<%BANNER%>

Imperata cylindrica Invasion in Juvenile Pinus taeda Forests

Permanent Link: http://ufdc.ufl.edu/UFE0021118/00001

Material Information

Title: Imperata cylindrica Invasion in Juvenile Pinus taeda Forests Implications of Diversity and Impacts on Productivity and Nitrogen Dynamics
Physical Description: 1 online resource (135 p.)
Language: english
Creator: Daneshgar, Pedram Patrick
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: cogongrass, cylindrica, diversity, imperata, invasive, loblolly, nitrogen, pinus, productivity, taeda
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Imperata cylindrica, a C4 rhizomatous perennial grass that invades a range of sites, is one of the most troublesome weed species in the world. Little information exists on its ecological impacts. The objective of this study was to examine the impacts of I. cylindrica on southeastern forests, specifically examining the role of species identity in community susceptibility to invasion and how invasion affects productivity and nutrient pools of establishing pine forests. The role of species richness, functional diversity, and species identity of native species were tested with I. cylindrica in mesocosms. Results showed negative relationships between the biomass and cover of the native species and I. cylindrica. Grasses proved to be the most resistant functional group. Andropogon virginicus was the most resistant to invasion over time, suggesting that resistance is a matter of species identity. The success of A. virginicus can be attributed to the fact that it had significantly greater root length, root length density, and specific root length than all of the native species and I. cylindrica. The root morphology characteristics allow it to be a great competitor belowground where I. cylindrica is most aggressive. The impacts of I. cylindrica and native vegetation competition on the productivity of loblolly pine (Pinus taeda) seedlings were evaluated in a 27-month long field study. At the end of the study, only 26% of the seedlings growing in I. cylindirca survived, half of what was observed with native species. After two growing seasons, the root collar diameter, height and stem volume index of pine seedlings growing with the invasive were significantly lower than seedlings competing with native vegetation. After 27 months, the pine seedling biomass in I. cylindrica was 18% of the seedlings grown with native vegetation. The I. cylindrica pine seedlings maintained the lowest levels of light saturated photosynthesis with reduced levels of stomatal conductance. Low foliar nitrogen may explain the reduced photosynthesis. Evidence from this study suggests that I. cylindrica competition significantly reduces the productivity and growth of loblolly pine seedlings compared to native vegetation. In the next study, 15N-labeled ammonium sulfate was used to compare how P. taeda seedlings compete for N in the presence of I. cylindrica and native vegetation competition. I. cylindrica competition led to significantly lower N content in the pine foliage and roots than the native treatment. Competition from I. cylindrica contributed to the pine seedlings taking up a greater percentage of the applied fertilizer than the seedlings competing with native vegetation. The belowground biomass of I. cylindrica was seven times higher than that of the native species despite its lower N concentration. While the native species retained more N aboveground, I. cylindrica held significantly more N belowground. Invasion by this grass could lead to a shift in N pools from above to belowground in infested ecosystems. Overall, the results indicate that communities with certain species such as A. vriginicus may better resist the invasion of I. cylindrica. However, once infested, I. cylindrica decreases productivity and alters nutrient pools. This can have serious negative implications for the health and integrity of infested southeastern forests.
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 Pedram Patrick Daneshgar.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Jose, Shibu.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021118:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021118/00001

Material Information

Title: Imperata cylindrica Invasion in Juvenile Pinus taeda Forests Implications of Diversity and Impacts on Productivity and Nitrogen Dynamics
Physical Description: 1 online resource (135 p.)
Language: english
Creator: Daneshgar, Pedram Patrick
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: cogongrass, cylindrica, diversity, imperata, invasive, loblolly, nitrogen, pinus, productivity, taeda
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Imperata cylindrica, a C4 rhizomatous perennial grass that invades a range of sites, is one of the most troublesome weed species in the world. Little information exists on its ecological impacts. The objective of this study was to examine the impacts of I. cylindrica on southeastern forests, specifically examining the role of species identity in community susceptibility to invasion and how invasion affects productivity and nutrient pools of establishing pine forests. The role of species richness, functional diversity, and species identity of native species were tested with I. cylindrica in mesocosms. Results showed negative relationships between the biomass and cover of the native species and I. cylindrica. Grasses proved to be the most resistant functional group. Andropogon virginicus was the most resistant to invasion over time, suggesting that resistance is a matter of species identity. The success of A. virginicus can be attributed to the fact that it had significantly greater root length, root length density, and specific root length than all of the native species and I. cylindrica. The root morphology characteristics allow it to be a great competitor belowground where I. cylindrica is most aggressive. The impacts of I. cylindrica and native vegetation competition on the productivity of loblolly pine (Pinus taeda) seedlings were evaluated in a 27-month long field study. At the end of the study, only 26% of the seedlings growing in I. cylindirca survived, half of what was observed with native species. After two growing seasons, the root collar diameter, height and stem volume index of pine seedlings growing with the invasive were significantly lower than seedlings competing with native vegetation. After 27 months, the pine seedling biomass in I. cylindrica was 18% of the seedlings grown with native vegetation. The I. cylindrica pine seedlings maintained the lowest levels of light saturated photosynthesis with reduced levels of stomatal conductance. Low foliar nitrogen may explain the reduced photosynthesis. Evidence from this study suggests that I. cylindrica competition significantly reduces the productivity and growth of loblolly pine seedlings compared to native vegetation. In the next study, 15N-labeled ammonium sulfate was used to compare how P. taeda seedlings compete for N in the presence of I. cylindrica and native vegetation competition. I. cylindrica competition led to significantly lower N content in the pine foliage and roots than the native treatment. Competition from I. cylindrica contributed to the pine seedlings taking up a greater percentage of the applied fertilizer than the seedlings competing with native vegetation. The belowground biomass of I. cylindrica was seven times higher than that of the native species despite its lower N concentration. While the native species retained more N aboveground, I. cylindrica held significantly more N belowground. Invasion by this grass could lead to a shift in N pools from above to belowground in infested ecosystems. Overall, the results indicate that communities with certain species such as A. vriginicus may better resist the invasion of I. cylindrica. However, once infested, I. cylindrica decreases productivity and alters nutrient pools. This can have serious negative implications for the health and integrity of infested southeastern forests.
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 Pedram Patrick Daneshgar.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Jose, Shibu.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021118:00001


