Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1985 Production and Nitrogen Dynamics of a Vallisneria Americana Grass Bed in Lake Pontchartrain, Louisiana (Nitrogen Cycle, Modeling, Seagrasses, Decomposition Phenology, Vallisneria-Celery Grass-Eelgrass, Environmental Factors). Diana Lee Steller Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Steller, Diana Lee, "Production and Nitrogen Dynamics of a Vallisneria Americana Grass Bed in Lake Pontchartrain, Louisiana (Nitrogen Cycle, Modeling, Seagrasses, Decomposition Phenology, Vallisneria-Celery Grass-Eelgrass, Environmental Factors)." (1985). LSU Historical Dissertations and Theses. 4165. https://digitalcommons.lsu.edu/gradschool_disstheses/4165

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS

This reproduction was made from a copy of a manuscript sent to us for publication and microfilming. While the most advanced technology has been used to pho­ tograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. Pages in any manuscript may have indistinct print. In all cases the best available copy has been filmed.

The following explanation of techniques is provided to help clarify notations which may appear on this reproduction.

1. Manuscripts may not always be complete. When it is not possible to obtain missing pages, a note appears to indicate this.

2. When copyrighted materials are removed from the manuscript, a note ap­ pears to indicate this.

3. Oversize materials (maps, drawings, and charts) are photographed by sec­ tioning the original, beginning at the upper left hand comer and continu­ ing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and is available, for an additional charge, as a standard 35mm slide or in black and white paper format.*

4. Most photographs reproduce acceptably on positive microfilm or micro­ fiche but lack clarity on xerographic copies made from the microfilm. For an additional charge, all photographs are available in black and white standard 35mm slide format. *

♦For more information about black and white slides or enlarged paper reproductions, please contact the Dissertations Customer Services Department

IM wrsity Mtaronhns International

8610675

Steller, Diana Lee

PRODUCTION AND NITROGEN DYNAMICS OF A VALLISNERIA AMERICANA GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA

The Louisiana State University and Agricultural and Mechanical. Col Ph.D. 1985

University Microfilms International 300 N. Zeeb Road, Ann Arbor, Ml 48106

PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages ______

2. Colored illustrations, paper or print ______

3. Photographs with dark background _____

4. Illustrations are poor copy ______

5. Pages with black marks, not original copy ______

6. Print shows through as there is text on both sides of page ______

7. Indistinct, broken or small print on several pages V

8. Print exceeds margin requirements ______

9. Tightly bound copy with print lost in spine______

10. Computer printout pages with indistinct print ______

11. Page(s) ______lacking when material received, and not available from school or author.

12. Page(s) ______seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled pages ______

15. Dissertation contains pages with print at a slant, filmed as received •

16. Other______

University Microfilms International

PRODUCTION AND NITROGEN DYNAMICS OF A VALLISNERIA americana

GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

in

Marine Sciences

by Diana Lee Steller B.A., Cornell University, 1972 M.S., University of Florida, 1976 December 1985 ACKNOWLEDGEMENTS

I"would like to thank each member of my committee

for their time and cooperation. This work would not

have been possible without their support.

Funding for the field work came from the U. S.

Army Corps of Engineers* New Orleans# Louisiana. Dr.

James Stone was instrumental in arranging funding#

student worker help* and equipment. 1 would like to

thank the American Association of University Women for

their fellowship which funded one year of this work.

Financial support was also given by the Mathematics

Department* L.S.U. and the Mathematics Departments of

Southern University and Southeastern Louisiana

University. Thanks go to the many tireless workers who

helped in the fieldwork (B. Johnson, G. Happ# J. Day*

L. Bahr* N. Walker* G. Rainey* T. Gayle* M. Casselman*

G. Root* M. Fannally# R. Allen# D. Smith* K. Hoffman*

A. Witzig* and A. Matherne) and particular thanks to

Albert J. Meier whose presence was a constant joy and

inspiration. Drs. J. Gosselink and W. Wiseman showed

patience and persistence in reading and correcting this

manuscript. TABLE OF CONTENTS

I. CHAPTER 1 1 INTRODUCTION 1 1.0 ORIGIN OF LAKE PONTCHARTRAIN GRASSBEDS 4 1.1 GRASSBED DEVELOPMENT RELATED TO SAND BEACH TRENDS 6 1.2 GRASSBED DECLINE 7 1.3 REFERENCES CITED 9 II. CHAPTER 2 24 LITERATURE REVIEW 2.0 INTRODUCTION 24 2.1 PRODUCTIVITY AND DISTRIBUTION RELATED TO ENVIRONMENTAL FACTORS 25 2.1.1 Light 26 2.1.2 Temperature 29 2.1.3 Salinity 31 2.1.4 Currents and Wave Action 31 2.1.5 Sediment 32 2.2 LIMITING NUTRIENTS 32 2.2.1 Carbon 32 2.2.2 Nitrogen 33 2.2.3 Phosphorus 37 2.2.4 Techniques 37 2.3 DECOMPOSITION 39 2.4 COMMUNITY 41 2.5 HUMAN INFLUENCE 42 2.6 LITERATURE CITED 44 III. CHAPTER 3 55 PRODUCTION OF A SUBMERGED GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA, INCLUDING LITTERFALL AND DECOMPOSITION USING A HARVEST TECHNIQUE 55 ABSTRACT 55 3.0 INTRODUCTION 56 3.1 THE STUDY AREA 58 3.2 METHODS 58 3.2.1 Harvest Methods 61 3.2.2 Litter 62 3.2.3 Decomposition 63 3.2.4 Harvest Techniques and Calculations 64 3.2.4.1 Peak Standing Crop 64 3.2.4.2 Harvest 65 3.2.4.3 Harvest Plus Litterfall 65 3.2.4.4 Harvest Plus Litterfall Plus Decomposition 66 3.3 RESULTS 67 3.3.1 Biomass 67 3.3.1.1 Area Along the Shore Effects 69 3.3.1.2 Distance from Shore iii Effects 70 3.3*1.3 Area Times Distance Effects 71 3.3.2 Litter 71 3.3.3 Decomposition 73 3.3.4 Production 75 3.4 DISCUSSION 76 3.4.1 Biomass 76 3.4.2 Litter 77 3.4.3 Decomposition 78 3.4.4 Production 79 3.4.4.1 Methods 79 3.4.4.2 Value of Harvest Results 82 3.5 LITERATURE CITED 85 IV. CHAPTER 4 110 ENVIRONMENTAL AND PHENOLOGICAL CORRELATES OF GROWTH OF THE SEAGRASS Vallisneria americana IN LAKE PONTCHARTRAIN no ABSTRACT 110 4.0 INTRODUCTION 112 4.1 DESCRIPTION OF STUDY AREA U 4 4.2 METHODS 114 4.3 RESULTS 116 4.3.1 Biomass and Net Annual Production Patterns n 6 4.3.2 Environmental Parameters Related to Distance from Shore 117 4.3.3 Environmental Parameters Related to Season ng 4.3.4 Phenology of the Grass Bed 122 4.4 DISCUSSION 124 4.5 REFERENCES CITED 130 V. CHAPTER 5 I57 SIMULATED NITROGEN DYNAMICS DURING GROWTH AND DECOMPOSITION OF A SUBMERGED VALLISNERIA AMERICANA GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA 157 ABSTRACT I57 5.0 INTRODUCTION I 5 9 5.1 DESCRIPTION OF THE STUDY AREA AND GRASS BED ECOSYSTEM i5g 5 .2 MODEL DEVELOPMENT 1 6 0 5.2.1 Model Boundaries 151 5.2.2 Construction of the Conceptual Model 1 6 2 5.2.2.1 Forcing Functions 152 5.2.2•2 Growth 1 g 4 5.2.2.3 Decay 154 5.2.3 Construction of the Simulation Models 166 5.2.3.1 GROW 1 6 6 5.2.3.2 DECAY 171

iv 5.2.4 Initializing GROW and DECAY 176 5.2.4.1 GROW 176 5.2.4.2 DECAY 176 5.2.5 Calibrating GROW and DECAY 177 5.2.5.1 GROW 178 5.2.5.2 DECAY 181 5.3 RESULTS AND DISCUSSION 134 5.3.1 GROW: Leaf and Root and Rhizome Nitrogen Simulation 185 5.3.2 DECAY: Inorganic and Organic Water and Sediment Nitrogen Simulations 187 5.3.3 Sensitivity Analysis 190 5.3.3.1 GROW Sensitivity Analysis 191 5.3.3.2 DECAY Sensitivity Analysis 193 5.3.4 Validation of GROW 197 5.3.5 Prediction of the Vallisneria Grass Bed's Response to Different Sunlight Conditions 198

5.3.6 Conclusions 199

5.4 REFERENCES CITED 201

VI. CHAPTER 6 264 CONCLUSIONS AND RECOMMENDATIONS TO MANAGERS 264 6.0 SUMMARY OF RESULTS 264 6.1 RECOMMENDATIONS TO ECOSYSTEM MANAGERS 266 6.2 LITERATURE CITED 269

v LIST OF FIGURES Chaoter 1

Figure 1. The physiography of the Pontchartrain Basin (from Saucier, 1963). 13 Figure 2. Origin of the Milton's Island and the Pine Island beach trends (from Saucier, 1963). Figure 3. Distribution of sediment types in Lake Pontchartrain, La., as determined by Bahr, et al., (1980). 17 Figure 4. Comparison of the Milton's Island and Pine Island Beach Trends with the location of submerged grassbeds in Lake Pontchartrain, Louisiana. 19 Figure 5. Land use in the Lake Pontchartrain watershed in 1954 (from Turner and Bond, 1980). 21 Figure 6 . Land use in the Lake Pontchartrain watershed in 1972 (from Turner and Bond, 1980). 23

Chapter 3

Figure 1. Location of the Lake Pontchartrain grassbeds (after Darnell, 1961, and Montz, 1978, and the control and nitrogen enrichment study sites. * 93 Figure 2. Submerged litter trap showing construction and position in the sediment. 95 Figure 3. Vallisneria americana biomass (means of all samples) versus time in g dry wt/square meter. 97 Figure 4. Daily average solar radiation, average water temperature, average water depth, mean salinity, and average currents for May, 1978, through April, 1979, at the Vallisneria grassbed (combined data for the control area and the nitrogen enrichment area. 99 Figure 5. Water column (a) and sediment lb) available inorganic nitrogen at the control area (symbol C) and the nxtrogen enrichment area (symbol N ) . 101 Figure 6 . Monthly production measured by the harvest method (NAPPH ), the harvest plus litter method (NAPPL), and the harvest plus litter plus decomposition method (NAPPp), at 25m from shore at vi the control area (a), 50m from shore at the control area (b), 25m from shore at the nitrogen enrichment area (c), and 5Um from shore at the nitrogen enrichment area (d). 103-106 Figure 7. Decomposition of Vallisneria americana leaves for summer (a) compared with Thalassia and upland marsh decomposi­ tion, and for winter (b). Months are numbered sequentially starting in May, 1978. (l=May, 2=June, 3=July, 4=August, 5=September, 6=0ctober, 7=November, 8=December, y=January, 10=February, ll=March, 12=April, 13= May, and 14=June.) 108-109

Chapter 4

Figure 1. Location of the study area in Lake Pontchartrain, Louisiana. 136 Figure 2. Monthly Vallisneria biomass versus distance from shore and time at the control area (a) and the nitrogen enrichment area. (Biomass values = grams dry weight/square meter). 138 Figure 3. Net annual production estimated by the simple harvest method at the control area (C) and at the nitrogen enrichment area (N). 140 Figure 4. Environmental parameters versus distance from shore (m)• 142-144 Figure 5. Water depth at 25m, 62.5m, and 100 m from shore versus time at the control area (a) and the nitrogen enrichment area (b). 146 Figure 6 . Secchi depth versus time at the control area (C) and the nitrogen enrichment area (N). 148 Figure 7. Solar radiation at the sediment surface (langleys) at 25m, 62.3m, and 100m from shore versus time at the control area (a) and the nitrogen enrichment area (b). 150 Figure 8 . Maximum (a) and minimum (b) water temperature ( C) at 25m, 62.5m, and lUOm from shore versus time. 152 Figure 9. Seedlings of Vallisneria americana emerging from fruit: a) showing seedling emerging from broken portion of the fruit; b) showing seedling emerging from the broken end of a fruit. 154 vii Figure 10. Clumps of seedlings still held together by the remnants of a decaying fruit: a) decaying fruit plus seedlings; b) small clump of seedlings held together in a gelatinous mass with only small portion of fruit remaining. 156

Chapter 5

Figure 1. Location map of the study area in Lake Pontchartrain. 225 Figure 2 . Conceptual model of nitrogen cycling in a Vallisneria grassbed. 227 Figure 3. Modeling symbols used in nitrogen models (after H. T. Odum, 1971, and Hall, et al., 1977). 229 Figure 4. Growth model (GROW) for a Vallisneria americana grassbed. 231 Figure 5. Decomposition model (DECAY) for a Vallisneria americana grassbed. 233 Figure 6 . Production versus available inor­ ganic nitrogen in the sediment, used in order to obtain values for maximum velocity of nitrogen uptake *v m a x ** and the half velocity (V. • 235 Figure 7. Results of measured (M) versus sim­ ulated (S) leaf nitrogen. 237 Figure 8 . Results of measured (M) versus simulated (S) root and thizome nitrogen. 239 Figure 9. Results for measured (M) versus simulated (S) organic nitrogen. 241 Figure 10 Results for measured (M) versus simulated (S) inorganic nitrogen. 243 Figure 11 Results for measured (M) versus simulated (S) sediment organic nitrogen. 245 Figure 12 Results for measured (M) versus simulated (S) sediment inorganic nitrogen. 247 Figure 13 Results of GROW sensitivity analysis for leaves. 249 Figure 14 Results of GROW sensitivity analysis for roots and rhizomes. 251 Figure 15 Results of simulating water organic nitrogen with and without allocthanous inputs. 253 Figure 16 Results of simulating sediment inorganic nitrogen using two different temperature and oxygen regimes. 255 viii Figure 17. Results for measured (M) versus simulated (S) leaf nitrogen for 50m from shore at the nitrogen enrichment area. 257 Figure 18. Results for measured (M) versus simulated (S) root and rhizome nitrogen for 50m from shore at the nitrogen enrichment area. 259 Figure 19. Results of altering sunlight reaching the bottom on leaf nitrogen. 261 Figure 20. Results of altering sunlight reaching the bottom on root and rhizome nitrogen. 263

ix LIST OF TABLES

I. CHAPTER 3

Table 1. Mean biomasa and analysis of variance data used in testing for differences in Vallisneria americana biomass with time, distance from shore, and area along the shoreline, for all biomass stations (^193). 89

Table 2. Annual production of the Vallisneria americana beds measured by four harvest techniques at four distances from shore in a control and a nitrogen addition area. 80

Table 3. Comparison of Lake Pontchartrain Vallisneria americana grass bed production and standing stocks with tropical and temperate grass bed estimates. 91

II. CHAPTER 4

Table 1. Correlation of environmental parameters with production and biomass. 132

Table 2. Sediment and water column nitrogen sources compared with plant growth requirements by months for 1978-1979 season for 50m from shore at the control area. 133

Table 3. Presence (+) or absence (-) of Vallisneria leaves, roots, rhizomes, flowers, fruit, Najas , and Ruppia from May, 1978, through June, 1979, at the Lake Pontchartrain grass bed. 134

III. CHAPTER 5

Table 1. Equations used in GROW and DECAY. Initial conditions and numerical coefficients of flows (k values) used in the equations are in Tables 3 and 4. Abbreviations are explained in the text. 205 X Table 2 . Values of forcing functions used in GROW and DECAY. 206

Table 3 . Initial conditions of the state variables in GROW and DECAY. 209 Table 4 . Numerical coeficients (k's) of flows in GROW and DECAY (Figures 4 and 5, Table 3). Abbreviations in text. 211

Table 5 . Simulated leaf and root and rhizome nitrogen storages (L and RR) for original forcing functions compared with L and RR for forcing functions plus 10% of their initial value. The number shown is the correlation coefficient. The closer the correlation coefficient is to one, the closer the amended simulation is to the original or the less sensitive the model's response is to changing the forcing function. All probabilities are less than 0.0003. 220

Table 6 . Results of sensitivity analysis of processes by altering coefficients of flows in GROW and DECAY. The greater the difference in peak or low values between the original and the altered condition, the greater the sensitivity of the model to that alteration. 221

Table 7 . GROW initial conditions and forcing functions used in the simulation of leaf and root and rhizome nitrogen of a nitrogen enriched grass bed 50m from shore at the nitrogen addition area. 223

xi ABSTRACT

Vallisneria americana grass beds of Lake

Pontchartrain contribute approximately 200 g dry 2 weight/m /yr primary production, a continuous

release of detrital material, and rapid nutrient

recycling. A harvest method was used to estimate production. The addition of litterfall to harvest production raised annual production estimates by up to

23%. When litterfall was corrected for decomposition,

the production estimates were raised by up to 85% over harvest estimates alone. Litterfall and decomposition

9 increased toward shore and occurred throughout the

year providing a continuous supply of energy and

nutrients. Grass bed production varied along the

coastline and was greatest at mid-depths.

Values of the environmenal factors were most

extreme near and far from shore. Severe environmental

conditions are related to the patterns of biomass and production. The bell-shaped depth distribution pattern is perpetuated annually by survival of grass

and by dormancy of fruit during the winter at mid-depths. The majority of fruit are produced at mid-depths where biomass and production are greatest.

The surviving plants or new shoots start growing first

at mid-depths, and later radiate out via the rhizome

system to revegetate the entire grass bed. Although

x i i depth distribution patterns are similar along the

coastline, the grass bed extends almost twice as far

from shore at the control area as at the nitrogen

enriched area. Water clarity was significantly

greater at the control area than at the nitrogen

enriched area. Light penetration may determine the

compensation depth of the grass bed.

Seasonal fluctuations of leaf and root and

rhizome nitrogen simulated by GROW are very similar to

field experimental values. GROW was most sensitive to

the forcing functions light, temperature, and sediment nitrogen. GROW was validated using initial conditions

from the nitrogen enriched grass bed and had

correlations better than 80% between simulated and measured leaf and root and rhizome nitrogen. Once

validated, GROW was run under normal and reduced light

conditions. When light reaching the sediment surface was reduced by 50%, the Vallisneria leaf and root and

rhizome nitrogen declined to zero within a year. This

suggests that conditions that reduce water clarity

could cause the loss of the grass beds.

xiii CHAPTER 1

INTRODUCTION

Seagrasses are a valuable natural resource growing in the shallow subtidal waters along protected coastal and estuarine shorelines. A multitude of ribbonlike leaves overlap and intertwine to form a

living carpet which unobtrusively undulates with the ebb and flow of tides and yet rivals productivity rates of most ecosystems in the world. Ironically, although grass bed's importance to fisheries was suggested by Peterson as early as 1918, their profound influence on processes occurring at the land-sea interface was not studied until the widespread death of Zostera grass beds in the North Atlantic in the

1930's. Since then, data documenting the many useful ecological functions of grass beds has increased while the grass beds themselves have declined due to human-induced disturbances.

Seagrass beds contribute to both biological and physical processes shaping vegetated intertidal and subtidal areas. Where seagrasses are abundant they are prolific primary producers (McRoy and McMillan,

1977). Distinctive associations of epiphytes growing on the seagrass leaf blades are also important contributors to seagrass community production. While seagrasses are rarely grazed, their rapid turnover and decomposition favor nutrient regeneration (Kenworthy, 1 et al ., 1982) and a detrital food chain (Zieman,

1968). The food supplied by epibionts and detritus

contributes to the diversity and productivity of the

secondary consumers in the grass beds. Thus, ample

food supplies as well as protection from predators

within the thick growth combine to make seagrasses an

excellent habitat and nursery ground for juvenile

forms of many commercially important fish and

shellfish species (Hoese and Jones, 1963; Kikuchi,

1974). Beds of submerged grasses are also a habitat

for waterfowl and muskrats (Perry, et al ., 1981).

Physically, the grass blades have a baffling effect on

water movements which reduces wave amplitudes and

water velocities within the beds. Thus, grass beds

aid in protecting adjacent shorelines from erosion and

locally increase sedimentation rates of fine material

(Scoffin, 1970), and thereby create a stable and

distinctive sediment environment (Orth, 1977).

Seagrass distribution and growth are regulated by

environmental factors such as available light,

turbidity, temperature, salinity, currents, wave

action, water depth, available nutrients, and sediment type (Moore, 1963; Westlake, 1967; Patriquin and

Knowles, 1972; and Phillips, 1974). Generally,

seagrasses tend to occur along low to moderate energy

coasts and grow out from shore to a depth determined by available light. Extreme temperatures, salinities or wave action are detrimental to grass bed

development.

Man's increasing development of coastlines into

residential, commercial and industrial sites has

altered nearshore physical and chemical environments

and resulted in the decline and loss of grass beds.

Turbidity increases, thermal effluents,

eutrophication, toxic chemical and waste additions,

and physical disturbances are among the alterations

related to human activities which are detrimental to

grass beds.

Human activities in Lake Pontchartrain are

related to the decline of the lake's grass beds over the last 25 years although the precise reasons for the decline are unknown (Turner, et al. , 1980). The grass bed's response to environmental factors and perturbations needs to be studied in order to development management criteria. Therefore, the annual growth and nutrient dynamics of a submerged

Vallisneria americana grass bed in Lake

Pontchartrain, Louisiana, were studied from May, 1978, through May, 1979. The Vallisneria grass bed, along with nearby marsh, swamp, and upland ecosystems, helps support a commercial fishery in Lake Pontchartrain which is valued at more than a million dollars per year (Thompson and Verret, 1980). Management of the

Lake Pontchartrain area must include preserving the grass bed ecosystem in order to perpetuate its fishery

potential.

The hypothesis tested in this paper is that grass

bed annual production depends on available light*

available nutrients* water depth, turbidity*

circulation and temperature which change with distance

from shore. Six major objectives were addressed in

this study:

1) To develop techniques to accurately measure growth in this submerged grass bed. 2) To measure overall productivity of the grass beds. 3) To describe the annual growth patterns of Vallisneria . 4) To identify variables which control growth of submerged grass beds. 5) To quantify macronutrient dynamics in the grass bed over an annual growth cycle. 6 ) To model nitrogen dynamics of the grass beds.

' 1.0 ORIGIN OF LAKE PONTCHARTRAIN GRASSBEDS

The location of grass beds in Lake Pontchartrain is determined by the development of sand beaches which were formed from five to ten thousand years ago. This concept of past geomorphic history determining grass bed distribution was first developed by Zieman (1972) who found that the circular grass beds along the south

Florida coast occured where they do because of a combination of rising sea level, a transgressing shoreline and submerged mangrove hammocks. These events led to shallow, circular sediment deposits advantageous for grass bed colonization. Since grass beds require shallow depths and particular sediment

types# it is not too surprising that physical

processes which operate to produce these conditions

are in turn responsible for the estabishment of grass

beds.

The geomorphic history of Lake Pontchartrain is

tied to the relationship of the marine waters of the

Gulf of Mexico with the shifting courses and

distributaries of the Mississippi River (Saucier#

1963) ? Lake Pontchartrain and Lake Maurepas occupy most of the Pontchartrain Basin (Figure 1)# a large

lowland area at the eastern edge of the deltaic plain of southeastern Louisiana. Most of the sediments in

the area are alluvia from the Mississippi River and rivers to the north of Lake Pontchartrain. Underlying these sediments and surfacing to the north and west of the basin is the Pleistocene Prairie Terrace.

The Milton*8 Island and Pine Island beach trends were instrumental in constructing the shorelines of

Lake Pontchartrain while it was still a coastal embayment. The Late Wisconsin Glacial Stage lowered sealevel to about 150m below the present level. About

30#000 years ago sea level began to rise again. As sea level rose two sand beach trends were initiated.

The rapid rise of sea level to its current level

(about 5,000 years ago)# caused the mouth of the

Mississippi River to move inland to near the present site of Baton Rouge. Southeastern Louisiana was

dominated by a shallow sea (Figure 2a) which has been

called the Pontchartrain Embayment (Fisk, 1947). The

Pine Island Beach Trend was started by a steep slope

caused by faults, and the Milton's Island Beach Trend

was started by irregularities in the Prairie

formation, both deflecting rising waters. These two

fingers of sand spread westward from near the Pearl

River to form the Milton's Island and Pine Island

beach trends (Figure 2b,c, and d). Sediment from the

Mississippi River quickly filled the shallow sea,

forming the Cocodrie Delta. The Cocodrie Delta

deteriorated after the Mississippi's course shifted to

the west and was cut off from the Lake Pontchartrain

area by levees and ridges. The Pine Island beach

trend and the Milton's Island beach trend maintained

the present location of the shorelines of Lake

Pontchartrain during the delta deterioration

(geomorphic history from Saucier, 1963).

1.1 GRASSBED DEVELOPMENT RELATED TO SAND BEACH TRENDS

The sand beach deposits of the Milton's Island and Pine Island beach trends were instrumental in

forming submerged grass bed habitats as well as

forming the shoreline configuration of Lake

Pontchartrain. Sand and silty-sand deposits are found today in the general locations of the Pine Island (southeastern lake shore) and Milton's Island

(northeastern lake shore) beach trends (Fig. 3).

These are probably ideal substrates for the submerged grass species ( Vallisneria americana (Michaux) and

Thalassia testudinum (Konig)) found in Lake

Pontchartrain, since Thalassia has the best survival when planted on sandy to silty-sand substrate

(Steller, 1976). The Lake Pontchartrain grass beds occur in almost perfect correspondence with the sandy beach trend deposits (Figure 4). Thus, the origin of the largest expanse of grass beds in Louisiana was tied to the development of sandy beach trend deposits in Lake Pontchartrain.

1.2 GRASSBED DECLINE

The formerly lush grass beds which mark the beach trends are not as extensive today as they once were, probably due to the effects of increasing urbanization and industrialization. Using, aerial photos taken at the maximum extent of annual standing stock in October of 1979, I estimated that less than 2,000 acres of grass beds were still present compared to the 20,000 acres previously recorded (Perret, et al ., 1971).

Most of the loss of grass beds was along the south shore of Lake Pontchartrain. As seen in Figures 5 and

6 , urban and industrial growth to the south of Lake

Pontchartrain has been extremely rapid in the last 25 years. New Orleans, which fronts the southern shore of Lake Pontchartrain, is the largest urban-industrial complex in Louisiana. Storm water drainage and treated sewage from this large metropolitan area are pumped directly into the lake. Numerous petro-chemical plants along the Mississippi River add chemical wastes which may reach the lake via the Inner

Harbor Navigation Canal (1HRC). The IHRC also connects the lake to the Mississippi River Gulf Outlet

Canal and the Intracoastal Waterway. These navigation canals introduce saline water. The resulting increase in the turbidity, salinity, and nutrient and pollutant loading to the lake have been postulated as reasons for the grass bed's decline (Turner, et al ., 1980).

This study was undertaken to better understand the functioning of the lake's grass bed ecosystem in order to help planners protect the grass beds which are important to Lake Pontchartrain's fishery. This general objective is addressed in the following chapters. Chapter 2 is a literature summary; Chapter

3 considers the productivity of the grass beds;

Chapter 4 describes the phenology of the species as it relates to survival and/or success in its environment; and in Chapter 5 a model of nitrogen cycling in

Vallisneria is developed. In Chapter 6 the

% conclusions from Chapters 3-5 are summarized and discussed to suggest the best practices to enhance

Lake Pontchartrain grass beds. 1.3 REFERENCES CITED

Bahr, L. M., J. P Sikora, and W. B. Sikora. 1980. Macrobenthic survey of Lake Pontchartrain, Louisiana, 1978. pp. 650-710. I_n : J. H. Stone (Ed.). Environmental Analysis of Lake Pontchartrain , Louisiana , Its Surrounding Wetlands , And Selected Land Uses . Vol II. U.S. Army Engineer Dist. N.O. Contract No. DACW 29-77-C-0253.

Fisk, H. N. 1947. Fine-grained alluvial deposits and their effects on Mississippi River activity. Waterways Experiment Station. Vicksburg, Mississippi. 82 pp.

Hoese, H. D. and R. S. Jones. 1963. Seasonality of larger animals in a Texas turtle grass community. Publ. Inst. Mar. Sci. Univ. of Texas. 9:37-47.

Kenworthy, W. J., J. C. Zieman, and G. W. Thayer. 1982. Evidence for the influence of seagrasses on benthic nitrogen cycle in a coastal plain estuary near Beaufort, North Carolina (U.S.A.). Oecologia, 54:152-158.

Kikuchi, T. 1974. Japanese contributions on consumer ecology in eelgrass ( Zostera marine L.) beds, with special reference to trophic relationships and resources in inshore fisheries. Aquaculture, 4:145-160.

McRoy, C. P. and C. McMillan. 1977. Production ecology and physiology of seagrasses. pp. 53-88. Ir: : C. P. McRoy and C. Helfferich (Eds.). Seagras Ecosystems. A Scientific Perspective . Vol 4. Marcel Dekker, N.Y.

Orth, R. J. 1977. The importance of sediment stability in seagrass communities, pp. 281-300. Iji : B. C. Couil (Ed.). Ecology of Marine Benthos . Univ. S. Carolina Press, Columbia, S. C.

Perret, W., B. Barrett, W. Latapie, J. Pollard, W. Mock, G. Adkins, W. Gaidry, and C. White. 1971. Cooperative Gulf of Mexico Estuarine Inventory and Study, Louisiana, Phase 1, Area Description, Louisiana Wild Life and Fisheries Comm., New Orleans.

Perry, M. C., R. E. Munro, and G. M. Haramis. 1981. Twenty-five year trends in Chesapeake Bay diving duck populations. Presented at 46th N. Amer. Wildl. Nat. Res. Conf.

Peterson* C. G. J. 1918. The sea bottom and its production of fish food. A survey of the work done in connection with the valuation of the Danish waters from 1883-1917. Rept. Danish Biol. Sta. 25:1-82.

Saucier* R. T. 1963. Recent Geomorphic History * of the Pontchartrain Basin.. L.S.U. Press. Baton Rouge. TT3 pp.

Scoffin* T. P. 1970. The trapping and binding of subtidal carbonate sediments by marine vegetation in Bimini Lagoon* Bahamas. Jour. Sed. Pet. 40(1):245-273.

Steller* D. 1976. Factors affecting the survival of transplanted Thalassia testudinum • pp. 1-20. In : (D. Cole, Ed.")* Proceedings of the Third Annual Conference on The Restoration of Coastal VegetatTon Tn Florida. Hillsborough Community College* Tampa* Florida.

Thompson* B. A. and J. S. Verret. 1980. Nekton of Lake Pontchartrain* Louisiana* and Its Surrounding Wetlands, pp. 711-865. Iri : (J. Stone* Ed.), Environmental Analysis of Lake Pontchartrain*' Louisiana* Its Surrounding Wetlands* and Selected Land Uses. Vol. 2. U.S. Army Engineer District* New Orleans, Contract No. DACW-29-77-C-0253.

Turner* R. E., R. M. Darnell and J. R. Bond. 1980. Changes in the submerged macrophytes of Lake Pontchartrain (Louisiana): 1954 - 1973. pp. 647 - 658. In :(J. Stone* Ed.)* Environmental Analysis of Lake Pontchartrain* Louisiana, Its Surrounding Wetlands* and Selected Land Uses. Vol. 2. U.S. Army Engineer District, New Orleans, Contract No. D&CW-29-77-C-0253.

Turner* R. E.* and J. R. Bond. 1980. Changes in the submerged macrophytes of Lake Pontchartrain (Louisiana): 1954-1973. Northeast Gulf Science 4(1):43-48.

