Proposal talk:Aerodrome Classification

From OpenStreetMap Wiki
Jump to navigation Jump to search

Edge cases

As is my tradition upon encountering a proposal to classify airports, I'm pulling out a few edge cases:

  • Allakaket Airport: no supporting buildings, no formal stands, gravel runway, used by light aircraft, a few thousand passengers a year. It would seem to be "aerodrome=basic", except the traffic is almost entirely commercial.
  • Grant County International: sizable supporting infrastructure, very small terminal, parking stands for hundreds of commercial aircraft, two long paved runways and three small ones, used by large and medium jets, almost exclusively general traffic (no passenger flights in the past 15 years). This one's all over the place in terms of classification.
  • Ellersport Airport: By my count, it's got at least thirteen hangars, but no other infrastructure beyond a windsock, no formal stands, a well-maintained paved runway, used by light aircraft for general traffic. It's unclear if this should be "basic", or if the paved runway and hangars are enough to move it up to "local" despite being used by only a dozen people.

--Carnildo (talk) 03:42, 22 January 2026 (UTC)


  • Allaket would be "aerodrome=basic" + "usage=commercial_aviation".
  • Sizable supporting (regional), small terminal (regional), hundreds of stands (global), 4 runways (global), used mostly by medium jets (continental), majority general traffic, 70 000 movements/600 passengers in 2024 (regional), seems to have only a couple connections (local). Grant County is a tough one, but it's closer to "aerodrome=regional". But I see how I should be more specific.
  • For Ellenport the runway would be enough but I see the issue, perhaps I should change the stands section. As it seems to be an airpark, the stands could be the "driveways", so it would be "aerodrome=local".
I will modify the descriptions a bit and get back to you.
--Telegram_Sam (talk) 04:20, 22 January 2026 (UTC)


I have cut the runway and stand criteria as they are often unreliable. I have also moved the usage criteria to their own section as that was confusing. How do they look now?
--Telegram_Sam (talk) 17:00, 22 January 2026 (UTC)
Looks better. But from my perpetual search for edge cases, how would you classify Memphis International? On the one hand, it has just a single seasonal international route and passenger numbers in the low millions; on the other, it's FedEx's global hub, making it the second-busiest cargo airport in the world. --Carnildo (talk) 05:35, 23 January 2026 (UTC)


I have thought about this issue. This is going to sound a bit weird, but take the annual transported cargo in kg and divide by 85 kg, which is the standard weight of a male adult for use in weight and balance for aircraft. This will give you a kind of passenger equivalent for cargo.
I have tested this and for most airports equivalent passengers will not change the total count by a substantial amount, however for cargo hubs it very much will. Memphis transported 3754236000 kg in 2024 which is equivalent to 44 million passengers, giving it the intercontinental rank. You can also check for large cargo terminals and all the big cargo planes.
--Telegram_Sam (talk) 06:09, 23 January 2026 (UTC)


Unfortunately, I don't think this classification system will work in reality. I have flown international commercial flights into many airports that see maybe only 4 - 10 arrivals and departures a day, all "continental" international flights (Bezierz, Lodz) with no local or national flights, or low volume (4 per hour max) aerodromes like Nuuk yet have 2,200 mile flights, not to mention when you start classifying private traffic.

When I was younger (my Googlefu is not working) I went to a tiny grass strip runway, which classed as an international runway, as they were accepting light aircraft from international departures, and had a "passport control" set up... which was little more than a folding table outside with a bloke sat under a sun umbrella. BUT... you had to book before you arrived. So this was a regional / continental aerodrome, seeing maybe 1000 flights a year, but access was not assumed to be yes, it was permissive only.

There is no mention of military aerodromes, even mention that that scheme doesn't cover them? Some military bases i used to fly from accepted civilian general aviation traffic, and had non-military units stationed there, like police and fire helicopters / observation aircraft.

I don't think that a classification that handles quantity / volume of traffic AND the destination charachteristics of those flights works, there are far too many variables that get muddled when being applied to Europe and Africa.

