Kelp forests are phyletically diverse, structurally
complex and highly productive components of cold-
water rocky marine coastlines. This paper reviews the
conditions in which kelp forests develop globally and
where, why and at what rate they become deforested.
The ecology and long archaeological history of kelp
forests are examined through case studies from
southern California, the Aleutian Islands and the
western North Atlantic, well-studied locations that
represent the widest possible range in kelp forest biodi-
versity. Global distribution of kelp forests is
physiologically constrained by light at high latitudes
and by nutrients, warm temperatures and other
macrophytes at low latitudes. Within mid-latitude
belts (roughly 40–60° latitude in both hemispheres)
well-developed kelp forests are most threatened by
herbivory, usually from sea urchins. Overﬁshing and
extirpation of highly valued vertebrate apex predators
often triggered herbivore population increases, leading
to widespread kelp deforestation. Such deforestations
have the most profound and lasting impacts on
species-depauperate systems, such as those in Alaska
and the western North Atlantic. Globally urchin-
induced deforestation has been increasing over the
past 2–3 decades. Continued ﬁshing down of coastal
food webs has resulted in shifting harvesting targets
from apex predators to their invertebrate prey,
including kelp-grazing herbivores. The recent global
expansion of sea urchin harvesting has led to the wide-
spread extirpation of this herbivore, and kelp forests
have returned in some locations but, for the ﬁrst time,
these forests are devoid of vertebrate apex predators.
In the western North Atlantic, large predatory crabs
have recently ﬁlled this void and they have become the
new apex predator in this system. Similar shifts from
ﬁsh- to crab-dominance may have occurred in coastal
zones of the United Kingdom and Japan, where large
predatory ﬁnﬁsh were extirpated long ago. Three
North American case studies of kelp forests were
examined to determine their long history with humans
and project the status of future kelp forests to the year
2025. Fishing impacts on kelp forest systems have been
both profound and much longer in duration than
previously thought. Archaeological data suggest that
coastal peoples exploited kelp forest organisms for
thousands of years, occasionally resulting in localized
losses of apex predators, outbreaks of sea urchin popu-
lations and probably small-scale deforestation. Over
the past two centuries, commercial exploitation for
export led to the extirpation of sea urchin predators,
such as the sea otter in the North Paciﬁc and predatory
ﬁshes like the cod in the North Atlantic. The large-
scale removal of predators for export markets
increased sea urchin abundances and promoted the
decline of kelp forests over vast areas. Despite
southern California having one of the longest known
associations with coastal kelp forests, widespread
deforestation is rare. It is possible that functional
redundancies among predators and herbivores make
this most diverse system most stable. Such biodiverse
kelp forests may also resist invasion from non-native
species. In the species-depauperate western North
Atlantic, introduced algal competitors carpet the
benthos and threaten future kelp dominance. There,
other non-native herbivores and predators have
become established and dominant components of this
system. Climate changes have had measurable
impacts on kelp forest ecosystems and efforts to
control the emission of greenhouse gasses should be a
global priority. However, overﬁshing appears to be the
greatest manageable threat to kelp forest ecosystems
over the 2025 time horizon. Management should focus
on minimizing ﬁshing impacts and restoring popu-
lations of functionally important species in these
Keywords: apex predators, biodiversity, herbivory, human
interactions, kelp forests, trophic cascades
Kelp forest ecosystems: biodiversity, stability, resilience and future
ROBERT S. STENECK1*, MICHAEL H. GRAHAM2, BRUCE J. BOURQUE3, DEBBIE
CORBETT4, JON M. ERLANDSON5, JAMES A. ESTES6 AND MIA J. TEGNER7†
1School of Marine Sciences, University of Maine, Darling Marine Center, Walpole, ME 04573, USA, 2Center for Population Biology, University
of California, Davis, One Shields Avenue, Davis, CA 95616, USA, 3Department of Anthropology, Bates College, Lewiston, ME 04240, USA,
4US Fish and Wildlife Service, 1011 East Tudor Road, Anchorage, AK 99503–6119, USA, 5Department of Anthropology, University of Oregon,
308 Condon Hall, Eugene, OR 97403–1218, USA, 6US Geological Survey, Long Marine Laboratory, 100 Shaffer Road, University of
California, Santa Cruz, CA 95060, USA and 7Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA
Date submitted: 7 November 2001 Date accepted: 25 September 2002
*Correspondence: Professor Robert Steneck Tel: 1 207 563 3146
ext. 233 Fax: 1 207 563 3119 e-mail: firstname.lastname@example.org
Environmental Conservation 29 (4): 436–459 © 2002 Foundation for Environmental Conservation DOI:10.1017/S0376892902000322
Kelp forest ecosystems 437
‘I can only compare these great aquatic forests . . . with
the terrestrial ones in the intertropical regions. Yet if in
any country a forest was destroyed, I do not believe
nearly so many species of animals would perish as would
here, from the destruction of the kelp. Amidst the leaves
of this plant numerous species of ﬁsh live, which
nowhere else could ﬁnd food or shelter; with their
destruction the many cormorants and other ﬁshing
birds, the otters, seals, and porpoise, would soon perish
also; and lastly, the Fuegian[s] . . . would . . . decrease in
numbers and perhaps cease to exist.’
Charles Darwin, 1 June 1834, Tierra del Fuego, Chile
(Darwin 1909, pp. 256–257).
Kelp forests dominate shallow rocky coasts of the world’s
cold-water marine habitats. They comprise primarily brown
algae in the order Laminariales and produce the largest
biogenic structures found in benthic marine systems (Dayton
1985a). Kelp forest ecosystems include structure-producing
kelps and their myriad associated biota such as marine
mammals, ﬁshes, crabs, sea urchins, molluscs, other algae
and epibiota that collectively make this one of the most
diverse and productive ecosystems of the world (Mann 1973).
Economically, kelp forest ecosystems have been signiﬁcant to
maritime peoples for thousands of years (Simenstad et al.
1978; Erlandson 2001).
Past reviews provided contemporary overviews of kelp
forests (Kain 1979; Dayton 1985a; Schiel & Foster 1986;
Witman & Dayton 2001). These rarely considered how kelp
forest ecosystems have changed at very large spatial scales
over very long periods of time spanning decades to millennia
( Jackson et al. 2001). A longer historical perspective allows us
to see how kelp forest ecosystems have changed and possibly
where they are headed in the future. Thus, it is with larger
spatial and temporal perspective that we embarked on this
The nature of kelp
Three morphological groups or ‘guilds’ of kelp are deﬁned by
the canopy height of their fronds (Dayton 1985a). ‘Canopy’
kelps are largest and produce ﬂoating canopies. Chief among
these is the giant kelp, Macrocystis spp. which grows to 45 m
long (Abbott & Hollenberg 1976) and dominates kelp forests
along the west coasts of North and South America and at
scattered locations in the South Paciﬁc Ocean including
South Africa, southern Australia, New Zealand and several
subantarctic islands (Fig. 1). Smaller canopy kelps include
Nereocystis leutkeana, which ranges from Central California
to Alaska, and its Southern Hemisphere counterpart,
Ecklonia maxima in South Africa, and Alaria ﬁstulosa in
Alaska and the Paciﬁc coast of Asia. These kelps reach about
10 m in length. ‘Stipitate’ kelp fronds are held above the
benthos with rigid stipes (Dayton 1985a). They include some
species of Laminaria sp. in Europe and the Paciﬁc North-
west, Ecklonia in southern Australia and New Zealand, and
Lessonia in Chile. Stipitate Laminaria dominates kelp forests
of the North Paciﬁc from Japan, north-east across coastal
Alaska to northern California. Most species of Laminaria are
less than 5 m long, but a few grow to 10 m in length. Other
stipitate genera found along the Paciﬁc coast of North
America include Pterygophora, Eisenia, Pleurophycus and
Figure 1 Kelp forest
distributions of the world and
their dominant genera (from
Raffaelli & Hawkins 1996).
Thalassiophyllum. ‘Prostrate’ kelps are most diminutive and
cover the benthos with their fronds. This guild includes
several species of Laminaria, which dominate most of the
Northern Hemisphere except for parts of the eastern North
Paciﬁc (Fig. 1). Prostrate Laminaria forests range from the
Gulf of Maine to Greenland in the western North Atlantic
and from Iceland to the high Arctic of Norway and south to
the north-westernmost corner of Africa, in the eastern North
Atlantic. Taxonomically, the kelps themselves are not
diverse. The most diverse kelp ﬂora occurs along the
California coast of North America. However, it only has 20
species of kelp distributed among 16 genera because most
genera are monotypic (Abbott & Hollenberg 1976).
Despite their low taxonomic diversity, kelps are highly
diverse structurally and functionally. They possess special-
ized cells for translocation (for example, ‘trumpet hyphae’
and sieve elements) that allow them to attain their great size
and diverse morphology. Even within species, developmental
stages and ploidy-phases of kelp span an unparalleled range
of sizes and shapes. Their gametophyte stage is an inconspic-
uous, microscopic ﬁlament living within the benthic
boundary layer (Neushul 1972). In this microhabitat fertiliz-
ation occurs and developing sporophytes grow. Depending
on the adult morphology of the species, the kelp fronds may
remain near the benthos (prostrate forms), occupy interme-
diate depths (stipitate forms), or grow to and ﬂoat on the
surface (canopy forms) (Neushul 1972). Morphology varies
widely among and within genera. The tiering of kelp
morphologies along with other associated taxa contributes to
the structural diversity of this system. All three kelp forms
can coexist (Dayton 1985a) together with a patchy carpet of
corticated macrophyte turf (Dayton 1975) and a pavement of
encrusting coralline algae. Each of these structural compo-
nents is habitat and food for associated organisms. Thus the
unique anatomy of kelp allows communities to attain the
stature of forests with wide-ranging implications for associ-
ated organisms and coastal communities.
The structure and function of kelp forests differ signiﬁ-
cantly from their terrestrial counterparts dominated by
angiosperm and gymnosperm trees. Compared to terrestrial
forests, kelp forests are more productive and diverse (at the
phyla level), but the average lifespan and structural height is
less. For example, many mature terrestrial forests attain
canopy heights of 10–30 m within 20 to 30 years. Individual
trees can have longevities of centuries to millennia.
Associated with terrestrial forests are animals primarily from
the three phyla Chordata, Arthropoda and Annelida. In
contrast, kelp forests reach canopy heights of 1–15 m
(although Macrocystis is larger than this) within 1–3 years.
Individual kelps have maximum longevities of about 25 years
(Steneck & Dethier 1994). Associated with kelp forests are
animals from more than 10 phyla, namely Chordata,
Arthropoda, Annelida, Echinodermata, Bryozoa, Cnidaria,
Mollusca, Platyhelminthes, Brachiopoda and Porifera.
The physical structure, algal biomass and organisms
associated with kelp forests profoundly alter local environ-
ments and ecologies. Kelp canopies dampen waves, which
inﬂuence water ﬂow and associated processes of coastal
erosion, sedimentation, benthic productivity (primary and
secondary) and recruitment (Duggins et al. 1990). The
canopies also reduce light, creating understorey conditions
favourable for a suite of species adapted to low light intensity
(Santelices & Ojeda 1984a); as a result, they can inﬂuence
interspeciﬁc competition among algae (Dayton 1985a). Kelps
are substratum for numerous sessile animals and algae
(Duggins 1980; Reed & Foster 1984; Dunton & Shell 1987)
and habitat for mobile organisms specialized to live and feed
directly on the kelp or its associated assemblages. For
example, trophically specialized limpets depend upon kelp
for their existence (Steneck & Watling 1982; Estes &
Steinberg 1988; Bustamante et al. 1995). Kelp forest archi-
tecture provides habitat, nursery ground and food for myriad
mobile pelagic and benthic organisms (Bernstein & Jung
1980; Bologna & Steneck 1993; Levin 1994; Anderson et al.
1997). Since predatory ﬁshes use canopies as habitat, canopy
loss can translate to increased survivorship of resident prey
organisms and their larvae (Gaines & Roughgarden 1987).
Thus, as kelp canopies vary, the ecological and oceanographic
processes associated with them will be altered.
Kelps concentrate biomass and are a signiﬁcant source of
nutrition for coastal marine ecosystems via food webs based
on macroalgal detritus (Duggins et al. 1989) because herbi-
vores rarely consume more than 10% of the living biomass
(Mann 2000). Large pieces of kelp commonly litter the
benthos in coastal zones and become food for detritivores and
microbes (Linley et al. 1981), thus making their carbon avail-
able to the coastal community of suspension feeders as well as
the herbivores and detritivores feeding directly on kelps
(Dunton & Schell 1987; Duggins et al. 1989). In effect, kelp
forests concentrate and magnify secondary production,
thereby supporting complex food webs in coastal zones
(Duggins et al. 1989; Mann 2000).
The combination of high kelp productivity in environ-
ments with the potential for high rates of disturbance can
make these forests surprisingly ephemeral. Entire kelp beds
can be eliminated by thermal events, storms or outbreaks of
herbivores and disappear without a trace within a year, but
signiﬁcantly, the entire community can return nearly as
quickly (Scheibling 1984; Harrold & Reed 1985; Hart &
Scheibling 1988; Witman 1988; Tegner et al. 1997).
As with most ecosystems, kelp forests are strongly inﬂu-
enced by both their physical and biological environment.
Fortunately, we know much more about how environmental
forcing functions operate in kelp forests than we do for most
other ecosystems. Kelp forests are ideally located and scaled
for human study; most components of the system are readily
observable, kelp forest phase-changes are rapid (i.e. occur-
ring within a few years) and manipulative experiments of
canopy removal or transplantation are easily accomplished.
Thus, hypotheses can be advanced, tested, reported and chal-
lenged within a decade.
438 R.S. Steneck et al.
LARGE SCALE PATTERNS: GLOBAL INTERPLAY
OF DIVERSITY, PRODUCTIVITY AND
The ecology of kelp forests
Kelp forests persist in a balance between ecological processes
driving their development and their deforestation. They wax
as a result of recruitment and net productivity and wane as a
result of biomass lost from disturbances both physical and
biological and competition. These factors are inﬂuenced by
properties intrinsic to, and extrinsic of, the kelp (Steneck &
Dethier 1994). Complex, multifactorial interactions make
long-term changes in these systems difﬁcult to demonstrate
and predict (Tegner et al. 1996a). Further, the population
density and community development in structurally diverse
(high canopy) kelp forests may decouple changes in the
canopy cover from changes in population densities, thus
obscuring linkages between patterns and processes (Schiel &
Foster 1986; Dayton et al. 1998). Nevertheless, several strong
ecological interactions control forest development and defor-
estation. These operate at varying spatial and temporal scales
and they will be the focus of this section. For this, we will
consider where kelps can live, where they develop forests,
and under what conditions kelp deforestation takes place.
Kelp forest development
Globally, kelp forests develop on shallow rocky shores in a
mid-latitude band where light and oceanographic conditions
allow the development and persistence of this growth form
(Fig. 1). While kelps can grow in Arctic and sub-Antarctic
regions (see Dunton & Dayton 1995), their abundance and
diversity are low (probably due to light limitations; Dunton
1990; Henley & Dunton 1997) and thus they rarely develop
forests above about 60° latitude. Similarly, warm tempera-
tures and low nutrient concentrations generally prevent kelp
forests from developing in subtropical or tropical regions
(Bolton & Anderson 1987; Gerard 1997). The lowest latitude
kelp beds (see Hatcher et al. 1987) usually correspond with
ocean current driven anomalies in latitudinal gradients of
warm temperatures and/or low nutrient conditions. For
example, kelp forests are found within the tropics of Cancer
and Capricorn only along the western coasts of southern
California to Mexico, northern Chile to Peru, western South
Africa and western Australia (Fig. 1). In these cases, cool
ocean currents ﬂowing toward the equator, or upwelling,
advect cool, nutrient-rich water to the kelp forests.
