By John McMillan
We should not be surprised by steelhead and salmon rushing upstream to pass former dams. That is their nature — to push boundaries, access new habitat. Unbuild it, and they will come.
And come they have in Washington’s Elwha River, where in 2014 two old dams were taken down. Not only have we observed hundreds of winter steelhead redds above the former dams in the Elwha, lately there have also been observations of numerous summer runs. This is particularly cool because summer steelhead were nearly (and possibly even completely) eradicated in this river.
All these steelhead above the former dams is a sign of success in terms of river and fisheries restoration. However, it has also posed a big challenge for scientists tasked with figuring out exactly where the steelhead were spawning, particularly early in the dam removal process. Why? Because there were many thousands of resident rainbow trout above those dams.
As one of the scientists participating in this effort, I quickly realized it might be easier to sample steelhead offspring during summer, when visibility was good and flows were lower. The challenge was distinguishing whether the fry were from rainbow trout or steelhead.
Why not focus on finding adult fish? Three reasons: a lot of ground to cover, not a lot of adults in the system early in the recolonization process, and poor stream visibility in winter and spring.
I went to work with colleagues to sample fry from resident rainbow above the dams and fry from steelhead below the dams (the paper published from this work is here: http://www.tandfonline.com/doi/abs/10.1080/02755947.2015.1074965 ). My hypothesis was that we could distinguish steelhead and rainbow fry using one of several metrics.
First, we could kill fry and extract their otoliths (ear bones). The chemistry of the ear bones can be used to tell whether the mother went to the ocean or not.
But killing a bunch of fry was not really an option for steelhead in the Elwha because we needed all the fish to help recolonize the habitat above the dams. So we turned to other options. The first was locating redds. While we might not see the adults, we could see redds in tributaries and sometimes in the mainstem Elwha once flows dropped and visibility improved. Steelhead are much larger than rainbow trout, so they should excavate larger redds.
The second was recording the size and density of fry. Female steelhead, being larger, tend to produce larger eggs than smaller rainbow and tend to spawn earlier in the season. Larger egg size and longer growing periods suggest steelhead fry should be larger during summer than rainbow fry.
Lastly, given that female steelhead carry thousands more eggs than a smaller female trout, we also predicted we would find higher densities of fry where steelhead spawned compared to where trout spawned.
To test these predictions I extracted otoliths from steelhead fry that unfortunately died during electrofishing, measured steelhead and trout redd sizes and the size of gravel in the redds, and then measured the length and density of known steelhead fry and rainbow trout fry prior to dam removal. All of this work was conducted prior to dam removal so we could ensure that we were only sampling steelhead below the dams and rainbow trout above.
Did any of those methods work?
Yes, testing otoliths worked as predicted. We could easily distinguish steelhead fry from trout fry because of the chemical signature. Easy, but not a sustainable technique for this project.
Yes, recording and comparing redd sizes also worked. Steelhead excavated much larger redds in larger gravel than smaller-sized trout. In fact, the difference was so significant that we could easily measure the length and width of each redd and denote from those measurements whether it was from a steelhead or trout.
But you can’t always see redds. So we thoroughly tested the size-and-density of fry method, which proved somewhat less conclusive. Yes, steelhead fry were a bit longer than trout fry, but not significantly enough to consistently distinguish between the two. However, fry densities were two to three times higher in areas where steelhead spawned than in areas where only rainbow trout spawned.This was critical because it meant we could use snorkeling or electrofishing to measure densities and a model to predict whether those fish were from steelhead or rainbows.
Ultimately we found that both redd size and density of fry were useful in distinguishing where female steelhead spawned compared to where female rainbow trout spawned. Since the dams came out in the Elwha the redd size metric has been most helpful for monitoring recolonization because it is less labor intensive to conduct redd counts than it is to sample juveniles. Redd counts have their limitations due to the influence of water clarity and flows, and the fact that it is generally not possible to sample entire watersheds. A combination of redd counts and surveys to monitor changes in fry densities has proven to be an effective method to document the return of steelhead above the dam sites.
This study highlights the considerable benefits of dam removal for fish and rivers. It also underscores the challenges for science when species with different life histories occupy the same areas above and below a dam. In such situations it can be difficult to pinpoint where fish are recolonizing. All things considered, this is a good “problem” to have.