Ecological Applications, 17(6), 2007, pp. 1752–1770
? 2007 by the Ecological Society of America
SERIAL DEPLETION OF MARINE INVERTEBRATES LEADS TO THE
DECLINE OF A STRONGLY INTERACTING GRAZER
ANNE K. SALOMON,1,4NICK M. TANAPE, SR.,2AND HENRY P. HUNTINGTON3
1University of Washington, Department of Biology, Box 351800, Seattle, Washington 98195-1800 USA
2Native Village of Nanwalek, Box 8003, Nanwalek, Alaska 99603 USA
3Huntington Consulting, 23834 The Clearing Drive, Eagle River, Alaska 99577 USA
leading to recent declines of the black leather chiton (Katharina tunicata) on the outer Kenai
Peninsula, Alaska (USA). This intertidal mollusk is a strongly interacting grazer and a
culturally important subsistence fishery for Sugpiaq (Chugach Alutiiq) natives. We took
multiple approaches to determine causes of decline. Field surveys examined the significant
predictors of Katharina density and biomass across 11 sites varying in harvest pressure, and an
integrated analysis of archaeological faunal remains, historical records, traditional ecological
knowledge, and contemporary subsistence invertebrate landings examined changes in
subsistence practices through time. Strong evidence suggests that current spatial variation in
Katharina density and biomass is driven by both human exploitation and sea otter (Enhydra
lutris) predation. Traditional knowledge, calibrated by subsistence harvest data, further
revealed that several benthic marine invertebrates (sea urchin, crab, clams, and cockles) have
declined serially beginning in the 1960s, with reduced densities and sizes of Katharina being the
most recent. The timing of these declines was coincident with changes in human behavior
(from semi-nomadic to increasingly permanent settlement patterns, improved extractive
technologies, regional commercial crustacean exploitation, the erosion of culturally based
season and size restrictions) and with the reestablishment of sea otters. We propose that a
spatial concentration in shoreline collection pressure through time, increased harvest
efficiency, and the serial depletion of alternative marine invertebrate prey have led to
intensified per capita predator impacts on Katharina and thus its recent localized decline.
We investigated the relative roles of natural factors and shoreline harvest
selection; prey switching; rocky intertidal; sea otter; serial decline; social–ecological system; subsistence
fisheries; traditional ecological knowledge.
Akaike Information Criterion (AIC); black leather chiton; Katharina tunicata; model
Competing hypotheses are often invoked to explain
the decline of marine species. Because effective conser-
vation plans and sustainable fisheries strategies necessi-
tate that causal mechanisms driving declines be
identified, a strong impetus exists to scrutinize the
strength of evidence for alternative causes (NRC 1999,
2003). This presents an enormous challenge because
ecosystems are affected simultaneously by multiple
drivers of change, both top-down (consumer-driven)
and bottom-up (resource-driven) (Fretwell 1977, Oksa-
nen et al. 1981), anthropogenic and natural (Dayton et
al. 1998), varying in magnitude and spatial extent (Levin
1992). Furthermore, present day perturbations operate
within the context of historical alterations (Lewontin
1969) and lastly, drivers of change can interact, leading
to complex and often synergistic effects (Hilborn and
Stearns 1982, Paine et al. 1998). Here we examine the
multiple factors driving the recent decline of a nearshore
benthic invertebrate, the black leather chiton (Katharina
tunicata) on the rocky shores of the outer Kenai
Peninsula, Alaska, USA.
In marine systems, species declines are often attribut-
ed to harvest pressure: direct and/or indirect, nonhuman
predation pressure, environmental forces, or a combi-
nation of these causal agents. For example, the effects of
fishing, natural predators, and large-scale forcing
functions (e.g., Pacific Decadal Oscillation) have been
implicated in the collapse of Steller sea lions in Alaska
(NRC 2003), Peruvian anchoveta off South America
(Clark 1981), and Atlantic cod in Canada (Hutchings
and Myers 1994). In temperate rocky intertidal ecosys-
tems, humans (Castilla and Dura ´ n 1985), sea otters
(Estes and Palmisano 1974), sea stars (Paine 1966), and
shorebirds (Wooton 1992) are all predators well known
to alter rocky intertidal community dynamics (Menge
and Branch 2001). In fact, in Alaska, humans and sea
otters have been implicated in dramatic localized
depletion of nearshore invertebrates leading to the
alteration of coastal ecosystems (Simenstad et al. 1978,
Duggins et al. 1989). Yet, at the same time, bottom-up
Manuscript received 11 August 2006; revised 30 January
2007; accepted 15 February 2007. Corresponding Editor: P. K.
4Current address: Marine Science Institute, University of
California, Santa Barbara, California 93106 USA.
processes such as nutrient availability and primary
production have also been shown to govern nearshore
dynamics elsewhere (Menge 2000, Nielsen and Navar-
Distinguishing between drivers of change in marine
systems is difficult because food webs are complex and
typically not well understood (Larkin 1978), natural
forcing functions are highly variable (Francis et al.
1998), fishing effort is spatially widespread, and true
controls and replicates are often absent (Ludwig et al.
1993). Indeed, the syndrome of shifting baselines, in
which the lack of information about the past can lead to
underestimates of overall declines (Pauly 1995, Dayton
et al. 1998, Jackson et al. 2001), may hinder the
detection of change in the first place. Furthermore,
emphasis on selected declines species-by-species or
stock-by-stock can lead to a myopic perception of what
is in fact a complex system and can thus mask more
general phenomena, such as serial depletions.
The serial decline of marine resources is a symptom of
ecosystem overfishing (Murawski 2000). Typically,
fisheries first target the most lucrative stock and switch
to the next most profitable stock when the marginal
value of the former becomes too small to make its
harvest worthwhile. This mechanism has been proposed
for the sequential decline of crustaceans in the Gulf of
Alaska (Orensanz et al. 1998) and abalone in California
(Karpov et al. 2000). A progression of multispecies
declines can also emerge when nonhuman predators
switch among alternative prey, from most preferred and
available to least preferred and rare. Certainly, switch
points for human and nonhuman predators may differ,
particularly when rare prey increase in commercial
value. Under these circumstances, commercial harvest
may continue unabated, often reducing prey to very low
levels, whereas nonhuman predators and subsistence
harvesters may switch to alternative prey earlier, when
preferred prey become rare. The concept of prey
switching by predators, a common phenomenon in
natural systems (Holling 1959, Holt and Lawton 1994),
has been proposed as one possible mechanism driving
the consecutive decline of marine mammals in the north
Pacific Ocean and Bering Sea (Springer et al. 2003). Yet,
sequential declines are often difficult to identify because
research tends to consider species singly over short
periods of time. Furthermore, multiple causation,
possible synergisms among drivers of change, and a
paucity of historical data tend to obfuscate the causal
mechanisms driving serial declines. Fortunately, retro-
spective analyses of archaeological data and traditional
ecological knowledge (TEK) can provide historical
context and insight on variability and patterns of change
within ecosystems, extending farther back than contem-
porary scientific research (Simenstad et al. 1978,
Cruikshank 2001, Jackson et al. 2001).
The value of TEK in contributing to ecological
problem solving and conservation planning has become
increasing recognized (Berkes et al. 2000, Drew 2005).
Resource users can provide qualitative observations on
the presence, absence, and/or relative abundances of
various species at particular points in time and space
when quantitative baseline information is scarce (Jo-
hannes 1998). Given their familiarity with an ecosystem
and awareness of its peculiarities, subsistence users can
also offer a synthesis of relative timing and rates of
ecosystem change, identify important species interac-
tions, raise entirely new scientific questions, or propose
alternative testable hypotheses (Berkes 1999, Pierotti
and Wildcat 2000, Moller et al. 2004, Drew 2005). In
addition, subsistence users can simultaneously offer
insight into the internal and external socioeconomic and
ecological factors driving their own dynamic behavior as
predators (Berkes et al. 2000, Turner and Berkes 2006).
Ultimately, ecosystems are driven by social–ecological
dynamics (Carpenter and Gunderson 2001, Gunderson
and Holling 2002), and therefore a strong need exists to
comprehend human behavior and the socioeconomic
factors that motivate it (Ludwig et al. 1993). Conse-
quently, TEK and social and ecological science have
been increasingly combined to enlighten marine ecology
research, inform management, and address conservation
issues (Ellis and Swan 1981, Huntington 2000, Johannes
2002, Moller et al. 2004, Sala et al. 2004, Berkes et al.
2005). Here, we use TEK to reveal historical patterns of
marine invertebrate abundance and subsistence harvest
practices in south-central Alaska, USA.
On the rocky shores of the outer Kenai Peninsula,
important traditional subsistence food item for Sugpiaq
natives (alsoself-referred to asChugachAlutiiq, see Plate
1; Stanek 1985).5This strongly interacting grazer is also
well known to play an important functional role in
structuring intertidal communities and reducing macro-
algal production in the Pacific Northwest (Paine 1992,
Markel and DeWreede 1998, Paine 2002). Known locally
as ‘‘bidarkis,’’ these chitons were harvested by early
inhabitants of this area, as suggested by shells found in
nearby middens, some dating back 3000 years or more
(de Laguna 1934). Furthermore, local Sugpiaq elders
report that villagers have been harvesting this chiton for
at least the past century. However, local declines of
ago, despite littlehuman populationgrowth inthe areaat
that time (U.S. Census 2000, Brown et al. 2001). Elders
further report that sea otters were absent locally in the
early 1900s but began to reestablish by the early 1960s.
In collaboration with the Sugpiaq villages of Port
Graham and Nanwalek, we examined the relative roles
of top-down, bottom-up, and abiotic natural factors (sea
otter predation, bird predation, macroalgal production,
ocean temperature, and wave exposure), and anthropo-
genic perturbations (shoreline collection by humans)
proposed by both village residents and ecologists as
5Sugpiaq (singular), Sugpiat (plural).
September 20071753SERIAL DEPLETION OF MARINE INVERTEBRATES
potential causal factors driving Katharina decline. To
examine the strength of evidence among alternative
hypothesized causes of black leather chiton declines, we
quantified the present-day spatial variation in Katharina
and the factors listed above. We then used archaeolog-
ical data, historical records, and semi-directed interviews
with Sugpiaq elders to document temporal trends in
subsistence harvest practices, relative nearshore benthic
invertebrate abundances, and the ecological and socio-
economic factors that likely triggered them.
This research was conducted in south-central Alaska,
USA, at 11 rocky intertidal sites on the outer tip of the
Kenai Peninsula, in lower Cook Inlet, near two Sugpiaq
native villages, Port Graham and Nanwalek (Fig. 1).
Sites were identified by tribal elders to span a gradient in
Katharina subsistence collection effort. Consequently,
sites were located at varying distances away from the
villages, from heavily exploited accessible sites located
close to the villages, to less accessible, moderately
exploited sites located farther from the villages (Fig. 1C).
Residents of Port Graham and Nanwalek rely heavily
on wild resources as the basis of their mixed subsistence
and cash economies. Although salmon and halibut
comprise the bulk of their subsistence harvest, shoreline
collection of marine invertebrates (Stanek 1985, 2006)
and macroalgae (Kari 1991) remain important, as with
other coastal native tribes in the northeastern Pacific
(Ellis and Swan 1981, Turner 2003). While an exemption
to the Marine Mammal Protection Act allows Alaska
natives to take marine mammals for subsistence or craft
purposes, marine mammals, primarily harbor seal and
sea lion, comprised only 2.2% and 3.7% of the total
subsistence biomass landed in 2003 in Nanwalek and
Port Graham respectively (Stanek 2006). On average,
5.3 6 1.7 sea otters (mean 6 SE) were harvested
annually from 1987 to 2003 in this region with the
exception of 1993 and 1997, when 62 and 30 individuals
were landed (Brown et al. 2001, Stanek 2006).
Present-day factors governing variation in Katharina
To determine the primary factors driving the spatial
variation in Katharina density and size structure, we
quantified the relative magnitude of the dominant
surrounding the Sugpiaq native villages of Port Graham and Nanwalek. Sites differ in their accessibility to human harvesters and
consequently encompass a gradient of shoreline collection effort. Sites are (1) Point Pogibshi, (2) Coal Mine, (3) Otter Rock, (4)
Romanoff Point, (5) Inner Nanwalek, (6) Outer Nanwalek, (7) Flat Island, (8) Magnet Rock, (9) Golden Rocks, (10) Jagged Rocks,
and (11) Point Adam.
(A) In south-central Alaska, USA, (B) on the outer tip of the Kenai Peninsula, we surveyed (C) 11 rocky intertidal sites
ANNE K. SALOMON ET AL.1754
Vol. 17, No. 6
factors known to influence Katharina density and size
across each site. Top-down factors included shoreline
collection effort and predation pressure by other known
Katharina consumers (sea otters, birds, and sea stars).
Bottom-up factors included macroalgal production,
while abiotic factors included wave exposure and water
Harvest effort.—We quantified the current spatial
variation in shoreline collection effort of Katharina
harvest by opportunistically surveying 39 village resi-
dents with a fixed questionnaire (a survey where all
respondents are asked the same questions in the same
order). Harvest surveys were conducted with a map of
the area so that local residents could identify where they
currently harvested and for how many days per year.
This allowed us to estimate per capita harvest effort
(number of days per person per year) at each site. We
also documented locally observed spatiotemporal trends
in Katharina density and size structure.
Nonhuman predators.—Site-specific sea otter and bird
presence was estimated based on sightings made on the
approach to each site and during each site visit within
100 m. Sightings were conducted in 2003 and 2004 for
sea otters (Enhydra lutris) and in 2004 for known avian
predators of Katharina: Glaucous-winged Gulls (Larus
glaucescens), Black Oystercatchers (Haematopus bach-
mani), and Northwestern Crows (Corvus caurinus). To
further describe the spatial distribution and abundance
of sea otters in this region, we conducted three small
boat surveys in June and July 2004. Beginning at our
northeastern most site, Point Pogibshi (59825.740N,
151851.350W) and ending at our southwestern most site
Point Adam (59815.110N, 151858.680W), we contoured
the shoreline, within 50 m of the shore at a speed of 20
km/h or less, and recorded the number and GPS
location of sea otters (adults and pups) observed by
eye and with binoculars. These spatial data were
displayed on maps using GIS (Appendix). At each site,
we also quantified densities of benthic predators
including predatory sea stars (Leptasterias spp., Evas-
terias troschelii, and Pisaster ochraceus) and predatory
snails (Nucella canaliculata, Lirabuccinium dirum, and
Fusitrition oregonensis) by recording individuals in 10
0.25-m2quadrats. Quadrats were randomly stratified
along a 50-m transect line placed horizontal to the shore
in Katharina habitat: the middle of the intertidal zone,
an area dominated by the brown algae Alaria marginata
and Hedophyllum sessile.
Macroalgal production.—We measured growth rates
of the dominant benthic macroalga, Alaria marginata,
the primary food source of Katharina. Thirty individuals
per site were tagged with small zip ties and vinyl
numbered tube tags secured around their stipe, below
their sporophylls. Initial length and maximum width
were measured. We estimated growth rates by punching
two small holes, one on either side of the kelp blade’s
midrib, 1 cm above the meristem, and returning one tide
series later to measure the distance between the punched
holes and the meristem (Pfister and Stevens 2002). This
allowed us to quantify absolute growth rates (i.e.,
growth/time) and relative growth rates (absolute growth
Wave exposure.—To quantify site-specific wave expo-
sure, a factor well known to influence intertidal
community assemblages, recruitment, and population
size structure (Dayton 1971), we used three maximum-
wave-force recorders (Bell and Denny 1994) deployed
and revisited five times per site in June 2005. Swell and
wind conditions during this sampling period represented
typical spring conditions, similar to those observed
during the June 2003 and 2004 sampling periods. On
each visit, spring extensions were measured to the
nearest 0.5 mm and reset. Spring extension data
collected in the field were converted into maximum
wave force (newtons), with calibration curves estab-
lished earlier, in the lab. When drag forces were too
small to cause observable spring extension, we assumed
maximum wave force values equivalent to the minimum
force required to overcome initial spring compression.
We estimated differences in wave exposure among sites
based on the mean maximum wave force recorded over
the sampling period. We also ranked sites in terms of
wave exposure based on the maximum wave force
experienced at each site over the sampling period.
Sea surface temperature.—Because temperature is
known to influence rocky intertidal species interactions
on a local and regional scale (Sanford 1999) and can
reflect upwelling events which are known to influence
recruitment rates and species interactions (Menge et al.
2004), we measured site-specific sea surface temperature
(SST) from June to September 2004 with temperature
loggers placed at mean low water (MLW). To estimate
daily SST, we averaged the temperatures recorded every
90 minutes during daily high tides when sea level was ?3
m above MLW. Monthly mean values were calculated
based on daily SST.
Spatial variation in Katharina
We quantified Katharina density and size structure in
June 2003 and 2004 by measuring the maximum length
of all individuals found in 10 0.25-m2quadrats
randomly stratified along a 50-m transect line placed
at two tidal elevations representing preferred habitat of
Katharina (O’Clair and O’Clair 1998; n ¼ 10 low-zone
and 10 high-zone quadrats per site). Intertidal elevation
was based on biological assemblages; low quadrats were
placed in the middle of the intertidal zone dominated by
the brown algae Alaria marginata and Hedophyllum
sessile, whereas high quadrats were placed in the middle
of the intertidal zone dominated by the red alga
Endocladia muricata and encrusting coralline species.
This was done to ensure that Katharina habitat was
adequately sampled given that Katharina size was
observed to vary with tidal elevation. Individual
Katharina biomass was estimated from a length–mass
regression: biomass (in g) ¼ 6 3 10?5length2.98(in mm),
September 2007 1755SERIAL DEPLETION OF MARINE INVERTEBRATES
where n ¼ 466, R2¼ 0.942. Estimates of site-specific
Katharina biomass per 0.25 m2were derived by summing
individual biomass estimates per quadrat and averaging
across all quadrats surveyed in the low (n¼10) and high
(n ¼ 10) intertidal zones.
Prehistoric and historic factors altering nearshore
Archaeological and historical records.—Based on the
faunal remains collected from a previously excavated
archaeological site (SEL-027) in Port Graham Bay, we
compared the relative percentage of invertebrate species
retrieved from deep strata representing a late prehistoric
occupation (;1300–1500 AD) to those retrieved from
shallow strata representing early historic occupation
(20th century; W. Workman and K. Workman, unpub-
lished manuscript). Temporal resolution was based on,
and limited by, three radiocarbon dates and the Katmai
volcanic ash layer of 1912. Relative species percentages
were based on minimum number counts. We also
reviewed archaeological literature and historical records
of settlement patterns in lower Cook Inlet to provide
further insight on the extent of prehistoric and historic
human habitation and subsistence harvest in the area.
Qualitative traditional knowledge.—We conducted
semi-directed interviews (Huntington 1998) with 10
tribal elders to document historical trends in nearshore
ecosystem dynamics from the 1920s onward, including
changes in subsistence shellfish resources and harvest
practices, commercial fishing effort, social and economic
drivers, and species interactions. The elders were
selected by recommendation from village Tribal Coun-
cils and by chain-referral. In semi-directed interviews,
researchers identify a set of topics to be addressed, but
the respondent can pursue his or her lines of thought
rather than being constrained by the format of a
questionnaire. In this way, the respondent may indicate
connections or additional information that the research-
ers did not or could not have anticipated. Furthermore,
the course of the interview can follow the respondent’s
understanding of the topic, rather than the researchers’
We used the information from the interviews in two
ways. First, we identified historical observations, both
ecological and socioeconomic, and developed a timeline
of events and trends. The timeline was presented to the
elders and others in both communities to confirm the
information. Second, based on the ideas proposed by the
elders, we developed a set of formal hypotheses to
explain the observed invertebrate declines. These hy-
potheses were also presented to the elders and others for
confirmation, and were used in our analysis along with
observations and hypotheses shared by local tribal
members during the harvest effort surveys.
