Welcome, Guest. Please login or register.

Login with username, password and session length
Forums Search:  


Author Topic: A question about accuracy for the gurus  (Read 8954 times)

tawood

  • Jr. Member
  • **
  • Posts: 7
  • Karma: 0
    • View Profile
A question about accuracy for the gurus
« on: October 28, 2010, 08:26:46 PM »
I'm using my new garmin 62s gps to do a "rough" survey of my 40 acres....basically, I'm marking (and re-marking/averaging) all four corners of my property.  Then I plan to put in the four to a route, and use the routing lines as a rough property line guide to post some "No Trespassing" signs.  To be on the safe side, I'll post them 10 feet or so in from the route line.
My question is:  would it matter, for utmost accuracy, which "position format" I use?  ie. hddd mm.mmm, vs hddd mm ss.s, vs hddd.ddddd?  Does the gps carry decimal places beyond what it displays?  If not, it would seem to me that the last choice (hddd.ddddd) would hold the most accuracy for the following:
hddd mm ss.s ......this carries minutes to 1/600, or degrees to 60,000th
hddd mm.mmm....this carries minutes to 1/1000, or degrees to 600,000th
hddd.ddddd..........this carries degrees to 1 millionth

or are my figures wrong?  (i'm no math major!!!)

maps4gps

  • Expert Advisor
  • *****
  • Posts: 1524
  • Karma: 17
    • View Profile
Re: A question about accuracy for the gurus
« Reply #1 on: October 29, 2010, 06:08:11 AM »
An arc minute is approx 6080 feet in the north-south direction; and 6080 ft times the cosine of the latitude in the east-west direction.  The decimal minutes format will be about 6 feet (NS).  Depending on the satellite configuration and other factors, the position the GPSr is 'showing' could be off by 10-20 ft or more.  

How open and level is the area along the property boundary?

« Last Edit: October 29, 2010, 06:23:38 AM by maps4gps »

Seldom

  • Expert Advisor
  • *****
  • Posts: 1852
  • Karma: 19
    • View Profile
Re: A question about accuracy for the gurus
« Reply #2 on: October 29, 2010, 07:09:40 AM »
Another question is whether or not the display units really affects how the data is stored in the GPSr.  I suspect the display is just a convenience for the user to view floating point data stored on the unit.  Since the accuracy of a commercial GPSr is around 10 meters, I'd suggest that to be on the safe side you post your signs 30' back from your presumed property line. (Or hire a surveyor with a Trimble and a ground station nearby to locate your line to within a couple of centimeters.)

Boyd

  • Expert Advisor
  • *****
  • Posts: 3906
  • Karma: 46
    • View Profile
Re: A question about accuracy for the gurus
« Reply #3 on: October 29, 2010, 11:02:08 AM »
I have spent quite a bit of time "surveying" my own land with a variety of GPS'es. Bottom line is that you are only going to get a very rough approximation, and your neighbors may not agree with you. Try visiting a known spot repeatedly and record a waypoint each time. You will see a surprising variation. Averaging them all *might* get you pretty close.

I started a thread about accuracy limits for Garmin's map format that might interest you: http://forums.gpsfiledepot.com/index.php/topic,1335.0.html

Making a custom map using accurate property data will be a better approach IMO. You can probably download property parcel GIS data from your town/county/state's website if you do a search. Overlaying this on high quality aerial imagery will give you some cues of what to look for when you mark your property line.

I used this technique to post my own land, however I border on a state Wildlife Management Area in dense woods, so I don't think anyone is likely to question it if my line isn't quite right. :)

tawood

  • Jr. Member
  • **
  • Posts: 7
  • Karma: 0
    • View Profile
Re: A question about accuracy for the gurus
« Reply #4 on: October 29, 2010, 01:15:55 PM »
Thanks Boyd...and good idea using the satelite images.  My property is VERY DENSE...hence the need to come up with some sort of approximation (I can not just "sight line" it, because I can only see 150 feet max.
There are some bigger trees that are actually visible on satelite pics, that I can use for better reference.

Also, since posting, I believe I have found an answer to my original question.  I can't now find the link I found before, but I now believe it is hdd.ddddd as I thought (taking into consideration that it may not matter if the gps unit carries more decimal places than it shows).

