral wall. The pattern extends beyond biology. Global average copper ore grades have fallen from roughly 2.5% in the early 1900s to approximately 0.5% today — meaning five times more rock must be mined, crushed, and processed to yield the same tonne of copper. Each halving of ore grade roughly doubles the energy required for extraction. Chromite, essential for stainless steel and superalloys, faces tightening supply with the highest-grade deposits concentrated in politically unstable jurisdictions. Metallurgical coking coal — irreplaceable in blast-furnace steelmaking — faces quality decline and rising extraction costs globally. Phosphate rock, the basis of agricultural fertilizer, has no substitute and is being depleted on a multi-decade trajectory that threatens global food production. In each case, the same dynamic applies: the high-grade, easy-to-access resource was extracted first, and what remains requires exponentially more energy, capital, and environmental destruction to recover — the identical EROI collapse we documented for the oil sands, playing out across the periodic table.

Converging thresholds. These are not independent problems with independent timelines. They are coupled systems approaching critical thresholds simultaneously. The energy required to extract degrading mineral ores rises just as the EROI of the energy supply itself is falling. The forests that stabilize the water cycle and sequester carbon are being converted to flammable monocultures and burning at unprecedented rates. The oceans that regulate climate and provide protein are being emptied from the top of the food chain downward. The soils that grow food are losing organic matter and depend on fertilizers derived from depleting phosphate and natural gas. Each system's failure accelerates the others. A fishery collapse increases pressure on agricultural protein, which increases fertilizer demand, which accelerates phosphate depletion and natural gas consumption, which degrades the energy return that supports the whole edifice.

The question is not whether these thresholds exist — the scientific evidence is unambiguous that they do. The question is whether they are crossed gradually, giving societies time to adapt, or whether the nonlinear dynamics of coupled systems produce cascading failures that arrive faster than institutions can respond. The ecological literature strongly suggests the latter. Systems under stress appear stable until they don't.