ArticlePDF Available

Spatial patterns of seagrasses and salinity regimes interact to structure marine faunal assemblages in a subtropical bay


Abstract and Figures

Anthropogenic disturbances in coastal and marine environments have resulted in the transformation of vegetated benthic habitat spatial patterns, which is thought to influence the distribution, community composition, and behavior of marine fauna. In Biscayne Bay, Florida, USA, freshwater discharges into nearshore areas have resulted in the fragmentation of the spatial patterning of submerged aquatic vegetation (SAV). To understand the ecological consequences of the freshwater-induced SAV seascape fragmentation, fish and crustaceans were sampled using seine nets across seascapes with continuous and highly fragmented SAV spatial configurations and across salinity regimes. Fragmented SAV seascapes supported significantly higher species diversity of fish and crustaceans, especially in areas influenced by freshwater discharges. Also, fragmented seascapes supported a higher abundance of the pink shrimp Farfantepenaeus duorarum and the goby Gobiosoma robustum, and higher biomass of generalist predatory fishes than seascapes with continuous SAV. In contrast, pinfish Lagodon rhomboides was more abundant in seascapes with continuous SAV. Faunal assemblage composition differed between zones of contrasting salinity regimes, and the contribution of species occurrence and abundance to the differentiation of assemblage composition between seascape types was associated with the salinity regimes of the seascapes. Thus, water salinity and spatial properties of SAV seascapes are factors that interact to influence faunal community structure in Biscayne Bay. These findings highlight the importance of understanding how environmental context (e.g. salinity regimes) can modulate the influence of benthic spatial patterning on the abundance and biodiversity of nekton communities.
Content may be subject to copyright.
Mar Ecol Prog Ser
Vol. 594: 21–38, 2018 Published April 26
Submerged aquatic vegetation (SAV), such as sea-
grasses and macroalgae, is distributed across time
and space in a variety of configurations. The spatial
configuration, or spatial pattern, of vegetated sea -
floor habitats influences the distribution and abun-
dance of many associated marine species (Pitt man et
al. 2004, Boström et al. 2011). The study of the eco-
logical consequences of spatial pattern is the core
theme of landscape ecology, with the key underlying
premise that both the composition and spatial ar -
range ment of a landscape mosaic affect ecosystem
structure and function in ways that are different
© Inter-Research 2018 ·*Corresponding author:
Spatial patterns of seagrasses and salinity regimes
interact to structure marine faunal assemblages
in a subtropical bay
Rolando O. Santos1,5,*, Diego Lirman1, Simon J. Pittman2, 3, Joseph E. Serafy1, 4
1Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Cswy, Miami, FL 33149, USA
2Biogeography Branch, Marine Spatial Ecology Division, National Centers for Coastal Ocean Science,
U.S. National Oceanic & Atmospheric Administration, 1305 East-West Highway, Silver Spring, MD 20910, USA
3Marine Conservation & Policy Research, Marine Institute, Plymouth University, Drake Circus, Plymouth PL4 8AA, UK
4NOAA, National Marine Fisheries Service, Southeast Fisheries Science Center, 75 Virginia Beach Drive, Miami, FL 33149, USA
5Present address: Southeast Environmental Research Center, Florida International University, Miami, FL 33199, USA
ABSTRACT: Anthropogenic disturbances in coastal and marine environments have resulted in the
transformation of vegetated benthic habitat spatial patterns, which is thought to influence the dis-
tribution, community composition, and behavior of marine fauna. In Biscayne Bay, Florida, USA,
freshwater discharges into nearshore areas have resulted in the fragmentation of the spatial pat-
terning of submerged aquatic vegetation (SAV). To understand the ecological consequences of the
freshwater-induced SAV seascape fragmentation, fish and crustaceans were sampled using seine
nets across seascapes with continuous and highly fragmented SAV spatial configurations and
across salinity regimes. Fragmented SAV seascapes supported significantly higher species diver-
sity of fish and crustaceans, especially in areas influenced by freshwater discharges. Also, frag-
mented seascapes supported a higher abundance of the pink shrimp Farfantepenaeus duorarum
and the goby Gobiosoma robustum, and higher biomass of generalist predatory fishes than sea-
scapes with continuous SAV. In contrast, pinfish Lagodon rhomboides was more abundant in sea-
scapes with continuous SAV. Faunal assemblage composition differed between zones of contrasting
salinity regimes, and the contribution of species occurrence and abundance to the differentiation of
assemblage composition between seascape types was associated with the salinity regimes of the
seascapes. Thus, water salinity and spatial properties of SAV seascapes are factors that interact to
influence faunal community structure in Biscayne Bay. These findings highlight the importance of
understanding how environmental context (e.g. salinity regimes) can modulate the influence of
benthic spatial patterning on the abundance and biodiversity of nekton communities.
KEY WORDS: Seascape ecology · Seagrasses · Submerged aquatic vegetation · Habitat
fragmentation · Species diversity
Resale or republication not permitted without written consent of the publisher
This authors' personal copy may not be publicly or systematically copied or distributed, or posted on the Open Web,
except with written permission of the copyright holder(s). It may be distributed to interested individuals on request.
Mar Ecol Prog Ser 594: 21– 38, 2018
where spatial patterns differ (Wiens et al. 1993,
Turner et al. 2001).
Considerable research in terrestrial landscape
ecology has focused on the process and conse-
quences of habitat fragmentation. Habitat fragmen-
tation is a landscape process that breaks apart large,
continuous patches into smaller units, changing the
amount of habitat available for associated organisms,
the ratio between the edge and core habitat, patch
size, and connectivity (Fahrig 2003, Lindenmayer &
Fischer 2007, Didham et al. 2012). According to the
static island biogeographic model, the spatial trans-
formation through fragmentation and habitat loss
can progress through several phases of spatial pat-
terns (i.e. perforation, dissection, subdivision, shrink-
age, and attrition), all ultimately resulting in the cre-
ation of fragmented or patchy landscape patterns
(Forman 1995, McGarigal et al. 2005). Species re -
spond to landscape patterns resulting from habitat
fragmentation in different ways (i.e. negative and/or
positive effects) and at different spatial scales
depending on their ecological needs (Betts et al.
2007, Villard & Metzger 2014). Here, we explore the
relationship between spatial patterning and the com-
munity structure of diverse nektonic organisms asso-
ciated with SAV.
Seascape ecology, the marine counterpart of land-
scape ecology, has focused predominantly on SAV
seascapes (i.e. mosaics of seagrass/algae patches
within a sediment matrix) revealing dynamic spatial
patterning and associated ecological consequences
(Boström et al. 2011). The dynamics of SAV patches
are influenced by a range of internal organismal and
external factors. Internal factors encompass a spe-
cies’ ability to resist and recover from disturbance
(recolonization ability, growth rates, and other life
history traits). External factors include physical and
biological disturbances (Fonseca & Bell 1998, Gil -
landers 2007, Jackson et al. 2017). For example,
SAV patches can become fragmented by wave ac -
tion and other hydrodynamic forces (Fonseca & Bell
1998), sedimentation events (Frederiksen et al. 2004),
diseases (Ralph & Short 2002), and herbivory (Bell et
al. 2007). Habitat loss and fragmentation of SAV sea-
scapes have also been driven by anthropogenic dis-
turbances, such as de clining water quality, nutrient
loading, sediment runoff, and changes in salinity
(Waycott et al. 2009, Santos et al. 2011), as well as
direct physical removal by dredging, vessel ground-
ings, and propeller scarring (Orth et al. 2006).
Seascape ecology studies have provided important
insights into how spatial patterns of SAV habitats
influence faunal assemblages (Turner et al. 1999,
Pittman et al. 2004, Boström et al. 2011, Hensgen et
al. 2014). Field studies and simulation modeling indi-
cate that both the composition (i.e. habitat amount
and types) and spatial configuration (i.e. spatial
arrangement, connectivity) of seascapes influence
key ecological processes such as faunal recruitment,
dispersal, and survivorship (Irlandi & Crawford 1997,
Pittman et al. 2004, Hovel & Regan 2008).
While habitat fragmentation and habitat loss are
typically reported as undesirable endpoints, some
degree of habitat fragmentation can, in fact, in -
crease species diversity and the abundance and
growth of certain species through positive edge
effects and increased spatial heterogeneity (Fahrig
2003, Ries & Sisk 2004). Edges are boundaries or
transition zones between adjacent habitat patches
that exhibit abrupt changes in physical structure,
biomass, and assemblage composition (Ries & Sisk
2004, Porensky & Young 2013). Edge effects in
vegetated seascapes are associated with higher fish
and epifaunal abundance (Bologna & Heck 2002,
Macreadie et al. 2010b, Boström et al. 2011, Pierri-
Daunt & Tanaka 2014). SAV habitat edges may pos-
itively affect faunal abundance by increasing move-
ment between patches, increasing the accumulation
of food resources, and modifying predation (Ries
& Sisk 2004, Macreadie et al. 2010b). In fragmented
or patchy habitats, edge effects can permeate the
entire seascape (Porensky & Young 2013). Fragmen-
tation increases the proportion of edge-to-interior
habitat, which may influence prey−predator inter -
actions, and the proportion of specialist and gener-
alist species (Bell et al. 2001, Ries & Sisk 2004).
Habitat fragmentation can also increase spatial het-
erogeneity by increasing the amount and diversity
of microhabitats that could be utilized by different
species (Horinouchi et al. 2009). Furthermore, frag-
mented SAV seascapes can influence prey accessi-
bility and predation success, which affect assemblage
structure and function of the nekton community
(Hovel et al. 2002, Connolly & Hindell 2006, Boström
et al. 2011).
Previous studies on the influences of SAV habitat
structure on faunal composition have often provided
contradictory results (Connolly & Hindell 2006, Bell
et al. 2007, Boström et al. 2011, 2017). Part of the dif-
ficulty encountered when examining faunal−habitat
relationships relates directly to the often limited spa-
tial scale of previous studies that have examined the
effects of fragmentation at the scale of individual
patches rather than the broader context of the sea-
scape (Boström et al. 2011). Habitat fragmentation,
however, is a process that occurs across mosaics of
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages
patches, where the size of patches, distances be -
tween patches, and composition of the matrix (e.g.
unvegetated sediments) influence organisms and
ecological processes (McGarigal & Cushman 2002,
Boström et al. 2011, Driscoll et al. 2013). Thus, a sea-
scape approach is needed to fully understand key
faunal−habitat relationships.
We examined the influence of the spatial configu-
ration of SAV seascapes on spatial patterns of marine
faunal diversity, abundance, and distribution, and
evaluated both increases and decreases of faunal
responses in seascapes with differing spatial configu-
rations. We assessed habitat−faunal relationships
using a field sampling design that takes into account
spatial variability in seascape patterning using a
binary patch-matrix model (Fig. A1 in the Appendix),
whereby patches are classified as either dominated
by seagrasses or unvegetated (i.e. dominated by sed-
iments). We also explored the interaction between
habitat patterning and salinity environments in near-
shore SAV habitats of Biscayne Bay (Florida, USA).
Biscayne Bay has been altered by water manage-
ment practices that release fresh water seasonally
from drainage canals. The pulsed release of fresh
water into littoral habitats has been linked directly to
the fragmentation of SAV seascapes in this coastal
lagoon (Santos et al. 2011, 2016). Concomitantly with
seascape transformation, changes in salinity regimes
across Biscayne Bay nearshore areas have been
linked to the reduction of estuarine fish abundance
and biomass, increased dominance of euryhaline
species (i.e. organisms that can tolerate a wide range
of salinity levels), and spatial changes in diversity
patterns (Serafy et al. 1997, 2003, Browder et al.
2005). A greater understanding of linkages among
water management practices, seascape fragmenta-
tion, and cascading effects on marine fauna is re -
quired to support science-based decisions within the
adaptive management framework proposed for the
restoration of the Florida Everglades. Thus, the
objective of this study was to determine the effects of
changes in SAV seascape patterns induced by fresh-
water discharges on marine fish and crustacean
assemblages. We combined field surveys with ana-
lytical tools from landscape ecology to test the
hypotheses that:
(1) Fish and crustacean abundance, biomass, and
diversity in more patchy, fragmented seascapes will
be higher than in continuous vegetated seascapes
(i.e. positive edge effects).
(2) Differences between seascapes would be ac -
centuated in areas with a salinity regime character-
ized by wider ranges of salinity by providing a more
spatially and temporally heterogeneous biophysical
environment which promotes coexistence of differ-
ent species.
Study area
Biscayne Bay is a shallow-water subtropical lagoon
(i.e. <3 m in depth) located adjacent to the city of
Miami and downstream of the Florida Everglades
system (Fig. 1a). Sampling of marine fishes and
crusta ceans focused on nearshore SAV seascapes
(<500 m from shore) in western Biscayne Bay, where
seagrasses are the dominant benthic macrophyte
(Lirman et al. 2008, 2014). SAV patches are mostly
composed of the seagrass Thalassia testudinum, with
some patches mixed with the seagrass Halodule
wrightii and rhyzophitic and drift macroalgae (Lir-
man et al. 2014). SAV communities have been stable
over the last 5 yr (Lirman et al. 2016). These vege-
tated communities, as well as the fringing man-
groves, provide habitat for a large number of com-
mercially and recreationally valuable species,
including pink shrimp Farfantepenaeus duorarum,
gray snapper Lutjanus griseus, hogfish Lachnolaimus
maximus, and spotted seatrout Cynoscion nebulosus
(Serafy et al. 1997, 2003, Faunce & Serafy 2008,
Browder et al. 2012).
Seascape mapping
To quantify the spatial patterning of seascapes and
to classify the study region into seascapes character-
ized by more continuous SAV and those with patchy,
more fragmented SAV configurations, the seafloor
(Fig. 1b) was mapped first using a Knearest neighbor
supervised classification applied to high-resolution
multispectral satellite images (Quickbird-2 satellite
images, 2.4 m pixel size) of the study region acquired
in November 2009 (ITT Visual Information Solutions
2008, Xie et al. 2008). The statistical classification
technique was based on an object-based (ENVI v4.5
Feature Extraction module, ITT Visual Information
Solutions 2008) approach that identified and delin-
eated patches with moderate to high macrophyte
cover (>40% of the bottom occupied by SAV). We
used >40% bottom cover as a threshold to define a
dense patch because studies have defined diffuse or
sparse SAV classes to areas with <40 % SAV cover,
and moderate and dense SAV classes to areas with
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
~40% SAV cover (Mumby & Edwards 2002, André-
fouët et al. 2003).
Object-based image classification optimized the
delineation of exterior and internal patch boundaries
(e.g. gaps, perforations) and provided better discrim-
ination between highly vegetated and sparse patch
classes under varying water depth and image quality
conditions than per-pixel based image classification
methods (Santos et al. 2011). Because objects (i.e.
image segments with distinct spatial, textural, and
spectral characteristics) are used instead of individ-
ual pixels, results do not have ‘salt-and-pepper’ ef -
fects or erroneously classified pixels across the image
(Kelly & Tuxen 2009).
The average percent cover of SAV from 153 geo -
referenced locations was used to inform the statisti-
cal classification algorithm and estimate the accuracy
of the seascape map produced. Approximately 20
benthic photos were taken at each of the 153 georef-
erenced locations for SAV community characteriza-
tion, following the methods of Lirman et al. (2014).
The delineation of the vegetated patches resulted in
an overall accuracy of 65%, and user accuracy of
64% (i.e. the probability that a pixel classified into a
given category represents that category on the
ground). Even though this accuracy demonstrated
only a moderate degree of agreement between the
maps and reference locations (Lathrop et al. 2006,
Lyons et al. 2010), maps were accepted as an accu-
rate representation of the seagrass seascape in Bis-
cayne Bay for the following reasons. (1) The accuracy
estimate of the maps may have been compromised
due to the horizontal disagreement between the
satellite images and the true position of the location
used for the accuracy assessment. (2) The values as -
signed to the locations used for this assessment were
average percent cover estimates of SAV using a
series of benthic photos taken around the assigned
reference location, thus increasing the probability of
spatial mismatch values between the reference loca-
tion and the habitat representation at the map scale
(Xie et al. 2008).
Fig. 1. (a) Study area and sampling design. The study area was divided into 2 zones based on salinity regimes: Zone 1 (high
and stable salinity, blue) and Zone 2 (low and variable salinity, green). (b) Submerged aquatic vegetation (SAV) seascape map
(dark green patches) with superimposed 500 × 500 m grid cells (i.e. seascape sampling unit; black line grid); some grid cells in
the south were excluded due to cloud cover interference with the image classification process. (c) Seascape sampling units
randomly selected and classified as continuous (light green) and fragmented (red) SAV seascapes. (d) Within each selected
grid cell, a 100 × 500 m plot was centered. Each plot was divided into 5 distance-to-shore strata (100 × 100 m) where 3
sampling replicates (yellow points) were randomly placed
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages
Spatial patterns of SAV seascapes
To assess the influence of seascape spatial
patterning (i.e. seascape effects) on fish and
crustacean assemblages, various spatial pattern
metrics were used to classify the survey sites
(n = 12 sites) into different SAV seascape cate-
gories: fragmented (FS) and continuous sea-
scape (CS). First, using a geographical informa-
tion system (GIS), a grid with 500 × 500 m cells
was superimposed over the seascape maps de -
line ating seascape sample units (SSUs) (i.e.
250 000 m2SSUs; Fig. 1b,c). The 500 m × 500 m
grid cells were identified as SSUs because this
spatial extent encompasses most of the sea-
scape heterogeneity, and because SAV habitats
<500 m from shore have been identified as the
most heterogeneous SAV habitats where sea-
scape differences can be observed along the
mainland coast of Biscayne Bay (Santos et al.
2011). In addition, the area up to 500 m from
shore has been identified as the area most likely
to be influenced by freshwater management
decisions and watershed restoration projects
(Lirman et al. 2008).
The seascape characteristics within each grid
cell were evaluated using spatial-pattern
metrics that quantify structural attributes of
seascape composition (variety and amount of
patch types) and spatial configuration (spatial
arrangement of patches). The 6 metrics calcu-
lated were: percentage of the seascape occu-
pied by a given habitat type (PLAND), mean
patch size (MPS), patch size coefficient of vari-
ation (PSCV), total edge (TE), area-weighted
mean patch fractal dimension (AWMPFD), and
patch density (PDENS) (Table 1). These metrics
have been widely used in landscape ecology
to investigate faunal− landscape associations
in terrestrial (Tischendorf 2001, Turner et al.
2001, McGarigal et al. 2005) and marine envi-
ronments (Pittman et al. 2004, Sleeman et al.
2005, Santos et al. 2011), and are robust and
stable across multiple spatial scales (Wu 2004).
Each cell or SSU was classified as either FS or
CS (Fig. 1c) using principal component analysis
(PCA) and hierarchical cluster analysis as de -
scribed by Santos et al. (2011) based on the sea-
scape pattern metrics de scribed above (Fig. 2).
The 500 m × 500 m cells classified as CS had a
higher proportion of the benthos covered by
larger SAV patches with lower shape complex-
ity. In contrast, FS cells had a higher density of
Table 1. Spatial pattern metrics used to quantify composition and configuration of submerged aquatic vegetation (SAV) seascape patterns in Biscayne Bay, Florida
(USA). Metrics were used in multivariate analysis to identify continuous and fragmented SAV seascapes. Variables in formulas: A: total landscape area (m2); aij: area
(m2) of patch jof patch type (class) i; ni: total number of patches of type (class) i; eik: total length (m) of edge in landscape involving patch type (class) i; pij: perimeter (m)
of patch ij
Metric Abbreviation Category Aspect Description Formula
Percentage of
PLAND Composition Area/density Percentage of the total landscape
made up of the corresponding class
PLAND (100)
Mean patch size (ha) MPS Composition/
Area/density Average size of a particular class
MPS 1a
Patch size coefficient
of variation (ha)
PSCV Configuration Area/density Variability in patch measures
Patch density
(patches ha−1)
PDENS Configuration Area/density Number of patches of a certain class
divided by the total landscape area
Total edge (m) TE Configuration Edge Sum of the lengths of all edges
TE 1eik
Area-weighted mean
patch fractal dimension
AWMPFD Configuration Shape Measure of patch shape complexity
2ln 0.25
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
smaller SAV patches with complex shapes and a
lower proportion of the substrate covered by SAV
patches (Fig. 3, Table A1 in the Appendix). A total of
12 cells was selected (6 in each of the salinity zones
described below) at random, with 6 cells classified as
FS and 6 as CS.
Due to the difference in salinity environments cre-
ated by the release of fresh water from canals in Bis-
cayne Bay (Caccia & Boyer 2005), sampling in this
study was also replicated within 2 zones with distinct
salinity regimes that have been previously described
by Lirman et al. (2008, 2014) (Zones 1 and 2, Fig. 1a).
Zone 1, an area with limited input of fresh water from
canal structures, was characterized by higher and
more stable salinity (wet season mean ± SD salinity:
26.6 ± 4.6 ppt, min−max: 13−36 ppt). In contrast, Zone 2
is influenced by pulsed freshwater flows from canals
that create a nearshore environment with low and
variable salinity (wet season mean salinity: 17.1 ± 8.2,
min−max: 2−33) (Lirman et al. 2008, 2014, Santos et al.
2011).This sampling design al lowed us to assess the
influence of seascape spatial patterns on fish and crus-
taceans under different salinity environments. Three
habitat grid cell replicates for each seascape type
(CS and FS) were randomly selected within salinity
Zones 1 and 2 (Fig. 1c,d). We were able to select differ-
ent seascape types within each salinity zone due to dif-
ferent factors. For example, FS and CS in Zone 2 were
identified adjacent and distant to canals, respectively,
thus enabling evaluation of the notion that freshwater
pulses are a major driver of seascape structuring in
Biscayne Bay. However, in Zone 1, seascapes classified
as FS were identified close to natural creeks, but also
in a portion of the bay characterized by shallow sedi-
ments and adjacent to exposed carbonate hardbottom
(Lirman et al. 2003, 2008, Browder et al. 2005).
Fish and crustacean sampling
For the faunal sampling, a 100 m × 500 m plot or
transect starting at the shore was randomly placed
Fig. 2. Principal component analysis (PCA) based on the spa-
tial pattern metrics (see Table 1) and performed to assess dis-
tinct seascape types. Based on this analysis, grid cells or sea-
scape sampling units were grouped into 2 seascape types:
continuous (green) and fragmented (red). The separation oc-
curred across the first axis (PC1) which explained 49.7 % of
the variation. The second axis (PC2) explained 31.4% of the
variation and illustrated heterogeneity in seascape patterns
among seascape types
Fig. 3. (a) Proportion of the seascape occupied by sub-
merged aquatic vegetation (SAV) patches (PLAND), (b)
mean patch size (MPS), (c) patch density (PDENS), and (d)
area-weighted mean patch fractal dimension (AWMPFD)
between seascape types (error bars: SE). Values are ex-
pressed as the standardized distance from the mean (value
subtracted by the mean and divided by the standard devia-
tion). Grid cells classified as continuous SAV seascapes
(green) tended to have a higher proportion of the bottom
covered by larger SAV patches with lower shape complex-
ity. In contrast, cells classified as fragmented SAV seascape
had a higher density of smaller SAV patches with complex
shapes and a lower proportion of the substrate covered by
SAV. All spatial pattern metrics were significantly different
between seascape types (see Table A1 in the Appendix)
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages 27
within each SSU (average depth = 1.1 m, average
depth range = 0.8 m; Fig. 1d). To document fish and
crustacean assemblage composition, abundance, bio -
mass, and diversity across the heterogeneous sea-
scape, each plot within the SSU was divided into 5
distance-to-shore strata (100 × 100 m), and 3 seine
sampling locations were randomly positioned within
each of these distance-to-shore strata (n = 15 points
plot−1, 3 deployment sites stratum−1; Fig. 1d). The fish
and crustacean sampling was conducted in 2012 dur-
ing the wet season (July−October). To maximize the
probability of capture, the sampling was performed
at night when many faunal species were actively for-
aging over seagrasses (Luo et al. 2009, Hammer-
schlag et al. 2010). In summary, the survey design
included: 2 seascape types (FS and CS), 2 salinity
zones (1 and 2), 3 replicates of each seascape type
per salinity zone, 5 distance-to-shore strata within
each plot, and 3 randomly located seine locations
within each distance-to-shore stratum, for a total
number of 180 seine locations.
Organisms were collected with a center-bag seine
net (21.3 m long, 1.8 m deep, 3 mm mesh) follow -
ing the Florida Marine Research Institute Fisheries-
Independent Monitoring Program Procedure Manual
(Florida Marine Research Institute 2007). The seine
was deployed and retrieved by motorboat, and each
seine haul swept a bottom area of approximately
210 m2, including both SAV patches and the matrix
(i.e. unvegetated or patches with low SAV cover)
within the hauling area. In addition, we randomized
the order of the seining hauls to promote the inde-
pendence of replicates across the 100 × 500 m sam-
pling plot and homogenize the distribution of environ-
mental conditions (e.g. tides, moonlight, temperature).
Sample and data processing
The fish and crustaceans collected were identified
to the lowest possible taxonomic level, counted,
and measured (total length, mm). Several metrics
were calculated to quantify species assemblages,
including species diversity indices (see below),
occurrence (presence/absence), and abundance
(count of ind. seine−1). Biomass (g seine−1) was esti-
mated using published length−weight relationships.
Peer-reviewed scientific publications and reports
were used to obtain allometric relationships to esti-
mate biomass. The delta approach was used to
account for positively skewed data and for zero-
inflation (Serafy et al. 2007). The data were thus
separated into a binary species occurrence matrix
(present = 1, absent = 0), and a species abundance
and biomass matrix when present (Clarke & War-
wick 2001, Serafy et al. 2007).
Different indices were used to assess the species
diversity between seascape types to minimize any
bias associated with any of the diversity indices
(Clarke & Warwick 1998, Izsák & Papp 2000). The
diversity indices considered were: number of species
per sample (species richness), Shannon-Wiener
index, Simpson diversity, and variation in taxonomic
distinctness. Variation in taxonomic distinctness is a
measure of ‘biological diversity’ that accounts for the
taxonomic differences in ‘relatedness’ among the
species rather than abundance. It is less biased by
site-to-site differences in sample size and has been
considered a proxy for functional diversity (Clarke &
Warwick 1998, Izsák & Papp 2000).
There was a difference of over 2 yr between the
date of capture of the imagery used to develop the
SAV seascape map (November 2009) and the seine
sampling (July−October 2012). While SAV abun-
dance changes (i.e. growth and/or dieback) may
have influenced some of the spatial pattern metric
estimates due to changes in patch area, total edge,
and perimeter:area ratios (Frederiksen et al. 2004,
Cunha & Santos 2009), no significant changes in
the SAV species composition or abundance across
Biscayne Bay were recorded over the time span
between the collection of the 2 data sources (Lirman
et al. 2014, 2016). In addition to the consistent levels
of SAV cover recorded in the area over the past 7−
10 yr, SAV seascape fragmentation in Biscayne Bay is
mostly a slow process that causes significant changes
in spatial characteristics over decadal scales (Santos
et al. 2016). Also, no major storms or hurricanes oc -
curred during this period (Lirman et al. 2016; https:// Thus, we be lieve that
seascape characteristics remained relatively constant
between the habitat classification and the collection
of fish and crustaceans.
Statistical analyses
Faunal responses (i.e. occurrence, abundance, bio-
mass) sampled from the 3 seine hauls per distance-
to-shore stratum were averaged. The averaged val-
ues per distance stratum were used as replicates
within each seascape sampling unit to capture the
species assemblage variability between the seascape
type and salinity zone treatments.
To examine if the assemblage composition differed
across seascape types and salinity zones, we used a
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
2-way crossed permutational multivariate analysis of
variance (PERMANOVA, type model I) with 1000 re -
sidual permutations. PERMANOVA allows the use of
dissimilarity measures to test for significance of spe-
cific effects (i.e. similar to a multivariate ANOVA) us-
ing a permutation test that does not require the data
to follow a particular distribution, and therefore is
more robust than parametric alternatives (Anderson
et al. 2008). We evaluated a habitat effect (seascape
types) and salinity effect (salinity zones) (both as fixed
effects), as well as their interaction. We used the
square-root of the estimated component of variation
ECV) as a measure of the relative importance be-
tween the fixed factors (seascape types and salinity
zones) and their interaction (Anderson et al. 2008).
Next, constrained ordinations (with Seascape × Zone
as a constrained factor) using a canonical analysis of
principal coordinates (CAP; Anderson & Willis 2003,
Anderson et al. 2008) were used to evaluate the differ-
ences in faunal assemblage composition within and
among seascape types (CS and FS) and salinity zones
(Zones 1 and 2). The correlations of individual species
(i.e. species occurrence, abundance, biomass) with
CAP axes were used to characterize the multivariate
effect by determining which species were driving the
assemblage composition differences between the sea-
scape types and salinity zones (Anderson & Willis 2003).
The means of the diversity indices were compared
among the seascape types and salinity zones using a
full-factorial 2-way ANOVA. Post hoc multiple com-
parisons were made using the Tukey multiple com-
parisons of means.
All ordination procedures (PERMANOVA, CAP)
were performed in PRIMER v6 with the add-on soft-
ware PERMANOVA+ (Anderson et al. 2008). ANOVA
and post hoc analysis were performed in R (R Core
Development Team 2010). We used α= 0.05 on all
tests to determine significant effects. Ordination mul-
tivariate procedures were based on a Bray-Curtis dis-
similarity matrix excluding the top 3 species (i.e. Euci-
nostromus spp., Atherinomorus stipes, Lucania parva).
These very abundant species were dominant across
all seascape and zone treatments and were removed
from subsequent analyses because they masked sub-
tler differences in the faunal assemblage (their re-
moval increased dissimilarity between seascape types
by 11%). The abundance and biomass data were log-
transformed to approach a normal distribution and
augment the contribution of rare species (i.e. buffer
the dominance of abundant species on the species
structure; Clarke & Warwick 2001). This transforma-
tion was also applied to diversity indices to validate
statistical assumptions for ANOVA.
Assemblage composition in continuous
vs. fragmented SAV seascapes
The species occurrence, abundance, and biomass
of the SAV-associated faunal assemblages differed
as a function of both seascape type and salinity
environment (Table 2). Significant interactions
between seascape type and salinity environment
were found for faunal occurrence and abundance
based on the PERMANOVA (pseudo-F1,56, p <
0.05), but were not significant for biomass at p <
0.05 (Table 2). For occurrence and abundance, the
interaction between seascape and salinity envi -
ronment was evident in the CAP plots that showed
the largest assemblage dissimilarity between sea-
scape types in Zone 2 (low and variable salinity;
Fig. 4a,b). In contrast, based on biomass, the largest
assemblage dissimilarity between seascape types
was in Zone 1 (Fig. 4c). Seascape types showed a
higher effect size (see higher
ECV in Table 2)
than salinity zones across all tests, indicating that
seascape types have a relatively higher importance
in explaining the assemblage composition.
Distinct seascape/salinity zone associations were
observed for fish and crustacean species. Based on
the correlations with both canonical axes (Fig. 4,
Source df
ECV Pseudo-F p (MC)
Seascape 1 11.05 7.42 < 0.001
Zone 1 8.41 4.71 <0.001
Seascape × Zone 1 7.51 2.48 <0.05
Residual 56 23.91
Seascape 1 18.96 11.95 < 0.001
Zone 1 11.73 5.19 <0.001
Seascape × Zone 1 15.07 4.46 <0.001
Residual 56 31.38
Seascape 1 10.59 4.25 < 0.001
Zone 1 8.02 2.86 <0.001
Seascape×Zone 1 6.83 1.67 0.106
Residual 56 32.19
Table 2. Results of the permutational multivariate analysis of
variance (PERMANOVA) conducted to assess the faunal re-
sponse to seascape type and salinity zone based on occur-
rence, abundance (number haul−1), and biomass.
square-root of estimated component of variation; pseudo-F
with associated p-values estimated from Monte Carlo per-
mutation (MC). Variables with p-values that are significant
at α= 0.05 are shown in bold
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages
Table 3), species such as Lagodon rhomboides, Op -
sa nus beta, and Harengula humeralis were associ-
ated with CS, especially in Zone 2, except for H.
humeralis which was highly associated with CS only
in Zone 1 (high and constant salinity; Fig. 4, Table 3).
Conversely, Lutjanus griseus, Farfantepenaeus duo-
rarum, and Callinectes spp. were associated with FS.
Species such as Gobiosoma robustum and Microgob-
ius gulosus, and L. griseus were highly associated
with FS in Zone 2.
Faunal diversity
Total number of species and number of unique spe-
cies (i.e. only observed in 1 of the seascape cate-
gories) differed between the seascape types and
salinity zones in western Biscayne Bay. FS seascapes
had a higher number of total and unique species
compared to CS habitats (i.e. 53 total species in FS vs.
44 in CS). Eight species were only found in CS and 15
species were only found in FS. A total of 45 and 43
species were identified in Zone 1 and Zone 2, respec-
tively, with 12 species unique to Zone 1 and 10 to
Zone 2.
Faunal assemblages in FS were significantly
more diverse than in CS, according to all of the
diversity indices (Fig. 5, Table 4, Table A2). While
significantly higher values were observed for all 4
metrics tested in FS in Zone 2 (low and variable
salinity), the Simpson and Shannon-Wiener diver-
sity metrics did not differ significantly between
seascapes in salinity Zone 1 (high and constant
Fig. 4. Plots of canonical analysis of principal coordinates
(CAP) for the: (a) occurrence, (b) abundance, and (c) biomass
of species assemblages in fragmented (stars) and continuous
(squares) submerged aquatic vegetation (SAV) seascapes. To
illustrate the interactive effect of SAV seascapes and salinity
regimes, species assemblages were placed in ordination
space based on seascape types within Zone 1 (high/stable
salinity; black symbols) and Zone 2 (low/variable salinity,
grey symbols). Vectors illustrate the strength and direction of
individual fish and crustacean species showing absolute cor-
relation ρ> 0.20 that contributed to the separation in species
assemblages between seascape types and zones. CALLI_ SPP:
Callinectes sp.; CARIDEA: caridean shrimps; FAR_ DUO:
Farfantepenaeus duorarum; GOB_ ROB: Gobiosoma robus-
tum; HAE_SCI: Haemulon sciurus; HAR_ HUM: Harengula
humeralis; LAG_RHO: Lagodon rhomboides; LUT_GRI: Lut-
janus griseus; MIC_GUL: Microgobius gulosus; OPS_ BET:
Opsanus beta
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
Fig. 5. Mean of species diversity indices (error bars: SE): species richness, Shannon-Wiener, Simpson diversity, and variation in
taxonomic distinctness (VarT) compared between continuous and fragmented submerged aquatic vegetation (SAV) seascapes
and salinity Zones 1 and 2. If interaction effects were significant, groups (letters) were identified using the Tukey multiple com-
parisons of means (see Table A2). Asterisks (*) identify significant seascape type effects only (2-way ANOVA, F1,159, p < 0.05)
Species ρaxis1 ρaxis2 y
Fragmented-Zone1 y
Continuous-Zone1 y
Fragmented-Zone2 y
Lutjanus griseus 0.80 −0.28 0.73 0.47 0.87 0.80
Harengula humeralis −0.52 0.16 0.73 0.80 0.47 0.40
Lagodon rhomboides 0.40 0.12 0.93 0.93 1 0.93
Opsanus beta 0.33 0.26 0.53 0.4 0.80 0.73
Haemulon sciurus −0.22 0.20 0.93 0.67 0.47 0.87
Callinectes sp. 0.16 −0.82 0.87 0.60 1 0.87
Farfantepenaeus duorarum 0.01 −0.42 1 1 1 1
Microgobius gulosus −0.90 0.34 0.40 0.01 4.44 0.12
Gobiosoma robustum −0.84 0.33 0.51 0.18 6.26 1.23
Caridean shrimps −0.60 0.39 5.13 5.60 19.67 1.53
Lagodon rhomboides −0.19 −0.39 5.49 5.00 6.49 13.28
Farfantepenaeus duorarum 0.01 0.68 45.22 31.63 25.79 19.67
Lutjanus griseus −0.69 0.37 27.53 15.44 100.25 43.10
Harengula humeralis 0.59 −0.16 3.17 14.50 1.31 4.31
Opsanus beta −0.37 −0.26 10.96 11.77 15.41 34.28
Farfantepenaeus duorarum −0.32 0.40 56.08 36.66 40.64 36.49
Lagodon rhomboides −0.29 < 0.1 100.84 109.52 153.73 222.77
Haemulon sciurus −0.18 −0.23 27.92 13.74 17.34 32.36
Callinectes sp. <0.1 0.92 15.32 21.64 19.87 8.07
Table 3. Correlations of species occurrence, abundance, and biomass with canonical axes 1 and 2 (ρaxis1 and ρaxis2) obtained
from canonical analyses of principal coordinates (CAP, see Fig. 4a−c). Only correlations of |r| 0.20 are shown. The last 4
columns present the average (y
) occurrence, abundance (number haul−1), and biomass (g) of these species within each combi-
nation of seascape type and salinity zone. Averages are used to illustrate how each species contributed to the separation of the
species assemblages between the seascape type and salinity zone
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages
SAV habitats and nekton communities have been
monitored in Biscayne Bay for over a decade to un-
derstand the ecological impacts of freshwater man-
agement on coastal habitats (Browder et al. 2012,
Lirman et al. 2014). Understanding the influence of
spatial patterns of SAV seascapes on associated fauna
is of growing importance due to increased seagrass
fragmentation and habitat loss, and the degradation
of SAV ecosystem services associated with changes in
freshwater regimes (Orth et al. 2006, Waycott et al.
2009, Boström et al. 2011). Here, we show that higher
relative abundance and biomass for several species
and higher diversity of fish and crustacean species
(based on species richness and variation in taxonomic
distinctness) in seagrass habitats were associated
with fragmented and patchy SAV seascapes. At the
spatial scale of this study of SAV seascapes (500 ×
500 m SSU), the fragmentation state observed in
nearshore Biscayne Bay revealed positive effects, in-
teracting with salinity regimes (i.e. positive effects
predominantly in Zone 2 and not Zone 1), on the di-
versity of the faunal community and the abundance
and biomass of certain trophic groups.
Seascape transformation (i.e. changes in the spatial
patterning of marine habitat patches) is a continuous
process driven by a series of disturbance/succession
events occurring at different scales (Cunha & Santos
2009, Santos et al. 2016, Jackson et al. 2017). Thus,
the positive fragmentation effects observed in this
study may not be static, and possible ecological
thresholds likely exist with respect to fragmentation
gradients that, when exceeded, can result in drastic
reductions in habitat value and provisioning (Fon-
seca & Bell 1998, Mizerek et al. 2011, Yeager et al.
2016). If SAV seascape fragmentation and habitat
loss proceed, most seagrass associated fishes could
disappear and the nekton assemblages may become
more similar to those over unvegetated sediments
(Horinouchi 2007, Boström et al. 2011). With the con-
tinuing decline in SAV extent, increased urbaniza-
tion, and the realized and projected global impacts of
climate change, the detection and characterization of
such thresholds based on a patch-mosaic model
should be a priority of SAV research.
Faunal assemblages in continuous vs.
fragmented seascapes
As expected, there were differences in assemblage
composition between seascape types, driven by the
higher abundance of Lagodon rhomboides and bio-
mass of Haemulon sciurus in continuous SAV sea-
scapes, and by higher abundance and larger individ-
uals of Lutjanus griseus and Farfantepenaeus duo ra -
rum in fragmented SAV seascapes. Ecological traits
and trophic interactions of these species suggest that
tradeoffs between food availability and predation
risk may be the mechanisms behind the faunal re -
sponses to seascape patterns (Connolly & Hindell
2006, Boström et al. 2011). For example, L. rhombo -
ides, an estuarine seagrass-dependent fish (Levin et
al. 1997, Potthoff & Allen 2003), uses continuous SAV
seascapes to avoid predation (Jordan et al. 1997) and
enhance the tradeoff between growth and foraging
efficiency (Levin et al. 1997). Larger individuals of L.
griseus, a generalist benthivore that forages at night
(Luo et al. 2009), were observed more frequently in
FS, suggesting that larger individuals of this species
prefer patchy seascapes for foraging activities. Gaps,
unvegetated patches, and macrophyte patches of low
complexity within FS may serve as corridors facilitat-
ing the movement of large predatory species (Irlandi
et al. 1995, Heck & Orth 2006). In Australia, King
George whiting Silliginodes punctatus consumed
more prey in areas within mosaics of seagrass sea-
scapes and unvegetated patches (Jenkins et al.
2011), indicating increased foraging efficiency within
patchy or fragmented SAV seascapes. Lastly, similar
to Syngnathidae and crustacean studies (Browder et
al. 1989, Macreadie et al. 2010b), F. duorarum tended
Biodiversity index Factor F1,159 p
Species richness Seascape 44.541 < 0.001
Zone 1.983 0.161
Seascape × Zone 5.584 0.019
Shannon-Wiener Seascape 15.705 < 0.001
Zone 0.748 0.388
Seascape × Zone 18.703 < 0.001
Simpson diversity Seascape 6.726 0.010
Zone 5.002 0.027
Seascape × Zone 9.880 0.002
VarT Seascape 0.900 0.002
Zone 0.110 0.600
Seascape × Zone 1.820 0.992
Table 4. Two-way ANOVA testing for differences of diversity
indices among the submerged aquatic vegetation (SAV) sea-
scape types, salinity zones, and their interaction. The results
present the F-statistic with associated degrees of freedom
and estimated p-values for the null hypothesis. Bold p-values
(α= 0.05) identify significant treatment effects. If interaction
effects were significant, groups were identified using the
Tukey multiple comparisons of means (see Table A2 in the
Appendix). VarT: variation in taxonomic distinctness
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
to be more abundant and larger in FS with smaller
patches with complex shapes (i.e. small patches with
high number of small edges) where its prey tend to
accumulate (Bologna & Heck 1999, 2002, Eggleston
et al. 1999, Healey & Hovel 2004).
The differences in assemblage composition be -
tween seascape types were also driven by the higher
abundance of small mobile species such as the
carideans and Gobiidae species in FS. Higher abun-
dances of shrimp and other small crustacean species
in fragmented SAV seascapes have been linked to (1)
the formation of isolated patches and seascapes with
small patches and higher edge density (Bologna &
Heck 1999, 2002, Healey & Hovel 2004); (2) con -
centration of individuals and species into remnant
patches after surrounding areas experienced habitat
loss or fragmentation (i.e. crowding effects) (Mac -
readie et al. 2010a,b); and (3) increased mobility
along edges and connectivity between patches
(Eggleston et al. 1998).
Faunal diversity
Manipulative studies have shown that sometimes
habitat fragmentation and edge effects can have a
positive influence on mobile marine fauna (Bologna
& Heck 2002, Pierri-Daunt & Tanaka 2014), whereas
some fauna appear unaffected by changes in patchi-
ness (Lefcheck et al. 2016) and others were only
affected when seagrass area was very low (< 25%
cover; Yeager et al. 2016). In Biscayne Bay, 2 diver-
sity indices indicated that faunal species diversity
was significantly higher in fragmented than in CS.
This study explored habitat−faunal characteristics
based on a patch-mosaic model; thus, the field sam-
pling design, on average, in FS sampled proportion-
ally more substrate types (i.e. seagrass patches and
the matrix composed of barren substrates or sub-
strates with low SAV cover) than in CS, potentially
influencing the number of species caught between
both seascape types. However, other studies using
either a patch or a patch-mosaic approach have also
de scribed higher species diversity in fragmented ver-
sus continuous marine habitats (Healey & Hovel
2004, Horinouchi et al. 2009).
The positive effects of fragmentation on species
diversity could be attributed to an increased co-exis-
tence of early and late successional stages, generalist
and specialist species, and high abundance of tran-
sient species (Debinski & Holt 2000, Fahrig 2003).
The coexistence of 2 competing species can be pro-
moted when the habitat is fragmented (Levin 1974,
Atkinson & Shorrocks 1981). This type of competition
relaxation was probably reflected by the higher oc -
currence and abundance of Gobiosoma robustum
and Microgobius gulosus in FS. These 2 species pre-
fer seagrass habitats over unvegetated sediment, but,
when competing directly for the same patch, G.
robustum can displace M. gulosus onto patches of
unvegetated sediment (Schofield 2003). In addition,
in the presence of the predator fish Opsanus beta,
both species prefer bare substrate to seagrass
patches (Schofield 2003).
FS may provide more niche space due to juxta-
posed microhabitat patches that generalist and tran-
sient predators could exploit (Ryall & Fahrig 2006).
For example, higher occurrence and abundance of
omnivore and generalist predators such as Floridic-
thys carpio, F. duorarum, L. griseus, C. sapidus, G.
robustum, and M. gulosus were observed in FS ver-
sus CS. Several studies have demonstrated how gaps
within seagrass meadows and edges of fragmented
patches can have species diversities and abundances
that are similar to or even greater than seagrass core
habitats (Horinouchi 2009), which may also partly
explain the enhanced diversity in FS. Thus, in accor-
dance with the habitat heterogeneity hypothesis
(MacArthur & MacArthur 1961), intermediate levels
of fragmentation may increase the diversity within
the seascape by increasing the number of microhab-
itats and species interactions in contrasting habitats
(Tscharntke et al. 2012).
Seascape and salinity interactions
We observed that the seascape patterns signifi-
cantly interacted with the salinity zones to modulate
differences in faunal assemblage composition and
diversity, highlighting the complexity and challenges
faced when trying to understand the response of fau-
nal assemblages to changes in their natural habitat.
Our study included salinity regimes as an explana-
tory factor because salinity can influence both spatial
attributes of SAV habitats (Santos et al. 2011) and
faunal responses in Biscayne Bay (Serafy et al. 2003,
Serrano et al. 2010, Browder et al. 2012). Salinity pre -
ferences and osmoregulation requirements can spa-
tially limit animals to remain within specific salinity
ranges and influence energy allocation (i.e. tradeoffs
between growth, reproduction, motility, and habitat
use; Hurst & Conover 2002, Serrano et al. 2010,
McManus et al. 2014). Results from a salinity labora-
tory experiment using the most abundant nearshore
fish species in Biscayne Bay suggested that the dif-
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages
ferential osmoregulatory abilities explain some, but
not all, of the differences in distribution and abun-
dance of fish assemblages in different salinity envi-
ronments (Serafy et al. 1997). In agreement with our
findings, this suggests that faunal composition and
distribution in nearshore Biscayne Bay are modu-
lated by both distinct combinations of seascape char-
acteristics and salinity environments.
The interaction between seascape types and salin-
ity regimes suggested that the largest contrast in
abundance and biomass between seascape types oc -
curred when the salinity regime was optimal for dif-
ferent species. Patterns of species composition, abun-
dance, and diversity can be a result of a spatial
hierar chy of interacting processes operating across
multiple ecological scales — for example, physiologi-
cal processes influenced by salinity, species interac-
tion, and movement dynamics (Pittman & McAlpine
2003, Howey et al. 2017). Salinity requirements may
primarily drive species distribution as well as the
species assemblage in a patchy or suboptimal salinity
environment. However, when salinity conditions are
favorable, biotic interactions and demographic pro-
cesses operating at the scale of the seascape may
influence structure, diversity, and distribution pat-
terns. This hierarchy of drivers was evidenced in our
study for species such as F. duorarum, L. griseus, and
L. rhomboides that have a wide salinity tolerance and
were ubiquitous across our study area (Serafy et al.
1997, Santos 2010, Serrano et al. 2010). For these spe-
cies, the highest contrast in abundance and biomass
was observed between the seascape types (and not
between salinity zones), especially within their opti-
mal salinity regimes. For F. duorarum, the largest
contrast in abundance between the seascape types
was observed in Zone 1, which exhibited stable poly-
haline regime optimal for this invertebrate species
(Browder et al. 2005, Zink et al. 2017). At the spatial
scale of this study, seascape types tended to out-
weigh the salinity regime effects, as demonstrated by
the response variance attributable to each factor (i.e.
ECV values in Table 2).
Our findings suggest that in Biscayne Bay, the
combination of seascapes with fragmented proper-
ties and variable salinity may support a better habitat
for some SAV-associated species. Within the context
of the intermediate disturbance hypothesis (Connell
1978), the right combination of seascape fragmen -
tation and variable salinity appear to be fostering
the co-existence of a diverse and productive faunal
community not recorded in continuous or more sta-
ble adjacent habitats. In addition, the interactive
effects of seascape types and salinity regimes on
diversity (species richness and Shannon-Wiener;
Fig. 5) could be attributed to an increase in facilita-
tive inter actions associated with moderate-stress
environments (Bruno et al. 2003, Holmgren & Schef-
fer 2010), and the expected increased presence of
generalist and euryhaline species in FS (Fahrig 2003,
Ryall & Fahrig 2006, Villard & Metzger 2014).
Further studies of fragmentation as a continuous
variable, and more detailed analyses of salinity (and
correlated variables like nutrients), are needed to
explore and predict potential thresholds in the inter-
action between seascape patterns and environmental
variables to determine when and under what condi-
tions habitat value decreases with increasing frag-
mentation (e.g. Yeager et al. 2016). Using a seascape
approach that combines statistical models, computer
simulations, fine-scale (10s of m) manipulations, and
broad-scale (100s of m to ha) inter-annual surveys
may provide the necessary information to identify
such critical thresholds that signal major ecosystem
shifts, and help conceptualize the potential future
effects of water management practices on the spatial
composition and configuration of SAV seascapes and
their associated nekton communities.
Urban coastal ecosystems are dynamic in distur-
bance regimes, and our findings show the relevance
of spatial patterning in the context of resource man-
agement and restoration strategies to evaluate es -
sential fish habitats and spatial distribution of marine
resources. Based on our study, the ecological re -
sponses to changes in the structure of SAV seascapes
should be incorporated into future studies on species
persistence and community assemblage stability un -
der anthropogenic disturbances. Forecasted in creases
of extreme disturbance events associated with cli-
mate change will likely expose seascapes to a series
of fragmentation/recovery events overlapping with
other environmental changes oc curring at broad
scales (i.e. freshwater discharges, nutrient loads, wave
or current exposure), highlighting the importance of
incorporating landscape eco logy concepts to under-
standing habitat pattern− process relationships at
relevant ecological scales.
Acknowledgements. Special thanks to James McCullars,
Matthew Dilly, and Amanda Guthrie for help in the field and
laboratory. This research was conducted under NPS permit
BISC-2011-SCI-0028. Funding was provided by NOAA’s
Educational Partnership Program and Living Marine
Resources Cooperative Science Center, the Army Corps of
Engineers, and the RECOVER Monitoring and Assessment
Program (MAP). The doctoral studies of R.O.S. were sup-
ported by the McKnight Doctoral Fellowship (Florida Edu-
cation Fund). S.J.P. was funded by NOAA National Centers
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
for Coastal Ocean Science’ Biogeography Branch of the
Marine Spatial Ecology Division. We thank the anonymous
reviewers, as well as the University of Miami’s Shark
Research & Conservation Program and Dr. Neil Hammer-
schlag for providing some of the equipment used in the field.
Anderson MJ, Willis TJ (2003) Canonical analysis of princi-
pal coordinates: a useful method of constrained ordina-
tion for ecology. Ecology 84: 511−525
Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA+
for PRIMER: a guide to software and statistical methods.
PRIMER-E, Plymouth
Andréfouët S, Kramer P, Torres-Pulliza D, Joyce KE and oth-
ers (2003) Multi-site evaluation of IKONOS data for clas-
sification of tropical coral reef environments. Remote
Sens Environ 88: 128−143
Atkinson WD, Shorrocks B (1981) Competition on a divided
and ephemeral resource: a simulation model. J Anim
Ecol 50: 461−471
Bell SS, Brooks RA, Robbins BD, Fonseca MS, Hall MO
(2001) Faunal response to fragmentation in seagrass
habitats: implications for seagrass conservation. Biol
Conserv 100: 115−123
Bell SS, Fonseca MS, Stafford NB (2007) Seagrass ecology:
new contributions from a landscape perspective. In:
Larkum AWD, Orth RJ, Duarte CM (eds) Seagrasses:
biology, ecology and conservation. Springer, Dordrecht,
p 625−645
Betts MG, Forbes GJ, Diamond AW (2007) Thresholds in
songbird occurrence in relation to landscape structure.
Conserv Biol 21: 1046−1058
Bologna PAX, Heck KL Jr (1999) Differential predation and
growth rates of bay scallops within a seagrass habitat.
J Exp Mar Biol Ecol 239: 299−314
Bologna PAX, Heck KL Jr (2002) Impact of habitat edges on
density and secondary production of seagrass-associated
fauna. Estuaries 25: 1033−1044
Boström C, Pittman SJ, Simenstad C, Kneib RT (2011) Sea-
scape ecology of coastal biogenic habitats: advances,
gaps, and challenges. Mar Ecol Prog Ser 427: 191−217
Boström C, Pittman SJ, Simenstad C (2017) Ecological conse-
quences of seagrass and salt-marsh seascape patterning
on marine fauna. In: Pittman SJ (ed) Seascape ecology.
Wiley & Sons, Hoboken, NJ, p 121−151
Browder JA, May LN Jr, Rosenthal A, Gosselink JG, Bau-
mann RH (1989) Modeling future trends in wetland loss
and brown shrimp production in Louisiana using The-
matic Mapper imagery. Remote Sens Environ 28: 45−59
Browder JA, Alleman R, Markley S, Ortner P, Pitts PA (2005)
Biscayne Bay conceptual ecological model. Wetlands 25:
Browder JA, Robblee MB, Liehr GA, Johnson D, Jackson TL
(2012) Epifaunal communities of mainland nearshore
South Biscayne Bay. Southeast Fisheries Science Center,
NOAA National Marine Fisheries Service, Miami, FL
Bruno JF, Stachowicz JJ, Bertness MD (2003) Inclusion of
facilitation into ecological theory. Trends Ecol Evol
Caccia VG, Boyer JN (2005) Spatial patterning of water
quality in Biscayne Bay, Florida as a function of land use
and water management. Mar Pollut Bull 50: 1416−1429
Clarke KR, Warwick RM (1998) A taxonomic distinctness
index and its statistical properties. J Appl Ecol 35:
Clarke KR, Warwick RM (2001) Change in marine commu-
nities: an approach to statistical analysis and interpreta-
tion, 2nd edn. PRIMER-E, Plymouth
Connell JH (1978) Diversity in tropical rain forests and coral
reefs. Science 199: 1302−1310
Connolly RM, Hindell JS (2006) Review of nekton patterns
and ecological processes in seagrass landscapes. Estuar
Coast Shelf Sci 68: 433−444
Cunha AH, Santos RP (2009) The use of fractal geometry to
determine the impact of inlet migration on the dynamics
of a seagrass landscape. Estuar Coast Shelf Sci 84:
Debinski DM, Holt RD (2000) A survey and overview of
habitat fragmentation experiments. Conserv Biol 14:
Didham RK, Kapos V, Ewers RM (2012) Rethinking the con-
ceptual foundations of habitat fragmentation research.
Oikos 121: 161−170
Driscoll DA, Banks SC, Barton PS, Lindenmayer DB, Smith
AL (2013) Conceptual domain of the matrix in frag-
mented landscapes. Trends Ecol Evol 28: 605−613
Eggleston DB, Etherington LL, Elis WE (1998) Organism
response to habitat patchiness: species and habitat-
dependent recruitment of decapod crustaceans. J Exp
Mar Biol Ecol 223: 111−132
Eggleston DB, Elis WE, Etherington LL, Dahlgren CP, Posey
MH (1999) Organism responses to habitat fragmentation
and diversity: habitat colonization by estuarine macro-
fauna. J Exp Mar Biol Ecol 236: 107−132
Fahrig L (2003) Effects of habitat fragmentation on biodiver-
sity. Annu Rev Ecol Syst 34: 487−515
Faunce CH, Serafy JE (2008) Selective use of mangrove
shorelines by snappers, grunts, and great barracuda.
Mar Ecol Prog Ser 356: 153−162
Florida Marine Research Institute (2007) The fisheries-inde-
pendent monitoring program procedure manual. Florida
Fish and Wildlife Conservation Commission, St. Peters-
burg, FL
Fonseca MS, Bell SS (1998) Influence of physical setting on
seagrass landscapes near Beaufort, North Carolina, USA.
Mar Ecol Prog Ser 171: 109−121
Forman RTT (1995) Land mosaics: the ecology of landscapes
and regions. Cambridge University Press, Cambridge
Frederiksen M, Krause-Jensen D, Holmer M, Laursen JS
(2004) Spatial and temporal variation in eelgrass (Zostera
marina) landscapes: influence of physical setting. Aquat
Bot 78: 147−165
Gillanders BM (2007) Seagrasses, fish, and fisheries. In:
Larkum AWD, Orth RJ, Duarte CM (eds) Seagrasses:
biology, ecology and conservation, 2nd edn. Springer,
Dordrecht, p 503−536
Hammerschlag N, Heithaus MR, Serafy JE (2010) Influ-
ence of predation risk and food supply on nocturnal fish
foraging distributions along a mangrove−seagrass eco-
tone. Mar Ecol Prog Ser 414: 223−235
Healey D, Hovel KA (2004) Seagrass bed patchiness: effects
on epifaunal communities in San Diego Bay, USA. J Exp
Mar Biol Ecol 313: 155−174
Heck KL Jr, Orth RJ (2006) Predation in seagrass beds. In:
Larkum AWD, Orth RJ, Duarte CM (eds) Seagrasses:
biology, ecology and conservation. Springer, Dordrecht,
p 537−550
Hensgen GM, Holt GJ, Holt SA, Williams JA, Stunz GW
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages
(2014) Landscape pattern influences nekton diversity
and abundance in seagrass meadows. Mar Ecol Prog Ser
507: 139−152
Holmgren M, Scheffer M (2010) Strong facilitation in mild
environments: the stress gradient hypothesis revisited.
J Ecol 98: 1269−1275
Horinouchi M (2007) Review of the effects of within-patch
scale structural complexity on seagrass fishes. J Exp Mar
Biol Ecol 350: 111−129
Horinouchi M (2009) Horizontal gradient in fish assemblage
structures in and around a seagrass habitat: some impli-
cations for seagrass habitat conservation. Ichthyol Res
56: 109−125
Horinouchi M, Tongnunui P, Nanjyo K, Nakamura Y, Sano
M, Ogawa H (2009) Differences in fish assemblage struc-
tures between fragmented and continuous seagrass beds
in Trang, southern Thailand. Fish Sci 75: 1409−1416
Hovel KA, Regan HM (2008) Using an individual-based
model to examine the roles of habitat fragmentation and
be havior on predator-prey relationships in seagrass
landscapes. Landsc Ecol 23: 75−89
Hovel KA, Fonseca MS, Myer DL, Kenworthy WJ, Whitfield
PE (2002) Effects of seagrass landscape structure, struc-
tural complexity and hydrodynamic regime on macro-
faunal densities in North Carolina seagrass beds. Mar
Ecol Prog Ser 243: 11−24
Howey LA, Wetherbee BM, Tolentino ER, Shivji MS (2017)
Biogeophysical and physiological processes drive move-
ment patterns in a marine predator. Mov Ecol 5: 16
Hurst TP, Conover DO (2002) Effects of temperature and
salinity on survival of young-of-the-year Hudson River
striped bass (Morone saxatilis): implications for opti -
mal overwintering habitats. Can J Fish Aquat Sci 59:
Irlandi EA, Crawford MK (1997) Habitat linkages: the effect
of intertidal saltmarshes and adjacent subtidal habitats
on abundance, movement, and growth of an estuarine
fish. Oecologia 110: 222−230
Irlandi EA, Ambrose WG Jr, Orlando BA (1995) Landscape
ecology and the marine environment: how spatial config-
uration of seagrass habitat influences growth and sur-
vival of the bay scallop. Oikos 72:307–313
ITT Visual Information Solutions (2008) ENVI Zoom tutorial:
ENVI feature extraction with supervised classification.
Boulder, CO
Izsák J, Papp L (2000) A link between ecological diversity
indices and measures of biodiversity. Ecol Model 130:
Jackson EL, Santos-Corujo RO, Pittman SJ (2017) Seascape
patch dynamics. In: Pittman SJ (ed) Seascape ecology.
Wiley & Sons, Hoboken, NJ, p 153−188
Jenkins GP, Syme A, Macreadie PI (2011) Feeding ecology of
King George whiting Sillaginodes punctatus (Perciformes)
recruits in seagrass and unvegetated habitats. Does diet
reflect habitat utilization? J Fish Biol 78: 1561−1573
Jordan F, Bartolini M, Nelson C, Patterson PE, Soulen HL
(1997) Risk of predation affects habitat selection by the
pinfish Lagodon rhomboides (Linnaeus). J Exp Mar Biol
Ecol 208: 45−56
Kelly M, Tuxen K (2009) Remote sensing support for tidal wet-
land vegetation research and management. In: Yang X
(ed) Remote sensing and geospatial technologies for
coastal ecosystem assessment and management. Springer,
Berlin, p 341−363
Lathrop RG, Montesano P, Haag S (2006) A multi-scale seg-
mentation approach to mapping seagrass habitats using
airborne digital camera imagery. Photogramm Eng
Remote Sensing 72: 665−675
Lefcheck JS, Marion SR, Lombana AV, Orth RJ (2016) Fau-
nal communities are invariant to fragmentation in exper-
imental seagrass landscapes. PLOS ONE 11: e0156550
Levin P, Petrik R, Malone J (1997) Interactive effects of habi-
tat selection, food supply and predation on recruitment
of an estuarine fish. Oecologia 112: 55−63
Levin SA (1974) Dispersion and population interactions. Am
Nat 108: 207−228
Lindenmayer DB, Fischer J (2007) Tackling the habitat frag-
mentation panchreston. Trends Ecol Evol 22: 127−132
Lirman D, Orlando B, Maciá S, Manzello D, Kaufman L,
Biber P, Jones T (2003) Coral communities of Biscayne
Bay, Florida and adjacent offshore areas: diversity, abun-
dance, distribution, and environmental correlates. Aquat
Conserv 13: 121−135
Lirman D, Deangelo G, Serafy J, Hazra A and others (2008)
Seasonal changes in the abundance and distribution of
submerged aquatic vegetation in a highly managed
coastal lagoon. Hydrobiologia 596: 105−120
Lirman D, Thyberg T, Santos R, Schopmeyer S and others
(2014) SAV communities of western Biscayne Bay, Miami,
Florida, USA: human and natural drivers of seagrass and
macroalgae abundance and distribution along a continu-
ous shoreline. Estuar Coast 37: 1243−1255
Lirman D, Monty J, Avila C, Buck E and others (2016) Sum-
mary report for Biscayne Bay. Seagrass Integrated Map-
ping and Monitoring Program, Mapping and Monitoring
Report No. 2.0. FWRI Technical Report TR-17. Fish and
Wildlife Research Institute, St. Petersburg, FL
Luo J, Serafy JE, Sponaugle S, Teare PB, Kieckbusch D
(2009) Movement of gray snapper Lutjanus griseus among
subtropical seagrass, mangrove, and coral reef habitats.
Mar Ecol Prog Ser 380: 255−269
Lyons MB, Phinn SR, Roelfsema CM (2010) Long term mon-
itoring of seagrass distribution in Moreton Bay, Australia,
from 1972-2010 using Landsat MSS, TM, ETM+. Int
Geosci Remote Sens Symp, Honolulu, HI, p 5−8
MacArthur RH, MacArthur JW (1961) On bird species diver-
sity. Ecology 42: 594−598
Macreadie PI, Connolly RM, Keough MJ, Jenkins GP, Hin-
dell JS (2010a) Short-term differences in animal assem-
blages in patches formed by loss and growth of habitat.
Austral Ecol 35: 515−521
Macreadie PI, Hindell JS, Keough MJ, Jenkins GP, Connolly
RM (2010b) Resource distribution influences positive
edge effects in a seagrass fish. Ecology 91: 2013−2021
McGarigal K, Cushman SA (2002) Comparative evaluation
of experimental approaches to the study of habitat frag-
mentation effects. Ecol Appl 12: 335−345
McGarigal K, Cushman S, Regan C (2005) Quantifying
terrestrial habitat loss and fragmentation: a protocol.
University of Massachusetts, Department of Natural Re -
sources Conservation, Amherst, MA
McManus LC, Yurek S, Teare PB, Dolan TE, Serafy JE
(2014) Killifish habitat suitability as a measure of coastal
restoration performance: integrating field data, behav-
ioral trials and simulation. Ecol Indic 44: 173−181
Mizerek T, Regan HM, Hovel KA (2011) Seagrass habitat
loss and fragmentation influence management strategies
for a blue crab Callinectes sapidus fishery. Mar Ecol Prog
Ser 427: 247−257
Mumby PJ, Edwards AJ (2002) Mapping marine environ-
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
ments with IKONOS imagery: Enhanced spatial resolu-
tion can deliver greater thematic accuracy. Remote Sens
Environ 82: 248−257
Orth RJR, Carruthers TTJB, Dennison WC, Duarte CM and
others (2006) A global crisis for seagrass ecosystems. Bio-
science 56: 987−996
Pierri-Daunt AB, Tanaka MO (2014) Assessing habitat frag-
mentation on marine epifaunal macroinvertebrate com-
munities: an experimental approach. Landsc Ecol 29:
Pittman SJ, McAlpine CA (2003) Movements of marine fish
and decapod crustaceans: process, theory and applica-
tion. Adv Mar Biol 44: 205−294
Pittman SJ, McAlpine CA, Pittman KM (2004) Linking fish
and prawns to their environment: a hierarchical land-
scape approach. Mar Ecol Prog Ser 283: 233−254
Porensky LM, Young TP (2013) Edge-effect interactions in
fragmented and patchy landscapes. Conserv Biol 27:
Potthoff MT, Allen DM (2003) Site fidelity, home range, and
tidal migrations of juvenile pinfish, Lagodon rhombo ides,
in salt marsh creeks. Environ Biol Fishes 67: 231−240
R Core Development Team (2010) R: a language and envi-
ronment for statistical computing.
Ralph PJ, Short FT (2002) Impact of the wasting disease
pathogen, Labyrinthula zosterae, on the photobiology of
eelgrass Zostera marina. Mar Ecol Prog Ser 226: 265−271
Ries L, Sisk TD (2004) A predictive model of edge effects.
Ecology 85: 2917−2926
Ryall KL, Fahrig L (2006) Response of predators to loss and
fragmentation of prey habitat: a review of theory. Ecol-
ogy 87: 1086−1093
Santos RO (2010) Linkage between mangrove fish commu-
nity and nearshore benthic habitats in Biscayne Bay,
Florida, USA: a seascape approach. MSc thesis, Nova
Southeastern University, Fort Lauderdale, FL
Santos RO, Lirman D, Serafy JE (2011) Quantifying fresh-
water-induced fragmentation of submerged aquatic veg-
etation communities using a multi-scale landscape ecol-
ogy approach. Mar Ecol Prog Ser 427: 233−246
Santos RO, Lirman D, Pittman SJ (2016) Long-term spatial
dynamics in vegetated seascapes: fragmentation and
habitat loss in a human-impacted subtropical lagoon.
Mar Ecol 37: 200−214
Schofield PJ (2003) Habitat selection of two gobies (Micro-
gobius gulosus, Gobiosoma robustum): influence of
structural complexity, competitive interactions, and pres-
ence of a predator. J Exp Mar Biol Ecol 288: 125−137
Serafy JE, Lindeman KC, Hopkins TE, Ault JS (1997) Effects
of freshwater canal discharge on fish assemblages in a
subtropical bay: field and laboratory observations. Mar
Ecol Prog Ser 160: 161−172
Serafy JE, Faunce CH, Lorenz JJ (2003) Mangrove shoreline
fishes of Biscayne Bay, Florida. Bull Mar Sci 72: 161−180
Serafy JE, Valle M, Faunce CH, Luo J (2007) Species-
specific patterns of fish abundance and size along a sub-
tropical mangrove shoreline: an application of the delta
approach. Bull Mar Sci 80: 609−624
Serrano X, Grosell M, Serafy JE (2010) Salinity selection and
preference of the grey snapper Lutjanus griseus : field
and laboratory observations. J Fish Biol 76: 1592−1608
Sleeman JC, Kendrick GA, Boggs GS, Hegge BJ (2005)
Measuring fragmentation of seagrass landscapes: which
indices are most appropriate for detecting change? Mar
Freshw Res 56: 851−864
Tischendorf L (2001) Can landscape indices predict ecologi-
cal processes consistently? Landsc Ecol 16: 235−254
Tscharntke T, Tylianakis JM, Rand TA, Didham RK and oth-
ers (2012) Landscape moderation of biodiversity patterns
and processes —eight hypotheses. Biol Rev Camb Philos
Soc 87: 661−685
Turner SJ, Hewitt JE, Wilkinson MR, Morrisey DJ, Thrush
SF, Cummings VJ, Funnell G (1999) Seagrass patches
and landscapes: the influence of wind-wave dynamics
and hierarchical arrangements of spatial structure on
macrofaunal seagrass communities. Estuar Coast 22:
Turner MG, Gardner RH, O’Neill RV (2001) Landscape
ecology in theory and practice: pattern and process.
Springer, New York, NY
Villard MA, Metzger JP (2014) Beyond the fragmentation
debate: a conceptual model to predict when habitat con-
figuration really matters. J Appl Ecol 51: 309−318
Waycott M, Duarte CM, Carruthers TJB, Orth RJ and others
(2009) Accelerating loss of seagrasses across the globe
threatens coastal ecosystems. Proc Natl Acad Sci USA
106: 12377−12381
Wiens JA, Stenseth NC, Van Horne B, Ims RA (1993) Ecolog-
ical mechanisms and landscape ecology. Oikos 66: 369−380
Wu J (2004) Effects of changing scale on landscape pattern
analysis: scaling relations. Landsc Ecol 19: 125−138
Xie Y, Sha Z, Yu M (2008) Remote sensing imagery in vege-
tation mapping: a review. J Plant Ecol 1: 9−23
Yeager LA, Keller DA, Burns TR, Pool AS, Fodrie FJ (2016)
Threshold effects of habitat fragmentation on fish diver-
sity at landscapes scales. Ecology 97: 2157−2166
Zink IC, Browder JA, Lirman D, Serafy JE (2017) Review of
salinity effects on abundance, growth, and survival of
nearshore life stages of pink shrimp (Farfantepenaeus
duorarum). Ecol Indic 81: 1−7
Author copy
Santos et al.: Seagrass seascapes and salinity structure faunal assemblages 37
Diversity index F1, 8 p
PLAND 36.054 < 0.001
MPS 14.466 0.003
PDENS 7.577 0.02
AWMPFD 14.736 0.003
Table A1. One-way ANOVA testing for differences in spa-
tial pattern metrics among the submerged aquatic vegeta-
tion (SAV) seascape types: fragmented (FS) vs. continuous
seascapes (CS). The results present Fwith associated de-
grees of freedom and estimated p-values for the null hypoth-
esis. These results illustrate that both FS and CS have differ-
ent spatial structure of SAV habitat patches. All p-values
were significant (p < 0.05). PLAND: proportion of the sea-
scape occupied by SAV patches; MPS: mean patch size;
PDENS: patch density; AWMPFD: area-weighted mean patch
fractal dimension
Diversity index Interaction term comparison Difference of means padjusted
Species richness Continuous:Zone 1−Continuous:Zone 2 0.449 0.903
Fragmented:Zone 1−Continuous:Zone 2 2.424 0.002
Fragmented:Zone 2−Continuous:Zone 2 4.167 0.000
Fragmented:Zone 1−Continuous:Zone 1 1.975 0.017
Fragmented:Zone 2−Continuous:Zone 1 3.718 0.000
Fragmented:Zone 2−Fragmented:Zone 1 1.743 0.042
Shannon-Wiener Fragmented:Zone 1−Continuous:Zone 2 0.276 0.005
Continuous:Zone 1−Continuous:Zone 2 0.302 0.002
Fragmented:Zone 2−Continuous:Zone 2 0.475 0.000
Continuous:Zone 1−Fragmented:Zone 1 0.025 0.990
Fragmented:Zone 2−Fragmented:Zone 1 0.199 0.075
Fragmented:Zone 2−Continuous:Zone 1 0.173 0.149
Simpson Fragmented:Zone 1−Continuous:Zone 2 0.119 0.005
Continuous:Zone 1−Continuous:Zone 2 0.134 0.001
Fragmented:Zone 2−Continuous:Zone 2 0.142 0.000
Continuous:Zone 1−Fragmented:Zone 1 0.015 0.976
Fragmented:Zone 2−Fragmented:Zone 1 0.022 0.922
Fragmented:Zone 2−Continuous:Zone 1 0.007 0.997
VarT Continuous:Zone 2−Continuous:Zone 1 8.179 0.981
Fragmented:Zone 1−Continuous:Zone 1 48.101 0.124
Fragmented:Zone 2−Continuous:Zone 1 55.959 0.049
Fragmented:Zone 1−Continuous:Zone 2 39.922 0.255
Fragmented:Zone 2−Continuous:Zone 2 47.780 0.117
Fragmented:Zone 2−Fragmented:Zone 1 7.857 0.983
Table A2. Table of computed Tukey HSD pairwise comparisons for the diversity indices observed between the Seascape × Zone
interaction terms. A bold padjusted value indicates significant differences between a distinct combination of Seascape × Zone factor
levels. Results were used to identify significant groups in Fig. 3. VarT: variation in taxonomic distinctness
Appendix. Comparison of the spatial pattern metrics between the seascape types and the post-hoc analysis for the diversity in-
dices observed between the seascape types and salinity zone interaction terms. A conceptual diagram illustrating the patch-
matrix approach considered in the study is also included
Author copy
Mar Ecol Prog Ser 594: 21– 38, 2018
Fig. A1. Illustration of a seascape analysis with (a) patch model and (b) patch-matrix approach. Red dots and black X illustrate
sampling points. In a patch model seascape analysis, faunal responses (Yi) are compared between patches (pi) with different
characteristics (e.g. area, shape, perimeter:area ratio) or between habitat core (dot) and edge (X) (right panel). In a seascape
analysis with a patch-matrix approach, faunal responses (Yi) are explored between different seascapes (Si) with distinct
amount and arrangement of patches or within an organism’s home range (dashed circles), thus incorporating faunal responses
across the entire mosaic of patches
Editorial responsibility: Romuald Lipcius,
Gloucester Point, Virginia, USA
Submitted: June 6, 2017; Accepted: January 25, 2018
Proofs received from author(s): April 3, 2018
Author copy
... At the landscape scale, the spatial pattern of seagrass meadows can be highly variable, ranging from continuous cover of vegetation to highly fragmented patches (Duarte et al. 2006). Although seagrass mapping efforts have historically focused on delineating generalized zones based on percentage cover of seagrass (Orth et al. 2006a, b;Roelfsema et al. 2009), recent technological advances, such as geographic object-based image analyses (Blaschke et al. 2014), have resulted in the development of higher resolution, patch-level seagrass maps (i.e., individual seagrass beds are mapped as discrete patches; Carter et al. 2011;Kenworthy et al. 2017;Santos et al. 2018) or maps with sub-pixel information (Uhrin and Townsend 2016). ...
... Consequently, seagrass maps that delineate discrete patches are of great interest to ecologists and natural resources managers because it allows the exploration of meadow spatial structure and configuration, both of which have important ecological implications (Robbins and Bell 1994;Bell et al. 1999;Johnson and Heck 2006;Hensgen et al. 2014). Spatial indices (i.e., patch statistics) from the field of landscape ecology have increasingly been used to assess the configuration of seagrasses and freshwater submerged aquatic vegetation (Frederiksen et al. 2004;Sleeman et al. 2005;Wedding et al. 2011;Yeager et al. 2011;Santos et al. 2018). However, these are typically sensitive to differences in scale (Wu 2004), which makes it important to select an appropriate scale given the ecosystem being studied and the purpose of the analysis. ...
... Although lacunarity has been applied in wetland communities (Wu et al. 1997;Dale 2000) and simulated seagrass environments (Mugan and MacIver 2020), it has not, to our knowledge, been used to evaluate seagrass landscape configuration using real-word data at multiple spatial scales over time. While lacunarity applications have waned (Gustafson 2019), the technique can provide helpful insight for understanding changes in coastal landscapes related to configuration (e.g., patches, edges, and ecotones), connectivity, and the consideration of scale (Pittman et al. 2021) in addition to understanding how the spatial pattern of seagrass patches influences use by organisms (McNeill and Fairweather 1993;Santos et al. 2018;Bilodeau et al. 2021). ...
Full-text available
Context Seagrasses are submerged marine plants that have been declining globally at increasing rates. Natural resource managers rely on monitoring programs to detect and understand changes in these ecosystems. Technological advancements are allowing for the development of patch-level seagrass maps, which can be used to explore seagrass meadow spatial patterns. Objectives Our research questions involved comparing lacunarity, a measure of landscape configuration, for seagrass to assess cross-site differences in areal coverage and spatial patterns through time. We also discussed how lacunarity could help natural resource managers with monitoring program development and restoration decisions and evaluation. Methods We assessed lacunarity of seagrass meadows for various box sizes (0.0001 ha to 400.4 ha) around Cat Island and Ship Island, Mississippi (USA). For Cat Island, we used seagrass data from 2011 to 2014. For Ship Island, we used seagrass data for seven dates between 1963 and 2014. Results Cat Island, which had more continuous seagrass meadows, had lower lacunarity (i.e., denser coverage) compared to Ship Island, which had patchier seagrass beds. For Ship Island, we found a signal of disturbance and path toward recovery from Hurricane Camille in 1969. Finally, we highlighted how lacunarity curves could be used as one of multiple considerations for designing monitoring programs, which are commonly used for seagrass monitoring. Conclusions Lacunarity can help quantify spatial pattern dynamics, but more importantly, it can assist with natural resource management by defining fragmentation and potential scales for monitoring. This approach could be applied to other environments, especially other coastal ecosystems.
... This includes those fishes that support Florida's iconic and highly valued recreational flats fisheries-which have experienced a severe decline in recent decades and may be impacted by the pink shrimp fisheries. Despite found in South Florida estuaries (Browder and Robblee 2009;Santos et al. 2018;James et al. This issue). ...
... Pink shrimp exhibit a life history pattern that includes migrating between nearshore juvenile nursery habitats and offshore adult habitats (Dall et al. 1990). This species is a major component of both the biomass and abundance of nearshore seagrass faunal communities (Santos et al. 2018). Pink shrimp are also economically important and support lucrative commercial fisheries throughout Florida (Johnson et al. 2012;Zink 2017). ...
... Pink shrimp densities in Biscayne Bay are influenced mainly by salinity, temperature, water depth, and submerged aquatic vegetation cover (SAV; dominated by seagrass species in Biscayne Bay; Zink et al. 2018). Since 2005, these environmental variables in Biscayne Bay have experienced dynamic changes that could influence the abundance Environ Biol Fish of pink shrimp populations, and subsequently, the pink shrimp fisheries within Biscayne Bay (Santos et al. 2018(Santos et al. , 2020. For instance, changes in water quality, invasive and harmful algal blooms, and Hurricane Irma have led to spatially heterogeneous changes in the environmental conditions and SAV cover throughout Biscayne Bay (Carey et al. 2011;Millette et al. 2019;Santos et al. 2020;Wachnicka et al. 2020). ...
Full-text available
Pink shrimp (Farfantepenaeus duorarum) are an economically important species in Biscayne Bay, FL, and support both food and bait commercial fisheries. Pink shrimp are also an important food resource for higher trophic level finfish species. This includes those fishes that support Florida’s iconic and highly valued recreational flats fisheries—which have experienced a severe decline in recent decades and may be impacted by the pink shrimp fisheries. Despite their economic and ecological importance, few studies have evaluated the long-term trends in Biscayne Bay’s pink shrimp fisheries. In this study, we evaluated over 30 years (1987–2020) of fisheries-dependent and economic data on the pink shrimp bait and food fisheries in Biscayne Bay with segmented regression to identify trends and potential breakpoints. We also evaluate trends in Biscayne Bay bonefish (Albula vulpes) over 25 years (1993–2018), based on recreational angler interview data, and assess potential interactions with the shrimp fisheries. We found that landings, value, effort, and participation (number of vessels and dealers) in both Biscayne Bay pink shrimp fisheries have exhibited declines from peaks in the late 1990s. No significant trends were detected in annual bonefish catch or catch per unit effort (catch/trip), but fishing effort declined over the time series. We did not find a significant relationship between annual bonefish catch per unit effort and commercial shrimp fishing landings or effort, suggesting that the pink shrimp fisheries are not a primary factor contributing to declines in the Biscayne Bay bonefish fishery.
... Our goal was to understand how the trophic niche and trophic level of an important generalist marine consumer, pinfish (Lagodon rhomboides), varied in response to the habitat amount and spatial configuration of submerged aquatic vegetation (SAV) seascapes. We concentrated our study in Biscayne Bay (Miami, Florida, USA), where SAV seascapes are influenced by freshwater management and where seascape properties can influence species abundance, diversity, and community assemblages (Santos et al., 2018). Pinfish were collected from two salinity zones of varying levels of anthropogenic influence in both continuous and fragmented seascapes. ...
... Our study concentrated on the nearshore habitats of the central-western section of Biscayne Bay, Miami, Florida, USA (Figure 1a), a shallow subtropical lagoon adjacent to the city of Miami (county population 2.5 million) and downstream of the Florida Everglades. We used preserved samples obtained by Santos et al. (2018), a study that sampled marine fishes and crustaceans on nearshore SAV seascapes (<500 m from shore) in western Biscayne Bay, where seagrasses are the dominant benthic macrophyte (Lirman et al., 2008(Lirman et al., , 2014. SAV patches are mostly composed of the seagrass Thalassia testudinum, with some patches mixed with Halodule wrightii (seagrass), rhizophytic macroalgae, and drift macroalgae (Lirman et al., 2014). ...
... SAV seascapes, as well as the fringing mangroves, provide habitat for a large number of commercially and recreationally valuable species (e.g., Farfantepenaeus duorarum, Lutjanus griseus, Lachnolaimus maximus; Diaz et al., 2001;Faunce & Serafy, 2008;Serafy et al., 1997Serafy et al., , 2003, including the focal species of this study-pinfish Lagodon rhomboides (Santos et al., 2018). Pinfish are an omnivorous species, which depends on seagrass seascapes (Jordan et al., 1997;Levin et al., 1997). ...
Full-text available
Habitat fragmentation of submerged aquatic vegetation (SAV) transforms the spatial pattern of seascapes by changing both the total area and spatial configuration of the habitat patches. The ecological effects of SAV seascapes are most often assessed using metrics of biological community composition (e.g., species and assemblage changes). We know considerably less about the effects of seascape structure on ecological processes such as food web function and energy flow. Here, we assess the difference in the trophic niche of Pinfish (Lagodon rhomboides, a generalist omnivore) across a spatial gradient of SAV from continuous to highly fragmented seascapes in Biscayne Bay (Miami, Florida, USA). The Bay seascapes are influenced by freshwater management practices that alter the distribution of SAV habitat and fish species abundance, diversity, and community assemblage. We combined SAV seascape maps with stable isotope and hypervolume analyses to determine how trophic niche size and overlap varied with changes in the seascape. We observed similar resource use across the seascape, but trophic niche size increased in more fragmented SAV seascapes, suggesting diversification of trophic roles and energy flow pathways. Pinfish collected from more continuous SAV habitats had smaller trophic niche size and higher trophic levels. Both trophic response metrics manifested a threshold response that depended on distinct SAV spatial characteristics (amount vs. spatial configuration) and environmental conditions. Our results suggest that habitat fragmentation of SAV seascape structure has ecological implications that could affect energy flow with cascading consequences for food web stability and ecosystem functioning.
... This is true in other coastal ecosystems as well, such as seagrass beds, which have seen habitat modification through altered light and salinity regimes, as well as by eutrophication from coastal development, human population growth, and water management practices (Waycott et al., 2009;Santos et al., 2011;Hall et al., 2016). Importantly, coastal habitats support many different species, and habitat loss from disturbance events has led to shifts and losses in consumer biomass (Boström et al., 2011;Yeager et al., 2016;Santos et al., 2018). ...
... Though we found minimal spatial variation in basal resource use, we did find spatial variation in species occurrence, and there was variation in basal resource use among species (Table 3). The seagrass die-off could have altered food web structure, especially at higher trophic levels, through a shift in the amount or configuration of seagrass habitat which has been shown to influence species occurrence through alterations in habitat quality, predation efficiency, and competition (Fahrig, 2003;Boström et al., 2011;Santos et al., 2018). Patterns in species occurrence could also be due to the spatial variability in environmental conditions across the bay (Sheridan et al., 1997;Kelble et al., 2021). ...
Macrophyte foundation species provide both habitat structure and primary production, and loss of these habitats can alter species interactions and lead to changes in energy flow in food webs. Extensive seagrass meadows in Florida Bay have recently experienced a widespread loss of seagrass habitat due to a Thalassia testudinum mass mortality event in 2015 associated with prolonged hypersalinity and bottom-water anoxia. Using stable isotope analysis paired with Bayesian mixing models, we investigated the basal resource use of seven species of seagrass-associated consumers across Florida Bay in areas affected by the 2015 seagrass die-off. Three years after the die-off, basal resource use did not differ for species collected inside and outside the die-off affected areas. Instead, consumers showed seasonal patterns in basal resource use with seagrass the most important in the wet season (58%), while epiphytes were the most important in the dry season (44%). Additionally, intraspecific spatial variability in resource use was lower in the wet season compared to the dry season. We were unable to detect a legacy effect of a major disturbance on the basal resource use of the most common seagrass-associated consumers in Florida Bay.
... Within many tropical and subtropical regions, saltmarshes, mangroves and seagrass meadows facilitate these key biogeochemical roles that in turn determine ecosystem productivity and ecological status (Wang and Gu, 2021b). These vegetated habitats transition along the estuary principally in response to salinity (Santos Rolando et al., 2018), but are also zoned according to a range of other factors, such as geomorphology and hydrodynamic conditions unique to the estuary that, across the spring-neap tidal cycle, can influence sediment deposition/resuspension and inundation/exposure (Lokhorst et al., 2018;Reed et al., 2020;Santos Rolando et al., 2018). Anthropogenic perturbations (e.g., pollution, disease, temperature extremes) are also known to influence these habitats and their structure (Goode et al., 2020;Hillyer et al., 2021). ...
... Within many tropical and subtropical regions, saltmarshes, mangroves and seagrass meadows facilitate these key biogeochemical roles that in turn determine ecosystem productivity and ecological status (Wang and Gu, 2021b). These vegetated habitats transition along the estuary principally in response to salinity (Santos Rolando et al., 2018), but are also zoned according to a range of other factors, such as geomorphology and hydrodynamic conditions unique to the estuary that, across the spring-neap tidal cycle, can influence sediment deposition/resuspension and inundation/exposure (Lokhorst et al., 2018;Reed et al., 2020;Santos Rolando et al., 2018). Anthropogenic perturbations (e.g., pollution, disease, temperature extremes) are also known to influence these habitats and their structure (Goode et al., 2020;Hillyer et al., 2021). ...
Full-text available
Vertical zonation within estuarine ecosystems can strongly influence microbial diversity and function by regulating competition, predation, and environmental stability. The degree to which microbial communities exhibit horizontal patterns through an estuary has received comparatively less attention. Here, we take a multi-omics ecosurveillance approach to study environmental gradients created by the transition between dominant vegetation types along a near pristine tropical river system (Wenlock River, Far North Queensland, Australia). The study sites included intertidal mudflats fringed by either saltmarsh, mangrove or mixed soft substrata habitats. Collected sediments were analyzed for eukaryotes and prokaryotes using small sub-unit (SSU) rRNA gene amplicons to profile the relative taxonomic composition. Central carbon metabolism metabolites and other associated organic polar metabolites were analyzed using established metabolomics-based approaches, coupled with total heavy metals analysis. Eukaryotic taxonomic information was found to be more informative of habitat type. Bacterial taxonomy and community composition also showed habitat-specificity, with phyla Proteobacteria and Cyanobacteria strongly linked to mangroves and saltmarshes, respectively In contrast, metabolite profiling was critical for understanding the biochemical pathways and expressed functional outputs in these systems that were tied to predicted microbial gene function (16S rRNA). A high degree of metabolic redundancy was observed in the bacterial communities, with the metabolomics data suggesting varying degrees of metabolic criticality based on habitat type. The predicted functions of the bacterial taxa combined with annotated metabolites accounted for the conservative perspective of microbial community redundancy against the putative metabolic pathway impacts in the metabolomics data. Coupling these data demonstrates that habitat-mediated estuarine gradients drive patterns of community diversity and metabolic function and highlights the real redundancy potential of habitat microbiomes. This information is useful as a point of comparison for these sensitive ecosystems and provides a framework for identifying potentially vulnerable or at-risk systems before they are significantly degraded.
... Across the global tropics, remote sensing data from air-and space-borne sensors have revealed the complex spatial and temporal patterns in the biological responses of corals to marine heat waves (Page et al. 2019). Such complex changes emerging at multiple scales justify the application of pattern-oriented scientific methods in attempts to understand and predict the consequences of changing seascape structure on ecological functions (Wu 2019, Bryan-Brown et al. 2020 and to identify spatial threshold effects (Yeager et al. 2016, Santos et al. 2018. Bridging science and practice for a better understanding of change will require innovative and integrative spatial frameworks with pattern-oriented indicators to inform spatial planning, restoration design, and ecosystembased climate adaptation strategies (Babí Almenar et al. 2018, Paulo et al. 2019. ...
... Seascape ecology recognises that environmental change plays out as a pattern-forming ecological process operating across multiple scales (Levin 1992). The application of concepts, spatial models, and spatial pattern metrics from landscape ecology has been transformative in understanding coastal ecosystem dynamics at spatial scales that are operationally relevant to management decision making (Browder et al. 1985, Costanza et al. 1990, Hovel & Regan 2008, Santos et al. 2018. Advances in computation are continually improving efforts to incorporate more complex patterns and processes into modelling at finer resolutions and across broader spatial and temporal scales. ...
... Biscayne Bay is a large (1110 km 2 ), shallow (depths generally < 3 m), subtropical lagoon system located downstream of the Florida Everglades and is surrounded by the city of Miami metropolitan area (population ~2.7 million) and multiple protected areas that preserve natural shorelines (Fig. 1). Thus, Biscayne Bay is heterogenous in urbanization and habitat distribution with distinct gradients of natural and anthropogenic stressors (Santos et al. 2011(Santos et al. , 2015(Santos et al. , 2018Lirman et al. 2014). Its western shoreline extends approximately 56 km from north to south. ...
Full-text available
The recreational flats fishery (bonefish, tarpon, and permit) in South Florida is economically and culturally important and has declined recently for unknown reasons. Biscayne Bay is a shallow subtropical lagoon system with a flats fishery bordered by a large urban center. The Bay also supports commercial fisheries, including the pink shrimp bait and food fisheries. These two shrimp fisheries represent Biscayne Bay’s most valuable fisheries, but how these fisheries interact with the recreational flats fishery is relatively unknown. We conducted a literature review to identify the potential direct and indirect effects of the two shrimp fisheries on the recreational flats fishery in the Bay. Our review found that there are likely minimal impacts of the Biscayne Bay pink shrimp fisheries on the flats fishery in Biscayne Bay since (a) the species are not caught by shrimping gear, (b) the shrimp fishery removes less than 10% of the Bay’s shrimp population, and (c) damage to seagrass is minimal (but hardbottom is damaged). Yet, the potential for indirect prey removal cannot be ruled out and requires quantification with additional diet data, food web, and mass balance models.
... Recent studies have identified increased chlorophyll-a and phosphate concentrations within the bay, which is more evident throughout the northern area and in nearshore areas of central Biscayne Bay, suggesting an urgent need for land use and land cover management to reduce local nutrient wash-off from the watershed to the Bay (Millette et al. 2019;Swart et al. 2013;Caccia and Boyer 2007). Santos et al. (2018) established that freshwater discharges into nearshore areas (contaminated by anthropogenic disturbances) have resulted in the fragmentation of the spatial patterning of submerged aquatic vegetation, which is thought to influence the distribution, community composition, and behavior of marine fauna. Man-made canals and waterways carry excess run-off and contaminants, from inland watersheds to Biscayne Bay. ...
Full-text available
Southeast Florida (SF) is among the most vulnerable regions to sea-level rise in the United States of America. The consequences associated with sea-level rise (SLR) are already apparent, including coastal inundation and erosion. The Coral Gables Canal watershed is located in SF and can be considered representative of the effects of combined mean and extreme SLR. In this research, the effect of concurrent mean and extreme sea-level rise on coastal inundation in the Coral Gables Canal watershed is explored. A three-dimensional hydrodynamic model for Biscayne Bay and the Coral Gables Canal is presented. The model is used to estimate water surface elevations throughout the model domain, and map inundation due to an extreme water-level event (Irma Hurricane) occurring alongside mean SLR scenarios. A comparison of the inundation coverage calculated in this research to estimations made by several online tools shows that the online simulators underestimate flooding areas by 72% to 85%. This is a consequence of underpredicting maximum water surface elevations occurring under combined SLR in the Coral Gables Canal. The model predicts that under the NOAA Intermediate High SLR scenario (year 2100), 40% of the CGC watershed will be inundated (water depths > 0.6 m), and 70% of the area will be flooded with water depths greater than 1.6 m in year 2120. Under the NOAA High SLR scenario at least 70% of the Coral Gables Canal watershed would be inundated in 2100 (water depths > 1.0 m). In year 2120, 90% of inland sub-basins will be flooded (0.6 m < depths < 2.2 m). These results are significant for planning flooding/inundation risk management strategies.
... Deployment of satellites with spectral bands specialized for high-resolution coverage of coastal environments (e.g., WorldView-2 and WorldView-3) now allows mapping of both land and below-water tidal marsh habitats (Rapinel et al. 2014;Santos et al. 2018). Incorporation of spectral sensors with drone technology also supports collection of high- Bulk tissue and compound-specific stable isotopes ...
Full-text available
Over the last 20 years, innovations have led to the development of exciting new technologies and novel applications of established technologies, collectively increasing the scale, scope, and quality of research possible in tidal marsh systems. Thus, ecological research on marshes is being revolutionized, in the same way as ecological research more generally, by the availability of new tools and analytical techniques. This perspective highlights current and potential applications of novel research technologies for marsh ecology. These are summarized under several themes: (1.) imagery — sophisticated imaging sensors mounted on satellites, drones, and underwater vehicles; (2.) animal tracking — acoustic telemetry, passive integrated transponder (PIT) tags, and satellite tracking, and (3.) biotracers — investigation of energy pathways and food web structure using chemical tracers such as compound-specific stable isotopes, isotope addition experiments, contaminant analysis, and eDNA. While the adoption of these technological advances has greatly enhanced our ability to examine contemporary questions in tidal marsh ecology, these applications also create significant challenges with the accessibility, processing, and synthesis of the large amounts of data generated. Implementation of open science practices has allowed for greater access to data. Newly available machine learning algorithms have been widely applied to resolve the challenge of detecting patterns in massive environmental datasets. The potential integration on digital platforms of multiple, large data streams measuring physical and biological components of tidal marsh ecosystems is an opportunity to advance science support for management responses needed in a rapidly changing coastal landscape.
Full-text available
Energetic resources and habitat distribution are inherently linked. Energetic resource availability is a major driver of the distribution of consumers, but estimating how much specific habitats contribute to the energetic resource needs of a consumer can be problematic. We present a new approach that combines remote sensing information and stable isotope ecology to produce maps of energetic resources (E‐scapes). E‐scapes project species‐ specific resource use information onto the landscape to classify areas based on energetic importance. Using our E‐scapes, we investigated the relationship between energetic resource distribution and white shrimp distribution and how the scale used to generate the E‐scape mediated this relationship. E‐scapes successfully predicted the size, abundance, biomass, and total energy of a consumer in salt marsh habitats in coastal Louisiana, USA at scales relevant to the movement of the consumer. Our E‐scape maps can be used alone or in combination with existing models to improve habitat management and restoration practices and have potential to be used to test fundamental movement theory.
Full-text available
Seagrass meadows and salt marshes are some of the most productive coastal habitats. With their global distribution they provide an important function for society through the provision of a number of goods and services. Both habitat types form geographically extensive seascapes commonly consisting of complex mosaics of vegetated patches interspersed among less structurally complex unvegetated sediments. Here we provide a theoretical framework for understanding the effects of the spatial configuration of seagrass and saltmarsh seascapes on marine fauna, and disentangle the complex mechanisms creating and maintaining the spatial configuration of seagrass and salt marsh vegetation. Although fish and invertebrates may show positive responses to habitat patch size, there are numerous examples of mixed results and non-linear responses, many of which appear to be species or life-stage specific, with large seasonal variation. Seagrass edges alter both physical and biological processes and may enhance larval settlement and fish predation. Salt marsh tidal channel edges play an important role for distribution of fish and shrimps, typically exhibiting a positive correlation between the total amount of marsh edge and nekton abundance and productivity. We show that seagrass meadows and tidal wetlands are interlinked through passive flows, active faunal movements and trophic interactions, with significant implications for management. We identify future research directions and challenges in seascape ecology.
Full-text available
Seascapes can be represented as hierarchically structured mosaics of relatively discrete areas (patches) recognisable and fluctuating across different spatial and temporal scales. Fluctuations are a result of disturbance regimes (frequency, severity and predictability), and temporal variability in colonisation, competition and dieback. The resulting shifting mosaics can result in complex ecological patterning and as such, patch dynamics are considered to be of profound importance in seascape ecology. The quantification of patch dynamics and the evaluation of change is required to understand and predict the effects of pattern on ecological processes, analyse differences between communities, and for assessing the implications for ecosystem resilience. This chapter reviews the evolution of seascape dynamics and fragmentation theory through a review of existing studies. Methods for quantifying and investigating dynamic spatial patterning and fragmentation are examined with illustrated examples and the future directions discussed with reference to ecological consequences of fragmentation and climate change effects on spatial patterning.
Full-text available
Background Blue sharks (Prionace glauca) are among the most abundant and widely distributed of oceanic elasmobranchs. Millions are taken annually in pelagic longline fisheries and comprise the highest component of auctioned fin weight in the international shark fin trade. Though studies of blue sharks outnumber those of other large pelagic sharks, the species’ complicated and sexually segregated life history still confound current understanding of Atlantic movement patterns. Lack of detailed information regarding movement and vertical behavior continues to limit management efforts that require such data for stock assessment and sustainable catch modeling. Therefore, this study aims to describe behavioral and ecological patterns distinct to aggregating and migrating blue sharks, and compare the findings to existing Atlantic movement models. Results Data collected from 23 blue sharks instrumented with pop-up satellite archival tags were used in statistical predictive regression models to investigate habitat use during a localized aggregation in the northwest Atlantic, while undergoing seasonal migrations, and with respect to environmental variables. Deployment durations ranged from 4 to 273 days, with sharks inhabiting both productive coastal waters and the open ocean, and exhibiting long-distance seasonal movements exceeding 3700 km. While aggregating on the continental shelf of the northwest Atlantic, blue sharks displayed consistent depth use independent of sex and life stage, and exhibited varied response to environmental (temperature and chlorophyll a) factors. As sharks dispersed from the aggregation site, depth use was influenced by bathymetry, latitude, demography, and presence in the Gulf Stream. Mature females were not observed at the New England tagging site, however, two mature females with recent mating wounds were captured and tagged opportunistically in The Bahamas, one of which migrated to the Mid-Atlantic Ridge. Conclusions Vertical behaviors displayed by blue sharks varied greatly among locales; depth use off the continental shelf was significantly greater, and individuals exhibited a greater frequency of deep-diving behavior, compared to periods of aggregation on the continental shelf. Sexual segregation was evident, suggesting mature and immature males, and immature females may be subjected to high levels of anthropogenic exploitation in this region during periods of aggregation. Analysis of the spatio-temporal tracks revealed that nine individuals traveled beyond the United States EEZ, including a mature female captured in The Bahamas that migrated to the Mid-Atlantic Ridge. These results reflect and augment existing Atlantic migration models, and highlight the complex, synergistic nature of factors affecting blue shark ecology and the need for a cooperative management approach in the North Atlantic. Electronic supplementary material The online version of this article (doi:10.1186/s40462-017-0107-z) contains supplementary material, which is available to authorized users.
Full-text available
Human-driven habitat fragmentation is cited as one of the most pressing threats facing many coastal ecosystems today. Many experiments have explored the consequences of fragmentation on fauna in one foundational habitat, seagrass beds, but have either surveyed along a gradient of existing patchiness, used artificial materials to mimic a natural bed, or sampled over short timescales. Here, we describe faunal responses to constructed fragmented landscapes varying from 4-400 m2 in two transplant garden experiments incorporating live eelgrass (Zostera marina L.). In experiments replicated within two subestuaries of the Chesapeake Bay, USA across multiple seasons and non-consecutive years, we comprehensively censused mesopredators and epifaunal communities using complementary quantitative methods. We found that community properties, including abundance, species richness, Simpson and functional diversity, and composition were generally unaffected by the number of patches and the size of the landscape, or the intensity of sampling. Additionally, an index of competition based on species co-occurrences revealed no trends with increasing patch size, contrary to theoretical predictions. We extend conclusions concerning the invariance of animal communities to habitat fragmentation from small-scale observational surveys and artificial experiments to experiments conducted with actual living plants and at more realistic scales. Our findings are likely a consequence of the rapid life histories and high mobility of the organisms common to eelgrass beds, and have implications for both conservation and restoration, suggesting that even small patches can rapidly promote abundant and diverse faunal communities.
The pink shrimp (Farfantepenaeus duorarum) has been selected as an ecological indicator to assess ecological effects on estuaries of implementation of the Comprehensive Everglades Restoration Plan that seeks to restore historical freshwater flows and nearshore salinity regimes in southern Florida. Concern over altered freshwater delivery impacts on pink shrimp productivity was expressed as early as the 1960s. The present review assessed pink shrimp scientific literature of the past 75+ years (>500 publications) to glean information relevant to understanding potential influence of freshwater management on pink shrimp productivity. The review was organized around “Essential Fish Habitat” metrics concerning abundance, growth, survival, distribution, productivity, and behavior. It summarizes previous pink shrimp field, laboratory, and modeling studies. Where possible, statistical analyses and meta-analyses of previously published data were performed to investigate consistency among independent findings. Pink shrimp occur in a wide range of salinities (0.5–67 ppt). A majority of studies (53.3%) reported maximal abundance between ∼20 to 35 ppt salinities. One laboratory study reported maximal growth at 30 ppt. Meta-analysis of reported growth rates did not yield results due to non-convergence of regression models. Reported survival was maximal at ∼30 ppt and remained high (>80% survival) across salinities of ∼15 to 40 ppt. A regression model that combined survival data across studies confirmed a previously reported parabolic relationship between salinity and survival; in this regression, 35 ppt maximized survival. Productivity, conditional upon survival and growth, was maximized at polyhaline (18–30 ppt) conditions. Inshore hypersalinity (>40 ppt) may elicit young pink shrimp behavioral cues counterproductive to settlement in nearshore areas. Virtually no information exists regarding postlarval pink shrimp movement or preference relative to salinity gradients. Realization and preservation of nearshore polyhaline conditions and elimination of hypersalinity should maximize growth, survival, and density, thus improving pink shrimp productivity. New and updated statistical models predicting pink shrimp distribution, abundance, growth, survival, and productivity relative to salinity conditions are needed to better guide freshwater management decisions.
Habitat fragmentation involves habitat loss concomitant with changes in spatial configuration, confounding mechanistic drivers of biodiversity change associated with habitat disturbance. Studies attempting to isolate the effects of altered habitat configuration on associated communities have reported variable results. This variability may be explained in part by the fragmentation threshold hypothesis, which predicts that the effects of habitat configuration may only manifest at low levels of remnant habitat area. To separate the effects of habitat area and configuration on biodiversity, we surveyed fish communities in seagrass landscapes spanning a range of total seagrass area (2-74% cover within 16 000-m2 landscapes) and spatial configurations (1-75 discrete patches). We also measured variation in fine-scale seagrass variables, which are known to affect faunal community composition and may covary with landscape-scale features. We found that species richness decreased and the community structure shifted with increasing patch number within the landscape, but only when seagrass area was low (<25% cover). This pattern was driven by an absence of epibenthic species in low-seagrass-area, highly patchy landscapes. Additional tests corroborated that low movement rates among patches may underlie loss of vulnerable taxa. Fine-scale seagrass biomass was generally unimportant in predicting fish community composition. As such, we present empirical support for the fragmentation threshold hypothesis and we suggest that poor matrix quality and low dispersal ability for sensitive taxa in our system may explain why our results support the hypothesis, while previous empirical work has largely failed to match predictions.
This work provides in-depth analysis of the origins of landscape ecology and its close alignment with the understanding of scale, the causes of landscape pattern, and the interactions of spatial pattern with a variety of ecological processes. The text covers the quantitative approaches that are applied widely in landscape studies, with emphasis on their appropriate use and interpretation. The field of landscape ecology has grown rapidly during this period, its concepts and methods have matured, and the published literature has increased exponentially. Landscape research has enhanced understanding of the causes and consequences of spatial heterogeneity and how these vary with scale, and they have influenced the management of natural and human-dominated landscapes. Landscape ecology is now considered mainstream, and the approaches are widely used in many branches of ecology and are applied not only in terrestrial settings but also in aquatic and marine systems. In response to these rapid developments, an updated edition of Landscape Ecology in Theory and Practice provides a synthetic overview of landscape ecology, including its development, the methods and techniques that are employed, the major questions addressed, and the insights that have been gained."
Landscape ecology is the study of processes occurring across spatially heterogeneous mosaics and the biotic responses to the resulting pattern. Spatial mosaics are made up of structural elements, biotic, and/or abiotic, which produce a set of patches set in a homogeneous matrix. Quantitative analyses of spatial and temporal patterns resulting from patch dynamics form the basis of landscape ecology. A landscape is larger than an individual's immediately observable area (Allen, 1998) and landscape studies typically address heterogeneity at very large spatial scales relative to the organism or process of interest.
Ecologists have used a variety of comparative mensurative and manipulative experimental approaches to study the biological consequences of habitat fragmentation. In this paper, we evaluate the merits of the two major approaches and offer guidelines for selecting a design. Manipulative experiments rigorously assess fragmentation effects by comparing pre- and post-treatment conditions. Yet they are often constrained by a number of practical limitations, such as the difficulty in implementing large-scale treatments and the impracticality of measuring the long-term (decades to centuries) responses to the imposed treatments. Comparative mensurative studies generally involve substituting space for time, and without pre-treatment control, can be constrained by variability in ecological characteristics among different landscapes. These confounding effects can seriously limit the strength of inferences. Depending on the scale of the study system and how "landscape" is defined, both approaches may be limited by the difficulty of replicating at the landscape scale. Overall, both mensurative and manipulative approaches have merit and can contribute to the body of knowledge on fragmentation. However, from our review of 134 fragmentation studies published recently in three major ecological journals, it is evident that most manipulative and mensurative fragmentation experiments have not provided clear insights into the ecological mechanisms and effects of habitat fragmentation. We discuss the reasons for this and conclude with recommendations for improving the design and implementation of fragmentation, experiments.