Wetland reclamation by accelerating succession

MISSING IMAGE

Material Information

Title:
Wetland reclamation by accelerating succession
Physical Description:
xiv, 267 leaves : ill. ; 28 cm.
Language:
English
Creator:
Rushton, Betty Toombs, 1929-
Publication Date:

Subjects

Subjects / Keywords:
Wetlands -- Florida   ( lcsh )
Ecological succession   ( lcsh )
Reclamation of land -- Florida   ( lcsh )
Phosphates -- Environmental aspects   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1988.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by Betty Toombs Rushton.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001115764
notis - AFL2486
oclc - 19924469
sobekcm - AA00004825_00001
System ID:
AA00004825:00001

Full Text











WETLAND RECLAMATION SY ACCELERATING SUCCESSION


By



BETTY TOOMBS RUSHTON























A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY
OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1988














ACKNOWLEDGMENTS


I thank Dr. H. T. Odum for his unique insight and perspective into

systems analysis and my faculty committee members Dr. Ronnie Best, Dr.

Mark Brown, Dr. Don Graetz, and Dr. Wayne Huber for guidance in their

specialized fields.

The project was supported by the Florida Institute of Phosphate

Research number 83-03-041R, "Interactions Between Wetlands and Phosphate

Mining," H. T. Odum and G. R. Best, principal investigators. A student

cooperative program between the University of Florida and The Florida

Bureau of Land Reclamation increased the scope of the project.

Several mining companies, government agencies, and private

individuals provided access to a wide variety clay settling ponds for

study sites. Cooperation is gratefully acknowledged. Special assistance

was given by Mike Preim and Larry Odom of Gardinier, Bill Hawkins at

Mobil, Selwyn Presnell at Agrico, Jim Sampson at CF Industries, Jay Allen

at IMC, and Jim Murriam at Tenoroc State Reserve. Tim King of The

Florida Game and Freshwater Fish Commission helped locate sites and

contributed ideas.

Tree planting, site preparation and seedling measurements were

assisted by M. Miller, C. Bersok, J. Feiertag, C. Peze! hkl,. Arrieta,

C. Irwin, P. Wallace, R. Wolfe, R. Hassoun, S. Best, R. Bartelson, D.

Segal, H. Meyer, M. Whitton, and especially Alfonso Hernandez and his

family and others. J. Ley and S. Tennenbaum helped with determining








elevations. B. Sargent installed wells. S. Swank and D. Cronwell

measured trees and water table levels. K. Merritt helped with fieldwork,

entered data, and ran soil analyses. M. Paulic and S. Best helped with

data entry. D. Segal and C. Pezeshki shared their experience with the

Wetland Site Index. Jim Feiertag explained statistics and SAS. Mary

Davis allowed the use of unpublished data about Florida wetlands and

provided ideas.

Linda Crowder and Jenny Carter helped with text and tables. Joanie

Breeze typed earlier drafts. Susan Ottersen and Steve Roguski revised

and improved figures and maps. R. Berger explained water level

recorders. The staff at CIRCA solved computer problems.




















TABLE OF CONTENTS


Page


ACKNOWLEDGMENT . . ... . ii

LIST OF TABLES . . .. ... vii

LIST OF FIGURES. . . ... . ix

ABSTRACT . ... . xiii

INTRODUCTION . . ... .. 1

A Systems View . . ... 2
Reclamation of Clay Settling Areas . 5
Previous Studies about Succession in Clay Settling Ponds 7


Basic Questions on Succession .
Succession Theory .
Wetland Succession .
Factors Controlling Wetland Species .
Hydrologic Regimes .
Fire . .
Biotic Disturbances .
Seed Dispersal . .
Light Availability .
Physical Soil Properties .
Chemical Soil Properties .
Wetland Organization at the Landscape
Self-Organization of a New Climax .
Description of Study Sites .
Vegetation Survey .
Cypress-Gum Seedling Transects .
Hydric Swamp Plots .


. .






Level

* .
* .


METHODS . . . .


Community Structure Measurements .
Vegetation Analysis .
Wetland Site Index .
Cypress-Gum Seedling Transects .
Seedling Plots .
Grazing Experiments .
Water Levels and Hydroperiod .
Monthly Rainfall Data .
Soil Measurements .


~
'rE
~
r









Statistical Evaluation . .
Hydric Swamp Seedling Plots . .
Plot Description . .
Mulch Treatments . .
Hydrology Measurements . .
Soil Analysis . .
Direct Seeding . .
Statistics . . .


RESULTS . . . .

Successional Sequences for Clay Pond Colonization .
Vegetation Organized by Soil Moisture . .
Trees Organized by Seed Availability . .
Seedlings for Enhancing Cypress-Gum Succession .
Patterns in Water Level Fluctuations . .
Rainfall Compared to Normal . .
Soil Characteristics . . .
Seedling Survival and Growth . .
Baldcypress Performance and Water Level . .
Moisture Gradient and Survival . .
Comparison of Tubelings and Bare Root Seedlings ..
Comparison of Cleared and Uncleared Seedling Transects
Effect of Cattle Grazing . . .
Comparison Between Pondcypress and Baldcypress .
Hydric Swamp Species for Reclamation . .
Average Water Table Depth . .
Growth and Survival . . .
Seedling Survival vs. Water Table Depth . .
Effect of Existing Vegetation . .
Tree Success and pH Values . .
Seedling Success in Clay Ponds Compared to Natural Systems
Mulch Treatments .. . . .
Direct Seeding . . .

SIMULATION OF WETLAND SUCCESSION . . .


System Diagram . .
Explanatory Narrative . .
Simulation Runs . .
Simulation Compared to Reality .


DISCUSSION . . . .

Ecosystem Development and Natural Succession .
Arrested Willow Succession .... .. .. .
Hydric Hardwood Succession . ... ..
Enhancing Cypress-Gum Succession ... .
Survival of Planted Cypress-Gum Seedlings .
Taxonomic Variation . . .
Which Wetland Type for Reclamation . .
Patterns of Survival for Planted Swamp Seedlings .
Predict Community Type from Substrate Characteristics .
Factors Influencing Seedling Success . .


67

67
67
78
83
83
90
90
105
108
116
121
127
127
127
133
133
133
142
142
146
146
148
148

155


. 155
. 160
. 162
. 165


167

167
168
170
176
177
179
181
182
183
187










Effects of Water Table . .
Effect of Competition . .
Nursery Stock Type . . .
Soil Amendments . . .
Cattle Grazing . . .
Economics of Reclamation . .


CONCLUSIONS

APPENDIX A

APPENDIX B

APPENDIX C


APPENDIX D



APPENDIX E



APPENDIX F


EXPLANATION OF SYMBOLS USED IN ENERGY DIAGRAMS .

ABBREVIATIONS AND SCIENTIFIC NAMES USED IN TEXT .

REGRESSION GRAPHS FOR EACH SPECIES IN EACH SITE FOR
CYPRESS-GUM TRANSECTS. . .

TREE SURVIVAL AND PERCENT GROWTH FOR EACH SPECIES
PLANTED IN HYDRIC PLOTS COMPARED TO AVERAGE DEPTH
OF WATER TABLE IN 1986 . .

TREE SURVIVAL AND PERCENT GROWTH FOR SOME SPECIES
PLANTED IN HYDRIC SEEDLING PLOTS COMPARED TO SOIL
pH (MEASURED IN WATER) . .

COMPUTER PROGRAM FOR SIMULATIONMODEL WRITTEN IN BASIC
FOR AN IBM PC COMPATIBLE COMPUTER. . .


REFERENCES . . . .

BIOGRAPHICAL SKETCH . . .


196

199

200


205



230



242


249

253

267


.JL 9
















LIST OF TABLES


Table Page

1 Summary information about study sites for cypress-gum transects .35

2 Example of wetland site index calculation .... .53

3 Tree species planted in experimental plots on clay settling
ponds listed according to moisture categories ... .62

4 Quadrats grouped by herbaceous wetland site index number
comparing tree species for moisture classes in clay settling
ponds 10 to 20 years since active use . .... .68

5 Quadrats grouped by herbaceous wetland site index number
comparing tree species for moisture classes in clay ponds over
20 years since active use . .... .70

6 Importance values for herbaceous vegetation used in the method
to delineate moisture classes for site 10 to 20 years old .74

7 Importance values for herbaceous vegetation used in the method
to delineate moisture classes for sites over 25 years old .76

8 Soil analysis for particle size and percent moisture in
cypress-gum seedling transects . .... .93

9 Soil analysis for organic matter, pH, and selected double acid
extractable nutrients in cypress-gum seedling transects ... .99

10 Statistical information from the Analysis of Variance used to
evaluate regressions between seedlings planted as tubelings
after one year's growth compared to average water depth .. 119

11 Statistical information from the Analysis of Variance used to
evaluate regressions between seedlings planted as tubelings
after one year's growth compared to average water depth for
1986. . ..... ........ ... 120

12 Comparison of growth and survival after theJirst year of
growth for bareroot seedlings with water table depths divided
into moisture classes . ...... 122

13 Growth and survival after the first year of growth for
tubelings compared to water table depths divided into moisture
classes . . . 125

vii










14 Statistical information for transects where cows were allowed
to graze compared to those where cows were excluded located at
Mobil pasture pond . . .. 129

15 Statistical information comparing survival, height, and growth
of planted baldcypress with planted pondcypress ... 131

16 Growth and survival of seedlings planted in experimental plots
measured in November 1986 eight months after planting .... .138

17 Summary table for average growth and survival of eleven hydric
hardwood species planted in six clay settling ponds .. 141

18 Seedling survival and percent growth arranged by depth to
average water table . .. .. 143

19 Growth and survival of tree seedlings compared to type of
vegetation cover existing in the plots at time of planting 145

20 Seedling survival and percent growth compared to pH .. 148

21 Survival and growth of planted hydric swamp seedlings compared
to type of mulch treatments . .... 149

22 Number of seeds germinating from litter transferred from
wetland forest to clay settling pond seedling plots .. 150

23 Direct seeding of hydric swamp species into clay settling ponds
and as garden experiments compares mulch treatments and water
regimes . ... ...... ... 153

24 Equations for the ecosystem type model in Figure 1 .. .157

25 Values and flows and storage used in initial calibration 158

26 Wetland forest communities on the south prong of Alafia River 172

27 Importance values for trees found in the floodplains of small
stream bottoms in Florida compared to trees colonizing clay
settling ponds . . ... .... 173

28 Wetland forest communities of north-central Florida compared to
vegetation colonizing clay settling ponds . .. 174

29 Distributional patterns for wetland tree species in Florida
compared to trees colonizing clay settling ponds ... 175

30 Soil analysis for nutrients and pH in Florida wetlapds compared
to clay settling ponds . .. . 184

31 Cost of tree seedlings using several methods ... .194


viii
















LIST OF FIGURES


Figure Page

1 Energy diagram of factors affecting succession on clay settling
ponds . . . 3

2 Energy diagrams comparing the mechanisms of succession for the
three models described by Connell and Slatyer (1977). .14

3 Location of study sites . . ... 32

4 Location of study sites at Tenoroc Zone 4A. .. .36

5 Location of study sites at Gardinier Ft. Meade complex, Area A. .38

6 Location of study sites at Gardinier's 0. H. Wright site. .39

7 Sketch showing locations of tree seedling transects at Alderman
Ford Ranch. . .. .. 41

8 Sketch showing location of tree seedling transects at IMC-H9. .43

9 Sketch showing location of bareroot seedling transects a) CF
Industries, b) Mobil pasture pond at Homeland.. ... .44

10 Sketch showing the location of seedling plots at Pruitt Ranch.. .46

11 Sketch showing location of seedling plots at IMC Peace River
Park. . . . .48

12 Sketch showing location of control plots on the Peace River
floodplain . .. .. ..49

13 Sketch of control plots at Camp Meeting Ground Creek. .50

14 Design for paired transects used in cypress-gum seedling
experiments . . .. .. .56

15 Design for hydric swamp experimental seedling plots .. ..61

16 Design for set of plots used in direct seeding experiments. 65

17 Tree dominance shown for moisture classes in younger and older
clay settling ponds . . .. .. 72









18 Hardwood tree species measured at Alderman Ford Ranch clay
settling pond . . .. 79

19 Hardwood tree species measured at Maine Ave. clay settling pond .80

20 Hardwood tree species measured at 0. H. Wright clay settling
pond . . . ..... .81

21 Hardwood tree species measured at a clay settling pond adjacent
to the south side of Lake Hancock . ... 82

22 Water levels measured in wells at Tenoroc State Reserve for
1985-86 growing season . . .. .. 84

23 Water levels measured in wells during the 1985-86 growing
season . . . 85

24 Water levels measured in wells at Alderman Ford Ranch during
1985-86 . . . .86

25 Water levels measured at a reference marker in ponds during
1985-86 growing season . . .. 87

26 Water levels measured at a reference marker in a pasture pond
owned by Mobil Mining and Minerals . .. ..88

27 Monthly rainfall data during the time seedlings were planted
and seeds introduced at a) Gardinier (area A) and 0. H. Wright
and b) Tenoroc. . . 91

28 Monthly rainfall data during the period when seedlings were
planted at a) CF Industries and b) IMC-H9 and Mobil pasture
pond . . . .. 92

29 Comparison of particle size from top 15 cm of soil column for 7
clay settling ponds used for cypress-gum seedling transplant
experiments . . .. 95

30 Regression analysis between water table depth and soil moisture
content a) all sites b) sites with no sand additions. .. .. 97

31 Soil moisture and percent clay for a) all sites and b) younger
sites.. . . . .98

32 Comparison of pH measured in water . .. .102

33 Comparison of average percent organic matter between 7 clay
settling ponds used for cypress-gum seedling transplant
experiments . . 103

34 Comparison of potassium in ppm between 7 clay settling ponds
used for cypress-gum seedling transplant experiments ... 104









35 Comparison of height and survival after one year of growth for
bareroot seedlings planted in clay settling ponds .. 106

36 Comparison of height and survival after one year of growth for
seedlings planted as tubelings in traditional clay settling
ponds . . . 107

37 Data for baldcypress seedlings planted at Gardinier, Area A.
a) An average cross section of 25 transects showing water table
levels for the 1985-86 growing season and seedling height for
1985 and 1986; b) Average survival of seedlings for 25
transects for 1985 after a fire and 1986 after a flood. ... 109

38 Data for baldcypress seedlings planted at IMC-H9. a) An
average cross section of 3 transects planted along a lakeshore
showing water table levels for the 1985-86 growing season and
seedling height for 1985 and 1986. b) Average survival of
seedlings in 3 transects for 1986 . .... 110

39 Data for baldcypress seedlings planted at Tenoroc, Zone 4A,
along a lakeshore. a) An average cross section of 6 transects
showing water table levels for the 1985-86 growing season and
seedling height for 1986. b) Average survival of seedlings in
6 transects.. . . ... 111

40 Data for baldcypress seedlings planted at Tenoroc, Zone 4A, dry
sites. a) An average cross section of 8 transects showing
water table levels for the 1985-86 growing season and seedling
height for 1985 and 1986. b) Average survival of seedlings in
8 transects.. . . ... ....... 112

41 Data for baldcypress seedlings planted at Gardinier's 0. H.
Wright site. a) An average cross section of 6 transects
showing water table levels for the 1985-86 growing season and
seedling height for 1985 and 1986. b) Average survival of
seedlings in 6 transects. . . ... 113

