Jump to content

Question about Data Sheet Description


andylphoto

Recommended Posts

I have looked briefly for RL0161 without success. I did only a brief search while in the area searching for other stations. I wasn't expecting success with the time I had. I did not have my metal detector, and the area had been logged within the past few years, leaving many brush piles. I also realized later that I actually misread the description on my first search attempt. Now, however, I'm not entirely sure where to measure from.

 

AT THE OUTSIDE OF A LONG CURVE WITH TANGENTS SOUTH AND NORTHWEST, ABOUT 125 YARDS NORTHWEST OF THE SOUTHEAST END OF THE NORTHWEST TANGENT OF CURVE

 

So my question is about where the SE end of the NW tangent would be, exactly. I've got two ideas that both seem plausible to me, and I wanted to get a consensus before I try another search in the spring. Would the SE end of the NW tangent be...

 

1. The point at which the straight-line portion of the road (NW/SE) begins its curve to the south, or

2. The point of intersection of the extended center lines of the NW and South tangents of the road described.

 

The first would be (geometrically speaking) the point of intersection of the line representing the NW tangent and the circle, a portion of which makes up the curve in the roadway. The second would be the point of intersection of both tangents, and would mark a point off the roadway, near the center of the curve.

 

The first would seem to me to make more sense, but then the placement of the station would then be well into the straight portion of the roadway, and would seem to be better described as "NW OF A LONG CURVE..." rather than "AT THE OUTSIDE OF A LONG CURVE..." When I read the latter, it seems to place the station at some point between the ends of the curve. I would guess that the boulder described later in the data sheet is probably still present, but getting started searching in the right area is always a good thing. I found a boulder during my search in the wrong area that seemed to match the description too. Thanks for any input!

Link to comment

Number 1 is correct. Number 2 is called the Point of Intersection or PI.

 

Outside would tell me to look on the NE side of the road. I agree that this is a vague way of describing it and "125 yards NW of the NW end of a long curve and xxx feet NE of the center of the road" would be clearer, if we are correct in the interpretation.

 

This figure from someone's college lecture may help.

Edited by Bill93
Link to comment

Bill93 seems to have figured it correctly. Apparently, the the writer was describing the direction of travel on the road, where there is a curve to the northwest. The mark shows on the topo map. The "X" is at:

 

43 34 18.8 N

 

88 21 02.7 W

 

or, a short distance south of the scaled coordinates (shown as a cross-hair inside a box).

 

-Paul-

 

 

 

17570045.jpg

Edited by PFF
Link to comment

Here is a screen capture from ScaredyCat's Benchmark Viewer on which I have drawn a 125 yard radius circle around the plotted BM. It falls about 125 yards N of the PC/PT of the curve, and 165 feet SW of the PI of the two tangents.

 

93478f63-4144-4015-92bd-a9f54963878b.jpg

 

As can be seen, the plotted BM location really doesn't correspond very well with any interpretation of the datasheet description. It is about 575 feet SE of what you called point 1, (the PC/PT of the curve), rather than 375 NW of it. It is about 165 feet SSW of point 2 (the PI of the two tangents). However, it is also about 375 feet (125 yards) North of the PC/PT of the curve coming from the South tangent.

 

Perhaps the picture will help you better than the words!

 

Bill93 points out that the BM icon, which represents the location of the scaled coordinates, lies NW of point 2 (the PI). The plotted BM location is marked by an X to the South of this icon.

Edited by Holtie22
Link to comment

The mistake in the datasheet description I see is that it should be

 

ABOUT 125 YARDS SOUTHEAST OF THE

SOUTHEAST END OF THE NORTHWEST TANGENT OF CURVE instead of Northwest.

 

It isn't that easy to specifically identify the tangent point of the curve to much better than 35 feet or so.

 

The other calls then to the road and trail as well as to being on a large rock and using the topo coordinates should be okay.

 

If you use the 125 yards that moves it NW along the road from where the topo sheet puts it so you have a few hundred feet to search along.

Link to comment

Thanks everyone for the advice. I had forgotten that this one appears on the topo map. I had actually adjusted the coordinates I used, which probably put me in the right area. After not finding it on the first time out, I went back home and re-read the data sheet, figuring on the next round, I'd do some more accurate measuring, and then found the data sheet conundrum.

 

This will be a good one for spring when there are no leaves and little underbrush. If I can track it down, I'll try to find this thread and give an update. I'm also going to try a spring search (for the same reasons for the one just south of this one, RL0012. These are two of the holdouts in my run on 1985 NGS "not founds" in this area. The second one may have been destroyed by a widening of the road or logging activity, or it may be just off the road in the underbrush. With the roads being gravel, measuring from the where the center of the road was 50 years ago is a bit of a trick. So was running a tape 99 feet through a forest with dense undergrowth.

Link to comment

Ah, but how do you get the coordinates off the topo map?

 

The popular Acme Mapper, with its Google Maps-like interface, allowed me to zoom in on the X that marks the location of RL0161 and gave me coordinates of 46.57189 , -88.35073 (46° 34' 18.804", -88° 21' 2.6274").