Zieman* J. C. 1968. A study of the growth and decomposition of the sea-grass Thalassia testudinum . M. S. Thesis. Univ. of Miami, Coral Gables, Fla. 50pp;

Zieman, J. C. 1972. Origins of circular beds of Thalassia (Spermatophytas Hydrocharitaceae) in South Biscayne Bay, Florida, and their relationship to mangrove hammocks. Bull. Mar Sci. 22(3):559-573. Figure 1. The physiography of the Pontchartrain Basin (from Saucier, 1963). B B E 3 I E 3 ffASIMW » H H - Figure 2. Origin of the Milton's Island and the Pine Island beach trends (from Saucier, 1963). psmi

.^.V.V.V.V.V.V.V.T

WIJHW im V” i’

/.v.w

C/I •• Figure 3. Distribution of sediment types in Lake Pontchartrain, La., as determined by Bahr, et al ., (1980).

I SILTY SAND a CLAYEY SAND

SANDY SILT CLAYEY SILT

CLAY SANDY CLAY □ SILTY CLAY Figure 4. Comparison of the Milton's Island and Pine Island beach trends with the location of submerged grass beds in Lake Pontchartrain, Louisiana. 19 Figure 5. Land use in the Lake Pontchartrain watershed in 1954 (from Turner and Bond, 1980).

I Figure 6. Land use in the Lake Pontchartrain watershed in 1972 (from Turner and Bond, 1980). 23

Pi CHAPTER 2

LITERATURE REVIEW

2.0 INTRODUCTION

Seagrasses are angiosperms adapted to shallow

seas and estuaries. Functions of seagrass ecosystems

include primary production, nursery grounds for fish

and shellfish, wave abatement, and shoreline stabilization. Seagrasses usually have well developed rhizome systems and linear or strap-shaped leaves (den Hartog, 1967). In North America, Zostera marina L. and Thalassia testudinum are the most abundant temperate and tropical seagrasses, respectively. Traditionally, seagrasses have included approximately 50 saline species in two families,

Potamogetonaceae and Hydrocharitaceae (den Hartog,

1977). A member of the Hydrocharitaceae, Vallisneria americana , growing in slightly brackish estuarine environments with functions similar to saline seagrasses has been included in the "seagrass" classification (Steller, 1981; and Kemp, et al .,

1984).

Seagrasses first appeared in the Cretaceous and developed their own unique assemblage of fauna and infauna. Due to rapid decomposition of the plants, the only permanent record of seagrass beds is the remains of their uniquely associated animals. The

24 25 rich fauna and infauna of seagrass beds have left a distinctive assemblage throughout the geologic record

(Brasier, 1975).

Although begun in the early 1900's

(Peterson,1918), seagrass research was rekindled as a result of the International Seagrass Workshop in 1973 and has attracted funding by the National Science

Foundation and the International Decade of Ocean

Exploration. Several general reviews of seagrass literature are now available (Phillips, 1960; Odum and

Copeland, 1964; MeRoy and Helffrich, 1977; Phillips and MeRoy, 1980; and Mann, 1982).

2.1 PRODUCTIVITY AND DISTRIBUTION RELATED TO

ENVIRONMENTAL FACTORS

Relative abundance, distribution and productivity are affected by environmental factors such as light, depth, temperature, salinity, currents and wave action, and sediment type (Phillips, 1960; Moore,

1963). Man's increasing influence in the coastal zone is altering many of the physical factors which determine seagrass distribution and productivity.

Available light and suitable substrate are the two most important physical factors determining seagrass distribution (Thayer, et al. , 1975). Many studies have shown that seagrasses decline in areas of increased anthropogenic pollution (Kikuchi, 1974; Phillips, 1974; Peres and Picard, 1975; Backman and

Barilotti, 1976; Phillips, et al. , 1978; Kemp, et al. , 1983).

Where environmental conditions are optimal seagrasses may be highly productive (Odum, 1956; Odum and Hoskins, 1958; Westlake, 1963; Ikusima, 1965,

1966; Jones, 1968; Qasim and Bhattathiri, 1971; Mann,

1972; McRoy and McMillan, 1977; and Zieman and Wetzel

1980). Standing crops range from 100 to 1000 grams 2 carbon per meter squared (gC/m ), except in 2 brackish areas where they rarely exceed 500 gC/m .

Daily rates of productivity range from 1 to 15 gC/m /day, so seagrasses are potentially as productive as upland grasslands where conditions are favorable.

2.1.1 Light

The intensity of photosynthetically active radiation (PAR) available for plant growth determines compensation depth and the upper limits of submerged grassbed production when other growth factors are not limiting. Sand-Jensen and Borum (1982, after Borum,

1980) reported that 75% of the variation in leaf growth rate in Zostera marina could be explained by surface irradiance. Light decreases with water depth and turbidity and, thus, restricts seagrass distribution. Light is the most limiting 27

environmental factor in turbid waters (Congdon and

McComb, 1979). Seagrasses adapt to different light

conditions by morphological changes (e.g., the shape

and length of leaves) or physiological changes (e.g., pigment distribution and photosynthetic efficiency)

(Wiginton and McMillan, 1979).

Morphology of seagrasses may determine productivity in different environments. Vallisneria grows from a basal rosette and elongates to the water surface, but generally does not form a surface canopy.

Vertical orientation allows more light to penetrate the water column than a surface canopy (light extinction coefficients of 0.0126 for vertical vs.

0.0235 for canopy in m /g dry wt leaf tissue)

(Titus and Adams, 1979). Where Vallisneria has a vertical orientation, 62% of Vallisneria biomass is within 30cm of the sediment surface (Titus and Adams,

1979). Other submerged plants, such as Myriophyllum , maintain the majority of their biomass near the water surface (68% within 30cm of water surface) by elongating and then sloughing their lower leaves.

Titus and Adams (1979) state that "we view the growth form of V^ americana as a constraint on its fitness in eutrophic waters with low transparency and dense macrophyte growths" (p. 274). They found that formerly pure stands of Vallisneria were being invaded 28 by Myriophyllum in eutrophic areas. This phenomena of replacement of strap-shaped vertical leaf systems by floating leaf plants also occurs annually in Thalassia grassbeds in the Gulf of Mexico which are invaded by floating algae during the winter months when light levels are lower (Zimmerman and Livingston, 1976b).

Seagrasses show a physiological response to different light levels by changing their pigment distributions and photosynthetic efficiency. For example, at low light levels Vallisneria shows shade-tolerant responses— i.e., it acclimates to low light levels, but it can re-acclimate within 24 hours of exposure to higher light levels (Titus and Adams,

1979). Such plasticity allows Vallisneria to survive during low levels of insolation and to adapt to higher light levels of relatively clear waters when PAR increases above saturation (400 to 600 uE/m * s). Productivity is also related to leaf age since chlorophyll content and activity are highest in mature Vallisneria leaves (Ikusima, 1966) and Thalassia leaves (Zieman, 1968). Light penetration may be inhibited by heavy epiphyte growth on older, senescent leaves (Sand-Jensen, 1977).

However, leaves are continually produced and older ones are replaced by new growth in Vallisneria

(Carpenter, 1980). Leaf production rates vary 29

significantly with productivity levels (Sand-Jensen,

1975), and, thus, during the summer when light levels

and productivity are high, new epiphyte-free leaves

are also abundant.

A combination of light and depth determine the

blade length and productivity of seagrasses. Where

water is shallow seagrass blades are generally shorter

and thinner (Zieman, 1972). Seagrass blade length

increases with depth until light becomes limiting. A

typical depth gradient of seagrasses would include

thin, strap-shaped leaves near shore, longer, thicker

strap-shaped leaves at intermediate depths, and short

ovate-shaped leaves farthest from shore (Strawn, 1961;

and den Hartog, 1970 and 1977). Patriquin (1973)

found a positive relationship in Thalassia between

leaf length and above and below ground production.

Thus, maximum production occurs at intermediate depths where leaves are longest.

2.1.2 Temperature

Temperature affects the rate of chemical and biochemical reactions, including photosynthesis and

respiration. Maximum daily rates of productivity of

seagrasses are similar throughout their latitudinal

range. Therefore, annual production is highest where

the most days with conditions favorable for growth

occur. Seagrass annual production is highest in the tropics where annual solar insolation is high and temperatures are within the optimal range for plant growth and rarely goes below or above critical limits.

Temperatures above optimal cause production to slow due to increased respiration rates and a breakdown of epidermal tissues (Jones, 1968; Schroeder, 1975).

Carbon uptake is limited below 10°C in subtropical areas. Generally the temperature for optimal growth increases toward the tropics. For example, rates of production reach a maximum at around 20°C for temperate Zostera (Thayer, et al ., 1975a) while they reach a maximum around 30°C for tropical Thalassia

(Zieman, 1975).

Vallisneria1s ability to carry on photosynthesis has been found to be specialized for midsummer temperatures (Titus and Adams, 1979). In response to altered temperatures, Vallisneria1s relative net photosynthetic rate increases from 10°C to the optimum rate at 32°C, above which it rapidly declines (Titus and Adams, 1979). This temperature response tested in the laboratory was consistent with the plant's phenology, because Vallisneria only had photosynthetic tissues in the water column for the four warmest months of the year in University Bay,

Wisconsin. This temperature range (10°C-32°C) is similar to field conditions encountered in Lake 31

Pontchartrain, Louisiana.

2.1.3 Salinity

Although most grassbed species have a salinity

optimum, saline species can generally survive in a

wide range of salinities. Most taxa have exchange of

water and selective ions across leaf surfaces due to a

lack of a cuticle. This exchange allows seagrasses to

withstand moderate variations in salinity. However,

most taxa have an optimal salinity range. Thalassia

has optimal photosynthesis at about 35 parts per

thousand (ppt) (Hammer, 1968) and can withstand from 0

ppt to 70 ppt (Gessner, 1971). Vallisneria is

basically a fresh to brackish water species and its

productivity increases from 0-8 ppt salinity.

Vallisneria does not grow at higher salinities. High

salinities increase the plant's respiration and may

limit net production. Hypersaline conditions may

compound the effects of thermal stress (Schroeder,

1975). McMillan and Moseley (1967) found that

Syringodium , Thalassia , Ruppia , and Halodule ,

respectively, were increasingly tolerant to hypersalinity. However, Ruppia also grows in the brackish environment in Lake Pontchartrain.

2.1.4 Currents and Wave Action

Density of the stand as well as net and gross primary productivities increase with current to a 32

maximum at about 1 knot (Westlake, 1963; Conover,

1968; and Hay, Wiedenmeyer and Marsalek, 1970), This

increase in productivity is related to disruption of

the stagnant zone next to the leaves. Circulation

increases nutrient availability and gaseous exchange

across the leaf surfaces. Because physical damage to

the plants may occur at high water velocities,

seagrasses generally occur in areas of moderate wave

energies (Moore, 1963).

2.1.5 Sediment

Seagrass beds are distributed among various

sediment types ranging from fine-grained clay to sand and coral cobble but they are generally found on a mix of sand and silt. Seagrasses actively change their

root zone sediments by producing, binding and trapping

fine-grained organic rich material (Ginsburg and

Lowenstam, 1968; Orth, 1977). Peak production occurs in the center of the seagrass beds where the fine-grained, organic rich sediments collect (Zieman,

1975).

2.2 LIMITING NUTRIENTS

2.2.1 Carbon

Carbon may be a limiting factor to growth since less carbon dioxide is dissolved in water than is available in the air, particularly in low salinity regions of estuaries (Kemp, 1983). In compensation, 33 some submerged species may use bicarbonate (Beer, et al ., 1977). Dissolved carbon dioxide or bicarbonate is absorbed either through the leases or roots. The relative absorbtion by leaves or roots is determined by the higher available concentration in the water column or interstitial waters (Sondergaard and

Sand-Jensen, 1979).

Epiphytism common to older seagrass leaves may impede leaf carbon uptake (Sand-Jensen, 1977). The mechanism of continual leaf production and sloughing assures a continual supply of new, epiphyte-free leaf area available for carbon exchange.

2.2.2 Nitrogen

Nitrogen may be a limiting nutrient for seagrass production and is quickly cycled through the seagrass ecosystem. It may be taken up via the leaves or roots in either oxidized or anaerobic forms. Uptake kinetics approximate the Michaelis-Menton hyperbolic response (McRoy and Alexander, 1975). The majority of the nitrogen assimilated generally comes from available nitrogen in the sediment or water column.

However, nitrogen fixation may be an important contributor to available nitrogen where other nitrogen sources are limiting.

The relative importance of nitrogen fixation is a controversial issue. Patriquin and Knowles (1972) 34 suggested that nitrogen fixation may be the primary source of nitrogen for Thalassia testudinum .

However, because they only sampled a single site in

Barbados, and their samples were transported over long distances (to Canada), were not immediately assayed (a two day delay before using the acetylene reduction technique), &nd were incubated for a long period of time (two to three days), their data have been questioned by McRoy, et al . (1973) who found only low levels of nitrogen fixation under similar conditions.

Although its relative importance to the plant's nitrogen budget has been questioned, nitrogen fixation in seagrass beds has been well documented. Nitrogen fixation may occur on the leaves (phyllosphere), in the root zone (rhizosphere) or on detritus (Patriquin and Knowles, 1972; Goering and Parker, 1972; Capone and Taylor, 1980; and Capone, et a^ ., 1979).

Nitrogen fixation in the phyllosphere has been attributed to the epiphytes growing on the leaves

(Goering and Parker, 1972). Goering and Parker (1972) recorded very high rates of nitrogen fixation (up to

300 mgN/m /day) on Thalassia , Cymodocea , Ruppia

,and Diplanthera leaves which were heavily colonized by Calothrix , a heterocystous cyanobacteria. When the leaf blades were wiped clean of epiphytes, no 35 nitrogen fixation was recorded - a result suggesting that the majority of nitrogen fixation was from the epiphytes, not the macrophytic plants. Capone and

Taylor (1977) also determined that nitrogen fixation associated with leaves of Thalassia was correlated with the presence of Calothrix and that maximum rates occurred during early summer and were related to light conditions. However, Capone and Taylor (1980) suggested that the condition of the plant sample and the size of the incubation container affected the t relative rate of fixation. When they used small pieces of Thalassia heavily encrusted with epiphytes

(containing Calothrix ) in a small (21ml) container they also recorded close to 300 mgN/m /day.

However, when they incubated whole leaves in larger containers (900ml), they recorded more realistic fixation rates for whole shoots that were about one third the previous rates (Capone and Taylor, 1980).

Capone and Taylor (1980) also suggest that leaves and their epiphytes are independent of each other with respect to nitrogen metabolism.

Rates of nitrogen fixation generally are higher in rhizosphere than in non-rhizosphere sediments

(Patriquin and Knowles, 1972; Brooks, et al ., 1971).

Even though Patriquin and Knowles (1972) may have enhanced nitrogen fixation rates by their techniques, Capone and Taylor (1980) noted nitrogen fixation rates at the low end of the range of values found by

Patriquin and Knowles (1972) near Bath, Barbados.

Capone and Taylor (1980) suggest that up to 50% of the nitrogen necessary for leaf production may be supplied by nitrogen fixation in the rhizosphere. The lowest rates of nitrogen fixation assayed were associated with detritus (Capone and Taylor, 1980). The variability of rates of nitrogen fixation are affected by factors such as available nitrogen in the sediment, abundance of nitrogen-fixers, available light, and oxygen (Buresh, e_t al ., 1980; Capone and Taylor,

1980). Nitrogen fixation does not always accompany seagrass beds and has been found to be low in the vicinity of some seagrass beds where other sources of nitrogen are not limiting (Lipschultz, et al .,

1979).

Sources of nitrogen in the sediment and overlying waters may vary seasonally. The maximum nitrogen uptake rate via leaves is approximately 10-20 ug-at/l.

Higher uptake rates occur in sediments where nitrogen supplies are greater. Evidence presented by Patriquin

(1972) suggests that the majority of nitrogen for seagrass growth comes from sediment because waters are generally nitrogen poor.

Nitrogen is quickly cycled in grassbeds since 37 uptake is rapid via leaves or roots, litterfall is continuous and and decomposition is rapid and complete

(Zieman, 1968). A large and diverse epifauna and infauna consume detritus (0'Gower and Wacasey, 1967).

The detritivores, in turn, are preyed upon by fish and crustaceans. Thus, sediment and water supplies which are quickly depleted by plant uptake are also quickly restocked by mobilization of inorganic nitrogen through decomposition.

2.2.3 Phosphorus

Phosphorus may also be a limiting factor for seagrass growth. Patriquin (1972) calculated that most of the phosphorus necessary for growth could be supplied by sediments. McRoy, et al . (1972) found both sediment and water to be important suppliers of phosphorus. Gentner (1977) found that Vallisneria took up phosphate equally by roots and shoots.

Seagrasses may act as phosphorus pumps by translocating phosphorus taken up from the sediments to the leaves where it may be exchanged with the water column (Barko and Smart, 1980 and 1981). However, translocation of phosphate occurs in both the root to shoot and shoot to root directions in Vallisneria and is highly variable according to radio-isotope tracer studies (Gentner, 1977).

2.2.4 Techniques There are a number of techniques available to determine whether water or sediment nutrient supplies are limiting factors for seagrass growth. The ratio of leaf to underground biomass indicates the relative importance of water or sediment nutrient supplies and is directly related to growth rate (Barko and Smart#

1978). Patriquin (1972) found that available nitrogen supplies could support growth for 0.3 days from water and 10 days from sediments based on measured productivity. Available phosphorus supplies could support growth for 3 days from water and 500 days from sediment supplies. These numbers give a simple estimate of which nutrient is likely to become limiting first. Gerloff and Krombholz, (1966) found that there was a critical level of nitrogen and phosphorus in plant tissue above which no more growth occurred (luxury consumption). Below the critical level of plant tissue nutrients maximum production did not occur. Fitzgerald (1968 and 1969) found that high ammonium uptake rates following a short period of starvation indicated a plant was nutrient limited prior to starvation. However# Boyd (1971) has shown that uptake rates and nitrogen and phosphorus leaf tissue contents follow an annual cycle and are typically higher in the spring. Thus, changes throughout the growing season make it necessary to 39

monitor seasonal patterns as well as absolute levels

in order to detect nutrient deficiencies. Also, since plant tissues contain many essential nutrients and

vitamins, assaying a single constituent and not the

entire suite of essential growth factors may not

necessarily give reliable results.

2.3 DECOMPOSITION

Decomposition leads to the eventual mineralization of the organic material. When leaves die the protein content drops at the time of death but rises sharply with bacterial colonization (Odum and

Heald, 1975). While live leaves of Zostera contain about 20% water soluble organics, recently dead leaves contain only 12% (Mann, 1972). The quick^loss of 8% dissolved organic carbon goes to bacterial degradation or is adsorbed onto particles and then degraded

(Fenchel, 1977). In Thalassia and Syringodium , the dissolved organic carbon immediately degraded was higher, 12.6% and 19.4%, respectively (Robertson, et al ., 1982). The relative proportion of quickly dissolved organic carbon determines the initial decomposition rate. After the soluble substances have dissolved, the remainder of the leaves are broken down by mechanical means and rapidly colonized by bacteria.

The attached microorganisms are assimilated by detrital feeders leaving the detrital material chemically unchanged (Hargrave, 1970b; Fenchel, 1970,

1972). Most detrital feeders lack digestive enzymes

capable of attacking structural organic material of

seagrass origin (Kristensen, 1972; Nielsen, 1966).

Rather, the effect of detrital feeding invertebrates

(mollusks, nematodes, etc.) is to increase the surface

area of the detritus by dividing the particles into

smaller units and thereby to increase bacterial and g fungal colonization. There are approximately 10

bacteria per gram dry weight of detritus (Fenchel,

1977). The number varies directly with surface area

and inversely with particle size (Fenchel, 1970),

oxygen content (Hargrave, 1972; Fenchel, 1972),

nutrient availability (Fenchel, 1972, 1977), and

competing bacterial consumers such as protozoa and

ciliates (Fenchel, 1970, 1977; Hargrave, 1970a).

Decomposition rate is faster under aerobic

conditions rather than in anaerobic conditions. Most

grassbeds occur in anaerobic sediments overlain by a

thin oxidized layer (Fenchel, 1969; Wood, 1965).

Decomposition probably occurs in the upper 20-30 cm of

the sediment where the dominant process is sulfate

reduction. The sulfate anion (SO^) is the most

abundant terminal hydrogen acceptor in the marine

environment. Other possible anaerobic hydrogen acceptors are carbon dioxide, nitrate and certain 41 bacteria. The products of anaerobic decomposition may diffuse upward or encounter oxidized areas near the roots and become oxidized and release energy. These decomposition processes occur rapidly in comparison with decomposition of emergent vegetation (Zieman,

1968, 1972? Wood, et al .,1969). If the detritus dries prior to decomposition the rate is even faster

(Burkholder and Doheny, 1968; Zieman, 1968).

2.4 COMMUNITY

Grassbeds contain considerably greater species diversity of infauna and epifauna than adjacent bare sand or mud bottoms (0'Gower and Wacasey, 1967).

Migratory species also take advantage of the shelter and abundant food in grass beds and further increase diversity by temporal niche displacement (Hoese and

Jones, 1965). The seasonality of fauna is interwoven with that of the grasses, which are most productive in early spring and fall (Darnell, 1961; Odum, 1967). A production lull may occur in summer due to the energy demands of flowering or exposure to high temperatures. Despite the almost total degradation of plant material, a distinctive sediment regime gradually accumulates (Scoffin, 1970). The distinctive sediment and associated faunal assemblage are bases for tracing the seagrass community through the geologic record (Brasier, 1975). 42

2.5 HUMAN INFLUENCE

Highly productive seagrass systems are usually located in highly desirable, shallow coastal waters which have been increasingly developed in recent years. Human influences on seagrass beds include addition of nutrients and pollutants, dredging, filling, and thermal, salinity and current modifications that alter the local ecology (Odum,

I960; Thayer, 1975a; Livingston, et al .,1976;

Turner, et al ., 1980; Schroeder, 1975; Peres and

Picard, 1975; Cambridge, 1975; and Zimmerman and

Livingson, 1976). Seagrasses are important to the local ecology of other nearshore habitats because they may contribute a major portion of the estuary's carbon source (Teas, 1976; Thayer, et al ., 1979).

Seagrasses decay faster than emergent species (Wood,

1969) and a larger percentage of production moves rapidly through the aquatic environment. Stored starch reserves (Zieman, 1975) and the ability to grow upward quickly may mitigate the effects of transient stresses such as sudden burial by dredging sediments

(Odum, 1963). However, long-term water quality degradation may cause grass beds to decline. For example, thermal stress from power plant cooling waters (Thorhaug, 1976) or lowered light levels due to dredging may result in loss of the vegetation and 43 succession to an algal community.

The need to make grass bed ecosystem data available for management decisions has become increasingly apparent. Models of growth and nutrient cycling under various environmental conditions can be helpful in defining and predicting grassbed responses.

Some recent modeling efforts include those of Titus. et al,./ (1975), Short (1980), and Wetzel (1985). 44

2.6 LITERATURE CITED

Backman, T. W. and P. C. Barilotti. 1976. Irradiance reduction: effects on standing crops of the eelgrass, Zostera marina in a coastal laqoon. Mar. Biol. 34:33-40.

Barko, J. W. and R. M. Smart. 1978. The growth and biomass distribution of the emergent freshwater plants, Cyperus esculentus and Scirpus validus , on different sediments. Aquat. Bot. 5:109-11?.

Barko, J. W. and R. M. Smart. 1980. Mobilization of sediment phosphorus by submerged freshwater macrophytes. Freshwater Biol. 10:229-238.

Barko, J. W. and R. M. Smart. 1981. Sediment based nutrition of submersed macrophytes. Aquat. Bot. 8:.

Beer, S., A. Eshel, and Y. Waisel. 1977. Carbon metabolism in seagrasses. J. Exp. Bot. 106:1180-1189.

Borum, J. 1980. Biomasse- og produktionsforholdhos Alegraes ( Zostera marina L.)og det tilknyttede epiohytsamfund. M.S. Thesis, Freshwater Biological Laboratory, Univ. of Copenhagen.

Boyd, C. 1971. The limnological role of aquatic macrophytes and their relationship to reservoir management. Reservoir Fish, and Limn. Spec. Pub. #8. Am. Fish. Soc. pp. 153-166.

Brasier, M. D. 1975. An outline history of seagrass communities. Paleontology. 18(4):681-702.

Brooks, R. H., P. L. Brezonik, H. D. Putnam, and M. A. Keirn. 1971. Nitrogen fixation in an estuarine environment:The Waccassassa on the Florida Gulf Coast. Limnol. Oceanogr. 16(5):701-715.

Buresh, R. J., M. E. Casselman, and W. H. Patrick, Jr. 1980. Nitrogen fixation in flooded soil systems, a review. Adv. Agron. 33:149-192.

Burkholder, P. R. and T. E. Doheny. 1968. The biology of eelgrass. Dept, of Cons, of Waterways, Town of Hempstead, Long. Is., N.Y. 120 pp.

Cambridge, M. L. 1975. Seagrasses of South-Western 45

Australia with special reference to the ecology of Posidonia australis Hook F. in a polluted environment. Aquat. Bot. 1:149-161.

Capone, D. G. and B. F. Taylor. 1977. Nitrogen fixation (acetylene reduction) in the phyllosphere of Thalassia testudinum . Mar. Biol. 40:19-28.

Capone, D. G. and B. F. Taylor. 1980. Microbial nitrogen cycling in a seagrass community, pp. 153-161. IN: V. S. Kennedy, (Ed.), Estuarine Perspectives. Papers presented at the 5th Annual International Estuarine Research Conf., Jekyll Island, Ga. Academic Press, N.Y.

Capone, D. G., P. Penhale, R. S. Oremland and B. F. Taylor. 1979. Relationship between productivity and ^(Cjl^) fixation in a Thalassia testuainum community. Limnol. Oceanogr. 21:117-125.

Carpenter, S. R. 1980. Estimating net shoot production by a hierarchical cohort method of herbaceous plants subject to high mortality. Am. Mid. Nat., 104:163-175.

Clark, S. H. 1970. Factors affecting the distribution of fishes in Whitewater Bay, Everglades National Park, Florida. Ph.D. Dissertation. Univ. of Miami, Coral Gables, Fla. 100 pp.

Congdon, R. A. and A. J. McComb. 1979. Production of Ruppia : seasonal changes and dependence on light in an Australian estuary. Aquat. Bot. 6:121-132.

Conover, J. T. 1968. Importance of natural diffusion gradients and transport of substances related to benthic marine plant metabolism. Bot. Mar. XI(1-4):1-9.

Darnell, R. M. 1961. Trophic spectrum of an estuarine community based on studes of Lake Pontchartrain, La. Ecology, 42(3):553-568. den Hartog, C. 1967. The structural aspect in the ecology of sea-grass communities. Helg. Wissen. Meersunter. 15:648-659. den Hartog, C. 1970. The Sea-grasses of the World. North-Holland, Amersterdam. 275 pp. den Hartog, C. 1977, Structure, function and classification in seagrass communities, pp. 89-121. £n sC. P. MeRoy and C. Helfferich (Eds.)., Seagrass Ecosystems. A Scientific Perspective. Marcel Dekker. N.Y.

Fenchel, T. 1969. The ecology of marine microbenthos. IV. Structure and function of the benthic ecosystem, its chemical and physical structure and the mcirofauna community with special reference to the ciliated protozoa. Ophelia, 6:1-182.

Fenchel, T. 1970. Studies on the decomposition of organic detritus derived from the turtle grass Thalassia testudinum . Limnol. Oceanogr. 15:14-20.

Fenchel, T. 1972. Aspects of decomposer food chains in marine benthos. Verh. Deutch. Zool. Gesell. 65. Jahresversamml. 14:14-22.

Fenchel, T. 1977. Aspects of the decomposition of seagrasses. pp. 123-146. Iji : C. P. McRoy and C. Helfferich, (Eds.), Seagrass Ecosystems ,A Scientific Perspective . Vol..4. Marcel. Dekker, Inc. N.Y.

Fitzgerald, G. P. 1968. Detection of limiting or surplus nitrogen in algae and aquatic weeds. J. Phycol. 4:121-126.

Fitzgerald, G. P. 1969. Field and laboratory evaluations of bioassays for nitrogen and phosphorus with algae and aquatic weeds. J. Phycol. 5:206-212.

Gentner, S. R. 1977. Uptake and transport of iron and phosphate by Vallisneria spiralis L. Aquat. Bot. 3:267-272.

Gerloff, G. C. and P. H. Krombholz. 1966. Tissue analysis as a measure of nutrient availability for the growth of angiosperm aquatic plants. Limnol. Oceanogr. 11:529-537.

Gessner, F. 1971. The water economy of the seagrass Thalassia testudinum .Mar. Biol. 19(3):258-260.

Ginsburg, R. N. and Ii. A. Lowenstam. 1958. The influence of marine bottom communities on the depositional environment of sediments. Jour. Geology. 66s310-318.

Goering, J. J. and P. L. Parker. 1972. Nitrogen fixation by epiphytes on seagrasses. Limnol. Oceanogr. 17:320-323.

Hammer, L. 1968. Salzgehalt und Photosynthese bei marinen Pflanzen. Mar Biol. 1(3):185-190.

Hargrave, B. T. 1970a. The effect of deposit feeding amphipods on the metabolism of benthic microflora. Limnol. Oceanogr. 15:21-30.

Hargrave, B. T. 1970b. The utilization of benthic microflora by Hyallela azteca (Amphipoda). J. Animal Ecol. 39:427-437.

Hargrave, B. T. 1972. Aerobic decomposition of sediment and detritus as a function of particle surface area and organic content. Limnol. Oceanogr. 17: 583-596.

Hay, W. W., F. Wiedenmeyer, and D. S. Marszalek. 1970 Modern organism communities of Bimini Lagoon and their relation to the sediments, pp. 19-25. IN Supko, S. D., Marszalek, and W. Bock, (Eds.), Sedimentary Environments and Carbonate Rocks of Bimini, Bahamas. Miami Geol. Soc. Miami, Fla.

Hoese, H. D. and R. S. Jones. 1963. Seasonality of larger animals in a Texas turtle grass community Publ. Inst. Mar. Sci. Univ. of Texas. 9:37-47.

Ikusima, I. 1965. Ecological studies on the productivity of aquatic plant communities I. Measurement of photosynthetic activity. Bot. Mag Tokyo. 78: 202-211.

Ikusima, I. 1966. Ecological studies on the productivity of aquatic plant communities II. Seasonal changes in standing crop and productivity of a natural submerged community of Vallisneria denseserrulata . Bot. Macj. Tokvo. 79:7-19.

Jones, J. A. 1968. Primary productivity of the tropical marine turtle grass, Thalassia testudinum , Koenig, and its epiphytes. Ph.D. Dissertation. Univ. of Miami, Coral Gables, Fla. 196 pp. 48

Kemp, W. M. 1933. Seagrass communities as a coastal resource: a preface. Mar. Tech. Soc. Jour. 17(2):3-5.

Kemp, W. M., R. R. Twilley, J. C. Stevenson, W. R. Boynton, and J. C. Means. 1983. The decline of submerged vascular plants in upper Chesapeake Bay: summary of results concerning possible causes. Mar. Tech. Soc. Jour. 17(2):78-89.

Kemp, W. M., W. R. Boynton, and R. R. Twilley. 1984. Influences of submersed vascular plants on ecological processes in upper Chesapeake Bay. pp. 367-394, Ijn :V. S. Kennedy (Ed.), The Estuary as a Filter . Academic Press, Inc., N.Y.

Kikuchi, T. 1974. Japanese contributions on consumer ecology in eelgrass ( Zostera marina L.) beds, with special reference to trophic relationships and resources in inshore fisheries. Aquaculture, 4:145-160.

Kristensen, J. H. 1972. Carbohydrates of some marine invertebrates with notes on their food and on the •natural occurrence 6f the carbohydrates studies.Mar. Biol. 14:130-142.

