1

OXYGEN MEDIATED GRAZING IMPACTS IN FLORIDA SPRINGS

By

KRISTIN DORMSJO

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

2008

© 2008 Kristin Dormsjo

2

To everyone who supported me.

3

ACKNOWLEDGMENTS

I would like to thank my parents for their love and support, as well as my husband.

Without them, none of this would have been possible. I would also like to thank my committee

members, Dr. Thomas Frazer, Dr. Bill Lindberg, Dr Mike Allen and Dr. Gary Warren, for all of

their guidance and expertise. Lastly, I would like to thank the Frazer lab, who were like my

family.

4

TABLE OF CONTENTS page

ACKNOWLEDGMENTS ...............................................................................................................4

LIST OF TABLES ...........................................................................................................................6

LIST OF FIGURES .........................................................................................................................7

ABSTRACT .....................................................................................................................................8

CHAPTER

1 INTRODUCTION ....................................................................................................................9

2 MATERIALS AND METHODS ...........................................................................................13

Study Site ................................................................................................................................13 Experimental Organism ..........................................................................................................14 Abundance and Distribution of Gastropods Including Elimia floridensis ..............................15 Grazing Effects .......................................................................................................................17 Statistical Analysis ..................................................................................................................20

3 RESULTS ...............................................................................................................................27

Environmental Parameters ......................................................................................................27 Abundance and Distribution of Gastropods Including Elimia floridensis ..............................28 Grazing Effects .......................................................................................................................29

4 DISCUSSION .........................................................................................................................48

APPENDIX ICHETUCKNEE MEAN ENVIRONMENTAL PARAMETERS ......................56

REFERENCE LIST .......................................................................................................................60

BIOGRAPHICAL SKETCH .........................................................................................................69

5

LIST OF TABLES

Table page 2-1 Transect Locations in the main river and feeder springs from 2007 transect

sampling.. ...........................................................................................................................23

3-1 Ichetucknee spring and river environmental parameters from caging experiment conducted in April, 2007. ...................................................................................................31

3-2 Gastropod species and total number of individuals surveyed for each species within river and feeder spring transects (2007). ............................................................................32

3-3 Differences of least squares means investigating the significant location*treatment interaction from mixed model analysis ..............................................................................33

A-1 Ichetucknee River mean environmental parameters for transect data collected in May 2003 and April 2004 (Kurz et al. 2003 and 2004). ............................................................57

6

LIST OF FIGURES

Figure page 2-1 Map of the Ichetucknee Springs State Park (ISSP) in Ft. White, Florida. .........................24

2-2 Sampling device designed for 2007 transect sampling.. ....................................................25

2-3 Total length data (mm) of Elimia floridensis from 2007 transect sampling. .....................26

3-1 Mean dissolved oxygen concentration (mg l-1) by transect from 2003, 2004 and 2007....................................................................................................................................34

3-2 Mean dissolved oxygen concentration (mg l-1 ± SD) per transect in the three feeder springs (2007).. ..................................................................................................................35

3-3 Mean current velocity (m s-1) vs. mean dissolved oxygen concentration (mg l-1).. ...........36

3-4 Macroinvertebrate abundance by transect (0.25 m2) from 2003. .......................................37

3-5 Mean Elimia floridensis abundance by transect (0.0625 m2) from 2003 and 2007. ..........38

3-6 Mean Elimia floridensis abundance (± SD) by transect in feeder springs from 2007 .......39

3-7 Density Ellipse for Singing Spring ....................................................................................40

3-8 Density Ellipse for Devil’s Eye. ........................................................................................41

3-9 Density Ellipse for Mill Pond. ...........................................................................................42

3-10 Density Ellipse for the main river. .....................................................................................43

3-11 Bivariate fit of mean of Elimia floridensis abundance by flow (m s-1) from 2007 ............44

3-12 Peripyton accumulation (µg Chl a/cm2, ± SD) in control dishes and snail treatment dishes within the main river.. .............................................................................................45

3-13 Weekly Estimates of chlorophyll a concentration (ug Chl a/cm2, ± SD) from control and snail dishes in the feeder springs .................................................................................46

3-14 Periphyton accumulation rates (µg Chl a/cm2 ± SD) in control dishes and snail treatment dishes within the main river. ..............................................................................47

7

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

OXYGEN MEDIATED GRAZING IMPACTS IN FLORIDA SPRINGS

By

Kristin Dormsjo

May, 2008

Chair: Thomas Frazer Major: Fisheries and Aquatic Sciences

Dissolved oxygen (DO) is one of several abiotic factors that exerts a strong influence on

the abundance and distribution of aquatic organisms. The Ichetucknee River and other spring-

dominated systems in Florida are fed, in large part, by groundwater emanating from the Floridan

Aquifer, one of the largest and most volumetrically productive karst aquifers in the world. The

DO concentrations in this discharged groundwater can be extremely low and near anoxic in some

cases. As a consequence, the abundance of grazing organisms may be markedly lower in and

around discharge points. In this study, I investigated the possibility that reduced grazer

abundances near spring vents results in reduced grazing pressure and facilitates the proliferation

of periphyton in these low DO areas. I first established the distributional patterns of gastropods

(>300 µm) within the Ichetucknee River and then carried out a manipulative experiment

involving the most abundant grazer, Elimia floridensis. Elimia floridensis were significantly

more abundant in the main river than in feeder springs, where they significantly reduced the rate

of periphyton accumulation on glass Petri dishes relative to controls. Within the feeder springs,

however, snails suffered substantial mortality (~50%) and were not able to significantly reduce

periphyton relative to controls. These findings provide new insights into the ecology of Florida’s

springs and demonstrate the potential importance of top down control of periphyton growth.

8

CHAPTER 1 INTRODUCTION

Nutrient over-enrichment is perhaps the most pervasive issue affecting aquatic ecosystems

around the globe (Pedersen & Borum, 1996; Hillebrand, 2002; Burkepile & Hay, 2006). Duarte

(1995) summarized popular thought concerning the effects of nutrient enrichment on submerged

aquatic vegetation (SAV), stating that as nutrient concentrations increase, fast-growing algae will

eventually cause losses of macrophytes through shading processes. Once the macrophytes within

a system start to decline, primary production within the system becomes dominated by algae

which, in turn, can lead to further shading and macrophyte losses (see also Jeppesen et al., 1991).

Increased nutrient delivery thus can significantly alter ecosystem structure and potentially

compromise ecological function (Gulis & Suberkropp, 2004; Gafner & Robinson, 2007; Wang et

al., 2007).

Bottom-up shifts from macrophyte to algal dominance are particularly possible in

systems throughout Florida where both recent human population growth and past and present

land use have lead to nutrient enrichment of ground-water (Jones et al., 1997). Florida’s highly

permeable karst geology, which includes more than 300 freshwater springs, facilitates the

exchange of this enriched ground-water with surficial water systems. Spring systems are fed

primarily by the Floridan aquifer which is one of the largest and most productive karst aquifers

in the world. The aquifer underlies an area of 100,000 square miles in southern Alabama,

southeastern Georgia, southern South Carolina and all of Florida (USGS, 2005). However,

contamination of this aquifer by nitrate and other contaminants has been linked to changes in

community composition, plant biomass, and increases in nuisance species (Wright &

McDonnell, 1986a, 1986b). In situ experimental manipulations have demonstrated that

9

additional enrichment may lead to significant increases in periphyton growth and abundance

(Notestein et al., 2003).

In many instances, however, periphyton growth is not regulated by nutrient inputs alone,

but by complex interactions between a variety of abiotic and biotic parameters with different

factors predominating under different conditions. Limited light availability, for example, can

regulate periphyton accrual in heavily shaded streams (Minshall, 1978; Hill et al., 1995;

Greenwood & Rosemond, 2005). In other cases, changes in hydrologic regimes (e.g., prolonged

periods of low flow) and/or cascading trophic effects resulting from higher order predators, can

influence algal abundance, including periphyton associated with macrophytes (Biggs et al., 2000;

Hargrave et al., 2006). While it is widely accepted that ‘top-down’ and ‘bottom-up’ factors are

not necessarily mutually exclusive (Leibold et al., 1997), there is limited consensus concerning

the relative importance of the two under varying natural conditions (Power, 1992) and there is

debate among ecologists as to which factors play the greatest role in limiting periphyton growth.

Despite this debate, many investigations conducted in controlled mesocosm environments

indicate the potential for mesograzers (e.g., amphipods, gastropods, crustaceans, and small fish)

to mitigate the effects of increased nutrient loads by controlling algal biomass and productivity

within stream and lake ecosystems (Feminella et al., 1989; Steinman et al., 1987; Feminella &

Hawkins, 1995; Steinman, 1996; Alavarez & Pardo, 2007). Within stream ecosystems,

mesograzer assemblages are often highly diverse and some species are classified into functional

groups (i.e. grazers, shredders, filterers, gatherers) based on their feeding strategy. When

considered at the community level, these species can have an important influence on nutrient

cycles, decomposition, and translocation of material (Wallace & Webster, 1996). With regard to

nutrient enrichment, however, it is often scraper species that are functionally most important for

10

controlling periphyton proliferation as, mechanistically, these species scrape away proliferating

periphytic algae (Cummins & Klug, 1979; Richardson, 1991) thereby maintaining nutrient and

light availability for SAV (Rosemond et al., 1993; Feminella & Hawkins, 1995). Complete ‘top-

down’ grazer control of periphyton proliforation has rarely been reported in situ however,

suggesting that there are other environmental factors influencing epiphyte-grazer relationships.

Dissolved oxygen (DO) has long been considered an important environmental factor in

aquatic systems and it has been shown, in fact, that low DO conditions (i.e. hypoxia and anoxia)

are limiting resources in ground-water and can influence the density, diversity, and composition

of invertebrate communities (Williams & Hynes, 1974; Strommer & Smock, 1989, Tyson &

Pearson, 1991). Few organisms can tolerate prolonged exposure to low DO concentrations

without experiencing adverse effects on respiration, growth, reproduction, or survival (Levin &

Gage, 1997; Eden et al., 2003; Watson & Ormerod, 2004; Bergquist et al., 2005; Wu & Or,

2005). For some invertebrate species, metabolic depression induced by low DO conditions can

lead to decreases in feeding rate, allowing for the build up of substantial amounts of organic

matter within a system and increased sedimentation rate (Gray et al., 2002; Churchill & Storey

1995; Bjelke, 2005). Increased sedimentation rate can, in turn, lead to further decreases in DO

concentrations, leading to anoxia and reductions in the abundance, biomass, diversity, and

species richness of benthic fauna (Gray et al., 2002; Montagna & Ritter, 2006).

