assessing the quality of alternative habitats used by …
TRANSCRIPT
1
ASSESSING THE QUALITY OF ALTERNATIVE HABITATS USED BY THE GRASS SHRIMP PALAEMONETES SPP.
By
MEREDITH MONTGOMERY
A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
2011
2
© 2011 Meredith Montgomery
3
To my family, human and canine
4
ACKNOWLEDGMENTS
I would like to thank Dr. Frazer and Dr. Jacoby for their mentorship and support of
this project. A special thank you is extended to Dr. Jacoby who fit me into his very busy
schedule and managed to be entertained patiently by my weekly, inevitable mistakes.
This project would have never been completed without him. I would like to thank my
committee for their comments, which served to greatly improve the quality of this
manuscript. I would like to thank Lauren Sweet for her hard work, unfailing laughter, and
excellent storytelling while scooping shrimp in the murky algae beds of the river. I dearly
thank my parents who have contributed their time, love, and free dog-sitting. I would
finally like to thank Norm and Mark who kept an eye out for my safety over long field
days and taught me the history of the Chassahowitzka River.
5
TABLE OF CONTENTS page
ACKNOWLEDGMENTS .................................................................................................. 4
LIST OF TABLES ............................................................................................................ 6
LIST OF FIGURES .......................................................................................................... 8
ABSTRACT ..................................................................................................................... 9
CHAPTER
1 INTRODUCTARY REMARKS ................................................................................. 11
2 MATERIALS AND METHODS ................................................................................ 13
Study Site ............................................................................................................... 13 Submerged Aquatic Vegetation .............................................................................. 13
Study Species ......................................................................................................... 14 Relative Abundance ................................................................................................ 17 In situ Growth Experiments ..................................................................................... 18
Reproductive Output ............................................................................................... 20 Relative Survival Rates ........................................................................................... 21
Statistical Analyses ................................................................................................. 21
3 RESULTS ............................................................................................................... 24
Environmental Conditions ....................................................................................... 24 Relative Abundance ................................................................................................ 24 Relative Survival ..................................................................................................... 25 In situ Growth Experiments ..................................................................................... 25
Modal Progression Analysis .................................................................................... 27
Fecundity ................................................................................................................ 27 Egg Volume ............................................................................................................ 28 Brood Volume ......................................................................................................... 28
Size at Maturity ....................................................................................................... 28
4 DISCUSSION ......................................................................................................... 47
LIST OF REFERENCES ............................................................................................... 56
BIOGRAPHICAL SKETCH ............................................................................................ 61
6
LIST OF TABLES
Table page 3-1 Two-way ANOVA based on log10-transformed salinities (ppt) measured along
three transects in February, May and August 2009.. .......................................... 30
3-2 Two-way ANOVA based on log10-transformed temperatures (°C) measured along three transects in February, May and August 2009.. ................................ 30
3-3 One-way ANOVA based on log10-transformed grass shrimp m-3 captured from filamentous algae, P. pectinatus, V. americana and M. spicatum in July 2009.. ................................................................................................................. 30
3-4 Two-way ANOVA based on log10-transformed grass shrimp m-3 captured from P. pectinatus, V. americana and M. spicatum in July and August 2009.. ... 30
3-5 One-way ANOVA based on arcsin transformed proportional survival rates in filamentous algae, P. pectinatus and V. americana.. .......................................... 31
3-6 Analysis of covariance based on log10-transformed wet weights with log10-transformed total length as a covariate. .............................................................. 31
3-7 Two-way ANOVA based on intermolt periods associated with positive growth for juveniles and males from filamentous algae, P. pectinatus and V. americana. .......................................................................................................... 31
3-8 Two-way ANOVA based on log10-transformed positive instantaneous growth rates for juvenile and male shrimp from filamentous algae, P. pectinatus and V. americana.. .................................................................................................... 31
3-9 Two-way ANOVA based on arcsin transformed percentage growth rates for juvenile and male shrimp from filamentous algae, P. pectinatus and V. americana. .......................................................................................................... 32
3-10 Regression analysis of instantaneous growth of all shrimp by holding time. ...... 32
3-11 Regression analysis of instantaneous growth of juvenile shrimp by holding time. .................................................................................................................... 32
3-12 Regression analysis of instantaneous growth of adult male shrimp by holding time ..................................................................................................................... 32
3-13 Analysis of covariance based on log10-transformed numbers of eggs and embryos with total length as a covariate ............................................................. 32
3-14 Analysis of covariance based on log10-transformed egg volumes with total length as a covariate. ......................................................................................... 33
7
3-15 Analysis of covariance based on log10-transformed brood volumes with total length as a covariate.. ........................................................................................ 33
3-16 Two-way ANOVA based on total lengths of reproductive females from filamentous algae, P. pectinatus, and V. americana in June and July 2009. ...... 33
3-17 Two-way ANOVA based on total lengths of reproductive females from P. pectinatus, and V. americana in June, July, and August 2009 ........................... 34
8
LIST OF FIGURES
Figure page 2-1 Location of the Chassahowitzka River along the west coast of peninsular
Florida. Diagram from Frazer et al. 2006. ........................................................... 23
3-1 Mean number of grass shrimp m-3 ± 95% confidence limits (CL). ...................... 35
3-2 Back-transformed mean proportional survival rates ± 95% confidence limits (CL) for grass shrimp tethered in three habitats. ................................................ 36
3-3 Size frequency distributions of grass shrimp collected from filamentous algae, P. pectinatus and V. americana in June-August 2009. ............................ 37
3-4 Log10-transformed wet weight versus log10-transformed total length for all Palaemonetes species with and without isopods. ............................................... 38
3-5 Total length (mm) vs. telson length (mm) of Palaemonetes spp. ........................ 39
3-6 Mean intermolt period ± standard error (SE) by habitat and life history stage. ... 40
3-7 Total number of molting shrimp by life history stage and holding time (12 hour intervals, 4 d). ............................................................................................. 41
3-8 Size frequency distributions of juveniles used in modal progression analysis. ... 42
3-9 Relationships between log10-transformed number eggs and embryos and total length for females collected in filamentous algae, P. pectinatus and V. americana. .......................................................................................................... 43
3-10 Relationships between log10-transformed egg volumes and total length for females collected in filamentous algae, P. pectinatus and V. americana. .......... 44
3-11 Back-transformed mean total lengths (mm) ± 95% confidence limits (CL) for mature females. .................................................................................................. 45
3-12 Size frequency distribution for mature female grass shrimp females from filamentous algae, P. pectinatus and V. americana. ........................................... 46
9
Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science
ASSESSING THE QUALITY OF ALTERNATIVE HABITATS USED BY GRASS SHRIMP PALAEMONETES SPP.
By
Meredith Montgomery
August 2011
Chair: Thomas Frazer Major: Fisheries and Aquatic Sciences
The biomass of native aquatic plants continues to diminish in Florida’s freshwater
and estuarine environments due to a variety of stressors. The Chassahowitzka River
exemplifies this trend with remnant patches of native macrophytes interspersed among
mats of filamentous algae and patches of Myriophyllum spicatum, a non-native
macrophyte. The relative quality of these alternative habitats was assessed by
measuring Palaemonetes spp. growth rates, reproductive output, relative survival, and
relative abundance. High quality habitats should support more grass shrimp, higher
survival, enhanced growth, and greater reproductive output. Growth rates were
statistically equal for juveniles and adult males from all habitats in which they were
collected, i.e., filamentous algae, Potamogeton pectinatus, and Vallisneria americana.
Intermolt periods were significantly shorter for juveniles relative to adult males, but the
time between molts did not differ significantly among for shrimp from different habitats.
Reproductive output was greatest for grass shrimp in V. americana, with small females
producing more numerous, small eggs. Relative survival was lower for shrimp tethered
in filamentous algae than those tethered in patches of rooted macrophytes. The native
V. americana yielded the greatest number of grass shrimp, and evidence suggested it
10
provided the highest quality habitat because grass shrimp survived better and
reproduced more quickly and more prolifically. Continued loss of native macrophytes,
e.g.,V. americana and P. pectinatus, with a shift to algae or non-native plants will
negatively affect grass shrimp populations and potentially cause detrimental effects to
populations of fish and invertebrates that prey on these species.
11
CHAPTER 1 INTRODUCTARY REMARKS
The biomass of native aquatic plants continues to diminish in many of Florida’s
freshwater and estuarine environments due to physical disturbance, changing salinity
regimes, eutrophication increased algal biomass, expansion of introduced species and
other stressors (Frazer, Notestein & Pine III, 2006). Surveys have documented such
changes in the Chassahowitzka River (Frazer et al., 2006). Historically dominated by
Vallisneria americana, the river also contained other native macrophytes, e.g., Sagittaria
kurziana and Potamogeton pectinatus, but it now contains significant amounts of
filamentous algae, including Lyngbya sp., and Myriophyllum spicatum, an introduced
plant. An understanding of the relative quality of these alternative habitats would
improve evaluations of costs and benefits associated with removing introduced and
nuisance species or restoring native vegetation (Beck et al., 2001; Castellanos and
Rozas, 2001).
