1 
 
Artificial structures versus mangrove prop roots: a general comparison of epifaunal 1 
communities within the Indian River Lagoon, Florida, USA.   2 
 3 
Dean S Janiak1*, Richard W Osman2, Christopher J Freeman1, Valerie J Paul1 4 
 5 
1Smithsonian Marine Station, Ft. Pierce, Florida 34949 USA 6 
2Smithsonian Environmental Research Center, Edgewater, Maryland 21037 USA 7 
 8 
Running Page Head: Comparison of artificial and mangrove habitat. 9 
 10 
*Corresponding author’s email: janiakd@si.edu 11 
 12 
Abstract 13 
 Urbanized coastal landscapes are becoming increasingly widespread throughout the 14 
world, and as a result essential habitat is being replaced with artificial structures.  Mangroves are 15 
threatened globally, and crucial ecosystem functions are being lost (e.g. habitat/refuge for 16 
associated species).  There remains a lack of understanding of how artificial structures function 17 
as habitat compared to natural substrates, particularly those being lost, such as mangrove prop 18 
roots.  The objectives of this study were to compare benthic epifaunal communities on artificial 19 
structures to those on mangrove prop roots across a large spatial scale and to assess seasonal 20 
trends of colonizing species within each habitat.  Identified species were also classified as either 21 
native or non-native to assess whether artificial structures harbor more non-natives compared to 22 
mangroves.  Results indicated that community composition differed significantly between habitat 23 
 2 
 
types as did richness and diversity.  More species were found and in higher percent cover on 24 
artificial structures.  Only a few species were dominant throughout the study and were present in 25 
both habitats, and these species varied in their abundance across sites and time.  Colonization at 26 
all sites was continuous throughout each season, particularly for those dominant species.  Non-27 
native species richness made up 30 – 50% of the community composition in each of the habitats 28 
and was significantly higher on artificial structures.  Overall, artificial structures appeared to 29 
provide a functional, unique surface for both dominant and rare species and could act as a buffer 30 
to biodiversity loss in a globally-threatened habitat.       31 
 32 
Introduction 33 
 Over the last several decades, humans have drastically altered the coastal marine 34 
landscape (Bulleri & Chapman 2010, Gittman et al. 2015).  With nearly one-third of the human 35 
population living within 100 km of the shoreline, rapid urban sprawl is resulting in concomitant 36 
losses of vital natural habitat and ecosystem services (Gittman et al. 2016).  Such losses are not 37 
habitat-specific and include seagrass beds (Orth et al. 2006), oyster reefs (zu Ermgassen et al. 38 
2012), coral reefs (Bellwood et al. 2004), and mangrove forests (Alongi 2002), with much of the 39 
damage resulting from anthropogenic influences.  Of these threatened ecosystems, mangrove 40 
forests are a unique foundational species found along the coastal shore throughout the tropics 41 
supporting marine and terrestrial biodiversity at all trophic levels (Kathiresan & Bingham 2001).  42 
Mangroves are globally threatened by a variety of human activities including aquaculture 43 
construction, timber harvesting, and coastal development (Alongi 2002).  Over the past four 44 
decades, there has been a steady global decline in mangrove forests with an estimated loss of 45 
35% (Valiela et al. 2001).  The worldwide destruction of mangrove forests is a major concern 46 
 3 
 
because they offer a diverse set of ecosystem provisions including raw materials for humans, 47 
coastal protection from storms, and habitat for a variety of terrestrial and marine species in 48 
shallow water environments (Barbier et al. 2011).  Specifically, prop roots and pneumatophores 49 
of some species extend into the water column and provide a complex, three-dimensional hard 50 
substrate for a diverse suite of invertebrates and a vital refuge for many juvenile fish species 51 
(Nagelkerken et al. 2008). 52 
Coastal ecosystems are also threatened by the construction of artificial structures, which 53 
in most cases replace natural shoreline, and can have local and regional ecological consequences.  54 
Artificial structures are implemented to harden coastal areas for protection of property against 55 
erosion, storms, flooding, sea-level rise, or for recreation, and are rarely built using ecological 56 
engineering which could reduce adverse impacts to the environment (Dafforn et al. 2015).  The 57 
inclusion of artificial structures in the marine environment is a global problem and is only 58 
expected to increase.  In the US alone, over 14% of the shoreline has been modified to some 59 
degree with the majority occurring in sheltered lagoons and estuaries (Gittman et al. 2015).  60 
Because of the ubiquity of artificial structures throughout the world, there has been a recent 61 
interest in how these hard structures function as habitat for marine species.  These additions 62 
provide a novel substrate in systems that are typically dominated by soft sediments.  Natural hard 63 
surfaces such as rocky or coral reefs, oysters, and mangrove prop roots are the primary habitat 64 
for a highly diverse suite of encrusting and mobile invertebrates and algae.  In areas where both 65 
natural and artificial habitat are present, comparisons between habitats are critical to better 66 
understand the potential negative effects of habitat displacement. This is important in predicting 67 
how continued shoreline hardening and the loss of natural habitat will affect the biodiversity and 68 
function of estuaries and coastal shores.       69 
 4 
 
Comparisons of marine communities in natural and artificial habitats have shown that 70 
they are strongly dissimilar and vary in their species abundance and composition (Connell & 71 
Glasby 1999, Connell 2001, Smith & Rule 2002).  Communities on artificial structures tend to be 72 
less diverse than those of natural hard substrates (Bacchiocchi & Airoldi 2003, Chapman 2003, 73 
Moschella et al. 2005).  Differences in community composition between substrate types can be 74 
due to differing ecological processes.  For example, artificial structures are found at different 75 
heights within the water column (Connell 2001), have a unique orientation, and are typically 76 
constructed of novel materials, all of which can influence recruiting species and community 77 
composition (Glasby & Connell 2001, Bulleri & Chapman 2010).  Artificial structures tend to 78 
favor large populations of non-native species (Lambert & Lambert 2003, Glasby et al. 2007, 79 
Tyrrell & Byers 2007, Airoldi et al. 2015) that are often well adapted to exploit open space 80 
(Simkanin et al. 2012).  These structures can also enhance dispersal abilities, acting as a stepping 81 
stone for non-natives to increase their spread (Bulleri & Airoldi 2005). This, along with a general 82 
reduction in native predators associated with artificial structures can have strong impacts on 83 
community development and overall community composition (Oricchio et al. 2016, Rogers et al. 84 
2016).     85 
In the Caribbean, southern Florida, and throughout the Gulf Coast, mangroves are a 86 
common component of the shoreline and provide a unique, critical habitat for many marine 87 
species.  Throughout this region, Rhizophora mangle is the dominant mangrove species with 88 
prop roots that extend into the subtidal, providing refuge for fish and substrate for a diverse suite 89 
in invertebrates.  Epifaunal communities on R. mangle have been well-studied in the Caribbean 90 
and shown to positively affect tree growth (Ellison et al. 1996), provide protection from infesting 91 
species (Ellison & Farnsworth 1990), and enhance overall ecosystem productivity (Nagelkerken 92 
 5 
 
et al. 2008).  The northern range of this species is limited by thermal tolerance to freezing and 93 
extends to the northern border of Florida (Cavanaugh et al. 2014).  Within the Indian River 94 
Lagoon (IRL), a subtropical estuary along the eastern shore of central Florida, mangrove forests 95 
are a significant component of the coast; however, as of 2007, 39% of the shoreline has been 96 
urbanized (Bricker et al. 2007).  The IRL has also been subjected to a variety of other human-97 
induced stressors, including water quality reduction, significant loss of seagrass beds, and 98 
continued algal blooms (Fletcher & Fletcher 1995, Nixon 1995, Lapointe et al. 2015), all of 99 
which have had lasting effects on the ecosystem.  The IRL is one of the most species-rich 100 
estuaries in North America (Swain et al. 1995); however, there exists very little data on the 101 
epifaunal communities associated with R. mangle roots.  There is also a general lack of data on 102 
established communities of marine species associated with artificial habitats in most parts of the 103 
world, including the IRL.  The objectives of the study were to (1) examine and compare the 104 
spatio-temporal trends of community composition and percent cover of different species in 105 
mangrove and artificial habitats within the IRL, (2) examine and compare the seasonal 106 
colonization of epifaunal species in the two different habitat types using recruitment panels, and 107 
(3) identify and compare native and non-native species presence within each of the habitats.                                     108 
  109 
Methods 110 
 Study Region 111 
 The Indian River Lagoon (IRL) is a shallow, narrow lagoon along the central eastern 112 
coast of Florida, USA, and serves as a transitional area between temperate and subtropical zones 113 
(Fig. 1).  It is comprised of three connected main water basins, the Mosquito Lagoon, the Indian 114 
River, and the Banana River, which together span a distance of 251 km from Ponce de Leon Inlet 115 
 6 
 
south to Jupiter Inlet.  The width of the lagoon varies from 2 - 4 km, and it has a depth range of 116 
approximately 1 - 3 m (Woodward-Clyde Consultants 1994).  The IRL has five inlets that allow 117 
estuarine/oceanic mixing to occur, while a sixth, located at a study site in Port Canaveral (B_03), 118 
is isolated from the lagoon by an on-demand lock system.  Besides B_03, all other sites are 119 
estuarine and located relatively far from inlets.  Tides are semi-diurnal and the tidal amplitude is 120 
approximately 10 - 30 cm with larger tides occurring closer in proximity to the inlets.  Where the 121 
shoreline has not been modified, mangrove forests tend to dominate throughout the Indian and 122 
Banana Rivers, becoming less abundant in the Mosquito Lagoon.  There is no strong salinity 123 
gradient throughout the IRL, but salinity is variable in certain regions resulting from proximity to 124 
oceanic and freshwater inputs and seasonality.  Environmental data including water temperature, 125 
salinity, and relative chlorophyll concentrations, gathered from the St. Johns Water Management 126 
District (http://webapub.sjrwmd.com/agws10/hdswq/), were monitored continuously from three 127 
sites (close proximity to sites M_06, IRL_02, and B_02) during the project within the study 128 
region (Fig. 2).   129 
 Seasonal Sampling of Epifaunal Communities 130 
 Sampling was conducted on a quarterly schedule from October 2014 to July 2016 (8 131 
sampling events).  Each sampling event occurred during a particular season: October (fall), 132 
January (winter), April (spring), and July (summer).  Twenty-nine sites spanning 150 km of the 133 
central and northern portions of the IRL were haphazardly chosen based on substrate type (either 134 
mangrove prop roots or artificial structures) and to account for a balanced coverage within the 135 
study region.  Eighteen sites, found in all three water basins, consisted of artificial habitat, and 136 
the other 11 sites, found in two of three water basins, consisted of dense Rhizophora mangle 137 
stands (Fig. 1).  All sites chosen had epifaunal communities that were subtidal and only exposed 138 
 7 
 
