The development and role of microtopography in constructed forested wetlands on phosphate-mined lands in central Florida

THE DEVELOPMENT AND ROLE OF MICROTOPOGRAPHY
IN CONSTRUCTED FORESTED WETLANDS ON PHOSPHATE-
MINED LANDS IN CENTRAL FLORIDA

By

BENJAMIN J. BUKATA III

A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

1999

ACKNOWLEDGEMENTS

I would like to thank Dr. M.T. Brown for his insight and perspective into this

study and for his guidance through my graduate career. I would also like to thank my

faculty committee members, Dr. T. Crisman and Dr. S. Doherty, for their guidance and

support.

This study was supported by the Florida Institute of Phosphate Research. I would

like to acknowledge private individuals at several mining companies who provided

access as well as assistance with site selection and background information: John Keifer

at CF Industries, Vance Pickard and Bill Hicks at IMC Agrico, and Rosemarie Garcia at

Cargill Inc.

I would like to thank fellow students at the Center for Wetlands for their

assistance with this study and support during my graduate career. In particular, Eliana

Bardi, Susan Carstenn, and Kristina Jackson provided countless hours of fieldwork

assistance and mental support for the numerous "judgement calls" that arose. Their

efforts helped make this study possible. I would also like to thank Matt Cohen and Susan

Carstenn for technical assistance during the simulation model development. Finally,

Eliana Bardi and my family supported me in many ways and made completion of my

graduate studies possible.

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS...............................................................................ii

LIST OF TABLES .................................................................................................vi

LIST OF FIGURES ..........................................................................................vii

ABSTRACT .....................................................................................................ix

INTRODUCTION ...........................................................................................1

Review of the Literature ....................................................... 2
Development of Microtopography in Natural and Constructed
Forested Wetlands...............................................................2
Role of Microtopography in Natural and Constructed Forested
Wetlands ......................................................................4
Created Microtopography (Hummocks) in Constructed
Forested Wetlands .................................................. .............6
Plan of Study ..............................................................................................7

M ETH O D S ............................................................................................................9

Description of Study Sites .............................................................. .......9
Microtopographic Development Sites.................................. ........... 9
Constructed Hummock Sites ............................................. ............. 14
FG-GSB2 ............................................................................15
Hal Scott ...........................................................................15
Field Data Collection-Microtopographic Development............................. 18
Ground Surface Elevations................................ ...... ............. 18
V egetation................................................... ..............................18
Canopy Cover ................................................................................. 19
Field Data Collection--Constructed Hummocks..........................................19
Ground Surface Elevations...........................................................19
V egetation.......................................................................................23
D ata A analysis .............................................................................................23
Vegetation Data ..................................................... ............... 23
Microtopographic Development .......................................... 23
Constructed Hummocks......................................................25
Microtopography Data ..................................................................27

iii

Page

Statistical Analysis ........................ .. ..... ..............................27
Simulation M odel ..................................................... ....................28

R E SU L T S....................................................................... ..................................29

Microtopographic Development ........................................29
Evaluation of Constructed Hummocks.....................................................35
IMC Agrico's FG-GSB2 Site........................................................35
V egetation ........................................................................... 35
Structural Characterization ..................................................40
IMC Agrico's Hal Scott Site ....................... ...............................46
V egetation .................................................. ......................46
Structural Characterization....................................... ...49
Simulation M odel ....................................... ........... ..........................50
M odel O verview ......................................................................... 50
Energy Source................................ .... .................. 50
Inundation and Hummock Storages ....................................50
Production and Producer Units ............................................ 50
Calibration and Simulation...........................................................53
Simulating Hummock Benefit........................... .......................54
Simulation Results .............................................. 55
Hum m ock Benefit......................................................................... 55

D ISCU SSION ........................................... ........................... .............................. 61

Microtopographic Development and Site Age........................................... 61
Microtopographic Development and Dominant Canopy Vegetation............62
Microtopographic Development in Natural and Constructed Forested
W etlands........................................................................................... 64
Species Richness and Divesity in Systems with Constructed Hummocks ....66
Species Richness and Diversity on Constructed Hummocks ......................66
Nuisance Species and Constructed Hummocks ........................................... 67
Structural Characterization of Constructed Hummocks.............................68
Simulation M odel .............................................. .. .................. 71
Limitations of the Study............................................. .....................71
Summary and Recommendations................................................................73

A PPEN D ICE S ......................................... ........... ..... ...... ..................... 75

A Summary of Microtopographic Development Data for the Twelve
Constructed System s............................ .............................................75

B Description and Evaluation of Pathway Flows, Storages, and Coefficients
Used in Simulation Mini-model ....................................................82

Page

LITERATURE CITED ..........................................................................................85

BIOGRAPHICAL SKETCH ..................................................................................90

LIST OF TABLES

Table Page

1 Classification of wetland plants in Florida ..........................................26

2 Summary of microtopographic development data for the twelve surveyed
sites ............................................................. .................................30

3 Summary of community descriptive indices comparing entire transects
intersecting a created hummock (ON) and sections of the same
transects which were only in adjacent wetland areas between
hummocks (OFF)................................................................. ............38

4 Community descriptive indice comparison between sections of transects
only on a constructed hummock and sections of the same transects
only in adjacent wetland areas between hummocks.............................39

5 Inventory and wetland classification of identifiable plant species found
exclusively on constructed hummocks at FG-GSB2............................41

6 Frequency of occurrence of nuisance species on and off constructed
hummocks at FG-GSB2 and Hal Scott....................................... ...42

7 Calculation table deriving average constructed hummock height at
FG-GSB2 and Hal Scott......................................................................45

8 Mean hummock and water level elevation for FG-GSB2 and Hal Scott.
Elevations are relative to the height of the laser level. Numbers in
Parenthesis represent hummock elevation relative ................................47

9 Inventory and wetland classification of identifiable plant species found
exclusively on constructed hummocks at Hal Scott.............................48

LIST OF FIGURES

Figure Page

1 Map showing locations of sampled constructed forested wetlands. Figure
not to scale..................................... ................. ..............................10

2 Plan view ofIMC Agrico's FG-GSB2 wetland showing topographic contours.
Constructed hummocks are shown in gray. Figure not to scale............. 16

3 Post reclamation plan view of Hal Scott showing forested wetland area
and location of constructed hummocks..........................................17

4 Example of a digitized photo in MAP FACTORY used to estimate
average percent canopy cover ................................................. .........20

5 Sampling method used in the collection of vegetative and elevation data
on large constructed hummocks at FG-GSB2.
a) Side view showing method of collecting elevation data; b) Plan
view showing frequency of sampling along elevation gradient .............21

6 Sampling method used to collect vegetative and elevation data on small
constructed hummocks at Hal Scott.
a) Side view showing method of collecting elevation data; b) Plan
view showing frequency of sampling along elevation gradient..............22

7 Microtopographic relief along an average transect.
a) Lizard Branch; b) FG-GSB2 .......................................................31

8 Relationship between site age and the rugosity index ofmicrotopographic
heterogeneity for the twelve systems surveyed......................................33

9 Comparison between systems having different dominant canopy
vegetation and the rugosity index of microtopographic heterogeneity....34

10 Relationship between the rugosity index of microtopographic
heterogeneity and community descriptive indices in systems having
an average canopy coverage less than 50 percent.
a) Species richness; b) Species diversity............................... .....36

Page

11 Relationship between the rugosity index of microtopographic
heterogeneity and community descriptive indices in systems having
an average canopy coverage greater than 70 percent.
a) Species richness; b) Species diversity........................................37

12 Typical cross section of hummocks showing heights and water levels
where constructed hummocks were evaluated.
a) FG-GSB2; b) Hal Scott.................................................................44

13 Complex systems diagram of a constructed forested wetland ....................51

14 Aggregated systems diagram of a constructed forested wetland used in
the simulation model............................ ............................................. 52

15 Interaction between planted wetland tree and nuisance species biomass
under calibrated conditions in the simulation mini-model....................56

16 Simulation results of the model in response to various I:H ratios.
a) Planted wetland trees (T); b) Nuisance species (N)..........................57

17 Relationship between the amount of hummock area in a wetland and the
hummock benefit index.
a) Under steady state conditions; b) For five different transformity
weighting factors multiplied to I and B in the hummock benefit index..58

18 Comparison of natural and constructed forested wetland systems using the
rugosity index of microtopographic heterogeneity................................65

Abstract of Thesis Presented to the Graduate School
Of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

THE DEVELOPMENT AND ROLE OF MICROTOPOGRAPHY
IN CONSTRUCTED FORESTED WETLANDS ON PHOSPHATE-
MINED LANDS IN CENTRAL FLORIDA

By

Benjamin J. Bukata III

May 1999

Chairman: Mark T. Brown
Major Department: Environmental Engineering Sciences

In this study, relationships between microtopographic heterogeneity and

community descriptive indices were investigated in 12 constructed forested wetlands of

various ages in Central Florida. The floristic and structural characteristics of two

constructed forested wetlands containing created hummocks were also studied. Ground

surface elevation, vegetation, and canopy cover data were collected on line transects.

Microtopographic heterogeneity was quantified using an index ofrugosity or

wrinkledness in a plane. A computer simulation model was developed to test the

hypothesis that there is an optimum amount of hummock area in a forested wetland,

beyond which, total value (water storage and gross production) is lower.

No strong relationships were found between rugosity values and site age.

Constructed forested wetlands of all ages had significantly lower microtopographic

heterogeneity than natural systems. There did not appear to be a correlation between

rugosity values and species richness or diversity of vegetation in wetlands having average

canopy cover less than 50% or greater than 70%. Rugosity values were influenced by

dominant canopy vegetation. Sites dominated by early successional species, primrose

willow (Ludwigiaperuviana) and carolina willow (Salix caroliniana), had significantly

greater rugosity values than systems dominated by other canopy vegetation species.

Constructed hummocks at both sites significantly contribute to the overall species

richness and diversity in wetlands constructed on phosphate-mined lands. Hummocks

provided sites for a diverse community of plant species, yet species deemed "nuisance" or

"undesirable" by the regulatory community were not commonly established. Species

richness and diversity was higher on hummocks than in areas between hummocks.

Frequency of occurrence of nuisance species on hummocks was 0.07 and 0.03, while it

averaged 0.13 and 0.23 in areas between hummocks. Additionally, constructed

hummocks in these systems appear to provide sites which may allow for increased

survivorship and growth rates of planted wetland tree stock by elevating them above

surrounding floodwaters. Results from a computer mini-model indicate that

incorporating hummocks as 20 percent of the surface area of a constructed wetland

provides the maximum benefit in terms of water storage and biomass production.

INTRODUCTION

Small-scale variations in forest floor topography, referred to as microtopography,

have been associated with the maintenance of species diversity within forested wetland

ecosystems (Beatty 1984). Microtopography is the result of hummocks, elevated

mounds, and hollows, low depressional areas of the forest floor. The formation of

microtopography has been attributed to a variety of causes such as wind thrown trees,

fallen tree limbs, and litter fall accumulation around tree trunks, root collars, shrub stems,

or cypress knees (Titus 1987). In the case of wind-thrown trees, the hummocks and

hollows that develop provide new colonization sites for plants. The resulting decaying

logs, pits and mounds can also be important colonization sites because they are initially

free of potential above or below ground competition from adjacent plants.

Microtopographic heterogeneity in forested wetlands results in a diversity of

microhabitats and subsequent variations in edaphic conditions. This provides a wide

variety of conditions available for both canopy tree species and understory vegetation

establishment and regeneration. Increased species diversity is often associated with

greater environmental heterogeneity, thus microtopography may be critical in promoting

and perpetuating diverse forested wetland systems.

It has been shown in numerous studies (Bell 1974, Chimner and Hart 1996, Davis

et al. 1991, Golet et al. 1993, Parately and Fahey 1986, Sloan 1998, Titus 1987) that

forested wetlands have microtopography and that it is an important element in the

structural and functional organization of wetlands. However, there is a dearth of

information concerning microtopographic development within newly constructed

wetlands or how it influences self-organization and vegetative community structure. In

Florida, many wetlands are constructed during the land reclamation process following

phosphate mining. These wetlands offer the opportunity to study the development and

role of microtopography in newly constructed systems. Several important questions arise

from observations of the importance of microtopography in forested wetland ecosystems

including:

1. Are forested wetlands constructed on phosphate-mined lands developing

microtopography? Is there a correlation between microtopography and site age or

species richness and diversity?

2. Does vegetation influence microtopographic development? Is there a difference in

microtopographic development within systems dominated by different vegetation?

3. Does microtopography influence species richness or diversity in constructed

wetlands?

4. Does microtopography provide sites and therefore increase establishment of nuisance

species in newly constructed wetlands?

Review of the Literature

Development of Microtopography in Natural and Constructed Forested Wetlands

The development of microtopography has been attributed to a variety of causes.

Subsequently, microtopographic heterogeneity can vary considerably from large, greater

than 1 m differences due to tree falls, to very small scale variation, 1-5 cm, resulting from

animal and vehicle tracks. Microtopographic heterogeneity in wetlands is frequently

caused as a result of disturbance (Ehrenfeld 1995). Examples of such disturbance include

tree falls, channel building of muskrats, differential patterns of litter accumulation and

erosion following a flood event, and animal or vehicle tracks. Microtopography often

forms as a result of natural processes such as accumulation of organic material around the

base of large trees or cypress knees, adventitious rooting above high water level, and

mechanical breakage of tree structure. Microtopography can also form due to

anthropogenic disturbances such as logging which results in depressed tire tracks and tree

stumps. In a study of microtopography in seven forested wetlands in North Central

Florida, Sloan (1998) found hummocks created from tree stumps after logging provided

large scale relief in systems with long hydroperiods.

