Dear list,
I received a lot of great hints for marking a grid in the field using GPS
methods. Thanks to everyone who replied off-list!! A few people asked me to
share the replies, so I thought I'd compile them and post them here for all to
see.
Thanks again, all.
Lauren Quinn
Sent:
Fri 6/18/10 6:54 PM
To:
[email protected]
You can probably create a track by walking the grid once, save it,
and then reload it to use it as a template for where to walk again.
Sent:
Fri 6/18/10 7:00 PM
To:
L Quinn ([email protected])
Hi Lauren,
One
thing to consider is the accuracy of the gps unit. The positional error
associated with a garmin can be a couple meters, especially if you are
moving around. If you are working on a fine scale you may want to stick
with measuring tapes. Good luck on your project.
Sent:
Fri 6/18/10 7:30 PM
To:
[email protected]
Hello,
>From my limited use of GPS units, I do not believe that
you will achieve anywhere near the accuracy you desire, using a handheld
Etrex. There are several factors to consider, and you would probably
find the GPS solutions either too cumbersome or too expensive. I will
leave it to the experts to advise you further on the model you would
need to achieve that accuracy.
You would probably do better with
getting good maps for wherever you go, and then using a compass
(compensate for magnetic versus true north) and your GPS to navigate the
area. If you are concerned about covering the same ground twice by
mistake, you can use a marking system to show where you have been (we
used toilet paper draped on low bushes in Minnesota. It's biodegradable
and will disappear the first time it rains). I would concentrate more on
the goal of the sampling system (achieve separation between
samples/cover the entire area) rather than focusing on whether I was in
exactly the right grid spot (unless that is the essential goal, hard to
say without knowing more details).
One final tip. If you know the
spacing, instead of using a measuring tape it may be quicker to
pre-measure a length of rope (one that doesn't stretch or shrink when it
gets wet) and use that in the field. Works better when you have a
person on either end of the rope.
Also, consider an infra-red range
finder.
Test your GPS in an open sports field. Walk the marked
lines and see how accurate the resulting GPS track is. That will give
you a rough idea of the accuracy you can expect.
Good luck!
Sent:
Fri 6/18/10 7:35 PM
To:
[email protected]
Cc:
[email protected]
Lauren,
It is possible to do what you are asking, but it is not easy. It
would be easier to arrive at your starting location, project one
waypoint in the direction you want to go (along a desired grid line), do
a "go to" to the projected point, and walk offsets of 0m, 20, 40, etc.
from the "go to" line. I suggest you read your garmin manual about these
topics, as I could not explain all of this from scratch to a
5-year-old.
Sent:
Fri 6/18/10 7:48 PM
To:
L Quinn ([email protected])
Dear Lauren,
I used to use a similar method for generating transects for surveying
birds with an eTrex Legend. I used Garmin's MapSource software (which
is simple and inexpensive) to program routes and then would sync them
with the GPS. You could create a grid simply using a series of
intersecting, straight-line routes. I would also use a feature in the
GPS (I think the eTrex can do this but I can't remember for sure) called
"project waypoint." After marking a waypoint, you can use the editing
tool for the waypoint and move the point by designating a starting
point, a distance, and a bearing. This was an easy way of creating
transects on the fly. Projecting a waypoint can be done anywhere as
long as you have a reference point for your field site. I'm not sure if
that is clear or helpful or quite what you need. If you would like
further details or help, please let me know.
Sent:
Fri 6/18/10 8:10 PM
To:
L Quinn ([email protected])
Hi Lauren,
You can load lines onto your Garmin
unit as tracks -- the best way to do it is using a free program called
DNR Garmin created by Minnesota DNR (can download online). If you
can use ArcGIS to generate a fishnet grid (XTools does this easily)
shapefile
for your study area, you can bring this into DNRGarmin, then save it as
a track file with a .gpx extension. This can be uploaded to your
Garmin using the Mapsource program that comes with the Garmin unit. The
track will show up on your display as a thin silver line. It loads
into your active track log, however, which means that you might run into
the problem if you use your Garmin a lot that the track will be
overwritten
with the track you create while walking around (you can avoid this by
turning
the active track log off before you start).
If you're not familiar with ArcGIS
or
XTools, you could ask a GIS person to generate the grid shapefile for
you,
then convert it in DNR Garmin yourself. I'm happy to help explain
the steps if you want to try it.
