<%BANNER%>

Demographic Responses of the Narrowly Endemic Conradina brevifolia (Short-Leaved Rosemary, Shinners) to Mowing in Florid...

Permanent Link: http://ufdc.ufl.edu/UFE0044813/00001

Material Information

Title: Demographic Responses of the Narrowly Endemic Conradina brevifolia (Short-Leaved Rosemary, Shinners) to Mowing in Florida Scrub
Physical Description: 1 online resource (57 p.)
Language: english
Creator: Proenza, Lynn M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: endangered -- endemic -- florida -- mechanical -- plant -- restoration -- scrub
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: Conradina brevifolia (short-leaved rosemary, Shinners) is state and federally listed as endangered. It is a subshrub narrowly endemic to the pyrogenic Florida scrub ecosystems on the Lake Wales Ridge where it is vulnerable to extinction due to human activities including development, agriculture, and fire suppression and exclusion. Fire suppression and exclusion can lead to fuel accumulation reducing biodiversity and can be hazardous near the wildland-urban interface. Mechanical treatments, such as mowing, is a land management tool that is becoming more widely used in upland ecosystems as a pre-fire treatment or a fire surrogate and can reduce vegetation cover, thus ameliorating the hazards of fuel accumulation and restoring fire regimes in order to accelerate habitat restoration and better protect this species.   Within the Lake Wales Ridge State Forest, various responses of C. brevifolia to mowing were tested including mortality, height and growth rate, flowering, and density. Bare sand and vegetation cover following mowing were also measured since changes in bare sand and vegetation cover could greatly impact C. brevifolia response, a potential gap specialist. Responses were measured prior to treatment, nine months post-treatment, and one year post-treatment. In response to mowing, C. brevifolia resprouted from aboveground stems. Total stem density and mortality of C. brevifolia was reduced in mechanically treated plots, but was not significantly lower than in control plots. This suggests that mowing may be used to restore scrub while simultaneously preserving C. brevifolia populations in areas where this species is locally abundant. In mechanically treated plots, flowering response was delayed or did not occur during the flowering season following treatment. Vegetation cover was not significantly reduced within one year of mowing. Due to pre-existing dense litter accumulation that was not greatly affected by treatment, bare sand cover did not differ between the two treatments. When restoring areas of scrub habitat that contain sufficient populations of C. brevifolia mowing can be used without causing significant loss of C. brevifolia individuals and resprouting is likely to occur. Results do not support the use of mechanical treatments as a fire surrogate because vegetation cover recovers within one year of treatment and litter accumulation is not reduced and, therefore, can result in the reduction of bare sand cover that may be critical to the survival of C. brevifolia.
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 Lynn M Proenza.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Andreu, Michael Gardner.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044813:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044813/00001

Material Information

Title: Demographic Responses of the Narrowly Endemic Conradina brevifolia (Short-Leaved Rosemary, Shinners) to Mowing in Florida Scrub
Physical Description: 1 online resource (57 p.)
Language: english
Creator: Proenza, Lynn M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: endangered -- endemic -- florida -- mechanical -- plant -- restoration -- scrub
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: Conradina brevifolia (short-leaved rosemary, Shinners) is state and federally listed as endangered. It is a subshrub narrowly endemic to the pyrogenic Florida scrub ecosystems on the Lake Wales Ridge where it is vulnerable to extinction due to human activities including development, agriculture, and fire suppression and exclusion. Fire suppression and exclusion can lead to fuel accumulation reducing biodiversity and can be hazardous near the wildland-urban interface. Mechanical treatments, such as mowing, is a land management tool that is becoming more widely used in upland ecosystems as a pre-fire treatment or a fire surrogate and can reduce vegetation cover, thus ameliorating the hazards of fuel accumulation and restoring fire regimes in order to accelerate habitat restoration and better protect this species.   Within the Lake Wales Ridge State Forest, various responses of C. brevifolia to mowing were tested including mortality, height and growth rate, flowering, and density. Bare sand and vegetation cover following mowing were also measured since changes in bare sand and vegetation cover could greatly impact C. brevifolia response, a potential gap specialist. Responses were measured prior to treatment, nine months post-treatment, and one year post-treatment. In response to mowing, C. brevifolia resprouted from aboveground stems. Total stem density and mortality of C. brevifolia was reduced in mechanically treated plots, but was not significantly lower than in control plots. This suggests that mowing may be used to restore scrub while simultaneously preserving C. brevifolia populations in areas where this species is locally abundant. In mechanically treated plots, flowering response was delayed or did not occur during the flowering season following treatment. Vegetation cover was not significantly reduced within one year of mowing. Due to pre-existing dense litter accumulation that was not greatly affected by treatment, bare sand cover did not differ between the two treatments. When restoring areas of scrub habitat that contain sufficient populations of C. brevifolia mowing can be used without causing significant loss of C. brevifolia individuals and resprouting is likely to occur. Results do not support the use of mechanical treatments as a fire surrogate because vegetation cover recovers within one year of treatment and litter accumulation is not reduced and, therefore, can result in the reduction of bare sand cover that may be critical to the survival of C. brevifolia.
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 Lynn M Proenza.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Andreu, Michael Gardner.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044813:00001


This item has the following downloads:


Full Text

PAGE 1

1 DEMOGRAPHIC RESPONSES OF THE NARROWLY ENDEMIC C onradina brevifolia (SHORT LEAVED ROSEMARY, SHINNERS) TO MOWING IN FLORIDA SCRUB B y LYNN PROENZA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PART IAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Lynn Proenza

PAGE 3

3 To my mother and father, my family and friends for their infinite support and guidance

PAGE 4

4 A CKNOWLEDGMENTS I am grateful to all those who have assisted me through this process, especially my mother and father who have supported me without reserve I am thankful to my committee members Dr. Michael Andreu, Dr. Kimberly Bohn, and Dr. Doria Gordon wh o have given me immense guidance. I would also like to express my gratitude to Dr. Wendell Cropper and Dr. Gerard Del Vecchio for their additional support and advice. I am especially thankful to Monica McGarrity for sacrificing he r time to review my work and to John Lagrosa for his endless suggestions and advice throughout this process ; Kenny Lopiano for his help with statistical design ; and my friends Monica McGarrity, Carolyn Cheatham Rhodes, Melissa Friedman, and Ashley Tyre for t heir time and effort to help collect data I would like to thank the Florida Department of Agriculture and Consumer Services, Florida Forest Service (FFS) for funding through the Florida Endangered and Threatened Plant Conservation Program. Without them, this project would not have been possible. I am also grateful to the FFS personnel including David Butcher, Keith Clanton, Mark Calhoun, and Pete Lewis for help with mechanical treatments and especially to Jennifer Navarra for all that she has done. My si ncere gratitude goes out to my family and to Dan iel for their love and support, my friends for listening, and last but not least Chico and Diego, my two wonderful dogs who have sacrificed many hikes and long walks throughout this process. I owe you one.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 LIST OF ABBREVIATIONS ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 12 The Lake Wales Ridge ................................ ................................ ................................ ............ 12 Fire ................................ ................................ ................................ ................................ .......... 13 Fire Suppression and Exclusion ................................ ................................ .............................. 13 Mechanical Treatment use as a Surrogate or Pre Fire Treatment ................................ .......... 15 Study Species ................................ ................................ ................................ .......................... 16 Objectives and Hypotheses ................................ ................................ ................................ ..... 18 2 MATERIALS AND METHODS ................................ ................................ ........................... 23 Study Site ................................ ................................ ................................ ................................ 23 Sampling Protocol ................................ ................................ ................................ .................. 23 Data Analysis ................................ ................................ ................................ .......................... 25 3 RESULTS ................................ ................................ ................................ ............................... 30 Conrad ina brevifolia Responses ................................ ................................ ............................. 30 Percent Mortality ................................ ................................ ................................ ............. 30 Height and Growth Rate ................................ ................................ ................................ .. 30 Flowering ................................ ................................ ................................ ......................... 31 Density ................................ ................................ ................................ ............................. 31 Vegetation Cover ................................ ................................ ................................ .................... 31 Bare Sand Cover ................................ ................................ ................................ ..................... 32 4 DISCUSSION ................................ ................................ ................................ ......................... 42 Conradina brevifolia Responses ................................ ................................ ............................. 42 Percen t Mortality ................................ ................................ ................................ ............. 42 Height and Growth Rate ................................ ................................ ................................ .. 42 Flowering ................................ ................................ ................................ ......................... 42

PAGE 6

6 Density ................................ ................................ ................................ ............................. 43 Resprouting ................................ ................................ ................................ ...................... 44 Vegetation Cover ................................ ................................ ................................ .................... 45 Bare Sand Cover ................................ ................................ ................................ ..................... 46 Research Needs ................................ ................................ ................................ ....................... 46 5 CONCLUSIONS AND MANAGEMENT IMPLICATIONS ................................ ................ 50 LIST OF REFE RENCES ................................ ................................ ................................ ............... 52 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 57

PAGE 7

7 LIST OF TABLES Table page 3 1 Percent mortal ity of Conradina brevifolia individuals at plot center (i.e., CPP) at one year post treatment ................................ ................................ ................................ ............. 34 3 2 Net change in height of Conradina brevifolia individuals at plot center (i.e., CPP) at one year post treatment ................................ ................................ ................................ ...... 35 3 3 Flowering response of Conradina brevifolia individuals at plot center (i.e., CPP) pre treatment and nine months post treatment ................................ ................................ ......... 36 3 4 Conradina brevifolia density (i.e., number of rooted stems) per treatment type at one year post treatment ................................ ................................ ................................ ............. 37 3 5 N et Change in perc ent vegetation cover at one year post treatment ................................ 39 3 6 Net change in percent bare sand cover at one year post treatment ................................ .... 41

PAGE 8

8 LIST OF FIGURES Figure pag e 1 1 Map of t he Lake Wales Ridge in Central Florida ................................ .............................. 20 1 2 Photograph of Conradina brevifolia in the L ake Wales Ridge ................................ ......... 21 1 3 Photograph showing Conradina brevifolia occur ring in its natural environment ............. 22 2 1 The Arbuckle Tract, one of four tracts comprising the Lake Wales Ridge State Forest, lies on the eastern border just below the cen ter of The Lake Wales Ridge ........... 26 2 2 Location of study plots accor ding to habitat type ................................ .............................. 27 2 3 Mechanical treatments to mimic mowing were carried out via hand tools inclu ding chainsaws, loppers, and metal bladed weed eaters ................................ ............................ 28 2 4 Photographs showing an exam ple of conditions immediately before and after mechanical treatment ................................ ................................ ................................ ......... 29 3 1 Average net change in height of Conradina brevifolia individual s at plot center (i.e., CPP) per treatment type ................................ ................................ ................................ ..... 34 3 2 Growth rate of Conradina brevifolia at one year post treatment. ................................ ..... 36 3 3 Average net change of Conradina brevifolia per treatment type at one year post treatment ................................ ................................ ................................ ............................ 38 3 4 Change in percent vegetation cover over time following a mowing treatment ................. 40 4 1 Photograph of r esprouting Conradina brevifolia at one year post treatment following mowing ................................ ................................ ................................ .............................. 48 4 2 Photograph of Conradina brevifolia resprouting following a low intensity fire ............... 49

PAGE 9

9 LIST OF ABBREVIATION S CPP Center Plot Plant FFS Florida Fo rest Service FNAI Florida Natural Areas Inventory LWR Lake Wales Ridge LWRSF Lake Wales Ridge State Forest SE Standard Error WUI Wildland Urban Interface

PAGE 10

10 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 DEMOGRAPHIC RESPONSES OF THE NARROWLY ENDEMIC Conradina brevifolia (SHORT LEAVED ROSEMARY, SHINNERS) TO MOWING IN FLORIDA SCRUB By Lynn Proenza August 2012 Chair: Michael Andreu Major: Forest Resources and Conservation Conradina brevifolia ( short leaved rosemary, Shinners) is state and federally listed as endangered. It is a subshrub narrowly endemic to the pyrogenic Florida scrub ecosystems on the Lake Wales Ridge where i t is vulnerable to extinction due to human activities including development, agriculture, and fire suppression and exclusion Fire suppression and exclusion can lead to fuel accumulation reduc ing biodiversity and can be hazardous near the wildland urban in terface M echanical treatments, such as mowing, is a land management tool that is becoming more widely used in upland ecosy s tems as a pre fire treatment or a fire surrogate and can reduce vegetation cover thus ameliorating the hazards of fuel accumulation and restoring fire regimes in order to accelerate habitat restoration and better protect this species. Within the Lake Wales Ridge State Forest, various responses of C. brevifolia to mowing were tested including mortality, height and growth rate, flower ing, and density B are sand and vegetation cover following mowing were also measured since changes in bare sand and vegetation cover could greatly impact C. brevifolia response, a potential gap specialist. Responses were measured prior to treatment, nine m onths post treatment, and one year post treatment

PAGE 11

11 In response to mowing C. brevifolia resprout ed from aboveground stems Total stem d ensity and mortality of C. brevifolia was reduced in mechanically treated plots, but was not significantly lower than in control plots This suggests that mowing may be used to restore scrub while simultaneously preserving C. brevifolia population s in areas where this species is locally abundant In mechanically treated plots, flowering response was delayed or did not occur during the flowering season following treatment. Vegetation cover was not significantly reduced withi n one year of mowing Due to pre existing dense litter accumulation that was not greatly affected by treatment bare sand cover did not differ bet ween the two treatments. When restoring areas of scrub habitat that contain sufficient populations of C. brevifolia mowing can be used without causing significant loss of C. brevifolia individuals and resprouting is likely to occur. Results do not support the use of mechanical treatments as a fire surrogate because vegetation cover recovers within one year of treatment and li tter accumulation is not reduced and, therefore, can result in the reduction of bare sand cover that may be critical to the survival of C. bre vifolia

PAGE 12

12 CHAPTER 1 INTRODUCTION The Lake Wales Ridge The relic dune system of the Lake Wales Ridge (LWR) in central Florida (Figure. 1 1) spans a mere 186 km long, 11.7 km wide at its broadest point, but encompasses more than 209 ,000 ha of diverse ecos ystems including the pyrogenic ( i.e. fire maintained) Florida scrub (Weekley et al., 2008a) Scrub habitat is characterized by deep, well drained, infertile soils and dominated by patches of bare sand (i.e. gaps) and xeromorphic plants such as Pinus clau sa ; o aks including Quercus chapmanii, Q. geminata, Q. inopina, and Q. myrtifolia ; Serenoa repens ; Ceratiola ericoides ; woody shrubs ; and sparse herbs and graminoids (Myers, 1990; Menges, 1999). The upland xeric habitats of the LWR are home to many endemic flora (Menges, 1999, Christman and Judd 1990), including 29 plant species listed as threatened or endangered (Turner et al., 2006; FNAI, 2010). The concentration of imperiled species on the LWR makes it one of the most imperiled hotspots in the United Stat es (Dobson et al., 1997). Even b efore human settlement of the LWR the upland xeric ecosystems within the LWR were rare (Myers, 1990) making them extremely vulnerable to habitat loss due to human activities. According to the most recent land use assessment for the LWR ( Weekley et al. 2008a), anthropogenic activities such as citrus agriculture, residential and commercial development, and ranching (Turner et al., 2006; USFWS, 1996) have resulted in the loss of 7 7 % of the original 134,711 ha of u pland xeric habitat (Weekley et al., 2008a). The remaining undeveloped sites now exist as fragments surrounded by residential, commercial, and agricultural land (Turner et al., 2006). As a result, Florida scrub is now listed as imperiled both globally and locally (Noss et al., 1995; FNAI, 2010).

