<%BANNER%>

Beach restoration in Florida

Permanent Link: http://ufdc.ufl.edu/UFE0041184/00001

Material Information

Title: Beach restoration in Florida effects on sea turtle nesting and hatchling physiology
Physical Description: 1 online resource (167 p.)
Language: english
Creator: Mota, Mario
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: beach, blood, carbon, carbonate, caretta, co2, compaction, concentration, dehydration, density, diffusion, dioxide, erythrocytes, escarpment, florida, gas, grain, hatching, hatchlings, incubation, loggerhead, melbourne, morphometric, nest, nourishment, o2, oxygen, pcv, porosity, protein, sand, sea, tryglicerides, tubes, turtle, water
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
Genre: Wildlife Ecology and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Beach restoration in Florida: Effects on sea turtle nesting and hatchling physiology Florida s beaches are constantly being threatened by erosional processes such as hurricanes. When a nearshore habitat is lost, a method commonly employed to restore coastlines is beach nourishment. Beach nourishment methods can vary, and depending on project details such as sand quality, beach slope and sand compaction, newly constructed beaches can cause beneficial or detrimental impacts to sea turtle nesting ecology. My dissertation investigates the physiological consequences different sand physical properties have on incubation gas concentrations of oxygen and carbon dioxide in loggerhead sea turtles. Sand characteristics were summarized for native and nourished beaches, east and west coast Florida beaches, and correlation analyses showed how changes in one sand parameter can influence changes in others. For example, an increase in percentage of smaller sand grains leads to an increase in sand compaction that decreases porosity and reduces the volume of air in beach sand. A reduction in total sand air volume can negatively impact gas diffusion rates between clutch and atmosphere, therefore decreasing clutch survivorship. Two sand physical properties, sand compaction and total calcium carbonate, were identified as most significant in explaining beach sand variability. Their influence on incubation gas exchange and hatchling physiology was researched in two experiments. The first experiment monitored gas concentrations of carbon dioxide and oxygen in sea turtle nests laid in low and high sand compaction. Data showed that diffusion gas rates were compromised in nests located in high sand compaction during the second half of incubation. Metabolism increases during this period and at approximately incubation day 32, the rates of embryonic oxygen demand and carbon dioxide offloading rise. High sand compaction restricts gas diffusion between clutch and atmosphere. This accumulates carbon dioxide in the clutch and prevents oxygen from reaching the eggs. This combination of gases is detrimental to sea turtle development and leads to lower hatching and emerging successes. In the second experiment I found that high concentrations of unbound sand calcium carbonate reduced clutch carbon dioxide concentrations during the second half of incubation. This buffering effect maintained incubation gases within safe physiological levels and reduced clutch mortality. However, hatchlings emerging from these nests had significantly smaller body depths, higher packed cell volumes, higher total protein levels, and higher triglycerides. This physiological data indicate these hatchlings were more dehydrated and metabolized more yolk than controls. Their physiological condition worsened after crawling 10 meters, intended to replicate traversing the beach to ocean. Data show that although clutch carbon dioxide levels can be reduced by unbound sand calcium carbonate, high sand compaction has a consistently negative impact on diffusion of oxygen and carbon dioxide molecules. Differences in atmospheric concentrations, diffusion rates, and solubility of oxygen and carbon dioxide molecules, intensify this effect for the latter. Regulation of specific sand characteristics is paramount for beach nourishment projects. Engineers and regulators should incorporate the information emanating from this dissertation into future beach restoration projects because it relates human needs to the consequences beach sand can have on the incubation of loggerhead sea turtles.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Mario Mota.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Carthy, Raymond R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041184:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041184/00001

Material Information

Title: Beach restoration in Florida effects on sea turtle nesting and hatchling physiology
Physical Description: 1 online resource (167 p.)
Language: english
Creator: Mota, Mario
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: beach, blood, carbon, carbonate, caretta, co2, compaction, concentration, dehydration, density, diffusion, dioxide, erythrocytes, escarpment, florida, gas, grain, hatching, hatchlings, incubation, loggerhead, melbourne, morphometric, nest, nourishment, o2, oxygen, pcv, porosity, protein, sand, sea, tryglicerides, tubes, turtle, water
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
Genre: Wildlife Ecology and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Beach restoration in Florida: Effects on sea turtle nesting and hatchling physiology Florida s beaches are constantly being threatened by erosional processes such as hurricanes. When a nearshore habitat is lost, a method commonly employed to restore coastlines is beach nourishment. Beach nourishment methods can vary, and depending on project details such as sand quality, beach slope and sand compaction, newly constructed beaches can cause beneficial or detrimental impacts to sea turtle nesting ecology. My dissertation investigates the physiological consequences different sand physical properties have on incubation gas concentrations of oxygen and carbon dioxide in loggerhead sea turtles. Sand characteristics were summarized for native and nourished beaches, east and west coast Florida beaches, and correlation analyses showed how changes in one sand parameter can influence changes in others. For example, an increase in percentage of smaller sand grains leads to an increase in sand compaction that decreases porosity and reduces the volume of air in beach sand. A reduction in total sand air volume can negatively impact gas diffusion rates between clutch and atmosphere, therefore decreasing clutch survivorship. Two sand physical properties, sand compaction and total calcium carbonate, were identified as most significant in explaining beach sand variability. Their influence on incubation gas exchange and hatchling physiology was researched in two experiments. The first experiment monitored gas concentrations of carbon dioxide and oxygen in sea turtle nests laid in low and high sand compaction. Data showed that diffusion gas rates were compromised in nests located in high sand compaction during the second half of incubation. Metabolism increases during this period and at approximately incubation day 32, the rates of embryonic oxygen demand and carbon dioxide offloading rise. High sand compaction restricts gas diffusion between clutch and atmosphere. This accumulates carbon dioxide in the clutch and prevents oxygen from reaching the eggs. This combination of gases is detrimental to sea turtle development and leads to lower hatching and emerging successes. In the second experiment I found that high concentrations of unbound sand calcium carbonate reduced clutch carbon dioxide concentrations during the second half of incubation. This buffering effect maintained incubation gases within safe physiological levels and reduced clutch mortality. However, hatchlings emerging from these nests had significantly smaller body depths, higher packed cell volumes, higher total protein levels, and higher triglycerides. This physiological data indicate these hatchlings were more dehydrated and metabolized more yolk than controls. Their physiological condition worsened after crawling 10 meters, intended to replicate traversing the beach to ocean. Data show that although clutch carbon dioxide levels can be reduced by unbound sand calcium carbonate, high sand compaction has a consistently negative impact on diffusion of oxygen and carbon dioxide molecules. Differences in atmospheric concentrations, diffusion rates, and solubility of oxygen and carbon dioxide molecules, intensify this effect for the latter. Regulation of specific sand characteristics is paramount for beach nourishment projects. Engineers and regulators should incorporate the information emanating from this dissertation into future beach restoration projects because it relates human needs to the consequences beach sand can have on the incubation of loggerhead sea turtles.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Mario Mota.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Carthy, Raymond R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041184:00001


This item has the following downloads:


Full Text

PAGE 1

BEACH RESTORATION IN FLORIDA: EFFECTS ON SEA TURTLE NESTING AND HATCHLING PHYSIOLOGY By MARIO JORGE MOTA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Mario Jorge Mota 2

PAGE 3

To my jogging buddies Cassy and Yuri 3

PAGE 4

ACKNOWLEDGMENTS I would like to thank the Disney Wildlife Conservation Fund, NASA, and Dynamac Corporation for funding and providing the necessary logistic assistance to complete this project. Many people contributed significantly to the fruition of this project. I am very thankful to my committee Chair, Ray Carthy, and committ ee members Franklyn Percival, Allen Foley, Robert Dean and Nat Frazer. I became a better biologist by inte racting with them throughout the years. Significant time was spent traversing Florida to mark nests and sample air from sea turtle nests located in different beaches. I am very gr ateful to the sea turtle permit holders and nest monitoring biologists who provided me with unforgettable assistance. In Melbourne Beach, I am in de bt to the University of Ce ntral Florida Marine Turtle Research crew, particularly Dr. L. Ehrhart, Dean Bagley, Karen Frutchey, Karen HollowayAdkins, Anne Marie Lauritsen, Kelly Roberts and St acy Cubis. In Naples, I am thankful to the Collier County Department of Park s and Recreation, particularly Maura Kraus, Mary Toro and Jason Seitz. In Anna Maria Is land, I am hugely thankful to Suzi Fox and the Anna Maria Turtle Watch Group. At Kennedy Space Center, I would like to thank the Merritt Island National Refuge and Cape Canaveral Air Force Station biol ogists for allowing me to work in their private beaches. Gas analyses were conducted at NASAs Space Life Sciences Laboratory under the guidance and assistance of Barbara Pete rson, Lanfang Levine, and Jeff Richards. A very special thank you goes to Meghan Kope rski for her help with the difficult and bureaucratic lengthy permitting process. This research was conducted under Florida Fish and Wildlife Conservation Commission Marine Turtle Permit # 94. 4

PAGE 5

Many friends volunteered time to help me with numerous parts of this project that often involved transporting lots of sand samples, drying sand, working long hours on beaches full of mosquitoes and sand fleas, or just offering good advise. I would like to thank Frank Knies, Debbie Jones, Marcia Fullylove, and Kim Reich for their support and encouragement throughout the years. Finally, I would like to thank my mother for all the support she has given me. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................8 LIST OF FIGURES .......................................................................................................................12 ABSTRACT ...................................................................................................................................13 CHAPTER 1 BEACH NOURISHMENT AND SEA TURTLE NESTING................................................16 Introduction .............................................................................................................................16 Beach Nourishment.........................................................................................................17 Beach Nourishment and Sea Turtles ...............................................................................20 2 FLORIDA BEACH SA ND PHYSIOGNOMY......................................................................27 Introduction .............................................................................................................................27 Materials and Methods ...........................................................................................................28 Description of Parameters and Method of Analysis ........................................................32 Sand Compaction and Shear Resistance ..........................................................................32 Percent Water ..................................................................................................................34 Grain Size Percent Composition .....................................................................................34 Particle Density, Bulk Density, and Sand Porosity .........................................................35 Sand pH...........................................................................................................................37 Percent Organic Carbon, Calcium Ca rbonate and Siliciclastic Matter ...........................37 Inorganic Metals..............................................................................................................37 Results .....................................................................................................................................38 Discussion ...............................................................................................................................74 3 A HUNDRED TURTLE EGGS ON THE NOURISHED BEACH.......................................81 Introduction .............................................................................................................................81 Materials and Methods ...........................................................................................................83 Results .....................................................................................................................................87 Discussion ...............................................................................................................................98 Conclusion ............................................................................................................................104 4 CARBON DIOXIDE, CARBONATE AND HATCHLING PHYSIOLOGY.....................106 Introduction ...........................................................................................................................106 Materials and Methods .........................................................................................................108 Results ...................................................................................................................................112 6

PAGE 7

Discussion .............................................................................................................................125 Conclusion ............................................................................................................................129 5 FUTURE BEACH NOURISHMENT AND SEA TURTLE NESTING.............................132 APPENDIX A GRAIN SIZE CLASS DATA FOR 10 FLORIDA SEA TURTLE NESTING BEACHES............................................................................................................................135 B INORGANIC METAL CONCENTRATIONS IN SEA TURTLE NESTING BEACH SAND...................................................................................................................................156 LIST OF REFERENCES .............................................................................................................161 BIOGRAPHICAL SKETCH .......................................................................................................167 7

PAGE 8

LIST OF TABLES Table page 2-1 Florida beaches sampled for Part II. ..................................................................................33 2-2 Beach survey data for Part I ...............................................................................................46 2-3 Descriptive statistics for Ca pe San Blas Beach transect sand ............................................47 2-4 Descriptive statistics for Anna Maria Island Beach transect sand .....................................47 2-5 Descriptive statistics for Casey Key Beach transect sand .................................................48 2-6 Descriptive statistics fo r Naples Beach transect sand ........................................................48 2-7 Descriptive statistics fo r Delray Beach transect sand ........................................................49 2-8 Descriptive statistics for Juno Beach transect sand ...........................................................49 2-9 Descriptive statistics for a nourishe d section of Melbourne Beach transect sand .............50 2-10 Descriptive statistics for a native se ction of Melbourne Beach transect sand ...................50 2-11 Descriptive statistics for Ke nnedy Space Center Beach transect sand ..............................51 2-12 Descriptive statistics fo r Flagler Beach transect sand ........................................................51 2-13 Descriptive statistics for transect sand collected from five native and five nourished beaches ...............................................................................................................................52 2-14 Descriptive statistics for transect sand collected from east and west coast Florida beaches ...............................................................................................................................53 2-15 Results of a non-parametric paired t test comparing sand data collected at the dune and high tide sampling sites of each transect and beach. ...................................................54 2-16 Results of a non-parametric t test compari ng differences between the transect data of each beach ..........................................................................................................................55 2-17 Results of a Kruskal Wallis ANOVA compari ng sand data from five native and five nourished beaches ..............................................................................................................56 2-18 Results of a Kruskal Wallis ANOVA compar ing sand data from east coast and west coast Florida beaches .........................................................................................................57 2-19 Results of a multiple comparison ANOVA te sting for interactions between Florida coast (east vs. west) and beach type (native vs. nourished) ...............................................58 8

PAGE 9

2-20 Results of a Principle Components Anal ysis done to extract beach sand properties that account for most of the varian ce in beach physical characteristics ............................59 2-21 Varimax rotation of the top four prin ciple components to maximize component factor loadings ....................................................................................................................59 2-22 Multiple regression analysis for the depe ndent variable sand water content in any Florida beach ......................................................................................................................61 2-23 Multiple regression analysis for the depe ndent variable sand water content in native Florida beaches..................................................................................................................61 2-24 Multiple regression analysis for the de pendent variable sand water content in nourished Florida beaches..................................................................................................62 2-25 Multiple regression analysis for the de pendent variable sand calcium carbonate in native Florida beaches ........................................................................................................62 2-26 Multiple regression analysis for the de pendent variable sand calcium carbonate in any Florida beach ...............................................................................................................63 2-27 Multiple regression analysis for the de pendent variable sand calcium carbonate in nourished Florida beaches..................................................................................................63 2-28 Multiple regression analysis for the de pendent variable sand organic content in native east coast Florida beaches .......................................................................................64 2-29 Multiple regression analysis for the de pendent variable sand organic content in nourished east coast Florida beaches .................................................................................64 2-30 Multiple regression analysis for the de pendent variable sand organic content in native west coast Florida beaches ......................................................................................65 2-31 Multiple regression analysis for the de pendent variable sand organic content in nourished west coast Florida beaches ................................................................................65 2-32 Results from a multiple regression analysis for the dependent variable sand organic content in any Florida beach ..............................................................................................66 2-33 Multiple regression analysis for the depende nt variable sand shear resistance in east coast Florida beaches .........................................................................................................66 2-34 Multiple regression analysis for the depe ndent variable sand shear resistance in any Florida beach ......................................................................................................................67 2-35 Multiple regression analysis for the de pendent variable sand compaction in any Florida beach ......................................................................................................................67 9

PAGE 10

2-36 Multiple regression analysis for the de pendent variable sand bulk density in east coast Florida beaches .........................................................................................................68 2-37 Multiple regression analysis for the depe ndent variable sand bulk density in west coast Florida beaches .........................................................................................................68 2-38 Multiple regression analysis for the depende nt variable sand particle density in any Florida beach ......................................................................................................................69 2-39 Results of a Spearman correlation analys is between sand physical parameters from 44 native and nourished sea turtle nesting beaches ...........................................................70 2-40 Results of a Spearman correlation analys is between sand physical parameters from 44 native and nourished sea turtle nesting beaches ...........................................................70 2-41 Results of a Spearman correlation analys is between sand physical parameters and 21 inorganic metals from 44 native and nourished sea turtle nesting beaches .......................71 2-42 Results of a Spearman correlation analysis between six sand grain size classes and 21 inorganic metals from 44 native and nourished sea turtle nesting beaches .......................73 3-1 Loggerhead sea turtle nests used to study incubation concentrations of carbon dioxide and oxygen in beaches with low (psi 1.5) and high sand compaction (psi > 1.5).....................................................................................................................................84 3-2 Descriptive statistics of nest physical dimensions meas ured from two treatments of low and high sand compaction ...........................................................................................89 3-3 Results of a Kruskal Wallis ANOVA compar ing nest cavity architecture data from nests in low and high sand compaction ..............................................................................89 3-4 Descriptive statistics of clutch egg data measured from nests of two treatments of low and high sand compaction. ..........................................................................................90 3-5 Average incubation concentrations of oxygen from loggerhead sea turtle nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5) .................................................................93 3-6 Average incubation concentrations of carbon dioxide (parts per million) from loggerhead sea turtle nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5) .............................95 3-7 Descriptive statistics of percent hatching success and emergence success data measured from nests of two treatmen ts of low and high sand compaction.. .....................97 3-8 Results of a Kruskal Wallis ANOVA comparing differences between percent hatching success and emergence success data measured from nests of two treatments of low and high sand compaction......................................................................................97 10

PAGE 11

4-1 Descriptive statistics of egg diameter and weight collected from six different loggerhead clutches ..........................................................................................................115 4-2 Results of a Kruskal-Wallis ANOVA compar ing egg diameters and weights from six different loggerhead clutches. ..........................................................................................115 4-3 Concentrations of carbon dioxide (log tr ansformed) measured in four different treatments that varied according to percentage of sand calcium carbonate .....................116 4-1 Time series of log transformed carbon dioxi de concentration data from four different treatments that varied according to percentage of sand calcium carbonate .....................117 4-4 Results of a Kruskal-Wallis ANOVA comparing incubation concentrations of carbon dioxide from four experimental treatments of carbonate sand. .......................................118 4-5 Post Hoc analysis of carbon dioxide con centrations (log) from four experimental treatments of carbonate sand using Tukeys HSD. ..........................................................118 4-6 Average incubation time (days), hatching success (%), and emergence success (%) of sea turtle clutches from four differe nt treatments of sand calcium carbonate .................119 4-7 Descriptive statistics of hatchling mor phometric data (mm) and weight (g) collected from three different experimental incubator treatments ..................................................119 4-8 Results of a Kruskal-Wallis ANOVA comparing hatchling morphometric data (mm) collected from three different e xperimental incubator treatments ...................................120 4-9 Descriptive statistics of hatchling bl ood data from three di fferent experimental incubator treatments. ........................................................................................................121 4-10 Results of a Kruskal-Wallis ANOVA comparing hatchling blood data from three different experimental incubator treatments ....................................................................122 4-11 Descriptive statistics of a 10-meter run trial and hatchling blood data from three different experimental incubator treatments. ...................................................................123 4-12 Results of a Kruskal-Wallis ANOVA comparing hatchling blood data from three different experimental incubator treatments. ...................................................................124 11

PAGE 12

LIST OF FIGURES Figure page 2-1 Five native and five nourished Florida beaches that were surveyed in Part I ....................29 2-2 Scree plot showing results obtained from a Principal Component Analysis performed to extract sand parameters that account fo r most of the variance in beach physical properties ............................................................................................................................60 3-1 Time series of log transformed carbon dioxide concentration data from nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5) .......................................................................................91 3-2 Time series of oxygen concentration data from nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5) .......................................................................................................92 4-1 Time series of log transformed carbon dioxi de concentration data from four different treatments that varied according to percentage of sand calcium carbonate .....................117 12

PAGE 13

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy BEACH RESTORATION IN FLORIDA: EFFECTS ON SEA TURTLE NESTING AND HATCHLING PHYSIOLOGY By Mario Jorge Mota December 2009 Chair: Ray Carthy Major: Wildlife Ecology and Conservation Floridas beaches are constantly being th reatened by erosional processes such as hurricanes. When a nearshore ha bitat is lost, a method commonly employed to restore coastlines is beach nourishment. Beach nourishment meth ods can vary, and depending on project details such as sand quality, beach slope and sand compaction, newly constructed beaches can cause beneficial or detrimental impact s to sea turtle nesting ecology. My dissertation investigates the physiological consequenc es different sand physical properties have on incubation gas concentrations of oxygen and carbon dioxide in loggerhead sea turtles. Sand characteristics were summarized for native and nourished beaches, east and west coast Florida beaches, and correlation analyses showed how changes in one sand parameter can influence changes in others. For example, an in crease in percentage of smaller sand grains leads to an increase in sand compaction that decreases porosity and reduces the volume of air in beach sand. A reduction in total sand air volume can negatively impact gas diffusion rates between clutch and atmosphere, therefore decreasing clutch survivorship. 13

PAGE 14

Two sand physical properties, sand compaction and total calcium carbonate, were identified as most significant in explaining beach sand variabilit y. Their influence on incubation gas exchange and hatchling physiology was researched in two experiments. The first experiment monitored gas concentr ations of carbon dioxide and oxygen in sea turtle nests laid in low and high sand compacti on. Data showed that diffusion gas rates were compromised in nests located in high sand comp action during the second half of incubation. Metabolism increases during this period and at approximately incubation day 32, the rates of embryonic oxygen demand and carbon dioxide offloa ding rise. High sand compaction restricts gas diffusion between clutch and atmosphere. Th is accumulates carbon dioxide in the clutch and prevents oxygen from reaching the eggs. This comb ination of gases is detrimental to sea turtle development and leads to lower hatching and emerging successes. In the second experiment I found that high concentrations of unbound sand calcium carbonate reduced clutch carbon diox ide concentrations during the second half of incubation. This buffering effect maintained incubation gases within safe physiological levels and reduced clutch mortality. However, hatchlings emer ging from these nests had significantly smaller body depths, higher packed cell volumes higher total protein levels, a nd higher triglycerides. This physiological data indicate these hatchlings were more dehydrated and metabolized more yolk than controls. Their physiologi cal condition worsened after cr awling 10 meters, intended to replicate traversing the beach to ocean. Data show that although clutch carbon dioxide levels can be reduced by unbound sand calcium carbonate, high sand compaction has a consistently negative impact on diffusion of oxygen and carbon dioxide molecules. Differences in atmospheric concentrations, diffusion rates, and solubility of oxygen a nd carbon dioxide molecules, intensify this effect for the latter. 14

PAGE 15

Regulation of specific sand char acteristics is paramount for beach nourishment projects. Engineers and regulators should incorporate the information emanating from this dissertation into future beach restoration pr ojects because it relates human needs to the consequences beach sand can have on the incubation of loggerhead sea turtles. 15

PAGE 16

CHAPTER 1 BEACH NOURISHMENT AND SEA TURTLE NESTING Introduction Beaches are dynamic geologic locations along coastlines. They are formed by eroded continental material such as sand and gravel that are washed to the ocean s by streams and rivers. This eroded material is suspended in the wate r column and then transported along the coast by littoral drift, a term given to longs hore currents that travel parall el to the coast and bring sand to the beach (Dean, 2002). Once deposited on the beach, constant onshore and offshore wave action gradually pushes sand and other sediments onto the back beach region, known as the planform. The intensity of this process varies seasonally and is more pr onounced during turbulent patterns of winter months that remove great swathes of sand and usually deposit them in offshore sandbars. This sand is returned to the beach during calmer spring and su mmer seasons thus recreating a sandy beach. These seasonal patterns dramatically change beach profile and appearance. The summer beach is typically wider with a flatter slope, whil e the winter profile is narrower and steeper. However, the shape of the beach planform does not depend solely on whether waves are constructive or destructive, but also on other parameters such as sand grain size composition, beach slope and sediment type (Dean, 2002). The evolution of a beach planform can be predicted by applying these parameters to the Pelnard-Cons idere (1956) equation. The approximately 825 miles of beaches frontin g the Atlantic Ocean, the Gulf of Mexico and the Straits of Florida are one of Floridas mo st valuable natural resources. However, this resource is heavily exploited by humans and the Florida Department of Environmental Protection, estimates that currently approximately 75% of residents live within 10 miles of the 16

PAGE 17

coast. This coastal settlement leads to extended development a nd pristine native beaches quickly get populated with houses, condominiums, hotels and inlets. Coastal development modifies the natural b eauty of a beach, but also has a secondary impact because it disrupts the natural succession of beach erosion and accretion. Erosion of developed coastlines occurs primarily from piers and inlet dredging, and is naturally exacerbated during the tropical hurric ane season (Clark, 1989). Coastal deve lopment, however, alters natural sand erosion/accretion movement and as consequence, eroded beach sand does not get restabilized. This leads to the destruction or loss of prope rty, threatens recreational or environmental interests and creates an economic burden for the State. To protect beaches and coastal property, the State legislature adopted the Florida Beach and Shore Preservation Act, Chapter161. This Act preserves and protects Floridas beaches and adjacent dune systems by bestowing authority on the Bureau of Beaches and Coastal Systems to monitor and evaluate beach erosion problems. It is estimated that more than 40% of the State's beaches are experiencing some degree of erosion (Clark, 1989). Once a region of critical erosion is identified, a beach management plan is developed and implemented through funding from the Florida Beach Erosion Control Program. One co mmon way to restore eroded beaches is through beach nourishment projects. Beach Nourishment Beach nourishment is a process that places la rge quantities of sand within the nearshore system. Several techniques can be used to nourish a beach. These can be accomplished by pipeline dredges, hopper dredges or truck hauls (Dean, 2002). Pipe line dredges have a pipeline positioned along side a vessel that moves along and excavates the bottom. Sand is then pumped in a water slurry to the beach via another pipeline. Hopper dre dges consist of a ship with a 17

PAGE 18

dredge pump and drag arm that suction sand and wa ter into the hopper. On ce the hopper is filled with sand, it is driven to the beach where it can be offloaded as a shoal or pumped again to shore. Truck hauls consist of transporting inland sand fr om a borrow area or sand mine to the beach. Pipeline dredging is the most commonly used method for beach nourishment in Florida (Dean, 2002). Once deposited on the beach, nourishment sand is spread and shaped to mimic native beaches. However, nourished beaches usually have flatter planforms and steeper profiles that change to resemble native profiles as nourishm ent projects equilibrate (Dean, 2002). Profile equilibration is the process where a beach takes its natural form and will depend on many factors such as project design, beach location, wave action, and sand compatibility. The time required for the planform to equilibrate is the primary factor that determines success of a nourishment project. Despite the best beach nourishment design plan s, nourished beaches inherently differ from native beaches. Some of these differences ar e in topography, sand grain size distribution, and sand temperature. The topography of nourished beaches differs from that of the local native coastline because when additional sand is deposited and spread out it forms a beach with a wider subaerial region, higher elevation, and flatter profile than those found in the native eroded coastline (Council, 1995; Dean, 2002). The increased subaerial width is the primary goal of the nourishment project, the higher elevation is due to the additional sand, and the flatter profile is a function of the project design. However, the newly formed beach is not very stable and as the planform settles, its physical characteristics change. Planform settlin g is primarily due to wave action that erodes 18

PAGE 19

and transports sand (Bruun, 2001; Dean, 2002), thus creating a very prominent escarpment near the high surf zone. Height and prominence of this escarpment vary by location and over time, but usually decrease as the project ages until they eventually assume the native coastline profile (Dean, 2002). Sand grain size distribution of a nourished be ach differs from that of a native beach because nourished sand grains are randomly mi xed, while in native beaches sand is usually separated in layers of different grain si zes (Nelson, 1985; Nelson and Mauck, 1986; Magron, 2000). This randomized sand mixture is a bypr oduct of the nourishm ent process itself, specifically the way sand is transported to shore. Sand is transported in a water slurry inside long pipelines that extend from the dredge to the active project area. This process mixes the different sized sand grains and creates a nourished beach planform with a fairly homogenous sand grain distribution (Nels on and Mauck, 1986; Magron, 2000). Separation of sand grains according to size, and subsequent formation of different planform layers are done primarily by wave action (Nelson and Mauck, 1986). As the nourished planform settles, wave action constantly erodes newly deposited sand, especially at the high surf zone. Erosion suspends sand grains in the wa ter column that then travel and precipitate according to their corresponding surface area and wei ght. This action distributes different sized sand grains along the beach slope and leads to the eventual form ation of layers. In a native beach, the planform slope is usually characteri zed by having larger sand grains closer to the surface and smaller bellow (Nel son and Mauck, 1986). The third property of nourished beaches that commonly differs from native beaches is sand temperature because nourished beach sand temperat ure is usually higher and fluctuates more than that of native beaches (Ackerman et al., 1992; Ackerman et al., 1992; Magron, 2000). This 19

PAGE 20

increase in sediment temperature is attributed primarily to two main factors: darker sand color and higher percentage of water content (Ackerman et al., 1992). Darker sand normally associated with a new b each nourishment project is a factor of its source location, specifically because it was dredged from a shoal or hauled from an inland mine. Sand that is not exposed to c ontinuous sunlight will be darker retain more light energy, and create hotter sand temperatures (Ackerman et al., 1992). As the nourished beach evolves, sun bleaches new sand and its color will gradually resemble that of th e native beach (Lucas, 2000). Nourished sand also has a higher percentage of water than that f ound in native beach sand (Ackerman et al., 1992). Because water retains heat very well, wetter sand has a high thermal conductivity and will have the capacity to incr ease sand temperature (Ackerman et al., 1985; Bustard and Greenham, 1968; Mann, 1977; McGehee, 1990). Beach Nourishment and Sea Turtles The coastal sandy beach system is home to many species of plants and animals, such as sea turtles, that are dependent upon beaches, dunes and near shore waters for the completion of important developmental stages. Floridas Atlan tic coast beaches provide nesting habitat to an estimated 30% of the world s population of loggerhead ( Caretta caretta ) sea turtles (Lutz et al., 1991). However, besides loggerheads other sea turtle species such as the Florida green ( Chelonia mydas ), and the leatherback (Dermochelys coriacea ), also commonly nest in Florida, although not as densely as the logg erhead. Because of their high se a turtle nesting density, it is paramount that Florida beaches be preserve d as an important global nesting rookery. Beach nourishment projects create beach planfo rms with different phys ical properties than those of native beaches and therefore, they have potential to cause beneficial or detrimental impacts to sea turtle nesting ecology. A review of current published literature shows that research has been conducted to evaluate the overall impact beach nourishment projects have on 20

PAGE 21

different aspects of sea turtle ethology, nest excavation be havior, incubation, and hatching success (Broadwell, 1992; Cornel isen, 1996; Crain et al., 1995; Magron, 2000; Rumbold et al., 2001). Nesting behavior of a sea turtle is impacted by beach nourishment projects because when a gravid turtle comes ashore on a nourished beach it encounters a habitat very different from the native. The nourished beach is typically charac terized by a sharp escarpment and a wider beach planform (Crain et al., 1995; Magron, 2000). Tu rtles must climb this escarpment and depending on several physiological factors su ch as their size and energy re serves, they might succeed or not. If the effort is unsuccessful, turtles can abandon their nesting attemp t for that evening or simply repeat it at a different location the same night (Bagley et al., 1994). These actions usually decrease the number of nests and increase the number of non-nesting events, commonly referred to as false-crawls, at the affected section of beach following a nourishment project (Rumbold et al., 2001). Sea turtles that manage to climb escarpments must use a lot of energy and get tired because they often nest near the escarpment edge (Bagley et al., 1994). This beha vior is more commonly observed in loggerheads than green sea turtles (p ersonal observation). Nests deposited near the beach escarpment have a higher potential to be washed out during storm events. Besides scaling a prominent escarpment, when sea turtles land on a nourished beach they encounter sand that usually has higher compac tion (Magron, 2000). Because sea turtles have a very complex nesting ritual (Carr, 1967), increas ed sand compaction has potential to impact this behavior. The usual consequence of increased sand co mpaction is deterrence of nest excavation (Mann, 1977), but Fletemeyer (1983) also reported an increase in the number of shallow nests 21

PAGE 22

following a beach nourishment project in Southeas tern Florida. Shallower nests have egg clutches that are closer to the sand surface and this makes them more prone to desiccation as well as more susceptible to pred ation (Rumbold et al., 2001). Carthy (1996) investigated logge rhead sea turtle nest architec ture in native and nourished sand in Melbourne Beach, Florida where increased sand compaction was related to differences in nest dimensions, particularly to nest neck le ngth, nest cavity depth and minimum egg depth. These parameters were correlated to turtle size (straight carapace length, carapace width and rear flipper length) and were factors of the turtles ab ility to pivot its plastron vertically in hard sand to add length to digging stroke s (Carthy, 1996). The native soft-s and of Melbourne Beach data did not show significant differences between th e variances of these three nest dimensions (Carthy, 1996). To reduce the impact that increased sand co mpaction has on sea turtle nesting, certain beach nourishment projects are ofte n required to till after sand deposition. Tilling softens sand in the subaerial region creating a less compact b each. However, tilling can have detrimental impacts if it is done too close to the beginning of the nesting season and the planform does not have sufficient time to settle. Examples of such occurrence are the 2002 Melbourne Beach and the 2004 Anna Maria Island beach nourishment projec ts that were completed in late April and were followed by heavy rain during May. This apparently lethal combination created troughs with standing water that suffocated nests laid in the immediate vicinity (p ersonal observation). Incubation of sea turtle eggs in nourished beaches is impacted by microhabitat conditions created by the new sand (Ackerman, 1981). This microhabitat is characterized by many physical properties, such as percent distribution of sand grain sizes and incr eased sand compaction. Distribution of sand grain sizes, sp ecifically an increas e in the proportion of larger diameters is 22

PAGE 23

associated with a decrease in hatching su ccess (Mortimer, 1990). An increase in sand compaction has potential to negatively impact incubation because it can restrict gas diffusion between clutch and atmosphere (Ackerman, 1977; 1981; Magron, 2000). These factors, as well as their interacti on, influence hydric conditions of sand that determine the water potential of a beach (A ckerman, 1981; Ackerman et al., 1992). Water potential of a beach can impact respiratory and osmotic properties of a clutch throughout incubation (Ackerman et al., 1992). In general, beaches with highly negative or highly positive water potentials will desiccate or suffocate clutches, respectively (Ackerman et al., 1992; McGehee, 1990; Mortimer, 1990). Another very important factor that can impact incubation of sea turtle clutches in nourished beaches is an increase in sand temperature. An increase in sand temperature can decrease incubation time; however, if the increase is extr eme it can have a lethal effect on the clutch (Yntema and Mrosovsky, 1980). Physiological temperature constraints for successful loggerhead sea turtle incubation are 26 -34 C (Yntema and Mrosovsky, 1980). Sand temperature has further implications in incubation because sea turtles have temperature-dependent sex determination (Yntema and Mrosovsky, 1980; Paukstis and Janzen, 1990). Temperaturedependent sex determination means that incubation temperature determines hatchling sex (Yntema and Mrosovsky, 1980; Paukstis and Janzen, 1990). Species have a specific pivotal temperature where sex ratios are similar, and for loggerhead sea turtles this is 29 C (Mrosovsky and Yntema, 1981). Below this temperature predominantly males are produced while above it produces mostly females (Mrosovsky and Yntema 1981). Higher incubation temperatures, such as those associated with nourished beaches, have potential to skew sex ratios of loggerhead sea 23

PAGE 24

turtles and feminize clutches (Mrosovsky a nd Yntema, 1981; Mrosovsky and Provancha, 1989; 1992; Mrosovsky, 1994). Skewing sex ratios can im pact the reproductive out put of a population. Altering sand physical characteristics, especial ly percent grain size distribution of a beach, influences its gas diffusion properties (Ackerman, 1977; 1980). Diffusion of oxygen and carbon dioxide occurs between the inside and the outsi de of the egg, as well as between the sand and atmosphere (Ackerman, 1980). Gas diffusion betw een the developing embryo and clutch air is accomplished by a vascular network within the chorioallantoic membrane of the egg. The diffusion relies on a concentration gradient that allows oxygen to move inside the semipermeable parchment eggshell and carbon dioxide to exit it This exchange supports normal hatchling development, but can be negatively impacted when beach nourishment projects alter gas diffusion properties (Ackerman, 1981). Ackerman (1981), found that embryonic development slowed and mortality increased in beaches with gas diffusion rates below the average levels of native beaches. Styrofoam castings of nest and egg mass in nourished and native beac hes allowed Carthy (1996), to calculate clutch volume as well as total chamber air space. Nests in very compact nourished sand have less air space as a consequence of altered nest geometry. Both altered nest geometry and reduction of air space significantly impacted incubation time, hatchling mass, and hatchling sex on the nourished beach (Carthy, 1996). Mortimer (1990), investigat ed the relationship between sand air volume and hatching success of green sea turtles ( Chelonia mydas ) at Ascension Island, and found a significant negative correlation between them: lower sand porosity values, such as those associated with very compact nourished sand, significantly decreased percentage hatching success. 24

PAGE 25

Hatchling emergence success can also be impact ed by sand alterations, such as the increase in sand compaction, that are associated with nour ishment beach projects. An increase in sand compaction can impede or slow the hatchling ascent (Raymond, 1984). However, other factors such as sand water content or sand porosity mu st lessen this effect because nourished beaches can have a higher percentage of hatchling emergences (Raymond, 1984; Broadwell, 1992). Despite extensive scientific research done to evaluate how beach nourishment projects affect sea turtle ecology much information is still unknown. This dissert ation takes this quest forward and investigates how beach nourishment impacts sea turtle incubation gas concentrations of oxygen and car bon dioxide and what influen ce this microenvironment has on hatchling physiology. I begin by surveying 44 native and nourished Flor ida sea turtle nesting beaches in order to characterize the physical properties of sand intrinsic to each beach type and coast. Relationships between physical parameters are determined, statistical correlations and interactions between sand physiognomy and sea turtle nesting ecol ogy are recognized. A Principal Components Analysis identifies parameters th at explain variance in sand prope rties, and these data are the foundation for two experiments described in Chapters 3 and 4. The incubation process is carefully studied in Chapter 3 by comparing gas concentrations of oxygen and carbon dioxide in loggerhead sea turt le clutches deposited in beaches with low and high sand compaction. This analysis determines how sand compaction impacts incubation gas concentrations, gas diffusion rates, embryoni c development and hatching success. Chapter 4 describes the abiotic influence in more de tail and determines how sand calcium carbonate impacts sea turtle incubation, hatchi ng success and hatchling fitness. 25

PAGE 26

Finally, Chapter 5 includes a synthesis su mmarizing how beach nourishment has potential to create different incubation environments that can affect loggerhead s ea turtle hatching success and hatchling physiology. Scientific knowledge of how beach nourishment activities impacts sea turtles is paramount for conservation, recovery plans, and future of these species. Data emanating from this project are also beneficial to the Florida Department of Environmental Protections Beaches and Coastal Systems Program because they rela te directly to the effectiveness of beach nourishment projects in meeting wildlife and human needs. 26

PAGE 27

CHAPTER 2 FLORIDA BEACH SAND PHYSIOGNOMY Introduction Florida beaches are a very important sea turtle nesting rookery. However, this rookery is not characterized by one homogeneous coastline, but a plethora of unique habitats that differ considerably in many properties such as sand qua lity, beach profile and sand compaction. These different habitats form very diverse beaches such as the crystalline-sand beaches of the Panhandle, hard-packed sandy beaches of the southw est coast, coralline beaches of the Keys, and soft coquina beaches of the north and central east coasts. This beach diversity creates very distinct sea turtle nesting habi tats characterized by different planform profiles and sa nd properties. These unique proper ties have potential to impact sea turtle nesting and incubation su ccess differently (Cornelissen, 1996). This project investigates how different sand physical characteri stics interact to affect sea turtle nesting ecology. It is divided into two part s, and the first surveys a subset of native and nourished nesting beaches to describe their intrin sic physical properties. By investigating the physical properties of these different habitats, it is possible to identify the physical characteristics that are shared by most Florida beaches and t hose unique to each coast, native or nourished beaches. Part II of this project surveys sand physical properties of 44 native a nd nourished sea turtle nesting beaches to describe corr elations and interactions that exist amongst sand properties and inorganic metals. Some of these beaches are not included in the first surveyed group of native and nourished beaches because their coastal restoration history is very diverse and prevents them from being categorized into one of th e two distinct beach categories. 27

PAGE 28

Materials and Methods According to the Florida Fish and Wildlife Conservation Commission, Florida has 190 sea turtle nesting beaches that are part of a Stat ewide Nesting Beach Survey. A sample of ten beaches, five native and five nourished, was identifie d for the first portion of this project. Native beaches were chosen because at the time of sa mpling, they were never nourished, and nourished beaches because they were restored within the la st three years. Native beaches were Cape San Blas in the Panhandles Gulf County, Casey Ke y in Sarasota County, Me lbourne Beach and the Merritt Island National Wildlife Refuge Beach in Brevard County, and Flagler Beach in Flagler County. Nourished beaches were Anna Maria Island in Manatee County, Parkshore Beach in Naples, Collier County, Delray and Juno Beaches in Palm Beach County, and Melbourne Beach (nourished section) in Brevard County ( Figure 2-1). Sampling of each beach was done using transe cts located between the dune and high tide lines following the procedures outlined in Acke rman et al. (1992), for examining the physical properties of a beach. Ackerman showed that hy dric and thermal characteristics of natural and nourished beaches could be determined by samp ling randomly between the high tide and dune lines and at depths ranging from 10-100 cm. Location of each sampling transect was determ ined using an ArcGIS program to randomly select geographical points within the nesting beach. Four trans ects were sampled per beach and each transect line was sampled at two locations one meter away from the dune line and one meter away from the high tide mark. Because leng th of exposed planform is influenced by ocean tide, water line data were rectified according to their deviation from the respective mean lower low water heights at the time of sampling. 28

PAGE 29

# # # # # $ $ $ $ $ NKennedy Space Center Melbourne Beach Cape San Blas Anna Maria Isl Flagler Park Shore Beach Casey Key Juno Delray Figure 2-1. Five native () and fi ve nourished () Florida beaches that were surveyed in Part I of this project to characterize the sand physical properties of these two different sea turtle nesting habitats. Melbourne B each had nourished and native sections. 29

PAGE 30

Data were collected at each transect site for beach width, sand compaction, shear resistance and sand temperature. Sand samples were also collected at a depth of 30 cm and brought to the laboratory for further analyses. These samples were analyzed for percent water content, grain size composition, bulk and particle density, porosity, pH, total organic content, and total calcium carbonate. Data analyses were performed using SPSS v.12 st atistical program software. Descriptive statistics as mean, standard deviation, standard error, variance, minimum, and maximum were calculated for each parameter and beach (Tables 2-3 through 2-12), for native and nourished beaches (Table 2-13), and for east an d west coast beaches (Table 2-14). A non-parametric paired t test analyzed for differences between values collected at the dune and high tide sampling sites of each transect a nd beach. This test verified if data were biased by varying sampling locations along beach width. To test if sand properties differed longitudinally along each beach, a non-parametric t test analyzed for differences between the transect data of each beach. The non-parametric Kruskal-Wallis analysis of variances (ANOVA) was used to compare differences in sand data collected for each pa rameter between native and nourished beaches, and between east and west coast beaches (Table 2-17 ). Because differences were observed in the sand physiognomy of east and west coast be aches, the General Lin ear Model (GLM) for multivariate analysis was used to check for interactions between beach type (native vs. nourished) and Florida coast (east vs. west). A Principal Components Analysis (PCA) was done to extract parameters that account for most variance in beach physical pr operties. Before the PCA was run and to assure data unbias, the Kaiser-Meyer-Olkin index of sampling adequacy and Bartlett's test of sphericity were done. 30

PAGE 31

Kaiser-Meyer-Olkin compares magnitudes of obse rved and partial correlation coefficients and a 0.60 Measure of Sampling Adequacy supports the PCA procedure. Bartlett's test of sphericity examines the null hypothesis that sand variab les in the population co rrelation matrix are uncorrelated and a si gnificance level of p 0.000 rejects the hypothesis. Data from all sand variables used in the PC A are not commonly available for every beach, nor is it logistically possible to collect all thes e data when assessing the suitability of a beach for sea turtle nesting and incubation. As such, ma thematical models were developed to obtain missing parameter data. To accomplish this, a Forward Stepwise multiple regression analysis was done to extract formulas to estimate vari ables identified by the top three principle components. All parameters in the original sand PCA analysis plus log 10 transformed data were included in the regression. Each paramete r was analyzed against all data, and then reanalyzed for east and west coasts, and for native and nourished beaches. Each regression model was studied for best fit that included evaluating its statistical significance, variation for model and residuals, and the Durban-Watson test for serial correlation of the residuals. Multicollinearity was analyzed by the variance inflation factor (VIF), variance tolerance, and condition index (CI). For Part II of this project a total of 44 b eaches were sampled, although not as extensively as those in Part I (Table 2-1) These sea turtle nesting beach es represent the unique range of native and nourished sand found throughout the st ate. Sand analyses performed for these beaches were identical to those done in Part I of this study, except that only one random sampling site was selected for each beach and sand was analyzed for inorganic metal composition. The metal composition analysis included aluminum, antimony, arsenic, barium, 31

PAGE 32

beryllium, boron, cadmium, calcium, chromi um, cobalt, copper, iron, lead, magnesium, manganese, molybdenum, potassium, sele nium, silicon, silver, and zinc. Description of Parameters and Method of Analysis Sand Compaction and Shear Resistance Sand compaction and shear resistance are ex tremely important properties because they serve as indicators of beach hardness. Beach hardness can be used to understand variations in parameters like sand water content, grain size composition, bulk density, and particle density (Cornelissen, 1996). Nourished be aches usually have extremely p acked and hard sand that can form a cement-like consistency (Nelson et. al., 1987; EAI, 1999; Ehrhart and Roberts, 2001). Sand compaction was measured in situ at a depth of 30 cm, using a Geotest pocket penetrometer fitted with a 2.54 cm adapter foot for weak soils. To perform the test, the penetrometer piston was slowly inserted into the sand until an engrav ed indicator mark was leveled with the soil. Compaction strength was read and recalibrated according to the foot adapter. The test was repeated three times and the average converted to kilograms per square centimeters (Bradford, 1986). Shear resistance was also collected in situ at a depth of 30 cm, using a Torsional vane shear tester with a midsize (2.54 cm) vane. Like a pe netrometer, this instrument also indicates sand hardness by specifically looking at tangential stress caused by sa nd viscosity. Data collection procedure consists of pressing the vane down in the sand to blade depth while maintaining constant vertical pressure. The knob is turned cloc kwise at a rate to develop failure within 5 to 10 seconds. After failure develops, spring tensi on is released and the dial indicates maximum shear value. The test was repeated three times and the average reported in kilograms per square centimeters (Bradford, 1986). 32

PAGE 33

Table 2-1. Florida beaches sampled for Part II. Coordinates are approxi mate sampling locations. Beach County Latitude Longitude Fort Pickens State Park Escambia 30 19.500 87 17.636 Pensacola Beach Santa Rosa 30 19.840 87 08.597 Fort Walton Beach Okaloosa 30 23.608 86 35.357 Panama City Beach-Nourished sand Bay 30 07.469 85 44.113 Panama City Beach-Native sand Bay 30 07.857 85 44.568 Mexico Beach Bay 29 56.539 85 24.645 Cape San Blas Beach Gulf 29 40.213 85 21.450 Anna Maria Island Beach Manatee 27 31.662 82 44.341 Coquina Beach Park/Bradenton Beach Manatee 27 26.814 82 41.840 Lido Beach Sarasota 27 18.983 82 34.943 Siesta Key Public Beach Sarasota 27 16.123 82 33.672 Casey Key Beach Sarasota 27 07.351 82 28.239 Venice Beach Sarasota 27 05.959 82 27.580 Captiva Island Public Beach Lee 26 31.581 82 11.649 Sanibel Island/ Blind Pass Beach Lee 26 28.899 82 10.958 Sanibel Island/ Lighthouse Beach Lee 26 27.108 82 00.877 Vanderbuilt Beach Collier 26 14.053 81 49.180 Naples/Parkshore Beach-Native sand Collier 26 12.258 81 48.982 Naples/Parkshore Beach-Nourished sand Collier 26 11.005 81 48.899 Marco Island Collier 26 64.733 81 43.718 Key West/Smathers Beach Monroe 24 33.089 81 46.274 Bahia Honda State Park (Ocean side) Monroe 24 39.261 81 16.797 Crandon Park South Dade 25 42.478 80 09.116 Spanish Park Beach Palm Beach 26 23.201 80 03.994 Delray Beach Palm Beach 26 27.584 80 03.466 Riviera Beach Palm Beach 26 47.076 80 01.930 Juno Beach Palm Beach 26 53.662 80 03.428 Fort Pierce Inlet-South side St. Lucie 27 27.850 80 17.373 Fort Pierce Inlet-North side St. Lucie 27 29.809 80 17.936 Vero Beach Indian 27 39.752 80 21.493 Sebastian Inlet-South side Indian 27 51.156 80 26.618 Sebastian Inlet-North side Brevard 27 51.826 80 26.900 Melbourne Beach-Native sand Brevard 28 00.792 80 31.857 Melbourne Beach-Nourished sand Brevard 28 04.182 80 33.422 Patrick Air Force Beach Brevard 28 14.594 80 36.092 Jetty Maritime Park Beach Brevard 28 24.365 80 35.431 Cape Canaveral Air Force Beach Brevard 28 25.197 80 34.828 Kennedy Space Center Brevard 28 34.329 80 34.233 Canaveral National Seashore Brevard 28 41.653 80 39.645 Flagler Beach Flagler 29 29.428 81 07.905 Matanzas Inlet BeachNorth side St Johns 29 42.793 81 13.711 Anastasia State Park St Johns 29 51.913 81 16.095 St. Augustine Inlet/ Vilano Beach St Johns 29 55.030 81 17.404 Fernandina Beach/ Amelia Island Nassau 30 39.891 81 25.864 33

PAGE 34

Percent Water Percent water content of a nesting beach is re lated to the ability of sand grains to hold water by adhesive and cohesive forces. Adhesion is defined as the attraction of solid soil particles to water, whereas cohesion relates to the mutual attraction between water molecules (Tan, 1995). Percent water content of sand is co rrelated to beach har dness, sand grain size composition, sand temperature, bulk density, a nd particle density (Cornelissen, 1996). Nourished beaches usually have higher percent water content than native beaches (Ackerman, 1981; EAI, 1999; Cornelissen, 1996). Percent water content was determined by th e gravimetric method (Gardner, 1986). This method involves measurement of water loss by weighi ng a sample before and after it is dried. Samples of approximately 150 mg of sand were colle cted at a depth of 30 cm and sealed in a preweighed bag. Bag and sample were immediatel y weighed and transported to the lab in an insulated cooler. All samples were dried in an oven at 105 C fo r at least 24 hours. Results are presented as volume percentage of water, which can be expressed in a dry mass percentage or a wet mass percentage. Each of these weight pe rcentages can be calcula ted using the equations: Dry mass % H2O = (mass H 2 O/mass oven-dry sand) x 100% (2-1) Wet mass % H2O = (mass H 2 O /mass moist sand) x 100% (2-2) Volume % H2O = dry mass % H 2 O x bulk density (2-3) Grain Size Percent Composition Grain size percent composition refers to th e proportion of differ ent-sized sediment particles present in a sand sample. The US De partment of Agriculture (Davis and Bennett, 1927) recognizes three major groups of soils separa tes: sand (2.0-0.050 mm), silt (0.050-0.002 mm), and clay (< 0.002 mm). These fractions can subse quently be subdivided in to finer size classes such as gravelly, coarse, or fine sand (Davis and Bennett, 1927). 34

PAGE 35

Grain size classes for each beach sand sample were determined by using a graduated nest of US Standard Sieves. These sieves break dow n sand into different classes using a combination of mesh sizes ranging from 16.0 mm to 0.063 mm based on the categories developed by Wentworth (1922) and Folk (1966). The pro cedure for this analysis first separates approximately 110-115 g of dried sand using a Carpo sand splitter. The sub-sample is placed inside a series of nested siev es decreasing in mesh size. Si eve sizes used were: 5 (4.0 mm), 10 (2.0 mm), 18 (1.0 mm), 40 (0.42 mm), 60 (0.25 mm), 120 (125 m), 230 (62.5 m), and 325 (44 m). This combination of sieve sizes separate s samples of different gr ain sizes ranging from pebbles to coarse silt. Nested sieves were placed inside a Tyler mechanical shaker for 10 minutes. The relative percentage of sediment retained by each sieve was then calculated. Data were analyzed for mean, mode, sorting, skewness, kurtosis and other statistics arithmetically, geometrically (mm), and logarithmically ( ) using the method of moments (Krumbein and Pettijohn, 1938) and the Folk and Ward (1957) graphical method. The shareware program GRADISAT 4.0 developed by Blott and Pye (2001) was used for intricate calculations. Mean grain size is described usi ng a modified Udden-Wentworth grade scale (Appendix A). Particle Density, Bulk Density, and Sand Porosity Particle density, bulk density, and sand porosity are factors that indicate the density of solid constituents, mass of material contained w ithin a given volume, and amount of pore space contained in a sample, respectively. These factors are important because they affect each other and also influence sand compacti on, shear resistance, percent wate r content, sand tortuosity, and gas exchange properties (Tan, 1995). Particle density is defined as the mass of a un it volume of particles, and is a measure of how tightly packed soil particle s are in any given space (Tan, 1995). Density of sand is a very useful property because it provide s information on the amount of pore space contained in a soil. 35

PAGE 36

Sands typically have total porosit y ranging from 35% to 50% whereas finer textures as silts or clays exhibit pore spaces ranging from 40% to 60% (Tan, 1995). Density and porosity provide clues concerning the ability of air to diffuse into sand and th e amount of pore space available in sand. Particle density was measured by placing a kno wn quantity (40 g) of dried sand in a 100 ml graduated cylinder and adding 50 ml of distille d water. After stirring to displace air, the volume was recorded. This volume minus total amount of water added equals volume of water displaced by sand, or volume of sand solids. Particle density was cal culated by dividing the weight of dry solid particles by the volume of solid particles and expressed in g/cm 3 Particle density = dry sand weight / sand volume (2-4) Bulk density is the mass of soil per unit volume of undisturbed or bulk soil volume (Tan, 1995). Bulk density is always less than particle density because bulk density considers density of solid volume and voids. Bulk density was determined by taking a co re sample of a known volume (100 ml) at a depth of 30 cm. This sample was brought to the lab and dried in an oven at 105 Celsius (C) for at least 24 hours. Bulk de nsity was determined by dividing the dry sample mass (grams) by the volume (cm 3 ) of the corer (Blake and Ha rtge, 1986) using the following formula: Bulk density = dry sand weight / corer volume (2-5) Porosity is defined as percentage of so il volume occupied by pore spaces, and was calculated through the measurements of bulk density and particle density (Tan, 1995) using the following formula: Porosity = 100 [(Bulk density / particle density) x 100] (2-6) 36

PAGE 37

Sand pH Sand pH is measured by first thoroughly mixing a suspension with an equal ratio of sand:water by weight and then placing a calib rated pH meter into the solution (Tan, 1995. Percent Organic Carbon, Calcium Ca rbonate and Siliciclastic Matter Percent organic carbon, calcium carbonate and siliciclastic ma tter were calculated using a loss on ignition method following the methods of Dean (1974). The protocol for this procedure ignites a pre-weighed desiccated sand sample at 550 C in a muffle furnace for one hour. Weight loss is attributed to organic matter. The same sample is then combusted at 1080 C for one hour and its post combustion weight recorded. We ight loss between 550 C and 1080 C corresponds to carbon dioxide evolved from carbonate minera ls according to the following reaction: CaCO 3 CO 2 + CaO (2-7) To calculate percent of calcium carbonate within a sample, the amount of carbon dioxide evolved is multiplied by the fraction of calcium that comprises calcium carbonate (0.40). Sand sample remaining after the complete second burn c onsists of siliciclastic matter (i.e. quartz). Inorganic Metals Inorganic metals were analyzed by an Induc tive Coupled Plasma-Mass Spectrometer at NASAs Space Life Sciences Laboratory of Kennedy Space Center. A sand sample of approximately 250 g of sand was collected, sealed in a bag and frozen. Samples were dried at 105 C and processed for inorganic analyses. The protocol for this method digests approximately 1.0 gram sediment sample in concentr ated nitric acid. The solution is filtered and diluted with deionized water to a concentration of 37% HNO 3 A subsample is injected into flowing argon and inductively heated to approximat ely 10,000 C. At this temperature, the gas is atomized and ionized, forming plasma whose ionic metal composition is measured ( Clesceri et 37

PAGE 38

al., 1999). Detailed methodology is outlined in EPA 200.7/6010 procedure for sediment analysis ( Clesceri et al., 1999). Results Beach survey data for Part I of this project are summarized in Table 2-2 and show date, time of sampling, wind speed, air temperature, number of transects and mean transect length for each beach. A non-parametric paired t test used to compar e values collected at the dune and high tide sampling sites of each transect and beach found statistically significant differences between parameters. These were: percent sand water co ntent, sand organic content, calcium carbonate, sand porosity, particle density, bulk density, pH, and the proportion of different sand grain size classes (Table 2-15). However, not all of these parameters differed in every beach and nourished beaches, particularly Naples and Delray, had a greater number of differences. Beach width appears to also be a confoundi ng factor controlling the homogeneity of nourished beach sand properties. Wider planform beaches such as Delray Beach (57.1 1.9 m) and Juno Beach (44.9 2.8 m) had more statistical differences than narrower beaches as Park Shore Beach, Naples (29.2 8.3 m). A non-parametric t test analyzed for differen ces between the transect data of each beach and showed statistically significant differences for: percent sand water content, sand organic content, calcium carbonate, sand porosity, partic le density, bulk density, pH, sand compaction, and the proportion of different sand grain size cla sses (Table 2-16). Unlike the results obtained for beach width, longitudinal differences in sand properties were more prominent and were observed in every beach. Kruskal-Wallis analysis of variances (ANOVA) results from the comparison of sand samples (n=40) from native beaches with samples (n=40) from nourished beaches show 38

PAGE 39

statistically significant differences for particle density ( F 1,78 = 5.22, p= 0.025), sand compaction ( F 1,78 = 34.64, p= 0.000), shear resistance ( F 1,78 = 27.38, p= 0.000), pH ( F 1,78 = 5.37, p= 0.023), percentage of coarse silt ( F 1,78 = 8.14, p= 0.006), and medium (< 44.0 m diameter) silt ( F 1,78 = 4.52, p= 0.037). Mean values were higher in nourished beaches, ex cept for sand shear resistance, particle density, and bulk density (Tables 2-13; Appendix A). Although particle density was higher in native beaches, mean va lues were very similar for native (3.14 g/cm 3 ) and nourished beaches (3.02 g/cm 3 ). Bulk density was also higher in native beaches, but as with particle density, the values we re very similar with 1.86 g/cm 3 for native and 1.78 g/cm 3 for nourished beaches. Analyses between sand samples (n= 48) collected in east coast beaches with those samples (n=32) from west coast beaches show statistically significant differences fo r percentage of water content ( F 1,78 = 26.6, p= 0.000), organic matter ( F 1,78 = 12.89, p= 0.001), calcium carbonate ( F 1,78 = 4.14, p= 0.045), bulk density ( F 1,78 = 5.93, p= 0.017), shear resistance ( F 1,78 = 91.50, p= 0.000), compaction ( F 1,78 = 27.03, p= 0.000), pH ( F 1,78 = 11.97, p= 0.001) and most sand grain size classes, except for medium silt (Table 2-18 ). Sand organic content, pH, calcium carbonate, particle density, bulk density, and percentage of coarse and medium sand grains were higher on the east coast, while the remaining parameters were higher on west coast beaches (Table 2-14; Appendix A). Statistically significant interactions were detect ed between coast and beach type for five of the 14 parameters tested (Table 2-19). These were: percent sand organic content ( F 1,76 = 14.15, p= 0.000), percent calcium carbonate ( F 1,76 = 9.14 p= 0.003), sand compaction (F 1,76 = 8.08, p= 0.006), pH ( F 1,76 = 5.61, p= 0.020), and very coarse ( > 1.0 mm) sand grains ( F 1,76 = 8.85, p= 0.004). 39