 

Acme says it's a "general purpose high-precision map application," but how accurate, exactly, is it? They don't tell us.

 

Another source for georeferenced topo maps is the USGS itself. First, install the free GeoPDF Toolbar, then download the map(s) you need.*

 

In this case, the location was slightly different: 46.57195 , -88.35068 (46° 34' 19.02", -88° 21' 2.4474").

 

The distance between the two sets of coordinates is about 25 feet (7.689 meters), probably about the width of this road.

 

I'm inclined to trust the USGS product, but I'm not really qualified to judge.

 

-ArtMan-

 

_____

*Click on "Map Locator," but don't be misled by the "BARAGA" quadrangle listed on the datasheet; this location is actually on the Nestoria map.

Link to comment

Hi, ArtMan,

 

Thanks for sharing your results. I got my coordinates by using a program developed by one of our GEOCAC members. It will overlay a topo map on GoogleEarth, and adds "balloons" for the NGS benchmarks.

 

With this system, I can zoom in on the "X". I also pan to keep the "X" in the center of my screen. With the grossly oversized "X", there is no doubt about having the cursor in the exact center. I find that this technique puts me within sight distance of a mark--generally 25 to 40 feet. I have used this method dozens of times with conventional benchmarks, and I recently used it to locate a 100-year-old meridian at the request of N.C.G.S.

 

The station was in a cemetery and the original description gave a name on a headstone. Would you believe that there were TWO persons with the same, rather unusual, name? They are buried several hundred feet apart. I was following what I thought was the original road and came to the first headstone. "Easy find," I thought to myself, although I was disappointed that my projected coordinates seemed to have missed by 300 feet.

 

But when I did not find the station 50 feet northeast of the grave marker, I thought I'd give the projected coordinates a try. Bingo! I won't say I tripped over the stone marker, but I was very, very, very close! [Grin.]

 

I liked the results from the Acme Mapper, Art. I'll have to give that a try. Here's how your numbers compared with mine:

 

46.57195, -88.35068 USGS Toolbar

 

46.57189, -88.35073 Acme Mapper

 

46.571892, -88.350747 GoogleEarth Topo Overlay

 

 

That's close enough for government work. [Chuckle.]

 

The image on the GE Overlay is very clean, as you can see below. (I zoomed in several stages closer than this to read the coordinates.) For comparision, the published (SCALED) coordinates for RL0161 are indicated by a cross-hair inside a grey box.

 

-Paul-

 

 

dece5692-85f6-4ad6-bffb-2cb14c4c7494.jpg

Edited by PFF
Link to comment

I use MacGPS Pro, and have USGS topos that I can access within the program. I've had very good accuracy with coordinates I've "adjusted" from the USGS maps in the past. In fact, I've been quite amazed at the accuracy the USGS incorporated into these maps, many quads still dating from the 50s, and how their placement of these benchmarks is so spot on.

 

I walked in along an abaondoned rail line in October with my dad--we had both taken coordinates for RL0015 from different topo programs, and both zeroed out within about 20 feet of the actual mark, which is about 1.8 miles along the track from the nearest accessible road.

 

Coordinates I had in the GPS for this one (RL0161) were 46 34 19.0 N 88 21 02.3 W

 

Time to show up with a GPS, a good tape, a metal detector and a compass for another looky see on this one. :-)

Link to comment

When people start comparing values down to 10's of feet, it becomes important to keep track of the datum, at least in some parts of the country.

 

Topos that I am familiar with are NAD27, so you need to set your GPS unit to that datum while entering the coordinates and then switch back, or else use some conversion program (NGS utilities or CORPSCON) beforehand so you can leave the GPS in NAD83 (correct for data sheets) or WGS84 (identical in many handhelds).

 

Does Google Earth (or some of the other on-line services) offer choice of datum? I don't use GE on my dial-up connection so don't know.

Link to comment

Here are some sample values

N42 W091 NAD27 is NAD83 N42 00 00.06 W91 00 00.54 (Iowa)

a shift of 1.75 meters lat, 12.40 m lon, a total of 12.52 m=41 feet.

 

N26 W081 NAD27 is NAD83 N26 00 01.33 W080 59 59.25 (south Florida)

a shift of 41.08 m lat, -20.96 m lon, total 46.12 m=151.3 ft

 

N36 W077 NAD27 is NAD83 N36 00 00.57 W076 59 58.84 (N. Carolina)

a shift of 17.63 m lat, -29.13 m lon, total 34.05 m=111.71 ft

 

N45 W069 NAD27 is NAD83 N45 00 00.24 W068 59 58.14 (Maine)

a shift of 7.47 m lat, -40.80 m lon, total 41.48 m=136.08 ft

 

N48 W124 NAD 27 is NAD83 N47 59 59.30 W124 00 04.73 (Wash. state)

a shift of -21.55 m lat, 98.06 m lon, total 100.40 m=329.40 ft

 

N33 W117 NAD27 is NAD83 N33 00 00.16 W117 00 03.11 (SW Calif)

a shift of 4.95m lat, 80.77 m lon, total 80.92 m=265.49 ft

 

N28 W098 NAD27 is NAD83 N28 00 01.06 W098 00 01.00 (S. Texas)

a shift of 32.54 m lat, 27.45 m lon, total 42.57 m=139.67 ft

 

N47 48 34.2 W83 33 32.4 (East of Lake Superior) Same in NAD27 and NAD83

Edited by Bill93
Link to comment

A couple of other points about issues raised in some of the recent posts in this thread. First remember that the USGS 1:24K map product was compiled using the National Map Accuracy Standards which basically says that most features shown are within 1/50th of an inch at the given map scale. This comes out to about 40 feet in this case.