Quantitative subsistence shellfish landings.—We com-
pared the traditional knowledge of marine invertebrate
abundance trends with a time series of invertebrate
subsistence landings data from 1987 and 2003 derived
from the Community Profile Database (CPDB) main-
tained by the Division of Subsistence of the Alaska
Department of Fish and Game (Brown et al. 2001,
Stanek 2006). The biomass of invertebrate landings from
both villages, Nanwalek and Port Graham, were
summed. Estimates of human population size for both
villages were derived from the CPDB and U.S. Census
Community Database (U.S. Census 2000, Brown et al.
2001). Only general qualitative comparisons were made
between contemporary invertebrate subsistence landings
data and archaeological invertebrate remains due to
variable preservation of invertebrate species in middens
(i.e., soft-bodied mollusks, like cephalopods, are not
represented in archaeological remains and fragile species
such as blue mussels are not quantifiable). Furthermore,
relative species proportions based on abundance vs.
biomass are not directly comparable.
To detect differences in the contemporary spatial
variation in Katharina, sea otter, bird, sea star, and
predatory snail density among sites and between years
(2003 and 2004), we used generalized linear models
(GLZs), each with a Poisson error distribution and a
log-link function fit by maximum likelihood (SAS Proc
Genmod V 9.1.3). Differences in Katharina biomass,
shoreline collection effort, Alaria growth rates, and
maximum wave exposure among sites were analyzed
with GLZs on log(x þ 1)-transformed data, a normal
error distribution and an identity link function, fit by
maximum likelihood. We also compared Katharina
population structure among sites and between years
with size–frequency distributions. Differences in month-
ly sea surface temperatures among sites were analyzed
with a repeated-measures GLZ. Post hoc comparisons
were made based on Bonferroni adjusted P values.
Linear regressions were used to quantify the relationship
between shoreline harvest pressure and Katharina
density and biomass.
To determine the strength of evidence among
alternative hypothesized causal factors contributing to
the contemporary variation in Katharina among sites,
we took an information theoretic model selection
approach (Burnham and Anderson 1998). We compared
alternative candidate models of Katharina density and
biomass as a function of four factors (harvest pressure,
sea otter presence, bird presence, and wave exposure),
because these factors varied significantly among sites
and were expected to most affect Katharina distribu-
tions. These models, derived from integrating traditional
knowledge and western science, represent a priori
competing hypotheses regarding the primary factors
governing the present spatial variation in Katharina
density and biomass. This approach was taken over
standard multiple-regression because the latter is sensi-
tive to stepwise selection criterion, direction of fitting,
and variable order. We used means in Katharina density
ANNE K. SALOMON ET AL.1756
Vol. 17, No. 6
and biomass and sea otter presence across years (2003
and 2004) when constructing each model.
We ranked candidate models based on small-sample
bias-corrected Akaike’s Information Criterion (AICc)
which we standardized to the best-fit model to produce
DAICcvalues. We normalized the likelihoods to a set of
positive Akaike weights (wi) representing the strength of
evidence in favor of a given model. We then examined
the relative importance of each variable in contributing
to the current spatial differences in Katharina by
calculating variable weights. This was done by summing
the Akaike weights (wi) of all of the models in which a
variable was found (Burnham and Anderson 1998).
Given that four explanatory variables were used to
construct all possible model combinations, a total of 16
models were compared for low- and mid-intertidal
estimates of Katharina density. We report the top five
models for each response variable.
Present-day factors governing spatial variation
Harvest effort.—The spatial variation in current-day
shoreline collection effort varied significantly among the
11 sites such that the most heavily harvested site, Inner
Nanwalek, experienced 60 times more collection effort
than Point Adam, the least harvested site (Fig. 2A,
Table 1). However, in the low intertidal, harvest pressure
alone explained only 22% of the spatial variation in
Katharina density (R2¼0.22, F¼2.56, df¼10, P¼0.14)
and only 18% of the spatial variation of Katharina
biomass (R2¼0.18, F¼1.97, df¼10, P¼0.19) over 2003
effort, (B) sea otter presence, and (C) black leather chiton
(Katharina tunicata) density (all data shown as mean þ SE).
Variation between years is shown in (B) and (C). Sites are
ordered according to shoreline collection effort, from most
heavily harvested to least heavily harvested.
Spatial variation in (A) annual per capita harvest
influence the spatial variation in current Katharina density
and biomass in the low intertidal.
Differences in multiple factors among sites that may
Shoreline collection effort?
Sea otter sightings
Site9 0.65 0.100
Maximum wave exposure
Site 10 185.39
Sea surface temperaturejj
Site 3 month
? Point Pogibshi was excluded from this analysis because,
when included, the algorithm failed to converge due to the lack
of shoreline collection visits to this site.
? Although a Poisson error distribution was assumed, this
model could not support an interaction term because of the
high abundance of 0 counts.
§ Point Adam had no individuals of Alaria marginata large
enough to tag at the appropriate tidal elevation.
|| Temperature data were not retrieved from Point Adam.
September 2007 1757 SERIAL DEPLETION OF MARINE INVERTEBRATES
Out of 39 village residents surveyed, 100% currently
collect and consume Katharina. Eighty-nine percent had
observed a decline in Katharina size while 66% had
observed a decline in the density of this chiton. Forty-six
percent of village residents surveyed currently send
Katharina to relatives residing outside of the village.
Nonhuman predators.—Sea otter presence varied
significantly among sites and between years (Fig. 2B,
Table 1). Coal Mine and Otter Rocks had the highest sea
otter sightings among sites in both years. At these two
sites, greater than seven sea otters on average were
sighted per visit in 2004. In contrast, less than one sea
otter was sighted per visit at Inner Nanwalek, Romanoff
Point, Golden Rocks, and Jagged Rocks in 2004. These
site-specific results are consistent with the three small
boat surveys (Appendix) which consistently revealed low
abundances of sea otters (0–1 individuals) in close
proximity to village sites (Inner Nanwalek, Romanoff
Point) and at the southwestern-most sites (Golden
Rocks and Jagged Rocks), and high abundances (rafts
of 10–42 individuals) at Coal Mine, Otter Rocks, Flat
Island, and Point Pogibshi. Based on the three small
boat surveys in 2004, on average we observed 173 6 14
adults and 43 6 8 pups (all data are mean 6 SE) along
our shoreline transect from Point Pogibshi to Point
Adam. Bird presence also varied significantly among
sites (Table 1). Flat Island, the location of a Glaucous-
winged Gull colony, had 286 times more bird sightings
than Point Adam, the site with the fewest sightings. On
Flat Island, Katharina shells were found scattered
around gull nests located on nearby cliffs. The sea stars
Evasterias troschelii and Pisaster ochraceus were exceed-
ingly rare (1 Evasterias and 0 Pisaster were recorded in
209 0.25-m2quadrats). Densities of the dominant sea
star (Leptasterias spp.) and predatory snails (Nucella
canaliculata, Lirabuccinium dirum, and Fusitrition ore-
gonensis) reached densities of 1.9 6 0.3 and 1.7 6 0.5 per
0.25 m2, respectively, and varied significantly among
sites. Leptasterias densities also varied significantly
among years (Table 1).
Macroalgal production.—The relative growth rate of
the dominant low intertidal macroalga, Alaria margin-
ata, did not vary significantly among sites (Table 1),
although absolute growth rates ranged from 1.10 6 0.05
cm/day (Otter Rocks) to 2.9 6 0.18 cm/day (Jagged
Wave exposure.—Wave exposure varied significantly
among sites (Table 1) and ranged from 43.4 N (Golden)
to ,3.9 N (Romanoff Point, Otter Rocks, Inner
Sea surface temperature.—Generally, sites experi-
enced similar sea surface temperatures (SSTs) from June
to August 2004, although in June, Point Pogibshi was
significantly cooler than the other sites and experienced
temperatures ?4.38C cooler than the warmest site, Flat
Island. This temperature discrepancy disappeared by
August. Water temperatures varied significantly among
months; June (7.88 6 0.18C), July (9.68 6 0.28C), August
(11.18 6 0.28C; Table 1).
Spatial variation in Katharina.—Katharina density in
the low intertidal varied significantly among sites with
no site difference between years but with a significant
site-by-year interaction (Fig. 2C, Table 2). Densities
ranged from as low as 1.0 6 0.4 chitons/0.25 m2to as
high as 11.3 6 1.0 chitons/0.25 m2. Spatial variation in
low intertidal Katharina biomass, integrating both
density and size, also varied significantly among sites
(Table 2). Compared with the most heavily harvested
site, Inner Nanwalek, densities of chitons in 2004 were
6.1 and 7.5 times greater at two rarely harvested sites
(Jagged Rocks and Point Adam; Fig. 2C), while biomass
was 6.5 to 8.7 times greater. Katharina were more
abundant in the high intertidal compared to the low
intertidal, but the mean size was smaller (Fig. 3A, B).
Both density and biomass of Katharina recorded in the
high intertidal varied significantly among sites and years
with a significant site-by-year interaction (Table 2). The
population structure of Katharina varied among sites
and between years (Fig 3A, B). Large individuals
(?61mm) were less abundant at sites where either
harvest effort or sea otter presence was high (Inner
Nanwalek, Coal Mine, and Otter Rocks). Furthermore,
the frequency of smaller (1–20 mm) and medium-sized
(21–60 mm) individuals, primarily in the high intertidal,
was greatest at sites where large individuals were
abundant (Golden Rocks, Jagged Rocks, Point Adam).
The proportion of smaller individuals was relatively
higher in 2004 than in 2003 among those sites where
human and sea otter predation pressure was low and
adult Katharina densities were high.
the low and high intertidal at 11 sites along a gradient of
subsistence shoreline collection pressure.
Spatial variation in Katharina density and biomass in
Model factor df
Site 3 year
Site 3 year
Site 3 year
Site 3 year
ANNE K. SALOMON ET AL.1758
Vol. 17, No. 6
Model selection: analysis of multiple present-day causes
Strong evidence exists that harvest pressure and sea
otter presence were the two most influential variables
describing the spatial variation in Katharina density and
biomass in the low intertidal and Katharina density in
the high intertidal. Harvest pressure in combination with
sea otter presence comprised the best-fit models,
indicated by their high Akaike weights (wi; Table 3).
Individually, these two factors had the greatest variable
weights in all three cases (Table 4). Furthermore, their
all sites in (A) 2003 and (B) 2004. Sample sizes (n) are given in sequence: low intertidal zone, high intertidal zone.
Size–frequency distributions of Katharina tunicata in the low (shaded bars) and high (open bars) intertidal zone across
September 20071759 SERIAL DEPLETION OF MARINE INVERTEBRATES
relative importance was practically equal. However, the
importance of wave exposure was considerable when
predicting high intertidal Katharina biomass (Tables 3
and 4). The strength of evidence for the effect of bird
presence on Katharina density and biomass was weak.
Prehistoric and historic factors altering nearshore
Archaeological and historical records.—The faunal
remains of archaeological site SEL-027 were primarily
composed of macroinvertebrates, although sea otter,
harbor seal, sea lion, porpoise, beluga, porcupine, sea
ducks, albatross, salmon, cod, and halibut remains were
also present (W. Workman and K. Workman, unpub-
lished manuscript). Out of all the invertebrate remains,
the large predatory whelk Neptunea spp. was found in
the greatest proportion, comprising 42% of the late
prehistoric and 37% of the early historic invertebrates
excavated (Fig. 4). Although the proportion of Neptunea
in historic strata dropped by 5% compared to prehistoric
strata, the proportion of smaller predatory whelks
(Nucella spp. and Volutharpa ampullacea) increased by
a factor of 1.5 and 2, respectively, collectively making up
29% of the historic invertebrate remains. The remaining
bulk of invertebrates comprised lower trophic level
bivalves including; softshell clams (Mya truncata), butter
clams (Saxidomus gigantea), cockles (Clinocardium
nuttallii), Pacific little neck clams (Protothaca staminea),
and grazing periwinkle snails (Littorina spp.). Species
each comprising ,2% of the total invertebrates exca-
vated included limpets (Acmaea mitra, Lottia spp.,
Tectura spp.), other clams (Macoma spp.), puppet
margarite snails (Margarites pupillus), barnacles (Bala-
nus spp.), surf clams (Mactromeris polynyma), other
whelks (Fusitriton oregonensis, Natica spp.), jingle shells
(Pododesmus macroschisma), and Katharina. While
Katharina represented a minor component of the
invertebrates excavated, its occurrence more than
doubled from the late prehistoric occupation (0.4%) to
the early historic occupation (1.1%). Blue mussels
(Mytilus trossulus) and sea urchins (Strongylocentrotus
spp.) were present but not quantifiable.
Historical records indicate that Nanwalek was estab-
lished as a fur trading post in 1786. Port Graham was
established as a permanent settlement between 1909 and
1912 as a transfer point for shipping and a cannery site,
the latter playing a major role in the local economy for
much of this century (Cook and Norris 1998). As
Sugpiat became economically enmeshed with the fur
trade followed by the fishing industry, formerly dis-
persed semi-nomadic populations settled in more
permanent villages (de Laguna 1934, 1956, Stanek
Qualitative traditional knowledge.—Historical ecolog-
ical and socioeconomic data collected from Sugpiaq
elders and village residents highlighted temporal changes
in the relative abundance of invertebrate resources,
ing the spatial variation in Katharina density and biomass in
the low and high intertidal zone among 11 sites.
Strength of evidence for alternative models explain-
Response and modelNK
Density, low intertidal
Harvest þ sea otter
Harvest þ sea otter þ bird
Harvest þ sea otter þ wave
Biomass, low intertidal
Harvest þ sea otter
Sea otter þ wave
Density, high intertidal
Harvest þ sea otter
Harvest þ sea otter þ wave
Sea otter þ wave
Harvest þ sea otter þ bird
Biomass, high intertidal
Harvest þ wave
Sea otter þ wave
Bird þ wave
Harvest þ sea otter þ wave
Note: Models with varying numbers of parameters (K), were
compared using small-sample bias-corrected Akaike Informa-
tion Criterion (AICc), AICcdifferences (DAICc), and normal-
ized Akaike weights (wi).
presence, and wave exposure) that contribute to the current spatial variation in Katharina density
and biomass in both the high and low intertidal zones.
The relative importance of four variables (harvest pressure, sea otter presence, bird
Density, lowBiomass, low Density, highBiomass, high
Sea otter presence
Notes: Relative importance was based on variable weights, which were calculated by summing
the Akaike weights (wi) over the subset of models for a specific response in which a variable was
found. The sign of each variable coefficient in parentheses indicates the direction of the relationship
between the response and each variable.
ANNE K. SALOMON ET AL.1760
Vol. 17, No. 6
changes in subsistence use, and sea otter presence from
the 1920s to 2003 (Fig. 5A, Table 5). These data revealed
a serial decline of marine invertebrates beginning in the
early 1960s, co-occurring with the recovery of the local
sea otter population and increased shoreline harvest
efficiency (Table 5). According to local observations, sea
urchin (Strongylocentrotus spp.) and sea cucumber
(Cucumaria spp.) were the first invertebrates to decline
in the 1960s, followed by Dungeness crab (Cancer
magister) and shrimp (Pandalus spp.), which began
declining in the late 1970s and were rarely harvested by
the mid 1980s. Saxidomus, Prototheca, and Clinocardium
were the next invertebrates to decline. Localized declines
of Katharina were the most recent in a chain of declines
(Fig. 5A). Traditional knowledge holders offered a
variety of hypotheses regarding causes of historical
marine invertebrate declines and contemporary Katha-
rina declines (Table 6).
Quantitative subsistence shellfish landings.—From
1987 to 2003, Katharina was the primary invertebrate
harvested in Port Graham and Nanwalek, comprising
40.4% (62.8%) of the mean annual subsistence inverte-
brate landings during this time period (Fig. 5B).
Saxidomus gigantea and giant pacific octopus (Enter-
octopus dofleini) made up 27.1% (63.6%) and 15.7%
(61.8%) of the mean annual catch, respectively,
although their collection along with Katharina and most
marine subsistence species dropped in 1989, following
the Exxon Valdez oil spill. The remaining species, each
comprising .2% of the annual catch during this time
included Protothaca staminea (4.0% 6 1.2%), Clinocar-
dium nuttallii (3.2% 6 0.9%), razor clams (Siliqua patula)
(3.0% 6 0.5%), and Nucella spp. (2.5% 6 0.3%). In
1992, landings of Protothaca staminea began to decline
followed by a steep decline in Saxidomus giganteus.
Species making up ,2% of the annual catch from 1987
to 2003 included mussels (Mytilus spp.; 1.9% 6 0.5%),
gumboot chitons (Crytpochiton stelleri; 1.7% 6 0.7%),
Cancer magister (0.4% 6 0.1%), Tanner crab (Chionoe-
cetes bairdi; 0.2% 6 0.1%), Pandalus spp. (0.2% 6 0.1%),
whelks (Fusitriton oregonensis and Neptunea spp.; 0.09%
6 0.03%), Strongylocentrotus spp. (0.05% 6 0.02%), and
Cucumaria spp. (0.01% 6 0.01%).
Recent localized declines of the black leather chiton,
Katharina tunicata, on the outer tip of the Kenai
Peninsula, Alaska, USA can be attributed to changes
in socioeconomic and ecological dynamics. Strong
evidence from field surveys suggests that present-day
spatial variation in Katharina density and biomass is
driven by a combination of human harvest and sea otter
predation, with the relative magnitude of these top-
down factors varying among sites (Figs. 2 and 3, Tables
3 and 4, Appendix). However, the likely long-term
causal mechanism driving Katharina declines was
revealed through an investigation into the historical,
ecological, social, and economic dynamics of the local
and regional marine ecosystem. This temporal depth was
provided by archaeological data, historical records, and
USA, representing both late prehistoric (ca. AD 1300–1500) and early historic (20th century) occupation. Note that blue mussels
(Mytilus trosulus) and sea urchins (Strongylocentrotus spp.) were present but not quantifiable (W. Workman and K. Workman,
Relative percentage of invertebrate species excavated from archaeological site SEL-027 in Port Graham Bay, Alaska,
September 20071761SERIAL DEPLETION OF MARINE INVERTEBRATES
traditional ecological knowledge (Figs. 4 and 5, Tables 5
Historical factors altering nearshore ecosystem dynamics
We propose that five salient historical events likely
triggered the serial decline of marine invertebrates
leading to localized reductions of Katharina: (1) spatial
restriction of human impacts, (2) extirpation and
subsequent recovery of sea otters, (3) new technologies
leading to increased fishing efficiency and effort, (4)
regional commercial exploitation of crustacean stocks,
and (5) indirect socioeconomic effects of the 1989 Exxon
Valdez oil spill. We begin by describing prehistoric
Archaeological evidence from sites in lower Cook
Inlet dating back 4500 years suggests that prior to the
Russian occupation in the 1780s, Suqpiaq natives were
semi-nomadic, traveling from small settlements to
numerous seasonal camps for specific harvest activities
(Cook and Norris 1998, Steffian 2001, Stanek 2006).
Faunal remains excavated from middens in this region
indicate that Sugpiaq hunters and gatherers relied
dashed line represents the increase in outboard motor use. Key to species: sea urchin, Strongylocentrotus spp.; sea cucumber,
Cucumaria spp.; crab, Cancer magistor; shrimp, Pandalus spp.; clams and cockles, Saxidomus giganteus, Prototheca staminea, and
Clinocardium nuttallii; bidarkis, Katharina tunicata; sea otters, Enhydra lutris). (B) Subsistence invertebrate landings of species
comprising .2% of the annual invertebrate catch from Nanwalek and Port Graham, Alaska, USA, from 1987 to 2003 (Katharina
tunicata, Saxidomus giganteus, Enteroctopus dofleini, Protothaca staminea, Clinocardium nuttallii, Nucella spp., and Siliqua patula),
and local human population size (U.S. Census 2000, Brown et al. 2001).
(A) Serial depletion of marine invertebrates from 1920 to 2003 revealed through qualitative traditional knowledge. The
ANNE K. SALOMON ET AL.1762
Vol. 17, No. 6
heavily on intertidal macroinveretbrates, marine mam-
mals, sea birds, and fish (de Laguna 1934, Klein 1996,
Steffian 2001). In Port Graham Bay, harvest of the large
predatory whelk Neptunea, the main invertebrate
excavated from a 700-year-old midden (Fig. 4), likely
had important ecological ramifications for low intertidal
and subtidal benthic community dynamics. Evidence
from this and other archaeological sites suggests that the
nearshore ecosystems of the Kenai Peninsula have been
modified by this maritime culture for at least the past
After the arrival of the Russian fur traders in the
1780s, both commercial fur trading companies and the
Russian Orthodox Church sought to centralize services
in larger villages. Thus, regional consolidation eventu-
ally led to the demise of smaller settlements and the
creation of larger, more permanently established villages
(de Laguna 1956, Cook and Norris 1998, Stanek 2006).
By the late 1880s, commercial fishing and canneries
gradually replaced fur trading as the major source of
local jobs and income, again promoting more stationary
settlement patterns. Consequently, subsistence hunting
and shoreline collection effort likely became increasingly
concentrated in space, leading to an increase in localized
harvest impacts (Table 6).
Due to the lucrative fur trade, sea otters became
locally extirpated from Alaska’s coastline by the early
1900s with only several pockets of animals remaining
(Estes and Palmisano 1974). From the early 1900s to the
1950s, sea otters were never observed in Port Graham
Bay or its surrounding rocky shores by today’s elders
(Table 5). However, sea otter invertebrate prey, includ-
ing sea urchins, crab, clams, cockles, mussels, octopus,
and chitons, were abundant and kelp beds were sparse
(Table 5). With the extirpation of sea otters from Alaska
south to California, benthic macroinvertebrate prey
and tribal residents from Port Graham and Nanwalek, Alaska; Katharina are known locally as ‘‘bidarkis.’’
Time series of historic ecological and socioeconomic observations based on representative quotes from Sugpiaq elders
Time EventTraditional ecological knowledge—observation
1800s–1960ssea otter extirpation ‘‘When the Russians came they cleaned the sea otters out. When I was 18 yrs
old  there were no sea otters around Port Graham.’’
1930s–1950shigh invertebrate densities and
low kelp biomass
‘‘We used to be able to get all the Dungeness we wanted. We used to collect
clams and cockles, nobody ever missed a tide. I didn’t have concept of poor
or rich in a western world sense. We were so rich because there was so
much out there.’’
‘‘The sea back then was a dinner table set at low tides.’’
‘‘There was not much kelp in front of Nanwalek when I was young.’’
1960ssea otter recovery ‘‘They came back in the early 60s. The population exploded in the late 70’s
‘‘Boy, those things multiply!’’
1960sinvertebrate decline begins ‘‘We used to see sea urchins all over Nanwalek Reef in the early 1940s. By
the late 60s sea urchins were mostly gone.’’
1964Great Alaska earthquake ‘‘After the earthquake, there was sunk land and no minus tides for about four
years. After that it came back to normal.’’
‘‘The earthquake damaged the clam beds. This quake did not take the
bidarkis, snails, and other invertebrates. If it did, they came back.’’
1970s increased harvest effort with
increased storage abilities
‘‘In the past we picked just enough to eat and snack on. But when electricity
and then freezers came to the village, people began to pick more because
they could store them.’’
1980s commercial crustacean crash ‘‘[Dungeness] were wiped out because of commercial crab fisheries and
dragging. They came right into this bay. Now they haven’t been able to
come back because of the sea otters.’’
1989 Exxon Valdez oil spill‘‘The oil spill impacted nature’s cycles, the seasonal clock work of our culture,
our life ways ... It had lingering effects, not only in our water but in our
‘‘Clams and cockles and Dungeness crab were declining before the oil spill.
The oil spill may have made it worse but they were already declining before
1989 increased harvest efficiency‘‘People locally were hired to help clean up the spill. Then there was more
money that came to the village. More money let more people own more
boats and bigger boats with better outboards. Many people could now go to
places that they couldn’t go to in the past.’’
1990s change in bidarki numbers and
‘‘I started noticing bidarki declines 10–15 years ago.’’
‘‘It’s harder to find the big ones now.’’
1990s–2000s compensatory growth‘‘There are more little ones but they are not big enough to pick. I used to not
see so many little ones.’’
1990s–2000sserial decline ‘‘The urchins were the first to go, then crab, then the clams. Bidarkis, they’re
the most recent change.’’
September 20071763 SERIAL DEPLETION OF MARINE INVERTEBRATES
populations likely flourished throughout the early 1900s
(Tegner and Dayton 2000).
With the protection of sea otters in 1911, this keystone
predator began to recover along Alaskan coastlines. Sea
otters returned to the nearshore of Port Graham and
Nanwalek by the early 1960s (Table 5). Sea urchins,
most likely Strongylocentrotus droebachiensis, and sea
cucumbers, which were once plentiful on Nanwalek Reef
in the 1940s, were mostly gone by the late 1960s.
With the introduction of the cash economy in the
early 1900s, fishing boats that were once wooden dories
were gradually replaced by motorboats for fishing and
travel, thereby increasing harvest efficiency. By the early
1980s, 10 years after the introduction of electricity to the
villages, freezers began to be used by subsistence
harvesters to store food. This storage ability allowed
people to increase their harvest effort (Table 5).
In Cook Inlet, commercial crab and shrimp fisheries,
landings of which peaked in the early 1960s with inshore
harvests in bays like Port Graham Bay (Cook and
Norris 1998; Table 5), required increased effort and
movement offshore to maintain harvest levels. By the
early 1980s, crustacean stocks began to collapse
sequentially in the Gulf of Alaska (Orensanz et al.
1998). Coincident with the serial collapse of crustaceans
was a conspicuous shift in benthic species composition
from shrimp in the 1970s to ground fish in the 1980s
(Anderson and Piatt 1999) and variations in the
distribution and abundance of marine mammals and
sea birds (Springer et al. 1999), in part attributed to the
Pacific Decadal Oscillation and the resulting climatic
regime shift of 1977 (Mantua et al. 1997). Whatever the
cause, by the 1980s, native subsistence users found
Dungeness crab (C. magister) increasingly hard to
collect while their main competitor, the sea otter,
populations of which were thriving, were observed
consuming juvenile Dungeness crab, among other
invertebrates (Table 5).
The 1989 Exxon Valdez oil spill in neighboring Prince
William Sound had severe cultural, social, and ecolog-
ical effects in Port Graham, Nanwalek, and the
surrounding area, even though relatively little oil came
into Port Graham Bay (Table 5; Peterson et al. 2003,
Stanek 2006). Local people hired to help clean up the
spill earned unusually high incomes over a short period
of time. Many people bought new and better boats and
outboard motors. Although initially people avoided
subsistence foods for fear of oil contamination, subsis-
tence harvest largely resumed within a few years (Fig.
3B). With better boats, people could visit more shoreline
benthic marine invertebrates and Katharina (bidarkis) specifically.
Competing hypotheses, formulated from traditional and local knowledge, regarding the factors causing the decline of
Formal hypothesis Traditional ecological knowledge—cause of decline
Historical shift in collection
‘‘[Before the Russian occupation] when resources became low, people moved on. They took all
of their camp out. Then they would go back when resources returned. Villages didn’t exist,
there were seasonal camps. They always traveled, from fall to spring. That’s what is
happening here, we’re not moving.’’
Sliding baselines ‘‘Maybe people’s range of acceptable harvest sizes has increased.’’
Effective population size, local
‘‘We ship bidarkis to friends and family. Most go to Anchorage in zip-lock bags.’’
‘‘There are limits, limits of what you can harvest. Some people go beyond it.’’
‘‘It’s harder to find the bigger ones so I’m getting the smaller ones.’’
Recruitment overfishing ‘‘They are getting wiped out and are having trouble reproducing.’’
Lack of seasonal restrictions ‘‘March was the month our elders stopped us from hunting. The animals had little ones inside.
If you want to see them in the future, leave them alone. New generation, it’s not that way.
They go out and get whatever they want whenever they want.’’
Breakdown in information
transfer from elders to
‘‘Now, the new generation doesn’t have an understanding or meaning. Poor kids don’t know no
better. We elders haven’t told the younger ones what the nature does.’’
‘‘We are blaming the younger generation but we are to blame. We are not teaching them.’’
Sea otter predation ‘‘Sea otters are part of the problem...they eat every thing we eat.’’
Bird predation‘‘The bidarki remains you find on the beach mostly in the spring when the spring birds are
around when the sea birds come, nesting birds will eat them. That is when the bidarkis are
Increased harvest efficiency ‘‘Now everyone has a skiff and we can see the immediate impact on the resource.’’
Multiple causation and sequential
‘‘It’s not just overharvesting. Declines are due to a chain reaction. There is still to this day, no
one reason for all of these declines.’’
Change in human and sea otter
prey species breadth
‘‘Years ago, people didn’t only go for bidarkis, everything was available. Why would they want
to just hit the bidarkis? They had crab, mussels, and urchins. The sea otter will change their
diet, like any other animal, like us. What are they going to turn to? They turn to bidarkis.
Because that’s our only diet from here now.’’
‘‘People always used to have native food. People eat less native food now, but people still eat
ANNE K. SALOMON ET AL.1764
Vol. 17, No. 6
per tide and access beaches in previously prohibitive
conditions (Table 5). Consequently, and ironically, a
delayed indirect effect of the spill was an increase in
shoreline harvest efficiency likely leading to an increase
in invertebrate mortality due to fishing. Although the
spill may have had some direct effects on local marine
invertebrates, elders observed that shellfish declines
began prior to the spill (Table 5).
In sum, historical subsistence harvest differed in
several ways from today’s practices. In prehistoric times,
prior to European contact, subsistence harvest effort
was less spatially concentrated because communities
shifted among seasonal camps. With migration of
regional clans to trading posts throughout the 1800s,
intentional consolidation of native villages by the
Orthodox religion in the mid to late 1800s, and the
creation of canneries in the late 1880s, central-place
subsistence foraging gradually replaced dispersed opti-
mal foraging. The introduction of modern technologies
(freezers and better boats) facilitated increased harvest
effort and efficiency, contributing to increased fishing
mortality. Yet, even with the increased ability to travel
with better boats, central-place foraging among subsis-
tence shoreline collectors remains common, particularly
in the winter when stored salmon supplies caught the
previous spring become low and dangerous weather
prohibits travel. Lastly, according to traditional knowl-
edge, subsistence prey items from the 1920s to 1950s
included a wider range of invertebrates, such as whelks,
sea urchins, sea cucumbers, crab, mussels, clams, and
cockles, because all of these invertebrates were present in
abundance (Fig. 5A). Over the last decades of the 20th
century, these resources became scarce sequentially,
likely due to intensified consumption by an increasing
sea otter population, rising subsistence and regional
commercial harvest effort, and prey switching by sea
otters and human harvesters.
Synergistic serial depletion
We propose that the recent localized depletion of
Katharina is a consequence of the serial decline of
alternative prey leading to increased per capita predator
impacts by both humans and sea otters on Katharina.
Sea otters are well known to reduce a diversity of
herbivorous epibenthic macroinvertebrates (sea urchins,
chitons, and limpets) and filter feeding bivalves (clams,
cockles, and mussels) (Estes and Palmisano 1974,
Simenstad et al. 1978, Kvitek et al. 1992). In the early
1900s, the breadth of Sugpiaq subsistence diets was
broader than it is today, due to the availability of a wide
range of nearshore benthic invertebrates. Consequently,
human and sea otter predation pressure was formerly
distributed over several species with preferred and most
accessible prey items targeted first. As preferred species
became less abundant, we postulate that predation
pressure intensified on those species that remained,
resulting in a gradual decline in the number of species
consumed. Sequential prey switching provides a mech-
anism for serial declines. Consequently, the recent
localized depletion of Katharina may in fact reflect the
serial depletion of nearshore benthic invertebrates by
both humans and sea otters resulting in increased per
capita Katharina mortality due to a scarcity of alterna-
Empirical and theoretical evidence suggests that the
presence of alternative prey can modify the direct
interaction between predator and prey by reducing per
capita predator impact through predator satiation or
increased handling time (Wootton 2002). In the past, the
availability of sea urchin, sea cucumber, crab, clams,
cockles, and other alternative benthic invertebrate prey
may have reduced the per capita impact of humans and
sea otters on Katharina (Fig. 6A). Consequently,
present-day per capita predation rates on this strongly
interacting grazer are likely higher because alternative
prey items are now scarce (Fig. 6B, Table 2).
This hypothesis of synergistic serial depletion is
substantiated by the multispecies trends in benthic
invertebrate subsistence landings in Port Graham and
Nanwalek from 1987 to 2003 (Fig. 5B). Landings of
alternative prey (whelks, sea urchin, sea cucumber,
mussels, Dungeness crab, Tanner crab, and shrimp)
were low between 1987 and 2003, remaining well below
2% of the mean annual catch during this period. These
quantitative landings data are consistent with the
qualitative traditional knowledge that these invertebrate
abundances were in short supply due to previous
declines (Fig. 5A). Relative to sea urchin and crab, a
greater biomass of clams and cockles was landed during
the late 1980s, in accordance with the traditional
knowledge that these bivalves declined after sea urchin
and crab. Landings of bivalves began to decline by the
early 1990s, yet Katharina landings were greater in 1991
alternative prey reduces the per capita predator impacts (solid
black arrow) by humans and sea otters on Katharina biomass.
(B) When alternative prey items are scarce, per capita predator
impacts on Katharina are magnified.
(A) Interaction modification (dashed gray arrow) by
September 2007 1765 SERIAL DEPLETION OF MARINE INVERTEBRATES
and 1992, compared with 1987, and have remained high,
despite the dip in quantity in 1989 due the scare from the
Exxon Valdez oil spill.
Looking deeper into the past, dramatic shifts in the
species composition, trophic level, and general size of
subsistence shellfish species landed have occurred over
the past 700 years in Port Graham Bay, Alaska (Figs. 4
and 5B). While the large, upper-trophic-level predatory
whelk Neptunea comprised the greatest proportion of
invertebrate midden remains during the late prehistoric
and early historic occupation, it and other whelks, such
as Volutharpa, are all but absent from today’s subsis-
tence shellfish landings (Fig. 5B). Rather, Katharina, a
considerably smaller, lower trophic level grazer, is now
the primary invertebrate harvested in Port Graham and
Nanwalek, comprising 40% 6 3% (mean 6 SE) of the
mean annual invertebrate catch. Certainly, both sea
otters and humans are size-selective predators, generally
targeting and removing larger prey species and larger
individuals within a species. This is evident from the
variation in Katharina size–frequency distributions
across sites (Fig. 3) and from the archaeological data,
which suggests that Neptunea, reaching maximum
lengths of 18 cm (O’Clair and O’Clair 1998), were
targeted over Nucella, Volutharpa, which reach lengths
no greater than 5 cm. Yet, with the 5% reduction in
Neptunea landings from late prehistoric to early historic
occupations came 1.5–2 fold increase in the landings of
the smaller whelks Nucella and Volutharpa (Fig. 4). This
transition in invertebrate landings from large high
trophic level predators to smaller, mid-trophic level
predators and finally to lower trophic level herbivores
suggests that fishing down marine food webs (Pauly et
al. 1998) may have begun as early as AD 1300.
In many ecological studies, diet breadth and prey
preferences of predators are typically assumed to remain
constant. However, increasing evidence suggests that
behavioral flexibility by predators may be a common
phenomenon (Estes et al. 2004). Evolutionary theory
implies that consumers have evolved flexible responses
to varying environmental conditions, such as the ability
to substantially alter foraging strategies and food
sources. Like sea otters, fishermen tend to switch
‘‘targets’’ as relative abundance and market values
change. This reinforces the importance of understanding
human and nonhuman predator behavior when devel-
did in the past before outboard engines, standing at the stern, facing forward to watch for oncoming seas. Vera is among many
subsistence resource users in the area who have observed changes in the abundance and behavior of species at particular points in
time and space, and can offer insight into the ecological and socio-economic factors driving ecosystem dynamics now and in the
past. Photo credit: Lisa Williams, June 2005.
On her way to jig for halibut in Port Graham Bay, Alaska, Vera Meganack, a Sugpiaq Elder, rows her skiff as they
ANNE K. SALOMON ET AL.1766
Vol. 17, No. 6
oping sustainable fisheries strategies and effective
marine conservation plans.
Compounding social complexities contributing
to invertebrate decline
Certainly, present-day causal mechanisms other than
overharvest and sea otter predation have contributed to
black leather chiton declines (Table 6). Elders point to
the deterioration in information transfer to the younger
generation of harvesters as a critical problem leading to
overall resource declines. Katharina sizes deemed ac-
ceptable by younger harvesters are smaller than those
used by today’s elders in their own youth (i.e., shifting
baselines). Furthermore, traditional subsistence man-
agement practices such as seasonal restrictions are no
longer being adhered to. Lastly, the effective human
predator population size in these villages is greater than
the local human population size. Teenagers, young
adults, and the ill often leave the villages for schooling
or health care provided by local city centers (Stanek
2006), yet are still sent subsistence food items, as are
relatives who have moved away (Table 6). This again
illustrates the importance of understanding the social
dynamics motivating human behavior.
Current ecological complexities contributing
to Katharina decline
Additional ecological complexities may be contribut-
ing indirectly to Katharina declines. We observed that
Katharina recruitment is typically high where adult
density is high (Fig. 3). This could be due to the
maintenance of GABA (gamma-aminobutyric acid)
producing coralline crusts by adult Katharina actively
grazing on turfing and foliose macroalgae. GABA is a
recognized chemical settling cue for many molluscan
species, including Katharina (Strathmann 1987); conse-
quently, coralline crusts form important settlement
habitat for invertebrate larvae, facilitating recruitment
and, in turn, future grazing. Conversely, where adult
Katharina densities have been reduced by harvest and/or
sea otter predation, Katharina recruitment tends to be
low (Fig. 3). Reduced densities of adult Katharina may
indirectly impede Katharina recruitment by releasing
macroalgae from grazing pressure and allowing species
like Endocladia muricatum in the high intertidal and
Alaria spp. in the mid to low intertidal to flourish. These
macroalgal species can overgrow and shade out encrust-
ing corallines thereby reducing valuable settlement
habitat and the production of the settling cue GABA.
Furthermore, macroalgae can mechanically sweep away
larvae competent to settle. In sum, food web alterations
caused by the exploitation of a dominant consumer may
in fact hinder recruitment of that very consumer.
Alternatively, when adult densities are low, local
recruitment may be reduced simply due to limited larval
supply. Katharina reach reproductive maturity at ;35
mm, their reproductive output increases cubically with
length, and larvae are pelagic for ;6 days (Strathmann
1987, Salomon et al. 2006). Katharina subpopulations
may be heavily reliant on local sources of larvae,
particularly where local currents, eddies, and boundary
layer effects impede widespread dispersal and facilitate
local recruitment. Certainly, local retention is more
common than previously thought (Jones et al. 1999,
Swearer et al. 1999), and the risk of recruitment
overfishing is becoming increasingly recognized (Walters
and Kitchell 2001; Table 6). In the context of fisheries
management, there is now broad empirical evidence
suggesting that low parental stock sizes can result in
lower mean recruitment (Myers et al. 1999).
Integrating multiple data sources
By drawing on multiple data sources, we have pieced
together a conceptual model of the local and regional
ecosystem that sheds insight into the likely causes of
recent Katharina declines. Contemporary field surveys
provided evidence that predation by humans and sea
otters influenced Katharina abundance, density, and
biomass. Interviews with elders offered evidence of serial
declines of nearshore invertebrates over time, together
with changes in human subsistence activity. A time series
of subsistence invertebrate landings afforded a quanti-
tative calibration and validation of traditional knowl-
edge from 1987 to 2003. Finally, archaeological data and
historical records revealed coarse changes in the
invertebrate catch and the spatial distribution of
subsistence harvest pressure, allowing us to scale our
contemporary research with a deeper time perspective.