Efforts to evaluate habitat quality in aquatic systems have focused primarily on
comparisons of vegetated and unvegetated habitats using measures of density and
diversity, with higher values assumed to be correlated with higher quality habitats (Van
Horne, 1983; Castellanos and Rozas, 2001). Castellanos and Rozas (2001) pointed to
the need to compare less dissimilar habitats, and Van Horne (1983) contended that
density and diversity do not assess habitat quality adequately. Species diversity does
not always indicate habitats of highest quality, e.g., old growth forests, with relatively
low diversity, provide significant value to specialists (Van Horne, 1983). In addition,
assuming a positive correlation between species abundance and habitat quality can be
misleading when habitats are used seasonally or when habitats of marginal quality
12
become sinks for individuals displaced from higher quality sites resulting in unstable
groups of immigrants that survive and reproduce poorly (Van Horne ,1983; Chockley,
St. Mary & Osenberg, 2008). To avoid drawing misleading conclusions from diversity
and abundance data, Van Horne (1983) recommended evaluating habitat quality with a
combination of density, growth, survival and reproduction. High quality habitat would
provide enhanced nutritional resources and refuge from predation, thereby supporting
higher densities, survival rates, growth rates and fecundity, with resident populations
gaining stability and the capacity to survive poor years (Van Horne, 1983).
Palaemonetes spp. or grass shrimp (Decapoda Palaemonidae) should yield
valuable insights into the relative quality of alternative, vegetated habitats in the
Chassahowitzka River. These shrimp are abundant and distributed widely in the rivers
and estuaries along the Gulf and Atlantic coasts. In this study, proxy measures of
fitness were compared among grass shrimp collected from alternative, vegetated
habitats in the Chassahowitzka River during the summer of 2009. Measures chosen to
assess the quality of available habitats were growth, reproductive output (fecundity, egg
volume, and brood volume), size at maturity, relative survival, and relative abundance.
13
CHAPTER 2 MATERIALS AND METHODS
Study Site
This study was conducted in the spring-fed and tidally influenced Chassahowitzka
River, Florida (Figure 2-1). In this river, the submerged aquatic vegetation is changing.
Native macrophytes, which included V. americana, P. pectinatus and Najas
guadalupensis, are being replaced by non-native vascular aquatic plants (e.g., M.
spicatum) and filamentous algae (e.g., Lyngbya spp.) (Frazer et al., 2006).
Submerged Aquatic Vegetation
Submerged aquatic vegetation provides much of the habitat structure within the
Chassahowitzka River. Vallisneria americana is a rooted native aquatic macrophyte
which historically dominated the Chassahowtizka River. It has long, tape-like leaves that
grow throughout the depth of the water column, is somewhat tolerant of brackish water
conditions, and persists year-round. Potamogeton pectinatus is thinly and extensively
branched, with short and strap-like associated leaves that grow to the height of the
water column. It was often found in fresh to brackish, low-flow areas of the
Chassahowitzka and persists year-round. The morphology of filamentous algae (e.g.,
Lyngbya spp.) varies among seasons; it is highly-structured and extends throughout the
water column during late fall and winter before becoming a mat-like layer in late spring
and summer. It often disappears in late summer and early fall. Myriophyllum spicatum
has finely branched leaves in whorls that often reach and float on the water’s surface. In
the Chassahowitzka River, this species creates a layer of tangled stems and whorls at
the water’s surface, with few branches or leaves persisting below the water’s surface.
14
Study Species
Grass shrimp, as their name implies, typically associate with seagrass or aquatic
macrophytes (Anderson, 1985). Aquatic macrophytes with different heights, growth
forms, and densities create habitats with differing levels of physical complexity that
mediate predation on grass shrimp to varying degrees (Coen, Heck & Abele, 1981; Bell
et al., 1984; Orth, van Montfrans & Fishman,1999; Castellanos and Rozas, 2001). Thus,
the assemblage composition of submerged aquatic vegetation will affect the survival of
grass shrimp. In fact, loss or removal of aquatic vegetation (e.g., due to coastal
dredging) has been shown to reduce their abundance (Anderson, 1985; Rosas and
Odum, 1987). Furthermore, Palaemonetes spp. are euryhaline, eurythermal and
capable of using sites with low dissolved oxygen as temporary refuge from predation or
places to forage on detritus (Welsh, 1975; Anderson, 1985; Lowe and Provenzano,
1990; Rowe et al., 2002). Therefore, the effects of habitat quality should not be masked
by variable responses to the range of salinities, temperatures or oxygen concentrations
that characterize tidally influenced rivers and other downstream estuarine habitats.
Grass shrimp have been classified as rapid dispersers, but directional movement may
be sporadic (Darcy, 2003). Howard (1985) reported that 60-75% of Palaemonetes spp.
remained in the same habitat patch six hours after they were stained, which indicated
Palaemonetes spp. may remain in a habitat long enough for measures of fitness to
reflect habitat quality.
Though these species are of little interest to human consumers, they are
extremely valuable in coastal ecosystems (Anderson, 1985). Grass shrimp serve as
important prey for fishes, invertebrates, and other carnivores in these systems, including
juveniles of valued species (Welsh, 1975; Kunz, Ford & Pung, 2006; Frazer, Lauretta &
15
Pine III, 2011). In fact, grass shrimp play a key role in transferring energy from the base
of food webs to higher trophic levels (Welsh, 1975). Adult Palaemonetes spp. have
been classified as opportunistic, generalist feeders, and they have been documented to
be detritivores, omnivores and facultative grazers on epiphytes, (Welsh, 1975; Beck and
Cowell, 1976; Bell and Coull, 1978; Morgan, 1980; Anderson, 1985; McCall, 2007).
Juvenile grass shrimp have been described as obligate, planktonic carnivores and
detritivores (Anderson, 1985). Furthermore, Palaemonetes pugio has been documented
to accelerate the breakdown of detritus, making nutrients and carbon available to
multiple trophic levels as fecal pellets, which enhance production of seagrass (McCall,
2007), dissolved organic matter, and shrimp biomass (Welsh, 1975; Anderson, 1985).
Because of their important ecological role, naturally high abundances and year-round
availability, grass shrimp have been used as an indicator of habitat quality in studies
focused on occupancy and density (Williams, 1984; Kneib, 1987; Glancy et al., 2003).
Decapods, including Palaemonetes spp., grow incrementally through ecdysis,
which can simplify studies of their growth. Ecdysis results from decreased levels of
molting inhibiting hormone (MIH) and synthesis of ecdysteroids (Fingerman, 1997;
Chang, Chang & Mulder, 2001). This hormonal cascade causes grass shrimp to shed
their cuticles, enabling both growth and copulation. The relative irreversibility of the
cascade makes grass shrimp suitable for short-term molting experiments designed to
estimate intermolt periods and growth rates (e.g., the methods of Ross et al., 2000).
Growth of Palaemonetes spp. peaks during the summer when temperatures are
consistently high and resources are abundant, with growth rates estimated to reach
0.133-0.143 mm d-1 in adults (Anderson, 1985). Grass shrimp achieve maximum
16
lengths of 42-50 mm and live 6-13 months (Anderson, 1985). Parasites (like the bopyrid
isopods observed on the gills of Palaemonetes spp. in the Chassahowitzka River) have
not been documented to limit the growth of Gulf grass shrimp (Anderson, 1985).
Adult female Palaemonetes spp. brood eggs. From February to October along the
Central Gulf coast Florida, females of Palaemonetes spp. grow extra setae in their
abdominal brood pouches to hold eggs during this reproductive season (Anderson,
1985; Raviv, Parnes & Sagi, 2008). Broods of Gulf Coast populations develop in 12-22
d, which allows a mature female to bear up to eight broods between April and
September (Anderson, 1985; Bauer and Abdalla, 2000). Ecdysis, mating, and
oviposition are coupled tightly; female grass shrimp have been observed to molt soon
after vitellogenesis and mating; and females will not molt later in the brooding cycle as
embryos could be lost (Bauer and Abdalla, 2000; personal observation). Females
continue somatic growth after reaching maturity at 22-44 mm, although growth rates
decrease (Beck and Cowell, 1976; Anger and Morriera, 1998). In Florida, Beck and
Cowell (1976) found that female Palaemonetes spp. hatched in early spring were able
to reproduce in late summer at 20-24 mm and in late winter at 35-43 mm. Overall, the
tendency to brood multiple batches of eggs during the year makes grass shrimp suitable
for evaluating differences in reproductive output as a metric of habitat quality.
Three species of Palaemonetes inhabit the Chassahowitzka River: Palaemonetes
pugio, P. intermedius, and P. vulgaris. These species exhibit similar morphologies,
growth and reproduction (Broad, 1957; Thorpe, 1976; Anderson, 1985). The species
may partition available habitat and alleviate niche overlap through agonistic behavior or
microhabitat choice based on salinity, e.g., P. vulgaris has been documented to expend
17
more energy in low salinity habitats (Thorpe, 1976; Skadsheim, 1984; Anderson, 1985;
Rowe, 2002). Similar to other Gulf coast populations, abundances and reproductive
effort of grass shrimp in the Chassahowitzka River peak in summer and early fall (Beck
and Cowell, 1976; Castellanos and Rozas, 2001), with increased availability of brooding
females making summer a suitable time to evaluate habitat quality.
