Articles


A Global Crisis for Seagrass
Ecosystems

ROBERT J. ORTH, TIM J. B. CARRUTHERS, WILLIAM C. DENNISON, CARLOS M. DUARTE, JAMES W.
FOURQUREAN, KENNETH L. HECK JR., A. RANDALL HUGHES, GARY A. KENDRICK, W. JUDSON KENWORTHY,
SUZANNE OLYARNIK, FREDERICK T. SHORT, MICHELLE WAYCOTT, AND SUSAN L. WILLIAMS


Seagrasses, marine flowering plants, have a long evolutionary history but are now challenged with rapid environmental changes as a result of coastal
human population pressures. Seagrasses provide key ecological services, including organic carbon production and export, nutrient cycling, sediment
stabilization, enhanced biodiversity, and trophic transfers to adjacent habitats in tropical and temperate regions. They also serve as “coastal canaries,”
global biological sentinels of increasing anthropogenic influences in coastal ecosystems, with large-scale losses reported worldwide. Multiple stressors,
including sediment and nutrient runoff, physical disturbance, invasive species, disease, commercial fishing practices, aquaculture, overgrazing, algal
blooms, and global warming, cause seagrass declines at scales of square meters to hundreds of square kilometers. Reported seagrass losses have led to
increased awareness of the need for seagrass protection, monitoring, management, and restoration. However, seagrass science, which has rapidly
grown, is disconnected from public awareness of seagrasses, which has lagged behind awareness of other coastal ecosystems. There is a critical need
for a targeted global conservation effort that includes a reduction of watershed nutrient and sediment inputs to seagrass habitats and a targeted
educational program informing regulators and the public of the value of seagrass meadows.

Keywords: seagrass, decline, sentinels, ecological services, monitoring




S                                            quantities of organic carbon. However, seagrasses and these
    eagrasses—a unique group of flowering plants that
    have adapted to exist fully submersed in the sea—                 associated ecosystem services are under direct threat from a
profoundly influence the physical, chemical, and biological en-             host of anthropogenic influences.
vironments in coastal waters, acting as ecological engineers                Seagrasses are distributed across the globe (figure 2), but
(sensu Wright and Jones 2006) and providing numerous im-                 unlike other taxonomic groups with worldwide distribution,
portant ecological services to the marine environment                  they exhibit low taxonomic diversity (approximately 60
(Costanza et al. 1997). Seagrasses alter water flow, nutrient cy-            species worldwide, compared with approximately 250,000
cling, and food web structure (Hemminga and Duarte 2000).                terrestrial angiosperms). The three independent lineages of
They are an important food source for megaherbivores such                seagrass (Hydrocharitaceae, Cymodoceaceae complex, and
as green sea turtles, dugongs, and manatees, and provide                 Zosteraceae) evolved from a single lineage of monocotyle-
critical habitat for many animals, including commercially                donous flowering plants between 70 million and 100 million
and recreationally important fishery species (figure 1; Beck               years ago (figure 3a; Les et al. 1997). This is in stark contrast
et al. 2001). They also stabilize sediments and produce large              to other plant groups that have colonized the marine envi-


Robert J. Orth (e-mail: jjorth@vims.edu) is a professor in the School of Marine Science, Virginia Institute of Marine Science, College of William and Mary, Gloucester
Point, VA 23062. Tim J. B. Carruthers is a science integrator and William C. Dennison is vice president for science applications and a professor at the Integration and
Application Network, University of Maryland Center for Environmental Science, Cambridge, MD 21613. Carlos M. Duarte is a research professor at the Instituto
Mediterráneo de Estudios Avanzados, Consejo Superior de Investigaciones Científicas/Universidad de las Islas Baleares, Calle Miquel Marqués 21, 07190 Esporles,
Islas Baleares, Spain. James W. Fourqurean is department chair of biological sciences and a member of the Southeast Environmental Research Center, Florida
International University, Miami, FL 33199. Kenneth L. Heck Jr. is a professor and chair of university programs at the Dauphin Island Sea Lab, Dauphin Island, AL
36528. A. Randall Hughes is a postdoctoral researcher, Suzanne Olyarnik is a graduate student, and Susan L. Williams is a professor at the University of California at
Davis; Hughes, Olyarnik, and Williams are also associated with the Bodega Marine Laboratory, Bodega Bay, CA 94923. Gary A. Kendrick is an associate professor at
the School of Plant Biology, University of Western Australia, Crawley 6009, Western Australia. W. Judson Kenworthy is a research biologist at the Center for Coastal
Fisheries and Habitat Research, National Ocean Service, National Oceanic and Atmospheric Administration, Beaufort, NC 28516. Frederick T. Short is a research
professor in the Department of Natural Resources and chair of the Natural Resources and Earth Systems Science PhD program at the University of New Hampshire,
Jackson Estuarine Laboratory, Durham, NH 03824. Michelle Waycott is a senior lecturer in the School of Marine and Tropical Biology, James Cook University,
Townsville, 4811 Queensland, Australia. © 2006 American Institute of Biological Sciences.


www.biosciencemag.org                                          December 2006 / Vol. 56 No. 12 • BioScience 987
Articles