One of the most common mechanisms for the formation ofmicrotopography cited

in the literature is due to wind-thrown trees (Beatty 1984, Ehrenfeld 1995, Golet and

others 1993, Lowry 1989). Wind-throw often occurs as a result of lightning strikes, old

age, disease, fire, or poorly anchored root systems (Wharton et aL 1977). Trees growing

in swamps are generally more shallow rooted than those in upland ecosystems and as a

result, they are more susceptible to wind-throw. Wind-throw results in a raised mass of

root and soil which, as the woody material decays, settles into a mound. Wind thrown

trees also produce variations in the wetland topography by providing downed biomass in

the form of limbs and tree trunks as well as the residual stumps. This woody debris is

often a large component of the organic matter pool in many forested wetlands and may

ultimately create drier conditions for seedling establishment and serve as a long term

reservoir in nutrient dynamics of a system (Goodall 1990).

While the formation of microtopography has been attributed to a variety of

causes, the main mechanisms are likely wind thrown trees and fallen woody debris (Titus

1987). However, in constructed wetland systems wind throw and fallen limbs or trunks

are absent for many years until a relatively mature canopy forms.

Role of Microtopography in Natural and Constructed Forested Wetlands

By creating small scale elevation variance in wetland ecosystems,

microtopography introduces variability in hydroperiod and depth of inundation. As a

result, microtopography can influence the spatial patterns and survivorship of individual

plant species (Harper et al. 1965; Sheldon 1974; Hamrick and Lee 1987; Eldridge et al.

1991) as well as effect the composition of wetland flora (Schlesinger 1978; Collins et al.

1982; Hardin and Wistendahl 1983; Huenneke and Sharitz 1986; Titus 1987; Ehrenfeld

1995). This is due to its effect on germination success and subsequent establishment as it

relates to seed placement and soil topography. Collins et al. (1982) found the distribution

of woody plants in a bog was strongly correlated with elevated microtopographic

positions. Stumps and logs found in deep-water hardwood swamps of the southeastern

U.S. are often colonized by distinctive vegetation (Dinnis and Batson 1974, Irwin 1975).

In Georgia's Okeefenokee Swamp, Schlesinger (1978) reported shrub species were

restricted to stumps and other emergent microsites. Vivian-Smith (1997) found that

small-scale variability in microtopography, on the order of only 1-3 cm, resulted in

significant differences in vegetation communities in experimental wetland mesocosms.

The mesocosms containing hummocks and hollows had greater species diversity,

richness, and evenness. Additionally, Vivian-Smith (1997) found that more species,

particularly rare woody perennials, favored drier sites on hummocks. Bragazza et al.

(1998), in a study in an Italian peat bog, reported well differentiated floristic communities

for hummock and hollows due to the differences in edaphic conditions. Hummocks were

drier, had lower pH and electrical conductivity values, and cation concentrations

underwent greater variations in hummocks than hollows.

Microtopography can affect the survival of emerged seedlings due to changes in

soil moisture status, bulk density, and infiltration rates (Eldridge et al. 1991). Goodall

(1990) found Mitragyna stipulosa tree seedlings grew preferentially on the top of mounds

formed in an African swamp forest clearing. This led to a zonation of species based on

elevation within this wetland. Komiyama et al. (1996) found that an elevation difference

of only 35 cm greatly affected the survival and growth rate of tree seedlings due to the

variation in hydroperiod and resulting edaphic conditions.

Microtopography, specifically hummocks, can also play an important role in the

ability of a system to regenerate. In a study of white cedar (Thuja occidentalis)

regeneration, Chimner and Hart (1996) determined that the majority of trees were found

growing on hummocks since they were often the only sites which provided unsaturated

soil conditions necessary for seedling establishment. This suggests that hummocks play

an important role in the successful regeneration of cedar in fen peatlands. In a study of

seedling distribution of six wetland tree species in a southeastern cypress-tupelo swamp,

Huenneke and Sharitz (1986) found that seedlings were found disproportionately on

emergent microsites which provided germination sites above summer water levels. In a

study ofmicrotopography and woody plant regeneration in a North Florida floodplain

swamp, Titus (1987) found that most shrub and tree seedlings occurred on hummocks

and in the spring, hummocks were carpeted by newly germinating seedlings.

Another important role ofmicrotopography in wetland systems is its influence on

surface water. Microtopographic heterogeneity influences water storage and movement.

Depressional areas, referred to as hollows, common to wetland systems act to store water

while hummocks impede overland flow away from a site (Rheinhardt et al. 1997).

This in turn affects water residence time and may ultimately influence water quality in

the wetland and downstream receiving bodies.

Created Hummocks (Microtopography) in Constructed Forested Wetlands

Microtopography is especially important on these newly constructed low relief

landscapes because it interacts with and influences the hydrology throughout the wetland,

creating a multitude of microsites (landscape heterogeneity) for understory plant and tree

seedling establishment (Barry et al. 1996). Small-scale microtopographic relief in the

form of constructed hummocks also increases the probability that established plants will

survive anoxic edaphic conditions and other stresses induced by a fluctuating hydrologic

regime.

Depth of floodwater is especially critical for seedlings and herbaceous species.

Water levels in newly constructed systems can often cover a substantial portion of the

wetland, resulting in significant stress or mortality to planted species (Teskey and

Hinckley 1977). In addition to providing sites for the regeneration of wetland tree

species (Chimmer and Hart 1996; Huenneke and Sharitz 1986; Titus 1987), hummocks

may be especially important in wetlands that develop long hydroperiods or high water

levels by providing elevated sites for planted trees. Thus, reducing the risk of high tree

mortality and the subsequent need for re-plantings. Hummocks, therefore, may provide

excellent sites for enhanced sapling tree growth and survivorship in newly constructed

forested wetlands where hydrologic conditions are still evolving.

Constructed hummocks also create microsites that help herbaceous understory

vegetation avoid anaerobic conditions associated with long hydroperiods (Barry et al.

1996). In addition, hummocks aid in seed "trapping" and retention, a particularly

important process in systems where water dispersal of seeds is common (Huenneke and

Sharitz 1986). By incorporating microtopography as hummocks into the wetland

construction process, a natural pattern of species recruitment and community

development is favored as the site matures and the chances of success in the reclamation

or mitigation project are increased (Barry et al. 1996).

Very little literature exists addressing the incorporation of microtopography and

subsequent plant community development and organization in constructed wetland

projects. At a site reported by Barry et al. (1996), a "mound and pool" topography was

constructed in a large mitigation project in New Hampshire. The design required

intensive earth-moving in order to create a dense arrangement of hummock and hollows

which may make such designs economically unfeasible. Hummocks were constructed

approximately 0.3 m above final design grade and 4.9 m wide with a 6.1 m on-center

hummock spacing. Pools were excavated to approximately 0.3 m below final grade

which matched the fill needed for hummock construction, thereby eliminating the need

for additional fill or offsite disposal of material.

Plan of Study

In this thesis, constructed forested wetlands on phosphate-mined lands in Central

Florida were studied to provide insight into the role and development of

microtopography. Twelve constructed forested wetlands of various ages having different

dominant canopy vegetation were surveyed. Elevation, vegetation, and canopy cover

data were collected to study: (1) development of microtopography in newly constructed

wetlands, (2) relationships between plant community and microtopographic

heterogeneity, and (3) vegetation affects on microtopographic development.

Two constructed forested wetlands having hummocks built during site

recontouring were surveyed to examine the structural and floristic characteristics on and

off constructed hummocks. At these sites, elevation and vegetation data were collected

to study (1) the contribution of hummocks to species richness and diversity in these

systems, (2) the potential of hummocks to increase the establishment of nuisance species,

and (3) the structural characteristics of these hummocks.

In addition, simulations of a computer mini-model were used to study forested

wetland succession in a constructed wetland with varying ratios of created hummocks

and inundated areas. An index was generated to study benefits of these ratios.

METHODS

In this study ofmicrotopography in wetlands constructed on Central Florida

phosphate mined lands, field sampling of ground elevations (microtopography),

vegetation, and canopy cover was conducted in twelve constructed forested wetlands of

various ages. Two constructed wetlands that had hummocks incorporated into the

recontouring phase were also studied. In these systems ground elevations and vegetation

were sampled. Site descriptions and detailed methods are given in the sections that

follow.

Description of Study Sites

All sites are located in the phosphate mining district of Central Florida (Figure 1).

Microtopographic Development Sites

Twelve constructed forested wetland systems of various ages were surveyed. Sites

were located on properties owned by IMC Agrico, CF Industries, and Cargill

Incorporated. FG-GSB2, which has constructed hummocks, was surveyed for the

purpose of both facets of this study. Site ranged in age from 2 to 28 years. Three of the

sites were constructed floodplain swamps, 1 was a lake border swamp, and 8 were still-

water wetlands.

AREA ENLARGED

YACXSONVILLE
* FFLORIDA

.00

SEast Lobe. West Lobe,
R6, R7, R9

] FG-GSB2
SHall's Branch, Lizard
SBranch, Jameson Jr.
E PR 1 (Section 12)
Parcel B

SHal Scott
[']sP6

Figure 1. Map showing locations of sampled constructed forested wetlands. Map not to scale.

IMC Agrico

Parcel B. Still-water wetland located south of Bartow, FL at the Clear Springs

Mine. The total site area is approximately 20 ha of which 4 ha is wetland. Mining was

completed in March 1968 and contouring was completed in November 1978. Overburden

was used to backfill this site and the site was not mulched. The primary water source to

the wetland is groundwater, surface water runoff, and minor amounts of flooding from the

Peace River. Revegetation with numerous wetland tree and 5 herbacous species was

completed in May 1979. At the time of sampling this site had an average canopy coverage

of 88 percent.

East of the Peace River (Section 1). This approximately 10 ha lake border

swamp is located south of Bartow, FL and just north of Highway 640 at the Clear Springs

Mine. Site was constructed in approximately 1970 using overburden as the primary

backfill. Site was not mulched. Approximately 1,000 trees/ha and no herbacous species

were planted in 1970. At the time of sampling this site had an average canopy coverage of

85 percent.

Jamerson Junior. A 1.3 ha forested wetland constructed in 1984 and located at

the Lonesome Mine. This site was considered a floodplain swamp due to seepage from

the west side of the site that converges into a small stream flowing through the system.

Site was back-filled with overburden fill consisting of clayey fine sand devoid of organic

matter. Stockpiled topsoil from the mined wetland was spread on the site as the final

reclamation task. In August 1990, 3,900 indigenous tree seedlings were planted.

However, 25 percent of the site was not planted due to its shallow inundation. Much of

this site remained permanently inundated due to ground water recharge allowing only

modest tree survival. In 1992 work began to raise the forest floor approximately 0.61 m

and reduce slopes of the valley walls. At least 70 percent of the site was affected. In

March 1993, red maple seeds were scattered over the land and 6,600 tree seedlings of 8

indigenous species were planted. In April and again on August 13, a few individuals of

several herbaceous species were transplanted from adjacent forests to the site in order to

replicate control wetlands. In June, approximately 2,500 herbaceous plants were planted

to stabilize the substrate and compete with cattails. At the time of sampling, this site had

an average canopy coverage of 48 percent.

This site along with Hall's Branch and Lizard Branch were released by the

regulatory agencies and incorporated into the Hillsborough River State Park.

Hall's Branch. This 1.5 ha still-water wetland is located in southeastern

Hillsborough County at the Lonesome Mine. Hall's Branch and the majority of its

watershed were mined in 1983. In 1984 the site was back-filled to approximately original

grade with sand tailings and overburden. In March 1985, topsoil from a donor wetland

was spread to a depth of 1 to 2.5 cm. Tree planting took place in two phases.

Approximately 2,500 containerized trees were planted in June 1985 and in July 1988. In

March and April 1989, the site was planted with 24 understory herbacous species

transplanted from a donor site. At the time of sampling this site had an average canopy

coverage of 86 percent.

Lizard Branch. This 2.5 ha still-water wetland was constructed in late spring

ofl992 and is also located at the Lonesome Mine. Due to a lack of additional fill

materials, Lizard Branch was back-filled with overburden rather than permit approved

sand/clay mix. This site is an extension to the 25 year floodplain of the Alafia River's

South Prong.

A minimum of 15 cm of muck was spread over the site. Stumps, logs, and clusters

of rocks were also placed on site in November 1993 after FDEP approval. Planting of

tree and herbacous species was completed in January of 1994. Trees were planted at a

density of 1,700 per ha and approximately 3.000 herbacous species were also planted.

CARGILL INC.

SP-12. This still-water wetland, located at the Ft. Meade Mine near Ft. Meade,

FL in southwestern Polk County was constructed for the purposes of testing design

principles. An array of wetland communities were designed to attenuate and cleanse

discharge waters from an adjacent clay settling pond before final discharge to Whidden

Creek.

The site consists of approximately 4 ha of wetlands, a 0.8 ha lake, and 1.7 ha of

uplands. Recontouring was completed in 1982. Mulch was spread on wetland areas, dead

wood was placed in numerous piles to enhance wildlife habitat, and snags were "planted".

Replanting of containerized, tubling, and bare-root trees was conducted throughout the

site in September of 1983 and May of 1984. At the time of sampling this site had an

average canopy coverage of 87 percent.

CF INDUSTRY

All sites are located at CF Industries Garwood Mine Complex in Hardee County.

West Lobe. Site was mined in 1978 and used as clay settling area. Site was back-

filled with a clay/sand mix until 1986. This still-water wetland which developed was not

mulched and allowed to naturally re-vegetate. Understory herbaceous species were

planted in 1996. Natural recruitment from an adjacent un-mined riverine system has

allowed this site to become dominated by Acer rubrum. At the time of sampling this site

had an average canopy coverage of 88 percent.