42 Data for baldcypress seedlings planted at CF Industries, SP-1.
a) An average cross section of 5 transects showing water table
levels for 1985-86 growing season and seedling height for 1986.
b) Average survival of seedlings in 5 transects.. .. 114

43 Data for baldcypress seedlings planted at Mobil pasture pond.
a) An average cross section of 8 transects which includes 4
grazed by cows showing water table levels for 1985-86 growing
season and seedling height for 1986. b) Average survival of
seedlings in 8 transects. . .... 115

44 Baldcypress survival and height compared to averap.water table
depth for 1986 shown for each clay settling pondp? moisture
class . . . .. 117

45 Baldcypress bareroot seedlings planted in the spring compared
to tubelings planted during the winter . .... 126










46 Baldcypress showed no significant difference in survival or
growth in plots cleared of all above-ground vegetation at time
of planting compared to those left uncleared ... 128

47 Comparison of seedlings planted at Mobil pasture pond where
cows were allowed to graze and those where cows were excluded 130

48 Survival and height of baldcypress (Taxodium distichum)
compared to pondcypress (Taxodium ascendens) planted as
seedlings in the same transects . .. 132

49 Monthly water table measurements for 1985-86 in hydric swamp
seedling plots at Gardinier and Tenoroc.. . .. 134

50 Monthly water table measurements for 1986 in hydric swamp
experimental seedling plots located in clay settling ponds at
Agrico and 0. H. Wright . . 135

51 Monthly water table measurements for 1986 in hydric swamp
experimental seedlings plots located in clay settling ponds at
IMC-Peace River Park and Pruitt Ranch .. .136

52 Monthly water table measurements for 1986 in hydric swamp
experimental seedling plots located at two control sites, Peace
River floodplain in climax vegetation, and Camp Meeting Ground
Creek in early successional myrtles and willows ... 137

53 Simulation model for hydrologic control of ecosystem type 156

54 Simulations of the model of ecosystem type for wet management
alternatives . . . 163

55 Simulation runs where water budget maintains water level
slightly below ground surface. . .164


xii















Abstract of Dissertation Presented to the Graduate School of
the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



WETLAND RECLAMATION BY ACCELERATING SUCCESSION

by

BETTY TOOMBS RUSHTON

April 1988

Chairman: Howard T. Odum
Major Department: Environmental Engineering Science

This research analyzed mechanisms and processes for accelerating

natural succession in order to restore soils and forests on clay settling

areas left from phosphate mining in central Florida. Field measurements

of succession on unreclaimed clay ponds showed wet sites dominated by

dense stands of small shrubby willows even after 60 years with succession

arrested because of a shortage of seeds for later stage trees. For drier

sites an orderly procession of pioneer wetland trees colonized when

wetland seed sources were within 200 meters. The first woody species

were willows, myrtles, and baccharis followed in 5 to 10 years by red

maple and elm. Oaks colonized slightly drier elevations. Hackberry,

cherry, and sweetgum were also found.

Experiments in which 3000 seedlings of 11 species were planted in

six clay settling areas demonstrated succession can be accelerated.

After the first growing season, results suggest thatpixed -iamp

vegetation typical of floodplains may be the most sul'able forested

wetland community for settling pond reclamation. Percent survival was
xiii








best for Carolina ash (98%), American elm (91%), and red maple (80%).

Some alluvial floodplain species were intermediate in success with 74%

survival for baldcypress, 61% for sweetgum, and 61% for laurel oak.

Trees from bayheads had the least survival with 52% for swampbay and 41%

for loblolly bay. Poorest survival for all species planted (39%) was

swamp tupelo. Floodplain species which require fairly dry conditions had

poor survival, i.e., southern magnolia (53%) and cabbage palm (43%).

A drought during the winter of 1984-85 caused 46% mortality for

baldcypress compared to 26% for baldcypress planted one year later. Plots

cleared of all above-ground vegetation were not significantly different

in percent survival or height of trees compared to those left uncleared

(P > 0.05). No significant differences (P > 0.05) in survival or growth

were seen for seedlings planted with hay or forest litter compared to

plots with no mulch treatments. Tubelings planted in adjacent rows

usually showed better survival for baldcypress than for pondcypress (P <

0.001). One site showed no significant difference. Planted tree

seedlings were more cost effective than placing seeds on the ground and

covering them with litter.

A simulation model with hydrologic regimes and outside seeding was

used to summarize the operation of the successional system. Simulation

that suggested trends for a longer time period than those observed in the

field trials are yet to be confirmed.


xiv














INTRODUCTION

The processes of natural succession may be the fastest, economical

way to restore soils and vegetation to a disturbed landscape.

Consequently, understanding and managing succession may be a practical

way to restore lands to useful functions after mining. This dissertation

contains field analyses of succession on clay settling areas, a landform

left by phosphate mining in central Florida. The major goals were to

understand vegetation colonization on clay settling ponds, to study

accelerated natural succession by planting tree seedlings, and to

determine the consistency of theories and data on forest succession with

a computer simulation model.

The underlying philosophy is to tailor natural ecosystem development

to accelerate forest reclamation. Procedures are developed to facilitate

the self-organizational processes of ecological succession in order to

regenerate vegetation and soils at low cost to the human economy.

Apparently, lack of adequate seeding is retarding succession in many

mined areas. Unless an adjacent forest was present, development has not

proceeded beyond an initial stage of herbaceous and shrub vegetation

confined to species which colonize by tiny, wind dispersed seed. Planted

seedlings may accelerate the normally slow process of primary succession

on phosphate mined land in central Florida.

Tree species planted in experimental plots included those founltin

wetland situations in Florida ranging from cypress d Ls and bayheads to

alluvial rivers and streams. Planted seedling plots *ere used to test









the effects of water table and moisture regimes, competition from

existing vegetation, grazing from cattle, mulch from floodplain forests,

and nursery seedling types.

Field studies were divided into three major parts. One consisted of

measuring existing vegetation found on clay settling ponds to better

understand natural succession. The second involved planted tree species

common to cypress forests in wet depressions to test the theory this was

a suitable community type for wet conditions in clay settling ponds. The

third included testing the suitability of a wider range of wetland tree

species common in Florida to determine if alluvial floodplain species or

trees found in cypress domes or bayheads would be the best choice to

plant for reclamation.



A Systems View


A systems diagram (Figure 1) shows some main features of a wetland

ecosystem with factors affecting succession on clay settling ponds. The

producer symbols represent mature wetland forests and the early

successional stage. The mature forest suggests the vegetation

communities which might result from wetlands succession on clay settling

ponds. The diagram uses systems language developed by H. T. Odum (1971,

1983). The convention of putting energy sources in order of increasing

quality from the left to right side of the diagram helps organize

interactions and makes it easier to understand. Symbols and their

meaning are included in Appendix A. Processes which e cA~sidered

important in directing vegetation colonization incliff~water level (which

controls soil oxygen), litter recycling, seed source and dispersal, fire,

grazing herbivores, soil components, and human intervention.









3












'o









0
00
5 r ,I, \











4-4
*0o





aa
402

-1b














C w C /
\ B / z s /













b o





S00







\ (\'a












1 \ ( a $
00 \\ Ul~loE a.









When vegetation colonizes bare ground it uses the energy of

sunlight, water, and nutrients to organize available adapted seeds into

structure and biomass. As time passes the system grows a forest with

larger biomass, metabolism, and diversity where many species are

replaced. The more developed forest is then called a climax. For

Florida the southern mixed evergreen hardwood forest is often climax in

ordinary moisture regimes (Quarterman and Keever 1962). For wetlands

several climax associations exist including bay swamps, cypress forests,

and hydric hardwood swamps. Changing environmental conditions, internal

consumer oscillations, and catastrophic events maintain the diversity of

vegetation types in the landscape. Floods, wind storms, fires, and

grazing herbivores may provide a pulse to the system that uses

destructive energy sources to facilitate productivity and self-

organization (Odum 1982, Campbell 1984). For example, along the southern

coastal plain, pine flatwoods, the dominant forest type, are maintained

by fire which kills mixed hardwood species. Other communities also adapt

to fire frequency and intensity. Cypress is maintained by low intensity

fires (Gunderson 1984, Duever et al. 1984) and sand pine by infrequent

fires (Monk 1968, Veno 1976, Peroni and Abrahamson 1986).

Human activity in reclamation projects can have positive or negative

effects. Often man sets back succession by tearing down the ecosystem

that nature has constructed. Kangas (1983) calculated that after 25

years more restoration work had accumulated through succession on spoil

piles left from phosphate mining than mechanical recl iop would

provide. Landscape diversity such as mounds and d sios in a

phosphate reclamation project, where peat and litter were added,

increased plant diversity even in the early stages of succession (Brown







5

and Odum 1985). An open question is whether large machine reclamation or

ecological methods are best. Large machines often eliminate uneven

topography and make a smooth homogeneous landscape. They are also an

expensive alternative for reclamation, costing from $300/acre for pasture

to $4,374/acre for woodlands on coal mining land (Kangas 1983).

Ecological engineering techniques for speeding up succession may use less

energy, depending on natural sources to produce large effects in

ecosystem patterns and processes (Odum 1971, Kangas 1983).



Reclamation of Clay Settling Areas


Clay settling ponds are a by-product of phosphate mining, a major

industry in central Florida. Typically one ton of clay waste (dry

weight) is produced for each ton of phosphate rock. The volume of clay

waste is many times greater than the original volume of clay and requires

large above-ground storage impoundments ranging from 80 to 400 ha

surrounded by earthen dams from 6 to 18 m in height (Haynes 1984).

Approximately 60 to 75% of the land proposed for mining in Florida is

designated for clay settling areas. Over 30,000 ha of settling

impoundments have already been built in central Florida with 1,000 ha of

new ponds constructed each year (Pittman et al. 1984). Since phosphatic

clays have poor load bearing capacity, possibilities for productive use

following mining are limited. Most reclamation projects have converted

clay ponds to pasture. The average settling area has a holding capacity

to receive by-product clays for approximately two y bu t-are used

intermittently for longer periods as clays subsidujL pore storage

becomes available. These areas may remain in service as water reservoirs
for the life of the mine (Yon 1983). Clays are deposited at 2 to 5%









solids, after several months of consolidation under their own weight,

they reach approximately 15% solids; and after several years about 25%.

Once decommissioned, vegetation develops rapidly with islands of cattail

(Typha latifolia) established in ponds still in the filling stage. When

the surface begins to dry, woody species invade dominated by willow

(Salix caroliniana) sometimes in association with groundsel-tree

(Baccharis halimifolia) and wax myrtle (Myrica cerifera).

Florida dominates phosphate mining in the U.S. in 1986 and over

66,800 ha have been or soon will be disturbed by mining (Marion 1986).

In 1987 reclamation was required for all current phosphate mining

activity. State laws under the jurisdiction of the Florida Department of

Natural Resources (Chapter 16C-16, Florida Statutes) mandate that all

lands mined after 1975 must be reclaimed. One goal of the legislation is

to develop a revegetation plan to minimize soil erosion, conceal the

effects of surface mining, and recognize the requirements for appropriate

habitat for fish and wildlife. The plan encourages special programs to

restore, enhance, or reclaim particular habitats lost as a result of

mining activities. It requires restoration of wetlands affected by

mining operations to cover at least the same amount of land as before

mining.

In addition, the Non-Mandatory Land Reclamation Trust Fund will use

monies derived from the severance tax for reclamation of nearly 36,423

additional ha mined before 1975 (Chapter 16C-17, Florida Statutes). Of

these, 18,616 ha were devoted to active and inactive clay settling ponds

in 1975 (Bromwell and Oxford 1980). The uneven top hy .and weedy

appearance of the natural vegetation make most abandoned clay settling










areas eligible for reclamation (Zellar-Williams and Conservation

Consultants 1980).

In 1987, the Department of Environmental Regulation drafted a rule

(Chapter 17-12, Florida Statutes) to mitigate adverse impacts of dredging

and filling in waters of the state. It makes knowledge about wetland

restoration an important aspect of the permitting process for phosphate

mining. The draft rule requires a certain percentage of target species

or a particular number of plants/acre. These tentative rules also

require knowledge about soils, hydrologic regimes, and a ten year plan

for any mitigation involving a mining area.

The Southwest Florida Water Management District reviews impacts on

wetlands not under the Department of Environmental Regulation

jurisdiction. For phosphate mining the District will review plans on a

case-by-case basis, balancing losses against the ability of the applicant

to mitigate those losses. Failure of the applicant to provide reasonable

assurance will mean the wetlands or environmentally sensitive areas will

not be mined.

County laws and ordinances are varied. Hamilton and Polk counties

are fairly lenient, while those in Hillsborough and Manatee counties are

more stringent. Marion (1986) described them as the most comprehensive

local regulations of any in the Country. He also observed that knowledge

about revegetation potentials of natural processes is needed to improve

management guidelines.



Previous Studies about Succession in Clay S ing'Ponds


The first vegetation studies of phosphate clay settling ponds, done

in the early 1980s, focused on descriptive natural history which reported







8

an initial cover of cattails (Typha sp.) and water hyacinths (Eichhornia

crassipes) followed by primrose willow (Ludwigia peruviana) and willow

(Salix caroliniana). Wax myrtle (Myrica cerifera) and a profusion of

vines including virgin's bower (Clematis virginiana), fern vine (Lygodium

japonicum), and climbing hemp vine (Mikania scandens) were found as sites

continued to dry (Zellar-Williams and Conservation Consultants 1980, King

et al. 1980, Schnoes and Humphrey 1980, Butner and Best 1981, Gilbert et

al. 1981, Rushton 1983). In some cases where a seed source was nearby,

sites had been invaded by hardwood species such as red maple (Acer

rubrum) and laurel oak (Quercus laurifolia) (Zellar-Williams and

Conservation Consultants 1980, Rushton 1983).

Several projects where hardwood tree species have been planted on

clay settling ponds show mixed results. Estech General Chemicals capped

a consolidated settling area with tailing sand in 1979. All of the five

spaded trees of red maple (Acer rubrum) and sweet bay (Magnolia

virginiana) survived while potted trees had 15% to 20% survival (Ruesch

1983). International Minerals and Chemical Corp. reclaimed a clay

settling area by partially capping the edges where they planted eight

hydric and mesic tree species with excellent results (Goodrich 1983).

Trees are healthy and 3 to 6 m tall 6 years later in 1987.

The Florida Division of Forestry has planted seedlings each year

from December 1980 to March 1984 using four different post-mining soil

types (Harrell in press). Fifty percent survival at the end of one

growing season was the principal reclamation standard used to recommend

tree species for planting since it was adequate to therequirements

of The Florida Department of Natural Resources, Bureau of Land

Reclamation. Sand-clay mix soils had a mean survival from 63 to 100% for






9

19 hardwood and softwood species. On phosphatic clay plots survival was

less, 33 to 66% for 14 tree species.

The following are forest plantings on clay settling ponds not

reported in the literature. Agrico planted trees several years ago at

their Ft. Greene Mine which are now robust and 2 to 3 m tall (Selwyn

Presnell, personal communication). They had 100% survival with cypress,

sweetgum, and ash which were potted seedlings. Oaks showed 92% survival

and blackgum 67%. Cypress had 75% survival when planted as bareroot

seedlings. No slash pine survived. Fertilizer treatments appeared to

have no effect on growth or survival. Cypress trees planted on a

sand-capped pasture ponds at Mobil were also doing well. Cottonwoods

planted on a clay pond in 1979 by Mobil have a closed canopy in 1987.