PAGE 40

Mean sand organic content in east coast native beaches was 0.77% (SD= 0.31) and was very similar to mean sand organic content of 0.75% (SD= 0.55) from west coast native beaches. Organic content in nourished beaches varied between coasts, and east coast nourished beaches had 1.24% (SD= 0.50) organic content while we st coast nourished beaches had 0.50% (SD= 0.16). Overall, east coast beaches had higher m ean organic content 1.0% (SD= 0.48) than west coast beaches 0.62 (SD= 0.42). Percent calcium carbonate was higher on east coast beaches 5.91% (SD= 3.11) than on west coast beaches 4.29 (SD= 4.41). However, there was a large difference between carbonate percentage according to beach type and coast. East coast native beaches had mean carbonate values of 4.72% (SD= 2.79) while nourished beac hes had 7.10% (SD= 3.0). On the west coast the opposite trend was observed and native beach es had mean carbonate values of 5.53% (SD= 5.8), while nourished beaches had 3.05% (SD= 1.8). Overall, native beaches had mean carbonate levels 5.04% (SD= 4.21) th at were similar to those of nourished beaches 5.48% (SD= 3.26). Mean sand compaction was higher on west coas t beaches 3.68 psi (SD= 0.46) than on east coast, 3.07 psi (SD= 0.55), regardless of beach type. East coast native beaches had mean sand compaction of 2.64 psi (SD= 0.34) while nourished beaches had 3.50 psi (SD= 0.34). West coast native beaches had mean sand compaction of 3.50 psi (SD= 0.51) and nourished beaches had 3.86 psi (SD= 0.34). Nourished beaches had higher mean sand compaction 3.64 psi (SD= 0.38) than native beaches 2.98 psi (SD= 0.59). Mean values for pH were higher on east coast beaches 9.05 (SD= 0.60) than on west coast beaches 8.58 (SD= 0.58). However, mean sand pH values on the east coast were very similar between native beaches 9.01 (SD= 0.71) and nourished beaches 9.09 (SD= 0.50). On the west 40

PAGE 41

coast there was a larger difference between th e sand pH of native b eaches 8.23 (SD= 0.49) and that of nourished beaches 8.92 (SD= 0.44). Overall, nourished beaches had higher mean sand pH 9.02 (SD= 0.48) than native beaches 8.70 (SD= 0.73). Mean percentage of coarse sand grains was lower in east coast beaches 8.75% (SD= 9.25) than on west coast beaches 21.32% (SD= 27.95). Native beaches also had higher mean values 17.6% (SD= 26.8) than on nourished beaches 9.94% (SD= 7.33). On the east coast, mean values for percentage of coarse sand in native beaches was 7.73% (SD= 11.24) and in nourished beaches was 9.76% (SD= 6.81). On the west coast, mean values for percentage of coarse sand in native beaches was 32.43% (SD= 35.8) and in nourished beaches was 10.2% (SD= 8.27). Principal Components Analysis (PCA) results indicate that four eigen values explain 79.8 % of the original variability (Table 2-20), but their corresponding scree plot shows that only three components are worth retaining (Figure 22). The fourth component only includes pH and its eigen value explains 10.4 % of the variance. Varimax component rotation was done to improve the interpretability of th e top four principal components that were labeled as: carbonate concentration, beach hardness, porosity index, and pH (Table 2-21). Carbonate concentration includes the paramete rs total calcium car bonate, total organic content, and very coarse sand; the second prin cipal component, beach hardness, includes sand compaction, shear resistance, particle density and coarse silt; the third, po rosity index, includes bulk density and sand porosity; and the fourth includes pH. A Forward Stepwise multiple regression analysis extracted regression formulas to estimate variables identified by the top three principle components. A regression model for percentage of water in sand was calculated for native (R 2 = 0.88, p= 0.000), nourished beaches (R 2 = 0.52, p= 41

PAGE 42

0.000), and a general model that can be applied to any Florida beach (R 2 = 0.64, p= 0.000). However, the latter is especia lly applicable to beaches whose nourishment history is unknown. Best regression equation to calculate percen tage of water in any Florida beach includes variables log CaCO 3 and beach compaction. Model II was chosen because when bulk density was added the CI became very high (32.9) indicat ing collinearity (Table 2-22). For native beaches, the best regression model was also Model II and includes variables log calcium carbonate and percent calcium car bonate (Table 2-23). Nourishe d beaches have a regression formula with log of shear resistance which corr esponds to Model I. Al though other models have a higher coefficient of determination, shear is the only variable not severely impacted by collinearity (Table 2-24). For percentage of calcium car bonate three regressi on equations were derived: one for native beaches, nourished beaches, and a general equation that can be applied to any Florida beach. Similarly what was previously mentioned, the general equation is especially applicable to beaches whose nourishment history is unknown. Regression model for native beaches and the ge neral model for any Florida beach include identical variables that are percentage of orga nic content and percentage of very coarse sand grains, Their coefficients of determination are R 2 = 0.88 ( p= 0.000) and R 2 = 0.83 ( p = 0.000), respectively. Model II was chosen for native b eaches because when bulk density is added to the equation, it increases the CI to unacceptable levels (Table 2-25 ). The regression model to estimate percentage of calcium carbonate in an y Florida beach was chosen for similar reason (Table 2-26). 42

PAGE 43

For nourished beaches, Model II that include s log of organic content and sand porosity (R 2 =0.76, p = 0.021) was chosen because when additional components are added, the CI increases to 20.68 which is very high (Table 2-27). Percent organic content in sand can be estimated using five different equations. Because of the interactions that exist between beach type and coast, regression equations were derived for east coast native (R 2 =0.78, p= 0.000) and nourished (R 2 =0.94, p= 0.000) beaches as well as for west coast native (R 2 =0.93, p= 0.000) and nourished (R 2 =0.95, p= 0.002) beaches. A general formula that can be applied to any Florid a beach was also calculated and has a R 2 =0.84 ( p= 0.05). Similarly to carbonate, the general e quation should be used for beaches whose nourishment history is unknown. On the east coast, Model I (R 2 =0.78, p= 0.000) that includes percentage of medium sand grains was chosen for native beaches because when bulk density is added the CI becomes 39.78 (Table 2-28). Model III (R 2 =0.94, p = 0.000) that includes percentage of coarse sand grains, percentage of water, and percentage of medium sand was chosen for nourished beaches (Table 229). On the west coast, organic content in nativ e beaches can be calculated using Model I (R 2 =0.93, p= 0.000) that includes only log of CaCO 3 because when pH is added it indicates collinearity (Table 2-30). In nourished beaches, the best model is number III (R 2 =0.95, p= 0.002) that includes log of CaCO 3 percentage of water, and percen tage of coarse sand (Table 231). The general model for any Florida beach uses variables percentage of CaCO 3 percentage of very fine sand, and percentage of coarse sand (Table 2-32). Shear resistance of sand can be calculated us ing two regression equations. One for east coast beaches that uses Model III and has a R 2 = 0.74 and p= 0.008 (Table 2-33). This model 43

PAGE 44

includes the parameters: percent of coarse sand, log of organic content and percent of medium sand. The second is a general equation that can be applied to any Florida beach. This regression uses Model II with a R 2 = 0.49 ( p= 0.000) that includes the va riables sand compaction and percentage of coarse sand grains. Additional variables inflate the CI beyond acceptable limits (Table 2-34). A regression equation could not be derived specifically for west coast beaches because of collinearity. One regression equation was derived to calcula te sand compaction for all Florida beaches (R 2 = 0.52, p = 0.046). This equation uses regression Mode l III (Table 2-35) that includes the parameters sand shear resistance; percent wa ter content and log of organic content. Regression equations for bulk de nsity were obtained for east and west coast beaches. The former uses Model II (R 2 = 0.31, p= 0.009) that includes percent sa nd water content and log of sand shear resistance (Table 2-36). The latter uses Model IV (R 2 = 0.62, p= 0.005) and parameters percent of very fine sand grains and log of sand organic content (Table 2-37). Particle density can be calculated for any Florida beach using a single regression equation (R 2 = 0.34, p = 0.000) that uses Model III with parameters percent of medium sand grains, log of sand calcium carbonate, and log of sand water content (Table 2-38). Statistically significant correla tions between 19 sand physical properties from a total of 44 native and nourished Florida beaches are presen ted in Tables 2-39 and 2-40. Correlations between these sand physical properties and 21 inor ganic metals are in Tables 2-41 and 2-42. Concentrations and distribution of these inor ganic metals varied wi dely throughout coastal Florida, but a few patterns were observed. First, silver was not detected (<0.000 g/g) in any beach, and cobalt and cadmium were detected in minute amounts in only a few beaches. 44

PAGE 45

45 Panhandle beaches had relatively low concentrati ons of any metals indicating that these sands are almost pure siliciclastic matter. Arsenic is present in higher concentrations in nourished sands and in beaches located in southwest Florida. A Spearmans analysis (Table 2-41), showed a strong correlation (r 43 = 0.75, p= 0.00) between arsenic and percentage of sa nd carbonate, as well as between arsenic and organic content (r 44 = 0.62, p= 0.00). Concentrations of all an alyte sampled are presented in Appendix B.

PAGE 46

46 Table 2-2. Beach survey data for Part I. Five native and five nourished beaches were selected for survey. Four transects were measured at each beach and data were co llected one meter seaward of the mean vegetation line down to one meter away from the mean high tide. Mean transect length ( SD) and mean beach width ( SD) for Cape San Blas have a large standard deviation because the east side of the southern poin t of the cape is a zone of sand accretion. This beach planform is considerably wider than that on the west side. CSB= Cape Sa n Blas; KSC= Kennedy Space Center. Beach Date Time Wind Air # Transect Beach Beach Speed Temp. Transects Length Width Type (Km/hr) C (meters) (meters) CSB 7/23/2004 1000 5.1 31.3 4 28.8 53.9 Native Anna Maria Island 7/23/2004 1500 4.9 32.1 4 21.4.3 52.8 Renourished Casey Key 7/24/2004 1300 4.9 32.1 4 15.4 28.3 Native Park Shore, Naples 7/24/2004 1500 6.9 33.3 4 11.3.3 29.2.3 Renourished Delray Beach 7/25/2004 1320 8.1 31.4 4 43.1.2 57.1.9 Renourished Juno Beach 7/25/2004 1500 9.4 32.5 4 39.5.6 44.9.8 Renourished Flagler Beach 7/26/2004 0930 2.6 29.0 4 18.7.0 29.5.9 Native KSC 7/28/2004 1120 11.4 26.3 4 15.9.7 25.7.1 Native Melbourne Beach 7/30/2004 0750 2.7 27.7 4 25.1.7 34.3.5 Renourished Melbourne Beach 7/30/2004 0950 2.4 30.4 4 18.3.6 31.1 Native

PAGE 47

Table 2-3. Descriptive statisti cs for Cape San Blas Beach transect sand (n= 8). Data are presented as mean, standard deviation (S D), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 14.2 4.48 1.58 20.1 4.30 18.3 % Organic Content 0.24 0.19 0.07 0.04 0.12 0.63 % CaCO 3 0.05 0.05 0.02 0.00 0.00 0.10 Bulk Density (g/cm 3 ) 1.55 0.16 0.06 0.03 1.26 1.76 Particle Density (g/cm 3 ) 3.24 0.31 0.11 0.10 2.85 3.88 % Porosity 51.7 6.01 2.12 36.1 46.5 62.2 Shear (psi) 7.10 0.00 0.00 0.00 7.10 7.10 Compaction (psi) 3.84 0.46 0.16 0.21 2.70 4.00 pH 8.13 0.36 0.13 0.13 7.57 8.60 Table 2-4. Descriptive statistics for Anna Maria Island Beach tran sect sand (n= 8). Data are presented as mean, standard deviation (S D), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 8.06 2.13 0.75 4.54 4.20 11.8 % Organic Content 0.55 0.21 0.07 0.05 0.20 0.83 % CaCO 3 3.53 2.28 0.81 5.19 0.73 7.90 Bulk Density (g/cm 3 ) 1.61 0.20 0.07 0.04 1.45 1.97 Particle Density (g/cm 3 ) 2.88 0.14 0.05 0.02 2.65 3.06 % Porosity 44.3 5.31 1.88 28.2 33.6 50.0 Shear (psi) 12.3 0.00 0.00 0.00 12.3 12.3 Compaction (psi) 3.79 0.46 0.16 0.21 2.70 4.00 pH 9.21 0.41 0.14 0.17 8.36 9.56 47

PAGE 48

Table 2-5. Descriptive statistics for Casey Key Beach transect sand (n= 8). Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction a nd shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 3.01 0.80 0.28 0.65 1.39 4.07 % Organic Content 1.26 0.15 0.05 0.02 0.93 1.38 % CaCO 3 11.0 1.91 0.68 3.65 7.12 13.4 Bulk Density (g/cm 3 ) 2.00 0.13 0.05 0.02 1.73 2.13 Particle Density (g/cm 3 ) 2.96 0.15 0.05 0.02 2.78 3.15 % Porosity 32.0 6.01 2.12 36.1 23.3 43.1 Shear (psi) 9.50 0.00 0.00 0.00 9.50 9.50 Compaction (psi) 3.16 0.30 0.11 0.09 2.70 3.60 pH 8.34 0.61 0.21 0.37 7.33 9.12 Table 2-6. Descriptive statistics for Naples Beach transect sand (n= 8). Data are presented as mean, standard deviation (SD), standard er ror (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 8.17 2.40 0.85 5.76 5.31 12.3 % Organic Content 0.46 0.11 0.04 0.01 0.23 0.55 % CaCO 3 2.57 1.11 0.39 1.23 1.17 4.34 Bulk Density (g/cm 3 ) 1.86 0.10 0.03 0.01 1.70 2.03 Particle Density (g/cm 3 ) 3.06 0.09 0.03 0.01 2.88 3.17 % Porosity 39.0 4.13 1.46 17.1 34.9 46.2 Shear (psi) 9.50 0.01 0.00 0.00 9.50 9.50 Compaction (psi) 3.93 0.14 0.05 0.02 3.60 4.00 pH 8.63 0.25 0.09 0.07 8.28 9.18 48

PAGE 49

Table 2-7. Descriptive statistics for Delray Beach transect sand (n= 8). Data are presented as mean, standard deviation (SD), standard er ror (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 4.42 0.91 0.32 0.82 3.09 5.33 % Organic Content 1.10 0.08 0.03 0.01 0.94 1.21 % CaCO 3 8.86 1.54 0.54 2.36 6.75 10.7 Bulk Density (g/cm 3 ) 1.95 0.13 0.05 0.02 1.73 2.10 Particle Density (g/cm 3 ) 3.09 0.12 0.04 0.02 2.92 3.29 % Porosity 37.4 5.70 2.02 32.5 32.5 47.6 Shear (psi) 7.10 0.00 0.00 0.00 7.10 7.10 Compaction (psi) 3.80 0.15 0.05 0.02 3.60 4.00 pH 9.18 0.18 0.06 0.03 8.83 9.36 Table 2-8. Descriptive statistics for Juno Beach transect sand (n= 8). Data are presented as mean, standard deviation (SD), standard er ror (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 5.93 1.13 0.40 1.27 4.78 7.84 % Organic Content 1.82 0.40 0.14 0.16 1.19 2.20 % CaCO 3 9.04 1.57 0.55 2.46 5.86 11.6 Bulk Density (g/cm 3 ) 1.70 0.08 0.03 0.01 1.60 1.80 Particle Density (g/cm 3 ) 2.92 0.02 0.01 0.00 2.89 2.95 % Porosity 41.8 2.30 0.81 5.30 38.9 45.2 Shear (psi) 7.10 0.00 0.00 0.00 7.10 7.10 Compaction (psi) 3.26 0.40 0.14 0.16 2.90 4.00 pH 9.37 0.09 0.03 0.01 9.27 9.56 49

PAGE 50

Table 2-9. Descriptive statisti cs for a nourished section of Me lbourne Beach transect sand (n= 8). Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and ma ximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 4.42 0.90 0.32 0.82 2.53 5.70 % Organic Content 0.79 0.20 0.07 0.04 0.58 1.13 % CaCO 3 3.42 1.33 0.47 1.76 1.70 5.71 Bulk Density (g/cm 3 ) 1.77 0.16 0.06 0.03 1.51 2.03 Particle Density (g/cm 3 ) 3.14 0.16 0.06 0.03 2.93 3.37 % Porosity 43.8 5.11 1.81 26.2 34.4 51.6 Shear (psi) 7.10 0.00 0.00 0.00 7.10 7.10 Compaction (psi) 3.43 0.20 0.07 0.04 3.10 3.60 pH 8.73 0.74 0.26 0.54 7.35 9.38 Table 2-10. Descriptive statistics for a native section of Melbourne Beach transect sand (n= 8). Data are presented as mean, standard devi ation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and sh ear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 5.25 2.06 0.73 4.23 3.64 9.96 % Organic Content 0.63 0.15 0.05 0.02 0.41 0.87 % CaCO 3 3.81 1.50 0.53 2.25 1.90 6.10 Bulk Density (g/cm 3 ) 1.88 0.12 0.04 0.02 1.72 2.06 Particle Density (g/cm 3 ) 3.19 0.46 0.16 0.22 2.75 4.27 % Porosity 40.0 9.39 3.32 88.2 25.0 55.0 Shear (psi) 6.20 0.00 0.00 0.00 6.20 6.20 Compaction (psi) 2.79 0.14 0.05 0.02 2.70 3.10 pH 8.36 0.82 0.29 0.68 6.91 9.21 50

PAGE 51

Table 2-11. Descriptive statisti cs for Kennedy Space Center Beach transect sand (n= 8). Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 4.05 1.82 0.64 3.33 1.13 6.39 % Organic Content 0.57 0.14 0.05 0.02 0.39 0.79 % CaCO 3 2.68 0.69 0.25 0.48 1.59 3.69 Bulk Density (g/cm 3 ) 1.88 0.13 0.05 0.02 1.64 2.07 Particle Density (g/cm 3 ) 3.18 0.21 0.07 0.04 2.94 3.50 % Porosity 41.0 4.24 1.50 18.0 34.0 47.8 Shear (psi) 4.70 0.00 0.00 0.00 4.70 4.70 Compaction (psi) 2.33 0.37 0.13 0.13 1.80 2.80 pH 9.18 0.39 0.14 0.15 8.33 9.51 Table 2-12. Descriptive statistics for Flagler Beach tr ansect sand (n= 8). Data are presented as mean, standard deviation (SD), standard er ror (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max % Water 4.30 0.71 0.25 0.50 3.23 5.64 % Organic Content 1.11 0.29 0.10 0.09 0.66 1.40 % CaCO 3 7.68 2.67 0.94 7.11 4.12 12.5 Bulk Density (g/cm 3 ) 1.99 0.11 0.04 0.01 1.78 2.10 Particle Density (g/cm 3 ) 3.10 0.12 0.04 0.02 2.85 3.25 % Porosity 36.3 4.04 1.43 16.4 29.25 41.0 Shear (psi) 2.80 0.00 0.00 0.00 2.80 2.80 Compaction (psi) 2.81 0.22 0.08 0.05 2.40 3.10 pH 9.50 0.20 0.07 0.04 9.05 9.65 51

PAGE 52

Table 2-13. Descriptive statistics for transect sand collected from five native (n= 40) and five nourished beaches (n= 40). Data are pres ented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max Native Beaches % Water 5.95 4.72 0.75 22.3 0.89 18.3 % Organic 0.76 0.42 0.07 0.18 0.12 1.40 % CaCO 3 5.04 4.21 0.67 17.7 0.00 13.4 Bulk Density (g/cm 3 ) 1.86 0.21 0.03 0.04 1.26 2.13 Particle Density (g/cm 3 ) 3.14 0.28 0.04 0.08 2.75 4.27 % Porosity 40.2 8.91 1.41 79.3 23.3 62.2 Shear (psi) 6.06 2.28 0.36 5.21 2.80 9.50 Compaction (psi) 2.99 0.59 0.09 0.35 1.80 4.00 pH 8.70 0.73 0.12 0.54 6.91 9.65 Nourished Beaches % Water 6.11 2.03 0.32 4.22 2.53 11.8 % Organic 0.94 0.54 0.09 0.30 0.20 2.20 % CaCO 3 5.61 3.37 0.53 11.4 0.73 11.6 Bulk Density (g/cm 3 ) 1.78 0.18 0.03 0.03 1.45 2.10 Particle Density (g/cm 3 ) 3.02 0.15 0.02 0.02 2.65 3.37 % Porosity 41.3 5.19 0.82 27.0 32.5 51.6 Shear (psi) 8.62 2.09 0.33 4.36 7.10 12.3 Compaction (psi) 3.64 0.38 0.06 0.15 2.70 4.00 pH 9.02 0.48 0.08 0.23 7.35 9.56 52

PAGE 53

53 Table 2-14. Descriptive statistics for transect sand collected from east (n= 48) and west coast (n= 32) Florida beaches. Data are presen ted as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction and shear resistance in pounds per square inch (psi). Parameter Mean SD SE Var Min Max East Coast % Water 4.55 1.25 0.18 1.57 0.89 7.84 % Organic 1.00 0.48 0.07 0.23 0.39 2.20 % CaCO 3 6.02 3.20 0.46 10.2 1.59 12.5 Bulk Density (g/cm 3 ) 1.86 0.15 0.02 0.02 1.51 2.10 Particle Density (g/cm 3 ) 3.10 0.24 0.03 0.06 2.75 4.27 % Porosity 40.0 5.86 0.85 34.3 25.0 55.0 Shear (psi) 5.83 1.62 0.23 2.63 2.80 7.11 Compaction (psi) 3.07 0.54 0.08 0.30 1.80 4.00 pH 9.05 0.61 0.09 0.37 6.91 9.65 West Coast % Water 8.25 4.73 0.84 22.4 1.39 18.3 % Organic 0.63 0.42 0.07 0.18 0.12 1.38 % CaCO 3 4.29 4.41 0.78 19.5 0.00 13.4 Bulk Density (g/cm 3 ) 1.75 0.23 0.04 0.06 1.26 2.13 Particle Density (g/cm 3 ) 3.04 0.23 0.04 0.05 2.65 3.88 % Porosity 41.8 8.97 1.59 80.5 23.3 62.2 Shear (psi) 9.60 1.87 0.33 3.50 7.10 12.3 Compaction (psi) 3.68 0.47 0.08 0.22 2.70 4.00 pH 8.58 0.58 0.10 0.34 7.33 9.56

PAGE 54

Table 2-15. Results of a non-parametric pa ired t test comparing sand data collected at the dune and high tide sampling sites o f each transect and beach. Only t values for sand properties that were statistically significant are shown. Parameters are: H 2 O= sand water content, Org= sand organic content, CO 3 = total sand carbonate, Por= sand por osity, PD= particle density, BD= bulk density, VCS= very coarse sand ( 1.0 mm diameter), CS= coarse sand, MS= me dium sand, and VFS= very fine sand. Sand Grain Sizes H 2 O Org CO 3 Por PD BD pH VCS CS MS VFS Cape San Blas -2.1 ** -7.3 ** Anna Maria Island -3.2 Casey Key -4.2 Park Shore, Naples -4.7 Delray Beach 11.4 ** -30.1 ** 3.4 12.4 ** 3.0 -4.3 -20.1 ** -3.2 15.0 ** Juno Beach 5.6 ** 5.4 ** 11.8 ** -5.4 ** 23.5 ** -17.8 ** Flagler Beach 3.5 8.0 ** -4.3 -3.2 KSC -6.8 ** Melbourne Beach (nou) Melbourne Beach (nat) 3.7 p= <0.05, ** p= <0.01 54

PAGE 55

55 Table 2-16. Results of a non-parame tric t test comparing di fferences between the transect data of each beach. Only t values f or sand properties that were statistically signi ficant are shown. Parameters are: H 2 O= sand water content, Org= sand organic content, CO 3 = total sand carbonate, PD= particle density, BD= bulk density, Por= sand porosity, Comp= sand compaction, VCS= very coarse sand ( 1.0 mm diameter), CS= coarse sand, MS= medium sand, and VFS= very fine sand. Sand Grain Sizes H 2 O Org CO 3 PD BD Por Comp pH VCS CS MS VFS Cape San Blas 13.5 9.0 52.5 44.9 32.5 22.9 84.2 3.2 9.6 12.8 Anna Maria Island 14.1 5.0 3.1 26.8 54.7 41.5 23.9 46.3 4.6 3.2 Casey Key 10.6 25.4 12.8 51.8 51.1 13.3 22.5 39.3 3.5 4.4 Park Shore, Naples 15.4 15.1 9.6 37.6 88.0 20.5 75.7 106 4.5 8.3 Delray Beach 35.2 99.9 36.5 125 75.6 32.2 42.1 124 4.4 5.8 ** 3.9 17.6 Juno Beach 24.1 23.4 14.3 138 463 81.8 46.5 593 40.3 80.7 20.1 7.2 ** Flagler Beach 14.7 9.7 6.1 ** 59.3 56.4 84.6 52.0 110 4.0 22.5 8.7 4.7 KSC 4.6 26.6 11.0 33.9 41.0 29.6 14.5 65.1 3.3 6.3 ** 21.8 5.0 Melbourne Beach (nou) 11.7 3.6 4.3 5.7 ** 6.5 ** 3.5 4.7 3.5 5.3 ** Melbourne Beach (nat) 7.1 ** 14.3 8.3 49.6 19.6 12.6 58.0 26.2 5.7 ** 29.3 10.8 p= <0.05, ** p= <0.01, all others p = <0.00

PAGE 56

Table 2-17. Results of a Kruskal Wallis ANOVA comp aring differences between sand data from five native (n=40) and five nourished (n=40) beaches. Only parameters that were statistically significant are shown. Desc riptive values for each parameter are presented in Table 2-13. Sum of Mean Squares Square F 1,78 P Particle Density 0.27 0.27 5.23 0.025 4.02 0.05 4.29 Sand Shear Resistance 131 131 27.39 0.000 373 4.78 504 Sand Compaction 8.59 8.59 34.64 0.000 19.33 0.25 27.92 pH 2.07 2.07 5.37 0.023 30.06 0.39 32.13 % Very Fine Sand 3338 3338 8.14 0.006 31984 410 35322 % Coarse Silt 0.26 0.26 4.52 0.037 4.42 0.06 4.68 56

PAGE 57

Table 2-18. Results of a Kruskal Wallis ANOVA comp aring differences between sand data from east coast (n=48) and west coast (n=32) Fl orida beaches. Only parameters that were statistically significant are shown. Desc riptive values for each parameter are presented in Table 2-14. Sum of Mean Squares Square F 1,78 P % Water 262 262 26.60 0.000 769 10 1031 % Organic 2.70 3 12.89 0.001 16.34 0 19.04 % CaCO 3 57.49 57 4.14 0.045 1084 14 1141 Bulk Density 0.21 0 5.93 0.017 2.82 0 3.03 Sand Shear Resistance 272 272 91.50 0.000 232 3 504 Sand Compaction 7 7 27.03 0.000 21 0 28 pH 4 4 11.97 0.001 28 0 32 % Very Coarse Sand 3035 3,035 8.38 0.005 28245 362 31280 % Coarse Sand 2184 2,184 21.08 0.000 8081 104 10265 % Medium Sand 10477 10477 40.91 0.000 19976 256 30453 % Very Fine Sand 8876 8876 26.18 0.000 26445 339 35322 57

PAGE 58

Table 2-19. Results of a multiple comparison ANOVA testing for interactions between Florida coast (east vs. west) and beach type (native vs. nourished). Only sand parameters that were statistically significant are shown. Descriptive values for each parameter are presented in Tables 2-13 and 2-14. Sum of Mean Squares Square F 1,78 P % Organic 2.46 2.46 14.15 0.000 % CaCO3 113 113 9.14 0.003 Sand Compaction 1.17 1.17 8.08 0.006 pH 1.77 1.77 5.61 0.020 % Very Coarse Sand 2,823 2,823 8.85 0.004 58

PAGE 59

Table 2-20. Results of a Princi ple Components Analysis done to extract beach sand properties that account for most of the variance in b each physical characteristic s. The first four components explain 79.8% of the variability. Component Eigenvalues % of Variance Cumulative % 1 4.31 39.19 39.19 2 2.15 19.50 58.69 3 1.18 10.72 69.41 4 1.15 10.44 79.85 5 0.78 7.05 86.90 6 0.63 5.73 92.62 7 0.33 3.01 95.63 8 0.21 1.89 97.53 9 0.17 1.51 99.04 10 0.10 0.89 99.93 11 0.01 0.07 100.0 Table 2-21. Varimax rotation of the top four principle components to maximize component factor loadings. Rotated components indica te that four eigen values explain 79.8 % of the variance in beach physical characteristics. This rotation used the Kaiser normalization and it converged in 6 iterations. Parameters Component 1 2 3 4 % CaCO 3 0.891 % Organic 0.877 Very Coarse Sand 0.743 Shear 0.840 Compaction 0.773 Coarse Silt -0.576 0.667 Particle Density -0.554 % Water Porosity 0.937 Bulk Density -0.838 pH 0.877 59

PAGE 60

1234567891011Component Number 0 1 2 3 4 5Eigenvalue Figure 2-2. Scree plot showing results obtained from a Principa l Component Analysis performed to extract sand parameters that account fo r most of the variance in beach physical properties. The top four principal components were labeled as carbonate concentration, beach hardness, porosity index, and pH. Carbonate concentration includes the parameters total calcium car bonate, total organic content, and very coarse sand; the second principal co mponent includes sand compaction, shear resistance, particle density and coarse si lt; the third, porosity index, includes bulk density and sand porosity; and the fourth in cludes only pH. The scree plot shows a curve after the third component suggesting that the fourth eigen value might not be worth retaining. The top three eigen va lues explain 69.4% of the variability. 60