PAGE 13

13 Fire Fire is one of the environmental factors that has shaped evolution of Florida scrub habitats (Abrahamson et al., 1984a). T h is pyrogenic ecosystem require s fire to maintain biodiversity and health of flora and f auna ( Schmalzer et al., 2003; Myers, 1990). Fire return intervals vary among p lant communities, but typically range from 5 to 30 years (Menges, 2007). Florida scrub is characterized by somewhat infrequent fires that commonly occur during severe weather and accumulated fuel conditions. F ires typically top kill most shrubs and oaks completely kill sand pines reducing vegetation densities and consumes litter on the forest floor This creates bare sand patches called gaps, which allow seedling recruitment of diverse, fire dependent flora many of which are endemic and imperiled (Menges et al., 2008; Abrahamson, 1984b; Greenberg et al., 1995; Menges et al., 2008; Myers, 1990). These gaps typically reduce in size with time since fire as plant communities recover (Greenberg et al., 1995; Menges et al., 2008). Fire Suppression and Exclusion In the past century, fire suppression and exclusion ha ve altered species composition in scrub habitat and reduced seedbank densities, production, and viability causing signific ant decrease s in populations of scrub vegetation species (Menges, 2007; Abrahamson and Abrahamson, 1996). In addition, lack of fire has resulted in overgrowth of scrub vegetation to subcanopy and canopy heights (Menges, 2007; Duncan and Schmalzer, 2004; Sc hmalzer et al., 1994; Guerin, 1993) creat ing extensive heavily shaded environments Over time, this overgrowth may result in conversion of scrub to a xeric hammock plant community (Abrahamson and Abrahamson, 1996) further promoting fire resistance (Myers 1985; Veno, 1976; Menges, 2007; Schmalzer et al., 2003). Today, the health of Florida scrub is threatened by species composition and structural changes due to lack of fire (Menges et al., 1993; Schmalzer et al., 2003; Abrahamson and

PAGE 14

14 Abrahamson, 1996) or disruption of the natural fire regime (Weekley and Menges, 2003). Fire suppression is positively correlated with increased vegetation cover and results in fuel accumulation (Williges et al., 2006). Increased fuel loads can influence fire intensities and ul t imately alter vegetation response (Dodge, 1972; Brown, 1983; Givens et al., 1984; Fernandes and Botelho, 2003; Schmalzer et al., 1994; Wally et al., 2006). Fire suppression may also h inder seed germination as accumulated fuel forms a layer of litter on th e forest floor (Petru and Menges, 2003). Furthermore, lichen populations tend to increase in density with time since fire inhibit ing seed germination (Hawkes and Menges, 1996; Hawkes and Menges, 2003). In areas with continued fire suppression plant specie s that require bare sand to germinate are relegated to habitat edges including firebreaks (Hall et al., 2002; Menges, 1999). In order to preserve the biodiversity and health of Florida scrub fire must be reintroduced ( Schmalzer et al., 2003; Myers, 1990). However, restoring fire regimes, especially in or near the wildland urban interface (WUI) can be challenging (Kalabokidis and Omi, 1998) In many cases fire suppression has resulted in unsafe prescribed fire conditions. In addition to safety concerns, n umerous fact ors including public opinion smoke management issues (e.g., proximity to smoke sensitive areas such as residential and commercial development) liability risk, and potential for fire escape can pose difficulties for land managers when attempti ng to execute prescribed fires (Haines and Busby, 2001; Myers, 1990, Rickey et al. 2007, Long et al., 2004). Habitat fragmentation further exacerbates the problem by reducing the extent of burned areas (Duncan and Schmalzer, 2004) because habitats lack con nect ivity and are now in capable of burning across vast areas (Myers, 1990). Many remaining scrub habitats, especially those near the WUI tend to be long unburned (Weekley et al., 2011; Rickey et al., 2007; Abrahamson and Abrahamson, 1996) L and managers ar e now turning towards the use of mechanical methods as

PAGE 15

15 pre fire treatments or fire surrogates to reduce fuel loads in order to maintain or restore scrub ecosystems (Menges and Gordon, 2010). Mechanical Treatment use as a Surrogate or Pre Fire Treatment M ec hanical treatments such as logging, mowing, and roller chopping have been shown to reduce prescribed fire hazards, assist in restoring natural fire regimes (Wally et al., 2006, Stephens and Ruth, 2005; Menges and Gordon, 2010), and accelerate restoration (Rickey et al., 2007) although weather conditions and time since treatment can alter outcomes of mechanical treatments. Furthermore mechanical treatments may be beneficial on lands with multiple management goals (e.g. ecosystem management and harvestin g) or in urban areas with remnant habitat (Greenberg et al., 1995; Long et al. 2004). Th e effectiveness and versatility of m echanical treatments make this a useful management tool especially when used prior to prescribed fire in restoration of long unburne d scrub (Rickey et al., 2007; Schmalzer et al., 2003) W ith out mechanical treatment canopy cover in long unburned scrub can rebound quickly within a few years of fire (Abrahamson and Abrahamson, 1996; Schmalzer et al., 2003) requiring numerous fire events to reduce fuel loads (Abrahamson and Abrahamson, 1996 ; Long et al., 2004) and litter cover (Rickey et al., 2007) As a pre fire treatment, mechanical treatments are as effective as or more effective than burning alone to reduce vegetation cover (Schmalzer et al., 1994, Weekley et al., 2011) Mechanical treatments affect numerous components of scrub ecosystems. Mechanical treatments can simulate ecosystem responses to fire disturbance s uch as reduced competition and increased light, soil temperature, and in some situations (e.g., logging treatments), bare sand coverage. F ire has been shown to cue some scrub plants to flower, germinate, and thrive and mechanical treatments can produce similar cues in some species ( Menges, 1999; Greenberg et al., 1995; Weekl ey et al., 2011; Menges, 2001; Williges, 2006).

PAGE 16

16 To date, little is known of the long term effects of using mechanical treatments as fire surrogates and species specific responses to mechanical treatments in Florida scrub (Menges and Gordon, 2010). Known d isadvantages to some forms of mechanical treatment use (e.g., logging) include an increase in soil disturbance which may cause increases in exotic plant abundance soil compaction, altered fire intensities ( Menges, 1999; Weekley et al., 2008b; Williges et al., 2006; Rickey et al., 2007, Wally et al. 2006), soil erosion, and water quality degradation (Long et al., 200 4 ). When used as a surrogate rather than a pre fire treatment, mechanical treatments can result in an increase of accumulated slash and litter thus reducing bare sand patches (Greenberg et al., 1995; Menges 2001; Weekley et al., 2011; Rickey et al. 2007). In addition, S repens is a desirable species in fire management and aids fire movement, but can be negatively affected by some forms of mec hanical treatments such as roller chopping or logging ( Schmalzer et al., 2003; Schmalzer et al., 1994; Greenberg et al., 1995). Fire can be more effective than using mechanical only treatments as fire surrogates for increasing percent bare sand cover by de creasing litter cover and depth, and is especially beneficial to rare, endemic plants found throughout the LWR (Weekley et al., 2011). In addition, herbaceous species richness is greatest in treatments with a fire component (Williges et al., 2006). Study S pecies Conradina brevifolia (short leaved rosemary Shinners) is narrowly endemic to Highlands and Polk Counties within the LWR ( Gray, 1965; USFWS, 2008) and is federally listed as endangered (USFWS, 1993). It is a small, woody mint shrub (Fig ure 1 2 ) that is specialized to white sand ridges of xeric scrub ecosystems ( Gray, 1965; Kral, 1983; Fig ure 1 3 ) that are dominated by xeromorphic oak species ( Quercus chapmanii, Q. geminata, Q. inopina, and Q. myrtifolia ) and typically have an open canopy of Pinus cla usa (USFWS 1996). The original distribution for C. brevifolia is not well known; however, it is one of the most narrowly endemic

PAGE 17

17 scrub plants on the LWR and its absence from suitable habitat beyond its present distribution suggests it may have always been narrowly distributed (USFWS, 1996). Due to its narrow distribution, C. brevifolia populations are s usceptible to extirpation from a ny given site or even extinction which could result from a variety of mechanisms including anthropogenic activities and cata strophic natural events ( Gray, 1965; USFWS 1996 ; Kral, 1983 ). As of July 2008 there were approximately 35 known populations of C. brevifolia 17 that are protected on private and publicly owned lands and 18 that are not protected and assumed to be un manag ed. A erial photography suggests that 10 of the 18 unprotected populations have recently been destroyed by development or remain in small remnants surrounded by development (USFWS, 2008) As of the completion of this study, a full population survey ha s not been conducted on all extant populations, and some old surveys have not been reassessed in more than two decades (USFWS, 2008). Currently, research on the biology and ecology of C. brevifolia is limited; such research is necessary in order to determine the best management tools for the recovery or maintenance of this endangered species (USFWS, 2008). Research has shown that many rare and endemic scrub species found throughout the LWR has similar requirements for gaps in order to thrive (Abrahamson, 1984c; Johnson et al., 1986; Quintana Ascencio et al., 1998; Quintana Ascencio and Morales Hernndez, 1997; Menges and Hawkes, 1998; Menges et al. 2008). Based on observations, it has been suggested that C. brevifolia may be a gap specialist and persist in areas with low vegetation cover and high bare sand cover (Clanton, 2010, personal communication). Conradina brevifolia was observed mainly in firela nes, plow lines, roadsides, and skid trails where gaps are prevalent (Weekley et al., 2008b), suggesting that this species may benefit from fire or other disturbances that help reduce litter and shrub cover and creat e gaps. Furthermore, C brevifolia has been classified as an

PAGE 18

18 obligate seeder, a plant that reproduces only from seed typically following a fire event (USF WS, 1996; Slapcinsky et al., 2010) ; however this status has not been experimentally tested. Conradina glabra a similar species is known to recruit via seedbank and resprout following a fire event ; h owever, as aboveground tissue damage or loss increased, resprouting and survival probability decreased ( Gordon 1996). Dicerandra frutescens is an obligate seeder that is killed when burned, cut, or otherwise defoliated (Menges, 1992) and some scientists suggest that C. brevifolia may exhibit a similar respons e to disturbance and may require similar land management strategies (USFWS, 1996 ; Kral, 1983 ). However, densities of C. brevifolia have been shown to increase following fire and logging disturbance events ( Weekley et al. 2008b; Slapcinsky et al. 2010) s uggesting that fire and mechanical treatments could prove beneficial. Objectives and Hypotheses In order to achieve C. brevifolia population protection and reach conservation goals for this species, additional research into the biology and ecology of t his species and its response to fire and other disturbances is critical. In response to this need for research, I studied demographic responses of C. brevifolia to mowing a type of low impact mechanical treatment to reduce vegetation while leaving some li ve tissue aboveground Variables were separated into two groups: those that were measured only on the C. brevifolia individual occurring at the plot center ( i.e., center plot plant; CPP) and those that were measured at the plot level. For v ariable s measure d on the CPP, I determined presence or absence of the plant to determine mortality at one year post treatment ; I measured height to determine annual recovery and growth rate at one year post treatment ; and I determined presence or absence of flowering to d etermine flowering response at nine months post treatment. At the plot level I counted the number of rooted stems to determine C. brevifolia net change in density ( i.e. net gain or net loss) at one year post treatment and measured percent vegetation cover and percent bare sand cover to determine how

PAGE 19

19 mechanical treatments affect the surrounding environment. These findings provide insights for guiding best management practices that may help to minimize population declines. Research on the responses of C. br evifolia to mowing has not been conducted Because s ome scientists believe C. brevifolia would have similar responses to mechanical treatments as that of D. frutescens and would, therefore, not survive if cut ( USFWS, 1996; Menges, 1992) I hypothesized that C. brevifolia would be killed by mechanical treatments within one year of treatment and that this species would experience greater mortality rates and significant density reduction in mechanically treated plots than in untreated control plots No previous studies or anecdotal evidence exists for flowering response, height, or growth rate of C. brevifolia ; t herefore, I hypothesized that flowering response, height, and growth rate would be stunted due to mechanical treatments. Based on previous studies, vege tation recovers from mechanical treatment within two to five years of treatment (Weekley et al., 2008; Weekley et al. 2011) and percent bare sand cover decreases (Greenberg et al., 1995; Rickey et al., 2007) without the additional treatment of fire ; theref ore, I hypothesized that vegetation cover would not reach pre treatment levels within one year post treatment and bare sand cover would be significantly reduced in mechanically treated plots than in untreated control plots

PAGE 20

20 Figure 1 1. The Lake Wa les Ridge in Central Florida is only 186 km long and 11.7 km wide yet harbors many rare, endemic species. ( Lake Wales Ridge mapping data provided by Archbold Biological Station )

PAGE 21

21 Figure 1 2 Conradina brevifolia is a small, woody mint shrub narrowly ende mic to the Lake Wales Ridge. ( Photo graph 2011 by: Lynn Proenza )

PAGE 22

22 Figure 1 3 Conradina brevifolia occurs in Florida scrub ecosystems characterized by deep, infertile, sandy soils and is commonly observed in bare sand patches. Numerous individuals can b e seen in the center of this photograph. ( Pho tograph 2011 by Lynn Proenza )

PAGE 23

23 CHAPTER 2 MATERIALS AND METHOD S Study Site The Lake Wales Ridge State Forest (LWRSF) is a multiple use state forest in southeastern Polk County, Florida (Fig ure 2 1 ) managed by the Florida Department of Agriculture and Consumer Services, Florida Forest Service (FFS). The LWRSF is comprised of various natural communities including Florida scrub (FDACS, 2006) This study was conducted in the southeastern portion of the Arbuckle Tra ct, one of four separate tracts that make up the LWRSF All plots were located in Florida scrub (Figure 2 2). Soils consisted of white and gray sands including: Satellite, Narcoossee, and Archbold which are somewhat poorly drained to moderately well draine d sandy soil s (Soil Conservation Service, 1990). Based on the most recent thirty year climate normals (1981 2012) for this area, t he ave rage annual temperature is 22.5 C and the average precipitation is 129.0 c m per year During the year following mechanic al treatment, the average annual temperature was 0.9C above average and annual precipitation was 32 cm below average (NOAA 2012) Sampling Protocol Using waypoints from FFS population surveys for Conradina brevifolia 20 circular 8 m diameter plots surro unding a single C. brevifolia individual were randomly selected and assigned a treatment type: mechanical only (10 plots) or an untreated control (10 plots) Pre t reatment s urveys w ere conducted in March 2011. Data collection consisted of detailed physical information for the CPP includ ing presence or absence, height of live growth and the presence or absence of flower s Additional data collection at the plot level included density of C. brevifolia percent cover of vegetation and percent cover of bare sa nd The lowermost decumbent branches of C. brevifolia occasionally root from the nodes (Gray, 1965) as

PAGE 24

24 Conradina verticillata a similar species, has exhibited. With time these branches are covered by substrate and separate plants are not distinguishable ( Roulston, 1994); t herefore, density of C. brevifolia at the plot level was determined by counting rooted stems. N ine month post t reatment s urveys occurred in February 2012 and the one year post t reatment s urveys occurred in May 2012. Data collection proced ures for these two survey events mimicked pre t reatment surveys but also included quantification of all C. brevifolia individuals that resprouted from mechanical treatments. With the exception of the CPP individual specific data on each C. brevifolia root ed stem w ere not collected during any of the three sampling surveys For this reason newly recruited or surviving individuals for which mechanical treatment was not obvious at one year post t reatment w ere grouped into one category. M echanical treatment wa s conducted during the spring season in May 2011. Using chainsaw s loppers, and metal bladed weed eaters this mechanical treatment mimicked mowing ; therefore, some aboveground plant tissue (i.e., cut stems) remained and soil disturbance was minimal (Fig ur e 2 3 ). All vegetation occurring within each mechanically treated plot was cut and left in the position in which it fell ; no vegetation was physically removed (Fig ure 2 4 ). This study initially included a fire component that was conducted following the pre treatment survey and the mowing treatment but the prescribed fire was incomplete and burned only three plots (2 mechanical, 1 control) ; therefore, the three plots were removed from all data analysis. In addition, a misidentification within one control plo t resulted in the removal of C. brevifolia rooted stem data for that location but data for the CPP and percent vegetation and bare sand cover for that location were included in analysis One CPP in mechanical treatments previously believed dead at nine mo nths post treatment was observed newly resprouted at one year post test, this datum was identified as an outlier. T o have a clear

PAGE 25

25 indication of actual height change, growth rates, and recovery of C. brevifolia to mechanical tre atment, data on CPP average net change in height and growth rate was analyzed without this datum. Data Analysis The variables tested on the C. brevifolia individual occurring at plot center included CPP percent mortality, CPP net change in height and growt h rate, and CPP flowering response The variables tested at the plot level included net change in C. brevifolia density percent vegetation cover, and percent bare sand cover. All variable s were compared between pre and one year post treatment with the ex ception of flowering response. Flowering response was compared between pre and nine month post treatment because the one year post treatment survey occurred at the end of the flowering period. level. test w as used to compare percent vegetation cover between treatments at the plot level test, test t hat account s for both unequal sample sizes and unequal variances (Welch, 1947), was used to compare CPP net change in height and growth rate, net change in C. brevifolia density at the plot level and percent bare sand cover at the plot level between treatments Due to a small sample size, a compare CPP percent mortality and CPP f lowering status between treatments

PAGE 26

26 Figure 2 1 The Arbuckle Tract, one of four tracts comprising t he Lake Wales Ridge State Forest, lies on the eastern border just below the center of The Lake Wales Ridge. ( Lake Wales Ridge boundar y data provided by Archbold Biological Station ; Arbuckle Tract boundary data provided by FFS)

