5
$\begingroup$

When comparing the average distance of the all known earthquake epicenters to the nearest pole, is the longitudinal distance to the equator or the poles the smallest of the two? Please note that any period for which earthquakes were able to be tracked must have taken place globally and allowed the location of the epicenters to be be estimated in a meaningful way as it relates to the question; meaning if the location is off by 3% that's likely okay, being off by 50% would not be okay.

The Global Historical Earthquake Archive (1000-1903)

enter image description here

Global earthquake epicenters, 1963–1998

enter image description here

Improve this question
$\endgroup$
5
  • 6
    $\begingroup$ When looking at global maps, be careful not to be misled by the map projection. If Earthquakes are equally spread over the globe, they will appear closer at lower latitudes on any projection that is not equal-area. $\endgroup$
    – gerrit ♦
    May 1, 2014 at 4:13
  • $\begingroup$ +1 True, good point, in fact, I'm not even apply to id the type of cylindrical projection the map above is using looking at this list on Wikipedia. If I did, then at the very least I could look at it in this animated map projection transitions viewer. $\endgroup$
    – blunders
    May 1, 2014 at 4:29
  • $\begingroup$ Don't the images you've provided basically already answer the question? Also, is there some kind of intent behind the question? Just wondering why this question is interesting... $\endgroup$
    – naught101
    May 1, 2014 at 6:32
  • 1
    $\begingroup$ Note that prior to 1903, there are no earthquakes mapped as located along the mid-ocean ridge ;) One reason I found this question interesting was because I think it illustrates our human nature is to rush to find patterns in incomplete data. When doing science, we need to guard against that. $\endgroup$
    – Mark Rovetta
    May 1, 2014 at 18:28
  • $\begingroup$ @MarkRovetta: Yes, data literacy is a major issue; both within science and within the general public. $\endgroup$
    – blunders
    May 1, 2014 at 18:32

2 Answers 2

Reset to default
7
$\begingroup$

Most earthquakes happen at the edges of the Earth's tectonic plates. Look at a better sample of earthquake epicenters collected using modern seismological techniques and compare to this List of tectonic plates

Earthquake epicenters are concentrated at the edges, and more sparse in plate interiors, but there are regions of high seismicity at high and low latitudes.

If you really want to know whether the concentration of earthquake epicenters is greater near the equator than near the poles, then you should try the following:

1) Divide the surface of the globe into ~30 regions of equal area.

2) Make a table with a row for each region. Give the table two columns: the region's latitude (L), and the number of epicenters in the region (E).

3) Compute the arithmetic average of E for all 30 regions, call this G. Take the square-root of G and call this D.

4) Now look at your table. How many regions have a concentration E>5*G? How many of these are located at 'polar' latitudes and how many are located at 'equatorial' latitudes?

5) If you find that 80% or more of the E>5*G regions are in 'equatorial' regions, then that is probably evidence that significantly more earthquakes are measured in 'equatorial' regions. If you don't find this much, the effect is small.

Improve this answer
$\endgroup$
4
  • 1
    $\begingroup$ +1 Thanks. Agree, that's my understanding too. Another way of asking the question would be, do tectonic plate edges exist at a higher frequency near the equator or poles? (I'd say no, by point is that your answer in my opinion needs to answer the intent of the question; possible I've missed something though.) $\endgroup$
    – blunders
    May 1, 2014 at 4:51
  • $\begingroup$ Not my field, but maybe there is "a tendency of the plates to move from the poles to the equator"? $\endgroup$
    – gerrit ♦
    May 1, 2014 at 13:29
  • $\begingroup$ Thanks for the update, though seems like a pretty complex formula, any reason why you couldn't just plot average longitudinal distance of the all known earthquake epicenters to the nearest pole, then see if this distance is closer to the equator or the poles? The distance from equator to the either pole appears to be 10000km, which would mean if the average is 5000km+, the epicenters of earthquakes happen closer to the equator. Might make sense to use a geometric mean instead of an average. Main issue is finding a dataset that's complete enough to use, and has the location metadate. $\endgroup$
    – blunders
    May 1, 2014 at 18:41
  • $\begingroup$ Earthquake data sets are available in earthquake catalogs. Many are available online such as the ANSS global earthquake catalog. Seismologists have to consider the suitability of the catalog they choose. These catalogs contain large amounts of data, there are reasons to 'block the data' rather than take an average of everything, both theoretical and practical. The procedure above is a simple one which would be relatively simple to implement. $\endgroup$
    – Mark Rovetta
    May 1, 2014 at 19:46
2
$\begingroup$

Here is one way to think about this problem.

Earthquakes that happen above latitude 45 are closer to the pole; the ones between +-45 degrees are closer to the equator.

Since the earth is a sphere, the total area of the earth's surface between +-45 degrees is about 70% of the total area (1/sqrt(2) to be precise). In other words, there is "a lot more earth near the equator than near the poles" - more than twice as much. This means that if earthquakes occurred with equal probability anywhere in the world, you would expect the answer to be a resounding "yes".

If you take into account the maps you are showing, I think that the answer is still yes - even without doing a detailed calculation.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.