Relative Abundance
In June, July and August 2009, sampling to assess the relative abundance of
grass shrimp was conducted in discrete patches of V. americana, P. pectinatus, M.
spicatum and filamentous algae. Habitat patches of similar scale were selected
according to their dominant vegetation and heterogeneous patches were avoided. Patch
dimensions to the nearest meter [length and width (±SE)] were 16 (± 2) m and 13 (± 1)
m for V. americana; 16 (± 1) m and 6 (± 2) m for P. pectinatus; 29 (± 9) m and 5 (± 3) m
for filamentous algae; and 8 (± 1) m; and 4 m for M. spicatum. Patches were located in
a uniform reach of the river as delineated by temperatures and salinities measured at
stations along three transects in February, May and August 2009 as part of a
complementary project documenting water quality and the distribution of vegetation.
Relative abundance of grass shrimp was assessed via standardized, 0.14-m3
sweeps of a hand-held dip net (1 mm × 1 mm mesh) made at the approximate center of
each habitat patch to avoid edge effects. A minimum of 4 sweeps was made, and
sweeping ceased if 150 individuals were collected (the target number for in situ growth
experiments) or after 10 sweeps. Shrimp were counted and held in aerated coolers prior
to growth experiments.
18
In situ Growth Experiments
Upon completion of standardized sweeps, non-standardized sweeps were made if
additional individuals were needed for growth experiments or studies of reproduction.
Collections continued until approximately 150 individuals were captured and stored in
an aerated cooler or until disturbance of the patch, high tide or sunset precluded
successful sampling. Individual shrimp were transferred from the aerated cooler into
separate small glass jars (6-8 oz). Aside from two initial trials in June that did not
include ovigerous females, shrimp were selected without regard to size, maturity, or
presence of eggs. Mesh screens (1 mm × 1 mm) affixed with rubber bands restrained
the shrimp within the jars, while allowing exchange of water.
Jars containing shrimp were held in a cooler (< 1 h) until 11-50 jars with shrimp
were placed in a slotted, plastic bin (60 × 38 × 20 cm) and returned to the river. These
bins were used to organize held shrimp and did not restrict each jarred shrimp’s access
to river water. Bins containing shrimp from all habitat types were held in a small V.
americana patch (28.71467 N, -82.58437 W) for approximately 96 h. The V. americana
patch was chosen because it had relatively high dissolved oxygen concentrations, was
located in shallow water, and was protected from boat traffic.
Shrimp were checked for evidence of molting or mortality at approximately 12-h
intervals (06:30 and 18:30) for 96 h. Upon evidence of molting, each shrimp and its
exuvia were placed in a labeled, plastic centrifuge vial containing 90% ethanol to
prevent deterioration of soft-shelled recently molted shrimp, and frozen. Shrimp that
died or did not molt during the experiment were frozen for morphometric analyses.
Grass shrimp and their exuviae were processed to yield measures of growth and
morphometric relationships between telson length and total length and wet weight and
19
total length. Shrimp were lightly washed with water to remove excess debris and dirt,
subsequently blotted with paper towels to remove excess water, and weighed to the
nearest 0.1 mg. The total length of each shrimp, i.e. from the tip of its rostrum to the tip
of its fully extended tail, was measured to the nearest 0.5 mm with a ruler. Total length
was not recorded if a rostrum or tail was broken. Exuviae typically comprised multiple or
poorly connected pieces; therefore, growth could not be assessed by comparing the
total lengths of the post-molt shrimp and its exuvia. Instead, lengths of unbroken telsons
were measured to the nearest 0.01 mm using a dissecting microscope and ocular
micrometer, with growth to be assessed by comparing telson lengths from post-molt
shrimp and their exuviae and by using the morphometric relationship between total
lengths and telson lengths. Intermolt periods, instantaneous growth rates and
percentage growth rates were estimated according to the methods of Ross et al. 2000
(Equations 2-1, 2-2, and 2-3):
( ) ( )
[
] (2-1)
( ) ( )
( ) (2-2)
( ) ( )
( ) (2-3)
During processing, grass shrimp were classified to species in accordance with
Williams 1984. In addition, grass shrimp were classified as being ovigerous, post-
ovigerous, or non-ovigerous.. Ovigerous and post-ovigerous shrimp were considered
sexually mature females. The minimum total length of these females across all samples
was used as the criterion for classifying non-ovigerous shrimp as either adult males or
juveniles.
20
Grass shrimp also were examined for the presence of bopyrid isopods on their
gills. Isopod lengths and widths were measured to the nearest 0.01 mm using Vernier
calipers held over the carapaces of the shrimp. Subsequently, isopods were removed
with tweezers and weighed to the nearest 0.1 mg.
Reproductive Output
All ovigerous females were categorized as having an open abdominal brood pouch
or a closed brood pouch. Closed brood pouches were assumed to contain an entire
brood, whereas, individuals with split or open brood pouches were assumed to have lost
eggs. The closed brood pouches were opened with a scalpel, and the scalpel and water
were used to free eggs and embryos. Care was taken to avoid removing claws, legs or
swimmerettes.
All eggs and embryos were counted with the aid of a compound microscope.
Embryos were distinguished from eggs by the presence of eyespots or visible cell
proliferation, and they were counted but not measured. Intact eggs that were not yolked,
were not counted or measured. Broken eggs were assumed to be yolked, and they
were counted. The lengths and widths of intact, yolked, elliptical eggs were measured to
the nearest 0.01 mm using an ocular micrometer fitted to a compound microscope. Egg
volume (treated as the sum of the volumes of two cones, Equation 2-4) and brood
volume (Equation 2-5) were calculated as:
( ) { [ ( )
] } [
( )
] (2-4)
( ) ( ) (2-5)
21
Relative Survival Rates
Tethering experiments to estimate relative survival rates were conducted in
September, October and November 2009. In each of four trials, grass shrimp were
collected from a patch of V. americana using hand-held dip nets (mesh: 1 mm × 1 mm).
Tethers were created by using a needle to attach 30-40 cm of thread between the base
of a shrimp’s rostrum and one of its eye stalks. Shrimp were held in an aerated cooler
for approximately 30 min to confirm the stability of tethers and absence of immediate
mortality.
Shrimp were transported to characteristic and proximate patches of V. americana
(7 X 5 m; 28.71648 N, -82.57844 W), P. pectinatus (17 X 5 m; 28.7159 N, -82.57745
W), and filamentous algae (7 X 6 m; 28.71638 N, -82.57689 W). Twenty-five shrimp
were used in all trials except the September 30 trial in P. pectinatus where 7 shrimp
were used. Each tether was attached approximately 20-25 cm above base of a wood
dowel. Dowels, separated by ~30 cm, were forced into the sediment within the
appropriate patch of habitat. Once situated, each shrimp was placed at the base of its
dowel within the vegetation, which simulated their normal daytime location. The shrimp
and habitat were observed for 15 minutes to ensure equilibration, and then the sites
were left undisturbed for 1 h. After that time, tethered shrimp were classified as present
(whether alive or dead) and absent. Relative survival was calculated as the total number
tethered grass shrimp present at the end of the trial divided by the total number grass
shrimp used in the trial.
Statistical Analyses
As appropriate, the assumptions of normality and homoscedasticity of residuals
were tested with Anderson-Darling and Cochran’s tests, respectively. Data were
22
transformed where necessary, and the results of analyses based on non-normal or
heteroscedastic data were interpreted with due regard for potentially inflated Type I
errors.
Variations in temperature and salinity were analyzed with two-way analyses of
variance (ANOVAs). Transects and months were treated as fixed factors. Relative
abundances of grass shrimp were analyzed similarly. Months of sampling and habitats
were treated as fixed factors. Differential survival rates among habitats were assessed
with a one-way ANOVA. Habitat was considered a fixed factor. Analysis of covariance
(ANCOVA) was used to determine if the relationship between wet weights and total
lengths differed among species and between shrimp with and without isopod parasites.
Species and parasitism were treated as fixed factors.
Data from in situ growth experiments were analyzed in two ways. Regression was
used to relate telson lengths to total lengths. Intermolt periods, instantaneous growth,
rates and percentage growth rates (expressed as proportions) generated from in situ
growth experiments were analyzed with two-way ANOVAs. Life history stage and
habitat were treated as fixed factors in these analyses.
In addition to estimates from the in situ experiments, monthly changes in mean
total lengths of visibly distinct cohorts of shrimp were used to estimate growth rates.
Between June and August, growth was calculated as (Equation 2-6):
( ) ( ) ( )
( ) (2-6)
Variation in size at maturity for females and measures of reproductive output were
analyzed with ANCOVAs or ANOVA that treated months and habitats as fixed factors.
Variation in size at maturity was analyzed with a two-way ANOVA. Fecundity (the
23
number of fertilized eggs), egg volume and brood volume were analyzed with
ANCOVAs treating total length as a covariate.
Figure 2-1. Location of the Chassahowitzka River along the west coast of peninsular Florida. Diagram from Frazer et al. (2006).
24
CHAPTER 3 RESULTS
Environmental Conditions
Salinity and temperature did not vary significantly across the three transects
bracketing the habitat patches sampled in this study (Tables 3-1 and 3-2). Salinities and
temperatures did vary significantly among months, with means differing by a maximum
of only 2.4°C and 1.6 ppt (Tables 3-1 and 3-2). Given these results, any differences in
measures of abundance, relative survival, growth, size at maturity or reproductive
output were not attributed to differences in environmental conditions among patches of
habitat.