PAGE 61

Table 2-22. Multiple regression analysis for the dependent variable sand water content (R 2 = 0.64, F 1,73 = 26.45, p= 0.000). Model II was chosen as a general equation to predict percent of sand water conten t in any Florida beach. SE = standard error; Tol= Tolerance; VIF= Variance inflation fact or; CI= Condition Index. Durbin-Watson value was 1.99 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 8.46 0.41 20.63 0.000 1.00 log CaCO 3 -4.68 0.54 -8.66 0.000 1.00 1.00 2.79 II (Constant) 1.54 1.39 1.10 0.274 1.00 log CaCO 3 -4.18 0.48 -8.77 0.000 0.96 1.04 2.90 Compaction 2.01 0.39 5.14 0.000 0.96 1.04 13.80 III (Constant) 9.41 2.85 3.30 0.002 1.00 log CaCO 3 -3.67 0.48 -7.66 0.000 0.85 1.18 3.30 Compaction 1.78 0.38 4.72 0.000 0.92 1.09 12.58 Bulk Density -4.04 1.30 -3.11 0.003 0.83 1.21 32.98 Table 2-23. Multiple regression analysis for the dependent variable sand water content (R 2 = 0.88, F 1,33 = 23.31, p= 0.000). Model II was chosen as a general equation to predict percent of sand water content in native Flor ida beaches. SE= standard error; Tol= Tolerance; VIF= Variance inflation fact or; CI= Condition index. Durbin-Watson value was 2.59 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 8.18 0.40 20.26 0.000 1.00 log CaCO 3 -5.73 0.50 -11.40 0.000 1.00 1.00 2.15 II (Constant) 6.78 0.43 15.89 0.000 1.00 log CaCO 3 -8.26 0.65 -12.64 0.000 0.36 2.80 2.67 % CaCO 3 0.48 0.10 4.83 0.000 0.36 2.80 5.79 III (Constant) 13.24 2.78 4.76 0.000 1.00 log CaCO 3 -8.47 0.62 -13.67 0.000 0.35 2.86 2.51 % CaCO 3 0.46 0.10 4.84 0.000 0.35 2.84 6.46 Particle Density -1.98 0.84 -2.34 0.025 0.85 1.17 31.38 61

PAGE 62

Table 2-24. Multiple regression analysis for the dependent variable sand water content (R 2 = 0.52, F 1,38 = 41.33, p= 0.000). Model I was chosen as a general equation to predict percent of sand water content in nourished Fl orida beaches. SE= standard error; Tol= Tolerance; VIF= Variance inflation fact or; CI= Condition index. Durbin-Watson value was 2.42 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) -7.81 2.18 -3.59 0.001 1.00 log Shear 15.06 2.34 6.43 0.000 1.00 1.00 19.28 II (Constant) -12.00 2.93 -4.10 0.000 1.00 log Shear 18.54 2.82 6.57 0.000 0.64 1.57 3.78 % Organic 1.03 0.51 2.04 0.048 0.64 1.57 30.46 Table 2-25. Multiple regression analysis for th e dependent variable sand calcium carbonate (R 2 = 0.88, F 1,37 = 21.04, p= 0.000). Model II was chosen as a general equation to predict percent of sand calcium carbonate content in native Florida beaches. SE= standard error; Tol= Tolerance; VIF= Variance in flation factor; CI= Condition index; VCS= Very coarse sand ( 1.0 mm diameter). Durbin-Watson value was 2.10 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) -1.84 0.61 -3.02 0.005 1.00 % Organic 9.06 0.71 12.85 0.000 1.00 1.00 3.93 II (Constant) -0.56 0.57 -0.99 0.328 1.00 % Organic 5.89 0.90 6.57 0.000 0.41 2.47 2.35 % VCS 0.06 0.01 4.59 0.000 0.41 2.47 6.61 III (Constant) -6.69 2.33 -2.87 0.007 1.00 % Organic 4.36 1.01 4.34 0.000 0.28 3.63 2.46 % VCS 0.07 0.01 5.34 0.000 0.39 2.54 7.51 Bulk Density 3.87 1.44 2.70 0.011 0.56 1.79 30.0 62

PAGE 63

Table 2-26. Multiple regression analysis for th e dependent variable sand calcium carbonate (R 2 = 0.83, F 1,77 = 50.67, p= 0.000). Model II was chosen as a general equation to predict percent of sand calcium carbonate content in any Florida beach. SE= standard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Condition index; VCS= Very coarse sand ( 1.0 mm diameter). Durbin-Watson value was 1.64 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) -0.27 0.46 -0.59 0.557 1.00 % Organic 6.57 0.47 14.13 0.000 1.00 1.00 3.76 II (Constant) -0.18 0.36 -0.51 0.609 1.00 % Organic 5.32 0.40 13.20 0.000 0.81 1.23 2.28 % VCS 0.07 0.01 7.12 0.000 0.81 1.23 4.48 III (Constant) -6.31 1.58 -3.99 0.000 1.00 % Organic 5.21 0.37 14.05 0.000 0.81 1.24 2.49 % VCS 0.06 0.01 6.47 0.000 0.76 1.32 4.74 Bulk Density 3.50 0.88 3.96 0.000 0.89 1.12 25.20 Table 2-27. Multiple regression analysis for th e dependent variable sand calcium carbonate (R 2 = 0.76, F 1,37 = 5.77, p= 0.020). Model II was chosen as a general equation to predict percent of sand calcium carbonate cont ent in nourished Florida beaches. SE= standard error; Tol= Tolera nce; VIF= Variance inflati on factor; CI= Condition index; MS= Medium Sand. Durbin-Watson value was 2.11 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 6.51 0.30 22.02 0.000 1.00 log Organic 10.91 1.10 9.89 0.000 1.00 1.00 1.44 II (Constant) 11.54 2.12 5.46 0.000 1.00 log Organic 10.79 1.04 0.36 0.000 1.00 1.00 1.67 Porosity -0.12 0.05 -2.40 0.021 1.00 1.00 16.98 III (Constant) 13.79 2.16 6.39 0.000 1.00 log Organic 11.86 1.06 11.22 0.000 0.84 1.19 1.83 Porosity -0.14 0.05 -2.85 0.007 0.98 1.02 5.67 % MS -0.04 0.02 -2.56 0.015 0.83 1.21 20.68 63

PAGE 64

Table 2-28. Multiple regression analysis for th e dependent variable sand organic content (R 2 = 0.78, F 1,22 = 79.36, p= 0.000). Model I was chosen as a general equation to predict percent of organic content in native east co ast Florida beaches. SE= standard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Condition Index; MS= Medium sand. Durbin-Watson value was 1.58 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 1.55 0.09 16.63 0.000 1.00 % MS -0.02 0.00 -8.91 0.000 1.00 1.00 5.93 II (Constant) 0.07 0.42 0.16 0.878 1.00 % MS -0.01 0.00 -9.62 0.000 0.92 1.09 6.11 Bulk Density 0.74 0.21 3.60 0.002 0.92 1.09 39.78 Table 2-29. Multiple regression analysis for th e dependent variable sand organic content (R 2 = 0.94, F 1,20 = 18.36, p= 0.000). Model III was chosen as a general equation to predict percent of organic content in nourished east coast Florid a beaches. SE= standard error; Tol= Tolerance; VIF= Variance in flation factor; CI= Condition index; CS= Coarse sand; MS= Medium sand. Durbin-Watson value was 1.78 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 0.27 0.17 1.62 0.120 1.00 % CS 0.06 0.01 6.16 0.000 1.00 1.00 5.07 II (Constant) -0.67 0.17 -3.92 0.001 1.00 % CS 0.05 0.01 8.83 0.000 0.95 1.06 5.54 % Water 0.22 0.03 6.75 0.000 0.95 1.06 10.38 III (Constant) -0.15 0.17 -0.88 0.391 1.00 % CS 0.05 0.00 12.31 0.000 0.94 1.07 6.02 % Water 0.21 0.02 8.81 0.000 0.94 1.06 8.48 % MS -0.01 0.00 -4.29 0.000 0.99 1.01 15.04 64

PAGE 65

Table 2-30. Multiple regression analysis for th e dependent variable sand organic content (R 2 = 0.93, F 1,10 = 139.7, p= 0.000). Model I was chosen as a general equation to predict percent of organic content in native west coast Florida beaches. SE= standard error; Tol= Tolerance; VIF= Variance inflati on factor; CI= Condition index. DurbinWatson value was 2.30 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 0.74 0.04 16.65 0.000 1.00 log CaCO 3 0.50 0.04 11.82 0.000 1.00 1.00 1.40 2 (Constant) 2.23 0.55 4.02 0.003 1.00 log CaCO 3 0.53 0.04 15.16 0.000 0.91 1.10 1.64 pH -0.18 0.07 -2.69 0.025 0.91 1.10 35.32 Table 2-31. Multiple regression analysis for th e dependent variable sand organic content (R 2 = 0.95, F 1,12 = 14.79, p= 0.002). Model III was chosen as a general equation to predict percent of organic content in nourished west coast Florida beaches. SE= standard error; Tol= Tolerance; VIF= Variance in flation factor; CI= Condition index; CS= Coarse sand. Durbin-Watson value was 2.19. Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 0.26 0.03 7.90 0.000 1.00 log CaCO 3 0.60 0.07 8.86 0.000 1.00 1.00 3.59 II (Constant) 0.38 0.06 6.38 0.000 1.00 log CaCO 3 0.63 0.06 10.41 0.000 0.95 1.06 3.95 % Water -0.02 0.01 -2.32 0.037 0.95 1.06 9.25 III (Constant) 0.50 0.05 9.66 0.000 1.00 log CaCO 33 0.65 0.04 15.27 0.000 0.94 1.07 3.27 % Water -0.02 0.01 -4.53 0.001 0.81 1.23 4.48 % CS -0.01 0.00 -3.85 0.002 0.86 1.17 12.15 65

PAGE 66

Table 2-32. Results from a multiple regression analysis for the dependent variable sand organic content (R 2 = 0.84, F 1,71 = 8.33, p= 0.05). Model III was chosen as a general equation to predict percent of organic content in any Florida beach. SE= standard error; Tol= Tolerance; VIF= Variance inflation factor ; CI= Condition index; CS= Coarse sand; VFS= Very fine sand. Durbin-Watson value was 1.46 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 0.29 0.05 5.93 0.000 1.00 % CaCO 3 0.10 0.01 14.08 0.000 1.00 1.00 3.34 II (Constant) 0.29 0.04 7.19 0.000 1.00 % CaCO 3 0.10 0.01 15.61 0.000 0.97 1.03 1.53 % VFS 0.52 0.09 5.79 0.000 0.97 1.03 3.50 III (Constant) 0.23 0.04 5.50 0.000 1.00 % CaCO 3 0.09 0.01 13.34 0.000 0.77 1.30 1.75 % VFS 0.54 0.09 6.32 0.000 0.96 1.04 4.02 % CS 0.01 0.00 2.89 0.005 0.80 1.25 4.14 Table 2-33. Multiple regression analysis for th e dependent variable sand shear resistance (R 2 = 0.74, F 1,44 = 7.65, p= 0.008). Model III was chosen as a general equation to predict shear resistance in east coast Florida beaches SE= standard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Condition index; CS= Coarse sand; MS= Medium sand. Durbin-Watson value was 1.67 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 7.65 0.36 21.04 0.000 1.00 % CS -0.09 0.02 -5.75 0.000 1.00 1.00 3.77 II (Constant) 8.57 0.31 28.08 0.000 1.00 % CS -0.13 0.01 -9.77 0.000 0.81 1.24 1.37 log Organic 4.73 0.75 6.30 0.000 0.81 1.24 4.36 III (Constant) 6.82 0.69 9.82 0.000 1.00 % CS -0.11 0.01 -8.64 0.000 0.69 1.44 1.64 log Organic 5.76 0.79 7.26 0.000 0.63 1.58 4.15 % MS 0.03 0.01 2.77 0.008 0.56 1.79 12.13 66

PAGE 67

Table 2-34. Multiple regression analysis for th e dependent variable sand shear resistance (R 2 = 0.49, F 1,73 = 18.8, p= 0.000). Model II was chosen as a general equation to predict shear resistance in any Florida beach. SE= standard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Condition index; CS= Coarse sand; MS= Medium sand. Durbin-Watson value was 1.22 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) -1.35 1.36 -0.99 0.325 1.00 Compaction 2.64 0.41 6.49 0.000 1.00 1.00 11.31 II (Constant) 2.55 1.52 1.68 0.098 1.00 Compaction 1.92 0.40 4.78 0.000 0.83 1.21 3.25 % CS -0.09 0.02 -4.34 0.000 0.83 1.21 15.58 II (Constant) 7.37 1.56 4.72 0.000 1.00 Compaction 1.19 0.37 3.24 0.002 0.72 1.40 3.53 % CS -0.11 0.02 -5.97 0.000 0.80 1.24 4.65 % MS -0.05 0.01 -5.45 0.000 0.86 1.16 20.53 Table 2-35. Multiple regression analysis fo r the dependent variable sand compaction (R 2 = 0.52, F 1,72 = 4.12, p= 0.046). Model III was chosen as a general equation to predict sand compaction in any Florida beach. SE= st andard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Conditi on index. Durbin-Watson value was 1.17 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 2.29 0.17 13.87 0.000 1.00 Shear 0.14 0.02 6.49 0.000 1.00 1.00 5.88 II (Constant) 2.07 0.16 13.26 0.000 1.00 Shear 0.11 0.02 5.73 0.000 0.92 1.09 4.24 % Water 0.07 0.02 4.33 0.000 0.92 1.09 7.05 III (Constant) 2.00 0.16 12.82 0.000 1.00 Shear 0.12 0.02 5.91 0.000 0.92 1.09 2.13 % Water 0.09 0.02 4.83 0.000 0.66 1.51 5.18 log Organic 0.44 0.22 2.03 0.046 0.70 1.43 7.57 67

PAGE 68

Table 2-36. Multiple regression analysis for the dependent variable sand bulk density (R 2 = 0.31, F 1,45 = 7.40, p= 0.009). Model II was chosen as a general equation to predict sand bulk density in east coast Florida beaches. SE= standard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Conditi on index. Durbin-Watson value was 1.34. Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 2.07 0.07 31.92 0.000 1.00 % Water -0.05 0.01 -3.37 0.002 1.00 1.00 6.25 II (Constant) 2.30 0.11 21.76 0.000 1.00 % Water -0.04 0.01 -3.05 0.004 0.97 1.03 6.68 log Shear -0.36 0.13 -2.72 0.009 0.97 1.03 12.51 Table 2-37. Multiple regression analysis for the dependent variable sand bulk density (R 2 = 0.62, F 1,24 = 9.51, p= 0.005). Model III was chosen as a general equation to predict bulk density in west coast Florida beaches. SE = standard error; Tol= Tolerance; VIF= Variance inflation factor; CI= Condition i ndex; VFS= Very fine sand. DurbinWatson value was 1.60 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 2.01 0.07 31.17 0.000 1.00 % Silt -0.01 0.00 -3.95 0.001 1.00 1.00 3.64 II (Constant) 1.99 0.06 32.46 0.000 1.00 % Silt 0.00 0.00 -2.62 0.015 0.76 1.32 2.05 % VFS -1.18 0.55 -2.14 0.043 0.76 1.32 4.44 III (Constant) 1.97 0.05 36.70 0.000 1.00 % Silt 0.00 0.00 -0.37 0.714 0.45 2.22 2.11 % VFS -1.63 0.50 -3.25 0.003 0.70 1.44 2.94 log Organic 0.35 0.11 3.08 0.005 0.59 1.70 6.06 68

PAGE 69

Table 2-38. Multiple regression analysis for th e dependent variable sand particle density (R 2 = 0.34, F 1,72 = 15.5, p= 0.000). Model III was chosen as a general equation to predict sand particle density in any Fl orida beach. SE= standard e rror; Tol= Tolerance; VIF= Variance inflation factor; CI= Condition i ndex; MS= Medium sand. Durbin-Watson value was 2.26 Collinearity Statistics Model SE t P Tol VIF CI I (Constant) 2.89 0.06 52.02 0.000 1.00 % MS 0.01 0.00 3.66 0.000 1.00 1.00 4.21 II (Constant) 2.99 0.07 42.88 0.000 1.00 % MS 0.00 0.00 3.08 0.003 0.94 1.07 2.52 log CaCO 3 -0.11 0.05 -2.14 0.036 0.94 1.07 5.76 III (Constant) 3.43 0.13 26.68 0.000 1.00 % MS 0.00 0.00 2.28 0.026 0.88 1.14 2.75 log CaCO 3 -0.24 0.06 -4.12 0.000 0.65 1.53 4.46 log Water -0.46 0.12 -3.94 0.000 0.69 1.44 12.80 69

PAGE 70

Table 2-39. Results of a Spearman correlation an alysis between sand physical parameters from 44 native and nourished sea turtle nesting beaches. Spearmans rho and significance level are presented only for statistically sign ificant correlations (n= 44). PD= particle density, Por= porosity, Comp.= sand co mpaction, Shear= sand shear resistance CO 3 PD Por Water Comp. Shear % Organic 0.88 -0.33 0.30 0.35 0.00 0.03 0.05 0.02 % CaCO 3 -0.32 0.31 0.03 0.04 Bulk Density -0.81 -0.40 -0.31 -0.36 0.00 0.01 0.04 0.02 Particle Density 0.35 -0.43 -0.31 0.02 0.00 0.04 % Porosity % Water 0.36 0.02 Compaction 0.79 0.00 Table 2-40. Results of a Spearman correlation an alysis between sand physical parameters from 44 native and nourished sea turtle nesting beaches. Spearmans rho and significance level are presented only for statistically significant correlatio ns (n= 44). FG= fine gravel, VFG= very fine gravel, VCS= very coarse sand, CS= coarse sand, MS= medium sand, FS= fine sand, VFS= very fine sand. FG VFG VCS CS MS FS VFS % Organic 0.31 0.42 0.58 0.04 0.00 0.00 % CaCO 3 0.39 0.49 0.63 0.01 0.00 0.00 Bulk Density 0.38 -0.46 -0.31 0.01 0.00 0.04 Particle Density -0.33 0.03 % Porosity 0.33 0.12 0.03 0.44 % Water -0.52 0.58 0.36 0.00 0.00 0.01 Compaction 0.39 0.46 -0.31 -0.61 0.36 0.68 0.01 0.00 0.04 0.00 0.02 0.00 Shear 0.30 0.32 -0.39 -0.54 0.44 0.56 0.05 0.03 0.01 0.00 0.00 0.00 70

PAGE 71

Table 2-41. Results of a Spearman correlation an alysis between sand physical parameters and 21 inorganic metals from 44 native and nour ished sea turtle nesting beaches. Spearmans rho and significance are presen ted only for statistically significant correlations (n= 44, except Ca and Fe where n= 43). Al As B Ba Be Ca Co Cr Cu % Organic 0.42 0.62 0.55 0.39 0.82 -0.41 -0.48 0.00 0.00 0.00 0.01 0.00 0.01 0.00 % CaCO 3 0.49 0.75 0.46 0.87 -0.45 -0.60 0.00 0.00 0.00 0.00 0.00 0.00 Bulk Density -0.32 -0.05 0.04 0.76 Particle Density -0.32 0.03 % Porosity 0.41 0.01 % Water 0.38 0.37 -0.29 0.01 0.01 0.06 Compaction 0.39 0.01 Shear 0.42 0.00 71

PAGE 72

Table 2-41. Continued Fe K Mg Mn Mo Pb Se Si Zn % Organic 0.45 0.52 0.84 0.50 0.84 0.76 0.58 0.00 0.00 0.00 0.00 0.00 0.00 0.00 % CaCO 3 0.59 0.47 0.84 0.63 0.86 0.78 0.51 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Bulk Density -0.36 -0.09 -0.33 -0.43 0.02 0.57 0.03 0.00 Particle Density -0.35 -0.43 -0.43 -0.37 -0.42 -0.31 0.02 0.00 0.00 0.01 0.00 0.04 % Porosity -0.38 -0.43 0.47 0.01 0.00 0.00 % Water 0.39 0.34 0.38 0.01 0.02 0.01 Compaction 0.54 0.00 Shear 0.36 0.55 0.02 0.00 72

PAGE 73

Table 2-42. Results of a Spearman correlation analysis between six sand grain size classes and 21 inorganic metals from 44 native and nour ished sea turtle nesting beaches. Spearmans rho and significance level are pr esented only for statistically significant correlations (n= 44, except for Ca and Fe where n= 43). Al As Be Ca Cd Co Cr Cu Fine Gravel 0.38 0.01 Very Fine Gravel 0.45 0.37 -0.33 0.00 0.02 0.03 Very Coarse Sand 0.32 0.41 0.53 -0.34 -0.32 0.03 0.01 0.00 0.03 0.03 Coarse Sand 0.31 0.04 Medium Sand -0.32 0.04 Fine Sand Table 2-42. Continued Fe Mg Mn Mo Pb Se Si Zn Fine Gravel 0.31 0.40 0.04 0.01 Very Fine Gravel 0.32 0.34 0.43 0.46 0.03 0.03 0.00 0.00 Very Coarse Sand 0.30 0.38 0.29 0.46 0.36 0.44 -0.35 0.05 0.01 0.05 0.00 0.02 0.00 0.02 Coarse Sand -0.32 0.03 Medium Sand -0.35 0.02 Fine Sand 0.40 0.01 73

PAGE 74

Discussion This research project surveyed nourished a nd native Florida beaches to identify what sand physical properties best characterize these two very distinct habitats. This characterization provided information on how the intrinsic properties of these beaches can influence sea turtle nesting ecology. Five factors are responsible for most of the variability between these two types of beach and these are: Florida coast, planform width and length, sand characteristics, and age of nourishment project. Because of the complex inte ractions between them, th ey have potential to influence all stages of se a turtle reproductive ecology. A nourished beach is a dynamic environment that is transitioning from an elevated flat planform to a beach with an inclined profile. This transformation occurs through continuous wave action in the swash zone where an escarp ment often forms. As the project ages, and depending on factors such as sand compaction an d intensity of wave action, the escarpment moves landward until it is assimilated into the beach profile (Nelson, 1985; Nelson and Mauck, 1986; Magron, 2000). In wider nourished beaches this assimilation is delayed to reach regions high on the berm crest and as a consequence, plan form equilibration can take several years to complete (Nelson and Mauck, 1986). Therefore, when describing the physical properties of a nourished beach, it is important they be refe renced to project age and beach width. When assessing the efficacy of a nourishment pr oject in regards to its impact on sea turtle reproduction, data such as nesting or hatching su ccess should be factored and analyzed according to proximity to escarpment. The escarpment separates homogeneous nourished sand from a planform with layered sand th at resembles a native beach. These two environments differ primarily in sand water content and percentage of larger sand grains th at consist mostly of broken shell material. 74

PAGE 75

Percentage of water in sand is influenced by matric, osmotic, and gravitational potentials that combined are known as the water potential of a beach. Water potential measures differential sand suction, the force that determines direct ion and flow of its wate r (Milton and Lutz, 2003). This component is extremely important because it can interfere with sea turtle nest excavation, egg chamber integrity, egg water potential, and incubation duration (Yntema and Mrosovsky, 1980; Magron, 2000; Ehrhart and Roberts, 2001; and Milton and Lutz, 2003) Nest excavation and egg chamber integrity can de disrupted if beach sand is very dry and has a negative water potential. When a sea turtle digs in dry l oose sand, lack of grain cohesiveness extends the nest ex cavation process and makes it difficult to shape an egg chamber (Milton and Lutz, 2003). Water potential of b each sand can also interfere with the water potential of sea turtle eggs. Higher (positive) sand water potential causes eggs to uptake water and swell. Conversely, lower (negative) sand wa ter potential will dehydra te the clutch (Milton and Lutz, 2003). Incubation time can be altered by water potential because when sand retains more water it increases its heat capacity which wi ll require more energy to maintain or raise its temperature (Farouki, 1986; Harrison and Morris on, 1993). This heat flux affects incubation temperature and can change its duration as we ll as skew sex ratios (Yntema and Mrosovsky, 1980; Mrosovsky and Yntema, 1981). Sand water can influence sea turt le nesting and incubation either directly or indirectly because it is correlated to other sand components. For example, the effect of sand water can be influenced (negative correlation) by the concentration of total sa nd carbonate. Sand carbonate is correlated to organic content that in turn, is influenced by perc entage of larger sand grains. These relationships show how the impact of one sand parameter can be cofounded by many other sand properties. Biologists must pay careful atte ntion to these relationships, described in Tables 75

PAGE 76

2-22 through 2-38 when evaluating what sand propert ies were responsible for data collected in their monitoring beach. Florida native beaches are very diverse, but this diversity is primarily attributed to geography and not necessarily because of sand characteristics. Differences between sand properties of native planforms are less promin ent and are dependent on degree of profile inclination. Steepness of a beach will determine th e distribution and sorting of its sand layers that in turn, influence most parame ters studied in this project. As in nourished beaches, biologists should m onitor water content and the proportion of larger sand grains (>1mm) because these parameters have greater potential to influence sea turtle nesting ecology in a native beach. Water cont ent will vary across beach width depending on planform inclination. Because of the correlations between sand pa rameters, percentage of larger sand grains will influence compaction, bulk density, porosity, and ultimately sand water potential. Table 2-40 shows relationships betw een these parameters and Tables 2-22 through 238 include regression formulas to calculate missing data. Sea turtle nests located near the dune base can be exposed to very different sand properties than those midbeach or low by the water. Alt hough intuitive that wate r content will be higher near the swash zone, differences in sand layering ar e not as obvious. Sea tu rtle biologists should characterize each sand layer acro ss their monitoring beach to understand the microenvironment at clutch depth. Beach productivity should be co mpared according to these microenvironments. It is important that these di fferences be pointed out so th at biases can be avoided. Native and nourished beaches should not be comp ared between Florida coasts because east and west coasts are very distin ct. East coast is subjected to more energetic wave action and erosional forces while Gulf of Mexico wate rs are normally calmer a nd less destructive. 76

PAGE 77

Differences in degree of wave action and sand mixing impact shorelines differently and are reflected in their respective beach sand properties. The sand characteristic most divergent between coasts is sand compaction and west coast beaches are much more compact. In fact, nour ished east coast beaches are more similar in compaction to west coast native beaches than they are to east coas t native beaches. Sand compaction is very important to sea turtle ne sting ecology because beaches with very compact sand will negatively impact all stages of sea turt le nesting ecology. Hard sand can deter nesting individuals from digging and increase the numbe r of non-nesting events (Nelson et. al., 1987; EAI, 1999; Ehrhart and Roberts, 2001). Clutch es deposited in hard sand can have lower hatching success because compaction can slow or prevent hatchlings from emerging thus lowering the productivity of a beach (Nelson et. al., 1987; EAI, 1999). In order to reduce sand compaction nourishment projects are often required to soften sand by tilling. However, this should be done early an d before the beginning of the nesting season so the planform has adequate time to settle. When comparing the effect sand compaction has on sea turtle nesting and incubation between east and west coasts beaches, compaction values and specific sand characteristics such as percentage of different gr ain size classes should be used instead of generalized beach types. Two other beach properties that can influence sea turtle nesting ecology and whose effects can be cofounded and difficult to isolate from coast or beach type are sand organic content and sand carbonate. Percentage of organic material in sand was very similar in native beaches regardless of coast, however, this trend was not observed in nourished sand. Nourished beaches had higher total organic loads than native, and among nouris hed beaches east coast sand had much higher 77

PAGE 78

values than west coast. Organic matter can accumulate easily on east coast shoreline because stronger oceanic waves and currents transport floating macroalgae to the beach planform. However, since high organic concentrations are not observed in the native east coast beaches sampled, organic load in nourished beaches must be attributed primarily to a byproduct of the nourishment project and secondarily to wave action. This indicates that beach nourishment projects on the east could be contributing to higher sand organic concentrations. Organic content includes all organic matter in sand including invertebrates and macroalgae that can be ina dvertently dredged from the ocean bottom and transported to beach planforms during the nouris hment process. Although dredging operations often rinse sand prior to its final placement on land, these sands inherently will have a certain amount of organismal bycatch. Once deposited on the beach, their decomposition increases concentrations of hydrogen sulfides, disulfites, and methanes in the incubation medium that can affect healthy gaseous environments of sea tu rtle clutches (Mota a nd Peterson, 2002; 2003). Calcium carbonate is a very dynamic factor in sea turtle nesting ecology and its concentration in beach sand can be detrimental or beneficial. Detrimental impacts are primarily related to increased sand compactness and cementa tion that reduces gas diffusion and the ability of hatchlings to emerge (Ackerman, 1981; Ehrhart and Roberts, 2001). It can also interfere with the formation of carbonic acid that softens eggs hells and aids pipping (Bustard and Greenham, 1968). Beneficial impacts are related to its ro le in buffering high concentrations of carbon dioxide, and as an embryonic sour ce of calcium (Carthy, 1996; Mota and Peterson, 2002; 2003). Percentage of total calcium carbonate in native beach sand was higher on west than on east coast beaches sampled. This increased concentra tion can be attributed to the abundant natural accumulation of seashell debris on west coast shor elines. However, carbonate can also originate 78

PAGE 79

from other sources because nourished beaches show a very different trend with east coast having higher concentrations th an west coast. Although the Florida Department of Environmental Protection agency has specifications regarding total calcium carbonate allowed in be ach nourishment fill, its concentration can be positively influenced by the amount of organic ma tter and percentage of larger sand grains. Other aspects of sand carbonate that are important to know are its physical form (powder, hard clumps, seashell debris) a nd reactive properties. Concentrations and distribution of inorganic metals in beach sand (Appendix B) revealed patterns that could be problematic for sea turtle incubation. One metal whose concentration stands out is arsenic. Arsenic is a metal of sp ecial concern because it can lead to circulatory and neurological disorders, can cause mutagenic, car cinogenic and teratogenic e ffects, and be lethal in relatively small concentra tions (Storelli et al., 1998). However, arsenic toxicity depends primarily on its oxidation state and molecular form, and four species of arsenic are present in seaw ater: arsenate, arsenite, methylarsonate and dimethylarsinate (Storelli et al., 1998). The dom inant form of arsenic in marine waters is arsenate (As V), and the more toxic and potenti ally carcinogenic arsenite (As III) rarely accounts for more than 20% of total arsenic in seawat er (Kubota et al., 2002). Arsenate dominates in oxidized marine sediments and can be reduced to arsenite or to less toxic organoarsenics (Storelli et al., 1998). Arsenic is present in higher concentrations in nourished sand and in beaches located in southwest Florida (Appendix B). However, this project is not implying that nourished sand is contributing directly to these concentrations. Instead, it appears that arsenic is cofounded in nourished beach concentrations of or ganic matter and calcium carbonate. 79

