Earth Travel On Time, Again

burgburgburg writes "The NY Times has an interesting article about a rather puzzling phenomena: for the fifth year in a row, the Earth's travel through space is right on time. The rate that the Earth travels through space has slowed ever so slightly for millenia. To compensate for this, since 1972, scientists have added a "leap second" at the end of each year. The problem: Since 1999, the Earth has been on time. The recognition of a need for a leap second was an unintended consequence of the invention of the atomic clock. Suggested reasons for the unexpected punctuality: the tides, weather and changes in the Earth's core."

Comments

Do atomic clocks keep perfect time?

[G4from128k]

It seems to me that physicists assume that their atomic clocks keep perfect time. But what if they don't? What if some key physical constants are changing in our neck of the universe. As an engineer I have found that most physical constants aren't (everything is a function of everything, its just an matter of the coefficient). In the case of the atomic clocks, a change of only 32 parts per billion would change the timebase by one second per year. Perhaps a particularly large, long-wavelength gravity wave has stretched spacetime and changed the clocks? Perhaps the four fundamental forces oscillate in undiscovered ways?

IANAP, so perhaps a professional could explain why the atomic clocks must be right -- why a 32 ppb variation in them is impossible (i.e. would manifest itself in other more obvious ways).

Re:Do atomic clocks keep perfect time?

[dublin]

They also have a plan to use pulsars to see just how imperfect the atomic clocks are.

Terrific. So how do they check the pulsars? Or is it just turtles all the way down?

Ultimately, it's *always* turtles all the way down when it comes to measurements. Sooner or later, you have to pick *something* and say, "This is my standard against which all things are measured." Sadly, we have (and can have) no truly absolute metrics of physical quantities against which to measure. Witness continual redefinitions of basic SI units over the past few centuries for a perfect example...

Tha atomic cesium clocks are the standard because they are *believed* to be the most reliable timekeepers we have. But they may well be inconsistent, and subject to many vagaries in their regularity: quantum fluctations, distortions of space-time causing local compressions or expansions of time itself, or Vietnamese hegemony in Southeast Asia - we just don't know.

In fact, we *can't* know, so we pick something, and hope it's good enough, vainly hoping that in this small way, man really can be the measure of all things...

Re:Full Text of Article

[jaxdahl]

Is the Earth's *rotation* slowing or speeding up at all, though?

Let's do some math and see how big of a distance difference a leap second is. Rough estimate of 93 million miles on average for r. 2*pi*r = 584,309,935 miles for the circumference of the orbit -- assuming a circular orbit. 60*60*24*365.25 = 31557600 seconds on average per year. circumeference/seconds ~= 18 miles. Interesting. But hardly seems like much.

Re:All wrong...

[notyou2]

No. God just likes to screw with us.

NO... God just likes to *fuck* with us.

Re:No more Feb 29th?

[TheF00]

Leap year has nothing to do with with leap seconds. Leap seconds are added due due to the slowing of the earths rotational rate. A leap year occurs when an extra day is added because the earth doesn't orbit the sun at exactly 365 days. Its 365.25... so every four years a extra day is added to keep the calender in sync. A leap second however only keeps the clock in sync.

Re:Physics for the rest of us

[Johnny Mnemonic]

You're assuming that the GPS satellites don't orbit at the exact same speed of the spin speed of the Earth; you're assuming that they in fact orbit a little faster than the Earth--that they orbit at a speed that is the exact same as the length of an ideal day.

I don't know which speed they orbit at, but it's not hard to think that they orbit exactly as fast as the Earth spins, so they don't need to be corrected as you mention. I would be interested to know which is true.

Re:It's Obvious

[pyr0]

Small enough to only account for a single second out of an entire year perhaps?