                                  ronment, such as salt marsh plants, mangroves, and marine
                                  algae, which are descended from multiple and diverse evo-
                                  lutionary lineages. In spite of their low species diversity and
                                  unique physiological characteristics, seagrasses have suc-
                                  cessfully colonized all but the most polar seas (figure 2).
                                  Compared with seagrass meadows, the other major coastal
                                  marine habitats are geographically restricted to much smaller
                                  latitudinal ranges (mangroves and coral reefs in tropical re-
                                  gions, kelp beds and salt marshes in temperate regions).
                                   Seagrasses have developed unique ecological, physiologi-
                                  cal, and morphological adaptations to a completely sub-
                                  mersed existence, including internal gas transport, epidermal
                                  chloroplasts, submarine pollination, and marine dispersal
                                  (den Hartog 1970, Les et al. 1997). To provide oxygen to their
                                  roots and rhizomes, often growing in highly reducing sedi-
                                  ments with toxic sulfide levels, and to support large amounts
                                  of nonphotosynthetic tissue (Terrados et al. 1999), seagrasses
                                  require some of the highest light levels of any plant group
                                  worldwide (approaching 25% of incident radiation in some
                                  seagrass species, compared with 1% or less for other angio-
                                  sperm species; Dennison et al. 1993). These extremely high
Figure 1. Examples of seagrass meadows and associated
                                  light requirements mean that seagrasses are acutely respon-
animals. (a) Seahorse (Hippocampus sp.) in temperate
                                  sive to environmental changes, especially those that alter
Cymodocea nodosa meadow, Mediterranean Sea. Photo-
                                  water clarity. Although it is true that the global distribution
graph: Gérard Pergent. (b) School of zebrafish (Girella
                                  and abundance of seagrasses have changed over evolutionary
zebra) over a temperate Posidonia australis meadow,
                                  time in response to sea-level change, physical modification of
Western Australia. Photograph: Gary A. Kendrick. (c)
                                  coastlines (figure 3a, 3b), and global changes in atmospheric
Manatee (Trichechus manatus) feeding in a tropical
                                  carbon dioxide (CO2) concentration and water temperature
Thalassia testudinum meadow, Puerto Rico. Photograph:
                                  (figure 3c; Crowley 1990, Berner and Kothavala 2001), the very
James Reid. (d) Green sea turtle (Chelonia midas) feeding
                                  gradual changes in environmental conditions over the history
in a tropical T. testudinum meadow, Yucatán. Photo-
                                  of seagrass evolution are overshadowed by current changes
graph: Robert P. van Dam.
                                  to the coastal zone resulting from increased human pres-
                                  sures. These pressures result in the degradation of estuaries
                                  and coastal seas, producing changes to species and habitats
                                                   (Lotze et al. 2006). These
                                                   rapid contemporary changes
                                                   have been caused by a mul-
                                                   titude of mechanisms, in-
                                                   cluding increased nutrient
                                                   and sediment runoff, in-
                                                   vasive species, hydrological
                                                   alterations, and commercial
                                                   fishing practices. As a result,
                                                   reported seagrass losses
                                                   worldwide have been accu-
                                                   mulating.

                                                   Seagrasses as ecologi-
                                                   cal service providers
                                                   and biological sentinels
                                                   Seagrass meadows have im-
                                                   portant ecological roles in
                                                   coastal ecosystems and pro-
                                                   vide high-value ecosystem
Figure 2. Current global distribution of seagrass in relation to mean ocean temperature.
                                                   services compared with other
Regional divisions are based on polar (< 4 degrees Celsius [°C]), temperate (4°C–24°C),
                                                   marine and terrestrial habi-
and tropical (> 24°C) climate (Green and Short 2003).

988 BioScience • December 2006 / Vol. 56 No. 12                               www.biosciencemag.org
                                                              Articles

tats (figure 4; Costanza et al. 1997). For example,
primary production from seagrasses and their as-
sociated macro- and microepiphytes rivals or ex-
ceeds that of many cultivated terrestrial ecosystems
(Duarte and Chiscano 1999). Seagrasses also pro-
vide an enormous source of carbon to the detri-
tal pool, some of which is exported to the deep
sea, where it provides a critical supply of organic
matter in an extremely food-limited environment
(Suchanek et al. 1985). Much of the excess organic
carbon produced is buried within seagrass sedi-
ments, which are hotspots for carbon sequestration
in the biosphere (Duarte et al. 2005). The structural
components of seagrass leaves, rhizomes, and roots
modify currents and waves, trapping and storing
both sediments and nutrients, and effectively fil-
ter nutrient inputs to the coastal ocean (Hem-
minga and Duarte 2000). Biodiversity in seagrass
meadows is greater than in adjacent unvegetated
areas, and faunal densities are orders of magnitude
higher inside the meadows (Hemminga and Duarte
2000). They also serve as a nursery ground, often
to juvenile stages of economically important species
of finfish and shellfish, although the species vary
by region and climate (figure 4; Beck et al. 2001, Figure 3. Seagrass evolution time line for the past 100 million years during
Heck et al. 2003). The large-scale loss of seagrass periods of changing (a) global ocean structure (Dietz and Holden 1970),
that occurred on both sides of the North Atlantic (b) mean sea level (Miller et al. 2005), and (c) atmospheric carbon dioxide
Ocean in the early 1930s, a result of “eelgrass wast- (CO2) concentration (Berner and Kothavala 2001) and mean global
ing disease” (Rasmussen 1977), had many effects temperature (Crowley 1990). Abbreviations: °C, degrees Celsius; KT,
on the ecosystem. Associated with this loss were a Cretaceous–Tertiary (approximately 65 million years ago [MYA]); m,
collapse of scallop fisheries and dramatic reductions meters; ppm, parts per million.
in waterfowl populations. In addition, it resulted
in the only known case of an extinction of a marine gastro-      shallow, protected coastal waters, directly in the path of water-
pod (Carlton et al. 1991). Finally, the proximity of seagrass     shed nutrient and sediment inputs, and are therefore highly
beds to other critical habitats, such as salt marshes (in tem-    susceptible to these inputs (figure 4), unlike mangrove forests
perate regions) or mangroves and coral reefs (in tropical re-     (which are largely unaffected by water quality) or coral reefs
gions), facilitates trophic transfers and cross-habitat utilization  (which occur farther away from the imputs).
by fishes and invertebrates (Beck et al. 2001). This provides      Another feature that makes seagrasses a valuable biologi-
an energy subsidy that may be essential in maintaining the      cal indicator is that they integrate environmental impacts
abundance of some coral reef fish species (Valentine and       over measurable and definable timescales (Longstaff and
Heck 2005).                              Dennison 1999, Carruthers et al. 2002), and a number of
  Moreover, seagrasses can be considered as biological sen-     key examples support this concept. Increased coastal develop-
tinels, or “coastal canaries.” Changes in seagrass distribution,   ment leading to nutrient inputs in Cockburn Sound, Australia,
such as a reduction in the maximum depth limit (Abal and       led to large-scale losses of seagrass into the 1990s, and sea-
Dennison 1996) or widespread seagrass loss (Cambridge and       grasses remain at low levels in the area today (Walker et al.
McComb 1984), signal important losses of ecosystem services      2006). The loss of seagrass led to sediment resuspension,
that seagrasses provide. Seagrasses are sessile, essentially in-   hampering restoration efforts and negatively affecting fish pop-
tegrating the relevant water quality attributes, such as chloro-   ulations. In this region of Australia, if seagrass density drops
phyll and turbidity, that affect the light reaching their leaves.   below the 25th percentile of the long-term average for two con-
Several features of seagrasses and seagrass meadows result in     secutive years, remedial action is now mandated by law in con-
their particular importance in this regard. The widespread dis-    fronting diffuse sources of pollution. Because of the
tribution of seagrasses throughout tropical and temperate re-     susceptibility of seagrasses to such stresses and the high level
gions (figure 2) allows better assessment of larger-scale trends   of ecosystem services they provide, seagrasses are also used as
than do other comparable coastal habitats, such as man-        one of the five sensitive indicators of pollution in the US
grove, corals, or salt marsh plants, which are limited to only    National Estuarine Eutrophication Assessment (Bricker et
one of these broad geographic regions. Seagrasses also live in    al. 2003). And in the Chesapeake Bay, historical levels of

