What I loved about surveying in Oklahoma was that if you had a ‘God Rock’* to start with, no survey coming from a different benchmark could over rule it.
In some cases the surveyors made errors. The Texas-New Mexico border (the north-south part) is not a straight line. This is more likely with N-S lines as an accurate clock is required to determine longitude, and segments not sighted from one reference will be offset by the clock error.
In cities, I have seen benchmarks as 4" brass/bronze discs embedded in sidewalks.
Until GPS toys, I doubt anyone recognized Lat/Long or elevation.
If you see a small “X” chiseled in a rock or pavement with the notation “TBM” - that was a Temporary Bench Mark, used at some point for a (usually complicated) survey - the crew had tracked position and elevation from some “kinda-sorta” at least, accurate point and were going to be around that spot for a while. By finding something hard and stationary, they could mark it and record in their notes its location and elevation. All subsequent (laying out a cloverleaf or such) work could then be thrown from that mark. Saves an incredible amount of time if you spent an entire day getting from the nearest good point to the area of interest.
You can easily lay out the Nazca lines with just a long string and measuring tape.
The Mason-Dixon line wasn’t just a survey line. You would think that it would be fairly easy to lay out lines on a map so that they don’t overlap, but they did an amazingly poor job of that in the early U.S. When Maryland and Pennsylvania got their charters, they both originally chose the 40th parallel as their border. But then when they did some more careful measuring, William Penn found out that this put their border north of Philadelphia, where Penn had already planned to build his capital city. Penn claimed the new border as just south of Philly, but Lord Baltimore kept with the original claim of the 40th parallel. This meant that a very long and thin stretch of land was claimed by both states.
For a long time, no one really cared much. It was basically an argument over lines on a map for land that was out in the wilderness where no one lived. But then folks started settling in those lands, with some getting their charter from Maryland and some from Pennsylvania. Some scuffles broke out (like a bunch of guys with rifles storming the courthouse and refusing to leave until the town pledged to belong to their state). Finally, Maryland and Pennsylvania sat down and settled the differences.
They didn’t just trust the local yahoos to draw the line on the ground, though. Instead, they sent for two respected professional astronomers and surveyors, by the names of Charles Mason and Jeremiah Dixon, all the way from London.
Note that these guys were astronomers in addition to being surveyors. They were doing some pretty sophisticated techniques using fairly simple tools.
As for the actual tools they used, I found this:
From here:
http://www.asce.org/People-and-Projects/Projects/Landmarks/Mason-Dixon-Line/
A navigator’s quadrant alone will get you the latitude. For longitude, you need to combine the quadrant with the accurate astronomical clock.
I don’t think Mason and Dixon were measuring anything. They were merely running and marking a true (celestial) east-west line.
The surveying job that really impresses me is Lewis and Clark Caverns, MT. The caverns have one natural entrance, where people go in. At some point, they decided they wanted a separate exit, so tourists wouldn’t have to turn around and come back out the way they came in. To do this, they had to dig a tunnel through a few hundred feet of solid rock. And they had to dig from the outside in, because if they dug from the inside out they’d have no way to dispose of the rubble. So they had to accurately survey the three-dimensional location of their target point, through a long network of twisty, turny passages, then aim the tunnel at that point. I don’t know how many lines of sight it takes to get from the natural entrance to the exit room, but it has to be hundreds at least, and in some places the traversable cavern is only a meter or so across.
That made my day.
What an interesting problem…
Modern surveying instruments are definitely up to the job (theodolites can do within an arc second of accuracy), but it seems insanely difficult to get it all right. I’d want to have two different teams do the surveying job independently, two or three times, before starting digging.
Is there any way they could have brought some compact boring machinery into the cave, drilling a small borehole straight up to mark the spot?
How about if they were to use some sort of fancy scientific way of sending a vibration through the chamber ceiling and using sensitive seismic instruments on the surface to get them close enough to start digging?
Could a proper geologist go in and look at the rock formations and just know from the formations where the spot directly over the chamber was? (e.g. “See that pink granite flow over there? It is visible on the surface right over our heads.”)
Or would a surveyer just laugh at all of that and say “Why bother? We do this all the time. We’ll get it spot on in one hour.”
Oh, and I should add that it wasn’t done using modern surveying instruments. The exit tunnel was blasted in 1935, by the CCC. And they certainly didn’t drill a vertical borehole… They could conceivably have drilled a horizontal one along the path of the full-sized tunnel, but they say they didn’t, and that would still have been 500 feet of rock to drill through.
I hadn’t considered the possibility of using seismic sounding to locate the room instead of surveying it, but I don’t think that would have been possible: Never mind the difficulty of doing the math without a computer, I’m pretty sure the necessary math didn’t even exist at that time.
I don’t know the details of how they did the Mason-Dixon, or indeed historical american surveys but here’s a bit of my conjecture on how they did it. Several of the elements of it have been touched on already.
The first problem was to decide where the boundary was defined from. From Wikipedia, it seems like the definition decided on was “the line of latitude 15 miles south of the southernmost house in Philadelphia.” Once the house was decided on, finding a point 15 miles to the south of it would have been a relatively easy survey that could have been done through a few methods available at the time. So they’d have surveyed south and marked the 15mi south point with some sort of block.
