<%BANNER%>

Recovery of Cypress Domes in North Central Florida 10 Years after Harvest

Permanent Link: http://ufdc.ufl.edu/UFE0042216/00001

Material Information

Title: Recovery of Cypress Domes in North Central Florida 10 Years after Harvest
Physical Description: 1 online resource (42 p.)
Language: english
Creator: Ricci, Nicole
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: coppice, cypress, diameter, dome, forested, harvesting, height, regeneration, stump, wetland
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Pondcypress (Taxodium distichum var. nutans (Ait.) Sweet) can naturally regenerate via stump sprouting, which is considered an important component of regeneration following harvesting because sprouts produce cones within two years that provide a seed source for additional seedling recruitment. However, there are few studies on the long-term success of pondcypress stump sprouts and their potential to act as a seed source. In this study I examined the effect of felling equipment (sawhead vs. shearhead) on the long-term success of stump sprouts and the likelihood of sprout vs. shrub development on cut stumps in relation to felling treatment, stump height and stump diameter in seven harvested cypress domes located in north central Florida, USA. Each of the study sites was harvested in spring of 1999 and re-examined in summer of 2009. I measured stump sprout characteristics such as numbers of sprouts per stump, sprout height and diameter, and cone presence on stumps tagged during the original study. I evaluated proportions of stumps with sprouts and cones using a two-factor Anova with felling equipment and time as the main effects. I also determined whether stump characteristics and felling treatment affected the probability of a stump containing a sprout, shrub, or being bare using a general linear mixed model. Long-term sprouting success varied by site, ranging from15%-42% of stumps having at least one sprout. The type of felling equipment used did not significantly affect the proportion of stumps producing sprouts or sprouts producing cones, and the proportion of stumps with at least one sprout changed very little over time. This study confirmed initial findings that sprouts were more likely to develop on smaller diameter stumps ( < 30 cm) and stumps cut 20 cm below or near the mean high water. The occurrence of shrubs or another tree species on a stump increased as both stump diameter and stump height increased. Survival of sprouts over ten years shows that initial sprout success after two growing seasons can be used as an indicator of coppice regeneration. Recommendations for harvesting to maximize sprouting include cutting in close proximity of the historic mean high water mark and on stumps smaller than 30 cm.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nicole Ricci.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Bohn, Kimberly Kirsten.
Local: Co-adviser: Cohen, Matthew.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042216:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042216/00001

Material Information

Title: Recovery of Cypress Domes in North Central Florida 10 Years after Harvest
Physical Description: 1 online resource (42 p.)
Language: english
Creator: Ricci, Nicole
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: coppice, cypress, diameter, dome, forested, harvesting, height, regeneration, stump, wetland
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Pondcypress (Taxodium distichum var. nutans (Ait.) Sweet) can naturally regenerate via stump sprouting, which is considered an important component of regeneration following harvesting because sprouts produce cones within two years that provide a seed source for additional seedling recruitment. However, there are few studies on the long-term success of pondcypress stump sprouts and their potential to act as a seed source. In this study I examined the effect of felling equipment (sawhead vs. shearhead) on the long-term success of stump sprouts and the likelihood of sprout vs. shrub development on cut stumps in relation to felling treatment, stump height and stump diameter in seven harvested cypress domes located in north central Florida, USA. Each of the study sites was harvested in spring of 1999 and re-examined in summer of 2009. I measured stump sprout characteristics such as numbers of sprouts per stump, sprout height and diameter, and cone presence on stumps tagged during the original study. I evaluated proportions of stumps with sprouts and cones using a two-factor Anova with felling equipment and time as the main effects. I also determined whether stump characteristics and felling treatment affected the probability of a stump containing a sprout, shrub, or being bare using a general linear mixed model. Long-term sprouting success varied by site, ranging from15%-42% of stumps having at least one sprout. The type of felling equipment used did not significantly affect the proportion of stumps producing sprouts or sprouts producing cones, and the proportion of stumps with at least one sprout changed very little over time. This study confirmed initial findings that sprouts were more likely to develop on smaller diameter stumps ( < 30 cm) and stumps cut 20 cm below or near the mean high water. The occurrence of shrubs or another tree species on a stump increased as both stump diameter and stump height increased. Survival of sprouts over ten years shows that initial sprout success after two growing seasons can be used as an indicator of coppice regeneration. Recommendations for harvesting to maximize sprouting include cutting in close proximity of the historic mean high water mark and on stumps smaller than 30 cm.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nicole Ricci.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Bohn, Kimberly Kirsten.
Local: Co-adviser: Cohen, Matthew.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042216:00001


This item has the following downloads:


Full Text





RECOVERY OF CYPRESS DOMES IN NORTH CENTRAL FLORIDA 10 YEARS
AFTER HARVEST




















By

NICOLE RICCI


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

2010

































2010 Nicole Ricci
































This thesis is dedicated to my parents, Daniel and Catherine Ricci, for their never-
ending support throughout my academic career.









ACKNOWLEDGMENTS

This research and my graduate assistantship were funded by a grant provided by

the Environmental Protection Agency's' Wetlands Region Four Program awarded to Dr.

Kimberly Bohn. Equipment and one field site was provided by the University of Florida,

as well as scientific advice through faculty and staff in the School of Forest Resources

and Conservation. The Florida Division of Forestry provided one field site for revisit on

this study. The USDA Forest Service National Tree Seed Laboratory in Macon, GA

provided access to their equipment. And a special thank you to Victor Vankus for his

technical advice and housing during the end of my research. The Campbell Group

LLC., Rayonier Inc., and Foley Timber Company also provided access to field sites.

The completion of this research was successful due to the many people that

contributed their knowledge and support. First and foremost, I would like to express my

gratitude to Dr. Kimberly Bohn for her guidance, supervision, and advice from the early

stages of this research. One simply could not wish for a better or friendlier adviser. I am

thankful to the additional members of my supervisory committee, Drs. Katherine Ewel

and Matt Cohen for their methodological advice and expertise in their respective areas

of study.

Furthermore, I would also like to convey thanks to Chad Foster, for his "foster

javelin method" and wetland flora knowledge, Rose Godfrey for her sterling attention to

detail, and Brandon Raczkoski for his heroic assistance in the field. Their help no

matter the weather or water level was invaluable. Particular acknowledgement also

goes to Wendy Adams for all her indispensable help with travel forms and

administration. I would also like to acknowledge James Colee of IFAS Statistics, who

was incomparable as a statistical consultant.









Finally, I would like to thank my friends and family for their emotional and technical

support during this research.









TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ....................................................................................................... 4

LIST O F TA BLES ......... ................ ..................... ...... ............... 7

LIS T O F F IG U R E S .................................................................. 8

A B S T R A C T .............. ...... ........... ................. ....... .. ....................................... 9

CHAPTER

1 INTRODUCTION .................. ...... ......... ......... 11

2 METHODS.............................................. ........ 16

Study Area and Experimental Design ................ ............... ....... 16
Stump Sprout Remeasurements....................... ............... 17
Data Analysis ........................ ........................ 18

3 RESULTS ......... .......... ....... ......................... 19

Characteristics of 10-yr old Pondcypress Sprout Regeneration......................... 19
Treatment Effect on Sprouts and Cones through Time...................................... 19
Influence of Stump Characteristics on Likelihood of Sprouting ............................. 20

4 DISCUSSION .................. ...................... 22

Sprouting Response after 10 years ......... ........................ ...... 22
Potential for Sprouts to Provide Seed Source................................. .. ....... 23
Predictors of Vegetative Development on Cypress Stumps.................. 24
Management Recommendations ...... ................... .......... 26

APPENDIX

A SAS CODE FOR GLIMMIX PROCEDURE ANALYZING SPROUTS AND
CONES THROUGH TIME ............... ... ........................ 35

B SAS CODE FOR GLIMMIX PROCEDURE ANALYZING STUMP
CHARACTERISTICS ........................ ......... ......... 36

C PERCENTAGE OF SPECIES FOUND ON STUMPS................................ 37

LIST O F R EFER EN C ES .......................... ........... ......... .......................... 38

BIO G RAPHICAL SKETCH ............... ............. ........... ... ......................... 42



6









LIST OF TABLES


Table page

2-1 Mean number of pondcypress sprouts per stump during 2002 and 2009 in
stumps of seven pondcypress sites in north central Florida (Figure 2-1)............ 28

2-2 Summary statistics by site for proportion of pondcypress stumps with live
sprouts, proportion of sprouts with cones, and mean dbh and height of
sprouts in 2009 .............. ............. ... .......... ... ............... .............. 28

2-3 The effects of harvesting treatments, time, and their interaction, on the
proportion of pondcypress sprouts per stump in seven pondcypress sites in
north central Florida........................................ .......... 28

2-4 The effects of harvesting treatments, time, and their interaction, on the
proportion of pondcypress sprouts with cones per stump in seven
pondcypress sites in north central Florida .................................... ................ 29

2-5 The effects of harvesting treatments, stump characteristics, and the
interaction of treatments*stump height classes, on the probability of
pondcypress sprouts occurring on a stump in seven pondcypress sites in
north central Florida........................................ .......... 29

2-6 The effects of harvesting treatments, stump characteristics, and the
interaction of treatments*stump height classes, on the probability of a stump
being bare in seven pondcypress sites in north central Florida ..................... 29

2-7 The effects of harvesting treatments, stump characteristics, and the
interaction of treatments*stump height classes, on the probability of other
vegetation occurring on a stump in seven pondcypress sites in north central
Florida. ............................................. 29









LIST OF FIGURES


Figure page

2-1 Locations of pondcypress wetland study sites in north central Florida by
county ................ ................. .............. ............... 30

2-2 Study design two 0.05 ha circular plots to each harvesting method (sawhead
or shearhead) and 3 stump heights per whole plot........................................... 31

2-3 Mean and standard error of proportion of stumps with sprouts through time in
seven pondcypress domes in north central Florida. ............. ................. 32

2-4 Mean and standard error of cone production through time in seven
pondcypress domes in north central Florida...... ........ ..... ................. 32

2-5 Treatment effect on stump height in seven pondcypress domes in north
central Florida ............... ......... ....................... ... ......... 33

2-6 Effects of stump diameter on likelihood of having a sprout or shrub on a
stump or being bare in seven pondcypress domes in north central Florida........ 33

2-7 Percentage of species growing on stumps in seven pondcypress domes in
north central Florida........................................ .......... 34

2-8 Probability of a stump being bare or having a shrub growing on a stump in
seven pondcypress domes in north central Florida. ............. ................. 34









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

RECOVERY OF CYPRESS DOMES IN NORTH CENTRAL FLORIDA 10 YEARS
AFTER HARVEST

By

Nicole Ricci

August 2010

Chair: Kimberly Bohn
Major: Forest Resources and Conservation

Pondcypress (Taxodium distichum var. nutans (Ait.) Sweet) can naturally

regenerate via stump sprouting, which is considered an important component of

regeneration following harvesting because sprouts produce cones within two years that

provide a seed source for additional seedling recruitment. However, there are few

studies on the long-term success of pondcypress stump sprouts and their potential to

act as a seed source. In this study I examined the effect of felling equipment (sawhead

vs. shearhead) on the long-term success of stump sprouts and the likelihood of sprout

vs. shrub development on cut stumps in relation to felling treatment, stump height and

stump diameter in seven harvested cypress domes located in north central Florida,

USA. Each of the study sites was harvested in spring of 1999 and re-examined in

summer of 2009. I measured stump sprout characteristics such as numbers of sprouts

per stump, sprout height and diameter, and cone presence on stumps tagged during the

original study. I evaluated proportions of stumps with sprouts and cones using a two-

factor Anova with felling equipment and time as the main effects. I also determined

whether stump characteristics and felling treatment affected the probability of a stump

containing a sprout, shrub, or being bare using a general linear mixed model.









Long-term sprouting success varied by site, ranging from 15%-42% of stumps

having at least one sprout. The type of felling equipment used did not significantly

affect the proportion of stumps producing sprouts or sprouts producing cones, and the

proportion of stumps with at least one sprout changed very little over time. This study

confirmed initial findings that sprouts were more likely to develop on smaller diameter

stumps (< 30 cm) and stumps cut 20 cm below or near the mean high water. The

occurrence of shrubs or another tree species on a stump increased as both stump

diameter and stump height increased.

Survival of sprouts over ten years shows that initial sprout success after two

growing seasons can be used as an indicator of coppice regeneration.

Recommendations for harvesting to maximize sprouting include cutting in close

proximity of the historic mean high water mark and on stumps smaller than 30 cm.









CHAPTER 1
INTRODUCTION

Adequate regeneration of second-growth cypress is important for both ecological

and economical reasons. Cypress timber has long been valued for its use as cross ties,

ladders, pilings, stakes, post wood and narrow boards (Wilhite and Toliver 1990) and

more recently as mulch for gardening. However, cypress swamps have multiple values

beyond timber production, including wildlife habitat, ground water recharge, flood

control, and water quality improvement (Ewel 1998). Recently, concerns over

ecosystem recovery following harvesting have prompted a need for better

understanding of regeneration dynamics in a variety of forested wetland systems in

order to ensure a quick recovery of these systems following harvesting disturbance.