Because DO is an important factor affecting the physiology of mesograzers, I investigated

the influence of DO on grazer-epiphyte interactions in a Florida spring-fed system (i.e. the

Ichetucknee River). This system, like many other spring-fed systems, provides an excellent

natural laboratory to investigate these processes as physical and chemical characteristics, other

than DO, are fairly constant (Odum, 1957). In several of Florida’s spring-fed systems, DO

11

concentrations in groundwater have been reported to be as low as 0.3 mg l-1 near the spring

vents, to as high as 8.0 mg l-1 in the adjoining rivers (Odum & Caldwell, 1955; Woodruff 1993;

Malard & Hervant, 1999). In a study of 58 Florida springs, more than 60% were found to be

hypoxic (Rosenau et al., 1977).

In the Ichetucknee River, DO values in the main channel range from 3.6 to 8.1 mg l-1 and

increase rapidly with distance away from the headspring due to photosynthesis by SAV and

atmospheric diffusion (Kurz et al., 2003 & 2004; Duarte et al. unpublished data). However, mid-

day DO concentrations in the feeder springs can be as low as 0.36 mg l-1. Kurz et al. (2003)

reported that the oxygen gradient within the main channel of the Ichetucknee River is positively

correlated with increasing macroinvertebrate abundance (i.e. macroinvertebrates increased with

distance downstream). The nature of this relationship, however, has not been fully explored.

Herein, I evaluate the hypothesis that low DO concentrations influence grazer distribution

in the Ichetucknee River and several of its feeder springs. Specifically, I tested the prediction that

Elimia floridensis, a primary grazer in this system, would be more abundant the main river than

in the feeder springs and spring runs that are characteristically low in DO. In addition, I evaluate

the hypothesis that low DO conditions within feeder springs influences both grazer survivorship

and grazing rates. Specifically, I tested the prediction that grazing on periphyton will be low in

these areas relative to the main river channel.

12

CHAPTER 2 MATERIALS AND METHODS

Study Site

The Ichetucknee River, a tributary of the Santa Fe River, is located 4.28 km west of Ft.

White, Florida; it is a spring-fed stream approximately 8.05 km long with an annual discharge of

109.7 ft3 sec-1 (Florida Geological Survey, 1977). The river is characterized by distinct zones, i.e.

the Headspring Reach, Rice Marsh, and downstream Floodplain (Kurz et al., 2003 and 2004).

Dominant forms of vegetation in both the Headspring and Rice Marsh are Sagittaria kurziana

and Zizania aquatica, while in the Floodplain, beds of S. kurziana become mixed with

Vallisneria americana (Kurz et al., 2003). The primary spring inputs come from a series of large

vents (Blue Hole, Singing Spring, Devil’s Eye, and Mill Pond) which lie within the upper 4.02

km of the river (Figure 2-1). Singing Spring is part of the Mission Springs Group and lies closest

to the river’s head spring while Devil’s Eye is located approximately 1.21 km downstream of the

Head Spring in the Rice Marsh. Mill Pond is the furthest downstream in the Floodplain. The

upper 4.82 km of the river, along with all of its major contributing springs, are contained within

the Ichetucknee Springs State Park (ISSP), a designated National Natural Landmark (Evans,

2006). All sampling for this study was carried out within the confines of the ISSP.

In recent years, vent areas and their associated spring runs along the Ichetucknee River

have exhibited a highly publicized increase in filamentous algae and subsequent loss of

macrophytes. It is widely assumed that increased nitrate concentrations are responsible for these

changes. However, within the system the mean nitrate concentration for the main river is 523 µg

l-1 while the mean concentration within the feeder springs is 519 µg l-1 (Kurz et al., 2004). Some

feeder springs, in fact, exhibit nitrate concentrations as low as 293 µg l-1 in the some of the

feeder springs. While these mean concentrations are within the range of values measured in other

13

spring-fed systems such as the Weeki Wachee and Homosassa Rivers (781 and 436 µg l-1,

respectively) where there have been recent declines in macrophyte abundance (Frazer et al.,

2006), it is unlikely that nitrate concentrations alone are causing the vegetation losses observed

within the feeder springs and their associated spring runs as mean nitrate concentrations within

the river are higher.

Relative to many other spring-fed systems, soluble reactive phosphorus (SRP)

concentrations within the Ichetucknee are high, owing in part to phosphorus rich deposits within

the springshed. In fact, before being acquired by the state, phosphate mining occurred on the

ISSP property during the early 20th century and the 1950 – 60’s (FDEP, 2000). Mean SRP values

of 47.3 µg l-1 and 49.9 µg l-1 have been reported for the Ichetucknee main river and springs,

respectively. In comparison, maximum mean SRP concentrations in the headwater areas of the

Weeki Wachee and Homosassa spring systems are 21 and 22 µg l-1, respectively (Frazer et al.,

2006). It is noteworthy that while SRP concentrations within the Ichetucknee are relatively high

compared to other spring-fed systems in the region, these concentrations may not stem from

anthropogenic activities in the springshed. There are too few data available that would allow a

determination of whether nitrogen or phosphorus has increased significantly in the Ichetuknee

River and its associated feeder springs.

Experimental Organism

To explore the possibility that dissolved oxygen concentrations influence grazer

distribution, abundance, and grazing rates, we targeted the most macroinvertebrate grazer within

the Ichetucknee, the prosobranch gastropod Elimia floridensis. Snails from the genus Elimia are

commonly found in freshwater systems throughout central and southeastern North America and

can play a role in the processes that structure communities (Newbold et al., 1983; Richardson et

al., 1988). Elimia, for example, has been shown to influence the standing crop and production of

14

periphyton and also the structure of the algal community (Tuchman & Stevenson, 1991; Hill et

al., 1992; Rosemond et al., 1993).

Although there is a limited amount of published information on the species of interest, E.

floridensis, several characteristics appear common to the genus. Egg production occurs in late

spring and early summer, with juveniles appearing in late summer and fall (Huryn et al., 1994).

Juveniles undergo two discrete periods of growth during their first and second summers resulting

in a minimum cohort duration of at least two years (Huryn et al., 1994). Although direct aging

techniques are not available at present, the snails appear to be long lived, with longevities being

reported anywhere from 2 to 11 years (Dazo, 1965; Mancini, 1978; Payne, 1979; Richardson et

al., 1988).

Abundance and Distribution of Gastropods Including Elimia floridensis

Field sampling. In the spring of 2007, chemical and physical parameters were sampled

along 14 regularly spaced transects within the ISSP. Coordinates for these transects (Table 2-1)

correspond to the first 14 transects (starting near the headspring and continuing downstream)

previously established by Kurz and colleagues (Kurz et al., 2003 and 2004). In order to quantify

gastropod distribution and abundance throughout the river, each transect was divided into three

stations, one in the middle of the channel and two on either side, approximately halfway between

the bank and middle of the river. Transect work was also carried out within the river’s three

major springs; Singing Spring, Devil’s Eye, and Mill Pond. These springs in particular were

chosen over the river’s other 14 springs based on the considerable anecdotal macrophyte losses

seen in each and ease of access. Within these areas, the first transect was located just below the

discharge point with the next two transects in the middle and end of the spring run, respectively.

Coordinates for all spring transects were recorded with a differentially corrected hand-held GPS

receiver.

15

At each transect station, water depth (m) was measured with a telescoping fiberglass

survey rod marked in 0.01 m increments. Temperature (°C), DO (mg l-1), and specific

conductivity (µS cm-1) were measured with a Yellow Springs Instrument Company model 650

hand-held meter. Average current velocity (m s-1) was measured at 60% of water column depth

with a Marsh-McBirney Model 2000 portable flow meter. All measurements were taken between

10 am and 3 pm.

Macrophyte above-ground biomass, periphyton, and gastropods were sampled at each

station within a 0.0625 m2 area delineated with a quadrat. The quadrat was comprised of one,

three sided, 0.10 m polyvinyl chloride pipe (PVC) frame cut transversely and attached to a 1.82

m, 425-μm Nitex mesh bag with zip ties (Figure 2-2). The bottom of the frame was weighted. In

its initial position, the bag was folded accordion-style and contained within the three sided PVC

frame. The fourth side of the quadrat was a movable flap of Nitex mesh with a zipper lining the

length of the side.

To deploy the quadrat, two divers on SCUBA were needed. One diver would gently guide

the quadrat through the SAV canopy and place it on the river bed with the open side oriented in

to the direction of flow. The second diver would then draw the loose flap of Nitex across the

open side of the PVC frame to the other Nitex panel with a matching half of zipper. The two

sides of the frame were pulled apart and zippered quickly to enclose the sample. Once the

sampler was closed and fully extended, the top half was folded over on itself to close the quadrat

and all SAV above-ground biomass was cut approximately 2.5 cm from sediment and allowed to

float up into the mesh bag. The two halves of the PVC frame were then gathered together and the

sampler was brought to the surface. On the cleared patch sediment, number of shoots and

number of blades per shoot were recorded. Because some snails could be seen falling off the

16

macrophytes within the mesh bag onto the sediment below, live snails on the cleared patch of

sediment were collected and brought to the surface where they were measured and identified. All

snails that were collected were measured to the nearest 0.1 mm with vernier calipers, identified

using an identification manual for the freshwater snails of Florida (Thompson, 2004), and

returned immediately to the river. Once all snails were processed, SAV was placed in pre-labeled

plastic bags and transported on ice to the laboratory for additional processing.

Laboratory analysis. Upon returning to the laboratory, plant material was blotted dry with

paper towels then sorted into species specific groups. Wet weight for each group was obtained,

after which time samples were dried at a constant temperature (65°C) for 48 hours and re-

weighed to acquire a dry weight measurement. The SAV surface area per quadrat was estimated

by applying a surface area to biomass correction factor to dry weights (see Hoyer & Canfield,

2001). Macrophyte biomass was expressed on an areal basis.

Grazing Effects

Experimental procedures. In order to quantify the effects of low DO on Elimia floridenis

survivorship and grazing rates, one hundred and eighty cages were made; each consisting of a

100 x 20 mm glass Petri dish with a conical wire lid constructed from cured galvanized steel

wire mesh (6.35-mm mesh). Prior to the beginning of the experiment, all Petri dishes were

sanded for 20 seconds with sandpaper attached to an electric drill bit to facilitate periphyton

colonization. The wire lids were cured by soaking them in a water bath for a period of three

months to remove zinc, which is known to be toxic to gastropods (Devi, 1997; Gay & Maher,

2003; Taylor & Maher, 2003).

After construction of the cages was complete, six identical PVC stands measuring 88 x 22

x 15 cm and holding five Petri dishes were arrayed at each of three river sites in a stand of S.

kurziana slightly above the confluence of one of three smaller spring runs (i.e. Singing Spring,

17

Devil’s Eye, or Mill Pond). In the three adjoining feeder springs, six PVC stands were arrayed

just downstream of the primary discharge point. Submersed aquatic vegetation, if present

beneath a stand, was trimmed to approximately 2.5 cm to allow adequate light penetration for

periphtyon growth. All Petri dishes on a stand were positioned at a 45° angle into the stream

flow.