Tim

heyyou325

  • Jr. Member
  • **
  • Posts: 38
  • Karma: 0
    • View Profile
Re: A question about accuracy for the gurus
« Reply #5 on: December 12, 2010, 09:29:06 AM »
I guess I didn't do my homework when I posted my surveying thread just a couple after here.   I shot my waypoints at least 3 times, most of them came to the same coordinates, averaged any different, and followed the directions back to them.  I was within 2 feet on every one I got good satellite readings on.  I realize there is still a margin of error.  I'm doing 146 acres for family and my neighbors are curious about the way it turns out, as every surveyor who comes along gets different results.  I used Easy gps , a free download, entered waypoints, figured distances and directions (it will do that for you) came up with where I felt the interior corners should be.  I checked with a brother and 2 sister-in-laws who live nearby and they agreed.  But now I've got 2+ feet of snow so I want to wait to go back and find the new coordinates.  I have 1 family boundary corner that I haven't been able to get a good reading on because of the tree canopy.  Hardest one to get to too.  Two of my neighbors want to see my results and maybe do the same thing.  If I know them it'll be me doing theirs but they are agreeable and nothing is going to be moved unless everyone agrees.  Putting it over Google Earth allows you to see some things, and the easy gps program was a big help.  You can figure square feet and http://www.convertunits.com/from/square+feet/to/acre you can figure the acres.  It's time consuming but worked good for me.  The gpsr can lead you right to where the coordinates should be.  I'd check 3 or 4 times and average tho.  I can't see far on most of mine either, nor can I walk thru big trees.

jbensman

  • Expert Advisor
  • *****
  • Posts: 422
  • Karma: 7
    • View Profile
Re: A question about accuracy for the gurus
« Reply #6 on: December 13, 2010, 08:20:55 AM »
Are you using your GPS's averaging function?  The newer Garmins have a really nice averaging program.  It will also allow you to go back to the waypoint on different days and get even better accuracy.  It sounds like you may be averaging by some other way.  Using the GPS's averaging function will get you the best accuracy.

heyyou325

  • Jr. Member
  • **
  • Posts: 38
  • Karma: 0
    • View Profile
Re: A question about accuracy for the gurus
« Reply #7 on: December 13, 2010, 07:40:23 PM »
I didn't know my GPSR had an averaging function.  I have an old 76CS which I bought for fishing trips and keep finding more uses.  The furthest off I ever was was only 1 number north/south and 1 number east/west.  When I returned home, after getting all the waypoints, I put them on the easy gps program.  Enlarged the file to the 2 foot level, and picked a spot right in the middle, copied the coordinates and used them.   Then I searched with the GPSR for the spot.    they were within 2 feet or less.   With my original coordinates, I had 18 of the 21 identical.  1 was one number off east and west, 1 was off both 1 east and west, and north and south, and the third I could only get 1 reading in the dense canopy.  All waypoints were shot on 3 different days.  When I plotted them, and then searched for the mark on the ground, the furthest off was  2 feet.  This was done on the third day.  Then I walked 50 or so feet away and searched again, and then a third time.  I am not moving any fence lines (at least yet, I don't figure to anyway).  The surveyor my parents and a neighbor hired didn't have any square corners and none of the quarter miles were the right distance.  I will have to look into the averaging feature this winter since I can't (don't want to with all the snow) finish now. 

jbensman

  • Expert Advisor
  • *****
  • Posts: 422
  • Karma: 7
    • View Profile
Re: A question about accuracy for the gurus
« Reply #8 on: December 14, 2010, 12:15:09 PM »
I confused you with the OP who said they had a 62 which has waypoint averaging.  Old GPSs like yours do not have the averaging capabilities of the newer Garmin GPSs.  I know the 60CSX had averaging but it just averages.  Yours may have the same thing.  The newer GPSs (starting with the CO) have intelegent averaging and do a much better job.  The newer GPS averaging functions take into account things like how good your signal is and can reject bad readings. 

Sonny

  • Jr. Member
  • **
  • Posts: 9
  • Karma: 0
    • View Profile
Re: A question about accuracy for the gurus
« Reply #9 on: December 18, 2010, 03:09:41 PM »
I am a Professional Forestry Consultant with 40 years experience. We use GPS to locate corners and lines. As far as establishing corners, I have had a Garmin 62S, a Magellan Mobile Mapper 6 and a Trimble  Geo XT together as a test of accuracy on an established survey corner. The known coordinate information was entered on the units and the result was not surprising, the Garmin 62s had the worst accuracy, the Magellan was better and finally the Trimble was the best. Not surprisingly, the price for the units vary upward with the accuracy. However, even with the Trimble, a single unit without a base station that has completely located its position down to the centimeter that  the handheld unit can  communicate with is not accurate enough for establishing corners. Also, in this state, you must be a licensed surveyor to do this.
We routinely use all of these units to find lines and corners to reestablish lines. A lot of times there is more out there in the field than many people will notice. Most land has been surveyed or had a line run through it in order to survey nearby land.  Putting posted signs around your property does not constitute establishing your lines.
We use Terrain Navigator to draw out the land parcels according to descriptions, download them to the units and proceed from there. However, drawing the parcels and transferring them to the units adds another level of error.

Seldom

  • Expert Advisor
  • *****
  • Posts: 1852
  • Karma: 19
    • View Profile
Re: A question about accuracy for the gurus
« Reply #10 on: December 19, 2010, 01:43:13 PM »
Thanks, Sonny, I needed that.

 

anything