Lipschultz, F. J., J. Cunningham and J. C. Stevenson. 1979. Nitrogen fixation associated with four species of submerged angiosperms in the central Chesapeake Bay. Est. and Coastal Mar. Sci. 9:813-818.

Livingson, R. J., R. S. Lloyd a.nd M. S. Zimmerman. 1976. Determination of sampling strategy for benthic macrophytes in polluted and unpolluted coastal areas. Bull. Mar. Sci. 26(4):569-575.

Mann, K. H. 1972. Macrophyte production and detritus food chains in coastal waters. Mem. 1st. Ital. Idrobiol. 29(Suppl.):353-383.

Mann, K. H. 1982. Ecology of Coastal Water . A Systems Approach . Univ. of Cal. Press. Berkeley, Ca. 312 pp.

McMillan, C. and F. N. Moseley. 1967. Saline tolerance of five marine spermatophytes of Redfish Bay, Texas. Ecology, 48:503-506.

McRoy, C. P. and C. Helfferich. 1977. Seagrass 49

Ecosystems, A Scientific Perspective . Vol. 4. Marcel Dekker, N.Y. 314 pp.

McRoy, C. P. and C. McMillan. 1977. Production ecology and physiology of seagrasses. pp. 53-83. In : P. C. McRoy and C. Helfferich (Eds.). Seagrass Ecosystems, A Scientific Perspective . Vol. 4. Marcel Dekker, N.Y.

McRoy, C. P., J. J. Goering and B. Chaney. 1973. Nitrogen fixation associated with seagrasses. Limnol. Oceanogr. 18:998-1002.

McRoy, C. P., R. J. Barsdate, and M. Nebert. 197.2. Phosphorus cycling in an eelgrass ( Zostera marina L.) ecosystem. Limnol. Oceanogr. 17:58-67.

McRoy, C. P. and V. Alexander. 1975. Nitrogen kinetics in aquatic plants in arctic Alaska. Aquat. Bot. 1:3-10.

Moore, D. R. 1963. Distribution of the seagrass, Thalassia ,in the United States. Bull. Mar. Sci. Gulf Carib. 13(2 ):329-342.

Nielsen, B. 0. 1969. Carbyhydrases of some wrack invertebrates. Natura Jutlandica. 12:191-194.

Odum, H. T. 1956. Primary productivity in flowing waters. Limnol. Oceanogr. 1:102-117.

Odum, H. T. 1960. Analysis of diurnal oxygen curves for the assay of reaeration rates and metabolism in polluted marine bays. pp. 547-555. In : Pearson, E.A. (Eds.), Proceedings of the First Intl. Conf. on Waste Disposal in the Marine Environment.

Odum, H. T. 1963. Productivity measurements in Texas turtle-grass beds and the effects of dredging an introcoastal channel. Inst. Mar. Sci. Pub. Univ. Texas. 9:48-58.

Odum, H. T. 1967. Biological circuits and marine systems of Texas, pp. 99- 157. In : T. A. Olson and F.J. Burgess (Eds.) Pollution and Marine Ecology. Proc. Conf. on Status of Knowledge . . . Relating to Ecology and Pollution Problems in the Marine Environment. Interscience Pub., N.Y.

Odum, II. T. and B. J. Copeland. 1974. A functional 50

classification of coastal ecological systems, pp. 5-84. In s Odum, H. T., B. J. Copeland and E. A.McMahan (Eds.), Coastal Ecological Systems of the United States . The Cons. Foundation, Wash., D.C.

Odum, li. T. and C. M. Hoskins. 1958. Comparative studies on the metabolism of marine waters. 29 pp.

Odum, W. E. and E. J. Heald. 1972. The importance of vascular plant detritus to estuaries, pp. 91-114. In : Coastal Marsh and Estuary Management Symposium.

O'Gower, A. K. and J. W. Wacasey. 1967. Animal communities associated with Thalassia , Diplanthera , and sand beds in Biscayne Bay. I. Analysis of communities in relation to water movements. Bull. Mar. Sci. 17(1):175-210.

Orth, R. J. 1977. The importance of sediment stability in seagrass communities, pp. 281-300. In : B. C. Coiul (Ed.), Ecology of Marine Benthos . Univ. S. Carolina Press, Columbia, S. C.

Patriquin, D. G. 1972. The origin of nitrogen and phosphorus for growth of the marine angiosperm Thalassia testudinum . Mar. Biol. 15:35-46.

Patriquin, D. G. 1973. Estimation of growth rate, production, and age of the marine angiosperm, Thalassia testudinum , Konig. Carib. Jour. Sci. 13:111-123.

Patriquin, D. G. and R. Knowles. 1972. Nitrogen fixation in the rhizosphere of marine angiosperms. Mar. Biol. 16:49-58.

Penhale, P. A. 1977. Macrophyte-epiphyte biomass and productivity in an eelgrass ( Zostera marina L.) community. J. Exp. Mar. Biol. Ecol. 26:211-224.

Peres, J. M. and J. Picard. 1975. Causes de la Rarefaction et de la Disparition des Herbiers de Posidonia oceanica sur les Cotes Francaises de la Mediterranee. Aquat. Bot. 1:133-139.

Petersen, C. G. J. 1918. The sea bottom and its production of fish food. A survey of the work done in connection with the valuation of the 51

Danish waters from 1883-1917. Rept. Danish Biol. Sta. 25:1-82.

Phillips, G. 1., D. Eminson and B. Moss. 1978. A mechanism to account for macrophyte decline in progressively eutrophicated freshwaters. Aquat.Bot. 4:103-126.

Phillips, R. C. 1960. Observations on the ecology and distribution of the Florida seagrasses. Fla. St. Bd. of Cons. Prof. Papers Series #2. 72 pp.

Phillips, R. C. 1974. Transplantation of seagrasses, with special emphasis on eelgrass, Zostera marina L. Aquaculture, 4:161-176.

Phillips, R. C. and C. P. McRoy. 1980. Handbook of Seagrass Biology. An Ecosystem Perspective . Garland STPM Press, N.Y. 353 pp.

Qasim, S. Z. and P. M. A. Bhattathiri. 1971. Primary production of a seagrass bed on Kavaratti Atoll (Laccadives). Hydrobiologia. 38:29-38.

Randall, J. E. 1965. Grazing effect on seagrasses by herbivorous reef fishes in the West Indies. Ecology. 46(3):255-260.

Robertson, M. L., A. L. Mills and J. C. Zieman. 1982. Microbial synthesis of detritus like particles from dissolved organic carbon released by tropical seagrasses. Mar. Ecol. Prog. Ser. 7:279-285.

Sand-Jensen, K. 1975. Biomass, net production and growth dynamics in an eelgrass ( Zostera marina L.) population in Vellerup Vig, Denmark. Ophelia, 14:185-201.

Sand-Jensen, K. 1977. Effect of epiphytes on eelgrass photosynthesis. Aquat. Bot. 3:55-63.

Sand-Jensen, K. and J. Borum. 1982. Regulation of growth of eelgrass ( Zostera marina L.) in Danish coastal waters. Marine Tech. Soc. Jour. 17(2):15-21.

Schroeder, P. B. 1975. Thermal stress in Thalassia testudinum . Diss. U. of Miami, Coral Gables. 157 pp.

Scoffin, T. P. 1970. The trapping and binding of 52

subtidal carbonate sediments by marine vegetation in Bimini Lagoon, Bahamas. Jour. Sed. Pet. 40(1)*249-273.

Short, F. T. 1980. A simulation model of the seagrass production system, pp.277-295. In * R. C. Phillips and C. P. McRoy. (Eds.TT Handbook of Seagrass Biology . An Ecosystem Perspective . Garland STPM Press, N.Y.

Sondergaard, M. and J. Sand-Jensen. 1979. Carbon uptake by leaves and roots of Littorella uniflora (L.) Aschers. Aquat. Bot. 6:1-12.

Steller, D. L. 1981. Production of a submerged grass bed related to depth, nitrogen, litterfall and decomposition. American Institute of Biological Sciences-Ecological Society of America Joint Meetings, Bloomington, Ind., August, 1981.

Strawn, K. 1961. Factors influencing the zonation of submerged monocotyledons at Cedar Key, Florida. Jour. Wildlife Mgmt. 25:178-189.

Teas, H. 1976. Productivity of Biscayne Bay mangroves, pp. 103-112. In : Thorhaug, A. (Ed.), Biscayne Bay: Past/ Present/ Future . U. of Miami Seagrant and ERDA, Miami, Florida.

Thayer, G. W., D. A. Wolfe and R. B. Williams. 1975a. The impact of man on seagrass systems. Am. Sci. 288-296.

Thayer, G. W., S. M. Adams, and M. W. LaCroix. 1975b. Structural and functional aspects of a recently established Zostera marina community. Proc. Sec. Intern. Estuarine Res. Conf., Myrtle Beach, S. C. Oct. 1973.

Thorhaug, A. 1976. Response of Thalassia testudinum grassbeds to thermal discharges from the Turkey Point Power Plant, pp. 22-34. IN * (A. Thorhaug, Ed.), Biscayne Bay : Past/ Present/ and Future . University of Miami Seagrant and ERDA, Miami, Florida.

Titus, J. E. and M. S. Adams. 1979. Coexistence and the comparative light relations of the submersed macrophytes Myriophyllum spicatum L. and Vallisneria americana Michx. Oecologia (Berl). 40*273-286. 53

Titus, J. E., R. A. Goldstein, M. S. Adams, J. B. Mankin, R. V. O ’Neill, P. R. Weiler, Jr., H. H. Shugart, and R. S. Booth. 1975. A production model of Myriophullum spicatum L. Ecology, 56:1129-1138.

Turner, R. E., R. M. Darnell and J. Bond. 1980. Changes in the submerged macrophytes of Lake Pontchartrain (Louisiana):1954-1973. N. E. Gulf Sci. 4(l):44-49.

Westlake, D. F. 1963. Comparisons of plant productivity. Biol. Rev., 38:385-425.

Westlake, D. F. 1967. Some effects of low velocity currents on the metabolism of aquatic macrophytes. Jour. Exp. Bot. 18(55):187-205.

Wetzel, R. L. 1985. Simulation model analyses of Zostera marina L. photosynthesis and growth: Effects of selected empirically-derived and hypothetical physical-chemical and biological interactions. Presented at the International Symposium on Aquatic Macrophytes, August, 1985, Silkeborg, Denmark.

Wiginton, J. R. and C. McMillan. 1979. Chlorophyll composition under controlled light conditions as related to the distribution of seagrasses in Texas and the U. S. Virgin Islands. Aquat. Bot. 6:171-184.

Wood, E. J. F. 1965. Marine Microbial Ecology . Chapman and Hall Ltd. London. 243 pp.

Wood, E. J. F., W. E. Odum, and J. C. Zieman. 1969. Influence of seagrasses on the productivity of coastal lagoons, pp. 495-502. In : Lagunas Costeras,un Simposio. Mem. Simp. Intern. Lagunas Costeras. UNAM-UNESCO. Mexico, D. F.

Zieman, J. C. 1968. A study of the growth and decomposition of the sea-grass Thalassia testudinum . M. S. Thesis. Univ. of Miami, Coral Gables, Fla.50 pp.

Zieman, J. C. 1972. Origins of circular beds of Thalassia (Spermatophyta:Hydrocharitaceae) in South Biscayne Bay, Florida, and their relationship to mangrove hammocks. Bull. Mar. Sci. 22(3):559-573. 54

Zieman* J. C. 1975. Seasonal variation of turtle grass* Thalassia testudinum Konig* with reference to temperature and salinity effects. Aquat. Bot. 1:107-123.

Zieman* J. C. and R. G. Wetzel. 1930. Productivity in seagrasses:methods and rates, pp. 87-116. In : R. C. Phillips and C. P. McRoy (Eds.). Handbook of Seagrass Biology: An Ecosystem Perspective. N.Y.

Zimmerman, M. S. and R. J. Livingston. 1976a. Effects of Kraft-mill effluents on benthic macrophyte assemblages in a shallow-bay system (Apalachee Bay, N. Fla, U.S.A.). Mar. Biol. 34:297-312.

Zimmerman* M. S. and R. J. Livingston. 1976b. Seasonality and physio-chemical ranges of benthic macrophytes from a north Florida estuary (Apalachee Bay). Contributions in Mar. Sci. 20:33-44. CHAPTER 3

PRODUCTION OF A SUBMERGED GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA, INCLUDING LITTERFALL AND DECOMPOSITION USING A HARVEST TECHNIQUE

ABSTRACT Production of a submerged Vallisneria americana bed in Lake Pontchartrain, Louisiana, was studied from

May, 1978, through May, 1979. Significant variations in biomass within the grass bed were attributed to distance from shore, time of year, and area along the shoreline. Production ranged from near zero to 282 g 2 dry weight/m /yr. The highest values occurred in the center of the grass beds.

The harvest technique and production calculations were adapted to the unique submerged grass bed environment. Monthly measurements of above and below ground biomass, litterfall and decomposition were used to estimate production. Positive increases in biomass were the base net production. Leaf litter that fell between harvest sampling trips was collected in littertraps and added to positive increases in biomass measured during the sampling period to give a harvest plus litterfall estimate of production.

Litterfall occurred throughout the year and was directly related to standing stock, nitrogen treatment and distance from shore. Rapid decomposition, particularly during the first two weeks after leaf

55 56 abscission, meant that some leaf litter decomposed while in the litter traps. Therefore, complete production estimates included harvested increases in biomass plus true litterfall corrected for losses due to litter decomposition. Decomposition rates were estimated using decomposition bags and were high in comparison to marine submerged grass beds and! emergent marshes. Virtually total decomposition of plant material occurred within seven months for experiments begun in spring and fall. Both litterfall and decomposition increased toward shore while biomass and production peaked in the center of the grass bed.

3.0 INTRODUCTION

Vallisneria americana Mich, is the dominant submerged estuarine angiosperm of Lake Pontchartrain,

La. Like Thalassia testudinum beds of more saline environments (Ginsburg and Lowenstam, 1958; Carr and

Adams, 1973; and Burrell and Schubel,1977),

Vallisneria beds stabilize shifting bottom sediments and create a habitat for a larye number of species.

More specifically, the submerged grass beds of Lake

Pontchartrain are a major nursery area for Callinectes sapidus (blue crabs), Cynoscion nebulosus (spotted sea trout) and Penaeus aztecus (penaeid shrimp) which 57 together account for a large percentage of the commercial fish catch in the lake (Thompson and

Verret, 1980). Submerged grasses decompose rapidly

(Zieman, 1968) and foster a detrital food chain

(Fenchel, 1970). Rapid decomposition and turnover rates make submerged grass beds an important site of nutrient and pollutant cycling.

Productivity of submerged grass beds is among the highest of any plant community (Zieman and Wetzel#

1980). However, geographic and sampling technique differences contribute to a high variability of productivity estimates. Productivity of grass beds 2 has been reported to range from 1 to 16 gC/m /day and standing crops to range from 100 to 5,000 gm dry wt/m (MeRoy and McMillan, 1977). Productivity estimates using oxygen techniques are among the highest estimates (Odum, 1957) but are inaccurate because they are biased by the leakage of oxygen bubbles, vary greatly even in small areas, and are hard to extrapolate over time. Carbon estimates using 14 C may underestimate production due to internal carbon cycling. Measuring the plant leaf growth by harvest or marking ignores underground production.

The main objective of this study was to measure the productivity of a submerged grass bed in Lake

Pontchartrain. To accomplish this it was necessary to 58

address the above problems by a) setting up a sampling

design which would yield accurate estimates; and b)

developing field techniques and production

calculations adapted to the submerged plant

environment.

3.1 TrIE STUDY AREA

Vallisneria americana occurs along sandy

shorelines out to 150m from shore on the northern

shore of Lake Pontchartrain at about 30°30'N

latitude, 90°W longitude (Figure 1). Zonation

across the bed showed Eleocharis parvula near shore,

then mixed Vallisneria and Ruppia maritima , and

finally only Vallisneria farthest from shore, sparsely

scattered among Rangia cuneata shells. Salinities

ranged from near 0 ppt to about 4 ppt over the year with considerable freshwater input occurring in the

spring, especially during openings of the Bonnet Carre

Spillway. Water temperatures were closely related to

air temperatures and ranged from 9°C to 34°C

annually. Water depths ranged spatially from 0m to

2m.

3.2 METHODS

Two study sites were chosen along the north shore of Lake Pontchartrain (Figure 1); a control site and a

nitrogen enrichment site. Both sites had sandy dunes on shore, undulating shorelines, and tidal creeks 59 which drained brackish marshes and whose mouths were about 20m from the edge of the sites.

Preliminary sampling, which started in December,

1977, revealed that the grasses had a patchy distribution. After taking 20 random grass samples from many depths, a "t" test was done to determine the minimum number of samples necessary to attain an 80% level of certainty in the mean harvested biomass.

This level of certainty was achieved with duplicate samples harvested at four distances from shore, and near or far from the tidal creeks (see Figure 1), to give a total of 16 samples at each of the two sites.

Two transects 50m apart and perpendicular to the shoreline were marked with permanent stakes at the four distances from shore (25, 50, 75, and 100m from shore) at the two treatment areas. Two samples were taken as replicates at each stake.

One hundred-thirty seven kilograms of ammonium nitrate were added to the nitrogen treatment area monthly beginning in May, 1978, and continuing through

September, 1979. Half of the NH^NO^ pellets

(pebble sized granules) were broadcast directly into the water over the area between the markers (quick introduction) and the remaining half of the pellets were scattered along the shore at the high tide mark to be washed out and distributed by the tides (slow 60 introduction). The latter introduction resulted in total distribution of the nitrogen in one to three weeks. In April, 1979, the Bonnet Carre Spillway was opened for five weeks and water high in nutrients

(Army Corps of Engineers, 1979) covered the entire study area. This disturbance raised water levels and prompted an algae bloom. After the closing of the

Spillway, water nutrient levels returned to normal and the algae bloom disappeared.

Sampling started in May, 1978. Monthly measurements included 32 grass harvest samples as well as physical and chemical measurements (see Figure 1).

The grass was collected along the two transects perpendicular to the shore which marked the boundary of each site. Dissolved oxygen, temperature, water depth, secchi depth, currents, and water chemistry were measured along each boundary halfway between the corner markers and directly in the center of the site.

This diamond shape allowed a spatial component of any parameter to be seen. For example, a flow of nutrients from a bordering creek would show high values along the creek side diminishing as the flow crossed the study area.

At the beginning of each field trip Kuus balls

(Doty, 1971) were placed in the diamond configuration to measure circulation (relative water flow), and 61

secchi depth# temperature, and oxygen were measured at

the same location every 3-4 hours for 24 hours. At

the end of the 24 hours# water samples were taken at

the site of the Muus balls and the Muus balls

collected, dried and weighed. Twelve detritus bags

were anchored in the spring and fall at 25m, 50m, and

75m from shore along a transect down the middle of

each treatment area. Sediment cores were taken

quarterly to a depth of approximately 25cm along the

same transect at all four distances from shore.

Nitrogen fixation measured by acetylene reduction

(Kasselman, 1980) was also measured quarterly in bare

sediment, rhizosphere sediment including plant parts,

and leaves. Litter traps were set out next to

permanent markers at 25m, 50m and 75m from shore.

3.2.1 Harvest Methods

A metal cylinder 30cm in diameter was forced

about 25cm into the sediment and all material inside

was collected with a posthole digger. It was sieved

through a 2mm mesh screen on the site to remove

sediment. The retained material was stored in a

ziploc bag on ice. Samples were then sorted into

species. Four major genera were distinguished: 1)

Vallisneria , 2) Ruppia , 3) Najas , and 4) Eleocharis

; and all else was lumped as "other." To further

categorize the Vallisneria parts, leaves, roots, 62

rhizomes, and reproductive material were separated.

All plant samples were then labeled, dried at

100°C and weighed. Biomass was harvested monthly.

3.2.2 Litter

Litter was collected in traps consisting of a

cubic metal frame 25cm on a side and covered with quarter inch hardware cloth and aluminum screen and anchored into the sediment (Figure 2). The anchors were sunk into the sediment far enough (35-50cm) to bury the bottom of the aluminum screen. Traps were placed over existing vegetation at 25m, 50m and 75m from shore adjacent to permanent harvest stakes in the control and nitrogen addition treatment areas. A door at the top of the trap allowed entry. Initially litter traps were cleared by hand of floating, unattached leaves, the sediment surface was sifted for fallen dead leaves and any leaves caught on the sides of the screen were also taken out. At each subsequent sampling, dead material was again cleared from the traps and retained. When time and weather conditions permitted, traps were cleared twice a month, once concurrent with the live biomass harvest. Any attached dead leaves were taken, but typically none were observed except in February, 1979. Litter was dried at 100°C and weighed.

When leaves fell off the Vallisneria plants some 63 decomposition occurred before they were removed from the litter traps. The amount of decomposition is related to the length of time before removal (usually two weeks) and the appropriate decomposition rate.

Since the leaf could have fallen any time during the two week interval, the average decomposition time was assumed to be one half the sampling interval, on the assumption that there was a uniform litterfall rate over the time period. The calculation for true litter

(TL) was equal to the observed litterfall (OL) plus decomposition loss (loss).

True Litterfall (TL) = Observed Litterfall (OL) + loss

TL = OL + loss

loss = TL x Decomposition Rate (D) x 1/2 of the time

or

TL = OL + (TLxDxl/2xtime)

by the distributive property

TLx(l-Dxl/2xtime) = OL

or

TL = 0L/(1 - Dxl/2xtime)

3.2.3 Decomposition

Decomposition was measured in 2mm mesh nylon bags filled with 120 gm wet weight of Vallisneria leaves and anchored on the bottom at 25m, 50m, and 75m from shore at the treatment and control sites. Initial sets of bags were immediately dried and weighed. Bags 64 were set out in October (fall-winter) and in June

(summer). Three bags per site were collected after two weeks and then at monthly intervals until all contents were gone. Animals were separated from the plant remains, identified and counted. Plant remains were dried at 100°C and weighed.

The decomposition rate in percent dry weight loss per day was used to estimate decomposition losses from the leaf litter in the littertraps. Decomposition rates used in the production calculations were computed using the average decomposition over the first two week time period (the littertrap time interval) for a given distance from shore. Using the littertrap time interval gave the decomposition rate of litter collected in the traps and added to harvest estimates of production.

3.2.4 Harvest Techniques and Calculations

Biomass, litterfall, and decomposition were the field measurements that were used to calculate production estimates. Production estimates were calculated separately for the two study sites to evaluate differences due to nitrogen enrichment. Four different estimates of production were calculated from the collected data.

3.2.4.1 Peak Standing Crop

The peak standing crop technique simply involved 65 recording the highest biomass at each of the sixteen stations for the study year. The net annual primary productivity (NAPP) was equal to this maximum biomass:

NAPPp = MAXIMUM BIOMASS

3.2.4.2 Harvest

Both above and below ground biomass were collected throughout the year to yield, production estimates by the harvest technique. In this technique plots were sampled once a month and positive changes in live material from one month to the next (T2-T1) were summed to give the annual NAPP. In this study both above and below ground material were measured so that:

NAPPn = 2(+A3I0MASS).

3.2.4.3 Harvest Plus Litterfall

Since in the harvest technique leaves might fall between sampling periods and would not be measured, litterfall during each sampling interim (mortality) was added to the harvest technique results to get a more accurate assessment of production. The method involved summing live biomass changes from the harvest technique plus the accumulation of dead litter in the litter traps. Calculations of the NAPP followed those of Smalley (1959) in that: l)if live biomass and litterfall were both positive, they were summed; 2)if either measured live biomass or litterfall were 66

negative, the algebraic sum of them was taken, and if

positive, added to annual NAPP? 3)negative changes

were ignored in the calculation.

This harvest plus litterfall technique assumes

that the site is homogeneous, that removing the dead

leaves and litter from a screened plot has no effect

on its production, and that no material escapes from

the screened plot via currents or decomposition. The

technique was tailored to the submerged grass bed by

collecting both above and below ground material and

constructing an underwater litter trap. There was

usually no standing dead material in this

ecosystem— i.e., once a leaf died it fell off of the

plant. The total annual NAPP was:

NAPP = 2(+ABI0MASS + MORTALITY)/time and since: MORTALITY = LITTER, therefore: NAPPL = 2(+ABI0MASS(T2 -Tl) + LITTER(T2 -T1))

3.2.4.4 Harvest Plus Litterfall Plus Decomposition

Leaves that fall may rapidly decompose before they are collected. Therefore the harvest plus

litterfall technique also underestimates true NAPP.

The litter-bag decomposition rates were used to correct the harvest plus litterfall technique. In addition to measuring litterfall, summer and winter rates of decomposition were also calculated. NAPP was calculated as:

NAPP = (+ BIOMASS + MORTALITY)/time MORTALITY = LITTER plus DECOMPOSITION 67

DECOMPOSITION =1/2 LITTER x (litter loss rate) x time 1/2 LITTER was an estimate of average litter biomass during the interval of decomposition, litter loss rate = % weight loss from T2-T1 NAPPd = 2(+ABI0MftSS(T2-Tl) + LITTER(T2-T1) + DEC0MP0SITI0N(T2-T1))

3.3 RESULTS

Production ranged from 81 to 280 gm dry 2 wt/m /yr in the control area and from 24 to 282 gm

dry wt/m /yr in the nitrogen addition area, as a

function of distance from shore and method of

calculation. Components of production - biomass,

litterfall and decomposition - are influenced by

environment, area along the shoreline, time of year,

and distance from shore. Given the large range of

concurrent production estimates from the same grass

bed, it is appropriate to look at each of these

environmental factors as they contribute to biomass,

litterfall, decomposition, and, thus, production.

3.3.1 Biomass

Vallisneria biomass generally started low in the

spring, rose to a peak during the early winter, fell

to a minimum in April, and rose markedly the second

spring (Figure 3). Environmental factors which might have affected this annual cycle such as sunlight,

temperature, salinity, water depth, and current are

shown in Figure 4. September is the first month when

temperature dropped significantly. The greatest 68

standing stock was present in December and January but

it dropped markedly toward the end of January,

probably due to a combination of short days (low

sunlight) and low temperatures. Biomass was at an

annual low in the late winter and early spring. In

April salinity decreased and sunlight, temperature,

and currents increased, encouraging new growth.

There were different patterns for the individual

parts of the plant. The roots reached their maximum

biomass in October. The fastest growth period for, the

leaves and rhizomes occurred at the root biomass

maximum. The minimum root biomass occurred in

February during the lowest temperatures, after which

the roots slowly began to recover. Rhizome biomass

increased steadily from May to late December with the

exception of late August. The proportion of biomass

in the rhizomes increased toward the end of the growing season, perhaps through reallocation of energy reserves from the top. Leaf biomass increased steadily through September when a burst of leaf growth occurred. In the first year leaf biomass was proportionally greatest in early October. Flowers first appeared in June, 1978, and fruit in September. .

Fruit remained attached until January. Flowers appeared earlier in 1979, a warmer year.

Average biomass values for a given site and depth 2 ranged from 0 to 199 gm dry wt/m . Maximum

individual plot biomass values for the major species 2 were: Vallisneria , 442.5g/m ; Ruppia , 2 2 13.6g/m ; Najas , 3.6g/m ; and Eleocharis , 2 0.2g/m . These figures are low when compared to coastal marsh vegetation like the nearby Goose Point marsh which ranged from 1000 to 2000 gm live above-ground dry wt/m (Cramer and Day, 1981).

The grass bed figures are probably particularly low due to the extremely harsh winter in the 1977-1978 season, as illustrated by the May, 1978, biomass of

14.8 + 5g/m^ compared with the May, 1979, biomass of 30.19 + 20 g/m^.

Analysis of variance using all biomass values including zeros showed that time of year, area along the shore, distance from shore and area along the shore*distance interactions were highly significant (p less than 0.001) contributors to variations in biomass

(Table 1). Time was the split plot variable, but because time was significant when analyzed in a completely randomized design, it was reasonable to consider time as completely random. We have already seen how biomass varied throughout the year and it is not surprising that the variation is highly significant.

3.3.1.1 Area Along the Shore Effects 70

The biomass at the nitrogen enriched area 2 (mean=32.1 g dry wt/m ) was significantly (p less

than 0.01) lower than at the control area (mean=55.9 g

dry wt/m ). This was particularly true during

spring, summer and fall. However, water column

available inorganic nitrogen and sediment available

inorganic nitrogen were always higher at the nitrogen

enrichment area (Figure 5). The lower biomass at the

nitrogen enrichment area could have been due to

intrinsic differences between the two areas or due to a negative effect of the nitrogen fertilization. An

increase in leaf blade epiphyte biomass was noted in the nitrogen enrichment area, particularly near shore where the introduction technique caused the inorganic nitrogen levels to be highest. Moderate 'to high nutrient loadings have been shown (Kemp, et al. ,

1983) to result in significant increases in growth of epiphytic and planktonic algae associated with submerged aquatic vegetation. Increased epiphytic algae have resulted in decreases in productivity of submerged macrophytic vegetation (Kemp, jit al. ,

1984).

3.3.1.2 Distance from Shore Effects

There were highly significant (p less than 0.01) differences in biomass means due to distance from shore (Table 1). The highest biomass occurred at 50m 71 and 75m from shore (center of the grass bed) and the least biomass occurred at 100m from shore.

3.3.1.3 Area Times Distance Effects

There was a highly significant (p less than 0.01) difference in the biomass means due to the interaction of area along the shore and distance from shore.

Sediment nitrogen values changed with distance from shore due to the introduction technique which concentrated nitrogen near shore (Figure 5). Any biotic differences due to enrichment, such as increased phytoplankton or ephiphytism, should also have changed with distance from shore. Therefore, it is not surprising that these two environmental factors showed interactive effects on biomass. Both the control and nitrogen enrichment areas showed highest biomass in the center of the grass beds. However, the nitrogen enrichment biomass curve was more compressed toward shore— i.e., the maximum biomass occurred at

50m (173 g dry wt/m ). The control grass bed actually extended farther from shore (150m), and its maximum biomass occurred at 75m from shore and was greater than at the enrichment area (199 g dry w t / m ^ ).

3.3.2 Litter

Rates of litterfall ranged from 0.004 to 0.2 g 2 dry wt/m /day. Since there were times when it was 72 impossible to collect litter due to poor weather conditions, broken traps, or vandalism, it was necessary to develop a relationship between litter and other more easily measured parameters. An analysis of variance revealed that litterfall rates were positively related to treatment and negatively related to distance from shore (p less than 0.05). At each treatment and distance from shore litter was found to be significantly correlated with average above ground standing stock. Since the biomass data were significantly affected by treatment and distance from shore it is not surprising that the litterfall was also affected by these factors. Litterfall was greater in the nitrogen treatment area and decreased away from shore. The regression equations for litterfall (L) versus standing stock (SS) were:

2 9 Control Area Nitrogen Treatment Area R

25m L=.00812*SS-.039 0.76 L=.00857*SS-.053 0.71

50m L=.00045*SS-.02Q 0.92 L=.00051*SS-.022 0.94

75m insufficient data

The negative term shows that there was some minimum standing stock present before litterfall begins or that the litter traps missed low levels of litterfall. The dramatic loss of live biomass that 73 occurred in January and February was not reflected in litterfall and was not adequately accounted for in the litterfall regression equations. The regression equations only account for litterfall (normal loss of leaves) correlated with standing stock during growth.

Shoots normally maintain a fairly constant number of leaves, with a leaf falling whenever a new one appears. The very rapid loss of leaves (and below ground biomass) in the winter was probably due to low temperatures and winter storms which ripped up the plants or buried them in sediment and caused them to die. For example, sediment cores showed 10cm losses or gains of sediment. This "litterfall" component was impossible to quantify. Since no positive growth was occurring during winter, the litterfall values calculated in the regression equations did not underestimate the litterfall contribution to net production (NAPP^ and NAPP^).

