Some dramatic video footage has emerged from a runway at Mumbai Airport, where two planes tried to take off and land on the same runway at virtually the same time. This incident is now being investigated by the Directorate General of Civil Aviation (DGCA) in India.
In this post:
An eye-catching runway video from Mumbai Airport
This incident happened on Saturday, June 8, 2024, at Chhatrapati Shivaji Maharaj International Airport in Mumbai (BOM). It involves two aircraft using runway 27R:
- A brand new Air India Airbus A320neo with the registration code VT-RTS was taking off, operating flight AI657 to Thiruvananthapuram (TRV)
- A two-year-old IndiGo Airbus A320neo with the registration code VT-ISV was landing, operating flight 6E5053 from Indore (IDR)
It’s common for the same runway to be used for takeoffs and landings, but it’s important that there’s sufficient separation between aircraft, to avoid two aircraft being on the runway at the same time. After all, you need to be prepared for the possibility that the aircraft taking off could abort takeoff, and that the aircraft landing could perform a go around.
While it’s not uncommon for a runway to see an aircraft land, and then see another aircraft take off a short time later, in this case there are just a couple of seconds between when the Air India jet takes off and the IndiGo plane lands. Suffice it to say that this is a bit too close for comfort.
Was this runway incident a big deal?
There’s no denying that this incident left a less than ideal buffer between aircraft, and that planes shouldn’t have been sequenced this closely. Air traffic controllers are under pressure to accommodate as many takeoffs and landings as possible, while still leaving a safe buffer. Obviously in this case, it was a bit too close to comfort.
That being said, is any party at fault here? What caused such little spacing here? Was the IndiGo jet approaching faster than expected, was the Air India jet slow to start its takeoff roll, or was this whole sequencing just overly optimistic?
I’ve seen some people suggest that the IndiGo jet should have performed a go around at the last minute. However, arguably that would have been even more dangerous. As an experienced pilot with tens of thousands of flight hours explained, the Air India A320neo was past V1 when the IndiGo jet landed, meaning it wouldn’t have aborted its takeoff anymore.
If the IndiGo plane had performed a go around, you would have had two jets airborne within very close proximity, which would have likely been even more dangerous. So arguably the decision of the IndiGo pilots to still land minimized the overall risk.
In response to this incident, the Air Traffic Controllers’ Guild of India has issued a statement, essentially saying that this was a non-event. The organization states that in good visibility, separation minimums can be reduced, and therefore there was “no air prox situation.”
The organization also stresses how controllers have stressful jobs, and how “the ATC has the discretion to allow arrival and departure within a few minutes on the same runway” (but… we’re talking about seconds, not minutes).
Bottom line
Airbus A320neos operated by IndiGo and Air India had a close call at Mumbai Airport on Saturday, as they were on the runway just seconds apart. While close sequencing is normal at major airports, this was definitely a bit too close for comfort. It’s nice to see a runway incident not be from the United States for once. 😉
What do you make of this runway incident at Mumbai Airport?
Is there anything good about India.....anything?
Assuming you’re not there, that’s definitely one thing that’s good about India but a negative for the world.
Way too many people in India! You should see the Trains.
US separation standards when it comes to these two aircraft would require 6000FT + Airborne before the landing aircraft crosses the runway threshold. Even if we as controllers were willing to let a situation with less than half that minima play out (that would be a terrible controller), the pilots would normally send themselves around before that.
If India's standards are different.... well, ok I guess.
But if any of you are among those calling...
US separation standards when it comes to these two aircraft would require 6000FT + Airborne before the landing aircraft crosses the runway threshold. Even if we as controllers were willing to let a situation with less than half that minima play out (that would be a terrible controller), the pilots would normally send themselves around before that.
If India's standards are different.... well, ok I guess.
But if any of you are among those calling the U.S standards unsafe because some airports (with ASDE) allow for landing clearances while aircraft are in position, then you sure as heck better not be defending this type of separation.
To put it another way: if you're aiming for certain separation, you're better off with a higher margin for error. This scenario leaves very little wiggle room.
Must add the disclaimer: "if it looks as close as it's made to look." ... and utilizing ATCs judgement of 3K feet. Maybe it was much more.
According to AVHerald it was 2130ft at the minimum:
https://avherald.com/h?article=519ac9a1&opt=0
As usual Sean M arrives with some sound info and common sense.
Aircraft separation standards are completely different for ground ops than once airborne.
Also separation minimums can vary by local procedure. Generally each country will have its own manual with base separation standards that apply country wide. These are then modified by unit procedures (with full approval of the regulator). For example sheer volume of aircraft and complexity of operations mean that...
As usual Sean M arrives with some sound info and common sense.
Aircraft separation standards are completely different for ground ops than once airborne.
