GPS Height Above elipsoid

[Saf...@gmail.com]

Dear guys,

I am plotting a line in GE using my gps data. I'm having problems with
altitude. The documentation says you can do absolute(realative to sea
level), glamptoground or relative to ground. But how can I achieve the
height above the elipsoid?
Thanks

[ManoM]

Hi,

Do you mean how do you add an altitude? Altitude is the third number
in a coordinate, longitude,latitude,altitude:

110, 45, 1000

For more information: http://code.google.com/apis/kml/documentation/kml_tags_beta1.html#coordinates

ManoM

[Jonathan van Tuijl]

ManoM wrote:
> Do you mean how do you add an altitude?

I think he means sea level is not the same as the ellipsoid. See
http://en.wikipedia.org/wiki/Geoid

Jonathan

[ManoM]

We represent the earth as a sphere (special case of an ellipsoid). The
surface of our sphere corresponds to 0 meters sea level. As far as the
KML coordinate system, we consider 0 meters altitude to be sea level,
and we draw KML and terrain in a way that's consistent with that.
Specifically, the EGM96 geoid is our sea level, a potato-like shape
that's smoothly varying but not perfectly smooth, and represents mean
sea level around the globe. The geoid (and therefore sea level) is
offset from the ideal WGS84 reference ellipsoid by as much as 200
meters or so in some places.

ManoM

[jpwade_bsu]

more info @ http://www.czmartin.com/home/i24/utm/earth_msl.htm

jpwade
www.czmartin.com/jpw

> > Jonathan- Hide quoted text -
>
> - Show quoted text -

[jpwade_bsu]

index page for the previous post.

http://www.czmartin.com/home/i24/utm/directory.html

On Sep 7, 3:23 am, jpwade_bsu wrote:
> more info @http://www.czmartin.com/home/i24/utm/earth_msl.htm
>
> jpwadewww.czmartin.com/jpw

> > - Show quoted text -- Hide quoted text -

[Saf...@gmail.com]

I have plotted a line representing a short boat in amsterdam using
sea level(m) for altitude. I obtained my data using an accurate gps
integrated sensor. When i plotted the line in GE (using absolute) the
results are not accurate. I know amsterdam is quite low but the line
was plotted ridulously high.

so my qn is, if i supply the altitude in m's above sea level,

what type of accuracy can I expect in GE ?

from my experiments the altitude is not represented realistically in
GE.

thanks

Saf

[barryhunter [KML Guru]]

As mentioned in the quote from ManoM:

> The geoid (and therefore sea level) [the value in Google Earth] is
> offset from the ideal WGS84 reference ellipsoid [the value from your GPS] by as much as 200

> meters or so in some places.

(I've added comments in [] to clarify)

Does that help?

It's also a relative misnomer that alitudes from GPS devices are
accurate, in practice the altitude isn't very accurate most of the
time. (if you get a lot of satalites visible it can be quite good
though)

[Saf...@gmail.com]

Hi. Is there any way to get a list of these offsets for all the
locations? so I can project more accurately in GE?

[ManoM]

Hi,

No, that list does not exist. Its not that we took Earth, looked at
the ideal geo and said "let's offset it here, here, and here."

ManoM

[Saf...@gmail.com]

Then the MSL values that you add into the kml file are just relevant
to GE? another words points with altitudes obtained from a gps device
will never be accurately portrayed in GE.

[barryhunter [KML Guru]]

You can but you will need to find the relevant transformation
formulas, probably as part of a datum transformation, as you will need
to do a 3D transformation. The point being its not a simple table
reference - although you could create such a table give the forumular,
but you would have to make a arbitrary decision on its precision.

Have a look at the proj4 libary - that can porbably do it, although
I've never tried.

(the other point being that frequently altitudes arent that accurate,
so a real tranformation is can be a waste of time)

[Nif]

Sorry to jump in here, but I was trying to find out how lat/long is
represented in GE and barryhunter pointed me to this thread...

ManoM -

So are you saying that the lat/long coordinates in GE earth are
represented by a sphere - x, y, z axis from the center all equal - not
a reference ellipsoid?

But that the altitude is referenced from the EGM96 geoid seems to be
used with respect to the WGS84 reference ellipsoid?

That doesn't make sense - to have lat/long (x,y on the surface)
coordinates derived from a sphere but lat/long (x,y on the surface)
used to determine position for geoid readings pulled from an
ellipsoid.

Can you give any more clarification?

Saf -

This link from NGA looks like it will allow you to enter WGS84
referenced coordinates (lat/long) and get EGM96 offsets from the WGS84
ref ellipsoid.

http://earth-info.nga.mil/GandG/wgs84/gravitymod/egm96/intpt.html

You're on your own to figure out WGS84 Ellipsoid height, Orthometric
height and what they mean. I don't know which one of these represents
the elevation/altitude values obtained from a GPS (if either).

The learning never stops...

Nif

On Sep 28, 1:11 pm, ManoM wrote:
> Hi,
>
> No, that list does not exist. Its not that we took Earth, looked at
> the ideal geo and said "let's offset it here, here, and here."
>
> ManoM
>

and