This item has the following downloads:


Full Text

PAGE 1

1 Imperata cylindrica INVASION IN JUVENILE Pinus taeda FORESTS: IMPLICATIONS OF DIVERSITY AND IMPA CTS ON PRODUCTIVITY AND NITROGEN DYNAMICS By PEDRAM PATRICK DANESHGAR A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Pedram Patrick Daneshgar

PAGE 3

3 To Farhad and Mina Daneshgar

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my gr aduate committee chair, Shibu Jose, for giving me an opportunity to pursue a doctoral degree and for his support, encouragement, suggestions, and guidance with my dissertation research and wri ting. I would also like to thank my committee members (Dr. Eric Jokela, Dr. Michelle Mack, Dr. Tim Martin, and Dr. Rick Williams) for their contributions to my research and for allowing me to pursue my degree even after some bumps in the road. I am grateful for the work done by Dr Craig Ramsey, who helped establish several of my projects; and to Robin, Leah, Jeremy, Jeff, Mike Don and Eric for helping with field work. I would like to thank all my fr iends and family for their neve r-ending support and love. I extend big thanks to my sister Shireen for her ea rs and her heart. Lastly, I would like to thank my mother Mina and my father Farhad for thei r unconditional love and support, for picking me up when I was down, and for letting me pursue my dream.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION................................................................................................................. .12 Review of Literature........................................................................................................... ....13 Biology of Imperata cylindrica........................................................................................13 Impacts of Imperata cylindrica .......................................................................................14 Community Susceptibility to Invasion............................................................................15 Invasive Impacts on Resources of Communities.............................................................17 Specific Objectives............................................................................................................ .....18 2 MECHANISMS FOR PLANT INVASION: A REVIEW.....................................................19 Biotic Resistance Hypothesis.................................................................................................20 Fluctuating Resource Availabil ity Theory of Invasibility......................................................21 Empty Niche Hypothesis........................................................................................................23 Diversity-Invasibility Hypothesis...........................................................................................24 Facilitation by Soil Biota..................................................................................................... ...27 Invasion Meltdown Hypothesis..............................................................................................29 Natural Enemies Hypothesis...................................................................................................30 Evolution of Increa sed Compe titive Ability...........................................................................31 Reproductive Traits............................................................................................................ ....32 Superior Competitor............................................................................................................ ...33 Novel Weapons Hypothesis....................................................................................................34 Integrated Mechanisms.......................................................................................................... .35 Conclusions.................................................................................................................... .........36 3 ROLE OF SPECIES IDENTITY IN PL ANT INVASIONS: EXPERIMENTAL TEST USING Imperata cylindrica ...................................................................................................42 Introduction................................................................................................................... ..........42 Methods........................................................................................................................ ..........46 Experimental Design and Study Site...............................................................................46 Data Collection................................................................................................................48 Statistical Analyses..........................................................................................................48 Results........................................................................................................................ .............49

PAGE 6

6 Discussion..................................................................................................................... ..........52 Conclusions.................................................................................................................... .........59 4 IMPACTS OF Imperata cylindrica AN ALIEN INVASIVE GRASS, ON THE PRODUCTIVITY OF AN ES TABLISHING PINE FOREST..............................................70 Introduction................................................................................................................... ..........70 Methods........................................................................................................................ ..........73 Site Description...............................................................................................................73 Cultural Treatments.........................................................................................................74 Growth and Gas Exchange Measurements......................................................................75 Water Potential Measurements........................................................................................76 Statistical Analysis..........................................................................................................76 Results........................................................................................................................ .............76 Survival and Growth........................................................................................................76 Above and Belowground Biomass..................................................................................77 Gas Exchange and Leaf Characteristics..........................................................................77 Discussion..................................................................................................................... ..........78 Conclusions.................................................................................................................... .........82 5 Imperata cylindrica AN ALIEN INVASIVE GRASS, MAINTAINS CONTROL OVER N AVAILABILITY IN AN ESTABLISHING PINE FOREST.................................91 Introduction................................................................................................................... ..........91 Methods........................................................................................................................ ..........94 Site Description and Experimental Design......................................................................94 Fertilizer Application.......................................................................................................95 Sampling Methods...........................................................................................................96 Statistical Analyses..........................................................................................................97 Results........................................................................................................................ .............98 Discussion..................................................................................................................... ........100 6 SUMMARY AND CONCLUSIONS...................................................................................117 APPENDIX MESOCOSM STUDY EXTRAS...........................................................................122 LIST OF REFERENCES.............................................................................................................123 BIOGRAPHICAL SKETCH.......................................................................................................135

PAGE 7

7 LIST OF TABLES Table page 3-1 Summary of the te n treatments used..................................................................................60 3-2 Analysis of variance of measurements taken at the end of the study between all treatments of by functional group......................................................................................61 3-3 Summary of the I. cylindrica shoots, cover and biomass in the functional group treatments as well as the reduction of these values from the control.................................62 3-4 Summary of root length, r oot length density and specifi c root length means(SE) for the native species and Imperata cylindrica in a) monoculture and in b) mixed communities.................................................................................................................... ...63 3-5 Percentage of total roots accounted for each species by depth in the soil profile.............64 4-1 Percent survival (%) of total loblolly pi nes seedlings planted in each treatments in March 2003..................................................................................................................... ...83 4-2 Mean (SE) leaf area, specific leaf ar ea (SLA) and % foliar nitrogen of 9-month-old loblolly pine seedlings for different treatments.................................................................84 5-1 Summary of competing vegetati on means (SE) for the native and I. cylindrica treatments..................................................................................................................... ....107 5-2 Mean nitrogen content (kg/ ha) in pine foliage, stem, and roots as well as foliage and roots of competing vegetation..........................................................................................108 5-3 Mean percentage of N derived from fe rtilizer (%NDF) for pine foliage, stem, and roots as well as foliage and roots of competing vegetation.............................................109 5-4 Mean percentage utilization of fertilizer N (% UFN) for pine foliage, stem, and roots as well as foliage and root s of competing vegetation......................................................110 A-1 Final mean % cover for each species in the ten treatments.............................................122

