# Earth's diameter is smaller than previously thought

1. Jul 6, 2007

### Staff: Mentor

I found this to be quite interesting.

continued

http://news.yahoo.com/s/afp/20070705/sc_afp/germanyscience_070705151649 [Broken]

Last edited by a moderator: May 3, 2017
2. Jul 6, 2007

### Mk

Wow, that's awesome, I had never thought of it before.

3. Jul 6, 2007

### Staff: Mentor

I was really surprised at how such a tiny difference could skew information.

4. Jul 7, 2007

### Gib Z

If i heard this about a year ago I would have been as well, but ever since I read about Chaos theory and Fractal's I'm quite scared about any changes at all! Fractal's are one example of where a tiny change effects the result greatly. Say $f(a_n)= g(a_{n-1})$ in other words, the function uses the previous value to generate the next, then as n approaches infinity, an initial small difference in a_n would be made very large, even though it was a tiny difference.

That's just fractal's, I read about a story concerned one of the discoverers of Chaos Theory. I can't remember the name but the guy was a meteorologist and printed some data, 10 decimals accuracy, that described a weather pattern. He wanted to re-create it later but didn't want his computer to take forever, so instead fed back in the data with 5 decimal places, thinking there would be little difference. To his shock it described a completely different system.

5. Jul 7, 2007

### matthyaouw

I wonder what the cause of the change could be... The article doesn't seem to say.

6. Jul 7, 2007

### PRDan4th

If the inner core is cooling slowly and condensing liquid iron from the outer core sinking to the inner core could slowly reduce the size of the earth and increase the density of the core. As the nuclear energy from radioactive elements runs down the earth should cool and decrease in size. Does this make sense?

7. Jul 8, 2007

### quark80

Heh. Fractals in geology...My area of expertise ;)

I can definately vouch for the fact that minute changes to the accuracy of some input numbers can drastically change outputs! Caused me endless pain trying to figure out why my results were completely obscure. All is well now thankfully :)

8. Jul 9, 2007

9. Jul 9, 2007

### wolram

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook