Citation
Quantifying the Effects of Epiphytic Algae on the Growth of a Submersed Macrophyte

Material Information

Title:
Quantifying the Effects of Epiphytic Algae on the Growth of a Submersed Macrophyte
Creator:
Guan, Jing
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (166 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Interdisciplinary Ecology
Committee Chair:
FRAZER,TOM K
Committee Co-Chair:
JACOBY,CHARLES A
Committee Members:
BRENNER,MARK
KIKER,GREGORY A

Subjects

Subjects / Keywords:
epiphytes -- light -- macrophytes -- model -- threshold
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Interdisciplinary Ecology thesis, Ph.D.

Notes

Abstract:
Declines in the abundance of submersed macrophytes in Florida's spring systems are attributed, in large part, to the proliferation of nuisance algae. Macroalgal mats and increased epiphytic burdens on the leaves of native macrophytes are particularly problematic as these algae intercept incident light necessary for photosynthesis. Prolonged shading leads to loss of macrophytes and the refuge, foraging habitat and other important ecosystem services they provide. In the Chassahowitzka River, a spring-fed system along the west coast of peninsular Florida, documented increases in epiphytes on macrophytes were temporally concordant with losses of important macrophytes, such as Vallisneria americana, and I explored the causal link between these events by studying the impacts of epiphytic loads on light attenuation and growth of V. americana. My results suggest that even low loads of epiphytes result in a marked reduction of the light available for growth. Therefore, to provide water resource managers with an objective tool to support decisions and improve water management activities, a simulation model of the growth of Vallisneria americana was developed. The model, based on observations in the laboratory and the field, predicts the growth of V. americana under different loads of epiphytes. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2017.
Local:
Adviser: FRAZER,TOM K.
Local:
Co-adviser: JACOBY,CHARLES A.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2018-06-30
Statement of Responsibility:
by Jing Guan.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
6/30/2018
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 QUANTIFYING THE EFFE CTS OF EPIPHYTIC ALG AE ON THE GROWTH OF A SUBMERSED MACROPHYTE By JING GUAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2017

PAGE 2

2 2017 Jing Guan

PAGE 3

3 To my parents and husband

PAGE 4

4 ACKNOWLEDGMENTS Certain things cannot be completed alone and completion of this dissertation definitely should be ranked as one of them. Without inspiration and guidance from my committee, this work would not have been done. My committ ee chair Dr. Thomas Frazer was always there with an open ear and sage advice on managing my progress and staying on track. He made me feel more confident and convinced me tha t what I am doing is worthwhile. I cannot thank him enough for his support. My c o chair Dr. Charles Jacoby provided encouragement and guidance me to follow up when things did work He was amazingly motivated, dedicated, and selfless with his time and energy. Dr. Frazer always challenged me to step back to take a look at the big picture of my research system, and Dr. Jacoby helped me home in on specific questions. I very much appreciate their consistent and combined efforts to provide me with a doctoral researc h experience that had freedom and direction. Dr. Greg Kiker was selfless with his time and energy and I thank him for his encouragement and willing ness to share his experience of the modeling world. Dr. Mark Brenner was always a welcoming presence and pos itive influence that encouraged me to keep going. His frank and insightful questions provoked thought and discussion that immensely improved this product. I also would like to thank my lab mates: Jessica Frost, Joelle Laing, and Morgan Farrell. Thanks for their selfless and amazing dedication in the field. Field days with them were some of the most enjoyable and memorable highlights of my doctoral study I could not ha ve asked for a better group.

PAGE 5

5 Most importantly, none of this could have happened without the loving support of my parents and husband. Their constant love, understanding, and encouragement gave me huge motivation. Finally, I thank all who have added spice to my life during these four years of study T heir encouragement and support made comple ting this research as enjoyable as possible.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 ................................ ................................ ......... 14 The Chassahowitzka River ................................ ................................ ..................... 17 Objectives ................................ ................................ ................................ ............... 19 2 LIGHT ATTENUATION BY EPIPHYTES ON VALLISNERIA AMERICANA ............ 22 Background ................................ ................................ ................................ ............. 22 Materials and Methods ................................ ................................ ............................ 26 Study Site and Environmental Data ................................ ................................ .. 26 Sampling ................................ ................................ ................................ .......... 26 Determination of Light Transmission Through Varying Epiphytic Loads .......... 27 Modeling Light Transmission Through Epiphytes ................................ ............. 29 Results ................................ ................................ ................................ .................... 32 Composition of Epiph ytic Communities ................................ ............................ 32 Determination of Light Transmission Through Different Epiphytic Loads ......... 32 Modeling Light Transmission Through Epiphytes ................................ ............. 33 Discussion ................................ ................................ ................................ .............. 34 3 THE EFFECTS OF E PIPHYTES ON THE GROWTH OF VALLISNERIA AMERICANA ................................ ................................ ................................ .......... 68 Background ................................ ................................ ................................ ............. 68 Materials and Methods ................................ ................................ ............................ 70 Study Sites ................................ ................................ ................................ ....... 70 Environmental Measurements ................................ ................................ .......... 71 Biomass of Vallisneria americana and Epiphytes ................................ ............. 71 Measurements of Vallisneria americana Growth and Epiphytic Loads ............. 72 Determination of the Plastochrone Interval (PI) and Leaf Age for Vallisneria americana ................................ ................................ ................................ ..... 74 Statistical Analyses ................................ ................................ .......................... 75 Results ................................ ................................ ................................ .................... 76 Environmental Data ................................ ................................ .......................... 76

PAGE 7

7 In situ Biomass of Vallisneria americana and Epiphytes ................................ .. 76 Effect of Leaf Age on Growth of Vallisneria americana ................................ .... 77 Effect of Epiphytic Load on growth of Vallisneria americana ............................ 77 Estimation of a Threshold for Detrimental Epiphytic Load ................................ 78 Discussion ................................ ................................ ................................ .............. 78 4 A MODEL OF VALLISNERIA AMERICANA GROWTH UNDER DIFFERENT EPIPHYTIC LOADS ................................ ................................ ................................ 99 Background ................................ ................................ ................................ ............. 99 Description of the Model ................................ ................................ ....................... 102 Growth of Vallisneria americana ................................ ................................ ..... 103 Photosynthesis of Vallisneria americana ................................ ........................ 104 Available Light for Vallisneria americana ................................ ........................ 104 Loss of Photosynthetic Product ................................ ................................ ...... 105 Distribution of Epiphytic Biomass ................................ ................................ ... 106 Simulations ................................ ................................ ................................ ..... 107 Sensitivity Analysis ................................ ................................ ......................... 108 Result s and Discussion ................................ ................................ ......................... 109 Use of Boundary Data to Estimate Detrimental Effects of Epiphytes ............. 109 Comparative Modeling for Salt Creek and Small Creek ................................ 110 Modeling Growth in the Absence of Epiphyte Loads ................................ ...... 111 Stratifying Epiphytic Loads ................................ ................................ ............. 113 Scenarios with Epiphytic Loads of 4 mg DW cm 2 and 5 mg DW cm 2 ........... 114 Sensitivity Analysis ................................ ................................ ......................... 115 Summary ................................ ................................ ................................ .............. 115 5 CONCLUSIONS ................................ ................................ ................................ ... 136 APPENDIX A CODE FOR CHAPTER 2 ................................ ................................ ...................... 137 B CODE FOR CHAPTER 4 ................................ ................................ ...................... 142 LIST O F REFERENCES ................................ ................................ ............................. 152 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 166

PAGE 8

8 LIST OF TABLES Table page 2 1 Parameter estimates and coefficients of determination (R 2 ) for unconstrained models ................................ ................................ ................................ ................ 40 2 2 Parameter estimates and coefficients of determination (R 2 ) for constrained models ................................ ................................ ................................ ................ 41 2 3 Candidate models ranked from best to worse based on AICc value, difference values ( ), and Akaike weights ( ). ................................ ................. 42 2 4 Comparison of models and key parameters relating light transmission to loads of epiphytes in this and previous studies. ................................ .................. 43 2 5 Comparison of methods in this and previous studies. ................................ ........ 44 3 1 Statistics for multiple regression involving epiphytic loads, leaf ages and their interaction for Salt Creek. ................................ ................................ ................... 82 3 2 Statistics for multiple regression involving epiphytic loads, leaf ages and their interaction for Small Creek. ................................ ................................ ................ 83 3 3 Estimated threshold of epiphytic loads that prevent the growth of Vallisneria americana ................................ ................................ ................................ .......... 84 4 1 Equations comprising the growth and production model for Vallisneria americana ................................ ................................ ................................ ........ 118 4 2 Parameters comprising the growth and production model for Vallisneria americana ................................ ................................ ................................ ........ 119 4 3 Conversions used in the growth and production model for Vallisneria americana ................................ ................................ ................................ ........ 120 4 4 Percenta ges of net photosynthetic production for sections of Vallisneria americana leaves. ................................ ................................ ............................ 121

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Location of the Chassahowitzka River and its estuary. ................................ ...... 21 2 1 L ocation of the sampling area and head springs in the Chassahowitzka River, Citrus County, Florida.. ................................ ................................ ............. 45 2 2 Apparatus used to measure attenuation of natural photosynthetically active radiation (PAR) by epiphytes. ................................ ................................ ............. 45 2 3 Schematic representation of measuring the effect of epiphytes on l ight penetration. ................................ ................................ ................................ ......... 46 2 4 The dominant green filamentous macroalgae Enteromorpha sp. in my study site ................................ ................................ ................................ ..................... 46 2 5 Other epiphytes in my study site. ................................ ................................ ........ 47 2 6 Scatter plots showing the relationships between epiphytic loads and light transmission. ................................ ................................ ................................ ...... 48 2 7 Comparison of relationships between epiphytic loads, as mg dry weight cm 2 of leaf, and light transmission, in previous studies. ................................ ............. 50 2 8 Decay equations fit to the relationships between total epiphytic loads (mg dry weight cm 2 of leaf) and light transmission. ................................ ......................... 51 2 9 Decay equations fit to the relationships between total epiphytic loads (mg ash free dry weight cm 2 of leaf) and light transmission. ................................ ..... 53 2 10 Decay equations fit to the relationships between total epiphytic loads (mg ash dry weight cm 2 of leaf) and light transmission. ................................ ................... 55 2 11 Decay equations fit to the relationships between total epiphytic loads (g chlorophyll a cm 2 of leaf) and light transmis sion. ................................ ............... 57 2 12 Decay equations fit to the relationships between total epiphytic loads (mg dry weight cm 2 of leaf) and light trans mission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. ...................... 59 2 13 Decay equations fit to the relationships between total epiphytic loads (mg ash free dry weight cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. ................................ ................................ ................................ ................... 61

PAGE 10

10 2 14 Decay equations fit to the relationships between total epiphytic loads (mg ash dry weight cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. ...................... 63 2 15 Decay equations fit to the relationships between tota l epiphytic loads (g chlorophyll a cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. ................... 65 2 16 Leaf from the Chassahowitzka river showing heterogeneous epiphytic loads on its two sides. ................................ ................................ ................................ .. 67 3 1 Location of study areas in Salt Creek and Small Creek ................................ ...... 85 3 2 Study areas in the Chassahowitzka River. ................................ ......................... 85 3 3 Equipment used to take measurements in the field.. ................................ .......... 86 3 4 Depiction of sampling events. ................................ ................................ ............. 86 3 5 Typical Vallisneria americana shoot during processing. ................................ ..... 87 3 6 Holes in leaves used to measure growth. ................................ ........................... 88 3 7 Labeled, pre weighed boats containing samples fo r processing. ....................... 88 3 8 Schematic of modified leaf marking technique. ................................ .................. 89 3 9 Schematic of the moving split window technique. ................................ .............. 89 3 10 Distribution of epiphytic loads along the leaves of Vallisneria americana .......... 90 3 11 The relationship between epiphytic load and leaf age for Vallisneria americana ................................ ................................ ................................ .......... 91 3 12 Percentage of Vallisneria americana leave s in different age classes (days) ...... 92 3 13 Growth of leaves of Vallisneria americana (cm 2 per week) of different ages (days) ................................ ................................ ................................ ................. 93 3 14 Maximum growth of Vallisneria americana leaves (cm 2 per week) of different ages (days) ................................ ................................ ................................ ......... 94 3 15 Relationships and trends for epiphytic loads and growth of Vallisneria americana shoots in Salt Creek. ................................ ................................ ......... 95 3 16 Relationships and trends for epiphytic loads and growth of Vallisneria americana shoots in Small Creek. ................................ ................................ ...... 96 3 17 Squared Euclidean distances (SED) versus epiphytic loads in Small Creek. ..... 97

PAGE 11

11 3 18 Squared Euclidean distances (SED) versus epiphytic loads in Salt Creek.. ....... 98 4 1 Unified modeling language diagram for the Chassahowitzka Spring system. .. 122 4 2 Maximum potential growth (mg dry weight [DW] per week) for Vallisneria americana leaves of different ages. ................................ ................................ .. 123 4 3 Observed growth of Vallisneria americana leaves (mg DW per week) as a function of epiphytic load (mg DW cm 2 of leaf). ................................ ............... 124 4 4 Variables used in the growth and production model for Vallisneria americana 126 4 5 Comparison of simulated and maximum observed growth of Vallisneria americana leaves with different epiphytic loads.. ................................ .............. 127 4 6 Comparison of model predictions with in situ measurements of growth for leaves of Vallisneria americana ................................ ................................ ....... 128 4 7 Differences in net photosynthetic production for leaves of Vallisneria americana with different epiphytic loads from Small Creek. ............................. 129 4 8 Differences in net photosynthetic production for leaves of Vallisneria americana with different epiphytic loads from Salt Creek.. ............................... 130 4 9 Distribution of epiphytic loads and differences in net photosynthetic production for sections of leaves of Vallisneria americana with different epiphytic loads from Salt Creek. ................................ ................................ ....... 133 4 10 Results of sensitivity analyses for parameters. ................................ ................. 135

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy QUANTIFYING THE EFFE CTS OF EPIPHYTIC ALG AE ON THE GROWTH OF A SUBMERSED MACROPHYTE By Jing Guan December 2017 Chair: Thomas K. Frazer Co chair: Charles A. Jacoby Major: Interdisciplinary Ecology Declines in the abundance of submersed macrophytes in systems are attributed, in large part, to the proliferation of nuisance algae. Macroalgal mats and increased epiphytic burdens on the leaves of native macrophytes are particularly problematic as these algae intercept incident light necessa ry for photosynthesis Prolonged shading leads to loss of macrophytes and the refuge foraging habitat and other important ecosystem services they provide In the Chassahowitzka River, a spring fed system along the west coast of peninsular Florida, documen ted increases in epiphytes on macrophytes were temporally concordant with losses of important macrophytes, such as Vallisneria americana and I explored the causal link between these events by studying the impacts of epiphytic loads on light attenuation an d growth of V americana My results suggest that even low loads of epiphytes result in a marked reduction of the light available for growth. Therefore, t o provide water resource managers with an objective tool to support decision s and improve water manage ment activities, a simulation model of the growth of Vallisneria

PAGE 13

13 americana was developed. The model based on observations in the laboratory and the field predict s the growth of V americana under different loads of epiphytes

PAGE 14

14 CHAPTER 1 INTRODUCTION Status of S ystems The landscape of peninsular Florida between latitudes 27 and 31 is punctuated by a large number and diversity of springs, due, in large part, to a highly permeable karst geology, abundant rainfall and a vast aquifer (Wetland Solutions Inc. 2010). These springs continuously discharge freshwater that supplies the downstream lakes, streams, rivers and estuaries that form These springs are characterized by cle ar water that is rich in dissolved nutrients and gases and constant in temperature and chemical content (Knight and Notestein 2008). These characteristics create and their downstrea m receiving waters Besides the se ecological values, clear water and associated plants and animals have been important economic resources that support various recreational activities attract a large number of visitors and generate millions of dollars of a nnual revenue for local economies (Bonn 2004, Scott et al. 2002). Unfortunately many springs in Florida show signs of substantial degradation including a proliferation of nuisance algae declines in native macrophytes, altered food webs and deterioratin g aesthetics ( Wetland Solutions Inc. 2010 ). In the famous Silver Springs, for example, fish populations have experienced significant declines (Munch et al. 2006). Human activities such as pumping groundwater and more intensive use of the land in watersheds of springs likely underlie many of the observed problems (Cohen et al. 2007). G roundwater depletion and nutrient pollution are of particular concern because they can reduce flows and lead to eutrophication (Knight and Notestein 2008). This predi cament creates significant negative feedbacks on ecological health public

PAGE 15

15 health, and local economies ; therefore, a greater emphasis on ecological research and monitoring in become a focus One focus for recent studies has bee n submersed macrophytes which always have been recognized as a ( Choice et al. 2014, Canfield and Hoyer 1988, Mattson et al. 1995 ) Submersed macrophytes are vascular aquatic plants that hav e internal transport structure s and true roots, and they are principally angiosperms that can flower and produce enclosed seeds on a season al basis (Sculthorpe 1985). Common native sagittaria ( Sagittaria kurziana ), w ild celery /eel grass ( Vallisneria americana ), southern naiad ( Najas guadalupensis ), coontail ( Ceratophyllum demersum ), and fanwort ( Cabomba caroliniana ), with Eurasian milfoil ( Myriophyllum spicatum ) and hydrilla ( Hydrilla verticillata ) being common invasi ve species The favorable substrate, flow and light environment in submersed macrophytes. For example, the total annual production of leaves for V. americana can reach 2704 g m 2 in Kings Bay, which belongs to the Crystal River/Kings Bay spring complex (Hauxwell et al. 2007). As some of the most important primary producers in submersed macrophytes play important ecological roles. They provide food for a number of grazers, enrich the water w ith oxygen that sustain s high abundance s of aquatic animals provide a refuge from predators, reduce turbidity by stabilizing the river bed and improve water quality by taking up nutrients (Smart et al. 1994, Rogers et al. 1995, Wigand et al. 2000). Give n the important ecological roles they play submersed macrophytes represent a useful indicator for evaluating the ecological

PAGE 16

16 recent decades, declines in the abundance of submersed macrophytes have been observed in several sp rings, lakes, estuaries, and rivers in Florida (Hauxwell et al. 2004, Knight and Notestein 2008). D eclines of native submersed springs have been attribute d to climate change, degradation of water quality (e.g. excessive loads of n utrients, herbicides and other chemical pollutants ), increased disturbance (e.g. trampling and propeller scars ), competition from exotic species (e.g. hydrilla), and light limitation (Canfield and Hoyer 1988 ) Among those factors contributing to declines of native submersed macrophytes algal proliferation has been widely recognized as a main causal factor I n recent decades, massive macroalgal mats and heavy loads of ecosystems. For example, a survey of 29 Florida springs found that almost all of them harbored macroalgae, and approximately 50% of the benthic substrate was covered by algal mat s with an average thickness of 0.5 m and a maximum thickness in some springs of 2 m (Stevenson et al. 2007). These algal communities consist of myriad microalgae, diatoms, cyanobacteria, microbes, and macroalgae (Notestein 2001) ring systems include the genera Chara and Nitella that resemble vascular plant s and filamentous genera principally green algae and cyanobacteria such as Cladophora Enteromorpha Lyngbya and Vaucheria (Knight and Notestein 2008). These epiphytic algae appeared to have proliferated rapidly, possibly stimulated by reductions in flow and nutrient enrichment in springs (Hauxwell et al. 2004, Hoyer et al 2004, Heffernan et al 2010 ). Once established, these algae have been shown to intercept a large proport ion of incident light necessary for photosynthesis

PAGE 17

17 and growth of submersed macrophytes (Phillips et al. 1978 ). E vidence suggests the occurrence of a fundamental shift in dominance from submersed macrophytes to nuisance algae with this event being a major concern for a broad group of environmental scientists and managers of natural resources (Springs Management Plan 2016). The Chassahowitzka River One place where increased epiphytic loads on macrophytes have been documented is the Chassahowitzka River (Lat a spring (Figure 1 1) These increases were temporally concordant with losses of important macrophyte species such as Vallisneria americana (Notestein 2001). S uch an observation sug gests a cause and effect relationship, so the system may be particularly suitable for a study of the impacts of epiphytes on macrophytes. The Chassahowitzka River is located in southwest Citrus County, FL. The climate in this area is subtropical, with mean annual precipitation ranging from 132 cm to 142 cm (United States Fish and Wildlife Service 1988). This spring fed, coastal river originates at the Chassahowitzka Spring, which is a first magnitude spring ( Yobbi and Knockenmus 1989 ) and it also is fed by several smaller spring vents located in its tributaries (Crab, Baird, and Potter Creeks Notestein 2001) The discharge from the main spring complex ranges from 2 to 8 m 3 s 1 and the mean discharge between 1930 and 1972 was 3.92 m 3 s 1 ( Rosenaur et al 1977). The river flows west approximately 4 km from the main spring boil to a coastal marsh complex and then another 4 km to the Gulf of Mexico (Notestein 2001) Given a gradient in elevation of 3 m or less, mean flow rates are generally less than 0.20 m s 1 (Frazer et al 2006)

PAGE 18

18 Such low flow rates mean that tidal cycles influence both spring discharge and flow within the river (Yobbi 1992), with tidal mixing creating brackish creeks and bays in the lower river. Above the marsh complex, the depth of the m idstream channel ranges from 0.5 m to 2.6 m, with a mean depth of 1.2 m (Notestein 2001). The upper portion of the river is narrow with a minimum width of 44 m and a maximum width of 175 m midway downstream, and in total, the wetted surface area of the r iver (above the marsh complex) is approximately 360,000 m 2 (Notestein 2001). Approximately 3% the wetted area is covered by a canopy of riparian vegetation (Notestein 2001). Water clarity in the river is good and light attenuation coefficients ( ) are generally less than 1.5 (Frazer et al. 2006) The primary substrate along the river bottom is sand ( ~ 54%), with varying mixtures of mud and rock (Frazer et al. 2006) Mud is more prevalent near the shoreline, and small patches of exposed limestone appear throughout the river ( ~ 1% of the total bottom area), which is indicative of the dominance of limestone in the underlying geology (Brooks 1981). The influence of groundwater keeps water temperature along the length of the river fairly uniform, ranging from 20.7 C to 26.4 C, but temporal variations caused by seasonal changes in air temperatures did occur during my sampling. In combination, the light environment and substrate in the Chassahowitzka River appear favorable for the growth of primary producers, wi th macrophytes, macroalgae, and periphyton observed throughout most of the river ( Notestein 2001 ) The density of submersed aquatic vegetation declines gradually with distance downstream because of higher salinities (Yobbi and Knochenmus 1989). Common macr ophytes include Vallisneria americana Potamogeton pectinatus Najas guadalupensis Myriophyllum

PAGE 19

19 spicatum and Hydrilla verticilla Some Sagittaria kurziana Ruppia maritima Potamogeton illinoensis and Ceratophyllum demersum also have been observed in this river (Frazer et al. 2006) Based on visual estimates, V. americana is the dominant macrophyte in the river (Notestein 2001). In addition, benthic algae represent approximately 43% of the total SAV biomass, composed primarily of Lyngbya sp. and Chaeto m orpha sp., with Gracilaria sp. and Enteromorpha sp. also abundant (Notestein 2001) Although a large portion of the river and estuary are protected in the Chassahowitzka National Wildlife Refuge the system has exhibited increasing anthropogenic impacts including nitrate concentrations at the headspring that have increased from 0.01 mg L 1 to over 0.5 mg L 1 or more than 50 fold since the 1960s (Jones et al. 1997 Frazer 2000, Scott et al 2004). These increased nutrient concentrations may have stimulate d macro algal blooms, which can impact the survival of macrophytes and ultimately the ecological, economic and social value of the resource (Pinckney et al 2001). Therefore, important questions regarding spring macrophytes and epiphytic algae are : How much light is intercepted by epiphytic algae on How much epiphytic biomass cause s damage or dysfunction for ; Can we effectively model the performance of macrophytes subject to different loads of epiphytes Objectiv es Toward this end, laboratory and field measurements and a model were used to answer these questions and achieve three main objectives: 1) empirically model the relationship between epiphytic biomass on V. americana leaves and light transmission (Chapter 2); 2) document the impacts of epiphytes on the growth of V. americana in the

PAGE 20

20 field (Chapter 3); and 3) employ these data to develop and calibrate a model that predict s the performance of V. americana under different loads of epiphytes (Chapter 4). The overarching objectives of this research were to develop a more comprehensive understanding of the impacts of epiphyte s on macrophytes and to provide water resource managers with objective tool s to assess the vulnerabil ity of macrophytes to the negative impacts of increased loads of epiphytes

PAGE 21

21 Figure 1 1. Location of the Chassahowitzka River and its estuary. Labels indicate spring vents and tributaries. Reprinted with permission from Chassahowitzka.net, http://www.chassahowitzka.net/rmap.htm (November 18, 2017).

