Map/projection datum precision questions

[jackchinook]

I'm not sure if this is the right forum for this but thought I'd take a stab anyways.

 

I'll also add that I'm currently working on getting my head around all of this stuff, so bear with me.

 

I'm trying to set up a database of important sites/waypoints we use in our fish surveying efforts (I'm a biologist) and I'm trying to draw up some protocols to help technicians all collect data simultaneously that are interchangable across several districts/employees/GPS units/computer units, etc.

 

How much variation should I expect across maps/units with NAD 1927 vs NAD 83 vs. WGS 84, etc.? Specifically, if waypoints are collected while a unit is set to one datum, then downloaded and plotted on a (computer) map that is project according to another datum, how much 'slop' should I expect?

 

Consider that these are these waypoints by no means require 'survey-level' precision/accuracy. We use them to mark things like bridge/culverts or locations of temperature logger devices. I'll have a temp logger tied to a tree on a riverbank (or something like that), a written description of where it is (or a waypoint plotted on a topo map) to get me in the near vicinity. Then I'll need to root around and get dirty.

 

I'm just trying to get a feel for how much effort I need to make to ensure that everyone's units are set to WGS 84, etc.

 

Thanks for any help you guys might be able to give. Oh, and any good basic overview text or other sources that can explain this to a NOOB are appreciated as well.

[Bill93]

The short answer is that NAD27 vs NAD83/WGS84 may be significant for your purposes, depending on where you are, but NAD83 vs WGS84 is not. Handheld units offer your choice.

You should probably be sure that your people stick with one datum or another.

 

Around here NAD27 is about 50 ft different for the same coordinate values. The differences between NAD83 and WGS84 are usually well under a meter and not measurable by handheld GPS units.

 

The NGS has a tool that will help you check this. Click on ToolKit and select NADCON and then again select the title NADCON. There are both on-line and downloadable versions of the converter.

[+Black Dog Trackers]

jackchinook -

 

To put it very bluntly, a mixture of (NAD 27) and (NAD 83 or WGS 84) coordinates should not be allowed.

 

NAD 27 is based on an antique (1866) ellipsoid of the wrong shape. Coordinates based on either NAD 83 (1980 ellipsoid) or WGS 84 (1984 ellipsoid) should be used instead.

 

The coordinate precision of handheld GPS units is 0.1 second, and this is about 10 feet in terms of distance. The accuracy of handheld GPS units is about +/- 10 meters, so they are a bit more precise than accurate.

 

The main answer to you is that if you're looking for a particular tree, it's better to be looking in the correct 10 meter wide area than looking in the wrong 10 meter wide area that might be 50 meters away!

 

I have looked for, but never have seen, any software that shows the difference in distance between NAD 83 and WGS 84 coordinates (or coordinate difference associated with a distance); it's just too small for people with handheld GPS receivers to be concerned with as Bill93 points out. So a mixture of NAD 83 coordinates and WGS 84 coordinates would cause zero problem for you.

 

As to maps, I imagine you might be referring to old 15 minute topo maps based on NAD 27? Determining coordinates from topo maps by scaling with a ruler is often seen to be up to a few hundred feet off. NGS datasheets for bench marks whose coordinates were determined by scaling from a topo map always have this caveat statement:

The horizontal coordinates were scaled from a topographic map and have an estimated accuracy of +/- 6 seconds. (A second translates to about 100 feet.)

[Papa-Bear-NYC]

I'm not sure if this is the right forum for this but thought I'd take a stab anyways.

...

How much variation should I expect across maps/units with NAD 1927 vs NAD 83 vs. WGS 84, etc.? Specifically, if waypoints are collected while a unit is set to one datum, then downloaded and plotted on a (computer) map that is project according to another datum, how much 'slop' should I expect?

...

Thanks for any help you guys might be able to give. Oh, and any good basic overview text or other sources that can explain this to a NOOB are appreciated as well.

You can do a real life "virtual" check on the difference using Topozone.

 

Find an area you are familiar with, bring up the Topozone map at the best resolution (probably 1:24000) and click on something. It doesn't matter what, really.

 

Now set the datum to NAD27 and the coordinate display to UTM. The coordinate value is displayed at the top of the window. UTM is nice for this check since it will tell you in meters the x and y coordinates from an origin.

 

Now change the Datum to NAD83/WGS64, and check the coordinates.

 

Now you can see the real life difference in units you understand.

 

This is actually the reverse of what Bill93 suggests. He said keep the coordinate values fixed and see how the point changes with the datum. the Topozone test keeps the point fixed and you see how the coordinate values change. Two ways of looking at the same thing.

 

An example. I live in New York City. Let's check the coordinates for the Empire State building. We'll use the north-east corner (34th St. & Fifth Avenue). Remember, the point of this test is not to see how well Topozone will display where the corner of the Empire Sate Building is. It has resolution constraints and I may have clicked the point badly. The point is to see the differences in the two datums (data?) at any particular point. In actuality, we're just using their conversion software.

 

Here's the map: http://www.topozone.com/map.asp?lat=40.748...7&layer=DRG

 

For NAD27, the location is UTM 18 585681E 4511094N (NAD27)

 

Now I'll change the datum (without changing the point)

 

New coordinates: UTM 18 585713E 4511317N (NAD83/WGS84)

 

So it looks like the coordinates moved 32 meters to the east and 223 meters to the north. Fairly substantial.

 

So, if you hung a thermometer there and gave someone the coordinates, and he went and looked for it with his GPS using the wrong datum, he'd be almost 3 blocks away. Of course if you use a GPS and report a location in a given datum, someone else can change to another datum and correctly find the point. The problem will (inevitably) arise when folks are not careful about which they use. Thus it's best to have a standard. I'd say stick with NAD83/WGS84.

Edited December 7, 2006 by Papa-Bear-NYC

[jackchinook]

Wow! Great responses. You've convinced me to make the effort to at least stick with NAD 83/WGS 84.

 

I'm using Maptech Terrain Navigator and another computer with TopoScout as my on screen tools for this project. Maptech appears to let me change the datum/projection while TopoScout does not. I created the same points, wrote the coordinates down to 8 digits or so, then plotted them, set the preferences back to Nad 27 and the two points were 113 feet apart. Lesson learned.

 

Many thanks.