East Lobe. This still-water wetland is approximately 150 m east of West Lobe. It

has the same site background and has been managed in the same manner as West Lobe

(SP-1). At the time of sampling this site had an average canopy coverage of 79 percent.

R-7. Stream system constructed in 1984 that received seepage from a large

adjacent clay settling pond. Site was back-filled with sand tailings and overburden and

then mulched. Approximately 1500 trees per ha were planted in 1985. No herbacous

plants were planted. At the time of sampling, this site had an average canopy coverage of

60 percent.

R-9. This site is a continuation of R-7. It was constructed in 1985, using sand

tailings and overburden, and planted in 1987 at a density of approximately 1500 trees/ha.

Site was also mulched and no understory plants were planted. At the time of sampling,

this site had an average canopy coverage of 45 percent.

R-6. Still-water wetland located just west of R-7 and R-9. Site was constructed

in 1993 using a sand-clay mix and was not mulched. Approximately 1500 trees/ha were

planted at this time. No herbacous understory plants have been planted. At the time of

sampling this site had an average canopy coverage of 72 percent.

Constructed Hummock Sites

Two constructed still-water forested wetlands having hummocks built during the

recontouring process were surveyed; IMC Agrico's FG-GSB2 and Hal Scott (Section 12).

FG-GSB2. Figure 2 provides a plan view of this site that includes topographic

contours and hummock locations. This 16 ha system is located in northwest Hardee

County. Site was constructed in 1995 using overburden and then mulched in the same

year. In 1996, wetland tree species were planted at a density of approximately 1,000 per

hectare. No herbacous vegetation was planted.

Fifteen hummocks were constructed ofmulch to meet permit requirements. The

majority of hummocks are approximately 0.2 ha in size with dimensions of 52 m long by

30 m wide. The largest constructed hummock is approximately 0.6 ha, having dimensions

of 107 m long by 52 m wide. Hummocks are approximately 0.3 to 0.46 m higher than the

designed site elevation of 36 m.

Hal Scott. Figure 3 provides a post reclamation plan view of this site showing the

constructed forested wetland areas and hummock locations. Site was constructed using

sand tailings in February and March of 1990. This site consists of approximately 61 ha of

wetland (herbacous and forested), 63 ha of deep lake, and 34 ha of upland.

Approximately 20 ha were designated as forested wetland. Site was mulched and planted

with wetland tree species at a density of approximately 1,00 trees per ha in the spring of

1990. Two supplemental plantings 1 year apart followed this initial planting effort.

Field Data Collection--Microtopographic Development

On the 12 microtopographic development sites, ground surface elevations,

vegetation, and canopy cover data were collected at all sites. In addition, these same data

were collected at FG-FSB2 on separate transects that did not intercept constructed

hummocks.

Wetland Boundary
4 Constructed hummock

Figure 2. Post reclamation contour map for FG-GSB2 showing topographic
contours. Constructed hummocks are shown in gray. Figure not to scale.

-ai e - - 4, 4, 4. 4, *
--- ._ IMCT PROPERTY
-------- S.i-U(C1Z -^uaoj. f
-SCALE .......-- .
B'A- B EDGE Or OISTUR
0 FEET TOO HOOKERS PRAIRIE

S- Upland Pasture/Forest

SMeandering Stream

] Herbacous Marsh

k- Proposed Forested Wetland Area

F- Constructed Hummocks

Figure 3. Post reclamation plan view of Hal Scott showing forested wetland area
and location of constructed hummocks.

Ground Surface Elevations

Ground surface elevations were measured using a laser level and stadia rod.

Within each wetland, a laser level was set up at two random locations and four 10 m

transects radiating out from the laser level were established. Ground surface elevations

were taken at 20 cm intervals along each transect. Surface water elevation was measured

at each of the two sampling locations within each wetland. Elevation of adventitious

rooting was recorded as an indicator of apparent maximum high water.

Vegetation

Vegetation were grouped into two strata; understory and canopy. Understory

consisted of any vegetation greater than 5 cm in height and having a DBH (diameter at

breast height) less than 5 cm. Canopy consisted of any vegetation having a DBH greater

than 5 cm.

At each 20 cm interval along transects, understory plant species were recorded

using the point-intercept method (Brower et al. 1990). The DBH, location, and identity of

all canopy vegetation lying within 1 m of either side of the transect were also recorded.

Canopy Cover

Hemispherical canopy photography was used to measure canopy cover within each

system. Photographs were taken, from beneath the canopy looking upward, using a

hemispherical 15 mm fisheye lens. Photographs are then analyzed to determine the

geometry of canopy openings, and in turn, to estimate light levels beneath the canopy

(Rich 1989). The camera was mounted on a tripod approximately 0.75 m above ground

level. Two canopy photographs with different exposures were taken at one randomly

determined location per transect, yielding sixteen photographs per site. After

development, the photograph having the sharpest contrast was selected for each transect.

Photographs were scanned and percent canopy cover was measured using a computer

mapping program. An example of the end product using this methodology is provided as

Figure 4.

Field Data Collection--Constructed Hummocks

Ground Surface Elevations

Methods used to survey hummocks are presented as Figures 5 and 6. All ground

surface elevation measurements were collected using a laser level and stadia rod.

Constructed hummocks at FG-GSB2 were extremely larger than hummocks at Hal Scott.

As a result, sampling intervals as well as transect lengths differed between the two sites.

Five transects were surveyed at FG-GSB2 and Hal Scott. Elevations of surface water at

the time of sampling were recorded on two separate occasions at each site. Elevation of

adventitious rooting was recorded as a hydrologic indicator for apparent maximum high

water.

At FG-GSB2, transects were also placed to bisect each hummock at a

perpendicular angle. Transects began at least 4 to 8 m before each hummock and

continued for 6 to 10 m across each hummock. Along each transect, field surveys of

ground elevations were taken at 40 cm intervals. At Hal Scott, transects were placed to

bisect each hummock at a perpendicular angle. Transects extended approximately 5 m on

either side of the hummock. Along each transect, field surveys of ground elevations were

taken at 20 cm intervals.

Figure 4. Example of a digitized photo in MAP FACTORY used to estimate average percent canopy cover.

Stadia Rod

Laser Level

Created
Hummock

(a.)

CI A

LLE

,j-I III U-I

40 cm
[---|
)1: 31v 31: 31 ^IC f: )IC :I

A

Transect

- Elevation measurement points,
vegetation data collected using
point intercept method.

(b.)

Figure 5. Sampling method used in the collection of vegetative and elevation data
on large constructed hummocks at the FG-GSB2 wetland.
a) Side view showing method of collecting elevation data; b) Plan view
showing frequency of sampling along elevation gradient.

Varied from;
C 0-,

\ E

3

Created
Hummock

Stadia Rod

Laser Level

Created
I Hummock I

(a.)

5m

Transect

B

Created
Hummock

SElevation measurement points,
vegetation data collected using
point intercept method.

(b.)

Figure 6. Sampling method used to collect vegetative and elevation data on small
constructed hummocks at the Hal Scott wetland.
a) Side view showing method of collecting elevation data; (b) Plan view
showing frequency of sampling along elevation gradient.

5m

20 cm

A

Vegetation

At each sampling interval along transects, plant species were recorded using the

point-intercept method (Brower et al. 1990). All vegetation greater than 5 cm in height

and having a DBH less than 5 cm intercepting the stadia rod were recorded.

Data Analysis

Vegetation Data.

Microtopographic Development. Species richness and a modified Shannon-

Weaver diversity index were calculated in characterize understory vegetation. Relative

density, relative dominance, and importance value (IV) were calculated to determine the

dominant canopy vegetation in a given system. Sites were grouped into two categories

based on average canopy coverage; sites having an average canopy coverage (1) greater

than 70 percent and (2) less than 50 percent. This was done in order to study relationships

between the rugosity index of microtopography and species richness and diversity in

systems having similar canopy cover.

Species richness was determined for each site and is expressed as the number of

naturally recruited or planted species present.

The Shannon-Weaver index is an information-theoretic index used to

characterize the diversity of a system by analyzing a random sample from the larger

system.

It is calculated as:

H'= (N log N-n ni log ni )/ N

where:

ni = the proportion of the number of individuals, ni, of species i

N = total number of individuals in all the species

This index was modified by replacing ni with n, and N with N, in order to generate a value

for the probability of sampling a species rather than an individual,

where:

n,= the probability of sampling a species

N, = the number of species per transect

Maximum diversity is achieved here when the species, N,, are distributed as evenly as

possible.

For each transect, basal areas and stem densities were calculated for each canopy

species. Relative density, relative dominance, and importance value (IV) for each canopy

species were then calculated as follows:

Importance value (IV) = (relative density + relative dominance) / 2 (2)

where:

relative density = number of stems / total number of stems (3)

relative dominance = species basal area / total basal area (4)

basal area = (DBH / 2)2 3.14 (5)

By averaging both relative density and dominance, the importance value is an estimate of

the overall influence or importance of each canopy species at any given site. Density or

basal area alone may not appropriately determine the dominant species, therefore, relative

importance is used to provide a more balanced representation of the species present

(Brower et al. 1990). Thus, the dominant canopy species for a system represents the

species having the highest importance value.

Constructed Hummocks. Methods used to analyze vegetation included

calculating species richness and the modified Shannon-Weaver Diversity index for all

transects. Identifiable vegetation on constructed hummocks was characterized according

to their wetland plant classification (Reed 1988) (Table 1).

Table 1. Classification of wetland plants in Florida.

Classification Frequency of Occurrence in Wetlands
Obligate (OBL) Always (greater than 99%)
Facultative Wet (FACW) Usually (67% 99%)
Facultative (FACW) Sometimes (34% 66%)
Facultative Upland (FACU) Seldom (1% 33%)
Upland (UPL) Never (less than 1%)

Frequency of occurrence and average frequency of occurrence of nuisance species on

hummocks was also calculated. Frequency of occurrence was calculated as follows:

Frequency of occurrence = ni / n (6)

where:

ni = the number of times species i occurred

n = the number of sampling points for a given transect length

The average frequency of occurrence was calculated by summing the frequency of

occurrence for a given nuisance species on all hummocks and dividing by the total number

of hummocks (5).

Microtopography Data

Rugosity (R), an index of the "wrinkledness" of a line, was used to determine

microtopographic heterogeneity of a site (Sloan 1998). It is calculated for each transect

as a deviation of elevation measurements within a system by calculating the ratio of a line

which follows the elevation gradient of a transect to a horizontal vector corresponding to

the length of the transect. The resultant formula is as follows:

R1.
R= L (en-en. )2+ ( in )2 100 (7)
i=1 In

where:
e = elevation at point n
I, = length of segment n

Transects within each system were then pooled to derive an average rugosity.

Statistical Analysis

The non-parametric Mann-Whitney Test was used to test if significant differences

existed between: (1) microtopographic heterogeneity in natural and constructed forested

wetlands, (2) species richness or diversity on entire transects which intersected a

constructed hummock versus sections of the same transects only in adjacent wetland areas

between hummocks and (3) species richness or diversity on constructed hummocks versus

adjacent wetland areas between hummocks.

Simulation Model

To test the hypothesis that there is an optimum amount of hummock area in

forested wetlands, a macroscopic minimodel was developed and simulated using data from

Florida wetlands for calibration. The following paragraph provides simulation

methodology.

The simulation model that was used to test the hypothesis of optimum hummock

area was developed in a three stage process. First, an overview diagram of the interaction

of hummocks with vegetation and water storage was drawn using the energy systems

language (Odum 1986) to organize thinking and highlight the system state variables and

interactions deemed important. Second, the overview diagram was aggregated into a

macroscopic mini-model that in essence simplified interactions to two or three variables

that could be varied to test the hypothesis dynamically. Finally, the structure of equations

in the model were translated directly from the aggregated systems diagram and

programmed in an EXCEL spreadsheet for simulation. The translation is such that each

storage in the diagram is a state variable and each pathway flowing into or out of a storage

becomes a term in the equation for the state variable that is either added or subtracted

during each time step.

RESULTS

The following sections provide results of calculated values using the rugosity index

of microtopographic heterogeneity and various parameters for the twelve constructed

systems. A comparison ofmicrotopographic heterogeneity, using the rugosity index,

between seven natural forested wetlands (Sloan 1998) and the twelve constructed forested

wetlands surveyed in this study also follows. Floristic and structural characterizations of

constructed hummocks at FG-GSB2 and Hal Scott are presented. Utilization of

constructed hummocks by nuisance species is also presented.

Microtopographic Development

Given in Table 2 are summarized data for each of the twelve constructed wetland

sites where microtopographic development was evaluated. Appendix A provides

microtopographic data summarized by transect for the twelve sites. Sites are given in

order of age. Site name is given in the first column followed by the rugosity index,

dominant canopy species, percent canopy cover, species richness, and finally species

diversity in the last column.

Calculated rugosity index values varied from 100.3 (FG-GSB2) to 104.7 (Lizard

Branch). Expanded elevation ranges encountered on transects at Lizard Branch in

comparison to systems such as FG-GSB2 resulted in greater rugosity index values (Figure

7).

