65
1
Current address: School of Animal, Plant and
Environmental Science, The University of Witwatersrand,
Private Bag 3, WITS 2050 South Africa
2
Current address: Department of Biological Sciences,
Idaho State Unilversity, Pocatello, ID 83209 USA
*
Corresponding author: [email protected]
Long-term Perspectives on Salmon Abundance:
Evidence from Lake Sediments and Tree Rings
De a n n e Dr a k e
1,*
a n D ro b e r t J. na i m a n
School of Aquatic and Fisheries Sciences
University of Washington, Seattle, Washington 98195, USA
br u c e Fi n n e y
2
Institute of Marine Science
University of Alaska Fairbanks, Fairbanks, Alaska 99775 USA
ir e n e Gr e G o r y -ea v e s
Department of Biology,
McGill University, Montreal, Quebec H3A 2T5, Canada
American Fisheries Society Symposium 71:65–75, 2009
© 2009 by the American Fisheries Society
Abstract.—Here we review developments in paleoecological reconstruction of Pacific salm-
on abundance and discuss the new management context and implications provided by
the reconstructions. Currently, two approaches are yielding long term reconstructions of
salmon abundance over the last hundreds to thousands of years. First, in sockeye salmon
Oncorhynchus nerka nursery lakes, the abundance of adult salmon is reflected in chemi-
cal and biological characteristics of lake sediments. These indicators have been used to
reconstruct patterns of salmon abundance over 2,500 years and are compared at several
points by archeological data. Second, emerging techniques using riparian tree-ring-growth
have produced sub-decadal resolution reconstructions of stream-spawning sockeye, Chi-
nook O. tshawytscha, pink O. gorbuscha, and chum O. keta salmon populations over the
last 150–350 years. Paleoecological reconstructions provide important insights into salmon
abundance and their variability prior to European settlement of western North America.
For example, sediment-based reconstructions show periods of naturally low sockeye salm-
on abundance at ~A.D. 1800 and from ~A.D. 0–700 in Alaskan lakes, and tree-ring based
reconstructions show river-specific patterns in abundance with cycles of 21–68 years in
duration. Both types of reconstruction also suggest relatively rapid, natural “recovery” of
salmon populations after periods of low abundance. As additional reconstructions become
available and a more synthetic understanding of them is developed, paleoecological re-
constructions will allow better evaluation of management paradigms (e.g., the long-term
fidelity of Pacific Decadal Oscillation cycles and regional salmon abundance) as well as
identification of additional patterns that cannot be extracted from limited historical data
sets. Paleoecological perspectives play a potentially important role in changing societal
expectations of salmon resources by recognizing natural variations in abundance. Such
expectations, if tempered by acknowledging natural changes in salmon productivity, can
be incorporated into flexible models, management and restoration strategies.
66
Drake et al.
Introduction
Any discussion of Pacific salmon On-
corhynchus spp. sustainability must include
fundamental questions about long-term pat-
terns of salmon abundance. Understanding
natural cycles and variation in discrete popu-
lations prior to the rise of industrial fisheries
and land use changes of the last century is
essential for assessing anthropogenic effects
and for making reasonable management de-
cisions. What portion of changes in salmon
abundance over the last century are a result of
our actions, and what portion can be attribut-
ed to natural environmental conditions? Can
natural cycles be incorporated into manage-
ment expectations and effective decisions?
A number of fish species are known to
fluctuate naturally in abundance over long
periods. The best examples may be Pacific
sardine Sardinops sagax and northern ancho-
vy Engraulis mordax. A 1700-year history
of each was reconstructed using scales pre-
served in anoxic ocean sediments (Baumgart-
ner et al. 1992). Scale deposition rates varied
by ~15x, suggesting very large, natural shifts
in fish abundance, and documented the Cali-
fornia anchovy fishery collapse in the 1940s.
Baumgartner et al.’s reconstructions showed
inferred peaks in sardine abundance every
150–280 years, while anchovies peaked in
abundance more frequently, every 60–100
years. Both show periods of very low abun-
dance (very few or no scales in sediments).
The anchovy and sardine reconstructions fu-
eled speculation that salmon abundance may
also have fluctuated naturally, but because
salmon scales are not preserved in sediments,
other approaches needed to be developed.
