Content uploaded by Scott Henderson
Author content
All content in this area was uploaded by Scott Henderson on Nov 17, 2017
Content may be subject to copyright.
El Nin˜ o, grazers and fisheries interact to greatly elevate
extinction risk for Galapagos marine species
GRAHAM J. EDGAR
*
wz,STUARTA.BANKS
*
, MARGARITA BRANDT
*
,RODRIGOH.
BUSTAMANTE§, ANGEL CHIRIBOGA
*
,SYLVIAA.EARLE}, LAUREN E. GARSKEk,
PETER W. GLYNN
**
,JACKS.GROVEww, SCOTT HENDERSONzz, CLEVE P. HICKMAN§§,
KATHY A. MILLER}},FERNANDORIVERA
*
,
kk andGERALD M. WELLINGTON
***
*
Charles Darwin Foundation, Puerto Ayora, Galapagos, Ecuador, wTasmanian Aquaculture and Fisheries Institute, University of
Tasmania, PO Box 252-49, Hobart, Tasmania 7001, Australia, zConservation International, 2011 Crystal Drive Suite 500,
Arlington, VA 22202, USA, §CSIRO Marine and Atmospheric Research, PO Box 120, Cleveland 4163, Qld, Australia, }National
Geographic Society, 1145 17th Street NW, Washington, DC 20036, USA, kBodega Marine Lab, University of California at Davis,
2099 Westside Road, Bodega Bay, CA 94923, USA,
**
Division of Marine Biology and Fisheries, Rosenstiel School of Marine &
Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149-1098, USA, wwSection of Fishes
Natural History Museum of Los Angeles County, 900 Exposition Blvd., Los Angeles, CA 90007 USA, zzConservation International,
Puerto Ayora, Galapagos, Ecuador, §§Department of Biology, Washington & Lee University, Lexington, VA 24450, USA,
}}University Herbarium, 1001 Valley Life Sciences Building #2465, University of California, Berkeley, CA 94720-2465, USA,
kkDepartment of Zoology, University of Melbourne, Parkville, Vic. 3010, Australia,
***
Department of Biology and Biochemistry,
University of Houston, Houston, TX 77204-5001, USA
Abstract
Comparisons between historical and recent ecological datasets indicate that shallow reef
habitats across the central Galapagos Archipelago underwent major transformation at the
time of the severe 1982/1983 El Nin˜o warming event. Heavily grazed reefs with crustose
coralline algae (‘urchin barrens’) replaced former macroalgal and coral habitats, resulting in
large local and regional declines in biodiversity. Following recent threat assessment work-
shops, a total of five mammals, six birds, five reptiles, six fishes, one echinoderm, seven corals,
six brown algae and nine red algae reported from coastal environments in Galapagos are now
recognized as globally threatened. The 2008 International Union for the Conservation of
Nature (IUCN) Red List includes 43 of these species, while two additional species (Galapagos
damsel Azurina eupalama and 24-rayed sunstar Heliaster solaris) not seen for 425 years also
fulfil IUCN threatened species criteria. Two endemic species (Galapagos stringweed Bifurcar-
ia galapagensis and the damselfish A. eupalama) are now regarded as probably extinct, while
an additional six macroalgal species (Dictyota galapagensis,Spatoglossum schmittii,Desmar-
estia tropica,Phycodrina elegans,Gracilaria skottsbergii and Galaxaura barbata) and the
seastar H. solaris are possibly extinct. The removal of large lobster and fish predators by
artisanal fishing probably magnified impacts of the 1982/1983 El Nin˜o through a cascade of
indirect effects involving population expansion of grazing sea urchins. Marine protected areas
with adequate enforcement are predicted to ameliorate but not eliminate ecosystem impacts
caused by increasing thermal anomalies associated with El Nin˜o and global climate change.
Keywords: climate change, effects of fishing, marine protected area, sea urchin, threatened species,
trophic cascade
Received 19 March 2009 and accepted 29 September 2009
Introduction
Through experimentation and ecosystem monitoring,
the Galapagos Archipelago potentially provides a glob-
ally unique ‘field laboratory’ for assessing impacts of
extreme oceanographic warming on marine biodiver-
sity. Galapagos sits near the centre of the most intense El
Nin
˜o events (Glynn & Ault, 2000) and is less impacted
by human activity than other large tropical and tempe-
rate archipelagos.
Dating back to the voyage of the Beagle in 1835 (Grove
& Lavenberg, 1997), Galapagos historically has also
attracted disproportionate scientific attention because
Correspondence: G. Edgar, Tasmanian Aquaculture and Fisheries
Institute, University of Tasmania, Private Bag 49, Hobart,
Tasmania 7001, Australia, tel. 161 3 6227 238, fax 161 3 6227
8035, e-mail: g.edgar@utas.edu.au
Global Change Biology (2010) 16, 2876–2890, doi: 10.1111/j.1365-2486.2009.02117.x
2876 r2009 Blackwell Publishing Ltd
of its fragmented landscapes, endemic flora and fauna,
isolation on the equator and reputation as a living
laboratory of evolution (Bensted-Smith, 2002). Ende-
mism rates for marine species are taxonomically vari-
able (8–67%) but generally much higher than most other
archipelagos for major marine groups (Bustamante
et al., 2000a; Hickman, 2009).
El Nin
˜o events appear to have strengthened in the
eastern tropical Pacific over the past millennium, with
unprecedented heating recorded over the past 50 years
(Conroy et al., 2009). Whether or not these changes are
directly caused by global warming, recent extreme El
Nin
˜o events provide case studies for assessing likely
impacts of climate change on ecosystems. Global warm-
ing will presumably affect communities primarily
through increasing magnitude and frequency of
extreme events comparable to El Nin
˜o rather than
through gradual changes in ocean climate (Reaser
et al., 2000; Boer et al., 2004).
Despite the comparative regularity of El Nin
˜o, field
observations indicate that Galapagos marine ecosys-
tems are not well adapted for extreme thermal impacts.
Intertidal shores and shallow rocky and coral reef
habitats across Galapagos changed substantially at the
time of the 1982/1983 El Nin
˜o (Robinson, 1985; Glynn,
1994). Although probably not unprecedented at the 100-
year scale (Enfield, 2001), this extreme thermal anomaly
was characterized by water temperatures up to 51C
above long-term averages (Barber & Chavez, 1986).
Elevated temperatures persisted for over 12 months
generating clear oceanic waters, major declines in dis-
solved nutrients and a decline in phytoplankton pro-
ductivity that ultimately resulted in a prolonged
reduction of biomass at the base of the marine food
chain (Robinson & Del Pino, 1985).
Vertebrate animals were much affected by changes to
the food web associated with the 1982/1983 El Nin
˜o.
Populations of endemic Galapagos penguins (Sphenis-
cus mendiculus), flightless cormorants (Phalacrocorax har-
risi), Galapagos fur seals (Arctocephalus galapagoensis)
and Galapagos sea-lions (Zalophus wollebaeki) were esti-
mated to have declined by 78%, 45%, 60% and 35%,
respectively (Robinson & Del Pino, 1985; Trillmich &
Limberger, 1985).
Habitat changes on shallow Galapagos reefs cata-
lysed by the 1982/1983 El Nin
˜o included loss of coral
reefs through bleaching and subsequent sea urchin
bioerosion (Glynn & Wellington, 1983; Glynn, 1990,
1994), and loss of most macroalgal beds (Robinson &
Del Pino, 1985). A total of 95–99% of reef coral cover
was lost from Galapagos between 1983 and 1985. All
known coral reefs based on calcareous frameworks died
and subsequently disintegrated to rubble and sand
(Glynn, 1994).
Oceanographic warming associated with both El
Nin
˜o and global climate change can operate directly
on species’ populations (Thomas et al., 2004, 2006;
Helmuth et al., 2006), or indirectly through interaction
with other factors (Occhipinti-Ambrogi, 2007; Poertner
& Knust, 2007; Crain et al., 2008), including factors
associated with human activity (Hughes et al., 2005).
In the marine context, interactions between climate
change and fishing have been suggested to have far-
reaching consequences for species persistence (Hughes
et al., 2003; Harley & Rogers-Bennett, 2004; Edgar et al.,
2005; Pandolfi et al., 2005).