Sent:
Fri 6/18/10 8:22 PM
To:
L Quinn ([email protected])
Hi Lauren,
I did exactly this using mapmaker (www.mapmaker.com). You can generate
cells or grids ( Utilities/Vector utilities/Generate/Square grids )
which you can overlay on maps. These can then be exported in a format
that Garmin units can use ( I was using different gps units so I
cannot help with this bit I'm afraid). But if you want basic help with
mapmaker, let me know.
Sent:
Fri 6/18/10 8:28 PM
To:
L Quinn ([email protected])
I think you can generate a grid of points in any GIS program and export
the points into a spreadsheet or text file that can be imported into the
garmin. We used use this script (but with an old version Arc View). I
think the new Arc GIS version would have such a system built in. Note
that equidistant points (in a triangular array) are a more efficient
system of packing points into space than a square grid.
www.hear.org/articles/pdfs/equidistantpoints.pdf
Sent:
Fri 6/18/10 8:42 PM
To:
L Quinn ([email protected])
Hi,
If you use GPS Trackmaker software you can create points in the
program and download them into your gps unit. The software is
garmin-compatible. For this approach you would need to go where you're
sampling ahead of time, though, in order to have a starting point to
build your grid around in the program. To do it on site you could bring
a laptop in your vehicle I suppose.
Hope this helps, good luck!
Sent:
Fri 6/18/10 8:57 PM
To:
L Quinn ([email protected])
Hi Lauren,
As youre probably finding out from other responses on
the listserv, the best way to do this is with a GPS pocket PC that runs
ArcPad. However, i work in sub-Saharan Africa where low tech is
necessary. You can generate a grid in a GIS (which is like generating
straight and perpendicular tracks in the GPS), and export that into a
GPS format - you then upload this format into your GPS and basically you
have what youre looking for. It is doable in the way that you describe.
Do you have an upload cable to connect the GPS to the PC? If so,
Ozi-Explorer is best software to help you out.
Let me know if i
can be of more assistance.
Sent:
Fri 6/18/10 9:05 PM
To:
L Quinn ([email protected])
If you set the GPS to the appropriate UTM Zone the coordinates are given
as Northings and Eastings in meters - take your first waypoint and add
(or subtract) 20 to the northing, walk due north til you get to the new
coordinates and waypoint it. When you get to the corner add 20 to the
easting etc etc
The Vista is only sub-3m so I would use the WAAS utility to be as
precise
as possible.
Sent:
Fri 6/18/10 9:13 PM
To:
L Quinn ([email protected])
Hi Lauren,
You should be able to do this even without a GPS, using a basic magnetic
orienteering compass
(http://www.forestry-suppliers.com/product_pages/View_Catalog_Page.asp?mi=1290)
and some pin-flagging and a meter tape (or just know your pace very well (ie
steps per 5m). Use the compass to make sure that you're always walking
perpendicular to your other transects and use the pin-flags to help you see the
intersections of the grid. I've used this method many times to setup smaller
plots where my grid was 5m x 5m spacing of 9 soil sampling locations (no GPS
needed, and consumer grade GPS units rarely have the accuracy to do much good
differentiating points below 20 m). If you're not clear, look up some sites
for instructions on how to use a magnetic compass for taking and walking along
a bearing.
Sent:
Fri 6/18/10 11:11 PM
To:
L Quinn ([email protected])
Hi Lauren,
I suspect that if you want a grid that
small, your gps would likely be inaccurate. It is often out by 3 m or
more and often much more if you are under trees, etc. Our
sampling/measuring requirements are somewhat different, but instead of
measuring tape we use a LEICA laser range finder. These are highly
accurate (to mm I believe) and allow you to shot towards a target, such
as some tall flags or a post. So if you put a post at a starting point,
walk away and take readings until you get to 20 m, plant another post,
you get the idea. The only potential problem is doing this at right
angles to get a grid. For that I use a compass, there may be other
solutions. The LEICA (less than $500) is far easier and more fun to work
with than either GPs or measuring tape,
just a suggestion,cheers,
Sent:
Sat 6/19/10 12:47 AM
To:
L Quinn ([email protected])
Hi Lauren,
You might be able to lay your grid out in MapSource as waypoints and
upload them to the Vista.
Another option would be to generate the grid in Excel, save the
co-ordinates as a .csv file and then convert it to a .gpx file using a
site like this : http://www.gps-data-team.com/convert.php.
A .gpx file is the format Garmin uses to save waypoints. You can then
upload the gpx file to the Vista using MapSource. You'd just need to
make sure the co-ordiates are in decimal degrees. This method might be
the best way to generate a really accurate grid as you can specify the
exact distance between co-ordinates, rather than trying to eyeball the
grid in MapSource
Good Luck!