Pacific salmon abundance appears to vary
with large-scale climate patterns, the best
known of which is the Pacific Decadal Oscil-
lation (PDO; Hare et al. 1999). The PDO is
defined by patterns of sea surface temperature
in the Pacific. During warm phases (positive
polarity PDO), coastal ocean biological pro-
ductivity is thought to be higher in Alaska and
relatively low in the southern Pacific salmon
range; regional salmon abundance reflects
these patterns of productivity. Historically,
polarity of the PDO shifts every 20–50 years,
and during cold phases (negative polarity)
the pattern of ocean productivity is reversed.
Although salmon abundance has tracked the
PDO over the last 80–90 years, the changes
reflect only 2–3 “regime shifts” (~1.5 cycles
of the PDO), and it is difficult to establish
cyclicity from these limited historical data.
These are also modern observations that are
limited to human-modified systems. Has the
PDO always been linked to cycles in regional
salmon abundance? Are there other cycles
that cannot be seen in 90 years of historical
salmon catch data? Is there any evidence in
the paleoecological record that salmon-cli-
mate links will continue to function as the
global climate changes?
Here we discuss the development and
potential management implications of long-
term cycles in salmon abundance recon-
structed from lake sediments (Finney 1998;
Finney et al. 2000, 2002; Gregory-Eaves et
al. 2003, 2004) and tree-ring growth (Drake
et al. 2002, 2005, Drake and Naiman 2007).
Paleoecological reconstructions provide a
unique source of information about natural
changes in salmon abundance over hundreds
to thousands of years, prior to the profound
changes imposed on the natural world by
human activities. These reconstructions can
help determine true baselines and frames of
reference for salmon restoration and assist
in understanding modern changes in salmon
populations.
Salmon Abundance
Reconstruction—Paleolimnological
Approaches
Unlike sardines and anchovies, salmon
do not reliably leave anatomical evidence of
their abundance—such as scales—in sedi-
67
Paleoecological Reconstruction of Salmon Abundance
ments. But Finney and others (2000, 2002)
demonstrated that lake-spawning sockeye
salmon Oncorhynchus nerka do leave evi-
dence of their abundance in the sediments of
some lakes. Nutrients accumulated by salm-
on in the ocean during growth and carried
to spawning areas can constitute as much as
70% of annual nitrogen (N) input (Naiman et
al. 2002) and 30–60% of annual phosphorus
(P) input (Koenings and Burkett 1987; Kroh-
kin 1975) to localized areas. Salmon-derived
nutrients are sequestered in phytoplankton
and zooplankton which typically die within
one season, settle, and form lake-bottom
sediments. Finney and others (1998, 2000,
2002) and Gregory-Eaves and others (2003,
2004) examined a number of nutrient sta-
tus indicators in sediments of Alaskan lakes
for evidence of salmon nutrient signals, the
best known of which is N-isotope composi-
tion (δ
15
N). N is supplied to most lakes from
N fixation by algae and plants (δ
15
N ~0‰),
decay of organic matter in the surrounding
watershed (bulk soil δ
15
N varies from ~2.1–
3.5‰ in the region; Amundson et al. 2003),
and atmospheric deposition (with δ
15
N from
–6.6 to 3.9; Kline et al. 1993). The N found
in salmon tissues, however, contains more of
the heavy isotope δ
15
N than these sources be-
cause adult salmon occupy a relatively high
trophic level (DeNiro and Epstein 1978), and
also because the heavy isotope of N is more
prevalent in marine environments than ter-
restrial environments. Adult sockeye salmon
tissue δ
15
N is ~12 ‰, so the relative contri-
bution of salmon N to lakes increases the
sediment δ
15
N. Sediment δ
15
N in a reference
lake (no salmon) was 1.5–3.5‰ (Finney et al.
2000), while in lakes supporting large sock-
eye salmon populations (contributing 4–24
kg N/ha/year) sediment δ
15
N was 5.0–9.5‰.
Measured salmon abundance over the last 75
years was closely reflected in sediment δ
15
N
(Figure 1) over time, and this relationship
was expanded to reconstruct salmon abun-
dance over 500 years (Finney 1998).