PAGE 8

8 LIST OF FIGURES Figure page 2-1 The balance between gross resource s upply and resource uptak e (as denoted by the isocline, where gross resource supply = res ource uptake) represents a communitys barrier to invasion............................................................................................................ ..38 2-2 Survival of Clidemia hirta in A) Costa Rica (where its native) and B) Hawaii (introduced range) in four natural enemy escape treatments in understory and open habitats....................................................................................................................... ........39 2-3 Mean (+ SE) dry biomass and height of purple loosestrife from Ithaca, New York, USA (Black) and Lucelle, Switzerland (gre y) after the growing season grown in common gardens under identical conditions......................................................................40 2-4 Biomass of roots, shoots, and total of Centaurea and Festuca grown together in grown in pots with and without activated carbon, which has a high affinity for organic chemicals.............................................................................................................. .41 3-1 Relationship between % cover of native understory species treatment means and mean % cover of Imperata cylindrica. ..............................................................................65 3-2 Relationship between native treatment total biomass means and biomass of Imperata cylindrica. ..........................................................................................................................66 3-3 Relationship between species richne ss as represented by species number and Imperata cylindrica biomass..............................................................................................67 3-4 Percent cover of Imperata cylindrica in each mesocosm for the ten treatments in the summer of 2005.................................................................................................................68 3-5 Above and belowground biomass means of native species in all the grass containing treatments..................................................................................................................... ......69 4-1 Pine seedling root collar diameter ( RCD), height, and stem volume index (SVI) means(SE) for the three treatments from planting through the end of the study...............85 4-2 Mean biomass(SE) of the pine seedli ng foliage, stems, and roots for the three treatments after one growing season ( 2003) and three growing seasons (2005)...............86 4-3 Mean light saturated photosynthesis (Amax), stomatal conductance (gs), and internal leaf CO2 concentration (Ci) summer means(SE) for pine seedlings in the three treatments..................................................................................................................... ......87 4-4 Monthly light saturated photosynthesis (Amax) of the pine seedlings in the VF, NC, and IC treatments.............................................................................................................. .88

PAGE 9

9 4-5 Soil water potential of the three treatments at two soil depths (30 and 60 cm).................89 4-6 Relationship between leaf nitrogen concentration and light saturated net photosynthesis (Amax).........................................................................................................90 5-1 Summary of mean monthly temperature (bars) and total monthly precipitation (line) for the study site in 2003..................................................................................................111 5-2 Amount of nitrogen in tissues (% of total biomass)(SE) of P. taeda seedlings and competing vegetation for the three treatments.................................................................112 5-3 Percentage of 15N recovery in soil at end of growi ng season in the VF (triangles), NC (squares), and IC (diamonds) treatments.........................................................................113 5-4 Percent total of N recovered at th e end of study for th e three treatments........................114 5-5 Monthly mean soil moisture of the th ree treatments for the summer of 2003.................115 5-6 Demonstration of the percent of total nitrogen occurring in a young emerging pine forest before and after invasion by I. cylindrica ..............................................................116

PAGE 10

10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Imperata cylindrica INVASION IN JUVENILE Pinus taeda FORESTS: IMPLICATIONS OF DIVERSITY AND IMPA CTS ON PRODUCTIVITY AND NITROGEN DYNAMICS By Pedram Patrick Daneshgar August 2007 Chair: Shibu Jose Major: Forest Resources and Conservation Imperata cylindrica a C4 rhizomatous perennial grass that invades a range of sites, is one of the most troublesome weed species in the worl d. Little information exists on its ecological impacts. The objective of this st udy was to examine the impacts of I. cylindrica on southeastern forests, specifically examining the role of sp ecies identity in comm unity susceptibility to invasion and how invasion affect s productivity and nutri ent pools of establishing pine forests. The role of species richness, functional dive rsity, and species identity of native species were tested with I. cylindrica in mesocosms. Results show ed negative relationships between the biomass and cover of the native species and I. cylindrica Grasses proved to be the most resistant functional group. Andropogon virginicus was the most resistant to invasion over time, suggesting that resistance is a matter of species identity. The success of A. virginicus can be attributed to the fact that it ha d significantly greater root lengt h, root length density, and specific root length than all of the native species and I. cylindrica The root morphology characteristics allow it to be a great competitor belowground where I. cylindrica is most aggressive. The impacts of I. cylindrica and native vegetation competition on the productivity of loblolly pine ( Pinus taeda ) seedlings were evaluated in a 27month long field study. At the end

PAGE 11

11 of the study, only 26% of the seedlings growing in I. cylindirca survived, half of what was observed with native species. After two growi ng seasons, the root colla r diameter, height and stem volume index of pine seedlings growing with the invasive were significantly lower than seedlings competing with native vegetation. Af ter 27 months, the pine seedling biomass in I. cylindrica was 18% of the seedlings grown with native vegetation. The I. cylindrica pine seedlings maintained the lowest levels of light saturated photosynthesis with reduced levels of stomatal conductance. Low foliar nitrogen ma y explain the reduced photosynthesis. Evidence from this study suggests that I. cylindrica competition significantly reduces the productivity and growth of loblolly pine seedlings compared to native vegetation. In the next study, 15N-labeled ammonium sulfate was used to compare how P. taeda seedlings compete for N in the presence of I. cylindrica and native vegetation competition. I. cylindrica competition led to significantly lower N cont ent in the pine foliage and roots than the native treatment. Competition from I. cylindrica contributed to the pine seedlings taking up a greater percentage of the applied fertilizer than the seedlings co mpeting with native vegetation. The belowground biomass of I. cylindrica was seven times higher than that of the native species despite its lower N concentra tion. While the native species retained more N aboveground, I. cylindrica held significantly more N belowground. Inva sion by this grass could lead to a shift in N pools from above to belowground in infested ecosystems. Overall, the results indicate that co mmunities with certain species such as A. vriginicus may better resist the invasion of I. cylindrica However, once infested, I. cylindrica decreases productivity and alters nutrient pools. This can have serious negative imp lications for the health and integrity of infested southeastern forests.