While baldcypress (Taxodium distichum (L.) Rich.) was highly valued for its high

proportion of decay-resistant heartwood (Wilhite and Toliver 1990), most of the large

older baldcypress trees that contained valuable heartwood has been harvested and

much of the remaining cypress is pondcypress (Taxodium distichum var. nutans (Ait.)

Sweet). With the sustained demand for cypress timber, and the improvements in

harvesting equipment, loggers have been able to access a variety of forested wetland

systems including small isolated cypress domes where pondcypress is frequently found.

These domes are typically embedded within pine flatwood communities and

occasionally intermixed within hardwood hammocks. The pondcypress is found in

association with swamp tupelo (Nyssa sylvatica var. biflora), slash pine (Pinus elliottii

Englem.), and various bays (Gordonia lasianthus (L.) Elis, Persea palustris (Raf.) Sarg.,

Magnolia virginiana L.) which regenerate on hummocks or along shallow edges

(Wharton et al. 1977, Brandt and Ewel 1989).









Additionally, it is now common for trees of all merchantable sizes to be harvested

because trends in the commercial usage of cypress have converted from lumber to pulp

and mulch for gardening (Ewel 1998). Logging operations typically involve whole-tree

harvesting of cypress (Duryea and Hermansen 1997). Though some studies have

documented that that pondcypress domes can regenerate back to their original basal

area and species composition by 50 years after harvesting (Terwilliger and Ewel 1986),

the majority of the literature has focused on baldcypress regeneration. Early survival of

pondcypress regeneration has been addressed in previous studies (Terwilliger and

Ewel 1986, Ewel et al. 1989, Ewel 1996), but the long-term success has not been

evaluated.

One unique characteristic of cypress is that it is one of the few conifer species that

can naturally regenerate via seed germination or vegetatively by stump sprouting after

harvest (Wilhite and Toliver 1990). Seed dissemination occurs during the fall to early

winter, and water and wildlife movement are the main dispersal mechanisms within

cypress domes. Seedlings need saturated but not submerged soil to germinate and

prefer full sun for vigorous growth. Competition from shrubs or a dense understory can

inhibit cypress regeneration by decreasing sufficient light and nutrients (Ewel 1989).

Limited seedling recruitment across a site can also be attributed to factors like poor

seed viability from parent stock, lack of seed sources after harvest, inability of seed to

be dispersed, and hydrologic limitations (Gunderson 1984, Schneider and Sharitz 1986,

Conner et al. 1986).

Stump sprouts are considered an important mechanism for immediately

regenerating a cypress stand after harvesting (Ewel et al. 1989, Randall et al. 2005). In









species that stump sprout, initial growth of sprouts is likely to be greater than seedling

regeneration due to the root systems of established trees and the amount of stored

carbohydrates in the root system (McCreary et al. 2002). Stump sprouts have vascular

connections between the sprouts themselves and the parent tree root system, with each

stem helping to sustain a portion of the root system (Kramer and Kozlowski 1979).

Conner et al. (1986) found that baldcypress seedling regeneration grew on average to

91 cm and stump sprouts grew to 152 cm in height after four growing seasons.

Additionally, since stump sprouts can produce cones as early as two years after

sprouting, this could provide an additional seed source for seedling recruitment when

environmental conditions are conducive for germination, so long as the seed source is

viable (Ewel 1998, Randall et al. 2005).

There are still several concerns regarding the long-term success of sprouting as a

mechanism for regenerating pond cypress in wetland systems. It is not clear whether

sprouts can survive to be recruited into the main canopy, and low sprout survival

between 17 and 23% has been observed in some cases (Conner et al. 1986, Conner

1988, Ewel 1996). In comparison, survival of bare-root pondcypress seedlings planted

in eight harvested swamps in north Florida ranged from 37% to 89% after four years

(Vince, Duryea and Randall unpub. data). Additionally, poor stem form and mechanical

damage from ground-based logging can limit vegetative reproduction (Hook and DeBell

1970, Lockaby et al. 1997). Stump decay could compromise sprouts although decay

may not necessarily move into the new sprout, particularly where the sprouts develop

off the side of the stump (Shigo, 1986). Moreover, sprout regeneration can be spatially









heterogeneous and sometimes insufficient in providing a continuous seed source

across the entire stand (Kennedy 1982, Keim et al. 2006).

While long-term survival of sprouts is affected by a multitude of factors, initial

establishment of sprouts are often a function of stump dimensions. Stump height and

diameter can influence the vigor of sprouting success regardless of species (Hook and

DeBell 1970, Kennedy 1982, Ewel 1996, Gardiner et al. 2000). For flood-prone species

stump height in relation to mean water depth has shown a strong correlation to sprout

establishment (Terwilliger and Ewel 1986) though relationships may vary by species.

Kennedy (1982) found that water tupelo cut 76 cm above the mean high water mark had

the highest survival of sprouts; while Ewel (1996) indicated stump less than 70 cm tall

from the ground yielded greater pondcypress sprout survival. Parent tree size can also

influence stump sprouting. Early sprout survivorship is greatest on smaller stumps

(Ewel 1996), while larger stumps over 40 cm in diameter did not show cypress sprout

survival after harvest (Keim et al. 2006).

Stumps may also provide a safe site for germination and establishment of other

plants such as shrubs or other tree seedlings which could potentially interfere with

sprout development. Most shrub species cannot tolerate being submerged in the early

stages of development and would naturally be recruited on stump surfaces above water.

Wetland shrub species such as Itea sp. prefer stumps because they are elevated above

floodwaters (Anderson et al. 2009), and dead or decaying stumps have also been

observed to serve as a foundation for shrub or bay establishment (Schlesinger 1978,

Casey and Ewel 2006). Furthermore, shrub invasion was noted on stumps in Louisiana

cypress-tupelo swamps decades after harvest (Conner et al. 1981). Whether a shrub is









recruited is on a stump may also be a function of the parent tree size. However, there

has been no definitive quantification of relating stump size and shrub occurrence.

Long-term monitoring is needed to better understand the recovery of cypress

domes after logging, yet there are few studies that have monitored sprouts or

regeneration success longer than six growing seasons (McCreary 2002). This work

builds upon initial two-year results reported by Randall et al. (2005). That study

examined, parent tree size, height of stump cut, and felling equipment on stump

sprouting (Randall 2001). Stump heights were cut below, at, or above mean high water,

with either a hydraulic shear-head or continuous disk saw-head on the interchangeable

head of the feller-buncher. The saw-head cuts with a disk blade and the shear-head,

which can be compared to pruning shears, snips the tree off at the appropriate stump

height. Using the saw-head attachment creates a smoother cut across the stump, while

stumps cut with a shear-head may increase damage by splitting or shattering stumps

(Porter et al. 1984). In general, two seasons after harvest, results indicated that greater

sprouting occurred on smaller diameter stumps close to the mean high water mark, and

sprout establishment was not significantly affected by felling equipment (Randall et al.

2005). These study sites were reinvestigated to assess the long-term survival of

cypress stump sprouts ten years after harvest in relation to felling method (sawhead

type) and to quantify the proportion of stumps with cone-bearing sprouts. An additional

objective of this study was to determine the likelihood of sprout vs. shrub development

as a function of felling equipment, stump diameter, and stump height.









CHAPTER 2
METHODS

Study Area and Experimental Design

This investigation was conducted in sites that were selected in 1999 to evaluate

factors that influenced stump sprouting in pondcypress trees in eight isolated cypress

domes. In 2009, the eight logged pondcypress domes were re-visited in north-central

Florida (Figure 1) (latitudes: 290 05' to 290 55'; longitudes: 820 10' to 820 35'). However,

after field investigations were completed, one site in Lafayette County was dropped due

to difficulty in stump identification and relocation. Prior to harvest, pondcypress and

swamp tupelo were the predominant species, along with slash pine, myrtle holly and

various bays in the canopy. Historical climate data ranging from 1950 2010 in each

county that contained a field site indicated that mean annual precipitation varied from

124 cm to 150 cm (Southeast Regional Climate Center 2010).

The original study used a split-plot design (Figure 2). Two 0.05 ha circular plots

were established on each half of the pondcypress dome and each half of the dome was

cut using a sawhead or shearhead harvesting technique. This was the whole-plot

treatment. Within each 0.05 ha plot, 45 trees were chosen to be cut at one of three

stump cutting heights as the subplot treatment. Heights were based on mean high

water level within that swamp. This was determined from stain and lichen lines on the

cypress trees (Randall et al. 2005). Stump heights were categorized as being cut low

(20 cm below mean high water), medium (at mean high water), and high (40 cm above

mean high water). Within each plot, permanent aluminum tags were used to mark each

study stump, and locations within each plot were mapped.









The pondcypress domes were clearcut in February and March of 1999 in

accordance with Florida's Best Management Practices in forested wetlands at that time,

which left 12 reserve trees per hectare. Trees outside the research plots were

harvested under operational standards. Stump height and diameter were recorded after

logging, and sprouting success was initially monitored in the summer of 1999 through

fall of 2002 (Randall et al. 2005).

Stump Sprout Remeasurements

The relocation of stumps from the previous study was conducted in May-July of

2009 using hand drawn plot maps and leave trees for orientation. We recorded whether

stumps were bare, had cypress sprouts, or had other vegetation growing on the stump.

If vegetation other than cypress was growing on the stump, the species was noted. For

stumps with sprouts, we recorded the number of live sprouts on each stump, the height

of all sprouts, and the dbh of stems > 5 cm (2 in) (stem diameter was measured at 1.4

meters from ground). We also noted whether or not each sprout was producing cones.

Eighty-seven percent of the initial stumps were relocated across all sites using plot

maps from the original study. Out of the 168 missing stumps from the original study, 75

(45%) were from sawhead treatment and 93 (55%) were from shearhead treatment out

of the total stumps in each category. A t-test confirmed that differences in numbers of

missing stumps across sites were not significantly different by treatment so were not

considered to be confounding factor in the subsequent analyses. Not surprisingly,

water level affected our ability to relocate some stumps. Stump cut height of missing

stumps was predominantly around mean high water level (65%), with the remaining

amount being cut at the 20 cm below mean high water (30%) and 5% cut 40 cm above

mean high water.









Data Analysis

Proc Glimmix (SAS 9.2) was used for all data analysis because this particular

model properly assesses random (swamp and plot) and fixed effects (treatment, time,

stump characteristics), and has the ability to estimate the appropriate standard errors

for binary response data. Individual swamps were used as a blocking factor All

statistical tests were run at a = 0.05.

To assess the number of sprouts per stump, the proportion of stumps with sprouts

and proportion of sprouts with cones, we used a generalized mixed model with

treatment and time as fixed-effect factors (Appendix A). The two time periods were from

the field observations in 2002 and 2009. Proportions of stumps with sprouts in 2009

were calculated only based on the total number of relocated stumps. Adjusted means

were plotted to visualize the change in sprouts or cones in time.

A mixed model with a binomial distribution was analyzed to determine the

probability of a stump having sprouts, other vegetation or nothing on it as a function of

stump height, diameter, treatment, and interactions of those factors across all sites,

(Appendix B). Stump diameters were broken into 5 categories that contained

approximately the same number of stumps. Diameter categories were 5 20 cm, 20-29

cm, 30-39 cm, 40-49 cm, and > 50 cm. Sprouting was not evident on stumps over 50

cm, and used as a cut off point for categorical analysis. Additionally, stump height was

categorized in the three cut height classes of the original study: 20 cm below mean high

water, at mean high water, and 40 cm above mean high water.









CHAPTER 3
RESULTS

Characteristics of 10-yr old Pondcypress Sprout Regeneration

The average number of sprouts per stump decreased over time (Table 2-1). In

2002, the number of sprouts per stump averaged 15 sprouts on sawhead stumps and

12 on shearhead stumps. High values were due to an anomaly at one site (F1), which

had an unusually large number of sprouts per stump, particularly for the sawhead

treatment (Table 2-1). Removing this site would result in similar numbers of sprouts per

stump (about 11) for both treatments in 2002. In 2009, the mean number of sprouts per

stump decreased to 7 for both harvest treatments. Changes in the number of sprouts

per stump with the removal of site F1 were not statistically significant between felling

treatments.

Individual swamps showed great variability in sprout dimensions and proportion of

stumps with sprouts. The mean diameter of the largest sprout per stump ranged from

7.1 to 18.1 cm across all of the sites, and mean height ranged from 5.3 to 7.1 m (Table

2-2). There was no correlation between the three stump heights used in the original

study and sprout characteristics such as the number of sprouts on a stump, sprout dbh

or height. The percentage of stumps with live sprouts varied between 10-40% in 2002

and between 11-41% in 2009.

Treatment Effect on Sprouts and Cones through Time

Time had a statistically significant effect on stump sprout success but felling

treatment did not (table 2-3). However, the change in proportion of stumps with sprouts

was likely not ecologically relevant, as the percentage of stumps with live sprouts in

2002 was 28% and rose only slightly to 31% in 2009, primarily because of additional









stumps with sprouts found on the sawhead cut stumps (Figure 2-3). This indicated that

stump sprouts primarily develop only within the first two years after harvesting.