Table 2. Summary of microtopographic development data for the twelve surveyed sites.
Dominant Canopy
Age at Time Rugosity Index2 Species Mean Canopy Species S-W Diversity2
Site of Sampling (Sloan 1998) (Using Importance Value) Coverage (%) Richness Index
FG-GSB2 2 100.4 (0.19) Mixed herbacous' 0 26 1.02 (0.07)
Lizard Branch 3 104.7 (1.30) Ludwigia peruviana 0 41 1.35 (0.09)
Jameson Jr. 4 101.1 (0.31) Taxodium distichum 48 15 0.66 (0.05)
R-6 4 103.7 (1.00) Salix caroliniana 72 15 0.72 (0.18)
East Lobe 10 102.7 (1.18) Salix carolinana 79 24 0.99 (0.18)
West Lobe 10 102.5 (0.90) Acer rubrum 88 20 0.79 (0.20)
R-9 10 100.3 (0.18) Fraxinms caroliniana 45 37 1.30 (0.18)
R-7 12 101.1 (0.68) Fraxiins caroliniana 60 40 1.00 (0.16)
SP-6 16 101.7 (1.16) Myrica cerifera 87 21 1.00 (0.11)
Hall's Branch 17 101.5 (0.54) Salix caroliniana 86 44 1.35 (0.27)
Parcel B 18 100.5 (0.38) Fraximns caroliniana 88 27 0.96 (0.17)
EPR 1 (Section 1) 27 101.6 (1.00) Myrica cerifera 85 21 0.87 (0.07)
1 Dominant species include L. peruviana and Juncus effusus.
2 Values in parenthesis represent 1 standard deviation.

0.40

0.30

0.20

0.10

0.00

0 1 2 3 4 5 6 7 8 9 10
Transect length (m)

(a.)

0.40 -

0.30 -

0.20 -

0.10 -

0.00

A

0 1 2 3 4 5 6 7 8 9 10
Transect length (m)

(b.)

Figure 7. Microtopographic relief along an average transect.
a)Lizard Branch; b) FG-GSB2.

I ~C ----I
I I I i 1 I I I 1 I T-

I

I I

J

Ages of constructed wetlands at the time of surveying ranged from 2 to 28 years.

As shown in Figure 8, no strong relationship was found between site age and the rugosity

index ofmicrotopographic heterogeneity. Older sites such as EPR 1 (Section 1), Parcel

B, and Hall's Branch had lower microtopographic development, based on rugosity index

values, than many younger constructed sites. EPR 1, the oldest site surveyed, had a

rugosity index value (101.6) which was lower than five other sites. Lizard Branch, a 4

year-old site, had the highest rugosity index value (104.7) of all sites.

Six different dominant canopy species were found in the twelve systems surveyed;

bald cypress (Taxodium distichum), carolina willow (Salix caroliniana), pop ash

(Fraxinus caroliniana), primrose willow (Ludwigiaperuviana), red maple (Acer rubrum),

and wax myrtle (Myrica cerifera). As shown in Figure 9, when comparing dominate

canopy species to the calculated rugosity index for that system, a system dominated by

primrose willow had the highest rugosity index value, 104.7. Sites having carolina willow

as the dominant canopy vegetation (East Lobe, R-6, Hall's Branch) had rugosity values of

102.7, 103.7, and 101.5, respectively. The mean rugosity for all carolina willow

dominated sites, 102.6, was greater than sites dominated by red maple (102.5), pop ash

(100.6), wax myrtle (101.7), and bald cypress (101.1) (Figure 9).

Using average canopy cover, sites were grouped into two categories; (1) sites

having an average canopy coverage greater than 70 percent and (2) sites having an

average canopy coverage less than 50 percent. This was done to examine relationships

between the rugosity index of microtopographic heterogeneity and species richness and

diversity. Seven sites had an average canopy coverage greater than 70 percent and five

sites had an average canopy coverage less than 50 percent.

105.0 -

104.5 -

104.0

103.5

103.0

102.5-

102.0-

101.5 -

101.0

100.5

100.0

Figure 8. Relationship between site age and the rugosity index of
microtopographic heterogeneity for the twelve systems surveyed.

I I I
10 15 20

Site age at the time of sampling (years)

105.0 -

104.0 -

103.0

102.0

101.0 -

100.0

Red maple

n=l

Wax myrtle

n=2

Pop ash Bald cypress

n=3

Dominant canopy species
(Using Importance Value)

Figure 9. Comparison between systems having different dominant canopy vegetation and the rugosity index of
microtopographic heterogeneity.

106.0

E _______

Primrose
willow

n=l

Carolina
willow

n=3

Using the modified Shannon-Weaver species diversity index, understory plant

diversity ranged from 0.66 to 1.35, found at Hall's Branch and Lizard Branch,

respectively. Species richness for understory vegetation ranged from 15 (R-6 and Hall's

Branch) to 44 (Jameson Jr.). No strong relationships were found between

microtopographic heterogeneity, using the rugosity index, and species richness or diversity

in sites having average canopy coverages less than 50 percent (Figure 10) or greater than

70 percent (Figure 11).

Evaluation of Constructed Hummocks

IMC Agrico's FG-GSB2 site

Vegetation. Results of descriptive indices are summarized in Table 3. Species

richness for entire transects which intersected a constructed hummock averaged 11 species

with a range of 7 to 16. Species diversity averaged 1.10 with a minimum diversity value

of 0.95 and a maximum of 1.27.

As shown in Table 3, species richness for the transect sections which did not

intersect hummocks and were only in the adjacent wetland areas between hummocks

averaged 5 species with a range of 3 to 7 species. Species diversity averaged 0.75 with a

range of 0.49 to 0.95.

Table 4 provides community descriptive indice results for the section of transects

only on constructed hummocks. Species averaged 10 species with a minimum richness of

6 species and a maximum of 14. Species diversity averaged 1.06 with a minimum value of

0.88 and a maximum of 1.22.

101.0

102.0

103.0

104.0

105.0

Rugosity index

(a.)

101.0

102.0 103.0
Rugosity index

104.0

105.0

(b.)

Relationship between the rugosity index of microtopographic
heterogeneity and community descriptive indices in systems having an
average canopy coverage less than 50 percent.
a) Species richness; b) Species diversity.

10

5

0
100.0

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0 +-
100.0

Figure 10.

102.0
Rugosity index

(a.)

102.0
Rugosity index

(b.)

103.0

103.0

104.0

104.0

Relationship between the rugosity index of microtopographic
heterogeneity and community descriptive indices in systems having an
average canopy coverage greater than 70 percent.
a) Species richness; b) Species diversity.

50
45

25
20
15
10
5
0-
100.0

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0 1
100.0

101.0

101.0

Figure 11.

Table 3. Summary of community descriptive indices comparing entire transects
intersecting a created hummock (ON) and sections of the same transects
which were only in adjacent wetland areas between hummocks (OFF).
Site Transect Richness1 Diversity'

IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2

IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2

IMC: Hal
IMC: Hal
IMC: Hal
IMC: Hal
IMC: Hal

ONI
ON2
ON3
ON4
ON5

Mean

OFF1
OFF2
OFF3
OFF4
OFF5
Mean

Scott
Scott
Scott
Scott
Scott

ON1
ON2
ON3
ON4
ON5

Mean

10
9
13
7
16
11(4)

3
3
6
5
7
5 (2)

19
15
14
11
16
15(3)

IMC: Hal Scott OFF1 11
IMC: Hal Scott OFF2 8
IMC: Hal Scott OFF3 9
IMC: Hal Scott OFF4 9
IMC: Hal Scott OFF5 10
Mean 9 (1)
1 Values in parenthesis represent 1 standard deviation.

1.08
1.03
1.18
0.95
1.27
1.10 (0.13)

0.49
0.62
0.89
0.82
0.95
0.75 (0.19)

1.34
1.24
1.21
1.12
1.25
1.23 (0.08)

1.12
1.01
1.03
1.04
1.06
1.05 (0.05)

Table 4. Community descriptive indice comparison between sections of transects
only on a constructed hummock (ON) and sections of the same transects
only in adjacent wetland areas between hummocks (OFF).
Site Transect Richness' Diversity1

IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2

IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2
IMC: FG-GSB2

IMC: Hal Scott
IMC: Hal Scott
IMC: Hal Scott
IMC: Hal Scott
IMC: Hal Scott

IMC: Hal Scott
IMC: Hal Scott
IMC: Hal Scott
IMC: Hal Scott
IMC: Hal Scott

ON1
ON2
ON3
ON4
ON5
Mean

OFF1
OFF2
OFF3
OFF4
OFF5
Mean

ON1
ON2
ON3
ON4
ON5
Mean

OFF1
OFF2
OFF3
OFF4
OFF5

Mean
1 Values in parenthesis represent 1 standard deviation.

10
9
10
6
14
10(3)

3
3
6
5
7
5 (2)

15
8
10
5
12
10(4)

1.08
1.03
1.09
0.88
1.22
1.06 (0.12)

0.49
0.62
0.89
0.82
0.95
0.75 (0.19)

1.25
0.88
1.08
0.95
1.12
1.04 (0.15)

1.09
1.12
0.94
1.04
0.98
1.03 (0.07)

10
11
7
9
8
9 (2)

A significant difference was found between species richness (p=0.02) and diversity

(p=0.01) for transects which intersected a constructed hummock and transect sections

which were only in adjacent wetland areas between hummocks. A statistically significant

difference was also found between species richness (p= 0.03) and diversity (p= 0.04) for

----

transect sections only on hummocks compared to sections which were only in adjacent

wetland areas.

As shown in Table 5, a mean of 4 identifiable plant species were found exclusively

on constructed hummocks with a maximum of 5 species and a minimum of 1. Forty-eight

percent of the identifiable plant species found exclusively on the constructed hummocks

were classified as FAC, FACW, or OBL while the remaining 52 percent were classified as

FACU (Reed 1988).

The frequency of occurrence of "nuisance species" on constructed hummocks is

presented in Table 6. Three "nuisance" species were found when sampling points from the

sections of transects only on constructed hummocks were pooled; primrose willow,

carolina willow, and cattail (Typha latifolia). Average frequency of occurrence for

primrose willow, cattail, and carolina willow was 0.26, 0.02, and 0.01, respectively (Table

5). Primrose willow, carolina willow, and cattail were also found in adjacent wetland

areas between hummocks and had an average frequency of occurrence of 0.35, 0.01, and

0.07, respectively. Additionally, hempweed (Mikania scandens) was found in wetland

areas, having a frequency of occurrence of 0.07. It was not found on hummocks.

Structural Characterization. Shown in Figure 12a is a typical cross section of

the hummocks at IMC Agrico's FG-GSB2 site. As shown in Table 7, mean hummock

height surveyed was 0.33 m with a minimum height of 0.22 m and a maximum of 0.49 m.

Apparent average water level, based on observations on two sampling dates, was

approximately 0.21 m below mean hummock elevation (Table 8). Apparent maximum

high water level, based on the elevation of primrose willow adventitious roots, is

approximately 0.55 m above mean hummock elevation (Table 8).

Table 5. Inventory and wetland classification of identifiable plant species found
exclusively on constructed hummocks at FG-GSB2.

Site
IMC: FG-GSB2

Transect
ON1

Plant species
Erechtites hieracifolia

Phytolacca americana
Paspalum notatum
Eupatorium capillifolium
Andropogon urvillei
Paspalum ridigulum

Wetland Vegetation
Classification (Reed 1998)
FAC
FACU
FACU
FACU
FAC
FACW

IMC: FG-GSB2

IMC: FG-GSB2

IMC: FG-GSB2

IMC: FG-GSB2

ON2 Ludwigia octovalis
Andropogon glomeratus
E. capillifolium
P. notatum

ON3 Crotalaria spp.
A. glomeratus
Salix caroliniana
P. notatum
Sesbania spp.

ON4 P. notatum

ON5 P. notatum
A. urvillei
E. capillifolium
Eclipta alba
Indigofera spp.
Mean number of
species found
per hummock = 4

OBL
FACW
FACU
FACU

FACU
FACW
OBL
FACU
FACW

FACU

FACU
FAC
FACU
FACW
FACU
FAC, FACW,
or OBL = 48%
FACU = 52%

Table 6. Frequency of occurrence (FOC) of nuisance species on and off constructed hummocks at FG-GSB2 and Hal Scott.
Frequency of Occurrence Frequency of Occurrence
Site Hummock Nuisance Species ON Hummocks OFF Hummocks

FG-GSB2

FG-GSB2

Ludwvigia pernviana
Typha spp.

L. peruviana
Salix caroliniana
Typha spp.

L. peruviana
Typha spp.

L. peruviana
Typha spp.
Mikania scandens

FG-GSB2

FG-GSB2

FG-GSB2

L. peruiviana
M. scandens
Typha spp.

0.03
0.00

0.64
0.04
0.00

0.22
0.00

0.16
0.11
0.00

0.39
0.20

0.47
0.00
0.06

0.08
0.06

0.67
0.00
0.10

0.25
0.00
0.00
Avg FOC
0.26
0.00
0.01
0.02

L. peruviana
M. scandens
S. caroliniana
Typha spp.

0.14
0.04
0.02
Avg FOC
0.35
0.03
0.01
0.07

~---

Table 6--continued.
Frequency of Occurrence Frequency of Occurrence
Site Hummock Nuisance Species ON Hummocks OFF Hummocks
Hal Scott 1 L. peruviana 0.00 0.08
M. scandens 0.06 0.58
S. caroliniana 0.17 0.04

Hal Scott 2 L. peruviana 0.00 0.21
M. scandens 0.08 0.25

Hal Scott 3 L. peruviana 0.00 0.21
M. scandens 0.13 0.50
S. caroliniana 0.00 0.04

Hal Scott 4 M. scandens 0.00 0.03 4
S. caroliniana 0.00 0.03

Hal Scott 5 S. caroliniana 0.00 0.10
L. peruviana 0.00 0.30
Avg FOC Avg FOC
L. peruviana 0.00 0.33
M. scandens 0.07 0.28
S. caroliniana 0.04 0.10

Range of hummock widths

10 107 m

0.55 m

0.33 m

I

Apparent
maximum
water level

Apparent
average water
level

(a.)

