8

Since the earth is slowing its rotation, and as far as I know, each day is 1 second longer every about 1.5 years, how long was an earth day near the formation of earth (4.5 billion years ago)?

I wouldn't assume to just do 4.5b/1.5 and subtract, because you would think the rate of change is changing itself, as seen here from wikimedia. It is a graphical representation of data from INTERNATIONAL EARTH ROTATION AND REFERENCE SYSTEMS SERVICE. They decide when its time for a leap second (the last one being on Jun 30, 2012) The data can be found here.

Qmechanic
  • 201,751
Cameron Aziz
  • 181
  • I think that BB1 is right that your figures are probably to large by several orders of magnitude. Notice that the green line does not show a steady increase (which would be expected if your figures wee right), but is instead dominated by noise. – dmckee --- ex-moderator kitten Jul 06 '12 at 04:41
  • The green line is a moving average of change, and is always positive (which it doesnt have to be, for days to get longer over time). At times, the slowing slows (as seen in the early 2000's) and was exceptionally fast in the 70's. If you notice the red line is generally increasing over time, which is the time of day lengthening. From the early 70s to today, they day IS 25 seconds longer. – Cameron Aziz Jul 06 '12 at 04:48
  • 1
    @Cameron Aziz Are you sure this info is accurate? Where is this from? – MadScientist Jul 06 '12 at 04:54
  • updated question. – Cameron Aziz Jul 06 '12 at 05:00
  • @Cameron Aziz, There isn't enough information here to give a much better estimate. Do you have more? Is it safe to assume that the fluctuations in the rate is random? Or does the rate seem to be changing slowly in one direction (because even if it is changing in one direction over the last 30 years- 30 years is a very short sample relatively speaking). That requires statistical analysis and more data. – MadScientist Jul 06 '12 at 05:00
  • Yes the data is correct for these 30ish years, it is also very noisy, and if we were to assume that rate over historical periods we get a total delta on order of 22 minutes over 2000 years, which would put solar eclipse tracks 5 degrees of arc (approximately 600 kilometers) away from their positions if the day remained constant, which should show up on the methodology in used in the link that BB1 gives. – dmckee --- ex-moderator kitten Jul 06 '12 at 05:02
  • @BB1 well yes there is more data, but no, I do not have anything else. – Cameron Aziz Jul 06 '12 at 05:04
  • @dmckee I think the issue is we cant assume / extrapolate the data over 2000 years or 4 billion years. – Cameron Aziz Jul 06 '12 at 05:05
  • @Cameron Aziz, Is there some explanation for the shifts in rate of change? Because if there is then maybe you can use physics to account for a minimum and maximum amount of shift. – MadScientist Jul 06 '12 at 05:08
  • I believe it has to do with tidal acceleration, the moon is essentially pulling back on the earth. Eventually one side of the earth will be able to see the moon and the other side will never have a moon, like the "dark side of the moon never sees earth" – Cameron Aziz Jul 06 '12 at 05:12
  • @Cameron, If, based on that, you could make an argument for when the rate reaches its maximum and mimimum value, then you could construct a good estimation. Also, If you could show that, for physical reasons, the rate is generally decreasing linearly (over larger time scales), then you could apply a linear approximation – MadScientist Jul 06 '12 at 05:36

3 Answers3

12

... each day is 1 second longer every about 1.5 years

That figure is way off.

According to this Scientific American article, the Earth's rotation rate just after the collision that formed the Moon was about once every 6 hours. At that time, the Moon would have been about 25,000 kilometers away. The tidal effect of the Moon is the major reason the day has been lengthening, and the Moon's orbit has been widening.

The collision is believed to have taken place about 4.5 billion years ago, not long after the formation of the proto-Earth.

There are still some open questions about the impact hypothesis (see the linked Wikipedia article), so this is uncertain.

I strongly suspect that the impact would have erased any information about the Earth's rotation rate before the impact. (It might be possible to estimate the pre-collision rotation rate by modelling the initial formation of the Earth; I don't know whether there's been any research in this area.)

