The Indonesian Throughflow: The Fifteen Thousand Rivers
December 1, 2016 3:43 AM   Subscribe

If Indonesia did not exist, the earth’s day would be significantly shorter than the approximately twenty-four hours to which we have become accustomed.
This would have a dramatic impact on the clock industry.
(In case you forgot what a Sverdrup was. - It is equivalent to 1 million cubic metres per second or 264,000,000 USgal/s)
posted by adamvasco (13 comments total) 16 users marked this as a favorite
 
If Indonesia did not exist, the earth’s day would be significantly shorter than the approximately twenty-four hours to which we have become accustomed. This would have a dramatic impact on the clock industry, of course, and familiar phrases like “twenty-four/seven” would have to be modified.

Hmm. It might be hard to imagine if you're a bad author and have never tried it yourself, but it's perfectly possible to divide also shorter time intervals in 24 parts. Heck, even the ancient people who first decided to split up the earth's day in 24 bits knew how to do that.

(the author also seems to have forgotten to explain how much shorter the day would be, but it could just be that I got a bit distracted by the idea of every animal on earth going to Bali for vacation...)
posted by effbot at 4:09 AM on December 1, 2016 [5 favorites]


You could try the first paragraph.
one of the most voluminous flows of water on earth, a task that consumes so much energy that it slows the very spinning of the globe.
posted by adamvasco at 4:19 AM on December 1, 2016


With so many opposing currents side by side, it looks like two ships starting out together could stretch something between them and launch Intercontinental Ballistic Rubber-bands.
posted by StickyCarpet at 4:21 AM on December 1, 2016 [1 favorite]


one of the most voluminous flows of water on earth, a task that consumes so much energy that it slows the very spinning of the globe.

Not sure how "it slows the spinning" answers the question "how much does it slow the spinning?"
posted by effbot at 4:26 AM on December 1, 2016 [4 favorites]


I thought that there was a bit of [citation needed] about that claim, so I went looking and found this. tl;dr--lots of weather/geological macroevents cause measurable (if very minute) changes in the earth's rotation.
posted by Halloween Jack at 4:48 AM on December 1, 2016 [2 favorites]


The article also failed to discuss careers in the clock industry
posted by thelonius at 5:24 AM on December 1, 2016 [3 favorites]


I will not feel safe until the scientist responsible for abbreviating Sverdrup as "Sv" is behind bars. There's no telling what else he's capable of.
posted by groda at 5:59 AM on December 1, 2016


The Boxing Day earthquake / Tsunami displaced ~30 cubic kilometers of water, made the Earth wobble on its axis, and slowed the day by ~3 microseconds. So it's not hard to imagine that a constant flow of 15 million cubic meters per second would have an effect over time.
posted by Bee'sWing at 6:49 AM on December 1, 2016 [2 favorites]


Um. A yearly flow of 15 sverdrups?
A yearly flow of cubic meters per second?
Cubic meters per second per year?
Wat.
posted by Horkus at 7:51 AM on December 1, 2016 [4 favorites]


That much water flow would be a good source of electricity.
posted by eye of newt at 9:08 AM on December 1, 2016


I've been diving in that area and the currents and flows are hard to imagine. The Gulf Stream moves about twice the water but doesn't have the restrictions the create the strong currents.

The fastest current I've swum against off of FL is about 4 knots. That is a very hard swim. I hit 8 knots in Bali. That is basically "you are going far out to sea today".
posted by pdoege at 11:55 AM on December 1, 2016 [3 favorites]


Hmm... the last time I pulled a bunch of numbers out of my ass to comment on a post I thought was dead, Metafilter seemed to appreciate it, and I'm as annoyed by the lack of quantification here as everyone else, so let's see if there's any more numbers deep in there, shall we?

Warning: even the crude phrasing above does not convey the magnitude of error which is likely to be involved in the following calculations, or the extent of the reaching which is required to get any answer at all. I'm about to taste dirty fingers at the back of my throat.

