Coral Reefs (2006)

DOI 10.1007/s00338-006-0122-9

P E R S PE C T I V E

Richard B. Aronson Æ William F. Precht

Conservation, precaution, and Caribbean reefs

Received: 29 November 2005 / Accepted: 24 April 2006

Ó Springer-Verlag 2006

Abstract Some authors argue that overfishing is an

important reason that reef corals have declined in recent
Introduction
decades. Their reasoning is that overfishing removes
No coral reef scientist can credibly deny that reef
herbivores, releasing macroalgae to overgrow and kill
ecosystems around the world are in jeopardy (Wilkinson
the corals. The evidence suggests, however, that global
2000, 2004). The Caribbean region has been hit partic-
climate change and emergent marine diseases make a far
ularly hard, with an average loss of 40% absolute coral
greater contribution to coral mortality, and that mac-
cover since the late 1970s (Gardner et al. 2003). Some
roalgae generally grow on the exposed skeletal surfaces
authors attribute these losses primarily to decades
of corals that are already dead. Macroalgal dominance,
or centuries of overfishing, coupled with the disease-
therefore, is an effect rather than a cause of coral mor-
induced, regional mass mortality of the echinoid Dia-
tality. Marine protected areas (MPAs), which are usu-
dema antillarum in 1983–1984 (Jackson 2001; Jackson
ally established to protect stocks of reef fish, foster
et al. 2001; Pandolfi et al. 2005). In their ‘‘herbivory
populations of herbivorous fish under at least some
scenario,’’ sharply reduced herbivory was the principal
circumstances. Increased herbivory can reduce algal
driver of coral mortality, because the loss of herbivores
cover, potentially accelerating the recovery of coral
released fleshy and filamentous macroalgae, especially
populations inside MPAs; however, establishing MPAs
the brown algae Lobophora variegata, Dictyota spp., and
will have only a limited impact on coral recovery unless
Sargassum spp., to overgrow the corals. Regional-scale
policymakers confront the accelerating negative effects
collapse of Caribbean reef ecosystems ensued. These
of the global-scale sources of coral mortality.
authors contend that coral assemblages of the Carib-
bean have lost their resilience—their capacity to recover
Keywords Coral reef Æ Coral disease Æ
following perturbation. Macroalgal dominance thus
Global climate change Æ Marine protected area Æ
constitutes an alternative community state, which may
Overfishing Æ Precautionary principle
or may not be stable (Knowlton 1992; Scheffer et al.
2001; Petraitis and Dudgeon 2004). Many Indo-Pacific
reefs are now losing their resilience as well, also
Communicated by Ecology Editor P.J. Mumby purportedly due to overfishing (e.g., Bellwood et al.
2004).
R. B. Aronson (&) The principal recommendation emerging from the
Dauphin Island Sea Lab, 101 Bienville Boulevard,
Dauphin Island, AL 36528, USA
herbivory scenario is that large, networked marine
E-mail: raronson@disl.org protected areas (MPAs) are the key to coral recovery,
Tel.: +1-251-8617567 based on the assumption that cascading food-web effects
Fax: +1-251-8617540 will resume after no-take regulations are established and
enforced (Nyström et al. 2000; Nyström and Folke 2001;
R. B. Aronson
Department of Marine Sciences, Pandolfi et al. 2003, 2005; Bellwood et al. 2004; Hughes
University of South Alabama, et al. 2005). In the absence of grazing by Diadema, lar-
Mobile, AL 36688, USA ger, protected populations of herbivorous parrotfish
(Labridae: Scarinae) and surgeonfish (Acanthuridae)
W. F. Precht
Ecological Sciences Division, PBS&J, should exert top-down control on algal growth and help
2001 NW 107th Avenue, Miami, FL 33172, USA corals recruit and recover. The corals should in turn
E-mail: bprecht@pbsj.com provide the physical structure the herbivores require,
and the resultant positive feedback should benefit both
corals and reef fish. Macroalgal overgrowth of Caribbean corals
Given that MPAs protect predators, an alternative
prediction is that cascading effects should reduce her- To claim that overfishing has driven reef degradation in
bivory and increase, rather than decrease, macroalgal the Caribbean via the herbivory scenario is to invert the
abundance inside reserves (e.g., Gibson et al. 2004). causal link between macroalgal growth and coral mor-
This prediction fails in most cases, for two reasons. tality. In the vast majority of cases, corals have not been
First, it tacitly and incorrectly assumes only one killed by macroalgal overgrowth. Corals have been and
predatory trophic level (cf. Bascompte et al. 2005). If are being killed by large-scale, external perturbations,
MPAs protect the predators of herbivorous fish, then and growth of macroalgae on the newly vacated sub-
the latter should decline and macroalgae should in- stratum is a secondary, collateral effect (Aronson
crease; however, because MPAs are also designed to and Precht 2001a; McCook et al. 2001; McManus and
protect the predators of those predators, and because Polsenberg 2004).
many predatory species feed at multiple trophic levels, The popular conception that Caribbean corals have
herbivorous fish should actually increase and macro- been killed primarily by macroalgal overgrowth has its
algae should decline. Second, populations of herbivo- roots in papers by Sammarco (1980, 1982) and Littler
rous fish are often exploited on unprotected reefs and Littler (1985), and especially in the recent reinter-
(Hughes et al. 1987; Gibson et al. 2004; Mumby 2006; pretation of a paper by Hughes (1989). Hughes (1989)
and many others), so the protections afforded by monitored permanent quadrats at three depths at Rio
MPAs tend to have a direct positive impact on her- Bueno on the north coast of Jamaica from 1983 to 1987.
bivorous fish as well. For both of these reasons, Following the mass mortality of Diadema, two coral
MPAs generally do not decrease herbivory. In some species that grow as small, low-relief colonies, Agaricia
cases MPAs increase the abundance of large, herbiv- agaricites and Madracis mirabilis, were susceptible to
orous reef fish and the intensity of herbivory (Hawkins overgrowth by macroalgae and declined (see also de
and Roberts 2003; Mumby et al. 2006), but this effect Ruyter van Steveninck and Bak 1986). Other coral
is by no means universal (Mosquera et al. 2000; species were less strongly affected. The least affected
Graham et al. 2003; Micheli et al. 2004; Robertson species were Colpophyllia natans, which is a large brain
et al. 2005). coral, and the Montastraea annularis species complex,
The idea of cascading effects is appealing because it which produces massive, mound-shaped colonies.
is simple, makes intuitive sense, and can be addressed Colpophyllia declined at one depth and increased at
on a local level through sound fisheries management. another, while Montastraea, which is one of the three
Even if herbivory is not the only important factor, important framework-building corals of the region,
MPAs could at least help corals persist in the short significantly increased at the one depth for which it was
term while we search for longer term solutions reported (Hughes 1989, table 1). Hughes et al. (1987)
(Palumbi 2005). Are these interpretations correct? Will explicitly stated that coral species producing high-relief
MPAs make reef assemblages more resilient by colonies, such as Colpophyllia and Montastraea, were
decreasing macroalgal cover, increasing coral cover, ‘‘relatively safe from algal overgrowth.’’ Hughes (1989)
and helping restore coral assemblages to a desirable, also pointed out that macroalgal dominance was
coral-dominated state? What else should we do about pre-empting space and, therefore, sharply reducing coral
coral populations and coral reefs that will make a recruitment and the potential for recovery.
