This project is part of the Permafrost Active Layer Seismic Interferometry Experiment. The goal is to track changes in the depth of the active layer. The part of the ground that remains frozen all year round is known as the permafrost. The part that thaws out is called the active layer. The depth of permafrost has important implications for infrastructure foundations. This project will test a new technique for mapping the permafrost layer. If successful it will be a cost efficient way to track time-varying thickness as it will negate the necessity of costly site visits.
Hard to believe this is my last week! The past few weeks has involved a lot code writing and processing data. We came back from Alaska with data from the suveys we took as well as from the stations that have been monitoring background noise. As I may have mentioned before there are 7 stations that have been installed since October of last year as part of the Permafrost Active Layer Seismic Interferometry Experiment (PALSIE). The stations are arranged in two circles of 25 and 50 meter radii composed of 3 stations each, plus a center station (see my early posts for a figure). Because of internet limitations at Poker Flats we can't download the data from Albuquerque, so one of the last things we did was to download up to the day before. We are also using data from a nearby PASCAL sight which is about 2 miles from our site on the top of a hill. We also did a ReMi (Refraction Microtremor) survey close to their station. The past few weeks has been spend processing the data from both PALSIE and PASCAL into the H/V spectral ratio. When we plot the mean and standard deviation we see fairly constant peaks in the lower frequencies and building fluctuations in higher frequencies. Today I'll be running the spectrum of the data to be sure that the changes we see are due to changing geology (as the ground freezes and thaws) and not a changing wave field. Hopefully we should also be getting data from the weeks since Alaska in the next day or two, although we will probably just do a quick check to see what has been happening.
Alaska: Days 1 and 2
Well today marks the end of a second busy day in Alaska. As I write this it is 10:30pm local time and the sun is still shining as if it were about 7pm. The sun will set at about 12:30 tonight but not enough to get completely dark before it rises again at 3:30am. We are staying at a hotel in Fairbanks as the site is just 30 miles outside of town at Poker Flat Research Range. Monday started out with a fairly steady rain in the morning. We geared up with our rain jackets in the midst of this and attempted to take some data. There was no problem in obtaining the data, we were just poking the ground with a stick until it hit permafrost, but there was a problem in recording it. Unfortunately, we did not bring a walkie talkie or a notebook you could use in the rain. We tried wrapping it in a plastic bag but very quickly found that this would not work. Fortunately the kind folks at the research range let us borrow one of theirs even though they were a little short as well. They also got to hear everything we said since it was hooked in their system. The rain actually started letting up after this and stopped altogether at some point. We did this across a couple of different lines on the field and tested the equipment from IRIS in preparation for the next day.
The idea for the second day was to lay down a line of 48 geophones, take data for ReMi (we just needed a source at the end of the line versus between every geophone). And then move the line and repeat. Set up however took much longer than anticipated. The ground we are working on is thawed ground above permafrost and is very wet (especially with the 6 inches of rain they’ve had since the middle of June) and spongy. Laying out the cables was difficult since we couldn’t take the entire roll and walk along dropping it as needed. Since the ground was also covered in a think moss we had drive the geophones some distance in the ground in order to get (we hoped) better coupling. In the end we finally got everything set up but it had taken I think 3 hours to set up. After lunch we got the hammer out to take some data only to find that it was not triggering! This wasn’t terrible since we could trigger it manually but a little inconvenient. The data started coming in finally but was not at all what was expected. We have some theories as to what caused what we saw and are going to test it out tomorrow.
Right now we are quite a bit behind schedule and have had to make some adjustments to the plan. Hopefully, the new sledge hammer with a metal tip will trigger better for one thing. Fortunately we had an extra day built into the schedule for contingencies. Looks like we won’t be having a “tourist” day but hopefully we will have learned a lot by the end. From what I understand, there has been very little seismic work done over permafrost. Tomorrow is a big day for us , especially as it is Rob’s last day, hopefully this time tomorrow we will be celebrating a successful day of solving the problems we encountered today!