PAGE 80

Organics and calcium carbonates are a byproduc t of the sand dredging process that can deposit marine organisms and seashell fragments on a beach planform. Native beaches in the southwest have high concentrations of sand carbonate as well as abundant seashells. Marine animals such as algae, bivalves, gastropods, a nd crustaceans accumulate varying concentrations or arsenic, however, its toxicity depends on th e chemical forms which the element occurs in tissues and the biogeochemical cycle it has in the marine environment (Lunde, 1977; Storelli et al., 1998). Carthy (1996) showed that the e ggshell is an important source of minerals such as calcium, to the developing embryos. This mineral transfer is facilitated when the eggshell is in direct contact with sand. Therefor e, it is plausible that if sand contai ns high concentrations of arsenic, that the eggshell could be a vector for the transf er of this analyte into the egg. Research on heavy metal concentrations in loggerhead and gr een sea turtle eggs s howed that the yolk contained the highest concentratio ns and burdens of heavy metals (Sakai et al., 1995). Because yolk provides nutrition for developing embryos, the eggshell serve as a pathway for sand arsenic to negatively impact the embryogenesis of sea turtle hatchlings. It is important that future research investig ates two important para digms related to heavy metals in sand. First, what are the conditions ne cessary for heavy metals, such as arsenic, to leach from seashell material to beach sand. Second, what are the pathways that induce absorption of sand arsenic into sea turtle eggs an d its subsequent transfer to developing embryos. 80

PAGE 81

CHAPTER 3 A HUNDRED TURTLE EGGS ON THE NOURISHED BEACH Introduction In So Excellent a Fishe, Ca rr (1967) writes an essay titled A Hundred Turtle Eggs to discuss sea turtle nesting ecology and the evolutiona ry adaptations that led to an optimal clutch size. Carr refers to biological constraints such as we ight of eggs in the turtle belly, amount of yolk per egg, and cost of predation that led to the selection of a clutch size that maximizes reproductive productivity. This chapter examines how the incubation of a hundred eggs is specifically impacted by beach nourishment projects through focusing on sa nd compaction. Compaction was identified in Chapter 2 as the main principal component in Florida beach sand, and data showed that nourished as well as west coast beaches usually have higher sand compaction than native or east coast beaches. Although higher sand compaction has potential to impact different aspects of sea turtle nesting and incubation, my objective is to assess its effects on incubation concentrations of oxygen and carbon dioxide. The gaseous environment of loggerhead s ea turtle egg incuba tion was described by Ackerman (1977), using nests in situ and in artificial incubators. Respiratory gas cycles and egg shell permeability were measured and related to embryonic metabolic activity (Ackerman, 1980; 1981). Data show that during the first half of incubation, gas concen trations of oxygen and carbon dioxide remain very similar to those in adjacent sand media (Ackerman, 1977). During this stage of development, embryos are very small and their correspond ing metabolic activity sufficiently low that embryonic oxygen consum ption is not restricted by egg shell gas conductance (Ackerman, 1981). 81

PAGE 82

As metabolic activity increases during the second half of incubation, oxygen demand rises peaking prior to hatching. Conc omitantly, respiratory carbon dioxi de levels increase within the clutch throughout incubation and also peak pr ior to hatching (Ackerman, 1981). Therefore, developing sea turtle embryos become increas ingly dependent on egg shell permeability to uptake oxygen and offload carbon dioxide. This exchange occurs across the parchment egg shell and between chorioallantoi c capillaries and nest gas (Ackerman and Prange, 1972). Nest gas refers to air space between individual eggs and will vary depending on clutch size, nest architecture and nest site selection. These factors are paramount because they will affect gas partial pressure gradients between the center and periphery of the clutch as well as between clutch and beach sand (Ackerman, 1975). Nests with larger clutches have higher demand for respiratory gas exchanges and establis h greater gas flux rates between beach sand air and clutch as metabolic activity increas es throughout incubation (Ackerman, 1980). Beach sand air, or sand porosity, is very important to sea turtle egg incubation. Porosity is the percentage of air in a known volume of sand (Tan, 1995) and is influenced by sand grain size, compaction, and percentage of water content (Tan, 1995). Sand with fi ner grains contains a larger volume of air than that with coarser grains However, as data from Chapter 2 show, finer grain sand also holds more water and is more compact. Increased compaction has potenti al to restrict sand gas di ffusion rates. Therefore, I hypothesize that gas diffusion between a sea tur tle clutch and atmosphere are negatively impacted in beaches with high sand compacti on. Lower gas diffusion rates can concentrate carbon dioxide and decrease oxygen concentrations within clutches, which can slow growth and increase embryo mortality (Ackerman, 1975; 1980). 82

PAGE 83

Materials and Methods Nourished and native beaches were chosen from both Florida coasts. East coast beaches were: Melbourne Beach (native and nourished sections), Patrick Air Force Base Beach (nourished), the Merritt Island National Wildlife Refuge Beach (native), and Canaveral National Seashore (native). West coast beaches were: An na Maria Island beaches (native and nourished), Parkshore Beach (native and nourished) and a nour ished section of Vanderbilt Beach in Naples. A total of 63 nests were studied from 20012004 (Table 3-1). Beach sand compaction data were used to separate nests into two experimental groups. There were 34 nests marked in beaches with high sand compaction and 29 in be aches with soft sand. High compaction was defined as sand with values > 1.5 psi and soft sand with < 1.5 psi, as measured with a pocket penetrometer. All nests used for this project we re located at least one meter away from heavy vegetation to avoid any negative interaction of plant roots as well as extraneous cellular respiration gases. Nesting loggerhead sea turtles were allowed to lay their nests undisturbed. The clutch was then counted and a subsample of 10 eggs was cleaned, weighed and their minimum and maximum diameters measured. These eggs were re turned to the nest a nd at approximately the 50 th egg, a 5 mm Teflon air-sampling tube was positio ned in the middle of the clutch. This tube was extended subterraneously one meter away fr om the clutch and was used to sample gas concentrations weekly. A Tidbit (Onset Corp.) temperature datalogger, programmed to record hourly temperatures, was deployed in the center of the clutch. A control nest was dug one meter away from the experimental nest. Dimensions for the egg chamber replicated those of the experimental nest, and golf balls were used in lieu of eggs. Golf balls were selected to simulate loggerhead sea turtle eggs because they have similar 83

PAGE 84

Table 3-1. Loggerhead sea turtle nests used to study incubation c oncentrations of carbon dioxide and oxygen in beaches with low (psi 1.5) and high sand compaction (psi > 1.5). Native and nourished beaches had nests in both experimental groups although west coast and nourished beaches had a higher number of nests in high compact sand. Beaches are: KSC, Kennedy Space Center ; PAFB, Patrick Air Force Base; MB, Melbourne Beach; VB, Vanderbilt Beach; PS, Park Shore Beach; AMI, Anna Maria Island. Compaction was measured with a poc ket penetrometer as pounds per square inch (psi). Year Low compaction High compaction 2001 12 KSC (native) 10 PAFB (nourished) 2002 5 KSC (native) 3 MB (nourished) 2003 6 MB (native) 5 MB (nourished) 2 MB (nourished) 3 MB (native) 2 VB (nourished) 3 PS (nourished) 2 PS (native) 2004 1 KSC (native) 6 AMI (nourished) 1 AMI (native) 2 KSC (native) Total 29 34 84

PAGE 85

diameters and weight. The number of golf balls us ed was equal to the number of eggs in each corresponding clutch. A second air-sampling tu be and Tidbit temperature datalogger were positioned at the center after the 50 th golf ball was dropped in. Air-sampling tubes were labeled, capped, their standing air volume calculated, and buried 8 cm below sand surface. Nest locations were marked with visible si gns on some beaches while on others their location was measured to hidde n stakes in the vegetation. Concurrently, data for beach compaction were measured at each nest location. Weekly samples of air were taken by first evacuating the standing ai r from the tubes and then extracting 10 ml using a gas ti ght syringe. Care was taken not to walk or stand directly on top of experimental or control nests. Air samples were transferred to 3" X 5" Call-5 Bond gas sampling bags made of five layers: polyester, polyvinylidenene chloride, aluminum foil, polyamide, and high-density polyethylene. Bags were fitted with a Luer stopcock and 0.5 mm septa for easy connection to the sampling syringe. Samples were analyzed for their concentrations of oxygen, carbon dioxide, hydrogen sulfide, and methane by gas chromatography. Oxygen and carbon dioxide were indicators of embryonic de velopment while hydrogen sulfide and methane were indicators of hatching and rotting eggs, respectively. Samples were quantified for carbon dioxide oxygen, methane and hydrogen sulfide by injecting 1ml samples with glass gas-tight syri nges into gas chromatographs. Carbon dioxide analysis was completed with a 6890 Hewlett Packard Gas Chromatograph with Thermal Conductivity Detector and a 30 m x 0.53 J&W GS -Q column. Oxygen and methane were analyzed on a 5880 Hewlett Packard Gas Chroma tograph with Thermal Conductivity Detector and a packed, 9X18 molecular sieve column. Hydrogen sulfide analysis was performed on a 85

PAGE 86

portable Photovac (Perkin Elmer) 10S Plus Gas Chromatograph with Photoionization Detector and a 15 m x 0.53 mm KCl/AlO2 column. All analyses were duplicated and the standard deviation between both samp les was less than 3%. After hatchling emergence, nests were excavated and their contents i nventoried. The number of hatched and unhatched eggs, live and dead hatchlings, and pipped eggs were counted. Unhatched eggs were opened and any discernabl e embryo was aged according to loggerhead sea turtle developmental stag es (Miller, 1982). Time of hatch ling death was important to know so that the number of live eggs in the incubating clutch could be corre lated to incubating gas levels. Percent hatching success (HS) and percent hatchling emergence success (HE) were calculated using the following: HS= (hatched eggs / total clutch) X 100 HE= (hatched eggs live hatchlings dead hatchlings)/ total clutch X 100 Data analyses were performed using SPSS v.12 st atistical program software. Descriptive statistics as mean, standard deviation, standard error, variance, minimum, and maximum were calculated for nest data from each sand treatment. The non-parametric Kruskal-Wallis analysis of variances (ANOVA) was used to compare differences in nest dimensions such as depth of nest neck, neck length, neck width, depth of nest chamber, chamber length, chamber width, distance from surface to top of eggs as well as beach sand compaction between nests from the soft and high sand compact nests. A similar ANOVA compared differences in clutch size, egg dimensions (minimum, maximum and average diameters) and egg weight between ne sts from soft and high sand compaction. 86

PAGE 87

Concentrations of carbon dioxide and oxyge n from each nest were averaged, log transformed and grouped according to treatment. A time series was plotted for each gas and trend lines derived for each sand compaction. Non-parametric paired t-tests (two-tailed) analyzed for diffe rences in concentrations of oxygen and carbon dioxide through the 8 weeks of incubation in both sand treatments. The nonparametric Kruskal-Wallis analysis of varian ces (ANOVA) compared di fferences in percent hatching success and emergence success be tween nests from both sand treatments. Results Descriptive statistics for nest data from each sand treatment are presented in Table 3-2. ANOVA results (Table 3-3) from the compar ison of nests from low (n=29) and high (n=34) sand compaction show statistically signi ficant differences for nest chamber length ( F 1,61 = 5.79, p= 0.019), distance from surface to top of eggs ( F 1,61 = 8.99, p= 0.002), and sand compaction ( F 1,61 = 138.0, p= 0.000). No statistically sign ificant differences in clutch size and egg dimensions were found between treatments and descriptive data are presente d in Table 3-4. Concentrations of oxygen (Table 3-5) and carbon dioxide (Table 3-6) showed statistically significant differences in oxygen ( t (32) = 3.2, p = 0.003), and carbon dioxide ( t (33) = 4.64, p = 0.000) concentrations. Time series analyses for carbon dioxid e and oxygen are in Figures 3-1 and 3-2, respectively. Concentrations of both gases re main very similar to atmospheric levels until incubation day 21 when average carbon dioxide levels in nests from high compact sand are double (2,831 ppm) the average from soft sand ne sts (1,262 ppm). Differences in carbon dioxide concentrations in high sand compaction continue to be more prominent during the fourth week and reach 14,585 ppm by day 35 (Table 3-6). Ne sts in softer sand also have carbon dioxide 87

PAGE 88

concentrations that increase, however, their le vels only reach 5,277 ppm (T able 3-6). At this time, oxygen concentrations also begin to show di fferences between sand treatments. Oxygen in nests from high sand compaction decreases to 18.4%, while in low sand compaction it stays fairly similar to atmosphere (Table 3-5). Clutch gas concentrations cont inue to increase through the sixth week of incubation and by day 49 (pip ping), nests in high sand compaction have peak carbon dioxide concentrations that average 30,777 ppm while those in soft native sand average 14,177 ppm (Table 3-6). Oxygen concentrations also continue to decrease and at day 49 nests in high sand compaction average 17.8%, while those in softer sand average 19.0% (Table 3-5). Trend equations for carbon dioxide are (y = 0.27 log(x) + 2.52) in low and (y= 0.46 log(x) + 2.28) in high sand compaction. Oxygen trend e quations are (y = -0.37 log(x) + 21.37) in low and (y= -1.05 log(x) + 22.43) in high sand compaction. Descriptive statistics of pe rcent hatching success and emergence success data measured from the two treatments are in Table 3-7. Stat istically significant difference (Table 3-8) was found for emergence success ( F 1,55 = 6.75, p= 0.012). 88

PAGE 89

Table 3-2. Descriptive statistics of nest physical dimensions (cm) measured from two treatments of low (n= 29) and high sand (n= 34) comp action. Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Compaction was measured with a pocket penetrometer as pounds per square inch (psi). Parameter Mean SD SE Var Min Max Soft Sand Nest chamber depth 52.7 6.12 1.53 37.5 40.0 63.0 Surface to nest neck 33.5 5.48 1.37 30.1 23.0 42.0 Neck length 20.7 2.90 0.72 8.46 16.0 26.0 Neck width 19.2 3.10 0.77 9.67 15.0 26.0 Nest chamber length 26.0 2.82 0.71 8.00 20.0 31.0 Nest chamber width 24.4 3.10 0.77 9.60 17.0 29.0 Surface to eggs 32.4 5.80 1.45 33.7 23.0 40.0 Sand compaction (psi) 0.97 0.49 0.12 0.25 0.10 1.50 Hard Sand Nest chamber depth 48.9 6.44 1.31 41.5 37.0 66.0 Surface to nest neck 30.7 5.56 1.13 30.9 22.0 45.0 Neck length 20.6 3.21 0.65 10.3 15.0 25.0 Neck width 18.2 2.73 0.56 7.50 14.0 24.0 Nest chamber length 24.3 3.38 0.70 11.4 18.0 30.0 Nest chamber width 23.1 4.66 0.95 21.7 16.0 35.0 Surface to eggs 26.4 7.81 1.59 61.1 8.00 46.0 Sand compaction (psi) 2.33 0.63 0.13 0.40 1.54 4.00 Table 3-3. Results of a Kruskal Wallis ANOVA co mparing nest cavity architecture data from nests in low (n=29) and high (n=34) sand compaction. Only parameters that were statistically significant are shown. Desc riptive values for each parameter are presented in Table 3-2. Sum of Mean Squares Square F 1,61 P Nest Chamber Length 94.6 94.6 5.79 0.019 994.6 16.6 1089 Surface to eggs 505.8 505.8 8.99 0.002 3316 56.2 3822 Sand Compaction 39.4 39.4 138 0.000 17.4 0.28 56.8 89

PAGE 90

Table 3-4. Descriptive statistics of clutch egg data measured from nests of two treatments of low (n= 29) and high sand (n= 34) compaction. Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Diameters are measured in millimeters and weight in grams. Parameter Mean SD SE Var Min Max Soft Sand Diameter 40.2 1.45 0.30 2.13 37.8 44.3 Minimum diameter 39.0 1.40 0.28 1.97 37.0 42.5 Maximum diameter 41.4 1.57 0.32 2.48 38.6 46.1 Weight 39.4 3.99 0.81 15.9 33.2 51.2 Clutch size 115.8 22.6 5.67 514 67.0 165 Hard Sand Diameter 39.4 2.29 0.39 5.24 31.7 43.7 Minimum diameter 37.9 2.48 0.42 6.17 30.4 42.3 Maximum diameter 40.8 2.21 0.38 4.92 33.0 45.1 Weight 39.2 5.15 0.88 26.5 30.5 51.1 Clutch size 118.3 18.5 3.77 341 79.0 148 90

PAGE 91

0.0 1.0 2.0 3.0 4.0 5.0 6.0 068141621242628303335384143454850525456586163Incubation DayLog CO2 >1.5 PSI <1.5 PSI Figure 3-1. Time series of log transformed carbon dioxide concentr ation data from nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5). Trend lin es were derived for each sand treatment of low (y = 0.27 log(x) + 2.52; R 2 = 0.32) and high (y = 0.46 log(x) + 2.28; R 2 = 0.64) compaction. 91

PAGE 92

15.0 16.0 17.0 18.0 19.0 20.0 21.0 22.0 23.0 24.0 06810141619212426283033353841434548505254565860 Incubation Day% Oxygen >1.5 PSI <1.5 PSI Figure 3-2. Time series of oxygen concentration data from nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5). Trend lines were derived for each sand treatment of low (y = -0.37 log(x) + 21.37; R 2 = 0.09 and high (y = -1.05 log(x) + 22.43; R 2 = 0.34) compaction. 92

PAGE 93

Table 3-5. Average incubation concentrations of oxygen from loggerhead s ea turtle nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with high sand compaction (psi > 1.5). Data are presented as daily % averages and log 10 transformed. Incubation Day ( 1.5 psi) (>1.5 psi) % O 2 log 10 % O 2 log 10 0 21.0 1.32 21.0 1.32 5 20.9 1.32 6 21.3 1.33 21.1 1.32 7 20.9 1.32 21.0 1.32 8 18.1 1.26 18.9 1.28 9 21.9 1.34 10 21.6 1.33 13 21.8 1.34 21.3 1.33 14 21.5 1.33 20.4 1.31 15 19.3 1.28 16 20.8 1.32 17 21.4 1.33 19 21.2 1.33 21.3 1.33 20 19.4 1.29 20.5 1.31 21 21.2 1.33 20.3 1.31 22 21.6 1.33 19.6 1.29 24 19.9 1.30 21.4 1.33 25 21.0 1.32 26 19.4 1.29 21.2 1.33 27 20.1 1.30 20.6 1.31 28 20.3 1.31 19.7 1.30 29 21.0 1.32 20.1 1.30 30 20.6 1.31 19.7 1.29 32 19.0 1.28 33 20.1 1.30 19.6 1.29 34 19.0 1.28 19.0 1.28 35 19.2 1.28 18.4 1.26 36 20.4 1.31 19.4 1.29 38 19.0 1.28 40 20.1 1.30 19.6 1.29 41 20.0 1.30 19.6 1.29 42 21.0 1.32 18.3 1.26 43 17.7 1.25 44 17.0 1.23 45 21.4 1.33 16.5 1.22 47 20.0 1.30 18.9 1.28 48 19.0 1.28 18.0 1.26 93

PAGE 94

Table 3-5. Continued Incubation Day ( 1.5 psi) (>1.5 psi) % O 2 log 10 % O 2 log 10 49 19.0 1.28 17.8 1.25 50 18.9 1.28 17.7 1.25 51 19.2 1.28 15.4 1.19 52 19.0 1.28 16.0 1.20 53 18.9 1.28 17.0 1.23 54 19.0 1.28 17.0 1.23 55 18.6 1.27 18.8 1.27 56 20.5 1.31 57 23.0 1.36 58 20.5 1.31 59 21.0 1.32 60 21.5 1.33 94

PAGE 95

Table 3-6. Average incubation concentrations of carbon dioxide (parts per million) from loggerhead sea turtle nests (n=29) in beaches with low sand compaction (psi 1.5) and nests (n=34) from beaches with hi gh sand compaction (psi > 1.5). Data are presented as daily ppm averages and log 10 transformed. Incubation Day ( 1.5 psi) (>1.5 psi) CO 2 ppm log 10 CO 2 ppm log 10 0 300 2.48 300 2.48 5 601 2.78 6 690 2.84 681 2.83 7 842 2.93 939 2.97 8 1118 3.05 1624 3.21 9 816 2.91 10 747 2.87 13 1203 3.08 1432 3.16 14 770 2.89 15 840 2.92 16 1141 3.06 3660 3.56 17 1227 3.09 19 1092 3.04 1909 3.28 20 1456 3.16 2297 3.36 21 1262 3.10 2831 3.45 22 1594 3.20 24 2739 3.44 1777 3.25 25 2033 3.31 2462 3.39 26 2332 3.37 5347 3.73 27 2067 3.32 4497 3.65 28 2918 3.47 3294 3.52 29 1575 3.20 2942 3.47 30 3550 3.55 7324 3.86 32 5149 3.71 33 1160 3.06 6661 3.82 34 4460 3.65 8586 3.93 35 5277 3.72 14585 4.16 36 2830 3.45 9017 3.96 38 11330 4.05 40 1027 3.01 22876 4.36 41 8233 3.92 18748 4.27 42 10484 4.02 25936 4.41 43 11513 4.06 44 15741 4.20 45 2081 3.32 14578 4.16 47 6644 3.82 48 6550 3.82 21509 4.33 95

PAGE 96

Table 3-6. Continued Incubation Day ( 1.5 psi) (>1.5 psi) CO 2 ppm log 10 CO 2 ppm log 10 49 14177 4.15 30777 4.49 50 11041 4.04 18381 4.26 51 4869 3.69 15894 4.20 52 17457 4.24 11519 4.06 53 9614 3.98 3828 3.58 54 2785 3.44 55 4802 3.68 56 4034 3.61 20276 4.31 57 3637 3.56 58 1875 3.27 59 3467 3.54 1580 3.20 60 61 1027 3.01 96

PAGE 97

Table 3-7. Descriptive statis tics of percent hatching success and emergence success data measured from nests of two treatments of low (n= 29) and high sand (n= 34) compaction. Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Parameter Mean SD SE Var Min Max Soft Sand Percent hatching success 85.6 11.8 2.41 140 54.1 95.8 Emergence success 85.2 11.8 2.41 140 54.1 95.8 Hard Sand Percent hatching success 76.2 26.3 4.57 689 0.0 99.3 Emergence success 66.8 33.0 5.74 1090 0.0 99.3 Table 3-8. Results of a Kruskal Wallis ANOVA comparing differences between percent hatching success and emergence success data measured from nests of two treatments of low (n=29) and high (n=34) sand comp action. Descriptive values for each parameter are presented in Table 3-5. Sum of Mean Squares Square F 1,55 P Percent hatching success 1242 1242 2.70 0.106 25296 460 26539 Emergence success 4680 4680 6.75 0.012 38096 693 42776 97

PAGE 98

Discussion Sea turtle nests in nourished beaches with high sand compaction (> 1.5 psi) had dimensions that were different from those in so ft native sand. These diffe rences were in nest chamber length and distance from sand surface to t op of eggs. Data from a similar comparison of nests from nourished and native sands in Mel bourne Beach, Florida, showed that nests in highly compact sand had smaller arch itectural dimensions (Carthy, 1996). Nest architecture is important to sea tu rtle egg incubation because it creates a subterraneous environment where carbon dioxide and oxygen concentrations diffuse between eggs and clutch air, and between clutch air a nd atmosphere. Nests in high sand compaction had a mean nest chamber length smaller (24.3 cm) than those in softer sand (26.0 cm). This allows fewer numbers of eggs in the nest chamber and dist ributes part of the clutch higher into the nest neck. Experimental data support this statement and mean distance from sand surface to top of eggs in soft native sand was 33.5 cm while in compact nourished sand 30.7 cm. Although the top clutch layer in nourished beaches with high sand compaction is closer to the sand surface, it does not necessarily suggest th at incubation gases diffuse faster within the clutch or between clutch and atmosphere. This is primarily because of two factors affecting incubation: high sand compaction and smaller air space per egg ratio. High sand compaction, commonly associated with nourished beaches, can hinder gas diffusion rates through beach planforms. This re striction slows exchange of incubation gases between clutch air and atmosphere that can in crease clutch mortality and lower reproductive success. Mean percent hatchling emergence in na tive soft sand was higher than that in nourished high compact sand (Table 3-7). Smaller air space per egg ratio refers to tota l air volume that surrounds a clutch. Air volume is created by spaces between individual eggs and is proportiona te to clutch size. This air 98

PAGE 99

is very important to incubation because it is the source of oxygen and where metabolic carbon dioxide is offloaded by developing embryos. Nests in soft native sand have larger egg chambers that can contain a greater number of eggs. Therefore, clutches can normally fit completely within the egg chamber and individual eggs will be exposed to a greater volume of cl utch air per egg ratio. Depending on biological clutch variations, sea turtle nests dug in nourishe d beaches may also have space to accommodate their clutches in the egg chamber. However, because the egg chamber of compact nourished beaches is smaller than that of soft native beaches, it will have an overall lower egg capacity. All compact nourished sand nests from this research project had total clutch volumes that were greater than their egg chamber volume. It is important that clutches fit entirely inside their egg chambe rs because clutch air diffuses outwardly from its center to periphery. This concentric diffusion creates a differential pressure inside the clutch and facilitates exch ange of incubation gases between nest and sand media (Ackerman, 1975). When this differential pr essure is altered, such as when part of the clutch is in the nest neck, it affects the diffusi on gradient of incubation gases. Therefore, sea turtle nests in high compact sand have clutches exposed to different air diffusion dynamics than those in soft native sand because of their smalle r egg chambers. These differences can slow the incubation gas exchange rate between clutch air and beach sand, and ca n potentially have a negative impact on hatching success. Weekly concentrations of carbon dioxide (Fig ure 3-1) and oxygen (Figure 3-2) from the two sand treatment nests were statistically different It is important to understand when and why differences appear during incubation. 99

PAGE 100

The first indication that incubation gas con centrations of oxygen and carbon dioxide are different between nests in the two sand treatments is seen at day 21. Ca rbon dioxide in high sand compaction nests is double that from soft sand ne sts. This period corresponds to stage 23 where embryos measure approximately 1.4 cm in stra ight carapace length, their carapace is almost formed, and irises develop their dark pigmentation (Miller, 1982). Stage 23 embryos are very small and their metabolic rate alone cannot ex plain this increase (A ckerman, 1980). Factors related to the diffusion rate for carbon dioxide in sand must account for its higher concentration in high sand compaction. This trend continues through incubation and it is during the fifth week that negative impacts of increased sand compaction on the diffusion of both gases are more prominent. At this stage hatchlings completed most of their deve lopment and metabolic rates are high (Ackerman, 1980; Miller, 1982). Incubation gases must be exchanged between chorioallantoic blood and clutch air, and secondarily betw een clutch air and beach sand. If beach sand restricts diffusion, it concentrates gases in the clutch forcing a greater differential pressure gradient between the inside and outside of eggs. For example, by incubation day 35 average carbon dioxide in clutch air of high sand compaction nests averages 14,585 ppm (Table 3-6). This indicates that carbon dioxide inside eggs must be higher so that diffusion is outwardly. Conversely, if oxygen clutch levels are 18.4%, oxygen inside eggs must be lower for it to diffuse inwardly. Although similar patterns is observed in low sand compaction nests, the differe ntial pressure gradients between inside and outside of eggs are smaller. Although carbon dioxide and oxygen averages fr om high compact sand nests are extreme during this incubation stage, some nests had gas levels that were more severe. For example, one nest in nourished sand (compaction= 2.65 psi) of Melbourne Beach during the 2003 summer 100

PAGE 101

(mb03-nou08), had 40,967 ppm carbon dioxide and 15.4% oxygen. This nest had only 60.2% hatching success and 39.8% emergence success. Embryos surrounded by very high carbon dioxide and low oxygen concentrations have slower growth and increased mortality during th is period because of reduced respiratory gas exchange (Ackerman, 1980). Respiratory gas exch ange occurs across the sea turtle parchment egg shell and its permeability re gulates rate of gas exchange (Ackerman and Prange, 1972). However, in high compact nourished beaches, re duced sand porosity appears to be a greater limiting factor for gas exchange than egg shell permeability. Reduced porosity, lower percenta ge of air volume, is negativ ely correlated to high sand compaction and slows gas diffusion between cl utch air and beach sand (Tan, 1995). Gas diffusion rates are influenced by molecular size and since carbon dioxide is a larger molecule than oxygen (molecular weights 46.0 g/mol and 32.0 g/mol, respectively), it diffuses slower. However, other properties such as temperature, pressure, and solubility will also affect gas diffusion rates. Generally, diffusion coefficients increase when temperature rises, and decrease when pressure rises (Denny, 1993). Because temperature and pressure will be fairly similar for concentrations of oxygen and carbon dioxide in clutch air, diffusi on coefficients and solubility properties will be most significant in determining clutch gas exchange rates. Diffusion coefficients of oxygen and carbon di oxide in air (1 atm, 30C), are 21.5 x10 -6 m 2 s -1 and 17.0 x10 -6 m 2 s -1 respectively (Denny, 1993). Concentr ation of oxygen in air is also higher (20.9%), than that of car bon dioxide (0.033%), indicating th at rate of delivery of oxygen by diffusion to clutch air is 635 times higher than that of carbon dioxide (Denny, 1993). Weekly levels of oxygen show that concentrations decrease throughout incubation, but rate of decrease is slow and never more than 2.4% between weeks (Table 3-5). Embryonic growth 101