Low latitude kelp (usually less than 40° latitude) are often
diminutive and share or lose community dominance to
fucoids such as Sargassum and other large brown algae that
become more diverse and abundant toward the tropics. In
North America, southern California kelps share space with
the fucoids Cystoseira osmundacea (40–30° N latitude) and
Sargassum spp. (35–25° N latitude) (Foster & Schiel 1985).
At Japan’s southern island, Honshu, scores of Sargassum
species become increasingly important and eventually domi-
nate macroalgal communities at latitudes less than about
40° N (Fujita 1998). One of the northernmost kelp of the
Southern Hemisphere is in Western Australia at 28–29° S
latitude (Hatcher et al. 1987), where diminutive Ecklonia
share space with brown algal species of Lobophora and
Sargassum. Elsewhere in South Australia (Shepherd &
Womersley 1970) and New Zealand (Choat & Schiel 1982),
large subtidal fucoids cohabit with and often dominate the
kelp. In New Zealand, fucoid densities were twice that of kelp
densities at low latitude (40° S) sites, but reversed to
kelp dominance at the higher latitude site (40° S), where
kelp densities were twice that of the fucoids (Choat & Schiel
1982). New Zealand fucoids frequently dominate shallow
zones (upper 5 m), displacing kelp dominance to below 10 m
(Choat & Schiel 1982). Only the relatively low diversity kelp
forests of South Africa (22–35° S latitude) and northern
Chile (40–20° S latitude) are without fucoids (Bolton 1996).
However, in Chile the large kelp-like Durvillaea sp.(closely
related to the fucoids) commonly dominates shallow zones
(Santelices 1990) and may outcompete Macrocystis there
In general, the three interacting processes that control the
development of kelp forests are recruitment, growth and
competition. Locally, kelp forests are established and main-
tained by successful settlement of zygotes, which grow and
are thinned by mortality from intraspeciﬁc competition
during their benthic life (Reed & Foster 1984; Chapman
1986). Recruitment is often seasonal and inﬂuenced by
environmental conditions at the time of settlement. In
complex kelp forests tiered with multiple levels (for example,
canopy, stipitate and prostrate forms; Table 1) such as the
California kelp forest, kelp recruitment and growth is regu-
lated by light available through breaks in the kelp canopy
(Reed & Foster 1984; Santelices & Ojeda 1984a; Graham et
al. 1997) as well as by available nutrients (Dayton et al. 1999).
Following intense storms that deforest or thin kelp canopies,
recruitment is usually strong, but the kelp species that grow
to dominance will depend upon nutrient conditions at the
time (Tegner et al. 1997).
Kelp growth depends on interactions among nutrient avail-
ability, temperature and light. Kelps dominate cold-water
coastal zones (Fig. 1) but can become physiologically stressed
at high sea temperatures, particularly when nutrient avail-
ability is low (Tegner et al.1996a;Gerard 1997). In some
regions without upwelling, periods of low nutrient concen-
trations correspond with warm summer temperatures when
the water is stratiﬁed. The combined effects of low nutrients
and high rates of respiration result in kelp plants that erode
more rapidly than they grow (Gagne et al.1982; R.S. Steneck,
unpublished data 2002). In kelp forests driven by the upwelling
of new nitrogen, such as those of southern California, warm
surface water temperature is a surrogate for low nutrient avail-
ability (Tegner et al.1996a). In this system when El Niño
events disrupt coastal upwelling, kelp becomes nutrient-
starved and dies back (Tegner & Dayton 1991). As a result, the
distribution, abundance and size of kelp plants decline as sea
surface temperatures increase (Dayton et al.1999).
Kelp forest ecosystems 439
As a group, kelps have a relatively low photosynthetic to
biomass ratio. This constrains them to relatively shallow,
well-illuminated zones compared to other functional groups
of algae (reviewed in Vadas & Steneck 1988). In environ-
ments free of herbivores or other agents of disturbance, kelp
frond size and density decline rapidly with depth making
kelps the shallowest of the major growth forms of marine
algae (Steneck & Dethier 1994). Extended periods of dark-
ness in the Arctic limit the northern distribution (Fig. 1),
diversity and maximum depth of kelps (Henley & Dunton
The limited development of kelps at high latitudes results
from interactions among light, temperature, ice scour, nutri-
ents and evolutionary biogeography. As a result, true kelps
(order Laminariales) grow much further north in the Arctic
(to about 70° North latitude) than they do south in the
Antarctic (to 55° South latitude) (Dunton & Dayton 1995).
Kelp growth in near-freezing seawater temperatures requires
dissolved inorganic nitrogen to physiologically accommodate
photosynthesis (Korb & Gerard 2000). The endemic Arctic
kelp Laminaria solidungula has a remarkable capacity to store
nitrogen that is only available in the winter when it has no
light (Henley & Dunton 1997). In contrast, in the Antarctic
where nitrogen is available all year round, true kelps are
displaced by morphologically similar brown algae,
Himantothallus sp. of the order Desmarestiales (Moe & Silva
1981; Dunton & Dayton 1995). This kelp-like alga has no
inherent capacity to store nitrogen, nor does it need to in the
nitrogen-replete waters of the Antarctic. Under experimental
nitrogen-starved conditions this alga quickly succumbs
(Korb & Gerard 2000). Thus Arctic and Antarctic distribu-
tions of kelp may hinge on physiological adaptations to light
and nutrient limitations in those areas. Arctic kelp illustrates
the remarkable range of physiological tolerance that can
evolve in the group. The effective replacement of
Laminariales in the high Antarctic by large brown algae in
the order Desmarestiales may be an evolutionary priority
effect of an older order in the geologically older Southern
Hemisphere ocean system. Nevertheless, polar populations
of kelps or kelp growth forms may grow to their physiological
limits at high latitudes but they do not attain forest there.
Seasonal change in oceanography and physical disturbances
from ice scour can limit or eliminate subarctic populations of
kelps (Himmelman 1980).
Widespread kelp deforestation can result from disease,
herbivory, and physiological stress or interactions among
those processes. At lower latitude kelp forests (usually less
than 40°), periodic deforestations results from oceanographic
anomalies in temperature, salinity or nutrients that either kill
kelps directly or trigger diseases that become lethal to physio-
logically-stressed plants. At mid-latitudes (about 40–60°),
herbivory by sea urchins is the most common and most
important agent of kelp deforestation. Latitudinal differences
in patterns and processes shaping kelp forests have resulted
in different researchers working in the same kelp forest
system but reaching different conclusions (Foster 1990).
Here we address the geography of kelp deforestation patterns
Kelp-free patches have probably always occurred at some
scale but those created by physical factors tend to be rela-
tively small and short-lived. The oldest term for algal
deforestation is the Japanese word isoyake, which means ‘rock
burning’ (D. Fujita, personal communication 2002). The
word was coined by Yendo (1902, 1903) to describe algal
deforestation in coastal zones of central Japan where the algal
decline was thought to have resulted from salinity anomalies
(Yendo 1903, 1914) rather than grazing, because herbivorous
440 R.S. Steneck et al.
Table 1 Dominant organisms, functional groups and diversity that deﬁne the structure and function of subtidal kelp forest ecosystems of
North America. *Steller’s sea cow (now extinct).
Western North Atlantic Aleutians, Alaska Southern California
Dominant kelp genera Laminaria (1 sp.), Agarum (1 sp.) Alaria (1 sp.), Laminaria (3 spp.), Macrocystis (1 sp.), Pterygophora
Thalassiophyllum (1 sp.), (1 sp), Laminaria (1 sp.), Eisenia
Agarum (1 sp.) (1 sp.), Pelagophycus (1 sp.),
Egregia (1 sp.), Agarum (1 sp.)
Structural tiering Prostrate Canopy, stipitate, prostrate Canopy, stipitate, prostrate
Sea urchins Strongylocentrotus droebachiensis Strongylocentrotus polycanthus Lytechinus anamesus,
Molluscs Haliotis (3 spp.), Tegula (3 spp.)
Fishes Medialuna californiesis, Girella
Marine mammals Hydrodamalis gigas* Hydrodamalis gigas*
Dominant carnivores Atlantic cod (Gadhus morhua) Sea otter (Enhydra lutris), killer Sea otter, spiny lobster (Panulirus
whales (Orcinus orca)interuptus), sheephead ﬁsh
sea urchins were rare. The isoyake killed all foliacious algae
ﬁrst and then all encrusting coralline algae; the algae recov-
ered several years later (Yendo 1914). Other mass mortalities
of Ecklonia- and Eisenia-dominated kelp forests resulted from
incursions of the Kuroshio Current along the central Japan
coast (D. Fujita, personal communication 2002). On Honshu
Island, near the southern limit of Japanese kelp, anomalous
incursions of the warm Tsushima Current periodically create
isoyake conditions. Such oceanographically-induced kelp
deforestations are usually short-lived and reversible, as was
the original isoyake case in Japan (Yendo 1914).
Kelp deforestations also result from El Niño events.
Strong El Niños halt coastal upwelling of nutrient-rich water
and cause surface waters to warm (Dayton et al. 1999). These
anomalies in California caused patchy deforestation followed
by rapid recovery (Tegner & Dayton 1987; Tegner et al.
1997). Such physiological stresses are likely to be more
common toward the low latitude limits of kelp ranges. For
example, the northern limit of three species of brown algae in
northern Chile shifted south toward higher latitudes
following the El Niño event of 1982–1983 (Peters & Breeman
1993). Such stresses may make kelps more susceptible to
disease. Low-latitude kelps in northern New Zealand have
succumbed to a disease that may have resulted from physio-
logical stress (Cole & Babcock 1996; Cole & Syms 1999).
Within mid-latitudes (roughly between 40° and 60° lati-
tude) where kelp development is less likely to be limited by
physical processes such as temperature, nutrients and light,
deforestation most often results from sea urchin herbivory
(Table 2; Leighton et al. 1966; Lawrence 1975; Duggins
1980; Himmelman 1980; Dayton 1985a,b; Estes & Duggins
1995; Mann 2000). This is most evident in the Northern
Hemisphere where the most widespread and long-lasting
herbivore-induced kelp deforestations have resulted from sea
urchin grazing (Table 3). These primarily Laminaria-domi-
nated kelp forests (Fig. 1) have been reduced in historical
times to coralline-dominated ‘urchin barrens’ in the Aleutian
Islands of Alaska, the Gulf of Maine, Canadian Maritimes (to
Newfoundland; Himmelman 1980), northern Japan
(Hokkaido Island), Iceland and northern Norway (Table 3).
South of those regions, forests either remain intact, such as in
southern California (Dayton et al. 1984), or are deforested
patchily in relatively few regions, such as in southern Norway
(Sivertsen 1997), Ireland (Kitching & Thain 1983), the UK
(Kain 1975) and southern Japan (Honshu Island; Fujita
Kelp deforestation from sea urchins is less common in the
Southern Hemisphere (Table 3). In Chile, extensive sea
urchin deforestation at mid latitudes (54°–46° S) was
reported by Dayton (1985b). However that study and others
(Table 3) report only patchy deforestation in north central
Chile (i.e. 32° S latitude; Ojeda & Santelices 1984). Dense
Macrocystis forests with few urchins have been described in
the southernmost regions of South America (around 55° S
latitude) in Chile (Castilla & Moreno 1982; Santelices &
Ojeda 1984b) and Argentina (Dayton 1985b). While south-
ernmost Chile has four sea urchin species (Loxechinus albus,
Pseudechinus magellanicus, Arbacia dufresnei and Austrocidaris
canaliculata), they subsist on drift kelp and rarely graze
attached Macrocystis plants. Along the east coast of southern-
most South America to the northern limit of kelps in
Argentina (42° S latitude; Barrales & Lobban 1975), the sea
urchin, Arbacia dufresnei is the dominant herbivore but its
grazing impact on kelp forests is minimal (Barrales & Lobban
Elsewhere in the Southern Hemisphere, sea urchin defor-
estation is patchy or restricted to particular depth zones. In
New Zealand, a band of sea urchin ‘barrens’ exists at depths
of around 10 m (Choat & Schiel 1982). Urchins there may be
prevented by wave turbulence from grazing shallower zones.
In south Australia, kelp deforestation is restricted to regions
of high spatial heterogeneity that provide shelters for noctur-
nally grazing sea urchins (Andrew 1993). Recent kelp
deforestation in Tasmania resulted from newly established
high-density sea urchin populations (C. Johnson, personal
communication 2002). In South Africa, sea urchins alone do
not overgraze kelp forests (Velimirov et al. 1977) but grazer-
induced phase shifts can occur from a diversity of herbivores
(G.M. Branch, personal communication 2002).
Whatever regulates sea urchin abundances or their grazing
behaviour often controls the distribution and abundance of
kelp forests (Tables 2 & 3). Predators are commonly strong
interactors (sensu Paine 1980) and as such are the single most
important agent controlling sea urchin populations (Cowen
1983; Duggins 1983; Tegner & Levin 1983; Estes & Duggins
1995; Sala et al. 1998; Steneck 1998). When sea urchin preda-
tors become the focus of intense and unsustainable ﬁshing
that extirpates them, hyperabundances of the sea urchins and
kelp deforestation often result (Lawrence 1975; Estes &
Duggins 1995; Steneck 1998). Exceptions to this pattern are
found at lower latitudes, where diverse guilds of kelp forest
herbivores and predators compensate for the loss of a single
predator species. Sea urchin abundance can also be inﬂu-
enced by other factors such as disease (Table 2; Fujita 1998;
Scheibling et al. 1999), turbulence (Choat & Schiel 1982) and
storms (Dayton 1985a; Ebeling et al. 1985) that can locally or
periodically reduce sea urchin abundance and thus control
kelp forest development indirectly (Foster & Schiel 1985).
Widespread, long-lasting kelp deforestation from sea
urchins may be a relatively recent phenomenon. In the
Aleutians of Alaska, the transition may have occurred early in
the 20th century (Estes & Duggins 1995). In Japan, ﬁsher-
folk observed deforestation and patches of corallines ﬁrst in
the early 1930s (Fujita 1987; 1998), although sea urchins
were not mentioned at that time. Later, growing sea-urchin-
grazed coralline patches were reported during the 1950s to
1960s (Ohmi 1951; Fujita 1998). In California during the
1960s, the term ‘barrens’ was coined to describe sea urchin-
induced kelp deforestation (Leighton et al. 1966). In the
North Atlantic, the ﬁrst gaps in kelp forests were reported in
the 1960s for the Gulf of Maine (Lamb & Zimmerman 1964),
Nova Scotia (Edelstein et al. 1969; Breen & Mann 1976),
Kelp forest ecosystems 441
442 R.S. Steneck et al.
Ireland (Ebling et al. 1966), and the UK ( Jones & Kain 1967).
Gaps in kelp forests in the western North Atlantic coalesced
and expanded during the 1970s and 1980s in Nova Scotia
(Mann 1977) and the Gulf of Maine (Steneck 1997).
Expansive coralline ‘barrens’ existed in Newfoundland in the
late 1960s (Himmelman 1980) and possibly were present
there earlier (Hooper 1980). In the eastern North Atlantic,
widespread urchin-induced deforestation was ﬁrst observed
Table 2 Present impacts on kelp forest ecosystems scored along a continuum: non-existent (blank), minimal (1), modest (2), great, but local
(3), both great and widespread (4). The sign of the impact index number reﬂects whether impact tends to reduce kelp forests () or cause
them to increase (). No sign indicates the impact can be positive or negative (for example, nutrient cycles).
Potential impacts Direct human North-west Atlantic Aleutians, Alaska Southern California
impacts impacts impacts impact
Direct kelp harvest Yes 1 (Dayton et al. 1998)
Herbivory: sea urchin grazing 4 (Steneck 1997) 4 (Estes & Duggins 3 (Leighton et al.
Fishing on sea urchins Yes 4, Maine, USA 1 (Tegner & Dayton
(Steneck 1997) 1991)
Disease 4, Nova Scotia, Canada 2 (Pearse & Hines
(Scheibling 1986; 1979)
Scheibling et al. 1999;
Levin et al. 2003)
Storms 2 (Harris et al. 1984;
Harrold & Reed 1985)
Herbivory: other grazers 2, Lacuna snails ( Johnson 1, Steller’s sea cow 1, Steller’s sea cow,
& Mann 1986) (Domning 1972) abalone (Tegner &
Levin 1983; Dayton
et al. 1999)
Predation on sea urchins
Reduced (ﬁshing pressure) Yes 4, cod (Steneck 1997) 4, sea otter (Estes & 3, sheephead ﬁsh
Duggins 1995; Dayton (Cowen 1983; Dayton
et al. 1998) et al. 1998)
Increased (predator switching) 2, crabs (Tegner & 3, killer whales (Estes
Levin 1983) et al. 1998)
Competition from non-native species Yes
Epiphytes Yes 2, bryzoan (Scheibling
1986; Scheibling et al.
1999; Levin et al. 2003)
Exploitative competition Yes 2, Codium sp. (Lambert 1, Sargassum sp.
et al. 1992; Steneck & (Dayton et al. 1998)
Eutrophication Yes 1 (Tegner et al. 1995)
Sedimentation Yes 1 (Tegner et al. 1995)
Oil spills Yes 1 (Dean & Jewett 1 (Dayton & Tegner
El Niño 3 (Tegner & Dayton
1987, 1991; Dayton
et al. 1998)
La Niña 2 (Dayton et al. 1998)
Global warming Yes 1 1 (Dayton et al. 1998)
Storm frequency and intensity 3 (Witman 1987) 2 (Ebeling et al. 1985; 3 (Pearse & Hines
Tegner et al. 1996a) 1979; Ebeling et al.
1985; Tegner et al.
1996a; Dayton et al.
Nutrient availability 1 (Chapman & Craigie 2 (Gerard 1982;
1977; Gerard 1997) Zimmerman &
Kremer 1984, 1986)
Kelp forest ecosystems 443
Table 3 Comparison of subtidal kelp forest ecosystems of the world. Numbers in parentheses denote number of ecologically important
species in subtidal kelp forests for speciﬁed taxa.
Kelps and their controlling agents Spatial and temporal scale of deforestation
Site and Dominant Deforesting Predators or Drift kelp, Regional Depth Local Duration References
latitude kelps herbivores diseases of kelp disease, distribution range distribution deforested
Western North Atlantic
Nova Scotia Laminaria (1), Echinoid (1) Fishes (2), Drift, Widespread Broad Homogenous Decades Tables 1, 2, 4;
43–45° N Agarum (1) urchin disease disease Fig. 5
Maine Laminaria (1), Echinoid (1) Fishes (2) Widespread Broad Homogenous Decades Tables 1, 2, 4;
43–44° N Agarum (1) (crabs) Fig. 4
Eastern North Atlantic
North Iceland Laminaria (1) Echinoid (1) ? Widespread Broad Homogenous ? Hjorleifsson
65° N et al. (1995);
W. H. Adey
North Norway Laminaria (1) Echinoid (2) Seabirds Widespread Broad Homogenous Decades Hagen (1983);
65–71° N Bustnes et al.
South Norway Laminaria (1) Echinoid (1) ? Restricted Broad Patchy Decades Sivertsen (1997)
Britain and Laminaria (3) Echinoids (2) Crabs Restricted Broad Patchy Kitching &
Ireland Ebling (1961);
52–55° N Ebling et al.
East North Paciﬁc
Alaska Alaria (1), Echinoid (1) Sea otter Widespread Broad Homogenous Decades Tables 1, 2, 4;
(Aleutians) Laminaria (3), Fig. 3
50–55° N Thalassiophyllum
(1), Agarum (1)
Southern Macrocystis (1), Echinoids (3), Sea otter (1), Drift, Restricted Broad Patchy Decade Tables 1, 2, 4;
California Laminaria (1), gastropods ﬁsh (1), Oceanographic Figs. 6, 7
30–35° N Pterygophora (1) (8), ﬁshes (2) lobster (1) (ENSO)
West North Paciﬁc
North Japan Laminaria (2) Echinoid (1– 3) Crabs, urchin Oceanographic Widespread Broad Homogenous Decades Fujita (1998);
(SW Hokkaido) disease event D. Fujita,
39–46° N personal
South Japan Undaria (1), Echinoid (3), ? Oceanographic Restricted Broad Patchy ? Fujita (1998); D.
(West Honshu) Eisenia (1), ﬁsh (1) event Fujita, personal
36–38° N Ecklonia (1) communication
East South Paciﬁc
North Chile Lessonia (1), Echinoids (2), Asteroids (3), Drift Widespread Shallow Patchy Decades Ojeda & Santilices
18°–42° S Macrocystis (1) ﬁshes (1), ﬁshes (3) (1984); Vasquez
gastropods (2) (1993); Vasquez
South Chile Macrocystis (1), Echinoids (1), Asteroids (1) Drift Restricted Shallow Homogenous Decades Dayton (1985b)
46°–54° S Lessonia (2) gastropods (1)
Southernmost Macrocystis (1) Echinoids (4) Asteroids (1) Drift None Castilla &
Chile Lessonia (2) Moreno (1982);
55° S Santelices &
Vasquez et al.
Argentina Macrocystis (1), Echinoid (1) ? ? None Barrales &
42°–55° S Lessonia (1) Lobban (1975)
444 R.S. Steneck et al.
in northern Norway in the early 1980s (Hagen 1983;
Sivertsen 1997) and in Iceland in the early 1990s
(Hjorleifsson et al. 1995). By the mid-1970s, sea urchins were
viewed as the major cause of kelp deforestation (Lawrence
1975) such that by the mid-1980s conferences were held to
discuss (among other things) how sea urchins could be erad-
icated (Pringle et al. 1980; Bernstein & Welsford 1982). Sea
urchin-induced kelp deforestation was widely reported in
mid-latitudes of the Northern Hemisphere from 40°–60° N
(higher in the eastern North Atlantic due to the Gulf Stream)
during the 1960s and 1980s. At the time, some researchers
openly wondered if kelp deforestation was an ‘irreversible
degradation’ (Mann 1977).
LONG TERM TRENDS: THREE NORTH
AMERICAN CASE STUDIES
Kelp forests live in a balance between forces contributing to
their development and deforestation. The geography of both
physical and biological forcing functions for kelp forest
development and persistence is complex. We know that
many kelp forests look and behave very differently today
from their historical counterparts. Many of these differences
are due, directly or indirectly to human perturbation of
physical or biological components of kelp ecosystems. Some
of the most impressive changes result from trophic cascades,
largely through the functional elimination of apex predators.
These play out over a period of decades, centuries or longer.
Here we illustrate this point with long-term historical
Table 3 Continued
Kelps and their controlling agents Spatial and temporal scale of deforestation
Site and Dominant Deforesting Predators or Drift kelp, Regional Depth Local Duration References
latitude kelps herbivores diseases of kelp disease, distribution range distribution deforested
West South Paciﬁc
Australia Ecklonia (1) Echinoids (1), Fishes (2) Widespread Moderately Patchy ? Andrew (1993,
(New South ﬁshes (1) deep 1994); Andrew
Wales) & Underwood
32°–35° S (1993); Andrew
Australia Macrocystis (1), Echinoid (1) Fish (1), Restricted Broad Homogenous Years C. Johnson,
(Tasmania) Ecklonia (1) lobster (1) personal
43° S communication
New Zealand Ecklonia (1), Echinoids (2), Fishes (?), Kelp disease Widespread Mid-depth Homogenous Decade
(North Island) Lessonia (1) gastropods (2) lobster (1)
34°–37° S Andrew &
Babcock et al.
New Zealand Ecklonia (1), Echinoids (1) ? Restricted Broad Patchy ? Schiel 1990;
(South Island) Lessonia (1), Schiel et al.
41°–47° S Macrocystis (1) (1995)
East South Atlantic
South Africa Ecklonia (1), Echinoids (1), Lobster (1), Widespread Deep only Patchy ? Anderson et al.
30°–35° S Laminaria (1), gastropods (1) ﬁsh (?) (1997); G.
Macrocystis (1) Branch,
East Indian Ocean
Western Ecklonia (1) Oceanography Hatcher et al.
Australia (high (1987)
28° S temperature
Kelp forest ecosystems 445
chronologies, some known and others inferred, from Alaska,
the western North Atlantic, and southern California. These
three systems are among the best studied in the world
and the only ones we know for which archaeological data
The kelp forest ecosystems of the eastern North Paciﬁc likely
arose during the last 20–40 million years with the evolution
of kelps, strongylocentrotid sea urchins, sea otters, and the
now extinct Steller’s sea cow (Estes et al. 1989). Kelp forests
probably dominated ice-free coasts throughout this region
since the last glacial period (Table 4). During the Pleistocene,
sea cows ranged from Japan, throughout the North Paciﬁc to
at least Monterey Bay in California ( Jones 1967). They may
have been extirpated from most of their range by aboriginal
hunting at the end of the Pleistocene and early Holocene,
because they survived thousands of years longer in the
Commander Islands, a region that was not peopled until the
time of European contact in 1741 (Estes et al. 1989).
European fur traders killed the last sea cow 27 years later.
The impact of sea cows on kelp forests is unknown. However
they had no teeth, probably were unable to dive and thus
were most likely a trophic specialist of canopy kelps
(Domning 1972; Clementz 2002). Kelp canopies are remark-
ably resilient to cropping of their most distal fronds and
currently, in California, they support a multi-million dollar
industry of canopy-cropping factory ships that sustainably
harvest kelp for their valuable alginates. If harvesting factory
ships do little permanent damage to kelp forests (Tegner &
Dayton 2000), it is unlikely that sea cow grazing of canopies
deforested kelp beds.
It is possible that the concentration of food associated with
highly productive North Paciﬁc kelp forests attracted early
maritime people and facilitated some of the earliest migration
of people from Eurasia to the Americas. It is now apparent
that anatomically modern humans had colonized the Ryuku
Islands south of Japan between 35 000 and 25 000 years ago
and that boats were in use along the Japanese coast by about
25 000 years ago (Erlandson 2001). The readily-available
shellﬁsh, ﬁnﬁsh, and marine birds and mammals may have
allowed human populations to become established, prolif-
erate and grow well before landward migration and the
eventual development of agriculture began. Furthermore,
climatic conditions along the land-sea margin were benign
compared with more inland areas. It is possible that the
whaling tradition of indigenous people of the North Paciﬁc
began with the overharvest of the predator-naive and
defenceless Steller’s sea cow, focusing thereafter on cetaceans
that were more difﬁcult to harvest, once the sea cows were
extirpated (Domning 1972). Currently, the earliest evidence
for human occupation of the southern Alaskan coast
(Aleutians to south-east Alaska) dates to between 9000 and
10 000 years ago, although rising post-glacial sea levels may
well have submerged earlier sites.
Alaskan kelp forests are likely to have been well developed
before human contact because sea otter predation on sea
Table 4 Spatial and temporal scale of change among three kelp forest ecosystems in North America.
Event North-west Atlantic West Aleutian islands Southern California
Pristine state (prior to Kelp forest Kelp forest Kelp forest
Scale (patch size) 100–500 km2(Johnson & Mann 200– 1000 km2(Estes et al. 1989) 10 km2(Harrold & Reed 1985;
1988) Tegner et al. 1996a)
First human 10 000 (Bourque 1995) 8000 (Simenstad et al. 1978) 12 000–13 000 (Erlandson
contact/occupation et al. 1996)
(years before present)
Marine organisms 5000 (Bourque 1996) 4500 (Simenstad et al. 1978) 11 600 (Erlandson et al. 1996)
present in diet
(years before present)
First known phase change ?–40 (Adey 1964) 2500 (Simenstad et al. 1978) 4000–6000 (Salls 1991, 1995;
(years before present) Erlandson et al. 1996;
Erlandson & Rick 2002)
First European 460–400 (Steneck 1997) 260 (Simenstad et al. 1978) 460–200 (Simenstad et al. 1978)
(years before present)
Kelp bed re-establishment 0.5 –4 ( Johnson & Mann 1988; 2 (Estes et al. 1989) 0.5 (Harrold & Reed 1985;
rate (years) Dayton et al. 1999) Tegner et al. 1996a)
Alternate (kelp free) state Coralline/urchin Coralline/urchin Coralline/urchin
Storms remove dominant Small scale removal (Johnson & — Large scale removal
kelp Mann 1988)
Recent apex predators Crabs (Leland & Steneck 2001) Killer whales (Estes et al. 1998) —
Introduced competitors Bryozoan, Codium (Lambert
et al. 1992)
446 R.S. Steneck et al.
urchins prevented overgrazing on kelp (Simenstad et al.
1978; Estes et al. 1998; left side Fig. 2). Aboriginal Aleuts
greatly diminished sea otters beginning around 2500 yr BP
with a corresponding increase in the size of sea urchins
(Simenstad et al. 1978). European and North American fur
traders subsequently hunted the remaining otters to the brink
of extinction in the 1700 and 1800s, causing the collapse of
kelp forests as they were grazed away by sea urchins released
from sea otter predation (middle of Fig. 2 and Fig. 3). Legal
protection of sea otters in the 20th century reversed their
decline and the resultant trophic cascade (Fig. 3), but kelp
forests have disappeared again over vast areas of the south-
west Alaska coast as sea otter populations have fallen prey to
killer whales (Estes et al. 1998). The whales apparently
shifted their diet to sea otters from seals and sea lions, after
the latter populations declined signiﬁcantly. The pinniped
declines are likely to have been caused by changes in the open
ocean. Whether these changes were natural or anthropogenic
Western North Atlantic
Kelps and sea urchins in the Gulf of Maine were derived
from the North Paciﬁc by way of the transarctic interchange
(Durham & MacNeil 1967; Vermeij 2001). Sea otters did not
make this journey. On arriving in the north-western Atlantic,
however, sea urchins encountered functionally-similar
predators, namely Atlantic cod and other large groundﬁsh,
which maintained the kelp-dominated state via a trophic
cascade. These predatory ﬁshes have been present in the Gulf
of Maine archaeological record for at least 5000 years
(Bourque 1995; Steneck 1997).