PAGE 12

12 CHAPTER 1 INTRODUCTION In 1958, Charles Elton, a pioneer in popul ation ecology, spoke of how ecological explosions were threatening the world. What he meant by ecological explosions was that living organisms were experiencing enormous increases in number. Today, we refer to species having explosive growth and spread outside their native range as invasive specie s. Biological invasions have caused more species extinctions than human induced climate change (DAntonio and Vitousek 1992) and are the second leading cause of species extincti ons after changes in habitat. Invasive plants, in particular, are to blame for much native species decline and ecosystem degradation (Wilcove et al. 1998). The invasion of native ecosystems by exotic plants can lead to alterations in nutrient cy cling, fire regime, hydrology, ener gy budgets, and native species abundance and survival (Mack et al. 2000). With few limits on range and distribution, co mbined with relatively easy transport of propagules, grasses in aggregate ma y be one of the most widespread invading plant groups in the world with serious implications for ecosystem function (DAntonio and V itousek 1992). Of the invasive grasses, Imperata cylindrica, a rhizomatous C4 perennial, is particularly troublesome. It has been spreading throughout natural and disturbed ecosystems all over the world, infesting over 500 million hectares and is considered a pe st in over 73 countries (MacDonald 2004). In the southeastern United States, I. cylindrica poses a serious threat to native ecosystems. Over 500,000 hectares are infested by the exotic in Florida alone (MacDonald 2004). I. cylindrica is also becoming a problem in establishing pine fore sts in the Southeast, thus becoming a dilemma for forest land managers (Jos e et al. 2002; Miller, 2000). To date, no studies have been condu cted to examine the impacts of I. cylindrica on establishing forests, which became the primary goal of this research. Our approach to

PAGE 13

13 understanding these impacts began with a review of all the curre nt theories on invasion success, particularly examining the factors that make co mmunities susceptible to invasion and also traits of successful invaders. To follow, we explored what characteristics of an establishing pine forest, its understory diversity in particular, in fluence the communitys su sceptibility to invasion. Next, we asked the question, what are the impacts of I cylindrica competition on young pine seedlings after a successful invasion and how do these impacts compare to competition from native vegetation? Lastly, with evidence that I. cylindrica competition reduced pine seedling physiology and growth, we used 15N enriched fertilizer to determine if competition for nitrogen explained the reduced pine seedling productivity. Review of Literature Biology of Imperata cylindrica Imperata cylindrica is a C4 grass that is found mainly in tropical and subtropical regions with annual rainfall between 75 and 500 cm (Bryson and Carter 1993). I. cylindrica is generally tolerant of variable so il conditions, but grows most favorably in acidic soil (pH 4.7) (Wilcut et al. 1988). Because of its C4 photosynthetic pathway, I. cylindrica is able to thrive in environments with high temperatures and light intensit y. In areas with low water supplies, C4 plants have higher water use efficiencies. I. cylindrica C4 photosynthetic pathway a llows it to be a great competitor with C3 plants. Imperata cylindrica has basal leaf blades that grow up to 1.5 m tall and 2 cm wide in conditions of good soil moisture and fertility (L ippincott 1997) with a noticeably off-center whitish mid-vein. The serrated leaf margins of I. cylindrica accumulate silicates, which deter herbivory (Dozier et al. 1998). Es timates of leaf biomass report 10 Mg per hectare consisting of 4.5 million shoots (Soerjani 1970).

PAGE 14

14 Imperata cylindrica considered a r-strategi st, is a prolific seed er producing as many as 3000 seeds per plant (Holm et al. 1977). Flowering and seed production of I. cylindrica in the U.S. occurs in late winter/early spri ng (Shilling et al. 1997; Willard 1998). I. cylindrica seeds are dispersed long distances ranging from 15m to 100m (Shilling et al. 1997). The viability of I. cylindrica seeds is highest within the first three m onths and the seeds do not have a dormancy period (Shilling et al. 1997). Because seeds complete ly lose viability after one year (Shilling et al. 1997), it is questioned as to whether or not they are the primary mechanism for spread. I. cylindrica can also reproduce asexually with its rhiz omes, which comprise greater than 60% of its biomass (Sajise 1976) giving I. cylindrica a low shoot/root ratio (B rook 1989). Terry et al. (1997) report fresh rhizome biomass in m onoculture to be 40 Mg per hectare. The regenerative capacity of rhizomes increases with maturity, allowing I. cylindrica to regenerate from fragments as small as 0.1g (Aye ni and Duke 1985). Resistant to heat and breakage, rhizomes occur in th e top 0.15 m in heavy clay soils and 0.1 m of sandy soils, but may penetrate the soil more than a meter deep (Holm et al. 1977). The rhizomes are long and tough with short internodes, which lead to the forma tion of dense mats just below the soil surface. Impacts of Imperata cylindrica The impacts of I. cylindrica in the Southeast have only re cently been examined. Several have reported evidence that I. cylindrica penetrates through the roots of other vegetation with the sharp apical ends of its rhizomes, which potentia lly leads to infection and/or death (Boonitee and Ritdhit 1984; Dozier et al. 1998; Jose et al 2002). By producing a llelopathic phenolic compounds, I. cylindrica is capable of preventing the growth of other plants including some pines by inhibiting germination (Sajise and Lales 1975). I. cylindrica s extensive rhizome network make it particularly co mpetitive belowground and as a re sult, monotypic patches of the grass form in infested areas, altering native sp ecies richness and diversity. The dense rhizome

PAGE 15

15 mat in the soil often is a mechanical hindrance to root growth of native speci es (Jose et al. 2002). In Florida sandhill savannas, it has been demonstrated that I. cylindrica replaced all native understory species (Lippincott 1997). Lippincott (2000) showed that I. cylindrica altered fire regimes by increasing flame heights and temperatur e, resulting in increased mortality of even fire-tolerant Pinus palustris Soil characteristics are altered with the invasion of I. cylindrica. In the Florida sandhills, soil moisture taken at 10 to 30 cm depth in the I. cylindrica plots was about 50% lower than the natural area plots (Lippinc ott 1997). This reduction in soil mo isture coincided with the 2 to 3 fold greater rhizome/root biomass, and 2-fold greater root length density in I. cylindrica plots relative to native species plots suggesting that in creased rhizome/root bi omass and root length leads to soil water depletion and intense below ground competiti on with native plant species and nearby pines. Collins and Jose (in review) demonstrated that I. cylindrica decreased soil nitrate and potassium levels and altered soil pH in inva ded pine flatwoods. Phosphorus levels may be altered in the soil after invasion as it was demonstrated that I. cylindrica is particularly competitive for phosphorus (Brewer and Cralle 2003). Community Susceptibility to Invasion The empty niche hypothesis states that exotic plants can be successful in invading a new community by accessing resources not being uti lized by the local species (Elton 1958, Levine and DAntonio 1999, Mack et al. 2000). A community then is susceptible to invasion if there are vacant niches. The success of Centaura solstitialis L. invasion into California grasslands is hypothesized to be due to its extensive, deep roots which access unused water below the shallow roots of the other vegetation (Roc he et al. 1994, Hierro et al. 2005) Unused resources are often to blame for invasion success. Davis et al. ( 2000) suggest that as long as the quantity of available resources is matched by the amount of resource uptake, a community will resist

