GPS Precision

Localization success depends on the accuracy of the point coordinates, the accuracy of the GPS ties to those points and the geometry of those points.

I've seen people localize to points under trees. Bad idea. Just like a total station resection you don't want to extrapolate and make ties beyond the extents of the resection. Don't make ties to points beyond the extents of the localization.

Thanks for responding! Okay. So if I were to localize on four well spaced corners of a parcel using a bipod to stabilize my rover and use corners that have been adjusted by the office, I might be in pretty good shape to set missing corners? On some of our larger parcels the rover switches bases. Will that throw any error into my localization?

How critical are the corners - a pasture in the boondocks or high-priced commercial land?

Localizing on four points could introduce some weird distortions, and gives you no checks.

If you localize on one point (two if there is ambiguity in the basis of bearings) and check the other known points, how closely do they match? That should tell you a lot about the repeatability of the data. If that doesn't check out, distorting the figure to fit four points may not be an improvement for the remaining one(s).

Thanks for responding! The corners are for high-priced commercial land. When you say "Localizing on four points could introduce some weird distortions, and gives you no checks.", what do you mean by "and gives you no checks" ? Would getting good residuals be good enough?

Good residuals in a localization can be misleading.

Thanks for responding! If good residuals in a localization can be misleading, what else can I rely on to determine the soundness of my localization?

Localizations are no substitute for following rules of dignity of calls. Localizations can be very effective for relating a GPS survey to an existing coordinate network.

Set And 3 GPS Positions Near The Point To Be Set

Occupy 1 of those points, backsight one and check to the other. If the check is 50% of your required precision set the corner. If you have used all of your error before stakeout, you may want to readjust or get more data.

Possible field readjustment is to resection to the other two points, mean your resection and GPS point, and redo your checks. Another adjustment is to average the backsight to check angles, given that you have reliable distances, i.e. an angular "wiggle in".

Paul in PA

I agree with you that localizations are no substitute for following rules of dignity of calls. The office has already adjusted the recovered corners and has decided where the missing corners are going. I'm new to GPS, so I'm trying to learn what I should or shouldn't use it for.

Set And 3 GPS Positions Near The Point To Be Set

Thanks for responding! Some of the existing corners are not intervisible. I would like to expedite setting the missing corners with my GPS.

Set And 3 GPS Positions Near The Point To Be Set

I have worked on surveys where the only tie in to the main traverse where 3 GPS traverse points tied to GPS traverse points. Using 3 GPS points you have 1 angle and 2 distance checks in the field. If possible do the GPS points so that the point to be set is somewhere within the inside angle and not more than 2 times your longer GPS to GPS distance.

Since you do not know exactly where the set point ends up, if trees block view from the angle point, you have two more possible points to occupy without extending your GPS traverse.

Paul in PA

You should never localize this type of project, use a good projection and locate the existing monuments, then it is time to determine the "errors" imbedded into the record measurements and decide how to replace missing monuments.

Localizing just makes a hash out of a project like this, the only time to do it is if you have existing NAD27 control that can't be changed, even then I would hesitate.

It's much better to put good state plane coordinates on existing monuments and go from there.

Calibration and localization was sold in the beginning of RTK as a way to "get on" existing instrumented control so you could just show up onsite and go, didn't make a lot of sense then and with better Geoid models it makes less sense today.

Bill has given you some good advice. I don't know if he will but I am sure that Kent could add a lot to this thread as well.

Good Luck

Soprry, I Missed The Localization Point

I have always used NAD83 into SPC as it came out of my post processing software or from OPUS/OPUS-RS.

Never had the need to localize, did squeeze a few into NAD27, but that was in my software. Why invent something that is not better than what has worked for many years?

Paul in PA