PAGE 22

22 CHAPTER 2 LIGHT ATTENUATION BY EPIPHYTES ON VALLISNERIA AMERICANA Background Florida has abundant spring resources, which have important economic and ecological value (Scott et al. 2002, Bonn 2004). In these valuable spring systems, submersed freshwater macrophytes historically dominate d primary production because of the shallow wa ter and suitable sediments. As some of the most important primary producers, submersed macrophytes play important roles in regulating the aquatic environment by providing food enriching the water with oxygen stabilizing the sediment cycling nutrients a nd acting as food and a refuge that sustains high abundances of animals (Smart et al. 1994, Rogers et al. 1995, Wigand et al. 2000). systems, they represent a useful indic a tor o f environmental health. In recent decades, environmental reports have documented declines in the abundance of macrophytes in several lakes, estuaries and rivers in Florida (Hauxwell et al. 2004). The decline of buted to many factors: climatic change, water quality degradation (e.g., excessive nutrients and agricultural herbicides), increased human activities (e.g., trampling and scarring from anchors or propellers), competition from exotic species (e.g., Hydrilla verticillata ), and reduced availability of light. Among those factors responsible for declines in macrophytes, a frequently cited influence is low light availability which decreases net photosynthesis (Kimber et al. 1995, Hauxwell et al. 2007). The mini mum light requirement for survival of freshwater macrophytes is approximately 13% of light

PAGE 23

23 (Carter et al. 2000, Kemp et al. 2004). The amount of light available to submersed aquatic macrophytes can be affected by many environme ntal factors, such as riparian canopies, phytoplankton and turbidity that reduce water clarity, and algae that grow on the leaves of macrophytes (Carter et al. 2000). A lgae can intercept a large proportion of incident light that macrophytes need to support their metabolism and growth, and available light has always been considered an important cause of loss of macrophytes (Sand Jensen 1977, Phillips et al. 1978, Orth and Moore 1983, Twilley et al. 1985, Cambridge et al. 1986, Silberstein et al. 1986, Canfie ld and Hoyer 1988). The influence of algae has become especially evident in recent decades, with massive macroalgal spring systems ( Knight and Notestein 2008 ). In some s prings there may have be en a fundamental ecological shift in the aquatic primary producer communities from native macrophytes to nuisance algae (Springs Management Plan 2016). Such ecological shift s can be attributed to chemical, physical, and biological changes in the aquatic systems, including excessive anthropogenic loads of nutrients reduced water velocities, and declines in abundance of grazers (Hauxwell et al. 2004, Hoyer et al. 2004, Hefferna n et al. 2010). Excessive proliferation of macroalgal mats and epiphytes on leaves of macrophytes are particularly problematic as these algae can intercept the incident light that macrophytes require for photosynthesis. Thus, it is necessary to quantify th e relationship between epiphytic loads on macrophytes and light attenuation and to formulate a numerical model incorporating the effect of epiphytes on the growth of macrophytes.

PAGE 24

24 The relationship between epiphytic loads and light attenuation has been quant ified using different methods to measure light transmission through epiphytes (Table 2 5). Most studies used an indirect method to measure transmission of light that involved removing epiphytes from leaves and measuring light transmission through a resuspe nded slurry (Sand Jensen and Borum 1984, Neckles et al. 1993, Dixon 2000). This method is easy to apply, but it destroys the three dimensional (3D) structure of the submersed epiphytic communities. This 3D structure recently was demonstrated to be a very i mportant influence on light attenuation, especially by filamentous epiphytes (Vermaat and Hootsmans 1994, Brush and Nixon 2002, Drake et al. 2003, Stankelis et al. 2003). Filamentous epiphytes that are underwater extend away from the leaves of macrophytes (Brush and Nixon 2002), whereas coralline algae and diatoms form flat sheets that may be more suitable for evaluation by the indirect method (Vermaat and Hootsmans 1994, Drake et al. 2003). Some other studies used artificial leaves, such as as an alternative to natural leaves (Phillips et al. 1978, Van Dijk 1993, Glazer 1999, Stankelis et al. 2003, Notestein 2001). Although this method can document the age of epiphytes on artificial leaves, the significant differences between artificial leave s and natural leaves may lead to important differences in the epiphytes that in turn, may affect estimates of light transmission (Bulthuis and Woelkerling 1983, Brush and Nixon 2002). In addition, artificial light sources also were used widely in previous studies (Sand Jensen and Borum 1984, Sand Jensen 1990, Van D ijk 1993), but Drake et al. (2003) found that artificial light sources do not mimic the quality of natural light accurately, which may be a very important influence on photosynthesis by aquatic p lants.

PAGE 25

25 After light attenuation was measured for different epiphytic loads, three equations were used to describe light attenuation as a function of the loads. The least used approach was the linear equation of Glazer (1999) that described light transmissio n through epiphytic bryozoans; however, the linear model fit well only at very low densities of epiphytes (0 to 5 mg DW cm 2 ). In aquatic science, the most widely used equation to model epiphyte light attenuation is an exponential decay function ( ) derived from the Beer Lambert Law (Kirk 1994, Burt et al. 1995, Stankelis et al. 2003), where is incident light, is the coefficient of epiphytic light attenuation, and is attenuated light that decreases exponentially with or loads of epiphytes. S ome studies however, found that a hyperbolic decay equation was applicable and superior in explaining variation in light attenuation with differences in loads caused by epiphytic communities with various morphologies (e.g., microa lgal films, coralline algal crusts, and filamentous macroalgae, Van D ijk 1993, Vermaat and Hootsmans 1994, Brush and Nixon 2002). In my study area, the spring fed Chassahowitzka River on the west coast of peninsular Florida, increases in epiphytes on macro phytes were temporally concordant with losses of important species, such as Vallisneria americana (Notestein 2001). This observation suggests a cause and effect relationship, although the direct effects of epiphytes on light attenuation and the performance of macrophytes in the system have not been investigated. The objectives of this chapter were to measure and model light transmission through different loads of epiphytes found on V. americana leaves. The method used by Brush and Nixon (2002) to directly m easure light attenuation was employed in this study. Thus, all measurements of light attenuation were obtained using leaves of V. americana that were collected in the field and held underwater so the 3D

PAGE 26

26 structure of epiphytic communities were maintained as they were exposed to natural light. The resulting measurements became the basis for fitting exponential and hyperbolic decay models to describe the relationship between the density of epiphytes and light attenuation. These two mathematical models were eva luated to identify the best fit to the results and the most useful numerical model for predicting growth of macrophytes under differing loads of epiphytes Materials and Methods Study Site and Environmental Data Sampling was conducted in a large, continuou s V. americana meadow in the Chassahowitzka River. The meadow covered ~ 2700 m 2 of the underlying sand/mud substrate. The study area is in the mid about 1.8 km west of the head springs and 2.2 km upstream of the coastal marsh complex (Figure 2 1). River flows are primarily from spring discharge so water temperatures remain about 25.8C. Although freshwater contributes 99% of the flow, the study area is a tidally influenced f reshwater oligohaline system with a m tidal range. Through my sampling period (August 2015 October 2015), light attenuation coefficients ( ) in the water and average concentrations of total nitrogen (TN), total phosphorus ( TP), and chlorophyll a (Chl a) ranged from 1.5 to 3.2, 640 to 950 g T N L 1 20 to 55 g T P L 1 and 2.79 to 16.09 g Chl a L 1 respectively. The riparian canopy does not shade much of the water at my site, and mean in situ surface irradiance ranged from 16 06 to 1914 E m 2 s 1 Sampling Between August and October 2015, samples of leaves were collected from the Chassahowitzka River. A snorkeler haphazardly tossed five quadrats (0.25 m

PAGE 27

27 by 0.25 m) within the meadow and within the quadrats, leaves of V. ameri cana with varying amounts of epiphytes were clipped carefully at their bases and placed in 1 L Nalgene jars filled with ambient water. Jars containing 4 6 leaves were stored on ice in a cooler prior to processing (within 24 hr of collection). Determinatio n of Light Transmission Through Varying Epiphytic Loads A ttenuation of natural photosynthetically active radiation (PAR) by epiphytes was measured Measurements were taken with a LI 1400 datalogger (Figure 2 2 A ) and two, 2 (LI COR UWQ5754 and LI COR UWQ 5692, Figure 2 2 B ). One sensor measured light through a leaf of V americana and the other simultaneously measured natural, incident irradiance at the same depth The sensors were submersed in an outdoor tank filled with wate r (108 cm long 62 cm wide 48 cm high) at the University of Florida (Figure 2 2 C ). In order to minimize errors caused by dramatic changes in incident light (e.g., edge effects introduced by passing clouds), all measurements were made under direct light on clear days when incident light was greater than 1300 E m 2 s 1 In addition to limiting variation in the quantity and quality of light, efforts were made to limit variation arising from the heter ogeneous distribution of epiphytes on V americana leaves. Therefore, leaves were separated into a series of sections approximately 6 cm long that had relatively homogen e ous distributions of epiphytes. The effect of epiphytes on light penetration was measured via a two step process (Figure 2 3). Before me asuring the penetration of PAR, one side of each section was scraped with a scalpel to remove epiphytes that were saved for taxonomic identification. Each processed section was placed on one of the submerged sensors, which ensured

PAGE 28

28 the epiphytes extended in to the water column. When the uncovered sensor indicated that incident irradiance was stable, the quantity of light hitting that sensor and the quantity of light passing through the section of leaf and its attached epiphytes were recorded. Recordings were made at 1 s intervals for 30 s at three points along each section. Next, the epiphytes on the other side of each section were removed and saved for determination of epiphytic biomass per unit area (mg dry weight cm 2 of leaf, mg ash free dry weight cm 2 of leaf, mg ash dry weight cm 2 of leaf, and g chlorophyll a cm 2 of leaf). The resulting clean sections of leaves were placed over the appropriate submerged sensor, and incident light and light passing through the section were measured. Sections were saved, and their lengths and widths were measured to determine their surface areas (cm 2 ). For each section, light attenuation by epiphytes (percent reduction in transmission of PAR) was calculated from the amount of incident light passing through both the epiphytes and the leaf ( ) relative to the amount of light passing th rough the leaf only ( ): The effects of scraping were determined by comparing transmission of incident light before and after scraping sections of five leaves that had no visible epiphytes. The epiphytes initially rem oved from leaves were processed to determine the species composition of assemblages. Prior to examination under an anatomical lens the samples were distributed as a single layer in a wet petri dish. Representative types of algae were chosen according to t heir morphology (e.g., filaments, mats, tufts, or slurry) and color (e.g., green, blue green, or red). Next, a small sample of each of the chosen algae was transferred with tweezers or an eyedropper to a separate wet glass

PAGE 29

29 slide for microscopic analysis. T hereafter, algae were examined at 10 0 X and 40 0 X and identified using standard keys. The epiphytes scraped from the other side of each section were used to determine biomass per unit area as chlorophyll a content and various measures of mass. To measure t he mass of epiphytes, half of each sample was placed in a pre weighed, 20 ml aluminum tray. Samples in aluminum trays were processed to yield dry weight (DW), ash weight (AW), and ash free dry weight (AFDW). Dry weights were measured after the pre weighed wet samples were held in a forced air drying oven maintained at approximately 65C for 36 to 48 hr. Ash weights were measured after dried samples had been heated to 450C in a muffle furnace for 4 hr. Dry weight and ash weight for each sample were measured to the nearest 0.001 g on a Mettler P 163 balance. Ash free dry weight is the difference between dry weight and ash weight. The other half of each sample was wrapped in 47 mm diameter Whatman GF/F glass microfiber filters that were stored in a freezer at 20C for no more than 2 d ays Chlorophyll pigments were extracted from the epiphytes with 90% ethanol in a 79C water bath, and chlorophyll a concentrations were determined spectrophotometrically ( Montana Department of Environmental Quality 2011 ). Using t hese data, various measures of epiphytic density were determined by dividing the DW, AFDW, AW and chlorophyll a concentrations by the relevant areas for sections of leaves (one side of each section only). Modeling Light Transmission Through Epiphytes Mean values of three replicate measures of percent light transmission for individual sections were used in subsequent analyses. Empirical relationships between

PAGE 30

30 light transmission through epiphytes and epiphytic biomass were investigated with regression bas ed on a standard least squares approach. Exponential and hyperbolic decay models were fit to the mean values, and the results were compared to determine the relative suitability of these different regression models. Exponential decay functions and negative hyperbolic functions were fit in both two parameter and three parameter forms: i.e., , and and In the relevant equations, y is the transmittance of PAR through epiphytes ( %) expressed as percent of incid ent light passing through both epiphytes and leaves relative to the amount of light passing through the leaves only; x is the epiphytic density expressed as mg DW cm 2 mg AFDW cm 2 mg AW cm 2 and g chlorophyll a cm 2 ; and a, b and c are constants. All models were evaluated in both constrained (100% light transmission at a load of zero) and unconstrained forms. Parameter estimates together with coefficients of determination ( R 2 ) were calculated with CurveExpert statistical software and Python matplotlib. All models were evaluated with the Akaike Information Criterion (AIC, Akaike 1973, Burnham and Anderson 2001). This criterion assesses goodness of fit with a likelihood function, and it includes a penalty related to overfitting that is calculated on the basis of the number of model parameters. Thus, AIC is used to determine if increased complexity significantly increases the amount of variation explained by a model. The preferred model is the one with the minimum AIC value. The general form for calculatin g AIC is : ; where K is the number of parameters included in

PAGE 31

31 the model, is the natural logarithm of likelihood of the model, and the likelihood (L) reflects the overall fit of the model (larger values indicate better fit). AIC also could be calculated in a more conventional formula: where K is the number o f parameters included in the model, n is the sample size, and RSS is the residual sum of squares. For small sample sizes (n/K< 40), AIC requires a bias adjustment, which yields a second order Akaike Information Criterion (AICc) or wh ere variables are as defined above. As sample sizes (n) increase, the last term of the AICc approaches zero, and AICc approximately equals AIC (Burnham and Anderson 2002). The AICc value itself has no meaning; AICc values become meaningful when they are co mpared among a series of candidate models. The model with the lowest AICc is the best model. To compare models, and Akaike weight ( ) are used. The difference between the model with the lowest AICc (the best fitting model) and the others is expresse d as: where is the AICc value of model and is minimum AIC value for all models. Akaike weights ( ) is an another method of measuring the suitability of each model, which is represented as the normalized relative likelih ood value of the model: where is the Akaike weight for model the numerator is the relative likelihood for model and the denominator is the sum of the relative likelihoods for the whole set of R candidate models

PAGE 32

32 Results Composition of Epiphytic Communities The epiphytes on V. americana leaves collected in the Chassahowitzka River were dominated by a green filamentous macroalgae Enteromorpha sp. (Figure 2 4) that was mixed with a small amount of Cladophora sp. (Figure 2 5 A ) and diatoms (Figure 2 5 B ). Epiphytes mostly adhered directly to the surfaces of leaves Determination of Light Transmission Through Different Epiphytic Loads Direct measurements of the light attenuation by epiphytic loads were obtained from a total of 120 samples. Blades without epiphytes attenuated 93.52 2.6% of incident light (mean SD, n = 60). A comparison of light passing through sections of leaves without epiphytes before and after they were scraped indicated that scraping did not a lter light attenuation substantially, with scraped blades transmitting approximately 0.48% less incident light. The epiphytic loads ranged from 0.21 to 16.66 mg DW cm 2 of leaf (Figure 2 6 A ), 0.11 to 8.14 mg AFDW cm 2 of leaf (Figure 2 6 B ), 0.05 to 8.52 m g Ash DW cm 2 of leaf (Figure 2 6 C ), and 1.05 to 47.35 g chlorophyll a cm 2 of leaf (Figure 2 6 D ). Biotic (AFDW) and abiotic (ash DW) components showed that abiotic materials the epiphytic communities comprised 21.90% to 77.10% of the epiphytic communities Epiphytic communities on macrophytes were not homogeneously distributed, with greater accumulation on the older portions of leaves near their tips and less or no epiphytes on the new or basal portions of leaves. Light transmission expressed as percentage of the incident irradiance (I 0 ) that penetrated to the leaves (I) ranged from 4.48 to 87.15%. Four scatter plots of light transmission versus different measures of epiphytic loads display similar trends (Figure 2 6): transmission of light deca yed rapidly at low epiphytic loads and then

PAGE 33

33 gradually leveled off as epiphytic loads increased. Incident light could be attenuated by 80% at ~ 6.29 mg DW cm 2 of leaf (or 3.43 mg AFDW cm 2 of leaf, 2.88 mg Ash DW cm 2 of leaf or 17.08 g chlorophyll a cm 2 of leaf), and transmission could be attenuated by 90% at ~ 17.73 mg DW cm 2 of leaf (or 4.53 mg AFDW cm 2 of leaf, 3.23 mg Ash DW cm 2 of leaf or 22.49 g chlorophyll a cm 2 of leaf). Modeling Light Transmission Through Epiphytes The 32 regression models captured 30.69 83.22% of the variation in observations of light transmission through different epiphytic loads (R 2 values in Table 2 1 and Table 2 2). The correlation coefficients (|r|) of all regressions were greater than 0.55, ranging from +0.61 to +0.91 T he |r| of regressions that expressed epiphytic loads as mg DW cm 2 of leaf were as high as 0.9. Models based on dry weight (mg DW cm 2 of leaf) had the best performance with R 2 as high as 0.83 Models based on loads expressed as chlorophyll a ( g c hlorophyll a cm 2 of leaf) had inferior fits relative to other models. Comparisons of determination coefficients (R 2 ) of all candidate models revealed that the unconstrained models captured more of the variation than constrained models. Candidate models we re ranked according to their suitability (Table 2 3) as determined by AICc values, differences ( AICc and the lowest AICc, and Akaike weights ( ). A value of < 2 suggested substantial evidence for the suitability of the model, values of 3 < < 7 indicated that the model had considerably portion of the variation in the data (Burnham and Anderson 2002). AICc values ranged from 268.68 to 351.88. The values of for the first five models

PAGE 34

34 (Unconstrained_DW_Three parameter exponential decay, Unconstrained_DW_Two parameter exponential decay, Unconstrained_DW_Three parameter hyperbolic decay, Unconstrained_DW_ Two parameter hyperbolic decay, and Unconstrained_AFDW_Two parameter hyperbolic decay) were < 2, suggesting these models explained a substantial amount of variation in the data. The values of for models 6 to 10 (Unconstrained_AFDW_Two parameter exponent ial decay, Unconstrained_AFDW_Three parameter hyperbolic decay, Constrained_DW_Two parameter hyperbolic decay, Unconstrained_AFDW_Three parameter exponential decay, and Constrained_DW_Three parameter hyperbolic decay) were between 3 and 7, so they explain ed considerably less of the variation in the data. Models with values 12 explained very little of the variation in the data. These results indicated that model 1 (Unconstrained_DW_Three parameter exponential decay) was the best of the 32 candidate mode ls, with the minimum and an Akaike weight ( ) of 0.23 (Table 2 3). Discussion Most of the e piphytic loads were < 10 mg DW cm 2 of leaf, with only a few samples reaching 16.66 mg DW cm 2 of leaf. Epiphytic loads up to 100 mg DW cm 2 of leaf were note d by Brush and Nixon (2002), and the lower epiphytic loads in my study may be a consequence of the relatively low phosphorus concentrations in the Chassahowitzka River. In addition, epiphytes may be more abundant on macrophytes with complex morphologies or rough leaf surfaces (Notestein 2001); therefore, lower epiphytic loads may reflect the simple structure of the leaves of V. americana

PAGE 35

35 The fou r scatter plots relating light transmission and epiphytic loads show similar trends: transmission decayed rapidly at low epiphytic loads and then reductions in transmission approached zero as epiphytic loads increased (Figure 2 6 ). These trends illustrated the dramatic capacity of epiphytes to produce shade, even at low loads (thin layer s ). Beyond a certain point, increasing epiphytic loads did not attenuate much more light than thinner layers. The regressions in Figures 2 8 to 2 15 are consistent with thos e of most previous studies (Figure 2 7 Brush & Nixon 2002). Linear regressions were used in a few studies (Glazer 1999, Bulthuis and Woelkerling 1983, Agust et al.1994), but they failed to describe light transmission through a wide range of epiphytic den sities (Brush and Nixon 2002), so a linear model may be suitable only for epiphytic communities composed of bryozoans or microalgae rather than macroalgae. Theoretically, light transmission should be 100% when the epiphytic load is zero, but all models to pass through that point yielded lower R 2 values ( Table 2 1 and Table 2 2). The unconstrained three parameter exponential decay model based on dry weight of epiphytes (model 1 in Table 2 3) captured most of the variation in the data, with the highest AICc value 268.68 and minimum and Akaike weight ( ). The reliability of three parameter, exponential decay models was noted in previous studies (Silberstein et al. 1986, Stankelis et al. 2003, Frankovich and Zieman 2005). However, model 2 also could be considered appropriate because model 1 is only 1.46 times as likely to be the best (evidence ratio = 0.23/0.16, Table 2 3). This model describes light transmission through epiphytes as which generates an epiphytic light attenuation coefficient that parallels the water column attenuation coefficient ( Thus, transmission of light

PAGE 36

36 through both the water column and a load of epiphytes can be communicated simply by modifying the standard Beer La mbert model (Frankovich and Zieman 1994, Kemp et al. 2000) : Although the regressions were similar between this study and previous investigations, there were still differences in slopes and asymptotes (Table 2 4). For a given epiphytic load, light transmission through epiphytes varied from previous studies, with the epiphytic loads that reduce light transmission by 50% ranging from 1.06 to 22.12 mg DW cm 2 of leaf (Table 2 4). These differences may be caused by the composition and morphologies of the epiphytic communities. In my study area, the green filamentous macroalga Enteromorpha sp. was the dominant species on leaves of V. americana which agreed with observations reported in e Chassahowitzka River. In my study, about 1.69 mg DW cm 2 of Enteromorpha sp. could attenuate 50% of incident light, whereas in Florida Bay and the Florida Keys coralline algae and associated carbonate sediment comprised most of the epiphytic load, and i t took ~ 4.36 mg DW cm 2 to reduce transmission of light by 50% (Frankovich and Zieman 2005). Furthermore, ~ 22.12 mg DW cm 2 of Cladophora sp. study of Zostera marina Enteromorpha sp. ma y have attenuated light more strongly because it is darker in color and forms long dense patches on leaves of macrophytes. In addition, coralline algae may absorb less light because pigment contents of symbiotic algae in corals are lower than in other algal species (Vsquez Elizondo and Enrquez 2017, Losee and Wetzel 1983). Light transmission decreases to an asymptote as epiphytic loads increase because the filamentous algae extend away from the surfaces of leaves so lengthening