PAGE 16

16 invasion. Thus each community has a threshold, where resource availability equals uptake, in which invasion can not occur, but this threshold is not fixed (Davis et al. 2000). Pulses in resources or declines in uptake offset this threshold allowing community invasion explaining why invasion success often occurs after disturba nces, when mortality contributes to a reduction in resource uptake. Efficient uptake of resources and occupancy of all niches in a community, both of which may prevent invasion, may be the result of high dive rsity. This is the con cept that Charles Elton (1958) used to propose the divers ity-invasibility hypothesis, which states that more diverse are less susceptible to invasion. Several authors have tested the relationshi p between diversity and invasion success producing conflicting results. A negative relationship, in support of Elton (Fargione and Tilman 2005; Milbau et al. 2005; Fargione et al. 2003; Ruijven et al. 2003; Dukes 2002; Kennedy et al. 2002; Naeem et al. 2000; Knops et al. 1999; Hooper 1998; Tilman 1997; Rejmanek 1989) a positive one, rejecting the hypot hesis, (Smith et al. 2004; Stohlgren et al. 2003; Foster et al. 2002; Pysek et al. 2002; McKinney 2001; Lonsda le 1999; Stohlgren et al. 1999) and no relationship (Collins et al. 2006; Crawley et al. 1999) have been shown. A number of community effects have been at tributed to the negative relationship between diversity and invasion. A crowding effect in a diverse community may prevent invasion when there simply is no room for introduced species. In experimental grasslan d plots, crowding from up to 24 species reduced the cover of exotic speci es by up to 98% (Kennedy et al. 2002). In a diverse community, a complimentary effect ma y be observed when multiple species utilize different resources while occupying the same area. Exotic grass leaf-blade length was reduced by enhanced complimentarity in an European grassland with increased neighborhood richness (Milbau et al. 2005). The possibility of havi ng a species, which is strong competitor with

PAGE 17

17 invading species increases with increasing diversity (a sampling effect). For example, the presence of Bromus diandrus was observed to prevent the es tablishment of the ruderal, Eschscholzia californica (Robinson et al. 1995). A sampling effect may also be the basis for di versity to have a positive relationship with plant invasion. Stachowizc et al. (2002) contests that diverse systems are more likely to have species that facilitate invasion, making the commun ity more susceptible. Others suggest that communities with increased species richness also have high turnover resulting in resource pulse which favor invasion (Stohlegren et al. 2003). The success of exotic colonization may be driven by the same factors that contri bute to high native diversity, resource availability and high propagule pressure. Invasive Impacts on Resources of Communities Invasive species can alter resource cycling a nd availability after successful introduction and spread. Simply through shading, invasives ca n reduce light availabil ity to other community species. For example, in New Zealand, increases in Tradescantia fluminensis biomass reduced light availability which led to an exponential decrease in native forest species richness and abundance (Standish et al. 2001). Gorchov and Tr isel (2003) reported th at light reduction by Lonicera maackii caused reduced survival and growth of Acer saccharum in Ohio, USA. By altering light availabi lity, invasive species also affect water availability. Introductions of grasses into open spaces can lead to less evaporative lo ss by covering the soil surface (Hughes et al. 1991; DAntonio et al. 1998). Changing the mi croclimate, including soil temperature and soil water, can have important implications for mine ralization rates and nutrient availability in the soil. Plant introductions into communities can lead to changes in nutrien t availability either directly or indirectly. For example, invasive species can change soil physical prope rties, the

PAGE 18

18 microbial community, species dominance and plant functional groups, all of which can lead to changes in soil nutrient dynamics (Ehrenfeld, 2003). The effects on nutrient availability of exotic plant invasions depend on how different the new species characteristics are from the native resident species (Chapin et al. 1996; Ehrenfeld 2003). Seve ral studies have shown that invasive plants maintain higher amounts of nitr ogen in their tissues (Vitousek et al. 1987; Vitousek and Walker 1989; Witkowski 1991; Ashton et al. 2005). Higher nitrogen concentrations in tissues could in dicate higher uptake, meaning less available nitrogen in the soil. In a study comparing competition for nitrogen betw een ponderosa pine and associated grasses it was found that alien grasses competed heavily for soil nitrate compared to native grasses which led to poor growth of ponderosa pine seedlings in the alien grass treatment (Elliott and White 1989). Higher nitrogen uptake could also imply be tter leaf litter quality, which could lead to increases in cycling through accelerated rate s of decomposition (Allison and Vitousek 2004: Ashton et al. 2005). Specific Objectives The impacts of I. cylindrica on forest productivity and di versity in the southeastern United States has yet to be expl ored. Therefore, the overall obj ective of the current study was to quantify the effects of I. cylindrica invasion on an estab lishing pine forest with the following specific objectives: Review the proposed mechanisms for successful plant invasions Test if the theory proposed by Charles Elton concerning the diversity-invasibility hypothesis holds true for I. cylindrica and native pine sandhil l understory species in controlled mesocosms Determine how I. cylindrica impacts the survival, physiology and productivity of establishing loblolly pine ( Pinus taeda ) seedlings Quantify the competitive usage of nitrogen between I. cylindrica pine seedlings and native vegetation using 15N labeled Ammonium Sulfate.

