I feel like I experienced a miracle last week.

Possibly I am throwing around the word “miracle” because I’ve got Herb Brooks on my mind (thanks to my fellow grad student and FW intramural soccer coach Matt who is obsessed with that guy). Or perhaps that is actually what happened.

Let me set the stage. Will and Otis, our two Seagliders, were deployed off the coast of Newport, for what should have been a brief, straightforward test of their passive acoustic systems before they were shipped off to the Gulf of Mexico for a project there. Of course, that would not be as exciting of a story if it all went as planned.

I can’t remember how much I’ve talked about it before (I looked it up…try here and here), but basically, the way these gliders work is they go out and dive in the ocean, listen for marine mammals, and every time they surface they call in to a basestation, offload their location and some log files, and continue on their way. Well. Otis (SG608) did exactly that. It was his first flight with us and all went smoothly, from a piloting stand point. Will (SG607) on the other hand….well, he went rogue. And I don’t mean to the brewery.

Will stopped calling in after only 5 dives. Did I tell you this was my first “solo” piloting of the gliders? Yes, I was sort of freaking out.

But what happened the next few days is not important (I blacked it out so I can’t tell you because I don’t remember).

The point is….WE FOUND HIM!!!!!!!!!!

So (1) the miracle part: Let me explain the chances of finding Will. Best case scenario we were searching in about a 1 km radius of a point we THOUGHT the glider would be diving to. Worst case, it was floating at the surface and had drifted who-knows how many miles offshore. But lets complicate things. Glider at the surface, great, easier to spot. Glider continuously diving = glider down for 1 hour 40 mins, at the surface for 20 mins. So lets say we ARE in the right place. Well then it has to be the right time, and you better spot the thing during that 20 mins and get the boat over there before it goes back down for an hour and 40 mins and pops up somewhere else in that 1 km radius. Lets add in some wind waves (We are 35 nm offshore here) and some fog. And this is the image you are looking for:

surface example


(2) the waiting part. Will was missing for 4 and a half days. That doesn’t seem like that long. But when everytime your phone beeps that you get a text message and your heart jumps thinking maybe its the glider, that is a long 108 hours. But that is a lot of what we had to do. This was exacerbated for me because I had to stay on land during the search trips. I had to be at my computer in case we heard from the glider and I could give updates on GPS locations or timing. This was a new experience for me. I’m not real good at sitting still and waiting.


(3) the teamwork part. To me, the greatest outcome of the whole thing. There is NO way we could have found Will without all hands on deck, without awesome grad students and scientists who went out to look (Laurie, Niki, Erin, Theresa, Curtis, Alex, Haru, Matt, Dave), Anatoli and Steve for answering my piloting questions, a chartered fishing boat (ok…we paid them, Sara thanks for coordinating), TWO trips out, the people at iridium for putting up with my incessant phone calls,  the dolphins that swam by the boat and provided moral support, Sharon and Holger for telling me not to freak out…I could go on. (and I’m SO SORRY if I am forgetting someone)


Today’s blog serves two purposes: (1) inform readers what’s going on in my research world and (2) an educational piece sharing some of my trials and tribulations with ArcGIS this week.

Right now we are preparing to deploy a glider up in the Gulf of Alaska, near Homer, in the US Navy’s Gulf of Alaska Temporary Maritime Activities Area. The glider’s acoustic system samples at 194 kHz allowing us to listen for vocalizations up to 97 kHz, which covers almost all cetacean species in the area, except porpoises which vocalize at really high frequencies (>150 kHz, we recorded them with a different glider though!).

I won’t be actually going out to deploy it or piloting this glider – we are collaborating with some folks from the University of Washington – but I am responsible for putting together the glider’s track and coming up with track points that are 5 km apart so we can set our ideal path for the glider. Why am I responsible for this, you ask? Well because I took an introductory GIS (Geographic Information System) course so….this becomes my job.

For those of you that have worked with GIS, you understand there is a STEEP learning curve. It may be one of the least intuitive programs on the planet. But, it is incredibly powerful for not only making maps but for spatial analyses too, so I am super happy to have learned even a tiny bit about it and get to learn more every week.