Range of hummock widths

1-3

A

w

0.29 m

0.17m \ le
f 0.07 m

apparent
maximumm
ater level

apparentt
average water
vel

Figure 12. Typical cross sections of hummocks showing heights and water levels
where constructed hummocks were evaluated.
a)FG-GSB2; b) Hal Scott.

Table 7. Calculation table deriving average constructed hummock height at
FG-GSB2 and Hal Scott. Elevations are relative to height of laser level.
Mean Ground Mean Hummock Mean Hummock
Site Transect Elevation (m)1 Elevation (m)' Height (m)1
FG-GSB2 1 1.38 1.64 0.26
FG-GSB2 2 1.41 1.63 0.22
FG-GSB2 3 1.47 1.82 0.35
FG-GSB2 4 1.50 1.82 0.33
FG-GSB2 5 1.31 1.79 0.49
Mean 1.41 (0.08) 1.74 (0.10) 0.33 (0.10)

Hal Scott 1
Hal Scott 2
Hal Scott 3
Hal Scott 4
Hal Scott 5
Mean

1.68
1.61
1.59
1.58
1.64
1.62 (0.05) 1

1.88
1.80
1.70
1.74
1.79
.78 (0.08)

0.20
0.19
0.14
0.16
0.15
0.17 (0.03)

1 Values in parenthesis represent 1 standard deviation.

Table 8. Mean hummock and water level elevation for FG-GSB2 and Hal Scott.
Elevations are relative to the height of the laser level. Numbers in
parenthesis represent hummock elevation relative to apparent water level.
Mean Hummock Apparent Mean Apparent Max
Site Elevation (m) Water (m)1 High Water (m)2
FG-GSB2 1.74 1.53 (+ 0.21) 2.29 (-0.55)

Hal Scott 1.78 1.68 (+0.10) 2.00 (-0.22)
1 Based on two site visits late in the growing season.
2 Based on the elevation of adventitious roots of primrose (FG-GSB2) and carolina
willow (Hal Scott).

IMC Agrico's Hal Scott (Section 12) site

Vegetation. Results of descriptive indices are summarized in Table 3. Species

richness on transects which intersected a constructed hummock had a mean of 15 species

with a minimum richness of 11 species and a maximum of 19. Species diversity averaged

1.23 with a minimum diversity value of 1.12 and a maximum of 1.34.

~

As shown in Table 3, species richness for the sections of transects only in adjacent

wetland areas between constructed hummocks averaged 9 species with a minimum

richness of 7 species and a maximum of 11. Species diversity averaged 0.92 with a range

of 0.94 to 1.12. A statistically significant difference was found between species richness

(p= 0.04) and diversity (p= 0.04) for entire transects which intersected a constructed

hummock and transect sections which were located only in adjacent wetland areas

between hummocks.

Species richness for the section of transects only on constructed hummocks

averaged 10 species with a minimum richness of 5 species and a maximum of 15 (See

Table 4). Species diversity averaged 1.04 with a range of 0.88 to 1.25. Species richness

and diversity for the section of transects only on hummocks was greater than the section

of transects only in adjacent wetland areas between hummocks.

As shown in Table 9, a mean of 4 identifiable plant species were found exclusively on

constructed hummocks with a maximum of 7 species and a minimum of 2. One- hundred

percent of these plant species were classified as FAC, FACW, or OBL (See Table 9).

Table 9. Inventory and wetland classification of identifiable plant species found
exclusively on constructed hummocks at Hal Scott.
Wetland Vegetation
Site Transect Plant species Classification (Reed 1988)
IMC: Hal Scott ON1 Eleocharis spp. OBL
Baccharis halimifolia FAC
Salix caroliniana OBL
Myrica cerifera FACW
Lythrum alatum OBL
Juncus effusus OBL
Thalia geniculata OBL

IMC: Hal Scott ON2 L. alatum OBL

Table 9-continued.

Site
IMC: Hal Scott

IMC: Hal Scott

IMC: Hal Scott

The presence and frequency of occurrence of "nuisance species" was also surveyed

at this site. As shown in Table 6, three "nuisance" species, hempweed, primrose willow,

and carolina willow were found at this site. Only hempweed and carolina willow were

recorded when data points collected from the five hummocks were pooled. On the five

hummocks surveyed, hempweed and carolina willow had an average frequency of 0.07

and 0.04, respectively. No nuisance species were found on hummock four or five. In

addition to hempweed and carolina willow, primrose willow were found in adjacent

wetland areas between hummocks. Primrose willow, hempweed, and carolina willow had

an average frequency of occurrence of 0.33, 0.28, and 0.04, respectively, in wetland areas

between hummocks (See Table 6). The average frequency of occurrence for the 3 species

found in adjacent wetland areas between hummocks was greater than frequencies recorded

on constructed hummocks.

s

T

ransect Plant species
ON3 Bacopa spp.
L. alatum
Pluchea rosea
Quercus nigra (seedling)
Hydrocotyle umbellata

ON4 Bacopa spp.
L. repens

ON5 L. alatum
P. rosea
Commelina diffiisa
Spartina bakeri
Mean number of
species found
per hummock = 4

Wetland Vegetation
Classification (Reed 1988)
OBL
OBL
FACW
FACW
OBL

OBL
OBL

OBL
FACW
FACW
FACW
FAC, FACW,
or OBL = 100%

Structural Characterization. Shown in Figure 12b is a typical cross section of

the hummocks on IMC Agrico's Hal Scott site. Mean hummock height surveyed was

0.17 m with a minimum height of 0.14 m and a maximum of 0.20 m (Table 7). Apparent

average water level, based on observations during two separate site visits, was assumed to

be approximately 0.10 m below mean hummock elevation (Table 8). Apparent maximum

water level, based on the elevation of adventitious rooting by carolina willow, was

assumed to be approximately 0.12 m above mean hummock elevation (Table 8).

Simulation Model

Model Overview

A computer model was developed to simulate planted wetland tree development,

while competing with nuisance species, in a constructed wetland having inundated areas

and created hummocks. For various percentages of wetland surface area inundated or as

constructed hummocks, the model simulated changes in tree and nuisance specie biomass

over time. The model predicts trends and long term results of incorporating hummocks

into constructed wetlands. A complex systems diagram is given in Figure 13 and the

aggregated system diagram that represents the simulation model is given in Figure 14. A

description and evaluation of energy flows and storage are provided in Appendix B.

Energy source. Renewable energy resources used by the system such as sun, rain,

and wind were aggregated into a single energy source (E). Only relative fractions of E

were used to examine the systems response.

Inundation and hummock storage. The two storage, I and H, are constants

used in the production functions of both producer units. "I" represents the amount of

4-
kRLY
CCESSIONAL
RESTED
ETLAN3 tD -- DIVERSITY

MATURE
FORESTED / ORGANIC
\Figure 13. Co x s s dia m of MATTER d fo d

Figure 13. Complex systems diagram of a constructed forested wetland.

J7P J9 J10

SNuisancespecies

DT= J3+J4-J5 = (k3*T*Jr*I)+(k4*T*Jr*H)-(k5*T)
DN= J8+J9-J10 = (k8*N*Jrr*I)+(k8*N*Jrr*H)-(k9*N)

Figure 14. Aggregated systems diagram of a constructed forested wetland used in the simulation model.

inundated area, while "H" represents the amount of hummock surface area in the

constructed wetland. Values for ""' and "I' varied from 0 to 100 with the sum of"I"

and "I' equaling 100.

Production and producer units. The model contains two producers, planted

wetland trees and nuisance species, in competition for sunlight. Trees receive sunlight first

and nuisance species receive light not absorbed or reflected by trees. This aggregation

ignores the initial shading of trees by nuisance species during the initial few years.

However, due to the shade tolerance of many wetland tree species this is a minor

omission.

Trees and nuisance species receive sunlight as a flow limited source. The larger

these producers grow the less sunlight is available. The remaining sunlight not captured

by planted trees and nuisance species are represented by Jr and Jrr, respectively. When

using a flow limited source, these remainders are the energy sources driving production.

Two production functions, A and B, are contained within each producer unit.

Gross production for A in each producer is a function of the sunlight remainder, the area

of the wetland inundated (I), and an autocatalytic feedback loop from the storage of that

producer (T or N). Gross production for B in the tree producer is a function of sunlight

(Jr), the square root of the area of constructed hummocks in the wetland (H), and a second

autocatalytic feedback loop from the storage. Gross production for B in the nuisance

species producer is a function of sunlight (Jr), H, and a second autocatalytic feedback loop

from the storage. The production ratio between the top and bottom production functions

at steady state for planted trees is approximately one to two and four to one for nuisance

species. Respiration for planted trees and nuisance species is subtracted from the storage

ofbiomass by a linear drain. Therefore, the amount of material flow for the linear drain

from the storage equals gross production.

Calibration and Simulation

Using data from natural climax forested wetland systems in Florida (Mitsch and

Gosselink 1995, Davis et al. 1991) and constructed wetlands in the phosphate-mining

district (Rushton 1988), pathway coefficients, energy flows, and initial conditions were

calculated in spreadsheets. The sum of inflow pathways to a storage were calculated by

dividing the climax storage value by its turnover time. Turnover times of 80 years and 2

years were used for wetland trees (T) and nuisance species (N), respectively. Steady state

was achieved by setting storage inflows equal to outflows. The model was simulated for

250 years with a 1 year time step. No changes in the flow of renewable energy (E) were

made during the simulations.

Simulating Hummock Benefit

The benefit of hummocks within forested wetlands was derived from the

simulation model as a product of two storage, biomass (tree and nuisance species) and

water. The rational is that both the storage ofbiomass and water are important benefits

of forested wetlands. With 100 percent inundation (maximum water storage), biomass is

mostly a result of herbaceous or floating aquatic vegetation. As the wetland ecosystem

dries out (becomes 100 percent hummock or upland area) biomass increases to maximum

terrestrial forested storage but water storage is minimal. Thus, the benefit of hummocks is

to provide some dryer area to maximize biomass, yet still allow water storage. The

following equation was used to calculate hummock benefit from model output:

Hummock Benefit = (1/100) (8)
B100:0
Bloo:o
where:

Hummock benefit = index

Bt= Planted wetland tree and nuisance species biomass at time t (t=250)

B oo:0 = Planted wetland tree and nuisance species biomass at time t for simulation
with 100% inundated area and 0% hummock area

I = Inundated area, (0 < I 100)

Simulation Results

Figure 15 is a graph of the simulation results using calibration values showing

wetland tree and nuisance species biomass for a 250 year simulation. Nuisance species

biomass peaks after 7 years and declines as tree biomass increases and begins to shade

these understory species. Tree biomass reaches steady state in approximately 190 years.

Figure 16 gives simulation results of tree and nuisance species biomass for

variations in ratios of I and H. In the lowest curve, the wetland has 100 percent inundated

area and 0 percent hummock. Each successively higher graph is the result of increasing

hummock area and a corresponding decrease in inundated area. The top curve results

from having 100 percent hummock area. As might be expected, when the percent of the

wetland that is hummock is increased the amount of tree biomass increases while nuisance

species biomass decreases. When the wetland is 100 percent inundated, representing a

herbaceous marsh, nuisance species biomass dominates for the course of the simulation.

-T (Wetland trees)
-N (Nuisance spp.)

0 50 100 150 200 250

Time (years)

Figure 15. Interaction between planted wetland tree and nuisance species biomass under calibrated conditions in the
simulation mini-model.

16,000

14,000

12,000

10,000

8,000

6,000

4,000

2,000

0

27,000

24,000

21,000

18,000

.5,000

.2,000

0 50 100 150 200

Time (years)

(a.)

9,000

6,000

3,000

0

2.000

1,800

1,600

1.400

1,200

n 1,000

E 800
.

0 50 100 150 200 250
Time (years)

(b.)
Figure 16. Simulation results of the model in response to various I:H ratios.
a) Planted wetland trees (T); b) Nuisances species (N).

0:100
/ 20:80

30:70
50:50
70:30
-80:20
90:10

100:0

250

100:0
90:10

70:30

50:50
30:70

0:100

Hummock Benefit

When the model is simulated repeatedly using different ratios of hummock to

inundated area, and the benefit is plotted for each simulation, the graph in Figure 17a

results. The vertical axis is the hummock benefit index expressed as a ratio between 0 and

1.0. Maximum benefit is equal to 1.0. The horizontal axis is percent hummock area. The

graph shows a maximum theoretical benefit of when percent hummock area is 20 percent

of total wetland area.

In order to study the relationship between hummock area and hummock benefit for

various ratios of I and H, the hummock benefit index was modified. The ratio of I/100

was weighted in the index to increase the importance of this variable. Ratios of Bt to

Bloo:o vary from 1 to 23 while 1/100 varies from 0.01 to 1. Before weighting I, it was first

necessary to normalize this variable in order to generate a range of 1/100 values which had

the same range as B,/Bloo:o ratios. Weighting factors were then incorporated into the

index. The resulting formula for the hummock benefit index is as follows:

Bt
HB= kB* k* (I/100) (9)
Bloo:1
where:

kB = Biomass weighting factor

ki = Inundation weighting factor

Values generated using this formula were then divided by the maximum value achieved for

all I:H ratios, resulting in a range of values from 0 to 1. Weighting factors are based on

the differences in transformity between biomass and water. The transformity, or energy

transformation ratio, is the amount of energy required to make a product (EMERGY) per

1.2

1.0

0.8 -

W 0.6 -

0.4

0.2

0.0
0 20 40 60 80 100
Percent hummock area

(a.)

