Or, in other words...

Perhaps that previous post was a bit jargon-dense. Let's try again.

I'm what's generally called a structural seismologist -- that is, the focus of my research isn't understanding earthquakes, it's understanding Earth structure. Seismic waves from earthquakes bend, bounce around, convert from one type to another, and repolarize, all depending on what structures they encounter as they travel through the Earth; because of this, it's possible to extract information about what's inside the Earth from the different arrivals that show up on a recorded earthquake (a seismogram). For this kind of work, it's not really a penalty to be located in a region that doesn't get many earthquakes (such as Manitoba); the energy from a good-sized earthquake (magnitude 5.5 or so) can be detected almost anywhere in the world.

Largely from this sort of information, we've known the basic layered structure of the Earth for many years. The top layer, and the most heterogeneous, is the crust; it's thick and light in the continents, thin and dense in the ocean basins. Underneath the crust (which is a few tens of kilometers thick, at most), there's the mantle, which is made of denser rock than is common at the surface; the bulk of the Earth is mantle, all the way down to 2890 km where the iron core appears. Because the mantle is less heterogeneous than the crust, it transmits seismic waves in a simpler fashion, without extra arrivals bouncing off of 3-D structures. Waves that spend most of their time in the mantle are simplest to analyze; the distance range (distance between the source and the instrument, that is) at which this happens is called the teleseismic range, and runs from about 30 to 100 degrees of arc.

There are a bunch of techniques for using teleseismic earthquakes to separate out receiver-side structure. One of the most basic is called the receiver-function technique, which takes advantage of some geometrical conveniences to effectively separate out source-side and receiver-side effects. The upshot of this is that a receiver function, derived from a teleseismic earthquake, ideally depends only on the structure of the crust and upper mantle beneath the instrument. So, basically, if you set up a bunch of seismometers and record teleseismic earthquakes for a couple of years, you can use this method to form images of Earth structure over the region you've covered with instruments.

Lately, I've been doing this for a bunch of instruments in Ontario, and I've been seeing some weird and unexpected stuff. I mostly expected to see a difference between the major geologic provinces of Ontario, as seen in this map: the ancient Superior province (speckled pink and dark green covering most of the map), and the somewhat less ancient but still old Grenville orogen, formerly a mountain belt (reg, grey, light green). But what I've been seeing is variation in structure within the Grenville: stations SADO and GAC, which both have really good data sets, are showing complicated layered structures at the top of the mantle -- different complicated layered structures.

Which is why I said what I did below.

Anyhow, I'm about to let those data be for a week while I bum around Vancouver and Seattle; it's my last chance for a vacation before the start of classes.

Comments

[Jacqueline]

Interesting. I would like to think that I know more about the earth than your average person, given the fact that I live in one of the most geologically active places on earth and am in a position where I occasionally have to explain said processes to people as part of my job, but I for some reason had never thought of people doing what you do. I don't what I thought people based their theories about the earth's structure from... it's not as if we can drill a hole big enough to check everything out. Thank you for posting this and clarifying a bit what it is that you do for a living.