The next task would have been to work out the latitude and longitude of the block. This would have been done using observations to the stars, a relatively new technique at the time. Wouldn’t have been too different from navigation of a ship, but the need for higher accuracy would mean that they’d have to use more stable instruments and do multiple measurements so that a single error didn’t throw the whole lot off.
Having determined ‘where they were’ they would have also oriented their instruments using the stars, aligned the telescope of their instrument due west, and sent their assistants to cut a line through the wilderness so that they could get a line of sight to the next point they wanted to mark. They would have had to have been very careful due to the long distance, and would have had to adjust the way the instrument was facing due to the curvature of the earth, etc.
Because there was no fixed point at the other end to check onto , they would have had to ‘daisy chain’ a series of segments, making sure each one was correct before continuing. (and indeed, the wiki article notes the line is actually a series of segments).
For each segment, they would have required a way of checking that they had got things right, and they must have had two ways of doing that - either they would have surveyed back to the start of the segment and compared the position calculated through the survey work to the position of the start of segment, or they could have surveyed from the start of segment to the next major marker and then check their position astronomically. Certainly, when they got to the end point, they would have used astro observations to check their work.
For the minor markers, because the distances were smaller, they might have got away with simply putting them directly in line between the major markers, simply set up on one of the markers, align the telescope with the next one, chain out the distance to the one you want to put in, and get the guy putting in the new marker to walk north or south until he gets right on the line that you are looking down. (but doing it all very carefully! Fucking up a state boundary line is not good for future business!)
Because all this is a hassle (seems like Mason and Dixon took about 5 years to do it?) between about 100 and 200 years ago most major powers started work on various survey systems where the local surveyors could start from a nearby benchmark instead of calculating their position the hard way. The Public lands system and the Great trigonometrical survey already mentioned are examples of the two major approaches to this.
Local surveys used many of the techniques already mentioned, they used more theodolite and chain and less astro and trig. The way properties used to be surveyed used to follow a more rigid pattern because that meant less computing trig by hand. They used to survey directly along boundary lines more because there were less immovable obstacles (You can’t rip down someone’s fence to see along the line anymore) and less/easier field calculations involved.
Tunnel surveying is relatively simple in principle - run a series of bearings and distances between the end points of the tunnel, and then calculate a missing line between the end points. once you’ve worked out the bearing between them (and the vertical (zenith) angle), line up your telescope with the missing line, get your assistant to move onto the crosshairs and mark the spot to start drilling, then tell them ‘Dig here!’ Repeat until one end of the tunnel hits the other, or until you die of boredom :smack: (Not that I’ve done any underground surveys, I believe there are a few other factors to do with light refraction to consider)
In the site survey any benchmarks that will be disturbed will be known. I believe the process is to let the owners know, place new benchmarks, and file the information with the appropriate agency, State/County, etc.
Yes, but in this case, one of the endpoints of the tunnel is inside of a mountain, and the series of bearings and distances to get to that endpoint must run through a twisty turny cave network, in much of which a wide baseline is impossible. I agree though, that once you find the coordinates of that endpoint, it’s simple and easy to aim the tunnel towards it.
The more I think about the more I wonder if there was some geological feature that helped them find where to dig. Regardless of how hey did it, I guess mining engineers know all about this kind of challenge, and there would have been plenty of good ones around at the time.
Isn’t that a form of dead reckoning? It sounds like the way a submarine navigates. I submarine knows where it is now because it knows where it has been before.
Maybe related to surveying:
I’ve noticed on construction sites that there will be three stakes arranged in a right triangle. These stakes will be located on the construction property, but usually quite a distance from the actual building going up. Are these used as some sort of reference to put the building in the right place?
I found a schematic of the caves here. There are some pretty gnarly looking constrictions there but I think it would be do-able. Possibly, rather than holding their chains level, they measured the diagonal distance and the slope of the chain. I don’t have much practical experience with survey chains. 
Lots of turns would mean lots of unevenly spaced setups, but that doesn’t mean it couldn’t have been done with careful observations. (Possibly multiple times)
It sort of is. The big difference is your survey has to end by checking onto a known point, otherwise you can’t determine how big or small your error is. In the case of the caves, that would mean surveying in and then back out to the start point.
Sorry, missed the edit window.
I’m not sure exactly how you mean they’re arranged, but if you mean three separate stakes hammered straight up and down, forming the corners of a small triangle on the ground, that’s quite a common way of marking a reference mark or datum point so that no-one dumps a pile of construction material there or parks on top of it.
Other patterns could indicate some sort of profile marker for the building. Builders often get the corners marked with an offset so that the marks aren’t destroyed prematurely.
If you’re in Jerusalem, take a walk through Hezekiah’s tunnel. It’s south of the Temple Mount in the City of David. The tunnel is several hundred feet long, was dug into the rock about 600BC to route a spring from outside the ancient city’s walls into the Pool of Siloam. It’s a fun walk, but in some places the water is over your knees.
Archaeologists figure they started from each end using dead reckoning and the survey tools of the time. In the middle, it does a few turns where they figure the workers at each end met up by listening for the hammering from the other side, so the surveying wasn’t perfect.
OTOH, as people remember from geometry class if they were paying attention, the Egyptians developed surveying to an art to re-plot fields etc. after the Nile’s annual floods. From this, the Pyramids for example are laid out almost perfectly aligned north-south. It’s amazing what you can do with simple tools - line of sight and poles, a rope acting as a compass, water in troughs for a level…