A review of conceptual plans approved by The Department of Natural

Resources since 1980 shows there will be a significant decrease in upland

forest and native rangeland and an increase in man-managed systems,

primarily improved pasture, which require high maintenance costs (Gilbert

et al. 1985). They further state research is needed to evaluate whether

forest habitat types can be reestablished, especially those native

woodlands which are essential to both game and non-game species. The

U.S. Fish and Wildlife Service (Haynes 1984) cites the reestablishment of

forested wetlands as one of the most neglected information gaps of

importance to surface-mining industries. Increased concern about wetland

losses has resulted in more stringent regulatory requirements for

reclamation.

Delay in reclaiming clay settling ponds becaus-e their expanded

volume and poor load-bearing strength has encouraged research to

determine methods for rapid dewatering and below grade disposal (Carrier








1982, Garlanger 1982, McLendon et al. 1983, Pittman and Sweeney 1983).

Disposal systems which involve mixing of sand with clay materials seem to

produce a better soil for forestry and agriculture (Marion 1986).

However, sand spray reclamation utilizes only 40% of the mining clay

waste; conventional settling areas are still required to store the

remaining 60% (McLendon et al. 1983).



Basic Questions on Succession


This study also pertains to principles of ecology concerned with

succession and self-organization of systems. The following contains a

brief review of viewpoints on wetland succession and discusses questions

which are related to this research.


Succession Theory

In some views of succession (Clement 1936, Odum 1969), wetlands were

considered a transient stage between shallow waters and terrestrial

forest climax communities. By this theory waters gradually fill in from

sediment deposition and accumulation of organic peat material from dead

plants. Shrubs and small trees appear. They continue to transform the

site to a terrestrial one, not only by adding organic matter, but also by

drying the soils through evapotranspiration. Where sediment raises land

above water levels, a change to drier vegetation is observed.

In many Florida locations, however, inorganic sediments are not

being added and organic deposition does not proceed beyond water levels

(Odum 1984b). Instead organic matter oxidizes or bur'n in dry weather so

that succession may be arrested. Wetland ecosystem in this situation

are a form of climax.








Vegetation changes are also regulated by subtle environmental

gradients especially water regimes. The species composition may depend

upon dispersal of seeds for those plants which are adapted to the

specific conditions (Gleason 1926, Whittaker 1953, McIntosh 1980). Since

each species has its own adaptation, no two occupy exactly the same

niche. This results in a continuum of overlapping sets of species, each

responding to environmental cues.

In general, succession progresses from simple systems with species

of small stature and short life spans toward organized systems with high

complexity and information content (Odum 1969). In mature communities

trees are larger, have more biomass, and retain more nutrients. In the

more mature system, energy is used more efficiently for maintenance of

forest of large structure with high gross production (Odum and Pinkerton

1955). What are the roles of early successional species? Do they

prepare the way for slow growing trees typical of a climax forest? Or

instead, are they an impediment to succession?

Connell and Slatyer (1977) describe three models of mechanisms for

forest establishment on bare soils. They assume no further changes in

the abiotic environment and that certain species usually appear first

because they have the ability to produce large numbers of easily

dispersed seeds, which are not adapted to germinating and growing on

occupied sites. The models differ in the mechanisms that determine the

methods in which new species appear. In model 1, early-succession

colonizers modify the environment so that later-succession species are

able to invade and grow to maturity. This is the "relay floristic" model

of Egler (1954). In models 2 and 3, any arriving pe'cies may be able to

colonize. This concept was identified by Egler (1954) as his "initial







12

floristics" model. In model 2 (Connell and Slatyer 1977) early colonists

neither increased nor decreased the rate of recruitment and growth of

later successional species. Trees that appear later were those that

arrived, either at the very beginning or later, and then grew slowly.

Seedlings survived and grew despite the presence of early successional

species because they were able to tolerate lower levels of resources. In

contrast, the third model postulated the detrimental effect of early

successional species which, once established, inhibited the invasion of

subsequent colonists or suppressed the growth of those already present.

In model 3, the replacement of species during succession requires death

of the early colonists.

Energy diagrams (Figure 2) show the interaction between early and

late successional species for the three models described by Connell and

Slatyer (1977). Model 1 has early successional species improving the

resource cycle stream so later successional species are able to colonize.

In this model mature species are unable to become established on bare

ground. Competition for sunlight results in poor reproduction of fast

growing early successional species, once conditions favor the invasion of

seeds and their germination for more mature species. Model 2 shows both

early and late successional species coexisting with the same resources.

However, the early vegetation modifies the environment so that it is less

suitable for its own reproduction, but has no effect on recruitment of

more mature forest. Competition from existing forest neither encourages

nor retards succession and more mature trees grow in the presence of

healthy early successional individuals. In model 3 tLe two community

types compete for a common energy source, sunlight, Where trees divide

the resource according to their biomass. Competition for nutrients and


























Figure 2. Energy diagrams comparing the mechanisms of succession for the
three models described by Connell and Slatyer (1977).

a) Model 1. Early successional colonizers modify the
environment so that later successional species are able to
colonize and grow to maturity.

b) Model 2. Early colonist neither increase nor decrease the
success of later successional species. Later species have
poorer seed dispersal and are slower growing but can grow
at lower levels of resources than early ones.

c) Early successional species modify the site so that it is
less suitable for seedlings of both early and late
successional species. Later species appear later because
they live longer and gradually replace earlier ones or
early ones persist in arrested succession.







14


Seeds
Soil Microbes
RNutrients
S T \Moisture


Late


Microbes
Seeds
Sun Early
S ,ast growth
short Wie


MODEL 1 2.

(a)




Seeds
Soil Microbes
-Nutrients
Moisture

Late v6
low growt
Wind) long life \Litte- Animals

Microbes
Early Seeds M
tshon life



MODEL 2 _L

(b)




Seeds
Soil Microbes
V+ i Nutrients



Sowrowt Litter

Microbes
Early Seeds
fast growt
Short life



MODEL 3

(c)








moisture cycled within the system is based on resource dilution by the

early successional vegetation which was able to become established first,

capture the resources, while the slower growing late successional trees

can only compete successfully once the short lived early invaders have

died.


Wetland Succession

Since disturbance is so common and frequent on the Coastal Plain of

the southeastern United States, it has been assumed that most wetland

forests have been in recent succession (Quarterman and Keever 1962).

Even though some wetland peat deposits have radiocarbon dates of more

than 1000 years (Odum 1984b), wetland forests were considered subclimax

communities on soils with standing water during at least part of the

year. In this context subclimax meant that vegetation had reached

maximum development so long as water conditions were maintained.

In a discussion of swamp plant succession Penfound (1952) described

the successional sequence in shallow waters and wetlands as proceeding

from a floating aquatic to marsh stage followed later by willow forest

(Davis 1937, Weaver and Clements 1938, Penfound and Hathaway 1938).

Depth of water table was cited as a controlling factor. Special

conditions caused other sequences, such as presence of floating mats

(Penfound 1952) or abrasive water flow (Brown and Montz 1986).

In the development of wetland hardwood forest along southern

alluvial floodplains the stages that follow willow on fine textured soils

are baldcypress (Taxodium distichum) and ash (Fraxinus app.) sometimes in

association with American elm (Ulmus americana), red maple (Acer rubrum),

water oak (Quercus nigra), sweetgum (Liquidambar styraciflua), tupelos

(Nyssa spp.), cabbage palm (Sabal palmetto), and sugarberry (Celtis









laevigata) followed much later by overcup oak (Quercus lyrata), water

hickory (Carya aquatica), southern magnolia (Magnolia grandiflora), and

persimmon (Diospyros virginiana) (Putnam et al. 1960). A similar

succession occurs as sloughs, swamps and oxbow lakes fill with sediments.

Where sediment inflow is large, however, succession is much slower

(Putnam et al. 1960).

On the southern coastal plain, swamps, not of alluvial origin, are

caused by impoundment of rainwater and seepage, which develops peat in

landlocked depressions (Putnam et al. 1960). These low nutrient swamps

support pondcypress (Taxodium ascendens) separately or together with

blackgum (Nyssa sylvatica var. biflora) and/or bays (Magnolia virginiana,

Persea palustris, and Gordonia lasianthus). Atlantic white cedar

(Chamaecyparis thyoides) is also found but is now rare in the south.



Factors Controlling Wetland Species


The main factors and interactions controlling the rate of succession

as well as the processes which select for specific wetland types are

shown in Figure 1 and discussed below. Wetlands in central Florida

include cypress swamps (two types baldcypress and pondcypress), bayheads,

and mixed hardwood swamps (Monk 1968). Severe conditions also cause

retrogression to earlier successional stages. In the case of wetlands,

major types are willow forest and herbaceous or perennial marshes.


Hydrologic Regimes

Water regimes exert a major influence on tree colonization.

Flooding and uneven topography create the diversity and complexity of

bottomland hardwood associations (Wharton et al. 1982). The type of








water flow from stagnant to abrasive flooding creates vegetation types

from still-water cypress domes and bay heads to alluvial river forest and

lake shores.

Dissolved oxygen concentration. Stagnant, poorly drained sites are

described as a stress to the system. For example, baldcypress (Taxodium

distichum) showed two to five times greater growth and weight increases

in moving water compared to stagnant water (Hook et al. 1970). Dickson

and Broyer (1972) found tupelo (Nyssa aquatica) root systems were more

damaged in anaerobic saturated soils than baldcypress. Both species,

however, failed to thrive under anaerobic conditions.

Similar results were seen in field studies where baldcypress,

pondcypress, and black tupelo (Nyssa sylvatica var. sylvatica) responded

differently to the addition of secondarily treated sewage. The study

suggested that the initial difficulty with early growth and establishment

as well as long-term growth of baldcypress and black tupelo in both

sewage domes was probably caused by low oxygen concentrations (Deghi

1978). In contrast, mortality of planted pondcypress seedlings was

greater in sewage domes, but growth of seedlings was better possibly

stimulated by increased nutrient input.

Depth of flooding. Deeply flooded lands as well as the high energy

systems described above more often support baldcypress and water tupelo

while less frequently flooded sites grade into various swamp hardwood

associations. A mean depth of 60 cm appears to be the threshold of

flooding a mixed hardwood swamp forest can tolerate before significant

changes in complexity of the forest occurs (Brown and Lugo 1982). The

majority of mixed hardwood swamp species will not survive two years of

continuous flooding (Broadfoot and Williston 1973). Bayhead communities









are also restricted to wet areas with minimum flooding (Monk 1966).

Bayheads are typical of seepage areas with steady increments of water

that fluctuate much less widely than cypress ponds (less than 33 cm vs

133 cm) (Wharton et al. 1977). The most deeply flooded sites favor

baldcypress (Taxodium distichum) and water tupelo (Nyssa aquatica).

Baldcypress has a critical depth of 3 to 6 m depending on duration of

flooding (Brown and Lugo 1982).

Pondcypress (Taxodium ascendens) show their best growth in the

center of ponds with depths from 30 to 130 cm. Domes become

progressively more shallow toward the edge where blackgum, bays and other

swamp hardwoods species also occur (Monk and Brown 1965). Blackgum

(Nyssa sylvatica var. biflora) tolerates frequent inundations to a depth

of 3 m or more (Hall and Penfound 1939). Although blackgum is found on

the banks of slackwater swamps and ponds, it attains its highest growth

rate in better aerated coves and seepage zones which remain wet year-

round (Fowells 1965).

Baldcypress appears more sensitive to dry conditions than some other

species. When exposed to drought, the succulent shoots of cypress

progressed from slightly wilted to complete collapse and irreparable

damage in 3 to 4 hours, whereas the shoots of similarly treated blackgum

plants would wilt, but recover following watering (Dickson and Broyer

1972).

In experiments designed to test moisture tolerance of swamp hardwood

species, seedlings were classified according to their tolerance to

saturated soil conditions as follows: tolerant--green ash, pumpkin ash,

water tupelo, and willow; intermediate--cottonwood, boxelder, red maple,

silver maple, pin oak, and sycamore; intolerant--shumard oak, cherrybark








oak, American elm, willow oak, sweetgum, hackberry, and sugarberry

(Hosner and Boyce 1962).


Fire

Trees found on flooded soils often have the capacity to grow on

drier sites but are controlled by fire. This is especially true for

hydric hardwood trees and bays. For example, the suppression of fire has

allowed fire-intolerant floodplain species to colonize open pine

flatwoods along the Alafia floodplain in central Florida (Clewell et al.

1982). Expansion of bayhead communities into adjacent flatwoods was

attributed to fire suppression in the Lake Wales Ridge section of central

Florida (Peroni and Abrahamson 1986). In the Big Cypress Swamp in south

Florida, fire maintains most sites as subclimax associations; otherwise

forest would succeed toward the climax mixed-hardwood forest (Duever et

al. 1984). In south Florida a surface fire in the cypress-mixed-hardwood

swamp kills the shallow-rooted hardwoods leaving monospecific stands of

more fire-tolerant cypress trees (Ewel and Mitsch 1978, Gunderson 1984).

In more severe burns where cypress are killed, willows and marsh plants

rapidly invade flooded sites, and are maintained as a sub-climax by

frequent recurring fires (Craighead 1971, Gunderson 1984). Fire

regenerates Atlantic white cedar (Chamaecyparis thyoides) (Brown and

Montz 1986). Without fire an older cedar stand will give way to bay

trees and swamp hardwood seedlings (McKinley and Day 1979).


Biotic Disturbances

Grazing activities of deer, cattle, sheep, hogs, and other large

animals result in substantial changes in forest sites. In well

established forest, large herbivores may open up space for forest








regeneration, but intense grazing pressures can change grasslands to

brushland, forest to thorn thickets, or woodlands to grasslands (Wagner

1969). The highly compacted soils of heavily grazed woodlots lower

initial moisture content after the rainy season and increase desiccation

during droughts by retarding permeability.

Cattle and deer browsing can kill seedlings in bottomland hardwood

forests in the southeast, especially if floodwaters concentrate browsing

on higher ground in the floodplain (Wharton et al. 1982). Baldcypress

and water tupelo have been reported to survive only one cropping by deer.

Swamp rabbits and nutria were such a problem in Louisiana that forest

plantings of baldcypress were suspended until something could be done

about these animals (Brown and Montz 1986). Connor and Day (1976)

discussed the effects of grazing by the forest tent caterpillar on

baldcypress-water tupelo and bottomland hardwood sites in Louisiana.

They speculate that the susceptibility of water tupelo to defoliation may

be one factor that favored the maintenance of nearly pure stands of

baldcypress.

Herbivores are by no means always detrimental to succession and

often play an essential role by dispersing seeds, creating gaps, and

selectively browsing or eliminating some species. For example, there are

rough correlations between successional stages which show an increased

palatability of early species compared to those later in the sere (Cates

and Orians 1975).