PS. see a few photos and read about the last few days at my other blog [url=http://www.myintroductiontoseismology.blogspot.com]http://www.myintroductiontoseismology.blogspot.com[/url]
I did a lot of reading last week trying to wrap my head around the theory behind Horizontal to Vertical Spectral Ratio (HVSR) which is what we will really be testing out to see if we can use it to monitor the thickness of the active layer. It was a little frustrating trying to find the references I had from a document that came out of European Research project (SESAME). I tried reading a few other articles with minor success. I still have a lot to study in order to become an "expert" on this topic (since this is what my poster is all about!), but that's what learning is all about so it's ok.
This past week has been a little more exciting. I have been working a lot with code. Some of the data collection for this project has been ongoing and we have data going back to October. I learned how to use Geopsy to make H/V plots and then how to do the same thing using a command line. My next task was to have this be an automated process. The trouble was that every time I ran the code it asked me if I wanted to override the previous parameters file. I tried running the code without a parameter file and this problem went away, but we weren't totally sure what parameters Geopsy was actually using. So Rob brought another programming language out called EXPECT. As you may infer, it "expects" to get an interactive question and answers that question according to what you tell it to do. I needed to run basically the same code over and over again and just change a number at the end in a couple of places. We decided to go with the easy choice in tis case and just write out all the commands we needed to run. I used excel to do this in just a few minutes since excel is smart at recognizing counting patterns! Then I saved it as a text file and we had our commands. Now we just needed expect to go through and run each line one at a time. This was a little more complicated, but with help from an expert coder (Thank you Kim!) we finally got the code to run! As I write this it is just finishing up the last couple lines! Now that we have a code that works we can reuse this again as we tweek the parameters and use it for all the stations! Now I just need to import into matlab and plot it ; ). For some reason there doesn't seem to be a way to save the image that you would get if you went through the interface. The documentation is rather sparse.
We've also been getting ready for our big trip up to Alaska! I am so excited! We'll be there all next week collecting data. Hopefully I'll be able to give quick updates since we wont be camping out in the wilderness! Just hope the weather holds up as there is a forecast for cool weather and rain! We've got a contingency day built in but I am hoping everything will go well and we'll get to do a little site seeing. I still can't believe I get to go to Alaska!!!
Wow, last week was a whirlwind ride for sure. I went to a conference right here in Albuquerque and then took a crazy trip to the Grand Canyon and a few other places over the weekend.
I was fortunate to be able to attend the Review of Monitoring Research (RMR) for Ground-Based Nuclear Explosion Monitoring last week. This was my first experience at a conference. It was mostly poster sessions with a plenary sessions in the mornings and workshops on the last day of the review. This was not conference but a review and so was not open to the public. I was glad to have the opportunity to attend. Since I am just beginning to learn about seismology I think the most valuable aspect of the review was the exposure both to the review and to different aspects of seismology. I think I had vaguely heard about the Nuclear-Test-Ban treaty but knew nothing about it. The review was geared for people who already knew a lot about it but I did learn a few things. The Comprehensive Nuclear-Test-Ban Treaty was adopted in 1996 but has not yet been ratified. As I understood it, there is a lot research going on to try to identify a nuclear weapon test and establish procedures for a possible on site investigation.
Reading the posters was a challenge since there were so many terms and techqniques I either haven't heard of or don't understand yet. However, being able to talk to the author really helped to be able to understand the gist of the posters and was really interesting. There were a lot of posters on developing better velocity models for various regions (mostly in Asia) and one on a technique to be able to easily recognize a bolide (meteorite) through a technique that honestly went way above me but was cool to see the results. The workshops were rather difficult to grasp but I did learn a little bit about the Source Physics Experiment (SPE) and the different phases. It's crazy to think about, but I could have a PHD in time for phase 3 when they will drill down to where an earthquake occured and set off an explosion to be able to directly compare the two kinds of events. I think that's the most exciting part of the experiment!
My goal this summer is to both work hard and play hard, so now here's a little bit of the play part. I had met another Sandia intern outside of SNL at a meetup hike who also wanted to go visit the Grand Canyon. She also had a conference last week but because of my schedule this past weekend seemed to be the best time for me to go (before things really get crazy with my project when we collect the data in July). We scrambled friday to switch gears and get packed and on the road to the south rim. We got a late start but managed to see so many things. This trip was an exploration more than a vacation to a particular place. I am still amazed at everything we managed to pack into the weekend. I'll just put a bulletined list of everything we did here.
