EPA 904/9-74-002
ECOSYSTEMS ANALYSIS OF THE BIG CYPRESS
SWAMP AND ESTUARIES
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGION IV
ATLANTA, GEORGIA 30309
JUNE, 1973
-------
ECOSYSTEMS ANALYSIS
OF THE
BIG CYPRESS SWAMP
AND ESTUARIES
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGION IV
ATLANTA, GEORGIA 30309
JUNE, 1973
-------
ECOSYSTEMS ANALYSIS OF THE BIG CYPRESS SWAMP
AND ESTUARIES
Michael R. Carter
Lawrence A. Burns
Thomas R,, Cavinder
Kenneth R. Bugger
Paul L. Fore
Delbert B. Hicks
H. Lavon RevelIs
Thomas W. Schmidt
Roberta Farley, Secretary
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Region IV
Surveillance & Analysis Division
South Florida Ecological Study
June 1973
Cover Credit:
Russ Smiley, Artist
12000 R. W. 22nd Place
Miami, Florida 33167
-------
FOREWORD
This report and the study it represents are replete with examples of the
sensitive and interdependent relationships necessary for survival of south Florida's
natural ecosystem. It is hoped that readers of the report will include planners,
government officials, land developers and/or owners, concerned citizens, as well
as the ecologists and scientists interested in the technical findings presented.
Environmental protection is a complicated task and it will take an educated public
to understand it and make it work. Even though part of a whole -- the multiagency
South Florida Environmental Project -- EPA's study findings can still be used to
draw specific recommendations. The reader may be able to suggest more as he
reviews the more detailed chapters.
-------
TABLE OF CONTENTS
I. SUMMARY « 1-1
BACKGROUND 1-1
PHYSIOGRAPHY 1-2
CLIMATOLOGY 1-2
HYDROLOGY 1-2
SEDIMENTATION 1-3
DETRITAL TRANSPORT , 1-3
WATER CHEMISTRY 1-4
PRIMARY PRODUCTION 1-5
PLANT SOCIOLOGY 1-6
FISH I-6
WASTE INVENTORY 1-8
II. CONCLUSIONS AND RECOMMENDATIONS II-1
CONCLUSIONS , , II-1
RECOMMENDATIONS H-5
III. DISCUSSION III-l
UPLAND ECOSYSTEM III-l
ESTUARINE SYSTEMS III-3
ANALYTICAL RESULTS III-6
IV. INTRODUCTION IV-1
GENERAL IV-1
BACKGROUND IV-2
OBJECTIVE IV-4
SCOPE IV-5
AUTHORITY ..... IV-5
ACKNOWLEDGEMENTS IV-6
V. DESCRIPTION OF STUDY AREA V-l
SOUTH FLORIDA V-l
BIG CYPRESS SWAMP V-7
FAHKAHATCHEE STBAND V-10
ECOSYSTEMS MODELS V-12
VI. CLIMATOLOGY VI-1
INTRODUCTION VI-1
RAINFALL VI-1
Methods , , VI-1
Background , VI-1
Results VI-4
TEMPERATURE AND RELATIVE HUMIDITY VI-6
Methods VI-6
Background ..... VI-6
Results , VI-6
SOLAR ENERGY AND PAN EVAPORATION VI-9
Methods . VI-9
Results VI-10
ill
-------
EVAPOTRANSPIRATION VI-11
Theoretical Methods VI-13
Results VI-16
Potential Evapotranspiration VI-18
Actual Evapotranspiration VI-20
VII. HYDROLOGY VII-1
INTRODUCTION VII-1
UPLAND HYDROLOGY VII-1
Methods VII-1
Results VII-3
Surface Water . VII-3
Groundwater VII-7
ESTUARY HYDROLOGY VII-22
Groundwater Input .... VII-22
Estuary Bathymetry VII-25
Introduction . „ VII-25
Methods VII-26
Results VII-26
Circulation Patterns . . „ VII-26
Introduction ..„ VII-26
Methods VII-29
Results VII-30
Salinity and Temperature Variations VII-35
Introduction VII-35
Methods . VII-37
Results VII-37
Tidal Fluctuations VII-52
VIII. BAY BOTTOM CHARACTERISTICS AND SEDIMENT DEPOSITION VIII-1
INTRODUCTION VIII-1
METHODS VIII-1
Bottom Characteristics VIII-1
Sedimentation Rates „ VIII-1
Coring „ VIII-3
RESULTS VIII-3
Bottom Characteristics VIII-3
Sedimentation Rates VIII-6
Coring . . VIII-11
DISCUSSION VIII-11
IX. DETRITAL TRANSPORT . IX-1
INTRODUCTION IX-1
METHODS IX-1
RESULTS AND DISCUSSION IX-3
X. WATER CHEMISTRY X-l
INTRODUCTION X-l
METHODS „ X-l
RESULTS X-4
Physical Characteristics X-4
Chemical Quality „ X-6
Nutrient Transport ... X-12
IV
-------
XI. METALS AND PESTICIDES XI-1
INTRODUCTION . . . XI-1
METHODS XI-1
Metals „ XI-3
Pesticides XI-3
RESULTS XI-3
Metals . . . . XI-3
Pesticides „ XI-7
XII. PLANT COMMUNITY BIOMASS AND METABOLISM XII-1
MARINE GRASSES AND ALGAE . „ „ XII-1
Introduction . , XII-1
Methods . . . . „ XII-1
Results o XII-5
Benthic Vegetation and Standing Crop Bioraass . XII-5
Light Extinction Studies XII-10
Gross Primary Productivity Studies XII-12
Discussion ........... XII-15
SALT MARSH AND MANGROVE METABOLISM XII-22
Introduction 0 . XII-22
Study Area and Site Locations XII-23
Methods . „ XII-25
Results XII-26
Discussion XII-35
UPLAND COMMUNITY BIOMASS AND METABOLISM XII-38
Biomass XII-38
Introduction XII-38
Methods XII-38
Results and Discussion ... XII-41
Litterfall in Woodlands XII-44
Introduction XII-44
Methods XII-44
Results „ XII-44
Discussion XII-47
Decomposition of Litter „ . XII-47
Introduction „ . XII-47
Methods XII-48
Results XII-48
Discussion „ XII-50
TERRESTRIAL COMMUNITY METABOLISM XII-51
Introduction .......... XII-51
Methods . . . XII-51
Results XII-52
XIII. PLANT COMMUNITY STRUCTURE XIII-1
COMMUNITY SURVEY AND CLASSIFICATION XIII-1
Habitat Classification . . . XIII-1
Plant Communities of Southwest Florida XIII-3
Plants Listed in the Community Description XIII-3
Plant Composition „ XIII-4
Cypress Strand XIII-4
Prairie XIII-5
The Pine and Palm Associations XIII-5
Disturbed Swamplands and Prairies ....... XIII-6
Freshwater Ditches and Canals ... XIII-8
v
-------
Aquatic Communities of the Cypress Strand
Lake and Ponds XIII-8
Salt Marsh and Mangrove Swamps XIII-10
Aquatic Coastal Communities XIII-12
XIV. MACROBENTHOS XIV-1
INTRODUCTION XIV-1
METHODS XIV-1
RESULTS AND DISCUSSION ... ..... XIV-1
XV. FISHES OF FAHKAHATCHEE STRAND AND TEN THOUSAND ISLANDS . . . XV-1
INTRODUCTION XV-1
STANDING CROP OF FISHES , . XV-1
Habitats, Sampling Procedure and Gear XV-1
Results . . . XV-4
Temporary Freshwater Habitats . XV-4
Canal Habitats . XV-7
Strand Lake Habitat . . XV-10
Bay Habitats . . „ XV-12
Tidal Stream Habitat XV-14
Discussion XV-15
RELATIVE ABUNDANCE OF ESTUARINE FISHES XV-16
Habitats, Sampling Gear, and Procedures XV-16
Results XV-19
Relative Abundance by Number . „ XV-19
Relative Abundance by Biomass XV-23
Distribution by Rank XV-25
Relative Abundance Between Bays XV-26
Distribution by Habitats XV-28
Distribution in Time . XV-31
Discussion XV-32
XVI. BIOLOGY OF JUVENILE AND ADULT SNOOK, CENTROPOMUS UNDECIMALIS,
IN THE TEN THOUSAND ISLANDS XVI-1
INTRODUCTION XVI-1
MATERIALS AND METHODS XVI-1
HABITATS OF JUVENILES XVI-2
HYDROGRAPHIC DATA ON JUVENILES XVI-3
PESTICIDES . . . . . ...'.»... XVI-3
ASSOCIATED ORGANISMS XVI-6
LENGTH CONVERSION FACTORS XVI-7
LENGTH-WEIGHT RELATIONSHIPS OF JUVENILES XVI-8
OCCURRENCE AND GROWTH XVI-9
FOOD HABITS XVI-10
FOOD HABITS RELATED TO SIZE XVI-15
FEEDING HABITS XVI-17
XVII. WASTE INVENTORY .... XVII-1
INTRODUCTION . . XVII-1
METHODS ...... . XVII-1
RESULTS XVII-2
XVIII. REFERENCES XVIII-1
vi
-------
Page
XIX. APPENDICES XIX-1
A. METHODS OF ECOSYSTEM ANALYSIS XIX-1
B. ECOSYSTEMS MODELS XIX-9
C. COMPUTATIONS OF SEDIMENTATION RATES FROM TRAP DATA . . . XIX-22
D. INHABITANTS FOUND IN SEDIMENTATION TRAPS XIX-23
E. SAMPLING STATION SITE DESCRIPTIONS AND LOCATIONS XIX-24
F. SPECIFIC TESTS USED FOR EACH WATER CHEMISTRY ANALYSIS . . . XIX-26
G. STANDING CROP BIOMASS OF BENTHIC VEGETATIOK IN FAHKA
UNION AND FAHKAHATCHEE BAYS, JANUARY 1972 XIX-28
H. STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
SAND-MUD BOTTOM OF FAHKAHATCHEE AND FAHKA UNION BAYS,
JULY, 1972 XIX-29
I. STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
MUD-SAND BOTTOM AREA IN FAHKAHATCHEE AND FAHKA UNION BAYS,
JULY, 1972 XIX-30
J. STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
SHELL BOTTOM IN FAHKAHATCHEE AND FAHKA UNION BAYS,
JULY, 1972 XIX-31
K. STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
MUD-SAND BOTTOM IN FAHKAHATCHEE AND FAHKA UNION BAYS,
SEPTEMBER, 1972 XIX-32
L. STANDING CROP^BIOMASS OF BENTHIC VEGETATION INHABITING
SAND-MUD BOTTOM IN FAHKAHATCHEE AND FAHKA UNION BAYS,
SEPTEMBER, 1972 XIX-33
M. STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
SHELL BOTTOM IN FAHKAHATCHEE AND FAHKA UNION BAYS,
SEPTEMBER, 1972 .' " XIX-34
N. SUMMARY OF MEAN LIGHT EXTINCTION COEFFICIENT WITH
RESPECT TO DATE, TIME OF DAY, AND WATER DEPTH, FAHKA
UNION BAY WITH MUD-SAND BOTTOM, 1972-73 XIX-35
0. SUMMARY OF MEAN LIGHT EXTINCTION COEFFICIENT WITH
RESPECT TO DATE, TIME OF DAY, AND WATER DEPTH,
FAHKAHATCHEE BAY WITH SAND-MUn w?TTOM, 1972-73 XIX-36
P. SUMMARY OF MEAN LIGHT EXTINCTION OEFFICIENT WITH
RESPECT TO DATE, TIME OF DAY, AND WATER DEPTH,
FAHKAHATCHEE BAY WITH MUD-SAND BOTTOM, 1972-73 XIX-37
vii
-------
Page
Q. BENTHIC GROSS PRIMARY PRODUCTIVITY (GPP) IN FAHKA
UNION AND FAHKAHATCHEE BAYS EMPLOYING LIGHT AND DARK
CHAMBERS, SPRING AND FALL, 1972 XIX-37
R. METHODS, METABOLIC STUDIES XIX-41
S. PHYSICAL DIMENSIONS AND RELATED BIOMASS (DRY WT) OF
TEN SPECIES INCLUDING SEEDLINGS XIX-51
T. REGRESSION EQUATIONS FOR WOODY BIOMASS BASED ON THE
MODEL Y = abx, WHERE Y IS TOTAL ABOVE GROUND WOODY
BIOMASS IN GRAMS DRY WEIGHT (CONSTANT WEIGHT AT
105° C) XIX-54
U, REGRESSION EQUATIONS FOR LEAF BIOMASS BASED ON THE
MODEL Y = ab , WHERE Y IS TOTAL LEAF BIOMASS IN GRAMS
DRY WEIGHT (CONSTANT WEIGHT AT 105° C) XIX-55
V. REGRESSION EQUATIONS FOR SABAL USING TOTAL HEIGHT .... XIX-56
W. DECOMPOSITION OF LITTER CONFINED TO FINE MESH BAGS
PLACED ON FOREST FLOOR AND SUBJECTED TO INUNDATION
DURING THE HYDROPERIOD, CENTRAL FAHKAHATCHEE STRAND,
1972 XIX-57
X. DECOMPOSITION OF LITTER CONFINED TO COARSE MESH BAGS
PLACED ON FOREST FLOOR AND SUBJECTED TO INUNDATION
DURING HYDROPERIOD, CENTRAL FAHKAHATCHEE STRAND, 1972 . . XIX-58
Y. DECOMPOSITION OF LITTER CONFINED TO FINE MESH BAGS
PLACED ON DEBRIS PILES AND NOT SUBJECTED TO INUNDATION
DURING HYDROPERIOD, CENTRAL FAHKAHATCHEE STRAND, 1972 . . XIX-59
Z. DECOMPOSITION OF LITTER CONFINED TO COARSE MESH BAGS
PLACED ON DEBRIS PILES AND NOT SUBJECTED TO INUNDATION
DURING HYDROPERIOD, CENTRAL FAHKAHATCHEE STRAND, 1972 . . XIX-60
AA. TREES AND SHRUBS OF THE CYPRESS STRAND XIX-61
BB. VINES OF THE CYPRESS STRAND XIX-63
CC. TERRESTRIAL HERBS OF THE CYPRESS STRAND (EXCLUDING VINES). XIX-65
DD. EPIPHYTIC HERBS OF THE CYPRESS STRAND XIX-67
EE. AQUATIC HERBS OF THE CYPRESS STRAND OR CYPRESS SLOUGH . . XIX-69
FF. PRAIRIE PLANTS XIX-71
GG. TREES AND SHRUBS OF THE SANDY MARL PINE AND PINE
ASSOCIATIONS XIX-76
HH. HERBACEOUS PLANTS OF THE WET PINE AND PALM ASSOCIATIONS . XIX-78
viii
-------
Page
II. AQUATIC PLANTS OF THE DITCHES AND CANALS XIX-81
JJ. ABUNDANCE OF MACROBENTHOS XIX-83
KK. LIST OF SCIENTIFIC AND COMMON NAMES OF FISHES FROM
FAHKAHATCHEE STRAND AND TEN THOUSAND ISLANDS, FLORIDA,
1972 XIX-86
LL. SUMMARY OF SELECTED AQUATIC BIOTA ASSAYED FOR ASH-FREE
WEIGHT AND ORGANIC NITROGEN, FAHKAHATCHEE STRAND AND
ADJACENT CANALS, FLORIDA, 1972 XIX-90
MM. SPATIAL DISTRIBUTION OF FISHES COLLECTED IN TEN
THOUSAND ISLANDS AND FAHKAHATCHEE STRAND, FLORIDA,
1972, BY HABITAT XIX-91
NN. TEMPORAL DISTRIBUTION OF FISHES COLLECTED IN TEN
THOUSAND ISLANDS AND FAHKAHATCHEE STRAND, FLORIDA,
1972 XIX-95
00. WASTE SOURCE INVENTORIES XIX-98
PP. DOMESTIC WASTE WATER INVENTORY XIX-99
QQ. INDUSTRIAL WASTE WATER SOURCES XIX-115
ix
-------
LIST OF TABLES
Page
III-l. WATER BUDGET FOR UPLAND STRAND, 1972 III-ll
III-2. CYPRESS STRAND NET PRODUCTION, 1972, FOR DRAINED AND
UNDRAINED SITES. (ALL VALUES GM DRY WT/M2/YEAR.). .... 111-14
III-3. LEAF LITTER TURNOVER RATES. (BASED ON GM DRY WT OF
MATERIAL.) 111-16
III-4. SUMMARY OF BIOMASS ACCUMULATION IN WET PRAIRIE ECOSYSTEMS
BASED ON PEAK-TO-PEAK HARVEST VALUES (GM ASH FREE DRY
WT/M2) 111-18
III-5. SUMMARY OF WET PRAIRIE CARBON AND WATER METABOLISM .... Ill-19
III-6. YEARLY RESOURCE INPUTS TO FAHKAHATCHEE BAY (METRIC
TONS/YEAR) 111-25
VI-1. ANNUAL AND MONTHLY MEAN RAINFALL IN THE VICINITY OF
FAHKAHATCHEE STRAND BASED ON ANALYSIS OF HISTORICAL
RAINFALL BY T. M. THOMAS ., VI-3
VI-2. AVERAGE MONTHLY TEMPERATURE (°C) FOR EVERGLADES CITY
AND NAPLES, FLORIDA VI-7
VI-3. AVERAGE MONTHLY TEMPERATURE AT STATIONS 2 AND 6
FOR 1972 VI-9
VI-4. AVERAGE MONTHLY RELATIVE HUMIDITY AT STATIONS
2 AND 6 FOR 1972 VI-10
VI-5. COMPUTED EVAPOTRANSPIRATION COMPARED TO PAN EVAPORATION
IN FAHKAHATCHEE STRAND DURING 1972 (CM H2O) ....... VI-15
VI-6. COMPARISON OF PRAIRIE GRASS EVAPOTRANSPIRATION
MEASURED BY HUMIDITY SENSORS TO PAN EVAPORATION
IN CM/DAY DURING AUGUST, 1972 VI-18
VI-7. CLIMATOLOGICAL DATA FOR 1972, FAHKAHATCHEE STRAND AREA . . VI-20
VI-8. EVAPOTRANSPIRATION CALCULATED FROM TEST WELL DATA
(MM H20) VI-22
VII-1. GROUND WATER CONTRIBUTION AT A SELECT SECTION OF THE
GAG CANAL SYSTEM VII-22
VII-2. GROUND WATER FLOWS IN M3 X 103 PER MONTH VTI-25
VII-3. RELATIVE DISCHARGE PER STATION VII-35
VIII-1. SUBSTRATE CHARACTERISTICS IN FAHKA UNION AND
FAHKAHATCHEE BAYS VIII-5
x
-------
VIII-2.
IX- 1.
IX-2.
IX- 3.
IX-4.
IX- 5.
IX- 6.
X-l.
X-2
X-3.
X-4
X-5.
X-6.
X-7.
X-8.
X-9.
X-10,
X-ll.
XI- 1.
SUMMARY OF SUBSTRATE CHARACTERISTICS IN FAHKA UNION AND
- FAHKAHATCHEE BAYS
SUMMARY OF DETRITAL CONCENTRATIONS SAMPLED DURING EBB
AND FLOOD TIDES IN FAHKA UNION AND FAHKAHATCHEE BAYS
DURING JULY AND OCTOBER, 1972
FAHKAHATCHEE BAY MASS TRANSPORT OF THREE CHEMICAL
PARAMETERS FOUND IN ONE TIDAL CYCLE IN OCTOBER, 1972 . . „
FAHKAHATCHEE BAY CARBON MASS BALANCE FOR ONE TIDAL
CYCLE IN OCTOBER 1972
SUMMARY OF PHYTOPIANKTON AND ZOOPLANKTON CONCENTRATIONS
IN MG/1 (WET WT BASIS) FOUND IN DETRITAL FRACTIONS
DURING MID -JULY IN FAHKAHATCHEE AND FAHKA UNION BAYS . . .
ZOOPLANKTON CONCENTRATIONS FOUND IN FAHKA UNION BAY AS
SAMPLED WITH A 500 MICRON MESH PLANKTON NET
PHYTOPLANKTON COLLECTED ON JUNE 15, 1972, WITH A 500
MICRON MESH PLANKTON 'NET
PHYSICAL AND CHEMICAL ANALYSES
APPARENT COLOR VALUES FOR EACH STATION PER MONTH
RANGE AND MEDIAN VALUES FOR CONDUCTIVITY IN MMHO/CM . . .
RANGE AND MEAN VALUES FOR SULFATES IN MG/1 .
NITRATE AND NITRITE NITROGEN CONCENTRATIONS FOR THE MONTH
OF AUGUST IN MG/1
THE PERCENTAGE OF TOTAL KJELDAHL NITROGEN THAT IS TOTAL
SOLUBLE KJELDAHL NITROGEN
THE PERCENTAGE OF TOTAL PHOSPHORUS THAT IS TOTAL
SOLUBLE PHOSPHORUS ........
THE AVERAGE TRANSPORT (KG/DAY) OF TOTAL KJELDAHL
NITROGEN, TOTAL PHOSPHORUS, AND TOTAL ORGANIC CARBON
INTO THE ESTUARY FROM THREE FRESH WATER AREAS FOR 1972 . .
THE AVERAGE TRANSPORT OF TOTAL PHOSPHORUS (KG/DAY) ....
THE AVERAGE TRANSPORT OF TOTAL KJELDAHL NITROGEN
(KG/DAY)
THE AVERAGE TRANSPORT OF TOTAL ORGANIC CARBON (KG/DAY) . .
LOCATION OF STATIONS FOR PESTICIDE AND METAL ANALYSIS . .
Page
VIII-6
IX-4
IX- 7
IX- 8
IX- 9
IX- 10
IX- 10
X-4
X-5
X-5
X-9
X-ll
X-12
X-12
X-14
X-18
X-19
X-20
XI- 1
XI
-------
Page
XI-2. PESTICIDE ANALYSES AND MINIMUM DETECTION LIMITS XI-4
XI-3. METAL ANALYSIS FOR WATER SAMPLES IN (jG/L XI-5
XI-4. METAL ANALYSIS FOR SEDIMENT SAMPLES IN (OG/G XI-6
XI-5. MEDIAN VALUES FOR METAL ANALYSIS OF THE SEDIMENT IN
MG/G XI-8
XI-6. RESULTS ((J-G/KG) OF FOUR PESTICIDES DETECTED IN THE
SEDIMENT SAMPLES „ XI-8
XI-7. PESTICIDE ANALYSES ON BIOLOGICAL SAMPLES FOR JANUARY . . . XI-10
XI-8. PESTICIDE ANALYSES ON BIOLOGICAL SAMPLES FOR APRIL .... XI-11
XI-9. PESTICIDE ANALYSES ON BIOLOGICAL SAMPLES FOR JULY .... XI-12
XI-10. PESTICIDE ANALYSES ON BIOLOGICAL SAMPLES FOR NOVEMBER . . XI-13
XI-11. PESTICIDE ANALYSES ON SPECIAL BIOLOGICAL SAMPLES XI-13
XII-1. STANDING CROP BIOMASS OF BENTHIC VEGETATION IN FAHKA
UNION AND FAHKAHATCHEE BAYS, JANUARY, 1972 XII-9
XII-2. GROSS PRIMARY PRODUCTIVITY RATES FOR FAHKA UNION AND
FAHKAHATCHEE BAY WITH RESPECT TO BOTTOM TYPES AND
RATES OF VISIBLE LIGHT TRANSMISSION, SPRING, 1972 .... XII-14
XII-3. GROSS PRIMARY PRODUCTIVITY RATES FOR FAHKA UNION AND
FAHKAHATCHEE BAY WITH RESPECT TO BOTTOM TYPES AND
RATES OF VISIBLE LIGHT TRANSMISSION, FALL, 1972 XII-15
XII-4. CALCULATED MONTHLY MEANS OF DEPTH DURING DAYLIGHT
HOURS, VISIBLE LIGHT TRANSMISSION AT BOTTOMS, AND
LIGHT EXTINCTION COEFFICIENT, FAHKA UNION AND
FAHKAHATCHEE BAYS, 1972 XII-18
XII-5. AVERAGE DAILY GROSS PRIMARY PRODUCTIVITY OF BENTHIC
PLANT COMMUNITIES ESTABLISHED ON THE MUD AND SAND
BOTTOMS OF FAHKA UNION AND FAHKAHATCHEE BAYS, 1972 .... XII-19
XII-6. AVERAGE DAILY RATES OF PRIMARY PRODUCTION AND
RESPIRATION FOR THE BENTHIC PLANT COMMUNITIES
ASSOCIATED WITH FAHKA UNION AND FAHKAHATCHEE BAYS
DURING 1972 XII-21
xii
-------
XII-7. RESULTS OF PLANT METABOLIC STUDIES TO DETERMINE PRIMARY
PRODUCTION OF THE SALT MARSH PLANT SPECIES ASSOCIATED
WITH THE UPPER FAHKA UNION RIVER BASIN, STATION D,
1972 XII-27
XII-8. MEAN RATES OF PRODUCTIVITY AND RESPIRATION FOR THREE
SPECIES OF PLANTS CHARACTERISTIC OF THE SALT GRASS
MARSHES ASSOCIATED WITH THE UPPER REACHES OF THE
FAHKA UNION RIVER, STATION D, 1972 XII-29
XII-9. RESULTS FROM PLANT METABOLIC STUDIES CONDUCTED TO
DETERMINE PRIMARY PRODUCTION OF THE MANGROVES
ASSOCIATED WITH THE UPPER FAHKA UNION RIVER, STATION D,
1972 XII-29
XII-10. DIURNAL METABOLIC RATES FOR VARIOUS COMPONENTS OF A
MANGROVE FOREST COMMUNITY ASSOCIATED WITH THE UPPER
REACHES OF THE FAHKA UNION RIVER, STATION D, 1972 .... XII-30
XII-11. RESULTS FROM PLANT METABOLIC STUDIES CONDUCTED TO
DETERMINE PRIMARY PRODUCTION OF SALT MARSH PLANTS
AND MANGROVES ASSOCIATED WITH THE MID-REACH OF FAHKA
UNION RIVER, STATION E, 1972 XII-31
XII-12. MEAN RATES OF PRODUCTIVITY AND RESPIRATION FOR PLANT
SPECIES CHARACTERISTIC OF THE SALTGRASS MARSH ASSOCIATED
WITH THE MID-REACHES OF THE FAHKA UNION RIVER, STATION E,
1972 XII-32
XII-13. DIURNAL PRODUCTIVITY RATES OF SUN LEAVES FOR RED AND
WHITE MANGROVE INHABITING THE STREAM SIDE OF THE MID-
REACH OF THE FAHKA UNION RIVER, STATION E, 1972 . . . . . XII-33
XII-14. RESULTS FROM PLANT METABOLIC STUDIES CONDUCTED TO
DETERMINE PRIMARY PRODUCTION OF MANGROVES ASSOCIATED
WITH THE LOWER FAHKA UNION RIVER BASIN, STATION F,
1972 XII-34
-<
XII-15, DIURNAL METABOLIC RATES FOR VARIOUS COMPONENTS OF THE
MANGROVE FOREST COMMUNITY ASSOCIATED WITH THE LOWER
REACH OF THE FAHKA UNION RIVER, STATION F, 1972 XII-34
XII-16. SUMMARY OF RESULTS FROM PLANT METABOLIC STUDIES CONDUCTED
TO DETERMINE PRIMARY PRODUCTIVITY OF MANGROVES ASSOCIATED
WITH A SMALL TIDAL STREAM OF FAHKAHATCHEE BAY, STATION G,
1972 XII-35
XII-17. DIURNAL METABOLIC RATES FOR VARIOUS COMPONENTS OF THE
MAINLAND MANGROVE FOREST COMMUNITY ASSOCIATED WITH A
SMALL TIDAL STREAM IN FAHKAHATCHEE BAY, STATION G,
1972 . XII-36
XII-18. MANGROVE METABOLISM WITH RESPECT TO CHANGES IN
FRESHWATER CONCENTRATION, 1972 ..... XII-37
xiii
-------
Page
XII-19. RESPONSE OF MANGROVE COMMUNITY METABOLISM TO CHANGING
CONCENTRATIONS OF CHLORIDES (FRESH WATER), 1973 XII-40
XII-20. COMMUNITY STAND TABLE FOR WOODY PLANT BIOMASS (WOOD AND
LEAF MASSES) IN AN UNDRAINED AREA OF A CYPRESS STRAND . . XII-42
XII-21. AVERAGE STANDING CROP BIOMASS WITH 95 PERCENT CONFIDENCE
LIMITS FOR WET PRAIRIES NEAR (AREA A) AND REMOTE (AREA B)
TO A DRAINAGE CANAL, 1972 XII-43
XII-22. LITTERFALL IN THE CENTRAL STRAND-UNDRAINED ........ XII-46
XII-23. LITTERFALL IN THE CYPRESS STRAND-DRAINED XII-46
XII-24. LITTERFALL IN THE CYPRESS STRAND-UNDRAINED XII-47
XII-25. TURNOVER RATES OF LITTER COMPONENTS, FROM LEAST SQUARES
REGRESSION OF DATA FROM DECOMPOSITION MATERIALS XII-50
XII-26. RESULTS OF STUDIES TO MEASURE WET PRAIRIE COMMUNITY
PRODUCTIVITY, RESPIRATION,, AND EVAPOTRANSPIRATION IN
AREAS NEAR (AREA A) AND REMOTE (AREA B) TO DRAINAGE
CANALS, 1972 XII-52
XII-27. MEAN RATES OF PRODUCTIVITY AND EVAPOTRANSPIRATION FOR
WET PRAIRIE COMMUNITIES ADJACENT (A) AND REMOTE (B) TO
MAJOR DRAINAGE CANALS. PRODUCTIVITY RATES NORMALIZED
FOR EQUAL CONDITIONS OF TOTAL SOLAR RADIATION, 1972 . . . XII-54
XIII-1. COMMUNITIES OF THE BIG CYPRESS AND THE ASSOCIATED
SOIL TYPES (FROM LEIGHTY, ET AL. , 1954) XIII-2
XIII-2. FRESHWATER CANALS AND DITCHES IN THE STUDY AREA XIII-9
XIV-1. BIOMASS OF MACROBENTHOS XIV-3
XIV-2. DISTRIBUTION AND RELATIVE ABUNDANCE OF DECAPOD CRUSTACEANS
COLLECTED IN TEN THOUSAND ISLANDS, FLORIDA, 1972 XIV-6,7
XIV-3. TEMPORAL DISTRIBUTION OF DECAPOD CRUSTACEANS COLLECTED
IN TEN THOUSAND ISLANDS, FLORIDA, 1972 XIV-8
XV-1. STANDING CROPS OF FLORA AND FAUNA OBTAINED MONTHLY FROM
TWO TEMPORARY WATERWAYS IN FAHKAHATCHEE STRAND, FLORIDA,
1972 ' XV-5,6
XV-2. SEASONAL CHANGES IN THE DENSITY OF AQUATIC ANIMALS IN A
TEMPORARY WATERWAY ADJACENT TO JANES SCENIC DRIVE,
FAHKAHATCHEE STRAND, 1972 XV-7
XV-3. STANDING CROP OF FISHES IN THE MOST EASTERN GAG CANAL,
FAHKAHATCHEE STRAND, FLORIDA, FEBRUARY, 1972. SURFACE
AREA OF SAMPLING SITE WAS 1,950 M2 XV-8
xiv
-------
XV-4. STANDING CROP OF FISHES IN THE TAMIAMI CANAL AT BIG
CYPRESS BEND, FLORIDA, MARCH, 1972. SURFACE AREA OF
THE SAMPLING SITE WAS 900 M2 XV-9
XV-5. STANDING CROPS OF FISHES AND AMPHIBIANS FROM A NATURAL
LAKE IN FAHKAHATCHEE STRAND, FLORIDA, APRIL, 1973.
SURFACE AREA OF THE SAMPLING SITE WAS 4,069 M2 XV-11
XV-6. STANDING CROPS OF FISHES FROM THREE SITES IN FAHKA
UNION BAY, FLORIDA, SEPTEMBER, 1972 XV-13
XV-7. STANDING CROPS OF FISHES FROM THREE SITES IN
FAHKAHATCHEE BAY, FLORIDA, SEPTEMBER, 1972 XV-13
XV-8. STANDING CROP OF FISHES FROM A TIDAL STREAM IN NORTH
FAHKAHATCHEE BAY, OCTOBER, 1972. SURFACE AREA OF THE
-SAMPLING SITE WAS 734 M2 ON A LOW EBB TIDE XV-15
XV-9. DISTRIBUTION OF THE TOTAL CATCH AT THE REGULAR STATIONS
BY SPECIES, NUMBER, AND BIOMASS FROM TEN THOUSAND ISLANDS,
1972 XV-20
XV-10. MOST ABUNDANT SPECIES OF FISHES AS PERCENT OF THE MONTHLY
CATCH FROM REGULAR STATIONS IN THE TEN THOUSAND ISLANDS,
1972, EXCLUDING ANCHOA SPP XV-21
XV-11. MOST ABUNDANT FAMILIES OF FISHES BY NUMBER AS PERCENT OF
MONTHLY CATCH FROM THE REGULAR STATIONS IN TEN THOUSAND
ISLANDS, 1972, EXCLUDING ENGRAULIDAE XV-21
XV-12. MOST ABUNDANT SPECIES OF FISHES BY BIOMASS AS PERCENT
OF THE MONTHLY CATCH FROM REGULAR STATIONS IN THE TEN
THOUSAND ISLANDS, 1972 XV-24
XV-13. MOST ABUNDANT FAMILIES OF FISHES BY BIOMASS AS PERCENT
OF MONTHLY CATCH FROM REGULAR STATIONS IN TEN THOUSAND
ISLANDS, 1972 XV-25
XV-14. PERCENT SPECIES AND BIOMASS COMPOSITION OF THE 40 MOST
ABUNDANT FISHES AT THE REGULAR MONTHLY STATIONS IN THE
TEN THOUSAND ISLANDS, 1972 XV-27
XV-15. COMPARISON OF YEARLY CATCHES OF FISHES TAKEN BY SURFACE
AND OTTER TRAWLS BETWEEN FAHKA UNION AND FAHKAHATCHEE
BAYS, FLORIDA, 1972 XV-28
XV-16. A SUMMARY OF THE FISH SPECIES FROM THE COASTAL WATER
HABITATS IN THE TEN THOUSAND ISLANDS, FLORIDA, 1972 . . . XV-29
XVI-1. DESCRIPTION OF HABITATS WHERE JUVENILE SNOOK, CENTROPOMUS
UNDECIMALIS, WERE COLLECTED IN TEN THOUSAND ISLANDS,
FLORIDA, 1971-72 . XVI-2
xv
-------
Page
XVI-2. TEMPORAL, SPATIAL, AND HYDRO-LOGICAL DATA ON ALL COLLECTIONS
OF JUVENILE SNOOK, CENTROPOMUS UNPEGIMALIS, FROM THE TEN
THOUSAND ISLANDS, FLORIDA, 1971-72 XVI-4
XVI-3. PHYSICAL AND CHEMICAL CHARACTERISTICS OF WATER COLLECTED
MONTHLY FROM THREE TIDAL STREAMS INHABITED BY JUVENILE
SNOOK, JUNE TO DECEMBER, 1972 XVI-5
XVI-4. ANALYSIS OF PESTICIDES IN SNOOK, CENTROPOMUS UNDECIMALIS.
IN TEN THOUSAND ISLANDS, FLORIDA, 1972 XVI-5
XVI-5. ASSOCIATED MACROCRUSTACEANS AND FISHES COLLECTED WITH
JUVENILE SNOOK DURING 15 RANDOMLY SELECTED SEINE HAULS
AT FOUR STATIONS IN TEN THOUSAND ISLANDS, FLORIDA, JUNE
THROUGH DECEMBER, 1972 ...... XVI-6,7
XVI-6. STOMACH CONTENTS OF 183 JUVENILE SNOOK, 142 OF WHICH
CONTAINED FOOD, FROM TEN THOUSAND ISLANDS, FLORIDA,
JUNE TO DECEMBER, 1972 XVI-11,12
XVI-7. STOMACH CONTENTS OF 271 ADULT SNOOK, 127 OF WHICH
CONTAINED FOOD, FROM TEN THOUSAND ISLANDS, FLORIDA,
MAY TO AUGUST, 1972 XVI-13,14
XVII-1. WASTE WATER TREATMENT FACILITIES IN SOUTHWEST FLORIDA
BY COUNTY AS OF JANUARY 1, 1973 XVII-3
XVII-2. RECEIVING BASINS XVII-4
xvi
-------
LIST OF FIGURES
Page
III-l. UPLAND ECOSYSTEM MODEL III-l
III-2. ESTUARINE ECOSYSTEM MODEL III-3
III-3. CLIMATIC DIAGRAMS III-7
3a - LONG TERM AVERAGE
3b - FOR 1972
III-4. MODEL OF EVAPORATION PROCESS III-8
III-5. MEAN DAILY INSOLATION, 1972 III-8
III-6. WINDSPEED AND SATURATION DEFICIT III-8
III-7. PAN EVAPORATION AND POTENTIAL EVAPORATION, 1972 III-9
III-8. HYDROLOGIC MODEL FOR UPLAND STRAND Ill-10
III-9. TEST WELL EVAPOTRANSPIRATION . „ Ill-10
III-10. RAINFALL AND ACTUAL EVAPOTRANSPIRATION „ Ill-11
III-ll. WATER TABLE DEPTHS, TEST WELLS 111-12
III-12. SWAMP FOREST DYNAMICS Ill-13
III-13. PRODUCTIVITY AND ACTUAL EVAPOTRANSPIRATION, WORLD-WIDE . . Ill-13
III-14. MODEL OF LITTER REMINERALIZATION SEQUENCES Ill-15
111-15. NITROGEN BUILD-UP IN DECAYING LITTER 111-16
111-16. MODEL OF WET PRAIRIE PRODUCTION Ill-17
III-17. PRODUCTIVITY AND ACTUAL EVAPOTRANSPIRATION, WET
PRAIRIE 111-20
111-18. WATER TABLE DEPTH AND EVAPOTRANSPIRATION 111-21
111-19. MONTHLY MEAN WATER TABLE DEPTH 111-21
111-20. SIMULATION OF WET PRAIRIE (GRASS) NPP 111-22
111-21. NPP AS A FUNCTION OF WATER TABLE DEPTH 111-22
111-22. MAP OF ESTUARY STUDY ZONE 111-23
111-23. RESPONSE OF METABOLIC BUDGET (SUN LEAVES TO cl" DYNAMICS). 111-26
111-24. FRESH WATER DYNAMICS 111-28
111-25. SPATIAL DISTRIBUTION OF TIDAL FLUSHING AND SWEET WATER . . 111-28
xvii
-------
Page
111-26. SPATIAL RESPONSE OF METABOLIC BUDGET 111-29
V-l. GROUND WATER EVAPORATION, EXPRESSED AS A PERCENTAGE
OF PAN EVAPORATION, AS A FUNCTION OF DEPTH TO THE
WATER TABLE (FROM TODD, 1959; AFTER WHITE, 1932) V-4
V-2. GENERAL LOCATION MAP OF THE BIG CYPRESS SWAMP AND
STUDY AREA V-6
V-3. MAJOR STRANDS OF THE BIG CYPRESS SWAMP V-7
V-4. THE THREE DISTINCT SUB-AREAS OF THE BIG CYPRESS
SWAMP SHOWING MAJOR DRAINAGE PATTERNS V-9
V-5. MAP OF STUDY AREA AND DRAINAGE WORKS IMMEDIATELY
TO THE WEST V-ll
V-6. PROPORTIONAL REPRESENTATION OF THE DISTRIBUTION OF
PLANT COMMUNITY TYPES IN THE FAHKAHATCHEE STRAND STUDY
AREA (6a) AND THAT PORTION OF THE BIG CYPRESS SWAMP
AND ITS ASSOCIATED ESTUARY LYING WITHIN COLLIER
COUNTY, FLORIDA (6b) V-12
VI-1. MAP OF STUDY AREA SHOWING CLIMATOLOGY STATION
LOCATIONS VI-2
VI-2. ANNUAL RAINFALL AT EVERGLADES CITY, FLORIDA,
1931-1972 VI-3
VI-3. ANNUAL RAINFALL AT NAPLES, FLORIDA, 1941-1972 VI-4
VI-4. SUMMARY OF ANNUAL AND MONTHLY RAINFALL AT EVERGLADES
CITY, FLORIDA, 1931-1972 VI-4
VI-5. SUMMARY OF ANNUAL AND MONTHLY RAINFALL AT NAPLES,
FLORIDA, 1940-1972 VI-5
VI-6. DEPARTURE FROM AVERAGE MONTHLY RAINFALL AT EVERGLADES
CITY AND NAPLES, FLORIDA, DURING 1972 VI-5
VI-7. RAINFALL DURING 1972 AT FOUR SITES IN FAHKAHATCHEE
STRAND VI-5
VI-8. RAINFALL DURING 1972 AT THREE SITES IN FAHKAHATCHEE
STRAND _ VI-5
VI-9. YEARLY TEMPERATURE RANGE AT EVERGLADES CITY, FLORIDA,
1931-1972 VI-7
VI-10. WEEKLY TEMPERATURE RANGE AT STATION 2 DURING 1972 .... VI-8
VI-11. WEEKLY TEMPERATURE RANGE AT STATION 6 DURING 1972 .... VI-8
xviii
-------
Page
VI-12. WEEKLY RELATIVE HUMIDITY RANGE AT STATION 2 DURING 1972 . VI-9
VI-13. WEEKLY RELATIVE HUMIDITY RANGE AT STATION 6 DURING 1972 . VI-10
VI-14. WEEKLY SOLAR ENERGY RANGE AT STATION 6 DURING 1972 .... VI-11
VI-15. MONTHLY PAN EVAPORATION AT STATIONS 2 AND 6 DURING 1972 . VI-11
VI-16. SOLAR ENERGY AND EVAPORATION IN FAHKAHATCHEE STRAND
DURING 1972 VI-12
VI-17. GRAPHICAL REPRESENTATION OF ACTUAL RAINFALL AND PAN
EVAPORATION IN FAHKAHATCHEE STRAND TO THREE COMPUTED
EVAPOTRANSPIRATION METHODS VI-15
VI-18. CONSUMPTIVE USE OF WATER BY BARLEY AND PAN EVAPORATION
DURING A TYPICAL GROWING SEASON VI-16
VI-19. RELATIONSHIP OF PERCENT OF FULL SOD EVAPOTRANSPIRATION
TO PERCENT SOD COVER BY STEWART AND MILLS (1967) VI-16
VI-20. COMPARISON OF SOIL TYPES TO EVAPORATION BY SLEIGHT
(1917) VI-17
VI-21. GRAPH OF GROUND WATER LEVELS IN EPA WELL W-l SHOWING
EFFECTS OF EVAPOTRANSPIRATION VI-17
VI-22. COMPARISON OF AVERAGE EVAPOTRANSPIRATION FROM EPA
WELLS W-2 AND W-3 TO AVERAGE MONTHLY THEORETICAL
ESTIMATES FROM TABLE VI-5 VI-17
VI-23. RELATIONSHIP OF CAPILLARY WATER SURFACE TO GROUND
WATER TABLE IN FAHKAHATCHEE STRAND DURING 1972 VI-18
VI-24. ESTIMATED SPECIFIC YIELD AS A FUNCTION OF THE DEPTH
OF THE GROUND WATER TABLE VI-21
VII-1. MAP OF STUDY AREA SHOWING HYDROLOGY STATIONS VII-2
VII-2. HYDROGRAPH AND CUMULATIVE DISCHARGE FOR ALLIGATOR
ALLEY BETWEEN S. R. 29 AND THE EASTERNMOST GAG CANAL . . . VII-3
VII-3. HYDROGRAPH OF MONTHLY MEAN DISCHARGES OF FLOWS PASSING
JANES SCENIC DRIVE VII-4
VII-4. HYDROGRAPH OF FLOWS PASSING JANES SCENIC DRIVE VII-5
VII-5. CUMULATIVE DISCHARGE PASSING JANES SCENIC DRIVE VII-6
VII-6. MAJOR DRAINAGE CANALS NEAR FAHKAHATCHEE STRAND VII-7
VII-7. HYDROGRAPH OF.MONTHLY MEAN DISCHARGE OF THE GAG CANAL
SYSTEM AT REMUDA RANCH VII-8
xix
-------
VII-8. INUNDATED AREAS OF FAHKAHATCHEE STRAND FROM JANUARY -
JUNE, 1972 „ VII-9
VII-9. INUNDATED AREAS OF FAHKAHATCHEE STRAND FROM JULY -
DECEMBER, 1972 „ VII-10
VII-10. HYDROGRAPH OF EPA WELL W-l VII-11
VII-11. HYDROGRAPH OF USGS WELL C-296 VII-11
VII-12. HYDROGRAPH OF EPA WELLS W-2 AND W-3 VII-12
VII-13. HYDROGRAPH OF USGS WELL C-496 VII-12
VII-14. HYDROGRAPH OF EPA WELL W-4 VII-13
VII-15. HYDROGRAPH OF EPA WELL W-5 VII-13
VII-16. NET CHANGE OF PIEZOMETRIC SURFACE INLAND FROM THE OCEAN . VII-15
VII-17. TIME LAG OF TIDAL EFFECT ON GROUND WATER LEVELS INLAND
FROM THE OCEAN VII-15
VII-18. GENERAL RESPONSE OF PIEZOMETRIC SURFACE TO TIDAL
EFFECTS VII-16
VII-19. GENERAL RESPONSE OF PIEZOMETRIC SURFACE TO TIDAL
RESPONSE VII-16
VII-20. GROUND WATER LEVELS IN FAHKAHATCHEE STRAND ON NOVEMBER 11,
1971 VII-17
VII-21. GROUND WATER LEVELS IN FAHKAHATCHEE STRAND ON MAY 4,
1972 VII-18
VII-22. GROUND WATER LEVELS IN FAHKAHATCHEE STRAND ON OCTOBER 10,
1972 VII-19
VII-23. GROUND WATER RECESSION AT EPA WELLS W-l, W-2, and W-4 . . VII-20
VII-24. EFFECTS OF PUMPAGE ON GROUND WATER LEVELS IN THE
VICINITY OF JANES SCENIC DRIVE AND THE EASTERNMOST
GAG CANAL VII-21
VII-25. INCREASE IN DISCHARGE DUE TO GROUND WATER INFILTRATION
IN THE EASTERNMOST GAG CANAL VII-23
VII-26. SUBSURFACE FLOW IN THE SOUTHERN END OF FAHKAHATCHEE
STRAND VII-24
VII-27. PAN EVAPORATION VERSUS RAINFALL IN FAHKAHATCHEE STRAND . . VII-24
xx
-------
Page
VII-28. FAHKA UNION BAY SHOWING MAJOR BATHYMETRIC FEATURES .... VII-27
VII-29. FAHKAHATCHEE BAY SHOWING MAJOR BATHYMETRIC FEATURES . . . VII-28
VII-30. DEPTH PATTERNS IN FAHKA UNION BAY AT MSL VII-29
VII-31. DEPTH PATTERNS IN FAHKAHATCHEE BAY AT MSL VII-29
VII-32. VOLUME, AREA, DEPTH RELATIONSHIP IN FAHKA UNION BAY . . . VII-30
VII-33. VOLUME, AREA, DEPTH RELATIONSHIP IN FAHKAHATCHEE BAY . . . VII-30
VII-34. MAP OF STUDY AREA SHOWING FLOW MEASUREMENT STATIONS . . . VII-31
VII-35. CROSS SECTIONAL AREA RELATIONSHIP TO TIDAL STAGE AT
STATION 10 VII-32
VII-36. VELOCITY CURVE AND TIDAL STAGE AT STATION 10 VII-32
VII-37. FLOW VERSUS TIME AT STATION 10 VII-32
VII-38. FLOW DIRECTIONS AND PERCENTAGE OF DISCHARGE IN
FAHKAHATCHEE BAY DURING FLOODING TIDE, OCTOBER 5 AND 6,
1972 VII-33
VII-39. FLOW DIRECTION AND PERCENTAGE OF DISCHARGE IN
FAHKAHATCHEE BAY DURING EBBING TIDE, OCTOBER 5 AND 6,
1972 VII-34
VII-40. SALINITY CURVES FOR STATIONS 1, 10 AND MID-BAY,
OCTOBER 5, 1972 VII-35
VII-41. DILUTION CURVE FOR FAHKAHATCHEE BAY. FRACTION OF
ORIGINAL CONCENTRATION REMAINING VERSUS TIDAL CYCLES . . . VII-36
VII-42. MONTHLY AVERAGE OF PERCENT OF A GIVEN CONTAMINANT
REMAINING IN FAHKAHATCHEE BAY AFTER 40 TIDAL CYCLES . . . VII-36
VII-43. MAP OF FAHKA UNION AND FAHKAHATCHEE BAYS SHOWING STATION
LOCATIONS FOR ESTUARY SALINITY STUDIES VII-37
VII-44. MAP OF ESTUARY SHOWING STATION LOCATIONS FOR "CANAL-
PASSES" SALINITY STUDIES VII-38
VII-45. ISOHALINES IN BAYS AT LOW TIDE (A.M.) ON FEBRUARY 1,
1972 , VII-39
VII-46. ISOHALINES IN BAYS AT HIGH TIDE .' .M.) ON FEBRUARY 1,
1972 VII-39
VII-47. ISOHALINES IN BAYS AT HIGH TIDE (A.M.) ON APRIL 7, 1972 . VII-40
VII-48. ISOHALINES IN BAYS AT LOW TIDE (P.M.) ON APRIL 7, 1972 . . VII-40
xxi
-------
VII-49 .
VII-50.
VII-51.
VII-52.
VII-53.
VII-54.
VII-55.
VII-56.
VII-57.
VII-58.
VII-59.
VII-60.
VII-61.
VII-62.
VII-63.
VII -64.
VII-65.
VII-66.
ISOHALINES IN BAYS AT LOW TIDE (A.M.) ON JUNE 27,
1972
ISOHALINES IN BAYS AT HIGH TIDE (P.M.) ON JUNE 27,
1972
ISOHALINES IN BAYS AT LOW TIDE (A.M.) ON AUGUST 10,
1972
ISOHALINES IN BAYS AT HIGH TIDE (P.M.) ON AUGUST 10,
1972
ISOHALINES IN BAYS AT LOW TIDE (P.M.) ON OCTOBER 31,
1972
ISOHALINES IN BAYS AT HIGH TIDE (A.M.) ON OCTOBER 31,
1972
ISOHALINES IN BAYS AT LOW TIDE (A.M.) ON DECEMBER 7,
1972
ISOHALINES IN BAYS AT HIGH TIDE (P.M.) ON DECEMBER 7,
1972
SALINITY VARIATIONS IN FAHKA UNION BAY AT LOW TIDE
DURING 1972
SALINITY VARIATIONS IN ^AHKA UNION BAY AT HIGH TIDE
DURING 1972
SALINITY VARIATIONS IN FAHKAHATCHEE BAY AT LOW TIDE
DURING 1972
SALINITY VARIATIONS IN FAHKAHATCHEE BAY AT HIGH TIDE
DURING 1972
SALINITY VARIATIONS IN THE FAHKA UNION CANAL AT LOW TIDE
DURING 1972
SALINITY VARIATIONS IN THE FAHKA UNION CANAL AT HIGH TIDE
DURING 1972 . . . . . .
SALINITY VARIATIONS IN FAHKAHATCHEE PASS AND WEST PASS
DURING 1972
TIME, STAGE, AND SALINITY VARIATIONS IN FAHKA UNION
CANAL AND FAHKA UNION PASS ON OCTOBER 17, 1972
TIME, STAGE, AND SALINITY VARIATIONS IN FAHKAHATCHEE
RIVER ON OCTOBER 18, 1972
TIME, STAGE, AND SALINITY VARIATIONS IN FAHKAHATCHEE
PASS AND WEST PASS ON OCTOBER 19, 1972 ....
Page
VII-41
VII-41
VII-42
VII-42
VII-43
VII-43
VII-44
VII-44
VII -46
VII-46
VII-46
VII-46
VII -47
VII-47
VII-48
VII-49
VII-49
VII-50
XXll
-------
VII-67. ANNUAL TEMPERATURE RANGES IN FAHKA UNION AND
VII-68.
VIII- 1.
VIII-2.
VIII-3.
VIII-4.
VIII-5.
VIII-6.
VIII-7.
VIII-8.
VIII-9.
VIII-10.
VIII-H.
VIII- 12.
VIII- 13.
IX- 1.
IX- 2.
IX- 3.
FAHKAHTACHEE BAYS DURING 1972
TIDAL AMPLITUDES WITH LUNAR PHASE FOR FAHKAHATCHEE BAY . .
GRID SYSTEM
SEDIMENTATION TRAP
SEDIMENT TRAP LOCATIONS
DISTRIBUTION OF BOTTOM CHARACTERISTICS
MEAN ORGANIC SEDIMENTATION RATE ISOPLETHS IN GM/M2/DAY,
JULY 17 THROUGH NOVEMBER 20, 1972
MEAN TOTAL SEDIMENTATION RATE ISOPLETHS IN GM/M2/DAY,
JULY 17 THROUGH NOVEMBER 20, 1972
BIWEEKLY TOTAL AND ORGANIC SEDIMENTATION RATES FOR
FAHKAHATCHEE BAY ....... 0
BIWEEKLY TOTAL AND ORGANIC SEDIMENTATION RATES FOR
FAHKA UNION BAY
BIWEEKLY MEAN ORGANIC FRACTION OF THE TOTAL TRAPPED
SEDIMENT FOR FAHKAHATCHEE AND FAHKA UNION BAYS
COMPARISON OF DAY AND NIGHT WIND SPEEDS. DATA FROM
UNIVERSITY OF MIAMI, ROOKERY BAY MARINE STATION
COMPARISON OF DAY AND NIGHT MEAN TIDAL STAGES
COMPARISON OF BAY SEDIMENTATION RATES TO WATER COLUMN
DEPTH IN FAHKAHATCHEE AND FAHKA UNION BAYS
ORGANIC CONTENT OF BAY BOTTOM CORES
DETRITAL SAMPLING STATIONS IN FAHKAHATCHEE AND FAHKA
UNION BAYS .....
TYPICAL 12 HOUR TIDAL CYCLE SHOWING DETRITAL SAMPLING
PERIODS
AVERAGE EBB TIDE TOG CONCENTRATION AT THE SIX SAMPLING
STATIONS IN JULY AND OCTOBER, 1972 .•
VII-51
VII-51
VIII-2
VIII-2
VIII-3
VIII-4
VIII-7
VIII-8
VIII-8
VIII-8
VIII-9
VIII-9
VIII-9
VIII-10
VIII- 11
IX- 2
IX- 3
IX-5
IX-4. AVERAGE EBB TIDE TANNIN AND LIGNINS CONCENTRATION AT
THE SIX SAMPLING STATIONS IN JULY AND OCTOBER, 1972 . . . IX-5
IX-5. AVERAGE EBB TIDE DETRITAL CONCENTRATIONS AT THE SIX
SAMPLING STATIONS IN JULY AND OCTOBER, 1972 IX-5
xxiii
-------
Page
X-l. LOCATION OF THE WATER QUALITY STATIONS X-2
X-2. LOCATION OF THE WATER QUALITY STATIONS X-3
X-3. pH VALUES FOR EACH STATION PER MONTH X-6
X-4. TEMPERATURE VALUES FOR EACH STATION PER MONTH X-7
X-5. THE MEDIAN AND RANGE VALUES FOR TURBIDITY X-8
X-6. THE MEAN AND RANGE VALUES FOR TOTAL ALKALINITY X-8
X-7. THE MEAN AND RANGE VALUES FOR CHLODIRES X-9
X-8. THE MEAN AND RANGE VALUES FOR DISSOLVED OXYGEN X-10
X-9. THE MEAN AND RANGE VALUES FOR TANNIN AND LIGNIN "LIKE"
COMPOUNDS X-10
X-10. THE MEAN AND RANGE VALUES FOR TOTAL KJELDAHL NITROGEN . . X-ll
X-ll. THE MEAN AND RANGE VALUES FOR TOTAL PHOSPHORUS X-12
X-12. THE MEAN AND RANGE VALUES FOR TOTAL ORGANIC CARBON . „ . . X-13
X-13. A SCATTERED DIAGRAM OF TOTAL KJELDAHL NITROGEN VERSUS
WATER FLOW X-13
X-14. AVERAGE TRANSPORT OF TOTAL KJELDAHL NITROGEN (KG/DAY)
FROM THREE FRESHWATER AREAS INTO THE ESTUARY X-15
X-15. AVERAGE TRANSPORT OF TOTAL PHOSPHORUS (KG/DAY) FROM
THREE FRESHWATER AREAS INTO THE ESTUARY X-16
X-16. AVERAGE TRANSPORT OF TOTAL ORGANIC CARBON (KG/DAY) FROM
THREE FRESHWATER AREAS INTO THE ESTUARY X-17
XI-1. LOCATION OF THE PESTICIDE AND METAL STATIONS XI-2
XII-1. LOCATIONS OF STATIONS USED IN THE BENTHIC PLANT GROSS
PRIMARY PRODUCTIVITY STUDIES AND IN THE SURVEYS OF LIGHT
EXTINCTION IN FAHKA UNION AND FAHKAHATCHEE BAYS, 1972 „ . XII-3
XII-2 BENTHIC RESPIROMETER XII-3
XII-3. RESPONSE AND CALIBRATION CURVE OF THE MARINE PHOTOMETER . XII-4
XII-4. TOTAL SOLAR RADIATION REPRESENTED BY VISIBLE LIGHT .... XII-4
XII-5. GENERAL DISTRIBUTION AND MEAN CONCENTRATION OF GREEN
FILAMENTOUS ALGAE IN FAHKA UNION AND FAHKAHATCHEE BAYS,
JANUARY, 1972 XII-6
xxiv
-------
XII-6. GENERAL DISTRIBUTION AND MEAN CONCENTRATION OF RED
MACRO-ALGAE IN FAHKA UNION AND FAHKAHATCHEE BAYS,
JANUARY, 1972 XII-7
XII-7 GENERAL DISTRIBUTION AND MEAN CONCENTRATION OF
MARINE GRASSES IN FAHKA UNION AND FAHKAHATCHEE BAYS,
JANUARY, 1972 XII-8
XII-8. GENERAL DISTRIBUTION AND CENTERS OF ABUNDANCE FOR
MARINE GRASSES IN FAHKA UNION AND FAHKAHATCHEE BAYS,
JULY, 1972. MEAN CONCENTRATION - (GM DRY WT/M2) XII-10
XII-9. GENERAL DISTRIBUTION AND CENTERS OF ABUNDANCE FOR
MARINE GRASSES IN FAHKA UNION AND FAHKAHATCHEE BAYS.
SEPTEMBER, 1972. MEAN CONCENTRATION - (GM DRY WT/M2) . . XII-11
XII-10. GENERAL DISTRIBUTION AND CENTERS OF ABUNDANCE FOR RED
AND GREEN ALGAE IN FAHKA UNION AND FAHKAHATCHEE BAYS,
JULY, 1972. MEAN CONCENTRATION (GM DRY WT/M2) XII-12
XII-11. GENERAL DISTRIBUTION AND CENTERS OF ABUNDANCE FOR RED
AND GREEN ALGAE IN FAHKA UNION AND FAHKAHATCHEE BAYS,
SEPTEMBER, 1972. MEAN CONCENTRATION (GM DRY WT/M2) . . . XII-13
XII-12. PRINCIPAL BOTTOM SUBSTRATE TYPES CHARACTERIZING
FAHKA UNION AND FAHKAHATCHEE BAYS, 1972 XII-14
XII-13. REGRESSION OF GROSS PRIMARY PRODUCTIVITY ON LIGHT
WITH RESPECT TO BOTTOM TYPE AND SEASON FOR FAHKA
UNION BAY, 1972 XII-16
XII-14. REGRESSION OF GROSS PRIMARY PRODUCTIVITY ON LIGHT
WITH RESPECT TO BOTTOM TYPE AND SEASON FOR
FAHKAHATCHEE BAY, 1972 XII-17
XII-15. MONTHLY AVERAGE BOTTOM ILLUMINATION AND TURBIDITY
WITH RESPECT TO AVERAGE DAYTIME DEPTHS IN FAHKA
UNION AND FAHKAHATCHEE BAYS, 1972 XII-18
XII-16. BENTHIC PLANT PRODUCTIVITY INTERRELATIONSHIP TO
AVERAGE DAILY NITROGEN LOADING PER MONTH, FAHKA
UNION BAY, 1972 XII-20
XII-17. NUTRIENT TRANSPORT ACROSS JANES SCENIC DRIVE AND
BENTHIC PLANT PRODUCTIVITY, FAHKAHATCHEE BAY, 1972 .... XII-21
XII-18. STATION LOCATIONS FOR THE STUDY OF MANGROVE AND
SALT MARSH COMMUNITY METABOLISM, 1972 XII-24
XII-19. SCHEMATIC OF DIURNAL RESPONSE OF PLANT METABOLISM .... XII-28
XII-20. RESPONSE OF MANGROVE GROSS PRIMARY PRODUCTIVITY TO
DECREASING CONCENTRATION OF FRESH WATER, NOVEMBER -
DECEMBER, 1972 XII-38
XXV
-------
XII-21. LOCATION OF SAMPLING AREAS XII-39
XII-22. EXAMPLE OF A SURVEYED QUADRAT XII-40
XII-23. AVERAGE STANDING CROP BIOMASS OF HERBS IN DRAINED
(AREA A, FIGURE XII-21) AND UNDRAINED (AREA B,
FIGURE XII-21) CYPRESS STRANDS, 1972 XII-43
XII-24. LITTER ACCUMULATION IN AN UNDRAINED CYPRESS STRAND
(AREA B, FIGURE XII-21), 1971-73 XII-45
XII-25. LITTER ACCUMULATION IN THE CENTRAL STRAND (AREA C,
FIGURE XII-21), 1971-72 XII-45
XII-26. LITTER ACCUMULATION IN A DRAINED CYPRESS STRAND
(AREA A, FIGURE XII-21), 1972 XII-45
XII-27. LITTER DECOMPOSITION ON CENTRAL STRAND FLOOR, FINE MESH . XII-49
XII-28. LITTER DECOMPOSITION ON CENTRAL STRAND FLOOR, COARSE
MESH , XII-49
XII-29. LITTER DECOMPOSITION ON DEBRIS PILE IN CENTRAL STRAND,
FINE MESH XII-49
XII-30. LITTER DECOMPOSITION ON DEBRIS PILES IN CENTRAL STRAND,
COARSE MESH XII-50
XIII-1. DISTRIBUTION OF COMMUNITIES IN THE STUDY AREA XIII-10
XIII-2. DISTRIBUTION OF COMMUNITIES IN THE BIG CYPRESS SWAMP
(INCLUDES FAHKAHATCHEE) .... XIII-11
XIV-1. STATION LOCATIONS FOR MACROBENTHOS STUDY XIV-2
XIV-2. ARITHMETIC MEAN FOR NUMBERS OF BENTHIC MACROINVERTEBRATES
INHABITING MUD SUBSTRATES IN APRIL, JUNE, OCTOBER, '1972,
FAHKA UNION AND FAHKAHATCHEE BAYS XIV-4
XIV-3. ARITHMETIC MEAN FOR NUMBERS OF BENTHIC MACROINVERTEBRATES
INHABITING SAND SUBSTRATES IN APRIL, JUNE, AND OCTOBER,
1972, FAHKA UNION AND FAHKAHATCHEE BAYS XIV-5
XIV-4. ARITHMETIC MEAN FOR NUMBERS OF BENTHIC MACROINVERTEBRATES
INHABITING SHELL SUBSTRATES IN APRIL, JUNE, AND OCTOBER,
1972, FAHKA UNION AND FAHKAHATCHEE BAYS XIV-5
XIV-5. MONTHLY NUMBERS OF GRASS SHRIMP (PALAEMONETES SPP.) TAKEN
COMBINED CATCHES OF SURFACE AND OTTER TRAWLS IN FAHKA
UNION AND FAHKAHATCHEE BAYS, 1972 XIV-7
XIV-6. MEAN MONTHLY BIOMASS, WET WEIGHT, OF PINK SHRIMP
(PENAEUS DUORARUM) AND GRASS SHRIMP XPALAEMONETES SPP.)
TAKEN FROM COMBINED CATCHES OF SURFACE AND OTTER TRAWLS
IN FAHKA UNION AND FAHKAHATCHEE BAYS, 1972 XIV-7
xxvi
-------
Page
XV-1. STUDY AREA AND SAMPLING SITES IN FAHKAHATCHEE STRAND
AND THE TEN THOUSAND ISLANDS, FLORIDA „ . . . . XV-2
XV-2. STUDY AREA AND SAMPLING STATIONS IN TEN THOUSAND
ISLANDS, FLORIDA XV-18
XV-3. DISTRIBUTION OF THE MEAN CATCHES BY NUMBER AT THE
REGULAR SURFACE AND OTTER TRAWLING STATIONS XV-22
XV-4. DISTRIBUTION OF THE MEAN CATCHES BY NUMBER AND BY
BIOMASS AT THE REGULAR SEINING STATIONS XV-22
XV-5. NUMBER OF FISH SPECIES FROM THE COLLECTIONS AT THE
REGULAR STATION BY TYPE OF GEAR AND BY TOTAL NUMBER
PER MONTH . XV-23
XV-6. DISTRIBUTION OF THE MEAN CATCHES BY BIOMASS AT THE
REGULAR SURFACE AND OTTER TRAWLING STATIONS . XV-24
XV-7. WATER TEMPERATURE AND SALINITY READINGS FOR ROUND
KEY AT SEINE STATION 2; FOR TIDAL STREAM OFF FAHKA
UNION CANAL AT SEINE STATION 5; FOR FAHKA UNION BAY
AT SURFACE TRAWL STATION 7; AND FOR FAHKAHATCHEE BAY
AT SURFACE TRAWL STATION 9 . ........ „ «, XV-26
XVI-1. THE LENGTH-WEIGHT RELATIONSHIP (LOG W = -4.7730
+ 2.8758-LOG FL) OF JUVENILE SNOOK IN THE TEN THOUSAND
ISLANDS, FLORIDA XVI-8
XVI-2. LENGTH-FREQUENCY DISTRIBUTION OF 193 JUVENILE SNOOK
FROM THE TEN THOUSAND ISIANDS, FLORIDA, 1971-72 XVI-9
XVI-3. PERCENT OF TOTAL VOLUME OF MAJOR FOOD ITEMS IN THE
STOMACH CONTENTS OF 173 JUVENILE (11-200 MM FL) AND
271 ADULT SNOOK (201-1020 MM FL) COLLECTED IN TEN
THOUSAND ISLANDS, FLORIDA, 1972 XVI-16
XVII-1. SCHEMATIC DIAGRAM OF TYPICAL EXTENDED AERATION AND
CONTACT STABILIZATION MODES OF "PACKAGE" TREATMENT
PLANTS ..... XVII-3
XVII-2. MAJOR WASTE WATER EFFLUENT BASINS IN SOUTHWEST FLORIDA . . XVII-5
xxv ii
-------
I - SUMMARY
Because of environmental considerations which arose when the Bade County
Port Authority began construction of a 39 square mile jetport on the eastern
edge of the Big Cypress Swamp, the U. S. Department of the Interior initiated
the South Florida Environmental Project. This multiagency effort was designed
to provide information for Federal, State and local decision makers in south
Florida to use in determining the best utilization of all area resources. The
overall objective of the U. S. Environmental Protection Agency's (EPA) role in
this study was to obtain necessary technical information for objective planning
of a program of wise use of south Florida's land, water, wildlife and fisheries
resources. The strategy selected by EPA was to obtain detailed technical
knowledge of the intricate interrelationships existing among the various com-
ponents of disturbed and relatively unaffected (by human activity) ecosystems.
This knowledge would then be used to formulate process studies and experimental
manipulation models for the various components of the ecosystem.
In order to accomplish the stated objectives, an in-depth study of a
representative area of the Big Cypress Swamp was conducted. The selected study
area was Fahkahatchee Strand (Figure V-3) and contiguous areas in Collier
County, including the bays and estuaries to the southwest. Field investigations
during 1971-1972 intensively examined biotic community interactions and were
concentrated on paired sites located within the Fahkahatchee Strand ecosystem
and contiguous tidal wetlands and estuaries with both contrasting natural and
hydrologically-disturbed ecosystems.
EPA assembled a team of scientists consisting of engineers, systems jecologists,
fishery biologists, freshwater biologists, botanists, chemists and technicians
to conduct the study. Offices and chemical and biological laboratories were
established at the Rookery Bay Marine Station located between Naples and Marco
Island. Additional technical support was obtained from the Chemical Services
Branch and the Biological Services Branch, Surveillance and Analysis Division,
EPA Region IV, in Athens, Georgia, and EPA's National Field Investigations
Center at Cincinnati, Ohio.
BACKGROUND
Unprecedented growth has taken place in the Collier County portion of the
Big Cypress Swamp during the last decade. From 1960 to 1970, Collier County
was the fifth fastest growing county in the United States with a population
increase from 15,753 to 36,568, or 132 percent. County planners expect the
population to reach 200,000 - 250,000 by the year 2000. Detrimental results
from this population increase on the environment is evident, especially in the
highly important estuarine areas. Onetime highly productive areas are now
being drained and mangrove forests cleared and filled for homesites and
commercial developments. Tidal streams have ceased to flow because of bulkheads.
Public access to many beaches has been blocked by oceanfront condominiums and
apartments.
1-1
-------
PHYSIOGRAPHY
The study area elevation is less than four meters above mean sea level and
land slopes gently south to southwest, varying from 8 to 16 cm/km. Natural
drainage of the land is slow and sheet flow occurs during much of the year,
especially during the hydroperiod. Geologically, the area is underlain by
Tamiami limestone of Miocene age. Soils of pure sand, marl or mixtures of
both with layers from 5 to 60 centimeters deep rest on the irregular surface
of the limestone. Peat muck fills the depressions in the bedrock.
CLIMATOLOGY
Most rainfall within the basin is of high intensity and short duration,
and the presence of artificial channels leading to tidewater effectively
exports much of this rainfall before maximum infiltration can occur.
During 1972, rainfall in Fahkahatchee Strand averaged 142 centimeters and
ranged from 120 to 194 centimeters. Maximum daily rainfall of 11.75 centimeters
occurred on June 19, 1972, during Hurricane Agnes. Rainfall occurred on 168
days during 1972. Air temperatures ranged from 4°C to 36°G with a yearly
average of 23°C. .Relative humidity ranged from 30 to 100 percent with an
average of 83 percent.
Evapotranspiration is a major consumptive user of groundwater in south
Florida. Depending upon the time of the year, evapotranspiration may account
for 90 to 95 percent of the groundwater drawdown in areas unaffected by drainage
canals. During 1972, pan evaporation in the study area averaged 139.5 centi-
meters with solar energy ranging from a low of 270 langleys/day in December to
a high of 745 langleys/day in June.
HYDROLOGY
Surface flows throughout 1972 were less than flows recorded during the
previous two years. No measurable surface water entered Fahkahatchee Strand
from above Alligator Alley during 65 percent of the year and no measurable flow
crossed Janes Scenic Drive 45 percent of the year despite higher than average
rainfall. With the exception of eight days in June, the main slough in
Fahkahatchee Strand was essentially void of surface water from April 1 until
August 27. The maximum area of inundation occurred in November, when approxi-
mately 51 percent of the study area north of U. S. 41 was inundated.
Based on historical rainfall information, calendar years 1970 and 1971 were
substantially below normal, but freshwater export from the Strand to its
dependent estuary continued at near-normal rates. Calendar year 1972 was above -
normal in rainfall, but surface water export from the Strand virtually ceased.
This lack of water export in itself is no cause for alarm and may represent a
natural cycle in which the upland ecosystem functions as a reservoir to main-
tain brackish water conditions in the estuarine zone. Wet-dry conditions are
essential to full productivity of the estuarine ecosystem. However, contrary
to expectation, groundwater tables in the upland zone failed to recover from
the depressed levels present at the end of the 1970-71 dry period, despite
rainfall inputs in 1972 well above long-term mean values. Drainage canals
1-2
-------
prevented recovery of groundwater levels. These canals had a catastrophic
effect on groundwater levels -- lowering natural levels as much as 122 centi-
meters. Maximum recession rates in the vicinity of canals were measured at
9.1 to 10.4 cm/day compared with 2.1 to 4.0 cm/day in areas unaffected by the
canals. Groundwater levels throughout the study area were below normal during
most of the year. The maximum water level drawdown below the surface was
1.79 meters in the vicinity of canals and 1.42 meters in locations not influenced
by canals. Direct groundwater drainage by canals was measured as high as 588
I/sec/km (31.7 cfs/mile). Groundwater gradients of 14.2 cm/km were found to be
typical in a north-south direction in Fahkahatchee Strand compared with 25.5
cm/km in an east-west direction near the easternmost GAG canal.
Fahka Union Bay, which receives massive discharges of freshwater drainage
from the Fahka Union Canal had salinity variations much greater than those of
Fahkahatchee Bay which receives natural runoff. During August, 1972, salinities
ranged from 27 to 32 ppt in Fahkahatchee Bay compared with 11 to 29 ppt in
Fahka Union Bay. Average salinities were always higher in Fahkahatchee Bay
than in Fahka Union Bay.
SEDIMENTATION
Fahkahatchee and Fahka Union Bay bottoms were found to be predominantly
mud, 52 and 76 percent, respectively, with sand blending to a lesser extent.
Oyster bars appear in isolated locations in both bays.
Sedimentation studies revealed mean rates of 126 and 160 gm/m^/day, the
greatest mean rates being in Fahka Union Bay. However, when these rates were
normalized for depth, the sedimentation rates for both bays were relatively
equal, indicating that resuspension of unconsolidated deposits was the principal
variable measured by this technique. Organic content of sediment deposits
decreased with depth.
DETRITAL TRANSPORT
Detritus entered Fahkahatchee Bay from salt marshes and mangrove forests
via tidal creeks and rivers at the rate of 9,000 Kg carbon/day in mid-October,
1972. At the same time, Fahkahatchee Bay released nearly 13,000 Kg carbon/day
into the Gulf of Mexico. Therefore, 4,000 Kg carbon/day came from the bay
itself. Ninety-five percent of the particulate organic carbon entering
Fahkahatchee Bay via the tidal rivers was found to be less than 0.8 micron
in size. A seasonal decrease in concentrations of total organic carbon (TOG),
tannins and lignins occurred from July to October. The quantity of organic
carbon entering Fahka Union Bay on an ebb tide was 2 to 14 times the quantity
entering Fahkahatchee Bay based upon mass per unit of bay area. However, a
majority of the organic carbon which entered Fahka Union Bay was "short
circuited" into the Gulf by the relatively deep Fahka Union Canal that
transects this bay.
1-3
-------
WATER CHEMISTRY
Fourteen physical and chemical water quality parameters were analyzed
monthly from sixteen stations:
Temperature
Turbidity
Apparent color
Conductivity
PH
Alkalinity
Chlorides
Dissolved Oxygen
Total and soluble Kjeldahl nitrogen
Nitrate and nitrite nitrogen
Total and soluble organic carbon
Total and soluble phosphorus
Sulfate
Tannin and lignin "like" compounds.
Surface waters in Fahkahatchee Strand consistently had greater concentrations of
total phosphorus, total organic carbon, total Kjeldahl nitrogen, apparent color
and tannin and lignin "like" compounds than did the drainage canals. Alkalinity,
dissolved oxygen and turbidity were higher in the canals. However, due to the
large flows in the drainage networks, the quantities of nutrients transported
by the Fahka Union Canal were much greater than those transported across Janes
Scenic Drive (five times more total Kjeldahl nitrogen, ten times more total
phosphorus, and seven times more total organic carbon). Rainfall and flow
variations produced annual trends in many of the water quality parameters.
During the wet season (summer and fall), tannin and lignin "like" compounds
and nutrient concentrations increased. During the dry season (winter and spring),
conductivity and chloride values increased.
Ten metals and 24 pesticides which have common use in south Florida were
monitored quarterly in water, sediment and biota. In some cases the median
concentration of metals found in the sediment of borrow canals was four to five
times greater in the .developed areas than in the undeveloped areas. The
concentration of all metals found in the sediment of Fahka Union Bay was at
least twice that observed in Fahkahatchee Bay. However, the water column in
Fahkahatchee Bay contained greater concentrations of metals than did Fahka Union
Bay.
Surface waters within the study area were free of detectable levels of
pesticides. In the sediments, only DDT, ODD, DDE, and PCB's were detected with
any significant frequency. The highest concentrations of DDT, DDD, and DDE
were 160, 160, and 96 ug/Kg, respectively, and were observed during January,
1972, in sediments of the Barron River Canal which drains a portion of the
agricultural area near Immokalee known as the Okaloacoochee Slough.
Only six pesticides (DDT, DDD, DDE, PBC's, Dieldrlu, and Myrex) were
observed in biotic samples. With the exception of DDT in the Florida gar and in
an Atlantic bonito, the detected levels of the above mentioned pesticides were
generally less than 50 t|g/Kg (wet wt). The most probable source of these com-
pounds was upland agricultural runoff.
1-4
-------
PRIMARY PRODUCTION
Studies of primary productivity were conducted on the benthic plant
communities inhabiting Fahka Union and Fahkahatchee Bays. Gross primary
productivity was determined with the use of benthic respirometers. Estimates
of net productivity were established from plant community biomass accumulation .
rates measured over the sampling year with plant harvesting techniques. Results
of these studies show that Fahka Union Bay maintained a benthic plant community
comprised mainly of green filamentous algae, Rhizoclonium hookeri, and red
macroalgae, Gracilaria spp. and Dasya harveyi. This algal community sustained
an average gross primary productivity of 0.24 gm C/m^/day for the year. Over
the same period, net productivity in terms of biomass accumulation was 0.17
gm dry wt/m^/day. Marked increases in primary productivity occurred concurrently
with seasonal flushing of nutrients into the bay from upland sources via Fahka
Union Canal.
Fahkahatchee Bay supported a benthic seagrass community composed primarily
of Diplanthera wrightii, with some Thalassia testudinum, and Halophila
engelmannii. Gross primary productivity of the community followed a slight
but steady decline through the sampling year with an average rate of 0.20 gm
dry wt/m^/day.
Gross primary productivity was obtained for salt marsh and mangrove
communities along a decreasing freshwater gradient. Diel metabolic rates were
determined from field measurements of carbon dioxide uptake and release by
plant components isolated in clear plastic metabolic chambers. Metabolic
measurements in the salt marshes were limited to the emergent portions of the
marsh plants. Gross productivity of the salt marsh community increased with a
reduction in fresh water availability and concomitant decrease in the number of
plant species inhabiting the marshes. Based on the measured increase in the
ratio of 24-hour respiration to gross primary productivity, and in the chloride
concentrations, the reduction in the number of marsh species appeared related
to the effects of metabolic stress caused by increased salinity.
Mangrove tree gross productivity also responded to a freshwater concentra-
tion gradient that ranged from a minimum of 4.7 ppt to a maximum of 1-6 ppt
chlorides. Gross metabolism of red mangrove leaves, Rhizophora mangle, decreased
steadily with increasing levels of chlorides. In contrast, leaf gross produc-
tivity of the black, Avicennia nitida, and white, Laguncularia racemosa, mangrove
species increased in response to greater chloride levels. Total mangrove red,
black and white community metabolism, trees only, responded to a decreasing
gradient of freshwater concentrations in the following manner: gross produc-
tivity increased, respiration increased, and net productivity decreased.
Mangrove community
Mean surface water metabolism (gm C/m2/day)
chlorides (ppt) GPP NPP R24
4.7 10.3 6.6 3.7
12.7 11.8 7.5 4.3
16.0 13.9 4.8 9.1
The effects of land drainage systems on annual increases in standing crop
biomass of a cypress forest community were assessed. Dramatic differences in
1-5
-------
standing crop biomass were evident between drained and undrained regions of the
cypress forest. Combined biomass of trees, shrubs, and herbs averaged nearly
9 kg dry wt/m2 in the drained region while the undrained area supported a
similar community type with a combined standing crop biomass averaging 17 kg
dry wt/m2. The net gain in community biomass for the sampling year averaged
3.1 gm dry wt/m2/day and 0.31 gm dry wt/m2/day for the undrained and drained
regions, respectively.
Litter accumulation on the forest floor in the cypress sites was assessed.
The undrained region showed an annual accumulation of 374 gm dry wt/m2 compared
to 267 gm dry wt/m2 in the drained area (all values corrected for losses to
decomposition). Leaf turnover rate in the undrained region was 0.791 turnovers
per year vs 1.035 turnovers per year in the drained area. Thus, the canopy in
the drained area is subject to thinning. Relatively large leaf-fall rates in
the drained area produces greater sunlight penetration and, with the abnormal
lack of surface waters, create a large stock of highly flammable materials on
the forest floor.
Preweighed parcels of litter (leaves, stems, etc.) were placed on the
forest floor to determine the effects of microclimatic conditions on litter
decay over a 12 month period. Results of this study show that the remineraliza-
tion of litter preceded at a reduced rate under drained conditions in the cypress
forest.
PLANT SOCIOLOGY
One of the most striking features of the plant communities in the study
area was the broad range of habitats present and the great species richness of
the area. Within each habitat, generally more than 20 plant species occurred
frequently enough to be considered common.
The cypress strand contained some 23 common species of woody trees and
shrubs, 17 species of common vines, 28 species of common terrestrial herbs, 18
species of common epiphytic herbs, and 15 species of common aquatic vascular
plants including a few macroalgae. The wet prairies, which seem at casual
glance to be monocultures of grass, in fact contained 28 species of common
monocots and 36 species of common dicots. The pine and palm association
contained some 13 species of common woody trees and shrubs, and 33 species of
common herbs. Ditches and canals contained some 34 common aquatic plant species.
Even roadsides adjacent to prairies and cypress strands contained over 28 common
species of plants. Strands, lakes and ponds contained 23 species of common
aquatics. The mangrove swamps contained, in addition to four species of common
trees, 25 species of common herbs, three species of common lianas and several
species of epiphytic herbs. Not included above are the many rare and exotic
plants.
FISH
Investigations of the fish fauna of Fahkahatchee Strand and the Ten Thousand
Islands were directed toward acquiring base line estimates of the absolute
abundance (standing crops of fishes) of freshwater and estuarine fish populations
and on the relative abundance of estuarine fishes.
1-6
-------
Estimates of the fish populations were obtained monthly at two sites in
Fahkahatchee Strand with a square meter frame and rotenone. Distribution and
abundance of the fishes in the freshwater areas of Fahkahatchee Strand were
directly related to the cyclic nature of the hydroperiod. The lowest number
and biomass of fish occurred during the wet summer and fall periods, whereas
the largest number and biomass occurred during the dry winter and spring months
when the fishes were concentrated in small ponds and sloughs.
Single estimates of fish populations were taken at ten other locations with
either block net and rotenone or an enclosed net and seines. The biomass
estimates for the fish populations were as follows:
Natural Lake in Fahkahatchee Strand - 14.1 gm/m2
GAG Canal - 6.6 gm/m2
Tamiami Canal - 170.7 gm/m2
Three sites in Fahka Union Bay - 0.7, 2.2, and 14.3 gm/m2
Three locations in Fahkahatchee Bay - 9.3, 1.4, and 6.3 gm/m2, and
A tidal stream in north Fahkahatchee Bay - 89.8 gm/m2.
During the year, a total of 273,270 fishes, representing 96 species and 41
families, were taken in seines, surface trawls and other trawls at 14 stations.
Their total biomass was 414.3 kg.
By number and biomass, the dominant species in the regular catches were the
anchovies, primarily the bay anchovy (Anchoa mitchilli). Excluding the anchovies
(Anchoa spp.), the most abundant fishes by number were yellowfin menhaden
(Brevoortia smithi), scaled sardine (Harengula pensacolae), pinfish (Lagodon
rhomboides). silver perch (Bairdiella chrysura), and silver jenny (Eucinostomus
gula). However, ranked by biomass, the dominant species were southern stingray
(Dasyatis americana), scaled sardine, fantail mullet (Mugil trichodon), striped
mullet (M. cephalus), and silver perch.
Analyses of the catches from the relative abundance and standing crop
studies showed that a greater abundance and diversity of fishes existed in
Fahkahatchee Bay, a relatively undisturbed environment, than in Fahka Union Bay,
a disturbed estuary. The composition of benthic fishes and the types of bottom
vegetation differed between bays. Abundant benthic fishes, such as pinfish and
silver perch, inhabited the dense stands of grasses (Thalassia and Diplanthera)
in Fahkahatchee Bay, while these grasses and their associated fishes were scarce
in the other bay.
A total of 193 juvenile snook (Centropomug undecimalis) were seined from
tidal streams in the Ten Thousand Islands between November, 1971, and December,
1972. These juveniles ranged in size from 14.0 to 196.0 mm fork length
(anterior extremity of closed jaws to the tip of the rays in the center of the
caudal fin). Most individuals (94.8 percent) were members of the 1972-year-
class.
For the first time, tidal streams were proven to be a major nursery area
for young snook. Shallow brackish stream habitats provided an abundant supply
of small forage organisms, flowing water, low salinities, favorable water
temperatures, and the general absence of piscivorous predators.
1-7
-------
The growth rate of juvenile snook was approximately 1 mm per day during
the warmer months. This estimate was based on the size of the juveniles at
the time of collection in conjunction with the duration of the spawning season.
By the end of December, a few individuals of the 0 age-class had attained a
length of 200 mm FL.
Stomach analyses of 183 juvenile snook (142 contained food) indicated that
the main food items, by volume, were fishes (81 percent), shrimps (16 percent),
crabs (2 percent), with lesser amounts of zooplankton and immature insects.
Dominant forage fishes were poeciliids, cyprinodonts, and atherinids. Feeding
intensified in moving water following a slack tide.
As a function of growth, juvenile snook exhibited two distinct feeding
stages. Smaller juveniles (< 20 mm FL) were exclusively planktivorous. A
mixture of food items, consisting of zooplankton, fishes, and palaemonid shrimp,
were consumed during a transitional period between 21 to 25 mm in length.
Larger juveniles (>26 mm FL) acquired a carnivorous diet (mainly piscivorous).
On a" volumetric basis, the summer diet of 271 adults (142 stomachs contained
food) consisted of fishes (55 percent), crabs (32 percent), and shrimps (10
percent). Peaks in feeding intensity generally occurred from sunset to sunrise
in conjunction with moving water.
WASTE INVENTORY
Current waste treatment practices and disposal methods for southwest
Florida were inventoried. Of the 309 facilities inventoried, 88 percent were
of a design capacity of less than 50,000 gpd, mostly package plants. Most
plants provided the equivalent of secondary treatment.
1-8
-------
II - CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
The potential consequences of poor surface water conservation practices
have serious economic as well as ecological implications. The apparent super-
abundance of water in south Florida is a pernicious illusion. Without adequate
water reserves, the inevitable series of low rainfall years will impose economic
hardships on the human population of south Florida. Effective surface water
control schemes should involve the utilization of natural ecological systems,
which incorporate water conservation "practices" as part of their normal
functioning. Use of the natural system should be an integral part of any
comprehensive conservation or development planning.
The effects of drainage works on Big Cypress regional ecosystems can be
summarized in terms of hydrologic variables and entrained influences on the
structure and metabolism of biological communities.
1. With favorable climatic conditions, recreational opportunities, and
availability of suitable home sites, Collier County will continue its rapid
rate of growth. However, the quality of the county land and water environment
will deteriorate at an accelerating rate as population pressures are applied to
the existing ecosystem of south Florida.
2. During 1972 rainfall was the major source of water input into the
Fahkahatchee study area. Overland flow entering the study area was negligible.
3. Evapotranspiration was the major source of water export from the study
area (76.6 percent during 1972), followed by canal export, surface flow and
subsurface flow.
4. Groundwater recession rates were approximately four times greater in
the_vicinity of canals than in an undrained area. As designed, canals remove
surface waters efficiently and rapidly and lower the groundwater table. Canals
alone cannot maintain higher than natural groundwater levels. Nearly two million
people (based on 100 gallons per person per day) could have been supported on
the quantity of fresh water discharged by the Fahka Union Canal during 1970-1972.
Discharges from drainage canals had a pronounced effect on salinity patterns
within the estuaries. Bays receiving upland drainage via canals had lower
salinities than did natural bays.
5. Mud is the primary substrate of the majority of the Ten Thousand
Islands Bays. Sedimentation rates were greater in Fahka Union Bay than in
Fahkahatchee Bay. Fahka Union Bay was more susceptible to substrate resuspension
due to its shallowness. The topography of Fahkahatchee Bay is significantly
different than Fahka Union Bay in that the latter bay is 30 percent shallower
and has only 25 percent of the area of Fa^-ihatchee Bay.
6. The volatile solids (organic) frac ion of the settling material
followed a seasonal pattern which peaked in the early fall. In terms of
organic content, no clear-cut differences were evident from the limited coring
in the two bays.
II-l
-------
7. The quantity of nutrients (carbon, nitrogen, phosphorus) reaching Fahka
Union Bay via Fahka Union Canal was more than five-fold the quantity passing
Janes Scenic Drive and potentially entering Fahkahatchee Bay. However, due to
short-circuiting of nutrients entering Fahka Union Bay, concentrations of nutrients
in Fahkahatchee Bay were generally greater than those in Fahka Union Bay.
Large amounts of organic carbon enter Fahkahatchee Bay from the salt
marshes and mangrove forests via tidal creeks. The carbon material is reduced
in the bay to more soluble forms. In mid-October, of the daily net quantity of
carbon released via Fahkahatchee Bay to the Gulf of Mexico, at least two-thirds
of this quantity originated from the mainland salt marshes and mangrove forests.
On ebb tides, Fahkahatchee Bay received considerably less organic carbon
per unit area than did Fahka Union Bay. The latter bay received large fresh-
water flows via the Fahka Union Canal from a network of drainage canals
extending as far north as Corkscrew Swamp Sanctuary. However, because Fahka
Union Canal channelizes its flow through the bay, a major portion of its load
is "short circuited" to the Gulf of Mexico.
8. Canals receiving runoff from developed areas showed a buildup of metals
in bottom sediments.
9. Concentration of metals in the sediments of Fahka Union Bay were on the
order of two to three times greater than those in Fahkahatchee Bay. Freshwater
discharge from the Fahka Union Canal and the lack of the potential filtering
mechanism of mangrove forests appear to be the cause of this disparity. The
long term effect of this buildup of metals is not known at this time.
10. No major difference in toxic metal concentrations was found in waters
from either canals or natural areas.
11. Low concentrations of pesticides were detected in sediments and
biological samples such as sport and commercially important fishes and shellfish.
This shows that pesticides are present in the aquatic environment of south
Florida.
12. Normal overland sheet flow of freshwater along a broad front infuses
minerals and nutrients into coastal salt marshes and mangrove communities.
Drainage canals short-circuit sheet flow and divert waterborne minerals and
nutrients directly to estuarine waters where they are rapidly incorporated into
sediment and are exported into the Gulf of Mexico. Mineral losses from short-
circuiting processes result in reduced primary productivity of salt marshes and
mangrove communities and thereby lessen the export to the estuary of detrital
materials that form the basic food source for fishery resources.
13. Mangrove ecosystems represent a key factor in the natural economy of
south Florida's estuarine ecosystems. They provide a habitat and substrate for
many fish and aquatic organisms and also provide detritus through leaf fall
throughout the year.
14. The community trend in mangrove metabolism is a function of a synergistic
action between nutrient availability and tidal flushing.
15. Fahka Union Bay supported a benthic plant community comprised mainly
of green filamentous and red macroalgae, whereas Fahkahatchee Bay maintained a
II-2
-------
benthic plant community which consisted primarily of marine grasses. These
grasses provided extensive habitat for numerous species of invertebrates and
juvenile and adult forage and sport fishes. There is no apparent reason to
suspect that Fahka Union Bay ever supported an extensive community of marine
grasses.
16. Gross primary productivity of benthic plant communities in Fahka Union
and Fahkahatchee Bays represented a small fraction of that maintained by the
contiguous mangrove ecosystems.
17. Drainage works effected a reduction in primary productivity in cypress
forests and wet prairie ecosystems. Reductions in productivity were directly
proportional to diminished freshwater availability. Cypress forests in undrained
regions maintained primary productivity rates nearly ten times that of drained
areas.
18. The draining of wet prairies that are normally inundated during the
wet season induced a series of undesirable side effects:
• Periphytic algal growths, which form the food base for much of the
population increases of small forage fishes, were completely
suppressed.
• The prairies became unavailable as habitats for populations of
small fishes because of the lack of standing water.
• The population levels of larger animals dependent on these
resources for their food supply necessarily declined, especially
the wading bird communities.
19. Drainage of the cypress forest initiates a canopy thinning process
which leads to greater light penetration to the forest floor. The consequences
of a dryer understory leads to changes in microclimatic conditions unfavorable
for litter decay. Thus, litter accumulation is accelerated and leads-to
increased fuel sources for destructive wildfires.
20. Dependence of the plant communities on direct rainfall alone for water
supply induces at the very least a 40 percent drop in net primary production,
which represents an equivalent drop in magnitude of the carrying capacity of the
ecosystem for wildlife.
21. Responses of the biological communities of the area to seemingly mild
drying shifts in hydrologic conditions are surprisingly large. Net production
is markedly depressed, posing a major threat to dependent wildlife populations.
22. The smallest standing crop of fishes per unit area of water in Fahka-
hatchee Strand occurred during the wet season (summer and fall) when opportunities
for dispersal were greatest. The concent—*tion of fishes per unit area steadily
increased as the surface waters receded during the dry months (winter and spring).
23. Predominant freshwater fishes in the Strand belonged to the following
families: Livebearers (Poeciliidae), killifishes (Cyprinodontidae), sunfishes
(Centrarchidae), and catfishes (Ictaluridae). Ecologically, these fishes
function as intermediate links in the food chain between primary producers and
numerous fish-eating predators, such as wading birds, alligators, turtles,
II-3
-------
snakes, and other animals.
24. Standing crop of fishes in the Tamiami and Fahkahatchee borrow canals
greatly exceeded the standing crop associated with GAG drainage canals.
25. A greater abundance and diversity of fishes inhabited Fahkahatchee
Bay, an essentially undisturbed estuary, than in Fahka Union Bay, a man-
influenced environment. Major dissimilarities between the bays existed in the
composition of benthic fishes and the bottom plant communities. Numerous
benthic fishes, such as pinfish and silver perch, were associated with the
stands of grass beds in Fahkahatchee Bay, whereas these grasses and their fish
inhabitants were sparce in the disturbed bay.
26. Tidal streams in the upper reaches of the estuary were identified for
the first time as the major nursery grounds for juvenile snook, Centropomus
undecimalis, in the Ten Thousand Islands.
27. Random land development patterns have led to the use of small wastewater
treatment facilities. An end result of this continuing trend is that many small
plants are serving areas which could presently be served by a central plant.
Nearly 90 percent of the wastewater treatment plants listed were of a design
capacity of less than 50,000 gpd.
28. A majority of the waste treatment facilities are "package plants"
designed to provide secondary treatment. However, the efficiency of these
plants is questionable due to a severe lack of qualified operators.
II-4
-------
RECOMMENDATIONS
1. Recognizing the potentially severe seasonal water shortages
of this region and the consequence of a unique local climate and
geomorphology, regional watershed management plans and applicability
to south Florida of the law of "riparian rights" are essential,
especially to the interaction of competing agircultural, residential,
recreational, and biologically-related water use patterns. Without
management plans and riparian principles, a major disruption in the
present diverse, stable (long-term), and productive natural ecosystems
will occur.
2. The philosophy of direct surface water removal should be
replaced by one of surface water redistribution. Canal drainage
systems and borrow canals that currently waste excessive quantities of
surface waters by open channel connections to the Gulf of Mexico
should be eliminated. Control structures should be installed in all
present north-south canal works in southern Collier County. These
structures should direct waters from GAG, Barron'River (S.R. 29), and
any other north-south canals into the Alligator Alley borrow canal.
The latter could serve as a distributor canal, reinjecting surface
flow on a broad front into the Fahkahatchee Strand and other areas of
the Big Cypress. The Tamiami Trail (U.S. 41) canal could serve a
similar function in restoring sheet flow to the mainland salt marshes
and mangrove forests.
3. Present borrow and canal drainage systems should be mpdified
by adding adjustable weirs and dikes or other water level control
structures designed to minimize the hydraulic gradient between canal
water and groundwater levels.
4. Developers of water supply well fields should be more acutely
conscious of the implications of the Ghyben-Herzberg relationship*,
and the interaction of man's patterns of water supply utilization with
regional hydrologic dynamics. Water management districts should
prohibit the over-utilization of groundwater resources by private
developers at the expense of the county as a whole.
5. All land development plans need to include studies of manage-
ment alternatives and recommended actions to control the effects of
development on both the local and regional hydrologic budgets. These
studies must include estimates of impacts of developments on both down-
stream and upstream water users, and must take into account both long-
term and short-term variations in climatic activity cycles. Water
*Because of a lower specific gravity, freshwater floats on top of saltwater.
The depth to the saltwater is related to the height of the freshwater above mean
sea level. Assuming a specific gravity of 1.025 for seawater, then for each foot
of freshwater which occurs above sea level, 40 feet of freshwater extends below
mean sea level. Therefore, a lowering of the freshwater table one foot would
increase the level of saline water by 40 feet.
II-5
-------
budgets based on long-term mean conditions do not adequately portray
potential impact on systems which require extreme hydroperiod
variations.
6. Slough and strand ecosystems are uniquely valuable as water
and wildlife reservoirs, and should have a special zoning category
designed for their preservation, including a substantial buffer fringe
of wet prairie or pineland.
7. Mangrove forests, especially those located in mainland zones
receiving overland freshwaters, are of key importance to overall
estuarine productivity, including commercial and sports fisheries.
Special zoning measures should be utilized to minimize destructive
interference with these resources.
8. Benthic marine grass communities must be recognized as uniquely
valuable habitats for the production of numerous species of inverte-
brates and fishes. Safeguards for protection of the natural ecosystem
should be inherent in all zoning in coastal and upland regions.
9. The effect of fluctuations (natural and man-induced) in the
hydrologic cycle on water quality should be monitored regularly.
Monitoring data should then be used in development of programs of
water quality management, integrating both technological and natural
ecosystem components as a composite system of water quality control.
10. Introduction of pesticides, herbicides, heavy metals, and
enteric bacteria into south Florida waters poses a major threat to all
water conservation and wildlife protection programs. Careful monitor-
ing of toxic elements in water, sediments, and biota, coupled with
regulations to prevent point source discharge of toxic materials, is
of the utmost importance. Also, programs of need-oriented agricultural
chemical application, for example those currently under development at
the Immokalee Experiment Station, should receive the widest possible
support and implementation.
11. Accelerated transport (as in canals) of toxic heavy metals to
estuarine ecosystems must be carefully controlled. Sheet flow across
mangrove ecosystems appears to provide one efficient downstream filter-
ing mechanism for heavy metal removal, but long-term effects within the
mangrove zone are virtually unknown.
12. Bond issues, cost sharing plans, and aid from other govern-
mental organizations should be used to equalize economic impact of
conservation measures on individual landowners.
13. As with so many noble efforts gathering dust on a shelf, the
lack of mechanisms to implement the above recommendations is a weakness
in south Florida at present. Some state interest has been directed
toward implementation mechanisms, i.e. institutional programs, but more
needs to be done. Federal legislation to protect unique parts of south
Florida also needs to be accelerated because the area is growing rapidly
and it may soon be too late.
II-6
-------
Ill - DISCUSSION
This chapter is devoted to development of an analysis of the Fahkahatchee
regional ecosystem, and a discussion of the consequences of cultural environ-
mental perturbations to that ecosystem. An exposition of the methods of
ecosystem analysis utilized in the study can be found in Odum (1971). A brief
discussion of some of the relevant concepts and the circuit language used in
model development is given in Appendix A. In this chapter, we first present a
set of overall ecosystem models, and describe our general hypotheses on system
functioning that are exemplified by the models. Next, some of our overall
study results are discussed using smaller process subunits of the models as a
device to place each individual study within its larger context. In addition,
we evaluate from the model concepts and accumulated data some longer term
patterns and trends for the Fahkahatchee ecosystem, emphasizing the consequences
of artificial drainage to ecosystem structure and dynamics.
UPLAND ECOSYSTEM
An ecosystem model of the upland terrestrial systems of the study area is
presented in Figure III-l. Sunlight and rainfall are shown as the basic source
Figure III-l. Upland ecosystem model.
III-l
-------
terms of the model. In many parts of the study area overland flow contributed
surface water to particular sites, but during the study year overland flow
into the study area as a whole was nearly negligible. Canals are shown as a
driving force diverting surface water and ground (soil) water to the estuary.
The work gate (Block 1) for canal action is taken here to have an intrinsic
conductivity, even in the absence of canals, that represents overland export
of water across U. S. 41 from the upland to the estuarine salt marshes and
mangrove forests. An export of mineral nutrients from the upland to the
estuary takes place in tandem with the flow of water (Block 2). The export
of minerals is a function of the volume of water flowing to the estuary and
the concentrations of mineral nutrients in the flowing waters.
Periphyton growth, a feature especially of the wet prairies in the study
area, is utilized by a grazing food chain that terminates in the production of
small forage fishes and a carnivore food chain. Periphyton populations require
standing surface water in the prairies in order to develop (Block 3), and their
productivity may be partially governed by the availability of mineral nutrients
(Block 4) for incorporation into photosynthate derived from sunlight. The
forest and prairie ecosystems draw dissolved nutrients and water for evapotrans-
piration primarily from soil water, but the effect of evapotranspiration water
loss can be observed in surface water levels as well (Blocks 5 and 6). These
communities support a grazing food chain that results, inter alia, in an annual
deer harvest by human hunters. Litter produced by forests and prairies fuels
a detrital food chain that, like the periphyton sequence, results in the
production of small forage fishes. The hydroperiod (surface water) controls
the rate of flow of litter into the food chain (Block 7). As is the case with
all the conductivity work gates in this set of models, there is a non-zero base-
line transfer rate that holds during the absence of standing water in the system.
The hydroperiod acts to accelerate this transfer (Block 7) by maintaining a
high moisture content of the litter and by making litter available to aquatic
and semi-aquatic macro-decomposer organisms.
Remineralization of dead plant material is accelerated by the activities
of the detrital food chain, forage fishes, and birds (Blocks 8, 9 and 10). In
the course of its metabolic activity each population accelerates the flow of
dead plant material back to mineral nutrients via decomposition process, which
are then newly re-available to the photosynthetic energy production process.
Once again, these blocks, e.g. Block 10, do not imply that the process ceases
in the absence of wading birds, it merely suggests that remineralization may
be slowed if bird populations are absent. The role of fire is shown by the
switch symbol controlling a flow to the mineral pool from the stock of dead
plant material. Several conditions must be met for this switch to be in a
conducting state. The condition that all input elements must be simultaneously
true to allow flow to occur through the gate is indicated by the logic symbol
"A" (logic AND) within the switch. First, the stock of litter must be above
the threshold value necessary to propagate fire. This condition is shown as
the fuel threshold "Tf" on the switch. Second, a spark must be provided,
either by a lightning strike or from some human source. Finally, the material
must be sufficiently dry to burn. This is shown by a control line from the
surface water pool entering the switch through a logical NOT symbol (small
circle). This simply means that litter in contact with surface water will not
burn. The role of fire could be expanded in the model to reflect, for example,
the burnoff of live prairie grasses, along with ground litter, and their
regrowth from storage reserves in the root system. Prairie fires that only
burn dead plant mass that is still standing upright (and the green shoots), in
III-2
-------
which fires may occur when there is some standing water in the prairies, could
also be shown. However, the simplified portrayal of Figure III-l is adequate
to our purpose, in that it shows the role of fire in a general way and indicates
some factors important in controlling the frequency of fire in the ecosystem.
Finally, the work gate (Block.11) that shows a controlling function of
surface water on the availability of the stock of forage fishes to wading birdsi
represents concentration of the fishes in pools and depressions as water levels
fall. It differs from the other blocks of this figure in that fish availability
increases as water level decreases. This feature is suggested by the division
(T ) sign placed next to the gate.
ESTUARINE SYSTEMS
The model of Figure III-2 is divided into three major blocks of land area.
On the left, the upland ecosystems are shown simply as a source of fresh water
and dissolved nutrients flowing to the estuary as in Figure III-l. The bottom
central portion of the figure represents mainland mangrove forests and salt
marshes, and the upper right-hand section of the figure shows the estuarine bay
UPLAND
ESTUARINE BAYS & ISLANDS
EVAPORATION
MAINLAND MANGROVE & SALT MARSH
Figure III-2. Estuarine ecosystem model.
III-3
-------
ecosystem and its fringing mangroves and island mangrove forests. The Gulf of
Mexico appears on the far right of the model as the ultimate sink for fresh
water, nutrients, and detritus in the estuary.
As in Figure III-l, the flow of water from the uplands entrains a flow of
nutrients in proportion to the concentrations of nutrients in the water (Figure
III-2, process Block 1). Under sheet flow conditions, these waters and minerals
enter the mainland mangrove and salt marsh zone, where they act as source terms
for evapotranspiration and as a mineral nutrient supply for the photosynthetic
activities of the plant communities. The "fresh water" storages in this model
are intended in the sense of a dilution of sea water, that is, if pure sea water
is present in the marsh it would be shown as zero fresh water, and conversely.
Marsh fresh water and nutrients are exchanged with the estuarine bay by
the action of the tides (Block 2)0 The net exchange, as shown by the use of a
two-way gate (Figure i - Appendix A), will depend on the relative salinities
and mineral concentrations between the two water bodies and the volume of tidal
exchange. Under drained conditions, the canal network acts as a switch (Block
3) that bypasses the mainland mangrove and salt marsh ecosystems and delivers
fresh water and nutrients directly to the bays. In this case, delivery of
upland fresh water and nutrients to mainland systems will only occur by tidal
exchange from the bays, and much of it may be lost by tidal exchange with the
Gulf of Mexico (Block 4). The extent of any loss of this kind would depend on
the flushing rate of the estuarine bays. The mainland zones would become
proportionally more dependent on rainfall inputs for their fresh water supply.
Within the mainland mangrove zone, plant production ultimately results in
a production of detritus that fuels a detrital food chain. This food chain is
similar in broad outline to the detrital food chain of the upland zone, although
the organisms involved may be quite different species. The basic sequence is
again a function of scavenger and decomposer action. An acceleration of the
nutrient regeneration sequence is shown, in the same way as in Figure III-l, by
a sequence of process blocks subtended to the populations of salt marsh and
mangrove animal residents (Blocks 5, 6, 7, and 8 of Figure III-2). Fresh
water in the marshes again controls the rate of incorporation of detritus
into the food chain (Block 9) in the sense described by Heald (1971). Break-
down of detrital materials was shown by this author to proceed most rapidly
under brackish water conditions, and was slowed in both pure fresh water and
full strength sea water. Detrital materials not consumed directly on the marsh
are exchanged with estuarine bay systems via the tidal exchange gate shown as
Block 10. Juveniles of many species ("Juvenile Sport Fishes" population) may
develop within these mainland salt ecosystems where brackish water conditions
protect them from predation by piscivorous adults of the same and other species.
Eventually they move to the bays, shown here as the transfer pathway marked
"recruitment" in Figure III-2.
Within the estuarine bay and island mangrove ecosystems two distinct plant
communities produce organic fuels for the ecosystem. Solar energy fixation by
the mangrove fringe and island systems may be gated by the availability of water
for evapotranspiration and by the mineral nutrient concentration of bay waters
washing this zone (Blocks 11 and 12).
Productivity of the marine "grasses" and algal communities of the bays
themselves is controlled by available nutrients in the bay (Block 13) and the
turbidity of the bay waters (Block 14). The extinction coefficient of the
III-4
-------
water governs the transmission of light to bay bottom communities by the well
known exponential relationship
I = Ioe~kz, (Eq. HI-1)
where if I is taken as visible light input at the water surface, z as the
depth of the water column, and k as the extinction coefficient; then I is the
light reaching the bottom community. Turbid waters are produced within these
shallow bays by the action of wind induced stirring of loose sediments from the
bottom into the water column. The transfer of momentum from moving air to the
sediment surface is less efficient as the depth of the water body that must first
be set in motion increases. Thus, the turbidity of a given bay is an independent
function of the average depth of the bay, assuming the inputs of stirring energy
from wind to any two bays is the same. In this sense, the light intensity
reaching bay-bottom communities is a "double" function of the depth of the bay.
First, depth determines the efficiency of the wind in producing turbid water.
Second, light transmission to the bottom is then a function of turbidity
(extinction coefficient) and the path length of incident light energy given by
the depth of the bay as in Beer's Law (Equation III-l above). This situation
is summarized by the source term to Block 14 marked "wind and water depth",
and the pathway notation "e~^z".
Both the bay bottom plant communities and the associated mangrove communities
contribute to the detritus pool in the bays. Paralleling the situation in the
upland mangrove zone, energy flow into the detrital food chain may be salinity
dependent (control line from "Bay Fresh Water" at Block 15). The remineralization
sequence proceeds as before through the decomposition sequence of Blocks 16, 17,
18, and 19, resulting in release of minerals bound in detritus for re-incorpora-
tion into new plant tissue. Once again, we caution the reader that each block
represents an acceleration factor in the remineralization sequence, so loss of
a block does not halt, but does slow, the total process.
The bird populations that appear at the top of both the bay and mainland
food chains are in many respects the same population. They have been separated
in this figure mostly for the sake of clarity in the figure. These bird
populations may act to transfer minerals from bays back to terrestrial mainland
communities. This factor could be of some importance to counter-current inter-
system mineral transfers. A possible magnitude for this process is suggested
by the observation that bird rookeries can produce sufficient quantities of
phosphorus in their excreta to result in phosphate rock concretions in nearby
waters (Lund, 1958). However, as we were not able to explicitly assess this
factor in our field investigations, we have omitted it from the present model.
Finally, the sports fish populations of the bay are harvested by an
extensive guide and tourist industry within the region. This industry is
portrayed by the economic transactor module at Block 20. The economic importance
and estuarine dependence of the penaeid shrimp populations and commercial
fisheries are not shown for two reasons. One, as with the birds, we did not
explicitly assess the commercial fisheries in our field work. Secondly, the
sports fishery will, we hope, serve as an adequate descriptor of mans dependence
on estuarine ecosystems for part of his economic well being.
Several major questions are immediately suggested by these models. For
example, in the model of Figure III-l, to what extent do canals remove surface
and soil water? How does this modify plant productivity, litter fall, and
III-5
-------
litter remineralization rates? What is the effect on the food chains that
ultimately support south Florida's populations of spectacular wading birds?
Within the estuarine system, do canal inputs substantially modify salinity
and dissolved nutrient regimes? To what extent are minerals merely exported to
the Gulf from these bays, and to what extent are they infused into mainland
marsh and mangrove ecosystems by tidal exchange? What are the relative
magnitudes of detrital inputs to bay systems from the resident plant populations
and from the mangrove communities?
In the next section of the report we summarize our findings directed at
providing magnitudes for some of the flow patterns shown in the models, as
steps toward answering the above questions.
ANALYTICAL RESULTS
Rainfall and sunlight act as the major forcing functions or input source
terms of Figure III-l. The U. S. Weather Bureau has maintained a station at
Everglades City in our study area for about the last 40 years. The month-by-
month long term average rainfall and temperature data from this station can be
seen in Figure III-3a, presented according to the scheme devised by Koppen (1931)
and used by Walter and Lieth in their worldwide climate atlas (1960-1967).
The monthly mean patterns for the Fahkahatchee Strand in 1972 are given in Figure
III-3b. These means are based on a network of four recording rain gages plus two
complete weather stations operated in the Strand during the study year, plus
records from Everglades City and a State Forest Service installation at
Copeland (Figure VI-1). Scales for both diagrams, for precipitation, temperature,
and time, are shown in Figure III-3b. This presentation scheme portrays the
general interaction of rainfall and temperature as it affects moisture avail-
ability to plant communities„ In those months in which the precipitation
curve dips below the temperature curve, the stippled area serves as a measure
of the intensity and duration of moisture stress to the plant community. The
hatched areas cover humid periods, and precipitation values greater than 110 mm
are printed in a scale of 1:10 against temperature and marked in black. Rain-
fall during the study year was slightly above the long term mean (1,440 mm vs
1,371 mm), with notable upward departures in February, March, and November,
and an unusually dry October. Another noticeable departure from the long-term
can be seen in the unusually high mean daily temperatures that prevailed in
early 1972.
Although these graphs indicate in a general way the climatic regime of the
Strand, the detailed relationship between water import in rainfall and export
via evaporative processes depends on sun, wind, and the saturation deficit of
the air. This relationship is shown in Figure III-4. The sun provides the
basic driving force to the process, vaporizing water at a conversion proportion
of about 590 calories per gram of water (Block 1). The evaporation process is
conditioned by the action of wind in removing water vapor, and by the relative
humidity or saturation deficit of the air mass (Block 2). Saturation deficit
expresses the ability of the air mass to absorb water vapor from free water
surfaces, or from the plant water release process.
Insolation during the study year is given in Figure EI-S. This curve
represents monthly mean daily solar energy input as measured by a pyrheliometer
at one of our weather stations. The high values for early 1972 appear to
account for the unusually high temperatures seen in Figure III-3b. Mean
III-6
-------
windspeeds and saturation
deficit are shown in Figure
III-6. The windspeed data
were provided by B. J. Yokel
and the Rookery Bay Project.
Saturation deficit in milli-
bars is the average for our
upland and lowland weather
stations. Daily air mass
characteristics were cal-
culated using noon tempera-
tures and relative humidity
from the two stations. We
used a modification of
Penman's method of calcula-
ting potential evapotrans-
piration from these data
(further discussed in
Chapter VI), and compared
this result to the average
evaporation from open water
surfaces. Free water
evaporation was measured by
two evaporation pans equipped
with recorders, one at each
weather station. In Figure
III-7, the solid line is
theoretically calculated
evaporation and the circles
are mean monthly evaporation
as an average of the two pans.
Least squares regression of
these two data sets yielded a
coefficient of correlation (r)
of 0..94 and the equation (in
mm H20)
Pan Evap. =1.03 (Potential
E.-T.) - 11.24, with PO.001.
E
£
37.8
33.3
12.8
O -2.8
I-
\-
5: 300
O 200
EVERGLADES CITY(lm) 23.9°
(32-41) 1371
Figure III-3a
UJ
cr
Q.
100
80
60
40
20
FAHKAHATCHEE
STRAND,
1972
I i i i i
30
20
10
JFMAMJJASOND
Climatic diagrams.
3b., for 1972.
Figure III-3b.
3a., long term average.
Pan evaporation and the
potential water loss calcu-
lated from climatological
parameters are thus in
reasonably good agreement. This information could be used to estimate
evaporative losses from estuarine bays, canals,. ponds, and other open water
surfaces in the study area. In the presence of extensive vegetation cover,
however, this pattern can be substantially modified.
Continuous recordings from ground water observation wells can be used to
estimate actual evapotranspirative water export. Detailed results for the
study area are supplied in Chapter VI, we will outline only the logic of the
calculation and the results in this chapter. Figure III-8 is a schematic
representation of the evapotranspiration process set in the overall hydrologic
dynamics of these swamp ecosystems. Overland surface water inputs to the
III-7
-------
POTENTIAL
-^EVAPORATIVE
LOSSES
Figure III-4. Model of evaporation process.
Q
CO
UJ
CD
<
500
400
300
- INSOLATION
JFMAMJ JASONO
strand are omitted since
they were negligible during
the study year. In general,
rainfall within the study
area enters a surface water
compartment that may infil-
trate into the soil or may
be exported in surface water
flow patterns. Soil water
may drain downward to the
water table as ground water
recharge, and then be
exported as subsurface flow.
