Grant_Cait_FINAL

Grant 1
Habitat Loss and Resulting Decrease of Species Abundance in
Seagrass Meadows
Caitlin Grant
University of Maine, Orono

Grant 2
Abstract: The goal of this study was to calculate the percent coverage of seagrass around the
island of South Caicos, and additionally, the abundance of invertebrates present at target sites.
Seagrass habitats supply food sources, protection, water quality maintenance and provide a
nutrient balance that also benefit neighboring ecosystems. The invertebrates chosen for
abundance analysis were: Eucidaris tribuloides, Lytechinus variegatus, Tripneustes ventricosus,
Diadema antillarum, Echinometra lucunter, Meoma ventricosa, and Holothuria mexicana. Coral
reef systems generate a very low amount of primary production and therefore need to receive
nutrients via an outisde system. Seagrass habitats provide nutrient advections that facilitate
cross-habitat exchanges of energy and materials. This trophic transfer is vital for the survival of
commercially important reef fish and island’s economy. An increase in tourism in the past
decade has been followed by plans to eradicate an expanse of the islands seagrass beds for a
‘cleaner’ appearance. We conducted this study with the aim of finding a correlation between
percent seagrass coverage and invertebrate abundance, to determine if the higher density of
seagrass holds a greater abundance of invertebrates. These invertebrates contribute to the
primary production that supports neighboring ecosystems. Seagrass data was collected at six
sites along multiple transects and quadrats were used to rank the percent coverage. Researcher
observation was used to record invertebrate abundance. Upon analysis of the data, it was
determined there were no significant correlations between the density of seagrass coverage and
invertebrate abundance for any of the species gathered at the six sites. Although literature has
recorded in previous studies that there is significance between these variables, we believe
methods of data collection, researcher error and less than ideal environmental conditions played
a major part in the results gained. Further research should be conducted with the current method
of data collection, with added guidelines and increased researcher awareness.
Introduction
The tropical islands of Turks and Caicos, located in the North Atlantic Ocean, are known
widely for their pristine white sand beaches and the diverse array of marine flora and fauna that
attract snorkelers and divers from all over the world. Turks and Caicos has fast become a hotspot
for a rapidly growing tourism market, veering away from its traditional fishing and agricultural
industry. The Turks and Caicos Islands (TCI) are comprised of eight islands that are located on
top of four sand banks. The islands of South Caicos, Providenciales and Grand Turk are the only
three inhabited islands, with a population of approximately 33,100 residents in total (2013 World
Bank). On South Caicos, a small limestone island, development has been markedly slower than
the greater populated island of Providenciales, but has begun creating plans to expand it’s value

Grant 3
to the increased tourism. Small developing islands such as this often depend on tourism as a
means to use its only asset in an international market (Zuidema 2011).
The marine environment surrounding South Caicos is home to a vast number of fish and
coral species, with an abundance of economically and ecologically important mangrove and
seagrass ecosystems. In terms of elemental dynamics, the latter play a crucial role in maintaining
the delicate balance of nutrients and trophic level transfers, vital to the health of the seagrass
communities and nearby ecosystems that rely on them either directly or indirectly. Additionally,
tourism and traditional coastal industries depend on the “unspoiled” marine environments and
products they produce, creating a link between healthy ecosystems and economy of the island
(Zuidema 2011).
Seagrasses have adapted to environments of high salinity and strong currents, preferring
soft sediments for root formation that is often the type of appealing sands tourists travel long
distances to enjoy, devoid of vegetation. Aside from their intrinsic value, seagrass meadows
provide water control and wave reduction as currents flow through the dense grasses. Their
branching root systems are comprised of horizontal rhizomes anchored in the sediment that trap
and reduce its mobility, physically affecting the substrate they grow upon (Hogarth 2007).
In addition to the physical benefits that seagrass beds provide to the ecosystem, the active
movement of high trophic level consumers among neighboring habitats and advection of
resources show strong evidence of influencing nutrient fluxes in cross-habitat exchanges of
energy and materials (Heck 2008), known as trophic transfer. Production and diversity of
commercially relevant fish are also strongly influenced from seagrass ecosystems (Duffy 2006).
Invertebrates inhabiting these ecosystems are able to use seagrass meadows as a means for
protection from predation among the dense shoots and enveloping canopy. When balanced and

Grant 4
healthy, seagrasses are a food source capable of supporting a great variety and abundance of
organisms. In addition to the species that permanently reside within the habitat, they also support
organisms that only utilize the resources periodically (Hogarth, 2007).
Biodiversity and Resilience
Biodiversity of species within an ecosystem is an important factor that affects the
stability and resilience of a system. Seagrasses possess functional groups that contribute to the
overall health of the ecosystem (Folke 2004). Marine dredging of seagrasses has become a
popular practice to eliminate the meadows that tend to be unfavorable to the growing number of
tourists visiting South Caicos each year. There are already plans in motion to uproot vegetation
in East Bay for a ‘cleaner’ appearance in the hopes of attracting more tourists to the area
(Zuidema 2011).
In their paper on regime shifts, resilience, and biodiversity in ecosystem management,
Folke et al. discuss the effect of human induced changes in dynamic ecosystems and the role of
functional groups in these systems. The importance of diversity in seagrass meadows can be
attributed to these functional groups that reflect the level of resilience within an ecosystem. The
presence of multiple “redundant” functional groups of invertebrates have the crucial stabilizing
ability to respond to different combinations of pressures which is crucial to stabilizing the
entirety of an ecosystem. If one functional group is eradicated, another redundant species present
in the system has the capability to maintain the same “jobs” that the former species generated.
Additionally, there have been studies carried out in multiple ecosystems monitoring the negative
results of habitat structure loss. Along with the resulting lower biomass and decline in species

Grant 5
richness present in the system, there is often colonizing species that prosper from the transition,
maintaining the new conditions and slowing the progress of recovery (Thrush 2002).
There have been multiple studies published proving significant differences in species
biodiversity in vegetated areas versus non-vegetated areas (Heck1997; Hemminga 2000) that
state “virtually without exception”, there were more species present in seagrass meadows than
adjacent habitats where seagrass was not present. Instead, to gauge if species richness and
abundance would be detrimentally affected if these meadows are dredged, this study investigates
whether the percent cover of seagrass correlates to biodiversity and resident invertebrate density.
Materials and Methods
Six sites were selected for collection of data on South Caicos in the Turks and Caicos
Islands, based on their proximity to the School for Field Studies Center and the availability of
target seagrass areas: Moxy Bush, Airport, Coastguard, Eastbay, Bell Sound West, and the
Dump. Data was collected for nine days between July 22nd, 2015 and August 4th, 2015 between 8
am and 4 pm. Air temperatures across the sites were approximately 31 degrees Celcius, with no
rainfall recorded during this period.
At each site, a 50 m transect was laid perpendicular to, the shore into a seagrass meadow,
a procedure previously carried out by English et al. in 1997. At every 5 m along the transect, two
researchers on opposing sides reported the seagrass cover within a 25 cm x 25 cm quadrat
divided into a 5 cm x 5 cm grid, recording the total coverage within a 50 cm x 50 cm square (4
quadrats) at each stop. To identify the invertebrates at each site, a belt transect method was used
to record the species that were readily visible within a 5 m area in total, with each researcher

Grant 6
recording 2.5 m on each side of their transect. In most sites, three transects were laid out and data
recorded, with only a few days where only two transects of data were able to be recorded.
To record the dominance of seagrass within each quadrat, they were divided into 25 5 x 5
cm sectors that were ranked from 0 to 5, depending on the biomass appearing in each square,
according to a method developed by Saito and Atobe (1970). The ranking criteria are described
in Table 1. displaying how coverage is established in each sector and used to show the frequency
and mid-point percent for each quadrat.
Figure 1. Example of estimates of dominance for each 25 sectors in a quadrat (Saito and Atobe
1977)
To calculate the percent coverage of seagrass in each transect, a method for the
estimation of cover adapted from Saito and Atobe (1970) was used. The formula is as follows:
Figure 2. Formula for calculation of percent coverage of seagrass (Saito and Atobe 1970)
where: Mi = mid point percentage of Class i:
f = frequency of each number with the same class of dominance

Grant 7
To determine if the coverage of seagrass is normally distributed we carried out a Shapiro-
Wilk test along each transect using JMP Pro 10 program. The Shapiro-Wilk test showed us that
our data was not normally distributed, so the median coverage and interquartile range (IQR), a
measure of statistical dispersion between the upper and lower quartiles, was established on each
transect at all sites. Next, we used a JMP scatterplot to create a graph of the percent seagrass
coverage versus species abundance, where each transect represented one data point. A regression
analysis was carried out to calculate the R2 value and a Spearman Rank test was applied to
estimate the p-value in JMP to assess any patterns between our two variables.
The invertebrates we decided to include for our abundance analysis, based on the
consistency they appeared within data, was comprised of multiple sea urchin species which we
combined into a single sea urchin category: Pencil Urchin (Eucidaris tribuloides), Variegated
Sea Urchin (Lytechinus variegatus), West Indian Sea Egg (Tripneustes ventricosus), Long Spine
Urchin (Diadema antillarum), Rock Boring Urchin (Echinometra lucunter), and Red Heart
Urchin (Meoma ventricosa). The second abundance analysis was carried out with the Donkey
Dung Sea Cucumber (Holothuria mexicana).
Results
The Shapiro-Wilk Goodness of Fit test was conducted to determine if the seagrass
percent coverage datasets from each site were normally distributed using the JMP Pro 10
program wgere any p-value above 0.05 would show the seagrass coverage being normally
distributed. Our p-values at each site were less than 0.001, confirming that the data was not
normally distributed. Given this information, we calculated the median coverage and IQR along
each transect at every date and location to continue further analysis (Appendix A). To calculate

Grant 8
the IQR, the 75% quartile was subtracted from the 25% quartile from the information provided
through the JMP analysis.
For each transect we created a scatterplot comparing the median seagrass coverage to the
abundance of sea urchins, and another comparing the median seagrass coverage to Donkey Dung
sea cucumbers. Using JMP Pro 10, we did a regression analysis for correlation and found the R2
value for sea urchins (0.004635, n = 48) and p-value through a Spearman Rank test (n = 48, p =
0.2972) and for Donkey Dung sea cucumbers (R2 = 0.001574, p-value = 0.2602, n = 48) (Figure
3 and 4). Since the p-value was greater than 0.05, there was no significant correlation between
seagrass coverage and invertebrate abundance, and our null hypothesis was rejected.
Figure 3. The effect of median percent seagrass coverage in relation to urchin abundance

Grant 9
Figure 4. The effect of median percent seagrass coverage in relation to Donkey Dung sea
cucumber adundance.
Discussion
We were not surprised to find that the percent seagrass coverage data sets for each
transect, location and date were not normally distributed. It would be rare to find a uniform
environment across seven separate sites, and within that, at difference transects at each site.
Differences in location, topography, water currents and quality all play a role in distributing
seagrass coverage and although all sites were on an 8½ square mile island, terrestrial and aquatic
factors differed.
The median percent seagrass coverage that was calculated for each transect at each site
consisted of a wide variation of complete coverage, to virtually no coverage. We expected to see
results such as this across the span of our target sites, but not necessarily the wide spread seen
within each individual site. Although certainly probable, we believe that a large majority of the
variation can be attributed to the ranking system used to identify seagrass coverage. A 0 – 5

Grant 10
ranking system was used (Figure 1.), and whilst this method has been successful in previous
seagrass coverage studies (Atobe and Saito 1997; English et al. 1970), researchers collecting data
for this project did not have a complete and uniform understanding of the method. The data
collected along one transect had the ability to be uniform, but time constraints and educational
purposes led to multiple researchers collecting data along each side of the transect. Again, for the
sake of keeping the project within a reasonable time period, often the third or fourth transect laid
would have two groups of researchers dividing the data collection along one transect. These
factors together leave room for a substantial amount of inaccuracies and the opportunity for
variation within data collected.
One of the locations visited for the dual collection of seagrass and mangrove data, Bell
Sound East, carried no seagrass cover to be measured, therefore it was left out of the analysis.
But unlike the sites that lacked vegetation, there were areas where seagrass beds had previously
been identified and measured by one group of researchers that later were assumed bare by other
research groups. Although these areas that lack vegetation are also naturally part of the sites
chosen to collect data, measuring the bare spots on the landscape was not conducive to the
direction of the study; to measure the percent cover of seagrass beds.
Our data showed no correlation between percent seagrass coverage and invertebrate
abundance, another result we understood could arise due to a few discrepancies while the data
was collected. Firstly, it was brought to the attention of the researchers on the project that the
transect number wasn’t recorded when data was collected, which made correlating the
invertebrate counts to the transect which it was measured at impossible. Secondly, the
researchers on the projects knowledge of Caribbean invertebrates were not uniform. This
resulted in some species being recorded while others were not, varying from day to day, transect,

Grant 11
location, and between the groups carrying out the observations. Another factor possibly
negatively affecting data collection was the presence of a vast abundance of Cassiopeia
Andromeda, rendering it nearly impossible to rank seagrass coverage where they were present.
Additionally, in areas where seagrass cover was particularly dense, researchers found it
difficult to observe invertebrates that were successfully using the habitat as cover, a pattern that
is observed on July 23rd, 2015 on transect 3 at Eastbay in the table from Appendix A. We know
from studies previously recorded that almost without exception, there is a greater abundance of
organisms present in vegetated areas compared to non-vegetated areas (Heck1997; Hemminga
2000). This leads us to believe that either the invertebrate counts were not conducted thoroughly,
or there are some seagrass sites present on South Caicos that follow a different habitat-
abundance pattern, which we find unlikely.
The data collected on July 24th, 2015 along transect 1 at Coastguard showed a completely
different pattern than the previous day at Eastbay, a very low percent seagrass coverage and a
abnormally high Sea Urchin count in relation to numbers recorded at all other sites (Appendix
A). A possible explanation for this observed pattern is the naturally occurring ‘flip-flop’ between
seagrass bed density and sea urchin populations commonly seen in tropical and subtropical areas
(Hemminga 2000). Sea urchins are herbivorous grazers who, when in high abundance, create an
extreme pressure on seagrasses by consuming large amounts of the vegetation. This results in
extensive bare patches of previously vegetated seagrass meadows, and high densities of sea
urchins, which in turn become easily predated in the newly open habitat. A recovery of
seagrasses follows after, with the eventual increase in sea urchin populations that continues the
cycle. Therefore, at the site where the high abundance of sea urchins were recorded, along with a
low percent seagrass coverage, it follows that we may be observing a transition between the

Grant 12
elimination of seagrass due to consumption, and the eventual predation of the grazing sea
urchins.
Further research should be conducted on the connection between percent seagrass
coverage and the abundance of invertebrates present within these ecosystems. The present
methods conducted in this study should continue to be utilized, but with additional guidelines
and increased knowledge gained before the research is carried out. To ensure that the sites being
studied are properly equipped with target habitats, prior surveying should be conducted. To make
certain that these areas are not being missed, and to prevent data overlaps, markers should be
placed evenly along the shoreline to indicate where transects should be laid. Weights placed
periodically along the transects would increase the stability and counteract strong currents if
present. Lastly, to aid with abundance counts, a clear list of the invertebrates to be recorded
should be memorized and present while data is being collected.
Acknowledgments
I would like to thank Andrea Murray, Kathy Lockhart, Heidi Hertler and my fellow Summer
2015 researchers for their help, support and patience during the duration of this study.

Grant 13
Bibiolography
Duffy JE. (2006) Biodiversity and the functioning of seagrass ecosystems Mar Ecol Prog Ser
311: 233-250
English et al. (1997). Seagrass Community Structure and Biomass, In: Survey Manual for
Tropical Marine Resources. Chapter 5: 246-258
Folke C, Carpenter S, Walker B, Scheffer M, Elmqvist T, Gunderson L, Holling CS. (2004)
Regime shifts, resilience, and biodiversity in ecosystem management. Annu Rev Ecol
Syst 35: 557-81
Heck KL Jr., Wetstone G. (1977) Habitat complexity and invertebrates species richness and
abundance in tropical seagrass meadows. Jour of Biog 4:135-142
Heck KL Jr., et al. (2008) Trophic transfers from seagrass meadows subsidize diverse marine
and terrestrial consumers. Ecosystems 11: 1198-1210
Hemminga, MA. Duarte, CM. (2000) Fauna associated with seagrass systems. In: Seagrass
ecology. Cambridge University, Camobridge UK
Saito, Y. and S. Atobe (1970). Phytosociological study of intertidal marine algae. I. Usujiri
Benten-Jima, Hokkaido. Bulletin of the Faculty of Fisheries, Hokkaido University, 21:
37-69.
Thrush SF, Dayton PK. (2002) Disturbance to marine benthic habitats by trawling and dredging:
implications for marine biodiversity. Annu Rev Ecol Syst 33:229-73
World Bank Group (2013) World Bank. Accessed 24 Jul.
http://data.worldbank.org/indicator/SP.POP.TOTL
Zuidema C, Plate R, Dikou A. (2011) To preserve or to develop? East Bay dredging project,
South Caicos, Turks and Caicos Islands. Jour of Coas Cons 15: 555-563

Grant 14
Appendix A
Table 1. Percent seagrass coverage median, interquartile range, and invertebrate abundance at
each location, transect number and date collected
Date Location Transect Median IQR Urchins Donkey Dung
22-Jul-15
Bell Sound
West 1 3.75 13.2484 0 1
22-Jul-15
Bell Sound
West 2 3.13 8.2484 0 1
22-Jul-15 Moxy 1 12.1282 8.4668 6 0
22-Jul-15 Moxy 2 35.4392 60.0279 13 1
22-Jul-15 Moxy 3 5.8768 34.5919 3 0
23-Jul-15 Airport 1 2.1896 6.8136 0 8
23-Jul-15 Airport 2 8.1904 16.7213 0 3
23-Jul-15 Airport 3 1.752 3.127 0 4
23-Jul-15 Eastbay 1 18.75 51.2794 2 0
23-Jul-15 Eastbay 2 50.626 36.4683 3 0
23-Jul-15 Eastbay 3 62.4378 41.8427 0 0
24-Jul-15 Coastguard 1 0.6256 1.251 46 0
24-Jul-15 Coastguard 2 0.3756 2.69 3 0
24-Jul-15 Coastguard 3 2.002 5.2534 5 0
24-Jul-15 Dump 1 0 3.9402 1 2
24-Jul-15 Dump 2 0 62.2503 2 0
24-Jul-15 Dump 3 0 75 0 0
24-Jul-15 Dump 4 0.2504 75 0 0
27-Jul-15 Moxy 1 7.8772 9.8121 4 0
27-Jul-15 Moxy 2 11.5028 9 0 0
27-Jul-15 Moxy 3 6.3778 28.7491 4 0
28-Jul-15 Airport 1 3.2526 7.1872 0 0
28-Jul-15 Eastbay 1 53.5004 47.1543 2 0
28-Jul-15 Eastbay 2 66.75 45.2489 2 1
28-Jul-15 Eastbay 3 30.0012 38.9042 0 0
30-Jul-15 Coastguard 1 0 1.4076 0 0
30-Jul-15 Coastguard 2 0 0.1252 12 0
30-Jul-15 Coastguard 3 1.2504 6.3764 0 0
30-Jul-15 Coastguard 4 3.0674 9.1297 0 0
30-Jul-15 Dump 1 0 75 0 1
30-Jul-15 Dump 2 0 0 0 5
30-Jul-15 Dump 3 0 16.7819 0 0
30-Jul-15 Dump 4 64.3756 75 2 9
31-Jul-15 Moxy 1 37.8754 52.9977 3 0
31-Jul-15 Moxy 2 12.1898 42.7774 5 1
3-Aug-15 Airport 1 3.1892 14.4389 0 0
3-Aug-15 Airport 2 5.9402 6.5633 0 0
3-Aug-15 Airport 3 1.7514 5.5326 0 6

Grant 15
3-Aug-15 Airport 4 2.3142 5.9075 0 7
3-Aug-15 Eastbay 1 61.8752 64.3106 0 0
3-Aug-15 Eastbay 2 71.625 37.5 0 0
3-Aug-15 Eastbay 3 46.751 70.8113 0 0
3-Aug-15 Eastbay 4 57.75 25.4058 0 0
4-Aug-15 Coastguard 1 5.066 16.846 1 0
4-Aug-15 Dump 1 0 14.3134 0 1
4-Aug-15 Dump 2 0.1252 1.2194 0 1
4-Aug-15 Dump 3 0.5634 1.7204 1 3
4-Aug-15 Dump 4 1.126 1.5626 0 0

Grant_Cait_FINAL

More Related Content

What's hot (20)

Viewers also liked (11)

Similar to Grant_Cait_FINAL (20)

Grant_Cait_FINAL