1.2 kB=l, kI= 1
kB= kI=4
1.0 --- *-- ---------kB=l, kI=6
-- kB=l, k1=8
t0.8 -- --*-kB=l, kI=10

0.64 .. ..---- ------

0.4

0.2

0.0
0 20 40 60 80 100

Percent hummock area

(b.)

Figure 17. Relationship between the amount of hummock area in a wetland and
the hummock benefit index.
a) Under steady state conditions; b) For five different transformity
weighting factors multiplied to I and B in the hummock benefit index.

unit energy of that product. The more energy transformations required to make or that

contribute to a product, the greater its transformity (Odum 1996). Transformity units are

expressed as solar emjoule per joule (sej/J).

The transformity of surface runoff is approximately 4 E4 sej/J while the

transformity ofbiomass is about 1 E4 sej/J (Odum 1996). By multiplying the factors in

the index by their transformities, results generated are in the same units. Curves shown in

Figure 17b are for different weighting factors (based on transformity) of water storage

between 1.0 and 10.0. The result of the weighting is to increase the importance of water

storage in the benefit index, shifting the curve of benefits up. Maximum hummock benefit

was achieved for all weighting factors of water storage when hummock area was 20

percent.

DISCUSSION

In this study of microtopographic development and characteristics of constructed

hummocks, several questions were poised including: Is there a progression of

microtopographic development in constructed forested wetlands? Do different dominant

canopy vegetation affect hummock development? Do constructed hummocks contribute

to species richness and diversity? Do constructed hummocks provide desirable sites for

the establishment of nuisance species? To answer these questions, numerous

measurements of physical and biological parameters were conducted and resulting data

were analyzed seeking correlations and significance. The main findings of this study are as

follows:

1. Constructed forested wetlands surveyed exhibited no apparent correlation

between microtopographic development and age of system.

2. Dominant canopy vegetation affects microtopographic relief Constructed

forested wetlands dominated by primrose willow and carolina willow had

higher relief than systems dominated by other canopy vegetation.

3. Constructed hummocks appear to increase species richness and diversity. Both

measures were higher in hummocks transects when compared to transects

without hummocks.

4. Constructed hummocks did not provide sites for extensive nuisance species

establishment. In fact, nuisance species were more prevalent in adjacent

wetland areas between hummocks.

Microtopographic Development and Site Age

As shown in Figure 10, no strong relationship existed between site age and the

rugosity index of microtopographic relief in the twelve systems surveyed. Young systems

surveyed, such as Lizard Branch (3 years), Jameson Jr. (4 years), and R6 (4 years) did not

have rugosity values lower than older systems. The presence of early successional species,

specifically primrose willow and carolina willow, seemed to have a more positive influence

on microtopographic development than age in the systems surveyed. Systems dominated

by primrose and carolina willow yielded the highest rugosity index values. Therefore,

young systems such as Lizard Branch, Jameson Jr., and R6 did not have rugosity values

lower than older systems as expected.

Microtopographic Development and Dominant Canopy Vegetation

As shown in Figure 11, a system dominated by primrose willow, Lizard Branch,

had the highest rugosity index of all sites. This was due to transects intersecting

hummocks formed by primrose willow. Hummocks in this system were formed by the

deposition of litter and woody biomass around the base of the shrub. Primrose willow has

very high growth rates and deposits significant amounts of biomass on the wetland floor.

When primrose willow forms a closed canopy, lower stems become weak and any future

loss of support from adjacent branches causes them to fall. This creates a maze of

branches at ground level which subsequently collect leaf litter and debris. Eventually, this

material creates organic mounds at the base of the shrub and ultimately raises ground

elevations (Bowmer 1991). Thus, primrose willow may be an important early successional

species to facilitate microtopographic development in constructed forested wetlands.

Systems dominated by carolina willow had a higher rugosity index than sampled

sites dominated by red maple, green ash, wax myrtle, and bald cypress (Figure 11).

Numerous worldwide studies of succession describe Salix spp. as the first pioneer woody

species to colonize saturated substrates laid bare by receding glaciers, earthquakes, forest

fires, gravel pits, and mine wastes (Rehder 1951, Smith and others 1978, Stott 1962,

Warren-Wren 1973). In Central and South Florida, carolina willow, the native species of

willow, is considered an early successional or pioneer species. Wharton et al. (1977)

described carolina willow as the dominant woody species in the early successional stages

of deep-water marsh succession to hardwood forest. Due to the relatively short life span

of carolina willow, broken limbs and branches are a common component of the wetland

floor. In this study, systems dominated by carolina willow had significant amounts of

fallen limbs and branches that increased site rugosity. While this material might

decompose completely, the woody debris will maintain a percentage of its elevated

position in relation to adjacent areas and continue to trap biomass deposited from

understory or overstory species. Fallen canopy biomass was noted by Titus (1987) as a

known formation catalyst for microtopographic development in forested wetland systems.

West Lobe, dominated by red maple, had the third highest rugosity value of the

twelve systems surveyed when comparing dominant canopy specie to microtopography

(Figure 11). However, this system was previously dominated by carolina willow and has

now been succeeded by red maple. This was apparent due to the standing biomass of

large mature, yet dying, carolina willow still present in this system. Red maple had a

higher importance value when compared to carolina willow and was considered the

dominant canopy tree specie. However, changes in ground elevations recorded during

surveying were often the result of fallen carolina willow biomass (branches and trunks). In

all, this suggests that carolina willow may also accelerate microtopographic development

in newly constructed forested wetlands due to the large amount of biomass which falls to

the wetland floor as the willow canopy matures or is succeeded by other species.

Microtopographic Development in Natural and Constructed Forested Wetlands

Seven natural forested wetland systems in North Central Florida surveyed by Sloan

(1998) were compared using the rugosity index of microtopographic heterogeneity to the

twelve constructed systems. As shown in Figure 19, constructed forested wetland systems

had significantly lower (p=0.004) microtopographic heterogeneity than natural systems.

The low rugosity index found in even the oldest constructed systems when compared to

natural systems suggests that the processes necessary to develop microtopography similar

to natural systems may take many years. Wind-throw of dead, poorly rooted, or diseased

trees, one of the most common formation mechanisms of microtopography, is absent in

these young systems because mature tree stands have not developed.

In the seven natural forested wetlands, Sloan (1998) surveyed five transects per

site. Microtopographic features were grouped into ten hummock types and the frequency

and density of each were calculated. Approximately five hummocks per transect, with a

mean height of 0.24 m, were found in these systems. Microtopographic variations in

constructed systems were not similar in magnitude or frequency when compared to natural

systems. Only ten topographic features in the twelve systems surveyed had heights equal

to or greater than the mean hummock height, 0.24, found in natural systems. These were

caused by fallen trunks and limbs of carolina willow at East and West Lobe.

140.0 -

130.0
Constructed systems Natural systems

120.0

0

90.0 T -rr-

..' . -S

c S

Figure 18. Comparison of natural and constructed forested wetland systems using the rugosity index of microtopographic
heterogeneity.

Species Richness and Diversity in Systems with Constructed Hummocks

At both FG-GSB2 and Hal Scott, species richness and diversity were significantly

greater (p=0.02) when entire transects which intersected a constructed hummock were

compared to the sections of transects which were only in adjacent wetland areas between

hummocks (Table 3). Based on these results, constructed hummocks appear to

significantly increase plant species diversity and richness in these constructed forested

wetland systems by providing elevated microsites for the colonization of other plant

species.

Species Richness and Diversity on Constructed Hummocks

Average species richness and diversity for the section of transects only on

constructed hummocks at FG-GSB2 were significantly higher than transect sections in the

adjacent wetland area (Table 4). At Hal Scott, average species diversity was greater for

the section of transects only on hummocks while average species richness were equal. By

analyzing species composition more closely on constructed hummocks, it was determined

that an average of four herbacous understory species at FG-GSB2 and Hal Scott were

found exclusively on constructed hummocks. As shown in Table 5, 48 percent of

understory species found at FG-GSB2 were FAC, FACW, or OBL. One hundred percent

of the identifiable plant species found exclusively on constructed hummocks at Hal Scott

were OBL or FACW (Table 9). The high percentage of identifiable species being

classified as FAC, FACW, or OBL indicates that elevations of constructed hummocks

relative to water level provide sites for wetland plants (OBL), as well as less flood tolerant

FAC and FACW plant species, which may otherwise not be present in these systems.

These species, combined with species found in the lower elevations of the wetland floor,

significantly increase plant richness and diversity at FG-GSB2 and Hal Scott.

Nuisance Species and Constructed Hummocks

A common concern voiced by the industry with regards to constructed hummocks

is that they become sites for nuisance species colonization. Results provided here suggest

that constructed hummocks at FG-GSB2 and Hal Scott do not provide sites for extensive

colonization by nuisance species (Table 6).

As shown in Table 5, primrose willow had a high frequency of occurrence (0.64)

on one constructed hummock at FG-GSB2. However, on the other four hummocks

surveyed, primrose willow had a frequency of occurrence of 0.17. Table 6 also shows that

of the four nuisance species found, three had higher relative frequencies in the wetland

areas adjacent to constructed hummocks than on the hummocks, while the fourth species

had the same frequency.

At Hal Scott, only the nuisance species hempweed or carolina willow were found

(Table 6). On three of the four hummocks on which they were found, hempweed and

carolina willow had an average frequency of occurrence of only 0.07 and 0.04,

respectively. In wetland areas adjacent to the constructed hummocks primrose willow,

hempvine, and carolina willow were found. As shown in Table 6, both hempvine and

primrose willow had significantly higher relative frequencies in these adjacent wetland

areas than on hummocks.

The presence of nuisance species such as primrose willow and cattail on

constructed hummocks, even at these relatively low frequencies of occurrence, could

warrant concern. However, Richardson and Johnson (1998) reported that primrose

willow and cattail biomass and cover decrease with increasing canopy development. As

planted and naturally recruited trees overtop and begin shading these early successional

species, they will be outcompeted by the maturing canopy. Supplemental planting, if

necessary, of desirable herbaceous shade tolerant species could then follow.

Structural Characterization of Constructed Hummocks

The density of constructed hummocks at FG-GSB2 and Hal Scott were not

comparable to values cited in the literature for natural forested wetland systems. The

density of constructed hummocks was approximately 1 per ha at FG-GSB2 and 350 per ha

at Hal Scott. Davis et al. (1991), in a survey of topographic relief in several wetland

communities of Central and North Florida, found "relatively rough microtopographic

relief" within forested wetlands due to the presence of hummocks. Hummock dimensions

within cypress domes, bayheads, and mixed hardwood swamps were similar and found at a

density of 50 to 64 per hectare.

Hummock dimensions at Hal Scott were similar to values cited in the literature

while hummocks at FG-GSB2 were significantly larger. Average hummock length at FG-

GSB2 was 52 m and 1.2 m at Hal Scott. In a survey ofmicrotopography from seven

natural forested wetlands in North Central Florida, Sloan (1998) found a mean hummock

length of 1.13 m. Chimner and Hart (1996), in a study of hydrologic and

microtopographic effects on northern white cedar regeneration, found a mean hummock

length of 1.75 m. Davis et al. (1991) found a mean hummock length of 1.9 m.

The average height of constructed hummocks at FG-GSB2, 0.33 m, and Hal Scott,

0.17 m, were comparable to values reported in the literature for natural hummocks; 0.24

m (Sloan 1998), 0.21 m (Chimner and Hart 1996), and 0.32 m (Davis et al. 1991).

However, hummock heights alone are not as useful as when they are reported relative to

water levels. If hummocks are to provide additional microsites for plant colonization in

forested wetland systems, hummocks need to be at an elevation which allow them to be

exposed or inundated as water levels fluctuate. Hummocks well below mean high water

or continually flooded, may not provide elevated sites for colonization by less flood

tolerant plant species. Alternatively, if the objective for constructing hummocks is to

provide sites for wetland vegetation and the elevation of hummocks is much greater than

mean water level, hummocks may provide edaphic conditions more suitable for upland

rather than wetland species. Hummock elevations relative to mean water levels at FG-

GSB2 and Hal Scott were comparable to those reported for natural systems. As shown in

Table 7, the mean elevations of constructed hummocks at FG-GSB2 and Hal Scott are

approximately 0.21 m and 0.10 m above apparent mean water level, respectively. In a

survey of two cypress domes by Sloan (1998), hummocks were 0.28 m above mean high

water. In a survey of a bayhead wetland in Central Florida by Davis et al. (1991),

hummocks were 0.17 m above mean water level.

Depth of floodwater is an important concern in the survivorship and growth of

newly planted trees in constructed forested wetland projects. Depth of floodwater is

especially critical for planted seedlings since the water will often completely cover them,

resulting in significant mortality (Teskey and Hinckley 1977). Constructed hummocks of

appropriate heights may provide sites which allow planted trees to avoid prolonged

inundation, thus, increasing survivorship and growth rates. Mean hummock elevation at

FG-GSB2 and Hal Scott was estimated to be 0.21 and 0.10 m, respectively, above

apparent mean water for these systems. Numerous studies indicate that planted wetland

trees have increased survivorship and growth rates in conditions where soils were

saturated or only slightly inundated. Rushton (1988) studied how ranges of water table

depths affected the first year of growth in tubeling wetland tree species in constructed

wetlands in Florida. At a site which had water table depths ranging from -200 cm to +10

cm, Taxodium ascendens had the highest survival in locations where the water table was

from -10 to +10 cm. Taxodium distichum had the highest survival in two locations where

the water table was -50 to -30 cm and -10 to +10 cm. Nyssa sylvatica var. bilora had

the highest survival in locations where the water table was 10 to 30 cm below the ground

surface. Best and Erwin (1984) found that bald cypress, one of the most common species

planted in constructed forested wetlands, showed the highest growth rates and

survivorship when soils were saturated or flooded to depths less than 10 cm. Kennedy

(1970) reported increased height growth of newly planted water tupelo seedlings in water

depths less than 5 cm than when planted in water depths of 15 to 25 cm. Constructed

hummocks at FG-GSB2 and Hal Scott appear to be appropriate sites for planted trees by

providing sites which remain above or close to mean water levels. These sites, therefore,

may reduce the potential for inundation stress during the important early stages of growth

and survivorship.