during the lowest tides of the year.  There were no mangrove sites in the Mosquito Lagoon as R. 139 
mangle stands are infrequent in this area, and those that are present lack dense subtidal roots.  140 
Artificial structures consisted of either wood pilings or seawalls constructed of concrete.  141 
Seawalls were either the bridge abutments (roughly 200 m in length) or supporting bridge pilings 142 
(< 50 m in length). The timing of construction for artificial structures could not be estimated, 143 
although all sites chosen appeared to have well-established epifaunal communities.  Despite 144 
differences in construction materials, the same methodology was used for data collection in this 145 
habitat type.  At each of the artificial structure sites, a GoPro video camera (1080p, 60fps) was 146 
used to record vertical video transects of sessile epifaunal communities.  The camera was 147 
attached to an L-shaped PVC frame and set at a fixed distance (25 cm) away from the substrate 148 
such that the same amount of area was recorded each time.  The shaft of the frame was 2 m long, 149 
and when handled from a boat, could record communities approximately 1.5 m below the water 150 
level, which corresponds to mean depth of the IRL.   151 
 Transects approximately 1 m in length were taken vertically to the sediment surface 152 
along three separate pilings at each site during each sampling event.  At sites with a seawall, 153 
triplicate transects were taken vertically to mimic video taken of pilings and were separated from 154 
each other by at least 1 m.  Epifaunal collections were also taken within each transect using a 155 
standard garden hoe and communities of approximately 200 - 300 cm2 were scraped off the 156 
surface.  Detached communities were collected into a 5 mm mesh net held below the scraped 157 
area.  Samples were not rinsed or sieved and were immediately placed in separate bags, put on 158 
ice, and brought back to the laboratory for species identification and to account for rare or small 159 
species not visible in the video.  There was no visible species loss resulting from the mesh size 160 
used.  From the video, a single screen capture was taken from each transect that was both of high 161 
 8 
 
resolution and representative of the surrounding community.  A 250-cm2 frame was placed 162 
around each screen capture, and percent cover was estimated using 100 randomly assigned points 163 
using the point-count software CPCe 4.1 (Kohler & Gill 2006).  At mangrove sites, the ends of 164 
three subtidal prop roots with epifauna were cut at a standard size (25 cm in length, roughly an 165 
area of 250 cm2) and brought back to the laboratory where each root was photographed for a 166 
permanent record and examined under a dissecting microscope.  All invertebrates present were 167 
identified and a visual estimate of percent cover (0-100) was made.     168 
  Quarterly Sampling of Colonizing Species 169 
 Colonization panels were deployed during the same intervals as seasonal sampling to 170 
measure seasonality of recruiting epifaunal species over time and between habitats.  When 171 
samples were first collected (October 2014), a set of colonization panels (100 cm2, PVC) were 172 
deployed in triplicate at 24 of the 29 sites.  Five sites were not used because of a lack of usable 173 
structure from which to deploy panels.  Because of the variation of habitat structure and 174 
construction at each of the sites, colonization panels were hung in a variety of configurations to 175 
optimize retrieval and stability.  At artificial structure sites, panels were either hung from a dock, 176 
with the colonization surface facing downward or wrapped around a piling with the colonization 177 
surface facing outward.  At mangrove sites, individual panels were attached to subtidal prop 178 
roots with the colonization surface facing outwards.  Regardless of habitat type or orientation, all 179 
panels were hung at the same distance from the bottom where seasonal samples were taken for 180 
each of the sites.  At each quarterly sampling event, panels were collected, placed into bags and 181 
put on ice with new panels being simultaneously deployed.  Panels were brought back to the 182 
laboratory, photographed, and examined under a dissecting microscope for species identification.  183 
 9 
 
Diversity, community composition, and percent cover of epifaunal species was estimated from 184 
photographs as described above and used as a proxy for colonization.   185 
 Statistical Analysis 186 
 Similar statistical methods were used for video transect and colonization panels to 187 
examine differences in communities.  For all analyses, sites for each particular habitat type were 188 
used as replicates for each season to get a robust estimate of community composition.  Included 189 
in all community analyses was the response variable open space, an important limiting factor in 190 
sessile benthic communities and a useful measure of seasonality.  Differences between the two 191 
types of habitat (artificial and mangrove) as well as how those differences varied over time were 192 
then compared.  To visually examine similarity of community composition over time and 193 
between the two habitat types, non-parametric multi-dimensional scaling (nMDS) was used.  An 194 
nMDS analysis was also conducted for each individual sampling event to examine specific site 195 
differences over time.  To examine differences in the abundances of percent cover in community 196 
composition, a non-parametric multivariate analysis of variance (PERMANOVA) was used to 197 
test the factors habitat and season.  When the interaction between habitat and season was 198 
significant, an analysis of similarity (SIMPER) was used to examine what particular species 199 
contributed the most dissimilarity between the two habitat types for each season (i.e. seasonal 200 
differences in species).  For those species that were found to be important or dominant within the 201 
overall study, a separate two-way analysis of variance (ANOVA) was used to examine 202 
differences in percent cover for the factors habitat and season.  To test for differences in richness 203 
and Shannon-Wiener diversity within communities, a two-way ANOVA was used for the factors 204 
habitat and season.  A two-way ANOVA was also used to test for differences in habitat and 205 
season for richness and Shannon-Wiener diversity for colonizing species during each quarterly 206 
 10 
 
sampling event.  When the interaction term was significant, a Student-Newman-Keuls pairwise 207 
comparison test was used to examine differences between habitat types for each season.   208 
A separate analysis was conducted for two sites, IRL_13mg (mangrove) and IRL_13a 209 
(artificial) (see Fig. 1), which were within 10 m of each other to examine communities between 210 
habitat types at a very small spatial scale.  The artificial habitat was a small dock (5 m) with a 211 
few wooden pilings surrounded by mangrove forest.  To visually examine similarities between 212 
communities and over time, an nMDS plot was constructed. To examine community composition 213 
in percent cover, a PERMANOVA analysis was conducted to test for differences in season and 214 
habitat.  A SIMPER analysis was also conducted to examine which variables caused the most 215 
dissimilarity between habitat types.   216 
Lastly, for all taxa identified to the species level from seasonal samples, a two-way 217 
ANOVA was used to examine differences in native and non-native species richness within the 218 
two habitats over time.  No analyses were done for colonizing species because artificial panels 219 
were being used as a substrate in both habitat types.  Native/non-native status for each species 220 
was assessed using WoRMS (WoRMS Editorial Board 2017), WRIMS (Pagad et al. 2017), and 221 
NEMESIS (Fofonoff et al. 2017) databases.   222 
All non-parametric analyses were conducted using Primer-E.  All parametric analyses 223 
were conducted using SigmaPlot v12.5.  All data were visually checked for normality and equal 224 
variances, and when assumptions were not met for parametric tests, data were transformed using 225 
either an acrsine square root transformation for percent cover or log transformation for 226 
continuous data to correct the issues.       227 
   228 
Results 229 
 11 
 
Seasonal Sampling of Epifaunal Communities  230 
 A total of 175 morphospecies, from 11 phyla, were identified from both habitats during 231 
the study (Table 1).  In artificial habitat, 164 taxa were found, while in mangroves, only 84 taxa 232 
were found. Out of all species found, 146 were identified to the species level (Table S1).  The 233 
most speciose group was ascidians (37 taxa), although the majority of these were found at only 234 
two sites, one near Port Canaveral inlet (B_03) and the other being the northern most site 235 
(M_01), near Ponce de Leon inlet.  Both sites were artificial structures and atypical in 236 
community composition.  Species richness was consistently greater in artificial habitats (habitat 237 
type: df = 1, F = 9.973, P = 0.002) and varied among seasons (df = 7, F = 26.773, P < 0.001) 238 
with a particularly strong decline in October 2015 (Fig. 3a).  Shannon-Wiener diversity showed 239 
some seasonal variability (df = 7, F = 13.108, P < 0.001), but otherwise there were no 240 
statistically significant differences between the two habitat types (Fig. 3b).  Community 241 
composition for each habitat type generally clustered together, although there was no strong 242 
similarity within both habitats among seasons (Fig. 4).  Composition based on percent cover was 243 
statistically different for both habitat (df = 1, Pseudo-F = 13.559, P(perm) < 0.001) and season 244 
(df = 7, Pseudo-F = 3.539, P(perm) < 0.001), and the magnitude of these differences varied over 245 
time (season x habitat, df = 7, Pseudo-F = 1.802, P(perm) < 0.001).           246 
         For each season, communities at the majority of sites clustered together, indicating that 247 
the same dominant species were present and in relatively similar abundances (Fig. S1).  Several 248 
sites were consistently dissimilar including M_01, the most northern site in the Mosquito Lagoon 249 
near Ponce de Leon inlet, and B_03, a large marina in the Banana River located adjacent to the 250 
Port Canaveral inlet.  Both sites were artificial habitats and in areas with strong water exchange 251 
and were composed of communities with higher diversity and low barnacle cover.    252 
 12 
 
 A SIMPER analysis was used to examine which species for each season were most 253 
important in the dissimilarity between habitat types (Table S2).  For most seasons, the barnacle, 254 
Amphibalanus eburneus, was the highest contributor to the dissimilarity of habitats with 255 
abundances typically higher in artificial habitat.  Throughout the entire study, A. eburneus was 256 
also the most dominant species found in both habitat types.  Barnacle percent cover was 257 
significantly different for season (df = 7, F = 4.069, P < 0.001) and habitat type (df = 1, F = 258 
37.946, P < 0.001) as well as for the interaction between the two (df = 7, F = 5.733, P < 0.001).  259 
Pairwise comparison tests indicated that for the majority of seasons, barnacle cover was 260 
significantly higher on artificial structures (Fig. 5a).  The bryozoans Conopeum chesapeakensis 261 
and C. tenuissimum (pooled as Conopeum spp.) comprised a significant proportion of the 262 
community throughout the study region.  Conopeum spp. were present in greater abundances in 263 
mangroves (Fig. 5b, df = 2, F = 10.331, P = 0.001).  These differences were not consistent 264 
though time (season, df = 7, F = 15.757, P < 0.001) with increased abundances found towards the 265 
end of the study (Fig. 5b).  Other less-significant contributing species were tube-building 266 
amphipods (Erichthonius brasiliensis and Monocorophium insidiosum), more abundant in 267 
mangroves, and hydroids (mostly Obelia spp.), more abundant in artificial habitat.  Other 268 
bryozoans including Schizoporella pungens and Victorella pavida were also abundant though 269 
differed between habitats. In artificial habitats, S. pungens was found in greater amounts, while 270 
V. pavida was found in greater amounts on mangrove prop roots (Table S2).  The amount of 271 
open space found between the two habitat types was an important factor in the dissimilarity 272 
when community composition was compared (Table S2).  Throughout the study, open space was 273 
consistently greater on mangrove prop roots (Fig. 5c, habitat type df = 1, F = 116.767, P < 0.001) 274 
and varied significantly by season (df = 7, F = 6.353, P < 0.001).   275 
 13 
 
A separate comparison for IRL_13mg and IRL_13a showed that replicates clustered 276 
together according to habitat type although this varied seasonally (Fig. 6).  Overall, communities 277 
significantly differed between the two habitat types (df = 1, Pseudo-F = 11.444, P(perm) = 278 
0.001) as well as during the different seasons (df = 7, Pseudo-F = 11.129, P(perm) = 0.001) 279 
despite being in close proximity to each other.  For some seasons, the communities in the two 280 
habitats were similar (10/14, 01/15, and 07/15), and for others, they were quite different (05/16 281 
and 08/16) indicating that the magnitude of difference was strongly dependent on season (df = 7, 282 
Pseudo-F = 5.846, P(perm) = 0.001).  An analysis of similarity (SIMPER) comparing the sites 283 
indicated that the barnacle, A. eburneus, open space, and Conopeum spp. were the most abundant 284 
species/variables at the sites and caused the largest dissimilarity between the habitats (Table 2).          285 
Quarterly Sampling of Colonizing Species 286 
Community composition measured as percent cover on recruitment panels was 287 
significantly different for season (df = 6, Pseudo-F = 3.995, P(perm) = 0.001) as well as for 288 
habitat type (df = 2, Pseudo-F = 3.406, P(perm) = 0.006), and the difference between habitats 289 
was consistent through time (season x habitat type, P(perm) = 0.759).  Richness for colonizing 290 
species varied by season (df = 6, F = 19.912, P < 0.001) and was generally greater in artificial 291 
habitats (df = 1, F = 18.413, P < 0.001), and these differences were consistent over time (season 292 
x habitat type, P = 0.681, Fig. 7a).  There was an overall difference in Shannon-Wiener diversity 293 
for season (df = 6, F = 7.737, P < 0.001) and habitat type (df = 7, F = 7.034, P = 0.008), although 294 
the difference between habitat types for each season was negligible (Fig. 7b) and no significance 295 
was found for pairwise comparisons.            296 
 An analysis of similarity (SIMPER) for communities revealed that A. eburneus, tube-297 
building amphipods, Conopeum spp., and open space were the most common variables measured 298 
 14 
 
within communities that caused dissimilarity among habitat types (Table S3), similar to what 299 
was found for epifaunal communities.  A. eburneus recruited in large numbers throughout the 300 
entire study, and abundances were significantly different for season (df = 6, F = 8.299, P < 301 
0.001) but not for habitat type (P = 0.056) or the interaction between the two (P = 0.191).  302 
Barnacles showed some seasonality with increased rates of colonization in the summer and 303 
decreases in the winter (Fig. 8a), however, colonization was continuous throughout the duration 304 
of the study.  The bryozoans Conopeum spp. recruited in large abundances in the late 305 
winter/early spring (January - April) and in particularly high amounts in 2016 following a brown 306 
tide (Fig. 2b and Fig. 8c), consistent with what was found for the established epifaunal samples.  307 
Conopeum spp. colonization was significantly different for season (df = 6, F = 30.047, P < 308 
0.001), habitat type (df = 1, F = 10.095, P = 0.002), and the interaction (df = 6, F = 8.565, P < 309 
0.001), indicating a seasonal trend for which the magnitude varied by habitat type.  Open space, 310 
or the seasonal rate of colonization, was different for season (df = 6, F = 4.496, P < 0.001), 311 
habitat type (df = 1, F = 6.257, P = 0.013) as well as the interaction between the two (df = 6, F = 312 
2.49, P = 0.022).  Open space tended to be relatively consistent on panels in artificial habitats 313 
while more variable on mangrove prop roots (Fig. 8c).   314 
Native-Non-native Comparison 315 
Of the 146 organisms identified to the species level, 97 were found to have a native range 316 
within the Florida/Caribbean region while 48 were classified as non-native (Table S1).  The 317 
remaining species were either not fully identified to the species level or had an unknown native 318 
range.  Differences in non-native species richness were found for habitat type, with more non-319 
natives found in artificial habitats (df = 1, F = 60.218, P < 0.001), and this varied by season (df = 320 
7, F = 24.022, P < 0.001) indicating some seasonality in colonization patterns (Fig. 9).   Non-321 
 15 
 
native species made up roughly 1/3 to 1/2 of the total community richness in each of the habitats 322 
for each season.  Non-natives that were relatively common and found in both habitats were the 323 
hydroid Obelia geniculata, the serpulid Ficopomatus enigmaticus, the barnacle Amphibalanus 324 
reticulatus, and the bryozoans Schizoporella pungens, Hippopodina indica, Victorella pavida, 325 
and Bugula neritina.    326 
 327 
Discussion     328 
Seasonal Sampling of Epifaunal Communities  329 
Mangrove forests are being lost at an alarming rate while man-made structures are 330 
becoming fairly ubiquitous, providing a novel habitat for epifaunal communities.  The goal of 331 
our study was to compare the epifaunal community structure on mangrove prop roots and 332 
artificial structures in the Indian River Lagoon (IRL).  There were strong differences in 333 
community composition and percent cover of individual species between the two habitats, and 334 
these results are similar to previous studies that demonstrate differences in communities in 335 
artificial and natural habitats (Torre & Targett 2016, Airoldi et al. 2015).  Artificial structures 336 
hosted a greater number of species, which generally covered a higher percentage of the substrate.  337 
This is in contrast to previous studies that have shown artificial structures have reduced species 338 
richness compared to natural substrates such as rocky shores (Connell & Glasby 1999, Chapman 339 
2003, Bulleri & Chapman 2010).  Mangroves roots are relatively small and discrete compared to 340 
rocky shores, and this may in part explain the differences in species richness.   341 
Unlike mangrove root communities in the nearby Florida Keys that show large amounts 342 
of variation in the short-term (Bingham & Young 1995), communities on mangroves in the IRL 343 
remained relatively similar through space and time.  Communities were dominated by species 344 
 16 
 
that are generally thought of as stress tolerant, similar to what has been found on a global scale 345 
(Australia: Bishop et al. 2012, Kenya: Crona et al. 2006, Costa Rica: Perry 1988, Jamaica: Elliot 346 
et al. 2012, and Philippines: Salmo et al. 2017) and were mainly composed of barnacles, 347 
bryozoans, hydroids, and tube-dwelling amphipods.  Similarly, the same dominant species were 348 
also found on artificial structures, though in larger percent cover.  Strong seasonal differences 349 
were also found for the minor taxa (< 10% of community), and unlike the dominant species, the 350 
composition of these species differed between the habitats.  Overall, community composition 351 
varied in two ways: 1) the abundance of the dominant taxa differed with percent cover being 352 
higher in artificial habitats and 2) the composition of the less-abundant taxa differed by habitat as 353 
well as by season (see Table S2).   354 
At a local scale, community differences between artificial structures and mangrove roots 355 
were also evident.  One site in the central Indian River, IRL_13, a small 5 m long wooden dock 356 
with few pilings and surrounded by mangroves, contained both habitats.  Despite their close 357 
proximity, communities still differed significantly over some seasons though this was not 358 
consistent between years.  Because of the spatial scale of the study, many artificial structures 359 
sampled were pilings and relatively small docks similar to IRL_13a that were distant from larger, 360 
more urbanized areas.  It is common to study large-scale artificial habitats, which have a variety 361 
of intrinsic factors that make them more than likely to host unique communities, but our study 362 
highlights the unique role that even small artificial structures in remote areas play as substrate for 363 
invertebrate communities.  A large number of artificial structures were only present at a single 364 
site in Port Canaveral (B_03), which contained communities that were the most diverse, most 365 
likely because of the proximity to the inlet and decreased environmental variability (Attrill 2002, 366 
Mook 1980).  The majority of sites sampled were estuarine, which differs from previous research 367 
 17 
 
that have focused on communities on artificial structures that are more coastal or have a strong 368 
oceanic influence similar to the Port Canaveral site (Gittman et al. 2016).   369 
Globally, gastropods and decapod crustaceans are the most common mobile groups found 370 
in mangroves while sponges are typically the dominant sessile group (Cannicci et al. 2008).  371 
While these groups are relatively diverse in the IRL, in this study they were relatively rare.  The 372 
most dominant species found in communities in both habitats throughout the entire study region 373 
was the barnacle Amphibalanus eburneus.  The dominance of barnacles was consistent at both 374 
the habitat and site level and caused community composition to be somewhat similar through 375 
time.  Unlike the positive effects of sponges in the Caribbean (e.g. Ellison et al. 1996), barnacle 376 
fouling has been shown to have negative effects on root growth (Perry 1988) and leaf and stem 377 
morphology, resulting in a reduction in gas exchange (Li & Chan 2008).   378 
Open space was found in significantly greater amounts on mangrove prop roots with no 379 
evidence of seasonal changes.  Besides barnacle abundances, the percent of open space found 380 
caused communities to be consistently dissimilar (see Table S2).  Open space can be created or 381 
maintained by both abiotic and biotic means and can provide insight into why communities in 382 
contrasting habitat types can differ.  Physical forces are particularly strong determinants of 383 
community structure in shallow mangrove communities (Bingham & Young 1995, Farnsworth & 384 
Ellison 1996), and this is presumably the case here as well.  Many of the artificial structures 385 
sampled in this study were slightly deeper and generally off the shoreline.  Pilings were situated 386 
several meters from the coast and seawalls extended deeper, which in both cases caused 387 
environmental stress to be reduced (e.g. increased flow, less wave action, and reduced sand 388 
scouring).  Other potential differences that can influence the amount of open space are substrate 389 
availability and age.  Prop roots are constantly adding substrate as they grow and can be of 390 
 18 
 
various ages, whereas artificial structures are a constant size, and at least locally at each site, 391 
were constructed at the same time. 392 
In recent years, the northern IRL has been subjected to increased algal blooms that have 393 
had devastating effects on fisheries and seagrasses (Proffitt 2017 and references within).  During 394 
our study, a brown tide occurred from December 2015 – May 2016 at the majority of sites, 395 
though it appeared to have no major effects on the majority of species.  However, one particular 396 
bryozoan, Conopeum spp. (mostly C. chesapeakensis), which was present in small amounts prior 397 
to the bloom, was found in large abundances at all sites and in both habitat types during and 398 
post-bloom, and this carried through until the end of the study.  In the Chesapeake Bay, this 399 
species (reported as C. seurati) had its greatest growth rates when food availability was highest 400 
and nearing bloom conditions (O’Dea & Okamura 1999).  Besides two barnacle species (A. 401 
eburneus and A. reticulatus), this was the only other species that occurred in relatively high 402 
abundances at the majority of sites, but this was only the case when bloom conditions were 403 
present.  404 
Quarterly Sampling of Colonizing Species 405 
 Colonization panels were used in this study as a standardized way to examine seasonal 406 
changes in recruitment as well as to determine if the larval pool was capable of reaching both 407 
habitat types.  Both artificial and mangrove habitats showed a similar seasonal trend in species 408 
richness on panels, with richness levels greater in the summer coinciding with warmer 409 
temperatures.  Despite only two significantly different sampling events, the general trend was 410 
that richness was greater in artificial habitats, mainly resulting from rare or low-abundance 411 
species.  Barnacle recruitment was consistent and ranged from 20 - 40% cover in the winter to 50 412 
- 60% cover in the summer.  Barnacles have a relatively long-lived pelagic larval phase and 413 
 19 
 
should have good dispersal capabilities, reaching both habitat types.  Consistent recruitment of 414 
barnacles in high numbers is most likely the cause of its continued dominance in established 415 
communities.  Other species present on panels at the majority of sites for both habitat types 416 
included bryozoans, tube-building amphipods, and hydroids, similar to what was found in 417 
sampled communities.  Conopeum spp. were found more in mangrove sites and unlike barnacles, 418 
have a shorter pelagic larval duration.  On mangrove roots, Conopeum spp. were typically on the 419 
root itself while in artificial habitats, colonies were growing on barnacles. 420 
 Open space was generally found in greater amounts on panels in mangrove habitats.  421 
Colonization rates of barnacles for each habitat were similar and it was expected that the 422 
available open space on mangrove roots would be filled by barnacles, but this was not the case.  423 
Open space in this case does not necessarily reflect a lack of recruitment, and this suggests that 424 
other mechanisms are limiting colonization on mangroves roots either resulting from predation 425 
or environmental stress.  Mangroves are an important refuge for a variety of fish species, 426 
particularly in estuarine habitats (Faunce & Serafy 2006), and therefore, epifauna on mangrove 427 
roots are presumably heavily consumed.  Mangrove roots in the IRL may also be simply too 428 
shallow to support many species.  The amount of open space found on mangrove roots was 429 
similar to that found on panels deployed in mangrove habitat, indicating that overall, mangrove 430 
habitat in the IRL might be more stressful for epifaunal communities.     431 
Native Non-native Comparison 432 
The majority of non-natives within the marine environment are found in estuaries, where 433 
they are located on a variety of artificial structures (see review by Ruiz et al. 2009).  In our study, 434 
non-native species richness was found to be higher in artificial habitats as well, and this pattern 435 
was consistent through space and time.  The IRL is a biodiverse estuary and it is not surprising 436 
 20 
 
that non-natives were found; however, this group made up roughly one-third to one-half of the 437 
species richness at most sites.  Surprisingly, the number of non-natives present was similar in 438 
both habitat types and not restricted to only artificial structures.  The taxonomic group in which 439 
most non-natives occurred were the Bryozoa (12 out of 23 species), though in many studies 440 
ascidians are the most common group of non-natives found (Airoldi et al. 2015, Gittenberger & 441 
van der Stelt 2011) because of their strong competitive abilities (Blum et al. 2007, Janiak et al. 442 
2013). Unique to our study, the dominance of the native A. eburneus in artificial habitats greatly 443 
reduced the amount of suitable space for non-natives to utilize.  In most cases, bryozoans were 444 
able to grow on top of barnacles, which likely contributed to their persistence in communities.         445 
Specific non-natives were not persistently dominant in communities, though typically 446 
there was at least one species that contributed significantly to the community composition during 447 
each of the sampling events.  Our sampling design utilized a variety of substrate types including 448 
seawalls, individual pilings, as well as docks and supports the generality that non-native species, 449 
at least in terms of richness, are more prevalent on artificial structures.  Similar studies have also 450 
shown that artificial substrates favor non-natives (Glasby et al. 2007, Tyrrell & Byers 2007).  451 
The reasoning for this is still unclear, but it has been suggested that artificial structures are a 452 
unique form of substrate and favorable for non-natives (Simberloff 1997, Connell & Glasby 453 
1999) because of a lack of use by natives, reducing competition.  Our study suggests that when a 454 
dominant native species utilizes artificial structures, the potential for non-natives to dominate 455 
that space is reduced.  In our study, this was primarily due to the continuous recruitment of 456 
barnacles throughout both years reducing any potential available space.  Of particular 457 
importance, predation in artificial habitats has received little attention, though it has been shown 458 
that predators as well as overall consumption strength are generally reduced in artificial habitats 459 
 21 
 
(Able et al. 2013, Kornis et al. 2017, Rodemann & Brandl 2017) leading to a reduction in 460 
potential biotic control. 461 
Conclusions 462 
In general, the composition and percent cover of species on colonization panels was 463 
similar to that in the sampled communities and recruitment was not a limiting factor structuring 464 
these communities.  However, there was strong evidence that either predation or environmental 465 
stress altered community composition and contributed to the differing amount of open space 466 
between the two habitats.  If these trends were driven by predation, it would suggest that 467 
predators are less abundant in artificial habitats.  Such shifts in trophic community structure 468 
could have consequences in urbanized estuaries at a larger scale.  Examining higher trophic 469 
levels is therefore important and can improve our understanding of non-recruitment processes in 470 
shaping communities.          471 
The IRL, like many other estuaries around the world, is in constant transition resulting 472 
from increased urbanization and decreased ecosystem health (Sime 2005).  Mangroves along the 473 
Florida coastline are slowly extending their range northward into new areas (Cavanaugh et al. 474 
2014), however in most parts of the world, mangroves are on the decline.  Mangrove roots 475 
provide an important hard surface for marine epifauna in an otherwise sedimentary system, and it 476 
is important to understand how artificial structures, now globally ubiquitous, can function as 477 
potential habitat in areas where natural habitat is being lost.  Results from our study support the 478 
general trend that communities on artificial structures are distinct at both the local and regional 479 
scale from mangrove roots, but not necessarily in a negative way.  Artificial structures had 480 
higher species richness and abundances of dominant species and could provide an important hard 481 
structure to help maintain biodiversity when natural structures are being lost.  A caveat to this is 482 
 22 
 
that artificial structures may select for non-native species that could influence the community 483 
composition of dominant native species.   Most importantly, when considering how to maintain 484 
biodiversity in systems that are losing important foundation species such as mangroves, it is 485 
critical to understand the role that artificial structures play in the preservation of biodiversity as 486 
well as the spread of non-native species.   487 
 488 
Acknowledgements 489 
 We would like to thank Sherry Reed and Woody Lee for field assistance.  Funding was 490 
provided by the St. Johns River Water Management District (Contract #27799). Permits for 491 
species collections were provided by the Florida Fish and Wildlife Conservation Commission 492 
(SAL-14-1567-SR).  Comments and suggestions from two anonymous reviewers greatly 493 
improved the manuscript.  This is contribution #XXXX from the Smithsonian’s MarineGEO 494 
network and contribution #XXXX from the Smithsonian Marine Station.      495 
 496 
References 497 
Able KW, Grothues TM, Kemp IM (2013) Fine-scale distribution of pelagic fishes relative to a 498 
large urban pier. Mar Ecol Prog Ser 476:185-198. 499 
 500 
Airoldi L, Turon X, Perkol-Finkel S, Rius M (2015) Corridors for aliens but not for natives: 501 
effects of marine urban sprawl at a regional scale. Diversity Distrib 21:755-768. 502 
 503 
Alongi DM (2002) Present state and future of the world’s mangrove forests. Environ Conserv 504 
29:331-349. 505 
 23 
 
 506 
Attrill MJ (2002) A testable linear model for diversity trends in estuaries. J Anim Ecol 71: 262-507 
269. 508 
 509 
Bacchiocchi F, Airoldi L (2003) Distribution and dynamics of epibiota on hard structures for 510 
coastal protection. Estuar Coast Shelf Sci 56:1157-1166. 511 
 512 
Barbier EB, Hacker SD, Kennedy C, Koch EW, Stier AC, Silliman BR (2011) The value of 513 
estuarine and coastal ecosystem services. Ecol Monogr 81:169-193. 514 
 515 
Bellwood DR, Hughes TP, Folke C, Nystrӧm M (2004) Confronting the coral reef crisis. Nature 516 
429:827-833. 517 
 518 
Bingham BL, Young CM (1995) Stochastic events and dynamics of a mangrove root epifaunal 519 
community. Mar Ecol 16:145-163. 520 
 521 
Bishop MT, Byers JE, Marcek BJ, Gribben PE (2012) Density-dependent facilitation cascades 522 
determine epifaunal community structure in temperate Australian mangroves. Ecology 93:1388-523 
1401. 524 
 525 
Blum JC, Chang AL, Liljesthrӧm M, Schenk ME, Steinberg MK, Ruiz GW (2007) The non-526 
native solitary ascidian Ciona intestinalis (L.) depresses species richness. J Exp Mar Ecol Biol 527 
342:5-14.  528 
 24 
 
 529 
Bricker S, Longstaff B, Dennison W, Jones A, Boicourt K, Wicks C, Woerner J (2007) Effects of 530 
nutrient enrichment in the nation’s estuaries: A decade of change. NOAA Coastal Ocean 531 
Program Decision Analysis Series No. 26. National Centers for Coastal Ocean Science, Silver 532 
Spring, MD. 328 pp. 533 
 534 
Bulleri F, Airoldi L (2005) Artificial marine structures facilitate the spread of a non-indigenous 535 
green alga, Codium fragile spp. tomentosoides, in the north Adriatic Sea. J Appl Ecol 42:1063-536 
1072. 537 
 538 
Bulleri F, Chapman MG (2010) The introduction of coastal infrastructure as a driver of change in 539 
marine environments. J Appl Ecol 47:26-35. 540 
 541 
Cannicci S, Burrows D, Fratini S, Smith III TJ, Offenberg J, Dahdouh-Guebas F (2008) Faunal 542 
impact on vegetation structure and ecosystem function in mangrove forests: A review. Aquat Bot 543 
89:186-200. 544 
 545 
Cavanaugh KC, Kellner JR, Forde AJ, Gruner DS, Parker JD, Rodriguez W, Feller IC (2014) 546 
Poleward expansion of mangroves is a threshold response to decreased frequency of extreme 547 
cold events. Proc Natl Acad Sci 111:723-727. 548 
 549 
Chapman MG (2003) Paucity of mobile species on constructed seawalls: effects of urbanization 550 
on biodiversity. Mar Ecol Prog Ser 264:21-29. 551 
 25 
 
 552 
Connell SD (2001) Urban structures as marine habitats: an experimental comparison of the 553 
composition and abundance of subtidal epibiota among pilings, pontoons and rocky reefs. Mar 554 
Environ Res 52:115-125. 555 
 556 
Connell SD, Glasby TM (1999) Do urban structures influence local abundance and diversity of 557 
subtidal epibiota? A case study from Sydney Harbour, Australia. Mar Environ Res 47:373-387. 558 
 559 
Crona BI, Holmgren S, Rönnbäck (2006) Re-establishment of epibenthic communities in 560 
reforested mangroves of Gazi Bay, Kenya. Wetlands Ecol Mange 14:527-538. 561 
 562 
Dafforn EA, Glasby TM, Airoldi L, Rivero NK, Mayer-Pinto M, Johnston EL (2015) Marine 563 
urbanization: an ecological framework for designing multifunctional artificial structures. Front 564 
Ecol Environ 13:82-90. 565 
 566 
Elliott T, Persad G, Webber M (2012) Variation in the colonization of artificial substrates by 567 
mangrove root fouling species of the Port Royal mangrove lagoons in the eutrophic Kingston 568 
Harbour, Jamaica. J Water Resource Prot 4:377-387. 569 
 570 
Ellison AM, Farnsworth EJ (1990) The ecology of Belizean mangrove root-fouling communities. 571 
I. Epibenthic fauna are barriers to isopod attack of red mangrove roots. J Exp Mar Biol Ecol 572 
142:91-104. 573 
 574 
 26 
 
Ellison AM, Farnsworth EJ, Twilley RR (1996) Facultative mutualism between red mangroves 575 
and root-fouling sponges in Belizean mangal. Ecology 77:2431-2444. 576 
 577 
Farnsworth EJ, Ellison AM (1996) Scale-dependent spatial and temporal variability in 578 
biogeography of mangrove root epibiont communities. Ecol Monogr 66:45-66. 579 
 580 
Faunce CH, Serafy JE (2006) Mangroves as fish habitat: 50 years of field studies. Mar Ecol Prog 581 
Ser 318:1-18. 582 
 583 
Fletcher SW, Fletcher WW (1995) Factors affecting changes in seagrass distribution and 584 
diversity patterns in the Indian River Lagoon complex between 1940 and 1992. Bull Mar Sci 585 
57:49-58. 586 
 587 
Fofonoff PW, Ruiz GW, Steves B, Simkanin C, Carlton JT (2017) National Exotic Marine and 588 
Estuarine Species Information System. https://invasions.si.edu/nemesis/ (accessed 04 Dec 2017) 589 
 590 
Gittenberger A, van der Stelt RC (2011) Artificial structures in harbors and their associated 591 
ascidian fauna. Aquat Invasions 6:413-420.  592 
 593 
Gittman RK, Fodrie FJ, Popowich AM, Keller DA, Bruno JF, Currin CA, Peterson CH, Piehler 594 
MF (2015) Engineering away our natural defenses: an analysis of shoreline hardening in the US. 595 
Front Ecol Environ 13:301-307. 596 
 597 
 27 
 
Gittman RK, Scyphers SB, Smith CS, Neylan IP, Grabowski JH (2016) Ecological consequences 598 
of shoreline hardening: a meta-analysis. Bioscience 66: 763-773.  599 
 600 
Glasby T, Connell S (2001) Orientation and position of substrata have large effects on epibenthic 601 
assemblages. Mar Ecol Prog Ser 214:127-135. 602 
 603 
Glasby TM, Connell SD, Holloway MG, Hewitt CL (2007) Nonindigenous biota on artificial 604 
structures: could habitat creation facilitate biological invasions? Mar Biol 151:887-895. 605 
 606 
Janiak DS, Osman RW, Whitlatch RB (2013) The role of species richness and spatial resources 607 
in the invasion success of the colonial ascidian Didemnum vexillum Knott, 2002 in eastern Long 608 
Island Sound. J Exp Mar Biol Ecol 443:12-20. 609 
 610 
Kathiresan K, Bingham BL (2001) Biology of mangroves and mangrove ecosystems. Adv Mar 611 
Biol 40:81-251 612 
 613 
Kohler KE, Gill SM (2006) Coral point count with Excel extensions (CPCe): a visual basic 614 
program for the determination of coral and substrate coverage using random point count 615 
methodology. Comput Geosci 32: 1259-1269.  616 
 617 
Kornis MS, Breitburg D, Balouskus R, Bilkovic DM, Davias LA, Giordano S, Heggie K, Hines 618 
AH, Jacobs JM, Jordan TE, King RS, Patrick CJ, Seitz RD, Soulen H, Targett TE, Weller DE, 619 
 28 
 
Whigham DF, Uphoff Jr, J (2017) Linking the abundance of estuarine fish and crustaceans in 620 
nearshore waters to shoreline hardening and land cover. Estuar Coasts 40:1464-1486. 621 
 622 
Lambert G, Lambert C (2003) Persistence and differential distribution of nonindigenous 623 
ascidians in harbors of the Southern California Bight. Mar Ecol Prog Ser 259:145-161.  624 
 625 
Lapointe BE, Herren LW, Debortoli DD, Vogel MA (2015) Evidence of sewage-driven 626 
eutrophication and harmful algal blooms in Florida’s Indian River Lagoon. Harmful Algae 627 
43:82-102. 628 
 629 
Li SW, Chan BKK (2008) Adaptations to barnacle fouling in the mangroves Kandelia obovata 630 
and Aegiceras corniculatum. Mar Biol 155:263-271. 631 
 632 
Mook, D (1980) Seasonal variation in species composition of recently settled fouling 633 
communities along an environmental gradient in the Indian River Lagoon, Florida. Harbor 634 
Branch Foundation Contribution 149: 573-581. 635 
 636 
Moschella PS, Abbaita M, Aberg P, Airoldi L, Anderson JM, Bacchiocchi F, Bulleri F, Dinesen 637 
GE, Frost M, Gacia E, Granhag L, Jonsson PR, Satta MP, Sundelof A, Thompson RC, Hawkins 638 
SJ (2005) Low-crested coastal defense structures as artificial habitats for marine life: Using 639 
ecological criteria in design. Coast Eng 52:1053-1071. 640 
 641 
 29 
 
Nagelkerken I, Blaber SJM, Bouillon S, Green P, Haywood M, Kirton LG, Meynecke JO, 642 
Pawlik J, Penrose HM, Sasekumar A, Somerfield PJ (2008) The habitat function of mangroves 643 
for terrestrial and marine fauna: a review. Aquat Bot 89:155-185. 644 
 645 
Nixon SW (1995) Coastal marine eutrophication: A definition, social causes, and future 646 
concerns. Ophelia 41:199-219. 647 
 648 
O’Dea A, Okamura B (1999) Influence of seasonal variation in temperature, salinity and food 649 
availability on module size and colony growth of the estuarine bryozoan Conopeum seurati. Mar 650 
Biol 135:581-588. 651 
 652 
Oricchio FT, Pastro G, Vieira EA, Flores AAV, Gibran FZ, Dias GM (2016) Distinct community 653 
dynamics at two artificial habitats in a recreational marina. Mar Environ Res 122:85-92. 654 
 655 
Orth RJ, Carruthers TJB, Dennison WC, Durte CM, Fourqurean JW, Heck Jr. KL, Hugh AR, 656 
Kendrick GA, Kenworthy WJ, Olyarnik O, Short FT, Waycott M, Willimas SL (2006) A global 657 
crisis for seagrass ecosystems. Bioscience 56:987-996.    658 
 659 
Pagad S, Hayes K, Katsanevakis S, Costello MJ (2017) World Register of Introduced Marine 660 
Species. (www.marinespecies.org/introduced) accessed 04 Dec 2017 661 
 662 
Perry DM (1988) Effects of associated fauna on growth and productivity in the red mangrove. 663 
Ecology 69:1064-1075. 664 
 30 
 
 665 
Proffitt CE (2017) Phytoplankton biomass in a subtropical estuary: drivers, blooms, and 666 
ecological functions assessed over space and time using structural equation modeling. Mar Ecol 667 
Prog Ser 569:55-75.  668 
 669 
Rodemann JR, Brandl SJ (2017) Consumption pressure in coastal marine environments 670 
decreases with latitude and in artificial vs. natural habitats. Mar Ecol Prog Ser 574:167-179. 671 
 672 
Rogers TL, Byrnes JE, Stachowicz JJ (2016) Native predators limit invasion of benthic 673 
invertebrate communities in Bodega Harbor, California, USA. Mar Ecol Prog Ser 545:161-173. 674 
 675 
Ruiz GM, Freestone AL, Fofonoff PW, Simkanin C (2009) Habitat distribution and 676 
heterogeneity in marine invasion dynamics: the importance of hard substrate and artificial 677 
structure. In: Wahl M, (eds) Marine Hard Bottom Communities. Ecological Studies 206. 678 
Springer-Verlag Berlin Heidelberg, p 321- 332.  679 
 680 
Salmo III SG, Tibbetts I, Duke NC (2017) Colonization and shift of mollusc assemblages as a 681 
restoration indicator in planted mangroves in the Philippines. Biodivers Conserv 26:865-881. 682 
 683 
Simberloff D (1997) The biology of invasions. In: Simberloff D, Schmitz DC, Brown TC (eds) 684 
Strangers in paradise: impact and management of nonindigenous species in Florida. Island Press, 685 
Washington, pp 3-19. 686 
 687 
 31 
 
Sime P (2005) St. Lucie Estuary and Indian River Lagoon conceptual ecological model. 688 
Wetlands 25:898-907. 689 
 690 
Simkanin C, Davidson IC, Dower JF, Jamieson G, Therriault TW (2012) Anthropogenic 691 
structures and the infiltration of natural benthos by invasive ascidians. Mar Ecol Evol Persp 692 
33:499-511. 693 
 694 
Smith SDA, Rule MJ (2002) Artificial substrata in a shallow sublittoral habitat: do they 695 
adequately represent natural habitats of the local species pool? J Exp Mar Biol Ecol 277:25-41. 696 
 697 
Swain HM, Breininger DR, Busby DS, Clark KB, Cook SB, Day RA, De Freese DE, Gilmore 698 
RG, Hart AW, Hinkle CR, McArdle DA, Mikkelsen PM, Nelson WG, Zahorcak AJ (1995) 699 
Indian River Lagoon conference – introduction. Bull Mar Sci 57:1-7. 700 
 701 
Torre MP, Targett TE (2016) Nekton assemblages along riprap-altered shorelines in Delaware 702 
Bay, USA: comparisons with adjacent beach. Mar Ecol Prog Ser 548:209-218. 703 
 704 
Tyrrell MC, Byers JE (2007) Do artificial substrates favor nonindigenous fouling species over 705 
native species? J Exp Mar Biol Ecol 342:54-60. 706 
 707 
Valiela I, Bown JL, York JK (2001) Mangrove forests: one of the world’s threatened major 708 
tropical environments. Bioscience 51:807-815. 709 
 710 
 32 
 
Woodward-Clyde Consultants Staff (1994) Physical features of the Indian River Lagoon. 711 
Tallahassee, Florida: Woodward-Clyde Consultants, Indian River Lagoon National Estuary 712 
Program: Project 92F274C 713 
 714 
WoRMs Editorial Board 2017 World Register of Marine Species. (www.marinespecies.org) 715 
accessed 04 Dec 2017 716 
 717 
Zu Ermgassen PSE, Spalding MD, Blake B, Coen LD, Dumbauld B, Geiger S, Grabowski JH, 718 
Grizzle R, Luckenbach M, McGraw K, Rodney W, Ruesink JL, Powers SP, Brumbaugh R 719 
(2012) Historical ecology with real numbers: past and present extent and biomass of an imperiled 720 
estuarine habitat. Proc R Soc B 279:1-8.  721 
 722 
 723 
 724 
 725 
 726 
 727 
 728 
 729 
 730 
 731 
 732 
 733 
 33 
 
Figure 1. Map of the study region within the Indian River Lagoon.  Letters for each site indicate 734 
the corresponding water body (M = Mosquito Lagoon, IRL = Indian River, and B = Banana 735 
River).  Symbols designate habitat type, circles indicate artificial habitat sites and triangles 736 
indicate mangrove sites.  Environmental data were collected in close proximity to M_06, 737 
IRL_02, and B_02.   738 
 739 
 740 
 741 
 742 
 743 
 744 
 745 
 746 
 747 
 748 
 749 
 750 
 751 
 752 
 753 
 754 
 34 
 
Figure 2. Environmental measurements for temperature (a), salinity (b), and chlorophyll levels 755 
(c) taken daily and then averaged by week and averaged from 3 stations within the range of the 756 
study region.  Error bars indicate ±1 S.E. 757 
 758 
a)  759 
Average Weekly Temperature
09
/1
4
11
/1
4
01
/1
5
03
/1
5
05
/1
5
07
/1
5
09
/1
5
11
/1
5
01
/1
6
03
/1
6
05
/1
6
07
/1
6
T
e
m
p
e
ra
tu
r e
 (
C
)
5
10
15
20
25
30
35
 760 
b) 761 
Average Weekly Salinity
09
/1
4
11
/1
4
01
/1
5
03
/1
5
05
/1
5
07
/1
5
09
/1
5
11
/1
5
01
/1
6
03
/1
6
05
/1
6
07
/1
6
S
a
li
n
it
y
10
15
20
25
30
35
40
 762 
c) 763 
Average Weekly Chlorophyll Levels
09
/1
4
11
/1
4
01
/1
5
03
/1
5
05
/1
5
07
/1
5
09
/1
5
11
/1
5
01
/1
6
03
/1
6
05
/1
6
07
/1
6
R
e
la
t i
v
e
 C
h
l o
ro
p
h
y
ll
 (
u
g
/L
)
0
100
200
300
400
 764 
 35 
 
Figure 3.  Species richness (a) and Shannon-Wiener diversity (b) for established communities.  765 
Asterisks indicate significance from SNK pairwise comparisons tests for each season.  Error bars 766 
indicate ±1 S.E. 767 
  768 
a) 769 
Species Richness
09
/1
4
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
S
p
e
c
ie
s 
R
ic
h
n
e
s s
0
5
10
15
20
Artificial 
Mangrove 
*
*
*
*
*
* *
 770 
b) 771 
Shannon-Wiener Diversity
Date
09
/1
4
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
S
h
a
n
n
o
n
 D
i v
e
rs
i t
y
0.0
0.5
1.0
1.5
2.0
2.5
Artificial 
Mangrove 
 772 
 773 
 774 
 775 
 776 
 777 
 778 
 779 
 780 
 781 
 782 
 783 
 784 
 785 
 786 
 36 
 
Figure 4. nMDS plot for established communities.  Data were log (X+1)-transformed and 787 
averaged by site.  Enclosures indicate the percent similarity among clusters and labels indicate 788 
sampling events.    789 
 790 
 791 
 792 
 793 
 794 
 795 
 796 
 797 
 798 
 799 
 800 
 801 
 802 
 803 
 804 
 805 
 806 
 807 
 808 
 809 
 810 
 811 
 812 
 813 
 814 
 815 
 816 
 817 
 818 
 819 
nMDS - Community Analysis
Transform: Log(X+1)
Resemblance: S17 Bray-Curtis similarity
Similarity
50
60
Habitat
A
MG
10/14
10/14
01/15
01/15
04/15 04/1507/15
07/15
11/15
11/15
02/16
02/16
05/16
05/16
08/16
08/16
2D Stress: 0.14
 37 
 
Figure 5.  Percent cover of barnacles (a), Conopeum spp. (b), and open space (c), for seasonal 820 
samples in each habitat type.  These groups were selected as they contributed a high amount of 821 
dissimilarity to communities in the different habitat types.  Asterisks indicate significant 822 
differences from SNK pairwise comparisons tests for each season.  Error bars indicate ± 1 S.E.   823 
 824 
a)                                                                          b) 825 
Amphibalanus eburnus Percent Cover
Date
09
/1
4
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
P
e
rc
e
n
t 
C
o
v
e
r
0
20
40
60
80
100
Artificial 
Mangrove 
*
*
*
**
Conopeum spp. Percent Cover
Date
09
/1
4
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
P
e
rc
e
n
t 
C
o
v
e
r
0
5
10
15
20
Artificial 
Mangrove *
*
 826 
c) 827 
Open Space Percent Cover
Date
09
/1
4
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
P
e
rc
e
n
t 
C
o
v
e
r
0
20
40
60
80
100
Artificial 
Mangrove 
*
*
* * *
*
*
 828 
 829 
 830 
 831 
 832 
 833 
 834 
 835 
 836 
 837 
 838 
 839 
 840 
 841 
 38 
 
Figure 6. nMDS plot for established communities from site IRL_13mg (mangrove habitat) and 842 
IRL_13a (artificial habitat) for each sampling event.  Both sites were in close proximity (<10 m) 843 
and data were selected to show differences in community composition at a relatively small 844 
spatial scale.     845 
 846 
 847 
 848 
 849 
 850 
 851 
 852 
 853 
 854 
 855 
 856 
 857 
 858 
 859 
 860 
 861 
 862 
 863 
 864 
 865 
 866 
 867 
 868 
 869 
 870 
 871 
 872 
 873 
 874 
 875 
 876 
 877 
 878 
 879 
 880 
 881 
 882 
 883 
 884 
 885 
 886 
 887 
Transform: Square root
Resemblance: S17 Bray-Curtis similarity
Date
10/14
01/15
04/15
07/15
11/15
02/16
05/16
08/16
A
A
A
MG
MG
MG
A
A
A
MG
MG MG
A
A
A
MG
MG
MG
A A
A
MG
MG
MG
A
A
A
MG
MG
MG
A
A
A
MG
MG
MG
A
A
A
MG
MG
AA
A
MG
MG
MG
2D Stress: 0.19
Transform: Square root
Resemblance: S17 Bray-Curtis similarity
Date
10/14
01/15
04/15
07/15
11/15
02/16
05/16
08/16
A
A
A
MG
MG
MG
A
A
A
MG
MG MG
A
A
A
MG
MG
MG
A A
A
MG
MG
MG
A
A
A
MG
MG
MG
A
A
A
MG
MG
MG
A
A
A
MG
MG
AA
A
MG
MG
MG
2D Stress: 0.19
 39 
 
Figure 7. Species richness (a) and Shannon-Wiener diversity (b) for colonization panels.  888 
Asterisks indicate significant differences from SNK pairwise comparisons tests for each season.  889 
Error bars indicate ± 1 S.E.   890 
   891 
a)                                                                       b)  892 
Species Richness - Colonization
Date
01
/1
5
04
/1
5
07
/1
5
10
/1
5
01
/1
6
04
/1
6
07
/1
6
S
p
e
c
ie
s 
R
ic
h
n
e
ss
0
5
10
15
20
Artificial 
Mangrove 
*
*
                  
Shannon-Wiener Diversity - Colonization
Date
01
/1
5
04
/1
5
07
/1
5
10
/1
5
01
/1
6
04
/1
6
07
/1
6
S
h
a
n
n
o
n
 D
iv
e
rs
it
y
0.0
0.5
1.0
1.5
2.0
2.5
Artificial 
Mangrove 
 893 
 894 
 895 
 896 
 897 
 898 
 899 
 900 
 901 
 902 
 903 
 904 
 905 
 906 
 907 
 908 
 909 
 910 
 911 
 912 
 913 
 914 
 915 
 916 
 917 
 918 
 919 
 920 
 921 
 40 
 
Figure 8.  Percent cover of barnacles (a), Conopeum spp. (b), and open space (c) for colonization 922 
panels in each habitat type.  These groups were selected as they contributed a high amount of 923 
dissimilarity to communities in the different habitat types.  Asterisks indicate significant 924 
differences from SNK pairwise comparisons tests for each season.  Error bars indicate ± 1 S.E.   925 
 926 
 927 
a)                                                                         b)  928 
Amphibalanus eburneus Colonization
Date
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
P
e
rc
e
n
t 
C
o
v
e
r
0
20
40
60
80
100
Artificial
Mangrove
*
Conopeum spp. Colonization
Date
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
P
e
rc
e
n
t 
C
o
v
e
r
0
20
40
60
80
100
Artificial
Mangrove
*
 929 
 930 
c)  931 
Open Space Colonization
Date
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
P
e
rc
e
n
t 
C
o
v
e
r
0
20
40
60
80
100
Artificial
Mangrove
*
*
 932 
 933 
 934 
 935 
 936 
 937 
 938 
 939 
 940 
 941 
 942 
 943 
 41 
 
Figure 9.  Total count of non-native species identified from established communities in both 944 
habitat types throughout the study.  Asterisks indicate significance from SNK pairwise 945 
comparisons tests for each season.  Error bars indicate ±1 S.E. 946 
      947 
 948 
Non-native Species Richness
Date
09
/1
4
01
/1
5
05
/1
5
09
/1
5
01
/1
6
05
/1
6
09
/1
6
S
p
e
c
i e
s 
R
ic
h
n
e
ss
0
2
4
6
8
Artificial 
Mangrove 
*
*
* *
*
*
 949 
 950 
 951 
 952 
 953 
 954 
Table 1.  Number of species and origin of all species found throughout the study.     955 
  956 
Phylum Native Non-native Unknown 
Porifera 10 1 4 
Platyhelminthes 2 0 0 
Cnidaria 7 10 6 
Mollusca 19 2 1 
Sipuncula 1 0 0 
Annelida 11 7 7 
Arthropoda 13 9 1 
Bryozoa 11 12 2 
Entoprocta 0 0 1 
Echinodermata 0 0 1 
Chordata 23 7 7 
Total 97 48 30 
 957 
 958 
 959 
 960 
 961 
 42 
 
Table 2. SIMPER analysis for IRL_13mg and IRL_13a for all variables that accounted for 75% 962 
of dissimilarity between habitats.  All sampling events were pooled together for analysis.   963 
 964 
Average Diss = 39.48 Mangrove Artificial                                
Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
Amphibalanus eburneus 48.75 68.28 13.19 1.11 33.42 33.42 
Open Space 18.13 16.42 7.06 1.27 17.89 51.31 
Conopeum spp. 9.96 1.88 4.07 0.52 10.3 61.61 
Amphibalanus reticulatus 4.75 3.56 3.11 0.62 7.88 69.49 
Hydroids 4.92 4.96 2.37 1.09 6 75.49 
 965 
 966 
 967 
 968 
 969 
 970 
 971 
 972 
 973 
 974 
 975 
 976 
 977 
 978 
 979 
 980 
 981 
 982 
 983 
 984 
 985 
 986 
 987 
 988 
 989 
 990 
 991 
 992 
 993 
 994 
 995 
 996 
 997 
 998 
 999 
 43 
 