The hypothesis of synergistic serial depletion emerged
from multiple lines of evidence, quantitative and
qualitative, current and historical, considered simulta-
Value of traditional knowledge and historical perspectives
Although often neglected in ecological studies, his-
torical data are vital for revealing the ‘‘ghosts’’ of
ecosystem past, the true magnitude of change of
ecosystems present, and the dynamics that link the two
(Pauly 1995, Dayton et al. 1998). Here, an analysis of
present-day ecological data alone could not have
explained the ultimate causal mechanisms governing
recent localized Katharina declines. Rather, knowing the
historical change in alternative invertebrate prey, human
settlement patterns, subsistence harvest practices, and
predatory population dynamics of sea otters was critical
to our current understanding of the factors leading to
recent declines of the Katharina, linking the results of
our fieldwork with longer term social–ecological dy-
namics. We argue that the explanation of decline was
revealed only through the examination of specific
historical events and an understanding of how these
events may have led to the presence or absence of key
consumers in the community.
Here, the value of western science was in developing
the quantitative relationship between predation pressure
and resource density. The value of traditional knowl-
September 2007 1767 SERIAL DEPLETION OF MARINE INVERTEBRATES
edge was in providing qualitative assessments of shifts in
predator and prey abundance and subsistence behavior
prior to ecological study. In this case, the two methods
agree on the major direct drivers of environmental
change, yet traditional knowledge provided important
historical ecological, socioeconomic, and cultural con-
siderations that were otherwise not available.
Indeed, debate exists over the relationship between
traditional knowledge and science (Berkes 1999) and
even over the degree to which a simple dichotomy
between the two withstands scrutiny (Agrawal 1995).
Certainly, both knowledge systems are not without their
limitations, and skepticism regarding the use of tradi-
tional ecological knowledge (TEK) exists for multiple
reasons (Huntington 2000, Moller et al. 2004). In
addition, drawing on TEK requires attention to cultural
context to avoid misinterpretation or misrepresentation
(Smith 1999, Huntington et al. 2006). Despite the
limitations, we bridge these two alternative yet comple-
mentary knowledge systems and profit from their
strengths to inform the causal mechanisms driving
Katharina declines. Through the documentation of
TEK via interviews and collaborative fieldwork, we
draw on TEK as one of several sources of information
needed to reconstruct ecological history. Perhaps most
importantly, exchanging and combining scientific and
traditional knowledge, while using a participatory
research approach, built community partnerships, raised
awareness, and provided quantitative and qualitative
data to inform future conservation strategies.
Ultimately, assessing the relative magnitudes and
relationships among human impacts, interspecific inter-
actions, and physical factors will reduce our uncertainty
in detecting drivers of change and increase our
likelihood of designing effective conservation strategies
for nearshore ecosystems (Salomon et al. 2001, Salomon
et al. 2002). However, management will fail if it focuses
on the most recent symptoms of decline rather than on
its deep historical causes (Jackson et al. 2001). This
research showcases the insight ecologists can glean from
delving into both ecological and social history. By
considering pivotal socioeconomic drivers across multi-
ple scales in time and space and integrating western
science and traditional knowledge, we obtained an
enhanced understanding of the causes driving Katharina
declines and consequently are now better equipped to
collaboratively develop an effective conservation plan
for the nearshore. This was and continues to be a
complex system subject to the vagaries of natural
predators, the physical environment, and socioeconomic
factors which motivate human behavior.
It is with deep gratitude that we thank the elders, Tribal
Councils, and residents of Port Graham and Nanwalek, Alaska,
for sharing their ecological insights and ocean home—
Quyanaa. Special thanks to Chief P. Norman, Second Chief
J. Kvasnikoff, past Chief E. Swenning, V. Yeaton, and L.
McMullen for their enthusiasm and assistance throughout this
research. Field assistance from D. Glahn, C. Harley, K.
Lestenkoff, P. McCollum, G. McMullen, M. McMullen, J.
McMullen, J. Miller, M. Norman, L. Villarreal, and N. Yeaton
was instrumental. This manuscript was greatly improved with
comments by J. Ruesink, E. Buhle, R. Paine, D. Boersma, D.
Schindler, J. Watson, and an anonymous reviewer. W.
Workman, R. Stanek, and the Alaska Department of Fish
and Game, Division of Subsistence contributed valuable data
and thoughtful suggestions. A. Malhotra kindly provided the
GIS maps. The Center for Alaskan Coastal Studies and the
Kachemak Bay National Estuarine Research Reserve offered
logistical support. This research was funded by the Gulf
Ecosystem Monitoring and Research Program, NOAA, and
an NSERC postgraduate fellowship to A. K. Salomon.
Agrawal, A. 1995. Dismantling the divide between indigenous
and scientific knowledge. Development and Change 26:413–
Anderson, P. J., and J. F. Piatt. 1999. Community reorgani-
zation in the Gulf of Alaska following ocean climate regime
shift. Marine Ecological Progress Series 189:117–123.
Bell, E. C., and M. W. Denny. 1994. Quantifying wave
exposure: a simple device for recording maximum velocity
and results of its use at several field sites. Journal of
Experimental Marine Biology and Ecology 181:9–29.
Berkes, F. 1999. Sacred ecology—traditional ecological knowl-
edge and resource management. Taylor and Francis,
Philadelphia, Pennsylvania, USA.
Berkes, F., J. Colding, and C. Folke. 2000. Rediscovery of
traditional ecological knowledge as adaptive management.
Ecological Applications 10:1251–1262.
Berkes, F., R. Huebert, H. Fast, M. Manseau, and A. Diduck.
2005. Breaking ice; renewable resource and ocean manage-
ment in the Canadian north. University of Calgary Press,
Calgary, Alberta, Canada.
Brown, L. A., G. B. Jennings, C. L. Scott, and C. J. Utermohle.
2001. Community profile database. Alaska Department of
Fish and Game, Division of Subsistence, Anchorage, Alaska,
Burnham, K. P., and D. R. Anderson. 1998. Model selection
and multimodel inference; a practical information–theoretic
approach. Springer, New York, New York, USA.
Carpenter, S. R., and L. H. Gunderson. 2001. Coping with
collapse: ecological and social dynamics in ecosystem
management. BioScience 51:451–457.
Castilla, J. C., and L. R. Dura ´ n. 1985. Human exclusion from
the rocky intertidal zone of central Chile: the effects on
Concholepas concholepas (Gastropoda). Oikos 45:391–399.
Clark, C. W. 1981. Bioeconomics of the ocean. BioScience 31:
Cook, L., and F. Norris. 1998. A stern and rock-bound coast;
Kenai Fjords National Park Historic Resource Study.
National Parks Service, Anchorage, Alaska, USA.
Cruikshank, J. 2001. Glaciers and climate change: perspectives
from oral tradition. Arctic 54:377–393.
Dayton, P. K. 1971. Competition, disturbance, and community
organization: the provision and subsequent utilization of
space in a rocky intertidal community. Ecological Mono-
Dayton, P. K., M. J. Tegner, P. B. Edwards, and K. L. Riser.
1998. Sliding baselines, ghosts, and reduced expectations in
kelp forest communities. Ecological Applications 8:309–322.
de Laguna, F. 1934. The archaeology of Cook Inlet, Alaska.
University of Pennsylvania Museum, Philadelphia, Pennsyl-
de Laguna, F. 1956. Chugach prehistory: the archaeology of
Prince William Sound, Alaska. University of Washington
Press, Seattle, Washington, USA.
ANNE K. SALOMON ET AL.1768
Vol. 17, No. 6
Drew, J. A. 2005. Use of traditional ecological knowledge in
marine conservation. Conservation Biology 19:1286–1293.
Duggins, D. O., C. A. Simenstad, and J. A. Estes. 1989.
Magnification of secondary production by kelp detritus in
coastal marine ecosystems. Science 245:170–173.
Ellis, D. W., and L. Swan. 1981. Teaching of the tides; uses of
marine invertebrates by the Manhousat People. Theytus
Books, Nanaimo, British Columbia, Canada.
Estes, J. A., E. M. Danner, D. F. Doak, B. Konar, A. M.
Springer, P. D. Steinberg, M. T. Tinker, and T. M. Williams.
2004. Complex trophic interactions in kelp forest ecosystems.
Bulletin of Marine Science 74:621–638.
Estes, J., and J. Palmisano. 1974. Sea otters: their role in
structuring nearshore communities. Science 185:1058–1060.
Francis, R., S. Hare, A. Hollowed, and W. Wooster. 1998.
Effects of interdecadal climate variability on the oceanic
ecosystem of the NE Pacific. Fisheries Oceanography 3:279–
Fretwell, S. D. 1977. The regulation of plant communities by
the food chains exploiting them. Perspectives in Biology and
Gunderson, L. H., and C. S. Holling. 2002. Panarchy:
understanding transformations in human and natural sys-
tems. Island Press, Washington, D.C., USA.
Hilborn, R., and S. C. Stearns. 1982. On inference in ecology
and evolutionary biology—the problem of multiple causes.
Acta Biotheoretica 31:145–164.
Holling, C. S. 1959. The components of predation as revealed
by a study of small mammal predation of the European pine
sawfly. Canadian Entomologist 91:293–320.
Holt, R. D., and J. H. Lawton. 1994. The ecological
consequences of shared natural enemies. Annual Review of
Ecology and Systematics 25:495–520.
Huntington, H. P. 1998. Observations on the utility of the semi-
directive interview for documenting traditional ecological
knowledge. Arctic 51:237–242.
Huntington, H. P. 2000. Using traditional ecological knowledge
in science: methods and applications. Ecological Applications
Huntington, H. P., S. F. Trainor, D. C. Natcher, O.
Huntington, L. DeWilde, and F. S. Chapin, III. 2006. The
significance of context in community-based research: under-
standing discussions about wildfire in Huslia, Alaska.
Ecology and Society 11.
Hutchings, J. A., and R. A. Myers. 1994. What can be learned
from the collapse of a renewable resource—Atlantic cod,
gadus-morhua, of Newfoundland and Labrador. Canadian
Journal of Fisheries and Aquatic Sciences 51:2126–2146.
Jackson, J. B. C., et al. 2001. Historical overfishing and the
recent collapse of coastal ecosystems. Science 293:629–638.
Johannes, R. E. 1998. The case for data-less marine resource
management: examples from tropical nearshore finfisheries.
Trends in Ecology and Evolution 13:243–246.