Seed Dispersal

Early successional species have large numbers of easily dispersed

seeds while later trees produce large seeds with more restricted range

and limited dispersal agents. Lack of adequate seeding can be a






21

significant impediment to succession. Bird dispersal of seeds is rarely

more than 100 to 200 m from the parent tree (Cruden 1966, Howe 1977,

Debussche et al. 1982, Glyphis et al. 1981, Smith 1975). In studies of

old field succession, bird-dispersed tree species were not established

until several years after wind-dispersed seedlings had matured indicating

bird perches are valuable for seed migration (Buell et al. 1971, Pickett

1982). Blue jays are important dispersers of acorns and other large nuts

by collecting only sound seeds and dispersing them over a wide range (up

to 4 km) from their source (Darley-Hill and Johnson 1981, Johnson and

Adkisson 1985). Even with water dispersed cypress and tupelo seeds the

majority of seeds are retained within 1800 m of their origin (Schneider

and Sharitz in press).

Wolfe (1987) cites proximity to seed source as perhaps the single

most important factor in natural succession on mined lands. In his study

few wind dispersed seeds were found beyond 45 m from a forest edge in

Polk County, Florida. Seed migration was significantly greater in

Alachua County, Florida, where patches of forest remain. He also

reported that although bird-dispersed seeds were less abundant with

increasing distance from seed source they were commonly concentrated

around perch sites which increased the distance they were found from a

parent tree.

Regeneration of baldcypress is hampered by poor seed germination and

its exacting requirements for light and moisture in early life (Brown and

Montz 1986). Cypress is difficult, but not impossible to reestablish

once it is removed by lumbering (Gunderson 1984). In deeply flooded

swamps successful germination and establishment of baldcypress and swamp

tupelo require an abundant supply of seed, plenty of moisture during









germination, a lack of predators, and enough growth to escape the next

flood event (Putnam et al. 1960, DeBell and Naylor 1972). These and

other studies emphasize the importance of seed source in forest

establishment.


Light Availability

Shading by the canopy becomes dominant as a factor influencing plant

succession after vegetation has advanced to a stage where plant cover is

sufficient to reduce light intensity to below 20% of full sunlight

(Shirley 1945a). For early successional species such as pines, growth is

best under full sunlight. Although pines grown at 50% shade survived,

they showed less photosynthesis and growth (Pearson 1936, 1940, Shirley

1945a, Kramer and Decker 1944). Other species, however, show as much or

more height under partial light. This is especially true for hardwood

species compared to early successional pine seedlings. Dogwood, white

oak, and red oak attained a maximum rate of photosynthesis at one-third

or less of full light intensity and further increase in light did not

increase photosynthesis (Kramer and Decker 1944). McDermott (1954)

showed the maximum height growth of sycamore was measured at 20% relative

illumination (RI), American elm at 33% RI, and river birch and alder at

50% RI. In a test to determine light requirements for seed germination

of hardwood species, McDermott (1953) found red maple was not sensitive

to low light intensity and American elm was only moderately sensitive.

Baldcypress height growth tended to increase as light intensity decreased

with its greatest growth at 25% of full sunlight, while pondcypress grew

at about the same rate for all light intensities except for the lowest,

5%, where it grew poorly (Neufeld 1983). Planting seedlings under a tree







23

canopy ameliorate harsh environmental conditions caused by high radiation

loads and may increase survival (Holbo et al. 1985).

Regardless of height growth and shade conditions all species show

much reduced, poorly developed root systems under shaded conditions

(Spurr and Barnes 1973). Shade increases susceptibility to summer

drought (Shirley 1945a).

Also, soil may affect the ability of plants to endure shaded

conditions. For example, Tubbs (1969) found good germination occurred

for yellow birch under full sunlight on mineral soil, but on heavy clay

or organic soils deep shade was required.

Putnam et al. (1960) described tolerance to shading for species in

the southern forest. The shade tolerant species common to wetlands

included hackberry, beech, dogwood, holly, magnolia, the bays, red maple,

the elms, persimmon, and the hickories. Although oaks and ashes were

reported to germinate beneath a complete canopy, the seedlings die back

within 3 years if not released by an opening in the canopy. Other

species, such as the tupelos and sweetgum, germinate sparsely especially

in small openings, but survive only so long as bright light is available.

Baldcypress regenerate in swamps where conditions are moist and

competitors are unable to cope with flooding. Baldcypress seedlings and

saplings are only moderately shade tolerant and may be suppressed by

faster growing species.


Physical Soil Properties

Soil texture affects tree growth by regulating pore space and

consequently both the water and air-holding capacity of the soil. Clays

and organic matter retain water and nutrient ions. Clay loam soils are

usually considered an ideal plant medium. Sandy soils without organic







24

matter retain water poorly and heavy clay is often anaerobic (Spurr and

Barnes 1973). Both conditions require plants with special adaptations.

Some of the most productive forests are found on clay soils

associated with alluvial floodplains (Broadfoot 1967). Texture per se

has little effect on tree growth as long as moisture, nutrients, and

aeration are adequate (Pritchett 1979). Many floodplain trees are

adapted to anaerobic conditions and the anaerobic gradient, rather than

moisture conditions, may be the more critical factor in plant

distribution (Wharton et al. 1982).

The Forest Service in Mississippi has conducted several experiments

to improve the silviculture value of Sharkey Clay soils since few

commercial alternatives for this type of site exists (Ferguson 1983).

These are reviewed here because Sharkey Clays have many of the same

properties as clay settling pond soils. The soil is poorly drained,

essentially flat, having 8-12% sand, 65-75% clay, 13-24% silt, and a pH

of 5.9 to 6.7 (Broadfoot 1976, Johnson and Krinard 1985). Growth of

green ash, water hickory, sugarberry and nutall oak was best on flooded

plots which were usually inundated from December to mid-May (Johnson

1975).

Water impoundments used to test tree growth on Sharkey Clays showed

variable results (Broadfoot 1967). Impounded species that made the

largest gains in growth, about 90%, were cottonwood, green ash, and

sweetgum. American elm and hackberry, two high-ranking species without

impounded waters, benefitted little from the treatment. Generally, the

oaks had the best growth of all species without impounded water, but only

small growth increases after water impoundment.






25

Sharkey Clay soils managed for silviculture were not as productive

as medium textured sites with the same treatment. Ten-year-old

cottonwood and sycamore grew about 55 to 60% as tall (Krinard and Kennedy

1983), 18-year-old sweetgums grew 57% as tall (Johnson and Krinard 1985),

and after 11 years green ash grew 70% as tall and nutall oak 90% as tall

compared to planted plantation trees on other soils (Krinard and Johnson

1981).

Krinard and Johnson (1976) described a cypress plantation at The

Delta Experimental Forest in Mississippi which showed good growth on

sharkey clay, which are generally not as suitable for silviculture as

more loamy soils. Plots were kept mowed for 10 years to reduce

competition. Survival of baldcypress at 4 years was 62%, at 11 years

41%, and at 21 years was still 41%. Height of trees after 4 years ranged

from 1 to 3 m and after 21 years the largest trees were over 20 m tall.

Compared to cypress plantations in Mississippi on other soil types,

survival was less. For example, plantings in Choctaw swamp had 80 to 95%

survival after 5 years (Rathborne 1951), a plantation along yellow creek

had 70% survival after 41 years, and a swamp in Natchitoches Parish,

Louisiana, had an initial 95% survival (Brown and Montz 1986).


Chemical Soil Properties

Nutrient deficiencies are not common in undisturbed forests. The

conservative nature of nutrient cycling, deep-rooting habit of trees, and

mycorrhizal fungal associations make efficient use of available resources

(Pritchett 1979). Problems are sometimes encountered with planted

seedlings especially if trees are not matched to site conditions. In

greenhouse experiments, baldcypress grew best with urea nitrogen

treatments, but no differences in tupelo dry weight were attributable to






26

nitrate or urea sources (Dickson and Broyer 1972). Wetlands are diverse

with respect to nutrients and pH. While many types of wetlands have more

than adequate supply of nutrients and are highly productive, some

wetlands are oligotrophic systems deficient in available nutrients which

were permanently tied up in organic deposits or sediments (Mitsch and

Gosselink 1986).

Monk (1966, 1968) divided Florida wetlands into two climax types

according to pH, nutrients, and depth of maximum flooding. In general

the bayhead swamp was more sterile, more acid, and not flooded as deeply

as the mixed swamp habitat. Dominated by broad-leaved evergreen trees

whose acid soils are high in organic matter, bayhead vegetation includes

sweet bay (Magnolia virginiana), swamp bay (Persea palustris), and

loblolly bay (Gordonia lasianthus). They may be found around spring

heads or ravines with constant seepage from higher terrain (Wharton et

al. 1977). In Florida, bay trees build up organic matter and elevation

making them typical of headwater streams while they are not common along

floodplains of larger rivers (Gross 1987, Clewell et al. 1982).

Mixed hardwood swamps are dominated by broad-leaved deciduous

species (Monk 1966, 1968). They occur along high energy, nutrient

enriched creeks, rivers, sloughs, and basins that are seasonally flooded.

Typical tree species are cabbage palm (Sabal palmetto), ash (Fraxinus

caroliniana), elm (Ulmus americana var. floridana), and baldcypress

(Taxodium distichum).

Other wetland species are generalists with a wide environmental

range. These included water oak (Quercus nigra), red maple (Acer

rubrum), sweetgum (Liquidambar styraciflua), and blackgum (Nyssa

sylvatica var. biflora).









Different conditions favor separate cypress species. Although

blackgum is found in association with pondcypress and baldcypress, the

two cypress species are almost never found growing together. The

critical indicator appears to be the low pH of soils in cypress domes

which range from 3.6 to 5.4 (Monk 1965). Laessle (1942) attributed

alkaline water in the St. Johns River, Florida, as an important factor

favoring baldcypress in an alluvial swamp and keeping acid loving plants

such as the bays from growing there. Low nutrient levels, especially

phosphorus, may limit the number of tree species able to colonize cypress

domes (Monk and Brown 1965).



Wetland Organization at the Landscape Level


Wetlands including lakes, rivers, and streams are organized in

hierarchial patterns (Leopold and Langbein 1962, Way 1973). The

landscape can be visualized as a web of energy flows converging from a

large support region where dilute energies are converged and cascaded

through the system with fewer and fewer components at each step until

there are only a few components at the high energy end of the scale

(Brown 1980, Odum 1982, 1983). For example, there are many more small

isolated wetlands and headwater streams than higher energy strands,

lakes, and rivers.

A wetland classification scheme (Odum 1984) based on source of water

and nutrients demonstrates the hierarchial concept for Florida wetlands.

Swamps without much inflow from runoff form bogs or bayheads. The soils

are low in nutrients, the water soft, the pH acidic, tree growth and net

productivity are low, and peat accumulates slowly. The most common tree

species are swamp bay (Persea palustris), loblolly bay (Gordonia






28

lasianthus), and sweetbay (Magnolia virginiana). Trees are evergreen, an

adaptation which may conserve nutrients (Monk 1966b).

With a larger watershed causing slightly increased nutrient levels

and alkalinity, still-water wetlands are colonized by pondcypress

(Taxodium distichum) singly or in association with blackgum (Nyssa

sylvatica var. biflora). With additional inflows slow moving channels or

sloughs are formed which differ little in vegetation from cypress domes

in the upper reaches but once strands converge with other streams and

sloughs enough water and nutrients are present for baldcypress (Taxodium

distichum) and its associates to become dominant. Eventually there is

adequate water energy to form a floodplain swamp with a depositional

environment having less peat and more clays. This swamp has many tree

species, abundant nutrient levels, higher alkalinity, higher pH, and a

wide diversity of water depths and hydroperiods. Although baldcypress

and water tupelo (Nyssa aquatic) are common they are not dominant. Lake

border swamps which typically have circulating waters and high nutrient

levels are also a characteristic environment for baldcypress.

Swamps observed over a long time period are seen to progress through

a succession of stages. The most developed is the climax, but there are

periods of regression, catastrophic reductions and changes caused by

hydrological and geological conditions. The concept of optimum frequency

of pulses for maximum power postulates an alternation of net production

and storage with pulses of consumption and use (Odum 1982). Analysis of

the pulses which control succession suggests a pattern of increasing

intensity and reduced frequency as the time scale is increased. For

example, pulses range from small diurnal changes of sun and rain; yearly

patterns of temperature and flooding; cyclical patterns of wet and dry






29

years; and catastrophic events which produce drastic changes by harvest,

drainage projects, or clearing (Odum 1984).

Clay settling ponds represent a large pulse: It is the beginning of

a cycle recovering from drastic change over the long time scale--total

earth upheaval, separation of constituent parts, and redeposition of the

unused products. Some material storage remain but are now disorganized.

An energy source includes high nutrient concentrations mixed with clays

which, in small amounts, are beneficial to plant growth and community

development. Initially industry supplies an energy subsidy in

circulating water which creates a productive wildlife habitat for water

birds. As time passes the water subsidy is shut off and weedy early

successional vegetation takes over. Some people perceive this community

as not useful so reclamation is encouraged or mandated.



Self-Organization of a New Climax


This dissertation is concerned with material processes of landscape

organization and the use of this information to develop low energy

alternatives to accelerate succession toward a more mature climax system.

What will be the properties of this new system? The configuration of the

drainage basin on clay settling ponds suggests pondcypress typical of

still-water isolated wetlands might be the best choice for wet

depressions. On the other hand high nutrient conditions and widely

fluctuating water tables with clay soils may favor a baldcypress or

alluvial floodplain association. The clays continue to dry over time,

differential settling occurs and distinct environments are created.

Additional species may be able to adapt to the new conditions. Sand

tailings on top of clays or spoil mounds left in clay ponds may create








seepage zones which meet the requirements for bayheads which could

function as headwater streams to the regional drainage network.

Perhaps a diversity of vegetation in a mosaic will develop. Clay

settling ponds create a unique environment not previously found in

central Florida. Species from other areas may be better adapted to

survival here, such as those found on clay soils in Mississippi or north

Florida.

Ecological engineering which involves the use of small energies

supplied by man may produce large effects in accelerating succession.

How limited are present successional patterns because there is a shortage

of genetic alternatives which have been unable to find their way to the

new kinds of disturbed soil (Odum 1981)? Can accelerated plant

succession enhance the building of soil structure in the same manner as

the restoration of native prairie grasses have provided a low cost, low

energy method of improving soils for agriculture in the midwest (Miller

and Jastrow 1986)? Whether succession on clay settling ponds is

intermingled with agriculture uses or directed toward some kind of climax

on otherwise derelict land, nature's work seems to be in symbiosis with

human purposes.



Description of Study Sites


Three field projects were conducted to better understand

successional processes that might enhance forest reclamation on clay

settling ponds. In order to gain a broad overview, an initial survey

documented vegetation found on 12 clay settling ponds of different ages.

Experiments designed to test the suitability of cypress-gum species in

wet depressions of clay ponds were tried in seven locations. A variety







31

of hydric tree seedlings were planted in six clay settling ponds and two

control sites. Locations of sites are shown in Figure 3. Three sites,

Gardinier (area A), 0. H. Wright, and Tenoroc (area 4A) were used for all

the research projects.


Vegetation Survey

Twelve clay settling ponds were chosen to represent sites with no

sand or overburden additions. Five sites were between 10 to 20 years

since decommissioned, six were 25- to 35-years-old, and one was a series

of old dredge pits backfilled with clay over 60 years ago. Since the

oldest site was no further advanced in the successional sequence, it was

included in the 25 to 35 year age group for analysis. Ages were

estimated from aerial photographs taken approximately every 10 years

since the late 1930's by the Agricultural Stabilization and Conservation

Service in the U.S. Department of Agriculture. Ages of sites represent

estimated time of abandonment or reclamation until 1984 when this project

was started.

