Part of our project involves sending a small number of juvenile Chinook salmon to our laboratory in Georgia, where we can test their feeding behavior in a controlled environment varying a single variable, such as water velocity, at a time. (Needless to say, we are in close contact with Fish & Game throughout this process and maintain the required permits in both states.) It was near the end of the Alaska field season before the lab was ready to receive fish.
A driftwood fire helped me and my field assistant (this time, my wife) keep warm while the minnow traps soaked in a logjam where we saw some juvenile Chinook salmon.
It didn’t take long for the traps to gather enough fish.
I packaged the fish very carefully in multiple heavy-duty aquarium bags, filled with oxygen, and surrounded by ice packs in insulated seafood shipping containers. Then I sent them on their way via Alaska Airlines, taking the same route to Atlanta that any human passenger would. In less than 24 hours they were acclimating in their new home.
As of this writing, four months later, they are still alive and well in the lab and providing valuable data in ongoing feeding experiments.
From August 23-25, 2014, my dad and I camped out on the Richardson Clearwater to try to collect 3-D video foraging data on adult grayling using the GoPros. We found one of the only good campsites on the river, although it was well hidden.
The weather and scenery were amazing the whole time.
In addition to filming, we also caught some fish to compare diet samples with the ones we caught a month before.
The video we shot of grayling turned out to be disappointing. The fish were mostly in water fast and deep enough that it was difficult to place the cameras nearby. When we did, our presence spooked the fish, and they eventually returned to the general area but not to their specific original positions in front of the cameras. Unlike Chinook salmon I’ve filmed previously with steady success, the grayling in this stream seem to range over an area large and uniform enough that when they’re disturbed they have no strong urge to return to the same spot anytime soon.
Although this stymied our efforts to shoot good footage on this trip, it was very valuable going into the off-season before our first real field season. It told us that we need to figure out a way to place the cameras in faster, deeper water than expected (i.e. devise a new anchoring system) and that we need to find a way to keep the cameras running for a very long time, so they’re ready when the fish eventually do return to suitable positions. This is exactly the kind of thing we needed to figure out while we still have the whole winter to engineer our solutions.
August 22, 2014 was a day trip to one of our Chena Chinook study sites to test the second of our two camera systems. Much less complicated than the big Nikons we tested and troubleshot previously, the GoPros were able to shoot pretty good video right away. (My experience filming Chinook salmon also made it easier than testing the big cameras on other fish.)
I had more family visiting and enlisted my father (left) and father-in-law (right) to help assemble the 3-D video calibration frame.
We were also joined by Bill Carter from the U.S. Fish and Wildlife Service, who wanted to learn how to deploy the 3-D video system with GoPros for use on a sheefish project later in the year.
It’s great that we had so much willing help for equipment testing during this pilot field season. Next year, we will have paid technicians.
In late August of 2014 my dad, a recently retired Wisconsin DNR fish biologist, flew up to Alaska to help with a week or so of intense fieldwork. We relied on volunteers this summer because we didn’t have enough fieldwork to justify hiring a technician, but it was very valuable to have him up here and be working with the same person for several days in a row.
On August 21st, we tested our main underwater video system in a relatively deep pool in Panguingue Creek.
The system consists of a side-by-side pair of Nikon DSLRs in underwater housings, with HDMI connections wired to Atomos viewer/recorders, so we can see what the cameras are seeing at all times and make sure they’re placed correctly.
We use 2 cameras instead of 1 because the side-by-side views allow us to use 3-D video measurement methods with the VidSync software and capture the exact 3-D coordinates of every action of the fish we want to measure. To enable those measurements, we also film a calibration grid, pictured below.
We didn’t get any good fish footage out of this trip, but that’s to be expected from the first test of complicated new equipment. We did learn a great deal about the quirks and limitations of our technology, and developed a good understanding of what I would need to buy, tweak, calibrate, or hack to make everything work smoothly when we begin real data collection next summer.
On August 7, 2014, I spent a long day on Panguingue Creek scouting, mapping, photographing, and taking notes on more than 30 pools to help identify the best representative study reaches for the remainder of our work. I brought along my fly rod and an underwater camera system to sample dolly varden and grayling and note their distribution and relative abundance throughout the creek.
The previous day I had briefly investigated (and ruled out) other possible study sites to the south. I car camped near Panguingue Creek along Stampede Road, where a wolf trotted past as I got ready to sleep, and a nice view of Mount Healy in Denali National Park greeted me in the morning.
The water on Panguingue Creek was high and tea-stained from this summer’s incessant rain, complicating the already difficult wading on its substrate of slippery boulders and cobbles.
Much of the stream flows too fast or too shallow even for the dwarf dollies we’re studying.
Grayling were mixed in with the dollies, and they dominated the larger pools. They were all very small compared to what we see in the larger rivers, but it’s good for our study to test our foraging models across a wide range of fish sizes and species.
I caught enough dwarf dolly varden to get a good feel for their distribution and the sections of the creek where we can best study their behavior and habitat. They are gorgeous fish, reminiscent of eastern Brook Trout, and even the very largest adults (like the 9.25″ male pictured below) retain their juvenile parr marks (dark vertical bars on their side).
At one point, far from the road, a massive beaver dam blocks the creek… but there were fish both above and below it.
I returned to the car after more than twelve hours on the stream and an hour-long bushwhack out, with exactly the data we needed to focus the rest of our study productively.
Testing a new understanding of drift-feeding fish behavior.