In Galapagos, published studies and our local obser-
vations extending over 40 years suggested that popula-
tions of many marine species had not recovered since
catastrophic declines at the time of the 1982/1983 El
Nin
˜o (see Robinson, 1985). Anecdotal observations also
indicated that Galapagos reefs had been ecologically
affected by excessive fishing pressure (Ruttenberg, 2001;
Sonnenholzner et al., 2009). We here consider the valid-
ity of these observations by assessing the hypothesis
that extreme heating events can cause regime shifts in
marine ecosystems and extinction of species. We use
historical data and reports, quantitative field surveys,
directed searches for threatened taxa and outcomes of
threat assessment workshops involving Galapagos ex-
perts to document the spatial and temporal scale of
recent changes in species’ distributions and important
benthic habitat types.
We also test the prediction that reef communities
exhibit changes along a spatial gradient of fishing
pressure. We use distance from port as a metric of
cumulative fishing pressure because fishers progres-
sively operate further and further from ports as stocks
are overfished (i.e., serial depletion). Effort near Gala-
pagos home ports has traditionally been greater than at
distance because relatively few fishers are prepared to
undertake multiday trips except for the lucrative sea
cucumber fishery (Born et al., 2003). Fishing in Galapa-
gos is primarily undertaken using small open fibreglass
and wooden boats, which are associated with larger
wooden motherships for travel on multiday trips to
outlying islands (Born et al., 2003).
In deciding on an appropriate index of fishing pres-
sure, we also considered direct measures of fishing
effort, as recorded in logbooks and by fishery observers
on board Galapagos fishing vessels (Danulat & Edgar,
2002). However, fishing effort is not independent of the
distribution of reef communities. Following tradeoffs
involving time and travel costs, fishers target sites
where they expect highest catch rates (i.e., sites with
highest perceived densities of fishery species). As a
consequence, no clear prediction could be made regard-
ing whether sites with high current fishing effort should
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2877
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
show positive or negative relationships with resource
species. Heavily fished sites could potentially possess
high densities of resource species because such sites
attract fishers, or low densities because fishing pressure
has depleted resource species at those sites.
We calculated Pearson’s correlation coefficients be-
tween distance from nearest fishing port and six ecolo-
gical metrics to test predictions of the hypothesis that
fishing has affected benthic community types. Predic-
tions tested were as follows:
(i) densities of large predatory fishes (sharks, serranid
groupers, carangid jacks, scombrid tunas and mack-
erel and lutjanid snappers) decline with fishing
effort (i.e., tend to be lowest off islands nearest to
ports),
(ii) densities of spiny lobsters (Panulirus penicillatus and
Panulirus gracilis) and slipper lobsters (Scyllarides
astori) increase with distance from fishing port,
(iii) catch per unit effort (CPUE) of spiny lobsters, as
recorded by observers in 2001 on board commercial
fishing vessels (Toral et al., 2002), increases with
distance from port,
(iv) densities of sea urchins (primarily Eucidaris galapa-
gensis,Tripneustes depressus,Lytechinus semitubercula-
tus,Echinometra vanbrunti and Diadema mexicanum)
decrease with distance from fishing port because of
increased levels of predation,
(v) coral cover increases with distance from fishing port
because of third-order trophic effects associated
with decreased predator numbers and increased
grazer numbers, and
(vi) cover of foliose macroalgae increases with distance
from fishing port because of third-order trophic
effects.
In addition to lobster numbers sighted on underwater
transects, we included CPUE of lobsters as a proxy for
lobster density because of the high degree of patchiness
in lobster distribution. Numbers recorded on transects
were greatly affected by stochastic sightings of groups
of individuals, particularly off islands with low lobster
densities. We nevertheless recognize that interpretation
of CPUE data also requires caution. The relationship
between CPUE and lobster density may not be linear
because commercial divers avoid areas with few lob-
sters, causing ‘hyperstability’ in catch rates (Harley
et al., 2001).
Methods
Data sources
All available historical and contemporary sources of
marine ecological data for Galapagos dating back to the
Allan Hancock and Templeton-Crocker (California
Academy of Sciences) expeditions of 1932 (Setchell,
1937; Setchell & Gardner, 1937; Taylor, 1945) were
searched for distribution and abundance data on threa-
tened species, corals, macroalgae, sea urchins, reef
fishes and lobsters. Limited information from earlier
expeditions was noted (1873 Hassler expedition, 1888,
1891 and 1904–1905 Albatross expeditions, 1898–1999
Hopkins Stanford expedition, 1905–1906 California
Academy of Science Expedition, 1925 Arcturus Expedi-
tion, and 1926 and 1928 Vanderbilt expeditions); how-
ever, these were excluded from temporal analyses
because few marine species were recorded on those
expeditions, and because a 50-year historical time slice
to 1982 was considered most relevant for comparisons
with recent patterns rather than applying an open-
ended historical time period.
Historical data sets that form the basis of threat
assessments presented here are summarized in Edgar
et al. (2008). Unpublished long-term data sets compiled
by the authors include (i) data on marine algae, inverte-
brates and fishes obtained by S. Earle during the 1966
Anton Bruun and 1972 Searcher expeditions (Mead et al.,
1972); (ii) results of an archipelago-wide study of inter-
tidal and subtidal fishes, invertebrates and plants in
1973–1974 (Wellington, 1975); (iii) coral monitoring data
obtained at intervals of 1–4 years at 30 sites established
in 1975 and 1976, plus 14 additional sites added in 1985
(Glynn & Wellington, 1983; Glynn, 1994, 2003); (iv)
observations and 42000 archived underwater photo-
graphs on fishes, invertebrates and macroalgae ob-
tained before, during and after the 1982/1983 El Nin
˜o
(Grove, 1985); (v) archipelago-wide fish observations
since 1987 (222 sites in total), including annual quanti-
tative transect data from 1994 to 2003 at 420 sites (F.
Rivera, unpublished results); and (vi) quantitative sur-
veys of fishes, mobile macroinvertebrates, sessile inver-
tebrates and floral communities at two depths at ca. 70
sites per year from 2000 to 2008 that were aimed at
assessing baseline conditions (Danulat & Edgar, 2002)
and population trends (S. Banks, unpublished results)
following establishment of the Galapagos Marine Re-
serve (GMR).
Additional data on threatened species and habitat
cover were obtained during two 10-day expeditions
undertaken between 26 November 2004 and 17 Decem-
ber 2004. A total of 56 sites were specifically searched
for 49 species considered potentially threatened, with
an average of five dive surveys per site. Sites were
selected to encompass the range of biogeographic and
environmental conditions across the archipelago, with
emphasis on locations with historical ecological data or
records of threatened taxa. In addition, while descend-
ing perpendicular to the shore, divers mapped the
2878 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
distribution of habitat types by recording the depth at
which habitat types changed from the low water mark
down to the reef edge or 30 m depth if reef habitat
extended passed that depth.
A total of 14 categories were recognized in the habitat
analysis – sand, open rock (including urchin barrens),
scleractinian (reef-building) corals, antipatharian (black)
corals, nonscleractinian corals, barnacles, anemones,
gorgonians, sponges, other sessile invertebrates, Sargas-
sum spp., other brown algae, red algae and green algae.
Along the depth profile, habitat types were scored on an
underwater slate using a semiquantitative habitat den-
sity scale – 0: absent; 1: single individual; 2: occasional
(0–2% cover, mode 1%); 3: common (2–25% cover, mode
10%); 4: abundant (25–75% cover, mode 50%); 5: com-
plete (75–100% cover, mode 100%). Mean percentage
cover of the different habitat types at different depths
was estimated by back-transforming semiquantitative
values using modes. Divers mapped an average of four
profiles at each site.
Assessment of threatened species and habitat change
following 1982/1983 El Nin
˜o
Threatened marine species in Galapagos were identi-
fied by, firstly, compiling a list of taxa recorded from
Galapagos that were included on the 2004 International
Union for the Conservation of Nature (IUCN) Red
List, then adding species belonging to groups not
previously assessed but which qualified as threate-
ned using IUCN Red List criteria (IUCN, 2001). The
latter were identified by (i) tabulating endemic Galapa-
gos marine species, (ii) gathering all available distribu-
tion and population trend data for these species, (iii)
undertaking a preliminary assessment to determine
whether these species fulfil IUCN Red List criteria as
globally threatened, (iv) instigating directed field
surveys at sites where species on this list had been
recorded historically or their presence suspected and (v)
reassessing the list of threatened species in the light of
new field data.