PAGE 27

27 Figure 2 2 Location of study plots according to habitat type; all plots were located in scrub habitat. (Community type data provided by FNAI)

PAGE 28

28 Figure 2 3 Mechanical treatments to mimic mowing were carried out via hand tools including chainsaws, loppers, and metal bladed weed eaters. Vegetation was cut and left in the position in which it fell. ( Photograph 2011 Lynn Proenza )

PAGE 29

29 (A) (B) Figure 2 4 Phot ographs showing an example of conditions immediately before ( A ) and after ( B ) mechanical treatment. ( Photographs 2011 by Lynn Proenza )

PAGE 30

30 CHAPTER 3 RE SULTS Conradina brevifolia Responses Percent Mortality Percent mortality was based on the presence or abs ence of the CPP at one year post treatment. Percent mortality was greater in mechanical treatment plots than in control plots, although this difference was not significant (n = 17, p = 0.212 ). Control plots averaged 11.1% ( SE = 11.1%) and mechanical treatm ent plots averaged 37.5% ( SE = 18.3% ; Table 3 1 ). Height and Growth Rate Live growth on only those surviving CPP individuals was measured; all others were excluded from data analysis. Average net change in height was not affected by treatment type (n = 12 t = 1.68, p = 0.185 ; Figure 3 1 ). In control plots (n = 8) one year post treatment, average height of CPP was 91.4 cm ( SE = 9.0 cm) represent ing an average net gain in height of 6.0 cm ( SE = 3.3 cm). In mechanical treatment plots (n = 4) one year post tr eatment, average height of CPP was 47.0 cm ( SE = 11.9 cm) represent ing an average net loss in height of 25.3 cm ( SE = 18.3 cm ; Table 3 1) Growth rates of CPP were significantly higher in mechanical treatment plots than in control plots (n = 12, t = 3.32, p = 0.036; Figure 3 2 ). Mechanically treated C. brevifolia was observed to resprout from the base of cut stems. Based on the assumption that all mechanically treated plants resprouted from the ground level (i.e., height was at 0 cm immediately following t reatment) average annual growth rate of CPP in control plots (n = 8) was 6.0 cm ( SE = 3.3 cm) by one year post treatment and 47.0 cm ( SE = 11.9 cm) in mechanically treated plots (n = 4) by one year post treatment (Figure 3.2)

PAGE 31

31 Flowering During pre treatm ent surveys, all 17 CPP in both treatments were flowering. At the nine month post treatment survey, significantly fewer of the surviving CPP (n = 5) were flowering in mechanical treatment plots as compared to CPP (n = 8) in control plots (n = 13, p = 0.032 ; Table 3 3 ). Density On average, density of C. brevifolia in control treatment plots increased by 1.6 ( SE = 2.43) rooted stems within one year post treatment, an average net gain of 6.9% ( SE = 11.7%). In mechanically treated plots net density of C. brev ifolia decreased by 13.5 ( SE = 6.29) rooted stems within the same time period, an average net decrease of 45.5% ( SE = 13.6%) Despite these differences in direction of change, a verage percent n et change in C. brevifolia density did not differ significantly between treatments (n = 1 6, t = 2.24, p = 0.051; Figure 3 3, Table 3 4 ). In response to mowing, C. brevifolia resprouted from aboveground stems. Prior to mowing, 217 rooted stems were recorded in the mechanical treatment plots. At one year post treatment 109 rooted stems were recorded in the mechanical treatment plots. Of the 217 rooted stems recorded in mechanical treatment plots prior to treatment, 49.8% (108 rooted stems) were killed, 34.6% (75 rooted stems) resprouted, and 15.7% (34 rooted stems) had either recruited or were accidentally uncut by the treatments. Vegetation Cover Common vegetation within the plots included Pinus clausa ; oaks and woody species including: Quercus geminata, Q. inopina, Q. chapmanii, Q. myrtifolia, Lyonia ferruginea, Vacc inium myrsinities and S erenoa repens ; monocots including Aristida stricta var. beyrichiana, Andropogon virginicus and Bulbostylis spp.; mosses including Selaginella arenicola ; and ground lichens including Cladina and Cladonia spp.

PAGE 32

32 Vegetation cover pre t reatment varied among plots ranging from 50 to 95% cover with 85% of all plots exhibiting greater than 60% cover (Table 3 5 ) A verage pre treatment vegetation cover was 74.4% ( SE = 6.0%) in control plots and 80.0% ( SE = 4.2%) in mechanical treatment plots and was not significantly different between treatments (n = 17, t = 0.74, p = 0.473) Average percent vegetation cover in control plots at one year post treatment was 76.7% ( SE = 6.5%) representing an average increase of 2.2%. A verage percent vegetation cover in mechanical treatment plots at one year post treatment was 65% ( SE = 5.9%) representing an average decrease of 15.0% However, average percent vegetation cover was not significantly different between treatments one year post treatment (n = 17, t = 1.32, p = 0.208). By graphing the percent vegetation cover over time one can begin to understand how these systems respond to perturbation (Figure 3 4 ). Statistically, vegetation cover between treatments at one year post treatment was significantly differ ent. During the last three months (March, April, May) of the first year following treatment there was very little additional growth observed with the average vegetation cover increasing by < 2%. As indicated by evidence from C. brevifolia density counts pr esented earlier these mechanical treatments did not directly impact 100% of the area on a given plot. Therefore, if one assumes that mechanical treatment effect s on vegetation cover was similar to the impact on C. brevifolia (15.7% uncut or recruited) the n the average immediate post treatment vegetation cover on these plots was 12.6%. At nine months post treatment, average percent vegetation cover in mechanically treated plots was 63.8% ( SE = 3.5%) and was significantly different between treatments (n = 17, t = 2.29, p = 0.037) Bare Sand Cover In control plots, b are sand cover pre treatment ranged from 1 to 65% ; 66 7 % of plots exhibit ed less than 10% bare sand cover and result ed in an average d 19.6% ( SE = 7.8%) In mechanical treatment plots bare san d cover pre treatment ranged from 3 to 25% ; 75.0 %

PAGE 33

33 exhibit ed less than 10 % bare sand cover and result ed in an average of 10.4% ( SE = 2.8%) D ifferences in percent bare sand cover between treatments pre treatment were not significantly different (n = 17, t = 1.11, p = 0.295). At o ne year post treatment, bare sand cover averaged 17.6% ( SE = 6.6%) in control plots and 7.3 % ( SE = 2.0%) in mechanical plots. Statistically, differences in bare sand cover between control and mechanical treatments were not significan t (n = 17, t = 0.39, p = 0.702)

PAGE 34

34 Table 3 1. Percent mortality of Conradina brevifolia individuals at plot center (i.e., CPP) at one year post treatment Number of Rooted Stems Pre treatment Number of Rooted Stems Post treatment Percent Mortality Cont rol 9 8 11.1 Mechanical 8 5 37.5 Figure 3 1. A verage net change in height of Conradina brevifolia individuals at plot center (i.e., CPP) per treatment type

PAGE 35

35 Table 3 2 Net change in height of Conradina brevifolia individuals at plot center (i.e., CPP) at one year post treatment Height Pre Treatment (cm) Height Post Treatment (cm) Net Change in Height (cm)* Control Mean 8 5 4 91.4 6 Standard Error 1 1 .4 9.0 3.3 Minimum 24 43 7 Maximum 127 123 19 Count 8 8 8 Mechanical Mean 72.3 47.0 2 5.3 Standard Error 15.7 11.9 18.3 Minimum 40 18 73 Maximum 111 71 16 Count 4 4 4 A negative number indicates a loss; in control treatments, a negative number indicates natural dieback

PAGE 36

36 Figure 3 2. Growth rate of Conradina brevifolia at one year post treatment. Table 3 3 Flowering response of Conradina brevifolia individuals at plot center (i.e. CPP) pre treatment and nine month s post treatment Control Mechanical Flowering Response Pre T reatment (n = 9) Nine Month Post Treatment ( n = 9) Pre T reatment (n = 8) Nine Month Post Treatment (n = 5) Yes 9 (100%) 8 (89%) 8 (100%) 1 (20%) No 0 (0%) 1 (11%) 0 (0%) 4 (80%)