The driving forces for
evapotranspiration exports
are sun energy, wind, and
saturation deficit as
already discussed in
connection with Figure
III-4. Water use by the
plant community can
appear as losses from
any of the three water
storage compartments, and
the sum of these water flows
represents total actual
evapotranspiration. Each
requires a separate
calculation. Diel changes
in the slope of ground
water recession curves
provide the basic datum,
and, when the water table
surface is above ground
level yield evapotrans-
piration discharge
directly in millimeters
of water (Block 1). When
the water table is below
ground level, apparent
evapotranspiration must be
corrected by the proportion
of the soil volume that is
gravity drainable water
(Block 3) ("specific yield",
Sy). We estimated Sy = 0.15
for the soils around our
wells based on coupled major
rainfall and ground water
rise observations (Chapter Figure III-6. Windspeed and saturation deficit.
VI). Finally, the soil
layer between the water table and the ground surface retains much water (the
"field capacity of the soil") that is available for plant transpiration (Block
2). This term must be measured somewhat indirectly. To obtain a maximum
Figure III-5. Mean daily insolation, 1972.
.Q
E
O
LJ ~
QQ
00.
20
15
10
CO
JFMAMJJASONO
III-8
-------
estimate of this factor, we assumed
that infiltration of surface water
would be perfect when the water
table was below the surface, i.e.,
that local surface runoff would be
negligible. If the soil water
compartment has been drawn below
field capacity at the time of a
given rainfall input, this
deficit must be made up before
much water can appear as a
ground water table rise.
Therefore, when the water table
is below ground, soil water
evapotranspiration can be no
more than the difference
between rainfall input and the
observed ground water rise as
a result of that rainfall,
corrected by the specific
yield. That is,
E.-T. (Field capacity) = rain -
Sy (water table rise)
(Eq. III-2)
when the analysis is restricted
to periods when the water table
is below the ground surface
and the water table rise is
directly attributable to paired
rainfall records.
O
CM
E
200
Ld
150
100
o:
o
Q.
S
LU
J — ' — I — J — I — I — ' '
' '
JFMAMJJASOND
Figure III-7. Pan evaporation and
potential evaporation, 1972.
The resulting curve for
actual evapotranspiration (mean of four wells) appears in Figure III-9. The
peaks in March, June, and October may represent periods when rainfall restored
soil field capacity that had been depleted in previous periods of water
scarcity, rather than true evapotranspiration peaks. The total evapotrans-
piration (1,100 mm) compares favorably with theoretical totals accumulated
using methods proposed by Thornthwaite (1,179 mm) and Stewart (1,117 mm).
This 1,100 mm of actual evapotranspiration represents 76 percent of total
rainfall (1,440 mm). With this information, coupled with the data of Chapter
VII (Hydrology), we can attempt to construct a complete water budget for the
Strand during 1972 (Table III-l).
The total upland area of the Fahkahatchee study site was 28,961.2 ha.
Using this figure, we convert mm of water in rainfall and evapotranspiration
to cubic meters of water in the watershed. Subsurface export is the total
ground water flow crossing U. S. 41 from the upland study zone, and the over-
land flow source term is a maximum estimator of water that entered the Strand
across Alligator Alley (17 percent of flow crossing Janes Scenic Drive). To
arrive at total surface water export from the Strand, 33 percent was added to
the surface flow crossing Janes Scenic Drive to account for flow increments
in the lower third of the strand as flowing waters approach U. S. 41. Yearly
III-9
-------
SURFACE
WATER
FLOW
INFILTRATION
GROUND WATER
RECHARGE
SOIL
WATER
E.T.
(FIELD
CAP)
GROUND
^WATER
EXPORT
canal increments/mile were cal-
culated using the regression
equation of Figure VTI-25 and
the monthly mean stage at EPA
Well W-2, corrected for effects
of construction (3.095 X 106
m /mile • yr). We then assume
that the total of 22 miles of
canal bordering the Strand
(Barren River Canal on the
east and GAG Properties canals
on the west) would act in this
fashion, producing the figure
in the table. This assumption
is, of course, badly flawed.
Canal weirs and changes in
head differences would reduce
ground water removal below
the tabulated value. However,
surface water flow into the
canals is not included in the
estimate, and would tend to
cancel decreases from the
listed value or perhaps even
produce an increase.
In any case, the balance
of sources and sinks is
reasonably good (sinks account
for 98 percent of source water),
and is supported by the lack of
any discernable major shift in
overall water table levels
between 1972 and 1973. In turn,
the establishment of an overall
balance without recourse to
difference terms supports the
approximate accuracy of the
evapotranspiration figure.
In the south Florida ecosystem,
although some major plant species
(e.g., cypress trees) drop their
leaves and remain dormant through-
out the winter season, others
(e.g., maples and willows) appear
to be active throughout the year.
To arrive at a yearly pattern of
actual evapotranspiration then,
correcting for the field
capacity water use that only
becomes manifest under rainy
conditions, we modified pan evaporation (PE)rates to give the same total export
as true evapotranspiration. Pan evaporation measures the driving forces
TOTAL
EVAPOTRANSPIRATION
Figure III-8. Hydrologic model for
upland strand.
zoo
I- ^ 150
O
Q.
100
50
J J
1972
Figure III-9. Test well evapotrans-
piration.
III-10
-------
Table III-l. Water budget for upland Strand, 1972.
Sources
108m3
mm HoO
Sinks
108m3
mm H£O
Rainfall
Overland flow
4.160
.034
4.194
1,440
12
1,452
Evapotranspiration
Subsurface export
Janes Drive surface
flow
+ 1/3
Canal flow
3.160
.011
.201
.067
.681
4.120
1,100
. 4
69
23
235
1,431
300
available to this form of water export, and we adjust the total to give 1,100 mm
export or 65.4 percent of total yearly pan evaporation (PE). This result is
given in Figure 111-10, along with the rainfall delivery rates for each month
of the year. In this figure, the
solid areas represent excess water
that is available for ground water
and field capacity recharge, the
hatched areas show periods when
plant community demand exceeds
the monthly supply. The presenta-
tion in this figure is similar to
that of the climate diagrams of
Figure III-3, except the total in
the present figure represents
estimated actual evapotranspiration
rather than a derived estimate based
on temperature. 100
CE
UJ
E
E
200
J F M A M
J J A
1972
S 0 N D
Figure III-10. Rainfall and actual
evapotranspiration.
This figure can be compared to
Figure III-11, which shows the
seasonal pattern of the ground
water table referred to ground
» surface elevation as a base line.
The seasonal patterns in the wells
generally agree with the relation-
ship shown in Figure III-5. The
early spring drawdown corresponds
to the period of heavy net water
export shown for the same period
in Figure III-10. Recharge peaks (Figure III-11) in February, June, August,
and November seem to correlate nicely with water availability peaks in those
same months in Figure 111-10. The discrepancy in magnitude correlation for
February probably represents levels of plant activity below the mean of 65
percent of PE.
Figure III-11 also serves as a comparison of ground water dynamics
between sites that were near canals and sites remote from canals. The upper
III-11
-------
Figure III-ll. Water table depths, test wells.
curve is the stage record from well W-5, a wet prairie area near Remuda Ranch,
1.8 km from the nearest canal. The lower curve represents the trace at well
W-3, located some 0.3 km from a canal. The portions of this (W-3) record that
were strongly affected by construction activities in the nearby canal have been
omitted. Polar planimetry of this figure for the period up to the recorder
failure in late August yielded an average difference between sites of 42 cm.
Thus, the approximate average effect of drainage canals during the study year
was to maintain the water table at a depth 42 cm below normal at sites near
canals.
Our paired sites for observation of plant community metabolism were located
at similar distances vis-a-vis canal locations. In early October we excavated
to the water table at both sites to check the accuracy of these well records as
indicators of water table levels at the metabolism sites. Our findings are
shown as the encircled Xs in Figure III-ll, and appear to at least partially
validate this assumption.
The role of these hydrologic factors within the swamp forest ecosystem is
shown in Figure 111-12. The availability of water to the plant community
controls the facility with which the system can convert sunlight to organic
energy (Block 1). In this sense the canal-driven water export process (Block 2)
competes with the plant community for water resources. Within the forest, the
rate of litter-fall may be related to net productivity of the plant community.
The forest floor litter is remineralized with an assist from a detrital food
chain (Block 3). The overall dynamics of the remineralization process may be
governed by moisture conditions in the forest (Block 4). Finally, minerals
enter the plant transpiration stream and are incorporated into new protoplasm
111-12
-------
(Block 5).
Annual net above-
ground productivity
increases in linear pro-
portion to yearly actual
evapotranspiration (AET).
Figure III-13, redrawn
from Rosenzweig (1968)
and data of Odum (1970) ,
sets forth this relation-
ship for the broad range
of terrestrial ecosystems.
Arctic tundra, deserts,
temperate forests, and
tropical rain forests are
all represented in this
graph. One of our major
objectives was to determine
the extent to which this
relationship holds true
for the detailed operation
of ecosystems within a given
climatic zone, and the
modifications that might be
induced by artificial
drainage.
Yearly net production
is the sum of several
factors. A large part of
the total energy fixation
in an ecosystem is utilized
directly by the autotrophic
community for maintenance
activities. The remainder,
net primary production,
may appear as a net change
in standing crop, as
removals by grazing food
chains, and as litter-fall
or detritus production
(Figure 111-12). The
effect of a 42 cm water table
lowering on the net produc-
tion of cypress strands is
summarized in Table III-2
(details in Chapter XII).
Figure III-12. Swamp forest dynamics.
3000
o
a
o _
cc k.
?
UJ
O
CD
2000
IOOO
Y = 2.290X - 366.8
r = 0.96
t = 1.64 on 25 d.f.
P < 0.0001
500 IOOO
ACTUAL EVAPOTRANSPIRATION
(mm/yr)
1500
Figure III-]0.. Productivity and actual evapo-
tr-inspi..ation, worldwide.
The biomass production
is broken down in Table
III-2 by woody plant species and herbaceous ground cover plants. All three
components of net production show a decrease in cross-site comparisons. The
herb layer shows an absolute decline in the drained site, suggesting that
ultimate loss of this layer of the community may occur. The total biomass in
111-13
-------
the natural strand was twice that in the drained strand, but this factor fails
to account for the difference in woody plant growth, as the undrained site
shows a growth four times that on the drained site.
Table III-2. Cypress strand net production, 1972, for drained and undrained
sites. (All values gm dry weight/m /year)
Component Near Canal Undisturbed
Litterfall
Woody Plant Growth
Understory Herbaceous
Plants
267.2
119.6
-20.0
366.8
373.2
485.0
311.6
1169.8
Mean Standing Crops
Woody Plants
Herbs
Total
(Kg dry weight/m^)
8.9
1.2
10.1
17.1
1.9
19.0
Interestingly, the relative litter-fall is 45 percent greater in the
drained site, based on the ratio of leaf litter-fall to mean standing crop of
leaf material (turnover rates of 0.713 vs 1.035). This suggests that the
response of this system to moisture stress may involve a process of thinning
in canopy density. This would reduce moisture loss by reducing the size of
the available water-exporting surface. If this is the case, two consequences
are immediately apparent. First, decreases in system leaf tissue will entrain
decreases in productivity. That this decrease in fact takes place is obvious
from the table. Second, canopy thinning permits more sunlight to reach the
forest floor, drying accumulated litter debris and making the system more
vulnerable to fire damage.
We have postulated a relationship between the cycling rate of mineral
nutrients and moisture conditions in the forest (Figures III-ll and 12). This
hypothesis is explicitly presented in the model of Figure 111-14. Litter
materials reaching the forest floor ("Duff", Q) have an energy content that
fuels a detrital food chain. In the process of utilizing this energy for
growth and reproduction, much of the original plant carbon is released as CC^
gas. However, other materials, for example phosphorus and nitrogen, are
retained within the system. This results in an increase in the relative ash
content of the remaining litter, which can be expressed as an increase in
111-14
-------
mineral nutrient concentration (Block 1). The build-up of nutrient concentra-
tion during the soil formation process makes these materials more available for
recycling (Block 2) to the plant community. The role of moisture in this
system is shown at Block 3. Water accelerates the process by producing
conditions favorable to small organisms (macro-decomposers) that break down the
litter to fine particles. The fine particles are further acted on by bacteria
and fungi that are also favored by moist conditions.
MINERAL
NUTRIENT
CONCENTRATION
LANT \ 'I DUFF
COMMUNITY AJTTERV Q
FALL
Figure 111-14. Model of litter remineralization sequences.
The breakdown rate of a unit parcel of litter materials can be described
by the equation
d£ = -kQ (Eq. III-3)
dt
where Q represents the standing crop of litter and k is in units of days" .
The parameter "k" is the turnover rate of the material, and is a function of
the activity levels of the decomposer populations„
We set out parcels of natural litter exposed to the range of relatively
moist and dry conditions found in the forest, and then studied the disappearance
rates of the parcels. In addition, the mesh bags containing the litter parcels
were of two sizes. This allowed assessment of the relative role of macro-
decomposers, as accelerators of decomposition, under the paired moisture
conditions. The turnover rates for leaf litter derived from this experiment by
least squares regression are presented in Table III-3. All of these values are
statistically significant at the 1 percent level.
There is a depression of turnover rate for both series in comparing moist
to dry conditions. Under moist conditions, the macro-decomposer faunal biomass
that became established within the litter bags was larger in the coarse mesh
bags, thus accounting for the increased turnover rate. The amphipods, limpets,
etc. that appeared in the coarse mesh bags thus accelerated the litter
decomposition rate by a factor of 1.6 times. The lack of a macro-decomposer
acceleration under dry conditions is something of an anomaly. We might hypothesize
III-15
-------
that fine mesh bags under dry conditions retained rain water more effectively,
thus inducing a moisture gradient within the dry condition experimental series,
Or, the fine mesh bags might have excluded predators that feed on the small
terrestrial insects we found associated with both dry condition series.
However, we lack sufficient hard data to critically evaluate either hypothesis,
Table III-3. Leaf litter turnover rates. (Based on gm dry wt of materials.)
Macro-decomposer access
Moist
Dry
Unrestricted
Restricted
2.686
1.723
1.049
1.334
In summary, we can conclude that moist conditions accelerate litter
remineralization in this forest by at least a factor of 1.3, and create
favorable conditions for macro-decomposers that further accelerate the process
by 1.6 times.
If mineral nutrients are retained by the system while carbon is being
"blown off" to the atmosphere, concentrations of nutrients will increase,
making these materials relatively more available to the autotrophic layer of
the community. Figure 111-15 shows the concentration of total organic .nitrogen
(Kjeldahl method), determined for litter bags at the time of collection. The
steady increase over time is presumably a consequence of the preferential
release of carbon from the Iitter0 The ratio of slopes in Figure 111-15 is
0.59 and the ratio of the
mean turnover rates from
Table III-3 is 0.54. The
more rapid turnover under
mois.t conditions thus
25
accounts for most of the
difference.
2
20
o>
<
In our studies of
net production in prairie
ecosystems, most of the
detail in our results was
irretrievably blurred by
wildfires that at various
times burned nearly half
the sites. In addition,
although preliminary
observations of flooded
prairies in late 1971
indicated that periphytic
algal growth was an
important part of overall
community metabolism, the
prairies failed to flood in 1972.
10
EXPLANATION
WET A
DRY O
N = 0 022) -I- 14.6
P< 0.001
N • O.OI3t + 10.9
P
-------
normal level of activity of this component of the ecosystem, although we can of
course state that in the absence of any appreciable hydroperiod there is,
unsurprisingly, no periphyton growth.
Figure 111-16 illustrates the complexity of the problem of estimating net
production in prairie grasses from sequential data on standing crop. Mineral
nutrients and sunlight combine (Block 1) to fix organic materials as gross
primary productivity (GPP) in green leaves. Some of this material is respired
directly (R^), some is translocated to the root system, and subsequently
respired (It,), and some may appear as an increase in standing crop of live
Figure 111-16. Model of wet prairie production.
leaves. In any given interval, some leaves die and are replaced by new shoots.
Leaf turnover will appear as increases in the "duff" and standing dead leaf
compartment, but at the same time dead materials are being broken down by the
detrital food chain and returned to the available mineral pool. When a fire
sweeps through the area, if dry litter stocks are sufficiently dense (Threshold
T), duff and green leaves both burn off and instantaneously build up a mineral
supply in the residual ash (Block 2). Stored root stocks may then be utilized
to rapidly regenerate the green leaf compartment, and the cycle resumes. The
seasonal cycle of yearly die-back of the grasses and spring restoration from
root storages adds still another level of complexity to the problem.
As a crude minimal estimate of prairie production, Table III-4 presents
the differences between yearly maxima and minima in each of these compartments.
Most values in Table III-4 represent the differences between means of four
determinations. Although detailed comparisons are not possible, note that
production values approach those listed above for cypress strands. The frequent
III-17
-------
Table III-4. Summary of biomass accumulation in wet prairie ecosystems based
on peak-to-peak harvest values. (gm ash free dry wt/m2)
Soil type
o
Ochopee marl,
Deep phase
Ochopee marl,
Deep phase
Ochopee -"marl ,
Deep phase
Ochopee marl
•L
Ochopee -marl
Leaves
206.6
195.9
234.6
458.4
229.3
Duff
84.8
189.9
46.0
104.4
139.8
Above ground
Sub-total
291.4
385.5
280.6
562.8
369.1
Roots
196.6
192.6
155.3
442.5
636.1
Total
488.0
578.4
435.9
1,005.3
1,005.2
a Canal influenced site
b Fire recovery values over 4-5 month period during summer.
fires in these prairie sites made possible an estimate of the fuel threshold
for propagation of fire (Figure 111-16, Block T). "Duff", as used here,
includes both loose ground litter and upright dead grass stalks, as grams ash-
free dry weight/m . Based on observations of prairie areas that failed to sustain
fire vs those that burned over shortly after biomass sampling, we find that
40
-------
where NPP = net primary production during daylight hours, I = total incident
solar radiation, ET = evapotranspiration, P is pan evaporation, n = number of
replicates and the subscript "o" indicates observed values. Pan evaporation
rates for the coastal weather station were used for correction factors as
this station location was much closer to the prairie sites. Further correction
of these estimates for standing crop variation had virtually no effect on the
resultant numerical values, and is therefore not reflected in the following
table (Table III-5).
Table III-5. Summary of wet prairie carbon and water metabolism. Standing
crop estimates are in grams ash-free dry weight/m2. Metabolic values are
in gm carbon/nT • day or
mm
Chamber
Metabolism biomass
Soil
f\
Ochopee marl,
deep phase
Ochopee marl,
deep phase
Ochopee marl,
deep phase
Ochopee marl,a
deep phase
Ochopee marl
Month n
June 2
Aug. 3
Aug. 1
Oct. 4
Aug. 4
GPP NPP R ET Leaf
10.92 0.98 9.94 3.5 153
6.01 2.84 3.17 3.9 267
14.50 4.61 9.89 4.1 216
4.60 1.94 2.66 2.6 231
15.28 4.39 10.89 4.35 298
Duff
369
621
40
128
176
a Canal affected
b Burned in 1972.
The metabolic values shown (Table III-5) are for the entire 24-hour
activity cycle. The basic data derived from the diel studies are in terms of
production during daylight hours and nighttime respiration. The NPP shown in
this table represents the balance between these two phases of the cycle. Total
respiration was calculated by assuming the mean nighttime rate would hold for
daylight hours as well, and GPP, gross primary productivity, is then the sum of
NPP and R. Net primary productivity shows an apparent relationship to either
seasonal effects, or drainage effects, within the data of the table. However,
the lack of synoptic measurements fully confounds the validity of direct
conclusions from the tabulated information. The evapotranspiration rates are
68, 82, and 59 percent of pan evaporative rates at the coastal station for June,
August, and October, respectively. This data supports the mean value of 65
percent derived from test well diel patterns. In addition, it suggests a
seasonal plant activity cycle, at least for the wet prairies, beyond the
reduction in functioning caused by low potential evapotranspiration.
Ill-19
-------
Y = 0962X-
P< 0.001
139
O
CM
a.
O CD
3 OC
Q <
0 °
a:
Q- E
o>
I- ^-
UJ
o
Evapotranspiration measures the simultaneous availability of sunlight and
water, which perhaps are the two most important factors controlling the produc-
tivity of terrestrial ecosystems (Rosenzweig, 1968; Figure 111-13). The
measured evapotranspiration and 24-hour net production values from eleven
complete diel cycles is shown in Figure 111-17. The linear relationship
between these variables is statistically significant at the 0.1% level, and
gives a conversion ratio of 0.962 (gm C/m^) per mm of water export. If we
note that the mean evapotranspiration rate for the entire strand was 3.00 mm/day
(lljlOO mm/hri/366 days in 1972), this relationship estimates yearly net produc-
tion at 548 gm C/m^ year. Conversion of this figure to ash-free dry weight by
the mole fraction of carbon in carbohydrate results in a value of 1.37 kg ash-
free dry weight/in^ year, a figure that is not greatly at odds with the minimal
estimates provided by harvest data (Table III-4).
The relationship of actual
evapotranspiration, as a fraction
of pan evaporation, to the
shallowness of the water table
is given in Figure III-18., In
this figure, the depth of water
(Z) is derived from the well
record pertinent to the date
and site of operation of the
metabolic equipment. Pan
evaporation was taken from
the nearer located weather
station. An exponential
model was fitted to this
data because the work of White
(1932, and Figure V-l) indicates
its suitability. Much of the
scatter in this diagram may be
attributable to utilization by
the plant community of field
capacity water when the water
table is low. Nonetheless,
the fitted intercept appears
accurate (1.007 vs a "true"
value of 1.000), so this equation may represent the average response of the
prairie community to water-table lowering.
This idea can be tested by deriving estimates of annual net primary
production from the posited relationships (Figures III-17 and 18) and comparing
them to the measured response of the cypress strand community to canal proximity.
The mean water table depth at the test well remote from a canal (W-5, Figure
III-11) was 49.8 cm. The estimated AET (actual evapotranspiration) would be
2468
EVAPOTRANSPIRATION
(mm H20/DAY)
Figure 111-17. Productivity and actual
evapotranspiration, wet prairie.
10
(AET) = (PE) (1.007 exp (-0.0067
49.8))
or 1,205 mm. This is in satisfactory agreement with the figure of 1,100 mm
from well data, and predicts (from Figure 111-17) an average net productivity
of 1.78 gm C/m day, or an annual net production of 651 gm C/nr. The mean
water table depth at well W-3 (0.3 km from a canal) was 90.8 cm. This figure
leads to an AET of 922 mm and an annual net production of 378 gm C/m^. The
latter figure gives a 42 percent drop in net productivity due to the 42 cm
111-20
-------
10
0.8
0.6
UJ
< 0.4
0.2
0
Y= l.007e-°0067E
P < 0.10
o
40
so
80
100
Figure 111-18. Water table depth and
evapotranspiration.
water table lowering induced
in the vicinity of the canal.
To compare this figure to the
data for cypress strands, the
latter must be corrected for
the difference in biomass
between the two sites (Table
III-2). The equivalent
expression for this situation
is
1 _ (NPP/SC)d
(NPP/SC)w
DEPTH TO WATER TABLE, 2 (cm)
where NPP is annual net
production, SC is standing
crop biomass, and d and w
represent dry and wet sites,
respectively. This expression
yields a value of 41 percent for the measured drop in annual production between
sites vs the 42 percent reduction for the theoretical calculation from data on
net productivity, actual evapotranspiration, and water table dynamics.
We can use the coupled equations of Figures 111-17 and 18 to simulate the
yearly cycle of production dynamics in the Fahkahatchee grassland ecosystem.
Figure 111-19 illustrates the (monthly mean) water table dynamics observed
during 1972„ The lower curve
represents conditions 0.3 km
(0.2 miles) from a canal, the
middle curve the more normal
dynamics at well W-5 (c.f. ~ o
Figure III-ll). The upper
curve shows a hypothetical
pattern, using the curve for
W-5 as a base, during a year
with a maximum water depth
of 13 cm above ground level
in the prairies. These
curves, in conjunction with
the mean monthly pan evap-
oration rates (Figure III-7)
were used to generate the
net primary productivity
curves of Figure 111-20.
The correspondence of NPP
with water table depth is
apparent. The upper curve,
which shows NPP for a year nearer the long-term average, results in a net annual
production of 931.2 gm C/m2. Using the mole fraction of carbon in carbohydrate
and the mean value of 5 percent ash content for prairie leaf samples collected
during the year (5.4 + 0.4, x + 1 SE), we can equate this figure to 2.45 dry Kg/m2
annual production. This value can be compared to that predicted by the world
wide patterns of Figure 111-13, which is 2.15 kg/m2 . year. The correspondence
is quite good, and we can probably account for the difference by our neglect of
the plant community activity cycle that has suggested itself at several
o
. 20
CO
<
UJ
I
40
60
80
100
120
a.
UJ
o
Figure III-19.
deptho
Monthly mean water table
111-21
-------
Figure 111-20. Simulation of wet prairie
(grass) NPP.
junctures in this chapter. We will continue to neglect it, except to point out
that the difference between late fall and very early spring NPP values in
Figure 111-20 probably represents
the artifact in the simulation
that produces the high estimate.
It should be pointed out,
however, that we have still not
attempted to include periphyton
production in estimates of NPP.
This factor would undoubtedly
elevate the NPP of these
prairies well above the world-
wide mean during years with a
normal hydroperiod.
The relationship of NPP
to water table depth is shown
in Figure 111-21, for months
when pan evaporation (PE,
measuring co-availability of
sun, wind, and saturation
deficit) is at 100 and 200 mm
H20/mo. The decline of NPP
over the 100 cm (3 foot)range
is exponential with depth,
and the percent loss is more
severe the more favorable are
conditions for plant community
production.
The implications of these
facts for wildlife populations
dependent on plant production
for their food resources are
sobering. Dependence of the
plant community on direct
rainfall for its water supply
induces at the very least a
40 percent drop in NPP, which
represents an equivalent drop
in magnitude of the carrying
capacity of the ecosystem for
wildlife. The removal of ponded
waters from the system eliminates periphyton production altogether, which,
through the suppression of forage fish populations, may eventually severely
depress the remaining bird populations dependent upon the resource during their
breeding season (c.f. Kahl, 1964; Leopold, ^t al., 1969, etc.). Some details
of the forage fish populations are reported in Chapter XV.
The comparison of estuarine systems subject to artificial drainage
perturbations is more difficult to assess, as a glance at Figures III-l and 2
will show. The location map of Figure 111-22 may help clarify the situation.
All the water that enters Fahka Union Bay via the canal would have gone to
tidewater through a mangrove zone somewhere in Collier County, but by no means
(mm/mo) PE
WATER TABLE DEPTH, £ (cm)
Figure 111-21. NPP as a function of water
table depth.
111-22
-------
Figure 111-22. Map of estuary study zone.
111-23
-------
all would pass through the Fahkahatchee estuary. To the extent that canals draw
water from the strand, the Fahkahatchee Bay system is deprived of fresh water
flow, some of which appears instead in Fahka Union Bay. From salinity studies
conducted throughout 1972 (Chapter VII), it became evident that fresh waters
entering Fahka Union Bay are partially short-circuited directly to the Gulf,
and to some extent spread laterally into the insular mangrove forest south of
the bays. However, the spread-out of fresh water was never sufficient to inject
fresh water from the canal into the mainland mangrove forest north of Fahkahatchee
Bay via the Fahkahatchee or East River (see isohaline summaries in Chapter VII).
It appears that fresh waters that bypass the mainland mangrove forest, although
they may appear in the waters washing the insular forest, are not back-pumped
into the mainland forest by tidal action to any appreciable extent.
Phytoplankton production seems to play a relatively minor role in these
estuarine bays. The bays are extremely shallow, and turbidity of the overlying
waters is quite high. Resuspension of the soft bottom sediments by wind action
is the major factor producing turbid waters in these bays, and is a function of
the depth of the overlying water column (Chapter VIII). Phytoplankton counts
were similar to those observed in other southeast estuaries (e.g. Lear and Smith,
1972). The mean depth or" Fahkahatchee and Fahka Union Bays is about 1 m (3 feet).
Thomas (1972) estimated phytoplankton NPP (^C method) to average 0.11, 0.77, and
0.54 gm C/m^/day for April, July, and October, respectively, in Port Royal Sound,
an extremely productive estuary in South Carolina. The euphotic zone in Fahka-
hatchee Bay, defined by the depth at which only 1 percent of incident light
remains, would on the average comprise the entire water column (Chapter XII,
mean k = ,0237/cm). Thomas' estimates therefore can be directly transformed to
an areal basis for these shallow bays. These estimates provided by Thomas'
work are in good agreement with similar studies in North Carolina and Georgia
estuaries (O'Hara, 1972). We can, therefore, estimate the annual mean NPP of
phytoplankton in Fahkahatchee Bay at 0.5 gm C/wr • day, accounting for the
higher light intensities in South Florida vs Georgia and the Carolinas.
The productivity of the marine grass and algal beds of the bay (Chapter XII)
averaged 0.376 gm C/m^ • day (GPP) for the year. NPP can be approximated as 50
percent of this figure, or 0.188 gm C/m^ • day (Odum, 1971), over the entire
bay ecosystem. This indicates that the basic food resources produced within
the bay ecosystem for dependent animal populations total about 0.688 gm C/nr/
day, or 1,870 mt/yr over the entire Fahkahatchee Bay ecosystem. Part of this
would enter the animal trophic levels via a grazing food chain based on phyto-
plankton dynamics, and part through detrital food chains based on grass-bed
leaf turnover and mortality in the grazing food chain.
To compare this in situ productivity to the inputs of detritus from the
mangrove forest, we conducted studies of water movement and detrital transport
over complete tidal cycles in the bay (Chapter IX). Heald (1971) found a
maximum detrital export, in a similar ecosystem, in the month of November that
amounted to 30 percent of yearly total export. Our study was conducted in mid-
October, when export in Heald's system was only 4.3 percent of the total; but
we use this 30 percent figure as a minimum estimator for calculating yearly
detritus input to Fahkahatchee Bay. We convert the net transport observed
(Chapter IX) to yearly values by the number of tidal cycles per year, the mole
fraction of carbon in carbohydrate, and Heald's (1971) 30 percent correction.
This process results in the reduction of the total resource base for Fahkahatchee
Bay to common units (Table III-6).
111-24
-------
Table III-6. Yearly resource inputs to Fahkahatchee Bay (metric tons/year)
Source mt/yr
o
Phytoplankton
Benthic plants
In situ subtotal 1,870
Mainland forest 2,100
Insular forest 400
Total resources 4,370
a Maximum estimate
b Minimum estimate
Net transport down the rivers accounts for system inputs derived from the
mainland forest. Net transport through the passes on the south side of the bay
accounts for inputs to the bay from the insular mangrove ecosystem. Bearing in
mind that the in situ productivity is a maximum estimate and the mangrove
estimate is a minimum estimate, we see that at least 57 percent of the total
resource base for the bay was derived from associated mangrove forests. Under
less restrictive assumptions this estimate could easily increase to 80 percent
or more. The contribution of the insular forest is only 16 percent of the total
mangrove forest detrital flow to the bay, even though insular mangrove forests
represent 36 percent of the total mangrove forest found in the study area.
From these considerations we can see that the mainland mangrove ecosystem
represents a key factor in the natural economy of the Fahkahatchee Bay ecosystem.
The fate of this forest as it responds to drainage effects will largely determine
the response of an estuarine bay to the loss of overland fresh water drainage.
Note especially that neither tidal re-injection of fresh waters from below the
forest nor compensating fresh water flow to the island mangrove forest can
affect the outcome. The first does not occur to any appreciable extent, and
the relative contribution of insular forests to the bay does not even match
their relative share of the total forest area.
In order to examine the response of mangroves to differential fresh water
inputs, we established four sites along a fresh-to-saline water gradient. Three
of these sites were in the drainage basin of the Fahka Union River, which enters
the Fahka Union Canal near its mouth (Figure 111-22). The fourth was near the
eastern shore of Fahkahatchee Bay. Community isolates, similar to those of the
prairie community experiments already described, were studied for net and gross
carbon fixation rates (details in Chapter XII). A result of this study is given
in Figure 111-23. This figure represents 14 complete diel cycles in mangrove
(4 species) canopy leaves in full sunlight, normalized, as before, for mean
daily insolation during the course of the complete experiment. The ordinate in
this figure (R24/GPP) is the total daily respiratory metabolism divided by total
daily gross carbon fixation (gm C/m^ leaf tissue • day). It thus represents the
fraction of the energy budget used for direct daily maintenance expressed within
111-25
-------
0 5
03
0 2
O.I
• RHIZOPHORA MANGLE (RED)
O AVICENNIA NITIDA (BLACK)
A LASUNCULARIA RACEMOSA (WHITE)
0) CONOCARPUS ERECTA (BUTTONWOOD)
Y = 0.438*
r =0.93
P<0.00001
- I26X
10
12
(CL-)0
(CL")t
X IOO
Figure 111-23. Response of metabolic budget (sun
leaves) to chloride dynamics.
the plant leaf tissue.
The abscissa is the gradient
in chloride concentration
between the overlying water
mass (IC1"}0 - {ci"3s) and
the soil water solution
(o stands for overlying,
s for soil); as a per-
centage of the chloride
concentration in the
soil water. This data
was determined partially
from diel studies of
soil and surface
salinities at the meta-
bolism sites, and
partially from synoptic
survey data between the
sites. We fitted an
exponential relationship
to this information
because the supposition
that the ratio (R^/GPP)
could exhibit negative values is logically implausible. (Negative values would
imply that the respiration process was fixing, rather than releasing, CC^). The
extrapolation of this relationship much beyond the range of measurement would
also be questionable, for (R24/GPP) probably stabilizes at some positive value
rather than continuing the slow decline of Figure 111-23. Within the measured
range, however, the probability of this result being due to chance alone is less
than 1 in 100,000.
The independent variable of Figure 111-23 is an expression of the inter-
action between tidal dynamics and water quality within the estuarine zone. The
range of possible significant interactions between these factors is rather
large. For example, Bowman (1917) found it impossible to culture red mangrove
seedlings on their natural soils until he adopted the habit of replacing the
water in his culture vessels daily. Under stagnant conditions, the buildup of
sulfide from anaerobic decomposition in the soil was toxic to the plants. This
author also found a non-linear decrease in transpiration rate with increasing
salinity of the plants' water supply. More recently Scholander, e_t al. (1955)
demonstrated that tidal action serves a physical role in pumping oxygen from
the atmosphere to the root systems, in both red and black mangroves. Lugo et al.
(in press) have suggested that macro-nutrient availability can strongly affect
the vigor of individual mangrove stands.
We might summarize some of the potential ecological factors affecting
mangrove ecosystem dynamics as including:
1. Tidal factors.
a. Transport of oxygen to the root system.
b. Physical exchange of the soil water solution with the overlying
water mass,, removing toxic sulfides and reducing the total salt
content of the soil water.
111-26
-------
c. Tidal flushing interacts with the surface water particulate load
to determine the rate of sediment deposition or erosion within a
given stand.
d. Vertical motion of the ground water table may transport nutrients
regenerated by detrital food chains into the root zone of the
mangroves.
2. Water chemistry factors.
a. Total salt content governs the osmotic pressure gradient between
the soil solution and the plant vascular system, thus affecting
the transpiration rate of the leaves.
b. A high macro-nutrient content of the soil solution has been
suggested (Kuenzler, 1969) as enabling the maintenance of high
productivity in mangrove ecosystems despite the low transpiration
rates caused by high salt concentrations in sea water.
c. Lugo, jBt: a.1. (in press) indicate that allocthonous matro-nutrients
contained in wet season surface runoff may dominate the macro-
nutrient budgets of mangrove ecosystems.
While this summary is a considerable simplication of the ecological
significance of water quality and tidal motion to the mangrove ecosystem, it
serves to indicate the great potential shifts in mangrove ecosystem dynamics
that could be induced by alteration of the natural hydroperiod.
In any case, the gradient of chloride concentration across the soil inter-
face, expressed as a ratio to the soil water chloride concentration, may serve
as an integrated indicator of the outcome of this complex dynamic process at
any particular point in the estuary. Our observed data on this variable were
used as the abscissa for Figure 111-23.
Some effects of tidal motion in producing vertical motion in the soil wata:
are discussed in Chapter VII. Vertical transport and tidal flushing would both
decline exponentially along a transect from the bay front to the salt marsh and
fresh water interface near U. S. 41. The relative proportion of phosphate and
chloride in solution may serve as an adequate preliminary water quality indicator
of some significance to the mangrove ecosystem. The water chemistry data of
Chapter X indicate an exponential decline in the concentration of total phosphorus
relative to chloride along the opposite gradient, from U. S. 41 to a point near
the ocean front. Figure 111-24 derived from water quality survey data for
October, is representative of the contrast between the Fahka Union (canal fed)
and Fahkahatchee systems. Note that in the mid-bay stations, the relative
concentrations of total phosphorus vs chloride between the two systems are
virtually identical. If this figure is taken in the sense of a resource
utilization curve for the mainland manp*"<-«T? ecosystem, it indicates that the
canal system short-circuits the "chemical resource potential" of flowing fresh
waters from the upland directly to the bay cosystem, where it is rapidly
dissipated by admixture with sea water. Although the insular mangrove forest
may receive some benefit from this process, the resource loss from the mainland
mangrove ecosystem is absolute.
111-27
-------
200
0. IOO
o
RIPARIAN BAY MID
FRONT BAY
•MAINLAND FOREST OR CANAL » «—
OCEAN
FRONT
-INSULAR FOREST-
POSITION IN ESTUARY
Figure 111-24. Fresh water dynamics.
to
We have schematized
the overall interaction
of tidal pumping and
water chemistry, as it
may affect the function-
ing of mangrove ecosystems,
in Figure 111-25. Both
the tidal amplitude and
the "chemical resource
potential" curve show an
exponential decay with
distance from the source.
The respiratory drain on
the system may be high
whenever either of these
functions is at a low
level.
If so, we can expect
that the fraction of the
mangrove ecosystem
budget that is devoted
to immediate needs,
(R24/GPP, Figure III-
23), could be related
to the sum of the
reciprocals of the
functions shown in
Figure 111-25. This
would predict that
(R24/GPP) should show
a generally U-shaped
response curve accord-
ing to position in the
estuary. Salinity of
the surface water mass
served as an adequate
indicator of position
during a synoptic
survey of surface water
chloride concentrations
at the metabolic study
sites. Hence, in Figure
111-26, we show (R24/GPP) as related to position or surface water chloride
concentration. The pattern in this figure accords fairly well with expectation.
However, this is at best a fairly rough approximation of an extremely complex
situation.
Some preliminary estimates of salt marsh and mangrove community metabolism
are presented in Chapter XII. In general, these results support the conclusion
of Lugo, et al. (at press) that zonation of mangrove communities has a strong
metabolic correlative that may help explain the related competitive abilities
of the several species of mangroves in their respective zones of dominance.
The data of ChapterXIIE indicate that this concept may supply equally well to
a:
<
a:
H
CD
cc
US. 41
BAYFRONT
POSITION IN ESTUARY
Figure 111-25. Spatial distribution of tidal
flushing and sweet water.
111-28
-------
salt marsh communities.
Although the evidence is
not conclusive, it seems
that the loss of overland
fresh water inputs to the
well developed mangrove
forests of southwest
Florida will at least
induce shifts in the
community structure of
these forests. The
impact of these shifts
on the detrital-based
economy of the lower
estuary would then
depend upon the
relative litter
production of the
several species and
the transport dynamics
of the mainland man-
grove ecosystem.
0.5
0.4
fc
o
S
cr
0.2
O.I
O BLACK MANGROVE
A WHITE ><
• RED
-------
IV INTRODUCTION
GENERAL
The decade ending in 1970 may well be remembered as "the beginning of the
end" for south Florida. Portions of south Florida had growth rates reminiscent
of the 1920 boom days in Miami. From 1960 to 1970 Collier County was the
fifth fastest growing county in the United States with an increase in popula-
tion from 15,753 to 36,568 or 132 percent. At the present time there are no
indications that the growth rate will not continue to "snowball". Based on
expectations of the major developers and industries in Collier County, the
1970's will surpass the 1960's in growth.
Several surveys (Hawkins, 1972; Candeub, Fleissig and Associates, 1973)
indicate that Collier County will grow in population from the present 40,000
to 200,000-250,000 in the next 14 to 27 years. Land is being sold which
normally is under water from fifty to ninety percent of the time. Upon
visiting the site of their retirement home, retirees are often dismayed at what
they find. Because of salinity barriers, access to the Gulf is not possible.
Drainage canals are often choked with dense vegetation. Adequate county roads
are few. The cost of providing telephone service to these remote areas is
horrendous -- running in the thousands of dollars. Municipal water supplies
and sewage treatment facilities are often unavailable.
Areas which were once highly productive are now drained. Vast mangrove
forests have been cleared and filled for homesites and commercial developments.
Tidal streams have ceased to flow because of bulkheads. Public access to
beaches is being curtailed by oceanfront condominiums and apartments. The use
of pesticides is threatening to upset the delicate balance of nature and to
eliminate entire species of animals.
What is the fate of the south Florida wilderness known as the Big Cypress
Swamp, the Everglades and the Ten Thousand Islands? Will man and his greed for
material possessions destroy these unrestorable resources for a short-term
monetary gain? Or will' the State or Federal Government, in their attempt to
preserve or protect these areas, impose such stringent regulations as to
prohibit all growth ,in south Florida? Somehow there must be an accomodation
between developers and conservationists. Man and nature have lived in union
for many thousands of years, and if both are to survive, man must realize
that he is but a small part of nature, and must not disrupt the vital organs
of his ecosystem.
This study examined the intricate relationships existing between the
various components of natural and disturbed ecosystems existing within the
Big Cypress Swamp and the Ten Thousand Islands. With a quantitative knowledge
of these inter-relationships, predictions may be made concerning the dependence
of one ecosystem upon another. With this knowledge county commissioners and
other decision makers in south Florida will be able to extend their decision
criteria beyond conflicting pressures from special-interest groups within their
constituencies and their own intuitions of the general public welfare.
IV-1
-------
BACKGROUND
In September 1968, the Bade County Port Authority began construction of a
39 square mile jetport six miles north of Everglades National Park and on the
edge of the Big Cypress Swamp. No environmental studies preceded the selection
of the site, and, although the Department of the Interior had expressed concern
for the effects of the jetport on Everglades National Park, the Miccosukee
Indians and the Big Cypress Swamp, final announcement of the selected site and
start of construction was learned through the news media. With a definite
knowledge of the site, Interior's fear for the future of Everglades National
Park, the Big Cypress Swamp and the Miccosukee Indians increased. Interior
and the Department of Transportation agreed to study the problem. The study
produced a report (U. S. Department of the Interior, 1969) prepared by an
Interior team of specialists under the direction of Dr. Luna B. Leopold of
the U. S. Geological Survey. The findings and recommendations of the report
are summarized below.
Development of the proposed jetport and its attendant facilities will
lead to land drainage and development for agriculture, industry,
housing, transportation and services in the Big Cypress Swamp which
will inexorably destroy the south Florida ecosystem and thus the
Everglades National Park. There are three alternatives for future
action:
1. Proceed with staged development of training, cargo, and com-
mercial facilities. Regardless of efforts for land-use regulation,
the result will be the destruction of the south Florida ecosystem.
Estimates of lesser damages are not believed to be realistic.
2. Proceed with final development and use of a training facility
of one runway, with no expansion for additional use. Obtain an
alternative site for expansion, probably through an exchange of
excess lands at the current site for public lands at a new site.
Permit no new or improved surface access to the current site.
This alternative would not preclude eventual development of lands
in the vicinity of the current site. It could, however, reduce
pressures for development and secure time for the formation of
sufficient public interest in environmental conservation to
achieve effective planning and land-use regulation.
3. An alternative site be obtained capable of handling the
training operation as well as the fully developed commercial
facility; and that when appropriate, the training activities
at the present site be abandoned and transferred to the new site.
Permit no new or improved surface access to the current site.
This would give an impetus to developing effective land-use
controls which could lead to permament protection of the south
Florida ecosystem.
In arriving at these recommendations, the study team came to a number of
conclusions regarding the impact of the jetport on the environment of south
Florida and on the general effect of development on the natural conditions of
any area or region. On the basis of the report, Secretarys Hickel and Volpe
agreed that an alternate site should be selected and that the single-runway
training strip could operate, with stringent environmental controls, until the
IV-2
-------
alternate site was selected. On January 15, 1970, the White House announced
that an agreement (The Everglades Jetport Pact, 1970) had been made with the
Dade County Port Authority and the State of Florida whereby the jetport would
be moved to a more desirable location.
Under the terms of the agreement (Jetport Pact) signed on January 16,
1970, between the Departments of Interior and Transportation and the State of
Florida, and the Dade County Port Authority, the Port Authority agreed to begin
immediately a search to relocate the commercial airport away from the planned
site in the Big Cypress Swamp area adjacent to the Everglades.
The agreement also provided that the facilities already constructed on
the site would be used only as a single-runway airport for flight training
purposes and that such operations would be carried out under strict environ-
mental safeguards designed to protect the Everglades National Park and the
Miccosukee Indians. Such training facilities would eventually be transferred
to the alternate commercial site.
The two Departments agreed that Dade County would not be required to
abandon the current airport site until an alternate site had been agreed upon
and acquired without cost to Dade County, and training facilities substantially
equivalent to those existing at the present airport had been constructed on the
site without cost to Dade County.
The agreement, recognized the need for a south Florida regional airport,
the construction of which must be completed before the end of the coming
decade, was effective for three years and was subsequently extended for a two-
year period ending January 16, 1975. This extension was primarily to allow
the parties to complete their obligations pertaining to the selection of an
alternate airport site. The agreement provided that the Department of the
Interior would undertake a comprehensive program to determine the present
condition of the environment, and monitor the effect of operations at the
training airport.
Dade County agreed that it would not construct additional runways, taxiways,
buildings, structures, or facilities of any type, except as required for fire,
rescue and security purposes and would not improve or extend the use of the
existing facilities already in place at the airport.
The Interior Department agreed to provide recommendations for uses of the
Big Cypress Swamp, while Dade and Collier Counties agreed that they Would
control, limit, and restrict all drainage in the Big Cypress Swamp area until
a comprehensive land use plan was agreed to among all parties.
Pursuant to the agreement between the United States (Department of Trans-
portation and Department of the Interior) the State of Florida, and the Dade
County Port Authority, the Secretary of the Interior agreed to perform the
following tasks: (1) Provide site criteria for locating an alternate jetport
site for south Florida and apply the criteria to the site offered for evaluation;
(2) Establish an environmental monitoring program for the area affected by the
presence and use of the existing jetport site; (3) Undertake the planning,
development, and coordination of an ecological study of the region, including
its hydrology, and providing recommendations for uses of Big Cypress Swamp,
which would be consistent with preserving and protecting the environment and
ecosystems of Everglades National Park, the water supply of the affected
IV-3
-------
communities and the marine resources of dependent estuaries.
The foregoing programs were accomplished as a joint effort by five bureaus
in the Department of Interior (National Park Service, Geological Survey, Bureau
of Sport Fisheries and Wildlife, Bureau of Outdoor Recreation, and the Bureau
of Indian Affairs) and two former bureaus, the Federal Water Quality Administra-
tion and the Bureau of Commercial Fisheries, now transferred to the Environmental
Protection Administration and National Oceanic and Atmospheric Administration,
respectively, under the direction of the Everglades-Jetport Advisory Board
composed of directors of each participating bureau and agency. The director
of the National Park Service was designated Chairman of the Everglades-Jetport
Advisory Board and the National Park Service was directed to coordinate the
program.
Public concern and pressure forced the Jetport to abandon its Big Cypress
location. The problem which forced the abandonment was not the direct effects
of the jetport itself, but the uncontrolled growth, with its inherent problems
around the periphery of the jetport. However, that very problem now appears
in other portions of the Big Cypress and Ten Thousand Islands. Massive growth
patterns are evident in western Collier County and are continuing easterly.
Potential sites include developments in Gum Slough, Marco Shores, and Ochopee.
OBJECTIVES
The overall objective of this study was to obtain the necessary technical
information for objective planning of a program for wise use of the natural
resources of south Florida's land, water, wildlife and fisheries. To accomplish
the objective it was necessary to predict the impact of different types and
levels of human activity on the characteristics of various south Florida
ecosystems.
The strategy selected was to obtain detailed knowledge of the intricate
inter-relationships existing between the various components of ecosystems
presently subject to a minimum of external perturbations resulting from human
activity. This detailed knowledge was used to formulate process studies and
experimental manipulation models for the various components of the ecosystem.
These models were then used for predicting some effects on the systems of
various inputs of change which could result from differing types and levels of
land use. These predictable effects are applicable to much of the remaining
undisturbed areas throughout the south Florida region which are ecologically
similar.
Concurrently in gathering information for the systems analysis and
modeling program, a detailed characterization of the study area was made.
This included background data on chemical quality of waters and sediments;
pesticide levels in water, sediments, fish and higher animals; and life history
aspects of several freshwater and marine fishes, especially the much sought
after snook. Several detailed studies relating to man-made changes to the
environment were conducted. These include salinity variations in natural
versus man-influenced estuaries, variations in benthic communities in natural
versus man-influenced estuaries, and the effects of canals and other drainage
on ground and surface waters.
IV-4
-------
SCOPE
In order to accomplish the stated objectives, a decision was made to
conduct an in-depth study of a representative area of the Big Cypress Swamp.
Logistics and time limitations prohibited studying the entire south Florida
region. The selected study area was Fahkahatchee Strand and contiguous areas
.including the bays and estuaries to the southwest.
The major biological and ecological balance of south Florida is predicated
upon the hydroperiod. Because of the dependence of response patterns of many
species of flora and fauna on the complete cycle of the hydroperiod, it was
initially intended that all phases of the study be conducted over a minimum one
year period. However, because of late arrival of some equipment, the minimum
one year period could not be met for all portions of the study. With the
exception of plant community productivity, all segments of the study were
conducted and completed as scheduled from October 1971 through January 1973.
AUTHORITY
The authority for the South Florida Ecological Study is a combination of
a Secretarial directive (Department of the Interior), normal Department of
Interior authorities and interagency memoranda of understanding, and a specific
appropriation by Congress. The National Environmental Policy Act of 1969
(PL-190, enacted January 1,'1970) sets forth policy and concepts on which the
South Florida Ecological Study was based. The purposes of the Act, in addition
to establishing a National Environmental Council, are to declare a national
policy which will encourage productive and enjoyable harmony between man and
his environment, to promote efforts which will prevent or eliminate damage to
the environment and biosphere, to stimulate the health and welfare of man, and
to enrich the understanding of the ecological systems and natural resources
important to the Nation.
IV-5
-------
ACKNOWLEDGEMENTS
The cooperation and assistance of the following individuals and agencies
are gratefully acknowledged:
Mr. Harmon Turner, County Manager, Collier County, Florida.
Mr. Tom Peek, County Engineer, Collier County, Florida.
Dr. Samuel C. Snedaker and Staff, Drs. Daniel Ward, H. T. Odum and
Suzanne Bayley, and Dr. Ariel Lugo and Staff, University of Florida.
Mr. Tom Buchanan and Staff, U. S. Geological Survey, Miami, Florida.
National Field Investigation Center, Environmental Protection Agency,
Cincinnati, Ohio (especially Messrs. Allan Lucas, Jerry Kaiser, and
Jim Steinfield).
Mr. Bernie Yokel, Director, Rookery Bay Marine Station, Naples, Florida.
Mr. H. E. Smith, Fahkahatchee Strand, Florida.
Mr. Joe Morrison, Big Cypress Bend, Florida.
Florida Department of Natural Resources, Division of Marine Resources.
Dr. Frank C. Craighead, Naples, Florida.
Mr. Robert R. Wheeler, County Sanitarian, Collier County, Florida.
Mr. Elwood Larsen, State of Florida Department of Pollution Control,
Punta Gorda, Florida.
Mr. E. Peter H. Wilkens, NOA, NMFS, Miami, Florida.
Personnel of the following marinas: Isles^ of Capri Marina, Marco River
Marina, and Remuda Ranch Resort Marina.
IV-6
-------
V - DESCRIPTION OF STUDY AREA
SOUTH FLORIDA
The vast peninsula of south Florida possesses many features that are
unique among American landscapes. First, it is a land that has been created
entirely through the efforts of living creatures. Neither the potent geological
forces that built the great mountain chains of the West, nor the patient
weathering of rocks by wind and water that eroded the Appalachian chain of
today's green hills, are anywhere in evidence. Instead, the foundation rock
of Florida is limestone, the result of million of years of activity by marine
plants and animals.
The landscape occupying this limey platform is equally unusual, and an
equally unique creation of the living world. The immense sweep of the saw-
grass plains of the Everglades and the fingering sloughs of the Big Cypress
Swamp are geologically very young. The area around Everglades National Park
emerged from the sea only 5,000 years ago, and the sawgrass plains just south
of Lake Okeechobee were marine meadows just 20,000 years ago (Craighead, 1971).
As the land emerged from the sea, it provided opportunities for colonization
to two very different groups of plant and animal life. North American
temperate zone species began to move down the Florida peninsula into the
newly created land, and tropical plants and animals of the West Indies region
began to arrive from the south. For the tropical species, Florida was the end
of a long chain of tropical islands ultimately connected to the mainland
forests of South America; the last island south of the zone where winter
frosts destroy life and require a yearly cycle of springtime renewal. The
temperate species' invasion of Florida was not hampered by ocean barriers,
but they may have encountered difficulties in competing successfully with the
better adapted animals and plants from the south.
The biological communities of south Florida today reflect the current
state of this continuing process. For example, the populations of large
mammals (white-tail deer, bear, cougar, raccoon) are derived from the north,
for the Florida current has prevented the migration of large tropical mammals
to Florida. In contrast, about 61 percent of the plant species are of tropical
origin, and 91 percent of these are species that occur in the Caribbean area
(Long, R. W., and 0. Lakela, 1971). Perhaps the most spectacular example of
this mixture occurs in the Fahkahatchee Strand of the Big Cypress area, where
native tropical royal palms exist side by side with bald cypress and swamp
hardwoods that are widespread throughout the Southeastern United. States.
Although ultimately these newly emerging ecosystems were to generate one of
the largest swamp and marshlands in the world, some major initial difficulties
had first to be overcome, primarily in terms of the special nature of the local
climate.
The climate of south Florida is broadly similar to that of oceanic islands.
Winter temperatures are mild, but the rainfall pattern throughout the year is
somewhat erratic. During the summer months, thunderstorms march across the
landscape and occasional hurricanes move in from the Caribbean; during the
winter the weather regime becomes a long succession of cloudless balmy days
broken by sporadic breakthroughs of cooler weather from the north. Thus, the
rainfall regime is marked by a summer rainy season from May through October
and a five month winter dry season. At first sight then, the pattern of water
V-l
-------
availability in south Florida seems similar to that of more northerly areas of
the United States, where winter rains are locked up in snow and ice and do not
become available to plants until much later in the year. Here, however,
another factor enters the picture: the extreme lack of topographic relief
within the south Florida basin. From Lake Okeechobee to Florida Bay the
seaward slope of the land averages only about two inches per mile (3 cm/km).
As a result, the flow of fresh water towards the sea is extremely slow and the
local thunderstorms spread a sheet of shallow water over large areas. Under
these restricted drainage conditions, dead plant material falling into the
shallow ponded waters is not readily decomposed, and thick deposits of organic
peats and muck soils gradually build up. The latest pattern that emerged
(Parker, 1955; Davis, 1943, 1946) was a vast basin of swamp and marsh
communities whose soils acted as a gigantic saturated sponge, absorbing the
summer rains and slowly releasing water during the winter drought season. The
importance of the development of this water storage capability to the biological
communities of the south Florida region (including man) cannot be overestimated.
Florida lies in the latitudinal belt of the great deserts of the world. If we
add the maritime effect of the Atlantic and Gulf water masses, consideration
of rainfall and temperature patterns leads to the.conclusion that south Florida's
natural landscape should be a "tropical savannah" (Hela, 1952). The sweeping
plains of Africa dotted with acacia and baobab trees are the best known example
of this kind of ecosystem; it is characterized by a long parched dry season
and lack of enough rain during the rainy season to make up losses in evaporation
and transpiration. During the dry season, life for the human members of these
communities is very difficult indeed, and both plant and animal communities are
very unlike what we see in south Florida today.
The variability within the average pattern is an important aspect of the
rainfall regime, and contributes to the need for water storage in any given
environment. Deserts are relatively lifeless not merely because they are
plagued by a low average rainfall (250 mm (10, in) or less), but because the
rain is so variable - some years it may not rain at all. The areas of North
America that support major human populations have mean annual rainfalls
ranging from 250 mm (10 in) to about 2,000 mm (80 in). The "coefficient of
variation" (ratio of variability (standard deviation) to the mean, s/x) varies
from as low as 0.1 for well watered regions to as high as 0.5 for the arid
regions (annual values) (Fair, Geyer, and Okun, 1966). This means that a year
with a rainfall deficiency as great as half the mean annual rainfall will occur
in arid regions as often as a deficiency of only one-tenth the mean in better
watered areas. If a stable consumptive-use pattern is to be maintained, high
values of the coefficient of variation signify that high storage requirements
exist and must be met simply to compensate for variation within the average
water-supply pattern. In other words, even when the overall pattern of water
use is in balance with the average supply, large populations of organisms
cannot be sustained in the absence of suitable "filters" to eliminate
environmental "noise".
The available rainfall records for south Florida covering the period
from 1825 to 1968 have been collected and analyzed by T. M. Thomas (1970).
These show that the coefficient of variation (c.v.) for annual rainfall amounts
to about 0.2 throughout the region. However, the variation on a month by
month basis is rather different. During the dry season, the c.v. runs from
0.75 up to values greater than unity, and wet season months show values from
0.3 to 0.7. A concrete example will serve to make the situation clear: for
Everglades City, in Collier County, the mean annual rainfall is 1,371 mm
V-2
-------
(54.41 in) over the period from 1931 to 1972, with a c.v. of 0.19. If we
choose January and July to represent the dry and wet seasons respectively, we
find that the rain varies from a low of 0.5 mm (.02 in) for January, 1950 to
173 mm (6.81 in) in January of 1958, with a mean for 1931 to 1972 of 42.2 mm
(1.66 in). The equivalent range for July is from 69.8 to 443.2 mm (2.73 -
17.45 in) on a mean of 209.8 mm (8.26 in). The coefficient of variation for
January is 1.0, for July values 0.4.
The total pattern that emerges is one of a regularly recurring wet and
dry season, with a highly variable detailed pattern of month by month delivery
imposed upon it. It is not our purpose here to make detailed comparisons of
south Florida with other areas of the country. Indeed, the month-by-month
variation in precipitation may in many regions be as great or greater than is
the case for south Florida. However, south Florida ecosystems are unique in
the extent to which the biotic community has developed an intrinsic water
reserve system in its massive organic peat and muck deposits. This appears
to have completely removed precipitation variability as a factor affecting
water availability to the ecosystems of this region. The weight of the
currently available evidence seems to indicate that environmental predictability
is a major factor in determining the diversity of biological communities
(Slobodkin and Sanders, 1969). Total abundance of organisms, by contrast, may
be a function of the size of the total resource base and the general rigor of
the environment. The lush ecosystems of south Florida are sustained by an
interaction of favorable conditions from both sides of this ledger. Moreover,
Futuyma (1973) has argued that highly diverse communities may be especially
unstable in the face of environmental perturbations, in the sense that
perturbation may produce irreversible changes in the species composition of the
ecosystem. He summarized the implications of his argument as "the bigger they
are, the harder they fall".
These effects sum up to the imposition of a necessary water storage
capacity as a condition of survival for south Florida ecosystems. The
apparent superabundance of water in south Florida is an illusion engendered
by the efficiency of ecological systems in storing water for subsequent
utilization in biotic processes of growth and reproduction. Without belaboring
the point, we should also point out that this mode of storage can be highly
efficient in terms of isolating the storage reserve from the evaporative
driving forces of sun, wind, and saturation deficit. Figure V-l shows the
exponential decline in evaporative water loss (expressed as a fraction of the
potential water loss from a free water surface) as the water table falls below
the surface of the ground (from Todd, 1959; after White, 1932). The advantages
over storage as surface water in lakes and reservoirs are apparent, for a lake
will, of course, continue to export water to the atmosphere at the free water
surface rate. In the case of a cypress slough there is an additional factor:
the cypress trees drop their leaves sometime around November, and do not leaf
out again until March, so the biological transpiration drain on the reserve
system during the dry season is very small. Considerations of this sort led
Odum, et al (1972), to recommend that the city of Naples control development
of a cypress slough within the city limits so as to retain its essential
characteristics as a water conservation tool.
The successional pattern of development of these peat and muck soils is
one of gradual increase in the thickness of the soil layer up to a point where
it comes into equilibrium with the surface water supply. Thereafter, the
increments of dead plant matter added during the wet season are decomposed by
V-3
-------
biochemical oxidation when the
surface layers are exposed during
the following dry season. The
inorganic nutrients that are
regenerated by and large remain
in the soil, and the plant carbon
is returned to the atmosphere as
C02 gas .
eo
tr
o
o_
UJ
40
UJ
o o
UJ
0_
0 20 40 60 80
DEPTH TO WATER TABLE (Inches)
Figure V-l. Ground water evaporation,
expressed as a percentage of pan
evaporation, as a function of depth to
the water table (from Todd, 1959; after
White, 1932).
At this point it is worth
recalling that quasi-mechanical
analogies, however useful in
organizing our thinking, ulti-
mately break down when we are
dealing with complex biological
systems. An ecosystem should
be regarded as a living
synthesis of the opportunities
and restrictions imposed by nature
at a given place and time, and in
a sense it continuously recomputes
its trajectory toward future states
of the system as these conditions fluctuate. Furthermore, in the face of
abrupt shifts in environmental conditions the ecosystem sends out many feelers,
as it were, toward new possible states; with the most probable end point
defined by the novel complex of plant and animal populations best adapted to
the new conditions. This situation has led to the introduction in ecology of
a number of descriptive terms for these states and processes. "Primary
succession" refers to the initial development of a biological community from
a more or less sterile initial point; this is the process described above for
south Florida, "Secondary succession" refers to the process of restoration
that follows a major local disturbance when environmental conditions are not
changed, for example, regrowth of a mature forest after a logging operation.
The "climax" state refers to the developmental endpoints of the successional
process, and may be of several types. Most notably, the "climatic climax" is
the ecosystem type dictated by climatic conditions, for example, tropical' rain
forests or arctic tundra. "Edaphic" climaxes are states of arrested succession
or local endpoints of ecosystem development dictated by special local conditions,
for example, unusually high concentrations of heavy metals in the soil or
regular dry season fires.
In south Florida continuing succession beyond the primary stage described
above has developed a mosaic of pine forest, cypress strands, hardwood hammocks,
coastal mangrove forest and salt marsh, sawgrass plains and wet prairies. The
role of edaphic factors, especially seasonal fires and soil depth, has been-
described by several authors (e.g. Egler, 1952, Craighead, 1971). In general,
the climatic climax of the region appears to be hardwood forests, and perhaps
cypress forests in deeper runoff "channels". The pine forests, sawgrass plains,
and wet prairies are maintained by a combination of shallow soils and periodic
light fires of sufficient intensity to kill the seedlings of broad-leaved
woody plants but not the grasses or pines. Sawgrass, for example, will often
sustain a prairie fire when there is still five centimeters of water standing
above the surface. The roots and peat soil thus remain undisturbed, but
invading shrubs are all killed. Whether eventually the sawgrass Everglades
and Big Cypress prairies would have given way to vast forests of cypress, pop
V-4
-------
ash, mahogany, gumbo limbo, and maple will never be known, for residential and
agricultural development, for good or ill, began a process of drainage through-
out the region that has profoundly changed the conditions of existence for
most of the ecosystems of south Florida,
All organic soils lose surface elevation, or subside, when drained.
Stephens and Speir (1969) have summarized experience with this phenomenon in
south Florida and made comparisons to other areas of the world. In the
Everglades, the peat and muck soils have a subsidence rate of 3.05 cm per
year after initial settlement. The loss of soil at a stone monument at the
Everglades Agricultural Experiment Station from 1924 to 1968 has amounted to
1.6 meters (5.25 ft). Stephens and Speir (1969) predicted that, based on the
overall soil loss pattern, this great winter vegetable growing zone will be
useless for agricultural purposes by about the year 2000. Recently some
growers in the region have been flooding their fields during the non-growing
season to delay this endpoint, but the process cannot be halted, only slowed.
The loss of soil when the water table is lowered by drainage is of course only
the natural process of biochemical oxidation that formerly kept the peat level
in equilibrium with the natural water supply, writ large. But for the ecosystem
it is an abrupt change in the conditions of existence that entrains the forces
of ecosystem succession toward a new endpoint whose outlines can be only dimly
perceived. Attempts to ameliorate the initial destructive changes that have
excited the public imagination have further confused the picture. Secondary
shifts of water balance, through the development of water conservation pools
and timed releases to other areas, have imposed still newer forces on the
transient states already induced by drainage. We are left with a myriad of
small scale experimental attempts by the ecosystem to adjust to new conditions,
but with very few scientific criteria for predicting or controlling the overall
outcome, and with still fewer ideas as to what would constitute a desirable or
ideal final endpoint. One possible endpoint is clear, the tropical savannah
imposed by the basic climatic regimen. This prospect, one of summer floods,
followed by a parched winter dry season marked by acute water shortages as
uncontrollable wildfires sweep the basin, is uniformly held to be undesirable.
Man today controls almost all the water that flows in the south Florida basin,
but our efforts thus far to manage it for the good of all have had grisly side
effects. During the drought of the early 1960's, more than half the alligator
population of Everglades National Park perished when the flow of water from
the water conservation areas to the north was cut off to maintain the city of
Miami's water supply (Craighead, F. C., Jr., quoted in Caulfield, 1970).
Conversely, more recently the southern parts of the conservation pools
and the Loxahatchee wildlife area have been invaded by purely aquatic and pond
ecosystems as water levels have deepened. What has been missing is not good
will, nor energetic measures ingeniously applied, nor the willingness to try
new solutions for novel problems. Rather we simply do not understand in
sufficient depth or detail the ways the past ecosystem built its own water
supply system; nor a detailed knowledge of the ecological processes that we
might enhance to prevent the destruction of the south Florida landscape in toto
as we use parts of it for our own economic and human ecological purposes.
As part of the South Florida Environmental Study, the Environmental
Protection Agency (EPA) decided to try to improve our knowledge of the
functioning of natural south Florida ecosystems and their initial responses
to drainage by a concentrated effort in one relatively restricted ecosystem
unit. The Fahkahatchee Strand and its associated wet prairies, salt marshes,
V-5
-------
and mangrove lined estuary was chosen for this purpose. Altogether, the study
area (Figure V-2) comprised 460 square kilometers (177 sq mi), mostly in the
Big Cypress and wholly within Collier County.
I
LAKE
OKEECHOBEE
CHARLOTTE CO. I GLADES CO.
PALM BEACH CO.
COLLIER CO.
BROWARD CO.
FT. LAUDERDALE
,S
TEN
THOUSAND
ISLANDS
I HOMESTEAD
0 5 10 IS 20 30 40
KILOMETERS
EVERGLADES
NATIONAL
PARK
Figure V-2. General location map of the Big Cypress Swamp and study area.
V-6
-------
BIG CYPRESS SWAMP
The Big Cypress Swamp is located in Southwest Florida within the
Floridian section of the Coastal Plain province (Figure V-2). Portions of
Collier, Lee, Monroe, Dade, Broward, and Hendry Counties are included in the
6,340 square kilometers (2,450 sq mi) of the Big Cypress Swamp. The Ten
Thousand Islands serve as the southwest boundary of the swamp.
Much of the Big Cypress Swamp is less than 4.9 meters above mean sea
level. The slope of the land surface is generally south to southwest and
varies from 8 to 16 cm/km (5 to 10 inches/mile). Natural drainage of the land
is slow. The swamp contains many small depressions or "potholes" which have
no surface drainage. These provide the only source of surface water during
droughts. Large areas covered by small to medium cypress trees, swamps con-
taining large cypress and deciduous broad leaf trees. pine islands, and wet
prairies are characteristic of plant communities in the region. Some note-
worthy cypress strands and swamps include Fahkahatchee Strand, Corkscrew
Swamp, Camp Keasis Strand, Kissimmee Billy Strand, and Wilson Strand
(Leighty et al, 1954) (Figure V-3).
I Mena
Hendry Co
Lee
• C :--'Uake
l$Mf-
'•< L?."5,
•"s:r-
»>.'..^
Jmmokalee
^^^s.
A
\^
|
Naples
Skillet
Strand/; Monroe
Strand
-7s Roberts N;
I*. LSK»_Strqnd_.
Monroe Co.
41
Figure V-3. Major strands of the Big Cypress Swamp.
V-7
-------
The Ten Thousand Island chain extends from Gordon Pass (South Naples)
southeastward for about 64 kilometers (40 miles) along the southwest coast of
Florida. This low coastal region features many tidal rivers, bays, lakes, and
thousands of small islands. Much of the area is covered by mangrove swamps
and salt-water marshes. Chokoloskee Island, consisting chiefly of ancient
shell mounds made by pre-Columbian Indians, and Marco Island are the best
known of the Ten Thousand Islands (Leighty et al., 1954).
Geologically the Big Cypress Swamp is characterized by Tamiami limestone
of Miocene age. The limestone is rather impervious, containing much quartz
sand of the Pamlico and Anastasia formations. The swamp is scalloped by many
basins and cut by sloughs (Craighead, 1971). The Tamiami formation consists
of the thin Tamiami Limestone, less than 150 feet thick (Parker, _et a\, 1955),
and a thick section of greenish and greenish-grey argillaceous marls and silty,
sandy, calcareous, shell marls and sands (Puri, 1964).
Soil of the Big Cypress Swamp is often pure sand, marl or mixtures of
both which layer from 5 to 61 centimeters (2 to 24 inches) deep on the
irregular surface of the limestone. Peat muck of various kinds fills the
depressions in the bedrock. A thin layer of marl or sand is usually found
just under the prairies and dwarf cypress. These soils deepen near the man-
groves (Craighead, 1971). Approximately 20.3 percent of Collier County con-
sists of Ochopee fine sandy marl, shallow phase (Leighty et al., 1954). The
limestone is generally found from 15 to 30 centimeters below the surface.
Cypress swamp lands account for approximately 18.6 percent of the Big Cypress
Swamp and contain a variety of soils consisting of an intermingling of black
or dark-grey mucky fine sand or peaty muck and brown peat with a subsoil of
grey or light-grey fine sand. Thirty-two soil types are found in the Big
Cypress Swamp and the southwest coast and Ten Thousand Islands.
The Big Cypress Swamp is unique in that nearly all surface waters
originate within the Swamp from rainfall. There are no significant surface
streams or rivers which flow into the swamp.
According to the U. S. Geological Survey (Klein et al., 1970) the Big
Cypress Swamp can be divided into three sub-areas each having reasonable
distinct internal drainage boundaries which are determined largely by topo-
graphic configuration and man-made drainage (Figure V-4). Sub-area A is
northeast of a low ridge and drains southeastward into the Everglades and
Conservation Area 3 of the Central and Southern Florida Flood Control District.
Sub-area B is located in the western portion of the Swamp. The historic
sheet flow has been altered by an extensive system of canals that drain the
water southward to Fahka Union Bay and Rookery Bay and westward to Naples Bay.
This sub-area contains the most extensive residential development in the Swamp.
Sub-area C occupies the central portion of the Big Cypress Swamp. With
the exception of the Turner River Canal, Barren River Canal and the easternmost
GAG Canals the area has been changed very little by man. Fahkahatchee Strand,
Okaloacoochee Slough, and Deep Lake Strand are the most well known natural
features of this sub-area. Sheet flow exists in a large portion of sub-area
C but the natural drainage is yielding to encroaching development along U. S.
41 and S. R. 84, Figure V-4 presents a generalized flow pattern for the Big
Cypress Swamp during a relatively high flow period from November 18-20, 1969.
V-8
-------
NAPLES
0 6 I'O
KILOMETERS
EVERGLADES
NATIONAL I
PARK I
Figure V-4. The three distinct sub-areas ;i. the Big Cypress Swamp showing
major drainage patterns.
V-9
-------
The estuaries of Southwest Florida are a complex system consisting of
thousands of islands connected by intermingling tidal streams and passes.
Salinities in these shallow estuaries depend on the freshwater input and may
range from hypersaline conditions during the dry winter season to very low
salinities during the wet summer season.
FAHKAHATCHEE STRAND
This study was focused on a small portion of the Big Cypress Swamp known
as Fahkahatchee Strand and contiguous areas that included salt marshes, man-
grove forests, and estuaries to the south. The study area was bounded to the
east by S. R. 29, and on the north by S. R. 84 (Alligator Alley); on the west
by the GAG drainage network, and by the Gulf of Mexico to the south. In total,
the study area covered approximately 460 square kilometers (177.5 sq mi). Its
unique native stand of Royal Palms is a well-known feature of the Fahkahatchee
Strand. The Strand is also widely known for its spectacular orchid populations,
especially in the deepest central slough. A more detailed map of the study
area is given in Figure V-5. Inspection of this map indicates that the
Fahkahatchee Strand is by no means an "undisturbed" ecosystem. Overland flow
from the Okaloacoochee Slough and other areas to the north is effectively cut
off by the system of highway borrow canals and GAG Properties drainage works
that together surround the Strand on three sides. This effect is especially
noticeable during storm periods, when waters that would normally be widely
distributed and flow slowly toward the Gulf are rapidly transferred to the
estuary by these canal systems. Virtually all of the virgin cypress was cut
from the Fahkahatchee during logging operations that ended in 1950 (Craighead,
1971). Altogether, about 85 percent of the lands in Collier County are
woodlands, and most have been subject to logging operations: of the virgin
pine forest of the area, only 10 percent remained in 1954; the rest of the
original pinelands were second growth stands of small size (Leighty, 1954).
In the Fahkahatchee, maple, willow and other swamp hardwoods have moved
from their secondary position in the virgin forest to full dominance today.
The logging operations involved the construction of a railway across the
Strand that today forms the roadbed for Janes Scenic Drive. Approximately
every 500 meters (1,600 ft) a perpendicular tramway built to facilitate removal
of the felled logs enters the swamp; these were constructed so that no point in
the swamp is more than 250 meters (800 ft) from dry land. This network of
tramways is what provides the herringbone appearance of the Strand in Figure V-5.
Immediately to the west, the platted roads and drainage ditches of Golden Gate
Estates are nearing completion.; these too are shown in Figure V-5.
Given these restrictions, we elected to concentrate on studies that might
identify and explore the basic driving forces of the system; the resultant
ecosystem metabolism and tuning to seasonal variation in these driving forces;
and some of the possible dependencies among subsystems that might be producing
integrated responses across the entire ecosyst-em from the upland forest to the
estuary. We made comparisons of similar subsystems according to their
location vis-a-vis the western boundary drainage works. Within the estuary,
we examined mangrove fringed bays being fed directly from upland drainage
works (Fahka Union Bay) and a bay somewhat removed from drainage inputs of
fresh water (Fahkahatchee Bay) (Figure V-5). By organizing these efforts
around compartmental models of ecosystem dynamics and energetics developed
by H. T. Odum (1972), we tried to format our studies so as to be able to
draw some tentative but general conslusions concerning ecosystem functioning
and responses to drainage.
V-10
-------
FAHKAHATCHEE BAY
Figure V-5. Map of study area and drainage works immediately to the west.
V-ll
-------
ECOSYSTEM MODELS
Early in the study period allocated to the South Florida Environmental
Study, the Department of the Interior (NFS) contracted a group at the University
of Florida Center for Aquatic Science to construct a set of models of the
south Florida regional ecosystem and some of its subsystems. That study (Lugo
et al., 1971) attempted to coalesce previous environmental studies done within
the region, and the views of individuals active in south Florida environmental
fields, into an overview of environmental processes and interactions important
to the south Florida basin. In order to use that study as an organizational
tool, we needed first to define the ecosystem types present within the
Fahkahatchee study area and in the Big Cypress as a whole. We used a published
map series on soil types and associated native vegetation (Leighty, 1954) for
this purpose. By polar planimetry of these maps we determined the areal
distribution of plant community types within the study area, which results are
given in Figure V-6a. To compare the study area to the whole Big Cypress
complex, we planimetered that portion of the Big Cypress lying within Collier
County; plus the salt marsh, mangrove, and estuarine areas from Naples south
to the Collier - Monroe County line (see Figure V-3). The result of this
procedure is presented in Figure V-6b. The greater representation of deep
-SWAMP HARDWOODS
AND CYPRESS
MANGROVE
MISC.-
-CYPRESS STRAND
GRASSY UNDERSTORY
SWAMP HARDWOODS
AND CYPRESS
CYPRESS STRAND
GRASSY UNDERSTORY
Figure V-6a,
-SABAL
WET
PRAIRIES'
FRESHWATER-
MARSH
PINE
Figure V-6b.
Proportional representation of the distribution of plant community types in
the Fahkahatchee Strand study area (6a) and that portion of the Big Cypress
Swamp and its associated estuary lying within Collier County, Florida (6b).
The Fahkahatchee area is included as part of the total Big Cypress analysis.
slough communities (swamp hardwoods and cypress) in the study area (Figure V-6a)
is due to centering on the Fahkahatchee slough; and the lesser representation
of mangrove and estuary in the overall pattern (Figure V-6b) is caused by two
factors: (1) the larger inland area involved, and (2) cut-off of estuarine
V-12
-------
areas connected to the Big Cypress that lie in Monroe County. Nonetheless
the community types of both areas are the same, and these acreages might be
used with estimates of productivity for the various subsystems to estimate
the total organic energy budget of the region.
Our criteria for determining the effort to be expended on any given site
type were based on several factors. The first was importance of each subsystem
to the overall region, indicated by its areal representation. Secondly, we
hoped to delineate the influence of-artificial drainage on each subsystem by
pairing sites strongly influenced by drainage with sites relatively remote from
the canals. In effect, we hoped to choose sites that would simulate an
experimental approach to drainage effects by studying similar communities
that differed simply in location vis-a-vis existing drainage works.
A subset of the original set of ecosystem models formulated by Lugo, et_ al.
(1971) can be found in Appendix B with the material they produced to describe
the models. That appendix is an abstraction from their report of subsystem
models appropriate to the Big Cypress area. For our purposes, these models
required considerable simplification to make them amenable to a field investi-
gation period of one calendar year using limited manpower. In addition, our
models were intended as hypotheses of system function that could ultimately be
used for computer simulation, and large models designed to serve as diagrams
for organizing and coordinating research are unwieldy in that context.
Several revisions of our models were also made necessary as field work and
analytical limitations became apparent during the course of the study period.
V-13
-------
VI - CLIMATOLOGY
INTRODUCTION
The subtropical climate in south Florida is directly responsible for many
of the biological and physical features of the Big Cypress Swamp and the Ever-
glades. This subtropical climate with its alternate wet and dry seasons, is
generally favorable to the phenomenal plant growth of the area. The high
relative humidity found in various swamps is at least indirectly responsible
for many of the native orchids which are found in no other location throughout
the world. Severe natural disturbances to the ecosystem occur only when the
normal climatic extremes are exceeded during excessive droughts and freezes,
and during hurricanes.
Some of the unique biological adjustments resulting from the climate and
the edaphic factors of south Florida are seen in the mangrove forests that
flourish under alternate baths of fresh and salt water; in the tropical hard-
wood hammocks that provide a mild microclimate of high humidity and frost-free
temperatures suitable to a tropical flora of trees, ferns, orchids, and bromeliads,
and in the pineland forests of the rock land that support more than 100 endemic
plants adapted to the fires of the dry period (Craighead, 1971).
RAINFALL
Methods
Six Universal Weighing Rain Gages were strategically located in the study
area (Figure VI-1). The rain gages were designed to receive a maximum of 30.5
centimeters (12 inches) of rainfall with a 196 hour recorder.
Background
Highly variable temporal and spatial rainfall patterns are well documented
for south Florida. A striking example occurred in 1947 when 298.7 centimeters
(117.6 inches) of rain fell east of the southern portion of Water Conservation
Area No. 1, while only 105 kilometers (65 miles) to the northeast some 142.2
centimeters (56.0 inches) of rainfall occurred over Taylor Creek (Department of
the Army, January 15, 1968).
The hydroperiod of southwest Florida is historically defined. Generally
60-65 percent of the annual precipitation occurs during the summer months of
June through August. The winter months of December through February account
for some 8-10 percent of the rainfall.
During the winter and early spring, frontal movements result in widespread
light-to-moderate rains in south Florida which may last two or three days.
From late spring to autumn, easterly warm, moist air masses move inland from
the Atlantic Ocean and the Straits of Florida. Low pressure areas over the
Atlantic Ocean and Gulf of Mexico also provide large quantities of moisture
inland. Prolonged periods of precipitation can result from continued inland
air-flow around slow moving pressure centers. Sea breezes contribute to the
moisture supply in a relatively narrow band along the coast. Much of the
summer rainfall occurs in the form of thunderstorms, and varies greatly in
intensity and locality. Winter thunderstorms are usually a closely spaced
VI-1
-------
STATE FORES7V
SERVICE V
COPELAND STA.V-
Figure VI-1. Map of study area showing climatology station locations.
VI-2
-------
series of cells in an extended squall line or in locally unstable air, and
rarely occur in any great number. Summer frequency of thunderstorms can be as
high as 60 percent. Hurricanes and less severe tropical storms are at times
major sources of precipitation (Department of the Army, January 15, 1968).
Several excellent reports have been prepared concerning analysis of
rainfall frequency for southern Florida. Of concern to the study area
Hughes, et al., (1971) reports the follow-
ing rainfall based on the period 1931-1955:
annual 132-142 centimeters (52-56 in'. ies);
winter 13 centimeters (5 inches) spxyng
20-25 centimeters (8-10 inches) £ suiru.._r
61 centimeters (24 inches); and autumn
36-41 centimeters (14-16 inches). Thomas
(1970) reports rainfall on an annual
monthly basis (Table VI-1).
Table VI -1. Annual and monthly
mean rainfall in the vicinity
of Fahkahatchee Strand based
on analysis of historical rain-
fall by T. M. Thomas.
Rainfall
Month
Centimeters Inches
The National Weather Service main-
tains two long term weather stations
in the vicinity of Fahkahatchee Strand.
One is located in Everglades City,
Florida, and the other in Naples,
Florida. Extreme variations in annual
rainfall are noted at both sites
(Figures VI-2 and 3). Everglades City
experienced a maximum annual rainfall
of 198.6 centimeters (78.19 in) during
1948 and a minimum of 92.7 centimeters
(36.50 in) in 1942. The average annual
rainfall at Everglades City based on
the period 1931-1972 is 138.2 centimeters
(54.41 in). Rainfall for 1972 was 157.1
centimeters (61.85 in) some 14 percent
above normal. Maximum rainfall at Naples
occurred in 1959 with 184.2 centimeters
(72/50 in) and a minimum in 1944 with 77.7 centimeters (30.60 in). Average
annual rainfall for the period of record 1941-1972 is 133.73 centimeters
(52.65 in). Naples received
141.7 centimeters (55.78 in) 2
during 1972, some six percent
above normal. '
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Annual
2.5- 5.1
2.5- 5.1
2.5- 5.1
5.1- 7.6
10.2-19.3
20.3-22.9
20.3-22.9
17.8-20.3
22.9-25.4
10.2-15.2
2.5- 5.1
2.5- 5.1
127 - 152
1-2
1-2
1-2
2-3
4-5
8-9
8-9
7-8
9-10
4-6
'1-2
1-2
50-60
Monthly rainfalls are more
dynamic than annual rainfalls.
As observed from Figure VI-4
seven months have records of
zero precipitation at Everglades
City. The most variable month
is June, when historic rainfalls
have ranged from a low of 6.45
centimeters (2.54 in) to a high
of 62.8 centimeters (24.73 in)
with a mean of 24.51 centimeters
(9.65 in). Monthly rainfall of
less than 2,54 centimeters
3 iso
_j
_i
2
? 125
CC
1950
YEAR
Figure VI-2, Annual rainfall at Everglades
City, Florida, 1931-1972.
VI-3
-------
(1 in) occurs 55 percent of the
time for January and 45 percent
for November. The rainfall may
be expected to exceed 25.4
centimeters (10 in) 44 percent
of the time for June and 26 per-
cent of the time for July.
Figure VI-5 is a summary of
annual and monthly rainfall at
Naples for the period May, 1940-
December, 1972. As expected,
the rainfall pattern at Naples
was different from Everglades
City. For example, the month
with the maximum and highest
average rainfall was September,
however, it is obvious that the
same hydroperiod is definable
from either set of records.
During 1972 both communities
received above average rainfall
(Figure VI-6). The greatest
positive departure from the
average monthly precipitation
at both locations was during
June due to the effects of
Hurricane Agnes. Seven months
produced positive departures
while five were negative.
200
175
E
•3. 150
125
<
o:
100
/\
1940
1950
I960
A
1970
YEAR
Figure VI-3. Annual rainfall at Naples,
Florida, 1941-1972.
Results
§
8)
o
EXPLANATION
!
i
1
100 ?j
I-
I—
o*
50 3
Figure VI-4. Summary of annual and monthly
rainfall at Everglades City, Florida,
1931-1972.
Rainfall (Figures VI-7 and
within the Fahkahatchee Strand
study area for 1972 varied from a
low of 120.13 centimeters (47.30
in) near the middle of Janes
Scenic Drive (Station 4) to a
high of 194.03 centimeters (76.39
in) at the State Forest Service
Tower near the eastern side of
Janes Scenic Drive. The yearly
average of the seven sites was 141.64 centimeters (55.74 in). Compared with
long term rainfall records at Everglades City only two of the seven stations
were "above average", yet, the yearly mean was 3.41 centimeters (1,35 in)
above average. The effects of Hurricane Agnes were obvious over the period
from June 16-19 when the rainfall ranged from 14.2 to 16.7 centimeters (5.61
to 6.59 in) and averaged 152.2 centimeters (5.98 in).
Based on the average rainfall of 141.64 centimeters during 1972, a total
of 650,877,000 cubic meters (527,672 ac-ft) of water fell on the study area
during the year. Of this, some 409,929,000 cubic meters (332,359 ac-ft)
arrived north of U. S. 41 and 240,958,000 cubic meters (195,313 ac-ft) south
of U. S. 41.
VI-4
-------
50
g30
Z
<
>20
I
H
O
EXPLANATION
ANNUAL-
i- MAXIMUM
--- MEAN
-- MEDIAN
I
lllllll'.l.
i • I • I M-!
I
I
200
Z
c
150 >
50 3
Figure VI-5. Summary of annual and monthly
rainfall at Naples, Florida, 1940-1972. |-'s
20
10
30
20
|
d |0
£
2
1
T 30
i
_j 20
<
1
10
20
10
STATION 1
' ALLIGATOR
ALLEY
J B . B
STATION 2
S.R. 29
R R
N M
R N H H R
. STATION 3
WEST JANES DRIVE
R R
J H R 1
STATION 4
MIDDLE JANES
DRIVE
R M R n
H Nl Kl bl H
^
\
\
\
s
p
r-
N
b
b
IS
N
S
N
N
S
* ^
J N
4 *•
1 ^
^
\
\
S
^
\
s
\
s
\
\
\
\
s
V
\
v R
r
s t
" F
s b
v b
s K
r*
K
k;
\
-\
1 F
i r
i k
n
i
1 F
r
|
r
] \
3
1
]
s n
^3
H
N
q R
s
V r
\
: n
; n «
^
71
<
/
•^
•
3 .
\
\ '
200
ISO
100
50
200
ISO
100 >
z
so c
J>
r
z
3
r
200 g"
3^
ISO
100
50
200
ISO
100
50
M A M J
J A S 0 N D 1972
1972
Figure VI-7. Rainfall during 1972 at
four sites in Fahkahatchee Strand,
J J
1972
Figure VI-6. Departure from average
monthly rainfall at Everglades
City and Naples, Florida during
1972.
p
40
30
20
10
40
30
20
10
30
20
10
' STATION
5
. EAST JANES DRIVE
|J
r
1
R
R
. STATE FOREST
SERVICE
COPELAND
STATION
R
s
v F
s b
v b-
v K
H
R
V
?
• STATIONS
US. 41
R k
q
3 n n
J F M
A
s
s
s
s
s
M
<;
s
s
s
s
N
N
S
n :
N
-
_
V
s
s
N
s
S
V
N
s
s
_
S
S
\
s
s
V
s,
\
\
S
s
\
s
\
P
^
: R r
a
^
^
^
s
s
s n
: 1
^ Kl
<
< .
<
1
<
<
\
\
\
\
*,
\
\
\
n
H
n H n
'
j
200
ISO
100
50
200
ISO
100
SO
ZOO
ISO
100
so
J J A S 0 N 0 1972
1972
I
Figure VI-8. Rainfall during 1972 at
three sites in Fahkahatchee Strand.
VI-5
-------
Maximum daily rainfall occurred on June 18 with an average of 11,75
centimeters (4.63 in). Rainfall exceeded 2.54 centimeters (1.0 in) on 45 days,
5.08 centimeters (2.0 in) on 10 days, 7.62 centimeters (3.0 in) on 3 days and
10.6 centimeters (4.0 in) on one day at various locations throughout the study
area. Seventy percent of the daily rain observations were less than 1.77
centimeters (0.5 in) and 29 percent less than 0.25 centimeter (0.1 in).
Rainfall was recorded on 168 days during 1972. Because of the spatial distri-
bution of rainfall, it was common to observe rain at only one of the six EPA
stations on any given day.
TEMPERATUREAND RELATIVE HUMIDITY
Methods
Two Belfort Instrument Company Hygrothermographs were installed in
Falkenberg Company Instrument Shelters at Station 2 located at the County
Prison Farm on S. R. 29 and Station 6 at Big Cypress Bend (Figure VI-1).
Each instrument was calibrated weekly with a standard thermometer and sling
psychrometer.
Background
Throughout south Florida temperatures are moderately high from July
through September and pleasantly warm to cool from October through May.
Because of temperature effects of the Gulf of,Mexico, coastal areas are cooler
than inland areas during the summer and warmer in the winter. Frost generally
occurs yearly in the Big Cypress Swamp, but rarely in the Ten Thousand Islands
and other coastal areas. With few exceptions relative humidity is moderate to
high in south Florida.
The National Weather Service maintains two weather stations near the
study area: one at Everglades City, Florida, and one at Naples, Florida.
The average temperature (1931-1972) at Everglades City (Figure VI-9) is
23.8° C with extremes ranging-from minus 4.4° C to 37.2° C. Temperatures
at Naples have averaged 0.2° C cooler than Everglades City (Table VI-2) based
on the period of record. As observed from Table VI-2 the temperature for
January 1972 was 3.3° C above average. Seven of the twelve months during
1972 were above average.
Results
Weekly ranges in temperatures for Stations 2 and 6 are presented in
Figures VI-10 and 11. Yearly temperature range at Station 6 was 4.4° C to
35,6° C with an average of 23.0° C.
Table VI-3 presents the 1972 average monthly temperatures for Stations
2 and 6. Since the nearest long term weather station at Everglades City is
located some 20.4 kilometers (12.7 mi) south of Station 2 and 9.6 kilometers
(6 mi) southeast of Station 6, it would be inappropriate to infer numerical
deviations from the long term>trend. However, it can be concluded that based
on the long term records from Everglades City and Naples that the months of
January, February, March, October, November, and December were above the long
term mean.
VI-6
-------
40 f
O 30
o
UJ
o:
QL
UJ
CL
UJ 10
MAX
AVG.
193O 1940 1950 I960 1970
Figure VI-9. Yearly temperature range at Everglades City, Florida, 1931-1972.
Table ¥1-2, Average monthly temperature (° C) for Everglades City and Naples,
Florida,
Month
Naples
Everglades City
1941-1972
1972
1931-1972
1972
January
February
March
April
May
June
July
Augus t
September
October
November
December
Average
18,5
IB. 8
20.7
23.1
25.1
27.2
28.0
28.3
27.7
24.8
21.7
19.3
23.6
21,8
19.3
21.2
23.1
25.3
26.6
27.6
27.8
27.6
25,4
22.7
20.7
24.1
18.9
19.3
20.9
23.2
25.2
27.0
27.9
28.1
27.7
25.4
22.0
19.5
23.8
20.9
18.0
*
22.3
24.9
26.2
*
*
27.7
*
23.2
20.4
*
* Unavailable
VZ-7
-------
40 r
Figure VT-10. Weekly temperature range at Station 2 during 1972.
40
30
UJ
cr
20
cc
UJ
Q.
UJ
10
JFMAMJ JASON
1972
Figure VI-11. Weekly temperature range at Station 6 during 1972.
As expected, relative humidity remained moderate to high at both Station
2 and 6 throughout the year (Figures VI-12 and 13). Maximum daily relative
humidity usually occurred in the late afternoon until mid-morning, and was
often close to 100 percent. Daily minimums occurred during the early afternoon
when temperatures were at a maximum. Average yearly relative humidity at
Station 2 was 88.6 percent with a range of 24-100 percent. At Station 6 the
average was 87.3 percent with a range of 30-100 percent. Table VI-4 presents
the average monthly relative humidity at both sites for 1972.
VI-8
-------
SOLAR ENERGY AND PAN EVAPORATION
Methods
Total solar radiation was measured
throughout the study period with a Bel-
fort Instrument Company Pyrheliograph
atop a standard instrument shelter at
Station 6 (Big Cypress Bend). Charts
were read weekly and the total solar
radiation reported daily.
Table VI -3. Average monthly
temperatures at Stations 2
and 6 for 1972.
Temperature ( C)
Month Station 2 Station 6
Pan evaporation was measured at
Station 2 and Station 6. A standard
1.83 meter (6 ft) diameter stainless
steel evaporation pan was used in
conjunction with a Stevens Type F
recorder for direct measurement of
evaporation. To minimize line shift
errors and float lag, a 41 centimeter
(16 in) diameter copper float with a
beaded float line was used. A 46
centimeter (18 in) stilling well
negated the effect of wind action on'the float.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
20.4
20.0
21.0
22.7
24.7
25.6
,1
.2
,1
26.
26.
26.
23.8
21.1
18.7
21.0
20.6
21.2
22.7
24.8
26.0
26.3
26.8
26.4
24.
22.
.5
.6
19.1
100
90
80
5 "°
3 SO
UI
> 4O
-I 30
UI
K
20
(0
MAX.
J J
1972
Figure VI-12. Weekly relative humidity range at Station 2 during 1972,
VI-9
-------
100
90
80
TO
t-
9 60
5
X 50
UJ
>
t- 40
ac. ao
20
10
MAX.
J J
1972
Figure VI-13. Weekly relative humidity range at Station 6 during 1972.
Table VI -4. Average monthly
relative humidity at Stations
2 and 6 for 1972.
Month
Relative humidity (%)
Station 2 Station 6
Results
Weekly ranges of solar energy
are presented in Figure VI-14. A
maximum daily solar radiation of 745
langleys/day occurred during the week
of July 9. The maximum average weekly
solar radiation occurred June 24-31
at 620 langleys/day and the minimum
during December 16-23 at 270 langleys/
day.
Total yearly pan evaporation
(Figure VI-15) varied from 124 centi-
meters (48,7 in) at Station 2 to 155
centimeters (61 in) at Station 6,
Maximum monthly evaporation at both
stations occurred during April and
May and minimum evaporation during
November and December. Station 6
is located at Big Cypress Bend
approximately 98 meters (100 yards)
south of U. S. 41. The site is in the
open and there are no immediate trees to
provide relief from the sun. Station 2, located some 49 meters (50 yards) west
of S. R, 29 at the Florida Department of Transportation Prison, received partial
shading during portions of the day, The difference in the two pan evaporation
rates is probably the consequence of shading. Nevertheless, the average pan
evaporation for the two sites was 24.06 centimeters greater than the average
precipitation within Fahkahatchee Strand for the same period (1972).
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
87.8
82.7
81.5
81.8
83.3
94.4
92.7
95.2
96.0
92.3
90.7
85.4
89.1
85.5
84.2
84.2
85,5
90.0
88.0
89.9
89.5
87.9
91.2
82.2
VI-10
-------
800
< TOO
£ 600
CO
Z 500
_l
^ 400
t| 300
a
a: zoo
a:
j 100
o
to
MAX.
J j
1972
1973
Figure VI-14. Weekly solar energy range at Station 6 during 1972.
Pan evaporation is directly
related to solar energy, wind speed,
temperature, and humidity. Figure
VI-16 presents pan evaporation and
solar energy versus time. As expected,
pan evaporation varies directly with
total solar energy. For the study
area, pan evaporation may be approxi-
mated by the following equation based
upon Figure VI-16:
Pan evaporation (cm/day) = (1.25 x 10~3)
(Solar energy (langleys/day)).
Refinements may be made to the above
equation by applying wind speed,
temperature and humidity factors.
These complexities will be cultivated
in the following discussion of evapo-
transpiration.
EVAPOTRANS PIRATION
£ i«o
UJ
STATION 6
BIS CYPRESS BEND
I.QO
l«
1.00
o.™
OH
on
0
STATION 2
OERMTMENT OF TRANSPORTATION
a 0
raifllr
[
r
r
r
r
r
r
N
r
rt
D
Figure VI-15. Monthly pan evapora-
•tion at Stations 2 and 6 during
1972.
Water is seemingly abundant in
south Florida, with large annual
average rainfalls and ground water
elevations near ground surface. However, expanding populations and agriculture
are demanding fresh water at ever increasing rates. In order to ascertain the
quantity of water any particular area can yield, several parameters must be
resolved. Rainfall readily passes into the shallow water table in areas
covered by sand, but in areas where the surface is covered by more impervious
substances such as marls or limerock much of the rainfall is lost by runoff.
VI-11
-------
10
0.1
0.6
0.4
0.2
0
1.0
01
_ o.i
g 0.4.
§«
— o.
z
o
t-
O.f •
0.
i.o
O.I
o.i
0.4
o.t
0.
JAN 72
SOLAR ENERGY
EVAPORATION
FEB 72
SOLAR ENERGY
MAR 72
SOLAR ENERGY
\
too
600
400
200
0
200 n,
-<
600>
200
0
BOO
600
400
600
II 13 19 IT 19 21 23 25 2T 29 31
APR 72
MAY 72
JULY 72
LO
a«
o.c
0.4
0.2
0
SOLAR ENERGY
600
•600
400
200
0
SEPT 72
13 19 IT 19
DEC 72
Figure VI-16. Solar energy and> evaporation in Fahkahatchee Strand during 1972,
VI-12
-------
A portion of the rainfall which reaches the aquifer then meets demands from
adjacent areas by hydraulic gradients to drainage canals. Another demand to
the rainfall is evapotranspiration. This term has been coined to be the sum
of evaporation and transpiration.
Evaporation is the movement of water back into the atmosphere from
surface areas. These surface areas include open water, rock, vegetation and
soil. When the ground water table is near the surface the evaporation from
soil is greatly enhanced due to capillary rise of the water and is almost
equal to evaporation from a free-water surface.
Transpiration is the return of water to the atmosphere by moving up
through and out from plants. Thus, the sum of the terms become evapotranspi-
ration, which may be interpreted as being the water moved back into the
atmosphere from moistened surfaces and by movement through plant cells.
Theoretical Methods
Numerous studies have been conducted concerning evapotranspiration and
its relation to climatic parameters and vegetation types. This report will
compare several evapotranspiration estimating methods to field evapotranspira-
tion measurements.
One of the most complete analysis was made by Penman (1948). His long
and tedious formula for potential evapotranspiration is based upon several
climatic parameters and is as follows:
Et = AH+0.27 Ea
A +0.27
with H = RA(l-r)(0.18+0.55n/N)-°Ta4(0.56-0.092 /e^) (0.10+0.90n/N)
and Ea = 0.35 (ea-e
-------
A = slope of saturated vapor pressure curve of air at mean monthly
temperature in mm Hg/° F.
Penman's coefficients were determined for a rather humid area, not far
from the ocean, covered with abundant growth. However, the method does not
include rainfall data, but assumes that the ground is completely shaded by
vegetation and never short of water.
Thornthwaite (1948) attempted to develop a simpler expression of
potential evapotranspiration assuming ideal conditions of soil moisture and
vegetation. He developed a monthly heat index related to monthly mean tempera-
ture and latitude. This index is then used to compute monthly evapotranspira-
tion. Thornthwaite found that his computed potential evapotranspiration is an
excellent index to stages of plant growth in that a given crop variety requires
normally a certain amount of water.
Four experimental watersheds located in the southern Florida flatwoods
were investigated (Stewart and Mills, 1948) in order to apportion runoff,
seepage losses, evapotranspiration, changes in soil moisture and ground water
storage from weather records on a watershed basis.
Stewart and Mills computed watershed evapotranspiration from the relation-
ship of:
Et = P-R-iAS
where P = precipitation
R = runoff
AS = change in storage.
Realizing that the maximum evapotranspiration is limited by solar radiation
when soil moisture is not limiting, and that rainfall limits actual evapo-
transpiration since it limits soil moisture they then related* evapotranspira-
tion to pan evaporation and rainfall. From the four experimental watersheds
the following equation was derived:
Et = Ep(°-081+0-133p)
where E = evapotranspiration in inches/month
E_ = pan evaporation in inches/month
P = precipitation in inches/month.
Table VI-5 presents a comparison of monthly mean pan evaporation (2 sites)
and precipitation (6 EPA sites) for Fahkahatchee Strand to computed potential
evapotranspiration. The Penman method estimates the potential evapotranspira-
tion to be 148.5 cm/yr or 89.08 percent of the pan evaporation. Thornthwaite
and Stewart methods yielded quite similar yearly results.
Figure VI-17 is a graphical representation of the information given in
Table VI-5. Stewart's method is shown to follow closely the monthly
precipitation while Thornthwaite and Penman methods follow more generally the
monthly pan evaporation.
VI-14
-------
Table VI-5. Computed evapotranspiration compared to pan evaporation and
precipitation in Fahkahatchee Strand during 1972. (cm I^O)
Month
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Pan
Evaporation
14.1
11.5
17.4
19.0
19.6
15.9
16.0
13.9
12.4
12.2
6.5
8.2
Precipitation Penman Thornthwaite
Method Method
2.9
7.7
7.2
4.1
6.6
29.0
16.8
24.5
11.4
6.8
13.5
3.0
7.8
7.9
14.0
14.9
17.1
15.5
17.4
14.4
13.9
11.9
7.4
6.3
6.3
4.4
6.3
9.0
11.3
12.8
13.9
13.6
12.0
9.5
7.8
11.0
Stewart
Method
3.3
5.5
8.0
5.6
8.4
25.6
15.4
19.1
8.4
5.3
5.2
1.9
Avg. of
3 Methods
5.8
5.9
9.4
9.8
12.3
18.0
15.6
15.7
11.4
8.9
6.8
6.4
Total
7o of pan
evapora-
tion
% of
precipi-
tation
166.7
100%
124.87
133.5
80.08
100
148.5
89.08
111.24
117.9
70.73
88.31
111.7
67.01
83.67
126.0
75.58
94.38
Various methods of approximating
evapotranspiration as presented
above certainly have their use-
fulness. Climatological infor-
mation is generally sufficient
for any particular locality to
make the computations. However,
variations in consumptive use
occur from season to season or
from day to day based upon para-
meters other than climatic
conditions.
Plant physiology and
morphology are factors governing
consumptive use. Figure VI-18
presents a typical consumptive
use and pan evaporation rates
during a growing season. The
consumptive use is shown to
increase, reach a peak rate and
then diminish. The peak rate,
UJ
Ul
PAN EVAPORATION
•»•
THORNTHWAITE
J F M * M J
0 i N 0
Figure VI-17. Graphical representation
of actual rainfall and pan evaporation
in Fahkahatchee Strand to three com-
puted evapotranspiration methods.
VI-15
-------
uj '
1.00
cn
tr
.so
UJ
.25
UJ
0
PAN EVAPORATION
CONSUMPTIVE
USE OF BARLEY
JUL
AUG
DATE
SEP
season of peak rate, and period
to maturity varies among types of
vegetation.
Variations in soil types,
nutrients and water availability
also control consumption.
Stewart and Mills (1967) worked
with evapotranspiration of
bermuda grass and St. Augustine
grass in controlled conditions.
Waterproof tanks were used with
controlled depth to water and
controlled sod cover. During a
five year period there were nine
separate months of less than one
inch of rain per month. Based on
the data for these months the evapotranspiration for sod crops with 0.91 meter
(36 in) water table (below surface) was found to be about 88 percent and 78
percent of the evapotranspiration for 0.61 meter (24 in) and 0.30 meter (12 in)
water table, respectively, during relatively rainfree months. Further, plant
density studies yielded the equation:
Y = 0.56X + 44
where Y = percent of full sod cover evapotranspiration
X = percent of sod cover.
Figure VI-19 shows this linear relationship. Note that the relationship
does not hold near the zero percent sod cover conditions.
Figure VT-18. Consumptive use of water
by barley and pan evaporation during
a typical growing season (from Todd,
1959).
= 0.56X+44
R. B. Sleight (1917) compared
soil types to evaporation (Figure
VI-20). Capillary rise in fine
soils can have a large effect on
evaporation and can provide water
from the water table to the
plants for transpiration.
Results
Direct measurement of evapo-
transpiration in Fahkahatchee
Strand was attempted by two field
techniques. The first of these
used ground water table fluctua-
tions as a measurement technique.
Figure VI-21 is a tracing of the
ground water well graph for EPA
Well W-l (Figure VI-1) located
near Alligator Alley above the
Strand. The graphical scale has
been expanded to accentuate the
effect of evapotranspiration on the ground water table. In this method the rate
100
;=• 80
60
40
20
20 40 60 80-100
PERCENT SOD COVER
Figure VI-19. Relationship of percent
of full sod evapotranspiration to
percent sod cover by Stewart and
Mills (1967).
VI-16
-------
of decline of the ground water
table is established by night
hours, approximately 1900 to
0700. Fluctuations from this
rate during the day then are
due to rainfall and/or evapo-
transpiration assuming that
other hydraulic parameters
remain constant. As shown
by Figure VI-21, tangents to
the curve were drawn for the
nighttime hours of April 2 to
April 3. Without rainfall or
evapotranspiration these two
tangents would coincide. For
April 3 the evapotranspiration
amounted to 1.52 centimeters.
In order to arrive at the actual
amount of water that was given
up from the water table the
above value must be adjusted
for the void ratio of the soil.
A void ratio of 0.7 was chosen
as being typical for sandy
marl. This void ratio equates
to a porosity of 0.41. The
actual evapotranspiration for
this date then was 0.41 x 1.52
or 0.62 centimeter.
en 3.0
or
§
o
Q_
o
to
FINE SANDY LOAM
0 25 50 75 100 125
WATER TABLE DEPTH BELOW SURFACE (CENTIMETERS)
Figure VI-20. Comparison of soil types
to evaporation by Sleight (1917).
z
o
O
Z
0800 P200 1600 2000 OOOO O400 08OO 1200 I60O 2OOO OOOO O40O
APRIL 2, 1972 APRIL 3,1972
TIME (HOURS)
to
30
10
• WELL LOGS
THEORETICAL
iLJ
JO
1972
Figure VI-21. Graph of ground water levels
in EPA Well W-l showing effects of
evapotranspiration.
Figure VI-22 presents a
comparison of evapotranspira- gf
tion from well logs to average
monthly theoretical estimates
from Table VI-5. The well
evapotranspirations are based
upon EPA Wells W-2 and W-3
adjusted for porosity. Well
evapotranspiration as shown
is generally higher than
theoretical evapotranspirations
except for two areas on the
graph. The first of these
areas is mid-April through mid-
June and the second area is mid-
August to early December. In both cases well evapotranspiration falls below
theoretical. The first period (April to June) can be explained by Figure
VI-23. During the period in question the vegetative covering would have to
extend 80 to 120 centimeters below the ground surface to reach the ground
water table and draw from it. Capillary rise reduces this depth somewhat.
Based upon Taylor's relationship (1948):
Figure VI-22. Comparison of average evapo-
transpiration from EPA Wells W-2 and
W-3 to average monthly theoretical
estimates from Table VI-5.
VI- 17
-------
2.4
where
K =
<
capillary rise in
cm
permeability in
cm/sec.
I-en w
0:0 h
LU2 ^
6ROUNO WATER TABLE
J J
1972
Figure VT-23. Relationship of capillary
water surface to ground water table in
Fahkahatchee Strand during 1972.
The soil at both well
sites is classified as fine
sand with intermittent layers
of muck. From Leonards (1962),
who reported permeabilities of
various soils, a typical value
for permeability of 1 x 10"3
cm/see was chosen. Using this
value the approximate capillary rise would be 10 centimeters.
The second field measurement technique for evapotranspiration attempted
in Fahkahatchee Strand was by means of humidity sensors. The discussion of
this technique was presented in Chapter II. Data received from this technique
is extremely limited and variable as depicted by Table VI-6.
Table VI -6. Comparison of prairie grass evapotranspiration measured by
humidity sensors to pan evaporation in cm/day during August, 1972.
Date
Evapotranspiration
Pan Evaporation
2 (August)
3
4
7
8
9
10
11
14
15
23
0.53
0.86
0.77
0.82
0.82
0.44
0.45
0.70
0.57
0.55
0.44
0.78
0.82
0.60
0.42
0.60
0.60
0.46
0.46
0.42
0.32
0.53
PotentialEvapotranspiration
The Penman (1963) method of computing potential water export attempts to
evaluate the effect of radiation heat gain and the drying power of local air
mass flow on evapotranspiration. The method assumes that the transport of
sensible heat is controlled by temperature differences in the same way as
transport of water vapor is controlled by vapor-pressure differences, arriving
at an expression for water export of
E =
x
VI-18
-------
where Ea involves wind speed and saturation deficit, A is the slope of the
saturation vapor-pressure curve at mean air temperature, H is the heat budget
in terms of mm of water vaporizable, and y a constant of the wet and dry bulb
psychrometer equation.
This equation can be reduced to
E = AH + 0.27 Ea
A + 0.27
The equation for heat budget involves a balance of input solar and cloud
radiation with back radiation, as a function of cloudiness, vapor pressure,
and temperature.
As we used pryheliometers to directly measure total radiation input, the
cloudiness factors can be eliminated from the equation to give:
H = R(l-r) -
-------
Table VI -7. Climatological data for 1972, Fahkahatchee Strand area.
R is the mean monthly radiation input in langleys/day, measured by a
pyroheliometer (Belfort Inst. Co.) at the coastal station (6). U is
the mean windspeed in kilometers per hour (provided by B. J. Yokel,
Rookery Bay Project).
Month
Jan.
Feb.
Mar.
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
R
ly/day
450 (est.)
468.3
503.0
496.7
539.4
498.7
535.9
455.9
438.2
429.7
330.6
315.7
u
kph
13.97
11.33
15.59
13.40
12.48
11.60
9.35
9.04
10.06
11.83
7.69
12.74
e
mm Hg
25.36
22.31
26.66
29.02
32.47
31.82
33.12
33.60
33.60
30.48
25.21
21.07
e
16.34
13.70
15.81
17.84
19.55
24.46
24.51
26.51
26.42
21.35
18.07
13.53
T
°C
26.1
24.0
27.0
28.4
30.2
30.0
30.7
31.0
31.0
29.2
26.0
23.0
Potential
E-T
mm H-O/mo
144.8
123.1
177.6
166.8
198.6
161.7
180.4
152.3
144.1
144.9
79.0
82.4
Pan
evap.
mm H20/mo
142.9
123.4
172.6
191.9
202.0
154.6
160.0
139.4
125.8
120.0
69.0
81.2
pressure, are in millimeters of mercury; eg was derived from standard tables.
Potential evapotranspiration was calculated by the Penman method and is in
millimeters of water/month; pan evaporation, in the same units, represents the
mean monthly evaporation from one pan at each station.
Comparison of potential evapotranspiration to pan evaporation by least
squares regression for the months of February through December yielded the
equation
PAN =1.03 (potential) - 11.2
with P <0.0001. The yearly totals from the two methods are in good agreement
as well. Pan evaporation totaled 1,682.8 mm, and total potential evapotrans-
piration was 1,755.7 mm.
Actual Evapotranspiration
Calculation of actual evapotranspiration from test well data requires a
knowledge of the specific yield of the soil. One method of evaluating this
characteristic of the soil utilizes a measured rise in the ground water table
in response to a known input of water over a relatively large area (Todd, 1959).
Paired test well and rainfall recording provided a means of estimating specific
yield (Sy) by an analogous procedure. In Figure VT-24, all rainfalls greater
than 0.5 inch and their associated water table rises are used to provide
estimates of Sy. These estimates are plotted against the shallowness of the
water table, -2. Perturbation of the estimate occurs at shallow water table
depths due to capillary rise and a lack of infiltration, leading to high
VI-20
-------
i.o
0.9
0.8
0.7
0.6
0.5
04
0.3
0.2
O.I
o 2.08
1.80
EXPLANATION
a EPA WELL W-l
o EPA WELL W-2
o USGS WELL C-496
25
5O
75
Z ( cm)
100
125
ISO
Figure VI-24. Estimated specific yield as a function of the depth of the
ground water table.
estimates. Similarly, when the water table is far below ground, transpiration
export of soil water reduces soil water content below field capacity. The
measured estimate of specific yield is again perturbed, again resulting in
artificially large values. The vest estimate of Sy from this figure, there-
fore, must be derived from values at an intermediate range of water table
depth. In this range, most values cluster between 0.1 and 0.2, thus 0.15 was
chosen as an approximate value of Sy for use in estimating consumptive water
loss from test well data.
Estimation of consumptive water use from the well log data required
fractionation of the data into three segments. Diel patterns of recession
rates estimated water export as evapotranspiration, exclusive of water derived
from field capacity. In the former case, water table changes when the water
table was above ground level directly estimated evapotranspiration. Diel
changes in recession rate when the water table was below ground level were
corrected by the Sy (0.15) to estimate evapotranspiration. Field capacity
recharge, the third component of evapotranspiration, was estimated by assuming
that all rainfall would infiltrate completely when the water table was below
ground level, and rainfall that failed to appear as a water table rise (using
Sy = 0.15) represented field capacity recharge. The result of this procedure
for four test wells is given in Table VT-8. The total evapotranspiration
estimated by this procedure amounts to 1,100.2 mm of water, some 76 percent
of total rainfall. The large peaks in March and June represent major field
capacity recharge peaks by rainfalls following extended dry periods.
VI-21
-------
Table VI -8. Evapotranspiration calculated from test well data (mm H.,0)
Well
Month
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
W-4
23.3
103.0
99.7
70.2
168.3
116.1
54.3
74.7
49.8
39.7
38.5
C-496
0
90.2
373.9
116.5
91.8
286.3
217.2
153.9
76.2
274.3
39.9
W-3
19.9
71.1
55.9
94.9
206.0
144.0
112.0
101.8
W-l
20.1
27.5
61.8
32.5
46.9
178.1
63.2
55.6
118.0
134.5
-1.6
32.5
Mean
10.0
40.2
152.4
76.2
75.6
209.6
135.1
94.0
92.7
152.9
26.0
35.5
1,100.2
VI-22
-------
VII HYDROLOGY
INTRODUCTION
Perhaps the most discussed aspect of land development within south
Florida is the effect on the natural hydrology. Due to the flatness of
terrain -- gradient approximates 3 cm/km (2 inches per mile) in the Everglades
system and 8 to 16 cm/km. ( 5 to 10 inches per mile ) in the Big Cypress
Swamp -- surface water moves at a very slow pace. The flatness of the ground
and the frictional effect of dense semi-aquatic vegetation combine to produce
a southerly water flow rate of less than 0.8 km/day (0.5 miles per day) in
the Big Cypress Swamp and less than 0.5 km/day (0.3 miles per day) in the
Everglades. This flatness and subsequent slow runoff rate necessitates
alterations to the natural system in order for man to exist and dominate in
a system unnatural to his liking. Man's basic alterations consist of dredge
and fill. The dredging procedures create drainage canals to remove surface
and subsurface water at a more rapid than normal rate. The drainage canals
often increase the flow velocity approximately one order of magnitude. Fill
operations create topographic relief that expedites runoff to the canals,
decreases the area of potential inundation and interrupts "sheet flow."
The network of canals within the Big Cypress Swamp is very efficient
at producing results, i.e., lowering of the ground water table. The U. S.