PAGE 37

37 filaments increase the epiphytic load more than they affect the areal coverage on leaves or the amount of light attenuation. Filamentous Enteromorpha sp. in this study reduced light transmission to approximately 10% of incident light, and three other filamentous algae, Ulothrix sp., Cladophora sp., and Polysiph o nia sp. reduced light transmission to 15%, 30%, and 18% of incident light, respectively (Brush and Nixon 2002). The asymptotic value may be determined by the detailed architecture of the epiphytic load, which is a topic for further study. Besides differences related to the composition and morphology of epiphytic communities, the method used to measure transmission of light could lead to differences among the results of investigations (Table 2 4). The method used to measur e transmission of light in this study kept the epiphytes in their natural orientations because they were submersed (Borum and Wium Anderson 1980, Sand Jensen and Borum 1984, Twilley et al. 1985, Neckles et al. 1993). This method is suitable for many algal morphologies, including film like, crustose and filamentous algae (V an Dijk 1993, Vermaat and Hootsmans 1994, Burt et al. 1995, Brush and Nixon 2002, Stankelis et al. 2003). Submergence is particularly important for algae that elongate from a relatively sm all base because the collapse of the three dimensional structure to a two dimensional structure can lead to overestimation of light attenuation (Brush and Nixon 2002). In fact, Cebrin et al. (1999) found that a given biomass of encrusting red algae attenu ated more light than a similar load of erect, brown algae because the brown algal blades floated and let more light pass. Therefore, the geometric structure of an epiphytic community is an essential factor when consider ing light attenuation

PAGE 38

38 Another consid eration is how samples are processed prior to measurement of light attenuation. A key element in processing is whether light attenuation is measured through the epiphytic load on one or both sides of a leaf. Based on field observations in my study area, I chose to measure light attenuation by epiphytes on one side of V americana leaves because epiphytic loads on the two sides of leaves were substantially different (Figure 2 1 6). Thus, my processing involved removing epiphytes from one side of each leaf bef ore measuring light passing through the remaining layer of epiphytes and the leaf. This approach eliminates the assumption that both sides of each leaf have the same composition and amount of epiphytes. In addition, two layer method relies on a different e quation: where is the amount of light passing through the leaf and two layers of epiphytes and is amount of incident light. This difference arises because two layers of epiphytes generate a multiplicative effect on light attenuation rather than an additive one (Vermaat and Hootsmans 1994). The two layer method may be most suitable for large accumul ations of crustose coralline algae and delicate film like algae. These firmly attached epiphytes generally are removed with acid after transmission of light has been measured. Another methodological consideration is the choice of the metric for characteriz ing epiphytic loads. Because photosynthetic pigments in epiphytes (e.g., chlorophyll a, chlorophyll b, chlorophyll c, fucoxanthin, and phycocyanins) are considered to be the dominant factors that influence light attenuation (Losee and Wetzel 1983, Agust et al 1994), many studies used total chlorophyll a mass per unit area to describe epiphytic loads (Stankelis et al. 1999). In my study, results expressing epiphytic loads as g chlorophyll a cm 2 of leaf yielded the poorest predictions of light

PAGE 39

39 transmissi on as shown by lower R 2 values (Table 2 1 and 2 2). Epiphytic loads described as total dry weight (mg DW cm 2 of leaf) yielded the best predictions, with R 2 values up to 0.83 Even other measures of epiphytic loads (mg AFDW cm 2 and mg Ash DW cm 2 ) yielded better results than those of chlorophyll a. This discrepancy may be a consequence of unpigmented mucous, frustules, calcium carbonate, and trapped detritus that all attenuate light without contributing chlorophyll a (Lin 1995). In my samples, ino rganic matter (ash dry weight) represented up to 77.10% of epiphytic loads. Such large amounts of abiotic matter may become trapped in the epiphytic matrix when sediments a re resuspended by diurnal tidal currents. Thus, useful estimates of epiphytic loads come from dry weights, without the need to extract chlorophyll a or use a muffle furnace. My results indicated that 90% of incident light is attenuated at ~ 7.73 mg DW cm 2 Thus, it is not necessary for epiphytic loads to be conspicuous before they have t he potential to affect the growth of macrophytes. Previous studies reported that 13% of incident light was the average minimal light requirement for survival of freshwater angiosperms (Chamber and Kalff 1985) and marine macroph ytes needed about 11% of incident light (Duarte 1991). Based on these minimum light requirements and my results, I hypothesize d that the critical epiphytic load for V americana is approximately 6 mg DW cm 2 i.e. the level at which transmission of PAR reaches 15% of incident light at a given depth This hypothesis was tested by measuring growth and epiphytic loads for V american a in the Chassahowitzka River.

PAGE 40

40 Table 2 1. Parameter estimates and coefficients of determination (R 2 ) for unconstrained models of light transmission through epiphytes. Two parameter exponential decay Three parameter exponential decay Two parameter hyperbolic decay Three parameter hyperbolic decay Epiphytic loads Equation mg DW cm 2 a=79.61 a=74.22 a=93.31 a=84.93 b=3.63 b=2.78 b=1.63 b=2.61 c=8.36 c=0.44 R 2 0.82 0.83 0.82 0.8 3 AICc 269.44 268.68 270.12 269.78 mg AFDW cm 2 a=73.31 a=67.53 a=83.37 a=78.41 b=2.36 b=1.79 b=1.09 b=1.57 c=8.01 c=0.56 R 2 0.82 0.82 0.82 0.82 AICc 271.21 271.6 270.22 271.27 mg Ash DW cm 2 a=87.71 a=81.14 a=101.05 a=95.63 b=1.25 b=0.84 b=0.6 b=0.82 c=13.05 c=0.6 R 2 0.72 0.33 0.73 0. 7 3 AICc 298.06 351.88 295.84 296.56 g Chl a cm 2 a=53.76 a=63.41 a=74.09 a=835.19 b=29.88 b=4.49 b=9.14 b=0.005 c=25.77 c=2.7 R 2 0.31 0.54 0.42 0.51 AICc 311.1 288.95 300 292.31

PAGE 41

41 Table 2 2. Parameter estimates and coefficients of determination (R 2 ) for constrained models of light transmission through epiphytes Two parameter exponential decay Three parameter exponential decay Two parameter hyperbolic decay Three parameter hyperbolic decay Epiphyte loads Equation mg DW cm 2 a=100 a=84.258 a=100 a=100 b=2.63 b=1.695 b=1.41 b=1.409 c=15.746 c=1.007 R 2 0.73 0.79 0.82 0.82 AICc 295.58 281.91 271.3 273.45 mg AFDW cm 2 a=100 a=79.73 a=100 a=100 b=1.45 b=0.693 b=0.725 b=0.52 c=20.254 c=1.53 R 2 0.60 0.74 0.79 0.81 AICc 319.28 295.64 280.81 277.55 mg Ash DW cm 2 a=100 a=85.97 a=100 a=100 b=1.05 b=0.76 b=0.613 b=0.725 c=14.08 c=0.69 R 2 0.69 0.75 0.73 0.73 AICc 302.49 291.97 295.86 296.83 g Chl a cm 2 a=100 a=73.277 a=100 a=100 b=8.403 b=3.778 b=4.926 b=2.217 c=26.72 c=2.342 R 2 0.31 0.53 0.38 0.50 AICc 337.51 289.887 304.632 293.63

PAGE 42

42 Table 2 3. Candidate models ranked from best to worse based on AICc value, difference values ( ), and Akaike weights ( ). Rank Model AICc 1 Uncon_DW_Three parameter exponential decay 268.68 0.00 2.28E 01 2 Uncon_DW_Two parameter exponential decay 269.44 0.76 1.56E 01 3 Uncon_DW_Three parameter hyperbolic decay 269.78 1.10 1.32E 01 4 Uncon_DW_Two parameter hyperbolic decay 270.12 1.44 1.11E 01 5 Uncon_AFDW_Two parameter hyperbolic decay 270.22 1.54 1.06E 01 6 Uncon_AFDW_Two parameter exponential decay 271.21 2.53 6.45E 02 7 Uncon_AFDW_Three parameter hyperbolic decay 271.27 2.59 6.26E 02 8 Con_DW_Two parameter hyperbolic decay 271.30 2.62 6.17E 02 9 Uncon_AFDW_Three parameter exponential decay 271.60 2.92 5.31E 02 10 Con_DW_Three parameter hyperbolic decay 273.45 4.77 2.10E 02 11 Con_AFDW_Three parameter hyperbolic decay 277.55 8.87 2.71E 03 12 Con_AFDW_Two parameter hyperbolic decay 280.81 12.13 5.31E 04 13 Con_DW_Three parameter exponential decay 281.91 13.23 3.06E 04 14 Uncon_Chl a_Three parameter exponential decay 288.95 20.27 9.06E 06 15 Con_Chl a_Three parameter exponential decay 289.89 21.21 5.67E 06 16 Con_Ash DW_Three parameter exponential decay 291.97 23.29 2.00E 06 17 Uncon_Chl a_Three parameter hyperbolic decay 292.31 23.63 1.69E 06 18 Con_Chl a_Three parameter hyperbolic decay 293.63 24.95 8.73E 07 19 Con_DW_Two parameter exponential decay 295.58 26.90 3.29E 07 20 Con_AFDW_Three parameter exponential decay 295.64 26.96 3.20E 07 21 Uncon_Ash DW_Two parameter hyperbolic decay 295.84 27.16 2.89E 07 22 Con_Ash DW_Two parameter hyperbolic decay 295.86 27.18 2.86E 07 23 Uncon_Ash DW_Three parameter hyperbolic decay 296.56 27.88 2.02E 07 24 Con_Ash DW_Three parameter hyperbolic decay 296.83 28.15 1.76E 07 25 Uncon_Ash DW_Two parameter exponential decay 298.06 29.38 9.53E 08 26 Uncon_Chl a_Two parameter hyperbolic decay 300.00 31.32 3.61E 08 27 Con_Ash DW_Two parameter exponential decay 302.49 33.81 1.04E 08 28 Con_Chl a_Two parameter hyperbolic decay 304.63 35.95 3.56E 09 29 Uncon_Chl a_Two parameter exponential decay 311.10 42.42 1.40E 10 30 Con_AFDW_Two parameter exponential decay 319.28 50.60 2.35E 12 31 Con_Chl a_Two parameter exponential decay 337.51 68.83 2.59E 16 32 Uncon_Ash DW_Three parameter exponential decay 351.88 83.20 1.96E 19

PAGE 43

43 Table 2 4. Comparison of models and key parameters relating light transmission to loads of epiphytes in this and previous studies. Author (Year) Equation Epiphytic loads (mg DW cm 2 of leaf) at 50% light transmission Asymptote % of incident light Present study Two parameter exponential 1.69 10% Brush and Nixon (2002) Three parameter hyperbolic 3.61 15% Three parameter hyperbolic 22.12 30% Three parameter hyperbolic 2.59 18% Frankovich and Zieman(2005) Two parameter exponential 4.36 10% Two parameter hyperbolic 4.27 15% Silberstein et al. (1986) Three parameter exponential 3.47 10% Burt et al. (1995) Two parameter exponential 1.80 20% Stankelis et al. (2003) Three parameter exponential 1.50 0% Van Dijk(1993) Two parameter hyperbolic 1.06 0%

PAGE 44

44 Table 2 5. Comparison of methods in this and previous studies. Sample processing Method of measurement Author ( year ) Study area One/Two layers Suspend/not Light Leaf Method Epiphytes Present study Florida Chass a howitzka Spring Freshwater one underwater natural natural directly Chaetomorpha sp. Brush and Nixon (2002) Rhode Island Mesocosms Marine one underwater natural natural directly Ulothrix sp., Cladophora sp., Polysiphonia sp. Frankovich and Zieman (2005) Florida Bay and Florida Key Marine one & two underwater artificial artificial directly calcium carbonate sediment & corallines, Polysiphonia sp. Burt et al. (1995) Australia Success Bank Marine two underwater artificial artificial directly Rhodophyta, Phaeophyta, Cyanophyta and some coralline algae Silberstein et al. (1986) Australia Cockurn Sound Marine two underwater artificial artificial directly combination of filamentous algae and coralline communities Stankelis et al. (2003) Maryland Patuxent estuary Mesohaline two underwater artificial artificial directly diatoms Van Dijk (1993) Lake Veluwe Freshwater one underwater artificial artificial directly d iatoms and green algae Cebrian et al. (1999) Spanish Cala Jonquet Marine one dry artificial artificial indirectly red encrusting algae, brown erect algae, bryozoa and hydrozoa

PAGE 45

45 Figure 2 1. L head springs in the Chassahowitzka River, Citrus County Florida The sampling area is a large continuous Vallisneria americana meadow (bright green area) that covers approximately 2700 m 2 and is located about 1.8 km west of the head springs. Reprinted with permission from Google Earth, https://www.google.com/earth/ (November 18, 2017) A B C Figure 2 2. A pparatus used to measure attenuation of natural photosynthetically active radiation (PAR) by epiphytes A) A LI 1400 datalogger B ) T wo, 2 underwater quantum sensors (LI COR UWQ5754 and LI COR UWQ 5692). C ) The apparatus was submerged in an outdoor tank filled with freshwater. Photo s courtesy of author.

PAGE 46

46 Figure 2 3. Schematic representation of measuring the effect of epiphytes on l ight penetration. Before measuring, one side of each leaf section was scraped to remove epiphytes. E ach scraped section was placed on an underwater sensor ( No. 1 ) When the uncovered sensor ( No. 2 ) indicated that incident irradiance was stable, irradiation passing through both the blade section and attached epiphytes was recorded. Next, the epiphytes on the other side of each section were removed and saved for determination of epiphytic load per unit area. The resulting clean sections were placed over the a ppropriate underwater sensor, and light passing through the blade was measured. A B C Figure 2 4. The dominant green filamentous macroalgae Enteromorpha sp. in my study site A ) Characteristic distribution of Enteromorpha sp. B ) Enteromorpha sp. branching filament at 100x magnification C ) Enteromorpha sp. branching filament at 400x magnification Photo s courtesy of author.

PAGE 47

47 A B Figure 2 5. Other epiphyte s in my study site A) Cladophora sp. branching filament at 4 00x magnification. B ) U biquitous diatoms at 400x magnification Photo s courtesy of author.

PAGE 48

48 A B Figure 2 6. Scatter plots showing the relationships between epiphytic loads and light transmission. Light transmission (I) expressed as percentage of incident irradiance (I 0 ). A) Epiphytic loads expressed as mg dry weight cm 2 of leaf (mg DW cm 2 ). B) Epiphytic loads expressed as mg ash free dry weight cm 2 of leaf (mg AFDW cm 2 ). C) Epiphytic loads expressed as mg ash dry weight cm 2 of leaf (mg ash DW cm 2 ). D) Epiphyti c loads expressed as g Chlorophyll a cm 2 of leaf.

PAGE 49

49 C D Figure 2 6. Co ntinued.

PAGE 50

50 Figure 2 7. Comparison of relationship s between epiphytic loads as mg dry weight cm 2 of leaf, and light transmission, in previous studies Reproduced with permission from Brush and Nixon ( 2002).

PAGE 51

51 A B Figure 2 8. Decay equations fit to the relationships between total epiphytic loads (mg dry weight cm 2 of leaf) and light transmission. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three parameter hyperbolic equation. Parameter estimates for regression equations are listed

PAGE 52

52 C D Figure 2 8. Continued

PAGE 53

53 A B Figure 2 9. Decay equations fit to the relationships between total epiphytic loads (mg ash free dry weight cm 2 of leaf) and light transmission. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three parameter hyperbolic equation. Parameter estimates for regression equations are listed.

PAGE 54

54 C D Figure 2 9. Continued

PAGE 55

55 A B Figure 2 10. Decay equations fit to the relationships between total epiphytic loads (mg ash dry weight cm 2 of leaf) and light transmission. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three parameter hyperbolic equation. Parameter estimates for regression equations are listed.

PAGE 56

56 C D Figure 2 10. C oun tinued

PAGE 57

57 A B Figure 2 11. Decay equations fit to the relationships between total epiphytic loads (g chlorophyll a cm 2 of leaf) and light transmission. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three parameter hyperbolic equation. Parameter estimates for regression equations are listed.

PAGE 58

58 C D Figure 2 11. Continued.

PAGE 59

59 A B Figure 2 12. Decay equations fit to the relationships between total epiphytic loads (mg dry weight cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three p arameter hyperbolic equation. Parameter estimates for regression equations are listed.

PAGE 60

60 C D Figure 2 12. Continued

PAGE 61

61 A B Figure 2 13. Decay equations fit to the relationships between total epiphytic loads (mg ash free dry weight cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three p arameter hyperbolic equation. Parameter estimates for regression equations are listed.

PAGE 62

62 C D Figure 2 13. Continued

PAGE 63

63 A B Figure 2 14. Decay equations fit to the relationships between total epiphytic loads (mg ash dry weight cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three p arameter hyperbolic equation. Parameter estimates for regression equations are listed.

PAGE 64

64 C D Figure 2 14. Continued.

PAGE 65

65 A B Figure 2 15. Decay equations fit to the relationships between total epiphytic loads (g chlorophyll a cm 2 of leaf) and light transmission that have been constrained to pass through 100% light transmission at epiphytic loads of zero. A) Two parameter exponential equation. B) Three parameter exponential equation. C) Two parameter hyperbolic equation. D) Three p arameter hyperbolic equation. Parameter estimates for regression equations are listed

PAGE 66

66 C D Figure 2 15. Continued.

PAGE 67

67 Figure 2 16. Leaf from the Chassahowitzka river showing heterogeneous epiphytic loads on its two sides Photo courtesy of author.

PAGE 68

68 CHAPTER 3 THE EFFECTS OF EPIPHYTES ON THE GROWTH OF VALLISNERIA AMERICANA Background Submersed macrophytes are crucial components of aquatic systems because they stabilize sediments and reduce turbidity, absorb and store nutrients, sequester carbon and provide refuge and foraging habitat for numerous aquatic organisms (Van Donk and van de Bund 2002, Gregg and Rose 1985). However, widespread reduction in the abundance of macrophytes has been observed since the 1900s as a consequence of eutrophicatio n caused by increased nutrient loads, reduced flows, and other factors (Seddon et al. 2000, Waycott et al. 2009). In particular, reductions in available light caused by suspended particles, phytoplankton blooms and growth of epiphytic algae are blamed for losses of macrophytes (Silberstein et al. 1986, Lauridsen et al. 1994). The influence of epiphytes on macrophytes has been demonstrated in several studies (Sand Jensen 1977, Bulthuis and Woelkerling 1983, Silberstein et al. 1986). Epiphytes inhibit perform ance of macrophytes in several ways, including shading leaves, increasing the boundary layer around leaves leading to slower exchange of carbon dioxide (CO 2 ) and oxygen (O 2 ), and competing for nutrients (Sand Jensen et al. 1985, Borowitzka et al. 2006). F or example, heavy loads of epiphytes on Ruppia cirrhosa caused decreased production and earlier seasonal dieback (Kirbe 1980). Epiphytic colonization of Potamogeton pectinatus was linked to cuticular erosion and peeling (Howard Williams et al. 1978), and epiphytes on Potamogeton crispus caused swelling and disorganization of epidermal cells (Rogers and Breen 1981). Fong et

PAGE 69

69 al. (2000) reported early leaf death and reduction of leaf flexibility in Zosteroids and Thalassia under heavy loads of epiphytes Epiphytes depress the productivity and reproduction of macrophytes mainly through diminishing the amount of light that reaches the leaves, which decreases photosynthesis (Orth and Van Montfrans 1984, Neckles et al. 1993, Drake et al. 2003). Many studies have confirmed that epiphytes can reduce light availability dramatically (Dennison et al. 1993, Vermaat et al. 1993, Czerny and D unton 1995, Fitzpatrick and Kirkman 1995, Kurtz et al. 2003), and my measurements of epiphytic light attenuation in Chapter 2 indicated that an epiphytic load of about 7.7 mg DW cm 2 can attenuate 90% of incident light. Most of the literature describing e piphyte macrophyte interactions has focused on brackish or marine environments. A few studies on relationships in freshwater systems were conducted in lakes (Phillips et al. 1978, Sand Jensen and Sndergaard 1981, Cattaneo et al. 1998) or in the laborator y (Asaeda et al. 2004). However, the influence of epiphytic loads on the growth and production of macrophytes in knowledge I conducted fieldwork in the spring fed Chassahowitzka River where a documented increase in epiphytes on macrophytes was temporally concordant with losses of important macrophytes, including V. americana (Notestein 2001). Vallisneria americana is the dominant macrophyte in the Chassahowitzka River, and it is dioec ious and perennial (Notestein 2001). These stoloniferous plants form expansive meadows via clonal extension. They can survive in fresh to mesohaline waters, and they grow in aquatic systems from Central America to Canada (Korschgen

PAGE 70

70 and Green 1988). Vallisneria americana has ribbon like leaves that reach up to 2 m or more in length depending on water movement and depth (Doust and LaPorte 1991). The leaves arise in a cluster from a short vertical stem that sends out rhizomes and stolons from which new shoots develop. Unbranched and fibrous roots are found at the base of each rosette (McFarland 2006). Knowing how loads of epiphytes affect the growth of macrophytes is important for protecting, restoring and identifying the threshold for detrimental epiphytic loads could serve as a valuable index of the health of a system Toward this end, this investigation measured growth rates for V. americana with different loads of epiphytes in the Chassahowitzka River. T his study was a further test of the conclusion from Chapter 2, which indicated that epiphytic loads of approximately 6 mg DW cm 2 of leaf are sufficient to cause detrimental effects on macrophytes, based on minimum light requirements derived from previous studies (Chamber and Kalff 1985, Duarte 1991). Materials and Methods Study Sites Measurements were made within large, continuous meadows of V. americana in were about 2 k m west of the head springs of the Chassahowitzka River (Figure 3 1). These meadows covered about 240 m 2 and 30 m 2 in Salt Creek and Small Creek, respectively (Figures 3 2 A and 3 2 B ). Both meadows consisted of monospecific stands of V. americana with epiphy tic algae on their leaves and little drift algae. The creeks were too shallow for motor boats and not near the main river so they were rarely disturbed. In the two areas, seven sets of measurements were collected from

PAGE 71

71 June 2016 to August 2016, with the f irst four sets occurring in Salt Creek and the last three sets taking place in Small Creek after manatees grazed the V. americana in Salt Creek. Small Creek was shallower, narrower and more shaded by riparian vegetation (Figure 3 2 B ), with a less dense co ver of shorter V. americana Environmental Measurements During each visit, depth, water temperature, pH, concentration of dissolved oxygen (DO), salinity, light attenuation and incident solar radiation were measured. Depth was measured to the nearest 0.1 m 3 A ). Temperature, salinity, pH, and DO were measured with a YSI model 650MDS meter and were recorded to the nearest 0.1C, 0.01, 0.01, and 0.1 mg O 2 L 1 To calculate light attenuation coefficients ( ), two quantum ligh t sensors (Li Cor Instruments Inc.) w ere used with a data logger to simultaneously measure photosynthetically active radiation (PAR, E m 2 s 1 ) at the surface and at one or more depths below the surface (Figure 3 3 A and B ). Values of were cal culated using the equation: where is the radiation at the surface and is the radiation at depth (Z, Kirk 1994). If the water was deeper than 1 m, measurements were made at three depths (Z = 0.5, 0.75, 1 m), and in shallow wa ter, three replicate measurements were recorded at Z = 0.5 m. Measurements were made at 10 fixed locations uniformly distributed in the meadows and marked with PVC pipes (Figure 3 2), and readings were not corrected for sun angle. Biomass of Vallisneria am ericana and Epiphytes Shoot densities and biomass of V. americana and biomass of the associated epiphytes within the study meadows were quantified within five, haphazardly tossed quadrats (0.25 m X 0.25 m). A snorkeler counted the number of shoots in the quadrats