After the initial deployment of stands and cages, periphyton were allowed to colonize the

Petri dishes in the absence of grazers for a one-week period. Following this periphyton

colonization period, individual snails were randomly assigned to cages associated with three of

the six stands at each site (one snail per cage approximated the natural density of snails in the

river; unpubl. data). Dishes on three remaining stands at each site did not receive a snail and

served as controls. Even though control dishes did not contain snails they each received a wire

lid to minimize any potential confounding issues related to caging effects All snails used in the

experiment were collected from the main river and were between 15 and 23 mm total length

(TL). This size range captured the numerically dominant size classes in this system (Figure 2-3)

and also ensured that snails were large enough so that they could not escape through the wire

mesh of the cages. While smaller gastropods were seen on the outside of the dishes, they were

rarely observed inside the wire lid. Those that were found inside the dishes were removed within

two days of colonization.

At the same time snails were added to the cages, one dish from every stand at all river and

spring sites was collected at random to quantify changes in Chl a. At the time of collection, a

diver placed each Petri dish in an individual quart sized plastic bag and transported the bag to the

surface. While underwater, bags were held shut until the dish was inserted to minimize the

introduction of ambient water. At the surface, each bag containing a Petri dish was placed in an

18

individual plastic sandwich container and stored in an ice-filled cooler for transport to the

laboratory for subsequent analysis for Chl a. This procedure was repeated every week for a one

month period. Depth, water temperature, dissolved oxygen concentration, pH and current

velocities rate were measured at each site between the hours of 10 am and 2 pm during each

collection period.

During the course of the study, all cages were inspected twice a week to remove fouling

material. During these inspections, the wire lids associated with the Petri dishes were gently

brushed with a soft bristle toothbrush to remove fouling material and the apparent health of

snails was assessed. With respect to the latter, cages were tilted to see if the snail’s foot was

attached. If the foot was unattached, the wire lid was removed and the snail picked up to see if

the body floated free from the shell. If a snail was determined to be dead before its associated

dish was selected for sampling, the shell was collected and brought back to the laboratory to be

weighed and measured. The associated Petri dish and wire cage, however, was left undisturbed

in the water until it was selected for analysis to avoid confusion in subsequent sampling events.

If the snail was unresponsive, but remained in its shell, covered by the operculum, it was put

back in the dish and the wire lid re-attached. Snails were replaced on three separate occasions

when they escaped from dishes at river sites. Replacement took place within two days of each

escape and these dishes were marked so the Chl a concentrations they yielded could be examined

later. In each case, Chl a concentrations fell within the range of concentrations from other dishes

collected during the same week. Thus, they were included in all analyses.

Laboratory analysis. Less than 24 hours after dishes had been collected from the river,

they were processed as follows. The bottoms of dishes (areas subject to grazing) were gently

brushed with a soft bristle tooth brush to loosen the attached periphyton. Loosened periphyton

19

was rinsed from the dish into a Nalgene holding reservoir using a squeeze bottle filled with

deionized (DI) water. This process was repeated until the bottom of the dish was visibly free of

periphyton. The resultant slurry was poured through a 1.0-mm screen into a similar reservoir to

remove large particulate debris. The slurry was then homogenized with a plastic paddle and a

sub-sample of known volume was filtered through a Whatman GF/F glass-fiber filter. Filters

were then frozen until further analysis, at which time chlorophyll a extraction was accomplished

with a hot ethanol method (Sartory & Grobbelarr, 1984). Chlorophyll concentrations were

determined spectrophotometrically and the acidification method was used to correct for

phaeophytin (APHA, 1992).

Statistical Analysis

When appropriate, mean transect values were used to summarize abiotic parameters,

gastropod abundance and vegetative measures in order to facilitate comparisons between the

river and feeder springs. Equal variance was determined using the Brown-Forsythe test for

unequal variance (Prob > F = 0.05). In JMP (v5.1, SAS), one-way analysis of variance

(ANOVA) was used to test differences in current velocities and snail abundance between river

and spring locations (α = 0.05). Snail density within the main river and feeder springs was

explored using elliptical contours in JMP (v5.1, SAS). Ellipses enclose the densest 50 % of the

estimated distribution and use correlation estimates to determine if snail number and transect are

related. In this case, transect is used as a proxy for distance from spring vents. Due to differences

in current velocities between the main river and feeder springs, regression analyses were run to

examine the impact of current velocity on snail distribution.

A Welch ANOVA F-test was used to compare mean DO concentrations between river and

spring transects (α = 0.05). For this analysis the main river was considered one site, therefore

there was no replication for this location. Thus, to facilitate comparison of DO concentrations

20

between the main river and feeder springs transects within the three feeder springs were

combined into one site and the means from the two locations (i.e., river and spring) were

compared. While this approach does not take into account transect variation due to distance from

the spring vent within the feeder springs, the purpose of the comparison was simply to reveal any

overall differences in DO concentrations between river and spring environments.

To analyze the effect of location (i.e., river or spring) on grazing effects from the grazer

manipulation experiment, an AR(1) covariance model was used in mixed-model repeated-

measures analysis (SAS, v 9.1). Data from each of the three feeder springs and each of the three

river sites were treated as replicates of location. Because mortality is one of the factors that

influences grazing effects, all data were included within the model, even if snails had expired.

All data were log10-transformed to satisfy the assumptions of normality and improve

heteroscedascity prior to analysis using the Shapiro-Wilk (W > 0.92) and Brown-Forsythe tests.

Time zero measurements of Chl a concentrations were used as a baseline and because mixed

model analysis can produce test statistics that are biased upwards and standard errors that are

biased downward, a Kenward-Roger (KR) correction factor was applied to data when

appropriate to limit the possibility of type I error (Littell et al., 2002). Within the model, log Chl

a was the response variable, location (river, spring) and treatment (control, snail) were the

between subject-terms and week (1-4) was the within-subject term. The subjects were individual

stands within locations. There were three interaction terms, treatment*week, treatment*location,

and treatment*week*location. Differences of least squares means were used to investigate

significant interaction terms.

Periphyton accumulation and grazing rate. Periphyton accumulation rate (µg Chl

a/cm2/day) and grazing rate (µg Chl a/cm2/day) were estimated using a least squares regression

21

method relating average weekly estimates of Chl a obtained by averaging Chl a concentrations

across replicate sites within the main river and feeder springs to the Chl a concentration for each

week in both treatment and control dishes. The slope of the linear equation describing this

relationship was the weekly rate. Values were subsequently converted to daily rates to facilitate

comparison with other reported work.

22

Table 2-1. Transect locations in the main river and feeder springs from 2007 transect sampling. Coordinates from the main river correspond to those previously established by Kurz et al. (2003 and 2004). Transect 1 within the main river was located 45.7 m from the river’s headspring. Within the individual feeder springs, transect 1 was located just below the spring vent, while transects 2 and 3 were taken in the middle and end of the respective spring runs.

Transect Location Latitude Longitude

1 Main River 29°58.986' 82°45.6426' 2 Main River 29°58.9242' 82°45.6162' 3 Main River 29°58.8474' 82°45.5568' 4 Main River 29°58.7904' 82°45.525' 5 Main River 29°58.695' 82°45.525' 6 Main River 29°58.608' 82°45.5316' 7 Main River 29°58.5246' 82°45.537' 8 Main River 29°58.4472' 82°45.5628' 9 Main River 29°58.3368' 82°45.5994'

10 Main River 29°58.269' 82°45.5988' 11 Main River 29°58.1952' 82°45.609' 12 Main River 29°58.131' 82°45.6516' 13 Main River 29°58.062' 82°45.6906' 14 Main River 29°58.0182' 82°45.7038' 1 Singing Spring 29°58.566' 82°45.495' 2 Singing Spring 29°58.559' 82°45.505' 3 Singing Spring 29°58.552' 82°45.510' 1 Devil's Eye 29°58.417' 82°45.602' 2 Devil's Eye 29°58.417' 82°45.591' 3 Devil's Eye 29°58.413' 82°45.583' 1 Mill Pond 29°57.995' 82°45.600' 2 Mill Pond 29°57.991' 82°45.641' 3 Mill Pond 29°57.999' 82°45.661'

23

Figure 2-1. Map of the Ichetucknee Springs State Park (ISSP) in Ft. White, Florida (Department of Environmental Protection: Florida Park Service, 2005).

24

Figure 2-2. Sampling device (collapsed form) used to sample submersed aquatic vegetation and associated gastropods. The movable flap of Nitex mesh (see text for deployment details) is held back by the blue bungee cord.

25

0

20

40

60

80

100

120

140

0.0 - 5.0 5.1 - 10.0 10.1 - 15.0 15.1 - 20.0 20.1 - 25.0 25.1 - 30.0

Total Length (mm)

Num

ber

of In

divi

dual

s

Figure 2-3. Size frequency distribution of Elimia floridensis collected during transect sampling.

26

CHAPTER 3 RESULTS

Environmental Parameters

Transect sampling in the main stem of the Ichetucknee River during May 2007 indicated a

general downstream increase in DO concentrations similar to the pattern previously reported

(Kurz et al. 2003 & 2004) (Figure 3-1). In 2007, mean transect values ranged from 3.24 to 7.39

mg l-1, with an overall mean value of 5.18 mg l-1. Dissolved oxygen concentrations in the feeder

springs discharging to the river were significantly lower (ANOVA, F=74.88, p < 0.0001) and

ranged between 0.52 and 2.23 mg l-1 (Figure 3-2); the lowest values were recorded near the point

of spring discharge and highest values closest to the river. Other environmental parameters for

both the main river and feeder springs are provided in Appendices 1 & 2. Of these, only mean

current velocity varied significantly between the main river and the smaller spring runs

(ANOVA, F = 24.10 p < 0.0001). Mean current velocity in the main river were generally > 0.1 m

s-1, while mean velocity in the feeder springs were generally < 0.05 m s-1 (Table A-1).

Regression analysis showed that DO concentration varied significantly with current velocity

(ANOVA, F = 19.81, p < 0.0002), with higher DO concentrations occurring at higher velocities

(Figure 3-3).