www.biosciencemag.org                                December 2006 / Vol. 56 No. 12 • BioScience 989
Articles




          Figure 4. Conceptual diagrams for (a) tropical and (b) temperate seagrass ecosystems,
          detailing key ecosystem services and major mechanisms of seagrass loss. (c) Temperate
          and tropical seagrass genera (and family names), from ephemeral to persistent.


seagrass abundance (based on an assessment of historical      prevents the shoreward migration of the seagrasses necessi-
photography) are being used as a target for the attainment of    tated by sea-level rise. In addition, significant seagrass habi-
improved water quality from comprehensive nutrient and sed-     tat continues to be lost to coastal development (marinas,
iment management strategies (Orth et al. 2002).           canal estates, and industry), leading to meadow fragmenta-
                                  tion, with unknown consequences for long-term survival
The challenge of rapid environmental changes            (Fonseca et al. 2000).
Seagrasses now live in a marine environment with a lower        Seagrass meadows are increasingly being recognized for
mean temperature and lower availability of CO2 than were ex-    their dynamic nature, in many cases on an interannual basis,
perienced by their ancestors (Beer and Koch 1996). The re-     with a dynamic equilibrium at broad spatial scales (square
cent trends of increasing global temperature, sea-level rise, and  kilometers) even in areas where water quality remains high
CO2 concentrations (figure 3c, 5a, 5b) could result in envi-    (Fonseca et al. 2000, Kendrick et al. 2000). But this awareness
ronments that are potentially more conducive for many sea-     is being overshadowed by rapid, large-scale seagrass losses
grass species. However, as a result of increased human       over relatively short temporal scales throughout the world, in
population (figure 5c) and concomitant increased anthro-      places such as the European Mediterranean (Marbà et al.
pogenic pressure to the coastal zone, the rates of change in    2005), Japan (Environment Agency of Japan 2000), the
coastal waters today are much faster than those experienced     Chesapeake Bay (Orth and Moore 1983) and Florida Bay
in the previous 100 million years of evolutionary history,     (Fourqurean and Robblee 1999) in North America, and Cock-
and may well be too fast to allow these species to adapt.      burn Sound (Walker et al. 2006) and Western Port (Bulthuis
Where human activities have led to a reduction in the genetic    1983) in Australia. Although there are places where seagrass
diversity of seagrasses, these species’ adaptation could be     loss has been reversed following improvements in water qual-
compromised (Williams 2001). In many areas, human alter-      ity, such as Tampa Bay, North America (Tomasko et al. 2005),
ations to the coastal zone (coastal hardening through break-    and Hervey Bay, Australia (Preen and Marsh 1995), the num-
waters, harbors, and groins) have led to a situation that      ber of declines far exceeds the reported increases, leading to

990 BioScience • December 2006 / Vol. 56 No. 12                               www.biosciencemag.org
                                                                        Articles

the concern that seagrasses are experiencing a global cri-
sis (table 1; Short and Wyllie-Echeverria 1996, Duarte
1999, 2002, Green and Short 2003).

Multiple stressors behind
seagrass declines
Environmental, biological, and extreme climatological
events have been identified as causes of seagrass losses in
temperate and tropical regions (table 1). Threats from
global climate change (e.g., increases in sea surface tem-
perature, sea level, and frequency and intensity of storms
and associated surge and swells), from regional shifts in
water quality (e.g., in the Chesapeake Bay; Kemp et al.
2005), and from more localized impacts due to increased
loading of sediment, contaminants, and nutrients (figure
6a) reaching coastal environments (e.g., Cockburn Sound;
Walker et al. 2006) have had demonstrable impacts on the
health of seagrass-dominated coastal ecosystems worldwide
(table 1). These global, regional, and local stressors can all
independently result in large-scale seagrass loss; however,
seagrasses are often simultaneously influenced by multi-
ple stressors at different temporal and spatial scales, and
studies that examine the interacting impacts of multiple
stressors are lacking. In all regions, the environmental
                                 Figure 5. Seagrass–human interaction time line for the past 10,000
effects of excess nutrients or sediments are the most com-
                                 years, showing (a) carbon dioxide (CO2) concentrations (Thoning
mon and significant causes of seagrass decline, and result
                                 et al. 1989, Petit et al. 1999), (b) mean sea level (Fleming et al.
in small to very large areas of seagrass being lost. The di-
                                 1998), and (c) global human population (Cohen et al. 1995).
rect influence of other organisms (e.g., brown tides, urchin
                                 Abbreviations: m, meters; ppm, parts per million; YBP, years
overgrazing, and disease) has also led to large-scale losses
                                 before the present.
and, when acting in concert with suspended sediments and
                                      Bay, Australia, which resulted in 1000 km2 of seagrass loss, high
nutrients, can accelerate the trajectory of seagrass loss for the
area in question. The greater diversity of causes attributed to       mortality and emigration of dugong eventually occurred
seagrass declines in temperate regions most likely reflects         (Preen and Marsh 1995). Recently, greater attention has
the much greater research and monitoring effort in Europe,         focused on the role of top-down control in seagrass declines,
North America, and southern Australia (Duarte 1999), rather         as cascading effects on trophic dynamics follow the loss of
than greater susceptibility in these regions (table 1).           higher-level consumers in seagrasses and other ecosystems
  Extreme climatic events (e.g., hurricanes, tsunamis) also can      (Heck et al. 2000, Jackson et al. 2001). Thus, seagrasses are
have large-scale impacts on seagrass communities and sub-          being influenced by both bottom-up and top-down processes
sequent effects on the ecosystem services provided by seagrass       (Heck and Orth 2006). Although our primary focus here is
meadows (table 1, figure 4). In the case of the pulsed turbidity      on the seagrasses themselves, seagrass-associated species are
events following the passage of tropical storms in Hervey          also threatened or vulnerable to extinction. Eleven of 28 fish