PAGE 37

37 Table 3 4 Conradina brevifolia density (i.e., number of rooted stems) per treatment type at one year post treatme nt Pre treatment Density One year Post treatment Density Net Change Density Percent Net Change Density Control Mean 23 24.6 1.6 106.9 Standard Error 8.0 9.5 2.4 11.7 Minimum 2 2 12 47.8 Maximum 73 85 12 155.6 Sum 184 197 13 Count 8 8 8 8 M echanical Mean 27.1 13.6 13.5 54.5 Standard Error 6.7 4.0 6.3 13.6 Minimum 6 0 53 0.0 Maximum 57 32 1 107.7 Sum 217 109 108 Count 8 8 8 8 *A negative number denotes a loss

PAGE 38

38 Figure 3 3. Average net change of Conradina brevifolia per treat ment type at one year post treatment

PAGE 39

39 Table 3 5. Net Change in percent vegetation cover at one year post treatment Pre Treatment One Year Post Treatment Net Change Control Mean 74.4 76.7 2.2 Standard Error 6.0 6.5 3.3 Minimum 50 45 5 Ma ximum 95 95 2 5 Count 9 9 9 Mechanical Mean 80 65 15 Standard Error 4.2 5.9 5.7 Minimum 65 45 35 Maximum 95 85 10 Count 8 8 8 *A negative value denotes a loss

PAGE 40

40 Figure 3 4. Change in percent vegetation cover over time following a mowing treat ment 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 Pre-Treatment At Treatment 9 Month Post-Treatment One Year Post-Treatment Percent Vegetation Cover Survey Event Control Mechanical

PAGE 41

41 Table 3 6 Net change in percent bare sand cover at o ne year post treatment Pre treatment Post treatment Net change* Control Mean 19.6 17.6 2 Standard Error 7.8 6.6 2.5 Minimum 1 0 15 Maximum 65 50 10 Count 9 9 9 Mechanical Mea n 10.4 7.3 3.1 Standard Error 2.8 2.0 1.2 Minimum 3 0 10 Maximum 25 15 0 Count 8 8 8 *A negative value denotes a loss

PAGE 42

42 CHAPTER 4 DISCUSSION Conradina brevifolia Responses Percent Mortality In control treatment plots, mortality due to natural causes was 11.1% whereas treatment induced mortality in mechanical treatment plots was 37.5%. Despite this difference, t he average CPP percent mortality did not differ between the treatments. This result supports the use of mowing since treatment induced mortali ty was not affected by the treatment However, land managers should use caution in areas where local abundance of C. brevifolia is low as these treatment induced mortality rates may not be ideal for all populations Height and Growth Rate Average net chang e in CPP height at one year post treatment was independent of the treatments Annual growth rate s of CPP individuals were significantly greater in mechanical treatments than in c ontrol treatments. This is important for land managers because th is suggest s t hat mowing influences C. brevifolia growth rate s allowing recovery within one year of treatment and increasing the chances of survival and post treatment reproduction. Flowering During the pre treatment surveys in March 2011, all CPP in both treatments we re flowering. Nine months post treatment, in February 2012, flowering response differed between the two treatments. In control plots, seven of eight surviving CPP were observed flowering whereas only one of five CPP in mechanical plots was in bloom. This s uggests that mechanical treatments may cause either a delay or absence of flowering post treatment. Treated plants may allocate stored resources to the regrowth of vegetation rather than expending it to producing flowers. However the added strain of droug ht on a plant already stressed by mechanical

PAGE 43

43 treatment may have influence d this response. Land managers should be cautious about applying fire within at least one year following treatment because plants have not had the opportunity to produce flowers and a re therefore, unable to increase the seedbank. Future monitoring studies should consider long term differences in flowering response and interactive effects of climate conditions. Density I found no significan t difference in net density of C brevifolia at the plot level between treatments despite the substantial difference between the means This suggests that recovery is within one year of mowing and therefore, mowing may be a min imally invasive treatment for restor ing the natural fire regime and maint aining the natural habitat when a sufficient number of individuals are present and burning alone is not an option However, variability was high among plots in mechanical treatments ; b ased on th e range of data for the mechanically treated plots 50% experi enced 30% or less reduction in density of C. brevifolia and only 25% of plots exper ienced a reduction greater than 90%. Based on plot level rooted stem counts from mechanically treated plots pre treatment, mortality was 49.8%. However, as previously menti oned above, CPP experienced 11.1% mortality (i.e., one individual) in control plots as a result of natural causes and 37.5% treatment induced mortality in mechanical treatment plots If the number of C. brevifolia killed based on density counts was adjuste d for natural mortality as was calculated from CPP percent mortality in control treatments then treatment induced mortality reduce s from 49.8% to 38.7% When comparing this adjusted rate to CPP treatment induced mortality in mechanical treatments, the dif ference is just 1.2% supporting that mortality rates of C. brevifolia may be less than 40.0%. However, as previously mentioned, land managers should use caution in areas where local

PAGE 44

44 abundance of C. brevifolia is low as these treatment induced mortality rat es may not be ideal for all populations. Resprouting Conradina brevifolia has been categorized as an obligate seeder following a fire event (USFWS, 1996; Slapcinsky et al., 2010); these authors assumed that it would be killed if cut (USFWS, 1996). However new evidence presented by this study shows that this species does have some potential to resprout from aboveground stems following m owing (Figure 4 1) recovering potentially through a combination of seeding and resprouting (Menges and Kohfeldt, 1995). Fo llowing a fire event, Gordon (1996) found that mortality rates for C. glabra were greater with the increase of tissue loss due to scorch. Mowing cut s vegetation above the ground su rface leaving some live tissue which may explain the ability for C. brevifol ia to resprout following treatment. Though some mortality is likely, some remaining plants, likely at least half, will either survive treatment or will resprout. This discovery is important as it may relieve apprehension about commencing restoration projec ts using mowing treatments when this endangered species is present. Response of C. brevifolia to fire was not studied, however, anecdotal observations during nine month post treatment surveys documented resprouting in areas of the LWRSF that experienced l ow intensity fire with no prior mechanical treatments; the resprouted individual continued to persist one year post treatment (Figure 4 2). However, in another area of the LWRSF that experienced mowing followed by a high intensity fire, all individuals wer e burned and no resprouting was observed within five months of the prescribed fire. This may indicate that either high intensity fire or the combination of mechanical treatment and fire may have contributed to the mortality of th ese plant s due to a substan tial amount of tissue loss. Similar results of fire treatment on C. glabra have been reported ( Gordon 19 96).

PAGE 45

45 During this study, C. brevifolia was commonly observed rooting from decumbent branches. One individual plant initially believed killed, was disco vered resprouting one year post treatment; branches of this individual also produced two separate ramets nearby and the parent plant may still be connected with these plants underground via a buried stem. Resprouting via decumbent branches is known to occu r for C. brevifolia (Gray, 1965) and has also been documented in the similar species Conradina verticillata (Roulston, 1994). Resprouting via decumbent branches could explain the tardiness of this resprout as these two ramets may be extracting resources th rough this stem and slowing resprout of the parent plant. Alternatively, the ramets may be distributing resources through the stem to the parent plant, allowing resprouting. These observations suggest that the percentage of resprouted plants may increase o ver time. Vegetation Cover There was no significant difference in average percent vegetation cover between mechanical and control treatment plots pre treatment and one year post treatment. This indicates that vegetation cover may recover within one year f ollowing mowing treatment. Based on previous studies, vegetation cover reaches pre treatment levels within five years using logging only mechanical treatments (Weekley et al., 2008b) and two to five years using mowing only mechanical treatments (Weekley et al., 2011); this study showed that vegetation cover reached close to pre treatment levels between nine months and one year and vegetation cover slows after nine months post treatment. However this may be attributed to the pre existing amount of vegetatio n cover among plots in both treatment types; vegetation cover ranged from 50 to 95% cover with 60% of plots exhibiting greater than 80% vegetation cover. The implicatio ns for land managers are that this rapid recovery of vegetation cover may hinder the abi lity to restore fire regimes or restore populations of C. brevifolia within one year of treatment and may indicate that fire should be applied quickly following treatment. Therefore, when using mowing as a pre fire