difference? Hughes (1994), citing his own 1989 paper, stated that
Here we dissect the herbivory scenario into its com- after the Diadema dieoff ‘‘most adult colonies’’ that had
ponent assertions: (1) that macroalgal overgrowth survived Hurricane Allen in 1980 had been ‘‘killed by
stemming from the loss of herbivores was the primary algal overgrowth, especially low-lying species...’’ In the
cause of coral mortality; and (2) that establishing MPAs same paragraph, Hughes (1994) discussed additional
will promote the recovery of coral populations. We then coral mortality from bleaching events. He went on to
evaluate in scientific and ethical terms the rationale for state in that paragraph that Montastraea declined
applying the precautionary principle to MPAs in the sharply, but he did not explicitly link this mortality to
face of incomplete evidence for the herbivory scenario. either macroalgal overgrowth or bleaching.
Finally, we propose a new agenda for promoting the Jackson et al. (2001) stated, ‘‘Large species of
recovery of coral reefs. Our agenda is based on accepting branching Acropora corals dominated shallow reefs in
the idea of multiple, interactive causality [Quinn and the tropical western Atlantic...until the 1980s when they
Dunham 1983; Holling and Allen 2002; see Hoegh- declined dramatically.’’ The first sentence of the fol-
Guldberg (2006) for application of these ideas to coral lowing paragraph then cited Hughes (1994) for the
reefs]. We depart from previous discussions by empha- statement, ‘‘Western Atlantic reef corals suffered sud-
sizing the urgent task of confronting environmental den, catastrophic mortality in the 1980s due to over-
globalization at the same time we act to solve problems growth by macroalgae that exploded after mass
at a local scale, including the preservation and mortality of the superabundant sea urchin D. antillarum
enhancement of stocks of reef fish. that [because of overfishing] was the last remaining
grazer of macroalgae.’’ These two sentences clearly im- other well-documented cases, widespread coral mor-
ply that macroalgae overgrew not just low-lying colonies tality has been the essential precondition for macro-
but also living colonies of the two framework-building algal dominance of reefs, and coral mortality has been
Acropora species: Ac. palmata (elkhorn coral) and Ac. decoupled from fishing pressure and herbivory (Aron-
cervicornis (staghorn coral). Moreover, Jackson (2001), son and Precht 2001a; Williams et al. 2001; Mumby
again citing Hughes (1994), explicitly stated that the et al. 2005).
principal cause of coral mortality across the Caribbean
region, including mortality of Montastraea spp. and
Acropora spp., was macroalgal overgrowth resulting Causal connections
from overfishing and the mass mortality of Diadema.
Neither Acropora species was reported from the study Coral mortality cannot in general be attributed to
quadrats in the original paper by Hughes (1989), and as macroalgal overgrowth, but once macroalgae are
mentioned above Montastraea had actually increased. established, they can suppress coral recruitment (Hughes
Subsequently, Pandolfi et al. (2003) stated that 1989; Hughes and Tanner 2000; McCook et al. 2001).
overfishing is the only reasonable explanation for the The question is whether we can manage for ecosystem
historical degradation of coral reefs. These authors did resilience by promoting the process of herbivory if we do
not propose a mechanism, but they cited Jackson et al. not simultaneously and directly confront the as-yet
(2001). Bellwood et al. (2004) stated that overexploita- uncontrolled global pressures causing coral mortality
tion of herbivorous fish can increase algal overgrowth of (e.g., McClanahan et al. 2002). Chief among these are
corals, citing Hughes (1994) and McCook et al. (2001). (1) climatic warming, increased cyclone intensity, and
Finally, Pandolfi et al. (2005) said essentially the same changes in ocean chemistry, all of which stem from
thing, but without attribution. The shifting baseline of greenhouse-gas emissions (Done 1999; Hoegh-Guldberg
emphasis in this sequence of papers brings to mind 1999; Kleypas et al. 1999; Pittock 1999; Walther et al.
previous warnings against accepting unsupported claims 2002; Emanuel 2005; McWilliams et al. 2005; Webster
about ecological effects (Elner and Vadas 1990; Hilborn et al. 2005); and (2) the spread of infectious marine
2004). diseases, which could be related to global warming and
In reviewing the literature, Nyström et al. (2000), may be exacerbated locally by nutrient loading (Aron-
Nyström and Folke (2001), and Scheffer et al. (2001) son and Precht 2001b; Harvell et al. 2002; Rosenberg
recognized the importance of coral mortality to the and Ben-Haim 2002; Bruno et al. 2003; Sutherland et al.
subsequent growth and dominance of macroalgae, but 2004). Nugues et al. (2004b) showed that macroalgae
they emphasized macroalgal overgrowth of living corals could be acting as reservoirs for coral disease, raising the
as well. Nyström and colleagues stressed that ‘‘even possibility of positive feedback in which coral mortality
large old coral colonies’’ were overgrown by macroalgae leads to increased macroalgal cover, which in turn leads
on Jamaican reefs. We also made this mistake (Aronson to more disease-induced coral mortality.
and Precht 2001a). Most of the 40% decline in coral cover on Caribbean
Lirman (2001) showed that macroalgae could over- reefs reported by Gardner et al. (2003) resulted from the
grow the living tissue of Montastraea faveolata colonies catastrophic mortality of Acropora palmata and Ac.
on heavily disturbed reefs in the Florida Keys. Although cervicornis, which were the dominant space occupants at
he did not assess the ecological impact on coral assem- shallow and intermediate depths, respectively. The pri-
blages, Lirman demonstrated experimentally that over- mary cause was a regional outbreak of white-band dis-
growth was more extensive when herbivores were ease (WBD), which began in the late 1970s and
excluded. Algal overgrowth, however, was predicated on continued through the early 1990s (Aronson and Precht
morbidity or partial mortality of the Montastraea colo- 2001b). The Caribbean-wide Acropora kill overwhelmed
nies. The degree to which macroalgae are able to over- the capacity of herbivores to respond behaviorally or
grow healthy coral tissue and, conversely, the degree to numerically to the vast expanses of newly opened space,
which established coral colonies monopolize space and and macroalgal abundance increased enormously
exclude algae remain equivocal in the context of com- (Knowlton 1992; Aronson and Precht 2001a). Although
munity-level dynamics (de Ruyter van Steveninck et al. the mass mortality of Diadema in 1983–1984 has been
1988; Umar et al. 1998; Lirman 2001; Jompa and promoted in the literature as the pivotal event that drove
McCook 2002; Nugues et al. 2004a; Mumby 2006). Caribbean reefs in the direction of macroalgal over-
There is only one experimental example from the growth and increased coral mortality (Jackson et al.