3.3.3 Decomposition

Compared to saline grass beds and the Goose Point salt marsh, decomposition of the Vallisneria americana leaves in litter bags was rapid and complete, percent leaves remaining actually went to zero (Figure 7a and 7b). Summer decomposition was approximately twice as fast as winter decomposition.

Decomposition rates ranged from 8-47 mg/g grass/day 74 loss. All of the biomass produced within a growing season could be decomposed before the next season.

Decomposition rates were higher at the nitrogen enrichment area than at the control area and increased toward shore.

At 25m from shore in the summer the control and nitrogen site rates (D) were similar after two weeks,

47+6 and 45+5 mg/g/day loss respectively. The nitrogen site decomposition bags were lost after this time and no further comparisons could be made.

However, in the winter, rates were faster at the nitrogen site (25 + 3 mg/g/day) than at the control site (16 + 4 mg/g/day). At 50m from shore decomposition rates at the control site increased from

11+2 mg/g/day in the first two weeks to 27 + 6 mg/g/day after two weeks in the summer and from 3 + 1 mg/g/day to 24 + 5 mg/g/day after three weeks in the winter, while the nitrogen area declined from 43+7 to 13 + 4 mg/g/day in the summer and stayed at 14 + 3 mg/g/day in the winter. After six weeks the percentages of plant material remaining at the control and nitrogen sites were similar. At 75m from shore a similar pattern occurred in the summer. The control site decomposition rate increased from 7 + 2 to 24 + 5 mg/g/day and the nitrogen site declined from 25+6 to

13+3 mg/g/day after three weeks. Decomposition bags 75 4 were washed away in the winter.

Initially, the nitrogen area generally had higher decomposition rates than the control area. However, after two to three weeks, the control area decomposition rates caught up with and surpassed those of the nitrogen area. This pattern occurred during both the summer and the winter. The decomposition rates for given seasons, treatments, and distances from shore were used in the production calculations for N A P P d .

3.3.4 Production

Production estimates ranged from 24 g dry 2 2 wt/m /year to 282 g dry wt/m (Table 2). The peak standing crop method (NAPPp ) gave the lowest estimate. The successive harvests of accumulated material (NAPP^) produced up to a 120% increase in calculated annual production compared with peak standing crop predictions. Adding litterfall estimates to the harvest technique (NAPP^) increased the estimate of annual production up to 28% over those of harvest alone. Litterfall was, particularly important near shore where the litterfall rates were highest. When observed litterfall was corrected for decomposition (NAPP q ), the harvest plus litterfall annual production estimates increased the simple harvest production estimates by up to 85%, 76

the greatest increase occurring near shore.

A weighted average production for the control and

the nitrogen addition areas was estimated by a

straight-line interpolation of areas under production

versus distance from shore curves at all distances

from shore including zero and 150m divided by the

-total distance from shore (Table 2). On a weighted average basis, litterfall plus harvest production

estimates (NAPPp) were only three to four percent higher than harvest estimates. Harvest plus

litterfall plus decomposition (NAPPp) gave the highest estimates of average grass bed production.

3.4 DISCUSSION

3.4.1 Biomass

Vallisneria americana in Lafce Pontchartrain appears to be an annual plant becausee a dramatic loss in both above ground and below ground biomass occurs in January and February. During the particularly harsh 1977-1978 winter, both above and below ground

Vallisneria biomass decreased to zero at the control and nitrogen addition areas. The 1973-1979 winter was mild and some plant biomass persisted throughout the year, particularly at mid-depths.

The Vallisneria grass bed annual pattern showed greater annual variation than marine grass beds from around the Gulf of Mexico. Marine grass beds in the 77

Gulf of Mexico are influenced by seasonal environmental conditions but are perennial— i.e., some plant biomass persists all year. In south Florida,

Zieman (1975) found that populations of Thalassia blades underwent seasonal fluctuations but always retained significant winter densities. Seagrass populations in north Florida also exhibited a seasonal fluctuation. Zimmerman and Livingston (1976 a and b) studied the seasonality and physico-chemical relationships of benthic plants in Apalachee Day and found that while Thalassia below ground biomass persisted all year, leaf densities decreased markedly below 15°. A similar pattern was found in Texas

(Conover, 1964). The loss of winter vascular plant biomass is often ofset by an increase in algae biomass in marine seagrass beds, but no influx of macro-algae occurred in the Vallisneria grass bed.

3.4.2 Litter

The litterfall rate for Vallisneria americana was correlated with above ground standing stock and showed seasonal variation. However, the rate of litterfall from January to February did not equal the decrease in harvested leaf biomass. The discrepancy probably resulted from a loss of litter which was not effectively recorded by the litter traps during erosion of the grass beds in winter storm fronts. 78

Closer monitoring of transport of plant material

during storm events needs to be done in order to

accurately quantify litterfall over the year.

3.4.3 Decomposition

Decomposition rates were initially higher at the

nitrogen addition area, but after a few weeks the

control area decomposition rates increased so that

both areas were similar. These results are examples

of what has been called the "priming effect" of added

inorganic nitrogen supplies on decomposition rates

(Smith, 1979). Microbe labile-nitrogen storage

determines the cell nitrogen available for microbial

growth. Addition of ammonium ions during the

microbes' early growth phase boosts their population

levels significantly and thus the rate of

decomposition of their substrate. This priming effect

occurs even when supplies of organic nitrogen for

decomposition are not limiting. Thus, in the nitrogen

addition area the initially increased decomposition

rate was a reflection of inorganic nitrogen

fertilization increasing the microbe populations.

After a few weeks the rate slowed again as the readily decomposed fraction of the detritus was exhausted

leaving more refractory forms of organic matter. The control area showed an initially lower decomposition

level as the microbe population increased and 79 eventually reached peak levels after two to three weeks. This differential effect disappeared after a few more weeks.

3.4.4 Production

3.4.4.1 Methods

The quantification of productivity proved to be a difficult task in submerged environments. Four basic methods are generally used: l)diurnal oxygen changes,

2)carbon fixation which is a measure of CC> 2 or

HCO^ via *4C or alkalinity, 3)measuring, clipping or stapling leaves and 4)harvest. Table 3 shows results of these four methods for productivity estimates and standing stocks of Thalassia , Zostera

,and Vallisneria grass beds.

The Vallisneria grass bed in Lake Pontchartrain had a standing stock ranging from about 20 g dry 2 2 wt/m to 400 g dry wt/m , which was within the range of standing stocks of both Zostera and Thalassia grass beds. V7ide leaf grass beds ( Thalassia ,

Zostera , and Vallisneria ) have very similar average standing stock values over their entire latitudinal range. MeRoy and McMillan (1977) reviewed standing stock and production values and estimated that the standing stocks turned over from one to three times a year in temperate grass beds and from two to five times a year in tropical grass beds. The Vallisneria turnover rate varied across the grass bed and ranged

from 1.0 to 2.2 times a year for combined above and

below ground biomass. These values are lower than

some literature values because below ground parts,

which turnover only once a year, were included in the

estimates and are typically neglected in literature

estimates of turnover rate. Thus, the turnover rate

reported for this study is among the lowest reported.

Overestimation or underestimation of annual

production by daily productivity methods (such as the 14 dissolved oxygen method or C) can occur due to

the inaccuracy of extrapolating daily production

values to an annual event. Multiplying turnover rates by the standing stocks of grass beds gives an estimate of net annual production. Multiplying the typical

turnover rates estimated by MeRoy and McMillan (1977) by standing stocks measured in the field (Table 3) gives net annual production estimates ranging from 40 2 2 g dry wt/m /yr to 4000 g dry wt/m /yr for 2 Thalassia and 62 g dry wt/m /yr to 5520 g dry wt/m /yr for Zostera . These annual production estimates can by compared with daily rate estimates of annual production by multiplying temperate daily rates by 120 days and tropical daily rates by 250 days to give an annual estimate of production (McRoy and

McMillan, 1977). However, daily productivity rates 01 estimated by the flowing water oxygen and enclosed chamber oxygen methods give estimates of net annual production higher than the range suggested by 14 . . measuring standing stocks. Daily C productivity rates give estimates of net annual production lower than the range suggested by measuring standing stocks.

Measurement of the longest five percent of the leaves or of clipped or stapled leaves as used by

Zieman (1968 and 1972) and Patriquin (1973) may be the least ambiguous approach to productivity; estimates fell within the standing crop times turnover rate but the method also has a variety of drawbacks. While the leaf length may be statistically related to productivity, leaf length varies with water depth and environmental conditions making a weighted average production estimate for an entire grass bed difficult.

Clipping may depress regrowth while stapling may increase likelihood of bacterial attack leading to breakage. Once measured, the leaf growth rate must be converted to productivity via assumptions of the average leaf length, number of leaves per unit area, growth per leaf, weight changes per unit growth, etc.— all of which may vary. Below ground production is also ignored by this method. Measuring leaf growth requires measurement of a statistically significant sample of leaves— a difficult task in Lake 82

Pontchartrain due to low water visibility and violent

winter storms.

The harvest method usually includes no estimate

of below ground production and no exact categorization

of "losses" among possible causes like respiration,

excretion, secretion, injury, grazing or death.

However, when below ground material (roots and

rhizomes), litterfall and decomposition are included

more direct and accurate measurements of net

production can be obtained. Production and standing

stock estimates generated at the same time can be

compared with the large number of literature values

for many ecosystems directly in g dry wt/m without having to make conversions. Thus, it is felt

that the harvest technique has many benefits as well as fewer inherent errors than the previously discussed techniques because it does not assume any physiological relationships and does not alter the

instantaneous growth rate of the plants. The

techniques discussed in this paper could be used to determine production of submerged grass beds in many different environments and the results could easily be compared to production literature.

3.4.4.2 Value of Harvest Results

From the. data collected during this study it is apparent that the grass beds turn over more than once 33 per season since peak standing stock underestimates

grass bed production. Patterns of production across

the grass bed can be determined and are useful in

estimating production of the entire grass bed. The addition of litterfall and decomposition to production estimates is particularly important near shore where the highest litterfall and decomposition rates occur.

Since litterfall and decomposition are continuously occurring throughout the year, the cycling of energy and nutrients to the rest of the food chain is continuous and no buildup of standing litter occurs.

Decomposition rates indicate that all of the annual production could be mineralized on the site. However, the sudden mortality in the winter produces a large amount of litter which apparently leaves the area, contributing organic carbon to other parts of the l a k e .

In order to assess the importance of grass bed productivity to Lake Pontchartrain, timing of the introduction of this organic matter to the next trophic level may be as critical as cumulative yearly production. Decomposition rates in the grass beds are high when compared with those of the nearby emergent ecosystems. The grass bed's continuous turnover of leaves via litterfall contributes a more constant supply of organic material to the estuary than the 84

emergent ecosystems which release detritus to the lake

primarily in the spring. Thus, although their

relative area and average productivity rates may be

small, the grass beds contribute an important source

of organic carbon to the system in the summer and fall

(Figures 3 and 4) when other sources of organic carbon may be low. 35

3.5 LITERATURE CITED

Army Corps of Engineers. 1979. Mississippi River Delta, Bonnet Carre' Spillway Near Laplace, La.: Water quality data water year October, 1978 - September, 1979. St. Charles Parish Hydrologic Unit, 6.4 km southeast of Laplace.

Buesa, R. J. 1972. Production primaria de las praderas de Thalassia testudinum de la platforma noroccidental de Cuba. INP, Cuba Cent. Inv. Pesqueras, Reva. Bal. Trab. CIP, 3:101-143.

Burrell, D. C., and J. R. Schubel. 1977. Seagrass ecosystem oceanography, pp. 195-227, I_n : C. P. McRoy and C. Helfferich, (Eds.) Seagrass Ecosystems. A Scientific Perspective . Marcel Dekker. N.Y.

Capone, D. G. and B. F. Taylor. 1980. Microbial nitrogen cycling in a seagrass community. Estuarine Perspectives. 153-161.

Carr, W. E. S. and C. A. Adams. 1973. Food habits in juvenile marine fishes occupying seagrass beds in the estuarine zone near Crystal River, Florida.Trans. Am. Fish. Soc. 102:511-540.

Conover, J. T. 1964. Biology, seasonal periodicity, and distribution of benthic plants in some Texas lagoons. Dot. Marina, VII:4-41.

Cramer, G. W., and J. W. Day. 1981. Productivity of four marsh sites surrounding Lake Pontchartrain, Louisiana. Amer. Mid. Naturalist. 106:65-72.

Darnell, R. M. 1961. Trophic spectrum of an estuarine community, based on studies of Lake Pontchartrain, Louisiana. Ecology 43:553-568.

Dillon, C. R. 1971. A comparative study of the primary productivity of estuarine phytoplanktonand macrobenthic plants. Ph.L. Dissert., Dept. Botany, Univ. No. Carolina, Chapel Hill. 112 pp.

Doty, M. S. 1971. Measurement of water movement in reference to benthic algal growth. Bot. Mar. X I V : 32-35.

Fenchel, T. 1970. Studies on the decomposition of organic detritus derived from the turtle grass Thalassia testudinum . Limnol. Oceanogr. 86

15:14-20.

Ginsburg, R. N. and H. A. Lowenstam. 1958. The influence of marine bottom communities on the depositional environment of sediments. Jour. Geol. 66:310- 318.

Kasselman, M. 1980. Nitrogen fixation in a south Louisiana Spartina salt marsh. M. S. Thesis. Louisiana State University. Baton Rouge, La. 110 pp.

Kemp, W. M., W. R. Boynton, and R. R. Twilley. 1984. Influences of submersed vascular plants on ecological processes in upper Chesapeake Bay. pp. 367-394, I_n :V. S. Kennedy (Ed.), The Estuary as a Filter . Academic Press, Inc.,

MeRoy, C. P. 1970. Standing stocks and related features of eelgrass populations in Alaska. J. Fish. Res. Bd. Canada 27(10):1311-1821.

MeRoy, C. P. and C. McMillan. 1977. Production ecology and physiology of sea- grasses, pp. 53-88, C. P. McRoy and C. Helfferich, (Eds.), Seagrass Ecosystems A Scientific Perspective . Marcel Dekker, Inc. N.Y.

Montz, G. N. 1978. The submerged vegetation of Lake Pontchartrain, La. Castanea 43:115-128.

Nixon, S. W. and C. A. Oviatt. 1972. Preliminary measurements of midsummer metabolism in beds of eelgrass, Zostera marina . Ecology, 53(1):150-153.

Odum, H. T. 1957. Primary production of eleven Florida springs and a marine turtle grass community. Limnol. Oceanogr. 2:85-97.

Odum, H. T. 1963. Productivity measurements in Texas turtle grass and the effects of dredging an intracoastal channel. Publ. Inst. Mar. Sci. Texas. 9:45-58.

Patriquin, D. G. 1973. Estimation of growth rate, production and age of the marine angiosperm, Thalassia testudinum , Konig. Carib. Jour. Sci. 13:111- 123.

Smalley, A. E. 1959. The role of two invertebrate 87

populations, Littorina irrorata and Orchelimum fidicinium in the energy flow of a salt marsh ecosystem. Ph.D. Thesis. Univ. of Ga., Athens. 126 p p .

Smith, 0. L. 1979. Application of a model of the decomposition of soil organic matter. Soil Bio. Biochem. 11:607-618.

Thompson, B. A. and J. S. Verret. 1980. Nekton of Lake Pontchartrain, Louisiana, and its surrounding wetlands. pp. 711-864, Ln : J. H. Stone, (Ed.), Environmental Analysis of Lake Pontchartrain, Louisiana, Its Surrounding Wetlands, and Selected Land Uses . Vol. 2. U.S. Army Engineer District, New Orleans, Contract No. DACW-29-77-C-0253.

Wetzel, R. G. 1964. A comparative study of the primary productivity of higher aquatic plants, periphyton and phytoplankton in a large, shallow lake. Int. Rev. Ges. Uydrobiol. 49:1-61.

Zieman, J. C. 1968. A study of the growth and decomposition of the sea-grass Thalassia testudinum . M.S. Thesis. Univ. of Miami, Coral Gables, Fla. 50pp.

Zieman, J. C. 1972. Origins of circular beds of Thalassia (Spermatophyta: Ilydrocharitaceae) in South Biscayne Bay, Florida, and their relationship to mangrove hammocks. Bull. Mar. Sci. 22(3):559-573.

Zieman, J. C. 1975. Seasonal variation of turtle grass, Thalassia testudinum Konig, with reference to temperature and salinity effects. Aquat. Dot. 1:107-123.

Zieman, J. C. and R. G. Wetzel. 1980. Productivity in seagrasses: methods and rates, pp. 87-116, Iri : R. C. Phillips and C. P. M e R o y , (Eds.), Handbook of Seagrass Biology: An Ecosystem Perspective . Garland STPM Press, N.Y.

Zimmerman, M. S. and R. J. Livingston. 1976a. Effects of Kraft-mill effluents on benthic macrophyte assemblages in a shallow-bay system (Apalachee Bay, north Florida, USA). Mar. Biol. 34:297-312.

Zimmerman, M. S. and R. J. Livingston. 1976b. Seasonality and physico-chemical ranges of benthic macrophytes from a north Florida estuary (Apalachee Bay). Contributions in Marine Science, 20s33-44. 89

Table 1. Mean biomass and analysis of variance data used in testing for differences in Vallisneria americana biomass with time, distance from shore, and area along the shore, for all biomass stations (n-193).

Treatment Mean Biomass Site (g dry wt/m )

s *6 • Control 56 + 6 Nitrogen enrichment 32 + 6

Distance from Shore 2 5m 29 + 5 50m 56 + 10 75m 75 + 10 100m 8 + 4

Analysis of Variance Source df F PR>F

Model 79 3.05 0.0001** Error 114 Time 9 2.81 0.0053** A r e a 1 13.55 0.0004** Distance 3 6.84 0.0003** Area * D i s t 3 11.12 0.0001**

**=Highly Significant (P<0.01);Time*area, time*distance,time*area*distance, distance to creek, creek*time, creek*area, and creek*distance, were not significant. 90

Table 2. Annual production of the Vallisneria americana beds measured by four harvest techniques at four distances from shore in a control and a nitrogen enrichment area.

ANNUAL PRODUCTION (g d ry w t / m /yr) Control Area Nitrogen Addition Area 25m 50m 75m* 100m* 25m 50m 75m* 100m* Weighted Weighted Average+ Average+

NAPPp 105 160 199 81 101 51 173 68 24 75

NAPP„ 151 192 250 178 148 60 274 68 24 102 11

NAPP l 179 201 75 276

NAPP d 281 225 100 280

*No litterfall or decomposition values are included at 75m and lOOnf from shore due to insufficient data. NAPPp=net annual primary production estimated by the peak production method NAPPH=net annual primary production estimated by the harvest production method NAPPL=net annual primary production estimated by the harvest plus litter method NAPPj=net annual primary production estimated by the harvest plus litter plus decomposition method +Weighted average: the area under the production versus distance from shore curve was calculated and divided by the total distance from shore. Distances included those both near and far from shore where zero production occurred. Table 3. Comparison of Lake Pontchartrain Vallisneria americana grassbed production and standing stocks with tropical and temperate suhmergent grassbed estimates. Location Dominant* Author Me thod Sdays** Annual Net Production *** Standing Stock Turnover**** Plant (g dry wt/m /yr) (g dry wt/m )(per year)

Louisiana V This IIAPPU 365 172-281 24-400 1.2 - 2.2 study Tlorida T Odun oxygen 250 37011-10,400 20-400 2 - 5 (1963) open water

n Cuba X I’.uesa oxygen 300 7233-9747 200-800 2 - 5 (1972) chamber Flor iua T Sieman+ leaf 300 1326-1793 50-250 2 - 5 Uiscaync (1966) narking Day • Taylor and Capone++ (19C0) Rhode Island n Nixon oxygen 120 e n 10C 1 - 3 and open water Oviatt (1972) Alaska ff lie Hoy oxygen 120 1030-1186 52-1340 1 - 3 (1970) chamber North Carolina Z billon 14C 120 53-606 175-545 1 - 3 (1970)

V = Thalassia , Z = Zostcra ** number of days o£ productivity annually, from estimates by Hclloy and McMillan (1977) *** converted from original data assuming .3xg(), . «jc and 2.6 g dry wt/gC (McHoy and McMillan. 1977) turnover estimated by McHoy and McMillan (T977) except in this study + and ++ values for I’.iscayne bay, I’lorida, taken at sane site but at different tines 92

Figure 1. Location of the Lake Pontchartrain grass beds (after Darnell, 1961, and Montz, 1978), and the control and nitrogen enrichment study sites. 93

LAKE PONCHARTRAIN

BONNET CARRE SPILLWAY 70

-N LOUISIANA CREEK

STUDY AREA MARSH

CONTROL 3 AREA

- NAJAS GUADALUPENSIS

. . 1 - GRASS SAMPLES (Z l ~ VALLISNERIA AMERICANA and ® - GRASS SANPLES and UTTER RUPPIA MARITIMA. 1954 a - MUUS BALL BRICKS p™. J AND WATER SAMPLES - VALLISNERIA AMERICANA and o - CORES RUPPIA MARTIMA. 1973. • - CORES and DETRITUS BAGS 94

Figure 2 Submerged litter trap showing construction and position in the sediment. 95

TOP LATCH

ALUMINUM SCREEN WIRE E o 10 SEWN TOGETHER CM WITH * V f f ALUMINUM WIRE

1 A 4 / HARDWARE CLOTH COVER ANCHORS

P LA C E D - N E X T T O PERMANENT STAKE

EDGE8 BURIED IN 8EDIMENT 96

Figure 3. Vallisneria americana biomass (means of all plots) in g dry wt/square meter versus time. BIOMASS (GM DRY WT/M2) 250- - 300 T 200 150-- 100 50-- A JN U AG E OT O DC A FB A AR A JUN MAY APR MAR FEB JAN DEC NOV OCT SEP AUG JUL JUN MAY -- -- REPRODUCTIVE 1978 (MONTHS) LEAVES RHIZOMES TIME STANDARD ERROR 1979 97 98

Figure 4. Daily average solar radiation, average water temperature, average water depth, mean salinity, and average currents for May, 1973, through April, 1979, at the Vallisneria grassbed (combined data for the control area and the nitrogen enrichment area). V. L0SS/24HR. PPT METERS °C LANGLEYS A JN U AG E OT O DC A FB A APR FEBMAR JAN DEC NOV SEPOCT JUL AUG JUN MAY 98 MONTH 1978 DAILYAVERAGE SOLAR RADIATION AT SLIDELLLA. 1941-1970 AVERAGECURRENTS MEANSALINITY AVERAGE WATER TEMP. AVERAGE WATER DEPTH 1979

99 100

Figure 5. Water column (A) and sediment (B) available inorganic nitrogen at the control area (symbol C) and the nitrogen enrichment area (symbol N). Water column nitrogen is shown versus time (months) and sediment nitrogen is shown versus distance from shore in meters. WATER COLUMN TOTAL AVAILABLE INORGANIC NITROGEN SEDIMENT INORGANIC NITROGEN IN PPT

•7*

t I

t ■

■ i ai I I i I 1 .2

I ) 4 I I 7 • « Ifl II 12 M S 40 M <0 IB m * 0 100 ■ n s . M I 11*911 - M I L CI«T*> BISTMRO 9BQR SBOKS 10 M TB S 101 (A) IB) 102

Figure 6 . Monthly production measured by the harvest method (NAPP„), the harvest plus litter method (NAPP.), and the harvest plus litter plus decomposition method (NAPPp), at 25m from shore at the control area (a), 50m from shore at the control area (b), 25m from shore at the nitrogen enrichment area (c), and 50m from shore at the nitrogen enrichment area (d). 103

HARVEST, LITTER, AND DECOMPOSITION PRODUCTION 25M FROM SHORE AT THE CONTROL AREA CRAMS DRY WEICHT PER SQUARE METER

110

100

90

70

60

50

40

30

20

-30

-40

-50 0 1 2 3 5 6 74 e 9 10 11 12 MONTHS, MAY 1978 - APRIL 1979 (a) 104

HARVEST, LITTER, AND DECOMPOSITION PRODUCTION 50M FROM SHORE AT THE CONTROL AREA CRAMS DRY WEIGHT PER SOJARE METER 60

50

40

30

20

-to

-20

-30

-40

-50 0 1 2 3 4 56 7 8 9 10 U 12 MONTHS. MAY 1978 - APRIL 1979 (b ) 105

HARVEST, LITTER, AND DECOMPOSITION PRODUCTION 25M FROM SHORE AT THE NITROGEN ADDITION AREA GRAMS DRY WEIGHT PER SQUARE METER 60

55

50

45

40

35

30

25

20

15

10

5

0

-5

-1 0

-15

-20 0 1 2 3 4 5 6 7 e 9 10 11 12 MONTHS, MAY 1970 - APRIL 1979 (c) 106

HARVEST, LITTER, AND DECOMPOSITION PRODUCTION 50H FROM SHORE AT THE NITROGEN ADDITION AREA GRAMS DRY WEIGHT PER SQUARE METER

140

120

100

80

60

40 -■

20 -j

0 v

•20 ■

-40 ■

-60 H

-80 ^

-100 0 2 3 4 5 6 7 8 9 10 12 MONTHS, MAY 1978 - APRIL 1979 (d) Figure 7. Decomposition of Vallisneria americana leaves in litterbags for summer (a) compared with Thalassia and upland marsh decomposition# and for winter (b). Months are numbered sequentially starting in May, 1978. (l=May, 2=June, 3=July, 4=August, 5=September, 6=0ctober, 7=November, 8 =December, 9=January, 10=February, ll=March, 12=April, 13=May, and 14=June.) DECOMPOSITION OF VALUSNERIA ■ REMAINING PER TIME

90

eo

•o

60

VJ—J

30

20

I 2 3 4 6 7 m m s lUT, 1970 - JUNE. 1979 LEGEND I SAMPLE I-2SM. CONTRU. Z-50H. CONTROL 3-7SM. CONTROL 4-2SM, NITROGEN ADDITION S-SOM, NITROGEN ADDITION 6-7SH. NITROGEN ADDITION

T-THALASS1A IfGOOSEPOINT HARSH 108 (•) DECOMPOSITION OF VALUSNERIA x KMixixc pea line vinteb dbcompositjon

90

70

•o

EO

40

30

20

7 a 10 II 12 13 14 MONTHS IUT, 1970 - JUNE, 1979 LeceMDi SAMPLE I-25H, CONTHOL 2-50H. CONTHOL 3-75M. CONTHOL

4-25M, MITOGEN ADDITION 5-50H, NITROGEN ADDITION 6-75H, NITROGEN ADDITION 109 (b) CHAPTER 4

ENVIRONMENTAL AND PHENOLOGICAL CORRELATES OF GROWTH OF THE SEAGRASS Vallisneria americana IN LAKE PONTCHARTRAIN

ABSTRACT The above ground and below ground biomass of

Vallisneria varies significantly with time of year,

area along the shore, and distance from shore. In

shallow water, biomass of Vallisneria increases

rapidly in the spring and summer and stabilizes at 200

to 400 g dry wt/m during the late fall. Growth

occurs throughout the year except in severe winters.

Seasonal biomass and production were not

significantly correlated with any one environmental parameter when compared at four distances from shore.

Seasonal variations in environmental parameters are

discussed in relation to their effect on biomass and production in the grass bed. Net production was found

to be negatively correlated in all seasons with

sediment available nitrogen and phosphorus. During

the biomass peak in late fall when sediment nitrogen was low, available nitrogen in the sediment may limit net production. Thus, timing of nutrient availability may affect biomass as a limiting factor.

The grass bed had a distinctive cross-bed profile, with maximum growth occurring in the center of the bed. Depth distribution appeared to be 110 Ill

affected by the seasonal extremes of the environmental

factors near to and far from shore and by the location

of the first revegetation in the spring at mid-depths.

Growth of grass in the spring is determined by

successful overwintering of existing vegetation and

seed development from the previous year's growth.

Minimum water depths and temperatures and violent

storms which occur in January and February often lead

to sudden decreases in biomass. Existing vegetation may be virtually eliminated during severe winters.

Fruits, however, overwinter well since they are

dormant during the winter. Fruits waterlog and fall

close to their site of origination. The majority of

fruits were found at mid-depths where net production was greatest— a result suggesting that only those areas with substantial net production produce fruit.

Sediment stability is also important for fruit survival during the dormant winter season. Sediments near shore were often disrupted during winter storm events since cores often showed a 10 to 15cm loss or gain of sediment. Associated experiments with seedlings showed that burial by 1 0 cm of sediment may decrease seed germination and seedling survival. At mid-depths environmental conditions are less severe.

Patches of vegetation surrounded by bare sediment are common at mid-depths in the early spring. These patches of vegetation may develop from clumps of

seedlings which germinate in the early spring. Growth

radiates out from these vegetated areas via the

rhizome system and by late fall the entire grass bed

is revegetated. The hardiest growth is in the center

of the grass bed which developes first.

The pattern of distribution with maximum biomass

at mid-depths is similar all along the coast, although

the width of the pattern varies along the coastline.

Compensation depth and average secchi depth at the

control area were found to be greater than the

compensation depth and average secchi depths at the

nitrogen enrichment area. Increased turbidity due to

dredging or eutrophication which accelerates algal

growth may decrease light penetration and limit the

maximum depth of the macrophytic community.

4.0 INTRODUCTION

Vallisneria americana is the dominant submerged

macrophyte in Lake Pontchartrain, Louisiana. The

productivity of the Vallisneria grass bed is important

to the local blue crab ( Callinectes sapidus ) and

shrimp ( Penaeus aztecus ) fishery. The grass bed has declined over the last 25 years due to

urbanization and industrialization in New Orleans. A 113

study to determine what factors affect growth of the

grass bed was undertaken in order to develop

management criteria for the remaining grass bads.

Vallisneria americana grass beds occur in

shallow water along the northeastern shoreline of Lake

Pontchartrain. Biomass within the grasssbed ranged

from near zero to 400 g dry wt/m . Production within the grass bed ranged from 100 to near 300 g dry wt/m /yr for the 1978-1979 growing season (Chapter

3). The biomass and production across the grass bed

affect estimates of total community production. The highest biomass and production estimates for the Lake

Pontchartrain grass bed were in the center of the grass bed, which occurred from 50m to 75m from shore.

This pattern of biomass and production was similar to that found in circular grass beds in Florida by Zieman

(1972), who suggested that the depth of sediment determined the distinctive central biomass peaks.

However, other environmental factors such as light penetration (Dennison, 1935) and minimum water temperature (Steller, 1976) may also be important in determining depth distribution.

The two objectives of this study were: l)to relate environmental and phenological factors to the biomass distribution with depth across the grass bed, and 2 )to determine which factors affect the width of 114

the grass bed. Vallisneria americana biomass and

production will be compared with environmental

parameters to determine the relative importance of

environmental parameters on the pattern of biomass

distribution with distance from shore and location

along the shoreline.

4.1 DESCRIPTION OF STUDY AREA

Vallisneria americana occurs along sandy

shorelines out to 150m from shore on the northern part

of Lake Pontchartrain at about 30°30'N latitude,

90°W longitude (Figure 1). Zonation across the

bed showed Eleocharis parvula near shore, then

Vallisneria and Ruppia maritima , and finally patches

of Vallisneria americana among Rangia cuneata

shells. Salinities ranged from near 0 ppt to about 4

ppt over the year with considerable freshwater input

occurring in the spring, especially during openings of

the Bonnet Carre Spillway. Water temperatures are

closely related to air temperatures and range from

9°C to 34°C annually. Water depths range from

0m to 2m spatially.

4.2 METHODS

.Two study sites were chosen along the north shore

of Lake Pontchartrain (Figure 1). One area was a

control site and the other a nitrogen enrichment site.

