Jump to content
Sign in to follow this  
Followers 1
Schuhhirsch

Accuracy of scaled benchmarks

Recommended Posts

Hi there,

for me as european Benchmarking is excotic, although there are (of course) similar installations here in the old world :-) This year I could manage to find two benchmarks in Los Angeles, and I'm preparing a longer journey to south west USA for next year.

 

And here's my question:

The location of a benchmark has to be known very accurate, as they are used to be the reference for other geodetic measurements.

On the other hand, coordinates for "scaled" BM are said to be (un-)exact for up to a quarter mile. That is, in my point of view, not really accurate :laughing:

 

Is it that just Groundspeak (or the public) does not receive the "accurate" data for these BMs? How is one checking the border of an estate with +/-200yrd?

 

A second question: the coordinates in the Data Sheets are given in NAD. I've read somewhere that the difference to WGS84 is way below the accuracy of a consumer GPS. Is that correct? Can I use the NAD coordinates in WGS without being directed to the neighbour state?

 

Thanks for informations,

Phil/Schuhhirsch, Austria

Share this post


Link to post
And here's my question:

The location of a benchmark has to be known very accurate, as they are used to be the reference for other geodetic measurements.

On the other hand, coordinates for "scaled" BM are said to be (un-)exact for up to a quarter mile. That is, in my point of view, not really accurate

The term Bench Mark is used to describe of point of known elevation in the Surveying profession. In Gecaching the term is used to describe all survey marks. The methods employed to determine elevation are different the those for determining positions.

 

It was not necessary to know the exact location of the BM (elevation only) and thus they were just plotted on a topographic map and the location was scaled. The most important aspect of the BM was the elevation, not the position. Determining the position of a mark involved a lot more work and time than determining elevations. There was no need to know the elevations of all position marks to accomplish the scope of the project.

 

At times and more common today with GPS, a point of known elevation (BM) may also be later used as a point for determining exact position and thus you are seeing more and more BM's with latitude, longitude and elevation. GPS is not very accurate for determining elevations and thus the more marks they use with know elevations the better it will be to predict the elevation of unknown points of elevation.

 

The opposite is also true, at time it was necessary (mapping) to have a precise elevation on a known position mark and thus it was leveled to. Also, they used them out of convenience rather then setting a new mark were some others already existed.

Edited by Z15

Share this post


Link to post

Hi there,

for me as european Benchmarking is excotic, although there are (of course) similar installations here in the old world :-) This year I could manage to find two benchmarks in Los Angeles, and I'm preparing a longer journey to south west USA for next year.

 

And here's my question:

The location of a benchmark has to be known very accurate, as they are used to be the reference for other geodetic measurements.

On the other hand, coordinates for "scaled" BM are said to be (un-)exact for up to a quarter mile. That is, in my point of view, not really accurate :laughing:

 

Is it that just Groundspeak (or the public) does not receive the "accurate" data for these BMs? How is one checking the border of an estate with +/-200yrd?

 

A second question: the coordinates in the Data Sheets are given in NAD. I've read somewhere that the difference to WGS84 is way below the accuracy of a consumer GPS. Is that correct? Can I use the NAD coordinates in WGS without being directed to the neighbour state?

 

Thanks for informations,

Phil/Schuhhirsch, Austria

 

THe quick answer to you second question is yes, NAD 83, used in the datasheets and WGS84 are closer (I've heard generally sub meter) than either consumer grade GPS, or any on-line maps can produce.

 

The first question is trickier. There are two types of geodetic survey markers and traditionally (before high grade GPS) were established by different teams using different methods.

 

1) Vertical Control points (aka "true" bench marks) measured elevation to very high accuracy - probably sub centimeter even 100 years ago. But these teams didn't know where they were setting the marks in a geodetic sense (that is latitude and longitude) but they did know where they were with respect to local features (10.71 feet from the west curb of Main Street, etc.)

 

2) Horizontal Control Points, (aka triangulation stations) measured location to a high degree of accuracy (some times modern checks on 100 year old marks are also at the centimeter level) but didn't know how high they were. The only way they estimated elevation was through vertical triangulation which is not too accurate.