Johannes, R. E. 2002. The renaissance of community-based
marine resource management in Oceania. Annual Review of
Ecology and Systematics 33:317–340.
Jones, G. P., M. J. Milicich, M. J. Emslie, and C. Lunow. 1999.
Self-recruitment in a coral reef fish population. Nature 402:
Kari, P. N. 1991. English Bay and Port Graham Alutiiq
plantlore: an ethnobotany of the peoples of English Bay and
Port Graham, Kenia Peninsula, Alaska. Pratt Museum,
Homer, Alaska, USA.
Karpov, K. A., P. L. Khaaker, I. K. Taniguchi, and L. Rogers-
Bennett. 2000. Serial depletion and the collapse of the
California abalone fishery. Pages 11–24 in A. Campbell,
editor. Workshop on rebuilding abalone stocks in British
Columbia. Canadian Special Publication of Fishes and
Aquatic Sciences, Guelph, Ontario, Canada.
Klein, J. R. 1996. Archaeology of Kachemak Bay, Alaska.
Kachemak Country Publications, Homer, Alaska, USA.
Kvitek, R. G., J. S. Oliver, A. R. DeGange, and B. S.
Anderson. 1992. Changes in Alaskan soft-bottom prey
communities along a gradient in sea otter predation. Ecology
Larkin, P. A. 1978. Fisheries management—essay for ecolo-
gists. Annual Review of Ecology and Systematics 9:57–73.
Levin, S. 1992. The problem of pattern and scale in ecology.
Lewontin, R. C. 1969. The meaning of stability. Brookhaven
Symposium in Biology 22:13–24.
Ludwig, D., R. Hilborn, and C. Walters. 1993. Uncertainty,
resource exploitation, and conservation: lessons from history.
Mantua, N., S. Hare, Y. Zhang, J. Wallace, and R. Francis.
1997. A Pacific interdecadal climate oscillation with impacts
on salmon production. Bulletin of American Metereological
Markel, R. W., and R. E. DeWreede. 1998. Mechanisms
underlying the effect of the chiton Katharina tunicata on the
kelp Hedophyllum sessile: size escapes and indirect effects.
Marine Ecology Progress Series 166:151–161.
Menge, B. A. 2000. Top-down and bottom-up community
regulation in marine rocky intertidal habitats. Journal of
Experimental Marine Biology and Ecology 250:257–289.
Menge, B. A., C. A. Blanchette, P. Raimondi, T. Freidenburg,
S. Gaines, J. Lubchenco, D. Lohse, G. Hudson, M. Foley,
and J. Pamplin. 2004. Species interaction strength: testing
model predictions along an upwelling gradient. Ecological
Menge, B. A., and G. M. Branch. 2001. Rocky intertidal
communties. Pages 221–251 in M. D. Bertness, S. D. Gaines,
and M. E. Hay, editors. Marine community ecology. Sinauer
Associates, Sunderland, Massachusetts, USA.
Moller, H., F. Berkes, P. Lyver, and M. Kislalioglu. 2004.
Combining science and traditional ecological knowledge:
monitoring populations for co-management. Ecology and
Murawski, S. A. 2000. Definitions of overfishing from an
ecosystem perspective. ICES Journal of Marine Science 57:
Myers, R. A., K. G. Bowen, and N. J. Barrowman. 1999.
Maximum reproductive rate of fish at low population sizes.
Canadian Journal of Fisheries and Aquatic Science 56:2404–
Nielsen, K. J., and S. A. Navarrete. 2004. Mesoscale regulation
comes from the bottom-up: intertidal interactions between
consumers and upwelling. Ecology Letters 7:31–41.
NRC. 1999. Sustaining marine fisheries. National Academy
Press, Washington, D.C., USA.
NRC. 2003. Decline of the stellar sea lion in Alaskan waters;
untangling food webs and fishing nets. The National
Academies Press, Washington, D.C., USA.
O’Clair, R. M., and C. E. O’Clair. 1998. Southeast Alaska’s
rocky shores. Plant Press, Auk Bay, Alaska, USA.
Oksanen, L., S. D. Fretwell, J. Arruda, and N. Pekka. 1981.
Exploitation ecosystems in gradients of primary productivity.
American Naturalist 118:240–261.
Orensanz, J. M. L., J. Armstrong, D. Armstrong, and R.
Hilborn. 1998. Crustacean resources are vulnerable to serial
depletion—the multifaceted decline of crab and shrimp
fisheries in the Greater Gulf of Alaska. Reviews in Fish
Biology and Fisheries 8:117–176.
Paine, R. T. 1966. Food web complexity and species diversity.
The American Naturalist 100:65–75.
Paine, R. T. 1992. Food-web analysis through field measure-
ment of per capita interaction strength. Nature 355:73–75.
Paine, R. T. 2002. Trophic control of production in a rocky
intertidal community. Science 296:736–739.
Paine, R. T., M. J. Tegner, and E. A. Johnson. 1998.
Compounded perturbations yield ecological surprises. Eco-
Pauly, D. 1995. Anecdotes and the shifting baseline syndrome
of fisheries. Trends in Ecology and Evolution 10:430.
September 20071769SERIAL DEPLETION OF MARINE INVERTEBRATES
Pauly, D., V. Christensen, J. Dalsgaard, R. Froese, and F. Download full-text
Torres, Jr. 1998. Fishing down marine food webs. Science
Peterson, C. H., S. D. Rice, J. W. Short, D. Esler, J. L. Bodkin,
B. E. Ballachey, and D. B. Irons. 2003. Long-term ecosystem
response to the Exxon Valdez oil spill. Science 302:2082–
Pfister, C. A., and F. R. Stevens. 2002. The genesis of size
variability in plants and animals. Ecology 83:59–72.
Pierotti, R., and D. Wildcat. 2000. Traditional ecological
knowledge: the third alternative (commentary). Ecological
Sala, E., O. Aburto-Oropeza, M. Reza, G. Paredes, and L.
Lopez-Lemus. 2004. Fishing down coastal food webs in the
Gulf of California. Fisheries 29:19–25.
Salomon, A. K., J. L. Ruesink, and R. E. DeWreede. 2006.
Population viability, ecological processes and biodiversity:
valuing sites for reserve selection. Biological Conservation
Salomon, A. K., J. L. Ruesink, B. X. Semmens, and R. T.
Paine. 2001. Incorporating human and ecological communi-
ties in marine conservation; an alternative to Zacharias and
Roff. Conservation Biology 15:1452–1455.
Salomon, A. K., N. Waller, C. McIlhagga, R. Yung, and C.
Walters. 2002. Modeling the trophic effects of marine
protected area zoning policies: a case study. Aquatic Ecology
Sanford, E. 1999. Regulation of keystone predation by small
changes in ocean temperature. Science 283:2095–2097.
Simenstad, C. A., J. A. Estes, and K. W. Kenyon. 1978. Aleuts,
sea otters, and alternate stable-state communities. Science
Smith, L. T. 1999. Decolonizing methodologies: research and
indigenous peoples. Zed Books, London, UK.
Springer, A. M., J. A. Estes, G. B. van Vliet, T. M. Williams,
D. F. Doak, E. M. Danner, K. A. Forney, and B. Pfister.
2003. Sequential megafaunal collapse in the North Pacific
Ocean: An ongoing legacy of industrial whaling? Proceedings
of the National Academy of Sciences (USA) 100:12223–
Springer, A. M., J. F. Piatt, V. P. Shuntov, G. B. van Vliet,
V. L. Valdimirov, A. E. Kuzin, and A. S. Perlov. 1999.
Marine birds and mammals of the subarctic gyers. Progress
in Oceanography 43:443–487.
Stanek, R. T. 1985. Patterns of wild resource use in English Bay
and Port Graham, Alaska. Technical Paper No. 104. Alaska
Department of Fish and Game Division of Subsistence,
Anchorage, Alaska, USA.
Stanek, R. T. 2006. Port Graham. In J. Fall, editor. Update of
the status of subsistence uses in Exxon Valdez oil spill area
communities. Exxon Valdez Oil Spill Restoration Project
Final Report (Restoration Project 040471). Alaska Depart-
ment of Fish and Game, Division of Subsistence, Anchorage,
Steffian, A. F. 2001. Cummillallret—Ancestors. In A. L.
Crowell, A. F. Steffian, and G. L. Pullar, editors. Looking
both ways: heritage and identity of the Alutiiq people.
University of Alaska Press, Fairbanks, Alaska, USA.
Strathmann, M. F. 1987. Reproduction and development of
marine invertebrates on the Northern Pacific coast: data and
methods for the study of eggs, embryos, and larvae.
University of Washington Press, Seattle, Washington, USA.
Swearer, S. E., J. E. Caselle, D. W. Lea, and R. R. Warner.
1999. Larval retention and recruitment in an island
population of a coral-reef fish. Nature 402:799–802.
Tegner, M. J., and P. K. Dayton. 2000. Ecosystem effects of
fishing in kelp forest communities. ICES Journal of Marine
Turner, N. J. 2003. The ethnobotany of edible seaweed
(Porphyra abbottae and related species; Rhodophyta: Ban-
giales) and its use by First Nations on the Pacific Coast of
Canada. Canadian Journal of Botany–Revue Canadienne De
Turner, N. J., and F. Berkes. 2006. Coming to understanding:
developing conservation through incremental learning in the
Pacific Northwest. Human Ecology 34:495–513.
U.S. Census. 2000. Community Database. Department of
Commerce, Community and Economics, Washington, D.C.,
Walters, C., and J. F. Kitchell. 2001. Cultivation/depensation
effects on juvenile survival and recruitment: implications for
the theory of fishing. Canadian Journal of Fisheries and
Aquatic Science 58:39–50.
Wooton, J. T. 1992. Indirect effects, prey susceptibility, and
habitat selection: impacts of birds on limpets and algae.
Wootton, T. J. 2002. Indirect effects in complex ecosystems:
recent progress and future challenges. Journal of Sea
Maps of the spatial distribution and abundance of sea otters (Enhydra lutris) observed during three small boat surveys conducted
in the summer of 2004 in and around Port Graham Bay, Alaska, USA (Ecological Archives A017-073-A1).
ANNE K. SALOMON ET AL.1770
Vol. 17, No. 6