Caribbean in which macroalgae, when released from 2001; Bellwood et al. 2004), fully half the decline of coral
the pressures of herbivory by fish, overgrew healthy cover reported by Gardner et al. (2003) occurred prior to
corals to an ecologically significant extent. Lewis that event. This is clearly evident from one of their
(1986) removed herbivorous fish from large enclosures graphs, which shows the regional loss of coral cover
in the back reef at Carrie Bow Cay, Belize and ob- from 1977 to 2002 as a step-function rather than a
served increased macroalgal cover and decreased cover gradual decline (Fig. 1). Interestingly, there was no de-
of the dominant coral, Porites astreoides. Diadema cline of coral cover immediately following the loss of
were not important herbivores in her study. In all Diadema. Subsequent declines were related to losses of
 Coral reefs of the Flower Garden Banks (FGB) in the
northwestern Gulf of Mexico provide an interesting
counterpoint to what has happened in the Caribbean.
The FGB are too far north for the cold-sensitive
acroporids to have established populations prior to the
warming conditions of the past few decades (Precht and
Aronson 2004). The cover of living hard corals has re-
mained stable at 40–60% from at least as early as the
1970s to the present, in large part because in the 1980s
there were no acroporids available to die of WBD and
thereby open large areas of reef surface for algal colo-
nization (Aronson et al. 2005). As was the case in Belize,
herbivorous fish are abundant on the FGB and histori-
Fig. 1 Long-term trajectory of coral cover on Caribbean reefs,
cally have not been exploited (Pattengill-Semmens and
based on a meta-analysis of ecological studies. Solid circles are Gittings 2003). Following the decline of Diadema in the
means, and error bars are 95% bootstrapped confidence intervals. early 1980s from an initial density of less than 2 ind/m2
The thin, solid line connects the means; the thick, solid line displays to essentially zero (S.R. Gittings, unpublished data),
the overall trend in the data. The initial decline in 1977–1982, macroalgae increased by approximately 10% absolute
representing a decrease in absolute coral cover from approximately
50 to 25%, corresponds to the regional loss of Acropora palmata cover (from 2.5 to 13%) within 2 years; however, coral
and Acropora cervicornis caused by white-band disease. The cover did not decline and the macroalgae were only able
declines in 1987–1990 and 1998–1999 are associated with major, to grow in the patches of open substratum between
El Niño-induced bleaching events. The latter two declines primarily living coral colonies (Gittings 1998).
represent the loss of non-acroporid corals, because much of the
damage to acroporids was already done by 1987. Redrawn from A logical prediction of the herbivory scenario is
Gardner et al. (2003) that reefs in closer proximity to people, or in prox-
imity to people for a longer time, are more likely to be
degraded (Jackson et al. 2001; Pandolfi et al. 2003,
Montastraea spp. and other non-acroporid corals, due 2005; Bellwood et al. 2004; Palumbi 2005). Evidence
to bleaching and disease (e.g., McClanahan and Muthiga from throughout the Caribbean shows this not to be
1998; Ostrander et al. 2000; Burke et al. 2004). the case (Gardner et al. 2003, 2005; Lang 2003). For
The recent ecological history of the Belizean barrier instance, acroporid corals died from WBD at the same
reef highlights the importance of coral mortality as the rate both near and far from centers of fishing activity
prerequisite for macroalgal dominance. Hurricane and other anthropogenic stressors (Aronson and
Hattie severely damaged coral populations within a Precht 2001a, b). Even the most remote, least fished
section of the barrier reef in 1961, more than two dec- areas of the Pacific have been affected by coral
ades before the mass mortality of Diadema. The salient bleaching, resulting in mass coral mortality (Walther
consequence was macroalgal dominance for more than a et al. 2002; Alling et al. 2006). Large-scale perturba-
decade despite the abundant, unexploited state of reef- tions continue to cause significant coral decline in the
fish populations (Stoddart 1963, 1965, 1969, 1974). world’s largest and best-managed reef system, Austra-
Twenty years later, reef-crest populations of Ac. palmata lia’s Great Barrier Reef (Bellwood et al. 2004; Pandolfi
(0–3 m depth) were wiped out by WBD and replaced by et al. 2005). No form of local stewardship, including
macroalgae along the length of the barrier reef. Ac. large MPAs, could have prevented the large-scale dis-
cervicornis populations in the deep spur-and-groove turbances responsible for the worldwide increase in
zone of the fore reef (9–15 m) suffered a similar fate, and coral mortality over the past few decades.
that zone also shifted to macroalgal dominance (Littler The cause-and-effect relationship between fishing and
et al. 1987; Aronson et al. 1994). Coral cover remained coral mortality is tenuous at best. The instances in which
high, however, in the intervening shallow spur-and- such connections have been clearly established are
groove zone (3–6 m), which was dominated by Agaricia almost entirely limited to the Indo-Pacific and include:
tenuifolia (Aronson and Precht 1995). Ag. tenuifolia is (1) blasting, muro ami, cyanide poisoning, and other
not susceptible to WBD, which apparently affects only destructive fishing techniques, in which corals are killed
the genus Acropora. Ag. tenuifolia persisted at high levels as part of the extraction process (McManus et al. 1997);
of percent cover, and macroalgae did not have the and (2) a positive relationship in Fiji between fishing
opportunity to become dominant in the shallow spur- pressure and density of the corallivorous seastar
and-groove. Herbivorous fish were not exploited and Acanthaster planci, and, possibly as a cascading conse-
were more abundant, not less abundant, in the deep quence, a negative relationship between fishing pressure
spur-and-groove than in the shallow spur-and-groove; and coral cover (Dulvy et al. 2004). Conversely, Jones
Diadema was not a significant herbivore in either zone; et al. (2004) observed that declining fish abundance and
and consumption rates of experimentally tethered algae diversity were direct consequences of catastrophic coral
were higher, not lower, in the deep zone (Lewis and mortality in Papua New Guinea, again regardless of
Wainwright 1985). protection status.
 Aronson 2006), but their ongoing recovery is having a
Marine protected areas and coral recovery positive impact on coral populations (Edmunds and
Carpenter 2001; Carpenter and Edmunds 2006). At very
There is no doubt that reef resources must be better high densities, however, Diadema populations tend to
managed, and we support efforts in that direction. overgraze reef substrata, to the detriment of coral
Establishing new MPAs, enlarging the existing ones, and populations (Bak and van Eys 1975; Sammarco 1980;
pursuing adaptive management and co-management Carpenter 1981). The analogous effect occurs in the
strategies surely have positive impacts on species that are western Indian Ocean, where overfishing increases
the targets of exploitation (Dayton et al. 2000; Mosqu- echinoid populations, leading to complex ecosystem
era et al. 2000; NRC 2001; Russ and Alcala 2003; Sobel effects. These include lower algal cover, but also lower
and Dahlgren 2004; Granek and Brown 2005). Common coral cover from overgrazing (McClanahan and Muth-
sense and modeling results (Mumby 2006) suggest that iga 1989; McClanahan et al. 1994, 1999; McClanahan
using MPAs to increase herbivory by fish should have and Shafir 1990).