Table 1. A synthesis of 47 representative case studies of seagrass loss.
                               Major mechanisms of loss (number of reports)
Area lost (km2)         Environmental                   Biological                   Extreme events

Temperate region
  < 1.0      Dredging, hydrological, dune migration (7)       Herbivory, introduced species, bioturbation (7)   Ice scour, heat waves (2)
  1.0–100     Eutrophication, sediment deposition (4)        Brown tide (1)                    No data
  > 100      Eutrophication, sea-level rise, high temperature (5)  Wasting disease (1)                 No data
Tropical region
  < 1.0      Vessel grounding, thermal pollution (5)        Herbivory (3)                    No data
  1.0–100     Eutrophication, boating, sedimentation (6)       Brown tide, urchin herbivory (2)           No data
  > 100      Hydrological, sediment resuspension (3)        No data                       Pulsed turbidity (1)

  Note: The seagrass genera studied in temperate regions include Cymodocea, Halodule, Heterozostera/Zostera, Posidonia, Syringodium, and Thalassia; gen-
era studied in tropical regions include Halodule, Halophila, Syringodium, Thalassia, and Zostera. An expanded table detailing the results of each study can
be found at www.vims.edu/bio/sav/bioscience_global_crisis_table_1.pdf.



www.biosciencemag.org                                       December 2006 / Vol. 56 No. 12 • BioScience 991
Articles

species vulnerable to extinction in the United States use        phication or cease dredge-and-fill activities, it is virtually
seagrass habitat during at least part of their life cycle (Musick    impossible to remove a nonnative species after establish-
et al. 2000).                              ment and spread (Lodge et al. 2006). Lastly, the rapid
  In addition to the well-documented causes of seagrass        expansion of fish farming and other aquaculture practices
declines, other threats to these species are emerging. Over the     (e.g., shellfish culture) can have serious consequences on
last 20 years, introductions of nonnative marine species have      local populations of seagrasses through physical disturbance
arisen as a major environmental challenge for the world’s        or increased deposition of organic matter and nutrients
oceans (Carlton 1989). Such introductions are accelerating       (Marbà et al. 2006).
worldwide (Ruiz et al. 2000), a trend that will continue as the
                                    Seagrass monitoring, management,
pathways for introductions widen and proliferate and as in-
                                    protection, and restoration
tervention lags (figure 6b; Naylor et al. 2001, Levine and
D’Antonio 2003, Padilla and Williams 2004). At least 28 non-      Reported cases of seagrass loss have increased almost tenfold
native species have become established in seagrass beds world-     over the last 40 years in both tropical and temperate regions
wide, of which 64% have documented or inferred negative        (figure 6c), suggesting increased rates of seagrass decline
effects (figure 6b). The concern about this emerging threat to     worldwide. In response to seagrass loss caused by increasing
seagrass beds is that, whereas it is possible to reverse eutro-    anthropogenic stresses on coastal seagrass meadows, during
                                              the last decade there has been a major in-
                                              crease in the number of marine protected
                                              areas that include seagrass (figure 6d)
                                              and in seagrass monitoring (figure 6e)
                                              and restoration projects throughout the
                                              world. The current challenges are to
                                              synthesize this information to enhance
                                              our understanding of global seagrass
                                              processes, threats, and change, and to
                                              apply this knowledge to develop effec-
                                              tive resource management programs.
                                              Efforts to protect seagrasses now include
                                              19 monitoring programs that encompass
                                              30 seagrass species in 44 countries
                                              (approximately 2000 sites).
                                                Perhaps the most difficult issue facing
                                              resource managers as they try to protect
                                              seagrasses is in implementing manage-
                                              ment plans to reduce nutrients and sed-
                                              iments from both diffuse and point
                                              sources in surrounding watersheds, es-
                                              pecially where watersheds cross jurisdic-
                                              tional boundaries. Seagrass distribution
                                              and abundance are being successfully in-
                                              corporated into water quality manage-
                                              ment programs and environmental
                                              impact studies in several areas, notably the
                                              Chesapeake Bay and Florida in North
                                              America, and Moreton Bay and the Great
                                              Barrier Reef Marine Park in Australia
                                              (Kenworthy et al. 2006). Management
                                              applications are based on the foundation
                                              of seagrass knowledge developed in each
                                              of those areas and are aimed at estab-
Figure 6. Time line showing pressures on seagrass populations and responses         lishing water quality standards to con-
over the last 150 years, including (a) nitrogen fertilizer use (Frink et al. 1999),     serve and restore seagrasses (Dennison et
(b) species introduced to the marine environment (Ruiz et al. 2000), (c) reported      al. 1993, Coles and Fortes 2001, Kenwor-
cases of seagrass loss in both tropical and temperate regions since 1965, (d) marine    thy et al. 2006).
protected areas (based on Spalding et al. 2003), and (e) monitoring effort (Duarte       A number of seagrass management
et al. 2004).                                        plans have objectives with quantitative