Geological Survey (USGS) (Klein, et al., 1970) found that water levels have
been lowered 61 to 122 centimeters (2 to 4 feet) in a 54 square-mile area
east of Naples.
UPLAND HYDROLOGY
Methods
Stevens Type A Model 71 stage recorders were used to provide continuous
records of stage at both surface and ground water stations throughout the
study area. Rating curves of flow versus stage were developed for the total
flow crossing Janes Scenic Drive at some 34 stations consisting of 8 bridges
and 34 culverts. Measurements were generally made on a monthly basis or
whenever there was a sufficiently significant increase or decrease in stage
to warrant a measurement. The rating curve is inclusive for all stations
located within line sigment 1-4 on Figure VII-1.
An additional rating curve was developed for a measurable cross section
of Fahkahatchee Strand which included a portion of Janes Scenic Drive and a
tramway known locally as West Main shown as line segment 2-3 on Figure VII-1.
This rating curve consisted of flows at one bridge and 22 culverts on Janes
Scenic Drive, 4 bridges and 1 culvert on West Main in addition to flow around
the western end of West Main. Information obtained from this section was of
prime value in determining areas of inundation in the southern end of
Fahkahatchee Strand.
Alligator Alley rating stations consisted of two culverts and three
bridges located west of SR 29 and east of the eastern-most GAG canal.
Stations A, B and C (Figure VII-1) were used to measure direct ground
water contributions to a portion of the GAG canals. Station A is located
VII-1
-------
JL
0 ' 2 REMUDA RANCH-*
KILOMETERS
TIDAL
RECORDER
EXPLANATIO
A WELL
• SURFACE WATER RECORDER
• STATION LOCATION
Figure VII-1. Map of Study Area Showing Hydrology Stations,
VII-2
-------
at the intersection of Janes Scenic Drive and the eastermost GAG canal.
Stations B and C are located 1.61 and 5.23 kilometers (1.0 and 3.25 miles)
north of Station A.
Results
X
UJ
o
I
o
en
Q
UJ
Surface Water
As discussed in Section VI the year 1972 would be described as an above
average year for rainfall. However, surface water flows and ground water
levels appear to be less than average.
Total flow for the _
year amounted to approxi-
mately 357,712 cubic meters
(2,900 Ac-Ft) with a maximum
daily discharge of 1,189
^liters/sec (42 cfs) being
observed on February 12
(Figure VII-2). During 65
percent of the year there
was no measurable flow.
No flow occurred from
February 24 until August 28,
a time lapse of over six
months. The average daily
flow for the year amounted
to less than 113 liters/sec
(4 cfs). Partially due to
the low flow characteristics
throughout the year a profuse
growth of aquatic ai*d semi-
aquatic vegetation has
occurred at all bridges.
This vegetative growth has
progressed to the point that
it inhibits much of the
natural flow north of
Alligator Alley from entering
Fahkahatchee Strand. During
1972, less than 17 percent
of the flow crossing Janes
Drive can be attributed to
input from north of Alligator
Alley.
o
O
CO
1.5
1.0
„
UJ
tr
ui
J
1972
Compared to recent
years, a minimal amount
Figure VII-2. Hydrograph and Cumulative
Discharge for Alligator Alley Between
SR 29 and the easternmost GAG canal.
of flow crossed Janes
Scenic Drive during
1972 (Figure VII-3). For the years 1970, 1971 and 1972 a total of 194,568,240
cubic meters, 111,661,680 cubic meters, and 20,112,054 cubic meters
(157,738 Ac-Ft, 90,525 Ac-Ft and 16,305 Ac-Ft), respectively, crossed Janes
Scenic Drive. No measurable flow passed Janes Scenic Drive from March 10
until August 13 (Figures VII-4 and 5). .A peak flow of 7,419 liters/sec (262 cfs)
occurred on November 12 and 13.
VII-3
-------
18
^ 16
UJ
X
O
V)
O
14
12
10
S
I
NJMMJ SNJMMJSNJMMJSN
1970 1971 1972
Figure VII-3. Hydrographic of iy[onthly Mean Discharge of Flows Passing
Janes Scenic Drive.
Figure VII-6 reveals the special relationship of Fahkahatchee Strand
to the major drainage ,canals in central Collier County. Records from the
USGS and aerial photographs indicate that at one time the Okaloacoochee
Slough was hydrologically tied to Fahkahatchee Strand. These ties must
certainly have been altered by the drainage canals shown on this figure.
The SR 29 borrow canal (Barron River) intercepts and diverts major flows
away from the Strand. SR 84 (Alligator Alley) borrow canal intercepts
upland flow; however, the canal "dead ends" on either side of the Strand.
Thus, this canal intercepts flow, but various structures along its reach
allow passpage of flows into the Strand. The GAG canal system to the west
probably intercepts little of the surface water flow to Fahkahatchee Strand,
These canals, however, intercept surface flows to other areas to the south-
west and they also draw heavily on the ground water (discussed later in
this Section).
Massive quantities of water are transported from the inland areas to
Fahka Union Bay via the GAG system of canals. The canals drain approxi-
mately 81,000 hectares (200,000 acres) in Collier County. Fahka Union Bay
receives direct drainage from as far north as Corkscrew Swamp, some 51.5
kilometers (32 miles) away. The total flow of Fahka Union Canal at Remuda
Ranch decreased to 177,263,000 cubic meters (143,708 Ac-Ft) in 1972 from
VII-4
-------
cr
o
3
rt
ro
0
3
09
rt
i-i
3
P
<
ro
i-t
P
OQ
ro
•
^o
P
H.
p
HI
P
h- '
t-1
CL
C
H-
3
I — t
V£)
-^1
to
5!
P
CO
h-*
Ul
^j
0
t— '
n
ro
3
rt
H-
3
ro
rt
ro
f-t
cn
an extreme 1
City was 11
V0 *
ro i—1
H
co CL
- C
i-i
H H-
ro 3
01 OQ
T3
ro i—1
O VD
rt — i
H' O
<^
ro P
I-1 3
•
TJ
0
G
i—1
P
rt
H-
0
3
•P-
-vj
B
H-
1— >
f«i
H-
0
3
"V
ro
i-i
CO
0
3
CO
P
3
^— *
•
CO
Ul
3
H-
I—1
l-»
H-
O
3
13
m
i-i
CO
o
3
CO
H'
3
i— «
^o
— J
0
v»
1 — 1
"O
-J
L__l
V*
P
3
CL
the dischar
^D
^J
O
p
p
I-1
**Q
^j
1— '
w
i-i
fD
CO
13
ro
o
rt
H-
-------
*2
x
"!§
K
X
0
Q
UJ
L
P^
O
24
22
20
18
16
14
12
10
e
6
4
2
0
J A
1972
Figure VII-5. Cumulative Discharge Passing Janes Scenic
Drive.
or 19.9 centimeters above average. With a decrease in surface water dis-
charge it would appear that ground water levels should increase during
1972. However, as will be seen later in this section, ground water levels
decreased instead of increasing during 1972.
Much of the flora and fauna of the Big Cypress Swamp depend upon the
hydroperiod for existence and for reproduction. Many of the amphibians
require inundation of the area before previously laid eggs will begin to
hatch. During the rainy season much of the swamp becomes inundated and
remains so for several months following a normal rainfall year. Approximately
10 percent of the area remains inundated year round.
A large portion (Figures VTI-8 and 9) of the study area remained rela-
tively dry throughout 1972. Approximately 37 percent of the study area
above US 41 was inundated during January and February. The inundated area
decreased to 20 percent in March and 4 percent in April. During May, June,
July and August there were no significant areas inundated within the
Fahkahatchee Strand study area. The only areas which contained water were
the canals and a few scattered ponds within Fahkahatchee Strand. Following
the heavy rains of June the interior of Fahkahatchee Strand became inundated
to a depth of approximately 12 centimeters (4 in ). However, some ten days
later the water had receded below the ground surface and remained so until
August 27. The area of inundation increased rapidly during September and
peaked during November. A slight decrease occurred in December. The percent
of the study area inundated during September, October, November and December
was 35, 45, 51, and 45, respectively. Because of the inaccessability of
much of the area north of US 41 and south of Janes Scenic Drive the area of
inundation is not completely delineated in Figures VII-8 and 9.
VII-6
-------
I Hendry Co
„ I <;'Trafford
Lee Co. l,'/,-^,
n7P/-''%4
1 /. >' *-J <-, x /
)> Kissimmee Billy
>'v Strand »..
II Willson
."• Strand
Skillet
>' Strand/J Monroe
' Strand
Naples
Monroe Co.
41
Fibure VII-6. Major Drainage Canals Near Fahkahatchee Strand.
Groundwater
The seven hydrographs in Figures VII-10 to 15 represent variations in
ground water levels throughout the study area. The seven well sites are
representative of the study area above US 41. The ground elevation presented
in each figure is the average of four to six elevations obtained within
100 meters of the well site.
EPA Well No. W-l (Figure VII-10) is located at the northwest corner of
the study area approximately 152 meters(500 ft) south of Alligator Alley.
With the exception of the first 23 days of November 1971 the ground water
was below surface. The ground water level continued to fall until the first
of June when, due to the increased rainfall, the level began to rise. The
maximum drawdown during 1972 was 1.48 meters (4.86 ft). A net decrease in
water level of 0.36 meters (1.19 ft) occurred during 1972. The maximum
distance below the surface was 1.48 meters (4.86 ft) on June 1, 1972.
The water level at USGS Well C-296 (Figure VII-11) was affected by the
operation of stop-log controls which regulate flow in the Barron River Canal
VII-7
-------
some 30.5 meters (100 ft) to the east. The hydrograph exhibited the same
general characteristics as did EPA Well W-l with the exception of a net
increase in water level of 0.29 meters (0.94 ft). The water level draw-
down for the year was 1.26 meters (4.13 ft).
Wells W-2 and W-3
(Figure VII-12) are
located at the western
edge of Fahkahatchee
Strand near the junction
of Janes Scenic Drive
and the easternmost GAG
canal. Well W-2 is
approximately 30.5
meters (100 ft) east
of the canal and W-3
some 320 meters (1,050
ft) east of the canal.
Due to ,the drainage
effects of the canal,
the water level at W-2
generally remains 0.12 -
0.18 meters (0.4 - 0.6
ft) below the level at
W-3. In late October
1972 the GAG Properties,
Inc. began construction
of a bridge at the inter-
section t>f Janes Scenic
Drive and the easternmost
GAG canal. In prepara-
tion for the footings of
the bridge the canal was
diverted and a sump was
dug at the canal site.
Several large pumps were
installed in order to
remove ground water from
the sump. On October 30,
1972, at 1300 hours the
water level in the sump
was 1.54 meters (5.06
ft) below MSL. Ground
water elevation at W-2
was 0,43 meters (1.41
ft) above MSL and W-3 at 1.27 (4.17 ft) above MSL. Level of water in the
diversion canal immediately adjacent to the sump was 1.34 meters (4.38 ft)
above MSL. Due to the pumping effect, Well W-2 had a net decrease in water
level of 0.90 meters (2.94 ft) during 1972. The net decrease at W-3 was
0.41 meters (1,34 ft). The maximum elevation observed at W-2 was 2=23 meters
(7.30 ft) on June 18 and a minimum of 0.27 meters (0.90 ft) on November 12.
By observing the vegetation in the immediate vicinity of W-2 and W-3 it can
be concluded that this area was frequently inundated in the past. The area
was inundated during a portion of late summer in 1971. Visual inspection
i
o
(O
.o
(T
UJ
001
Figure VII-7. Hydrograph of Monthly Mean Discharge
of the GAG Canal System at Remuda Ranch.
VII-8
-------
Figure VII-8. Inundated Areas of Fahkahatchee Strand From January - June 1972
(Cross-Hatching Represents Areas of Standing Water).
VII-9
-------
Figure VII-9. Inundated Areas of Fahkahatchee Strand From July - December 1972.
(Cross-Hatching Represents Areas of Standing Water.)
VII-10
-------
5J37S
CO
IE
UJ
Ul
>3.00
< 275
GROUND ELEVATION 3.95 m
during that time revealed
that the easternmost GAG
canal was intercepting
surface water from the
Fahkahatchee Strand side
of the canal. As design-
ed, the canal very rapidly
removed the surface water
and lowered the ground
water table. During 1972
the water table was always
lower than 0.54 meters
(1,77 ft) and 0.49 meters
(1.62 ft) below the ground
elevation at W-2 and W-3,
respectively.
USGS Well C-496 is
located on Janes Scenic
Drive near the center
of Fahkahatchee Strand.
The Strand at this site
(Figure VII-13) is
generally inundated
with the exception
of extreme dry periods.
During 1972 the area
was not inundated
from April to September
except for several days
in July following heavy
rains. On June 1 the
ground water receded to
0.90 meters (2.96 ft)
above MSL some 0.79
meters (2.60 ft) below
the surface. The main
slough of Fahkahatchee
Strand was inundated
from January through
May with a maximum water
depth of 0.36 meters
(1.18 ft) on January 1. The slough was again inundated from August 27 until
the end of the year with a maximum depth of 0.44 meter (1.45 ft) on November 18.
A net decrease in water level of 0.05 meter (0=16 ft) occurred during the
year.
EPA Well W-4 is located on Janes Scenic Drive near the eastern edge
of Fahkahatchee Strand. The ground water levels (Figure VII-14) exhibited
the same general characteristics as USGS Well C-496. However, inundation
did not occur until August 26 with a maximum depth of 0.09 meter (0.30 ft)
on September 1. The maximum depth to the water table below the surface was
1»16 meters (3.79 ft) on June 1.
NDJFMAMJ J A 3 0 N D J
1972
Figure VII-10. Hydrograph of EPA Well W-l.
GROUND ELEVATION 3.60 m
3.28
3.00
I
J J
1972
Figure VII-11. Hydrograph of USGS Well C-296.
VII-11
-------
EPA Well W-5 (Fig-
ure VII-15) is located on
an access road to the
Remuda Ranch pumping
station at the same
latitude as Well W-4.
The area is classified
as a wet prairie (low
lying areas of Ochopee
marl or Ochopee fine
sandy marl consisting
mainly of short grasses).
However, during 1972 the
area was anything but a
wet prairie. The prairie
was inundated for approx-
imately 25 days during
portions of June, August,
and September with a
maximum depth of 0.11
meter (0.35 ft) on
June 18. The minimum
level of the ground water
table was 0.17 meter
(0.55 ft) above MSL or
1.02 meters (3.36 ft)
below ground on May 1.
A net increase in water
level of 0.16 meters
(0.51 ft) occurred dur-
ing the year. With
ground water levels
approaching MSL at this
location the possibility
GROUND ELEVATION AT WELL No W-2 2.76m
2 75
2.50 •
Z.25 -
^ 2.00 •
V)
or
LU
I-
UJ
UJ
UJ
_l
St.
UJ
1.75
1.50 •
1.25 •
LOO -
0.75 •
0.50 -
0.25 -
Figure VII-12. Hydrograph of EPA Wells W-2 and W-3.
07S
Figure VII-13. Hydrograph of USGS Well C-496.
of salt water intrusion
exists in the area. Assum-
ing that ground water
levels do not increase
downstream from this site,
the expected interface
between fresh and saline
water on May 1 would occur
some 6.8 meters (22.0 ft)
below the surface, based up-
on the Ghyben-Herzberg prin-
cipal that for each foot of
water which occurs above MSL
40 feet of fresh water
occurs below MSL.
The elevation of the
top of the weir on the
VII-12
-------
Fahka Union Canal immedi-
ately downstream from USGS
surface water gaging sta-
tion 2-2911.43 (Figure
VII-1) is 0.61 meter (2.0
ft) above MSL. EPA Well
W-5 is 2.1 kilometers
(1.3 miles) east of the
body of water impounded
by the weir. As observed
by Figure VII-15 the ground
water elevation at EPA Well
W-5 was less than 0.61
meter (2.0 ft) during part
of January, February, March.
May, November and December,
in addition to the entire
month of April in 1972.
With the exception of
sixteen days from April 21
to May 6, 1972, the eleva-
tion of the water surface
in the canal at USGS stage
recorder 2-2911.43 (Fig-
ure VII-1) was 0.61 meter
(2.0 ft) above MSL or
higher. The water surface
elevation in the canal fell
to a minimum of 0.52 meter
(1.71 ft) above MSL on
May 5. The water eleva-
tion in EPA Well W-5 fell
to a minimum of 0.17
meter (0.55 ft) above
MSL on the same date.
Based on a positive dif-
ferential in stage
between the canal and EPA
Well W-5 ground water
recharge may be occur-
ring during a portion
of the year.
i.so
N D J
Figure VII-14. Hydrograph of EPA Well W-4.
N 0 J
Figure VII-15. Hydrograph of EPA Well W-5.
In a report (Gee and Jenson, 1970) prepared by a consulting engineering
firm for GAG Properties, Inc., the following statement appears on page 9.
"However, a review of the proposed development south of Alligator Alley and
east of the natural divide indicates that a canal should be constructed on
the east boundary of the development to assist the natural flow of water
southward into the Strand." The results of such action would be to induce
removal of water out of the Strand at approximately one order of magnitude
faster than normal during the hydroperiod. As such, canals in Collier County
are designed, constructed, maintained, and operated to rapidly remove
surface water and lower ground water levels. However, during the dry winter
months canals may be usable as ground water recharge sources. For this to
VII-13
-------
take place would require the canal water level to be higher than the surrounding
ground water level. This would further require the importation of ground or
surface waters from upland sources, via a canal network, into the Strand.
Canals, by nature, are not self-substaining and as separate entities cannot
maintain higher than natural ground water levels. Thus, the usefulness of
canals as a recharge source for ground water in Fahkahatchee Strand is
questionable.
The merging of fresh water aquifers and the Gulf of Mexico is an
extremely important and complex system. In this study we investigated the
possible effects of the ocean dynamics at a ground water well (W-6) located
approximately two miles above US 41.
Coastal aquifers respond to the sinusoidal dynamics of the tides.
Effects of the tidal movements are proportional to the amplitude of the
tide, the duration of the tide, distance inland, storage coefficient of
the aquifer, and the coefficient of transmissibility of the aquifer.
A relationship has been developed (Todd, 1959) that yields the amplitide
of the fluctuations of the ground water table at any distance from the shore.
This amplitude is expressed as:
h = h e -Vrrs/t0T
nx n0e
where hx = ground water amplitude at distance x from shore
ho = tidal amplitude at the shore in feet
X = distance from shore in feet
s = aquifer storage coefficient
tQ = tidal period in days
T = transmissibility of the aquifer in gallons/day/foot
In a report of wells and ground water resources in Collier County (Klein,
1954), a transmissibility value of 92,000 gpd/ft and a storage coefficient
of 0.001 was developed for the shallow artesian aquifer existing at the Naples
well field. He also related in his report that the semi-impermeable stratum
between the shallow artesian aquifer and the upper non-artesian aquifer may
not exist south of Naples well field. EPA Wells 2 and 3 were used to compute
the transmissibility of the non-artesian aquifer at their site which is at
the western edge of the Strand. The transmissibility was 4.1 cm/sec based
upon three observations. Since this value is quite similar to Klein's, a
transmissibility of 4.3 cm/sec and a storage coefficient of 0.001 was used.
Figure VII-16 was developed using these values, along with a tidal period
(high to low) of six hours and a tidal amplitude or half range of 0.43 meter
(1.4 ft). As seen from this figure, tidal effects on ground water table is
rapidly dampened inland from the shore. A tidal range of 0,91 meter (3.0 ft)
produces only a 0.86 centimeter (0.34 inch) fluctuation in the ground water
table 3.22 kilometers (2 miles) inland.
Todd also demonstrated a relationship for the time lag of a given
maximum or minimum tide at the ocean to the time that this effect is seen
inland. The time lag is: tL = x/"toS/4rrTo This relationship is shown on
Figure VII-17 as a plot of time lag in hours versus kilometers inland from
shore. At 3.22 kilometers (2.0 miles) inland the time lag is approximately
3.75 hours.
VII-14
-------
Figure VII-18 is a graph of
ground water levels at EPA
Well W-5 for December 8-9,
1972. This well is located
3.22 - 4.02 kilometers (2 -
2.5 miles) above the tidal-
ly influenced portion of
the Fahka Union Canal.
This graph clearly shows a
response to the tidal
fluctuations. Responses
during daylight hours are
overpowered by evapotrans-
piration as shown by the
log for the daylight
period of 0700 - 1800
on December 8. However,
early morning responses
to flood tides are shown
to produce a range of 0.76
centimeter (0.025 ft) or
PIEZOMETRIC SURFACE
t
0.125 cm
234
KILOMETERS INLAND FROM SHORE
Figure VII-16. Net Change of Piezometric Surface
Inland From the Ocean.
2345
KILOMETERS INLAND
amplitude (h) of 0.38 centimeter
(0.0125 ft). Entering this ampli-
tude into Figure VII-16, a value of
3.86 kilometers (2.4 miles) inland
is received. Thus, the response of
the well log during low transpiration
interferences compares favorably with
the calculated response. Klein,
during pumping test of the old Naples
well field, noticed responses of the
wells to tidal movement. He reported
a time lag of 1.5 hours and a daily
fluctuation range of 6 to 21 centi-
meters (0.2 to 0.7 ft) for a well
located 0.86 kilometers from shore.
The water level in this well, however,
was influenced somewhat by well field
pumping. Figures VII-16 and 17 give
a value of 0.9 hour time lag and a
tidal range of 0.34 meters (1.1 ft)
or 0.17 meters (0.55 ft) = h,, Again,
field measured values and calculated
values favorably compare.
Figure VII-17. Time Lag of Tidal Effect
on Ground Water Levels Inland From the
Ocean.
Figure VII-19 is a graphical
representation of the dynamics of
piezometric surface. Although tidal caused fluctuations of.ground water
levels are of little significance in the overall hydrology they do create
problems in direct measurement of evapotranspiration at specific locations.
VII-15
-------
70
UJ
> 69
CD
67
66
FLOOD FLOOD
EB3 ' EBB
DEC 8, 1972
DEC 9, 1972
Figure VII-18. General Response of
Piezometric Surface to Tidal Effects.
Figures VII-20, 21, and 22 are
contour maps showing ground water
levels in Fahkahatchee Strand on
different dates. It should be
recognized that a number of ground
water gradients could be construct-
ted in Figures VII-20, 21, and 22.
The ground water gradients as shown
are based on straight line extra-
polation between the various wells.
Therefore, flow line analysis and
flow net analysis would be incon-
clusive. During November 1971
water levels were at a maximum,
decreasing to a minimum in
June 1972 and again maximizing
in October 1972. The most pro-
nounced effect of the GAG drain-
age canals on water levels is exhibited at the intersection of Janes Scenic
Drive and the easternmost canal near wells W-2 and W-3. On November 8, 1971,
(Figure VII-20) the 2.00 meter ground water contour is approximately 6.5
kilometers north of the expected location of the contour. Without the canal,
the water level at the same' latitude would be approximately 3.1 meters. In
the center of Fahkahatchee Strand the hydraulic gradient decreases approxi-
mately 0.12 m/km compared to 0.16 m/km near the canal. The same general
characteristics are evident on June 14 and October 10.
Ground water recession rates are greater in areas near the canals than
in areas distant from canals. Figure VII-23 presents ground water levels
for approximately seven days following hurricane Agnes in June 1972. During
June 16, 17 and 18, Fahkahatchee Strand received from 14.0 to 16.5 centimeters
(5.5 to 6.5 inches) of rain. Ground water levels peaked at most wells on the
afternoon of June 18. The recession rate at EPA Well W-2 was 9.1 centimeters
per day (0.30 ft/day) for the five days immediately following the peak in
ground water stage. This rate of 9.1 centimeters per day near the GAG Canal
r
GROUND SURFACE
YATER TABLE
k
- FLUCTUATION (2h) OF
0.76cm SEEN 3.75 HOURS
LATER (tL) 3.22 km
INLAND
AQUIFER
//
/
V7
* * 9
s
/
/
/
/
/
/
1 1 IL/C1U. nCIIUVJl
f\ ""7Y~7
i
OCEAN
- vc.il/ \jr O0.9*t Cm
rMEAN
V" y SEA LEVEL
NOT TO SCALE
Figure VII-19. General Response of Piezometric Surface to Tidal Response
VII-16
-------
0 I 2
N
t
_
^v^
^^^^
_l
Z
t>
0
4
^J-^:
~x
-E5 X.
0 ^^^^
? ^^-^
2 ^\_
PERS
Z
o
- X
>
jf ^
f _-! J> , !
^ —
1.78m
2.36m
^— -
l
!
A
\
V
r — \
\
0.92m
\
. N
x
X
2 SG S.^
< C* PRESS
SEND
'5?
^
A
3.98 m
\
\
\
V Sx
\
\
\ \
\
'•^0
3.88m
\
\
\ X\
U\ ^.
^A
GO
X
*»
X
EXPLANATION
A Observed ground woter
level* on November 8, 1971
Figure VII-20. Ground Water Levels in Fahkahatchee Strand on November 11, 1971,
VII-17
-------
EXPLANATION
Observed ground water
levels on May 4, 1972
Figure VII-21. Ground Water Levels in Fahkahatchee Strand on May 4, 1972.
VII-18
-------
EXPLANATION
A Observed ground water
levels on October 10, 1972
Figure VTI-22. Ground water levels in Fahkahatchee Strand on October 10, 1972.
VII-19
-------
EPA WELL W-l
RECESSION RATE-3.96 cm/DAY
24
in
a:
LU
UJ 2.0
EPA WELL W-2
RECESSION RATE-9.14 cm /DAY
UJ
>
IB
tr
UJ
I
14
EPA WELL W-4
RECESSION RATE - 2 13 cm/OAY
compares with 4.0 centi-
meters per day (0.13 ft/
day) at Well W-l near
Alligator Alley and 2.1
centimeters per day (0.07
ft/day) at Well W-4 on
Janes Scenic Drive near
the eastern side of
Fahkahatchee Strand. The
rates include evapotrans-
piration as shown by
Figure VII-23. The reces-
sion rates at EPA Well W-5
and USGS Well C-496 (Fig-
ure VII-1) were 2.1 centi-
meters per day (0.7 ft/
day).
As previously dis-
cussed, GAG Properties
began construction of a
bridge at the intersec-
tion of Janes Scenic Drive
and the easternmost GAG
canal on October 24, 1972.
Initial pumping began on
the afternoon of October 24
and continued until the
morning of October 26
(Figure VII-24). During
this period the pumping
operations lowered the
water level at W-2 by
0.61 meter per day (2.00
ft) and W-3 by 0.14 meter per day (0.45 ft). Upon cessation of pumping, the
ground water level rapidly returned to near "normal" level. The maximum
recession rate at W-2 was approximately 5.21 meters per day (17.1 ft/day
for 1.8 hours) compared with 0.20 meter per day (0.67 ft/day) at W-3.
Maximum recovery rate was 4.08 meters per day (13.4 ft/day) at W-2 and 0.26
meter per day (0.84 ft/day) at W-3. Conceivably, if pumping had continued for
several weeks the ground water level at W-2 would approach the 1.54 meters
(5.06 ft) below MSL in the pumping sump. This is evidenced by the second
pumping period which began at approximately 8:00 a.m. on October 27 and
continued until 8:00 a.m. on October 31. During this period the maximum
drawdown at W-2 was 1.34 meters (4.51 ft) and 0.22 meter (0.72 ft) at W-3.
Again, the recovery rate was very rapid upon cessation of pumping. As
observed in Figure VII-24 additional pumps ware installed on October 28.
Based on knowledge of water levels in the sump and at W-2 and W-3, the
pumpage between October 27 and 31 would have an effect on ground water levels
approximately 0.91 kilometer (3,000 ft) from the canal.
Direct measurements of ground water contribution to a selected section
of the GAG canal system (Section A-C on Figure VII-1) were made throughout
the year. These measurements were made only when there was no visible
20
21 22
JUNE 1972
24
25
Figure VII-23. Ground Water Recession at EPA
Wells W-l, W-2 and W-4.
VII-20
-------
l.6r
1.4
12
O
m
DC
UHO
UJ
UJ
£T
LU
I
0.6
04
BEGIN
PUMPING
STOP
PUMPING
• EPA WELL W-3
EPA WELL W-2
STOP
PUMPING
24
25
26 27
OCT
28
29
30
31
NOV
Figure VII-24. Effects of Pumpage on Ground Water Levels in the Vicinity of
Janes Scenic Drive and the Easternmost GAG Canal.
surface water contribution to the canal. Table VII-1 is a summary of all
measurements made during the study period.
The largest contribution of ground water occurred on October 8, 1971,
at a rate of 558 I/sec/km (31.7 cfs/mi). However, the water level recorders
had not yet been installed at the wells,and consequently, the stage is not
known. Figure VIII-25 is a graphical representation of the increase in
canal discharge per mile versus ground water stage at W-2. The mathematical
equation of Q=297H-356 is based upon the four observations near the straight
line where canal flow was unobstructed. The linear regression coefficient
(R) is 0.96. Conceivably, it would be possible for water in the canal to
recharge the ground water in time of droughts. However, for this to occur
the water in the canal would have to originate from an external source in
order for its stage to be higher than ground water stage. As observed from
Figure VII-12, the stage at W-2 (near the canal) was always lower than W-3
(0.32 km [0.2 mile} away) during low stage periods, and consequently ground
water recharge was non-existent. Only during periods of high canal flows
following rapid surface runoff from adjacent areas was the stage higher at
W-2 than W-3.
VII-21
-------
Table VTI-1. Ground Water Contribution at a Select Section of the GAG
Canal System.
Ground Water Stage at EPA
Contribution Well W-2
Date (I/sec/km) (Meters Above MSL)
Oct. 8, 1971
Dec. 1, 1971
Jan. 4, 1972
June 28, 1972
July 10, 1972
July 26, 1972
Aug. 25, 1972
Sept. 9, 1972
Nov. 2, 1972
558
118
95°
121
88
*•»
70C
49C
58
186
a
1.57
1.36
1.63
1.47
1.63
1.46
1.42
0.34
a Water level recorder not installed this date.
b Dense algae growth in canal.
c Canal partially blocked at Janes Scenic Drive (Station A on Figure
VIII-1).
d GAG pumping in sump adjacent to well.
ESTUARY HYDROLOGY
Groundwater Input
Subsurface flow from Fahkahatchee Strand into the estuary below is
extremely small. The hydraulic gradient of the ground water table from
Janes Scenic Drive to the Gulf of Mexico was on the order of 0.001 m/m
(0.0001 ft/ft) or 9.47 cm/km (0.5 ft/mi) during 1972. To obtain an approxi-
mation of this subsurface flow the general hydraulic equation Q = TIL was
used.
Where: Q = Flow in liters per day
T = Transmissibility of the aquifer in cm/sec
I = Hydraulic gradient in m/m
L = Length of the flow section in meters taken
perpendicular to the direction of flow
A transmissibility value (T) of 4.3 cm/sec (92,000 gpd/ft) was used.
This value was discussed previously in the groundwater section and is
considered typical of the area in question.
Several parameters are not included in this approximation of subsurface
flow. Canal drainage, storage, artesian flows, variations in hydraulic
gradient and in transmissibility were not determined. However, the approxi-
mations should indicate an order of magnitude of the flows.
VII-22
-------
120
I 100
\
o
o>
m
O
tr
<
o
CO
<
<
o
80
60
40
UJ
to
<
LJ
o 20
ALGAL GROWTH
PARTIALLY DAMMED
1.2 1.3 1.4 1.5 1.6
WATER LEVEL (METERS ABOVE MSL)
1.7
Figure VII-25. Increase in Discharge Due to Ground
Water Infiltration in the Easternmost GAG Canal.
Two sections perpendicular to the north-south direction of flow were
used. These sections are shown on Figure VII-1. The first section to
intercept the north-south flow is US 41 from SR 29 westerly 16 kilometers
(10 miles). The second (estuary) section is parallel to and of the same
approximate length as the US 41 section. This section is at the northern
general edge of Fahkahatchee Bay. The hydraulic gradient for the subsurface
flow passing the US 41 section was determined by use of USGS Well C-496 on
Janes Scenic Drive and EPA Well W-5 north of Remuda Ranch. The gradients
were adjusted to the direction of flow. The hydraulic gradient for the
estuary section was determined by Well W-5 and the average water elevations
in Fahkahatchee Bay.
Figure VII-26 is a plot of monthly subsurface flow in cubic meters for
the two sections. Subsurface flow passing US 41 ranged from 140,060 nrVmo
(37 million gal/mo) for January 1972 to 30,283 m3/mo (8 mgpm) for June 1972
with an average value of 95,392 m-^/rno (25.2 mgpm). Flow passing into the
VII-23
-------
estuary ranged from
100,313 m3/mo (26.5
mgpm) for September
1972 to 25,741 m3/mo
(6.8 mgpm) for April
1972 with an average
of 70,303 m3/mo (18.5
mgpm). As shown on
this figure the flow
passing into the estu-
ary exceeded the flow
passing U.S. 41 from
mid-May through mid-
September.
For the period
of mid-May through
early September the
monthly rainfalls ex-
ceeded the pan evapo-
ration, thus, account-
ing for the increase
in flows (Figure VII-
21). However, for the
year (1972) only 75
percent of the subsur-
face flow passing
U.S. 41 reaches the
estuary. Again, this
is due in large part
to the imbalances of
evaporation-transpir-
ation and rainfall.
~ 14
I
O 12
O 10
X
n
O
UJ
o:
en
SUBSURFACE FLOW
PASSING US, 41
SUBSURFACE FLOW
PASSING INTO THE
ESTUARY
M
J J
1972
Figure VII-26. Subsurface Flow in the Southern End of
Fahkahatchee Strand.
.cr
30
20
10
Table VII-2 sho^ws;
the monthly magnitudes^ o
of flows passing the ^ ~~*
two sections discussedz
above and also the ^
drainage canal system
bordering the western
end of the Strand.
The column marked
"Canal Drainage" is
the monthly averages
of the ground water
flow into a 1.6 kilometer (one mile) stretch of the easternmost GAG canal near
its intersection with Janes Scenic Drive. The values shown in this column are for
50 percent of the ground water interception, assuming that the ground water inflow
to this canal section comes equally from the east and from the west. It should be
emphasized that column 4 is for only a 1.6 kilometer (1 mi) section of the
canal. Since this section is at the western edge of one of the "wettest"
JFMAMJJASON
1972
Figure VII-27. Pan Evaporation Versus Rainfall in
Fahkahatchee Strand.
VII-24
-------
Table VII-2.
3 3
Ground Water Flows in m x 10 Per Month.
Month
Subsurface
Flow Passing
US 41
Subsurface
Flow Passing
into the Estuary
Canal
Drainage
Fahka-
Union
Canal
Dec.,, 1971
Jan., 1972
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Totalb
508
137
114
133
132
56
29
50
56
74
108
124
128
1,114
72
80
51
25
31
83
69
81
100
88
77
83
840
504
367
242
776
1,888
5,736
2,235
699
4,213
35,309
24,220
22,434
30,335
25,142
15,098
9,954
177,263
a Flows into 1.609 kilometers (one mile) section of GAC Canal near Janes
Scenic Drive. Incomplete due to construction during a portion of tlie
year.
b Totals are for 1972.
portions of the GAC properties in Collier County no attempt was made to
extrapolate to the entire drainage system estimated at 305.7 kilometers
(190 mi). However, it is evident from Table VII-2 that ground water inter-
ception by a 1.6 kilometer (1 mi) section of a north-south drainage canal
is approximately six times greater than the "normal" ground water input
to the estuaries from a 16 kilometer (10 mi) east-west section. The final
column is monthly flows passing the.weir on Fahka Union Canal at its inter-
section with US 41. This canal is fed by the vast GAC drainage system. These
flows give an indication of the relative water inputs to the esturarine system.
Estuary Bathymetry
Introduction
Soundings were made in bpth bays in order to: (1) make topographic
comparisons of the bays, (2) obtain volume, area and mean depth relationships
to tidal stages. From these finding numerous applications will be made
throughout the report such as circulation patterns, turnover rates, water
column depths, and sediment accumulations to name a few.
VII-25
-------
Methods
A Bludworth Marine Portable Echo Sounding Survey Recorder, Model
ES-130AVF, was used in mapping bay topography and obtaining channel cross
sections whenever needed. The bathymetry study consisted of 99 transects
totaling 33.0 kilometers (20.5 mi) in Fahka Union Bay and 93 transects
totaling 40.2 kilometers (25 mi) in Fahkahatchee Bay.
A Stevens Type A, Model 71, Continuous Stage Recorder was mounted in
Fahkahatchee Bay. Since no bench marks were available, the datum was
approximated by comparing mean daily and flood tide levels with records
from two stage recorders located at Rookery Bay near Marco Island. These
Rookery Bay recorders were previously referred to MSL (1929 datum). Selected
days were used when wind action was negligible and tidal patterns were
regular during the spring tide period. The estimated datum should be accurate
within 13.0 centimeters (0.1 ft).
Results
Both Fahkahatchee and Fahka Union Bays are classified as shallow
inland bays. The bays are characteristically shallow in the northern
half averaging 0.6 - 0.9 meter (2-3 ft) in Fahka Union Bay (Figure V1I-28)
and 0.9 - 1.2 meters (3-4 ft) in Fahkahatchee Bay (Figure VII-29). The
average depth increases by approximately 0.30 meter (1.0 ft) in the southern
half of each bay.
The average depth of Fahkahatchee Bay is 1.18 meters (3.88 ft) compared
to 0.93 meter (3.06 ft) in Fahka Union Bay. Ninety percent of Fahkahatchee
Bay has a depth greater than 0.82 meter (2.70 ft) compared to 0.75 meter
(2.45 ft) in Fahka Union Bay (Figures VII-30 and 31).
At MSL Fahka Union Bay has a surface area of 186 hectares (460 ac)
and a volume of 1,800,000 cubic meters (1,459 ac-ft) (Figure VII-32). At
approximately 0.61 meter (2.0 ft) above MSL the majority of the islands
become inundated and the surface area increased to 200 hectares (494 ac)
with a box volume of 2,950,000 cubic meters (2,351 ac-ft). When the low ebb
decreases to 0.61 meter (2.0 ft) below MSL the inundated portion of the bay
is 164 hectares (405 ac). During the winter, a spring tide and a strong
northerly wind decreases the inundated area to approximately 50 percent.
At this condition the water level has fallen to some 0.81 meter (2.65 ft)
below MSL and the volume decreased to 300,000 cubic meters (243 ac-ft).
At MSL Fahkahatchee Bay has a surface area of 740 hectares (1,829 ac)
and a volume of 8,800,000 cubic meters (7,134 ac-ft) (Figure VII-33). At
0.61 meter (2.0 ft) above MSL the surface area increased to 765 hectares
(1,890 ac) with a volume of 13,300,000 cubic meters (10,782 ac-ft). Some
680 hectares (1,680 ac) are inundated at 0.61 meter (2.0 ft) below MSL.
Circulation Patterns
Introduction
Fahkahatchee and Fahka Union Bays are shallow bodies of water with
average depths at MSL of 1018 meters and 0.93 meter, respectively. The bays
VII-26
-------
N
FAHKA UNION
BAY
NOTE
DEPTH IN FEET. MULTIPLY BY
0.3048 TO OBTAIN METERS
Figure VII-28, Fahka Union Bay Showing Major Bathymetric Features*
VII-27
-------
Figure VII-29. Fahkahatchee Bay Showing Major Bathymetric Features.
VII-28
-------
O 1.75
UJ
5
30 40 50 60 70
PERCENT EQUAL TO OR GREATER THAN
Figure VII-30.
Bay at MSL.
Depth Patterns in Fahka Union
are characterized by muddy
bottoms occasionally
relieved by a more sandy
bottom or by oyster bars.
To the north the bays are
bordered by mangrove flats
and tidal creeks which
blend into brackish tidal
marshes further inland. Due
to extreme flatness, much
of the area to the north
becomes inundated during
flood tides making bound-
aries for the bays non-
del ineable. The southern
extreme of the bays are
separated from the Gulf
of Mexico by the "Ten
Thousand Island" chain.
Numerous passes meander
through these islands
allowing circulation
between the bays and outer
keys. Many of these pas-
ses are six to eight meter?
in depth with well scoured
sandy bottoms.
Methods
A circulation study of
Fahkahatchee Bay was con-
ducted on October 5 and 6,
1?72. This was a joint
study with the University
of Miami's Rosenstiel
School of Marine and
Atmospheric Sciences. The
purpose of the study was to:
(1) quantify the water move-
ment between Fahkahatchee
Bay and the various open-
ings to the uplands and
the Gulf of Mexico;
(2) determine the exchange
of flows between
Fahkahatchee Bay and
adjacent bays; and
(3) determine the tidal prism and flushing rate of Fahkahatchee Bay.
The study consisted of utilizing two two-man sampling teams each equipped
with a sixteen foot boat, Beckman RS5-3 Salinometer, Bendix ducted current
meter and sample bottles for laboratory salinity check. The location of the
eleven sampling stations and a Stevens Type A-71 Stage Recorder are shown in
Figure VII-34.
30 4O 50 60 70
PERCENT EQUAL TO OR GREATER THAN
Figure VII-31.
Bay at MSL.
Depth Patterns in Fahkahatchee
VII-29
-------
Staff gages were
installed at Stations 1,
2 and 5 in order to relate
stage at sampling times
to the continuous record-
er. Sampling was carried
out over a full tidal cy-
cle during a period when
there were two equal
tides per day. Station
10 was sampled over two
tidal cycles. A fatho-
meter was used at each
station so that rela-
tionships between sec-
tional area and water
elevations could be com-
puted.
The eleven stations
were divided between the
two field parties with
each party attempting
to sample each of its
stations hourly. Verti-
cal salinity profiles
were measured at 61
centimeters (2 ft) incre-
ments from 30 centimeters
(1 ft) below the surface
to 30 centimeters above
the bottom. Surface
salinity samples were
intermittently checked
by standard laboratory
procedures.
SURFACE AREA (HECTARES)
140 120 100 80
Figure VII-32. Volume, Area, Depth Relationship
in Fahka Union Bay.
SURFACE AREA (HECTARES)
TOO SCO »QO 400 30O
a 10 12
VOLUME (m'xiO6)
Figure VII-33. Volume, Area, Depth Relationship
in Fahkahatchee Bay.
The following pro-
cedure was used at each
station for calculating
flows. Flows passing
each station were compu-
ted from field measured
velocity and cross sectional area relationships and tidal stage (Figure
VII-35). Field measured velocity curves are shown in Figure VII-36. As
seen by the velocity curve for Station 10, the transition between tides
or slack time was typically a very short period. Flow curves, Figure VII-37,
are then plotted against time and integrated for total volume passing a
particular station over the required time period.
Results
Table VII-3 lists the flow delineations of each station as measured
during this study. The percentages shown are relative to the total flow
VII-30
-------
Figure VII-34. Map of Study Area Showing Flow Measurement Stations.
VII-31
-------
passing all eleven
stations during either
a flooding or ebbing
tide. The term flood
designates a rising
water surface and ebb
a period of falling
water.
in
a:
LU
I- MSL .
LJ
Figures VII-38
and 39 show the direc-
tion of flow for each
station and percent-
age of total discharge
passing a particular
station relative to
the total discharge
of the eleven sta-
tions during one-half
of a tidal cycle. As
shown by Figure VII-38, _
during a flooding tide, Q?
approximately 80 percent t-
of the total discharge 2
was entering Fahkahat- ~
chee Bay via Stations o
5-11 and 20 percent ^
of the total discharge uj
was departing via
Stations 1-4, As
observed from Figure
VII-39, the reverse
of the above was true
on the ebb tide.
20
30 40 50 60 70
DISTANCE (METERS)
80
90
100
Figure VII-35„ Cross Sectional Area Relationship
Tidal Stage at Station 10.
to
-VELOCITY CURVE
Fahka Union Bay
to the west (through
Station 1) accounts
for two to five per-
cent of the total
Fahkahatchee Bay dis-
charge. Gate Bay to
the southwest exchang-
es flow via Station 5
at only one percent of
the total discharge.
Stations 2, 3 and 4
pass flow in and out
of the salt marsh by
way of tidal creeks.
Discharge during this
particular study
through the creeks at
these three stations
1.0
0.5
MSL
TIDE GAGE RECORDING
0.5
UJ 1.0
1.0
m<
nm
mi-
o
09OO
I2OO 1500
OCT 5, 1972
1800 21OO 0000
TIME (HOURS)
03OO 06OO
OCT 6, 1972
0900
Figure VII-36.
Station 10.
400
Velocity Curve and Tidal Stage at
200
o
-------
ARROW INDICATES DIRECTION OF FLOW
DURING FLOODING TIDE
STATION NUMBER
PERCENTAGE OF TOTAL BAY DISCHARGE
MOVING THROUGH PARTICULAR PASS
Figure VII-38. Flow Directions and Percentage of Discharge in Fahkahatchee
Bay During Flooding Tide, October 5-6, 1972.
accounted for 16 percent of the total discharge. These three stations are
dependent not only upon stage and wind, but also upon other hydrologic
conditions such as high runoff and droughts. Also evident is that the
adjacent bays (Gate and Fahka Union) account for only three to six percent
of the total discharge.
Water exits Fahkahatchee Bay through the passes during the ebb tide,
mixes with offshore water and re-enters during the subsequent flood tide.
Exchange ratio is the volume of "new" offshore water entering the bay divided
by the total volume entering the bay during the flood tide. Parker (1972)
developed the relationship with salinity as a tracer, where:
,. _ CF-CE
Cg = Weighted average salinity on ebb tide
Cp = Weighted average salinity on flood tide
C0 = Background salinity
VII-33
-------
ARROW INDICATES DIRECTION OF FLOW
DURING EBBING TIDE
STATION NUMBER
PERCENTAGE OF TOTAL BAY DISCHARGE
MOVING THROUGH PARTICULAR PASS
FAHKAHATCHEE
ISLAND
Figure VII-39. Flow Direction and Percentage of Discharge in Fahkahatchee Bay
During Ebbing Tide, October 5 and 6, 1972„
Station 10 was monitored over two tidal cycles which produced sufficient
data to arrive at a reasonable estimate for exchange ratio. Several offshore
salinity measurements were made during the study. The general location of these
samples were at Round Key, 8 and 4 kilometers (2.5 and 5.0 mi) seaward of Round
Key. The "offshore salinities ranged from 27 to 34 ppt. Station 10 salinities
(Figure VII-40 ranged from approximately 24 ppt to somewhat greater than 24 ppt.
Computation based on offshore salinities and Station 10 salinities yield
a reasonable estimate of exchange ratio to be 10 percent. Therefore, on a
flood tide approximately 10 percent of the water is "new" offshore water, and
90 percent has been in Fahkahatchee Bay or adjacent bays before. Assuming that
the above ratio is typical of the other inlets, a flushing rate can be approx-
imated.
Flushing rate in this case may be defined as the dilution of a given con-
centration by tidal action in a completely mixed system. This dilution, or
decaying may be approximated by:
VII-34
-------
A M
_2 = [1 - (r TP/(VLW + TP)}N
where: A = Concentration after N cycles
A = Initial concentration
Vjrj = Bay volume at low water
TP = Tidal prism volume
N = Number of tidal cycles
r = Exchange ratio
Table VII-3. Relative Discharge Per Station
High water and low water
volumes were obtained from Figure
VII-33. The tidal prism is the
difference between low and high
water volume. Figure VII-41
presents the calculated flushing
rate in Fahkahatchee Bay assum-
ing complete mixing and two tides
of equal magnitude per day. As
shown by this figure, a consti-
tuent would be reduced to 10 per-
cent of the original concentration
after 42 tidal cycles or approxi-
mately 21 days.
Seasonal variations affect
tidal prism volumes, therefore,
affecting flushing rates. Fig-
ure VII-42 was developed by
applying monthly average tidal
ranges to the flushing equation. The exchange ratio was held constant at 10 per-
cent and the number of tidal cycles at 40. As observed on Figure VII-42, after
40 cycles in February, 8.5 percent of a given contaminant remains compared to
13 percent in August.
Salinity and Temperature Variations
Station
Number
1
2
3
4
5
6
7
8
9
10
11
Flood
% Q
2
1
8
7
1
10
10
6
13
33
9
Ebb
% Q
5
4
6
6
1
11
6
4
13
33
11
Introduction
An easily measurable
parameter for assessing the
physical effects of drainage on
estuaries is differences in
salinity patterns in areas sub-
ject to rapid, normal or below
normal runoff. In general,
Fahka Union Bay is subject to
sporadic massive fresh water
inputs from the GAG drainage
system. In contrast, Fahka-
hatchee Bay receives normal or
below normal runoff from the
Fahkahatchee and East Rivers.
30
Q.ZO
Q.
10
STATION IO
• i ••
BAY STATION
STATION
08OO IOOO 1200 1400 I6OO I8OO 2OOO
TIME (HOURS)
Figure VII-40. Salinity Curves for Stations
I, 10 and Mid-Bay, October 5, 1972.
VII-35
-------
Less than average runoff may
occur in Fahkahatchee Bay due to
the interception of surface and
subsurface waters by the GAG
canals to the west and the Barren
River Canal to the east of
Fahkahatchee Strand.
Salinity patterns in South
Florida estuaries are generally
very dynamic because of the
seasonality of the hydroperiod.
During winter drought periods
the salinity in some estuaries
increase above that of the Gulf
of Mexico due to evaporation.
Salinities exceeding 40 ppt
have been measured in the Ten
Thousand Islands area. During
the summer, when rainfall and
runoff is at a maximum, salin-
ities at low ebb are often
below 10 ppt.
For fish and other aquatic
life to survive in the estuaries
they must have a high salinity
tolerance or be able to leave
certain estuaries whenever the
salinity becomes intolerable.
Certain marine species migrate
into the estuaries seeking opti-
mum salinities and temperatures
for reproductive purposes.
Extreme variations in the normal
spatial and temporal distribution
0.9
0.8
O.T
0.6
g
u.
o
0.5
< 0.4
CC
U.
O 0.3
0.2
o
o
0.1
UJ
-------
of salinity may adversely affect the aquatic communities of an estuary. Some
estuarine animals such as the blue crab and fiddler crab can live in or near
water that is essentially fresh but must return to the ocean to spawn, since
their eggs cannot develop in fresh water. Field studies of distribution show
that the number of species decline as salinity falls, and the weight of
physiological evidence strongly indicates that salinity is a limiting factor
for marine organisms, especially when it varies downward (Pearse, 1957).
Methods
Data for salinity profiles of Fahkahatchee Bay, Fahka Union Bay, Fahka
Union Pass, Fahkahatchee River, Fahkahatchee Pass and West Pass were collected
every two months on both the high and low tides. Each study utilized two
sampling crews. One crew concentrated in the bays and the other crew concen-
trated in the canals and passes. Studies were conducted from approximately
1.25 hours before to 1.25 hours after both the ebb and flood slack. Except
as noted on the salinity profiles all measurements were made at the one foot
depth using Beckman Model RS5-3 Portable Salinometers. Simultaneous temperature
measurements were also conducted.
Results
Salinity station locations for the estuaries and canal-passes are shown
in Figures VII-43 and 44. The bay study consisted of 55 stations (1 through 55)
and the canal-passes 33 stations (56 through 88).
Figure VII-43. Map of Fahka Union and Fahkahatchee Bays Showing Station
Locations for Estuary Salinity Studies.
VII-37
-------
Inflow to the bays was relatively low in January with a minimum occurring
in April (Figures VII-45 through 56).
Figure VII-44. Map of Estuary Showing Station Locations for "Canal-Passes"
Salinity Studies.
VII-38
-------
Figure VII-45. Isohalines in Bays at Low Tide (a.m.) on Feb 1, 1972.
Figure VII-46. Isohalines in Bays at High Tide (p.m.) on Feb 1, 1972.
VII-39
-------
Figure VII-47. Isohalines in Bays at High Tide (a.m.) on April 7, 1972
Figure VII-48. Isohalines in Bays at Low Tide (p.m.) on April 7, 1972
VII-40
-------
Figure VTI-49. Isohalines in Bays at Low Tide (a.m.) on June 27, 1972.
Figure VII-50. Isohalines in Bays at High Tide (p.m.) on June 27, 1972.
VII-41
-------
N
Figure VII-51. Isohaline in Bays at Low Tide (a.nu) on August 10, 1972.
Figure VII-52. Isohaline in Bays at High Tide (p.m.) on August 10, 1972.
VII-42
-------
Figure VII-53,, Isohaline in Bays at Low Tide (p.m.) on October 31, 1972.
Figure VII-54. Isohaline in Bays at High Tide (a.m.) on October 31, 1972.
VII-43
-------
Figure VII-55. Isohalines in Bays at Low Tide (a.m.) on Dec 7, 1972.
Figure VII-56. Isohalines in Bays at High Tide (p.m.) on Dec 7, 1972.
VII-44
-------
On April 7, 1972 the average salinity in Fahka Union Bay on the ebb and
flood tides (Figures 57 and 58) were 35.3 ppt and 36.5 ppt, respectively,
compared to 36.7 ppt and 36.9 ppt in Fahkahatchee Bay (Figures VII-59 and 60).
During both the ebb and flood only 37 ppt isohaline was evident in Fahkahatchee
Bay. Salinities in the Gulf of Mexico off Round Key approximated 36.5 ppt.
It is evident that upland runoff was small and that very little mixing occurred
between Fahkahatchee Bay and the Gulf of Mexico during this period. Bay
salinities became greater than off-shore salinities due to these small runoffs
and minor exchange rates (Figure VII-42) permitting evaporation in the upper
bay and hence a build-up of salts and other solids. Salinities at Station 53,
some 4.34 kilometers (2.7 mi) above the mouth of the Fahkahatchee River, were
in the order of 36.3 ppt to 36.8 ppt compared to 32 ppt near the mouth of the
Fahka Union Canal on the flood tide, indicating that eVen during periods of
extreme low flows in Fahka Union Canal, effects of drainage are evident.
Extreme differences in salinity patterns in the bays were noted in the
June 27 survey (Figures VII-49 and 50). Isohalines in Fahka Union Bay ranged
from 3 ppt to 8 ppt compared with 9 ppt to 21 ppt in Fahkahatchee Bay (Figure
VII-44). From the isohalines it appears that during high runoff periods the
GAG Canal has an effect on the western most portion of Fahkahatchee Bay. For
instance, the 9 ppt to 13 ppt range is not evident in the vicinity of the
Fahkahatchee and East Rivers. Ignoring the exchange effect from Fahka Union
Bay would give a salinity range of 14 ppt to 21 ppt in Fahkahatchee Bay,,
During the ebb tide the average salinity in Fahka Union and Fahkahatchee Bays
were 6.5 ppt and 17.1 ppt, respectively, a difference of 10.6 ppt. Again,
ignoring exchange effects from Fahka Union, would increase the average salinity
in Fahkahatchee Bay to 17.9 ppt, therefore, increasing the difference between
the two bays to 11.4 pptD The same general characteristics exist on the flood
tide (Figure VII-50).
The greatest observed salinity gradient in Fahka Union Bay occurred on
a low ebb tide during the morning of August 10. During this ebb the majority
of water entering Fahka Union Bay remained in the channel, or was transported
out of the bay via the channel. There was very little evidence of canal water
being transported across the bay or through the small cut at the extreme north-
east side of the bay. Within this cut the salinity gradient was approximately
1.8 ppt per 30.5 meters (100 ft). During the ebb the salinity range in Fahka
Union and Fahkahatchee Bays were 11 ppt to 29 ppt and 27 ppt to 33 ppt,
respectively. On the flood tide the salinity range in Fahka Union Bay increased
from 22 ppt to 30 ppt and the average salinity increased from 20.5 ppt to 22.0
ppt. Only a minor change took place in Fahkahatchee Bay with the salinity
range increasing from 28 ppt to 33 ppt and the average increasing from 26.4
ppt to 27.3 ppt.
The concentric isohalines in Fahkahatchee Bay the mornings of August 10
(Figure VII-51) and October 31 (Figure VII-53) are indicative of a changing
tide. In general, the change of tide during the morning run would occur while
the sampling team was in the eastern side of Fahkahatchee Bay. Since the
afternoon run was conducted in reverse of the morning run the tide change
occurred while sampling in Fahka Union Bay.
The average salinity was always lower in Fahka Union Bay than Fahkahatchee
Bay (Figures VII-57 through 60). The maximum observed differences occurred
during the June 27 survey when flows in the Fahka Union Canal were near a maximum.
VII-45
-------
40
35
30
25
Q.
O.
20
15
10
JFMAMJJASONDJ
1972
40
35
30
25
30
15
10
JFMAMJJASONDJ
1972
Figure VII-57. Salinity Variations in Figure VII-58. Salinity Variations in
Fahka Union Bay at Low Tide During 1972. Fahka Union Bay at High Tide During 1972.
40
35
30
25
20
15
10
40
35
30
25
Q.
>-
20
15
10
"JFMAMJJASONDJ
1972
FMAMJJASONDJ
1972
1972 1972
Figure VII-59. Salinity Variations in Figure VII-60. Salinity Variations in
Fahkahatchee Bay at Low Tide During Fahkahatchee Bay at High Tide During 1972.
1972.
VII-46
-------
Flows peaked in the Fahka Union Canal in September (Figure VII-7). Conceivably,
had a salinity run been conducted during the peak runoff period the differences
would have been even greater.
Salinities in the Fahka Union Canal-Bay-Pass transect always increased
downstream from U.S. 41 on the ebb tide (Figure VII-61). During periods of
high flow the maximum gradient occurred downstream from the mouth of the
Fahka Union Canal. During periods of low flow the maximum gradient occurred
in the canal. During the low runoff study on April 7 the salinity increased
from 11.6 ppt at U.S. 41 to 32.2 ppt at the mouth of Fahka Union Canal as
compared to 0.9 ppt to 2.1 ppt during the wet month of June. Salinities at
the mouth of Fahka Union Pass varied from a low of 30.3 ppt on December 7 to
a high of 36.6 ppt on April 7 on ebb tide. On flood tide (Figure VII-62) the
salinity range increased to a low of 33 ppt on August 10 to a high of 36.9 on
April 7.
For comparative purposes Figure VII-63 is comparable to that portion of
Figures VII-61 and 62 below 8.5 kilometers (5.3 mi) labled "PASS." As would
be expected, the salinity gradient below Fahka Union Bay was always greater
than that below Fahkahatchee Bay. On April 7 ebb tide, the salinity actually
decreased seaward of Fahkahatchee Bay. The greatest salinity gradient seaward
of Fahkahatchee Bay occurred on the June 27 ebb tide with an increase of 10.4
ppt from 21.6 ppt to 33.0 ppt. This compared with a change of 18.6 ppt from
12.1 ppt to 30.7 ppt in Fahka Union Bay on the same date. With the exception
40 •»
CANAL
FAHKA-UNION
^L BAY ^L PASS
FAHKA-UNION
I 23456789 10
KILOMETERS BELOW U S HWY 41
12 13 14
Figure VII-61. Salinity Variations in
the Fahka Union Canal at Low Tide
During 1972.
40 .*
I 2345678910
KILOMETERS BELOW US HWY 41
12 13 14
Figure VII-62. Salinity Variations in
the Fahka Union Canal at High Tide
During 1972.
VII-47
-------
APR 7
JUNE 27
.OCT3I
fc/v" -FEBI
•/'—• CDEC7
AUG 10
APR 7
JUNE 27
of June 27 the salinity differences seaward of Fahkahatchee Bay were less
than 1.5 ppt compared to 12.7 ppt in the Fahka Union Pass.
In conjunction with the detrital transport study of October 17, 18 and 19
(Section IX), six time series vertical salinity measurements were conducted
over a tidal cycle in Fahka Union Bay and Fahkahatchee Bay. The Fahka Union
Canal and Fahka Union Pass stations (Figure VII-64) are at the same general
locations as are salinity stations 1 and 9 (Figure VII-43), respectively.
The Fahkahatchee River and East River stations (Figure VII-65) are located
approximately half way between salinity stations 50 and 51 and approximately
183 meters (600 ft) above the mouth of the East River, respectively. Fahka-
hatchee Pass and West Pass stations (Figure VII-66) are located between
salinity stations 31 and 32 and at station 40, respectively.
As observed from Figure VII-64, surface salinities in the canal varied
from a high of 18 ppt on the flood tide to a low of 2»5 ppt on the ebb tide.
Within the pass, some 1.1 miles downstream, the surface salinities varied from
a low of 13 ppt to a high
of 28 ppt. Both stations
exhibited stratification t_
during portions of the
tidal cycle. Within the
canal the greatest strati-
fication occurred some two
hours after the low slack
with salinity at the sur-
face of 3 ppt and the five
foot depth of 15 ppt. Ca-
nal water was well mixed
for approximately six
hours during the ebb.
Maximum stratification
within the Pass occurred
at the beginning of the
ebb. Salinities at the
one, five and seven foot
depth were 18 ppt, >26 ppt
and 27 ppt, respectively.
The Pass remained strati-
fied until two hours after
the ebb slack. Restrati-
fication did not occur on
the high slack at 2300
hours. Fresh water input
from the GAG drainage sys-
tem was approximately
14,151 I/sec (500 cfs)
during the study. Had the
study taken place around
the first of September
with fresh water input
in the order of 6,295
I/sec (2,200 cfs) the
FAHKAHATCHEE PASS
HIGH TIDE
KILOMETERS SEAWARD OF FAHKAHATCHEE BAY
. ^ ^..~. AUG 10
:-^Z—- -- .OCT3I
APR 7
FAHKAHATCHEE PASS
LOW TIDE
01 23456
KILOMETERS SEAWARD OF FAHKAHATCHEE BAY
Figure VII-63. Salinity Variations in Fahkahatchee
Pass and West Pass During 1972.
VII-48
-------
E°2
— 0
LU
002
en °4
06
25
20
ji 15
b 10
^ 5
30
g. 25
£20
§ 15
<
1/5 10
FAHKA-UNION CANAL
FAHKA-UNION PASS
1000 1200 1400 1600 1800 2000 2200 2400
TIME (HOURS)
§04
I 02
~- 0
0.6
26
„ 25
^ 22
<
(n
20
26
•^ 25
»- 23
21
20
FAHKAHATCHEE RIVER
EAST RIVER
1200 1400 1600 1800 2000 2200 2400 0200
TIME (HOURS)
Figure VII-64. Time, Stage and Salin-
ity Variations in Fahka Union Canal
and Fahka Union Pass on Oct 17, 1972.
Figure VII-65. Time, Stage and Salinity
Variations in Fahkahatchee River on
Oct 18, 1972.
degree of stratification at both stations would have been presumably greater.
Salinity stratification existed at all four stations in Fahkahatchee
Bay (Figures VII-65 and 66) but to a much lesser degree than in Fahka Union
Bay. Differences at the 0.30 and 1.52 meter (1 and 5 ft) depth were generally
less than 1.0 ppt. Over the tidal cycle, salinities in Fahkahatchee River
ranged from a high of 25.8 ppt to a low of 23 ppt as compared with 25.7 ppt
to 20.6 ppt in the East River. Based upon salinity differentials it is
apparent that the greatest input of fresh water into the northern end of
Fahkahatchee Bay is from the East River. This was confirmed by the previously
discussed circulation study.
As observed from the salinity profiles in Figure VII-66, a greater
percentage of fresh water exits via Fahkahatchee Pass than West Pass, The
minimum salinities in Fahkahatchee Pass and West Pass during the October 19
survey were 26 ppt and 27 ppt, respectively. Likewise, maximum salinities
were 30 ppt and 30.7 ppt, respectively at the flood slack.
In conjunction with the bi-monthly salinity studies, measurement of
estuary water temperatures were also made. Regardless of the tidal phase,
the yearly temperature profiles (Figure VII-67) are presented on a morning
and afternoon basis. The average water temperature differences between
VII-49
-------
-, °6
co 04
-i 1 1 1 1 1
_ 30
•&.
,£• 29
£ 28
? 27
$*
25
FAHKAHATCHEE PASS
31 r
£30
H ^
| 28
^ 27
26
WEST PASS
1200
1400
2200 0000 0200
Fahka Union and Fahkahatchee
Bays are less than one degree
centigrade (Figure VII-67).
Morning water temperatures
ranged from a low of 24.1° C to
a high of 32.7°C in Fahka Union
Bay compared with 23.6°C to
32.1 C in Fahkahatchee Bay. The
afternoon temperature range in
Fahka Union and Fahkahatchee
Bays were 24.7°C to 33.1°C and
24.3°C to 32.7°C, respectively.
The slightly higher water tem-
peratures in Fahka Union Bay
are probably due to the fact
that it is somewhat more shallow
than Fahkahatchee Bay.
Tidal Fluctuations
The most important eco-
logical factor in an estuary
is the tide, since it is the
prime agent which effects the
exchange of water, and its ver-
tical range generally determines
the extent of tidal flats which
may be exposed or submerged
with each tidal cycle. The
action of the tides against
the volume of river water
discharged into an estuary
results in a complex system
whose properties vary according to the structure of the estuary as well as the
magnitude of the river flow and range of tide (Emery and Stevenson, 1971).
A Stevens Type A, Model 71, Stage Recorder was installed at the southern
side of Fahkahatchee Bay in the vicinity of Fahkahatchee Pass (Figure VII-1)
in order to assess the tidal fluctuations in the bay and to provide information
from which to conduct other studies. Due to lack of stage recorders, this
portion of the study was conducted only from June to December 1972.
There is a greater tidal differential in Fahkahatchee Bay during the
spring tide periods than the neap tide period (Figure VII-68),,
Due to the shallowness of both Fahkahatchee and Fahka Union Bays, wind
produces a very pronounced effect on the water elevations of the bays. During
northern (continental) winds the salt marshes above the bays, and
the bays themselves are depleted of a significant portion of their water.
During the study period the lowest observed tides occurred on December 16,
1972, when the surface water elevation fell to 81.4 centimeters (2.67 ft)
below MSL. At this time only 78 percent of Fahkahatchee Bay and 47 percent
of Fahka Union Bay were inundated.
1600 1800 2000
TIME (HOURS)
Figure VII-66. Time, Stage and Salinity Varia-
tions in Fahkahatchee Pass and West Pass on
Oct 19, 1972.
VII-50
-------
35
The highest elevation
occurred on June 19, 1972,
during the influence of
Hurricane Agnes. During
this period of high
southwesterly winds the
water elevation peaked
at 88.4 centimeters
(2.90 ft) above MSL. The
highest low tide occurred
on this date at 39.6
centimeters (1.30 ft)
above MSL. The greater
difference in tidal mag-
nitude also occurred on
June 19 with the low
high tide at 83.2 centi-
meters (2.73
MSL, and the
tide at 49.7
ft) above
low low
centimeters
(1.63 ft) below MSL.
This 1.33 meter (4.36 ft)
decrease in tidal ampli-
tude accounted for over
9,800,000 cubic meters
(7,950 ac-ft) of water
being removed from Fahka-
hatchee Bay and contiguous
uplands.
o
30
CE
25
20
35 r
35
30
LJ
20
FAHKAHATCHEE BAY
A.M.
JFMAMJJ ASOND
FAHKAHATCHEE BAY
PM
J FMAMJJ ASONO
20
35
30
25
20
FAHKA-UNION BAY
A.M.
JFMAMJJASONO
FAHKA-UNION BAY
P.M.
JFMAMJJASOND
Figure VII-67. Annual Temperature Ranges in
Fahka Union and Fahkahatchee Bays During 1972.
+ 80
+4O
> +20
UJ
-I 0
<
UJ -20
CO
2 -40
<
^ -60
-80
O
€
22
24
26
28
30
4 6 8 IO
JUNE - JULY 1972
14
16
IS
20
Figure VII-68. Tidal Amplitudes With Lunai Phase for Fahkahatchee Bay.
VII-51
-------
During these periods of oceanic winds, tidal amplitudes are greatly
increased, raising saline waters inland above U.S. 41 in the vicinity of
lower Fahkahatchee Strand,, Due to the extremely small topographic relief
of the area, a water elevation variation of a few inches can inundate or
drain vast areas of highly productive mangrove and salt marsh areas.
VII-52
-------
VIII - BAY BOTTOM CHARACTERISTICS AND SEDIMENT DEPOSITION
INTRODUCTION
Southwestern Florida coastal areas are of increasing concern to those who
are interested in the changes to the natural and esthetic qualities of the
estuarine environments. One of the major factors affecting the estuarine
system is sedimentation; however, little information is available.
The purpose of this section is to identify the bottom characteristics
of Fahkahatchee and Fahka Union Bays. Further, the Bays were investigated
for sedimentation rates followed by coring based upon sedimentation rates.
Sediment and core samples were analyzed for organic and inorganic func-
tions on dry weight basis.
METHODS
Bottom Characteristics
Fahkahatchee and Fahka Union Bays were staked on a 300 by 300 meter grid
system (Figure VIII-1) which resulted in 103 and 24 sampling sites in Fahka-
hatchee and Fahka Union Bays, respectively.
Bottom deposits were collected from each grid intersection by means of a
15 by 15 centimeter Ekman dredge. Bottom sediments were classified by visual
examination into three somewhat arbitrary classifications of mud, sand or shell.
Combinations of the three types were delineated by noting the predominant,
secondary and tertiary characteristic.
Sedimentation Rates
To assess the amount of organic and inorganic sediment depositing in the
two bays, sediment traps were installed on the bay bottoms. The traps
(Braidech, Gehring and Klevens, 1972) were constructed from PVC pipe and
contained removable inner cylinders to reduce turbulence and to increase
settling efficiency within the traps (Figure VIII-2).
The traps were placed on the bay bottoms at somewhat arbitrary grid loca-
tions. Twenty-eight sites presented on Figure VIII-3 were used with 14 sites
per bay. Fahka Union Bay was reasonably proportioned by the traps; however,
the 14 traps used in Fahkahatchee Bay were only sufficient to cover the area
below the mouths of Fahkahatchee and East Rivers.
The traps were installed on July 17, 1972, and were serviced every two
weeks until removal on November 20, 1972. When servicing, a protective cap
was placed over the open end of the traps to prevent any resuspended material
from entering the traps during retrieval and installation. Contents of the
traps were then transferred to a large container (approximately 10 liters),
resuspended by hand mixing, and sub-sampled by use of a 300 milliliter BOD
bottle. Subsequently, the traps were cleaned and replaced on the bay bottoms.
VIII-1
-------
Figure VIII-1. Grid System.
Sub-samples were analyzed
for salinity, dry weight
and ash free dry weight.
Appendix C describes the
laboratory methodology
and computations to
arrive at the organic
and inorganic sedimenta-
tion rates. The computed
bi-weekly rates are in
terms of grams per meter
squared per day for each
of the 28 stations.
i S
2 o
D oj
f to
H
~i
I
-r
i
I5.25cm I.D.
2.84cm I.D.
-25.40cm
SIDE VIEW ' TOP VIEW
Figure VII-2. Sedimentation Trap.
VIII-2
-------
Figure VIII-3. Sediment Trap Locations,
Coring
Inorganic and organic content of the top 40 centimeters of the bay bottoms
was assessed by coring. The coring devices used were 3.5 centimeter I. D. PVC
pipe cut to one meter lengths. These open ended pipes were eased into the soft
bay bottoms to approximately 2/3 of their length. Rubber stoppers were then
placed on the top and bottom of the cores. The bottom stopper was installed
by digging alongside of the in-place core to the tube end. Samples were then
returned to the laboratory for analysis.
RESULTS
Bottom Characteristics
The distribution of bottom types is illustrated in Figure VIII-4. The
boundaries drawn between the three types were not clearly defined by the field
survey and hence serve only to show the general trend in the distribution of
the sediments. Table VIII-1 presents the observed substrate characteristics
VIII-3
-------
EXPLANATION
": :'v: MUD
HSTS SHELL
...-.-...- SAND
Figure VIII-4. Distribution of Bottom Characteristics.
of Fahka Union and Fahkahatchee Bays in order of predominant substrate. For
example, the substrate designated MSO at Grid P-9 indicates that the sediment
sample was predominantly mud with sand to a lesser extent and also some shell
fragments present. A summary of the substrate observations is presented in
Table VIII-2.
Inland bays of the Ten Thousand Islands are generally characterized as
"muddy" bays. Fahkahatchee and Fahka Union Bays are no exception. Approxi-
mately 52 percent of Fahkahatchee Bay and 76 percent of Fahka Union Bay is
mud. Mud is present in all of Fahka Union Bay and in 79 percent of Fahka-
hatchee Bay.
Oyster bars are found in both bays, especially on bottoms contiguous to
the shallow mangrove islands. Where oysters were found only as extensions of
the mangroves they were not considered to be characteristic of the bottom.
The largest areas of oysters were found in the mouths of Fahkahatchee and
East Rivers and the southeastern side of Fahkahatchee Bay. Smaller concen-
trations of oysters were found on the northeast and west side of Fahka Union
Bay near the mouth of 'the Fahka Union Canal.
VIII-4
-------
Table vill-l. Substrate Characteristics in Fahka Union and
Fahkahatchee Bays.
Grid
Fahka
P-9
P-10
P-ll
P-12
P-13
Q-9
Q-ll
Q-12
Substratea
Union Bay
MSO
SO
MSO
MO
MO
MOS
M
M
Grid
Q-13
Q-15
R-10
R-ll
R-12
R-13
R-14
S-10
Substratea
MO
0
MSO
M
M
MS
MSO
SM
Grid
S-ll
S-12
S-13
T-ll
T-12
T-13
U-12
U-13
ft
Substrate
SMO
M
MS
OS
M
M
MS
0
Fahkahatchee Bay
C-l
C-3
C-4
C-5
C-6
C-7
D-l
D-2
D-3
D-4
D-5
D-6
D-7
D-8
E-l
E-2
E-3
E-4
E-5
E-6
E-7
E-8
E-9
F-2
F-3
F-4
F-5
F-6
F-7
F-8
F-9
G-0
G-l
G-2
M
M
MO
MS
MS
SM
M
MO
SOM
M
MS
M
M
MSO
MSO
SM
SMO
SM
SM
SM
SOM
SO
M
SMO
SMO
SMO
SMO
SMO
SMO
MSO
MO
0
OS
OS
G-3
G-4
G-5
G-6
G-7
G-8
G-0
H-l
H-2
H-3
H-4
H-5
H-6
H-7
H-8
H-9
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
J-2
J-3
J-4
J-5
J-6
J-7
J-8
K-l
K-2
K-3
SO
SO
SM
SMO
MS
MS
M
OS
SMO
SMO
SM
SO
SM
MO
MS
M
SMO
SMO
OS
SM
SM
MS
M
MS
SO
SO
S
SM
M
M
MO
MOS
SO
OS
K-4
K-5
K-6
K-7
K-8
L-l
L-2
L-3
L-4
L-5
L-6
L-7
L-8
M-3
M-4
M-5
M-6
M-7
M-8
N-4
N-5
N-6
N-7
0-4
0-5
0-6
0-7
P-5
P-6
P-7
P-8
Q-5
Q-7
Q-8
SO
SO
MO
M
MO
0
SO
SM
MO
SO
MS
MS
MSO
OS
SM
MS
MSO
MS
M
MO
MS
MS
M
MO
MS
MS
SO
MS
MS
MSO
SO
M
MO
M
a M - mud, S - sand, 0 - shell.
VIII-5
-------
Table VIII-2. Summary of Substrate Characteristics in Fahka Union and
Fahkahatchee Bays.
Primary Secondary Tertiary
characteristic characteristic characteristic
Area-ha. % of bay Area-ha. % of bay Area-ha. % of bay
Fahka Union Bay
Mud
Sand
Shell
143.1
20.4
25.8
75.6
10.8
13.6
20.4
67.7
5.8
10.8
35.8
3.1
25.8
16.8
17.6
13.6
16.8
17.6
Totals 189.3 100.0 93.9 49.7 60.2 48.0
Fahkahatchee Bay
Mud
Sand
Shell
Totals
386.8
299.6
56.8
743.2
52.1
40.3
7.6
100.0
183.3
215.2
200.7
599.2
24.7
28.9
27.0
80.6
14.4
11.7
124.8
150.9
1.9
1.5
16.8
20.2 •
Sandy substrates represent 40.3 percent of the primary substrate of
Fahkahatchee Bay and only 10.3 percent of Fahka Union Bay. The largest single
area of sandy bottom is in the eastern side of Fahkahatchee Bay from the mouths
of Fahkahatchee and East Rivers to Fahkahatchee Pass and West Pass.
Sedimentation Rates
Mean organic and total sedimentation rates for both bays are shown as
contours on Figures VIII-5 and 6, respectively. The rates for each station
over the sampling period of July 17 to November 20, 1972 (9 biweekly periods)
were averaged and plotted so that "sedimentation rate contours" could be
developed as shown. Both inorganic and organic rates decrease seaward from
the mouths of the tidal rivers and canals. Large velocity changes occur at
these apertures to the bays. High rates also occurred near the pass between
Fahka Union and Fahkahatchee Bays and also near the pass between Fahkahatchee
Bay and Gate Bay to the east. The high concentrations at the first pass can
be attributed to velocity decreases in the embayment at this pass. Deposi-
tions near the pass between Gate and Fahkahatchee Bays are caused by a
stilling basin created during ebbing tides, by the junction of the flows from
Fahkahatchee River, East River, Gate Bay and othe.r flows from the east.
Fahka Union Bay consistently had greater sedimentation rates than Fahka-
hatchee Bay with mean rates of 160.0 gm/m^/day versus 126.5 gm/m2/day (Figures
VIII-7 and 8). Throughout the sampling period both bays had gradual rate
increases with one two-week period (August 15 - August 29, 1972) having rates
nearly twice that of the previous period. Rainfall was relatively high pre-
ceding and during this period of high sedimentation rates. Apparently a
large quantity of sediment was flushed into the bays at this time.
VIII-6
-------
f\
Figure VIII-5. Mean Organic Sedimentation Rate Isopleths in gm/m /day,
July 17 Through November 20, 1972.
Biweekly mean organic organic fraction, as a percentage of the mean
sediment rate, differed little between the two bays (Figure VIII-9). In fact,
both bays had approximately 19 percent organic content in their sediment traps
in late July and the organic content increased to approximately 33 percent in
early October, then decreased to 27 percent in mid-November. Floating peri-
phyton in a decomposing state was observed in both bays during the fall.
Possibly a good portion of the increased organic percentage can be attributed
to decomposed periphyton.
In order to quantify the sedimentation rates into deposition rates (cm/yr)
the specific gravity of the sediment was required. This determination was made
by use of a pycnometer bottle (Taylor, 1948). The contents, both organic and
inorganic fractions, of all 28 traps for the sampling period of October 10 to
October 27, 1972, were composited in order to get a sufficient quantity of
material for the test. Subsequent analyses yielded a specific gravity of
2.23 for the material. Applying the density to the mean sedimentation rates
for the sampling period (July 17 - November 20, 1972) of 126.5 gm/m2/day for
Fahkahatchee Bay and 160.1 gm/m^/day for Fahka Union Bay discloses deposition
VIH-7
-------
Figure VIII-6. Mean Total Sedimentation Rate Isopleths in gm/m^/day, June 17
Through November 20, 1972.
SEP OCT
1973
Figure VIII-7. Biweekly Total and
Organic Sedimentation Rates for
Fahkahatchee Bay.
Figure VIII-8. Biweekly Total and
Organic Sedimentation Rates for
Fahka Union Bay.
VIII-8
-------
rates of 2.31 cm/yr and 2.94
cm/yr, respectively. Resus-
pension of bottom sediments
is obviously occurring in
these shallow bays. Several
factors such as boat traffic
near the traps, bottom for-
aging marine fishes, shrimp
and crabs using the traps as
living and feeding habitats,
and wind induced wave action
are contributors to turbu-
lence and resuspension.
Sis
3
cc
o
40
O 30
UJ
in
20
FAHKAHATCHEE
FAHKA UNION
AUG
SEP
1972
NOV
Figure VIII-9. Biweekly Mean Organic
Fraction of the Total Trapped Sediment
for Fahkahatchee and Fahka Union Bays.
Commercial fishermen,
when deploying their gear,
probably created some sedi-
ment turbulence near the traps; however, tourist and sport-fishing boat traffic
was at a low during the sampling period. Fishes and other marine animals found
the traps to be suitable homes. Better than 40 percent of the traps serviced
(102/252) yielded at least one inhabitant. Appendix D lists the various species
found in the traps and also the frequency of occurrence. Certainly the above
two contributors are small. However, the third, wind induced wave action, is
a strong parameter in creating resuspension in these shallow bays.
A gradual sedimentation rate increase occurred in both bays in conjunction
with a seasonal decrease in mean daytime tidal stages. These stages made the
bays seasonably shallower and more susceptible to resuspension by strong day-
time winds. Figures VIII-10 and 11 present the variation in day and nighttime
(0600 - 1800 and 1800 - 0600) windspeeds and tidal stages during the sampling
period. The former figure clearly shows that the average daylight windspeed
(7.0 mph) exceeded the nightly averages (5.4 mph) over the span of the stud}'
period. Unlike the wind, a seasonal decrease in daytime mean tidal stage
occurred while the mean nighttime tidal stage increased (Figure VTII-11). The
10
Q
UJ
co
Q
A 8 0
1972
+ 0.5
+0.4
+O-3
ul +0.1
O
fEMSL
05 -0.1
_;
g -0.2
•- -03
-0.4
-0.5
1800-06OO
0600- I8OO
S
1972
Figure VIII-10. Comparison of Day
and Night Wind Speeds. Data From
University of Miami, Rookery Bay
Marine Station.
Figure VIII-11. Comparison of Day
and Night Mean Tidal Stages.
VIII-9
-------
culmination then is that over the sampling period (July 17 - November 20, 1972)
a 30 percent greater turbulence driving force (wind) occurred in the bays dur-
ing the daylight hours than during the night hours. This, coupled with the
accompanying seasonal decrease in daytime tidal stages, appears to have led
to a seasonal increase in sedimentation rates.
A comparison of mean trap depths and observed sedimentation rates between
the two bays reveal that water column depths are a dominant factor affecting
sedimentation rates. The average water column depths at the traps were 87 and
106 centimeters for Fahka Union and Fahkahatchee Bays, respectively. The ratio
of the depths of the two bays, 87/106 (.83), is relatively the same as the
inverse of the ratio of the average sedimentation rates of the two bays,
126.5/160 (.79). Therefore, one might conclude that during this sampling
period the sedimentation rates of the two bays were relatively equal when
normalized for depth.
In order to statistically test the validity of the above hypothesis, a
linear regression analysis of water column depth to sedimentation rates for
each bay was performed. A comparison of the two linear regressions revealed
that the slopes and elevations of the linear equations were not significantly
different at the five percent level. Figure VIII-12 was constructed from data
of both bays with the linear plot described by the equation y = -192.6 X +321.8
(P O.001),
260 -
240 -
N£200
>«.
•S 180
UJ
5.60
g 120
LJ
i 100
^J
CO
_i 80
60
EXPLANATION
O FAHKAHATCHEE BAY
& FAHKA UNION BAY
Y = -I92.6X + 321.8
P< 0.001
0.7
06 0.9 1.0 I.I
DEPTH OF WATER COLUMN (METERS)
1.2
Figure VIII-12. Comparison of Bay Sedimentation Ratss to Water Column Depth
in Fahkahatchee and Fahka Union Bays.
VIII-10
-------
Coring
Three cores were analyzed for organic and inorganic fractions. The cores
were taken at sediment Stations 6 and 14 in Fahka Union Bay, and the third at
Station 14 in Fahkahatchee Bay (Figure VIII- 3). Fahka Union Bay Station 6
and Fahkahatchee Bay Station 14 were chosen because of the relatively equal
organic content found in the sediment traps at these locations, 30.86 percent
and 28.95 percent organic content, respectively. Fahka Union Bay Station 14
was selected as being remote to the man-made canal running through it and,
therefore, generally free of any possible canal effects.
Results of the core
analyses are depicted on
Figure VIII-13. Below a
sediment depth of 15 centi-
meters the cores were nearly
identical in color and organic
content with the organic
percentage leveling off at
four - six percent. How-
ever, above 15 centimeters
the cores differed consid-
erably.
The organic content
of Fahka Union Bay core 6
was much greater than
Fahkahatchee Bay core 14
above 15 centimeters even
though the organic content
of the sediment traps at
these two locations was
nearly equal. The Fahka
Union Bay core had much slower
disruption in the zone from 10
FAHKA- UNION BAY FAHKAHATCHEE BAY FAHKA-UNION BAY
STATION* 14 STATION # 14 STATION J*6
IMETERS)
o
z lo
o
5
o
1- 2Q
0
m
m
5 30
Q
m
X
D- 4O
° /
0 /
O ^SEDIMENT
0 TRAP
O
O
O
0
0
' 0
0
p
0
°0 ^SEDIMENT
0 0 TRAP
8
o
°o°
o
0
o
p
o
o
0
o
o
o
• 0
o
o
0
o
• o
o
G
o
•^SEDI
0 TR
O
UJ
o
10
20 30
0 10 20 30 0 IO
% ORGANIC CONTENT
Figure VIII-13. Organic Content of Bay
Bottom Cores.
organic degradation and also showed a possible
to 14 centimeters.
The core taken at Fahka Union Bay Station 14 (remote to the canal) revealed
very little organic content. Possibly, sediments at this site were quickly
being resuspended and transported away from the site. This is suggested by
the vast difference of the organic content of the sediment trap (26 percent)
and the organic content of the first centimeter of the sediment (five percent).
An unexplained phenomenon appears in all three cores at a sediment depth
of three centimeters down to a depth of six centimeterr.. In this section of
each core the organic content increased rather than decreased. Certainly a
possibility is a greater abundance of infaunal organisms existing in this zone.
DISCUSSION
Fahkahatchee and Fahka Union Bays revealed few surprises, having predomi-
nantly mud bottoms blending frequently with somewhat more sandy bottoms. Oyster
bars overlay these bottoms in several locations in both bays. Fahka Union Bay
bottoms had a larger percentage of total area composed of mud and shell than
Fahkahatchee Bay bottoms.
VIII-11
-------
Sedimentation traps revealed seasonal variation's in organic content of
the sedimenting material, over the sampling period of July 17 - November 20,
1972, with peak organic fractions occurring in early to mid-October. During
this period, decomposing periphyton was observed floating in the bays, thus
obviously contributing to the organic content of the sediment.
Sedimentation rates from the traps coupled with sediment density revealed
deposition rates of 2.31 cm/yr and 2.94 cm/yr for Fahkahatchee and Fahka Union
Bays, respectively. These numbers being unrealistically high lead to the
belief that resuspension of bottom sediments is occurring in the shallow bays.
Water column depth seems to correlate well with the inverse of the sedi-
mentation rate. Mean daylight hour tidal depths decreased seasonably over the
sampling period during which time the sedimentation rate increased. Also, the
ratio of the trap depths in Fahka Union Bay to the trap depths in Fahkahatchee
is relatively the same as the inverse of the ratios of sedimentation rates in
the bays. Fahka Union Bay being the more shallow of the two was more suscep-
tible to wind induced turbulence.
The limited number of cores that were taken did not yield any conclusive
differences in the bays. The top 15 centimeters of the cores had wide ranges
in percent organic content; however, below 15 centimeters all of the cores
revealed relatively equal organic content. All cores disclosed a zone where
the organic content increased with depth. Possibly, macro-organisms could
have caused this irregularity. This zone was approximately three centimeters
below the surface.
VIII-12
-------
IX. DETRITAL TRANSPORT
INTRODUCTION
Detritus in estuary systems is a chief energy link between salt marshes
and bays below them fronting the ocean. Carbon containing detritus in various
forms of decomposition is carried into the bays via tidal creeks which meander
through the salt marshes. During periods of high freshwater runoff these tidal
creeks serve as "catch basins" to funnel the detrital load into the bay systems.
During periods of low runoff they serve to carry tidal water which cyclically
flushes a portion of the salt marsh.
Six major aspects of the detrital transport and decomposition scheme were
investigated during this study: 1) the chemical and physical makeup of the
detrital material; 2) the concentrations and quantities of detritus, total
organic carbon (TOG) and tannin and lignin "like" compounds that were imported
to and exported from the estuarine system; 3) the interchanging of carbon
forms (detritus, TOC, tannin and lignin) as they interacted and were acted upon
in the system; 4) standing crop of zooplankton and phytoplankton; 5) sea-
sonal variations in carbon transport dynamics; and 6) a mass balance of carbon
for Fahkahatchee Bay,
METHODS
In order to obtain an estimate of the detrital (including TOC and tannin
and lignin) transport mechanism and rate, six sampling stations were established
in the salt marsh and bay systems. Four of these sites were in Fahkahatchee Bay
and two sites in Fahka Union Bay (Figure IX-1). Half of the stations were
located at the interfaces between the saltmarsh and the bays and the ramaining
three stations were at the interfaces of the bays and the Gulf of Mexico. Within
the time limitations of the detrital study these six stations were felt to be
reasonably representative of the detrital transport dynamics.
A sampling scheme was devised such that a two man boat crew could alter-
nately draw samples from two stations over a 12 hour tidal period starting at
either high or low slack water. The sampling procedure which follows required
approximately 30 minutes on station. While anchored on station at mid-channel
a submersible pump was employed to pump 300 liters, as measured by 100 liter
containers, of tidal water through two stacked sieves. An integrated sample
of the water column was obtained by varying the elevation of the pump inlet
(2.54 cm I.D.). The detrital material remaining on the two sieves, U. S. Std.
#50 and #230 (297 micron and 63 micron) was then transferred to labeled con-
tainers and returned to the laboratory for analysis. From the filtrate which
passed through the sieves a two liter sample was taken for analysis of the
more soluble constituents.
Several times during each pumping period, a ducted current meter was used
to measure the mid-channel velocity profile. These velocity measurements
coupled with channel cross sectional characteristics and tidal stages enabled
flow quantities to be computed as described earlier in the "circulation pattern
section" of Chapter VII.
IX-1
-------
Figure IX-1. Detrital sampling stations in Fahkahatchee and Fahka Union Bays.
IX-2
-------
Analysis of the samples returned to the laboratory followed American Public
Health Association Standard Methods procedures. The detrital fractions that
remained on the two sieves were analyzed for ash free dry weight. A third
detrital fraction was obtained by passing one liter of the before mentioned
two liter filtrate sample, through an 0.8 micron millipore filter. Ash free
dry weight analysis was subsequently carried out on this nanno fraction.
Subsamples for tannin and lignin and total organic carbon (TOG) analysis
were taken from the two liter filtrate sample. The TOG analysis was performed
using a mass spectrometer by the Chemical Services Branch, EPA^ Athens, Georgia.
Phytoplankton concentrations were determined for two size fractions of the
filterable material. The larger of the sizes being the material that passed
through the 297 micro sieve and was retained on the 63 micron sieve. The
smaller being the fraction that passed through the 63 micron sieve and was
retained on a 0.8 micron filter.
Zooplankton concentrations were also broken down into two size ranges.
The larger being the material that remained on the 297 micron sieve and the
smaller being the material that passed through and was retained on the 63 micron
sieve.
Plankton concentrations were volutnetrically determined by the Sedgwick-
Rafter method and reported as mg/1 on a wet weight basis.
RESULTS AND DISCUSSION
Detrital sampling was rigorously conducted at two periods when the ampli-
tudes of consecutive high or low tides were relatively equal. Nearly 400
separate analyses were performed for each period. The first period was July
18-20, 1972, during which time two stations were sampled each day* The second
period was October 17-18, 1972, with the two Fahka Union Bay stations sampled
on the 17th and the four Fahkahatchee Bay stations sampled on the 18th using
two boats. A two man boat crew was able to make approximately eight sampling
runs on each of the two stations over a 12 hour tidal cycle. Figure IX-2
presents a typical sampling
regime during which time
the beginning and ending
high slack tides were of
relatively equal ampli-
tudes.
+ 06
+0.4
The July sampling -
period was chosen as being »
typical of the summer condi- «
tions and at a time when <
freshwater runoff should ^
have been maximizing the
"flushing" of the uplands
and the salt marshes.
However, as discussed in
Chapter VII, upland runoff
STATION I FAHKAHATCHEE PASS SAMPLING PERIODS
Rf\
STATION 2 WEST PASS SAMPLING PERIODS
TIME (hours)
Figure IX-2. Typical 12 hour tidal cycle showing
detrital sampling periods.
IX-3
-------
was small at this time. The October sampling was preceded by a detailed
circulation study (reference made earlier) in Fahkahatchee Bay. Tidal ampli-
tudes during the circulation study and during the October detritus study were
of the same magnitude and the runoff conditions were unchanged. Therefore,
the circulation patterns could be applied to the detrital load to compute
mass balance in Fahkahatchee Bay.
Table IX-1 presents a summary of the average concentrations of TOG, tannin
and lignin "like" compounds, and three detrital fractions for each of the six
stations. Average concentrations shown are for both ebbing and flooding tides
during the sampling periods in July and October.
Table IX-1. Summary of detrital concentrations sampled during ebb and flood
tides in Fahka Union and Fahkahatchee Bays during July and October 1972.
Average Concentration in mg/1
Station
1
Location
Fahkahatchee Pass
Ebbing Tide
Flooding Tide
Ebbing Tide
Flooding Tide
Date
(1972)
July 19
Oct 18
TOG
(mg/1)
10.50
11.71
8.0
7.57
Tannin and
Lignin
(mg/1)
0.45
0.47
0.39
0.32
297|j.+
Detrital
Fraction3 b
0.54
0.14
0.10
0.40
297n - 63)0,
Detrital
Fraction3
0.50
0.13
63|j, - O.SM.
Detrital
Fraction3
9.10
4.69
1.58
2.30
West Pass
Ebbing July 19 12.25 0.47 0.44
Flooding 15.44 0.42 0.15
Ebbing Oct 18 8.40 0.47
Flooding /.75 0.37
Fahkahatchee River
Ebbing July 18 17.50 1.09 0.27
Flooding 18.76 1.62 0.18
Ebbing Oct 18 11.67 0.96 0.45
Flooding 12.00 0.91 0.21
East River
Ebbing July 18 20.50 1.62 0.23
Flooding 23.67 1.97 0.12
Ebbing Oct 18 13.22 1.18 0.42
Flooding 12.25 1.11 0.27
Fahka Union North
Ebbing July 20 15.00 0.83 0.24
Flooding 14.66 0.88 0.35
Ebbing Oct 17 9.83 0.74 0.49
Flooding 9.33 0.62 1.69
0.092
0.248
0.50
0.15
0.22
0.36
0.43
0.18
0.21
0.22
0.52
0.17
0.41
0.32
0.31
0.49
7.76
2.18
3.12
2.48
9.47
20.06
3.47
2.37
14.24
14.26
2.90
1.65
6.21
7.88
2.60
1.40
6 Fahka Union South
Ebbing
Flooding
Ebbing
Flooding
July 20
Oct 17
14.80
16.43
9.20
7.40
0.69
0.51
0.46
0.44
0.17
0.33
0.34
0.26
0.37
0.19
0.31
0.23
7.90
3.76
4.08
4.76
a Ash free dry weight
b Combination of 297[i -' detrital fraction and 297|i - 63(X detrital fraction
IX-4
-------
Odum and de la Cruz (1967) reported detrital fractions in the ratio of 1,
4, and 95 for material retained on 239 micron, 63 micron, and 0.8 micron filters,
respectively. They were measuring detrital transport off a spartina salt marsh
near Sapelo Island in Georgia. The ratio of the detrital size fractions shown
in Table IX-1 is 5, 4, and 91 with relatively the same size regime as the Georgia
study. Obviously then, the predonderance of the detrital material carried in the
tidal waters in both studies was in a highly fractionated state. This is further
evidenced by the TOG concentrations which were greater than the detrital sieved
fractions. TOG analyses in this study represents carbon which passed through a
63 micron sieve.
Ebbing tide average detrital concentrations (July and October) of TOG, tannin
and lignins, and detrital fractions are presented in Figures IX-3, 4 and 5, respec-
tively. All six stations are shown with their average ebb tide concentrations for
July and October.
20r
15
o>
O I0
o
STA 4
CD
OT
3 =
STA
1972
Figure IX-3. Average ebb tide TOG
concentration at the six sampling
stations in July and October 1972.
A comparison of ebb tide TOG
concentrations at the rivers to those
at the passes in each bay reveal that
concentration gradients across
Fahkahatchee Bay (Stations 3 to 1)
are much greater than those seen in
Fahka Union Bay (Stations 5 to 6).
TOG concentrations decreased 37
percent from salt marsh/bay inter-
face (rivers) to bay/Gulf inter-
face (passes) in Fahkahatchee Bay
while in Fahka Union Bay the TOG
concentration decrease was only
J A S 0
1972
Figure IX-4. Average ebb tide tannin
and lignins concentration at the
six sampling stations in July and
October 1972.
is r
CO
Ul
o
STA 6
STA 3
STA A
STA 5
STA 2
STA I
1972
Figure IX-5. Average ebb tide detrital
concentration at the six sampling sta-
tions in July and October 1972.
IX-5
-------
four percent. This may be due to several complexities such as: 1) flow "short
circuiting" in Fahka Union Bay, due to the relatively deep Fahka Union Canal;
2) upland flows based upon flow/bay area differ considerably between the bays;
3) fractionization of detritus into more soluble forms of carbon.
Short circuiting of tidal waters through Fahka Union Bay can easily occur
due to the open channel characteristics found in that bay. At mean tide the
average depth of the canal at mid-bay is three times greater than the average
depth of the bay. Fahkahatchee Bay has no such channels (see Figures VII-28
and VII-29). Short circuiting via Fahka Union Canal is also seen in the salinity
patterns reported in Figures VII-45 through VII-56. In particular, Figures VII-
51, VII-54, and VII-55 reveal strong convergents of isohalines near the Fahka
Union Canal.
The second factor involves considerations of the bay dilution of in-flowing
concentrations of detritus.
In October, during ebb tides, the flow passing into Fahka Union Bay past
Station 5 was 1,765 X 106 liters or 9.5 X 10° liters/hectare based upon the
surface area of the bay at MSL. While in Fahkahatchee Bay the flow passing
into the bay via Fahkahatchee and East Rivers was only 386 X 106 liters or
0.5 X K>6 liters/hectare. Fahkahatchee Bay thus has a greater diluting capacity
for upstream flows than does Fahka Union Bay.
The third complexity, transformation of carbon forms, will be discussed
later in this section.
TOG and detrital mass flow computations reveal vast differences of carbon
transport into the two bays in terms of mass per unit area of bay. During the
sampled ebb tide in July the mass of TOC entering Fahkahatchee Bay via the two
tidal rivers was 25,600 kg or 34.6 kg/ha based on the bay surface area at MSL.
In October, this value decreased to 6.7 kg/ha while in Fahka Union Bay on an
ebb tide the Fahka Union Canal discharged 75.8 kg/ha and 93.3 kg/ha of TOC
into the bay in July and October, respectively. Therefore, again on an area
basis, Fahka Union Bay received 2.2 times the mass/area and 13.9 times the
mass/area as did Fahkahatchee Bay during ebb tides in July and October, respec-
tively.
Detrital inputs from the tidal rivers in Fahkahatchee Bay during ebb tides
were 23.5 kg/ha and 2.1 kg/ha for July and October, respectively. In Fahka
Union Bay the detrital inputs during the ebb tides were 34.7 kg/ha and 32.2
kg/ha for July and October. Therefore, on an area basis Fahka Union Bay
received 1.5 and 15.3 times the mass/area as did Fahkahatchee Bay in the July
and October sampling periods. However, the above comparisons must be tempered
by the fact that much of the flow into and out of Fahka Union Bay stays within
the channel and does not mix with the water in the bay
Mass transport of the three chemical parameters (TOC, detritus, and tannin
and lignin) into and out of Fahkahatchee Bay was computed from concentrations,
flow measurements and circulation relationships. Table IX-2 reveals only mass
transport of each chemical parameter since the chemical analyses overlap one
another.
IX-6
-------
Table IX-2. Fahkahatchee Bay mass transport of three chemical parameters found
in one tidal cycle in October 1972.
Net Transport3. Net Transport^
Chemical
Parameter
TOC
Tannins and Lignins
Detritus
Via
Import
3806
396
4431
the
kg
Rivers
Export
Via the
Import
855
Passes
Export
6588
782
kg
Net
Bay Balance
Deposition
5286
Export
2782
386
kg
jj Net Transport to the Bay
It is apparent that Fahkahatchee Bay is quite a dynamic system with TOC
and tannin and lignin analysis showing production and detritus analysis showing
deposition in the bay (Table IX-2). Tannins and lignins being degradation prod-
ucts of woody plants cannot be produced in the bay system other than from the
mangrove islands. However, woody plant fragments deposited in the bay as
detrital fractions can undergo decomposition into tannins and lignins and other
carbon bearing compounds.
In order to arrive at a true carbon mass balance for the October tidal
cycle in Fahkahatchee Bay the carbon forms must be adjusted due to overlapping
in their chemical and physical makeup. The TOC analyses reports concentrations
of organic carbon compounds ranging from soluble form to 63 microns in size.
The detrital fractions analyzed were, in decreasing order of size: 1) that
which remained on the 297 micron sieve; 2) that which passed through the 297
micron sieve and remained on the 63 micron sieve, and 3) that which passed
through the 63 micron sieve and remained on a 0.8 micron millipore filter.
Clearly then the TOC analyses contains carbon forms also found in the third
form above. Secondly, a conversion factor is required to convert detritus
concentrations (ash free dry weight) to equivalent carbon concentrations. This
is accomplished by assuming an empirical formula of C^O (40 percent carbon)
for detritus. The carbon mass balance for Fahkahatchee Bay is presented in
Table IX-3.
The tidal rivers yielded large quantities of carbon (Table IX-3) to
Fahkahatchee Bay in October. Most of the material was in carbon forms of
less than 0.8 micron in size with a significant quantity in the size range
of 0.8 to 63 microns and with some forms existing above 63 microns in size.
However, when the tidal water exits the bay via the passes to the Gulf nearly
all (98 percent) of the organic carbon was contained in the 0.8 micron or less
fraction. Thus, large quantities of detrital material were deposited in the
bay and then converted - broken down - into smaller fractions and to more
soluble forms before being exported from the bay.
IX-7
-------
Table IX-3. Fahkahatchee Bay carbon mass balance for one tidal cycle in
October 1972.
Carbon fraction
size range
Net transport —
via the rivers
Import Export
Net transport —
via the passes
Import Export
Net bay balance
Deposition Export
Soluble thru
0.8 micron —
0.8 microncthru
63 micron —
63 micrgn and
larger —
Net carbon
2864
638Q
435
4368 kg
6294 kg
1926 kg
a. Net transport to the bay
_b By assuming an empirical formula of CH~0 (40 percent carbon) for detritus
this fraction is computed from: TOG - (0.40 X (0.8 to 63y fraction));
£ 0.40 X (0.8 to 63y fraction);
d 0.40 X (63 - 297y and 297p + fractions).
The above degradation process can be further amplified by again referring
to the bay circulation pattern study, in particular the bay turnover rates on
page VII-46. From this figure, of an initial concentration entering the bay
in a soluble form, 10 percent will be remaining after 42 tidal cycles, or 21
days (Figure VII-40). From this same figure the half-life of the concentration
is 13 cycles. Therefore, after six days, 50 percent of an initial concentra-
tion would be remaining in the bay available for microbal degradation and
detrital feeders. Since a large portion of the detrital material entering
the bay was not in a soluble form, the half-life, due to flushing would be
greater than six days.
Along with the fractionation of the organic material there appears to be
net carbon production taking place in the bay itself. Evidenced by a net car-
bon export from the bay of 1,926 kg per tidal cycle or 3,852 kg/day. Based
upon the surface area of Fahkahatchee Bay at MSL the net production of carbon
was 0.5 gm/m^ /day.
Table IX-4 presents a summary of the phytoplankton and zooplankton con-
centrations found in the tidal waters at the six sampling stations. Plankton
samples were taken both in the July and October study. However, due to time
limitations only the July samples were analyzed.
No clearcut trends are apparent in the phytoplankton results. A concen-
tration range of 0.066 to 5.166 mg/1 was found in Fahkahatchee Bay while in
Fahka Union Bay the range was only 0.136 to 2.626 mg/1. Little significance
can be derived from these ranges other than the fact that phytoplankton
IX-8
-------
Table IX-4. Summary of phytoplankton and zooplankton concentrations in mg/1
(wet weight basis) found in detrital fractions during mid-July in Fahkahatchee
and Fahka Union Bays.
Station
Location
Phytoplankton*
297y - 0.8y
Zooplankton
29 7y
29 7\i - 63y
Fahkahatchee Pass
Ebbing tide
Flooding tide
West Pass
Ebbing
Flooding
Fahkahatchee River
Ebbing
Flooding
East River
Ebbing
Flooding
Fahka Union North
Ebbing
Flooding
Fahka Union South
Ebbing
Flooding
0.066
0.741
5.
3.
166
790
0.786
0.372
0.296
0.961
0.136
2.626
0.447
1.402
0
0.001
0
0
0
0
0
0.007
0
0
0
0
0.008
0.020
0.005
0.011
0.016
0.025
0.005
0.006
0.003
0.002
0.007
0.008
* Combined sample from 297 - 63 micron and 63 - 0.8 micron detrital fractions.
concentration at the Gulf-Bay interfaces were greater on flood tides than on
ebb tides. It follows that phytoplankton concentrations were greater in the
Gulf than in the bays.
Of the two detrital fractions analyzed for zooplankton concentrations the
larger fraction (297 micron and greater) contained essentially no zooplankton.
However, the smaller detrital fraction (297 - 63 micron) did contain zooplankton
concentrations, but only at level of a few micrograms/liter. Two points are
apparent from the zooplankton concentrations found in the 297 - 63 micron
seston In general, zooplankton concentrations found in Fahkahatchee Bay
were greater than those found in Fahka Union Bay. Evidenced by unweighted
concentration averages of 0.024 mg/1 for the four stations in Fahkahatchee
Bay and 0.010 mg/1 for the two stations in Fahka Union Bay, the above is based
upon all observations (flooding and ebbing tides) at the six stations.
Secondly, zooplankton concentrations on flooding tides were greater than
those for ebbing tides, therefore producing a concentration gradient seaward
from the salt marshes.
IX-9
-------
Tables IX-5 and IX-6 are presented as further evidence that plankton con-
centrations were not abundant in either Fahkahatchee or Fahka Union Bay.
Table IX-5. Zooplankton concentrations found in Fahka Union Bay as sampled
with a 500 micron mesh plankton net.
Date Concentration PPM
May 16, 1971 °
August 2, 1971 °
November 1, 1971 °-006
February 14, 1972 °
(1) Gulf Coastal Fisheries Center, St. Petersburg, Florida
Table IX-6. Phytoplankton collected on June 15, 1972, with a 500 micron mesh
plankton net.
Station
Fahkahatchee Bay
Fahkahatchee River
East River
Fahkahatchee Pass
West Pass
Fahka Union Bay
Fahka Union Canal
Fahka Union Bay
Algae
0 %nl
0
6
0
0
0
Diatoms
104 C/ml
44
358
264
29
201
IX-10
-------
X. WATER CHEMISTRY
INTRODUCTION
Water plays a dominant role in the life cycle, function, and diversity of
all animal species found in south Florida. The chemical characteristics of
water regulates and is regulated by the biota. It therefore follows that this
study would not be complete until the chemical characteristics were evaluated.
The study area is not completely unaffected by man. There are many
hunting camps within the area; some of the water entering the study area
originates from farmlands where fertilizers and pesticides have been used; and
many canals have been dug, draining surface and ground water from the surrounding
lands. Despite these intrusions there are still sections within the study area
where only minor effects of man may be observed.
Data obtained from this study will be used to show areas where degradation
of the environment is taking place, as background information for future studies
and for the suitability of this area for a particular use or multipurpose uses.
METHODS
During November 1971, 16 sampling stations were established (Figures X-l
and 2). These stations were selected as being representative of the aquatic
area and for determining any changes in water chemistry within the drainage
system. Location and description of each station are presented in Appendix E.
Stations 1, 2, and 3 were selected to determine the chemistry of water
entering the study area; Stations 4, 5, 6, 7 and 10 were selected to determine
the chemical characteristics of the fresh water in the study area; Stations 3,
4, and 10 were also selected to determine the chemical characteristics of
water in man-made canals; Stations 8 and 9 were selected to determine the
general chemical features of water at the freshwater-salt marsh interface;
Stations FU-1 and F-l were selected to determine the chemical characteristics
of water entering the Fahka Union and Fahkahatchee Bay, respectively; Stations
FU-2 and FU-3 were selected to determine the chemical characteristics of water
in an "unnatural" bay system (Fahka Union Bay system), which receives most of
its water from a man-made canal (Fahka Union Canal); and Stations F-l and F-2
were selected to determine the chemical characteristics of water in a "natural"
bay system (Fahkahatchee Bay system), which receives its water from natural
drainage.
Sampling and analyses were initiated in December 1971 and continued
through December 1972. The type, frequency, and location of analyses are
presented in Table X-l. In addition to the above samples, weekly samples were
taken at Stations 4 and 10 and monthly samples were taken at Stations 5, 6, and
7 to determine the nutrients transported by water from the uplands into the
estuaries.
Except for dissolved oxygen determination, grab samples were taken and
then transferred into clean, 2-liter, polyethylene bottles. Preservatives
were added to only those samples scheduled for shipment to the Southeast
Environmental Research Laboratory in Athens, Georgia. Analyses conducted in
X-l
-------
Figure X-l. Location of the water quality stations.
X-2
-------
Figure X-2. Location of the water quality stations,
X-3
-------
Table X-l. Physical and chemical analyses.
Parameter
Frequency of measurement
Where determined
Physical
Temperature
Turbidity
Apparent color
Conductivity
Monthly (starting Dec., 1971)
Monthly (starting Dec., 1971)
Monthly (Dec., 1971 to Feb., 1972)
Monthly (starting Feb., 1972)
Monthly (starting Dec., 1971)
Chemical
Alkalinity
Chloride
Dissolved oxygen
Total and soluble
Kjeldahl nitrogen
Nitrate and nitrite
nitrogen
Total and soluble
organic carbon
Total and soluble
phosphorus
Sulfate
Tannin and lignin
"like" compounds
Monthly (starting Dec.,
Monthly (starting Dec.,
Monthly (starting Dec.,
Monthly (starting Mar.,
Once (August, 1972)
Monthly (starting Mar.,
Monthly (starting Mar.,
Monthly (starting Dec.,
1971)
1971)
1971)
1972)
1972)
1972)
1971)
Monthly (starting Jan., 1972)
Field
Naples lab
Naples lab
Field
Naples lab
Naples lab
Naples lab
Naples lab
Naples lab
SERL,
Athens, Ga.
SERL,
Athens, Ga.
Naples lab
Naples lab
Naples lab
the Naples laboratory were usually initiated within a few hours of sampling.
All analyses except for the tannin and lignin "like" test followed one of the
methods given in the EPA manual Methods for Chemical Analysis of Water and
Wastes, 1971 and/or Standard Methods for the Examination of Water and Wastewater,
13th Ed. The specific test used for each parameter is presented in Appendix F.
Mention of trade names or commercial products does not constitute endorsement by
the Environmental Protection Agency.
RESULTS
Physical Characteristics
Apparent color values ranged from 50 to 100 at all freshwater stations
except Station 10 (Fahka Union Canal), where a consistent value of 30 was found.
Values found at the Fahkahatchee Strand stations are not unusual for swamp water.
For marine waters, the lowest value (25) was found in Fahkahatchee and Fahka
Union Bays, and the highest value (120) was found at Station F-l which receives
direct drainage from a mangrove and salt-marsh swamp. A.pparent color values
for each station are given in Table X-2.
X-4
-------
Table X-2. Apparent color values for each station per month.
Station Dec. Jan. Feb.
Station Dec. Jan.
Feb.
1
2
3
4
5
6
7
8
90
100
100
90
50
90
70
80
80
100
100
60
60
90
60
90
90
80
90
50
60
90
50
100
9
10
F-l
F-2
F-3
FU-1
FU-2
FU-3
100
30
120
40
35
50
45
30
100
30
100
30
25
45
35
30
80
30
100
30
25
45
30
25
The conductivity of the freshwater stations ranged from a high of 1.70
mmho/cm at Station 5 to a low of 0.22 mmho/cm at Station 2. The estuarine
stations had their highest values during May (>60 mmho/cm) and their lowest
values during September. At all stations, conductivity was found to be
inversely related to rainfall. During the months of low rainfall, salt
concentrations increased which causes an increase in conductivity; and during
the months of high rainfall, the reverse would occur. The maximum, minimum and
median values are presented in Table X-3.
Table X-3. Range and median values for conductivity in mmho/cm.
Station
Minimum
Median
Maximum
Station
Minimum
Median
Maximum
1
2
3
4
5
6
7
8
0.26
0.22
0.48
0.50
0.31
0.53
0.47
1.30
0.46
0.44
0.60
0.65
0.50
0.57
0.58
15
1.24
1.36
1.44
1.44
1.70
0.70
0.75
31
9
10
F-l
F-2
F-3
FU-1
FU-2
FU-3
1.90
0.81
23
37
38
2
6
35
6
0.81
43
52
52
35
38
52
40
1.62
<60
< 60
<60
<60
< 60
< 60
Values of pH between 7 and 8 were observed at all freshwater stations
during the year except for acidic readings during April at Stations 1, 2, and 3
and a reading of 8.1 at Station 10 during February. Except for a value of 8.8
found in Fakha Union Bay during November, the pH readings for both bay systems
ranged from 7.6 to 8.1. Monthly variations of pH at all stations are presented
in Figure X-3.
There was a definite annual trend in water temperature, with December,
January, and February being the coldest months, and June, July, and August
being the hottest months. The greatest temperature range occurred at Station 5
with a low of 20° C in February and a high of 35.5° C in August. Because of
the greater water flow, Station 10 had the least temperature fluctuation compared
X-5
-------
STA. I
STA. 2
•—•STA. 3
o- — o STA. 4
• STA. 5
o STA. 6
STA. 7
STA. 10
STA.8
o~— o STA. 9
*=
.•.ul»n«.».Mlu.....:v_j
to the other freshwater
stations. There were no NO DATA
major differences in
temperature between the
bays. The annual trend
in temperature for each
station is presented in
Figure X-4.
Median values of
turbidity for the undeveloped
strand area were very low
(2.0 JTU). In areas affected
by dredging and development
operation, which is represented
by the Fahka Union Canal and
Fahka Union Bay, there was an
increase in the median tur-
bidity (72.0 JTU). The Fahka
UnioQ Bay system, which
receives water from the Fahka
Union Canal, had median values
of 5.6 and 5.8 JTU; and at the
same time, the median values
for Fahkahatchee Bay system
were 4.4 and 5.0 JTU. Turbidity
followed no yearly trend,
Maximum, minimum, and median
values of turbidity for each
station are presented in
Figure X-5.
Chemical Quality
At most stations, a
decrease in alkalinity was
observed from April through
June and from September through
October. Mean values for the Figure X-3.
Fahka Union Canal and Barren month.
River stations were higher
than most other stations in the Strand. This is an indication of groundwater
drainage into these canals, since three well stations had total alkalinity
values approximately 150 percent higher than the surface water. Stations 6 and
7 were dry during the period when the other stations had their lowest values;
therefore, their mean values were much higher than the other strand stations.
The range and mean values for Fahka Union Bay were greater than that for
Fahkahatchee Bay. These greater values indicate the influence the Fahka Union
Canal discharge has on Fahka Union Bay. The maximum, minimum, and mean values
are presented in Figure X-6.
•—• STA. F-l
o—-•o STA. FU-I
-*•;£; "'.v.v.vn'w->•*'•••""'.'
. • STA. F-2
o—«o STA. FU-2
• • STA. F-3
o—"O STA. FU-3
pH values for each station per
X-6
-------
o
40
30
20
40
30
20
40
30
20
40
HJ 30
o:
H 20
ui
H
40
30
20
40
30
20
40
30
20
40
30
20
« • STA. I
o—-o STA. 2
•STATION DRY
• • STA.7
o—--o STA.iO
J^
... o-"'
.JO O O
• • STA. 8
o o STA. 9
STA. F-l
STA. FU-I
.-*'•,
At most stations, chloride
concentrations reached a
maximum during May and
minimum during August through
October. This trend excludes
March and April, since no
chloride analyses were
conducted. Chloride levels
varied inversely to the amount
of rainfall for the year, the
concentration of salts during
the dry period, and the
dilution of salts during the
wet period. Chloride concen-
trations in the Fahka Union
Canal at Station 10 exceeded
all other freshwater stations.
The yearly mean chloride
concentration of 72 mg/1 at
Station 10 was the result of
saltwater intrusion occurring
during high tides. Fresh-
water discharges from Fahka
Union Canal effected a
reduction in chloride levels
in the Fahka Union Bay system.
Mean chloride concentration
in the Fahka Union Bay was
approximately 25 percent less
than reported for the
Fahkahatchee Bay. The maximum,
minimum, and mean values for
all stations are presented in
Figure X-7.
C\J • •
DJFMAMJJASONO
Sulfate concentrations
at Stations 1 through 7
ranged from below the minimum
detection limit (1.0 mg/1) to
a maximum of 9.7 mg/1. During
May for the freshwater stations and March to May for the estuarine stations,
no sulfate analysis was conducted. A major difference occurred between the
sulfate concentrations at Station 10 and the other freshwater stations.
Here again, this was mainly due to the intrusion of salt water during high
tides. As with the chloride values, the drainage of the Fahka Union Canal
into the Fahka Union Bay was reflected in the lower maximum and mean values
for the bay compared to the Fahkahatchee Bay. Sulfate concentrations were
nearly 25 percent less in Fahka Union Bay. The maximum, minimum, and mean
values for each station are presented in Table X-4. In calculating the mean,
all values less than the minimum detection limit was treated as zero.
• • STA. F-2
o—-o STA. FU-2
—8-—
• • STA. F-3
o o STA.FU-3
Figure X-4. Temperature values for each
station per month.
X-7
-------
At all stations except 6
and 7, the highest dissolved
oxygen concentrations were found
during March through June. Con-
centrations at some stations
reached super-saturation during
this period (Stations 10 and
F-2 during March; Stations 10,
F-2, F-3, FU-2, FU-3 during
April; and Stations 10, F-2,
FU-2 during May). This was
attributed to increased plant
metabolic activity. The
highest concentrations were
found at Station 10 during
April and Station F-2 during
May, with values of 9.8 and
10 mg/1 respectively. The
lowest concentrations at most
stations were found during
September and October, with
Stations 2 and F-l having
concentrations (<0.50)
below the minimum detection
limit. During most of the
year, oxygen concentrations
in the strand water were
well below saturation
(usually between 1 and 3
mg/1). The abundance of
aquatic vegetation and the
high rate of flow were the
main cause for near- and
super-saturation concen-
trations at Station 10
during most of the year.
The drainage of salt-marsh
water was the main ,cause
for low mean value for
Station F-l compared to
mean values at the other
estuarine stations. No
major difference was found
in dissolved oxygen levels
between the bays. Most
dissolved oxygen samples
were taken between 1000
and 1500 hours, which
biased the results toward
the high side. The
maximum, minimum, and mean
values are presented in Figure X
17
15
3 13
KEY
MAXIMUM
MEAN
I MINIMUM
s
o
I
FAHKAHATCHE
O
z
id
•hi
I 2 S 6
1
"
10 FU-I FU-2 FU-3 F-l F-2 F-3
STATIONS
3489
Figure X-5. The median and range values
for turbidity.
O
100
KEY
MAXIMUM
tO
MINIMUM
Ill
II
!|
12967
+
IE .
ii
10 FU-I FU-2 FU-3 F-l F-2 F-3 3489
STATIONS
Figure X-6. The mean and range values for
total alkalinity.
-8.
X-8
-------
A general increase in the
concentrations of tannin and
lignin "like" compounds at all
stations occurred from June
through November due to flushing
of upland swamps during periods
of heavy rainfall. Concentrations
in the strand ranged from 0.6 to
2.8 mg/1. Runoff from marsh-
lands caused the high mean
concentrations at Stations 8, 9
and F-l. The decrease in mean
concentrations (1.7 mg/1 to 0.4
mg/1) between Stations F-l and F-2
indicates the rapid degradation
of these compounds in an environ-
ment exposed to long periods of
sunlight and high dissolved
oxygen levels. There were no
apparent differences in mean
concentrations between the bays.
The maximum, minimum, and mean
values for each station are
presented in Figure X-9.
Ct TO
O
KEY.
MEAN
' MINIMUM
•f
ffi
£
.1
II
in 14
I
i
li
ti
10 S 9 FU-IFU-2FU-3
STATIONS
Figure X-7. The mean and range values
for chlorides.
Table X-4. Range and mean values for sulfates in mg/1.
Station Minimum Mean Maximum
Station Minimum Mean Maximum
1 <1.0
2 <1.0
3
-------
The highest concentration
of total and soluble Kjeldahl
nitrogen was 6.80 mg/1 and was
found at Station 1 during May,
a period of minimum surface
water flow. The mean concen-
trations for total Kjeldahl
nitrogen for the freshwater
stations ranged from 0.56 mg/1
at Station 10 to 1.66 mg/1 at
Station !„ Total Kjeldahl
nitrogen concentrations for the
two bays were approximately
equal (0.36 mg/1 at Station F-2
versus 0.43 mg/1 at Station FU-2).
The higher mean total Kjeldahl
nitrogen concentration at Stations
8, 9 and F-l compared to the
other estuarine stations was
mainly due to direct marshland
drainage. In most cases, the
soluble fraction of total
Kjeldahl nitrogen represented
from 80 to 90 percent of the
total Kjeldahl nitrogen. The
percent of total soluble
Kjeldahl nitrogen was less
in Fahka Union Bay than in
Fahkahatchee Bay. The
maximum, minimum, and mean
concentrations for each station
are shown in Figure X-10. In
calculating the mean, all
values less than the minimum
detection limit (0.06 mg/1)
were .treated as zero. The
percentage of the total
Kjeldahl nitrogen reported as
total soluble Kjeldahl
nitrogen for each station is
presented in Table X-6.
Sporadically throughout
the year, the total phosphorus
concentration in most areas
fell below the minimum detection
limit (0.02 mg/1). The highest
concentration of 0.40 mg/1 was
found at Station 9 during April,
a period of minimum flow. Mean
concentrations for total
phosphorus for all stations
ranged from 0.02 to 0.08 mg/1,
with Station 1 having the
highest mean concentration.
(O
CO
I
05
KEY
MAXIMUM
— MEAN
'MINIMUM
iu
" •
I
il
i
t
I
I
i
^ I MINIMUM DETECTION LIMIT 1--J
IOFU-1 FU-2FU-3 F-l F-2 F-3
STATIONS
Figure X-8. The mean and range for
dissolved oxygen.
tf 1.0
* as
z 0.2
KEY
MAXIMUM
.
3
ac
|i
i(.
I
II
1 I1
•••
IOFU-IFU-2 FU-3 F-l F-2 F-3
STATIONS
Figure X-9. The mean and range values for
tannin and lignin "like" compounds.
X-10
-------
Table X-5. Nitrate and nitrite nitrogen concentrations for the month of
August 1972 in mg/1.
Station
NO-NO nitrogen
O j£
Station
NO--NO nitrogen
•3 2.
I <0.01*
2 0.01
3 0.03
4 0.08
5 <0.01
6 **
7 -*•*
8 <0.01
9
10
F-l
F-2
F-3
FU-1
FU-2
FU-3
0.01
<0.01
<0.01
<0.01
<0.01
0.01
<0.01
<0.01
* Less than the minimum detection limit of 0.01 mg/1.
** Stations were dry.
Mean concentrations for
the Fahkahatchee Strand stations
were much higher than the mean
concentration in the Fahka Union
Canal (Station 10). Mean concen-
trations in the strand ranged
from 0.04 to 0.08 mg/1, compared
to 0.02 mg/1 at Station 10. The
Fahka Union Bay maintained a
lower mean concentration of total
phosphorus than did Fahkahatchee
Bay. Less than 50 percent of
the total phosphorus was in the
soluble form at all stations
except Stations F-l and FU-3,
where it was 60 percent. The
maximum, minimum, and mean
concentrations for each station
are presented in Figure X-ll.
In calculating the mean, all
values less than the minimum
detection limit (0.02 mg/1)
were treated as zero. The
percentage of total phosphorus
reported as, total soluble
phosphorus for each station is
presented in Table X-7.
2.5O
230
2 10
-- I 90
E"
— I.7O
Z
O 150
K- I.3O
Z
_l I 10
I
Q O.90
tu
2 o.ro
< O.50
P 0.30
O.IO
KEY
I MAXIMUM
— MEAN
• MINIMUM
g
Z
"006 MINIMUM DETECTION LIMIT, f
cc
111
111
<
S
I
i
i
i
IOFU-1 FU-2 FU-3 F-l F-2 F-3
STATIONS
Figure X-10. The mean and range values for
total Kjeldahl nitrogen.
The highest average total organic carbon (TOG) concentration (26 mg/1)
was found at Stations 6 and 9. During April, the highest TOC concentration
(36 mg/1) was found at the latter station. Also during this same month, the
highest phosphorus concentration and next-to-the-highest Kjeldahl nitrogen
concentration was found at Station 9. This occurred during a period of minimum
water flow. Of the freshwater stations, the Fahkahatchee Strand stations had
the highest average TOC concentrations (15 to 26 mg/1), the Barron River stations
X-ll
-------
Table X-6. The percentage of total Kjeldahl nitrogen that is total soluble
Kjeldahl nitrogen.
Station
Percentage
Station
Percentage
1 81
2 84
3 87
4 90
5 90
6 79
7 83
8 88
9
10
F-l
F-2
F-3
FU-1
FU-2
FU-3
82
95
85
100
79
82
81
68
were next (13 to 17 mg/1), and
Station 10 had the lowest of
12 mg/1.^ No apparent differences
in TOG concentrations occurred
between the bays. The high
average T.OC concentration of
17 mg/1 at Station F-l was
caused by the direct drainage
of a salt marsh. No insoluble
form of total organic carbon
was detected. The maximum,
minimum, and mean concentrations
for total organic carbon for
each station are presented in
Figure X-12.
Nutrient Transport
Originally, four of the
16 water quality stations were
incorporated into a special
study to determine surface
Q.
KEY
MAXIMUM
—-MEAN
'MINIMUM
i—••*-«) 02
£
lU-
£
d.
I Z 5 6
Id FU-I FU-2 FU-3 F-l F-2 F-3
STATIONS
Figure X-ll. The mean and range
total phosphorus.
3489
values for
Table X-7. The percentage of total phosphorus that is total soluble phosphorus
Station
Percentage
Station
Percentage
1
2
3
4
5
6
7
8
38
30
14
55
30
40
17
20
9
10
F-l
F-2
F-3
FU-1
FU-2
FU-3
7
0
60
32
48
13
22
60
X-12
-------
water transport of nutrients from
the freshwater areas to the
estuaries. Of these four
stations (10, 9, 8 and 4), two
(Stations 8 and 9) were found
to be subject to tidal
influence. Consequently,
these two stations were
replaced with the three Janes
Drive stations (Stations 5, 6
and 7). Stations 10 and 4
(Fahka Union Canal and Barron
River respectively) were
sampled weekly, and the stations
on Janes Drive (Stations 5, 6
and 7) were sampled monthly.
Each sample was analyzed for
total organic carbon, total
Kjeldahl nitrogen, and total
phosphorus. The results of
each analysis are reported as
either the average kg/day of
nutrients per week transported
by the Fahka Union Canal and the
Barron River or as the average
kg/day per month transported
across Janes Drive. Because
nutrient concentrations were
not linearly related to flow (Figu
during the week or month that the
36
34
32
3O
26
26
24
C 2Z
| 20
*~ 18
O
O 16
14
12
10
8
6
4
KEY
MAXIMUM
i!
'
I
(
ii
12567 IO FU-I FU-2 FU-3 F-l F-2 F-3 3469
STATIONS
Figure X-12. The mean and range values for
total organic carbon.
re X-13), an average flow, based on flows
sample was taken, was calculated. Then, the
1-2 p
0.6
0.4-
0.2
a'
O <§>
O
•& 0
EXPLANATION
O FAHKA-UNION CANAL
A BARRON RIVER
Q JANES DRIVE
o o
e o
&
o
200 400 600 800 IOOO
FLOW (I/MC)
Figure X-13. Total Kjeldahl nitrogen versus flow.
ItDO
MOO
X-13
-------
average flow was multiplied by each nutrient concentration for the Fahka Union
Canal and Barren River and by the average of each nutrient concentration of
Stations 5, 6, and 7 for Janes Drive to give the average transport (kg/day) of
each nutrient for each week or each month, as the case may be. This study
started in March 1972 and continued through December 1972.,
Because of an absence of surface flow, no nutrients were transported
across Janes Drive from the middle of March to the middle of August. The
greatest amount of nutrients was transported by the Fahka Union Canal during
the week of June 20, with an average of 1,200 kg/day total Kjeldahl nitrogen,
290 kg/day total phosphorus, and 52,000 kg/day total organic carbon. This
was the result of heavy rains received from Hurricane Agnes during the week.
More nutrients were also transported by Fahka Union Canal for the year than by
either of the other two systems (Fahkahatchee Strand or the Barren River). The
average amount of nutrients (kg/day) transported by or across each freshwater
area during 1972 is given in Table X-8. The average amount of nutrients (kg/day)
transported by Fahka Union Canal and Barren River for each week and across Janes
Table X-8. The average transport (kg/day) of total Kjeldahl nitrogen, total
phosphorus, and total organic carbon into the estuary from three freshwater
areas during 1972.
Station Nitrogen Phosphorus Carbon
Fahka Union Canal
Barren River
Janes Scenic Drive
260
130
58
20
10
2.2
7,600
2,900
1,100
Drive for each month is presented in Figures X-14 through 16 and is given in
Tables X-9 through 11.
X-14
-------
H-
era
i-i
l-ti fD
§
AVERAGE TRANSPORT OF TOTAL KJELDAHL NITROGEN (IO«xkg/DAY)
«
Ln
&?
i-i
fD
ro >
Hi fD
fD PJ
CD (TO
P1 fD
(a rt
rt i-i
fD pi
i-i 3
CO
H O
fD i-i
P) ft
CO
O
H- Hi
3
rt rt
O O
fD <->•
CO fD
ft h-1
3
I-1-
rt
fD
P
OQ
(O
ro
-------
O OFAHKA-UNSON CANAL
O O BARRON RIVER
JANES DRIVE
Figure X-15. Average transport of total phosphorus (kg/day)
from three freshwater areas into the estuary.
X-16
-------
AVERAGE TRANSPORT OF TOTAL ORGANIC CARBON
X
I
Mi (D
i-i
h- '
d- O-v
IT1 •
H
CD
n> t>
-------
Table X-9. The average transport of total phosphorus (kg/day).
Date
Mar. 2
9
16
23
30
Apr. 4
11
18
26
May 4
8
15
22
30
June 7
12
20
26
July 3
11
19
25
Fahka
Union
Canal
3.4
0.93
0.61
0.73
0.44
0.93
1.6
0.88
0.0
0.0
0.15
12
, 8.2
2.1
37
14
290
33
10
7.0
14
ND*
Janes
Barren Scenic
River Drive
0.8 0.67
2.8
0.94
1.4
0.76
0.75 0.0
2.8
1.2
1.7
0.44 0.0
0.56
1.1
0.94
0.55
0.58 0.0
1.7
26
6.5
13
6.4 0.0
18
75
Date
Aug. 1
7
15
23
29
Sept. 6
12
20
25
Oct. 3
10
16
24
31
Nov. 6
15
20
28
Dec. 5
11
18
26
Fahka
Union
Canal
22
15
84
18
25
37
21
9.2
7.1
17
35
15
17
16
6.6
ND*
13
5.6
8.4
3.3
8.0
5.4
Barren
River
7.0
8.3
15
16
15
26
20
14
18
16
29
21
13
5.7
7.8
8.0
ND*
17
4.8
6.0
6.5
3.9
Janes
Scenic
Drive
0.32
1.3
4.2
11
4.1
* No data
X-18
-------
Table X-10. The average transport of total Kjeldahl nitrogen (kg/day).
Date
Mar. 2
9
16
23
30
Apr. 4
11
18
26
May 4
8
15
22
30
June 7
12
20
26
July 3
11
19
25
Fahka
Union
Canal
98
43
31
42
19
30
14
2.5
0.0
0.0
3.2
210
43
45
310
230
1,200
950
350
150
320
ND*
Janes
Barren Scenic
River Drive
52 12
84
66
18
13
58 0.0
54
30
9.0
9.4
9.3 0.0
6.6
4.7
5.7
12 0.0
37
120
160
120
110 0.0
140
ND*
Date
Aug. 1
7
15
23
29'
Sept . 6
12
20
25
Oct . '3
10
16
24
31
Nov. 6
15
20
28
Dec. 5
11
18
26
Fahka
Union
Canal
450
300
320
480
330
480
520
410
360
540
510
440
240
190
200
ND*
370
260
180
110
100
150
Barren
River
230
170
250
170
150
390
400
350
330
380
250
280
160
120
150
ND*
160
200
150
150
76
120
Janes
Scenic
Drive
6.8
77
77
250
160
* No data
X-19
-------
Table X-ll. The average transport of total organic carbon (kg/day).
Date
Mar. 2
9
16
23
30
Apr. 4
11
18
26
May 4
8
15
22
30
June 7
12
20
26
July 3
11
19
25
Fahka
Union
Canal
1,100
ND*
860
510
480
600
430
220
0
0
59
2,800
2,100
780
8,700
6,800
52,000
34,000
17,000
7,700
7,200
ND*
Janes
Barren Scenic
River Drive
800 170
ND*
760
480
460
1,200 0.0
940
360
.0 310
.0 160 0.0
190
140
120
120
320 0.0
480
4,200
4,300
3,100
3,400 0.0
4,300
6,500
Date
Aug. 1
7
15
23
29
Sept. 6
12
20
25
Oct. 3
10
16
24
31
Nov. 6
15
20
28
Dec. 5
11
18
26
Fahka
Union
Canal
10,000
9,200
9,000
14,000
17,000
17,000
14,000
8,300
7,900
6,600
12,000
7,500
5,700
3,600
4,300
ND*
5,700
5,000
3,800
3,000
2,700
3,800
Barron
River
6,300
4,700
5,700
5,100
6,900
7,400
7,800
6,300
6,200
4,800
2,900
5,500
2,500
1,900
1,800
2,100
3,100
2,900
2,600
2,400
2,300
2,900
Janes
Scenic
Drive
170
1,600
960
4,300
3,300
* No data
X-20
-------
XI - METALS AND PESTICIDES
INTRODUCTION
This section documents the. distribution and concentration of -metals and
pesticides found in the Barren River, Fahkahatchee Strand, and associated estu-
aries. At present, no significant use of pesticides or metals was found within
the study area; however, the study area has been affected to some extent by
extensive agriculture applications of these chemicals in immediate upland areas.
The primary objective of this study was to determine if unnatural levels of pes-
tecides and metals were present and to obtain background data for future studies.
METHODS
During December 1971, seven stations which were representative of the study
area were established—two stations (Stations 4 and 7) in the undeveloped areas
and five stations (Stations 1, 2, 3, 5, and 6) in the developed areas. These
stations were sampled on a quarterly basis for 1 year starting January 24, 1972.
Station locations are given in Table XI-1 and shown on Figure XI-1. Methods for
collection and analysis, and a list of each pesticide and metal analyzed for are
discussed below.
Table XI-1. Location of stations for pesticide and metal analyses.
Station Location
1 GAG Canal beside Alligator Alley (S. R. 84) 16.3 km (10.1
miles) east of the west-end toll booth. A freshwater station.
2 Bridge on Alligator Alley (S. R. 86) 31.4 km (19-5 miles)
east of the west-end toll booth. Off-shoot of the Borrow
Canal draining southward into the Fahkahatchee Strand. A
freshwater station.
3 Barren River Canal in front of Jane's Resturant on S. R. 29.
A freshwater station.
4 The first wooden bridge (bridge E) 4.8 km (3 miles) on
Janes Scenic Drive. A freshwater station on the outer
edge of the Fahkahatchee Strand.
5 Bridge #?1 on U. S. 41 (Tamiami Trail). An off-shoot of
the Borrow Canal draining southward into a salt marsh. It
has brackish water and subject to tidal influence.
6 Fahka Union Bay west of Black Marker #35, which is a
shallow area away from the main channel.
7 Approximate middle of Fahkahatchee-Bay.
XI-1
-------
Figure XI-1. Location of the pesticide and -metal stations.
XI-2
-------
Metals
Water and sediment samples were collected at each station. Water samples
were taken directly in clean, 1-quart, wide"juouth jars; and the sediment sam-
ples were taken by use of an Ekman dredge and then placed into the same type
jar. Upon return to the Naples laboratory, the water samples were preserved
with 5 ml HN03. No preservative was used for the sediment samples. The sam-
ples were then immediately shipped to the Southeast Environmental Research
Laboratory (SERL) in Athens, Georgia, for analysis.* The EPA's atomic absorp-
tion method for total metals was used for each metal analysis. Both sediment
and water samples were analyzed for arsenic, chromium, copper, lead, nickel,
zinc, manganese, cadmium, and mercury. Only water samples were analyzed for
iron,
Pesticides
Water, sediment, and fish samples were collected at each station. Water
samples were taken directly into a pesticide-free, 1-quart, wide-mouth glass
jar with aluminum foil-lined lids. Sediment samples were taken using an Ekman
dredge and then placed in the same type jar. Fish samples were collected using
a seine or "hook and 3ine" and then immediately put on ice. Upon return to the
Naples laboratory, the water and sediment samples were immediately shipped to
the SERL in Athens, Georgia. The fish were kept on ice until they were counted,
weighed, and measured; and then they were frozen and subsequently packed in dry
ice and sent to the Athens laboratory. EPA's gas chromatographic method was
used for each pesticide analysis. In addition to the quarterly samples, vari-
ous other animal samples, which were dead when found, were collected and analyzed
for the same pesticides as the fish.
The pesticide analysis for each type sample and the minimum detection lim-
its are presented in Table XI-2.
RESULTS
Metals
Analytical results for both water and sediment samples are presented in
Tables XI-3 and 4 respectively.
Arsenic was not detected,in the water at any station; however, at most of
the stations, some was detected in the sediment. At Stations 3 and 6, arsenic
was detected 100 percent of the time; but at Stations 4, 5, and 7, none was
detected. The highest concentration of 11.5 Mg/g was found at Station 3 dur-
ing April.
No cadmium was detected in the water at Stations 1 through 4, and it was
found only during January at Station 5. However, it was found in almost all
the samples from Stations 6 and 7, with the highest value of 50 ug/1 being
detected during January and April at Station 7. At the Stations where cadmium
was detected, a decreasing trend was seen from January through November,
* Analysis performed by Chemical Services Branch, Surveillance and Analysis
Division, Region IV, Environmental Protection Agency, Athens, Georgia.
XI-3
-------
Table XI-2. Pesticide analyses and minimum detection limits.
Minimum detection
. Minimum detection limit in sediment
Pesticide limit in water and biological
(yg/1) samples (yg/kg)
Aldri-n
Lindane
Chlordane
Chlorobenzilate
DDD
DDE
DDT
Dieldrin
Endrin
Heptalchlor Epoxide
Heptachlor
Methoxychlor
Toxaphene
Diazinon
Guthion
Methyl Parathion
Parathion
Malathion
Ethion
Trithion
Mirex
Dimethoate (cygon)
Lannate*
PCB's
0.005
0.002
0.05
0.5
0.01
0.01
0.02
0.01
0.02
0.01
0.005
0.1
0.25
0.2
0.5
0.2
0.02
0.04
0.02
0.1
0.05
1
2
0.05
0.25
0.1
2.5
25
0.5
0.5
1.0
0.5
1.0
0.5
0.25
5.0
15
10
25
1
2
4
4
2
1
50
100
5.0
* Analyzed for in sediment and water only.
Cadmium was detected in the sediment at every station except Station 7.
The highest median value of 3.9 yg/g and the highest concentration of 6.0 yg/g
were found at Station 2. There was also a decreasing trend through the year.
Chromium was detected in the water only at Station 5 during July at a con-
centration of 60 yg/1; however, it was detected in the sediment at every station
during the year. The highest median values of 34.8 Ug/g and 36.0 yg/g were
found at Stations 2 and 6 respectively, but the highest concentration of 67 yg/g
was found at Station 1 during January.
No copper was detected in the water at Station 1 through 4. However, it
was detected at Station 6 and 7 throughout the year and at Station 5 during
November. The highest concentration of 60 yg/1 was found at Station 7 during
November.
Copper was detected in the sediment at every station throughout the year.
The highest concentration of 18,5 yg/g was found at Station 1 during July. The
highest median values of- 9.8 yg/g and 8.6 jig/g were found at Stations 1 and 3
respectively, and lowest median value of 2.8 yg/g was found at Station 7.
XI-4
-------
Table XI-3. Metal analysis for water samples in-yg/1.
Nickel
Station Jan.
1 <50
2 <50
3 <50
4 <50
5 <50
6 115
7 140
Apr.
<100
<100
<100
<100
280
630
670
July
<50
<50
<50
<50
50
110
210
Nov.
<20
<20
_
50
120
200
Jan.
:"
< 10
< 10
30
30
Manganese
Station Jan.
1 <10
2 <10
3 <10
4 <10
5 130
6 40
7 50
Apr.
<30
50
<30
<30
200
90
80
July
<50
<50
<50
<50
70
<50
70
Nov.
60
80
-
_
50
40
70
Jan.
<10
<10
<10
<10
30
40
50
Copper
Station Jan.
1 < 10
2 < 10
3 < 10
4 < 10
5 <10
6 30
7 30
Apr.
<30
<30
<30
<30
<30
36
43
July
<50
<50
<50
<50
<50
<50
50
Nov.
<20
<20
_
_
<20
<20
60
Jan.
<50
<50
<50
<50
<50
<50
<50
Zinc
Apr.
30
40
<30
50
100
100
100
July
<50
<50
<50
70
<50
<50
100
Nov.
50
<20
-
30
<20
40
Cadmium
Apr.
<25
<25
<25
<25
<25
45
50
July
<25
<25
<25
<25
<25
<25
35
Nov.
<20
<20
-
-
<20
20
30
Chromium
Apr.
<50
<50
<50
<50
<50
<50
<50
July
<50
<50
60
<50
<50
<50
<50
Nov.
<40
<40
-
-
<40
<40
<40
Lead
Jan. Apr. July
250 <100 <100
<50 <100 <100
<50 <100 <100
<50 <100 <100
<50 200 <100
330 370 100
330 300 270
Mercury
Jan. Apr. July
<0.2 <0.5
<0.2 <0.5
<0.2 <0.5
<0.2 <0.5
<0.2 <0.5
<0.2 <0.5
<0.2 <0.5
Arsenic
Jan. Apr. July
<20
<20
<20
<20
<20
<20
<20
Nov.
<100
<100
-
<100
250
500
Nov.
<0.2
<0.2
-
-
<0.2
<0.2
<0.2
Nov.
<29
<29
-
-
<29
<29
<29
Iron
Station
1
2
3
4
5
6
7
Jan.
-
-
-
-
-
-
Apr.
<300
800
380
<300
5,600
460
540
July
140
300
460
100
570
220
270
Nov.
1,150
620
-
-
170
330
380
NOTE: < indicates results were below minimum detection limit, which changes
periodically depending upon the condition of the atomic absorption unit.
- indicates analysis was not conducted.
XI-5
-------
Table XI-4. Metal analysis .for sediment samples in yg/g.*
Nickel
Zinc
Lead
Station Jan. Apr. July Nov. Jan. Apr. July Nov. Jan. Apr. July Nov.
1
2
3
4
5
6
7
58
50
26
14
26
20
8
8.0
12.0
4.0
6.0
3.6
8.0
5.6
14.4
10.8
6.4
3.6
2.8
14.0
6.0
13.0
6.7
10.3
<1.5
-
12.8
7.3
63
74
30
11
16
5
4
12.4
19.2
34.8
16.6
10.8
7.2
5.4
9.2
8.8
62.7
10.0
4.0
10.6
4.4
7.3
5.9
101
<1.5
-
11.2
5.5
40
230
28
17
52
20
16
30.8
44.7
64.8
28.0
1,079
9.6
9.4
16.0
28.8
160
7.2
4.0
24.3
9.2
25.3
7.9
41.7
2.4
-
23.9
8.0
Manganese
Cadmium
Mercury
Station Jan. Apr. July Nov. Jan. Apr. July Nov. Jan. Apr. July Nov.
1
2
3
4
5
6
7
152
56
34
12
38
20
8
16
46.5
20.4
20.6
14.4
20.2
15.6
15.6
13.2
46.7
6.7
5.6
36.0
15.6
27.7
18.9
44.0
2.0
-
39.9
17.0
3
5
2
1
1
2
<1
1.2
6.0
0.8
1.2
0.8
0.8
<0.8
0.8
2.8
0.8
<0.8
<0.8
1.2
<0.8
2.0
0.8
1.6
<0.8
-
0.8
<0.8
0.47
.08
.07
.07
<.04
.26
.09
<.04
<.04
.10
<.04
.22
.04
.06
<.10
.20
.20
.14
.19
1.42
.44
<.04
<.04
<.04
<.04
-
< .04
.08
Copper
Chromium
Arsenic
Station Jan. Apr. July Nov. Jan. Apr. July Nov. Jan. Apr. July Nov.
1
2
3
4
5
6
•7
10
6
9
2
6
3
3
5.6
6.0
6.0
4.8
6.8
3.2
2.7
18.5
5.2
9.6
4.4
1.6
5.0
2.6
5.3
2.0
9.9
1.2
-
4.6
2.9
67
47
20
8
12
20
8
15.2
25.6
14.0
14.8
4.8
27.2
17.6
12.4
34.4
14.8
8.8
11.9
52.7
14.4
24
32
28
2
-
44
20
.9
.0
.2
.8
.3
.5
<1.6
8.4
2.6
<1.4
< 1.3
6.4
<1.4
<2.0
<2.0
11.5
<2.0
<2.0
3.4
<2.0
2.6
<2.0
5.8
<2.0
<2.0
3.5
< 2.0
2.8
<2.0
3.0
<2.0
-
4.8
< 2.0
* Analyzed and reported on a dry-weight basis except mercury which was reported on
a wet-weight basis.
NOTE: < indicates results were below minimum detection limit, which changes
periodically depending upon the condition of the atomic absorption unit.
- indicates analysis was not conducted.
Iron was consistently detected at Stations 2, 3, 5, 6, and 7 and in over
50 percent of the samples from Stations 1 and 4. No unusual concentrations
were found at any station except at Station 5 where 5.60 mg/1 was found during
April. In the study area and surrounding areas, iron is -used in many fertiliz-
ers and fungicides (ex. furbam for both domestic and commercial use). Also, in
this area, iron concentrations of 900 ug/1 were found in shallow wells; and
higher concentrations were found in deeper wells (Finney and Miller, 1960).
In the water, 250 vg/1 of lead was found at Station 1 during January, 200
yg/1 was found at Station 5 during April; and lead was detected in all samples
XI-6
-------
at Stations 6 and 7, The highest concentration of 500 yg/1 was found at Sta-
tion 7 during November. When lead was detected, it was from two to ten times
higher than the USPHS Drinking Water Standards (0.05 mg/1).
Lead was detected in all sediment samples. The highest concentration of
1,079 yg/g was fotind at Station 5 during April. The two highest -median values
of 53.3 and 52.0 iig/g were found at Stations 3 and 5 respectively, and the two
lowest median values of 21.1 and 9.3 vg/g were found at Stations 4 and 7 respec-
tively. In the study area, lead is found in the pesticide (lead arsenate) and
as lead tetraethyl in car exhaust and boat fuel.
Manganese was detected in the water 75 to 100 percent of the time at Sta-
tions 5 through 7 and 50 percent or less at the other stations. The highest
concentration of 200 yg/1 was found at Station 5 during April. Manganese was
detected in the sediment at every station during the year. The highest con-
centration of 152 yg/g was detected at Station 1 during January. The highest
median value of 45.3 yg/g was found at Station 2.
Mercury was not detected in the water at any time, but it was detected at
least once during the year in the sediment at each station. The highest con-
centration of 1.42 yg/g was found at Station 6 during July.
Nickel was not detected in the water at Stations 1 through 4, but it was
consistently detected at Stations 6 and 7 and 75 percent of the time at Sta-
tion 5. The highest concentration of 670 yg/1 was found at Station 7 during
April.
Nickel was detected in the sediment at each station throughout the year
except during November at Stations 4 and 5. The highest concentration of 58
yg/g was found at Station I during January. The highest median values of 13.7
and 13.4 yg/g were found at Stations 1 and 6 respectively, and the lowest median
values of 4.8 and 3.6 yg/g were found at Stations 4 and 5 respectively.
The highest concentration of zinc of 100 yg/1 was found at Stations 5, 6,
and 7 during April and also at Station 7 during July. Zinc was not detected
at Station 3. It was detected in all sediment samples except samples from Sta-
tions 4 and 5 during November. Station 3 had the highest sediment median value
of 48.7 yg/g, which was three to ten time higher than the median values for the
other stations. Station 7 had the lowest median value of 4.9 yg/g.
The median values for each metal at each station is given in Table XI-5.
None of these metals were found to be used extensively within the study area;
therefore, it is believed that the unnatural concentrations of most of these
metals was caused by the drainage from upland agricultural industry. The
unnatural concentration of lead is thought to be caused by the combination of
car exhaust and drainage from upland agricultural industry.
Pesticides
Of the 24 pesticides analyzed for in the water and sediment samples, none
were detected in the water at any tijtje; and only six pesticides (DDT, DDD, DDE,
Methoxychlor, PCB's, and Dieldrin) were detected in the sediment. Dieldrin was
detected at Station 6 (13 ^yg/kg) during April and at Station 3 (2.3 vg/kg) dur-
ing July, Methoxychlor was detected at Station 3 (2.1 Vg/kg) during April.
XI-7
-------
Table XI-5. Median values for metal analysis of the sediment in yg/g.
Stations
Metal
Arsenic
Cadmium
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Zinc
1
2.3
1.6
20.0
7.8
28.0
21.8
<.07
13.7
10.8
2
<2.0
3.9
33.2
5.6
36.7
32.7
0.06
11.4
14.0
3
4.4
1.2
17.6
9.3
53.3
45.3
0.08
8.4
48.7
4
2.0
0.9
8.4
3.2
12.1
9.3
0.06
4.8
10.5
5
C2.0
0.8
11.9
6.0
52.0
14.4
0.19
3.6
10.8
6
4.1
1.0
35.7
3.9
21.9
28.1
.15
13.4
8.9
7
<2.0
<0.8
16.0
2.8
9.3
15.6
.08
6.6
4.9
The concentration of DDT, DDD, DDE? and PCB's for each sediment sample is given
in Table XI-6. The highest concentration of DDT, DDD, and DDE was 160, 160,
and 96 yg/kg respectively; and they were found at Station 3 during April. The
highest concentration of PCB's (130 yg/kg) was found at Station 2 during January.
Both Stations 2 and 3 receive runoff from farmland in the Immokalee area where
pesticides are used extensively.
Table XI-6. Results (yg/kg) of four pesticides detected in the sediment samples.*
DDT
Station
1
2
3
4
5
6
7
Jan.
ND**
ND
2.7
0.78
ND
ND
ND
Apr.
3.2
2.9
160
0.22
0.69
0.58
0.22
July
ND
ND
ND
ND
ND
ND
ND
Nov.
2.2
ND
ND
ND
ND
ND
ND
Station
1
2
3
4
5
6
7
Jan.
0.48
ND
13
0.68
1.8
0.61
ND
DDE
Station
1
2
3
4
5
6
7
Jan.
0.14
ND
4.2
0.39
1.1
0.23
0.09
Apr.
2.3
2.4
96
0.09
0.29
0.28
0.21
July
ND
0.73
ND
ND
ND
ND
ND
Nov.
0.40
1.1
1.1
ND
0.37
0.89
ND
Station
1
2
5
4
5
6
7
Jan.
ND
130
ND
8
ND
ND
ND
DDD
Apr.
3.2
2.9
160
0.22
0.69
0.58
0.22
PCB
Apr.
20
5.7
5.2
1.5
4.1
ND
2.6
July
ND
1.0
ND
ND
ND
ND
ND
July
ND
ND
ND
ND
ND
ND
ND
Nov.
0.88
1.4
2.6
ND
0.44
1.7
ND
Nov.
ND
ND
11
ND
4.2
ND
ND
* Sediment was calculated on a dry weight basis.
** ND - None detected'.
XI-8
-------
In the biological samples, only six pesticides (DDT, DDE, DDD, PCB's,
Dieldrin, and Mirex) of the 23 analyzed for were detected. Of these six, only
four pesticides (DDT, DDE, DDD, and PCB's) were detected with significant
frequency. Dieldrin was found in two fish during July and three fish during
November. Mirex was observed once in a bobcat found dead on "U.S. 41 near
Ochopee, Florida. When a pesticide was detected, it was usually less than 50
jjg/kg. The pesticide DDE was found in the highest concentrations of 320 yg/kg
and 380 jig/kg; these concentrations were found in a Florida gar and a Bonito
fish respectively. The concentration for each pesticide detected in each bio-
logical sample is presented in Tables XI-7 through 11. The scientific name for
each fish taken for pesticide analysis can be found in Appendix MM.
The consistent absence of detectable pesticides in the water and the low
concentrations found in the sediment and most of the biological samples suggest
a lack of significant pesticide contamination at the present time. However,
studies have shown that residue levels of chlorinated hydrocarbon pesticides in
the water can build up in fish and other aquatic organisms to toxic concentra-
tions for animals which feed upon those organisms (Cope, 1966; Butler, 1966;
Keith, 1966). None of these pesticides were found to be used extensively within
the study area; therefore, it is believed that most of the pesticides detected
entered the study area from drainage of the upland agriculture industry, mainly
by way of the Barren River.
XI-9
-------
Table XI-7. Pesticide analyses on biological samples for January.
Size Range
Station
No.
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
_6
6
6
7
7
7C
8
8
a Based
b NT> - N
Type of Total Length
Sample
Largemouth bass
Brook silversides
Spotted sunfish
Florida gar
Golden shiner
Redear sunfish
Brook silversides
Gambusia
Florida gar
Flagfish
Gambusia
Marsh killifish
Gambusia
Florida gar
Largemouth bass
Sheepshead
Mangrove snapper
Eastern oysters
Sheepshead
Mangrove snapper
Eastern oysters
Snook A
Snook B
on whole fish analyses
'one detected.
(mm)
84
32-82
122-134
362
162
179
40-69
19-40
486
22-36
22-46
83
17-44
471
197
256
215
-
380
240
-
228
156
, calculated
DDE
UgAg
11.
6.8
4.2
8.6
0.6
2.3
9.5
26.
200.
8.1
18.
23.
16.
24.
9.2
8.0
5.8
0.7
12.
7.8
1.7
42.
40.
on a wet
ODD
yg/kg
2.8
0.7
1.1
52.
0.3
4.8
5.7
15.
94.
6.8
11.
14.
12.
20.
11.
7.2
3.2
1.0
3.1
3.8
1.8
SI.
56.
weight
DDT
UgAg
7.0
2.3
2.3
25
NDD
2.3
1.7
9.1
40.
4.6
10.
11.
6.5
4.0
5.4
2.0
1.6
0.76
2.8
2.0
1.0
11.
12.
basis.
PCB's
Vg/kg
35.
12.
12.
38.
8.6
22.
22.
42.
110.
24.
31.
52.
28.
34.
104.
37.
47.
22.
47.
35.
16.
85.
120-
c Station 8 was a station on Henderson Creek.
XI-10
-------
Table XI-8, Pesticide analyses on biological samples for April.'
Size Range
Station Type of Total Length
No. Sample (mm)
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
6
6
6
7
7
7
Bowfin
Florida gar
Yellow bullhead
Bluegill
Warmouth
Spotted sunfish
Bluegill
Redear
Gambusia
Spotted sunfish
Striped mullet
Gambusia
Sail fin molly
Striped mullet
Redfin needlefish
Sheepshead
Gaff tops ail catfish
Eastern oyster
Sheepshead
Grevalle jack
Eastern oyster
561
662
313
101-236
116
86-201
122-135
98-129
18-34
112-120
378
19-38
28-56
87
316-318
387
428
-
300
288
*~
DDE
y.g/kg
8.7
320.
3.2
14.
3.4
7.4
6.1
6.7
34.
8.9
7.4
36.
14.
2.2
24.
13.
12.
18.
8.
32.
12.
ODD
yg/kg
9.1
59.
3.9
7.8
ND
3.8
8.8
3.4
48.
8.0
13.
13.
19.
1.0
22.
9.3
12.
1.6
1.2
4.8
1.3
DDT
yg/kg
6.3
4.5
2.3
17.
ND
6.8
8.6
4.0
7.2
5.7
5.5
5.7
13.
ND
16.
3.8
7.6
7.4
6.3
7.2
ND
PCB's
y.g/kg
NDb
110.
ND
26.
ND
23.
16.
ND
38.
13.
76.
11.
14.
ND
ND
ND
ND
ND
ND
76.
20.
a Whole fish analysis calculated on a wet weight basis.
b ND - None detected.
XI-11
-------
Table XI-9. Pesticide analyses on biological samples for July.'
Station
No.
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
5
6
6
6
6
7
7
Type of Sample
Bluegill
Spotted sunfish
Brook silverside
Redear sunfish
Bluegill
Largemouth bass
Florida gar
Largemouth bass
Bluegill
Striped mojarra
Redfin needlefish
Tidewater silverside
Snook
Redfin needlefish
Eastern oyster
Blue crab
Spadefish
Fantail mullet
Eastern oyster
Blue crab
Alligator - A.c
B.d
Size Range
Total Length (mm)
189
76-132
26-55
152
139-168
331
464
62-99
175
242
196-353
22-61
371
261-340
-
156-184
(carapace width)
163-190
195
-
138-175
(carapace width)
1,520
DDE
lag/kg
55.
2.8
1.4
2.9
8.1
73.
300.
3.7
4.9
19.
7.3
12.
13.
1.4
0.76
2.2
2.8
2.6
1.1
2.1
16.
6.4
ODD
Pg/kg
30.
Nlf
1.3
2.4
3.9
49.
85.
ND
3.7
17.
4.4
7.2
5.4
ND
ND
4.1
1.5
2.9
ND
ND
ND
ND
DDT
Mg/kg
10.
ND
2.0
ND
4.9
30.
53.
ND
4.4
4.0
2.6
2.7
ND
ND
ND
2.3
ND
ND
ND
ND
6.8
ND
Dieldrin
Vig/kg
0.9
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
2.3
ND
PCB's
Hg/kg
25.
ND
16.
17.
18.
80.
ND
ND
ND
ND
36.
21.
ND
ND
ND
ND
ND
ND
ND
ND
3.9
ND
* Samples based on wet weight basis
ND - None detected
, A. - Subcutaneous fat and muscle tissue from tail
B. - Subcutaneous fat from abdominal region
XI-12
-------
Table XI^IO. Pesticide analyses on biological samples for November.
Size Range
Station Total Length DDE
No. Type of Sample (mm) yg/kg
1
1
2
2
2
3
3
3
4
4
4
5
5
5
6
6
6
6
7
7
7
7
Blue gill
Spotted sunfish
Re dear sunfish
Bluegill
Spotted sunfish
Spotted sunfish
Redear sunfish
Bluegill
Gambusia
Yellow bullhead
Bluegill
Flagfish
Sailfin molly
Florida gar
Common jack
Sheep she ad
Mangrove snapper
Oysters
Grevalle jack
Mangrove snapper
Oysters
Marine catfish
75-140
85-95
134-159
132-165
123-138
84-115
101-107
105-113
16-42
36-48
143
17-33
20-52
505
327
215
209-280
-
446
289
-
~
NDb
1.6
12.0
2.3
1.6
9.0
2.6
0.75
1.8
ND
12.
4.1
11.
41.
24.
2.1
14
1.3
23.
6.6
0.9
110.
ODD
yg/kg
ND
ND
8.5
1.5
1.6
3.0
0.83
ND
ND
ND
3.3
3.7
10.
16.
4.0
1.9
6.5
ND
11.
ND
ND
9.1
DDT
y-g/kg
ND
3.2
29.0
8.3
3.6
3.8
2.3
ND
ND
ND
6.6
2.5
14.
4.4
3.7
2.7
5.1
ND
5.2
ND
ND
1.9
Dieldrin PCB's
y.g/kg yig/kg
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
3.2
ND
ND
2.2
ND
1.6
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
22.
20.
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
a Whole fish analysis, calculated on a wet weight basis.
b ND - None detected.
Table XI-11, Pesticide analyses on special biological samples.3
Station
U.S. 41
(Ochopee, Fla.)
Shell Island Rd.
Naples, Fla.
Gulf of Mexico,
1.6 km off Marco
Island, Fla.
Date of
Sample Type of Sample
4/26/72 Bobcat
a . sub-epithel-
ial fat
b . mesenteric
fat
c . liver
8/08/72 Armadillo (1150 g.)
a . liver
b. fat tissue
(subcutaneous)
c. muscle (thigh)
11/16/72 Bonito fish (765 T.L.
mm)
DDE
yg/kg
106.
52.
39.
5.4
ND
ND
380.
DDD
Vg/kg
4.3
ND
12.
ND
ND
ND
54.
DDT
yg/kg
9.8
ND
23.
1.1
ND
ND
170.
Mirex Dieldrin
yg/kg ye/kg
4.0
23.
1.5
ND
ND
ND
ND
11.
77.
9.3
ND
ND
ND
ND
PCB's
ug/kg
43.
230.
19.
ND
ND
ND
300.
Calculated on a wet weight basis
bND - None detected
XI-13
-------
XII - PLANT COMMUNITY BIOMASS AND METABOLISM
MARINE GRASSES AND ALGAE
Introduction
The role of benthic marine plant communities in estuarine productivity
requires little substantiation (Odum, 1968; Phillips, 1960; and Thorhaug and
Stearns, 1972),, These plants, apart from providing shelter and food for
invertebrates and juvenile fishes, initiate biotic cycling of inorganic carbon.
To date little information has been gathered on the rate at which these plants
synthesize and use organic carbon. Hence, to properly judge an estuary's
productivity would first require experimental work to determine net and gross
primary productivity of these estuarine plant communities.
This section reports the experimental work conducted in Fahka Union and
Fahkahatchee Bays in the Ten Thousand Island, Florida, for the purposes of
determining the annual net and gross primary productivity of the benthic
plant communities. Also, included is an assessment of various environmental
factors effecting primary production.
To determine annual net primary productivity of the bay systems, the
biomass accumulation approach was adopted. This method relies on biomass
harvesting techniques the merits of which are discussed by Westlake (1971).
Sampling was conducted in January, March, July, September, and November 1972
to determine changes in standing crop biomass. In determining gross primary
productivity, specially designed benthic respirometers (Lucas and Thomas,
1970) were employed to measure dissolved oxygen exchange rates for the benthic
communities during the spring and fall seasons 1972.
Methods
Survey of standing crop biomass in January first involved the development
of a general map showing the distribution of plant densities in the two bays.
This was accomplished during the bottom characterization study (Chapter VIII).
At each of the established grid intersections, a bottom grab, obtained with a
Petersen dredge, was sieved for its content of benthic vegetation. A No. 18
U. S. Standard sieve effectively separated vegetation from sediment. Vegetation,
if present, was reported on a qualitative basis as trace, medium, or heavy
amounts. This classification scheme was used to construct a map showing the
distribution of plant densities in the bays. Soon after the qualitative cate-
gories were quantified in the following manner.
Bay areas designated as supporting trace, medium, or heavy amounts of
vegetation were sampled with a screened scoop device which effectively
harvested plants within a 0.28 m^ plot. The plant samples included roots
and rhizomes which extended to a sediment depth of approximately 6 to 7 cm.
Plant material was then separated from remaining sediment with a No. 18 U. So
Standard sieve, stored on ice, and returned to the laboratory at Rookery Bay
for processing. Additional samples of vegetation were obtained, preserved,
and later forwarded for identification to the Biological Services Branch,
Surveillance and Analysis Division at the Southeast Environmental Research
Laboratory in Athens, Georgia.
XII-1
-------
Processing of samples entailed first a sorting of the plant material
by vegetative types - marine grasses, red marine algae, and green filamentous
algae„ Subsequently, all material was oven dried for three to four days at
103°C., cooled to room temperature and weighted. Average concentrations
(gm dry wto/m^) were calculated for each of the identified density categories.
The sampling scheme heretofore described was altered for the remaining
surveys (March, July, September, and November). An Ekman dredge modified
with a two meter shaft and special release mechanism was used for sampling.
Two bottom grabs were obtained and composited at each of the grid intersections
sampled. The composite sample represented two 15.2 X 15.2 cm bottom plots„
As before, sediment penetration was restricted to 6 - 7 cm» Processing of
the samples remained the same as in the January survey.
Vegetational concentrations were plotted at each of the respective grid
intersections and appropriate isopleths drawn for the July and September
surveys. The March and November survey results were not treated in this
manner because they served as only spot checks on changes in standing crop
biomass and limited to a fewer number of sampling sites. The isopleths
indicate an average, weighted for bottom area and concentration, for each of
the three substrate and vegetative types of Fahkahatchee and Fahka Union Bays.
Benthic gross primary production rates were determined for mud-sand and
sand-mud bottoms in Fahka Union and Fahkahatchee Bays. Mud-sand bottoms were
characterized by a soft substrate mainly composed of mud with fine sand as
a secondary feature. The sand-mud substrate, of course, was the reverse in
compositional features, but still soft in texture. The productivity studies
were conducted during the spring of 1972 (February 15 to 19 and March 7 to 26)
and during the fall of 1972 (October 26 to November 5). Study sites are
indicated in Figure XII-1.
Production rates were determined using a black and a clear plexiglass
chamber to measure the exchange rates of dissolved oxygen (DO) between bay
water and the benthic community. Community in the sense used in the present
study includes the biotic and abiotic elements, such as chemical oxygen
demand, associated with the bottoms. Both chambers were identical in construc-
tion (Figure XII-2).
The chambers were constructed to maximize the ratio of bottom area to
water volume in order to permit rapid detection of dissolved oxygen changes.
Also, the construction elevated the circulation system sufficient distance
from the bottom to minimize resuspension of fine bottom particles during the
operation^ Each chamber when sitting on a horizontal plane was slightly
bias by design to allow gases to collect at one corner and escape via a check
valveo The plexiglass portions of the chamber were bolted to a stainless
steel flange and cutting edge which effectively seals the test water within
the chambers when they are lowered onto any soft bottom sediment. Each
chamber, in position, covered 0.186 m of bottom and contained 14=5 liters of
bay water. The entrapped water was circulated through the chamber and a
clear plexiglass manifold by a 12 volt DC submersible pump mounted externally
to the chamber. No seal monitoring system was used rluring the operation of
the chamber when in place. Past experience in using the system in other
studies indicated that the substrate types encountered in the present study
would permit an effective chamber tu sediment seal. Changes in dissolved
XII-2
-------
FAHKA UNION BAY
EXPLANATION
LJ6HT INTENSITY STATION
GPP SITES
O MUD--SAND BOTTOM
A SAND-MUD BOTTOM
FAHKAHATCHEE ••;
Figure XII-1. Locations of Stations Used in the Benthic Plant Gross Primary
Productivity Studies and in the Surveys of Light Extinction in Fahka Union
and Fahkahatchee Bays, 1972
STAINLESS STEEL
SEALING FLANGE
Figure XII-2. Benthic Respirometer.
XII-3
-------
oxygen content of the test water were measured and recorded with portable DO
analyzers and strip chart recorders.
Prior to placement of chamber on the bottom, the apparatus was first
suspended in the water just beneath the surface and the circulation pump
engaged. The system was operated a few minutes until the chamber and manifold
were purged of trapped air bubbles. At this time, the DO meter was adjusted
to the surface water dissolved oxygen content which was first determined with
a Winkler titration. Once the system was correctly adjusted and operating,
the circulation pump was disengaged and the unit carefully lowered onto the
bottom. There, the chamber rested for several minutes to allow for any
resuspended bottom particles to settle again.
Resuspension of bottom material while placing and operating the unit was
always a threat to the success of test. Extensive resuspension would lead to
an accelerated oxygen demand (Lucas and Thomas, 1970). To account for
possible resuspension the unit's circulation was first stopped just prior
to its retrieval so that water trapped in the clear circulation manifold
could be examined for suspended or settled solids.
Test runs with the opaque (black) chamber provided estimates of the
benthic community respiration rate. The clear chambers yielded net oxygen
production rates for its enclosed benthic vegetation. In either case, the
actual rate of change of DO with time was determined from regression analyses.
To convert oxygen production or respiration rates to rates of carbon assimila-
tion on an aerial basis the following calculation was used:
g . K • °-06V = gm C/m2/hr
Where:
|3 = Regression Coefficient (mg 02/1/min)
V = Volume of Chamber (14.5 1)
ry
A = Bottom Area Confined to Chamber (Do 186 m )
K = Mole Fraction of C12 and 02 (0.375)
To establish a response relationship of gross primary productivity to
light, visible light illumination at the chamber depth was monitored with a
marine photometer3 equipped with sea and deck cells. In advance of the study,
the cells were checked for linear response to visible light intensities and
the photometer readings calibrated to express radiant energy (within visible
light range of 400 - 700 nm) as langleys per minute. The response and calibra-
tion curve is shown in Figure XII-3„
Additional work was performed to establish the fraction of total solar
radiation represented by visible light spectra. For these experiments, total
solar radiation was measured by a pyrheliometer under various conditions of
a The instrument was constructed by personnel of the project to meet special
application,,
XII-4
-------
cloud cover and sun elevation during the month of August 1972. Simultaneously,
a spectral radiometer was deployed to measure visible light radiation. The
faction of photosynthetically available energy included in the total daylight
radiation received at Roekery Bay was 48.47 percent (Figure XII-4). This
value is considered as a yearly constant for South Florida.
Fahka Union and
Fahkahatchee Bays are
relatively shallow
estuaries. The inter-
action of wind velocity
and tidal changes in
depth were major
factors effecting ~ fl
turbidity, hence, |~
effecting light
extinction^ To arrive
at a model describing
the relationship of
light extinction to
wind and depth, surveys
were conducted during
1972 and 1973 to measure
light extinction in
both bays,
*Q4
00
i
*
03
¥•0.0020 + 0.0I33X
P O0OOGOI
K>0 ZOO 300
SEA CELL OUTPUT t a AMPS)
400
Diminution of light
intensity through the
water column was deter-
mined in both bays at
selected locations
(Figure XII-1) and on
various dates. The
marine photometer used
in the productivity
tests was used for
the surveys. Using
the light intensity
data coupled with
measurements of depth,
the following equation
provided a measure
of light extinction
coefficients (K).
Figure XII-3, Response and Calibration Curve of
the Marine Photometer.
I,Or
,0.8
.*0.6
3 0.4
UJ
m
oc
to
Y = 0.4847X + 0.0141
P<0.0O000I
04 0.6 O.8 1.0 1.2 1.4
PYRMEU0METER READING (ly/min)
1.6
Figure XII-4« Total Solar Radiation Represented
by Visible Light.
XII -5
-------
K = I0 Where: Iz = Light Intensity at Z (n amps)
I0 = Light Intensity at Surface (p, amps)
Z = Depth (cm) of Water Column
Scheduling of surveys was random from the standpoint of considering
time of day, tides, and winds; hence, light extinction was measured under a
wide range of conditions»
RESULTS
Benthic Vegetation and Standing Crop Biomass
Qualitative samples of the two bays yielded the following list of plant
species:
Marine grasses
Tracheophyta
Potamogetonacea
Dipl.anthera wrightii
Hydrocharitaceae
Thalassia t e studinum
Halophila engelmannii
Red macroalgae
Rhodophyta
Ceramiaceae
Gracilaria verrucosa
Gra.clla.ria sjoestedii
Gracilaria foliifera
Dasya harveyi
Green filamentous algae
Chlocophyta
Cladophoraceae
RhizocIonium hookeri
Taxonomically, vascular plants designated as marine grasses are not members
of the family Graminae. However, for purposes of this study they were
referred to as such.
In January, green filamentous algae maintained the greatest standing
crop biomass in both bays. Definitive concentration (gm dry wt/m^) for the
categories of densities indicated in Figure XII-5 are following: Trace - 0.6,
medium - 8.9, and heavy - 46.9.
Comparative to the green filamentous algae., the standing crop of grasses
and red macro-algae biomass were sparse during January. As before, the dis-
tribution and abundance of these two plant groups were broken down into cate-
gories of trace, medium, and heavy densities (Figures XII-6 and 7). Respectively,
definitive concentrations were 0.6, 1.6, and 8.2 for the red macro-algae, 0.3
and 1.6 gm dry wt./m^ for the grasses. Estimates of total standing crop biomass
in January for the three plant types follows:
XII-6
-------
EXPLANATION
/ / / / / TRACE
MEDIUM
//////////////////// HEAVY
Figure XII-5. General Distribution and Mean Concentration of Green Filamentous
Algae in Fahka Union and Fahkahatchee Bays, January 1972
Fahkahatchee Bay -
Fahka Union Bay
green filamentous algae
grasses
red macro-algae
green filamentous algae
grasses
red macro-algae
13,541 gm dry wt.
9,756 gm dry wt.
3,371 gm dry wt.
16,849 gm dry wt.
776 gm dry wt.
192 gm dry wt.
A summary of the January 1972 standing crop biomass survey is provided
Appendix G.
in
In March, spot checks on standing crop biomass during the course of the
benthic chamber work yielded an average concentration of 6.2 and 14.9 gm
dry wt/m^ for Fahka Union and Fahkahatchee Bays, recpectively. These values
represent an average concentration of the combined grasses and red and green
algae. Sampling was limited to those areas of the two bays designated in
the January survey as supporting trace, medium, and heavy concentrations
of plants„
XII-7
-------
FAHKA UNION BAY EXPLANATION^
TRACE
/////////////////,. MEDIUM
HEAVY
Figure XII-6. General Distribution and Mean Concentration of Red Macro-Algae
in Fahka Union and Fahkahatchee Bays, January 1972.
Major changes in total standing crop biomass occurred between July and
September 1972 (Table XII-1). In Fahkahatchee Bay both marine grasses and
red macro-algae showed steady increases in biomass whereas filamentous algae
declined. Marine grasses totaled 169,930 kg in July and 245,685 kg in
September; likewise, the red macro-algae increased from a total of 13,402 kg
in July to 56,120 kg in the fall. Green filamentous algae decreased from a
maximum in January to 1,817 kg in July and a low of 597 kg in September.
In Fahka Union Bay red macro-algae were the only plant type to show a
steady increase in biomass from January to September. In July the standing
crop of red macro-algae increased nearly 11 times over the total 192 kg
reported for January. The September biomass exceeded by about 580 times the
January estimate.
Little change occured with the marine grasses of Fahka Union Bay in
July and September - 3,733 kg vs 3,788 kg, respectively; although, the two
values were about 5 times that reported for January.
XII-8
-------
FAHKA UNION BAy EXPLANATION
TRACE
'//////////////////, MEDIUM
Figure XII-7. General Distribution and Mean Concentration of Marine Grasses
in Fahka Union and Fahkahatchee Bays, January 1972.
Unlike Fahkahatchee Bay, green filamentous algae in Fahka Union Bay
first decreased in biomass to an apparent low of 9,238 kg in July but
promptly increased to 15,472 kg in September -- a level comparable to
estimates in January (16,849 kg).
The general distribution and centers of abundance of benthic vegetation
in Fahka Union and Fahkahatchee Bays during the July and September surveys
are shown in Figures XII-8, 9, 10, and 11. Three basic types of substrates
characterized the two bays -- predominantly, mud, sand-mud mixes, and shell
(Figure XII-12). The former substrate having the greatest representation of
either bay, supported the greatest standing crop in July. In Fahka Union
Bay, the mud substrate was colonized principally with red and green algae;
whereas, in the other bay the mud bottom support mainly marine grasses.
The details of standing crop biomass data vs substrate types are given in
Appendices H, I, J, K, L and M.
XII-9
-------
Table XII-1. Comparison of Standing Crop Biomass Surveyed in Fahkahatchee
and Fahka Union Bays, 1972.
Vegetational Types Jan. Mar.a July Sept. Nov.a
Fahkahatchee Bay
Marine grasses 9,756 169,930 245,685
Red macroalgae 3,371 13,402 46,120
Green filamentous algae 15,541 3,634 597
Total standing crop
biomass (kg dry wt.) 26,668 186,966 292,372
Avg. bay cone. -
(gm dry wt./m ) 3.6 14.9 25.2 39.3 75.6
Fahka Union Bay
Marine grasses 776 3,733 3,788
Red macroalgae 192 2,229 108,422
Green filamentous algae 16,849 9,258 15,472
Total standing crop
biomass (kg dry wt.) 17,817 15,200 127,682
Avg. bay cone.
(gm dry wt./nT 9.4 6.2 8.0 67.4 66.8
a
Results of abbreviated surveys during GPP studies.
The immediate question at this point is why does the mud bottom of
Fahka Union Bay support mainly a benthic community of algae and not a marine
grass community such as found in Fahkahatchee Bay? Discussion of this point
will be delayed until after the results of the gross productivity have been
reported.
The abbreviated survey in November during the benthic chamber studies
yield the following average plant biomass for the three plant types combined
for Fahka Union and Fahkahatchee Bays: 66.8 and 7.5.6 gm dry wt/m2, respectively.
The November average for Fahka Union Bay would be about the same for the
September survey. However, the November average for Fahkahatchee Bay would
be nearly twice that determined for the September study.
In summary, Fahka Union Bay was characterized by a mud bottom which
supported a benthic plant community primarily comprised of red and green
algae. Standing crop biomass remained nearly steady for the first six months
of the year, followed by an outburst of algal mass accumulation in the fall
XII-10
-------
Figure XII-8. General Distribution and Centers of Abundance for Marine Grasses
in Fahka Union and Fahkahatchee Bays, July 1972. Mean Concentration (gm Dry Wt/M^)
months (Table XII-1). Fahkahatchee Bay, although it maintained a benthic plant
community comprised'mainly of algae in January, the marine grasses steadily
increased in abundance and became the major plant community of the estuary.
Unlike Fahka Union Bay, marine grasses were the principal plant type inhabit-
ing the mud and mud-sand mixed substrate.
Light Extinction Studies
From February through April 1972 and from October 1972 to'February 1973,
surveys were conducted to assess light extinction characteristics of water
masses in Fahka Union and Fahkahatchee Bays and to develop a model for predicting
the visible light delivery to the benthic plant communities for photosynthesis.
During the two survey periods no difference in daytime turbidity (as
measured by light extinction coefficient) could be reported between bays or
between substrates within Fahkahatchee Bay. These judgments were based on
results of analyses of variance using the reported mean extinction coefficients
given in Appendices N, .0 and P.
XII-11
-------
Figure XII-9. General Distribution and Centers of Abundance for Marine Grasses
in Fahka Union and Fahkahatcb.ee Bays, September 1972. Mean Concentration
(gm Dry Wt/m2) .
A linear regression model was developed for relating light extinction
to depth. Based on 87 paired observations of light extinction coefficients
(K) and depth of water column (Z), the following regression equation had a
correlation coefficient of 0.64 with a t value of 6.00 at the 95 percent confi-
dence level.
K = -1.40 X 10~4 Z + 3.49 X 10'2
Where: K = Coefficient of Light Extinction (cm) at Bottom
Z = Depth (cm) of Water Column
Wind as an integrated factor with depth was viewed as the principal
agent of vertical mixing in these shallow bays. Resuspension of sediments
led to changes in the light scattering properties of the water; hence, the
effects of wind on the diminution of light as it passed through the water
column were manifested in the measurements of light extinction,, The reported
XII-12
-------
Figure XII-10. General Distribution and Centers of Abundance for Red and Green
Algae in Fahka Union and Fahkahatchee Bays, July 1972. Mean Concentration
(gm Dry Wt/m^).
equation, therefore, includes wind, not as an implicated term, but as an
integrated factor with depth.
Gross Primary Productivity Studies
Gross primary productivity (GPP) was assessed for benthic plant communities
growing in Fahka Union and Fahkahatchee Bays during the months of February and
March (spring) and again in October and November (fall) 1972,, Using light and
dark chambers, rates of benthic community oxygen demand and benthic plant
production of oxygen were determined and are reported in Appendix Q.
From these data, rates of GPP and visible light transmission were converted
to grams carbon per meter square per hour and langleys per hour, respectively,
and reported in Tables XII-2 and 3. Data from these two tables were used to
establish linear regression equations describing response of benthic GPP to
visible light illumination at the bottom. The model used in these calculations
is following:
XII-13
-------
Figure XII-11. General distribution and centers of abundance for red and
green algae in Fahka Union and Fahkahatchee Bays, September 1972. Mean
concentration (gin
Y = mX + b
where: Y = GPP (gm/C/m2/hr), b = Y intercept
X = visible light (ly/hr)
m = slope
Under survey conditions, the linearity of the GPP response to visible light
transmission was good. In all cases, correlation coefficient of regression
was greater than 0.88. At this point an examination of the reported regression
equations (Figures XII-13 and 14) will indicate that in all cases the Y inter-
cept is not equal to zero. Theoretically in a broad sense, the intercept should
be zero on the basis that gross primary productivity ceases when light goes to
zero. Unfortunately, theoretical consequences do not take into consideration
possible threshold responses of productivity to light and also some unavoidable
XII-14
-------
EXPLANATION
MUD
SHELL
SAND
Figure XII-12. Principal bottom substrate types characterizing Fahka Union
and Fahkahatchee Bays, 1972.
Table XII-2. Gross primary productivity rates for Fahka Union and Fahkahatchee
Bays with respect to bottom types and rates of visible light transmission,
Spring 1972.
Date
Feb 15
Feb 15
Feb 15
Feb 16
Feb 17
Feb 18
Feb 18
Feb 18
Mar 7
Mar 7
Mai* R
nar O
MQT* ft
nar o
Mar 8
Mar 24
Fahka Union
GPP
-2 9
10 gm C/m /Hr
0.9685
0.2925
0.2967
0.9536
-
_
I
-
Bay - MSa
Bottom
Light (Vis.)
Ly/Hr
3.7512
0.8562
0.2141
3.2351
-
_
_
„
-
Fahkahatchee
GPP
10" 2 gm C/m2/Hr
-
-
-
-
0 . 8417
1.0217
2.7596
1.9941
_
_
.
0.1821
Bay - SMb
Bottom
Light (Vis.)
Ly/Hr
-
-
-
-
0.4362
0.8179
2.4000
2.4338
_
_
-
0.1894
Fahkahatchee
10" 2 gm C/m2/Hr
-
-
"*
~
1.5697
8.0293
7.4504
-
-
4.1983
4.4738
3.2377
Bay - MSa
Bottom
Light (Vis.)
Ly/Hr
-
"
"
1.5632
4.6073
4.3338
•
-
1.8184
2.6936
2.6172
a Mud-sand bottom
b Sand-mud bottom
XII-15
-------
Table XII-3« Gross primary productivity rates for Fahka Union and Fahkahatchee
Bay with respect to bottom types and rates of visible light transmission,
Fall, 1972.
Date
Oct 26
Oct 26
Oct 27
Oct 27
Oct 29
Oct 29
Oct 29
Oct 29
Oct 31
Oct 31
Oct 31
Oct 31
No- 1
riov 2
Nov 2
Nov 2
Nov 4
Nov 4
Nov 4
Nov 4
Nov 5
Fahka Union
GPP
10"2 gm C/m2/Hr
1.8143
0.7023
1.2240
6.0312
-
-
-
-
-
-
.
-
14.7264
-
-
-
8.3206
9.3121
21.3636
22.1880
1.0017
Bay - MSa
Bottom
Light (Vis.)
Ly/Hr
1.8705
2.9272
2.4652
3.4508
-
-
-
-
-
-
-
-
6.8578
-
-
-
5.9174
6.3127
13.6014
12.2077
1.6202
Fahkahatchee
2 GPP •)
10 gm C/m /Hr
_
_
.
-
-
-
-
1.4224
1.0159
0.3944
0.1236
_
_
-
-
0.7032
0.5931
_
_
3.8276
Bay - SMb
Bottom
Light (Vis.)
Ly/Hr
.
-
.
-
5.0000
4.6374
3.4980
1.1239
„
-
.
2.9063
1.9863
_
7.4664
Fahkahatchee
GPP
10"2 gm C/m2/Hr
10.8240
5.6770
8.9704
4.1876
_
5.7566
4.5679
3.8794
-
Bay - MSa
Bottom
Light (Vis.)
Ly/Hr
3.2780
1.9241
3.1973
1.4648
1.3379
1.3368
0.8979
-
a Mud-sand bottom
b Sand-mud bottom
experimental error will always accompany field measurements. Reviewing Tables
XII-2 and 3 will indicate that the data base for these curves were generated
over a variety of test conditions: e.g., differences in date, time, and places
of measurements. The effects of differences in depth, wind, and turbidity on
productivity would then be mediated by incoming visible light to the chamber.
It is apparent from the reported response curves that gross primary pro-
ductivity in Fahka Union Bay responded much more favorably to environmental
conditions during the fall then in the spring. In Fahkahatchee Bay, however,
gross productivity remained nearly the same in both seasons. Some environ-
mental factors effecting these difference responses are considered in the
forthcoming discussion.
Discussion
Our objectives in these studies were to establish estimates of the annual
net and gross primary productivity for the benthic plant communities in Fahka
Union and Fahkahatchee Bays and identify some specific environmental factors
effecting plant community metabolism.
Light is the principal regulator of autotrophic metabolism. Hence, monthly
estimates of gross productivity could be based on the average daily quantity of
photosynthetic light delivered to the plant community over the month, providing
other environmental factors are not limiting. The meaningfullness of this
proposition, of course, depends on the existence of a quantitative relationship
XII-16
-------
between light and gross photo-
synthesis. Nutrients can also
be a major factor effecting
photosynthesis. Not to be
excluded from consideration are
the plant species themselves
and how they fit in the meta-
bolic picture. These, although
major factors, are but a few of
the many possible considerations
that can confound judgments
regarding plant primary pro-
ductivity.
Experimental results of our
study have clearly demonstrated
a quantitative response between
benthic plant metabolism and
light. Also shown was the
degree to which turbidity
governs the quantity of visible
light at the bottoms in Fahka
Union and Fahkahatchee Bays.
Naturallys turbidity would be
a secondary factor with respect
to the total incoming solar
radiation.
0.10
O.08
0.06
0.04
0.02
0.00
0.22
MUD-SAND BOTTOM
SPRING, 1972
V0.0022X +0.0019
{r =0.98)
0.00
0.0
2.0 4.0 6.0 8.0 10.0 12.0 14.0
VIS. LIGHT AT BOTTOM (ly-h'1)
Figure XII-13. Regression of gross primary
productivity on light with respect to
bottom type and season for Fahka Union
Bay, 1972.
Turbidity, as in our case
measured by coefficient of light
extinction, was a consequence
of wind and depth interactions
which can ultimately lead to
resuspension of sediment. The
previously reported equation K = 1.40 X 10~4 Z + 3.49 X IQ~Z provides a model
from which estimates of light extinction coefficients can be calculated under
given conditions of depth. By substitution of this expression into the light
extinction model established in Beer's Law, the quantity of photosynthetic
energy (langleys) available to the benthic plants can be determined for given
depth and total solar radiations.
where:
I = Io-e~KZ
I = quantity of photosynthetic light received at the
bottom (langleys)
lo = quantity of photosynthetic light received at the
water surface (langleys)
K = light extinction coefficient (K = -1.40 X 10~4
Z + 3.49 X 10~2)
Z = depth of water (cm) over bottom
XII-17
-------
A complete expression for
benthic light illuminations
would be I = I0-e-d.40 X 10-4
Z + 3.49 X 1C-2) Z.
From the preceding expres-
sion, the mean daily visible
light (langleys/day) delivered
to the benthic plant communities
was calculated (Table XII-4).
These data were based on monthly
averages for daytime depths
(0600 to 1800 hours) and total
solar radiation. Inspection
of Figure XII-15 reveals that
the average benthic light
illumination for both bays
followed a general pattern of
decreasing light from January
to December 1972. Benthic
plant gross production, how-
ever, generally proceeded at
greater rates in November
than in March 1972 (Figures
XII-13 and 14).
A month by month estimate
of productivity for the year
was possible on the bases of
linear interpolation over time
of the slopes and intercepts
derived from the spring and
fall surveys. This operation
yielded separate regression
equations for each month.
Each equation coupled with
the respective mean light
value given in Table XII-4
provide an estimate of mean daily
study year (Table XII-5).
0.12
0.10
0.08
0.06
0.04
0.02
0.00
SAND-MUD BOTTOM
MUD-SAND BOTTOM
SPRING, 1972
= O.OI84X-0.0058
(r-0.94)
Y=0.0089X + 0.0024
(r=0.95)
N
I
£
o
E
Q.
Q.
O
0.12
0.10
0.08
O.06
0.04
0.02
0.00
FALL, 1972
Y=0.0265X+O.OII8
(r=0.95)
Y=0.0054X-0.0088
(r=0.88)
0.0 2.0 4.0 6.0 8.0 10.0
VIS. LIGHT AT BOTTOM (ly-h'1)
Figure XII-14. Regression of gross primary
productivity on light with respect to
bottom type and season for Fahkahatchee
Bay, 1972.
gross productivity for each month of the
A seasonal pattern emerges for gross productivity and the accumulation of
biomass of the benthic algae in Fahka Union Bay (Figure-16). In the case of
gross productivity, the steady increase from March to November appeared to be
independent of the diminishing quantities of photosynthetic light indicated
in Figure XII-15. Apparently, environmental factors other than light were
stimulating an increase in gross productivity.
During the same time frame, an outburts in the accumulation of standing
crop biomass occurred. The exact point in time at vhich this activity took
place is difficult to ascertain, except to say that it occurred between the
time of the July and September biomass surveys.
XII-18
-------
^ahka Union Bai
Fahkahatchee Be
cFahkahatchee Ei
X X
1 1
to B C
ft C O.
3 a- i
a. i m
i n m
cr IB 3
rt a.
rt o*
§ 0
O rt
rt o
December 299
i-i
ON
vl
1-1
to
VI
Ul
I—1
ro
00
o->
ro
i-i
00
ro
h-
ON
00
o
o
ro
Ul
Ul
o
0
ro
1-4
LO
0
0
to
I--
September 436
October 431
November 333
ON Ul 10
vl 00 00
LO ro oo
I-1 LO .p-
0 l-i l-i
LO ro oo
ON 00 tO
(-• ro LO
00 00 .p-
to O l-i
ro LO ro
ON O 00
ON Ul to
i-i ro ro
to *- .p-
ui ro f
i-i ro ro
00 LO .p-
•P- 00 LO
0 O O
0 O O
ro ro ro
P- LO ro
vl .p- Ui
o o o
o o o
ro i— 1-1
O to 00
•P- I-1 LO
O O O
o o o
00 VI ON
ro o 1-1
££
c£
en
rt
•P- Ul
Ul LO
00 O
to .p-
h-i
£ 8
Ul ON
ro LO
Ul h-i
O l-»
-P- -P-
O ON
hO LO
ro LO
00 I-i
ro ro
Ul to
LO .p-
ro LO
ON h- 1
t- ro
0 0
0 O
ro ro
1- O
vl 00
O 0
o o
1-1 1-1
.p- Ul
o o
o o
Ui -P-
LO .p-
<-, S >
CUT)
3 ^ fl
ro H-
h-*
*• Ul JS
to ro to
tO ON O
l-l
O to 00
LO vl Ul
ON Ul LO
LO ro t-"
•P- 00 Ul
f O 00
•P- ^- LO
tO LO h-i
•P- LO l-i
ro LO LO
to ro .p-
ro ro o
ro ro ro
vl to vl
00 H-i VI
ro LO ro
to O vi
to Ul Ul
0 O O
o o o
ro ro ro
O )-• LO
*- O O
O 0 O
O O O
ON vl 00
!-• O vl
O O O
O O O
J> -P- ON
O CO Ul
February 468
March 502
to to
vl ON
LO -P-
ro o
O to
LO .p-
ON Ul
!-• O
00 tO
tO VI
*- -P-
O LO
f I-1
ro ro
to to
Ul LO
to ro
vl ON
00 ON
0 0
0 0
ro ro
t- Ul
vl tO
o o
0 O
ro ro
?5
0 O
0 0
oo to
ro ui
CH
3
g
H
Ul
O
O
ON
t-i
O
to
^
O
ON
VI
*-
LO
vl
ro
00
to
ro
Ul
to
o
o
to
O
O
ro
ro
i-i
O
o
§
o
3
rt
31
C
i
B
OS
1
55
V
CO
i
K
o
C
i
en
m
w
i
o"
td
I
S
o
•n
i
o-
i
H 0
a o a
to rt H-
i* pi t— i
-i.i-.vj
U
•* o ro
0 i- H
ro
0 §
* a
f ro
H--O
^ 3-
rr
pel n
o 3
3
m
M
3
CD <
3 H-
!-•' en
^-N 01 I-1'
I-1 01 CT1
-^ H- l-i
-*v o ro
(i t-<
*3 P H-
•-^ rt oq
3*
W rr
O
rt
*
O
O
ro f
Hi H-
l-h 00
H- 3"
0 n rt
o ro M
rt 3 X
3* rt rt
!-••
n o
0
rt
CO I—1
SI H-
p^OQ
^ IT-
CO
y rt
^r" M
p3 fD
p" p
smission
ttchee Ba
"<; \»
co rt
a4
M O
tO rt
CO
Co
P
CU
i-1
H-
CW
rt
extinc
rt
H-
§
O
oef f ici
ro
P
IT-
CO
d
p
H-
o
p
H
cr
ro
X
M
H
1
4^
•
Calcula
rt
ro
a.
i
rt
3"
t— '
q
ft)
P
CO
O
l-h
fD
rt
ct.
uring
dayligh
rt
p4
O
<
H-
CO
cr
r— '
ro
H-
rt OQ
O C
co ro
ro IX
I-! H
CO M
OQ 1
fD M
("I* i^x
H- 0
fD rt
&» M
rt Co
P" <
co re
H- B3
fJ C»
fD
83 CT"
P* O
?? rt
CO rt
O
P
H- H-
O 1— '
P M
g 1
Cu P
Co
*^ rt°
m ui.
§• o
pf p
§•
03 0
n* ex*
Br,
fD C
fD H
tr
W H-
CO rt
vo H«
vj rt
fD
CO
fD
O
rt
EXTINCTION DEPTH (cm)
COEFFICIENT FOR DAYLIGHT HOURS
sis SoSoSoisil
LIGHT AT BOTTOM
(ly-DAY-')
(O
^J
ro
-------
Table XII-5. Average daily gross primary productivity of benthic plant
communities established on the mud and sand bottoms of Fahka Union and
Fahkahatcb.ee Bays, 19720
Avg. daily GPP by mo. (gm C/m2/day)
Fahka Union Bay Fahkahatchee Bay Fahkahatchee Bay Fahkahatchee Bay
Month MSa (143.1 ha.) MSa (386.8 ha.) SMb (299.6 ha.) (743.2 ha.)
January
February
March
April
May
June
July
August
September
October
November
December
0.349
0.206
0.069
0.110
0.153
0.183
0.253
0.266
0.311
0.376
0.366
0.281
0.650
0.576
0.537
0.534
0.592
0.596
0.662
0.600
0.598
0.627
0.529
0.453
0.183
0.212
0.250
0.235
0.244
0.225
0.217
0.170
0.147
0.131
0.090
0.010
0.413
0.385
0.381
0.372
0.406
0.401
0.432
0.380
0.370
0.379
0.311
0.277
Daily average
for year 0.243 0.580 0.176 0.376
Mud-sand bottom
Sand-mud bottom
Algal populations can be characterized by a rapid turnover rate opposed
to vascular plants. Under optimum environmental conditions of light, nutrient,
temperature, and possibly salinity, an algal community can enter into an
exponential growth phase that may occur over a few days; a phenomenon frequently
associated with algal blooms.
Thus, the question is whether or not the fall pulse of benthic algal
productivity was related to seasonal freshwater discharges from Fahka Union
Canal or Fahka Union Bay.
I
The view that the fall pulse in productivicy was stimulated by nutrients
entering the bay via the canal has certain appeal (Figure XII-16); although,
close examination of the given data indicates that the community was slow,
or failed to respond to the initial slug of nutrients arriving at the bay in
June when hurricane Agnes passed through the region. Following the "big
flush," nutrient loading to the bay proceeded steadily to a peak (at least
nitrogen) in September when the greatest standing crop biomass of algae was
recorded. Likewise, gross productivity followed a trend fairly similar to
nutrient loading.
On the other hand, the fall pulse in productivity may possibly be an
inherent growth property of the community and unrelated to seasonal discharges
of the canal. Returning to Table XII-1, it is apparent that iff both bays a
XI1-20
-------
fall increase in benthic algal
standing crop occurred. For
example, in Fahkahatchee Bay,
the green and red algae combined
showed a 100 percent increase in
biomass from July to September.
During the same period, the
algal community of Fahka Union
Bay increased many times.
Nutrient loading of Fahkahatchee
Bay was not as clearly evident as the
case in Fahka Union Bay. The former
bay receives the major share of its
freshwater input from upland sources
such as Fahkahatchee Strand (Chapter
VII, Figures VII-8 and 9). During
the study year, measurable surface
water flow across U.S. 41 (Tamiami
Trail) failed to develop. Hence,
the bay was dependent for its major
freshwater supply on rainfall
accumulation and runoff in the
marsh lands south of U.S. 41.
O--OBAY LOADING OF PHOSPHORUS
• NITROGEN
a o
-p u>
O
20 5
EZZZ3 STANDING CROP BIOMASS
CALCULATED GPP
O OOBSERVED GPP
1972
Figure XII-16. Benthic plant
productivity interrelationship
and daily nitrogen and phosphorus
loading per month, Fahka Union
Bay, 1972.
The potential of the uplands as
a source of nutrients to the bay by
way of runoff is evident in Figure
XII-17 where nutrient transport across
Janes Drive is illustrated. Under
normal hydrological conditions, a
similar transport picture would be
expected at U.S. 41 and in the freshwater reaches of the Fahkahatchee River and
also the East River. Both rivers lead to the bay. Nutrient enrichment of the
bay was somewhat clarified in the water chemistry section (Chapter X). Total
Kjeldahl nitrogen concentrations reported for Fahkahatchee River were on the
average twice that observed in the bay (Figure X-10).
For Fahkahatchee Bay, measured gross productivity rates in March were
nearly the same as reported in November. The yearly pattern, however, shows
a slight but steady decrease in gross productivity (Figure XII-17). The
opposite trend was true for the accumulation of plant biomass. The marine
grasses increased their standing crop biomass over the year at a steady rate
and peaked in the fall of the year. It is quite possible that after the
community approached its peak biomass, the plants were only photosynthesizing
at a maintenance rate; hence a slight decline in production rates would not
be totally unexpected.
Measurements of net productivity when assessed by biomass accumulation
techniques represents, at best, a minimal estimate of the parameter. This is
true because the estimates fail to account for plant carbon losses to respira-
tion, grazing, and leaf or cell mass turnover. The latter consideration would
be extremely salient when dealing with an algal community. With these
XII-21
-------
I
Pi
z
0.0
TRANSPORT ACROSS JANES DRIVE
» • NITROGEN
3 O PHOSPHORUS
t
,0.0
"
_
1
2.0 i
10 -|
3 £
3,1-
gj
O
consequences in mind, estimates
of benthic net productivity in
Fahka Union and Fahkahatchee
Bays were based on the low and
high points of standing crop
biomass.
With Fahka Union Bay, the
September standing crop minus
the March estimates yielded an
average biomass increase of
61.2 gm dry wt/m of bay bottom.
Although this gain occurred over
a six month period, nevertheless,
it represents the only measured
net increase for the sampling
year. This figure divided by
365 gives, an estimate of 0.168
gm dry wt/m /day as net produc-
tion for the bay. An estimate
of annual net productivity for
Fahkahatchee Bay extended from
January to November 1972, and
the average gain in standing
crop biomass was 0.197 gm dry
wt/m^/day. Summaries of aver-
age daily rates of primary
productivity are given in Table XII-6 and are based on values reported in
Table XII-5 and the above estimates of net production rates. A comparison
of the two estimates of net productivity figures would tend to indicate that
the two bays supported similar plant communities in terms of net productivity
capabilities. However, Fahka Union Bay supported mainly benthic algae whereas
Fahkahatchee Bay maintained marine grasses.
Table XIl-6. Average daily rates of primary production for the benthic plant
communities associated with Fahka Union and Fahkahatchee Bays during 1972.
0.8
07
X
1- — 0.6
Z >j
iSo.4
• E 0.3
| "-
«. W
3 ~~ 02
4 01
UJ
S 00
n
IV //Vl STANDING CROP BIOMASS
• — •CALCULATED GPP
O ©OBSERVED GPP
•^-^-—^-•^•Cj
* ^ • * ^
171
/
/
R /
.171. . M, y
^
/
/
/
/
/
/
/
,_^^
"*S
7
/
y
/
^
/
1
/
f
/
^
<
k.
8O
7O
60
40
30
20
10
JFMAMJJASONO
1972
Figure XII-17. Nutrient transport across
Janes Scenic Drive and benthic plant
productivity, Fahkahatchee Bay, 1972.
Fahka Union Bay Fahkahatchee Bay
gm C/m2/day gm dry wt/m^/day gm C/m^/day gm dry wt/m2/day
Gross productivity
Net productivity
0.243
0.168
0.376
0.197
a Fahka Union Bay total bottom area of 189.3 ha.
Fahkahatchee Bay total bottom area of 743.2 ha.
Reasons for the difference in plant community types between the two bays
are not obvious. First, turbidity of the waters as a factor has little appeal.
XII-22
-------
As indicated earlier, we found little differences in the turbidity regimen of
the two bays. Possibly, water depth could have been a contributing factor to
the disparity in community types.
Fahka Union, on the average, was nearly 30 percent shallower than
Fahkahatchee Bay. A review of Figure XII-15 will show that average depths
during daylight hours were always greater than 60 cm in Fahka Union Bay.
This depth is not uncommon to marine grass beds in the Ten Thousand Island
region (B. Yokel, personal communication).
An examination of the information previously reported on sediment char-
acteristics indicates that the bottom of Fahka Union Bay served to a greater
degree as a sink for heavy metals. For example, nickel, zinc, lead, cadmium,
copper, chromium, and arsenic were nearly two to three times greater in con-
centration in sediments in Fahka Union than in Fahkahatchee Bays (Chapter XI).
The reverse was true for metal concentrations in the water columns of the two
bays. Concnetrations of nickel, zinc, lead, cadmium, and copper were con-
sistently greater in waters of Fahkahatchee Bay than Fahka Union Bay. From a
toxicological viewpoint, possibly the concentrations of metals in the sediments
of Fahka Union Bay inhibited development and growth of marine grasses. Marine
grasses are vascular plants and maintain a root system which derives much of
their nutrient requirements from the sediment whereas algae obtain their
nutrients principally from the water. .
Finally, we might hypothesize that the additional nutrient loading of
Fahka Union Bay from the canal stimulates algal growth to a level of standing
crop biomass that leads to a smothering or shading out of the vascular plants.
This possibility has merit when considering the scheduling of nutrient input
to both bays.
During most of the winter months (dry season), Fahka Union Bay received
nutrients via freshwater discharges from the Fahka Union Canal while Fahkahatchee
Bay received almost no inputs of nutrients from the uplands. In both bays a
winter die back of the grasses occurred, thus, leaving the major portion of
the winter standing crop biomass as algae (review Table XII-1). With the
approach of the spring growing season, the algae in Fahka Union had the
competitive edge for colonization of the bay because of additional nutrients
provided by the canal and reduced competition for these nutrients in the
absence of marine grasses.
SALT MARSH AND MANGROVE METABOLISM
Introduction
The functional properties of mangrove ecosystems in south Florida has
recently become a topic of extensive analytical investigation. Studies by
Heald (1971), and Odum (1971), have linked in quantitative terms the regional
role of mangrove forests in the detrital-based economy of an estuary. Man-
grove detritus is ultimately derived from the communities' ability to produce
organic matter. Limitations of this work process are governed by numerous
environmental controls such as light and nutrients.
The nutrient budget supporting mangrove metabolism in the region may be
ultimately derived from nutrient stores generated in the freshwater uplands.
XII-23
-------
This same view is indicated in recently reported work on mangrove metabolism
by Lugo, et_ -al. (in press). Overland flow of fresh water serves as the
principal transporter of nutrients from the interior regions to the tidal
zone. As fresh water seeks its way to the Gulf, its nutrient load is
relieved through biotic uptake and dilution.
The object of this study was to determine the response of mangrove pro-
ductivity to a decreasing freshwater gradient which implies that across this
gradient, nutrient availability for plant metabolism decreased with increas-
ing chloride concentrations. An additional purpose of this study was to
assess saltgrass marsh community productivity.in response to increasing
levels of chlorides.
Study Area and Site Locations
Two areas were selected for study on the basis of accessibility with
equipment and proximity to a freshwater gradient. The Fahka Union River
drainage system served as the principal area of investigation (Figure
XII-18). The upper reaches of the river receive surface runoff from two
freshwater sources, rainfall in the immediate salt marsh regions, and
surface water runoff from the upland terrestrial systems north of U.S. 41.
In both cases, the river serves as a tidal waterway for collection and
delivery of nutrient-bearing fresh water to the estuary. The second area
of study was Fahkahatchee Bay, which represents the final stage, or receiv-
ing point (prior to the Gulf), for natural freshwater runoff. The Fahkahat-
chee River and the East River discharge to this bay and share the same
drainage features as Fahka Union River. The four study sites were positioned
along a north to south freshwater gradient (Figure XII-18). For reason of
limited accessibility, the Fahkahatchee River drainage was excluded from this
study. The northern most site (Station D) represented the area of greatest
freshwater concentration, while the southern most location (Station G)
represented the least concentration. A general description of each study
site location follows:
Station D. The general habitat was composed of saltgrass marshes and
scattered stands of mature mangroves. Both communities bordered a shallow
brackish lake, whose water level varied slightly with tidal amplitudes.
Tidal water level changes ranged from about 8 to 16 cm.
The mangrove community followed a general landward zonation pattern along
one side of the lake. Red mangroves predominated at the water's edge. Their
proproots and often the base of their trunks were subject to semidiurnal tidal
flushing. The interior zone was higher ground, and was mainly occupied by
black mangroves and scattered individuals of buttonwood. Regular flushing of
this zone was not apparent.
The saltmarsh community occupied the shore and landward areas opposite
to the mangrove community. The marsh at this point in the Fahka Union River
basin was subject to daily flooding - mainly a consequence of tidal amplitude.
The area supported a variety of semiaquatic plants that are identified in
Chapter XIII.
Station E. At this point in the basin, the river channel was well defined
and ranged from 5 to 15 meters in width. Colonizing the river's edge were
XI1-24
-------
Figure XII-18. Station locations for the study of mangrove and salt marsh
community metabolism, 1972.
XII-25
-------
immature red and white mangroves, backed by a salt marsh community comprised
of two plant species - Spartina apartina and Pistichlis spicata (saltgrass).
Both communities received semidiurnal ti.dal flushing.
Station F. A forest of mature mangroves flanked the river in this reach
of the basin. The river channel ranged from 15 to 20 meters in width. Species
zonation was pronounced. Red mangroves dominated the river frontage vegetation,
but coexisted with the black and white mangroves in the interior zones. Semi-
diurnal tidal flushing did not appear to be a characteristic feature of the
community, except in the case of red mangroves occupying the edge of the river
channel.
Station G. The study site was situated at the mouth of a small tidal
stream located on the northeastern shore of Fahkahatchee Bay. The mainland
mangroves represented a mature forest. This particular area supported some
of the largest sized mangroves found in the Fahkahatchee region. Species
zonation was similar to Station F. However, semidiurnal tidal flushing of
the forest floor was a typical feature of the community.
Methods
Plant taxonomic features of the communities are dealt with in Chapter
XIII. Net and gross primary productivity of manland mangrove forests and
salt marshes were determined from measurements of plant uptake and release
of carbon dioxide. At this point, only an abbreviated version of the method-
ology employed is presented. Details of the subject are given in Appendix R.
All metabolic measurements were limited to above—ground vegetation and included
only principal plant species of the community. '
Clear plastic chambers were used to isolate a portion of the plant. Each
chamber received a continuous supply of ambient air at a known rate of flow
and carbon dioxide concentration. The chambers were fitted with an exhaust
port at the distal end of enclosure where exiting air was sampled for measure-
ment of carbon dioxide concentrations. In the case of groundcover, such as
salt marsh plants, a rigid cylindrical chamber (1 meter in diameter by 1 meter
in height) covered with a clear, thin (0.05 cm) polyacetate material was set
over a segment of the plant community and sealed to the substratum by the
cutting edge of the chamber. When working with woody plants, a clear plastic
sheath was used to .enclose leafy branches, trunk segments, and proproots. The
fresh air flow to the chamber ballooned the sheath into a semi-rigid form.
These sheaths, when inflated, measured about 1.2 meters in length by 0.6 meters
in diameter.
Both the polyacetate and plastic material were assessed for filtering
effect on the light spectrum. In both cases, the material acted as a neutral
filter, but suppressed light transmission by approximately three percent.
The fresh air supply to each chamber was regulated to maintain an average
turnover rate of about five per minute. This rate proved optimum for main-
taining across the chamber a €62 concentration (ppm) gradient within the limits
of detection and precision of the infrared gas analyzer employed. The gradient
was usually fess than five ppm. This turnover rate also proved effective in
controlling air temperatures within the chambers. In all cases, the tempera-
ture differential between ambient air and air exhausting from the chamber
remained less than 4°C, and usually ranged from 2° to 3°C.
XII-26
-------
Plant photosynthesis was determined by measuring the decrease in th.e
chamber CO? gradient, and respiration rate was determined from measured
increases in the gradient. These changes were recorded at hourly intervals
over a 24-hour period. In most cases, four chambers were deployed. The CC^
analyzer received daily calibration by introducing a known concentration of
carbon dioxide to the detection system.
Assessment of plant transpiration was planned for this study but at the
time electronic problems with the measuring devices prevented determinations.
Productivity and respiration values are reported as totals derived from
integration of hourly rates over a complete diurnal period. Salt marsh pro-
ductivity is reported as per unit land area of the community, while mangrove
productivity is given per unit area of metabolic surface, e.g. leaf, trunk,
and proproot area. To make productivity comparisons between species or
communities would be inappropriate unless based on per unit land area. To
render the initial data comparable, the mangrove metabolic values must be
multiplied by the appropriate land area index (e.g. total metabolic surface
area per" unit land area). Indices for making these conversions were obtained
from other work. Their implementation into our study is discussed later in
the report.
These metabolic studies were conducted over the period from November 17
to December 21, 1972. To establish the freshwater regime at each of the
metabolic sites, chloride concentrations were determined for surface water
and soil solutions. When possible, an automatic, sequential sampler was used
to collect hourly samples over a 24-hour period. In the case of soil waters,
a perforated casing was inserted into the soil to a depth of approximately
0.75 meters. Seepage water was sampled in the same manner as above. In the
absence of the automatic sampler, quasi-synoptic samples were obtained. During
the course of the metabolic studies total solar radiation was recorded with a
pyrheliometer.
All studies were conducted from a 12 by 38 foot pontoon barge equipped
with an outborad motor/ At each station, the barge was motored into position
and securely anchored next to the plant community to be studied. Tents erected
on the barge served as a base of operation and as shelters for the metabolic
equipment.
Results
Station D, upper Fahka Union River, represented the least saline region
of the drainage system in our study of mangroves. Soil water chlorides for
the study site ranged in concentrations from 4.6 ppt to 5.0 ppt and averaged
4.9 ppt over the 24-hour sampling period. Simultaneously, surface water
chlorides ranged from 4.4 ppt to 4.8 ppt and averaged 4.7 ppt.
Twenty diurnal rates of photosynthesis and respiration were measured over
the period of November 26 to December 8, 1972. The saltmarsh community hosted
several plant species but restricted accessibility with the equipment limited
our study efforts to the following species: Spartina spartina, Eleocharis
cellulosa, and Distichlis spicata (saltgrass). Results of the metabolic
measurements are reported in Table XII-7. To provide a common basis for
comparibility, the reported net primary productivity and respiration rates
XII-27
-------
Table XII-7. Results of plant metabolic studies to determine primary production
of the salt marsh plant species associated with the upper Fahka Union River
basin, Station D, 1972.
Chamber Biomass Composition
gm/m2 (Ash Free Wt.l
Date
(1972)
Nov 27
Nov 28
Nov 29
Nov 27
Nov 28
Nov 27
Nov 28
Nov 27
Nov 28
Nov 29
Green Above Ground
Vegetation
/
Spartina
Spartina
Spartina
Spartina
Spartina
Eleocharis
Eleocharis
Distichilisb
Distichilis!>
Distichilisb
Total
1,451.6
1,451.6
1,451.6
1,762.1
1,762.1
688.4
688.4
3,094.6
3,094.6
3,094.6
Lvs.
531.8
531.8
531.8
813.9
813.9
92.8
92.8
551.2
551.2
551.2
Dead
324.8
324.8
324.8
541.5
541.5
52.4
52.4
856.8
856.8
856.8
Roots
595.0
595.0
595.0
406.7
406.7
543.2
543.2
1,686.6
1,686.6
1,686.6
Total
Solar
Radiation
ly/day
339
313
360
400
412
339
313
351
325
373
NPP
Lp
gm C
m'.Lp
0.860
0.777
1.044
1.196
1.497
0.674
0.625
1.646
1.700
1.794
V
gm C
m2-Dp
0.263
0.649
0.807
0.572
0.714
0.387
0.572
0.735
0.875
0.670
NPP,
d
gm C
m2-d
0.597
0.128
0.236
0.624
0.784
0.287
0.053
0.911
0.824
1.124
Rd
gm C
m2.d
0.477
1.176
1.462
1.035
1.293
0.702
1.037
1.331
1.585
1.214
GPP.
d
gm C
m2-d
1.074
1.304
1.698
1.659
2.076
0.988
1.090
2.242
2.410
2.338
aNPP - net primary production
R - respiration
GPP - gross primary production
Lp - photoperiod
Dg - dark period (13.2 hours)
d - 24 hour day
Saltgrass
requires normalization for differences in solar radiation received and chambered
biomass. Average gross production or respiration rates for the chambered plant
component can be derived from the following relationships:
I
NPP
LP • ly
«' Bo
n
N
n
I
"24
-PP
• R
Dp
• Bo
ZBo"
• n
N .
GPP = NPP24 +
where
NPP = total observed net productivity for photoperiod (gm C/m2/Lp)
Lp
Bo = observed chamber biomass (on carbon fixation rates, Bo refers
to green leaves and for respiration rates, Bo refers to total
biomass)
XII-28
-------
Iy0 = observed total solar radiation for day
N = number of replications (N>1)
ly = daily average (293 langleys) of total solar radiation for November
17 through December 21, 1972
R,, = total observed nighttime respiration
c
Dp = dark period in hours
24 hrs = complete day
2
= systems mean net productivity for day (gm C/m /day)
(Figure XII-19; NPP24 =1-2)
= systems mean respiration for day (gm C/m^/day)
(Figure XII-19; R24 = 2 + 3)
R
24
GPP = systems mean gross productivity for day (gm C/m2/day)
Figure XII-19 represents diagrammatically the general metabolic scheme for plant
photosynthesis and respiration over a complete diurnal period.
Based on the above relationships, an average daily production and respira-
tion were calculated for the three saltmarsh plant species studies and reported
in Table XII-8.
Results of the mangrove
metabolic measurements at
Station'D are presented in
Table XII-9. As noted,
measurements were limited
to red and black mangrove
and buttonwood (Rhizophora
mangle, Avicennia nit_ida,
and Conocarpus ere eta,
respectively). White
mangroves (Laguncularia
racemosa) were absent from
this community. The
normalization procedures
outlined above were also
applicable to the mangrove
data except that the area
of metabolic surface, e.g.
leaf surface area, replaces
the term for observed biomass.
Average productivity and
respiration rates for selected
components of the community
are presented in Table XII-10.
PHOTOPERIOD-
0200
0600 1000 I4OO
TIME (HOURS)
I8OO
2200
I NET PHOTOPERIOO PRODUCTION
2 NIGHTTIME RESPIRATION
3 PHOTOPERIOO RESPIRATION
Figure XII-19. Schematic of diurnal response
of plant metabolism.
XII-29
-------
Table XII-8. Mean rates of productivity and respiration for three species of
plants characteristic of the saltgrass marshes associated with the upper
reaches of the Fahka Union River, Station D, 1972.
Chamber Biomassa, Ash Free Wt.
Cgm/m2)
Green
Plant
Distichlis sp.
Spartina sp.
Eleocharis sp.
Total
3,094.
1,606.
688.
6
8
4
Lvs
551.
672.
92.
2
8
8
Above-
Ground
Dead
856.8
433.2
52.4
Metabolism
Roots
1,686.
500.
543.
6
8
2
(gin
GPP0.
24
2.06
1.41
0.976
C/m2/day)b
NPP,,,.
24
0.679
0.293
0.104
R
1.
1.
0.
24
38
12
872
a Average of two replications for Spartina sp.
b rates normalized for biomass and light.
Table XII-9. Results of plant metabolic studies conducted to determine primary
production of the mangroves associated with the upper Fahka Union River,
Station D, 1972.
Plant Metabolism
Total NPP
Date
(1972)
Dec
ii
"
"
"
"
ii
"
"
"
a
. 5
5
5
5
5
5
7
8
8
8
NPP
R
GPP
Lp
Dp
d
Vegetation
Rhizophora
shade Ivs .
sun Iv's .
proproot
trunk
Conocarpus
sun Ivs .
shade Ivs.
shade Ivs ,
Avicennia
sun Ivs .
shade Ivs .
trunk
= net primary
= respiration
n i • • 2m C
Radiation &-=
ly/Lp m -L,p
373 3
304 4
NA
NA
353 4
400 3
390 3
360 1
360 0
NA
production
= gross primary production
= photoperiod
= Dark period
= 24 hour day
(13.4 hours)
.879
.530
NA
NA
.035
.352
.774
.179
.859
NA
b
c
d
V
__
m -Dp
0.486
0.541
NA
NA
0.691
0.717
0.589
0.317
0.238
NA
NPP,
d
gm C
2
m «d
3.393
3.990
NA
NA
3.344
2.634
3.184
2.136
0.621
NA
Rd
gm C
7
m
0
0
0
1
1
1
1
0
0
1
•d
.873
.972
.592
.049
.243
.290
.059
.570
.429
.919
GPP,
d
gm C
2
m *d
4.267
4.962
NA
NA
4.587
3.924
4.244
2.707
1.049
NA
red mangrove
buttonwood
black
mangrove
XII-30
-------
Table XII-10. Diurnal metabolic rates for various components of a mangrove
forest community associated with the upper reaches of the Fahka Union River,
Station D, 1972.
Plant Type
Plant Component
a 2
Metabolism gm C/m /day
GPP
24
NPP
24
R,
24
Red mangrove
Black mangrove
Buttonwood
Shade Ivs.
Sun Ivs.
Proproot
Trunk
Shade Ivs .
Sun Ivs.
Trunk
Shade Ivs .
Sun Ivs.
3.43
4.80
NA
NA
0.890
1.21
NA
3.17
3.90
2.56
3.82
NA
NA
0.461
0.643
NA
2.00
2.66
0.873
0.972
0.592
1.05
0.429
0.570
1.92
1.17
1.24
a Productivity based on average daily solar radiation of 293 langleys.
As indicated previously, species zonation was apparent at Station D. Red
mangroves fronted the river's edge, and black mangroves dominated the interior
zone with a scattering of mature buttonwood. The interior region was obviously
subjected to only occasional tidal flushing. Extensive litter accumulation was
in progress on the forest floor. The absence of black mangrove pneumatophores
indicated the zone was well drained. The presence of black mangrove pneumato-
phores is usually associated with poorly drained soils. Pneumataphores have
been recognized as special adaptational structures for a gas exchange mechanism
in which oxygen is transported to the root assemblages (Scholander, et^ al.,
1955). Species zonation would seem to be related to metabolism. The red man-
groves maintained a gross productivity nearly four times that of the black man-
groves under equal conditions of incoming solar radiation. Likewsie, the gross
productivity of the buttonwood was less than that observed for the red mangroves.
Lugo, et al. (in press) reported a similar scheme of mangrove zonation on a
basis of species metabolism.
The next downstream site (Station E) selected for study was also subject
to slight changes in chlorides, but at higher concentrations than Station D.
Soil water chlorides varied from 10.9 ppt to 12.5 ppt and averaged 11.6 ppt
over a 24-hour period. To establish a chloride relationship between surface
and soil water, paired sampling was conducted two months later. At this time,
chloride concentration of the soil water was 13.7 ppt vs 12.7 ppt for the sur-
face waters. This site was subject to semidiurnal tidal flushing.
The assemblage of plant species representing the saltgrass marsh community
at this station was markedly divergent from that of the upstream marsh station.
The latter area was shared between several plant species, whereas at the down-
stream site only Spartina spartina and Distichlis spicata (commonly known as
XII-31
-------
saltgraas) were found. The increase in chloride gradient from the upstream
site to the present location may be the environmental factor dictating the
difference in species composition. A marked increase in chloride levels
(salinity) may pose a potential stress condition. Those species lacking the
physiological machinery to cope with this new situation yield their position
in the community to more suited populations. Hence the surviving populations
may respond with added vigor to the diminished competition for space and
nutrients.
Possibly this is the case for the two saltmarsh plant species reported
above. Observed rates of photosynthesis and respiration are reported for
the community in Table XII-11. These rates after normalization for differences
in chamber biomass and incident solar radiation appear in Table XII-12.
Possibly coupled to the measured increase in Spartina and saltgrass pro-
ductivity was an auxiliary nutrient source. A mussel population was found
inhabiting the root masses of Spartina. Their presence was not noted at
Station D. Odum (1971) cites the work of Kuenzler (1961, 1961a) and Pomeroy,
et al. (1967) concerning mussel populations as biotic mechanisms for recycling
phosphorus. The latter investigators demonstrated the role of mussles as
auxiliary nutrient sources in a Spartina marsh in Georgia.
Table XII-11. Results from plant metabolic studies conducted to determine
primary production of salt marsh plants and mangroves associated with the
mid-reach of Fahka Union River, Station E, 19720
Chamber Biomass Composition
gm/m2 (Ash Free Wt.)
Date
(1972)
Nov 17
Nov 18
Nov 17
Nov 18
Nov 17
Nov 18
Nov 17
Nov 18
Green Above Ground
Vegetation
Spartina
Spartina
Spartina
S*partina
Distichilis
Distichilis
Distichilis
Distichilis
Total
2,002.9
2,002.9
3,272.8
3,272.8
3,284.8
3,284.8
3,846.5
3,846.5
Lvs.
626.4
626.4
919.0
919.0
951.9
951.9
1,725.8
1,725.8
Dead
781.5
781.5
824.7
824.7
481.3
481.3
730.1
730.1
Total
Solar
Radiation
Roots
595.0
595.0
1,528.2
1,528.2
1,851.6
1.851.6
1,390.6
1,390.6
ly/Lp
398
320
398
297
392
360
392
360
NPPT
Lp
gm C
m . Lp
2.572
2.038
3.305
2.253
4.273
5.194
4.221
5.008
Plant
gm C
m2.Dp
1.395
1.554
1.700
1.736
2.015
2.066
2.799
2.366
Metabolism3
NPP,
d
gm C
m2.d
1.177
0.484
1.604
0.517
2.258
3.127
1.422
2.642
Rd
gm C
m2.d
2.560
2.852
3.120
3.185
3.697
3.791
5.135
4.342
GPPd
gm C
m2.d
3.737
3.336
4.725
3.703
5.712
6.919
6.557
6.984
Rhizophora0
Nov 21 sun Ivs. NA NA NA NA 306 2.670 0.438 2.233 0.798 3.031
Nov 22 sun Ivs. NA NA NA NA 75 1.095 0.454 0.641 0.829 1.469
Nov 22 La!unClvs!ia NA NA NA NA 75 1.539 0.400 1.139 0.706 1.845
net primary production
R - respiration
GPP - gross primary production
Lp - photoperiod
Dp - dark period (13.1 hours)
d - 24 hour day
bSaltgrass
CRed mangrove
^White mangrove
XII-32
-------
Table XII-12. Mean rates of productivity and respiration for plant species
characteristic of the saltgrass marsh associated with the mid-reaches of
the Fahka Union River, Station E, 1972.
Mean Chamber Biomass,
Ash Free Wt. (gm/m2)
Plant
Distichlis sp.
Spartina sp.
Total
2,637.8
3,565.6
Green
Lvs.
772.7
1,338.8
Above-
Ground
Dead
803.1
605.7
Metabolism
(gm C/m2/day)a
Roots
1,061.6
1,621.1
24
5.07
3.52
NPP24
0.847
0.474
R24
4.22
3.05
a Rate normalized for biomass and light
By comparison, both species showed a marked increase in gross productivity
over their counter-parts in the less saline marshes located upstream (Station D).
These increases in total carbon budget (GPP) may be of some consequence to
community structure.
For example, in the upstream marsh, approximately 89 percent of the carbon
economy of Eleocharis sp. was diverted to maintenance of community respiration;
a percentage derived from the ratio »2A. The percentages of the carbon budget
devoted to maintenance by Spartina sp. and Distichlis sp. were 79 and 67 percent,
respectively.
In contrast, where fresh water was less available for metabolism, the
plants perhaps rely to a greater degree on their physiological machinery to
cope with the added environmental stress. The added metabolic "strain" might
place additional drains on the plant's carbon budget. In the case of Eleocharis
sp., it appears that the added cost for survival on its carbon economy was over-
whelming, hence, the population dropped from the community at Station E. On the
other hand, Spartina sp. and Distichlis sp. maintained their position in the
community by increasing their gross carbon budget to meet the additional
constraints of respiration, which in this case were 87 and 83 percent of gross
production, respectively. This increase in gross productivity correlates with
a diminished competition for space and nutrients.
Assessment of mangrove metabolism was limited to an immature red and white
mangrove community, which fronted the edge of the river channel. Observed rates
of metabolism are reported in Table XII-11.
After normalizing the measured net productivity rates for equal solar
radiation, major differences in their gross metabolism of carbon can be seen in
Table XII-13.
The structure of the white mangrove population at Station E may account for
the high net productivity reported. Although most of the individual trees main-
tained a canopy height similar to the reds, their overall structure indicated an
immature population. Lateral branching began near ground level, and the upper
XII-33
-------
Table XII-13. Diurnal productivity rates of sun leaves for red and white
mangrove inhabiting the stream side of the mid-reach of the Fahka Union
River, Station E, 1972.
a 2
Metabolism gm C/m /day
Plant
Red mangrove
White mangrove
GPP24
3.79
6.32
NPP24
2.97
5.61
R24
0.817
0.706
a Productivity based on average daily solar radiation of 293 langleys,
story canopy was not a crown-like feature typical of more mature populations.
High net productivity and low expenditures of energy for respiration might be
consonant with work processes directed at a rapid accumulation of biomass.
The lower Fahka Union River site (Station F) represented a position in
the drainage basin of less freshwater availability. This is substantiated by
the following reported chloride concentrations.
Chloride concentrations of the river water at this site averaged 12.7 ppt
and ranged from 8.2 ppt to 17.0 ppt over a 24-hour period. The relative wide
range in chloride level reflects the presence of fresh water from the Fahka
Union Canal backing up into the lower reach of the river during flooding tide.
Paired analyses of soil solution and surface water chloride yielded a concentra-
tion of 18.4 and 16.2 ppt. These analyses were conducted approximately two weeks
after the start of the metabolic studies at this location. Species zonation was
apparent. As before, red mangroves lined the shore of the river. Black and white
mangroves coexisted in the interior zone with a few scattered individual reds.
Some black mangrove pneumatophores were present on the forest floor of the
interior zone, but their numbers were relatively few with respect to the profuse
numbers usually found in the poorly flushed black mangrove forests. Results of
the metabolic measurements are reported in Table XII-14. Productivity rates
adjusted for equal light conditions are given in Table XII-15. From Table XII-15,
species zonation on the basis of metabolism was indicated.
Station G represented the estuarine position in the freshwater gradient
line established for the study of mangrove metabolism. At this site, bay water
flooded the land area on a semidiurnal frequency. Immediate effects of this
tidal flushing were evident. The forest floor was virtually swept clean of
debris. At the time of study, the measured chloride content of the flooding
water averaged 16 ppt and ranged from 15.0 to 17.0 ppt over a 24-hour period.
Subsequent paired samples of soil solution and surface water were obtained and
analyzed for chloride content. Concentration of chlorides were 18.2 and 18.0
ppt, respectively.
The mangrove forest maintained a scheme of species zonation similar to the
other study sites. Red mangroves were predominant along the edge of the stream
while black and white mangroves were representative of the interior zone.
Measured metabolic rates for this community are given in Table XII-16. To render
XII-34
-------
Table XII-14. Results from plant metabolic studies conducted to determine
primary production of mangroves associated with the lower Fahka Union
River Basin, Station F, 1972.
o
Plant Metabolism
Date
(1972)
Dec
it
M
"
n
. 14
15
. 15
14
15
Vegetation
Rhizophora
sun Ivs .
trunk
proproot
Avicennia
sun Ivs .
trunk
Total
Solar
Radiation
IX/LP
213
NA
NA
213
NA
NPPT
Lp
m -Lp
1.958
NA
NA
1.523
NA
V
gm C
m -Dp
0.149
NA
NA
0.140
NA
NPP,
d
gm C
m -d
1.808
NA
NA
1.383
NA
R
gm
m
0.
0.
0.
0.
0.
d
_C
•d
265
406
747
250
323
GPP,
d
gm C
m *d
2.074
NA
NA
1.632
NA
Laguncularia
n
"
a.
14
15
NPP =
R =
GPP =
Lp =
Dp =
d =
sun Ivs .
trunk
net primary
respiration
240-
NA
production
gross primary production
photoperiod
dark period
24 hour day
(13.5 hours)
1.277
NA
b
c
d
0.126
NA
1.151
NA
0.
0.
225
126
1.376
NA
Red mangrove
Black
White
mangrove
mangrove
Table XII-15. Diurnal metabolic rates for various components of the mangrove
forest community associated with the lower reach of the Fahka Union River,
Station F, 1972.
Plant
Red mangrove
Black mangrove
White mangrove
a 2
Metabolism gm C/m /day
Components
Sun Ivs.
Trunk
Proproot
Sun Ivs.
Trunk
Sun Ivs.
Trunk
GPP24
2.81
NA
NA
2.20
NA
1.66
NA
NPP
24
2.54
NA
NA
1.96
NA
1.43
NA
R24
0.265
0.406
0.747
0.250
0.323
0.225
0.126
a Productivity based on average daily solar radiation of 293 langleys.
XII-35
-------
Table XII-16. Summary of results from plant metabolic studies conducted to
determine primary productivity of mangroves associated with a small tidal
stream of Fahkahatchee Bay, Station G, 1972.
Metabolism3
Date
1972
Dec.
Dec.
Dec.
Dec.
Total
Solar
NPPT
Lp
D..^.,*^ Sm ^ &
20
21
19
19
Vegetation
Rhizophora
Sun Ivs .
Sun Ivs .
Trunk
Proproot
ly/L m -L
P
248
173
NA
NA
P
0.719
0.596
NA
NA
m
0
0
0
0
%
m C
2-D
P
.263
.274
.315
.493
NPP,
a
gm C
m2-d
0.350
0.322
NA
NA
R
gm
2
m
0.
0.
0.
0.
d
C
•d
469
496
559
876
GPP,
a
gm
2
m
0.
0.
NA
NA
C
•d
818
809
Dec. 20
Dec. 21
Dec. 20
Avicennia
Sun Ivs.
Sun Ivs.
Trunk
248 2.35 0.782 1.57 1.39 2.96
183 1.11 0.568 0.538 1.01 1.55
NA NA 0.630 NA 1.12 NA
Laguncularia
Dec. 20 Sun Ivs.