--O0235 (talk) 10:06, 22 January 2026 (UTC)

What's wrong with military=airfield + military_service=* ? service_branch=* can be further discussed.
—— Kovposch (talk) 14:07, 22 January 2026 (UTC)


  • Beziers had 260K movements / 280K passengers in 2024 (regional), has a small terminal (regional), small supporting infrastructure (local), small number of connections around a region of its continent (regional), aircraft types in use is mixed so inconclusive. Beziers would be classified as regional.
  • Lodz had 2500 movements / 430K passenngers in 2024 (regional), has a small terminal (regional), small supporting infrastructure (local), very small number of connnections in its continent (interpreted as regional), high passenger to movement ratio (172) means medium aircraft are being used (continental). Lodz would be classified as regional.
  • Had trouble finding Nuuk statistics, but Grønlands Statistik seems to say 11K passengers in 2024 (regional), has a small terminal (regional), small supporting infrastructure (local), small number of connectionals in a continental sized area (interpreted as regional), used by medium aircraft as it seems (continnental). Nuuk would be classified as regional.
  • The grass strip you mentioned has no paved runway (basic), had 1000 movements a year (local or basic), has few or no supporting buildings (basic), connections are unknown, used by light aircraft (local or basic). That grass strip would be classified as basic.
  • I saw military aerodromes as outside the scope of this proposal. Joint aerodromes are acceptable as they are now with a aeroway + military tag mix.
  • I actually based these characteristics based on aerodromes from Brazil, France, Malaysia and the USA. They should be able to be applied worlwide and if there are issues, they can modified slightly.
--Telegram_Sam (talk) 17:06, 22 January 2026 (UTC)

Structuring

Problems

  • aerodrome=* : Not the most suitable for this. Mixing different aspects (infrastructure, aircraft types, traffic volumes, connections, significance) makes it messy, difficult to classify and use. Criteria is furthermore subjective. importance=* already exists for the "influence". Proposal:Aerodrome#Tagging_an_international_aerodrome mentioned these ideas. Judges by passengers only, without freight.
    • *=global : It's often used as an advertising/marketing and even propaganda term. Best to avoid it.
      • 8. "Connected to airports all over the world": Above *=continental , does this needs to be connected to all continents?
  • usage=* : Vague special word should be avoided for specific aspects, and again reserved for characteristics difficult to categorize. If railway=* + usage=main is referred to, the closest analogy might be hubs, which would have to use the lowest standard any one airline calling it a hub for this depends on each airline.
    • *=*_aviation : Although GA is the whole term, it looks repetitive
    • A single GA vs Commercial distinction doesn't seem enough. It's more interesting and needed to know whether busy hubs accept plenty or limited GA traffic.

Suggestions

  • aerodrome=* →
  • usage=* : Could be ignored for now. It's difficult to organize. As others mentioned, emergency services, law enforcement, and government isn't handled, GA can be inferred from lack or low passenger=* and cargo=* implicitly in the meanwhile.


—— Kovposch (talk) 14:02, 22 January 2026 (UTC)


