Physicists Unveil World's Most Precise Clock
technologyreview.com
technologyreview.com
I remember when the current Primary Standard, NIST-F1, was meeting the press in the late 90s. I was in undergrad and one of their junior guys came through interviewing for a faculty job. His lecture remains one of the most memorable hours of my life. Probably helped by the fact that I had just heard about the F1 in Scientific American a few weeks prior.
It is unclear that this particular clock will take over as the Primary Standard, whose data is compared to a competing project in France and coordinated with the astronomical measurements at the US Naval Observatory to construct the Universal Coordinated Time standard.
That is then what the USNO transmits to coordinate the GPS satellites, radio services, etc.
All four of the NIST Nobel luareates won for the work they've done in laser cooling for the atomic clocks.
This seems like a pretty big deal! Perhaps not for measuring height, but measuring gravity field with such a precision can have more applications
With enough precise enough clocks, I imagine it might be possible to do the same by measuring timing differences. Funny thought, really.
(Also, the inverse problem for the Laplacian is hard. Hiding a dense object inside a sphere of lower density (in a background of uniform moderate density) is easy, and fully cloaks against external gravitational detection.)
> Based on current models of astronomical events, and the predictions of the general theory of relativity, gravitational waves that originate tens of millions of light years from Earth are expected to distort the 4 kilometer mirror spacing by about 10−18 m, less than one-thousandth the charge diameter of a proton.
Longer answer: there might be a clever way to use clocks to pick off gravitational waves, particularly at low frequency, but I doubt that they'll be competitive with pulsar timing/LISA.
There are proposals to use atom interferometers for gravitational wave detection. The community is quite divided on the chances for success (I'm in the pessimistic camp at the moment). Barring any surprises, an atom interferometer is much more likely than a timing-based measurement to see any gravitational wave signal. On the LIGO front, so long as the source estimates for gravitational wave signals are correct and as long as Advanced LIGO hits its sensitivity goals, a LIGO GW detection is imminent.
There's always room for cleverness.
Also, is it obvious how one could use an ultra precise clock to see evidence for GWs? Such clock can reveal spatial gradients in the rate of ticks (and so reveal things like local mass densities), but GWs should be spatially homogeneous, no?
Thanks for any help!
I haven't studied the atom interferometer GW stuff in great detail, but the measurements look challenging from a feasibility and systematic perspective. Any one objection can be addressed, but there are a lot of them, making it challenging in the aggregate. Continued work on the ground in the atom interferometer field over longer baselines will explore both. If that work can show that it works as proposed, then pessimism will turn to optimism; experiment is the arbiter of truth.
As for the second point, regarding clocks, my initial response of "no" came from exactly your argument. I hesitated when considering a global array of clocks. For wavelengths comparable to the size of Earth and smaller, there's enough phase difference that perhaps a sufficiently precise clock, read out sufficiently fast, might pick off a signal. Additionally, if there are signals in the micro-to-nanohertz regimes, it might appear as an unexplained fit residual to an overall model of the gravitational potential at any one clock. My GR-fu is insufficient to make reliable estimates of the sizes of these effects (I bet they're small!).
>They say their new clocks can keep time with an unprecedented precision of one part in 10^18.
That seems to a measure of accuracy, not precision.
I can make a clock that gives the time in trillions of years since the Big Bang, and it will be perfectly accurate for the distant future as long as it always returns "0". But that's not a great precision.
I can also make a clock that gives the time in one 10^256th of a second, but it won't be accurate.
Accuracy: getting as close to the target as you can. Precision: hitting the same spot repeatedly.
The paper is correct - these measurements hit the same spot repeatedly. Accuracy in this context is a tricky thing. The SI second is defined by the Cesium standard, but the best Cesium clocks are >100 times less precise. These two clocks are hitting the same tiny spot over and over again somewhere within the fuzzy bullseye that is the Cesium target. The SI standard is not defined as well as the new generation of clocks can measure.
Once the world settles in on an optimal optical clock design (the previous leader has been an Aluminum clock, this one uses Ytterbium), the SI standard will move to the new technology.
The impressive success of these clocks as a measure of gravitational redshift is also an Achilles heel. I'm not sure that anyone really knows how to do reliable time transfer with them yet, as one's depth in Earth's gravitational potential relative to another clock must be known at the centimeter scale. Another factor of 10-100 in precision will require that such metrology will improve by the same factor. It's a challenge.
Earth tides (dirt, not water) are larger than centimeters. With further improved precision, they'll be sensitive to small changes in the locations of local gravitational sources, even big ones at long range. In our gravity lab in Seattle, we see the effects of annual variation in water table. All of these things must be properly characterized in order to compare two clocks that are not co-located.
These are good problems to have, and are inherent to measuring time at this precision, but they're very daunting. Not insurmountable, but often frustrating.
Another reminder the amount of information hitting us every moment is absolutely astounding. That we're building devices which can collect & present that information is equally astounding.
>And physicists rely on clocks to test the fundamental laws of the universe to ever deeper levels.
What will this instrument allow us to further discover, confirm, or understand?
This project also reminds me of the 10,000 Year Clock-- http://longnow.org/clock/
In the end, the real answer to your question might well be something that we currently know nothing about. In other words: "If we already knew the answer to that question, we wouldn't need the clock."