During the experimental manipulation, carried out one month after the transect sampling,

DO concentrations and current velocities were lowest (2.75 – 4.98 mg l-1 and 0.15 - 0.20 m s-1,

respectively) in the main river at the upstream locations adjacent to Singing Spring and highest

(5.87 – 9.46 mg l-1 and 0.17 - 0.54 m s-1, respectively) at the downstream locations adjacent to

Mill Pond Spring (Table 3-1). As expected, DO concentrations and current velocities were both

lower in the feeder springs than in the main stem of the river. Lowest DO concentrations (0.36 –

0.67 mg l-1) were consistently recorded in Mill Pond Spring. Dissolved oxygen concentrations in

27

Singing Spring and Devil’s Eye Spring ranged between 0.75 - 0.92 mg l-1 and 0.98 -1.25 mg l-1,

respectively. Current velocities tended to be lower in Mill Pond Spring and Devil’s Eye (0.01 –

0.05 m s-1 and 0.01 – 0.06 m s-1, respectively) than in Singing Spring (0.03 – 0.09 m s-1). In each

of these feeder springs, there was a general decline in velocity during the 4-week investigation

(see Table 3-1).

Abundance and Distribution of Gastropods Including Elimia floridensis

In total, seven gastropod species were collected from the Ichetucknee River and the three

aforementioned spring runs (Table 3-2). Elimia floridensis was numerically dominant and

comprised 85.7% of the individual records. One species of pulmonate gastropod, Planorbella

scalaris, was found in the river, where it represented only 0.3 % of the individual records. In the

feeder springs, P. scalaris was the most frequently recorded taxon, i.e. 81.9% of the individuals

sampled. Elimia floridensis comprised only 9.7% of the snails recorded in the three feeder

springs. Prosobranch snails from the family Hydrobiidae were observed near the vent in Mill

Pond Spring, but because of their small size, genus and species could not be determined in the

field.

The previously reported pattern of a downstream increase in total macroinvertebrates per

sample in the Ichetucknee River (see Kurz et al. 2004) did not hold for E. floridensis, the most

abundant gastropod sampled in 2007 (cf. Figure 3-4 and 3-5). Moreover, the relationship

between E. floridensis abundance and DO concentration in the main stem of the river was not

statistically significant (ANOVA, F = 1.47, p < 0.25). Elimia floridensis was, however, less

abundant in the feeder springs than in the river (ANOVA, F = 8.11, p <0.01) and the mean

number of E. floridensis in each of the feeder springs approached zero near the spring vents

(Figure 3-6) where both DO concentrations and plant abundance was lowest (Appendix 1).

Within all three feeders springs, snail density was positively correlated with distance from the

28

spring vent while in the main river, there was a slight negative correlation between density and

transect (Figures 3-7, 3-8, 3-9, 3-10).

Due to the significant differences in current velocities between the main river and feeder

springs (ANOVA, F = 21.10, p > 0.0001), bivariate fit models were run to determine if snail

abundances varied with current velocity. In neither case was there a statistically significant

relationship between velocity and the abundance of E. floridensis (main river ANOVA, F = 2.14,

p > 0.15; feeder springs ANOVA, F = 0.70, p > 0.40), though there was a qualitative increase in

snail abundance with velocity in the main river (Figure 3-11).

Longitudinal variation in gastropod abundance and distribution stemming from variation

between quadrat samples within transects was not examined in detail. It should be noted

however, that there was no significant relationship between either macrophyte surface area (cm2)

per quadrat and snail abundance per quadrat (ANOVA, F = 2.77, p = 0.1031) or quadrat location

(middle of river or near bank) and snail abundance (ANOVA, F = 1.13, p = 0.3285).

Grazing Effects

Snail mortality. Snail mortality in the main river was low, with only two snails expiring in

week 4. In the feeder springs, snail mortalities were higher than in the main river. In Mill Pond,

all 12 snails died in week 1. In Devil’s Eye, one treatment stand was destroyed in week 3 of the

experiment, therefore two snails were lost. Five of the remaining 10 snails died, 4 within week 1

and 1 in week 4. Only 2 of 12 snails died in Singing Spring, 1 in week 2 and 1 during week 4.

Effect of grazers on periphton accumulation. Within the overall mixed model analysis,

location (ANOVA, F = 99.32, p < 0.0001), treatment (ANOVA, F = 6.39, p = 0.016), week

(ANOVA, F = 12.61, p < 0.0001), and the location*treatment interaction (ANOVA, F = 5.96, p

=0.022) were significant. Analysis of the location*treatment interaction using differences of least

squares means indicated that while the rate of accrual of algae on the Petri dishes (as indicated

29

by the change in Chl a) increased with time in both snail and control dishes within the main

river, accrual was significantly greater (Table 3-3) in the absence of snails (Figure 3-12). Within

the feeder springs, there was no significant effect of snail treatment (Table 3-3) and Chl a in both

treatment and control dishes increased significantly over time (Figure 3-13). In control dishes

alone, periphyton accumulation over time within the feeder springs was significantly greater than

within the main river (Table 3-3), with respect to grazing between the two locations, snails

within the river grazed substantially more periphyton over time than snails within the feeder

springs (Table 3-3) where there were no significant treatment effects.

Rates of periphyton accumulation and grazing. Within the main river, the rate of

periphyton accrual in control dishes was almost twice that of those dishes containing snails, i.e.

0.08 µg Chl a day-1 and 0.05 µg Chl a day-1, respectively (Figure 3-14). The difference of these

rates yields a grazing rate of 0.03 µg Chl a day-1. Periphyton accrual rate in the feeder springs

was estimated to be 0.21 µg Chl a day-1. This rate is substantially higher (by a factor of ~ 2-4)

than the accumulation rates in the river. Grazing rates were not calculated for the feeder springs

due to either high snail mortality or a non-significant treatment effect.

30

Table 3-1. Mean values for environmental parameters in three feeder springs and the main stem of the Ichetucknee River during the caging experiment conducted in April, 2007.

Section Location Week Depth (m) Temp (°C)

DO (mg/L) pH

Flow (m/s)

Spring Singing Spring 0 1.00 21.77 0.75 7.77 0.09 1 0.90 21.71 0.96 7.70 0.03 2 1.00 21.74 0.84 7.83 0.05 3 0.80 21.81 0.90 7.60 0.03 4 0.90 21.76 0.92 7.77 0.06 Devil's Eye 0 1.30 21.97 1.21 7.60 0.03 1 1.20 21.77 1.08 7.70 0.05 2 1.40 21.86 1.25 7.67 0.06 3 1.40 21.93 0.93 7.52 0.01 4 1.30 21.84 0.98 7.69 0.02 Mill Pond 0 1.20 21.92 0.36 7.49 0.04 1 1.10 21.87 0.67 7.58 0.05 2 1.20 21.87 0.52 7.80 0.01 3 1.20 21.93 0.57 7.49 0.01 4 1.00 21.89 0.53 7.76 0.02 Grand Mean 1.13 21.84 0.83 7.66 0.04 River Singing Spring 0 1.10 21.95 4.98 7.83 0.02 1 1.10 21.75 3.26 7.68 0.20 2 1.20 21.67 4.38 7.78 0.15 3 1.50 21.71 2.75 7.67 0.17 4 1.00 21.77 4.15 7.76 0.13 Devil's Eye 0 2.20 22.24 5.96 7.73 0.11 1 2.40 21.32 4.79 7.75 0.14 2 2.70 22.18 6.25 7.88 0.19 3 2.50 22.65 6.18 7.77 0.16 4 2.60 21.96 5.24 7.87 0.31 Mill Pond 0 2.10 23.33 9.46 8.00 0.17 1 1.80 21.11 5.87 7.86 0.20 2 1.50 22.34 8.30 8.12 0.52 3 2.40 23.48 8.99 8.05 0.10 4 1.20 22.46 7.96 8.00 0.54 Grand Mean 1.82 22.13 5.90 7.85 0.21

31

32

Table 3-2. Listing of gastropod species, total number of individuals surveyed for each species and percent of total abundance determined for each species within the Ichetucknee River and three feeder springs.

Location Species Total Number of Individuals % of Total Abundance River Elimia floridensis 251 85.7

Pomacea paludosa 8 2.7 Lioplax pilsbryi 2 0.7 Notogillia wetherbyi 29 9.9 Hydrobiidae 0 0.0 Planorbella scalaris 1 0.3 Tarebia granifera 2 0.7

Spring Elimia floridensis 27 9.7 Pomacea paludosa 0 0.0 Lioplax pilsbryi 0 0.0 Notogillia wetherbyi 10 3.6 Hydrobiidae 13 4.7 Planorbella scalaris 227 81.9 Tarebia granifera 0 0.0

Table 3-3. Differences of least squares means investigating the significant location*treatment interaction from mixed model analysis of Elimia floridensis grazing effects on periphyton accrual. C = control dishes containing no snails, S = treatment dishes containing one E. floridensis.

Effect Location Treatment Location Treatment Pr > |t| Location*Treatment River C River S 0.0014 Location*Treatment River C Spring C <0.0001 Location*Treatment River S Spring S <0.0001 Location*Treatment Spring C Spring S 0.9263

33

0

1

2

3

4

5

6

7

8

9

0 2 4 6 8 10 12 14 16

Transect

Mea

n D

O (m

g l-1

)

200320042007

Figure 3-1. Mean dissolved oxygen concentration (mg l-1) by transect. Data for 2003 and 2004

are from Kurz et al. (2003 and 2004) and re-plotted her for comparative purposes.

34

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 1 2 3 4

Transect

Mea

n D

O (m

g l-1

)

Singing SpringDevil's EyeMill Pond

Figure 3-2. Mean dissolved oxygen concentration (mg l-1 ± SD) by transect in the three feeder

springs (2007). Transect 1 is located below the spring vent while transects 2 and 3 are in the middle and end of the feeder spring runs, respectively.

35

y = 24.878x + 1.0236R2 = 0.7793

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Mean Current Velocity (m s-1)

Mea

n D

O (m

g l-1

)

Figure 3-3. Mean current velocity (m s-1) vs. mean dissolved oxygen concentration (mg l-1). Data

from both the main river and feeder springs surveyed in 2007.

36

0

2000

4000

6000

8000

10000

12000

0 2 4 6 8 10 12 14 1

Transect

Mea

n #

Indi

vidu

als

per 0

.25

m2

6

Figure 3-4. Macroinvertebrate abundance by transect re-plotted from Kurz et al. (2003). Data

presented are for midr-iver stations only (n =1). Transects start upstream in the Headsprings Reach and continue downstream into the Floodplain Reach (see text for additional details related to river description and sampling locations).

37

0.0

20.0

40.0

60.0

0 2 4 6 8 10 12 14 16

Transect

Mea

n E.

flor

iden

sis

Abu

ndan

ce/0

.062

5 m

2

20072003

Figure 3-5. Mean Elimia floridensis abundance by transect in 2003 (Kurz et al. 2003) and 2007;

N = 3 for each data point. Transects start upstream in the Headsprings Reach and continue downstream into the Floodplain Reach (see text for additional details related to river description and sampling locations).