PAGE 19

19 CHAPTER 2 MECHANISMS FOR PLANT INVASION: A REVIEW With increases in transport and commerce over the last thousand years, humans have been accidentally and deliberately dispersing and intr oducing plants to ecosystems beyond their native range (Mack et al. 2000). Plants making the tran sition to a new habitat must undergo a series of filters in order to become established; a historical filterwhich asks whether or not the species arrives, a physiological filterwhich asks wh ether or not the species can germinate, grow, survive, and reproduce and lastly a biotic filte rwhich address whether or not the species can compete and defend itself successfully (Lambers et al. 1998). Not every introduction results in naturalizat ion and only a few of those that become naturalized become invasive. As a statistic al generalization, Williamson and Fitter (1996) proposed the tens rule on the succ ess of plants and animals as invaders when introduced to new ranges. This rule suggests that 1 in 10 of the biota brought into a region will escape and appear in the wild, 1 in 10 of those will become naturalized as a self-sustaining population and 1 in 10 of those populations will become invasive Although the percentage of plants crossing borders becoming invasive seems low, the few th at eventually do have radical effects on native species populations, communities and ecosystem processes. Since 1958, Charles Elton and other ecologist s have made attempts to understand how introduced species become invasive in order to predict where and when invasions could occur. Dispersal, establishment, and survival are necessary for successful invasion of natural communities (Hobbs 1989), but what are the underly ing mechanisms for invasion? There are wide array of reasons as to w hy plant invaders may have rapid growth and spread in their new environments. Disturbance may reduce competition allowing for the establishment of invaders. Exotic plants may escape herbivores or parasi tes, which keep their populations low in their

PAGE 20

20 native lands. The invaders may alter their ne w environment in order to promote their own growth while suppressing the growth of others. Empty niches may occur in a community that can be filled by an introduced plant. There are several plau sible explanations and several mechanisms for invasion have been proposed. In this chapter, many of the foremost theo ries of plant invasion of new communities will be reviewed. Several have been proposed in r ecent years and some of the more prominent ones with regards to plant invasion will be addressed. The discussion will begin with some of the theories of ecosystem susceptibility to invasi on and the factors that may determine whether or not a community is invaded. Some of the theori es on how invasion is facilitated will be then portrayed followed by some of the suspected attri butes of invading plants. It should be noted that some of these theories have been heavily researched and supported or refuted, while some of the more recent ones lack experimental proof. So me of the following ideas that will be discussed are overlapping in concept and theory, while others are quite distinct. Biotic Resistance Hypothesis While some theories suggest that some plant species are able to easily invade new habitats because they do not encounter any herb ivores that threaten their establishment and spread ( natural enemies hypothesis discussed later), the biotic re sistance hypothesis says that exotics fail to establish and spr ead due to negative interactions between the introduced species and the native biota (Maron and Vila 2001). Enem ies of the intruders oc cur in their introduced habitat, which can suppress their spread and esta blishment. The native communities are able to resist invasion. In a common garden experime nt conducted in southern Ontario, Canada, the impacts of herbivory were tested on 30 old-field plant species. It was observed that non-native species experienced equal or greater herbi vory than natives (Agr awal and Kotanen 2003) suggesting some evidence of biotic resistance.

PAGE 21

21 The hypothesis holds true as l ong as there are generalist he rbivores in the community, which can attack the invaders and the amount of the invaders does not exceed the amount the herbivores can consume. Maron and Vila (2001) suggest that th ere is a threshold of exotic species abundance that generalist native herb ivores can successfully limit. Beyond this threshold, the biotic resistance no longer exists. A species co uld rise above the threshold by means of propagule pressure. If a species is contribut ing large amounts of seed to a community, there is greater insuranc e of its establishment, survival and spread (Hierro et al. 2005). This concept has lead to the propagul e pressure hypothesis. Many acknow ledge that the difference in the number of propagules arriving in a community plays a role in th e level of invasion (Williamson 1996; Lonsdale 1999; Mack et al. 2000). Fluctuating Resource Availability Theory of Invasibility Resource availability is one the driving fact ors determining what sp ecies occur within a community. When resources are limited, less species are able to establish themselves within a community and when resources increase either du e to disturbance, heavy herbivory, or even fertilization the window opens up fo r the establishment of new species It was this concept that led Mark Davis and his colleagues (2000) to devel op the fluctuating resource availability theory of invasibility. This states that an increase in the quantity of unus ed resources will allow a plant community to be susceptible to invasion. Thei r theory relies on the assu mptions that available resources such as light, nutrient s and water are accessible to inva ding species and that as long as there is no severe competition from resident species for those resources, the species should successfully invade the community. The theory rests on the concept that a commun itys susceptibility to invasion is not fixed (Davis et al. 2000). Fluctuati ons in resource availability will determine how prone a community is to invasion. Increases in resource availability can be driven by two means. First, resident use

PAGE 22

22 of resources could decline. Damage and mortality to resident species could occur as the result of a disturbance thus reducing the upt ake of resources. The other way resource availability can increase is by increasing the suppl y of resources at a rate faster than the uptake of the resident species. Examples of this include higher prec ipitation than normal that will increase water supply, eutrophication of soils increasing nutrien t availability or loss of upper canopy trees allowing for greater light availabi lity. A community can maintain its resistance to invasion with increases in resource availability as long as the species in that community increase their uptake. Decreases in resource availability will incr ease competition between resident species in a community and make that community even harder to invade. According to Davis et al. (2000) the invasibility of a plant community is based upon a balance between re source uptake and gross resource supply (Figure 2-1). As long as th ese two are equivalent, the community should be resistant to invasion. Fluctuations away from this balance either increase or decrease the communitys invasibility. The literature has demonstrated several times, scenarios in which fluctuations in resource availability have affected an ecosystems inva sibility. It was demonstrated, in Gros Morne National Park (Canada), a boreal eco system, that resources essential for alien plants were either not limiting to the resident species or were s upplied by recent disturbances (Rose and Hermanutz 2004). The light availability a nd percent of bare ground partially produced by moose trampling were significantly higher than the undisturbed sites in this boreal ecosystem, which suggests that the increases in light availability allowed for invasion of aliens. Fluctuating light availability in a podocarp/broad-leaved forest in New Zealand was also shown to be a driving factor for the invasion of Tradescantia fluminensis which would reach its maximum biomass at 10 to 15 percent full light (Standish et al. 2001).