For the 10 to 20 year age group, two ponds were located on mining

property of Gardinier, Inc., near Ft. Meade, Florida. Area A (GA-A) is a

rectangular clay settling pond on unmined land which had a perimeter

ditch and culverts to accelerate drying of the clays. Area B (GA-B) had

been reclaimed by knocking the dikes down around the edge of the pond and

re-contouring. The vegetation was sampled in the wet interior where no

dike material had been placed. Both ponds were in active use in 1968.

Noralyn-1lA (N11A) and Noralyn-11C (N11C) were adjacent ponds owned

by International Mining and Chemical Corporation. They were both in

active use in 1968. The N11C site represents a method called stage
























































'Map area


0 4
mi s I
miles


SAGRICO


STENOROC


AVE.


BAR'


N11CE
CAMP
MEETING
GROUND


AREA BE I
WRI(
AREA A
GARDINER


POLK___
FIND. HARDEE


Location of study sites.


Figure 3.








filling where ponds are allowed to dry out and re-vegetate between

filling intervals.

A clay settling pond adjacent to the Florida Institute of Phosphate

Research south of Highway 60 and owned by U.S. Steel (USS-FIPR) was also

in the 10 to 20 year age group. The pond was built on mined land. It

has several tall spoil piles in the interior and is shown full of water

in 1968.

Of the six clay settling ponds in the 25 to 35 year age group a site

owned by U.S. Steel on the south shore of Lake Hancock is the youngest

(USS-LH). In aerial photographs it is shown during the filling stage in

1958. It is reported to have been stage filled where clays were allowed

to dry between filling cycles. It was built on unmined land.

A clay settling pond owned by Hopewell and leased as part of

Alderman Ford Ranch (HO-AFR) was under construction on unmined land in

1948. It is adjacent to the Alafia River Floodplain.

A clay settling pond (MA) located at the end of Maine Ave. east of

Eaton Park was constructed on unmined land adjacent to the Saddle Creek

Floodplain and in aerial photographs was shown being filled with clays

during 1952.

Zone 1 (TR1) located at Tenoroc State Reserve was filled with clays

around 1952. It has a large sand tailing pile 8 m thick deposited on one

end and a lake on the other. The transect surveyed is between these two

features.

The 0. H. Wright site (GA-WR), owned by Gardinier, Inc., is located

on an old mine site adjacent to the Whidden Creek Floodplain. It has

many protruding spoil piles. Aerials show it being used for clay

disposal in 1952. Only vegetation on clays in the mine cuts was sampled.







34

The Alderman Ford Ranch site (AFR) had the most diverse forest of

all the sites surveyed. It is located adjacent to a wide undisturbed

floodplain. The clay settling pond was constructed on unmined land and

aerial photographs show it being filled with clays in 1948.

The Clark James Site (IMC-CJ) is a series of old dredge pits

backfilled with clay. They were estimated to be greater than 60 years

old when the vegetation survey was taken.


Cypress-Gum Seedling Transects

Seven clay settling ponds representing different ages and

reclamation techniques were planted during the winter of 1984-85. The

locations are shown on the map in Figure 3 and summary information is

listed in Table 1.

Tenoroc (Area 4A) is a large clay settling pond located in a State

Reserve under the jurisdiction of the Dept. of Natural Resources. The

west end, where most of the trees were planted, was mined and has many

protruding spoil piles. Aerial photographs show the site being mined in

1958 and clays filling the mine cuts in 1968. Three paired transects of

bareroot seedlings were located on the edge of a pond. Four paired drier

transects of tubelings and one paired transect of bareroot seedlings were

planted in willows growing at the northwest corner. The location of

study sites is shown in Figure 4.

Gardinier (Area A), located at their Ft. Meade mine, was abandoned

as a spoils area in 1973. The pond is rectangular in area with clays

approximately 10 m thick and a surface area of 130 ha (Blue and Mislevy

1982). During the filling stage coarser materials settle near the inflow

pipe with the finest material migrating to the lower end. In 1975 a

perimeter ditch drained the higher ground and the dikes were lowered as a












CC



n -4
moo


mQ
MM
SE-4Z


















S
H
O





0
l#l


r- N


cWn



W0

M w
EH E-


w en
N z
c~o

M
-U
E-4 M
EU


U
4 9z
Z E-


M Q. P-
O





U ( 2
04 2 4


0 C 0 C E- Clo
Z U Z ( COl d


N
NeZ
-0








E-l






I-4
U


-44
4 00

c ZN
c oo
U E--
0 8


pa as
E-4 M
ETin
00 00


Has
r4O
E-4 fc


oloo
-4-'


0 o o %0oo -a- co
in% Uln 11 Ln r-- Wn
u u -4
i-I3 -43 c


r- LM
r-r- in


Q





1x1





1. W
i--




M

1-
UHOQ arO B
S4 04 u

w E4 E4

0 mH E-4
SEE-4

















X IAU B .<
0o 0







'0




% 9z
&- &(


0


C-I
z



E-4
0
CCO



E-4
a& W




4 .-it
'4'













,-4

,4 C O
to 0
M o
W 0 r.





a) O0
W 0
c-
u 4








>. 0






C40 0.
C4"
$4 0 M

o IC


00
rl 0o



E-4
SQW
) 4a w
541.


0 m



V 400

ao o 4



Sca 0 0C
'0 3


2 0









000
IM-4

a)40 0W
0 .. c








'0
1r. 0S 4


0fl 0 >
-45-I I















02
0a2 o u a














a0.
uac









CO 0 J <1
M C 0C U
E- f- 00







37

reclamation method. Outfall pipes are now above the level of the clays

providing drainage only during extremely flooded conditions. Trees were

planted in the wetter lower end which has been flooded for the past

several years. During the early establishment phase, however, a drought

and subsequent fire caused considerable mortality. In 1987 work is in

progress to break the dike, drain the pond, and mine the land underneath

the pond. Figure 5 shows the location of the tree seedling plots.

0. H. Wright, owned by Gardinier, is located adjacent to the Whidden

Creek floodplain. It is an old strip mine backfilled with clay. Aerial

photographs from 1957 show the mine cuts being filled. It has well-

defined rows of spoil piles with mine cuts about 10 m wide. The clay

filled cuts are colonized by willows, intermediate zones support red

maple, elm, and a few other hydric species, while the spoils have a more

diverse vegetation dominated by oaks. As shown in Figure 6, some of the

study transects are located in a periodically flooded low area while

others are in the drier narrow mine cut. Some of the uncleared plots were

partially placed on spoil piles covered with clay, but with higher

elevations and only 80 to 100 cm of clay. These were not used in the

paired plot evaluations. Although it is not known how deep the clays are

in the swales, during well installation it was determined plots are

planted on at least 3 m of clay, except transect 6, where clays were 182-

cm thick.

Alderman Ford Ranch was strip mined by the American Agriculture and

Chemical Co. Aerial photographs show the clay settling area being filled

in 1948. This unreclaimed clay settling pond is approximately 70 ha and

is placed on unmined land. It is located immediately above the

confluence of the north and south prong of the Alafia River. A






















,rIta














o 3\


-- -






-

Y --

I
a-



e-
I-
0-

0-


C
(0 (0
o


0



C
0 a)


0q
0C
0.



m 0



0.
ca






9C
0
U-4.
0m







II II
n (zic


0- -
e--


I-
0-
a -



0- 1
a-j
___


V a? 0


N


mm


0
ta


> *0

0
CO


0



> 0
4) Wo
0u4

.o



00
w 0

a


S00.


0 0o
4)
0
oo


a0.








CD 0


0












m
0 0















00
*M
0 A

Q 4)4
00








z4-4 o
u- 0d 0)


0 ^3 SD


J


IMV (



































.
WZ




0-

0

I LL

Z


*HWc
cis ca










0
a











4-O O
0c
pa W


ow

















41
S1-4
s a







-rl ( 4






0 0
















a o
c


ot
a)



























r4
o i
Sa)















-'-4
Oa 4-







a o^-
0 C
*ili-
4- M
n)c
u <






Page
Missing
or
Unavailable














0 125 250 50
mtI Ir
meters


ALAFIA
RIVER
FLOODPLAIN


= PAIRED SEEDLING TRANSECTS
WELLS USED FOR WATER TABLE DATA



Figure 7. Sketch showing locations of tree seedling transects at
Alderman Ford Ranch.







42

transects are in a willow forest adjacent to the lake and two are planted

in the lake. Figure 8 is a sketch showing the location of the tree

seedling transects.

CF Industries used a sand/clay mix for clay disposal at their Hardee

mining site which allows for much faster consolidation of the clays than

traditional methods. Filling of SP-1 (also designated Area R-2) was

begun in October 1980 and was discontinued in September of 1982

(Garlanger 1982). The pond covers approximately 44.5 ha and the mix

averages about 11 m deep (Garlanger 1982). The pond was constructed with

a divider dike located halfway between the east and west walls. The

study site shown in Figure 9 is located at instrument station number 3

established by Ardaman and Associates for an evaluation of sand/clay mix

reclamation. Six plots have been placed more or less perpendicular to

the shoreline of a wet depression. The plots were planted across a

gradient of open water, marsh vegetation, and then willow as the

elevations increased. The site is characterized by enormous cracks.

Mobil has reclaimed a series of clay settling ponds west of the

Peace River floodplain. The site used in this study is a pasture pond

located south of Highway 640 near Homeland. From aerial photographs it

appears to have been abandoned from active use as a clay settling area

about 1960 and capped with sand tailings about 1979. At least part of

the site had been mined before clays were deposited. It is now leased

for pasture and the pond is partially fenced which keeps cows off half of

the transects. Seedlings were planted in a small shallow lake that

seldom goes dry. Soil at the surface in the transects was 100% sand. A

sketch of the study plots is shown in Figure 9.





























LAKE~ SPOIL

SPOIL

3.


Figure 8. Sketch showing location of tree seedling transects at IMC-H9.















































Pasture


0\
0
Ct


Pasture


Figure 9. Sketch showing location of bareroot seedling transects
a) C. F. Industries, b) Mobil pasture pond at Homeland.


Pea









Hydric Swamp Plots

Six clay settling ponds and two control sites were selected to plant

eleven species of tree seedlings (Figure 3). Sites included Gardinier

area A, 0. H. Wright, and Tenoroc area 4A already described in the

cypress-gum seedling experiments. Additional sites are described below.

Pruitt Ranch is a conventional clay settling pond deactivated

approximately 36 years ago. It is located in an area where the regional

water table is close to the surface which had kept the clays wet and

maintained a herbaceous cover of cattails over most of the pond. The

pond was built on mined land and protruding spoil piles were vegetated

with myrtle. The land surrounding the mined area is a well maintained

pasture. During 1984-85 perimeter and interior ditching was implemented

as a reclamation technique and in late 1985 existing vegetation was

removed with a controlled burn. Although the site was considered too wet

for cows, there was evidence that cows have been grazing on the planted

tree seedlings. The clays were not thick over a large portion of the

site. In the northeast corner where the trees are planted, plots 1 and 2

have clays over 3 m deep, but plots 3 and 4 are on shallow clays of about

50 cm and a mixture of sand and clays below that depth. A sketch of the

seedling plot locations is shown in Figure 10.

The IMC-Peace River Park is a conventional clay settling pond

estimated from aerial photographs to have been mined from 1952 to 1956

and used as a settling area until 1968. Several spoil piles are above

the elevation of the clays. It was leased for pasture until 1986 when

plans were made to establish a county park on the site. The study plots

(Figure 11) were located next to a small lake.



















































Figure 10.


Sketch showing location of seedling plots at Pruitt Ranch.









47












z
o
0\ 0








-JJ











S----









CY
4.
















0
c















0
0
C =O 0


























o
U 5











-0
0 0
B .l
sl(U



1-
i r I-z a

: *
-


sc
M~








--------------------* --------------------____________ f







48

The Agrico site is a 193 ha clay settling pond located at their Ft.

Greene mine. Pumping of clays was terminated in the late 1970s but the

pond was used for tailings and debris until decommissioned from active

use in 1983. It is the third pond in a series so the finest clays

migrated to this site. Depth of clays vary from 1 to 10 m. A successful

agriculture operation was started in 1984 which is part of a reclamation

project. In 1986 the surface soils have been dried with ditches 5 to 15

m apart and alfalfa has been planted to accelerate drying further. Three

of the tree seedling plots were planted on shallow clays (1 m) where

alfalfa was already established and three plots were on deeper clays (10

m) near a drainage spillway.

Seedlings were planted at two control plots to compare to seedlings

planted in clay settling ponds. Two plots were located in a backswamp on

the undisturbed Peace River floodplain. Soils are muck typical of

flooded lands. Location of the study plots is shown in Figure 12.

Another control site was selected at Camp Meeting Ground Creek south

of Homeland. Mining has impacted what had been described as a beautiful

setting where, in the old days, people came from miles around to camp

before special church events. Today the creek has been diverted and now

runs a straight course at the base of a clay settling pond. Vegetation

is in a willow and myrtle early successional stage. Sandy soils are the

substrate. A sketch of the study plots is shown in Figure 13.

















































z
0

-50

WWOO
oua
a.. m, a.













































Figure 13. Sketch of control plots at Camp Meeting Ground Creek.
Plot 2 is divided into 3 sections to fit it along the
creek floodplain.















METHODS


The research methods are described by task. The first task was

designed to better understand natural succession by measuring vegetation

on ponds 10 to 20 years since decommissioned compared to those over 25-

years-old. Two tasks were used to test success of planted tree

seedlings. Cypress-gum swamp species were planted in wet depressions

during the winter of 1984-85 and a wider range of wetland trees was

planted in plots representing different moisture regimes in February of

1986.



Community Structure Measurements


Vegetation Analysis

Trees and shrubs were sampled by measuring diameter at breast height

(dbh). Quadrat size was 100-m2 for trees greater than 10 cm dbh while

shrubs with dbh from 1 to 10 cm were measured in a 10-m2 area within the

larger quadrat. Herbaceous vegetation was estimated using percent cover

in three 0.25-m2 areas in each quadrat.

Quadrats were selected systematically along compass lines across the

clay settling area. The distance between paired quadrats was typically

50 to 150 m depending on the size of the pond. From 6 to 15 quadrats

were sampled in each site. Five sites were between 10 to 20 years since

decommissioned, six were 25- to 35-years-old, and one was a series of old

dredge pits backfilled with clay over 60 years ago. Quadrats were

51









52

usually sampled in pairs. Adjacent plots in each clay settling pond have

the same number but one is A and the other is B. Site abbreviations are

given in site descriptions.


Wetland Site Index

A form of gradient analysis using weighted averages divided quadrats

into 5 moisture classes. Statistical analysis using Duncan's Multiple

Range Test for willow biomass further refined distinct community types

found on clay settling ponds. The Fish and Wildlife Service National

Wetland Inventory (FWSNWI) list (Reed 1986) provided an ecological

ranking for each species in Florida which in this study divided plants

into 4 categories from obligate wetland to facultative upland. The

method weighs each species contribution to the index by its relative

abundance in wetland habitats. The list was used to produce a single

number which represents the range of upland to wetland by values from

zero to one, respectively. In this study herbaceous vegetation was used

for the site index since this gave a more realistic estimate of relative

moisture than trees which once established can persist for a considerable

time as ponds continue to dry out. An example of the calculations for

one quadrat in this study is shown in Table 2.