PAGE 102

slows in low oxygen conditions and egg mortality increases (Ackerman, 1975; 1980). However, sea turtles evolved adaptations to survive low oxygen concentrations, and Ackerman and Prange (1972) showed that loggerhead eggs can hatch in artificial nests c ontaining 5% oxygen if it decreases slowly. One nest (04-noami5-9) in a nourished section of Anna Maria Island (compaction= 3.5 psi) in 2004 had oxygen concentr ations that dropped pr ecipitously from 16% to 6% in one week. This clutch had 3.9% hatching success, and when it was excavated, unhatched eggs had embryos in ag ed-appropriate stages (stages 2629). Therefore, it appears that sea turtle eggs can tolerate a great degr ee of hypoxia if its onset is slow. This adaptive behavior enables sea turtle eggs survive low oxygen concentrations in clutch air caused by diffusion restrictions of different sand media. Critical period appears to be between incubation days 42-49 that correspond to embryonic stages 2830, respectively. This is the period prior to pipping when embryonic metabolism is more intense (Ackerman, 1977; 1980). The rate of carbon dioxide diffusion in air is slower than that of oxygen. This difference compromises the exchange between clutch air an d atmosphere and helps concentrate elevated levels in nests with high sand compaction. Howe ver, the harmful physiological impact of high carbon dioxide concentrations can be abated if clutch oxygen concentrations are near atmospheric levels. In fact, the Bohr Effect depends on high carbon dioxide to facilitate hemoglobin release oxygen to tissues. For ex ample, a 2003 nest (mb03-nou04) from a nourished Melbourne Beach section (compaction= 2.6 psi) had 66,229 ppm car bon dioxide and 20.0 % oxygen concentrations prior to pipping. This ne st had 96.6% hatching and emergence success. As mentioned in the previous example of nest mb03-nou08, when concentration of carbon dioxide is high and oxygen low, embryonic mortality increases. 102

PAGE 103

As clutch air diffuses outwardly through beach sand, it contacts water molecules. Sand water content ranges from an average of 5.9% in native beaches to 6.1% in nourished beaches, and from 4.5 % in east coast beaches to 8.2% in west coast beaches (Chapter 2). Diffusion rates of oxygen and carbon dioxide in water (1 atm, 30C), are 2.67 x10 -9 m 2 s -1 and 2.08 x10 -9 m 2 s -1 respectively, that corre spond to coefficients 10,000 times lo wer than in air (Denny, 1993). Lower diffusion coefficients of oxygen and carbon dioxide are a factor of their solubility rates and concentrations in water. Carbon diox ide is approximately 28 times more soluble in water than oxygen, and its concentration in wate r (0.02%) is similar to that in air (0.03%). However, the concentration of oxygen in water is only about 5% of that in air (Denny, 1993). Therefore, the movement of carbon dioxide mol ecules through water is reduced because of its decreased diffusion coefficient alone, whereas movement of oxygen is reduced because of its decreased concentration and solubili ty rate. This indica tes that the transpor t rate of oxygen in water is slower than th at of carbon dioxide. Because carbon dioxide is more soluble in wate r than oxygen, it is intuitive to think that sand water aids its diffusion from clutch air to atmosphere. However, when carbon dioxide dissolves in water a great percentage reacts to produce carbonic acid which then dissociates to bicarbonate ions (Denny, 1993). Ca rbonic acid is important to s ea turtle nesting ecology because it aids egg pipping by softening eggs hells (Bustard and Greenham, 1968). After pipping, hatchlings emerge from thei r eggshells and this activity increases sand porosity surrounding the nest chamber. Increased porosity reduces or equalizes gas pressure differences between clutch air and atmosphere. A sand depression is often visible at the beach surface above the nest. 103

PAGE 104

Conclusion A nourished beach with high sand comp action changes normal incubation gas concentrations and gas exchange rates of oxyge n and carbon dioxide in loggerhead sea turtle nests. Normal gas exchange occurs in two cycles: between parchment eggshell and clutch air, and between clutch air and atmosphere. Parchment eggshell permeability restricts diffusion between the chorioallantoic membrane and clutch air. However, this is not the para mount factor controlling gas diffusion rates from nests in nourished beaches with high sand compac tion. In these nests, diffusion of clutch air through sand media establishes primary gas partial pressure gradients. This is because of a smaller egg chamber, lower total clutch ai r volume, reduced sand porosity, and sand water content. Smaller egg chamber and lower total clutch ai r volume change normal gas partial pressure dynamics between the center and periphery of the clutch. Reduced sand porosity and increased sand water content slow diffusion of clutch oxygen and carbon dioxide in sand media. Because of differences in concentrations, diffusion rates, and solubility of the two molecules, oxygen is less affected than carbon dioxide. These impacts can be reduced or avoided if beach nourishment projects are designed to minimize changes in sand porosity and sand water c ontent. Porosity is affected by sand density and grain size. Reduced density or compaction, and a higher percen tage of larger sand grains lessen this effect by increasing tota l air volume in beach sand. Sand water content is affected by total sand calcium carbonate and sand compaction. Although water is necessary to maintain proper egg hydration, excess sand moisture is negatively correlated to porosity and will reduce total sand air volume. Sand water also acts as a pathway to dissipate elevated carbon dioxide concentrations from the clutch. Carbon dioxide dissolves well 104

PAGE 105

in water and as previously mentioned, the resu lting carbonic acid helps weaken the eggshell and aid in hatching (Bustard and Greenham, 1968). However, high water content hinders the diffusion rate of oxygen through sand. Growth and development of embryos are strongly dependent on proper physiological levels of oxygen and carbon dioxide. The most crit ical period is between incubation days 42-49 when embryonic metabolism is more intense. Al though sea turtle eggs are adapted to survive a certain degree of hypoxia its onset must be slow Beach nourishment projects that reduce total sand air volume and increase sand water conten t create conditions beyond what sea turtle clutches are evolved to survive. It is very important that beach restoration projects maintain the integrity and quality of the beach habitat. The population of loggerhead sea turtles in Florida is declining (Witherington et al., 2009), and the recovery of this species will depend on several factors including increasing beach productivity. 105

PAGE 106

CHAPTER 4 CARBON DIOXIDE, CARBONATE AND HATCHLING PHYSIOLOGY Introduction Chapter 3 data showed that sea turtle nests with higher incubation concentrations of carbon dioxide and lower oxygen were primarily located in nourished sand with high compaction. These nests had a lower hatching success, and a statistically significant lower emergence success than those in soft sand. This difference was attributed to higher concentrations of carbon dioxide and low oxygen in clutch air during the second half of incubation. Unlike oxygen, carbon dioxide accumulate s because it has a lower diffusion rate and moves more slowly between clutch air and atmosphere. Aside from its diffusion in cl utch air, carbon dioxide mol ecules also encounter water molecules in sand. As discussed in Chapte r 3, carbon dioxide dissolves well in sand water leading to the formation of carbonic acid that aids in eggshell dissolution and pipping. Besides water content, other sand properties can also po tentially influence the exchange of incubation gases between sea turtle cl utches and atmosphere. A Principal Components Analysis (PCA) was done in Chapter 2 to determine the parameters that accounted for most variance in beach physical properties. Sand calcium carbonate was identified as primary factor. Calcium carbonate is very dynamic because it has a dual role in sea turtle incubation. It can be detrimental because it increases sand compaction, lowering gas diffusion rates, and the ability of hatchlings to emerge (Ackerman, 1981; Ehrhart and Roberts, 2001). It can be beneficial because of its ability to buffer concentrations of carbon dioxide (Mota and Peterson, 2002; 2003). Total percentage of calcium carbonate in Fl orida beaches varies. Higher percentages (14.4 30.0%) were measured in nourished beac hes in the southwest coast, and 0.0 % was 106

PAGE 107

measured throughout native Panhandle beaches (Chapter 2). However, its role is not only dependent on total concentration, but also on physi cal form and reactivity. Physical forms of carbonate can vary from fine powder to hard clum ps to seashell carbonate, and each of these will have different surface areas a nd reactivity rates (Henderson et al., 2001; Kraimer et al., 2005; Landi et al., 2004; Moreno et al., 2006). B ecause nourishment sand sources vary, knowing carbonate reactivity rates can prev ent the use of highly reactive carbonate that might adversely affect a sea turtle nesting beach (Henderson et al ., 2001; Horvath et al., 20 05; Mosaddeghi et al., 2005; Siewert, 2004). This project, however, does not address the question of how different carbonate reactivity rates influence sand compaction. This question merits further research and should be brought to fruition in a laboratory wh ere all experimental variables can be properly controlled. Aside from lowering percent hatching and em ergence success as described in Chapter 3, high concentrations of carbon dioxide during incubation can al so influence other hatchling physiological characteristics. For example, sex in sea turtles is determined by the proportion of time spent at a certain temper ature during incubation (Yntema and Mrosovsky, 1980). However, it has been shown that in Chrysemys scripta higher carbon dioxide le vels can lengthen total incubation time and along with pH, can skew sex ratios (Etchberger et al., 1992). Chrysemys and Pseudemys incubated at elevated levels of carbo n dioxide absorbed less yolk, had different pigmentation and plastral patterns, had depressed metabolism and were smaller than turtles incubated at lower levels of carbon dioxide (Etchberger et al., 1992; 1993; Kam, 1993). In painted turtles, high carbon dioxide leads to an increase in lactic acid that is buffered by the breakdown of bone and shell compon ents (Jackson et al., 2000). 107

PAGE 108

Successful incubation requires proper physiologi cal levels of carbon dioxide. Elevated concentrations not only have a negative impact on incubation itself, but ca n also potentially lead to developmental abnormalities. The present project investigates two major questions pertaini ng to the influence elevated concentrations of carbon dioxide can have on loggerhead sea turtle incubation. The first question is whether levels of carbon dioxi de can significantly be reduced by unbound sand carbonate. Secondly, I investigated the possible impacts that elevat ed levels of carbon dioxide may have on hatchling physiology and fitness. Materials and Methods Loggerhead sea turtle eggs were collected on July 22 nd 2006 on the Merrit Island National Wildlife Refuge Beach. They were removed from different clutches during oviposition and placed in commercial egg cartons, covered by a wet towel and transported inside an airconditioned vehicle to the incubation laboratory at NASAs Kennedy Space Center. Transport time was less than 30 minutes. All eggs were measured (minimum and maximum diameters), weighed, and marked with a pencil to identify clutch. They averaged 40.1 mm (SD = 1.86) in diameter and 39.6 g ( SD = 4.59) in weight. Eggs were randomly distributed amongst 12 incubators (40/incubator) that were constructed from 10-gallon insulated commercial Igloo coolers. Coolers were 50 cm deep with an inner diameter of 34 cm that co rresponds to a volume of approximately 0.0453 m 3 Eight experimental incubators were housed in a humidity and temperature-controlled room with a 12-hour light cycle (on at 7AM). Inc ubation temperature and moisture were measured using commercial sensors buried in the sand. Ea ch experimental incubator was filled with approximately 47 Kg ( SD = 1.77) of sand that was weathered fo r at least two years to reduce the possibility of introducing highly re active carbonates. Calcium carbonate concentrations in the 108

PAGE 109

two experimental sand treatments varied. Merr itt Island National Wildlife Refuge Beach sand averaged 3.11 % ( SD = 0.61), and Jensen Beach sand averaged 10.74 % ( SD = 0.97). Four control incubators were buried on th e beach. These replicated experimental conditions except that the bottom of the coolers was removed to prevent water from pooling on eggs. Each experimental treatment consisted of a sample size of 40 eggs/incubator with four replicates. This design produced a statistically appropriate number of hatchlings for the second portion of this experiment. Because nest chamber architecture can aff ect gas diffusion, each nest had the same architecture, depth, and all eggs were positioned 25 cm below sand surface and 15 cm above the incubator bottom. This eliminated any variability due to degree of sand contact and total volume of air spaces. Nests on the Merrit Island Beach used as egg sources were marked and their reproductive success was followed. Additionally, an equal number of control in situ nests (not used for egg collection) were marked and their reproductive su ccess recorded. This was done in collaboration with the Merritt Island National Wildlife Refuge. During incubation, weekly 2 ml air samples were collected and analyzed for oxygen and carbon dioxide. Gas samples were extracted from the center of each clutch via an air-sampling tube and using a gas-tight syringe. Samples we re transferred to Call-5 Bond gas sampling bags fitted with a Luer stopcock for easy c onnection to the sampling syringe. Carbon dioxide analysis was completed with a 6890 Hewlett Packard Gas Chromatograph with Thermal Conductivity Detector and a J&W GS-Q column. Oxygen was analyzed on a 5880 Hewlett Packard Gas Chromatograph with Th ermal Conductivity Detector and a packed molecular sieve column. Methodologies used to sample clutch air from each incubator and to 109

PAGE 110

analyze it for oxygen and carbon dioxid e concentrations replicated those described in Chapter 3 of this research project. In collaboration with the College of Veterinary Medicine at the Univ ersity of Florida, a hatchling physical condition and health assessme nt protocol was designed. After emergence, hatchlings were weighed, measured (straight carapace length, straight carapace width, body depth), and data correlated to sand treatment, incubation time, oxygen and carbon dioxide levels. A sample of 10 hatchlings/incubator had 200 microl iters of blood drawn from the dorsal cervical sinus using a 25 gauge syringe according to proc edures outlined in Wibbels et al. (1998). Blood samples were transferred to a micropipette, spun and analyzed for glucose, total cholesterol, packed cell volume, triglycerides, lactate a nd total protein. Analyses were done with commercial testers: cholesterol, triglycerides and glucose were measured with a CardioChek portable analyzer with PTS Panels test strips in mg/dL; protein was measured with a portable refractometer in g/dL, and lactic acid with an Arkray Lactate Pro blood test meter in mmol/L. A different subset of five hatchlings from each incubator were placed on a sand-covered runway and allowed to crawl 10 meters. Observ ations were made on ha tchling behavior during the crawl and total crawl time r ecorded. Afterwards, these hatc hlings also had blood drawn and similar blood chemistry analyses done. After all hatchlings emerged, nest contents were excavated and inventoried. Number of hatched and unhatched eggs was counted and all d ead embryos were aged according to Millers life stages (1982). Gas concentr ations of oxygen and carbon dioxide measured in each clutch throughout incubation were adjusted to reflect mort ality at each respective age. Sand from both experimental incubator treatments was rean alyzed for calcium carbonate after hatching. 110

PAGE 111

Data analyses were performed using SPSS v.12 statistical program software. A nonparametric Kruskal-Wallis analysis of variances (ANOVA) was used to compare differences in average egg diameter and weights between clutches. Concentrations of carbon dioxide from each inc ubator were averaged, log transformed and grouped according to sand treatment. A time seri es was plotted and trend lines derived for carbon dioxide levels from each carbonate sa nd treatment, control beach incubators, and in situ nests. A Kruskal-Wallis ANOVA analyzed for di fferences in incubation concentrations of carbon dioxide, incubation time, percent hatchi ng success and emergence success between all incubation sand treatments. Post Hoc analysis using Tukeys HSD was performed to find which means were significantly different from one another. Descriptive statistics as mean, standard devi ation, standard error, variance, minimum, and maximum were calculated for hatchling morphomet ric data from each incubation treatment. A non-parametric Kruskal-Wallis ANOVA was used to compare differences in these data. A sample of hatchlings from each incubator treatment had blood drawn and analyzed for glucose, total cholesterol, packed cell volume, triglycerides, lactat e and total protein. Packed cell volume/protein was also calculated and is a ratio that indicates dehydration st ate. Descriptive statistics as mean, standard deviation, standard error, variance, minimum, and maximum were calculated for blood data from each incubation treatment. Non-parametric Kruskal-Wallis ANOVA analyzed for differences in blood data according to incuba tion experimental treatments. Post Hoc analysis using Tukeys HSD was perf ormed to find which treatment groups were significantly different from one another. A subset of five hatchlings from each incuba tor were placed on a sand-covered runway and allowed to crawl 10 meters. Total crawl times were recorded, blood was drawn and analyzed for 111

PAGE 112

similar chemistry. Non-parametric KruskalWallis ANOVA analyzed for differences in blood data according to incubation expe rimental treatments. Post Hoc Tukeys HSD analysis identified which treatment groups were differed. Results Descriptive statistics as mean, standard devi ation, standard error, variance, minimum, and maximum for each clutch are in Table 4-1. A non-parametric Kruskal-Wallis analysis of variances (ANOVA) showed statistically significant differences (Table 4-2) for average egg diameter ( F 5,54 = 51.25, p= 0.000), and egg weight ( F 5,54 = 86.7, p= 0.000). Post Hoc analysis using Tukey s HSD test shows that eggs from clutches numbers six and three were statistically different from all others including each other in both diameter and weight. Average concentrations of carbon dioxide from each sand treatment are in Table 4-3. A time series and carbon dioxide trend lines for nests in sand with low carbonate (y = 0.24 log(x) + 2.97; R 2 = 0.42), high carbonate (y = 0.54 log(x) + 2.92; R 2 = 0.70), control beach incubators (y = 0.12 log(x) + 2.92; R 2 = 0.17), and in situ nests (y = 0.05 log(x) + 2.95; R 2 = 0.04) are in Figure 4-1. A non-parametric Kruskal-Wa llis ANOVA showed statistica lly significant differences ( F 3,47 = 16.23, p= 0.000) between the four sand treatments (Table 4-4). Post Hoc Tukeys HSD test show that carbon dioxide concentrations m easured from eggs incubating in high carbonate sand differed from all others. Carbon dioxide measured from eggs incubating in low carbonate sand did not differ from that recorded from b each control incubators, but differed from carbon dioxide levels from in situ nests. Carbon dioxide measured from beach control incubators did not differ from that measured from in situ nests (Table 4-5). 112

PAGE 113

Non-parametric Kruskal-Wallis ANOVA (Table 4-6) found statistically significant differences for average incubation time ( F 3,36 = 11.37, p= 0.001), percentage of hatching success ( F 3,36 = 3.98, p= 0.035), and percentage of hatchling emergence ( F 3,36 = 20.14, p= 0.000) between all incubation sand treatments. Incubation time wa s shorter by an average of two days for eggs in high carbonate sand. Despite efforts to mini mize differences between incubation treatments, low carbonate sand temperature averaged 0.7 C higher. Hatching and emergence successes were different (lower) for in situ nests, however, this was due to storm erosion and nest flooding. Descriptive statistics for hatchling morphomet ric data from each incubation treatment are in Table 4-7. A non-parametric Kruskal-Wa llis ANOVA (Table 4-8) shows statistically significant differences for straight carapace width ( F 2,239 = 3.53, p= 0.031), and body depth ( F 2,240 = 36.38, p= 0.000). Post Hoc analysis using Tukeys HSD test shows that hatchlings from in situ nests had smaller widths; and hatchlings from low carbonate sand incubators had smaller body depths. Descriptive statistics of hatc hling blood data for each incubator treatment are in Table 4-9. Non-parametric Kruskal-Wallis ANOVA found statistic ally significant differences (Table 4-10) for packed cell volume ( F 2,93 = 11.25, p= 0.000), triglycerides ( F 2,93 = 5.07, p= 0.008), packed cell volume/protein ratio ( F 2,93 = 18.90, p= 0.000), and total serum protein ( F 2,93 = 8.46, p= 0.000). Post Hoc analysis using Tukeys HSD s howed that packed cell volume differed between hatchlings incubated in low a nd high carbonate sand, triglyceride s differed between hatchlings incubated in high carbonate sand and in situ, and total serum protein as well as PCV/protein ratio differed in hatchlings fr om high carbonate sand. Descriptive statistics for ha tchling crawl time and blood data from a subset of five hatchlings from each incubation treatment are in Table 4-11. Non-parametric Kruskal-Wallis 113

PAGE 114

ANOVA (Table 4-12) found statistically significant differences for run time ( F 2,57 = 61.97, p= 0.000), packed cell volume ( F 2,57 = 7.76, p= 0.001), triglycerides ( F 2,57 = 23.34, p= 0.000), and packed cell volume/protein ratio ( F 2,57 = 5.71, p= 0.005). Post Hoc Tukeys HSD analysis showed that crawl time differed betw een each of the three treatment groups. Packed cell volume, triglyce rides, and PCV/protein ratio differed between hatchlings incubated in high carbonate sand and the other two treatment groups. 114

PAGE 115

Table 4-1. Descriptive statisti cs of egg diameter (mm) and weight (g) collected from six different loggerhead clutches (n= 40 from each clutch). Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Clutch Mean SD SE Var Min Max 1 Diameter 38.4 0.82 0.26 0.67 36.7 39.6 Weight 35.6 1.62 0.51 2.64 31.4 37.4 2 Diameter 38.4 0.67 0.21 0.46 37.6 39.5 Weight 35.8 1.78 0.56 3.17 33.2 37.8 3 Diameter 42.6 0.69 0.22 0.48 41.5 43.3 Weight 46.6 1.55 0.49 2.40 44.2 48.7 4 Diameter 38.8 0.64 0.20 0.41 38.2 40.0 Weight 36.7 1.21 0.38 1.46 35.0 38.7 5 Diameter 39.4 0.96 0.30 0.92 37.8 41.2 Weight 38.3 1.98 0.62 3.92 35.0 40.9 6 Diameter 41.3 0.73 0.23 0.54 40.3 42.3 Weight 43.7 1.07 0.34 1.16 42.2 45.2 Table 4-2. Results of a Kruskal-Wallis ANOVA co mparing egg diameters and weights from six different loggerhead clutches (n=40 from each clutch). Post Hoc analysis using Tukeys HSD showed that eggs from clutch es number three and six were different from all others including each ot her in both diameter and weight. Sum of Mean Squares Square F 5,54 P Diameter 148.9 29.8 51.2 0.000 31.39 0.58 180.36 Weight 1067.2 213.4 86.7 0.000 132.9 2.46 1200.2 115

PAGE 116

Table 4-3. Concentrations of carbon dioxide (log transformed) measured in four different treatments that varied according to percentage of sand calcium carbonate. Experimental incubators (n=4 each treatme nt) had sand with low (3.11 %) carbonate collected at the Merritt Island National Wildlife Refuge Beach (MINWR), and high (10.74 %) carbonate sand collected at Jensen Beach in 2006. Beach incubators were incubators buried at the MINWR Beach and in situ nests served as controls. Incubation Day Low CaCO 3 High CaCO 3 Beach Incub. in situ 0 2.63 2.63 2.63 2.63 9 3.17 3.29 3.07 3.04 15 3.34 3.47 3.22 3.18 22 3.46 3.74 3.23 3.25 29 3.68 3.84 3.27 3.05 33 3.64 4.27 3.21 3.29 37 3.54 4.38 3.35 3.22 42 3.40 4.44 3.27 3.06 45 3.53 3.99 2.80 2.93 48 3.47 3.87 2.93 49 3.45 3.94 2.94 2.91 51 3.41 3.98 3.16 2.91 52 3.82 4.08 53 3.17 3.27 116

PAGE 117

2.0 2.5 3.0 3.5 4.0 4.5 5.0 09152229333742454849515253Incubation DayCO2 concentration (log) Log. (Inc Low CaCO3) Log. (Inc High CaCO3) Log. (Beach Inc) Log. (in situ Nests) Figure 4-1. Time series of log transformed carbon dioxide concentra tion data from four different treatments that varied according to percentage of sand calcium carbonate. Experimental incubators (n=4 each treatme nt) had sand with low (3.11 %) carbonate collected at the Merritt Island National Wildlife Refuge Beach (MINWR), and high (10.74 %) carbonate sand collected at Jensen Beach in 2006. Beach incubators were incubators buried at the MINWR Beach and in situ nests served as controls. Trend lines were derived for each sand treatment of low (y = 0.24 log(x) + 2.97; R 2 = 0.42) and high (y = 0.54 log(x) + 2.92; R 2 = 0.70) carbonate, as well as control beach incubators (y = 0.12 log(x) + 2.92; R 2 = 0.17) and in situ nests (y = 0.05 log(x) + 2.95; R 2 = 0.04). 117

PAGE 118

Table 4-4. Results of a Kruskal-Wallis ANOVA comparing incubation concentrations of carbon dioxide from four experimental treatments of carbonate sand. Sum of Mean Squares Square F 3,47 P Carbon Dioxide 5.00 1.67 16.23 0.000 4.84 0.10 9.84 Table 4-5. Post Hoc analysis of carbon dioxide concentrations (log) from four experimental treatments of carbonate sand using Tukeys HSD. Results show carbon dioxide from clutches in high carbonate sand incubators di ffered from all others. Carbon dioxide from clutches in low carbonate sand did not differ from clutches in beach incubators, but differed from in situ nests. Beach incubators did not differ from in situ nests Treatment Tukeys Grouping In situ nests 3.03 Beach incubators 3.12 3.12 Low carbonate incubator 3.40 High carbonate incubator 3.84 118

PAGE 119

Table 4-6. Average incubation time (days), ha tching success (%), and emergence success (%) of sea turtle clutches from four different treatments of sand calcium carbonate. Experimental incubators (n=4 each treat ment) had sand with low (3.11 %), and high (10.74 %) carbonate sand. Beach incubators (n=4) were buried in low carbonate sand, and in situ nests (n=28) were controls. Tukeys Post Hoc analysis was done to find which treatments differed. Asterisks (*) indicate statistically significant differences p < 0.03. Treatment Incubation % Hatching % Emergence Low CaCO 3 52.0 85.6 85.6 High CaCO 3 49.2 86.9 86.3 Beach Incubator 54.7 95.0 94.2 in situ 54.2 81.0 79.5 Table 4-7. Descriptive statistics of hatchling morphometric data (mm) and weight (g) collected from three different experimental incubato r treatments (n=4 each treatment). One group was incubated in low (3.11 %) carbon ate sand (Nat), other in high (10.74 %) carbonate sand (Nou), and the third had incubators burie d in beach sand that had identical low carbonate concentrations (Beach). Data are presented as mean, standard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maximum (Max). Parameter Mean SD SE Var Min Max Weight Nat sand 17.6 1.42 0.13 2.03 14.4 20.5 Nou sand 17.6 1.58 0.16 2.49 13.2 21.4 Beach 18.3 1.49 0.39 2.23 16.1 20.5 Length Nat sand 43.9 1.31 0.12 1.70 40.6 46.3 Nou sand 43.9 1.51 0.15 2.28 39.4 47.4 Beach 44.5 0.77 0.20 0.59 43.2 45.5 Width Nat sand 34.0 1.10 0.10 1.22 30.8 36.7 Nou sand 34.1 1.27 0.13 1.61 30.9 37.2 Beach 33.3 0.52 0.14 0.27 32.5 34.2 BD Nat sand 18.2 0.77 0.07 0.59 16.4 20.2 Nou sand 17.3 0.93 0.09 0.87 15.1 19.5 Beach 18.6 0.78 0.20 0.61 17.2 20.1 119

PAGE 120

Table 4-8. Results of a Krus kal-Wallis ANOVA comparing hatchling morphometric data (mm) collected from three different experimental in cubator treatments (n=4 each treatment). Statistically significant differences were found between straight carapace width and body depth. Post Hoc analysis using Tukey s HSD showed that width was smaller in hatchlings from in situ nests and body depth was sma ller in hatchlings from high carbonate sand incubators. Sum of Mean Squares Square F 2,239 P Width 9.3 4.66 3.53 0.031 315.6 1.32 325.0 Body Depth 52.2 26.10 36.38 0.000 172.1 0.72 224.3 ______________________________________________________________________________ 120

PAGE 121

Table 4-9. Descriptive statisti cs of hatchling blood data from three different experimental incubator treatments. One group was incuba ted in low (3.11 %) carbonate sand (Nat), other in high (10.74 %) carbonate sand (Nou), and the third had incubators buried in beach sand that had identical low carbona te concentrations (Beach). Data are presented as mean, standard deviation (S D), standard error (SE), variance (Var), minimum (Min), and maximum (Max). PC V= Packed cell volume, PCV/Protein= dehydration state indicator. Parameter Mean SD SE Var Min Max PCV Nat sand 0.39 0.07 0.01 0.00 0.25 0.60 Nou sand 0.47 0.08 0.01 0.01 0.33 0.66 Beach 0.43 0.08 0.02 0.01 0.26 0.54 Triglycerides (mg/dL) Nat sand 91 31 5 975 58 191 Nou sand 106 52 9 2706 54 258 Beach 69 7 2 50 56 81 PCV/Protein Nat sand 5.08 1.27 0.19 1.61 2.22 7.5 Nou sand 6.97 1.77 0.30 3.13 4.11 11 Beach 5.19 0.97 0.25 0.95 3.26 7.18 Cholesterol (mg/dL) Nat sand 228 155 23 23876 50 401 Nou sand 178 164 27 26859 50 401 Beach 127 88 23 7706 50 266 Lactic acid (mmol/L) Nat sand 6.79 2.80 0.42 7.81 0.9 12.7 Nou sand 5.38 3.41 0.57 11.60 0.4 10.8 Beach 7.31 3.81 0.99 14.55 1.3 16.4 Protein (g/dL) Nat sand 7.92 1.61 0.24 2.59 5 13.5 Nou sand 6.89 1.15 0.19 1.32 4.7 9.2 Beach 8.39 1.04 0.27 1.08 7 11.4 121

PAGE 122

Table 4-10. Results of a Krus kal-Wallis ANOVA comparing hatchling blood data from three different experimental incubator treatments PCV= Packed cell volume, PCV/Protein ratio indicates dehydration state. Only st atistically significant parameters are shown. Sum of Mean Squares Square F 2,92 P PCV 0.13 0.06 11.25 0.000 0.53 0.01 0.66 Triglycerides 15090 7545 5.07 0.008 138302 1487 153392 PCV/Protein 78.7 39.34 18.90 0.000 193.6 2.08 272.3 Protein 31.9 15.95 8.46 0.000 175.3 1.89 207.2 122

PAGE 123

Table 4-11. Descriptive statisti cs of a 10-meter run trial and hatchling blood data from three different experimental incubator treatmen ts. One group was incubated in low (3.11 %) carbonate sand (Nat), other in high (10.74 %) carbonate sand (N ou), and the third had incubators buried in beach sand that ha d identical low carbonate concentrations (Beach). Data are presented as mean, sta ndard deviation (SD), standard error (SE), variance (Var), minimum (Min), and maxi mum (Max). PCV= Packed cell volume, PCV/Protein= dehydration indicator. Parameter Mean SD SE Var Min Max Run time (min) Nat sand 2.62 0.81 0.47 0.66 2.07 3.55 Nou sand 3.97 0.27 0.13 0.07 3.59 4.21 Beach 2.08 0.25 0.18 0.06 1.90 2.26 PCV Nat sand 0.45 0.09 0.05 0.01 0.37 0.55 Nou sand 0.63 0.15 0.08 0.02 0.40 0.73 Beach 0.44 0.16 0.11 0.02 0.33 0.55 Cholesterol (mg/dL) Nat sand 283 54.8 31.7 3009 244 346 Nou sand 93.7 87.5 43.7 7656 50 225 Beach 152 144 102 20808 50 254 Triglycerides (mg/dL) Nat sand 60.5 14.0 3.1 195 45 104 Nou sand 78.7 12.5 2.8 156 61 112 Beach 53.3 9.5 2.1 90.2 36 72 Lactic acid (mmol/L) Nat sand 5.74 3.72 0.83 13.8 0.4 15.8 Nou sand 5.72 3.80 0.87 14.4 0.8 11.7 Beach 6.44 3.49 1.95 12.17 1.1 13.9 Protein (g/dL) Nat sand 7.07 1.10 0.64 1.21 5.8 7.8 Nou sand 6.53 2.01 1.00 4.04 4.6 8.5 Beach 6.4 1.41 1.00 2.00 5.4 7.4 PCV/Protein Nat sand 5.55 2.02 0.45 4.09 2.04 8.77 Nou sand 8.15 3.79 0.87 14.4 3.88 15.8 Beach 5.79 2.16 0.54 4.64 3.33 12.2 123

