On Wed, 13 Feb 2008, Old Central wrote:

IIRC the use of GPS to determine heights is a complex topic. You need to
determine the spheroid and geoid separation in relation to the grid used
and so on. Remember that many countires use by the different versions of
these for their mapping and with different origins.

If you want to know the height above local sea level, then yes, you need a
map of the geoid. But nobody uses that. In the UK, we use height above the
OSGB36 datum, which can be computed from the WGS84-based GPS height fairly
easily (not trivially, but a computer can do it without breaking a sweat).
Ditto for any other reference frame.

Additionally as identified above there are satellite fix issues.

I think this is the killer. There's just so much inaccuracy in a typical
height measurement that it's not very useful.

I suspect then, as a relative method it would be reliable but more
difficult to be reliable as an absolute method.

Interesting point.

tom

--
Me ant a frend try'd to WALK the hole unterrgrand but was putting off -
sometime we saw a trane! -- Viddler Sellboe

On Wed, 13 Feb 2008, Tom Anderson wrote:

On Wed, 13 Feb 2008, Old Central wrote:

IIRC the use of GPS to determine heights is a complex topic. You need to
determine the spheroid and geoid separation in relation to the grid used
and so on. Remember that many countires use by the different versions of
these for their mapping and with different origins.

If you want to know the height above local sea level, then yes, you need a
map of the geoid. But nobody uses that. In the UK, we use height above the
OSGB36 datum,

Hang on, no, that's rubbish. We do use the local sea level, aka Ordnance
Datum Newlyn.

So yes, you're right.

Clearly, the solution is just to switch to using geometric rather than
gravitational heights.

The use of an irregular height datum kind of freaks me out. It's fine for
going up and down without moving across the planet, but it means that you
can't relate a height in one place to a height in another, in terms of
position in space, without knowing the shape of the datum. It means our
coordinate system isn't really a coordinate system.

But if you used geometric coordinates, then you'd find that sometimes,
walking along a contour was walking up or down the gravity well. And the
maths for working out distance is still hard, because it's all on the
surface of a spheroid!

Cartography is hard.

tom

--
Me ant a frend try'd to WALK the hole unterrgrand but was putting off -
sometime we saw a trane! -- Viddler Sellboe

Tom Anderson wrote:
On Wed, 13 Feb 2008, Tom Anderson wrote:

On Wed, 13 Feb 2008, Old Central wrote:

IIRC the use of GPS to determine heights is a complex topic. You need
to determine the spheroid and geoid separation in relation to the
grid used and so on. Remember that many countires use by the
different versions of these for their mapping and with different
origins.

If you want to know the height above local sea level, then yes, you
need a map of the geoid. But nobody uses that. In the UK, we use
height above the OSGB36 datum,

Hang on, no, that's rubbish. We do use the local sea level, aka Ordnance
Datum Newlyn.

Well whenever I am using GPS these days [1], I can find my altitude by
reference to my watch and a copy of Reed's Almanac. And that leads me
to the question, what sea level are you taking? Certainly most charts
I've found (Admiralty and Imray) use LAT [2] as their datum for points
below MHWS [3], and MHWS for heights on dry land.

[1] and, no, I don't rely on it, I always have several alterntive
methods of navigating at the same time, just in case.
[2] lowest astronomical tide
[3] mean high water springs

Robin

R.C. Payne wrote:

Well whenever I am using GPS these days [1], I can find my altitude by
reference to my watch and a copy of Reed's Almanac.

Remind me again...how does that work?

John Rowland ("John Rowland" )
gurgled happily, sounding much like they were saying:

Well whenever I am using GPS these days [1], I can find my altitude by
reference to my watch and a copy of Reed's Almanac.

Remind me again...how does that work?

Dead easy... Reed's Almanac is a tide table...

I'm not quite sure how it'll tell you whether the transporter that your
boat is on the back of is on the M4 or the A4 underneath it, though...
Oh, wait. If there's a horrible scraping cracking noise, it's the remains
of the top of the mast against the underside of the M4 flyover.

Tom Anderson wrote "Cartography is hard."

