[Taxacom] Estimation in GPS positioning

[Dusty]

I apparently failed completely at communicating what I was trying to
communicate. I even drew a picture....

To steal your population analogy, the population of Alaska is knowable. Say
it's 698473. You are rounding off to 700000. That's OK, if a little strange.
Then you are explicitly adding error that does not encompass the true value
- say 700000 +/- 1000. That's very much not OK. You can round all you want,
but you must increase your error estimates accordingly.

>From another message...
*I'm glad you're Google Maps images are so nicely georegistered. By GPS'ing
landmarks, I find that Google satellite images in my field area in NW
Tasmania are as much as 20 m out of register with the ground. I'm also glad
that both you and Dusty do your fieldwork around mailboxes and apartments in
town. Mine's down in more challenging conditions of topography, access to
satellites. I sometimes have to wait 5 minutes to get a GPS reading under
forest cover, and with a large accuracy declaration. Sometimes the unit says
'Weak signal. Are you indoors?'*

Yes, most of my field work happens around mailboxes. Except the part about
flying around Alaska, where my GPS continues to fail to kill me despite the
mountains, trees, clouds, and it being a known instrument of pure evil. In
fact, GPS continually fails to kill several hundred other pilots who land on
it several hundred times per day. It's almost like it's a mature technology
or something. And I'm sure Google, what with it's limited budget and
propensity for closed-mindedness, has indeed mucked up mapping the entire
southern hemisphere, and the mistake can not possibly be in some later
misunderstanding of how technology meets reality...

-D

On Sat, Dec 4, 2010 at 1:50 PM, Bob Mesibov <mesibov at southcom.com.au> wrote:

> Hi, Dusty.
>
> Here's a simple example: the estimated population of Alaska is 698473 +/-
> 1000 people. What I've just said is that the true population of Alaska is
> estimated to be between 697473 and 699473. I have *not* said that the
> population is 698473.
>
> The statement looks a little funny because the population is given to the
> nearest person, but there is a 1000-person uncertainty in the estimate. The
> standard practice in dealing with significant figures is for the estimate to
> agree with the uncertainty, so I'll round off: 698000 +/- 1000. I have *not*
> just said that the population is 698000.
>
> There is no difference between these two estimates. I have not moved the
> population of Alaska from 698473 to 698000. I have not created a new
> population figure.
>
> When you write a lat/long +/- an uncertainty, you are not specifying a
> point, you are specifying a data space within which the true position
> exists. Sometimes this space is represented by a circle; I sometimes do this
> in GIS. The estimate is not the circle, it is the set of possible positions
> within that circle.
>
> You cannot, as you seem to have done in your last post, treat possible
> positions within the estimate as real, and compare them. It is just as
> likely that a rounded-off position will be closer to the true position than
> the non-rounded-off position, as for the rounded-off position to be further
> away. We have no idea what is happening within the estimate.
>
> It's different with statistical estimates. 698473 +/- 1000 might be a
> sample mean +/- average deviation, in which case we know something about the
> likely distribution of population samples. We *do not know this* for single
> GPS readings, which are not means.
>
> I hope this clarifies for you the use of uncertainty in GPS positioning. If
> you want to argue that an ordinary handheld GPS unit is more accurate and
> has less uncertainty than the manufacturer and all the GPS user guides are
> claiming, then I suggest you test that idea in a range of situations and
> publish your results, so that the GPS manuals and GPS user guides can be
> revised.
> --
> Dr Robert Mesibov
> Honorary Research Associate
> Queen Victoria Museum and Art Gallery, and
> School of Zoology, University of Tasmania
> Home contact: PO Box 101, Penguin, Tasmania, Australia 7316
> Ph: (03) 64371195; 61 3 64371195
> Webpage: http://www.qvmag.tas.gov.au/?articleID=570
>