SkyscraperPage Forum

SkyscraperPage Forum (https://skyscraperpage.com/forum/index.php)
-   Diagrams (https://skyscraperpage.com/forum/forumdisplay.php?f=30)
-   -   Precision of height values determined with Google Earth (https://skyscraperpage.com/forum/showthread.php?t=245404)

Alpha Jan 8, 2021 2:51 PM

Precision of height values determined with Google Earth
 
How precise are height values determined with Google Earth? Is the precision for all areas, where 3D is available, the same or are there some differences?

Interesting comparision:

Stuttgart TV Tower ( http://skyscraperpage.com/cities/?buildingID=2661 ): 216.6 metres according entry, 216 metres according Google Earth

Nevertheless Google Earth has a problem with lattice towers like electricity pylons as such objects are not modelled properly and it does not work in rural areas as there are often no 3D-information available.

Anders Franzén Jan 8, 2021 4:04 PM

My experience is that it isn't too precise, but good enough.

In Google Earth you always substract the given ground level from the given height (above sea level) of the building. So it can always differ a meter or so.
An example:

Google Earth shows the ground level to be 10 m, when it is 9,6 m.

The height of a building is shown as 50 m above sea when it's 50.4 m.

This means that according to Google Earth, the building is 40 m, but in reality 40.8 m, or approx. 41 m.

Alpha Jan 8, 2021 7:16 PM

However, there is the question how good modelling works at Google Earth. I believe problems can also occur for objects situated at inclined area.

Artemco Jan 9, 2021 8:32 AM

my opinion is to ignore these values at all

Quilmeño89 Jan 9, 2021 9:37 AM

I have verified that the Google 3D Laser Scanning is extremely accurate (with rounded values). Only the official modeling. User-made is obviously not that reliable.

The 3D in Argentina is not the best (nothing comparable with the USA and Japan), however, it is incredibly accurate.

Examples from Buenos Aires City:

Building - Official Height (Google Earth Height)

• Le Parc Figueroa Alcorta - 168,8 m (169 m)

• Mulieris 1 and 2 - 164,4 m (164 m)

• Renoir II - 171,1 m (171 m)

• El Faro I and II - 160,2 m (160 m)

• Torre YPF - 160 m (160 m)

• Chateau Puerto Madero - 155,7 m (156 m)

• Mirabilia II - 142,2 m - 142 m


Also with lower buildings:

• World Trade Center I and II - 80,1 m (80 m)

• Buenos Aires Stock Exchange Annex - 71,2 m (71 m)

• AFIP - 51,4 m (51 m)

• Colonos Norte and Sur - 35,3 m (35 m)


Also in the city of Rosario we have:

• Maui I - 140,7 m (141 m)


And I have to say that many of the heights of Buenos Aires were wrong here and everywhere, because the architects or the publications showed wrong values. Google Earth revealed this and I got the official heights thanks to a survey of the city government, which endorsed the values obtained with GE.

So, in my opinion, Google Earth is more reliable than many official publications, than Emporis, and of course, than Skyscraper Center/CTBUH, which is a joke (I saw they put official heights based on my estimations in SSP —inventing decimals that I never gave, like 69.5 when my estimation was 70— and most of the heights they show are wrong, which is why I think it is one of the least reliable sources).

And about Emporis, I sent them information citing Google Earth as my source and they accepted it, using the values as the official architectural height.
So if you trust Emporis, you trust Google Earth (not to mention they have a lot of totally wrong information too).

I think it is preferable and much more useful to have a rounded height than to have a completely wrong height.

Quilmeño89 Jan 9, 2021 9:52 AM

Quote:

Originally Posted by Anders Franzén (Post 9153504)
My experience is that it isn't too precise, but good enough.

In Google Earth you always substract the given ground level from the given height (above sea level) of the building. So it can always differ a meter or so.
An example:

Google Earth shows the ground level to be 10 m, when it is 9,6 m.

The height of a building is shown as 50 m above sea when it's 50.4 m.

This means that according to Google Earth, the building is 40 m, but in reality 40.8 m, or approx. 41 m.

In Google Earth Pro you can measure the height of the building without having to depend on the elevations, and that allows greater precision, which can avoid that minimum error (which is also nothing, taking into account that in the diagram of Buenos Aires there are buildings up to 15 m apart from their actual height).

Oberhafenjunge Jan 9, 2021 10:18 AM

Quote:

Originally Posted by Quilmeño89 (Post 9154444)
[...](which is also nothing, taking into account that in the diagram of Buenos Aires there are buildings up to 15 m apart from their actual height).

Thats a more general problem around here, not only Buenos Aires. I also started a list of false heights to post them as corrections and suggested a few corrections already.

What is interesting about all this to me is: can a measurement in google maps/earth be a sufficient source to add the height to a building here at SSP?
There is quite a bit of incongruence regarding heights and the requirement of sources and there is a lot of buildings without specified heights. if measuring them in google would be regarded a viable source this could be an easy way to fill in many of the missing heights (due to lack of source) for many buildings

Anders Franzén Jan 9, 2021 10:19 AM

I guess you still have to find the lowest point on the ground to measure from, no?
I always treat Google Earth values as estimates, but yeah, architectural publications can be completely off. GE is better than them.
Many cities nowadays have made their building permit archives digitally available to everybody. Those would be the best sources imho.

About Emporis, the one data that I found are wrong very often is the construction year. Many times Emporis is one year off from every other sources.