The percentage of sprouts with cones increased significantly (p=0.0002) from the

last period of monitoring in 2002 to 2009, but was not significantly influenced by

treatment (Table 2-4). The proportion of sprouts with cones was between 10 and 13%

for shearhead and sawhead stumps in 2002 and 17 and 26% for those treatments in

2009 (Figure 2-3).

Influence of Stump Characteristics on Likelihood of Sprouting

We calculated the probability, and tested the significance, that a stump of a given

size (diameter and height class) and felling treatment would have sprouts, other

vegetation or nothing growing on it 10 years after harvesting. Diameter was broken into

5 classes based on the range of diameters of sprouts that the majority of stump sprouts

occurred. Height classes were based on the initial study categories related to mean

water depth. Initially the interaction between stump diameter x height was tested, but it

was dropped when all insignificant and were removed from the final analysis.

Evaluation of stump characteristics (diameter and height classes) and harvesting

technique were conducted using only those stumps relocated in 2009 since that is the

only time period for which data on other vegetation on stumps was recorded. A

significant interaction effect between treatment x stump height was only found for

stumps with sprouts (Figure 2-5, Table 2-5) (p=0.048). Multiple comparisons indicated

that stumps cut with the shear head at mean high water and 40 cm above mean high

water were significantly different from each other and significantly different from the

sawhead stumps cut at the same heights. Harvesting technique did not significantly









affect the likelihood of a stump being bare or having a shrub growing on it. (Table 2-6,

Table 2-7).

Stump diameter significantly affected the probability of a stump having a cypress

sprout growing on it (p<0.0001). There was a decrease in the likelihood of a sprout

occurring on a stump as the stump diameter increased (Figure 2-6). Sprouting was

more likely to occur on stumps within a diameter range of 0-20 cm. Stump diameter also

significantly affected the likelihood of a stump being bare (p=0.0004) or having other

vegetation (p<0.0001). The likelihood of a stump being bare nearly doubled as diameter

increased from less than 20 cm wide to 21 cm and greater (Figure 2-6). There was no

statistical difference in the likelihood of a stump being bare for stumps between 21 and

50 cm. The likelihood of a stump having other vegetation on it increased significantly

with diameter (Figure 2-6). The highest probability of a shrub occurring on a stump was

at stump diameters of 30 cm and greater. Lyonia lucida and Myrica cerifera were the

dominant shrub species found on stumps (Figure 2-7).

Stump height had no significant effect on the likelihood of stumps being bare

(Table 2-6) but did significantly affect the likelihood of having a shrub occur on a stump

(p<0.0001) (Figure 2-8). Stumps cut at 40 cm above mean high water were the most

favorable for shrub recruitment.









CHAPTER 4
DISCUSSION

Sprouting Response after 10 years

Cypress stump sprouts are an important source of regeneration following logging

in forested wetlands. Stump sprouts can have an early growth advantage over

regeneration by seeds; however, there has been concern about the longevity of coppice

sprouts in other species,and their ability to be recruited into the overstory (Mostacedo et

al. 2009). Additionally, it is not clear how survivorship and growth of pondcypress

sprouts compares to that of baldcypress sprouts. Visser and Sasser (1995) found that

the recruitment of baldcypress second growth trees did not keep up with mortality;

however, findings from our study suggest that 10 years after harvesting, survival of

coppice sprouts was high. All stumps that sprouted immediately following harvesting

still had at least one if not several dominant sprouts 10 years later. Pondcypress sprout

sizes measured in this study were consistent with those reported for baldcypress of a

similar age. On average, dominant pondcypress sprouts in our study sites were 10 cm

in diameter and over 4.7 m tall. Keim et al. (2006) found that baldcypress coppice

sprouts from 10-50 years after harvesting averaged 10 cm in diameter and 6.8 m tall.

Additionally, the number of sprouts per stump decreased over time though this was

expected because dominant sprouts will eventually outcompete the others. Kennedy

(1982) found a decrease in water tupelo stump sprouts over 6 growing season that led

to an average of 5 sprouts per stump.

It has been suggested that sprouts originating from the top or cut surface of the

stumps will not develop into quality timber producing trees (Conner et al. 1986). Tree

form was not analyzed in our study, but casual observation suggested that the majority









of surviving sprouts did, in fact, regenerate as a top sprout. The cypress sprouts in our

field sites appeared to be growing in good form at this time, but it is too soon to

determine their potential as a timber source.

Harvesting technique did not have a significant impact on sprout regeneration and

survival. The average number of sprouts per stump was virtually the same for both

treatments. It had been originally hypothesized that long-term survival would be notably

higher on sawhead cut stumps due to shear-related damage by the shearhead feller.

The stumps of trees can be damaged by a shearhead in the form of ring-shake, stump-

pull, shatter, and splitting resulting in an increased risk of disease and decay because of

the amount of damage caused by the blades (Porter et al. 1984). The severity of

splitting damage can damage the stump, providing an entry-way for pathogens that

eventually translocate to the coppice sprouts. However, our results indicate that at least

10-year survival of coppice sprouts was not affected by felling technique.

Potential for Sprouts to Provide Seed Source

Cypress sprouts are also considered important for providing and additional seed source

for seedling recruitment after harvesting. At our study sites, approximately one quarter

of the sprouts were cone-bearings, and a preliminary seed viability test from cone

samples taken in early October of 2009 indicated that on average 41% of the seeds

were viable (varying from 20% to 58% across sites). Both pondcypress and baldcypress

are considered to have similar reproductive characteristics, and several factors may

contribute to variability and magnitude of seed viability of these trees in any given year.

Mattoon (1915) stated that baldcypress produces a larger crop of seeds approximately

every 3 years. Other factors such as drought or overall health of the tree may also affect

seed viability in any given year. There is very little literature on viability of cypress









seeds in reference conditions, although at least one other study has reported bald

cypress seeds had consistently low viability over a three-year period (Sharitz et al.

1986). Although few naturally recruited seedlings were observed in the study sites,

additional studies would be needed to determine if this was due to inadequate seed

source, inadequate safe sites for germination, or some other factors.

Predictors of Vegetative Development on Cypress Stumps

Stump dimensions were better predictors than felling technique alone on the

probability that the stump would have a cypress sprout versus some kind of other

vegetation growing on it. For instance, stump diameter was a significant predictor of

whether a stump would have a sprout or a shrub on it. My results were not unexpected

as this trend had already been documented for cypress in several stands (Ewel 1996,

Randall 2005) and for many other hardwood species (Golden 1999, Sands and Abrams

2009). Pondcypress of saw-timber size with a diameter larger than 40 cm produce fewer

sprouts (Keim et al. 2006). Diameter is an important factor because older and larger

trees have fewer basal dormant buds (Kauppi et al. 1988).

Stump height in relation to mean high water level has also been noted to be an

important predictor of successful sprouting in swamps and flood-prone sites, and in this

study stump height was also a significant predictor of the presence of either sprouts or

other vegetation on stumps. Bald cypress stump sprouts were more likely to survive on

shorter stumps (Keim et al. 2006), and in this study pondcypress sprouts were

successful on stumps both at or 20 cm below mean high water. Water levels were not

monitored under the duration of our study, but drought conditions were present during

the initial study when most sprouts regenerated (Randal 2001). The low cut stumps

may be submerged in years with typical precipitation, but typical effects of inundation on









stump sprout survival and growth on those cut stumps may not have occurred on our

study sites. Stump sprouts at our study sites had an opportunity to establish before

normal to high water levels occurred again. Perhaps a higher mortality for low cut

stump sprouts due to inundation stress would have been evident across all sites if these

dry conditions did not occur (Ewel 1996). Because of those conditions, Randall (2001)

suggested from observations outside the study plots, to cut cypress around the mean

high water mark.

One concern of logging is that cypress swamps may convert to hardwood or shrub

dominated swamps due to human and environmental influences. Most of the

understory plants and shrubs in cypress and tupelo forests occur on elevated sites,

which protect susceptible seeds and seedlings from flooding (Huenneke and Sharitz

1986). In Louisiana cypress-tupelo swamps, shrub invasion occurred where the

canopy opened over elevated micro sites, including stumps or logs (Conner et al. 1981).

Short-term studies suggested that harvesting in baldcypress and water tupelo swamps

could lead to a conversion to hardwood species (Gellerstedt and Aust 2004). Conner et

al. (1986) also found that after logging, all sites were dominated by small, shrubby, non-

timber species after 1 to 4 growing seasons. On the other hand, other studies, such as

those in Okefenokee Swamp indicate that shrubs only rooted on dead stumps and did

not invade the entire site (Schlesinger 1978). At our study sites, shrub or other tree

species tended to grow on stump heights greater than 40 cm above the mean high

water level. Lyonia lucida and Myrica cerifera were the most prevalent shrubs found on

stumps, and these species are typical of pondcypress stumps in the southeast.









Shrubs are faster growing, and may have an advantage over cypress seedlings or

might compete with sprouts on stumps. However stumps that do contain sprouts

usually do so because of the rapid early growth of stump sprouts (Johnson 1978). In

this study, we found relatively few, if any stumps that had both sprouts and shrubs

growing on them. It is possible that the shrubs developed on taller stumps that were

simply not conducive to sprout development early on; however we do not have data

from 2002 to determine whether competition existed during the early phases of

vegetation development.

Management Recommendations

In our study, cypress coppice sprouts survived over ten years past harvest. These

results suggest that early coppice regeneration counts during the first few years post-

harvesting should provide an appropriate assessment of the coppice regeneration

potential of that stand. The position of sprouts on stumps, and the condition of sprouts

were not recorded in this study. Future condition and establishment position may result

in additional sprout mortality in the future as sprouts develop into mature trees.

Our recommendations regarding the size of trees to cut in order to increase the

probability of stump sprouting remain consistent with those of previous studies. Trees

of a smaller diameter (< 30 cm) should have a better sprouting response than larger

trees. If a site contains only larger cypress trees, sprouting may occur at low levels.

Other recommendations suggest that he greatest likelihood of sprout success will occur

when stumps are cut at or near the high water mark; although sprout success on

stumps cut below the mean high water mark were high in this study, we attribute this to

the extreme drought conditions immediately following harvest. Still, cutting at mean high

water levels appears to be a safe recommendation unless accurate predictions of









climate within the few years following harvesting (when sprouts develop) could be

obtained.









Table 2-1. Mean number of pondcypress sprouts per stump during 2002 and 2009 in
stumps of seven pondcypress sites in north central Florida (Figure 2-1).
Number of sprouts per stump
Cutting
Year Method AC F1 F2 G P1 P2 R
2002 sawhead 9 43 9 12 12 10 11
shear 8 23 8 17 12 11 13
2009 sawhead 3 8 7 10 7 7 10
shear 6 7 5 10 8 8 8


Table 2-2. Summary statistics by site for proportion of pondcypress stumps with live
sprouts, proportion of sprouts with cones, mean, dbh, and height of sprouts in
2009.
Mean measurement by site
Sprout Measurement AC F1 F2 G P1 P2 R
Proportion stumps w/ sprouts
(2002)a 10% 40% 27% 39% 31% 24% 28%
Proportion stumps w/ sprouts
(2009)a 11% 39% 34% 41% 32% 27% 29%
Proportion sprouts w/ cones
(2002)c 17% 9% 49% 19% 47% 54% 68%
Proportion sprouts w/ cones
(2009)c 55% 40% 67% 56% 46% 70% 71%

Mean/ SE dbh of largest sprout 8.7 8.5 11.1 7.1 8.5 18.1 9.5
(cm)b (0.61) (0.37) (0.43) (0.25) (0.31) (1.65) (0.34)

Mean/ SE height of tallest sprout 5.8 5.0 7.1 5.3 6.0 6.2 6.1
(m)b (0.36) (0.20) (0.18) (0.15) (0.17) (0.16) (0.19)
a (n=1222)
b (n=1070)
c (n=1070)


Table 2-3. The effects of harvesting treatments, time, and their interaction, on the
proportion of pondcypress sprouts per stump in seven pondcypress sites in
north central Florida.
Effect on proportion of stumps w/ sprouts No. DF F Value Pr> F
Treatments 1 2.73 0.11
Time 1 14 <0.001
Treatments*Time 1 1.29 0.27









Table 2-4. The effects of harvesting treatments, time, and their interaction, on the
proportion of pondcypress sprouts with cones per stump in seven
pondcypress sites in north central Florida.
Effect on Proportion of Sprouts with Cones No. DF F Value Pr> F
Treatments 1 1.71 0.21
Time 1 19.27 0.002
Treatments*Time 1 0.31 0.58


Table 2-5. The effects of harvesting treatments, stump characteristics, and the
interaction of treatments*stump height classes, on the probability of
pondcypress sprouts occurring on a stump in seven pondcypress sites in
north central Florida.
Effect on probability of stumps w/ sprouts No. DF F Value Pr> F
Treatments 1 5.8 0.02
Stump Diameter Classes 4 42.69 <0.001
Stump Height Classes 2 16.94 <0.001
Treatments*Stump Height Classes 2 3.06 0.048


Table 2-6. The effects of harvesting treatments, stump characteristics, and the
interaction of treatments*stump height classes, on the probability of a stump
being bare in seven pondcypress sites in north central Florida.
Effect on probability of bare stumps No. DF F Value Pr> F
Treatments 1 0.5 0.49
Stump Diameter Classes 4 4.65 <0.001
Stump Height Classes 2 2.59 0.08
Treatments*Stump Height Classes 2 1.09 0.34


Table 2-7. The effects of harvesting treatments, stump characteristics, and the
interaction of treatments*stump height classes, on the probability of other
vegetation occurring on a stump in seven pondcypress sites in north central
Florida.
Effect on probability of stumps w/ other vegetation No. DF F Value Pr> F
Treatments 1 2.33 0.14
Stump Diameter Classes 4 27.76 <0.001
Stump Height Classes 2 19.1 <0.001
Treatments*Stump Height Classes 2 0.41 0.67


















































30 15 0 30 Miles


Figure 2-1. Locations of pondcypress wetland study sites in north central Florida by
county.










































1il


3 Stump Heigts: 20 cm beow, mean,40 ~c above mean high wata

ShaWow well


Leavetrees









Figure 2-2. Study design two 0.05 ha circular plots to each harvesting method (sawhead
or shearhead) and 3 stump heights per whole plot.