The authors also examined several bio-
logical indicators of nutrient-driven lake pro-
ductivity including diatoms and invertebrate
remains. They found that sediments with
high δ
15
N (deposited during periods of high
salmon abundance) also contained greater
abundance of diatoms indicative of nutrient
enrichment (e.g., the dominant taxa were
Stephanodiscus; Figure 2) and greater fluxes
of invertebrate remains (e.g., Bosmina). To-
gether, these productivity and isotope indica-
tors constitute strong, multi-proxy evidence
that salmon abundance was recorded in the
sediments of several lakes supporting large
spawning populations. But salmon abun-
dance can only be reflected in the sediments
of lakes where flushing rates are low and
spawning populations contribute enough nu-
trients to affect the nutrient mass balance of
the whole lake system (Holtham et al. 2004).
The potential contribution of salmon to lake
nutrient budgets of the Karluk Lake system
was recognized over 70 years ago by Juday
and others (1932), but the recent advances
in paleolimnology described here quantify a
fundamental link between long-term variabil-
ity of sockeye salmon populations and lake
ecology.
The first reconstruction extended to
300 years before present (Finney 1998 and
Finney et al. 2000). Both biological and
chemical indicators suggest that, over this
period, the lowest abundance of sockeye
salmon spawners in both Karluk and Akalura
lakes (south-central Alaska) occurred over
the last 30 years. A short period of relatively
low abundance was also apparent at ca. A.D.
1800 (Figure 2). When the reconstructions
were extended to 2,200 years before present
(Finney et al. 2002; Figure 3), not only were
the decadal-scale fluctuations apparent, but a
much lower-frequency pattern also emerged;
sockeye abundance in both lakes was excep-
tionally low from ca A.D. 0–600, shifting to
consistently higher mean abundance after ca.
A.D. 800.
68
Drake et al.
0
2
4
6
8
10
0510 15 20 25 30 35
Escapement/lake areas
(1000s fish/km
2
)
Sedimentary
15
N (‰)
Correlation on log(X+1)
transformed data of both
X and Y: r = 0.84,
P <0.01, n = 12
δ
Fi g u r e 1. Recent sockeye salmon escapement per unit lake area vs. δ
15
N in lake sediments.
A clear, positive relationship forms the basis for sediment-based reconstructions. Modified
from Finney et al. (2002).
Reconstructed sockeye salmon abun-
dance in Akalura and Karluk lakes is corrobo-
rated by archeological evidence (Mills 1994).
The lowest inferred abundance of sockeye
occurred between A.D. 0 and 600, a period
when a smaller human population was build-
ing smaller dwellings, and producing fewer
fishing tools. These characteristics began to
reverse at about A.D. 800, when an increase
in inferred salmon abundance was prob-
ably important in supporting larger human
populations. Since A.D. 800, inferred mean
sockeye salmon returns have varied around
a higher mean in both Karluk and Akalura
lakes. The sediment-based reconstructions,
however, show declines in salmon-derived
nutrient loading over the commercial fishing
era since about 1880 (Finney et al. 2000; Fig-
ure 3), and this could influence longer-term
lake production in some systems. Ongoing
work is extending sediment reconstructions
to span the entire Holocene.
Perspectives from Tree-rings
A relatively new approach to reconstruct-
ing salmon abundance is based on links be-
tween salmon and riparian tree-ring growth.
Trees along salmon streams are potentially
affected by salmon nutrients in ways fun-
damentally similar to lakes—varying nu-
trient contributions from year to year may
be reflected in primary productivity (tree-
ring growth—a fertilization effect) or the
isotopic composition of tree rings. These
characteristics, along with soil processes
regulating availability of salmon nutrients to
riparian trees, are examined in an ongoing
69
Paleoecological Reconstruction of Salmon Abundance
series of studies by Drake and colleagues
(Drake et al. 2002, 2005, 2006, Drake and
Naiman 2007).
Lake sediment and tree-ring approaches
differ fundamentally and provide useful con-
trasts. First, lake sediment approaches target
lake-spawning sockeye populations, while
tree-ring approaches are used to target stream-
spawning salmon, including pink O. gorbus-
cha, chum O. keta, Chinook O. tshawytscha,
and coho O. kisutch. Second, tree-rings are
of higher resolution (annual) than sediments
(~decadal) and are far easier to date, but are
generally limited to 500 years or less, where-
as lake sediments can span much longer pe-
riods—potentially thousands of years. Third,
there are simply more sediment records avail-
able for analysis because few salmon rivers
meet the criteria for tree-ring reconstruction:
streams must support riparian trees at least
150 years old, must have 15+ years of high-
quality (e.g., weir count) escapement data
for validation, and must be associated with a
good quality reference (salmon-free) riparian
forest site.