Relative Abundance
Standardized sweeps yielded reliable data for four habitats in July, V. americana,
P. pectinatus, M. spicatum and filamentous algae, and three habitats in July and
August, V. americana, P. pectinatus and M. spicatum. Therefore, a one-way ANOVA
was conducted using data from July, and a two-way ANOVA was conducted using data
from July and August (Tables 3-3 and 3-4). Log10-transformed densities (individuals m-3)
met the assumptions of homoscedasticity and normality. In July, densities varied
significantly among the four habitats, and in July and August, densities varied
significantly among combinations of month and habitat (Tables 3-3 and 3-4). Grass
shrimp were never collected in M. spicatum despite similar sampling effort (Figure 3-1).
In July, filamentous algae harboured198× fewer shrimp than V. americana and 3× fewer
than P. pectinatus (Figure 3-1). Vallisneria americana yielded 75 and 5× more shrimp
than P. pectinatus in July and August, respectively (Figure 3-1). In August, shrimp
25
densities in V. americana decreased four-fold and shrimp densities in P. pectinatus
increased four-fold (Figure 3-1).
Relative Survival
Significantly more grass shrimp survived when they were tethered in P. pectinatus
and V. americana as compared to filamentous algae (Table 3-5; Figure 3-2). The
proportion of shrimp surviving in the first two habitats was over 5× as great (Figure 3-2).
No difference in relative survival was observed in grass shrimp tethered in V. americana
and P. pectinatus.
In situ Growth Experiments
Across June, July and August, in situ growth experiments involved 1839 grass
shrimp from three species (1137 Palaemonetes pugio, 373 P. intermedius and 217 P.
vulgaris) and three habitats, V. americana (786 shrimp), P. pectinatus (723 shrimp) and
filamentous algae (330 shrimp). Some individuals of all life history stages molted, i.e.,
juveniles, males, and females, but only juveniles and males regularly exhibited positive
growth. Total lengths of shrimp from V. americana, P. pectinatus and filamentous algae
ranged from 11 mm to 50 mm, and the size frequency distribution contained multiple
modes (Figure 3-4). Smaller shrimp were more numerous in V. americana, with mean
total lengths (± standard deviations) of 28.13 mm (± 5.63 mm), 27.09 mm (± 5.84 mm),
and 20.70 mm (± 4.99 mm) for shrimp from filamentous algae, P. pectinatus and V.
americana, respectively.
The percentage of grass shrimp infected by isopods also varied among habitats,
with infection rates of 22.49%, 30.62% and 10.88% for shrimp from filamentous algae,
P. pectinatus and V. americana, respectively. Parasites were never observed on
reproductive females, but they were observed on 15.19% of juveniles and 32.23% of
26
males. In addition, 21.18% of P. pugio, 17.42% of P. intermedius and 20.27% of P.
vulgaris were parasitized.
In an effort to determine if growth needed to be analyzed separately for species
and parasitized shrimp, an ANCOVA was used to compare linear regressions based on
log10-transformed wet weights and log10-transformed total lengths of infected and
uninfected shrimp across the three species. Wet weights were related to total lengths,
and the presence of an isopod parasite altered this relationship by increasing the
intercept for wet weight by about 0.06 g (Table 3-6; Figure 3-4). The average wet weight
of removed isopods was 0.01 g. The lack of significance for the species and interaction
terms indicated P. pugio, P. intermedius and P. vulgaris all exhibited similar
morphometrics even if infected by an isopod. These results indicated that the three
species and shrimp with and without parasites could be pooled for analyses based on
total lengths.
Telson lengths reliably predicted total lengths (Figure 3-5). Therefore, growth was
expressed as change in total length, with the total lengths of pre-molt shrimp estimated
from the regression. Intermolt periods did not vary significantly among habitats or
combinations of habitats and months (Table 3-7). Mean intermolt periods were
significantly shorter for juveniles (1.7×), and this relationship held across all habitats
(Figure 3-6).
Only 32.19% of the 1839 shrimp used in the growth experiments increased in size,
with 52.52% showing no measurable change and 15.29% shrinking. In addition, females
seldom grew, so analyses were limited to data from juveniles and males. Positive
instantaneous growth rates did not vary significantly for juveniles or males from any of
27
the three habitats that yielded shrimp (Table 3-8). Overall, the mean positive
instantaneous growth rate (± standard error) for all juvenile and male shrimp was 0.44
(± 0.11) mm d-1. Similarly, instantaneous growth expressed as a percentage increase in
size did not vary significantly for juveniles and males from the three habitats (Table 3-9).
The overall mean (± standard error) instantaneous percentage growth rate was 2.34
( 0.73)% d-1.
Juvenile and adult male molting predominately occurred during overnight holding-
time intervals (12, 36, 60, 84 hr) (Figure 3-7), and decreased from 12 to 96 hr. A
regression of instantaneous growth versus holding time for all shrimp was not significant
(F1, 453 = 0.72, p = 0.397) (Table 3-10). Similar regressions were not significant for
juveniles (F1, 265 = 0.71, p = 0.401) (Table 3-11) or adult males (F1, 186 = 1.55, p = 0.215)
(Table 3-12).
Modal Progression Analysis
Only one cohort of shrimp could be clearly identified, with those shrimp captured in
V. americana (Figure 3-8). Therefore; juvenile growth rates from June to August 2009
were estimated with modal progression analysis of total lengths for this cohort of
shrimp. In addition, the cohort was distinguished most easily in June and August. The
analysis indicated that these juveniles grew 0.05 mm d-1.
Fecundity
Available data supported analysis of fecundity, i.e., numbers of eggs and embryos,
for females captured in filamentous algae, P. pectinatus, and V. americana during June
and July 2009. Analysis indicated that fecundity varied significantly with total length and
among habitats (Table 3-13; Figure 3-9). The lack of a significant interaction indicated
that fecundity increased similarly with total length for shrimp from the three habitats and
28
two months, with regressions for shrimp from different habitats indicating increases of
0.01-0.03 eggs per mm increase in total length (Figure 3-9). The regressions also
indicated that shrimp from V. americana carried more eggs and embryos (Figure 3-9),
with back-transformed means and 95% confidence limits of 99.1, 23.1 and 425.1 in
June and 87.8, 66.0 and 116.8 in July for V. americana; 48.0, 28.3 and 81.4 in June and
43.2, 13.9 and 134.6 in July for P. pectinatus; and 46.7, 32.5 and 67.1 in June and 43.3,
21.4 and 87.6 in July for filamentous algae.
Egg Volume
An ANCOVA indicated that the relationship between egg volumes and total length
varied according to the month of sampling and habitat sampled (Table 3-14; Figure 3-
10). Variation among habitats was of primary interest. Back-transformed least squares
mean egg volumes (i.e., mean egg volumes for females captured in each habitat with
total length and month held constant) were 1.66 mm3 for filamentous algae, 1.64 mm3
for P. pectinatus and 1.21 mm3 for V. americana.
Brood Volume
An ANCOVA indicated that the relationship between brood volumes and total
lengths varied according to the combination of month of sampling and habitat sampled
(Table 3-15). Variation among habitats, the measure of primary interest, was
approximately 1 mm3. Back-transformed least squares mean brood volumes were 9.90
mm3 for filamentous algae, 9.72 mm3 for P. pectinatus and 10.73 mm3 for V. americana.
Size at Maturity
Mature female grass shrimp were defined as females brooding eggs or embryos
and females with extended brood pouches indicating recent hatching of eggs. Available
data supported two analyses. One analysis focused on shrimp collected from
29
filamentous algae, P. pectinatus and V. americana in June and July, and the other
analysis focused on shrimp collected from P. pectinatus and V. americana in June, July
and August. In both cases, mean total lengths of mature females varied significantly
among combinations of months and habitats (Tables 3-16 and 3-17). On average,
mature females from V. americana were smaller, with the difference ranging from 2.0
mm in June to 6.3 mm in July (Figure 3-11). In addition, size frequency distributions for
mature females separated by habitats showed the presence of numerous, small, mature
females in samples from V. americana (Figure 3-12).
30
Table 3-1. Two-way ANOVA based on log10-transformed salinities (ppt) measured along three transects in February, May and August 2009. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.01) and homoscedastic (p > 0.01).
Factor df Sum of squares Mean square F-value p-value
Month 2 0.00929 0.00464 170.12 < 0.001
Transect 2 0.00018 0.00009 3.35 0.058
Month × Transect 4 0.00028 0.00007 2.58 0.073
Error 18 0.00049 0.00003
Table 3-2. Two-way ANOVA based on log10-transformed temperatures (°C) measured
along three transects in February, May and August 2009. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.01) and homoscedastic (p > 0.01).
Factor df Sums of squares Mean square F-value p-value
Month 2 0.38184 0.19092 11.23 0.001
Transect 2 0.04892 0.02446 1.44 0.263
Month × Transect 4 0.01267 0.00317 0.19 0.942
Error 18 0.30599 0.01700
Table 3-3. One-way ANOVA based on log10-transformed grass shrimp m-3 captured
from filamentous algae, P. pectinatus, V. americana and M. spicatum in July 2009. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Habitat 3 8.0596 2.6865 323.89 < 0.001
Error 4 0.0332 0.0083
Table 3-4. Two-way ANOVA based on log10-transformed grass shrimp m-3 captured
from P. pectinatus, V. americana and M. spicatum in July and August 2009. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and homoscedastic (p > 0.05).
Factor df Sum of squares Mean square F-value p-value
Month 1 0.0006 0.0006 0.05 0.835
Habitat 2 10.2357 5.1179 388.45 < 0.001
Month × Habitat 2 0.7273 0.3636 27.60 0.001
Error 6 0.0791 0.0132
31
Table 3-5. One-way ANOVA based on arcsin transformed proportional survival rates in filamentous algae, P. pectinatus and V. americana. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Habitat 2 1.116 0.558 22.13 < 0.001
Error 9 0.227 0.025
Table 3-6. Analysis of covariance based on log10-transformed wet weights with log10-
transformed total length as a covariate. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were non-normal (p < 0.01) and heteroscedastic (p < 0.01).
Factor df Sums of squares Mean square F-value p-value
Length 1 213.190 213.190 22000.00 < 0.001
Species 2 0.016 0.008 0.83 0.436
Isopod 1 0.241 0.241 24.50 < 0.001
Species × Isopod 2 0.005 0.002 0.25 0.778
Error 1676 16.521 0.010
Table 3-7. Two-way ANOVA based on intermolt periods associated with positive growth
for juveniles and males from filamentous algae, P. pectinatus and V. americana. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Life history stage 1 116.392 116.392 12.69 0.006
Habitat 2 1.301 0.650 0.07 0.932
Life history stage × Habitat 2 0.982 0.491 0.05 0.948
Error 9 82.557 9.173
Table 3-8. Two-way ANOVA based on log10-transformed positive instantaneous growth
rates for juvenile and male shrimp from filamentous algae, P. pectinatus and V. americana. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Life history stage 1 0.0742 0.0742 0.62 0.455
Habitat 2 0.0825 0.0413 0.34 0.720
Life history stage × Habitat 2 0.0468 0.0234 0.19 0.827
Error 8 0.9634 0.1204
32
Table 3-9. Two-way ANOVA based on arcsin transformed percentage growth rates for juvenile and male shrimp from filamentous algae, P. pectinatus and V. americana. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and heteroscedastic (p < 0.01).
Factor df Sums of squares Mean square F-value p-value
Life history stage 1 0.0066 0.0066 1.06 0.333
Habitat 2 0.0107 0.0053 0.80 0.457
Life history stage × Habitat 2 0.0009 0.0005 0.08 0.928
Error 8 0.0493 0.0062
Table 3-10. Regression analysis of instantaneous growth of all shrimp by holding time
(Instantaneous growth w/ month TL = 0.00282 -0.000037 * Holding time). Source DF SS MS F P
Regression 1 0.0005208 0.0005208 0.72 0.397 Residual Error 453 0.3284688 0.0007251
Table 3-11. Regression analysis of instantaneous growth of juvenile shrimp by holding
time (Instantaneous growth w/ month TL = - 0.00698 +0.000039 * Holding time).
Source DF SS MS F P
Regression 1 0.0003393 0.0003393 0.71 0.401 Residual Error 265 0.1268240 0.0004786
Table 3-12. Regression analysis of instantaneous growth of adult male shrimp by
holding time (Instantaneous growth w/ month TL = 0.0144 -0.000098 * Holding time).
Source DF SS MS F P
Regression 1 0.0014377 0.0014377 1.55 0.215 Residual Error 186 0.1729565 0.0009299
Table 3-13. Analysis of covariance based on log10-transformed numbers of eggs and
embryos with total length as a covariate. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were non-normal (p < 0.01) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Total length 1 0.5193 0.5193 20.78 < 0.001
Month 1 0.0039 0.0039 0.16 0.693
Habitat 2 1.5736 0.7868 31.49 < 0.001
Month × Habitat 2 0.0027 0.0014 0.05 0.947
Error 96 2.3984 0.0250
33
Table 3-14. Analysis of covariance based on log10-transformed egg volumes with total length as a covariate. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were non-normal (p < 0.01) and heteroscedastic (p < 0.01).
Factor df Sums of squares Mean square F-value p-value
Total length 1 50.381 50.381 1366.94 < 0.001
Month 1 5.183 5.183 140.63 < 0.001
Habitat 2 49.801 24.900 675.60 < 0.001
Month × Habitat 2 14.786 7.393 200.59 < 0.001
Error 6523 240.415 0.037
Table 3-15. Analysis of covariance based on log10-transformed brood volumes with
total length as a covariate. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were normal (p > 0.05) and heteroscedastic (p < 0.01).
Factor df Sums of squares Mean square F-value p-value
Total length 1 0.043 0.043 2.59 0.116
Month 1 0.024 0.024 1.44 0.238
Habitat 2 0.046 0.023 1.40 0.261
Month × Habitat 2 0.130 0.065 3.89 0.030
Error 35 0.584 0.017
Table 3-16. Two-way ANOVA based on total lengths of reproductive females from
filamentous algae, P. pectinatus, and V. americana in June and July 2009. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were non-normal (p < 0.01) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Month 1 0.016 0.016 9.13 0.003
Habitat 2 0.274 0.137 78.26 < 0.001
Month × Habitat 2 0.060 0.030 17.03 < 0.001
Error 358 0.628 0.002
34
Table 3-17. Two-way ANOVA based on total lengths of reproductive females from P. pectinatus, and V. americana in June, July, and August 2009. According to an Anderson-Darling test for normality and Cochran’s test for homoscedasticity, the data were non-normal (p < 0.01) and homoscedastic (p > 0.05).
Factor df Sums of squares Mean square F-value p-value
Month 2 0.045 0.022 10.35 < 0.001
Habitat 1 0.210 0.210 98.64 < 0.001
Month × Habitat 2 0.021 0.010 4.87 0.008
Error 272 0.590 0.002
35
Figure 3-1. Mean number of grass shrimp m-3 ± 95% confidence limits (CL).
36
Figure 3-2. Back-transformed mean proportional survival rates ± 95% confidence limits (CL) for grass shrimp tethered in three habitats.
0.0
0.2
0.4
0.6
0.8
1.0
Algae Potamogeton Vallisneria
Ba
ck-t
ran
sfo
rme
d m
ea
n p
rop
ort
ion
al su
rviv
al ra
te
95
% C
L
37
Figure 3-3. Size frequency distributions of grass shrimp collected from filamentous
algae, P. pectinatus and V. americana in June-August 2009.
38
Figure 3-4. Log10-transformed wet weight versus log10-transformed total length for all
Palaemonetes species with and without isopods.
Without isopody = 3.1758x - 5.4175
r² = 0.9273
With isopody = 3.1586x - 5.3582
R² = 0.9401
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Lo
g10[W
et w
eig
ht (
g)]
Log10[Total length (mm)]
Without isopod With isopod Linear (Without isopod) Linear (With isopod)
39
Figure 3-5. Total length (mm) vs. telson length (mm) of Palaemonetes spp.
0
5
10
15
20
25
30
35
40
45
50
0 1 2 3 4 5 6
Telson length (mm)
To
tal le
ngth
(m
m)
Total length=7.2356 x (Telson length) + 1.0069 r2 = 0.8879
40
Figure 3-6. Mean intermolt period ± standard error (SE) by habitat and life history stage.
41
Figure 3-7. Total number of molting shrimp by life history stage and holding time (12
hour intervals, 4 d).
42
Figure 3-8. Size frequency distributions of juveniles used in modal progression analysis.
0
20
40
60
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
June N
um
be
r o
f sh
rim
p
0
20
40
60
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
July
Num
be
r o
f sh
rim
p
0
20
40
60
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
August
Total length (mm)
Num
be
r o
f sh
rim
p
43
Figure 3-9. Relationships between log10-transformed number eggs and embryos and
total length for females collected in filamentous algae, P. pectinatus and V. americana.
44
Figure 3-10. Relationships between log10-transformed egg volumes and total length for
females collected in filamentous algae, P. pectinatus and V. americana.
Filamentous algaey = 0.7241x + 7.565
r² = 0.5653
Potamogeton pectinatusy = 0.9427x + 7.2952
r² = 0.7054
Vallisneria americanay = 1.157x + 6.9939
r² = 0.8536
0
2
4
6
8
10
0.0 0.5 1.0 1.5 2.0 2.5
Lo
g10[E
gg
vo
lum
e (
m3)]
Total length (mm)
Filamentous algae Potamogeton pectinatus Vallisneria americana
45
Figure 3-11. Back-transformed mean total lengths (mm) ± 95% confidence limits (CL)
for mature females.
46
Figure 3-12. Size frequency distribution for mature female grass shrimp females from
filamentous algae, P. pectinatus and V. americana.