-*- Sawhead


S -Shearhead


- H


2002


2009


Time




Figure 2-3. Mean and standard error of proportion of stumps with sprouts through time
in seven pondcypress domes in north central Florida.


SSawhead

- Shearhead


--


2002


Time


2009


Figure 2-4. Mean and standard error of cone production through time in seven
pondcypress domes in north central Florida.


0.5
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0


0.35

0.3

0.25

0.2

0.15

0.1

0.05












0.6 ,


0.5

0.4

0.3

0.2

0.1


-- sawhead
- -*- shearhead


20cm Below
MHW


Stump Height


Figure 2-5. Treatment effect on stump height in seven pondcypress domes in north
central Florida.



0.8

4 .
E 0.7 -





0.2 -shrub
4-I
0.4 ----- -n-sprout

4 0.3 -*--nothing

|$ 0.2 'f''' -k--shrub

o o.1 -. / .

0
<20 21-30 31-40 41-50 >51


Mean P-ihf Water


40cm Above
MHW


Figure 2-6. Effects of stump diameter on likelihood of having a sprout or shrub on a
stump or being bare in seven pondcypress domes in north central Florida.


Stump Diameter (cm)














45
40 -
S35
S30
o 25 -
S20
. 15
4- 10
H0 F-] n n _


E



U U (0
w = E



(U
a= ao


rc -o i u n -r u

^ 3 te > -o -a ja U
73 0
tn ro u '"n C V" E ro
(U C UO Co (U3o; iri

C 0 CII
TO o c.
>. c U a


-o -o
o
U C
C
MT~iU,,3EH-


Species on Stump


Figure 2-7. Percentage of species growing on stumps in seven pondcypress domes in
north central Florida.


Figure 2-8. Probability of a stump being bare or having a shrub growing on a stump in
seven pondcypress domes in north central Florida.


0.6
a
E

C
6 0.4 -


u 0.3
o ------ nothing
_ 0.2 -
S' -A-shrub

M 0.1 -
0
a-
0

20cm Below MHW Mean I-igh Water 40cm Above MHW

Stump Height (cm)










APPENDIX A
SAS CODE FOR GLIMMIX PROCEDURE ANALYZING SPROUTS AND CONES
THROUGH TIME

PROC IMPORT OUT= proproportions
DATAFILE= "e:\thesis\proportions.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;

ods rtf;
ods graphics on;

proc glimmix data=proportions plots=pearsonpanel;
weight stump_count;
title 'with time effect included swamp is random sprouts';
class plot swamp treatments time;
model sprouts = treatmentsltime/ddfm=sat;
random swamp plot(swamp);
Ismeans treatments*time/slice=treatments plot=meanplot(cl join sliceby=time) pdiff;
Ismeans treatmentsltime/pdiff;
run;

proc glimmix data=proportions plots=pearsonpanel;
title 'with time effect included swamp is random cones';
weight stump_count;
class plot swamp treatments time;
model cones = timeltreatments time/ddfm=sat;
random swamp plot(swamp);
Ismeans treatments*time/slice=treatments plot=meanplot(cl join sliceby=time) pdiff;
Ismeans treatmentsltime/pdiff;
run;

title;
ods graphics off;
ods rtf close;










APPENDIX B
SAS CODE FOR GLIMMIX PROCEDURE ANALYZING STUMP CHARACTERISTICS

ods rtf;
ods graphics on;
PROC IMPORT OUT= b4
DATAFILE= "e:\thesis\book4.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
proc glimmix data= b4;
where stump_ht < 300;
title 'sprouts';
class size cat_diameter swamp treatment plot cat_depth cat_ht;
model sprout = treatment cat_diameter cat_depth treatment*cat_depth/dist=bin solution
ddfm=kr;
random swamp*plot;
Ismeans treatment cat_diameter cat_depth/ilink pdiff plots=mean(cl join ilink);
Ismeans treatment*cat_depth/ilink pdiff plots=mean(cl join sliceby=treatment ilink);
output out=p pred=p;
run;
proc glimmix data= b4;
title 'bare stumps';
where stump_ht < 300;
class size cat_diameter swamp treatment plot cat_depth cat_ht;
model bare = treatmentlcat_diameterlcat_depth@1 treatment*cat_depth/dist=bin
solution ddfm=kr;
random swamp*plot;
Ismeans treatment cat_diameter cat_depth/ilink pdiff plots=mean(cl join ilink);
Ismeans treatment*cat_depth/ilink pdiff plots=mean(cl join sliceby=treatment ilink);
output out=pb pred=p;
run;
proc glimmix data= b4 ;
title 'other stumps';
where stump_ht < 300;
class size cat_diameter swamp treatment plot cat_depth cat_ht;
model other = treatmentlcat_diameterlcat_depth@1 treatment*cat_depth/dist=bin
solution ddfm=kr;
random swamp*plot;
Ismeans treatment cat_diameter cat_depth/ilink pdiff plots=mean(cl join ilink);
Ismeans treatment*cat_depth/ilink pdiff plots=mean(cl join sliceby=treatment ilink);
output out=po pred=p;
run;
ods graphics off;
ods rtf close;










APPENDIX C
PERCENTAGE OF SPECIES FOUND ON STUMPS


Species
Eupatorium capillifolium
Woodwardia virginica
Smilax sp
Cladium jamaicense
Persea borbonia
Ilex glabra
Acer rubrum
Toxicodendron radicans
Nyssa silvatica
Crataegus aestivalis
Pinus elliotti
Rubus sp
Cephalanthus occidentalis
Lyonia ligustrina
Myrica cerifera
Lyonia lucida


Count
1
1
2
3
3
4
6
15
16
20
21
30
36
39
138
214


% of Species on Stump
0.2
0.2
0.4
0.5
0.5
0.7
1.1
2.7
2.9
3.6
3.8
5.5
6.6
7.1
25.1
39.0









LIST OF REFERENCES


Anderson, J.T., Landi, A.A., and P.L. Marks. 2009. Limited flooding tolerance of
juveniles restricts the distribution of adults in an understory shrub (Itea virginica;
Iteaceae) Am. J. Bot. 96(9):1-9.

Brandt, K., and K.C. Ewel. 1989. Ecology and management of cypress swamps: A
review. Bull. 252. Florida Cooperative Extension Service, Institute of Food and
Agriculture Science, University of Florida, Gainesville. 19 p.

Casey, W. and K. Ewel. 2006. Patterns of succession in forested depressional wetlands
in North Florida, USA. Wetlands J. 26(8): 147-160.

Conner, W.H., 1988. Natural and artificial regeneration of baldcypress (Taxodium
distichum (L.) Rich.) in a Louisiana Swamp. For. Ecol. Manag. 14(8): 305-317.

Conner, W.H., Toliver, J.R., and F.R. Sklar. 1986. Natural Regeneration of Baldcypress
(Taxodium distichum (L.) Rich.) in a Louisiana Swamp. For. Ecol. Manag. 14:
305-317.

Conner, W.H., J.G. Gosselink and R.T. Parrondo. 1981. Comparison of the vegetation
of three Louisiana swamp sites with different flooding regimes, Am. J. Bot. 68,
pp. 320-331.

Duryea, M.L. and L.A. Hermansen. 2006. Cypress: Florida's majestic and beneficial
tree. Florida Cooperative Extension Service, Institute of Food and Agriculture
Science, University of Florida, Gainesville. Circular 1186.

Ewel, K.C. and H.T. Odum (eds.). 1984. Cypress swamps. University Presses of
Florida, Gainesville FL. 34-56.

Ewel, K.C., H.T. Davis, and J.S. Smith. 1989. Recovery of Florida cypress swamps from
clearcutting. South. J. Appl. For. 13(3):123-126.

Ewel, K.C. 1996. Sprouting by pondcypress (Taxodium distichum var. nutans) after
logging. South. J. Appl. For. 20():209-213.

Ewel, K.C. 1998. Pondcypress swamps. pp. 405-420 in M. G. Messina and W. H.
Conner (eds.). Southern forested wetlands: ecology and management. CRC
Press, Boca Raton FL.

Gardiner, E.S., Russell Jr., D.R., Hodges, J.D., T.C. Fristoe., 2000. Impacts of
mechanical tree felling on development of water tupelo regeneration in the
Mobile Delta, Alabama. South. J. Appl. For. 24, 65-69.









Golden, M.S. 1999. Factors affecting sprouting success in a bottomland mixed
hardwood forest. pp 157-163 In: Haywood JD (ed) Proceedings of the Tenth
Biennial Southern Silvicultural Research Conference. USDA For. Gen. Tech.
Rep. SRS-30, 618 p.

Gunderson, L.H. 1984. Regeneration of cypress in logged and burned strands at
Corkscrew Swamp Sanctuary, Florida. pp. 349-357. In: K.C. Ewel and H.T.
Odum (Eds) Cypress Swamps. The University of Florida Press, Gainesville,
Florida. 472 p.

Huenneke, L.F. and R.R. Sharitz. 1986. Microsite abundance and distribution of woody
seedlings in a South Carolina cypress-tupelo swamp. Am. Midi. Nat. 115: 328-
335.

Hook, D.D., and D.S. DeBell. 1970. Factors influencing stump sprouting of swamp and
water tupelo seedlings. USDA For. Ser. Res. Pap. SE-57. Southeastern Forest
Experiment Station, Asheville, NC. 9 p.

Johnson, R.L. 1978. Planning for natural regeneration of hardwoods in the Coastal
Plain. In Proc. 2nd Symposium on Southeast hardwoods, April, 1977. Dotham,
AL. Atlanta, GA: Southeast Area State and Private Forestry, Forest Service, U.S.
Department of Agriculture. 114-119.

Kauppi, A., Rinne, P., and A. Ferm. 1988. Sprouting ability and significance for
coppicing of dormant buds on Betula pubescens Ehrh. Stumps. Scand. J. For.
Res. 3:343-354.

Keim, R.F., Chambers, J.L., Hughes, M.S., Dimov, L.D., Conner, W.H., Shaffer, G.P.,
Gardiners, E.S., and J.W. Day. 2006. Long-term success of stump sprouts in
high-graded baldcypress-water tupelo swamps in the Mississippi Delta. For. Ecol.
Manag. 234:24-33.

Kennedy Jr., H.E., 1982. Growth and survival of water tupelo coppice regeneration after
six growing seasons. South J. Appl. For. 6, 133-135.

Kramer, Paul J., T.T. Kozlowski. 1979. Physiology of woody plants. New York:
Academic Press. 811 p.

Lockaby, B.G., Stanturf, J.A., and M.G. Messina. 1997. Effects of silvicultural activity on
ecological processes in floodplain forests of the southern United States: a review
of existing reports. For. Ecol. Manag. 90(2,3): 93-100.

Mattoon, W. R. 1915. The Southern Cypress. U.S. Department of Agriculture Bulletin
272. Washington DC. 74p.









McCreary, D.D., Tietje, W.D., Schmidt, RH, Gross, R., Weitkamp, WH, Willoughby, BL.,
and F.L. Bell. 2002. Stump sprouting of blue oaks in California. In: Standiford
R.B. (Tech. coord.) Proceedings of the Symposium on Oak Woodlands and
Hardwood Rangeland Management. USDA Forest Service, Pacific SW Research
Station, Berkley, California. Gen. Tech. Rep. PSW-126, pp. 64-69.

Mostacedo, M., Putz, F.E., Fredericksen, T.S., Villca, A., and T. Palacios. 2009.
Contributions of root and stump sprouts to natural regeneration of a logged
tropical dry forest in Bolivia. For. Ecol. Manag. 258(6): 978-985.