One might expect a riparian tree growth
response to salmon abundance because nutri-
ent limitation of tree growth occurs widely
in the eastern North Pacific, and tree-ring
growth responses to fertilization are well
documented in forestry science (see Nason
and Myrold 1992). The empirical evidence
supporting salmon nutrient fertilization of
riparian trees is substantial. Transport of
Fi g u r e 2. Lake sediment indicators of salmon abundance: δ
15
N, diatom-based total phos-
phorus (TP), common diatoms, and summed benthic diatoms (including both common and
uncommon taxa) over 2200 years in Karluk Lake, Alaska. Recent declines in sockeye es-
capement track decreases δ
15
N, TP, and some diatoms. Modified from Finney et al. (2002)
and Gregory-Eaves et al. (2003).
Karluk Lake ~2200 yr record
Finney & Gregory-Eaves et al., 2002.
Gregory-Eaves et al., 2003.
70
Drake et al.
Fi g u r e 3. Sediment δ
15
N, an indicator of sockeye salmon abundance, over the last 2,500
years in two sockeye nursery lakes (Karluk and Akalura) on Kodiak Island, Alaska. Frazer
Lake did not support salmon until the 1950s when sockeye were stocked and a fish ladder
was built, as reflected in sediment δ
15
N. Modified from Finney et al. (2002).
salmon nutrients from streams to riparian
forests by predators and scavengers has been
widely observed and discussed (e.g., Ceder-
holm et al. 1999; Gende et al. 2002). Addi-
tionally, natural abundance N isotope studies
have consistently shown that salmon-derived
nitrogen is incorporated into riparian vegeta-
tion on spawning streams (e.g., Bilby et al.
1996; Ben-David et al. 1998; Hilderbrand et
al. 1999, Helfield and Naiman 2002). Drake
and others conducted mechanistic studies of
riparian soils and tree physiology that build
on the empirical evidence. They quantified the
nutrient contributions (ammonium, nitrate, S,
P, K, Ca, Mg) of decaying salmon carcasses
to riparian soils on a salmon stream (Drake et
al. 2005) and, in a companion study, used an
N isotope tracer (δ
15
N ~ 30,000) to quantify
the fate of ammonium (the main nitrogenous
product of salmon decay) in riparian soils
and trees. Within 6 months, riparian western
red cedar Thuja plicata took up at least 37%
of the pulse of ammonium applied during
salmon spawning (Drake et al. 2006).
Annual tree ring growth was found to be
positively related to salmon nutrient inputs
(weir counts of escapement) by Drake et al.
in a pilot study (2002). The authors compared
tree ring growth at a salmon-influenced site
and a reference site beyond a barrier to salm-
on passage on each stream. Growth patterns
common to all trees at a site (chronologies)
were calculated using standard dendroecol-
ogy techniques and programs (cross-dating
verified with COFECHA; Holmes (1983),
and chronologies calculated with ARSTAN;
Cook (1985)). More recently, salmon abun-
dance was reconstructed for five Pacific
Northwest streams (Drake and Naiman 2007).
Tree growth-escapement relationships (e.g.,
71
Paleoecological Reconstruction of Salmon Abundance
Figure 4) were used as calibration sets to re-
construct adult salmon abundance, as in Fig-
ure 5, and the reconstructions were validated
by comparing them to local histories (e.g.,
construction of dams and salmon canneries),
regional salmon landings, and the Pacific
Decadal Oscillation climate index (Mantua et
al. 1997). The reconstructions showed lower
population cycle maxima in recent decades in
two of the five populations: Chinook in the
Umpqua River, Oregon, and pink salmon in
Sashin Creek, southeast Alaska (Drake and
Naiman 2007). Reconstruction of Drinkwa-
ter Creek B.C. sockeye suggested that abun-
dance since the mid-1990s has been 15–25%
greater than any time since 1850 (Figure 5b).
No long-term deviations from natural cycles
were detected for salmon in the Kadashan
River or Disappearance Creek in southeast
Alaska. Decadal-scale cycles in salmon
abundance with periods of 25–68 years were
detected in all reconstructions.