positive effects on corals and other living structural Populations of Acropora spp. are now recovering at
components (i.e., ecosystem impacts), but those effects localities dispersed throughout the Caribbean (Precht
have not been established empirically (Jameson et al. and Aronson 2006). Increases in the abundance of
2002; Graham et al. 2003; Sale et al. 2005). A further acroporids bear no relation to protection status or fish
complication is that some species of parrotfish in the abundance, and they could be related to the recovery of
Caribbean eat corals as well as algae (Littler et al. 1989; Diadema. Acropora spp. are even increasing dramatically
Miller and Hay 1998; Rotjan and Lewis 2005). On a on some reefs along the heavily fished north coast of
positive note, Mumby et al. (2006) showed that in- Jamaica (Idjadi et al. 2006; Precht and Aronson 2006). A
creased densities of herbivorous fish in a MPA in the question of critical importance is whether the recovery
Bahamas controlled macroalgal growth in the wake of of these coral populations will outpace future damage
widespread coral mortality from bleaching in 1998 (see from hurricanes, bleaching events, and disease out-
also Williams and Polunin 2001). In the Florida Keys, breaks.
however, Miller et al. (2003) found no difference in
macroalgal abundance between no-take reserves and
reference sites. Precaution, metaphor, and the politics of conservation
If the largest-scale sources of disturbance are not
addressed, they will continue to kill remnant coral Managers are watching corals die for reasons having
populations and suppress recruitment. Judging from nothing to do with the abundance of fish; the location,
recent observations of the extent of coral mortality from size, or connectivity of MPAs; or the presence or ab-
disease and bleaching, there is every reason to suspect sence of fishers. Should we play it safe, invoke the pre-
that under such circumstances higher levels of herbivory cautionary principle, and manage fisheries for reef
will have little or no influence on coral recovery. The resilience just in case it has some value in the end? For
same can be said for hurricane damage: Mumby’s (2006) most conservation-minded people, a positive response to
modeling results predict that enhanced herbivory by this question hinges on the metaphorical comparison of
parrotfish inside MPAs could promote coral recovery in reef degradation to human illness (e.g., Palumbi 2005).
subregions of the Caribbean with a low incidence of There are two types of errors in medical testing: false
hurricanes, but not where hurricanes frequently damage positives and false negatives. A false positive is a case in
the reefs. We know of only one case from St. Croix, US which the test result indicates a disease that is not
Virgin Islands in which the trophic control exerted by present. In a false negative, the disease is present but the
herbivorous fish not only suppressed macroalgal growth test fails to detect it. A false positive can be very
following an outbreak of WBD and subsequent hurri- upsetting but a false negative is far more dangerous, and
canes, but also promoted the recruitment and growth of this latter aspect of the medical analogy has dictated
corals (Bythell et al. 2000). Other recent studies in both attitudes toward the conservation and management of
the Caribbean and the Indo-Pacific have found that coral reefs.
coral decline was not ameliorated, nor was recovery In statistical testing, false positives (false alarms)
enhanced, in MPAs (McClanahan et al. 2001, 2005; correspond to type I statistical errors and false negatives
Jones et al. 2004). (defective alarms) correspond to type II errors (e.g.,
Sufficient densities of Diadema and another herbivo- Gonick and Smith 1993). Until recently, ecologists have
rous echinoid, Echinometra viridis, can control macro- been obsessed with avoiding type I error: falsely con-
algal growth and promote coral recruitment and growth cluding that there is an effect when in fact there is none.
in the Caribbean on both small and large spatial scales Fear of type I error has led to ultraconservative testing,
(Sammarco 1982; Hughes et al. 1987; Aronson and which has come at the cost of equally problematic type
Precht 1997, 2000). Whether or not Diadema popula- II error: falsely concluding that there is no effect when in
tions in the Caribbean were artificially inflated by fact there is one. The rationale for the precautionary
overfishing prior to the mass mortality is debatable (Hay principle is to avoid type II errors in conservation and
1984; Levitan 1992; Lessios et al. 2001; Precht and management by easing the requirement of an excessively
low rate of type I errors; in other words, the burden of simply will not make progress rapidly enough to deflect
proof is shifted from those who would protect the the downward trajectory of coral reefs.
environment onto those who would damage it (Buhl- Attending to the most workable issues—herbivory
Mortensen 1996; Dayton 1998). Cast in terms of coral and other local problems such as nutrient loading and
reefs, some aspects of reef degradation and recovery may pollution—could be salutary but in the long term may
not be completely understood or fully detectable right not be sufficient to reverse the decline of coral popula-
now, but the risks to (metaphorical) reef health are tions in the Caribbean. There is already an abundance of
potentially so severe that precaution is the only rea- creative thought on more effective, ecosystem-based
sonable option. approaches (Allison et al. 2003; Roberts et al. 2003;
In our view, reefs should not be construed as sick Pikitch et al. 2004; Guerry 2005; Fernandes et al. 2005;
superorganisms, with overfishing as etiology and McClanahan et al. 2006). One promising idea is to locate
MPAs as therapy. Neither the metaphor of human MPAs in areas where physical conditions are less con-
illness nor the statistical rationale for the precaution- ducive to bleaching and other physical damage (Done
ary principle applies to the idea that maintaining 1999; West and Salm 2003; but see Kim et al. 2005 on
abundant stocks of reef fish will necessarily promote marine diseases). Managing for socio-ecological resil-
coral recovery in the Caribbean. Without substantial ience (Adger et al. 2005) is also a step in the right
sacrifice from the rest of the world on global envi- direction. There is wisdom in integrated planning for
ronmental issues, MPAs will not fix the problem of environmental–economic–cultural disasters; however,
globally declining coral populations. only by addressing the root causes of those catastrophes
Facile application of the precautionary principle is a at the highest levels of government and society will we be
rich man’s game. As Cooney (2004) pointed out, ‘‘For able to find lasting solutions (McClanahan et al. 2002;
precaution to contribute to, rather than conflict with, Ehrlich and Kennedy 2005).
sustainable development, the burden of the precaution- The central agenda item of a new conservation ethic
ary principle must be borne by those most able to afford for all of us as coral reef scientists must be direct action
it.’’ In the case of industrial-scale fishing operations, on the global problems that now overshadow all local
mandated precautionary actions are affordable and issues, transcending even the largest conceivable marine
appropriate, but for the sake of subsistence fishers on reserves. It goes without saying that the situation is
Caribbean reefs, our goals must be compassion and considerably more dire than when Glynn (1991), Smith
accuracy rather than reflexive conservatism in either and Buddemeier (1992) and others sounded the alarm
direction. In other words, we should not insist on about global change and coral reefs. Knowing what we
unrealistically low error rates of either type (Dayton already know about climate change, it would be disin-
et al. 1995; Gray 1996; Holm and Harris 1999), and genuous to argue that such legislation is required on
meta-analysis may be capable of providing the desired purely precautionary grounds; the type I error rate is
balance (Côté et al. 2005). From a practical standpoint, already low for predicted and observed effects. Hope-
conserving populations of Caribbean reef fish and sus- fully, even the most recalcitrant of developed nations
taining fisheries for the benefit of local people are worthy can be convinced to help bear the costs of prevention, by
and attainable goals in their own right, so why compli- taking such actions as adopting the Kyoto Protocol and
cate the issue with premature promises about coral then adhering to it.
