OT: GPS altitude question

[N_Cook]

Anyone know of a site like heavensabove for astronomical predictions
including Iridium flares.
So you input a latitude and longitude and the application returns a date
and time when 3 or more GPS satellites are within (selectable) say 30,
degrees of the zenith. Wish to determine the height of a patch of ground
. It is away from roads, so daftlogic.com etc and old OS maps (before
they got precious with their data) and proper spot heights, is no use.

[Pat]

Google Earth?

[mrob...@att.net]

N_Cook <div...@tcp.co.uk> wrote:
> Anyone know of a site like heavensabove for astronomical predictions
> including Iridium flares.

> So you input a latitude and longitude and the application returns a
> date and time when 3 or more GPS satellites are within (selectable)
> say 30, degrees of the zenith.

This doesn't let you select the degrees-from-zenith, but you can get a
good idea by looking at the 2D or 3D map.

http://in-the-sky.org/satmap.php

> Wish to determine the height of a patch of ground.

Pretty much anywhere in the UK should have decent GPS coverage - take a
GPS receiver to the spot and let it sit there for a while. GPS altitude
is not as accurate as latitude and longitude, but the longer you can
keep the receiver still, the better it will get.

If you have really a lot of time, let it sit there for many hours
logging data, and plot the results... there will be a definite
distribution and you can pick the number closest to the center. If you
dig into the settings enough, some receivers have options for stationary
vs pedestrian vs vehicle - this tweaks the averaging/filtering that they
do on the raw data, and may help you get a reliable answer sooner.

> It is away from roads, so daftlogic.com etc and old OS maps (before
> they got precious with their data) and proper spot heights, is no use.

Google Maps has a "terrain" feature. You can only see the contour lines
in a specific zoom range - at roughly 1.5" to 1000' scale, I get 100 m
major contours and 20 m minor contours. Zooming out one mouse-wheel to
roughly 1.5" to 2000' scale, I get 200 m major and 40 m minor. Two more
mouse-wheels out and no contours at all.

Oh, this one is better. http://openstreetmap.org and click the "layers"
icon (a stack of paper) at the right and select "cycle map". 50 m major
contours and 10 m minor ones if you zoom in enough, at least in England
and Scotland.

Matt Roberds

[N_Cook]

I'll try again later today. ISTR that whan it first came out, I tried it
on a piece of ground with known survey-level and it came back rubbish.
Like anything based on google maps, (just linear interpolation between
road junctions so doesn't catch humps and dips )

[N_Cook]

I've tried that, in effect. Known survey-level piece of ground and for
ten minute sessions on 3 days. However I averaged the results, the
answer was no better than .5m or so of the known surveyed level.
I wanted to repeat this on some day and time with known good zenithal
satellites, get a good agrreing result, and then move to the unknown
patch of ground on another day, but again zenithal for that spot, it is
not nearby to me

[Leif Neland]

N_Cook forklarede den 19-11-2014:

If you really need this data to less than 0.5m, then try asking a
surveyor to come by with a proffessional gps, which have
millimeter-accuracy.

Leif

--
Husk kørelys bagpå, hvis din bilfabrikant har taget den idiotiske
beslutning at undlade det.

[josephkk]

On Tue, 18 Nov 2014 10:40:35 +0000, N_Cook <div...@tcp.co.uk> wrote:

Just take your personal GPS receiver out there and read it. Now that the
whole constellation of 24 satellites is up, with many in orbit spares also
up there are very few places that do not see 4 to 8 satellites at any
time. Also at this point the number of satellites visible can be solved
analytically of with any number of computer algebra and computer geometry
programs or even an spreadsheet.

?-)

[josephkk]

On Wed, 19 Nov 2014 19:10:54 +0100, Leif Neland <le...@neland.dk> wrote:

>
>> I've tried that, in effect. Known survey-level piece of ground and for ten
>> minute sessions on 3 days. However I averaged the results, the answer was no
>> better than .5m or so of the known surveyed level.
>> I wanted to repeat this on some day and time with known good zenithal
>> satellites, get a good agrreing result, and then move to the unknown patch of
>> ground on another day, but again zenithal for that spot, it is not nearby to
>> me
>
>If you really need this data to less than 0.5m, then try asking a
>surveyor to come by with a proffessional gps, which have
>millimeter-accuracy.
>
>Leif

No. It does not mm grade accuracy, never did. I have discussed this with
my professional surveyor friends. It is good to about 1 to 2 cm in
horizontal and 5 cm in vertical with survey grade equipment. Moreover, to
get that you must use known equipment in the differential mode.
Handheld/automobile units are within 10 m moving, but with an averaging
centroid of maybe 2 m. Pretty much the same for all non-survey grade
equipment. It is just the physics of the system.

?-)

[N_Cook]

So not much different to what anyone can do with a calibrated sighting
telescope on a camera tripod and a 20m long polythene pipe manometer
tube and homemade staffs , if there is a convenient circa 1910 OS map
with decimal-foot spot height nearby or ,very rare these days, a still
extant benchmark.
Its intenely annoying that later series publically available OS maps
have no bench mark heights at all, less spot heights, and when they are
there, much less quoted accuracy resolution.

I'm only assuming that if you have 3 GPS satellites near the zenith then
the altitude would be a more accurate determination. But as we would not
know the bias of useage in the algorithm then only 1 additional
satellite near the horizon may well cock it all up anyway

[Leif Neland]

josephkk udtrykte præcist:

> On Wed, 19 Nov 2014 19:10:54 +0100, Leif Neland <le...@neland.dk> wrote:

>> If you really need this data to less than 0.5m, then try asking a

>> surveyor to come by with a professional gps, which have

>> millimeter-accuracy.
>>
>> Leif
>
> No. It does not mm grade accuracy, never did.

Well, perhaps I wasn't accurate enough :-)

http://water.usgs.gov/osw/gps/:
Survey-grade GNSS receivers in static mode has accuracy of
Horizontal: 5mm + 1 ppm Vertical: 10 mm + 1 ppm

1ppm is relative to the known reference, each km gives 1mm inaccuracy.

[Ian Malcolm]

N_Cook <div...@tcp.co.uk> wrote in news:m4hla9$ruc$1...@dont-email.me:

<http://en.wikipedia.org/wiki/Earth_tide>
Tidal forces cause semidiurnal elevation changes of the Earth's crust
with an ampltude up to 55cm at the equator. Unless you subtract the
current GPS elevation of a nearby benchmark, averaging over an arbitary
period will be fairly meaningless. --
Ian Malcolm. London, ENGLAND. (NEWSGROUP REPLY PREFERRED)
ianm[at]the[dash]malcolms[dot]freeserve[dot]co[dot]uk
[at]=@, [dash]=- & [dot]=. *Warning* HTML & >32K emails --> NUL