Home >Opinion >Columns >A day in the life of a familiar planet

One last exploration of our days here on Earth and I’ll stop. Well, for now.

What actually counts as a day, on our planet? 24 hours, you say. That’s right, but it’s really an approximation, not a definition. The way we define a day is by the rotation of the Earth. A day is the time it takes to spin on its axis exactly once. That’s very close to 24 hours, which is why we use that number. But it’s worth remembering two things about that: yes, it isn’t exactly 24 hours, and the length of the day changes all the time. That is, the Earth spins faster sometimes, slower some other times. Faster rotation gives us a shorter day; slower rotation makes it longer.

By “day" here, incidentally, I don’t necessarily refer to the stretch of clock time between sunrise and sunset, meaning the daylight hours. I mean instead what I mentioned briefly in my last column: the time from one solar noon to the next. That’s close to 24 hours. And naturally, that length of time varies with the speed of the planet’s rotation.

All of which is not to suggest that we’re going to see dramatically shorter or longer days, nor that we’ve seen them already. You almost certainly have never noticed when there has been a change in the length of our day; nor will you notice if it happens in the future. Why am I so sure? Because these changes we’re talking about are by an amount of the order of a thousandth of a second each time. I defy you to take note of such a change with any means available to you.

So how do we measure the length of a day, and then how it changes by these minuscule amounts? For a first pass, of course, we define a day as the time interval between two identical happenings: two successive sunrises or sunsets, for example. If you see two such, you measure the time between them and that’s how long your day is. Still, to do this we need to be at some location with a clear view of the horizon to the east or the west. For many of us, that’s not possible. So perhaps a better metric might be the time between two successive solar noons—that is, from when the sun is at its highest in the sky today to when it is at its highest tomorrow. Though even that has its limitations. Perhaps it’s an overcast day? And think about it, how do you pinpoint that precise moment when the sun is at its highest?

But let’s say you do have a view of the horizons on a perfectly clear day; let’s say you can tell with precision when the sun is at its peak. Taking an average through the year, we define this measure as the “mean solar day", assigning it 24 hours or 86,400 seconds.

But there are some caveats about this that are worth thinking through. Consider that sunrise, sunset and noon timings—as I explained in the previous two columns—do not depend solely on the Earth’s rotation. Our planet’s motion around the Sun also has some bearing on them. That’s because our position in relation to the Sun is changing all the time. So because the Earth’s axis is tilted, at any given spot on the Earth, you will see the Sun at a slightly different spot in the sky every day. On our floor at home, we have a starkly visual reminder of this. We’ve been marking where a particular shadow is cast at 9:10 every morning since the winter solstice. In just over a month, that shadow has plotted a deep, foot-long “C".

The upshot of all this is that using these solar phenomena—sunrise, noon, sunset —is not good enough, if we want to measure a day with any precision. So how can we measure how long a day is? That is, how do we measure how long the Earth takes to make precisely one rotation?

Part of the answer is to use not the Sun, but far more distant celestial objects, like stars. Stars are so far from us compared to the Sun that we can treat them as effectively fixed in relation to the Earth from day to day. We can’t do that with the Sun. Strictly, nothing in the universe is “fixed"—every single object is moving in different ways. But astronomers find ways to define and use certain frames of reference, like distant stars. This gives us what is known as sidereal time.

It goes something like this. From a given location on Earth at night, look out at a given star. Let’s say you choose Betelgeuse, the red giant in the constellation of Orion: bright and hard to miss. You note its celestial coordinates (one kind are called right ascension and declination, which I’ll leave for another time). Make a precise note of the time of this observation. The next night, you wait for Betelgeuse to reach that same spot in the sky—meaning the point at which it has the same celestial coordinates that we noted the previous night. Make a precise note of the time again. The difference between today’s note and yesterday’s is one day, in sidereal time. Again, there’s more involved here, meaning a more accurate definition. But for now, this is a basic understanding of sidereal time.

Sidereal time gives us a more accurate measure of how long one rotation of the Earth takes. A sidereal day, then, is about 86,164 seconds, or 23 hours, 56 minutes and 4 seconds.

Note that that’s about 3 minutes and 56 seconds less than the mean solar day. That’s the difference I alluded to above, due to the Earth’s orbit around the Sun. Over a year, because of the circuit it makes around the Sun, the Earth makes one extra rotation on its axis, for a total of 366—not 365. (Aside: To understand this, place two identical coins on a table, touching each other. Hold one down with a finger and rotate the other, gear-fashion and slowly, around the first, making sure they are touching all the time. Watch carefully. You’ll find the moving coin makes not one, but two rotations around its centre.)

Incidentally, scientists also define a “stellar day", which is about 8.4 milliseconds (ms) longer than the sidereal day. I’ll leave that for another stellar day.

Two final nuggets about all this and I shall call it a day.

One: over the last several hundred million years, the Earth’s rotation has been slowing down, giving us longer days. Some fossil records suggest that 380 million years ago, there were about 400 days in a year; or just under 22 hours in a day. One possible explanation for this lengthening is that the cycle of tides is a slight but constant drag on the planet’s rotation.

But two: last year, the Earth rotated faster than it has for many years. Consider this: until the end of 2019, the shortest solar day we’d had since 1973 was 5 July 2005, 1.0516 ms less than 86,400 seconds. But in 2020, we beat that record —wait for it—28 times. 2020’s shortest day was 19 July, at 1.4602 ms less than 86,400 seconds.

But wait for it again: The average day in 2021 is expected to be about 0.14 ms shorter than we experienced in 2020.

It’s really true. Never enough hours in a day.

Once a computer scientist, Dilip D’Souza now lives in Mumbai and writes for his dinners. His Twitter handle is @DeathEndsFun

Subscribe to Mint Newsletters
* Enter a valid email
* Thank you for subscribing to our newsletter.

Never miss a story! Stay connected and informed with Mint. Download our App Now!!

Edit Profile
My ReadsRedeem a Gift CardLogout