Indigenous ﬁshers exploited cod by hook and line for
thousands of years (Bourque 1995, 2001; Steneck 1997).
They maintained a varied diet of marine organisms such as
cod, other ﬁsh, oysters and clams, as well as terrestrial
animals such as deer and sea mink, the latter now extinct
(Bourque 1995). When the ﬁrst Europeans explored the Gulf
of Maine, the abundance of large ﬁsh impressed them (Rosier
1605). Vespucci marked the western North Atlantic coast on
his 1526 map of the New World Bacallaos, which is
Portuguese for ‘land of the codﬁsh’. In 1602, Bartholomew
Gosnold named Cape Cod for the myriad ﬁsh that ‘vexed’ his
ship. By all accounts, cod and other large predatory ﬁsh were
stable components of coastal zones throughout the western
Figure 2 Timing of phase
changes in community state
of kelp forests of North
America. Kelp with
vertebrate predators, sea
urchins without kelp and kelp
without predators have been
identiﬁed for some or all of
the case study locations. Kelp
forests are listed from the
greatest number of trophic
levels on the left to fewest
trophic levels on the right.
Case studies are listed from
lowest species diversity in
Maine to highest diversity in
southern California. See text
or Table 4 for references.
Figure 3 Temporal trends in kelp forests, predators and sea
urchins of Amchitka, Alaska. Abundances estimated from several
studies (see text) and Estes and Duggins (1995). Arrows indicate
the timing of change in major community-changing forcing
functions. Width of arrowheads indicates magnitude of the forcing
North Atlantic (Steneck 1997). Further, in rare offshore
habitats where large predatory ﬁshes still persist, urchins are
rare, kelp is abundant (Vadas & Steneck 1988, 1995) and
attack rates on tethered sea urchins are high (Vadas &
Steneck 1995). Signiﬁcantly, very small urchins (a few
millimetres in diameter) were found at this site indicating
that they can recruit but do not, probably because of high ﬁsh
predation rates (Vadas & Steneck 1995).
We surmise that kelp forests dominated the benthos while
predatory ﬁnﬁsh were abundant in coastal zones through at
least the 1930s. The earliest reports of algae support that
supposition. Hervey (1881) described all three dominant kelp
species (Table 1) as being ‘very abundant from Greenland to
Cape Cod’ and often ‘washed ashore in great numbers’.
Windrows of kelp detritus are good indicators of a kelp-
forested state (Novaczek & McLachlan 1986), there was no
mention of expansive patches of coralline algae at that time
and the earliest scientiﬁc study in the region ( Johnson &
Skutch 1928) reported that kelps were the ‘most character-
istic plant in the midlevels of the sublittoral zone.’ Similarly,
Nova Scotia was described as kelp-dominated in the early
1950s (MacFarlane 1952).
Extensive ﬁshing grounds for cod and other predatory
ﬁshes were ﬁrst mapped for coastal zones in Maine in the
1880s and then again in the 1920s, with remarkably little
change in areal extent or location (Steneck 1997). Cod stocks
persisted until mechanized ﬁshing technology and on-board
refrigeration allowed spawning aggregations of cod to be
targeted in the 1930s (Rich 1929; Conkling & Ames 1996).
This set off a rapid decline in the numbers and body size of
coastal cod in the Gulf of Maine (Steneck 1997; Jackson et al.
2001). Data from 5000 year old Indian middens, and from
ﬁsheries over the past century document a relatively recent
but rapid decline in the average cod body-size, coincident
with their extirpations from coastal zones ( Jackson et al.
2001; Steneck & Carlton 2001). Dominant ﬁsh predators in
the coastal zone were replaced by small, commercially less
important species such as sculpins (Steneck 1997). Today,
large predatory ﬁnﬁshes remain functionally absent from
coastal regions of the western North Atlantic (Steneck 1997).
Predatory ﬁshes consume and control the distribution and
abundance of sea urchins (Keats et al. 1986; Vadas & Steneck
1995). The extirpation of coastal cod and other ﬁshes by the
1940s in the Gulf of Maine resulted in functional loss of apex
predators, which fundamentally altered coastal food webs as
lobsters, crabs and sea urchins all increased in abundance
(Steneck 1997). In the 1960s, scuba diving allowed coastal
ecosystems to be observed and described in situ for the ﬁrst
time. The coastal Gulf of Maine was described then as a
mosaic of kelp forests (Lamb & Zimmerman 1964) and
widely spaced ‘barren’ patches of sea urchins and coralline
algae (Adey 1964; W.H. Adey, personal communication
2001; Fig. 4). Similar patches were described a decade later in
Nova Scotia (Breen & Mann 1976; Fig. 5). Over the next two
decades, sea urchin abundances increased throughout the
Gulf of Maine, kelp forests declined and coralline barrens
grew and coalesced (Steneck 1997; Fig. 2). From the mid
1980s to the early 1990s, kelp forests reached an all-time low
in their distribution and abundance throughout the region
(Steneck 1997). Similar developments were observed in Nova
Scotia, except that the system there was punctuated with
disease-induced mass mortality of urchins causing it to
rapidly oscillate between forested and deforested states (Fig.
6, discussed below). Arguably, the conditions for high densi-
ties of sea urchins that repeatedly denuded coastal zones of
eastern Nova Scotia were only possible after the system
became functionally free of apex-predators (middle and right
sides of Fig. 2).
Kelp forest ecosystems 447
Figure 4 Temporal trends in kelp forests and sea urchins in the
Gulf of Maine in the western North Atlantic. Width of arrowheads
indicates the magnitude of the forcing function’s impact.
Figure 5 Temporal trends in kelp forests and sea urchins of Nova
Scotia. Abundances estimated from Edelstein et al. (1969), Breen
and Mann (1976), Warton and Mann (1981), Scheibling and
Stephenson (1984), Scheibling (1986) and Johnson and Mann
(1988). Width of arrowheads indicates the magnitude of the forcing
In 1987, a ﬁshing industry developed to harvest Maine’s
sea urchins for their highly valued roe. This caused a remark-
ably rapid decline in urchin distribution and abundance
(Steneck 1997; Vavrinec 2003). In response, kelp forests
recovered, but were now without a functional herbivore
trophic level (right side of Fig. 2). As a result, primary
production was enhanced, and the revegetated habitat also
increased the recruitment potential for some ﬁsh (Levin
1991, 1994) and crabs (McNaught 1999). Newly-settled crabs
(Cancer spp.) survived in high numbers and became micro-
predators of settling sea urchins. The micropredator impact
of crabs on urchins is striking because they can consume
entire cohorts of settled sea urchins over large areas. Areas
with rates of urchin settlement exceeding 10 000 individuals
m2had no survivors a year later due to crab micropredation
(McNaught 1999). The phase change to macroalgae in the
absence of large predatory ﬁnﬁsh allowed large populations of
adult-sized crabs to accumulate. Experiments to reintroduce
adult sea urchins to population densities historically main-
tained prior to the sea urchin ﬁshery resulted in attacks by
swarms of large crabs (Leland 2002). In 2000, all 24 000
introduced urchins were consumed within two months.
Similar patterns were observed in 2001 on a second attempt
to reintroduce adult urchins to the region (Leland 2002).
Thus a new ‘apex’ predator has emerged in coastal zones of
the Gulf of Maine with functionally the same impact of verte-
brate apex predators. Fished-down coastal food webs in the
western North Atlantic resulted in an herbivore-free alter-
nate stable state dominated for the time being by kelp.
The highly diverse food web of southern California kelp
forests historically included several herbivorous species of sea
urchins, small crustaceans, abalone and other snails as well as
their predators including sea otters, spiny lobsters and large
sheephead labrid ﬁsh (Table 1). At the time of early
European contacts (AD 1542–1769), the kelp-laden and
protected coastlines of the southern California Bight
supported one of the highest concentrations of hunter-gath-
erers in human history. Archaeological evidence shows that
maritime Indians colonized California’s northern Channel
Islands at least 12 000–13 000 years ago (Erlandson et al.
1996; left side of Fig. 2; Table 4). These early maritime
people had seaworthy boats and by 10 000 years ago devel-
oped the ﬁrst hook and line ﬁshery in the Americas (Rick et
al. 2001). They relied heavily on marine resources associated
with kelp forests such as abalone shellﬁsh, sheephead and
marine mammals (Erlandson et al. 1996). Human settlement
proliferated through the Holocene and later Indian peoples
may have exploited kelp forest predators and herbivores with
sufﬁcient intensity to cause localized phase-shifts. In San
Clemente Island middens, the size of sheephead bones
decreased as prehistoric human populations increased with
increased ﬁshing intensity and a later overlying stratum was
composed almost entirely of purple sea urchin remains (Salls
1991, 1995). Similar urchin ‘lenses’ have been noted at
several San Miguel Island sites, all of which, so far, date to
the last 3500–4500 years. This suggests that Indians may
have created local sea-urchin barrens by hunting such sea
urchin predators as sea otters and sheephead (Erlandson et al.
1996). If these were the ﬁrst human-induced phase shifts in
the system, they occurred thousands of years after ﬁrst
human contact and appear to have been localized and short-
duration events (Table 4). After European contact,
traditional ﬁshing economies of Native American peoples
were severely disrupted and effectively ended as their popu-
lations were decimated by old-world disease epidemics and
colonial oppression (Tegner & Dayton 2000; Erlandson &
The maritime fur trade functionally eliminated sea otters
from southern California by the early 1800s (Tegner &
Dayton 2000). However, widespread phase shifts to the
deforested state were not observed until 150 years later
(Tegner et al. 1996a; Fig. 6). This apparent lag timing
between local sea otter extinction and urchin-induced phase
shifts in kelp forest probably resulted from the buffering
inﬂuences of alternate predators, herbivores and competitors
(Table 1; Cowen 1983; Tegner & Levin 1983; Schmitt 1987).
Spiny lobsters and sheephead are both subtropical generalist
predators that feed on sea urchins and reach their northern
range limits in southern California (Dayton et al. 1998;
Tegner & Dayton 2000). Sheephead became larger and more
abundant after Native Americans stopped ﬁshing on them. It
is also possible that their populations along with that of spiny
lobsters increased in abundance following the sea otter’s
448 R.S. Steneck et al.
Figure 6 Temporal trends in kelp forests of Point Loma
California. Abundance estimates summarized in Leighton et al.
(1966), Tegner et al. (1996a), McGowan et al. (1998) and Tegner
and Dayton (2000). Width of arrowheads indicates the magnitude
of the forcing function’s impact. The boxed area on the right of the
ﬁgure indicates a period of high resolution subtidal data (see
demise. Several species of abalone (Haliotis spp.; Table 1)
share food and habitat with sea urchins and these competitors
may also have kept urchin populations in check (Tegner &
Levin 1983). Released from predation by otters and Native
Americans, populations of several abalone species expanded
and became targets of a new ﬁshery beginning in 1850 (Cox
1962; Tegner & Dayton 2000).
During the mid-20th century, intensiﬁed ﬁshing pressure
on the remaining predators (spiny lobster and sheephead)
and herbivorous competitors (ﬁve species of abalones) may
have relaxed the predation and competition that had been
controlling sea urchin populations and phase shifts became
widespread for a relatively brief period of time (Fig. 4).
Spiny lobster landings peaked in 1894, and then stabilized at
lower levels. However, the larger lobsters, those most adept
at killing adult sea urchins (Tegner & Levin 1983), had
become rare. Fishing pressure on large male sheephead
increased after the advent of skin diving in the 1940s and
accelerated in the 1950s with the loss of alternative targets
(Dayton et al. 1998). Abalone harvests accelerated in the
1950s, causing widespread population declines by the late
1960s such that these grazers are also now functionally
extinct. El Niño events, pollution discharge and sedimen-
tation accelerated the loss of kelp, which, along with an
increase in destructive urchin grazing, resulted in a phase
shift to a largely kelp-free state in the 1950s and 1960s (Fig.
6). Finally, a sea urchin ﬁshery developed and expanded
rapidly in the early 1970s, reducing grazing pressure in some
areas (Tegner & Dayton 1991). Commercial harvesting
reduced the distribution, abundance, and body size of
exploited urchin stocks, leading toward another phase shift
back to a forested state. In 1988, a market developed for live
sheephead, which resulted in their virtual elimination as a
predator in this system (Tegner & Dayton 2000). Thus the
diversity of functionally important species in southern
California continues to decline, and, in time this could make
the system as a whole less resistant to phase shifts. The few
breaks in the canopy kelp that occurred since 1965 resulted
from strong storms related to El Niño or La Niña events
(Fig. 6) or occasional outbreaks of small herbivorous crus-
taceans released from their predators due to El Niño
southern oscillation (ENSO) changes in coastal oceanog-
raphy. Intense storms in 1983 and 1988 reduced the density
of the three dominant kelp genera (Macrocystis, Pterygophora
and Laminaria; Fig. 7), but recovery was rapid due to high
recruitment into the breaks in the Macrocystis canopy
(Tegner et al. 1997). In general, the greater diversity of
southern California kelp forests, including urchin predators
no longer harvested by Native Americans after Spanish colo-
nization, may have buffered the phase shift to a deforested
state and facilitated recovery from physical (oceanographic)
disturbances. However, this system has experienced serial
trophic-level dysfunction, beginning with sea otters and
more recently including virtually all other functionally-
important predators and herbivores.
Biodiversity, trophic cascades and rates and
consequences of kelp deforestation
In the North American case studies, the extirpation of preda-
tors led to increased herbivory by sea urchins resulting in
kelp deforestation at local to widespread spatial scales (Fig.
2). In the western North Atlantic and Alaska, where predator
diversity is low (Table 1), the transition between kelp forests
and coralline communities was rapid, frequent (Fig. 5), wide-
spread and in some cases long-lasting (Table 4). These
patterns differ from southern California, where the diversi-
ties of predators, herbivores and kelps are high (Table 1),
deforestation events have been rare or patchy in space and
short in duration (Harrold & Reed 1985), and no single domi-
nant sea urchin predator exists (Fig. 6; Tegner & Dayton
2000). The biodiversity within functional nodes, such as
trophic levels, is critical to the structure and functioning of
kelp forest ecosystems. Nevertheless, even the most diverse
systems can and are losing their functional diversity as over-
ﬁshing reduces the ecologically effective population densities
of important species rendering them ecologically-extinct
(Estes et al. 1989). All of this suggests that the fragility and
rate of change in kelp forest ecosystems may depend on local
biodiversity. It remains an open question whether diverse
kelp forests will persist or if serial disassembly and instability
will inevitably result.
The consequences of kelp deforestation can affect
surrounding marine and terrestrial habitats. Drift from giant
kelp (Macrocystis pyrifera) dominates nearshore-produced
phytodetritus in the Southern California bight, contributing
between 60 and 99% of beach-cast autotrophic detritus
(Zobell 1971). Similar estimates were made for areas adjacent
Kelp forest ecosystems 449
Figure 7 Temporal trends in the kelp forest of Point Loma,
California, USA, 1983–1996 at 12 m depth (from Tegner et al.