992 BioScience • December 2006 / Vol. 56 No. 12                                 www.biosciencemag.org
                                                               Articles

metrics aimed at restoring seagrass to target levels that allow    efforts should be aimed at systemwide approaches to protect
resource managers, who are making critical decisions, to ex-      these ecosystems.
pend public funds. One key example is the process of seagrass
                                    Science and public awareness
restoration, for which costs are very high (Kenworthy et al.
                                    of seagrasses: A disconnect
2006) and success is uncertain. Worldwide, the success of
seagrass transplantation and restoration is around 30% (Fon-      Over the last 35 years, scientists have responded to the need
seca et al. 1998), although in some regions higher success rates    for more information on seagrasses and their contribution to
have been reported (Green and Short 2003). Numerous          the productivity of coastal and estuarine systems with more
restoration projects have been attempted or are being planned     research and monitoring programs that have resulted in a 100-
at mostly small scales (< 1 ha) using a variety of techniques     fold increase in the annual number of papers published
with both adult plants and seeds, although interest in larger-     during this time period. This increase represents a sustained
scale transplant programs is growing as resource managers be-     publication growth rate of 12.8% per year (figure 7a) and in-
come more aware of the value of seagrass and develop          cludes a seagrass atlas (Green and Short 2003), a methods book
mitigation programs to offset losses from activities such as      (Short and Coles 2001), and two research syntheses (Hem-
dredging (Fonseca et al. 1998). However, some species are so      minga and Duarte 2000, Larkum et al. 2006).
difficult to transplant that restoration is not logistically or eco-   Despite the increase in scientific publications on seagrasses,
nomically feasible, and longer-term studies that compare the      the level of public awareness, as reflected by the number of
functioning of transplanted areas with that of natural systems     reports on seagrass ecosystems in the media, is far less than
are rare (Fonseca et al. 1998).                    that for other coastal habitats. Salt marshes, mangroves, and
  Seagrass loss is usually the symptom of a larger problem.      coral reefs receive 3-fold to 100-fold more media attention
To effectively reverse the decline of seagrasses, conservation     than seagrass ecosystems, although the services provided by
plans must first identify and resolve the problems at a scale     seagrasses, together with algal beds, deliver a value at least twice
that includes the interconnectivity of coastal systems and       as high as the next most valuable habitat (figure 7b; Costanza
the mechanisms affecting the declines and gains (e.g., water      et al. 1997). This difference in media attention partly reflects
quality, land use practices). Once this is done, restoration      disproportionate research effort, as the number of scientific
efforts should be balanced against the capacity of seagrasses     documents on seagrass is also below those on salt marshes,
to recover naturally. Strategic restoration can introduce       mangroves, and coral reefs (figure 7b). Reports on seagrasses
founder populations that can accelerate the overall recovery      in the New York Times and National Geographic are 3 to 50
of the ecosystem (Orth et al. 2006). At present, our knowledge     times lower than those for salt marshes, mangroves, and
of the population dynamics of seagrasses remains poor for the     coral reefs. Nevertheless, these data indicate that translating
majority of species and regions (Kenworthy 2000). As a         scientific understanding of seagrass ecosystems into public
result, considerable research efforts will be required to guide    awareness has not been as effective as for other coastal eco-
effective restoration and preserve genetic diversity (Williams     systems.
2001). Better ecological information on such approaches is        Much of this disconnect between available information and
required, and especially the trajectories on how rapidly ecosys-    public awareness undoubtedly stems from the invisibility of
tem services are restored. Until this is achieved, management     seagrasses, as they grow underwater, and from the avoidance




Figure 7. Comparison of seagrass, salt marsh, mangrove, and coral reef habitats in terms of (a) journal publications (Web of
Science 1950–2006) and (b) citations in more broadly accessed media (Google and Web of Science), and estimated monetary
value of ecosystem services provided by these habitats (Costanza et al. 1997).


www.biosciencemag.org                                December 2006 / Vol. 56 No. 12 • BioScience 993
Articles

