Heathrow - separation distance/time ?

[Daytona]

I was mucking about with satellite mapping tools and at my first
attempt with Microsoft's tool I accidently got a nice picture of a 747
in flight, so I played some more and found this -
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>

One runway, 2 planes coming in to land, 2 on the runway. I thought
that there was meant to be 3 minutes separation ? Is this normal
accepted/good practice ?

Daytona

[Ross Younger]

Post by Daytona
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>
One runway, 2 planes coming in to land, 2 on the runway. I thought
that there was meant to be 3 minutes separation ? Is this normal
accepted/good practice ?

What appears just doesn't seem legal by my understanding of the law - but
Google's satellite images can be deceiving. What you get from Google is
a composite of a great many satellite pictures, taken at varying times
by various satellites. I think that what you're seeing there is the
same aircraft appearing in four consecutive frames taken by a satellite
passing overhead Heathrow. (It may also be four different aircraft at
different times, but I have to say they do look rather similar, and the
spacing between them is very regular.)


Ross

[Peter]

Post by Ross Younger

Post by Daytona
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>
One runway, 2 planes coming in to land, 2 on the runway. I thought
that there was meant to be 3 minutes separation ? Is this normal
accepted/good practice ?

What appears just doesn't seem legal by my understanding of the law - but
Google's satellite images can be deceiving. What you get from Google is
a composite of a great many satellite pictures, taken at varying times
by various satellites. I think that what you're seeing there is the
same aircraft appearing in four consecutive frames taken by a satellite
passing overhead Heathrow. (It may also be four different aircraft at
different times, but I have to say they do look rather similar, and the
spacing between them is very regular.)
Ross

That image is completely impossible as a real time representation of
aircraft spacing.

I wonder if the imagery involves a synthetic aperture. This uses
multiple images, separated in time but crucially done from different
camera positions, which are then correlated. The resolution
improvement is of the order of the square root of the number of images
correlated so e.g. if you take 100 pics you get a 10x resolution
improvement. It also gets rid of atmospheric distortions, on a similar
ratio.

These pics appear to have been taken about 10 secs apart. One could
work it out - the jets are doing about 140kt just before landing.

This is done commonly with airborne/spaceborne radar (where moving the
imaging antenna through say 10km yields the resolution of a 10km
diameter antenna) and it's done with radio astronomy all the time but
I was not aware it was done with the visible spectrum.

It requires massive processing power, bandwidth, and storage.

One would expect to see the same effect with cars on the roads, but I
see that somebody has painted them over :)

What suprises me is that nobody has yet done this with the JFK
assasination movie :)

[Stephen]

Post by Peter
I wonder if the imagery involves a synthetic aperture. This uses
multiple images, separated in time but crucially done from different
camera positions, which are then correlated. The resolution
improvement is of the order of the square root of the number of images
correlated so e.g. if you take 100 pics you get a 10x resolution
improvement. It also gets rid of atmospheric distortions, on a similar
ratio.

Does this explain why some aircraft in flight appear to have white shadows?
There's one at 53 24'56.62"N 6 10'10.98"W. This is the only one I can
find quickly but unfortunately it's in the middle of some other hideous
imaging artefact that has recently appeared.

Stephen

[Daytona]

Post by Stephen
There's one at 53 24'56.62"N 6 10'10.98"W.

<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=53+24%2756.62%22N+6+10%2710.98%22W&ie=UTF8&ll=53.415722,-6.169724&spn=0.009131,0.026951&t=k&om=1>

[Stephen]

Some people seem to doubt that what we're seeing is multiple images of the
same aircraft. Here's another example:
52 12 22.17N 0 6 45.71W This is a glider tug and both images are the same
aeroplane - Cambridge Gliding Club only have one PA25. If you zoom right in
you can see the rope but there's no sign of the glider though.

(After going to this point zoom in further and look for the images to the
east and west of the point. They are 0.28 miles apart. The eastern image
is obscured by the label Google Earth puts on it so it's best to hide that
when you get there.)