1996a). Population density data are shown for canopy (Macrocystis)
and stipitate (Pterygophora) kelps. Percentage cover data are shown
for the prostrate kelp Laminaria.
to kelp forests of eastern Nova Scotia (Mann 2000). Offshore
contributions are facilitated by gas-ﬁlled ﬂoats and stipes
which, when adult sporophytes are detached from the bottom
due to grazing or physical disturbance, provide for long-
distance dispersal by rafting (Harrold & Lisin 1989; Hobday
2000). When ﬂoating kelp rafts are deposited on the shore,
the ﬂoats break, and they wash into shallow nearshore habi-
tats and ultimately into offshore basins (Graham et al. 2003).
Secondary productivity of both shallow (Vetter 1995) and
deep-sea (Harrold et al. 1998) soft-sediment systems is
consequently driven in a large part by allochthonous food
subsidies from regional kelp resources. Kelp detritus can also
make its way into nearby intertidal food webs through either
the capture of ﬁne kelp particles by ﬁlter feeders (for
example, mussels [Duggins et al. 1989] or clams [Soares et al.
1997]) or large pieces of drift kelp by limpets (Bustamante et
al. 1995; Bustamante & Branch 1996) and sea urchins (Day &
Branch 2002). Kelp detritus enhances the inherently low
productivity of terrestrial ecosystems on arid islands (Polis &
Hurd 1996). During dry years, carbon and nitrogen from
marine bird and mammal faeces and beach cast marine
detritus fuel terrestrial productivity, with the greatest impact
on islands with large ratios of shoreline to area (Graham et al.
2003). The importance of marine subsidies lessens during
rainy years when high precipitation increases terrestrial
production. Excellent examples are found in southern
California where numerous islands of low productivity (for
example, the Channel Islands) are embedded within a highly
productive marine system. In addition to localized areas of
high accumulation of guano and pinniped excrement, the
shoreline is loaded with large quantities of kelp detritus
(Graham et al. 2003).
POTENTIAL STATES IN 2025
Extrapolation of known trends
It is likely that climate change, human population growth,
coastal development, oil spills, ﬁsheries-induced impacts,
and invasions of non-native species, will continue and poss-
ibly accelerate over the 2025 time horizon. All of these may
well contribute to the continuing disassembly of kelp forest
ecosystems. It is difﬁcult to extrapolate known trends into the
future, because non-linear thresholds and complex interac-
tions can cause ecosystems to behave unpredictably (see
Scheffer et al. 2001). Some activities may change the nature
of functional relationships. For example, overﬁshing of
predators of sea urchins can cause kelp forests to decline, but
overﬁshing of sea urchins themselves can have the reverse
effect (discussed above). Since the primary market for sea
urchin roe is Japan, the currency value of the Yen could regu-
late ﬁshing pressure on this driver of many kelp forest
ecosystems. However it is impossible to predict how global
economic markets will evolve over the next several decades.
While global temperature is expected to increase over the
period to 2025, related patterns of droughts, ﬁres, heat waves,
storms and precipitation are expected to increase in some
areas but decline in others (Houghton et al. 1996). This
underscores the limitations of attempting to generalize about
kelp forest ecosystems globally. However, by considering the
future of the three case studies described above, we may gain
insight into some of the potential risks that befall these
ecosystems with widely varying diversities and scales in space
and time. Our review considers trends from the largest spatial
and temporal scales to the smallest.
Ocean-climate change: global warming, regime shifts
Ocean temperature regulates the physiology and biogeog-
raphy of marine algae (Adey & Steneck 2001). Global
warming, regime shifts and ENSOs are climate-driven
thermal effects that can impact kelp forest ecosystems at a
wide range of intensities that operate at several temporal and
spatial scales. Global warming operates at the largest
temporal and spatial scales (see Fig. 8), but the projected
changes over the next two decades are modest compared to
ocean-basin-scale regime shifts. Regime shifts cause
temperature ﬂuctuations nearly an order of magnitude
greater and persist at the temporal scale of decades (Fig. 9).
Superimposed on both of those changes are ENSOs, which
can cause the greatest temperature anomalies but impact
coastal zones at smaller scales and over periods of only a year
or two. The interaction among these three ocean climate
effects is complicated because each varies differently in space
and time. However, when two or more thermal anomalies
coincide, the compounded perturbations to kelp forests can
be staggering (Paine et al. 1998).
450 R.S. Steneck et al.
Figure 8 Temperature trends over the past 600 years in the
Northern Hemisphere based on multiproxies of palaeoclimates
from atmospheric and oceanographic sources (Mann et al. 1998).
The average temperature over the 600-year period is 0.0°C. Long-
term trends are evident in the average trends in 50 year running
Mean annual temperature in the Northern Hemisphere
has increased by 0.8°C over the past century (Mann et al.
1998) and is expected to increase by 0.28°–0.58°C globally by
2025 (Houghton et al. 1996). The rate and magnitude of
global temperature rise during the past century is unprece-
dented over at least the past 600 years (Fig. 8; Mann et al.
1998). As a group, kelps are limited to coldwater coastal zones
(Fig. 1). Thus kelps living close to their upper thermal limits
will be likely to recede to higher latitudes during protracted
Temperature effects are complicated by a variety of
ocean-atmosphere interactions. While the long-term
frequency of strong storms in the eastern North Paciﬁc
Ocean has changed relatively little since AD 1625 (Enﬁeld
1988), recently the frequency and intensity of extreme
cyclones have increased markedly (Graham & Diaz 2001).
Two storms in the 1980s were described as storms of the
century (Seymour et al. 1989; Fig. 9). The recent steady
(rather than stepwise) increase in storm activity may be
related to rising sea surface temperatures in the western trop-
ical Paciﬁc (Graham & Diaz 2001). Thus, abiotic
storm-induced disturbances of kelp forests could increase
commensurate with global warming (Fig. 9). However, kelp
forests usually recover rapidly from such disturbances
(Tegner et al. 1997).
In 1982–1983, a 4–5°C ENSO warming halted coastal
upwelling and created one of the most severe disturbances of
a giant kelp forest ever documented (Paine et al. 1998).
Nitrogen concentrations in seawater inﬂuence the welfare of
kelps and vary with ocean temperature (Gerard 1997). When
surface waters stratify due to ENSO or as a result of global
warming, nitrogen concentrations decline and kelps become
nitrogen limited and may die (Dayton et al. 1999).
Macrocystis beds are particularly vulnerable because they
possess limited nitrogen storage capacity (Gerard 1982).
Although nitrogen in groundwater run-off and atmospheric
inputs has increased in coastal oceans ( Jickells 1998), its
impact is modest or undetectable relative to natural ﬂuxes in
all but enclosed basins with limited ﬂushing ( Jickells 1998).
Thus, trends toward increasing temperature and decreasing
nitrogen availability in kelp forests are likely to continue.
The combination of warming coastal oceans and increased
stratiﬁcation will be likely to shrink the geographic range of
kelp beds living closest to the tropics. Indirect effects of
temperature on species that inﬂuence kelp forests will be
discussed below. Sea level might rise by 20 cm by the year
2050 as a result of global warming (Houghton et al. 1996).
This could cause increased coastal erosion and turbidity,
which could then affect kelp depth distribution (Vadas &
Steneck 1988) and possibly other processes such as photo-
synthesis and recruitment (Graham 1996). Increased
sedimentation could also reduce the area of substrate avail-
able for kelp settlement.
Regime shifts and ENSO events can cause thermal anom-
alies of shorter duration but potentially greater impact. In
general, ENSO events create warmer sea temperatures along
the southern California coast, but the events themselves are
somewhat more frequent during regime shifts that them-
selves create warmer conditions on that coast (Fig. 9).
Following El Niño warming comes La Niña cooling
(McGowan et al. 1998), which is often associated with violent
storms (Tegner & Dayton 1987, 1991; Seymore et al. 1989;
Tegner et al. 1997).
One of the best-studied climatic regime shifts began in the
mid-1970s and ended at the end of the 1990s (Steele 1998). If
the new regime’s duration is similar to the last two (Fig. 9),
then cooler than average temperatures and perhaps lower
than average ENSO and storm events may be expected over
much of the next quarter century (McGowan et al. 1998; Fei
Chai, personal communication 2002). In our case studies,
only the California system was signiﬁcantly affected by
storms. However the duration of storm impacts is usually
brief and system recovery is rapid; thus, they should have
little lasting impact (Figs. 6–7; Tegner et al. 1997).
Changing coastal biodiversity: new apex predators
Changes in kelp forest biodiversity that affect functional
components of kelp beds can disrupt the system in both
predictable and unpredictable ways. The most conspicuous
changes to the kelp forest result from overﬁshing key drivers
such as apex predators and sea urchins (Estes & Duggins
1995; Jackson et al. 2001). Reductions in either driver can
lead to trophic-level dysfunction, and alternate stable states
or large-scale instabilities. For example, extirpation of sea
urchin predators led to hyperabundances of sea urchins that
have been stable for decades in Alaska (Fig. 3) and remark-
ably unstable in Nova Scotia (Fig. 5) due to epizootic disease
cycles (Scheibling et al. 1999). Arguably, predator loss led to
Kelp forest ecosystems 451
Figure 9 Recent decadal to annual thermal sea surface
temperatures (SST) for southern California (from McGowan et al.
1998). The variable bold line over the duration of the time interval
indicates the 27-month running mean. The dark horizontal lines
indicate regime-shift averages of mean temperature. Superimposed
boxes identify periods of regime shift and ENSOs. S intense
storms (Seymour et al. 1989; Dayton et al. 1999).
hyperabundances of species setting the stage for disease-
related mass mortalities (McNeill 1976).
Apex predators in pristine kelp forests were probably
relatively large vertebrates. Our case studies show a consis-
tent trend of ﬁshing down food webs, such that large
vertebrates were often targeted and extirpated relatively
rapidly. Today, ﬁsh are the most commonly identiﬁed
predator, but sea otters and sea ducks are also important
vertebrate predators in some northern regions (Table 3). In
all cases these vertebrate predators are smaller in body size
and/or fewer in number than they were in the past (i.e. at
ﬁrst human contact). Crustaceans such as spiny lobsters and
crabs are among the most important invertebrate predators
(Table 3). If we extrapolate from the known trend in the
Gulf of Maine case study in which extirpation of large
predatory ﬁnﬁsh led to the dominance of crabs as apex
predators, then it is possible that crab predation elsewhere is
the result of a disrupted trophic cascade. Crabs are dominant
predators of sea urchin in Japan and in the UK (Table 3).
Both of these regions have a long history of ﬁshing that
targeted and extirpated coastal groundﬁsh. In Europe,
coastal groundﬁsh stocks were fought over in the 13th and
14th centuries and their depletion is thought to have
contributed to the development of distant ﬁsheries in Iceland
and eventually North America (Kurlansky 1997). Thus it is
possible that the very early extirpation of apex predators in
Europe’s coastal zones led to the rise of crabs as predators
just as has more recently happened in the Gulf of Maine
(Leland & Steneck 2001).
Overﬁshing in kelp forests leads to ecological (Estes et al.
1989) and possibly absolute (Tegner et al. 1996b) extinctions.
This loss in biodiversity may make these systems more
susceptible to invasion from non-native species (Stachowicz
et al. 1999). Recently, kelp forests in the western North
Atlantic have been invaded by the green alga Codium fragile,
an introduced competitor that could replace the kelp species
in this region, which have a long history of resilience and
dominance (Fig 5; Chapman & Johnson 1990; R.E.
Scheibling, personal communication 2001). The possible
replacement of Laminaria sp. by Codium sp. requires sufﬁ-
cient breaks in the canopy for the latter species to take hold.
This has been facilitated by the introduction in the 1970s of
a non-native encrusting bryozoan that coats, embrittles and
opens the kelp canopy every summer (Lambert et al. 1992;
Levin et al. 2003). However, these two invaders join a long
list of invading species that have become important players in
kelp forest ecosystems of the western North Atlantic
(Steneck & Carlton 2001). Species such as the common peri-
winkle (Littorina littorea) and the green and shore crabs
(Carcinus maenas, Hemigrapsus sanguineus) have not only
invaded, but in many cases have come to dominate the
ecosystem (Steneck & Carlton 2001). Whereas other marine
systems have a history of invasion, few have seen the large-
scale changes in dominance evident in the western North
Atlantic. The successful series of invasions there stands in
stark contrast with patterns observed in the species-rich
southern California kelp forests, where introduced species
generally remain subordinate to native dominants.
Declining water quality
Coastal development often reduces the permeability of soil in
the watershed resulting in greater run-off and increased
turbidity from plankton and particulates. If this occurred, the
areal extent of kelp beds would shrink as the areal extent of
the photic zone and thus their habitable area declined.
Extremely low iron availability may reduce the potential
productivity of kelp. If nitrogen compounds increase due to
run-off, sewage disposal from population centres, or nitrogen
input from atmospheric sources, nutrient availability could
increase. However, human inputs of nutrients are evident
only in coastal areas of restricted water exchange and most
coastal zones ‘appear to be still dominated by large inputs
from the open ocean and there is little evidence of anthro-
pogenic perturbations’ ( Jickells 1998).
In heavily urbanized areas of Japan, terrestrial deforesta-
tion and damming of rivers is thought to starve coastal zones
of the iron and humic substances necessary for kelp develop-
ment (Suzuki et al. 1995; Matsunaga et al. 1999). This is
hypothesized to create a phase shift from kelp dominance to
coralline dominance without any changes in herbivory, sea
temperature or macro-nutrients (Matsunaga et al. 1999). We
know of no other urbanized area (for example, southern
California or Boston, Massachusetts, USA) where crustose
coralline algae dominate shallow rocky shores without herbi-
Point-source pollution is often very conspicuous, but
rarely has it resulted in a serious deforestation of kelp ecosys-
tems. A broken sewer in southern California in the late 1950s
and early 1960s may have resulted in the increase of sea
urchins and the decline in Macrocystis canopy there during
that period (see Fig. 6 and previous discussion), but the
impact was confounded by several other factors. The massive
Exxon Valdez oil spill in 1989 occurred in the vicinity of kelp
forests of south-west Alaska. However, kelps were minimally
impacted and they recovered rapidly (Dean & Jewett 2001).
For most components of the kelp forests, full recovery took
two years or less (Dean & Jewett 2001). Oil covered sea otters
in the spill area, and while the accounts of impacts on them are
debated (Paine et al. 1996), the greatest impact was a decrease
from around 5 otters km1of shoreline to between 2–3 otters
km1(Paine et al. 1996). Otters were never absent from the
system, but they may have suffered a long-term and lingering
impact from oil in at least parts of spill area in Prince William
Sound (Monson et al. 2000). While otters and other air-
breathing predators may be most susceptible to oil spills, with
effects similar to overﬁshing, that may not hold for apex
predators that are ﬁsh. For example, following the Exxon
Valdez oil spill, mortality reports estimated thousands of sea
otters and sea birds died, but listed fewer than 10 dead ﬁsh
from the event (Paine et al. 1996). Thus ﬁsh may be inherently
less susceptible to oil spills, but they are by no means immune.
452 R.S. Steneck et al.
Given the likelihood that both non-point source and
point-source pollution are likely to increase with increased
human population growth, water quality is expected to
continue to decline. We do not yet know if thresholds of
accelerated mortality exist for declining water quality as they
apparently do for increased ﬁshing pressure. However, it is
currently top-down impacts on ecosystem drivers (such as
sea urchins and their predators) that most consistently
denude kelp forests.
