Proposal talk:Aeroway=stopway

From OpenStreetMap Wiki
Jump to navigation Jump to search

Isn't a stopway or blast pad part of the runway?

It seems that many mappers are including this area as part of the aeroway=runway - sometimes adding the tag runway=blast_pad (see Is this incorrect? --Jeisenbe (talk) 04:47, 10 November 2020 (UTC)

Almost nothing is incorrect in OSM. It's just different approaches to capturing the same information in data :) The problem I see with using a secondary sub-tag to aeroway=runway is that most data usage and renderers seems to only check for the primary aeroway=runway and are unaware that subtags exist and they might be looking at another part of a runway. Or to put it differently: A tag used since the beginning of OSM was redefined to mean "either runway or stopway or displaced_threshold". I see an analogy to aeroway=runway where the service=* subtag allows to specify the type. If we would decide for this approach there should also be a subtag value for the core runway, e.g. runway=main alongside runway=stopway and runway=displaced_threshold. But this would mean that 53000+ occurances of aeroway=runway would need to be reviewed and updated accordingly to be sure that this is a main runway. I therefor think the cleaner approach is introducing distinct tags aeroway=stopway and aeroway=displaced_threshold.
And as to whether stopways/blast pads are part of the runway: I guess that's a philosophical question since planes shouldn't land on them. --Claudius (talk) 10:35, 10 November 2020 (UTC)
The US FAA defines a stopway as "An area beyond the takeoff runway no less wide than the runway and centered upon the extended centerline of the runway, able to support the airplane during an aborted takeoff, without causing structural damage to the airplane, and designated by the airport authorities for use in decelerating the airplane during an aborted takeoff." (AIM PCG S-7) The FAA also defines a runway as "A defined rectangular area on a land airport prepared for the landing and takeoff run of aircraft along its length." (AIM PCG R-6) That fairly clearly excludes stopways from the runway, so introducing a separate aeroway=stopway tag makes sense. It would also seem to exclude displaced thresholds, but displaced thresholds are defined as being "on the runway" (AIM PCG D-3), so tagging displaced thresholds as part of the runway might make more sense. --AntiCompositeNumber (talk) 18:35, 11 November 2020 (UTC)
Thanks for providing these FAA definitions. I've updated the "Rationale" section of the proposal to include this. --Claudius (talk) 08:03, 12 November 2020 (UTC)
In addition to the information above, about the displaced treshold. To remention the FAA, they define a runway as "A defined rectangular area on a land airport prepared for the landing and takeoff run of aircraft along its length." (AIM PCG R-6) For the displaced treshold the FAA states "Displacement of a threshold reduces the length of runway available for landings." (AIM 2.3.4) So if I interpret this correctly you're not allowed to land on the displaced treshold. If that is the case then technically the displaced treshold does not fit the definition of a runway (despite it being metioned as "on the runway" (AIM PCG D-3)). Then it should be aeroway=displaced_treshold as well. --E de Wit (talk) 22:56, 24 November 2020 (UTC)
A displaced threshold may be used for takeoffs, and a displaced threshold at the far end of the runway can be used when slowing down after landing. Displaced thresholds are considered part of the runway. --BigPeteB (talk) 23:11, 3 March 2021 (UTC)

Several questions

This looks good. The tag is already in use, and its usage should be pretty straightforward anyway. Good job looking up the FAA and ICAO definitions to clarify how this relates to and is distinct from the runway.

You suggested that the direction of the way should point towards the runway. While I like the consistency, is this necessary? The stopway should have exactly one end that shares a node with the end of a aeroway=runway; the chevrons must necessarily point towards that runway. I don't see how the directionality of the way should matter. If the purpose of its directionality is only so that the chevrons would be rendered following the direction of the way, I'd consider that tagging for the renderer and an error in the design; there would never be a scenario in the real world where a stopway would run "backwards" like that.

(If we do keep this requirement, I think the requirement as stated is overly precise. It's redundant to say it points "towards the closest threshold of the runway it belongs to." Stopways are only ever at the end of a runway, so both thresholds would be in the same direction from the stopway. You could simply say it points "towards the runway it belongs to.")