The shores at both sites were characterized by sandy 115 dunes, an undulating shoreline, and tidal creeks which drained brackish marshes and originated about 2 0 m from the edge of the sites.

Biomass was measured monthly at 25m, 50m, 75, and

1 0 0m from shore along two transects at each site from

May, 1978 through May, 1979. Presence or absence of leaves, roots, rhizomes, flowers, fruit, newly originated short shoots (apical meristems), and attached algae were recorded monthly for the two study sites.

Over 120 Vallisneria fruit were collected from

50m to 75m from shore in a grass bed near Mandeville,

Louisiana. The fruits were placed in fresh water and allowed to germinate. A box was built and placed 25 cm below the water surface in a fresh water pond in the Lake Pontchartrain drainage basin about 25km west and 10km north of Lake Pontchartrain in December in order to study germination and seedling survival over the winter. One hundred twenty fruits were placed in the box on sand and sixty fruits were covered with 10 cm of sand to simulate burial by storms during the winter.

Cores for nutrient determinations and grain size analysis were taken quarterly at 25m, 50m, 75m, and

100m from shore at each of the study sites. Dissolved oxygen, temperature, water depth, secchi depth, 116 current, and water chemistry were sampled monthly at the stations indicated in Figure 1. Surface irradiance data were obtained from the National

Climatic Center for Slidell, Louisiana. Details of methods can be found in Chapter 3.

4.3 RESULTS

4.3.1 Biomass and Net Annual Production Patterns

It was shown in Chapter 3 that significant (0.01 level) differences in Vallisneria biomass were contributed by distance from shore, area along the shore, and time. Figure 2 shows monthly biomass at the two study areas.at each distance from shore.

Biomass generally increased from low values in the spring through late summer, and was minimal in late winter. However, the biomass peaks were higher at mid-depths.

Maximum net annual primary production of

Vallisneria at two study areas occurred in the center of the grass bed (Figure 3). Measuring distance from the shoreline to the maximum point perpendicular to the shore where the grass bed ceases, the "center" is here defined as half the maximum distance from shore.

The center of the grass bed occurs at different distances from shore at the two areas. At the control site, the highest production and the center of the grass bed coincided at about 75m from shore. At the 117 nitrogen enrichment area the highest production again coincided with the center of the grass bed, which is at 50m from shore. The width of the grass bed also varied with location along the shoreline, and was almost twice as wide at the control area as at the nitrogen enrichment area.

4.3.2 Environmental Parameters Related to Distance from Shore

Environmental parameters related to growth are light, water depth, water temperature, available nutrients, sediment type, and wave action and currents

(Phillips, 1960; :ioore, 1963; Thayer, et al. , 1975).

When these factors were graphed versus distance from shore for the two study areas (Figure 4), no single factor followed the bell shaped production curves.

When light at the sediment surface, average water depth, minimum water temperature, maximum water temperature, sediment type, sediment nitrogen, sediment phosphorus, and currents were correlated with biomass and production at four distances from shore

(Table 1), none of these physical factors were highly correlated with biomass or production (magnitude of correlation coefficients (r) was less than 0.60, probability (p) greater than 0 .1 ).

The extreme values of sunlight at the sediment 118

surface, water depth, and maximum water temperature were either near or far from shore— a finding which may in part account for the depression of maximum production near and far from shore. These

environmental factors were highly correlated with each other. Sunlight reaching the sediment surface was negatively correlated (r=-0.95, p=.0.0002) with water depth. Maximum water temperatures were also highly correlated with bottom sunlight(r=0.82), as well as negatively correlated with water depth (r=-0.83).

4.3.3 Environmental Parameters Related to Season

Environmental parameters vary both spatially and seasonally and may, therefore, influence seasonal patterns of grass bed production. Water depth generally increases with distance from shore and also undergoes a seasonal change (Figure 5). Depth gradually increases during the summer and fall and is at a seasonal low in the winter. In May the nitrogen enrichment area had water depths exceeding those at the control area at all distances from shore. The two study areas had very similar seasonal patterns of depth distributions with distance from shore the rest of the year.

Secchi depth differed significantly (p<0.05, paired Student's "t" test) between the two study sites

(Figure 6 ). The control area generally had greater 119

secchi depths (i.e., lower turbidity) most of the

year. Lower secchi depths at the nitrogen addition

area, particularly during August, could have been in

response to increased phytoplankton and algal growth

due to the nitrogen enrichment. There was an increase

in attached algae on the Vallisneria leaves at the

nitrogen enrichment area, particularly at 25m from

shore. However, since no direct measurements of

phytoplankton were made, it can only be stated that

turbidity was significantly higher at the nitrogen

enrichment area.

Water depth and secchi depth can be related to

light attenuation in water through the equation (Pond

and Emery, 1982):

I^e"*1* 7*Z>/SD (Equation 1)

I=surface irradiance

IQ=light at the bottom

Z=water depth

SD=secchi depth

Thus, light reaching the bottom is dependent upon the

surface irradiance (I), water depth (Z), and secchi

depth (SD). Sunlight reaching the bottom provides the

sunlight energy available for phytosynthesis of the

seedlings and young plants. Solar radiation reaching

the sediment surface was graphed versus distance from

shore and time (Figure 7). There was less light at 1 2 0 the sediment surface at the nitrogen enrichment site during May and June? the beginning of the growing season. This v/as partially due to increased water depths (Figure 5) and to decreased secchi depths

(Figure 6 ).

Water temperatures during the study exceeded the thermal production maximum and approached the thermal production minimum. Photosynthesis reaches a peak around 29°C, above which it declines and ceases above 32°C. Maximum water temperatures at the

Lake Pontchartrain grass bed exceeded this upper limit during June, July, and August of 1978 at 25m from shore, and in July at all depths (Figure 8 a).

Production rates decreased near shore (Fig. 5, Chapter

3) during July, possibly in response to the high water temperatures. High temperatures increase respiration rates and cause net production to decline. Low water temperatures near or below the 1 0 ° lower limit for growth (Barko, 1980) occurred during January (Figure

8b). Average plant biomass declined markedly in

January (Fig. 3, Chapter 3).

The time of year when available nitrogen was lowest in supply was December (early winter), although generally nitrogen and phosphorus did not limit production in the grass bed. Increases in seagrass biomass nitrogen over time compared with available 121

sediment nitrogen have been used by Patriquin (1973)

to estimate sediment nitrogen reserves (days of

nitrogen supply available for growth). This technique was used for the control area, 50m from shore, to

estimate the sediment reserves available for plant

growth over the year. Table 2 shows that the month with the lowest supply of sediment nitrogen for plant

growth (on the assumption that all nitrogen for growth

came from the sediments, and not from remineralization

in the sediments) was December. December has high

biomass (Figure 3) per square meter and nitrogen

necessary to maintain the level of growth at that time

is approximately 38 gN/m /day. At that level of

growth, the sediment nitrogen supplies would be used

in ten days. December was time of year of the highest

standing stock, relatively low mineralization rates

(Chapter 3), and low sediment nitrogen supply. The

Patriquin (1973) technique was a rough indication of '

the time of year which would be the most critical for

nitrogen supply. Sediment available inorganic nitrogen and phosphorus at the two study sites were

compared with quarterly grass bed net production.

Linear regression of sediment nitrogen and phosphorus with quarterly production revealed that production was not significantly correlated (r=-0 .6 ) with either

sediment available nitrogen or phosphorus. 122

4.3.4 Phenology of the Grassbed

Fruit dormancy during the winter, fruit germination near site of origination and overwintering of surviving vegetation ensure revegetation of the grass bed in the spring. Vallisneria leaves, roots, and rhizomes are present most of the year, but may be absent in severe winters (Table 3). March, 1978, was the spring of a particularly severe winter and no

Vallisneria leaves, roots, or rhizomes remained present at the control or nitrogen addition areas after that winter. However, Vallisneria revegetated the control and nitrogen addition areas from seedlings starting in April, 1978. Vallisneria fruits are dormant during the winter (Kaul, 1978). Fruits develop in the late fall and are attached to the plants by long springy peduncles. Fruits eventually waterlog and fall to the sediment surface, still attached to the plants. The peduncles remain intact for two to three months and keep the fruit at the same location where they originated. When 50 wracks of

Vallisneria leaves along the shoreline were inspected for fruit, no fruit were found in the litter— a result suggesting that very few fruit break loose and float to other locations. The fruit remain dormant until the early spring, and the achenes eventually break open to allow germinating seedlings to emerge (Figure

9). Up to 150 viable seedlings were counted emerging 123

from a fruit. Holes or brown damaged areas on the

fruit were the first sites where seedlings emerged.

As the seedlings emerged,- the achene decayed into a

gelatinous mass which held the seedlings together in

clumps (Figure 10) until they attached to the sediment

surface via root hairs (Kaul, 1978). The emerging

seedlings are easily transported by currents when they

are separate, but in clumps they are not easily

dislodged from the sediment surface.

Seedlings could either be buried or washed away when surface sediment was disturbed during winter

storms. Cores taken at the control and nitrogen

enrichment sites in Lake Pontchartrain in December,

1978, and March, 1979, showed from 10cm to 15cm of

sediment lost or gained at 25m and 50m from shore.

Fruits and seedlings are swept away when the upper

15cm of the sediment surface are lost. Burial decreases fruit germination and survival. None of 60

fruit buried under 10 cm of sand germinated. Young seedlings need sufficient light and oxygen for growth and stable sediment conditions to maximize their chances of survival. Of 60 fruit not buried under sand, less than 15% germinated (many of the fruits were lost to predation by a fish population which took up residence over the box).

The most productive areas appear to produce the most reproductive material. The first appearance of 124 2 substantial Vallisneria biomass (25 g dry wt/m ) occurred in May at the control area at 75m from shore

(Figure 2a), and in August at the nitrogen addition area at 50m from shore (Figure 2b). These distances also had the highest annual biomass (Figure 2, a and b), the greatest production (Figure 3), the maximum number of flowers (>200 staminate flowers/m at 75m from shore at the control site in July), and the maximum number of fruit (>15/m at the control site in October). Net production was also fairly high at 25m from shore at the control area (Figure 3) and fruits were present, although not as numerous as at deeper water depths.

4.4 DISCUSSION

The two questions posed in this paper were: l)how do environmental and phenological factors relate to the distinctive biomass depth distribution pattern, and 2 )why is this pattern narrower at some locations along the shoreline? The highest biomass and production occurred at mid-depths, but none of the environmental parameters tested had the same depth distribution (Figure 4). However, the extreme values of most parameters (light available at the sediment surface, water depth, and maximum and minimum temperatures) all occurred near or far from shore— a result suggesting that the survival of vegetation at mid-depths was in response to the environmental 125

extremes near and far from shore.

Maximum seed production at mid-depths, mechanisms which result in the majority of seeds remaining at mid-depths, and increased chances for survival at mid-depths result in increased production at mid-depths and the bell-shaped biomass depth distribution profile of the grass bed. Although annual variations in environmental factors can explain the general grass bed biomass fluctuations over an annual cycle, the mechanisms relating biomass depth distribution with environmental factors are not as easily explained. However, plant growth stages may be useful in understanding the underlying mechanisms resulting in the plant responses to environmental variables and ensuring plant survival (Phillips,

1976).

Since the majority of fruit developed and survived at mid-depths in the area of highest biomass and net production and remained close to the area of origination, the biomass depth distribution pattern was perpetuated. Fruit remained dormant during the winter and germinated in the spring. Successful clumps of seedlings produced patches of vegetation in swales typical of the grass bed at mid-depths in the spring and early summer. New shoots in the spring also emerged from vegetation which survived the winter, and biomass was greatest at mid-depths (Figure 126 2a and 2b).

Sediment disturbances were most prevalent near

shore and limited fruit germination near shore. Fruit buried by 10cm of sediment did not germinate. These

results are similar to Kaul's (1978) for Ottelia , a

species in the same family as Vallisneria . He covered Ottelia seeds with only 0.5 cm of sediment, and only a few germinated, and none formed seedlings.

When he buried seeds from which the hypocotyl had protruded, little further growth continued and all soon died. Even though many fruits were produced 25m

from shore, apparently few survived winter storms, and patches of vegetation were scarce at 25m from shore in the spring.

Maximum fruit production and initial seedling growth occurred where environmental conditions were least severe. Gradually, the grass bed radiated out from the initial patches via the rhizome system. By late fall a fairly continuous grass bed had developed but was most lush in the center. Once the pattern of maximum growth at mid-depths was established, the pattern was perpetuated by fruit dormancy and seedling survival and by vegetation persisting through the winter.

Lower light at the nitrogen enrichment area may have limited seedling survival and plant growth to depths nearer shore than at the control area. 127

The width of the entire pattern may be influenced by bottom topography and the relative clarity of the water. The nitrogen enrichment area had greater water depths than the control area in May. The nitrogen enrichment area also had lower secchi depths than the control area for the majority of the year. The initially greater water depths at the nitrogen enrichment area combined with the effect of lower water clarity decreased the bottom light available to the plants at the nitrogen enrichment area in the spring. After germination, emerging seedlings are particularly vulnerable to low light conditions. In his study of seedling germination in Ilydrocharitaceae,

Kaul (1973) stated that food was provided for the young cotyledon by the early onset of photosynthesis in the cotyledon. Therefore, light is critical to

Vallisneria seedlings from the time that they emerge from the seed coat and may limit their depth distribution away from shore. Biomass at the nitrogen enrichment area had lower values and was closer to shore than at the control area, particularly in the spring. Lower initial biomass occurring nearer shore may have been related to the lower light available for seedling growth at the nitrogen enrichment area.

Since the maximum depth of occurrence is an indicator of the compensation depth, the compensation depth at the nitrogen addition site was assumed to be less than at the control site. The decrease in compensation depth at the nitrogen enrichment area could have been affected by the generally lower secchi depth (i.e., greater turbidity) at the nitrogen enrichment area than at the control area. The compensation depth (maximum depth of occurrence) for established plants is also affected by minimum light intensity required for growth, as well as the presence of suitable substrate (Thayer, £t a^ .,1975). Of these factors, photon flux density (minimum light intensity) has been found to be the primary environmental determinant (Wiginton and McMillan,

1979). It has been shown by many authors (Kikuchi,

1974; Phillips, 1974; Peres and Picard, 1975; Packman and Barlotti, 1976; Sand-Jensen and Borum, 1982 after

Bak, 1979) that when light penetration decreases the compensation depth for minimum plant growth also decreases. When plants are taken from their typical depth and transplanted at depths exceeding their natural depth of occurrence, production slows

(Sand-Jensen and Borum, 1982 after Bak, 1979) and death may occur (Steller, 1976). However, decreased light for plant growth generally is caused by factors other than increasing depth. Increasing human activities in coastal waters are often the causes of decreased light penetration. Lake Pontchartrain water quality has been degraded by New Orleans storm seweraye, industrialization, increasing nitrogen and phosphorus loadings, and shell dredging. Management practices which allow increased nutrient loading or turbidity in Lake Pontchartrain may be detrimental to grass bed survival. 130

4.5 REFERENCES CITED

Backman, T. W. and P. C. Barilotti. 1976. Irradiance reduction: Effects on standing crops of the eelgrass, Zostera marina in a coastal lagoon. Mar. Biol. 34:33-40.

Bak, H. P. 1979. Alegraes, Zostera marina L., Limfjorden 1973. M.S. Thesis, Botanical Institute, Univ. of Aarhus (in Danish).

Barko, J. W. 1980. The influence of selected environmental factors on submersed macrophytes - a summary. A.S.C.E. Symposium, June, 1980. Minneapolis, Minn. 5 pp.

3arko, J. W., D. G. Hardin, and H. S. Matthews. 1982. Growth and morphology of submersed freshwater macrophytes in relation to light and temperature. 60(6):877-887.

Dennison, W. C. 1985. Effect of light and nutrients on the depth distribution of Zostera marina . Presented at the International Symposium on Aquatic Macrophytes, August, 1985, Silkeborg, Denmark.

Kaul, R. B. 1978. Morphology of germination and establishment of aquatic seedlings in Alismataceae and Hydrocharitaceae. Aquat. Dot., 5:139-147.

Kikuchi, T. 1974. Japanese contributions on consumer ecology in eelgrass ( Zostera marina L.) beds, with special reference to trophic relationships and resources in inshore fisheries. Aquaculture, 4:145-160.

Moore, D. R. 1963. Distribution of the seagrass, Thalassia ,in the United States. Bull. Mar. Sci. Gulf Carib. 13(2 ):329-342.

National Climatic Center, 1980. Daily average solar radiation at Slidell, La., 1941 - 1970. N.O.A.A. Env. Data and Information Service, Asheville, North Carolina.

Patriquin, D. 1973. Estimation of growth rate, production and age of the marine angiosperm Thalassia testudinum Konig. Carib. J. Sci., 13(1-2):111-123.

Peres, J. M. and J. Picard. 1975. Causes of decrease and disappearance of the seagrass Posidonia oceanica on the French Mediterranean coast. Aquat. Bot., 1:133-139. 131

Phillips, R. C. 1960. Observations of the ecology and distribution of the Florida seagrasses. Pof. Pap. Florida 3d. Conserv. 2:1-72.

Phillips, R. C. 1974. Transplantation of seagrasses, with special emphasis on eelgrass, Zostera marina L. Aquaculture, 4:161-176.

Phillips, R. C., 1976. Preliminary observations on transplanting and a phenological index of seagrasses. Aquat. Bot. 2:93-101.

Pond, G. L., and E. Emery. 1932. Descriptive Physical Oceanography An Introduction . Pergamon Press, Inc., N. Y. 249 pp.

Sand-Jensen, K. 1977. Effect of epiphytes on eelgrass photosynthesis. Aquat. Bot., 3:55-63.

Sand-Jensen, K. and J. Borum, 1982. Regulation of growth of eelgrass ( Zosera marina L.) in Danish coastal waters. Marine Tech. Soc. Jour., 17(2):15-21.

Steller, D. 1976. Factors affecting the survival of transplanted Thalassia testudinum . pp. 1-20. I_n :(D. Cole, Ed.), Proceedings of the Third Annual Conference on the Restoration of Coastal Vegetation in Florida . Hillsborough Community College, Tampa, Florida.

Thayer, G. W., D. A. Wolfe, and R. B. 'Williams. 1975. The impact of man on seagrass systems. Am. Sci. 288-296.

Wiginton, J. R. and C. McMillan. 1979. Chlorophyll composition under controlled light conditions as related to the distribution of seagrasses in Texas and the U.S. Virgin Islands. Aquat. Dot*, 6:171-184.

Zieman, J. C. 1972. Origins of circular beds of Thalassia (Spermatophyta: llydrocharitaceae) in South Biscayne Bay, Florida, and their relationship to mangrove hammocks. Bull. Mar. Sci. 22(3):559-573. Table 1. Correlation of environmental parameters with production and biomass.

Correlation Coefficients/Probability n-8

I z . HinT MaxT STSIN SPhos Prod Bio

Z 1.0000 -0 .9 5 3 8 0.4339 0.8169 0.2930 0.3173 0.7716 0.2480 0.4154 0.0000 0.0002 0.2828 0.0133 0.4813 0.4438 0.0249 0.5537 0.3061

X 1.0000 -0 .5 9 5 6 -0 .8 3 4 0 -0 .3 0 6 5 -0 .4 8 6 7 -0 .7 0 5 9 -0 .1 3 5 2 -0 .3 4 7 2 0.0000 0.1192 0.0101 0.4604 0.2213 0.0504 0.7495 0.3995

HinT 1.0000 0.4382 -0 .0 3 1 0 0.4983 0.0241 0.3058 0.4361 0.0000 0.2775 0.9419 0.2089 0.9548 0.4614 0.2800

NaxT 1.0000 0.5486 0.6850 0.8223 0.0036 0.1140 0.0000 0.1592 0.0608 0.0122 0.9933 0.7867

ST 1.0000 0.3064 0.4360 -0 .5 9 1 9 -0 .2 1 7 8 0.0000 0.4605 0.2802 0.1221 0.6044

SIN 1.0000 0.5410 -0 .1 2 2 9 -0 .2 2 6 7 0.0000 0.1662 0.7719 0.5892

SPhos 1.0000 0.0674 0.0469 0.0000 0.8741 0.9122

Prod 1.0000 0.7975 0.0000 0.0177

Bio 1.0000 0.0000

I-sunlight at the bottom (langleys) Z-water depth MinT-minimum water temperature ( C) HaxT-maximum water temperature ( C) ST-sediment type SIN-sediment inorganic nitrogen (ppt) SPhos-sediment phosphorus (ppt) , Prod-harvest net annual primary production ig dry wt/m /yr) Bio-mean biomass (g dry wt/m ) 133

Table 2. Sediment and water column nitrogen sources compared with plant growth requirements by months for the 1978-1979 season for 50m from shore at the control area.

Time Water Column Sediment Plant Growth # Days Nitrogen Nitrogen Requirements Supply (ppm) (a) (b) (mgN/n^) (mgN/n^.day)

Jun 0.060 319 0 ---

Jul 0.061 416 6.6 63

Aug 0.028 513 9.1 57

Sep 0.045 513 12.7 40

Oct 0.037 513 27.9 18

Nov 0.030 513 27.9 18

Dec 0.033 360 37.6 10

Jan 0.036 360 0 —

Feb 0.108 360 0 —

Mar 0.080 360 24.1 15

Apr 0.045 931 31.4 30

May 0.193 931 399.8 2 a - The # of days supply is calculated as follows: plant parts are multiplied by their % nitrogen and by positive differences in plant part biomass from one month to the next. This value is divided by the sediment nitrogen available. b - Water column nitrogen data from Miller, 1979. Values include nitrate, nitrite, ammonia and ammonium. 134

Table 3. Presense (+) or absence (-) of Valllsnerla leaves, roots, rhizomes, flowers, fruit, Naias , and Ruppla from May, 1978, through June, 1979, at the Lake Pontchartrain grassbed.

Month Vallisneria Najas Ruppia leaves roots rhizomes flowers fruit

March ------April + + + - - -- May + + + - - - + June + + + +1 - - + July + + + +2 - - + August + + + + - + + Septemeber + + + + - - + October + + + + + + + November + + + + - December + + + - + - - January + + + - + -- February +3 +3 + - + --

March +3 +3 + - - --

April +3 +3 + - - - + May + + + + -- - June + + + + —— + lfirst appearance at the control site 2first appearance at the nitrogen addition site 3part present in less than five samples out of 32 135

Figure 1 Location of the study area in Lake Pontchartrain, Louisiana. 136

LAKE PONCHARTRAIN

BONNET CARRE SPILLWAY XEW drums

-N CREEK

MARSH

- NAJAS GUADALUPENSIS

& - GRASS SAMPLES g - GRASS SAMPLES and UTTER A VALLISNERIA AMERICANA and RUPPIA MARITIMA. 1954 - MUUS BALL BRICKS AND WATER SAMPLES VtNl — VALLISNERIA AMERICANA and o - CORES RUPPIA MARTIMA. 1973. - CORES and DETRITUS BAGS 137

Figure 2. Monthly Vallisneria biomass versus distance from shore and time at the control area (A) and the nitrogen enrichment area (B). (Biomass values=grams dry weight/square meter). MONTHLY GRASS BIOMASS VERSUS DISTANCE FROM SHORE AND TIME OONTROl.Ara III m

CD CD

a CD

a a CD TT W

MONTHS, MAY W78 - APRL W79 MONTHLY GRASS BIOMASS VERSUS DISTANCE FROM SHORE AND TIME MTO0GB4 BnCHyBff AREA I I I

IM CD CD CD CD CD CDCD

CD CD CD a CDCD

CD a V na MONTHS, MAY «78 - APRL 1979 139

Figure 3 Net annual production estimated by the simple harvest method at the control area (C) and at the nitrogen enrichment area (N). NET ANNUAL P R O D U C T S VERSUS DISTANCE FROM SHORE DISTANCE FROM VERSUS S T C U D O R P ANNUAL NET RDCINETAE YSML AVS ETHOD M SIMPLE HARVEST BY ESTMATED PRODUCTION

ta>m-<\»mHni3E mroeowx-)*; -<»3 C 200 260' 240' 2804 100 120 1604 140 1WV 204 604 404 ' ' 0 20 EED TT » 1 2 » - — 1 • » * TRT LEGEND) 060 40 ITNEFO HR CM) FROM DISTANCE SHORE 60 100 120 140 140

141

Figure 4. Environmental parameters versus distance from shore (m). (a) sunlight (langleys), (b) water depth (m), (c) maximum water temperature ( C), (d) minimum water temperature ( C), (e) sediment inorganic nitrogen (ppt), (f) sediment phosphorus (ppm), (g) grain size (l=sand, 2 =silty sand, 3=silt, 4=silty clay, 5=clay), and (h) currents (% loss after 24 hr.). 25.0" o METERS LANGLEYS 22.5 27.5" 0 0 2 0 0.5" 0 0.7-- 20 0 r 22 0.3 - 140" 160" 180 * * 100 120 0 2 18" . . 8 6 - 20 " " " " "

30 ITNE FROMDISTANCE (METERS) SHORE ULGT T EIET SURFACE SEDIMENT AT SUNLIGHT 05 70 50 40 WATER TEMPERATURE MINIMUM WATER MAXIMUM TEMPERATURE HATER DEPTH CD) (C) (A) (B) 60 80 80 100 142 PPM 0.08 0 0 0.16" 0.18" 0 0.22 Q. .10-- . . 0 2 12 10 12 " 4 1 18 r -- - 20 " " -- -- 30 DISTANCE FROM SHORE (METERS) 40 SEDIMENT INORGANIC NITROGEN SEDIMENT PHOSPHORUS 0 0 0 80 70 60 50 (F) SO 143 144

3.25 t SEDIMENT TYPE

3.00”

2.75”

2.50” ui £2.25 + in

£2.00 + C9

1.75”

1.50”

1.25”

(G) 1.00 -1 H

40” CURRENT

cc•35” X »»• CM ^30 + tncn o * 2 5 ”

(H) 20 20 30 40 50 60 70 80 90 100 DISTANCE FROM SHORE (METERS) 145

Figure 5 Water depth at 25mf 62.5m, and 100m from shore verses time at the control area (A) and the nitrogen enrichment area (B). WATER DEPTH VERSUS TIME WATBR DEPTH VERSUS TIME CONTROL AREA NTROGBi BdCHMENT AREA 1.0-

0.0-

0 .7 0 .7 -

' 0.4 0 .4

T 0 . * 0 .5 ; lARHHna 0 .4 0.4

0 .3 - 0 .5 ;

0.2 0.2-

0.1 0.1

0.* 0.0: o lUTJUNJULAUsssrocTMOvcecjAarEsruiAPt 0 HAT J IM J IIL AUC SEP OCT MOV DEC JAN FED HAH APR rams, hat mo - apiil mo rams, hat ma - awil mo LSCOBi DISTAKE 25 42.5 100 LECEHDi D1STAKS 25 ♦ ♦-* 42.5 100 147

Figure 6 . Secchi depth versus time at the control area (C) and the nitrogen enrichment area (N). 148 TIME LEGEND i TRT TRT 1 2 i —■> •—■ » • LEGEND TIME IN MONTHS. MAY Cl978) - APRIL Cl979) 0 MAY JUK JUL AUG SEP OCT NOV DEC JAN .FEB MAX APR MAX .FEB JAN DEC NOV OCT SEP AUG JUL 0 JUK MAY - - 10 40- 30- 2 0 190 140 170 ICO 130 150 100 230-j 210 220 200 120 110

cnuiouxi-t o u c u h x p-*z o x SECCHI DEPTH VERSUS TIME BY TREATMENT 149

Figure 7. Solar radiation at the sediment surface (langleys) at 25m, 62.5m, and 100m from shore versus time at the control area (A) and the nitrogen enrichment area (B). SOLAR RADIATION AT THE SEGMENT SURFACE SOLAR RADIATION AT THE SBXMENT SURFACE CONTROL AREA MTROGOJ BRCHMS'T AREA 360- 360-

330- 330

300- 300-

no- 270-

240- 240- PIO- L A■ CL ISO^ 180- Te

S I BOH M^nroi>r 1 BO-

120- 120-

90-

60- 60^

30 - 30-

0 lUTJUHJXAUSSEPOCTIKJVOeCJUFEaHUAPS 0 RAT J\JR JUL AUC SEP OCT NOT DEC JAN FEB RAX APR NORTHS. HAT 1978 - APRIL 1979 FOITAS, RAT 1978 - APRIL 1979 LEGENOi DISTANCE «-«-> 25 * « « 82.5 m-m-m 100 I A I 151

Figure 8 . Maximum (A) and minimum (B) water temperatures (°C) at 25m, 62.5m, and 100m £rom shore versus time. Line at 32 C in (A) refers to the upper limit of net productivity with temperature, and the line at 10°C in (B) refers to the lower limit of net productivity with temperature. MAXIMUM TEMPERATURE VERSUS TIME MNMUM TEMPERATURE VB1SUS TIME 30.0:

32.6-

*0.0-

22.6- 27.6 -

2 5 .0- 20.0- OaO«GBtd«

2 2 .6- 17.6-

2 0 .0- 15.0: 17.6-

12.6- 10.0

10.0- 7.6-

7.6- 6.0

2.S: 2.6-

0.0 0.0- o uijiajawsffoamrDecjMPaiuitn 0 HATJUNJULAUGSEPOCTNOVOECJANFEBHARAPR IUT, 1970 - APRIL 1979 RAT. >970 - APRIL 1979 LEGENDi DISTANCE ■» . «■ 25 » »-• 62.6 » » » 100 LEGENOi DISTANCE 25 « « » 62.5 100 I A 1 152 153

Figure 9. Seedlings of Vallisneria americana emerging from fruit: (top) seedling emerging from broken portion of the fruit; (bottom) seedling emerging from the broken end of a fruit. 154 155

Figure 10. Clumps of seedling still held together by the remnants of a decaying fruits (top) decaying fruit plus seedlings; (bottom) small clump of seedlings held together in a gelatinous mass with only small portion of fruit remaining. 156 CHAPTER 5

SIMULATED NITROGEN DYNAMICS DURING GROWTH AND DECOMPOSITION OF A SUBMERGED VALLISNERIA AMERICANA GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA

ABSTRACT A model of nitrogen dynamics was developed as part of a study of growth and decomposition of

Vallisneria americana in Lake Pontchartrain,

Louisiana. A conceptual model was used to help design

and plan field research. The data from the field

study were then used to initialize, calibrate and run

.the nitrogen simulation model of growth (GROW).

Specifically, GROW used experimental results from a

sampling station 50m from shore in the control area,

and a high correlation (>90%) was found between

experimental and modeled leaf and below ground biomass nitrogen changes over the year. A decay model (DECAY) used the output (litterfall) of GROW to predict

cycling of available nitrogen. GROW was validated using initial conditions from a fertilized grass bed

50m from shore tested concurrently with the control.

The simulated leaf and below ground biomasses were correlated (>80%) with experimental results.

After validation, the relative effects of

sunlight, water depth, secchi depth, circulation,

temperature and nutrients were tested. GROW was most

sensitive to changes in sunlight (or, inversely, to

157 158 turbidity). When simulated light reaching the plants diminished by 25%, total leaf nitrogen declined over the year while below ground plant nitrogen only reached 14% of the control value. When 50% sunlight was simulated, the grass beds went into a steady decline.

The decomposition model (DECAY) predicted water organic and inorganic nitrogen concentrations with greater than 90% accuracy when driven by grass bed litterfall and outside sources of organic matter.