47
CHAPTER 4 DISCUSSION
Following the recommendation of Van Horne (1983), the relative quality of four
habitats vegetated habitats in the Chassahowitzka River was compared by combining
measures of abundance with proxy measures of grass shrimp fitness. Myriophyllum
spicatum served as the poorest habitat, and it never yielded shrimp for measurements
of growth or reproductive output despite equivalent collection effort during the peak of
grass shrimp abundance. Nearby (within 5-50 m) patches of other habitats did yield
shrimp. Filamentous algae also was considered to provide poor habitat as it supported
relatively low numbers of grass shrimp compared to rooted plants. Shrimp associated
with filamentous algae also had the lowest relative survival rate, and exhibited lower
reproductive output than shrimp from V. americana. Potamogeton pectinatus served as
habitat of intermediate quality, with abundance of grass shrimp and their reproductive
output second only to collections from V. americana. In addition, relative survival rates
for grass shrimp tethered in P. pectinatus were not different from those of grass shrimp
tethered in V. americana. Vallisneria americana provided the highest quality habitat,
with higher densities of shrimp collected from this native macrophyte exhibiting high
survival rates and the greatest reproductive output due to production of smaller and
more numerous eggs at a smaller size. Intermolt intervals and growth rates were
variable and did not differ significantly for grass shrimp from all habitats. Each measure
of habitat quality yielded some insight, and each had associated challenges.
Relative abundance can be a misleading metric of habitat quality when used in
isolation (Van Horne, 1983; Heinrichs et al., 2010). Competition among life history
stages within a species and among different species for limited space and resources in
48
quality habitats may force animals into lesser quality habitat (Van Horne, 1983; Pulliam
and Danielson, 1991) Abundances also yield only a snapshot of distributions without
information on past events leading to or future effects arising from the observed
distributions (Van Horne, 1983; Pulliam and Danielson, 1991; Chockley et al., 2008;
Heinrichs et al., 2010). In addition, differential capture efficiencies can bias estimates of
relative abundance (Pierce et al., 1990; Bayley and Austen, 2002). The habitats in this
study did differ in form. Nevertheless, the larger differences in relative abundance (75×-
198×) were unlikely to arise solely from gear bias.
Differences in the structure of habitats probably affected relative survival rates, an
important metric of fitness, as evidenced by the results of tethering experiments.
Palaemonetes spp. commonly face strong predation pressure, and they seek refuge in
structurally complex habitats (Khan, Merchant & Knowton, 1997). For example, P. pugio
remained closer to and made greater use of refuge in the presence of predators (Davis,
Metcalfe & Hines, 2003). Therefore, the evaluation of relative percent survival by habitat
type serves as an important metric contributing to a grass shrimp’s fitness. In the
Chassahowitzka River, grass shrimp survived better in habitats where structure
extended throughout the water column, in contrast to mat-like filamentous algae.
Filamentous algae may prove to be a better refuge in late fall and early winter when
patches are more extensive and deeper (Camp. unpublished data). Even if filamentous
algae can provide a habitat of suitable quality, its tendency to wane during periods of
peak grass shrimp abundances and reproduction in the Chassahowitzka River (Camp.
unpublished data) leads to a mismatch between the needed refuge and its availability,
with potentially detrimental consequences for grass shrimp.
49
Survival rates as assessed by tethering experiments come with caveats. In this
study, the selection of similarly sized and co-located patches of filamentous algae, P.
pectinatus, and V. americana reduced confounding effects from variation in type and
density of predators, as well as variation in access due to differing perimeter to area
ratios (Darcy, 2003; Kneib and Scheele, 2000; Moore and Hovel, 2010). In fact,
experiments were performed in the centers of patches to further diminish the effect of
increased predation along perimeters (Darcy, 2003; Moore and Hovel, 2010). In an
effort to control for differences in visibility and movement of the grass shrimp prey
(Davis, 2003; Paine, 1976), tethering experiments were limited to non-ovigerous grass
shrimp with total lengths of 25-35 mm and no evidence of isopod infection.
Nevertheless, tethering of grass shrimp did involve suturing, and this process may have
resulted in the release of body fluids possibly altering predator behavior (Peterson &
Black, 1994). It is unlikely, however, that a release of body fluids would have led to
differential predation across habitats because the patches were co-located and subject
to similar assemblages of mobile predators. Tethers may limit or alter movement of
shrimp if they become entangled (Aronson and Heck, 1995), but entanglement of
tethers would likely increase in complex habitats leading to decreased avoidance of
predators and lower estimates of relative survival.
Unlike tethering, experimental assessments of growth rates using standard
techniques (Ross et al., 2000) did not yield significant insights into habitat quality.
Growth rates expressed as either instantaneous change in total length or percentage
change in total length, and components of these measures, i.e. intermolt period and
growth increment, were variable with no statistically significant differences among
50
habitats. Despite efforts to minimize stress on shrimp in this study (i.e., short elapsed
time from collection to maintenance, in situ maintenance, and four day holding period),
approximately 56% of the shrimp that molted did not grow or became smaller. Fewer
juveniles grew (31%) despite molting more frequently than adult males (64% grew).
Mean intermolt periods did not vary significantly among shrimp from alternative habitats,
but intermolt periods for juveniles were consistently shorter, which indicated that
experiments yielded useful data. In addition, mean intermolt periods (± standard error)
in this study were similar to those documented for P. pugio in Vernberg and
Piyatiratitivorakul (1998): i.e., for males 12.3 (± 1.0) d in this study compared to 11.2
(± 3.9) d and for juveniles 8.1 (± 0.6) d in this study compared to 7 days.
Another method to estimate growth was applied to juveniles collected from V.
americana. Modal progression analysis of a cohort yielded 0.05 mm d-1 as an estimate
of instantaneous growth, which was less than the estimate of 0.44 mm d-1 obtained from
positive results of molting experiments. The difference in these estimates may be due to
size dependent mortality, overlapping cohorts and the differential treatment of shrimp
that shrink or do not grow. Grass shrimp experience different predation rates at different
sizes (Kneib, 1987), so cohorts can be disrupted. Grass shrimp demonstrate the
capacity to grow or shrink in response to environmental conditions, so cohorts could
overlap. In addition, the estimate from molting experiments excluded shrimp that did not
grow, which would lead to a larger mean value.
Intermolt period and growth are affected by temperature and salinity (Quetin et al.,
2003; Tarling et al., 2006; Vernberg and Piyatiratitivorakul; 1998). Intermolt periods tend
to be shorter where resources are at their peak (Ross, 2000) and water is warmer,
51
within the physiological limits of the species (Vernberg and Piyatiratitivorakul, 1998). For
any given species, growth tends to be maximized at optimal salinities and temperatures
determined by the energetic demands associated with maintaining homeostasis, and
such demands increase either side of this range. As temperature and salinity were
similar among habitats in this study, these influences should not have generated
differences in intermolt periods or growth.
Molting frequency from growth experiments will be an accurate estimate of
readiness to molt at the population level only if molting is random (Hoenig and
Restrepo, 1989; Miller, Huntley & Brooks, 1984; Quentin et al., 2003; Ross et al., 2000;
Tarling et al., 2006). Crustaceans have been shown to exhibit synchronicity in molting
related to inter-individual, tidal, diel, and lunar cues, which may serve to protect newly
molted individuals by swamping potential predators (Miller et al., 1984). For example,
copepods have been documented to produce “bursts” of molting at frequencies higher
than those predicted by stage determination (Miller et al., 1984). Therefore, synchrony
may have affected the evaluation of intermolt periods for grass shrimp collected from
alternative habitats, although the effect of synchrony was diminished by using results
from multiple experiments conducted over three months to calculate intermolt periods.
Molting experiments have been used to evaluate shrimp and krill growth (Quetin
and Ross, 1991; Ross et al., 2000; Kneib, 1987). The method assumes that the
physiological basis for ecdysis and growth are set a few days before molting when the
animal reaches a nutritional saturation point. If this “point of no return” has been
reached, then intermolt period and growth increment are assumed to be independent of
collection and incubation (Ross et al., 2000). However, the validity of these assumptions
52
is drawn into question by the results of various studies. For example, growth increment
decreased rapidly after capture and isolation in one study of krill (Tarling et al., 2006).
Furthermore, both metrics, intermolt period and growth increment, were affected by food
supply (Buchholz et al., 1991; Gorokhova, 2002; Ikeda and Dixon, 1982; Tarling et al.,
2006). In Buchholz et al. (1991), doubling food availability in the laboratory increased
the growth rate of krill and addition of natural phytoplankton produced the highest
growth rates. In Ikeda and Dixon’s study (1982), krill starved for 211 consecutive days
experienced no greater mortality than controls, continued to molt, and demonstrated
negative growth (up to -50% wet weight). Negative growth increments may increase
survival of crustaceans that do not sequester lipids during periods of limited food. In
addition, Tarling et al. (2006) observed similar patterns, with growth increments of krill
falling by 26% and 50% and negative growth at days 3 and 5 of their experiment.
Growth increments decreased most in individuals that had been growing the fastest
(Tarling et al., 2006), which would equate to juveniles in this study. In comparison to
adult males, the high percentage of grass shrimp juveniles that did not grow suggested
that smaller grass shrimp have less metabolic reserves to cope with collection and
isolation. The grass shrimp in this experiment were held in situ, but they were isolated
from their natural food sources, which could have affected their growth (Kneib, 1987).
