Reading this, I wonder if the melting of the ice caps causes a measurable slow down in the Earth's rotation and an increase in the length of a day by moving mass farther from the Earth's axis of rotation.
Excellent link, thanks! Here are some key points for convenience (although the article's so short that this is practically half of it):
> the rate of rotation varies by up to a millisecond per day [regardless of ice melting]
> Melting land ice ... will change the Earth’s rotation only if the meltwater flows into the oceans
> For example, if the Greenland ice sheet were to completely melt and the meltwater were to completely flow into the oceans, then ... the length of the day [would become] longer ... by about two milliseconds
> Melting sea ice, such as the Arctic ice cap, does not change sea level because the ice displaces its volume and, hence, does not change the Earth’s rotation.
It doesn’t mention what sort of frequency those variations happen at. If the variation is caused primarily by plate tectonics, it might take a year or more to swing between the two extremes.
Small changes in the rotation rate of the Earth (primarily due to seismic activity and the tidal interaction with the Moon, I think) are the reason why sidereal (or Solar) time diverges from time as kept by an atomic clock (the atomic clock has errors too, but they're orders of magnitude smaller).
This is how you get different definitions of "Universal Time":
* UT1 = sidereal time, based on measurements of the angles of distant quasars.
* UTC = atomic clock time, with offsets added in whenever it gets too far (about 1 second) from UT1.
One final fun thing about UTC. Why is it UTC?
* Coordinated Universal Time = CUT ?
* Universal Coordinated Time = UCT ?
* Temps Universel Coordonné = TUC ?
UTC was chosen precisely because it doesn't correspond to either the English or French acronym.
> UTC was chosen precisely because it doesn't correspond to either the English or French acronym.
That's fascinating, I assumed it was a French acronym. "Universale Temps Coordinaté" (I do not speak French, but I have seen Kentucky Fried Chicken signs in Québec!) :)
Regardless of the political process of selection, I am glad things worked out the way they did. The other acronym possibilities are really unpleasant in English.
* Sidereal time is different from solar time; sidereal time measures the Earth's rotation relative to the stars, while solar time measures it relative to the Sun, which moves relative to the stars. Thus, UT1 and sidereal time have different rates.
* Wikipedia's page on day length fluctuations (https://en.wikipedia.org/wiki/Day_length_fluctuations) has a discussion of the influences involved. It also has a handy graph, from which we can see that the average day length is around 2 milliseconds shorter than in the 1970s (hence why we had a leap second every year from 1973-1979, but the last leap second was over four years ago).
* There's also UT2 and UT0. UT2 has a corrections for seasonal variations, while UT0 is derived from raw observations, without correcting for polar motion like UT1. These are rarely used.
I lumped Solar and sidereal time together because they're both essentially dependent on the Earth's rotation. Their rates are different by ~1 day/year, of course.
If you ever want to be bored out of your mind (at whatever level of detail that occurs for you), just ask someone in this field about the definitions of geoid, ellipsoid, etc. etc. and you will have hours (well, maybe minutes) of fun.
My PhD supervisor told a funny story about this to illustrate the phenomenon of learning more, but seemingly knowing less.
First you learn that the Earth is "round". Then you learn about shapes and that the Earth is actually a sphere. Then you learn more shapes and that the Earth isn't actually a sphere but is slightly squashed at the poles: it's spheroid. Then you learn that it is actually more squashed at the top than at the bottom. It is in fact a geoid. So you look up geoid in the dictionary and find that geoid means "Earth shaped".
And for added fun if you look up some of the reference points maintained by NOAA they'll list a position, a date, and a velocity. Due to the relative movement between different parts of the Earth even "fixed" points will be moving so you need to calculate it based off of the velocity and length of time since the position was surveyed. Geodesy is weird.
It's unfortunate and potentially confusing to readers that they've used the Washington Monument and the Flatiron Building as illustrations in the article and suggested that their heights might change. The reference point for measuring a structure's height is its base, not the geoid. Unless the definition of a foot changes the height of these structures will not change.
Curiously, the definition of a foot is changing. Or rather, one of the two definitions of a foot (the U.S. survey foot) is being deprecated in favor of the other (the international foot). The two differ by about two parts per million.
My understanding was that an inch is exactly 2.54cm, and a foot is exactly 12 inches ever since the metrification of US customary units. Is that not correct?
No, the original metric definition of US units in 1894 [0] made 1 yard = 3600/3937 metre (i.e. 1 inch = 100/3937 metre). The US survey foot is 12 of those inches.
The 1 inch = 25.4 mm definition was adopted by the British Standards Institution in 1930 and by the American Standards Association in 1933 (with the name "industrial inch"); in 1959 the US, UK, Canada, Australia, New Zealand and South Africa signed a treaty agreeing on that definition [1] (and also a definition of the pound). The international foot is twelve of those inches, but the US allowed (but recommended against) the continued use of the old foot in surveying.
The foot was defined in terms of meters only in 1959, and the definition was slightly different than prior use. (The US and the Commonwealth previously used slightly different feet.) The survey foot is the definition prior to the definition in terms of meters, and is still widely used in surveying.
I would have expected that the recalibration have big impact in middle of continent and no impact near coasts but strangely it's not the case. How coastline altitude could have been incorrectly calibrated before ?
This caused problems when a bridge was being built between Germany (which uses the North Sea) and Switzerland (which uses the Mediterranean) - someone made a mistake including the known height difference between the two seas:
> That’s because height is only height compared to a reference point — and geodesists...are redefining the reference point, or vertical datum, from which height is derived.
So it isn't getting shorter. We're changing our reference point. Those are very different things!