However, annual changes in the sediment organic and inorganic nitrogen storages were not accurately predicted. Although Vallisneria litterfall and root and rhizome death were included in the model, other factors such as resurfacing of buried peat layers, carbon influx during storms, spring high water upland runoff, and spillway openings were not considered in the model. Further field work is needed to quantify these factors as well as to quantify the processes of nitrification and denitrification. 159

5.0 INTRODUCTION

A project to model the nitrogen dynamics of

Vallisneria americana grass beds in Lake

Pontchartrain was undertaken as part of a field study

in order to better understand how the grass beds

function, as well as to answer questions concerning

how perturbations would influence them. Submerged

grass beds of Vallisneria atnericana historically

occupied the shallow, sandy shorelines of Lake

Pontchartrain, La. (Saucier, 1963) and supported an

extensive fishery (Thompson and Verret, 1980).

The three main objectives of this modeling

exercise were 1) to develop a conceptual model as a

basis for the experimental design of field work to

generate data needed for a simulation model, 2) to run a simulation model and to compare simulated with

experimental results, and 3) once the simulation model was validated, to use it as a tool to predict the

grass bed system's response to specific alterations.

For example, if increasing turbidity (due to eutrophication or shell dredging) decreases light availability, how will the grass bed respond?

5.1 DESCRIPTION OF TIIE STUDY AREA AND GRASSBED ECOSYSTEM

The most extensive grass beds in Lake

Pontchartrain occur along the northern shore from 160

Mandeville eastward, the most luxurious growth occurring near Goose Point (see Figure 1). The accompanying field study (Chapters 3 and 4) was conducted in the grass beds west of Goose Point from

May, 1978 to May, 1979. Vallisneria grew from near shore out to about 150m from shore in waters up to about 2 meters deep. Salinities ranged from near 0 ppt to 4 ppt and temperature from about 10°C to

30°C during the year. Currents were generally 0 to 10 cm/sec and alongshore from the west. There was generally little wave action except during winter cold fronts (T. Gayle, personal communication, Center for

Wetland Resources, L.S.U.). The portion of the grass beds chosen for modeling was 50m from shore in a zone of high productivity (see Chapters 3 and 4). The site was bordered by a very gradually sloping beach, then a sand dune and a brackish marsh on the shore. The grass bed sediments were silty sand underlain by peat.

5.2 MODEL DEVELOPMENT

The nitrogen modeling effort was carried out in nine steps: 1)definition of system boundaries;

2)construction of a conceptual model; 3)development of simulation models of growth and decomposition (GROW and DECAY); 4)initializing the models with values determined from control plot field data; 5)calibration 161

of the models by adjusting coefficients to produce the

best fit with one year of experimental data from 50m

from shore at the control area yrass bed; 6)running

GROW and DECAY; 7)sensitivity analysis of the models which determined which parameters caused the major perturbations in the models; 8)validation of the

(GROW) model using initial conditions 50m from shore at the nitrogen fertilized area, to compare simulation

results with field samples; and, finally, 9)prediction of the control grass bed's responses to changes in

light intensities.

5.2.1 Model Boundaries

An important first step in modeling is to define the boundaries of the system in question. In this case the system modeled included above and below ground Vallisneria biomass in a square meter, the sediment down to the bottom of the root zone, and the water column and atmosphere above them. The model was run with parameters set for conditions at 50m from shore in the control area grass bed (Figure 1). This was in the zone of maximum production and the only depth where significant plant tissue and growth responses to nitrogen fertilization occurred (Chapter

3) and where, therefore, growth under different nitrogen regimes could be compared. The root zone at

50m from shore extended down to about 25cm below the 162 surface and the water depths ranged from 0.34 to 0.65m over the year.

5.2.2 Construction of The Conceptual Model

Nitrogen cycling in the Vallisneria grass bed was modeled because nitrogen is generally thought to be the macronutrient which most often limits production

(Patriquin and Knowles, 1972), nitrogen turns over quickly in submerged grass beds in comparison with other macronutrients (McRoy and McMillan, 1977), and nitrogen may be in abundant or short supply depending upon time of the year (Chapter 4). The conceptual model of nitrogen cycling in the Vallisneria americana bed is shown in Figure 2. The grass bed ecosystem was conceptually divided into three major functional compartments: 1) outside forcing functions;

2) growth (GROW); and 3) decomposition (DECAY). These major compartments were organized and diagrammed with the modeling language developed for ecosystem analysis by fi.T. Odum (1971, see Fig. 3).

5.2.2.1 Forcing Functions:

A number of material and energy flows (forcing functions) cross the system boundaries and change nitrogen or affect its cycling within the system

(Figure 2, top). Four forcing functions affect plant nitrogen assimilation and growth in the model: sunlight, water and secchi depth, temperature, and circulation (Westlake, 1967; and Salisbury and Ross,

1969). Sunlight reaching the plant affects the

photosynthetic, metabolic and mineral assimilation

rates. However, incident radiation is attenuated with

increasing water depth and turbidity. Secchi depth is

proportional to light penetration through the water

and inversely related to light attenuation. The

combined effect of water depth and secchi depth is

indicated in the conceptual model by a division sign.

Temperature affects the rate of chemical reactions and

the rate of plant metabolism and may limit plant

growth above or below critical limits. Waves, tides,

and currents are included in a general catagory called

circulation which is a measure of integrated water

flow over the area per unit time. Water circulation

breaks up the stagnant boundary layer at the plant

surface and increases the supply of nutrients passing

the plant surface (Belyaev, ej: al ., 1977).

Six forcing functions are related to plant decay

and associated nitrogen cycling. Circulation affects

the movement of allocthanous organic and inorganic

nitrogen supplies across the boundaries of the grass bed system, and influences the gaseous exchanges between the atmosphere and water or sediment.

Temperature increases the rate of decomposer metabolism and chemical reactions of decomposition. 164

Precipitation is an independent external source of

inorganic nitrogen.

5.2.2.2 Growth:

The growth compartment of the conceptual model

(GROW) simulates nitrogen dynamics during plant

growth. Spatial relationships are indicated by the > . separation of above and below ground compartments.

For example, it is assumed that the sources of

available nitrogen are water column inorganic nitrogen

(WIN) for the leaves (L) and sediment inorganic

nitrogen (SIN) for the roots and rhizomes (RR).

Nitrogen assimilation (ASSIM) is increased by

sunlight, temperature, and circulation, and decreased

by water depth and turbidity (the inverse of secchi

depth). Nitrogen assimilated by the plants is also

proportional to the plant biomass present (L and RR).

Once incorporated into the plant, nitrogen moves by

translocation (TRANS) or by xylem flow (XYFLO) from

one part of the plant to another. When the plant dies

(litterfall and death) the organic nitrogen in dead plant material becomes part of the organic nitrogen

storages in the water (WON) or sediments (SON).

5.2.2.3 Decay:

The decay compartment of the conceptual model

(DECAY) simulates nitrogen dynamics of organic matter decomposition. Nitrogen in leaves (L) and roots and rhizomes (RR) enters the organic nitrogen pools in the

water column (WON) or sediment (SON) when the plants

die (litterfall and death). Leaves consequently

settle and become part of the sediment organic

nitrogen storage (SON). Waves# tides, and currents

(circulation) bring organic nitrogen to the grass bed

from outside the system boundaries— e.g., from the

adjacent marsh. Nitrogen fixation (NF) may increase

water column or sediment organic nitrogen

concentrations. However, extremely low nitrogen

fixation rates were measured during the field

experiment (Chapter 3) and, therefore, nitrogen

fixation was not included in the simulation model.

Organic material is quickly colonized by decomposer

organisms which break it into smaller pieces which are

more amenable to bacterial attack. Thus, the organic

nitrogen pool includes the micro-organisms actively

using the detritus as an energy source. Many of these

decomposer organisms require inorganic nitrogen which

is supplied from water (WIN) or sediment (SIN)

inorganic nitrogen storages and immobilized (IM).

While immobilization converts inorganic nitrogen into

organic nitrogen, mineralization degrades organic nitrogen into inorganic forms. The rate of mineralization (MIN) is dependent upon- temperature, oxygen, and the biomass of decomposer organisms. 166

Allocthanous inorganic nitrogen is brought across the boundaries of the system by circulation.

Precipitation increases inorganic nitrogen in the water (WIN). Inorganic nitrogen in the water column is assimilated (ASSIM) by leaves and in the sediment is assimilated by roots, to complete the nitrogen pathways within the grass bed system.

5.2.3 Construction of The Simulation Models

5.2.3.1 GROW:

The growth model (GROW) simulates nitrogen pathways during growth (Figure 4) and was developed using the growth compartment of the conceptual model

(Figure 2). The model has two storage compartments or state variables; leaves above ground (L) and roots and rhizomes below ground (RR). The complete equation for leaf nitrogen storage is given in Table 1 and can be simplified as:

(Equation 1)

dL/dt = ASSIMl - TRANS + XYFLO - LITTERFALL dL/dt : the rate of change of leaf nitrogen (gN/m2 )/day per square meter (L) with respect to time (t)

ASSIM^: leaf assimilation of nitrogen "

TRANS : translocation of nitrogen from leaves " to below ground parts of the plant

XYFLO : xylem flow of nitrogen from below " ground parts of the plant to the leaves 167

LITTERFALL : dead leaves falling from the " plant The complete equation for root and rhizome nitrogen storage is given in Table 1 and can be simplified as:

(Equation 2) dRR/dt = ASSIMr r + TRANS - XYFLO - DEATH 2 dRR/dt : the rate of change of root and (gN/m )/day rhizome nitrogen per square meter (RR) with respect to time (t)

ASSIMRR : root assimilation of nitrogen "

TRANS : translocation of leaf nitrogen (L) to " below ground parts (RR) of the plant

XYFLO : xylem flow from the below ground parts " of the plant (RR) to the leaves (L)

DEATH : dead roots and rhizomes falling from " the plant

Sequentially, the leaf nitrogen (L) and root and rhizome nitrogen (RR) storage equations include assimilation of nitrogen as it is affected by outside forcing functions, translocation and xylem flow moving nitrogen within the plant, and litterfall and death releasing nitrogen.

Assimilation: The nitrogen assimilation rate is controlled by outside forcing functions and the amount of plant material available to absorb the incoming nutrients. In Vallisneria americana , as in other seagrasses, nitrogen may be taken up across leaf surfaces or root surfaces (Salisbury and Ross, 1969;

MeRoy and 3arsdate, 1970). Therefore, GROW separates storages of above ground leaves (L) and below ground 168 root and rhizoues (RR).

The assimilation across the leaf surfaces

(ASSIMj^) is proportional to the concentration of available water inorganic nitrogen (WIN). However, physical forcing functions such as light, depth, and temperature are also important factors controlling assimilation rate. Sunlight, water depth, and secchi depth are related in the equations

I=Ioe“*1,7/SE>)X(Z) (Equation 3)

I = sunlight reaching the bottom (langleys)

I = sunlight incident to the water surface

SD = secchi depth (meters)

Z = v/ater depth "

This equation calculates the solar radiation reaching the sediment surface, a conservative estimate of sunlight reaching the leaves (Figure 4).

Assimilation exponentially increases up to a light compensation point where growth becomes linear until light reaches a maximum level above which uptake declines (Barko and Smart, 1981). Temperature affects the assimilation rate of Vallisneria . Assimilation is very low below 10°C, linearly increases with temperature in the midrange, and ceases above a thermal maximum around 32°C (Barko, et al .,

1982). The assimilation rate of Vallisneria is related to current speed and increases exponentially 169 from low water flow up to a velocity of about

0.5cm/sec (Westlake, 1967) and then linearly increases until at high velocities growth ceases. The final factor affecting assimilation is the amount of leaf surface area available for nutrient exchange.

Nutrient uptake is generally proportional to thallus surface area or biomass (L) (Belyaev, et al ., 1977).

Nitrogen assimilation across root surfaces

(ASSIMRR) is dependent upon available sediment inorganic nitrogen (SIN). The mathamatical relationship of the assimilation of SIN is a limiting nutrient curve approximated by Michaelis-Menton nutrient uptake dynamics. Physical factors affecting root and rhizome (RR) nitrogen assimilation were assumed to be the same as for the leaf surface, except that circulation was not included. The root and rhizome nitrogen storage (RR) was used to approximate the root surface area in the model which is proportional to assimilation rate.

Transport within the Plant: During photosynthesis nitrogen is converted to organic compounds and translocated (TRANS) to the roots, rhizomes, and other shoots on the same rhizome. The rate of translocation is dependent upon the rate of metabolism, age of the leaves, temperature and light 170

(Salisbury and Ross, 1969). The highest rates are from mature, non-flowering leaves (Harrison, 1978).

The majority of photosynthate is translocated to flowering shoots and young vegetative shoots during spring and summer, and, thus, remains in the above-ground leaf storage (L). Translocation during the late fall and winter, when mainly mature or senescent leaves are present, goes to the below-ground rhizome nitrogen storage (RR), since mature vegatative leaves receive little photosynthate (Harrison, 1978).

The xylem stream mobilizes assimilated nitrogen from the roots and rhizomes to the leaves (XYFLO). Xylem flow is dependent upon the nutrient uptake rate and gradient between the concentrations in the roots and leaves (Salisbury and Ross, 1969).

Litterfall and Root Death: Litterfall is proportional to instantaneous live leaf standing stock

(Chapter 3). The high leaf turnover rate results in higher average leaf productivity because leaf photosynthesis and metabolism decreases with leaf age

(Ikusima, 1965) and light and carbon exchanges across the leaf surface decrease with epiphytism which increases with age (Sand-Jensen, 1977).

Root and rhizome death is also proportional to the standing stock available. However, root and rhizome death or turnover rate is generally lower than 171 leaf turnover rate (tlcRoy and McMillan, 1977) since below ground parts are not subject to the extremes of temperature, currents, and faunal attack that the leaves experience in the water column.

5.2.3.2 DECAY:

The decay model (DECAY) simulates nitrogen pathways in the grass bed during decomposition (Figure

5). The decomposition compartment of the conceptual model (Figure 3) was the basis for the DECAY simulation model (Figure 5). The model has four storage compartments or state variables: water organic nitrogen ('WON), water inorganic nitrogen (WIN), sediment organic nitrogen (SON), and sediment inorganic nitrogen (SIN). The complete equation for water organic nitrogen is given in Table 1 and can be simplified as:

(Equation 4)

dWON/dt = LITTERFALL + AON + IM - MIN - BURIAL dWON/dt : rate of change of water column (gN/m )/day organic nitrogen with respect to time (t)

LITTERFALL : abscission of dead leaves from " plant adding to organic nitrogen in water column (WON)

AON : allocthanous organic nitrogen entering the " square meter boundary from surrounding water

IM : immobilization of inorganic nitrogen to " organic nitrogen in decomposer organisms 172

MIN : mineralization of organic nitrogen to " inorganic nitrogen

BURIAL ; settling of organic nitrogen in the " water column to organic nitrogen in the sediment (SON)

The complete equation for water inorganic nitrogen

(WIN) is given in Table 1 and can be simplified as:

(Equation 5)

dWIN/dt = MIN + AIN - IM - ASSIM 2 dWIN/dt : rate of change of water column (gN/m )/day inorganic nitrogen with respect to time (t)

MIN : mineralization of organic nitrogen to " inorganic nitrogen

AIN : allocthanous inorganic nitrogen entering " the square meter boundary from the surrounding water

IM : immobilization of inorganic nitrogen to " organic nitrogen in decomposer organisms

ASSIMl : assimilation of inorganic nitrogen " by grass bed leaves in the water column

The complete equation for sediment organic nitrogen

(SON) is given in Table 1 and can be simplified as:

(Equation 6)

.dSON/dt = LITTERFALL + DEATH + BURIAL + IM - MIN o dSON/dt : rate of change of sediment organic (gN/m^)/day nitrogen with respect to time (t)

LITTERFALL : abscission of dead leaves from plant " adding to organic nitrogen in the sediment organic nitrogen (SON)

DEATH : dead roots and rhizomes (RR) falling from " the plant and adding organic nitrogen to the sediment organic nitrogen (SON) 173

BURIAL : incorporation of organic nitrogen from the water column (WON) into the sediment organic nitrogen (SON)

IM s immobilization of inorganic nitrogen to organic nitrogen by decomposer organisms

MIN : mineralization of organic nitrogen to inorganic nitrogen

The complete equation for sediment inorganic nitrogen

(SIN) is given in Table 1 and can be simplified as:

(Equation 7)

dSIN/dt = MIN - IM - ASSIM 2 dSIN/dt : rate of change of sediment (gN/m )/day inorganic nitrogen (SIN) with respect to time (t)

MIN : mineralization of organic nitrogen to " inorganic nitrogen

IM : immobilization of. inorganic nitrogen to " organic nitrogen in decomposer organisms

ASSIMRR : assimilation of inorganic nitrogen " by grass bed roots in the sediment

Sequentially, the water organic nitrogen (WON) and sediment organic nitrogen (SON) storage equations include litterfall and death introducing plant organic nitrogen; exchanges with outside sources of organic nitrogen or allocthanous organic nitrogen (AON) and burial of water column organic nitrogen into upper sediment layers; inorganic nitrogen immobilization or use by decomposer organisms; and mineralization of organic nitrogen to inorganic nitrogen. The water inorganic nitrogen (WIN) and sediment inorganic 174 nitrogen (SIN) equations include mineralization of organic nitrogen to inorganic nitrogen, loss of inorganic nitrogen used by decomposer organisms (IM), introduction of allocthanous inorganic nitrogen from outside sources, exchanges of inorganic nitrogen between the water column and sediment, and uptake or assimilation of available inorganic nitrogen by macrophytic plants.

Litterfall and Death: Dead leaves were assumed to contribute organic nitrogen (LITTERFALL) to both the water column and sediment organic nitrogen storages (WON and SON). It was found in litterfall experiments at the study site that litter traps contained dead leaves both floating in the water column and lying within the top few millimeters of the sediment. Therefore, litterfall contributed to both the water column and sediment organic nitrogen storages in equal proportion. Litterfall was also proportional to the standing crop of live leaves (L)

(Chapter 3). Dead roots and rhizomes only contributed organic nitrogen (DEATH) to the sediment organic nitrogen storage (SON). The DEATH rate was proportional to the organic nitrogen present in roots and rhizomes (RR), but was assumed to be somewhat lower than the leaf litterfall rate reported in a review of production (McRoy and McMillan, 1977). 175

Allocthanous Organic and Inorganic Nitrogen (AON and AIN)s Allocthanous or outside sources of organic and inorganic nitrogen included runoff from upland swamps and marsh areas, water from other areas in the lake, rainfall, and pollution.

Immobilization (IM): Inorganic nitrogen is assimilated by micro-organisms and metabolized into nitrogenous constituents in their cells via a process known as immobilization. In a review article on seagrass detritus-decomposition relationships, Klug

(1980) stated that the most likely source of inorganic nitrogen is from surrounding water in the water column, or from interstitial water. In the model immobilization depleted the supply of inorganic nitrogen in the water column and sediment (WIN and

SIN) and was dependent upon the mass of decomposers which was a function of organic detritus and estimated in the model by the size of the organic nitrogen compartments (WON and SON).

Mineralization (MIN): Organic nitrogen is transformed into inorganic nitrogen via mineralization. This process is dependent upon the supply of organic material (WON and SON) which is available in the seagrass system from litterfall and root death. Through experiments on the temperate seagrass Zostera , Godshalk and Wetzel (1973) found 176

that the relative decay rate of seagrasses (MIN) was

also proportional to temperature (T), dissolved oxygen

(WOX or SOX), and inorganic nutrients available for

immobilisation. Immobilization of inorganic nitrogen

increases micro-organism biomass which increases

mineralization and, thus, the two processes tend to

counteract each other (Brannon, 1973).

5.2.4 Initializing GROW and DECAY

Initial conditions (IC) are shown in the GROW and

DECAY models (Figures 4 and 5). Values, references and calculations using data compiled during the field experiments are shown in Table 3.

5.2.4.1 GROW Initial leaf nitrogen storage (L) in May, 1978, 2 was 0.456 gN/m and initial root and rhizome nitrogen storage (RR) in May, 1978, was 0.258 gN/m . Plant nitrogen was estimated without

including zero biomass values (as in Chapter 3) because the object was to model the response of grass

to environmental parameters, not to model the patchiness of the grassbed.

5.2.4.2 DECAY

Initial conditions in May, 1978, for the water

inorganic nitrogen storage (WIN), water organic nitrogen storage (WON), sediment inorganic nitrogen 177

storage (SIN), and sediment organic nitrogen storage

(SON) were calculated using data from the field

experiment done for this study or from a companion

study done on phytoplankton production in the

Goosepoint grassbed during the same time period by

Miller (19S0). The water column available inorganic 2 nitrogen (WIN) was 0.060 gN/m and the water 2 column organic nitrogen (WON) was 2.980 gN/m .

Sediment available inorganic nitrogen in the root zone

(SIN) was 2.900 gN/m , calculated by multiplying

the nitrogen concentration by the bulk density of the

sediment. Sediment organic nitrogen was similarly

calculated and had a value of 190.000 gN/in for

the root layer.

5.2.5 Calibrating GROW and DECAY

Calibration of the models involved calculating

the numerical coefficients (k values) of flows as

shown in Figures 4 and 5 and Table 4. Coefficients were determined by one of four methods: 1)solving for

the only unknown (k) in an expression for a known

flow; 2)using values from the literature or from this

field experiment; 3)optimization by trial and error where no literature or field values were available;

and 4)dimensional analysis (combining known

relationships of dimensioned parameters to obtain an

expression for an unknown parameter). Where unlike 178

units were combined in a single flow, it was necessary

to calculate a coefficient which would normalize all

values to grams of nitrogen per square meter per day

(Hall and Day, 1977). Details of the calculation of k

values are given in Table 4.

5.2.5.1 GROW

GROW was first run as a separate model by itself with no coupling with the DECAY model. Therefore all

values for coefficients were derived independently for

GROW.

Assimilation: The coefficient of leaf

assimilation (k^) was calculated using the first

method by solving the assimilation expression (Table

1, Equation 8) for kj^ given the average value for

nitrogen assimilation per day in May and the values of

the variables (Table 2) for May. The value for k^ was 6.0E-07.

Root and rhizome assimilation was also calculated using the first method. Plant production versus available sediment inorganic nitrogen (SIN) followed a

limiting nutrient Michaelis-Menton form (S-shaped curve levelling off and parallelling the x-axis) rather than being linearly proportional to concentration (shown in Figure 4 and enlarged in

Figure 6). The Michaelis constants were calculated by estimating where production leveled off (the SIN 179 concentration at which no further increases in production occurred as sediment nitrogen increased =

VMAX^ anc* the concentration by two to get To calculate 1 the root and rhizome assimilation expression (Table 1, equation

9) was solved for using the daily root and rhizome assimilation rate and the May average of all the factors. The value of ] < 2 was 1.5E-05 (Table

4).

Translocation: The coefficient of translocation

(k^) was calculated using methods two and three.

The rate of translocation was assumed to be proportional to the nitrogen present in the leaves

(Salisbury and Ross, 1969). Decause no literature values for nitrogen translocation rates were available and translocation was not directly measured in the field experiment, a value for the coefficient was arrived at by using growth data (Table 4) and was calculated to be 0.03. It was assumed that translocation would increase during the period of leaf senescence in the winter when rhizome biomass reached its relative peak (Figure 3, Chapter 3). The winter translocation coefficient (lc^) was increased from

0.03 by 0.005 intervals until the modeled leaf and root and rhizome nitrogen storages (L and RR) most closely fit the true values. The winter k^ was 180 set at 0.07. A switch changed the values from growing season to winter values on day 130 starting from May 1 since temperature is assumed to affect senescence and the magnitude of translocation and temperatures first dropped appreciably in December.

Xylem Flow: Values for the coefficient of xylem flow (k^) were calculated using method two and were proportional to the root and rhizome nitrogen present. Xylem flow or mobilization can be indirectly measured by the accumulation of leaf nutrients over time (Barko and Smart, 1980) and is higher during the growing season than during winter when production is low. Barko and Smart (1980) measured a xylem flow coefficient of approximately 0.01 for phosphorus in

Myriophyllum , a plant with proportionally less root biomass than Vallisneria . Using their technique, the xylem flow coefficient was calculated from the field experiment (Chapter 3) to range from 0.1 at its peak to near zero at its low. The values for k^ used in the model were average values for the growing season and the winter, and were set to 0.07 and 0.03, respectively. A switch changed the values from growing season to winter values on day 180 starting from May 1, when temperatures first dropped appreciably.

Litterfall and Boot and Rhizome Death: The coefficients of litterfall (kg) and death (kg)

were calculated using literature values and

optimization. Leaves were assumed to break off every

fifteen days (Zieman, 1968 and Patriquin, 1973).

Grazing is included in the calculation of the

litterfall coefficient (kg). Winter storms

increased litterfall and were modeled by increasing

the rate of litterfall for one day in February to half

the standing stock to simulate a violent storm. V/hen

the sites were sampled in late February, 1979,

extensive shoreline erosion suggested that a violent

storm had passed since the previous sampling trip in

early February, 1979. The coefficient of root and

rhizome death (kg) was calculated on the

assumption of a turn over rate of once a growing

season, or once every 200 days.

5.2.5.2 DECAY

Coefficients were derived independently for

DECAY.

Allocthanous Organic and Inorganic Nitrogen: The coefficients for allocthanous organic and inorganic nitrogen input (k^ and k^g) were calculated using literature and field experiment values. During the field study pulses of nitrogen were introduced at the nitrogen enriched grassbed 1/2 km west of the control grassbed (Figure 1). These nitrogen additions 182

resulted in elevated nitrogen concentrations in the water column. Thirty days later, the water nitrogen

concentration was not significantly different (p>0.05)

from the control site. It was assumed that the nitrogen concentrations in the water column simulated

from conditions at the study site would only be affected by nitrogen concentrations of outside sources of organic or inorganic nitrogen for thirty days (a conservative estimate of residence time).

Allocthanous organic and inorganic nitrogen concentrations (AON and AIN) were compared with the simulated grassbed water column organic or inorganic nitrogen storage (WON and WIN). The higher concentrations (AON or WON; AIN or WIN) were multiplied by 1/30 and added to the lower organic or inorganic nitrogen storages until their differences were zero. The values of ky and were both, therefore, 0.033. When WON equalled AON and when WIN equalled AIN, the flows of nitrogen from outside ceased until allocthanous sources were different than study area concentrations.

Burial: The coefficient of burial (kg) of the nitrogen in dead leaves was calculated using field data. Burial was assumed to follow first order kinetics and was proportional to the organic nitrogen in the water column. Because new leaves were observed 133 to fall and rest on or slightly below the sediment surface (settling or burial) after seven days, the coefficient was set at 0.07.

Mineralization: The coefficients of mineralization of nitrogen in the water column

(kg) and in the sediment ( ^ q ) were solved by setting mineralization equal to the decomposition rate measured in the field study (Chapter 3). The mineralization rate was proportional to the temperature and oxygen present. Given the values for mineralization rate, temperature, and oxygen, the expression (Table 1, equation 10) was solved for the only missing value. The coefficient of mineralization in the water column (kg) was 5.08E-05. The coefficient of mineralization in the sediment

(k^Q ) was set by trial and error.

Immobilization: The coefficients of immobilization of inorganic nitrogen (k^^ and kij) were calculated using a dimensional approach to convert grams organic nitrogen of microbes in the detritus to grams inorganic nitrogen used by the microbes during the decay process. The coefficients of immobilization (k^ and k ^ ) were calcuated to be 0.001. The rates of immobilization, proportional to the amount of decomposing material present (WON and SON) caused a flow from the storages 184 of water and sediment inorganic nitrogen (WIN and SIN) to the organic nitrogen storages (WON and SON) to be used by the bacteria. These flows from the inorganic nitrogen storages v/ere not dependent upon the value of the stored inorganic nitrogen.

Once the values for the forcing functions, initial conditions, and flow coefficients were calculated, GROW and DECAY were run using the

Continuous System Modeling Program (CSMP) on an ISM model 360 digital computer. Data for forcing functions (sunlight, water depth, and secchi depth) were entered daily using a Fortran subroutine or monthly (all other forcing functions) and were interpolated by the CMSP function generator which provided a continuous input of forcing function values to GROW -and DECAY. Simulations were run for 365 days beginning in May, 1978, and ending in April, 1979.

The equations were solved using the rectangular method of integration, a CSMP option. Internal switches were operated using CSMP logic functions.

5.3 RESULTS AND DISCUSSION

The results of GRO'W and DECAY indicate that they fairly realistically simulate many, but not all, of the nitrogen components of the natural submerged grass bed ecosystem. While GROW accurately predicts leaf and root and rhizome nitrogen storages, DECAY does not 185 predict nitrogen cycling in the water column and sediment with as much accuracy. Often, the discrepancies suggest further field or laboratory experiments. Possible reasons for noted discrepancies are given in the following discussion.

5.3.1 GROVJ: Leaf and Root and Rhizome Nitrogen

Simulation

Simulated and measured leaf nitrogen were very closely correlated (r=0.95), as were simulated and measured root and rhizome nitrogen (r=0.98) (Figures 7 and 8). Simulated leaf nitrogen did not rise as quickly as measured nitrogen in August. The secchi depths were much greater in August than in June or

July (Chapter 4), and perhaps the model was not responsive enough to changes in light at the bottom.

Maximum temperatures also started to drop below a critically high level in August (Chapter 4), but maximum temperatures were not considered in the model as a negative influence on growth as Short (1977) suggested. Zieman (1975) found that Thalassia productivity was greatly depressed during times when temperatures were high (31.5°C-35.5°C), the same high temperature range encountered in this experiment (Chapter 4). The simulated leaf nitrogen values did not drop as low as the measured leaf 186 nitrogen values during late winter. Two probable reasons for the rapid loss of leaves are storm erosion and freezing temperatures. In his model of Zostera production, Short (1977) modeled loss of leaf carbon as a function of respiration and storm erosion. The quantification of the effects of winter storms is needed.

Low temperatures affect both productivity and standing crop. Temperatures below the 10° productivity minimum were encountered during January and February (Chapter 4). Low temperatures are particularly devastating to standing stock levels when they occur simultaneously with very low water levels

(also encountered in late winter, Chapter 4) or subaereal exposure (Zieman, 1975). Only daily average temperatures were used as forcing functions in the model, which did not include values above or below the temperatures critical for plant production (Chapter

4). Inclusion of maximum and minimum temperatures as well as average temperatures and a more precise estimation of storm erosion in the leaf nitrogen equation would probably make the model more responsive to seasonal cycles. Simulated root and rhizome nitrogen were very close to the measured nitrogen values all year— a result suggesting that fine tuning the forcing functions for below ground parts would 187

offer no better solution than the present model.

5.3.2 DECAY: Inorganic and Organic Water and Sediment

Nitrogen Simulations

Results of simulated seasonal fluctuations in

water organic nitrogen (WON) and water inorganic

nitrogen (WIN) agree well with field data (Figures 9

and 10, r=0.9o, p=0.0001 for WON, and r=0.96, p=0.0001

for WIN). The simulated water organic nitrogen did

not rise as quickly as measured values during the

summer (Figure 9). This could have been due to more

rapid exchange of dissolved and particulate organic

nitrogen between the outside sources and the grass bed

than estimated in the model or to the lack of a

phytoplankton nitrogen storage compartment in the model. Swenson (1980) estimated that flushing or

exchange of water in Lake Pontchartrain occurred more

rapidly during high water flow events. No attempt was

made in the model to increase the exchange rate in

response to v/eather conditions. Miller (1930)

measured an increase in chlorophyll a concentrations

indicative of phytoplankton biomass in the Vallisneria

grass bed from June through August of 1978. However,

summer was the only season when an increase in phytoplankton biomass coincided with an increase in water column organic nitrogen.

Differences between measured and simulated 188 organic nitrogen values in the water column in the winter may be an indication of leaf litter efflux from the grass beds. In December and January, the simulated organic nitrogen levels were higher than outside sources and the flow coefficient of exchange

(k^) became negative— a result indicating an efflux of organic nitrogen from the grass bed to surrounding waters. There was an increase in litterfall at that time (Chapter 3) which put large amounts of organic nitrogen in the form of leaves in the water column. Water samples taken in the field did not include fallen leaves, although dead leaves were simulated as part of the organic nitrogen storage in the water column in the model.

