Kat

August 23rd, 2018

Hello all! Sorry about the delay in this blog post. During my last couple of weeks at Penn State, my computer crashed and after two weeks of repairs I finally got it back. Sadly, they had to wipe my computer and, though I have all my data and figures backed up, all of my programs are gone and need to be redownloaded.

Since my last post, I finished creating my many, many velocity models, wrote and submitted my abstract for AGU, and spent the last of my time at PSU working on my poster. While we still need to interpret my results, I’m extremely proud of the progress I made this summer on my project! I am incredibly grateful for the opportunity to participate in the IRIS program this summer, and am especially thankful for the wonderful mentorship I received from Dr. Natalie Accardo and Dr. Andrew Nyblade during my time at Penn State. I learned a great deal about seismology research, including my first taste of field work, and hope to carry this knowledge forward into my last two years of undergraduate work, as well as into graduate school.

Thanks for reading!

July 20th, 2018

Hello again!

It’s been awhile since my last post and as such my project has progressed quite a bit since then. As previously stated my project this summer uses the data from our geophone deployment during my first week here; specifically I’m working with the data from our longest line of geophones, which consisted of 216 geophones. A contour map of our critical zone area in Shale Hills is shown below, with every geophone that was deployed represented by a blue dot. The line that I am studying is outlined by the purple rectangle. The Shale Hills Critical Zone Observatory was chosen as our study area as it has been studied by multiple other research groups, so we can compare our data to other geophysics studies conducted in the region and supplement our data with various other data about the region, such as borehole data. Secondly, we wanted to use this region for our study as it was only twenty minutes away and Penn State has access to it, as they are a primary partner of the observatory.

My first step in data processing was to pick the seismic wave arrivals for all the data that was collected on my line of geophones. This included 216 sledge hammer sources and 23 betsy gun sources, so I had a sizable amount of picking to do over the course of a few weeks. An example of what this looks like can be seen in the wiggle plot of one of my sources below. The red lines are the wave arrivals I manually picked.  

In order to identify outliers or bad picks, two things were done. First, I wrote a script in matlab to plot the picks that also identified any picks that were close in distance but significantly far apart in arrival time, as those would be outliers and should be removed. Second, I ran reciprocity checks on the data, where I compared the picks from one sledge hammer source to the picks from the 215 other sledgehammer sources to identify where they did not match up. An example of a reciprocity check is seen below. Most of the picks are fairly close, but some, such as the purple pick from shot 9551, is slightly too high meaning I would go back to sledge hammer shot 9551 and try to fix the problem.

After reciprocity checks were completed and the picks where fixed as much as possible, I loaded my pick files into SeisImager in order to run an inversion on the data to create a velocity model of the critical zone beneath my line of geophones. A slight problem I have run into is that the program can only run an inversion with a maximum number of 65 pick files. As I have over 200 files, we decided to break my line up into multiple sections and run it piece by piece and use the 23 betsy gun sources to create a model for the entire line. The betsy gun shots were used as they are significantly stronger than the sledgehammer shots and the wave arrivals can be picked to much farther distances. Below is a preliminary example of a velocity model of the entire line using the betsy gun picks. Interestingly, as of right now there appears to be a change in the thickness of the intermediate layer, shown in green in the model, but the upper, slower layers have a more uniform thickness, shown in red and pink in the model. We plan to explore this feature and its possible causes more as I create additional models of the region.

Alongside my research, I have also been working on presenting my work in my lab, to my peers, and to various friends and family. Some forms of this include an “elevator speech”, or a 60 second description of my work that I could pass along in an elevator ride, a two minute presentation to the rest of the IRIS interns limited to three slides, and a discussion of my work thus far to my lab at Penn State. As all of these talks have different time lengths, access to visual aids, and audience background knowledge, they have been an educational exercise in how to change how I present depending on who I am talking to and what I want them to get out of my talk. Though I had some experience discussing my research to others before in formal settings, I'm greatful for the new knowledge I have gained this summer.

June 21st, 2018

Hello again random person reading my blog!

This week was certainly more calm than last week, as we started our foray into data processing and analysis, but there were quite a few bumps along the way. Since this study is breaking new ground in 3D seismic processing of the critical zone, it’s been a learning process for everyone involved.

We started processing the active-source data our array collected and spent a day or so experimenting with filters and gains, before eventually deciding to pick our waves with little-to-no processing of the data. We then tried various picking GUIs, with varying levels of success, before finding one that worked. From then on, I picked wave arrivals, coded a bit in Matlab and Seismic Unix, and looked into various software packages to analyze surface waves with the help of Natalie, my advisor. As of now, we’ve had quite a bit of trouble finding software that we have access to and will do what we need and it seems like the surface wave analysis we planned for me to do might not be feasible. Though the week hasn’t gone as smoothly as I had hoped technology and software-wise, I am still super excited to be working with the data that I helped create and learning how to use so many new programs!

My current goal is to get comfortable with the all the software we plan to use this summer, as almost everything we’ve used so far is new to me. I’ve powered through quite a few manuals and papers in order to gain more knowledge on the packages and the code they contain under the hood (ah, SeisImager manual and your 314 pages, you did not make my life easy…), but I’ve definitely improved as the week has gone on and hopefully I will get to use this information soon in my future work!

June 10th, 2018

Hello whomever has stumbled upon this blog!

I’m Kat, one of the IRIS interns this year and a rising junior at Yale University. So far, this internship has been a wild ride and I can’t wait to see what the rest of the summer brings!

A couple of weeks ago, all nineteen interns had a weeklong orientation in Socorro, New Mexico. Though the days were quite long, I learned a great deal about geology and geophysics, data processing in SAC and Matlab, and seismic station installation. I was a bit wary of the hiking portions of orientation, as this astrophysics major has never done a day of fieldwork in her life and is not the most… outdoorsy person, but those were actually my favorite parts of the week! Hiking at an elevation of 10,000 feet on Magdalena Mountains and exploring San Lorenzo Canyon were amazing experiences, though the fact that I got to operate a telescope at Magdalena Mountain Observatory and star gaze in the canyon might mean I’m slightly biased. I already miss the friends I made during orientation and definitely plan to stalk their blogs for updates.

After the orientation, I immediately started work the next day for my project at Penn State. That week we deployed 2,100 geophones and dataloggers, sledgehammered between them to create active source data, moved all 2,100 geophones and dataloggers, sledgehammered again, them removed all the geophones and dataloggers and packed them up to ship back to IRIS.

In one week.

I know. I’m honestly shocked we managed it too.

Thankfully, we had the help of about twenty-five or so volunteers throughout the week, and Joey, the other IRIS intern at PSU, was always there with music at our lunch break to keep our spirits up (insert shameless plug for the new Kid Cudi album here). I also, somehow, managed not to get bitten by any ticks and avoid Lyme Disease, though there are definitely a few more bruises and mosquito bites on my body than I started the week with. With the majority of our data collected, I can’t wait to start processing and interpreting it this upcoming week!