Scientists with expert local knowledge formally as-
sessed the threat status of tropical eastern Pacific corals
and endemic Galapagos macroalgae at IUCN Red List
Marine Threat Assessment Workshops in Galapagos
(28–31 May 2006). Endemic Galapagos corals and
macroalgae identified as threatened were added to the
2007 Red List following these workshops, and an addi-
tional four nonendemic corals added to the 2008 Red
List following additional workshops facilitated world-
wide by IUCN and the Global Marine Species Assess-
ment (Carpenter et al., 2008).
Changes in habitat types during the course of the
1982/1983 El Nin
˜o were quantified using digitized
information on habitats present in the background of
underwater images photographed across the GMR by J.
Grove. Of 42000 images available, a total of 239 were
considered suitable for analysis, comprising 71, 46 and
122 images photographed in the years 1982, 1983 and
1984, respectively. These images, which were photo-
graphed at sites distributed haphazardly across the
archipelago, included background areas of benthic ha-
bitat sufficiently sharp to be classified into 16 classes
(green foliose algae, red foliose algae, Sargassum spp.,
other foliose brown algae, crustose coralline algae, fine
mixed turf algae, sponges, barnacles, gorgonians, Pocil-
lopora spp., Pavona spp., other scleractinian corals, Tu-
bastraea spp., other nonscleractinian corals, black corals
and bare rock). The point intercept method based on a
grid with 50 points overlaid on each image was used to
estimate the percentage cover of different taxa and
substratum types. The image subject, generally a fish,
was ignored.
The significance of changes in mean density from
1982 to 1984 were assessed using one-way analysis of
variance (ANOVA), with significant outcomes further
analysed using a posteriori Tukey’s test (Zar, 1996) to
identify individual years of significant change. Habitat-
level changes were investigated using nonmetric multi-
dimensional scaling (nMDS), a graphical procedure that
provides the best depiction of faunal relationships
possible within a two-dimensional plot (Carr, 1996),
based on mean cover of major habitat groups in differ-
ent years. In order to allow comparison with data
obtained during threatened species surveys in 2004,
taxa were grouped into ten functional categories: green
foliose algae, red foliose algae, brown foliose algae,
open reef (crustose coralline algae, fine mixed turf algae
plus bare rock), sponges, barnacles, gorgonians, scler-
actinian corals, nonscleractinian corals and black corals.
The similarity matrix used in the nMDS, which de-
scribed pairwise patterns of faunal similarity between
years, was calculated using the Bray-Curtis index (Faith
et al., 1987).
The set of 239 underwater images was also used to
assess changes in density of the abundant urchins
Eucidaris galapagensis and Lytechinus semituberculatus.
Abundance of each urchin species was counted within
the largest rectangular area of reef substratum clearly
discernible within the image. An estimate was made by
eye of the area of the rectangle using sizes of organisms
present as references. To reduce bias, all image counts
were made by a single person, whose overall accuracy
in estimating distances was 97.5 15.7% (mean esti-
mated distance across image/measured distance SD)
using a set of 22 images with a calibrated survey tape
that was concealed on the image when the distance
estimate was made.
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2879
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
Assessment of impact of fishing
We tested the prediction that densities of predatory
fishes, spiny lobsters and urchins vary with distance
from fishing port using data obtained between 13 May
2000 and 13 December 2001 during GMR baseline
surveys (Danulat & Edgar, 2002; Edgar et al., 2004a, b).
Data comprised densities of fishes and mobile macro-
invertebrates sighted along 50 m transect lines set at two
depths (5–16 m) at 4250 sites during underwater visual
censuses. Two transect blocks (50 m 5 m for fishes;
50 m 1 m for macroinvertebrates) at two depths were
generally surveyed at each site. Overall, a total of 1158
and 1138 transect blocks were surveyed archipelago-
wide for fishes and macroinvertebrates, respectively.
Data on estimated coral and macroalgal cover were
obtained from 111 sites visited during threatened spe-
cies surveys undertaken in 2004 plus 118 sites visited in
2000–2001 during the GMR baseline surveys.
Survey data were aggregated as the mean of transects
for each of the 15 major islands to reduce the likelihood of
spatial autocorrelation confounding analyses, as occurs
when sites equally contribute data to probability tests but
with numerous sites concentrated in a particular location
(Legendre, 1993). Because of the great distance around
the largest island (Isabela) from the populated south to
the unpopulated north (ca. 120 km north–south), data
from this island were subdivided into independent north
and south components using a boundary at 0.681Slati-
tude. Distance to port for each island was calculated as
minimum sea distance from nearest port to sites sur-
veyed, with mean of distances to sites surveyed calcu-
lated for each island. The four recognized fishing ports in
Galapagos are Puerto Baquerizo Moreno (San Cristobal),
Puerto Ayora (Santa Cruz), Puerto Villamil (southern
Isabela) and Canal Itabaca (Santa Cruz).
Results
Threatened species and habitat change following 1982/
1983 El Nin
˜o
Initial analysis based on data to 2004 indicated a total of
49 Galapagos marine species as potentially globally
threatened. Nineteen of these species had been listed
on the 2004 IUCN Red List (the mammals, birds,
reptiles and two fishes), with the other thirty species
considered to fulfil IUCN-threatened species criteria on
preliminary screening.
Two expeditions undertaken in 2004 in search of
threatened species successfully located seven endemic
algae on the preliminary threatened species list that had
not been observed for over 30 years. None of these algae
(Eisenia galapagensis,Sporochnus rostratus,Chondria flex-
icaulis,Laurencia oppositoclada,Myriogramme kylinii,Pseu-
dolaingia hancockii and Schizymenia ecuadoreana) were
found in high densities. The other taxa on the prelimin-
ary list that were not seen since the 1982/1983 El Nin
˜o
were not encountered despite searches at their historic
locations.
The preliminary list of threatened species was mod-
ified following targeted searches. Macroalgal species
recorded at more than one site (Sporochnus rostratus
and Chondria flexicaulis) were removed from the list
under the assumption that rare species observed at
two or more sites were also likely to be present at
multiple additional sites not searched by divers, and
hence were best regarded as data deficient (sensu
IUCN, 2001) until population trend information was
available. Survey and historical data were presented at
expert threat assessment workshops for corals and
macroalgae, leading to the addition of three endemic
Galapagos corals and 15 macroalgae to the 2007 IUCN
Red List. A total of 45 Galapagos marine species are
currently recognized as threatened, comprising five
mammals, six birds, five reptiles, six fishes, one echi-
noderm, seven corals, six brown algae and nine red
algae (Table 1). Two of these species, the damselfish
Azurina eupalama and sunstar Heliaster solaris, remain to
be added to the Red List, but clearly fulfil criteria as
they have not been sighted during the past 30 years
despite searches.
The nMDS two-dimensional display of relationships
between habitat types present in the background of
underwater photographs indicated major ecosystem-
level changes between 1982 and 1983, with subsequent
relative stability of ecosystems in 1983 and 1984 (Fig. 1).
The stress value associated with this plot is extremely
low (Po0.01), indicating that the plot provides an
extremely good two-dimensional depiction of multi-
variate patterns (Clarke, 1993). Habitat data obtained
by divers in 2004 showed much greater similarity to
data digitized from underwater photographs in 1983
and 1984 than data from 1982, despite the very different
methods of assessment.
Among the major habitat classes examined, the cover
of fine mixed turf algae increased significantly from
1982 to 1983, the cover of foliose brown algae and
sponges declined precipitously over the same period,
the cover of barnacles increased greatly between 1983
and 1984, and the amount of bare rock was significantly
higher in 1983 than in either 1982 or 1984 (Fig. 2). Other
sessile taxa showed no significant trends when assessed
using one-way ANOVA. The density of the urchin Eu-
cidaris galapagensis increased significantly by a factor of
about two from 1982 to 1984 (Fig. 2).
