Ask a crew what sizes the fuel on a long over-water sector and the answer comes back, almost without exception: the engine failure. It is the failure the training footprint is built around, the one every simulator detail contains, and the one the aeroplane is certified against. It is also, on a great many routes, not the case that fixes the number. The scenario that fixes the number is usually the depressurisation — a healthy aeroplane, two engines producing full thrust, condemned to spend the whole of a long diversion at an altitude where it burns fuel at a rate it was never designed to sustain. That inversion is not an exotic case, and it is visible on the flight plan of anybody who knows which line to read.
What the critical fuel scenario is actually asking
The question is not « how much fuel do I need to reach the en-route alternate ». It is narrower and considerably more demanding: from the most penalising point on the route, having suffered the most limiting of a defined set of failures, can the aeroplane still reach a usable en-route alternate and arrive with its prescribed reserves untouched? The answer is not a fuel figure but a difference — the amount, if any, by which that requirement exceeds the fuel already planned to be on board at that point. That difference is what appears on the flight plan as additional fuel, and on many sectors it is zero.
A word on names. ICAO writes the framework as extended diversion time operations, EDTO, which applies to aeroplanes of any number of engines. Europe and the industry generally still say ETOPS for two-engine aeroplanes and reserve the newer term for the general case. The architecture underneath is the same: an approved maximum diversion time, an area of operations derived from it, a set of adequate aerodromes, and a fuel calculation built on named failure scenarios. Approval steps are commonly expressed as sixty, one hundred and twenty or one hundred and eighty minutes at the approved one-engine-inoperative speed in still air, with longer times available under further conditions.
Three scenarios, and an instruction to take the worst
The regulation does not ask which failure is likely. It names the cases and tells the operator to size on the most limiting of them, which is a very different instruction and the source of most of the confusion around this subject.
| Scenario | What is lost | Altitude flown for the diversion | Dominant penalty |
|---|---|---|---|
| Engine failure at the critical point | Half the installed thrust | The one-engine-inoperative net level-off altitude, well up in the flight levels for a modern twin | Lower speed, longer diversion time |
| Depressurisation at the critical point | Cabin pressurisation only | A level sustainable without supplemental oxygen, unless additional oxygen is installed | Severely degraded specific range for the whole diversion |
| Depressurisation combined with engine failure | Both | The lower of the two, at reduced speed | Longest time at the worst specific range |
For a two-engine aeroplane all three are examined. For three- and four-engine aeroplanes the depressurisation cases are the ones that do the work, since losing one engine of four does not force the same descent. The combined case is not a prediction that both will happen together; it is a boundary the fuel plan is required to stand on.
The whole argument is altitude
Everything that follows rests on one property of a transport aeroplane: its fuel efficiency is a strong function of altitude, and the optimum is high. Near the optimum cruise level the wing is working close to its best lift-to-drag ratio at a high true airspeed, and the engine is running at a high, efficient rating in cold, thin air. Move the same aeroplane down to a low level and both terms deteriorate at once. It flies further from its best lift coefficient at the speed available, and the engine burns more for each unit of thrust in dense, warmer air at a lower rating. The two effects push the same way.
The result is that specific range at a low diversion altitude is not slightly worse than cruise specific range; it is dramatically worse, often by something approaching a factor of two, and the exact figure is a type and weight question that only the aeroplane flight manual can answer for you. That single fact is the whole of the argument. Any scenario forcing a long segment at low altitude is expensive, and its cost grows in direct proportion to the time spent there — which is to say, in proportion to the diversion distance.
The key idea is that flying low burns fuel much faster than flying high. If you have to divert, staying at high altitude (even with one engine out) is better than being forced down low (even with both engines working). The worst case is often a depressurisation, because it forces you to descend to a lower, less efficient altitude for the whole diversion.
On a Boeing 787 flying from Los Angeles to Tokyo, a depressurisation at the critical point might force the aircraft to descend to 10,000 feet (FL100) for a 3-hour diversion to Anchorage. At this altitude, the specific range could be half of what it would be at the optimal cruise altitude of FL390. This means the aircraft would burn fuel at twice the rate, making the depressurisation scenario more fuel-intensive than an engine failure at a higher altitude.
Engine failure: you lose thrust and keep most of the altitude
After an engine failure at cruise the aeroplane cannot hold its level and drifts down to the net one-engine-inoperative level-off altitude for its weight and the ambient temperature. For a modern twin at realistic diversion weights that altitude is still well up in the flight levels, and it improves as fuel burns off. The aeroplane is therefore not condemned to the bottom of the atmosphere; it merely loses the top of its performance envelope.
The fuel arithmetic is correspondingly moderate. One engine working hard burns at a rate not far removed from two engines at cruise thrust, so hourly consumption does not collapse in the operator’s favour, and the aeroplane is slower, so the diversion takes longer. Specific range degrades — but it degrades from a good starting point, at an altitude where the airframe and engine still function near a reasonable part of their envelope. What actually drives the engine-failure case is time, not burn rate.
Depressurisation: you keep the thrust and lose the altitude
The depressurised case inverts every term. Both engines are healthy and the aeroplane’s performance is unimpaired; what is impaired is the cabin, and the descent that follows is not a performance decision at all. It is an oxygen decision. The aeroplane must reach and hold a level at which the occupants can be sustained for the remainder of the diversion, and the supplemental oxygen available sets how long any intermediate level can be held. Passenger oxygen from chemical generators is finite and comparatively short-lived; crew oxygen lasts longer but is also finite; neither was sized for a multi-hour diversion.