Simulation Model

As shown in Figure 16a, progression of the system to a climax forested wetland is

dramatically slowed when no hummocks are incorporated into the wetland. However,

when hummocks accounted for 10 percent of the wetland surface area, the time required

to reach a climax wetland system was significantly reduced. The curves in Figure 16a

show that for every 10 percent incremental increase in hummock area, the greatest

increase in tree biomass is achieved when 20 percent of the wetland area is hummocks.

Using the hummock benefit index, this ratio of inundated to hummock area also maximizes

benefit in terms of water storage and biomass. As shown in Figure 17, weighting factors

based on biomass and water transformities were used to increase the importance of water

in the index. Maximum hummock benefit resulted when constructed hummocks were 20

percent of the wetland area for all ratios of the weighting factors. This suggests that this

may be the optimum amount of hummock area to incorporate into a newly constructed

forested wetland.

Limitations of the Study

Several factors associated with the sampling methodology used in this study may

have influenced results presented here. First, the clustering effect that occurred due to

transects radiating out from a central point, the laser level, sampled the central zone of a

particular node more intensively than the peripheral transect areas of the wetland. A more

appropriate method might have been to begin sampling a greater distance away from the

laser level in an attempt to reduce a concentration of sampling effort.

Random placement of the two sampling nodes per site at CF Industries' East and

West Lobe resulted in nodes being placed in areas which were relatively dry. Sampling in

the dry conditions witnessed at East and West Lobe may have negated the ability to study

vegetative trends induced by the interaction of hydrology and the systems

microtopographic heterogeneity. This is based on the assumption that the dry hydrologic

conditions at the time of sampling were responsible for the vegetative community present.

The small sample size used in studying the relationship between dominant canopy

vegetation and the rugosity index of microtopographic heterogeneity limits confidence in

results presented here. A larger sample size, particularly for systems where n=l, would

have also provided a more equal representation of system types.

Methodology used to survey hummock vegetation may have also influenced

results. Transect sections in wetland areas adjacent to constructed hummocks at FG-

GSB2 were not of consistent or equal length to transect sections placed on constructed

hummocks. This, in some cases, resulted in greater sampling effort on longer transects.

The absence of water level data for sites surveyed in this study limited results

presented here. Mean water levels were determined based on site visits and adventitious

rooting rather than monitoring wells or stage recorders. The use of wells or stage

recorders would have allowed a more accurate depiction of the relationship between

constructed hummock elevations and water levels as well as provide more insight into

relationships between the plant communities and microtopographic heterogeneity.

Summary and Recommendations

This study examined the development and role of microtopography in constructed

forested wetlands. Numerous studies have shown that microtopography is an important

component in the structural and functional organization of forested wetland systems.

Results from this study indicate that microtopography is not developing in constructed

systems. Even in the oldest sites surveyed, microtopographic heterogeneity was

significantly lower than that observed in natural systems. However, microtopography was

found to be developing in early successional communities dominated by Salix caroliniana

and Ludwigiaperuviana. These species are currently deemed as undesirable by the

regulatory community and efforts are made for their control. It may therefore be

counterproductive to control these species due to their apparent contribution to

microtopographic development. Results from this study also indicate that incorporating

microtopography, hummocks, into constructed forested wetlands increases species

richness and diversity of these systems. Furthermore, hummocks do not appear to provide

sites for the extensive establishment of nuisance or undesirable plant species.

Microtopography may be incorporated into constructed systems in a variety of

methods besides the construction of hummocks. One possibility lies in allowing an

irregular surface topography to exist following recontouring. Incorporating woody debris,

such as trees and limbs, from mined sites into the constructed wetland may be another

alternative to help increase the microtopographic heterogeneity of these systems.

APPENDIX A

SUMMARY OF MICROTOPOGRAPHIC DEVELOPMENT DATA FOR THE
TWELVE CONSTRUCTED SYSTEMS

IMC: EPR 1
IMC: EPR 1
IMC: EPR 1
IMC: EPR 1
IMC: EPR 1
IMC: EPR 1
IMC: EPR 1
IMC: EPR 1

Site
(Section 1)
(Section 1)
(Section 1)
(Section 1)
(Section 1)
(Section 1)
(Section 1)
(Section 1)

Transect
1.1
1.2
1.3
1.4
2.1
2.2
2.3
2.4

Avg
SD

Rugosity Index
(Sloan 1998)
101.5
103.8
101.0
101.0
101.8
100.6
101.9
101.0
101.6
1.0

Canopy
Coverage (%)
88.8
88.6
90.2
80.4
84.5
84.9
77.6
80.9
84.5
4.6

IMC: Parcel B
IMC: Parcel B
IMC: Parcel B
IMC: Parcel B
IMC: Parcel B
IMC: Parcel B
IMC: Parcel B
IMC: Parcel B

Avg
SD

100.4
100.3
101.4
100.5
100.3
100.3
100.4
100.2
100.5
0.4

85.9
82.9
92.0
92.8
87.5
90.8
80.8
94.2
88.4
4.9

0.85
0.53
0.97
1.14
0.88
0.82
1.05
0.86
0.89
0.18

Total 21

Diversity
0.87
1.03
0.62
1.10
1.22
0.94
1.07
0.96
0.98
0.18

Species
Richness
6
6
9
3
11
14
7
10
27

Total

Site
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch

Transect
1.1
1.2
1.3
2.1
2.2
2.3
2.4

Avg
SD

Rugosity Index
(Sloan 1998)
101.3
101.4
102.6
101.7
101.1
100.8
101.4
101.5
0.6

Canopy
Coverage (%)
86.7
90.2
84.4
81.5
89.1
86.5
88.8
85.7
3.0

IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.

Avg
SD

101.5
100.6
100.7
101.2
101.0
101.3
101.1
101.3
101.1
0.3

43.8
56.2
54.7
35.9
29.1
26.7
77.1
62.4
48.2
17.5

1.33
1.29
1.33
1.27
1.40
1.39
1.41
1.35
1.35
0.05

Total 44

Diversity
0.34
0.95
0.82
0.41
0.86
0.85
0.40
0.66
1.34

Species
Richness
2
7
5
2
6
6
2
15

Total

Site
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch
IMC: Hall's Branch

Transect
1.1
1.2
1.3
2.1
2.2
2.3
2.4

Avg
SD

Rugosity Index
(Sloan 1998)
101.3
101.4
102.6
101.7
101.1
100.8
101.4
101.5
0.6

Canopy
Coverage (%)
86.7
90.2
84.4
81.5
89.1
86.5
88.8
85.7
3.0

IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.
IMC: Jameson Jr.

Avg
SD

Diversity
0.34
0.95
0.82
0.41
0.86
0.85
0.40
0.66
1.34

Species
Richness
2
7
5
2
6
6
2
15

Total

101.5
100.6
100.7
101.2
101.0
101.3
101.1
101.3
101.1
0.3

43.8
56.2
54.7
35.9
29.1
26.7
77.1
62.4
48.2
17.5

1.33
1.29
1.33
1.27
1.40
1.39
1.41
1.35
1.35
0.05

Total

Site
IMC: Lizard Branch
IMC: Lizard Branch
IMC: Lizard Branch
IMC: Lizard Branch
IMC: Lizard Branch
IMC: Lizard Branch
IMC: Lizard Branch
IMC: Lizard Branch

Transect
1.1
1.2
1.3
1.4
2.1
2.2
2.3
2.4

Avg
SD

Rugosity Index
(Sloan 1998)
105.0
105.4
102.3
103.2
102.7
103.5
111.9
103.9
104.7
3.1

Canopy
Coverage (%)
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy

FG-GSB2
FG-GSB2
FG-GSB2
FG-GSB2
FG-GSB2
FG-GSB2
FG-GSB2
FG-GSB2

No Canopy
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy
No Canopy

Total 26

Diversity
1.10
1.09
1.06
1.02
0.88
1.11
0.91
1.05
1.03
0.09

Species
Richness
26
22
21
19
14
23
18
17
41

IMC:
IMC:
IMC:
IMC:
IMC:
IMC:
IMC:
IMC:

Total

100.3
100.7
100.5
100.6
100.4
100.4
100.2
100.2
100.4
0.2

Avg
SD

0.98
1.04
1.11
1.02
0.93
0.98
0.94
1.15
1.02
0.08

Avg
SD

Site

Cargill: SP-6
Cargill: SP-6
Cargill: SP-6
Cargill: SP-6
Cargill: SP-6
Cargill: SP-6
Cargill: SP-6
Cargill: SP-6

Transect
1.1
1.2
1.3
1.4
2.1
2.2
2.3
2.4

Avg
SD

Rugosity Index
(Sloan 1998)
100.7
101.4
100.9
104.4
101.5
101.9
102.0
101.3
101.7
1.2

Canopy
Coverage (%)
85.7
79.6
88.4
91.5
86.7
94.1
82.7
84.9
86.7
4.6

CFI: West Lobe
CFI: West Lobe
CFI: West Lobe
CFI: West Lobe
CFI: West Lobe
CFI: West Lobe
CFI: West Lobe
CFI: West Lobe

Diversity
1.24
0.93
0.94
0.97
1.03
1.11
0.94
0.87
1.00
0.11

Species
Richness
15
7
7
8
9
11
7
6
21

Total

1.1
1.2
1.3
1.4
2.1
2.2
2.3
2.4
Avg
SD

101.0
102.0
101.9
102.0
103.0
103.9
102.5
103.9
102.5
0.9

91.2
90.5
84.1
90.5
81.0
85.3
92.1
91.4
88.3
4.2

0.63
0.89
0.94
1.00
0.41
0.83
0.89
0.69
0.78
0.20

Total

Rugosity Index Canopy Species
Site Transect (Sloan 1998) Coverage (%) Diversity Richness

1.36
1.38
1.32
1.15
1.24
1.32
1.37
1.26
1.30
0.08

0.61
0.94
0.87
0.85
0.45
0.82
0.55
0.70
0.72
0.17

Total 37

100.3
100.1
100.7
100.3
100.2
100.3
100.4
100.2
100.3
0.2

104.7
103.2
103.6
103.8
102.1
102.8
105.1
104.2
103.7
1.0

CFI: R-9
CFI: R-9
CFI: R-9
CFI: R-9
CFI: R-9
CFI: R-9
CFI: R-9
CFI: R-9

Avg
SD

CFI: R-6
CFI: R-6
CFI: R-6
CFI: R-6
CFI: R-6
CFI: R-6
CFI: R-6
CFI: R-6

Avg
SD

68.4
63.0
59.7
49.8
35.7
22.8
33.9
29.7
45.4
17.1

71.6
79.0
73.3
71.7
64.8
70.2
75.5
67.5
71.7
4.4

Total 15

Site
CFI: East Lobe
CFI: East Lobe
CFI: East Lobe
CFI: East Lobe
CFI: East Lobe
CFI: East Lobe
CFI: East Lobe
CFI: East Lobe

Transect
1.1
1.2
1.3
1.4
2.1
2.2
2.3
2.4

Avg
SD

CFI: R-7
CFI: R-7
CFI: R-7
CFI: R-7
CFI: R-7
CFI: R-7
CFI: R-7
CFI: R-7

Avg
SD

Rugosity Index
(Sloan 1998)
102.5
102.9
104.1
102.1
104.4
101.6
103.3
100.9
102.7
1.2

100.7
101.3
100.4
102.2
102.0
100.6
100.6
101.0
101.1
0.7

Canopy
Coverage (%)
72.4
85.6
75.1
76.5
77.2
86.7
75.9
84.2
79.2
5.4

58.5
71.8
21.4
85.5
57.1
69.7
61.5
55.8
60.2
18.5

Diversity
1.16
0.97
1.04
0.98
1.05
0.59
1.15
1.00
0.99
0.18

0.84
1.13
0.99
0.89
0.84
0.94
1.14
1.21
1.00
0.15

Total

Total

Species
Richness
12
8
9
8
9
3
12
8
24

10
20
16
11
9
15
25
27
40

APPENDIX B

DESCRIPTION AND EVALUATION OF PATHWAY FLOWS, STORAGE, AND
COEFFICIENTS USED IN SIMULATION MINI-MODEL

Description of item
Sources and flows used in steady state calibrations:

Sunlight (renewable energies)

Sunlight flow

Storages and flows used in steady state calibrations:

Forested wetland:

Sunlight used by trees (percent)

Sunlight not intercepted by trees, remainder (percent)

Sunlight used by trees inundated (percent)

Sunlight used by trees on hummocks (percent)

Biomass

Gross production:

Production A (inundated)

Production B (on hummocks)

Respiration

Nuisance species:

Sunlight used by nuisance species (percent)

Sunlight not intercepted by nuisance species, remainder
(fraction)

Sunlight used by nuisance species inundated (percent)

Sunlight used by nuisance species on hummocks
(percent)

Biomass

Expression and Value

E=100

J=100

JT=70

Jr-30

J1=35

J2=35

T=15,400 kg/ha
(Mitsch and Gosselink 1995)

J3=62.5 (Rushton 1988)

J4=130 (Rushton 1988)

J5=192.5

JN=10

Jrr-20

J6=5

J7=5

N=100 kg/ha

Description of item
Gross production:

Production A (inundated)

Production B (on hummocks)

Respiration

Wetland surface area inundated
(a constant, fraction of 100)

Wetland surface area as hummocks (fraction of 100)
(a constant, fraction of 100)

Coefficient calculations:

Trees at steady state:

Sunlight used by inundated trees

Sunlight used by trees on hummocks

Gross production:

Production A (inundated)

Production B (on hummocks)

Respiration

Nuisance species at steady state:

Sunlight used by inundated nuisance species

Sunlight used by nuisance species on hummocks

Expression and Value

J8=45

J9=5

J10=50

I=96 (Davis et al. 1991)

H=4 (Davis et al. 1991)

kl*Jr*I*T = 35
kl = 8.1 E-07

k2*Jr*H*T = 35
k2 = 1.9 E-05

k3*Jr*I*T = 62.5
k3 = 1.4 E-06

k4*Jr*H*T = 130
k4 = 5.14 E-04

k5*T = 192.5
k5 = 0.125

k6*Jrr*I*N = 5
k6 = 2.6 E-05

k7*Jrr*H*N = 5
k7 = 6.3 E-04

Description of item
Production A (inundated)

Expression and Value
k8*Jrr*I*N=45
k8 = 2.3 E-04

Production B (on hummocks)

k9*Jrr*H*N = 5
k9 = 6.3 E-04

Respiration

kl0*N= 50
kl0 = 0.5

--

LITERATURE CITED

Barry W.J., A.S. Garlo, and C.A. Wood. 1996. Duplicating the mound-and-pool
microtopography of forested wetlands. Restoration and Management Notes,
14(1): 15-21.