Also separation minimums can vary by local procedure. Generally each country will have its own manual with base separation standards that apply country wide. These are then modified by unit procedures (with full approval of the regulator). For example sheer volume of aircraft and complexity of operations mean that what is appropriate at Albany or Syracuse doesn't work to keep planes moving at JFK or LGA.
It is very common for busy airports to have a procedure called "Land After" used when 2 landing aircraft are close together and the first landing a/c has not cleared the runway or "Land after the Departing" which is what is seen in this case where a landing a/c can be cleared with a departure on the takeoff roll.
In every case there are major limitations on when they can be used. Each airport is slightly different, but commonly can only be used in daylight with good visibility and there has to be an expectation that a certain distance of ft or meters will be kept.
All perfectly legal and approved by the relevant regulator.
One of the key components of these procedures is that the pilot given the clearance has to agree to it. A pilot can refuse a land after or land after the departing clearance if they are not comfortable.
This does look a little "snug" shall we say but at the end of the day the lander touched down once the departure was already airborne and its very unwise to judge horizontal distance solely from zoom lenses which as any photographer will tell you will fore-shorten and distort the view significantly - making objects further away seem much closer than they are in reality.
Using Google Maps to estimate the distance between the 2 aircraft when the Indigo touches down (just shy of the penultimate touchdown zone marker) and where the Air India is at the time (Notably already airborne so already legal at this point) which seems to be adjacent to the N6 and F1 rapid exits then that would make the distance between them almost 3,000ft horizontally (Just shy of 1km).
The absolute worst case scenario would be for the Indigo to have gone around. At that point you have 2 aircraft the landing aircraft going faster than the accelerating departure then flying on the same track with the 2nd aircraft gaining and overtaking the first. Neither pilot would be able to see the other and TCAS would be inhibited that close to the ground.
The Indigo a/c would be pitched up nose high so would lose the departure below them and the Air India crew would never be able to see the faster aircraft above and behind them.
So all in all I kind of suspect this is indeed a non-event. A Land after Departing clearance was given and accepted and normal margins were likely maintained or breached slightly. It's definitely tight, but almost certainly perfectly safe within the set criteria. And Sean is correct, this happens globally multiple times a day at airports that are busy and have mixed mode runway operations.
Oh and Ben, you seemed to question the Guild on using minutes instead of seconds ("ATC has the discretion to allow arrival and departure within a few minutes on the same runway” (but… we’re talking about seconds, not minutes)"..........Well their statement states they can clear 4 movements every 3 minutes, so that works out to 45 seconds per clearance so they are talking seconds and not minutes!
@ATC - the local minima for reduced separation at VABB are DRY runway, 3000m visibility, tailwind of 5kt or less, nil Windshear, and reasonable expectation that departing aircraft will have vacated runway before landing arrival crosses threshold. The investigation will determine if these have been met.
Excellent, thank you Sean.
I know at Gatwick and Heathrow there is a specific "Land After the Departing" clearance which is different and separate to the "Land After" and "Cleared to Land" clearances, do you know if VABB has similar or do they simply use "Cleared to Land"? I would be pretty surprised if they didn't with such intense single runway ops.
Such a clearance usually has specific conditions beyond general reduced visual...
Excellent, thank you Sean.
I know at Gatwick and Heathrow there is a specific "Land After the Departing" clearance which is different and separate to the "Land After" and "Cleared to Land" clearances, do you know if VABB has similar or do they simply use "Cleared to Land"? I would be pretty surprised if they didn't with such intense single runway ops.
Such a clearance usually has specific conditions beyond general reduced visual separation on the aerodrome.
For example at Gatwick these are the rules for land after a departing:
Special landing procedures may be in force in conditions hereunder, when the use
will be as follows:
- When the RWY-in-use is temporarily occupied by other traffic, landing clearance
will be issued to an arriving ACFT provided that at the time the ACFT crosses the
THR of the RWY-in-use the following separation distances will exist:
- Landing following departure - The departing ACFT will be airborne and at
least 2000m/1.1 NM from the THR of the RWY-in-use, or if not airborne, will be
at least 2500m/1.35 NM from the THR of the RWY-in-use.
- Conditions of Use
The procedures will be used by DAY only under the following conditions:
- When 26L/08R is in use;
- When the controller is satisfied that the pilot of the next arriving ACFT will be
able to observe the relevant traffic clearly and continuously;
- When the pilot of the following ACFT is warned;
- When there is no evidence that the braking action may be adversely affected;
- When the controller is able to assess separation visually or by radar derived
information.
Of course I'm not sure what the ops manual at VABB contains so it's impossible really to guess whether this whole thing was "legal" or not. It's certainly very close.
What the public need to understand about the job is that it's just not pleasant to be working at minimum separation 100% of the time with no margin for error. We'd love to be able to not have to keep doing this and space everything out more, but airlines and the traveling public want more and more flights, and there's a finite amount of runways and airspace to use.