The procedure is described in detail by Michener (1983) whose

purpose was to find a simple means of using plant species lists to

delineate boundaries between wetlands and adjacent upland areas.

Wentworth and Seneca (1986) have evaluated the method for two data sets

where results could be validated with quantitative measurements. They

found the ranking of vegetation type made by a skilled worker was nearly

identical to that computed using the weighted average technique.








Table 2. Example of wetland site index calculation.


Name of Plant Species Percent Total Cover Category Product
Category in Category Cover in Category Value
(1) (2)


FACU Thelepteris normalis 5 10 X 0.18 = 1.80
Eupatorium capillifolium 5

FAC Parthenocissus quinquefolia 15 20 X 0.50 = 10.00
Acer rubrum 5

FACW None 0 0 X 0.82 = 0.00

OBL Hydrocotyle umbellata 5 26 x 1.00 = 26.00
Lemna minor 1
Salvinia rotundifolia 20



Total Cover all Species: 56 Total product: 37.80



To calculate Wetland Site Index:
Total Product/Total Cover = 37.80/56 = 0.675


NOTES:

(1) Wetland plant list for Florida compiled from regional
wetland plant manuals for the U.S. Fish and Wildlife Ser-
vice (Reed 1986) assigns plants to the following categories:


Obligate (OBL) found in wetlands under natural (not planted)
conditions frequency (>99% frequency)
Facultative Wetland (FACW) Usually found in wetlands (67%-99%)
frequency
Facultative (FAC) sometimes found in wetlands (34%-66% freq-
quency) Facultative Upland (FACU) seldom found in wetlands
(1%-33% frequency)


(2) Median of wetland frequency listed under (1)









In general, direct gradient analysis is a method of smoothing raw

field data for cases in which major environmental factors are evident

(Gauch 1982) in this case a moisture gradient. The application of

weighted averages as a method in plant ecology dates back to the 1950's

(Curtis and McIntosh 1951, Whittaker 1967). Whittaker (1978) summarized

its development and use.

Since the results seemed appropriate, quadrats in this study were

grouped into wetland classes using the same scheme as in the northeastern

United States (Michner 1983). Well-drained soils produce communities

with index values between 0.1 and 0.5; poorly drained soils between 0.5

and 0.7; and very poorly drained soils between 0.7 and 0.9. The group

from 0.9 to 1.0 was modified for this study. The group from 0.90 to 1.0

was observed to be permanently inundated in the U.S. Northeast while

these sites were only frequently flooded for one group and periodically

inundated for another in clay settling ponds. Frequently inundated sites

which were often too wet for trees to survive were given a value of 1.0

and those flooded for a shorter time period ranged from 0.9 to 0.99.



Cypress-Gum Seedling Transects


Seedlings characteristic of cypress wetlands were planted in wet

depressions of clay settling ponds during the winter of 1984-85.

Survival height was statistically related to measurements of water

regime, soil characteristics, cultural treatments, and grazing.


Seedling Plots

Elongated quadrats (4-m X 30-m) were established through an

environmental gradient from dry to wet where possible. Cleared and







55

uncleared transects were places adjacent to each other (Figure 14). One

hundred transects with 93 trees each were planted with a KBC planting

bar. A KBC planting bar is triangular in cross section which tapers to a

point making a hole slightly larger than the plug of soil around the

tubeling roots. Seedlings were arranged in 3 columns on 1-m centers and

one of 3 species was randomly assigned to each column. In paired

experiments, the tree order was duplicated. Survival and tree height

were measured for 50% of the trees 3 months after planting in May 1985

and for all of the trees approximately one year after planting in April

1986.

Both bareroot and tubeling seedlings were planted. Tubelings were

grown by Pete Wallace's Nursery, Rt 1 338F, Gainesville, Fla 32608.

Tubelings were grown for 6 to 8 months in styrofoam or plastic flats

which hold from 80 to 100 individuals in each flat. These were held in a

shade house and kept well-watered until planted. The soil was a nursery

potting medium and seedlings were fertilized as needed while in the

nursery. Species were baldcypress (Taxodium distichum), pondcypress

(Taxodium ascendens), and blackgum (Nyssa sylvatica var. biflora).

Bareroot seedlings came from the Division of State Forestry at

Chiefland. The species included baldcypress (Taxodium distichum), green

ash (Fraxinus pennsylvanica), and water tupelo (Nyssa aquatiaca). They

were grown in the ground from seed, fertilized at planting time with one

or two more top dressings applied during the growing season. Seedlings

were approximately one year old when pulled from the ground, tied into

bundles of 50 or 100 seedlings, and kept in a cooler at 4 C until

planted. The time of storage ranged from several days to several months.





























































Figure 14. Design for paired transects used in cypress-gum
seedling experiments. Trees were also planted
at PVC pipe and flagging. The transect which
included columns 4, 5, and 6 were cleared of all
above-ground vegetation at time of planting.









Paired plots at 4 clay settling ponds were used to compare

baldcypress seedling types. Bareroot seedling transects were placed

within 10 m and parallel to the tubeling plots for comparison.

To understand the role of competition from existing vegetation,

other paired plots were planted with one of each pair cleared of all

above-ground vegetation with a machete, bank blade, or chain saw. Cleared

plots were placed 6 m from and parallel to uncleared plots. Three clay

settling ponds were used for these experiments.


Grazing Experiments

An experiment was designed at Mobil pasture pond to test the

influence of cows on seedling establishment. A fenced area excluded cows

from half of the pond where seedlings were planted. In the other half of

the transects cows were allowed to graze freely. Four transects with 93

seedlings each were planted for each treatment. The transects with cows

were not included in any of the comparisons except for the one

specifically testing the effect of cows.


Water Levels and Hydroperiod

Water table levels were measured monthly during the 1984-85 growing

season (Oct, Mar, Apr, May, Jun, Jul, Aug). On flooded sites water

depths were taken at a reference marker. Where the water table was below

the surface, shallow piezometer wells were installed which reached to the

water table. Wells were made of 3.8 to 5.08-cm diameter PVC pipe fitted

with 50 to 60 cm of well screen on the bottom and capped.

For dry sites, elevations were taken with a level and stadia rod at

each cluster of transects. A moisture measurement for each tree was

related to the elevation of the wells and its water table reading. Where









trees were flooded, the depth of water was measured at each tree. By

these methods a relative moisture measurement was estimated for each tree

for comparison within and between clay settling ponds. An average of the

seven monthly readings during the growing season of 1984-85 was used for

statistical analysis. At Alderman Ford Ranch a grid of wells was

established. Here, water studies were done in collaboration with M.

Miller with monthly monitoring.

In order to compare survival and height to water table depth, sites

were divided into moisture classes. Several divisions were tested and

the following were selected because they usually had sufficient

observations in each group for meaningful statistical analysis. Water

table depths in cm were divided into the following classes: more than

-200 cm below ground surface, -200 to -100, -100 to -50, -50 to -30, -30

to -10, -10 to +10, +10 to +30, +30 to +50, and greater than +50 cm.


Monthly Rainfall Data

Rainfall data for 1985 were supplied by mining companies for

measurements as close as possible to the study sites. For Gardinier the

station was about one mile from Area A and two miles from 0. H. Wright.

Agrico supplied data from their Saddlecreek mine adjacent to Tenoroc.

The station located near 1-4 was about two miles from the study sites.

Rainfall data from International Minerals and Chemical Corp. Clear

Springs Mine was used for IMC-H9 about one mile distant and Mobil pasture

pond about 4 miles away. CF Industries weather station was about a half

mile from the study sites at SP-1.

The average rainfall from 1951 to 1980 is reported by the National

Weather Service for specific observation sites. An average of 4 weather







59

stations in the phosphate district was used to compare average rainfall

with rainfall during the planting and establishment year.


Soil Measurements

Soil cores were taken with a soil sampler using a mud auger head.

The ground was cleared of all leaf litter prior to sampling. Spoon size

aliquots from the top 15 cm of soil were combined to form a sample.

Soils were analyzed for water content, particle size, organic matter, pH,

and selected nutrients. Also a depth to water table on the day the

samples were taken is included. Particle size was determined by the

hydrometer method (Day 1965). The results were divided into percent

sand, silt, and clay using the U.S. Dept. of Agriculture scheme (<0.002

mm clay, 0.002-0.05 mm silt, and >0.05 mm sand). Soil moisture was

measured gravimetrically (Gardner 1965). Water content was computed by

the following formula:


DW = [(Wt of wet soil + tare) (wt of dry soil + tare)]
[(Wt of dry soil + tare) taree)] x 100.


Other measurements were made by the Institute of Food and Agriculture

Sciences (IFAS) at their extension soil testing laboratory in the Wallace

Building on the University of Florida Campus. Standard methods were used

as described by Rhue and Kidder (1983). Soil pH was determined in a 1:2

soil:water suspension. The Walkley-Black technique was used to measure

organic matter. The double acid method, Mehlich I, was used for the

extraction of potassium, calcium, magnesium, and phosphorus.


Statistical Evaluation

Statistics were performed using the Statistical Analysis System
(SAS) (Ray et al. 1982). The main-frame computer of the Northeast









Regional Data Center located on the University of Florida was used to

process data. Significance tests of frequencies for survival data were

determined using the chi-square test. A t-test for the difference

between two means was used to analyze height data. To test differences

between multiple means the Duncan multiple-range test determined

significant differences at the P = 0.05 percent level. Regressions

between tree height and water table depth were determined with an

analysis of variance procedure.



Hydric Swamp Seedling Plots


A variety of trees suitable for accelerating restoration of native

wet habitat in central Florida were planted on clay ponds during February

of 1986.


Plot Description

Six clay settling ponds and two control sites were selected to plant

eleven species of tree seedlings. The plot design is shown in Figure 15.

Thirty-eight plots were planted during late February 1986. Sites were

divided into 3 moisture categories, wet, dry, and transitional. Trees

for each moisture class are listed in Table 3. Abbreviations for trees

uses the first three letters of the genus followed by the first letter of

the species. Eighteen individuals of 6 species were planted in each plot

on 1-m centers. Plots were cleared of all herbaceous vegetation at time

of planting, but the tree canopy where it existed was not disturbed.

Plots were labeled by site abbreviation, plot number, and existing

vegetation. See Appendix B for a complete list of abbreviations.

Vegetation designations are H for herbaceous plots and types are P =


























5 ft PVC pipe
O 21/2 ft PVC pioe
+ tree plonlinM


-- -4- + + -4-
CONTROL
-4- + + -4- + + +
+ + + + 0 + + -
+4 + + 0 + +
S- -4- + + + + -
MULCH
+ + + + + + +
4- + 0 + + 0
4. 4- 0 -4- + 4-
+ + 4- + + + 4-
LITTER
.4- 4- + + 4- 4

Pipe Dimension
- 8-8m
Plot Dimension
-- r9m


Figure 15.


Design for hydric swamp experimental seedling plots.
Eighteen individuals representing six species were planted
in each plot on 1-m centers in a randomized pattern.
Litter was collected from a floodplain forest, mulch
was a bale of hay and control was no treatment. Seedlings
were also planted at PVC pipes used to mark plot bound-
aries. Understory vegetation was cleared and mulch
treatments covered a 9 x 12 area although plot boundaries
mark an 8 x 11 area. A water table well was installed
near the center of each plot.


I O






62

Table 3. Tree species planted in experimental plots on clay settling ponds
listed according to moisture categories. Abbreviations used in table
and graphs are also included.



Type of plot Abbreviation Scientific Name Common Name


Wet
(10 plots)





Transitional
(22 plots)





Dry
(6 plots)


TAXD
FRAC
ULMA
PERP
QUEL
NYSS

QUEL
SABP
ACER
GORL
TAXD
NYSS

MAGG
LIQS
QUEL
SABP
ACER
^^"fl


Taxodium distichum
Fraxinus caroliniana
Ulmus floridana
Persea palustris
Quercus laurifolia
Nyssa biflora

Quercus laurifolia
Sabal palmetto
Acer rubrum
Gordonia lasianthus
Taxodium distichum
Nyssa biflora

Magnolia grandiflora
Liquidambar styraciflua
Quercus laurifolia
Sabal palmetto
Acer rubrum


Baldcypress
Carolina ash
Florida elm
Swamp bay
Laurel oak
Blackgum

Laurel oak
Cabbage palm
Red maple
Loblolly bay
Baldcypress
Blackgum

Southern magnolia
Sweetgum
Laurel oak
Cabbage palm
Red maple


considerable undisturbed floodplain has been left at this point providing

an abundant seed source for revegetation located immediately across the

dike. Sand tailings have been added in a small area to the north side.

The clay settling pond has an interior ditch which recirculated water

immediately inside the dike during the active life of the pond. This

feature created several swales which are the wettest areas. The swale

adjacent to the dike near the upper end is drier and colonized by hydric

species. This is also the case for much of the uplands near the

floodplain. The deeper swales are characterized by a shrubby willow

community typical of many clay settling ponds. The clays are less than 3

m over much of the site, but where the seedlings are planted a test hole

indicates clays are at least 5 m thick. Twelve seedling transects were

planted in two swales which were known to be periodically flooded. The

outside swale was naturally colonized by maples (Acer rubrum) and oaks

(Quercus laurifolia) and the interior swale by scrub willows (Salix









Polygonum, L = Ludwigia, J = Juncus, T = Typha, I = Imperata, E =

Eupatorium, and A = alfalfa. For plots with a tree cover the dominant

trees are given with the most dominant species listed first, S = Salix, M

= yrica, B = Baccharis, and A = Acer.

Seedlings were grown by Wallace's Nursery in plastic flats and

maintained in a shade house until planted with a KBC planting bar. Tree

height was measured at time of planting and 8 months later in Nov. 1986.


Mulch Treatments

Plots (9 X 12-m) were divided into three mulch treatments. Leaf

litter from a nearby floodplain forest representing the same size area (9

X 4-m) was transported to the sites in garbage bags (nine 113 liter bags

for each plot). One bale of hay was spread over a 9 X 4-m area for the

mulch treatment and no ground cover was used for the 9 X 4-m control

section. A plot layout is shown in Figure 15.


Hydrology Measurements

A shallow (2.8-m) water table monitoring well with .5-m of well

screen at the bottom was installed near the center of each plot. Water

level readings were made monthly during the summer of 1986. Average

readings from April through October of 1986 were used for comparison

between plots.


Soil Analysis

Soil cores were collected in each plot (except Pruitt Ranch) during

a one day period in August 1986. Cores were taken from the portion of

the plot with no mulch treatments using a sampler with a mud auger head.

The top 5-cm was discarded and cores from the 5 to 15 cm depth were air

dried and mixed for analysis. Organic matter and pH were analyzed by







64

IFAS at their extension soil testing laboratory in the Wallace Building

on the University of Florida Campus. Standard methods were used as

described by Rhue and Kidder (1983). Soil pH was determined in a 1:2

soil:water suspension. The Walkley-Black technique measured organic

matter.