Although significant changes in coral cover were not
detected for the 1982–1984 period (Fig. 2), long-term
2880 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
Table 1 Threatened Galapagos marine and shoreline species included on the 2008 IUCN Red List, plus two additional endemic species that, although not yet formally
evaluated, qualify for listing as they have not been sighted for 425 years and are possibly extinct
Threat Threat Threat
Critically endangered species Endangered species Vulnerable species
Camarhynchus heliobates Mangrove finch
*
XBalaenoptera musculus Blue whale F Arctocephalus galapagoensis Galapagos fur seal
*
E,F
Pterodroma phaeopygia Galapagos petrel
*
F,I Zalophus wollebaeki Galapagos sea lion
*
E,F Physeter macrocephalus Sperm whale F
Dermochelys coriacea Leatherback turtle F Spheniscus mendiculus Galapagos penguin
*
EMegaptera novaeangliae Humpback whale F
Eretmochelys imbricata Hawksbill turtle F Phalacrocorax harrisi Flightless cormorant
*
ELarus fuliginosus Lava gull
*
X
Azurina eupalama Black-spotted damselfishwEChelonia mydas Green turtle F,I Phoebastria irrorata Waved albatross
*
F
Heliaster solaris 24-rayed sun starwELepidochelys olivacea Olive ridley turtle F Amblyrhynchus cristatus Marine iguana
*
E,I
Rhizopsammia wellingtoni coral
*
ESargassum setifolium String sargassum
*
EF
$
Rhincodon typus Whale shark F
Tubastraea floreana coral
*
EMycteroperca olfax Bacalao grouper F
Bifurcaria galapagensis Galapagos stringweed
*
EF
$
Sphyrna mokarran Giant hammerhead F
Desmarestia tropica Tropical acidweed
*
EF
$
Hippocampus ingens East Pacific seahorse F
Dictyota galapagensis brown alga
*
EF
$
Taeniura meyeni Speckled stingray F
Spatoglossum schmittii brown alga
*
EF
$
Polycyathus isabelae coral
*
E
Phycodrina elegans red alga
*
EF
$
Pocillopora elegans coral E
Laurencia oppositoclada red alga
*
EF
$
Pocillopora inflata coral E
Myriogramme kylinii red alga
*
EF
$
Fungia curvata coral E
Schizymenia ecuadoreana red alga
*
EF
$
Psammocora stellata coral E
Gracilaria skottsbergii red alga
*
EF
$
Eisenia galapagensis Galapagos kelp
*
EF
$
Galaxaura barbata red alga
*
EF
$
Acrosorium papenfussii red alga
*
EF
$
Pseudolaingia hancockii red alga
*
EF
$
Austrofolium equatorianum red alga
*
EF
$
The most probable threats to species survival are also shown. (E, El Nin
˜o; F, overfishing; EF
$
, herbivore overgrazing associated with interactions between El Nin
˜o and overfishing;
I, predation by introduced species; X, unknown threatening process that probably includes predation by introduced species.)
*
Endemic and near endemic (499% of known global population) Galapagos marine species;
wEndemic species that qualify but have not yet been placed on the International Union for the Conservation of Nature (IUCN) Red List.
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2881
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
monitoring indicated a precipitous decline in reef corals
over the subsequent 3- year period (Fig. 3, Glynn, 1990,
1994). Little regeneration of surviving corals or estab-
lishment of new colonies 420 cm diameter have since
occurred (P. Glynn and G. Wellington, unpublished
results), with corals now covering significant areas
(i.e., patches of tens or hundreds of square metres) only
off the northern islands of Darwin and Wolf, and
Marchena in the central region (Table 2). Solitary corals
have also declined greatly in recent decades, including
the endemic taxon Madrepora oculata f. gamma, which
was recorded as abundant until the 1982/1983 El Nin
˜o
(Glynn & Wellington, 1983; Robinson & Del Pino, 1985),
but has not been seen in any surveys since.
Recent data also contrast with historical observations
extending back to 1932 and photographs (Fig. 4) that
indicate lower intertidal and shallow subtidal habitats
of the central and western archipelago were densely
covered by large fucoid algae before 1983, most con-
spicuously by Bifurcaria galapagensis and Sargassum spp.
(Houvenaghel & Houvenaghel, 1974; Wellington, 1975).
Foliose macroalgae are now highly localized in Galapa-
gos, with patches of Sargassum spp. virtually absent
outside the cool western upwelling region (Table 2).
Diverse red algal beds grew luxuriantly at many sites
in depths 415 m before the 1980s (Taylor, 1945; Mead
et al., 1972; Wellington, 1975). Large red algae are now
uncommon, while urchin ‘barrens’ composed of open
reef with encrusting coralline algae and cropped algal
turfs have predominated since at least 1987 (Kendrick,
1988) across most of the archipelago (Table 2).
The present distribution of threatened Galapagos
species coincides with loss of macroalgal habitat from
the central and southeastern region of the archipelago.
Species now recognized as threatened were recorded as
present off many more islands before 1983 than since
that year (Fig. 5), despite a 20-fold increase in search
effort in terms of diver hours between the periods pre-
and post-1983 (300 h cf. 6000 h underwater). Since
the 1982/1983 El Nin
˜o, macroinvertebrate and macro-
algal species now recognized as threatened have largely
disappeared from the central archipelago, with greatest
losses at the four islands with human habitation (Flor-
eana, Santa Cruz, San Cristobal and Isabela). Most
remnant populations of threatened species now persist
only in the relatively isolated western region of the
archipelago. Few threatened species reside in the north-
ern region.
Impact of fishing
Based on data from the 2000–2001 GMR baseline sur-
veys, densities of large predatory fishes and lobsters
were both significantly lower at islands located near
fishing ports than at more distant islands (Fig. 6). The
opposite relationship was found for sea urchins. Coral
cover was significantly positively correlated with dis-
tance of islands from port, while macroalgal cover was
not significantly affected, albeit trending in a positive
direction (Fig. 6).
The above relationships were strongly influenced by
data from the two most distant islands of Darwin and
Wolf (Fig. 6). Most notably, lobster densities were very
high at Darwin and Wolf while low and variable else-
where. If the islands of Darwin and Wolf are removed
from analyses, the power of statistical tests declines,
with only the correlation between predatory fishes and
distance among the plots shown in Fig. 6 remaining
significant for two-tailed tests (R50.59, 0.054P40.01).
A much stronger trend is indicated for macroalgal
density, albeit one that is only marginally significant if
a one-tailed test (i.e., for a positive correlation) is
applied (R50.44, 0.01oPo0.05).
Discussion
Threatened species and habitat change following 1982/
1983 El Nin
˜o
Of a total of 45 Galapagos marine species now recog-
nized as Vulnerable, Endangered or Critically Endan-
gered (Table 1), the majority were assessed as
threatened because of their susceptibility to El Nin
˜o
events, at least in part. Information provided during
threat assessment workshops and now included with
entries on the Red List (http://www.iucnredlist.org/)
indicates that the major threatening processes affecting
marine species are overfishing (15 species threatened),
El Nin
˜o (14 species threatened) and interactive effects
between these two factors (15 species threatened; Table 1).
Although climate change interacts with both El Nin
˜o
Fig. 1 Results of nonmetric multidimensional scaling showing
biotic relationships between communities of sessile benthic
organisms in different years.
2882 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
Fig. 2 Changes between years in mean cover ( SE) of bare rock and major sessile floral and functional groups, and mean density
(SE) of the urchins Eucidaris galapagensis and Lytechinus semituberculatus. Probability values associated with one-way analysis of
variance F-tests are also shown (density data log transformed), with significant differences between years indicated by differing
associated letters.
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2883
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
and fishing, and potentially represents the gravest
threat of all, this process was not considered during
assessments because of the magnitude of current un-
certainties, as has also been the case with most other
IUCN threat assessments (Akcakaya et al., 2006).
The probable extinction of the Galapagos damsel
Azurina eupalama is particularly notable given that the
IUCN Red List has yet to recognize a fully marine fish
as ‘Extinct’ (Edgar et al., 2005). This endemic plankti-
vorous species was formerly common in localized ag-
gregations on the islands of Floreana, Espan
˜ola, Isabela,
Marchena, Santiago, San Cristobal, Santa Cruz, and
Santa Fe, but declined precipitously in 1983 (Robinson
& Del Pino, 1985). It has not been sighted since, neither
in Galapagos nor Isla del Coco, the extra-limital location
for one museum record 600 km north of Galapagos
(Grove & Lavenberg, 1997) that is possibly misattributed.