Reconstruction of Chinook salmon in
the Umpqua River shows lower than aver-
age abundance over the last 100 years when
compared to the period prior to intensive
industrial fisheries and cannery operations
in the area (about 1750–1880). This pat-
tern fits well with documented large-scale
decline of salmon in continental U.S. (see
Nehlsen et al. 1991), but the reconstruction
also provides new information (Figure 3a).
A strong, natural cyclicity extending back
at least 350 years is evident, as is a clear
interruption in natural cycle maxima after
the Winchester dam was constructed on the
Umpqua in 1890. Reconstructions of popu-
lations north of the Canada-U.S. border also
conform to general expectations (see Slaney
et al. 1996), suggesting that the abundance
of one intensively managed stock (Drinkwa-
ter Creek B.C. sockeye salmon) is 15–25%
higher than any time since the 1850 s, two
of the four stocks are maintaining essen-
tially natural abundance levels and patterns
(Kadashan River and Disappearance Creek,
southeast Alaska), and the last shows lower
reconstructed maxima in recent decades
(Sashin Creek pink salmon, southeast Alas-
ka).
0
5,000
10,000
15,000
0.50.7 0.91.1 1.31.5
Escapement (weighted, lag-1)
Growth index
r = 0.37
2
Fi g u r e 4. An example of non-climatic tree-ring growth index vs. escapement (weighted,
lag-1) at the Umpqua River site where a positive relationship between tree-ring growth and
salmon escapement was detected. Reproduced from Drake and Naiman (2007) with per-
mission from the Ecological Society of America.
72
Drake et al.
B. Drinkwater Creek
1972
1939
1915
0
5,000
10,000
15,000
18501900 1950 2000
Year
Sockeye abundance
A. Umpqua River
0
5,000
10,000
1750 18001850190019502000
Year
Chinook abundance
Winchester
Dam 1890
+ PDO
+ PDO
+ PDO
+ PDO
1915
1939
1972
Fi g u r e 5. Reconstructed salmon abundance (—) including extrapolated values outside of the
calibration range (---), and lag-1 weighted escapement (–□–). The Umpqua River, Oregon
(A) lies within the Southern Pacific salmon range, Drinkwater Creek, British Columbia (B) is
central. Benchmark years, or historic “boom (■) and bust (•)” years in regional salmon sh-
eries are marked for comparison; 1915 was a year of very high salmon catch/abundance
in Washington and Oregon, and very poor catch/low abundance in Southeast Alaska. 1939
was the reverse—a year of generally poor catch in Washington and Oregon (Pacific Fisher-
man 1939), while 1933–1935 were peak years for pink and chum catch in Southeast Alaska
(Mantua et al 1997). 1972 was a year of very high Chinook catch in the southern range
(National Fisherman 1972), and 1970–1975 were poor years for chum and pink catch in
Southeast Alaska. Shaded areas mark periods when the PDO was it its positive/warm phase
(pre-1925 values are from Gedalof and Smith 2001, and post-1925 values are from Mantua et
al. (1997) and Biondi et al. (2001)). Reproduced from Drake and Naiman 2007 with permis-
sion from the Ecological Society of America.
73
Paleoecological Reconstruction of Salmon Abundance
Over the last decade, there has been a
large collective investment in examining the
use of N isotopes in tree rings to reconstruct
salmon abundance, but no significant rela-
tionship between tree-ring δ
15
N and salmon
escapement over time has been demonstrated
in the peer-reviewed literature. A
15
N tracer
study showed that isotopically labeled N tak-
en up by riparian trees in 1998 was distrib-
uted throughout at least 10 years of adjacent
tree-rings in western hemlock Tsuga het-
erophylla, even when the mobile sap N was
removed (Drake 5005). Further, historic ex-
tirpation of salmon from the Metolius River,
Oregon (~1920) did not result in a significant
change in tree-ring δ
15
N (Drake 2005).
