Salmon are everywhere: why salmon matter and how metapopulation perspectives can help

            Salmon are everywhere. That’s the problem; and the beauty of salmon.  Salmon species use every part of a river system from headwaters to massive rivers out into the ocean.  That’s why salmon are the “canary in the coal mine” for the health of our ecosystems.  When the whole system is healthy, salmon thrive. But as parts of that system are degraded or connections lost, salmon that might be “everywhere” aren’t.  

            From wilderness areas and managed forests through small rivers meandering across farm-filled valleys into large rivers that pass through cities and past dams, salmon touch upon a wide array of human enterprise. That’s why salmon are everywhere in the newspapers as well.  Perhaps not on the front page, but if you start noticing the not-front-page news, you begin to notice that that efforts to save salmon have wide-reaching consequences.  

            While populations of Pacific salmon may be healthy across much of their range, (in Alaska and Russia), populations along the West Coast have seen sharp declines.  These declines led to listings of species as threatened or endangered under the U.S. Endangered Species Act during the 1990s.  These listings are leading to government regulations and an enormous mobilization toward the goal of recovering salmon.  Whether you realize it or not, you’re contributing to this effort when you pay your power bills.  From 2000-2009, the Bonneville PowerAdministration that administers the dams supplying our electricity have spent $108–178 million/year on direct expenses toward fish projects and have incurred $12.6–397 million/year in lost opportunity costs due to mandated changes in dam operations to assist salmon passage.  That’s a lot of money!  

            Past the front page, you begin seeing more stories of the rippling effects of salmon.  This past fall, two different dams, once impassable for salmon, were removed in rather spectacular fashion.  In January, the Oregonian newspaper trumpeted theColumbia River Trust’s acquisition of a 920acre farm adjacent to the river; the dykes that once protected the farm will be removed and flooded fish habitat created.  Last week, a judge threw out the state of Oregon’s river temperature standards approved by the EPA.  Why?  The temperature standards were deemed insufficient for salmon that need cold water.  It’s a decision that could significantly impact farmers, businesses and timber operations. 

      

             Outside the Pacific Northwest, places that still have abundant salmon are fighting to keep it that way.  Bristol Bay in Alaska is home to the most valuable salmon fishery in the world.  Bad news for salmon: a massive deposit of gold and copper was found in the headwaters and one of the biggest open pit mines in the world has now been proposed for.  An Olympian struggle has ensued between residents and the proposed developers.

 

            Not only is Bristol Bay home to an invaluable sockeye fishery, the rivers feeding the bay have been the focus of one of the best, long-term monitoring programs of salmon population dynamics.  And what has been learned in Bristol Bay about salmon metapopulation dynamics is providing critical insights for conservation efforts in the Pacific Northwest.

            But first: metapopulations? What are those? If you’re a basketball fan and you hear the prefix “meta,” the first thing that might come to mind is the basketballplayer formerly known as Ron Artest. I’m not sure if the newly minted Mr. Metta World Peace cares at all about salmon, but naturally functioning metapopulations can also contribute to harmony and stability.

            A “population” is a group of organisms of the same species that interbreeds given their overlap in space and time.  For salmon, a population is defined by the river in which the salmon spawn and the timing of their spawning.  A “metapopulation” is a collection of populations which are partially isolated but can exchange individuals.  Within a metapopulation, individual populations may go extinct but these locations may be recolonized. Metapopulations may display emergent properties such as stability distinct from the properties of the contributing populations which may themselves be unstable.

            As a species, salmon are perfectly suited to be understood in terms of metapopulations.  In a 2007 concept paper, Schtickzelle and Quinn outlined how salmon fit the three criteria for the metapopulation concept: habitats associated with populations must be discrete (you can’t have habitats and their associated populations grading seamlessly from one to another); while habitats must be discrete, there must also be dispersal between populations to allow for recolonization; finally, the dynamics of local populations must not be synchronized, they can’t behave identically.  

            Salmon meet all of these criteria. Populations are associated with discrete habitats, stretches of gravel beds beneath shallow water suitable for spawning separated by miles of unsuitable river habitat.  And while salmon are famous for “always” returning to the birthing stream, thankfully, some make mistakes.  They take a wrong turn.  That’s called “straying.”  It sounds bad, as if these salmon had failed the one fundamental test of being a salmon, but it’s these prodigal salmon that may be the recolonizers of populations about to blink out.

            Back to Bristol Bay.  The University of Washington investigators and affiliated researchers who have been gathering this data recently published articles in Nature and Conservation Letters synopsizing key findings and implications of their past decade of research (Schindler et al. 2010, Moore et al. 2010). They use an analogy from economic theory, the “portfolio effect.”  That is, a portfolio of investments performs best and minimizes risks when its holdings are diverse.  

            Bristol Bay is fed by nine major rivers.  Each of these rivers has monitored salmon returns for more than 50 years.  And in addition to monitoring at river outlets, one system, the Wood River, has seen monitoring of individual streams.  In 2003, Ray Hilborn and colleagues documented an interesting phenomenon: key rivers such as the that once dominated the Bristol Bay fishery through much of the 20^th century declined precipitously in the 1990s whereas other rivers that had been minor contributors grew by leaps and bounds in salmon productivity.  However, while all of these rivers were going up and down, the Bristol Bay fishery as a whole, the metapopulation, remained stable. One might expect salmon populations within this ecoregion to behave similarly since they are all experiencing a similar climatic regime.  But the investigators point to the “biocomplexity” of salmon life histories: some spawn on sandy beaches, others in large cobble beaches and others in gravel streambeds.  Each river is unique and associated salmon populations with their unique life histories respond differently to region-wide changes in climate.