Well I’ve made these maps before for Guam and Hawaii, so Gulf of Alaska, easy peasy! I’m finally starting to remember how to make the track from the initial way points, then turning the track into 5 km spaced points. But, news flash! The earth is round. And measuring things at higher latitudes gets weird/complicated/annoying/inaccurate/etc.

So this week (really the last two days) I taught myself about projections in ArcGIS. Projections are basically trying to show our round, 3D Earth in 2D. At the equator this isn’t so bad, but up (and down) by the poles things can become really distorted.

Take this image of the US for example. Depending on what projection you use, it looks slightly different! And those differences are more pronounced the further north you get. So by the time you get to Alaska…well, you’ve got to do something about it.

Fortunately, lots of people have made hundreds of projections for different areas and different spatial scales that reduce distortion, either in area, distance, angles, etc.

So then all I had to do was find the one I needed (this took much research and trial and error), then redo all my mapping/measuring/GPS coordinate extracting steps on a correctly projected map. You know, once you make sure all parts of your map are in the same projection, that the data frame has the right projection, and that you saved it every 5 seconds in case it crashed. Once I got past the frustration, I ended up pretty proud of myself, and now I learned my lesson for next time: only work in areas near the equator.

Want to know what projection I used? Of course you do. The lovely Alaska Albers Equal Area Conic! Sorry I can’t share a picture of the pretty map…I’m not sure I should show you where our glider will be I don’t want anyone going up to Alaska and stealing it.




Greetings extensive readership!

In the midst of the summer and early fall when I was traveling a bunch and doing field work, I remember thinking how nice the term would be to be in one place for a while and get some analysis/other work done. What I didn’t realize was how unexciting my life would be for blog posts….

I guess excitement depends on your interests, though, because for me there have been SOME exciting moments standing in front of my computer. I’ve spent the last month putting my master’s on hold, instead analyzing acoustic data collected from one of our gliders that was deployed back in March, and then deploying and analyzing another glider all within the month of October. Want to see what I found? Good. I was going to put in the images anyway.

Here's a Stejneger's beaked whale click.
Here’s a Stejneger’s beaked whale click. The top image is a long term spectrogram, or LTSA, that shows 15 minutes. All the little bits around 50 kHz are beaked whale clicks. The middle spectrogram just shows one click during a fraction of a second, and the bottom shows the wave form, or the amplitude of the click.

From the March deployment, the excitement came in the form of TONS of beaked whales. Like so many. Like all the time. Including the super weird looking Stejneger’s beaked whale (Mesoplodon stejnegeri). I can tell the species by what frequency the click is at, how much time there is between clicks (inter click interval, aka ICI fyi), and the duration of the click. They are all unique features for this species of beaked whale, which I know thanks to other people confirming that by combining visual and acoustic data like was done by theses lovely folks at Scripps.


Here's two porpoise species detected together - Dall's porpoise and harbor porpoise
Here’s two porpoise species detected together – harbor porpoise (the two higher frequency red specs) and Dall’s porpoise (the middle, slightly lower frequency). All together a bunch of those clicks make up that light blue section in the LTSA on the top.

The March deployment also brought excitement through porpoise recordings! Did I mention that glider was the first of its kind to record ultra high frequencies? We used a 394 kHz sampling rate, which means we could detect vocalizations up to 196 kHz, which is where porpoise and a few other odontocetes (toothed whales) vocalize. Most equipment doesn’t sample that high (memory gets filled too fast) so this was pretty neat-o. I’m a big fan of looking for these ultra high frequency encounters because they are so obvious in the upper part of the LTSA, far above the background noise.

And like I mentioned, I did go out in the field one day. We deployed one of our new gliders for a few days just outside of Newport in early October, and I went out on the recovery. I took this one super exciting picture of these gulls on the back of the ship. You’re welcome.

Piper helped Holger and Alex prep Will the glider before he got deployed in early October.
Piper helped Holger and Alex prep Will the glider before he got deployed in early October.
These gulls agree with "no excitement November". Until I threw pistachio shells over the side. Sadly this is the only picture I took of the whole glider deployment.
These gulls agree with “no excitement November”. Until I threw pistachio shells over the side. Sadly this is the only picture I took of the whole glider recovery.