recovery that promote fishing restrictions of question- At the same time, single-taxon management remains a
able efficacy (see McClanahan 1999; Agardy et al. 2003; powerful and relatively inexpensive approach to achiev-
Hilborn 2004)? ing certain goals of conservation (Power et al. 1996;
Simberloff 1998; Hilborn 2004). Conserving and
enhancing populations of Diadema in tandem with action
A new agenda on global change could be the most direct route to coral
recovery, provided their densities are not elevated to the
Pandolfi et al. (2003) characterized coral mortality from point that they graze juvenile corals. Echinoids appear to
disease and bleaching as distractions from the main ef- be more effective than fish at reducing macroalgae and
fect, which they perceived to be overfishing. In a follow- enhancing coral recruitment in the Caribbean (Sammarco
up paper, Pandolfi et al. (2005) evidently reversed 1980, 1982; Carpenter 1986; Aronson and Precht 1997;
themselves and stated that all identified causes of deg- Carpenter and Edmunds 2006), so they could be our
radation are important and should be addressed. They best hope for increasing herbivory rapidly and effectively
recommended, furthermore, that ‘‘...scientists should in an overfished world. An unresolved question is
stop arguing about the relative importance of different whether MPAs, by protecting the predators of Diadema,
causes of reef decline.’’ Quite to the contrary, scientific will diminish the resilience of coral assemblages
debate and discussion are essential if we are to determine (McClanahan et al. 2001). In the absence of Diadema,
the most effective courses of action. We all know how no-take restrictions on parrotfish populations, such as
limited the resources are that can be brought to bear on those currently in place in Florida and Bermuda, could
reef conservation, especially in developing countries. If limit macroalgal overgrowth and promote coral recovery
we attempt to address all issues simultaneously, we on at least some Caribbean reefs (Mumby 2006).
 We cannot lapse into paralytic nihilism and assume Bascompte J, Melián CJ, Sala E (2005) Interaction strength com-
that global environmental issues—greenhouse-gas binations and the overfishing of a marine food web. Proc Natl
Acad Sci USA 102:5443–5447
emissions in this case—and their ultimate cause, the Bellwood DR, Hughes TP, Folke C, Nyström M (2004) Con-
human population (Birkeland 2004), are intractable fronting the coral reef crisis. Nature 429:827–833
problems. If predictions of escalating coral mortality Birkeland C (2004) Ratcheting down the coral reefs. BioScience
from bleaching and related causes are realized, then lo- 54:1021–1027
Bruno JF, Petes LE, Harvell CD, Hettinger A (2003) Nutrient
cal management in vacuo will amount to little more than enrichment can increase the severity of coral diseases. Ecol Lett
a series of rear-guard actions, which will at best delay 6:1056–1061
the demise of coral populations and the dissolution of Buhl-Mortensen L (1996) Type-II statistical errors in environ-
reef ecosystems. Reversing the global-scale causes of mental science and the precautionary principle. Mar Pollut Bull
coral mortality will be critical to saving coral reefs. 32:528–531
Burke CD, McHenry TM, Bischoff WD, Huttig ES, Yang W,
Thorndyke L (2004) Coral mortality, recovery and reef degra-
Acknowledgements We are grateful to N.D. Bood, G.F. Crozier, dation at Mexico Rocks Patch Reef Complex, Northern Belize,
M.R. Dardeau, K.J.P. Deslarzes, K.L. Heck Jr., N.L. Hilbun, L.S. Central America: 1995–1997. Hydrobiologia 530/531:481–487
Kaufman, S.L. Miller, R.M. Moody, P.S. Petraitis, C.S. Rogers, Bythell JC, Hillis-Starr Z, Rogers CS (2000) Local variability but
and J.F. Valentine for advice and discussion, and to S.R. Gittings landscape stability in coral reef communities following repeated
for access to his unpublished data from the Flower Garden Banks. hurricane impacts. Mar Ecol Prog Ser 204:93–100
Comments from J.F. Bruno, T.R. McClanahan, T.J.T. Murdoch, Carpenter RC (1981) Grazing by Diadema antillarum (Philippi) and
R.T. Paine, M.L. Robbart, R. van Woesik and an anonymous its effects on the benthic algal community. J Mar Res 39:749–765
reviewer improved the manuscript. We thank R.M. Moody and Carpenter RC (1986) Partitioning herbivory and its effects on coral
R.P. Dunne for help in drafting Fig. 1. Financial support was reef algal communities. Ecol Monogr 56:345–363
provided by the Dauphin Island Sea Lab (DISL) and PBS&J. This Carpenter RC, Edmunds PJ (2006) Local and regional scale
is DISL Contribution 376. recovery of Diadema promotes recruitment of scleractinian
corals. Ecol Lett 9:271–280
Cooney R (2004) The precautionary principle in biodiversity con-
References servation and natural resource management: an issues paper for
policy-makers, researchers and practitioners. IUCN, Gland
Côté IM, Gill JA, Gardner TA, Watkinson AR (2005) Measuring
Adger N, Hughes TP, Folke C, Carpenter SR, Rockström J (2005) coral reef decline through meta-analyses. Philos Trans R Soc
Social–ecological resilience to coastal disasters. Science Lond B 360:385–395
309:1036–1039 Dayton PK (1998) Reversal of the burden of proof in fisheries
Agardy T, Bridgewater P, Crosby MP, Day J, Dayton PK, management. Science 279:821–822
Kenchington R, Laffoley D, McConney P, Murray PA, Parks Dayton PK, Thrush SF, Agardy MT, Hofman RJ (1995) Envi-
JE, Peau L (2003) Dangerous targets? Unresolved issues and ronmental effects of marine fishing. Aquat Conserv 5:205–232
ideological clashes around marine protected areas. Aquat Dayton PK, Sala E, Tegner MJ, Thrush S (2000) Marine reserves:
Conserv 13:353–367 parks, baselines, and fishery management. Bull Mar Sci 66:617–
Alling A, Doherty O, Logan H, Feldman L, Dustan P (2006) 634
Catastrophic coral mortality in the remote Central Pacific de Ruyter van Steveninck ED, Bak RPM (1986) Changes in
Ocean: Kiribati, Phoenix Islands. Atoll Res Bull (in press) abundance of coral-reef bottom components related to mass
Allison GW, Gaines SD, Lubchenco J, Possingham HP (2003) mortality of the sea urchin Diadema antillarum. Mar Ecol Prog
Ensuring persistence of marine reserves: catastrophes require Ser 34:87–94
adopting an insurance factor. Ecol Appl 13:S8–S24 de Ruyter van Steveninck ED, van Mulekom LL, Breeman AM
Aronson RB, Precht WF (1995) Landscape patterns of reef coral (1988) Growth inhibition of Lobophora variegata (Lamouroux)
diversity: a test of the intermediate disturbance hypothesis. Womersly by scleractinian corals. J Exp Mar Biol Ecol
J Exp Mar Biol Ecol 192:1–14 115:169–178
Aronson RB, Precht WF (1997) Stasis, biological disturbance, and Done TJ (1999) Coral community adaptability to environmental
community structure of a Holocene coral reef. Paleobiology change at the scales of regions, reefs and reef zones. Am Zool
23:326–346 39:66–79
Aronson RB, Precht WF (2000) Herbivory and algal dynamics on Dulvy NKD, Freckleton RP, Polunin NVC (2004) Coral reef
the coral reef at Discovery Bay, Jamaica. Limnol Oceanogr cascades and the indirect effects of predator removal by
45:251–255 exploitation. Ecol Lett 7:410–416
Aronson RB, Precht WF (2001a) Evolutionary paleoecology of Edmunds PJ, Carpenter RC (2001) Recovery of Diadema antilla-