PAGE 46

46 treatment, fire should be applied within one year of treatment; however, as a fire surrogate, annual thinning treatments would need to be conducted to retain control of vegetation growth. Bare Sand Cover In p revious studies (e.g. Weekley et al., 2011) mechanical treatments have been shown to lead to litter and thatch accumulation reducing bare sand patches This study did not show a significant difference in the average net change of percent bare sand cover one year post treatment between treatments This may likely be attributed to the consid erable amoun t of pre existing litter cover among plots in both treatment types ; 7 6 % of the plots contained 10% or less cover However, repeated mechanical treatments could lead to litter and thatch accumulation over time (Rickey et al. 2007; Weekley et al. 2011) and may warrant additional research In a previous log and burn study C. brevifolia abundance increase d dramatically following logging treatments (Weekley et al., 2008b) Tree harvest can increase bare sand coverage from skid trails and heavy equ ipment and could produce conditions more conducive to C. brevifolia recruitment. The present study was not designed to test whether this species is in fact a gap specialist but rather to test the effect of this thinning technique on bare sand coverage H owever C. brevifolia was observed most frequently in areas of bare sand and was rarely found in areas of dense vegetation and litter. In addition, individuals occurring in dense cover were larger than those occurring in bare sand, perhaps due to the lengt h of time since establishment ; initial recruitment may have occurred when litter accumulation was minimal Further research is needed to determi ne the relationship between leaf litter accumulation and C. brevifolia recruitment Research Needs To date, no long term study has been conducted to understand the response of Florida scrub or C. brevifolia to long term mechanical treatments as fire surrogates; therefore, frequent

PAGE 47

47 monitoring of a variety of element s (e.g. plant responses, nutrient cycling, etc.) i s recommended (Menges and Gordon, 2010). Although logging techniques have been shown to increase C. brevifolia density (Weekley et al., 2008 b ) the nature of this response (i.e. resprouting vs. recruitment) is unknown and may warrant additional research C omparisons of mechanical and burn or burn only treatments to mechanical only treatments could provide invaluable insights for restoration efforts Incidental field observations indicate that C. brevifolia resprouted after a low intensity fire. However, re sprouting was not observed in areas where mechanical treatment was followed by a high intensity fire and aboveground tissue loss and damage was significant This suggests that resprouting may also vary with fire intensity. Knowledge of numerous biological and ecological traits of C. brevifolia and its associations with the surrounding environment is severely lacking; without an understanding of the general ecology and physiological requirements of this species, development of effective management strategie s is difficult. Furthermore, although this species is believed to be a gap specialist (Gray, 1965) evidence of this habitat requirement is still lacking. Until further studies are conducted and appropriate management techniques for C. brevifolia recovery and protection are in place, regular fire management strategies should be used when possible to best mimic the natural disturbance regimes.

PAGE 48

48 Figure 4 1 R esprouting Conradina brevifolia at one year post t reatment following mowing ( Photograph 2012 by Lynn Proenza )

PAGE 49

49 Figure 4 2 Conradina brevifolia resprouting following a low intensity fire. This picture was taken four months post burn ( Photograph 2012 by Lynn Proenza )

PAGE 50

50 CHAPTER 5 CONCLUSIONS AND MANA GEMENT IMPLICATIO NS Many remnants of the fr agmented Florida scrub on the Lake Wales Ridge have been affected both directly and indirectly by human activit ies like agriculture, development, and fire suppression. As a result many plants, including C brevifolia are listed as endangered by both state and federal governments. In an effort to conserve and restore populations of this native endemic scrub species, critical habitats must be protected and managed ( Gray, 1965; USFWS, 2008). Because little is known about the ecology and biology of C. brevifo lia species specific management techniques cannot be implemented. Low impact mowing treatments can help attain several goals in preserving and restoring C. brevifolia populations as some live tissue remains aboveground soil disturbance that may lead to i nvasions by non native species was minimal, and maintaining individual plants was possible These results demonstrate that at least half of the remaining plants present prior to treatment will likely remain uncut or will resprout following treatment and re sprouting may continue to occur more than a year following treatment. However, land managers should use caution to ensure that potential restoration sites contain an adequate population size and this species is in fact, locally abundant Furthermore, m owi ng reduce d vegetation cover and associated hazardous fire conditions making it a useful pre fire treatment for maintaining habitat between prescribed burns when fire is applied within one year of mowing. However, flowering for C. brevifolia is delayed or a bsent within the first year following a mowing treatment. If reintroduction of fire following treatment is the desired goal, caution should be taken to ensure sufficient time for C. brevifolia to flower and increase the seedbank. Fire has been repeatedly s hown as an important aspect of Florida scrub. Based on this study, t he use of mowing as a fire surrogate is not recommended since vegetation can recover quickly and

PAGE 51

51 l itter may accumulate thus reducing bare sand cover age and resulting in the loss of habita t for a potential gap specialist

PAGE 52

52 LIST OF REFERENCES A BRAHAMSON W.G 1984b. Post fire recovery of Florida Lake Wales Ridge vegetation. American Journal of Botany. 71: 9 21. ________________. 1984c. Species responses to fire on Florida Lake Wales Ridge American Journal of Botany. 71:35 43. ________________., A. F. J OHNSON J. N. L AYNE AND P. A. P ERONI 1984a. Vegetation of the Archbold Biological Station, Florida: An example of the southern Lake Wales Ridge. Florida Scientist. 47:209 250. ____________ ____., AND C. R. A BRAHAMSON 1996. Effects of fire on long unburned Florida uplands. Journal of Vegetation Science 7:565 574. B ROWN J. K. 1983 128. In: Proceedings of a symposium and workshop on wilderness fire General technical report INT 182. U.S. Forest Service, Ogden, Utah. C HRISTMAN S. P. AND W. S. J UDD 1990. Notes on plants endemic to Florida scrub. Biological Sciences. 53:52 73. C LANTON K. 2010. FDACS, Florida Forest Service, Lake Wales Ridge State Fo rest, Frostproof, FL Pers. Comm. D OBSON A. P., J. P. R ODRIGUEZ W. M. R OBERTS AND D. S. W ILCOVE 1997. Geographic distribution of endangered species in the United States. Science 275:550 553 D ODGE M 1972. Forest fuel accumulation: A growing problem. Sc ience. 177:139 142. D UNCAN B. W. AND P. A. S CHMALZER 2004. Anthropogenic influences on potential fire spread in a pyrogenic ecosystem of Florida, USA. Landscape Ecology 19:153 165 FDACS, FFS. 2006. Ten year resource management plan for the Lake Wales Rid ge State Forest, Polk County. Available online: http://www.floridaforestservice.com/ state_forests/management_current_plans.html#reg3 Date Accessed: O ctober 22, 2012. F ERNANDES P. M. AND H. S. B OTELHO 2003. A review of prescribed burning effectiveness in fire hazard reduction. International Journal of Wildland Fire 12:117 128 F LORIDA N ATURAL A REAS I NVENTORY (FNAI) 2010. Guide to the natural communiti es of Florida: 2010 edition. Florida Natural Areas Inventory, Tallahassee, FL. G IVENS K. T., J. N. L AYNE W. G. A BRAHAMSON S. C. W HITE S CHULER 1984. Structural changes and successional relationships of five Florida Lake Wales Ridge plant communities. Bu lletin of the Torrey Botanical Club Vol 111, No. 1, pp. 8 18.

PAGE 53

53 G ORDON D. R. 1996 Experimental translocation of the endangered shrub Apalachicola rosemary Conradina glabra to the Apalachicola Bluffs and Ravines Preserve, Florida. Biological Conservation 7 7:19 26. G RAY T. C. 1965. A monograph of the genus Conradina A. Gray (Labiatae). Ph.D. Thesis. Vanderbilt University, Nashville, TN. G REENBERG C. H., D. G. N EARY L. D. H ARRIS S. P. L INDA 1995. Vegetation recovery following high intensity wildfire and silvicultural treatments in sand pine scrub. 1995. American Midland Naturalist 1133:149 163. G UERIN D. N. 1993 Oak dome structure and fire ecology in a Florida longleaf pine dominated community. Bulletin of the Torrey Botanical Club 120:107 114. H AINES T. K. AND R. L. B USBY 2001. Prescribed burning in the south: Trends, purpose, and barriers. South. J. Appl. For. 25:149 153. H ALL J. M., T. W. G ILLESPIE D. R ICHARDSON S. R EADER 2002. Fragmentation of Florida scrub in an urban landscape. Urban Ecosys tems 6:243 255. H AWKES C. V. AND E. S. M ENGES 1996. The relationship between open space and fire for species in a xeric Florida shrubland. Bull. Torrey Bot. Club 123:81 92. ____________. AND E. S. M ENGES 2003. Effects of lichens on seedling emergence in a xeric Florida shrubland. Southeastern Naturalist. 2:223 234. J OHNSON A. F., W. G. A BRAHAMSON AND K. D. M C C REA 1986. Comparison of biomass recovery after fire of a seeder ( Ceratiola ericoides ) and a sprouter ( Quercus inopina ) species from south cent ral Florida. American Midland Naturalist. 116: 423 428. K ALABOKIDIS K. D., AND P. N. O MI 1998. Reduction of fire hazard through thinning/residue disposal in the urban interface. Int. J. Wildland Fire 8:29 35. K RAL R. 1983. A report on some rare, threa tened, or endangered forest related vascular plants of the South. United States Department of Agriculture, Forest Service, Southern Region, Atlanta, GA. L ONG A. J., D. D. W ADE AND F. C. B EALL 2004. Managing for fire in the interface: Challenges and oppo rtunities. In : S. W. Vince, M. L. Duryea, E. A. Macie, L. A. Hermansen (eds.) Forests at the Wildland Urban Interface. Boca Raton, FL: CRC Press. M ENGES E. S 1992. Habitat preferences and response to disturbance for Dicerandra frutescens, a Lake Wales R idge (Florida) Endemic Plant. Bulletin of the Torrey Botanical Club 119:308 313. ___________., W. G. A BRAHAMSON K. T. G IVENS N. P. G ALLO J. N. L AYNE 1993. Twenty years of vegetation change in five long unburned Florida plant communities. Journal of Veg etation Science. 4:375 386.