PAGE 72

72 and collected six shoots to yield 30 samples for estimates of biomass of V. americana and epiphytes. Samples were stored in labeled plastic bags on ice during transport to the laboratory wh ere they were frozen until processing. In the laboratory, biomass es of shoots and the epiphytic algae on them were measured. After counting the number of leaves per shoot and measuring the heights and maximum widths of leaves, each leaf was placed under a transparent sheet of plastic marked with a 1 cm grid. The number of grid cells within the outline of the leaf were counted to yield the surface area of the leaf, that is, leaf area equaled the number of cells multiplied by 1 cm 2 Leaf area allowed me to ex press epiphytic biomass per unit area. To account for the heterogeneous distribution of epiphytes on V americana leaves, leaves were separated into 10 cm sections before epiphytes were gently scraped off and saved in individual plastic weighing boats that had been pre weighed and labeled. The clean leaves and the epiphytes were dried at 60C to a constant weight as measured to the nearest 0.001 with an electronic balance. A real biomass (g DW m 2 ) for V. americana and its epiphytes was calculated by multiplying the mean dry weight of V. americana shoots or the mean dry weight of epiphytes on shoots by the mean density of shoots Measurements of Vallis n eria americana Growth and Epiphytic Loads The growth of V. americana was documented w ith a modified leaf marking technique commonly used for submersed macrophytes with wide leaves and a basal meristem (Figure 3 8 Odum 1957, Zieman 1968, Zieman and Wetzel 1980, Hauxwell et al. 2007). This technique was selected instead of the oxygen exchan ge method or incorporation of 14 C isotopes because of variation induced by oxygen ( O 2 ) lacunal system, recycling of O 2 transport and release of O 2 the theoretical and practical

PAGE 73

73 challenges associated with use of 14 C, and the fact that these other techniques would not yield independent estimates of photosynthesis for epiphytes and macrophytes (Westlake 1978, Capone et al. 1979, Kelly et al. 1981, Kremer 1981, Ramus 1981, Lindeboom and De Bree 1982, Sand Jensen et al. 1982). At the beginning of a sampling period, a quadrat (0.25 m by 0.25 m) was tossed haphazardly five times within the selected meadow and six shoots in each quadrat were identified with a pink flag on a stake and a buoy that marked the area (Figure 3 4 ). All leaves in each tagged shoot were marked by carefully punching two holes with a syringe needle (18 gauge) approximately 3 cm above the rhizome (Hauxwell et al. 2007). Tagged shoots were retrieved one to two weeks after punching, at which time new shoots were tagged and punched. Shoots and their associated belowground tissue were collected by hand, rinsed in ambient water and placed in labeled plastic bags. All samples were stored on ice in a cooler during transport to the laboratory and frozen until they were processed. Freezing for 12 h or more made the epiphytes easier to remove. After thawing, each leaf was gently scraped with a scalpel, and the epiphytes were saved. Leaves comprising each shoot were separated and ranked by age from senescent to new before their total lengths and maximum widths were measured (Figure 3 5 ). Leaf areas also were estimated with a grid ded sheet For a given shoot, senescent leaves are located at the outside of the bundle, and they bore holes that were at the same height as they were initially punche d because these leaves did not grow. These marks served as reference points for measuring growth of younger l eaves (Figure 3 6 ). New growth for younger leaves was identified as the material between their holes and the reference point. All leaves without ho les were

PAGE 74

74 considered new growth. Epiphytes, new growth and old leaf tissue were placed in separate pre weighed, labeled plastic weighing boats and dried at 60C to a constant weight as determined to the nearest 0.0 01 g with an electronic balance (Figure 3 7 ). Growth of V. americana shoots (R) was calculated as in cm 2 per day or mg DW per day where G is the quantity of new material (cm 2 or mg dry weight) and N is the number of days between pu nching and retrieval (Figure 3 8 ). Determination of the Plastochrone Interval (PI) and Leaf Age for Vallisneria americana Age is another influence on the growth of leaves of macrophytes, since many important physiological processes change with age ( Cebrian and Duarte 1994), including photosynthesis (Mazzell a and Alberte 1986) and synthesis of proteins (Thayer et al. 1984, Zieman et al. 1984). Thus, I sought to estimate the age of leaves. Conventionally, leaf age is expressed as leaf rank (Patriquin 1973, Ott 1980), but Duarte (1991) found that leaves with th e same rank may differ in age by 1.1 to 47.2 d ays To avoid such errors, leaf age was estimated with the method used by Erickson and Michellini (1957), which converts leaf ranks into absolute age in days using the Plastochron e Interval (PI, Askenasy 1880, Lamoreaux et al. 1978). Plastochrone intervals (PI, day per leaf) for V. americana were derived from observations of newly emerged and unmarked leaves in shoots ( Duarte et al. 1994). The interval represents the time between the appearance of two consecuti ve leaves, and it is calculated as: ; where Time Interval is the number of days between punching and retrieval for shoots with at least one new leaf.

PAGE 75

75 Once the plastochr one interval for V. americana was known, estimates of leaf age in days could be calculated for all leaves (Erickson and Michellini 1957). These calculations involved a Plastochron e Index (PI Index): ; where R is the rank of a given leaf among all the leaves in a shoot arranged from youngest to oldest, represents the fractional age of the youngest leaf (which is < 1), and are the lengths of the two youngest leaves in the shoot (ranks 1 and 2), and is the reference length or the length at which a new leaf can be detected. Estimates of leaf age in days were made by multiplying the PI Index by the Plastochrone Interval. Statistical Analyses Regression analyses were used to examine the relationships be tween new growth per shoot per week and epiphytic loads expressed as total loads per shoot or loads per unit area. These analyses were performed using Microsoft Excel. A boundary analysis was employed to identify epiphytic loads that impeded the growth of V. americana (Ludwig and Tongway 1995). The analysis involved (1) ranking growth rates according to epiphytic loads from highest to lowest, (2) bracketing contiguous sets of growth rates in a window of preassigned width (w = 4 for Salt Creek and 6 for Sma ll Creek), (3) splitting each window into two equal groups ( ), (4) averaging the growth rates in the two groups (H 1 and H 2 ) (5) computing a dissimilarity index as squared Euclidean distance (SED) = (6) moving the window one position further along the ordered series of growth rates, (7) computing another dissimilarity index, (8) repeating this process for the entire dataset (Figure 3 9 ), and (9) plotting the dissimilarity indices against loads of epiphytes

PAGE 76

76 Thresholds for the effect of epiphytic loads were identified as the first large increase in the dissimilarity indices. As a test of the thresholds, growth rates below the threshold test. Resu lts Environmental Data The environmental conditions at these study sites were similar to those documented in Chapter 2, which was about 294 m away. During my sampling, water temperatures ranged from 25.9C to 31.1C Average daytime dissolved oxygen concen trations (DO) were 8.84 mg L 1 The tide and led to variation in attenuation coefficients ( ), with the mean being 3.27. Incident light at the two sites differed, with the mean and maximum in Small Creek 26.43% and 53.57% respectively, of values in Salt Creek at noon. In situ Biomass of Vallisneria americana and Epiphytes In Salt Creek, the areal biomass of V. americana was 143 g DW m 2 as estimated from mean shoot dry weight and mean shoot density ( 182 shoots m 2 ). The areal biomass in Small Creek was 37.35 g DW m 2 with a mean shoot density of 86 shoots m 2 New leaves appeared in shoots every 5.25 0.63 days, i.e., about every 5 days Loads of epiphytes per shoot varied, with a range of 10.9 to 15 211 mg DW per plant (0.08 to 32.37 mg DW cm 2 ), and an overall mean standard deviation of 1 73 5 ( 2 55 6 ) mg DW per plant (6.66 7.26 mg DW cm 2 ). Epiphytic loads accumulated near the ti ps of leaves, and sections between the base of the shoots and 7 cm wer e free of epiphytes (Figure 3 10 ). Mean epiphytic loads per leaf differed with leaf age, with an increase to a maximum value of 6.86 mg DW cm 2 on 40 day old

PAGE 77

77 leaves and a decre ase on older leaves (Figure 3 11 ). Epiphytes were not present on all leaves younger than 3 days 85.71% of leaves aged 3 to 9 days, and 22.63% of leave s aged 9 to 15 days (Figure 3 11 ). Effect of Leaf Age on Growth of Vallisneria americana The reference length ( ) was set as 1.8 cm because it was the length of the shortest new leaf. Based on estimated ages, leaves were grouped into four age classes: young (< 20 days), early ( 20 29 days), mature ( 30 39 days), and senescent ( 40 days). Of all leaves sampled in Sma ll Creek and Salt Creek, respectively, young leaves represented 26.46% and 35.11%, early leaves represented 17.95% and 25.29%, mature leaves represented 17.95% and 21.96%, and senescent leaves represented 37.64% and 17.64% (Figure 3 12) New growth was re duced in leaves older than 30 days and essentially absent in the few leaves older than 40 days (Figure 3 13 ). The largest amounts of growth were 22.92 cm 2 per week in Small Creek and 30.24 cm 2 per week in Salt Creek, which were recorded for leaves that we re 15 days old (Figure 3 13 ). Maximum values for new growth increased up to 15 days and reached approximately zero after 40 days (Figure 3 14 ). Effect of Epiphytic Load on growth of Vallisneria americana F ieldwork provided seven sets of measurements representing a total of 134 shoots and 1236 leaves. Linear regressions yielded low coefficients of determination (R 2 ) regardless of whether epiphytic loads were standardized to unit area (Figures 3 15 and 3 16 ). Nevertheless, growth expressed as gain in dry weight or surface area was negatively correlated with epiphytic loads in Salt Creek and positively correlated with epiphytic loads in Small Creek.

PAGE 78

78 Given the variation in growth related to the age of leaves, multiple regressions wer e employed to examine relationships of growth to leaf age (X 1 ), epiphytic load (X 2 ), and the interaction of leaf age and epiphytic load (X 1 x X 2 ). Growth was negatively correlated with both leaf age and epiphytic load in both Salt and Small creeks (Tables 3 1 and 3 2 ). Estimation of a Threshold for Detrimental Epiphytic Load Threshold epiphytic loads that inhibited growth of V. americana were estimated for leaves in different age classes in Salt Creek and Small Creek. Lack of epiphytes on young leaves or l ack of growth for mature leaves restricted these analyses to two age classes in each creek (Figures 3 17 and 3 18 ). For Small Creek, boundary analysis using a window of six datapoints indicated thresholds at ~ 0.4 mg DW cm 2 for early leaves ( 20 29 days) a nd between 0.85 and 0.79 mg DW cm 2 for mature leaves ( 30 39 days, Figures 3 17 A and 3 17 B ). For Salt Creek, boundary analysis using a window of four datapoints indicated thresholds between 3.84 and 4.76 mg DW cm 2 for young leaves (< 20 days) and between 4.92 and 5.23 mg DW cm 2 for early leaves ( 20 29 days, Figures 3 18 A and 3 18 B ). These thresholds were confirmed by significant tests (Table 3 3 ). Discussion My environmental data (temperat ure, DO, salinity, and ) fell within the ranges of data from two transects (5 and 6) that were surveye d from 1998 to 20 05 ( Frazer 2000, Frazer et al. 2001 Frazer et al. 2006 ). My temperatures, salinities and values were above the average values because my sampling always occurred during low tides.

PAGE 79

79 My measures of biomass for V americana and its epiphytes illustrated changes in the Chassahowitzka system since previous sampling ( Frazer 2000, Frazer et al. 2001 Frazer et al. 2006 ). During my sampling, maximum biomass of V. americana (143 g DW m 2 ) was about 10 % of the historic average (1295.53 189.68 g DW m 2 ) and close to the historic low. In contrast, my mean epiphytic biomass was twice that of the historic average (2.25 0.21 g Chl a g 1 DW host plant), and my maximum epiphyte load was 26.06 g Chl a g 1 DW host plant as compared to the historic maximum of 6.53 g Chl a g 1 DW host plant. These differences suggest degradation of V. americana due to increa sed epiphytic loads. Furthermore, average shoot biomass in Salt Creek was higher than values recorded in Small Creek. This difference probably arises from differences in incident solar radiation, which was 75% lower in Small Creek. Similar differences were reported for V. americana by Blanch et al. (1998) in the River Murray in South Australia where light regimes were affected by turbidity, and in a shading experiment conducted in Perdido Bay at the border between Alabama and Florida (Kurtz et al. 2003). Numbers of new leaves per shoot and total leaves per shoot in Small Creek were higher than values in Salt Creek. This result supports the conclusion that low light availability leads to production of new leaves in V. americana which was reported by Kimber et al. ( 1995 ) and Muenscher (1936). Additional leaves per shoot increase photosynthetic production in Small Creek where less light is available. Epiphytic loads were higher in Salt Creek, and it has been shown that light has a strong positive effect on t he growth of epiphytes (Burnell et al. 2014). Epiphytic loads on V. americana initially increased with leaf age and then declined after 40 days

PAGE 80

80 (Figure 3 11 ), which agrees with observations for seagrass species (Borum 1987, Hootsmans and Vermaat 1991). T he increase in epiphytic biomass indicates that epiphytic growth exceeds losses from grazing or sloughing, and the presence of a maximum load points to a balance between growth and loss at about 40 days (Borum 1987), with the subsequent decrease likely ca used by loss of older and more heavily epiphytized sections. Epiphytes tend to increase near the tips of leaves (Figure 3 10 ), which are older and exposed to colonization and growth for a longer time Borowitzka et al. (1990) found similar results for the seagrass Amphibolis griffithii Growth rates for leaves varied between Salt Creek and Small Creek primarily as a consequence of the amount of incident light, but leaf age and epiphytic loads influenced growth rates at both sites. Leaf age appears to limit overall growth potential, and epiphytic loads above a certain threshold essentially reduce growth to zero, with thresholds varying according to the age of leaves. Young leaves (< 20 day s) in Salt Creek stopped growing at epiphytic loads of ~ 4.3 mg DW cm 2 and substituting this load into the unconstrained two parameter exponential decay equation from Chapter 2 ( ) indicated that the leaves were receiving 24.35% of incident light. Growth of early leaves ( 20 29 days) in Salt Creek ceased at ~ 5.08 mg DW cm 2 which equates to 19.67% of incident light. In Small Creek where incident light was 75% less, thresholds for early leaves (20 days < le af age < 30 days) and mature leaves (30 days < leaf age< 40 days) were 0.4 and 0.8 mg DW cm 2 respectively, which equates to 17.83% and 15.97% of available incident light. Overall, the threshold epiphytic load that limited growth of V. americana in the Chassahow i tzka River was between 4 and 5 mg DW cm 2 I n other words, the minimum

PAGE 81

81 light requirement was about 22% of incident light In Chapter 2, the critical threshold for epiphytic loads was estimated to be 6 mg DW cm 2 base d on values for light requirements drawn from the literature. Results based on measurements of growth in the field were < 6 mg DW cm 2 in part because of light attenuation by the overlying water column. My results were similar to previous reports for seagr asses near the Homosassa and Weeki Wachee rivers where sites that had a median light penetration > 20% of incident light supported the most abundant and diverse beds of seagrass (Choice et al. 2014). To my knowledge, this is the first study to quantify a threshold for the influence of epiphytes on growth of freshwater macrophytes using in situ measurements. The results can be applied to improve management of water quality, light attenuation and the health of macrophytes, which are critical components of Fl fed aquatic systems.

PAGE 82

82 Table 3 1. Statistics for multiple regression involving epiphytic loads, leaf ages and their interaction for Salt Creek. ANOVA df SS MS F Significance F R 2 Adjusted R 2 Term Coefficient t value P value Regression 3 9038 3013 128 7.32E 64 0.39 0.39 Intercept 10.57 25.08 2.47E 95 Residual 597 14096 24 Epiphytic load 0.23 14.64 1.03E 41 Total 600 23133 Leaf age 0.57 8.93 5.11E 18 Interaction 0.01 7.43 3.75E 13

PAGE 83

83 Table 3 2. Statistics for multiple regression involving epiphytic loads, leaf ages and their interaction for Small Creek. ANOVA df SS MS F Significance F R 2 Adjusted R 2 Term Coefficient t value P value Regression 3 5727 1909 134 4.26E 67 0.39 0.39 Intercept 8.41 26.58 5.19E 105 Residual 631 9001 14 Epiphytic load 0.15 15.84 7.71E 48 Total 634 14727 Leaf age 4.58 6.63 7.33E 11 Interaction 0.08 6.58 1.02E 10

PAGE 84

84 Table 3 3. Estimated threshold of epiphytic loads that prevent the growth of Vallisner i a americana tests assessed differences in mean epiphytic loads below and above each threshold. Creek Age Epiphytic loads (mg dry weight cm 2 ) Threshold Values versus threshold Mean SE P value Small Early 0.40 0.41 below 0.34 0.00 2.08E 03 above 0.61 0.04 Mature 0.79 0.85 below 0.57 0.01 1.14E 07 above 1.32 0.25 Salt Young 3.84 4.76 below 1.27 0.06 8.07E 03 above 13.83 52.28 Early 4.92 5.23 below 2.74 1.35 1.06E 18 above 16.13 70.20

PAGE 85

85 Figure 3 1. Location of study areas in Both areas were about 2 km west of head springs of the Chassahowitzka River, Florida Reprinted with permission from Google Earth, https://www.google.com/earth/ (November 18, 2017). A B Figure 3 2. Study areas in the Chassahowitzka River. A) The area in Salt Creek was wide (about 32 m) with less shade from riparian vegetation B ) The area in Small Creek was shallow narrow (about 5 m) and shaded by riparian vegetation Anchored PVC bars evenly distri buted in the two areas as fixed spots for measurement s of irradiation and light attenuation Photos courtesy of author.

PAGE 86

86 A B Figure 3 3. Equipment used to ta ke measurements in the field. A ) an underwater quantum light sensor to measure water depth and light attenuation coefficient ( ). B ) A n under water quantum light sensor with a data logger for measuring and recording photosynthetically active radiation (PAR, E m 2 s 1 ). Photos courtesy of author. Figure 3 4 Depiction of sampling events. During each sampling event a snorkeler punched two holes through the leaves of 30 shoots of V allisneria americana approximately 3 cm above the ir rhizome s and they were marked with pink flags and a buoy. Photos courtesy of author.

PAGE 87

87 Figure 3 5 Typical Vallisneria americana shoot during processing. Leaves on each shoot were separated and ranked by age from senescent leaves ( right ) to new leaves ( left ) before their total length s and width s of leaves were measured Photo courtesy of author.

PAGE 88

88 Figure 3 6 Holes i n leaves used to measure growth. The scars on senescent leaves served as reference point s for measuring growth of younger leaves. Photos courtesy of author. Figure 3 7 Labeled, pre weighed boats containing samples for processing. Dry weight s were determined for epiphytes, new growth and original material Photo courtesy of author.

PAGE 89

89 Figure 3 8 Schematic of modified leaf marking technique Modified from Short and Duarte (2001) Figure 3 9 Schematic of the moving split window technique. In a window eight datapoints wide a dissimilarity value was calculated using the means o f the four datapoints in the re d and green frames before the window was moved one datapoint and a new index was calculated.

PAGE 90

90 Figure 3 1 0 Distribution of epiphytic loads along the leaves of V allisneria americana Orange points represent the epiphytic loads on different sections of leaves, blue points and bars represent the means and standard error s for epiphytic loads on consecutive 5 cm long leaf sections. Blue dotted line depicts the trend. y = 0.1384x 1.8317 R = 0.9374 -2 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 70 80 90 Epiphytic loads (mg DW cm 2 ) Distance from leaf basal part (cm) Mean of epiphytic loads Epiphytic loads Linear (Mean of epiphytic loads)

PAGE 91

91 Figure 3 11 The relationship between epiphytic load and leaf age for V allisneria americana Orange points represent the epiphytic loads on leaves with different ages, blue points and bars represent the means and standard error s for epiphytic loads Blue line depicts the trend. 0 2 4 6 8 10 12 14 16 18 20 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 80 90 Epiphytic loads (mg DW cm 2 ) Leaf age (Days) Epiphytic densities Mean Epiphytic densities

PAGE 92

92 Figure 3 12 Percentage of V allisneria americana leaves in different age classes (days) in Small Creek (left) and Salt Creek (right). 15.0 10.0 5.0 0.0 5.0 10.0 15.0 0~3 3~9 9~15 15~21 21~27 27~33 33~39 39~45 45~51 51~57 57~63 63~69 69~75 75~81 Percentage of total leaves (%) Leaf age (days) Salt Creek Small Creek

PAGE 93

93 Figure 3 13 Growth of leaves of V allisneria americana (cm 2 per week) of different ages (days), in Small Creek (blue triangles) and Salt Creek (orange circles). 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60 Leaf new growth (cm 2 ) per week Leaf age(days) Small Creek Salt Creek

PAGE 94

94 Figure 3 14 Maximum growth of Vallisneria americana leaves (cm 2 per week ) of different ages (days) in Small Creek (blue) and Salt Creek (orange). -5 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60 Maximum growth (cm 2 per week) Leaf age(days) Small Creek Salt Creek

PAGE 95

95 A B C D Figure 3 15 Relationships and trends for epiphytic loads and growth of Vallisneria americana shoots in Salt Creek. A) and B) Epiphytic loads expressed as mg dry weight (DW). C) and (D) Epiphytic loads expressed as mg DW cm 2 A) and C) Growth expressed as mg DW per week B) and D ) Growth expressed as cm 2 per week. y = 0.0017x + 65.539 R = 0.0225 0 50 100 150 200 250 0 5000 10000 15000 20000 New growth per shoot (mg DW per week ) Epiphytic loads per shoot (mg DW) Salt Creek y = 0.0011x + 32.276 R = 0.0364 0 10 20 30 40 50 60 70 80 90 100 0 5000 10000 15000 20000 New growth per shoot (cm 2 per week ) Epiphytic loads per shoot (mg DW) Salt Creek y = 1.4858x + 75.407 R = 0.0995 0 50 100 150 200 250 0 10 20 30 40 New growth per shoot (mg DW per week ) Epiphytic loads per shoot (mg DW cm 2 ) Salt Creek y = 0.8344x + 37.385 R = 0.1206 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 New growth per shoot ( cm 2 per week ) Epiphytic loads per shoot (mg DW cm 2 ) Salt Creek

PAGE 96

96 A B C D Figure 3 16 Relationships and trends for epiphytic loads and growth of Vallisneria americana shoots in Small Creek. A ) a nd B ) Epiphytic loads expressed as mg dry weight (DW) C) and D ) Epiphytic loads expressed as mg DW cm 2 A) and C) Growth expressed as mg DW per we ek B) and D) Growth expressed as cm 2 per week. y = 0.0906x + 43.033 R = 0.2132 0 20 40 60 80 100 120 140 0 200 400 600 800 New growth per shoot (mg DW per week ) Epiphytic loads per shoot (mg DW) Small Creek y = 0.0457x + 26.789 R = 0.1648 0 10 20 30 40 50 60 70 80 0 200 400 600 800 New growth per shoot (cm 2 per week ) Epiphytic loads per shoot (mg DW) Small Creek y = 17.568x + 47.559 R = 0.0666 0 20 40 60 80 100 120 140 0 1 2 3 New growth per shoot (mg DW per week ) Epiphytic loads per shoot (mg DW cm 2 ) Small Creek y = 8.9422x + 29.035 R = 0.0523 0 10 20 30 40 50 60 70 80 0 1 2 3 New growth per shoot (cm 2 per week ) Epiphytic loads per shoot (mg DW cm 2 ) Small Creek

PAGE 97

97 A B Figure 3 17 Squared Euclidean distances (SED) versus epiphytic loads in Small Creek. A) P lants in the early age class B) P lants in the mature age class. Moving window (W) included six datapoints. 0 10 20 30 40 50 60 70 0 0.2 0.4 0.6 0.8 1 1.2 SED Epiphytic loads (mg DW cm 2 ) Early group in Small Creek (W=6) 0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 SED Epiphytic loads (mg DW cm 2 ) Mature group in Small Creek (W=6)

