Does replacing wood in logged streams produce more fish?
Lessons from the Twelvemile Creek Fish Monitoring Project
(Report downloads are available at the end of this page)
Our streams in Southeast Alaska naturally have lots of trees, directly in the river and in the surrounding floodplain forests. And for the past several thousand years native fish populations have adapted and evolved to that reality. We know streams have big trees because of our direct experience walking through these forests and streams. We also know this because we have counted and measured these trees. In this case “we” is used to collectively refer to all of us that care about our fishery resources, including communities, agencies (like the U.S. Forest Service and others), commercial fishers, businesses, etc.
Most of us are also familiar with the story of how, before the rules changed in 1990, logging occurred right up to the stream banks, and in some cases trees were even pulled directly out of streams (as in the above photo). We know this is not a “natural” condition, and the objective for nearly all in-stream restoration projects is to get these streams back to a condition that is more like the “natural” condition (often termed the “reference” condition) of lots of trees. While there are certainly other considerations and caveats to that statement, it is generally true. The photo below provides an idea of what a typical restoration project might look like, in this case Twelvemile Creek before and after restoration in 2013 (from Twelvemile Creek Monitoring Summary, USFS Tongass National Forest, March 2015 – available at this link).
It is a routine practice to measure wood (and many other physical measures) in streams before and after an in-stream restoration project to make comparisons to the “reference” conditions. These comparisons dictate the restoration design and help to evaluate the level of project success. In general, monitoring studies have shown that replacing wood helps our streams get closer to the reference condition. In the case of Twelvemile Creek, wood placed in the stream even survived 2 major flood events. The wood shifted around, some pieces were lost to the ocean, and others piled up in log jams. Wood naturally moves during storm events and good restoration designs allow for that.
However, we eat fish, not habitat. We are fish-centric people and naturally we want to see a direct result in fish numbers. So understandably there is a very strong push to understand how these projects actually affect fish. Sometimes it is not enough to simply accept that more natural habitat conditions (the “reference” condition) are “good” for all fish species.
Increasing our understanding of fish responses to restoration, and how to do it, was the objective of the Twelvemile Creek Fish Monitoring Project – a partnership between the Tongass National Forest and Sitka Conservation Society. Coho salmon smolts that were going to the ocean were counted and tagged, and marine commercial harvest was tracked over 3 years. Returning adults were counted over several generations of fish. These actions took a lot of time and a lot of money.
However, even these efforts were insufficient to track changes in fish production in the Twelvemile Creek watershed. We worked closely with restoration scientists from both Alaska and the Pacific Northwest (where they have done a lot more restoration work than we have) to evaluate this project and determined that a more intensive effort was needed to evaluate a fish response to restoration (here is a link to the report). To evaluate a fish response to in-stream restoration, the general recommendations included (“general” because it depends on which species and the size of the watershed):
- Commit to 10 years of monitoring, 4 before and 6 after the actual restoration work. This length of study is necessary to account for all the variability in the life of a salmon
- At the same time, monitor a stream where no restoration work has occurred, for comparison
- Carefully design the monitoring plan to measure the most useful metrics, that will help answer a specific question, such as “will restoration increase coho smolt production by 50%”
It is easy to be critical of our monitoring efforts on Twelvemile Creek, yet the challenge for our project is shared throughout the Pacific Northwest and other regions. A recent study found more than 400 papers addressing stream restoration effectiveness, yet only 4 report on fish responses and only one was a robust study that is applicable to Twelvemile (Roni et al. 2014. NOAA Tech. Memorandum NMFS-NWFSC-127). That particular study did show a strong positive fish response to in-stream wood replacement.
Therefore, if we want to better understand the fish response to in-stream restoration we will need to collectively commit to long-term projects (minimum 10 years), and pony up the cash to get it done. Otherwise, we can focus on habitat changes and continue to learn about habitat-fish interactions in our streams.
For example, we did learn that each year the Twelvemile Watershed produces approximately 40,000 to 60,000 coho smolt, and that the watershed contributes approximately 2,600 coho to our commercial and sport charter fisheries.
With the Twelvemile Creek Fish Monitoring Project, we also had the opportunity to engage local communities on Prince of Wales Island to participate in monitoring. We were reminded that the people in these communities, that rely upon fish for their livelihoods, are keenly interested in their streams and want to be involved in their management. We worked with over 90 school kids from Hydaburg, Klawock, and Craig in “Stream Team” (a program run by the Alaska Natural Heritage Institute), where kids collected monitoring data and interacted with fish biologists and other professionals. We also developed an intern training program that integrated 3 generations of fisheries people! With this program, high school students from the rural communities of Port Protection, Naukati, Thorne Bay, Kasaan, and Klawock worked on site with professional fish biologists and the Fisheries Technology Program at the University of Alaska Southeast to learn practical skills and gain exposure to fisheries careers. Both these activities help spread awareness of the challenges and opportunities of stream restoration and fisheries management, and inspire local young people to pursue careers in these fields.
So we can sum up the lessons we learned as follows:
Lesson 1: Communities surrounding restoration projects are keenly interested in project outcomes and engaging in the work
Lesson 2: Measuring how fish respond to restoration is very hard and costly (and this sounds like a broken record for most of us in this line of work).
Lesson 3: Measuring habitat changes (for example, counting wood, pools, or gravel) is relatively easier and changes have already been noted.
Lesson 4: In-stream restoration projects assume that streams closer to the “reference” conditions are better for fish – have greater diversity, increased resilience, and therefore maintain better habitat and biodiversity over the long term.
Lesson 5: Engaging in restoration requires that we accept the uncertainty in the above assumption while at the same time commit to continue learning and adapting restoration practices appropriately.
At the close of this project, we contracted with Phil Roni of Cramer Fish Sciences to conduct data analysis, report results, and review our progress towards meeting objectives. The report can be downloaded here:
The Tongass National Forest conducted an independent study of physical habitat changes to the in-stream restoration work done at Twelvemile Creek. That report can be downloaded here: