Similarly, the border between California and Mexico is not a straight east-west line, but tilts slightly southwest-northeast. This is because the Americans wanted to be sure all of San Diego Bay was American territory.
And in Nebraska. There’s a chunk of Iowa (Carter Lake) in Nebraska now, on the other side of the Missouri.
As I pointed out upthread, this phenomenon happens on the Missouri, Ohio, and even the Wabash. For example, there’s a largish chunk of Kentucky near Evansville on the north side of the Ohio. And there’s two chunks of Indiana on the Illinois side of the Wabash.
One thing that makes this kind of thing more likely is that not all river borders are drawn through the deepest part of the river. Some are drawn on one bank. The best known of these is the Potomac, where the border between Maryland and Virginia is on the Virginia bank. But it’s also on the Ohio, where the borders are along the northern bank. But note it’s the northern bank as of 1793, before any dams raised the height of the river and moved the banks outward.
A third such dislocated river border is the Chattahoochee between Georgia and Alabama, where Georgia owns the entire river. Going by GoogleMaps, there’s a few slivers of Georgia on the west bank of the river.
OK, so after further research, I’ll correct my former statement and say that the border was actually drawn in 1906 - 20th Century, but still pre-WW1 and pre-Ottoman breakup.
Another oddity of surveying: Zone of Death (Yellowstone) - Wikipedia
That’s not a surveying issue. It’s poorly written federal law.
Yeah, the fact that there’s a little bit of Idaho (and Montana) in Yellowstone wasn’t an accident, but a deliberate decision to get more states in favor of creating the park.
Sure, but that doesn’t mean that the Wyoming District Court should get jurisdiciton over parts of Montana and Idaho, contrary to the Constitution.
I agree with Saint_Cad that it’s a statutory error, not a constitutional loophole. Congress has not complied with the Constitution, not that there’s a problem in the Constitution.
Also, Yellowstone is exclusively Federal jurisdiction meaning you can’t even have state authority there. Sometimes I fantasize about doing something there minor, just bad enough to warrant a jury trial, to test it. The only case I’m familiar with, the defendant wimped out and plead out and as part of the agreement agreed not to use the 6th Amendment as a defense.
It’s not contrary to the Constitution. There is no corresponding Zone of Death in Montana because there are enough people there to form a jury if need be.
I didn’t say it was an accident, but it is an oddity.
I understand what you are saying, but I have always assumed that it is the angle between a line drawn to the center and a line from the equator drawn to the equator. That would also be the angle determined by astronomical observations. When teams of French geographers went to Ecuador and to northern Norway to document the oblateness by measuring the length of a degree, I assume that’s the definition they were using.
No, the astronomical measurement would be from the normal to the surface, not from the line to the center.
Oh, is that because they use a plumb line (which is normal to the surface) to set up the sextant? In the northern hemisphere that will point to some place south of the axis, owing to oblateness.
Or the horizon, or an artificial horizon. I’m not sure what method is used most nowadays, but all three can be used, and historically have been. I don’t think there’s any way, with a purely local experiment, to detect the direction towards the center.
Cite? Yellowstone was established in 1872, when the “states” involved were sparsely populated territories with no voting representation in Congress. I’ve never heard that the geologist who proposed the original boundaries, Ferdinand Hayden, was moved by politics in doing so. He likely wasn’t even aware that his boundaries slopped over into Idaho; that wasn’t determined until some years later.
Congress added extra bits of Montana (the squiggles in the northwest and northeast) to the park in 1929 and 1932.
And which centre of the Earth did you have in mind? Just this question exemplifies the grief inherent in mapping when you want high resolution.
The Earth isn’t just oblate, but also slightly pear shaped. And it isn’t homogeneous. So plumb lines are not plumb. Any mapping dependent on the gravitational field has an intrinsic limit to its usefulness.