PAGE 98

98 A B Figure 3 18 Squared Euclidean distances (SED) versus epiphytic loads in Salt Creek. A) P lan ts in the young age class. B) P lants in the early age class. Moving window (W) included four datapoints. 0 20 40 60 80 100 120 140 0 5 10 15 20 25 SED Epiphytic loads (mg DW cm 2 ) Young group in Salt Creek (W=4) 0 20 40 60 80 100 120 140 160 180 0 10 20 30 40 SED Epiphytic loads (mg DW cm 2 ) Early group in Salt Creek (W=4)

PAGE 99

99 CHAPTER 4 A MODEL OF VALLISNERIA AMERICANA GROWTH UNDER DIFFERENT EPIPHYTIC LOADS Background Meadows of submersed macrophytes are found in many aquatic systems around the world. They are major contributors to primary production, provide nursery habitats for aquatic organisms (Orth and V an Montfrans 1984), stabilize sediments, and assimilate and store nutrients (Oshima et al. 1999). In recent decades, proliferation of nuisance algae has seriously threatened the survival of macrophytes (Silberstein et fed sy stems, temporal concordance between increased epiphytic loads and loss of submersed macrophytes has been widely observed, with the Chassahowitzka River being a prime example (Notestein 2001). In the Chassahowitzka River, Vallisneria americana meadows repre sent an important and formerly extensive habitat (Frazer et al. 2001). The leaves of V. americana support many epiphytes, which can be responsible for up to 20 30% of the total primary production (Notestein 2001). Evidence suggests that epiphytic loads h ave several detrimental impacts on the productivity and growth of the host plant, especially through competition for available light and nutrients (Mazzella and Ott 1984, Dunn et al. 2008, Frankovich and Zieman 2005, Orth and Moore 1983). In addition, epip hytes can increase drag on leaves of macrophytes causing shear stress and loss of biomass (Doyle 2001). A primary limitation on the distribution and productivity of macrophytes, and other photosynthetic organism s is the amount of light available to support metabolism and growth ( Spence 1976, Chambers and Kalff 1987, Duarte 1991, Lorenti et al. 1995, Pirc 1985). In some aquatic systems, strong seasonal variations in sunlight,

PAGE 100

100 temperature, and precipitation produce seasonality in the growth of macroph ytes (Zimmerman et al. 1994). I fed systems, however, the major influence is sunlight, a consequence of the stabilizing effect of the groundwater that maintains a relatively constant temperature, volume, and quality of water (Odum 1957). In addition to seasonal variation caused by changes in insolation, the amount of incident light reaching the photosynthetic tissues of macrophytes is affected by attenuation caused by material in the water column and epiphytes growing on their leaves (Kirk 1994). In particular, results in Chapter 2 and 3 indicate that epiphytes cause dramatic reductions in light available to macrophytes, and thereby their productivity. Hence, light availability, which varies across waterbodies, represents an essential facto r modulating the productivity of macrophytes (Nelson and Waaland 1997, Buia and Mazzella 1991, Zupo et al. 1997). In this study light availability controlled gross photosynthetic productivity of V. americana with the relationship between available light and productivity being strong regardless of season and latitude. One way to make such relationships useful is to develop a simulation model of macrophytic production (Elkalay et al. 2003, Plus et al. 2003, Short 1980, Zimmerman et al. 1994, Madden and Kemp 1996, Bocci et al. 1997, Coffaro and Sfriso 1997, Best and Boyd 2001). The majority of such models have been generated for coastal areas (e.g., Elkalay et al. 2003, Plus et al. 2003) and do not take into account loss of biomass caused by the release or tr ansfer of dissolved organic carbon (DOC). In addition, past models measured growth for whole plants (Elkalay et al. 2003) over time steps of one day (Plus et al. 2003), which may compromise their accuracy. Besides the effects of light, most of these models included nutrient concentrations in the water column (e.g.,

PAGE 101

101 nitrogen or phosphorus) as factors limiting production. T here are, however, few substantiated reports of growth of submersed plants in natural freshwater systems being limited by nutrient concent rations (Sytsma and Anderson 1993). In many of a re not positively correlated with algal blooms or reductions of macrophytes (Heffernan et al. 2010). In addition, macrophytes can abso rb nutrients both through their leaves and their roots, so they can grow under oligotrophic concentrations (Iizumi and s can be considered to have ample nutrients for macrophytes. In addition, few models focus o n V americana (e.g., Best and Boyd 2001), and to the best of my knowledge, no model has been developed for submersed I chose to develop I construct ed a simulation model relating the ability of V. americana to convert available light to biomass under the negative influence of epiphytic loads (a V E mo del). This model employed a 15 min time step and focused on production by 1 cm sections of leaves. The preliminary model was tested by comparing the results of simulations with field observations from Chapter 3. Furthermore, the influence of epiphytic load s on growth of V. americana growth was assessed with scenarios involving plants with no loads of epiphytes In two other scenarios, plants were considered to have epiphytic loads of 4 and 5 mg DW cm 2 of leaf to verify the proposed threshold for epiphytic loads that affect the survival of V. americana In addition, the contribution to photosynthetic production for the youngest sections of leaves were differentiated given a heterogeneous distribution of epiphytes. The ultimate aim of this modeling is to pro vide

PAGE 102

102 water resource managers with an objective tool to assess the vulnerability of V. americana to epiphytic loads. For example, managers can use an upper threshold for non detrimental loads of epiphytes as an early warning indicator for the need to prote ct existing macrophytes from additional epiphytization. Description of the Model The model is composed of four main ecological elements, Water, Epiphytic Algae, Light, and V. a mericana the latter treated hierarchically as Section, Leaf, and Plant (Figure 4 1). In the Chassahowitzka River, Light is the limiting factor for photosynthesis and addition of biomass by V. americana Light availability is influenced by conditions in the Water and by loads of Epiphytic Algae, with both the water column and epiphytes on V. americana attenuating incident light. Whatever light reaches the leaves of V. americana is used to convert inorganic matter, which is not considered to be limiting, into organic matter that generates addition of biomass (growth) if production exceeds the requirements associated with maintenance of existing biomass, which includes respiration, a loss term associated with the epiphytic load, and a loss term associated with the age of the leave s. Thus, in the model, the growth of V. americana is a result of an interaction between a key extrinsic environmental factor, i.e., light, and key intrinsic characteristics associated with maintenance. The linkages among extrinsic and intrinsic factors are codified via mathematical expressions and statistical relationships related to light attenuation, photosynthesis, respiration, and losses of biomass associated with epiphytic loads and senescence of sections of leaves (Table 4 1).

PAGE 103

103 Growth of V allisneria a m ericana In its simplest form, growth of leaves of V. americana is calculated on the basis of the amount of net photosynthetic production allocated to aboveground biomass, which is expressed as the net photosynthetic production of a section of a leaf multip lied by a factor that determines the quota of production allocated to new leaf biomass ( Table 4 2). Based on field observations in Chapter 3, the potential for growth varies with leaf age (Figure 4 2). In this model, leaves of V. americana are divide d into 1 cm sections, with the number of sections (n, Table 4 1) depending on the length of each leaf The overall net photosynthetic production per leaf is given by the sum of the net production in all sections ( Table 4 1). Similarly, the net production per shoot ( Table 4 1) is the sum of the net photosynthetic production for all leaves ( ) in that shoot. The net photosynthetic production for a section ( Table 4 1 ) is calculated by subtracting degradative processes that consume photosynthetic product from gross photosynthetic production. In this model, respiration, excretion of dissolved organic carbon (DOC) and senescence and dehiscence induced by epiphytes are th e main degradative processes. Therefore, variation in net photosynthetic acquisition ( Table 4 1 ) by V. americana can be described by the differential equation: ; where is the ph otosynthetic rate per 15 min time step for sections of leaves, is the respiration rate per section and time step, is the loss of photosynthetic product from DOC excretion, is the loss of product induced by epiphytes, which varies with epiphytic load, and (cm 2 ) is the area of a 1 cm 2 section of leaf.

PAGE 104

104 Photosynthesis of V allisneria a mericana Since light is the main factor that determines production of V. americana biomass in the Chassahowitzka River, gross photosynthetic production is estimated through a Photosynthesis Irradiance relationship, which is the most commonly used tool to estimate biological productivity in aquatic systems (Kirk 1994). The Photosynthesis Irradiance relationship is an empirical relationship between light radiation ( Table 4 1) and photosynthetic rate ( Table 4 1), generally expressed as ; where is the specific photosynt hetic rate per 15 min time step for leaves receiving a given amount of photosynthetically active radiation (PAR), is the maximum potential photosynthetic rate, [ ] is a given amount of incident PAR, and is the irradiance half saturatio n constant or the amount of PAR needed for photosynthesis to proceed at The values of , (Table 4 2) the light intensity at the compensation point, and (Table 4 2) the light intensity at the saturation point are taken from the literature. Light availability [ ] is a crucial factor in estimating photosynthetic production, and it is generated by the light availability module. Available Light for V allisneria a mericana The amount of light reaching the surface of leav es is a function of the light measured at the water surface, and it decreases because of extinction in the water column caused by water molecules, turbidity, and dissolved color and extinction due to shading by epiphytes. In the water column, incident light decreases exponentially with increasing depth, and increasing epiphytic biomass generates a similar exponential decrease in available light (Kemp et al. 2004). The equation for transmission of light through th e water column is given by the Lambert Beer Law:

PAGE 105

105 (Kirk 1994); where is the irradiance at depth is the amount of incident light, is the extinction coefficient and is the depth of the leaf section in question. A similar equation describing transmission of light through epiphytes is derived from the results in Chapter 2, and it is expressed as: (Kirk 1994); where is the amount of light that reaches the surface of the le af after penetrating the epiphytic load is the amount of incident light reaching the epiphytes, and are constants that can vary among macrophytes and epiphytic communities, and is the epiphytic load on the section in question. The equat ions can be combined to calculate the amount of light available for photosynthesis as: (Batiuk et al. 1992). The values of constants are derived from fieldwork in Chapters 2 and 3. Loss of Photosynthetic Prod uct Respiration, excretion of DOC and loss of biomass caused by epiphytization reduce gains of biomass through photosynthesis in this model. A fraction of the gross photosynthetic production generated by V. americana is used to satisfy the metabolic demands of existing biomass, i.e., respiration ( Best and Boyd 1999). In general, respiration ( ) is influenced by temperature, but I applied a constant respiration rate because groundwater discharges into the Chassahowitzka River create a relatively const ant, year round water temperature of about 25C (Frazer et al. 2001, Figure 4 4 C ). The respiration rate ( ) is the sum of losses associated with above ground and below ground tissues. The excretion of dissolved organic carbon ( Table 4 1) represe nts another non negligible component of the loss of biomass from macrophytes ( Demarty and Prairies 2009, Duarte et al. 2010 ). In the model, it is described as: ;

PAGE 106

106 where is the rate of DOC excretion and is the amount of carbon fixed in photosynthesis. In addition, epiphytic loads not only reduce the amount of light reaching leaves, their colonization enhances senescence and dehiscence of leaves, which represent additional factors that reduce growth. In the model, thes e losses are calculated as: which assumes that a given epiphytic load ( ) generates a loss ( ) that is proportional to the amount of gross photosynthetic production ( ) scaled by a coefficient ( ) estimated during calibration (Elkalay et al. 2003). Distribution of Epiphytic Biomass Loads of epiphytes are generally low on young, fast growing leaves, whereas older, slower growing leaves are largely colonized; therefore, epiphytic biomass on sections of leaves ( ) is calculated using a linear function: where is the rank of the section in question (from youngest to oldest), is a scaling coefficient for epiphytic bio mass, , and are the area of a section, epiphytic biomass found on the whole leaf, and the total number of sections comprising the leaf in question, respectively. The coefficients in these two equations are derived from the r esults in Chapter 3 (F igure 3 10 ) that indicated that epiphytic biomass increased linearly with distance above a bottommost 7 cm section that was generally free of epiphytes.

PAGE 107

107 Simulations In this simplified model, the V. american a meadows were assumed to exist in an environment free of pests, diseases, and competitors, with suitable weather and an ample supply of nutrients. Intraspecific competition for light (self shading) is not considered because water currents move the leaves constantly, kee ping exposure to available light fairly even. In addition to these extrinsic factors, the growth of V. americana relies on the intrinsic factors as well (Ott 1979). All plants are assumed to be at the same developmental phase, and substantial transfers of organic matter do not occur between leaves and non photosynthetic rhizomes and roots. Moreover, losses through herbivory, sloughing of leaves and fragmentation were not included in this model given the focus on a short time interval (~7 days). All calcula tions were performed for 1 cm long sections of leaves at 15 min time step s using the Python programing language. Simulations were run from 13 June 2016 to 22 August 2016 (62 days), which was the period when meadows in the Chassahowitzka River were sampled (Chapter 3). This approach allowed comparisons between simulat ed and measured data. These measurements of growth (mg DW per week) as a function of epiphytic load (mg DW cm 2 of leaf) displayed a marked exponential decline, with scattered values below this boundary (Figure 4 3). Transmission of light through epiphytes displayed a similar relationship (Chapter 2), which suggests a cause and effect relationship. Thus, I believe that data constituting the upper boundary of the growth versus epiphytic load curve represent those leaves influenced primarily by epiphytic loads, whereas data under the boundary represent those leaves influenced by epiphytes and other factors (e.g., grazers, substrates, and

PAGE 108

108 nutrients). The model was calibrated initially using data on t he boundary to establish parameters related to epiphytic loads, and then run for the rest of data without changing any parameters. The estimated values of some parameters were similar to values reported in the literature (Table 4 2), which gives me confide nce in my ability to estimate other parameters. Other data related to V americana (e.g. ages of leaves, lengths and widths of leaves, dry weights of leaves, epiphytic loads on leaves, and areas and dry weights of new growth were taken from the results of field experiments in Chapter 3. Values for relevant driving variables were obtained from several sources (Figure 4 4). Photosynthetically active radiation (PAR) reaching the water surface was obtained from the Florida Automated Weather Network (FAWN) Leca nto station temperature were obtained from the USGS station at the head spring of the calibrated to match in situ measurements recorded during each sampling event (Chapter 3) that were proportional to the in situ measurements scaled by a coefficient. The discrepancy was computed between the in situ measurements and the calibrated data on the temporal period of in situ measur e ments. The calibrated data for PAR and water depth can capture 84.58% and 69.48% of in situ measurements Sensitivity Analysis A sensitivity analysis investigated the impact of variation in parameters on the model outputs. Each parameter was modified by 20%, and the model was run using one modified parameter at a time. Then, results of the series of runs were analyzed with a sensitivity index (SI Chapelle et al. 2000 ):

PAGE 109

109 where is the percentage of variation applied to a parameter ( 20%), is the number of simulated plants (60), is the new state value generated with the new parameter value, and is the reference stat e value derived using the or i ginal calibrated parameters. Afterwards, mean SI values were calculated for of +20% and 20%. Results and Discussion The model support s different simulations. I focus ed on representative scenarios that exemplify insights that can be derived using this model. Use of Boundary Data to Estimate Detrimental Effects of Epiphytes The set of data bounding measurements of growth for leaves with differing epiphytic biomass was used to calibrate the model (Figure 4 3), with the resulting predictions showing good agreement with field measurement ( = 85%, Figures 4 5 A and 4 5 B ). The steep slopes of the initial declines imply that epiphytes on V. americana strongly inf luence growth of leaves (Figure 4 5 A ). Growth was near zero at loads of approximately 300 mg DW per leaf (~4.6 mg DW cm 2 of leaf). Given these results, a method to estimate the vulnerability of V. americana to epiphytic loads involves application of the n egative exponential curve that fits the data on the boundary (Figure 4 3). The equation describes the growth of V. americana leaves ( ) as a function of epiphytic loads ( ). This empirical model provides an estimate of growth potential g iven a stipulated epiphytic load, based on an assumption that epiphytes are the only environmental influence. Thus, the estimates do

PAGE 110

110 not account for influences of other factors (e.g. light and leaf age). Nonetheless, the approach can provide manager s with useful information regarding the health of macrophytes. Comparative Modeling for Salt Creek and Small Creek to data from Salt Creek and Small Creek. A total of 63 V. americana plants (607 leaves) from Salt Creek and 47 plants (524 leaves) from Small Creek yielded inputs for the model. Predicted values showed good concordance with field observations from both creeks, with = 67% and = 62%, respectively (Figure 4 6). The r oot mean square deviation (RMSD) assesses how well a model describes a system, and the values were 19.14 and 39.23 for Salt Creek and Small Creek, respectively (Figures 4 6 A and 4 6 B ). The higher RMSD for Small Creek (Figure 4 6 B ) highlights a more scatter ed distribution than that for Salt Creek (Figure 4 6 A ). One potential reason for this scatter may be the more complex light regime in Small Creek caused by shading from riparian vegetation For example, the degree of shade changed with sun angle, but the m odel did not account for this variation The maximum predicted growth per week is 154 and 256 g DW per shoot for Salt Creek and Small Creek, respectively (Figure 4 6). More growth in Small Creek can be attributed to lower epiphytic densities (Chapter 3). To differentiate the contributions to net photosynthetic production by 1 cm sections of leaves, plants were separated into a Salt Creek batch and a Small Creek batch, and the net photosynthetic production of each section of leaf was calculated (Table 4 4). For Salt Creek, positive (+) net photosynthetic production disappeared above the section at 15 cm from the base or ( 0 cm), which implied that sections from 15 cm to the apex of the leaf (87 cm) did not contribute to accumulation of energy

PAGE 111

111 by a plant, but rather those sections consumed energy. For Small Creek, where epiphytic loads were lower (Chapter 3), the negative ( ) net photosynthetic production began at 36 cm to 48 cm from the leaf base. The reason Salt Creek exhibited larger negative net photosynth etic production is because higher light levels generated heavier loads of epiphytes on younger sections of leaves (Mazzella and Ott 1984, Dunn et al. 2008, Frankovich and Zieman 2005, Orth and Moore 1983, Doyle 2001). Thus, the model suggests that younger photosynthetic tissues near the base of V. americana leaves serve as the most important contributors to the energy needed to support growth. This conclusion contrasts with results from other studies (Titus et al. 1975, Elkalay et al. 2003) and contradict s the intuitive hypothesis that the optimum photosynthetic tissues for submersed macrophytes are near the water surface where there is greater light availability. Evidence also suggests that V. americana can survive and grow with low levels of available light because of its efficient use of light (Best and Boyd 2001), and in my study area, shallow, clear water may ensure sufficient light reaches the base of shoots. Modeling Growth in the Absence of Epiphyt e Loads The negative impacts of epiphytic loads were evaluated with a comparison of the predictions from a scenario with no epiphytic loads and one using observed epiphytic loads. Without impacts from loads of epiphytes all sections had positive net photo synthetic production (Table 4 4). The basal sections (0 10 cm) remained the most important areas for photosynthesis, generating +30.55% and +40.43% of the net photosynthetic production in Salt Creek and Small Creek, respectively. Although positive net phot osynthetic production was generated by every section, less total

PAGE 112

112 production was generated by older sections near the tips of leaves, which is probably a consequence of a decrease in the number of sections contributing. Two methods were used to assess the d egree of impact generated by epiphytes: and epiphytic epiphytic epiphytic which represents a relative difference scaled to the scenario with no epiphytes absolute differences in net photosynthetic production increased linearly with intensity of epiphytization in Small Creek (Figure 4 7 A and 4 7 B ), and the relative differences displayed a posi tive, logarithmic relationship to increasing epiphytic loads (Figures 4 7 C and 4 7 D ) V. americana epiphytic epiphytic V. americana epiphytic

PAGE 113

113 epiphytic Stratifying Epiphytic Loads When studying the influence of epiphytic loads on growth of V. americana it is important to express density of epiphytes in a useful way In this chapter, epiphytic loads were expressed in mg DW per plant and mg DW cm 2 of leaf; however, these two methods do not describe the characteristic distribution of epiphytes on plants. Some larger plants have larger surface area that, in total, carries more epiphytic biomass, but weights of epiphytes do not translate directly into cover on leaves. Leaf area is an important variable for most ecophysiological processes, including absorbtion of light, respiration, photosynthetic efficiency, absorption of nutrients, and plant growth (Blanco and Folegatti 2005); therefore, standardized epiphytic loads per unit area of leaves was employed. This method effectively eliminates the influence of plant size, but it cannot reflect the heterogeneous distribution of epiphytic biomass in the real world. I used field data to stratify epiphytic loads (mg DW cm 2 of section) for 1 cm sections because they should experience similar depths, incident light intensities, and macrophyte architecture (Gosselain et al. 2005). Epiphytic loads for sections varied from algal free in sections 1 7, through a linear increase to 35 mg DW cm 2 for sections 8 50, and then stabilization until section 84, with a few sections beyond this point having up to 50 mg DW cm 2 of section (Figure 4 9 A ). Using the stratified method to express epiphytic loads, I evaluated absolute and relative differences in net photosynthetic production caused by different epiphytic loads (Figures 4 9 B and 4 9 C ). At low epiphytic loads (< 6 mg DW cm 2 of layer), t epiphytic

PAGE 114

114 (Figures 4 9 B and 4 9 C ). After this point, absolute differences epiphytic 1.1 mg DW cm 2 of section (Figure 4 9C ). These phenomena imply that epiphytic loads in the Chassahowitzka River make a maximum impact on growth, in large part because the filamentous epiphytic algae add biomass by extending away from the lea f surface without covering more surface area. Some ) are greater than 1 is because layers with higher epiphytic loads (>8 mg DW cm 2 of section) have negative net photosynthetic production, which makes Scenarios with Epiphytic Loads of 4 mg DW cm 2 and 5 mg DW cm 2 To test for a threshold epiphytic load that prevents growth of V. americana simulations were performed with total epiphytic loads of 4 mg DW cm 2 of leaf and 5 mg DW cm 2 of leaf for fourteen V. americana plants. Loads of epiphytes were assumed to increase linearly from the base of the leaves to their tip. With epiphytic loads of 4 mg DW cm 2 of leaf, two of the fourteen plants generated positive net photosynthetic production, and the others exhibited no net photosynthetic accumulation. Furthermore, no net photosynthetic production was achieved by any of the 14 plants with epiphytic loads of 5 mg DW cm 2 of leaf. To a certain extent, these results indicate that the threshold for detrimental epiphytic loads for V. americana is between 4 and 5 mg DW cm 2 of leaf. This conclusion agrees with results in Chapter s 2 and 3. Based on this critical threshold, I calculated that, in the Chassahowitzka River, the minimum light required to support growth of V. americana is 20 26% of incident

PAGE 115

115 surface irradiance, which is a value that is in agreement with the 21% proposed by Chambers and Kalff (1985). Sensitivity Analysis To test the sensitivity of the model to v ariation in its component parameters, I increased and decreased each of the main parameters by 20%. The model is sensitive ( increasing leaf tissue ( ), the maximum potential rate of photosynthetic production ( ), respiration rate ( ), coefficient of epiphytic light attenuation, and coefficient of epiphyte induced biomass loss ( Figure 4 10). The model based on the photosynthesis irradiation curve is most sensitive to The sensitivity analyses highlighted the need for accurate estimates of parameters that influence production and link production to epiphytic loads. Other parameters, such as the coefficient for light attenuation in the water co lumn ( ), the half saturation constant ( ), and saturating light intensity ( <0.25), probably because the shallow water at my study sites attenuate s only a small amount of light and incident light is always sufficient for maximum growth of submersed macrophytes like V. americana a shade tolerant plant. Further evidence that light is not limiting comes from the fact that the model is not sensitive to variation in the compensation point ( <0.05). Summary The model produced predictions that agreed well with in situ measurements ( = 67% ), and it supported useful simulations of growth of V. americana subject to different epiphytic loads in the Chassahowitzka River. The inclusion of light, water

PAGE 116

116 depth, and epiphytic biomass, seemed to capture the most important influences on the growth of V. americana Moreover, the importance of epiphytic loads has been confirmed by the model parti cularly regarding competition for light. The model considers 1 cm sections of leaves and a 15 min time step, which was meant to increase the accuracy of predictions relative to previous studies (Elkalay et al. 2003, Plus et al. 2003). In particular, the i mportance of the basal part of the leaves of V. americana in generating a substantial portion of the total photosynthetic production of a whole plant is indicated clearly by this model but seldom considered in other physiological or ecological studies. In addition, DOC excretion appears to be a non negligible loss of biomass (Demarty and Prairies, 2009), so it is important to include this component in models of growth for V. americana Various scenarios highlighted the processes incorporated into the model and provided insights into potentially important ecological effects. The scenario with no epiphytic load indicated that the detrimental effect of epiphytes on growth occurs at a relati vely low biomass. When the effect reaches a certain level, adding filamentous algae adds biomass, but it does not shade more tissue or reduce growth of the macrophyte substantially. In addition, scenarios demonstrated that epiphytic loads of 4 to 5 mg DW cm 2 of leaf represent a likely threshold beyond which leaves of V americana do not grow. Because of the sensitivity of V. americana growth to increasing epiphytic loads, epiphytes can be used key bioindicators for monitoring the health of spring fed sys tems. The model presented here serves as a tool for simulating the effects of epiphytic loads

PAGE 117

117 on V. americana and it provides support for water resource managers to improve management of these important systems.