PAGE 23

23 Empty Niche Hypothesis Davis et al. (2000) proposed that the fluctuation of resources is what leads to the invasion of plant communities by exotics. Disturbances often caused these fluctuations and the imbalance of resource supply and uptake. However, what if a relatively stable community just happened to have resources not being utilized by the native community that are accessible to newcomers? The empty (or vacant) niche hypothesis states that exotic plants can be successful in a new community by accessing resources not being uti lized by the local species (Elton 1958, Levine and DAntonio 1999, Mack et al. 2000). In order to test the viability of the empty niche hypothesis, Hierro et al. (2005) suggested that parallel studies should be conducted of the invasive in it s native and introduced range in order to show that the invasive species is using the unused resources in the new community while also showing that those re sources were being utilized by other plants in the native community. No such studies have been documen ted, but several studies of invasives in their introduced communities have alluded to the upta ke of unused resources by invading plants. Centaura solstitialis L. is believed to dominate California grasslands with its extensive, deep roots which access unused water below the shallow roots of the other vegetation (Roche et al. 1994, Hierro et al. 2005 and the sources there in). Studies involving cover crops pr eventing the spread of invasive s seem to be utilizing the empty niche hypothesis. If cover crops are plante d to prevent spread, they are essentially being used to occupy an unused niche and uptake unused resources that could be utilized by exotic invaders. In field experiments c onducted in Nigeria, it was observed that after twel ve months of planting there was up to a 71% reduction in Imperata cylindrica (invasive grass invading agricultural fields) biomass when cover crops such as velvetbean ( Mucuna pruriens (L.) were planted (Chikoye and Ekeleme 2001).

PAGE 24

24 Diversity-Invasibility Hypothesis Species rich communities are considered to ut ilize more resources and thus there are less empty niches to occupy. With fewer resources to tap and less niches to invade, species-rich communities may be less prone to invasion. This is one of the main concepts that led to the development of Charles Eltons di versity-invasibili ty hypothesis, which states that more diverse communities are less vulnerable to invasion. Several studies have been conducted, as of late, to determine whether or not Eltons theory holds true. In most cases, species richness was used as a surrogate for species diversity. Stohlgren a nd his colleagues (1999) have been leading the argument that the hypothesis does not hold true and instead, diversity and invasibility are positively related, while Tilman (2004) di sagrees and supports Eltons theory. Using two data sets of native and non-native plant distributions from throughout the United States, Stohlgren et al. (2003) ra n correlations of na tive and non-native pl ant species richness on multiple scales to determine if there was any re lationship between diversity of native species and non-native invasions. The results of this study de monstrated that native species richness was, in fact, positively correlated with i nvasive species distributions and as spatial scales increased the correlations grew stronger betw een native and non-native distri butions. They found that areas high in native species richness s upported large numbers of non-na tive species and propose the primary mechanism by which diverse habitats are able to support non-native plants is through rapid turnover. Increases in ric hness lead to increases in turnove r within the habitat resulting in pulses of available resources, which promote the growth of natives and non-natives. This concept supports the fluctuati ng resource hypothesis. Stohlgren et al. (1999) proposed another mechanism in which the success of exotic col onization is driven by th e same factors that contribute to high native ric hness, including high levels of propagule supply and resource availability as well as favor able environmental conditions.

PAGE 25

25 Several other studies have produced results in support of Stohlgren (Wiser et al. 1998, Higgins et al. 1999, Lonsdale 1999, Smith and Kn app 1999). Stachowizc et al. (2002) suggest that there are more likely to be facilitators or important habi tat forming species that make conditions ideal for an invasive in diverse comm unities. In a Kansas grassland, it was observed that a reduction in the dominance of the C4 grasses resulted in the reduction of the invasive Melilotus officinalis (Smith et al. 2004). Removal of th e dominants resulted in higher light availability of up to 35%, which nega tively affected the establishment of Melilotus (Smith et al. 2004). Tilman (2004) contends that as diversity incr eases, invasibility d ecreases. Recently, he justified this theory by citing stochastic theory for community assembly (Tilman, 2004). In this theory, every new species entering a community is treated as an inva der. There are three requirements for the establishment of an individu al in a community. First, community assembly results from the success and failu res of propagules of invaders. Next, an invading propagule will survive and reproduce by only utilizing unconsumed resources. Lastly, successful establishment of an invader depends on the resource requirement of an invader relative to other species in its community. During assembly of a community, mo re and more invaders are utilizing unused resources and as a result when the number of invaders increases the amount of available resources decreases making it harder for new inva ders to establish themselves (Tilman 2004). With these assumptions the number of invaders in a community will plateau as it becomes more diverse. This theory imitates the logistic theory and the idea of carrying capacity. As species numbers increase, the maximum number of speci es a system can support is approached. As species number increases, the probability of in vasion by a new invader decreases (Tilman 2004).

PAGE 26

26 Several mechanisms have been proposed that explain how diverse systems are resistant to invasion including the crowding e ffect, the complimentary effect, and the sampling effect. The crowding effect is one mechanism by which diverse systems reduce their susceptibility to invasion. In a crowded commun ity, there is little room for establishment of invasive seedlings. Kennedy et al. (2002) tested the relationship between species diversity and invasion in 147 experimental grassland plots of varying diversit y, from one species to 24 species. The success of 13 species of exotic plants wa s assessed. There was a 98% reduction in invader cover in the most diverse plots compared to monocultures, which was attributed to crowding (Kennedy et al. 2002). The complementary effect refers to the abil ity of multiple species to utilize different resources or different sources of resources, in such a way that they can coexist in the same area. Plants that complement each other within a comm unity efficiently utiliz e the various resources allowing for the community to be resistant to invasion. Plants complement each other by occupying different niches (empty niche hypothesis) In Belgium, the effects of three invader grasses on European grasslands were assessed for varying levels of dive rsity and it was observed that when increasing neighborhood richness, comp lementarity was enhanced, which negatively affected invader leaf length (Milabau et al. 2005). By complementing each other plants in a community can discourage invasion. Resources may be used more efficiently in di verse plots because they are more likely to have a species that is highly effective in capturing resources. This is re ferred to as the sampling effect and it can play a role in a communitys susceptibility to invasion. A highly diverse community is more likely to include a species, wh ich is capable of outcompeting an invader. The sampling effect stands as a possible mechan ism for why more divers e communities are less