PAGE 54

54 ___________. AND C. V. H AWKES 1998. Interactive effects of fire and microhabitat on plants of Florida scrub. Ecological Applications 8: 935 946. __________. AND N. K OHFELDT 1995. Life history strategies of Florida scrub plan ts in relation to fire. Bulletin of the Torrey Botanical Club 122:282 297. ___________ 1999. Ecology and conservation of Florida scrub. Pp. 7 22 In: A NDERSON R. C., J. S. F RALISH AND J. M. B ASKIN (eds), Savannas, barrens, and rock outcrop plant c ommunities of North America. Cambridge University Press, UK. ___________ 2001. Comparative ecology of Florida scrub plants. In : Zattau, D. P., Proceedings of the Florida Scrub Symposium 2001. Orlando, Florida. U.S. Fish and Wildlife Service. Jacksonville FL. ___________. 2007. Integrating demography and fire management: an example from Florida scrub. Australian Journal of Botany 55: 261 72. ___________., A. C RADDOCK J. S ALO R. Z INTHEFER AND C. W. W EEKLEY 2008. Gap ecology in Florida scrub: Species occurrence, diversity, and gap properties. Journal of Vegetation Science 19:503 514. ___________. AND D. R. G ORDON 2010. Should mechanical treatments and herbicides be used Florida Scientist 73:147 174. M YERS R. L. 1985 Fire and the dynamic relationship between Florida sandhill and sand pine scrub vegetation. Bulletin of the Torrey Botanical Club 112:241 252. __________ 1990. Scrub and high pine. Pp. 150 193. In : Myers, R.L. and J.J. Ewel (eds.), Ecosystems of Florida. University of Central Florida Press, Orlando, Fl. NOAA NATIONAL CLIMATIC DA TA CENTER 2012. Summary of monthly normals, 1981 2012. Data received July 3, 2012. N OSS R.F., T.E. L A R OE III, AND J.M. S COTT 1995. Endangered ecosystem s of the United States: A preliminary assessment of loss and degradation. National Biological Service Biol. Report 128:58 pp. P ETRU M., E. S. M ENGES 2003. Seedling establishment in natural and experimental Florida scrub gaps. Journal of the Torrey Botan ical Society 130:89 100. Q UINTANA A SCENCIO P.F. AND M. M ORALES H ERNNDEZ 1997. Fire mediated effects of shrubs, lichens and herbs on the demography of Hypericum cumulicola in patchy Florida scrub. Oecologia 112: 267 271. __________________., R.W. D OLA N AND E.S. M ENGES 1998. Hypericum cumulicola demography in unoccupied and occupied Florida scrub patches with different time since fire. Journal of Ecology 86: 8640 651.

PAGE 55

55 R ICKEY M. A., E. S. M ENGES AND C. W. W EEKLEY 2007. Effects of mechanical treat ments and fire on litter reduction in Florida scrub and sandhill. Pp. 104 108. In: R. E. Masters and K. E. M. Galley (eds.), Proceedings of the 23 rd Tall Timbers Fire Ecology Conference: Fire in Grassland and Shrubland Ecosystems. Tall Timbers Research Sta tion, Tallahassee, Florida, USA. R OULSTON T. H. 1994. Reproductive ecology of Conradina verticillata Jennison, a rare, Knoxville, TN. S CHMALZER P. A., D. R. B REININGER F. W. A DRIAN R. S CHAUB B. W. D UNCAN 1994. Development and implementation of a scrub habitat compensation plan for Kennedy Space Center. NASA Technical Memorandum 109202. John F. Kennedy Space Center, Florida. ______________., T. E. F OSTER AND F. W. A DRIAN 2003. Responses of long unburned scrub on the Merritt Island/Cape Canaveral barrier island complex to cutting and burning. In : Proceeding of the Second International Wildland Fire Ecology and Fire Management Congress. S HINNERS L. H 1962. Synopsis of Co nradina (Labiatae). Sida 1:84 88. S LAPCINSKY J. L., D. R. G ORDON E. S. M ENGES 2010. Responses of rare plant species to fire in Natural Areas Journal. 30:4 19 S OIL C ONSERVATION S ERVICE 1990. Soil survey of Polk County, F lorida. Soil Conservation Service. S TEPHENS S. L. AND L. W. R UTH 2005. Federal forest fire policy in the United States. Ecological Applications. 15:532 542. T URNER W. R., D. S. W ILCOVE AND H. M. S WAIN 2006. State of the scrub: Conservation progress management responsibilities, and land acquisition priorities for imperiled species of Florida's Lake Wales Ridge. USFWS. 1996. Recovery Plan: Nineteen central Florida scrub and high pineland plant species. U.S. Fish and Wildlife Service. Atlanta, Georgi a. _______ 1993. Endangered or threatened status for five Florida plants. Final Rule. Federal Register 58 FR:37432 37443. _______ 2008. Short leaved rosemary ( Conradina brevifolia ) 5 year review: Summary and evaluation. U.S. Fish and Wildlife Service. V ero Beach, Florida. V ENO P. A. 1976. Successional relationships of five Florida plant communities. Ecology. 57:498 508. W ALLY A. L., E. S. M ENGES AND C. W. W EEKLEY 2006. Comparison of three devices for estimating fire temperatures in ecological studies Applied Vegetation Science 9:97 108.

PAGE 56

56 W EEKLEY C. W., AND E. S. M ENGES 2003. Species and vegetation responses to prescribed fire in a long unburned, endemic rich Lake Wales Ridge scrub. Journal of the Torrey Botanical Society 130:265 282. ______________ _., E. S. MENGES, AND R. L. PICKERT. 2008a. An ecological map of Columbian habitat loss. Florida Scientist 71:45 64 _____________., E. S. M ENGES M. A. R ICKEY G. L. C LARKE S S MITH 2008b. Effects of mechanical treatments and fire on Florida scrub vegetation, Final Report to U.S. Fish and Wildlife Service, Vero Beach Office, Vero Beach, Fl. _____________., E. S. M ENGES D. B ERRY G REENLEE M. A. R ICKEY G. L. C LARKE AND S. A. S MITH 2011. Burning more effective than mowing in restoring Florida scrub. Ecological Restoration. 29:357 373. W ELCH B. L variances are involved. Biometrika 34:28 35. W IL LIGES K., J. B AKER N. G OODHOPE T. S EMONES A. T ORAL AND A W AGNER 2006. Effects of Management Regimes on Successionally Advanced Scrub Habitat. Annual Report. Fish and Wildlife Research Institute, Florida Fish and Wildlife Conservation Commission, Tall ahassee, Florida, USA.

PAGE 57

57 BIOGRAPHICAL SKETCH Lynn Proenza was raised in Orlando, Florida where she spent much of her time outdoors. After graduating high school, she attended college seeking her s ign l anguage i nterpretation degree, but then quickly mov ed to the field of natural resources. She graduated with her Bachelor of Science degree in e nvironmental s cience from the University of South Florida and began working in the environmental field gaining experience in consulting and natural resource land management. After eight years with experience and hands on learning, she returned to school to seek her Mast er of Science degree at the University of Florida, School of Forest Resources and Conservation with hopes to work in the field of ecological restoration and land management