Subject: nyonqc datum shift try by computation & other thoughts
Date: Jan 05, 2006 @ 00:12
Author: aletheia kallos (aletheia kallos <aletheiak@...>)
Prev    Post in Topic    Next
Prev    Post in Time    Next


the promised help hugh


by chance
in the immediate nyonqc area
a shift in datum from nad27 to nad83 or vice versa
produces a surface displacement of just under 30
meters
or as i understand it
just about exactly half the breadth of the normal tho
not necessarily guaranteed target window produced by a
typical commercial grade gps receiver

& please correct me if i am off

but what these numbers seem to indicate is that by
shifting between the 2 datums while trying on location
there to reach the probably irreducible 3 meter target
window for this unmarked tripoint as it is legally
defined
a trypointer would shift from the center of one window
to the periphery of the other or vice versa

& there is a 33 percent overlap of coverage apparently

which means that there is at least some probability in
this case that the datum shift would not turn a hit
into a miss or a miss into a hit
but also a more substantial probability that the shift
would indeed cause just that difference

however
both of these probabilities pale equally against what
is at best only the small fraction of a percent chance
one would have of correctly placing let alone
correctly centering oneself & the gps receiver in the
3 meter nyonqc window within the 60 meter window of
coverage produced by using either datum
or indeed any other datum

that is
one seems to be stuck in class e or d or c at the very
best & most hopeful


the overall impression of the try then is like
shooting at a fly on a large distant wall with both
barrels of a shotgun

the datum
or the barrel if you will
wouldnt seem to make much if any difference

what matters more is whether the fly just happens to
be in the lens of overlapping coverage

but even that doesnt matter much because the
commercial gps accuracy level is so diffused in
relation to the fly

& the most significant fact of all
i suppose
is that there is precious little chance of any shot
from either barrel hitting the fly


if the datum shift reaches or exceeds 60 meters
then the circles of coverage cease to overlap or touch
at all

at 200 meters it might as well be 2 different shotguns
for any cumulative or synergistic effect of the
proximity of the coverages has long since ceased

but there are still 2 scattershot discharges & 1 fly

& guess what

still the most significant fact appears to be
very little chance the fly gets hit in any case


so i think trying for unmarked points may not be the
way to nail this thesis
& i respectfully withdraw the suggestion


better to seek marked but especially humbly marked
points in dense woods i would think now
say like those of the berkshire county persuasion
where reaching the correct 60 meter circle as opposed
to the incorrect one could make a difference between a
hit or a miss
simply because the humble marker might never even be
glimpsed from the wrong window

& yes perferably load the deck as much as possible
with the largest datum shift you can find too

i wouldnt blame but only encourage you in that

for you will still have to deal with the fact that the
trypointer in your experiment can be expected to bring
his brains & eyes & maps & all his other favorite
navigational aids & tricks
including even compensating for the known difference
in your datums
& not doing anything counterintuitive just because the
satellites say so
etc etc
for he is still going to be trying 100 percent in
earnest
& all the more so now that he is accepting to handicap
himself in this bizarre but fun way
so i think this demonstration may still not be an easy
thing to give even under the best of circumstances


but i also think trypointing with a heavy reliance on
gps tech isnt always necessarily the best or brightest
or most fun way to go in any case
so if anything is ever missed or lost in this
translation from one datum to another
it may in fact not be much


well anyway i hope i have contributed to your try in
some way





__________________________________________
Yahoo! DSL – Something to write home about.
Just $16.99/mo. or less.
dsl.yahoo.com