How do they figure out elevation?

Cpt_Kirk said:
p4pb3877659.jpg
Click to expand...

since I grew up there - the funniest thing about that is nobody remembers that there's yet another state between those butts and the Canadian border. Poor, poor Nor'Dakota ... never gets any respect. :D

LDJones said:
That would actually be the Wyoming side.

But still very funny!
Click to expand...

you are correct, sir! :)
 
Last edited:
The altimeter is very inaccurate compared to GPS. Someday the FAA will ban them as too inaccurate, I hope. :D
 
Capt. Geoffrey Thorpe said:
Nope.

To use a barometric device to determine elevation / altitude you first need to correct for the local "sea level" barometric pressure which changes as "highs" and "lows" roll through independent of the temperature - that is done by measuring the pressure at a known elevation and applying the standard correction for the known elevation. So, basically, if you know your elevation, you can measure the pressure, and if you don't wait too long you can measure the pressure again and determine that you haven't moved since the last measurement.
Click to expand...

Thanks for the correction. So it really does come down to brute force measurement, then? Pretty interesting.
 
So can they now measure with extreme precision using satellites or are there still errors that work their way in?

Is the base refrence (MSL) even a precise known or is that still subject to deviation? And where is the standard marked? Lots of places or is there a master refrence somewhere that calibrates the whole system?
 
gkainz said:
since I grew up there - the funniest thing about that is nobody remembers that there's yet another state between those butts and the Canadian border. Poor, poor Nor'Dakota ... never gets any respect. :D
Click to expand...

Actually it's Wyoming (as previously mentioned) and Montana that would be between the butts and Canada. The faces look south east. The "ass-end" so to speak of the Black Hills National Forest borders Wyoming. The only thing that would end up in North Dakota would be the presidential farts if there was a prevailing westerly wind.
 
It's a computed value. That's what the M is, and yes the USGS fixes what it is. Note that that isn't necessary where the water is. There are tidal datums for various bodies of coastal water in addition the the geographical datum.
 
flyingron said:
It's a computed value. That's what the M is, and yes the USGS fixes what it is. Note that that isn't necessary where the water is. There are tidal datums for various bodies of coastal water in addition the the geographical datum.
Click to expand...

I know what the 'M' is and I know it's computed. What I'm asking is if there is a marker somewhere that is master reference for the whole shebang?

I've seen markers nailed down in the earth for lat/long and I can wrap my mind around that. But up down seems pretty hard to mark with any great precision as the reported THREE markers on the steps of the Capitol in Denver point out.
 
MAKG1 said:
Yes, that's corrected GPS. The correction had to come from somewhere. The earth is not an ellipsoid on the scale of a few feet.
Click to expand...

The technical name for the earth's shape in an oblate sphereoid. ask any geologist. :yes:
 
Captain said:
So can they now measure with extreme precision using satellites or are there still errors that work their way in?

Is the base refrence (MSL) even a precise known or is that still subject to deviation? And where is the standard marked? Lots of places or is there a master refrence somewhere that calibrates the whole system?
Click to expand...

I was reading the history of the TRANSIT system, the predecessor to GPS. They had to put in offsets for all of the major geographic areas that were not tied together by surveys. And they found out that Hawaii was over a kilometer from where it was thought to be.
 
MAKG1 said:
An oblate spheroid is a special case of an ellipsoid, where the two larger principal axes are equal.

The earth isn't really either at the scale you want.
Click to expand...

At the scale you're talking about, there's no name for that shape that I'm aware of.
 
Captain said:
I know what the 'M' is and I know it's computed. What I'm asking is if there is a marker somewhere that is master reference for the whole shebang?

I've seen markers nailed down in the earth for lat/long and I can wrap my mind around that. But up down seems pretty hard to mark with any great precision as the reported THREE markers on the steps of the Capitol in Denver point out.
Click to expand...