Well, I’ve had a successful initial go at using GMT to create a map of where my research will take place. It took a couple of days but I finally got the map projections right and the positions for all the labels just right. The cool thing is I’ve already got something that will go on my poster (after I put the maps together in another program)!
My data for the summer will come from 3 sources: a permanent installation of seismometers, an array of geophones and a sharp, pointed stick (I know, high tech, right?).
Why a stick?
The use of the stick is simple: poke the ground and see how far down it goes. When the pole gets stuck, you’ve hit permafrost.
About the permanent stations: H/V Experimental technique
There are currently 7 stations installed with Nanometrics Trillium Posthole Seismometers and Reftek-130 digitizers. These stations are set up in a triangular array with one station in the center (see one of the fancy maps I created in GMT). They’ve been taking data from the ambient noise since October.
I’ll be using a technique with this data called H/V Spectral Ratio Analysis to track changes in the depth of the active layer (see my profile). While H/V has been used to categorize the vulnerability of an area to shaking caused by an earthquake, using H/V for the depth of permafrost is a new technique.
Geophones: The ReMi Method
In addition to the data from the permanent stations, I’ll be collecting new data with the geophones. This will be an active survey that will, in addition to the pole measurements, act as ground truth for the experimental H/V technique.
A FEW TECHNICAL DETAILS
Normally geophones are used in refraction surveys but I’ll be using the ReMi (Refraction Microtremor) method. The data will be recorded in the time domain and then transferred to the p-tau domain. The p-tau domain is slowness (inverse of velocity) vs what is essentially time. In this domain noise as well as the reflected and refracted waves get stacked out so that just the direct waves are left. This data is then transferred to the p-frequency domain via Fast Fourier Transform (FFT) and then inverted until a velocity model is obtained that matches the data.
Tools of the Trade
I’ve already mentioned that I’m using GMT to create nice maps for visuals for my poster. I’ll also be using GEOPSY for the H/V analysis along with Matlab. I don’t know all the details just yet.
So the point off all this is try to see if this new technique will work for measuring depth to permafrost. If it works it will be a cost efficient way to map and track the permafrost layer through time. There are other geophysical techniques to map it, but these require multiple site visits if you want to track it through time. H/V could work on a daily basis and especially could be used as the US Array is installed in Alaska beginning this year.
Well, I've got a lot to accomplish this summer, and it is always good to set goals to make sure I'm on track to accomplish everything. You can read a description of my project on my profile page. I won't be collecting data for a few weeks so for the first third of the summer I'll be reading a lot to build my up my background knowledge. The topics I'll be reading about include permafrost, ReMi (Refraction Microtremor) and H/V (don't really know what that is at this point!). I'll also be attending a conference here in albuquerque which not only will help with building background knowledge but also help understand one of the major projects going on here at Sandia National Labs.
That brings me to another big goal for the summer: start to zero in on what it is I want to do in Geophysics. To accomplish this I'll be meeting as many people as possible and ask a lot of questions about what they do. There are opportunities to meet people informally here at SNL, the conference I'll be attending and also in an additional mentor program I've signed up for!
The second and last third of the summer will blend together. I'll be collecting data in Alaska and then it's crunch time.Once I get back I'll be working hard to:
1. Process all the data
2. Finish my poster
The goal is to have the poster finished by the end of my internship. To that end, I've already started on it by beginning to prepare a basic map of the location of my research for the poster. That may sound simple but I'm using a program called GMT (Generic Mapping Tools) and it is anything but simple! This is a tool that will come in handy in the future as well as it is commonly used for preparing presentations (or posters!)
Finally, looking beyond my time this summer as an IRIS intern, I'd like to identify some possible advisors I'd like to work under for graduate school. So, I need to come up with a list of 10 or so things I'll be looking for in a potential advisor and their research. This should be finished by the second third of my internship. My mentors will give me suggestions of people or places that might be a good match, then I will start to reach out to them either in the last part of the internship or right after. I would love to meet a few at the AGU meeting in December if possible!