CONCLUSIONS AND MANAGEMENT
Kelps are the largest bottom-dwelling organisms to occupy
the euphotic zone. Their size and photosynthesis to biomass
ratio constrain their distribution globally and locally. Kelps
are among the shallowest of the subtidal macrophytes. Kelp
forests fail to develop at high latitudes due to light limitations
and at low latitudes due to limitations in nutrients, high sea
temperatures and competition from other macrophytes. In
shallow mid-latitude rocky marine shores worldwide, phylet-
ically diverse, structurally complex and highly productive
kelp forests develop. These are uniquely capable of altering
local oceanography and ecology by dampening wave surge,
shading the sea ﬂoor with their canopy, providing a physical
habitat for organisms above the benthic boundary-layer and
by distributing trophic resources to surrounding habitats. In
this context, the three kelp forest case studies from North
America represent ecosystems along a continuum of natural
biodiversity and human interactions. In each system we
reviewed archaeological literature in an attempt to recon-
struct an ecological baseline for the structure and function of
kelp forests prior to contact with modern humans.
Consumer animals structure kelp forest interactions via
two primary ‘drivers’, namely (1) herbivory by sea urchins
and (2) carnivory from predators of sea urchins. Other
forcing functions can be important. For example, kelps are
prone to destruction and thinning by storms and competi-
tors. Further, their growth and survival are sensitive to
temperature, light and nutrient availability. However, the
spatial scale and magnitude of these impacts on kelp forests
are small relative to those of the consumers. Kelp deforesta-
tion worldwide results from sea urchin grazing, which is
controlled by predation in kelp forests where human
harvesting impacts have been minimal.
Kelp forests of the eastern Paciﬁc may have facilitated an
early coastal migration of humans into the Americas. The
concentration and high productivity of vertebrates and inver-
tebrates along this coast would have provided early human
settlers with a stable source of food between 15 000 and
10 000 years ago. Archaeological data indicate that coastal
settlements exploited organisms associated with kelp forests
for thousands of years and this occasionally resulted in the
localized loss of apex predators, outbreaks of sea urchin
populations and deforestation. However, these human
impacts on kelp bed systems were probably localized and
Over the past two centuries, the commercial exploitation
of kelp forest consumers led to the extirpation of sea urchin
predators such as the sea otter in the North Paciﬁc and
groundﬁsh such as Atlantic cod in the North Atlantic. In
those systems, sea urchin abundances increased and kelp
forests were denuded over vast stretches of coast. In the
southern California system, the high diversities of predators,
herbivores and kelp appear to have buffered this system from
Biodiversity of kelp forests may also help resist invasion of
non-native species. In the species-depauperate western
North Atlantic, introduced algal competitors carpet the
benthos and threaten the dominance of kelp. Other intro-
duced herbivores and predators have taken hold and have
increased to dominate components of the system.
Global and regional climate changes have measurable
impacts on kelp forest ecosystems. Increasing frequencies of
ENSO events, oceanographic regime shifts and violent
storms cause deforestation. This, in combination with the
serial loss in biodiversity from overﬁshing, appears to be the
greatest threat to structure and functioning of these systems
over the 2025 time horizon.
Management for the conservation of kelp forest ecosys-
tems should focus on restoring biodiversity and especially on
minimizing ﬁshing on predators. In particular, species such
as sea otters, sheephead and cod should be restored to fulﬁl
their functional role in the Alaska, California and western
North Atlantic systems, respectively. While sea otters are
already protected, other commercially valuable species such
as Atlantic cod will be unlikely to be preserved for this
ecosystem role that they perform. Ultimately, human values
and political will determine the conservation agenda.
Signiﬁcant investment in education for stakeholders, the
general public and policy makers will be necessary for this
conservation goal to succeed.
This work was conducted as part of the Long-Term
Ecological Records of Marine Environments, Populations
and Communities Working Group developed by Jeremy B.
C. Jackson and supported by the National Center for
Ecological Analysis and Synthesis (funded by NSF grant
DEB–0072909, the University of California, and the
University of California, Santa Barbara). Additional support
was provided by The Pew Foundation for Marine
Conservation (to RSS, JAE and MJT), NOAA’s Sea Grant
program to the University of Maine (to RSS), a University of
California faculty fellowship (to MHG) and the National
Undersea Research Program’s National Research Center at
the University of Connecticut at Avery Point (Grant No.
NA46RU0146 and UCAZP 94–121 to RSS). Colleagues
shared insights and unpublished data for several kelp forest
ecosystems worldwide. In particular we thank, Neil Andrew
(New South Wales, Australia), Russ Babcock (New Zealand),
George Branch (South Africa), Bob Cowen (California), Ken
Kelp forest ecosystems 453
454 R.S. Steneck et al.
Dunton (Arctic Alaska), Daisuke Fujita ( Japan), Craig
Johnson (Tasmania, Australia), Pato Ojeda (Chile), Bob
Scheibling (Nova Scotia), Peter Steinberg (New South
Wales, Australia) and Julio Vasquez (Chile). Rodrigo
Bustamante, Geoff Jones, Paul Dayton, Nicholas Polunin
and an anonymous reviewer provided terriﬁc guidance for
revising this paper. Amanda Leland, John Vavrinec and
Chantale Bégin critiqued the manuscript. Sarah Maki
provided technical assistance. To all we are grateful.
Abbott, I.A. & Hollenberg, G.J. (1976) Marine Algae of California.
Stanford, CA, USA: Stanford University Press.
Adey, W.H. (1964) The genus Phymatolithon in the Gulf of Maine.
Hydrobiologia 24: 377–420.
Adey, W.H. & Steneck, R.S. (2001) Thermogeography over time
creates biogeographic regions: a temperature/space/time-inte-
grated model and an abundance-weighted test for benthic marine
algae. Journal of Phycology 37: 677– 698.
Anderson, R.J., Carrick, P., Levitt, G.J. & Share, A. (1997)
Holdfasts of adult kelp Ecklonia maxima provide refuges from
grazing for recruitment of juvenile kelps. Marine Ecology Progress
Series 159: 265–273.
Andrew, N.L. (1993) Spatial heterogeneity, sea urchin grazing, and
habitat structure on reefs in temperate Australia. Ecology 74:
Andrew, N.L. (1994) Survival of kelp adjacent to areas grazed by sea
urchins in New South Wales, Australia. Australian Journal of
Ecology 19: 466–472.
Andrew, N.L. & Choat, J.H. (1982) The inﬂuence of predation and
nonspeciﬁc adults on the abundance of juvenile Evechinus
chloroticus (Echinodea: Echinodermata). Oecologia 54: 80–87.
Andrew, N.L. & O’Neill, A.L. (2000) Large-scale patterns in
habitat structure on subtidal rocky reefs in New South Wales.
Marine and Freshwater Research 51: 255–263.
Andrew, N.L. & Underwood, A.J. (1993) Density-dependent
foraging in the sea urchin Centrostephanus rodgersii on shallow
subtidal reefs in New South Wales, Australia. Marine Ecology
Progress Series 99: 89–98.
Babcock , R.C, Kelly, S., Shears, N.T., Walker, J.W. & Willis, T.J.
(1999) Changes in community structure in temperate marine
reserves. Marine Ecology Progress Series 189: 125– 134.
Barrales, H.L. & Lobban, C. (1975) The comparative ecology of
Macrocystis pyrifera with emphasis on the forests of Chubut
Argentina. Ecology 63: 657– 677.
Bernstein, B.B. & Jung, N. (1980) Selective pressures and coevolu-
tion in a kelp canopy community in Southern California.
Ecological Monographs 49: 335–355.
Bernstein, B.B., & Welsford (1982) An assessment of feasibility of
using high-calcium quicklime as an experimental tool for research
into kelp bed/sea urchin ecosystems in Nova Scotia. Canadian
Technical Report of Fisheries and Aquatic Sciences 968: 1–51.
Bologna, P.A.X. & Steneck, R.S. (1993) Kelp beds as habitat for
American lobster Homarus americanus. Marine Ecology Progress
Bolton, J.J. (1996) Patterns of species diversity and endemism in
comparable temperate brown algal ﬂoras. Hydrobiologia 326/327:
Bolton, J.J. & Anderson, R.J. (1987) Temperature tolerances of two
southern African Ecklonia species (Alariaceae: Laminariales) and
of hybrids between them. Marine Biology 96: 293–297.
Bourque, B.J. (1995) Diversity And Complexity In Prehistoric
Maritime Societies: A Gulf Of Maine Perspective. New York, USA:
Bourque, B.J. (2001) Twelve Thousand Years: American Indians In
Maine. Lincoln, Nebraska, USA: University of Nebraska Press.
Breen, P.A. & Mann, K.H. (1976) Changing lobster abundance and the
destruction of kelp beds by sea urchins.Marine Biology 34:137–142.
Bustamante, R.H. & Branch, G.M. (1996) The dependence of inter-
tidal consumers on kelp-derived organic matter on the west coast
of South Africa. Journal of Experimental Marine Biology and
Ecology 196: 1–28.
Bustamante, R.H., Branch, G.M. & Eekhout, S. (1995)
Maintenance of an exceptional intertidal grazer biomass in South
Africa: subsidy by subtidal kelps. Ecology 76: 2314–2329.
Bustnes, J.O., Lonne, O.J., Skjoldal, H.R., Hopkins, C., Erikstad,
K.E. & Leinass, H.P. (1995) Sea ducks as predators on sea urchins
in a northern kelp forest. In: Ecology of Fjords and Coastal Waters,
ed. H.R. Skjoldal, C. Hopkins & K.E. Erikstad, pp. 599–608.
Amsterdam, the Netherlands: Elsevier Science.
Castilla, J.C. & Moreno, C.A. (1982) Sea urchins and Macrocystis
pyrifera: experimental test of their ecological relations in southern
Chile. Proceedings of the International Echinoderm Conference,
Tampa Bay, Florida, USA, ed. J.M. Lawrence, pp. 257–263.
Rotterdam, the Netherlands: Balkema.
Chapman, A.R.O. (1986) Age versus stage: an analysis of age- and
size-speciﬁc mortality and reproduction in a population of
Laminaria longicruris Pyl. Journal of Experimental Marine Biology
and Ecology 97: 113–122.
Chapman, A.R.O. & Craigie, J.S. (1977) Seasonal growth in
Laminaria longicruris: relations with dissolved inorganic nutrients
and internal reserves of nitrogen. Marine Biology 40: 197–205.
Chapman, A.R.O. & Johnson, C.R. (1990) Disturbance and organ-
ization of macroalgal assemblages in the Northwestern Atlantic.
Hydrobiologia 192: 77–121.
Choat, J.H. & Ayling, A.M. (1987) The relationship between habitat
and ﬁsh faunas of New Zealand reefs. Journal of Experimental
Marine Biology and Ecology 110: 228–284.
Choat, J.H. & Schiel, D.R. (1982) Patterns of distribution and abun-
dance of large brown algae and invertebrate herbivores in subtidal
regions of northern New Zealand. Journal of Experimental Marine
Biology and Ecology 60: 129–162.
Clementz, M. (2002) The evolution of herbivorous marine
mammals: ecological and physiological transitions during the
evolution of the orders Sirenia and Desmostylia. Ph.D. disserta-
tion, Department of Earth Sciences, University of California at
Santa Cruz, USA.
Cole, R.G. & Babcock, R.C. (1996) Mass mortality of a dominant
kelp (Laminariales) at Goat Island, North-eastern New Zealand.
Marine Freshwater Resource 47: 907–911.
Cole, R.G. & Syms, C. (1999) Using spatial pattern analysis to
distinguish causes of mortality: an example from kelp in north-
eastern New Zealand. Journal of Ecology 87: 963–972,
Conkling, P.W. & Ames, T. (1996) Penobscot ﬁsheries in the 20th
century. In: Penobscot: the Forest River and Bay, ed. D.D. Platt,
pp. 46–65. Rockland, ME, USA: Island Institute.
Cowen, R.K. (1983) The effect of sheephead (Semicossyphus pulcher)
predation on red sea urchin populations: an experimental analysis.
Oecologia 58: 249–255.
Cox, K.W. (1962) California abalones, family Haliotidae. California
Department of Fish and Game, Fishery Bulletin 118, California,
USA: 133 pp.
Darwin, C. (1909) The Voyage of the Beagle. The Harvard Classics
Volume 29. New York, USA: P.F. Collier & Son Company.
Day, E. & Branch, G.M. (2002) Effects of sea urchins (Parechinus
angulosus) on recruits and juveniles of abalone (Haliotis midae).
Ecological Monographs 72: 133–149.
Dayton, P.K. (1975) Experimental studies of algal-canopy interac-
tions in a sea otter dominated kelp community at Amchitka
Island, Alaska. Fishery Bulletin, United States 73: 230–237.
Dayton, P.K. (1985a) Ecology of kelp communities. Annual Review
Ecology Systems 16: 215–245.
Dayton, P.K. (1985b) The structure and regulation of some South
American kelp communities. Ecological Monographs 55: 447–468.
Dayton, P.K. & Tegner, M.J. (1990) Bottoms below troubled
waters: benthic impacts of the 1982–84 El Niño in the temperate
zone. In: Ecological Consequences of the 1982–83 El Niño to Marine
Life, ed. P.W. Glynn, Elsevier Oceanography Series No. 52, pp.
433–472. Amsterdam, the Netherlands: Elsevier: 563 pp.
Dayton, P.K., Currie, V., Gerrodete, T., Keller, B.D., Rosenthal,
R., & Ven Tresca, D. (1984) Patch dynamics and stability of some
California kelp communities. Ecological Monographs 54: 253–289.
Dayton, P.K., Tegner, M.J. Edwards, P.B. & Riser, K.L. (1998)
Sliding baselines, ghosts, and reduced expectations in kelp forest
communities. Ecological Applications 8: 309– 322.
Dayton, P.K., Tegner, M.J., Edwards, P.B. & Riser, K.L. (1999)
Temporal and spatial scales of kelp demography: the role of
oceanography climate. Ecological Monographs 69: 219–250.
Dean, T.A. & Jewett, S.C. (2001) Habitat-speciﬁc recovery of
shallow subtidal communities following the Exxon Valdez oil
spill. Ecological Applications 11: 1456– 1471.
Domning, D.P. (1972) Steller’s sea cow and the origin of North
Paciﬁc aboriginal whaling. Syesis 5: 187–189.
Duggins, D.O. (1980) Kelp beds and sea otters: an experimental
approach. Ecology 61: 447–453.
Duggins, D.O. (1983) Starﬁsh predation and the creation of mosaic
patterns in a kelp dominated community. Ecology 64: 1610–
Duggins, D.O., Eckman, J.E. & Sewell, A.T. (1990) Ecology of
understory kelp environments. II. Effects of kelps on recruitment
of benthic invertebrates. Journal of Experimental Marine Biology
and Ecology 143: 27–45.
Duggins, D.O., Simenstad, C.A. & Estes, J.A. (1989) Magniﬁcation
of secondary production by kelp detritus in coastal marine ecosys-
tems. Science 245: 101– 232.
Dunton, K.H. (1990) Growth and production in Laminaria solidun-
gula: relation to continuous underwater light levels in the Alaskan
High Arctic. Marine Biology 106: 297–304.
Dunton, K.H. & Dayton, P.K. (1995) The biology of high latitude
kelp. In: Ecology of Fjords and Coastal Waters, ed. H.R. Skjoldal,
C. Hopkins, K.E. Erikstad & H.P. Leinass, pp. 499–507.
Amsterdam, the Netherlands: Elsevier Science.
