September 2004 / Vol. 54 No. 9 ?  BioScience 843
Articles
Recent reviews describing human impacts on coastal and marine ecosystems in the Caribbean em-
phasize two major problems. First is the lack of information
about processes controlling the productivity of diverse coastal
ecosystems. Second is the potential long-term negative impact
of human disturbances on the goods and services these
ecosystems provide (Richards and Bohnsack 1990, Ogden
1997). The Caribbean Sea has been diagnosed as an eco-
system showing ?signs of environmental stress? (Richards
and Bohnsack 1990), representing a global trend in which 
human impacts on marine ecosystems are more the rule
than the exception. A current debate in environmental re-
source management concerns how best to understand the
magnitude of ecosystem change associated with human 
impacts and to designate the global regions that are more 
susceptible to human activities (Vitousek et al. 1997). In the
Caribbean Sea and its watersheds, degradation of water and
land resources is related to major declines in biodiversity
Victor H. Rivera-Monroy (e-mail: riverav@louisiana.edu) is a senior research scientist, and Robert R. Twilley was director, at the Center for Ecology and  Environmen-
tal Technology, University of Louisiana, Lafayette, LA 70504; Twilley is now director of the Wetland Biogeochemistry Institute, Louisiana State University, Baton Rouge,
LA 70803. Rivera-Monroy and Twilley are also associated with the Department of Biology at the University of Louisiana. David Bone is a professor at the Universidad
Sim?n Bol?var, Instituto de Tecnologia y Ciencias Marinas, Caracas 1080-A, Venezuela. Daniel L. Childers is an associate professor at the Department of Biological 
Sciences and the Southeast Environmental Research Center, Florida International University, Miami, FL 33199. Carlos Coronado-Molina is a senior research scientist
at the South Florida Water Management District, West Palm Beach, FL 33416. Ilka C. Feller is a senior scientist at the Smithsonian Environmental Research Center, Edge-
water, MD 21037. Jorge Herrera-Silveira is a professor at the Centro de Investigaci?nes y de Estudios Avanzados del Instituto Polit?cnico Nacional (CINVESTAV-IPN),
Unidad M?rida, M?rida, Yucat?n, Mexico 97310. Rudolf Jaffe is a professor at the Department of Chemistry and director of the Southeast Environmental Research 
Center, Florida International University, Miami, FL 33199. Ernesto Mancera is director of the Instituto de Investigaciones Tropicales, Universidad del Magdalena, Santa
Marta, Colombia. Eliska Rejmankova is a professor at the Department of Environmental Science and Policy, University of California, Davis, CA 95616. Joseph E.
Salisbury is a research scientist at the Ocean Processes Analysis Laboratory, University of New Hampshire, Durham, NH 03824. Ernesto Weil is a professor at the Marine
Sciences Department, University of Puerto Rico, Lajas, Puerto Rico 00667. ? 2004 American Institute of Biological Sciences.
A Conceptual Framework to
Develop Long-Term Ecological
Research and Management
Objectives in the Wider
Caribbean Region
VICTOR H. RIVERA-MONROY, ROBERT R. TWILLEY, DAVID BONE, DANIEL L. CHILDERS, CARLOS CORONADO-
MOLINA, ILKA C. FELLER, JORGE HERRERA-SILVEIRA, RUDOLF JAFFE, ERNESTO MANCERA, ELISKA REJMANKOVA,
JOSEPH E. SALISBURY, AND ERNESTO WEIL
The Caribbean Sea and its watersheds show signs of environmental degradation. These fragile coastal ecosystems are susceptible to environmental
impacts, in part because of their oligotrophic conditions and their critical support of economic development. Tourism is one of the major sources of
income in the Caribbean, making the region one of the most ecotourism dependent in the world. Yet there are few explicit, long-term, comprehen-
sive studies describing the structure and function of Caribbean ecosystems. We propose a conceptual framework using the environmental signature
hypothesis of tropical coastal settings to develop a series of research questions for the reef?sea-grass?wetland seascape. We applied this approach
across 13 sites throughout the region, including ecosystems in a variety of coastal settings with different vulnerabilities to environmental impacts.
This approach follows the strategy developed by the Long Term Ecological Research program of the National Science Foundation to establish eco-
logical research questions best studied over decades and large spatial areas.
Keywords: Caribbean, eutrophication, LTER (long-term ecological research), coastal settings, human influence
(Solow 1995), species extinction (Brooks and Smith 2001),
mass mortality of organisms (Lessios 1988, Harvell et al.
1999), and reduction of marine resource productivity (Jack-
son et al. 2001). Furthermore, watersheds of the Caribbean
Sea are one of the top five hotspots of marine and terrestrial
biodiversity identified around the world, representing an
area with an exceptional concentration of endemic species.
Caribbean watersheds are undergoing an exceptional loss of
habitat; only about 10% of their primary terrestrial vegeta-
tion remains (Myers et al. 2000, Brooks and Smith 2001).
The Caribbean Sea consists of fragile coastal ecosystems that
are susceptible to environmental impacts, in part because of
their oligotrophic conditions (Richards and Bohnsack 1990,
Lapointe 1997) and their critical support of economic de-
velopment. The wider Caribbean region includes 36 to 40
countries, with a total population of 38 million (figure 1; CIA
2003). Travel and tourism are major sources of income in the
Caribbean, making the region one of the most ecotourism 
dependent in the world. The Caribbean travel industry had
gross earnings of $17 billion in 1999, which provided about
900,000 direct and indirect jobs and contributed about a
quarter of the region?s foreign exchange profits. Travel and
tourism were expected to generate $34.6 billion of economic
activity in 2003, and this amount will grow to $78.4 billion
by 2013 (WTTC 2003). Employment in the travel industry in
2003 was estimated at 1,857,000 jobs, or 12% of total em-
ployment (1 in every 8 jobs) in the region. Approximately 100
million tourists visit the Caribbean annually, contributing
about 43% of the combined gross national product and one-
third of the export earnings for the region (including the
Dominican Republic, Puerto Rico, Jamaica, the Bahamas,
Cuba, Barbados, Aruba, Bermuda, Trinidad and Tobago, and
Cura?ao; WTTC 2003). In addition to supporting intense
tourism activities, the Caribbean Sea is one of the most heav-
ily traversed seas worldwide. Shipping activities in the region
include the transport of oil, which generates pollution in the
844 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles
Figure 1. Location of study sites in the wider Caribbean region: 1, Everglades, Florida; 2, Puerto Rico (wet); 3, Puerto Rico
(dry); 4, San Juan, Venezuela; 5, Morrocoy National Park, Venezuela; 6, Los Roques, Venezuela; 7, Ci?naga Grande, Colom-
bia; 8, Bocas del Toro, Panama; 9, Twin Cays, Belize; 10, northern Belize wetland complex; 11, Sian Ka?an, Mexico; 12,
Celestun Lagoon, Mexico; 13, Terminos lagoon, Mexico.
form of bilge water and garbage dumped
into the ocean (Atwood et al. 1987). In
addition, there is a significant amount of
pollution from nonpoint sources within
the region, including contaminants such as
sewage, solid-waste leachates from landfills,
industrial and agricultural runoff, and pe-
troleum products (Atwood et al. 1987).
These impacts threaten the biodiversity
and environmental quality of coastal
ecosystems in the region. The extent of
the loss of biological diversity is still poorly
known (Loreau et al. 2001), particularly in
the Caribbean, where species extinction
occurs at an accelerated rate. However, we
know that of the 197 endemic mammals
and birds across the Lesser and Greater
Antilles, at least 43 have become extinct
over the last 500 years (Myers et al. 2000).
It is estimated that in the last 150 years
eight species of vertebrates have become ex-
tinct in Jamaica alone, and more than 100
plant species indigenous to Trinidad and
Tobago may be threatened by extinction.
Despite much more widely publicized sup-
port for biodiversity conservation, the lack
of regional governmental and institutional
support for research suggests that trends 
of declining biological diversity in the
Caribbean will continue over the next 
several decades (Myers et al. 2000, Brooks
and Smith 2001). This trend is ongoing
despite the presence of 22 Ramsar Con-
vention sites and 20 coastal biosphere re-
serves located throughout the Caribbean
region (table 1).
There are few comprehensive, long-term
research programs aimed at understanding
the ecosystem services of coastal settings in
the Caribbean, despite the strong ecolog-
ical and economic importance of coastal
ecosystems to the region. Regional research
projects initiated in the 1970s and 1980s,
such as CARICOMP (Caribbean Coastal Marine Productiv-
ity) and CARIPOL (Caribbean Petroleum Pollution Moni-
toring Project), have provided useful information about
changes in the distribution, structure, and productivity of coral
reef, sea grass, and mangrove ecosystems as influenced by
petroleum pollution. Similarly, the Association of Island 
Marine Laboratories of the Caribbean, or AIMLC, has played
an important role in assisting and initiating collaborative 
research among its members (Ogden 1987). More recently, the
TEAM (Tropical Ecology, Assessment, and Monitoring) Ini-
tiative at Conservation International (www.conservation.
org/xp/CIWEB/home) has developed the first early-warning
system for global biodiversity, which will provide much-
needed knowledge about tropical ecosystems. This program
is designed to supply information necessary for assessing the
overall effectiveness of conservation efforts.
We have developed a conceptual framework to analyze
the diverse ecological properties of coastal settings in the
Caribbean region and evaluate the ecological vulnerability of
these ecosystems to a variety of human and natural distur-
bances. In creating this framework, we emphasized the 
biogeochemical properties and primary productivity of
the ecosystems that form the reef?sea-grass?wetland seascape.
Our approach follows the strategy developed by the Long 
Term Ecological Research (LTER) program of the National 
Science Foundation to establish research questions that are 
September 2004 / Vol. 54 No. 9 ?  BioScience 845
Articles
Table 1. Locations of selected Ramsar sites and biosphere reserves around the
wider Caribbean Sea.
Ramsar Biosphere 
Country or territory Ecosystem sitea reserveb
Guatemala Maya ?
Sierra de las Minas ?
Honduras Rio Pl?tano ?
Nicaragua Cayos Miskitos and coastal zone ?
Wetlands, Bluefield Bay ?
San Juan River ?
Costa Rica Northeast wetlands ?
Gandoca-Manzanillo ?
Cordillera Volc?nica Central ?
La Amistad ?
Panama Dari?n ?
La Amistad ?
Colombia Ci?naga Grande de Santa Marta lagoon ? ?
San Andres Seaflower ?
Venezuela Los Olivitos lagoon ?
Cuare ?
Los Roques ?
Restinga lagoon ?
Tacarigua Lagoon ?
Aruba and Netherlands Antilles ?
Saint Lucia ?
Guadeloupe ?
British Virgin Islands ?
Puerto Rico Gu?nica Commonwealth Forest ?
Luquillo Experimental Forest ?
North, Middle, and East ?
Caico Islands
Cayman Islands ?
Jamaica Black River ?
Cuba Zapata lagoon ? ?
Baconao ?
Buenavista ?
Cuchillas del Toa ?
Pen?nsula de Guanahacabibes ?
Sierra del Rosario ?
Mexico Ria Lagartos ?
Dzilam ?
Sian Ka?an ?
Calakmul ?
Florida (United States) Everglades and Dry Tortugas ? ?
a. See www.wetlands.org/RDB/neotropics/Caribbean.html.
b. See www.unesco.org/mab.
best studied over decadal time periods and large spatial areas
(Kaiser 2001, Hobbie et al. 2003). The sites described in this
article are national reserves, parks, or research areas from
the Caribbean region that are described as vulnerable or
heavily influenced by human activities. Ecological research at
some of these sites has been performed on a regular basis since
the 1970s. However, the long-term records of selected eco-
logical processes for these sites are inadequate to quantitatively
define the magnitude and direction of ecosystem change,
which limits the development of appropriate coastal man-
agement plans.
The ocean ecosystem of the wider Caribbean region
The Caribbean Sea is considered the smallest ocean basin and
classified as the second largest sea in the world (2,512,950
square kilometers [km2]) (figure 1; Dillon et al. 1987). Located
in the Western Hemisphere between North and South Amer-
ica, the Caribbean Sea is bordered on the north and east by
the West Indies archipelago, on the south by South America,
and on the west by the Central American isthmus and the 
Yucat?n Peninsula. All Central American countries except
El Salvador have extensive coastal areas along the Caribbean
Sea. Several oceanic islands are a major feature of the
Caribbean seascape, including Dominica,
Martinique, and the Barbados on the eastern
boundary; Cuba and Hispaniola (Haiti and
the Dominican Republic) to the north; and
Jamaica and the Cayman Islands to the north-
west. Although south Florida is not geolog-
ically part of the Caribbean Sea, it is
considered part of the region because of the
strong oceanic connection downstream
though east-west water circulation. The wider
Caribbean region thus includes coastal re-
gions in the Caribbean Sea proper and up-
stream areas in south Florida and the Yucat?n
Peninsula (including the Usumacinta?
Terminos lagoon complex).
The Caribbean Sea includes four major
basins, with approximately half of the water
deeper than 3600 meters (m) (Richards and
Bohnsack 1990). Many areas exceed 4000 m
in depth, with the deepest region?in the
Cayman Trench, located between Cuba and
Jamaica?ranging from 7535 to 7686 m.
Because of the active semi-independent
movement of the Caribbean plate, there are
earthquakes and volcanic activity throughout
the region (Dillon et al. 1987). Seasonal vari-
ation in temperature is small, ranging from
25 degrees Celsius (?C) in the winter to 28?C
in the summer. Ocean circulation in the 
region is strongly influenced by the trade
winds that form westward patterns, causing
upwelling along the South American conti-
nent. Water flows into the Caribbean Sea
mostly through the Grenada, St. Vincent, and St. Lucia pas-
sages in the southeast. The water continues westward as the
Caribbean Current, the main surface circulation in the
Caribbean Sea. The flow turns westward as it crosses the
Cayman Basin, and it enters the Gulf of Mexico as a narrow
boundary current through the Yucat?n Straits, where it forms
both the loop current in the Gulf of Mexico and the Gulf
Stream south of Florida (Fratantoni 2001). The average flow
of these currents is 24 million m3 per second, with near-
surface velocities approaching 80 to 150 centimeters per 
second in the Yucat?n Current (Coats 1992). This water 
circulation results in high connectivity among the different
basins.
Vast regions of the Caribbean are influenced by river 
discharge. Mean discharge to the region that includes the
coastal areas of Venezuela, Trinidad and Tobago, and Grenada
(figure 2) is more than 2100 km3 per year (Fekete et al. 2002),
but a significant volume of Amazon water is also delivered to
the Caribbean via the North Brazilian Current (Hellweger and
Gordon 2002). The two largest rivers of the region, the
Orinoco (1103 km3 per year; figure 2) and the Magdalena (228
km3 per year; figure 3) (Restrepo and Kjerfve 2000), account
for well over 60% of the fresh water discharged directly into
846 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles
Figure 2. Spatially mapped correlation coefficients between the monthly clima-
tology of Orinoco River (Venezuela) discharge and the ocean color (reciprocal
of monthly averaged water-leaving radiance at 412 nanometers, from SeaWiFS
[Sea-viewing Wide Field-of-view Sensor] data) between November 1997 and
October 2002. The region of high positive correlation shows that the hydro-
logical regime of the Orinoco is in phase with the ocean color field, suggesting 
a vast region of fluvial influence of the Orinoco?s highly colored waters. The 
adjacent region of negative correlation west of the plume is an area where 
upwelling-induced phytoplankton blooms tend to be out of phase with the
Orinoco discharge.
the Caribbean. The influence of the Orinoco
River plume on the ecology and biogeo-
chemistry of the western Caribbean extends
to areas close to Puerto Rico and the Do-
minican Republic, again demonstrating the
degree of connectivity in the Caribbean basin
(Salisbury et al. 2001). These rivers repre-
sent point sources for large fluxes of dis-
solved and particulate constituents (figures 
2, 3). These river constituents typically have
dramatic effects on the primary productiv-
ity of coastal waters. Turbid conditions lim-
iting productivity, with a zone of high
productivity at the distal end of turbid
plumes, have been reported by several 
investigators (Demaster et al. 1996).
Coastal ecosystems of the 
wider Caribbean region
Geomorphic features and geophysical ener-
gies of coastal settings (Thom 1982), along
with regional climate, have been identified as
environmental signatures that constrain the
structure and function of tropical coastal
ecosystems (Thom 1982, Twilley 1995). These
environmental signatures have been used to
generalize patterns across coastal geomor-
phological types, including reefs, lagoons,
estuaries (drowned river valleys), and deltas
(Downing et al. 1999). The vulnerability of
different coastal ecosystems to human dis-
turbance can be associated with these environmental signa-
tures. The combination of geophysical processes (e.g., river
discharge, wind, and water residence time) and biogeo-
chemical properties of a coastal setting provides a way to
summarize how stressors limit the structure and function of
coastal ecosystems. For example, changes in light availability
and nutrients are two key factors that determine both the type
and the rate of primary production among four geomopho-
logical types of coastal settings (figure 4).