of their very shallow habitat by many boaters (unless they run     ecosystems. Furthermore, developing models that incorpo-
aground, whereupon the boat propellers damage seagrass).        rate the landscape scale of seagrass dynamics and can link to
In addition, although a high diversity and abundance of or-      watershed runoff models will help inform resource man-
ganisms live in seagrass beds, the animals are often small       agers about the consequences of various watershed
and cryptic, in contrast to the large and dazzling organisms      activities on seagrass dynamics.
that attract the general public to coral reefs. The few charis-      Our major recommendation is to respond to the global
matic megafauna that do inhabit seagrass meadows (mana-        seagrass crisis with extensive conservation efforts involving
tees, dugongs, and sea turtles; figure 1) are elusive and not easily  comprehensive nutrient management schemes, sanctuaries or
viewed in the wild, and because they are endangered by over-      protected areas, and education for the public and resource
harvesting and habitat destruction, they are not nearly        managers (Kenworthy et al. 2006). The majority of seagrass
as abundant as the fish and invertebrates on coral reefs        losses are a result of human activities in the adjacent water-
(Jackson et al. 2001). Without strong public support for        sheds, which lead to increased nutrient and sediment runoff.
seagrasses and the uncharismatic but highly productive         The isolated case studies of seagrass recoveries when inputs
animals they shelter, conservation efforts will continue to      of nutrients (e.g., Tampa Bay, Florida; Tomasko et al. 2005)
lag behind those of other key coastal ecosystems.           or sediments (Hervey Bay, Australia; Preen and Marsh 1995)
                                    are curtailed demonstrate the potential effectiveness of con-
The need for a global conservation                   servation efforts. The preservation of seagrasses and their
effort for seagrasses                         associated ecosystem services—in particular, biodiversity,
We have presented the case that seagrasses are facing a crisis     primary and secondary production, nursery habitat, and
due to a diverse array of pressures from human activities in      nutrient and sediment sequestration—should be a global
the coastal zone, as well as the increased frequency and in-      priority. We believe that the crisis facing seagrass ecosystems
tensity of natural disasters such as hurricanes, which may also    can be averted with a global conservation effort, and this
be indirectly associated with human activities (i.e., global      effort will benefit not just seagrasses and their associated
warming). Although seagrasses have experienced considerable
                                    organisms but also the entirety of coastal ecosystems.
environmental changes in sea level, CO2, and temperature over
the past 100 million years of their evolutionary history, these
                                    Acknowledgments
historical changes were gradual. How well seagrasses can
                                    This work was conducted as a part of the Global Seagrass
adapt to the unprecedented rates of change they are cur-
                                    Trajectories Working Group supported by the National
rently experiencing is unknown. In view of the many cases of
                                    Center for Ecological Analysis and Synthesis, a center funded
documented seagrass loss, predictions for the future of
                                    by the National Science Foundation (grant DEB-0072909), the
seagrass-dominated coastal systems cannot be optimistic.
                                    University of California at Santa Barbara, and the state of
  While the global science community has focused on pre-
                                    California. The genesis of this project was in discussions with
dicting future change to the oceans and to coastal ecosystems
                                    colleagues at multiple International Seagrass Biology Work-
for iconic groups like corals, seagrasses have generally been
                                    shops. This product is contribution no. 334 from the South-
ignored by all but marine scientists, except in the most highly
                                    east Environmental Research Center at Florida International
developed countries. Given the importance of seagrasses to
                                    University; no. 2772 from the School of Marine Science, Vir-
humans (Costanza et al. 1997, Larkum et al. 2006), it is im-
                                    ginia Institute of Marine Science, College of William and
perative to assess the future of seagrasses under the expo-
                                    Mary; no. 2355 from Bodega Marine Laboratory, University
nentially increasing pressures of human growth and
                                    of California–Davis; no. 439 from the Jackson Estuarine Lab-
development in the watersheds and coastal zones of the
                                    oratory, University of New Hampshire; no. 382 from the
world. A quantitative analysis of seagrass trajectories could
                                    Dauphin Island Sea Lab; and no. 4015 from the University of
form the foundation to incorporate seagrasses into a global
                                    Maryland Center for Environmental Science.
science policy for the world’s oceans.
  Monitoring seagrass meadows is a necessary but insufficient
                                    References cited
conservation activity, because remedial actions are not fully
                                    Abal EG, Dennison WC. 1996. Seagrass depth range and water quality in
effective in stopping declines once they are detected (Short and
                                      Southern Moreton Bay, Queensland, Australia. Marine and Freshwater
Burdick 1996, Delgado et al. 1999). It will be critically im-       Research 47: 763–771.
portant to forecast the likely cumulative effects of the known     Beck MW, et al. 2001. The identification, conservation, and management of
and emerging stressors of seagrasses. Present scenarios for fu-      estuarine and marine nurseries for fish and invertebrates. BioScience
                                      51: 633–641.
ture seagrass trends are either of limited geographic extent
                                    Beer S, Koch EW. 1996. Photosynthesis of seagrasses vs. marine macroalgae
(Fourqurean et al. 2003) or limited to qualitative statements
                                      in globally changing CO2 environments. Marine Ecology Progress Series
(Short and Neckles 1999, Duarte 2002, Duarte et al. forth-
                                      141: 199–204.
coming). Quantitative forecasts, together with risk analysis      Berner RA, Kothavala Z. 2001. GEOCARB III: A revised model of atmospheric
identifying the most vulnerable areas, can inform conserva-        CO2 over Phanerozoic time. American Journal of Science 301: 182–204.
tion and management strategies and help determine the most       Bricker SB, Ferreira JG, Simas T. 2003. An integrated methodology for
cost-effective allocation of resources to conserve seagrass        assessment of estuarine trophic status. Ecological Modelling 169: 39–60.