38

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

0 1 2 3 4

Transect

Mea

n E.

flor

iden

sis

Abu

ndan

ce p

er 0

.062

5 m

2

Singing SpringDevil's EyeMill Pond

Figure 3-6. Mean Elimia floridensis abundance (± SD) by transect in feeder springs in from

2007; N = 3 for each data point. Transect 1 is located below the spring vent while transects 2 and 3 lie in the middle and end of the spring run, just before the confluence of the run and river, respectively.

39

Figure 3-7. Density Ellipse for Singing Spring. Transect 1 lies closest to the spring vent while transect 2 and 3 lie in the middle and end of the spring run, respectively. Ellipse encircles the densest 50% of the estimated distribution. Correlation analysis reveals a 49% correlation between the number of individual snails per quadrat (0.0625m2) and distance from the spring vent.

40

Figure 3-8. Density Ellipse for Devil’s Eye. Transect 1 lies closest to the spring vent while transect 2 and 3 lie in the middle and end of the spring run, respectively. Ellipse encircles the densest 50% of the estimated distribution. Correlation analysis reveals a 57% correlation between the number of individual snails per quadrat (0.0625m2) and distance from the spring vent.

41

Figure 3-9. Density Ellipse for Mill Pond. Transect 1 lies closest to the spring vent while transect 2 and 3 lie in the middle and end of the spring run, respectively. Ellipse encircles the densest 50% of the estimated distribution. Correlation analysis reveals a 25% correlation between the number of individual snails per quadrat (0.0625m2) and distance from the spring vent.

42

Figure 3-10. Density Ellipse for the main river. Transect 1 was located near the headspring. Ellipse encircles the densest 50% of the estimated distribution. Correlation analysis reveals a -6% correlation between the number of individual snails per quadrat (0.0625m2) and distance from the downstream.

43

y = 42.396x + 0.795R2 = 0.2365

0.0

5.0

10.0

15.0

20.0

25.0

30.0

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

Current Velocity (m/s)

Mea

n E.

flor

iden

sis

abun

danc

e pe

r 0.0

625

m2

Figure 3-11. Bivariate fit of mean of Elimia floridensis abundance by flow (m s-1) from 2007

transect sampling in both the main river and feeder springs.

44

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 5 10 15 20 25 30

Day

ug C

hl a

/cm

2

ControlTreatment

Figure 3-12. Peripyton accumulation (µg Chl a/cm2, ± SD) in control dishes and snail treatment

dishes within the main river. Estimates of Chl a obtained by averaging weekly Chl a concentrations across the three replicate sites within the river.

45

0

2

4

6

8

10

12

0 5 10 15 20 25 30

Day

ug C

hl a

/cm

2

ControlTreatment

Figure 3-13. Weekly Estimates of chlorophyll a concentration (ug Chl a/cm2, ± SD) from control

and snail dishes in the feeder springs. Estimates of Chl a obtained by averaging weekly Chl a concentrations across replicate sites within the three feeder springs.

46

y = 0.0757x + 0.9707

y = 0.047x + 0.8417

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 5 10 15 20 25 30

Week

ug C

hl a

/cm

2 Control

Treatment

Figure 3-14. Periphyton accumulation rates (µg Chl a/cm2 ± SD) in control dishes and snail

treatment dishes within the main river. Estimate of grazing rate obtained by subtracting the slope of the treatment line from the slope of the control line.

47

CHAPTER 4 DISCUSSION

Although previous data collected in the main stem of the Ichetucknee River implicate

dissolved oxygen as an important determinant of macroinvertebrate abundance (Kurz et al.

2003), data collected in 2007 and reported herein suggest that this is not the case for E.

floridensis. I found no evidence of a downstream gradient in E. floridensis abundance and no

statistically significant relationship with dissolved oxygen within the main river. There was,

however, a marked reduction in snail abundance within the feeder springs which are chronically

hypoxic and it is noteworthy that snail density was positively correlated with DO concentration.

In fact, no live E. floridensis were found in the immediate vicinity of the spring vents in two of

the three feeder springs and only one live snail was recorded at the other. These observations are

consistent with my hypothesis that low dissolved oxygen concentrations influence the broader

distributional pattern of E. floridensis in this system as a whole. It is likely that snail abundances

are also reduced in the uppermost reach of the river in close proximity to the Ichetucknee Spring

which serves as the river’s point of origin. Access to this portion of the river was restricted

during this study and thus their abundances in this area could not be confirmed.

Results from the caging experiments illustrate more definitively the ecological

consequences of hypoxic conditions on E. floridensis. Snails caged in the feeder springs suffered

high mortality. In these springs, DO concentrations were extremely low, with concentrations as

low as 0.36 mg l-1. In the main river, where mean oxygen concentrations were almost six fold

higher, only two snails died and only after a four week exposure period. Although there are

concerns with any study in which enclosures are employed (Walde & Davies, 1984; Quinn &

Keough, 1993; MacNally, 1996), snail mortality in the main river enclosures was negligible and

suggests that potential caging artifacts were not responsible for the high mortality observed in the

48

feeder springs. It is noteworthy that in Mill Pond Spring, where mean DO concentrations were

lowest, i.e. 0.57 ± 0.11 mg l-1, all snails died within the first week of the experiment. In Devil’s

Eye Spring, mean DO concentrations were the highest of the three springs, however, variation in

concentration was also highest (1.10 ± 0.14 mg l-1), which may account for 50 % snail mortality

seen after four weeks.

In Singing Spring, where only 2 of 12 snails died, observational data are suggestive of a

sub-lethal effect of low DO concentration. While snails in this spring lived, they were rarely

observed attached to the bottoms of the Petri dishes and were not able to significantly decrease

periphyton accrual rates relative to control dishes without snails. Therefore, in the context of this

study, I suggest that chronic exposure to hypoxic conditions led to decreased feeding rates in E.

floridensis at the individual level. This result would almost certainly translate to the population-

level over time, as reductions in food intake would reduce the energy available for egg

production (Brethes et al. 1994; Ito et al. 1996). In addition to reduced food intake, a number of

other potential side effects of low DO have also been reported and are likely to have important

physiological and population-level consequences. Low oxygen availability, for example, can

determine clutch size in some species (Lardies & Fernandez, 2002, Fernandez et al. 2006), while

prolonged exposure to hypoxia can reduce development rate in both larval and juvenile

gastropods (Cancino et al. 2003). Reduced clutch size is most likely a result of oxygen

competition between siblings in jelly masses or ovicapsules (Fernandez et al. 2006). In extreme

cases, hypoxic events lasting several days can decimate entire snail populations (Laboy-Nieves

et al. 2001).

Oxygen mediated affects on grazer abundance and feeding rate can have profound

ecological consequences in aquatic systems. Reductions in both of these factors may lead to

49

increases in periphyton load and subsequent macrophyte losses. Within the main river, where

major losses in macrophyte biomass have not been documented, snails grazed at a rate of 0.03 µg

Chl a day-1, which is lower than rates reported for other prosobranch gastropods. For instance,

individual Potamopyrgus antipodarum can graze at rates between 0.09 and 0.37 µg Chl a day-1

(James et al. 2000) while Rounick & Winterbourn (1983), found that P. antipodarum ingested

0.40 µg Chl a day-1. Jujubinus striatus, a marine gastropod often associated with the seagrass

Zostera marina, can graze as much as 300 to 1300 µg organic matter day-1 (Hily et al. 2004).

Even though the grazing rates reported here for E. floridensis are lower than those previously

reported for other gastropods, snails did have a significant effect on Chl a accrual rates in the

main stem of the Ichetucknee River. At high densities, it is possible that E. floridensis may

regulate periphyton accumulation on macrophytes in this system. However, the snail density

used in the caging experiments was assumed to be representative of mean snail densities on

macrophytes in the main river. Thus, it is likely that grazing by E. floridensis in combination

with some other factor(s) is responsible for controlling periphyton growth.

Within the feeder springs, where macrophyte biomass has decreased and filamentous algal

coverage increased (Ringle, 1999; Ritchie, 2006), it is clear that E. floridensis is not having a

regulatory effect of periphyton growth. In Devil’s Eye Spring and Mill Pond Spring, snails

suffered 50 and 100% mortalities, respectively, while in Singing Spring, snails survived but

overall, there was no significant effect of grazing on Chl a accrual. In the absence of grazers,

periphyton accrual rates in the feeder springs were significantly greater than in the main river. If

grazing was the sole determinant of periphyton accumulation, I would have expected no

difference. Clearly, factors other than grazing by E. floridensis allow for a more rapid growth of

periphyton in these feeder springs relative to the river.

50

On of the factors that may be responsible, in part, for differences in periphyton accrual

between the river and spring sites is variation in current velocities. Velocity rates in rivers and

streams can affect algal colonization and accumulation, ecosystem production and other biotic

interactions through scouring and drag processes (Biggs et al. 2005). Velocities in excess of 0.55

m s-1 have been shown to remove filamentous algae from artificial substrates (Ryder et al. 2006),

leaving only crustose epiphytes behind. Current velocities rates in the main stem of the

Ichetucknee River were typically an order of magnitude greater than in the spring runs. While the

epiphyte and algal species which accumulated in the dishes were not characterized

taxonomically, qualitative analysis of algal composition indicated that loosely attached

filamentous algae was found almost exclusively on dishes in the feeder springs, while harder,

crustose algae was more typical on dishes within the river, especially in Devil’s Eye and Singing

Spring runs.

The efficacy of artificial substrates, such as the ones used in the present study, has long

been debated in ecology (Schagerl & Donabaum, 1998; Barbiero, 2000; Fisher & Kelso, 2007;

Nakamura et al. 2007) and it is possible that the community composition of algae that

accumulated on the glass Petri dishes employed in this study was not representative of the algal

communities associated with other natural substrates, macrophytes in particular, in this spring-

fed river system. Natural substrates often exhibit greater species richness than their artificial

counterparts (Cattaneo & Amireault, 1992, Barbiero, 2000) and additional variation may have

occurred in algal and bacterial community composition on the Petri dishes due to herbivory. For

instance, long, filamentous chain forming algal species in the treatment dishes were most likely

more at risk as gastropods tend to prey preferentially on the upper layers of the periphyton

51

community, leaving the more calcareous and crustose algal species behind (Lamberti et al. 1989;

Steinman et al. 1992; Hillebrand et al, 2000).

In spite of the potential artifacts associated with the use of artificial substrates to collect

periphtyon and measure grazing rates, there is a long history of this approach in the literature.

Glass microscope slides, which have been utilized in many studies due to the presumed inertness

of their surface and relative ease of epiphyte removal (Vandijk, 1993; Claret, 1998; Ghosh &

Gaur, 1998; Notestein et al. 2003; Boisson & Perrodin, 2006), are generally accepted as a

suitable substrate for periphyton and biofilm collection (Danilov & Ekelund, 2001; Lane et al.