A large pulse of organic and inorganic nitrogen in the water column and the sediment occurred in April of 1979. The Bonnet Carre Spillway was opened in

April, 1979, (Army Corps of Engineers, 1980) introducing high amounts of nitrogen and other nutrients to Lake Pontchartrain. The simulated nitrogen levels did not rise as high as recorded nitrogen levels which rose dramatically during this pulse of incoming nitrogen. Thus the model is more responsive to seasonal biological cycles than to sudden, extreme perturbations.

Simulated water inorganic nitrogen (WIN) was very 189 similar to measured water inorganic nitrogen levels throughout the year except in April during the opening of the Bonnet Carre Spillway (Figure 10). Simulated

WIN did not reach the high values that the measured water inorganic nitrogen attained in April. The dramatic decrease of simulated water inorganic nitrogen in the model from April to May was probably due to plant assimilation. Simulated leaf nitrogen

(L) increased dramatically from April to May and caused assimilation to increase proportionally.

Again, DECAY predicts seasonal nitrogen cycles more accurately than sudden, extreme perturbations.

Simulated sediment organic nitrogen (SON) and sediment inorganic nitrogen (SIN) concentrations were weakly negatively correlated with measured sediment organic and inorganic nitrogen (Figures 11 and 12, r=-0.52, p=0.07 for SON and r=-0.53, p=0.08 for SIN).

Simulated sediment organic nitrogen (SON) generally stayed within the range of measured values which fluctuated above and below the simulated values

(Figure 11). Perhaps segregating the upper layer of sediment where the majority of the processes take place from the lower layers of sediment where organic nitrogen is stored but is relatively undisturbed would make the model more responsive to seasonal fluctuations. Quantifying the pulses of organic 190 detritus into the area from the contiguous marsh and spillway openings and incorporating them in the model as forcing functions would also increase the model's responsiveness to seasonal events.

Simulated sediment inorganic nitrogen (SIN) tracked measured values from May through January, but deviated thereafter (Figure 12). Measured increases may reflect the fact that certain inputs were not included in the model. These inputs would include runoff from the marshes and spillway openings in the spring. The simulated growth pulse of Vallisneria from ,\pril to May (which closely followed the measured growth pulse) caused simulated SIN to decrease due to plant assimilation. because measured SIN values did not decrease even when assimilation due to plant uptake increased, the Bonnet Carre Spillway opening may have provided sufficient nutrients for plant growth as well as increasing sediment inorganic nitrogen storages. Thus, the plants and sediment act as a filter for pulses of incoming nitrogen. Field measurements of sediment-water column nutrient fluxes need to be made in order to adequately model these p r o c e s s e s .

5.3.3 Sensitivity Analysis

Sensitivity analyses were done in order to determine the model's response to environmental 191

parameters and processes. Varying the values of the

forcing functions tested the model's response to

environmental parameters. In order to test the

sensitivity of the model to processes, the flow

coefficients were varied since they could either

increase or decrease the relative value of flows.

5.3.3.1 GROW Sensitivity Analysis

Leaf and root and rhizome nitrogen storages were

found to be most sensitive to the forcing function

sunlight. The effect of increasing each forcing

function individually by 10% (irradiance at the

sediment surface, temperature, circulation, water

column nitrogen, and sediment nitrogen) on the leaf

and root and rhizome nitrogen storages (L and RR) was

investigated. The results were compared to simulated

L and RR with original forcing functions using the

Statistical Analysis System (SAS) correlation procedure. The nearer the correlation coefficient of

the forcing function was to one, the less response L

or RR had to altering that forcing function. This

simple test revealed that L and RR were most sensitive

to changes in sunlight (r= 0.896 for L and 0.844 for

RR, Table 5, Figures 13 and 14)— a findng which agrees well with Sand-Jensen and Borum's (1983) findings that

75% of the variation in leaf growth rates can be

accounted for by changes in irradiance. Leaves and 192 roots and rhizomes were also very sensitive to temperature (r= 0.91 for L and 0.90 for RR, Table 5,

Figures 13 and 14). This result agrees with Zieman's

(1975) finding that seasonal changes in Thalassia grass beds' growth could be explained by seasonal temperature variations. Sediment nitrogen had an intermediate effect when 12 months were considered.

When the dramatic pulse of nitrogen in the spring was considered, sediment nitrogen was found to have a great effect on both root and rhizome and leaf nitrogen. Thus, sunlight and temperature were of major importance during a typical seasonal cycle, while sediment nitrogen became important during pulse events. Water column nitrogen and circulation altered the original L and RR values very little (Table 5,

Figures 13 and 14).

Leaf and root and rhizome nitrogen storages were most sensitive to changes in the flow coefficients of assimilation (k^ and k 2 )» The coefficients of the flows (k values) were altered in order to test the sensitivity of leaf and root and rhizome storages (L and UR) to changing the rates of processes. (A summary of L and RR responses to changing k's is in Table 6).

The leaf (L) and root and rhizome (RR) nitrogen storages in the GROW model were very sensitive to the rate of assimilation of nitrogen, as shown by the 193 model's response to changing the assimilation coefficients (k^ and k2). When ^ was increased by an order of magnitude (xlO), L increased 18 by 10 . When k^ and k2 were increased

(2x) simultaneously, both L and RR increased by

10 . These results demonstrated the sensitivity of the model to these assimilation coefficients.

5.3.3.2 DECAY Sensitivity Analysis

The DECAY sensitivity analysis showed that

Vallisneria leaf organic nitrogen and allocthanous nitrogen inputs are important in maintaining nitrogen levels in the water column, the rate of mineralization is important in maintaining nitrogen levels in the sediments, and increasing the rate of plant assimilation could deplete available inorganic nitrogen. DECAY flow coefficients (k values) and forcing functions were altered individually (Figure

5).

To test the sensitivity of DECAY water organic nitrogen (WON) and water inorganic nitrogen (WIN) storages to the forcing functions of allocthanous organic and inorganic nitrogen (AON and AIN), the coefficients of exchange (ky and k^2 ) were set to zero (Figure 15). Water organic nitrogen (WON) simulated with no allocthanous inputs had an annual cycle similar to the measured WON curve but with higher amplitude (Figure 15). The high organic nitrogen burst during the Bonnet Carre Spillway opening in April of 1979 was not reflected in the '.JON values simulated without outside inputs. The WON curve simulated without outside inputs (Figure 15) did not rise in April, 1979, but rose sharply in May,

1979, when the leaf nitrogen (L) curve (Figure 7) rose sharply and associated litterfall increased. This finding demonstrates that to adequately simulate water column organic nitrogen, both exchange with allocthanous inputs and locally produced vascular plant material must be included. These results show that the influence of outside sources of organic and inorganic nitrogen are important to maintain realistic values over the grass bed.

The coefficients of mineralization (k^ and k ^ ) were altered in order to test their effects upon water organic nitrogen (V70N), water inorganic nitrogen (WIN), sediment organic nitrogen (SON) and sediment inorganic nitrogen (SIN). The sediment nitrogen storages were much more sensitive to changes in the mineralization rate than the water column storages. Doubling the coefficient of mineralization in the sediment (h^g) caused SON to steadily 2 decrease (from 190 to 151 gN/m ) and SIN to increase (peak SIN increased from 5.7 to 24.9 195 2 gN/m ). Decreasing k^Q by one half caused SON to steadily increase (from 190 to 261 gN/m ) and

SIN to decrease. (SIN was less than zero from June through August, and from February until May.)

Sediment nitrogen compartments in the model were insensitive to both sediment oxygen and temperature until the sediment mineralization rate (k^) was lowered to bracket appropriate levels for SON and SIN.

The rate of mineralization under anaerobic conditions is much lower than under aerobic conditions (Klug,

1980). In a review article on decomposition in seagrass beds, Fenchel (1977) noted that the majority of seagrass decomposition occurs under anaerobic conditions. I originally assumed that simulating lower concentrations of oxygen in the sediment and keeping water column and sediment mineralization rates equal would sufficiently lower mineralization rates in the sediment to simulate measured concentrations of

SON and SIN. However, it was found that changing oxygen levels in the sediments altered the SON and SIN concentrations very slightly. After the relative rate of mineralization in the sediment was determined, SIN was sensitive to changes in both temperature and oxygen. Sediment temperatures set the same as the measured water column temperatures caused wide seasonal fluctuations in SIN and simulating oxygen 196

concentrations below 5 ppm in the sediments caused SIN

to drop to negative values during the spring (Figure

16). Therefore, a smoothed temperature function

(sediment temperature, Table 2) and sediment oxygen

levels which did not fall below 5 ppm were used as

forcing functions for sediment mineralization. Data

for both sediment temperature and oxygen levels are

needed. The sediment nitrogen compartments were not

responsive to seasonal fluctuations in Vallisneria

detritus loadings or to changes in immobilization and

mineralization. The initial value of the organic 2 nitrogen storage (190 gN/m ) was large in relation

to the values of the fluxes in and out of the storage.

The coefficients of assimilation (k^ above ground and k 2 below ground) were altered to test the sensitivity of water inorganic nitrogen (WIN) and sediment inorganic nitrogen (SIN) to plant uptake. It has already been shown that the burst of plant growth in the spring of 1979 coincided with a dramatic drop in simulated available inorganic nitrogen. When k^ and k 2 were simultaneously doubled, WIN and

SIN storages steadily decreased to near zero. Thus, simulating an increase over the entire year of the plant uptake by twofold would deplete available nitrogen supplies. 197

5.3.4 Validation of GROW

GROW was validated by running the model with

initial conditions and forcing functions for 50m from

shore at a nitrogen enriched grass bed on the north

shore of Lake Pontchartrain approximately one

kilometer to the northwest of the control grass bed.

Inorganic nitrogen was added monthly for a year to

fertilize the area (Chapter 3). Values for the

initial conditions, flow coefficients, and forcing

functions are shown in Table 6.

Simulated values for leaf nitrogen exceeded

experimental values, but the seasonal trends were very

similar (r=0.96, p=0.0001, Figure 17). Simulated

values for root and rhizome nitrogen were lower than

experimental values but were also closely correlated

(r=0.80, p=0.0056, Figure 13). GROW, run under the

conditions at the nitrogen addition area, partitioned

relatively more leaf nitrogen and less root and

rhizome nitrogen per square meter than did the actual

Vallisneria population. The effects of nitrogen

limitation stress on submergent aquatic plants (Denny,

1972), emergent salt marsh plants (Smart and Barko,

1978), and emergent freshwater plants (Barko and

Smart, 1979) is to relatively increase belowground

biomass. The GROW model could be modified to reflect

different available nitrogen conditions by altering 198

the the Michaelis constants and the translocation and

xylem flow interchanges between above and below ground parts in response to different levels of nitrogen

stress, llore field data under a range of nutrient

conditions is needed to fully understand allocation of

nitrogen within submergent plants.

5.3.5 Prediction of the Vallisneria Grassbed's

Response to Different Sunlight Conditions

The forcing function bottom sunlight was decreased to predict the Vallisneria grass bed's

response to increasing turbidity in Lake

Pontchartrain. Leaf and root and rhizome nitrogen storages were most sensitive to the forcing function bottom sunlight when, compared with temperature, circulation, water nitrogen and sediment nitrogen.

Because the grass bed is sensitive to changes in sunlight and shell dredging and industrial and urban development in Lake Pontchartrain increase turbidity, the grass bed's response to altered sunlight has both basic and practical management implications.

Therefore, the GROW model was run for 50m from shore at the control area with all initial conditions, flow coefficients, and forcing functions the same as listed in Tables 2,3, and 4 except for sunlight. Three simulations were run: 1)75% of normal sunlight, 2)50% of normal sunlight, and 3)25% of normal sunlight. 199

Results of seasonal fluctuations in Vallisneria

grass bed leaf and root and rhizome nitrogen are shown

in Figures 19 and 20. When bottom sunlight was

diminished by 25% (75% of normal sunlight), leaf

nitrogen reached a peak nitrogen of only 12% of the peak nitrogen at the control area under full sunlight,

and root and rhizome nitrogen reached a peak of only

14% of the control area nitrogen under full sunlight.

When bottom sunlight was diminished by 50% and 75%.

(50% and 25% of full sunlight), both leaf and root and rhizome nitrogen declined steadily over the year.

5.3.6 Conclusions

The following conclusions can be drawn from this simulation study of growth and decay of a Vallisneria grass bed.

1) Seasonal fluctuations in leaf and root and rhizome nitrogen simulated by GROW are very similar to field experimental values.

2) GROW and DECAY are more responsive to seasonal variations in forcing functions than to sudden, extreme events, such as the Bonnet Carre Spillway opening.

3) More data on the effects of wind induced erosion on the grass bed under winter storm conditions are needed.

4) While DECAY simulated water organic and 200 inorganic nitrogen storages were close to field values, sediment organic and inorganic nitrogen storages had very poor correlations with field values.

5) Further field data including nitrogen fixation, sediment-water column nitrogen fluxes, denitrification, allocthanous inputs, and mineralization rates within the Vallisneria grass bed are needed.

6) GROW was sensitive to the forcing functions, light at the bottom, temperature, sediment nitrogen, currents, and water nitrogen, in descending order of sensitivity.

7) GROW was validated using the same model and simulating seasonal leaf and root and rhizome nitrogen storages in a fertilized grass bed 50m from shore.

Both L and RR were closely correlated with field r e s u l t s .

8 ) The relative partitioning of nitrogen above and below ground under different nitrogen loadings needs to be studied.

9) If light at the bottom decreases due to increased turbidity, the Vallisneria grass bed may be drastically reduced. 201

5.4 REFERENCED CITED

Anderson, J. M. and A. Macfadyen, Eds., 1976. The Role of Terrestrial and Aquatic Organisms in Decomposition Processes . Blackwell Scientific Pub., London. 450pp.

Army Corps of Engineers, 1980. Daily discharges for 1979. Bonnet Carre' Floodway near New Orleans, La., computed daily discharge. Exhibit 5, New Orleans District, Engineering Division, New Orleans, La.

Barko, J. W., D. G. Hardin, and M. S. Matthews, 1982. Growth and morphology of submersed freshwater macrophytes in relation to light and temperature. Can. J. Bot., 60(6):877-887.

Barko, J. W. and R. M. Smart, 1979. The nutritional ecology of Cyperus esculentus , an emergent aquatic plant, grown on different sediments. Aquat. Bot., 6:13-28.

3arko, J. W. and R. M. Smart, 1930. Mobilization of sediment phosphorus by submersed freshwater macrophytes. Freshwater Biology, 10:229-238.

Barko, J. W. and R. M. Smart, 1981. Comparative influences of light and temperature on the growth and metabplism of selected submersed freshwater macrophytes. Ecol. Monographs, 51:219-235.

Belyaev, V. I., K. M. Khailov, and I. N. Okhotnikov, 1977. Mathematical simulation of a marine coastal ecosystem containing macrophytes. Aquat. Bot. 3:315-328.

Brannon, J. M., 1973. Seasonal variation of nutrients and physicochemical properties in the salt marsh soils of Barataria Bay, La. M. S. Thesis. L. S. U., Baton Rouge, La. 113pp.

Denny, P., 1972., Sites of nutrient absorption in aquatic macrophytes. J. Ecol., 60:819-829.

Fenchel, T., 1977. Aspects of the decomposition of seagrasses. pp. 123-146. I_n :C.P. MeRoy and C Helfferich (Eds.), Seagrass Ecosystems : A Scientific Perspective . Marcel Dekker, Inc. New York.

Godshalk, G. L. and R. G. Wetzel, 1978. Decomposition of aquatic angiosperms. III. Zostera marina L. and a conceptual model of decomposition. Aquat. Bot. 202 5:329-354.

Hall, C. A. S. and J. W. Day, Jr., 1977. Systems and models: Terms and basic principles, pp. 5-36. In :C. A. S. Hall and J. W . Day, (Eds.), Ecosystem Modeling in Theory and Practice : An Introduction with Case Histories . John Wiley and Sons, Hew York.

Hall, C. A. S., J. W. Day, Jr., and 11. T. Odum, 1977. A circuit language for energy and matter, pp. 37-43. Iji :C. A. S. Hall and J. W. Day, (Eds.), Ecosystem Modeling in Theory and Practice : An Introduction with Case Histories . John Wiley and Sons, New York.

Harrison, V/. G., 1978. Experimetal measurements of nitrogen remineralization in coastal waters. Limnol. Oceanogr. 23(4): 684-694.

Ikusima, I., 1965. Ecological studies on the productivity of aquatic plant communities I. Measurement of photosynthetic activity. Bot. Mag. Tokyo. 78:202-211.

Klug, M. J., 1980. Detritus-decomposition relationships, pp. 225-246. I_n :R. C. Phillips and C. P. McRoy (Eds.). Handbook of Seagrass Biology : An Ecosystem Perspective . Garland STPM Press. New York.

McRoy, C. P. and R. J. Barsdate, 1970. Phosphate absorption in eelgrass. Linnol. Oceanogr. 15(1):6-13.

McRoy, C P., and C. McMillan, 1977. Production ecology and physiology of seagrasses. pp. 53-88. I n :P. C. M c R o y and C. Uelfferich (Eds.). Seagrass Ecosystems . A Scientific Perspective . Marcel Dekker, New York.

Miller, R. L., 1930. Phytoplankton production within the grass beds off Goose Point, Lake Pontchartrain, Louisiana. M. S. Thesis. L. S. U., Baton Rouge, La. 123pp.

National Climatic Center, Washington, D. C., 1979. Total hemispheric mean daily solar radiation for New Orleans, 1941-1970 base period, in langleys by month, from SOLMET Volume 1, Tape 9724.

Nixon, S. W., C. A. Oviatt, and S. S. Hale, 1976. Nitrogen regeneration and the metabolism of coastal marine bottom communities, pp. 269-283, !tn :J. M. Anderson and A. Macfadyen, (Eds.), The Role of Terrestrial and Aquatic Organisms in Decomposition Processes . The 17th Symposium of the British Ecological Soceity, April, 1975. Blackwell Scientific Pub., 203 London.

Mixon, S. W., 1980. Between coastal marshes and coastal waters - a review of twenty years of speculation and research on the role of salt marshes in' estuarine productivity and water chemistry, pp. 438-525. In. :P. Hamilton and K. B. MacDonald, (Eds.), Estuarine and Wetland Processes . Plenum Pub. Corp. New York.

Odum, H. T., 1971. Environment , Power and Society . Wiley-Interscience. New York. 331pp.

Patriquin, D. G., 1973. Estimation of growth rate, production, and age of the marine angiosperm, Thalassia testudinum ,Konig. Carib. Jour. Sci. 13:111-123.

Patriquin D. G. and R. Knowles, 1972. Nitrogen fixation in the rhizosphere of marine angiosperms. Mar. Biol. 16:49-58.

Portier, R., 1934. Personal communication. Environmental Studies Institute, L.S.U. Baton Rouge, La.

Salisbury, F. B., and C. Ross, 1969. Plant Physiology . Wadsworth Pub. Co., Inc., Belmont, Ca. 750pp.

Sand-Jensen, K., 1977. Effects of epiphytes on eelgrass photosynthesis. Aquat. Bot. 3:55-63.

Sand-Jensen, I\. and J. Borum, 1933. Regulation of growth of eelgrass ( Zostera marina L.) in Danish coastal waters. Marine Technology Society Journal. 17(2):15-21.

Saucier, R. T., 1963. Recent Geomorphic History of the Pontchartrain Basin . L. S. U. Press, Baton Rouge, La. 114pp.

Short, F. T., 1980. A simulation model of the seagrass production system, pp. 277-295. In :R. C. Phillips and C. P. McRoy. (Eds.), Handbook of Seagrass Biology . An Ecosystem Perspective . Garland STPM Press, New York.

Smart R. M. and J. W. Barko, 1978. Influence of sediment salinity and nutrients on the physiological ecology of selected salt marsh plants. Estuarine and Coastal Mar. Sci. 7:437-495.

Swenson, E. M., 1930. General hydrography of Lake Pontchartrain, Louisiana. pp. 57-156, I_n :J. il. 204 Stone, (Ed.), Environmental Analysis of' Lake Pontchartrain , Louisiana , Its Surrounding Wetlands , and Selected Land Uses . U. S. Army Engineer District, New Orleans. Contract No. DACW29-77-C-0253.

Thompson, B. A., and J. S. Verret, 1930. Nekton of Lake Pontchartrain, Louisiana, and its surrounding wetlands, p. 711-864. In :J. H. Stone, (Ed.), Environmental Analysis of Lake Pontchartrain , Louisiana , Its Surrounding Wetlands , and Selected Land Uses . U. S. Army Engineer District, New Orleans. Contract No. DACW29-77-C-0253.

Westlake, D. F., 1967. Some effects of low velocity currents on the metabolism of aquatic macrophytes. Jour. Exp. Bot. 10(55):187-205.

Zieman, J. C., 1968. A study of the growth and decomposition of the sea-grass Thalassia testudinum . M. S. Thesis. Univ. of Miami, Coral Gables, Fla. 50pp.

Zieman, J. C., 1975. Seasonal variation of turtle grass, Thalassia testudinum Konig, with reference to temperature and salinity effects. Aquat. Bot. 1:107-123. Table 1. equations used in CHOU and DECAY. Initial conditions and numerical coefficients of flows (k values) used in the equations are in Tables 3 and 4. Abbreviations are explained in the text.

CROH EQUATIONS Equation 1 dL - ASSIM. —TRAMS 4 XYFLO - LITTERFALL 1 dt L dE » kl"*WIN*I e_1*7*Z/SD*T»C*E - k3*L 4 k4*RR - k5*L B dt ° dHR - ASS1M dd 4TRANS - XYFLO - DEATH 2 dt“ RR dnit - k2MSIM-2.9)/((SIN-2.9)47.25)*I e~1,7*Z/SD*T»C*HR 4 k3»L - k4*RR - k6*RR 9 dt

DECAY EOUATIONS Equation 1 dllOH -LITTERFALL 4 AON 4 1M MIN - UURIAL 4 dt • dWOH « kS*L 4 k7»AON 4 kll*HOM - k9*WONMJOX»T - kBM/ON la dt dWIN - HIN 4 AIN IH ASSIM. 5 dt dWIN - k9*HON*HOX*T 4 kl3*A!N - kll’WON - kl*UIN*I e-1,7*kz/SD«T»C»L 11 dt dSOH -LITTERFALL 4 DEATH 4 UURIAL 4 IH - Mill 6 dt dSOII - k5*L 4 k6*RR 4 kBMJON 4 kl2*SON - k 10*SON«SOX*ST 12 dt dSIM - MIN IH ASSIM,,„ 7 “dt" MR dSIM - klU#SON*SOX*ST - k 12‘SON - k2*SIM«I e”1*7*Z/SD*C*T»RR 13 dt ° 205 Table 2. Values of forcing functions used in GROW and DECAY.

Date Sun Depth Secchi Current Water Sediment Water Sediment Water Sediment (1) (m) Depth (mg/24 Temper­ Tempera­ Nitro­ Nitrogen Oxygen Oxygen (in) hours) ature ture gen - (gN/m ) (ppm) (ppm) ro (8C) (gN/m )

May 534 0.44 0.62 33.8 23.6 23.6 0.06 2.98 9.25 5.00

June 544 t 0.48 1.24 33.8 26.1 24.4 0.06 2.98 7.61 5.23

July 492 0.45 0.68 33.8 27.2 25.6 0.08 2.98 7.70 5.23

Aug 466 0.52 2.01 47.1 28.6 26.7 0.04 5.30 7.00 5.00

Sep 411 0.65 1.47 34.2 28.3 26.7 0.03 5.30 7.21 5.00

Oct 362 0.60 2.00 28.8 27.8 26.1 0.03 5.30 8.63 5.00

Nov 264 0.57 1.89 28.8 15.6 22.2 0.03 5.30 8.98 5.00

Dec 211 0.46 1.42 28.8 10.6 18.3 0.04 3.69 9.84 5.00

Jan 226 0.34 1.42 28.8 16.1 17.2 0.06 3.69 11.41 5.00

Feb 301 0.41 1.40 18.2 10.0 15.6 0.07 3.69 10.51 5.00

Mar 384 0.47 0.57 36.6 18.1 18.3 0.07 6.01 8.59 5.00

Apr 483 0.63 2.00 66.1 20.6 18.9 0.11 9.31 10.00 5.00 206 207

References in Table 2.

Sun (I) - National Climatic Center, mean daily total solar radiation, 1941 - 1970, for Slidell, La., in langleys.

Depth (Z) - Average depth at Mandeville for 1978 - 1979, estimated from tapes from 6 a.m. - 6 p.m. on the dates when field measurements were taken, the average depth at Mandeville was 10cm greater than the average depth at 50m from shore in the control area (the area modeled) with a correlation coefficient of 0.85, validity 99.9%.

Secchi Depth (SD) - Secchi depth was measured once a month over a 24 hour period and averaged to give a rough estimate of yearly cyclic phenomena affecting water clarity such as color, phytoplankton biomass, suspended sediment, etc.. Usually secchi depth was the same offshore as in the grassbeds, but where differences occurred, the readings over the beds were used.

Current (C) - The current or circulation was measured using Muus balls (Doty, 1971) composed of plaster of Paris made with deionized, distilled water. The loss of plaster in grams over a measured time is indicative of all the water movement over thje block during that time period. Therefore, circuation measured by Muus balls is an integrated velocity value with no directional component.

Water Temperature (T) - Temperatures were average daily air temperatures averaged for given months from Slidell, La.. When these air temperatures were compared with water temperatures taken in the field for those days in the field, the mean water temperature was 0.9 F warmer than the air temperature and the relative changes in temperature were very similar, the correlation coefficient being 0.95, with a validity of 99.9%.

Sediment Temperature (ST) - Sediment temperatures were based on water temperatures, but the variations were s m o o t h e d .

Water Nitrogen (WIN) - Water inorganic available nitrogen for the first six months wa taken from Miller (1930) who measured phytoplankton production and physical parameters in the Lake Pontchartrain grassbed during this study. An average of his surface and bottom dissolved ammonia, nitrate, and nitrite at about 50m 208 from shore were summed to give inorganic nitrogen. Dissolved organic nitrogen and particulate forms were disregarded. The rest of the year's data were from field trips in this study and analyzed on a Technicon Autoanalyzer in the Center for 'Wetland Coastal Ecology Lab.f L. o« U ..

Sediment Nitrogen (SIN) - Sediment nitrogen was measured from sections of cores taken quarterly. Details of techniques are described in Chapter 3. To determine available nitrogen on a per square meter basis, a known volume of dried core adjacent to that used for nutrient analysis was weighed2 The densities ranged from 1.8 to 2.5 gm sediment/m . The root layer was assumed to be 25cm deep from analysis of2 all cores, and this depth multiplied by the gm/m density and by the concentration of nitrogen per gram of sediment gave values in gN/m .

Water Oxygen (WOX) - Water oxygen readings were averaged for 24 hour periods during the field study. A hydrolab was used to measure oxygen readings 5cm from the sediment surface at 50m from shre in the control area every three to four hours for 24 hours once a month for the study year.

Sediment Oxygen (SOX) - Sediment oxygen was calculated assuming that the oxygen concentration exponentially decaysed from the sediment-water surface to the depth at which the sediment assumed a uniformly black color (anaerobic condition). Then the value of oxygen at 10cm from thh sediment surface was calculated, since this was the depth at which most of the roots an rhizomes occurred. A minimum 5ppm was assumed if the calculated values fell below that level. 209

Table 3. Initial conditions of the state variables in GROW and DECAY.

State Variable Symbol Value Reference

Leaf Nitrogen L 0.456 gN/m2 Chapter 3

Root and Rhizome RR 0.258 gN/m2 Chapter 3 Nitrogen

'Water Inorganic WIN 0.06 gN/m2 Miller, 1980 Nitrogen

Water Organic WON 0.29 gN/m2 Miller, 1980 Nitrogen

Sediment SIN 2.93 gN/m2 ^ Chapter 4 Inorganic Nitrogen

Sediment Organic SON 190.0 gN/m2 ^ Chapter 4 Nitrogen

Leaf nitrogen was calculated from the average biomass of leaf material collected 50m from shore at the control site on April 30, 1978, (lOg dry wt/m ) multiplied by average percent nitrogen in the leaves in .-lay (4.56%) to give a leaf nitrogen storage of 0.456 gN/m .

The average combined2 biomass of roots and rhizomes (I5.75g dry wt/m ) was multiplied by the percent nitrogen in roots and rhizomes (1.64%) to give an initial root and rhizome nitrogen storage (RR) of 0.258 gN/m. cWater column available inorganic nitrogen (WIN) included ammonia-ammonium, nitrate, and nitrite and was the average of May water samples taken from the top and bottom of the water column by Miller (1930). The average value for inorganic nitrogen was 0.16 mg/l N which was multiplied by the depth2in meters at 50m from shore (.40m) to give 0.06 gN/m . dWater column organic nitrogen (WON) assumed to be the difference between Kjeldahl nitrogen and ammonia nitrogen was 0*72 5 mg/l multiplied by 0.40m depth to give 0.29 gN/m . 210

0 Sediraent2 available inorganic nitrogen was 2.93 gN/m , calculated by a straight line extrapolation between the concentration of ammonia-ammonium, nitrate, and nitrite (4.65 mg/l) measured 5cm from the sediment surfac^ in March with a bulk density of 2.04 g dry weight/cm and the concentration in June (7.65 mg/l) multiplied by the bulk density of 1.72 g dry weight/cm to give a value of 11.92 gN/m . The depth of the ro^t layer never exceeds .25m, therefore, the?gN/m was multiplied by .25m to give 2.93 gN/m..

Sediment organic nitrogen was2 similarly calculated and had a value of 190.0 gN/m assuming a .25m root layer. Table 4. Numerical coefficients (k's) of flows in GROW and DECAY (Figures 4 and 5, Table 1). Abbreviations in text.

Coefficient Flow Value (Per Day) Reference 0 o u 1 kl Leaf ASSIMLa . 6 • Chapter 3

k2 Root and Rhizome*3 1 .5E-05 Chapter 3 ASSIM r r

k3 TRANS Growing Season0 .03 Chapter 3 Winter .07 trial and error

k4 XYFLO Growing Seasone .03 trial and error Winter .07 trial and error

k5 Litterfall Normal^ .067 Zieman,1968 Stornr3 .5 Chapter 3 h k6 Death .005 Chapter 3

k7 AON1 .033 Chapter 3

k8 Burial3 .14 Chapter 3 V k9 MIN Water Column 5.03E-05 Godshalk and Wetzel, 1978 Chapter 3

klO MIN Sediment*" 3.10E-06 trial and error

kll IM Water Columnm .001 Portier, 1984 Anderson and McFayden, 1976

kl2 IM Sedimentn .001 Portier, 1984 Anderson and McFayden, 1976

kl3 AIN° .033 Miller, 1980

aLeaf assimilation was modeled as

k,*WIN*Il o *e“ *1*7/SD5*Z*T*C*L. Factors affecting assimilation are available water inorganic nitrogen (WIN), available

sunlight d 0 )» water depth (Z) and secchi

depth (SD), temperature (T), circulation (C), and

leaf nitrogen (L). The values of the physical

factors are shown and explained in Table 2.

Water inorganic nitrogen (WIN) measured monthly

during this field experiment or by Miller (1980)

were used for available nitrogen since the values

predicted by DECAY were not yet verified. Leaf

nitrogen (L) is instantaneously calculated by the

model. The only quantity not accounted for is

kj^. To calculate k^, the increase in

leaf nitrogen per square meter over the first

month of the study period (from May to June,

1978) was calculated. This value was divided by

31 to arrive at a daily nitrogen assimilation

rate. Since this was equal to the expression

shown above, the daily rate divided by the May

average values of the known factors in the

expression gave the value for kj^.