Stephen

[Greg]

Post by Stephen
Some people seem to doubt that what we're seeing is multiple images of the

Surely these multiple images are simply due to the way the image was
photographed in strips?, if something is moving in the right direction at
the right speed it'll appear on more than one strip.
Greg

[Peter]

Post by Greg
Surely these multiple images are simply due to the way the image was
photographed in strips?, if something is moving in the right direction at
the right speed it'll appear on more than one strip.

I better get my coat. You may well have the answer there :) :)

[Daytona]

On Sun, 16 Jul 2006 22:29:23 +0100, Peter

Post by Peter

Post by Daytona
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>

That image is completely impossible as a real time representation of
aircraft spacing.

Not impossible, but unlikely I hope ! :-)

The gaps, from left to right are -

1-2 ~580m
2-3 ~570m
3-4 ~470m

I'm a little surprised by the gap between the 2nd (flying) & 3rd
(landed).

Post by Peter
I wonder if the imagery involves a synthetic aperture. This uses
multiple images, separated in time but crucially done from different
camera positions, which are then correlated. The resolution
improvement is of the order of the square root of the number of images
correlated so e.g. if you take 100 pics you get a 10x resolution
improvement. It also gets rid of atmospheric distortions, on a similar
ratio.

Interesting, I wasn't aware of that, I'm surprised at the increase in
resolution.

Post by Peter
These pics appear to have been taken about 10 secs apart. One could
work it out - the jets are doing about 140kt just before landing.
This is done commonly with airborne/spaceborne radar (where moving the
imaging antenna through say 10km yields the resolution of a 10km
diameter antenna) and it's done with radio astronomy all the time but
I was not aware it was done with the visible spectrum.
It requires massive processing power, bandwidth, and storage.
One would expect to see the same effect with cars on the roads, but I
see that somebody has painted them over :)

What do you mean ? Looking at adjacent areas I can see some large
vehicles with white roofs that seem to bring this into question. The
question is whether they are taken using the same method, and if not,
why not ?

Daytona

[Peter]

Post by Daytona
Interesting, I wasn't aware of that, I'm surprised at the increase in
resolution.

I don't think they take anywhere near 100 images of the same spot
though.

Post by Daytona
What do you mean ? Looking at adjacent areas I can see some large
vehicles with white roofs that seem to bring this into question. The
question is whether they are taken using the same method, and if not,
why not ?

The section of the area that URL showed had a street/road map
superimposed, covering up any vehicles on the roads.

Those jets could have landed as shown, but this would never be done in
civil aviation :) You don't get a "clear to land" until the previous
one is off the active runway.

[David Wright]

Post by Daytona
Not impossible, but unlikely I hope ! :-)

Zoom right in to maximum - all but one of the aircraft "disappear".

D.

[David Wright]

Post by David Wright
Zoom right in to maximum - all but one of the aircraft "disappear".

Scrub that - looking at the wrong runway lol

They are all, however, the same plane - Air France it says on it.

D.

[Richard Herring]

Post by Peter

Post by Ross Younger

Post by Daytona
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF
8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>
One runway, 2 planes coming in to land, 2 on the runway. I thought
that there was meant to be 3 minutes separation ? Is this normal
accepted/good practice ?

What appears just doesn't seem legal by my understanding of the law - but
Google's satellite images can be deceiving. What you get from Google is
a composite of a great many satellite pictures, taken at varying times
by various satellites. I think that what you're seeing there is the
same aircraft appearing in four consecutive frames taken by a satellite
passing overhead Heathrow. (It may also be four different aircraft at
different times, but I have to say they do look rather similar, and the
spacing between them is very regular.)
Ross

That image is completely impossible as a real time representation of
aircraft spacing.
I wonder if the imagery involves a synthetic aperture. This uses
multiple images, separated in time but crucially done from different
camera positions, which are then correlated. The resolution
improvement is of the order of the square root of the number of images
correlated so e.g. if you take 100 pics you get a 10x resolution
improvement. It also gets rid of atmospheric distortions, on a similar
ratio.
These pics appear to have been taken about 10 secs apart. One could
work it out - the jets are doing about 140kt just before landing.
This is done commonly with airborne/spaceborne radar (where moving the
imaging antenna through say 10km yields the resolution of a 10km
diameter antenna) and it's done with radio astronomy all the time but
I was not aware it was done with the visible spectrum.