994 BioScience • December 2006 / Vol. 56 No. 12                                    www.biosciencemag.org
                                                                            Articles
                                            Administration (NOAA) Coastal Ocean Office. NOAA Coastal Ocean
Bulthuis DA. 1983. Effects of in situ light reduction on density and growth
  of the seagrass Heterozostera tasmanica (Martens ex Aschers.) den Har-       Program Decision Analysis Series no. 12.
  tog in Western Port, Victoria, Australia. Journal of Experimental Marine    Fonseca MS, Kenworthy WJ, Whitfield PE. 2000. Temporal dynamics of
  Biology and Ecology 67: 91–103.                           seagrass landscapes: A preliminary comparison of chronic and extreme
Cambridge ML, McComb AJ. 1984. The loss of seagrasses in Cockburn            disturbance events. Biologia Marina Mediterranea 7: 373–376.
  Sound, Western Australia, 1: The time course and magnitude of seagrass     Fourqurean JW, Robblee MB. 1999. Florida Bay: A recent history of ecological
  decline in relation to industrial development. Aquatic Botany 20: 229–243.     changes. Estuaries 22: 345–357.
Carlton JT. 1989. Man’s role in changing the face of the ocean: Biological in-   Fourqurean JW, Boyer JN, Durako MJ, Hefty LN, Peterson BJ. 2003.
  vasions and implications for conservation of near-shore environments.        Forecasting responses of seagrass distributions to changing water
  Conservation Biology 3: 265–273.                          quality using monitoring data. Ecological Applications 13: 474–489.
Carlton JT, Vermeij G, Lindberg DR, Carlton DA, Dudley EC. 1991. The first     Frink CR, Waggoner PE, Ausubel JH. 1999. Nitrogen fertilizer: Retrospect and
  historical extinction of a marine invertebrate in an ocean basin: The        prospect. Proceedings of the National Academy of Sciences 96: 1175–1180.
  demise of the eelgrass limpet, Lottia alveus. Biological Bulletin 180:     Green EP, Short FT, eds. 2003. World Atlas of Seagrasses. Berkeley: Univer-
  72–80.                                       sity of California Press.
Carruthers TJB, Dennison WC, Longstaff BJ, Waycott M, Abal EG,           Heck KL, Orth RJ. 2006. Predation in seagrass beds. Pages 537–550 in
  McKenzie LJ, Long WJL. 2002. Seagrass habitats of Northeast Australia:       Larkum AWD, Orth RJ, Duarte CM, eds. Seagrasses: Biology, Ecology, and
  Models of key processes and controls. Bulletin of Marine Science 71:        Conservation. Dordrecht (The Netherlands): Springer.
  1153–1169.                                   Heck KL, Pennock J, Valentine J, Coen L, Sklenar SS. 2000. Effects of nutri-
Cohen JE. 1995. How Many People Can the Earth Support? New York:            ent enrichment and large predator removal on seagrass nursery habitats:
  Norton.                                       An experimental assessment. Limnology and Oceanography 45:
Coles RC, Fortes M. 2001. Protecting seagrass—approaches and methods.          1041–1057.
  Pages 445–463 in Short FT, Coles RG, eds. Global Seagrass Research       Heck KL, Hays C, Orth RJ. 2003. A critical evaluation of the nursery role
  Methods. Amsterdam: Elsevier.                            hypothesis for seagrass meadows. Marine Ecology Progress Series 253:
Costanza R, et al. 1997. The value of the world’s ecosystem services and        123–136.
  natural capital. Nature 387: 253–260.                      Hemminga M, Duarte CM. 2000. Seagrass Ecology. Cambridge (United
Crowley TJ. 1990. Are there any satisfactory geologic analogs for a future green-
                                            Kingdom): Cambridge University Press.
  house warming? Journal of Climate 3: 1282–1292.
                                          Jackson JBC, et al. 2001. Historical overfishing and the recent collapse of coastal
Delgado O, Ruiz J, Perez M, Romero R, Ballesteros E. 1999. Effects of fish
                                            ecosystems. Science 293: 629–638.
  farming on seagrass (Posidonia oceanica) in a Mediterranean bay:
                                          Kemp WM, et al. 2005. Eutrophication of Chesapeake Bay: Historical trends
  Seagrass decline after organic loading cessation. Oceanologica Acta 22:
                                            and ecological interactions. Marine Ecology Progress Series 303: 1–19.
  109–117.
                                          Kendrick GA, Hegge BJ, Wyllie A, Davidson A, Lord DA. 2000. Changes in
den Hartog C. 1970. The Seagrasses of the World. Amsterdam: North-
                                            seagrass cover on Success and Parmelia Banks, Western Australia between
  Holland.
                                            1965 and 1995. Estuarine, Coastal and Shelf Science 50: 341–353.
Dennison WC, Orth RJ, Moore KA, Stevenson JC, Carter V, Kollar S,
                                          Kenworthy WJ. 2000. The role of sexual reproduction in maintaining
  Bergstrom PW, Batiuk RA. 1993. Assessing water quality with submersed
                                            populations of Halophila decipiens: Implications for the biodiversity
  aquatic vegetation. BioScience 43: 86–94.
                                            and conservation of tropical seagrass ecosystems. Pacific Conservation
Dietz RS, Holden JC. 1970. Reconstruction of Pangea: Breakup and disper-
                                            Biology 5: 260–268.
  sion of continents, Permian to present. Journal of Geophysical Research
                                          Kenworthy WJ, Wyllie-Echeverria S, Coles RG, Pergent G, Pergent-Martini
  75: 4939–4956.
                                            C. 2006. Seagrass conservation biology: An interdisciplinary science for
Duarte CM. 1999. Seagrass ecology at the turn of the millennium: Challenges
                                            protection of the seagrass biome. Pages 595–623 in Larkum AWD, Orth
  for the new century. Aquatic Botany 65: 7–20.
                                            RJ, Duarte CM, eds. Seagrasses: Biology, Ecology and Conservation.
———. 2002. The future of seagrass meadows. Environmental Conserva-
                                            Dordrecht (The Netherlands): Springer.
  tion 29: 192–206.
                                          Larkum WD, Orth RJ, Duarte CM, eds. 2006. Seagrasses: Biology, Ecology
Duarte CM, Chiscano CL. 1999. Seagrass biomass and production: A
                                            and Conservation. Dordrecht (The Netherlands): Springer.
  reassessment. Aquatic Botany 65: 159–174.
                                          Les DH, Cleland MA, Waycott M. 1997. Phylogenetic studies in the
Duarte CM, Alvarez E, Grau A, Krause-Jensen D. 2004. Which monitoring
                                            Alismatidae, II: Evolution of the marine angiosperms (seagrasses) and
  strategy should be chosen? Pages 41–44 in Borum J, Duarte CM, Krause-
                                            hydrophily. Systematic Botany 22: 443–463.
  Jensen D, Greve TM, eds. European Seagrasses: An Introduction to
                                          Levine JM, D’Antonio CM. 2003. Forecasting biological invasions with
  Monitoring and Management. (13 November 2006; www.seagrasses.org/
                                            increasing international trade. Conservation Biology 17: 322–326.
  handbook/european_seagrasses_high.pdf)
                                          Lodge DM, et al. 2006. Biological invasions: Recommendations for U.S.
Duarte CM, Middelburg J, Caraco N. 2005. Major role of marine vegetation
                                            policy and management. Position paper, Ecological Society of America.
  on the oceanic carbon cycle. Biogeosciences 2: 1–8.
                                            (13 October 2006; http://weblogs.nal.usda.gov/invasivespecies/archives/
Duarte CM, Borum J, Short FT, Walker DI. Seagrass ecosystems: Their
                                            2006/06/biological_inva.shtml)
  global status and prospects. In Polunin NVC, ed. Aquatic Ecosystems:
                                          Longstaff BJ, Dennison WC. 1999. Seagrass survival during pulsed turbid-
  Trends and Global Prospects. Cambridge (United Kingdom): Cam-
                                            ity events: The effects of light deprivation on the seagrasses Halodule
  bridge University Press. Forthcoming.
                                            pinifolia and Halophila ovalis. Aquatic Botany 65: 105–121.
Environment Agency of Japan. 2000. Threatened Wildlife of Japan: Red
                                          Lotze HK, et al. 2006. Depletion, degradation, and recovery potential of
  Data Book. 2nd ed., vol. 8: Vascular Plants. Tokyo: Japan Wildlife Research
                                            estuaries and coastal seas. Science 312: 1806–1809.
  Center.
                                          Marbà N, Duarte CM, Díaz-Almela E, Terrados J, Álvarez E, Martínez R,
Fleming K, Johnston P, Zwartz D, Yokoyama Y, Lambeck K, Chappell J.
                                            Santiago R, Gacia E, Grau AM. 2005. Direct evidence of imbalanced
  1998. Refining the eustatic sea-level curve since the last glacial maximum
                                            seagrass (Posidonia oceanica) shoot population dynamics along the
  using far- and intermediate-field sites. Earth and Planetary Science
                                            Spanish Mediterranean. Estuaries 28: 53–62.
  Letters 163: 327–342.
                                          Marbà N, Santiago R, Díaz-Almela E, Álvarez E, Duarte CM. 2006. Seagrass
Fonseca MS, Kenworthy WJ, Thayer GW. 1998. Guidelines for the conser-
                                            (Posidonia oceanica) vertical growth as an early indicator of fish
  vation and restoration of seagrasses in the United States and adjacent
  waters. Silver Spring (MD): National Oceanic and Atmospheric            farm-derived stress. Estuarine, Coastal and Shelf Science 67: 475–483.