The five threatened species most widely encountered
during recent surveys (Galapagos sea lion, green turtle,
marine iguana, bacalao grouper and Galapagos pen-
guin) are all higher vertebrates that have been compara-
tively well studied and were placed on the Red List
because of documented population declines. If popula-
tion trend data were available for fishes, invertebrates
and macroalgae, many more threatened species would
likely be recognized. Numerous cryptic reef-associated
species undoubtedly declined with the loss of coral
reefs, as did populations of inconspicuous algal epi-
phytes and grazing invertebrates associated with
Bifurcaria galapagensis and other macroalgal beds
Fig. 3 Sheltered embayment at Devils Crown, Floreana, showing well developed Pocillopora damicornis reef in 1982 with calcareous
framework extending for 0.6–0.8 m below photographed reef surface (a; photo P. Glynn); and remnants of coral framework and rubble
with numerous urchins (Eucidaris galapagensis) in 1986–1987 (b; photo P. Glynn; c; photo F. Rivera). The seabed at this site now consists of
sand and coral fragments (most o5 cm long) overlaying basalt.
Table 2 Estimated mean percentage ( SE) cover of habitat-forming reef corals, macroalgae and open reef (including bare rock,
fine algal turf and crustose coralline algae) for sites studied from 0 to 30 m depth in 2004 within the three major Galapagos marine
biogeographical zones (Fig. 5, Edgar et al., 2004a)
Zone Reef corals (%) Sargassum spp. (%) Other foliose algae (%) Open reef (%)
Northern 5.8 2.6 0 2.6 2.5 76.3 4.7
Central 2.3 0.6 0.02 0.01 3.3 1.1 82.7 2.3
Western 0.02 0.01 2.2 0.5 14.9 2.6 75.2 2.7
Macroalgal species recognized as habitat-forming are foliose species with fronds 45 cm length. Percentage cover data relate to sites
visited during threatened species surveys in 2004, where all sites with known concentrations of corals and macroalgae were visited,
hence mean estimates for corals and macroalgae are inflated compared with true archipelago-wide means.
2884 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
(Houvenaghel & Houvenaghel, 1974; Robinson & Del
Pino, 1985).
The majority of threatened plants and animals within
the region are endemic Galapagos taxa (Table 1). The
trend for endemic species to show disproportionately
large population declines compared with nonendemic
taxa is indicated by the 2008 IUCN Red List, which
currently lists as threatened nine of 10 endemic Gala-
pagos marine mammals, birds and reptiles, compared
with only five of the 33 nonendemic marine species
belonging to these higher vertebrate groups that regu-
larly frequent the archipelago.
Following the 1982/1983 El Nin
˜o, much of the Gala-
pagos region underwent major ecosystem changes that
have apparently persisted to date. These changes are
reflected in the nMDS plot (Fig. 1), where biotic patterns
in 2004 much more closely approach patterns at the end
of the 1982/1983 El Nin
˜o, and also during the 1984 La
Nin
˜a, than they do patterns at the start of the warming
event. The similarity of observed patterns in 1983 with
2004 was unexpected given the different techniques
used to assess cover of habitat classes in those 2 years,
emphasizing the magnitude of real differences between
either of those years and 1982.
Major habitat changes detected over the 1982–1984
period include substantial declines in cover of sponges
and foliose brown algae, particularly Sargassum, and an
increase in the cover of a layer of mixed turf algae
(intermingled diatoms, Hildenbrandia,Hincksia,Jania
and other algae forming a fine turf o10 mm tall) on
exposed reef surfaces (Fig. 2). Barnacle cover increased
substantially in 1984, presumably a consequence of the
strong La Nin
˜a in that year and the rapid population
responses shown by barnacles in periods with cool
upwelling conditions (Witman & Smith, 2003).
Although not found to be significant, the greatly in-
creased cover of foliose green algal species in 1984 was
also probably real given parallels with the massive
increase in cover of Ulva (including Enteromorpha) that
was documented in lower intertidal habitats during the
transition into La Nin
˜a conditions at the end of the
1997/1998 El Nin
˜o (Vinueza et al., 2006).
Little scientific attention or management concern
followed observations by Robinson (1985) during the
course of the 1982/1983 El Nin
˜o that Bifurcaria galapa-
gensis, Sargassum spp. and other canopy-forming algae
had completely disappeared from all 16 sites studied.
Any major reduction in Galapagos macroalgal habitat
nevertheless represents a significant loss of global bio-
diversity. This diverse habitat comprises a small tempe-
rate ecosystem with a very high level of endemism that
is isolated on the equator. Approximately 90 of 300
(33%) macroalgal taxa recorded from Galapagos have
not been recorded elsewhere.
A total of seven large and conspicuous macroalgal
species have not been sighted since the 1982/1983 El
Nin
˜o. Bifurcaria galapagensis was assessed by specialists
at the 2006 Galapagos threat assessment workshop as
Fig. 4 The rocky intertidal shore in front of the Charles Darwin Research Station in 1974 (left; photo G. Wellington) and 2003 (right;
photo F. Rivera). The dominant brown alga in the left photo is Bifurcaria galapagensis, an endemic species that once formed extensive
growths from the low intertidal to 6 m depth on moderately exposed coasts of southern and central islands (Wellington, 1975). Despite
extensive searches archipelago-wide, this alga has not been observed since dieback and catastrophic decline between January and March
1983 (Robinson & Del Pino, 1985). It is now probably extinct.
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2885
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
probably extinct, while Dictyota galapagensis,Spatoglos-
sum schmittii,Desmarestia tropica,Phycodrina elegans,
Gracilaria skottsbergii and Galaxaura barbata were re-
garded as possibly extinct. Many other poorly known
species were listed as data deficient, and may be threa-
tened as well.
Some macroalgal species in Galapagos are clearly at
great risk from extinction, a conclusion at odds with a
recent suggestion that deep water macroalgal commu-
nities in Galapagos are resilient to impacts of El Nin
˜o
(Graham et al., 2007). That suggestion was, however,
based on an estimated depression of about 20 m in the
Galapagos thermocline during typical El Nin
˜o events,
in which case most of the population of the kelp Eisenia
galapagensis and other deepwater algae could persist
within a refuge below the zone of El Nin
˜o nutrient
depletion. However, the 20m depression in the thermo-
cline referred to by Graham et al. (2007) is based on a
study of the mild 1991 El Nin
˜o (Kessler et al., 1995), and
is close to normal intra-annual variation (Wellington
et al., 2001). By contrast, during the extreme 1982/1983
El Nin
˜o, the thermocline declined by 50–70 m for many
months (Halpern et al., 1983), putting all known Gala-
pagos kelp populations at risk.
In contrast to changes in macroalgal communities,
changes in coral communities have been well documen-
ted in recent decades (see Feingold, 2001; Glynn, 2003).
The major observed cause of coral loss following the
1982/1983 El Nin
˜o was bioerosion by the urchin Eu-
cidaris galapagensis, which apparently increased in abun-
dance (Fig. 2) and destroyed reef frameworks weakened
by bleaching (Glynn et al., 1979, 2001; Glynn, 1990,
1994). Although the 1997/1998 El Nin
˜o generated a
thermal anomaly of similar magnitude to the 1982/
1983 event (Wellington et al., 2001), its ecological impact
on corals was less severe, probably because no large
coral reef frameworks remained in the archipelago in
1997, and because genotypes of remnant coral and
associated zooxanthellae were more resilient to heat
stress (Glynn et al., 2001).
In addition to its role in accelerating coral bioerosion,
the urchin Eucidaris galapagensis now appears to be
Fig. 5 Number of species on the current Galapagos threatened marine species list (Table 1) with historical records at different islands in
the three major biogeographic zones before (dark bars) and after (light bars) the 1982/1983 El Nin
˜o. Threatened marine vertebrates and
corals (Pocillopora elegans and P. inflatus) that are widely distributed across the archipelago have been excluded from totals. Conservation
and tourism zones with prohibition on fishing are shown as darkened areas of coast.