And it's not likely to be, unless you can think of a way of maintaining
phase coherence at optical frequencies between all the measurements :-/

Post by Peter
It requires massive processing power, bandwidth, and storage.
One would expect to see the same effect with cars on the roads, but I
see that somebody has painted them over :)
What suprises me is that nobody has yet done this with the JFK
assasination movie :)

[Peter]

Post by Richard Herring
And it's not likely to be, unless you can think of a way of maintaining
phase coherence at optical frequencies between all the measurements :-/

No need for that at all.

[Richard Herring]

Post by Peter

Post by Richard Herring
And it's not likely to be, unless you can think of a way of maintaining
phase coherence at optical frequencies between all the measurements :-/

No need for that at all.

OK, so go ahead and tell us why _you_ think synthetic-aperture
techniques aren't in routine use for optical imaging.

[Peter]

Post by Richard Herring
OK, so go ahead and tell us why _you_ think synthetic-aperture
techniques aren't in routine use for optical imaging.

SA might be the wrong term for correlating multiple images to remove
noise, but the technique works for optical wavelengths too. It relies
on the camera and/or the object moving a bit between images.

I was speculating that the NASA imagery (which is what most of google
earth and nasa world wind uses) used this technique but perhaps they
don't need to.

[Richard Herring]

Post by Peter

Post by Richard Herring
OK, so go ahead and tell us why _you_ think synthetic-aperture
techniques aren't in routine use for optical imaging.

SA might be the wrong term for correlating multiple images to remove
noise,

It is, unless you're correlating the full complex signal, which isn't
available from most optical sensors ;-)

Post by Peter
but the technique works for optical wavelengths too. It relies
on the camera and/or the object moving a bit between images.

Or the intervening atmosphere moving. That's image stacking or speckle
interferometry.

Post by Peter
I was speculating that the NASA imagery (which is what most of google
earth and nasa world wind uses) used this technique but perhaps they
don't need to.

I'm too hot to sit down and do the sums ;-(

[david]

Well not much to add here except: late lanbding clearences are very very
common...3 minutes? ha! Forget it. At LGW we often get very close to the
Cat 1 minimum of 200ft before we get the clearance to land, and then just as
the preceding aeroplane clears the runway. So, his tail is just off, tower
tell him to go to ground then they tell us to land. bam, bam, bam! Thats it.

Or immediately the preceding a/c's gear has left the ground!

It gets VERY close.

BUT...not as close as the picture shows! They'd not clear to land with one
(two!) still on.


David

Post by Daytona
I was mucking about with satellite mapping tools and at my first
attempt with Microsoft's tool I accidently got a nice picture of a 747
in flight, so I played some more and found this -
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>
One runway, 2 planes coming in to land, 2 on the runway. I thought
that there was meant to be 3 minutes separation ? Is this normal
accepted/good practice ?
Daytona

[Clive]

Post by Daytona
I was mucking about with satellite mapping tools and at my first
attempt with Microsoft's tool I accidently got a nice picture of a 747
in flight, so I played some more and found this -
<URL:http://maps.google.co.uk/maps?f=q&hl=en&q=Heathrow+airport&ie=UTF8&ll=51.477948,-0.479558&spn=0.009542,0.026951&t=h&om=1>
One runway, 2 planes coming in to land, 2 on the runway. I thought
that there was meant to be 3 minutes separation ? Is this normal
accepted/good practice ?
Daytona

It's the same aircraft (Air France) - been discussed many times on various
sites

Clive