  • Mixing of different aspects is necessary because aerodromes are can be quite varied and many variables need to be taken into account. It is part of why they have been difficult to classify. The criteria I give in the proposal are meant to help reach a conclusion.
  • Can you elaborate on why "importance" should be used? From my point of view, it is so abstract and encompassing that it could be applied to anything. We could start classifying urban settlements and roads using it.
  • Understood. I have changed "global" to "intercontinental".
  • No, they only have to have connections outside a continental-sized area.
  • The purpose of the usage tag is to differentiate between aerodromes you use to travel and aerodromes you use for leisure, which is a practical application. Small aerodromes used for general aviation can be considered to be used for leisure while all others don't have those services or are too busy for them.
  • I understand "*_aviation" is repetitive but is it not useful for the tags to be readily apparent in midst of all the "usage" tags?
  • All public airports will accept GA traffic if you pay enough. Fees and scheduling are not in the scope of this proposal. The usage tag is meant to differentiate between aerodromes you use to travel and aerodromes you use for leisure. Aerodromes predominantly commercial are used for travel while those predominantly general are used for leisure.
  • The "passenger" tag cannot be used as the "aerodrome" tag is meant to take multiple variables into account.
  • Can you explain why the "basic" aerodrome tag should be removed? It is important to be able to tag airstrips.
  • The "cargo" tag could be used but it's outside the scope of this proposal.
  • In your last points you say infrastructure should be ignored and statistics are unacceptable, however these are the most important criteria for classifying an aerodrome. If you take these out, you have almost nothing and the proposal is basically just giving tags to be used at your own discretion. On top of that, most mappers would instintually be basing their own classification on possible knowledge of the airport's statistics.
--Telegram_Sam (talk) 17:28, 22 January 2026 (UTC)
aerodrome=* shouldn't be occupied by this. If you insist on aggregating the functional and physical significance, it should be limited to importance=* . This leaves aerodrome=* open for other use to be discussed. GA may be aerodrome=general , which seems the most characteristic, as "continental airport" isn't a term. Otherwise it has to invent eg aviation=general anew.
As I said, infrastructure can be determined from spatial join of the features inside. If this is too demanding and necessary, referring to capacity=* or platforms=* , you should add the number of runways, stands, and gates directly. The latter two would be useful as an initial level-of-detail before all individual aeroway=* features are added in micro. Terminals are difficult to quantify for different sizes. Aircraft types can be explicitly proposed separately.
Do have 5 TagInfo runways=* to begin with. stands=* and gates=* would be confusing, so have to be eg aerodrome:stands=* + aerodrome:gates=* (relying on plural form) / aeroway:parking_position:count=* + aeroway:gate:count=* ( *:count=* is more obvious, but aeroway:*=* is not ideal). This allows anyone to decide on each airport's significance is by everyone's own standards.
If you need traffic volumes, you should go as far as arguing for the stats to be added directly as population=* does. However, it won't be welcomed in OSM in general.
—— Kovposch (talk) 18:07, 23 January 2026 (UTC)


I'm aggregating function and form because they are both sides of the same coin. Aerodromes are an open book, you can predict the function of an aerodrome using a visual analysis and if it's at hand, a check of its statistical data. All the main tag criteria do is give users unfamiliar with aviation starting points on how to judge an aerodrome.
Yes I know the usage tag is a bit obtuse, but there are important exceptions to the size rule in terms of travel vs leisure and they need to be tagged somehow. I also know continental is an unusual term but it was the best generic bigger than regional term I could come up with.
I think you're misunderstanding how the criteria fit into the classification of the aerodromes. They are not hard rules and they are not meant to be added as data to OSM. They are guidelines on how to judge an aerodrome. I'm sure this type of micromapping would be useful but it's outside the scope of what I'm trying to accomplish.
Everyone is free to use their own judgement in choosing a tag as per the proposal. However unfamiliar users need a frame of reference when approaching an unknown topic and the guidelines help them make a decision.
Edit: I've been thinking about what you said about wanting to know if there is GA at an aerodrome. I now see you are right. It is too reductive to say an aerodrome focuses on one or the other. I have removed the usage section and replaced it with commercial_aviation and general_aviation tags.
--Telegram_Sam (talk) 18:46, 23 January 2026 (UTC)

Misunderstandings

  • access=*
    • access=no : "Aerodrome owner does not allow members of the general public to use the aerodrome." is access=private
    • access=permissive : It's a tolerated (full) access=yes without legal right. "Aerodrome owner allows limited access to the aerodrome to members of the general public after negotiation." should not be handled by access=* , and should be more specific in what's limited.
  • leisure=sports_centre : Not necessary on an aeroway=* feature. The sports don't match the "take place within an enclosed area" definition either. They don't happen inside. It's a base only.


—— Kovposch (talk) 14:03, 22 January 2026 (UTC)


Understood. I have changed both the access and sports sections.
--Telegram_Sam (talk) 17:11, 22 January 2026 (UTC)
Retrieved from "https://wiki.openstreetmap.org/w/index.php?title=Proposal_talk:Aerodrome_Classification&oldid=2947198"