Have you (or has anyone else) considered how EMAS ([W] Engineered materials arrestor system) would be tagged? The only reference I find to them is a brief mention on Talk:Tag:aeroway=runway#runway=blast_pad. Looking at the FAA's AIM, figure 2-3-42 near the bottom clearly depicts an EMAS that overlaps with the stopway/blastpad. That would preclude using a tag like aeroway=EMAS (and I oppose combining tags like aeroway=stopway;EMAS). The next obvious tag might be runway=EMAS, except that as you already showed, a stopway is not part of a runway. :-/

Come to think of it, is there even a need to go through a proposal and approval process? Although the proposed aeroway=stopway isn't listed on aeroway=*, it is mentioned on Aeroways.

--BigPeteB (talk) 19:52, 1 December 2020 (UTC)

P.S.: Ouch, I just found the FAA's Quick Reference to Airfield Standards. According to this, it's clear that a "blast pad" and "stopway" are not the same thing, as has been previously stated at Talk:Tag:aeroway=runway#runway=blast_pad. It doesn't provide a convenient all-in-one definition for either term, but what I gather is that
  • A blast pad is solely to provide protection against jet blast but is not suitable as a movement area, while a stopway is designed to support an aircraft. (Quote from page 36: "Stopways look like blast pads but are considered full-strength pavement and are suitable to support aircraft during an aborted take-off.")
  • The widths of the stopways and blast pads are not the same. Stopways equal runway width. Blast pads equal runway width plus runway shoulders. (Quoted from page 12)
  • Blast pads are unlit, while stopways may have red edge lights.
  • Blast pads are not included in calculated/declared runway stopping distances, while stopways are.
This makes the OSM rule of "tag what's on the ground" a little difficult to apply in practice. It seems there is a distinction between a blast pad and a stopway, but it would require some research for a mapper to figure out which is which. Mapping by satellite imagery isn't sufficient to distinguish the two. The presence of lighting would positively indicate a stopway, but that lighting is optional. Presumably you could use the declared distances to determine whether it's a blast pad or stopway, although that's complicated by all the other runway areas and restrictions that go into those calculations.
I then found FAA Advisory Circular 150/5300-13A, Section 102 which clearly defines the term and explicitly states that they are not equivalent:
  • Blast Pad. A surface adjacent to the ends of runways provided to reduce the erosive effect of jet blast and propeller wash. A blast pad is not a stopway.
  • Stopway (SWY). An area beyond the takeoff runway, no less wide than the runway and centered upon the extended centerline of the runway, able to support the airplane during an aborted takeoff, without causing structural damage to the airplane, and designated by the airport authorities for use in decelerating the airplane during an aborted takeoff. A blast pad is not a stopway.
I'm a little torn as to whether we ought to care about the distinction. On the one hand, Aviation#Rationale makes a clear argument for why OSM should not try to map some features used for aviation. The concern I find most compelling there is that anyone who would care about the difference is probably a pilot or an airport engineer, and both of them legally must use official sources of information to do their jobs rather than OSM. On the other hand, the OSM rule is to "map what's observable on the ground", although I'm not sure if this distinction counts as "observable"; at worst a mapper would need some expertise in materials engineering, get access to the site, and examine the material in question, which is obviously impractical. And it would certainly be confounded if there's ever an instance of something built to the capabilities of a stopway but declared and used only as a blast pad; would that count as a difference you couldn't observe from the ground, or is that along the lines of a political boundary defined by law that still falls within OSM's scope? --BigPeteB (talk) 20:18, 1 December 2020 (UTC)
Thanks you for going all the way down the rabbit hole and checking FAA documentation. I was only looking at ICAO documentation since I tried to make it an international and not a US American definition. But your findings explain why the majority of blast_pad tagging usage (from what I could tell remotely used for both genuine stopways and blast pads alike) is used on US airports.
Regarding the direction: I came across a few airports with runways located in a crossing angle where stopways connect the end of one runway with another runway. That's why I tried to be overly explicit in defining the direction. But good call out for simplifying the language for OSMs international audience. In the next proposal revision I will change to your suggested "towards the runway it belongs to".