Quilmeño89 Jan 9, 2021 10:28 AM

Quote:

Originally Posted by Oberhafenjunge (Post 9154451)
Thats a more general problem around here, not only Buenos Aires. I also started a list of false heights to post them as corrections and suggested a few corrections already.

What is interesting about all this to me is: can a measurement in google maps/earth be a sufficient source to add the height to a building here at SSP?
There is quite a bit of incongruence regarding heights and the requirement of sources and there is a lot of buildings without specified heights. if measuring them in google would be regarded a viable source this could be an easy way to fill in many of the missing heights (due to lack of source) for many buildings

Well, Dylan Leblanc usually puts Google Earth as the source. And since GE is more accurate than almost any other source, I think it's worth it (I think a couple of centimetres doesn't make a big difference, since even official sources can differ by 1 or 2 metres from each other).

Oberhafenjunge Jan 10, 2021 9:09 AM

Quote:

Originally Posted by Quilmeño89 (Post 9154453)
Well, Dylan Leblanc usually puts Google Earth as the source. And since GE is more accurate than almost any other source, I think it's worth it (I think a couple of centimetres doesn't make a big difference, since even official sources can differ by 1 or 2 metres from each other).

Thats good to know! I'll use this then in the future when no other reliable measurements are available!

Oberhafenjunge Jan 12, 2021 7:23 PM

Quote:

Originally Posted by Alpha (Post 9153424)
How precise are height values determined with Google Earth? Is the precision for all areas, where 3D is available, the same or are there some differences?

Interesting comparision:

Stuttgart TV Tower ( http://skyscraperpage.com/cities/?buildingID=2661 ): 216.6 metres according entry, 216 metres according Google Earth

Nevertheless Google Earth has a problem with lattice towers like electricity pylons as such objects are not modelled properly and it does not work in rural areas as there are often no 3D-information available.

Are we talking about the classic old google earth app or the new browser one?
If latter is the case, I would love how you can measure heights there since the tool, similarly to google maps only allows for horizontal measurements and not vertical ones...

koops65 Jan 12, 2021 7:51 PM

Many Toronto buildings in the database use a satellite based height system already...

Alpha Jan 14, 2021 12:33 AM

But are there satellite based height systems working also in rural areas? And how good do they work for lattice structures?

WhipperSnapper Jan 15, 2021 6:21 AM

It's aerial photography from planes and drones put through GIS mapping software. ESRI is, by far, the largest provider.


Measuring from google earth or other GIS providers will give you a truer height of a structure from a specific point than most official sources. It has become extremely accurate. Lattice shouldn't be any issue.

Quilmeño89 Jan 15, 2021 1:48 PM

Exactly. It reveals how wrong some sources are, but also how many of them are correct.

And I think it is important to show that height, because if not, an illustrator can intrerpret it as one more estimate and change it to his (or her) own estimate when drawing, leading to the wrong height.

Imagine a 33-storey building that, according to Google Earth, is 87 m tall because it has a low floor-to-floor height. But then an illustrator doesn't trust that supposed estimate and changes it to 100 m, giving each floor 3 m and rounding up.
That would lead to an unrealistic and imprecise diagram.

La_Parca Jan 15, 2021 8:35 PM

Also for example if you measure in Google Earth the tallests buildings in Barcelona you will realize that the "official" height is the height of the building over the sea level, but is not the real height

Thanks to Google Earth we can know the official height of various buildings around the world, and yes is very usseful to add and draw buildings to SSP with their real heights

Quilmeño89 Jan 15, 2021 11:19 PM

The same happened with some buildings in Buenos Aires, such as the Edificio República.

The height published everywhere was 99 m, but thanks to Google Earth we were able to know that its real height is 92 m and that 99 m was the height above sea level.
Then, the survey of the Buenos Aires City Government gave us the exact height: 91.8 m.

Oberhafenjunge Jan 16, 2021 11:25 AM

Measuring buildings from the sea level is odd... I didn't know that they did this anywhere?
Here in Germany it's usually everything upwards of the street level, that is measured/considered as height of a building. Here in Germany but also in other places in Europe the height of the eaves/the lower edge of a roof/top edge of the façade is also a big thing, since many building regulations, like in Berlin, determine a maximal height for the Eaves in certain areas etc. (In Berlin its their infamous 22m "Traufhöhe"/eaves height) Those things are also measured from street level.
I thought that was the usual way to measure those things...

Imagine Madrid or Zurichs "fake" Skyline with measurements being from sea level, both cities being more than 600 and 400m above the sea ... Goodbye Dubai ;)

When I draw or I request buildings I try to get hands on some sections or elevations of the buildings (which isn't always possible) and measure those things in CAD. If they divert more than +/-0,5m from the database I put them on my list to submit corrections at some point if I manage to finde more verifiable data...

Alpha Jan 18, 2021 7:46 PM

There are sometimes funny height definitions in some sources: I found one from South Korea in the internet, which said that the height of Stuttgart TV tower ( https://en.wikipedia.org/wiki/Fernsehturm_Stuttgart ) is 391 metres. They measured the height instead from its base from Schlossplatz ( https://en.wikipedia.org/wiki/Schlossplatz_(Stuttgart) )!

However height definitions of man-made structures using sea level as reference seem to be from some importance in aviation, as these values are published in most official air-traffic obstacle lists.

koops65 Jan 24, 2021 3:00 AM

Quote:

Originally Posted by WhipperSnapper (Post 9160552)
It's aerial photography from planes and drones put through GIS mapping software. ESRI is, by far, the largest provider.

I stand corrected...


All times are GMT. The time now is 10:55 AM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2024, vBulletin Solutions, Inc.