2003). Politano (Univ. Florida, unpubl. data) measured periphyton accumulation rates on

Sagittaria kurziana, the most dominant macrophyte in the Ichetucknee River. Results from that

study suggest periphyton accumulation rates during a comparable time period, i.e. 4 weeks, were

similar to values reported here. In the main river, the average Chl a content on the surface of

control Petri dishes (standardized to surface area) after 4 weeks of exposure was 3.1 ± 2.1 µg cm-

2, while the average Chl a content on macrophytes at the same sample locations was 5.3 ± 2.8 µg

cm-2. In the feeder springs, Chl a content on Petri dishes and macrophytes averaged 6.68 ± 3.4

µg cm-2 and 25.4 ± 38.3 µg cm-2, respectively. These comparisons suggest that periphyton

accumulation on Petri dishes within the river adequately represented the rate of periphyton

accrual on natural substrates, while in the feeder springs, rates of periphyton accrual on dishes

were likely an underestimate.

While this study focused primarily on periphyton accrual rates and oxygen mediated

grazing impacts by E. floridensis, it should be noted that there are many other grazer species

within the river whose spatial patterns of abundance and distribution may be structured by

hypoxia. For instance, four other species of prosobranch gastropods are known to occur in the

52

main stem of the river. Only one of those species was recorded as live, however, in any of the

feeder springs (Hydrobiidae sp.). In addition to gastropods, other grazer distributions within the

Ichetucknee may be influenced by lowDO concentrations. For example, hypoxic conditions can

result in changes in chironomid assemblages (Quinlan et al. 1998, Little & Smol, 2001; Quinlan

& Smol, 2002) which, along with gastropods, are some of the most abundant invertebrate grazers

within the river (Kurz et al. 2003, G. Warren pers. comm.). In addition to changes in invertebrate

distribution, fish distribution and relative abundance may also be influenced by hypoxia. In fact,

in the area surrounding Singing Spring the fish community is typically characterized by species

considered to be low oxygen tolerant (e.g. Gambusia holbrooki, Heterandria formosa, and

Notropis harperi). The diversity of fishes in this area only increases with distance from the

spring vent (Mckinsey & Chapman, 1998). This differential susceptibility to hypoxia in grazer

species may leave habitat niches unused, allowing colonization by invasive species (Fox & Fox,

1986; Jewett et al. 2005).

As part of this study, other parameters in addition to DO concentration, such as water

chemistry and human disturbance, were not examined. Like DO concentration, these parameters

may influence snail distribution and abundance within the river and warrant further investigation.

For example, it is noteworthy that there is a gradient of sulfate between the three springs studied.

Concentrations are highest in Mill Pond (35.5 mg l-1) and subsequently decrease is Devil’s Eye

(19.6 mg l-1) and Singing Spring (10.2 mg l-1) (SRWMD, 2007). Hydrogen sulfide, the reduced

form of sulfate, can influence snail distribution and abundance and at concentrations of 3.4 mg l-1

and above can be toxic to freshwater gastopods (Eden et al., 2003, Heller & Ehrlich, 1995). At

this time however, hydrogen sulfide concentrations within the Ichetucknee have not been

measured. With respect to human disturbance, it is almost certain that the results presented

53

herein were in some way influenced by this factor. Following years of substantial vegetation loss

in the main river caused by recreational park visitors, maximum daily loads, or carrying

capacities, were established in 1979 (DuToit, 1979). However, while the macrophytes within the

river recovered, the park is still visited by 3,000 visitors annually and it is possible that their

presence influences gastropod distribution, especially in Singing Spring and Devil’s Eye where

access is not restricted.

Due to the large number of springs in Florida, the findings presented herein have several

important implications for management. The potential interactions between oxygen availability,

periphyton accrual and grazing merit further investigation. In addition, because the rate of

periphyton accumulation was likely underestimated in the springs, and the lack of grazing

pressure alone cannot account for the periphyton accumulation seen within the springs, current

velocity should be considered as well. This environmental variable appears to co-vary with

dissolved oxygen concentration and there exists also a qualitative positive relationship between

current velocity and density of E. floridensis. High current velocities in sections of the main river

removes flocculent material, leaving denser, sandier sediment and limestone outcrops behind.

These microhabitats may allow for increased survival or growth relative to areas of high floc

build up and therefore represent preferred gastropod habitat (Olabarria et al. 2002). In these

preferred areas, grazing pressure, in combination with current velocity, may exert control over

periphyton growth.

In a broader sense, managers working within any riverine system should not forget that

changes in physical parameters are just as influential in producing ecosystem change as nutrient

load. For example, in the Weeki Wachee River, a highly spring-influenced system in Hernando

County, mean total nitrogen and phosphorous concentrations within the system have increased

54

55

from 1998-2000 to 2003-2005 (from 500 to 700 µg N l-1; 24 to 27 µg P l-1) (Frazer et al., 2006).

However, mean stream velocity within the river also increased from an average of 0.23 m/s to

0.28 m/s and in conjunction with this, instead of a nutrient stimulated increase in periphyton

load, there was a decrease in mean periphyton abundance from 0.078 to 0.035 mg Chl a g wet

weight host macrophyte-1 (Frazer et al., 2006). In the Homosassa River, mean total nitrogen and

phosphours also increased from 1998-2000 to 2000-2005 (from 425 to 500 µg N l-1; 24 to 26 µg

P l-1). However, mean stream velocity in this system is much lower (0.08 m s-1 from 2003 to

2005) than in the Weeki Wachee and Ichetucknee rivers and, in this case, periphyton loads

increased from 0.029 to 0.165 mg Chl a g wet weight host macrophyte-1. These examples

demonstrate how chemical and physical forces might act in concert to affect periphyton

abundance. This fact should be considered by managers whenever changes such as reductions in

nutrient load, changes in minimum flows and levels, and diversions of riverine waters are

proposed.

APPENDIX ICHETUCKNEE MEAN ENVIRONMENTAL PARAMETERS

56

Table A-1. Ichetucknee River mean environmental parameters for transect data collected in May 2003 and April 2004 (Kurz et al. 2003 and 2004). In 2003, N = 5 for depth, flow and canopy and N = 3 for each of the other parameters. In 2004, N = 3 for depth, flow and canopy and N ≥ 3 for other parameters.

Year Transect Depth

(m) Temp (°C)

DO (mg/L) pH

Flow (m/s) Conduct (uS/cm) # Snail/Transect SAV Surface Area (cm2)

2003 1 0.98 21.9 4.3 7.46 0.11 320 0.49 . 2 0.98 21.8 4.5 7.46 0.18 320 . . 3 0.98 21.8 4.7 7.57 0.22 320 . . 4 1.18 21.9 3.7 7.27 0.26 312 . . 5 1.00 21.9 4.1 7.45 0.23 311 0.58 . 6 0.82 22.1 5.0 7.49 0.20 309 0.58 . 7 1.12 22.2 4.9 7.50 0.23 314 0.71 . 8 1.00 22.3 5.2 7.59 0.24 315 . . 9 1.00 22.5 5.2 7.67 0.15 323 . . 10 1.04 22.6 6.2 7.61 0.17 322 . . 11 1.03 23.0 6.7 7.75 0.13 321 . . 12 1.44 23.2 7.4 7.70 0.13 322 . . 13 1.20 23.3 7.6 7.79 0.14 321 . . 14 1.16 23.5 7.8 7.84 0.15 321 . .

2004 1 1.40 21.8 4.2 7.53 0.09 309 1.57 . 2 1.00 21.9 4.6 7.60 0.09 309 1.9 . 3 1.00 22.0 5.0 7.58 0.12 309 1.69 . 4 1.70 21.9 4.5 7.64 0.12 299 . . 5 1.60 22.0 4.3 7.63 0.20 297 0.58 . 6 1.20 22.1 5.2 7.64 0.09 296 0.42 . 7 1.40 22.2 3.7 7.64 0.15 297 0.63 . 8 1.60 22.4 5.2 7.72 0.18 297 0.4 . 9 1.40 22.4 5.2 7.73 0.10 303 0.47 . 10 1.90 22.6 6.1 7.80 0.32 299 0.52 . 11 1.70 22.7 6.3 7.83 0.22 299 0.58 .

57

Table A-1 Continued.

Year Transect Depth

(m) Temp (°C)

DO (mg/L) pH

Flow (m/s) Conduct (uS/cm) # Snail/Transect SAV Surface Area (cm2)

2004 12 1.60 22.8 7.1 7.85 0.21 300 0.39 . 13 1.30 22.8 6.9 7.89 0.24 300 1.38 . 14 1.30 22.9 7.4 7.87 0.10 300 0.72 .

2007 1 1.06 21.8 4.3 7.49 0.04 318 4.00 2493 2 1.10 21.9 4.8 7.45 0.10 317 1.00 4347 3 1.10 21.9 5.3 7.46 0.10 317 4.00 4532 4 0.93 21.5 4.2 7.70 0.15 291 22.33 2245 5 1.43 21.5 4.3 7.72 0.08 288 5.67 8199 6 1.13 21.5 5.4 7.78 0.12 287 5.67 6736 7 1.70 21.7 3.2 7.71 0.14 271 5.00 7519 8 1.47 21.9 4.8 7.87 0.29 285 2.33 15223 9 1.63 22.0 4.4 7.80 0.11 289 13.00 14585 10 2.30 21.3 5.1 7.88 0.30 285 18.67 2534 11 1.53 21.4 6.0 8.02 0.06 284 1.00 10897 12 1.60 21.8 6.9 8.08 0.09 285 2.67 11330 13 1.33 20.8 6.4 7.97 0.13 284 4.33 10023 14 1.47 21.3 7.4 8.14 0.16 284 3.67 9248

58

59

Table A-1 Continued.

Year Transect Depth

(m) Temp (°C)

DO (mg/L) pH

Flow (m/s) Conduct (uS/cm) # Snail/Transect SAV Surface Area (cm2)

2007 SS-1 0.87 21.8 1.0 7.69 0.02 282 4.000 412 SS-2 0.83 21.8 1.8 7.70 0.04 282 3.667 3115 SS-3 1.27 21.8 1.9 7.74 0.06 282 10.000 2946 DE-1 1.17 21.8 0.9 7.66 0.03 297 3.333 845 DE-2 1.17 21.8 1.0 7.63 0.04 297 1.667 1054 DE-3 1.13 21.9 1.3 7.64 0.06 296 54.333 3667 MP-1 0.83 21.9 0.5 7.77 0.04 324 2.667 906 MP-2 0.50 21.9 1.4 7.86 0.08 323 3.667 2596 MP-3 0.63 21.9 2.2 7.84 0.03 323 6.000 1224

REFERENCE LIST

Alvarez M. & Pardo I. (2007) Factors controlling epilithion biomass in a temporary, karst stream: the interaction between substratum and grazing. Journal of the North American Benthological Society, 26, 207-220.