Caribbean coral reefs. In: Allmon WD, Bottjer DJ (eds) rum reduces macroalgal cover and increases abundance of
Evolutionary paleoecology: the ecological context of macro- juvenile corals on a Caribbean reef. Proc Natl Acad Sci USA
evolutionary change. Columbia University Press, New York, 98:5067–5071
pp 171–233 Ehrlich PR, Kennedy D (2005) Millennium assessment of human
Aronson RB, Precht WF (2001b) White-band disease and the behavior. Science 309:562–563
changing face of Caribbean coral reefs. Hydrobiologia 460:25– Elner RW, Vadas RL Sr (1990) Inference in ecology: the sea urchin
38 phenomenon in the northwestern Atlantic. Am Nat 136:108–125
Aronson RB, Edmunds PJ, Precht WF, Swanson DW, Levitan DR Emanuel K (2005) Increasing destructiveness of tropical cyclones
(1994) Large-scale, long-term monitoring of Caribbean coral over the past 30 years. Nature 436:686–688
reefs: simple, quick, inexpensive techniques. Atoll Res Bull Fernandes L, Day J, Lewis A, Slegers S, Kerrigan B, Breen D,
421:1–19 Cameron D, Jago B, Hall J, Lowe D, Innes J, Tanzer J,
Aronson RB, Precht WF, Murdoch TJT, Robbart ML (2005) Chadwick V, Thompson L, Gorman K, Simmons M, Barnett B,
Long-term persistence of coral assemblages on the Flower Sampson K, De’Ath G, Mapstone B, Marsh H, Possingham H,
Garden Banks, northwestern Gulf of Mexico: implications for Ball I, Ward T, Dobbs K, Aumend J, Slater D, Stapleton K
science and management. Gulf Mex Sci 23:84–94 (2005) Establishing representative no-take areas in the Great
Bak RPM, van Eys G (1975) Predation of the sea urchin Diadema Barrier Reef: large-scale implementation of theory on marine
antillarum Philippi on living coral. Oecologia 20:111–115 protected areas. Conserv Biol 19:1733–1744
Gardner TA, Côté IM, Gill JA, Grant A, Watkinson AR (2003) Jompa J, McCook LJ (2002) The effects of nutrients and herivory on
Long-term region-wide declines in Caribbean corals. Science competition between a hard coral (Porites cylindrica) and a
301:958–960 brown alga (Lobophora variegata). Limnol Oceanogr 47:527–534
Gardner TA, Côté IM, Gill JA, Grant A, Watkinson AR (2005) Jones GP, McCormick MI, Srinivasan M, Eagle JV (2004) Coral
Hurricanes and Caribbean coral reefs: impacts, recovery pat- decline threatens fish biodiversity in marine reserves. Proc Natl
terns, and role in long-term decline. Ecology 86:174–184 Acad Sci USA 101:8251–8253
Gibson J, McField M, Heyman W, Wells S, Carter J, Sedberry G Kim K, Dobson AP, Gulland FMD, Harvell CD (2005) Diseases
(2004) Belize’s evolving system of marine reserves. In: Sobel J, and the conservation of marine biodiversity. In: Norse EA,
Dahlgren C (eds) Marine reserves: a guide to science, design, Soulé ME (eds) Marine conservation biology: the science of
and use. Island Press, Washington, DC, pp 287–315 maintaining the sea’s biodiversity. Island Press, Washington,
Gittings SR (1998) Reef community stability on the Flower Garden DC, pp 149–166
Banks, northwest Gulf of Mexico. Gulf Mex Sci 16:161–169 Kleypas JA, Buddemeier RW, Archer D, Gattuso J-P, Langdon C,
Glynn PW (1991) Coral reef bleaching in the 1980s and possible Opdyke BN (1999) Geochemical consequences of increased
connections with global warming. Trends Ecol Evol 6:175–179 atmospheric carbon dioxide on coral reefs. Science 284:118–120
Gonick L, Smith W (1993) The cartoon guide to statistics. Har- Knowlton N (1992) Thresholds and multiple stable states in coral
perCollins, New York reef community dynamics. Am Zool 32:674–682
Graham NAJ, Evans RD, Russ GR (2003) The effects of marine Lang JC (ed) (2003) Status of coral reefs in the western Atlantic:
reserve protection on the trophic relationships of reef fishes on results of initial surveys, Atlantic and Gulf Rapid Reef
the Great Barrier Reef. Environ Conserv 30:200–208 Assessment (AGRRA) program. Atoll Res Bull 496:1–630
Granek E, Brown MA (2005) Co-management approach to marine Lessios HA, Garrido MJ, Kessing BD (2001) Demographic history
conservation in Mohéli, Comoros Islands. Conserv Biol of Diadema antillarum, a keystone herbivore on Caribbean
19:1724–1732 reefs. Proc R Soc Lond B Biol Sci 268:2347–2353
Gray JS (1996) Environmental science and a precautionary ap- Levitan DR (1992) Community structure in times past: influence of
proach revisited. Mar Pollut Bull 32:532–534 human fishing pressure on algal–urchin interactions. Ecology
Guerry AD (2005) Icarus and Daedalus: conceptual and tactical 73:1597–1605
lessons for marine ecosystem-based management. Front Ecol Lewis SM (1986) The role of herbivorous fishes in the organization
Environ 3:202–211 of a Caribbean reef community. Ecol Monogr 56:183–200
Harvell CD, Mitchell CE, Ward JR, Altizer S, Dobson AP, Ostfeld Lewis SM, Wainwright PC (1985) Herbivore abundance and
RS, Samuel MD (2002) Climate warming and disease risks for grazing intensity on a Caribbean coral reef. J Exp Mar Biol
terrestrial and marine biota. Science 296:2158–2162 Ecol 87:215–228
Hawkins JP, Roberts CM (2003) Effects of fishing on sex-changing Lirman D (2001) Competition between macroalgae and corals:
Caribbean parrotfishes. Biol Conserv 115:213–226 effects of herbivore exclusion and increased algal biomass on
Hay ME (1984) Patterns of fish and urchin grazing on coral reefs: coral survivorship and growth. Coral Reefs 19:392–399
are previous results typical? Ecology 65:446–454 Littler MM, Littler DS (1985) Factors controlling relative domi-
Hilborn R (2004) Ecosystem-based fisheries management: the car- nance of primary producers on biotic reefs. Proc 5th Int Coral
rot or the stick? Mar Ecol Prog Ser 274:275–278 Reef Cong 4:35–39
Hoegh-Guldberg O (1999) Climate change, coral bleaching and the Littler MM, Taylor PR, Littler DS, Sims RH, Norris JN (1987)
future of the world’s coral reefs. Mar Freshw Res 50:839–866 Dominant macrophyte standing stocks, productivity and com-
Hoegh-Guldberg O (2006) Complexities of coral reef recovery. munity structure on a Belizean barrier reef. Atoll Res Bull
Science 311:42–43 302:1–24
Holling CS, Allen CR (2002) Adaptive inference for distinguishing Littler MM, Taylor PR, Littler DS (1989) Complex interactions in
credible from incredible patterns in nature. Ecosystems 5:319– the control of coral zonation on a Caribbean reef flat. Oeco-
328 logia 80:331–340
Holm S, Harris J (1999) Precautionary principle stifles discovery. McClanahan TR (1999) Is there a future for coral reef parks in
Nature 400:398 poor tropical countries? Coral Reefs 18:321–325
Hughes TP (1989) Community structure and diversity of coral McClanahan TR, Muthiga NA (1989) Patterns of predation on a
reefs: the role of history. Ecology 70:275–279 sea urchin, Echinometra mathaei (de Blainville), on Kenyan
Hughes TP (1994) Catastrophes, phase shifts, and large-scale coral reefs. J Exp Mar Biol Ecol 126:77–94
degradation of a Caribbean coral reef. Science 265:1547–1551 McClanahan TR, Muthiga NA (1998) An ecological shift in a re-
Hughes TP, Tanner JE (2000) Recruitment failure, life histories, mote coral atoll of Belize over 25 years. Environ Conserv
and long-term decline of Caribbean corals. Ecology 81:2250– 25:122–130
2263 McClanahan TR, Shafir SH (1990) Causes and consequences of sea
Hughes TP, Reed DC, Boyle M-J (1987) Herbivory on coral reefs: urchin abundance and diversity in Kenyan coral reef lagoons.