Dec. 21 Sun Ivs.
Dec. 19 Trunk
a Productivity based on
b Red mangrove
c Black mangrove
d White mangrove
275 2.55 0.918 1.64
191 0.813 ND NA
NA NA 1.16 NA
1.63
NA
2.07
3.27
NA
NA
average daily light of 293 langleys.
the observed production rates of each specj.es more comparable, net production
was normalized for equal light. The adjusted data is given in Table XII-17.
Discussion
For the purpose of the study, the freshwater gradient as assessed by
chloride concentration was established along a north to sr>uth line. Starting
at Station E, the northernmost point on the gradient, surface water chlorides
averaged 4.7 ppt. Chloride levels for both soil and surface waters steadily
increased at successive downstream sites with a maximum observed concentration
of 18 ppt occurring at Fahkahatchee Bay (Station G).
Mangrove metabolism responded to the freshwater gradient in two directions,
First, a lateral zonation by species was evident on the basis of leaf gross
productivity and tidal flushing. Secondly, a near linear response in species
metabolism occurred with decreasing availability of fresh water.
XII-36
-------
Table XII-17. Diurnal metabolic rates for various components of the mainland
mangrove forest community associated with a small tidal stream in
Fahkahatchee Bay, Station G, 1972.
a 2
Metabolism gm C/m /day
Plant
Red mangrove
Black mangrove
White mangrove
Components
Sun Ivs.
Trunk
Proproot
Sun Ivs .
Trunk
Sun Ivs.
Trunk
GPP24
1.14
NA
NA
2.80
NA
3.43
NA
NPP24
0.661
NA
NA
1.60
NA
1.80
NA
R24
0.477
0.559
0.876
1.20
1.12
1.63
2.07
a Mean diel productivity based on average daily light of 293 langleys.
A review of Tables XII-lO and 15 will indicate that leaf metabolism was
greater for the red mangroves than for either the black or white species. In
both cases, the reds were exposed to tidal dynamics more frequently than the
other species, which occupied the interior zones of the community. At Station
G, Fahkahatchee Bay, the three species appeared to be equally subjected to
tidal flushing; however, the metabolic response by species was reversed (Table
XII-16). Red mangrove production was secondary to both the black and white
specimens. A similar species zonation scheme based on mangrove metabolism was
reported by Lugo, et al. (In Press).
By species, the response of metabolism to decreasing freshwater concen-
tration was evident in Table XII-17. The data reported have been converted to
unit area of metabolic surface per unit land area. This procedure takes into
account structural and density dissimilarities between and within species
populations. To accomplish the land area conversion requires multiplication of
the initial metabolic rates by the appropriate land area index (i.e., total
leaf surface area per unit land area).
Indices for leaf area were obtained from the reported work of Snedaker,
et al. (in press). These investigators studied the structural arrangement of
different mangrove stands in the region of the present study. Their report
included estimates of leaf area indices representative of codominant red-black
and red-white mangrove stands. The forest stand yielded leaf area indices of
1.9 to 2.1, and 2.1 to 2.4, respectively. These indices were assumed to be
appropriate for the mangrove stands at Stations D and E; however, Stations F and
G were represented by a roughly codominate forest of red, black and white
mangroves. For the purpose of this study, we took the average total of the
leaf area indices for the two mangrove stands (4.2) reported by Snedaker, et al.
and assumed it representative of the three-way mixed forest at Stations F and G.
The average divided by three provided a crude estimate of leaf area index (1.4)
for each species.
XII-37
-------
Indices for mangrove trunk and red mangrove proproot area were obtained
from reported investigations by Lugo, et al. (in press). These investigators
studied mangrove metabolism and structure of a three-way mixed forest at
Rookery Bay, Naples, Florida.
It is evident that the gross productivity of mature red mangroves decreases
with increasing salinity (Table XII-18). The opposite trend was true for mature
black and white mangroves, in which gross metabolism increased with salinity.
These production-salinity relationships are graphically shown in Figure XII-20.
In the case of the white and red mangrove response, the GPP values reported for
Station E were omitted from the figure. As indicated previously, the metabolic
Table XII-18. Mangrove metabolism with respect to changes in chloride concen-
tration, 1972. Metabolic values corrected for leaf, trunk, and proproot
area indices.
Plant Metabolism
Species
Red Mangrove
Black Mangrove
White Mangrove
Station
D
E
F
G
D
E
F
G
D
E
F
G
Mean
Chloride
(ppt)
4.7
11.6
12.7
16.0
4.7
11.6
12.7
16.0
4.7
11.6
12.7
16.0
Proproot
R24
0.40
ND
0.50
0.59
NA
NA
NA
NA
NA
NA
NA
NA
Trunk
R24
0.07
ND
0.03
0.04
0.24
ND
2.7
3.6
ND
ND
0.08
0.14
(gm c/m2/day)
Leaves
R74
1.8
1.7
0.37
0.67
1.1
ND
0.34
1.7
ND
1.7
0.36
2.3
NPP24
5.7
6.3
3.0
0.25
0.94
ND
2.7
2.2
ND
13.0
1.8
2.4
Species
GPP
8.0
8.0
3.9
1.6
2.31
ND
5.7
7.5
ND
15.0
2.2
4.8
Compartment Indices a
Proproot
0.67
0.67
0.67
0.67
NA
NA
NA
NA
NA
NA
NA
NA
Trunk
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
Leaves
1.9
2.1
1.4
1.4
2.1
2.1
1.4
1.4
1.4
2.4
1.4
1.4
a m ^ metabolic surface per m land area
b representative of immature specimens
c calculated estimate
measurements were for immature trees not comparable structurally or metabolically
to the mature individual trees studied. To view the response of community
metabolism to changing conditions of fresh water only requires a summing of the
species GPP rates by station, as reported in Table XII-18. Community metabolism
is given in Table XII-19.
From Table XII-19 it is evident that the carbon budget (GPP) generated by
the mangrove system grows larger as fresh water becomes less available. This
appears to be a plus for the community. However, a closer examination of the
situation reveals that increased respiration applies heavy pressures against
the carbon economy of the system. This is apparent by comparing the percentage
of the carbon budget devoted to respiration at Stations D, F and G. Respectively,
respiration demands are 35.9, 36.9, and 65.5 percent of the total budget. This
aspect of the data has already been discussed in terms of the salt marsh
community. Some further discussion of mangrove community response is given in
Chapter V.
XII-38
-------
UPLAND COMMUNITY BIOMASS AND
METABOLISM
Biomass
Introduction
This section reports
studies conducted to assess
drainage effects on upland
terrestrial plant communities
of the Fahkahatchee Strand
and associated areas. Study
emphases were placed on the
cypress and wet prairie
communities. Of the two, only
the former community was
studied thoroughly enough to
yield sufficient data on
biomass to make comparisons
between drained and undrained
regions (Figure XII-21 areas
A and B, respectively). Details
of the plant taxonomic features
of the areas are reported in
Chapter XIII.
Methods
10
WHITE MANGROVES
10
CM
o
a.
(D
BLACK MANGROVES
10 r
Two techniques were used
to assess changes in community
biomass. The least difficult
of the two involved the herba-
ceous plants associated with
the ground cover of the cypress
strands and the wet prairies.
In these areas, the plants,
including roots in a 0.25 m
plot were harvested at selected
points along a transect that
measured 100 meters long and
was sequentially numbered at
one meter intervals. Sampling
was with replication, usually
four. Points of harvesting
were selected from a book of random numbers.
RED MANGROVES
5 10 15 20
MEAN SURFACE WATER CHLORIDES
(ppt)
Figure XII-20. Response of mangrove gross
primary productivity to decreasing con-
centration of fresh water. Nov.-Dec.
1972.
The harvesting method included separating the above ground vegetation
into dead and green components. The roots, from a maximum depth of about 30 cm,
were first separated from the soil with a coarse mesh sieve (6 mm mesh) and then
washed and drained. All material was oven dried at 103 + 2° C for three to four
days, cooled to room temperature, weighed, and analyzed for ash-free dry weight.
Biomass results are reported in both dry and ash-free weight per square meter of
community.
XII-39
-------
Figure KII-21. Location of sampling areas,
-------
Table XII-19. Response of mangrove community metabolism to changing concentra-
tions of chlorides (fresh water), 1973.
Station
Mean Surface
Water Chlorides
(PPt)
Community Metabolism
(gm C/m2/day)
GPP
24
R
24
D 4.7
F 12.7
G 16.0
10.3
11.8
13.9
6.6
7.5
4.8
3.7
4.3
9.1
Harvesting of large numbers of trees and shrubs to assess changes in
community biomass was impractical. To save both time and landscape, several
large quadrats (10 m X 10 m) were established in the two cypress strand areas
(areas A and B, Figure XII-21). In each plot, all trees and shrubs were
identified and their position in the plot marked by coordinates. Figure XII-22
gives an actual example of a surveyed quadrat in area B. Each specimen was
0 123456789
I 23456789
1
2
3
4
5
6
7
8
9
,0
A J3 •
A///
— // V ,*t/
U •
X \\\\*
• X '/ 0 ®
o X. ®® /
• • • ^ •
® b ®
A», Q
-.•*'''
o
•0 ®©®
®
O
0 .'. ® c
/// .
i '* A
T •!
o D x gr
I ^
o
•<»/»•
(29
t 0 / A -
O ®
f ®
® / 0
• ® A ' ««
1 I.I I 1 1 1 1 I (VI O
1
2
3
4
5
6
r
8
9
EXPLANATION
OTaxodium \ Magnolia
A Sabal x Baccharis
• Annona ® Taxediurn -SAPLING DIA. <3.0cm
/ Persea x Salix
W
A
SCALE
Him N
Figure XII-22. Example of a surveyed quadrat.
XII-41
-------
assigned an aluminum tag bearing physical dimensions of the plant, date of
survey, and coordinate numbers. This tag served to identify the individual at
a later date when the indicated dimensions were remeasured. Changes in biomass
were determined from the before and after measurements. The relationship of
selected physical dimensions to biomass was derived in the following manner.
Regression equations describing biomass as a function of selected physical
dimensions were established specifically for this study. Data for the regression
analysis were obtained by measuring, harvesting and weighing different sized
individuals of a given tree or shrub.
First, tree trunk diameters were measured at a point of 137 cm above
ground with a DBH meter (Diameter at Breast Height). To measure small diameters
(< two cm) a caliper was used. In the case of shrubs or seedlings, trunk diameters
were measured at a point of 20 cm above ground.
After the tree or shrub was cut down with a chain or hand saw, the total
height of the plant was measured. Then the leaves and woody components were
separated. The woody parts were separated further into twigs (< one cm in
diameter) which included leaf buds and stems, limbs, and trunk. All vegetative
fractions were weighed for fresh weights and subsampled for dry weight conversion
and chemical analyses. Fresh weights were obtained in the field with the use of
a 22 and 2.6 kilogram capacity triple beam balances.
Calculation of the regression equations and correlation coefficients was
accomplished with a statistical programmable calculator. The calculator programs
test various regression forms (linear, exponential, or parabolic) using various
physical dimensions (diameter, height, diameter times height or diameter squared
times height). Data used in the regression analyses are given in Appendix S.
The power curve relationship of biomass vs square root of diameters
multiplied by height was found to yield the highest correlation value of the
different regression models tested (Appendices T, U, and V). Only the Sabal
palmetto (cabbage palm) failed to conform well to the (D-H)^'^ regressions.
One might expect the palm, with its columnar growth, not to conform with the
regression patterns of dicots and conifers.
Generally, the woody and leafy biomass regression equations show correlation
coefficients well above 0.9. The exceptions were leafy biomass regressions for
a few deciduous tree species. Since the deciduous species were sampled in the
late fall, this is not unexpected because of possible differences in leaf-fall.
Results and Discussion
Dramatic differences in standing crop between the drained and undrained
cypress strand areas were evident. The drained area (area A, Figure XII-21)
was consistently lower in biomass and showed an average standing crop of woody
plants of 8,935.5 grams dry wt/m2 (from March 15, 1972 to January 16, 1973)
compared to the undrained area (area B, Figure XII-21) of 17,120.2 grams dry
wt/m2 during the same period. Comparison of the growth of biomass was equally
impressive. The drained area of woody plant biomass increased by 119.6 grams/m
during the 307 day period. The undrained area yielded a biomass increase of
485.0 gm/m2 during a 302 day period (Table XII-20). This annual increase trans-
lates into rates of 0.390 gm/m2-day and 1.606 gm/m2.day, respectively. The
undrained site had a rate of increase in biomass over four times that of the
drained area.
XII-42
-------
Table XII-20. Community stand table for woody plant biomass (wood and leaf
masses) in an undrained and drained area of a cypress strand.
Undrained Area
2 2
Genus Mean Biomass gm/m Change in Biomass gm/m
Taxodium
Myrica
Baccharis
Sabal
Annona
Magnolia
Salix
Persea
15,790.3
262.4
11.6
890.7
8.1
8.6
2.0
146.8
81.995
106.599
5.164
252.029
0.327
2.506
0.187
36 . 339
Total 17,120.2 485.Oa
a For 302 days = 1.606 gm/day.
Drained Area
- _ _
Genus Mean Biomass gm/m Change in Biomass gm/m
Taxodium
Myrica
Salix
Cephalanthus
Persea
Fraxinus
Acer
8,167.6
613.7
100.2
42.5
10.3
0.2
1.1
44.106
73.482
-2.773
-0.405
3.36
0.005
0.245
Total 8,935.5 119.1
b For 307 days = 0.390 gm/day.
The herbaceous ground cover plants of the cypress strands followed a
similar trend (Figure XII-23). The average standing crop of herbs in the drained
area was 1,153.5 gm/m2 from January 31, 1972 to October 12, 1972 compared to
1,945.3 gm/m2 in the undrained area (from February 11, 1972 to September 9, 1972).
The change in biomass in the drained area was -20.0 gm/m2 for the 255 day period
compared with +311.6 gm/m2 for the 211 day period in the undrained area. This is
a rate of -0.078 gm/day and 1.477 gm/day, respectively.
Combining the woody plants with the herbaceous plants the net gain in
biomass becomes 0.312 gm dry wt/day in the drained area and 3.083 gm dry wt/day
in the undrained area. The net increase in biomass was 10 times greater in the
XI1-43
-------
undrained cypress strand
than the drained cypress
strand.
A wet prairie in the
drained region of study area
(area A, Figure XII-21)
was sampled for estimates
of standing crop biomass
on four occasions during
the period of January 31
to October 5, 1972. Two
wet prairies remote to
canal drainage were
sampled in area B (Figure
XII-21) during the period
February 11 to August 9,
1972. Results from
sampling are presented in
Table XII-21. Prairie
fires in area B nullified
any reasonable judgements
concerning changes in
standing crop biomass and
differences in biomasses
between drained and undrained
prairies.
*g 2400
£
* 22OO
EXPLANATION
SOIL TYPE CF A UNDRAINED
O DRAINED
I = ± 2 STANDARD ERRORS
1972
FigureiXII-23. Average standing crop biomass of
herbs in drained (area A, Figure XII-21) and
undrained (area B, Figure XII-21) cypress
strands, 1972.
Table XII-21. Average standing crop biomass with 95 percent confidence limits
for wet prairies near (area A) and remote (area B) to a drainage canal,
1972.
Date Study Soil
1972 Area Type
Jan
Mar
. 31
. 10
June 12
Oct
Feb
Apr
Aug
Au
Feb
Apr
Aug
Apr
Aug
a
b
. 5
. 11
. 14
' 2a
. 7a
. 16
' 14a
' 14b
. 17b
. 9
Prairie
it
A
ii
11
ii
B
it
ti
it
it
it
ii
ti
it
burned
ti
Oe 154
101
" 151
" 308
" 87
" 63
" 259
11 234
Od 122
" 93
" 458
" 29
" 229
April 20
March 21
Lvs.
.1 ±
.6 ±
.7 ±
.2 ±
.0 ±
.7 ±
.6 ±
.6 ±
.1 ±
.0 ±
.4 ±
.0 ±
.3 ±
Biomass, Ash Free Wt. (gm/m2)
Duff
59.7
41.7
62.3
343.7
38.8
25.8
58.6
234.9
30.5
36.3
533.6
5.6
45.7
301.2 ±
284.0 ±
368.8 ±
293.2 ±
500.4 ±
430.7 ±
620.6 ±
46.0 ±
296.2 ±
426.2 ±
104.4 ±
40. 4 ±
139. 8 ±
37.6
80.7
170.4
510.6
183.4
167.7
148.1
73.2
119.6
163.4
22.2
10.6
129.1
Roots
461.5 ±
513.8 ±
412.2 ±
608.8 ±
517.0 ±
461.1 ±
653.7 ±
616.4 ±
428.6 ±
453.3 ±
871.1 ±
781.6 ±
1,064.7 ±
143.9
168.0
128.7
262.7
178.7
111.0
106.4
188.7
86.9
165.6
923.7
172.4
250.0
Total
916.8 ±
899.4 ±
932.7 ±
1,210.2 ±
1,104.5 ±
955.4 ±
1,533.9 ±
897.0 ±
843.2 ±
972.4 ±
1,334.0 ±
850.9 ±
1,433.8 ±
177.3
194.3
237.6
1,105.3
256.2
257.8
289.7
496.8
291.7
328.3
1,435.1
172.9
302.5
, 1972
ft
»
XII-44
-------
Litterfall in Woodlands
Introduction
Litterfall represents the fraction of plant net production which is not
grazed, remineralized through decomposition, and accumulated as biomass. In
stable mature woodlands much of the woody plants biomass ultimately becomes
litterfall which enters into a detritus food chain (Odum, E. P., 1969).
Because litterfall is a function of net production in woodlands, it becomes
a basis for assessing the effects of land drainage on woodlands. Litterfall is
an important part of energy and nutrient budget for woodland ecosystems; a
subject considered in Chapter II (Discussion and Conclusions). This section
reports the results of litterfall studies in the various woodland communities
of the study area. These woodlands included a segment of the central Fahkahatchee
Strand (Figure XII-21, area C), a cypress strand remote to canal drainage (Figure
XII-21, area B) and a cypress strand near a drainage canal (Figure XII-21, area
A). For purposes of this study, areasB and C are referred to as undrained and
area A as drained.
Methods
Collections of lirtterfall were with catch baskets measuring either one
meter square or 0.25 m . Six of the latter units were used in the central strand
and 18 to 20 of the smaller baskets were placed in each of the cypress strands.
The m2 baskets were shallow trays of fiberglass screening over a galvanized
hardware cloth molded with six-inch sides. The smaller baskets were wooden
frames of 3/4" x 2" firring strips with fiberglass screen bottoms. Mesh size
of the screening was approximately 1.5 by 1.5 mm. The baskets were placed above
the ground (to reduce moisture and decomposition of the litter) at regular
intervals. Litter was collected periodically, separated, dried and weighed.
Subsamples of the separate components, leafy, woody and fine (small or fragmented
particles), were analyzed for organic nitrogen and phosphorus content and ash
free dry weight. Litterfall rates reported were corrected for decompositional
losses during periods between collection times (see the following section on
decomposition.
Results
The following litter accumulation graphs (Figures XII-24 through 26) present
the litter accumulation in the wooded communities. Tables XII-22, 23 and 24 give
average daily litterfall based on litter collected throughout the year and
corrected for decomposition in the baskets.
Comparison of the woody sites indicates that the central strand produced
the greatest amount of litter (756 gm/m2.yr). This area was characterized by a
dense canopy of vegetation, rich organic soil, remote to drainage canal, and a
lengthy hydroperiod. The central strand was considered the most stable and
mature woodland community studied for litterfall in this report.
The cypress strand areas were characterized by fewer trees and more
herbaceous plants than the central area. The undrained cypress strand showed an
annual average rate of litterfall of 374 gm/m2/yr compared to somewhat less than
267 gm/m2/yr in the drained area. The drained area also maintained less standing
crop biomass and developed the least amount of community growth.
XII-45
-------
500
j-. 400
6
!> 300
I-
» 200
LU
*
£ 100
500
AM
^P= 1.02309 gm/DAY
374.45 gm/YEAR
Figure XII-24. Litter accumulation in an undrained cypress strand (area B,
Figure XII-21) 1971-73.
900
800
Z" 700
E
» 600
£ 500
IT
UJ 400
?
o
300
ZOO
too
AM . 2.0655 gm/DAY
AT " 755.973 gm/YEAR
900
800
500
300
100
o n o
1971
J J A S
1972
Figure XII-25. Litter accumulation in the central strand (area C, Figure XII-21),
1971-72.
.AM..0.7303 gm/DAY
DT " 267.28 gm/YEAR
^.300
UJ
-I >
^QI
O o>
300
200
100
j J
1972
Figure XII-26. Litter accumulation in a drained cypress strand (area A, Figure
XII-21), 1972.
XII-46
-------
Table XII-22. Litterfall in the central strand - undrained.
Mean Litter
Collection Period of Wood
Date Accumulation (gm/m2 Dry Wt.)
Nov. 11, 1971 27 13.00 ± 18.44
Dec. 3, 1971 22 8.45 ± 10.37
Jan. 12, 1972 40 6.28 ± 4.82
Mar. 3, 1972 51 71.13 ± 72.78
Apr. 12, 1972 39 32.56 ± 58.15
May 12, 1972 31 5.62 ± 5.11
June 12, 1972 30 60.78 ± 53.32
July 26, 1972 45 9.37 ± 17.58
^
Oct. 18, 1972 84 107.35 ± 196.52
Nov. 2, 1972 15 4.02 ± 3.30
Dec. 13, 1972 41 4. 75 ± 6.31
Total litterfall
Mean yield per day
20.
Drop (x ± ^r-j
(gm/m2 Dry Wt.)
68.78 ± 9.66
51.23 ± 11.96
29.98 ± 7.11
46.52 ± 17.02
16.18 ± 6.40
25.85 ± 7.54
29.85 ± 9.56
20.93 ± 17.06
144.48 ± 17.31
32.77 ± 10.78
57.63 ± 11.18
Table XII-23. Litterfall in the cypress strand -
Mean Litter
Collection Period of Wood
Date Accumulation (gm/m2 Dry Wt. )
Mar. 3, 1972 58 4.60 ± 5.04
Apr. 10, 1972 38 0.84 ± 0.88
May 12, 1972 32 0.96 ± 0.12
June 12, 1972 31 2.24 ± 1.52
Jan. 15, 1973 217 1.36 ± 0.16
Total litterfall
Mean yield per day
Drop (x ± -fy
Leaf and
(gm/m2 Dry Wt.)
49.84 ± 23.00
15.56 ± 10.44
17.16 ± 6.92
23.76 ± 8.60
136.36 ± 44.84
Mean Litter Drop: Corrected for
Losses to Decomposition Subtended
by Yield Per Day of Litter
Wood Leaves
(gm/m2 Dry Wt.) (gm/m2 Dry Wt.)
13.31 69.89
(0.49) (2.59)
8.96 51.90
(0.39) (2.36)
6.50 30.70
(0.16) (0.77)
74.33 47.95
(1.46) (0.94)
33.68 16.56
(0.86) (0.42)
5.77 26.33
(0.19) (0.85)
62.38 30.39
(2.08) (1.01)
9.74 21.50
(0.22) (0.48)
115.39 151.83
(1.37) (1.81)
4.07 33.06
(0.27) (2.20)
4.92 59.05
(0.12) (1.44)
755.97
2.06
drained .
Mean Litter Drop: Corrected for
Losses to Decomposition Subtended
by Yield Per Day of Litter
Wood Leaves
(gm/m2 Dry Wt.) (gm/m2 Dry Wt.)
4.836 51.60
(0.083) (0.89)
0.868 15.92
(0.023) (0.42)
0.987 17.48
(0.031) (0.55)
2.301 24.20
(0.074) (0.78)
1.633 154.76
(0.008) (0.71)
267.28
0.73
XII-47
-------
Table XII-24. Litterfall in the cypress strand - undrained.
Mean Litter Drop (x ±
Leaf and fine
Collection Period of Wood material
Date Accumulation (gm/m2 Dry Wt. ) (gm/nT Dry Wt. )
Mean Litter Drop: Corrected for
Losses to Decomposition Subtended
by Yield Per Day of Litter
Leaves
Wood
(gm/m2 Dry Wt,) (gra/m2 Dry Wt.)
Feb.
Mar.
Apr.
May
July
Sept
Nov.
Jan.
11, 1972 65
21, 1972 39
18, 1972 28
20, 1972 32
31, 1972 72
. 8, 1972 39
2, 1972 55
17, 1973 76
1.30 + 4.28 29.90 ± 5.96 1.375
(0.021)
3.40 ± 5.80 24.18 ± 15.06 3.516
(0.090)
0.22 ± 0.20 5.32 ± 2.45 0.225
(0.008)
0.44 ± 0.56 14.66 ± 9.34 0.452
(0.014)
12.16 ± 15.05 ' 12.18 ± 5.79 12.938
(0.180)
0.53 ± 0.59 8.44 ± 4.02 0.548
(0.014)
1.12 ± 1.23 78.40 ± 16.44 1.174
(0.021)
29.26 t 54.46 177.16 ± 42.69 31.238
(0.411)
Total litterfall
Mean
yield per day
31.17
(0.48)
24.75
(0.63)
5.41
(0.19)
14.94
(0.47)
12.71
(0.18)
8.64
(0.22)
81.00
(1.47)
185.30
(2.44)
374.45
1.02
Discussion
It is difficult to make a judgement about a plant productivity based on
litterfall unless something is known about community structure. Based on
results of the biomass studies of cypress strands, it is logical to assume that
the drained area with its lower biomass, lower community growth and lower leaf-
fall than the undrained area is in a reduced state of activity.
The turnover rate of leaf litterfall was greater in the drained cypress
than in the undrained even though the absolute amount was less. The turnover
rate of leaves for the undrained was 0.791 turnover per year. This value was
arrived at by dividing leaffall per year (364 gm/m ) by community mean leaf
mass (460 gm/m ). In the drained area the leaf turnover rate was 1.035 turn-
overs per year. This value was generated in the same manner given above. The
45 percent increase in turnover rate in the drained area may be due to stress
from drainage resulting in a process of canopy thinning.
Decomposition of Litter
Introduction
Decomposition is an important process necessary in the cycling of minerals
and nutrients. The rate of decomposition of litter in a given community
determines the amount of carbon, nitrogen, phosphorus, and other minerals and
XII-48
-------
nutrients that are tied-up in the litter component of a system and not available
for converting to living protoplasm. Generally, high rates of remineralization
accompany high productivity unless the decomposition attacks a substantial part
of living matter. Enhancing the cycling of nutrients by rapid decomposition of
litter speeds along productivity process.
This section reports decomposition rates of litter accumulations in the
central strand, comparing microclimatic conditions on the strand floor and the
dry elevated debris piles. The conditions on these elevated debris piles
resembled those on the floor of dry or drained strands. In the latter situation
there was a build-up of poorly decomposed litter and a reduction of humus, as
compared with the central strand floor (Chapter XIII).
Methods
Decomposition was measured by setting out bags containing known quantities
of litter and collecting sets of bags periodically. The initial litter used was
collected at the same time from litter baskets. The initial litter was sub-
sampled for analyses of organic nitrogen and phosphorus, dry weight, and ash
free dry weight, and then separated on the basis of component (leaves, wood,
and fine litter). Each component was divided equally by weight among 12 bags.
The bags were of two types: six with 1/32 inch mesh size (fine mesh), and six
with 1/16 inch mesh (coarse mesh). All bags were the 12 x 12 inch size with
delta style knit from a commercial fisherman supply house that sold the items
as "bait bags". A total of 24 bags were prepared: 12 were placed on the
strand floor and 12 on debris piles. Both sets contained six coarse mesh bags
and six fine mesh bags. This number of bags allows sampling of a fine and
coarse mesh bag for both sites every two months for one year. The quantity (Q)
of litter remaining in the bags at collection time is assumed to follow an
exponential relationship with time (t): Q = aebt (Figures XII-27 through 30
and Appendices W, X, Y and Z) . This relationship assumes that decay is dependent
on quantity. Decay equals quantity present times a constant. It follows that
litter falling into the litter baskets (J) is related to time (t), quantity
present (Qt) and the decomposition coefficient (k = -b) according to the equation
J = (Qtk)//l-e~kt, i£ the quantity at time zero is zero and the litter falls at
a steady rate. The decomposition coefficient equals -b in the equation Q=ae"'Dt:
describing the quantity of litter in a debris pile decay bag after a length of
time.
Results
The decomposition rate on the strand floor was substantially higher than
that on the debris piles. The coefficient of decomposition was 1.72 x 10~3
gm/gm-day dry weight for leaves on the strand floor compared to 1.33 x 10"-*
gm/gnrday on the debris pile, both in fine mesh bags. For the coarse mesh
bags, the difference was even greater; 2.69 x 10~3 gm/gm-day on the strand floor
and 1.049 x 10"3 gm/gm.day on the debris pile (Table XII-25). Processes of
remineralization as shown leads to a reduction in carbon content of litter;
however, these processes result in a build-up of organic nitrogen. Th^'s is
evident in Figure 11-15.
XII-49
-------
40
E
o>
X
o
in 30
LL)
U
CC
< 20
Q
(E
O
10
DRY WEIGHT
O DRY WEIGHT
A ASH FREE WEIGHT
•ASH FREE WEIGHT
N D
1971
J J
1972
Figure XII-27. Litter decomposition on central strand floor, fine mesh. Figure?
based on data given in Table XII-34, 1972.
40r
o>
X
O
UJ
$
UJ
30
20
Q
10
EXPLANATION
° DRY WEIGHT
A ASH FREE WEIGHT
A
1—ASH FREE WEIGHT
N D
1971
J J
1972
Figure XII-28. Litter decomposition on central strand floor, coarse mesh. Figure
based on data given in Table XII-35, 1972.
"i so
O
UJ
x
<
O
10
EXPLANATION
O DRY WEIGHT
A ASH FREE WEIGHT
r
DRY WEIGHT
D
1971
1972
Figure Xll-29. Litter decomposition on debris pile in central strand, fine mesh.
Figure based on data given in Table XII-36, 1972.
XII-50
-------
30
X
UJ
UJ
u.
r
a
a:
o
. o
10
RY WEIGHT
EXPLANATION
O DRY WEIGHT
A ASH FREE WEIGHT
ASH FREE WEIGHT
D
1971
J J
1972
Figure XII-30. Litter decomposition on debris piles in central strand, coarse
mesh. Figure based on data given in Table XII-37, 1972.
Table XII-25, Turnover rates of litter components, from least squares regression
of data from decomposition materials. Units are in 10"3 (turnovers/day).
Litter Decay
Bags
Coarse mesh
Dry wt.
Ash- free wt.
Fine mesh
Dry wt.
Ash- free wt.
Total
2.1309a
2.2799
1.339lf
1 . 6047b
Wet
Leaf § Fine
2.6860a
2.9839
1.7227a
2.0580b
Wood
1.3098a
1.2581a
0.5877C
0,5490C
Total
1.0637a
1.4037b
1.4542a
1.8309b
Dry
Leaf § Fine
1.0492a
1.3528C
1.3338a
2.3154b
Wood
1.1408a
2.3412a
2.9524b
Dry wt.
Ash-free dry wt.
Combined data (coarse
Total
1.259a
1.618a
fine mesh bags) for debris pile (dry)
litter bags _
Leaf § fine
Wood
1.192a
1.989a
1.741a
2.318a
a Significant at 1% level.
b Significant at 5% level.
c Not significant at 5% level.
Discussion
Litter decomposition on the strand floor generally progressed at a greater
rate than litter placed on the debris pile. This was probably due to increased
activity of bacteria and detrital feeding organisms. The strand floor bags
XII-51
-------
contained numerous limpits, amphipods, and even a sizeable crayfish during the
hydroperiod. The debris pile bags also contained a number of species of
invertebrates who were mainly small insects but their total mass was much
smaller than that found in the strand floor bags. One collection of strand
floor litter bags contained 0.07 gm in the coarse mesh bag. A later collection
contained 0.15 gm invertebrate dry wt in the fine mesh bag and 0.25 gm dry wt
of invertebrates in the coarse mesh bag. The large difference in faunal mass
between the coarse and fine mesh bags helps explain why decomposition was more
rapid in the coarse mesh.
This decomposition experiment does not take into account seasonal effects
on remineralization because the initial litter collected for the study was
obtained at only one point of the season. A comprehensive study of litter
decay should examine the decomposition of litter collected at every season.
The information gained in this study, however, does establish that microclimatic
conditions associated with inundation effects remineralization.
TERRESTRIAL COMMUNITY METABOLISM
Introduction
Wet prairie communities constitute the second largest areal representation
in the study area (Figure V-5a). This, coupled with ecotonal involvement of the
grassy understory of cypress strands makes the wet grasslands, on a basis of
land coverage, the principal plant community of the Big Cypress Swamp (Figure
V-5b). Fortunately, these areas were considered ideal for metabolic studies
since segments of the community could be readily isolated in metabolic chambers
for field investigation.
Field studies were conducted to assess community primary productivity and
evapotranspiration. The objective of these investigations was to relate
community metabolism to ground water availability. Thus, study sites were
established in prairie regions both near and remote to drainage canals.
Methods
Measurements of wet prairie community metabolism were.accomplished in the
same general manner previously described for the saltgrass marshes. Details
of the methodology and equipment applied are given as an appendix to this
chapter. An abbreviated version of the methodology and materials is reported
in this section. In this case only the above ground portion of the wet prairie
community was subjected to measurements of metabolic activity.
Open-ended chambers constructed of 6 mm diameter steel rod framing and
covered with clear polyacetate (0.05 cm thick) were set over a segment of the
prairie community and sealed to the ground. A fresh air supply to each chamber
was regulated at a fixed rate that maintained a chamber overturn rate of five
per minute. Temperatures within the chambers were usually less than 5° C above
ambient temperatures.
Community photosynthesis, respiration, and evapotranspiration were assessed
on the basis of measured changes in the carbon dioxide and water vapor gradients
across the chamber. Rates for community net primary productivity, respiration,
XII-52
-------
and evapo'transpiration are reported as totals derived from trapezodial inte-
gration of hourly rates and in terms of a per unit land area of the community.
Results
Field studies of wet prairie metabolism were conducted in June and
October, 1972, at prairie site "A" and in August, 1972, at site "B". As
indicated in the study of terrestrial biomass, Figure XII-21, the former area
represented a region of the swamp adjacent to a major drainage canal and the
latter area was a relatively undrained region. Results of the measurements of
prairie activity are presented in Table XII-26.
Table XII-26. Results of studies to measure wet prairie community productivity,
respiration, and evapotranspiration in areas near (area A) and remote
(area B) to drainage canals. 1972.
Site Date Soil
Area 1972 Type
A June
ii it
" Oct.
II It
II II
11 II
B Aug.
M II
II II
II II
B "
ft 11
it it
M II
12 Oe
II M
4 ••
M tl
5 "
6
2 "
II It
3j "
7d „
9e Od
tr tf
it ii
14d "
Chamber Biomass, Ash
fem/m^l
Green
Civs.)
178.2
128.3
193.1
269.8
193.1
269.8
283.7
258.6
259.6
216.1
230.5
247.0
400.4
316.4
Duff
290.5
397.7
142.2
210.1
142.2
210.1
674.1
631.3
556.4
40.3
199.6
105.6
102.7
106.2
Roots
429.5
453.2
501.9
610.9
501.9
610.9
660.4
692.8
607.9
601.5
1111.6
949.2
943.8
798.4
Free Wt.
Total
822.7
979.2
837.2
1,090.8
837.2
1,090.8
1,618.2
1,582.7
1,400.8
857.9
1,541.7
1,301.8
1,446.9
1,221.0
Photo-
Period
(hrs.)
13.7
If
11.9
II
II
II
13.2
II
II
13.1
M
13.0
M
Community Metabolism
Light
Uy/day)
547
267
"
426
248
684
It
508
453
480
II
453
413
NFP,
P2
gm C/m
6.22
5.33
1.08
2.97
1.20
3.11
7.32
8.86
2.21
9.00
6.16
6.05
9.18
15.05
%>
gm C/Tsf
4.31
4.26
1.97
0.97
1.14
1.27
1.87
1.12
1.30
4.37
3.50
3.29
4.44
8.64
tl
2
mm H20/m /day
3.92
4.63
ND
it
1.68
4.29
5.28
S.57
ND
8.24
4.38
4.51
7.00
5.74
PEC
mm H 0/m /day
6.00
6.00
II
3.90
6.38
8.87
ND
8.51
6.38
It
5.32
a NPP. = net primary productivity during photoperiod.
Lp
R_ = respiration during darkness.
ET = evapotranspiration for 24 hours
b Area A = drained region
Area B = undrained region
c PE = pan evapon tion at Big Cypress Bend.
d Prairie burned April 20, 1972.
e Prairie burned M irc.h 21, 1972.
To expand the reported activity rates into estimates of mr.an daily rates
of gross primary productivity and community evapotranspiration, f..he observed
values required correction for average daily solar radiation and evapotranspira-
tion for the month. The latter parameter can be normalized on the basis of
average daily pan evaporation. The relationship between communil.y evapotrans-
piration and pan evaporation was previously established in Chapter V. The
conversions and calculations of mean gross productivity were based on the
following series of equations.
XII-53
-------
_ - NPP
NPPT = I v_
P
n
*DP =
n
= NPPLp - RDp
= 24'(RD/Dp)
GPP = NPP24 + R24
- ET
P p o
ET24 = Po
n
where:
ET
„, = mean evapotranspiration in 24 hour day for month
NPP = mean net primary productivity during photoperiod
lip
I = mean daily total solar radiation for month
NPP = observed net primary productivity
I0 = observed total solar radiation (ly/day)
n = replicates
R~ = mean nighttime respiration
R^ = observed nighttime respiration
NPP,-,, = mean net productivity in 24 hour day for month
R24 = mean respiration in 24 hour day for month
Dp = mean nighttime period for month
GPP = mean gross productivity in 24 hour day for month
P = mean 24 hour pan evaporation for month
ET = observed evapotranspiration for 24 hours
Po = observed 24 hour pan evaporation at Big Cypress Bend station.
From the,above series of conversions, the following Table XII-27 presents the
mean observed net productivity and respiration rates with one standard error of
mean indicated. The second part of the table shows mean estimates of daily
community carbon and water metabolism for the months indicated.
XII-54
-------
Table XII-27. Mean rates of productivity and evapotranspiration for wet prairie
communities adjacent (A) and remote(B) to major drainage canals. Produc-
tivity rates normalized for equal conditions of total solar radiation.
1972.
Metabolism (gm C/m )
Site
Area
A
"
B
ii
"
Soil
Type
Oe
H
1!
II
Od
Date
1972
June 12
Oct. 4-6
Aug. 2-3
Aug. 7a
Aug. 9-14b
NPP
W Lp
±SX
5.26 ± 0.44
3.28 + 0.60
4.26 ± 2.0
9.06
9.36 + 2.1
R
Dp
±Sx
4.28 ± 0.22
1.34 ± 0.22
1.42 ± 0.22
4.45
4.97 ± 1.2
No.
Replication
2
4
3
1
4
Metabolism (gm C/m /day or mm H-O/day)
Month
GPP
24
R
24
NPP
24
ET
24
A
ii
B
ii
it
a
b
Oe June
" Oct.
" Aug.
11 Aug.a
Od Aug.b
Prairie burned April 20,
Prairie burned March 21,
10.92
4.60
6.01
14.50
15.28
1972
1972
9.24
2.66
3.17
9.89
10.89
0.98
1.94
2.84
4.61
4.39
3.5
2.6
3.9
4.1
4.35
The above reported results have been incorporated into the modeling section.
details of discussion refer to Chapter VI.
For
XII-55
-------
XIII - PLANT COMMUNITY STRUCTURE
COMMUNITY SURVEY AND CLASSIFICATION
Communities are often described by their dominant vegetation types.
Vegetation, an element both conspicuous and more or less permanent, provides
perhaps the most convenient means of characterizing a biological community
during brief on-site inspections.
When large tracts of land are involved it is difficult to examine the
entire study area and inspect every parcel of land within a reasonable length
of time. Some substitute means of survey, other than detailed on-site
inspection, must be employed. Soil types and vegetation types are frequently
strongly correlated, for vegetation has a great influence on the soil and vice
versa. Soil survey maps are often well illustrated and periodically updated
by aerial photographs.
The Collier County soil survey (Leighty, _et al., 1954) briefly lists
the more dominant plants found in each soil type. To further describe the
community as to composition and magnitude, one might survey examples of the
various communities, and attach soil types and a more detailed list of the
plants found in each community. From the aeral extent of the associated soil
types (Table XIII-1), the extent of associated plant communities can be determined,
The soil type designations in this table refer to the system used in the 1954
soil survey, and are included here to document the area partition of community
types used elsewhere in this report.
Habitat Classification
Many communities can be divided into subtly varying habitats. Even
though a community as a whole maintains a certain continuity inherent by
definition, it still contains distinct zones of different types of vegetation.
Vertical zones or habitats may be caused, inter alia, by stratification of
light through the various canopy levels. Horizontal zonation may be related
to variations in exposure, standing water duration and depth, or by a variety
of substrata for rooting and attachment of the plants.
For example, the cypress strand community contains many microhabitats
apparently based on exposure to different levels of light, temperature, water,
and substratum. The cypress strand contains both aquatic and terrestrial
habitatso The strand floor is in many places inundated most of the year. Here
aquatic herbs thrive. In this same area, large terrestrial woody plants also
live. The cypress strand provides a home for both terrestrial and aquatic
plants and animals. The strand floor may contain fishes, Crustacea, and
insect larvae considered aquatic; but the trees and elevated parts of the
strand support terrestrial insects, birds, reptiles, and mammals. Light
stratification also separates the strand into various habitats. The various
canopy levels contain distinct vegetation types, as is the case for most
woody communities. In addition to various light strata and hydroperiods (time
of exposure to water), the cypress strand encompasses a variety of substrate,
to which distinct plant groups become1 rooted and attached. These include soil,
rock, logs and stumps, and other plants (epiphytic), or surface water (free
floating plants).
XIII-1
-------
Table XIII-1. Communities of the Big Cypress and the associated soil types
(from Leighty, et al., 1954).
Community
Soil Types
Cypress and hardwood swamp (Strand)
Cypress with grassy understory
Wet prairie
Sawgrass prairie
Pine
Sabal (Cabbage palm)
Freshwater marsh
Salt m.arsh
Mangrove forest
Estuary and lagoons
Misc. (coastal beach, shell mounds,
made lands, and coastal hammock)
Cf
ob5,
, Beg*, and Rag*
Ob, Oc, Od, Oe, Of, Be* Ra3, Ra *
Fa, Aa_, Ca , and Pa 9
Oa, and Tb
Ob2, Aa, Be, Be * Ce, Bb, Be,
Bd, Kb, la, Me, Ra2, Ra *, Sa,
Sc, Ba, Ca, La, ana Pa
Be, Cd, Bb, Be, Ka, Kb, Me,
Sc, Ca, Cc, Bc4, Bd4, Ia^,
and Pa
Fb
Ta
Mangrove
Estuary
Cb, Ra, Sb, Ma, and Cb&
* All soils subscripted "8" are divided between pine and cypress;
"9" divided among pine, cypress and prairie; "7" divided between
prairie and saw palmetto.
Other subscripts indicate the following: "2" - pine; "3" - cabbage
palm; "5" - cypress; "6" - coastal hammock
The entire study area can be divided into broad habitat categories as
well, such as coastal and upland, or terrestrial and aquatic. These broad
categories contain numerous community types. The upland terrestrial habitat
includes prairies, hammocks, swamp, marshes, and scrubland communities. The
coastal terrestrial habitats include the bay estuary, tidal creeks, tidal
pools, coastal river, and brackish pond communities.
These broader habitat descriptions, coupled with community typology and
more specific habitat classification within the community, help describe the
habitat and niche of individual plant species.
XIII-2
-------
Plant Communities of Southwest Florida
The communities studied were by no means inclusive of all the
communities represented in the area. Certain predominant communities were
studied, and some less extensive and less "important" communities were
excluded.
The coastal intertidal communities studied were mangrove swamp, saltgrass
marsh (Distichilis spicata), and the spartina marsh (Spartina spartinae). The
coastal intertidal communities not studied extensively were the black juncus
march (Juncus roemerianus), the eleocharis marsh (Eleocharis cellulosa), and
the coastal beach area. The coastal aquatic communities studied included
brackish ponds, bays, and inlets of the estuary. All of the area outside of
the inland waterways protected by mangrove islands was excluded from study --
for example, offshore reefs and the open sea.
The upland communities studied were cypress swamps, wet prairies, pine
and palm hammocks, ponds, sloughs, canals, and roadside ditches. The upland
communities excluded from study were the various freshwater marshes, saw grass
marsh (C1ad ium jamacensis), cattail marsh (Typha angustifolia), and spartina
marsh (Spartina bakerii). Scrub lands, containing scrub pine (Pinus clausa)
and the scrub oaks (Quercus myrtifolia and Quercus virginiana variety geminata),
were also excluded.
Plants Listed in the Community Descriptions
The plants listed in this description of the communities are restricted
phyletically to the vascular plants (Tracheophyta), which includes the ferns,
fern allies, and the seed plants. The list excludes all algae, some of which
are important and are mentioned later in the sections concerning the community
in which they are found. The estuary, for example, is particularly rich in
marine algae. Canals and ponds contain numerous green and blue-green algae
species, diatoms, desmids, and the like. Various blue-green algae are nitrogen
fixers in the aquatic and semi-aquatic communities. Certain blue-green algae
are producers of soil marl as a metabolic by-product. Marl soils are an
important and widespread group of soil types.
The list excludes mosses and liverworts (Bryophyta). Masses of moss
occur on the trunks of hardwoods and the sabal palm, as epiphytes, in the
cypress slough. Lichens also are excluded, but are common epiphytes in wooden
areas. The lichens are generally considered important nitrogen fixers. Fungi
are also excluded, even though they are unsurpassed as decomposers of cellulose
in both woody and grassy communities. Some fungi are important in mycorrhizal
associations, aiding the capture of phosphorus and other nutrients by the roots
of many higher plants.
The list of plants in the community is by no means complete, nor should
it be considered a checklist for even the plants in the phyla included. The
list merely names the more dominant plants and their importance with respect
to size, abundance, and where they can be found.
XIII-3
-------
Plant Composition
Cypress Strand
The cypress swamp or cypress strand is composed of cypress (Taxodium)
and various hardwood trees and shrubs, together with a variety of ferns,
orchids, bromeliads, woody vines, and herbs. The cypress swamp is influenced
by a long period of inundation. Only some of the freshwater marshes exceed
the cypress strand in length of hydroperiod. The soil and vegetation density
and type is highly variable in the cypress strand. In the deep interior of
the strand, the soil is rich in organic matter, the level of soil and bedrock
is relatively deep, and the vegetation lush, with a dense canopy. Nearer the
edge of the cypress strand, the soil becomes sandy to rocky and the vegetation
is more sparse. The bedrock and soil are elevated, thus having a shorter
period of inundation, resulting in less of the humus accumulation characteristic
of the interior strand, due perhaps to more frequent exposure to aerobic
conditions. The strand edge is occasionally visited by fire from the adjoining
fire sub-climax communities, the prairie and pinelands.
The Strand edge in many ways fits the classic description of an ecotone.
However, it is not clearly distinct from the interior cypress strand; rather,
there exists a gradual graduation of one into the other. Many of the smaller,
more isolated strands contain nothing that resembles a strand interior. Rather,
they consist of vegetation only characteristic of the Strand edge. The "Strand
edge" type of community includes a large part of the Big Cypress Swamp, perhaps
as much as one-half. This massiveness in area does not fit the classic concept
of an ecotone as a thin-lined border between two different communities.
Many of the small shrubs, herbs, and trees that are found along the
Strand edge cannot become established deeper in the Strand. Their seeds rarely
have time to germinate during the dry season before the deep Strand's floor is
inundated, drowning the seedlings. Many small trees, shrubs, ferns, and other
herbs do, however, become established on the elevated piles of debris on stumps,
logs, and at the base of trees in the deep Strand.
Many plants found in great abundance in the deep Strand may not occur at
all in the Strand edge or vice versa. For example, many of the ferns found in
the deep Strand do not occur at all along the edge of the Strand. The wax
myrtle (Myrica cerifera) is the most abundant shrub along the Strand edge, but
it is completely absent in the deep Strand. The willow (Salix caroliniana)
is most abundant in shallow strand ponds or depressions along marshy borders,
but it occurs only occasionally through the rest of the Strand.
The royal palm (Roystonea elata) is found in groves on strand ridges, but
is almost completely absent everywhere else in the Strand. Perhaps the only
plant found uniformly throughout the cypress strand is cypress itself. Even
the cypress show a great variability from the inner strand to the strand edge.
Many taxonomists consider the cypress as two distinct species -- the pond
cypress (Taxodium ascendens), inhabiting the outer strand, and bald cypress
(Taxodium distincum), inhabiting the inner strand.
A list of the plants commonly found in the cypress strand is presented in
Appendices AA, BB, CC, DD and EE. Abundance is based on the Strand as a whole,
which is by no means homogeneous. The identification of the substratum should
give the reader an idea of where the plant can be found in the Strand. Size and
XIII-4
-------
abundance is rated relative to other members of the particular group of plants
in each table. The species lists were accumulated via field collecting conducted
throughout the study period. Abundance values are based on gradient sampling
further discussed in Chapter XII.
Prairie
The prairie is characterized by low slender grasses, sedges, rushes, and
other herbs. It is basically flat, slightly elevated above the cypress strand
which often borders it. The increase in elevation results in a shorter hydro-
period which usually lasts several months, but may be completely absent some
years. The bedrock is very near the surface in the shallow phase soils. Rock
outcroppings are common. The thin soil layer over the bedrock is poor in humus
due to exposure to air and fire. The soil texture is 'dominated by marl and
sometimes fine sand. The shorter hydroperiod apparently is conducive to
frequent fires, which may burn the prairies as often as every other year.
Regular fire is essential to the maintenance of the prairie vegetation
and soil. Fire regularly kills tree seedlings which have invaded the grassland.
Fire removes the debris of dead plant leaves. This removal enhances light
penetration for both the herbs and the blue-green algae of the prairie soil.
The blue-green algae is responsible for the production of the soil marl. The
fire also burns most of the humus deposits on the soil surface. A humus
build-up would destroy the alkaline marl and change the entire nature of the
soil. In this way, a fire restricts the vegetation to those slender herbs
adapted to fire and maintains the soil type.
The soil survey maps list the prairie soil types as various depth phases
of Ochopee marl and fine sandy marl (types Oa, Ob, Od, Oe, and Of).
Appendix FF lists vegetation in the prairie (subdivided by monocots and
dicots). Some of the prairie plants are normally considered aquatic, and this
is mentioned under the column "Growth habitat". These aquatic plants occur
most frequently in shallow depressions of the prairie, or in prairies with a
relatively long hydroperiod. Like the cypress strand, prairies are not completely
homogeneous.
The Pine and Palm Associations
The pine and palm associations are woody plant communities consisting
primarily of fire-adapted trees, shrubs, and herbs. The fire sub-climax
community is characterized by the South Florida slash pine (Pinus elliottii
variety densa), the cabbage palm (Sabal gajLmetto), and the saw palmetto (Seranoa
repens) together with various other shrubs and small trees.
The soil for these pine and palm communities is characteristically sandy,
rocky, or fine sandy marl. These soils may be very wet or even inundated during
part of the year; but they are periodically very dry during the dry season, which
encourages fire, and may of itself retard hardwood tree growth.
The various soil types associated with the pine and palm communities are:
Arzell fine sand (Aa), Blanton fine sand (Ba), Charlotte fine sand (Ca), Copeland
fine sand (Cc, Cd, and Ce), Immokalee fine sand (la), Keri-Copeland complex (Ka),
Keri fine sand (Kb), Matmon loamy fine sand (Me), Ochopee fine sandy marl (shallow
phase) pine (Ob2), Pompano fine sand (Pa), Rockland (Ra), and Sunniland fine sand
XIII-5
-------
(Sc). The particular pine and palm communities that were investigated in this
study are those which are intermingled with the wet prairies and cypress strand.
The particular soil associated with these pine and palm communities is the
Ochopee fine sandy marl (pine) (Ob2).
Many of the herbs and shrubs in the pine and palm community are also
found in the wet prairie and the cypress strand. These plants exhibit a
variety of adaptations to fire. Some shrubs and small trees are adapted to
fire not by resisting it, but by being able to quickly reseed and complete
their life cycle immediately after the fire. The sea myrtle (Baccharis
halimifolia), myrsine (Myrsine quianensis), red bay (Persea borbonia), and
sweet bay (Magnolia virginiana) are all examples of shrubs and small trees
which regularly invade the pine areas after a fire. Wax myrtle (Myrica
ceritera), on the other hand, is only killed above ground by fire. Its roots
survive and are able to send up new shoots soon after the fire.
The herbs of the pine association are adapted to fire recovery in much
the same way as the prairie plants. In fact, many prairie herbs occur in pine
and palm associations. The grasses, sedges, and rushes frequently have
extensive underground stems (rhizomes) that survive a fire and send up new
shoots. Other grass plants re-establish after a fire by seed. Other herbs
have various underground parts (tubers, bulbs, corms, and rhizomes) that send
up new shoots after the fire. Still others re-establish by seed.
Fire has much to do with the vegetation of the pine and palm communities.
Generally, the less frequent the fire or the longer the time since the last fire,
the more shrubs and small trees dominate the understory vegetation. After a
fire, the slender herbs tend to dominate the ground-cover vegetation.
The pine, cabbage palm, and saw palmetto are well adapted for fire
survival. The pine has a thick insulating bark that protects the vital vascular
material and cambium of the trunk. The cabbage palm and saw palmetto both have
vascular material deeply imbedded throughout their trunks, as numerous vascular
bundles. The dispersion of vascular material throughout the stem is characteristic
of monocots, and it appears essential to the success of woody monocots like the
palm. In addition to diffuse vascular bundles, the saw palmetto's trunk is next
to and partly covered by the soil, which affords added protection against fire.
Appendices GG and HH list many of the plants commonly found in the pine
and palm associations. The tables are categorized by trees and shrubs, then
herbs. The tables give the general size range (even though this is highly
variable depending on fire frequency, particularly for woody plants), the
abundance (this also is strongly dependent on fire), and the substratum on
which the plant, lives.
Disturbed Swamplands and Prairies
Plant communities can be stressed and disturbed by many forces, caused by
nature or man. The most prevalent activities disturbing the swamplands and
prairies of the study area are roadbuilding, drainage, logging, burning, and
farming.
Roadbuilding and drainage are the most extensive activities disturbing
the natural communities of the study area. Drainage is essential to the genesis
of land suitable for residential and commercial development. Roadbuilding is
XIII-6
-------
needed to provide access to the developing areas.
Drainage in the swamplands shortens or eliminates the natural hydroperiod,
essential for maintaining the community's identity. The continued absence of
water on the swamp floor may permit the germination of hardwoods, such as red
maple (Acer rubrum). The absence of water may reduce the vitality of bald
cypress (Taxodium disticum) in the swamp.
The drained area cypress slough is marked by a noticeable disturbance
of the humus soil layer. There is preponderance of hardwood seedlings and
annual and perrenial weeds on the swamp floor; and in many cases a buildup
of undecomposed litter on the soil surface is noticeable from the presence of
leaves and stems that have not turned into humus.
Roadbuilding through wooded cypress strands involves the physical removal
of vegetation and the relocation of soil for the making of roadbeds and ditches.
The roadside is first invaded by annual and perrenial weeds, which are fast-
growing and quickly seed into the exposed soil. These weeds include Epatorium
(several species including those called dog-fennel), Ambrosia artemis i i f o 1 i a
(ragweed), Bidens pilosa (beggar ticks), Pluchea rosea, Phragmites australis
(reed), Arundo donax (giant reed), Rhynchelytrum repens (Natal grass),
Andropogon virginicus (broom sedge), Cyperus planifolius, Dichromena colorata,
Juncus biflorus, Juncus polycephalus, Boehmeria cylindria (button hemp),
Amaranthus sp., Urena lobata (Caesar weed), Sida sp., Lantana sp., and Hedyotis
coryombosa. After the annual and perrenial weeds, a few vines begin to invade,
e.g. Ipomoea alba (moon flower), Sarcostemma clausa, Cynanchum sp. (milkweed
vines), and Mikania batatifolia (hemp vine).
The cypress tree, a slow grower requiring rather special conditions for
seed germination and seedling survival, cannot quickly re-establish on the
roadsides. However, other woody plants with fast-growing and quick reseeding
capabilities can invade the roadside. The most conspicuous tree which invades
the roadside of the cypress strand is the Florida trema (Trema. micrantha). It
is a strikingly fast-growing tree which regularly invades disturbed soil left
opejn to sunlight and exposed to reseeding. Other woody plants invading the
cypress strand roads are Baccharis halimifolia (sea myrtle) and Ludwigia peruviana
(water primrose). The latter is particularly prevalent on poorly drained road-
sides with wet ditches.
Logging of swamps for cypress timber also has an effect on the vegetation
of the cypress strand. The cypress do not quickly re-establish after logging.
Hardwoods such as Salix caroliniana (willow), Acer rubrum (red maple),
Fraxinus caroliniana (pop ash), and Ficus aurea (Florida strangler fig) quickly
expand and multiply to fill the gap left b~y the removal of cypress. Stumps
and logging slash provide an excellent substratum for the germination and
growth of shrubs, trees, ferns and other herbs. The extreme abundance of ferns,
herbs, and small shrubs on debris-covered logs' and stumps is noticeable in
logged-over cypress swamps.
Burning of the cypress swamp is a rare occurrence. It happens only under
abnormally harsh and sustained hot, dry, and drought conditions. The cypress
are destroyed and do not recover, and the seedlings grow slowly. The first
trees to re-establish after a swamp fire are SaHx caroliniana (willow),
Fraxinus caroliniana (pop ash), and Acer rubrum (red maple). Burned areas of
a cypress swamp show a conspicuous absence of large cypress and an abundance of
XIII-7
-------
one or all of the hardwoods mentioned above, especially willow.
Freshwater Ditches and Canals
Freshwater roadside ditches and drainage canals are aquatic communities
of the uplands which might also be classified as disturbed communities. The
vegetation of these communities is highly variable, depending on water depth,
age, water velocity, turbidity, seasonal fluctuation of the water level and
bottom type.
Water depth and turbidity together determine the amount of light reaching
the bottom of the canal (shoreline vegetation contributes some additional
shading effect). Generally, the more light reaching the bottom, the denser is
the rooted aquatic vegetation for a given bottom type and current velocity.
Water depth and bottom type also affect the form of the rooted vegetation.
Generally emergent vegetation does not occur in water deeper than some minimum
depth, nor does it occur on a bottom too hard for suitable anchorage.
Swift-moving canals and ditches tend to have hard or shifting sandy bottoms,
The softer muds and silts are washed away. The combination of swift current
and poor anchorage eliminates many rooted aquatics.
Floating aquatics tend to inhabit deep, sluggish canals and ditches. The
depth discourages rooted aquatics, which compete with the floating plants.
However, if the current is swift the floating aquatics are washed downstream.
Seasonal fluctuation of the water level has the greatest effect on the
vegetation of shallow canals and ditches, and shallow margins of deeper canals
and ditches. These shallow water bottoms are often left exposed to the
atmosphere during the dry season. This exposure is fatal to most floating
aquatics and many rooted aquatics. However, some rooted aquatics actually
benefit from this exposure, e.g. Diodia virginiana (buttonweed), cannot flower
and produce seed in its emergent form.
The periodic stagnation of shallow roadside ditches and canals results
in very high temperatures during the warmer months. This high temperature may
stress both plant and animal life in the water by reduction of dissolved oxygen
and stepped-up respiratory metabolism.
Table XIII-2 lists the freshwater canals and ditches in the study area,
and Appendix II their aquatic plant associations.
Aquatic Communities of the Cypress Strand Lakes and Ponds
The cypress strand typically includes depressions which are filled with
water and form a pond or lake. The smaller, shallower depressions form shallow
ponds,, inhabited by rooted emergent and submergent aquatic plants. The larger,
deeper depressions form small lakes inhabited by floating aquatic vegetation
and sometimes floating islands. The lakes normally contain water even during
the driest of dry seasons, but the ponds usually dry up during the dry season.
Both communities are important to fishes as breeding grounds and a place to live
during droughts. Such ponds and lakes are often inhabited by alligators.
Alligators habitually dig out and deepen the ponds and lakes during the dry
season, thus aiding the survival of fish and fish-eating birds.
XIII-8
-------
Table XIII-2. Freshwater canals and ditches in the study area.
Station
Water Depth
Water Velocity
Seasonal Flux
Bottom
GAG canal system
(GAC) (above the
weir)
Janes Scenic
Drive (JD)
Tamiami Canal (TC)
(west of Royal Palm
Hammock)
Alligator Alley (AA)
Miscellaneous road-
side ditches and
canals (misc)
2-4 meters, narrow
margins except at
the weir
Usually less than
1% meters
1-4 meters, nar-
row margins
1-4 meters, mostly
narrow margins,
some broadened
margins
Variable up to 3
meters
Sluggish headwaters,
moderate downstream,
sluggish margins
Mostly sluggish,
seasonally stagnate
to moderate
Mostly sluggish
Mostly sluggish
Variable, mostly
sluggish
Water velocity variable,
water depth less variable
Most parts exposed dur-
ing dry season
Seasonally variable but
rarely exposed except the
margins
Seasonally variable mar-
gins, exposed during dry
season
Variable, mostly seasonal
exposure
mostly sandy
limestone,
muds over
limestone
Sandy, rocky
and organic
muds and
silts
Sandy, rocky
and organic
Sandy , rocky
and organic
Variable
The strand lake studied in the 'study site is referred to as Ballard's
Lake, The floating vegetation consisted mostly of Pistia stratiotes (water
lettuce), along with Eichhornia crasspies (water hyacinth), Lemna perpusilla
(duckweed), and Limnobium spongia (frogbit). The lake's margin consisted of
Zizaniopsis miliacea (giant cutgrass), Saururus cernuus (lizard's tail), and
Cephalanthus occidentalis (button bush). Ballard's Lake also had numerous
small floating islands, mostly less than a meter in diameter and about 10 to
20 centimeters thick. These islands were composed of a dark, well-decomposed
organic material. The vegetation on them consisted of Ludwigia leptocarpa
and L,. peruviana (water primrose), Flue he a purpurascens, Amaranthus sp., and
Cyperus sp.
These floating islands are not typical of all small strand lakes, e.g.
Corkscrew Swamp's water lettuce lake has few such floating islands. Floating
islands of this sort are described by Craighead (1971). He mentions their
distribution in the Everglades National Park and speculates on their origin.
Several strand ponds were observed in the study area. These ponds appear
to be depressions similar in depth to the sawgrass marsh, but much less extensive
and protected from fire and sawgrass invasion by the surrounding cypress swamp.
The emergent vegetation consists of Thalia geniculata (fire flag), and Sagittaria
lancifolia (arrowhead), and occasionally Canna flaccida. This herbaceous aquatic
zone is often fringed by Salix caroliniana (willow). The pond margin consists
of a few small emergent and submergent aquatics such as Juneus polycephalus
(rush), Sagittaria graminea (arrowhead), Bacopa caroliniana, Ludwigia repens,
and Pontideria lanceolata (pickeralweed). These ponds are often circled by a
line of Annona glabra (pond apple) and Fraxinus caroliniana (popash).
XIII-9
-------
Salt Marshes and Mangrove Swamps
-SWAMP HARDWOODS
AND CYPRESS
The coastal intertidal habitat consists of grassy communities (salt marshes)
and woody communities (mangrove swamp, coastal hammocks, and coastal beach). The
coastal hammocks and coastal beach were not studied, as they were not considered
extensive enough or important enough in the study area to warrant priority over
the mangrove swamp. Mangrove swamps are quite extensive throughout the coastal
habitat (Figures XIII-1 and 2).
The salt marsh types
studied most extensively
were the saltgrass marsh,
consisting of mostly
Distichlis spicata (salt-
grass) and Salicornia
bigelovii (glasswort)
along with scattered
Spartina spartinae, and
the cordgrass marsh,
consisting mostly of
Spartina spartinae (cord-
grass) along with
scattered patches of
Distichlis spicata
(salgtrass), Eleocharis
cellulosa (spike rush)
and Sesuvium portulacas-
trum (sea purslane).
Various other
saltmarsh communities
include: the black
juncus marsh (Juncus
roemerianus) , the spike
rush marsh (Eleocharis
cellulosa) , and the
saltwort marsh (Bat is
martima) (Craighead,
1971).
The saltgrass
marsh is character-
ized by moderate
salinity (20 ppt);
daily or twice daily
flooding with high tide; numerous marine invertebrates such as snails, mussels,
and polychetes; and a noticeable amount of periphytic green algae.
In contrast to the saltgrass marsh, the cordgrass marsh is characterized
by low salinity (5 ppt); few or no snails, mussels, or polychetes; irregular
inundation occurring only during very high tides and o-v^rland freshwater flow;
and little periphytic algae. In general, the marine influence in the cordgrass
marsh is greatly reduced by the miles of river, mangrove swamp, bays, and
mangrove islands separating the marsh from the Gulf of Mexico.
CYPRESS STRAND
GRASSY UNDERSTORY
Figure XIII-1. Distribution of communities in
the study area.
XIII-10
-------
This cordgrass marsh
is of a different species
and a completely different
habitat from the Spartina
alterniflora marsh which is
common from Central Florida
to Texas westward, and to
New England northward.
The Spartina
alterniflora marsh, in
contrast to the Spartina
spartinae marsh in South
Florida, tolerates high
salinity, wave action, and
tidal fluctuation. It is
located right on the shores
of the Atlantic and the Gulf
of Mexico. A small Spartina
alterniflora marsh is
reported to exist near Little
Marco Pass (near Marco
Island), but this marsh
type as a whole is rare in
the Ten Thousand Islands.
-CYPRESS STRAND
GRASSY UNDERSTORY
SWAMP HARDWOODS
AND CYPRESS
PINE
Figure XIII-2. Distribution of communities in
the Big Cypress Swamp (includes Fahkahatchee)
The mangrove swamps
and islands of South Florida
consist of four major tree
species: Rhizophora mangle
(red mangrove), Laguncularia
racemosa (white mangrove),
Avicennia nitida (black
mangrove) and Conocarpus
erecta (buttonwood). The
distribution of these species
is dependent on water depth and salinity. The red mangrove, with its extensive
prop root system and special reproductive fruits, can inhabit the deepest waters.
It is conspicuous along shorelines of bays, rivers, and islands. The black
mangrove is adapted to shallower water with its pneumatophores (protrusions of the
roots through the soil to aid root respiration). The black mangrove is conspicuous
interior to the red mangroves on shores of bays, rivers, and larger islands. It
is also present on the steeper, more elevated banks of rivers and canals.
The white mangrove, like the black, shows a pneumatophore adaptation. The
white usually inhabits sandy or shelly soil, in contrast with the rich organic
soil inhabited by the black mangrove. The white'mangrove is a fast grower and
quickly invades areas disturbed by storm or dredging.
Buttonwood is another woody plant associated with mangroves. Buttonwood
is not commonly found in areas of high salinity or high tidal influence.
Buttonwood has no special root adaptation for standing water, but it can inhabit
the more elevated coastal areas where salinity eliminates upland trees. Button-
wood occurs right up to the point of termination of salinity and tidal influence.
XIII-11
-------
Numerous salt-tolerant herbs are associated with mangroves, particularly
on elevated portions of the mangrove swamp. These herbs include Acrosticum
aureum (leather fern), Acrosticum danaeaefolium (leather fern), Fimbristylis
spathacea, Fimbristylis castanea, Yucca alpifolia (Spanish bayonet),
Hymenocallis latijcjlia (spider lily), Batis maritima (saltwort), Salicornja
bigelovii (glasswort), Suaeda linearis (sea blite), Alternanthera ramoissima
(chaff flower), Philozerus vermicularis (samphire), Iresine celosia (bloodleaf),
Irianthema portulacastrum, Sesuvium portulacastrum (sea pursland), Opuntia
compressa (pricklypear cactus), Opuntia stricta (pricklypear cactus), Iva
frutescens (marsh elder), Borrichia frutescens (sea daisie), and Borrichia
arborescens (sea oxeye). In addition to these ground-cover herbs, several
species of bromeliads (Bromeliaceae) are found as epiphytes in mangroves in
the moderate to lower salinity areas. Also three lianas are common among the
mangroves in the study area -- Hippocratea yolubilis, Rhabdadenia biflora
(rubber vine), and Ipomoea tuba (moon vine).
Aquatic Coastal Communities
The aquatic coastal communities in the study area comprise bays, inlets,
coastal rivers, brackish-water lakes and lagoons, and coastal canals. The
distribution of the various plants in the coastal aquatic communities is
strongly related to salinity, bottom type, water depth, and currents.
Benthic estuarine plants occur only in shallow waters with weak currents
and good light penetration. The rooted benthic vegetation cannot survive on
bottoms that are constantly shifting or experiencing a large amount of erosion
or sedimentation.
The bay estuary is characterized by rooted vegetation such as Thalassia
testudium (turtle grass), Halophila engelmannii (sea grass), Diplanthera
wrightii, and several species of marine algae. The bay estuary experiences
high salinities and a large degree of tidal fluctuation. It is separated from
the Gulf only by scattered mangrove islands. Coastal rivers contain rooted
benthic vegetation only on shallow, stable,, sandy bottoms. The depth and
current of the mainstream river produces a bottom unsuitable for benthic
vegetation. The river vegetation is highly variable, depending on salinity.
Near the bays, where salinity is highest, the vegetation is similar to that
found ii the bay. Diplanthera wrightii is particularly prevalent.
Brackish-water lakes and lagoons^ in the study area are located in the
upper reaches of the rivers. They are actually broad, shallow expanses of the
river bordered by mangrove and saltmarshes, principally Spartina spartinae
marsh and Distichilis spicata marsh. These brackish-water lakes and lagoons
are characterized by low to moderate salinity and reduced tidal influence.
The benthic vegetation of moderate salinity (10-20 ppt) lakes and lagoons
consist mostly of Ruppia maritima (widgeon grass). The lakes and lagoons farther
up the rivers with lower salinity (less than 5 ppt) consist of Chara sp. and
Najas marina (marine niad).
Coastal canals generally are lacking in benthic vegetation except along
the margins. The combination of depth, current, and boat traffic, reduce the
amount of light reaching the bottom and the stability of the bottom (increased
sedimentation and resuspension).
XIII-12
-------
XIV - MACROBENTHOS
INTRODUCTION
A study of macrobenthic invertebrates was undertaken with several views
in mind. First, numbers and kinds of invertebrate animals inhabiting an area
can be indicators of changes in the physical and chemical quality of the sub-
strate and surrounding water. Secondly, the benthic faunal community serves
as a food resource for numerous fish species, many of which are of important
commercial and sport value. Finally, results of such a study would serve as
background information for determining future changes in the benthic inverte-
brate resources.
METHODS
Both Fahkahatchee and Fahka Union Bays were sampled for macrobenthos dur-
ing March 28-30, June 20-21, and September 24-28, 1972. A stratified sampling
program design resulted in 18 sampling stations in Fahka Union Bay (six mud,
six sand, and six shell) and 26 stations in Fahkahatchee Bay (ten mud, ten
sand, and six shell). Because of the absence of large areas of sand and shell
substrates in Fahka Union Bay, it was necessary to duplicate samples at two
shell sites in order to obtain the desired number of samples. However, when
more than one sample was taken from one location (within a predetermined grid),
the boat was moved approximately 10 to 15, meters away from the first site.
The sampling station locations are presented in Figure XIV-1.
Two samples were collected and composited at each station by use of a
15.2 cm by 15.2 cm (6 in. by 6 in.) Ekman dredge, resulting in a total area
of 232 square centimeters (0.5 square feet). A slurry was made from each
composite sample contents and sieved through a No. 30 sieve. The organisms
and other material retained in the sieve were removed into a glass jar which
was subsequently filled with ethanol and a small amount of sodium bicarbonate.
The samples were forwarded to the Biological Services Branch, Surveillance and
Analysis Division, Environmental Protection Agency, in Athens, Georgia, for
identification and biomass determinations.
Decapod crustaceans collected in the monthly fish trawling samples at
regular stations in the estuary were sorted, identified, and tabulated. Dur-
ing the year, the collections numbered 24 otter and 23 surface trawls in Fahka
Union Bay; while in Fahkahatchee Bay, there were 67 otter and 69 surface trawls.
Freshwater decapods were obtained monthly in samples from a square-meter drop
frame at two sites in Fahkahatchee Strand. Some special collections were made
at estuarine locations not included in the routine fish sampling scheme.
Descriptions of collection gear, sampling procedure, and station locations are
given in Chapter XV.
RESULTS AND DISCUSSION
A taxonomic breakdown of benthic organisms found during the course of the
study is presented in Appendix JJ. Table XIV-1 summarizes the data on biomass
in the two bays. The data were analyzed by analysis of variance, using a log
transformation (Snedeson, 1956). Two-way ANOVA (neglecting interaction)
XIV-1
-------
N
EXPLANATION
M-3— STATION NUMBER
M -I SUBSTRATE
s LM-MUO
0 J 8-SAND
0-SHELL
Figure XIV-1. Station locations for macrobenthos study.
revealed a lack of significant (5-percent level) differences either by bay, by
date, or by substratum both within and between bays.
The (arithmetic) mean number of individuals captured per sampling, segre-
gated by bay and by substratum, is shown in Figures XIV-2 through 4. The only
significant differences (based on paired-sample t tests) reside in the obser-
vations on Isopoda. However, most of the difference is due to the June
sampling points. This coincided with an increase in the standing crop of
decapods as measured in the catches of the otter and surface trawls.
The relative numerical abundance of decapods from the bays and adjacent
areas is presented in Table XIV-2. The monthly occurrences of crustaceans are
listed in Table XIV-3.
During our investigation, the commercially important pink shrimp (Penaeus
duorarum) was the most abundant decapod crustacean in our collections. Pink
shrimp were extremely numerous in the marine grass beds in north Fahkahatchee
Bay. Peak catches were trawled during July and November, which indicated a
seasonal pattern of availability.
XIV-2
-------
Table XIV-1. Biomass of macrobenthos.
(geometric mean) , gm/m2.
Values as arithmetic mean ±1 S.E.
Mud Substrate
Sand Substrate
Shell Substrate
Fahkahatchee Bay
March - a
March - b
June - a
June - b
September - a
September - b
March - a
March - b
June - a
June - b
September - a
September - b
1.7810.38(1.42)
1.04+0.27(0.77)
2.59±0.53(2.09)
1.12*0.28(0.82)
3.4411.34(1.89)
1.22*0.40(0.78)
Fahka
2.3610.68(1.40)
1.38*0.38(0.84-)
5.31±1.03(4.?0)
2.8610.63(2.48)
2.19*0.4-7(1.93)
1.06*0.22(0.94')
3.02+0.74(2.18)
1.24+0.42(0.82)
5-77±3.50(2.56)
2.70*1.81(1.06
5.33*1.81(3-95)
2.23±0.84(1.45)
Union Bay
2.02*0.68(1.25)
1.01*0.33(0.60)
4.26±1.28(3.40)
1.83*0.4-6(1.57)
3.0811.75(1-74)
1.04*0.4-2(0.78)
10.7±3.09(8.11)
3.55*1-28(2.33)
2.34+0.52(2.33)
1.10i0.30(0.9l)
5.91+3.80(2.48)
2.13+1.38(0.92)
0.84+0.21(0.68)
0.42±0.10(0.36)
1.7810.52(1.37)
0.87±0.27(0.49)
1.4010.27(1.19)
0.5010.10(0.45)
a - Dry weight values
b - Ash-free dry weight
The mean number and biomass (wet weight) of pink shrimp and grass shrimps
(Palaemonetes vulgaris, P_. intermedius, and P_. pugio) per unit-of-effort are
compared in Figures XIV-5 and 6 respectively.
The decopods are important faunal components in estuarine food chains.
Numerous crustaceans—for instance, Palaemonetes intermedius and Rithropanopeus
harrisii—consume large quantities of vascular plant detritus and algae (Odum
and Heald, 1972). In turn, grass shrimps, pink shrimp, swimming crabs
(Portunidae), mud crabs (Xanthidae), and other decapods often are major food
items in the diets of sport and commercial fishes, including spotted seatrout
(Stewart, 1961); gray (mangrove) snapper (Croker, 1960); red drum (Yokel, 1966);
snook (Marshall, 1958); ladyfish; and silver perch.
XIV-3
-------
400r-
OJ
300-
a: 200
LJ
Q.
CO
CO
O
o;
o
100
MUD SUBSTRATE
FAHKAHATCHEE BAY
TANAIDACEA
MARINE WORMS
OTHER
J
M A M J J A S
MONTH
1,000
900
800
700
CM
in
ci
CO
eo
o:
o
600
500
400
300
200
100
FAHKA UNION BAY
M A M J J AS
MONTH
Figure XIV-2. Arithmetic mean for numbers of benthic macroinvertebrates
inhabiting mud substrates in April, June, and October, 1972, Fahka Union and
Fahkahatchee Bays.
XIV-4
-------
500 r
400
in
O
DC
UJ
a.
200
22
O 100
SAND SUBSTRATE
500
FAHKAHATCHEE BAY
FAHKA UNION BAY
MARINE WORMS-,
OTHER—i
MAMJJASO
MONTH
MAMJJASO
MONTH
Figure XIV-3. Arithmetic mean for numbers of benthlc macroinvertebrates
inhabiting sand substrates in April, June, and October, 1972, Fahka Union
and Fahkahatchee Bays.
SHELL SUBSTRATE
500
400
PJ
3OO
ta-
in
200
UJ
Q.
CO
tr
o
5 300
FAHKAHATCHEE BAY
MARINE WORMS-
OTHER-i
UJ
a.
CO
200
o
oc.
O (00
FAHKA UNION BAY
MARINE WORMS-
OTHER—,
MAMJJASO
MONTH
MAMJ JASO
MONTH
Figure XIV-4. AritfeABtic »ean for numbers of benthic macroinvertebrates
inhabiting shell M*ttr*tes in April, June, and October, 1972, Fahka and
Fahkahatchee
XIV-5
-------
Table XIV-2. Distribution and relative abundance of decapod crustaceans col-
lected in Ten Thousand Islands, Florida, 1972. Most individuals were taken
in association with monthly fish collections while employing otter and sur-
face trawls. Habitat designations were as follows: Fahka Union Bay (A);
Fahkahatchee Bay - West (B), - North (C), - East (D); other estuarine areas
(E); and freshwater sites (F).
Habitat
Family and Species ABC D E F
Penaeidae
Penaeus duorarum 355 357 1655 522
Palaemonidae
Periclimenes longicaudatus 2 5 246 14
Periclimenes americanus 3 9 42 6
Leander paulensis 32 8 81
Palaemonetes p_aludosus 24
Palaemonetes vulgaris 15 6 14 58
Palaemonetes intermedius 21 91 1430 162
Palaemonetes pugio 1 2 26 12
Alpheidae
ALpheus normanni 1 8
Alpheus heterochaelis 1 55
Hippolytidae
Thor floridanus 4
Hippolyte pleuracantha 45 422 134
Tozeuma carolinense 57 365 114
Astacidae
Procambarus alleni 514
Porcellanida^
Petrolisthes armatus 2 87
P or ce 11 ana s ay an a 2
Paguridae
Paguristes spp. 2 9 33
Pagurus longicarpus a a
Pagurus bonairensis a a a a a
Portunidae
Portunus sayi 226
Portunus gibbesii 272
Callinectes sapidus 27 22 12 24
Callinectes ornatus 6 14 11 25
Xanthidae
Menippe mercenaria 1 1
Rhithropanopeus harrisii 1
XIV-6
-------
Table XIV-2. Distribution and relative abundance of decapod crustaceans col-
lected in Ten Thousand Islands, Florida, 1972. Continued.
Family and Species
Habitat
C
D
Neopanope texana
Eurypanopeus depressus
Panopeus herbstii
Goneplacidae
Eucratopsis crassimanus
Grapsidae
Aratus pisonii
Ocypodidae
Uca rapax
Uca pugilator
Majidae
Libinia dubia
23 161
1
1
70
7
221
15
10
51
12
17
38
42
a Numerical data on Pagurus spp. were only recorded from a few
collections.
40
I
1-30
n
320
5
O
m
2 10
PINK SHRIMP
GRASS SHRIMP
J J A
MONTH
Figure XIV-5. Monthly numbers of Figure XIV-6. Mean monthly biomass, wet
pink shrimp (Penaeus duorarum)
and grass shrimp (Palaemonetes^
spp.) taken from combined catches
of surface and otter trawls in
Fahka Union and Fahkahatchee Bays,
1972.
weight, of pink shrimp (Penaeus
duorarum) and grass shrimp (Palae-
monetes spp.) taken from combined
catches of surface and otter trawls
in Fahka Union and Fahkahatchee Bays,
1972.
XIV-7
-------
Table XIV-3. Temporal distribution of decapod crustaceans collected in Ten
Thousand Islands, Florida, 1972. Species not taken in association with
monthly fish collections appear in parentheses.
Family and Species
Penaeidae
Penaeus duorarum
Palaemonidae
Periclimenes longicaudatus
Periclimenes americanus
Leander paulensis
Palaemonetes paludosus
Palaemonetes vulgaris
Palaemonetes intermedius
Palaemonetes pugio
Alpheidae
Alpheus heterochaelis
Alpheus normanni
Hippolytidae
Thor floridanus
Hippolyte pleuracantha
Tozeuma carolinense
Astacidae
Procambarus alleni
Porcellanidae
Petrolisthes armatus
Porcellana sayana
Paguridae
Paguristes spp.
Pagurus longicarpusa
Pagurus bonairensisa
Portunidae
Portunus sayi
Portunus gibbesii
Callinectes sapidus
Callinectes ornatus
Xanthidae
Menippe mercenaria
Rhithropanopeus harrisii
Neopanope texana
Eurypanopeus depressus
Panopeus herbstii
Goneplacidae
Eucratopsis crassimanus
Grapsidae
Aratus pisonii
Ocypodidae
Uca rapax
Uca pugi later
Majidae
Libinia dubia
Month
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
XXXXXXXXX XXX
X X X X XXX
X'X XX XXXX
X XXX
XX XXXX
XXXXXXXX X
XXXXXXXXX XXX
XX XX
XX XX
XXX
X
XXXXX XXX
XXXXXXXXX XXX
XXXX XXXXXX
XX XXX X X
(x)
X XXX
X
XXXXXXXXX XXX
X X
xx x
XXXXXXXX XXX
XXXXXX XXX
X
X
XXXXXXXXX XXX
xx x
X
(x)
(x)
(x) (x) (x)
(x) (x) (x)
XXXXXXXXX XXX
a
Data Incomplete
XIV-8
-------
XV - FISHES OF FAHKAHATCHEE STRAND AND TEN THOUSAND ISLANDS
INTRODUCTION
There have been few systematic surveys of the marine fishes in South-
western Florida. Historically, Lonnberg (1894) and Longley and Hildebrand
(1941) described the fish fauna of the southwest coasts and Florida Keys,
respectively. Evermann and Kendall (1900) prepared a checklist of Florida
fishes. In recent times, the most comprehensive surveys of fishes were
conducted by Tabb and Manning (1961) in Whitewater and Florida Bay and by
Roessler (1967) in Buttonwood Canal; these areas are in Everglades National
Park near Flamingo. Gunter and Hall (1965) investigated the fishes of the
Caloosahatchee River and adjacent estuaries near Fort Myers, Florida. The
fishes of the Charlotte Harbor area were studied by Wang and Raney (1971).
Various aspects of the life histories of single fish species have been
conducted in the Everglades area. Investigations by Croker (1960) on gray
snapper (Lu11anus griseus), Steward (1961) and Moffett (1961) on spotted sea-
trout (Cynosion nebulosus), Yokel (1966) on red drum (Sciaenops ocellata),
Waldinger (1968) on three species of mojarras (Gerreidae), and Volpe (1959)
and Marshall (1958) on snook (Centropomus undecimalis) have contributed
important information on the biology ,of these fishes. In addition, Odum and
Heald (1972) studied energy flow pathways of 45 fish species in Whitewater Bay
and North River, and Hunt (1953) reported on the food habits of the Florida gar
(Lepisosteus platyrhincus) and associated freshwater fishes in the Tamiami Canal.
Little information is available concerning the fishes of the Fahkahatchee
Strand and Ten Thousand Islands. The importance of these waters as aquatic
habitats for freshwater and estuarine fishes has not been documented. To help
fill this void in the information, we initiated fisheries investigations to
establish baseline data for the year 1972. The objectives of our studies were
(1) inventory of fishes for these two areas, (2) determine the seasonal
distribution and relative and absolute abundance of these fishes, and (3) assess
fish community structure as related to habitat and other aspects of environmental
conditions.
STANDING CROPS OF FISHES
Habitats, Sampling Procedures and Gear
Freshwater fish communities were sampled at two temporary waterways
(seasonally inundated borrow canals) in Fahkahatchee Strand (Stations A-^ and
in a drainage canal adjacent to the western edge of Fahkahatchee Strand (Station
B), in a tidal canal paralleling the Tamiami Trail (Station C), and in a natural
lake, locally known as Ballard's Lake (Station D) in Fahkahatchee Strand.
Estuarine fish communities were studied at three sites in Fahka Union Bay
(Stations E, F, and G), at three sites in Fahkahatchee Bay (Stations H, I, and J),
and in a tidal stream in north Fahkahatchee Bay (Station K). These stations are
depicted in Figure XV-1.
Several techniques were used to enclose or block the water columns. Each
was designed to obtain the maximum return from a given area. Because of the
shallow depths and intermittent periods of inundation, the gear for sampling
XV-1
-------
Figure XV-1. Study area and sampling sites in Fahkahatchee Strand and the Ten
Thousand Islands, Florida. Dots designate the stations for the standing
crops of fishes.
XV-2
-------
animal populations at the two temporary waterways in Fahkahatchee Strand (Stations
A-, and A~) consisted of a mr-frame and 5.0 percent emulsified rotenone, a fish
toxin. The frame was dropped over an undisturbed area, mud was stacked around
the outer bottom edges of the frame to prevent escapement, and emulsifiable
rotenone was added to produce a dosage of about 1.5 ppm in the entrapped water.
Following the removal of the aquatic vegetation in the m -area, dip nets with
square hoops were repeatedly swept back and forth until all the fishes, amphibians,
and crustaceans were collected. The top 2.5 cm of bottom debris was examined for
killifishes (Fundulus spp.) and crayfish (Procambarus alleni) that often burrowed
into the substrate.
Two nets were used to block off a 100 meter stretch of stream in the
eastern branch of the GAG Canal (Station B) and in the Tamiami Canal (Station C).
One block net was 30.5 m long and 1.8 m in depth with 6.4 mm-bar mesh netting,
while the other was 21.3 m long and 1.8 m in depth with 12.8 mm-bar mesh. Nets
were stretched across the canal and anchored securely to the mud bottoms with
steel fence posts. The studyareas received emulsifiable rotenone at a concen-
tration of approximately 1.5 ppm. Immobilized fishes that surfaced were dipped
from the water. An estimate of the number of dead fishes that sank to the
o
bottom was obtained from the recovery of individuals in m -wire baskets that
were randomly spaced throughout the study section. Later, to aid in the
detoxification of the rotenone, potassium permanganate was dispersed downstream.
Emulsifiable rotenone was applied to the waters of Ballard's Lake (Station
D) at a concentration of 2.0 ppm. Floating fishes were dipped from the lake on
the day of poisoning, as well as on the following day. Due to natural processes
of biological decay, we assumed that the majority of dead fishes that initially
sank to the bottom had floated to the surface by the second day.
Estimates of the standing crops of estuarine fishes in Fahka Union (Stations
E, F, and G) and Fahkahatchee Bays (Stations H, I, and J) were obtained by
enclosing the water column with a 6.4 mm-bar mesh net that was 274.3 m (900 feet)
long and 3.0 m deep. A rectangular sampling area was formed using four corner
posts. The net was then eased off the bow of the boat around the corner posts.
To prevent escapement, the lead line was pushed into the soft substrates.^- '
Fishes were recovered by repeatedly seining through the enclosed areas
until the yield was negligible. The number of seine hauls per study site varied
from three to seven. The few fishes that were not recovered generally were small
individuals such as gobies (Gobiidae), pipefishes (Syngnathidae), and anchovies
(Engraulidae).
Block nets and rotenone were used to sample the fish community of the tidal
stream (Station K) that emptied into Fahkahatchee Bay. The downstream net was
set across the mouth of the stream early in the morning on a high tide. As the
waters receded, the net prevented escapement of fishes from the stream and into
the bay. On the following low ebb tide, the shallow upstream segment of the
study area was blocked off, and rotenone was applied to the water to give a
(1) Equipment for this phase of the study was provided on loan by the Florida
Freshwater Fish and Game Commission. The study was conducted as a cooperative
effort with the University of Florida, Center for Aquatic Sciences.
XV-3
-------
concentration of 2.0 ppm.
The temporary waterways in the Strand (Stations A, and A,-,) were sampled
monthly, while the other study sites were only sampled once.
Fishes were separated by species, counted, measured, and weighed.
Generally, large fishes were processed in the field, while small individuals
were preserved in a 10 percent solution of formalin and returned to the
laboratory for examination.
All length measurements in this chapter were total length (anterior
extremity with the jaws closed to the tip of the compressed caudal lobes) to
the nearest 1.0 mm, unless stated otherwise. Wet weights (biomass) of the
fishes were taken with either a beam balance (20-kg capacity) or a dial type
balance (1,600-gm capacity) to the nearest 1.0 or 0.1 gm, respectively.
The names of fishes and their phylogenetic order follow the recommendations
of the American Fisheries Society (Bailey, _et al., 1970). A tabulation of the
common and scientific names of fishes collected from Fahkahatchee Strand and the
Ten Thousand Islands appears in Appendix KK.
Results
Temporary Freshwater Habitats
At the two freshwater sites (Stations A^ and A£), the major biological
components were fishes (12 species), amphibians (2 species), macrocrustaceans
(2 species), and aquatic vascular plants (10 species) (Table XV-1).
The standing crops of fishes by biomass (wet weight) ranged from 1.1 to
172.8 gm/m , excluding the months of May and June when the study sites were dry.
The mean biomass for the year was 44.3 gm/m . The mean number of fishes per m
varied from 3.5 to 575.0 individuals. The lowest observed values by biomass and
by number occurred during the wet season (summer and fall), while the highest
values were recorded during the dry season (winter and spring).
The freshwater fishes in the temporary waterways belonged to the following
families: Poeciliidae (livebearers), Cyprinodontidae (killifishes), Centrarchidae
(sunfishes), and Ictaluridae (catfishes). The species composition in decreasing
order of abundance was composed of mosquitofish (Gambusia affinis), flagfish
(Jordanella floridae), sailfin molly (Poecilia latipinna), least killifish
(Heterandria formosa), marsh killifish (Fundulus confluentus), yellow bullhead
(Ictalurus natalis), spotted sunfish (Lepomis punctatus), bluefin killifish
(Lucania goodei), Everglades pygmy sunfish (Elassoma evergladei), warmouth
(Lepomis gulosus), bluegill (L. macrochirus), and golden topminnow (Fundulus
chrysotus). Mosquitofish, flagfish, sailfin molly, and least killifish were
the major fishes and represented 61.2, 20.9, 6.6 and 4.0 percent, respectively,
of the total catch. On occasion, the Florida gar (Lepisosteus platyrhincus)
was sighted in the waterways, but was never taken in the sampling gear.
The abundance of fishes in Fahkahatchee Strand was directly related to the
seasonal water patterns. During the year, the fewest number of fishes per unit
area were present in July, after the initial flooding of the waterways which
dispersed the fish communities. Throughout the remainder of the wet season,
XV-4
-------
Table XV-1. Standing crops of flora and fauna obtained monthly from two
temporary waterways in Fahkahatchee Strand, Florida, 1972.
Month
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Biological Category
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Sampling sites dried-up
Sampling sites dried-up
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Total
Number
of Species
9
1
4
7
1
1
4
8
2
2
6
8
2
4
2
2
1
8
4
2
1
4
8
1
2
3
6
2
2
4
Mean
Number of
Individuals
/m2
187.0
35.5
24.5
2.5
23.5
239.0
1.5
32.0
575.0
14.0
3.5
24.5
86.5
7.5
1.5
11.0
18.0
2.5
19.0
28.0
2.0
18.5
Mean
Biomass
/m2
(gm)
77.6
75.3
3,810.1
26.0
2.9
72.6
4,226.2
101.1
0.8
113.6
3,495.5
172.8
54.2
4,284.2
1.1
30.3
44.2
2,664.1
1.5
1.2
6.3
2,696.5
18.4
4.0
25.3
2,726.5
10.0
7.1
49.3
5,420.8
XV-5
-------
Table XV-1. Standing crops of flora and fauna obtained monthly from two
temporary waterways in Fahkahatchee Strand, Florida, 1972. (Continued)
Month Biological Category
Nov. Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Dec. Fishes
Other vertebrates
Macrocrustaceans
Aquatic plants
Total
Numb er
of Species
4
2
2
3
7
1
2
3
Mean
Number of
Individuals
/m2
9.0
2.0
13.0
27.5
0.5
14.0
Mean
Biomass
/m2
(gm)
3.6
3.0
36.9
2,640.5
30.8
1.0
40.2
2,990.1
there was a steady increase in the size of the community. By January, the
water had ceased to flow at the study sites with a gradual drawdown in the
water levels during the remainder of the dry season. The progressive loss of
surface water concentrated the fishes; they reached their maximum density in
April (Table XV-1). The study sites were dry during May and June. Changes in
the density of various fishes and other aquatic animals during the wet and dry
seasons are tabulated for selected months at the study site (Station A-^) along
Janes Scenic Drive (Table XV-2).
The impact of the seasonal fluctuation of fishes on the ecology of the
Strand perhaps can be best illustrated by extrapolating the mr-sample data to
the numbers and biomass of fishes per hectare (1 ha = 10,000 m ="2.5 acres).
The standing crops of fishes were equivalent to an average of 132,000
individuals /ha with a biomass of 69,200 gm/ha during the wet season (June
through November) and an average of 2,106,000 fishes with a biomass of 817,000
gm/ha during the dry season (December through May) .
Other aquatic vertebrates included the peninsular newt (Diemictylus
viridescens) and various tadpoles (mostly Rana spp.). The largest catch of
juvenile frogs occurred in July and was composed of recently hatched individuals.
During the other months, both the tadpoles and peninsular newt were sporadic in
occurrence and were a minor component of the animal populations in the waterways .
The macrocrustaceans in the monthly collections were dominated by crayfish
(Procambarus alleni) with a sparse occurrence of the freshwater shrimp (Palaemonetes
paludosus) . In fact, crayfish were the major faunal organism in the waterways,
exceeding the biomass of fishes in 80.0 percent of the monthly collections
(Table XV-1). Mean monthly biomass of crayfish varied from 6.3 to 112.5 gm/m^,
with the peak abundance of 112.7 gm/m in March.
Two aquatic vascular plants, creeping water primrose (Ludwigia repens) at
Station A-^ and Diodia yirginiana at Station
dominated the plant communities
in the waterways. Eight additional plant species, listed in Table XV-1, occurred
XV-6
-------
Table XV-2. Seasonal changes in the density of aquatic animals in a temporary
waterway adjacent to Janes Scenic Drive, Fahkahatchee Strand, Florida,
1972.
Wet Season
Dry Season
Species
Aug.
Number/m2
Oct.
Number/in2
Feb.
Apr.
Number/in Number/nr
Ictalurus natalis
Fundulus confluentus
Jordanella floridae
Lucania goodei
Gambusia af finis
Heterandria formosa
Poecilia latipinna
Elassoma everglade!
Lepomis macro chirus
Lepomis punctatus
Diemictylus viridescens
Rana spp.
Palaemonetes paludosus
Procambarus alleni
2
Total number/m
Total biomass (gm/m )
2
2
2
1
1
8
16
14.8
3
1
34
1
8
3
16
66
89.4
1
25
22
8
1
4
1
1
44
107
133.7
1
4
101
5
199
13
9
1
3
25
361
257.1
in the monthly collections. No relationship was demonstrated between the
density of plants and the abundance of fishes or crayfish.
Additional data on percent ash-free dry weight and percent organic nitrogen
for animal and plant species occurring in the temporary waterways of Fahkahatchee
Strand are provided in Appendix LL. Similar values for a few fishes from canals
adjacent to the strand are included in the same appendix.
Canal Habitats
The standing crops of fishes by number and by biomass (wet weight) in the
two GAG and Tamiami canals were strikingly dissimilar (Tables XV-3 and 4). In
the Tamiami Canal, 34,999 fishes (38.9 individuals/m2) with a total biomass of
153.6 kg (170.7 gm/m2) were recovered in the tidal canal. The fishes from the
sampling site in the GAG Canal totalled 3,559 individuals (1.8 fish/m2) with a
biomass of 12.8 kg (6.6 gm/m ). Thus, on a m2-basis, the number of individuals
and biomass of fishes in the Tamiami Canal exceeded the estimate from the GAG
Canal by 95.4 percent and 96.1 percent, respectively.
Dissimilar species and biomass compositions were demonstrated between the
canals. The dominant fishes in the Tamiami Canal were mosquitofish and flag-
fish which represented, respectively, 59.6 percent and 25.8 percent of all the
fishes (Table XV-4). Dominance was reversed when examining the biomass
composition. By weight, the Florida gar comprised 80.5 percent of the total
biomass, but represented only 0.5 percent of the total number of fishes.
Conversely, the biomass of mosquitofish constituted only 2.3 percent of the total
catch.
XV-7
-------
Table XV-3. Standing crop of fishes in the most eastern GAG Canal, Fahkahatchee
Strand, Florida, February, 1972. Surface area of the sampling site was
1,950 m2.
Species
Lepisosteus platyrhincus
Notemigonus crysoleucas
Notropis maculatus
Erimyzon sucetta
Ictalurus natalis
Fundulus chrysotus
Fundulus
-------
Table XV-4. Standing crop of fishes in the Tamiami Canal at Big Cypress Bend,
Florida, March, 1972. Surface area of the sampling site was 900 m2.
Species
Lepisosteus platyrhincus
Notemigonus crysoleucas
Notropis maculatus
Erimyzon sucetta
Ictalurus natalis
Strongylura notata
Cyprinodon variegatus
Fundulus chrysotus
Fundulus confluentus
Jordanella floridae
Lucania goodei
Lucania parva
Gambusia affinis
Heterandria formosa
Poecilia latipinna
Menidia berylllna
Centropomus undecimalis
Elassoma everglade!
Lepomis gulosus
Lepomis macrochirus
Lepomis marginatus
Lepomis microlophus
Lepomis punctatus
Micropterus salmoides
Diapterus plumieri
Mugi 1 cephalus
Microgobius gulosus
Trinectes maculatus
Number
189
2
2
1
1
1
2
1
1,156
9,020
278
974
20,868
253
1,539
1
4
4
46
75
12
17
515
7
11
17
2
1
Size Range
(mm)
235-631
62-106
48-53
74
70
288
22-28
23
22-41
19-36
20-35
20-36
15-39
12-21
18-35
42
330-641
19-28
29-62 •
29-115
45-71
89-180
32-145
238-376
92-291
315-383
47-63
30
Species Biomass
Composition Biomass Composition
(%) (gm) (%)
0.54
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
3.30
25.77
0.79
2.79
59.62
0.72
4.40
<0.01
0.01
0.01
0.13
0.21
0.03
0.05
1.47
0.02
0.03
0.05
<0.01
<0.01
123,649.7
9.0
1.7
3.5
3.2
35.5
1.0
0.3
671.2
2,880.4
86.3
250.1
3,550.7
19.9
700.3
0.5
5,289.0
1.5
1,241.0
161.0
36.0
508.0
1,101.0
3,171.0
2,694.0
7,573.8
2.2
1.5
80.48
<0.01
<0.01
<0.01
<0.01
0.02
<0.01
<0.01
0.44
1.87
0.06
0.16
2.31
<0.01
0.46
<0.01
3.44
<0.01
0.81
0.10
0.02
0.33
0.72
2.06
1.75
4.93
<0.01
<0.01
Biomass/m2
(gm)
137.389
0.010
0.002
0.004
0.004
0.039
0.001
<0,001
0,746
3.200
0.096
0.278
3.945
0.022
0.778
<0.001
5.877
0.002
1.379
0.179
0.040
0.564
1.223
3.523
2.993
8.415
0.002
0.002
Total
34,999
153,643.3
170.715
Another major difference in the two canals was attributed to the water
type. The source of fresh water in the GAG (new) Canal was runoff (primarily
during the wet season) and by the discharge of ground water (throughout the
year). Thus, flowing water, although reduced at the time of sampling on
February 28, was present throughout the year.
The Tamiami Canal in contrast lies in an east-west direction parallel to
U. S. Hwy 41, and perpendicular to Fahkahatchee Strand. Ground water, surface
water, and tidal waters mix in the canal and ultimately flow by way of numerous
bridges to the saltmarshes and by a north-south interceptor canal near S.R. 29
to the estuary below. Waters in the canal are generally sluggish and have high
salinity variations depending upon the hydroperiod and tidal cycle. On March 2,
the salinity at the time of sampling near Big Cypress Bend was 3.0 ppt. The
salinity at the sampling sites in the Tamiami Canal steadily increased during
the dry season, and in 1972, reached a high of 27.7 ppt in May. Following the
onset of the rainy season late in June, freshwater conditions returned to this
waterway for the remainder of the year.
XV-9
-------
The main dissimilarities between the GAG and Tamiami Canals were attributed
to the time of construction and to annual, as well as to semi-diurnal, fluctua-
tions in the salinity levels of Tamiami Canal. Both of these factors affected
the abundance and species composition of the fish communities. Twenty-three
species of freshwater fishes were identified from the GAG Canal, whereas 21
freshwater species, plus an additional 6 estuarine species, were taken in the
Tamiami Canal. Species of estuarine fishes included snook, striped mullet
(Mugil cephalus), striped mojarra (Piapteras plumieri), sheepshead minnow
(Cyprinodon variegatus), clown goby (Microgobius gulosus), tidewater silverside
(Menidia beryllina), and hog choker (Trinectes maculatus) (Table XV-4). By
weight, these saltwater fishes represented 10.2 percent of the total catch.
The given canal data represents the only reported information on standing
crops of fishes from the Tamiami and GAG Canals. The fish communities of other
Everglades canals (Miami Canal and South New River Canal) were investigated by
Clugston (1962) during the dry season, eight months after the occurrence of a
summer fish kill. Clugston reported biomass estimates of 19.4, 37.0 and 104.5
gm/m^ from three canal sites. His mean biomass value of 53.6 gm/m was eight
times greater than the biomass of fishes from the GAG Canal and three times less
than the biomass of fishes in the Tamiami Canal. Near Miami, the composition of
fishes in the Tamiami Canal, based on angling and net collections (Hunt, 1953),
was similar to the fish species, excluding the saltwater fishes, in the Tamiami
Canal at Big Cypress Bend.
The two major piscivores in the man-made canal systems of south Florida
are Florida gar and bowfin (Amia calva). Clugston (1962) reported that bowfin
represented 96.4 (18.7 gm/m^) and 97.2 percent (101.6 gm/nr) of the biomass
composition at two sites in the Miami Canal, while the Florida gar constituted
91.4 percent (33.8 gm/m^) of total biomass at a single site in South New River
Canal. O'Connell (1958), as cited by Hunt (1960), used electro-fishing gear
during a rough fish removal program in 12.8 km (8 mile) section of the Tamiami
Canal. By weight, Florida gar and bowfin populations represented approximately
20.0 and 60.0 gm/m , respectively, but the Florida gar occurred in greater
numbers than the bowfin. In the present study, the biomass of Florida gar in
the Tamiami Canal at Big Cypress Bend was 137.4 gm/m2- which greatly exceeded the
estimates from other canals in South Florida.
Review of Table XV-4 will indicate that an overwhelming portion of the
total reported standing crop biomass is comprised of piscivores. At first
glance, this may seem to be an impossible situation if the classical concept
of a trophic pyramid is invisioned. However, Tamiami Canal is not a closed
system. The reported estimates of standing crop (biomass) fail to take into
account recruitment to the system of forage animals from adjoining marshlands
and waterways. Also to be considered is the role of the macroinvertebrate
community in the diet of the predators. Finally, in light that the data
represents only one sampling period of the year, the reported standing crop
biomass could be a reflection of a seasonal concentration of predators in the
canal.
Strand Lake Habitat
The fish population of a natural lake (Station P or Ballard's Lake) was
sampled in early April (Figure XV-1). Ballard's Lake is a shallow depression
with a surface area of 4,069 m and a mean water depth of 1.2 m at the time of
sampling. Floating aquatic plants, mainly water lettuce (Pistia stratiotes)
XV-10
-------
and duckweed (Lemna perpusilla) intermixed with small floating islands (1.0 to
2.0 m in diameter) covered approximately 70.0 percent of the water surface.
Stands of giant cutgrass (Zizaniopsis millacea) and button bush (Cepha1anthus
occidentalis) characterized the boggy shoreline. The main source of water
for this lake was surface runoff. Estimates of fish populations were conservative
because of the dense stands of floating aquatics, which hampered the recovery of
fishes.
The standing crop of fishes in Ballard's Lake consisted of 2,483 individuals
(0.6 fish/m^) with a total biomass of 57.4 kg (14.1 gm/m^) distributed among 16
species (Table XV-5). Mosquitofish and yellow bullhead (Ictalurus natalis) were
the most abundant species, comprising 57.7 percent and 33.1 percent, respectively,
of the total number. The biomass composition was dominated by yellow bullhead
(58.7 percent), Florida gar (24.1 percent), bowfin (7.4 percent), and brown
bullhead (Ictalurus nebulosus) 5.3 percent).
Table XV-5. Standing crops of fishes and amphibians from a natural lake in
Fahkahatchee Strand, Florida, April, 1973. Surface area of the sampling
site was 4,069 m^.
Species Biomass
Size Range Composition Biomass Composition Biomass/m2
Number (mm) (%) (gm) (%) (gm)
Fishes
Amia calva 4 460-505 0.16 4,268.0
Lepisosteus platyrhincus 24 198-642 0.97 13,860.6
Notemigonus crysoleucas 12 97-134 0.48 196.9
Ictalurus natalis 823 69-332 33.14 33,748.4
Ictalurus nebulosus 74 89-348 3.00 3,024.7
Fundulus chrysotus 1 35 0.04 0.4
Fundulus confluentus 9 35-52 0.36 13.4
Jordanella floridae 5 26-47 0.20 5.9
Lucania goodei 7 24-34 0.28 2.5
Gambusia affinis 1,432 14-46 57.67 328.6
Heterandria formosa 5 16-27 0.20 0.7
Poecilia latipinna 5 33-60 0.20
Lepomis gulosus 24 90-181 0.97 1,128.0
Lepomis macrochirus 29 71-148 1.17
Lepomis microlophus 1 103 ^ 0.04
Lepomis punctatus 2£ 50-103 ' 1.13 259.8
Subtotal 2,483 57,452.1
0.
12.
128.
579.
22,
7.42
24.12
0.34
58.74
5.26
<0.01
0.02
0.01
<0.01
0.57
<0.01
0.02
1.96
1.01
0.04
0.45
1.049
3.406
0.048
8.294
0.743
<0.001
0.003
0.001
<0.001
0.081
<0.001
0.003
0.277
0.142
0.006
0.064
14.119
Amphibians
Amphiuma means
Siren lacertina
Subtotal
Total
1 529
22 208-630
23
2,506
4.35 435.2 6.96
95.65 5,819.2 93.04
6,254.5
63,706.6
0.107
1.430
1.537
15.656
XV-11
-------
Two large salamanders, greater siren (Siren lacertina) and amphiuma
(Amphiuma means), were present in the lake (Table XV-5). This was the only
habitat in which these two species were encountered during the year. Although
only 23 individuals were caught, their total biomass of 6.2 kg represented 1.5
gm/m2. When combined with the fish data, the standing crops of aquatic
vertebrates in Ballard's Lake by weight represented 15.7 gm/m . Fishes and
amphibians served as a potential food supply for a reproducing population of
alligators (Alligator mississippensis) inhabiting the lake.
Bay Habitats
The standing crops of estuarine fishes (combined total catch estimate) in
Fahka Union Bay consisted of 686 fishes (0.1 fish/m2) with a biomass of 33.8 kg
(5.2 gm/m2), while the combined estimate in Fahkahatchee Bay totalled 1,958
individuals (0.4 fish/m2) with a biomass o£ 25.9 kg (5.3 gm/m2) (Tables XV-6 and
7). On an areal basis, the biomass of fishes in Fahkahatchee Bay exceeded the
estimates from Fahka Union Bay at two of three sampling sites. However, the
mean estimate of biomass from both bays was essentially identical.
A single biomass value, unless accompanied by analysis of the various
species of the community is often misleading. In these two bays, the species
compositions were strikingly different. In Fahka Union Bay, over half the fishes
in the total catch were mojarras (Eucinostomus spp.) with lesser numbers of
yellowfin menhaden (Brevoortia smithi) (15.2 percent), code goby (Gobiosoma
robusturn) (4.4 percent), and Gulf pipefish (Syngnathus scovelli) (3.8 percent),
plus an assortment of other fishes (Table XV-6). In Fahkahatchee Bay, the dominant
species of fishes were composed of pinfish (Lagodon rhomboides) (55.8 percent),
silver perch (Bairdiella chysura) (12.8 percent), majorras (12.3 percent), code
goby (5.2 percent), and Gulf pipefish (2.5 percent). Also, more species of
fishes were recorded from Fahkahatchee Bay (33) than from Fahka Union Bay (26)
during this survey.
Compositions by biomass were quite dissimilar between the bays. Over half
of the total biomass in Fahka Union Bay was dominated by the Atlantic stingray
(Dasyatis sabina) which represented only eight individuals. In Fahkahatchee Bay,
the major component of the biomass (50.0 percent) was composed of pinfish,
numbering 1,092 individuals. Other major components of the biomass in Fahka
Union Bay were ladyfish (Elops saurus) (7.7 percent), striped mullet, and sea
catfish (Arius felis). The last two species each represented 7.4 percent of the
total biomass. In Fahkahatchee Bay, the abundant pinfish were followed by
Atlantic stingray (19.3 percent), striped mullet (8.4 percent), and silver perch
(4.6 percent). In addition, popular sport and food fishes, such as snook,
spotted seatrout, and gray or mangrove snapper were represented in the sample
catches from both bays (Tables XV-5 and 6).
Initially, we had intended to analyze the data from the enclosure net by
the DeLury regression method (Lagler, 1956). This technique provides an estimate
of the size of the population based on the catch per unit-of-effort. However,
DeLury's method is only considered valid if the following three conditions are
met: closed population, constant unit-of-effort, and equal catchability of
fishes. Although we fulfilled two of the conditions, the data were not amenable
to DeLury's method, since some fish species, such as Atlantic stingray and
striped mullet, exhibited a differential response to the gear. Also, at two
study sites, the catch per seine haul steadily declined, as expected, during
the first three or four tows, but suddenly increased during the last couple of
XV-12
-------
Table XV-6. Standing crops of fishes from three sites in Fahka Union Bay,
Florida, September, 1972.
Station E Station F
Area - 1,840 m2 Area - 2,788 m2
Species
Dasyatis sabina
Elops saurus
Brevoortia smithi
Anchoa hepsetus
Synodus foetans
Arius felis
Hyporhamphus unifasciatus
Syngnathus scovelli
Centropomus undecimalis
Lutianus griseus
Diapterus plumieri
Eucinostomus argenteus
Eucinostomus gula
Eucinostomus spp.
^Archosargus probatocephalus
Lagodon rhomboides
Bairdiella chrysura
Cynoscion nebulosus
Mugil cephalus
Gobiosoma robustum
Scomberomorus maculatus
Prionotus sp.
Paralichthys albigutta
Achirus lineatus
Symphurus plaeiusa
Sphoeroides nephelus
Total
Mean biomass (gm/m2)
Number Biomass Number
(gm)
1 840.4 2
2 421.4 8
30
3
10
2
6
2 11.9
173
4
1
32
1
1
3 1.7 2
4
8 1,275.4 279
0.7
Biomass
(gm)
1,027.4
2,073.8
82.4
3.9
2,390.7
14.3
0.8
504.5
0.8
0.7
5.1
175.4
0.7
0.2
3.0
6.283.7
2.2
Station G
Area - 1,840 m2
Number Biomass
(gm)
5
1
104
3
1
20
4
4
4
49
179
2
3
13
4
1
1
1
399
17,293.7
109.0
679.8
44.1
95.7
5.0
1,568.5
1,076.3
803.9
196.9
992.6
1.5
634.6
139.8
2,518.6
75.8
4.3
37.0
26,277.1
14.3
Number
8
11
104
30
6
11
2
26
4
4
4
49
181
175
3
13
4
1
4
32
1
1
1
6
4
1
686
Combined Estimate
Total Area - 6,468 m2
Species Biomass
Composition Biomass Composition
(%) (gm) m
1.17
1.60
15.16
4.37
0.87
1.60
0.29
3.79
0.58
0.58
0.58
7.14
26.38
25.51
0.44
1.89
0.58
0.14
0.58
4.66
0.14
0.14
0.14
0.87
0.58
0.14
19,161.5
2,604.2
679.8
82.4
48.0
2,486.4
14.3
5.8
1,568.5
1,076.3
803.9
196.9
1,004.5
506.0
634.6
139.8
0.8
0.7
2,518.6
5.1
175.4
0.7
75.8
6.2
3.0
37.0
33,836.2
56.63
7.70
2.01
0.24
0.14
7.35
0.04
0.02
4.64
3.18
2.38
0.58
2.97
1.49
1.88
0.41
<0.01
<0.01
7.44
<0.01
0.52
<0.01
0.22
0.02
<0.01
0.11
Biomass/m2
(gm)
2.962
0.403
0.105
0.013
0.007
0.384
0.002
0.001
0.242
0.166
0.124
0.030
0.155
0.078
0.098
0.022
<0.001
<0.001
0.389
<0.001
0.027
<0.001
0.012
0.001
<0.001
0.006
5.2
Table XV-7. Standing crops of fishes from three sites in Fahkahatchee Bay,
Florida, September, 1972.
Station H
Area - 1,208 m2
Species
Dasyatis sabina
Myrophis punctatus
Brevoortia smithi
Harengula pensacolae
Opisthonema oglinum
Anchoa mitchilli
Synodus foetans
Opsanus beta
Hyporhamphus unifasciatus
Strongylura notata
Strongylura timucu
Syngnathus scovelli
Centropomus undecimalis
Oligoplites saurus
Luc janus griseus
Diapterus plumieri
Eucinostomus argenteus
Eucinostomus gula
Eucinostomus spp.
Orthopristis chrysoptera
Archosargus £robatocej)halus
Lagodon rhomboides
Bairdiella chrysura
Cynoscion nebulosus
Nugil cephalus
Chasmodes saburrae
Gobiosoma bosci
Gobiosoma robustum
Microgobius gulosus
Etropus crossotus
Achirus lineatus
Symphurus plagiusa
Sphoeroides nephelus
Total
Mean hiomass (gm/m2)
Number
2
1
4
18
1
12
54
4
18
10
781
83
10
1
1
1
1,001
Biomass
(gm)
15.5
5.2
15.9
6.3
285.6
42.0
240.0
0.2
323.5
45.9
9,336.9
760.2
196.3
0.3
0.1
0.3
11,274.
9.