Keith Thompson
  • 7,643
8

Actually we only gain 1.3 milliseconds every 96-100 years, not 1 second every 1.5 years! :) the shortest known Earth day was 6 hours and the longest is 24 hours & 2.5 milliseconds (today's current day), in 1820 the day was exactly 24 hours, but since it's been nearly 200 years we've gained 2.5 milliseconds to our day. So the days get longer just very shortly, I believe it'll be 15 minutes longer in 50 million years.

darknessra
  • 89
  • 1
    this is completely false. – Cameron Aziz Sep 02 '14 at 05:20
  • 4
    While not quite accurate, darknessra's answer is close. According to http://tycho.usno.navy.mil/leapsec.html the rate of deceleration of the Earth's rotation is about 1.06 milliseconds per day per century. If the rate of deceleration was constant, which it is not, it would take (15 min * 60 s/min * 1000 ms/s) / (2 ms / 188 y) = 84.6 million years for the day to lengthen by 15 minutes.
    It would be incorrect to compare the deceleration rate, conventionally given in milliseconds per day per century (1/time), with the clock drift rate, given in seconds per year (dimensionless).     – Clement Cherlin Mar 09 '15 at 07:05
  • 1
    These information are correct since I came with the exact result when I derived a model for the length of the day over the entire Earth's history that is available at https://www.researchgate.net/publication/26541355_The_Length_of_the_Day_A_Cosmological_Perspective – Maxwell Nov 18 '19 at 21:15
-1

Shouldn't each day be one second longer every 1.5 years, if the Earth is rotating slower? Assuming your info is accurate, (1 sec)/(1.5 years) * (4.5 billion years) = The Number of Seconds Shorter The Day Was 4.5 billion years ago. Subtract that from the number of seconds in a day now. I would convert the final answer to minutes.

But your numbers are wrong and lead to an absurd result. "The average day has grown longer by between 15 millionths and 25 millionths of a second every year" http://www.popsci.com/jessica-cheng/article/2008-09/ive-heard-earths-rotation-slowing-how-long-until-days-last-25-hours

So, do the same calculation with these numbers instead.

MadScientist
  • 1,505
  • 1
    shorter = typo. also, as I stated, that assumption is inheritly false. The rate of change isn't constant. I have added a link to the rate of change for the last 50 years. – Cameron Aziz Jul 06 '12 at 04:02
  • 1
    The numbers don't make sense btw. A day cannot be more than 606024 seconds shorter. This calculation implies that a day in the past was 3 billion seconds shorter than a day today (days were negative length?). – MadScientist Jul 06 '12 at 04:08
  • 2
    precisely why its not linear... – Cameron Aziz Jul 06 '12 at 04:10
  • 1
    Although the rate is changing, you can take the average and this should give you a good estimate: "The average day has grown longer by between 15 millionths and 25 millionths of a second every year" (http://www.popsci.com/jessica-cheng/article/2008-09/ive-heard-earths-rotation-slowing-how-long-until-days-last-25-hours) – MadScientist Jul 06 '12 at 04:19
  • You have failed to understand the question. The value given is the rate of slowing in this epoch, not some permanently fixed global value. – dmckee --- ex-moderator kitten Jul 06 '12 at 04:20
  • 1
    The effect is inherently a tidal one (the moon is pushed further away at the cost of some of the Earth's angular momentum), and as such was larger in the distant past when the moon was closer. The article you link takes data (nice methodology, BTW) for all of 2500 years which is a drop in the bucket. – dmckee --- ex-moderator kitten Jul 06 '12 at 04:24
  • 1
    But, he said "each day is 1 second longer every 1.5 years" and that is false. Are you saying that rate was accurate at some time? – MadScientist Jul 06 '12 at 04:27
  • 1
    If the rate of change is changing, the reasonable thing to do would be to take the average rate of change. There could be physical reasons to weight some rates higher. – MadScientist Jul 06 '12 at 04:31
  • 1
    The data the OP linked show a figure which could be misread as the value he quotes. I believe I've read that the day was shorter by on order of one hour during the era of the dinosaurs which suggests a rate of change more like your reference than the OP's. But just noticing that the OP has the wrong figures does not answer the question. – dmckee --- ex-moderator kitten Jul 06 '12 at 04:38
  • 1
    @dmckee There is a lack of data here. This is my best estimate given the amount of data provided. It might be a decent lower bound (assuming the Earth is generally rotating slower with nothing to crazy happening periodically- like huge meteorites hitting it a making it spin significantly faster). How fast do you think the earth was rotating at its fastest in the distant past? If you answer that then the average of that rate with the modern one would probably give a good estimate. – MadScientist Jul 06 '12 at 04:48