 

Then sometime in the 1960s, clerks in the CGS (now known as the NGS) office had the task to put all the reports from the preceding 130 years into a database, and put it on a computer. This was the advent of the data sheet. But for each mark they had to enter both vertical and horizontal information. So how did they get horizontal information for bench marks and vertical information for triangulation stations (which were not measured by the original surveys)? They looked at USGS topographic maps. The USGS is a separate agency from the CGS and made all the topo maps we know and love, but they obviously include locations and elevations for the areas where the NGS marks are located. So for bench marks, the clerks looked at a map, read the description, found the point (e.g. Main Street, etc.) and with a ruler, got a latitude and longitude. And for triangulation stations, the looked at where the mark was on the topo map and measured by eye where it was with respect to the contour lines. Obviously there is lots of room for error in this process.

 

So Horizontal marks have scaled elevations (scaled from a topo map) and Vertical marks have scaled location (same).

 

But the accuracy of the scaled information was not critical to the surveyor on the job - as long as he/she could find the mark. The land surveyor using an NGS bench mark for control, doesn't need to know accurate geodetic location, he/she just needs to know if it's the right street, house lot, building project, etc. He needs to make sure the pipes coming out of the building match the town water pipes. And the NGS surveyor measuring harbor lines, needs to have good horizontal control, but elevation is secondary.

 

Now with modern surveying methods (meaning high grade GPS) the surveying can get everything in one package, so a number of recently surveyed marks are accurate in both dimensions.

Edited by Papa-Bear-NYC

Share this post


Link to post

Schuhhirsch,

 

In addition to the above excellent answers, I want to point out:

 

1) Bench Mark (with a space) is the term the surveying community uses for vertical control stations.

 

2) Benchmark (with no space) is the term we hobbyists use for both vertical control stations and horizontal control stations.

 

3) Perhaps in Europe there is no analog of a vertical (vertical-only) control station?

Share this post


Link to post

Hi all,

thanks for all your answers, they all helped a lot :(

 

Here in Austria the situation seems to be a little different than in the US, as the stations are set up by just one single authority, and they do it since over two centuries.

 

So as a summary: a foreign tourist, that does not want to get in too deep into "Benchmarking" (without the space), but wants to find some "drive-by" benchmarks (because he wants to concentrate on locations like Arches or Bryce NP :D )

1. should avoid the "scaled" BMs, as they may need a lot of effort to find them

2. should not bother to transform NAD to WGS

 

For going for benchmarks at home I will try to deal with Waymarking.com

 

Thanks again,

Philipp

Share this post


Link to post

So for bench marks, the clerks looked at a map, read the description, found the point (e.g. Main Street, etc.) and with a ruler, got a latitude and longitude ...

Obviously there is lots of room for error in this process.

I remember this discussion about two or three years ago. Seems to me that the points were scaled using UTM, right?

Lots of times the numbers (digits) got transposed and caused many errors.

If NGS Surveyor or someone would like to do a writeup on this, it would make pretty interesting reading.

 

~ Mitch ~

Share this post


Link to post

1. should avoid the "scaled" BMs, as they may need a lot of effort to find them

2. should not bother to transform NAD to WGS

Concerning #1, I would have the opposite point of view. The "adjusted" benchmarks are often on private property, which requires obtaining permission, and are often not near a road. Neither of these are a big problem but are things that add to time spent. The "scaled" benchmarks are often on a bridge or very near a road or railroad. Don't bother using a GPS receiver to find them; use a tape measure instead. The use of a tape measure is not in itself a significant difference - looking for "adjusted" stations often requires tape measuring too. The main thing is to read the description to see how easy or difficult a mark will be. It could be either an "adjusted" or a "scaled" one that is easier.

 

Concerning #2, that is certainly true. There used to be some NAD27 (1927) but not any more, so for a handheld GPS receiver being used in benchmark search mode, you can consider NAD=WGS.

Share this post


Link to post

And, you can run into other strange problems. I found one that was a quarter mile off. Someone wrote N 41 09 42 W 074 23 40. The keyboard entry person typed in N 41 09 42 W 074 23 10. But it was easy to find, by the description.

On geocaching trips, my companions condescend to allow me to look for some benchmarks. But only if they were easy, or near geocaches. So I look for ones on bridges or publc buildings. As someone said, good descriptions can make for easy finds. Train station in Salem, Virginia. Railroad bridge in Pike County, Ky. Highway bridge in Sherman Mills, Maine.

