The Storm Passes Friday. Why Your Monday Flight Is Still Cancelled: Aircraft Out of Position, How Airlines Clear a Backlog, and Your Place in the Queue
Typhoon Dolphin is forecast to clear Okinawa on 7 August. A lot of people are assuming that means flying resumes on the 8th.
It usually does not. The airport reopens, the sky is blue, and your flight is still cancelled two days later.
This is the part of typhoon disruption that almost no coverage explains, and it is the part that actually determines when you get home.
The storm is a weather event that lasts hours. The disruption is a logistics event that lasts days.
This article covers why that gap exists, how airlines actually clear a backlog, and what determines your position in the queue.
Why the disruption outlasts the storm
Because aircraft are not at the airport waiting for the weather to improve.
An airline's fleet is a chain. Each aircraft flies a sequence of legs across a day, and the crew flies a sequence alongside it. When a typhoon closes an airport, that chain breaks in the middle.
The aircraft that was supposed to operate your Monday flight is now sitting in a different city, because it was diverted on Thursday or never departed on Friday. Getting it back is not one flight. It is a repositioning problem across the whole network.
Crew is the harder constraint. Pilots and cabin crew are limited by legally enforced duty and rest rules. A crew that got stuck somewhere unplanned may be out of hours, and cannot simply be told to fly home. They need rest, and sometimes they need to be flown somewhere as passengers first, which consumes seats.
Then there is the airport itself. Ground equipment gets secured or moved before a storm. Debris has to be cleared. Runway and navigation aids get inspected. Fuel supply and catering restart on their own schedules. An airport reopening is a process, not a switch.
And underneath all of it: aircraft on the ground during the storm may need inspection before they fly again.
So the sequence is roughly — storm passes, airport reopens, aircraft return, crews reset, and only then does the schedule rebuild. Each of those steps takes hours to a day, and they run partly in sequence rather than all at once.
Two to four days is a normal recovery for a significant closure. A week is possible if the storm hits a major hub during a peak travel period.
How airlines actually clear a backlog
Not in the order people expect.
The first priority is not passengers. It is restoring the network. An airline that spends day one flying stranded passengers home in random directions still has its aircraft in the wrong places on day two, and the problem repeats.
So the early flights after a reopening are often positioning flights and the regular schedule, not extra recovery flights. This looks unfair from the terminal and it is the fastest route back to normal for everyone.
Once the schedule is running, the backlog gets cleared by whatever seats are left over on normal flights. That is the key thing to understand. Your rebooking is competing for spare capacity on flights that were already mostly sold.
On a busy route in August, spare capacity is close to zero. This is why a two-day closure can produce a five-day delay in getting people out — there is simply nowhere to put them.
Extra sections get added on high-volume routes when it makes commercial sense, usually with larger aircraft rather than more flights. Do not count on it.
Within that constrained space, the ordering is broadly: passengers already rebooked and confirmed, then by fare class and status, then by when the rebooking request was made.
That last one is the part you control. The person who rebooked on Thursday sits ahead of the person who rebooked on Saturday, even if their original flights were on the same day.
Cargo and mail also compete for space in the hold, which affects some routes more than passengers realise, though not usually the seat count.
Your place in the queue, and how to improve it
Act on the first day, not the worst day.
The single biggest determinant of when you fly is when you asked. Rebooking requests are processed roughly in order, and the queue on day one is far shorter than the queue on day three.
Use the app or the website before the phone. Airline apps often allow self-service rebooking during a disruption, and the app is not competing with a hold queue. If self-service offers you something acceptable, take it immediately.
If you must call, call the airline's overseas office rather than the domestic one. Wait times differ enormously between call centres and there is no rule that you must use the one in your own country.
Be flexible on airport rather than on date. Accepting a nearby alternative airport usually gets you out days earlier than waiting for a seat at your original one. In this region that can mean flying into a different Japanese city, or routing via Seoul, Taipei, Manila or Hong Kong.
Do not decline a poor option without securing a better one. Once you release a confirmed seat, you rejoin the queue at the back.
Keep your original booking intact until you have the replacement confirmed. This sounds obvious and people break it constantly under stress.
And book accommodation early. Hotels near a disrupted hub fill in hours, and the ones left at midnight are expensive. If you are going to be stuck for three days, decide that on day one rather than extending night by night.
On compensation: this is weather, which almost always sits outside statutory compensation schemes. Airlines are generally still obliged to get you there or refund you, and duty-of-care obligations vary by jurisdiction and carrier. Insurance is a separate question with its own timing rules.
What to expect, realistically
If you are flying out of an airport that closed for a day, plan on two to four days before things look normal.
If your flight is cancelled and the airline offers you something four days out, that is not the airline being lazy. That is usually what the seat map actually looks like.
If the sky is clear and your flight is still cancelled, the aircraft is somewhere else. That is the explanation almost every time.