Discussion: Long-term Views
Lake sediment-based reconstructions
show that natural fluctuations in salmon abun-
dance occur at frequencies spanning decades
to millennia. Recent tree-ring analyses show
that patterns at higher frequencies (years to
decades) can also be reconstructed and ap-
pear to be stream-specific. Although recon-
structions of salmon abundance cannot be
used to make absolute predictions for the fu-
ture, they can be used to modify societal and
commercial expectations. Reconstructions of
salmon abundance provide new perspectives
for management and scientists interested in
modeling salmon abundance and manage-
ment. Most notably:
1) Modern salmon population dynam-
ics in the context of reconstructions: salmon
clearly fluctuate naturally in abundance over
multiple time scales. In the reconstructions
presented here, natural periods of low salmon
abundance have always been followed by
recovery. The industrialization of fisheries,
damming of rivers, proliferation of hatcher-
ies, and land use practices that compromise
riverine ecosystems, however, have driven
some salmon populations to exceptionally
low levels during times when abundance may
otherwise have been high, resulting in dimin-
ished baseline populations. Paleoecological
reconstructions allow the incorporation of
natural cycles and these diminished baselines
into analyses and models, and a better under-
standing of modern patterns of abundance
and survival.
2) Human expectations: During the period
when salmon were less abundant in Karluk
Lake (A.D. 0–600), archeological evidence
suggests that the local human population also
declined, and that the people remaining in the
area shifted their foraging strategies away from
salmon. Modern human response to declining
salmon abundance is often mediated by mar-
kets rather than either necessity or physical
limitations of gear, and increased fishing ef-
fort in response to declines has historically re-
sulted in collapse of fisheries. Paleoecological
perspectives have a potentially important role
in changing societal expectations of salmon
resources—providing a model for understand-
ing and accepting natural variations in fish
abundance. Such expectations can be better
incorporated into management strategies by
acknowledging temporal changes in salmon
abundance and productivity.
3) Natural salmon variability/cycles and
nutrient fertilization: “nutrient enhancement”
is a currently fashionable response to low
salmon abundance that promises salmon re-
covery and has generated much excitement;
but it also draws attention away from proven
threats such as overfishing and habitat deg-
radation. Reconstructions show that salmon
abundance can change from low to high within
a few years (e.g., Figures 3 and 5)—popula-
tions can increase or recover quickly, without
the benefit of human-mediated fertilizer ad-
dition, as long as suitable habitat is available.
The role of nutrient enhancement should be
reconsidered in the context of natural popu-
lation variations (reconstructions) along with
other issues (downstream nutrient pollution,
bioaccumulation of toxins, and human-health
considerations).
74
Drake et al.
Last, we should remember that ideas
about cycles in animal abundance are not new
and are not limited to fish. Classical studies
in ecology still underpin the modern manage-
ment of upland game animals – for example, in
the 1930s Elton described periodic maxima in
vole populations (e.g., Elton and Davis 1935),
and by 1954 Andrewartha and Birch (1954)
had published a seminal book on animal pop-
ulation dynamics based largely on studies of
violent fluctuations in “plague” insects.
The reconstructions presented here, in
combination with ongoing paleoecologi-
cal studies throughout the range of Pacific
salmon, will continue to improve our under-
standing of patterns in salmon abundance.
We hope that new, long-term perspectives
will be incorporated by science, policy and
management, and that they will help attain
sustainability of salmon and the ecosystems
that support them.
Acknowledgments
We thank the many colleagues, students
and technicians that contributed to this work,
and we thank Brady Green and one anony-
mous reviewer for helpful comments on the
manuscript. We also thank the Pacific North-
west Research Station of the U.S. Forest Ser-
vice, the University of Washington Sea Grant
Program, the University of Arizona Labora-
tory of Tree-ring Research, Alaska Sea Grant,
NSF/NOAA–GLOBEC, the NOAA Auke
Bay Laboratory, the Alaska Department
of Fish & Game, the U.S. Fish & Wildlife
Service, Bonneville Power, and the Shoban
Tribes for support.
References
Amundson, R., A. T. Austin, E. A. G. Schurr, K. Yoo,
V. Matzek, C. Kendall, A. Ubersax, D. Brenner,
and W.T. Baisden. 2003. Global patterns of the
isotopic composition of soil and plant nitrogen.
Global Biogeochemical Cycles 17(1): 1031,
doi:10.1029/2002GB001903.
Andrewartha, H. G., and L. C. Birch. 1954. The Dis-
tribution and Abundance of Animals University
of Chicago Press, Chicago.
