I realize that there is always the risk of positive ID of far off points. To that end usually there are more than one tower/tank/structure to choose from. So, I always turn the angle between them. Then you know which are which.
As for a training mission, there is none better than a disc and couple of towers. And you can pick the disc that works since it's not a project. AM radio towers are always tall and skinny, so they aren't wider than a couple of seconds. Churches are good because you can usually visit the structure, and common on the east coast. you could even ask the vicar about the structure. The NGS sheet for the tank/tower/etc. will generally have good description, and maybe even a few recovery notes over the years.
The inverse from even a 2nd order disc to a 3rd intersection is still a good azimuth assuming a few miles away. Especially for a beginner.
Also, there is another tack. Set up anywhere convenient and take a solar astro azimuth to a landmark. Figure the azimuth. Then take a Polaris shot to the landmark and compare. Pick somewhere where you can see a tower light, and make it your own. So, solar and Polaris. Get about the same answer and you must be doing it right.
((I've seen a lot of descriptions to a lot of tanks. I've never seen a tangent of a tank. To be an intersection station it has to be center, otherwise how would they position the tank? Maybe a tangent was in lieu of an azimuth mark.))
.
> >It's a lot easier on land than at sea
>
> That's for sure, but then the requirements get tightened on land. I expect you were pretty happy with 1 arc minute at sea, but on land people are going to expect 5 or 10 arc seconds.
Well, yes and no. If you're in the middle of no where in the South Atlantic Ocean, then yes, that'd be great...about a nautical mile of latitude...
But, if it's the last sight you take prior to making landfall at, say Lizard Point, and your dead reckoning leads you to believe you're some 5 miles off the coast, but your sight puts you at 49° 57' 19.62" N x 5 ° 12' 32.59 W, well, then you have a big problem. 10 arc seconds may well be the difference between success and failure. It's not possible to achieve that accuracy with celestial observation alone, of course, so, as on land the "sea surveyor" will use other data to confirm (or challenge) the information you have.
Reading several discussions on SPC, UTM, etc. and tying local surveys into the bigger world, in the attempt to establish the absolute location of a monument vs. its location relative only to something else on the plat, I find the parallels interesting:
At sea, when you approach land, you DO want to know your absolute position as well as the absolute position of objects and locations outside of your "local coordinate system".
Looking back at some of those situations now, I realize I'd give a lot to have had a handheld GPS looking at 8-10 birds with WAAS, and a 9' circle of error! I really don't know what we were thinking back then!
It's a lot easier on land than at sea
Back when I was doing a lot of azimuths, I would often use a planet, usually Jupiter if it was up. I have also used the moon, but there are some extra comps to do so that the obs are corrected to the surface point (ephemerides give a geocentric position of the object, which doesn't matter for stars but can matter for planets and definitely matters for the moon. My paper I wrote years ago has a lunar example worked up, but I think it might have an error or misprint in it, I will have to check. But the steps shown are what is needed.
>I've never seen a tangent of a tank
Nor have I ever seen that stated on the NGS data sheets. For smokestacks and standpipes, I would expect them to use the mean of the right and left edges (tangent lines). For water tanks there is almost always some feature on top to be used (and hope it hasn't been moved).
For the most part, using a landmark for azimuth on paying work is about obsolete. GPS has done that in. It's just the art of the field work, and then there is the academics of previous methods. The understanding that a good Polaris shot, properly attended to with Laplace, DUT, and getting what theoretically should be the same to a practical degree of precision.
ON a paying job in the late 80s, the client disallowed gps for azimuth. It was experimental and there were so few satellites that there was only 4 hours of usable reception. And dual frequency equipment wasn't commercially available. The old 5 channel single frequency Trimble 4000SL. So, we shot Polaris to 1st order, 2nd class specs. Multiple nights, 16 sets, T3. The Laplace correction was 9.5 seconds. I observed a 1st order mark less than 2 miles away (too close for 1st order). the corrected azimuth by Polaris was only 0.4 seconds from inverse. Several of the county surveyors that I knew, that didn't trust GPS, said that couldn't even be done reliably. However, it was demonstrated on the multiple nights.
A good astro azimuth is something that fewer, and fewer surveyors have done, or even anticipate that they ever will. The main reason for doing the math is to take the PLS test. And to walk in the foot steps of previous surveyors.
What I was trying to say is that rather than a light or vent, the center of tank was often the position. This was obtained by observing tangents. The mean of the left and right was used as the angle to the object.
Remember as well that these intersection stations were no part of the adjustment of data for the more modern versions of NAD 83. At one time they were to be dropped from the NGS database.
Of course, if enough redundant observations to other intersection stations are made it is possible to identify "bad" ones.
HTH,
DMM