Dunton, K.H. & Schell, D.M. (1987) Dependence of consumers on
macroalgal (Laminaria solidungula) carbon in an Arctic kelp
community: delta-carbon-13 evidence. Marine Biology 93:
Durham, J.W. & MacNeil, F.S. (1967) Cenozoic migrations of
marine invertebrates through the Bering Strait region. In: The
Bering Land Bridge, ed. D.M. Hopkins, pp. 326–349. Stanford,
CA, USA: Stanford University Press.
Ebeling, A.W., Laur, D.R. & Rowley, R.J. (1985) Severe storm
disturbance and reversal of community structure in a southern
California kelp forest. Marine Biology 84: 287–294.
Ebling, F.S., Hawkins, A.D., Kitching, J.A., Muntz, P. & Pratt,
V.M. (1966) The ecology of Lough Ine XVI. Predation and
diurnal migrations in Paracentrotus community. Journal of Animal
Ecology 35: 559–566.
Edelstein, T., Craigie, J.S. & McLachlan, J. (1969) Preliminary
survey of the sublittoral ﬂora of Halifax county. Journal of
Fisheries Research Board of Canada 26: 2703–2713.
Enﬁeld, D.B. (1988) Is El Niño becoming more common?
Oceanography 59: 123–127.
Erlandson, J.M. (2001) Anatomically modern humans, maritime
voyaging, and the Pleistocene colonization of the Americas. In:
The First Americans: The Pleistocene Colonization of the New
World, ed. N.G. Jablonski, pp. 1–9. San Francisco, CA, USA:
California Academy of Sciences.
Erlandson, J.M. & Rick, T.C. (2002) Human impacts on ancient
environments: a case study from California’s Northern Channel
Islands. In: The Archaeology of Insularity: Examining the Past in
Island Environments, ed. S.M. Fitzpatrick (in press). New York,
Erlandson, J.M., Kennett, D.J., Ingram, B.L., Guthrie, D.A.,
Morris, D.P., Tveskov, M.A., West, G.J. & Walker, P.L. (1996)
An archaeological and palaeontological chronology for Daisy
Cove (CA-SMI-261), San Miguel Island, California. Radiocarbon
Estes, J.A. & Duggins, D.O. (1995) Sea otters and kelp forests in
Alaska: generality and variation in a community ecological para-
digm. Ecological Monographs 65: 75–100.
Estes, J.A. & Steinberg, P.D. (1988) Predation, herbivory, and kelp
evolution. Paleobiology 14: 19– 36.
Estes, J.A., Duggins, D.O. & Rathbun, G.B. (1989) The ecology of
extinctions in kelp forest communities. Conservation Biology 3:
Estes, J.A., Tinker, M.T., Williams, T.M. & Doak, D.F. (1998)
Killer whale predation on sea otters linking oceanic and nearshore
ecosystems. Science 282: 473– 476.
Foster, M.S. (1990) Organization of macroalgal assemblages in the
Northeast Paciﬁc: the assumption of homogeneity and the illusion
of generality. Hydrobiologia 192: 21–33.
Foster, M. & Schiel, D.R. (1985) The ecology of giant kelp forests
in California: a community proﬁle. US Fish and Wildlife Service
Biological Report 85 (7.2), USA: 152 pp.
Fujita, D. (1987) The report of interviews with ﬁshermen
on Isoyake in Taisei-cho, Hokkaido. Suisanzoshoku 35: 135–
Fujita, D. (1998) Strongylocentrotid sea urchin-dominated barren
grounds on the Sea of Japan coast of northern Japan. In:
Echinoderms, ed. R. Mooi & M. Telford, pp. 659–664. Rotterdam,
the Netherlands: A.A. Balkema.
Gagne, J.A., Mann, K.H. & Chapman, A.R.O. (1982) Seasonal
patterns of growth and storage in Laminaria longicruris in relation
to differing patterns of availability of nitrogen in the water.
Marine Biology 69: 91–101.
Gaines, S.D. & Roughgarden, J. (1987) Fish in offshore kelp forests
affect recruitment to intertidal barnacle populations. Science 235:
Gerard, V.A. (1982) Growth and utilization of internal nitrogen
reserves by the giant kelp Macrocystis pyrifera in a low nitrogen
environment. Marine Biology 66: 27– 35.
Gerard, V.A. (1997) The role of nitrogen nutrition in high-tempera-
Kelp forest ecosystems 455
ture tolerance of kelp, Laminaria saccharina. Journal of Phycology
Graham, M.H. (1996) Effect of high irradiance on recruitment of
the giant kelp Macrocystis (Phaeophyta) in shallow water. Journal
of Phycology 32: 903–906.
Graham, M.H., Dayton, P.K. & Erlandson, J.M. (2003) Ice ages
and the shifting ecology of temperate coasts. Trends in Ecology and
Evolution (in press).
Graham, M.H., Harrold, C., Lisin, S., Light, K. Watanabe, J.M. &
Foster, M.S. (1997) Population dynamics of giant kelp
Macrocystis pyrifera along a wave exposure gradient. Marine
Ecology Progress Series 148: 269–279.
Graham, N.E. & Diaz, H.F. (2001) Evidence for intensiﬁcation of
North Paciﬁc winter cyclones since 1948. Bulletin of American
Meteorological Society 82: 1869–1893.
Hagen, N.T. (1983) Destructive grazing of kelp beds by sea urchins
in Vestfjorden, northern Norway. Sarsia 68: 177– 190.
Harris, L.G., Ebeling, A.W., Laur, D.R. & Rawley, R.J. (1984)
Community recovery following storm damage: a case of facilita-
tion in primary succession. Science 224: 1336–1338.
Harrold, C. & Lisin, S. (1989) Radio-tracking rafts of giant kelp:
local production and regional transport. Journal of Experimental
Marine Biology and Ecology 30: 237–252.
Harrold, C. & Reed, D.C. (1985) Food availability, sea urchin
grazing, and kelp forest community structure. Ecology 66:
Harrold, C., Light, K. & Lisin, S. (1998) Organic enrichment of
submarine-canyon and continental-shelf benthic communities by
macroalgal drift imported from nearshore kelp forests. Limnology
and Oceanography 43: 669–678.
Hart, M.W. & Scheibling, R.E. (1988) Heat waves, baby booms, and
the destruction of kelp beds by sea urchins. Marine Biology 99:
Hatcher, B.G., Kirkman, H. & Wood, W. F. (1987) Growth of the
kelp Ecklonia radiata near the northern limit of its range in
Western Australia. Marine Biology 95: 63–73.
Henley, W.J. & Dunton, K.H. (1997) Effects of nitrogen supply and
continuous darkness on growth and photosynthesis of the arctic
kelp Laminaria solidungula. Limnology and Oceanography 42:
Hervey, A.B. (1881) Sea Mosses: a Collector’s Guide and an
Introduction to the Study of the Marine Algae. Salem, MA, USA:
Himmelman, J.H. (1980) The role of the green sea urchin
Strongylocentrotus droebachiensis in the rocky subtidal region of
Newfoundland. In: Proceedings Of The Workshop On The
Relationship Between Sea Urchin Grazing And Commercial
Plant/Animal Harvesting, ed. J.D. Pringle, G.J. Sharp & J.F.
Caddy, pp. 92–119. Halifax, NS, Canada: Fisheries and Oceans
Canada, Resource Branch.
Hjorleifsson, E., Kassa, O. & Gunnarsson, K., (1995) Grazing of
kelp by green sea urchins in Eyyjafjordu, North Iceland. In:
Ecology of Fjords and Coastal Waters, ed. H.R. Skjoldal, C.
Hopkins, K.K. Erikstad & H.P. Leinass, pp. 593–597.
Amsterdam, the Netherlands: Elsevier Science.
Hobday, A.J. (2000) Abundance and dispersal of drifting kelp
Macrocystis pyrifera rafts in the Southern California Bight. Marine
Ecology Progress Series 195: 101–116.
Hooper, R. (1980) Observations on algal-grazer interactions in
Newfoundland and Labrador. In: Proceedings Of The Workshop
On The Relationship Between Sea Urchin Grazing And Commercial
Plant/Animal Harvesting, ed. J.D. Pringle, G.J. Sharp & J.F.
Caddy, pp. 120–124. Halifax, NS, Canada: Fisheries and Oceans
Canada, Resource Branch.
Houghton, J.T., Meira Filho, L.G., Bruce, J., Hoesung Lee
Callander, B.A., Haites, E. & Maskell, K. (1996) Climate Change
1995. World Meteorological Organization/Intergovernmental
Panel on Climate Change, Cambridge, UK: Cambridge
Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford,
L.W., Bourque, B.J., Bradbury, R., Cooke, R., Erlandson, J.,
Estes, J.A., Hughes, T.P., Kidwell, S., Lange, C.B., Lenihan,
H.S., Pandolﬁ, J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J.
& Warner, R. (2001) Historical overﬁshing and the recent collapse
of coastal ecosystems. Science 293: 629–638.
Jickells, T.D. (1998) Nutrient biogeochemistry of the coastal zone.
Science 281: 217–222.
Johnson, C.R. & Mann, K.H. (1986) The importance of plant
defense abilities to the structure of subtidal seaweed communities:
the kelp Laminaria longicruris de la Pylaie survives grazing by the
snail Lacuna vincta (Montagu) at high populations densities.
Journal of Experimental Marine Biology and Ecology 97: 231–267.
Johnson, C.R. & Mann, K.H. (1988) Diversity, patterns of adap-
tation, and stability of Nova Scotian kelp beds. Ecological
Monographs 58: 129–154.
Johnson, D.S. & Skutch, A.F. (1928) Littoral vegetation on a head-
land of Mt. Desert Island, Maine. I. Submersible or strictly
littoral vegetation. Ecology 9: 188–215.
Jones, N.S. & Kain, J.M. (1967) Subtidal algal colonization
following the removal of Echinus. Helgolander Wissenshaftlichen
Meeresuntersuchungen 15: 160–166.
Jones, R.E. (1967) A Hydrodamalis skull fragment from Monterey
Bay, California. Journal of Mammalogy 48: 143–144.
Kain, J.M. (1975) Algal recolonization of some cleared subtidal
areas. Journal of Ecology 63: 739– 765.
Kain, J.M. (1979) A view of the genus Laminaria. Oceanographic
Marine Biology Annual Review 17: 101–161.
Keats, D.W., Steele, D.H. & South, G.R. (1986) Atlantic wolfﬁsh
(Anarhichas lupus L.; Pisces: Anarhichidae) predation on green sea
urchins (Strongylocentrotus droebachiensis (O. F. Mull.):
Echinodermata: Echinoidea) in eastern Newfoundland. Canadian
Journal of Zoology 64: 1920–1925.
Kitching, J.A. & Ebling, F.J. (1961) The ecology of Lough Ine XI.
The control of algae by Paracentrotus lividus (Echinoidea). Journal
of Animal Ecology 30: 373–383.
Kitching, J.A. & Thain, V.M. (1983) The ecological impact of the
sea urchin Paracentrotus lividus (Lamarck) in Lough Ine, Ireland.
Philosophical Transactions of the Royal Society of London B 300:
Korb, R.E. & Gerard, V.A. (2000) Effects of concurrent low
temperature and low nitrogen supply on polar and temperate
seaweeds. Marine Ecology Progress Series 198: 73– 82.
Kurlansky, M. (1997) Cod: a Biography of the Fish that Changed the
World. New York, USA: Walker and Co.
Lamb, M. & Zimmerman, M.H. (1964) Marine vegetation of Cape
Ann, Essex County, Massachusetts. Rhodora 66: 217–254.
Lambert, W.J., Levin, P.S. & Berman, J. (1992) Changes in the
structure of a New England (USA) kelp bed: the effects of an
introduced species? Marine Ecology Progress Series 88: 303–307.
Lawrence, J.M. (1975) On the relationships between marine plants
and sea urchins. Oceanography and Marine Biology Annual Review
456 R.S. Steneck et al.
Leighton, D.L., Jones, L.G. & North, W. (1966) Ecological
relationships between giant kelp and sea urchins in southern
California. In: Proceedings 5th International Seaweed Symposium
Oxford, England, ed. E.G. Young & J.L. McLachlan, pp.
141–153. Oxford, UK: Pergamon Press.
Leland, A. (2002) A new apex predator in the Gulf of Maine: large,
mobile Jonah crabs (Cancer borealis) control benthic community
structure. Masters Thesis, University of Maine. Orono Maine,
Leland, A. & R. S. Steneck. (2001) Are crabs in Maine a new apex
predator? Abstracts of the 82nd Annual Meeting of the Western
Society of Naturalists, Monterey, CA, USA.
Levin, P. (1991) Effects of microhabitat on recruitment variation in
a Gulf of Maine reef ﬁsh. Marine Ecology Progress Series 75:
Levin, P.S. (1994) Fine-scale temporal variation in recruitment of a
temperate demersal ﬁsh: the importance of settlement versus
post-settlement loss. Oecologia 97: 124–133.
Levin, P.S., Coyer, J.A., Petrik, R. & Good, T.P. (2003)
Community-wide effects of non-indigenous species on temperate
rocky reefs. Ecology (in press).
Linley, E.A.S., Newell, R.C. & Bosma, S.A. (1981) Hetertrophic
utilization of mucilage released during fragmentation of kelp
(Ecklonia maxima and Laminaria pallida) development of micro-
bial communities associated with the degradation of kelp
mucilage. Marine Ecology Progress Series 4: 31– 41.
MacFarlane, C. (1952) A survey of certain seaweeds of commercial
importance in southwest Nova Scotia. Canadian Journal of Botany
Mann, K.H. (1973) Seaweeds: their productivity and strategy for
growth. Science 182: 975– 981.
Mann, K.H. (1977) Destruction of kelp beds by sea urchins: a
cyclical phenomenon or irreversible degradation? Helgolander
Wissenshaftlichen Meeresuntersuchungen 30: 455–467.
Mann, K.H. (2000) Ecology of Coastal Waters, with Implications for
Management. Volume 2. Oxford, UK: Blackwell Science:
Mann, M.E., Bradley, R.S. & Hughes, M.K. (1998) Global-scale
temperature patterns and climate forcing over the past six
centuries. Nature 392: 779– 787.
Matsunaga, K., Kawaguchi, T., Suzuki Y. & Nigi, G. (1999) The
role of terrestrial humic substances on the shift of kelp
community to crustose coralline algae community of the southern
Hokkaido Island in the Japan Sea. Journal of Experimental Marine
Biology and Ecology 241: 193–205.
McGowan, J.A., Cayan, D.R., Dorman, L.M. (1998) Climate-ocean
variability and ecosystem response in the Northeast Paciﬁc.
Science 281: 210–217.
McNaught, D.C. (1999) The indirect effects of macroalgae and
micropredation on the post-settlement success of the green sea
urchin in Maine. Ph. D. Dissertation, Department of
Oceanography, University of Maine, USA.
McNeill, W.H. (1976) Plagues and Peoples. New York, USA:
Anchor Books: 340 pp.
Moe, R. & Silva, P. (1981) Morphology and taxonomy of
Himantothallus (including Phaeoglossum and Phyllogigas), an
Antarctic member of the Desmarestiales (Phaeophyceae). Journal
of Phycology 17: 15–29.
