I just got back from field work (digging holes, mixing cement, and talking to landowners) over the weekend.  it was fun, exciting, and exhausting.  we even had some time to make a few tourist stops. 

 After digging a 3 ft hole, 15 inches in diameter we placed the black vault in the ground on top of a layer of cement.  After leveling the vault, we would then mix more cement to surround the sides of the vault a few inches.  Grout was also mixed to be poured into the vault.  The grout was a more liquid consistency so that when dried, it would create a more level surface for the seismometer to be placed upon.  The grout was also self-expanding so that as it dried it pushed the inside walls out against the concrete and stablized the vault structure.  The concrete and grout are used to couple with the ground and create more accurate readings.

A finished hole looks like the following picture.  Just an inch or so of the vault is visible, along with the lid.  Each lid was covered with a tarp to take precautions against water entering the vault.  Dirt from the hole was then piled onto the tarp, so that when we left each site just a small mound of dirt was visible.  In October the seismometers will be installed as well as an Action Packer and a solar panel.  

While I was in Oregon and Washington I got to learn a lot about searching for the best sites.  I also helped with a preliminary seismic test we did on the surface of each station.  We set up a seismometer on the surface and took data for about half an hour.  We haven't looked at that data yet, but the kids we encountered really liked jumping around it and seeing the waves they created.

Finally, we made some time to visit Mt. St. Helens and Crater Lake.  Here is a picture of our group at Crater Lake... Hector, Mike, Me, and Devin.

I am now going to switch from looking at Mexico data to Cascadia.  I'm hoping to find some correlations between the activity in both areas. 

For the past week or so I've been looking at low frequency events instead of tremor.  While I was looking for patterns of tremor frequency, I noticed a bunch of large amplitude, low frequency events, mostly between 1-10hz.  So right now I've been trying to look at a large sequence of days to see if these low frequency events have some relationship with when the tremor occurs.  The low frequency events on seismographs look pretty much like earthquakes (however earthquakes are usually at higher frequencies 5-20hz).  We are also going to see if these low frequency events happen in certain distinct locations.

Other than that, I'm looking forward to field work with Mike and a couple of his grad students in Oregon and Washington.  I think we'll be gone for about a week and a half to look for seismic station sites.  We'll be digging and installing vaults, but we don't have the instruments yet, due to some sort of mix up with PASSCAL.

The following are 2 spectrograms from the OXNC station. The first spectrogram shows tremor and the second does not. On the spectrograms the x-axis is time, here a full day, the y-axis is frequency, and the different colors represent different amplitudes. Notice lots of purple in the first spectrogram.

The next two spectrograms are the same days from another stations (OXPL).

The next two images show two plots, one from station OXNC and the other from OXPL. The x-axis is frequency and the y-axis is amplitude. The first graph shows loud tremor, which can be noted by the large ampl range and the second graph shows a time inbetween tremor event, possibly quiet tremor. We can see similar spikes in ampl at both stations.

after looking at another station in the area, it was difficult to see the frequency content of tremor within the spectrogram.  hypothesized patterns based on spectrograms from station OXNC were unable to be seen at station OXPL.   we think that we weren't able to see as much at OXPL because of the changes in ampl and various cultural noise at each station.  within the spectrogram feature in SAC we aren't able to alter the ampl settings, so we have changed our methods to look at frequency content after performing fast fourier transforms.

so since the 4th of july i've been looking at the fourier transforms of individual hours of tremor events.  it has been slightly tedious and while i am getting better at taking notes on what i'm seeing and what to look for, there is just a lot of information to take note of.  I have been caught between wanting to just do all possible days making cohesive notes about what the frequency content looks like throughout an event and between doing some days and then comparing those to other stations.  At station OXNC, I came across several plots that contained multiple bumps between 4 to 10 hz.  Mike and I talked about how something like may be harmonic and simulating a process like volcanoes, fluid flowing through a conduit, causing a sort of oscillation.  Mike has warned me to stay skeptical, so I will.

also, after looking at the data from Oaxaca, I will look at some of the data from Cascadia.  I think this will be a great opportunity for me to learn about other plate boundaries and find out more about frequency content.

Right now I'm hoping to go through five or six stations looking at the same tremor events.  I'm not sure how many records they have on hand, but I suppose the more events that I'm able to look through, the better.  Right now, it is going kind of slowly looking at spectrograms made on SAC for each day, but as I look at more and more information I am getting better at identifying frequency content characteristics.  I am really intrigued by the Oaxacan tectonic region as well.  Like Andy said, it is relatively shallow and I think it will be interesting to compare our findings with other tectonic regions and hopefully find more constraints on what causes tremor.  I feel like I should do some more reading on other results, different tectonic regions, potential Moho effect.