Differences in growth among life history stages also may be related to differences
in metabolism. Molting and somatic growth, along with ingestion, assimilation,
respiration, excretion and locomotion, represent the key metabolic demands for grass
shrimp that must be met by nutrition (Tarling et al. 2006). Juvenile and adult life history
stages have different metabolic demands, with respiratory demands increasing from
53
25% to 52% as shrimp increase in size from juveniles to adults. Energy in juveniles is
devoted predominately to somatic growth (14% of total metabolism), while energy in
adults is divided between reproduction (14%) and growth (5%). The energetic cost of
molting remains similar for adults and juveniles, and it is estimated to be only 2% of
their overall metabolism (Vernberg and Piyatiratitivorakul, 1998). Given this relatively
low metabolic cost, shrimp could be expected to molt in response to physiological and
other cues, even if they do not have the reserves needed to yield growth. Fortunately,
other proxy measures of fitness, i.e., relative survival and reproductive output, were
used to evaluate habitat quality in this study.
In combination with survival, reproductive output appeared to represent a good
proxy for fitness. Grass shrimp from V. americana brooded smaller and more numerous
eggs than shrimp collected from filamentous algae and P. pectinatus. Gulf coast grass
shrimp have been estimated to produce up to eight broods between April and October
(Bauer and Abdalla, 2000). At this rate, grass shrimp from V. americana would have
more than double the reproductive output of grass shrimp from filamentous algae or P.
pectinatus.
The number of eggs and size of eggs produced by Palaemonetes spp. in this
study were consistent with literature reports (Beck and Cowell, 1975; Broad, 1957;
Corey and Reid, 1991) ranging from 47-99 eggs per brood. A trade-off seemed to exist
between egg size and number of eggs per brood. Limits to reproductive output include
anatomical constraints and energetic resources (Hines, 1982). In numerous studies of
crustacean reproduction, female body size has proven to be an important factor in
determining reproductive output, brood weight, number of eggs per brood, and gonad
54
capacity (Clarke, 1993; Lara & Wehrtmann, 2009; Hines, 1982; Tallack, 2007). Brood
weight, and consequently brood volume, often was constrained to approximately 10% of
total body weight due to these allometric constraints and constraints on penetration of
dissolved oxygen into egg masses (Hines, 1982; Fernandez et al., 2006). Such
restrictions create a tradeoff between egg volume and the number of eggs per brood,
with total brood volume being relatively constant. An inverse relationship between egg
size and egg number per brood was seen in this study and reported for amphipods,
mysids, fairy shrimp, and brachyurans (Hines, 1982). In decapods, larger egg size and
decreased egg number, indicative of greater parental investment in each egg, have
been documented to decrease the number of zoeal stages and associated predation
proceeding metamorphosis to the first juvenile stage (Broad, 1957; Hines, 1982). In fact,
females may be able to read environmental cues and adjust their investment in each
egg (Fox and Czesak, 2004; Plaistow et al., 2004). In environments where food is
scarce or of poor quality, females may invest more energy or yolk per egg to improve
growth and survival of the resulting embryos (Clarke, 1993). In this study,
Palaemonetes spp. from V. americana produced more and smaller eggs, which may
indicate the presence of enhanced nutritional resources that will support development
through all zoeal and juvenile stages. In fact, grass shrimp in alternative habitats within
the Chassahowitzka River may be using the plasticity associated with egg size and egg
number to create reproductive strategies that maximize fitness.
Size at maturity is influenced by relative growth rates and nutritional resources
(Anger and Moreira, 1998). Size at maturity in Palaemonetes spp. is described by
Plaistow (2004) as being an overhead threshold: grass shrimp reach a minimum size or
55
state of maturity after which total body length exerts minimal effect on reproduction. The
results of this study were consistent with this theory, because brood volume did not
differ significantly for shrimp from different habitats, although females from V. americana
exhibited a smaller mean total length at maturity.
Hines (1982) described the possibility that a crustacean can “fine tune” fecundity
at low latitudes by producing many broods of many small eggs, as long as limitations of
small size and seasonality are counter-balanced. Grass shrimp collected from V.
americana brooded more eggs, with smaller volumes, at a smaller mean total length
than shrimp collected from alternative habitats. Grass shrimp from V. americana extend
their reproductive season and overall reproductive output by maturing more rapidly. This
pattern of early maturity and increased fecundity could increase fitness (Bertness,
1981). Few grass shrimp survive their first winter, in large part because they are
vulnerable to predation across all life history stages and body lengths, so shrimp benefit
from maturing and reproducing quickly and prolifically.
This study indicates that wide-scale habitat change from the native macrophyte V.
americana to nuisance levels of filamentous algae and M. spicatum could generate
negative consequences for populations of grass shrimp in the Chassahowitzka River.
Lower abundances, decreased reproductive output, and reduced survival, especially
during the critical summer months may obviate the shrimp’s role in the trophic ecology
of the Chassahowitzka River, resulting in detrimental effects on predators, including
commercially and recreationally important fish species. Thus, those responsible for
managing or restoring vegetation in the Chassahowitzka River and other systems
undergoing similar changes in vegetation should be concerned about such trends.
56
LIST OF REFERENCES
Anderson, G. (1985) Species profiles: Life histories and environmental requirements of coastal fishes and invertebrates (Gulf of Mexico): grass shrimp. U.S. Fish and Wildlife Service Biological Report 82(11), Slidell, Louisianna, Pp 1-19. Anger, K. and Moreira, G.S. (1998) Morphometric and reproductive traits of tropical caridean shrimps. Journal of Crustacean Biology, 18(4), 823-838. Aronson, R.B. and Heck, K.L. (1995) Tethering experiments and hypothesis testing in ecology. Marine Ecological Progress Series 121, 307-309. Bauer, R.T. and Abdalla, J.H. (2000) Patterns of brood production in the grass shrimp Palaemonetes pugio (Decapoda: Caridea). Invertebrate Reproduction and Development, 38(2), 107-113. Bayley, P.B. and Austen, D.J. (2002) Capture efficiency of a boat electrofisher. Transactions of the American Fisheries Society, 131, 435-451. Beck, J.T. and Cowell B.C. (1976) Life history and ecology of the freshwater Caridean shrimp, Palemonetes paludosus (Gibbes). American Midland Naturalist, 96(1), 52-65. Beck, M.W., Heck, K.L., Able, K.W., Childers, D.L., Eggleston, D.B., Gillanders, B.M., Halpern, B., Hays, C.G., Hoshino, K., Minello, T.J., Orth, R.J., Sheridan, P.F. & Weinstein, M.P. (2001) The identification, conservation, and management of estuarine and marine nurseries for fish and invertebrates. BioScience, 51(8), 633-641. Bell, S.S. and Coull, B.C. (1978) Field evidence that shrimp predation regulates meiofauna. Oecologica, 35,141-148. Bertness, M.D. (1981) Pattern and plasticity in tropical hermit crab growth and reproduction. The American Naturalist, 117(5), 754-773. Broad, A.C. (1957) Larval development of Palaemonetes pugio Holthuis. Biological Bulletin, 112(2), 144-161. Buchholz, F. (1991) Molt cycle and growth of Antarctic krill Euphausia superba in the laboratory. Marine Ecology Progress Series, 69, 217-229. Castellanos, D.L. and Rozas, L.P. (2001) Nekton use of submerged aquatic vegetation, marsh, and shallow unvegetated bottom in the Atchafalaya River Delta, a Louisiana tidal freshwater ecosystem. Estuaries, 24(2), 184-197. Chang, E.S., Chang, S.A. & Mulder, E.P. (2001) Hormones in the lives of crustaceans: an overview. American Zoologist, 41, 1090-1097.
57
Chockley, B. R., St. Mary, C.M. & Osenberg, C.W. (2008) Population sinks in the Upper Florida Keys: the importance of demographic variation in population dynamics of the marine shrimp Stenopus hispidus. Marine Ecology Progress Series, 360, 135-145. Clarke, A. (1993) Reproductive trade-offs in Caridean shrimps. Functional Ecology, 7(4), 411-419. Coen, L.D., Heck, K.L. & Abele, L.G. (1981) Experiments on competition and predation among shrimps of seagrass meadow. Ecology, 62(6), 1484-1493. Corey, S. and Reid, D.M. (1991) Comparative fecundity of Decapod crustaceans I. The fecundity of the thirty-three species of nine families of the Caridean shrimp. Crustaceana, 60(3), 270-294. Darcy, M.C. (2003) Influence of habitat corridors on dispersal success, predation, induced mortality, and colonization of estuarine macrofauna in seagrass. MS Thesis, North Carolina State University, North Carolina. Davis, J.L.D., Metcalfe, W.J., & Hines, A.H. (2003) Implications of a fluctuating fish predator guild on behavior, distribution, and abundance of a shared prey species: the grass shrimp Palaemonetes pugio. Journal of Experimental Marine Biology and Ecology, 293, 23-40. Fernández, M., Calderόn, R., Cifuentes, M., Pappalardo, P. (2006) Brooding behaviour and cost of brooding in small body size brachyuran crabs. Marine Ecology Progress Series, 309, 213-220. Fingerman, M. (1997) Crustacean endocrinology: A retrospective, prospective, and introspective analysis. Physiological Zoology, 70(3), 257-269. Fox, C.W. and Czesak, M.E. 2000. Evolutionary ecology of progeny size in arthropods. Annual Review of Entomology, 45, 341-369. Frazer, T.K., Notestein, S.K. & Pine III, W.E. (2006). Changes in the physical, chemical and vegetative characteristics of the Homosassa, Chassahowitzka and Weeki Wachee Rivers Final Report. Southwest Florida Water Management District, Brooksville, Florida, 163 pp. Frazer, T.K., Lauretta, M.V. & Pine III, W.E. (2011) Increased nutrient loading of spring-fed, coastal rivers: effects on habitat and faunal communities Final Report. Florida Fish and Wildlife Conservation Commission, State Wildlife Grants Program, Tallahassee, Florida. 152 pp. Gillanders, B.M., Able, K.W., Brown, J.A., Eggleston, D.B. & Sheridan, P.F. (2003) Evidence of connectivity between juvenile and adult habitats for mobile marine fauna: an important component of nurseries. Marine Ecological Progress Series, 247, 281-295.