2886 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
present in sufficient numbers to prevent re-establish-
ment of coral and macroalgal habitat, thereby facilitat-
ing a regime shift in local benthic habitats. Studies
elsewhere indicate that urchin barrens typically devel-
op at sites with urchin densities exceeding about 3 m
2
and are maintained at densities 41m
2
(Tegner &
Dayton, 2000; Shears & Babcock, 2003). Densities of
urchins are now sufficiently high to preclude regrowth
of macroalgae and coral at most sites in Galapagos, with
Eucidaris galapagensis densities averaging 3.2 m
2
(Ed-
gar et al., 2004a), and total urchin densities averaging
45m
2
across the GMR (Fig. 6). Urchin densities
o1m
2
were consistently recorded only off the far-
northern islands of Darwin and Wolf.
Impact of fishing
One possible answer to the pivotal question ‘why did
the 1982/1983 El Nin
˜o generate such catastrophic en-
vironmental impacts compared with previous and sub-
sequent events?’ is that observed impacts represent
naturally recurrent perturbations that are regularly
associated with strong El Nin
˜o events, but with a cycle
extending longer than our period of study. We consider
this unlikely given that the 1982/1983 El Nin
˜o de-
stroyed all structural coral reefs in the archipelago,
including some that had persisted since at least AD
1600 (Dunbar et al., 1994). Moreover, Azurina eupalama
and other marine species that presumably frequented
Galapagos for tens of thousands of years have appar-
ently now disappeared.
An alternative explanation is that overfishing in
recent decades has reduced ecosystem resilience. Spe-
cifically, overfishing may have weakened predatory
control exerted by lobsters and fishes on populations
of urchins, resulting in greatly increased grazing pres-
sure and loss of macroalgae and corals (Sonnenholzner
et al., 2009). This hypothesis appears consistent with
observations and food web modelling (Bustamante
et al., 2008).
The hypothesis is also largely consistent with out-
comes of our analysis of relationships between ecosys-
tem metrics and distance from fishing port. As fishing
pressure increases, the density of predatory fishes and
lobsters significantly decreases, the density of sea urch-
ins increases and the density of coral decreases (Fig. 6).
Fig. 6 Scatterplots showing relationships between total densities of large predatory fishes, spiny lobsters, sea urchins, corals and
macroalgae and distance from fishing port. Data shown are means for islands where 415 field surveys were undertaken during 2000–
2001 GMR baseline surveys. Pearson’s correlation coefficients, with associated significance values based on two-tailed tests, are also
shown.
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2887
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
In the case of corals, the relationship between cover and
fishing pressure is likely affected by a lack of coral
growth in the cool upwelling western archipelago,
whereas the distribution of macroalgae is greatly en-
hanced by nutrient upwelling in the west. Thus, pat-
terns of density of corals and macroalgae probably
reflect both environmental and anthropogenic drivers,
the latter including fishing pressure and its correlate
analysed here.
An overfishing hypothesis that includes the premise
that predators, particularly lobsters, mediate commu-
nity interactions involving urchins in Galapagos is
supported by field studies across the eastern tropical
Pacific. Significantly lower densities of Eucidaris urchins
are present in marine-protected areas compared with
fished reference sites, both for the oceanic region that
encompasses Galapagos, Isla del Coco and Malpelo,
and for the continental coast extending from Panama
to Ecuador (G. Edgar, S. Banks, S. Bessudo, J. Corte
´s, H.
Guzman, F. Rivera, G. Soler, F. Zapata, unpublished
results). A keystone predator/urchin relationship in
Galapagos is also consistent with studies in South
Africa, New Zealand and United States, where lobsters
have been found to control invertebrate prey popula-
tions, particularly urchins (Tegner & Levin, 1983; Barkai
& Branch, 1988; Tegner & Dayton, 2000; Shears & Bab-
cock, 2002). In Galapagos, lobsters were directly ob-
served feeding on urchins at night, and high
abundances of lobster were found in our 2001 surveys
only around the northern islands of Darwin and Wolf
(Fig. 6), where urchins were few, and corals relatively
abundant. No clear relationship between distance from
port and lobster density is evident in Fig. 6 when
Darwin and Wolf are excluded from the plot, an indica-
tion that lobster overfishing is probably pervasive other
than off the northern islands.
A predator control hypothesis is additionally consis-
tent with the timeline of the development of the Gala-
pagos lobster and reef fish fisheries. The number of
registered Galapagos fishers increased from about 100
in 1971 to about 240 in 1985 and to about 1000 in 2001
(Bustamante et al., 2002). The semi-industrial lobster
fishery developed rapidly in the 1970s following the
introduction of compressed air surface supply diving
equipment to the islands (Bustamante et al., 2000b).
Predatory fish populations also have been over-
exploited, including Mycteroperca olfax, the most heavily
targeted reef species and biomass dominant amongst
the predatory fishes. This Vulnerable species of grouper
has been characterized as functionally extinct in the
central region of Galapagos (Ruttenberg, 2001; Okey
et al., 2004).
The next major El Nin
˜o will provide an unwelcome
test for a predator control hypothesis, given that the
Galapagos lobster fishery has recently expanded to the
northern islands, where spiny lobster numbers were an
order of magnitude higher than elsewhere in Galapagos
in 2001 (Fig. 6). We predict that unless new conservation
strategies are enacted to protect northern populations of
lobsters within sanctuary zones, coral habitat will di-
minish and urchin barrens will become more prevalent
in the north. Macroalgal beds are also predicted to
decline further unless exploitation of lobsters and pre-
datory fishes is controlled more effectively.
Results of this Galapagos study strengthen the grow-
ing contention that marine ecosystems are far from
isolated from threatening processes and the risk of
species extinction (Roberts & Hawkins, 1999; Dulvy
et al., 2003). Lack of distributional and population
information, ignorance about genetically distinct sibling
species, lack of IUCN Species Survival Commission
Specialist Groups to undertake formal Red List assess-
ments, few biodiversity inventories, and failure to re-
view scattered historical data, have presumably
contributed to the small number of marine species
worldwide currently recognized as extinct or threa-
tened. While major declines in terrestrial plant commu-
nities can be directly observed, the Galapagos situation
indicates that marine habitats can transform over large
areas virtually without notice or documentation. Such
oversights are illustrated by a ‘low’ threat ranking for
Galapagos in the global analysis of threats to coral reefs
(Bryant et al., 1998). A Global Marine Species Assess-
ment has now been initiated by IUCN to begin to
address these issues (http://www.sci.odu.edu/gmsa/).
For future studies, the GMR possesses a near ideal
environment for quantifying effects of oceanographic
anomalies and fisheries on marine biodiversity, and for
modelling future impacts of climate change. It also
possesses a natural resource management framework
well suited to the design of scientific studies, and one that
should generate positive conservation outcomes, with
subdivision of the coastal environment into interspersed
conservation, tourism and fishing zones. Unfortunately,
neither field monitoring data nor observations of the
distribution of fishing effort indicate that the ‘no-fishing’
conservation zones are well respected by local fishers
(Danulat & Edgar, 2002). If these zones were to be
effectively enforced, and new conservation zones added
to safeguard sites with populations of threatened species
(Edgar et al., 2008), then the GMR should provide invalu-
able insights into interactive human effects on marine
ecosystems associated with climate change.
Acknowledgements
We would particularly like to thank K. Carpenter, S. Livingstone,
C. Pollock and other staff associated with the Global Marine
2888 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
Species Assessment and IUCN Red List; P. C. Silva (algae) and S.
D. Cairns (ahermatypic corals) for taxonomic determinations; P.
Humann, J. Cortes and H. Guzman for information on coral
distribution; J. S. Feingold, J. M. Farin
˜a, M. Calvopin
˜a, S. A.
Shepherd, L. Kerrison, G. Toti, J. Delgado, M. V. Toral-Granda, R.
B. Mawbey, M. Sugden, L. Vinueza, C. M. Crawford, F. Smith
and V. Francesco for field assistance; T. M. Brooks, S. Baker, J.