Baumgartner, T. R., A. Soutar, and V. Ferreira-Bartri-
na. 1992. Reconstruction of the history of Pa-
cific sardine and northern anchovy populations
over the past two millennia from sediments of
the Santa Barbara Basin, California. California
Cooperative Fisheries Investigations Reports
33:24–40.
Ben-David, M., T. A. Hanley, and D. M. Schell. 1998.
Fertilization of terrestrial vegetation by spawn-
ing Pacific salmon: the role of flooding and
predator activity. Oikos 83:47–55.
Bilby, R. E., B. R. Fransen, and P. A. Bisson. 1996.
Incorporation of nitrogen and carbon from
spawning coho salmon into the trophic system
of small streams: evidence from stable isotopes.
Canadian Journal of Fisheries and Aquatic Sci-
ences 53:164–173.
Biondi, F., A. Gershunov, and D. R. Cayan. 2001.
North Pacific decadal climate variability since
1661. Journal of Climate 14:5–10.
Cederholm, C. J., M. D. Kunze, T. Murota, and A.
Sibatani. 1999. Pacific Salmon carcasses: es-
sential contributions of nutrients and energy
for aquatic and terrestrial ecosystems. Fisheries
24:6–15.
Cook, E. R. 1985. A time series approach to tree-
ring standardization. Doctoral dissertation.
University of Arizona, Tucson.
DeNiro, M. J., and S. Epstein. 1978. Influence of
diet on the distribution of nitrogen isotopes
in animals. Geochimica et Cosmochimica Acta
42:495–506.
Drake, D.C. 2005. The fate and influence of salmon
nutrients in riparian soils and trees: seasonal to
multi-century patterns. Doctoral dissertation.
University of Washington, Seattle.
Drake, D. C., and R. J. Naiman. 2007. Reconstruc-
tion of Pacific salmon abundance from ripar-
ian tree-ring growth. Ecological Applications
17:1523–1542.
Drake, D. C., R. J. Naiman, and J. S. Bechtold. 2006.
Fate of nitrogen in riparian forest soils and
trees; a 15N tracer study simulating salmon de-
cay. Ecology 87:1256–1266.
Drake, D. C., R. J. Naiman, and J. M. Helfield. 2002.
Reconstructing salmon abundance in rivers; an
initial dendrochronological evaluation. Ecology
83:2971–2977.
Drake, D. C., J. V. Smith, and R. J. Naiman. 2005.
Salmon decay and nutrient contributions to ripar-
ian forest soils. Northwest Science 79:60–70.
Elton, C., and C. H. S. Davis. 1935. An epidemic
75
Paleoecological Reconstruction of Salmon Abundance
among voles (Microtus agrestis) on the Scottish
border in the Spring of 1934. Journal of Animal
Ecology 4:277–288.
Finney 1998. Long-term variability in Alaskan sock-
eye salmon abundance determined by analysis
of sediment cores. North Pacific Anadromous
Fish Commission Bulletin 1:388–395.
Finney, B. P., I. Gregory-Eaves, M. S. V. Douglas, and
J. P. Smol. 2002. Fisheries productivity in the
northeastern Pacific Ocean over the past 2,200
years. Nature (London) 416:729–733.
Finney, B. P., I. Gregory-Eaves, J. Sweetman, M. S.
V. Douglas, and J. P. Smol. 2000. Impacts of
climatic change and fishing on Pacific salmon
abundance over the past 300 years. Science
290:795–799.
Gedalof, Z., and D. J. Smith. 2001. Interdecadal cli-
mate variability and regime-scale shifts in Pacif-
ic North America. Geophysical Research Letters
28:1515–1518.
Gende, S. M., R. T. Edwards, M. F. Willson, and M. S.
Wipfli. 2002. Pacific salmon in aquatic and ter-
restrial ecosystems. BioScience 52:917–928.
Gregory-Eaves, I., B. P. Finney, M. S. V. Douglas, and
J. P. Smol. 2004. Inferring sockeye salmon (On-
corhynchus nerka) population dynamics and
water-quality changes in a stained nursery lake
over the past ~500 years. Canadian Journal of
Fisheries and Aquatic Sciences 61:1235–1246.