Supplement 1000 
 1001 
Supplement Figure 1. nMDS plots for each sampling event using Bray-Curtis similarity matrix.  1002 
Each point represents the average of 3 replicates taken at each site.  Triangles, labeled A, 1003 
represent artificial habitat and open circles, labeled MG, represent mangrove habitat.  Some 1004 
events have two graphs, the first to include all sites and the second, a subset with outliers 1005 
removed.     1006 
 1007 
a)                                                                              b) 1008 
  1009 
 1010 
c)                                                                    d)  1011 
  1012 
 1013 
e)                                                                    f) 1014 
  1015 
Quarterly Sites
10/2014
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_03
B_04
B_05
B_06
B_06alt
B_07
B_08
IRL_01
IRL_02
IRL_03
IRL_04
IRL_05
IRL_06IRL_07IRL_08
IRL_09
IRL_10
IRL_11
IRL_12
IRL_13d
IRL_13mg
M_01
M_02
M_04
M_05
M_06
2D Stress: 0.11
Quarterly Samples
01/2015
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_03
B_04
B_05
B_06
B_06alt
B_07
B_08
IRL_01
IRL_02
IRL_03
IRL_04
IRL_05
IRL_06
IRL_07
IRL_08
IRL_09IRL_10
IRL_11
IRL_12IRL_13dIRL_13mg
M_01
M_02M_04
M_05
M_06
2D Stress: 0.13
Quarterly Samples
04/2015
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_03
B_04
B_05
B_06
B_06alt
B_07
B_08
IRL_01
IRL_02
IRL_03
IRL_04
IRL_05
IRL_06IRL_07
IRL_08
IRL_09IRL_10I L_11
IRL_12
IRL_13d
IRL_13mg
M_01
M_02
M_04
M_05
M_06
2D Stress: 0.12
Quarterly Samples -  B_03 Included
07/2015
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_012 B_0345B_06B_06alt78IRL 1I 2I _0345789IRL_101I 2IRL 13dIRL_13mgMM 6
2D Stress: 0.01
Quarterly Samples - B_03 Removed
07/2015
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_04
B_05
B_06
B_06alt
B_07
B_08
IRL_01
IRL_02
IRL_03
IRL_04
IRL_05
IRL_06
IRL_07
IRL_08
IRL_09
IRL_10
IRL_11
IRL_12
IRL_13d
IRL_13mg
M_01M_02
M_04
M_05
M_06
2D Stress: 0.13
Quarterly Samples -  B_03 & IRL_03 Included
10/2015
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_012
B_03
45B_06altB_06
78IRL_01I 2
IRL_03
I 456789013dIRL_13mgM 1_02
2D Stress: 0.01
 44 
 
 1016 
 1017 
g)                                                                   h) 1018 
  1019 
 1020 
i) j) 1021 
  1022 
 1023 
k)                                                                        l) 1024 
  1025 
 1026 
 1027 
 1028 
 1029 
 1030 
 1031 
Quarterly Samples - B_03 & IRL_03 Removed
10/2015
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_04
B_05
B_06alt
B_06
B_07B_08
IRL_01
IRL_02
IRL_04
IRL_05
IRL_06IRL_07
IRL_08
IRL_09
IRL_10
IRL_11
IRL_12
IRL_13d
IRL_13mg
M_01
M_02
M_04
M_05
2D Stress: 0.1
Quarterly Samples - B_03 Inlcuded
01/2016
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_012 B_03456B_06alt78IRL_01I L_023IRL_0456I 8902I 13dIRL_13mg
M_01M 2_06
2D Stress: 0.01
Quarterly Samples - B_03 Removed
01/2016
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_04 B_05
B_06
B_06alt
B_07
B_08
IRL_01
IRL_02
IRL_03
IRL_04
IRL_05
IRL_06
IRL_08
IRL_09
IRL_10
IRL_11
IRL_12
IRL_13d
IRL_13mg
M_01
M_02
M_04
M_05
M_06
2D Stress: 0.12
Quarterly Samples
04/2016
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_03
B_04
B_05
B_06
B_06alt
B_07
B_08
IRL_01
IRL_02
IRL_03 IRL_04
IRL_05
IRL_06
IRL_07
IRL_08
IRL_09
IRL_10
IRL_11
IRL_12
IRL_13d
IRL_13mg
M_01
M_02
M_04
M_05
M_06
2D Stress: 0.15
Quarterly Samples - B_03 Included
07/2016
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_012 B_0356_06a78IRL_012I 34I L_0567890I L_12IRL_1 dIRL 13mgM 024
2D Stress: 0.01
Quarterly Samples - B_03 Removed
07/2016
Resemblance: S17 Bray-Curtis similarity
Habitat
A
MG
B_01
B_02
B_05
B_06
B_06a
B_07
B_08
IRL_01
IRL_02
IRL_03
IRL_04
IRL_05
IRL_06IRL_07
IRL_08
IRL_09IRL_10
IRL_11
IRL_12
IRL_13d IRL_13mg
M_01
M_02
M_04
M_05
M_06
2D Stress: 0.07
 45 
 
Supplement Table 1. Species list for all taxa identified at least to genus.  Letters indicate species 1032 
origin status (N = native, I = introduced, and U = unknown or cryptogenic) 1033 
  1034 
Porifera Clathrina sp. U 
  Cliona sp. U 
  Darwinella rosacea N 
  Halichondria cf magniconulosa N 
  Halichondria melanadocia N 
  Haliclona curacaoensis N 
  Haliclona implexiformis N 
  Haliclona manglaris N 
  Halisarca sp. U 
  Hymeniacidon heliophila N 
  Leucosolenia sp. U 
  Mycale microsigmatosa N 
  Suberites aurantiacus N 
  Sycon ciliatum I 
  Tedania ignis N 
Platyhelminthes Euplana gracilis N 
  Stylochus ellipticus N 
Cnidaria Actinia bermudensis N 
  Anemone sp. A  U 
  Anemone sp. B U 
  Anemonesp. C U 
  Bimeria vestita I 
  Bunodeopsis antilliensis N 
  Campanularia macroscypha N 
  Clytia linearis I 
  Clytia noliformis I 
  Diadumene franciscana I 
  Diadumene leucolena N 
  Dynamena cornicina I 
  Eudendrium annulatum I 
  Eudendrium carneum I 
  Exaiptasia pallida N 
  Halecium sp. U 
  Hebella scandens N 
  Obelia bidentata I 
  Obelia dichotoma  I 
  Obelia geniculata I 
  Obelia sp. U 
 46 
 
  Plumularia floridana N 
  Tubularia sp. U 
Mollusca Anomia simplex N 
  Arcuatula papyria N 
  Astyris lunata N 
  Bostrycapulus aculeatus N 
  Brachidontes exustus N 
  Cardotes floridanus N 
  Cerithium sp. U 
  Cinctura hunteria N 
  Costoanachis avara N 
  Crassostrea virginica N 
  Crepidula atrasolea N 
  Crepidula depressa N 
  Crepidula fornicata N 
  Diodora cayenensis N 
  Geukensia demissa N 
  Hiatella arctica N 
  Mytella charruana I 
  Mytilopsis sallei N 
  Nassarius vibex N 
  Perna viridis I 
  Petaloconchus varians N 
  Urosalpinx cinerea N 
Sipuncula Phascolion cryptum N 
Annelida Alitta succinea N 
  Bispira brunnea N 
  Branchiomma bairdi N 
  Cirratulus grandis N 
  Fabricia stellaris I 
  Ficopomatus enigmaticus I 
  Ficopomatus miamiensis N 
  Hydroides dianthus I 
  Hydroides dirampha N 
  Hydroides elegans I 
  Hydroides floridana N 
  Hydroides sanctaecrucis N 
  Hydroides sp.  U 
  Janua sp.  U 
  Lumbrinereis sp.  U 
 47 
 
  Marphysa sanguinea N 
  Pileolaria berkeleyana I 
  Polydora cornuta N 
  Sabella sp. U 
  Salmacina huxleyi N 
  Serpula sp. U 
  Spirobranchus minutus I 
  Terebella sp. A U 
  Terebella sp. B. U 
  Vermiliopsis infundibulum I 
Arthropoda Amphibalanus amphitrite I 
  Amphibalanus eburneus N 
  Amphibalanus improvisus N 
  Amphibalanus reticulatus I 
  Amphibalanus subalbidus N 
  Amphibalanus venustus N 
  Balanus trigonus I 
  Caprella penantis I 
  Chthamalus fragilis N 
  Cirolana sp. U 
  Erichsonella attenuata N 
  Ericthonius brasiliensis I 
  Ericthonius punctatus N 
  Eurypanapeus depressus N 
  Leptochelia rapax N 
  Libinia dubia N 
  Megabalanus coccopoma I 
  Menippe mercenaria N 
  Monocorophium insidiosum N 
  Pachygrapsus transversus I 
  Palaemonetes pugio N 
  Petrolisthes armatus I 
  Sphaeroma terebrans I 
Bryozoa Acanthodesia savartii N 
  Aeverrillia armata N 
  Alcyonidium polyoum N 
  Amathia distans I 
  Amathia gracilis N 
  Amathia verticillata I 
  Amathia vidovici I 
 48 
 
  Anguinella palmata N 
  Bugula neritina I 
  Bugula stolonifera N 
  Conopeum chesapeakensis N 
  Conopeum tenuissimum N 
  Cradoscrupocellaria atlantica N 
  Cradoscrupocellaria bertholleti I 
  Hippoporina indica I 
  Jellyella tuberculata I 
  Membranipora sp. U 
  Membranipora tenella N 
  Nolella stipata I 
  Parasmittina nitida N 
  Savignyella lafonti I 
  Schizoporella pungens I 
  Triticella sp. U 
  Victorella pavida I 
  Watersipora subtorquata I 
Entoprocta Barentsia sp.  U 
Echinodermata Ophiactis sp. U 
Chordata Aplidium antillense N 
  Aplidium sp. A U 
  Aplidium sp. B U 
  Ascidia curvata N 
  Botrylloides nigrum N 
  Botrylloides sp. U 
  Botryllus schlosseri N 
  Botryllus sp.  U 
  Clavelina oblonga N 
  Cystodytes dellechiajei N 
  Didemnum perlucidum I 
  Didemnum sp. U 
  Diplosoma glandulosum N 
  Diplosoma listerianum I 
  Diplosoma sp. U 
  Distaplia bermudensis N 
  Distaplia corolla N 
  Ecteinascidia turbinata N 
  Eudistoma clarum N 
  Eudistoma hepaticum N 
 49 
 