2
.3
Station I Station J
Area - 1,717 m2 Area - 1,947 m2
Number Biomass Number
(gm)
1
2
6 35.0
2 0.5 3
12
1 392.9 39
1
2
2
31
1
1
2 332.0 3
4
13
3 16.8 126
29
3 15.2
8
76 1,415.4 235
2 20.6 166
12
3
4
1
12 3.6 89
5
2
1
52
2 177.3
109 2,409.3 848
1.4
Biomass
(gm)
5, 000.0s
7.9
2.4
96.2
194.6
7.1
21.6
92.3
11.5
494.0
1.2
39.5
464.8
30.6
558.9
24.0
66.8
2,197.4
418.4
230.9
2,184.0
1.8
0.4
18.0
1.2
7.8
0.9
43.3
12,217.5
6.3
Number
1
2
2
1
10
5
12
40
1
2
2
49
1
1
6
4
25
183
33
21
18
1,092
251
22
3
4
1
102
6
2
2
52
2
1,958
Combined
Total Area
Estimate
- 4,868 m2
Species Biomass
Composition Biomass Composition
(%) (gm) (%)
0.05
0.10
0.10
0.05
0.51
0.25
0.61
2.04
0.05
0.10
0.10
2.50
0.05
0.05
0.30
0.20
1.28
9.34
1.68
1.07
0.92
55.77
12.82
1.12
0.15
0.20
0.05
5.21
0.30
0.10
0.10
2.66
0.10
5,000.0a
7.9
15.5
5.2
50.9
2.9
96.2
587.5
7.1
21.6
92.3
17.8
494.0
1.2
657.1
464.8
72.6
815.7
24.2
338.7
112.7
12,949.7
1,199.2
427.2
2,184.0
1.8
0.4
21.9
1.3
7.8
1.2
43.3
177.3
25,901.
0
19.30
0.03
0.05
0.02
0.19
0.01
0.37
2.26
0.02
0.08
0.35
0.06
1.90
<0.01
2.53
1.79
0.28
3.14
0.09
1.30
0.43
49.99
4.62
1.64
8.43
<0.01
<0.01
0.08
<0.01
0.03
<0.01
0.16
0.68
Biomass /m2
(gm)
1.027
0.002
0.003
0.001
0.010
0.001
0.020
0.121
0.001
0.004
0.019
0.004
0.010
<0.001
0.135
0.094
0.015
0.168
0.005
0.070
0.023
2.660
0.246
0.088
0.447
<0.001
<0.001
0.004
<0.01
0.002
<0.001
0.009
0.036
5.3
Estimated weight, observed only.
XV-13
-------
tows. This reversal in the catch was attributed in part to the earlier removal
of large quantities of vegetation which initially hampered the seining operation.
Thus, the standing crop of fishes were analyzed on the basis of the actual total
catch.
The distribution of bottom vegetation at several study sites influenced
the composition of fishes. In Fahka Union Bay, the benthic vegetation was
composed of sparse stands of Diplanthera wrightii at Station F and a moderate
amount of red algae, such as Gracillaria spp., at Station G, while Station E
was devoid of vegetation. Benthic vegetation was more abundant in Fahkahatchee
Bay. Mixed stands of Diplanthera - Thalassia beds and various algae were
prominent at Station H, and thick stands of marine algae covered the bottom at
Station J, while moderate amounts of algae were present at Station I. The
largest number of benthic fishes, such as pinfish, silver perch, and pig fish
(Orthopristis chrysoptera) were taken in the Diplanthera - Thalassia beds at
Station H. Other fishes such as lady fish, anchovies (Anchoa spp.), needlefishes
(Strongylura spp.), snook, and clupeids are pelagic species that feed primarily
in the water column, and thus, bottom types generally have little effect on
their distribution.
In summary, the fish communities differed between the two bays in several
respects: (1) there were four times as many fishes per unit area in Fahkahatchee
Bay than in Fahka Union Bay even though the mean biomass totals were similar; ,
(2) a greater number of species were present in Fahkahatchee Bay as opposed to
the other bay; and (3) the four dominant species by weight contributed over
70.0 percent of the total biomass in each bay. However, in Fahkahatchee Bay,
this biomass was distributed among 1,347 individuals, while in Fahka Union Bay,
a similar percentage of the biomass was attributed to 34 individuals, mostly
adults .
Tidal Stream Habitat
The standing crop of fishes in a tidal stream (Station K) contiguous with
north Fahkahatchee Bay was composed of 27 species. Numerically, the mo j arras
(Eucinostomus %ula and JE. argenteus) were dominant species representing 81.0
percent of the total fish communities. Other abundant species were bay anchovy
(Anchoa mitchilli) (9.0 percent), juvenile sheepshead (Archosargus probatocephalus )
(4.7 percent), pinfish (1.4 percent), striped mojarra (Diapterus plumleri)
(1.1 percent), and striped mullet (1.1 percent).
The biomass composition was predominantly striped mullet and mojarras
(Eucinostomus spp.) which represented 42.0 and 40.2 percent, respectively, of
the total catch. Other major representatives in decreasing order by weight were
sheepshead, striped mojarra, sea catfish, redfin needlefish (Strongylura notata) ,
and timucu (j> . timucu) (Table XV-8) .
Red mangrove trees lined the shoreline of this natural stream while the
surrounding terrain was densely forested. The area adjacent to the stream was
flooded during periods of high tides. It was not possible to obtain an estimate
of the area of inundation from aerial photographs because of the thick canopy.
The standing crop of fishes from the tidal stream represented 98.9
(Table XV-8). This value was an overestimate and not comparable with other
habitats, since an unknown portion of the fish community had moved from the
flooded lowlands to the stream in ebbing waters. Nevertheless, the fishes
XV- 14
-------
Table XV-8. Standing crop of fishes from a tidal stream in north Fahkahatchee
Bay, October, 1972. Surface area of the sampling site was 734 m^ on, a low
ebb tide.
Species
Myrophis punctatus
Anchoa mitchilli
Arius felis
Ops anus beta
Strongylura marina
Strongylura notata
Strongylura timucu
Cyprinodon variegatus
Lucania p_arva
Centropomus undecimalis
Caranx hippos
Oligqplites saurus
Lut j anus griseus
Diapterus pjumieri
Eucinostomus argenteus
Eucinostomus gula
Eucinostomus spp.
Archosargus probatocephalus
Lagodon rhomb oides
Cynoscion nebulosus
Menti cirrus littoral is
Scianops ocellata
Mugil cephalus
Bathygobius sopor at or
Gobionellas shufeldti
Gobiosoma robustum
Lophogobius cyprinoides
Microgobius gulosus
Total
Number
1
473
2
2
1
7
20
2
1
6
1
13
7
58
423
3,544
282
246
73
19
1
1
56
2
3
1
1
3
5,249
Size Range
(mm)
278
14-65
337-370
68-120
380
195-410
118-431
15-20
26
66-138
62
53-91
62-184
50-323
28-97
45-99
13-39
62-380
68-91
77-212
91
249
155-503
76-101
40-57
20
49
21-27
Species Biomass
Composition Biomass Composition
(%) (8™) (%)
0.02
9.01
0.04
0.04
0.02
0.13
0.38
0.04
0.02
0.11
0.02
0.25
0.13
1.10
8.06
67.52
5.37
4.69
1.39
0.36
0.02
0.02
1.07
0.04
0.06
0.02
0.02
0.06
10.5
332.4
780.0
31.1
75.2
729.4
528.6
1.0 -
0.9
31.9
4.2
27.3
365.3
1,732.5
1,782.0
24,705.4
24.4
5,716.7
615.5
532.6
5.1
150.0
27,683.4
21.5
1.8
0.5
1.0
1.2
65,891.4
0.02
0.50
1.18
0.05
0.11
1.11
0.80
<0.01
<0.01
0.05
<0.01
0.04
0.55
2.63
2.70
37.49
0.04
8.68
0.93
0.81
0.08
0.23
42.01
0.03
<0.01
<0.01
<0.01
<0.01
Biomass/m2
(gin)
0.014
0.453
1.063
0.042
0.102
0.994
0.720
0.001
0.001
0.043
0.006
0.037
0.498
2.360
2.428
33.658
0.033
7.788
0.838
0.726
0.007
0.204
37.716
0.029
'0.002
0.001
0.001
0.002
89 . 770
recovered in this survey indicated that tidal streams serve as an important
habitat for numerous sport and commercial species, such as spotted seatrout,
gray snapper, snook, crevalle jack (Caranx hipposp, sheepshead, striped mullet,
pinfish, and striped mojarra.
Discussion
The temporary waterways (shallow borrow canals) that served as study
sites in Fahkahatchee Strand were analogous in many respects to other freshwater
habitats, such as prairie ponds, sloughs, swamp depressions, marsh areas, lakes,
and alligator ponds. Many of these upland aquatic habitats were initially
inundated during the rainy season. The accumulation of surface runoff in the
early summer serves as a mechanism of dispersal, opening up new feeding and
spawning areas for poecilids, cyprinodonts, centrarchids, and other fishes.
With the commencement of the dry season in late fall, the surface waters
gradually recede and the fishes retreat from the shallow margins of the wetlands
XV-15
-------
and collect in the remaining pockets of water. Permanent aquatic areas, such
as deeper ponds, lakes, and canals that remain inundated throughout the dry
season, function as reservoirs for the fish populations of the Strand. Thus,
the distribution and abundance of the fishes were directly related to the
cyclic nature of the hydroperiod.
Food chain interrelationships of freshwater fishes play an important role
in the ecology of the Strand. Fishes incorporate both plant and animal material
into tissue and as a group, serve as an intermediate link between primary
producers (plants) and the tertiary consumers (higher predators). Sailfin molly
and flagfish are herbivores which feed upon green and bluegreen algae, diatoms,
vascular plant material, and detritus with lesser amounts of rotifers, protozoans
and ostracods (Hunt, 1953 and Odum and Heald, 1972). Omnivorous fishes, such as
mosquitofish, least killifish, and bluefin killifish, feed on a mixed diet of
algae, some phytoplankton, copepods, oligocheates, rotifers, mites, amphipods,
snails, and other invertebrates including immature insects. Carnivorous fishes,
which were generally more numerous in the permanent aquatic habitats (Strand
lakes and adjacent canals), consisted of spotted sunfish, redear sunfish,
warmouth, largemouth bass, and Florida gar. Forage fishes and generally the
larger invertebrates (mayfly and dragonfly nymphs, crayfish, ostracods, snails
and freshwater shrimp) form the bulk of their diet (Hunt, 1953).
Fishes and macrocrustaceans are important food items in the diet of other
vertebrate animals such as birds, raccoon (Procyon lotor), opossum (Didelphis
virginiana), turtles, snakes, and alligators. Numerous wading birds, e.g.,
great blue heron (Ardea herodius), little blue heron (Florida caerula), wood
stork (Mycteria americana), American egret (Casmerodius albus), white ibis
(Eudocimus albus), and others, were often observed foraging in shallow waters
for fishes and crayfish. The winter nesting season of wood stork, herons,
ibises, and egrets occurs concurrently with the lowered water levels and the
resultant concentration of forage organisms during the dry season. The breeding
success of many of these birds is related to the availability and abundance of
an adequate piscivorous diet. For example, a nesting colony of wood stork,
composed of 12,000 adults and their altricial young, consumed more than 1.2
million kg of small forage fishes during the two months of the breeding season
(Kahl, 1962, 1964 quoted from Odum, 1971).
Alligators subsist on a variable diet that is either directly or indirectly
dependent upon the aquatic environment. Immature alligators (less than 152.5 cm
in length) from an Everglades canal fed mainly on invertebrates (crayfish,
snails, and insects) and lesser quantities of fishes (Focarty and Albury, 1967).
Larger alligators prefer vertebrate animals, as Mcllhenny (1934) found gar,
herons, egrets, and snakes in the stomachs of alligators collected near bird
rookeries on Avery Island, Louisiana. Species of fishes in alligators' diets
reported by these authors and others included flagfish, mosquitofish, large-
mouth bass, other sunfishes (Lepomis spp.), seminole killifish (Fundulus
seminolis), bowfin, and gar.
RELATIVE ABUNDANCE OF ESTUARINE FISHES
Habitats, Sampling Gear, and Procedures
Twenty-two fish collections were obtained monthly at 14 regular stations
from January through December, 1972, in the estuarine waters of the Ten Thousand
XV-16
-------
Islands (Figure XV-2). As a measure of relative abundance, these collections
comprised the monthly standard unit-of-effort. Sampling was usually conducted
during the first week of each month at the time of the new moon when maximum
tides prevailed.
Several ecological classifications have been proposed for estuarine
environments, many of which are only appropriate for specific areas. To
compare zones, the estuary was divided into three somewhat arbitrary physical
divisions (lower reaches, middle reaches, and upper reaches) which follow a
modification of the terminology proposed by Carriker (1967).
In the Ten Thousand Islands, the lower reaches of the estuary referred
primarily to the outer beaches in the surf zone, as well as to the shallow
nearshore waters and numerous inlets between the islands. This zone was
characterized by sand and shell bottoms, pronounced tidal currents, and
salinities generally above 30.0 ppt. Benthic vegetation varied from none to
dense stands of marine algae. Middle reaches of the estuary were represented
by bays, channels, and tidal streams with thick stands of red mangrove
(Rhizophora mangle) lining the shorelines. Typical features of this zone were
soft mud and mud-sand bottoms with varying amounts of marine algae (Gracillaria
spp.) and Thalassia - Diplanthera beds. Upper reaches of the estuary were
characterized by soft muddy bottoms, currents ranging from slow to rapid, and
usually low salinities in a range of 0.1 to 10.0 ppt during the rainy season.
Bottom vegetation, if present, was usually restricted to sparse stands of
filamentous green algae. This zone included streams, canals, and marsh lands
that receive freshwater runoff from the uplands but nevertheless were subjected
to tidal influences.
The estuary in the Ten Thousand Islands is a rather narrow belt, which
extends from the ocean front to the salt marshes in the extreme upper reaches,
an average distance of 8.0 km (5.0 miles). The upper reaches are delimited by
man-made barriers, in most instances, such as roadbeds (U. S. Hwy 41) and
stream weirs. The major source of fresh water enters the estuary through either
drainage canals (Tamiami and GAG Canals) or altered stream beds (Barren River
and Blackwater River Canals). Natural freshwater runoff enters the estuary from
the wet prairies and marshes south of U. S. Hwy 41.
At the regular stations, sampling gear consisted of two bag seines, an
otter trawl or try net, and a surface trawl0 One bag seine was 30.5 m long
and 1.8 m deep with 0.64 cm-bar mesh, while the other was 15.2 m X 1.2 m with
0.64 cm-bar mesh in the wings and 0.3 cm-bar mesh ii the bag. The otter trawl,
constructed of 2.54 cm-bar mesh netting, was 3.05 m wide with a 0.64 cm-bar
mesh liner in the codend. The surface trawl, which was towed by two boats, was
6.5 m wide and 0.8 m deep with a codend length of 6.0 m; the nylon netting was
composed of 0.64 cm-bar mesh.
A standardized sampling procedure was followed each month. Surface and
otter trawls were towed in a straight line for a 5-minute period at the sampling
sites. Surface trawling usually took place on a rising tide, while otter
trawling was conducted during periods of high water. At the seining sites, the
seine was spread parallel to the shoreline and hauled perpendicular to the
shore. Seining at the regular stations on the Gulf beaches and in the bays
was conducted during flood tides, while fishes were collected with seines at
the two stations in the upper reaches in ebbing waters.
XV-17
-------
Figure XV-2. Study area and sampling station in Ten Thousand Islands, Florida.
Dots mark the regular seine stations (Numbers 1-6), surface trawl stations
(Numbers 7-10), and otter trawl stations (Numbers 11-14).
XV-18
-------
During 1972, there were a total of 259 regular fish collections consisting
of 72 seine hauls, 91 otter trawl tows, and 92 surface trawl tows. Ideally,
the monthly standard unit-of-effort was composed of 22 collections represented
by single hauls at each of six seine stations and duplicate tows at four otter
trawl, as well as at four surface trawl stations. Because of either mechanical
problems or damaged gear, five otter trawl and four surface trawl collections
were missed during the year.
Study sites were spread throughout the estuary to represent a variety
of habitats (Figure XV-2). In the lower reaches, the beach seining sites were
at Panther Key (Station 1) and at Round Key (Station 2) facing the open Gulf
of Mexico. Seining sites in the middle reaches were located over a soft mud
bottom on the south side of a small island in Fahka Union Bay (Station 3) and
at the mouth of the Fahkahatchee River in Fahkahatchee Bay over a broken shell
substrate (Station 4). In the upper reaches, seining sites consisted of a
small embayment at the mouth of a shallow tidal stream with a soft mud bottom
on the east side of Fahka Union Canal (Station 5) and at the end of a short
canal off Henderson Creek near East Naples (Station 6). Surface trawl sites
(Stations 7 through 10) and otter trawl sites (Stations 11 through 14) were
in the middle reaches of the estuary in Fahka Union and Fahkahatchee Bays.
Data on fishes were also obtained during preliminary surveys from October
through December, 1971, from other studies (standing crops of fishes), and from
special collections at other estuarine sites during 1972, such as Tiger Key and
Camp Lulu Key in the lower reaches, Gaskin Bay and East Bay in the middle reaches,
and numerous brackish water streams in the upper reaches. These random collec-
tions were employed to supplement both the ecological data and the list ;of fish
species from the Ten Thousand Islands, but they were not used in calculating the
monthly totals from the regular stations.
Results
Relative Abundance by Number
A total of 273,270 individuals, distributed among 96 species and 41
families, were collected by surface trawling, otter trawling, and seining from
the 22 regular monthly collections during 1972 (Table XV-9). An additional 38
species, which were collected during the preliminary survey in 1971 and in
special studies at other estuarine and freshwater locations, are reported later
in the results.
The eight most numerous fish species, in decreasing order of abundance,
were bay anchovy, yellowfin menhaden (Brevoortia smithi), scaled sardine
(Harengula pensacolae), striped anchovy (Anchoa hepsetus), pinfish (Lagodon
rhomboides), silver perch (Bairdiella chrysura), and silver jenny (Eucinostomus
gula). The bay anchovy was the dominant fish species in the estuary accounting
for 70 percent of the total catch. Excluding the anchovies, the other above
species represented 64 percent of the year's total catch (Table XV-10).
In an investigation of the fish fauna of Charlotte Harbor (approximately
120.0 km northwest of our study site in the Ten Thousand Islands), Wang and
Raney (1971) reported that the five most abundant species, other than the
anchovies, were pinfish, silver perch, silver jenny, pigfish (Orthopristis
chrysoptera), and sand seatrout (Cynoscion arenarius). They employed an otter
trawl in collecting most of these fishes which accounted for the preponderance
XV-19
-------
Table XV-9. Distribution of the total catch at the regular stations by species,
number, and biomass from Ten Thousand Islands, 1972.
Sampling Gear
Seines
Subtotal
Surface trawl
Subtotal
Otter trawl
Subtotal
Station
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Number of
Species
47
43
37
23
34
38
79
30
29
41
40
55
22
29
30
27
45
Number of
Individuals
2,729
7,983
2,497
22,309
81,037
72,339
188,894
22,282
15,721
24,487
17,192
79,682
615
717
1,029
2,333
4,694
Biomass
(gm)
45,060.9
132,085.0
47,152.2
16,455.1
29,612.9
28,221.4
298,587.5
16,338.5
15,796.8
41,627.4
24,870.7
98,653.4
1,328.2
3,325.2
3,679.5
8,769.9
17,102.8
Total
96
273,270
414,343.7
of benthic species comprising 76.1 percent (calculated) of their total catch.
They did not adequately sample the pelagic component of the fish fauna since
species such as yellowfin menhaden, scaled sardine, needlefishes, and
silversides were seldom collected.
Numerically, fishes representing seven families dominated the monthly
collections. The most abundant fish families were anchovies (Engraulidae),
herrings (Clupeidae), needlefishes (Belonidae), mojarras (Gerridae), porgies
(Sparidae), drums (Sciaenidae), and mullets (Mugilidae). Excluding the
anchovies, these families contributed from 48 to 97 percent of the standard
monthly catch during 1972 (Table XV-11).
XV-20
-------
Table XV-10. Most abundant species of fishes as percent of the monthly catch
from regular stations in the Ten Thousand Islands, 1972, excluding Anchoa
spp.
Species Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Opisthonema oglinum 0.3 0.3 0.4 22.9 1.5 9.7 7.2 0.1 0.3
Brevoortia smithi 1.7 93.5 94.6 87,0 91.1 1.4 5.0 38.2 0.4 <0.1 0.5 15.1
Harengula pensacolae <0.1 <0.1 0.8 3.6 1.2 50.2 22.2 28.1 12.7
Eucinostomus gula 20.8 0.4 0.5 0.1 0.2 1.9 2.5 2.6 5.8 7.6 10.8 10.5
Lagodon rhomboides 0.2 0.6 0.2 1.2 2.5 57.8 5.0 1.0 0.8 4.9 0.7 4.6
Bairdiella chrysura 0.1 0.7 <0.1 0.2 0.6 12.6 9.6 5.5 5.4 25.0 1.7 1.5
Total 23.1 95.1 95.2 88.5 94.6 54.9 48.4 47.8 68.3 65.0 41.9 44.5
Table XV-11. Most abundant families of fishes by number as percent of monthly
catch from the regular stations in Ten Thousand Islands, 1972, excluding
Engraulidae.
Family Jan, Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Clupeidae 2.0 93.3 94.6 87.0 91.4 2.6 31.5 40.9 60.3 29.5 28.7 28.1
Belonidae 4.2 0.3 0.3 0.4 0.2 0.4 2.5 2.1 1.4 3.4 3.2 12.5
Gerridae 40.2 2.7 1.4 3.4 2.2 10.7 20.9 12.8 16.5 24.0 34.3 23.2
Sparidae 0.5 0.6 0.2 1.3 2.3 38.2 5.2 1.1 0.8 5.1 0.9 0,6
Sciaenidae 0.3 0.9 <0.1 0.3 0.7 13.0 12.7 4.2 3.9 23.6 2,0 1.7
Mugilidae 0.3 0.1 0.2 0.6 0.5 1.2 1.8 5.8 2.6 0.5 7.9 15.4
Total 47.5 97.9 96.8 93.0 97.1 66.1 74.6 66.9 85.5 85.9 77.0 79.5
Fishing devices are notoriously selective by species and by size.
Extrinsic factors (type of gear and sampling procedure) and intrinsic factors
(behavior of the fishes, size, and diurnal sampling times) are responsible for
variations in the catches (Pope, 1966). The surface trawling gear was
extremely effective for smaller-sized pelagic fishes, such as clupeids, engraulids,
strongylurids, and atherinids, while the otter trawl captured larger numbers of
bottom dwelling fishes, such as sciaenids, gerreids, and sparids. Thus, both
trawls complemented each other in the open bay areas. Seines were generally
non-selective since they caught juvenile and adult fishes which inhabited the
water column as well as the benthos.
XV- 21
-------
The most abundant catches by gear and month were usually associated with
the seasonal dominance of one or two species. The preponderance of juvenile
yellowfin menhaden in March and of anchovies in June and October contributed
to the peak surface trawl catches in those months (Figure XV-3). The largest
mean catch with the otter trawl occurred in June and was due to an abrupt
increase in the numbers of pinfish and pigfish in the collections. Peak
catches at the seining stations were attributed to anchovies in January and to
the scaled sardine at the beach stations in November (Figure XV-4).
1,200
01,000
800
600
400
UJ
200
SURFACE TRAWL
OTTER TRAWL
J J
MONTH
S
N
Figure XV-3. Distribution of the mean catches by number at the regular surface
and otter trawling stations.
IO.OOO
i 8,000
1 6/300
4.0OO
2,000
-.0.0
2.0 ~
M
J J
MONTH
Figure XV-4. Distribution of the mean catches by number and by biomass at the
regular seining stations.
XV-22
-------
By species, the largest total number in the standard monthly collections
were recorded in June (61) and November (54) (Figure XV-5). Similar increases
in the number of species occurred in the same two months with all three pieces
of gear. The fewest species in the monthly catches occurred in March and April.
6O
50
CO
UJ
CO
u.
O 30
cc
20
10
TOTAL NUMBER
SEINE
SURFACE TRAWL
OTTER TRAWL
M
M
J J
MONTH
Figure XV-5. Number of fish species from the collections at the regular
stations by type of gear and by total number per month.
• Relative Abundance by Biomass
Biomass (wet weight) values are usually not determined in surveys of
estuarine fishes (Tabb and Manning, 1961; Koessler, 1970; Gunter and Hall,
1965: Wang and Raney, 1972; and Springer and Woodburn, 1960), although
McFarland (1963) maintained that the biomass of fishes yielded more ecological
information than the numerical data. The current study represents the first
time that measures of biomass were employed to assess the relative abundance
of fishes in Southwestern Florida. The total biomass of fishes collected at
the regular stations in the Ten Thousand Islands was 414.3 kg (Table XV-9).
By biomass, the most abundant fishes in our'monthly collections, in
decreasing order, were bay anchovy, southern stingray (Dasyatis americana),
scaled sardine, fantail mullet (Mugil trichodon). striped mullet (M_. cephalus),
silver perch, and yellowfin menhaden. The monthly distribution of these
dominant fishes in the regular catch, except for the southern stingray which
only occurred sporadically, are presented in Table XV-12. These six species by
biomass comprised 49.7 percent of the total catch. The distribution of the
total mean monthly catches by biomass for the seine, surface trawl, and otter
trawl are presented in Figures XV-4 and 6.
XV-23
-------
Table XV-12. Most abundant species of fishes by biomass as percent of the
monthly catch from regular stations in the Ten Thousand Islands, 1972.
Species
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Brevoortia smithy <0.1 7.4 20.7 2.2 15.2 0.1 0.6 20.7 0.2 <0.1 0.3 0.1
Harengula pensacolae <0.1 <0.1 0.1 0.5 0.5 31.4 16.4 17.9 1.6
Anchoa mitchilli 58.6 28.2 19.4 3.0 7.8 10.2 16.1 8.1 16.7 19.7 15.8 22.9
Bairdiella chrysura <0.1 0.2 <0.1 <0.1 0.3 7.4 0.8 0.6 2.3 24.9 0.5 0.5
Mugil cephalus 2.4 26.6 1.9 29.7 10.0 6.3 0.9 14.4 2.6
Mugil trichodon 0.2 0.1 5.2 5.5 0.5 1.0 1.6 14.7 14.0 0.8 17.0 57.0
Total
59.0 38.3 43.4
5.8 50.5 20.7 49.3 70.9 62.8 62.8 65.9 64.7
4.4
o>
O
1.2
0.8
O
QQ
0.4
<
LJ
SURFACE TRAWL
OTTER TRAWL
M
MJJASOND
MONTH
Figure X7-6. Distribution of the mean catches by biomass at the regular surface
and otter trawling stations.
XV-24
-------
The monthly catch data at the regular stations indicate that the species
of anchovies (Engraulidae) also dominated the fish biomass. In fact, the same
seven families of fishes dominated the fish fauna of the estuaries as measured
by biomass, as well as by numbers (Tables XV-13 and 11), respectively. On a
biomass basis, these families comprised 23 to 96 percent of the monthly catch.
Table XV-13. Most abundant families of fishes by biomass as percent of monthly
catch from regular stations in Ten Thousand Islands, 1972.
Family Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Clupeidae 0.1 7.4 20.7 2.2 15.3 0.2 1.9 21.4 36.1 19.9 18.2 1.8
Engraulidae 60.2 28.3 19.9 3.3 10.2 12.8 16.3 9.3 17.5 20.4 16.6 23.1
Belonidae 6.2 2.8 9.0 1.6 1.4 2.2 2.5 4.2 2.4 8.0 4.7 15.9
Gerridae 10.6 2.9 6.6 6.7 15.7 14.6 2.3 8.5 4.5 5.6 4.3 5.3
Sparidae 8.8 6.8 10.4 0.2 1.4 20.4 2.5 1.0 0.6 8.0 10.5 4.1
Sciaenidae 0.2 38.4 <0.1 5.5 OiS 8.1 1.1 0.8 2.5 25.1 0.5 5.8
Mugilidae 0.3 2.9 3.2 3.5 27.2 2.9 32.1 25.5 20.4 2.0 31.4 39.6
Total 86.4 89.5 69.9 23.0 72.0 61.2 58.7 70.7 84.0 89.0 86.2 95.6
In June, increases in the number of individuals, number of species, and
biomass (Figures XV-3 through 6) were partially attributed to a reduction in
the salinities of the estuarine waters. The highest salinity readings (39 ppt)
were recorded at the regular stations in May; salinities had dropped 6 to 12
ppt at the time of the June sampling (Figure XV-7). June traditionally is the
start of the rainy season in South Florida. In addition, shifts in the species
composition such as 7- and 10-fold increases in the abundance of pigfish and
pinfish along with a 100-fold decrease in the number of yellowfin menhaden,
accompanied the seasonal decline in the June salinities (Table XV-10).
Distribution by Rank
The analyses of the percent composition by number and by biomass of the
40 most abundant fishes from the regular collections are presented in Table
XV-14. The wide distribution and abundance of the bay anchovy was demonstrated
by the dominance of this fish in both the numerical (75 percent) and the biomass
compositions (18 percent).
Numerical and biomass abundance of many species showed a similar ranking.
However, for the majority of species, their ranking by biomass and numerical
composition as a percent of total catch was dissimilar. For example, the
fantail and striped mullet ranked 3rd and 4th by biomass, but numerically, they
ranked 10th and 25th, respectively. Snook, red drum, and gray (mangrove)
snapper by weight ranked 14th, 18th and 30th; by number, none of these three
XV-25
-------
i-
22.0
iao
O SALINITY
A TEMPERATURE
M A M J J A
1972
40.0
30.0
20.0
s- 2
0.0
O SALINITY
A TEMPERATURE
le.oi i-
10.0
JFMAMJJASOND
1972
O 30.0
UJ
(E
SI
UJ 22.0
O SALINITY
A TEMPERATURE
JFMAMJJASOND
1972
30.0 w O 30.0
C u
2 IT
20.0 H ? 26.0
^ IT
•O UJ
~ 24D
10.0
0.0
UJ
18.0
SALINITY
A TEMPERATURE
JFMAMJJASOND
1972
Figure XV-7. Water temperature and salinity readings for Round Key at seine
Station 2 (upper left), for tidal stream off Fahka Union Canal at seine
Station 5 (upper right), for Fahka Union Bay at surface trawl Station 7
(lower left), and for Fahkahatchee Bay at surface trawl Station 9 )lower
right),
popular sport fishes were among the 40 most numerous fishes in the regular
collections (Table XV-14). In the present study, neither biomass nor numerical
species compositions used alone were entirely adequate as a means of analyzing
relative abundance. However, these data show the need to incorporate biomass
values in future assessments of fish fauna.
Relative Abundance Between Bays
Relative abundance of fishes by species, number, and biomass with surface
and otter trawling gear was compared between Fahka Union and Fahkahatchee Bays.
Throughout the year, almost twice as many species of fishes were caught with
surface and otter trawls in Fahkahatchee Bay than in Fahka Union Bay (Table
XV-15). Numerically, surface trawl catches were slightly greater in Fahka
Union Bay which was attributed to abundant catches of yellowfin menhaden in
March and April, but the mean otter trawl catch in Fahkahatchee Bay was 58
percent greater than the returns from the other bay. The mean biomasses with
the surface and otter trawls in Fahkahatchee Bay were substantially greater than
in Fahka Union Bay.
XV-26
-------
Table XV-14. Percent species and biomass composition of the 40 most abundant
fishes from the regular monthly stations in the Ten Thousand Islands, 1972,
Species
Anchoa mitchilli
Brevoortia smith!
Harengula pensacolae
Anchoa hepsetus
Lagodon rhoraboides
Bairdiella chrysura
Eucinostomus gula
Opisthonema oglinum
Menidia beryllina
Muglil trichodon
Hyporhamphus unifasciatus
Strongylura notata
Merabras martinica
Eucinostomus argenteus
Orthopristus chrysoptera
Syngnathus scovelli
Lucania parva
Floridichthys carpi o
Trachinotus falcatus
Oligoplites saurus
Jordanella floridae
Poecilia latipinna
Cyprinodon variegatus
Strongylura timucu
Mugil cephalus
Gobiosoma robustum
Hippocampus zosterae
Diapterus plumieri
Cynoscion arenarius
Arius felis
Sphoeroides nephalus
Cynoscion nebulosus
Strongylura marina
Synodus foetans
Archosargus probatocephalus
Heterandria formosa
Lut janus synagris
Microgobius gulosus
Fundulus similis
Chloroscombrus chrysurus
Composition
by Number
C%)
74.79
15.09
1.78
1.59
1.13
0.97
0.61
0.54
0.41
0.32
0.31
0.27
0.23
0.19
0.18
0.17
0.16
0.14
0.12
0.12
0.11
0.09
0.08
0.05
0.04
0.04
0.04
0.04
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
<0.01
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Composition
by Biomass
Species (%)
Anchoa mitchilli
Dasyatis americana
Harengula pensacolae
Mugil trichodon
Mugil cephalus
Bairdiella chrysura
Brevoortia smithi
Pogonius cromis
Archosargus probatocephalus
Strongylura notata
Lagodon rhomboides
Eucinostomus argenteus
Diapterus plumieri
Centropomus undecimalis
Dasyatis sabina
Trachinotus falcatus
Eucinostomus gula
Sciaenops ocellata
Hyporhamphus unifasciatus
Opisthonema oglinum
Chaetodipterus faber
Arius felis
Anchoa hepsetus
Negaprioh brevirostris
Strongylura timucu
Strongylura marina
Orthopristis chrysoptera
Lepisosteus platyrhincus
Sphoeroides nephalus
Lut janus griseus
Sphyrna tiburo
Tylosurus crbcodilus
Membras martinica
Bagre marinus
Floridichthys carpio
Oligoplites saurus
Mugil curema
Caranx hippos
El ops saurus
Paralichthys lethostigma
18.40
10.63
8.09
7.83
7.14
4.71
4.03
3.72
3.56
3.30
2.95
2.66
2.32
2.24
2.00
1.90
1.59
1.37
1.18
1.10
1.05
1.02
0.86
0.81
0.68
0.64
0.53
0.46
0.44
0.35
0.33
0.32
0.31
0.29
0.21
0.20
0.20
0.12
0.10
0.09
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Dissimilarities between these two bays existed in the composition of the
bottom fishes and plants. In Fahkahatchee Bay, the presence of abundant grass
beds (Thalassia - Diplanthera) provided an important habitat for numerous fishes,
particularly pinfish and silverperch. These two fishes often dominated the
otter trawl collections in this bay. Grassy areas are apparently a preferred
habitat for pinfish as Hoese and Jones (1963) stated that the pinfish was the
major fish species with the greatest biomass in the grass communities of Texas.
In the present study, the sparse occurrence of these grasses and their
associated fish fauna was judged to be responsible for the decreased catches in
Fahka Union Bay. In summary on the basis of the catch data, we concluded that
more favorable ecological conditions, particularly in reference to the structure
of the bottom fish and plant communities, existed in Fahkahatchee Bay (essentially
XV-27
-------
Table XV-15. Comparison of yearly catches of fishes taken by surface and otter
trawls between Fahka Union and Fahkahatchee Bays, Florida, 1972.
Gear
Number
of Species
Number
of Tows
Mean Number
of Individuals
Mean
Biomass (gm)
Surface trawl'
Otter trawlb
Fahka Union Bay
30 23
22 24
Fahkahatchee Bay
968.8
25.6
710.4
55.3
Surface trawl0
Otter trawl"
a
b =
c =
d =
Station 7
Station 11
Stations 8,
Stations 12
51 69 831.9 1,193.0
42 67 60.9 235.4
9, and 10
, 13, and 14
an undisturbed environment) than in Fahka Union Bay (a disturbed or stressed
environment).
Distribution by Habitats
The distribution of fishes was related to environmental differences among
the various estuarine and freshwater habitats. The total number of fish species
collected with all types of gear from six aquatic habitats are summarized in
Table XV-16. The occurrence of individual species by habitat are tabulated in
Appendix MM. Fishes taken in the scheduled monthly collections (catch per unit
effort) and in the monthly collections from the waterways in Fahkahatchee
Strand are distinguished from those species collected at irregular intervals
throughout the year during special studies, such as standing crop estimates,
fish collections for pesticide analyses, and juvenile snook investigations.
Also, the tabulation of fishes in Appendix MM serves as a checklist of fishes
from the coastal waters of the Ten Thousand Islands and Fahkahatchee Strand.
Generally, sandy beach habitats in the chain of islands known as the Ten
Thousand Islands are not extensive. In fact, many of the islands that boarder
the open Gulf are devoid of sandy beaches, because of extensive stands of red
mangrove trees lining the water edge. The surf zone in the Ten Thousand Islands
is generally a low energy type; the occurrence of pounding surfs is usually
restricted to storms or other hydrographic disturbances. The longest stretches
of beach are found on Morgan Beach near Cape Romano, Kice Island, Marco Island,
and Naples Beach, which are northwest of the main study area.
Shallow water reefs are characteristic of the Florida Keys, but they are
absent from the area of the Ten Thousand Islands. Thus, members of the tropical
XV-28
-------
Table XV-16. A summary of the fish species from the coastal water habitats in
the Ten Thousand Islands, 1972.
1.
2.
3.
4.
5.
6.
Habitat
Beach and adjacent waters
Middle reaches
Tidal streams in upper reaches
Tidal canals in upper reaches
Freshwater canals
Freshwater habitats in Fahkahatchee
Strand
Number
of
Species
68
89
47
59
24
17
Number
of
Families
34
42
24
25
10
7
reef fauna, such as angelfishes (Chaetodontidae), grunts (Pomadasyidae), and
wrasses (Labridae), were not represented in our faunal collections.
A total of 68 species were recorded from the beach habitat and adjacent
waters. A number of clupeids, sciaenids, carangids, gerreids, strongylurids,
and others accounted for 57 species collected in the surf zone during the
regular monthly collections. Eleven additional species were taken in special
collections along the beach or in nearshore waters.
Species unique to the catches from the beach habitat included lemon
shark (Negaprion brevirostris), blacktip shark (Carcharhinus limbatus), permit
(Trachinotus falcatus), smooth butterfly ray (Gyjimura micura), northern kingfish
(Menticirrus saxtilis), least puffer (Sphoeroides parvus), tarpon snook
(Centropomus pectinatus), longsnout seahorse (Hippocampus reidi), and houndfish
(Tylosurus crocodilus).
Two species, longsnout seahorse and least puffer, from the beach
represented range extensions. The longsnout seahorse is common in the Carribean.
Starck (1968) has the only previous record from the continental United States
which was collected in the Florida Keys. Our single specimen was taken in a
special tow off Round Key in April, 1972. Shipp and Yerger (1969) reported
that the least puffer occurs in northern waters of the Gulf of Mexico and south
to Apalachicola Bay. This species was also recorded from the Gulf of Campeche,
Mexico. Our only specimen was collected at the beach seine station on Round
Key in January, 1972. The identification of the least puffer was verified by
Dr. Thomas McKinney, National Marine Fisheries Service, on the basis of taxononic
characteristics provided by Shipp and Yerger.
Of the six habitats (Table XV-16), the greatest number of species were
recorded from the middle reaches of the estuary. This increase was attributed
to greater sampling effort, as well as to environmental conditions in the bays,
XV-29
-------
such as moderate salinities, a more abundant food supply, and a greater number
of communities including Diplanthera - Thalassia beds, stands of red algae,
oyster reefs, shallow mud flats, and tidal streams. Of the 89 fish species
from this habitat, 16 species were taken in special studies not associated with
the standard monthly unit-of-effort.
Other fishes that were only collected in the middle reaches of the estuary
included the spotted eagle ray (Aetobatus narinari), jewfish (Epinephelus
itajara), gag (Mycteroperca microlepis), highfin goby (Blennius nicholsi) ,
Florida blenny (Chasmodes saburrae), Atlantic bumper (Chloroscombrus chrysurus),
lookdown (Selene vomer), emerald parrotfish (Nicholsina usta), fringed flounder
(Etropus crossotus), and great barracuda (Sphyraena barracuda). Other fishes
inhabiting the middle reaches of the estuary, as well as the other specified
habitats, were frillfin goby (Bathygobius soporator) from a high salinity tidal
stream in north Fahkahatchee Bay and two species, naked goby (Gobiosoma bosci)
and skilletfish (Gobiesox strumosus), which were retrieved from sedimentation
traps in the bays. The skilletfish, a member of the clingfish family, is rarely
captured with trawling gear or seines because of their secretive habits and
preference for oyster communities or the underside of submerged wooden
structures.
A few of the more abundant fishes in the middle reaches of the estuary
were yellowfin menhaden, thread herring, scaled sardine, halfbeak (Hvporhamphus
unifasciatus), bay anchovy, leatherjack, mojarras (four species), pinfish,
silver perch, needlefishes (three species), pigfish, code goby, dwarf seahorse,
fantail mullet (Mugil trichodon), Gulf pipefish (Syngnathus scovelli), gold-
spotted killifish, and rough silverside (Membras martinica).
In past studies, many investigators, such as Phillips and Springer (1960),
Tabb and Manning (1960), Gunter and Hall (1965), Roessler (1970), and others,
encountered difficulties in differentiating between some of the above closely
related species or their various ontogenetic stages. The major difficulty was
the lack of adequate keys or taxonomic descriptions. Recent papers on juvenile
and larval clupeids (Dahlberg, 1970; and Houde and Fore, 1973), needlefishes
(Collette, 1968), and^mullets (Futch, 1966) were invaluable in distinguishing
these species. Meristic variations in specimens of halfbeak in our collections
demonstrated that (the genus Hyporhamphus (Miller, 1945) represented a species
complex, that is, unifasciatus probably included at least two sympatric species
in the Ten Thousand Islands. Following the present procedure, all members of
this genus in Southwest Florida were designated as unifasciatus (Springer and
Woodburn, 1960; and Bohlke and Chaplin, 1968).
Two distinct brackish water habitats were present in the upper reaches.
Generally the tidal streams were natural waterways characterized by shallow
depths (<1.0m) , strong currents, and narrow widths, while the tidal canals
were dredged (man-made) channels possessing moderate depths (1.0 to 2.0 m),
sluggish to moderate currents, and wider widths. The upper ends of many of
these streams and canals were confluent with freshwater marshes and thus,
represented the headwaters of the estuary (Carriker, 1967). Salinity patterns
in these upper reaches continually shifted during the wet and dry seasons, as
well as during the semi-diurnal tidal phases. Many freshwater and marine fishes
moved up and down these headwater regions in response to the seasonal and daily
changes in the salinity concentrations.
XV-30
-------
During the year, a greater number of species (59) were collected in the
tidal canals than in the tidal streams (47). Most fishes in the upper reaches
were either oligohaline or euryhaline species and thus, they exhibited a wide-
spread distribution among the different aquatic habitats (Appendix MM) .
Florida gar, yellowfin menhaden, bay anchovy, tidewater silverside, mosquito-
fish, sailfin molly, spotfin mojarra, and rainwater killifish were the dominant
fishes in these brackish waters. Favorable conditions of the habitat, such as
an abundant food supply and few piscivorous predators, enhanced the well-being
of young snook, redfish, and ladyfish, three popular sport fishes, that
utilized these habitats as nursery grounds. In the snook investigation, all
but 6 of 193 juveniles were captured in the upper reaches (mainly, tidal
streams).
Dissimilarities between the freshwater habitats (man-made or dredged
canals, such as GAG, Alligator Alley, and Tamiami Canals, and the various
aquatic habitats in Fahkahatchee Strand) were discussed earlier in this
chapter. More species of fishes were collected in the canals than in Fahkahatchee
Strand (Table XV-16 and Appendix MM), possibly because the canals are inundated
throughout the year, while many areas of the Strand dry-out during the dry
season. Fishes unique to the canal habitats included the seminole killifish and
the bluespotted sunfish from the eastern GAG Canal and the brook silverside
from the Alligator Alley Canal, while the brown bullhead was only taken in a
Strand lake (Ballards). More extensive sampling in these areas undoubtedly
would have increased the list of fishes from these habitats.
Distribution in Time
The monthly distribution of 134 species of fishes, largely collected
during the 1972 biological year, are indicated in Appendix NN. Fishes
identified from the regular estuarine and freshwater stations (Stations A-^ and
A2) were distinguished from those taken in other special collections. Because
of the number of fishery studies in diverse habitats and the deployment of
several types of sampling gear, no attempt was made to affix numbers to
individual species, since comparisons of these data would be meaningless
because of inherent gear selectivity. The addition of biomass values would
have compounded this problem. However, seasonal patterns of distribution of
the most abundant species and families by number and by biomass were satisfied
to a certain extent in the prior tables and figures in the results.
Although subjective, a comparison of the estuarine species listed by
habitat (Appendix MM) and by time of collection (Appendix NN) will furnish some
general information on seasonality and relative abundance of the less numerous
species. For instance, uncommon species, such as the bull shark (Carcharinus
leucas), tarpon snook (Centropomus pectinatus) and highfin blenny (Blennius
nicholsi) were recorded from a single habitat in a given month, whereas common
species usually were observed in several habitats during a number of months.
All of the freshwater fishes were permanent residents and probably would have
been recorded in every month of the year (Appendix NN), had these areas been
sampled more frequently.
XV-31
-------
Discussion
The present investigation represents the only extensive survey of the
fish fauna of the Ten Thousand Islands and adjacent freshwater areas. Of the
134 species in our checklist (Appendix KK), subsequent collections will
undoubtedly increase the number of fishes from these waters.
Along the southwestern coast of Florida, similar species of fishes
dominate the fish fauna. In the Ten Thousand Islands, the most abundant
fishes are anchovies, yellowfin menhaden, scaled sardine, pinfish, silver
perch, and silver jenny. By far, the pelagic bay anchovy is the dominant
species occurring in all zones of the estuary throughout the year. In a 3-year
study of the fishes in Whitewater and Florida Bays, Tabb and Manning (1961)
observed that the most abundant fishes were anchovies, mojarras, silver perch,
pinfish, and code goby. Studies in other adjacent areas, such as Charlotte
Harbor (Wang and Raney, 1971), the Caloosahatchee River and nearby estuaries
(Gunter and Hall, 1965), and the Buttonwood Canal (Roessler, 1967), recorded
a similar assemblage of dominant fishes.
Fish production in the mangrove-lined estuaries of the Ten Thousand
Islands was illustrated by the occurrence of numerous juvenile sport, commercial,
and forage fishes and shellfishes in the collections. Many of these species
are estuarine-dependent, that is, either a portion of or in a few instances
their entire life cycles, are spent in the estuaries.
Young of popular sport fishes in southwest Florida that utilize these
waters as nursery grounds include the red drum, snook, ladyfish, gray (mangrove)
snapper, lane snapper, sheepshead, tarpon, and spotted seatrout. Most of these
species were spawned in the offshore waters of the Gulf and first entered the
estuaries as post-larval or juvenile stages. In this group, only the spotted
seatrout spawns in the bays, while the snook (Marshall, 1958) and the red drum
(Yokel, 1966) reportedly spawn in the deep outer passes leading to the Gulf of
Mexico. The early juvenile stages of gray snapper, spotted seatrout, red drum,
sheepshead, and lane snapper initially are found in the grass (Diplanthera -
Thalassia) beds in Fahkahatchee Bay and later move to other estuarine habitats.
Post-larval and juvenile ladyfish, snook, and tarpon (Wade, 1969) inhabit the
brackish water areas of the mangrove estuaries, such as tidal streams, ditches,
canals, and marshes, which provide favorable environmental conditions for their
growth and survival. As these young increase in size, they move to the higher
salinity regions of the estuary. A description of tidal streams as a major
nursery habitat for juvenile snook is presented in the following chapter.
Estuarine finfishes and shellfishes dominate the commercial fishery
landings in the Gulf of Mexico. Twenty-three major commercial species were
found to inhabit the estuary of Tampa Bay as immatures (Sykes and Finucane,
1964). All of these species were taken in various inshore habitats in our
study area, except the red grouper, Epinephelus morio. Although not listed by
Sykes and Finucane (op. cit.), juvenile jewfish, also a member of the grouper
family, were fairly common in the bays of the Ten Thousand Islands during the
summer and fall.
Pink shrimp, Penaeus duorarum, blue crab, Callinectes sapidus, and eastern
oyster, Crassostrea virginica, are all dependent on the estuarine habitat for
survival. Although oysters were not sampled with the trawling gear, oyster
bars were present along the shorelines of the bays, in the channels, and at the
XV-32
-------
mouths of rivers and creeks. Pink shrimp and blue crabs were taken at all
trawling stations in the bays. Fahkahatcb.ee Bay supports a large population
of immature pink shrimp which were particularly abundant in the seagrass beds.
Juvenile shrimp from the estuaries in this area are harvested as adults on the
Tortugas and Sanibel fishing grounds (Costello and Allen, 1966). Shrimp are
the most valuable fisheries resource in south Florida.
The different regions in the estuary varied in importance as nursery
grounds (Appendix MM) . Based on the distribution and abundance of juvenile
fishes, these areas are listed in the decreasing order of importance as follows:
bays in the middle reaches of the estuary; brackish water areas, such as
tidal streams, canals, and marshes, in the upper reaches; and the open beaches
in the lower reaches. Of the two bays that were intensely studied, Fahkahatchee
Bay was a more productive nursery than was Fahka Union Bay. Fahkahatchee Bay
featured a greater abundance of benthic vegetation and invertebrate fauna as
compared to the stressed environment of Fahka Union Bay. Although not as many
juvenile species inhabit the low salinity zone in the upper reaches, these
areas provide a vital habitat for many fishes, such as snook, ladyfish, sheeps-
head minnow, tidewater silverside, and rainwater killifish. The sandy outer
beaches served as a nursery ground for the least number of species, partially
due to a less abundant supply of food and shelter. However, some species
favored this region at certain times in their early life history. For example,
juvenile permit were only collected at the beach stations in our area. Young
scaled sardine and threadherring inhabited the bays during the summer; during
the fall months, they were concentrated in the surf zone of the outer beaches.
To date, a serious threat to the well-being of the estuarine ecosystem
in the study area concerns the presence of an extensive system of drainage and
borrow canals, such as GAG and Tamiami Canals. These canals disrupt the
seasonal water patterns and reduce the overland flow of fresh water to the
estuaries. At the height of the winter drought in May, 1972, the lack of
freshwater runoff created hypersaline conditions in Fahkahatchee Bay with
salinities approaching 39.0 ppt. Even higher levels have been reported in
previous years. The onset of the summer rains lowered the salinity levels in
June. Hypersaline conditions upset the salinity transport mechanism by which
larval and juvenile fishes and shrimp find their way into the estuary and
remove the low salinity barrier which allows marine predators to enter the
nursery grounds. Thus, the protective nature of the estuary is lost. Odum
(1970) stated that the maintenance of a salinity gradient is essential for the
survival and protection of estuarine-dependent species, particularly those that
are spawned outside the estuary.
On the basis of our evaluation of the data collected during the present
study, it is concluded that the estuary in the Ten Thousand Islands provides
favorable habitats which supply food and shelter for numerous fishes and
macroinvertebrates. Part of the catch taken by the sport and commercial
fisheries is directly related to the production and growth of young individuals
in the Ten Thousand Islands.
XV-33
-------
XVI - BIOLOGY OF JUVENILE AND ADULT SNOOK, CENTROPOMUS UNDECIMALIS,
IN THE TEN THOUSAND ISLANDS
INTRODUCTION
The snook, Centropomus undecimalis, is an extremely popular sport and
food fish from the mangrove-lined estuaries of southwest Florida. The fight-
ing qualities of this species make it a worthy adversary for both novice and
seasoned fisherman, and its fillets are a gourmet's delight. This species is
prevalent in the tropical and subtropical waters of south Florida along both
the east and west coasts. The most abundant concentrations are found in the
Ten Thousand Islands. This area also supports the main spawning grounds which
reportedly occur near the outer passes leading to the Gulf of Mexico (Marshall,
1958). A thorough understanding of the biology and ecology of the snook is a
necessary prerequisite for the proper management of the species.
Mos't of the prior investigations on the snook dealt primarily with the
adults. Information on migration, age, and growth (Volpe 1959); general
biology and historical data on the commercial fishery (outlawed in 1957)
(Marshall, 1958); osteology (Eraser, 1958); and taxonomy (Rivas, 1962) were
among the more comprehensive works. The few references on the biology of the
juveniles included those on the food habits from a marsh habitat (Harrington
and Harrington, 1961); on range extension (Lunz, 1953, and Martin and Shipp,
1971); and on general ecology (Springer and Woodburn, 1960, and Linton and
Rickards, 1965).
Many aspects of the biology of the snook are still unknown. This paper
presents the results of a study on tidal-stream habitats, length-weight rela-
tionships, food habits, and seasonal occurrences of juveniles and food and
feeding habits of the adults from the Ten Thousand Islands, Florida.
MATERIALS AND METHODS
The study commenced in November 1971 and continued through December 1972.
Juvenile snook were,collected from tidal streams and canals in the upper
reaches of the estuary. The selection of sampling sites was based on accessi-
bility and the frequent presence of juveniles.
All juveniles were captured with a 15.2 m by 1.2 m nylon (Delta style)
bag seine. The wings were constructed of 0.64-cm bar mesh and the bag of 0.3-
cm bar mesh. Young snook were immediately preserved in a 5-percent solution
of formalin in saline water. Biological data on standard, fork, and total
lengths; wet weights; and stomach contents were recorded in the laboratory.
Specimens of adult snook were obtained at marinas on the Isle of Capri,
Marco Island, and the GAC Remuda Ranch; from seine collections; and by hook-
and-line during the summer. Data on lengths, sex, stages of gonadal develop-
ment, and stomach contents were usually taken in the field; whereas, f :>od items
were analyzed in the laboratory.
A total of 454 snook were examined during an intensive study of the food
relationships of this species. The 183 juveniles (142 stomachs with food)
ranged in length from 14 to 196 mm, and the 271 adults (127 stomachs with food)
XVI -1
-------
varied from 224 to 1,020 mm in length. The adults were collected during the
summer, and the juveniles were taken during the summer and fall.
All length measurements of fishes in this paper were of fork length
(anterior extremity of closed jaws to the tip of the rays in the center of the
caudal fin) to the nearest 1.0 mm, unless stated otherwise. Wet weights were
taken with a dial-type balance (1,600-gm capacity) to the nearest 0.1 gm.
The analysis of the physical and chemical characteristics of estuarine
waters followed the procedures outlined in the manual by the Environmental
Protection Agency (1971), except for salinity and water temperatures which
were taken with a salinometer.
HABITATS OF JUVENILES
Numerous aquatic zones (outer beaches, open bays, and freshwater canals)
were searched before the nursery grounds of juvenile snook were located in the
Ten Thousand Islands. Marshall (1958) was unsuccessful in his attempts to
capture juveniles from this area. In our collections, young snook were only
taken in the brackish waters of tidal streams and in dredged canals. Other
features of these habitats included shallow depths, currents, shifting salini-
ties, warm water temperatures, soft mud bottoms with little vegetation, often
shoreline stands or clumps of red (Rhizophora mangle) or white (Laguncalaria
racemosa) mangroves, an abundance of small food organisms, and few aquatic
predators. Some stream habitats were contiguous with adjacent marshy areas
during the rainy season. The sampling stations are described in Table XVI-1.
The waters of all stations drained into the lower bays of the estuary and
eventually into the Gulf of Mexico.
Table XVI-1. Descriptions of habitats where juvenile snook, Centropomus
undecimalis, were collected in Ten Thousand Islands
Station
Number
1
2
3
4
5
Location
Tamiami Trail,
1.6 km east of
Collier-Seminole
State Park
East Naples,
Barefoot Williams
Road
East Naples,
Barefoot Williams
Road
East Naples,
access road to
KOA Campgrounds,
west of Hwy. 951
North Fahkahat-
chee Bay, 120 m
east of mouth of
East River
Mean Water
Description Depth (m)
Tidal canal at 1.0
upper reaches of
Blackwater River
Tidal canal con- 1.0
fluent with
Henderson Creek
Tidal ditch 0.3
empties into
upper end of
Station 2
Tidal stream 0.6
branching off
Henderson Creek
Tidal stream 1.2
Bottom
Type
Soft mud and
filamentous
algae
Soft mud and
sand
Firm mud, and
sand
Soft mud
Firm mud and
sand
, Florida, 1971-1972.
Shoreline
Vegetation Tidal Current
Roadside weeds Moderate
with scattered
red mangrove and
marshes
Grasses Sluggish
Roadside weeds Sluggish to
with scattered moderate
palms and button-
wood
Stands of white Moderate
mangrove
Stands of red Rapid
mangrove
Juveniles were reported from similar stream habitats by other investiga-
tions. On Sapelo Island, Georgia, Linton and Rickards (1965) collected
specimens from the headwaters of tidal creeks; but none were taken at the
XVI-2
-------
creek mouths, in the sounds, or on the outer beaches. In Tampa Bay, Springer
and Woodburn (1960) seined 35 specimens, 26.6 to 191.0 mm SL, from a man-made
bayou canal. They stated that "the preferred habitats of the young are pro-
tected bodies of water, usually of small surface area and shallow depths."
Based on juvenile specimens stored in various museums, Marshall (1958) con-
cluded that young snook in Florida inhabited the marginal areas of the estuary,
such as drainage streams, creeks, shorelines, small ditches, and man-made
tributaries.
Marshlands also serve as major nursery areas for juvenile snook. This
type of habitat was described by Harrington and Harrington (1961). They found
juveniles in a seasonally inundated salt marsh that was adjacent to the brack-
ish waters of the Indian River, Florida. Drought conditions existed in the
marsh until flooding occurred in September due to a combination of heavy rain-
fall and high tides. Invading fishes included juvenile snook, as well as young
tarpon (Megalops atlantica), ladyfish (Elops saurus), and several other species.
Major characteristics of the marsh habitats were salinities ranging from 20.0
ppt to hypersaline conditions, high water temperatures, sand and peat substrate,
plentiful supply of forage organisms, and rooted vegetation consisting of thick
beds of saltwort (Batis maritima) and pneumatophores of the mangroves.
HYDROGMPHIC DATA ON JUVENILES
Juveniles were taken at five stations during numerous stages of the tidal
cycle (Table XVI-2). The salinity ranged from lows of 0.3 ppt to a high of
29.7 ppt. A statistical relationship was not demonstrated between the positive
seine hauls (46) and the salinity of the water (r = 0.23, 44 d.f.). It was
thought that juveniles of a given size (length) might be taken more frequently
at a preferred salinity. But again, no correlation was shown between salinity
and individual sizes (r = 0.21,191 d.f.). However, it was evident from the
collections that juvenile £. undecimalis are usually found in brackish waters
in the Ten Thousand Islands.
The warmest water temperatures were recorded during the summer months;
temperatures gradually declined to a low of 18.7°C in the late fall (Table
XVI-2). Colder water temperatures occurred at night, but all sampling took
place during daylight hours. A statistical correlation could not be estab-
lished between the positive collections and the water temperature. Also, no
correlation existed between the size of the juveniles and the temperature.
Marshall (1958) remarked that snook could not tolerate water temperatures less
than 15.5°C (about 60°F). Mortality of snook following sudden cold spells near
Sanibel Island, Florida, were discussed by Storey and Gudger (1936) and Storey
(1937).
Measurements of selected chemical parameters were made monthly from three
tidal stream habitats from June through December 1972 (Table XVI-3). No com-
parative chemical parameters for juvenile snook habitats were found in the
literature.
PESTICIDES
Three snook were analyzed for the presence of 23 pesticides distributed
among a class of compounds known as chlorinated hydrocarbons, organophosphates,
and organic carbamates. Concentrations of four chlorinated hydrocarbons (DDE,
XVI-3
-------
Table XVI-2. Temporal, spatial, and hydrological data on all collections of
juvenile snook, Centropomus undecimalis, from the Ten Thousand Islands,
Florida, 1971-72.
Date
Station
Salinity
(PPt)
Temperature
(C)
Tidal
Stage
Juvenile Fish
Number
Size Range
(mm)
1971
17
24
13
Nov.
Nov.
Dec.
1,
2
1,
2
2, 3
1.
15
5-19.7
.9
18.9-20.7
22.
27.
27.
7-25
4
5-27
.4
.9
Ebb, Flood
Flood
Ebb
5
1
4
123-178
104
89-184
1972
22
23
6
20
1
11
15
22
24
25
28
31
14
15
22
28
29
5
12
17
20
27
8
13
15
24
2
6
8
12
29
June
June
July
July
Aug.
Aug.
Aug.
Aug.
Aug.
Aug.
Aug.
Aug.
Sept.
Sept.
Sept.
Sept.
Sept.
Oct.
Oct.
Oct.
Oct.
Oct«
Nov.
Nov.
Nov.
Nov.
Dec.
Dec.
Dec.
Dec.
Dec.
3
3
2
1,
4
4
2,
2,
4
4
4
3,
4
3,
3
4
3,
3,
4
3,
3
5
4
3,
3
4
4
4
4
1
4
3
4
4
4
4
4
4
4
4
0.
0.
5.
0.
1.
7.
0.
8.
3.
2.
2.
0.
0.
0.
8.
0.
0.
0.
0.
1.
2.
29
3.
6.
10
1.
1.
2.
12
7.
14
5
5
3
6-3.0
2
4
5-4.9
5-9.3
4
4
2
4-0.5
4
4-0.5
7
3
3-0.5
3-0.4
3
6-7.0
0
.7
0
9-13.0
.3
6
5
4
.9
1
.4
32.
32.
35.
27.
30.
31.
29.
27.
30.
28.
31.
26.
30.
28.
28.
27.
28.
23.
25.
29.
25.
26.
28.
28.
27.
19.
18.
24.
28.
23.
19.
6
9
4
1-27
3
8
9-32
2-27
4
2
1
8-26
7
8-29
7
7
5-28
5-24
7
2-29
0
8
9
6-29
6
4
7
2
0
5
0
.3
.5
.8
.9
.3
.6
.3
.9
.6
Ebb
Ebb
Ebb
Flood, Ebb
Ebb
Flood
Ebb
Flood
Flood
Flood
Flood
Ebb
Ebb
Ebb
Flood
Ebb
Ebb
Ebb
Ebb
Ebb
Flood
Ebb
Ebb
Ebb
Ebb
Ebb
Ebb
Ebb
Flood
Ebb
Ebb
2
1
3
4
2
3
3
5
5
2
3
12
2
11
2
6
10
7
19
3
3
6
26
6
7
14
'4
2
1
6
3
24-25
22
14-35
27-37
14-57
35-91
30-r62
40-104
38-92
22-99
29-98
31-101
67-94
35-107
116-121
60-95
67-153
22-110
55-161
76-124
92-148
58-121
70-196
51-146
82-117
63-149
30-70
46-47
41
26-49
48-64
DDD, DDT, and PCB's) were detected in two subadulults and one adult snook (Table
XVT-4). No detectable concentrations were found for 19 other pesticides
(aldrin, lindane, chlordane, chlorobenzilate, deldrin, endrin, heptachlor
epoxide, heptachlor, methoxchlor, toxaphene, diazinon, guthion, methyl para-
thion, malathion, ethion, trithion, dimethoate, mirex and parathion).
Because of the small sample size and the lack of comparative data on
snook from the literature, no firm conclusions on the persistence and effects
of DDE, DDD, DDT, and PCB's on the well-being of snook can be made at this
time. The carnivorous diet of the snook and low levels of similar pesticides
XVI-4
-------
Table XVI-3. Physical and chemical characteristics of water collected monthly
from three tidal streams inhabited by juvenile snook, June to December, 1972.
Parameter Mean Range
pH 7.5 7.1 to 7.6
Temperature (C) 28.8 23.5 to 33.7
Salinity (ppt) 5.4 0.2 to 21.0
Conductivity (millimhos/cm) 9.5 0.3 to 34.8
Chloride (mg/1) 2,115.8 38.4 to 9,322.0
Dissolved oxygen (mg/1) 5.9 3.1 to 9.3
Total hardness (mg/1 as CaC03) 956.9 70.0 to 3,400.0
Total alkalinity (mg/1 as CaC03) 206.8 52.9 to 309.0
Total phosphorus (mg/1) 0.10 0.04 to 0.17
Total soluble phosphorus (mg/1) 0.03 <0.02 to 0.05
Total Kjeldahl nitrogen (mg/1) 1.36 0.87 to 1.80
Total soluable Kjeldahl nitrogen (mg/1) 0.93 0.80 to 1.24
Table XVI-4. Analyses of pesticides in snook, Centropomus undecimalis, in Ten
Thousand Islands, Florida, 1972.
Chlorinated Hydrocarbons
Length DDE ODD DDT PCB's
Date Location (mm) (yg/kg) (ug/kg) (yg/kg) (yg/kg)
Jan. 26
Jan. 26
July 28
Henderson Creek
Henderson Creek
Fahkahatchee Bay
140
206
335
40.0
42.0
13.0
56.0
51.0
5.4
12.0
11.0
NDa
120.0
85.0
NDa
a = Not detectable
in forage fishes from nearby waterways (see Chapter XI on pesticides), lead to
the conclusion that this predator acquired these pesticides through the food
chain. These results also showed that some pesticides have become an integral
component of the coastal marshes, streams, and bays in southwest Florida.
XVI-5
-------
ASSOCIATED ORGANISMS
The occurrence of other fishes and macrocrustaceans taken in the same col-
lection along with juvenile snook from four sites (Stations 1, 2, 3, and 4)
are listed in Table XVI-5. These associated animals were sorted from 15 ran-
domly selected collections or 33 percent of the total number of positive hauls.
Only a few fishes, such as redfin needlefish (Strongylura notata), Florida gar
(Lepisosteus platyrhincus), and perhaps other young-of-the-year snook, were
of sufficient size to prey upon juvenile snook. The fact that these juveniles
inhabit low salinities, largely free of piscivorous predators, may operate as
a survival adaptation for the species. In turn, most of the smaller fishes
were utilized as forage by the snook.
Table XVI-5. Associated macrocrustaceans and fishes collected with juvenile
snook during 15 randomly selected seine hauls at four stations in Ten
Thousand Islands, Florida, June through December, 1972.
Species
Number of
Times Present
Percent of
Total Number
of Hauls
Total Number
of Individuals
Macrocrustaceans
Penaeus duorarum
Palaemonetes intermedius
Palaemonetes pugio
Callinectes sapidus
Fishes
Lepisosteus platyrhincus
Elops saurus
Brevoortia smithi
Anchoa hepsetus
Anchoa mitchilli
Arius fells
Strongylura notata
Strongylura timucu
Adinia xenica
Cyprinodon variegatus
Floridichthys carpio
Fundulus confluentus
Pundulus grandis
Jordanella floridae
Lucania goodei
Lucania parya
Gambusia affinis
Heterandria formosa
Poecilia latipinna
Menidia beryllina
Lepomis gulosus
Lepomis macrochirus
Lepomis microlophus
Lepomis punctatus
2
2
8
8
1
1
2
2
4
1
1
1
3
8
2
5
2
2
2
12
8
1
12
15
2
1
2
1
,7
,7
13.3
13.3
53.3
53.3
6.7
6.7
13.3
13.3
26.
6.
6.7
6.7
20.0
53.3
13.3
33.3
13.3
13.3
13.3
80.0
53.3
6.7
80.0
100.0
13.3
6.7
13.3
6.7
2
38
261
40
1
2
366
6
404
2
2
1
3
279
10
30
6
6
2
448
516
2
773
3,629
2
1
24
2
XVI-6
-------
Table XVI-5. Associated macrocrustaceans and fishes collected with juvenile
snook during 15 randomly selected seine hauls at four stations in Ten
Thousand Islands, Florida, June through December, 1972. Continued.
Species
Number of
Times Present
Percent of
Total Number
of Hauls
Total Number
of Individuals
Micropterus salmoides
Diapterus plumieri
Eucinostomus argenteus
Eucinostomus gula
Archosargus probatocephalus
Lagodon rhomboides
Cyjnoscion arenarius
Cynoscion nebulosus
Leiostomus xanthus
Pogonias cromis
Sciaenops o cell at a
Mugil cephalus
Mugi 1 curema
Mugil trichodon
Gobiosoma rob us turn
Lophogobius cyprinoides
Microgobius gulosus
Achirus lineatus
Trinectes maculatus
1
6
14
1
1
2
1
1
2
1
2
4
1
2
4
8
2
1
7
6.7
40.0
93.3
6.7
6.7
13.3
6.7
6.7
13.3
6.7
13.3
26.7
6.7
13.3
26.7
53.3
13.3
6.7
46.7
2
72
256
8
3
18
3
1
3
5
6
77
1
2
6
148
11
1
11
LENGTH CONVERSION FACTORS
To provide comparisons of length measurements with the published data of
other investigations, conversion factors were calculated to convert one type of
measurement to another. Empirical data on standard-, fork-, and total-length
measurements were recorded from 175 juvenile snook within a size range of 14 to
196 mm FL. Linear regressions were determined by the method of least squares.
Conversion factors and statistics describing the length relationships in mm
were as follows:
SL = -0.369 + 0.870 FL r = 0.999 p = 0.01
FL = -0.424 + 1.150 SL
FL = -1.531 + 0.909 TL r = 0.998 p = 0.01
TL = 1.684 + 1.100 FL
SL = -0.577 + 0.789 TL r = 0.999 p = 0.01
TL = 0.732 + 1.268 SL.
Juvenile snook, less than 14 mm FL, may not conform exactly to the above con-
version factors.
XVI-7
-------
LENGTH-WEIGHT RELATIONSHIPS OF JUVENILES
The relationship of fork length
to weight was derived from the fol-
lowing equations (Lagler, 1956):
100.0
log a =
Zlog WS(log L)2
-Slog L«E(log L'log W)
n = Zlog W-(N-log a)
Zlog L
log w = log a + n*log L,
where a and n are constants, W is
the wet weight in gm, and L is the
fork length in mm. In the current
study, the calculated length-weight
formula was :
log W = -4.7730 + 2.8758'log FL.
The correlation coefficient for this
formula was 0.998.
The log-log graph of weight as
a function of length for 193 juve-
nile snook is shown in Figure XVI- 1.
If either the weight or the length
of the fish is unknown, the log-log
graph can be used to estimate the
other value. However, the loga-
rithmic formula is more precise in
determining an unknown weight or
length measurement.
The coefficient of condition
(K) is a numerical measure of the
well-being or relative robustness of
fish (Lagler, 1956). Coefficient
values of juveniles were calculated
from the following equation:
where W is the weight in gm, L is
the fork length in mm, and the fac-
tor of 105 reduces the number of
decimals in the K value. Condition
values are useful in comparing the
same fish from different areas to
determine the suitability of habi-
tats or environmental changes.
10.0
o»
I
—
LU
1.0
O.I
0.01
10 20
FORK LENGTH (cm)
Figure XVI-1. The length-weight relation-
ship (log W = -4.7730 + 2.8758-log FL)
of juvenile snook in the Ten Thousand
Islands, Florida.
XVI-8
-------
The average condition value for juvenile snook from tidal stream habitats
in the Ten Thousand Islands was 1.05. Condition factors ranged from a low of
0.77 to a high of 1.52. There was a slight, but non-significant, increase in
condition values associated with an increase in length. Overall, the snook
appeared to be in good condition. No comparative values on young snook from
other regions were found in the litereature.
OCCURRENCE AND GROWTH
Study sites established in the fall of 1971 were sampled twice a month
from January 1972 until the initial members of the 1972-year-class (two speci-
mens 24.0 and 25.0 mm in length) were captured on June 22 (Table XVT-2). The
sampling effort intensified after this date as time permitted, since a new
year-class was entering the nursery grounds. A steady increase in the size of
the juvenile population occurred
1971
during the remainder of the sum-
mer and into the fall, reaching
peak abundance in November. The
monthly'mean number of juveniles
per positive seine haul was as
follows: June - 1.5; July - 2.3;
August - 3.2; September - 4.4;
October - 5.4; November - 10.6;
and December - 3.2. No defini-
tive records were kept on the
negative hauls that occurred on
several occasions. These studies
clearly show that juveniles are
found in the upper reaches of the
estuary during the summer and
fall months.
The monthly length-frequency
distribution for 193 juvenile
snook that were captured during
the study are depicted in Figure
XVT-2. Combined to,tals were pre-
sented for a couple of months
because of the small sample size.
Juveniles taken in November and
December 1971 were considerably
larger than those obtained in
December 1972. The smaller-sized
juveniles in the December collec-
tions in 1972 were attributed
either to colder water tempera-
tures which caused the larger
juveniles to migrate to the bays,
or to a late fall spawning period
in 1972 compared to 1971 (Table
XVI-2).
Juvenile snook grpw at the
rate of about 1.0 mm per day, at
N=IO
3.0
15.0 17.0 19.0
5.0 70 9.0 11.0 13.0
FORK LENGTH (cm)
Figure XVI-2. Length-frequency distribu-
tion of 193 juvenile snook from the Ten
Thousand Islands, Florida, 1971-72.
XVI-9
-------
least during the warmer months. This estimate was derived from the following
data: (1) the monthly length^frequency graph (Figure XVI-2), (2) the size of
the juveniles at the time of collection (Table XVI-2), and (3) the commencement
of the spawning season based on gonadal stages of development in the adults.
The spawning season of the snook in the Ten Thousand Islands occurs from
the first of May to about the middle of November. In May, several adult males
were sexually mature (running-ripe), while many females were in advanced stages
of maturing. Marshall (1958) stated that spawning started in June, but he was
unable to examine gonadal development during May.
The growth rate of 1.0 mm per day correlated with our estimate of the start
of the spawning season. For example, the juvenile snook (102 mm TL) that was
collected on August 11, 1972, and that grew at a daily rate of 1.0 mm, would
back calculate to an estimated spawning date of about May 1. Also, a 204-mm TL
fish taken on December 13, 1971, would give an estimated spawning tine of about
May 23, 1971. Since the metabolic requirements of fish are related to tempera-
ture, a slower growth rate presumably occurs during the colder periods in
November and December.
FOOD HABITS
Fishes and crustaceans (shrimps, crabs, and zooplankton) comprised the main
food items in the diet of juvenile snook. Fishes occurred in 77.5 percent of
juvenile stomachs and represented 80.8 percent of the total food volume. Thus,
fishes were the most important item in the diet. Piscivorous prey of the juve-
niles consisted mostly of livebearers (Poecilidae), killifishes (Cyprinodontidae),
and silversides (Atherinidae) (Table XVI-6). By total volume, the most'abundant
fishes in the diet of the juveniles were Poecilia latipinnia (34.9 percent),
Gambusia affinis (12.7 percent), Lopliogobius cyprinoides (7.7 percent), and
Menidia beryllina (3.8 percent). Other identifiable fishes included Fundulus
confluentus, Cyprinodon variegatus, Adenia xenica, and Scianops ocellata. Of
the 138 forage fishes appearing in the stomach contents, 70.2 percent were
identifiable to species. In their study of the food habits of juvenile snook
(up to 45 mm SL), Harrington and Harrington (1961) failed to identify the species
of fishes that juveniles had preyed upon.
Fishes represented about half of the total diet of the adults (by occurrence
and volume), but fishes were less important as food in comparison to the juve-
niles (Table XVI-7). The majority of fishes identified in adult stomachs, such
as Lagodon rhomboides, Anchoa mitchilli, Harengula pensacolae, Eucinostomus gula,
Synodus foetens, and Floridicthys carpio, typically inhabit the higher salinity
regions of the estuary. Forage fishes found in brackish and freshwater areas
were also preyed upon. Snook are classed as a euryhaline species because they
possess physiological adaptations enabling them to travel back and forth between
highly saline and freshwater environments.
Of the 93 fishes distributed among 10 families in the adult stomachs, we
were only able to identify 23.6 percent of these to species; the. unidentifiable
fishes were in various stages of digestion. Marshall (1958) encountered similar.
difficulties in his examination of the stomach contents of 128 adults (52.3 per-
cent had empty stomachs), as he only identified seven species belonging to six
families, Two of the fishes, Mugil cephalus and Orthopristis chrysoptera, were
not noted in the current study.
XVI-10
-------
Table XVI-6, Stomach contents of 183 juvenile snook, 142 of which contained
food, from Ten Thousand Islands, Florida, June to December 1972.
Food Item
Arthropoda
Crustacea
Cladocera
Ostracoda
Copepoda
Amphipoda
Decapoda
Penaeidae
Penaeus duorarum
Palaemonidae
Palaemonetes pugio
P. vulgarls
P^ spp.
Portunidae
Insect a
Corixidae
Unidentifiable nymphs
Vertebrata
Osteichthyes
Cyprinodontidae
Adenia xenica
Cyprinodon variegatus
Fundulus con fluent us
Lucania parva
Poeciliidae
Gambusia af finis
Poecilia latipinnia
Numerical
Total
2
3
38
13
6
9
1
28
5
5
3
1
2
3
6
47
25
Percent
Frequency of
Occurrence
0.7
0.7
6.3
0.7
4.2
5.6
0.7
18.3
2.8
2.8
2.1
0.7
1.4
2.1
4.2
23.2
16.2
Percent
of Total
Volume
<0.1
<0.1
<0.1
<0.1
3.8
3.6
0.2
7.9
1.8
0.5
0.9
0.7
1.2
1.4
5.0
12.7
34.9
XVI-11
-------
Table XVI-6. Stomach contents of 183 juvenile snook, 142 of which contained
food, from Ten Thousand Islands, Florida, June to December 1972. Continued.
Food Item
Numerical
Total
Percent
Frequency of
Occurrence
Percent
of Total
Volume
Atherinidae
Menidia beryllina
Sciaenidae
Scianops ocellata
Gobiidae
Lophogobius cyprinoides
Unidentifiable fishes
Planta
Algae
Vascular plants
SUMMARY
Microcrustaceans
Shrimps
Crabs
Insects
Fishes
Plant material
4
41
2
6
56
44
5
8
138
8
5.6
0.7
2.8
26.8
0.7
2.1
6.3
24.6
2.8
4.2
77.5
2.8
3.8
0.9
7.7
12.5
0.4
0.1
15.5
1.8
1.4
80.8
0.4
Crustaceans were of secondary importance in the juvenile diet, since this
group only represented approximately 20 percent of the food intake by percent
occurrence and volume (Table XVI-6). Microcrustaceans (copepods, cladocerans,
ostracods, and amphipods) were selected as food by the smaller juveniles. Of
the shrimps, grass shrimps (Palaemonetes pugio and P_. vulgaris) constituted a
greater porportion of the diet than the pink shrimp (Penaeus duorarum). Volu-
metrically, palaemonid and panaeid shrimps comprised 11.7 percent and 3.8
percent of the total food intake, respectively. Crabs were of little impor-
tance ii the diet of juveniles (1.8 percent by volume). Also, juveniles preyed
upon aquatic insects (corixid nymphs and others) to a limited extent.
Overall, crustaceans, composed of crabs and shrimps, were of greater impor-
tance in the diet of adult snook than in juveniles. Crabs occurred in 48.0
percent of all adult stomachs and represented 32.3 percent of food volume,
while various shrimps comprised 35.4 percent by occurrence and 9.6 percent by
volume of the total diet. The importance of crabs as a food item was not in
agreement with Marshall (1958), who reported that crabs only represented 4.4
percent by volume of the snooks' diet in the Ten Thousand Islands.
XVI-12
-------
Table XVI~7. Stomach contents of 271 adult snook, 127 of which contained food,
from Ten Thousand Islands, Florida, Hay to August 1972.
Food Item
Crustacea
Isopoda
Decapoda
Astacidae
Procambarus alleni
Penaeidae
Penaeus duorarum
Palaemonidae
Palaemonetes intermedius
P. spp.
Alpheidae
Alpheus heterochaelis
A. spp.
Hippolytidae
Hippolyte pleuracantha
Paguridae
Paguristes spp.
Pagurus spp.
Portunidae
Callinectes ornatus
C. sapjldus
Portunus gibbesii
P. sayi
H SPP-
Unidentifiable portunids
Xanthidae
Menippe mercenaria
Neopanope texana
/ertebrata
Amphibia
Ranidae
Rana spp.
Numerical
Total
1
1
85
3
6
2
1
1
1
1
9
11
89
5
26
36
1
2
1
Percent
Frequency of
Occurrence
0.8
0.8
29.1
1.6
2.3
1.6
0.8
0.8
0.8
0.8
4.7
7.9
11.0
2.4
11.0
18.1
0.8
1.6
0.8
Percent
of Total
Volume
0.1
1.0
9.1
0.1
0.1
0.3
<0.1
<0. 1
5.0
8.6
10.2
0.6
4.3
3.0
0.2
0.3
0.2
Osteichthyes
Clupeidae
Harengula pensacolae
0.8
0.2
XVI-13
-------
Table XVI-7. Stomach contents of 271 adult snook, 127 of which contained food,
from Ten Thousand Islands, Florida, May to August 1972. Continued.
D , T, Numerical
Food Item „ . ,
lotal
Engraulidae
Anchoa mitchilli
Synodidontidae
Synodus foetens
Antherinidae
Menidia beryllina
Cyprinodontidae
Cyprinodon variegatus
Floridicthys carpio
Poeciliidae
Poecilia latipinna
Centrarchidae
Lepofflis punctatus
Gerreidae
Eucinostomus gula
Sparidae
Lagodon rhomboides
Gobiidae
Lophogobius cyprinoides
Unidentifiable fishes
Planta
Algae
Phaeophyta
Vascular plants
Diplanthera wrightii
Thalassia testudinum
Bivalve shell and coral fragments
SUMMARY
Shrimps
Crabs
Other crustacea
Fishes
Other vertebrates
Plant material
Inorganic material
4
1
4
1
1
1
2
1
5
1
71
2
11
1
18
98
181
2
93
1
14
18
Percent
Frequency of
Occurrence
2.4
0.8
1.6
0.8
0.8
0.8
0.8
0.8
2.4
0.8
38.6
0.8
3.1
0.8
7.9
35.4
48.0
1.6
48.8
0.8
4.7
7.9
Percent
of Total
Volume
0.3
0.5
0.2
0.5
<0.1
3.4
1.0
5.3
0.2
43.3
<0.1
0.3
0.2
1.1
9.6
32.3
1.1
55.1
0.2
0.6
1.1
-------
The dominant crustaceans in the stomach contents of the adults were crabs
of the family Portunidae, composed of Callinectes sapidus, C_. ornatus, Portunus
gibbesii, and 1?. sayl. Mud crabs (Xanthidae), represented by Menippe mercenaria
and Neopanope texana, and hermit crabs (Paguridae) formed minor components of
the snook's diet. Portunids are swimming crabs that frequently inhabit pelagic
zones, while xanthids and pagurids are generally associated with the benthos.
Of several shrimp species in the stomachs of adults, the pink shrimp (85
by number, 29.1 percent by occurrence, and 9.1 percent by volume) was the domi-
nant organism. Other identifiable shrimps of minor importance as food included
Palaemonetes intermedius, Alpheus heterochaelis, and Hippolyte pleuracantha.
Incidental food items occurring in the stomach contents of snook were a
crayfish (Procambarus alleni), an isopod, and a tadpole (Rana sp.). Non-food
items consisted of strands of algae and vascular plant material, pieces of
bivalve shells, and coral fragments.
Harrington and Harrington (1961) provided the only known quantitative data
on the food habits of juvenile £. undecimalis. They examined 172 snook, ranging
in size from 5.3 to 51.3 mm FL, from a high salt marsh in northeast Florida
during September and October. They reported that the smallest snook (5.3 to
16.8 mm) fed almost exclusively on copepods (cyclopoids, harpacticoids, and
calanoids) with lesser amounts of mosquito larvae and fishes. Zooplankton com-
prised about 87 percent by volume of the food material in snook of this size
range. In the 18.0 to 28.3 mm size group, fishes accounted for 85 percent of
the total food volume, copepods for 13 percent, and grass shrimp (Palaemonetes
intermedius) for 2 percent. Juveniles between 29.5 and 51.3 mm in length fed
exclusively on fishes (89 percent) and Palaemonetes shrimp (11 percent). Unfor-
tunately, they did not identify the species of forage fishes that the snook had
preyed upon. The observations of Harrington and Harrington that juvenile snook
initially fed on zooplankton, followed by a transitional shift (around 18.0 to
28.3 mm in length) to a piscivorous diet were in agreement with our study.
Qualitative studies on the food of juvenile snook included the investiga-
tion of Harrington and Harrington (1961) and Linton and Rickards (1965). The
latter authors examined the stomach contents of 62 specimens (27.3 to 85.7 mm
FL) from tidal creeks in Georgia. Food items were found in the stomachs of 24
snook. They reported that Palaemonetes^ shrimps accounted for 88 percent .(fre-
quency of occurrence) of the food and Gambusia affinis for the remainder.
Their data were neither in agreement with Harrington and Harrington nor with
the present study which demonstrated that juveniles within the above size range
were predominantly piscivorous.
FOOD HABITS RELATED TO SIZE
Ontogenetic shifts in the food habits of the-snook were evaluated by
grouping the juveniles into 12 size classes between 11.0 and 200.0 mm in length
and the adults into four size classes between 201 and 1,020 mm (Figure XVI-3).
Juveniles of this species exhibited two distinct feeding stages: planktivorous
and carnivorus. Numerous shifts in the contribution (by percent volume) of the
major food items (fishes, shrimps, and crabs) were observed in the carnivorous
diet of the larger juvenile and adult size classes.
Juveniles in the smallest size class (11.0 to 20.0 mm) were exclusively
planktivorous, with copepods (Calanidae) and cladocerans accounting for all of
XVI-15
-------
LU
5
ID
§
fe
I-
Z
LU
O
cr
ID
Q.
100-
80-
60-
40 H
20-
0-1
3/3
6/6
8/8
13/16
80-
60-
40-
20-
0-1
6/11
LQ
11-20
13/17
21-25
16/21
E??^^
26-30
28/39
9/14 18/21 15/21
n
r
41-50
20/40
I
81-90 91-100 101-150 151-200 201-400 401-600 601-800 801-1020
FISHES ^H SHRIMPS E3 ZOOPLANKTON
CRABS
INSECTS
OTHER ITEMS
Figure XVI-3. Percent of total volume of major food items in the stomach con-
tents of 173 juvenile (11-200 mm FL) and 271 adult (201-1,020 mm FL) snook
collected in Ten Thousand Islands, Florida, 1972. Values above each histo-
gram indicate the number of stomachs with food (numerator) over the total
number of snook that were examined (denominator). Size class groups appear
below each histogram.
XVI-16
-------
the food material. A transitional period was apparent at a size of 21.0 to
25.0 mm in which juveniles became quite carnivorous (mainly piscivorous), con-
suming lesser quantities of zooplankton and a greater volume of fishes (Figure
XVI-3). During this transitional period, a variety of food items were present
in the juvenile diet. Microcrustaceans included copepods, ostracods, amphipods,
and insect nymphs. In Figure XVT-3, amphipods were placed in the zooplankton
fraction, since these immature specimens presumably occurred as suspended mate-
rial in the tidal streams. The abruptness of the transitional feeding stage
was indicated by the predominantly carnivorous diet of the snook in the sub-
sequent size class (26 to 30 mm), which was composed of fishes and palaemonid
shrimp with only a trace of zooplankton. The initial forage fishes preyed upon
by juveniles were Gambusia affinis and Cyprinodon variegatus.
As a function of growth, changes in the carnivorous diet of the juveniles
were frequently related to the size of the prey. The larger sized fishes, such
as adult Poecilia latipinna and Lophogobius cyprinoides, were eaten by juve-
niles over 100.0 mm in length, while smaller-sized fishes, such as Gambusia
affinis>and Menidia beryllina, were more frequently preyed upon by juveniles
less than 100.0 mm. Also, the data revealed a change in emphasis in relation
to the shrimps; the smaller palaemonid species were readily consumed by juve-
niles between 26.0 and 100.0 mm; whereas, larger penaeid shrimp formed the bulk
of the diet of juveniles in 101.0 to 200.0 mm size classes.
Relative changes in the summer dietary habits of the adults were shown
among the four size classes (Figure XVI-3). Fishes were the major food source
in three of the four groups, ranging from 27.5 to 70.8 percent by volume.
Crabs, mainly portunids, steadily increased in importance among the adults,
reaching a peak abundance (54.7 percent) in 601.0 to 800.0 mm size class.
Shrimps, primarily Penaeus duorarum, ranked third in importance (by volume) as
food, but were essentially insignificant in the diet of adults in the 801.0 to
1,020.0 mm size class.
FEEDING HABITS
An inquiry into the feeding habits of juvenile and adult snook was based
on the analyses of food habits (Table XVI-6 and 7) and on observations in
nature. The preponderance of food organisms in the diet, e.g., calanoid cope-
pods, cladocerans, swimming crabs (portunids), penaeid shrimp, and poecilid,
atherinid, and sparid fishes, were associated either entirely or at least par-
tially with some portion of the. water column. Some animals, such as the pink
shrimp and the blue crab, that burrow into the substrate are not exclusively
benthic inhabitants, since they spend a portion of their time off the bottom.
In addition, Darnell (1958) indicated that certain organisms (polychaetes,
bivalve molluscs, hydroids, xanthid crabs, and isopods) and the presence of
bottom material (sand grains, mud, and debris) were characteristic items in
the stomach contents of benthic feeding fishes. These items were noticeably
absent or if present, of minor importance in the dietary requirements of the
snook. Thus, analysis of the data indicated that the snook is a pelagic feeder.
A periodicity in feeding intensity occurred in this species, which was
realted to time of day, tidal flow, and habitat. Adults exhibited two peak
feeding periods, as suggested by increased catches (hook-and-line) and by the
condition of food in the stomachs. Feeding intensified in the early morning,
specifically a couple of hours before sunrise, and in the early evening, two
XVI-17
-------
or three hours after sunset. Juvenile snook did not appear to conform to this
pattern.
Minor periods of feeding intensity in adults were associated with changes
in the tidal cycle. A feeding response was triggered, apparently by flowing
waters, following a standing ebb or flood tide. A similar behavior was observed
in the feeding of the larger juveniles.
The predominant feeding grounds of the adults were in narrow passes or
cuts between islands during periods of strong currents. The vulnerability of
fishes, crabs, and shrimps increased as these organisms were swept along in
the flowing waters. Even under artificial lights, feeding was most intense
when a current was present but generally ceased during periods of slack tide.
The feeding of juveniles also intensified in moving waters, which may partially
explain why the narrow confines of tidal streams served as a major habitat for
the young.
xyi-is
-------
XVII - WASTE INVENTORY
INTRODUCTION
The current waste treatment practices and disposal methods for southwest
Florida were inventoried as a stated requirement of the South Florida Ecological
Study. The area of consideration is generally defined as south of Lake
Okeechobee and includes Charlotte, Glades, Hendry, Lee, Collier, Monroe, Dade,
Broward, and Palm Beach Counties. The time budget of the study required that
the scope of the inventory be limited to the southwesternmost counties con-
tiguous with and north of the Big Cypress Swamp. These counties included
Charlotte, Glades, Hendry, Lee, and Collier. Although the counties of Monroe,
Dade, Broward, and Palm Beach were not inventoried, a catalog of inventories
carried out by various groups over the past few years in these counties is
presented in Appendix 00.
Southwest Florida, in particular Charlotte, Lee, and Collier Counties, is
experiencing phenomenal increases in population growth, Obviously, demands
for water and wastewater disposal are increasing proportionally and possibly
at even greater rates. Wastewater disposal is of prime importance because
fresh water is at a premium, groundwater is near the ground surface, and water
percolates rapidly through the overlying sands to groundwater sources.
Current practices of scattered growth lead developers into using small
package wastewater treatment plants. The cost of pumping sewage great dis-
tances into existing wastewater treatment plants if often prohibitive. As
growth increases, so is the demand for more package plants, with an end result
of many small plants serving an area which could be better served by one
medium- to large-sized plant. Once a developer has made an investment in a
treatment plant, it is extremely difficult to convince him to abandon this
investment and to connect onto a central plant.
Although master planning is being carried out in these counties, the con-
cept in many cases is following growth, not leading it. Even after sewage
districts have been established, the pains and delays of bonding and achieving
state and federal funding is often overwhelming.
METHODS
Information for this inventory was obtained from the files of the Florida
Department of Pollution Control (DPC). The current practice of the DPC is to
issue permits to construct and to operate wastewater treatment facilities. A
registered engineer is required to sign these permits. In general, the permits
are quite comprehensive and were certainly most ample for the limited amount of
time that could be spent on the inventory. Treatment facilities not subject to
the permit system are not contained in this report.
Domestic and industrial treatment plants were inventoried as to n-vme,
location, type of waste (industrial only), influent parameters (industrial
only), population served (domestic only), average and design flows, percent
removal of the influent concentrations, type of treatment, receiving stream,
and receiving basin, The waste type was limited to the general categories of
laundry waste, cooling water, dairy waste, etc. Influent parameters and flows
xvn-i
-------
were taken directly from the engineer's estimates contained in DPC permit
system reports. Percent removal of influent parameters was derived from two
sources. The first of these sources is a monthly operating report of each
wastewater treatment plant required by the DPC. This report has no specific
format and is basically a once-a-month log of 5-day biochemical oxygen demand
(BOD5) and suspended solids (SS) removals. The report is predicated upon the
DPC requirement of 90-percent BOD^ and 90-percent SS removals. Approximately
60 to 70 percent of the wastewater treatment plant owners were submitting this
report to the DPC. If this report was not available, then the second source
was, again, the permit system. BOD5 and SS removals were than taken as the
design engineer's estimate, if available, of the plant's efficiency.
Types of wastewater treatment are listed in general terms such as acti-
vated sludge (extended aeration, contact stabilization), trickling filters,
and stabilization lagoons. Chlorination is also listed when it appeared on
the permits.
Effluent discharge points are listed as drainfields (drain tile), specific
and undefined streams, the Gulf of Mexico, total retention ponds, or surface
irrigation.
The final inventoried parameter is the basin receiving the effluent. This
information was compiled by following the water course of the receiving streams.
In general, the ultimate basin is the Gulf of Mexico; but in order to be more
specific, the basins listed are major rivers, passes, and harbors at the ocean
frontage. Seven basins are listed along with undefined tidal canals and also
on-site disposal.
RESULTS
The DPC permitting system has listed for the five counties of Charlotte,
Collier, Glades, Hendry, and Lee a total of 309 wastewater treatment facilities,
of which 265 were domestic waste plants (Appendix PP) and 44 were industrial
waste treatment plants (Appendix QQ). The actual number of plants complying
with the permitting system was unknown. Also, the DPC does not require a per-
mit for domestic wastewater treatment plants that have a designed flow of less
than 3,000 gallons per day (gpd).
Of the 309 listed wastewater treatment plants, 271 (or 87.7 percent) were
of a design capacity of less than 50,000 gpd (Table XVII-1). Typically, the
small plants are extended aeration "package" sewage treatment facilities.*
When these plants reach or exceed their design capacity, they can be converted
to a contact stabilization process which will approximately double the capacity
of the extended aeration process (Figure XVII-1).
In the extended aeration mode, the plants can attain 90-percent BOD5 and
SS removals with a reasonable amount of operational control. However, when
these plants are converted to a contact stabilization mode, greater control is
required to maintain desired efficiencies of treatment. Frequently, the plants
are operated by untrained personnel. Also presented in Table XVII-1 are the
effluent discharge points.
* The designation "package" refers to prefabricated plants readily available
from several vendors.
XVII-2
-------
Table XVTI-1. Wastewatej: treatment facilities in southwest Florida by county
as of January 1, 1973.
Treatment
County Plants
Design Flow
50,000 GPD
or Less
Effluent Discharge Point
Streams
or Gulf
Drainfields
Total Retention
Ponds
Surface
Irrigation
Unknown
INDUSTRIAL WASTE
Charlotte
Collier
Glades
Hendry
Lee
Sub total
DOMESTIC WASTE
Charlotte
Collier
Glades
Hendry
Lee
Sub total
Total
Percentage
of total
5
15
4
2
18
44
50
79
14
10
112
265
309
~
5
15
2
-
16
"38
47
71
14
8
93
233
271
87.7
2
1
1
1
7
12
27
27
5
3
17
~79
91
29.5
.
3
-
-
3
6
10
11
2
1
31
55
61
19.7
_
5
1
1
5
12
10
39
6
5
51
ITT
123
39.8
1
2
-
-
2
5
2
1
-
1
7
11
16
5.2
2
4
2
-
1
~~9
I
1
1
-
6
~9
18
5.8
Total retention (evaporation/
percolation) ponds are the.most
common disposal means, with 40
percent of the plants using this
method. High evaporation and per-
colation rates make these ponds an
acceptable method of effluent dis-
posal. However, increasing land
values in some areas may serve as
a deterrent to implementation of
such disposal methods.
Thirty percent of the treat-
ment facilities discharge their
effluent to streams or to the Gulf
of Mexico. If operational control
becomes lax or plant overloading
takes place, this method of waste-
water disposal may present a
pollutional hazard to many public
beaches and recreational areas.
Ironically, the public beaches
have their maximum use during the
peak tourist season, at which
time the greatest likelyhood of
wastewater treatment plant over-
loading occurs.
Drainfields account for 20
percent of the effluent discharge
points. The final method listed
EXTENDED AERATION
12 HOUR DETENTION
'CHLORINE! EFFLUENT
CONTACT STABILIZATION
HOUR DETENTION
INFLUENT
IDIGESTOR]
HOUR DETENTION V^ /
Figure XVII-1. Schematic diatram of typical
extended aeration and contact stabiliza-
tion modes of "package" treatment plants.
XVII-3
-------
is surface irrigation. Disposal by irrigation is used by 5 percent of the
plants, with the typical disposal source being golf courses.
Seven basins (Figure XVII-2) in southwest Florida receive significant
quantities of effluent discharge. These basins are Lake Okeechobee, Charlotte
Harbor, Caloosahatchee ^iver, Estero Bay, Naples Bay, Marco Pass, and
Chokoloskee Pass (Table XVII-2). Tidal canals refers to undefined or minor
tidal waterways. On-site disposal includes drainfield, evaporation-percolation
ponds, and surface irrigation. Each basin is listed with the number of waste-
water treatment plants (WTP) and their flows, in million gallons per day (mgd),
discharging into them.
Table XVII-2. Receiving basins.
1
Lake
County
INDUSTRIAL
Collier
Lee
Subtotal
Okeechobee
No. of
WTP
2
Flow
(MGD)
1.220
234
Charlotte Caloosahatchee Estero
Harbor River B;
No. of Flow No, of Flow No. of
WTP (MGD ) WTP (MGD ) WTP
- - 5 *572.054 2
1-5 *572.054 2
ly
Flow
(MGD)
0.011
0.011
567
Naples Marco Chokoloskee
Bay Pass Pass
No. of Flow No. of Flow No. of Flow
WTP (MGD) WTP (MGD) WTP (MGD)
------
1 0.010
8 9
Tidal On Site
Canals Disposal
No. of Flow No. of
WTP (MGD) WTP
- - 8
10
1 - 21
Flow
(MGD)
0.086
0.067
3.215
DOMESTIC
Charlotte
Collier
Hendry
Lee
Subtotal
TOTAL
* Includes
-
3 0.
_
6 0.
8 1.
571.4 MGD
-
313
-
341
561
26
-
26
27
circulation
2.
2.
2.
438
-
,438
.438
of Caloo
-
13
13
18
sah;
-
11.
11.
*583.
itchee
-
933 4
933 4
987 6
River for
:
0.060
0.060
0.071
cooling
9 2.945
- -
9 2.945
9 2.945
water.
1 0.015 22 0.283
3 0.029 2 0.115 15 0.281 (49 1.416
~ ~ - - - - 89 3.473
3 0.029 2 0.115 18 0.334 175 5.431
3 0.029 3 0.125 19 0.334 196 8.646
The basin, other than tidal canals or on-site disposal, receiving the
largest number of waste treatment plant discharges is Charlotte Harbor. This
harbor receives effluent from 27 plants which have a total design flow of 2.4
mdg. The Caloosahatchee River received the greatest wastewater flow (12.6 mgd
not including cooling water discharge) from 18 plants.
On-site disposal is employed at two-thirds of the plants, not including
one 3-mgd industrial waste plant. The average flow of these 195 plants, which
use on-site disposal, is less than 30,000 mgd.
XVI1-4
-------
N
EXPLANATION
MAJOR EFFLUENT BASINS
LAKE OKEECHOBEE
CHARLOTTE HARBOR
CALOOSAHATCHEE RIVER
ESTERO BAY
NAPLES BAY
MARCO PASS
CHOKOLOSKEE PASS
Figure XVII>2, Major wastewater effluent basins in southwest Florida.
XVII-5
-------
XVIII - REFERENCES
Alexander, T. R. 1967. A tropical hammock on the Miami (Florida) limestone -
A twenty-five-year study. Ecol. 48: 863-867.
Bailey, Reeve M., John E. Pitch, Earl S. Hearld, Ernest A, Lachner, C. C. Lindsey,
C. Richard Robins, and W. B. Scott. 1970. A list of common and scientific
names of fishes from the United States and Canada. (Third Edition). Amer.
Fish. Soc., Spec. Publ. No. 6. 149 pp.
BShlke, James E. and Charles C. G. Chaplin. 1968. Fishes of the Bahamas and
adjacent tropical waters. Livingston Publ. Co., Wynnewood, Pa. 771 pp.
Braidech, T., P. Gehring, and C. Kleveno. 1972. Project Hypo - An intensive
study of the Lake Erie Central Basin hypolimnion and related surface water
phenomena, 1972. U.S. EPA Tech. Rep. TS-05-208-24. 182 pp.
Butler, Philip A. 1966. Pesticides in the marine environment. J. Appl. Ecol.
3 (suppl.): 253-259.
Carriker, Melbourne R, 1967. Ecology of estuarine benthic invertebrates: A
prospective, pp. 442-487. In; Estuaries, Edited by George H. Lauff.
Amer. Assoc. Adv. Sci., Publ. No. 83: 757 pp.
Caulfield, P. 1970. Everglades. Sierra Club, San Francisco. 143 pp.
Collette, Bruce B. 1968. Strongylura timucu (Walbaum): A valid species of
western Atlantic needlefish. Copeia 1968 (1): 189-192.
Cope, Oliver B. 1966. Contamination of the freshwater ecosystem by pesticides.
J. Appl. Ecol. 3 (suppl.): 33-43.
Costello, T. J. and D. M. Allen. 1966. Migrations and geographic distribution
of pink shrimp, Penaeus duorarum, on the Sanibel and Tortagus grounds,
Florida. U.S. F&WS, Fish. Bull. 66(3): 491-502.
Cralghead, F. C., Sr. 1963. Orchids and other air plants of the Everglades
National Park. Univ. Miami Press, Coral Gables, Fla.
Craighead, F. C., Sr. 1971. The trees of south Florida. Vol. 1, The natural
environments and their succession. Univ. Miami Press, Coral Gables, Fla.
212 PP.
Croker, Robert A. 1962. Growth and food of the gray snapper, Lutjanus griseus,
in Everglades National Park. Trans. Amer. Fish. Soc. 91(4): 379-383.
Dahlberg, Michael D. 1970. Atlantic and Gulf of Mexico menhadens, genus
Brevoortia (Pisces: Clupeidae). Bull. Fla. St. Mus., Biol. Sci. 15: 91-162.
Darnell, Rezneat M. 1958. Food habits of fishes and larger invertebrates of
Lake Pontchartrain, Louisiana, an estuarine community. Publ. Inst. Mar.
Sci., Univ. Tex. 5: 353-416.
XVIII-1
-------
Davis, J. H,, Jr. 1943. The natural features of southern Florida. Geol.
Survey Bull. No. 25, Dept. of Conservation, Tallahassee, Fla.
Davis, J. H., Jr. 1946. The peat deposits of Florida. Geol. Survey Bull. No.
30, Dept. of Conservation, Tallahassee, Fla.
Department of the Army, Jacksonville District Corps of Engineers. February 15,
1968. Water resources for central and southern Florida. U.S. GPO,
Jacksonville, Fla. Main Report.
Egler, F. E. 1952. Southeast saline everglades vegetation, Florida, and its
management. Vegetatio acta Geobotanica 3: 213-265.
Emery, K. 0. and R. E. Stevenson. 1971. Estuaries and lagoons - Treatise on
marine ecology and paleoecology. Volume 1, Ecology, p. 679.
Everglades Jetport Pact, The. January 16, 1970. Articles of agreement by and
between the United States, State of Florida, Dade County Port Authority,
Collier County.
Evermann, B. W. and W. C. Kendall. 1900. Checklist of the fishes of Florida.
Rep. U.S. Comm. Fish. 23: 37-103.
Fair, G. M., J. C. Geyer, and D. A. Okum. 1966. Water and wastewater engineer-
ing. Vol. 1, Water supply and wastewater removal. John Wiley & Sons, Inc.,
New York.
Fassett, N. C. 1940. A manual of aquatic plants. Univ. Wis. Press, Madison,
Wis.
Finney, S. N., Jr. and J. B. Miller. 1961. Some physical and chemical char-
acteristics of selected Florida waters. Fla. St. Bd. Health, Jacksonville.
108 pp.
Focarty, Michael, Jr. and J. David Albury. 1967. Late summer foods of young
alligators in Florida. 21st Ann. Conf. S.E. Assn. Game and Fish Comm.
pp. 220-222.
Fraser, T. H. 1968. Comparative osteology of the Atlantic snooks (Pisces,
Gentropomus). Copeia 1968(3): 433-460.
Futch, Charles R. 1966. Lisa - The Florida black mullet. Fla. St. Bd. Consrv.,
Salt Water Fish. Leafl. 6: 1-6.
Futuyma, D. J. 1973. Community structure and stability in constant environ-
ments. Amer. Nat. 107: 443-446.
Gee & Jenson Consulting Engineers, Inc. 1970. Report covering review of water
management systems of GAG Properties, Inc. in Collier and Lee Counties,
Florida; West Palm Beach, Florida.
Gunter, Gordon and Gordon E. Hall. 1965. A biological investigation of the
Caloosahatchee estuary of Florida. Gulf Res. Rep. 2(1): 1-72.
XVII1-2
-------
Harrington, Robert W., Jr. and Eleanor S. Harrington, 1961. Food selection
among fishes invading a high subtropical salt marsh; From onset of flood-
ing through the progress <£ a^mosquito brood. Ecol. 42(4): 646-666.
Hawkins, Mary Ellen. June 18, 1972. Conservationists versus developers - Can
the twain meet to achieve balance growth. The Naples Daily News, p. 2.
Heald, E. J. 1971. The production of organic detritus in a south Florida
estuary. Sea Grant Tech. Bull. No. 6, Univ. Miami. 110 pp.
Hela, I. 1952. Remarks on the climate of southern Florida. Bull. Mar. Sci.
Gulf and Carib. 2: 438-447.
Hitchcock, A. S. 1951. Manual of the grasses of the United States, Vol. 1
and 2. 2nd Ed., revised by Agnes Chase. TJSDA Misc. Publ. 200, Washington,
D.C.
Hoese, H. D. and R. S. Jones. 1963. Seasonality of larger animals in a Texas
turtle grass community. Publ. Inst. Mar. Sci. Univ. Tex. 9: 37-47.
Houde, Edward D. and Paul L. Fore. 1973. Guide to identity of eggs and larvae
of some Gulf of Mexico clupeid fishes. Fla. Dept. Nat. Resources, Mar.
Res. Lab. Leafl. Ser, Vol. IV, Pt. 1, No. 23. 14 pp.
Hughes, G. H., E. R. Hampton, and D. F. Tucker. 1971. Annual and seasonal
rainfall in Florida. USGS Map Ser. No. 40.
Hunt, Burton P. 1953. Food relationships between Florida spotted gar and other
organisms in the Tamiami Canal, Bade County, Florida. Trans. Amer. Fish.
Soc. 82: 13-33.
Kahl, M. P., Jr. 1962. Bioenergetics of growth in nestling wood storks.
Condor 64: 169-183.
Kahl, M. P., Jr. 1964. Food ecology of the wood stork (Mycteria americana) in
Florida. Ecol. Mong. 34: 97-117.
Keith, James 0. 1966. Insecticide contaminations in wetland habitats and their
effects on fish-eating birds. J. Appl. Ecol. 3 (suppl.): 71-85.
Keup, L. E. 1968. Phosphorus in flowing water. .Water Res. 2: 373.
Klein, Howard. 1954. Ground water resources of the Naples area, Collier County,
Florida. Fla. Geol. Survey Rep. of Investigation No. 11. 63 pp.
Klein, H., at al. May 1970. Some hydrologic and biologic aspects of the Big
Cypress Swamp drainage area, southern Florida. Openfile Rep. 70003. USGS,
Tallahassee, Fla.
Koppen, W. 1931. Grunduss der Klimakunde. Walter de Gruyter & Co., Berlin.
Kuenzler, E. J. 1961. Phosphorus budget of a mussel population. Limnol.
Oceanogr, 6: 400-415,
XVIH-3
-------
Kuenzler, E, J. 1961$. Structure and energy flow of a mussel population.
Limnol. Oceanogr. 6: 191^204.
Lagler, Karl F. 1956. freshwater fishery biology. Wm. C. Brown Co., Dubuque,
la. 421 pp.
Lear, D. W. and S. K. Smith. 1972. Analysis of phytoplankton standing crop.
In; Port Royal Sound Environmental Study, pp. 217-223, 502-518. S.C.
Water Res. Comm.
Leighty, R. G., M. B. Marco, G. A. Swenson, et_ al_. 1954. Soil survey (detailed
reconnaissance) of Collier County, Florida. USDA, SCS and Fla. Ag. Exp.
Sta. Soil Survey Ser. 1942 No. 8, Washington, B.C. 72 pp. and maps.
Leonards, G. A. 1962. Foundation engineering. McGraw Hill, Inc., New York.
p. 134.
Linton, Thomas L. and William L. Rickards. 1965. Young common snook on the
coast of Georgia. Quart. J. Fla. Acad. Sci. 28(2): 185-189.
Long, R. W. and L. Lakela. 1971. A flora of tropical Florida. Univ. Miami
Press, Coral Gables, Fla.
Longley, W. H. and S. F. Hildebrand. 1941. Systematic catalogue of the fishes
of Tortugas Florida. Carnegie Inst. Wash. Publ. 535. 331 pp., 134 pis.
Lonnberg, A. J. E. 1894. List of fishes observed and collected in south
Florida. Ofvers. Kougl. Akad. Forh. 3: 109-131.
Lucas, A. M. and N. A. Thomas. 1970. Sediment oxygen demand in Lake Erie's
central basin. In; Project Hydro - An intensive study of the Lake Erie
central basin hypolimnion and related surface water phenomena. 1972. Ed.
by Burns, N. M. and C. Ross. U.S. EPA, Tech. Rep. TS-05-71-208-24. 182 pp.
Lugo, A. E., G. Evink, M. M. Brinson, A. Broce, and S. C. Snedaker. 1973.
Diurnal rates of photosynthesis, respiration, and transpiration in mangrove
forests of south Florida. In press. Dept. Bot., Univ. Fla.
Lugo, A. E., S. C. Snedaker, S. Bayley, and H. T. Odum. 1971. Models for plan-
ning and research for the south Florida environmental study. Univ. Fla.,
Gainesville. 123 pp.
Lunz, G. R. 1953. First record of the marine fish Centropomus undecimalis in
South Carolina. Copeia 1953(4): 240.
Mackenthun, K. M. 1965. Nitrogen and phosphorus in water. U.S. DHEW, PHS,
Div. Water Supply and Pollution Cont. Ill pp.
Marshall, Arthur R. 1958. A survey of the snook fishery of Florida, with
studies of the biology of the principal species, Centropomus undecimalis
(Bloch). Fla. St. Bd. Conserv., Tech. Ser. No. 22. 39 pp.
Martin, James R. and Robert L. Shipp. 1971. Occurrence of juvenile snook,
Centropomus undecimalis, in North Carolina waters. Trans. Amer. Fish. Soc.
100(1): 131-132.
XVHI-4
-------
McFarland, William N, 1963. Seasonal change in the number and the biomass of
fishes from the surf at Hustang Island, Texas. Publ. Inst. Mar. Sci.,
Univ. Tex. 9; 91-105,
Mcllhenny, E. A. 1934. Notes on alligators. Copeia 2: 85-86.
McKee, Jack Edward and Harold W, Wolf. April 1971. Water quality criteria.
2nd Ed. Calif. St. Water Resrcs. Cont. Bd. Publ. 3-A. 548 pp.
Miller, Robert R. 1945. Hyporhamphus patris, a new species of hemiramphid
fish from Sinaloa, Mexico, with an analysis of the generic characters of
Hyporhamphus and Hemiramphus. Proc. U.S. Nat. Mus. 96(3195): 185-193.
Moffett, A. W. 1961. Movements and growth of spotted seatrout, Cynoscion
nebulosus (Cuvier), in west Florida. Fla. St. Bd. Conserv. Tech. Ser. 36:
1-35.
Muenscher, W. C. 1944. Aquatic plants of the United States. Comstock Publish-
ing Assoc., Cornell Univ. Press, Ithaca, N.Y.
O'Connell, Timothy R., Jr. 1958. Experimental electrical fishing gear develop-
ment. Fla. Dept. Game & Freshwater Fish (mimeo).
Odum, E. P. 1971. Fundamentals of ecology. W. B. Saunders Co., Philadelphia,
Pa. 3rd Ed, 574 pp.
Odum, E. P. and de la Cruz. 1967. Particulate organic detritus in a Georgia
salt marsh-estuarine ecosystem. In; Estuaries. Amer. Assn. Advancement
Sci., Washington, D.C. pp. 383-388.
Odum, H. T. 1963. Productivity measurements in Texas turtle grass and the
effects of dredging an intracoastal channel. Univ. Tex. Inst. Mar. Sci.
9: 48-58.
Odum, H. T. 1967a. Biological circuits and the marine systems of Texas. In;
Pollution and marine ecology, pp. 99-157. Olson, T. A. and F. J. Burgess,
Eds. Interscience Publishers, New York. 364 pp.
Odum, H. T. 1967b. Energetics of world food production. In: The world food
problem, Vol. 3, pp. 55-94. U.S. GPO, Washington, D.C. 332 pp.
Odum, H. T. 1968. Work circuits and system stress. In; Primary productivity
and mineral cycling in natural ecosystem, pp. 81-138. Univ. Maine Press,
Orono, Me.
Odum, H. T. 1971. Environment, power, and society. John Wiley & Sons, Inc.,
New York. 331 pp.
Odum, H. T. 1972. An energy circuit language for ecological and social systems:
its physical basis. In; Systems analysis and simulation in ecology, Chapt.
4, (II) BC Patten, Ed. Academic Press.
Odum, H. T., C. Littlejohn, and W. C. Huber. 1972. An environmental evaluation
of the Gordon River area of Naples, Florida, and the impact of developmental
plans. Rep. to Co. Comm. of Collier Co., Fla. 101 pp. (mimeo).
xyin-5
-------
Odum, H. T,, A. Lugo, and L, A. Burns. 1970. Metabolism of forest floor
microcosms. In; Odum, H, T. (Ed.), A tropical rain forest, Chap. 1-3.
USAEC Div, Tech, Info., Washington, D.C.
Odum,H. T. and R. C. Pinkerton, 1955. Time's speed regulator: The optimum
efficiency for -maximum power output in physical and biological systems.
Amer. Sci. 43: 331-345.
Odum, William E. 1970. Insidious alteration of the estuarine environment.
Trans. Amer. Fish. Soc. 99(4): 836-847.
Odum, W. E. 1972. Pathway of energy flow in a south Florida estuary. Sea
Grant Tech. Bull. No. 7. 162 pp.
Odum, William E. and Eric Heald. 1972. Tropic analyses of an estuarine man-
grove community. Bull. Mar. Sci., Gulf and Carib. 22(3): 671-738.
O'Hara, J. 1972. The Port Royal Sound ecosystem - A review. In; Port Royal
Sound Environmental Study, pp. 305-310. Port Royal Sound Environ. Study,
S.C. Water Rercs. Comm. 555 pp.
Oosting, H. V. 1956. The study of plant communities, 2nd Ed. W. H. Freeman
& Co., San Francisco, Galif.
Parker, D. C., D. P. Norris, and A. W. Nelson. April 1972. Tidal exchange at
Golden Gate. J. San. Engr. Div., Amer. Soc. Civil Engr. pp. 305-309.
Parker, G. G., G. E. Ferguson, and S. K. Love. 1955. Water resources of
southern Florida. USGS Water Supply Paper 1255, Washington, D.C. 965 pp.
Pearse, A. S. and Gordon Gunter. December 30, 1957. Treatise on marine ecology
and paleoecology. In: Treatise on Marine Ecology, Chap. 7. pp. 149-150.
Penman, H. L. 1948. Natural evaporation from open water, bare soil, and grass.
Proc. Royal Soc.', Series A.