Sent:
Sat 6/19/10 6:46 AM
To:
L Quinn ([email protected])
Hi Lauren
I wanted the exact same thing for my vegetation study and never found a
satisfactory way to do it using my Garmin. I ended up buying a trimble
Juno and making a grid in gis and importing it into the gps. It worked
beautifully, BUT, the TerraSync software for the trimble is expensive.
(and I had to have some serious GIS help making the grid..., oh and it
was a 1 km grid, you might need a better trimble unit for a 20 m grid
since that is really small resolution) Don't know how helpful this is,
but let me know if you have any questions.
Sent:
Sat 6/19/10 11:45 AM
To:
[email protected]
Hi Lauren,
I
understand you want the GPS to generate a grid pattern wherever you're
standing. Although newer sophisticated models may have such a function, a
grid pattern likely will be something you'll have to create in a GIS
and upload to your GPS before you set out. If you don't know where
you're going to be sampling, you could try generating a random grid of
squares or even points (sampling stations) over your sampling area which
is easy to do in a GIS and shouldn't be hard to upload along with a map
or georeferenced aerial photo to your GPS... I don't know if this is of
much help to you... however, feel free to write me back if you want to
discuss this further.
Sent:
Mon 6/21/10 3:23 PM
To:
L Quinn ([email protected])
Hi Lauren;
I generally agree with the response I saw on the list regarding
using Hawths tools and DNR Garmin to get waypoints loaded into your GPS
unit, I have spent a lot of time using both tools and they are very
good, but I also agree with the comment that efficiently navigating to
waypoints with any accuracy is pretty difficult with any GPS unit.
What I generally do for GPS-based forest sampling is to use the "Go
To" feature to find the distance and azimuth to my first/next sample
point, then use compass and pacing to get close. Depending on the
importance of my accuracy in choosing a plot center, I may do this 2-3
more times to zero in before settling on a plot center. I then let the
GPS sit on the chosen plot center long enough for the precision to reach
a desired level and mark a waypoint for the "actual" plot center. You
can chase your waypoints around a couple of square foot area for hours
if you have poor precision on a consumer GPS unit, but generally getting
close to your grid point is good enough as long as you don't
introduce bias, just choose a method and stick with it (i.e. your method
could be to take 3 headings and choose the point: compass and pacing,
then wait 2 minutes for unit to acquire a given PDOP then navigate using
compass and measuring tape, then hold the GPS at that position for 2
minutes for the unit to acquire precision, then use compass and
measuring tape to determine plot center. Place GPS at plot center and
wait for PDOP to reach 1.x and record waypoint)
If you are unable to go back and forth between your GPS and ArcGIS
to determine sample points, I would recommend just using a little
division with your geographic or UTM grid to select waypoints. Get the
generalized corners of your sample area, determine how many points you
want to put in it, divide x1-x2 by how many vertical lines your grid
has, and y1-y2 by how many horizontal lines your grid has, then navigate
to the x,y for each point that you add up. It's even easier if you
decide what spacing you want the points to have ahead of time (rather
than number of points per sample area of unknown size). If you want
your points on a 20m grid, go to the SW corner of your sample area and
add 20 to each of the UTM values for your first point, 20 to the x value
for your next point to the east, 20 to the y value for the sample point
north of that, etc etc.
Sorry for writing a book...I really dig sampling methodology and
GPS/GIS stuff...
And feel free to drop me a line if anything is unclear or if
there's any other way I can help.
PS I was doing a quick search for the appropriate projection for
Japan and ran across this:
http://cast.uark.edu/home/research/data-distribution-and-discovery/gis-data-resource-guides.html
might be worthwhile to try and track down a copy of that Japan
Mapping Resources Guide....
Sent:
Mon 6/21/10 6:12 PM
To:
L Quinn ([email protected])
Hi, Lauren,
I recommend that you rent or borrow a mapping-grade GPS unit (like a Trimble
GEO). you can get the fine grain you want, and can set out a grid anywhere you
want by using a process called offsetting. Very cool.
You also might want to consider taking a mapping GPS workshop when you get a
chance. I happen to teach such, so I'll send you some info to consider
Cheers!
Hotmail has tools for the New Busy. Search, chat and e-mail from your inbox.
Learn more.
_________________________________________________________________
Hotmail is redefining busy with tools for the New Busy. Get more from your
inbox.
http://www.windowslive.com/campaign/thenewbusy?ocid=PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_2