Our approach provides a conceptual framework to 
generalize the impacts of land-use changes on ecosystem
health across different coastal settings in the Caribbean basin.
Ecosystem health is the condition in which ?an ecosystem is
active and maintains organization and autonomy over time
and is resilient to stress? (Costanza 1992). To characterize 
environmental signatures, we considered 10 categories,
including regional climate, geomorphological type, and 
anthropogenic characteristics (table 2). These categories have
been recognized as forcing functions that regulate the 
structure and function of coastal ecosystems (Thom 1982,
Twilley 1995).Values for each category are represented in nom-
inal and ordinal scales, since information is lacking across most
coastal settings (e.g., nutrient loading rates, sea-level rise). The
ordinal scales are ranges of values using established classifi-
cation systems. Another characteristic of each site is the 
initial effort to establish long-term ecological research. The
selected sites were grouped along this continuum of envi-
ronmental signatures based on geomorphological type 
(figure 4; Twilley 1995, Downing et al. 1999). The main 
assumption of this framework is that the susceptibility of
tropical coastal settings to eutrophication will vary along
this continuum of environmental signatures.
Availability of nutrient and light resources, along with 
water residence time, defines environmental signatures
from marine to river-dominated coastal settings (figure
4). The dual resource gradient (light and nutrients), along
with the variance in salinity, determines the diversity of
ecosystems in the coastal setting. The coastal seascape 
consists of reef, sea grass, and wetland ecosystems distrib-
uted across longitudinal and topographic gradients of any
coastal setting. The longitudinal gradient in salinity occurs
from tidal freshwater to euhaline environments. Within
each of these salinity regimes, there are topographic gra-
dients from submerged (subtidal) to emergent (intertidal
and floodplain) zones. Emergent zones include fringe (edge
of intertidal) to interior (supratidal) and floodplain re-
gions. Reefs and sea grasses colonize the submerged zone,
and wetlands (mangrove, marshes, and forested wetlands)
colonize the fringe, interior, and floodplain regions of the
emergent zone.
September 2004 / Vol. 54 No. 9 ?  BioScience 847
Articles
Figure 3. Spatially mapped correlation coefficients between the monthly clima-
tology of Magdalena River (Caribbean coast of Colombia) discharge and the
ocean color (monthly averaged water-leaving radiance at 555 nanometers
(nm), from SeaWiFS [Sea-viewing Wide Field-of-view Sensor] data) between
November 1997 and October 2002. The 555-nm band has been used as a proxy
for light scattering by particles (sediment and detritus). The region of high 
positive correlation suggests that the Magdalena River is the source of a large
sediment plume.
M
ag
da
le
na
 R
iv
er
The coastal seascape can be described by the ratio of emer-
gent to submerged area (wetland-to-water ratio). The relative
distribution of reef?sea-grass?wetland ecosystems across a
coastal seascape varies with the four geomorphological types
used in this classification of coastal settings, which can be
linked to specific combinations of geophysical energies. Reefs
and lagoons are marine-dominated coastal settings, with
coral reef and sea grass communities dominating the coastal
seascape. In these coastal settings, the emergent zone is lim-
ited to the intertidal area, with a reduced wetland-to-water ra-
tio. Light is plentiful in the submerged zone of this coastal
setting, yet nutrients are low, resulting in oligotrophic con-
ditions. In river-dominated coastal settings, coral reefs and sea
grasses are less abundant in the submerged zone, and wetlands
dominate the more extensive emergent zone. Depending on
the degree of freshwater input and tidal range, the ratio of
emergent to submerged zone increases in these coastal settings,
as in a delta, for example.
Many of the sites we reviewed have more than one geo-
morphological type, with multiple gradients resulting in di-
verse combinations of coastal ecosystems (figure 4). For
example, Terminos lagoon, in Mexico, is a delta-lagoon com-
plex that has extensive marshes and mangroves in its river-
dominated seascapes, in contrast with the extensive sea grasses
in its marine-dominated regions. The environmental signa-
ture concept describes how each of these ecosystems in a
coastal seascape has a characteristic structure and function that
can be related to available resources (e.g., nutrients), regula-
tors (e.g., salinity), and hydrologic gradients unique to a
coastal setting.
Environmental signature of selected coastal settings
The environmental signature of a coastal (regional) setting is
a combination of geomorphological type, geophysical ener-
gies, and levels of disturbance. There are conspicuous differ-
ences in topographic relief throughout the entire Caribbean
region.Areas with average altitudes from 2000 to 3000 m above
mean sea level (msl), associated with volcanic activity, can 
be found in Honduras, El Salvador, Costa Rica, Panama (e.g.,
Bocas del Toro; table 2), and the Lesser Antilles. The highest
altitude adjacent to the coastal zone in the Caribbean, located
in the Ci?naga Grande de Santa Marta deltaic region of
Colombia, is 5775 m above msl. In contrast, lower altitudes
(less than 500 m above msl) are geologically associated with
extensive karstic topography, as in the Yucat?n Peninsula
(Celestun, Sian Ka?an), Belize (Twin Cays), and south Florida
(the Everglades; table 2). Topographic relief classes were 
assigned to each site: low (e.g., the Everglades), medium (e.g.,
San Juan River), or high (e.g., Ci?naga Grande), depending
on the major regional geological features. Four of the 13 sites
are oceanic islands; two sites are on high islands and two on
low islands.
The combination of regional climate and topographic 
features influences the amount of freshwater input at each 
site. Although the Caribbean coast of Colombia is arid, the 
Ci?naga Grande has a deltaic geomorphology because of the
848 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles
Ta
bl
e 
2.
C
ha
ra
ct
er
is
ti
cs
 o
ft
he
 s
tu
dy
 s
it
es
.
P
re
ci
pi
ta
ti
on
-to
-
Fr
es
hw
at
er
To
po
gr
ap
hi
c
N
ut
rie
nt
Ty
pe
 o
f
ev
ap
ot
ra
ns
pi
ra
ti
on
M
ea
n 
ti
da
l
D
is
tu
rb
an
ce
Su
sc
ep
ti
bi
lit
y 
to
La
nd
sc
ap
e
Ec
os
ys
te
m
 
in
flo
w
re
lie
f
flu
x
M
in
er
al
og
y
la
nd
sc
ap
e
ra
ti
o
ra
ng
e
(h
ur
ric
an
es
)
se
a-
le
ve
l r
is
e
m
an
ag
em
en
t
Ev
er
gl
ad
es
,F
lo
rid
a
Lo
w
 t
o 
m
ed
iu
m
Lo
w
Lo
w
K
ar
st
ic
C
on
tin
en
ta
l
M
ed
iu
m
0.
5
M
ed
iu
m
M
ed
iu
m
R
es
to
ra
tio
n
Pu
er
to
 R
ic
o/
w
et
M
ed
iu
m
 (
riv
er
)
H
ig
h
M
ed
iu
m
N
on
ka
rs
tic
O
ce
an
ic
M
ed
iu
m
0.
5
H
ig
h
M
ed
iu
m
Pu
er
to
 R
ic
o/
dr
y
Lo
w
H
ig
h
Lo
w
N
on
ka
rs
tic
O
ce
an
ic
Lo
w
0.
5
M
ed
iu
m
H
ig
h
S
an
 J
ua
n,
Ve
ne
zu
el
a
H
ig
h 
(ri
ve
r)
M
ed
iu
m
H
ig
h
N
on
ka
rs
tic
C
on
tin
en
ta
l
H
ig
h
3
Lo
w
Lo
w
M
or
ro
co
y 
N
at
io
na
l P
ar
k,
Lo
w
Lo
w
Lo
w
K
ar
st
ic
C
on
tin
en
ta
l
M
ed
iu
m
0.
5
Lo
w
H
ig
h
R
eh
ab
ili
ta
tio
n
Ve
ne
zu
el
a
Lo
s 
R
oq
ue
s,
Ve
ne
zu
el
a
Lo
w
Lo
w
Lo
w
K
ar
st
ic
O
ce
an
ic
Lo
w
0.
5
Lo
w
H
ig
h
C
i?
na
ga
 G
ra
nd
e,
C
ol
om
bi
a
H
ig
h 
(ri
ve
r)
H
ig
h
H
ig
h
N
on
ka
rs
tic
C
on
tin
en
ta
l
Lo
w
0.
5
Lo
w
M
ed
iu
m
R
eh
ab
ili
ta
tio
n
B
oc
as
 d
el
 T
or
o,
Pa
na
m
a
Lo
w
H
ig
h
Lo
w
N
on
ka
rs
tic
C
on
tin
en
ta
l
H
ig
h
0.
25
Lo
w
a
M
ed
iu
m
Tw
in
 C
ay
s,
B
el
iz
e
Lo
w
Lo
w
Lo
w
K
ar
st
ic
O
ce
an
ic
Lo
w
0.
25
M
ed
iu
m
H
ig
h
N
or
th
er
n 
B
el
iz
e 
w
et
la
nd
Lo
w
 t
o 
m
ed
iu
m
Lo
w
Lo
w
K
ar
st
ic
C
on
tin
en
ta
l
M
ed
iu
m
?
M
ed
iu
m
H
ig
h
co
m
pl
ex
S
ia
n 
K
a?
an
,M
ex
ic
o
H
ig
h 
to
 m
ed
iu
m
Lo
w
M
ed
iu
m
K
ar
st
ic
C
on
tin
en
ta
l
Lo
w
0.
5
Lo
w
Lo
w
C
el
es
tu
n 
La
go
on
,M
ex
ic
o
M
ed
iu
m
 t
o 
hi
gh
Lo
w
Lo
w
K
ar
st
ic
C
on
tin
en
ta
l
M
ed
iu
m
0.
5
M
ed
iu
m
H
ig
h
Te
rm
in
os
 la
go
on
 
M
ed
iu
m
 t
o 
hi
gh
Lo
w
H
ig
h
N
on
ka
rs
tic
C
on
tin
en
ta
l
M
ed
iu
m
0.
5
Lo
w
M
ed
iu
m
a.
V
ol
ca
ni
c 
ac
ti
vi
ty
.
Magdalena River, a large freshwater subsidy to the region.Arid
regions of the Yucat?n Peninsula receive extensive groundwater
flow in the Sian Ka?an and Celestun Lagoons (Herrera-Silveira
and Comin 2000). More moist regions include the San Juan
River estuary in Venezuela, Bocas del Toro in Panama,
Terminos lagoon in Mexico, and the Everglades watershed in
Florida (table 2). Puerto Rico is a high island with extreme
differences in precipitation, representing arid and moist sites
at similar geographical locations. Although the tidal regime
across sites is small (0.2?0.5 m, except in San Juan River 
estuary; table 2), water residence time differs among the sites
depending on the magnitude of river discharge, as in the
Ci?naga Grande de Santa Marta, Colombia (7?11 days), and
Celestun Lagoon, Mexico (25 days). We define a fertility 
gradient across these geomorphological types as associated
with nutrient input, residence time, and light regime.
The frequency and magnitude of disturbance, both natural
and human, are also important considerations in the envi-
ronmental signature of a coastal setting. Hurricanes have a
strong effect on the ecological conditions of the Caribbean 
region. Although hurricane impacts are generally greater in
oceanic (e.g., Puerto Rico, Twin Cays, Belize) than in conti-
nental seascapes (e.g., San Juan,Venezuela; Sian Ka?an, Mex-
ico; table 2), the latter can also receive considerable damage.
September 2004 / Vol. 54 No. 9 ?  BioScience 849
Articles
Figure 4. Conceptual framework of the interaction between the amount of light, nutrient in-
puts, and water residence time across different types of estuaries, based on geomorphology
from carbonate reefs to river-dominated deltas. The horizontal arrows indicate an increase
or decrease of sediment and nutrient inputs, wetland-to-estuary ratio, and benthic and
pelagic productivity. The presence or absence of mangroves, sea grasses, and coral reefs in
each coastal setting is indicated by the bars, as are the influences of fresh water and plankton
productivity. Numbers representing study sites were placed along the bars depending on the
dominant ecosystem found at each site: 1, Everglades, Florida; 2, Puerto Rico (wet); 3, Puerto
Rico (dry); 4, San Juan, Venezuela; 5, Morrocoy National Park, Venezuela; 6, Los Roques,
Venezuela; 7, Ci?naga Grande, Colombia; 8, Bocas del Toro, Panama; 9, Twin Cays, Belize;
10, northern Belize wetland complex; 11, Sian Ka?an, Mexico; 12, Celestun Lagoon, Mexico;
13, Terminos lagoon, Mexico. Modified from Downing and colleagues (1999).
For example, Hurricane Andrew hit southern Florida in 1992,
affecting extensive urban areas and mangrove wetlands (Smith
et al. 1994). Hurricane Mitch, in 1998, was a category 5 storm
when it hit coral reefs, mangrove forests, and urban centers
in Honduras, Nicaragua, and Guatemala. Forecasting the
condition of ecosystem functions, ecosystem services, and nat-
ural resource capital in the Caribbean region requires knowl-
edge of large disturbances such as El Ni?o?Southern
Oscillation (ENSO) events and hurricanes. These forcing
functions have spatially specific effects on coastal settings, in-
creasing the diversity in ecosystem structure and function
across the coastal seascape. Since the 1970s, ENSO events
have occurred more frequently and persisted longer (Trenberth
and Hoar 1997). Temporal patterns in precipitation are
strongly modulated by episodic events such as hurricanes
and ENSO. Because these global phenomena are controlled
by climate change, the frequent occurrence of extreme climate
events in the Caribbean could be more relevant than the
fluctuations of the mean climate (Stenseth et al. 2002).
Impacts of sea-level rise on coastal settings are better as-
sessed using regional values rather than global averages
(Church 2001). Most intertidal wetlands in south Florida
and the Caribbean can survive the present regional rates of
sea-level rise, which range from 1.2 to 2.3 millimeters per year.
The sustainability of these intertidal wetlands relies on migrat-
ing inland to areas of decreasing tidal inundation along 
undeveloped shores. However, activities associated with 
urbanization, such as tourism, infrastructure, and agriculture,
can remove supratidal regions, restricting the migration of in-
tertidal wetlands in response to climate change. An increase
in sea level alone would not be enough to ?drown?coral reefs
(Kleypas et al. 2001), but the combination of sea-level rise and
decreased water quality (e.g., increase in nutrient enrich-
ment and turbidity) would result in flooded coastlines that
could affect many coral reefs and change reef distribution at
local and regional scales. Thus, sea-level rise in combination
with eutrophication and increasing sea surface temperatures
has the potential to modify reef distribution not only in the
Caribbean region but globally (Kleypas et al. 2001). Caribbean
countries, like other small island and coastal areas, face dif-
ficult decisions in confronting the adverse effects of global cli-
mate change and associated sea-level rise. Consequently, the
vulnerability of coastal resources and infrastructure to sea-level
rise increases constantly, highlighting the urgent need for an
integrated framework for coastal management (Twilley et
al. 2001).
Ecological vulnerability across 
environmental signatures
The 13 sites in this overview illustrate the ecological vulner-
ability of the coral reef?sea-grass?wetland seascape across
different environmental signatures in coastal settings. Areas
with different levels of disturbance need to be included in
ecosystem assessment to advance our understanding of the
trajectories that ecosystems follow after major changes 
attributable to human effects, natural causes, and their 
interaction (Vitousek et al. 1997). Our selected sites across the
wider Caribbean include different degrees of human impact
on the structure and function of ecosystems in diverse envi-
ronmental signatures. The management variable (table 2)
refers to projects that have been implemented to rehabilitate
and restore coastal ecosystems at the coastal seascape level.
Coral reefs are one of the most conspicuous coastal ecosys-
tems in the Caribbean Sea, where the second longest barrier
reef in the world and largest in the Northern Hemisphere is
located off the coast of Belize (Hughes 1994). Recent evalu-
ations of the global decline in reef area describe the western
Atlantic Ocean and the Caribbean Sea as the regions where
the most serious losses are occurring (Gardner et al. 2003).
The reefs of the Caribbean and adjacent coastal waters con-
stitute about 12% of the global total and are major indicators
of the environmental stress in the region (Wilkinson 2000).
Degradation and destruction of reefs are caused by sedi-
mentation, destructive fishing, poorly regulated mining and
construction, and anthropogenic nutrient inputs (Szmant
2002). Current estimates indicate that 29% of the Caribbean
reef areas are considered at high risk because of increased
runoff and sedimentation caused by deforestation, nutrient
contributions from hotel and shipping wastes, coastal con-
struction, and mining.
Studies in the selected sites and throughout the Caribbean
region have focused on monitoring changes in reef extension,
in species composition, and in the frequency of events such
as bleaching, herbivory, and invasion by fleshy macroalgae
(Aronson et al. 2002). The sites with large areas of coral reef
cover are Everglades National Park, Florida; Morrocoy National
Park (Parque Nacional Morrocoy) and Los Roques,Venezuela;
Bocas del Toro, Panama; Twin Cays, Belize; Celestun Lagoon,
Mexico; and Sian Ka?an, Mexico (table 2).
Considerable efforts in the wider Caribbean have been
dedicated to evaluating the damage caused by coral diseases
such as white band, and by coral bleaching due to an in-
crease of 1?C to 2?C in sea surface temperatures above mean
monthly summer values (Skirving and Guinott 2001). A cur-
rent paradigm suggests that a fixed bleaching temperature
threshold of 1?C to 2?C above the present average sea tem-
perature needs to be reevaluated, since a broad spectrum of
responses indicates that the temperature threshold changes
over time as a result of acclimation and evolution (Hughes et
al. 2003). Current reports suggest a massive regionwide de-
cline of corals across the Caribbean basin, with the ?average
hard coral cover on reefs being reduced by 80%, from about
50% to 10% cover, in three decades?(Gardner et al. 2003).Yet
it is still not clear what the relative roles of climatic and local
human factors are, and how they interact in determining the
long-term trajectories of reef decline (Buddemeier et al.
2003).
Recent studies on species turnover in the Belize Barrier Reef
have shown a significant shift of coral reef species over hun-
dreds of square kilometers in less than a decade (Aronson 
et al. 2002). These results confirm the short temporal scale 
on which human activities affect coral reef habitats. For 
850 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles
example, Hughes (1994) discussed the role of overfishing
and reduced herbivory as the main factors inducing 
dramatic shifts from coral-dominated to algae-dominated 
systems in Jamaica. According to new surveys, all of the 
classic reef zonation patterns found in Jamaica in the 1950s
have disappeared. Documented structural and functional
changes in Jamaican reefs illustrate how overfishing, her-
bivory, and large physical disturbances (e.g., hurricanes) in-
teract to modify reef communities in the Caribbean region
at the coastal seascape level. More recently, Lapointe (1997)
argued that in addition to these causes, studies of macroalgal
blooms need to evaluate how excess nutrient concentrations
in water columns deteriorate reef areas. Documentation of
nutrient enhancement in the Caribbean is scarce, despite the
environmental stress caused by eutrophication (Aronson
and Precht 2000). Because of the lack of long-term data on
nutrient enrichment in coral reef habitats, controversies 
remain about the relative importance of eutrophication and
herbivory in promoting macroalgae blooms in coral reefs
(Hughes et al. 1999, Lapointe 1999, McCook 1999, Aronson
and Precht 2000, Szmant 2002).
The shallow-water communities of Morrocoy National
Park, Venezuela, provide a good example of the complexity
involved in assessing the interaction of large-scale distur-
bances (human and natural), overfishing, herbivory, and
eutrophication on coral reef habitats in the wider Caribbean
(table 2). The 320-km2 park consists of semienclosed, inter-
connected embayments (Bone et al. 1998). Early in 1996,
mass mortalities were recorded for coral reefs (60%?90%),
gastropods, sipunculans, polychaetes, and sponges, following
unusually low temperatures apparently caused by upwelling
of cold water along the Venezuelan coast (Laboy-Nieves et al.
2001). Evaluation of coral recovery 3 years after the distur-
bance shows that there are still large extensions of dead reefs
and of reefs covered by algae and sand (Villamizar 2000).
The disturbance of reef communities associated with this
upwelling event along the Venenzuelan coast occurred along
with increased human impacts in the region. These included
urban and industrial development around the park, sewage
inputs from tourism facilities, unregulated recreational and
commercial fishing, and coral bleaching and other diseases
(Villamizar 2000). These impacts, along with the large inputs
of sediments, nutrients, and organic matter transported by
rivers during the rainy season, all slowed the recovery of
coral reefs. Sea temperature was probably the trigger that
initiated the massive mortalities (Laboy-Nieves et al. 2001),
but lower temperatures increased the chance of mortality for
already stressed coral reef habitats and associated commu-
nities. Additional die-offs were observed at the end of 1996,
when the highest precipitation in 28 years was registered.
Low salinity and hypoxic conditions triggered mass mortal-
ities of fishes, holothurians, sea urchins, and sea stars, along
with a significant reduction in the area of sea grass (Laboy-
Nieves et al. 2001). The high precipitation was apparently 
associated with the 1995?1996 ENSO, an event that further
confounded the identification of causes for the massive coral
mortality in this park system.
Sea grasses are also affected by excessive nutrient inputs
(Duarte 1995). Regional differences in nutrient availability 
determine the distribution of different sea grass species along
salinity gradients (Fourqurean et al. 2001). High nutrient
loading causes algal blooms, which increase turbidity and 
reduce the photosynthetic efficiency and growth of sea grasses.
The excess nutrients also result in greater growth and pro-
duction of epiphytes, to the extent of accelerating sea grass
dieback. Phosphorus (P) has been identified as a limiting 
nutrient in karstic substrates, so an increase in this nutrient
will certainly increase sea grass biomass in the short term
(Jensen et al. 1998).
Some sea grass species have a narrow tolerance for salin-
ity changes, which can trigger major shifts in species com-
position (Lirman and Cropper 2003). Organisms are affected
more by extreme salinities during major disturbances than by
values observed under average environmental conditions.
Salinity in association with nutrient enrichment can also 
become a stressor when freshwater inputs are drastically re-
duced. For example, the mass mortality of sea grasses in
Florida Bay has been attributed to long-term salinity stress
caused by reduced freshwater flow. Moreover, studies show that
salinity stress can exacerbate susceptibility to other factors 
(e.g., pathogens, temperature), leading to extensive die-offs.
Studies in Florida Bay have been critical in understanding how
changes in salinity and P limitation affect species composi-
tion and productivity patterns (Fourqurean et al. 1995).
Florida Bay studies have also evaluated the physiological re-
sponse of sea grasses to stress resulting from iron deficiency,
low oxygen concentrations, and drastic reductions in salin-
ity. Computer simulation models for sea grasses are being 
developed for Florida Bay; they incorporate different hier-
archical levels, from physiological to spatially explicit seascape
processes (Fong et al. 1997).
The high biological diversity found in the coral reefs of the
Caribbean Sea is strongly influenced by the presence of man-
grove forests and sea grasses. These three ecosystems form
strongly coupled habitat complexes, which are not com-
pletely understood, along the coastal seascape (Twilley et al.
1998, Koch and Madden 2001, McKee et al. 2002). There is a
continuum across these ecosystems in which complex nutrient
exchanges define the spatial and temporal distribution of
mangroves, sea grasses, and coral reefs. Nutrient availability
is highly controlled by nutrient recycling in mangrove forests
and sea grasses (Duarte 1995, Rivera-Monroy and Twilley
1996, Feller et al. 2003a). The resulting high rates of primary
production and organic material production sustain complex
trophic food chains. Therefore, negative impacts in one
ecosystem can cascade across the coastal seascape, affecting
other areas. For example, mangrove deforestation in the
coastal zone can significantly reduce the productivity of sea
grasses and coral reefs by causing excessive sediment loads that
increase turbidity.
September 2004 / Vol. 54 No. 9 ?  BioScience 851
Articles
The total mangrove area for the wider Caribbean region is
estimated at 25,882 km2 (Spalding et al. 1997), representing
about 50% of the total mangrove area in the Neotropics.
The physiognomy of mangrove forests in the Caribbean re-
gion is diverse, from scrub forests (less than 1.5 m tall) in the
interior of islands to well-developed forests (more than 30 m
tall) in river-dominated coastal seascapes. These gradients pro-
vide insights into the multiple stressors that can affect man-
grove forest growth. Mangrove forests growing under
nutrient-limited conditions respond quickly to fertilization.
Recent studies in karstic environments in the Caribbean
(Twin Cays, Belize; table 2) have found that nitrogen (N) and
P are not uniformly distributed within mangrove ecosys-
tems (Feller et al. 2003b). At Twin Cays, both N and P en-
richment significantly increased productivity along a tidal
gradient; trees were generally P limited in the interior zones
of the forest but N limited in the fringe. This response is as-
sociated with low P availability in carbonate-rich sediments.
Nutrient-enrichment field experiments, like the one per-
formed in Belize, are limited in the Caribbean, although it ap-
pears that mangrove forests in Bocas del Toro, Panama, are
also prone to P limitation (Lovelock et al. 2004). Chen and
Twilley (1999) showed how patterns of forest development in
Shark River, Florida, differ along a fertility gradient defined
by total P concentration. This fertility gradient is apparently
caused by greater inputs of mineral material (enriched with
P) at the mouth of the estuary, where mangrove trees grow
significantly larger (more than 10 m tall) than at the head of
the estuary (less than 5 m tall). The fact that the main P
source at this site originates at the mouth of the estuary un-
derscores the importance of nutrient sources for mangrove
development.
Rehabilitation of ecosystems in 
the wider Caribbean region 
A rehabilitation project in the Ci?naga Grande de Santa
Marta, Colombia, and a restoration project in the Florida
coastal Everglades are the largest such projects in the coastal
Neotropics (table 2). The degradation of water quality,
alteration of hydrological regime, and loss of species and
habitat in both coastal regions are the main conditions that
prompted the planning, development, and financial support
of these projects. The Ci?naga Grande region provides a dra-
matic example of negative anthropogenic effects; the loss of
an extensive mangrove area (about 350 km2) and the reduc-
tion or disappearance of important commercial fish stocks
have increased the region?s poverty and social unrest.
Hydrological alteration in the Ci?naga Grande was the result
of the construction of roads and levees in the 1950s to improve
transportation and ranch farming. In the Everglades, an ex-
tensive canal system was developed in the 1940s to optimize
water management for agriculture practices, urban devel-
opment, and flood protection (Davis et al. 1994). Negative 
environmental impacts in the Everglades include massive sea
grass die-offs, coral reef mortality, low water quality, noxious
algae blooms, habitat fragmentation, and major changes in
vegetation distribution and diversity that have altered the
density and spatial dispersion of wild and commercial animal
species (Davis et al. 1994). Freshwater rediversion is the main
restoration measure in both the Ci?naga Grande and the
Everglades, but the geomorphological differences between the
two sites (figure 4) make the outcome of both projects com-
paratively uncertain. Because there have been no previous
long-term projects of this type, it is unclear what the actual
ecological response, and the effectiveness of restoration 
measures, will be.
The overall objective of the Ci?naga Grande rehabilitation
project was to reduce soil salinity by reconnecting the flood-
plain to the Magdalena River through five dredged channels
in areas of historical distributaries (Botero and Salzwedel
1999). All five planned channels were dredged by 1998, and
mangrove recovery was apparent after the practical salinity
level dropped from more than 100 to 40?50, a range normally
observed in mangrove forests (Rivera-Monroy et al. 2004). In
the Everglades, restoration aims to redivert fresh water to al-
low a more natural overland flow (sheet flow) through the
freshwater marshes of Taylor and Shark River sloughs to the
estuaries and coastal zone. The full Everglades restoration plan
will be completed over the next 10 to 30 years; some hydro-
logic structures have already been removed, and freshwater
flow has been restored in some regions.
Research questions to test the environmental 
signature hypothesis 
We propose a series of questions describing the vulnerabil-
ity of coastal ecosystems to human and natural disturbance
across the coral reef?sea-grass?wetland coastal seascape,
based on findings at each of our research sites (figure 5).
The vulnerability of this coastal seascape depends on the en-
vironmental signatures of tropical coastal settings. These
questions follow our main hypothesis that the susceptibility
of tropical coastal systems to disturbance will vary the envi-
ronmental signature of coastal settings. The questions are
grouped by ecosystem to facilitate their location in our con-
ceptual framework, but ecosystems should be viewed as
closely interconnected to each other in a seascape. Figure 5 in-
dicates the location of the ecosystems and illustrates their hy-
drologic connectivity. For example, sea grasses are absent on
deltaic systems because of high sediment loads (figure 4). As
mentioned above, the study sites listed in table 2 represent dis-
crete points across the range of coastal settings, allowing an
intersite comparison of factors that control nutrient cycling
and primary productivity. Thus, multiple questions could
be tested simultaneously at multiple sites, which represent a
combination of environmental factors. Rather than trying to
establish an exhaustive list of questions, we focused on the re-
search problems that integrate different ecological processes
at different temporal scales along an environmental signature,
from uplands to oligotrophic coastal waters.
The general research questions that encompass the envi-
ronmental signature concept are as follows:
852 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles
? How do long-term
changes in freshwater
flow control the magni-
tude of nutrient and
organic-matter inputs to
the estuarine zone? 
? How will human-
induced changes in
nutrient enrichment,
together with these
changes in hydrology,
affect patterns and mag-
nitudes of primary and
secondary production in
diverse types of coastal
settings? 
? How will these direct
impacts by cultural
eutrophication interact
with long-term changes
in climate drivers (sea-
level rise, hurricanes,
fires) to modify ecolog-
ical patterns and pro-
cesses across different
coastal settings? 
Any application of the
approach and research
agenda proposed here to
carry on long-term ecolog-
ical studies in the Caribbean
will need to take into con-
sideration the financial re-
quirements of the study and
the strong cultural and po-
litical differences among all
nations forming the wider
Caribbean region. One of
the major impediments to
reducing environmental
degradation of the Carib-
bean Sea is the complex
process needed to establish
a regional management
plan for marine resources.
Richards and Bohnsack
(1990) pointed out more
than a decade ago that the
real problem was the lack
of coordinated support
among the 36 to 40 Caribbean nations and territories for mon-
itoring the rich and diverse ecosystems in the region. Some
of the major impediments to such coordination are the wide
cultural, educational, and economic differ-
ences among these countries. Of particular
importance are the diverse management poli-
cies (or complete lack thereof) that each coun-
try has developed, based on economic
priorities that do not embrace sustainable 
development.
One of the major priorities for sustain-
able management of the rich diversity of eco-
systems in the wider Caribbean is to develop
cost-effective research programs addressing
common management issues in different 
locations. Cross-domain approaches and col-
laborations could foster the development of analytical and
experimental tools to address common environmental prob-
lems in the region. The approach and research agenda 
September 2004 / Vol. 54 No. 9 ?  BioScience 853
Articles
Freshwater and brackish marshes
1. How do differences in freshwater flow modify the spatial 
distribution of wetlands in the upper regions of the watershed?
2. What is the nutrient (phosphorus and nitrogen) removal 
capacity of wetlands around agricultural and urban areas?
3. What changes in species community composition and pro-
ductivity occur in response to an increase in nutrient loads?
4. How does the quality and quantity of organic matter 
change along a hydrological gradient from freshwater 
marshes to mangrove forests?
Mangroves
1. Does aboveground (wood, litterfall) and belowground 
productivity change in response to differential nutrient 
enrichment (phosphorus versus nitrogen)?
2. Are mangrove species-specific growth rates modified by 
changes in hydroperiod and along salinity gradients, and 
what physiological mechanisms are developed or optimized 
to maintain these growth rates?
3. How fast are productivity and nutrient flux patterns within 
mangrove forests reestablished after major human (e.g., fresh-
water diversions) and natural (e.g., hurricanes) disturbances?
4. Does the relative importance of organic matter exported from 
mangrove forests as a major food source to sustain secondary 
production vary across geomorphological settings?
5. How do accretion and elevation rates in mangrove forests vary 
in response to change in sea-level rise across the Caribbean?
Figure 5. Proposed research questions along an 
upstream-downstream salinity gradient encom-
passing marshes, mangroves, sea grasses, and 
coral reefs. (continued on next page)
Sea grasses
1. What is the minimum nutrient loading rate that triggers 
the replacement of sea grass?dominated systems with 
phytoplankton-dominated systems? Is the magnitude of
this rate similar across different geomorphological 
settings?
2. How are the growth and reproduction rates of sea grass 
modified as nutrient availability increases? Are species-
specific responses related to site-specific hydrological and 
salinity regimes?
3. What is the role of the interaction between sediment 
oxygen concentrations and nutrient availability in con-
trolling the spatial distribution and mortality of sea grass?
4. How are the growth rate and spatial distribution of sea 
grass affected when mangrove forests are removed?
5. How fast do sea grasses recover after major natural 
(hurricanes) and human (nutrient enrichment) 
disturbances?
presented in this article aim to strengthen research initiatives
already present in the region, such as CARICOMP. We envi-
sion our research questions as a template to organize and syn-
thesize current ecological information and design system-level
ecological research in the long term. Accomplishing these
tasks will require the active involvement of organizations
such as the International Long Term Ecological Research
(ILTER) network. The ILTER network can help to consolidate
and organize regional research efforts with collaboration and
support from the US LTER program, in addition to other 
international organizations (e.g., UNESCO?s MAB [Man 
and the Biosphere] Programme or the LOICZ [Land-Ocean
Interactions in the Coastal Zone] Project).