APHA (American Public Health Association). (1995) Standard methods for the examination of

water and wastewater. 19th edition. American Public Health Association, American Water Works Association, and Water Pollution Control Federation, Washington, DC.

Barbiero R.P. (2000) A multi-lake comparison of epilithic diatom communities on natural and artificial substrates. Hydrobiologia, 438, 157-170.

Bergquist D. C., Fleckenstein C., Knisel J., Begley B., Macdonald I. R. & Fisher C. R. (2005) Variations in seep mussel bed communities along physical and chemical environmental gradients. Marine Ecology-Progress Series, 293, 99-108.

Biggs B.J.F., Francoeur S.N., Huryn A.D., Young R., Arbuckle C.J. & Townsend C.R. (2000) Trophic cascades on streams: effects of nutrients enrichment on autotrophic and consumer benthic communities under two different fish predator regimes. Canadian Journal of Fisheries and Aquatic Sciences, 57, 1380-1394.

Biggs B.J.F., Nikora V.I. and Snelder T.H. (2005) Linking scales of flow variability to lotic ecosystem structure and function. River Research and Applications, 21, 283-298.

Boisson J.C. & Perrodin Y. (2006) Effects of road runoff on biomass and metabolic activity of periphyton in experimental streams. Journal of Hazardous Materials, 132, 148-154.

Brethes J.C., Ferreyra G. & Delavega S. (1994) Distribution, growth and reproduction of the limpet Nacella-(Patinigera)-concinna (Strebel 1908) in relation to potential food availability, in Esperanza Bay (Antarctic peninsula). Polar Biology, 14, 161-170.

Burkpile D.E. & Hay M.E. (2006) Herbivore vs. nutrient control of marine primary producers: Context-dependent effects. Ecology, 87, 3128-3139.

Cancino J. M., Gallardo J. A. & Torres F. A. (2003) Combined effects of dissolved oxygen concentration and water temperature on embryonic development and larval shell secretion in the marine snail Chorus Giganteus (Gastropoda: Muricidae). Marine Biology, 142, 133-139.

Canfield D.E. & Hoyer M.V. 1998. Surface area of aquatic macrophytes. Aquatics, 26-27.

Cattaneo A. & Amireault M.C. (1992) How artificial are artificial substrata for periphyton. Journal of the North American Benthological Society, 11, 244-256.

Claret C. (1998) Hyporheic biofilm development on artificial substrate, as a tool for assessing trophic status of aquatic systems: First results. International Journal of Limnology, 34, 119-128.

60

Canfield D.E. & Hoyer M.V. 1998. Surface area of aquatic macrophytes. Aquatics, 26-27.

Cummins K.W. & Klug M.J. (1979) Feeding ecology of stream invertebrates. Annual Review of Ecology and Systematics, 10, 147-172.

Danilov R.A. & Ekelund N.G.A. (2001) Comparison of usefulness of three types of artificial substrata (glass, wood and plastic) when studying settlement patterns of periphyton in lakes of different trophic status. Journal of Microbiological Methods, 45, 167-170.

Dazo B.C. (1965) The morphology and natural history of Pleurocera acuta and Goniobasis livescens (Gastropoda: Cerithiacea: Pleuroceridae). Malacologia, 3, 1-80.

Devi V.U. (1997) Heavy metal toxicity to an intertidal gastropod Morula granulate (Duclos): Tolerance to copper, mercury, cadmium and zinc. Journal of Environmental Biology, 18, 287-290.

Duarte C.M. (1995) Submerged aquatic vegetation in relation to different nutrient regimes. Ophelia, 41, 87-112.

DuToit C.H. (1979) The carrying capacity of the Ichetucknee Springs and River. University of Florida, MS Thesis. 176 pp.

Eden N., Katz T. & Angel D. L. (2003) Dynamic response of a mud snail Nassarius sinusigerus to changes in sediment biogeochemistry. Marine Ecology-Progress Series, 263, 139-147.

FDEP, (2000) Ichetucknee Springs State Park approved unit management plan. Florida Department of Environmental Protection, Division of Recreation and Parks. 54 pp.

Feminella J.W., Power M.E. & Resh V.H. (1989) Periphyton responses to invertebrate grazing and riparian canopy in three northern California coastal streams. Freshwater Biology, 22, 445-457.

Feminella J.W. & Hawkins C.P. (1995) Interactions between stream herbivores and periphyton: A quantitative analysis of past experiments. Journal of the North American Benthological Society, 14, 465-509.

Fernandez M., Vallet D. & Castells P. (2006) The effects of temperature and oxygen availability on intracapsular development of Acanthina mondon (Gastropoda: Muricidae). Revista Chilena de Historia Natural, 79, 155-167.

Fisher J.C. & Kelso W.E. (2007) Effectiveness of artificial plants in subsurface exclosures as a substrate for hydrilla-dwelling macro invertebrate communities. Journal of Freshwater Ecology, 22, 33-39.

Fox M.J. & Fox B.J. (1986) Susceptibility of natural communities to invasion. In: Groves RH, Buron JJ (editors) Ecology of biological invasions. Cambridge University Press, Cambridge, pp. 57-66.

61

Frazer T.K., Notestein S.K. & Pine W.E. Jr. (2006) Changes in the physical, chemical and vegetative characteristics of the Homosassa, Chassahowitzka and Weeki Wachee Rivers. . Final Report. Suwannee River Water Management District, Live Oak, Florida.

Gafner K. & Robinson C.T. (2007) Nutrient enrichment influences the response of stream macroinvertebrates to disturbance. Journal of the North American Benthological Society, 26, 92-102.

Gay D. & Maher W. (2003) Natural variation of copper, zinc, cadmium and selenium concentrations in Bembicium nanum and their potential use as a biomonitor of trace metals. Water Research, 37, 2173-2185.

Ghosh M. & Gaur J.P. (1998) Current velocity and the establishment of stream algal periphyton communities. Aquatic Botany, 60, 1-10.

Gray J.S., Wu R.S.S. & Or Y.Y. (2002) Effects of hypoxia and organic enrichment on the coastal marine environment. Marine Ecology Progress Series, 238, 249-279.

Gulis V. & Suberkropp K. (2004) Effects of whole-stream nutrient enrichment on the concentration and abundance of aquatic hyphomycete conidia in transport. Mycologia, 96, 57-65.

Greenwood J.L. & Rosemond A.D. (2005) Periphyton response to long-term nutrient enrichment in a shaded headwater stream. Canadian Journal of Fisheries and Aquatic Sciences, 62, 2033-2045.

Gregory S.V. (1980) Effects of light, nutrients, and grazing on periphyton communities in streams. Dissertation Oregon State University, Corvallis, Oregon, USA.

Hargrave C.W. (2006) A test of three alternative pathways for consumer regulation of primary productivity. Oecologia, 149, 123-132.

Heller J. & Ehrlich S. (1995) A fresh-water prosobranch, Melanoides tuberculata, in a hydrogen-sulfide stream. Journal of Conchology, 35, 237-241.

Hill W.R. & Knight A.W. (1988) Nutrient and light limitation of algae in two northern California streams. Journal of Phycology, 24, 125-132.

Hill W.R., Boston H.L. & Steinman A.D. (1992) Grazers and nutrients simultaneously limit lotic primary productivity. Canadian Journal of Fisheries and Aquatic Sciences, 47, 2307-2314.

Hill W.R., Ryon M.G. and Schilling E.M. (1995) Light limitation in a stream ecosystem: Responses by primary producers and consumers. Ecology, 76, 1297-1309.

62

Hillebrand H., Worm B. & Lotze H.K. (2000) Marine microbenthic community structure regulated by nitrogen loading and grazing pressure. Marine Ecology Progress Series, 204, 27-38.

Hillebrand H. (2002) Top-down versus bottom-up control of autotrophic biomass – a meta-analysis on experiments with periphyton. Journal of the North American Bethological Society, 21, 349-369.

Hilly C., Conan S., Raffin C. and Wylie-Echeverria S. (2004) In vitro experimental assessment of the grazing pressure of two gastropods on Zostera marina L. ephiphytic algae. Aquatic Botany, 78, 183-195.

Huryn A.D., Koebel J.W. and Benke A.C. (1994) Life history and longevity of the Pleurocerid snail Elimia: A comparative study of eight populations. Journal of the North American Benthological Society, 13, 540-556.

Ito K., Goshima S. & Nakao S. (1996) Growth and reproduction of the generalist opisthobranch Halo japonica: Effect of algal seasonality on growth rate. Marine Biology, 126, 395-401.

James M.R., Hawes I. & Weatherhead M. (2000) Removal of settled sediments and periphyton from macrophytes by grazing invertebrates in the littoral zone of a large oligotrophic lake. Applied Issues, 44, 322-326.

Jeppesen E., Kristensen P., Jensen J.P., Søndergaard M., Mortensen E. and Lauridsen T. (1991) Recovery resilience following a reduction in external phosphorus loading of shallow, eutrophic Danish lakes: duration, regulating factors and methods for overcoming resilience. In Giussani G., Van Liere L.V. & Moss B. (editors). Ecosystems Research in Freshwater Environment Recovery. Memorie dell’Istituto italiano de idrobiologia dott. Marco de Marchi. 48, 127-148.

Jewett E.B., Hines A.H. and Ruiz G.M. (2005) Epifaunal disturbance by periodic low levels of dissolved oxygen: native vs. invasive species response. Marine Ecology Progress Series, 304, 31-44.

Jones G.W., Upchurch S.B., Champion K.M. & Dewitt D.J. (1997) Water quality and hydrology of the Homosassa, Chassahowitzka, Weeki Wachee and Aripeka spring complexes, Citrus and Hernando Counties Florida (Origin of increasing nitrate concentrations). Ambient Ground-Water Quality Monitoring Program. Southwest Florida Water Management District. Brooksville, FL. 167 pp.

Kurz R.C., D.C. Woithe, S.K. Notestein, T.K. Frazer, J.A. Hale and S.R. Keller (2004) Mapping and monitoring submerged aquatic vegetation in Ichetucknee Springs – 2004. Final Report. Suwannee River Water Management District, Live Oak, Florida.

Kurz R.C., P. Sinphay, W.E. Hershfeld, A.B. Krebs, A.T. Perry, D.C. Woithe, S.K. Notestein,

T.K. Frazer, J.A. Hale and S.R. Keller (2003) Mapping and monitoring submerged aquatic vegetation in the Ichetucknee and Manatee Springs. Final Report. Suwannee River Water Management District, Live Oak, Florida.

63

Laboy-Nieves E.N., Klein E., Condoe J.E., Losada F., Cruz J.J. & Bone D. (2001) Mass mortality of tropical marine communities in Morrocoy, Venezuela. Bulletin of marine Science, 68, 163-179.