Porter, G., Martin, B., Wright, L., Gallagher, T., Stuart, W., and B. Izlar. 1984.
Occurance of shear damage on southern pine sawlogs. American Pulpwood
Association Technology Release 84-P-5. 6 p.

Randall C.K. 2001. Factors influencing stump sprouting in pondcypress (Taxodium
distichum var. nutans (Ait.) Sweet). M.Sc. thesis, Univ. of Florida, Gainesville,
79p.

Randall, C.K., Duryea, M.L., Vince, S.W., and R.J. English. 2005. Factors influencing
stump sprouting by pondcypress (Taxodium distichum var. nutans (Ait.) Sweet).
New Forests. 29:245-260.

SAS Institute. 2008. SAS software, version 9.2. SAS Institute Inc. Cary, NC, USA.

Schlesinger, W.H. 1978. On the relative dominance of shrubs in Okefenokee Swamp.
The Am. Nat. 112(987): 949-954.

Schneider, R. L. and R. R. Sharitz. 1986. Seed bank dynamics in a southeastern
riverine swamp. Am. J. Bot. 73:1022-30.

Sharitz, R.R., R.L Schneider, K.W. Dyer, and N.C. Martin. 1986. Wetland Forest
Regeneration and Hydrologic Regime. Annual Rep. of Ecol. Rest. DOE Report
SR-819-17. P. 47-53.

Shigo, Alex L. 1986. A new tree biology. Shigo and Trees, Associates, Durham, New
Hampshire. 595 p.

Southeast Regional Climate Center (SERCC). 2007. Historical Florida Weather.
Available online at http://www.sercc.com/climateinfo/historical/historical_fl.html;
last accessed Dec. 12, 2009.

Terwilliger, V.J. and K.C. Ewel. 1986. Regeneration and growth after logging in Florida
cypress domes. For. Sci. 32:493-506.









Visser, J. M., and C. E. Sasser. 1995. Changes in tree species composition, structure,
and growth in a bald cypress-water tupelo swamp forest: 1980-1990. For. Ecol.
Manag. 72:119-129.

Wharton, C.H., H.T. Odum, K. Ewel, M. Duever, A. Lugo, R. Boyt, J. Bartholomew, E.
DeBellevue, S. Brown, M. Brown, and L. Duever. 1977. Forested Wetlands of
Florida: Their management and use. Div. State Plan., Tallahassee, FL.

Wilhite, L., J. Toliver. 1990. Taxodium distichum (L.) Rich. Baldcypress. In: Burns, R.;
Honkala, B.; tech. cords. Silvics of North America, Vol. 1. Conifers. Agricultural
Handbook 654. Washington, DC: U.S. Department of Agriculture, Forest Service:
563-572.









BIOGRAPHICAL SKETCH

Nicole Ricci, a native to Macomb, Michigan, attended Macomb Community

College as an Art and Design major, before transferring to Northern Michigan University

to follow a bachelor's degree in Earth Science. During her studies at NMU, she earned

the Farrell Cartography Scholarship for achieving the highest GPA in all mapping

science courses.

After completing her Bachelor of Science, she attended Michigan Technological

University in pursuit of a graduate degree in forestry and wetland science. She

graduated with a course-work Master of Science in December of 2007.

In June of 2008, she entered the School of Forest Resources and Conservation at

the University of Florida to pursue a thesis track Master of Science. Throughout her

time at UF, she enjoyed learning many southern traditions, and gained lifelong

friendships.





PAGE 1

1 RECOVERY OF CYPRESS DOMES IN NORTH CENTRAL FLORIDA 10 YEARS AFTER HARVEST By NICOLE RICCI 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 2010

PAGE 2

2 2010 Nicole Ricci

PAGE 3

3 This thesis is dedicated to my parents, Daniel and Catherine Ricci, for their never ending support throughout my academic career.

PAGE 4

4 ACKNOWLEDGMENTS This research and my graduate assistantship were funded by a grant provided by the Environmental Protection Agencys Wetlands Region Four Program awarded to Dr. Kimberly Bohn. Equipment and one field site was provided by the University of Florida, as well as scientific advi ce through faculty and staff in the School of Forest Resources and Conservation. The Florida Division of Forestry provided one field site for revisit on this study. The USDA Forest Service National Tree Seed Laboratory in Macon, GA provided access to their equipment. And a special thank you to Victor Vankus for his technical advice and housing during the end of my research. The Campbell Group LLC., Rayonier Inc., and Foley Timber Company also provided access to field sites. The completion of this resear ch was successful due to the many people that contributed their knowledge and support. First and foremost, I would like to express my gratitude to Dr. Kimberly Bohn for her guidance, supervision, and advice from the early stages of this research. One simp ly could not wish for a better or friendlier adviser. I am thankful to the additional members of my supervisory committee, Drs. Katherine Ewel and Matt Cohen for their methodological advice and expertise in their respective areas of study. Furthermore, I would also like to convey thanks to Chad Foster, for his foster javelin method and wetland flora knowledge, Rose Godfrey for her sterling attention to detail, and Brandon Raczkoski for his heroic assistance in the field. Their help no matter the weather or water level was invaluable. Particular acknowledgement also goes to Wendy Adams for all her indispensable help with travel forms and administration. I would also like to acknowledge James Colee of IFAS Statistics, who was incomparable as a statistic al consultant.

PAGE 5

5 Finally, I would like to thank my friends and family for their emotional and technical support during this research.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 ABSTRACT ..................................................................................................................... 9 CHAPTER 1 INTRODUCTION .................................................................................................... 11 2 METHODS .............................................................................................................. 16 Study Area and Experim ental Design ..................................................................... 16 Stump Sprout Remeasurements ............................................................................. 17 Data Analysis .......................................................................................................... 18 3 RESULTS ............................................................................................................... 19 Characteristics of 10yr old Pondcypress Sprout Regeneration .............................. 19 Treatment Effect on Sprouts and Cones through Time ........................................... 19 Influence of Stump Characteristics on Likelihood o f Sprouting ............................... 20 4 DISCUSSION ......................................................................................................... 22 Sprouting Response after 10 years ........................................................................ 22 Potential for Sprouts to Provide Seed Source ......................................................... 23 Predictors of Vegetative Development on Cypress Stumps .................................... 24 Management Recommendations ............................................................................ 26 APPENDIX A SAS CODE FOR GLIMMIX PROCEDURE ANALYZING SPROUTS AND CONES THROUGH TIME ...................................................................................... 35 B SAS CODE FOR GLIMMIX PROCE DURE ANALYZING STUMP CHARACTERISTICS .............................................................................................. 36 C PERCENTAGE OF SPECIES FOUND ON STUMPS ............................................. 37 LIST OF REFERENCES ............................................................................................... 38 BIOGRAPHICAL SKETCH ............................................................................................ 42

PAGE 7

7 LIST OF TABLES Table page 2 1 Mean number of pondcypress sprouts per stump during 2002 and 2009 in stumps of seven pondcypress sites in north central Florida (Figure 21). ........... 28 2 2 Summary statistics by site for proportion of pondcypress stumps with live sprouts, proportion of sprouts with cones, and mean dbh and height of sprouts in 2009. .................................................................................................. 28 2 3 The effects of harvesting treatments, time, and their interaction, on the proportion of pondcypress sprouts per stump in sev en pondcypress sites in north central Florida. ........................................................................................... 28 2 4 The effects of harvesting treatments, time, and their interaction, on the proportion of pondcypress sprouts with cones per stump in seven pondcypress sites in north central Florida. ......................................................... 29 2 5 The effects of harvesting treatments, stump characteristics, and the interaction of treatments*stump height classes, on the probability of pondcypress sprouts occurring on a stump in seven pondcypress sites in north central Florida. ........................................................................................... 29 2 6 The effects of harvesting treatments, stump characteristics, and the interaction of treatments*stump height classes, on the probability of a stump being bare in seven pondcypress sites in north central Florida. ......................... 29 2 7 The effects of harvesting treatments, stump characteristics, and the interaction of treatments*stump height classes, on the probability of other vegetation occurring on a stump in seven pondcypress sites in north central Florida. ............................................................................................................... 29

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Locations of pondcypress wetland study sites in north central Florida by county. ................................................................................................................ 30 2 2 Study design two 0.05 ha circular plots to each harvesting method (sawhead or shearhead) and 3 stump heights per whole plot. ............................................ 31 2 3 Mean and standard error of proportion of stumps with sprouts through time in seven pondcypress domes in north central Florida. ........................................... 32 2 4 Mean and standard error of cone production through time in seven pondcypress domes in north central Florida. ...................................................... 32 2 5 Treatment effect on stump height in seven pondcypress domes in north central Florida. .................................................................................................... 33 2 6 Effects of stump diameter on likelihood of having a sprout or shrub on a stump or being bare in seven pondcypress domes in north central Florida. ....... 33 2 7 Percentage of species growing on stumps in seven pondcypress domes in north central Florida. ........................................................................................... 34 2 8 Probability of a stump being bare or having a shrub growing on a stump in seven pondcypress domes in north central Florida. ........................................... 34

PAGE 9

9 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 RECOVERY OF CYPRESS DOMES IN NORTH CENTRAL FLORIDA 10 YEARS AFTER HARVEST By Nicole Ricci August 2010 Chair: Kimberly Bohn Major: Forest Resources and Conservation Pondcypress ( Taxodium distichum var. nutans (Ait.) Sweet) can naturally regenerate via stump sprouting which is considered an important component of regeneration following harvesting because sprouts produce cones within two years that provide a seed source for additional seedling recruitment. However, there are few studies on the long term success of po ndcypress s tump sprouts and their potential to act as a seed source. In this study I examined the effect of felling equipment (sawhead vs. shearhead) on the long term success of stump sprouts and the likelihood of sprout vs. shrub development on cut stumps in relat ion to felling treatment, stump height and stump diameter in seven harvested cypress domes located in north central Florida, USA. Each of the study sites was harvested in spring of 1999 and reexamined in summer of 2009. I measured stump sprout character istics such as numbers of sprouts per stump, sprout height and diameter, and cone presence on stumps tagged during the original study. I evaluated proportions of stumps with sprouts and cones using a twofactor Anova with felling equipment and time as the main effects. I also determined whether stump characteristics and felling treatment affected the probability of a stump containing a sprout, shrub, or being bare using a general linear mixed model.

PAGE 10

10 Long term sprouting success varied by site, ranging fro m15% 42% of stumps having at least one sprout. The type of felling equipment used did not significantly affect the proportion of stumps producing sprouts or sprouts producing cones, and the proportion of stumps with at least one sprout changed very little over time. T his study confirmed initial findings that sprouts were more likely to develop on smaller diameter stumps (< 30 cm) and stumps cut 20 cm below or near t he mean high water. The occurrence of shrubs or another tree species on a stump increased as both stump diameter and stump height increased. Survival of sprouts over ten years shows that initial sprout success after two growing seasons can be used as an indicator of coppice regeneration. Recommendations for harvesting to maximize sprouting include cutting in close proximity of the historic mean high water mark and on stumps smaller than 30 cm.

PAGE 11

11 CHAPTER 1 INTRODUCTION Adequate regeneration of secondgrowth cypress is important for both ecological and economical reasons. Cypress timber has long been valued for its use as cross ties, ladders, pilings, stakes, post wood and narrow boards ( Wilhite and Toliver 1990) and more recently as mulch for gardening. However, cypress swamps have multiple values beyond timber production, including wildlife habitat, ground water recharge, flood control, and water quality improvement (Ewel 1998). Recently, concerns over ecosystem recovery following harvesting have prompted a need for better understanding of regeneration dynamics in a variety of forested wetland systems in order to ensure a quick recovery of these systems following harvesting disturbance. While baldcypress ( Taxodium distichum (L.) Rich.) was h ighly valued for its high proportion of decay resistant heartwood (Wilhite and Toliver 1990) most of the large older baldcypress trees that contained valuable heartwood has been harvested and much of the remaining cypress is pondcypress ( Taxodium distichum var. nutans (Ait.) Sweet ). With the sustained demand for cypress timber, and the improvements in harvesting equipment, loggers have been able to access a variety of forested wetland systems including small isolated cypress domes where pondcypress is frequently found. These domes are typically embedded within pine flatwood communities and occasionally intermixed within hardwood hammocks. The pondcypress is found in association with swamp tupelo ( Nyssa sylvatica var. biflora ), slash pine ( Pinus elliot tii Englem.), and various bays ( Gordonia lasianthus (L.) Elis, Persea palustris (Raf.) Sarg., Magnolia virginiana L.) which regenerate on hummocks or along shallow edges (Wharton et al. 1977, Brandt and Ewel 1989).