Root and Rhizome assimilation was modeled as

k,*((SIN-2.9)/(SIN-2.9)+7.25))*I e"{1,7/ « o *T*C*RR. Nitrogen production and SIN were

measured in the companion field experiment.

Average values of growth were used where there

were more than one value for a given level of SIN to give the production versus SIN curve (Fig. 6).

The maximum velocity of uptake was found and the

sediment nitrogen level present at half the

maximum rate of uptake (7.25gN/m ) was used

as the Michaelis constant. The growth-SIN curve

best fit the limiting nutrient Michaelis-Menton

form when a lower limit of available sediment 2 nitrogen was set at 2.9 mgN/m . The other

physical parameters are discussed above and their

values shown in Table 1. The nitrogen in roots

and rhizomes (RR) was instantaneously calculated

by the model.

Translocation is a source-sink system whereby

photosynthates are transported to locations of

growth (Salisbury and Ross, 1969). During the

growing season the quantity of nitrogen

synthesized daily in the leaves ranged from about

10% to 50% of the nitrogen present, or about 30%

on the average. This was estimated from

instantaneous growth which ranged from .01 to

.125 gN/m /day (measured by the diurnal

oxygen technique and using conversion factors

from McRoy and McMillan, 1977). It is assumed

that growth will be the same above and below

ground. There are approximately ten leaves per

shoot, so if 10% of each leaf's new nitrogen went to below ground parts, the above and below ground

new nitrogen would be equal. The percentage of

total nitrogen in leaves which was new nitrogen

(0.30) was multiplied by 10% to give a

translocation coefficient (kg) of 0.03 for

the growing season.

^There was a dramatic death of leaves from late

December to early February. Senescense is

accompanied by translocation of organic compounds

containing nitrogen to the below ground parts of

the plant. The translocation coefficient

(kg) for winter was increased from the

growing season value (0.03) by 0.005 intervals

until the modeled leaf and root and rhizome

nitrogen storages (L and RR) most closely fit the

measured values. eXylem flow was highest when there was high

nitrogen assimilation during growth (Salisbury

and Ross, 1969). Therefore, the value for the

coefficient of xylem flow was set higher during

the growing season than during the winter dormant

period. Since no literature values or direct

measurements were available, the coefficient was

assumed to have the same magnitude as the

translocation coefficient, except that when

translocation was high the xylem flow was low, and vice versa.

^The coefficient of litterfall was set at 0.067

since leaves fall from the plant approximately

once every fifteen days (Zieman, 1968, and

Patriquin, 1973).

^Violent winter storms may rip leaves from the

plants or bury leaves under sediment contributing

to water or sediment organic nitogen. Cores

taken from the site showed a 15cm loss or gain of

surface sediment during the winter and litterfall

accelerated at that time. Plant material found

in March was generally buried under a layer of

sediment which caused its death.

Root and rhizome biomass did not turn over

quickly as did the leaf biomass so that the

coefficient of root and rhizome death was lower

than that of litterfall. Roots and rhizomes were

assumed to turn over once a growing season

(Chapter 3). Generally, root biomass peaked in

October and rhizome biomass peaked in late

December (Figure 3, Chapter 3). Calculating from

the beginning of the experiment this meant that

the coefficient of death was once per

approximately 200 days.

1It was known from water samples taken at the

nitrogen enrichment site that the nitrogen added on one trip would not be detectable the next

sampling trip roughly thirty days later. It was

assumed, therefore, that total mixing occurred

within thirty days. The concentration of the

allocthanous organic nitrogen source (AON) was

multiplied by the exchange coefficient (k^)

and added to the study area water column value

until it was equal to the allocthanous water

concentration. This increased the study area

organic nitrogen concentration to that of the

outside in about 30 days. When the study area

was equal or exceeded the outside organic

nitrogen, the flow in of outside organic nitrogen

ceased.

■^Leaf litter was collected in litter traps every

two weeks. Some leaf litter would be floating in

the water column, and some was raked from the

upper layers of sediment. Since it was not known

when the leaves fell or how long it took them to

settle or be buried and become part of the

sediment, it was assumed that one week (the

average time in the trap = 1/2 x two week period)

was the time it took for leaf burial. v The relative rate of mineralization or

decomposition of seagrasses have been shown to be

proportional to temperature, dissolved oxygen, and mineral nutrients available £or microbes and

inversely proportional to particle size and

initial tissue refractility (Godshalk and Wetzel,

1978). The mineral nutrients used by microbes

during decay are calculated in the immobilization

expression. The particle size and initial tissue

refractility are assumed to remain the same

throughout the year because the decomposing

substrate is always Vallisneria leaves. The

necessary parameters are temperature and

dissolved oxygen whose values for the water

column and sediment are shown in Table 2. The

decay rate measured in the field (Chapter 3) for

summer, the control area, at 50m from shore was

1 .1 1 % of the organic nitrogen present, or .0111

(per day) x WON (gN/m ) equalled the grams

nitrogen per square meter decomposed per day.

However, k^xTxWOX was substituted for .0111

to account for temperature and dissolved oxygen

changes over the year. Thus:

.0111 (per day) = kg x T (°C) x WOX (ppm)

T=average temperature in May

WOX=average water column dissolved oxygen in May

k(j=5.08E-05 (per day/(°Cxppm))

^The coefficient of mineralization in the sediment

(k^) was estimated using a trial and error approach. An increase in the value of k1Q

would raise SIM and lower SOM. Conversely, if

1^10 were decreased, SOM would increase and

SIM would decrease. The value of SOM steadily

decreased if k^Q equalled kg. Therefore,

kiQ was decreased until both SON and SIN were

close to their real values. Seasonal variations

were dependent upon the sediment temperature and

sediment oxygen. Dissolved oxygen in the

sediment (Table 2) was lower than in the water

column. mandnThe coefficients of immobilization of

inorganic nitrogen by microbes during the decay

process (k^^ and kj^) were calculated

using a dimensional approach.

WON or SON (gN)/m^ x g detritus/gN x #bacteria/g

detritus x

g dry wt/bacteria x gN used/g dry wt bacteria/hr x

hr/day 2 = gN used by bacteria/m /day.

Numerically, the equation was:

WON or SON x 1/.025 gN/gdetritus(a) x

1 0 ^bacteria/gdetritus(b)

x 4E-12g/bacteria(b) x .025gNused/gbacteriaxhr(c) x

24hr/day = 0.00096 rounded to 0.001 x WON 2 gN/m /day a)average %N in detritus at 50m from shore at the

control area was 2.5%.

b)average figures from Louisiana salt marsh,

provided by Portier, personal communication,

and about lOx smaller than reported for

tropical seagrasses (Fenchel, 1977).

c)p.397, Anderson and McFayden, 1976.

°The coefficient of AIN (k*^) was calculated 7 the same way as the coefficient of AON (k ). Table 5. Simulated leaf and root and rhizome nitrogen storages (L and RR) from original forcing functions compared with L and RR from forcing functions plus 10% of their initial value. The number shown is the correlation coefficient. The closer the correlation coefficient is to one, the closer the amended simulation is to the original or the less sensitive the model's response is to changing the forcing function. All probabilities are less than 0.0003.

Forcing Function Changed Leaf Nitrogen (l ) Root and Rhizome Nitrogen (r r )

Bottom Sunlight (I) 0.8963 0.8440

Temperature (T) 0.9076 0.9012

Sediment Inorganic N (SIM) 0.9845 0.9873

Water Inorganic N (WIN) 0.9977 0.9972*

Circulation (C) 0.9981 0.9380*

*not a forcing function used in the equation for RR Table 6 . Results of sensitivity analysis of processes by altering coefficients of flows in GROW and DECAY. The greater the difference in peak or low values between the original and the altered condition« the greater the sensitivity of the models to that alteration.

Parameter %Change or State Variable Peak Value Time Changed Value Change original^altered Shift (gN/m )

GROW kl xlO L 2.45 1 0 18 kl and k 2 x2 L 2.45 107 x2 RR 6.65 1 0 7 k3 +0.01 L 2.45 3.42 RR 6.65 9.72 k4 +0.01 L 2.45 1.23 -7days RR 6.65 3.06 -7days k3 and k4 + .01 L 2.45 1.78 -7days RR 6.65 4.63 -7days k3 and k4 switch in December L 2.45 1.46 +28days instead of November RR 6.65 4.72 +7days k5 x2 V L 2.45 0.2 RR 6.65 0.95 k5 (wind

DECAY

X5 x2 WON 0.79 0.58 SON 216 199 k6 x2 WON 0.79 0.64 SON 216 208 k9 xl/2 WIN -0.003 WON +0.03 k9 x2 WIN +0.005 WON -0.05 klO xl/2 SIN <0 spring and summer SON 216 261 klO x2 SIN 5.71 24.9 SON$ 216 151 kl x2 WIN steady decline to near 0 k 2 x2 SIN steady decline to near 0 222 *low value instead o£ peak value $ending value Table 7. GROM in itia l conditions and forcing functions used in the s im u la tio n o f le a f and ro o t and rhizome n itro g e n o f a n itro g e n enriched grassbed 50m from shore at the nitrogen addition area.

Param eter Symbol Value R eference

Leaf nitrogen L 0.02 gN/m2 field experiment In itia l Condition

Root and Rhizome n itro g e n RR 0 .0 3 gN/m2 field experiment In itia l Condition

Leaf Assimilation k l 6.2E-08 calculated by solving Coefficient for kl given nitrogen area in itial L and WIN

Root and Rhizome k2 1.525E-06 calcuated by solving Assimilation Coefficient for k2 given nitrogen area i n i t i a l RR and SIN

F o rc in g ,F u n c tio n Nay Jun i J u l Aug1 Sep Oct Nov Dec Jan Feb Mar Apr (gN/m-4) 1978 1979

HIM 0.16 3.29 2.81 2.07 2.07 2.96 3.68 1.11 3.12 1.38 3.60 3.10

SIN* 2.98 4.68 3.65 19.03

■•■Quarterly SIN values were input to CSMP function generator which interpolated between points and gave model values daily. Figure 1. Location of the study area in Lake Pontchartrain, Louisiana. 225

LAKE PONCHARTRAIN

BONNET CARRE SPILLWAY N[w ormrs

-N LOUISIANA CREEK

STUDY AREA MARSH

CONTROL 3 AREA

- NAJAS GUADALUPENSIS

jfc - GRASS SAMPLES VALLISNERIA AMERICANA and ffl - GRASS SAMPLES and UTTER RUPPIA MARITIMA. 1954 a - MUUS BALL BRICKS AND WATER SAMPLES VALLISNERIA AMERICANA and O “ CORES RUPPIA MARTIMA. 1973. • - CORES and DETRITUS BAGS 226

Figure 2. Conceptual model of nitrogen cycling in a Vallisneria grasabed. ASSIMaasslmilation TRANS-translocation XYFLO*xylem flow NF-nitrogen fixation IMalsoobillzatlon MIN-mineralization KliMS II...... IREMIC 1 11111(11

1

GROW Figure 3 Modeling symbols used in nitrogen models (after H. T. Odum, 1971, and Hall, et al., 1977). 229

S y n b o l Name M ea n in g

S o u rc e Source of energy or naterials fr'on outaide th e s y a te n

H e a t S in k Energy degraded into 7 heat neceaaary for any real proceaa to occur

S to r a g e Paaalve atorage aodule, ( S t a t e no new energy la V a r i a b l e ) generated loading or O u n lo a d in g Work G a te Energy or aaterial intereectlon points, aay have aany aatheaatjcal relations; e g . e , e t c .

P r o d u c e r Green plants, capture sunlight as a self- > aalntalnlng, cycling receptor system

F lo w Flow without backforce

A d d i t i o n Two flows added t o g e t h e r

S w itc h Control Switch nay turn flow on or off

P a a a lv e Level of storage exerts C o n t r o l a control on X, without < £ _ < » > actual loss of Q

C o n a u n e r Consumer nodule, a salf-nalntainlng nodule ■ Q *

Flows can go either d i r e c t i o n depending on v a lu e s o f X| A I j 230

Figure 4. Growth model (GROW) for a Vallisneria americana grassbed 50m from shore in the control area for nitrogen flows and initial leaf and below ground root and rhizome nitrogen storages. Graphs of the effects of forcing functions on flows are shown in rectangles along the pathways of the flows. ^574

9571 GROW 232

Figure 5. Decomposition model (DECAY) for a Vallisneria americana grassbed 50m from shore in the control area for nitrogen flows and initial water column and sediment nitrogen storages. WOX-water oxygen SOX^sediment oxygen 233

AllOCT Atmospheric Auocthanous> ORGANIC INORGANIC NITROGEN mnm f TEMPERATURE NITROGEN ^(AON, °2 A ITI (AIN)

■ © 3AS >/ waterSs WATER v miL ^UTTERFAll/ OR6ANIC IN INORGANIC V^=(K. NITROGEN NITROGEN , (WON) (WIN I \IC=0.a6^ JC'-HCGN* Ikjrial > JL (h ) \ ^OIMENr SEDIMENT ORGANIC INORGANIC DEATH f NITROGEN IN NITROGEN iSfi. ISONI IS INI I V2. 1C = 190.06N/M IC*2.IIGN/N* DECAYNIN 234

Figure 6 . Production versus available inorganic nitrogen in the sediment, used in order to obtain values for maximum velocity of nitrogen uptake (v m a x ) anc* t^ie velocity * MONTHLY VALLISNERIA PRODUCTION VERSUS SEDIMENT AVAILABLE NITROGEN

) 03\3 OWNHC *<500 ZOHHOCDOSOT) 350- 375- 325- 200 300- 225- 275- 175- 100 25- DASHED LINES SHOW A TYPICAL PRODUCTION VERSUS NUTRIENT CURVE NUTRIENT VERSUS PRODUCTION SHOWA TYPICAL LINES DASHED - ' AUSFO AC, 98 TRUHSPEBR 1979 1978, SEPTEMBER, THROUGH FROM MARCH, VALUES 50M FROM SHORE AT THE CONTROL AREA AT CONTROL THE SHORE FROM 50M SEDIMENT AVAILABLE NITROGEN CN/SQ M CN/SQ NITROGEN AVAILABLE SEDIMENT ^ S T E A D Y S T A T E VELOCITY

MAX

Figure 7 Results of measured (M) versus simulated (S) leaf nitrogen per square meter 50m from shore at the control area. 0 1 = --- , S = ---— ). oi/)\zo zmooajHMZ Tj>pir^ MEASURED NITROGEN SIMULATED VERSUS LEAF 0.25- 2.00- 3.50-j 0.00- 0.50-i 0.75-; 2.25' 2.50- 2.75-i 3.00' 3.25- 1 1.25-i 1.75 1.50-; . 00 -= o |nrwi 1 2 3 4-5 50 M FROM SHORE AT THE CONTROL AREA FROMM AT SHORE CONTROL THE 50 CRAMS NITROGEN PER SQUAREMETER PER CRAMS NITROGEN OTSMT 17 PI, 1979 APRIL, - 1978 MAT, MONTHS 6 7 6 •| |WWI |»»HHII| H H » I|f» IM W |IW «• M |l 1 1 1 13 12 11 10 9 23? 238

Figure 8. Results for measured (M) versus simulated (S) root and rhizome nitrogen per square meter for 50m from shore at the control area. ( M = --- , S = ---- ). M ow\zo znoosiHHz razoNHia oz> -hoojo MEASURED VERSUS SIMULATED ROOT RHIZOME NITROGEN AND 0.0- 0.5-i 2.5 3.0 3.5 4.0 5.0 4.5 5.5 6.0 6.5 7. OH 0 T""' 2 50 M FROMSHORE THE50 AREA AT CONTROL 3 CRAMS NITROGENCRAMS SQUAREMETERPER OTSMY 17 PI. 1979 APRIL. 1978 - MAY. MONTHS 4 7 5 6 r 1 11 10 9

2 13 12 . 239 240

Figure 9 Results for measured (M) versus simulated (S) water organic nitrogen for 50m from shore at the control area. (M --- , S = ---- ). zmoownwz nHZ>o»o somH>« NITROGEN ORGANIC R E T A W SIMULATED S U S R E V D E R U S A E M 0.9- 0 0 0.7- 0.5' 0.2- 0.3- 0.4- 0.0- t.o-j . . 6 8 - - 0 2 50 M FROMSHOREAREA AT THE50 CONTROL 3 CRAMS NITROGENCRAMS PERSQUARE METER OTSMY 17 A. 1979 MAY. 1978- MAY. MONTHS 4 5 6 7 6 10 9 11 2 13 12 242

Figure 10. Results for measured (M) versus simulated (S) water inorganic nitrogen for 50m from shore at the control area. 0 1 = --- , S = ---- ) . zmoopQHMz nwz>oaozwpophx : NITROGEN INORGANIC WATER SIMULATED VERSUS D E R U S A E M 5.0-i 5.5-; 6.0-i 6.5-1 7.0-J 2.0-j 4.0-1 o.o-; 0.5- 2.5-; 3.0 3.5-i t.5-i 0 2 HUNDREDTHS OF GRAMS PER SQUARE METERHUNDREDTHS OF GRAMS 50 M FROMSHORE ATTHEAREA 50 CONTROL 3 OTSMY 17 A. 1979 MAY. - 1978 MAY. MONTHS 4 5 6 7 8 10 9 11 2 13 12 243 244

Figure 11. Results fo r measured (M) versus simulated (S) sediment organic nitrogen for 50m from shore at the control area. CM , S ) . MEASURED VERSUS SIMULATED SEDIMENT ORGANIC NITROGEN ORGANIC SEDIMENT SIMULATED VERSUS MEASURED z d o o ?o h h z nwz>oao Hzniwont/) 200 220 240 260-j 100 120 140 160 1 BO' 20H 40- 60-i 80 0- 50 M FROM SHORE AT SHOREAT AREATHE M FROM 50CONTROL CRAMS NITROGEN PER SQUARE NITROGEN METERCRAMS OTSMY 17 A. 1979 MAY. 1978 - MAY. MONTHS nrrT' 10 11 rrTTTTT 12 rTTTT 13 245 246

Figure 12. Results for measured CM) versus simulated (S) sediment inorganic nitrogen for 50m from shore at the control area. (M = ------, S = ------) . MEASURED VERSUS SIMULATED SEDIMENT INORGANIC NITROGEN INORGANIC SEDIMENT SIMULATED VERSUS MEASURED z p i o o z h m z nwz>oao2H m z p i z h d i d h r*»pi» 10 6 - H 0 1 2 50 M FROM SHORE AT THE CONTROL AREAAT THE SHORE FROM M CONTROL 50 3 GRAMS NITROGEN PER SOUARE PER METER NITROGEN GRAMS OTSMT 17 A. 1979 MAY. 1978 - MAT. MONTHS 4 5 6 7 e 1 1 1 13 12 11 10 9 247 248

Figure 13, Results of GROW sensitivity analysis of forcing functions. The impact on nitrogen content in leaves of increasing each forcing function by 10%. (0 = original simulation, W = increasing water nitrogen by 10%, C = increasing currents by 10%, S = increasing sediment nitrogen by 10%, T = increasing temperature by 10 %, and I = increasing solar radiation at the sediment surface by 10%), The original simulation is that shown in Figure 7, 249

FORCING FUNCTION SENSITIVITY ANALYSIS FOR LEAVES 50 M FROM SHORE AT THE CONTROL AREA CRAMS NITROGEN PER SQUARE METER

2H

22-.

20 -

18-:

16r

0 12-i

0 2 3 4 5 6 7 8 9 10 11 12 13 MONTHS MAY. 1978 - MAY. 1979 LEGENDS TRT 4

O-ORIGINAL I-SUN+10* T-TEMPERATURE-*-10X C-CIRCULATIONHOX S“SEDIMENT NITROGEN^ I OX U-WATER NITROGEN*-10X 250

Figure 14, Results of GROW sensitivity analysis of forcing functions. The impact on nitrogen content in roots and rhizomes o f increasing each forcing function by 10%, ( (0 = original simulation, W = increasing water nitrogen by 10%, C = increasing currents by 10%, S = increasing sediment nitrogen by 10%, T = increasing temperature by 10%, I = increasing solar radiation at the sediment surface by 10%). The original simulation is that shown in Figure 8. 251

FORCING FUNCTION SENSITIVITY ANALYSIS FOR ROOTS AND RHIZOMES 50 M FROM SHORE AT THE CONTROL AREA CRAMS NITROGEN PER SQUARE METER

19-1 18- 17- 16- 15-

0 14-

A 13-

12-

I 11-

M 10-

JTTTTTTTTTjr 0 1 2 3 4 5 6 7 a 9 10 11 12 13 MONTHS MAY. 1978 - MAY, 1979 LEGENDt TRT

O-ORIGINAL I-SUNvlOX T-TENPERATURE+10X__ C-CIRCULATION+IOX S-SEOIMENT NITROGEN*'0* W-WATER NITROGEN*10X 252

Figure 15. Results of simulating water organic nitrogen CgN/m^) with (----- ) and without (_____ ) allocthanous inputs. zmoo»Hwz nnzxnajo s o p ih x : S I M U L A T E D W A T E R O R G A N I C N I T R O G E N W I T H A N D W I T H O U T A L L O C T H A N O U S I N P U T S 0.55- 0.70- 0.80- 0.60-j 0.65- 0.75- 0.40- 0.45^ 0.50-j 0.30 0.35 o.ooH 0.05 0 0.15-1 0.20 0.25 . 10 TTJTT 2 50M FROM SHORE AT THE CONTROL AREACONTROLAT THE SHOREFROM 50M CRAMS VJT INPUTS /VlJITH NITROGEN PER SOUARE METER NITROGEN PER A. 98-MY 1979 - MAY. 1978 MAY. ITHOUT INPUTS tttt 7 ""T"* 10 11 "■'I'" 12 T " 13 253

254

Figure 16. Results of simulating sediment inorganic nitrogen using two different temperature and sediment oxygen regimes, (smoothed temperature and oxygen minimum of 5 ppm ------; sediment temperature equal to water temperature and sediment oxygen near 0 ppm * 3 • 255

SIMULATED SEDIMENT INORGANIC NITROGEN ALTERING T AND SOX 50M FROM SHORE AT THE CONTROL AREA CRAMS NITRCCEN PER SOUARE METER

\SM00THE0 TEMPERATURE 4-

D--.

E -I-

S -2-

-4-

-5-

MAY. 1970 - MAY. 1979 256

Figure 17. Results for measured (M) versus simulated (S) leaf nitrogen of a Vallisneria americana grassbed 50m from shore at a nitrogen enriched area. (M = ------S = ------). 257

MEASURED VERSUS SIMULATED LEAF NITROGEN 50 M FROM SHORE AT THE NITROGEN ADDITION AREA GRAMS NITROGEN PER SQUARE METER

5.5-j

5.0-

4.5-

4.0-

L E A F N I T R 0 G E N G N / S. 0 M

1.5-

1. 0-

0.5-

0. 0-

MONTHS HAY. 1978 - APRIL. 1979 Figure 18. Results for measured (M) versus simulated (S) root and rhizome nitrogen for a Vallisneria americana grassbed 50m from shore at a nitrogen enriched area. ( M = ...... , S = ------). 259

MEASURED VERSUS SIMULATED ROOT AND RHIZOME NITROGEN 50 M FROM SHORE AT THE NITROGEN ADDITION AREA CRAMS NITROGEN PER SQUARE METER 7.5-1

7.0-

6.5-

6. 0-

R 0 5.5- 0 T A 5.0- N D R 4.5- H 1 ' Z 0 4.0- M E N 3.5- 1 T R 0 3.0- C E N 2.5- C N / S 2.0- Q M t .5-

1. 0-

0.5-

0.0- 0 2 3 4 675 8 9 10 11 12 MONTHS MAY. 1978 - APRIL, 1979 260

Figure 19. Results of simulation of leaf nitrogen after altering sunlight reaching the bottom compared with leaf nitrogen at the control area under normal light conditions, 50m from shore, (original sunlight = 100%, 75% of original sunlight = 75%, etc.). FET O ATRN SNIH O VALLISNERIA ON SUNLIGHT ALTERING OF EFFECTS M OWNseo zmoo^-)«z •n>nr 2.50' 3.50H 2 2.25- 3.00' 3.25' 0 0.25' 0.50- 0.75- 2.75' 1.50- 1 1.25- 1.75- . . . 00 00 00 - - - 50 - i e t a " GRAMS NITROGEN PER SQUARE SQUARE PER METER NITROGEN GRAMS ■ r■ ■ ■ 1 . i 1 1i ■.i OTSNY 17 A, 1979 MAY, - 1975 NAY. MOWTHS *

LEAF NITROGEN NITROGEN LEAF SUNLIGHT ORIGWAl, _ y...... __ , 261 Figure 20. Results of simulation of root and rhizome nitrogen after altering sunlight reaching the bottom compared with root and rhizome nitrogen at the control area under normal light conditions, 50m from shore, (original sunlight = 1 0 0 %, 75% of original sunlight = 75%, etc.). 263

EFFECTS OF ALTERING SUNLIGHT ON VALLISNERIA ROOT AND RHIZOME NITROGEN GRAMS NITROGEN PER SQUARE METER

7.0-1

6.5-

6.0-

5.5- R 0 0 T 5.0- A N D 4.5- R H 1 4.0- Z 0 M E 3.&; N 1 T 3.0- R 0 G E 2.5- N C N 2.0- / S 0 1.5- M

0.5-

0.0'

MONTHS MAY. 1978 - MAY. 1979 LEGENDi TRT 2 «-»-► 3 4

O-ORIGINAL 7-75* SUNLIGHT 5-50* SUNLIGHT 2-25* SUNLIGHT CHAPTER 6

CONCLUSIONS AND RECOMMENDATIONS TO MANAGERS

6.0 SUMMARY OP RESULTS

Vallianeria americana grass beds of Lake

Pontchartrain contribute approximately 200 g dry weight/m /yr primary production, a continuous release of detrital material, and rapid nutrient recycling. In Chapter 3 it was concluded that the harvest method for estimating production can be used successfully in a submerged grass bed ecosystem. The harvest method does not assume any physiological relationships and is an unambiguous estimate of production. The addition of litterfall to harvest production raised annual production estimates by up to

28% over harvest estimates alone. When litterfall was corrected for decomposition, the production estimates were raised by up to 85% over harvest estimates alone.

Litterfall and decomposition increased toward shore and occurred throughout the year in the grass bed providing a continuous supply of energy and nutrients to the system. Grass bed production varied along the coastline (in areas of different nutrient loadings) and was greatest at mid-depths.

It is concluded in Chapter 4 that no one environmental factor was highly correlated with biomass or production gradients across the grass bed.

264 265

Values of the environmenal factors were most extreme

near and far from shore. Severe environmental

conditions including winter storm frontal passages may

be related to the pattern of maximum biomass and

production at mid-depths in the grass bed. The

bell-shaped depth distribution pattern is perpetuated

annually by survival of grass at mid-depths and by

dormancy of fruit during the winter. The majority of

fruit are produced at mid-depths where biomass and production are greatest. The surviving plants or new

shoots start growing first at mid-depths, and later

radiate out via the rhizome system to revegetate the entire grass bed. Although depth distribution patterns are similar along the coastline, the grass bed extends almost twice as far from shore at the control area as at the nitrogen enriched area. Water clarity (as measured by Secchi depths) was significantly greater at the control area than at the nitrogen enriched area. Light penetration may determine the compensation depth of the grass bed, and the width of the distribution pattern. Management of the grassbeds should emphasize water quality and particularly water clarity.

The modeling exercise in Chapter 5 concluded that the seasonal fluctuations of leaf and root and rhizome nitrogen simulated by GROW are very similar to field 266

experimental values. Both the GROW and DECAY models were more responsive to seasonal variations in forcing

functions than to sudden, extreme events, such as the

Bonnet Carre Spillway opening. GROW was most

sensitive to the forcing functions light reaching the

sediment surface, temperature, and sediment nitrogen.

GROW was validated using initial conditions from the

nitrogen enriched grassbed and had correlations better

than 80% between simulated and measured leaf and root

and rhizome nitrogen. Once it was validated, GROW was

run under normal and reduced light conditions. When

light reaching the sediment surface was simulated to be reduced by 50%, the Vallisneria leaf and root and

rhizome nitrogen declined to zero within a year. This

suggests that conditions that reduce water clarity could cause the loss of the grass beds.

6.1 RECOMMENDATIONS TO ECOSYSTEM MANAGERS

To "manage" the Lake Pontchartrain grass beds means to ensure their continued survival. The environmental factors which this study found to be most critical to their survival are available sunlight, temperature, and sediment nitrogen. Other researchers working in similar grass bed environments have suggested that other environmental factors,

(e.g., eutrophication, introduction of detergents, and heavy metal pollution) are detrimental to seagrass growth and survival (Peres and Pickard, 1975; Agami,

et al ., 1976; and Sand-Jensen and Borum, 1983).

Therefore, any practices which decrease water clarity

or increase water pollution should be avoided.

The following recommendations are made to state,

local and federal ecosystem managers concerned with managing the Lake Pontchartrain area:

1. All the parishes bordering the lake should pass laws requiring residential, commercial, and all other point sources of water discharge of pollution to have water treatment facilities. Storm water drainage

from towns bordering the lake (e.g., Mandeville,

Madisonville) should be rerouted to flow through natural (e.g., swamp) or man-made filter systems before being discharged into the lake.

2. Shell dredging should not be allowed in Lake

Pontchartrain until it is proven sediment plumes will not come into grass bed areas. Dredging should not be allowed within any areas from which currents would tend to transport the sediment contaminated water plumes toward the grass beds.

3. The grass beds should be included in the areas protected and considered sensitive to industrial disasters (e.g., oil spills or toxic chemical releases).

4. Since spring and summer are the most critical 268

times for grass bed establishment and growth,

enforcement of pollution laws shoud be concentrated

during this period.

5. The Sonnet Carre Spillway stimulated grass bed

growth when opened during the spring and could be

considered a safe management practice.

6 . The shorelines bordering grass beds should be

protected from local disturbances such as dredge and

fill residential communities, or any residential or

commercial construction which would cause storm water

and sewage drainage into the grass beds, sediment

disruption, and/or continually increased boat traffic.

7. Facilities which discharge thermal (heated) water (such as power plants) should not be located

along shorelines which would allow the thermal effluent to flow into grass bed areas.

0. The contiguous grass bed along the north shore of Lake Pontchartrain should be designated as a State recreation area and protected. 269

G.2 LITERATURE CITED

Agami, M, M. Litav, and Y. Waisel. 1976. The effects of various components of water pollution on the behaviour of some aquatic macrophytes of the coastal rivers of Israel. Aquat. Bot., 2:203-213.

Peres, J. M. and J. Picard. 1975. Causes of decrease and disappearance of the seagrass Posidonia oceanica on the French Mediterranean coast. Aquat. Dot., 1:133-123.

Sand-Jensen, K. and J. Borum. 1933. Regulation of growth of eelgrass ( Zostera marina L.) in Danish coastal waters. Marine Tech. Soc. Jour. 17(2):15-21. VITA

Diana Lee Steller was born June 28, 1950, in

Evansville Indiana. After spending her childhood in Florida, she attended Cornell University where she received her undergraduate degree in anthropology.

She attended University of Florida where she received a M.S. degree in Environmental Engineering

Sciences. Her career goals are to teach marine science related courses and do research concerning tropical aquatic plants.

270 DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: Diana Lee Steller

Major Field: Marine Sciences

Title o f Dissertation: PRODUCTION AND NITROGEN DYNAMICS OF A VALLISNERIA americana GRASS BED IN LAKE PONTCHARTRAIN, LOUISIANA

Approved:

Major Professor and Chairman

Dean of the Graduate sehool

EXAMINING COMMITTi

r i r-

j f • ( j J 1-

Date of Examination: November 22, 1985