Future strategies to advance research on the coastal and
oceanic resources of the Caribbean region will need the 
active participation of scientists, not only in posing ecolog-
ically relevant questions but also in designing multi- and 
interdisciplinary studies in which political and social issues
are major components (Balmford et al. 2002). This is partic-
ularly true given the diverse geopolitical borders and economic
importance of the region. We believe that because of the
economic value, large-scale interconnectivity, and complex-
ity of the ecological processes in the wider Caribbean re-
gion, finding solutions to the area?s environmental problems
within the next 10 to 15 years represents a major challenge to
estuarine and coastal science.
Acknowledgments
This work was supported by funding from the National 
Science Foundation (NSF) through supplemental funds 
provided by the LTER Network Office (University of New
Mexico) and the Florida Coastal Everglades (FCE) LTER
program (grant no. DEB-9910514). Funding by the Insti-
tuto Colombiano para el Desarrollo de la Ciencia y la 
Tecnolog?a (COLCIENCIAS; code no. 2105-13-080-97) to 
V. H. R.-M., R. R. T., and E. M., and by NSF (DEB-9981535)
to I. C. F., contributed to the preparation of the manuscript.
This work developed from workshops held at Florida Inter-
national University (FIU) in March 2001 and at the Center
for Ecology and Environmental Technology, University of
Louisiana at Lafayette (ULL), in March 2002. We wish to
thank Stephen E. Davis III (Texas A&M University), James
Fourqurean (FIU), and Joel C. Trexler (FIU) for their par-
ticipation and suggestions during the workshop at FIU.
Christopher Madden (South Florida Water Management
District [SFWMD]) and Jennifer H. Richards (FIU) con-
tributed ideas and suggestions for information analyses in both
workshops and during manuscript preparation. The manu-
script greatly benefited from comments by Ernesto Medina
(Instituto de Investigaciones Cient?ficas and Universidad
Central de Venezuela), Carl Fitz (SFWMD), Paula Spiniello
(Instituto de Zoolog?a Tropical, Facultad de Ciencias, Uni-
versidad Central de Venezuela), and Barbara Hasek (ULL). The
authors acknowledge their respective institutions for allow-
ing time to travel to the workshops.We want to thank Ed Prof-
fitt and Ken Krauss from the USGS National Wetland Research
Center, and Paola Reyes, Edward Casta?eda, and Osvaldo
Perez from ULL, for their help hosting the workshop in
Lafayette. Special thanks to Pam Griego (LTER Network 
Office) and Mike Rugge (FCE LTER program) for help with
travel arrangements, and to Vaclav Hasek for preparing 
figures 1 and 5. This article was greatly enhanced by the 
critical reviews of Stephen V. Smith and two anonymous 
reviewers. This is SERC (Southeast Environmental Research
Center) and Smithsonian Environmental Research Center
contribution 230 and 681, respectively.
References cited
Aronson RB, Precht WF. 2000. Herbivory and algal dynamics on the coral
reef at Discovery Bay, Jamaica. Limnology and Oceanography 45:
251?255.
Aronson RB, Macintyre IG, Precht WF, Murdoch TJT, Wapnick CM. 2002.
The expanding scale of species turnover events on coral reefs in Belize.
Ecological Monographs 72: 233?249.
854 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles
1. Can a significant increase in nutrient (phosphorus and 
nitrogen) loading into coastal areas with extensive coral 
reefs lead to coral mortality because of excessive macro-
algal growth?
2. What is the interactive role of nutrient flux increases 
(phosphorus and nitrogen) and herbivory in maintaining 
coral reef areas and reducing macroalgal colonization?
3. What is the effect of high sediment load in controlling 
herbivory and macroalgal abundance in coral reefs?
4. How fast can coral reefs recover from large-scale (e.g.,
hurricanes, sedimentation) and local (e.g., coral bleach-
ing) disturbances? And how do human impacts affect the 
rate of recovery?
5. Are there significant long-term differences in nutrient 
loading in coral reef areas across the Caribbean?
Coral reefs
Figure 5. (continued)
Atwood DK, Burton FJ, Corredor JE, Harvey GR, Mata-Jimenez AJ,Vasquez-
Botello A, Wade BA. 1987. Cooperative coastal ecology at Caribbean 
Marine Laboratories. Oceanus 30: 25?32.
Balmford A, et al. 2002. Economic reasons for conserving wild nature.
Science 297: 950?953.
Bone D, Perez D, Villamizar A, Penchaszadeh PE. 1998. Parque Nacional 
Morrocoy, Venezuela. Pages 151?159 in B. Kjerfve, ed. CARICOMP?
Caribbean Coral Reef, Seagrass and Mangrove Sites. Paris: United Na-
tions Educational, Scientific and Cultural Organization.
Botero L, Salzwedel H. 1999. Rehabilitation of the Ci?naga Grande de Santa
Marta, a mangrove-estuarine system in the Caribbean coast of Colom-
bia. Ocean and Coastal Management 42: 243?256.
Brooks T, Smith ML. 2001. Caribbean catastrophes. Science 294: 1469?1471.
Buddemeier RW, Ware JR, Gardner TA, C?t? IM, Gill JA, Grant A, Watkin-
son AR. 2003. Coral reef decline in the Caribbean. Science 302: 391?393.
Chen R, Twilley RR. 1999. Patterns of mangrove forest structure and soil 
nutrient dynamics along the Shark River Estuary, Florida. Estuaries 22:
955?970.
Church JA. 2001. Climate change: How fast are sea levels rising? Science 294:
802?803.
[CIA] US Central Intelligence Agency. 2004. The World Factbook. (25 July
2004; www.cia.gov/cia/publications/factbook/)
Coats DA. 1992. The Loop Current. Chapter 6 in Milliman JD, Imamura E,
eds. Physical Oceanography of the U.S. Atlantic and Eastern Gulf of
Mexico. Herndon (VA): US Department of the Interior.
Costanza R. 1992. Toward an operational definition of ecosystem health. Pages
239?256 in Costanza R, Norton BG, Haskell BD, eds. Ecosystem Health:
New Goals for Environmental Management. Washington (DC): Island
Press.
Davis SM, Gunderson LH, Park WA, Richardson JR, Mattson JE. 1994.
Landscape dimension, composition, and function in a changing Ever-
glades ecosystem. Pages 419?444 in Davis SM, Ogden JC, eds. Everglades:
The Ecosystem and Its Restoration. Delray Beach (FL): St. Lucie Press.
Demaster DJ, Smith WOJ, Nelson DM, Aller JY. 1996. Biogeochemical
processes in Amazon shelf waters: Chemical distributions and uptake rates
of silicon, carbon and nitrogen. Continental Shelf Research 16: 617?628.
Dillon WP, Edgar NT, Scanlon KM, Klitgord KD. 1987. Geology of the
Caribbean. Oceanus 30: 42?52.
Downing JA, et al. 1999. The impact of accelerating land-use change on the
N-cycle of tropical aquatic ecosystems: Current conditions and pro-
jected changes. Biogeochemistry 46: 109?148.
Duarte CM. 1995. Submerged aquatic vegetation in relation to different 
nutrient regimes. Ophelia 41: 87?112.
Fekete BM,V?r?smarty CJ, Grabs W. 2002. High-resolution fields of global
runoff combining observed river discharge and simulated water balances.
Global Biogeochemical Cycles 16 (3): 1042.
Feller IC, Whigham DF, McKee KL, Lovelock CE. 2003a. Nitrogen limitation
of growth and nutrient dynamics in a disturbed mangrove forest, Indian
River Lagoon, Florida. Oecologia 134: 405?414.
Feller IC, McKee KL,Whigham DF, O?Neill JP. 2003b. Nitrogen vs. phosphorus
limitation across an ecotonal gradient in a mangrove forest. Bio-
geochemistry 62: 145?175.
Fong P, Jacobson ME, Mescher MC, Lirman D, Harwell MC. 1997. Investi-
gating the management potential of a seagrass model through sensitiv-
ity analysis and experiments. Ecological Applications 7: 300?315.
Fourqurean JW, Powell GVN, Kenworthy WJ, Zieman JC. 1995. The effects
of long-term manipulation of nutrient supply on competition between
the seagrasses Thalassia testudinum and Halodule wrightii in Florida
Bay. Oikos 72: 349?358.
Fourqurean JW, Willsie A, Rose CD, Rutten LM. 2001. Spatial and tempo-
ral pattern in seagrass community composition and productivity in
south Florida. Marine Biology 138: 341?354.