And if you take one thing from this: the first twelve hours after a cancellation are worth more than the next three days of effort. Rebook early, take the seat, sort out the details afterwards.
Where to check this yourself
Japan Meteorological Agency: https://www.jma.go.jp/en/typh/
Korea Meteorological Administration: https://www.kma.go.kr/eng/weather/typoon/typhoon_05.jsp
Joint Typhoon Warning Center: https://www.metoc.navy.mil/jtwc/jtwc.html
Hong Kong Observatory: https://www.hko.gov.hk/en/index.html
China National Meteorological Center: http://www.nmc.cn/publish/typhoon/typhoon.html
For the recovery phase specifically, the airport's own website is more useful than the weather agencies. Airports publish operational status and it updates faster than airline schedules do.
The queue, worked out in numbers
Everything above explains the mechanism. The question it does not answer is the one people actually ask: how many days? That can be estimated from published figures, and the answer is uncomfortable.
Step one: how big is one day of Naha
Naha Airport handled 21,705,234 passengers and 158,822 aircraft movements in 2024. Divided out:
| Naha Airport, 2024 | Value |
|---|---|
| Passengers for the year | 21,705,234 |
| Aircraft movements for the year | 158,822 |
| Average passengers per day | 59,466 |
| Average movements per day | 435 |
| Passengers per movement | 136.7 |
Step two: compare that to one typhoon day
During Typhoon Khanun in August 2023, more than 400 flights were cancelled on a single day and more than 65,000 passengers were affected.
| Khanun, one day | As a share of a normal Naha day |
|---|---|
| 400+ flights cancelled | About 92% of average daily movements |
| 65,000+ passengers affected | About 109% of average daily passengers |
Read the second row again. A single typhoon day removed slightly more than one entire normal day of the airport's passenger throughput. Not a fraction of it. All of it, and a little over.
That backlog does not land on an empty airport. It lands on the following days, which are already carrying their own full load of people who booked months ago.
Step three: how fast can spare seats absorb it
The only way the queue clears is through seats that would otherwise have flown empty. So the recovery speed depends entirely on load factor, which is the share of seats actually filled.
Published figures: ANA's domestic load factor for fiscal 2024 was 75.0 per cent. JAL recorded a high of 82.9 per cent in a single quarter of fiscal 2025, and 88.8 per cent on international routes in August 2025. Peak summer therefore runs well above the annual average.
Taking Naha's average of 59,466 passengers a day, the spare capacity at each load factor and the time to absorb 65,000 stranded passengers works out as follows.
| Load factor | Spare seats per day | Days to clear 65,000 |
|---|---|---|
| 75% — ANA domestic, fiscal 2024 | About 19,800 | 3.3 days |
| 80% | About 14,900 | 4.4 days |
| 85% — realistic August | About 10,500 | 6.2 days |
| 90% | About 6,600 | 9.8 days |
The arithmetic, so you can check it: seats per day equals 59,466 divided by the load factor; spare equals seats minus 59,466; days equals 65,000 divided by spare.
What that actually means for a Monday flight
If the storm closed the airport on a Thursday in August, you are not waiting for the weather. You are waiting in a queue that is three days long at the industry's annual average load factor, and around six days long at the load factor August actually runs at.
Monday is day three or four of that. Your Monday flight being cancelled is not a surprise or a failure. It is the middle of the curve.
This also explains a detail that confuses people. During Khanun, Naha resumed operations after shutting for two days, and hundreds of flights were still cancelled after it reopened. The airport reopening is not the end of the disruption. On these numbers it is closer to the halfway point.
Three honest caveats
This is an estimate built from annual averages, and it should be read as an order of magnitude rather than a prediction. Three specific weaknesses are worth naming.
August is above the annual average for Naha, so dividing the year by 365 understates a real August day in both passengers and movements. That makes the comparison in step two conservative. The load factor figures are airline-wide rather than Naha-specific, and Okinawa routes in August will sit at the top of the range rather than the middle. And the model assumes every spare seat goes to a stranded passenger, which no airline can achieve, because spare seats are distributed across routes and times that do not match where the queue is. All three of those push the real recovery time up rather than down.
The practical conclusion survives all of them: plan a buffer of several days after a typhoon, not several hours, and do not treat the airport reopening as your departure date.
Sources for the figures above
- Naha Airport 2024 traffic: Naha Airport (21,705,234 passengers, 158,822 movements)
- Khanun cancellation and passenger counts: Al Jazeera, Typhoon Khanun knocks out power, grounds flights in Japan's Okinawa, 2 August 2023
- Naha reopening with flights still cancelled: Honolulu Star-Advertiser, Typhoon Khanun lashes Okinawa, grounding flights
- ANA domestic load factor, fiscal 2024: ANA HOLDINGS financial results
- JAL load factor figures: JAL Group consolidated financial results
Comments
Post a Comment