www.biosciencemag.org                                        December 2006 / Vol. 56 No. 12 • BioScience 995
Articles
Miller KG, Kominz MA, Browning JV, Wright JD, Mountain GS, Katz ME,        Short FT, Neckles H. 1999. The effects of global climate change on
  Sugarman PJ, Cramer BS, Christie-Blick N, Pekar SF. 2005. The Phanero-       seagrasses. Aquatic Botany 63: 169–196.
  zoic record of global sea-level change. Science 310: 1293–1298.        Short FT, Wyllie-Echeverria S. 1996. Natural and human-induced disturbance
Musick JA, et al. 2000. Marine, estuarine, and diadromous fish stocks at risk     of seagrasses. Environmental Conservation 23: 17–27.
  of extinction in North America. Fisheries 25: 6–30.              Spalding M, Taylor M, Ravilious C, Short F, Green E. 2003. Global overview:
Naylor R, Williams SL, Strong DR. 2001. Aquaculture—a gateway for exotic        The distribution and status of seagrasses. Pages 5–26 in Green EP, Short
  species. Science 294: 1655–1656.                          FT, eds. World Atlas of Seagrasses: Present Status and Future Conserva-
Orth RJ, Moore KA. 1983. Chesapeake Bay: An unprecedented decline in
                                            tion. Berkeley: University of California Press.
  submerged aquatic vegetation. Science 222: 51–53.
                                          Suchanek TH, Williams SW, Ogden JC, Hubbard DK, Gill IP. 1985. Utiliza-
Orth RJ, Batiuk RA, Bergstrom PW, Moore KA. 2002. A perspective on two
                                            tion of shallow-water seagrass detritus by Caribbean deep-sea macrofauna:
  decades of policies and regulations influencing the protection and
                                            δ 13C evidence. Deep Sea Research 32: 2201–2214.
  restoration of submerged aquatic vegetation in Chesapeake Bay, USA.
                                          Terrados J, Duarte CM, Kamp-Nielsen L, Agawin NSR, Gacia E, Lacap D,
  Bulletin of Marine Science 71: 1391–1403.
                                            Fortes MD, Borum J, Lubanski M, Greve T. 1999. Are seagrass growth and
Orth RJ, Luckenbach ML, Marion SR, Moore KA, Wilcox DJ. 2006. Seagrass
                                            survival affected by reducing conditions in the sediment? Aquatic Botany
  recovery in the Delmarva coastal bays. Aquatic Botany 84: 26–36.
                                            65: 175–197.
Padilla DK, Williams SL. 2004. Beyond ballast water: Aquarium and orna-
                                          Thoning KW, Tans PP, Komhyr WD. 1989. Atmospheric carbon dioxide at
  mental trades as sources of invasive species in aquatic ecosystems.
  Frontiers in Ecology and the Environment 2: 131–138.                Mauna Loa observatory, 2: Analysis of the NOAA GMCC data, 1974–1985.
Petit JR, et al. 1999. Climate and atmospheric history of the past 420,000 years    Journal of Geophysical Research 94: 8549–8565.
  from the Vostok ice core, Antarctica. Nature 399: 429–436.           Tomasko DA, Corbett CA, Greening HS, Raulerson GE. 2005. Spatial and tem-
Preen AR, Marsh H. 1995. Responses of dugongs to large-scale loss of          poral variation in seagrass coverage in southwest Florida: Assessing the
  seagrass from Hervey Bay, Queensland, Australia. Wildlife Research 22:       relative effects of anthropogenic nutrient load reductions and rainfall in
  507–519.                                      four contiguous estuaries. Marine Pollution Bulletin 50: 797–805.
Rasmussen E. 1977. The wasting disease of eelgrass (Zostera marina) and      Valentine JF, Heck KL. 2005. Perspective review of the impacts of overfish-
  its effects on environmental factors and fauna. Pages 1–51 in McRoy CP,
                                            ing on coral reef food web linkages. Coral Reefs 24: 209–213.
  Helfferich C, eds. Seagrass Ecosystems. New York: Marcel Dekker.
                                          Walker DI, Kendrick GA, McComb AJ. 2006. Decline and recovery of
Ruiz GM, Fofonoff PW, Carlton JT, Wonham MJ, Hines AH. 2000. Invasion
                                            seagrass ecosystems—the dynamics of change. Pages 551–565 in Larkum
  of coastal marine communities in North America: Apparent patterns,
                                            AWD, Orth RJ, Duarte CM, eds. Seagrasses: Biology, Ecology and
  processes, and biases. Annual Review of Ecology and Systematics 31:
                                            Conservation. Dordrecht (The Netherlands): Springer.
  481–531.
                                          Williams SL. 2001. Reduced genetic diversity in eelgrass transplantations
Short FT, Burdick DM. 1996. Quantifying eelgrass habitat loss in relation
                                            affects both individual and population fitness. Ecological Applications
  to housing development and nitrogen loading in Waquoit Bay,
                                            11: 1472–1488.
  Massachusetts. Estuaries 19: 730–739.
                                          Wright JP, Jones CG. 2006. The concept of organisms as ecosystem engineers
Short FT, Coles RG, eds. 2001. Global Seagrass Research Methods.
  Amsterdam: Elsevier.                                ten years on: Progress, limitations, and challenges. BioScience 56: 203–209.




996 BioScience • December 2006 / Vol. 56 No. 12                                           www.biosciencemag.org