Jackson and L. Kaufmann for comments on the draft manuscript;
R. Bensted-Smith, G. Merlen and E. Cruz for facilitating the
project; and J. Seigel and R. Feeney at the Natural History
Museum of Los Angeles County for access to images and
assistance with photo digitization. Threatened species cruises
were supported by the National Geographic Society, ecological
baseline surveys by USAID, Beneficia Foundation, the CSIRO
Climate Adaptation Flagship and Pew Charitable Trusts (to
RHB), Galapagos threat assessment workshops by the Gordon
and Betty Moore Foundation (Marine Management Area Science
project), and other aspects of the project by the Galapagos
National Park Service, Charles Darwin Foundation, Darwin
Initiative, Walton Family Foundation, Australian Research
Council, Tom Haas and the New Hampshire Charitable Founda-
tion and Conservation International. US National Science Foun-
dation grant OCE 0526361 and earlier grants supported the coral
studies of P. W. Glynn.
References
Akcakaya HR, Butchart SHM, Mace GM, Stuart SN, Hilton-Taylor C
(2006) Use and misuse of the IUCN Red List Criteria in projecting
climate change impacts on biodiversity. Global Change Biology,12,
2037–2043.
Barber RT, Chavez FP (1986) Ocean variability in relation to living
resources during the 1982–83 El Nin
˜o. Nature,319, 279–285.
Barkai A, Branch GM (1988) The influence of predation and substratal
complexity on recruitment to settlement plates: a test of the theory of
alternate states. Journal of Experimental Marine Biology and Ecology,124,
215–237.
Bensted-Smith R (2002) A Biodiversity Vision for the Galapagos Islands.
Charles Darwin Foundation and World Wildlife Fund, Puerto Ayora,
Galapagos, Ecuador.
Boer GJ, Yu B, Kim S-J, Flato GM (2004) Is there observational support for
an El Nin
˜o-like pattern of future global warming? Geophysical Research
Letters,31, L06201, doi: 10.1029/2003GL018722.
Born AF, Espinoza E, Murillo JC, Nicolaides F, Edgar GJ (2003) Effects of
the Jessica oil spill on artisanal fisheries in the Galapagos. Marine
Pollution Bulletin,47, 319–324.
Bryant D, Burke L, McManus J, Spalding M (1998) Reefs at Risk: A Map-
Based Indicator of Potential Threats to the World’s Coral Reefs. World
Resources Institute, Washington, DC.
Bustamante RH, Branch GM, Bensted-Smith R, Edgar GJ (2002) The status
of and threats to marine biodiversity. In: A Biodiversity Vision for the
Galapagos Islands (ed. Darwin Foundation and World Wildlife Fund),
pp. 80–95. CDF, Puerto Ayora, Galapagos.
Bustamante RH, Collins KJ, Bensted-Smith R (2000a) Biodiversity con-
servation in the Gala
´pagos Marine Reserve. Bulletin de l’Institut Royal
des Sciences Naturelles de Belgique, (Suppl. 70), 31–38.
Bustamante RH, Okey TA, Banks S (2008) Chapter 9. Biodiversity and
food-web structure of a Gala
´pagos shallow rocky reef ecosystem. In:
Food Webs and the Dynamics of Marine Reefs (eds McClanahan TR, Branch
GM), pp. 1–23. Oxford University Press, New York.
Bustamante RH, Reck GK, Ruttenberg BI, Polovina J (2000b) The Gala
´-
pagos spiny lobster fishery. In: Spiny Lobsters: Fisheries and Culture (eds
Phillips BF, Kittaka J), pp. 210–220. Fishing News Books, Blackwell
Science, Oxford.
Carpenter KE, Abrar M, Abey G et al. (2008) One third of reef-building
corals face elevated extinction risk from climate change and local
impacts. Science,321, 560–563.
Carr MR (1996) PRIMER User Manual. Plymouth Routines in Multivariate
Ecological Research. Plymouth Marine Laboratory, Plymouth, UK.
Clarke KR (1993) Non-parametric multivariate analyses of changes in
community structure. Australian Journal of Ecology,18, 117–143.
Conroy JL, Restrepo A, Overpeck JT, Steinitz-Kannan M, Cole JE, Bush MB,
Colinvaux PA (2009) Unprecedented recent warming of surface tem-
peratures in the eastern tropical Pacific Ocean. Nature Geoscience,2,46–
50. Available at http://www.nature.com/ngeo/journal/v2/n1/abs/
ngeo390.html
Crain CM, Kroeker K, Halpern BS (2008) Interactive and cumulative
effects of multiple human stressors in marine systems. Ecology Letters,
11, 1304–1315.
Danulat E, Edgar GJ (2002) Reserva Marina de Gala
´pagos, Lı
´nea Base de la
Biodiversidad. Charles Darwin Foundation and Gala
´pagos National
Park Service, Gala
´pagos, Ecuador.
Dulvy NK, Sadovy Y, Reynolds JD (2003) Extinction vulnerability in
marine populations. Fish and Fisheries,4, 25–64.
Dunbar RB, Wellington GM, Colgan MW, Glynn PW (1994) Eastern Pacific
sea surface temperature since 1600 A.D.: the delta18O record of climate
variability in Galapagos corals. Paleoceanography,9, 291–315.
Edgar GJ, Banks S, Bensted-Smith R et al. (2008) Conservation of threa-
tened species in the Galapagos Marine Reserve through identification
and protection of marine Key Biodiversity Areas. Aquatic Conservation:
Marine and Freshwater Ecosystems,18, 955–968.
Edgar GJ, Banks S, Farin
˜a JM, Calvopin
˜a M, Martı
´nez C (2004a) Regional
biogeography of shallow reef fish and macro-invertebrate communities
in the Galapagos Archipelago. Journal of Biogeography,31, 1107–1124.
Edgar GJ, Bustamante RH, Farin
˜a J-M, Calvopin
˜a M, Martinez C, Toral-
Granda MV (2004b) Bias in evaluating the effects of marine protected
areas: the importance of baseline data for the Galapagos Marine
Reserve. Environmental Conservation,31, 212–218.
Edgar GJ, Samson CR, Barrett NS (2005) Species extinction in the marine
environment: Tasmania as a regional example of overlooked losses in
biodiversity. Conservation Biology,19, 1294–1300.
Enfield DB (2001) Evolution and historical perspective of the 1997–1998 El
Nin
˜o-Southern Oscillation event. Bulletin of Marine Science,69, 7–25.
Faith DP, Minchin PR, Belbin L (1987) Compositional dissimilarity as a
robust measure of ecological distance. Vegetatio,69, 57–68.
Feingold JS (2001) Responses of three coral communities to the 1997–98 El
Nin
˜o-Southern Oscillation: Galapagos Islands, Ecuador. Bulletin of
Marine Science,69, 61–77.
Glynn PW (1990) Coral mortality and disturbances to coral reefs in the
tropical Eastern Pacific. In: Global Ecological Consequences of the 1982–83
El-Nin
˜o-Southern Oscillation (ed. Glynn PW), pp. 55–126. Elsevier,
Amsterdam.
Glynn PW (1994) State of coral reefs in the Galapagos Islands: natural vs
anthropogenic impacts. Marine Pollution Bulletin,29, 131–140.
Glynn PW (2003) Coral communities and coral reefs of Ecuador. In: Latin
American Coral Reefs (ed. Corte
´s J), pp. 449–472. Elsevier Science,
Amsterdam.
Glynn PW, Ault JS (2000) A biogeographic analysis and review of the far
eastern Pacific coral reef region. Coral Reefs,19, 1–23.
Glynn PW, Mate JL, Baker AC, Calderon MO (2001) Coral bleaching and
mortality in Panama and Ecuador during the 1997–1998 El Nin
˜o-
Southern Oscillation event: spatial/temporal patterns and comparisons
with the 1982–1983 event. Bulletin of Marine Science,69, 79–109.
Glynn PW, Wellington GM (1983) Corals and Coral Reefs of the Gala
´pagos
Islands. With an Annotated List of the Scleractinian Corals of the Gala
´pagos.
University of California Press, Berkeley.
Glynn PN, Wellington GM, Birkeland C (1979) Coral reef growth in the
Galapagos: limitation by sea urchins. Science,203, 47–49.