Share this post


Link to post
Then sometime in the 1960s, clerks in the CGS (now known as the NGS) office had the task to put all the reports from the preceding 130 years into a database, and put it on a computer. ... So for bench marks, the clerks looked at a map, read the description, found the point (e.g. Main Street, etc.) and with a ruler, got a latitude and longitude.

Ah, is that how they did it? I've wondered.

 

So as a summary: a foreign tourist, that does not want to get in too deep into "Benchmarking" (without the space), but wants to find some "drive-by" benchmarks (because he wants to concentrate on locations like Arches or Bryce NP :D )

1. should avoid the "scaled" BMs, as they may need a lot of effort to find them

Phil, as BDT said, scaled marks are often in easy-to-find locations, so don't write them off.

 

Also, if there's an area you're particularly interested in, it's pretty easy these days to determine very accurate coordinates for the survey marks that are on USGS topo maps by using calibrated electronic versions of the topos. Granted, only a fraction of marks in the NGS database are shown on topos, but at least those are some you can work with.

 

That's great that you're planning another trip to the U.S. What areas are you thinking about visiting?

 

Patty

Share this post


Link to post

I vote also for bench marks along roads and on bridges being among the easiest to find. You can even see some of these marks with Google Earth's Street View feature - I haven't decided if this takes the fun out of it yet, but for a long trip it's nice to know the mark is still there. :D

Share this post


Link to post

There were a couple of different ways to get the scaled horizontal positions for bench marks. For those marks that were used by USGS as control for the topographic mapping series it's usually fairly easy to scale their position since they are shown on the topo map. Any other mark had to be manually plotted. If the mark was sent in a building or bridge these were often easy to locate and scale. Regrettably many thousands had to be plotted by the cartographer reading the description and trying to decipher the most probable location for the mark. Hence the rational for using a standard of 6 arc seconds for scaled positions.

 

Also note that the difference between NAD 83 and WGS 84 is approximately 1m each in the horizontal and ellipsoid height. This varies a little bit but it's a good ball park number to remember. Since the difference for mapping and charting is small the Federal Geodetic Control Subcommittee published a Federal Register Notice in 1995 that allows them to be considered the same for all mapping products with scales of 1:5000 or smaller -- NAD 83/WGS 84 Federal Register Notice

Share this post


Link to post

Dave, were the coordinates for the "X"s on USGS topo maps really derived by using rulers and measuring from the ticks on the sides of the maps? If so, what was the typical +/- accuracy of those coordinates?

 

Patty

Share this post


Link to post

Wintertime - That's pretty close to the way it was done. The NGS cartographic branch (long since disbanded) used a few other tools, but it wasn't much more sophisticated that than. The accuracy of point that was plotted on the map with an X is usually about 1 arc second (approximately 100 ft).

Share this post


Link to post

Some of the scaled errors make for real good head scratching exercises.

 

SR0221, one of a very few Missouri River Commission (MORC) bench marks (BM 48-2) to achieve PID status in Montana was recovered over 1000 feet west of the DATASHEET L/L. It was found about one third of a mile from the called powerhouse, very close to a BM X on the USGS Quad map and within 50 feet of the Army's published coordinates after applying an adjustment provided by SquareNail.

 

The DS position is about 0.20 mi SW of the power on the east side of a gully.

Share this post


Link to post

One source of large errors in scaled coordinates is when the location was described with reference to a road number that got moved to a different route between the time the disk was described (often 1935) and the coordinates were plotted (1960's). I've seen several examples of that.

Share this post


Link to post

Having watched the Quartermaster plotting the location of the ship while on watch on the bridge, I would imagine the actual scaling was done with dividers and not a ruler. One leg of the divider would be placed on the location of the benchmark and the other leg would be placed on the nearest lat or lon line on the chart (depending on which one you are scaling). Then the dividers would be moved to the side of the chart where the tic marks are and the coordinate read. Then the operation would be repeated for the other coordinate.

 

How did I do DaveD? :rolleyes:

Share this post


Link to post

68-eldo -- That was the way it was done early on. It didn't take too long for the Cartographic unit to acquire a digitizing tables. Using this method they would tape the map to the table, index the four corners and provide the computer with the corner coordinates (latitude and longitude). By moving the plotting pen anywhere within the map you could obtain the coordinates of the object. The majority of marks were plotted in this manner. As kayakbird noted previously, if roads had been moved since the marks were originally set this would induce a significant error. I had an experience like that once on a field project in New Mexico back in the mid-70's. I found a string of bench marks offset from the road by several hundred feet, yet the descriptions clearly stated that they were within a few feet of the road centerline.