community structure following mass mortalities of sea urchins. Oecologia 83:362–370
J Exp Mar Biol Ecol 113:39–59 McClanahan TR, Nugues M, Mwachireya S (1994) Fish and sea
Hughes TP, Bellwood DR, Folke C, Steneck RS, Wilson J (2005) urchin herbivory and competition in Kenyan coral reef lagoons:
New paradigms for supporting the resilience of marine eco- the role of reef management. J Exp Mar Biol Ecol 184:237–254
systems. Trends Ecol Evol 20:380–386 McClanahan TR, Muthiga NA, Kamukuru AT, Machano H,
Idjadi JA, Lee SC, Bruno JF, Precht WF, Allen-Requa L, Edm- Kiambo RW (1999) The effects of marine parks and fishing on
unds PJ (2006) Rapid phase-shift reversal on a Jamaican coral coral reefs of Northern Tanzania. Biol Conserv 89:161–182
reef. Coral Reefs 25:209–211 McClanahan TR, McField M, Huitric M, Bergman K, Sala E,
Jackson JBC (2001) What was natural in the coastal oceans? Proc Nyström M, Nordemar I, Elfwing T, Muthiga NA (2001) Re-
Natl Acad Sci USA 98:5411–5418 sponses of algae, corals and fish to the reduction of macroalgae
Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, in fished and unfished patch reefs of Glovers Reef Atoll, Belize.
Bourque BJ, Bradbury RH, Cooke R, Erlandson J, Estes JA, Coral Reefs 19:367–379
Hughes TP, Kidwell S, Lange CB, Lenihan HS, Pandolfi JM, McClanahan TR, Polunin NVC, Done TJ (2002) Resilience of
Peterson CH, Steneck RS, Tegner MJ, Warner RR (2001) coral reefs. In: Gunderson LH, Pritchard L Jr (eds) Resilience
Historical overfishing and the recent collapse of coastal eco- and the behavior of large-scale systems. SCOPE Series, vol 60,
systems. Science 293:629–638 Island Press, Washington, DC, pp 111–163
Jameson SC, Tupper MH, Ridley JM (2002) The three screen McClanahan TR, Maina J, Starger CJ, Herron-Perez P, Dusek E
doors: can marine ‘‘protected’’ areas be effective. Mar Pollut (2005) Detriments to post-bleaching recovery of corals. Coral
Bull 44:1177–1183 Reefs 24:230–246
McClanahan TR, Verheij E, Maina J (2006) Comparing the man- Pikitch EK, Santora C, Babcock EA, Bakun A, Bonfil R, Conover
agement effectiveness of a marine park and a multiple-use col- DO, Dayton P, Doukakis P, Fluharty D, Heneman B, Houde
laborative fisheries management area in East Africa. Aquat ED, Link J, Livingston PA, Mangel M, McAllister MK, Pope J,
Conserv 16:147–165 Sainsbury KJ (2004) Ecosystem-based fishery management.
McCook LJ, Jompa J, Diaz-Pulido G (2001) Competition between Science 305:346–347
corals and algae on coral reefs: a review of evidence and Pittock AB (1999) Coral reefs and environmental change: adapta-
mechanisms. Coral Reefs 19:400–417 tion to what? Am Zool 39:10–29
McManus JW, Polsenberg JF (2004) Coral-algal phase shifts on Power ME, Tilman D, Estes JA, Menge BA, Bond WJ, Mills LS,
coral reefs: ecological and environmental aspects. Prog Ocea- Daily G, Castilla JC, Lubchenco J, Paine RT (1996) Challenges
nogr 60:263–279 in the quest for keystones. BioScience 46:609–620
McManus JW, Reyes RB Jr, Nanola CL Jr (1997) Effects of some Precht WF, Aronson RB (2004) Climate flickers and range shifts of
destructive fishing methods on coral cover and potential rates of reef corals. Front Ecol Environ 2:307–314
recovery. Environ Manage 21:69–78 Precht WF, Aronson RB (2006) Death and resurrection of Carib-
McWilliams JP, Côté IM, Gill JA, Sutherland WJ, Watkinson AR bean reefs: a palaeoecological perspective. In: Côté I, Reynolds
(2005) Accelerating impacts of temperature-induced coral J (eds) Coral reef conservation. Cambridge University Press,
bleaching in the Caribbean. Ecology 86:2055–2060 Cambridge, pp 40–77
Micheli F, Halpern BS, Botsford LW, Warner RR (2004) Trajec- Quinn JF, Dunham AE (1983) On hypothesis testing in ecology
tories and correlates of community change in no-take marine and evolution. Am Nat 122:602–617
reserves. Ecol Appl 14:1709–1723 Roberts CM, Andelman S, Branch G, Bustamante RH, Castilla JC,
Miller MW, Hay ME (1998) Effects of fish predation and seaweed Dugan J, Halpern BS, Lafferty KS, Leslie H, Lubchenco J,
competition on the survival and growth of corals. Oecologia McArdle D, Possingham HP, Ruckleshaus M, Warner RR
113:231–238 (2003) Ecological criteria for evaluating candidate sites for
Miller MW, Aronson RB, Murdoch TJT (2003) Monitoring coral marine reserves. Ecol Appl 13:S199–S214
reef macroalgae: different pictures from different methods. Bull Robertson DR, Choat JH, Posada JM, Pitt J, Ackerman JL (2005)
Mar Sci 72:199–206 Ocean surgeonfish Acanthurus bahianus. II. Fishing effects on
Mosquera I, Côté IM, Jennings S, Reynolds JD (2000) Conserva- longevity, size and abundance? Mar Ecol Prog Ser 295:245–256
tion benefits of marine reserves for fish populations. Anim Rosenberg E, Ben-Haim Y (2002) Microbial diseases of corals and
Conserv 4:321–332 global warming. Environ Microbiol 4:318–326
Mumby P (2006) The impact of exploiting grazers (Scaridae) on the Rotjan RD, Lewis SM (2005) Selective predation by parrotfishes on
dynamics of Caribbean coral reefs. Ecol Appl 16:747–769 the reef coral Porites astreoides. Mar Ecol Prog Ser 305:193–201
Mumby PJ, Foster NL, Glynn Fahy EA (2005) Patch dynamics of Russ GR, Alcala AC (2003) Marine reserves: rates and patterns of
coral reef macroalgae under chronic and acute disturbance. recovery and decline of predatory fish, 1983–2000. Ecol Appl
Coral Reefs 24:681–692 13:1553–1565
Mumby PJ, Dahlgren CP, Harborne AR, Kappel CV, Micheli F, Sale PF, Cowen RK, Danilowicz BS, Jones GP, Kritzer JP,
Brumbaugh DR, Holmes KE, Mendes JM, Broad K, Sanchi- Lindeman KC, Planes S, Polunin NVC, Russ GR, Sadovy YJ,
rico JN, Buch K, Box S, Stoffle RW, Gill AB (2006) Fishing, Steneck RS (2005) Critical science gaps impede use of no-take
trophic cascades, and the process of grazing on coral reefs. fishery reserves. Trends Ecol Evol 20:74–80
Science 311:98–101 Sammarco PW (1980) Diadema and its relationship to coral spat
NRC (2001) Marine protected areas: tools for sustaining ocean mortality: grazing, competition, and biological disturbance.