PAGE 12

12 Additionally, it is now common for trees of all merchantable sizes to be harvested because trends in the commercial usage of cypress have converted from lumber to pulp and mulch for gardening (Ewel 1998). Logging operations typically involve wholetree harvesting of cypress (Duryea and Hermansen 1997). Though some studies have documented that that pondcypress domes can regenerate back to their original basal area and species composition by 50 years after harvesting (Terwilliger and Ewel 1986), the majority of the literature has focused on baldcypress regeneration. Early survival of pondcypress regeneration has been addressed in previous studies (Terwilliger and Ewel 1986, Ewel et al. 1989, Ewel 1996), but the long term success has not been evaluated. One unique characteristic of cypress is that it is one of the few conifer species that can naturally regenerate via seed germination or vegetatively by stump sprouting after harvest ( Wilhite and Toliver 1990). Seed dissemination occurs during the fall to early winter, and water and wildlife movement are the main dispersal mechanisms within cypress domes. Seedlings need saturated but not submerged soil to germinate and prefer full sun for vigorous growth. Competition from shrubs or a dense understory can inhibit cypress regeneration by decreasing s ufficient light and nutrients (Ewel 1989). Limited seedling recruitment across a site can also be attributed to factors like poor seed viability from parent stock lack of seed sources after harvest inability of seed to be dispersed and hydrologic limitations (Gunderson 1984, Schneider and Sharitz 1986, Conner et al. 1986) Stump sprouts are considered an important mechanism for immediately regenerating a cypress stand after harvesting (Ewel et al. 1989, Randall et al. 2005). In

PAGE 13

13 species that stump sprout, initial growth of sprouts is likely to be greater than seedling regeneration due to the root systems of established trees and the amount of stored carbohydrates in the root system ( McCreary et al. 2002). Stump s prouts have vascular connections between the sprouts themselves and the parent tree root system, with each stem helping to sustain a portion of the root system ( Kramer and Kozlowski 1979). Conner et al. (1986) found that baldcypress seedling regeneration grew on average to 91 cm and stump sprouts grew to 152 cm in height after four growing seasons. Additionally, since stump sprouts can produce cones as early as two years after sprouting, this could provide an additi onal seed source for seedling recruitment when environmental conditions are conducive for germination, so long as the seed source is viable (Ewel 1998, Randall et al. 2005). T here are still several concerns regarding the long term success of sprouting as a mechanism for regenerating pond cypress in wetland systems. It is not clear whether sprouts can survive to be recruited into the main canopy, and low sprout survival between 17 and 23% has been observed in some cases ( Conner et al. 1986, Conner 1988, Ewel 1996). In comparison, survival of bareroot pondcypress seedlings planted in eight harvested swamps in north Florida ranged from 37% to 89% after four years (Vince, Duryea and Randall unpub. data). Additionally, poor stem form and mechanical damage fro m ground based logging can limit vegetative reproduction (Hook and DeBell 1970, Lockaby et al. 1997). Stump decay could compromise sprouts although decay may not necessarily move into the new sprout, particularly where the sprouts develop off the side of the stump (Shigo, 1986). Moreover, sprout regeneration can be spatially

PAGE 14

14 heterogeneous and sometimes insufficient in providing a continuous seed source across the entire stand (Kennedy 1982, Keim et al. 2006). While longterm survival of sprouts is affec ted by a multitude of factors, initial establishment of sprouts are often a function of stump dimensions. Stump height and diameter can influence the vigor of sprouting success regardless of species (Hook and DeBell 1970, Kennedy 1982, Ewel 1996, Gardiner et al. 2000). For floodprone species stump height in relation to mean water depth has shown a strong correlation to sprout establishment (Terwilliger and Ewel 1986) though relationships may vary by species. Kennedy (1982) found that water tupel o cut 76 cm above the mean high water mark had the highest survival of sprouts ; while Ewel (1996) indicated stump less than 70 cm tall from the ground yielded greater pondcypress sprout survival. Parent tree size can also influence stump sprouting. Early sprout survivorship is greatest on smaller stumps (Ewel 1996), while larger stumps over 40 cm in diameter did not show cypress sprout survival after harvest (Keim et al. 2006). Stumps may also provide a safe site for germination and establishment of other plants such as shrubs or other tree seedlings which could potentially interfere with sprout development. M ost shrub species cannot tolerate being submerged in the early stages of development and would naturally be recruited on stump surfaces above water W etland shrub species such as Itea sp. prefer stumps because they are elevated above floodwaters (Anderson et al. 2009) and dead or decaying stumps have also been observed to serve as a foundation for shrub or bay establishment ( Schlesinger 1978, Casey and Ewel 2006) Furthermore, shrub invasion was noted on stumps in Louisiana cypresstupelo swamps decades after harvest (Conner et al. 1981). Whether a shrub is

PAGE 15

15 recruited is on a stump may also be a function of the parent tree size. However, there has b een no definitive quantification of relating stump size and shrub occurrence. Long term monitoring is needed to better understand the recovery of cypress domes after logging, yet there are few studies that have monitored sprouts or regeneration success longer than six growing seasons (McCreary 2002). This work builds upon initial twoyear results reported by Randall et al. (2005). That study examined parent tree size, height of stump cut, and felling equipment on stump sprouting (Randall 2001). Stump heights were cut below, at, or above mean high water, with either a hydraulic shear head or continuous disk saw head on the interchangeable head of the feller buncher. The saw head cuts with a disk blade and the shear head, which can be compared to pruning shears, snips the tree off at the appropriate stump height. Using the saw head attachment creates a smoother cut across the stump, while stumps cut with a shear head may increase damage by splitting or shattering stumps (Porter et al. 1984). In general, t wo seasons after harvest results indicated that greater sprouting occurred on smaller diameter stumps close to the mean high water mark and sprout establishment was not significantly affected by felling equipment (Randall et al. 2005) These study sites were reinvestigated to assess the longterm survival of cypress stump sprouts ten years after harves t in relation to felling method (sawhead type) and to quantify the proportion of stumps with conebearing sprouts. An additional objective of this study was to determine the likelihood of sprout vs. shrub development as a function of felling equipment, stump diameter, and stump height.

PAGE 16

16 CHAPTER 2 METHODS Study Area and Experimental Design This investigation was conducted in sites that were selected in 1999 to evaluate factors that influenced stump sprouting in pondcypress trees in eight isolated cypress domes In 2009, the eight logged pondcypress domes were revisited in northcentral Florida (Figure 1) (latitudes: 29 05 to 29 55; lon gitudes: 82 10 to 82 35). However, a fter field investigations were completed, one site in Lafayette County was dropped due to difficulty in stump identification and relocation. Prior to harvest pondcypress and swamp tupelo were the predominant species, along with slash pine, myrtle holly and various bays in the canopy. Historical climate data ranging from 1950 2010 in each county that contained a field site indicated that mean annual precipitation varied from 124 cm to 150 cm (Southeast Regional Cli mate Center 2010). The original study used a split plot design (Figure 2). Two 0.05 ha circular plots were established on each half of the pondcypress dome and each half of the dome was cut using a sawhead or shearhead harvesting technique. This was the wholeplot treatment. Within each 0.05 ha plot, 45 trees were chosen to be cut at one of three stump cutting heights as the subplot treatment. Heights were based on mean high water level within that swamp. This was determined from stain and lichen lines on the cypress trees (Randall et al. 2005). Stump heights were categorized as being cut low (20 cm below mean high water), medium (at mean high water), and high (40 cm above mean high water). Within each plot, permanent aluminum tags were used to mark eac h study stump, and locations within each plot were mapped.

PAGE 17

17 The pondcypress domes were clearcut in February and March of 1999 in accordance with Floridas Best Management Practices in forested wetlands at that time, which left 12 reserve trees per hectare. T rees outside the research plots were harvested under operational standards. Stump height and diameter were recorded after logging, and sprouting success was initially monitored in the summer of 1999 through fall of 2002 (Randall et al. 2005) Stump Sprout Remeasurements The relocation of stumps from the previous study was conducted in May July of 2009 using hand drawn plot maps and leave trees for orientation. We recorded whether stumps were bare, had cypress sprouts, or had other vegetation growing on t he stump. If vegetation other than cypress was growing on the stump, the species was noted. For stumps with sprouts, we recorded the number of live sprouts on each stump, the height of all sprouts, and the dbh of stems > 5 cm (2 in) (stem diameter was measured at 1.4 meters from ground). We also noted whether or not each sprout was producing cones Eighty seven percent of the initial stumps were relocated across all sites using plot maps from the original study. Out of the 168 missing stumps from the ori ginal study, 75 (45%) were from sawhead treatment and 93 (55%) were from shearhead treatment out of the total stumps in each category. A t test confirmed that differences in numbers of missing stumps across sites were not significantly different by treatment so were not considered to be confounding factor in the subsequent analyses. Not surprisingly, water level affected our ability to relocate some stumps. Stump cut height of missing stumps was predominantly around mean high water level (65%), with the remaining amount being cut at the 20 cm below mean high water (30%) and 5% cut 40 cm above mean high water.

PAGE 18

18 Data Analysis Proc Glimmix (SAS 9.2) w as u sed for all data analysis because this particular model properly assesses random (swamp and plot) and fixed effects (treatment, time, stump characteristics), and has the ability to estimate the appropriate standard errors for binary response data. Individual swamps were used as a blocking factor All statistical tests were run at = 0.05. To assess the num ber of sprouts per stump, the proportion of stumps with sprouts and proportion of sprouts with cones, we used a generalized mixed model with treatment and time as fixedeffect factors (Appendix A). The two time periods were from the field observations in 2002 and 2009. Proportions of stumps with sprouts in 2009 were calculated only based on the total number of relocated stumps. Adjusted means were plotted to visualize the change in sprouts or cones in time. A mixed model with a binomial distribution was analyzed to determine the probability of a stump having sprouts, other vegetation or nothing on it as a function of stump height, diameter, treatment, and interactions of those factors across all sites, (Appendix B). Stump diameters were broken into 5 categories that contained approximately the same number of stumps. Diameter categories were 29 cm, 30 39 cm, 4049 cm, and > 50 cm. Sprouting was not evident on stumps over 50 cm, and used as a cut off point for categorical analysis. Additional ly, stump height was categorized in the three cut height classes of the original study: 20 cm below mean high water, at mean high water, and 40 cm above mean high water.

PAGE 19

19 CHAPTER 3 RESULTS Characteristics of 10yr old Pondcypress Sprout Regeneration The average number of sprouts per stump decreased over time (Table 21). In 2002, the number of sprouts per stump averaged 15 sprouts on sawhead stumps and 12 on shearhead stumps. High values were due to an anomaly at one site (F1), which had an unusuall y large number of sprouts per stump, particularly for the sawhead treatment (Table 21). Removing this site would result in similar numbers of sprouts per stump (about 11) for both treatments in 2002. In 2009, the mean number of sprouts per stump decrease d to 7 for both harvest treatments. Changes in the number of sprouts per stump with the removal of site F1 were not statistically significant between felling treatments. Individual swamps showed great variability in sprout dimensions and proportion of stumps with sprouts. The mean diameter of the largest sprout per stump ranged from 7.1 to 18.1 cm across all of the sites, and mean height ranged from 5.3 to 7.1 m (Table 2 2 ). There was no correlation between the three stump heights used in the original study and sprout characteristics such as the number of sprouts on a stump, sprout dbh or height. The percentage of stumps with live sprouts varied between 1040% in 2002 and between 1141% in 2009. Treatment Effect on Sprouts and Cones through Time Time had a statistically significant effect on stump sprout success but felling treatment did not (table 23). However, the change in proportion of stumps with sprouts was likel y not ecologically relevant, as the percentage of stumps with live sprouts in 2002 was 28% and rose only slightly to 31% in 2009, primarily because of additional

PAGE 20

20 stumps with sprouts found on the sawhead cut stumps (Figure 2 3) This indicate d that stump sprouts primarily develop only within the first two years after harvesting. The percentage of sprouts with cones increased significantly (p=0.0002) from the last period of monitoring in 2002 to 2009, but was not significantly influenced by treatment (Tabl e 2 4). The proportion of sprouts with cones was between 10 and 13% for shearhead and sawhead stumps in 2002 and 17 and 26% for those treatments in 2009 (Figure 23). Influence of Stump Characteristics on Likelihood of Sprouting We calculated the probabi lity, and tested the significance, that a stump of a given size (diameter and height class) and felling treatment would have sprouts, other vegetation or nothing growing on it 10 years after harvesting. Diameter was broken into 5 classes based on the rang e of diameters of sprouts that the majority of stump sprouts occurred. Height classes were based on the initial study categories related to mean water depth. Initially the interaction between stump diameter x height was tested, but it was dropped when all insignificant and were removed from the final analysis. Evaluation of stump characteristics (diameter and height classes) and harvesting technique were conducted using only those stumps relocated in 2009 since that is the only time period for which data o n other vegetation on stumps was recorded. A significant interaction effect between treatment x stump height was only found for stumps with sprouts (Figure 25, Table 25) (p=0.048). Multiple comparisons indicated that stumps cut with the shear head at me an high water and 40 cm above mean high water were significantly different from each other and significantly different from the sawhead stumps cut at the same heights. Harvesting technique did not significantly

PAGE 21

21 affect the likelihood of a stump being bare or having a shrub growing on it. (Table 26, Table 27) Stump diameter significantly affected the probability of a stump having a cypress sprout growing on it (p<0.0001). There was a decrease in the likelihood of a sprout occurring on a stump as the stum p diameter increased (Figure 26). Sprouting was more likely to occur on stumps within a diameter range of 020 cm. Stump diameter also significantly affected the likelihood of a stump being bare (p=0.0004) or having other vegetation (p<0.0001). The likeli hood of a stump being bare nearly doubled as diameter increased from less than 20 cm wide to 21 cm and greater (Figure 26). There was no statistical difference in the likelihood of a stump being bare for stumps between 21 and 50 cm. The likelihood of a stump having other vegetation on it increased significantly with diameter (Figure 26). The highest probability of a shrub occurring on a stump was at stump diameters of 30 cm and greater. Lyonia lucida and Myrica cerifera were the dominant shrub species found on stumps (Figure 27). Stump height had no significant effect on the likelihood of stumps being bare (Table 26) but did significantly affect the likelihood of having a shrub occur on a stump (p<0.0001) (Figure 28). Stumps cut at 40 cm above mean high water were the most favorable for shrub recruitment.

