Hello All!

I finally arrived in Australia last Sunday (the 19th), and I was greeted at the airport by one of Professor Tkalcic’s graduate students, Marija. The first thing that struck me upon getting on the ground was how cold it was! For some reason it hadn’t really sunk in that I was heading into winter until I had the wonderful, “my nose and ears are about to freeze off” feeling. After getting picked up from the airport, Marija gave me a brief tour of the city and we went and got lunch. The university is located right next to downtown, and there are lots of shops and restaurants as well as tourist-y locations that I hope to visit soon! My residence is about 15 minutes away from the university by car, and the bus has been an efficient means of getting to and from the school. Professor Tkalcic was out until Wednesday of my arrival week, so Marija told me to take Monday and Tuesday off to settle in and adjust to the new season and time zone. I took those few days as an opportunity to do some grocery shopping and exploring, and I was amazed at how diverse the wildlife is here! Within 24 hours of arriving I saw kangaroos and nearly a dozen different species of colorful birds.

Here are a couple of cockatoos I found near my house (they’re really beautiful but they’re also VERY loud in the morning - actually all of the birds are very loud in the morning…)

I would’ve taken a picture of the kangaroos I saw but they were kind of far away so I couldn’t get any good shots, but I will definitely have some kangaroo pictures by the end of my time here!

And here is a picture of Telstra tower, one of the landmarks of Canberra:

Some interesting things that I observed within my first few days of arriving: In Australia, they don’t really have “strip malls” like we have in the U.S, and the grocery stores are found in an indoor plaza along with various restaurants and other shops. Also, being from California, I’m used to seeing something priced at say, $9.99, and having to pay something like $10.78 for it. Conveniently, here in Canberra, the price you see is the price you pay, and they even include a “rounding” on receipts so that all prices are rounded to the nearest 5 cents (no pesky pennies!). Lastly, seeing cars driving on the left side of the road has been very tough to adapt to. My instincts about which way to look before crossing streets are betraying me, and I’ve had to stop and think a lot more at intersections to try and figure out where cars are moving (yes this has attracted me some stares and looks from drivers).

When I finally started working on Wednesday we hit the ground sprinting seeing as AGU abstract deadlines are quickly approaching! The group that I am working with is quite diverse (there actually aren’t even any native Australians!), and there is another intern from India that I am sharing an “office” with. (It’s more of a computer lab) Luckily he already has a masters degree, so he has been an invaluable resource when it comes to programming questions. The project that I will be working on consists of calculating moment tensor solutions using a Hierarchical Bayesian Inversion on an event in the Geysers geothermal region in California. The advantage of this technique is that it creates a model with more realistic parameter uncertainties by factoring in aspects such as noise. In order to do this, I will be working with a wide variety of computing tools including: SAC, GMT, Matlab, Fortran (although most of the Fortran codes are already written, I just make minor tweaks to them), and most importantly UNIX, lots and lots of UNIX.

We have event data in the form of SAC files from 30 stations around Northern California/Southern Oregon, and my first task has been to compute Signal to Noise ratios for a particular frequency band in these waveforms. Since the Bayesian Inversion is computationally intensive, we want to narrow down our number of stations to something around 10, so we want to find an ideal combination of stations (hence looking at the quality through Signal/Noise ratios). After computing these ratios, and ranking them, I then began to play around with different combinations of stations in a linear inversion, which is much less computationally demanding. I am still running different combinations and trying to find the ideal array in terms of azimuthal distribution (we want a good ~circular spread), distance from the event, signal quality, and quality of fit. This phase will likely keep me busy for the upcoming week while I also scramble to write an abstract within the week! (luckily I’m getting lots of help from Marija and Professor Tkalcic on this!)

And here are some fun figures I've produced. These are some GMT plots that I made which show the locations of the stations in our study region. (I still need to add a star for the actual event location...)

And here are some moment tensor solutions computed from a linear inversion (note that the 30 station and 10 station arrangements produce quite different solutions, I'll get back to you on why that is when we figure it out).

Well that’s it for now, stay posted for much more to come!

Cheers!

By Rob on August 8th, 2015

Glad to see you settling into Australia, Alex! You've already got a lot done and it should be a very interesting project. On the moment tensor inversion, it's interesting to note that it looks as though each inversion (10 and 30 stations) actually gives 2 solutions - the beachball and the lines. Challenge question - what is the significance and which would you expect given that you are studying geothermal events?

By Alex Burky on August 17th, 2015

When you ask about the significance, do you mean in a broader sense? Or do you mean the significance of the results of the inversion? And given the geothermal region, we're sort of expecting to find events dominated by tensile failure, since the region is constantly being pumped with water to maintain hydrostatic pressures at depth - so this pressure probably cause cracks to propagate and open up in the subsurface.

By Rob on August 21st, 2015

Good answer, lets go deeper. In your most recent post, you talked about focal solution mechanisms. What would one of the crack opening events driven by hydrostatic pressure look like?

By Alex Burky on September 2nd, 2015

Depending on the crack geometry, it would have an Isotropic sort of component. Pure expansion would give a pure isotropic solution, so a beachball that's all black - sort of like a little explosion occurring underground. But the crack opening would have some dilatational component, so it would be a mostly black beachball with either a sliver of white in the middle, or with two small white "lobes." And then if this is oblique, as it likely would be, then it would be some tilted variant of one of these.

*You must be logged in to post a comment.*