Lamberti G.A., Gregory S.V., Ashkenas L.R., Steinman A.D & McIntire C.D. (1989) Productive capacity of perihyton as a determinant of plant-herbivore interactions in streams. Ecology, 70, 1840-1856.

Lane C.M., Taffs K.H. & Corfield J.L. (2003) A comparison of diatom community structure on natural and artificial substrata. Hydrobiologia, 493, 65-79.

Larides M.A. & Fernandez M. (2002) Effect of oxygen availability in determining clutch size in Acanthina monodon. Marine Ecology Progress Series, 239, 139-146.

Levin L. A. & Gage, J. D. (1998) Relationships between oxygen, organic matter and the diversity of bathyal macrofauna. Deep-Sea Research Part Ii-Topical Studies in Oceanography, 45, 129-163.

Leibold M.A., Chase J.M., Shurin J.B. & Downing A.L. (1997) Species turnover and the regulation of trophic structure. Annual Review of Ecology and Systematics, 28, 467-494.

Littell R.C., Stroup W.W. & Freund R.J. (2002) SAS for linear models, fourth edition. Cary, NC: SAS Institute Inc.

Little J.L. & Smol J.P. (2001) A chironomid-based model for inferring late-summer hypolimnetic oxygen in southeastern Ontario lakes. Journal of Paleolimnology, 26, 259-270.

MacNally R. (1996) Scaling artifacts in confinement experiments: a simulation model. Ecological Modeling, 99, 229-245.

Malard F. & Hervant F. (1999) Oxygen supply and the adaptations of animals in groundwater. Freshwater Biology, 41, 1-30.

Mancini E.R. (1978) The Biology of Goniobasis semicarinata (Say) (Gastropoda: Pleuroceridae) in the Mosquito Creek Drainage System. Ph.D. Dissertation, University of Louisville, Louisville, Kentucky.

Minshall G.W. (1978) Autotrophy in stream ecosystems. BioScience, 28, 776-771.

Montagna P.A. & Ritter C. (2006) Direct and indirect effects of hypoxia on benthos in Corpus Christi Bay, Texas, USA. Journal of Experimental Marine Biology and Ecology, 330, 119-131.

Nakamura Y., Kawasaki H. & Sano M. (2007) Experimental analysis of recruitment patterns of coral reef fishes in seagrass beds: Effects of substrate type, shape, and rigidity. Estuarine Coastal and Shelf Science, 71, 559-568.

64

Newbold J.D., Elwood J.W., O’Neill R.V. & Sheldon A.L. (1983) Phosphorus dynamics in a woodland stream ecosystem: A study of nutrient spiraling. Ecology, 64, 1249-1265.

Notestein S.K., Frazer T.K., Hoyer M.V. & Canfield D.E. (2003) Nutrient limitation of periphyton in a spring-fed, coastal stream in Florida, USA. Journal of Aquatic Plant Management, 41, 57-60.

Odum H.T. & Caldwell D.K. (1955) Fish respiration in the natural oxygen gradient of an anaerobic spring in Florida. Copeia, 104-106.

Odum, H. T. (1957) Primary production measurements in eleven Florida springs and a marine turtle-grass community. Limnology and Oceanography, 2, 85-97.

Olabarria C., Underwood A.J. & Chapman M.G. (2002) Appropriate experimental design to evaluate preferences for microhabitat: an example of preferences by species of microgastropods. Oecologia, 132, 159-166.

Payne B.S. (1979) Bioenergetic Budgeting of Carbon and Nitrogen in the Life-histories of Three Lake Populations of the Prosobranch Snail, Gonobasis livescens. Ph.D. Dissertation, Syracuse University, Syracuse, New York.

Pedersen M.F. & Borum J. (1996) Nutrient control of algal growth in estuarine waters: Nutrient limitation and the importance of nitrogen requirements and nitrogen storage among phytoplankton and species of macroalgae. Marine Ecological Progress Series, 142, 261-272.

Power M.E. (1992) Habitat heterogeneity and the functional significance of fish in river food webs. Ecology, 73, 1675-1688.

Quinn G.P & Keough M.J. (1993) Potential effect of enclosure size on field experiments with herbivorous intertidal gastropods. Marine Ecology Progress Series, 98, 199-201.

Quinlan R. & Smol J.P. (2002) Regional assessment of long-term hypolimnetic oxygen changes in Ontario (Canada) shield lakes using subfossil chironomids. Journal of Paleolimnology, 27, 249-260.

Quinlan R., Smol J.P. & Hall R.I. (1998) Quantitative inferences of past hypolimnetic anoxia in south-central Ontario lakes using fossil midges (Diptera: Chironomidae). Canadian Journal of Fisheries and Aquatic Sciences, 55, 587-596.

Richardson T.D., Scheiring J.F. & Brown K.M. (1988) Secondary production of two lotic snails (Pleuroceridae: Elimia). Journal of the North American Benthological Society, 7, 234-245.

Richardson J.S. (1991) Seasonal food limitation of detritovores in a montane stream: an experimental test. Ecology, 72, 873-887.

65

Ringle K (1999) Unlocking the labyrinth of north Florida springs. National Geographic March: 168, 40-59.

Ritchie, B. 2006, June 19. Ichetucknee Springs is feeling the pain too. Tallahassee Democrat. Rosemond A.D., Mullholland P.J. & Elwood J.W. (1993) Top-down and bottom-up control of

stream periphyton: Effects of nutrients and herbivores. Ecology, 74, 1264-1280.

Rosenau J.C., Faulkner G.L., Hendry C.W. & Hull R.W. (1977) Springs of Florida. Bulletin 31 (revised), Division of Management, Bureau of Geology, Florida Department of Natural Resources, Tallahassee. 461 pp.

Rounick J.S. & Winterbourn M.J. (1983) The formation, structure and utilization of stone surface organic layers in two New Zealand streams. Freshwater Biology, 13, 57-72.

Ryder D.S., Robyn J.W., Errol N. & Burns A. (2006) Can flow velocity regulate epixylic biofilm structure in a regulated floodplain river? Marine and Freshwater Research, 57, 29-36.

Sartory D.P. & Grobbelaar J.U. (1984) Extraction of chlorophyll a from freshwater phytoplankton for spectrophotometric analysis. Hydrobiologia, 114, 177-187.

Schagerl M. & Donabaum K. (1998) Patterns of major photosynthetic pigments in freshwater algae. 1. Cyanoprokaryota, Rhodophyta and Cryptophyta. International Journal of Limnology, 39, 35-47.

SRWMD. Springs Data Base, Suwannee River Water Management District. http://www.srwmd.state.fl.us/countyspringsummary/countyspringsummary.asp?county_code=F023&Submit=GO

Steinman A.D., McInitire C.D., Gregory S.V. Lamberti G.A & Ashkenas L.R. (1987) Effects of herbivore type and density on taxonomic structure and physiognomy of algal assemblages in laboratory streams. Journal of the North American Benthological Society, 6, 1640-1648.

Steinman A.D. (1992) Does an increase in irradiance influence periphyton in a heavily-grazed woodland stream. Oecologia, 91, 163-170.

Steinman A.D., Boston H.L. & Hill W.R. (1992) Functional-responses associated with growth form in stream algae. Journal of the North American Benthological Society, 11, 229-243.

Steinman A.D. (1996) Effects of grazers on benthic freshwater algae. pp 432-373 in R.J. Stevenson, M.L. Bothwell & R.L. Lowe (editors). Algal ecology: freshwater benthic ecosystems. Academic Press, San Diego.

Strommer J.L. & Smock L.A. (1989) Vertical distribution and abundance of invertebrates within the sandy substrate of a low-gradient headwater stream. Freshwater Biology, 22, 263-274.

66

Taylor A. & Maher W. (2003) The use of two marine gastropods, Austrocochela constrida and Bembicium auratum as biomonitors of zinc, cadmium and copper exposure: Effects of mass, within and between site variability and net accumulation relative to environmental exposure. Journal of Coastal Research, 19, 541-549.

Thompson F.G. (2004) An identification manual for the freshwater snails of Florida. For the Florida Museum of Natural History. 97 pp.

Tuchman N.C. & Stevenson R.J. (1991) Effects of selective grazing by snails on benthic algal succession. Journal of the North American Benthological Society, 10, 430-443.

Tyson R.V. & Pearson T.H. (1991) Modern and ancient continental shelf anoxia: an overview. In Tyson R.V & Peason T.H. (editors) Modern and Ancient Continental Shelf Anoxia. Geological Society Special Publication, 58, 1-24.

USGS, Ground Water Atlas of the United States. http://capp.water.usgs.gov/gwa/index.html Vandijk G.M. (1993) Dynamics and attenuation characteristics of periphyton upon artificial

substratum under various light conditions and some additional observations on periphyton upon Potamogeton pectinatus. Hydrobiologia, 252, 143-161.

Walde S.J. & Davies R.W. (1984) Invertebrate predation and lotic prey communities: evaluation of in situ enclosure/exclosure experiments. Ecology, 65, 1206-1213.

Wallace J.B. & Webster J.R. (1996) The role of macroinvertebrates in stream ecosystem

function. Annual Review of Entomology, 41, 115–139. Watson A. M. & Ormerod S. J. (2004) The microdistribution of three uncommon freshwater

gastropods in the drainage ditches of British grazing marshes. Aquatic Conservation-Marine and Freshwater Ecosystems, 14, 221-236.

Wang L.Z., Robertson D.M. & P.J. Garrison (2007) Linkages between nutrients and assemblages of macroinvertebrates and fish in wadeable streams: Implication to nutrient criteria development. Environmental Management, 39, 194-212.

Williams D.D. & Hynes H.B.N. (1974) The occurrence of benthos deep in the substratum of a stream. Freshwater Biology, 4, 233-256.

Woodruff A. (1993) Florida springs chemical classification and aquatic biological communities. M.S. Thesis. University of Florida, Gainesville. 117 pp.

Wright R.M. & McDonnell A.J. (1986a) Macrophyte growth in shallow streams: Field investigations. Journal of Environmental Engineering, 112, 952-966.

Wright R.M. & McDonnell A.J. (1986b) Macrophyte growth in shallow streams: Biomass model. Journal of Environmental Engineering, 112, 967-981.

67

68

Wu R. S. S. & Or Y. Y. (2005) Bioenergetics, growth and reproduction of amphipods are affected by moderately low oxygen regimes. Marine Ecology-Progress Series, 297, 215-223.

BIOGRAPHICAL SKETCH

Kristin was born in Long Island and grew up in Massachusetts. She attended Florida State

University for her undergraduate degrees in biology and history before enrolling at University of

Florida for her masters degree. She currently lives in Virginia with her husband and dogs.

69