Direct Seeding

Direct seeding experiments were designed so results could be

2
compared with the results of planted seedlings. Plots measured 5 m and

contained nine 1-m2 quadrats for 3 replicate treatments in each plot

(Figure 16). The treatments were the same as the seedling plots --

litter, mulch, and control. In the case of litter, meter square areas in

floodplain forests were raked up and bagged for each litter quadrat in

the pond; For hay, 1 bale was divided between the 18 quadrats. Seeds in

the control plot were scratched into the soil with a pointed rake, while

seeds were placed on top of the clays and under the mulch in the other

treatments.

Three plots were placed at both Gardinier and Tenoroc: one in each

of 3 community types, myrtle, willow, and herbaceous vegetation. Seeds

of each species were counted into lots of 50 and placed in the plots

twice during the winter of 1984-85. In addition, seeds from the same

batch as those planted during February were compared in garden

experiments using gro-mix, a medium especially designed for seed

germination, compared to clay soils. To measure the effect of adequate

moisture half the garden experiments were watered while the other half

received only rainfall. Seeds were spread in flats (6 x 31 x 52 cm) on

top of soils in the same manner as field plots. Four flats were used for
each moisture treatment.




















CONTROL

(> ---- <


1 m -


MULCH E

4 ---- ,11


MULCH CONTROL








CONTROL LITTER
EE1 -


HEIGHT OF MARKERS

( 1.5m PVC PIPE
O 3.0m PVC PIPE
.75m PVC PIPE


Figure 16. Design for set of plots used in direct seeding
experiments.







66
Statistics

Statistics tested significant differences using the Statistical

Analysis System (SAS) (Ray et al. 1982). Facilities at the Northeast

Florida Regional Data Center located on the University of Florida campus

were used to process data. Significant differences between growth data

for seedlings was analyzed with Duncan's Multiple Range Test. Frequency

data were compared using the chi-square test. Linear regressions

determined relationships between two variables.















RESULTS


Sections follow that describe succession on clay settling ponds when

no reclamation attempts are made. Also included are two sets of tree

seedling experiments. One tested the feasibility of planted tree species

suitable for cypress-gum wetlands. The other compared trees from a

broader range of wetland types.



Successional Sequences for Clay Pond Colonization


Measurements of community structure for individual quadrats describe

vegetation organization on clay settling ponds. Then comparisons are

made about succession in relation to closest seed source.


Vegetation Organized by Soil Moisture

Data from field studies of succession are related to soil moisture.

Gradient analysis using herbaceous vegetation divided sites into five

moisture classes. Duncan's New Multiple Range Test separated the five

moisture classes into two types based on willow tree dominance. It was

noted willow is present in all groups, but grew best in quadrats with a

herbaceous wetland site index between 0.70 and 0.99 (Tables 4 and 5).

Tree species grouped according to the herbaceous moisture site index

show differences between two different age classes (Figure 17). Since

willow was found at almost all sites, it was used as an indicator species

to test for significant differences between moisture classes. Although








Table 4. Quadrats grouped by herbaceous wetland site index number
comparing tree species for moisture classes in clay ponds
10 to 20 years since active use.



SITE PLOT WETLAND TREE SPECIES *Basal Area m2/ha WATER
SITE REGIME
INDEX

Herb SALC MYRC BACH ACER FACW FAC


GA-B A6 1.000 FREQUENTLY
GA-B B5 1.000 0.71 INUNDATED
N11A D30 1.000
N11A D15 1.000 3.70
N11A B15 1.000 5.70
GA-B B6 1.000
N11A B30 1.000


N11C Al 0.995 11.61 2.28 PERIODICALLY
USS B4 0.987 2.92 FLOODED
USS B13 0.981 2.40
N11A B14 0.960 5.72
GA-A A2 0.919 3.43


GA-A B2 0.892 1.09
USS B9 0.878 8.36 0.11 0.13 VERY
GA-A B1 0.848 14.26 0.86 0.08 POORLY
GA-A A6 0.845 2.11 0.91 DRAINED
Nll-A D14 0.839 9.80
GA-A Al 0.807 3.76
USS A13 0.805
USS A9 0.762 3.68 0.38
GA-A B6 0.748 2.02 0.33
GA-A B9 0.727


N11C
GA-A
N11C
GA-B
N11C


0.680
0.663
0.568
0.521
0.515


5.97
0.84
3.21
0.17
3.54


0.51


2.92


0.27
2.50


0.28


BETTER
DRAINED








Table 4. Continued.


SITE PLOT WETLAND TREE SPECIES *Basal Area m2/ha WATER
SITE REGIME
INDEX

Herb SALC MYRC BACH ACER FACW FAC


GA-B B4 0.488 2.79 8.42
N11C B8 0.384 0.38 3.56 WELL
N11C B2 0.378 2.39 7.50 DRAINED
GA-B B2 0.327 2.01 4.26
N11C B5 0.326 3.21 1.70 0.71
USS A4 0.316 0.71 6.22 13.20
USS B7 0.287 3.40 0.86
USS A7 0.236 2.76 1.76
N11C A2 0.233 9.62 2.57
GA-B A2 0.182 0.90 0.52
GA-B A4 0.180


SALC
MYRC
BACH
ACER
FACW
FAC


Salix caroliniana Willow
Myrica cerifera Myrtle
Baccharis halimifolia Saltbush
Acer rubrum none present
Sambucus candensis Elderberry
Facultative trees none present








Table 5. Quadrats grouped by herbaceous wetland site index number com-
paring tree species for moisture classes in clay ponds over
20 years since active use.




SITE PLOT WETLAND TREE SPECIES Basal Area m2/ha WATER
SITE REGIME
INDEX

Herb SALC* MYRC BACH ACER FACW FAC


GA-WR B2 1.000 6.40 0.77 1.77 FREQUENTLY
IMC-CJ D3 1.000 18.67 INUNDATED
GA-WR B5 1.000 9.07
MA B3 1.000 3.70


HO-AFR A10 0.993 14.69 PERIODICALLY
IMC-CJ D2 0.989 8.06 FLOODED
HO-AFR B10 0.988 22.15
HO-AFR B7 0.987 5.47 1.84 0.47
HO-AFR Al 0.987 23.44
GA-WR A8 0.954 19.53
IMC-CJ B5 0.933 29.66
GA-WR A5 0.931 20.73
IMC-CJ B3 0.919 9.06
HO-AFR A7 0.913 7.75 0.08


VERY
POORLY
DRAINED


1.31 1.26


3.82 1.45
1.23


4.60
0.73


1.96
8.63


15.77
12.67
2.73
7.55
3.14


1.42
5.75
3.20

16.11
4.19
1.28


0.24
3.27
0.42



3.50

1.32


0.89
1.96 0.23


32.28
37.51
5.23
1.16
11.84
0.86
10.11
13.63
0.96
4.54
2.95


AFR
AFR
IMC-CJ
AFR
GA-WR
HO-AFR
GA-WR
GA-WR
AFR
HO-AFR
AFR
AFR
AFR
AFR
HO-AFR
MA
IMC-CJ
AFR
IMC-CJ
TR1


A15
B15
D5
A17
AB1
B4
Bl1
AO
A3
A4
B16
Al
A16
B7
Bl
B4
B4
A12
D4
A2


0.833
0.829
0.820
0.795
0.736
0.722
0.692
0.583
0.573
0.559
0.559
0.546
0.540
0.539
0.525
0.510
0.500
0.500
0.500
0.500


0.23

0.24

1.88


3.30


2.08
8.46


5.58
21.40








Table 5. Continued.


SITE PLOT WETLAND TREE SPECIES Basal Area m2/ha WATER
SITE REGIME
INDEX

Herb SALC* MYRC BACH ACER FACW FAC


1.99

2.84 3.32
0.67 4.15
3.02
2.54 1.28 10.12


0.95
15.98

0.64
13.31
3.42


0.94
1.32 5.62

4.91 1.16
1.68 4.98


AFR
AFR
AFR
USS-LH
MA
GA-WR
AFR
TR1
TR1
USS-LH
IMC-CJ
USS-LH
MA
GA-WR
MA
TR1
USS-LH
MA
TR1
TR1
MA
USS-LH
MA
USS-LH
USS-LH
MA
AFR
MA


1.53
5.96
2.69




3.41
1.89
2.21
0.92

9.18


BETTER
3.80 DRAINED












0.79


B12
A9
B9
LH1
A5
Al
A7
A5
B2
LH4
B2
LH3
A3
A2
B2
B5
LH6
A4
B8
A8
B5
LH7
A2
LH5
LH2
Al
B17
B1


0.491
0.488
0.478
0.477
0.474
0.473
0.469
0.468
0.459
0.456
0.452
0.451
0.447
0.439
0.436
0.425
0.421
0.420
0.400
0.400
0.396
0.393
0.373
0.340
0.319
0.285
0.248
0.211


1.14

4.70
5.31


* SALC Salix caroliniana
MYRC Myrica cerifera
BACH Baccharis halimifolia
ACER Acer rubrum
FACW Facultative wet trees Sambucus canadensis,
Ulmus americana var. floridana
FAC Facultative trees Liquidambar styraciflua,
virginiana, Prunus caroliniana.


Quercus laurifolia

Quercus nigra, Quercus


1.13



0.23
16.03

2.15
3.39
2.23
14.19
4.34


1.36
0.60 4.59
2.74 6.81
0.36
3.41
6.18
2.41
0.32
4.52
0.34
0.45


2.82


9.93


3.69 2.95
0.39


1.29 0.57
2.34
8.96 1.47
20.87 1.00
0.60
1.50
3.43


6.08
2.99
5.68
4.66
5.48
1.43
1.59
1.96
7.74








































1.00 0.99 0.90 0.a9 0.70 0.09 0.50 <.50
MOISTURE SITE INDEX
2' SALC MYRC BACH


0.99 0.90


0.89 0.70


0.e6 o.so


Z2 SALC


=2J MYRC


MOISTURE SITE INDEX
SBACH = ACER


CxM FACW


Mg IFAC


Figure 17. Tree dominance shown for moisture classes in younger and older
clay settling ponds. (a) 10- to 20-year-old clay settling ponds,
(b) ponds over 25 years old. See Tables 4 and 5 for abbreviations
of trees.


N=7 N-6 N-9 N-5 N 11

A



A

A

A A




A ffl /


N=4 N=10 N=6 N=14 N=28

B


zC


BC
-0

-0
- 9





,


19
18
17
18
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0


1.00


<.50


1


W.u'









general trends are apparent in the 10- to 20-year-old clay settling

ponds, trees did not show a significant difference (P > 0.05). However,

in ponds over 25 years since deactivated, differences (P < 0.05) are

seen. At the wettest sites, trees grow poorly because of too much water.

Here, cattail (Typha latifolia), primrose willow (Ludwigia peruviana), or

open water are dominant. A wetland site index between 0.70 and 0.99

favors growth of willows more than any other classification in both age

groups. The trees in this category are usually shrubby and much branched

at the base. They have grown considerably in size from the younger to

older sites. A wetland site index below 0.70 reduces willow dominance

and provides suitable conditions for the growth of other species.

Wax-myrtle (Myrica cerifera) and saltbush (Baccharis halimifolia) are

typical in younger sites while the older ponds include several species

common to bottomland hardwood forests such as red maple (Acer rubrum),

American elm (Ulmus americana var. floridana), and oak species especially

(Quercus laurifolia). Willows in these drier categories are usually

large, single stemmed, and often senescent.

The numbers of quadrats which occurred in each of the moisture

classes indicates a general trend of drier conditions over time.

Forty-two percent of the quadrats in the 10- to 20-year-old sites are in

the drier categories below 0.70 while 68% of the quadrats in the older

ponds meet this criterion. The two stage-filled sites IMC-N11C in the

younger sites and USS-HC in older ponds are much drier than the rest of

the clay settling ponds.

Importance values for herbaceous vegetation which were used in the

method to delineate moisture classes are given in Tables 6 and 7 for the

two different aged clay settling ponds. Some trends are worthy of








Table 6. Importance values for herbaceous vegetation used in the method
to delineate moisture classes for site 10 to 20 years old.



IMPORTANCE
VALUE SPECIES COMMON NAME CATEGORY*


INDEX 1.00
30.17 Typha latifolia
27.83 Salvinia rotundifolia
12.07 Ludwigia peruviana
8.22 Lemna minor
7.92 Juncus effusus
7.54 Azolla caroliniana
5.28 Salix caroliniana
0.98 Lythrum alatum

INDEX .90 TO .99
32.93 Ludwigia peruviana
23.78 Salix caroliniana
22.44 Lythrum lineare
5.00 Mikania scandens
4.27 Typha latifolia
3.41 Polygonum punctatum
2.80 Hydrocotyle umbellata
2.44 Cyperus odoratus
1.71 Pluchea purpurascens
1.22 Eupatorium capillifolium

INDEX .70 TO .90
35.66 Ludwigia peruviana
20.21 Mikania scandens
8.98 Typha latifolia
5.84 Polygonum punctatum
5.77 Cyperus odoratus
4.81 Imperata cylindrica
4.17 Baccharis halimifolia
2.89 Eupatorium capillifolium
2.57 Salix caroliniana
1.60 Hydrocotyle umbellata
1.28 Juncus effusus
1.28 Rubus betulifolius
0.96 Acer rubrum
0.96 Conyza parva
0.96 Clematis virginiana
0.77 Myrica cerifera
0.64 Momordica charantia
0.64 Pluchea purpurascens


cattail
water spangles
primrose willow
lesser duckweed
soft rush
mosquito fern
coastal plain willow
loosestrife


primrose willow
coastal plain willow
loosestrife
climbing hempweed
cattail
water smartweed
water pennywort
fragrant flatsedge
fleabane
dogfennel


primrose willow
climbing hempweed
cattail
water smartweed
fragrant flatsedge
cogon grass
eastern baccharis
dogfennel
coastal plain willow
water pennywort
soft rush
blackberry
red maple
dwarf horseweed
virgins bower
wax myrtle
balsam apple
fleabane


OBL
OBL
OBL
OBL
FACW
OBL
OBL
OBL


OBL
OBL
OBL
FACW
OBL
OBL
OBL
FACW
FACW
FACU


OBL
FACW
OBL
FACW
FACW
FACU
FAC
FACU
OBL
OBL
FACW
FAC
FAC
FACU
FAC
FAC
FAC
FACW








Table 6. Continued.


IMPORTANCE
VALUE SPECIES COMMON NAME CATEGORY*


INDEX .50
35.11
19.68
17.02
5.85
4.79
4.26
2.66
2.13
2.13
1.06
1.06
1.06
1.06
1.06
0.53
0.53

INDEX 0.0
42.16
26.91
14.49
6.03
3.73
2.66
1.48
0.71
0.59
0.41
0.41
0.35
0.06


TO .70
Imperata cylindrica
Ludwigia peruviana
Clematis virginiana
Thelypteris normalis
Salix caroliniana
Rubus betulifoilius
Mikania scandens
Hydrocotyle umbellata
Eupatorium capillifolium
Polygonum punctatum
Parthenocissus quinquefolia
Pluchia purpurascens
Juncus effusus
Azolla caroliniana
Lemna minor
Typha latifolia

TO .50
Imperata cylindrica
Thelypteris normalis
Clematis virginiana
Rubus betulifolius
Ludwigia peruviana
Lygodium japonicum
Ampelopsis arborea
Hydrocotyle umbellata
Salix caroliniana
Parthenocissus quinquefolia
Myrica cerifera
Eupatorium capillifolium
Juncus effusus


* See Table 2 for explanation of category type


cogon grass
primrose willow
virgins bower
maiden fern
coastal plain willow
blackberry
climbing hempweed
water pennywort
dogfennel
water smartweed
virginia creeper
fleabane
soft rush
mosquito fern
lesser duckweed
cattail


cogon grass
maiden fern
virgins bower
blackberry
primrose willow
climbing fern
peppervine
water pennywort
coastal plain willow
virginia creeper
wax myrtle
dogfennel
soft rush


FACU
OBL
FAC
FACW
OBL
FAC
FACW
OBL
FACU
FACW
FAC
FACW
FACW
OBL
OBL
OBL


FACU
FACW
FAC
FAC
OBL
FAC
FAC
OBL
OBL
FAC
FAC
FACU
FACW







76
Table 7. Importance values for herbaceous vegetation used in the method
to delineate moisture classes for sites over 25 years old.