Beatty, S. W. 1984. Influence ofmicrotopography and canopy species on spatial
patterns of forest understory plants. Ecology 65(5): 1406-1419.

Bell, D.T. 1974. Studies on the ecology of a streamside forest: composition and
distribution of vegetation beneath the tree canopy. Bulletin of the Torrey
Botanical Club 101(1):14-20.

Bersok, C.J. 1986. Conditions controlling wetland development on a phosphate-mined
reclamation site in Florida. M.S. thesis. Center for Wetlands, University of
Florida, Gainesville, FL.

Best, G. R. and K. L. Erwin. 1984. Effects ofhydroperiod on survival and growth
of tree seedlings in a phosphate surface-mined constructed wetland. 1984
Symposium on Surface Mining, Hydrology, Sedimentology, and Reclamation.
University of Kentucky, Lexington, Kentucky.

Bowmer, K.H. 1991. Management of aquatic plants in the botany wetlands, I. Survey,
biology, and review of options. Charles Stuart University, Wagga Wagga NSW,
Australia.

Bragazza, L., R. Alber, and R. Gerdol. 1998. Seasonal chemistry of pore water in
hummocks and hollows in a poor mire in the southern Alps (Italy). Wetlands
18(3):320-328.

Bridgham, S.D. and C.J. Richardson. 1993. Hydrology and nutrient gradients in
North Carolina Peatlands. Wetlands 13(3):207-218.

Brower, J.H., Zarr, J.H., van Ende, C.N. 1990. Field and LaboratoryMethods for
General Ecology. Wm. C. Brown Publishers, Dubuque, IA.

Chimner, R. A. and J. B. Hart. 1996. Hydrology and microtopography effects on
Northern White-Cedar Regeneration in Michigan's Upper Peninsula. Canadian
Journal. Of Forest Resources, 26:389-393.

Collins, B.S. and Pickett, S.T.A. 1982. Vegetation composition and relation to
environment in an Allegheny hardwood forest. American Midland Naturalist,
108:117-123

Collins, B.S., J.V. Perino, and J.L.Vankat. 1982. Woody vegetation and
microtopography in the bog meadow association of Cedar Bog, a west-central
Ohio fen. American Midland Naturalist, 108(2):245-249.

Dakshini, K.M. 1965. A study of the vegetation of Mothronwala Swamp Forest,
Dehra Dun, India. J. Ind. Bot. Soc. 44(4):411-428.

Davis, M.M., M.T. Brown, and G.R. Best. 1991. Vegetation and structural
characteristics of native ecological communities. Chapter 4. In, Techniques and
guidelines for reclamation of phosphate mined lands. Publication No. 03-044-
095. Florida Institute of Phosphate Research, Bartow, FL.

Davis, M.M., M.T. Brown, S. Doherty, and R.E. Tighe. 1991. Vegetation and
structural characteristics of constructed wetland and upland communities.
Chapter 9. In, Techniques and guidelines for reclamation of phosphate mined
lands. Publication No. 03-044-095. Florida Institute of Phosphate Research,
Bartow, FL.

Dinnis, W.M. and W.T. Batson. 1974. The floating log and stump communities in
the Sutee Swamp of South Carolina. Castanea 39:166-170.

Ehrenfeld, J.G. 1995. Microtopography and vegetation in Atlantic white cedar swamps:
the effect of natural disturbances. Canadian Journal of Botany 73:474-484.

Eldridge, D.J., M. Westoby, and K.G. Holbrook. 1991. Soil surface characteristics,
microtopography and proximity to mature shrubs: Effects on survival of several
cohorts of Atriplex seedlings. Journal ofEcology 78:357-364.

Farmer, R.E., M. Cunningham, and M.A. Barnhill. 1982. First-year development of
plant communities originating from forest topsoils placed on southern
Appalachian mine soils. Journal of Applied Ecology 19:283-294.

Golet, F.C., .J.K. Calhoun, W.R. DeRagon, D.J. Lowry, and A.J. Gold. 1993. Ecology
of Red Maple Swamps in the glaciated Northeast: A community profile. U.S.
Fish and Wildlife Service Biological Report 12.

Goodall, D.W. 1990. Ecosystems of the World 15; Forested Wetlands. Elsevier
Science Publishing Company INC., New York, N.Y..

Hamrick, J.L. and J.M. Lee. 1987. Effect of soil surface topography and litter cover on
the germination, survival, and growth of musk thistle (Carduws nutans).
American Journal of Botany, 74:451-457.

Hardin, E.D. and W.A. Wistendahl. 1983. The effects of floodplain trees on herbaceous
vegetation patterns, microtopography, and litter. Bulletin of the Torrey Botanical
Club, 110:23-30.

Harper, J.L., J.T. Williams, and G.R. Sagar. 1965. The heterogeneity of sol surfaces and
its role in determining the establishment of plants from seed. Journal of Ecology
53:273-286.

HillierH.G. 1973. Hilliers' Manual of Trees and Shrubs. Salix, Ed. 3. A.S. Barnes
Publishing, South Brunswick.

Huenneke, L.F. and R.R. Sharitz. 1986. Microsite abundance and distribution of
woody seedlings in a South Carolina cypress-tupelo swamp. The American
Woodland Naturalist 115(2):328-335.

Irwin, J.E. 1975. Structure of stump communities in a stream affected by thermal
effluent. M.A. Thesis, University of North Carolina, Chapel Hill, N.C.

Johnson, A. I. 1962. Methods of measuring soil moisture in the field. Geological
Survey Water-Supply Paper 1619-U, United States Government Printing Office,
Washington, D.C.

Keddy, P.A. and Ellis, T.H. 1984. Seedling recruitment of eleven wetland plant species
along a water level gradient: shared or distinct response? Canadian Journal of
Botany, 63:1876-1879.

Kennedy, H. E. 1970. Growth of newly planted water tupelo seedlings after flooding
and siltation. Forest Science 16:250-256.

Komiyama, A., T. Santiean, M. Higo, P. Patanaponpaiboon, J. Kongsangchai, and K.
Ogino. 1996. Microtopography, soil hardness and survival of mangrove
(Rhizophora apiculata) seedlings planted in an abandoned tin-mining area. Forest
Ecology and Management, 81:243-248.

Lamb, G.N. 1915. Willows: their growth, use, and importance. U.S. Department of
Agriculture, Bulletin No. 316, Professional Paper, Washington, D.C.

Lowe, E.F. 1983. Distribution and structure of floodplain plant communities in the
upper basin of the St. Johns River, Florida. Technical Publication SJ 83-8,
Department of Water Resources, St. Johns River Water Management District,
Palatka, FL.

Lowry, D.J. 1989. Restoration and creation ofpalustrine wetlands associated with
riverine systems of the glaciated Northeast, In J. A. Kunsler and M.E. Kentula,
eds. Wetland Creation and Restoration: The Status of the Science, Vol. I. U.S.
EPA, EPA 600/3-89/038a.

Odum, H.T. 1996. Environmental Accounting: EMERGY and Environmental Decision
Making. John Wiley and Sons, Inc., New York, NY.

Odum, H.T. 1994. Ecological and General Systems: An Introduction to Systems
Ecology. University Press of Colorado, Niwot, CO.

Ott, R-L. 1993. An Introduction to Statistical Methods and Data Analysis. Wadsworth
Publishing Company, Belmont, CA.

Parately, R. D. and T. J. Fahey. 1986. Vegetation-environment relations in a conifer
swamp in central New York. Bulletin of the Torrey Botanical Club 113:357-371.

Rawson, J.W. 1974. Willows. In, Shrubs and vines for northeastern wildlife. USDA
Forest Service, General Technical Report NE-9, Washington, D.C.

Reed, P.B. 1988. National list of plant species that occur in wetlands: 1988 Florida.
U.S. Fish and Wildlife Service, Biological Report 88(18.9).

Rehder, A. 1951. Manual of Cultivated Trees and Shrubs Hardy in North America.
MacMillan and Company, New York, N.Y.

Rheinhardt, R.D., M.M. Brinson, and P.M. Farley. 1997. Applying wetland reference
data to functional assessment, mitigation, and restoration. Wetlands 17(2): 195-
215.

Rich, Paul M. 1989. A Manual for Analysis of Hemispherical Canopy Photography.
Los Alamos National Laboratory, Los Alamos, N.M.

Richardson, S.G. and C.D. Johnson. 1998. Forested wetland restoration and
"nuisance" plant species management on phosphate mined lands in Florida.
Paper presented at the 1998 National Meeting of the American Society for
Surface Mining and Reclamation, St. Louis, MO.

Rushton, B.T. 1988. Wetland reclamation by accelerating succession. Ph.D.
dissertation. University of Florida, Gainesville, FL

Schlesinger, W.H. 1978. On the relative dominance of shrubs in Okeefenokee Swamp.
The American Naturalist 112(987):949-954.

Schumm, S.A. 1963. Sinuosity of alluvial rivers of the Great Plains. Geological
Society of American Bulletin, 74:1089-1100.

Sheldon, J.C. 1974. The behavior of seeds in the soil. 3. The influence of seed
morphology and the behavior of seedlings on the establishment of plants from
surface-lying seeds. Journal of Ecology 62:47-66.

Sloan, M. 1998. The role of microtopography in maintaining speciesdiversity. M.S.
Thesis, University of Florida, Gainesville, FL.

Smith, F. F., D.K. Smith, and G.W. Argus. 1978. Willows for pleasure and benefit.
American Horticulturist, 57:22-33.

Stauffer, A.L. and R.P. Brooks. 1997. Plant and soil responses to salvaged marsh
surface and organic matter amendments at a constructed wetland in Central
Pennsylvania. Wetlands, 17(1):90-105.

Stott, K.G. 1962. New uses for willow. Coedwigwt: The Forester, 4:83- 88.

Teskey, R.O.and T.M. Hinckley. 1977. Impact of water level changes on woody
riparian and wetland communities, Vol. I: Plant and soil responses to flooding.
Fish and Wildlife Service,U.S. Government Printing Office, Washingotn, D.C..

Titus, J.H. 1987. The effect of microtopography on seedling regeneration in a Florida
hardwood floodplain swamp. M.S. Thesis, University of Florida, Gainesville, FL.

Vivian-Smith, G. 1997. Microtopographic heterogeneity and floristic diversity in
experimental wetland communities. Journal of Ecology, 85: 71-82.

Warren-Wren, S.C. 1973. The Complete Book of Willows. A. S. Barnes and Co., Inc.,
Cranbury, N.J.

Wharton, C.J., H.T. Odum, K. Ewel, M. Duever, A. Lugo, R. Boty, J. Bartholome, E.
DeBellevue, S. Brown, M. Brown, and L. Duever. 1977. Forested Wetlands of
Florida: Their Management and Use. Center for Wetlands, University of Florida,
Gainesville, FL.

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
70	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file

MILLISECOND

CLASS.METHOD

MESSAGE

sobekcm_page_globals.constructor

Application State validated or built

sobekcm_database.verify_item_lookup_object

sobekcm_page_globals.constructor

Navigation Object created from URI query string

sobekcm_database.verify_item_lookup_object

sobekcm_page_globals.display_item

Retrieving item or group information

sobekcm_page_globals.get_entire_collection_hierarchy

Retrieving hierarchy information

sobekcm_assistant.get_entire_collection_hierarchy

cached_data_manager.retrieve_item_aggregation

Found item aggregation on local cache

item_aggregation_builder.get_item_aggregation

Found 'all' item aggregation in cache

system.web.ui.page.page_load (ufdc.page_load)

sobekcm_page_globals.constructor.on_page_load

html_echo_mainwriter.add_style_references

Adding style references to HTML

html_echo_mainwriter.add_text_to_page

Reading the text from the file and echoing back to the output stream

html_echo_mainwriter.add_text_to_page

Finished reading and writing the file