EL NIN
˜O, GRAZERS AND FISHERIES INTERACT WITH GALAPAGOS MARINE SPECIES 2889
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890
Graham MH, Kinlan BP, Druehl LD, Garske LE, Banks S (2007) Deep-
water kelp refugia as potential hotspots of tropical marine diversity
and productivity. Proceedings of the National Academy of Science, USA,
104, 16576–16580.
Grove JS (1985) Influence of the 1982–1983 El Nin
˜o event upon the
ichthyofauna of the Gala
´pagos archipelago. In: El Nin
˜o in the Gala
´pagos
Islands: the 1982–1983 Event (eds. Robinson G, Del Pino EM), pp. 191–
198. Charles Darwin Foundation for the Gala
´pagos Islands, Quito,
Ecuador.
Grove JS, Lavenberg RJ (1997) The Fishes of the Gala
´pagos Islands. Stanford
University Press, Stanford, USA.
Halpern D, Hayes SP, Leetmaa A, Hansen DV, Philander SGH (1983)
Oceanographic observations of the 1982 warming of the Tropical East-
ern Pacific. Science,221, 1173–1175.
Harley SJ, Myers RA, Dunn A (2001) Is catch-per-unit-effort proportional
to abundance? Canadian Journal of Fisheries and Aquatic Sciences,58,
1760–1772.
Harley DG, Rogers-Bennett L (2004) Effects of climate change and fishing
pressure on exploited invertebrates. CalCOFI Report,45, 98–110.
Helmuth B, Broitman BR, Blanchette CA et al. (2006) Mosaic patterns of
thermal stress in the rocky intertidal zone: implications for climate
change. Ecological Monographs,76, 461–479.
Hickman CP (2009) Evolutionary responses of marine invertebrates to
insular isolation in Galapagos. Galapagos Research,66, 32–42.
Houvenaghel GT, Houvenaghel N (1974) Aspects e
´cologiques de la
zonation intertidale sur les cotes rocheuses des iles Galapagos. Marine
Biology,26, 135–152.
Hughes TP, Baird AH, Bellwood DR et al. (2003) Climate change, human
impacts, and the resilience of coral reefs. Science,301, 929–933.
Hughes TP, Bellwood DR, Folke C, Steneck RS, Wilson J (2005) New
paradigms for supporting the resilience of marine ecosystems. Trends in
Ecology and Evolution,20, 380–386.
IUCN (2001) IUCN Red List Categories and Criteria, Version 3.1. Species
Survival Commission, IUCN, Gland, Switzerland.
Kendrick G (1988) Unpublished Trip Report: Preliminary Survey of the Marine
Benthos of the Western Parts of Isla Isabela and Isla Fernandina. Unpub-
lished report to Charles Darwin Foundation, Gala
´pagos, Ecuador.
Kessler WS, McPhaden MJ, Weickmann KM (1995) Forcing of intraseaso-
nal Kelvin waves in the equatorial Pacific. Journal of Geophysical
Research,100, 10613–10631.
Legendre P (1993) Spatial autocorrelation: trouble or new paradigm?
Ecology,74, 1659–1673.
Mead GW, Silva PC, Earle SA et al. (1972) Report of the R/V Searcher Cruise
72-1 Gala
´pagos Islands and Cocos Island. January–February 1972. Unpub-
lished report to Charles Darwin Foundation, Gal a
´pagos, Ecuador.
Occhipinti-Ambrogi A (2007) Global change and marine communities:
alien species and climate change. Marine Pollution Bulletin,55, 342–352.
Okey TA, Banks SJ, Born AF et al. (2004) A trophic model of a Gala
´pagos
subtidal rocky reef for evaluating fisheries and conservation strategies.
Ecological Modelling,172, 383–401.
Pandolfi JM, Jackson JBC, Baron N et al. (2005) Are U.S. coral reefs on the
slipperyslope to slime? Science,307, 1725–1726.
Poertner HO, Knust R (2007) Climate change affects marine fishes
through the oxygen limitation of thermal tolerance. Science,315, 95–97.
Reaser JK, Pomerance R, Thomas PO (2000) Coral bleaching and global
climate change: scientific findings and policy recommendations. Con-
servation Biology,14, 1500–1511.
Roberts CM, Hawkins JP (1999) Extinction risk in the sea. Trends in Ecology
and Evolution,14, 241–246.
Robinson G (1985) Influence of the 1982–83 El Nin
˜o on Gala
´pagos marine
life. In: El Nin
˜o in the Gala
´pagos Islands: the 1982–1983 Event (eds
Robinson G, Del Pino EM), pp. 153–190. Charles Darwin Foundation
for the Gala
´pagos Islands, Quito, Ecuador.
Robinson G, Del Pino EM (1985) El Nin
˜o in the Gala
´pagos Islands: the 1982–
1983 Event. Charles Darwin Foundation for the Gala
´pagos Islands,
Quito, Ecuador.
Ruttenberg BI (2001) Effects of artisanal fishing on marine communities in
the Galapagos Islands. Conservation Biology,15, 1691–1699.
Setchell WA (1937) The Templeton Crocker expedition of the California
Academy of Sciences, 1932. No. 31. A preliminary report on the algae.
Proceedings of the California Academy of Science,22, 65–98.
Setchell WA, Gardner NL (1937) The Templeton Crocker expedition of the
California Academy of Sciences, 1932. No. 34. Report on the Sargas-
sums. Proceedings of the California Academy of Sciences,22, 127–158.
Shears NI, Babcock RI (2002) Marine reserves demonstrate top-down
control of community structure on temperate reefs. Oecologia,132,
131–142.
Shears NT, Babcock RC (2003) Continuing trophic cascade effects after 25
years of no-take marine reserve protection. Marine Ecology Progress
Series,246, 1–16.
Sonnenholzner JI, Ladah LB, Lafferty KD (2009) Cascading effects of
fishing on Galapagos rocky reef communities: reanalysis using cor-
rected data. Marine Ecology Progress Series,375, 209–218.
Taylor WR (1945) Pacific marine algae of the Allan Hancock Expeditions
to the Gala
´pagos Islands. Allan Hancock Pacific Expeditions,12, 1–528.
Tegner MJ, Dayton PK (2000) Ecosystem effects of fishing in kelp forest
communities. ICES Journal of Marine Science,57, 579–589.
Tegner MJ, Levin LA (1983) Spiny lobsters and sea urchins: analysis of a
predator-prey interaction. Journal of Experimental Marine Biology and
Ecology,73, 125–150.
Thomas CD, Cameron A, Green RE et al. (2004) Extinction risk from
climate change. Nature,427, 145–148.
Thomas CD, Franco AMA, Hill JK (2006) Range retractions and extinc-
tion in the face of climate warming. Trends in Ecology and Evolution,21,
415–416.
Toral MV, Espinoza E, Hearn A, Martinez C (2002) Langostas espinosas.
In: Reserva Marina de Gala
´pagos, Lı
´nea Base de la Biodiversidad (eds
Danulat E, Edgar GJ), pp. 199–221. Charles Darwin Foundation and
Gala
´pagos National Park Service, Gal a
´pagos, Ecuador.
Trillmich F, Limberger D (1985) Drastic effects of El Nin
˜o on Gala
´pagos
pinnipeds. Oecologia,67, 19–22.
Vinueza LR, Branch GM, Branch ML, Bustamante RH (2006) Top-down
herbivory and bottom-up El Nin
˜o effects on Galapagos rocky-shore
communities. Ecological Monographs,76, 111–131.
Wellington GM (1975) The Gala
´pagos coastal marine environment. Resource
report, Department of National Parks and Wildlife, Quito, Ecuador.
Wellington GM, Strong AE, Merlen G (2001) Sea surface temperature
variation in the Galapagos Archipelago: a comparison between
AVHRR nighttime satellite data and in situ instrumentation (1982–
1998). Bulletin of Marine Science,69, 27–42.
Witman JD, Smith F (2003) Rapid community change at a tropical
upwelling site in the Galapagos Marine Reserve. Biodiversity and
Conservation,12, 25–45.
Zar JH (1996) Biostatistical Analysis. Prentice-Hall, Englewood Cliffs,
NJ.
2890 G. J. EDGAR et al.
r2009 Blackwell Publishing Ltd, Global Change Biology,16, 2876–2890