  Eudistoma obscuratum N 
  Eudistoma sp. N 
  Eusynstyela sp. U 
  Herdmania momus N 
  Lissoclinum fragile N 
  Microcosmus exasperatus N 
  Molgula manhattensis N 
  Molgula occidentalis N 
  Perophora viridis N 
  Phallusia nigra N 
  Polyandrocarpa zorritensis I 
  Polyclinum constellatum I 
  Pyura vittata N 
  Styela canopus I 
  Styela plicata I 
  Symplegma brakenhielmi I 
  Symplegma rubra N 
   
   
   
   
   
   
   
 
 
 
 
 
 
 
 
   
   
   
   
   
   
   
   
   
 1035 
 1036 
 50 
 
Supplement Table 2. SIMPER analysis for established communities. 1037 
 1038 
10/14 Ave Diss = 55.48 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 52.37 55.52 15.62 1.57 28.16 28.16 
  Open Space 21.17 13.5 9.38 1.01 16.91 45.06 
  Mytilopsis sallei 8.33 0 3.86 0.39 6.97 52.03 
  Amphipod tubes 8.4 3.06 3.67 0.72 6.61 58.64 
  Hydroids 2.73 7.61 3.22 1.07 5.81 64.45 
  Diplosoma listerianum 3.93 4.15 2.84 0.69 5.12 69.56 
  Branchiomma bairdi 2.17 2.91 1.51 0.63 2.71 72.28 
01/15 Ave Diss = 58.77 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 54.83 50.48 15.88 1.14 27.02 27.02 
  Anemone sp. C 0 12.06 6 0.57 10.21 37.23 
  Amphipod tubes 8.7 1.41 4.22 0.88 7.18 44.41 
  Open Space 8.7 4.81 3.86 1.32 6.57 50.98 
  Bugula neritina 1.93 4.31 2.81 0.5 4.78 55.76 
  Schizoporella pungens 0 5.39 2.65 0.33 4.52 60.28 
  Victorella pavida 3.7 0.89 2.18 0.59 3.71 63.99 
  Amphibalanus improvisus 3.47 0.19 1.79 0.28 3.05 67.04 
  
Halichondria cf 
magniconulosa 0.1 2.5 1.23 0.42 2.1 69.14 
  Molgula occidentalis 0.37 2.07 1.17 0.26 1.99 71.13 
04/15 Ave Diss = 52.59 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 62.17 48.96 13.3 1.31 25.28 25.28 
  Open Space 17.23 9.72 6.92 1.02 13.16 38.44 
  Amphipod tubes 5.57 2.19 2.64 0.68 5.02 43.47 
  Various Algae 2.27 4.33 2.51 0.66 4.77 48.23 
  Anemone sp. B 0 4.7 2.17 0.3 4.12 52.35 
  Schizoporella pungens 0 4.41 2.02 0.37 3.83 56.19 
  Suberites aurantiacus 0 4 1.85 0.25 3.51 59.7 
  Hydroids 3.33 1.67 1.76 0.82 3.34 63.04 
  Conopeum spp. 1.37 1.83 1.41 0.4 2.67 65.71 
  
Halichondria cf 
magniconulosa 0.47 2.65 1.32 0.61 2.5 68.22 
  Bugula neritina 2.13 0.65 1.24 0.43 2.35 70.57 
07/15 Ave Diss = 52.97 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 43.4 64.44 15.07 1.45 28.45 28.45 
  Open Space 19.53 11.74 7.03 1.17 13.28 41.73 
  Amphibalanus reticulatus 5.77 0.72 2.98 0.46 5.62 47.35 
  Hydroids 5.23 2.13 2.66 0.85 5.02 52.38 
  Amphipod tubes 2.97 2.39 1.84 0.9 3.48 55.86 
  Various Algae 1.53 2.3 1.5 0.76 2.82 58.68 
  Conopeum spp. 3 0.07 1.47 0.24 2.78 61.46 
  Mytilopsis sallei 2.9 0 1.37 0.29 2.59 64.05 
  Diadumene spp. 2.57 0.93 1.22 0.89 2.31 66.35 
  Diplosoma listerianum 1.17 1.69 1.22 0.48 2.31 68.66 
  
Halichondria cf 
magniconulosa 2.3 0.41 1.22 0.42 2.29 70.95 
10/15 Ave Diss = 42.82 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 62.83 67.9 12.85 1.25 30.01 30.01 
 51 
 
  Open Space 16.87 4.96 6.66 0.96 15.56 45.57 
  Amphipod tubes 7.4 8.35 4.9 1.03 11.44 57.01 
  Victorella pavida 4.93 0.12 2.16 0.39 5.04 62.05 
  Amphibalanus reticulatus 3.9 1.14 1.97 0.61 4.61 66.66 
  Diplosoma listerianum 0.07 3.63 1.76 0.3 4.11 70.77 
 
 
01/16 Ave Diss = 47.15 
 
 
 
Mangrove 
 
 
 
Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 51 58.57 11.72 1.4 24.86 24.86 
  Open Space 21.58 7.94 8.21 1.27 17.41 42.27 
  Amphibalanus reticulatus 10.88 6.86 6.06 1 12.86 55.13 
  Conopeum spp. 8.67 4.35 4.52 0.85 9.58 64.71 
  Amphipod tubes 1.27 6.75 3.47 0.57 7.37 72.08 
04/16 Ave Diss = 64.16 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 42.7 53.93 15.99 1.27 24.92 24.92 
  Open Space 25.1 0.11 11.56 1.13 18.02 42.93 
  Bugula neritina 13.03 5.43 6.5 0.68 10.13 53.06 
  Alcyonidium polyoum 10.23 7.06 5.65 0.98 8.81 61.87 
  Amphibalanus reticulatus 4.97 10.89 5.38 0.86 8.38 70.25 
07/16 Ave Diss = 48.74 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 52.33 72.92 13.84 1.18 28.4 28.4 
  Open Space 16.67 0.04 7.97 1.38 16.35 44.74 
  Conopeum spp. 12.5 3.45 6.18 0.71 12.69 57.43 
  Amphibalanus reticulatus 5.53 3.3 3.2 0.75 6.57 64 
  Hydroids 4.77 3.04 2.15 1.05 4.41 68.41 
  
Halichondria cf 
magniconulosa 2.77 2.03 1.98 0.55 4.06 72.47 
 1039 
 1040 
 1041 
 1042 
 1043 
 1044 
 1045 
 1046 
 1047 
 1048 
 1049 
 1050 
 1051 
 1052 
 1053 
 1054 
 1055 
 1056 
 1057 
 1058 
 52 
 
 1059 
Supplement Table 3. SIMPER analysis for colonization panels. 1060 
 1061 
01/15 Ave Diss = 73.15 Mangrove Artificial         
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 39.68 30.81 18.92 1.28 25.86 25.86 
  Open Space 26.92 13.96 14.75 1.02 20.16 46.02 
  Bugula neritina 2.32 12.06 6.59 0.57 9.01 55.03 
  Alcyonidium polyoum 7.92 9.65 5.8 0.82 7.93 62.97 
  Hydroids 2.08 7.37 3.84 0.54 5.25 68.22 
  Styela plicata 0 7.02 3.41 0.4 4.66 72.89 
04/15 Ave Diss = 71.56 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Open Space 34.26 17.52 15.57 1.19 21.76 21.76 
  Amphibalanus eburneus 22.39 27.62 14.4 1.08 20.12 41.89 
  Conopeum spp. 15.83 12.64 9.11 1.06 12.73 54.62 
  Amphipod tubes 8.22 9.46 6.26 0.74 8.75 63.37 
  Hippoporina indica 7.78 0.24 3.74 0.54 5.23 68.6 
  Hydroids 2.43 4.72 2.65 0.66 3.71 72.31 
07/15 Ave Diss = 55.34 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 60 47.41 13.75 1.2 24.85 24.85 
  Open Space 10.25 11.7 7.95 0.64 14.37 39.22 
  Ficopomatus spp. 9.25 0.37 4.42 0.81 7.98 47.21 
  Diplosoma listerianum 0.95 6.5 3.32 0.64 5.99 53.2 
  Amphibalanus reticulatus 2.6 4.8 3.21 0.52 5.8 59 
  Hydroids 3.7 3.41 2.6 0.91 4.71 63.71 
  Amphipod tubes 3.15 4.02 2.32 0.82 4.19 67.9 
  Branchiomma bairdi 1.05 4.17 1.92 0.73 3.47 71.36 
10/15 Ave Diss = 55.26 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 50 47.28 15.35 1.35 27.78 27.78 
  Open Space 27.25 7.86 12.18 1.02 22.05 49.83 
  Amphipod tubes 8.38 12.62 6.59 1.08 11.93 61.76 
  Schizoporella pungens 0 8.1 4.05 0.39 7.33 69.09 
  Amphibalanus reticulatus 1.25 7.45 3.84 0.66 6.95 76.04 
01/16 Ave Diss = 70.20 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 41.88 22.14 16.97 1.41 24.17 24.17 
  Conopeum spp. 26.12 25.02 16.45 1.14 23.44 47.6 
  Open Space 13.4 8 6.7 1.1 9.54 57.14 
  Amphipod tubes 0.36 11.86 5.93 0.69 8.45 65.59 
  Bugula neritina 0.64 10.44 5.35 0.52 7.62 73.21 
04/16 Ave Diss = 70.11 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Conopeum spp. 46.88 13.02 19.91 1.37 28.41 28.41 
  Amphibalanus eburneus 25.88 30.88 14.63 1.29 20.87 49.27 
  Open Space 15.92 14.66 10.33 0.85 14.73 64.01 
  Amphipod tubes 2.92 15.68 7.43 0.58 10.6 74.61 
07/16 Ave Diss = 58.91 Mangrove Artificial                                
  Species Av.Abund Av.Abund Av.Diss Diss/SD Contrib% Cum.% 
  Amphibalanus eburneus 53.52 47.57 17.22 1.4 29.23 29.23 
  
Halichondria cf 
magniconulosa 12.37 4.24 6.89 0.73 11.7 40.94 
 53 
 
  Open Space 3.52 10.52 5.63 0.55 9.56 50.5 
  Schizoporella pungens 0 8.9 4.22 0.38 7.15 57.65 
  Hydroids 11.22 7.29 4.05 1.31 6.88 64.53 
  Conopeum spp. 5 4.69 3.93 0.65 6.67 71.2 
 1062 
 1063 
 1064 
 1065 
 1066 
 1067 
 1068 
 1069 
 1070 
 1071 
 1072 
 1073 
 1074 
 1075 
 1076 
 1077 
 1078 
 1079 
 1080 
 1081 
 1082 
 1083 
 1084 
 1085 
 1086 
 1087 
 1088 
 1089 
 1090 
 1091 
 1092 
 1093 
 1094 
 1095 
 1096 
 1097 
 1098 
 1099 
 1100 
 1101 
 1102 
 1103 
 54 
 
 1104 
 1105