Share this post


Link to post

Thanks DaveD, that makes sense that the later work was done with a digitizing table. I seem to see things like digitizing tables as new stuff. :-)

Share this post


Link to post
I found a string of bench marks offset from the road by several hundred feet, yet the descriptions clearly stated that they were within a few feet of the road centerline.

 

We would encounter that quite often in the the UP of Michigan. Most control was set in 1948 and most all roads have been rebuilt, grades changed (cuts and fills) and many realigned since then with bad curves removed etc. People read the descriptions and it says 75 ft from c/l of US-41 but fail to realize US-41 was realigned and the current highway is maybe 1/4 mile off in that particular location. Found many marks like this even ones that NGS failed to locate in a 1985 level run across the da UP. Working for the MDOT we had the plans and R.O.W Maps to go from as well as techs on the crews who recall the roadwork and could easily determine the old roads location and weather is was obliterated or still in use as another name. Often that just call it Old US-41 for instance. I know of one stretch of US2 that the grade was lifted some 10 feet from 1948 and other areas were the hills were cut to flatten the road.

 

I found a lot of marks the were listed as not found just because the looker had no idea the location he was looking in was not the same as in 1948. NGS was usually good at finding them but I suppose like our crews had to rely on seasonal workers for a lot of work. The recon crew I met in 1996 was just tech and a summer intern and the level crew also has 2 seasonal's and 2 tech's.

 

e.g.

RK0395 HISTORY - Date Condition Report By

RK0395 HISTORY - 1975 MONUMENTED NOS

RK0395 HISTORY - 19970721 MARK NOT FOUND LOCENG

RK0395 HISTORY - 19981216 MARK NOT FOUND MIDT

RK0395

RK0395 STATION DESCRIPTION

RK0395

RK0395'DESCRIBED BY NATIONAL OCEAN SERVICE 1975

RK0395'2.9 MI NORTH FROM K I SAWYER AFB.

RK0395'2.9 MILES NORTH ALONG CO. RD. 553 FROM THE MAIN ENTRANCE TO K.I.

RK0395'SAWYER AFB, 71 FT. WEST OF UTILITY POLE NUMBER 7131 ON EAST SIDE

RK0395'OF HIGHWAY, 50 FT. SOUTH OF CENTERLINE OF DIRT FIRE LANE WHICH RUNS

RK0395'IN A NORTHWESTERLY DIRECTION, 45.3 FT. NORTH CENTERLINE OF EAST WEST

RK0395'FIRE LANE, 43 FT. WEST CENTERLINE OF CO. RD. 553, 21 FT. EAST OF

RK0395'4 INCH PINE, 1 FT. BELOW GROUND SURFACE.

RK0395

RK0395 STATION RECOVERY (1997)

RK0395

RK0395'RECOVERY NOTE BY LOCAL ENGINEER (INDIVIDUAL OR FIRM) 1997 (GNV)

RK0395'SEARCHED EXTENSIVELY FOR THIS STATION BUT UNABLE TO FIND. I THINK

RK0395'THIS MONUMENT WAS PROBABLY TORN OUT DURING NEW COUNTY ROAD

RK0395'CONSTRUCTION. THE ES 7 BM STICKS UP ABOVE GROUND AND IS NOW ON A

RK0395'SMALL RISE - REMAINDER OF OLD ROAD BED LOOKS TO HAVE BEEN EXECAVATED

RK0395'OUT FOR NEW ROAD BED FILL.

RK0395

RK0395 STATION RECOVERY (1998)

RK0395

RK0395'RECOVERY NOTE BY MICHIGAN DEPARTMENT OF TRANSPORTATION 1998 (SHN)

RK0395'A THOROUGH SEARCH REVEALED NO EVIDENCE OF THE MARK. MARK IS ASSUMED

RK0395'DESTROYED AND BURIED WHEN THE NEW ROAD WAS MOVED 40 FT WEST.

Edited by Z15

Share this post


Link to post

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...
Sign in to follow this  
Followers 1

×
×
  • Create New...