IMPORTANCE
VALUE SPECIES COMMON NAME CATEGORY*


INDEX 1.00
46.39 Ludwigia peruviana
32.50 Lemna minor
11.77 Hydrocotyle verticillata
7.85 Cephalanthus occidentalis
1.57 Typha latifolia
INDEX .90 TO .99
37.16 Lemna minor
33.90 Salvinia rotundifolia
11.78 Bacopha sp.
8.16 Polygonum punctatum
4.53 Ludwigia peruviana
1.51 Vitis rotundifolia
1.03 Sambucus canadensis
0.66 Mikania scandens
0.66 Spirodela polyrhiza
0.60 Hydrocotyle verticillata
INDEX .70 TO .90
37.55 Sambucus canadensis
26.20 Polygonum punctatum
11.64 Ludwigia peruviana
8.01 Ampelopsis arborea
4.51 Lygodium japonicum
3.78 Clematis virginiana
3.64 Eupatorium capillifolium
1.75 Parthenocissus quinquefolia
1.46 Erectides styraciflua
0.73 Ulmus americana
0.58 Acer rubrum
0.15 Mikania scandens
0.00 Juncus effusus
INDEX .50 TO .70
32.39 Clematis virginiana
15.82 Ampelopsis arborea
12.38 Lygodium japonicum
5.99 Parthenocissus quinquefolia
5.74 Vitis rotundifolia
5.59 Drymaria cordata
4.49 Myrica cerifera
3.99 Valeriana scandens
1.50 Salvinia rotundifolia
1.35 Sambucus canadensis
1.25 Toxicodendron radicans
1.25 Erectidies styraciflua
1.25 Quercus laurifolia


primrose willow
lesser duckweed
whorled pennywort
button bush
cattail


lesser duckweed
water spangles

water smartweed
primrose willow
muscadine grape
American elder
climbing hempweed
big duckweed
whorled pennywort


American elder
water smartweed
primrose willow
peppervine
climbing fern
virgins bower
dogfennel
virginia creeper
fireweed
American elm
red maple
climbing hempweed
soft rush


virgins bower
peppervine
climbing fern
virginia creeper
muscadine grape
heartleaf drymary
wax myrtle
valerian
water spangles
American elder
poison ivy
fireweed
laurel oak


OBL
OBL
OBL
OBL
OBL


OBL
OBL
OBL
OBL
OBL
FAC
FACW
FACW
OBL
OBL


FACW
OBL
OBL
FAC
FAC
FAC
FACU
FAC
FAC
FACW
FAC
FACW
FACW


FAC
FAC
FAC
FAC
FAC
FAC
FAC
FAC
OBL
FACW
FAC
FAC
FACW








Table 7. Continued.


IMPORTANCE
VALUE SPECIES COMMON NAME CATEGORY*


INDEX .50
1.15
1.05
1.00
0.75
0.65
0.50
0.50
0.35
0.25
0.25
0.25
0.10
0.10
0.05
0.05
0.05
INDEX 0 TC
29.01
15.60
12.20
6.73
6.00
5.12
5.04
4.73
3.31
2.44
1.75
1.58
0.73
0.73
0.62
0.62
0.54
0.52
0.45
0.42
0.42
0.31
0.31
0.31
0.21
0.10
0.02


TO .70
Mikania scandens
Ulmus americana
Eupatorium capillifolium
Polygonum punctatum
Acer rubrum
Celtis laevigata
Baccharis halimifolia
Thelypteris normalis
Momordica charantia
Hydrocotyle umbellata
Ludwigia peruviana
Spirodela polyrhiza
Ambrosia artemisiifolia
Cornus foemina
Conyza parva
Lemna minor


Clematis virginiana
Thelypteris normalis
Lygodium japonicum
Rubus betulifolius
Parthenocissus quinqufolia
Toxicodendron radicans
Ampelopsis arborea
Vitis rotundifolia
Valeriana scandens
Erechtites hieracifolia
Ambrosia artemisiifolia
Momordica charantia
Imperata cylindrica
Lantana camera
Myrica cerifera
Sambucus canadensis
Drymaria cordata
Mikania scandens
Acer rubrum
Passiflora incarnata
Quercus laurifolia
Eupatorium capillifolium
Crotalaria rotundifolia
Baccharis halimifolia
Ulmus americana
Celtis laevigata
Lepidium virginicum


climbing hempweed
American elm
dogfennel
water smartweed
red maple
sugarberry
eastern baccharis
maiden fern
balsam apple
water pennywort
primrose willow
big duckweed
ragweed
gray dogwood
dwarf horseweed
lesser duckweed


virgins bower
maiden fern
climbing fern
blackberry
virginia creeper
poison ivy
peppervine
muscadine grape
valerian
fireweed
ragweed
balsam apple
cogon grass
lantana
wax myrtle
American elder
heartleaf drymary
climbing hempweed
red maple
passion flower
laurel oak
dogfennel
rabbit-bells
eastern baccharis
American elm
sugarberry
virginia pepperweed


*See Table 2 for explanation of category type


FACW
FACW
FACU
OBL
FAC
FACW
FAC
FACW
FAC
OBL
OBL
OBL
FACU
FACW
FACU
OBL


FAC
FACW
FAC
FAC
FAC
FAC
FAC
FAC
FAC
FAC
FACU
FAC
FACU
FACU
FAC
FACW
FAC
FACW
FAC
FACU
FACW
FACU
FACU
FAC
FACW
FACW
FACU









mention. In the frequently flooded category, cattails along with

floating aquatics are dominant in young ponds while primrose willow which

grows on less frequently flooded sites was more commonly found in older

ponds. As sites became drier, woody vines were dominant. This is

especially true for older sites where tree species were sparse and vines

often form impenetrable thickets. Dry sites in the younger ponds were

often dominated by cogon grass (Imperata cylindrica), which forms dense

stands excluding all other vegetation. This pest species has only

recently invaded central Florida and may further retard succession where

it becomes firmly entrenched.


Trees Organized by Seed Availability

When a seed source was nearby, hydric hardwood tree seedlings were

often a component of the herbaceous vegetation. Distance to seed source

and moisture regime appeared to be controlling factors in species

colonization. When quadrats with a site index less than 0.70 were

arranged as distance to seed source, a strong trend was seen for trees to

be larger and more numerous closer to the floodplain. Figures 18 to 21

show this relationship for four clay settling ponds located adjacent to a

floodplain seed source. Alderman Ford Ranch had larger trees with more

diversity than the other sites studied (Figure 18). It is also located

near a large floodplain and was older (35 years). Lake Hancock (Figure

21) is a stage filled pond and is estimated to have been in active use

more recently (20 years since decommissioned) than the other 3 sites, and

yet it had some of the largest trees. After the initial invasion by

early successional species such as willow and wax-myrtle, red maple is

the most frequent tree to colonize.













JU
28
28
24
22
20
18
16
14
12
10
8
6
4
2


50 70 100 125 150 200 225 250 300 400 425 475 500 525 700 725
DISTANCE FROM SIZEABLE SEED SOURCE (M)
SLS ( PC 3 CL E QS = UA B AR


PLANT SPECIES:
Liquidambar styraciflua (sweetgum)
Prunus caroliniana (cherry)
Celtis laevigata (hackberry, sugarberry)
Quercus spp. (oaks) includes Q. laurifolia, Q. nigra
Ulmus americana var. floridana (elm)
Acer rubrum (red maple)


Figure 18.


Hardwood tree species measured at Alderman Ford Ranch clay
settling pond (greater than 35 years since active use).
Quadrats are arranged in distance from the floodplain forest
of the Alafia River. Early successional species such as
willow and myrtle were usually present but are not included.
Blank spaces indicate no later successional tree species were
present in quadrat.


KEY FOR
LS
PC
CL
QS
UA
AR













10
17
16
15
14
13
12
11
10
9
a
7



4
3




100 175 200 225 300 600 625 700 725
DISTANCE FROM SIZEABLE SEED SOURCE (M)
7 LS PC eZ CL On as = UA AR


KEY FOR
LS
PC
CL
QS
UA
AR


PLANT SPECIES:
Liquidambar styraciflua (sweetgum)
Prunus caroliniana (cherry)
Celtis laevigata (hackberry, sugarberry)
Quercus spp. (oaks) includes Q. laurifolia, Q. nigra
Ulmus americana var. floridana (elm)
Acer rubrum (red maple)


Figure 19.


Hardwood tree species measured at Maine Ave. clay settling
pond (greater than 30 years since active use). Quadrats are
arranged in distance from the Saddle Creek floodplain forest.
Early successional species such as willow and myrtle were
usually present but are not included in the graph. Blank
spaces indicate no later successional tree species were
present in the quadrat.




































50 75 150 175 275 300 575 600 700
DISTANCE FROM SIZEABLE SEED SOURCE (M)
7 LS K PC ET CL SM QS = UA AR


KEY FOR PLANT SPECIES:
LS Liquidambar styraciflua (sweetgum)
PC Prunus caroliniana (cherry)
CL Celtis laevigata (hackberry, sugarberry)
QS Quercus spp. (oaks) includes Q. laurifolia,
UA Ulmus americana var. floridana (elm)
AR Acer rubrum (red maple)


Figure 20.


Hardwood tree species measured at 0. H. Wright clay settling
pond (greater than 25 years since active use). Quadrats are
arranged in distance from Whidden Creek floodplain. Early
successional species such as willow and myrtle were usually
present but are not included in graph. Blank spaces indicate
no later successional trees were present in quadrat.


Q. nigra





































50 60 75 100 150 175 200
DISTANCE FROM SIZEABLE SEED SOURCE (M)
SLS PC CL QS 2 UA AR




KEY FOR PLANT SPECIES:
LS Liquidambar styraciflua (sweetgum)
PC Prunus caroliniana (cherry)
CL Celtis laevigata (hackberry, sugarberry)
QS Quercus spp. (oaks) includes Q. laurifolia, Q. nigra
UA Ulmus americana var. floridana (elm)
AR Acer rubrum (red maple)


Figure 21.


Hardwood tree species measured at a clay settling pond
adjacent to the south side of Lake Hancock (greater than 20
years since active use). Quadrats are arranged in distance
from the floodplain. Early successional species such as
willow and myrtle were usually present, but are not included
in graph.









Seedlings for Enhancing Cypress-Gum Succession


Cypress-gum swamp seedlings were planted in wet depressions of clay

settling ponds to test the theory they were suitable to follow willows,

but were not found, because their seeds were unable to reach the site in

great enough quantity for germination. Data on water levels and soil are

presented first. Then data on seedling performance are given and related

to water regime and soil.


Patterns in Water Level Fluctuations

Results of the monthly water level readings are shown in Figures 22

to 26. Levels dropped dramatically in May, the normal dry season in

central Florida which occurs after the time of normal frontal storms in

winter. Water levels quickly rebounded in June with the onset of summer

convectional activity. Highest water levels were in late August

following summer rains, but varied between sites because of local rains

and differing characteristics between ponds. At Tenoroc dry sites

(Figure 22) water levels were 20 to 75 cm below land surface except for

one drier transect where the water table was over a meter below the

ground. At this well and several others, the further the water table was

below the surface, the greater the tendency for levels to fluctuate

widely. Tenoroc wet sites had wells at either end of a transect which

ran perpendicular to a small lake.

Hydrology at CF Industries (Figure 23a) did not follow the general

pattern found at other sites. Water levels did not fall in May and

instead increased steadily during the study. Reclamation activity during

the spring of 1986 knocked down dikes and smoothed the material around

the perimeter of the site. This activity extended for about 300 m into















-10
-20
-30
-40
-SO -
-60
-70
-80-
-90
-100-
-110
-120-
-130
-140
-150-
-160 -
-170
-180
-190
-200
OCT


o 2 TB (5)

o 1 TB 5)


APR MAY JUN
GROWING SEASON 198a-86
A 2 TB (25)


I
MAY
SEASON
+4


Figure 22.


Water levels measured in wells at Tenoroc State Reserve for
1985-85 growing season. (a) Wells in drier transects. (b)
Wells in transects planted perpendicular to a pond. (Wells
identified by transect number, seedling type [bareroot or
tubeling], and distance in meters from beginning of transect
in parenthesis. Exact well locations can be determined by
looking at the site plans.)


MAR
MAR


x TB(25)


AUG


94 TB (10)


x 3 T8 (25)


0 -

-20 -

-40 -
-G-
-60 -

-80 -

--00 -


-120-

-140 -

-160-

-1a0 -

-200 ---
OCT


MAR APR
GROWING
5 S BR (5)


JUN
19 85-a
5 BR (25)


AUG


I


~---- c~-



r














-W

20

0

-20

-40

-60
-ao
-80

-100

-120

-140

-160

-180

-200


5 BR (0)


GROWING SEASON 1985-80
SS BR (25) o 2 BR (0)


A 2 BR (30)


20 -

10

0 --

-10

-20

-30 -

S-40-

-so50

-60

-70-

-80ao -

-90 -

-100 --
OCT


WATER LEVEL IN WELLS
O. H. WRIGHT WET SITES


MAR APR MAY JUN JUL AUG


1 TB (25)

Figure 23.


GROWING SEASON 1985-86
+ 1 BR 25 2 BR (15)


Water levels measured in wells during the 1985-86 growing
season. (a) Water levels at CF Industries show a gradual
increase during the monitoring period attributed to
reclamation activity. (b) Water levels at 0. H. Wright
continue to decline in June and July unlike other sites.


AUG


A 2 BR (0)

















-0.4

--0.5

-0.6-

-0.7

-0.8


-1

-1.1

-1.2

-1 .


-1.5
--1.6


FEB MAR APR MAY JUN JUL SEP OCT NOV DEC JAN FEB APR JUN JUL

1985 1l88
a SWALE 2 + SWALE 1


-0.1
-0.2

-0.3
--0.4
-0.5

-0.6 -
--0.7
-0.8 -

-0.9
-1 -
-1.1 -

-1.2


-1.4.
-1.5-
-i.e ---- ---- ---- ---- ---
FEB MAR APR MAY JUN JUL


0 SWALE 2 WEU..6


SEP OCT NOV DEC JAN FES A
1985 1986
-+ SWALE 1 WEL.LS5


-r JUN JUI
PR JUN JUL


Figure 24.


Water levels measured in wells at Alderman Ford Ranch during
1985-86. Hydroperiod fluctuates more widely in drainage
swale 2 than in swale 1 next to the dike. See Figure 7 for
well locations. Data for the north well are not shown. The
wells in (b) are to the south of (a).


I _


*