Phillips, R. C. I960. Observations on the ecology and distribution of the
Florida seagrasses. Prof. Paper, Series No. 2, Fla. St. Bd. Consr., Mar.
Lab., St. Petersburg, Fla. 72 pp.
Phillips. Ronald C. and Victor G. Springer. 1960. A report on the hydrography,
marine plants, and fishes of the Caloosahatchee River area, Lee County,
Florida. Fla. St. Bd. Consr. Spec Sci. Rept. No. 5. 34 pp.
Pomeroy, L. R., R. E. Johannes, E. P. Odum, and B. Roffman. 1969. The phos-
phorus and zinc cycles and productivity of a salt marsh. In; Proc. 2nd
Sym. on Radioecology. D. J. Nelson and F. C. Evans, Eds. Clearinghouse
Fed. Sci. Tech. Info., Springfield, Va. pp. 412-419.
Pope, J. A. 1966. Manual of methods for fish stock assessment. Part III.
Selectivity of fishing gear. FAO Fish. Tech. Paper No. 41. 50 pp.
Puri, Harbans S. and Robert 0. Vernon, 1964. Summary of the geology of Florida
and a guidebook to the classic exposures. Spec. Publ. No. 5, Fla. Geo. Sur.,
Tallahassee, Fla.
XVIII-6
-------
Rivas, Luis R. 1962. The Florida fishes of the genus Centropomus, commonly
known as snook. Quart. J. Fla. Acad. Sci. 25(1): 53-64.
Roessler, Martin A. 1970, Checklist of fishes in Buttonwood Canal, Everglades
National Park, Florida, and observations on the seasonal occurrence and
life histories of selected species. Bull. Mar. Sci. 20: 860-893.
Sawyer, Clair N. and Percy L. McCarty, 1967. Chemistry for sanitary engineers,
2nd Ed. McGraw-Hill, Inc., New York, N.Y. 518 pp.
Shipp, Robert L, and Ralph W, Yerger. 1969. A new puffer fish, Sphoeroides
parvus. from the western Gulf of Mexico, with a key to sepcies of
Sphoeroides from the Atlantic and Gulf coasts of the United States. Proc.
Biol. Soc. Wash. 82: 477-488.
Sleight, R. B. 1917. Evaporation from the surfaces of water and riverbed
materials. J. Agr. Res. 10.
Slobodkin, L. B. and H. L. Sanders. 1969. On the contribution of environmen-
tal predictability to species diversity. Brookhaven Symp. Biol. 22: 82-93.
Small, J. K. 1918. Ferns of Royal Palm Hammock. Published by the author, New
York, N.Y.
Small, J. K. 1918. Ferns of tropical Florida.
Small, J. K. 1933. Manual of the southeastern flora. Univ. N.C. Press, Chapel
Hill, N.C.
Snedecor, G. W. 1956. Statistical methods applied to experiments in agricul-
ture and biology. Iowa St. Univ. Press, Ames, la. 534 pp.
Soil Conservation Service, USDA. March 1954. Soil survey - Detailed reconnais-
sance, Collier County, Florida. Series 1942, No. 8, p. 3.
Springer, Victor G. and Kenneth D. Woodburn. 1960. An ecological study of the
fishes of the Tampa Bay area. Fla. St. Bd. Consvr. Prof. Paper Ser. No. 1,
104 pp.
Starck, Walter A., II. 1968. A list of fishes of Alligator Reef, Florida, with
comments on the nature of the Florida reef fish fauna. Undersea Biol.
1(1): 4-40.
Starck, W. A., II. 1968. Fishes of Alligator Harbor, Florida. Undersea Biol.
1(1): 18.
Stephens, J. C. and W. H. Speir. 1969. Subsidence of organic soils in the
U.S.A. Assn. Int. d'Hydrologic Sci. Publ. No. 89, Colloque de Tokyo. 1969.
Stewart, E. N. and W. C. Mills. 1967. Effect of depth to water table and plant
density on evapotranspiration rate of southern Florida. Trans. Amer. Soc.
Ag. Engrs. 10(6): 746-747
XVIII-7
-------
Stewart, Kenneth W. 1961, Contribution to the biology of the spotted sea-
trout, Cynoscion nebulosus, in the Everglades National Park, Florida.
M.S. Thesis, Univ. Miami, Coral Gables, Fla, 103 pp.
Storey, Margaret. 1937. The relation between normal range and mortality of
fishes due to cold at Sanibel Island, Florida. Ecol. 18(1): 10-26.
Storey, Margaret and E. W. Gudger, 1936. Mortality of fishes due to cold at
Sanibel Island, Florida, 1886-1936. Ecol. 17(4): 640-648.
Sykes, James E. and John H. Finucane. 1966. Occurrence in Tampa Bay, Florida,
of immature species dominant in Gulf of Mexico commercial fisheries. U.S.
F&WS Fish. Bull. 62(2): 369-379.
Tabb, Durbin C. and Raymond B. Manning. 1961. A checklist of the flora and
fauna of northern Florida Bay and adjacent brackish waters of the Florida
mainland collected during the period July 1957 through September 1960.
Bull. Mar. Sci., Gulf and Carib., 11(4): 552-649.
Taylor, D. W. 1948. Fundamentals of soil mechanics. John Wiley & Sons, Inc.,
London. PP. 24-25 and 152-154.
Thomas, N. A. 1972. A study of the primary productivity of Port Royal Sound.
In; Port Royal Sound Environmental Study, pp. 213-215 and 494-500. S.C.
Water Rsrcs. Comm. 555 pp.
Thomas, Terence M. October 1970. A detailed analysis of climatological and
hydrological records of south Florida with reference to man's influence
upon ecosystem evaluation. Univ. Miami, Rosenstiel School Mar. & Atm. Sci.
Thorhaug, A. and R. D. Stearns. 1973. An ecological study of Thalassia
testudium in unstressed and thermally stressed esutaries. Manuscript sub-
mitted to Ecol. School of Med., Univ. Miami, Miami, Fla. 30 pp.
Thornthwaite, C. W. 1948. An approach toward a rational classification of
climate. Geog. Rev. 38(1).
Thornthwaite, C. W. 1951. The water balance in tropical climates. Bull.
Amer. Meteoral. Soc. 32: 166-173.
Todd, D. K. 1959. Ground water hydrology. John Wiley & Sons, Inc., New York.
336 pp.
U.S. Department of the Interior. September 1969. Environmental impact of the
Big Cypress Swamp Jetport.
U.S. Public Health Service. 1962. Drinking water standards. PHS Publ. No. 62,
U.S. DREW, Washington, D.C.
Volpe, Alfred V. 1959. Aspects of the biology of the common snook Centropomus
undecimalis (Bloch) of southwest Florida. Fla. St. Bd. Consrv. Tech. Ser.
No, 31. 37 pp.
Wade, Richard A. 1969. Ecology of juvenile tarpon and effects of dieldrin on
two associated species. U.S. F&WS Tech. Paper 41: 1-85.
XVIII-8
-------
Waldinger? F. J. 1968, Relationships of environmental parameters and catch
of three species of the mojarra family (Gerreidae), Eucinostomus gula,
Eucinost omus argeriteus, and Diapterus plumieri, collected in 1963 and 1964
in Buttonwood Canal, Everglades National Park, Florida. M.S. Thesis, Univ.
Miami, Coral Gables, Fla. 68 pp.
Walter, H. and H. Leith. 1960-1967. Klima diagramm Weltatlas. V. B, Gustav
Fisher Verlag. Jena.
Wang, Johnson C. S. and Edward C. Raney. 1971. Distribution and fluctuations
in the fish fauna of the Charlotte Harbor estuary, Florida. Mote Mar. Lab.,
Sarasota, Fla. 56 pp and Appendix, 38 pp.
Westlake, D. W. 1971. A manual on methods for measuring primary production in
aquatic environments. In: IBP Handbook No. 12, R. A. Vollenweider, Ed.,
Blackwell Scientific Publications, Oxford and Edinburgh. 213 pp.
White, W. N. 1932. A method of estimating ground water supplies based on dis-
charge by plants and evaporation from soil. USGS Water Supply Paper No.
659, Washington, D.C. 105 pp.
Yokel, Bernard J. 1966. A contribution to the biology and distribution of the
red drum, Sciaenops ocellata. M.S. Thesis, Univ. Miami, Coral Gables, Fla.
160 pp.
XVIII-9
-------
APPENDIX A - METHODS OF ECOSYSTEM ANALYSIS
The energy circuit language developed by H. T. Odum (Odum and Pinkerton
1955, Odum 1967a, 1967bs 1968, 1970, 1971, 1972) and concepts of systems ecology
developed over the past several decades (e.g. Patten 1972 and contained
references) by many investigators formed the methodological basis of our
approach to the analysis of the Fahkahatchee ecosystem. In the following
discussion we shall recapitulate a few of the main trends of this branch of
ecology as a means of making our methodological biases explicit. Our additional
goal here is to clarify the logic underlying the inferences we draw from the
data concerning the overall structure and functioning of the Fahkahatchee
ecosystem.
An ecosystem can be defined for analytical purposes as a set of functional
components, biotic and abiotic, which are coupled by flows of energy and materials
whose magnitude, direction, and functional form are amenable to direct investiga-
tion. Within this definition there is a major disparity between energetic
coupling and mineral flow coupling that must be borne firmly in mind. It is
summarized by the text book distinction that minerals circulate in ecosystems;
but the flow of energy is one-way through the system. Incoming energy is
ultimately dissipated in heat. A major area of investigation in systems
ecology is embedded within this definition, that of the nature of the inter-
action between these two apparently dissimilar processes.
The flow of energy and materials in ecosystems is roughly parallel. For
example, the transfer of energy along food chains occurs via the consumption of
high-energy materials contained in the food. The concentration of mineral
nutrients in the soil system is paralleled by a dissipation of energy in the
detrital food chain. This interaction of energy flow and mineral cycling
serves to organize collections of plant and animal populations into ecosystems:
the population gene pools act as an information storage 'that establishes and
maintains the range of successful interactions we see operating in ecosystems
today.
Transformation of energy into useful work is the basic process governing
the operation of all systems. Artificial systems serving man's purposes rely on
the ingenuity of the engineer to establish useful pathways of energy flow, while
natural systems enploy selective pressures acting over a lengthy time span to
achieve what may be similar end results. These may include the development of
processes that enhance long-term stability, some set of particular resource
utilization rates with a balance of throughput and recycling, and progressively
greater internal control of environmental variables. In both cases the laws of
thermodynamics establish the boundary conditions that constrain system operation.
These laws help form the underlying guidelines for constructing system circuit
diagrams for ecosystem analysis.
In brief, the first (conservation) law imposes a requirement that the
energy budget be balanced, i.e., it is not possible for any incoming energy to
simply disappear. This is not so trivial a point as it appears, for it serves
as a means of keeping track of the completeness of a system definition: if
major terms of the energy budget do not appear in the model, this can be taken
as an indication that the circuit is not yet complete. The degradation of
potential energy to dispersed forms (second thermodynamics law) accompanies
every energy transformation. This forces systems to treat the utilization of
XIX-1
-------
potential as a prime resource. Perhaps a profitable way of viewing the second
(entropy) law in this context is that potential represents the opportunity to
perform useful work, and some of this opportunity is irretrievably expended
each time it is employed.
These two considerations make possible the calculation of balanced
energy budgets for whole systems, and lend assurance that the opportunity to
transform energy to useful work must be apportioned in a systematic, regular
fashion. The loss of work opportunity with each transformation forces the
system to treat this eminently exhaustible commodity as its most precious
resource. Moreover, it assures us that there is scant likelihood that any
substantial energy consuming process is incidental to basic system function.
Natural systems subsidize the operation of human systems in many ways.
The subsidies range from the generation in past ages of the fossil fuels that
power today's technological revolution to the provision of the multitude of
services subsumed under the classical economic definition of a "free good".
We must include in this category the esthetic and cultural values derived by
the individual from recreational opportunities afforded by access to green
space and wilderness.
Estimation of the value of these subsidies, and the development of means
of interweaving the strands of the social and ecological fabric to enhance
their mutual survival, are still in their infancy. Too often the valuation
of natural systems is based merely on the dollar cost to man of operations
that serve to exploit a resource. The primary expenditures of energy that
originally produced and continually maintain the resource base are usually
forgotten.
The circulation of mineral elements in ecosystems may act as a feedback
governor that helps control the direction and speed of energy flow. A similar
mechanism may be at work in economic systems, insofar as currency circulation
may be said to effect changes in potential work opportunities. This feedback
effect lets us define an energetic value for minerals and nutrients in terms
of the energy flows they control. A similar treatment of dollar flow and
energy flow may be possible as well. Dollar flow and mineral flow are in any
case both accompaniments to the flow of energy and its dissipation in the
course of the useful work that has developed and maintains the world as we know
it. This parallelism allows preliminary comparisons of natural and artificial
systems to be made in terms of a common measure of value. For example, a
quantitative equivalence between currency and energy can be defined from a ratio
of kilocalories of energy to the U. S. dollar, derivable from available infor-
mation on the annual real gross national product and total yearly power
consumption in the U. S. economy.
Photosynthesis, respiration, and matter transport are the basic processes
of ecosystems that account for energy fixation, utilization, and transformation;
and for the cycling of crucial limiting minerals. The primary productivity of
the ecosystem represents the energy available to operate all the energy consuming
processes and biological work powered by organic respiration throughout the
system. For this reason primary consideration must be given to direct measure-
ment of gross photosynthesis as a most important system parameter; for gross
primary productivity establishes the magnitude of the organic energy budget
available for all other purposes. Ecosystems may be coupled to auxiliary
energy sources as well, and in some cases may be totally dependent on them
XIX-2
-------
(for example, benthic detrital systems). Auxilliary energy sources provided by
physical and chemical environmental processes may reduce the quantities of
energy the system expends in maintenance, and are often of sufficient magnitude
to produce selective pressures that create ecosystems specifically adapted to
utilize them. In general, these subsidies may make additional energies avail-
able for increased complexity of biological organization or increased net
productivity. At the other end of the scale, physical processes often produce
conditions that require a commitment of energy resources to counter destructive
effects - this situation serves to define the concept of a "stress" in energetic
terms.
A word is in order here concerning the modular components used in the
energy circuit language for constructing compartmental models of ecosystems.
A set of 15 basic symbols of this language is given in Figure A-l. Each module
(a)
OUTPUT
PASSIVE
STORAGE
(b)
CONTROL
FACTOR
HEAT
SINK
WORK GATE
INPUT [ JOUTPUT
A HEAT
w SINK
SELF-
MAINTENANCE
CONSUMER
(d)
INPUT
OUTPUT
,HEAT
SINK
CYCLING
RECEPTOR
,NPUT|STORA6E
NPUTjSTC
JLHEAT
SINK
ACTIVE
STORAGE
(f)
(9)
INPUT
JOUTPUT
OUTPUT
SINK
GREEN
PLANT
POPULATION
SOURCE
(I)
(i)
DRIVING
FORCE
STRESS
HEAT
SINK
TWO-WAY
GATE
. HEAT
SINK
MISCELLANY
(fc)
STEADY
STATE FLOW
(o)
CONSTANT
GAIN
PATHWAY
HEAT SINK
SWITCH
CONTROL
THRESHOLD
CONTROL
ECONOMIC
TRANSACTION
Figure A-l. Modules of the energy circuit language. These modules are used as
block components in constructing compartmental models of ecosystems.
Association of mathematical functions with compartment interactions leads
to computer simulation of ecosystem dynamics. Details of these mathematical
relationships and derivation of some interaction terms can be found in Odum
(1972). Figure is redrawn from Lugo, e_t al. (1971).
XIX-3
-------
represents a particular cluster of similar ecological entities, and each has a
particular group of associated mathematical functions. Most basic ecological
processes can be portrayed by combinations of these operators, which are defined
as follows:
la)
INPUT
OUTPUT
PASSIVE
STORAGE
(a). The passive storage module stores potential
energy as the balance of its inputs and outputs. A
complete definition requires a mathematical expression
for the quantity of potential energy stored, and an
expression of storage functions for establishing that
quantity. The expression for the outflow usually is
a function of the circuit connections as well as the
quantity stored, that is, the flow of energy from
the storage will depend on the process to which it is
coupled. The input functions may also be of several
kinds. For example, water stored in a tank against
gravity has an energetic value related to its
hydrostatic head; energy in compressed gas is
related to pressure by a logarithmic function. These storages are state
variables of the system, and may represent storages of water, chemicals,
biological populations, air, money, information, etc. In an environmental
system, with its great diversity-of possible utilization pathways, there are
losses inherent in any storage. Almost any concentration of a quantity
constitutes an incipient energy source by virtue of its low entropy, and tends
to be used by one agency or another. For example, a storage of food tends to
develop chemical reactions, populations of consumer insects, etc. Work is
done to protect the source if containers or some other means are used to
insulate an energy source from losses. This is, in effect, a drain on the
source. Leaf litter dynamics in forest ecosystems is an example of a process
describable by this module. The input to forest floor litter accumulation, a
storage, is given by the simple sum of leaf-fall rates from the trees. Litter
decomposition dynamics are adequately described by a rate equation based on the
quantity of litter available to decomposer populations (see Chapter XII for an
example of this application).
(b)
(b). The "work gate" represents a control
function on the transmission characteristics of
pathways of energy flow. The action of the control
factor modifies the input-output transfer of energy
through the block by regulating the conductivity of
the contained pathway. For example, the transport
rate of carbon dioxide gas to plants from atmospheric
reservoirs is strongly controlled by wind driven air
circulation in the forest. If the control agent is
in short supply, the controlling factor can be
represented as a passive storage. The dynamic
behavior of the work-gate module can be shown to
be a limiting factor hyperbola under these conditions (Odum, 1968). For example,
photosynthetic production under phosphorus-limited conditions is controlled by
the rate of supply of phosphorus to the plant community. Phosphorus supply may
be governed by the rates of accretion from exogenous sources or in situ
recycling of previously generated materials.
HEAT
SINK
WORK GATE
XIX-4
-------
(c) The "self maintaining consumer population"
symbol representa a combination of "active storage"
(Figure e) and a work-gate (Figure b). In the
population module, potential energy stored within
the population is utilized to maintain a continuing
input of new potential for growth and reproduction.
Several formulations of the detailed operation of
this module lead naturally to mathematical functions
that exhibit "logistic" properties, i.e. populations
that grow rapidly at first and subsequently reach
a stable plateau of population size.
(0
INPUT
OUTPUT
HEAT
SINK
SELF~
MAINTENANCE
CONSUMER
(a)
INPUT
(d) The "pure energy receptor" symbol
represents a coupling of pure energies such as
sound, light, and water wave trains to an energy
storage process. Chlorophyll dynamics are an
example of this process. In this module incoming
energy interacts with some cycling material to
produce-an excited or activated state, which then
decays to its ground state while transferring
energy to the next step in a chain of linked
reactions. The kinetics of this module were
first described in terms of enzyme-catalyzed
biochemical reactions, and is generally labelled
"Michaelis-Mentor" kinetics. Its transfer
function is hyperbolic. It is usually used in the present ecological context
to depict the reception of light energy by plants in the initial steps of the
photosynthetic reaction. Stable whole ecosystems whose mineral cycles are
fully isolated from allocthonous interchange exhibit dynamic properties and a
functional organization that can be described with this module (Odum, Lugo,
and Burns, 1970).
.HEAT
SINK
CYCLING
RECEPTOR
(e) This "active storage" module expresses
the generation of stored potential from an input
energy flow. In this ^case the re-storage requires
that some transformation process be performed on
the input stream. In this module work is being
done against some resisting force, and the
potential may be stored in a new form. For
example, conversion of the kinetic energy of
falling water to electric power in hydroelectric
engineering projects can be represented as an
active storage. The law of entropy requires that
such conversions cannot be 100 percent efficient,
thus some of the input energy is necessarily
diverted to the "heat sink" of the module.
(f) The symbol for a plant population
combines a cycling receptor module (Figure d)
and self maintaining properties (Figure c). In
addition, work must be done to maintain the
photosynthetic machinery of the cycling receptor.
All this internal organization is summarized in
the bullet-shaped symbol of Figure f.
(e)
INPUT [STORAGE
JLHEAT
SINK
ACTIVE
STORAGE
(f)
INPUT
\OUTPUT
SINK
GREEN
PLANT
POPULATION
XIX-5
-------
(fl)
OUTPUT
(g) Energy sources from outside the boundaries of the
system of interest are indicated with a circle symbol. Names
for this type of module from other system languages include:
forcing function, driving function, environmental variable,
etc. In order to completely describe the source, the
nature of the energy flow (whether light, a flow of organic
matter, the drying power of an air mass, etc.), and the
manner of its delivery, must be specified. For example,
one class of sources exert a constant driving tendency
even when large flows are being delivered, e.g. the
pressure from large water reservoirs, or chemical reactions
taking place under conditions where reactant concentrations
are maintained at constant levels. Another class is sources that deliver a
constant flow regardless of the utilization pattern downstream. Examples
include the flux of sunlight, and water flowing past a waterwheel. Whatever
the case, the delivery function must be indicated mathematically or graphically
to complete the description of the source.
SOURCE
(h)
ENERGY
DRAIN
STRESS
FACTOR
(h) Stress factors require the commitment of
metabolic energies to counter their destructive effects.
This symbol summarizes the process as a drain of potential
that would otherwise be available for other ecological
purposes. Although the rate of energy drain may be a
linear function of stress intensity, the response of the
system to long-term stress may show a lengthy period of
minimal effect followed by an abrupt collapse. Such a
pattern could emerge from the gradual exhaustion of
energy reserves used in the ecosystem to weather normal
intermittent short periods of unfavorable environmental
conditions, such as drought. The cruve for a stress
factor might then follow the hyperbolic relationship of
a limiting-factor work-gate, but in the opposite sense. The "stress" symbol,
then, is an inverted work-gate draining system potential to the heat sink.
The loss rate is controlled by an environmental factor entered on the opposite
side of the gate.
HEAT
SINK
STRESS
(i) The "two-way gate" or forced diffusion module
portrays a bi-directional transport of materials that
may be proportional to a concentration gradient or a.
casual force operating the gate. The heat sink allows
for conformance of the process to the second law of
thermodynamics. Tidal flushing of an estuary is an
example of such a process. Tidal forces move water
in and out of the bay, and net concentration changes
in dissolved materials are a function of concentration
differences between estuarine and off-shore waters.
(j) Any miscellaneous transfer function
appropriate to special cases is represented with this
box symbol. This module is used when the function is
unknown or of no consequence to the point being made.
It can also be used when the mathematical transfer
function is known but a specific symbol is not
available.
XIX-6
DRIVING
TWO-WAY
GATE
(i)
FUNCTION | PU
HEAT
SINK
MISCELLANY
-------
(k) In a "constant gain" amplifier the input
force acts as a tap on an unlimited supply source.
Output dynamics are expressed as a constant increment
of the input force called the "gain" of the module.
The exponential growth of biological populations
under completely unlimiting conditions can be por-
trayed with this module.
(1) Pathways of energy delivery are repre-
sented by the interconnecting lines of the circuit
diagram. The existence of the pathway generally
implies that work is done to maintain it, and this
must be shown somewhere in the diagram. The nature
of the casual force producing the energy flow is
not automatically defined by the pathway. As in
the case of the energy source, additional notation
must be added to define the transmission charact-
eristics of the pathway. However, this question need
not be explored if the coupled responses of the driving
agent (source) and the module receiving the energy are
known. For example, in creating sets of differential
equations from a circuit diagram for simulation, the
time dependent behavior of each module is frequently
derived by writing expressions for the balance of
inflows and outflows to the unit. Each algebraic term
describes a flow as a function of the causal actions.
(k)
INPUT
HEAT
SINK
CONSTANT
GAIN
(I)
STEADY
STATE FLOW
HEAT
SINK
PATHWAY
HEAT SINK
The second energy principle requires that all
spontaneous processes include a dispersal of potential
energy into distributed heat of the environment, unavailable as a driving
impetus for any further process. The heat sink representing this must flow
from every spontaneously operating module of the energy system. When some
simple system such as water flow or electrical current flow is being represented,
the heat sink receives flows directly from the pathway. In complex modules that
represent groups of processes, the heat sink is a miscellaneous conduit of heat
dispersal from the several processes. For example, the respiration of an
organism is the sum of many processes of work and heat dispersal. Respiration
can be measured by the oxygen consumption or C02 production that accompanies
these processes. Measurement of total energy dissipation can be a good starting
point in evaluating the energy flows of complex systems.
Particular component processes, whenever they are of interest, may be
isolated, labelled, measured, and represented by separate flow lines and heat
sinks. If the magnitude of the total process is known, this leaves, by
substraction, a lesser flux of heat dispersal represented by the main miscellaneous
heat flow of the module.
(m) The "switch control" symbol, a special case of work-gate, is used when
discontinuities have been identified in a conductivity control process. The
design of complex switching functions used in modern highspeed digital computers
has provided a thorough theoretical background to the use of this module
XIX-7
-------
in ecological models. Some switched responses are of a
simple on-off character, but many reflect the simultaneous
presence of several necessary conditions for the initia-
tion of an ecological process.
(n) Many responses of ecosystems to environmental
variables involve threshold effects. For example, the
light ground fires characteristic of many southeastern
ecosystems will not propagate unless standing stock of
dry ground litter is greater than some threshold level
(T). The small circle symbol represents a logical "not"
operation by which cutoff points can be represented by
this module„ It portrays a flow of energy that abruptly
ceases when some control threshold is exceeded.
(o) The "economic transactor" is used to relate
the flow of energy to the flow of money, both in
economic systems and at the interface between economic
and ecological systems. Unlike the flow of minerals
that accompanies energy flow, the parallel flow of
money is in the opposite direction, as in a grocery
store. The price structure and market mechanisms
of the economic system regulates the' conductivity
and kinetics of this module. The direct heat losses
of these transactions are small since the work
involved generally is small, as for example in the
purchase of goods in a market. However, the
maintenance of complex regulatory agencies for
controlling the prices of goods and services may
induce substantial costs of coupling within the
system.
We can summarize the useful features of the
energy circuit language for compartmental modeling
of ecological systems as follows:
1. Information is presented in a diagram-
matic form that conveniently summarizes
large quantities of data.
2. Energy flow pathways and functional
interactions are presented in an explicit
form.
(m)
OFF-ON
HEAT
SINK
SWITCH
CONTROL
Cn)
OUTPUT
HEAT
SINK
THRESHOLD
CONTROL
(o)
HEAT
SINK
ECONOMIC
TRANSACTION
3. Energy circuits can be rapidly transferred to a set of mathematical
equations for simulation.
4. Each module implies a particular type of process describable by one
of a limited set of mathematical functions.
5. Research directed toward calibration of the model and calculation of
necessary coefficients for simulation can be coordinated by referring to
the diagrams.
XIX-8
-------
APPENDIX B - ECOSYSTEMS MODELS
This appendix contains models of 5 ecosystem types that are common
in the Fahkahatchee area.: shallow estuarine ecosystems, mangroves, wet
prairie, cypress, and pineland. It is reprinted (retyped) from: Lugo, A. E.,
S. C. Snedaker, S. Bayley, and H. T. Odum, 1971. Models for planning and
research for the South Florida Environmental Study. University of Florida,
Gainesville. 123 pp.
Shallow Estuarine Ecosystems - Three types: marine meadows, hypersaline
lagoons, and estuaries enriched with municipal wastes.
In Figure (11) are diagrammed some of the main parts and processes of
shallow estuarine ecosystems. In clear shallow waters with good water movement,
the bottom plants dominate and the marine meadow system results. When evapora-
tion exceeds freshwater inflows, salinities rise above that of the open sea
(36 ppt) causing a decrease in diversities of many species: this is the
hypersaline ecosystem type. When enriched with municipal wastes containing
fertilizer elements from treated sewage, phytoplankton growths predominate,
shading out many bottom plants, producing some oxygen extremes and other
properties associated with the concept of eutrophication. The diagram becomes
modified in various ways for the various kinds of shallow estuaries of south
Florida depending on which pathways are the most important.
In the estuarine model (Figure 11), the sources of the forcing
functions are shown to be: tides, winds, nutrient and toxic wastes, sunlight,
air, dry air, river water and rain, human and bird population, saltwater and
fisherman. The natural forces all interact among themselves and with the
biota to maintain a highly productive system. Since estuarine ecosystems are
so strongly coupled to many other systems .their values have not been understood
and they have become one of the most abused ecosystems.
Phytoplankton and bottom meadows, consisting of submerged macrophytic
aquatic plants are shown as the major producers. The red tide is also shown as
a producer population even though its role in the system is not completely
understood. As discussed previously its role could be related to acute
eutrophication. Because of its severe ecological and economic impact, it
deserves intensive study.
The interactions among the physical and chemical forces and the biotic
components are the primary features of estuarine systems. An attempt has been
made to diagram some of the interactions utilizing one-way and two-way workgates,
energy sources and storages. For example, tide is illustrated as a major source
of (1) motion (a storage symbol), (2) diffusion of oxygen and carbon dioxide
between air and water (aided by wind), (3) generation of currents and internal
mixing, and (4) movement of saltwater into the system. Tidal motions are also
shown to be responsible for the redistribution of nutrients in river, sea and
rain water and nutrient wastes, and then acting as an amplifier on plant
XIX -9
-------
X
* * TOTAL
•=|- RESPIRATION
Figure 11. Estuary including eutrophic and hypersaline types.
-------
productivity. In addition to its contribution of nutrients, sea water is also
shown to interact with dry air (air with a high saturation deficit) and wind
which through evaporation increases salinity and may produce brine if dilution
is prevented. Salinity differences also affect water currents and red tide.
The metabolic gases, oxygen and carbon dioxide, products of photo-
synthesis and respiration, respectively, are shown resulting directly from
biological processes and indirectly as a result of diffusion. The whole
system's respiratory carbon dioxide is shown as a multiplier on photosynthesis
and the oxygen thus produced as a multiplier on all consumers. (For simplicity
we have omitted several multipliers.) The importance of motion as a gas
transport mechanism is also shown to have amplifier effect.
When an interaction function is unknown or unspecified, a box symbol
is sometimes used. In this example, phytoplankton are shown shading the
bottom meadows. Phytoplankton also support a portion of the food web by
direct feeding and through the production of detrital organic material.
Microbial decomposition converts this material for subsequent use in the
filter feeder chain. Bottom meadows also generate organic material used in
the food web.
From these diverse food chains, many species of game and commercial
fish are derived. As shown in the model many, like the shrimp depend on
water currents and outside sources of larvae for their residence in the
estuarine system. Waterfowl are also shown to have an outside source of
individuals through migration. In both cases, the organisms return to the
open sea or terrestrial environments after completion of a portion of their
life cycles in the estuary.
Because of the many gradients of stress already present in the estuary,
and the demonstrated capacity of these organisms to concentrate different kinds
of abiotic substances, addition of toxic substance to this ecosystem stress
many of the component populations. This is shown in the model with the stress
symbol, operated by toxic wastes, acting on many species of consumers. Man
is shown, through his economic system, to derive a harvest of fishes and shrimp.
Tourists values are also gained in proportion to population pressure.
Mangrove
Low-energy coastlines are characterized by bays, estuaries, salt and
mud flats, salt marshes and mangrove swamps as opposed to the high-energy
coastlines of sand beaches and dunes. The majority of south Florida's perimeter
is composed of a wide band of mangroves juxtaposed between shallow estuaries
and bays, and the upland land mass. Mangroves are a dominent natural feature
due to the flat, slightly sloping relief of the south Florida peninsula. Unlike
many of the other subsystems, the mangroves have clearly defined roles in the
ecology of south Florida. Their ecological importance, however, is rarely
taken into consideration by planners and developers who tend to view this land
area as of marginal value and evaluate it in terms of human use and occupation.
Some 400,000 acres of red mangroves have been so converted to other uses in
Florida. The natural, regional roles of mangrove ecosystems are summarized
within the following statements:
XIX-11
-------
(1) Mangrove ecosystems intercept and filter surface water runoff
thereby preventing large-scale nutrient losses to the open sea.
A by-product of this filtering action is the maintenance of
water quality.
(2) Their complex root structure serves in the accumulation of
organic and inorganic materials. The systems also act as
mechanical energy buffers to strong winds and storm tides,
and thus prevent periodic devastations of the coastline.
(3) They have a high rate of leaf net-production (13 metric tons
per hectare per year, Heald, 1970) which is largely exported
into the estuaries and forms the basis of detrital food chains.
This net production can also be evaluated in terms of the
fisheries it supports.
(4) The shallow waters around mangroves providing breeding and
feeding areas for many estuarine dependent fisheries.
(5) Mangroves support large bird populations by providing rookeries
adjacent to productive estuarine feeding areas.
Classical mono-specific mangrove ecosystems are composed of three
dominant species whose locational relationships are closely correlated with
salinity and topographic relief. The Red Mangrove (Rhizophora mangle)forms
the outermost band associated with the lowest elevations and highest salinity.
It is characterized by extensive prop roots. Inward from the Red Mangrove is
the Black Mangrove (Avicennia nitida) growing at higher elevations which are
subjected to smaller diurnal tidal variations. The Black Mangrove is characterized
by pneumatophores which are suggested to have functions analogous to prop roots.
The last landward zone is influenced only by maximum-high and storm tides and is
composed of the White Mangrove (Laguncularia racemosa) and Buttonwood (Conocarpus
sp.) in varying proportions. Mangroves, world wide, have been extensively
studied, but only a very few of these investigations have attempted to
quantitatively describe their structure and function. The most significant
works are those of Davis (1940), Golley, Odum and Wilson (1962), Heald (1970),
W. E. Odum (1970) and Golley, et al. (1968).
The energy sources powering the mangrove complex, modeled in Figure (14),
are: oxygen, rain, river flow, tides, mixing action, sunlight and heat. The
importance of oxygen as an energy source is shown by its effect in driving the
many respiratory processes in the substrate of the mangroves which are limited
by low oxygen concentrations. In the absence of oxygen the organic muds become
anaerobic as shown in the submodel describing anaerobic processes (Figure 15).
In addition, oxygen is required to sustain root respiration. Because of the
limited amount of oxygen, roots are shown pumping oxygen into the sediments.
Structurally this pumping is done via pneumatophores, lenticels, and diffusion
through aerenchyma cells.
Rain is a source of freshwater into the system. Freshwater is shown to
flow by the action of rivers and to become loaded with terrestrially-derived
sediments and man-introduced substances such as heavy metals. In the mangrove
areas, freshwater combines with saltwater and by the action of vigorous mixing,
form brackish waters rich in nutrients which support a spectrum of organisms„
The brackish waters are shown to act as a. multiplier function on the energy
XIX-12
-------
X
I—>
10
TO ESTUARY
Figure 14. Mangroves.
-------
X
THIO -
BACILL
THIO-
XI DANS
PSEUDO
MONAS
DENITRI
ICAN
Figure 15. Aerobic and anaerobic processes in mud.
-------
flows to those consumers adapted to the specific conditions associated with
mangroves. Heald for example (1970) has shown that certain detritus decomposers
function at faster rates in brackish waters in contrast to either saline or
freshwaters. In the model, brackish waters are also shown to import nutrients
which in turn act as multipliers on plant photosynthesis; they transport seston
which are the base of a filter-feeder chain and are also shown to import the
extraneous materials which may stress mangrove trees. They also provide
sediments which affect plant zonation and in turn exert an influence on plant
photosynthesis.
Tides are shown as an important source of energy in transporting
waters and minerals in the estuarine-mangrove complex. For example, the
model shows the tides serving as a "pump" of detrital particles from the
mangroves to the estuaries. Another important role is the maintenance of the
water level in the mangrove system. The water level is also shown to have a
switching effect on the amount of oxygen available for respiratory processes
in the sediments.
Solar energy is shown powering plant photosynthesis and the mangrove
trees are depicted subdivided into their basic functional parts. Reproductive
structures are shown as providing a feedback into productivity by maintaining
a vigorous plant population. Roots are shown as important in pumping atmospheric
oxygen into the sediments, and as a multiplier on sessile organisms which use
them as a substrate. They also serve as multipliers on the trees themselves by
their role in anchorage. Both roots and leaves contribute to the liter which
in turn serves as a supplier of organic matter for sediment respiratory
processes, peat formation, and as a base for decomposer and detrital food chains.
Leaves, in addition, serve as a base for the grazer food chain which links the
mangroves to the terrestrial ecosystem. The organisms in the terrestrial food
chain also feed on sessile organisms on mangrove roots as shown in the model.
The sessile organism exhibit an On-Off activity cycle determined by water-level
fluctuations.
Aquatic food chains, both aerobic and anaerobic, are shown as being
dependent on the detritus and sediments of the system. These two distinct
processes are indicated in the diagram. In the detritus food chains, one
consumer is shown doing work for the next by breaking up the detrital particles
into smaller pieces and thus facilitating feeding. As the detrital particles
are reduced in size due to decomposition and fragmentation by consumers,
nutrients are released and incorporated into the system through plant photo-
synthesis. A multiplier on the mangroves illustrates this effect.
Heat is shown as an On-Off swtich on many of the metabolic processes of
the system since they are sensitive to low temperatures which periodically stress
these systems.
The sediments of mangrove systems, with their extensive organic matter
storages and oxygen requirements, are complex biological networks where aerobic
and anaerobic processes switch on and off depending on the particular environ-
mental circumstances. The details of these processes are far from being
completely understood and are perhaps one of the most exciting frontiers of
ecology,, In Figure (15) we have modeled some of the anaerobic and aerobic
processes expected to occur in these muds. On the right side of the diagram,
reduced compounds are shown as a storage being acted upon by self-maintaining
organism populations. On the left, oxygen is shown as an energy source powering
XIX-15
-------
the oxidation of reduced compounds. This is illustrated as a multiplier on
the aerobic-sediment flora. Note that 804, H2S, FeSjNOjj and NH+ couple the
aerobic and anaerobic processes through their roles as the products and raw
materials of various bacterial metabolic processes. Other details of the
system's function are shown as they relate to the mangrove system. Illustrated
are liter as a carbon source to the sediments and POi production from anaerobic
decomposition acting as a multiplier on plant productivity as does NOg. The
roles of CC^j tides and water level in the system are also depicted.
Wet Prairie
Wet prairies are frequently designated to be those areas composed of
Rhynchos pora, Eleocharis, and Panicum and are characterized by water regimes
which are intermediate between the sawgrass and slough ecosystems. In addition,
this ecosystem is frequently found on a marl soil as compared with the peaty
soils of sawgrass and slough ecosystems. Descriptions of this ecosystem are
found in Davis (1943) and Loveless (1959).
In the model of Figure (23) marl solution and deposition are depicted
in addition to the mechanism of evapotranspiration, and the biological inter-
actions among macrophytes, periphyton, wildlife and man. The major forcing
functions shown are: rain, inflow of water, ground water composition, sunlight,
wind, dry air, and fossil fuels. The'main producers are the aquatic macrophytes
and periphyton. The important consumers are microbes and invertebrates, frogs,
and other carnivores. Frogging is a major activity in this ecosystem and man
is shown harvesting frogs and other game animals. The major biological inter-
actions shown are the production of oxygen and organic matter (detritus) by
photosynthesis and the role of these two substances in powering the food chains
and the respiratory processes of both plants and animals. Respiration in turn
is shown releasing carbon dioxide and minerals and these are depicted recycling
as feedbacks into the photosynthetic process. Many of these feedbacks as well
as organic export from the system are shown to be operated through the switching
and pumping action by water levels in the system. Water levels are shown to be
dependent on rainfall and run-off for inputs and to evapotranspiration for
losses. Standing water is also shown as an unlimited energy source into a
constant gain amplifier powering high evapotranspiration. The process of
evapotranspiration is powered by heat (derived from sunlight), dry air (which
provides a high saturation deficit) and wind. Wind maintains the diffusion
gradient and drives moist air away. Plants contribute to the flow through
plant transpiration.
Finally, the chemistry of marl solution and deposition is shown with
the energy language, in the lower right side of the model. An important role
in this equilibrium is played by periphyton through their actions in the carbon
equilibrium and pH and the quality of ground water as a source of calcium. The
dynamics of marl formation is now being studied by researchers in the department
of Biology at the University of Miami.
Cypress
The diagram in Figure (25) shows the salient features of the cypress
swamp and its interactions with man. Illustrated as the main producers (powered
by sunlight) are cypress trees, aquatic plants, other component tree species,
and many epiphytes and herbs. These producers provide both local and migratory
wildlife with an abundant diversity of foods s_hown in the model as a storage.
XIX-16
-------
X
M
X
^±r TOTAL
RESPIRATION
Figure 23. Wet prairie.
-------
X
M
X
Figure 25. Cypress swamp.
-------
These wildlife species are then shown to have specialized roles in the mineral
cycles of the system. A feedback entering the nutrient storage symbol illustrates
this function. Also contributing to the nutrient pool are the decomposers which
do work on the pool of dead organic matter and peat (each shown as a separate
storage). Notice that the water level has a switching action on decomposition;
when the water is high, oxygen becomes limiting to decomposition and the switch
is Off. The cypress trees are the dominant species and have interesting
adaptations and interactions with the other components of the system. Shown
in the model are the characteristic cypress knees which are presumed to
contribute to high productivity by pumping oxygen into the roots (growing in
anaerobic conditions) and therefore stimulate mineral and water movement
through the trees. In addition, cypress trees are known to produce hydrocarbons
and dissolved organic matter (such as tannic acid) which might be important to
the system's energy economy (Greenfield, el: al., 1970). In the model these
substances are shown to interact with water level which serves as a medium for
dispersal. In addition, the hydrocarbon films pick up dust from the air and
therefore become enriched with organic matter which may serve as a fuel for
the decomposers. These interactions are shown in the model. Peat, wildlife
components, and the hydrocarbon-dust films are shown as export pathways to
other systems. This emphasizes the coupling of the cypress swamps with
surrounding ecosystems.
On the left portion of the diagram, the large workgate represents all
system interactions that promote and maintain productivity. Shown as performing
these functions are; the many specialized activities of wildlife (grazing for
example), nutrient availability, water level (which also acts as a selective
force on plants), and the many specialized functions of other plant components,
such as the role of epiphytes in nutrient removal from rainfall. Rainfall is
shown as the only source of water to the basin as has been discussed in many
reports.
The interactions of man are many. In terms of benefits from the system,
we show his harvest of wood, aquatic and terrestrial wildlife, and water from
the aquifer. Notice that man must work to obtain water from the aquifer. This
is shown by a workgate on the water flow. On the negative side, man stresses
the systems as his intrusions increase. In this model we show stress as
dependent on two factors: demand and availability of power, and the many uses
shown on the left as energy sources. As the demand for power in the urban
system increases; the use of power goes up; and as the use of power increases,
the activities of man, shown on the left, also increase. These, for example,
are: use of pesticides, jetports, cattle grazing, agriculture, housing,
siltation, waste production, fertilizer use, minimg, and fire frequency. All
these of course, have an effect on the system by stressing the producers and
wildlife.
Pineland
Pinelands, like many other ecosystems, exist in a precarious balance
with their environment as illustrated in Figure (26). Any shift in environmental
conditions, particularly the frequency of fire, permits hardwoods to invade and
replace pine as the dominant species (Alexander and Dickson, 1970).
In order to emphasize the internal biological detail, only solar energy,
rain, and such external forces such as lightening, are shown as forcing functions
in the systems model. The dominant components of the vegetation are pines
XIX-19
-------
(canopy),'a diverse understory of palms and hardwood, and a herbaceous layer
composed of seedlings and repair (successional) species. The important environ-
men factors are soil moisture, available nutrients, fire and the substrate.
Fire, in this example, is shown as a consumer of organic matter.
Robertson (1953) has described fire frequencies and the effects on the pineland
ecosystem in the Evergaldes National Park. Fires occur when moisture is low
and an ignition source is operational. This is shown with two switches and the
presence of dry fuel (litter). Because of the relatively high frequency of fire,
the litter storage is kept low even though it is continually being replenished
by programmed leaf fall. The programming (phenology) of the forest events have
not been sufficiently documented. Pines are fire adapted species, illustrated
in the model as thick bark, specialized phenological patterns (e.g., growth)
and the production of combustible needles. Fire is also shown to stress hard-
woods causing heavy mortality and preventing their assumption of dominance. In
addition, fire releases (shown as a switch) and repair specialist species which
are adapted to germination in burned or otherwise with several stressed areas.
Some of these species also fix nitrogen and as a group act as a multiplier on
the whole system. This is shown as a feedback on the producers. The rapid-
growing producers in particular provide food and cover to a diversity of
consumers associated with the pineland ecosystem. The consumers are shown to
have an important role in the system through their mineral recycling work.
The interaction of soil, water and plants is thought to have a role in
determining plant species composition. The first, of course, is the obvious
role of water in photosynthesis and transpiration shown by a multiplier.
Because of the shallow substrate, root space is limited and thus the unavailability
of water and nutrients is dependent upon a relatively small storage. This
limitation of critical resources exerts an influence on species densities and
survival. These interactions are illustrated in the lower right hand portion
of the model.
Also shown in the model are suggested specialized roles for the species
comprising the diverse hardwood understory. For example, each species could
have an important function in accumulating and storing a specific mineral as
has been shown for other forest components elsewhere in the United States. In
general the pineland ecosystem like all others present lacks quantitative
description of its structure and function in a specialized role of the component
species.
XIX-20
-------
X
N3
FIRE SUCCESS ION ALS
Figure 26. Pinelands.
-------
APPENDIX C - COMPUTATIONS OF SEDIMENTATION RATES
FROM TRAP DATA
I. Laboratory methodology
A. Let sample settle for 24 hours
B. Test for Halides (CL~) on a sample of supernatant
C. Decant off supernatant to near solids level
D. Tare burned and cooled crucible (crucible weight)
E. Add solids and liquid to crucible (use known volume of rinse water)
F. Weigh (wet weight + crucible)
G. Dry at 105° C for 24 hours
H. Weigh (dry weight + crucible)
I . Burn for one hour at 550° C
J. Wet the ash and dry at 105° for 24 hours
K. Weigh (ash weight + crucible)
II. Calculations
L.* Salt concentration = (1.850)(CL~ mg/1) + 30
M. Total weight salts = (volume salt water dried) (salt concentration)
= (F - H - rinse water) (g) L
N. Organic weight = H - K in 300 ml sample
0. Inorganic weight = K - D - M in 300 ml sample
P. Total organic weight = N (. -*1??*8 °! sfmPle , Q = gm
& & V0.333 liter (sub sample)^ s
Q. Organic deposition rate = P * (elapsed time in days) * (area of
2
sampler) = gm/m /day
R. Total inorganic weight = 0 (. -*1*?*3 ° . J = gm
to & V0.333 liter (sub sample) J 6
S. Inorganic deposition rate = R * (elapsed time in days) •=- (area of
2
sampler) = gm/m /day
* Knudsen formula - S ppt = 1.850 CL ppt + 0.030 in ppt
XIX-22
-------
APPENDIX D - INHABITANTS FOUND IN SEDIMENTATION TRAPS
Fahka Union Bay
Unidentified crab - 1 Silver perch - 10
Skilletfish - 1
Goby - 17
Sheepshead - 2
Snapping shrimp - 1
Toadfish - 1
Snapper - 1
Blue crab - 21
Number of species - 11
Number of individuals - 125
Frequency of occurrence - 50/126
Jewfish - 2
Shrimp - 68
Fahkahatchee Bay
Snail - 1
Goby - 23
Toadfish - 6
Snapper - 2
Number of species - 8
Number of individuals - 78
Frequency of occurrence - 52/126
Total
Blue crab - 11
Sheepshead - 2
Skilletfish -
Shrimp -32
Number of species - 12
Number of inidividuals - 203
Frequency of occurrence - 102/252
XIX-23
-------
APPENDIX E - SAMPLING STATION SITE DESCRIPTIONS AND LOCATIONS
Station Description and Location
1 Southeast corner of "bridge at mile 18.5 on Alligator Alley
(S. R. 84). An off-shoot from the Borrow Canal draining
southward into the Fahkahatchee Strand. A freshwater station.
2 North side of "bridge at mile 19-5 on Alligator Alley (S. R.
84). An off-shoot from the Borron Canal draining southward
into the Fahkahatchee Strand. A freshwater station.
3 South side of "bridge east of the junction of S. R. 29 and
Alligator Alley (S. R. 84) on the Borron River. A fresh-
water station.
4 South side of "bridge in front of Janes Resturant on S. R.
29 on the Barren River. A freshwater station.
5 South side of wooden "bridge on Janes Scenic Drive 4.83
kilometers (3 miles) from the fire tower. A freshwater
station on the outer edge of Fahkahatchee Strand.
6 25 meters (2?.4 yards) north of Fort Myers Tram at a point
on the Tram 100 meters (109 yards) from the junction of Fort
Myers Tram and Janes Scenic Drive. A freshwater station
close to the middle of Fahkahatchee Strand.
7 A culvert on Janes Drive 14.48 kilometers (9 miles) from
the fire tower. A freshwater station slightly outside of
the Fahkahatchee Strand, but occasionally receives water
from the^ strand.
8 North side of bridge ?1 on U. S. 41 (Tamiami Trail). An
off-shoot of the Borron Canal draining southward into a
salt marsh. It has brackish water and is subject to tidal
influence.
9 North side of bridge 64 on U. S. 41 (Tamiami Trail). An
off-shoot of the Borron Canal draining southward into a
salt marsh. It has brackish water and is subject to tidal
influence.
10 North side of weir on Fahka Union Canal close to U. S. 41.
A freshwater station.
F-l Fahkahatchee River at first fork going upstream. An
estuarine station.
F-2 Approximate center of Fahkahatchee Bay.
XIX-24
-------
APPENDIX E - SAMPLING STATION SITE DESCRIPTIONS AND LOCATIONS (CONT'D)
Station Description and Location
F-3 South of Fahkahatchee Bay in the middle of Fahkahatchee pass.
FU-1 Midstream of Fahka Union Canal at the junction -,/f Fahka Union
River approximately 2.90 kilometers (1.8 miles) downstream
from U, S. 41. An estuarine station.
FU-2 Fahka Union Bay east of Black Marker 3J> in mid-channel.
FU-3 Fahka Union Bay east side of Black Marker $| in the middle of
Fahka Union pass.
XIX-25
-------
APPENDIX F ^ SPECIFIC TEST USED FOR EACH WATER CHEMISTRY ANALYSIS
Temperature; In situ using a Beckman Model RS5-3 salinometer.
Conductivity; In situ using a Beckman Model E.S5-3 salinometer.
pH: Fisher Accumet Model 220 pH -meter using a glass electrode in combination
with a reference potential (Saturated calomel electrode).
Apparent color: Visual comparison of nonclarified sample with paltinum-cobalt
standards.
Turbidity: Hach Model 2100A turbidimeter, which was calibrated with a formazin
standard freshly prepared every week,
Dissolved oxygen; The modified Winkler with full-^bottle technique. Samples
were fixed in the field with 2 ml of manganous sulfate solution and 2 ml alkali
azide solution. In the laboratory, samples were acidified and titrated with
standardized sodium thiosulfate.
Alkalinity: An unaltered sample was manually titrated to an electrometrically
determined end point of pH 4,5.
Chloride: Acidified samples with diphenylcarbazone bromophenol blue indicator
and titrated with mercuric nitrate solution.
Sulfate: The sulfate ion was converted to a barium sulfate suspension which
was read as turbidity by the Hach Model 2100 A turbidimeter and compared to a
curve prepared from standard sulfate solution.
Total and soluble Kjeldahl nitrogen: A portion of the sample was filtered by
passing it through a 0.45-y Millipore filter membrane. The unfiltered portion
was analyzed for total Kjeldahl nitrogen, and the filtrate was analyzed for
total soluble Kjeldahl nitrogen. The nitrogen was determined using the micro-
glassware-nesslerization procedure.
Total and soluble phosphorus; A portion of the sample was filtered by passing
it through a 0.45-u Millipore filter membrane. The unfiltered protion was ana-
lyzed for total phosphorus, and the filtrate was analyzed for total soluble
phosphorus. The single reagent method, which is found in the EPA manual, was
used for the analyses of both total and soluble phosphorus.
Total and soluble organic carbon; A portion of the sample was filtered by pass-
ing it through a 0,45-y Millipore filter paper. Both filtered and unfiltered
portions were preserved with 2 ml concentrated E^SO^. per liter and then sent to
the Southeast Environmental Research Laboratory in Athens, Georgia, for analysis.
The unfiltered portion was analyzed for total organic carbon, and the filtrate
was analyzed for soluble organic carbon. Both forms of carbon were analyzed on
the Beckman Total Organic Carbon Analyzer according to EPA procedures,
Nitrates-nitrite nitrogen; Samples were preserved with 2 ml concentrated H2SOjt
per liter and then shipped to the Southeast Environmental Research Laboratory in
Athens, Georgia, for analysis by the automated cadmium reduction method,
XIX-26
-------
APPENDIX F - SPECIFIC TEST USED FOR EACH WATER CHEMISTRY ANALYSIS (CONT'D)
Tannin and llgnin "like" compounds; For both the freshwater and estuarine sam-
ples, the procedure in the Standard Methods for the Examination of Water and
Wastewater, 13th Ed. was followed except that for the estuarine samples, a
filtering procedure was employed after the addition of the sodium carbonate
solution. This was to eliminate the precipitate from metal carbonates. Tannin
and lignins and other hydroxylated aromatic compounds are detected by this
method.
XIX-27
-------
APPENDIX G - STANDING CROP BIOMASS OF BENTHIC VEGETATION
IN FAHKA UNION AND FAHKAHATCHEE BAYS. JANUARY 1972
Plant type and
relative density
Percent of total bay area
supporting vegetation
Standing crop
biomass
(kg dry wt)
Green filamentous algae
Trace
Medium
Heavy
Total
Red macro-algae
Trace
Medium
Heavy
Total
Marine grasses
Trace
Medium
Heavy
Total
Green filamentous algae
Trace
Medium
Heavy
Total
Red macro-algae
Trace
Medium
Heavy
Total
Marine grasses
Trace
Medium
Heavy
Total
Fahkahatchee Bay
31.2
6.8
2.2
40.2
39.5
20.9
0.0
60.4
29.4
18.1
10.1
57.6
Fahka Union Bay
42.4
14.1
10.9
67.4
23.8
3.9
0.0
27.7
17.8
0.0
5.3
23.1
1,393
4,503
7,645
13,541
883
2,488
0
3,371
1,311
2,154
6,191
9,756
4,812
2,376
9,661
16,849
102
90
0
192
153
0
623
776
XIX-28
-------
APPENDIX H - STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
SAND-MUD BOTTOM OF FAHKAHATCHEE AND FAHKA UNION BAYS, JULY 1972
Total standing
Area Mean cone. crop biomass
Plant types (ha ) (gm dry wt /m2) (kg dry wt )
Fahkahatchee Bay (299.6 ha )
Marine grasses 250.4 23.6 59,094
(1.1 - 150.1)
Red macroalgae 256.3 2.5 6,407
(0.0 - 47.4)
Green filamentous 256.3 0.7 1,794
algae (0.0 - 0.8)
No vegetation 43.3
a Mean concentration subtended by range in concentrations
Total area of bay represented by substrate.
Total 67,295
Fahka Union Bay (20.4 ha )
Marine grasses 20.4 16.1 3,288
(0.0 - 31.2)
Red macroalgae 0.0 0.0 0
Green filamentous
algae
No vegetation
Total
0
0
.0
.0
0
0
.0
.0
0
0
3,288
XIX-29
-------
APPENDIX I - STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING MUD-SAND
BOTTOM AREA IN FAHKAHATCHEE AND FAHKA UNION BAYS, JULY 1972
Total standing
Area Mean cone. crop biomass
Plant types (ha ) (gm dry wt /m2) (kg dry wt )
Fahkahatchee Bay (586.8 ha )b
Marine grasses 224.2 48.8 109,410
(0.4 - 296.0)
Red macroalgae 339.0 1.9 6,441
(0.0 - 30.0)
Green filamentous 114.8 0.2 23
algae (0.0 - 2.2)
No vegetation 47.8
Total 115,874
Fahka Union Bay (145.1 ha )b
Marine grasses
Red macroalgae
Green filamentous
algae
No vegetation
Total
60.1 7.4
(1.3 - 11.9)
121.2 1.8
(0.0 - 8.6)
121.2 7.6
(0.0 - 37.8)
21.9
445
2,229
9,238
11,912
o
Mean concentration subtended by range in concentrations
Total area of bay represented by substrate
XIX-30
-------
APPENDIX J - STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING
SHELL BOTTOM IN FAHKAHATCHEE AND FAHKA UNION BAYS, JULY 1972
Standing
Area Mean cone. crop biomass
Plant types (ha ) (gm dry wt /nr) (kg dry wt )
FahkahatcheeBay .(56.8 ha )
Marine grasses 11.2 12.7 1,426
(3.7 - 17.1)
Red macroalgae 34.0 1.6 554
(0.0 - 3.1)
Green filamentous
algae 0.0 0.0 0
No vegetation 22.8
Total 1,980
Fahka Union Bay (25.8 ha )
Marine grasses 0.0
Red macroalgae 0.0
Green filamentous
algae 0.0
No vegetation 25.8
o
Mean concentration subtended by range in concentrations
r.^
Total area of bay represented by substrate
XIX-31
-------
APPENDIX K - STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING MUD-SAND
BOTTOM IN FAHKAHATCHEE AND
FAHKA UNION BAYS,
SEPTEMBER 1972
Area
Plant types (ha )
Fahkahatch.ee Bay (386.8 ha )b
Marine grasses 317.4
Red macroalgae 317.4
Green jrilamentous
algae 5.4
No vegetation 64.0
Total
Fahka Union Bay (143.1 ha )b
Marine grasses 57.0
Red macroalgae 49.4
Green filamentous 40.6
algae
No vegetation 36.7
Total
Mean cone.
(gm dry wt /m )
43.1
(0.0 - 250.1)
2.0
(0.0 - 28.1)
10. 5C
0.0
5.0
(3.2 - 13.3)
21.9
(0.0 - 132.5)
38.1
(10.3 - 96.1)
0.0
Standing
crop biomass
(kg dry wt )
136,799
6,396
567
0
143,762
2,850
108,341
15,472
0
126,663
o
Mean concentration subtended by range in concentrations
b
Total area of bay represented by substrate
c
Only one observation
XIX-32
-------
APPENDIX L - STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING SAND-MUD
BOTTOM IN FAHKAHATHCEE AND FAHKA UNION BAYS, SPETEMBER 1972
Standing
Area Mean conc.a _ crop biomass
Plant types (ha ) (gm dry wt /m ) (kg dry wt )
Fahkahatchee Bay (299.6 ha )
b
Marine grasses 277.5 39.1 108,502
(0.0 - 152.5)
Red macroalgae 277.5 13.9 38,572
(0.0 - 207.1)
Green filamentous
algae 0.0 0.0 0
No vegetation 22.1 0.0 0
Total 147,074
Fahka Union Bay (20.4 ha )
Marine grasses 20.4 4.6° 938
Red macroalgae 0.0 0.0 0
Green filamentous
algae 0.0 0.0 0
No vegetation
Only one observation
Total 938
cl
Mean concentration subtended by range in concentrations
b
Total area of bay represented by substrate
XIX-33
-------
APPENDIX M - STANDING CROP BIOMASS OF BENTHIC VEGETATION INHABITING SHELL
BOTTOM IN FAHKAHATCHEE AND
FAHKA UNION BAYS, SEPTEMBER 1972
Plant types
Fahkahatchee Bay (56.8
Marine grasses
Red macroalgae
Green filamentous
algae
No vegetation
Total
Area
(ha )
iha)b
25.6
44.3
0,0
12.5
Mean cone. _
(gm dry wt /m )
1.5
(0.3 - 2.9)
2.6
(0.0 - 10.4)
0.0
0.0
Standing
crop biomass
(kg dry wt )
384
1,152
0
0
'l,536
Fahka Union Bay (25.8 ha )
Marine grasses
Red macroalgae
Green filamentous
algae
No vegetation
Total
0.0
9.5
0.0
16.3
0.0
0.8°
0.0
0.0
0
81
0
0
81
rt
Mean concentration subtended by range in concentrations
Total area of bay represented by substrate
Only one observation
XIX-34
-------
APPENDIX N - SUMMARY OF MEM LIGHT EXTINCTION COEFFICIENT WITH RESPECT TO DATE,
TIME OF DAY, AND WATER DEPTH, FAHKA UNION BAY WITH MUD-SAND BOTTOM, 1972-71
Mean light extinction
coefficient (K) and standard
Date
Feb. 16, 1972
Feb. 16
Feb. 19
Mar. 24
April 6
April 25
Oct. 26
Oct. 27
Nov. 1
Nov. 4
Nov. 4
Nov. 5
Dec. 7
Feb. 14, 1973
Time •
of
day
p.m.
p.m.
a.m.
p.m.
p.m.
a.m.
a.m.
a.m.
a.m.
a.m.
a.m.
a.m.
a.m.
a.m.
Mean
depth
(cm)*
64.9
68.1
32.1
101.4
87.8
88.1
75.6
61.0
85.0
ND
33.1
113.4
58.5
92.8
error of mean
Mean
K
0.0262
0.0242
0.0496
0.0179
0.0265
0.0215
0.0188
0.0226
0.0157
0.0213
0.0143
0.0274
0.0318
0.0141
Sx
0.0014
0.0053
0.0021
0.0004
0.0026
0.0026
0.0017
0.0015
0.0009
0.0006
0.0010
0.0030
0.0025
0.0010
Sample
size
9
3
3
11
10
10
5
4
2
2
3
2
10
10
Mean K
0.0237
mean depth during time interval of light extinction measurements
XIX-35
-------
APPENDIX 0 - SUMMARY OF MEAN LIGHT EXTINCTION COEFFICIENT WITH RESPECT TO DATE.
TIME OF DAY, AND WATER DEPTH, FAHKAHATCHEE BAY WITH SAND-MUD BOTTOM, 1972-73
Mean light extinction
coefficient (K) and standard
Date
Feb. 18, 1972
Mar. 7
Mar. 7
Mar. 16
Mar. 16
Mar. 24
April 25
Oct. 3i
Nov. 2
Nov. 4
Dec. 7
Feb. 14, 1973
Time
of
day
p.m.
a.m.
p.m.
a.m.
p.m.
p.m.
a.m.
p.m.
p.m.
p.m.
p.m.
p.m.
Mean
depth
(cm)*
114.3
74.4
91.7
106.7
161.3
112.8
122.2
112.9
71.9
164.3
63.0
117.1
error of mean
Mean
K
0.0359
0.0255
0.0294
0.0171
0.0149
0.0131
0.0158
0.0198
0.0378
0.0107
0.0310
0.0129
Sx
0.0034
0.0001
0.0006
0.0001
0.0007
0.0014
0.0020
0.0043
0.0018
0.0037
0.0021
Sample
size
2
3
6
1
2
6
6
6
2
4
6
6
Mean K
0.0220
mean depth during time interval of light extinction measurements
XIX-M
-------
TIME OF DAY, AND
WATER DEPTH, FAHKAHATCHEE BAY WITH MUD-SAND BOTTOM, 1972-73
Mean light extinction
coefficient (K) and standard
Date
Feb. 18, 1972
Feb. 18
May 7
May 8
May 8
May 10
May 14
May 16
May 24
May 24
April 6
April 25
Oct. 29
Oct. 30
Nov. 1
Nov. 4
Nov. 4
Dec. 7
Feb. 14, 1973
Mean K
Time
of
day
a.m.
p.m.
p.m.
a.m.
p.m.
p.m.
p.m.
p.m.
a. iii.
p.m.
p.m.
a.m.
p.m.
p.m.
p.m.
a.m.
p.m.
p.m.
a.m.
Mean
depth
(cm)*
53.1
213.4
120.4
86.9
76.2
102.9
142,0
146.3
121.6
77.8
96.3
116.1
78.5
80.8
86.7
117.0
121.0
67.5
115.1
Mean
K
0.0323
0.0216
0.0319
0.0198
0.0208
0.0244
0.0160
0.0208
0.0138
0.0293
0.0232
0.0169
0.0194
0.0208
0.0201
0.0338
0.0250
0.0325
0.0150
0.0230
error of mean
Sx
0.0032
0.0002
0.0005
0.0003
0.0005
0.0020
0.0010
0.0026
0.0018
0.0016
0.0013
0.0012
0.0062
0.0007
0.0015
0.0023
0.0016
Sample
size
6
1
4
3
3
2
3
1
9
10
10
10
6
2
2
4
5
10
10
mean depth during time interval of light extinction measurements
XIX-37
-------
APPENDIX Q - BENTHIC GROSS PRIMARY PRODUCTIVITY (GPP) IN FAHKA UNION AND FAHKAHATCHEE BAYS
X
00
EMPLOYING LIGHT AND DARK CHAMBERS,
SPRING AND FALL 1972
Chamber conditions
Initial
System Date
FU-MSa Feb. 15
n n
n n
n n
n n
" Feb. 16
n n
F-MSb Feb. 18
ii n
n n
ii M
H n
M u
" Mar. 8
n n
n ii
n n
" Mar. 10
n n
F-SMC Feb. 17
n M
" Feb. 18
ii ii
" Mar. 7
n n
M ii
H ii
" Mar. 24
n n
Type
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Temp. °C
22.5
22.4
23.5
22.0
23.6
23.8
23.5
22.1
23.0
22.0
21.9
21.9
23.0
24.0
ND
23.0
25.3
25.5
24.0
22.5
23.8
22.5
23.0
24.0
24.0
24.0
24.0
24.0
23.2
23.0
D.O.
(mg/1)
7.2
n
7.3
n
7.4
5.8
n
4.2
1!
4.7
n
5.4
n
6.8
n
7.2
n
7.4
5.8
n
7.8
7.4
ii
n
it
u
7.2
Time interval
for rate
(hrs . )
1330 -
1332 -
1459 -
1501 -
1625 -
1632 -
1110 -
1148 -
0932 -
0943 -
1117 -
1122 -
1306 -
1245 -
1103 -
1144 -
1545 -
1544 -
1438 -
1430 -
1358 -
1406 -
1453 -
1442 -
1130 -
1204 -
1332 -
1353 -
1526 -
1558 -
1409
1405
1541
1543
1702
1704
1226
1250
1035
1032
1132
1157
1327
1337
1127
1238
1615
1626
1448
1445
1448
1502
1521
1521
1206
1257
1430
1335
1648
1638
Rate of
D.O. meta-
bolism
ID'2 mg/1/
min
-1.29578
-1.84738
-1.22880
-1.39539
-1.69990
-1.86818
-0.37392
-0.91699
+0.09447
-0.79950
+2.53807
-2.03494
+2.84615
-1.39719
+1.56667
-0.82447
+0.10952
-2.43900
+0.51266
-1.33136
-0.41467
-0.89405
-1.77350
-2.35539
+1.04167
-0.53005
-0.05430
-1.19000
-0.59630
-0.70000
GPP
mg/1/
min
0.5516
0.1666
0.1683
0.5431
0.8940
4.5730
4.2433
2.3911
2.5485
1 . 8440
0.4794
0.5819
1.5717
1.1357
0.1037
Vis. light
transmission
at bottom
10" 2 ly/min
6.252
1.427
0.357
5.392
2.605
7.679
7.223
3.0307
4.489
4.362
0.727
1.363
4.000
4.000
0.316
-------
X
H
X
(jO
vo
ArrrjlNUJLA i^ — j3rjiN±nio wix>.
EMPLOYING 1
j i_> ij JL j.v-i_i- un.i
LIGHT AND
DARK CHAMBERS,
SPRING AND FALL 197
Chamber conditions
Initial
System
FU-MSa
ti
ii
"
"
n
"
it
n
M
tt
"
"
n
n
1!
It
It
11
tt
F-MSb
n
n
n
tt
tt
n
n
Date
Oct. 26
n
11
it
Oct. 27
it
M
n
Nov. 1
11
n
n
Nov. 4
11
n
n
n
n
Nov. 5
n
Oct. 29
n
n
n
n
it
it
ti
Type
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Temp. °C
27.3
27.8
28.0
27.8
28.8
28.8
29.0
30.0
27.2
27.6
27.1
28.0
27.0
27.0
27.0
27.0
27.2
27.0
27,8
28.0
28.5
29.4
29.9
29.4
29.9
30.2
30.7
30.2
D.O.
(mg/1)
5.0
7.5
11
ti
5.1
n
5.8
n
5.9
"
5.7
n
5,5
6.8
5.5
6.8
5.5
6.8
5.1
1!
4.6
n
5.5
5.9
5.7
5.9
6.1
5.9
Time interval
for
rate
(hrs.)
0815
0905
0915
0905
0957
1010
1100
1245
1018
1034
1213
1251
0848
1753
0917
1753
0945
1753
0745
0746
0915
0945
1015
1230
1145
1230
1300
1230
- 0900
- 0940
- 0945
- 0940
- 1046
- 1055
- 1128
- 1315
- 1039
- 1058
- 1231
- 1315
- 0857
- 1810
- 0925
- 1810
- 0950
- 1810
- 0818
- 0813
- 1015
- 1020
- 1115
- 1300
- 1209
- 1300
- 1330
- 1300
Rate of
D.O. meta-
bolism
1(T2 mg/1/
min
-0.86666
-1.90000
-1.50000
-1.90000
+0.11820
-0.57894
+0.27714
-3.15789
+6.00396
-2.38333
+2.83333
-1.90559
+2.66660
-2.63698
+10.10000
-2.63698
+10.00000
-2.63698
-1.25174
-1.82222
+3.23140
-2.93333
+0.30000
-2.93330
+2.44220
-2.66667
-0.28111
-2.66667
2 CONT'D
GPP
U "
mg/1/
min
1.
0.
0.
3.
8.
4.
5.
12.
12.
0.
6.
3.
5.
2.
033-3
4000
6971
4350
3873
7389
3036
7370
6370
5705
1647
2333
1090
3856
Vis
. light
transmission
at bottom
10-2
3
4
4
5
11
9
10
22
20
2
5
3
5
2
ly/min
.118
.879
.109
.751
.430
.862
.521
.669
.346
.700
.463
.207
.329
.441
-------
APPENDIX Q - BENTHIC GROSS PRIMARY PRODUCTIVITY (GPP) IN FAHKA UNION AND FAHKAHATCHEE BAYS
EMPLOYING LIGHT AND DARK CHAMBERS, SPRING AND FALL 1972 CONT'D
X
H
X
Chamber conditions
Initial
System
F-MSb
ti
it
tt
"
ti
F-SMC
"
tt
tt
it
it
ii
tt
F-SMC
it
it
ti
it
it
Date
Nov. 2
it
ti
it
ti
ti
Oct. 31
it
ti
it
it
ii
ii
tt
Nov. 4
ii
ii
it
Nov. 5
"
Type
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Light
Dark
Clear
Dark
Temp. °C
27.8
27/6
28.0
27.8
27.5
27.8
28.4
28.4
28.7
28.4
29.0
28.4
29.0
29.0
28.2
28.4
28.2
28.5
28.2
27.4
D.O.
Cmg/1)
6.0
ti
5.6
ii
5.2
"
5.8
tt
6.0
it
6.1
ti
7.3
ti
5.8
11
6.8
ii
5.1
it
Time interval
for rate
(hrs
1515 -
1510 -
1613 -
1609 -
1720 -
1725 -
1128 -
1239 -
1227 -
1239 -
1306 -
1239 -
1546 -
1512 -
1500 -
1445 -
1603 -
1600 -
0910 -
0914 -
0
1530
1528
1631
1633
1738
1737
1221
1330
1257
1330
1345
1330
1607
1600
1545
1545
1651
1650
0949
0950
Rate of
D.O. meta- GPP
/^
bolism
Vis
. light
10 z transmission
ID'2 mg/1/ mg/1/
min
-1.25714
-4.53571
-2.04285
-4.64444
-2.85714
-5.06666
-0.24471
-1.05478
-0.47619
-1.05478
-0.83023
-1.05478
-2.02333
-2.09375
-1.16507
-1.56557
-1.43677
-1.77456
+1.17998
-1.00000
min
'
2
£• •
.
o
\s •
n
\J *
0
\J •
0.
0.
2.
2786
<£• / \J\J
6016
\J\J Jk.\J
Rim
OJ.U1
5786
mj 1 \J\J
??46
£> £* T"VJ
0704
\J / \j *-f-
4005
3378
1800
at bottom
ID'2
2.
2.
1.
8.
7.
5.
1.
4.
3.
12.
ly/min
230
228
496
333
729
830
8731
844
310
444
a Fahka Union Bay - Mud-sand bottom
Fahkahatchee Bay - Mud-sand bottom
c Fahkahatchee Bay - Sand^Kud bottom
-------
APPENDIX R - METHODS, METABOLIC STUDIES
A general exposition of the methodology employed in this study for the
determination of plant community carbon and water dynamics has been presented
earlier in Chapter XII. In this appendix we detail some modifications of
standard methods developed to deal with local environmental conditions, and
describe the particular methodologies and equipment configurations utilized.
We attempted to incorporate most of the precautionary measures advocated by
Brown (1968), with appropriate modifications to adapt the procedures he
received for physiological investigations, to a field ecological study.
Our primary objective was to minimize the disturbance of the ecosystem
involved in our attempts to measure the dynamics of metabolic behavior. This
required, efforts to maintain environmental parameters of significance to plant
community metabolism at near-ambient levels, especially C02 concentration,
light levels, and temperature. In addition, mobile field deployment of sub-
stantial quantities of complex equipment required the development of a general
standardized procedure, designed to minimize potential sources of error in the
measurement system, although these were of unequal magnitude in the several
communities that were studied.
In general, the method consisted of isolating a segment of the plant com-
munity in a rigid, clear polyacetate chamber. The chambers were run as open
systems, and the gradient in C02 and water vapor concentration induced between
the inlet and outlet ports were analyzed in terms of grams of carbon and grams
of water fixed or released per unit time. The effect of the materials ,(0.5 cm
polyacetate) used for chamber construction on the ambient solar energy regime
was examined with an Isco Spectroradiometer, kindly lent for this purpose by
the University of Florida. All materials acted as neutral filters, and sup-
pressed incoming visible light radiation (400 to 700 nm) by approximately 3
percent. The chambers employed for studies of wet prairie, salt marsh, and
ground cover vegetation were of two types. The first was a cylindrical chamber
1 meter in diameter and 1 meter high. Occasionally, two of these chambers were
stac'ked for tall-grass communities, to give a single chamber 1 meter in diame-
ter and 2 meters high. The second type was an inverted "U" shaped chamber that
was 3 feet high, 2 feet wide, and 3 feet long. These chambers were often cou-
pled end-to-end to create one 6-foot chamber, especially for short grass prairies.
All of these chambers were constructed of welded steel rod (one-fourth inch in
diameter) to minimize shading in the chambers due to rigid structural elements.
For woody plants too large for these chambers, plastic sheaths were used to
enclose portions of the plant, and the dynamic behavior of the organs studied
was extrapolated to the entire stand from independent studies of the structural
properites of the stand. These sheaths ballooned to a rigid form when the air
supply was introduced to them; and under these conditions, measured about 1.2
meter in length by 0.6 meter in diameter.
The air flow to any given chamber was in almost all cases constant during
any particular diel run. Flow rates were commensurate with observed daytime
wind speeds and of sufficient magnitude to maintain five to six turnovers per
minute in the chambers. These turnover rates kept the temperature gradient
across the chamber within 4 to 5°C of ambient temperatures under the worst con-
ditions encountered (early afternoons of clear, hot summer days on dry short-
grass prairies) and usually limited noon temperature gradients to about 2°C.
XIX-41
-------
Carbon dioxide concei tration gradients generally were limited to 5 to 10 ppm,
thus eliminating artificial C02 starvation as a factor supressing observed net
photosynthesis. The air flow was introduced at the proximal end of the chamber,
near ground level, via a plastic container cut to provide divergence of the
incoming air stream at the inlet port. Outlet parts were located at the distal,
top side of the chamber and provided with clear plastic "chimneys" to minimize
turbulent air exchange through the port. Air flow rates were monitored daily,
with a hot-wire anemometer placed at a third-point in the inlet duct to measure
average flow. Cross-sectional area of the duct multiplied by flow rate then
gave air mass delivery rate. Generally, four chambers (often of varying type)
were deployed during any given diel run; and each was monitored for metabolic
activity over a 15-minute cycle every hour.
Calculation of carbon dynamics from the CC>2 gradient across the chamber is
given by (derived from Brown, 1968):
g C/hr = 0.14625 (Appm) (F) /T (Eq. 1)
where (Appm) is the value of the C02 gradient, F is the air flow rate in m3/hr,
and T is the temperature measured at the same point in the inlet duct as was
flow in °K.
The frequent occurrence of nighttime (ambient) air mass stagnation (due to
temperature inversions) posed special problems to this analysis. Under these
conditions, ambient C02 concentrations frequently exceeded 600 ppm and occasion-
ally exceeded 800 ppm during short time intervals. Given that noon concentrations
were usually 290 to 300 ppm, the simultaneous maintenance of high instrument
sensitivity (for chamber gradients) and an extremely large range (for monitoring
ambient concentration) offers certain technical difficulties in that the objec-
tives are basically incompatible. In addition, nighttime small-scale atmospheric
eddies introduced transient temporal C02 gradients as large as 50 ppm/min at the
inlet ports, which effectively quenched the 5-ppm signal of interest derived
from plant metabolism. The latter problem was resolved by deriving the source
air stream from an elevation of some 10 meters above ground level in the prairie
communities and about 4 meters above the canopy in wooded stands. In addition,
the source air was pumped to a 13.6-m3 buffer (tent) and then distributed to the
chambers by individual, electrically driven pumps located inside the buffer (Fig-
ure R-l).
Equal-length cables, each consisting of three lengths of 7.9-mm I.D. tygon
tubing, were used to conduct the air streams to the infared gas analyzer (IRGA,
Beckman model 215B). In order to maintain acceptable precision and range in
the analysis, a tripartite gas-flow sequence was used over each 15-minute meas-
urement cycle.
1. A standard bottled gas (300 ppm C02 in H2) was introduced to the ref-
erence side of the IRGA, and comparison with the ambient air stream established
the absolute level of C02 in the air. This part of the analysis was performed
using "Range 1", the least sensitive range of the instrument. (IRGA calibra-
tion for this sector of the analysis was accomplished by least squares regression
(parabolic model) of IRGA response to a series of bottle gases of differing CO^
concentration. Calibration procedures are given in more detail later in this
appendix. )
XIX-42
-------
LEGEND
(D DUCTING 10 em
RELATIVE HUMIDITY SENSOR
Figure R--1. Simplified schematic of gas flow control.
2. A selectroswitch, programmable timer then switched the input gases (Fig-
ure R-l) to flow ambient air to the reference side of the analyzer.and chamber
exhaust air to the "read" side of the IRGA. At the Same time, the timer switched
the IRGA from range 1 to range 3, the latter with a very small total range (about
60 ppm total span at 300 ppm baseline).
3. In the final step, ambient air was channeled to both-sides of the IRGA
to give a hard value for the "no difference" point in the readings. Range 3
calibration of the IRGA essentially involved parameterization (LSR) of instru-
ment sensitivity as a function of the C02 content in the reference cell.
This procedure provided knowledge of the absolute concentration of C02 in
the ambient air stream and the instrument deflection produced by passage of the
gas through the plant chamber (step 2) versus the "no difference" deflection
(if any—this step compensated for zero drift of the IRGA). With calibrated
values for the sensitivity of the instrument, it was possible to calculate the
absolute level of C02 in the chamber exhaust stream and, thus, the gradient
across the chamber. This gradient corresponds to the factor (Appm) in Eq. 1.
Success of this analytical scheme hinged on several key features of the
IRGA. First, the instrument is indifferent to which of the cells is regarded
as the "reference" and which the "read" cell, except that the meter gives off-
scale (negative) readings for one of the alternatives. The use of an external
recorder (Honeywell Electronik 194) permitted a setting of instrument "zero"
at mid-range on the recorder. In this way, daytime carbon fixation and night-
time carbon release could be monitored with equal facility. Secondly, range 3
(R3) amounts to nothing more than a porportional expansion of a given segment
of the range 1 (Rl) response curve. Treatment of a small length of this curve
(under the range of calibrations here utilized) as a straight-line segment
introduced an error of no more than 0.2 to 0.3 ppm and permitted efficient
parametrization of the IRGA sensitivity as a function of ambient C02 concen-
tration.
A manifold and a large vacuum pump was used to maintain constant purging
of the sampling lines, thus minimizing water condensation in the lines. One—way
XIX-43
-------
check valves prevented cross-contamination of chamber air streams within the
manifold. Pumps used to transport the air stream into the IRGA were diaphragm
pumps of a type that does not leak air into the sample stream or contaminate
the stream with oil, etc. Range 1 zero and span were checked daily by intro-
duction of standard gases to the IRGA in the field, and complete reparameterization
of IRGA response was performed approximately monthly. The same air flow rates
(and gauge pressures) were used for field analysis and for calibration and were
of sufficient magnitude to fully purge the instrument chambers during each phase
of the analytical cycle. Flow rates were maintained at stable (1.5 1/min) levels
by pumping gas into the IRGA chambers and frequently monitoring output flow rates
with permanently installed rotameters.
Compensation for interference by water vapor, due to light absorption at
infrared wavelengths within the analyzer, poses special problems to this kind
of analysis. In-line dessicant chambers are often used to predry the sample
streams prior to injection into the analyzer. However, there does not appear
to be any fully satisfactory drying agent available. Brown (1968) judged silica
gel to be inadequate for this application, and we found Drierite (CaSO^) to
be extremely reactive in terms of intrinsic CC>2 uptake and release properties
under the conditions of this analysis. This was despite extensive trials involv-
ing varying column sizes, preconditioning schedules, etc. Ultimately, we adapted
a procedure of relying upon the internal temperature control of the IRGA as a
means of preventing condensation in the analyzer chambers and did not predry
the air streams.
Calibration data provided by the manufacturer with the instrument indicate
an interference by water vapor of an apparent concentration change of 10.5 ppm
at 3.5 mol percent in range 1. This water vapor content is equivalent to that
of a fully saturated air stream at 87°F (B. R. Strain, personal communication)
run against a fully dried air stream. The error introduced to the Rl readings
of this study was therefore about 2 percent. In R3, the 10.5 shift represents
about 20 percent of full scale. However, the change in relative humidity between
the chamber input and output streams rarely exceeded 10 percent and was more
usually on the order of 4 to 7 percent. In general then, the mode of analysis
used in this study limited the range 3 error introduced by neglect of water vapor
interference to a point maximum of about 10 percent of the measured CC>2 value
under worst case conditions, with a more usual error of about 2 percent of the
measured value for both range 1 and range 3. Furthermore, the error was not
fully systematic; for under many conditions, small quantities of water vapor
condensation occurred in the chambers and were reflected in the readings.
Water export from the plant community was measured by routing the sample
stream from chamber inlet and outlet ports through relative humidity transducers
(Hygrodynamics 15-7012 W). Each sensor received a constant air supply from
either the-,inlet or outlet port during the entire 15-minute measurement cycle.
Output of one sensor was recorded (analog trace) for the first 8 minutes of the
cycle, at which point the output was electircally switched to the other sensor.
Values of the difference were read from charts at the switching point to give
fully synpotic and fully stabilized pairs of RH values for comparison of inlet
and outlet air streams. Each sensor had a paired thermocouple probe to estab-
lish the true temperature at the sensor. Saturated water vapor content at that
temperature was derived from logarithmic interpolation of the Smithsonian Tables
(Handbook of Chemistry ahd Physics, CRC Publishing Co.). Throughout the analy-
sis (both for C02 and for H20), atmospheric pressure was taken as a constant
XIX-44
-------
one bar. Water export rate during any 15 minute measurement cycle can then be
expressed as
g H2P/hr = f(RH)i ST_ - (RH)o sj F (Eq. 2)
O
where RH is relative humidity, S is saturated water content (g IL^O/m) at the
relevant temperature T, i and o refer to inlet and outlet chamber port air-
streams respectively, and F is the air flow delivery rate in naP/hr.
In the case of ground chambers, carbon and water export rates were corrected
by ground areas occupied by the chambers to give g C/m^-hr and g HoO/nr-hr,
respectively. In the latter case, conversion to mm of water export is directly
from the relationship
1 kg t^O/m^.hr = mm/hr .
All temperature records were made with copper constontan (Honeywell Type T)
thermocouples or polyvinyl over polyvinyl insulated extension wire (16 gauge,
#S-12876C) coupled to thermocouple reference junctions of a Honeywell multipoint
thermocouple recorder (Model 112).
A simplified schematic of the electrical control mechanisms run by the
programmable ("Selectroswitch") microtimer is given in Figure R-2. The timer
was a series of 14 gauged SPDT switches programmed by plastic blocks inserted
on a revolving drum (1 rph). Solenoid valves for gas flow control, RH sensor
outputs, etc., were switched, in the patterns described above, by this time.
In addition, positive identification of each step of the full hour cycle was
provided by recorder event marks programmed into the timer.
IRGA Calibration and Sample Calculations
Range 1 parameterization of the IRGA response was conducted by least
squares regression of a parabolic model on observed recorder response levels.
The IRGA performed in a differential mode, based on departures from a 300 ppm
zero point (i.e. 300 ppm COo in No on both "reference" and "read" sides of the
analyzer). The analyzer gain was set to give a meter deflection of 25 units for
the 100 ppm shift in concentration of 300 ppm (reference) vs 400 ppm ("read" or
"sample" cell). A sample calibration run is given in Table R-l, with the
resultant calibration curve in Figure R-3. The slight deviations of the
observed points from the calibration curve are primarily due to variation in the
COo content of the bottled gases. Composition of these gases as delivered is
rated at + 3 ppm. Also, caution must be used when bottle pressures get low, as
concentration shifts occur due to the differing molecular weights of the CO™
and carrier N2-
Calibration of range 3 was accomplished by parameterization of IRGA
sensitivity (expressed as ppm/scale division) against a series of standard gases
introduced to the reference cell. A closed system (Figure R-4) of known total
volume was injected serially with specific volumes of pure CO . To eliminate
cumulative errors, each injection was considered in terms of the concentration
change thereby induced, and paired with the corresponding meter or recorder
change. This resulted in a paired data set of (Appm/Ascale divisions), i.e.
sensitivity as ppm/scale division, vs the C02 concentration in the reference
XIX-45
-------
®
Q
o.
LEGEND
CHAMBER NO I LEAD
CHAMBER NO 2 LEAD
CHAMBER NO 3 LEAD
CHAMBER NO 4 LEAD
SOLENOID VALVE 3- WAY
SWITCHING TIMER
BATTERY 12 V DC
RELATIVE HUMIDITY SENSOR
RANGE CONTROL SENSOR
GAS ANALYZER
TEMPERATURE RECORDER
DUAL PEN RECORDER
EVENT PEN LEADS
EVENT PEN LEADS
RELATIVE HUMIDITY LEADS
EVENT MARK LEADS
Figure R-2. Simplified schematic of electrical controls.
Table R-l. Sample calibration of C02 analyzer range 1 using standard gases and
300 ppm reference.
Sample gas
ppm
200
300
350
400
Meter
reading
-------
600
500
4OO
I. 300
a.
zoo
100
C0» (ppm) =0.0252XE + I.I04IX + 184.529
X= RECORDER READING
RANGE I
10/28/72
10 20 30 40 5O 60
RECORDER READING
Figure R-3. Sample Range 1 calibration curve.
TO
80 90
100
It
LEGEND
STANDARD GAS
ROTAMETER
GAS ANALYZER
RECORDER
FILTER
PUMP
MIXING CHAMBER
PURE COg INJECTION
CD MANOMETER
Figure R-4. Schematic of closed-system range 3
IRGA calibration.
XIX-47
-------
cell. Table R-2 presents a simplified version of one calibration run utilizing
this procedure.
Table R-2. Calibration of C0? analyzer range 3 using closed system and standard
gas references,,
Reference
gas (ppm)
0
0
100
100
200
200
300
300
300
350
350
350
350
400
400
400
ppm per
Sample Sample
gas (cc) gas (ppm)
0
0.4
2.0
2.4
4.0
4.4
6.0
6.4
6.8
6.8
7.2
7.6
8.0
8.0
8.4
8.8
scale
0
20.139
100.695
120.834
201.390
221.528
302.084
322.223
342.362
342.362
362.501
382.640
402.779
402.779
422.918
443.057
division = ae
Meter Recorder
reading reading
<0
57.5
9.0
71.4
<0
31.9
<0
1.7
37.9
<0
<0
19.3
51.7
-2.0
29.7
61.3
bx
44.2
117.5
59.0
121.4
34.8
81.9
14.0
51.7
87.9
1.7
35.5
69,3
101.7
48.0
79.7
111.3
ppm per
scale division
20.139/73.3
20.139/62.4
20.139/47.1
20.139/37.7
20.139/36.2
20.139/33.8
20.139/33.8
20.139/32.4
20.139/31.7
20.139/31.6
= 0.275
= 0.323
= 0.428
= 0.534
= 0.556
= 0.596
- 0.596
= 0.622
= 0.635
a 0.637
x = ambient ppm
a = 2.
b = 2.
727 x lO'1
225 x 10~3
This data set was fitted to an exponential model by LSR. The calibration curve
resulting from the data of Table R-2 is given in Figure R-5.
Sample calculations incorporating the above calibration data, hypothetical
recorder outputs, and Eq. 1 are shown in Table R-3. All computations were
performed using the LSR equations programmed on a Monroe Model 1760 desk-top
calculator.
XIX-48
-------
o
>s
E
o.
o.
1.0
0.9
0.8
0.7
0.6
O
O 0.5
0.4
0.3
ppm/ SD = 0.273 e ao°" ™m AMB-
0.2
ai
RANGE 3
10/28/72
100 200 300
AMBIENT AIR (ppm)
400
Figure R-5. Sample range 3 calibration curve.
XIX-49
-------
Table R-3. Sample calculations of CC^ exchange in chambers.
Ambient to
Ambient to
sample, Chamber exhaust to sample
300 ppm to sample, ambient to and ambient
reference Ambient reference to reference Exhaust
(Range 1) ppm (Range 3) (Range 3) ppm
50.0
49.8
50.2
50.1
50.0
297.6
296.9
298.3
298.0
297.6
48.5
40.2
52.3
50.8
39.8
50.2
49.9
50.2
49.6
50.0
296.7
291.8
299.4
298.6
292.2
Temperature
of air Carbon
entering metabolism
chamber C gm C/hra
0
43.3
48.9
37.8
47.8
-0.0481
-0.2368
0.0505
0.0299
-0.2459
Flow = 100 nr/hr
Ambient ppmb = 0.0232X2 + 1.1041X + 184.329
X = recorder reading (range 1)
ppm/scale division13 = 0.273 e°-0022 Tppm amb.)
Appro = (Ascale divisions) (ppm/scale division)
Exhaust ppm = ambient ppm + Appm
gm C/hr = [(0.14625) (Appm)/V)- F
(Eq. 1) Negative values = P
Positive values = R
Taken from calibration data Tables
R-l and R-2.
XIX-50
-------
APPENDIX S - PHYSICAL DIMENSIONS AND RELATED BIOMASS (DRY WT) OF TEN SPECIES
INCLUDING SEEDLINGS.
Genus species DBH (cm)
Taxodium distichum
var. nutans (pond
cypress)
Sabal palmetto
(cabbage palm)
Acer rubrum*
(red maple)
Fraxinus caroliniana*
(pop ash)
Salix caroliniana*
(water willow)
14.3
24.5
1.2
18.8
10.6
22.9
24.9
23.5
19.0
19.2
28.3
23.1
20.1
24.2
23.5
27.0
15.6
21.7
30.3
15.2
13.2
7.6
8.5
5.3
9.1
7.9
10.3
5.4
3.3
4.6
3.4
14.2
15.2
18.4
9.8
7.9
11.2
12.1
8.3
9.7
15.6
18.8
17.7
18.4
Height (cm)
820
1,156
224
1,150
565
1,486
600
877
1,110
602
500
411.8
363
1,030
925
1,171
1,236
1,222.5
1,483
1,161
947
1,025
954
755
1,008
831
1,010
760
470
640
570
1,300
1,400
1,350
813
742
798
848
782
1,095
894
1,053
950
1,153
Wood mass (gm)
38,291.5
154,939.5
292.1
94,059.9
11,301.8
60,964.0
90,503.0
115,229.0
125,893.0
50,892.0
72,653.0
46,933.0
24,169.0
128,296.0
114,272.0
328,838.3
70,494.0
153,857.1
366,538.9
64,981.4
44,166.8
15,274.6
15,264.0
1,643.5
15,128.4
10,143.0
34,069.4
6,502.0
1,928.4
4,623.6
1,939.4
69,821.9
85,911.2
120,893.2
15,717.3
14,963.3
21,468.2
31,605.3
8,115.3
18,702.1
28,253.4
125,167.1
97,687.2
128,649.5
Green leaves (gm)
607.3
1,685.2
9.3
1,291.4
780.7
2,382.7
3,700.0
4,267.0
8,827.0
4,521.0
6,487.0
4,451.0
2,454.0
8,935.0
6,031.0
1,085.2
1,320.1
977.1
5,389.0
1,170.9
625.1
1,363.6
2,013.4
822.4
293.6
271.2
72.5
14.3
none
8.5
none
none
1,587.9
280.9
131.5
490.4
166.1
135.9
• 142.3
657.2
462.8
455.4
1,310.4
694.8
XIX-51
-------
APPENDIX S - PHYSICAL DIMENSIONS AND RELATED BIOMASS (DRY WT) OF TEN SPECIES
INCLUDING SEEDLINGS. CONT'D.
Genus species
Annona glabra
(pond apple)
DBH (cm)
15.6
6.5
12.5
19.5
14.2
15.3
7.8
7.2
Height (cm)
715
780-
819-
1,000
950
840
970
840
Wood mass (gm)
25,007.2
7,455.0
29,596.2
75,589.4
50,770.7
38,417.8
8,845.7
10,432.4
Green leaves (gm)
459.2
90.6
715.4
1,615.7
1,103.3
1,004.5
360.8
276.1
Diameter (on)
Myrica cerifera
(wax myrtle)
Baccharis halimifolia
(sea myrtle)
Magnolia virginiana
(sweet bay)
Persea borbonia
(red bay)
20 cm from
ground
5.1
5.2
4.4
2.0
1.8
4.2
5.9
6.5
8.1
1.8
4.0
4.6
6.6
1.22
0.71
1.63
0.91
0.66
0.53
2.38
1.05
0.63
0.32
0.49
3.70
1.29
1.52
1.51
2.58
0.57
0.62
0.33
4.25
260
260
260
190
160
350
400
400
400
270
300
250
410
160
143
230
100
92
60
260
132
44
40
38
332
198
177
230
283
90
80
59
560
XIX-52
2,412.2
2,895.6
1,233.8
364.8
220.8
2,204.2
5,236.4
5,759.2
12,361.8
219.2
1,113.1
1,385.0
4,092.8
98.5
22.6
204.3
22.8
11.2
4.1
264.8
32.6
4.0
1.8
2.2
955.8
99.5
101.5
105.5
379.4
9.6
7.9
2.1
2,818.8
333.3
403.7
179.6
33.3
42.7
106.3
607.3
759.5
1,784.4
21.6
120.8
314.5
557.7
22.7
8.7
66.0
8.8
6.0
2.2
69.5
8.3
0.3
0.5
0.6
187.6
35.4
56.2
25.6
73.1
5.8
5.0
1.4
700.5
-------
APPENDIX S - PHYSICAL DIMENSIONS AND RELATED BIOMASS (DRY WT) OF TEN SPECIES
INCLUDING SEEDLINGS. CONT'D.
Diameter (cm)
20 cm from
Genus species ground Height (cm) Wood mass (gm) Green leaves (gm)
Taxodium distichum* 2.65 219 126.8 none
var. nutans 1.80 169 58.6 none
(pond cypress)
* Acer rubrua, Fraxinus caroliniana, Salix caroliniana and Taxodium distichum are
deciduous trees, some individuals were taken during or after leaffall.
XIX-53
-------
APPENDIX T - REGRESSION EQUATIONS FOR WOODY BIOMASS BASED ON THE MODEL y = axb,
WHERE y IS TOTAL ABOVE GROUND WOODY BIOMASS IN GRAMS DRY WEIGHT (CONSTANT
WEIGHT AT 105° C). THE INDEPENDENT VARIABLE IS x = JD-H, WHERE D is DBH
IN CENTIMETERS AND H IS TOTAL TREE HEIGHT (CM) . SAMPLE SIZE AND CORRELATION
COEFFICIENT ARE GIVEN FOR EACH SPECIES INVESTIGATED.
Species
Acer rubrum
(red maple)
Annona glabra
(pond apple)
Baccharis halimifolia*
(sea myrtle)
Fraxinus caroliniana
(pop ash)
Magnolia virginiana*
(sweet bay)
Myrica cerifera*
(wax myrtle)
Persea borbonia*
(red bay)
Sabal palmetto
(cabbage palm)
Taxodium distichum
(cypress)
Taxodium distichum*
(cypress seedlings)
b
4.3145
3.7139
3.2742
3.0645
2.6315
3.1955
2.9426
2.6560
2.4953
2.3895
a r
4.7382 x 10"5 0.9852
8.7872 x 10"4 0.9653
1.3818 x 10"2 0.9952
2.1181 x 10"2 0.9978
5.2914 x 10~2 0.9985
1.983 x 10"2 0.9665
2.6116 x 10"2 0.9983
2.1394 x 10"1 0.9770
2.725 x 10"1 0.9834
6.328 x 10"2 1.0000
n
11
8
6
8
5
13
9
9
6
2
* Diameter taken 20 cm from ground. All others are calculated using DHB
diameter at 137 cm.
XIX-54
-------
APPENDIX U - REGRESSION EQUATIONS FOR LEAF BIOMASS BASED ON THE MODEL y = axb,
WHERE y IS TOTAL LEAF BIOMASS IN GRAMS DRY WEIGHT (CONSTANT WEIGHT AT
105° C). THE INDEPENDENT VARIABLE IS x = fD•H, WHERE D IS DBH IN
CENTIMETERS AND H IS TOTAL TREE HEIGHT (CM). SAMPLE SIZE AND CORRELATION
COEFFICIENT ARE GIVEN FOR EACH SPECIES INVESTIGATED.
Species
Acer rubrum
(red maple)
Annona glabra
(pond apple)
Baccharis halimifolia
(sea myrtle)
Fraxinus caroliniana
(pop ash)
Magnolia virginiana
(sweet bay)
Persea borbonia
(red bay)
Sabal palmetto
(cabbage palm)
* Some Acer rubrum were
Only 5 of the Fraxinus
b
1.2208
3.9190
2.6775
4.2056
2.7722
2.8700
1.5383
a
3.0284
7.7717 x 10"6
2.1218 x 10"2
3.8224 x 10"7
8.2460 x 10"5
4.5687 x 10"2
3.1299
undergoing leaf fall.
caroliniana trees had leaves .
r n
0,5432* 11
0.9476 8
0.9948 6
0.9346 5f
0.9666 5
0.9825 9
0.7709 9
At the time of
harvest 3 were leafless.
-------
APPENDIX V - REGRESSION EQUATIONS FOR SABAL USING TOTAL HEIGHT,
barn
(Y = axb, Y = wood mass, x = height (m))
6.843 1.2687 0.9087 9
(Y = ax + b, Y = leaf mass (kg), x = height (m))
1.0979 0.6199 9
XIX-56
-------
APPENDIX W - DECOMPOSITION OF LITTER CONFINED TO FINE MESH BAGS PLACED ON FOREST
FLOOR AND SUBJECTED TO INUNDATION DURING THE HYDROPERIOD, CENTRAL FAHKA-
HATCHEE STRAND, 1972,
No. of days
material
subject
to decomp,
0
61
121
182
243
311
366
Vegetative
component
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Dry wt.
(gm)
23.6
11.6
1.2
36.4
21.9
8.1
0.36
30.36
14.3
9.3
0.73
24.33
15.2
8.8
-
24.0
16.15
8.8
24.95
12.5
8.3
1.1
21.9
12.5
8.3
20.8
Ash free wt.
Cgm)
21.89
10.99
1.10
33.98
19.57
7.90
12.90
9.15
13.34
8.60
21.94
14.06
8.59
22.65
11.11
8.07
10.63
8.08
18.71
Organic
nitrogen
(gm)
0.2596
0.0821
0.0229
0 . 3646
0.3307
0.0278
0.0061
0.3646
0.2475
0.0294
0.2622
0.0387
0.3009
0.3487
0.0271
0.3758
0.2310
0.0262
0.0246
0.2818
0.2875
0.0299
0.3174
XIX-57
-------
APPENDIX X - DECOMPOSITION OF LITTER CONFINED TO COARSE MESH BAGS PLACED ON
FOREST FLOOR AND SUBJECTED TO INUNDATION DURING HYDROPERIOD, CENTRAL
FAHKAHATCHEE STRAND, 1972.
No. of days
material
sub j ect
to decomp.
0
61
121
182
243
311
366
Vegetative
component
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Dry wt.
(gm)
23.6
11.6
1.2
36.4
17.5
9.9
0.60
28.00
13.9
9.5
0.59
23.99
11.7
8.3
0.59
20.59
10.9
8.2
19.10
7.1
7.6
14.7
11.0
6.85
17.85
Ash free wt.
Cgm)
21.89
10.99
1.10
33.98
15.62
9.51
12.63
9.36
10.37
7.26
0.50
18.13
9.46
8.00
17.46
6.27
7.42
13.69
9.44
6.63
16.07
Organic
nitrogen
(gm)
0.2596
0.0821
0.0013
0.3430
0.2899
0.0941
0.0093
0.3933
0.2449
0.0390
0.0116
0.2955
0.2064
0.0365
0.2457
0.0476
0.2933
0.1372
0.0300
0.1672
0.2629
0.0379
0.3008
XIX-58
-------
APPENDIX Y - DECOMPOSITION OF LITTER CONFINED TO FINE MESH BAGS PLACED ON DEBRIS
PILES AND NOT SUBJECTED TO INUNDATION DURING HYDROPERIOD, CENTRAL FAHKA-
HATCHEE STRAND, 1972.
No. of days
material
subject
to decomp.
0
64
120
193
305
366
Vegetative
component
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Dry wt.
Cgm)
20.5
3.6
1.4
25.5
16.9
3.1
1.22
21.22
17.4
2.9
-
20.3
12.45
2.00
2.10
16.55
13.3
2.1
-
15.4
10.6
1.4
2.8
14.8
Ash free wt.
(gm)
18.80
3.41
1.31
23.52
15.20
3.02
15.59
2.80
18.39
10.93
1.89
1.88
14.70
13.78
Organic
nitrogen
Cgm)
0.2255
0.0255
0.0267
0.2777
0.2073
0.0174
0.0204
0.2451
0.2339
0.0155
0.2494
0.1869
0.0151
0.0341
0.2361
0.2341
XIX-59
-------
APPENDIX 2 - DECOMPOSITION OF LITTER CONFINED TO COARSE MESH BAGS PLACED ON
DEBRIS PILES AND NOT SUBJECTED TO INUNDATION DURING HYDROPERIOD, CENTRAL
FAHKAHATCHEE STRAND, 1972.
No. of days
material
subject
to decomp.
0
64
120
193
305
366
Vegetative
component
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Leaf
Wood
Fine
Total
Dry wt.
(g"0
20.5
3.6
1.4
25.5
16.4
3.4
0.74
20.54
17.5
2.98
-
20.48
16.0
2.57
0.65
19.22
14.7
2.68
-
17.38
12.7
2.3
1.2
16.2
Ash free wt.
(PO
19.01
3.41
1.28
23.70
14.64
3.22
15.87
2.84
-
18.71
14.26
2.49
0.58
17.33
15.37
Organic
nitrogen
(gm)
0.2255
0.0255
0.0267
0.2777
0.1630
0.0228
0.0087
0.1945
0.2107
0.0235
-
0.2342
0.2110
0.0231
0.0106
0.2447
0.2294
XIX-60
-------
APPENDIX AA - TREES AND SHRUBS OF THE CYPRESS STMND.
Key to Table
Abundance
very rare
rare
infrequent
abundant
very abundant
Growth habit
less than 1 in 1000
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 1000
Substratum
Organic soil ~~\strand
Sandy organic soils_Jedge
Sandy organic, sandyjsttand
and rocky soils _/interior
large tree -
tree -
small tree -
large shrub -
shrub -
small shrub -
over 10 meters
5 to 10 meters
3 to 5 meters
2 to 3 meters
1 to 2 meters
less than 1 meter
Genus species
Growth habit
Abundance
Substrate
Roystonea elata
(Royal palm)
Sabal palmetto
(Cabbage palm)
Myrica cerifera
(wax myrtle)
Taxodium distichum
(Bald-cypress)
Salix caroliniana
(Willow)
Ficus aurea
(Strangler fig)
Magnolia virginica
(Sweetbay)
Annona glabra
(Pond apple)
Persea barbonid
(Red bay)
Itea virginica
(Virginia willow)
large tree
very rare to rare
small tree to tree rare to infrequent
shrub to small
tree
small to large
tree
shrub to large
tree
shrub to tree
shrub to tree
abundant to very
abundant
abundant to very
abundant
infrequent to
abundant
rare to infrequent
very rare to rare
small tree to tree infrequent to
abundant
shrub to small
tree
shrub to large
shrub
infrequent to
abundant
organic to sandy or-
ganic soil
organic, rocky, and
sandy soil
only on sandy organ-
ic, sandy and rocky
soil
organic, sandy or-
ganic, rocky and
sandy soil
organic soil
organic, rocky and
sandy soil, debris
and other plants
organic, sandy or-
ganic, rocky or
sandy soil
organic soil
organic, rocky or
sandy soil
very rare to rare organic soil
XIX-61
-------
APPENDIX AA - TREES AND SHRUBS OF THE CYPRESS STRAND. CONT'D.
Genus species
Growth habit
Abundance
Substrate
Quercus nigra
(Water oak)
Ximenia americana
(Hog plum)
Chrysobalanus icaco
(Coco plum)
Ilex cassine
(Dahoon holly)
Acer rubrum
(Red maple)
Cornus foemina
(Dogwood)
Myrsine quianensis
Ardisia escallo-
nioides
Fraxinus caroliniana
(Ash)
Cephalanthus occi-
dentalis (Button
bush)
Hamelia patens
(Fire bush)
Psychotria unudata
(Coffee bush)
Psychotria salzneri
(Coffee bush)
large tree
shrub to small
tree
shrub to small
tree
shrub to large
tree
tree to large
tree
small tree
small to large
shrub
small to large
shrub
large shrub to
tree
shrub to small
tree
small shrub to
shrub
small shrub to
shrub
small shrub to
shrub
very rare to rare
rare to infrequent
rare to infrequent
very rare to rare
abundant
organic, sandy organ-
ic soil
organic, rocky or
sandy soil debris
organic, rocky or
sandy soil debris
organic, rocky or
sandy soil
organic to sandy
organic soil
very rare to rare organic soil
infrequent to
abundant
rare to infrequent
organic, rocky and
sandy soil debris
organic soil and
debris
rare to infrequent organic soil
very rare to rare organic or sandy
organic soil
very rare
infrequent to
abundant
infrequent to
abundant
sandy organic soil
organic, sandy or-
ganic soil and debris
organic, sandy or-
ganic soil and debris
XIX-62
-------
APPENDIX BB - VINES OF THE CYPRESS STRAND.
Key to Table
Abundance
very rare
rare
infrequent
abundant -
very abundant -
Growth habit
very small
small
medium
large
- less than 1 in 1000
- 1 to 5 per 1000
- 5 to 50 per 1000
- 50 to 500 per 1000
- over 500 per 100
less than 2 mm diameter
2 to 5 mm diameter
5 to 10 mm diameter
over 10 mm diameter
liana - vines reaching the top of the strand canopy
vine - vines not usually reaching the top of the strand canopy
Genus species
Growth habit
Abundance
Substrate
Smilax laurifolia medium to large
(Green brier) liana
Smilax bona-nox
(Green brier)
medium liana
Passiflora suberosa very small her-
(Passion flower) baceous vine
Passiflora pallens
(Passion flower)
Sarcostemma clausa
(Milkweed vine)
Melothria pendula
(Wild cucumber)
medium to large
herbaceous vine
small to medium
herbaceous vine
very small her-
baceous vine
Momordica charantia small herbaceous
(Wild balsam apple) vine
Cucurbita okeecho-
beensis (Indian
pumpkin)
Ipomoea alba
(Moon vine)
large herbaceous
vine
massive, large
woody liana
rare to abundant
rare to abundant
very rare
very rare
rare to abundant
very rare to rare
very rare to rare
very rare
organic, sandy organ-
ic, rocky and sandy
soil
organic sandy, organ-
ic rocky and sandy
soil
sandy organic, sandy
and rocky soil
sandy organic soil
sandy organic, sandy
and rocky soil
sandy and rocky soil
and debris
sandy soil
sandy and rocky soil
rare to abundant sandy organic, sandy,
and rocky soil
XIX-63
-------
APPENDIX BB - VINES OF THE CYPRESS STRAND. CONT'D.
Genus species
Growth habit
Abundance
Substrate
Hippocratea volubilis large woody liana
Ampelopsis arborea
(Pepper vine)
Parthenocissus quin-
quefolia (Virginia
creeper)
Cissus sicyoides
(Opossum grape)
Vitis rotundifolia
(Grape)
Vitis aestivalis
(Grape)
Toxicodendron radi-
cans (Poison ivy)
small woody to
herbaceous vine
small course her-
baceous vine
abundant to very
abundant
very rare to rare
rare to infrequent
small to large rare
woody vine or liana
small to large
woody liana
small to large
woody liana
small to large
woody liana
Mikania batatifolia small herbaceous
(Hemp vine) vine
rare to infrequent
very rare to rare
abundant to very
abundant
rare to abundant
organic soil
sandy organic and
organic soil
sandy organic, sandy
and rocky soil
organic and sandy
organic soil
organic and sandy
organic soil
organic and sandy
organic soil
organic, sandy or-
ganic, sandy and rocky
soil
sandy organic, sandy,
and rocky soil, debris
XIX-64
-------
APPENDIX CC - TERRESTRIAL HERBS OF THE CYPRESS STRAND (EXCLUDING VINES)
Key to Table
Abundance
very rare -
rare -
infrequent -
abundant -
very abundant -
Growth habit
very small
small
medium
large
very large
less than 1 in 1000
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 100
less than 10 cm
10 to 50 cm
50 to 100 cm
100 to 200 cm
over 200 cm
Genus species
Growth habit
Abundance
Substrate
Osmunda regalis
(Royal fern)
medium to large
Campyloneurum phylli- medium
tidis (Strap fern)
Nephrolepsis exal- medium to large
tata (Boston fern)
Acrostichum danaeae- large
folium (Leather fern)
Pteridium aquilinum medium to large
(Bracken fern)
Blechnum serrulatum medium
Woodwardia virginica medium
(Virginia chain fern)
Thelypteris kunthii small to medium
(Swamp fern)
Chloris neglecta
(Finger grass)
Panicum sp.
(Panic grass)
medium to large
very small to
small
very rare to rare sandy organic soil,
deb re
very rare to rare debris
rare to abundant
very rare
very rare
abundant to very
abundant
very rare
infrequent to
abundant
infrequent to
abundant
very rare to rare
sandy organic soil
and debris
sandy organic
sandy organic, sandy
and rocky soil
sandy organic, sandy
and rocky soil, debris
sandy and rocky soil
sandy organic soil
and debris
sandy organic soil
sandy organic soil
and debris
XIX-65
-------
APPENDIX CC - TERRESTRIAL HERBS OF THE CYPRESS STRAND (EXCLUDING VINES). CONT'D.
Genus species
Growth habit
Abundance
Substrate
Andropogon virgini- medium to large
cus (Broom sedge)
Arundo donax
(Giant reed)
large to very
large
Tripsacum dacty- large
loides (Grama grass)
Cyperus planti-
folius
Cladium jamaicen-
sus (Saw grass)
Malaxis spicata
(Ground orchid)
medium to large
medium to large
very small to
small
Boehmeria cylin- small to medium
drica (Button hemp)
Kosteletzkya altha- medium to large
eifolia
Lythrum lineare
(Loose strife)
Asclepias incarnata medium to large
(Milkweed)
Lantana camara
Callicarpa ameri-
cana (French mulberry)
Lippia nodiflora
(Capeweed)
Hyptis alata
(Bittermint)
Pluchea rosea
(Marsh fleabane)
Pluchea purpuras-
cens (Camphorweed)
medium to large
medium to large
very small
medium to large
medium
medium
Solidago angustifolia large
(Golden rod)
Aster caroliniensis large
(Climbing aster)
infrequent to
abundant
rare to infrequent
very rare to rare
sandy organic, sandy
and rocky soil
sandy organic, sandy
and rocky soil
sandy organic, sandy
and rocky soil
very rare to rare sandy and rocky soil
infrequent
sandy organic, sandy
and rocky soil
very rare to rare damp organic debris
infrequent
sandy and rocky soil
debris
rare to infrequent sandy soil
medium to large infrequent
rare
very rare to rare
rare
rare
rare
rare
rare
sandy organic, sandy
and rocky soil
sandy organic, sandy
and rocky soil
sandy organic soil
sandy organic soil
organic, sandy organ-
ic soil
sandy organic soil
sandy and rocky soil
sandy and rocky soil
very rare to rare sandy organic soil
very rare to rare
sandy organic and
sandy soil
XIX-66
-------
APPENDIX DD - EPIPHYTIC HERBS OF THE CYPRESS STRAND,
Key to Table
Abundance
very rare -
rare -
infrequent -
abundant -
very abundant -
Growth habit
very small
small
medium
large
very large
less than 1 in 1000
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 1000
less than 2 cm
2 to 10 cm
10 to 20 cm
30 to 100 cm
over 100 cm
Genus species
Growth habit
Abundance
Substrate
Psilotum nudum medium
(Wish fern)
Vittaria lineata medium
(Shoestring fern)
Polypodium polypo- small to medium
diodes (Resurrection
fern)
Phlebodium aureum
(Golden polypody)
medium to large
Tillandsia usneoides large
(Spanish moss)
Tillandsia fascicu- large
lata (Common wild pine)
Tillandsia setacea
(Needle leaf bro-
meliad)
small
Tillandsia utricu- large to very
lata (Giant wild pine) large
Tillandsia circinata small to medium
(Twisted wild pine)
Tillandsia balbisi-
ana (Reflexed wild
pine)
small to medium
very rare to rare
very rare to rare
very rare to rare
rare
Sabal and cypress
Sabal
Pond apple and ash
Sabal
very rare to rare Cypress and hardwoods
abundant to very
abundant
rare to infrequent
very rare
very rare to rare
very rare to rare
Cypress and hardwoods
Cypress, pond apple,
ash and other hard-
woods
hardwoods
Cypress, pond apple,
ash and other hard-
woods
Cypress, pond apple,
ash and other hard-
woods
XIX-67
-------
APPENDIX DD - EPIPHYTIC HERBS OF THE CYPRESS STRAND. CONT'D.
Genus species
Growth habit
Abundance
Substrate
Tillandsia valenzue- small to medium
lana
very rare to rare
Harrisella porrecta very small (leaf- very rare
(Orchid)
less)
Polyrrhiza lindenii very small to
(Ghost orchid) small (leafless)
rare
Epidendrum anceps
(Brown epidendrum)
medium to large rare to infrequent
Epidendrum rigidum small
(Matted epidendrum)
rare to infrequent
Epidendrum difforme small to medium rare to infrequent
Encyclia tampensis medium to large
(Butterfly orchid)
Encyclia cochleata medium to large
(Shell orchid)
infrequent to
abundant
very rare to rare
Cypress, ash, pond
apple and other small
hardwoods
Wax myrtle
Pond apple, ash and
other hardwoods
Pond apple, ash and
other hardwoods
Pond apple, ash and
other hardwoods
Pond apple, ash and
other hardwoods
Pond apple, ash,
other hardwoods and
cypress
Pond apple, ash and
other hardwoods
rrx-68
-------
APPENDIX EE - AQUATIC HERBS OF THE CYPRESS STRAND OR CYPRESS SLOUGH.
