The butterfly effect in deep places

I recently found another thought-provoking feed to follow on Facebook and Tumblr – the Earth Story. In comparison to NASA Earth Observatory, which I’ve posted about before, this feed delves below the mappable surface of the earth into deeper aspects of its history and processes.

Today this article popped up on my feed and I couldn’t resist reblogging. This argument weaves an intricate, tangled web of interactions between solar activity, glacier formation, pressure, and volcanism that seems almost too good to be true, so I’m delving deeper into it.

Their hypothesis hinges on one assumption and two simple mass-balance concepts

  1. the Milankovitch Cycle’s effects on the amount of sun the earth gets influence glaciation
  2. When the earth is colder, water accumulates in glaciers on the continents, lowering sea level. When the earth warms up, that water flows back into the sea, raising sea level.
  3. Magma beneath the earth’s crust flows towards areas of low pressure, and away from areas of high pressure.

Combining the three, the author reasons that during cold periods where more mass in on the continents, the mantle will tend to flow towards the relatively lower-pressure oceans and increase the rate of plate movement over a scale of a thousand or so years. Conversely, when those glaciers melt the hot mantle will ooze back towards the continents, increasing volcanism on land, also on a thousand-year timescale. The author then takes the step of associating the Milankovitch cycles with the changes in climate, and therefore with the changes in volcanism.

Described like this, it’s easy to imagine that the solid plates float on top of a layer of Hawai’ian-pahoehoe-style lava – but that’s not the case. The mantle is more like a slowly flowing solid, under enough pressure from the rock above it that even at high heat it stays solid instead of melting. Decompression melting, as mentioned in this article, occurs when a parcel of rock rises into a lower-pressure area at a faster rate than its temperature can equilibrate to the rock around it, and therefore begins to melt properly.

The “gotcha!” moment is that researchers have actually found a way to test that hypothesis! Using high-resolution mapping and isotope data from mid-ocean ridges, researchers at the University of Oxford and at Harvard correlate surges in plate creation to glacial cycles. Unfortunately, the author of the Earth Story article doesn’t cite his source, so I had to do some sleuthing. Turns out that the article is based on Glacial cycles drive variations in the production of oceanic crust by Crowley, Katz, et al., published in Science.

Decompression melting

Bathymetry (depth survey) of the Southwest Indian Ridge, taken from Crowley et al.

I think Crowley, Katz, et al. have found sound evidence that glaciation/ocean depth pressure differences affect crust production at mid-ocean ridges, but I haven’t managed to find an article with any evidence that volcanoes on land might be affected as the system swings in the opposite direction. It’s always much more difficult to get good chronologies from land-based systems, though, because of that pesky little thing called erosion.

Also, the author of the Earth Story article seems to imply that Milankovitch cycles cause those changes in glaciation, while the Science article only mentions glaciation cycles. While there are excellent mathematical models to chart the interactions between eccentricity of Earth’s orbit, changes in the planet’s tilt, and precession, their combined signal is relatively weak and doesn’t correlate with any glaciation before the Pleistocene.


One thought on “The butterfly effect in deep places

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s