PAGE 124

Table 4-12. Results of a Krus kal-Wallis ANOVA comparing hatchling blood data from three different experimental incubator treatmen ts. Blood was sampled after hatchlings crawled a 10-meter sand track. PCV= P acked cell volume, PCV/Protein ratio indicates dehydration state. Only statistically significant parameters are shown. Sum of Mean Squares Square F 3,47 P Run Time 33.85 16.92 61.97 0.000 15.57 0.27 49.41 PCV 0.31 0.15 7.76 0.001 1.13 0.02 1.44 Triglycerides 6862 3431 23.33 0.000 8381 147 15244 PCV/Protein 81.64 40.82 5.71 0.005 407 7.14 489 124

PAGE 125

Discussion Concentrations of oxygen and carbon dioxide re corded in all experimental treatments reflected the incubation cycles described in Chapte r 3 of this dissertation. These cycles showed a pattern with oxygen remaining similar to atmosp heric levels (20.9 %) duri ng the first month of incubation, and then gradually decreasing until hatching. Concomitantly, carbon dioxide concentrations remained relatively low dur ing the first month of incubation, but then increased sharply until hatching. These patt erns result from increased me tabolic demand by hatchlings in the second half of incubation. Because carbon dioxide levels from in situ nests did not differ statistically from those in beach incubators, it indicates that the use of artificial incubators did not influence results. Also, housing incubators in a laboratory room did not skew data because carbon dioxide did not differ statistically between beach incubators and expe rimental incubators with low carbonate. Carbon dioxide levels recorded in incuba tors with high carbonate sand we re statistically different than those from incubators with low carbonate sand, indicating that the sand treatment impacted the amount of carbon dioxide levels (Figure 4-1). This disparity is attributed to increased compaction that is a factor of higher carbona tes. Compaction hinde rs diffusion of carbon dioxide that accumulates and can negatively affect hatching. However, a very significant finding from this experiment is that average carbon dioxide levels from high carbonate sand were much lower than those reported in Chapter 3. In this experiment, up to incubation day 42 carbon dioxid e was similar to that recorded in high compaction sand in Chapter 3. These concentrat ions continued to increase in high compaction sand nests through the sixth week of incubation (30,777 ppm), however, in incubators with high carbonate sand levels decreased to a low average of 8,710 ppm by day 49. This corresponds to 125

PAGE 126

developmental stage 30 when pipping occurs and when embryonic metabolism is very high (Ackerman, 1980). This decrease was due to a buffering eff ect on carbon dioxide concentrations by unbound calcium carbonate in experimental sand. This re action leads to the form ation of carbonic acid which aids to soften eggshells and aid egg pi pping. Low carbonate sand lost an average of 1.92 g (61.7 %) of its original amount while high carb onate sand lost an average of 4.19 g (42.3 %). Approximately twice as much unbounded sand calciu m carbonate was lost from the incubators with high carbonate sand supporting the hypothesis that it reacted to buffer carbon dioxide. A reduction of very high clutch ca rbon dioxide concentrations duri ng late incubation stages is beneficial to reduce embryonic mortality. These data suggest that when choosing a sa nd source for beach nourishment projects, it is important to measure total ca lcium carbonate as well as unbound calcium carbonate in sand. Further research should investigate the rati o or minimum concentr ation of unbound calcium carbonate that maximizes its role in buffering elevated levels of carbon dioxide and catalyzes the formation of carbonic acid while preventing sand from cementing sea turtle nests. Beach incubators had mean oxygen levels fairly identical to in situ nests and very close to atmospheric levels. Oxygen concentrations from the two treatment incubators were slightly lower than oxygen levels from beach incubators but similar to each other. Oxygen in these incubators decreased slowly after the second mo nth of incubation, but unlike those from highly compacted sand, never got below 17.2 %. This oxygen concentration is within the tolerance range of sea turtle embryos (Ackerman, 1980). Nests in highly compacted nourishment sand (Chapter 3) had oxygen levels decreased to 15.4 % and had fairly success ful hatching rates (60.2 %). 126

PAGE 127

In 2003, one nest in a nourished section of Park Shore Beach with very compact sand (compaction= 3.2 psi), had a clutch oxygen concen tration of 6.4 % at incu bation day 45. This level is extremely low for human survival and comparable to that at the summit of Mount Everest. However, this nest was washed out by Tropical Storm Henri so it is unknown what its hatching success wouldve been. When some of its unhatched eggs were recovered from the surf and opened, embryos showed stage-appropriate an atomical development indicating that they were viable despite lo w oxygen concentrations. Hatchlings from in situ nests had smaller carapace widths than those from other treatments. However, this difference must be disregarded because each in situ clutch came from a different individual female. Experimental incubators had eggs that were co llected from six different nests and randomly assigned to each treatment. Differen t egg sizes produce different sized hatchlings and egg randomization negated this effect. In in situ nests this was not controlled and therefore, statistically significant differences in hatchli ng straight carapace width were confounded by the experiment and should be ignored as having biologi cal significance. Packed cell volume (PCV) refers to percen tage of total blood volume occupied by erythrocytes and low values can indicate anemia. Hatchlings from both incubator treatments as well as those from beach control incubators ha d PCV values that were higher than the mean reported (28.3 %) for subadult loggerhead sea turtles (Lutz and Dunbar-Cooper, 1987). It appears that loggerhead sea turtle embryos have plasticity to adapt to hypoxic conditions by increasing the number of erythrocytes. This increase was 8% higher in hatchlings from high carbonate than in those from low carbonate incu bators (47 % and 39%, re spectively), indicating that the increase could be corr elated to hypoxic conditions. 127

PAGE 128

Serum triglycerides are an energy source and body fat consists almost entirely of triglycerides. Triglyceride blood levels in sea tu rtle hatchlings are influe nced by diet or residual yolk (Jacobson, personal communication). Triglycer ides differed between hatchlings incubated in high carbonate sand inc ubators (106 mg/dL) and in situ nests (69 mg/dL). Hatchlings from high carbonate sand, that were also thinner than the others, were metabolizing more yolk energy reserves. Total serum protein measures total protein, albumin, and globulin in blood. This test indicates if an animal is malnourished or if the liver and kidneys are functioning properly. Hatchlings incubated in high car bonate sand had statistically lowe r protein levels (6.89 g/dL) than those from low carbonate sand (7.92 g/dL) and beach incubators (8.39 g/dL). High values indicate lower plasma levels and a con centration of protei ns (Cotter, 2001). PCV/protein ratio is used to indicate dehydration, anemia or polycemia, and normal physiological blood levels usually have a ratio near 5.0 (Cotter, 2001). If PCV values are elevated, as in the case of the majority of e xperimental hatchlings, a low protein level gives a high PCV/protein ratio, suggesti ng dehydration. High carbonate sand had a higher percentage of dehydrated hatchlings (89.1 %) with a higher mean PCV/protein ratio of 6.97 than low carbonate sand (43.3 %, 5.08) and beach incubators (53.4%, 5.19). Differences were found in crawl time for each treatment group. Hatchlings from the beach incubators completed the sand course in the fast est time (2.08 min), followed by hatchlings from low carbonate sand (2.62 min) and high carbonat e sand (3.97 min). Hatchlings from high carbonate sand incubators were more sluggish and took more rest breaks than those from the low carbonate sand incubators. Hatchlings from beach incubators did not take any rest breaks and crawled non-stop until they reached a flashlight positioned at the end of the sand runway. 128

PAGE 129

Although blood lactate levels were not statistically significan t between treatments, mean values confirm behaviors observed during crawl tria ls (Table 4-11). Lactat e levels measured in hatchlings from beach incubators were highest (6.44 mmol/L), reflecting tired muscles after a non-stop 2.08 minutes crawl. Hatchlings from low carbonate sand had lactate values of 5.74 mmol/L and high carbonate sand hatchlings had lact ate values of 5.72 mmol/L. These values are lower because hatchlings crawled slower, rested more and longer which helped dissipate lactic acid build up in muscle tissue. This shows th at the ocean crawl exerts considerable stress on hatchlings as indicated by an increase in the PCV/protein dehydration indices as well as in the number of dehydrated hatchlings. Conclusion The goals of this experiment were to evaluate the effects unbound sand calcium carbonate had on clutch gas concentrations of oxygen and carbon dioxide, embryonic development, hatching success, emergence and phy siology of loggerhead sea turtle hatchlings. It appears that sand carbonate buffers elevated carbon dioxide a nd increases hatching success. However, the incubation microenvironment created by sand carbona tes also has a negative impact on hatchling physiology and fitness. High carbonate sand concentrations buffered elev ated carbon dioxide levels that reduced clutch mortality. This buffering effect was not observed throughout in cubation, but only after carbon dioxide increased to very high levels whic h occurred approximately at incubation day 42. This shows the dual impact carbonate has on s ea turtle incubation: increasing sand compaction and buffering carbon dioxide. Increasing compacti on hinders gas diffusion and as described in Chapter 3, can increase clutch mortality; bufferi ng carbon dioxide counteracts this effect, but does not prevent other physiological implications. 129

PAGE 130

This presents a very interesting challenge for beach nourishment projects. Total sand carbonate is a parameter that is regulated in beac h fill because of its propensity to increase sand compaction. However, total carbonate includes carbonate in seashell material which is not free to react. Therefore, unbound calcium carbonate is the variable that should be carefully monitored. Although high unbound carbonate buffered carbon dioxide, high concentrations are not desirable because they increase sand compac tion that hiders the diffusion of clutch carbon dioxide. High concentrations of carbon dioxi de increase chances of embryonic mortality, especially between incubation days 30-42. Lo wer concentrations of unbound calcium carbonate are less harmful to sea turtle incubation. Blood data show that sea turtle embryos evolve d adaptations to surviv e low concentrations of oxygen and high carbon dioxide by increasing their number of erythr ocytes. The increase appears to be correlated to degree of hypoxia. This physiolo gical adaptation contributes to successful incubation and explains why some sea turtle nests ha ve a high hatching success in nourished beaches with high sand compaction. Howe ver, it should not be used as justification for a poorly designed beach nourishment project. As described in Chapter 2, beaches in the Panhandle have very little or no sand carbonate. Sea turtle nests hatch in these beaches which indicates that calcium carbonate in sand media is not required for successful incubation. Despite being adapted to increa se their total number of erythr ocytes, hatchlings exposed to higher carbon dioxide concentrations were thin ner, more dehydrated and metabolized more fat energy reserves. Their physiological condition worsened after the crawl from nest to ocean and raises the question of hatchling fitness. While it is important to monitor hatching and emergence success, it is just as important to understand th e effects that incubation conditions created by beach nourishment projects can have on hatchling quality. 130

PAGE 131

After emerging, hatchling crawl to the ocean and begin their next physical challenge, the swimming frenzy. Hatchlings exposed to higher carbon dioxide levels we re physiologically less fit for this challenge. Fitness was measur ed as crawl speed, dehydr ation, and amount of triglycerides that could be converted to en ergy. Despite the buffe ring effect that unbound carbonate had on carbon dioxide, hatchli ngs were still exposed to elev ated levels. This exposure, especially between incubation days 30-42, negatively impacted their physiology. The crawl to ocean and swimming frenzy are a period in the sea turtle life cycle when very high selection occurs. Detrimental change s in hatchling physiology will reduce their fitness and changes of survival. Therefore, beach nour ishment projects must be designed to reduce impacts on nesting turtles (number of false craw ls), egg incubation (hatching success), and any factor with the possibility to cause nega tive consequences on hatchling physiology. 131

PAGE 132

CHAPTER 5 FUTURE BEACH NOURISHMENT AND SEA TURTLE NESTING Beach restoration projects increase beach width and stabilize coastal erosion. To accomplish this, a large amount of fill material is pl aced on the shore. Properties of this fill are designated to increase the longevity of the nourishment project and may not necessarily be conducive to sea turtle nesting ecology. Although there are regulations to minimize impacts of beach nourishment projects on sea turtle nesting, they create habitats with physical properties that are different than those in native coastlines Because the beach ecosy stem is utilized by sea turtles, these differences can have benefici al or detrimental impacts on their reproductive ecology. The goals of this dissertation were: 1) to assess what sand physical characteristics differ mostly between native and no urished beaches, 2) determined statistical correlations and interactions between sand prope rties and sea turtle incubation, 3) compare how sand compaction influences incubation gas concentrations of oxygen and carbon dioxide, and 4) determine how unbound calcium carbonate impacts sea turtle incubation and hatchling fitness. Beach nourishment projects create nesting habitat in locations where it was previously limited. However, that habitat quality may be s uboptimal which reduces its beneficial value. Project engineers and regulators st ipulate acceptable values for sand properties such as total sand carbonate and beach compaction. These parameters contribute most variance in beach characteristics and have a str onger influence on sea tu rtle ecology because of the potential they have to impact incubation. This impact can be directly correlated to th ese two parameters or de rived from interactions between several physical properties. For exam ple, different proportions of sand grain sizes regulate sand compaction. Compaction influences sand porosity and tortuosity which determine sand water content. The inter action between all these sand physi cal characteristics impacts the 132

PAGE 133

diffusion coefficients, solubility rates and concentrations of oxygen and carbon dioxide. It is important that these relationships be deciphere d to minimize selecting beach nourishment sands that are likely to cause detrimental effects to the incubation of sea turtle clutches. An effect commonly associated with new b each nourishment projects is increased sand compaction. This negatively impacts gas diffusion between clutch ai r and sand, and between sand and atmosphere. Although sea turtle embryos evolved adaptations to survive adversities such as increased hypoxia, beach nourishment projects are pushing them beyond their physiological limitations. The severity of this impact increases in the second half of incubation, especially between incubation days 42-49 wh en metabolic oxygen demand and carbon dioxide offloading peak. This is the mo st critical period for the surviv al of sea turtle embryos in nourished beaches with high sand compaction. Unfortuna tely, this often leads to increased clutch mortality. Unbound sand calcium carbonate is a parameter that must also be carefully regulated because it can have a dual influence in sea turtle incubation: high concentrations increase sand compaction, but also buffer carbon dioxide by form ing carbonic acid. However, this reaction only occurred after incubation day 42 and when carbon dioxide levels were extremely high. The exposure to elevated carbon dioxide and lo w oxygen caused significant differences in body width and blood chemistries of hatchlings. Th ese hatchlings were thinner, slower, more dehydrated and metabolized more yolk energy than controls. Obviously th is compromises their chances to survive. After emergence, hatchl ings enter the ocean and begin their swimming frenzy. This period in the life of a sea turtle life has very high selection pressure and beach nourishment projects should not create incubation conditions that compromise their chances to survive. 133

PAGE 134

It is important to understand how differen t beach nourishment practices can impact different aspects of sea turtle nesting, incubation, and hatchli ng physiology in order to minimize negative effects. Understandi ng the correlations between sand properties and sea turtle survivorship is paramount for Floridas Department of Environmental Protections Beaches and Coastal Systems Program in order to better manage beach nourishment projects to meet sea turtle and human needs. 134

PAGE 135

APPENDIX A GRAIN SIZE CLASS DATA FOR 10 FL ORIDA SEA TURTLE NESTING BEACHES 135

PAGE 136

A-1. Grain size class data for Ca pe San Blas Beach. Sand was sort ed in a graduated nest of US Standard Sieves and analyzed for arithmetical (A rith), geometrical (Geo), and logarithmic (Log) statistics using the Me thod of Moments and the Folk & Ward Method. Cape San Blas sand is defined as bimodal, moderately sorted very fine sand. m Grain Size Distribution Mode 1: 76.5 3.73 Gravel:0.0% Coarse Sand: 1.2% Mode 2: 302.5 1.75 Sand: 100.0% Medium Sand: 36.9% Mode 3: Mud: 0.0% Fine Sand: 0.0% D10: 66.74 1.61 Very Fine Sand: 61.9% Median/D50: 84.06 3.57 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 326.8 3.91 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 4.90 2.42 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 260.0 2.29 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 3.89 2.08 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 210.6 1.96 Very Coarse Sand: 0.0% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 166.9 128.5 2.96 121.5 3.04 Very Fine Sand Sorting 121.4 1.98 0.99 1.87 0.90 Moderately Sorted Skewness 1.14 0.55 -0.55 0.71 -0.71 Very Coarse Skewed Kurtosis 5.42 1.43 1.43 0.50 0.50 Very Platykurtic 136

PAGE 137

A-2. Grain size distribution for Cape San Blas. Sand was sorted in a graduated nest of US Standard Sieves and is defined as bim odal, moderately sorted very fine sand. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter ( mm ) 137

PAGE 138

A-3. Grain size class data for A nna Maria Island Beach. Sand was sorted in a graduated nest of US Standard Sieves and analyzed for arithmeti cal (Arith), geometrical (Geo), and logarithmic (Log) statistics using the Met hod of Moments and the Folk & Ward Method. Anna Maria Island sand is defined as trimodal, poorly so rted very fine sand. m Grain Size Distribution Mode 1: 76.5 3.73 Gravel:0.0% Coarse Sand: 6.7% Mode 2: 1200 -0.24 Sand: 100.0% Medium Sand: 11.5% Mode 3: 302.5 1.75 Mud: 0.0% Fine Sand: 0.0% D10: 66.4 -0.16 Very Fine Sand: 67.1% Median/D50: 82.2 3.60 Very Coarse Gr avel: 0.0% Very Coarse Silt: 0.0% D90: 1114 3.91 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 16.8 -25.02 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 1048 4.07 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 4.43 2.30 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 246.9 2.15 Very Coarse Sand: 14.8% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 303.7 152.1 2.72 155.4 2.69 Fine Sand Sorting 399.0 2.89 1.53 2.76 1.47 Poorly Sorted Skewness 1.58 1.04 -1.04 0.84 -0.84 Very Coarse Skewed Kurtosis 3.85 2.38 2.38 0.82 0.82 Platykurtic 138

PAGE 139

A-4. Grain size distribution for A nna Maria Island. Sand was sorted in a graduated nest of US Standard Sieves and is defined as tr imodal, poorly sorted very fine sand. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 139

PAGE 140

A-5. Grain size class data for Ca sey Key. Sand was sorted in a gr aduated nest of US Standard Sieves and analyzed for arithmetical (Arith), ge ometrical (Geo), and logarithmic (Log) statistics using the Method of Moments and the Folk & Ward Method. Casey Key sand is defined as trimodal, poorly sorted very coarse sand. m Grain Size Distribution Mode 1: 1200 -0.24 Gravel:0.0% Coarse Sand: 16.0% Mode 2: 605 0.75 Sand: 100.0% Medium Sand: 11.4% Mode 3: 302.5 1.75 Mud: 0.0% Fine Sand: 0.0% D10: 250.7 -0.41 Very Fine Sand: 9.9% Median/D50: 1070 -0.10 Very Coarse Gr avel: 0.0% Very Coarse Silt: 0.0% D90: 1327 2.00 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 5.29 -4.89 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 1076 2.40 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 2.26 -3.03 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 682.3 1.18 Very Coarse Sand: 62.7% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 890.9 689.2 0.54 745.6 0.42 Coarse Sand Sorting 422.2 2.38 1.25 2.23 1.16 Poorly Sorted Skewness -0.82 -1.60 1.60 -0.79 0.79 Very Fine Skewed Kurtosis 2.02 4.40 4.40 1.46 1.46 Leptokurtic 140

PAGE 141

A-6. Grain size distribution for Ca sey Key. Sand was sorted in a graduated nest of US Standard Sieves and is defined as trimodal, poorly sorted very coarse sand. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000 P a r t ic le D i a m e t e r ( m m ) 141

PAGE 142

A-7. Grain size class data for Pa rkshore Beach, Naples. Sand was sorted in a graduated nest of US Standard Sieves and analyzed for arithmeti cal (Arith), geometrical (Geo), and logarithmic (Log) statistics using the Met hod of Moments and the Folk & Ward Method. Parkshore Beach sand is defined as trimodal, poorly so rted very fine sand. m Grain Size Distribution Mode 1: 76.5 3.73 Gravel:0.0% Coarse Sand: 12.8% Mode 2: 302.5 1.75 Sand: 100.0% Medium Sand: 38.9% Mode 3: 605.0 0.75 Mud: 0.0% Fine Sand: 0.0% D10: 66.4 0.70 Very Fine Sand: 43.6% Median/D50: 264.8 1.92 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 613.9 3.87 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 8.98 5.50 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 545.5 3.17 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 4.29 2.32 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 254.5 2.10 Very Coarse Sand: 4.7% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 284.7 190.6 2.39 214.8 2.22 Fine Sand Sorting 267.2 2.41 1.27 2.35 1.23 Poorly Sorted Skewness 1.90 0.23 -0.23 -0.25 0.25 Fine Skewed Kurtosis 6.87 1.79 1.79 0.67 0.67 Very Platykurtic 142

PAGE 143

A-8. Grain size distribution for Parkshore Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defined as trim odal, poorly sorted very fine sand. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 143

PAGE 144

A-9. Grain size class data for Delray Beach. Sand was sorted in a graduated nest of US Standard Sieves and analyzed for arithmetical (A rith), geometrical (Geo), and logarithmic (Log) statistics using the Me thod of Moments and the Folk & Wa rd Method. Delray Beach sand is defined as polymodal, poorly sorted very fine sand. m Grain Size Distribution Mode 1: 76.5 3.73 Gravel:0.0% Coarse Sand: 14.8% Mode 2: 302.5 1.75 Sand: 100.0% Medium Sand: 35.1% Mode 3: 605 0.75 Mud: 0.0% Fine Sand: 0.0% D10: 69.1 -0.05 Very Fine Sand: 38.9% Median/D50: 279.4 1.84 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 1038 3.86 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 15.04 -70.85 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 969.4 3.91 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 6.46 3.79 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 433.0 2.69 Very Coarse Sand: 11.3% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 360.4 225.9 2.15 234.7 2.09 Fine Sand Sorting 347.4 2.64 1.40 2.67 1.42 Poorly Sorted Skewness 1.43 0.15 -0.15 -0.12 0.12 Fine Skewed Kurtosis 4.05 1.74 1.74 0.64 0.64 Very Platykurtic 144

PAGE 145

A-10. Grain size distribution for Delray Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defined as polymodal, poorly sorted very fine sand. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 145

PAGE 146

A-11. Grain size class data for Juno Beach. Sand wa s sorted in a graduated nest of US Standard Sieves and analyzed for arithmetical (Arith), ge ometrical (Geo), and logarithmic (Log) statistics using the Method of Moments and the Folk & Ward Method. Juno Beach sand is defined as polymodal, moderately sorted medium sand. m Grain Size Distribution Mode 1: 302.5 1.75 Gravel:0.0% Coarse Sand: 24.4% Mode 2: 605.0 0.75 Sand: 100.0% Medium Sand: 51.4% Mode 3: 76.50 3.73 Mud: 0.0% Fine Sand: 0.0% D10: 79.12 0.52 Very Fine Sand: 15.7% Median/D50: 316.0 1.66 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 695.2 3.66 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 8.79 6.98 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 616.1 3.14 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 2.10 2.29 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 294.1 1.07 Very Coarse Sand: 8.5% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 417.6 320.1 1.64 369.7 1.44 Medium Sand Sorting 293.0 2.13 1.09 1.93 0.95 Moderately Sorted Skewness 1.41 -0.52 0.52 0.21 -0.21 Coarse Skewed Kurtosis 4.68 2.99 2.99 1.54 1.54 Very Leptokurtic 146

PAGE 147

A-12. Grain size distribution for Juno Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defined as polymodal, moderately sorted medium sand. 0.0 10.0 20.0 30.0 40.0 50.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 147

PAGE 148

A-13. Grain size class data for a native secti on of Melbourne Beach. Sand was sorted in a graduated nest of US Standard Si eves and analyzed for arithmetical (Arith), geometrical (Geo), and logarithmic (Log) statistics using the Method of Moments and the Folk & Ward Method. Melbourne Beach native sand is defined as trimodal, poorly sorted medium sand. m Grain Size Distribution Mode 1: 302.5 1.75 Gravel:0.0% Coarse Sand: 11.5% Mode 2: 76.50 3.73 Sand: 100.0% Medium Sand: 60.1% Mode 3: 605.0 0.75 Mud: 0.0% Fine Sand: 0.0% D10: 72.75 0.77 Very Fine Sand: 24.8% Median/D50: 289.6 1.79 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 584.5 3.78 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 8.03 4.88 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 511.7 3.01 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 1.34 1.27 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 84.79 0.42 Very Coarse Sand: 3.6% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 313.8 241.2 2.05 200.9 2.32 Fine Sand Sorting 228.7 2.10 1.07 2.06 1.04 Poorly Sorted Skewness 2.07 -0.38 0.38 -0.50 0.50 Very Fine Skewed Kurtosis 8.78 2.54 2.54 3.24 3.24 Extremely Leptokurtic 148

PAGE 149

A-14. Grain size distribution for a native secti on of Melbourne Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defined as trimodal, poorly sorted medium sand. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 149

PAGE 150

A-15. Grain size class data for a nourished sect ion of Melbourne Beach. Sand was sorted in a graduated nest of US Standard Si eves and analyzed for arithmetical (Arith), geometrical (Geo), and logarithmic (Log) statistics using the Method of Moments and the Folk & Ward Method. Melbourne Beach nourished sand is defined as polymodal, poorly sorted medium sand. m Grain Size Distribution Mode 1: 302.5 1.75 Gravel:0.0% Coarse Sand: 10.8% Mode 2: 76.50 3.73 Sand: 100.0% Medium Sand: 54.4% Mode 3: 605.0 0.75 Mud: 0.0% Fine Sand: 0.0% D10: 72.57 0.51 Very Fine Sand: 25.2% Median/D50: 293.3 1.77 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 700.6 3.78 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 9.65 7.37 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 628.0 3.27 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 3.84 2.26 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 254.9 1.94 Very Coarse Sand: 9.6% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 364.2 259.1 1.95 237.3 2.08 Fine Sand Sorting 310.1 2.30 1.20 2.53 1.34 Poorly Sorted Skewness 1.72 -0.13 0.13 -0.17 0.17 Fine Skewed Kurtosis 5.28 2.40 2.40 0.87 0.87 Platykurtic 150

PAGE 151

A-16. Grain size distribution for a nourished section of Melbourne Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defined as polymodal, poorly sorted medium sand. 0.0 10.0 20.0 30.0 40.0 50.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 151

PAGE 152

A-17. Grain size class data for Kennedy Space Cent er Beach. Sand was sorted in a graduated nest of US Standard Sieves and analyzed for arithmetical (Arith), geometrical (Geo), and logarithmic (Log) statistics using the Met hod of Moments and the Folk & Ward Method. Kennedy Space Center sand is defined as tr imodal, poorly sorted medium sand. m Grain Size Distribution Mode 1: 302.5 1.75 Gravel:0.0% Coarse Sand: 20.1% Mode 2: 605.0 0.75 Sand: 100.0% Medium Sand: 62.3% Mode 3: 76.50 3.73 Mud: 0.0% Fine Sand: 0.0% D10: 77.58 0.73 Very Fine Sand: 17.1% Median/D50: 300.8 1.73 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 601.6 3.69 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 7.75 5.03 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 524.0 2.95 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 1.33 1.27 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 84.95 0.41 Very Coarse Sand: 0.5% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 329.1 272.4 1.88 242.9 2.04 Fine Sand Sorting 174.9 1.92 0.94 2.21 1.15 Poorly Sorted Skewness 0.91 -0.92 0.92 -0.33 0.33 Very Fine Skewed Kurtosis 5.18 3.18 3.18 3.26 3.26 Extremely Leptokurtic 152

PAGE 153

A-18. Grain size distribution for Kennedy Space Cent er Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defi ned as trimodal, poorly sorted medium sand. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (mm) 153

PAGE 154

A-19. Grain size class data for Flagler Beach. Sand was sorted in a graduated nest of US Standard Sieves and analyzed for arithmetical (A rith), geometrical (Geo), and logarithmic (Log) statistics using the Meth od of Moments and the Folk & Ward Me thod. Flagler sand is defined as polymodal, poorly sorted coarse sand. m Grain Size Distribution Mode 1: 605.0 0.75 Gravel:0.0% Coarse Sand: 38.1% Mode 2: 302.5 1.75 Sand: 100.0% Medium Sand: 28.9% Mode 3: 1200 -0.24 Mud: 0.0% Fine Sand: 0.0% D10: 81.32 -0.23 Very Fine Sand: 14.0% Median/D50: 533.7 0.91 Very Coarse Gravel: 0.0% Very Coarse Silt: 0.0% D90: 1173.0 3.62 Coarse Gravel : 0.0% Coarse Silt: 0.0% (D90 / D10): 14.42 -15.73 Medium Gravel: 0.0% Medium Silt: 0.0% (D90 D10): 1091.7 3.85 Fine Gravel: 0.0% Fine Silt: 0.0% (D75 / D25): 2.35 3.15 Very Fine Gravel: 0.0% Very Fine Silt: 0.0% (D75 D25): 386.2 1.23 Very Coarse Sand: 19.0% Clay: 0.0% Method of Moments Folk & Ward Method Arith. Geo. Log. Geo. Log. Description m m m Mean 556.8 416.0 1.27 524.5 0.93 Coarse Sand Sorting 361.3 2.31 1.20 2.21 1.14 Poorly Sorted Skewness 0.65 -0.81 0.81 -0.22 0.22 Fine Skewed Kurtosis 2.42 2.94 2.94 1.38 1.38 Leptokurtic 154