58
Glancy, T.P., Frazer, T.K., Cichra, C.E. & Lindberg, W.J. (2003) Comparative patterns of occupancy by decapod crustaceans in seagrass, oyster, and marsh-edge habitats in a northeast Gulf of Mexico estuary. Estuaries, 26(5), 1291-1301. Gorokhova, E. (2002) Moult cycle and its chronology in Mysis mixta and Neomysis integer (Crustacea, Mysidacea): implications for growth assessment. Journal of Experimental Marine Biology and Ecology, 278(2), 179-194. Heinrichs, J.A., Bender, D.J., Gummer, D.L. & Schumaker, N.H. (2010) Assessing critical habitat: Evaluating the relative contribution of habitats to population persistence. Biological Conservation, 143, 2229-2237. Hines, A.H. (1982) Allometic constraints and variables of reproductive effort in Brachyuran crabs. Marine Biology, 69, 309-320. Hoenig, J.M. and Restrepo, V.R. (1989) Estimating the intermolt periods in asynchronously molting crustacean populations. Biometrics, 45, 71-82. Howard, R.K. (1985) Measurements of short-term turnover of epifauna within seagrass beds using an in situ staining method. Marine Ecology Progress Series, 22,163-168. Ikeda, T. and Dixon, P. (1982) Body shrinkage as a possible overwintering mechanism of the Antartic krill, Euphausia superba Dana. Journal of Experimental Marine Biology and Ecology, 62, 143-151. Khan, R.N., Merchant, H.C. & Knowton, R.E. (1997) Effects of macrophytic cover on the distribution of grass shrimps, Palaemonetes pugio and P. vulgaris. Invertebrate Biology, 166(3), 243-247. Kneib, R.T. (1987) Predation risk and use of intertidal habitats by young fishes and shrimp. Ecology, 68(2), 379-386. Kneib, R.T. and Scheele, C.E.H. (2000) Does tethering of mobile prey measure relative predation potential? An empirical test using mummichogs and grass shrimp. Marine Ecological Progress Series, 198, 181-190. Kunz, A.K, Ford, M., & Pung, O.J. (2006) Behavior of the grass shrimp Palaemonetes pugio and its response to the presence of the predatory fish Fundulus heteroclitus. American Midland Naturalist, 155(2), 286-294. Lara, L.R. and Wehrtmann, I.S. (2009) Reproductive biology of the freshwater shrimp Macrobrachium carcinus (L.) (Decapoda: Palaemonidae) from Costa Rica, Central America. Journal of Crustacean Biology, 29(3), 343-349.
59
Lowe, B.T. and Provenzano, A.J. (1990) Survival and reproduction of Palaemonetes paludosus (Gibbes, 1850) (Decapodas: Palaemonidae) in saline water. The Journal of Crustacean Biology, 10(4), 639-647. McCall, D.D. and Rakocinski, C.F. (2007) Grass shrimp (Palaemonetes spp.) play a pivotal trophic role in enhancing Ruppia maritime. Ecology, 88(3), 618-624. Miller, C.B., Huntley, M.E. & Brooks, E.R. (1984) Post-collection molting rates of planktonic, marine copepods: Measurement, application, problems. Limnology and Oceanography, 29(6), 1274-1289. Moore, E. C. and Hovel, K. A. (2010) Relative influence of habitat complexity and proximity to patch edges on seagrass epifaunal communities. Oikos, 119, 1299-1311. Morgan, M.D. (1980) Grazing and predation of the grass shrimp Palaemonetes pugio. Limnology and Oceanography, 25(5), 896-902. Paine, R.T. (1976) Size-limited predation: An observational and experimental approach with the Mystilus-Pisaster interaction. Ecology, 57, 858-873. Peterson, C.H. and Black, R. (1994) An experimentalist’s challenge: when artifacts of intervention interact with treatments. Marine Ecological Progress Series, 111, 289-297. Pierce, C.L., Rasmussen, J.B., & Leggett, W.C. (1990) Sampling littoral fish with a seine: Corrections for variable capture efficiency. Canadian Journal of Fisheries and Aquatic Sciences, 47,1004-1010. Plaistow, S.J., Lapsley, C.T., Beckerman, A.P. & Benton, T.G. (2004) Age and size at maturity: sex, environmental variability and development thresholds. Proceedings of the Royal Society London B, 271, 919-924. Pulliam, H.R. and Danielson, B.J. (1991) Sources, sinks, and habitat selection: A landscape perspective on population dynamics. The American Naturalist, 137 Supplement, S50-S66. Orth, R.J., van Montfrans, J., & Fishman, J. (1999) Virginia Institute of Marine Science. Report to the Virginia Marine Resources Commission: A preliminary study of predation on blue crabs by three fish predators in a seagrass bed. http://web.vims.edu/bio/sav/bluecrabpred/?=www. Accessed 12 May, 2011. Quetin, L.B. and Ross, R.M. (2003) Episodic recruitment in Antarctic krill Euphausia superba in the Palmer LTER study region. Marine Ecology Progress Series, 259, 185-200.
60
Raviv, S., Parnes, S. & Sagi, A. (2008) Coordination of reproduction and molt in decapods. In: Reproductive Biology of Crustaceans (Ed E. Mente), pp. 365-390.: Science Publishers, Enfielied, New Hampshire. Rosas, L.P. and Odum, W.E. (1987) Use of tidal freshwater marshes by fishes and macrofaunal crustaceans along a marsh stream-order gradient. Estuaries, 10, 36-43. Ross, R.M., Quetin, L.B., Baker, K.S., Vernet, V. & Smith, R.C. (2000) Growth limitation in young Euphausia superba under field conditions. Limnology and Oceanography, 45, 31-43. Rowe, C.L. (2002) Differences in maintenance energy expenditure by two estuarine shrimp (Palaemonetes pugio and P. vulgaris) that may permit partitioning of habitats by salinity. Comparative Biochemistry and Physiology Part A, 132, 341-351. Skadsheim, A. (1984) Coexistence and reproductive adaptations of amphipods: The role of environmental heterogeneity. Oikos, 43, 94-103. Tallack, S.M.L. (2007) Size-fecundity relationships for Cancer pagurus and Necora puber in the Shetland Islands, Scotland: How is reproductive capacity facilitated? Journal of the Marine Biological Association of the United Kingdom, 87, 507-515. Tarling, G.A., Shreeve, R.S., Hirst, A.G., Atkinson, A., Pond, D.W., Murphy, E.J. & Watkins, J.L. (2006) Natural growth rates in Antarctic krill (Euphausia superba): I. Improving methodology and predicting intermolt period. Limnology and Oceanography, 51, 959-972. Thorpe, J.H. (1976) Interference competition as a mechanism of coexistence between two sympatric species of the grass shrimp Palaemonetes (Decapoda: Palaemonidae). Journal of Experimental Marine Biology and Ecology, 25,19-35. Van Horne, B. (1983) Density as a misleading indicator of habitat quality. The Journal of Wildlife Management, 47(4), 893-901. Vernberg, F.J. and Piyatiratitivorakul, S. (1998) Effects of salinity and temperature on the bioenergetics of adult stages of the grass shrimp (Palaemonetes pugio Holthius) from the North Inlet Estuary, South Carolina. Estuaries, 21(1), 176-193. Welsh, B.L. (1975) The role of grass shrimp, Palaemonetes pugio, in a tidal marsh ecosystem. Ecology, 56(3), 513-530. Williams, A.B. (1984) Shrimps, Lobsters, and Crabs of the Atlantic Coast of the Easter United States, Maine to Florida. Smithsonian Institution Press, Washington D.C.
61
BIOGRAPHICAL SKETCH
Meredith Montgomery was born and raised in Gainesville, Florida. After graduating
from high school in 2001, she headed to Duke University to study biology and Spanish,
study abroad as much as possible, and sleep in a few tents in Krzyzewskiville. After an
influential semester and summer studying marine science at the Duke University Marine
lab and graduating in May 2005, Meredith took job opportunities in the marine sciences
that offered new skills, adventure, and travel; working on projects varying from marine
mammal and sea turtle ecology to seagrass and water quality projects. Finding
employment and a home in the Frazer laboratory at the University of Florida, Meredith
decided to stay at UF to begin graduate school. While in graduate school, she spent a
great deal of time thinking about Florida’s Gulf Coast rivers, taking a brief departure to
study lionfish ecology in the mangroves and patch reefs of San Salvador, Bahamas.
Meredith has decided to return to her two first loves: animals and medicine. She will
enroll at the University of Florida’s College of Veterinary Medicine Fall 2011.