PAGE 22

22 CHAPTER 4 DISCUSSION Sprouting Response after 10 years Cypress stump sprouts are an important source of regeneration following logging in forested wetlands. Stump sprouts can have an early growth advantage over regeneration by seeds; however, there has been concern about the longevity of coppice sprouts in other species,and their ability to be recruited into the overstory (Mostacedo et al. 2009). Additionally, it is not clear how survivorship and g rowth of pondcypress sprouts compares to that of baldcypress sprouts. Visser and Sasser (1995) found that the recruitment of baldcypress second growth trees did not keep up with mortality; however, findings from our study suggest that 10 years after harvesting, survival of coppice sprouts was high. All stumps that sprouted immediately following harvesting still had at least one if not several dominant sprouts 10 years later. Pondcypress sprout sizes measured in this study were consistent with those repor ted for baldcypress of a similar age. On average, dominant pondcypress sprouts in our study sites were 10 cm in diameter and over 4.7 m tall. Keim et al. (2006) found that baldcypress coppice sprouts from 1050 years after harvesting averaged 10 cm in di ameter and 6.8 m tall. Additionally, the number of sprouts per stump decreased over time though this was expected because dominant sprouts will eventually outcompete the others. Kennedy (1982) found a decrease in water tupelo stump sprouts over 6 growing season that led to an average of 5 sprouts per stump. It has been suggested that sprouts originating from the top or cut surface of the stumps will not develop into quality timber producing trees (Conner et al. 1986). Tree form was not analyzed in our st udy, but casual observation suggested that the majority

PAGE 23

23 of surviving sprouts did, in fact, regenerate as a top sprout. The cypress sprouts in our field sites appeared to be growing in good form at this time, but it is too soon to determine their potential as a timber source. H arvesting technique did not have a significant impact on sprout regeneration and survival The average number of sprouts per stump was virtually the same for both treatments. It had been originally hypothesized that long term sur vival would be notably higher on sawhead cut stumps due to shear related damage by the shearhead feller. The stumps of trees can be damaged by a shearhead in the form of ring shake, stump pull, shatter, and splitting resulting in an increased risk of disease and decay because of the amount of damage caused by the blades (P orter et al. 1984). The severity of splitting damage can damage the stump, providing an entry way for pathogens that eventually translocate to the coppice sprouts. However, our results indicate that at least 10year survival of coppice sprouts was not affected by felling technique. Potential for Sprouts to Provide Seed Source Cypress sprouts are also considered important for providing and additional seed source for seedling recruitment after harvesting. At our study sites, approximately one quarter of the sprouts were cone bearings, and a preliminary seed viability test from cone samples taken in early October of 2009 indicated that on average 41% of the seeds were viable (varying from 20% to 58% across sites). Both pondcypress and baldcypress are considered to have similar reproductive characteristics, and several factors may contribute to variability and magnitude of seed viability of these trees in any given year. Mattoon (1915) stated that baldcypress produces a larger crop of seeds approximately every 3 years. Other f actors such as drought or overall health of the tree may also affect seed viability in any given year. There is very little literature on viability of cypress

PAGE 24

24 seeds in reference conditions, although at least one other study has reported bald cypress seeds had consistently low viability over a threeyear period (Sharitz et al. 1986). Although few naturally recruited seedlings were observed in the study sites, additional studies would be needed to determine if this was due to inadequate seed source, inadequate safe sites for germination, or some other factors. Predictors of Vegetative Development on Cypress Stumps Stump dimensions were better predictors than felling technique alone on the probability that the stump would have a cypress sprout versus some kind of other vegetation growing on it. For instance, stump diameter was a significant predictor of whether a stump would have a sprout or a shrub on it. My results were not unexpected as this trend had already been documented for cypress in several stands ( Ewel 1996, Randall 2005) and for many other hardwood species ( Golden 1999, Sands and Abrams 2009). Pondcypress of saw timber size with a diameter larger than 40 cm produce fewer sprouts (Keim et al. 2006). Diameter is an important factor because older and larger trees have fewer basal dormant buds (Kauppi et al. 1988). Stump height in relation to mean high water level has also been noted to be an important predictor of successful sprouting in swamps and floodprone sites, and in this study stump height w as also a significant predictor of the presence of either sprouts or other vegetation on stumps. Bald cypress stump sprouts were more likely to survive on shorter stumps ( K eim et al. 2006) and in this study pondcypress sprouts were successful on stumps both at or 20 cm below mean high water. Water levels were not monitored under the duration of our study, but drought conditions were present during the initial study when most sprouts regenerated (Randal 2001). The low cut stumps may be submerged in years with typical precipitation, but typical effects of inundation on

PAGE 25

25 stump sprout survival and growth on those cut stumps may not have occurred on our study sites. Stump sprouts at our study sites had an opportunity to establish before normal to high water levels occurred again. Perhaps a higher mortality for low cut stump sprouts due to inundation stress would have been evident across all sites if these dry conditions did not occur (Ewel 1996). Because of those conditions, Randall (2001) suggested from observations outside the study plots, to cut cypress around the mean high water mark. One concern of logging is that cypress swamps may convert to hardwood or shrub dominated swamps due to human and environmental influences. Most of the understory plants and shrubs in cypress and tupelo forests occur on elevated sites, which protect susceptible seeds and seedlings from flooding (Huenneke and Sharitz 1986). In Louisiana cypress tupelo swamps, shrub invasion occurred where the canopy opened over elevated micro sites, including stumps or logs (Conner et al. 1981). Short term studies suggested that harvesting in baldcypress and water tupelo swamps could lead to a conversion to hardwood species (Gellerstedt and Aust 2004). Conner et al. (1986) also found that aft er logging, all sites were dominated by small, shrubby, nontimber species after 1 to 4 growing seasons. On the other hand, other studies, such as those in Okefenokee Swamp indicate that shrubs only rooted on dead stumps and did not invade the entire site (Schlesinger 1978). At our study sites, shrub or other tree species tended to grow on stump heights greater than 40 cm above the mean high water level. L yonia lucida and M yrica cerifera were the most prevalent shrubs found on stumps, and these species ar e typical of pondcypress stumps in the southeast.

PAGE 26

26 Shrubs are faster growing, and may have an advantage over cypress seedlings or might compete with sprouts on stumps. However stumps that do contain sprouts usually do so because of the rapid early growth of stump sprouts (Johnson 1978). In this study, we found relatively few, if any stumps that had both sprouts and shrubs growing on them. It is possible that the shrubs developed on taller stumps that were simply not conducive to sprout development early on; however we do not have data from 2002 to determine whether competition existed during the early phases of vegetation development. Management Recommendations In our study, cypress coppice sprouts survived over ten years past harvest. These results suggest that early coppice regeneration counts during the first few years post harvesting should provide an appropriate assessment of the coppice regeneration potential of that stand. The position of sprouts on stumps, and the condition of sprouts were not recorded in this study. Future condition and establishment position may result in additional sprout mortality in the future as sprouts develop into mature trees. Our rec ommendations regarding the size of trees to cut in order to increase the probability of stump sprouting remain consistent with those of previous studies. Trees of a smaller diameter (< 30 cm) should have a better sprouting response than larger trees. If a site contains only larger cypress trees, sprouting may occur at low levels. Other recommendations suggest that he greatest likelihood of sprout success will occur when stumps are cut at or near the high water mark; although sprout success on stumps cut b elow the mean high water mark were high in this study, we attribute this to the extreme drought conditions immediately following harvest. Still, cutting at mean high water levels appears to be a safe recommendation unless accurate predictions of

PAGE 27

27 climate within the few years following harvesting (when sprouts develop) could be obtained.

PAGE 28

28 Table 21. Mean n umber of pondcypress sprouts per stump during 2002 and 2009 in stumps of seven pondcypress sites in north central Florida (Figure 21). Number of sprouts per stump Year Cutting Method AC F1 F2 G P1 P2 R 2002 sawhead 9 43 9 12 12 10 11 shear 8 23 8 17 12 11 13 2009 sawhead 3 8 7 10 7 7 10 shear 6 7 5 10 8 8 8 Table 22. Summary statistics by site for proportion of pondcypress stumps with live sprouts proportion of sprouts with cones, mean, dbh, and height of sprouts in 2009. Mean measurement by site Sprout Measurement AC F1 F2 G P1 P2 R Proportion stumps w/ sprouts (2002) a 10% 40% 27% 39% 31% 24% 28% Proportion stumps w/ sprouts (2009) a 11% 39% 34% 41% 32% 27% 29% Proportion sprouts w/ cones (2002) c 17% 9% 49% 19% 47% 54% 68% Proportion sprouts w/ cones (2009) c 55% 40% 67% 56% 46% 70% 71% Mean/ SE dbh of largest sprout (cm) b 8.7 (0.61) 8.5 (0.37) 11.1 (0.43) 7.1 (0.25) 8.5 (0.31) 18.1 (1.65) 9.5 (0.34) Mean/ SE height of tallest sprout (m) b 5.8 (0.36) 5.0 (0.20) 7.1 (0.18) 5.3 (0.15) 6.0 (0.17) 6.2 (0.16) 6.1 (0.19) a (n=1222) b (n=1070) c (n=1070) Table 23. The effects of harvesting treatments, time, and their interaction, on the proportion of pondcypress sprouts per stump in seven pondcypress sites in north central Florida. Effect on proportion of stumps w/ sprouts No. DF F Value Pr> F Treatments 1 2.73 0.11 Time 1 14 < 0.001 Treatments*Time 1 1.29 0.27

PAGE 29

29 Table 24 The effects of harvesting treatments, time, and their interaction, on the proportion of pondcypress sprouts with cones per stump in seven pondcypress sites in north central Florida. Effect on Proportion of Sprouts with Cones No. DF F Value Pr> F Treatments 1 1.71 0.21 Time 1 19.27 0.002 Treatments*Time 1 0.31 0.58 Table 25 The effects of harvesting treatments, stump characteristics, and the interaction of treatments*stump height classes, on the probability of pondcypress sprouts occurring on a stump in seven pondcypress sites in north central Florida. Effect on probability of stumps w/ sprouts No. DF F Value Pr> F Treatments 1 5.8 0.02 Stump Diameter Classes 4 42.69 <0.001 Stump Height Classes 2 16.94 < 0 .001 Treatments*Stump Height Classes 2 3.06 0.048 Table 26 The effects of harvesting treatments, stump characteristics, and the interaction of treatments*stump height classes, on the probability of a stump being bare in seven pondcypress sites in north central Florida. Effect on probability of bare stumps No. DF F Value Pr> F Treatments 1 0.5 0.49 Stump Diameter Classes 4 4.65 < 0.001 Stump Height Classes 2 2.59 0.08 Treatments*Stump Height Classes 2 1.09 0.34 Table 27 The effects of harvesting treatments, stump characteristics, and the interaction of treatments*stump height classes, on the probability of other vegetation occurring on a stump in seven pondcypress sites in north central Florida. Effect on probability of stumps w/ other vegetation No. DF F Value Pr> F Treatments 1 2.33 0.14 Stump Diameter Classes 4 27.76 <0.001 Stump Height Classes 2 19.1 <0.001 Treatments*Stump Height Classes 2 0.41 0.67

PAGE 30

30 Figure 21. Locations of pondcypress wetland study sites in north central Florida by county.

PAGE 31

31 Figure 22. Study design two 0.05 ha circular plots to each harvesting method (sawhead or shearhead) and 3 stump heights per whole plot.

PAGE 32

32 Figure 23. Mean and standard error of proportion of stumps with sprouts through time in seven pondcypress domes in north central Florida. Figure 24. Mean and standard error of cone production through time in seven pondcypress domes in north central Florida.

PAGE 33

33 Figure 25. Treatment effect on stump height in seven pondcypress domes in north central Florida. Figure 26. Effects of stump diameter on likelihood of having a sprout or shrub on a stump or being bare in seven pondcypress domes in north central Florida.