The consequence is a fully serviceable aeroplane obliged to fly the entire remaining diversion low. The emergency descent itself costs almost nothing, since it is flown at idle thrust — the cost is everything that comes after it. Where the engine-failure case pays for time at a degraded but respectable specific range, the depressurised case pays for the same distance at the worst specific range the aeroplane can produce, and it does so for as long as the diversion lasts.
The trap. « Two engines running, so it must be the cheaper case » is the intuition to unlearn. Engine thrust is not the scarce commodity in a depressurisation; altitude is. A healthy aeroplane held down at a low level for three hours can easily consume more than the same aeroplane covering the same distance on one engine at a driftdown level.
The combined case, and the equipment that decides it
Where the combined scenario is required, it is usually the one that binds: the lowest achievable level, one engine, the lowest speed, therefore the longest time at the worst specific range. It is worth noticing, though, that this is not a fixed conclusion — it is the output of a calculation whose inputs an operator can change.
The most powerful of those inputs is the oxygen installation. A fleet fitted with extended-duration supplemental oxygen can hold a higher level after a decompression, which lifts the specific range for the entire diversion and can move the dimensioning case back to the engine failure. That fit is not a refinement; on routes crossing high terrain — the Andes, the Himalaya, the Tibetan plateau — it is what makes the route possible at all, because a descent to a level sustainable without oxygen is simply not available above such terrain. There, the depressurisation scenario stops being a fuel question and becomes an escape-route question, planned laterally as much as vertically.
| Variable | Effect on the critical fuel requirement | Which scenario it favours |
|---|---|---|
| Longer approved diversion time | Increases it, roughly with the time spent diverting | Strengthens the depressurised cases |
| Extended-duration supplemental oxygen | Reduces it, sometimes substantially | Can return the engine failure to dominance |
| Headwind on the diversion track | Increases it, and moves the critical point | Whichever case is slowest |
| Higher aeroplane weight at the critical point | Lowers the driftdown level and raises the burn | Strengthens the engine-failure case |
| Faster approved diversion speed schedule | Shortens time, raises hourly burn — net effect is type-specific | Depends on the aeroplane |
| High terrain on the diversion track | Constrains the achievable level after decompression | Makes the depressurised cases decisive |
The critical point is not the equal-time point
These two are routinely conflated and they are different constructions. The equal-time point between two alternates is a time construct: the position from which the flying time to either is the same, computed at a declared speed schedule and in the forecast wind. The critical point for fuel is the position from which the required diversion fuel is greatest. In still air with symmetric alternates they nearly coincide. Add a wind field and they separate, because the fuel-worst point is displaced toward whichever diversion the wind punishes.
A route with several alternates has several such points, and the flight plan should say which scenario dominates at each. The additional fuel that reaches the tanks is the largest excess found anywhere along the track, not a sum and not an average.
Why the answer is so often zero
Here is the part that surprises crews new to the operation. On a long sector, the fuel physically on board at the critical point is dominated by the trip fuel still needed to reach the destination, and that is frequently larger than anything the critical scenario demands. The critical fuel requirement is then satisfied by the fuel already there, and no additional fuel appears on the plan at all.
The requirement bites in the opposite conditions: a comparatively short sector with a long approved diversion time, a critical point close to the destination where little trip fuel remains, a distant en-route alternate, or a strong wind on the diversion track. Those are the flights on which the additional fuel figure is real, and on which a late reroute or an alternate becoming unusable changes the number in the air rather than on the ground.
What sits on top of the cruise to the alternate
The critical fuel figure is never simply the cruise burn from the critical point to the alternate. A defined set of allowances is stacked on top, and each of them is there because a diversion is flown by a degraded aeroplane in imperfectly known conditions.
| Element | Why it is in the calculation |
|---|---|
| Ice accretion on unprotected surfaces | A long low-level diversion may cross icing conditions the cruise profile would have avoided |
| Operation of anti-icing and de-icing systems | Engine and wing protection costs thrust and fuel, and is likely to be needed |
| Error in the forecast wind | The diversion track was not the planned track, so its wind was never verified |
| Deterioration in cruise fuel burn | The performance data describes a new aeroplane; yours is not |
| Auxiliary power unit operation where required | Some configurations need it after the failure being modelled |
| Holding, approach and landing at the alternate | Customarily a short hold above the alternate followed by an approach and landing; the exact construction is fixed in the operator’s approval |
| Arrival with final reserve intact | The reserve is not available to the scenario — it is what the scenario must not touch |
The right reflex. When an en-route alternate degrades below its planning minima in flight, the arithmetic does not simply lose one aerodrome. The critical points move, the diversion distances change, and the dimensioning scenario may change with them. Recompute against the alternates that remain usable, rather than assuming the fuel already on board still covers the case it was built for.
What to read off your own flight plan
Four questions turn all of this into something checkable rather than theoretical. Which scenario does the plan name as dimensioning at each critical point, and does that change across the route? Is the additional fuel non-zero, and if so, which of the variables above put it there? What level can your fleet actually hold after a decompression, given the oxygen fit — and does the plan assume that level or a lower one? And which of the en-route alternates is closest to being unusable on the day, given that en-route alternate planning minima are deliberately more demanding than the aerodrome’s operating minima?

Crews who ask those four questions once tend to read the fuel section of an extended-range flight plan differently afterwards, because the answer reframes the whole operation: the aeroplane is not being fuelled against the failure it was certified for, but against the one that keeps it low. The planning and performance reasoning behind this sits in the type-agnostic operations material at FL330 training, while the figures that bind you are in your own manual and your own operational approval.

Leave a Reply
You must be logged in to post a comment.