Gregory-Eaves, I., J. P. Smol, M. S. V. Douglas, and
B. P. Finney. 2003. Diatoms and sockeye salmon
(Oncorhynchus nerka) population dynamics:
Reconstructions of salmon-derived nutrients in
two lakes from Kodiak Island, Alaska. Journal
of Paleolimnology 30: 35–53.
Hare, S. R., N. J. Mantua, and R. C. Francis. 1999.
Inverse production regimes: Alaska and west
coast Pacific salmon. Fisheries 24:6–14.
Helfield, J. M., and R. J. Naiman. 2001. Effects of
salmon-derived nitrogen on riparian forest
growth and implications for stream productiv-
ity. Ecology 82:2403–2409.
Hilderbrand, G. V., T. A. Hanley, C.T. Robbins, and C.
C. Schwartz. 1999. Role of brown bears (Ursus
arctos) in the flow of marine nitrogen into a ter-
restrial ecosystem. Oecologia 121:546–550.
Holmes, R. 1983. Computer assisted quality control
in tree-ring dating and measurement. Tree-Ring
Bulletin 44: 69–75.
Holtham, A. J., I. Gregory-Eaves, M. G. Pellatt, D. T.
Selbie1, L. Stewart, B. P. Finney, and J. P. Smol.
2004. The influence of flushing rates, terrestrial
input and low salmon escapement densities on
paleolimnological reconstructions of sockeye
salmon (Oncorhynchus nerka) nutrient dynam-
ics in Alaska and British Columbia. Journal of
Paleolimnology 32:255–271.
Juday, C. W. H. Rich, G. I. Kemmerer and A. Mean.
1932. Limnological studies of Karluk Lake,
Alaska 1926–1930. U.S. Bureau of Fisheries
Bulletin 28: 485–488.
Kline, T. C., J. J. Goering, O. A. Mathisen, P. Poe,
P. L. Parker, and R. S. Scalan. 1993. Recycling
of elements transported upstream by runs of
Pacific salmon. II. δ
15
N and δ
13
C evidence in the
Kvichak River watershed, Bristol Bay, Alaska.
Canadian Journal of Fisheries and Aquatic Sci-
ences 50:2350–2365.
Koenings, J. P. and R. D. Burkett. 1987. An aquatic
Rubic’s cube: restoration of the Karluk Lake
sockeye salmon (Oncorhynchus nerka). Pages
419–434 in Smith, H. D., L. Margolis, and C. C.
Wood editors Sockeye salmon (Oncorhynchus
nerka) population biology and future manage-
ment. Canadian Special Publication of Fisheries
and Aquatic Sciences 96.
Krohkin, E. M. 1975. Transport of nutrients by salmon
migrating from the sea into lakes. Pages 153–
156 in A. D. Hasler, editors Coupling of Land
and Water Systems. Springer-Verlag, New York.
Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace
and R. C. Francis. 1997. A Pacific interdecadal
climate oscillation with impacts on salmon pro-
duction. Bulletin of the American Meteorologi-
cal Society 78:1969–1079.
Mills, R. O. 1994. Radiocarbon calibration of archae-
ological dates from the central Gulf of Alaska.
Arctic Anthropology 31:126–149.
Naiman, R. J., R. E. Bilby, D. E. Schindler, and J. M.
Helfield. 2002. Pacific salmon, nutrients, and
the dynamics of freshwater and riparian eco-
systems. Ecosystems 5:399–417.
Nason, G. E. and D. D. Myrold. 1992. Nitrogen fertil-
izers: fates and environmental effects in forests.
Pages 67–81 in H. N. Chappel, G. F. Weetman,
and R. E. Miller, editors. Forest fertilization: sus-
taining and improving nutrition and growth of
western forests. College of Forest Resources,
University of Washington, Seattle.
National Fisherman. 1972. September 1972. Journal
Publications, Inc. Portland, Oregon.
Nehlsen, W., J. E. Williams, and J. A. Lichatowich.
1991. Pacific salmon at the crossroads: stocks
at risk from California, Oregon, Idaho and
Washington. Fisheries 16:4–21.
Pacific Fisherman. 1939. Yearbook. M. Freeman,
Portland, Oregon.
Slaney, T. L., K. D. Hyatt, T. G. Northcote, and R. J. Fielden.
1996. Status of anadromous salmon and trout in
British Columbia and Yukon. Fisheries 21:20–35.