 

While digitizing points off of a quad may be better than that, it is a general indication of about how good the product was intended to be. Add to that paper base which changes size and so forth, you should not expect to be able to get coordinates from it to 5 feet accuracy for example.

 

The fact that we now have by virtue of scanned images of the maps, the ability to drill down and measure easily less than this does not change the inherent issue in the product itself, but makes it easier to forget about it than the days when you had to physically scale off the paper copy.

 

Most of the USGS maps were compiled using some projection based on a graticule in the NAD27 datum. However, unless you are using a paper map, many of the online systems as well as PC based mapping software allow for the map to be registered in NAD83 and in that case the coordinates you digitize are in NAD83 assuming that it was all done correctly. One way to check is to find one of the 2'30" tick marks on the face of the map and see what kind of coordinate you have for it.

 

The shifts between the datums are small in some areas, but signicantly large in many areas of the country.

 

- jlw

Edited by jwahl
Link to comment
I use MacGPS Pro, and have USGS topos that I can access within the program. I've had very good accuracy with coordinates I've "adjusted" from the USGS maps in the past. In fact, I've been quite amazed at the accuracy the USGS incorporated into these maps, many quads still dating from the 50s, and how their placement of these benchmarks is so spot on.

I, too, have had excellent luck with marking the locations of "X" marks on USGS topo maps within MacGPS Pro. I use free downloadable DRG files from the California Spatial Information Library.

 

Patty

Link to comment

I've also found that USGS quads are often still in NAD27 datum, and even after correcting from the datum, the map can still be shifted by tens of feet.

 

One way to check is to find some NGS horizontal control points (preferably ones that have GPS adjusted coordinates) and find the USGS quad that has the corresponding points marked on the map. Check the coordinates taken from your favorite map tool with the coordinates as published on the datasheet.

 

Another way is to look at the map, find a recognizable landmark like the exact center of a road intersection, and compare the coordinates from the map with coordinates taken from one of the aerial image programs like Google Earth or Microsoft Virtual Earth. The "satellite images" (usually actually orthographic aerial photos) are usually pretty accurate. The USGS map is often quite a bit less accurate.

Link to comment
One way to check is to find some NGS horizontal control points (preferably ones that have GPS adjusted coordinates) and find the USGS quad that has the corresponding points marked on the map. Check the coordinates taken from your favorite map tool with the coordinates as published on the datasheet.

 

Excellent suggestion, Jim. This would be especially effective if you use a trianglulation station which is shown as a triangle on the map.

 

According to Alan-Jon of the SCGS, errors are introduced by photographing a round object (Earth) and displaying it as a flat photo or map. One of his tasks is to go throughout South Carolina to take precise GPS readings at places which can be identified in aerial photos. They are using this data to do corrections--although I don't know anything about the process. I would think that this is being done in other places, as well. If so, the accuracy will get better with time. But it also points out the wisdom of jwahl's comments about getting different results in different parts of the United States.

 

The interesting thing is that in spite of the imprefections, using the coordinates derived from a topo map are a great help in looking for marks with SCALED coordinates (or where no coordinates are available).

 

It makes you wonder about those SCALED coordinates, doesn't it? I mean, if they knew where to put the "X" on the map, why are the data sheet coordinates so far off? [Grin.]

 

-Paul-

Link to comment

According to Alan-Jon of the SCGS, errors are introduced by photographing a round object (Earth) and displaying it as a flat photo or map.

 

An orthorectified aerial image has had several adjustments: 1) the image plane has been adjusted for tilt and camera characteristics, 2) the elevation of each pixel has been determined from a DEM or some other source of elevation data and projected to a level vertical datum, and 3) the adjusted pixels are then projected to a plane using some map projection.

 

When you look at an aerial image such as Google Earth, Virtual Earth, or Terraserver, those have all been orthorectified. If you incautiously take some tall landmark such as a water tower, building dome, or radio antenna, you will find that the top of that landmark in the image is NOT georegistered correctly, since invariably the image was taken from a camera located somewhere to the side of the object, and while the various processing steps have resulted in correct georegistration of ground level objects, but objects that have significant height above ground are not correctly georegistered.

 

A number of states (my own New Jersey, and neighboring New York, as examples) have provided excellent 1-foot resolution images in the local state plane coordinate system as downloadable files. Often those same images have been used for Google Earth or Microsoft Virtual Earth, who reprojected them to UTM or some other projection. Terraserver uses the images from the USGS.

 

edit:typos

Edited by holograph
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...