Tom

Sorry, whilst catography is a related topic, this problem normally
comes under geomatics these days (old fashioned land surveying).

OC

On Thu, 14 Feb 2008, R.C. Payne wrote:

Tom Anderson wrote:
On Wed, 13 Feb 2008, Tom Anderson wrote:

On Wed, 13 Feb 2008, Old Central wrote:

IIRC the use of GPS to determine heights is a complex topic. You need to
determine the spheroid and geoid separation in relation to the grid used
and so on. Remember that many countires use by the different versions of
these for their mapping and with different origins.

If you want to know the height above local sea level, then yes, you need a
map of the geoid. But nobody uses that. In the UK, we use height above the
OSGB36 datum,

Hang on, no, that's rubbish. We do use the local sea level, aka Ordnance
Datum Newlyn.

Well whenever I am using GPS these days [1], I can find my altitude by
reference to my watch and a copy of Reed's Almanac. And that leads me
to the question, what sea level are you taking? Certainly most charts
I've found (Admiralty and Imray) use LAT [2] as their datum for points
below MHWS [3], and MHWS for heights on dry land.

Really? I know about LAT, but i'm surprised to hear that land heights are
measured from MHWS. OS maps use the Newlyn datum, which is the mean sea
level at Newlyn back in 1915 or something; that's carried through the
country by levelling, so the datum is an gravitational isopotential
surface. MHWS is not only a high, not mean, tide, but is something that's
affected by local seabed topography, and so is not an isopotential
surface. That means it won't be parallel to the Newlyn datum, so not only
will Admiralty heights be different to OS heights, but the difference will
vary across the country!

Horses for courses, though. Nautical charts use LAT as a datum because
depths are there so you can work out if you're going to run aground and
that lets them have tide values which are always positive. Plus, it means
that when you see a blue bit on a chart, you know it's always underwater.
You couldn't use LAT for land heights, because it's not defined on land. I
suppose they use MHWS on land because it has a similar property - anything
with a positive height is always above water.

Hang on, how do they determine MHWS on land? Are you sure they don't use
ODN?

It irks me that the Newlyn datum is a mean sea level, and not LAT. But
then i suppose it's natural to define an isopotential surface that way,
because it's the sea level you'd have if you got rid of the moon. Except
it's not, because of topographic effects. I think.

In conclusion, geomatics is hard.

Anyway, my proposal is for *all* heights to be measured as distance from
the centre of mass of the earth. SOLVED!

tom

--
Work alone does not suffice: the efforts must be intelligent. -- Charles
B. Rogers

Tom Anderson wrote:

Anyway, my proposal is for *all* heights to be measured as distance
from the centre of mass of the earth. SOLVED!

Why not measure heights from the centre of mass of the earth-moon system?
That would abolish the need for tide tables because the tide would be at a
fixed height... although the land would go up and down.

;-)

On Fri, 15 Feb 2008, John Rowland wrote:

Tom Anderson wrote:

Anyway, my proposal is for *all* heights to be measured as distance
from the centre of mass of the earth. SOLVED!

Why not measure heights from the centre of mass of the earth-moon system?
That would abolish the need for tide tables because the tide would be at a
fixed height...

SPLENDID IDEA.

although the land would go up and down.

A minor detail.

A slightly bigger 'although' is that it wouldn't work - sea level isn't
just a constant distance from the earth-moon barycentre. If it was,
there'd only be one tide a day, wouldn't there?

tom

--
natural disasters, unexplained phenomena, chaos, chance, tattooing,

In article ,
John Rowland wrote:
Tom Anderson wrote:

Anyway, my proposal is for *all* heights to be measured as distance
from the centre of mass of the earth. SOLVED!

Why not measure heights from the centre of mass of the earth-moon system?
That would abolish the need for tide tables because the tide would be at a
fixed height... although the land would go up and down.

Notice the way in which the sea stays steady as a rock whilst the
buildings keep washing up and down.

Nick
--
Serendipity: http://www.leverton.org/blosxom (last update 4th February 2008)
"The Internet, a sort of ersatz counterfeit of real life"
-- Janet Street-Porter, BBC2, 19th March 1996