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.