See my previous posts. What, you thought the Earth was static?:dunno:
 
Captain said:
I know what the 'M' is and I know it's computed. What I'm asking is if there is a marker somewhere that is master reference for the whole shebang?

I've seen markers nailed down in the earth for lat/long and I can wrap my mind around that. But up down seems pretty hard to mark with any great precision as the reported THREE markers on the steps of the Capitol in Denver point out.
Click to expand...
Not a single reference point, but there are many benchmarks (especially near the coasts) that have the sea level datum calcualated for them. that serve as a starting point.
 
warthog1984 said:
See my previous posts. What, you thought the Earth was static?:dunno:
Click to expand...

The earth is static (more or less) except when it moves.

I'm not being obtuse here, I'm saying that obviously if there were an earthquake in Denver rising the ground 15 feet then the marker would need to be moved when they rebuild the Capitol.

But I'm discounting that. Just taking the earth as is in a snapshot how do they figure elevation? Sounds like the answer is in old days they used survey crew which did introduce error along the way but were surprisingly (to me) accurate.

Now there's sattallites that are more precise, but just nailing down a MSL seems to be problematic, especially since the earth isn't even a sphere.

So let me ask this...when I look at a benchmark that says its 3,015 feet above sea level what tolerance do you suppose there is on that? I'd assume less than a foot given it's marked to the foot (I think). Given the non-sphere shape of the earth, compounded survey errors, GPS sattallite position issues and such, are we talking inches, feet, millimeters or what?

I suppose in the end it doesn't really matter. In our line of work if the threshold is really 40' MSL but is marked 37' then so what? All our instruments are calibrated to the marked elevation so in the end we're all on the same page and breakout 200' above the ground regardless what it's listed as or is in reality.
 
Captain said:
I figure today they can use GPS to pinpoint the elevation above mean sea level of any location on earth. But how did they do it years ago? The 'mile high city' was called that way before GPS. Even has a marker on the Capitol steps at the exact spot.

So, how'd they figure it out? Anyone know?
Click to expand...

Really ?
 
Captain said:
The earth is static (more or less) except when it moves.

Now there's sattallites that are more precise, but just nailing down a MSL seems to be problematic, especially since the earth isn't even a sphere.
Click to expand...

Looks like some info here that might help you: http://www.ngs.noaa.gov/heightmod/

One nit...GPS does not calculate your elevation. GPS calculates your distance from the center of the Earth even as the Earth hurtles around the Sun. Then your receiver uses whatever model of the Earth's shape it has to subtract what it thinks MSL is (or would be at your lat/long) to give you an elevation (or altitude, if flying). So even with GPS, the calculated elevation is still only as good as the estimate of MSL, which as you say could be problematic.

You may want to read the book The Measure of All Things about the 7 yr surveying expedition that became the foundation of the meter and the supposed error they made.
 
Captain said:
I figure today they can use GPS to pinpoint the elevation above mean sea level of any location on earth. But how did they do it years ago? The 'mile high city' was called that way before GPS. Even has a marker on the Capitol steps at the exact spot.

So, how'd they figure it out? Anyone know?
Click to expand...


We always called it "differential leveling" using mean sea level as zero.

Dwayne
 
When I got my first Garmin GPS, and I don't even remember what year that was, I wanted it to help me find holes and bars on a lake in Minnesota where I would go fishing. It was accurate to something like 30', and that was on a good day. Fast forward to the present, I sometimes go geocaching with my much newer Garmin GPS, and I often come within a couple of feet of the exact coordinates of what I'm looking for. To me, it is just magic. That is all I need to know. :yes:
 
And if you really want to give yourself a headache, consider the implications of the melting ice caps and the rise in MSL......
:stirpot:
 
Captain said:
The earth is static (more or less) except when it moves.

I'm not being obtuse here, I'm saying that obviously if there were an earthquake in Denver rising the ground 15 feet then the marker would need to be moved when they rebuild the Capitol.