If things were not always being pushed to the maximum efficiency then delays would go out of control and people would be up in arms. It can't work both ways.
For example in 2022 NATS in the UK handled 2,136,904 flights. If every flight was delayed 1 minute for additional protection that would equate to 35,615 hours of delays, and that would just be flights in the UK.
The FAA handles 45,000 flights a day, 16.4 million flights a year. Delay each of them by just a single minute and you have 273,750 hours of delays.
We can make the system less likely to have these kind of incidents, more separation can always be applied so that things like this don't happen. But the public need to be aware of the consequence of just what that would look like.
Go around is still the safest thing to do. Pilots behind call out and add a sidestep to give lateral separation. ATC will deconflict maneuvers from that point on. Most student pilots will experience this maneuver at least once on a busy day doing pattern work at their local airport. I would expect this has been practiced many times by the pilots of these aircraft.
This is closer than normal, but separation of 10-12 seconds between departures and arrivals is not uncommon at single runway airports like Mumbai or London Gatwick.
Certainly investigate the ATC instructions and the decision making process, but in the end this was something that happens multiple times daily at many busy airports. As long as all parties involved were fully aware of the situation (ie. ATC and both sets of pilots) and it didn't violate...
This is closer than normal, but separation of 10-12 seconds between departures and arrivals is not uncommon at single runway airports like Mumbai or London Gatwick.
Certainly investigate the ATC instructions and the decision making process, but in the end this was something that happens multiple times daily at many busy airports. As long as all parties involved were fully aware of the situation (ie. ATC and both sets of pilots) and it didn't violate any specified separation minima, it was just professionals doing their job.
I mean, there's a huge difference between 10-12 seconds between 1st aircraft airborne and second aircraft over threshold... vs 10-12 seconds of literal seperation between aircraft. (I.E, 10-12 seconds from hitting each other)
The Indigo should've went around, but not last second. They should've done it much earlier seeing the runway was still occupied.
If the Air India would've aborted takeoff, this might have been a fatal crash. Even if the Indigo would go around, it still takes a bit until the plane actually starts climbing again.
A lot of should've, might've here. But worst for me is the reaction of the ATC Guild. With this attitude,...
The Indigo should've went around, but not last second. They should've done it much earlier seeing the runway was still occupied.
If the Air India would've aborted takeoff, this might have been a fatal crash. Even if the Indigo would go around, it still takes a bit until the plane actually starts climbing again.
A lot of should've, might've here. But worst for me is the reaction of the ATC Guild. With this attitude, it's just a matter of time until a crash happens. Still hoping there will be an actual investigation into the occurence.
Did you not read the part of the article that said the IndiGo plane landed when the Air India plane was past the point at which it could have aborted takeoff?
Did you not read the part in my comment that said a go around will take a while until effective (any normal person would understand that even going around a few seconds earlier wouldn't help). Besides, a) rejected takeoffs beyond V1 have happend in the past and b) the Indigo crew wouldn't know at all when the other plane reached V1.
Your comment is a great example of "nobody died, so let's just continue...
Did you not read the part in my comment that said a go around will take a while until effective (any normal person would understand that even going around a few seconds earlier wouldn't help). Besides, a) rejected takeoffs beyond V1 have happend in the past and b) the Indigo crew wouldn't know at all when the other plane reached V1.
Your comment is a great example of "nobody died, so let's just continue and not question anything". Great example of what's wrong in some parts of the world.
I've explained above why Indigo going around would have been the most dangerous possible outcome. Both aircraft would lose sight of each other and TCAS would be unable to intervene.
If the Air India aborted above V1 that would make everything moot anyway because at that point he won't be stopping on the runway anyway!
I can only think of 2 rejects above V1 I have ever heard about, a Spantax DC10 in...
I've explained above why Indigo going around would have been the most dangerous possible outcome. Both aircraft would lose sight of each other and TCAS would be unable to intervene.
If the Air India aborted above V1 that would make everything moot anyway because at that point he won't be stopping on the runway anyway!
I can only think of 2 rejects above V1 I have ever heard about, a Spantax DC10 in Malaga and the USA Jet MD82 at Willow Run. Both of them went off the end of the runway.
In situations like these keeping the other aircraft in visual sighting is critical to avoiding something way worse.
People in aviation question things all the time, it's got nothing to do with anyone burying their head in the sand. There's always things that need being looked at again and changed if necessary. As I also explained above this was all very likely completely legal, determined as safe in advance by the regulator and importantly considered as safe by the pilot involved.
That is a wild guess. In reality for the A320 family on a dry runway decently long runway or sea level V1 and VR are usually the same or within a knot or two…. Either way it’s not something ATC would know.
Good luck finding the truth in India, where lying is a national sport.
…in a sport where the US is the champ.
Yeah just like how we found the WMDs in Iraq.
Like all the major sports in US, winning the domestic title gives you the entitlement of being world champions.