PAGE 155

A-20. Grain size distribution for Flagler Beach. Sand was sorted in a graduated nest of US Standard Sieves and is defined as polymodal, poorly sorted coarse sand. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 -7.0 -5.0 -3.0 -1.0 1.0 3.0 5.0Particle Diameter ( )Class Weight (%) 100 1000 10000 100000Particle Diameter (m) 155

PAGE 156

156 APPENDIX B INORGANIC METAL CONCENTRATIONS IN SEA TURTLE NESTING BEACH SAND

PAGE 157

B-1. Inorganic metal concentrations measured in sand from 18 west coast of Florida sea turtle nesting beaches. Some beaches had native and nourished sand and were sample d at both sections. Sampling was conducte d during the last week of July 2004 and represents conditions at time of sampling. Inorganic metals were anal yzed by Inductively Coupled Plasma Mass Spectrometry as outlined in EPA 200.7/6010 procedure for sediment analysis. Units are g/g. Ag Al As B Ba Be Ca Cd Co Cr Fort Pickens State Park <0.000 0.075 <0.000 0.002 0.002 0.003 0.980 0.001 0.003 0.672 Pensacola Beach <0.000 0.160 <0.000 0.004 0.002 0.003 3.690 <0.000 0.003 0.657 Fort Walton Beach <0.000 0.253 <0.000 0.004 0.006 <0.000 0.433 <0.000 <0.000 0.708 Panama City Beach-Nourished <0.000 0.215 <0.000 0.004 0.003 0.002 27.40 <0.000 <0.000 0.712 Panama City Beach-Native <0.000 0.268 <0.000 0.204 2.068 0.004 0.622 0.001 0.003 5.452 Mexico Beach <0.000 1.660 <0.000 0.122 1.131 <0.000 2.694 <0.000 <0.000 4.030 Cape San Blas <0.000 0.947 <0.000 0.185 1.706 <0.000 2.172 <0.000 <0.000 5.522 Anna Maria Island-Nourished <0.000 0.631 7.394 0.192 1.665 0.005 118.9 <0.000 <0.000 4.712 Coquina Beach Park <0.000 2.732 44.04 0.053 0.055 0.002 530.2 <0.000 <0.000 0.376 Lido Beach <0.000 11.26 228.4 0.228 0.190 <0.000 3118 <0.000 <0.000 0.089 Siesta Key Beach <0.000 0.464 2.072 0.115 1.161 0.003 28.26 <0.000 0.004 3.342 Casey Key <0.000 4.352 322.4 0.109 0.250 0.031 3754 <0.000 <0.000 0.330 Venice Beach <0.000 4.662 10.34 0.228 1.618 <0.000 962.8 <0.000 <0.000 4.060 Captiva Island Public Beach <0.000 3.418 280.0 0.124 0.239 0.019 3460 <0.000 <0.000 0.335 Sanibel Island/Blind Pass Beach <0.000 8.374 350.8 0.391 0.303 <0.000 3510 <0.000 <0.000 0.076 Sanibel Island/Lighthouse Beach <0.000 1.829 222.2 0.067 0.134 0.024 2262 <0.000 <0.000 0.338 Parkshore Beach-Native <0.000 3.844 9.670 0.045 0.053 0.001 969.6 0.051 <0.000 0.862 Parkshore Beach-Nourished <0.000 3.022 <0.000 0.023 0.081 <0.000 17.48 0.038 <0.000 0.330 Vanderbuilt Beach-Nourished <0.000 3.804 <0.000 0.017 0.142 0.002 23.98 0.019 <0.000 0.678 Marco Island <0.000 4.588 208.0 0.053 0.200 0.019 2306 <0.000 <0.000 0.369 157

PAGE 158

B-1. Continued Cu Fe K Mg Mn Mo Pb Sb Se Si Zn Fort Pickens State Park 0.010 0.157 11.25 0.518 0.002 0.003 0.073 <0.000 0.059 0.014 0.001 Pensacola Beach 0.006 0.213 11.29 0.867 0.005 0.003 0.077 0.018 <0.000 0.020 0.002 Fort Walton Beach 0.006 0.378 11.39 0.618 0.005 0.003 0.088 <0.000 0.022 0.021 0.005 Panama City Beach-Nourished 0.008 1.087 11.51 3.408 0.002 0.003 0.148 <0.000 0.023 0.016 0.004 Panama City Beach-Native 0.001 0.228 13.20 0.930 0.002 0.003 0.086 <0.000 0.001 0.055 0.467 Mexico Beach <0.000 0.685 12.55 4.554 0.007 0.002 0.127 <0.000 0.047 0.029 0.318 Cape San Blas <0.000 2.188 12.91 3.064 0.029 0.003 0.209 <0.000 0.052 0.029 0.456 Anna Maria Island-Nourished <0.000 0.436 12.82 9.870 0.007 0.012 0.157 <0.000 0.024 0.014 0.384 Coquina Beach Park 0.007 3.532 12.51 35.36 0.066 0.053 0.505 <0.000 0.050 0.500 0.029 Lido Beach 0.009 46.32 14.09 519.8 0.272 0.174 5.658 <0.000 0.148 0.091 0.054 Siesta Key Beach <0.000 0.282 12.28 6.772 0.005 0.003 0.114 <0.000 0.014 0.014 0.241 Casey Key 0.005 12.10 13.30 565.2 0.181 0.174 4.424 <0.000 0.059 0.106 0.040 Venice Beach <0.000 8.936 13.60 87.96 0.126 0.073 1.090 <0.000 0.043 0.181 0.371 Captiva Island Public Beach 0.004 16.95 13.25 265.6 0.170 0.135 2.716 <0.000 0.118 0.187 0.037 Sanibel Island/Blind Pass Beach 0.003 27.50 14.40 408.4 0.223 0.158 3.856 <0.000 0.055 0.287 0.044 Sanibel Island/Lighthouse Beach 0.002 2.086 13.35 132.8 0.073 0.099 1.093 <0.000 0.031 0.617 0.040 Parkshore Beach-Native <0.000 3.652 11.77 59.92 0.045 0.078 0.643 <0.000 0.022 0.033 0.072 Parkshore Beach-Nourished 0.013 3.068 12.17 4.478 0.006 0.021 0.209 <0.000 0.080 0.120 0.024 Vanderbuilt Beach-Nourished 0.008 2.772 12.18 3.616 0.009 0.023 0.190 <0.000 <0.000 0.049 0.090 Marco Island 0.004 5.306 13.12 142.5 0.109 0.106 1.271 <0.000 0.040 0.071 0.046 158

PAGE 159

B-2 Inorganic metal concentrations measured in sand from 21 eas t coast of Florida sea turtle ne sting beaches. Some beaches ha d native and nourished sand and were sample d at both sections. Sampling was conducte d during the last week of July 2004 and represents conditions at time of sampling. Inorganic metals were anal yzed by Inductively Coupled Plasma Mass Spectrometry as outlined in EPA 200.7/6010 procedure for sediment analysis. Units are g/g. Ag Al As B Ba Be Ca Cd Co Cr Key West/Smathers Beach <0.000 4.478 122.0 0.087 0.127 <0.000 612.6 <0.000 <0.000 0.487 Bahia Honda (Ocean side) <0.000 2.056 12.64 0.415 0.936 0.023 3220 <0.000 <0.000 3.732 Crandon Park South <0.000 2.490 330.6 0.174 0.080 0.015 2644 <0.000 <0.000 0.245 Spanish Park Beach <0.000 3.364 13.04 0.203 0.809 <0.000 1754 <0.000 <0.000 3.056 Delray Beach <0.000 6.730 13.43 0.144 0.149 <0.000 2868 0.002 <0.000 <0.000 Riviera Beach <0.000 5.596 238.4 0.133 0.148 <0.000 3100 <0.000 <0.000 0.240 Juno Beach <0.000 9.992 14.59 0.292 0.184 <0.000 3876 <0.000 <0.000 <0.000 Fort Pierce Inlet-South <0.000 14.00 12.90 0.312 0.230 <0.000 3228 0.003 <0.000 <0.000 Fort Pierce Inlet-North <0.000 4.802 154.4 0.126 0.083 <0.000 2350 <0.000 <0.000 0.091 Vero Beach-Native <0.000 3.932 191.9 0.150 0.075 <0.000 2972 <0.000 <0.000 0.178 Vero Beach-Nourished <0.000 14.51 61.24 0.048 0.048 <0.000 424.4 <0.000 <0.000 0.273 Sebastian Inlet-South <0.000 5.394 9.790 0.086 0.053 <0.000 1653 0.023 <0.000 0.452 Sebastian Inlet-North <0.000 7.890 225.4 0.188 0.121 <0.000 3468 0.128 <0.000 0.173 Melbourne Beach-Native <0.000 5.562 125.1 0.155 0.082 <0.000 2410 0.080 <0.000 0.400 Melbourne Beach-Nourished <0.000 11.78 270.8 0.268 0.155 <0.000 2810 <0.000 <0.000 0.052 Patrick Air Force Beach <0.000 9.538 190.9 0.191 0.111 <0.000 2244 0.034 <0.000 <0.000 Jetty Maritime Park Beach <0.000 11.12 170.6 0.101 0.056 <0.000 491.2 0.013 <0.000 0.271 Cape Canaveral Air Force Beach <0.000 8.074 148.3 0.119 0.068 <0.000 1285 0.026 <0.000 0.469 Kennedy Space Center <0.000 2.592 <0.000 0.062 0.047 <0.000 1410 0.039 <0.000 0.482 Canaveral National Seashore <0.000 5.084 9.528 0.103 0.065 <0.000 1554 0.024 <0.000 0.517 Flagler Beach <0.000 3.060 8.746 0.120 0.059 <0.000 2730 <0.000 <0.000 0.704 Matanzas Inlet Beach <0.000 4.856 3.404 0.050 0.035 <0.000 238.8 0.051 <0.000 0.742 Anastasia State Park <0.000 4.386 2.812 0.047 0.049 <0.000 78.08 0.001 <0.000 0.645 St. Augustine Inlet/Vilano Beach <0.000 3.626 10.67 0.076 0.059 <0.000 1883 <0.000 <0.000 0.728 159

PAGE 160

160 B-2. Continued Cu Fe K Mg Mn Mo Pb Sb Se Si Zn Key West/Smathers Beach 0.011 28.94 12.79 36.62 0.150 0.063 1.552 <0.000 <0.000 0.076 0.047 Bahia Honda (Ocean side) <0.000 2.590 14.36 998.0 0.129 0.178 5.940 <0.000 <0.000 0.072 0.372 Crandon Park South 0.011 12.21 13.19 825.0 0.148 0.180 5.886 <0.000 0.005 0.130 0.081 Spanish Park Beach <0.000 17.77 12.82 270.0 0.219 0.096 2.522 <0.000 <0.000 0.019 0.321 Delray Beach <0.000 27.52 12.62 436.6 0.397 0.150 4.038 <0.000 0.048 0.052 0.054 Riviera Beach <0.000 37.68 13.64 264.2 0.293 0.143 3.542 <0.000 <0.000 0.065 0.095 Juno Beach <0.000 28.92 12.72 1450 0.532 0.220 9.226 <0.000 0.143 0.068 0.036 Fort Pierce Inlet-South <0.000 89.06 13.62 681.2 0.768 0.189 6.924 <0.000 <0.000 0.035 0.066 Fort Pierce Inlet-North 0.006 38.82 13.06 160.9 0.296 0.108 2.970 <0.000 0.035 0.115 0.054 Vero Beach-Native <0.000 26.06 12.58 113.7 0.334 0.113 1.934 <0.000 <0.000 0.033 0.037 Vero Beach-Nourished <0.000 11.12 12.31 27.98 0.108 0.072 0.751 <0.000 <0.000 0.035 0.033 Sebastian Inlet-South 0.001 24.66 12.09 133.2 0.312 0.099 1.984 <0.000 <0.000 0.041 0.044 Sebastian Inlet-North <0.000 54.38 13.12 180.1 0.549 0.137 3.936 <0.000 0.064 0.067 0.058 Melbourne Beach-Native <0.000 45.28 13.34 179.8 0.494 0.125 3.232 <0.000 <0.000 0.049 0.120 Melbourne Beach-Nourished <0.000 82.74 13.93 452.8 0.920 0.160 5.954 <0.000 0.060 0.066 0.079 Patrick Air Force Beach <0.000 62.10 12.97 393.4 0.867 0.148 5.200 <0.000 <0.000 0.338 0.062 Jetty Maritime Park Beach <0.000 30.64 13.78 128.4 0.333 0.077 2.528 <0.000 0.036 0.068 0.077 Cape Canaveral Air Force Beach <0.000 38.02 12.86 169.1 0.427 0.102 2.808 <0.000 <0.000 0.070 0.121 Kennedy Space Center <0.000 40.05 12.39 69.68 0.198 0.070 1.367 <0.000 <0.000 0.028 0.070 Canaveral National Seashore <0.000 35.40 12.37 143.7 0.451 0.091 2.528 <0.000 <0.000 0.046 0.045 Flagler Beach <0.000 39.68 11.86 55.50 0.315 0.097 2.310 <0.000 0.004 0.031 0.034 Matanzas Inlet Beach 0.005 15.77 12.18 48.96 0.130 0.048 1.026 <0.000 0.057 0.050 0.051 Anastasia State Park <0.000 13.20 12.07 34.54 0.139 0.026 0.915 <0.000 <0.000 0.045 0.040 St. Augustine Inlet/Vilano Beach <0.000 22.50 12.08 279.4 0.425 0.116 2.846 <0.000 0.053 0.058 0.039

PAGE 161

LIST OF REFERENCES Ackerman, R. 1975. Diffusion and gas exchange of sea turtle eggs. Ph.D. dissertation, University of Florida, Gainesville, Florida, 124 pp. Ackerman, R. 1977. The respiratory gas exchange of sea turtle nests ( Chelonia, Caretta ). Respiration Physiology 31, 19-38. Ackerman, R. 1980. Physiological and ecological aspects of gas exchange by sea turtle eggs. American Zoologist 20, 575-583. Ackerman, R. 1981. Growth and gas ex change of embryonic sea turtles ( Chelonia, Caretta ). Copeia 4, 757-765. Ackerman, R., and H. D. Prange. 1972. Oxygen diffusion across a sea turtle ( Chelonia mydas ) egg shell. Comparative Bioche mistry and Physiology 43, 905-909. Ackerman, R., R. Seagrave, R. Dmi'el, and A. Ar. 1985. Water and heat exchange between parchment-shelled reptile eggs and their surroundings. Copeia 3, 703-711. Ackerman, R., T. Rimkus, and R. Horton. 1992. Hydr ic and thermal characteristics of natural and renourished sea turtle nesti ng along the Atlantic coast of Florida. Unpublished report submitted to Florida Department of Natural Resources Contract # 6407, Iowa State University, Ames, Iowa, 125 pp. Bagley, D., T. Cascio, R. Owen, S. Johnson, and L. Ehrhart. 1994. Marine turtle nesting at Patrick Air Force Base, Flor ida; 1987-1993: trends and issues. In: Bjorndal, K., A. Bolten, and P. Eliazar (Eds.), Fourteenth annual symposium on sea turtle biology and conservation. National Oceanographic and Atmospheric Administration, Washington D.C., pp. 180-181. Blake, G.R., and K.H. Hartge. 1986. Bulk density. In : Klute, E. (Ed.), Met hods of soil analysis: Part I: Physical and mineralogical methods. Soil Science Society of America Series No. 5. Madison, Wisconsin, pp. 363-376. Blott, S., and K. Pye. 2001. Gradistat: A grain si ze distribution and statis tics package for the analysis of unconsolidated sediments. Earth Surface Processes and Landforms. 26, 12371248. Bradford. J.M. 1986. Penetrability. In: Klute, E. (E d.), Methods of soil analys is: Part I: Physical and mineralogical methods. Soil Science Soci ety of America Series No. 5. Madison, Wisconsin, pp. 463-472. Broadwell, A.L. 1992. Effects of beach renouri shment on the survival of loggerhead sea turtle nests. In: Salmon, M. and J. Wyneken (compilers). Proceedings of the eleventh annual workshop on sea turtle biology and conservation. NOAA Tech. Mem. NMFSSEFSC-302, pp. 21-23. 161

PAGE 162

Bruun, P. 2001. The development of downdrift erosion: an update of paper in JCR, Vol. 11 (4), Journal of Coastal Research 17 (1), 82-89. Bustard, H. R., and P. Greenham. 1968. Physical and chemical factors affecting hatching in the green turtle, Chelonia mydas Ecology 49, 269-276. Carr, A. F. 1967. So excellent a fishe. Natura l History Press, Garden City, New York, 248 pp. Carthy, R. 1996. The role of the eggshell and nest chamber in loggerhead turtle ( Caretta caretta ) egg incubation. Ph.D. dissertati on, University of Florida, Ga inesville, Florida, 121 pp. Clark, R.R. 1989. Beach conditions in Florida: A st atewide inventory and identification of the beach erosion problem areas in Florida. Beaches and Shores Technical and Design Memorandum, 89, 1-167. Clesceri L., A. Greenberg and A. Eato n (Eds.), 1999. Standard methods for examination of water & wastewater: 20 th Edition. American Public Health Association. United Book Press, Inc., Baltimore, Maryland, 1325 pp. Cornelisen, C. 1996. Effects of beach renourishmen t on physical attributes of a sea turtle nesting beach, east-central Florida, USA. M. S. Thesis. Florida In stitute of Technology, Melbourne, Florida, 109 pp. Cotter, S. 2001. Hematology: Quick look series in veterinary medicine. Teton NewMedia, Jackson Hole, Wyoming, 137 pp. Council, N. R. 1995. Environmental issues associated with beach nourishment. Beach Nourishment and Protection. C. o. b. n. a. protection. National Academy Press, Washington D.C., pp. 107-126. Crain, D.A., A.B. Bolten and K.A. Bjorndal. 1995. Effects of beach nourishm ent on sea turtles: Review and research initiatives. Restoration Ecology 3 (2), 95-104. Davis, R. and H. Bennett. 1927. Grouping of soils on the basis of mechanical analysis. United States Department of Agriculture Departmental Circulation No. 419. Dean, W. 1974. Determination of carbonate and or ganic matter in calcareous sediments and sedimentary rock by loss on ignition: co mparison with other methods. Journal of Sedimentary Petrology 44 (1), 242-248. Dean, R. 2002. Beach Nourishment: Theory and Pr actice. World Scientific, River Edge, New Jersey, 399 pp. Denny, M. 1993. Air and Water: The biology and physics of lifes media. Princeton University Press, Princeton, New Jersey, 360 pp. 162

PAGE 163

EAI (Ecological Associates, Inc). 1999. Martin county beach nourishment pr oject, sea turtle monitoring and studies, 1997 annual report and fi nal assessment. Ecological Associates, Inc., Jensen Beach, Florida, 323 pp. Ehrhart, L. M. and K. Roberts. 2001. Marine Tur tle nesting and reproductive success at Patrick Air Force Base; Summer 2001, 45 CES/CEV: 39. Etchberger C.R., M.A. Ewert, J. Phillips, C. Nelson, and H. Prange. 1992. Physiological response to carbon dioxide in embr yonic red-eared slider turtles Trachemys scripta Journal of Experimental Zoology 4 (1), 1-10. Etchberger C.R., M.A. Ewert, J. Phillips, C. Nelson, and H. Prange. 1993. Environmental and maternal influences of embryoni c pigmentation in a turtle ( Trachemys scripta elegans ). Journal of Zoology 230, 529-539 Part 4. Farouki, O.T. 1986. Thermal Properties of Soils. In, Series on Rock and Soil Mechanics. Trans Tech Pub. Clausthal-Zekkerfekdm Germany 11, 12-27. Fletemeyer, J. R. 1983. The impact of beach ren ourishment on sea turtle nesting. In: Tait, L. S. (Ed), 1983 Joint annual meeting of the American Shore and Beach Preservation Association and Florida Shore and Beach Pres ervation Association: The new threat to beach preservation, Boca Raton, Florida, pp. 168-177. Folk, R.L. 1966. A Review of grain-si ze parameters. Sedimentology 6, 73-93. Folk, R. and W. Ward. 1957. Br azos River Bar: A study in th e significance of grain size parameters. Journal of Sedimentary Petrology 27 (1), 3-26. Gardner, W.H. 1986. Water content. In: Klute, E. (E d.), Methods of soil analys is: Part I: Physical and mineralogical methods. Soil Science Soci ety of America Series No. 5. Madison, Wisconsin, pp. 493-544. Harrison, S.J. and P. Morrison. 1993. Temperature in a sandy beach under strong solar heating; Patara Beach, Turkey. Estuarine, Coastal and Shelf Science. 37, 89-97. Henderson, T., R.H Merry and R Murray. 2001. Reactivity of the Southern Mu rray Mallee soil carbonates. Proceedings of the 10 th Australian Agronomy Conf erence, Australian Society of Agronomy. Hobart, Tasmania, 354 pp. Horvth, B., O. Opara-Nadi, and F. Beese. 2005 A simple method for measuring the carbonate content of soils. Soil Science Society of America Journal 69, 1066-1068. Jackson, D., A. L. Ramsey, J. M. Paulson, C. E. Crocker, and G. R. Ultsch. 2000. Lactic acid buffering by bone and shell in anoxic softshell and painted turtles. Physiological and Biochemical Zoology 73, 290-297. 163

PAGE 164

Kam, Y. 1993. Physiological effects of hypoxia on metabolism and growth of turtle embryos. Respiratory Physiology May: 92 (2), 127-138. Kraimer, R. A., H. Curtis Monger, and Robert L. Steiner. 2005. Mineralogical distinctions of carbonates in desert soils. Soil Science Society of America Journal 69, 1773-1781. Kubota, R., T. Kunito, and S. Tanabe. 2002. Chemi cal speciation of arsenic in the livers of higher trophic marine animals. Marine Pollution Bulletin 45, 218-223. Krumbein, W. C., and F. J. Pettijohn. 1938. Manual of sedimentary petrography. D. AppletonCentury, New York, New York, 549 pp. Landi, A., A. R. Mermut, and D. W. Ande rson. 2004. Carbon distribution in a hummocky landscape from Saskatchewan, Canada. Soil Science Society of America Journal 68, 175184. Lucas, L. L. 2000. Results and assessment of a ph ysical monitoring program associated with beach nourishment on high-density marine turt le nesting beaches in the center of the Archie Carr National Wildlife Refuge, Sebast ian, Florida, USA. M.S. thesis, Florida Institute of Technology, Melbourne, Florida, 294 pp. Lunde, G. 1977. Occurrence and transformation of arsenic in the marine environment. Environmental Health Perspectives 19, 47-52. Lutz, P. and G. Lapennas, 1982; Effects of pH, CO 2 and organic phosphates on oxygen affinity of sea turtle hemoglobins. Respiratory Physiology 48 (1), 75-87. Lutz, P., and A. Dunbar-Cooper. 1987. Variations in the blood chemistry of the loggerhead sea turtle, Caretta caretta Fishery Bulletin 85, 37-43. Lutz, P, A. Schulman, and S. Shaw. 1991. Fisher Island sea turtle project annual report. Submitted to Evans Environmental and Geological Science and Management, Inc by University of Miami, 35 pp. Magron, J.-P. R. 2000. Impact assessment of beac h nourishment on the physical environment of a high-density marine turtle ne sting beach, Sebastian Inlet, Florida. MS Thesis. Florida Institute of Technology. Melbourne, Florida, 172 pp. Mann, T. M. 1977. Impact of developed coastlin e on nesting and hatch ling sea turtles in southeastern Florida. MS Thesis. Florida A tlantic University, Bo ca Raton, Florida, 156 pp. McGehee, M. A. 1990. Effects of moisture on eggs and hatchlings of lo ggerhead sea turtles ( Caretta caretta ). Herpetologica 46, 251-258. 164

PAGE 165

Miller, J. D. 1982. Embryology of ma rine turtles. Ph.D. dissertati on, University of New England, Armidale, New South Wales, Australia v. 1, 213 pp., v. 2, 248 pp. Milton, S. and P. Lutz. 2003. Physiological and gene tic responses to environmental stress. In: Lutz, P., J. Musick, and J. Wyneken. (Eds), The biology of sea turtles, Vol II. CRC Press LLC. Boca Raton, Florida, pp. 163-198. Moreno, F., J.M. Murillo, F. Pelegrn, and I.F. Giron. 2006. Long-term impact of conservation tillage on stratification ratio of soil organic carbon and loss of total and active CaCO 3 Soil & Tillage Research 85, 86. Mortimer, J. A. 1990. The influence of beach sand characteristics on the nesting behavior and clutch survival of green turtles. Copeia 3, 802-817. Mosaddeghi, M.R., M.A. Hajabbasi, and H. Khademi. 2005. Tensile strength of sand, palygorskite and calcium carbonate mixtures and interpretation with the effective stress theory. Geoderma 134, 160-170. Mota, M., and B. Peterson. 2002. Beach renourishement and sea turtle nesting microenvironments. Presented at the 22 nd annual sea turtle symposium. Miami, Florida. Mota, M. and B. Peterson. 2003. Beach renourishment and its impact on gas concentrations in loggerhead sea turtle nests in Florida. Presented at the 23 rd annual sea turtle symposium. Kuala Lumpur, Malaysia. Mrosovsky, N. 1994. Sex ratios of sea turtles. The Journal of Experimental Zoology 270, 1627. Mrosovsky, N. and J. Provancha. 19 89. Sex ratio of loggerhead sea turtles hatching on a Florida beach. Canadian Journal of Zoology 67, 2533-2539. Mrosovsky, N. and C. L. Yntema. 1981. Temperature dependence of se xual differentiation in sea turtles: Implications for conservation practices. In: Bjorndal, K. (ed), Biology and conservation of sea turtles. Smithsonian Institution Press, Washington D.C., pp. 59-65. Nelson, D. A. 1985. Beach nourishment sand comp atibility with logge rhead sea turtle nesting. Fifth annual workshop on sea turtle biology and conservati on, Waverly, Georgia. Nelson, D. A. and K. Mauck. 1986. St. Lucie inlet dredged material disposal effects on the firmness of sand used by nesting turtles. Te chnical Report U.S. Army Corps of Engineers Waterways Experiment Stati on., Vicksburg, Mississippi. Nelson, D. A., Mauck, K., and J. Fletemeyer. 1987. Physical effects of beach nourishment on sea turtle nesting, Delray Beac h, Florida. Technical Repor t EL-87-15, USACE Waterways Experiment Station, Vicksburg, Mississippi. 165

PAGE 166

Paukstis, G. and F. Janzen. 1990. Sex determination in reptiles: Summary of effects of constant temperatures of incubation on sex ratios of offspring. Smithsonian Herpetological Information Service, Smithsonian: 29. Pelnard-Considere, R. 1956. Essai de theorie de l' evolution des formes de rivage en plages de Sable et de Galets. 4th J ournees de l'Hydraulique, Les Energies de la Mer. Raymond, P.W. 1984. The effects of beach restora tion on marine turtles nesting in South Brevard County, Florida. MS Thesis. University of Central Florida, Orlando, Florida, 121 pp. Rumbold, D. G., Davis, P. W., and C. Perretta. 2001. Estimating the effect of beach nourishment on Caretta caretta (loggerhead sea turt le) nesting. Restoratio n Ecology 9 (3), 304-310. Sakai, H., H. Ichihashi, H. Suganuma, and R. Tatsukawa. 1995. Heavy metal monitoring in sea turtles using eggs. Marine Pollution Bulletin 30, 347-353. Siewert, C. 2004. Rapid screening of so il properties using thermogravimetry. Soil Science Society of America Journal 68, 1656-1661. Storelli, M., Ceci, E., and G. Marcotrigiano. 1998. Distribution of heavy metal residues in some tissues of Caretta caretta (Linnaeus) specimen beached along the Adriatic Sea (Italy). Bulletin of Environmental Contamina tion and Toxicology 60 (4), 546-552. Tan, K. 1995. Soil sampling, preparation, and analysis. Marcel Dekker, Inc. New York. 432 pp. Valette-Silver, N. J., Riedel, G. F., Crecelius, E. A., Windom, H., Smith, R. G., and S. S. Dolvin. 1999. Elevated arsenic concentrations in bivalv es from the southeast coasts of the USA. Marine Environmental Research 48, 311-333. Wentworth, C.K. 1922. A scale of grade and class terms for clastic sediments. Journal of Geology 30, 377-392. Wibbels, T. R., Hanson, J., Balazs, G.H., Hillis-Starr, Z. and B. Phillips. 1998. Blood sampling techniques for hatchling chel oniid sea turtles. Herpetol ogical Review 29 (4), 218-220. Witherington, B. E., Kubilis, P., Brost, B. a nd A. Meylan. 2009. Decreasing annual nest counts in a globally important loggerh ead sea turtle population. Ecol ogical Applications 19 (1), 30-54. Yntema, C. L. and N. Mrosovsky. 1980. Sexual differentiation in hatchling loggerheads ( Caretta caretta ) incubated at different controlled temp eratures. Herpetologica 36 (1), 33-36. 166

PAGE 167

BIOGRAPHICAL SKETCH Mario was born and raised in Angola, where he developed an appr eciation and love for nature. Because of the civil war, his family was forced to leave and seek refuge in Portugal and later the United States. He attended Rutgers University in New Jersey where he got his Bachelor of Science degree in Zoology and Physiology. After Rutgers, Mari o took time off between studies and worked for an airline. This allowed him to travel the wo rld which further ingrained in him the beauty and natural wonders of nature. Four years later, he enrolled at the University of Central Florida where he got his Master of Science degree in Biology under the mentorship of Dr. L. Ehrhart. Ma rios thesis research described the anatomy of the adrenal glands in the Florida manatee. While at UCF Mario interned with the UCF Marine Turtle Research crew in Melbourne Beach, where he learned sea turtle ecology. Working at NASAs Kennedy Space Center he got interested in beach nourishment and how it impacts the incubation of sea turtle clut ches. This topic became the foundation for his dissertation research at the University of Florida.