But I'm discounting that. Just taking the earth as is in a snapshot how do they figure elevation?

So let me ask this...when I look at a benchmark that says its 3,015 feet above sea level what tolerance do you suppose there is on that?
Click to expand...

Leaving aside the fact that the earth is always in motion (plate tectonics, earthquakes, etc.) and the need for updated datums due to climate changes and other stuff (http://en.wikipedia.org/wiki/North_American_Vertical_Datum_of_1988), very simplistically, a good survey will be generally be within +/-0.02ft vertically and horizontally when it is established. There are more precise surveys of course, but even older survey markers will generally still be within 0.3ft or so of the originating datum unless somebody messes with the benchmark.
 
warthog1984 said:
Leaving aside the fact that the earth is always in motion (plate tectonics, earthquakes, etc.) and the need for updated datums due to climate changes and other stuff (http://en.wikipedia.org/wiki/North_American_Vertical_Datum_of_1988), very simplistically, a good survey will be generally be within +/-0.02ft vertically and horizontally when it is established. There are more precise surveys of course, but even older survey markers will generally still be within 0.3ft or so of the originating datum unless somebody messes with the benchmark.
Click to expand...

Watch out for the occasional blunder though...I once found a half section line 1/4 mile from it's intended location...I found it *after* I'd already caused some paperwork to be filed - got the lucky task of correcting the paperwork
 
warthog1984 said:
Leaving aside the fact that the earth is always in motion (plate tectonics, earthquakes, etc.) and the need for updated datums due to climate changes and other stuff (http://en.wikipedia.org/wiki/North_American_Vertical_Datum_of_1988), very simplistically, a good survey will be generally be within +/-0.02ft vertically and horizontally when it is established. There are more precise surveys of course, but even older survey markers will generally still be within 0.3ft or so of the originating datum unless somebody messes with the benchmark.
Click to expand...

Would the elevation measurement degrade with time as the survey crew works away from a known benchmark?

I don't know why but in my head I see the verticle azimuth not being as precise as the horizontal. I suppose because;

A. Typically you're only dealing with a few degrees of vertical change per measurement whereas on the horizontal they have the whole half circle of the scale to work with. That just seems more precise,

B. The vertical plane (ground) is pretty wavy. Lots of hills and such seems like a constant source to introduce more error,

C. The earth ain't flat. So you have to account for the earth falling away as you march West from Boston or wherever they started. Did they even know the exact size of the earth when the survey crews were benchmarking out to Denver? I suppose they must have.

So, with all these chances to introduce error coupled with plain old human error it's shocking to me that they were able to determine elevation with any accuracy 'back in the day'.
 
Last edited:
SixPapaCharlie said:
But in the sky it isn't using the height of the terrain in my XY location.
Or did I misunderstand you?
Click to expand...

It is using your height above the datum plane, typically WGS-84, but there are others depending on the mapping system.
 
warthog1984 said:
Leaving aside the fact that the earth is always in motion (plate tectonics, earthquakes, etc.) and the need for updated datums due to climate changes and other stuff (http://en.wikipedia.org/wiki/North_American_Vertical_Datum_of_1988), very simplistically, a good survey will be generally be within +/-0.02ft vertically and horizontally when it is established. There are more precise surveys of course, but even older survey markers will generally still be within 0.3ft or so of the originating datum unless somebody messes with the benchmark.
Click to expand...

Hmm, what about when the survey standard is a 'Cigarette'? From what I was told, it was benchmarked as the time it would take a man on horseback to roll and smoke a cigarette. I looked at some Godforsaken property in west TX that still had that survey.
 
thehulk06 said:
Of course people will downplay its accuracy. How the hell else can they explain it? Its pretty crazy though.Its a map that is depicted as if its seen from space..... anyways, theres lots of stuff on it as its pretty well known. But here's something else for you to read.http://theunexplainedmysteries.com/piri.htmlIm just saying, an old map with accuracy beyond its time..... gets you thinking
Click to expand...