Monson, D.H., Doak, D.F., Ballachey, B.E., Johnson, A.M. &
Bodkin, J.L. (2000) Long-term impacts of the Exxon Valdez oil
spill on sea otters, assessing through age-dependent mortality
patterns. Proceedings of the National Academy of Sciences 97:
Neushul, M. (1972) Functional interpretation of benthic marine
algal morphology. Contributions to the systematics of benthic
marine algae of the North Paciﬁc, ed. I.A. Abbott & M. Kurogi,
pp 47–74. Kobe, Japan: Japanese Society of Phycology
Novaczek, I. & McLachlan, J. (1986) Recolonization by algae of the
sublittoral habitat of Halifax County, Nova Scotia, following the
demise of sea urchins. Botanica Marina 24: 69–73.
Ohmi, H. (1951) Studies on Isoyake or decrease of seaweeds along
the coast of Northern Japan. Bulletin Faculty Fisheries Hakodate,
Hokkaido University 2: 109–117.
Ojeda, F. & Santelices, B. (1984) Ecological dominance of Lessonia
nigrescens (Phaeophyta) in central Chile. Marine Ecology Progress
Series 19: 83–91.
Paine, R. T. (1980) Food webs: linkages, interaction strength and
community infrastructure. Journal of Animal Ecology 49:
Paine, R.T., Ruesink, J.L., Sun, A., Soulanille, E.L., Wonham,
M.J., Harley, C.D.G., Brumbaugh, D.R. & Secord, D.L. (1996)
Trouble on oiled waters: lessons from the Exxon Valdez oil spill.
Annual Review Ecology Systematics 27: 197–235.
Paine, R.T., Tegner, M.J. & Johnson, E.A. (1998) Compounded
perturbations yield ecological surprises. Ecosystems 1: 535–545.
Pearse, J.S. & Hines, A.H. (1979) Expansion of a central California
kelp forest following the mass mortality of sea urchins. Marine
Biology 51: 83–91.
Peters, A.F. & Breeman, A.M. (1993) Temperature tolerances and
latitudinal range of brown algae from temperate Paciﬁc South
America. Marine Biology 115: 143– 150.
Polis, G.A. & Hurd, S.D. (1996) Linking marine and terrestrial food
webs: allochthonous inputs from the ocean supports high
secondary productivity on small islands and coastal land
communities. American Naturalist 147: 396–523.
Pringle, J.D., Sharp, G.J. & Caddy, J.F., eds. (1980) Proceedings of
the Workshop on the Relationship Between Sea Urchin Grazing and
Commercial Plant/Animal Harvesting. Canadian Technical Report
of Fisheries and Aquatic Sciences, 954: 273 pp.
Raffaelli, D. & Hawkins, S. (1996) Intertidal Ecology. London, UK:
Chapman & Hall.
Reed, D.C. & Foster, M.S. (1984) The effects of canopy shading on
algal recruitment and growth in a giant kelp forest. Ecology 65:
Rich, W.H. (1929) Fishing grounds of the Gulf of Maine. Bureau of
Fisheries Documentary 1959: 51–117.
Rick, T.C., Erlandson, J.M. & Vellanoweth, R. (2001) Paleocoastal
marine ﬁshing on the Paciﬁc coast of the Americas: perspectives
from Daisy Cave, California. American Antiquity 66: 593–613.
Rosier, J. (1605) True Relations of Captain George Weymouth’s
Voyage. London, UK: George Bishop.
Sala, E., Boudouresque, C.F. & Harmelin-Viven, M. (1998)
Fishing, trophic cascades, and the structure of algal assemblages:
evaluation of an old but untested paradigm. Oikos 82: 425–439.
Salls, R.A. (1991) Early Holocene maritime adaptation at Eel Point,
San Clemente Island. In: Hunter-Gathers Of Early Holocene
Coastal California, ed. J.M. Erlandson & R.H. Colton, pp. 63–80.
Perspectives in California Archaeology, Volume I. Los Angeles,
USA: Institute of Archaeology, University of California.
Salls, R.A. (1995) Ten-thousand years of ﬁshing; the evidence for
alternate stable states in nearshore ecosystems as the result of
overexploitation of the California sheephead (Semiscossyphus
Kelp forest ecosystems 457
pulcher) by prehistoric ﬁshermen on San Clemente Island,
California. In: Memorias IX Simposium Internacional Biologia
Marina ( June 1992), pp. 205–214. Universidad Autonoma Baja
California Sur, La Paz, Baja California Sur, Mexico.
Santelices, B. (1990) Patterns of organizations of intertidal and
shallow subtidal vegetation in wave exposed habitats of central
Chile. Hydrobiologia 192: 35– 57.
Santelices, B. & Ojeda, F.P. (1984a) Effects of canopy removal on
the understory algal community structure of coastal forests of
Macrocystis pyrifera from southern South America. Marine
Ecology Progress Series 14: 165–173.
Santelices, B. & Ojeda, F.P. (1984b) Population dynamics of coastal
forests of Macrocystis pyrifera in Puerto Toro, Isla Navarino,
Southern Chile. Marine Ecology Progress Series 14: 175–183.
Scheffer, M., Carpenter, S., Foley, J.A., Folke, C. & Walter, B.
(2001) Catastrophic shifts in ecosystems. Nature 413: 591–596.
Scheibling, R.E. (1984) Echinoids, epizootics, and ecological stab-
ility in the rocky subtidal off Nova Scotia, Canada. Helgolander
Wissenshaftlichen Meeresuntersuchungen 37: 233–242.
Scheibling, R.E. (1986) Increased macroalgal abundance following
mass mortalities of sea urchins (Strongylocentrotus droebachiensis)
along the Atlantic coast of Nova Scotia. Oecologia 68: 186–198.
Scheibling, R.E. & Stephenson, R.L. (1984) Mass mortality of
Strongylocentrotus droebachiensis (Echindermata: Echinoidea) off
Nova Scotia, Canada. Marine Biology 78: 153–164.
Scheibling, R.E., Hennigar, A.W. & Balch, T. (1999) Destructive
grazing, epiphytism, and disease: the dynamics of sea urchin-kelp
interactions in Nova Scotia. Canadian Journal of Fisheries and
Aquatic Sciences 56: 1–15.
Schiel, D.R. (1990) Macroalgal assemblages in New Zealand: struc-
ture, interactions and demography. Hydrobiologia 192: 59–76.
Schiel, D.R. & Foster, M.S. (1986) The structure of subtidal algal
stands in temperate waters. Oceanographic Marine Biology Annual
Review 24: 265–307.
Schiel, D.R., Andrew, N.L. & Foster, M.S (1995) The structure of
subtidal algal and invertebrate assemblages at the Chatham
Islands, New Zealand. Marine Biology 123: 355–367.
Schmitt, R.J. (1987) Indirect interactions between prey: apparent
competition, predation, predator aggregation, and habitat segre-
gation. Ecology 68: 1887– 1897.
Seymour, R.J., Tegner, M.J., Dayton, P.K. & Parnell, P.E. (1989)
Storm wave induced mortality of giant kelp, Macrocystis pyrifera,
in Southern California. Estuarine, Coastal and Shelf Science 28:
Shepherd, S.A. & Womersley, H.B.S. (1970) The sublittoral
ecology of West Island, South Australia. I. Environmental
features and the algal ecology. Transactions of the Royal Society of
South Australia 94: 105–138.
Simenstad, C.A., Estes, J.A. & Kenyon, K.W. (1978) Aleuts, sea
otters, and alternate stable-state communities. Science 200:
Sivertsen, K. (1997). Geographic and environmental factors
affecting the distribution of kelp beds and barren grounds and
changes in biota associated with kelp reduction at sites along the
Norwegian coast. Canadian Journal of Fisheries and Aquatic
Sciences 54(12): 2872– 2887.
Soares, A.G., Schlacher, T.A. & McLachlan, A. (1997) Carbon and
nitrogen exchange between sandy beach clams (Donax serra) and
kelp beds in the Benguela coastal upwelling region. Marine
Biology 127: 657–664.
Stachowicz, J.J. Whitlatch, R.B. & Osman, R.W. (1999) Species
diversity and invasion resistance in a marine ecosystem. Science
Steele, J.H. (1998) Regime shifts in marine ecosystems. Ecological
Applications 8: S33–S36.
Steneck, R.S. (1997) Fisheries-induced biological changes to the
structure and function of the Gulf of Maine ecosystem. In:
Proceedings of the Gulf of Maine Ecosystem Dynamics Scientiﬁc
Symposium and Workshop, RARGOM Report 91-1, pp. 151–165.
Hanover, NH, USA: Regional Association for Research in the
Gulf of Maine.
Steneck, R.S. (1998) Human inﬂuences on coastal ecosystems: does
overﬁshing create trophic cascades? Trends in Ecology and
Evolution 13: 429–430.
Steneck, R.S. & Carlton, J.T. (2001) Human alterations of marine
communities: students beware! In: Marine Community Ecology,
ed. M. Bertness, S. Gaines & M. Hay, pp. 445–468. Sunderland,
MA, USA: Sinauer Press.
Steneck, R.S. & Dethier, M.N. (1994) A functional group approach
to the structure of algal-dominated communities. Oikos 69:
Steneck, R.S. & Watling, L.E. (1982) Feeding capabilities and limi-
tations of herbivorous molluscs: a functional group approach.
Marine Biology 68: 299–319.
Suzuki, T., Kuma, K., Kudo, I. & Matsunaga, K. (1995) Iron
requirement of the brown macroalgae Laminaria japonica,
Undaria pinnatiﬁda and the crustose coralline algae, and their
competition in the northern Japan Sea. Phycologia 34: 201–205.
Tegner, M.J. & Dayton, P.K. (1987) El Niño effects on southern
California kelp forest communities. Advances in Ecological
Research 17: 243–279.
Tegner, M.J. & Dayton, P.K. (1991) Sea urchins, El Niños, and the
long-term stability of Southern California kelp forest communi-
ties. Marine Ecology Progress Series 77: 49– 63.
Tegner, M.J. & Dayton, P.K. (2000) Ecosystem effects of ﬁshing in
kelp forest communities. ICES Journal of Marine Science 57:
Tegner, M.J. & Levin, L.A. (1983) Spiny lobsters and sea urchins:
analysis of a predator-prey interaction. Journal of Experimental
Marine Biology and Ecology 73: 125–150.
Tegner, M.J., Basch, L.V. & Dayton, P.K. (1996b) Near extinction
of an exploited marine invertebrate. Trends in Ecology and
Evolution 11: 278–279.
Tegner, M.J., Dayton, P.K., Edwards, P.B. & Riser, K.L. (1996a)
Is there evidence for long-term climatic change in southern
California kelp forests? California Cooperative Oceanic Fisheries
Investigations Reports 37: 111–126.
Tegner, M.J., Dayton, P.K., Edwards, P.B. & Riser, K.L. (1997)
Large-scale, low-frequency oceanographic effects on kelp forest
succession: a tale of two cohorts. Marine Ecology Progress Series
Tegner, M.J., Dayton, P.K., Edwards, P.B., Riser, K.L., Chadwick,
D.B., Dean, T.A. & Deysher, L. (1995) Effects of large sewage
spill on a kelp forest community: catastrophe or disturbance?
Marine Environmental Research 40: 181–224.
Vadas, R.L. & Steneck, R.S. (1988) Zonation of deep water benthic
algae in the Gulf of Maine. Journal of Phycology 24: 338–346.
Vadas, R.L. & Steneck, R.S. (1995) Overﬁshing and inferences in
kelp-sea urchin interactions. In: Ecology of Fjords and Coastal
Waters, ed. H.R. Skjoldal, C. Hopkins, K.E. Erikstad & H.P.
Leinaas, pp. 509–524. Amsterdam, the Netherlands: Elsevier
458 R.S. Steneck et al.
Kelp forest ecosystems 459
Vasquez, J.A. (1993) Abundance, distributional patterns and diets of
main herbivorous and carnivorous species associated with
Lessonia trabeculata kelp beds in northern Chile. Serie Ocasional
Universidad Catolica del Norte 2: 213–229.
Vasquez, J.A. & Buschmann, A.H. (1997) Herbivore-kelp interac-
tions in Chilean subtidal communities: a review. Revista Chilena
de Historia Natural 70: 41–52.
Vasquez, J.A., Castilla, J.C. & Santelices, B. (1984) Distributional
patterns and diets of four species of sea urchins in giant kelp forest
(Macrocystis pyrifera) of Puerto Toro, Navarino Island, Chile.
Marine Ecology Progress Series 19: 55–63.
Vavrinec, J. (2003) Resilience of green sea urchin (Strongylocentrotus
droebachiensis) populations following ﬁshing mortality: marine
protected areas, larval ecology and post-settlement survival.
Ph.D. dissertation, School of Marine Sciences, University of
Velimirov, B., Field, F.G., Grifﬁths, C.L. & Zoutendyk, P. (1977)
The ecology of kelp bed communities in the Benguela upwelling
system. Helgolander Wissenshaftlichen Meeresuntersuchungen 30:
Vermeij, G.J. (2001) Community assembly in the sea: geologic
history of the living shore biota. In: Marine Community Ecology,
ed. M. Bertness, S. Gaines & M. Hay, pp. 39–60. Sunderland,
MA, USA: Sinauer Press.
Vetter, E.W. (1995) Detritus-based patches of high secondary
production in the nearshore benthos. Marine Ecology Progress
Series 120: 251–262.
Warton, W.G. & Mann, K.H. (1981) Relationship between destruc-
tive grazing by the sea urchin, Stronylocentrotus droebachiensis, and
the abundance of American lobster, Homarus americanus, on the
Atlantic coast of Nova Scotia. Canadian Journal of Fisheries and
Aquatic Sciences 38: 1339–1349.
Witman, J.D. (1987) Subtidal coexistence: storms, grazing, mutu-
alism, and the zonation of kelps and mussels. Ecological
Monographs 57: 167–187.
Witman, J.D. (1988) Stability of Atlantic kelp forests. Trends in
Ecology and Evolution 3: 285–286.
Witman, J. & Dayton, P. (2001) Rocky subtidal communities. In:
Marine Community Ecology, ed. M. Bertness, S. Gaines & M.
Hay, pp. 339–366. Sunderland, MA, USA: Sinauer Press.
Yendo, K. (1902) Kaiso Isoyake Chosa Hokoku. Suisan Chosa
Hokoku 12(1): 1–33. (In Japanese.)
Yendo, K. (1903) Investigations on ‘Isoyake’ (decrease of seaweed).
Journal of the Imperial Fisheries Bureau 12: 1–33. (In Japanese.)
Yendo, K. (1914) On the cultivation of seaweeds, with special
accounts of their ecology. Economic Proceedings of the Royal Dublin
Society 2(7): 105–122.
Zimmerman, R.C. & Kremer, J.N. (1984) Episodic nutrient supply
to a kelp forest ecosystem in Southern California. Journal of
Marine Research 42: 591–604.
Zimmerman, R.C. & Kremer, J.N. (1986) In situ growth and
chemical composition of the giant kelp, Macrocystis pyrifera:
response to temporal changes in ambient nutrient availability.
Marine Ecology Progress Series 27: 227–285.
Zobell, C.E. (1971) Drift seaweeds on San Diego county beaches.
Nova Hedwigia 32: 269–314.