PAGE 34

34 F igure 27 Percentage of species growing on stumps in seven pondcypress domes in north central Florida. Figure 28. Probability of a stump being bare or having a shrub growing on a stump in seven pondcypress domes in north central Florida.

PAGE 35

35 A PPENDIX A SAS CODE FOR GLIMMIX PROCEDURE ANALYZING SPROUTS AND CONES THROUGH TIME PROC IMPORT OUT = proproportions DATAFILE = "e: \ thesis \ proportions.csv" DBMS =CSV REPLACE; GETNAMES=YES; DATAROW= 2 ; RUN; ods rtf ; ods graphics on; proc glimmix data =proportions plots =pearsonpanel; weight stump_count; title 'with time effect included swamp is random sprouts' ; class plot swamp treatments time; model sprouts = treatments|time/ ddfm =sat; random swamp plot(swamp); lsmeans treatments*time/slice =treatments plot=meanplot(cl join sliceby=time) pdiff ; lsmeans treatments|time/ pdiff ; run; proc glimmix data =proportions plots =pearsonpanel; title 'with time effect included swamp is random cones' ; weight stump_count; class plot sw amp treatments time; model cones = time|treatments time/ ddfm =sat; random swamp plot(swamp); lsmeans treatments*time/slice =treatments plot=meanplot(cl join sliceby=time) pdiff ; lsmeans treatments|time/ pdiff ; run; title ; ods graphics off ; ods rtf close ;

PAGE 36

36 APPENDIX B SAS CODE FOR GLIMMIX PROCEDURE ANALYZING STUMP CHARACTERISTICS ods rtf ; ods graphics on; PROC IMPORT OUT = b4 DATAFILE = "e: \ thesis \ book4.csv" DBMS =CSV REPLACE; GETNAMES=YES; DATAROW= 2 ; RUN; proc glimmix data = b4; where stump_ht < 300; title 'sprouts' ; class size cat_diameter swamp treatment plot cat_depth cat_ht; model sprout = treatment cat_diameter cat_depth treatment*cat_depth/ dist =bin solution ddfm =kr; random swamp*plot; lsmeans treatment cat_diameter cat_depth/ ilink pdiff plots =mean(cl join ilink ); lsmeans treatment*cat_depth/ ilink pdiff plots =mean(cl join sliceby=treatment ilink ); output out =p pred=p; run; proc glimmix data = b4; title 'bare stumps' ; where stump_ht < 300; class size cat_diameter swamp treatment plot cat_depth cat_ht; model bare = treatment|cat_diameter|cat_depth@ 1 treatment*cat_depth/ dist =bin solution ddfm =kr; random swamp*plot; lsmeans treatment cat_diameter cat_depth/ ilink pdiff plots =mean(cl join ilink ); l smeans treatment*cat_depth/ ilink pdiff plots =mean(cl join sliceby=treatment ilink ); output out =pb pred=p; run; proc glimmix data = b4 ; title 'other stumps' ; where stump_ht < 300; class size cat_diameter swamp treatment plot cat_depth cat_ht; model other = treatment|cat_diameter|cat_depth@ 1 treatment*cat_depth/ dist =bin solution ddfm =kr; random swamp*plot; lsmeans treatment cat_diameter cat_depth/ ilink pdiff plots =mean(cl join ilink ); lsmeans treatment*cat_depth/ ilink pdiff plots =mean(cl join sliceby =treatment ilink ); output out =po pred=p; run; ods graphics off ; ods rtf close ;

PAGE 37

37 APPENDIX C PERCENTAGE OF SPECIES FOUND ON STUMPS Species Count % of Species on Stump Eupatorium capillifolium 1 0.2 Woodwardia virginica 1 0.2 Smilax sp 2 0.4 Cladium jamaicense 3 0.5 Persea borbonia 3 0.5 Ilex glabra 4 0.7 Acer rubrum 6 1.1 Toxicodendron radicans 15 2.7 Nyssa silvatica 16 2.9 Crataegus aestivalis 20 3.6 Pinus elliotti 21 3.8 Rubus sp 30 5.5 Cephalanthus occidentalis 36 6.6 Lyonia ligustrina 39 7.1 Myrica cerifera 138 25.1 Lyonia lucida 214 39.0

PAGE 38

38 LIST OF REFERENCES Anderson, J.T., Landi, A.A., and P.L. Marks. 2009. Limited flooding tolerance of juveniles restricts the distribution of adults in an understory shrub ( Itea virginica; Iteaceae) Am. J. Bot. 96(9):1 9. Brandt, K., and K.C. Ewel. 1989. Ecology and management of cypress swamps: A review. Bull. 252. Florida Cooperative Extension Service, Institute of Food and Agriculture Science, University of Florida, Gainesville. 19 p. Casey, W. a nd K. Ewel. 2006. Patterns of succession in forested depressional wetlands in North Florida, USA. Wetlands J. 26(8): 147 160. Conner, W.H., 1988. Natural and artificial regeneration of baldcypress ( Taxodium distichum (L.) Rich.) in a Louisiana Swamp. For. Ecol. Manag. 14(8): 305317. Conner, W.H., Toliver, J.R., and F.R. Sklar. 1986. Natural Regeneration of Baldcypress ( Taxodium distichum (L.) Rich.) in a Louisiana Swamp. For. Ecol. Manag. 14: 305317. Conner, W.H., J.G. Gosselink and R.T. Parrondo. 1981. Comparison of the vegetation of three Louisiana swamp sites with different flooding regimes, Am. J. Bot. 68 pp. 320 331. Duryea, M.L. and L.A. Hermansen. 2006. Cypress: Floridas majestic and beneficial tree. Florida Cooperative Extension Service, Institute of Food and Agriculture Science, University of Florida, Gainesville. Circular 1186. Ewel, K.C. and H.T. Odum (eds.). 1984. Cypress swamps University Presses of Florida, Gainesville FL. 3456. Ewel, K.C., H.T. Davis, and J.S. Smith. 1989. Recovery of Florida cypress swamps from clearcutting. South. J. Appl. For. 13(3):123126. Ewel, K.C. 1996. Sprouting by pondcypress ( Taxodium distichum var. nutans ) after logging. South. J. Appl. For. 20():209213. Ewel, K.C. 1998. Pondcypress swamps. pp. 405420 in M. G. Messina and W. H. Conner (eds.). Southern forested wetlands: ecology and management CRC Press, Boca Raton FL. Gardiner, E.S., Russell Jr., D.R., Hodges, J.D., T.C. Fristoe., 2000. Impacts of mechanical tree felling on development of water tupelo regeneration in the Mobile Delta, Alabama. South. J. Appl. For. 24, 65 69.

PAGE 39

39 Golden, M.S. 1999. Factors affecting sprouting success in a bottomland mixed hardwood forest. pp 157 163 In: Haywood JD (ed) Proceedings of the Tenth Biennial Southern Silvicultural Research Conference. USDA For Gen Tech Rep SRS 30, 618 p. Gunderson, L.H. 1984. Regeneration of cypress in logged and burned strands at Corkscrew Swamp Sanctuary, Florida. pp. 349 357. In: K.C Ewel and H.T. Odum (Eds) Cypress Swamps The University of Florida Press, Gainesville, Florida. 472 p. Huenneke, L.F. and R.R. Sharitz. 1986. Microsite abundance and distribution of woody seedlings in a South Carolina cypress tupelo swamp. Am. Midl. Nat 115: 328 335. Hook, D.D., and D.S. DeBell. 1970. Factors influencing stump sprouting of swamp and water tupelo seedlings. USDA For Ser Res P ap. SE57. Southeastern Forest Experiment Station, Asheville, NC. 9 p. Johnson, R.L. 1978. Planning for natural regeneration of hardwoods in the Coastal Plain. In Proc. 2nd Symposium on Southeast hardwoods April, 1977. Dotham, AL. Atlanta, GA: Southeast Area State and Private Forestry, Forest Service, U.S. Department of Agriculture. 114119. Kauppi, A., Rinne, P., and A. Ferm. 1988. Sprouting ability and significance for coppicing of dormant buds on Betula pubescens Ehrh. Stumps. Scand. J. For. Res. 3:343354. Keim, R.F., Chambers, J.L., Hughes, M.S., Dimov, L.D., Conne r, W.H., Shaffer, G.P., Gardiners, E.S., and J.W. Day. 2006. Long term success of stump sprouts in high graded baldcypress water tupelo swamps in the Mississippi Delta. For. Ecol. Manag. 234:24 33. Kennedy Jr., H.E., 1982. Growth and survival of water tup elo coppice regeneration after six growing seasons. South J. Appl. For. 6, 133 135. Kramer, Paul J., T.T. Kozlowski. 1979. Physiology of woody plants New York: Academic Press. 811 p. Lockaby, B.G., Stanturf, J.A., and M.G. Messina 1997. Effects of silvi cultural activity on ecological processes in floodplain forests of the southern United States: a review of existing reports. For. Ecol. Manag. 90(2,3): 93100. Mattoon, W. R. 1915. The Southern Cypress U.S. Department of Agriculture Bulletin 272. Washingt on DC. 74p.

PAGE 40

40 McCreary, D.D., Tietje, W.D., Schmidt, RH, Gross, R., Weitkamp, WH, Willoughby, BL., and F.L. Bell. 2002. Stump sprouting of blue oaks in California. In: Standiford R.B. (Tech. coord.) Proceedings of the Symposium on Oak Woodlands and Hardwood Rangeland Management USDA Forest Service, Pacific SW Research Station, Berkley, California. Gen. Tech. Rep. PSW 126, pp. 6469. Mostacedo, M., Putz, F.E., Fredericksen, T.S., Villca, A., and T. Palacios. 2009. Contributions of root and stump sprouts to na tural regeneration of a logged tropical dry forest in Bolivia. For. Ecol. Manag. 258(6): 978985. Porter, G. Martin, B. Wright, L., Gallagher, T. Stuart, W. and B. Izlar 1984. Occurance of shear damage on southern pine sawlogs American Pulpwood Asso ciation Technology Release 84P 5. 6 p. Randall C.K. 2001. Factors influencing stump sprouting in pondcypress ( Taxodium distichum var. nutans (Ait.) Sweet). M.Sc. thesis, Univ. of Florida, Gainesville, 79p. Randall, C.K., Duryea, M.L., Vince, S.W., and R.J. English. 2005. Factors influencing stump sprouting by pondcypress ( Taxodium distichum var. nutans (Ait.) Sweet). New Forests. 29:245260. SAS Institute 2008. SAS software, version 9.2. SAS Institute Inc. Cary, NC, USA. Schlesinger, W.H. 1978. On the relative dominance of shrubs in Okefenokee Swamp. The Am. Nat. 112(987): 949954. Schneider, R. L. and R. R. Sharitz. 1986 Seed bank dynamics in a southeastern riverine swamp. Am. J. Bot. 73: 1022 30. Sharitz, R.R., R.L Schneider, K.W. Dyer, and N.C. Ma rtin. 1986. Wetland Forest Regeneration and Hydrologic Regime. Annual Re p. of Ecol. Rest. DOE Report SR 81917. P. 4753. Shigo, Alex L. 1986. A new tree biology Shigo and Trees, Associates, Durham, New Hampshire. 595 p. Southeast Regional Climate Center (SERCC). 2007. Historical Florida Weather Available online at http://www.sercc.com/climateinfo/historical/historical_fl.html ; last a ccessed Dec 12, 2009. Terwilliger, V.J. and K.C. Ewel. 1986. Regeneration and growth after logging in Florida cypress domes. For. Sci. 32:493506.

PAGE 41

41 Visser, J. M., and C. E. Sasser. 1995. Changes in tree species composition, structure, and growth in a bald cypress water tupelo s wamp forest: 19801990. For. Ecol. Manag. 72:119 129. Wharton, C.H., H.T. Odum, K. Ewel, M. Duever, A. Lugo, R. Boyt, J. Bartholomew, E. DeBellevue, S. Brown, M. Brown, and L. Duever. 1977. Forested Wetlands of Florida: Their management and use. Div. State Plan., Tallahassee, FL. Wilhite, L., J. Toliver 1990. Taxodium distichum (L.) Rich. Baldcypress. In: Burns, R.; Honkala, B.; tech. cords. Silvics of North America Vol. 1. Conifers. Agricultural Handbook 654. Washington, DC: U.S. Department of Agricult ure, Forest Service: 563572.

PAGE 42

42 BIOGRAPHICAL SKETCH Nicole Ricci, a native to Macomb, Michigan, attended Macomb Community College as an Art and Design major, before transferring to Northern Michigan University to follow a bachelor s degree in Earth Science. During her studies at NMU, she earned the Farrell Cartography Scholarship for achieving the highest GPA in all mapping science courses. After completing her Bachelor of Science, she attended Michigan Technological University in pursuit of a graduate degree in forestry and wetland science. She graduated with a coursework Master of Science in December of 2007. In June of 2008, she entered the School of Forest Resources and Conservation at the University of F lorida to pursue a thesis track Master of Science. Throughout her time at UF, she enjoyed learning many southern traditions, and gained lifelong friendships.