Key to Table
Abundance
very rare
rare
infrequent
abundant
very abundant
Growth habit
less than 1 in 1000
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 1000
small - less than 20 cm
medium - 20 to 100 cm
large - over 100 cm
Genus species
Growth habit
Abundance
Substrate
Rhynchospora inundata medium to large
(Beak rush)
Pontederia lanceo- medium to large
lata (Pickerel weed)
Juncus polycephalus medium
(Rush)
Sagitarria lancifolia medium to large
(Arrowhead)
Sagittaria graminea small to medium
(Arrowhead)
Saururus cernuus medium to large
(Lizard's tail)
Polygonum hydropip-
eroides (Smartweed)
Nasturtium officin-
ale (¥ater cress)
Ludwigia repens
medium to large
medium to large
small to medium
Proserpinaca palus- small
tris (Mermaid weed)
very rare to rare
infrequent
very rare to rare
infrequent
rare to infrequent
rare to infrequent
rare to infrequent
very rare to rare
abundant to very
abundant
infrequent to
abundant
rooted emergent in
standing water
rooted emergent in
standing water
rooted emergent in
shallow water
rooted emergent in
standing water
rooted emergent or
emergent
rooted emergent in
standing water or
soggy soil
rooted emergent in
shallow to moderately
deep water
rooted submergent or
emergent in shallow
water
rooted submergent in
shallow to moderately
deep water
rooted emergent or
submergent in shallow
water or soggy soil
XIX-69
-------
APPENDIX EE - AQUATIC HERBS OF THE CYPRESS STRAND OR CYPRESS SLOUGH. CONT'D.
Genus species
Growth habit
Abundance
Substrate
Centella erecta
small
Hydrocotyle umbell- small
ata
Bacopa caroliniana small to medium
Utricularia inflata small
(Bladder wort)
Biodia virginiana small
(Button weed)
infrequent to
abundant
very rare to rare
infrequent to
abundant
very rare to rare
infrequent to
abundant
rooted emergent in
soggy soil
rooted emergent in
soggy soil or shallow
water
rooted submergent in
shallow to moderately
deep water
free-floating in
shallow to moderately
deep water
rooted emergent in
soggy soil or very
shallow water
XIX-70
-------
APPENDIX FF - PRAIRIE PLANTS.
Key to Table
Abundance
very rare -
rare -
infrequent -
abundant -
very abundant -
Growth habit
very small
small
medium
large
very large
less than 1 per 1000 individuals
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 1000
less than 10 cm
10 to 50 cm
50 to 100 cm
100 to 200 cm
over 200 cm
Monocots of the Prairie
Genus species
Growth habit
Abundance
Location
Sagittaria lanci-
folia (Arrowhead)
Sagittaria graminea
(Arrowhead)
medium to large
emergent aquatic
small to medium
submergent or em-
ergent aquatic
Muhlenbergia capill- medium to large
aris
Spartina bakerii
(Cord grass)
Setaria corrugata
(Foxtail grass)
Panicum hemitomen
(Maiden cane)
Andropogon virgini-
cus (Broom sedge)
Andropogon sp.
Manisuris rugosa
(Necklace grass)
medium to large
small to medium
medium to large
emergent aquatic
large
medium to large
small to medium
very rare to rare
rare
abundant to very
abundant
infrequent
infrequent to
abundant
prairie and prairie
depressions
prairie and prairie
depressions
throughout prairie,
prairie depressions
and elevations
prairie and prairie
depressions
prairie and prairie
elevations
very rare to rare prairie depressions
infrequent
infrequent
rare to infrequent
prairie and prairie
elevations
prairie and prairie
elevations
prairie and prairie
elevations
XIX-71
-------
APPENDIX FF - PRAIRIE PLANTS. CONT'D.
Genus species
Growth habit
Abundance
Location
Eleocharis cellulosa
(Spike rush)
Eleocharis atropur-
purea
Fuirena longa
Dichromena colorata
(White-eyed grass)
Cladium jaraaicensis
(Saw grass)
Rhynchospora inun-
data (Beak rush)
Rhynchospora sp.
Schoenus nigricans
Xyris sp.
Pontederia lanceo-
lata
Juncus biflorus
Juncus polycepha-
lus
Hypoxis juncea
Aletris lutea
Hymenocallis pal-
meri (Spider lily)
Crinum americanum
(Swamp lily)
Sisyrinchium sp.
small to medium
emergent aquatic
very small to
small
small to medium
small
large
medium to large
emergent aquatic
medium
medium to large
small to medium
medium to large
emergent aquatic
medium to large
medium to large
emergent aquatic
very small to
small
medium
small to medium
small to medium
small to medium
very rare to rare
prairie and prairie
depressions
rare to infrequent prairie
infrequent
infrequent
infrequent to
abundant
very rare to rare
prairie
prairie
prairie and prairie
depressions
mostly prairie de-
pressions
rare to infrequent prairie
infrequent to
abundant
very rare
very rare
throughout prairie
and in prairie de-
pressions
prairie
prairie depressions
very rare to rare prairie and prairie
depressions
rare to infrequent prairie and prairie
depressions
very rare
very rare
infrequent to
abundant
prairie
prairie and prairie
elevations
prairie
rare to infrequent prairie
infrequent to
abundant
prairie
XIX-72
-------
APPENDIX FF - PRAIRIE PLANTS. CONT'D.
Genus species
Growth habit
Abundance
Location
Calopogon pulchellus medium to large
(Marsh pink orchid)
Spiranthes lacinia- medium
ta (Lady's trusses
orchid)
very rare to rare prairie
very rare prairie
Dicots of the Prairie
Cassytha filiformis small parasitic
(Love vine) vine
Polygala baldwinii small
Polygala grandiflora small
Linum carteri var-
iety smallii
Stillingia syl-
vatica
Proserpinaca palus-
tris
Eryngium aromaticum
Eryngium yuccifol-
ium
Eryngium prostra-
tum
Centella erecta
very small to
small
small to medium
to large
small emergent or
submergent aquatic
medium
medium
small vine like
herb
very small
Oxypolis filiformis medium to large
Sabatia grandiflora small
Sabatia bartramii small to medium
Asclepias longifolia slender medium
(Milkweed) in height
infrequent to
abundant
very rare to rare
rare to infrequent
very rare to rare
prairie and prairie
elevations
prairie and prairie
elevations
throughout prairie
scattered through
the prairies
very rare to rare throughout prairie
very rare to rare
very rare to rare
rare to infrequent
very rare
prairie and prairie
depressions
prairie
prairie
prairie elevations
rare to infrequent prairie and prairie
depressions
rare to infrequent throughout prairie
rare (seasonal) throughout prairie
very rare (seasonal) throughout prairie
very rare (seasonal) throughout prairie
XIX-73
-------
APPENDIX FF - PRAIRIE PLANTS. CONT'D.
Genus species
Growth habit
Abundance
Location
Asclepias lanceo- slender medium to
lata {Milkweed) large in height
Asclepias incarnata medium to large
(Milkweed)
Heliotropium poly- very small to
phyllum small
Physostegia denti-
culata
Bacopa caroliniana
Buchnera elongata
medium
very small to
small submergent
aquatic
small
Agalinis purpurea medium
Elytraria corolin-
iensis variety an-
gustifolia
small
Borrefia terminalis medium
Lobelia glandulosa small to medium
Pluchea rosea
Pluchea purpuras-
cens
Rudbechia sp.
Coreopsis leaven-
worthii
Solidago angusti-
folia (Goldenrod)
small to medium
small to medium
small
small to medium
large
very rare to rare
(seasonal)
very rare
infrequent
(seasonal)
very rare
rare (seasonal)
infrequent
very rare
(seasonal)
rare to infrequent
(seasonal)
very rare
(seasonal)
infrequent to
abundant
very rare to rare
rare
very rare to rare
rare to infrequent
very rare to rare
(seasonal)
throughout prairie
mostly low areas
of prairie
throughout prairie
and prairie ele-
vations
mostly prairie
elevations
mostly prairie de-
pressions and low
lying prairies
mostly elevated
prairies near pine-
lands
mostly elevated
prairies riear pine-
lands
mostly near pinelands
on drier elevated
prairies
mostly elevated
prairies near pine-
lands
common throughout the
prairies
mostly on drier
prairies
mostly on drier
prairies
mostly near pinelands
throughout all prair-
ies
mostly near cypress
strands
XIX-74
-------
APPENDIX FF - PRAIRIE PLANTS. CONT'D.
Genus species
Growth habit
Abundance
Location
Solidago sp.
(Goldenrod)
Mikania batati-
folia (Hemp vine)
Liatrus garberi
(Blazing star)
large to very large generally very rare
extensive herba-
ceous vine
small to medium
rare to infrequent
very rare
(seasonal)
extensive in some
prairies
widely scattered in
all prairies
mostly near pine-
lands or palm
hammocks
Vernonia blodgettii
(Ironweed)
Cirsium horridulum
(Giant thistle)
Flaveria latifolia
(Rabbit weed)
Eupatorium mikan-
ioides
medium to large very rare
medium to large,
very stout
small to medium
very rare to rare
(seasonal)
infrequent to
abundant
small to medium very rare to rare
mostly drier rocky
or sandy sites near
pinelands
scattered over most
prairies
found in patches on
most prairies partic-
ularly the drier
areas
mostly on drier
prairies
XIX-75
-------
APPENDIX GG - TREES AND SHRUBS OF THE SANDY MARL PINE AND PALM ASSOCIATIONS.
Key to Table
Abundance
very rare •
rare -
infrequent -
abundant -
very abundant -
Growth habit
large tree
tree
small tree
large shrub
shrub
small shrub
less than 1 per 1000
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 1000
- over 10 meters
- 5 to 10 meters
- 3 to 5 meters
- 2 to 3 meters
- 1 to 2 meters
- less than 1 meter
Genus species
Growth habit
Abundance
Substrate
Pinus elliottii var.
densa (Slash pine)
Sabal palmetto
(Cabbage palm)
Serenoa repens (Saw
palmetto)
Myrica cerifera
(Wax myrtle)
medium to large
tree
small, medium or
large tree
low growing pros-
trate shrub
medium shrub,
large shrub to
small tree
abundant to very
abundant
abundant to very
abundant
infrequent to
abundant
infrequent,
abundant to very
abundant
soil
soil
higher more elevated
soil
soil
Ficus aurea
(Fig)
Magnolia virgini-
ana (Sweet bay)
Persea borbonia
(Red bay)
Ilex cassine
(Dahoon holly)
small shrub to
small tree
medium shrub, large
shrub to small
tree
very rare to rare
very rare to rare
medium to large very rare, rare to
shrub infrequent
large shrub, small very rare (mostly
tree or medium on the pine and
tree palm islands of
prairies)
soil debris piles and
often epiphytic on
the Sabal palmetto
soil, often more
elevated soil
soil
soil, often more
elevated soil
XIX-76
-------
APPENDIX GG - TREES AND SHRUBS OF THE SANDY MARL PINE AND PALM ASSOCIATIONS. CONT'D.
Genus species
Growth habit
Abundance
Substrate
Ilex glabra
(Holly)
Lyonia ferruginea
Myrsine quianensis
Bumelia reclinata
Baccharis halimi-
folia (Sea myrtle)
medium shrub
medium to large
shrub
medium to large
shrub
large shrub,
small tree, or
medium tree
small, medium, or
large shrub
very rare (mostly on
the pine and palm soil, often more
islands of prairies) elevated soil
very rare (mostly on
the pine and palm
islands of prairies) soil
rare to infrequent soil
very rare to rare
mostly on more
elevated soil
rare, infrequent or
abundant soil
XIX-7 7
-------
APPENDIX HH - HERBACEOUS PLANTS OF THE WET PINE AND PALM ASSOCIATIONS,
Key to Table
Abundance
very rare -
rare -
infrequent -
abundant -
very abundant -
Growth habit
very large
large
medium
small
very small
less than 1 per 1000
1 to 5 per 1000
5 to 50 per 1000
50 to 500 per 1000
over 500 per 1000
over 200 cm
100 to 200 cm
50 to 100 cm
10 to 50 cm
less than 10 cm
Genus species
Growth habit
Abundance
Substrate
Phlebodium aureum
(Golden polypody
fern)
small to medium
Blechnum serrulatum
(fern) small to medium
Muhlenbergia capil-
laris medium
Setaria geniculata
(Foxtail grass) small to medium
Paspalum monosta-
chyum
medium
Andropogon virgini-
cas (Broom sedge) medium to large
Andropogon sp. medium
Manisuris rugosa
(Necklace grass) medium
Dichromena colorata small to medium
very rare to rare
epiphytic on Sabal
palmetto
sandy organic soil,
very rare to rare prefers some shade
abundant
infrequent to
abundant
abundant
very rare to rare
rare
rare to
infrequent
very rare to rare
soil, prefers sun
soil, prefers some
direct sunlight
soil, prefers a
well lighted area
soil, tolerates
partial shade
soil, tolerates
some shade
soil, prefers sun
mostly rocky sandy
areas, tolerates'
partial shade
XIX-78
-------
APPENDIX HH - HERBACEOUS PLANTS OF THE WET PINE AND PALM ASSOCIATIONS. CONT'D.
Genus species
Growth habit
Abundance
Substrate
Rhynchospora grayii medium
Tiliandsia fascicu-
lata (Common wild
pine)
Tiliandsia balbisi-
ana (Reflexed wild
pine)
Smilax laurifolia
(Green brier)
Aletris lutea
small
woody vine
small to medium
(leaves basal)
Polygala grandiflora small to medium
Kosteletzkya altha-
lifolia medium to large
Eryngium yuccifolium medium
very small to
small, somewhat
Eryngium prostratum viney at times
Centella erecta
very small
small to medium
Sabatia grandiflora (slender herb)
Trichostema dicho-
tomum medium to large
Buchnera elongata small
small slender
Hedyotis fasiculata herb
Elytraria carolini-
ensis variety angus-
tifolia small
rare to infrequent
mostly rocky, sandy
soil, prefers sun
small to medium very rare to rare epiphytic on pine
very rare to rare epiphytic on pine
mostly on shrubs and
very rare to rare small trees
very rare
mostly in sunny areas
mostly sunny to part-
rare to infrequent ly shaded areas
very rare to rare tolerates some shade
rare, infrequent to
abundant
very rare to rare
very rare to rare
rare to infrequent
(seasonal)
very rare
tolerates moderate
shade, prefers some
direct light
mostly on rocky
surfaces, needs
some sun
tolerates shade,
needs moisture
soil needs good
light
needs good light
rare to infrequent needs some direct
(varies seasonally) light
very rare
very rare
shallow soil, needs
good light
shallow soil, needs
some direct sun
XIX-79
-------
APPENDIX HH - HERBACEOUS PLANTS OF THE WET PINE AND PALM ASSOCIATIONS. CONT'D.
Genus species
Growth habits
Abundance
Substrate
Lobelia glandulosa
Coreopsis leaven-
worthii
Melanthera aspera
var. glabrius-
cula
Rudbeckia fulgida
(Coneflower)
Pluchea rosea
(Marsh fleabane)
Mikania batatifolia
(Hemp vine)
Liatrus garberi
(Blazing star)
Eupatorium mikani-
oides
small to medium
(slender herb)
small
medium to large
medium
medium
Pluchea purpurascens
(Camphorweed) medium
slender herba-
ceous vine
medium to large
medium
rare (seasonal)
shallow rocky soil,
needs sun
very rare to rare shallow soil, needs
(varies seasonally) sun
very rare
sandy soil, needs sun
rare to infrequent soil, tolerates
(varies seasonally) partial shade
very rare
very rare
mostly on shallow
rocky soil, needs
sun
mostly on shallow
rocky soil, needs
sun
soil tolerates par-
rare to infrequent tial shade
very rare
very rare
sandy soil, needs sun
mostly shallow rocky
soil, needs sun
XL>;-80
-------
APPENDIX II - AQUATIC PLANTS OF THE DITCHES AND CANALS.
Genus species
Azolla caroliniana
(red water fern)
Salvinia rotundi-
folia (water fern)
Typha angustifolia
(narrow leaf cat-
tail)
Typha latifolia
(broad leaf cat-
tail)
Potampgeton illi-
noensis (pondweed)
Najas quadalupensis
(naiad)
Sagittaria lancifolia
(arrowhead)
Sagittaria graminea
(arrowhead)
Zizaniopsis milia-
cea (giant cutgrass)
Phragmites austra-
lis (reed)
Panicum hemitomen
(maidencane)
Cyperus haspan
(sedge)
Eleocharis acicu-
laris
Cladium jamaicen-
sis (sawgrass)
Rhynchospora inu-
data (beakrush)
Lemna perpusilla
(duckweed)
Eichhornia crassipes
(water hyacinth)
Water depth
1 meter or more
1 meter or more
less than 1 meter
less than 1 meter
0.1-2.0 meters
0.5-3.0 meters
less than 0.5
meters
less than 1.0
meters
less than 0.5
meters
less than 0.5
meters
less than 0.5
meters
less than 0.2
meters
less than 0.5
meters
less than 1.0
meters
less than 1,0
meters
ovei 1 meter
over 1 meter
Water velocity
sluggish
sluggish
sluggish to mod-
erate ditches,
canals and margins
sluggish
sluggish to mod-
erate
sluggish moder-
ate or swift
sluggish canal.
ditches and mar-
gins
sluggish to mod-
erate
sluggish margins
sluggish margins
(also along banks)
sluggish canals,
ditches and mar-
gins (also along
banks)
sluggish margins
and on banks
sluggish margins
sluggish margins
or shallows
sluggish margins
or shallows
sluggish canals
and ditches
sluggish to mod-
erate canals and
ditches
Seasonal flux
cannot tolerate
exposure
cannot tolerate
exposure
tolerates season-
al exposure
tolerates brief
seasonal exposure
cannot tolerate
exposure
cannot tolerate
exposure
tolerates exposure,
requires only
moist soil
tolerates ex-
posure
tolerates exposure
tolerates continu-
ous exposure
tolerates continu-
ous exposure
tolerates continu-
ous exposure
tolerates brief
exposure
tolerates exposure
(exposure required
for seedling growth
and germination)
tolerates exposure
cannot tolerate ex-
posure
cannot tolerate ex-
posure
Bottom type
free floating
aquatic
free floating
emergent, deep-
ly rooted in soft
mud or organic
bottoms
emergent, deep-
ly rooted in soft
mud or organic
bottoms
Submergent
rooted in sandy
or sandy mud
bottoms
variable sandy.
rocky or sandy
muds
emergent rooted
in sandy soil,
sandy muds or
organic bottoms
emergent or sub-
mergent rooted in
sandy rocky, sandy
mud, or organic
soils
emergent rooted
in sandy organic,
to organic soil
emergent aqua-
tic and on banks
rooted in sandy,
sandy mud, and
sandy organic
soils
emergent and on
banks rooted in
sandy, rocky and
sandy organic
soils and muds
emergent or on
banks rooted in
sandy, rocky
soils and muds
sandy organic
muds emergent or
submergent
emergent rooted
in sandy organic
muds and organic
soils
emergent rooted
in sandy, sandy
organic or organ-
ic soils and muds
free floating
aquatic
free floating
aquatic (often
webbed across
the surface of
a canal)
Station
Misc.
Misc.
GAC,
JD, TC,
AA,
Misc.
Misc.
AA
GAC,
JD, TC,
AA,
Misc.
GAC,
JD, TC,
AA,
Misc.
GAC
JD, TC,
AA,
Misc.
JD,
Misc.
GAC,
JD, TC,
AA,
Misc.
JD, TC,
AA,
Misc.
GAC
GAC
(at the
weir)
JD, AA,
Misc.
GAC,
JD,
Misc.
Misc.
TC
(near
agri
cultu-
ral ac-
tivi-
ties)
XIX-81
-------
APPENDIX II - AQUATIC PLANTS OF THE DITCHES AND CANALS. CONT'D.
Genus species
Pontederia lanceo-
lata (pickeralweed)
Juncus polycephalus
(rush)
Canna flaccida
(wild canna)
Thalia geniculata
(fire flag)
Saururus cernuus
(lizzard's tail)
Polygon urn hydropi-
peroides (smartweed)
Nymphaea oderata
(white waterlily)
Nymphaea elegans
(Blue waterlily)
Ludwigia repens
Ludwigia peruviana
(water primrose)
Ludwigia leptocarpa
(water primrose)
Myriophyllum brasili-
ense (water milfoil)
Proserpinaca palustris
(mermaid weed)
Bacopa caroliniana
'tricularia foliosa
bladderwort)
Jtricularia purpurea
(purple flowered
bladderwort)
Dioda virginiana
(buttonweed)
Water depth
less than 0.5
meters
less than 0.5
meters
less than 1.0
meters
less than 1.0
meters
less than 0.5
meters
less than 1.0
meters
over 1 meter
over 1 meter
less than 1
meter
less than 0.5
meters
less than 0.5
meters
over 0.5 meters
less than 1.0
meters
less than 1.0
meters
over 1.0 meters
over 1.0 meters
less than 1.0
meters
Water velocity
sluggish canals,
ditches and
margins
sluggish margins
and ditches
sluggish margins
and ditches
sluggish margins
and ditches
sluggish ditches
sluggish ditches
and margins
sluggish to mod-
erate
sluggish to mod-
erate
sluggish to mod-
erate
sluggish ditches
and margins mostly
on banks
sluggish ditches
moderate to swift
sluggish margins
and shallow dit-
ches (also on
banks)
sluggish to mod-
erate margins and
shallow ditches
sluggish margins
and ditches
sluggish margins
and ditches
sluggish to mod-
erate ditches and
margins (also on
banks)
Seasonal flux
tolerates exposure
tolerates continu-
ous exposure
tolerates exposure
tolerates exposure
tolerates brief
seasonal exposure
tolerates seasonal
exposure
cannot tolerate
exposure
cannot tolerate
exposure
tolerates sea-
sonal exposure
tolerates exposure
tolerates exposure
cannot tolerate
long exposure
tolerates continu-
ous exposure
tolerates very
brief exposure
cannot tolerate
exposure
cannot tolerate
exposure
Must be exposed
seasonally for
flowering
Bottom type
emergent rooted
in sandy organic
and organic soils
and muds
emergent rooted
in sandy and
sandy organic
bottoms, margins,
and on banks
emergent in sandy
organic and or-
ganic soils
emergent rooted
in sandy organic
and organic
soils
emergent in sandy
organic or organ-
ic soils and muds
emergent in organ-
ic soils and muds
emergent with
floating leaves,
rooted in organ-
ic soil or mud
emergent with
floating leaves
rooted in or-
ganic soil or
mud
submergent or
emergent rooted
in sand, rocky
organic soils and
muds
sandy, sandy or-
ganic and organic
soils
sandy, sandy or-
ganic, and organic
soils
sandy, rocky, and
organic soils and
muds , submergent
emergent or sub-
mergent in sandy
rocky, and organ-
ic soils and muds
submergent rooted
in sandy, or or-
ganic soils and
muds
free floating
aquatic
free floating
Submergent or
emergent (seasonal)
and on dry banks,
sandy, sandy organ-
ic soils and muds
Station
JD, AA,
Misc.
GAC,
JD,
Misc.
JD,
Misc.
JD, AA,
Misc.
JD
JD,
Misc.
JD, AA,
JD
GAC,
JD,
Misc.
GAC,
JD, TC,
AA,
Misc.
JD,
Misc.
GAC
GAC
JD, AA,
Misc.
GAC
JD, AA,
Misc.
GAC,
JD, AA,
Misc.
GAC,
JD,
Misc.
JD,
Misc.
XIX-82
-------
APPENDIX JJ - ABUNDANCE OF MACROBENTHOS (INDIVIDUALS PER 0.046 M ).
FAHKAHATCHEE BAY (MARCH 28-30, 1972)
Grid Locations
Grid Locations
Grid Locations
Mud Substrate
ORGANISMS Q-6 C-5 D-7 F-8 G-7 1-6 K-5 K-7 M-6 N-4
Arthropoda
Crustacea
Malacostraca
Amphipoda 6 21 305 10 32 4 374 195 458 130
Isopoda 611 55 28 10 6 1 38 50 575
Cumacea 1 2
Thoracic a
Decapoda 11 1
Ostracoda 1 14
Insecta
Ephemeroptera
Baetidae
Diptera
Chironomidae 12
Tendipedidae 2
Trichoptera
Psychomyildae
Annelida
Polychaeta 20 8 13 2 5 8 10 10 13 13
Rhynchocoela 5 13 3 5 71
Mollusca
Gastropoda 1 1 43 2
Amphineura
Pele-cypoda 18 11 17 14 1 2 195
Echinodermata
Ophiuroidea 2
Porifera
Sand
E-l E-3 E-5 G-3 G-5 1-4 J-2 J-3 L-2 L-4
11 403 10 920 6 14 461 628 70 14
7 1 6 138 1 1 6 80 69 12
1 69
1 121415
43 22 13
4 2 1 21 4 10 40 26 1 4
2 15 4 10 8 10
1 211
52 4 2 13 1
1 1
Shell
G-0 G-l H-0 1-3 K-2
13 19 9 59 203
74159
1
10 2 13 1
1 5
1
2
42
2 1 8 2 64
10 7
1
1
2 1 1
1
1 2
M-3
9
4
1
5
29
2
22
FAHKA UNION BAY (MARCH 28-30, 1972)
ORGANISMS
Arthropoda
Crustacea
Malacostraca
Amphipoda
Isopoda
Cumacea
Mysidacea
Decapoda
Ostracoda
Copepoda
Insecta
Diptera
Annelida
Polychaeta
Hirudinea
Rhynchocoela
Mollusca
Gastropoda
Pelycepoda
Chordata
Ascldiacea
Grid Loca.tj.ons
Mud Substrate
22 23 1 1 7
282 644 148 195 684
2
5
2
1
23
1
36
5
7
35
3
Grid Locations
Sand
S-9A
6
114
2
1
S-9B P-9A
5 9
186 498
1
1
1
34 3
6
2 15
2
P-9B
4
102
1
13
5
7
1
S-10A
56
590
1
2
4
74
1
8
12
11
S-10B
49
19
44
6
1
Grid Locations
Shell
T-10A
4
145
2
3
8
36
4
1
1
T-10B T-10C U-12A U-12B
24 248 10 202
153 140 165
211
1.
1
27 34
1 1 8
1 2
U-12C
7
91
3
2
1
1
XIX-8 3
-------
APPENDIX jj - ABUNDANCE OF MACROBENTHOS (INDIVIDUALS PER 0.046 M ). CONT'D.
FAHKAHATCHEE BAY (JUNE 20-21, 1972)
Grid Locations
Grid Locations
Grid LocatIons
ORGANISMS
Arthropoda
Crustacea
Malacostraca
Amphipoda
Isopoda
Cumacea
Mysidacea
Thoracica
Decapoda
Qstracqda
Copepooa
Insecta
Diptera
Anne lida
Polychaeta
Rhynchocoe la
Mollusca
Gastropoda
Amphineura
Feltcypoda
Echinodermafca
Ophiuroidea
Porlfera
Chordata
Ascidiacea
Osteichthyes
Acanthocephala
Mud Substrate
Q-6 C-5 D-7 F-8 G-7 1-6 K-5 K-7 M-6 N-4
17 12 12 1 7 8 10 9 13 9
889 2 49 2 3 8 5 1 2 527
1 12
1
6 2121
1
6 16 8 3 5 12 17 10 15 16
2 12 2 5 11 1
1 4 2 6 6 10
10 31253 21
1
1 1
1
2
E-l
126
12
4
3
4
26
3
5
2
Sand
E-3 E-5 G-3 G-5 1-4 3-2 J-3 L-2 L-4
20 13 33 25 2 1 219 4 171
12 5 647 1 6 143 3 61
11 I
1
1 61211 3
3
1
20 27 27 8 7 1 70 2 7
3 4
1 2517
11 64
3
1
Shell
G-0 G-l H-0
58 62 22
454 445 23
4 1
123
15
9 23 11
1
11
1 1
135
1-3 K-2 M-3
20 5 18
80 9 76
1
3 1
24 2 13
1 3
23
FAHKA UNION BAY (JUNE 20-21, 1972)
_.Grid__Locat:ions_.
Grid Locations
ORGANISMS
Arthropoda
Crustacea
Malacostraca
Amphipoda
Isopoda
Cumacea
Mysidacea
Decapoda
Annelida
Polychaeta
Rhynchocoela
Mollusca
Gastropoda
Pe 1 ec-ypoda
Echinodermata
Ophiuroidea
Mud Substrate
Grid Locations
Sand
Shell
P-8 P-10 Q-ll Q-12 R-ll R-.I2 S-9A S-9B P-9A F-9B S-1QA S-1QB T-10A T-10B T-1QC U-12A U-12B U-12C
283
1 1
3 20
44 11
56 68
41
10 29
18
39 284 1229 961 1358 1172 483 803 818 148 120
10
1
1 1
10 11
1
34
17
18
12
14
1
34
5
5
15
14
3
38
85
45
3
13
19
57
2
11
206
8
21
10
6
44
2
12
932
1
33
7
236
10
2
7
786
XIX-84
-------
APPENDIX JJ - ABUNDANCE OF MACROBENTHOS (INDIVIDUALS PER 0.046 M2). CONT'D.
FAHKAHATCHEE BAY (SEPTEMBER 25-28, 1972)
ORGANISMS
Arthropoda
Crustacea
Malacostraca
Amphipoda
Isopoda
Cumacea
Mysidacea
Tanaidacea
Decapoda
Annelida
Polychaeta
Oligochaeta
Mollusca
Gastropoda
Amphineura
Pelecypoda
Echinodermata
Ophiuroidea
Ectoprocta
Sipunculida
Porifera
Grid Locations
Mud Substrate
Q-6 C-5 D-7 F-8 G-7 1-6 K-5 K-7 M-6 N-4
202 67 32 194 3 5 55 189
33 18 6
2
398 448 41
14 33 33
8 43 8 13 19 13 1 2 4
2 9 1
6 1 14 1
212 3
1 1
2
148
E-l
4
3
4
28
11
1
2
Grid Locations
Sand
E-3 E-5 G-3 G-5 1-4 J-2 J-3 L-2 L-4
29172 59 7 12 9
124 12
1 1
1
2 3
3754 4342
44 25 21 24 8 1 26 22 25
14 10 4 10
4 4 7 13 32 2 7
13 1
1 1
3 367
Grid Locations
Shell
G-0 G-l H-0 1-3 K-2 M-3
12 13 6 21
1 2 4 92 7
21484
24 614
1 58 1 10
2 2
3 1
1
1 1
FAHKA UNION BAY (SEPTEMBER 25-28, 1972)
...Grid Locations
^Locations
Grid Locations
ORGANISMS
Arthropoda
Crustacea
Malacostraca
Amphipoda
Isopoda
Cumacea
Tanaidacea
Decapoda
Insecta
Diptera
Ceratopogonidae
Chiron omidae
Collembola
Annelida
Polychaeta
Oligochaeta
Mollusca
Gastropoda
Pelecypoda
Sipunculida
Rhynchocoela
Mud Substrate
P-8 P-10 Q-ll Q-12 R-ll R-12 S-9A S-9B
152 118 303 117 44 166 317 202
1 25 8 47 11 70
118 69 229 67 28 130 29
21 15 2
1
413 1 16 16
3
7 1
7741 7
4 3
Sand
P-9A P-9B
74 147
18 2
1
10
1
1
4 1
1
3
9 1
1
2
Shell
S-10A S-10B
45 5
18 1
10
3
10 7
5
4
19 6
T-10A T-10B
1 83
12
1
17
1 1
1
6 14
25
1
1
6
T-10C
26
9
49
7
8
4
4
1
U-12A
13
7
53
1
1
6
1
7
4
U-12B U-12C
218 72
9 22
273 39
1
8 4
2 1
XIX-85
-------
APPENDIX KK - LIST OF SCIENTIFIC AND COMMON NAMES OF FISHES FROM FAHKAHATCHEE
STRAND AND TEN THOUSAND ISLANDS, FLORIDA, 1972.
Family
Species
Common name
Carcharhinidae
Sphyrnidae
Dasyatidae
Myliobatidae
Lepisosteidae
Amiidae
Elopidae
Ophichthidae
Clupeidae
Engraulidae
Synodontidae
Cyprinidae
Catostomidae
Ictaluridae
Ariidae
Batrachoididae
Gobiesocidae
Carcharhinus leucas
Carcharhinus limbatus
Negaprion brevirostris
Sphyrna tiburo
Dasyatis americana
Dasyatis" sabina
Gymnura micrura
Aetobatus narinari
Lepisosteus platyrhincus
Amia calva
Elops saurus
Megalops atlantica
Myrophis punctatus
Brevoortia patronus
Brevoortia smithi
Harengula pensacolae
Opisthonema oglinum
Anchoa cubana
Anchoa hepsetus
Anchoa mitchilli
Synodus foetens
Notemigonus crysoleucas
Notropis maculatus
Erimyzon sucetta
Ictalurus natalis
Ictalurus nebulosus
Arius felis
Bagre marinus
Opsanus beta
Gobiesox strumosus
Bull shark
Blacktip shark
Lemon shark
Bonnethead
Southern stingray
Atlantic stingray
Smooth butterfly ray
Spotted eagle ray
Florida gar
Bowfin
Ladyfish
Tarpon
Speckled worm eel
Gulf menhaden
Yellowfin menhaden
Scaled sardine
Atlantic thread herring
Cuban anchovy
Striped anchovy
Bay anchovy
Inshore lizardfish
Golden shiner
Taillight shiner
Lake chubsucker
Yellow bullhead
Brown bullhead
Sea catfish
Gafftopsail catfish
Gulf toadfish
i
Skilletfish
XIX-86
-------
APPENDIX KK - LIST OF SCIENTIFIC AND COMMON NAMES OF FISHES FROM FAHKAHATCHEE
STRAND AND TEN THOUSAND ISLANDS, FLORIDA, 1972. CONT'D.
Family
Species
Common name
Ogcocephalidae
Exocoetidae
Belonidae
OgcQcephalus radiatus
Hyporhamphus unifasciatus
Strongylura marina
Strongylura notata
Strongylura timucu
Tylosurus crocodilus
Cyprinodontidae Adinia xenica
Poeciliidae
Atherinidae
Syngnathidae
Centropomidae
Serranidae
Centrarchidae
Cyprinodon variegatus
Floridichthys carpio
Fundulus chrysotus
Fundulus confluentus
Fundulus grandis
Fundulus seminoljs
Fundulus similis
Jordanella floridae
Lucania goodei
Lucania parva
Gambusia affinis
Heterandria formosa
Poecilia latipinna
Labidesthes sicculus
Membras martinica
Menidia beryllina
Hippocampus erectus
Hippocampus reidi
Hippocampus zosterae
Syngnathus floridae
Syngnathus louisianae
Syngnathus scovelli
Syngnathus springeri
Centropomus parallelus
Centropomus pectinatus
Centropomus undecimalis
Diplectrum formosum
Epinephelus itajara
Mycteroperca microlepis
Elassoma evergladei
Enneacanthus gloriosus
Polka-dot batfish
Halfbeak
Atlantic needlefish
Redfin needlefish
Timucu
Houndfish
Diamond killifish
Sheepshead minnow
Goldspotted killifish
Golden topminnow
Marsh killifish
Gulf killifish
Seminole killifish
Longnose killifish
Flagfish
Bluefin killifish
Rainwater killifish
Mosquitofish
Least killifish
Sailfin molly
Brook silverside
Rough silverside
Tidewater silverside
Lined seahorse
Longsnout seahorse
Dwarf seahorse
Dusky pipefish
Chain pipefish
Gulf pipefish
Bull pipefish
Fat snook
Tarpon snook
Snook
Sand perch
Jewfish
Gag
Everglades pygmy sunfish
Bluespotted sunfish
XIX-87
-------
APPENDIX KK - LIST OF SCIENTIFIC AND COMMON NAMES OF FISHES FROM FAHKAHATCHEE
STRAND AND TEN THOUSAND ISLANDS, FLORIDA, 1972. CONT'D.
Family
-Species
Common name
Centrarchidae
Contd.
Percidae
Rachycentridae
Echeneidae
Carangidae
Lutjanidae
Gerreidae
Pomadasyidae
Sparidae
Sciaenidae
Ephippidae
Scaridae
Lepomis gulpsus
Lepomis macrochirus
Lepomis majrginatus
Lepomis microlophus
Lepomis punctatus
Micropterus s a lino ides
Etheostoma fusiforme
Rachyc entron canadum
Echeneis neucratoides
Caranx crysos
Caranx, hippos
Chiproscombrus chrysurus
Oligoplites saurus
Selene vomer
Trachinotus carolinus
Trachinotus falcatus
Lutjanus griseus
Lutjanus synagris
Diapterus olisthostomus
Diapterus plumierT
Kur.i nngitnimic; argeTlteuS
Eucinos tomus gula
Orthopristis chrysoptera
Archosargus probatocephalus
Lagodpn rhomboides
Bairdiella chrysura
Cynoscion arenarius
Cynoscion nebulpsus
Le ids t omus xanthurus
Menticirrhus americanus
Menticirrhus saxatilis
Pogonias cromis
Sciaenop's ocelTata
Chaetodipterus faber
Nicholsina usta
Warmouth
Bluegill
Dollar sunfish
Redear sunfish
Spotted sunfish
Largemouth bass
Swamp darter
Gobi a
Whitefin sharksucker
Blue runner
Crevalle jack
Atlantic bumper
Leatherjacket
Lookdown
Florida pompano
Permit
Gray snapper
Lane snapper
Irish pompano
Striped mojarra
Spotfin mojarra
Silver jenny
Pigfish
Sheepshead
Pinfish
Silver perch
Sand seatrout
Spotted seatrout
Spot
Southern kingfish
Northern kingfish
Black drum
Red drum
Atlantic spadefish
Emerald parrotfish
XIX-88
-------
APPENDIX KK - LIST OF SCIENTIFIC AND COMMON NAMES OF FISHES FROM FAHKAHATCHEE
STRAND AND TEN THOUSAND ISLANDS, FLORIDA, 1972. CONT1 D.
Family
Species
Common name
Mugilidae
Sphyraenidae
Blenniidae
Gobiidae
Trichiuridae
Scombridae
Triglidae
Bothidae
Soleidae
Cynoglossidae
Balistidae
Ostraciidae
Tetraodontidae
Diodontidae
Mugil cephalus
Mugil curema
Mugil trichodon
Sphyraena barracuda
Blennius nicholsi
Chasmodes saburrae
Bathygobius soporator
Gobione11us shufeldti
Gobiosoma bosci
Gobiosoma robustum
Lophpgobius cyprinoides
Microgobius gulosus
Microgobius thalassinus
Trichiurus lepturus
Sarda sarda
Scomberomorus maculatus
Prionptus scitulus
Prionotus tribulus
Etropus crossotus
Paralichthys al_bigutt_a
P.aralichthys lethostigma
Achirus lineatus
Trinectes macuTatus
Symphurus plag!usa
Monacanthus hispidus
Lactophrys quadricornis
Sphoeroides nephelus
Sphoeroides parvus
Chilomycterus antillarum
Chilomycterus schoepfT
Striped mullet
White mullet
Fantail mullet
Great barracuda
Highfin blenny
Florida blenny
Frillfin goby
Freshwater goby
Naked goby
Code goby
Crested goby
Clown goby
Green goby
Atlantic cutlassfish
Atlantic bonito
Spanish mackerel
Leopard searobin
Bighead searobin
Fringed flounder
Gulf flounder
Southern flounder
Lined sole
Hogchoker
Blackcheek tonguefish
Planehead filefish
Scrawled cowfish
Southern puffer
Least puffer
Web burrfish
Striped burrfish
XIX-89
-------
APPENDIX LL - SUMMARY OF SELECTED AQUATIC BIOTA ASSAYED FOR ASH-FREE DRY WEIGHT
AND ORGANIC NITROGEN, FAHKAHATCHEE STRAND AND ADJACENT CANALS, FLORIDA,
1972.
Species
Size Wet Percent
Date range weight dry
collected (mm) (gm) weight
Ash-free, Organic N,
percent of percent of
dry weight dry weight
Fishes
Lepisosteus p1atyrhincus
Notemigonus crysoleucas
Erimyzon sucetta
Ictalurus natalis
Ictalurus nebulosus
Fundulus confluentus
Fundulus seminolis
Jordanella floridae
Lucania gpodei
Gambusia affinis
Heterandria formosa
Poecilia latipinna
Enneacanthus gloriosus
Lepomis gulosus
Lepomis macrochirus
Lepomis microlepis
Lepomis punctatus
Amphibians
Diemictylus viridescens
Rana spp.
Crustacean
Procambarus alleni
Feb.
Feb.
Feb.
Mar.
Apr.
Mar.
Feb.
Jan.
Feb.
Jan.
Jan.
Mar.
Feb.
Feb.
Feb.
Feb.
Feb.
2
28
28
7
4
7
28
12
12
12
12
7
28
28
28
28
28
384
62-104
34-68
70
34-61
123
17-33
19-53
15-45
13-25
23-51
46-52
171
204
45-120
145-191
13.7
26.7
589.0
14.2
14.5
50.2
152.5
77.1
5.0
37.6
19.0
112.3
166.2
210.8
73.4
19.7
18.0
22.4
23.3
23.2
24.8
19.9
18.6
19.7
22.0
24.5
17.9
26.2
26.4
21.1
17.1
Feb. 8
Jan. 12
Mar. 7
37-74
21-50
5.8
8.4
18-22a 197.4
12.1
11.9
20.4
68.5
84.0
82.6
93.0
86.
85.
84.
82.9
82.
84.
96.9
86.9
77.
75.
77.
76,0
79.0
,1
,3
,2
,1
.3
68.1
7.6
9.6
9.4
8.5
11.4
8.4
9.5
9.2
8.8
8.6
9.4
7.8
9.1
8.9
9.7
9.9
10.0
8.0
8.6
8.0
Plants
Saggittaria graminea
Cyperus spp.
Pontederia lanceolata
Polygonum hydropiperoides
Ludwigia rep ens
Proserpinaca palustris
Bacopa caroliniana
Utricularia foliosa
Diodia virginiana
Mikania batatifolia
July 27
July 27
Jan. 12
Jan. 12
Jan. 13
Jan. 12
Feb. 8
Jan. 12
Feb. 8
Jan. 12
250.3
15.0
527.5
813.0
660.0
57.6
90.0
224.5
202,0
234.9
8.5
11.3
7.2
25.1
5.7
22.2
10.1
5.3
10.6
32.9
80.1
86.8
78.6
93.8
77.6
72.4
74.6
58.5
79.8
85.3
1.0
1.3
1.4
0.6
1.1
0.9
1.4
1.2
1.8
0.9
a Measured from anterior orbit to posterior edge of carapace.
XIX-90
-------
APPENDIX MM - SPATIAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972, BY HABITAT: (1) BEACHES AND
ADJACENT WATERS IN THE LOWER REACHES; (2) BAYS IN THE MIDDLE REACHES;
(3) TIDAL STREAMS AND (4) TIDAL CANALS IN THE UPPER REACHES; (5) FRESH-
WATER CANALS; AND (6) FRESHWATER LAKES, PONDS, AND TEMPORARY WATERWAYS IN
FAHKAHATCHEE STRAND. SPECIES TAKEN IN THE MONTHLY COLLECTIONS AT THE
REGULAR STATIONS ARE INDICATED BY THE LETTER X, WHEREAS PARENTHESIS
DESIGNATE FISHES THAT WERE ONLY COLLECTED DURING OTHER INVESTIGATIONS.
HABITAT
SPECIES
Carcharhinus leucas
Carcharhinus limbatus
Negaprion brevirostris
Sphyrna tiburo
Dasyatis americana
Dasyatis sabina
Gymnura micrura
Aetobatus narinari
Lepisosteus platyrhincus
Amia calva
Elops saurus
Mega lops atlantica
Myrophis punctatus
Brevoortia patronus
Brevoortia smith!
Harengula pensacolae
Opisthonema oglinum
Anchoa cubana
Anchoa hepsetus
Anchoa mitchilli
Synodus foetens
Notemlgonus cryso leucas
Notropis maculatus
Erimyzon sucetta
Ictalurus natalis
Ictalurus nebulosus
Arius fells
Bagre marinus
Ops anus beta
Goblesox strumosus
Ogcocephalus radiatus
Hyporhamphus unifasclatus
Strongylura marina
Strongylura notata
Strongylura timucu
Tylosurus crocodilus
Adinia xenica
1
(X)
X
(X)
X
X
X
(X)
X
X
X
X
X
X
X
X
X
X
X
X
X
2
X
(X)
X
X
X
(X)
X
X
X
X
X
X
X
X
X
X
X
(X)
(X)
X
X
X
X
3
X
X
(X)
X
X
X
X
X
(X)
X
X
X
(X)
4
(X)
(X)
X
X
(X)
X
X
X
X
(X)
(X)
(X)
(X)
X
X
X
X
5 6
(X) X
(X) (X)
(X) (X)
(X)
(X)
(X) X
(X)
XIX-91
-------
APPENDIX MM - SPATIAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972, BY HABITAT: (1) BEACHES AND
ADJACENT WATERS IN THE LOWER REACHES; (2) BAYS IN THE MIDDLE REACHES;
(3) TIDAL STREAMS AND (4) TIDAL CANALS IN THE UPPER REACHES; (5) FRESH-
WATER CANALS; AND (6) FRESHWATER LAKES, PONDS, AND TEMPORARY WATERWAYS IN
FAHKAHATCHEE STRAND. SPECIES TAKEN IN THE MONTHLY COLLECTIONS AT THE
REGULAR STATIONS ARE INDICATED BY THE LETTER X, WHEREAS PARENTHESIS
DESIGNATE FISHES THAT WERE ONLY COLLECTED DURING OTHER INVESTIGATIONS. CONT'D,
SPECIES
Cyprinodon variegatus
Floridichthys carpio
Fundulus chrysotus
Fundulus confluentus
Fundulus grandis
Fundulus seminolis
Fundulus similis
Jordanella floridae
Lucania goodei
Lucania parva
Gambusia af finis
Heterandria formosa
Poecilia latipinna
Labidesthes sicculus
Membras martinica
Menidia beryllina
Hippocampus erectus
Hippocampus reidi
Hippocampus zosterae
Syngnathus floridae
Syngnathus louisianae
Syngnathus s cove Hi
Syngnathus springeri
Centropomus para lie lus
Centropomus pectinatus
Centropomus undecimalis
Diplectrum formosum
Epinephelus ita.lara
Mycteroperca micro lepis
Elassoma evergladei
Enneacarthus gloriosus
Lepomis gulosus
Lepomis machrochirus
Lepomis marginatus
Lepomis microlophus
Lepomis punctatus
Micropterus salmoides
Etheostoma fusiforme
Rachycentron canadum
Echeneis neucratoides
Caranx crysos
Caranx higpos
Chlorpscombrus chrysurus
Oligoplites saurus
Selene vomer
1
X
X
X
X
(X)
(X)
X
X
(X)
X
X
(X)
X
X
X
XIX- 92
2
X
X
X
X
X
X
X
X
X
X
X
X
X
(X)
X
X
(X)
(X)
(X)
X
X
X
HABITAT
3 4
X (X)
X X
(X)
X X
(X) X
X
X (X)
X X
X X
(X)
X X
X X
X X
(X)
X X
(X)
(X) (X)
(X) (X)
(X)
X
(X)
(X)
X
X X
5
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
(X)
6
X
X
X
X
X
X
X
X
X
X
(X)
X
-------
APPENDIX MM - SPATIAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972, BY HABITAT: (1) BEACHES AND
ADJACENT WATERS IN THE LOWER REACHES; (2) BAYS IN THE MIDDLE REACHES;
(3) TIDAL STREAMS AND (4) TIDAL CANALS IN THE UPPER REACHES; (5) FRESH-
WATER CANALS; AND (6) FRESHWATER LAKES, PONDS, AND TEMPORARY WATERWAYS IN
FAHKAHATCHEE STRAND. SPECIES TAKEN IN THE MONTHLY COLLECTIONS AT THE
REGULAR STATIONS ARE INDICATED BY THE LETTER X, WHEREAS PARENTHESIS
DESIGNATE FISHES THAT WERE ONLY COLLECTED DURING OTHER INVESTIGATIONS. CONT'D,
HABITAT
SPECIES
Trachinotus carolinus
Trachinotus falcatus
Lut i anus griseus
Lut j anus synagris
Diapterus olisthostomus
Diapterus plumieri
Eucinostomus argenteus
Eucinostomus gula
Orthopristis chrysoptera
Archosargus probatocephalus
Lagodon thomboides
Bairdielia chrysura
Cynoscion arenarius
Cynoscion nebulosus
Leiostomus xanthurus
Menticirrhus americanus
Menticirrhus littoralis
Menticirrhus saxatilis
Pogonias cromis
Sciaenops ocellata
Chaetodipterus faber
Nicholsina usta
Mugil cephalus
Mugil curema
Mugil trichodon
Sphyraena barracuda
Blennius nicholsi
Chasmodes saburrae
Bathygobius soporator
Gobionellus shufeldti
Gobiosoma bosci
Gobiosoma robustum
Lophogobius cyprinoides
Microgobius gulps us
Microgobius t ha las sinus
Trichiurus lepturus
Sarda sarda
Scomberomorus mac u la t us
Prionotus scitulus
Prionotus tribulus
Etropus crossotus
Paralichthys albigutta
Paralichthys lethostigma
1
(X)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
(X)
(X)
X
X
X
X
2
X
X
X
X
X
X
X
X
X
X
X
X
(X)
X
(X)
(X)
X
X
X
X
X
(X)
(X)
X
(X)
X
(X)
X
X
X
X
X
X
X
X
X
3
(X)
X
X
X
(X)
(X)
X
X
(X)
(X)
X
X
(X)
X
X
(X)
X
X
456
X
X
X
X
X
X
X
(X)
X
X
X
X
X
X
(X)
X
XIX-93
-------
APPENDIX MM - SPATIAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972, BY HABITAT: (1) BEACHES AND
ADJACENT WATERS IN THE LOWER REACHES; (2) BAYS IN THE MIDDLE REACHES;
(3) TIDAL STREAMS AND (4) TIDAL CANALS IN THE UPPER REACHES; (5) FRESH-
WATER CANALS; AND (6) FRESHWATER LAKES, PONDS, AND TEMPORARY WATERWAYS IN
FAHKAHATCHEE STRAND. SPECIES TAKEN IN THE MONTHLY COLLECTIONS AT THE
REGULAR STATIONS ARE INDICATED BY THE LETTER X, WHEREAS PARENTHESIS
DESIGNATE FISHES THAT WERE ONLY COLLECTED DURING OTHER INVESTIGATIONS. CONT'D.
SPECIES
HABITAT
2
4
Achirus lineatus
Trinectes maculatus
Symphurus plagiusa
Monacanthus hispidus
Lactophrys quandricornis
Sphoeroides nephelus
Sphoeroides parvus
Chilomycterus antillarus
Chllomycterus schoepfi
X
X
X
(X)
X
X
X
X
X
X
(X)
X
X
X
X
X
(X)
X
XIX-94
-------
APPENDIX NN - TEMPORAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972. SPECIES TAKEN AT REGULAR ESTUARINE
AND FRESHWATER STATIONS EACH MONTH ARE INDICATED BY THE LETTER X. PARENTHESIS
DESIGNATE SPECIES THAT WERE COLLECTED DURING OTHER INVESTIGATIONS.
Month
Species
Carcharhinus leucas
Carcharhinus limbatus
Negaprion brevirostris
Sphyrna tiburo
Dasyatis americana
Dasyatis sabina
Gymnura micrura
Aetobatus narinari
Lepisosteus platyrhincus
Amia calva
Elops saurus
Megalops atlantica
Myrophis punctatus
Brevoortia patronus
Brevoortia smith!
Harengula pensacolae
Qpisthonema oglinum
Anchoa cub ana
Anchoa hepsetus
Anchoa mitchilli
Synodus foetens
Notemigonus crysoleucas
Notropis maculatus
Erimyzon sucetta
Ictalurus natal is
Ictalurus nebulosus
Arius felis
Bagre marinus
Opsanus beta
Gobiesox strumosus
Ogcocephalus radiatus
Hyporhamphus unifasciatus
Strongylura marina
Strong^lura notata
Strongylura timucu
Tylosurus crocodilus
Adinia xenica
Cyprinodon variegatus
Floridichthys carpio
Fundulus chrysotus
Fundulus confluentus
Fundulus grandis
Fundulus seminolis
Fundulus similis
Jordanella floridae
Lucania goodei
Lucania parva
Gambusia affinis
Heterandria formosa
Poecilia latipinna
Labidesthes sicculus
Membras martini ca
Menidia beryllina
Hippocampus erectus
Hippocampus reidi
Hippocampus zosterae
Syngnathus floridae
Syngnathus louisianae
Syngnathus scovelli
Syngnathus springori
Centropomus parallelus
Centropomus pectinatus
Jan.
X
(x)
X
(x)
X
X
X
X
ex)
X
X
X
X
X
X
X
X
Cx)
X
X
X
X
(x)
X
X
X
X
Feb.
X
(x)
X
X
X
X
X
X
(x)
(x)
ex)
(x)
X
X
X
X
X
X
(x)
X
(x)
X
X
Cx)
X
X
X
X
(x)
X
X
X
X
Mar.
(x)
(x)
X
X
X
X
X
X
X
ex)
(x)
Cx)
X
X
X
X
X
(x)
X
(x)
X
X
X
X
X
X
X
X
X
X
X
X
X
Apr.
(x)
(X)
X
(x)
Cx)
(x)
(x)
X
X
X
X
Cx)
X
Cx)
X
X
X
x
X
X
Cx)
X
X
X
X
X
X
X
X
Cx)
Cx)
X
X
X
May
(x)
X
X
Cx)
X
X
X
X
X
X
X
X
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
June
X
X
X
X
X
Cx)
X
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
(x)
X
X
X
X
X
Cx)
X
X
X
X
X
X
X
Cx)
(x)
July
X
(x)
X
Cx)
(x)
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
X
X
X
(x)
Aug.
X
Cx)
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
X
X
Cx)
X
X
X
X
X
X
Sept.
X
Cx)
(x)
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
Oct.
Cxj
Cx)
(x)
X
X
X
X
X
X
X
X
X
(x)
Cx)
X
Cx)
X
X
X
X
X
Cx)
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
Nov.
r 1&
Cx)
X
X
X
X
X
X
X
X
X
X
(x)
Cx)a
X
X
X
X
X
(x)
X
X
X
X
X
X
X
X
X
X
X
X
Dec.
X
X
X
X
X
X
X
X
X
X
(x)a
X
X
X
Cx)
(x)
Cx) -
X
Cx)
Cx)
X
X
Cx)
X
X
X
X
X
XIX- 95
-------
APPENDIX NN - TEMPORAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972. SPECIES TAKEN AT REGULAR ESTUARINE
AND FRESHWATER STATIONS EACH MONTH ARE INDICATED BY THE LETTER X. PARENTHESIS
DESIGNATE SPECIES THAT WERE COLLECTED DURING OTHER INVESTIGATIONS. CONT'D,
Month
Species
Centropomus undecimalis
Diplectrum formosum
Epinephelus itajara
Mycteroperca microlepis
Elassoma everglade!
Enneacanthus gloriosus
Lepomis gulosus
Lepomis maciochirus
Lepomis marginatus
Lepomis microlophus
Lepomis punctatus
Micropterus salmoides
Etheostoma fusiforme
Rachycentron canadum
Echeneis neucratoides
Caranx crysos
Caranx hippos
Chloroscombrus chrysurus
Oligoplites saurus
Selene vomer
Trachinotus carolinus
Trachinotus falcatus
Lut janus griseus
Lut janus synagris
Diapterus olisthostomus
Diapterus plumieri
Eucinostomus argenteus
Eucinostomus gula
Orthopristis chrysoptera
Archosargus probatocephalus
Lagodon rhonboides
Bairdiella chrysura
Cynoscion arenarius
Cynoscion nebulosus
Leiostomus xanthurus
Menticirrhus americanus
Menticirrhus littoralis
Menticirrhus saxatilis
Pogonias cromis
Sciaenops ocellata
Chaetodipterus faber
Nicholsina usta
Mugil cephalus
Mugil curema
Mugil trichodon
Sphyraena barracuda
Blennius nicholsi
Chasmodes saburrae
Bathygobius soporator
Gobionellus shufeldti
Gobiosoma bosci
Gobiosoma robustum
Lophogobius cyprinoides
Microgobius gulosus
Microgobius thalassinus
Trichiurus lepturus
Sarda sarda
Scomberoraorus maculatus
Prionotus scitulus
Prionotus tribulus
Etropus crossotus
Paralichthys albigutta
Paralichthys lethostigma
Jan.
X
(x)
X
(x)
(x)
X
(x)
(x)
(x)
X
X
X
X
X
X
X
X
X
(x)
X
X
(x)
X
X
X
Feb.
X
X
Cx)
(x)
(x)
(x)
(x)
X
(x)
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx}
Mar.
X
Cx)
Cx)
Cx)
Cx)
(x)
X
(x)
Cx)
(x)
Cx)
X
X
X
X
X
X
X
(x)
X
Cx)
X
X
X
X
Cx)
X
Apr.
Cx)
(x)
X
Cx)
Cx)
Cx)
Cx)
Cx)
Cx)
Cx)
X
X
X
X
X
X
X
(x)
X
X
Cx)
X
X
X
X
May
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
X
X
June
X
X
X
(x)
X
(x)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
X
July
X
Cx)
Cx)
(x)
Cx)
X
X
X
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
Aug.
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
(x)
X
Cx)
X
X
X
X
X
Cx)
X
X
X
X
Sept,
Cx)
(x)
X
X
Cx)
X
Cx)
X
X
X
X
X
X
Cx)
X
X
X
X
X
Cx)
X
X
Cx)
Cx)
X
(x)
X
Cx)
X
X
X
X
Cx)
Cx)
Oct.
X
Cx)
X
Cx)
Cx)
Cx)
X
(x)
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
(x)
X
X
X
X
X
Cx)
(X)
Cx)
X
Cx)
X
Nov.
(x)
X
X
Cx)a
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx)
X
Cx)
X
Cx)
X
X
X
Cx)a
Dec.
Cx)
X
Cx)
X
Cx)
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Cx)
X
X
XIX-9 6
-------
APPENDIX NN - TEMPORAL DISTRIBUTION OF FISHES COLLECTED IN TEN THOUSAND ISLANDS
AND FAHKAHATCHEE STRAND, FLORIDA, 1972. SPECIES TAKEN AT REGULAR ESTUARINE
AND FRESHWATER STATIONS EACH MONTH ARE INDICATED BY THE LETTER X. PARENTHESIS
DESIGNATE SPECIES THAT WERE COLLECTED DURING OTHER INVESTIGATIONS. CONT'D.
Species
Achirus lineatus
Trinectes maculatus
Symphurus plagiusa
Monacanthus hispidus
Lactophrys quadricornis
Sphoeroides nephelus
Sphoeroides parvus
Chilomycterus antillarus
Chilomycterus schoepfi
Jan.
x
x
x
x
x
(x)
Month
Feb. Mar. Apr. May June July Aug.
x x
(x) x x x x
X
XX X
X X X X X
(x) x
X X X X X
Sept.
(x)
X
(x)
(x)
X
Oct.
X
X
X
X
x
Nov.
X
X
X
X
x
x
Dec.
(x)
(x)
X
X
X
Collected during preliminary surveys in 1971.
XIX-97
-------
APPENDIX 00 - WASTE SOURCE INVENTORIES.
Broward County
1. Broward County Domestic Wastewater Source Inventory, Broward
County Health Department, Fort Lauderdale, Florida, July, 1971.
2. Interim Plan for Water Quality Management, Vicinity of Broward
County, Florida, Fort Lauderdale, Florida, May, 1972.
3. Broward County Wastewater Source Inventory, Broward County
Health Department, Fort Lauderdale, Florida, October, 1972.
Dade County
1. Report of Waste Source Inventory and Evaluation, Dade County,
Florida. Environmental Protection Agency, Southeast Region,
Southeast Environmental Research Laboratory, Technical Programs,
Athens, Georgia, June, 1971.
Monroe County
1. Environment and Identity - A Plan for Development in the
Florida Keys. Milo Smith Associates, Inc., Planning Consultants,
Tampa, Florida, June, 1970.
Palm Beach County
1. Existing Facilities Analysis, Chapter 3 of Interim Water Quality
Management Plan, Palm Beach County. William M. Bishop Engineering,
Inc., February, 1973.
Florida Wide
1. Municipal Waste Facilities - Region IV, U. S. Department of
Health, Education and Welfare, Public Health Service Division
of Water Supply and Pollution Control, Basic Data Branch,
Washington, D. C., 1962.
2. Inventory of Public Sewerage Systems in Florida. Florida
State Board of Health, Bureau of Sanitary Engineering, Jacksonville,
Florida, 1966.
XIX-98
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - CHARLOTTE COUNTY.
Population
Name § location served
1. Admiralty Villas
N. Beach Rd. , Englewood 109
2. Bay Vista Restaurant
McCall Rd., Englewood 261
3. B's Restaurant
Edgewater Dr. , Punta Gorda
4. Charlotte County Country Club
385 miles east of Punta Gorda 200
5. Charlotte County Development
Authority, County Airport In-
dustrial Park 150
6. East Elementary School,
Punta Gorda 750
7. El Galeon, Englewood
8. Englewood Beach Condominum
Englewood Beach 500
9. Fran S Frank's Sail-inn
Charlotte Harbor
10. Gasparilla Mobile Estates
Placida 150
11. General Development Utilities,
Inc., Toledo Blade Blvd. § Pellam
Waterway, Port Charlotte 40
12. Gulf to Bay Trailer Park
Manasota Key 30
13. Harbour Heights Subdivision
Highlands Rd. , Punta Gorda 332
14. Harbour Inn Motel
Tamiami Trail, Charlotte Harbor SO
15. Harbor View Trailer Park,
North of Ft. Myers
16. Holiday Inn Travel Park
Englewood
17. Hot Springs Development
Charlotte Harbor
18. Lazy Lagoon Trailer Park
Punta Gorda 300
19. Lemon Bay Elementary
School, Englewood
20. Marylou Trailer Park
Punta Gorda 342
Flow (GPD)
Avg. Design
15,000
5,000
1,000
5,000
7,500
9,000
1,000 4,000
50,000,
26,000
15,000
2,000 30,000
3,000
5,300 42,000
2,000 8,300
25,000
15,000
40,000
15,000
9,000
18,000
Percent
removal
9/72
BOD 91
SS 61
8/72
BOD 92
SS 94
10/72
BOD 88
SS 41
11/72
BOD 95.5
7/72
BOD 99
SS 92
11/72
BOD 96
SS 68
Engineer's
estimate
BOD 94
10/72
BOD 90
SS 96
11/72
BOD 62
SS 70
9/72
BOD 98
SS 90
11/72
BOD 99
SS 80
7/72
BOD 92
SS 57
7/72
BOD 91
SS 98
11/72
BOD 99
SS 36
Type of
treatment
a.) Extended aeration
b.) Chlorination
a.) "MAROLF" package
b.) Extended aeration
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
a. ) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
a.) Extended aeration
b.) Chlorination
c.) Sand filter
a.) Contact Stabili-
zation
b.) Gas Chlorination
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
a.) "DEFIANCE" package
b.) Extended aeration
c.) Polishing pond
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
a.) "IMHOFF" tank
b.) Chlorination
c.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Sand filters
a.) Extended aeration
b.) Chlorination
a. ) Extended aeration
b.) Chlorination
a.) Extended aeration
a.) Extended aeration
b.) Chlorination
a.) "MAROLF" package
b.) Extended aeration
a.) "DAVCO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization
pond
Receiving
stream
Irrigation
Tile drain-
field
Peace River
Evaporation
percolation
pond
Evaporation
percolation
pond
Drain field
Evaporation
percolation
pond
Evaporation
percolation
pond
Charlotte
Harbor
Salt water
canal
Pellam
Waterway to
Charlotte
Harbor
Lemon Bay
DeSota Canal
to Peace
River
Peace River
Evaporation
percolation
pond
Evaporation
percolation
pond
_, , t
Harbor
Shell Creek
Drainfield
Peace River
Receiving
basin
9
9
2
9
9
9
9
9
2
8
2
2
2
2
9
9
2
2
9
2
XIX-99
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - CHARLOTTE COUNTY . CONT'D.
Population
Name S location served
21. Fred McCullough Motel
Manasota Key 24
22. B.C. Nuzum Apt. Complex
Punt a Gorda
23. Oakwater Cove Condominium
Manasota Key
24. Palm Marina Travel Trailer
Park, Cape Haze
25. Palm § Pines Trailer Park
§ Cottages, Punta Gorda 300
26. Palm Plaza
Punta Gorda 100
27. Palmetto Mobile Park
Charlotte Harbor 300
28. Palmetto Park Trailer
Palmetto Park 300
29. Park Hill Manor #1
Mobile Home Park, Burnt Store
Rd. , Punta Gorda 300
30. Park Hill Manor #2
Mobile Home Park, Burnt Store
Rd. , Punta Gorda
31. Park Point Mobile Home Park
Englewood
32. Pelican Harbor Mobile
Home Estates, Punta Gorda 372
33. Pine Terrace Trailer Park
Punta Gorda
34. Polynesian Palms
35. Port Charlotte Murdock
Plant, Port Charlotte
4,310
36. Port Charlotte Subdivision connec-
Port Charlotte tions.
37. Punta Gorda, Punta Gorda 100
38. City of Punta Gorda 10,000
39. City of Punta Gorda
Charlotte County Airport
40. Punta Gorda Camp Grounds
Punta Gorda 198
Flow (GPD)
Avg. Design
2,000
6,500
6,000
5,000
15,000
7,025 10,000
13,300
16,000
15,000
20,000 30,000
8,000
20,000
15,000
40,000
30,000
500,000 800,000
10,000
0.75 MGD 1 MGD
260,000
5,000
Percent
removal
10/72
BOD 93
SS 97
9/72
BOD 95
SS 93
12/72
BOD 94
SS 96
11/72
BOD 94
SS 98
5/72
BOD 98
SS 92
10/72
BOD 97
SS 95
10/72
BOD 95
SS 77
12/72
BOD 99
SS 98
12/72
BOD 92
SS 95
12/72
BOD 93
SS 90
9/72
BOD 98
SS 91
12/72
BOD 99
SS 98
Type of Receiving R'
treatment stream
a.) "MAROLF" package
b . ) Extended aeration
c.) Chlorination
d.) Stabilization pond Lemon Bay
a.) "AERO-JET" package
b.) Extended aeration
c.) Chlorination Drainfield
a.) Extended aeration
b.) Chlorination
c.) Sand filters Drainfield
a.) "DEFIANCE" package Evaporation
b.) Extended aeration percolation
c.) Chlorination pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination Peace River
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization Charlotte
pond Harbor
a.) Extended aeration
b.) Chlorination
c.) Stabilization Charlotte
pond Harbor
a.) Extended aeration Alligator
b.) Chlorination Creek
a.) "DEFIANCE" package
b.) Extended aeration Evaporation
c.) Chlorination percolation
d.) Stabilization pond pond
a.) Extended aeration
b.) Stabilization
lagoon Lemon Bay
a.) "SCHWIND" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond Peace River
a.) Extended aeration
b.) Chlorination Peace River
a.) "DAVCO" package
b.) Contact stabili-
zation Peace River
a.) Extended aeration Olman Canal
b.) Chlorination to Peace
c.) Stabilization pond River
a.) "INHOFF" tank Alligator
b.) Stabilization pond Creek
a.) Contact stabilization
b.) Chlorination
c.) Sand filters Peace River
Charlotte
a.) Trickling filter Harbor
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination Alligator
d.) Stabilization pond Creek
eceiving
basin
2
9
9
9
2
2
2
2
9
2
2
2
2
2
2
2
2
2
XIX-100
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - CHARLOTTE COUNTY. CONT'D.
Population
Name § location served
41. River Forest Mobile
Home Park, Punta Gorda 700
42. River Haven Mobile Home
Park, Rt. 2, Punta Gorda
Flow (GPD)
Avg . Des ign
35,000
15,000
Percent
removal
9/72
BOD 99
SS 94
11/72
BOD 92
SS 97
a.)
b.)
c.)
a.)
b.)
c.)
d.)
a.)
Type of
treatment
Extended aeration
Chlorination
Stabilization pond
"DEFIANCE" package
Extended aeration
Chlorination
Stabilization pond
"AER-0-FLOW"
package
43. Rotonda West
Cape Haze
44. Sea Cove Motels
Punta Gorda
45. Seahorse Apartments
Englewood
46. Shell Creek Park Camp
Punta Gorda 200
47. Spanish Cove
Englewood
48. Sun 'N Shade Family
Campground, Punta Gorda
49. Tiki Condominium
Englewood 80
50. Windmill Village of
Punta Gorda, Punta Gorda 400
25,000
3,000
7,250
10,000
15,000
10,000
4,000
17,000
10/72
BOD 98
SS 88
Engineer's
estimate
BOD 90
10/72
BOD 97
SS 89
11/72
BOD 99
SS 98
Engineer's
estimate
BOD 90
Engineer's
estimate
BOD 90
9/72
BOD 95
SS 80
9/72
BOD 98
SS 92
b.)
= 0
d.)
a.)
Extended aeration
Chlorination
Stabilization pond
"AER-0-BIC"
Receiving Receiving
stream basin
Peace River
Alligator
Creek
Evaporation
percolation
pond (Golf
Course irri-
gation
2
2
9
package
b.)
c.)
a.)
b.)
c.)
a.)
b.)
c.)
d.)
a.)
b.)
a.)
b.)
a.)
b.)
a.)
b.)
c.)
d.)
Extended aeration
Chlorination
"DEFIANCE" package
Extended aeration
Chlorination
Drainfield
Drainfield
1
9
"LIFTAIRE-3" package
Extended aeration
Chlorination
Stabilization pond
Extended aeration
Chlorination
Extended aeration
Chlorination
Extended aeration
Chlorination
"DEFIANCE" package
Extended aeration
Chlorination
Stabilization pond
Shell
Creek
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Drainfield
2
9
9
<9
9
XIX-101
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - COLLIER COUNTY.
Population
Name § location served
1. Avalon Elementary School
Thomasson Dr., E. Naples 750
2. Baker Carroll Point
Naples 120
3. Brookside Bowling Alley
Davis Blvd. , Naples
4. Capri Village Trailer
Park, Cove Dr., Naples 180
5. Captain's Table of Ever-
glades City, Everglades City 332
6. Caribbean Court Trailer Park
Tamiami Trail, N. Naples 500
7. Coconut River Estates,
Airport Road, Naples
8. Collier County Administra-
tion § Court Bldg. , Immokalee
9. Collier County Government
Center, E. Naples
10. Collier County Housing
Authority, Immokalee Immokalee
11. Collier-Seminole State
Park, near Royal Palm Hammock 190
12. Crawford Apartments
Immokalee 75
13. Duda § Sons Migrant
Workers Camp, Tamiami Trail,
South Naples Oo
14. East Naples Middle School
Estey Dr. , Naples
15. El Rancho Mobile Village
Radio Rd. , Naples 250
16. Enchanting Acres Trailer
Park, Enchanting Acres Blvd.
Naples 475
17. Everglades City, Ever-
glades City 1,000
18. Everglades Subdivision
Ochopee
19. Florida-Vanderbilt Devel-
opment Corp., Gulf shore Drive,
Naples 240
Flow (GPD)
Avg. Design
8,000
70,000
2,600
10,000
34,000
25,000
50,000
5,000
12,000
60,000
5,000
7,500
6,000
13,500
15,000
25,000
100,000
35,000
12,000
Percent
removal
Engineer's
estimate
BOD 96
SS 97
12/72
BOD 89
SS 98
11/72
BOD 99
SS 92
8/72
BOD 92
SS 75
Engineer's
estimate
BOD 90
12/72
BOD 91
Engineer's
estimate
BOD 90
11/72
BOD 92
SS 93
12/72
BOD 75
SS 91
12/72
BOD 97
SS 98
11/72
BOD 87
SS 81
7/72
BOD 92
SS 94
11/72
BOD 99
SS 93
7/72
BOD 89
SS 78
Engineer ' s
estimate
BOD 90
Type of
treatment
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "DAVCO" package
b.) Contact stabili-
zation
c.) Chlorination
d.) Microstrainer
a.) Extended aeration
b.) Chlorination
a.) "AERO-JET" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "DAVCO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) "IMHOFF" tank
b.) Chlorination
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Contact stabili-
zation
b.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
Receiving Receiving
stream basin
Naples Bay
Canal to
Gulf
Canal to
Gulf
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Haldeman
Creek
Evaporation
percolation
pond
Addison Bay
to Big Marco
Pass
Overflow to
storm sewer
Overflow to
farm ditch
Evaporation
percolation
pond
Overflow to
ditch
Haldeman
Creek
Tidal canal
to Chokolos-
kee Bay
Evaporation
percolation
pond
Drainfield
5
8
8
9
9
9
9
9
5
9
6
8
8
9
8
5
7
9
9
XIX-102
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - COLLIER COUNTY. CONT'D.
Population
Name § location served
20. Forest Lake Estates, Pine
Ridge Rd. , Naples
21. Glades Apt. Subdivision,
Tamiami Trail, E. Naples 400
22. City of Golden Gate Subdi-
vision, Golden Gate 3,000
23. Golden Lion Motor Inn 50 unit
Ochopee motel
24. Greystone Park, Henderson
Creek, Rt. 1, Naples 292
25. Gulf Winds East Develop-
ment, Marco Island 90
26. Harbor Lakes Co-op Club,
Sandpiper St . , Naples 80
27. Harmony Shores Trailer Port
Bamboo Dr., E. Naples 360
28. Highland Elementary School,
Lake Trafford Rd. , Immokalee 90
29. Highland Elementary School,
Lake Trafford Rd. , Immokalee 750
30. Hitching Post Trailer Park
Barefoot Williams Rd. , Naples 510
31. Holiday Manor Trailer Park
Henderson Creek Dr. , Naples 680
32. Home's Store § Restau-
rant, Tamiami Trail, Naples
33. I.D. Corp. (Isle of Capri
Tarpon Village, Isle of Capri
34. Immokalee Apartments
Immokalee 480
35. Immokalee High School,
9th St., Immokalee 135
36. Immokalee Stockade
Immokalee
37. Immokalee Vegetable Har-
vesting § Hauling, Immokalee 100
Flow (GPD)
Avg. Design
25,000
40,000
70,000 300,000
6,900 15,000
14,000
9,000
6,000
13,500
9,000
10,000
40,000
34,000
2,600
15,000
36,000
13,500
10,000
5,000
Percent
removal
11/72
BOD 94
SS 95
9/72
BOD 92
SS 88
9/72
BOD 90
SS 87
11/72
BOD 87
'SS 87
12/72
BOD 59
SS 44
9/72
BOD 87
SS 93
11/72
BOD 99
SS 91
11/72
BOD 96
SS 93
10/72
BOD 95
SS 87
10/72
BOD 89
SS 78
9/72
BOD 79
SS 68
9/72
BOD 97
SS 80
Type of
treatment
a.) Extended aeration
b.) Chlorination
a.) "DEFIANCE" package
b.) Activated sludge
c.) Chlorination
a.) Activated sludge
b.) Chlorination
c.) Stabilization pond
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
Pond only loading
of 12. S Ib. of BOD per
acre per day
a.) "DEFIANCE" package
b.) Contact stabili-
zation
c.) Chlorination
d.) Stabilization pond
a.) "DAVCO" package
b.) Contact stabili-
zation
c.) Chlorination
d.) Stabilization pond
a. 3 "SECO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Sand filters
a.) "SECO" package
b.) Contact stabili-
zation
c.) Chlorination
d.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
Receiving Receiving
stream basin
Evaporation perc-
olation pond (Golf
course irrigation) 9
Evaporation
percolation
pond 9
Gordon River 5
Borrow pit 9
Henderson
Creek 6
Evaporation
percolation
pond 9
Evaporation
percolation
pond 9
Evaporation
percolation
pond 9
Drainfield 9
Evaporation
percolation
pond 9
Evaporation
percolation
pond 9
Evaporation
percolation
pond 9
Ditch 8
Drainfield 9
Drainfield 9
Drainfield 9
Ditch 9
Evaporation
percolation
pond 9
XIX-103
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - COLLIER COUNTY. CONT'D.
Population Flow (GPD)
Name § location served Avg. Design
38. King's Crown Motel of
Naples, Vanderbilt Beach,
Naples 69
39. Lake Trafford Elementary
School, Lake Trafford Rd.,
Immokalee 750
40. Lake Trafford Marina
Immokalee 100
41. LaPlaya Motel, Gulf
Shore Drive, Vanderbilt
Beach, Naples 200
42. Lely Tropical Estates
Subdivision, E. Naples 2,000
43. M § E Mobile Home Park
Isle of Capri
44. Marco Beach Subdivision
Marco Island 2,000
45. Marco Island Inn
Marco Island
46. Marco Island Toll Bridge,
Toll Plaze, Marco Island
47. Marco Towers Apartments
Isle of Capri 465
48. McGinnis Apartment Com-
ples, Kelly Rd. , Naples 120
49. Moorhead Manor Trailer
Park, Kelly Rd. , Naples 400
50. Mozzone Mobile Home Park
12 mi. south of Naples, U.S. 41
51. City of Naples, 5th Avenue
North, Naples 20,000
52. Naples Fruit 5 Vegetable
Co-op (Camp Happy) Rt.2, Naples 200
S3. Naples Kampgrounds, Bare-
foot Williams Rd. , Naples 300
54. Naples Land Yacht Harbor
Naples 500
55. Naples Mobile Estates,
Radio Rd. , Naples 534
56. Naples Shopping Plaza
Tamiami Trail, Naples 50
3,000 6,900
10,000
5,000
6,100 15,000
200,000
123,000 200,000
4,500 10,000
1,650
30,000 35,000
9,000 10,000
17,000 20,000
30,000
2.535
1.9 MGD MGD
20,000
15,000
7,500 25,000
28,000
10,000
Percent
removal
4/72
BOD 94
SS 77
12/72
BOD 96
SS 94
11/72
BOD 86
SS 71
12/72
BOD 94
SS 93
9/72
BOD 90
SS 69
11/72
BOD 88
SS 51
9/72
BOD 97
SS 70
11/72
BOD 94
SS 68
11/72
BOD 99
SS 97
11/72
BOD 90
SS 79
Engineer's
estimate
BOD 90
9/72
BOD 98
SS 98
11/72
BOD 94
SS 99
11/72
BOD 96
SS 90
11/72
BOD 95
SS 96
11/72
BOD 99
SS 94
11/72
BOD 98
SS 95
Type of
treatment
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a. ) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.} Stabilization pond
a.) "DAVCO" package
b.) Contact stabili-
zation
c.) Chlorination
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) "DAVCO" package
b.) Contact stabili-
zation
c.) Chlorination
d.) Stabilization pond
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
Receiving Receiving
stream basin
Canal to Gulf
Drainfield
Evaporation
percolation
pond
Tributary to
Gulf
Evaporation
percolation
pond
Spray irriga-
tion on Golf
Course
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Evaporation
percolation
pond
Evaporation
percolation
pond, overflow
to Gulf
Evaporation
percolation
pond
Gordon River
Evaporation
percolation
pond
Evaporation
percolation
pond
Haldeman
Creek
Evaporation
percolation
pond
Oyster Bay
8
9
9
8
9
9
9
9
9
9
8
9
5
9
9
5
9
5
XIX-104
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - COLLIER COUNTY, CONT'D.
Population Flow (GPD)
Name § location served Avg. Design
57. Naples Tomato Growers,
Inc. (Camp Sandy) , 12 miles
north of Naples 114
58, Palm River Estates Subdi-
vision, Palm River Blvd.,
Naples 500
59. Palm River Mobile Home
Park, Palm River Estates, Naples
60. Pinecrest Elementary
School, 9th St.,-Iimnokalee
61. Plantation Island No. 3
Trailer Park, Everglades City 303
62. Port-Au-Prince
Naples 400
63. Remuda Ranch Estates, Tarn-
iami Trail South, Naples
64. River Bend Mobile Home
Park, North of Naples
65. Riverwood Mobile Home
Park, Isle of Capri Rd. , Naples
66. Riviera Colony Mobile Home
Subdivision, East Naples 800
67. Royal Cove, U.S. 41
North, Napjes
68. Royal Park Villas Condo-
miniums, Weeks Ave., Naples 500
69. Sandy Ridge Camp, Basso
Farms, North Naples 100
70. Sea Gate Elementary
School, Sea Gate Dr., Naples 65
71. Shadowlawn Elementary School,
Shadowlawn Ave., Naples 780
72. Sorrento Villas Apartments
Naples
73. Sunny Acres Mobile Home
Park, Radio Rd. , Naples
74. Tara Heights 8th Avenue
Trailer Park, Naples 146
75. Trafford Pine Estates
Immokalee
4,000
50,000
5,000 8,000
7,500
10,000 15,000
40,000
North
35,000
South
50,000
6,600
10,000
40,000
5,000
25,000
10,000
6,500
8,000
20,000
7,500
12,000
37,500
Percent
removal
11/72
BOD 98
SS 92
4/72
BOD 92
SS 93
11/72
BOD 91
SS 99
9/72
BOD 81
SS 82
11/72
BOD 97
SS 92
11/72
BOD 98
SS 99
9/72
BOD 93
SS 90
7/72
BOD 97
SS 99
4/72
BOD 76
SS 80
12/72
BOD 84
Engineer's
estimate
BOD 95
Type of
treatment
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Extended aeration
b.) Stabilization pond
a.) "MAROLF" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) "MARPLF" package
b.) Extended aeration
c.) Chlorination
d.) Stabilization pond
a.) Septic tank
b.) Sand filter
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) "SECO" package
b.) Extended aeration
c.) Chlorination
a.) Extended aeration
b.) Chlorination
Receiving
stream
SCK drainage
canal
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Barron River
Evaporation
percolation
pond
Evaporation
percolation
pond
Overflow to
tidal canal
Evaporation
percolation
pond
Drainage
canal
Evaporation
percolation
pond
Haldeman
Creek
Evaporation
percolation
pond
Drainage
ditch
Ditch
Drainfield
Evaporation
percolation
pond
Drainfield
Evaporation
percolation
pond
Receiving
basin
8
9
9
9
7
9
9
8
9
8
9
5
9
8
8
9
9
9
9
XIX-105
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - COLLIER COUNTY. CONT'D.
Population
Name § location served
76. Vanderbilt Beach Club
Vanderbilt Beach, Naples 72
77. Vanderbilt lovers, Van-
derbilt Beach, Naples
78. Villa De Marco
Marco Island 200
79. Windmill Village of Naples,
Mobile Home Park, Naples 475
Flow (GPD) Percent
Avg. Design removal
8/71
BOD 92
5,000 SS 97
12/72
BOD 89
70,000 SS 98
11/72
BOD 60
10,000 SS 75
9/72
BOD 93
33,000 SS 90
Type of Receiving Receiving
treatment stream basin
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination Wiggins Pass
d.) Stabilization pond Waterway
a.) Contact stabili-
zation
b.) Chlorinfition River to Gulf
a.) Extended aeration
b.) Chlorination
c.) Micro filter Marco River
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond Drainfield
5
8
6
9
XIX-106
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - GLADES COUNTY.
Population Flow (GPD) Percent
Name § location served Avg. Design removal
1. Benbow Village
Moore Haven 400
2. Caloosa Lodge Mobile
Home Park, Glades
3. Canal Front Trailer Park
Moore Haven 90
4. Citrus Center
Moore Haven 60
5. Devel's Garden Indian
Reservation
6. Fisheating Creek
Palmdale
7. Glades County Sugar Growers
Co-op Association, Moore Haven 360
8. Glades County Sugar Growers
Cp-op Plant, Moore Haven 200
9. Lykes Park Campgrounds
10. Moore Haven School
Moore Haven 1,000
11. River Oakes Subdivision
fj Campground, Moore Haven
10/72 a.)
BOD 99 b.)
28,000 SS 85 c.)
Engineer's
estimate a.)
10,000 BOD 90 b.)
a.)
5,400 b.)
a.)
b.)
c.)
3,000 d.)
a.)
a.)
10,000 b.)
3/72 a.)
BOD 99 b.)
15,000 SS 97 c.)
a.)
b.)
5,000 c.)
a.)
10,000 b.)
a.)
12,000 15,000 b.)
Engineer's
estimate a.)
40,000 BOD 90 b.)
a.)
Type of
treatment
"YEOMAN" package
Extended aeration
Stabilization pond
Extended aeration
Chlorination
Extended aeration
Chlorination
"SECO" package
Extended aeration
Chlorination
Stabilization pond
"IMHOFF" tank
Extended aeration
Chlorination
Extended aeration
Chlorination
Stabilization pond
Extended aeration
Chlorination
Stabilization pond
Extended aeration
Chlorination
"DEFIANCE" package
Extended aeration
Extended aeration
Chlorination
"AER-0-FLOW"
Engineer's package
12. Shawnee farms,
Clewiston
13. Sportsman Village Trailer
Park, Moore Haven 160
14. Sugar Cane Harvesting, Inc.
Clewiston 120
estimate b.)
25,000 BOD 90 c.)
a.)
b.)
c.)
7,500 10,000 d.)
a.)
b.)
c.)
12,000 d.)
Extended aeration
Chlorination
"DEFIANCE" package
Extended aeration
Chlorination
Stabilization pond
"MAROLF" package
Extended aeration
Chlorination
Stabilization pond
Receiving Receiving
stream basin
Drainage
canal
Drainfield
Tile
drainfield
Ditch to Caloosa-
hatchee River
Fisheating
Creek
Evaporation
percolation
pond
Evaporation
percolation
pond
Ditch to Caloosa-
hatchee River
Canal to Caloosa-
hatchee River
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
8
9
9
1
8
9
9
1
1
9
9
9
9
XIX-107
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - HENDRY COUNTY.
Population Flow (GPD) Percent Type of
Name § location served Avg. Design removal treatment
1. Air Glades Air Park, Inc.
Clewiston
2. City of Clewiston,
S.R. 80, Clewiston 2,500
3. A. Duda § Sons
LaBelle
4. Farm Workers Community
Clewiston 275
5. Harlem Academy
Clewiston
6. Harvesters Village
Clewiston 370
7. Hooker Point Labor
Quarters, South Bay
8. City of LaBelle,
Citrus St., LaBelle 600
9. LaBelle Elementary School
LaBelle
10. Shumacher KOA Camp
Clewiston 200
a.) Extended aeration
47,000 b.} Chlorination
a.) Contact stabili-
zation
8/72 b.) Chlorination
BOD 96 c.) Stabilization
80,000 250,000 SS 70 lagoon
a.) Extended aeration
5,000 b.) Chlorination
a.) Extended aeration
b.) Chlorination
20,000 c.) Stabilization pond
9/72
BOD 97 a.) Extended aeration
5,000 SS 87 b.) Chlorination
a.) "DAVCO" package
b.) Extended aeration
26,000 c.) Chlorination
Engineer's a.) "AERO-JET" package
estimate b.) Extended aeration
25,000 BOD 90 c.) Chlorination
a.) Contact stabili-
5/72 zation
BOD 52 b.) Chlorination
60,000 SS 84 c.) Stabilization pond
a.) "DEFIANCE" package
b.) Extended aeration
2,600 c.) Chlorination
a.) Extended aeration
b.) Chlorination
9.000 c.j Stabilization pond
Receiving Receiving
stream basin
Evaporation
percolation
pond
Canal to Lake
Okeechobee
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Spray irri-
gation
Evaporation
percolation
pond
Canal to Lake
Okeechobee
Canal to Caloosa-
hatchee River to
Lake Okeechobee
Evaporation
percolation
pond
9
1
9
9
9
9
9
1
1
9
XIX-108
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - LEE COUNTY.
Population
Name § location served
1. Bamboo Mobile Village,
Tamiami Trail, Bonita Springs 102
2. Bay Shore Estates, Mobile
Homes, N. Ft. Myers
3. Bayside Estates, Ft.
Myers Beach
4. Bermuda Cay Condominium
Apartments, Ft. Myers
5. Bishop Verot High School
Ft. Myers
6. Bonita Beach Apartments
Bonita Springs
7. Bonita Beach Trailer Park
Bonita Springs
8. Bonita Egret, Bonita
Beach
9. Bonita Springs Camp-
ground, Bonita Springs
10. Buccaneer Resort Inn
Ft. Myers Beach
11. Cabover Diner, 1383 Tam-
iami Trail, Ft. Myers
12. Caloosa School
Cape Coral
13. Captain's Walk Condomin-
iums, Sanibel Island 96
14. Carl's Sunrise Village
Ft, Myers
15. Carriage Village Mobile
Home Park, Ft. Myers
16. Casa Bonita Apartments
Bonita Springs
17. Century 21, North
Ft. Myers 664
18. Chateau Estates, North
Ft. Myers
19. Cherry Estates, St. James
20. Colony, Sanibel Island
21. Covered Wagon Trailer
Park, Estero
22. Develco (Plantation Estates)
Ft. Myers
Flow (GPD)
Avg. Design
1,067 5,000
20,000
30,000
7,500
4,000
5,000
5,000
17,000
3,360
2,000
20,000
10,000
40,000
20,000
40,000
125,000
50,000
5,000
15,000
15,000
Percent
removal
9/72
BOD 96
10/72
BOD 98
SS 84
12/72
BOD 99
SS 98
Engineer's
estimate
BOD 90
Engineer ' s
estimate
BOD 90
12/72
BOD 76
SS 90
10/72
BOD 90
SS 40
12/72
BOD 85
SS 82
Engineer's
estimate
BOD 90
12/72
BOD 98
SS 86
12/72
BOD 97
SS 97
Engineer's
estimate
BOD 90
Engineer's
estimate
BOD 90
11/72
BOD 97
SS 93
12/72
BOD 99
SS 99
Engineer's
estimate
BOD 90
Type of Receiving
treatment stream
a.) Extended aeration
b.) Chlorination Leitner's
c.) Stabilization pond Creek
Evaporation
a.) Extended aeration percolation
b.} Chlorination pond
a.) Contact stabili-
zation Land spray
b.) Chlorination irrigation
a.) Extended aeration Ditch to Ca-
b.) Chlorination loosahatchee
a.) Extended aeration
b.) Chlorination Drainfield
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
a.) Extended aeration
b.) Chlorination Drainfield
a.} Extended aeration
b.) Chlorination Drainfield
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
a.) Extended aeration
b.) Chlorination Drainfield
a.) Contact Stabili- Evaporation
zation percolation
b.) Chlorination pond
a.) "DEFIANCE" package Evaporation
b.) Extended aeration percolation
c.) Chlorination pond
a.) Contact stabili- Evaporation
zation percolation
b.) Chlorination pond
a.) Contact stabili- Evaporation
zation percolation
b.) Chlorination pond
a.) Contact stabili- Evaporation
zation percolation
b.) Chlorination pond
a.) Extended aeration
b.) Chlorination Drainfield
a.) Extended aeration Surface irri-
b.) Chlorination gation
a.) Extended aeration Irrigation,
b.) Chlorination drainfield
Receiving
basin
4
9
9
3
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
XIX- 109
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - LEE COUNTY. CONT'D.
Population Flow (GPD)
Name § location served Avg. Design
23. Diamond Head Inn, Ft.
Myers Beach 5,000
24. Dodrill, Rufus, Sanibel
Island 25,000
25. Estero Beach Club
Apartments, Ft. Myers Beach 11,250
26. Estero Beach Club East
Ft. Myers Beach 15,000
27. Estero Island Inn
Estero Island
Percent
removal
12/72
BOD 99
SS 84
Engineer's
estimate
BOD 90
12/72
BOD 99
SS 88
Engineer's
estimate
BOD 90
12/72
BOD 99
SS 87
a.)
b.)
c.)
a.)
b.)
a.)
b.)
a.)
b.)
Type of
treatment
"AER-0-BIC" package
Extended aeration
Chlorination
Extended aeration
Chlorination
Extended aeration
Chlorination
Extended aeration
Chlorination
Receiving
stream
Drainfield
Evaporation
percolation
pond
Drainfield
Drainfield
Receiving
basin
9
9
9
9
28. Estero 7000, Ft. Myers
Beach
29. Eventide Motel, Ft.
Myers Beach
30. Everglades Lanes, Tam-
iami Trail, Ft. Myers
31. Family Estates Inc.
N. Ft. Myers
32. Fiesta Village, Cypress
Lakes
33. Flamingo Bay Inc.
Ft. Myers
34. Florida Cities Water
Company, Tropicanna Dr.,
Fiesta Village
35. Florida Cities Water
Company, Tropical Isles
Subdivision, Ft. Myers
36. City of Ft. Myers,
South of Bowling Green
Subdivision, Ft. Myers
37. City of Ft. Myers,
Raleigh St., Ft. Myers
38. Ft. Myers Beach
Hotel, Ft. Myers Beach
39. Four Seasons, N. Ft.
Myers
40. Fox Trailer Park,
Queen St., Ft. Myers
41. Garden Cove Mobile
Home Park, Ft. Myers
100,000
6,500
3,400 4,370
25,000
300,000
35,000
1,400 94,000 150,000
3,300 146,000 330,000
14,000 3.0 MGD 6.0 MGD
14,000 .3.0 MGD 4.0 MGD
30,000
15,000
124 1,800 4,910
12,250
12/72
BOD 96
SS 71
11/72
BOD 99
SS 56
Engineer's
estimate
BOD 97
12/72
BOD 99
SS 81
Engineer's
estimate
BOD 90
SS 90
Engineer's
estimate
BOD 95
Engineer's
estimate
BOD 97
10/72
BOD 94
SS 45
Engineer's
estimate
BOD 70
Engineer's
estimate
BOD 95
Engineer's
estimate
BOD 90
SS 90
8/72
BOD 99
SS 92
7/71
BOD 93
Evaporation
percolation
a.) "MAROLF" package pond § Golf
b.) Extended aeration course irrl-
c.) Chlorination gation
a.) Extended aeration
b.) Chlorination
Drainfield
a.) Extended aeration Evaporation
b.) Chlorination percolation
c.) Stabilization pond pond
a.). Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
Evaporation
percolation
pond
Canal to
Caloosahat-
chee River
a.) "AER-0-BIC" package Evaporation
b.) Extended aeration percolation
c.) Chlorination pond
a.) Contact stabili-
zation Lagoon to
b.) Chlorination Caloosahat-
c.) Stabilization pond chee River
a.) Activated sludge
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Trickling filters
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
Caloosahat-
chee River
Caloosahat-
chee River
Caloosahatchee
River
Drainfield
a.) "AER-0-BIC" package Evaporation
b.) Extended aeration percolation
c.) Chlorination pond
a.)
b.)
Extended aeration
Chlorination
a.) Extended aeration
b.) Chlorination
Evaporation
percolation
pond
Evaporation
percolation
pond
XIX-110
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - LEE COUNTY. CONT'D.
Population
Name § location served
42. Gasparilla Island Water
Assoc., Gasparilla Island
43. Gulf Air Trailer Park
Ft. Myers
44. Gulf Bay Estates, Ft.
Myers Beach
45. Gulf Shores Resort Condo-
miniums, Ft. Myers Beach
46. Holiday Inn of Ft. Myers,
Beach, Ft. Myers Beach
47. Holiday Inn Trav-L-Park
Ft. Myers
48. Hyde Park Condominiums
Ft. Myers
49. Indian Creek Mobile
Home Park, Ft. Myers
50. lona Trailer Ranch, Rt. 69 trail -
#5, Ft. Myers ers
51. Island Beach Club
Sanibel Island 260
52. Island Inn, Sanibel
Island
53. John A. Johnson Travel
Trailer Park, Golden Lake Dr.,
Ft. Myers 72
54. Lazy Days Mobile Home
Park, Ft. Myers
55. Leasure Village Mobile
Home Park, N. Ft. Myers
56. Lehigh Utilities Inc.,
Lehigh Acres 14,800
57. Leonardo Arms
Ft. Myers Beach
58. Mariner Properties
Sanibel Island
59. McGregor Mobile Manor
Ft. Myers
60. Mid-Island Villas
Ft. Myers Beach
Flow (GPD)
Avg. Design
275,000
20,000
15,000
20,000
25,000
15,000
30,000
25,000
3,300 5,000
2,600
10,000
400 3,600
3,600
150,000
413,000 480,000
40,000
30,000
30,000
7,500
Percent
removal
12/72
BOD 98
SS 62
12/72
BOD 98
SS 91
Engineer's
estimate
BOD 90
SS 90
9/72
BOD 99
SS 91
12/72
BOD 98
SS 93
12/72
BOD 99
SS 85
Engineer's
estimate
BOD 90
SS 90
Engineer's
estimate
BOD 94
7/72
BOD 98
SS 97
12/72
BOD 99
SS 89
12/72
BOD 95
SS 90
12/72
BOD 99
SS 99
10/72
BOD 92
SS 90
9/72
BOD 98
SS 91
11/72
BOD 98
SS 95
Engineer's
estimate
BOD 95
Type of Receiving
treatment stream
Evaporation
percolation
a.) Contact stabili- pond 6 Golf
zation course irri-
b.) Chlorination gation
a.) Extended aeration
b.) Chlorination Drainfield
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
a.) "AER-0-FLOW" package
b . ) Extended aeration
c.) Chlorination Drainfield
a.) "AER-0-JET" package
b . ) Extended aeration
c.) Chlorination Drainfield
a.) Extended aeration
b.) Chlorination Drainfield
a.) "DEFIANCE" package
b.} Extended aeration
c.) Chlorination Drainfield
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
Canal to
a.) Extended aeration Caloosahat-
b.) Chlorination chee River
a.) Extended aeration Evaporation
b.) Chlorination percolation
c.) Stabilization pond pond
a.) Extended aeration
b.) Chlorination Drainfield
a.) Extended aeration
b.) Chlorination Drainfield
a.) Extended aeration
b.) Chlorination Drainfield
a.) Contact stabili- Evaporation
zation percolation
b.) Chlorination pond
a.) Contact stabili-
zation
b.) Chlorination
c.) Stabilization pond Able Canal
a.) "AER-0-FLOW" package
b.) Extended aeration
c.) Chlorination Drainfield
Evaporation
a.) Extended aeration percolation
b.) Chlorination pond
Evaporation
a.) Extended aeration percolation
b.} Chlorination pond
a.) Extended aeration
b.) Chlorination Drainfield
Receiving
basin
9
9
9
9
9
9
9
9
3
9
9
9
9
9
3
9
9
9
9
61. Mobile Land 6 Title Co.,
Bayshore Rd., Ft. Myers
800
Engineer's
estimate
10,000 20,000 BOD 66
a.) Contact stabili-
zation
b.) Chlorination
Drainfield
XIX-111
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - LEE COUNTY. CONT'D.
Population
Name 6 location served
62. Mound Key Development
Corp. , Ft. Myers
63. Naples-Ft. Myers Kennel
Club, Bonita Springs
64. Nationwide Realty
Sanibel Island
65. Neptune Inn, Beach 68 motel
Estates units
66. Oak Creek Trailer Park
Bonita Springs
67. Ohio Medical Products,
Tice St. , Tice
68. Old Bridge Park Corp.,
N. Ft. Myers
69. Orange Harbor Mobile
Home Park, Palm Beach Blvd.,
Ft. Myers 325
70. Page Mobile Village,
Cleveland Ave. , Ft. Myers 400
71. Peaceful Pines Mobile
Village, Ft. Myers
72. Pine Acres Travel Trailer
Park, Pine Island
73. Pines Mobile Home Park II
N. Ft. Myers
74. Pink Shell Cottages
Ft. Myers Beach
75. Poinsettia Mobile
Home Park, Ft. Myers
76. Privateer Condominium,
Estero Blvd., Ft. Myers Beach 192
77. R § R Trailer Park
Ft. Myers
78. Red Coconut Trailer
Park, Estero Blvd., Ft.
Myers Beach 380
79. Jerry Ritter Trailer
Park, Ft. Myers 97 trailers
80. Riverdale High School
Ft. Myers
81. River Gardens Mobile Home
Park, Ft. Myers
Flow (GPD)
Avg. Design
51,000
40,000
800,000
5,350 7,500
5,000
4,000
80,000
8,000 25,000
15,000 30,000
30,000
9,000
160,000
10,000
50,000
5,000 10,000
30,000
15,000 20,000
10,000
20,000
Percent
removal
12/72
BOD 99
SS 99
Engineer's
estimate
BOD 90
12/72
BOD 99
SS 90
Engineer's
estimate
BOD 96
SS 99
12/72
BOD 99
SS 99
10/72
BOD 91
SS 96
7/72
BOD 99
SS 98
Engineer ' s
estimate
BOD 83
Engineer's
estimate
BOD 90
SS 90
Engineer's
estimate
BOD 90
SS 90
Engineer's
estimate
BOD 90
SS 90
11/72
BOD 98
SS 95
11/72
BOD 96
SS 90
Engineer's
estimate
BOD 90
SS 90
11/72
BOD 95
SS 93
Engineer's
estimate
BOD 90
12/72
BOD 92
SS 91
8/72
BOD 97
SS 97
Type of
treatment
a.) Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
a.) Trickling filters
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
Receiving Receiving
stream basin
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Imperial River
to Creek
Evaporation
percolation
pond
Evaporation
percolation
pond
Canal to
Caloosahat-
chee River
Ditch to
Whiskey Creek
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
9
9
9
9
4
9
9
3
3
9
9
9
a.) "AER-0-FLOW" package
b.) Extended aeration
c.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
-
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
Drainfield
Evaporation
percolation
pond
Drainfield
Evaporation
percolation
pond
Ma t an z as Pass
Evaporation
percolation
pond
Evaporation
percolation
pond
9
9
9
9
4
9
9
XIX-112
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - LEE COUNTY. CONT'D.
Population
Name 8 location served
82. Royal Palm Plaza Shopping
Center, Ft. Myers
83. San Carlos Subdivision
San Carlos Park
84. San Carlos Trailer Park,
San Carlos Blvd., Ft. Myers 73
85. Sandpiper Gulf Resort,
Estero Blvd., Ft. Myers 100
86. Sanibel Center Building
Sanibel Island 9 stores
87. Serendipity South Mobile
Home Park
88. Shell Harbor Inn 36 apart -
Sanibel Island ments
89. Sherwood Forest (Vista
Village), N. Ft. Myers
90. Shibui Apartments
Sanibel Island
91. Skyvilla Mobile Home Park
N. Ft. Myers 80
92. South Seas Plantation
Captiva Island
93. Southwest Florida, Bonita
Springs 4,000
94. Spring Creek Village
Estero
95. Star Plaza Shopping Center
N. Ft. Myers
96. Stevens 5 Layton Mobile
Home Park, Alva
97. Sun § Fun Travel Trailer
Park, Palm Beach Blvd., Ft.
Myers 440
98. Sunland Training Center 1,200
99. Swifts Trailer Park
N. Ft. Myers
100. Tahiti Mobile Village
Estero
101. Tamishaw Trailer Park,
2557 N. Tamiami Trail, Ft. 45 trail-
Myer^ ers
102. Tangl'swoo.i Elementary
School, Tanglewood, Ft. Myers
Flow (GPD)
Avg. Design
18,000
150,000
7,250 10,000
2,000 5,000
3,600
5,000
50,000
10,000
3,000 50,000
20,000
15,000 20,000
40,000
3,500
40,000
21,000 25,000
250,000
10,000
30,000
3,375 5,000
7,500
Percent
removal
Engineer's
estimate
BOD 95
SS 95
7/72
BOD 99
SS 99
7/72
BOD 99
SS 94
9/72
BOD 99
SS 97
12/72
BOD 99
SS 47
12/72
BOD 97
SS 99
11/72
BOD 99
SS 77
Engineer's
estimate
BOD 93
Engineer's
estimate
BOD 90
SS 90
Engineer's
estimate
BOD 95
Engineer's
estimate
BOD 97
10/72
BOD 90
SS 87
6/72
BOD 88
SS 0
11/72
BOD 97
SS 94
12/72
BOD 94
Engineer's
estimate
BOD 90
SS 90
6/72
BOD 99
SS 97
8/72
BOD 99
SS 99
12/72
EOD S3
SS 90
Type of
treatment
a.) Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) "DEFIANCE" package
b.) Extended aeration
c.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.)"AER-0-BIC" package
b.) Extended aeration
c.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Activated sludge
b.) Chlorination
a.) Trickling filters
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
stream
Evaporation
percolation
pond S spray
irrigation
Evaporation
percolation
pond S Golf
course irri-
gation
Drainfield
Drainfield
Evaporation
percolation
pond
Drainfield
Evaporation
percolation
pond
Drainfield
Drainfield
Drainfield
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Orange River
Nine mile run
to Caloosahat-
chee River
Evaporation
percolation
pond
Estero River
Evaporation
percolation
pond
Evaporation
percolation
pond
basin
9
9
9
9
9
9
9
9
9
9
9
9
9
9
3
3
9
4
9
9
XIX-113
-------
APPENDIX PP - DOMESTIC WASTE WATER INVENTORY - LEE COUNTY. CONT'D.
Population
Name § location served
103. Thunderbird Utilities, Inc.
Ft . Myers
104. Tip Top Trailer Village,
Tamiami Trail, Ft. Myers 150
105. Trailwinds Subdivision
Ft. Myers
106. Tremar Motel Apartments,
First St., Ft. Myers 27 units
107. Tropicana Mobile Manor
Ft. Myers 100
108. Twenty Nine Palms,
Estero Blvd., Ft. Myers 60 units
109 . Upriver Campground
N . Ft . Myers
110. Waterway Estates
N. Ft. Myers
111. Whiskey Creek Club
Estates, Ft. Myers
112. Windmill Cottages
Ft. Myers
Flow (GPD) Percent
Avg. Design removal
7/72
BOD 99
50,000 SS 94
5/72
BOD 97
4,000 7,500 SS 79
7/72
BOD 89
125,000 SS 75
Engineer's
estimate
3,360 BOD 96
12/72
BOD 78
8,500 18,750 SS 67
Engineer's
estimate
14,000 15,000 BOD 90
10,000
12/72
BOD 99
330,000 SS 81
2/72
BOD 90
150,000 SS 40
10/72
BOD 95
SS 92
Type of
treatment
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
c.) Stabilization pond
a.) Extended aeration
b.) Chlorination
a.) Extended aeration
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Contact stabili-
zation
b.) Chlorination
a.) Extended aeration
b.) Chlorination
Receiving Receiving
stream basin
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Evaporation
percolation
pond
Drainfield
Evaporation
percolation
pond
Caloosahat-
chee River
Evaporation
percolation
pond
9
9
9
9
9
9
9
3
9
XIX-114
-------
APPENDIX QQ - INDUSTRIAL WASTE WATER SOURCES - CHARLOTTE COUNTY.
Name 5 Location
Type of
waste
Influent
parameters (MG/L)
Flow (GPD)
Avg. Design
Percent
removal
Type of
treatment
Receiving Receiving
stream basin
1. Eagle's Nest
Mobile Home Reverse
Park, North Ft. osmosis
Myers brine
2. Mar- Jon
Laundry, Grove
City
3. Rotonda West,
Rotonda West,
Florida
4. Spinosa
Laundry, Grove
City
5. Thrifty Mart
Coin Laundry,
Englewood
BOD 330, DO 9.4,
Ph 7.6, TS 696,
Laundry SS 224
Reverse
osmosis
brine
Laundry BOD 300, SS 350
Laundry
Engineer's
estimate
BOD 95
12,000 SS 97
Engineer's
estimate
SS 85
a.) Trickling
filter
b.) Chlorination Drainfield
a.) Trickling
filter
b.) Chlorination Lemon Bay
a.) Port
Charlotte STP
9
8
2
XIX-115
-------
APPENDIX QQ - INDUSTRIAL WASTE WATER SOURCES - COLLIER COUNTY.
Name § location
1. Byron Adams
Laundry, 2269
Tamiami Trail,
Naples
2. Ange Laundry,
Immokalee
3. Chesser's
Laundromat ,
Naples
4. Cochran Laun-
dry, Naples
5. Cooper
Slaughterhouse,
Immokalee
6. Everglades
Fisheries, Ever-
glades City
7. H.L.H. Pro-
ducts, Immokalee
8. Moody Laundry,
Immokalee
9. Naples' Coin
Laundry, Whit-
man Blvd. , Naples
10. Naples Quick
Clean Laundry, N.
Tamiami Tr. , Nap
Naples
11. B.K. Popejoy
Laundry, SR 29N,
Immokalee
12. Jack Queen
Laundry, Immok-
alee
13. Shell Mound
Coin Laundry,
Chokoloskee
14. Wash House
Laundry Immok-
alee
15. No name
N. 4th St.,
Immokalle
Influent
Type of parameters
waste (mg/1)
Laundry
Laundry
Laundry
Laundry
Hog § cattle BOD 2000
processing TSS 1000
Fish 6
shrimp pro-
cessing
Canned
tomatoes
processing
Laundry
Laundry BOD 140
Laundry BOD 160
Laundry
Laundry
Laundry BOD 220
Laundry
Laundry
Flow (GPD) Percent
Avg. Design removal
11,000
Engineer's
estimate
BOD 93
12,000 SS 90
Engineer* s
estimate
BOD 90
20 ,,000 SS 90
Engineer's
estimate
5,050 SS 85
3/72
BOD 83
11,300 SS 94
4/72
BOD 86
11,300 SS 81
Engineer's
estimate
14 , 880 BOD 86
2/72
BOD 73
7,000 10,000 SS 74
Type of
treatment
a. J Sand filter
a.) Septic tank
a.) Grease removal
b.) Activated sludge
c.) Chlorination
a.) Stabilization
ponds
a.) Trickling filter
a.) Trickling filter
a.) Trickling filter
a.) Trickling filter
a. ) Koroseal media
aeration
b.) Chlorination
a.) Trickling filter
None
Receiving Receiving
stream basin
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Irrigation
Spray irri-
gation
Drainfield
Drainfield
Evaporation
percolation
pond
Bay
Evaporation
percolation
pond
9
9
9
9
3
9
9
9
7
Q
XIX-116
-------
APPENDIX QQ - INDUSTRIAL WASTE WATER SOURCES - GLADES COUNTY.
Name £
i location
Type or
waste
Influent
parameters
(mg/1)
Flow (GPD)
Avg. Design
Percent
removal
Type of
treatment
Receiving
stream
Receiving
basin
1. Glades County
Sugar Growers Co- Sugar cane
op, Moore Haven processing
2. Glades County Cooling wa-
Sugar Growers Co- ter § boiler
op, Moore Haven blow-down
50,000
1.1 MGD
a.) Evaporation
percolation pond
3. Lawhon's
Laundry, Moore
Haven
4. Moore Haven
Sugar House,
Moore Haven
Laundry
Sugar cane
processing
BOD 1500
1.0 MGD
Engineer's
estimate
BOD 80
a.) Ripple aeration
b.) Stabilization
lagoon
Canal to
Lake Okeecho-
bee
APPENDIX QQ - INDUSTRIAL WASTE WATER SOURCES - HENDRY COUNTY.
Name S location
1. Everglades
Sugar Refinery,
Clewiston
2. U.S. Sugar
Corp, Clewiston
Type of
waste
Sugar cane
processing
BOD
Sugar cane
processing
Influent
parameters Flow (GPD)
(mg/1) Avg. Design
Temp. 24°C
DO 0.9
COD 1052
BOD 400 220,000 140,000
BOD 118
TSS 106
TVS 904
DS 1098
Temp. 71 F
DO 2.6
COD 736
Total N 7.52
Total P 2.0
Chlorides 82
Ph 6.9 3.0 MGD
Percent Type of
removal treatment
Engineer ' s
estimate a.) Five oxidation
BOD 97 ponds
Engineer's
estimate a.) 200 acres of
BOD 96 lagoons
Receiving
stream
Farm canal
to Lake
Okeechobee
Evaporation
percolation
ponds
Receiving
basin
1
9
XIX-117
-------
APPENDIX QQ - INDUSTRIAL WASTE WATER SOURCES - LEE COUNTY.
Type of
Name § location waste
1. Aewett Center
Washarama, 6069 Mc-
Gregor Blvd. Ft.
Myers Laundry
2. Bayshore Coin
Laundry, 1260 N.
Trail, N. Ft.
Myers Laundry
3. Bob's Coin Laun-
dry, 1120 5th St.,
Ft. Myers Beach Laundry
4. Busch Maytag
Coin Laundry,
Terry St, Bonita
Springs Laundry
5. Florida Power
§ Light Co., SR Cooling
80, Tice water
6. Ft. Myers Villa
Laundry, 880 S. Tara-
iami Trail, Ft.
Myers Laundry
7. Handy Corner
Laundry, Nalle Rd.
§ Bayshore, Ft.
Myers Laundry
8. Island Shores
Coin Laundry, 940
3rd St. , Ft. Myers
Beach Laundry
9. Island Water
Association, Brine
Sanibel Island discharge
10. J.F. Land
Coin Laundry,
Pine Island Laundry
11. Mid-Island
Coin Laundry, 165
Sterling Ave. , Ft.
Myers Beach Laundry
12. Paluck Laundry,
Ft. Myers Shores Laundry
13. Pondella Coin
Laundry, Ft. Myers Laundry
14. Prathers
Laundry, 1828
Evans Ave . , Ft .
Myers Laundry
15. San Carlos
Park Laundromat ,
San Carlos Park Laundry
16. Star Plaza
Coin Laundry, U.S.
41 § Willis St.,
Nt. Ft. Myers Laundry
Influent
parameters
fmg/l)
TSS 57
TVS 796
TS 3216
Temp. 35°C
DO 5.4
BOD 41
PH 7.8
BOD 402
SS 159
TS 1703
DO 6.7
PH 8.2
At=9°F
to av.
condi-
tions
BOD 224
TSS 300
Temp. 140 F
BOD 145
SS 190
TS 1434
DO 2.7
PH 7.7
TDS 25,000
CL 10,600
S04 3,450
BOD 205
TSS 198
TS 1977
BOD 200
Temp. 92°F
PH 7.7
DO 4.2
BOD 226
SS 228
TS 1660
Flow (GPD)
Avg. Design
2,900 6,000
4,000 6,000
Circulation
of Caloosa-
hatchee River
571.4 MGD
4,000 10,000
5,800 6,000
640,000
15,000
5,400
14,880
Est.
5,000
10,000
13,700
Percent
removal
2/72
BOD 45
SS 67
2/72
BOD 92
4/72
BOD 88
SS 10
2/72
BOD 88
SS 89
Engineer ' s
estimate
BOD 85
11/72
BOD 89
SS 88
Type of
treatment
a.) Trickling filter
a.) Trickling filter
a.) Trickling filter
b.) Chlorination
a.) Trickling filter
b.) Chlorination
a.) Trickling filter
b.) Chlorination
c.) Stabilization
pond
a.) Trickling filter
b.) Chlorination
Receiving Receiving
stream basin
Evaporation
percolation
pond
Evaporation
percolation
pond
Evaporation
percolation
pond
Drainfield
Caloosahat-
chee River
On central
sewer system
Drainfield
Canal to
Matanzas Pass
Direct discharge to Gulf
3/72
BOD 86
SS 78
4/72
BOD 78
SS 51
1/72
BOD 71
SS 82
2/72
BOD 84
SS 54
4/72
BOD 68
SS 75
a.) Trickling filter
b.) Chlorination
a.) Trickling filter
b.) Chlorination
a.) Trickling filter
b.) Pond
a.) Trickling filter
City sewer
a.) Trickling filter
b.) Sand filter
a.) Trickling filter
Evaporation
percolation
pond
Canal to
San Carlos
Bay
Irrigation
Drainfield
Pond
Powell Creek
9
9
9
9
3
3
9
4
3
9
4
9
9
3
9
3
XIX-118
-------
APPENDIX QQ - INDUSTRIAL WASTE WATER SOURCES - LEE COUNTY. CONT'D.
Name S location
Influent
Type of parameters Flow (GPD) Percent
waste (mg/1) Avg. Design removal
Type of
treatment
Receiving Receiving
stream basin
17. Tic Toe Laun-
dromat, Jean Dr.,
N. Ft. Myers Laundry
18. Trail Dairy,
N. Ft. Myers Dairy
a.) Lagoon
Irrigation
G. P. O. 1974- 74O-2SO / 354S. REGION NO. 4
XIX-119
------- |