PAGE 118

118 Table 4 1 E quations comprising the growth and production model for Vallisneria americana Parameter Process Equation References Shoot growth (mg DW per week) Leaf growth (mg DW per week) Best and Boyd (2001) Net production per section (mg DW per 15min) Elkalay et al. (2003) Gross photosynthetic production per section (mg CO 2 m 2 per 15 min) Kirk (1994) Available light for section (E m 2 s 1 ) Batiuk (1992) Biomass loss induced by epiphytes ( mg CO 2 m 2 per 15 min) Elkalay et al. (2003) Biomass loss for DOC excretion Penhale and Smith (1977) Epiphytic biomass (mg DW cm 2 of leaf) of section This study Water depth (m) of section This study Epiphytic biomass ration per leaf (mg DW cm 2 of leaf) This study Section area (cm 2 ) Number of layers

PAGE 119

119 Table 4 2 P arameters comprising the growth and production model for Vallisneria americana Parameter Description Value References Fraction of net photosynthetic production allocated to leaf 0.8 Best and Boyd (2001) Respiration Rate (mg CO 2 m 2 per 15 min) 2.57 Penning de Vries and Van Laar (1982) The maximum potential photosynthetic production (mg CO 2 m 2 s 1 ) 0.12 Titus and Adams (1979) The irradiance half saturation constant (E m 2 s 1 ) 123.08 Titus and Adams (1979) The light intensity at compensation point (E m 2 s 1 ) 9.4 Titus and Adams (1979) The light intensity at saturation point (E m 2 s 1 ) 846.15 Titus and Adams (1979) Water column light attenuation coefficient ( ) 1.56 Measurements Epiphytic load light attenuation constant 1 0.3 This study Epiphytic loads light attenuation constant 2 2 This study Coefficient of epiphyte induced biomass loss (cm 2 of leaf/mg DW) 0.2 0.75 This study Coefficient of biomass loss for DOC excretion 0.4 Penhale and Smith (1977) Number of leaves per shoot Measurements Section number Length of leaf (cm) Measurements Width of leaf (cm) Measurements Water depth (cm) Calibrated with data from USGS Incident light radiation (E m 2 s 1 ) Calibrated with data from FAWN Leaf epiphytic biomass (mg DW cm 2 of leaf ) Measurements Time step (min) 15 Fixed Temperature ( ) 25 Fixed

PAGE 120

120 Table 4 3. Conversions used in the growth and production model for Vallisneria americana Description Units conversion rate References Solar radiation unit conversion Environmental Growth Chambers Leaf dry weight and area unit conversion Old leaves: New leaves: This study Metabolite unit conversion This study Synthetic product unit conversion Duarte (1992)

PAGE 121

121 Table 4 4. P ercentages of net photosynthetic production for sections of Vallisneria americana leaves positive net photosynthetic product ion which implies biomass photosynthetic product ion which implies loss of biomass. Stratified leaves (cm) Salt Creek Small Creek Section count Field observed epiphytic loads Epiphytic load = 0 Section count Field observed epiphytic loads Epiphytic load = 0 81 90(Apex) 21 0.24% +0.11% 71 80 75 0.77% +0.44% 61 70 288 0.98% +1.65% 51 60 719 2.56% +4.06% 40 50 1393 4.67% +7.98% 97 0.16% +0.56% 31 40 2147 6.98% +12.21% 524 +0.5% +5.80% 21 30 3226 8.44% +18.08% 1873 +7.27% +20.41% 11 20 4654 1.77% +24.91% 3343 +25.91% +32.81% 0 10(Base) 6126 +125.72% +30.55% 4621 +66.49% +40.43%

PAGE 122

1 22 Figure 4 1. Unified modeling language diagram for the Chassahowitzka Spring system Diagram shows the structure and dynamic behavior among: Water, Epiphytic Algae, Light Irradiation, and Vallisneria americana (hierarchically expressed as Section, Leaf, and Plant). The objects in the ecosystem are represented by compartments and interactions are represented by the links

PAGE 123

123 Figure 4 2. Maximum potential growth (mg dry weight [DW] per week) for Vallisneria americana leaves of different ages. 0 10 20 30 40 50 60 0 10 20 30 40 50 60 70 80 Maximum leaf growth (mg DW per week) Leaf age (day) Maximum potential growth of leaves

PAGE 124

124 Figure 4 3. Observed growth of Vallisneria americana leaves (mg DW per week) as a function of epiphytic load (mg DW cm 2 of leaf). Observations on the boundary (orange dots) display an exponential decline ( green line) of the form with 0 10 20 30 40 50 60 0 2 4 6 8 10 Leaf growth (mg DW per week) Epiphytic loads (mg DW cm 2 of leaf) Relationship between epiphytic load and leaf growth TrendLine Observations

PAGE 125

125 A B 0 200 400 600 800 1000 1200 1400 1600 1800 2000 6/13/16 6/23/16 7/3/16 7/13/16 7/23/16 8/2/16 8/12/16 8/22/16 Irradiance (E cm 2 s 1 )) Time Water surface photosynthetically active radiation (PAR) 0 10 20 30 40 50 60 70 80 90 100 6/13/16 6/23/16 7/3/16 7/13/16 7/23/16 8/2/16 8/12/16 8/22/16 Water depth (cm) Time Water depth at study area

PAGE 126

126 C Figure 4 4. Variables used in the growth and production model for Vallisneria americana A ) P hotosynthetically active radiation ( PAR ) at the surface of the water B ) W ater depth C ) Water temperature 20 21 22 23 24 25 26 27 28 29 30 6/13/16 6/23/16 7/3/16 7/13/16 7/23/16 8/2/16 8/12/16 8/22/16 Temperature (C) Time Water temperature

PAGE 127

127 A B Figure 4 5. Comparison of simulated and maximum observed growth of Vallisneria americana leaves with different epiphytic loads. A ) A ctual growth (mg dry weight [DW] per week) B) M odel predictions versus in situ measurements (n=33). 0 10 20 30 40 50 60 0 200 400 600 800 Leaf growth (mg DW per week) Epiphytic loads per leaf (mg DW) Simulated and observed leaf growth at boundary Simulated Observed y = 0.8898x R = 0.8527 0 10 20 30 40 50 60 70 -10 10 30 50 70 Observed Simulated Simulated and observed leaf growth (mg DW per week) on the boundary curve X=Y Simulated vs. Observed Linear (Simulated vs. Observed)

PAGE 128

128 A B Figure 4 6. Comparison of model predictions with in situ measur ements of growth for leaves of Vallisneria americana A) Plots are for leaves from Salt Creek with = 67% (n=63). B ) Plots are for leaves from Small Creek with = 62% (n=47). y = 0.87x R = 67% 0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160 Observed Simulated Simulated and observed leaf growth (mg DW) (n=63) in Salt Creek x=y Simulated vs. Observed Linear (Simulated vs. Observed) y = 0.87x R = 62% 0 50 100 150 200 250 300 0 50 100 150 200 250 300 Observed Simulated Simulated and observed leaf growth (mg DW) (n=47) in Small Creek x=y Simulated vs. Observed Linear (Simulated vs. Observed)

PAGE 129

129 A B C D Figure 4 7. Differences in net photosynthetic production for leave s of Vallisneria americana with different epiphytic loads from Small Creek. A) and B) Plots show absolute differences ( net photosynthetic production) C) and D ) Plots show relative differences (R). y = 0.495x R = 0.7324 0 50 100 150 200 250 300 350 400 0 200 400 600 800 Absolute differences net photosynthetic production (mg DW) Epiphytic loads (mg DW per plant) y = 142.61x R = 0.4664 0 50 100 150 200 250 300 350 400 0 1 2 3 Absolute differences net photosynthetic production (mg DW) Epiphytic loads (mg DW cm 2 of leaf) R = 0.8428 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 200 400 600 800 Relative differences ( R) Epiphytic loads (mg DW per plant) R = 0.8303 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.5 1 1.5 2 2.5 Relative differences (R) Epiphytic loads (mg DW cm 2 of leaf )

PAGE 130

130 A B Figure 4 8. Differences in net photosynthetic production for leaves of Vallisneria americana with different epiphytic loads from Salt Creek. A) and B) Plots show absolute differences ( net photosynthetic production). C), D) and E) Plots show relative diff erences (R). 0 100 200 300 400 500 600 700 800 0 2000 4000 6000 8000 10000 12000 14000 16000 Absolute differences net photosynthetic production (mg DW) Epiphytic loads (mg DW per plant) 0 100 200 300 400 500 600 700 800 0 5 10 15 20 25 30 35 Absolute differences net photosynthetic production (mg DW) Epiphytic loads (mg DW cm 2 )

PAGE 131

131 C D Figure 4 8. Continued. 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 0 2000 4000 6000 8000 10000 12000 14000 16000 Relative differences (R) Epiphytic loads (mg DW per plant) 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 0 5 10 15 20 25 30 35 Relative differences (R) Epiphytic loads (mg DW cm 2 )

PAGE 132

132 E Figure 4 8. Continued. y = 0.0606x + 0.5146 R = 0.4545 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 0 1 2 3 4 5 6 Relative differences (R) Epiphytic loads (mg DW cm 2 )

PAGE 133

133 A B Figure 4 9. Distribution of epiphytic loads and differences in net photosynthetic production for sections of leaves of Vallisneria americana with different epiphytic loads from Salt Creek. A) Plots show heterogeneous distribution of epiphytes along the lengths of leaves. B) Plots show absolute differences ( net photosynthetic production). C) Plots show relative differences (R). 0 10 20 30 40 50 60 0 10 20 30 40 50 60 70 80 90 Averaged epiphytic loads per layer (mg DW cm 2 of layer) Leaf layers 0 100 200 300 400 500 600 700 800 0 10 20 30 40 50 60 Absolute differences net photosynthetic production (mg DW) Averaged epiphytic loads of layer (mg DW cm 2 )

PAGE 134

134 C Figure 4 9. Continued. 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 0 10 20 30 40 50 60 Relative differences (R) Averaged epiphytic loads of each layer (mg DW cm 2 )

PAGE 135

135 Figure 4 10. Results of sensitivity analyses for parameters 1 0.18 1.53 0.2 0.47 0.0033 0.15 0.26 0.9 1 0.17 1.46 0.23 0.53 0.0024 0.24 0.26 1.02 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Sensitivity 20% -20%

PAGE 136

136 CHAPTER 5 CONCLUSIONS In Chapter 2, I showed that epiphytes on V. americana can attenuate a considerable portion of incident light. D irect measurements of light transmission through submersed epiphytes produced an empirical model that represents an important component of the numerical model developed in Chapter 4, and along with light requirements drawn from the literature, yielded a pr ediction regarding the threshold epiphytic load ( 4 to 5 mg DW cm 2 of leaf ) that essentially stops growth of V americana leaves. In Chapter 3, the detrimental impacts of epiphytic loads on the growth of V. americana were quantified and evaluated in the fi eld. This work confirmed the hypothesis from Chapter 2 regarding a threshold epiphytic load. This threshold represents a valuable indicator of the health of spring fed aquatic eco systems that complements traditional indicators based on water quality. In Ch apter 4, a simulation model was developed that relates available light to production of biomass for V americana under different epiphytic loads. The model incorporated the statistical relationship from Chapter 2, the field data from Chapter 3, and relationships and parameters from the literature to predict growth of V americana under different epiphytic loads. The model can guide water resource managers as they implement actions to protect macrophytes from epiphytic loads, restore healthy meado fed systems.

PAGE 137

137 APPENDIX A CODE FOR CHAPTER 2 CurveExpert Code for Chapter 2 : # this preamble is optional, but it makes things nicer. Here, you can choose the # name of your model, the equation (if applicable), and the latex form of the equation # (for nice rendering) name = ur"Forced_3_Parameter_Exponential_Decay" nindvar = 1 equation = r"b+q0*exp( x/a)" latexequation = r"b+q0 \ mathrm{exp}{ \ left( x/a \ right)}" def evaluate(x,b ,q0,a): xf = 0.0 yf = 100.0 b = yf q0*exp( xf/a) return b + q0 exp( x/a) def initialize(x,y): """ The initialize function is in charge of initializing the parameters, given the raw data x and y (which are columns of data). Obviously, any Python functions can be used here. The return value from this function should be anything that can be translated into a numpy array. If you don't know what this means, don't worry; just fo llow the examples. """ b = 20.55000000000000000E+00 q0 = 100.000000000000000E+00 a = 1.000000000000000E+00 return (b,q0,a) Code for Python Matplotlib Appendix: import numpy as np from scipy.optimize import curve_fit import matplotlib ma tplotlib.use('Agg') import matplotlib.pyplot as plt from functions import if __name__ == "__main__": #Total DW x = np.array([0.2182539683, 0.2678062678, 0.3336703741, 0.4011899703, 0.4126750184, 0.5556921564, 0.6975414523, 0.7233273056, 0.7333682 556, 0.774025974, 0.7792860734, 0.8483516484, 0.9597818678, 1.073080481, 1.074923324, 1.185185185, 1.411042945, 1.631623213, 1.68281489, 1.731175229,

PAGE 138

138 1.759543659, 1.884057971, 1.976047904, 1.983957219, 2.157407407, 2.911340206, 2.973262032, 2.980769231, 3. 06763285, 3.127272727, 3.130374957, 3.255555556, 3.344068706, 3.484176279, 3.501922508, 3.628577868, 3.680555556, 3.75187294, 3.975225225, 4.018640351, 4.17989418, 4.233664559, 4.424390244, 4.529411765, 4.740740741, 5.025839793, 5.155502392, 5.171014493, 5 .171945701, 5.313981616, 5.982142857, 6.203296703, 6.35625, 6.455026455, 7.804444444, 7.933333333, 8.354385965, 9.489230769, 11.33156966, 16.66147725]) y = np.array([76.35382426, 77.32572675, 76.85355017, 87.14624233, 75.41629998, 70.25901862, 66.60361 481, 54.86474553, 67.84346474, 60.24143158, 54.14845937, 57.35518262, 53.45346221, 62.05054586, 53.57278321, 60.33802166, 47.11178937, 59.50845763, 50.53201086, 44.62358724, 47.19197733, 38.73192891, 46.98181413, 45.01706591, 51.0922529, 27.42021176, 26.65 652522, 40.1285566, 35.87361739, 25.40533619, 62.22260031, 24.42602135, 20.13887855, 37.23497851, 61.45734292, 13.625503, 26.91658861, 25.32638282, 14.14494296, 32.71531587, 19.02114614, 43.14799943, 13.01767064, 26.77101523, 14.56419496, 18.85469551, 24.9 8520182, 26.50361415, 11.45962702, 32.01269484, 13.54327294, 8.075660015, 9.344904253, 6.213800643, 8.68022516, 18.70946595, 4.484490645, 9.566883309, 12.3959923, 17.9341622]) #=== Plot fig = plt.figure() ax = fig.add_subplot(111) xlabel = 'Epiphyte Biomass, mg DW $cm^{ 2}$ of leaf' ax.set_xlabel(xlabel) ax.set_ylabel('Light Transmission (I/$I_0$) %') plt.scatter(x, y) xdata = np.linspace(0, 20, 1000) plt.xlim([0,20]) plt.ylim([0,100]) plt.grid() plt.savefig('scatter_DW.png') no_force_point(x, y, 'DW', xlabel) force_point(x,y,'DW', 0 100, xlabel) import numpy as np from scipy.optimize import curve_fit import matplotlib.pyplot as plt def no_force_point(x, y, file_name, xlabel, chla=None): if chla: x_max=50 else: x_max=20 for i in xrange(4): t = i+1 print t popt, pcov = model_func_curve_fit(x, y, None, t, chla) residuals = y model_func_one(x, popt[0], popt[1], popt[2]) ss_res = np.sum(residuals**2)

PAGE 139

139 ss_tot = np.sum((y np.mean(y))**2) r_squared = 1 (ss_res / ss_tot) print "unconstraint model%s r_squared = %s"%(t, r_squared) #=== Plot fig = plt.figure() #fig.suptitle('Light Transmittion', fontsize=14, fontweight='bold') ax = fig.add_subplot(111) ax.set_xlabel(xlabel) ax.set_ylabel('Light Transmission %') plt.scatter(x, y) #plt.savefig('testplot.png') xdata = np.linsp ace(0, x_max, 1000) plt.plot(xdata, model_func(xdata, popt[0], popt[1], popt[2], t)) text = model_text(popt, r_squared, t) plt.annotate(text, xy=(0.03, 0.95), xycoords='axes fraction', fontsize=12) plt.xlim([0,x_max]) plt.ylim([0,100]) plt.grid() plt.savefig('unconstraint_%s_model%s.png'%(file_name, t)) def force_point(x, y, file_name, force_x, force_y, xlabel, chla=None): x = np.insert(x, 0, force_x) y = np.insert(y, 0, force_y) if chl a: x_max=50 else: x_max=20 for i in xrange(4): t = i+1 sigma = np.ones(len(x)) sigma[[0]] = 0.01 popt, pcov = model_func_curve_fit(x, y, sigma, t, chla) residuals = y model_func_ one(x, popt[0], popt[1], popt[2]) ss_res = np.sum(residuals**2) ss_tot = np.sum((y np.mean(y))**2) r_squared = 1 (ss_res / ss_tot) print "constraint model%s r_squared = %s"%(t, r_squared) #=== Plot fig = pl t.figure() #fig.suptitle('Light Transmittion', fontsize=14, fontweight='bold') ax = fig.add_subplot(111) ax.set_xlabel(xlabel) ax.set_ylabel('Light Transmission %') plt.scatter(x, y) #plt.savefig('testplo t.png') xdata = np.linspace(0, x_max, 1000) plt.plot(xdata, model_func(xdata, popt[0], popt[1], popt[2], t))

PAGE 140

140 text = model_text(popt, r_squared, t) plt.annotate(text, xy=(0.03, 0.95), xycoords='axes fraction', fon tsize=12) plt.xlim([0,x_max]) plt.ylim([0,100]) plt.grid() plt.savefig('constraint_%s_model%s.png'%(file_name, t)) def model_func_curve_fit(x, y, sigma, t, chla=None): if t == 1 and sigma != None: return curve_fit (model_func_one, x, y, p0=(100 1, 1), sigma=sigma) elif t == 2 and sigma != None: return curve_fit(model_func_two, x, y, p0=(100 1, 1), sigma=sigma) elif t == 3 and sigma != None: return curve_fit(model_func_three, x, y, p0=(100 1, 1), sigma=sigma) elif t == 4 and sigma != None: return curve_fit(model_func_four, x, y, p0=(100 1, 1), sigma=sigma) elif t == 1 and sigma == None: return curve_fit(model_func_one, x, y, p0=(100 1, 1)) elif t == 2 and sigma == None: return curve_fit(model_func_two, x, y, p0=(100 1, 1)) elif t == 3 and sigma == None: return curve_fit(model_func_three, x, y, p0=(100 1, 1)) elif t == 4 and sigma == None: if chla: return curve_fit(model_func_four, x, y, p0=(835, 0.001, 2.7)) else: return curve_fit(model_func_four, x, y, p0=(100, 1, 1)) def model_func(x, a, b, c, t): if t == 1: return model_func_one(x, a, b, c) elif t == 2: return model_func_two(x, a, b, c) elif t == 3: return model_func_three(x, a, b, c) elif t == 4: return model_func_four(x, a, b, c) def model_text(popt, r_squared, t): if t == 1: return model_text_one(popt, r_squar ed) elif t == 2: return model_text_two(popt, r_squared) elif t == 3: return model_text_three(popt, r_squared) elif t == 4: return model_text_four(popt, r_squared) def model_text_one(popt, r_squared): return "$f(x)=ae^{ x/b}$ | a=%.3f, b=%.3f" % (popt[0],popt[1])

PAGE 141

141 #return "$f(x)=ae^{ x/b}$ | a=%.3f, b=%.3f, $R^{2}$=%.3f" % (popt[0],popt[1], r_squared) def model_func_one(x, a, b, c): return a*(np.exp( x/b)) def model_text_two(popt, r_squared): return "$f(x)= c + ae^{ x/b}$ | a=%.3f, b=%.3f, c=%.3f" % (popt[0],popt[1],popt[2]) #return "$f(x)= c + ae^{ x/b}$ | a=%.3f, b=%.3f, c=%.3f, $R^{2}$=%.3f" % (popt[0],popt[1],popt[2], r_squared) def model_func_two(x, a, b, c): return c + a*(np. exp( x/b)) def model_text_three(popt, r_squared): return "$f(x)= a/(1+x/b)$ | a=%.3f, b=%.3f" % (popt[0],popt[1]) #return "$f(x)= a/(1+x/b)$ | a=%.3f, b=%.3f, $R^{2}$=%.3f" % (popt[0],popt[1], r_squared) def model_func_three(x, a, b, c): return a/(1+x/b) def model_text_four(popt, r_squared): return "$f(x)= a*(1+c*x/b)^{ 1/c}$ | a=%.3f, b=%.3f, c=%.3f" % (popt[0],popt[1],popt[2]) #return "$f(x)= a*(1+c*x/b)^{ 1/c}$ | a=%.3f, b=%.3f, c=%.3f, $R^{2}$=%.3f" % (popt[0],popt[1],popt[2], r_squa red) def model_func_four(x, a, b, c): return a*(np.power((1+c*x/b), ( 1/c)))

PAGE 142

142 APPENDIX B CODE FOR CHAPTER 4 Section: # coding: utf 8 from __future__ import unicode_literals from math import exp, pow from django.db import models from django.utils import timezone # Create your models here. class Section(models.Model): leaf = models.ForeignKey('Leaf', on_delete=models.CASCADE) light = models.ManyToManyField('Light', null=True) section_rank = models.IntegerField() gr oss_photo = models.FloatField(null=True) respiration = models.FloatField(null=True) net_photo = models.FloatField(null=True) #loss_by_epi = models.FloatField(null=True) def setCalNetPhoto(self): sum_gross_photo =0 sum_resp iration =0 sum_net_photo = 0 #sum_loss_by_epi = 0 QL = self.leaf.plant.experiment.ql #1 KD = self.leaf.plant.experiment.kd #1.56 #(1/m) PMAX = self.leaf.plant.experiment.pmax #0.077 mg CO2 m 2s 1 IC = self.leaf.plant.experiment.ic #9.4 #(uE/s/m^2) LIGHT_SAT = self.leaf.plant.experiment.light_sat #846.15 #(uE/s/m^2) KM =self.leaf.plant.experiment.km #123.08 #(uE/s/m^2) EPI_LIGHT_COEF = self.leaf.plant.experiment.epi_l ight_coef #1 EPI_BIO_LOSS_COEF = self.leaf.plant.experiment.epi_bio_loss_coef #1 RESPIRATION = self.leaf.plant.experiment.respiration # 2.57mg CO2/(m2*15min) ALGAE_BIOMASS = self.leaf.plant.experiment.algae_biomass for time in self.light.all().order_by('date_and_time'): time_depth = (time.water.depth self.section_rank + 0.5)/100 #input depth(cm), after/100, (m) if time_depth<0: time_depth=0 water_attenuation_pct=exp( 1*K D*time_depth) if ALGAE_BIOMASS == None:

PAGE 143

143 if self.section_rank>7: algae_biomass= self.leaf.algae_biomass_ratio*(self.section_rank 7)*(float(self.leaf.width)) algae_attenuation_pct = (EPI_LIGHT_COEF*0.8)*exp( 1*algae_biomass/(EPI_LIGHT_COEF*3.63)) #0.8 #0.3*exp( 1*algae_biomass) else: algae_biomass=0 algae_attenuation_pct =1 else: #algae_biomass = ALGAE_BIOMASS algae_biomass = ALGAE_BIOMASS*(self.section_rank*float(self.leaf.width)*float(self.leaf.length))/(0.5*se lf.leaf.section_num*(self.leaf.section_num+1)) algae_attenuation_pct = ( EPI_LIGHT_COEF*0.8)*exp( 1*algae_biomass/(EPI_LIGHT_COEF*3.63)) if algae_biomass==0: algae_attenuation_pct =1 #print algae_biomass #print algae_attenuation_pct #print "#####" #print algae_attenuation_pct #################################### #print "%s %s %s"%(time.surface_light, water_attenuation_pct, algae_attenuation_pct) #if time_av ailable_light>846.15: P=Pmax=16.5*8.96*10^ 4 #else (if 140=LIGHT_S AT: gross_photo_rate=PMAX #mgCO2/(m^2*s) elif IC
PAGE 144

144 gross_photo = gross_photo_rate*60*15*(1*self.leaf.width)/10000*0.68 #mgCO2/(15min) ->mgDW sum_gross_photo = sum_gross_photo + gross_photo if algae_biomass > 1: if algae_biomass >= 4.5: # most epi respiration = EPI_BIO_LOSS_COEF*0.15*5*gross_photo + 0.4* gross_photo + RESPIRATION*(1*self.leaf.width)/10000*0.68 #respiration = 0.25*algae_biomass*gross_photo + 0.4* gross_photo + RESPIRATION*(1*self.leaf.width)/10000*0. 68 elif 2 < algae_biomass <= 4.5: respiration = EPI_BIO_LOSS_COEF*0.2*algae_biomass*gross_photo + 0.4* gross_photo + RESPIRATION*(1*self.leaf.width)/10000*0.68 elif 1< algae_biomass <=2: respira tion = EPI_BIO_LOSS_COEF*0.4*algae_biomass*gross_photo + 0.4* gross_photo + RESPIRATION*(1*self.leaf.width)/10000*0.68 else: #less epi respiration = EPI_BIO_LOSS_COEF*0.3*algae_biomass*gross_photo + 0.4* gross_photo + RESPIRATIO N*(1*self.leaf.width)/10000*0.68 #if self.section_rank <= 45: #if algae_biomass >= 0.2: # mg DW/cm^2 #respiration = (0.4* gross_photo + 4.15*0.2*gross_photo) + RESPIRATION*(1*self.leaf.width)/10000*0.68 #elif 0.1mgDW sum_net_photo = sum_net_photo + net_photo #1mgDW=1.467mgCO2~1.65mgCO2 #1mgC02=0.606~0.682mgDW #print "water_attenuation_pct %s"%water_attenuation_pct current_time = timezone.localtime(time.date_and_time).strftime("%Y %m %d %H:%M")

PAGE 145

145 #print_out = "%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s"%(self, current_time,time.surface_light, # time_depth, time.water.depth, water_attenuation_pct, # algae_biomass,algae_attenuation_pct, # time_available_light, # gross_photo_rate,net_photo,sum_net_photo) #with open("log/section.csv", "a") as myfile: # myfile.write(print_out+' \ n') #print print_out #print 'time: %s net_photo:%s time_availible_light: %s time_dep th: %s'%(time.date_and_time, net_photo, time_availible_light, time_depth) #print "sum_net_photo: %s"%sum_net_photo self.gross_photo = sum_gross_photo self.respiration = sum_respiration self.net_photo = sum_net_photo #exit() return self.net_photo #|7 days sum net photo real data |^2=r2 #outplot: x list all parameter and r^2 sum net photo*growth rate=new growth #(CO2) (WHOLE LEAVE) ''' def save(self, *args, **kwargs): i f self.id != None: self.setCalNetPhoto() super(Section, self).save(*args, **kwargs) def __unicode__(self): return "Plant %s: Leaf: %s Section: %s" % (self.leaf.plant.id, self.leaf.leaf_rank, self.section_rank) #./recreate.sh #(clean old data and input new data) #./createsuperuser.sh #if I want to see database (website) #python manage.py cal_section # calculate #python manage.py continue # continue #https://macrophyte model yangx.c9users.io/admin/macrophyte/light/ #website #https://macrophyte model yangx.c9users.io/admin/macrophyte/section/ #sudo service postgresql start #sudo service redis server start #https://bitbucket.org/y_xu /macrophyte_model #./start.sh #./recreate.sh #./celery_run.sh # http://24.91.12.125/admin/ #bitbucket.org

PAGE 146

146 #./sql/section_sum.sh (ps:Calculate section net photo) #./sql/plant_epi.sh (ps:Calculate epi density total blade) # 1.cd data > 2.#./drop_create.sh (remove old data) #3.#./rollback.sh (Run project see if it is right) #3.open a new terminal run#./sql/plant_epi.sh) #4. find result in file(sql/plant_epi.csv) #cd sql > bash section_sum.sh #pg_dump macrophyte | gzip > macrophyte.gz ( download .gz) Leaf: # coding: utf 8 from __future__ import unicode_literals from django.db import models from macrophyte.models import Section from django.db.models import F, FloatField, Sum # Create your models here. class Leaf(models.Model): plant = models.ForeignKey('Plant', on_delete=models.CASCADE) algae = models.OneToOneField('Algae', on_delete=models.CASCADE, null=True) leaf_rank = models.IntegerField() age = models.Intege rField(null=True) length = models.FloatField(null=True) width = models.FloatField(null=True) section_num = models.IntegerField(null=True) algae_biomass_ratio = models.FloatField(null=True) #algae_density = models.FloatField(null=True) real_growth= models.FloatField(null=True) net_photo = models.FloatField(null=True) self_used_net_photo = models.FloatField(null=True) gross_photo_product = models.FloatField(null=True) respiration = models.FloatField(null=True) #r = models.FloatField(null=True) def setCalSectionNum(self): if int(float(self.length)) == float(self.length): self.section_num = int(float(self.length)) else: self.section_num = int(float(self.length)) + 1 return self.section_num

PAGE 147

147 # self.algae_biomass_ration=float(self.algae.algae_biomass)/(float(self.width)*(section_nu m 7)(section_num 7+1)/2) #(epi DW mg)/(1*W)*(n*(n+1)/2) #def setCalAlgaeDensity(self): # self.algae_densi ty= float(self.algae.algae_biomass)/(float(self.length)*float(self.width)) #(mg/cm^2) # return self.algae_density def setCalAlgaeBiomassRatio(self): if int(self.section_num) > 7: #self.algae_biomass_ratio =(f loat(self.algae.algae_biomass)*float(self.width)*float(self.length))/(float(self.width)*(se lf.section_num 7)*(self.section_num 7+1)/2) self.algae_biomass_ratio =float(self.algae.algae_biomass)/(float(self.width)*(self.section_num 7)*(self.secti on_num 7+1)/2) #print "%s = float(%s)/(float(%s)*(%s 7)*(%s 7+1)/2)"%(self.algae_biomass_ratio, self.algae.algae_biomass, self.width, self.section_num, self.section_num) else: self.algae_biomass_ratio = 0 return self .algae_biomass_ratio def setNetPhoto(self): result = Section.objects.filter(leaf=self).aggregate(sum_net_photo=Sum(F('net_photo'), output_field=FloatField())) self.net_photo = result['sum_net_photo'] if self.net_photo == No ne: self.net_photo = 0 #self.r = result['sum_net_photo'] self.real_growth #print "R^2 for leaf %s is %s"%(self.leaf_rank, r) return self.net_photo # def R=abs(sum net photo nb_dw) return R # each leaf has one R^2=R*R # I want to know the sum of all leaves R^2 and return the value def setGrossPhotoProduct(self): result = Section.objects.filter(leaf=self).aggregate(sum_gross_photo_product=Sum(F('gr oss_ph oto'), output_field=FloatField())) self.gross_photo_product = result['sum_gross_photo_product']

PAGE 148

148 if self.gross_photo_product == None: self.gross_photo_product = 0 return self.gross_photo_product def set Respiration(self): result = Section.objects.filter(leaf=self).aggregate(sum_respiration=Sum(F('respiration'), output_field=FloatField())) self.respiration = result['sum_respiration'] if self.respiration == None: self.res piration = 0 return self.respiration def setSelfUsedNetPhoto(self): if self.net_photo < 0: self.self_used_net_photo = 0 return self.self_used_net_photo if self.age <= 3: self.sel f_used_net_photo = self.net_photo elif 3 < self.age <= 9: if self.net_photo > 52.3: self.self_used_net_photo = 52.3 else: self.self_used_net_photo = self.net_photo elif 9 < self.age <= 15: if self.net_photo > 56.16: self.self_used_net_photo = 56.16 else: self.self_used_net_photo = self.net_photo elif 15 < self.age <= 21: if s elf.net_photo > 40.56: self.self_used_net_photo = 40.56 else: self.self_used_net_photo = self.net_photo elif 21 < self.age <= 27: if self.net_photo > 16.55: se lf.self_used_net_photo = 16.55 else: self.self_used_net_photo = self.net_photo elif 27 < self.age <= 33: if self.net_photo > 11.70:

PAGE 149

149 self.self_used_net_photo = 11.70 else: self.self_used_net_photo = self.net_photo elif self.age > 33: self.self_used_net_photo = 0 return self.self_used_net_photo #return self.LeaveIncreasePerShoot #@property #def setTide(self): # return 1 def save(self, *args, **kwargs): if self.algae != None: self.setCalSectionNum() self.setCalAlgaeBiomassRatio() #self.setCalAlgaeDensity() super(Leaf, self).save(*args, **kwargs) def __unicode__(self): return "Plant %s Leaf %s" % (self.plant.id, self.leaf_rank) #https://macrophyte model yangx.c9users.io/admin Plant: # coding: utf 8 from __future__ import unico de_literals from django.db import models from macrophyte.models import Leaf from django.db.models import F, FloatField, Sum # Create your models here. class Plant(models.Model): experiment = models.ForeignKey('Experiment', on_delete=models.CASCADE) predicted_growth = models.FloatField(null=True) gross_photo_product = models.FloatField(null=True) respiration = models.FloatField(null=True) real_growth = models.FloatField(null=True, default=0) net_photo = models.FloatField(null=True) new_leaf = models.FloatField(null=True) # predicted growth = growth rate* plant net photo #predicted growth vs. real growth (real photo product)

PAGE 150

150 #@property #def setTide(self): # return 1 def setNewLeaf(self): se lf.new_leaf = 0 for leaf in Leaf.objects.filter(plant=self): if leaf.self_used_net_photo == 0: pass elif leaf.age <= 3: self.new_leaf = self.new_leaf + leaf.net_photo else: self.new_leaf = self.new_leaf + (leaf.net_photo leaf.self_used_net_photo) return self.new_leaf def setNetPhoto(self): result = Leaf.objects.filter(plant=self, net_photo__gte=0).aggregate(sum_net_photo=Sum(F('net_photo'), output_field=FloatField())) self.net_photo = result['sum_net_photo'] if self.net_photo == None: self.net_photo = 0 return self.net_photo def setGrossPhotoProduct(self): result = Leaf.objects.filter(plant =self).aggregate(sum_gross_photo_product=Sum(F('gross_phot o_product'), output_field=FloatField())) self.gross_photo_product = result['sum_gross_photo_product'] return self.gross_photo_product def setRespiration(self): resul t = Leaf.objects.filter(plant=self).aggregate(sum_respiration=Sum(F('respiration'), output_field=FloatField())) self.respiration = result['sum_respiration'] return self.respiration def setPredictedGrowth(self): #predicted_Growth #L_FRACTION = self.experiment.l_fraction self.predicted_growth = self.net_photo #* L_FRACTION # Ftaction of net photo allocated to leaves return self.predicted_growth def save(self, *args, **kwargs):

PAGE 151

151 super(Pl ant, self).save(*args, **kwargs) class Meta: ordering = [' experiment','id'] def __unicode__(self): return "Experiment: %s Plant: %s" % (self.experiment.id, self.id)

PAGE 152

152 LIST OF REFERENCES Agust, S., S. Enrquez, H. Frost Christensen, K. Sand Jensen and C.M. Duarte. 1994. Light harvesting among photosynthetic organisms. Functional Ecology 8: 273 279. Akaike, H. 1973. Information theory as an extension of the maximum likelihood principle In Second International Symposium on Information Theory ed. B. Petrov, and F. Cski Akadmiai Kiad Asaeda, T., M. Sultana, J. Manatunge, and T. Fujino. 2004. The effect of epiphytic algae on the growth and production of Potamogeton perfoliatus L. in two light co nditions Environmental and Experimental Botany 52(3): 225 238. Askenasy, E. 1880 ber eine neue Methode, um die Vertheilung der Wachsthumsintensitt in wachsenden Theilen zu bestimmen. Varl Winter's Universittsbuchhandlung in Heidelberg. Batiuk, R.A., R.J. Orth, K.A. Moore, W.C. Dennison, and J.C. Stevenson. 1992. Chesapeake Bay submerged aquatic vegetation habitat requirements and restoration targets: A technical synthesis. Report number PB 93 196665/XAB Gloucester Point: Virginia Instit ute of Marine Science. Best, E.P., and W.A. Boyd. 1999 A simulation model for growth of the submersed aquatic macrophyte Eurasian watermilfoil (Myriophyllum spicatum L.). No. WES TR A 99 3. Vicksburg : Army Engineer Waterways Experiment Station. Best, E.P ., and W.A. Boyd. 2001. A simulation model for growth of the submersed aquatic macrophytes American wildcelery (Vallisneria americana Michx.). US Army Crops of Engineers. Engineer Research and Development Center. Blanch, S.J., G.G. Gant, and K.F. Walker. 1998. Growth and recruitment in Vallisneria americana as related to average irradiance in the water column Aquat ic Bot any 61: 181 205. Blanco, F.F., and M.V. Folegatti. 2005. Estimation of leaf area for greenhouse cucumber by linear measurements under salinity and grafting. Sci ence of Agric ulture 62(4) : 305 309 Bocci, M., G. Coffaro, and G. Bendoricchio. 1997. Modelling biomass and nutri ent dynamics in eelgrass ( Zostera marina L.): application to the lagoon of Venice (Italy) and Oresund (Denmark). Ecol ogical Model ing 102: 67 80. Bonn, M.A. 2004. Visitor profiles, economic impacts and recreational aesthetic values associated with eight pri ority Florida springs located in the St. Johns River Water Management District. Special Publication SJ2004 SP35. Palatka: St. Johns River Water Management District.

PAGE 153

153 Borowitzka, M.A., P.S. Lavery, and M. v an K eulen. 2006. Epiphytes of seagrasses. In Seagras ses: B iology, Ecology and Conservation ed. A. Larkum, R.J. Orth, and C. Duarte. Springer Netherlands 441 461 pp Borowitzka, M.A., R.C. Lethbridge, and L. Charlton. 1990. Species richness, spatial distribution and colonisation pattern of algal and invertebrate epiphytes on the seagrass Amphibolis griffithii Marine Ecology Progress Series 64: 281 291. Borum, J. 1987. Dynamics of epiphyton on eelgrass ( Zostera marina L.) leaves: relative roles of algal growth, herbivory, and substratum turnover Limnology and Oceanography 32(4): 986 992. Borum J. and S. Wium Andersen. 1980. Biomass and production of epiphytes on eelgr ass ( Zostera marina L.) in the resund, Denmark Ophelia 1: 57 64. Brook, H.K. 1981. Guide to the physiographic divisions of Florida. Gainesville: Florida Cooperative Extension Service Institute of Food and Agricultural Sciences, University of Florida Brush, M.J. and S.W. Nixon. 2002. Direct measurements of light attenuation by epiphytes on eelgrass Zostera marina Marine Ecology Progress Series 238: 73 79. Buia, M.C., and L. Mazzella. 1991. Reproductive strategies of the Mediterranean seagrasses, Posi donia oceanica (L.) Delile, Cymodocea nodosa (Ucria) Aschers., and Zostera noltti Hornem. Aquat ic Bot any 40: 343 362. Bulthuis, D.A. and W.J. Woelkerling. 1983. Biomass accumulation and shading effects of epiphytes on leaves of the seagrass, Heterozostera tasmanica in Victoria Australia Aquatic Botany 16: 137 148. Burnell, O.W., B.D. Russell, A.D. Irving, and S.D. Connell. 2014. Seagrass response to CO 2 contingent on epiphytic algae: indirect effects can overwhelm direct effects Oecologia 176(3): 871 8 82. Burnham, K.P., and D.R. Anderson. 2001. Kullback Leibler information as a basis for strong inference in ecological studies. Wildlife Research 28: 111 119. Burnham, K.P., and D.R. Anderson. 2002. Model selection and multimodel inference: A practical inf ormation T heoretic Approach Springer Verlag Burt, J.S., G.A. Kendrick, R.J. Masini, and C.J. Simpson. 1995. Light and Posidonia sinuosa seagrass meadows in the temperate coastal waters of Western Australia. II. Effect of epiphyte species assemblages and biomass on attenuating light to the leaf surface Technical Series 61 Perth: Department of Environmental Protection.

PAGE 154

154 Cambridge, M.L., A.W. Chiffings, C. Brittan, L. Moore, and A.J. McComb. 1986. The loss of seagrass in cockburn sound, Western Australia. II Possible causes of seagrass decline Aquatic Botany 24: 269 285. Canfield, D.E. and M.V. Hoyer. 1988. Influence of nutrient enrichment and light availability on the abundance of aquatic macrophytes in Florida streams. Canadian Journal of Fisheries an d Aquatic Sciences 45: 1467 1472. Capone, D.G., P.A. Penhale, R.S. Oremland, and B.F. Taylor. 1979. Relationship between productivity and N 2 (C 2 H 2 ) fixation in a Thalassia testudinum community Limnology and Oceanography 24(1): 117 125. Carter, V., N.B. Ry bicki, J.M. Landwehr, and M. Naylor. 2000. Light requirements for SAV survival and growth. Chesapeake Bay submerged aquatic vegetation water quality and habitat based requirements and restoration targets: a second technical synthesis. Chesapeake Bay Program 4 15. USEPA. Cattaneo, A., G. Galanti, and S. Gentinetta. 1998. Epiphytic algae and macroinvertebrates on submerged and floating leaved macrophytes in an Italian lake Freshwater Biology 39(4): 725 740. Cebrin, J., and C.M. Duarte. 1994. The depen dence of herbivory on growth rate in natural plant communities Functional Ecology 518 525. Cebrin J., S. Enrquez, M. Fortes, N. Agawin, J.E. Vermaat, and C.M. Duarte 1999. Epiphyte accrual on Posidonia oceanica (L.) Delile leaves: implications for ligh t absorption. Bot anica Mar ina 42: 123 128. Chambers, P.A. and J. Kalff. 1985. Depth distribution of biomass of submerged aquatic macrophyte communities in relation to Secchi depth Canadian Journal of Fisheries and Aquatic Sciences 42: 701 709. Chambers, P.A. and J. Kalff. 1987. Light and nutrient in the control of aquatic plant community structure: I. In situ experiments. J ournal of Ecol ogy 75 : 611 619. Chapelle, A., A. Mnesguen, J.M. Deslous Paoli, P. Souchu, N. Mazouni, A. Vaquer, and B. Millet. 2000 Modelling nitrogen, primary production and oxygen in a Mediterranean lagoon. Impact of oysters farming and inputs from the watershed. Ecol ogical Model ing 127: 161 181. Choice, Z.D., T.K. Frazer, and C.A. Jacoby. 2014. Light requirements of seagrasses det ermined from historical records of light attenuation along the Gulf coast of Peninsular Florida Marine Pollution Bulletin 81(1): 94 102. Coffaro, G. and A. Sfriso. 1997. Simulation model of Ulva rigida growth in shallow water of the Lagoon of Venice. Ecol ogical Model ing 102 : 55 66.

PAGE 155

155 Cohen, M.J. 2007. Sources, transport and transformations of Nitrate N in the Flo rida environment. Final Report SJ2007 SP10. St. Johns River Water Management District. Czerny, A.B. and K.H. Dunton. 1995. The effects of in s itu light reduction on the growth of two subtropical seagrasses, Thalassia testudinum and Halodule wrightii Estuaries and Coasts 18(2): 418 427. Demarty, M. and Y.T. Prairies. 2009. In sity dissolved organic carbon (DOC) release by submerged macrophytes epiphyte communities in southern Quebec lakes. Can adian J ournal of Fish eries and Aquat ic Sci ences 66: 1522 1531 Dennison, W.C., R.J. Orth, K.A. Moore, J.C. S tevenson, V. Carter, S. Kollar, P.W. Bergstrom, and R.A. Batiuk. 1993. Assessing water quality with submersed aquatic vegetation BioScience 43(2 ): 86 94. Dixon, L.K. 2000. Establishing light requirements for the seagrass Thalassia testudinum : an example f rom Tampa Bay, Florida. CRC Press, LLC Doust, J.L. and G. LaPorte. 1991. Population sex ratios, population mixtures and fecundity in a clonal dioecious macrophyte, Vallisneria americana The Journal of Ecology 79: 477 489. Doyle, R.D. 2001. Effects of waves on the early growth of Vallisneria americana Freshwater Biology 46: 289 397. Drake, L.A., F.C. Dobbs, and R.C. Zimmerman. 2003. Effects of epiphyte load on optical properties and photosynthetic potential of the seagrasse s Thalassia testudinum banks ex Konig and Zostera marina L. Limnology and Oceanography 48: 456 463. Duarte, C.M. 1991. Seagrass depth limits. Aquat ic Bot any 40: 337 363. Duarte, C.M. 1992. Nutrient concentration of aquatic plants: patterns across species. Limnology and Oceanography 37: 882 889. Duarte, C.M., N. Marba, N. Agawin, J. Cebrian, S. Ennquez, M.D. Fortes, M.E. Gallegos, M. Merino, B. Olesen, K. Sand Jensen, J. Uri, and J.E. Vermaat. 1994. Reconstruction of seagrass dynamics: age determinations and associated tools for the seagrass ecologist. Mar ine Ecol ogy Prog ress Ser ies 107: 195 209. Duarte, C.M., Y.T. Prairie, T.K. Frazer, M.V. Hoyer, S.K. Notestein, R. Martinez, A. Dorsett, and D.E. Canfield. 2010. Rapid accretion of dissolved organic carbon in the springs of Florida: the most organic poor natural waters. Biogeosciences 7: 4051 4057.