I'm not seeing where that chart is showing 'coastline under the ice'.:dunno: Many of what we know as 'first discoveries' are of course no such thing. Anyone that sailed on a beam reach off of Africa was going to find Antarctica, if they turned down wind they would be swept up the archipelago, again a beam reach will take them to South America and up the coast.
 
Captain said:
I know what the 'M' is and I know it's computed. What I'm asking is if there is a marker somewhere that is master reference for the whole shebang?

I've seen markers nailed down in the earth for lat/long and I can wrap my mind around that. But up down seems pretty hard to mark with any great precision as the reported THREE markers on the steps of the Capitol in Denver point out.
Click to expand...

Yes, several primary physical sea level datum points on islands around the world that are manned during times of oceanographic survey to provide the datum for the reference offset.
 
Captain,
I'm with ya buddy. I'm thinking about how (if GPS is the *new* reference standard) we consider THAT accurate.

GPS Satellites were put in space by man and they are supposed to be in a certain position in orbit. They are not in geosynchronous orbit. However, they can (I guess) be positionally calibrated using Celestial Navigation. Except the stars are moving too.

I think in the end it boils down to what my father told me when I constantly asked him "Why?"...."Because I said so." A meter is a meter because at NIST someone devised the standard and everything else lines up with that. Same for our unit of time, seconds....

http://en.wikipedia.org/wiki/Caesium_standard said:
Caesium clocks are the most accurate commercially produced time and frequency standards, and serve as the primary standard for the definition of the second in SI (the metric system). By definition, radiation produced by the transition between the two hyperfine ground states of caesium (in the absence of external influences such as the Earth's magnetic field) has a frequency of exactly 9,192,631,770 Hz. That value was chosen so that the caesium second equalled, to the limit of human measuring ability in 1960 when it was adopted, the existing standard ephemeris second based on the Earth's orbit around the Sun.[2] Because no other measurement involving time had been as precise, the effect of the change was less than the experimental uncertainty of all existing measurements.
Click to expand...
 
Captain said:
If you lose something but know exactly where it is did you really lose it?
Click to expand...


Layman's terminology set in a reference most folks would understand because they've had an inaccurate clock that ran slow in their possession before.

Really the loss (or gain) isn't reality unless you're measuring it against something else.

Most folk don't realize there's a slight error correction in GPS necessary because a clock in orbit will run slightly slower than a clock on the surface of the planet, due to higher gravitational pull and good ol' General Relativity...

http://www.physics.org/article-questions.asp?id=55

One of many corrections necessary to increase the accuracy of the math such that the made up spheroid world of GPS actually puts you in a specific geographic location, consistently.

GPS as a system is one of the most interesting examples of how to build a system that assumes from the start that it's wrong, and allows for built in corrections... Something very few engineers do, nor do well.
 
denverpilot said:
Layman's terminology set in a reference most folks would understand because they've had an inaccurate clock that ran slow in their possession before.

Really the loss (or gain) isn't reality unless you're measuring it against something else.

Most folk don't realize there's a slight error correction in GPS necessary because a clock in orbit will run slightly slower than a clock on the surface of the planet, due to higher gravitational pull and good ol' General Relativity...

http://physicsworld.com/cws/article/news/2010/feb/17/gravitys-effect-on-time-confirmed

One of many corrections necessary to increase the accuracy of the math such that the made up spheroid world of GPS actually puts you in a specific geographic location, consistently.

GPS as a system is one of the most interesting examples of how to build a system that assumes from the start that it's wrong, and allows for built in corrections... Something very few engineers do, nor do well.
Click to expand...

I was kidding.

By the way, I wasn't impressed with your link. I replaced it with what I think is a better one. Let me know what you think.
 
About you not being impressed? Par for the course.

Oh you meant the new link...
 
You must log in or register to reply here.
Back
Top