            Rogers and Shindler (2008) pushed this further by demonstrating that even within a single river system, the Wood River, there was no more correlation in annual salmon productivity among streams close to each other than between the Wood River as a whole and rivers on the other side of this vast bay.  Asynchrony reigns!  And that’s a good thing.  If all of the populations went up and down together at the same time, the metapopulation would be vulnerable to a crash.  Biocomplexity and asyncrhrony confer resilience, a portfolio effect.  

            Bristol Bay provides an example of how a natural metapopulation is “supposed” to work.  The worrying thing is that in the highly altered systems of the Pacific Northwest, evidence suggests that salmon populations are becoming more synchronized with time (Moore et al. 2010).  Dams that homogenize river flow regimes and straying hatchery fish may be facilitating this synchronization.

            Focusing on the Willamette River basin, Fullerton et al. (2011) explore the effects of dams, habitat degradation and hatcheries on metapopulation structure by using a modeling method called graph analysis, a formal way to assess what might otherwise look like doodles and back-of-the-napkin drawings.  Circles represent populations and their size.  Lines between circles represent connections among populations (immigration/emigration) with line thickness representing the degree of exchange.  The overall structure of the metapopulation is represented in graphical space with the x axis representing increasing connectivity between populations and the y axis representing increasing variance in population size. 

              Graph models reveal historic Chinook and steelhead metapopulations as falling between “classic” or “Mainland-island” metapopulation structure, that is, having moderate connectivity and either roughly equal population sizes (“classic”) or highly variable population sizes (“Mainland-island”).  Remember that key criteria for metapopulations are discrete habitat patches and some limited dispersal between patches, criteria that would be characterized as moderate connectivity.  The modeled dam scenario consistently reduced connectivity between populations leading to a “nonequilibrium” metapopulation state.  “Nonequilibrium” is not a good thing:  with less connectivity, vulnerable populations are less likely to get reinforcements.  In contrast, in the hatchery scenario, the high straying rates for hatchery fish led to increased connectivity.  The resulting structure might be described as “patchy” or “panmitic” with so much flow between populations that the metapopulation acts as one big population with everyone in step.  A model that combined dams, hatcheries and habitat degradation, our current state of affairs, displayed metapopulation structure with low connectivity and variance in population size, again the “nonequilibrium” state.

            Why are metapopulation perspectives important?  A metapopulation perspective can help us assess issues like connectivity and asynchrony among populations, permitting evaluation of whether our actions are moving salmon metapopulations toward or away from resiliency and stability.  Metapopulation perspectives are being used by the government agencies to shape strategies for recovering salmon; metapopulation analyses will be used to determine when a salmon metapopulation listed under the ESA is “out of the woods” and can be de-listed as no longer threatened.  Conservation organizations are also using metapopulation perspectives to prioritize populations for conservation, the portfolio in which they will invest.  

            Salmon are “everywhere” throughout our river systems.  But unless we’re living in the natural wonder of a place like Bristol Bay, we rarely see salmon, and it’s unlikely we have a sense that salmon are everywhere.  In places like the Willamette Valley, populations are on the ropes due to a myriad of human alterations.  However, using lessons from natural systems and metapopulation perspectives to develop restoration strategies, it’s possible that salmon populations may eventually rebound and metapopulation resiliency restored.

Images

1.  Ben Knight. http://ourbristolbay.com/photo-gallery.html

2.  Ryan Peterson. http://pool32mag.blogspot.com

3.  from  Schindler et al. (2010)

4.  from Hilborn et al. (2003)

5.  from Fullerton et al. (2011)

6.  Steve Cowgen, Michael Milstein. The Oregonian. http://blog.oregonlive.com/news_impact/2008/10/dam.jpg
 

References

1. Fullerton, A.H., S.T. Lindley, G.R. Pess, B.E. Feist, E.A. Steel, and P. Mcelhany. 2011. “Human Influence on the Spatial Structure of Threatened Pacific Salmon Metapopulations.” Conservation Biology.

2. Hilborn, R., T.P. Quinn, D.E. Schindler, and D.E. Rogers. 2003. “Biocomplexity and Fisheries Sustainability.” Proceedings of the National Academy of Sciences 100 (11): 6564.

3. Moore, J.W., M. McClure, L.A. Rogers, and D.E. Schindler. 2010. “Synchronization and Portfolio Performance of Threatened Salmon.” Conservation Letters 3 (5): 340–348.

4. Rogers, L.A., and D.E. Schindler. 2008. “Asynchrony in Population Dynamics of Sockeye Salmon in Southwest Alaska.” Oikos 117 (10): 1578–1586.

5. Schindler, D.E., R. Hilborn, B. Chasco, C.P. Boatright, T.P. Quinn, L.A. Rogers, and M.S. Webster. 2010. “Population Diversity and the Portfolio Effect in an Exploited Species.” Nature 465 (7298): 609–612.