So you can define all sorts of centres, various geometric, gravitational, all different. Which is partly why we end up with the WSG-84 spheroid defined by a series of spherical harmonics. And even it gets tweaked. That does at least give us a reference body that everyone can agree to work with. But it wasn’t a thing back when borders were drawn.
Here in Oz we have a really good mapping anomaly. If you look at a map you would get the impression that the Eastern border of South Australia is a straight line. It should run from Queensland at the North edge straight down the 141 degree line of longitude. It is intersected from the East by the Queensland New South Wales border, which is a straight line, and further South by the Victoria border, which is a river, and thence down to the bottom of the continent. Except that it doesn’t. There was a mistake in surveying the location of that part of the boarder between SA and Victoria. Out by a few miles, and too far West. Huge disputes that went right to the Privy Council. But the mistake stands. So the border does not just have a kink in it, because the Victoria - NSW border is a river, there is a tiny river delineated diagonal wiggle that forms the edge of the kink, and it isn’t pretty, essentially forming a tiny island marooned from the state it belongs in.
Right; I was answering in the context of a uniform ellipsoid, which does at least still have a single unambiguous center. But you’re right, of course, that the real Earth is more complicated, in multiple ways (both in shape and in varied density).
I think that the official definition of latitude is the angle from the plumb-line direction (i.e., the local normal to the geoid) to the equatorial plane. And the equatorial plane is probably defined either as perpendicular to the minor axis of the moment of inertia, or as perpendicular to the angular momentum vector (these two would be extremely close together, but probably not exactly the same).
The U.S. Naval Observatory seems to imply
that one should keep track of the difference between “astronomical coordinates” [“the longitude and latitude of the point on Earth relative to the geoid. These coordinates are influenced by local gravity anomalies.”] and terrestrial latitude and longitude. E.g., they mention distinct astronomical, geocentric, and geodetic zeniths.
That’s not the officialest definition. The latitude on a map (geodetic latitude, it’s called) doesn’t use the local normal to the geoid. Why not?
We start by asking: what’s the point of knowing the latitude and longitude at a point? For most of us, the only point is: if we know the lat-lon at this point, we can calculate the distance and direction from this point to some other point whose lat-lon we already know. And that calculated distance is supposed to be close to correct – if we have the lat-lons at point A and point B, and calculate the distance, and then measure the actual straight-line distance (correcting for the slope of the line, and for altitude, etc) the two distances should agree. In the US, calculated distances between third-order survey stations are supposed to be correct to something like 1 in 10000.
That won’t work if your lat-lons are based on the local vertical. People have the idea that the traditional way to determine latitude-longitude was by sighting the stars, meaning, using the local vertical. The so-called astronomical lat-lons that you get that way give you calculated distance that are much less accurate than we want. In an extreme case (the Big Island, in Hawaii) the calculated east-west length of the island will be off by more than a mile.
So what to do instead? Traditionally, the answer was triangulation – measuring distances and angles on Earth, and trying to calculate the corresponding distances and angles on the spheroid that we’re using to approximate the geoid. Choosing that spheroid comes first, and that’s where we need the stars. One way or another, the guys in charge in the US decided to use the Clarke 1866 spheroid for surveying in the US (and, later, the rest of North America). For all we know the Clarke 1866 might still be the best fit to the geoid in North America, but they eventually decided that using the spheroid that best fit the whole world was better, even though it didn’t fit the US as well as the Clarke.
Using the new spheroid, they had to recalculate the whole lat-lon system, to get calculated distance to still agree with actual. That’s why California got “moved” about 100 meters west when the US switched from NAD27 to NAD83. As I recall, the calculated distance across the US in NAD83 is 8 meters different than in NAD27, just because triangulation can’t do better than that.
Sure, but that just pushes the question back to what formula you’re using to find distances, given latitudes and longitudes. For any method for defining latitude and longitude, if you know the full definition (including all of the spherical harmonics defining the reference geoid, and so on), you can derive a formula (possibly a very complicated formula, if you’re using a very detailed model).