How do current flows change the length of the day? In one way: The Earth has a fixed amount of angular momentum that determines how fast it spins, so if parts of the earth are spinning faster or slower than others, then the rest of the Earth must be spinning slower or faster than it otherwise would to make everything equal. In other words, water flowing westward makes the day shorter, water flowing eastward makes the day longer.

But here's what's incredibly tricky: on average, the current flowing into a place has to equal the current flowing out of that same place. So to look at the effect of current flowing through Indonesia, we have to try to guess how that current gets replenished. We can't just look at water flowing through Indonesia, we have to look at water flowing around the whole planet.

Looking at a larger map of ocean currents, it doesn't appear that there's any significant westward current circumnavigating the globe other than the Antarctic Subpolar current, which is pretty well shielded off from the rest of the oceans by the Antarctic Circumpolar current. This suggests that flow from the Pacific to the Indian oceans is probably being returned to the Pacific via the South Pacific current or the neighboring Antarctic Circumpolar current.

As a wild guess (looking it up in the appendix, so to speak), and trying to err on the side of overestimation, let's say that blocking up Indonesia would stop all 15 Sv of flow and that that flow currently travels in a loop west all the way from South America to Africa, south almost to Antarctica, east back to South America, and back up north again there. We'll say the path is about 24e6 m west near the equator, then about 12e6 m east near the antarctic.

I have no idea what the velocity of this flow is, but fortunately momentum is all we care about, conservation of momentum applies, and so it doesn't matter whether parts of the flow are narrow and faster while others are wide and slower. 15 Sv is 15e6 m^3/s, which for water at 1e3 kg/m^3 gives a mass flux of 15e9 kg/s. A flow of that magnitude over 24e6 m has 360e15 kg*m/s of linear momentum, and if that's happening 6e6 meters from the Earth's axis then it accounts for 2.16e24 kg*m^2/s of angular momentum. (Note: using 3 digits here is an attempt to avoid rounding error, not an attempt to pretend that we have 3, or even 1, significant figure of accuracy. We're deep into the small intestine now.)

That's reduced on the eastward component by the same 15e9 kg/s but this time multiplied by 12e6 m and 3e6 m: 540e21 kg*m^2/s. The net angular momentum is then 1.62e24 kg*m^2/s.

If we assume that the Earth is a perfect sphere of uniform density (is this burning sensation on my hands stomach acid?) then it has an moment of inertia of 2/5 * 6e24 kg * (6.4e6 m)^2, or 98e36 kg*m^2. Earth is rotating at 2*pi/24/3600 = 72.7e-6 rad/s, so the total angular momentum is 7.1e33 kg*m^2/s.

The ratio of angular momentum between the Indonesian current and the Earth is thus 1.62e24/7.1e33, or 2.28e-10. That's also the ratio describing the change in the day which would occur if the current was stopped: with the water moving faster in an inertial frame of reference (because the current was mostly opposed to the Earth's rotation), the Earth would be turning slower, and the day would be 2.28e-10*24*3600 = 19.7e-6 seconds longer.

Okay, I'm clearly performing a poorly-guided self-tonsillectomy at this point, but I have been trying to err on the side of larger numbers rather than smaller, and I still only get a clock effect measured in tens of microseconds per day. If we didn't want to redefine the second, we'd be able to compensate for this effect by adding a new leap second every 1/19.7e-6/365 ~= 140 years.

Clock makers everywhere can breathe a sigh of relief.
posted by roystgnr at 7:29 AM on December 2, 2016 [1 favorite]


Thanks for the numbers roystngr. I'm glad I don't have to worry about filling those microseconds (and actually, that's a bigger effect than I was expecting, even if you're a couple of orders of magnitude out.)
posted by ambrosen at 8:31 AM on December 2, 2016


« Older SLNYT   |   A woman's face with Nature's own hand painted Newer »


This thread has been archived and is closed to new comments