As for EMAS: I have been using a subtag stopway=EMAS for the 5 occurances I found documented since they seemed to fall into a "special kind of stopway" group. BUT: Given that we have established that there are three distinctively different types of "runway end aeroways" I am now considering to expand this proposal to document three tags: aeroway=stopway aeroway=blast_pad aeroway=EMAS. From what I can tell from looking at various international airports the distinction would be:
  • stopways: full runway width and most of the time include chevron markings (apart from most airports in the former Eastern Block)
  • blast_pads: often not full runway+shoulder width and never chevron markings
  • EMAS: tile patterns on the surface, no chevron markings but occasionally yellow outline marking
Do you think this would make sense or overcomplicate the matter?
--Claudius (talk) 09:33, 4 December 2020 (UTC)
Well, your descriptions aren't quite right. Stopways and blast pads are both marked with yellow chevrons. And the definitions seem to say that a blast pad includes runway shoulder width (which makes sense, as engines for the largest aircraft like the A380 or An-225 nearly overhang a standard 150ft/45m runway), while a stopway is only meant for the wheels of the aircraft and is at least runway width but may be wider.
Your description of EMAS isn't right, either. They also have yellow chevrons (just like stopways and blast pads). True, some types of EMAS are visually distinguishable as they're made from lighter-colored concrete produced in small tiles, but it's an area of ongoing research, and new types (or even some existing types) may appear the same as the non-EMAS parts of a stopway or blast pad.
I'll point again to AIM 2-3, figure 2-3-42. The image shows an EMAS at the end of a stopway or blast pad. The whole thing is marked with yellow chevrons, which doesn't tell us anything. It's runway width + shoulders, which also doesn't tell us whether it's a wide stopway or just a blast pad. The runway end indicator lights are mounted on short poles to place them level with the runway, which suggests to me that this is probably a blast pad (meant only to protect against jet blast and to possibly stop landing aircraft in an emergency while possibly damaging them in the process) rather than a stopway (meant for stopping aborted takeoffs without damage to the aircraft). The only way to verify it, though, would be to figure out what airport this is and determine whether the declared runway lengths include this area or not.
I did check AC 150/5220-22, which clarifies EMAS a bit. "An EMAS is not intended to meet the definition of a stopway... When a stopway is provided, [the runway safety area and runway object free area] lengths begin at the stopway end." An EMAS does have to be designed to withstand jet blast. Thus, I think we can safely say that an EMAS is never part of a stopway, and moreover it is always part of a blast pad. If you agree with that conclusion, then unfortunately it means that tagging these as aeroway=blast_pad and aeroway=EMAS doesn't work (unless we want to allow aeroway=blast_pad;EMAS, which I do not). --BigPeteB (talk) 21:18, 16 December 2020 (UTC)
I looked at "ICAO Annex 14 - Volume I - Aerodrome Design and Operations Seventh Edition, July 2016". The only occurance of the term "blast pad" in the document is in section 3.4 "runway strips":
3.4.11 Recommendation.That portion of a strip to at least 30 m before the start of a runway should be prepared against blast erosion in order to protect a landing aeroplane from the danger of an exposed edge.
Note 1.— The area provided to reduce the erosive effects of jet blast and propeller wash may be referred to as a blast pad.
So summarizing the exchange here on the talk page and FAA's stopway definition it sounds like a stopway is the part of a blast pad that has full strenght pavement and is therefor suitable as a movement area. So as a taxonomy: blast pad > stopway > EMAS
I will focus with this proposal on stopways as a subsection of blast pads only. That means there is a potential for an additional tag e.g. aeroway=blast_pad for the prepared area beyond the runway+stopway that can withstand jet blast. I will consider this out of scope since it is too hard to tell them apart.--Claudius (talk) 11:21, 23 February 2021 (UTC)
Uh, no, that still isn't correct. Stopways can support an aircraft without damaging the aircraft or the surface. EMAS and blast pads cannot. The stopway=EMAS tag that you added to this proposal therefore makes no sense, and directly contradicts FAA definitions. --BigPeteB (talk) 23:03, 3 March 2021 (UTC)
Thanks once again for checking again. After reading through it once more indeed EMAS are not special versions of stopways and thus are out of scope for this proposal. Instead they would deserve their own aeroway=EMAS tag most likely. --Claudius (talk) 22:15, 2 April 2021 (UTC)