Fratantoni DM. 2001. North Atlantic surface circulation during the 1990?s
observed with satellite-tracked drifters. Journal of Geophysical Research
106: 22067?22093.
Gardner TA, C?t? IM, Gill JA, Grant A, Watkinson AR. 2003. Long-term 
region-wide declines in Caribbean corals. Science 301: 958?960.
Harvell CD, et al. 1999. Emerging marine diseases?climate links and 
anthropogenic factors. Science 285: 1505?1510.
Hellweger FL, Gordon AL. 2002. Tracing Amazon River water into the
Caribbean Sea. Journal of Marine Research 60: 537?549.
Herrera-Silveira JA, Comin FA. 2000. An introductory account of the types
of aquatic ecosystems of Yucat?n Peninsula (SE Mexico). Pages 213?227
in Munawar M, Lawrence SG, Munawar IF, Malley DF, eds. Aquatic
Ecosystems of Mexico: Status and Scope. Leiden (The Netherlands):
Backhuys.
Hobbie JE, Carpenter SE, Grimm NB, Gosz JR, Seastedt TR. 2003. The US
Long Term Ecological Research program. BioScience 53: 21?32.
Hughes T, Szmant AM, Steneck R, Carpenter R, Miller S. 1999. Algal blooms
on coral reefs: What are the causes? Limnology and Oceanography 44:
1583?1586.
Hughes TP. 1994. Catastrophes, phase shifts, and large-scale degradation of
a Caribbean coral reef. Science 265: 1547?1551.
Hughes TP, et al. 2003. Climate change, human impacts, and the resilience
of coral reefs. Science 15: 929?933.
Jackson JBC, et al. 2001. Historical overfishing and the recent collapse of coastal
ecosystems. Science 293: 629?638.
Jensen HS, McGlathery KJ, Marino R, Howarth RW. 1998. Forms and avail-
ability of sediment phosphorus in carbonate sand of Bermuda seagrass
beds. Limnology and Oceanography 43: 799?810.
Kaiser J. 2001. An experiment for all seasons. Science 293: 624?627.
Kleypas JA, Buddemeier RW, Gattuso JP. 2001. The future of coral reefs in
an age of global change. International Journal of Earth Sciences 90:
426?437.
Koch MS, Madden CJ. 2001. Patterns of primary production and nutrient
availability in a Bahamas lagoon with fringing mangroves. Marine 
Ecology Progress Series 219: 109?119.
Laboy-Nieves EN, Klein E, Conde JE, Losada F, Cruz JJ, Bone D. 2001. Mass
mortality of tropical marine communities in Morrocoy, Venezuela.
Bulletin of Marine Science 68: 163?179.
Lapointe BE. 1997. Nutrient thresholds for bottom-up control of macroalgal
blooms on coral reefs in Jamaica and southeast Florida. Limnology and
Oceanography 42: 1119?1131.
???. 1999. Simultaneous top-down and bottom-up forces control
macroalgal blooms on coral reefs. Limnology and Oceanography 44:
1586?1592.
Lessios HA. 1988. Mass mortality of Diadema antillarum in the Caribbean:
What have we learned? Annual Review of Ecology and Systematics 19:
371?393.
Lirman D, Cropper WP. 2003. The influence of salinity on seagrass growth,
survivorship, and distribution within Biscayne Bay, Florida: Field,
experimental, and modeling studies. Estuaries 26: 131?141.
Loreau M, et al. 2001. Biodiversity and ecosystem functioning: Current
knowledge and future challenges. Science 294: 804?808.
Lovelock CE, Feller IC, McKee KL, Engelbrecht BMJ, Ball MC. 2004. The 
effect of nutrient enrichment on growth, photosynthesis and hydraulic
conductance of dwarf mangroves in Panama. Functional Ecology 18:
25?33.
McCook LJ. 1999. Macroalgae, nutrients and phase shifts on coral reefs:
Scientific issues and management consequences for the Great Barrier Reef.
Coral Reefs 18: 357?367.
McKee KL, Feller IC, Popp M, Wanek W. 2002. Mangrove isotopic (delta 
N-15 and delta C-13) fractionation across a nitrogen vs. phosphorus 
limitation gradient. Ecology 83: 1065?1075.
Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J. 2000.
Biodiversity hotspots for conservation priorities. Nature 403: 853?858.
Ogden JC. 1987. Cooperative coastal ecology at Caribbean marine labora-
tories. Oceanus 30: 9?15.
???. 1997. Marine managers look upstream for connections. Science 278:
1414?1415.
Restrepo J, Kjerfve B. 2000. Magdalena River: Interannual variability
(1975?1995) and revised water discharge and sediment load estimates.
Journal of Hydrology 235: 137?149.
September 2004 / Vol. 54 No. 9 ?  BioScience 855
Articles
Richards WJ, Bohnsack JA. 1990. The Caribbean Sea: A large marine ecosystem
in crisis. Pages 44?53 in Sherman K, Alexander LM, Gold BD, eds. Large
Marine Ecosystems: Patterns, Processes, and Yields. Washington (DC):
American Association for the Advancement of Science.
Rivera-Monroy VH, Twilley RR. 1996. The relative role of denitrification and
immobilization in the fate of inorganic nitrogen in mangrove sedi-
ments. Limnology and Oceanography 41: 284?296.
Rivera-Monroy VH, Twilley R, Medina E, Moser EB, Botero L, Francisco AM,
Bullard E. 2004. Spatial variability of soil nutrients in disturbed riverine
mangrove forests at different stages of regeneration in the San Juan
River estuary, Venezuela. Estuaries 27: 44?57.
Salisbury JE, Campbell JW, Meeker LD,V?r?smarty C. 2001. Ocean color and
river data reveal fluvial influence in coastal waters. Eos, Transactions of
the American Geophysical Union 82: 221?224.
Skirving W, Guinott J. 2001. The sea surface temperature story on the Great
Barrier Reef during the coral bleaching event of 1998. Page 350 in
Wolanski E, ed. Oceanographic Processes of Coral Reefs: Physical and 
Biological Links in the Great Barrier Reef. Boca Raton (FL): CRC Press.
Smith TJ III, Robblee MB, Wanless HR, Doyle TW. 1994. Mangroves,
hurricanes, and lightning strikes. BioScience 44: 256?262.
Solow AR. 1995. Estimating biodiversity: Calculating unseen richness.
Oceanus 38 (2): 9?10.
Spalding M, Blasco F, Field C. 1997.World Mangrove Atlas. Okinawa (Japan):
International Society for Mangrove Ecosystems.
Stenseth NC, Mysterud A, Ottersen C, Hurrell JW, Chan KS, Lima M. 2002.
Ecological effects of climate fluctuations. Science 297: 1292?1296.
Szmant AM. 2002. Nutrient enrichment on coral reefs: Is it a major cause of
coral reef decline? Estuaries 25: 743?766.
Thom BG. 1982. Mangrove ecology: A geomorphological perspective. Pages
3?17 in Clough BF, ed. Mangrove Ecosystems in Australia: Structure,
Function, and Management. Canberra: Australian National University
Press.
Trenberth KE, Hoar TJ. 1997. El Ni?o and climate change. Geophysical 
Research Letters 24: 3057?3060.
Twilley RR. 1995. Properties of mangrove ecosystems and their relation to
the energy signature of coastal environments. Pages 43?62 in Hall CAS,
ed. Maximum Power. Boulder: University Press of Colorado.
Twilley RR, Rivera-Monroy VH, Chen R, Botero L. 1998. Adapting an 
ecological mangrove model to simulate trajectories in restoration 
ecology. Marine Pollution Bulletin 37: 404?419.
Twilley R, Barron EJ, Gholtz HL, Harwell MA, Miller RL, Reed DJ, Roser JB,
Siemann EH, Wetzel RG, Zimmerman RJ. 2001. Confronting Climate
Change in the Gulf Coast Region. Cambridge (MA): Union of Concerned
Scientists.
Villamizar E. 2000. Estructura de una comunidad arrecifal en Falcon,
Venezuela, antes y despues de una mortalidad masiva. Revista de Biolo-
gia Tropical 47: 19?30.
Vitousek PM, Mooney HA, Lubchenco J, Melillo JM. 1997. Human domi-
nation of Earth?s ecosystems. Science 277: 494?499.
Wilkinson C. 2000. Status of Coral Reefs of the World: 2000. Townsville 
(Australia): Australian Institute of Marine Science.
[WTTC] World Travel and Tourism Council. 2003. Caribbean Travel and
Tourism: A World of Opportunity. London: WTTC.
856 BioScience  ?  September 2004 / Vol. 54 No. 9
Articles