PAGE 156

156 Dunn, A.E., D.R. Dobberfuhl, and D.A. Casamatta. 2008. A survey of algal epiphytes from Vallisneria americana michx. (Hydrocharitaceae) in the Lower St. Johns River, Florida. Southeastern Naturalist 7(2): 229 244 Elkalay, K., C. Frangoulis, N. Skliris, A. Goffart, S. Gobert, G., Lepoint, and J. Hecq. 2003. A model of the seasonal dynamics of biomass and production of the seagrass Posidonia oceanica in the Bay of Calvi (Northwestern Miditerranean). Ecological Modeling 167: 1 18. Environmental Growth Chamber s. Accessed at: http://www.egc.com/useful_info_lighting.php Erickson, R.O., and F.J. Michelini. 1957. The plastochron index. Am erican J ournal of Bot any 44: 297 305. FAWN, Florida Automated Weathe r Network. 2015. Institute of Food and Agricultural Sciences University of Florida Gainesville, Florida. Accessed at: http://fawn.ifas.ufl.edu/data/reports/ Fitzpatrick, J.R., and H. Kirkman. 1995. Effects of prolonged shading stress on growth and survival of the seagrass Posidonia australis in Jervis Bay, New South Wales, Australia. Marine Ecology Progress Series 127: 279 289. Fong, C.W., S.Y. Lee, and R.S. W u. 2000. The effects of epiphytic algae and their grazers on the intertidal seagrass Zostera japonica Aquatic Botany 67(4): 251 261. Frankovich, T.A. and J.C. Zieman. 1994. Total epiphyte and epiphytic carbonate production on Thalassia testudinum across Florida Bay Bull etin of Ma rine Sci ence 54: 679 695. Frankovich, T.A., and J.C. Zieman. 2005. Periphyton light transmission relationships in Florida Bay and the Florida Keys, USA. Aquatic Botany 83: 14 30. Frazer, T.K. 2000. Coastal nitrate assessment: Nu trient assimilation capacity of five gulf coast rivers. Second annual project summary. Tampa: Southwest Florida Water Management District Surface Water Impr ovement and Management Program Frazer, T.K., M.V. Hoyer, S.K. Notestein, J.A. Hale, and D.E. Canf ield. 2001. Physical, chemical and vegetative characteristics of five gulf coast rivers. Gainesville: University of Florida Frazer, K.T., S.K. Notestein, and W.E. Pine, Jr. 2006. Changes in the physical chemical and vegetative characteristics of the Homosassa, Chassahowitzka and Weeki Wachee Rivers Submitted to Southwest Florida Water Management District Surface Water Improvement and Management Program.

PAGE 157

157 Glazer, B.T. 1999. Analysis of physical, chemical, and biological factors inhibiting growth and re and Rehoboth Bays. Lewes: University of Delaware Gosselain, V., C. Hudon, A. Cattaneo, P. Gagnon, P. Planas, and D. Rochefort. 2005. Physical variables driving epiphytic algal biomass in a dense macrophyte bed of the St. Lawrence River (Quebec, Canada). Hydrobiologia 534: 11 22. Gregg, W.W., and F.L. Rose. 1985. Influences of aquatic macrophytes on invertebrate community structure, guild structure, and microdistribution in streams. Hydrobio logia 128: 45 56. Hauxwell, J.A., C.W. Osenberg and T.K. Frazer. 2004. Conflicting management goals: manatees and invasive competitors inhibit restoration of a native macrophhyte. Ecol ogical App lications 14: 57 586. Hauxwell, J.A., T.K. Frazer, and C.W. O senberg. 2007. An annual cycle of biomass and productivity of Vallisneria americana in a subtropical spring fed estuary. Aquatic Botany 87 : 61 68. Heffernan, J.B., D.M. Liebowitz, T.K. Frazer, J.M. Evans, and M.J. Cohen. 2010. Algal blooms and the nitrogen enrichment hypothesis in Florida springs: evidence, alternatives, and adaptive management. Ecological Applications 20: 816 829. Hootsmans, M.J.M. and J.E. Vermaat. 1991 Macrophytes, a key to understanding changes caused by eutrophication in shallow freshwater ecosystems. IHE, Delft, The Netherlands, IHE Report Series. Howard Williams, C., B.R. Davies and R.H.M. Cross. 1978. The influence of perip hyton on the surface structure of a Potamogeton pectinatus L. leaf. Aquat ic Bot any 5: 87 91. Hoyer, M.V., T.K. Frazer, S.K. Notestein, and D.E. Canfield. 2004. Vegetative characteristics of three low lying Florida coastal rivers in relation to flow, light salinity and nutrients. Hydrobiologia 528: 31 43. Iizumi, H., and A. Hattori. 1982. Growth and organic production of eelgrass ( Zostera marina L.) in temperature waters of the Pacific Coast of Japan. III. The kinetics of nitrogen uptake. Aquatic Bot any 12: 245 256. Jones, G.W., S.B. Upchurch, K.M. Champion, and D.J. Dewitt. 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. Southwest Florida Water Management Program Ambient Ground Water Quality Monitoring Program.

PAGE 158

158 Kelly, M.G., B. Moeslund, and N. Thyssen. 1981. Productivity measurement and the storage of oxygen in the aerenchyma of aquatic macrophytes Archi v f r Hydrobiologie 92: 1 10 Kemp, W.M., R. Bartleson, L. Murray. 2000. Epiphyte contributions to light attenuation at the leaf surface. In: Chesapeake Bay submerged aquatic vegetation water quality and habitat based requirements and restoration targets: a second technical synthesis. United States Environmental Protection Agency 55 69. Kemp, W.M., R. Batiuk, R. Bartleson, P. Bergstrom, V. Carter, C.L. Gallegos, W. Hunley, L. Karrh, E.W. Koch, J.M. Landwehr, K.A. Moore, L. Murray, M. Naylor, N.B. Rybicki, J .C. Stevenson, and D.J. Wilcox. 2004. Habitat requirements for submerged aquatic vegetation in Chesapeake Bay: water quality, light regime, and physical chemical factors. Estuaries 27: 363 377. Kimber, A., C.E. Korschgen, and A.G. Van Der Valk 1995. The d istribution of Vallisneria americana seeds and seedling light requirements in the Upper Mississippi River. Canadian Journal of Botany 73: 1966 1973. Kirk, J.T.O. 1994. Light and photosynthesis in aquatic ecosystems. Cambridge University Press. Kirbe, T. 1980. Production of Ruppia cirrhosa (Petagna) Grande in mixed beds in Ring Kobing Fjord (Denmark). Aquat ic Bot any 9:135 143. Knight, R.L. and S.K. Notestein. 2008. Springs as Ecosystems. In: Brown, M.T., K.C. Reiss, M.J. Cohen, J.M. Evans, K.R. Reddy, P.W. Inglett, K.S. Inglett, T.K. Frazer, C.A. Jacoby, E.J. Phlips, R.L. Knight, S.K. Notestein, and K.A. McKee. 2008. Summary and synthesis of the available literature on the Effects of nutrients on spring organisms and systems. Report to the Flori da Department of Environmental Protection Tallahassee, Florida Korschgen, C.E. and W.L. Green. 1988. American wild celery (Vallisneria americana): ecological considerations for restoration. U.S. Fish and Wildlife Service. Fish and Wildlife Technical Rep ort 19 24. Kremer, B.P. 1981. Metabolic implications of non photosynthetic carbon fixation in brown macroalgae Phycologia 20(3): 242 250. Kurtz, J.C., D.F. Yates, J.M. Macauley, R.L. Quarles, F.J. Genthner, C.A. Chancy, and R. Devereux. 2003. Effects of light reduction on growth of the submerged macrophyte Vallisneria americana and the community of root associated heterotrophic bacteria Jou rnal of Experimental Marine Biology and Ecology 291(2): 199 218. Lamoreaux, R.J., W.R. Chaney, and K.M. Brown. 1978. The plastochron index: a review after two decades of use American Journal of Botany 65: 586 593.

PAGE 159

159 Lauridsen, T.L., E. Jeppesen, and M. Snd ergaard. 1994. Colonization and succession of submerged macrophytes in shallow Lake Vaeng during the first five years following fish manipulation Hydrobiologia 275(1): 233 242. Lin H.J. 1995. Responses of epiphytes on eelgrass ( Zostera marina L.) to nutri ent enrichment. PhD thesis, University of Rhode Island, Kingston, RI. Lindeboom, H.J. and B.H.H. De Bree. 1982. Daily production and consumption in an eelgrass ( Zostera marina ) community in saline Lake Grevelingen: discrepancies between the O 2 and 14 C meth od Netherlands Journal of Sea Research 16: 362 379. Lorenti, M., L. Mazzella, and M.C. Buia. 1995. Light limitation of Posidonia oceanica (L.) Delile leaves and epiphytes at different depths. Rapports de la Commision Internationale de la Mer M diterrane 34. Losee R.F. and R.G. Wetzel. 1983. Selective light attenuation by the periphyton complex. In Periphyton of freshwater ecosystems Wetzel, R.G. (ed). Springer Ludwig, J.A. and D.J. Tongway. 1995. Spatial organization of landscapes and its function i n Semi Arid Woodlands, Australia. Landscape Ecology 10: 51 63. Madden, C.J., and W.M. Kemp. 1996. Ecosystem model of an estuarine submersed plant community, calibration an simulation of eutrophication responses. Estuaries 19: 457 474. Mattson, R.A., J.H. E pler, and M.K. Hein. 1995. Description of benthic communities in karst, spring fed streams of north central Florida Journal of the Kansas Entomological Society 68: 18 41. Mazzella, L. and J. Ott. 1984. Seasonal changes in some features of Posidonia oceanica (L.) Dellile leaves and epiphytes at different depths. In Proceedings of the International workshop on Posidonia oceanica beds ed. C.F. Boudouresque A.J. de Grissac, J. Olivier, and G I S Posidonie G.I.S. Posidonie Mazzella, L. and R.S. Albe rte. 1986. Light adaptation and the role of autotrophic epiphytes in primary production of the temperate seagrass, Zostera marina L Journal of Experimental Marine Biology and Ecology 100(1 3): 165 180. McFarland, D.G. 2006. Reproductive ecology of Vallisn eria americana Michaux SAV 06 4 Vicksburg: Engineer research and development center Montana Department of Environmental Quality. 2011. Sample collection and laboratory analysis of Chlorophyll a standard operation procedure. Accessed at: http://deq.mt.gov/Portals/112/Water/WQPB/QAProgram/Documents/PDF/SOPs/ WQPBWQM 011v6_FNL.pdf

PAGE 160

160 Muenscher, W.C. 1936. The germination of seeds of Potamogeton Annals of Botany 50(200): 805 821. Munch, D.A., D.J. Toth, C. Huang, J.B. Davis, C.M. Fortich, W.L. Osburn, E.J. Phlips, E.L. Quinlan, M.S. Allen, M.J. Woods, P. Cooney, R.L. Knight, R.A. Clarke, and S.L. Knight. 2006. Fifty year retrospective study of the ecology of Silver Springs, Florida Publication Number: SJ2007 SP4. Palatka: St. Johns R iver Water Management District Neckles, H.A., R.L. Wetzel, and R.J. Orth. 1993. Relative effects of nutrient enrichment and grazing on epiphyte macrophyte ( Zostera marina L ) dynamics. Oecologia 93: 28 5 295. Nelson, T.A., and J.R. Waaland. 1997. Seasonality of eelgrass, epiphyte, and grazer biomass and productivity in subtidal eelgrass meadows subjected to moderate tidal amplitude. Aquat ic Bot any 56: 51 74. Notestein S.K. 2001. Physical, chemical, and vegetative characteristics of the Chassahowitzka river. Gainesville: University of Florida. Odum, H.T. 1957. Trophic structure and productivity of Silver Springs, Florida. Ecological Monographs 27(1): 55 112. Orth, R.J., and J. van Montfrans. 1984. Epiphyt es seagrass relationships with an emphasis on the role of micrograzing: a review. Aquat ic Bot any 18: 43 69. Orth, R.J., and K.A. Moore. 1983. Chesapeake Bay: an unprecedented decline in submerged aquatic vegetation. Science 222: 51 53. Oshima, Y., M.J. Ki shi, and T. Sugimoto. 1999. Evaluation of the nutrient budget in a seagrass bed. Ecol ogical Model ing 115: 19 33. Ott, J.A. 1979. Persistence of a seasonal growth rhythm in Posidonia oceanica (L.) Delile under constant conditions of temperature and illumina tion. Mar ine Biol ogy Lett ers 1: 99 104. Ott, J.A. 1980. Growth and production in Posidonia oceanica (L.) Delile Marine Ecology 1(1): 47 64. Patriquin, D. 1973. Estimation of growth rate, production and age of the marine angiosperm Thalassia testudinum Konig Caribbean Journal of Science 13: 111 123 Penhale, P.A. and W.O. Smith. 1977. Excretion of dissolved organic carbon by eelgrass ( Zostera marina ) and its epiphytes Limnology and Oceanography 22(3): 400 407.

PAGE 161

161 Penning de Vries, F.W.T., and H.H. VanLaa r. 1982. Simulation of growth processes and the model BACROS. Simulation of plant growth and crop production, Pudoc, Wageningen 114 131. Phillips, G., D. Eminson, and B. Moss. 1978. A mechanism to account for macrophyte decline in progressively eutrophica ted freshwaters. Aquat ic Bot any 4: 103 126. Pinckney, J., H. Paerl, P. Tester, and T. Richardson. 2001. The role of nutrient loading and eutrophication in estuarine ecology. Environmental Health Perspectives 109: 699 706. Pirc, H. 1985. Growth dynamics in Posidonia oceanica (L.) Delile Marine Ecology 6(2): 141 165. Plus, M., A. Chapelle, A. Mnesguenb, J.M. Deslous Paolic, and I. Auby. 2003. Modelling seasonal dynamics of biomasses and nitrogen contents in a seagrass meadow ( Zostera noltii Hornem.): application to the Thau lagoon (French Mediterranean coast). Ecological Modeling 161(3): 211 236. Ramus, J. 1981. The capture and transduction of light energy In The biology of seaweeds ed C.S. Lobban, and M.J. Wynne. Botanical Monographs, Bla ckwell Scientific Publications Rogers, J.W., D.G. McFarland, and J.W. Barko. 1995. Evaluation of the growth of Vallisneria americana Michx. in relation to sediment nutrient availability. Lake and Reservoir Management 11: 57 66. Rogers, K.H. and C.M. Bre en. 1981. Effects of epiphyton on Potamogeton crispus L. leaves Microbial Ecology 7(4): 351 363. Rosenau, J.C., G.L. Faulkner, C.W. Hendry Jr., and R.W. Hull. 1977. Springs of Florida. Florida Department of Natural Rescources. Bureau of Geology Bulletin N o.32: 461. Sand Jensen, K. 1977. Effect of epiphytes on eelgrass photosynthesis. Aquat ic Bot any 3: 55 63. Sand Jensen, K. 1990. Epiphyte shading its role in resulting depth distribution of submerged aquatic macrophytes. Folia geobotanica and phytotaxonomica 25: 315 320. Sand Jensen, K., C. Prahl, and H. Stokholm. 1982. Oxygen release from roots of submerged aquatic macrophytes Oikos 38: 349 354. Sand Jensen, K. and J. Borum. 1984. Epiphyte shading and its effect on photosynthesis and diel me tabolism of Lobelia dortmanna L. during the spring bloom in a Danish lake Aquatic Botany 20: 109 119.

PAGE 162

162 Sand Jensen, K., N.P. Revsbech, and B.B. Jorgensen. 1985. Microprofiles of oxygen in epiphyte communities on submerged macrophytes. Mar ine Biol ogy 89: 55 62. Sand Jensen, K. and M. Sndergaard. 1981. Phytoplankton and epiphyte development and their shading effect on submerged macrophytes in lakes of different nutrient status International Review of Hydrobiology 66(4): 529 552. Scott, T.M., G.H. Means, R.C. Means, and R.P. Meegan. 2002. First magnitude springs of Florida: Florida Geological Survey Open File Report 85. Scott, T.M., G.H. Means, R.P. Meegan, R.C. Means, S.B. Upchurch, R.E. Copeland, J. Jones, T. Roberts, and A. Wil let. 2004. Springs of Florida. Florida Geological Survey Bulletin No. 66. Sculthorpe, C.D. 1985. The biology of aquatic vascular plants. Koeltz Scientific Books. Konigstein, Germany. Seddon, S., R.S. Connolly, and K.S. Edyvane. 2000. Large scale seagrass d ie back in northern Spencer Gulf, South Australia Aquat ic Bot any 66: 297 310. Short, F.T. 1980. A simulation model of seagrass production system. In: Handbook of seagrass biology : an ecosystem perspective Phillips, R.C., and C.P. McRoy (eds). Garland ST PM Press. Short, F.T., and C.M. Duarte. 2001. Methods for the measurement of seagrass growth and production. In Global seagrass research methods ed. F.T. Short, and R.G. Coles. Amsterdam: Elsevier Science Silberstein, K., A. Chiffings, and A. McComb. 198 6 The loss of seagrass in Cockburn Sound, Western Australia. III. The effect of epiphytes on productivity of Posidonia australis Hook. F. Aquat ic Bot any 24: 355 371. Smart, R.M., J.W. Barko and D.G. McFarland. 1994. Competition between Hydrilla verticillata and Vallisneria americana under different environmental conditions. Technical report NO. WES/TR/EL A 94 1. Vicksburg: Army engineer waterways experiment station. Spence, D.H.N. 1976. Light and plant response in fre shwater In Light as an ecological factor ed. R. Bainbridge, G.C. Evans, and O. Rackham Blackwell Science Publications. Springs Management Plan. 2016. Southwest Florida water Management District Accessed at: http://www.swfwmd.state.fl.us/publications/files/Springs_Management_Plan.pdf Stankelis, R.M., M.D. Naylor, and W.R. Boynton. 2003 Submerged aquatic vegetation in th e mesohaline region of the Patuxent estuary: Past, present, and future status. Estuaries 26: 186 195

PAGE 163

163 Stankelis R.M., W.R. Boynton, and J.M. Frank. 1999. Submerged aquatic vegetation (SAV) habitat evaluation. Chapter 1. In Ecosystems processes component le vel interpretive report N o 16 Solomons: Chesapeake Biological Laboratory, Universi ty of Maryland System Stevenson, R.J., A. Pinowska, A. Albertin, and J.O. Sickman. 2007. Ecological condition of algae and nutrients in Florida springs: The Synthesis Repo rt. Tallahassee: Prepared for the Florida Department of Environmental Protection. Sytsma, M.D., and L.W.J. Anderson. 1993. Transpiration by an emergent macrophyte:source of water and implications for nutrient supply. Hydrobiologia 271: 97 108. Thayer, G.W ., W.J. Kenworthy, and M.S. Fonseca. 1984 Ecology of eelgrass meadows of the Atlantic coast: a community profile (No. FWS/OBS 84/02). National Marine Fisheries Service, Beaufort, NC (USA). Beaufort Lab. Virginia Univ., Charlottesville (USA). Dep artment of Environmental Sciences. Titus, J., and M.A. Adams. 1979. Coexistence and the comparative light relations of the submersed macrophytes Myriophyllym spicatum L. and Vallisneria americana Michx. Oecologia 40: 273 286. Titus, J., R.A. Goldstein, M.S. Adams Shugart, and R.S. Booth. 1975. A production model for Myriophyllum spicatum L. Ecology 56: 1129 1138. Twilley, R., W. Kemp, K. Staver, J. Stevenson, and W. Boynton. 1985 Nutrient enrichment of estuarine subm ersed vascular plant communities. I. Algal growth and effects on production of plants and associated communities. Mar ine Ecol ogy Prog ress Ser ies 23: 179 191. United States Fish and Wildlife Service. 1988. Crystal River National Wildlife Refuge annual narra tive report Chassahowitzka National Wildlife Refuge. Homosassa, Florida. van Dijk, 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 L. Hydrobiologia 252: 143 161. van Donk, E. and W.J. van de Bund 2002. Impact of submerged macrophytes including charophytes on phyto and zooplankton communities: allelopathy versus other mechanisms Aquatic Botany 72(3): 261 274. Vsquez Elizondo, R.M. and S. Enrquez. 2017. Light absorption in coralline algae ( Rhodophyta ): a morphological and functional approach to understanding species distribution in a coral reef lagoon Frontiers in Marine Science 4: 297.

PAGE 164

164 Vermaat, J.E ., J.A.J. Beijer, R. Gijlstra, M.J.M. Hootsmans, C.J.M. Philippart, N.W. van den Brink and W. van Vierssen. 1993. Leaf dynamics and standing stocks of intertidal Zostera noltii Hornem. and Cymodocea nodosa (Ucria) Ascherson on a). Hydrobiologia 258: 59 72. Vermaat, J.E. and M.J.M. Hootsmans. 1994. Periphyton dynamics in a temperature light gradient. Lake Veluwe, a macrophyte dominated system under Eutrophication Stress, Springer Netherlands. Waycott, M., C.M. Duarte, T.J. Carruthers, R.J. Orth, W.C. Dennison, S. Olyarnik, A. Calladine, J.W. Fourqurean, K.L. Heck, A.R. Hughes, and G.A. Kendrick. 2009. Accelerating loss of seagrasses across the globe threatens coastal ecosystems Proceedings of the National Academy of Science s 106(30): 12377 12381. Westlake, D.F. 1978. Rapid exchange of oxygen between plant and water Proceedings International Association of Theoretical and Applied Limnology Verh andlungen des Int ernationalen Verein Limnol ogie 20 : 2363 2367 Wetland Solutions, Inc. 2010. An ecosystem Project Agreement No.08010 Wigand, C., J. Wehr, K. Limburg, B. Gorham, S. Longergan, and S. Findlay. 2000. Effect of Vallisneria americana (L.) on community structure and ecosystem function in lake mesocosms. Hydrobiologia 418: 137 146. Yobbi, D.K. 1992. Effects of tidal stage and ground water levels on the discharge and water quality of springs in costal Citrus and Hernando Counties United States Geological Survey. WRI 92 4069. Yobbi, D.K. and L.A. Knockenmus. 1989. Salinity and flow relations and effects of reduced flow in Chassahowitzka River and Hommossassa River estuaries, Southwest Florida. U.S. Geological Survey. Water Resources Investigations Report 88 4044. Zieman, J.C., 1968 A stud y of the growth and decomposition of the seagrass: Thalassia testudinam. Miami: University of Miami. Zieman, J.C., and R.G. Wetzel. 1980. Productivity in seagrasses: methods and rates. In Handbook of seagrass biology : an ecosystem perspective ed. R.C. Phillips and C.P. McRoy Garland STPM Press. Zieman, J.C., S.A. Macko, and A.L. Mills. 1984. Role of seagrasses and mangroves in estuarine food webs: temporal and spatial changes in stable isotope composition and Amino acid content during decomposition B ulletin of Marine Science 35(3): 380 392.

PAGE 165

165 Zimmerman, R.C., A. Cabello Pasini, and R.S. Alberte. 1994. Modelling daily production of aquatic macrophytes from irradiance measurements: a comparative analysis. Mar ine Ecol ogy Prog ress Ser ies 114: 185 196. Zupo, V., M.C. Buia, and L. Mazzella. 1997. A production model for Posidonia oceanica based on temperature. Estuarine Coastal Shelf Sci ences 44: 483 492.

PAGE 166

166 BIOGRAPHICAL SKETCH Jing Guan was born in Qingdao, China, a beautiful coastal city. Camping, hiking, and swimming instilled in her a fundamental interest and enchantment with the natural world. Her interests in natural science took hold, and since secondary school, she has in dulged in all kinds of scientific experiments. As part of the generation born in the 1980s in China, she witnessed overpopulation, deterioration of water resources and exhaustion of fishery resources. Her enthusiasm for natural sciences and a desire to mak e a difference motivated her to pursue an associate s degree focused on Fisheries Sciences and Aquaculture. Undergraduate studies and an internship at a fisheries research institution deepened her understanding of Environmental Science and brought her into a new realm, aquatic science. Subsequently, she went to the University of Florida to study water treatment, management of water quality and hydro ecological restoration. She obtained her M.S from the Department of Soil and Water Sciences. While in Florida she was particularly drawn toward the unique spring systems and felt heartbroken when she saw the anthropogenic impacts they endure. Her strong interest macrophytes, and spring fed systems. Jing plans to continue working on applied aquatic science around the world.