A Question From a Newbie (way to measure GPS and Google Earth error?)

 

Hello everyone. I have followed this forum for several years but since I have little to contribute, I never saw the need to join until now.

I am a real estate agent who often deals with large undeveloped land tracts. Thanks to all the great information I have learned from your forum, and some help from my husband, I create Basecamp files using topographic maps, deed information and tax office data to mark waypoints for property corners and draw property lines. I view these files with clients using Google Earth Pro. Afterward, I download the files to my GPS and show the property to interested clients.

There are professional software packages which will do this but since I am a one woman operation, the cost is prohibitive. The Google Earth Pro property line overlay I find to be both inaccurate and incomplete.

I know about GPS error, my handheld unit estimates that. Due to the wide open skies here in the southwest, I often get accuracy of less than 20 feet, which is good enough to give a client a rough idea of where a property line is. I cannot afford to pay for a survey when the sale may not go through.

My question is this: What about the error in Google Earth? I was led to believe it was highly accurate when I paid $400 for the Google Earth Pro license. Now it is free! Just my luck. Internet research I have done shows it is not that accurate and can be way off in many areas. Since the posters here use it a lot, how do you handle this error in your files? I understand precision is not that important when you are trying to find say a gas station. That big old Exxon sign is a big giveaway. I work with physical reference points listed in deeds like “center of stream bed”, “rock wall” or “highway centerline” which are visible to Google Earth. I use offsets from these points to create waypoints for my property corners.

Google Earth error has been mentioned here several times in the past but a recent post caught my eye. It mentioned a way to measure GPS and Google Earth error but now I can’t find the post. I need to know how far off my property corner waypoints actually are.

I would really appreciate any tricks you can show me that would help with this error. Thank you all in advance.

Jeannie

Waypoint averaging.

There are some experts on this site and I am not one of them, so hopefully the more knowledgable folks will respond.

From my experience, the accuracy measure on the satellite screen of consumer grade GPS devices and Google maps get you close, but you can't really depend on them for things like property lines.

You mentioned that you use some type of known reference points to determine the points you wish to mark. You also said you have a "handheld" gps.

Most of the higher end Garmin handhelds have a waypoint marking feature called "waypoint averaging". This is often used by geocachers who want to get very accurate coordinates for their cache placement. This takes a lot of patience. To use it correctly, you have to take readings of the same location at different times over several days. And it still depends on the conditions of the location (canopy cover, clear sky obstructions such as cliffs and buildings, etc.) But under good conditions, the averaging technique has been shown to get very accurate coordinates for a given point. In general, the more readings you take, the more accurate the average becomes.

You can use the averaging technique on a reference point (like a tree you think is near the property line), then use coordinate calculation formulas from your known reference point (middle of road for example) to calculate how close your reference point is to the line. These can be some fairly complex calculations but there are online tools and downloadable programs that will do these coordinate based distance calculations.

The other thing I will mention is to make sure your handheld GPS supports Glonass, as that may increase the number of GPS satellites that can be "seen" at a given place and time, which will help improve the accuracy in difficult locations.

--
Alan - Android Auto, DriveLuxe 51LMT-S, DriveLuxe 50LMTHD, Nuvi 3597LMTHD, Oregon 550T, Nuvi 855, Nuvi 755T, Lowrance Endura Sierra, Bosch Nyon

Compare

You could also try comparing your results from Google to Bing Maps. On Bing maps, to get the GPS coordinates, right click on the spot and a small windows will pop up with the info.

--
Garmin Nuvi 2450

More Information

First, welcome to POI Factory! Since you’ve been following the forum for some time, you probably know the basics on how things work around here. Take advantage of the search box at the top left. It may take a couple of tries but you can usually find what you want. There is also a great FAQ section with a wealth of good information.

You’ve obviously mastered the basics of creating and working with .gpx files. Who knows, you may become a contributor!

I think this is the thread you are looking for:

http://www.poi-factory.com/node/44197?page=1

I mentioned the Google Earth accuracy experiments in a lengthy, off topic post near the bottom of page 2.

Most file creators here either don’t know about G E error or just ignore it. For the reason you mentioned, it doesn't adversely affect the overall usefulness of these file since you can get a good visual reference on most waypoints or POI’s while navigating.

The process you’re using with consumer grade GPS is pushing the limits as far as accuracy is concerned. Alandb offered some good advice about position averaging and Glonass to minimize GPS error. Your Montana 600T is a good unit and will position average but it is not Glonass capable. Some information on Glonass and a list of Garmin compatible units is available here:

https://support.garmin.com/support/searchSupport/case.faces?...

Without expensive equipment, Google Earth error is difficult to measure. You can see part of it by using it to zoom in on a well-defined object in your area. Then use the time slider to go back to previous images and see how your object appears to move while the coordinates do not. You can use G E’s ruler tool to get a rough idea of this error by measuring the object shift.

As you well know, highly accurate commercial GPS equipment, which use RTK and differential beacons, are available but cost thousands of dollars. Perhaps you could ask one of the surveyors you deal with who use this type of equipment to do an accurate survey on a point visible with Google Earth, like a pole or highway marking. Then simply go to the coordinates he gives you in G E and measure the offset from the visible object. Surveyors often know the location of points like this and may be willing to give you the coordinates.

There are places where “bluebooked” benchmarks, which are used to align the G E coordinate system with the standard WGS84 datum, are visible using G E. These are often the starting points surveyors use in their work. If you can find one, you can get the exact (within a quarter inch) coordinates by looking it up here:

http://www.ngs.noaa.gov/CORS-Proxy/NGSDataExplorer/

Again, go to these coordinates in G E and measure the offset.

I know of several such points but they are here in the east and would be of no use where you are located.

You may be able to get more advice by asking a surveyor friend.

Some Good Advice

Thank you alandb, jfossy and bdhsfz6 for your replies. You have given me some good suggestions.

Yes, that was the post I was looking for. The Washington Monument was a good example.

I may try the EPE software you all think so highly of. It could simplify what I do.

Thank you again

An observation

When I create a POI, I attempt to place the pin in the center of the driveway entering the business about 20-50' back from the curb using Google Earth. I have observed that over time (maybe a couple of years) my points have drifted so that when displayed in Google Earth now, they are no longer in the driveway but off in the trees a bit. The GPS is not in play here, only coordinate values and how they appear on GE. Is this my imagination? For my needs, this is not critical, but it may be a problem for blue-bird if Google Earth's images float around a bit under the coordinates.

Related to this is GE often projects a road over the image. Sometimes it is offset so that it runs parallel to the road in the image. I've assume this is a road one of the GE camera cars has traveled on and the difference is a matter of the onboard gps position calculation but I really don't know the source of the projected road. I've also noted that the GE image is a patchwork with different image resolutions and perhaps from different years. As you scroll across the image you can often spot the junctures of these images and the road or other linear features will be noticeable offset at that point.

--
"There's no substitute for local knowledge" nüvi 750, nüvi 3597