SpruceRoots Magazine - February 2000


Dog, Salmon, Rolling, Rolling, Rolling

by David Loewen

Research on salmon has been conducted close to 100 years, but not until the last decade has one of the most important links been discovered between spawning salmon and ecosystems; a link which is obvious to many, especially gardeners. The return of salmon to their natal stream (place of hatch) is vitally important to the function of ecosystems on the coast and has been for thousands of years.

While preparing a report for some recent stream assessment work in Rennell Sound, I discovered some rather surprising information. A comparison of salmon escapements by decade from 1950-1999 in six streams draining into Rennell Sound showed dramatic declines in spawning salmon. The most astounding, a 97% decrease in spawning chum salmon, in four decades!

Salmon escapement is the number applied to adult spawning salmon which manage to return to their spawning stream. Salmon escapement is determined by a fisheries officer or creek guardian walking a stream and estimating by species, the total number of spawning salmon. These numbers are then published in a catalogue produced by Department of Fisheries and Oceans Fish Habitat Inventory and Information Program (FHIIP). It is widely recognized that escapement estimates are obtained with a variety of techniques, so differences in enumeration methods from year to year can affect the reliability of data. Therefore, escapement data is only a general indicator of species composition, distribution, and abundance.

Here on Haida Gwaii, escapement numbers are generally concentrated on larger, easier to count salmon runs such as chum and pink. Coho are difficult to enumerate as there are far fewer fish in the stream, they are quite secretive hiding under cut banks and coho migrate much further upstream than chum and pink.

Table 1. A comparison of escapement counts in Rennell Sound (17 streams) by decade, from 1950-1999
   Coho Pink Chum
 1950-59  4,648 374,875 222,650
 1960-69  1,935 169,225 78,180
 1970-79  513 438,095 94,401
 1980-89  649 216,838 46,791
 1990-99  1,502 122,764 18,394
 Decline in 50 years  68%  67%  92%

In four decades, salmon populations in Rennell Sound have declined dramatically. Coho by 68%, pink by 67%, and chum by 92%. A comparison of total salmon compostition shows that approximately 85% or 14 of 17 streams in Rennell Sound are comprised of over 90% chum. An important and interesting statistic why?


Fishing and logging, two former economic pillars of the coast, have dramatically reduced salmon numbers and despite nearly a century of research, not until the last decade has the destination and importance of salmon derived nutrients been discovered. The main gist in this research has been tracing nitrogen (N) and carbon (C) derived from salmon carcasses, through the ecosystem. Yes, those smelly, stinky, maggot infested salmon carcasses that dogs love to roll in are very much a keystone of the ecosystems they rot in.

Research conducted by Dr. Tom Reimchen in Gwaii Haanas and other regions of coastal BC which still have intact watersheds, has traced the importance of salmon carcasses to riparian vegetation (i.e. the massive spruce trees once found along Haida Gwaii streams) and all the little critters surrounding streams (eagles, ravens, martens, and even ground beetles). One of the most important individuals in this chain is the bear, as the amount of salmon carcasses and salmon waste they transfer from the stream to the surrounding environment is staggering.

Over a 45 day period in Bag Harbour (Gwaii Haanas) where approximately 5800 chum return, each bear transferred close to 700 salmon (1600 kg) of salmon from stream to forest. Within 2 metres of the stream, over 4000 kg of salmon biomass per hectare was identified. Just imagine a garden with this much fertilizer. Analysis of nitrogen isotopes (N15 - only found in marine environments) in riparian vegetation such as ferns, salmon berry, hemlock needles, found approximately 13% of nitrogen in plant tissue to be salmon derived. Analysis of growth rings in trees located near major spawning areas and trees located completely away from salmon areas, showed a direct correlation with years of increased spawners. For example, years of large chum runs correlated with growth spurts in the tree.

In other studies conducted in the United States (Bilby, Bisson, etc), N and C isotope tracing demonstrated up to 40% of N and C derived from eggs and carcasses of salmon were present in tissues of juvenile salmon. Significant amounts of marine N and C were also found in other fish, invertebrates, and riparian vegetation. In a comparison between one stream with carcasses deliberately added and one stream without, juvenile salmon showed increases in numbers, size and condition. Analysis of juvenile stomach contents while carcasses were present, demonstrated approximately 39% of contents was derived from eggs and carcass flesh.

Some scientists suggest that current salmon populations are 95% less than historical returns. What does this mean for an area such as Rennell Sound where the influx of chum carcasses has been reduced by approximately 92% in only four decades, in streams composed of 90% chum. What was the historical abundance in Rennell Sound? What are the long-term impacts on these streams losing over 90% of the marine derived nutrients they relied on for thousands of years?

Some hypothesize that the carrying capacity (the maximum population a given environment can sustain) of coastal old-growth forests and streams has been greatly reduced. If this is true, the diversity of fungi, invertebrates and vertebrates must be much lower. Many of the impacts on stream environments and surrounding ecosystems are yet to be seen. However, one of the impacts on vertebrates is quiet plainly visible. Ask a local commercial fishermen how much fishing they've done over the last few years.

There is no question that salmon populations have declined; the question is by how much? As fewer rotting carcasses are available for all critters we have a self-perpetuating cycle of lower and lower productivity in an ecosystem that includes, lest we forget; us.

Bilby, R.E., B.R. Fransen, and P.A. Bisson. 1996. Incorporation of nitrogen and carbon from spawning coho salmon into the trophic systems of small streams: evidence from stable isotopes. Canadian Journal of Fisheries and Aquatic Sciences 53

Bilby, R.E. et al. 1998. Response of juvenile coho salmon (Onchorynchus kisutch) and steelhead (Onchorynchus kisutch) to the addition of salmon carcasses to two streams in southwestern Washington , U.S.A.

Reimchen, T. and D.D. Mathewson. 1999. Salmon, bear, and nitrogen isotopes ­ Linking ocean nutrients to forest ecosystems. An unpublished report.

SpruceRoots Magazine - February 2000