ecosystems. National Research Council, National Academy J Exp Mar Biol Ecol 45:245–272
Press, Washington, DC Sammarco PW (1982) Echinoid grazing as a structuring force in
Nugues MM, Delvoye L, Bak RPM (2004a) Coral defense against coral communities: whole-reef manipulations. J Exp Mar Biol
macroalgae: differential effects of mesenterial filaments on the Ecol 61:31–55
green alga Halimeda opuntia. Mar Ecol Prog Ser 278:103–114 Scheffer M, Carpenter S, Foley JA, Folke C, Walker B (2001)
Nugues MM, Smith GW, van Hooidonk RJ, Seabra MI, Bak RPM Catastrophic shifts in ecosystems. Nature 413:591–596
(2004b) Algal contact as a trigger for coral disease. Ecol Lett Simberloff D (1998) Flagships, umbrellas, and keystones: is single-
7:919–923 species management passé in the landscape era? Biol Conserv
Nyström M, Folke C (2001) Spatial resilience of coral reefs. Eco- 83:247–257
systems 4:406–417 Smith SV, Buddemeier RW (1992) Global change and coral reef
Nyström M, Folke C, Moberg F (2000) Coral reef disturbance and ecosystems. Annu Rev Ecol Syst 23:89–118
resilience in a human-dominated environment. Trends Ecol Sobel J, Dahlgren C (eds) (2004) Marine reserves: a guide to sci-
Evol 15:413–417 ence, design, and use. Island Press, Washington, DC
Ostrander GK, Armstrong KM, Knobbe ET, Gerace D, Scully EP Stoddart DR (1963) Effects of Hurricane Hattie on the British
(2000) Rapid transition in the structure of a coral reef com- Honduras reefs and cays, October 30–31, 1961. Atoll Res Bull
munity: the effects of coral bleaching and physical disturbance. 95:1–142
Proc Natl Acad Sci USA 97:5297–5302 Stoddart DR (1965) Re-survey of hurricane effects on the British
Palumbi SR (2005) Germ theory for ailing corals. Nature 434:713– Honduras reefs and cays. Nature 207:589–592
714 Stoddart DR (1969) Post-hurricane changes on the British Hon-
Pandolfi JM, Bradbury RH, Sala E, Hughes TP, Bjorndal KA, duras reefs and cays: re-survey of 1965. Atoll Res Bull 131:1–25
Cooke RG, McArdle D, McClenachan L, Newman MJH, Stoddart DR (1974) Post-hurricane changes on the British Hon-
Paredes G, Warner RR, Jackson JBC (2003) Global trajectories duras reefs: re-survey of 1972. Proc 2nd Int Coral Reef Symp
of the long-term decline of coral reef ecosystems. Science 2:473–483
301:955–958 Sutherland KP, Porter JW, Torres C (2004) Disease and immunity
Pandolfi JM, Jackson JBC, Baron N, Bradbury RH, Guzman HM, in Caribbean and Indo-Pacific zooxanthellate corals. Mar Ecol
Hughes TP, Kappel CV, Micheli F, Ogden JC, Possingham HP, Prog Ser 266:273–302
Sala E (2005) Are U.S. coral reefs on the slippery slope to slime? Umar MJ, McCook LJ, Price IR (1998) Effects of sediment
Science 307:1725–1726 deposition on the seaweed Sargassum on a fringing coral reef.
Pattengill-Semmens CV, Gittings SR (2003) A rapid assessment of Coral Reefs 17:169–177
the Flower Garden Banks National Marine Sanctuary (stony Walther G-R, Post E, Convey P, Menzel A, Parmesan C, Beebee
corals, algae and fishes). Atoll Res Bull 496:500–510 TJC, Fromentin J-M, Hoegh-Guldberg O, Bairlein F (2002)
Petraitis PS, Dudgeon SR (2004) Detection of alternative states in Ecological responses to recent climate change. Nature 416:389–
marine communities. J Exp Mar Biol Ecol 300:343–371 395
Webster PJ, Holland GJ, Curry JA, Chang H-R (2005) Changes in Williams ID, Polunin NVC (2001) Large-scale associations
tropical cyclone number, duration, and intensity in a warming between macroalgal cover and grazer biomass on mid-depth
environment. Science 309:1844–1846 reefs in the Caribbean. Coral Reefs 19:358–366
West JM, Salm RV (2003) Resistance and resilience to coral Williams ID, Polunin NVC, Hendrick VJ (2001) Limits to grazing
bleaching: implications for coral reef conservation and man- by herbivorous fishes and the impact of low coral cover on
agement. Conserv Biol 17:956–967 macroalgal abundance on a coral reef in Belize. Mar Ecol Prog
Wilkinson C (ed) (2000) Status of coral reefs of the world: 2000. Ser 222:187–196
Australian Institute of Marine Science, Cape Ferguson and
Dampier
Wilkinson C (ed) (2004) Status of coral reefs of the world: 2004.
(vols 1 and 2). Australian Institute of Marine Science, Cape
Ferguson and Dampier