PAGE 27

27 susceptible to invasion. One way to determine wh ether or not the sampling effect is playing a role in resistance is to test an invader in m onoculture of a wide variety of species. A single species may be tolerant of an invader and this may be further demonstrated in communities of varying diversity containing that spec ies. If the invader is consistently failing to be successful in growth or establishment each time it is paired with that particular species along different diversity gradients, then it is evident that there may be a sampling effect. Fargione and Tilman (2005) demonstrated ev idence of a sampling effect with prairiesavannah communities at varying levels of dive rsity (1,2,4,8,16 species). It was observed that invader biomass was inhibited in plots in the presence of st rongly competitive C4 bunchgrasses (Fargione and Tilman 2005). Soil nitrate concentrations decreased and root biomass of resident species increased with the presence of C4 grasses across a diversit y gradient, leaving the researchers to believe that communities are more resistant to invasion when they contain C4 grasses. Eltons theory of i nvasibility has brought on much de bate since 1958 and currently the discrepancy is still unresolved. The diversity-invasib ility hypothesis may only hold for certain types systems or only on certain spatial scales. The debate will likely continue and research will continue to test it with a variety of exotic species and invaded communities. Facilitation by Soil Biota Soil-borne mutualists could facilitate the inva sion of exotic plants. Soil biota can alter the soil conditions enough to favor the spread of an exotic over a native species. Reinhart and Callaway (2006) proposed the enhanced mutu alisms hypothesis, which acknowledges the possibility that there may be str onger facilitation of grow th of invasives by soil microbes in new habitats than what the plants e xperienced in their native range. The mutualisms this hypothesis refers to are that formed with my corrhizal fungi and nitrogen-fixers.

PAGE 28

28 The engineers of the enhanced mutualisms hypothesis suggest that the mutualism of invasive plants with mycorrhizae may not be ju st a two-way associati on (Reinhart and Callaway 2006). Instead, nonnative plants may use the associations of mycorrhizae and multiple plants and gain advantage by tapping into the mycelial network without providing the essentials for maintaining such a symbiosis. In the pres ence of arbuscular mycorrhizae and the North American native grass F. idahoensis the invasive Centaurea maculosa showed a 66% increase in growth compared to when grown in the absence of the fungi (Marler et al. 1999). The invasion of new habitats by plants particularly legumes may be facilitated by nitrogen-fixing bacteria. Nodul e production by invading legumes re quires a certain threshold of nitrogen-fixing bacteria (Parker 2001 ). Some legumes are able to invade with the aide of native bacteria and some succeed by bringing the nitrogenfixing bacteria with them. For example, the nitrogen-fixing actinomycetes Frankia maintains a symbiosis with Myrica faya (both from the same habitat) allowing it to invade and alte r the nitrogen cylce in ecosystems in Hawaii (Vitousek et al. 1987). The presence of certain soil bi ota in the exotic habitat may not facilitate greater growth and spread of plant species, but ra ther provide less restraint than biota in their native habitat. The activity of soil microbes may limit plant growth not only by limiting available nutrients, but also by providing negative feedback, wh ich keeps the species under cont rol. Thus, when the species is introduced elsewhere it may not be constraine d by the same mechanisms that disallow the plant growth and spread in its native ra nge. Callaway et al. (2004) reported that Centuarea maculosa experienced greater inhibitory effects by soil microbes in its native European soils than in North America. They attribute the differences in performance by the in vasive in the two soils

PAGE 29

29 to different feedback mechanisms. This study ma y be a demonstration of the plant escaping the inhibitory effects of its native soil biota, a theory which will be discussed later in this chapter. Invasion Meltdown Hypothesis Facilitation of invasion by plants may not only be facilitated by soil microbes and mycorrhizal fungi, but also by a variety of flor a and fauna. The phenomenon of already invading exotic biota opening the door for the invasion by other aliens by a ltering site conditions and by providing a positive feedback has often been obs erved. The invasion meltdown hypothesis states that increasing numbers of exotic s species facilitate a dditional invasions (Col autti et al. 2004). A meltdown of an ecosystem occurs as the num ber of invasive species increases. Invasion meltdowns could be facilitated by plants or animals. Plants that alter the soil characteristics may also facilitate invasion of ot her species. In Hawaii, several studies have demonstrated how Myrica faya a nitrogen-fixing shrub has inva ded volcanic nitrogen-poor soils and altered soil properties (V itousek 1986; Vitousek and Wa lker 1989). Vitousek (1986) suggested that M. faya could further facilitate additional plant invasion. Hughes et al. (1991) experimentally showed that there was signifi cant increase in biomass accumulation of the invader Psidium cattleianum in M. faya infested communities. Plants may also alter the soil characteristics by introducing chemicals, wh ich may hinder the growth of native species (allelopathy, which will be discussed later), which may allow for the establishment of exotics. The introduction of pollinating wasps to south Florida has allowed for the establishment of Ficus species, that depend on the pollinators fo r reproduction (Simberloff and Von Holle 1999). Several exotics plants invade eco systems after herbivores from their native land have feasted on the new plants. Exotic herbivores can reduce competition of native plants for the exotic plants. In a meta-analysis of 63 manipulative field studie s, Parker et al. (2006) observed that grazing by exotic herbivores allowed for 52% greater abunda nce of exotic plants in native communities.

PAGE 30

30 They also observed that grazing by exotic herbivores led to an in crease in exotic plant species richness, which they attributed to a reduction in the abundance of native species (Parker et al. 2006). An invasion meltdown of this sort requi res the invasive plants to be preceded by generalist herbivores. Sp ecialist herbivores are unlikely to impact the plants in a new habitat enough to allow for the introduction a nd success of new plant species. Natural Enemies Hypothesis With the invasion meltdown hypothesis, some plant species were successful invaders because they followed exotic generalist herbivores into new habitats. Some invasive plants may be successful because instead of following genera list herbivores, they escape from specialist herbivores that keep them from spreading in their native habitats. The natural enemies hypothesis assumes that natural enemies suppress plan ts in their native range and it is the escape from these enemies that allows exotic populatio ns to explode in thei r new habitat (Maron and Vila 2001). These natural enemies are not limited to herbivores; fungal pathogens and destructive soil biota may also be considered enemies. The na tural enemies hypothesis has also been referred to as the enemy release, enemy escape, herbivore escape, predator escape, or ecological release hypotheses. Three points dr ive the basis of this hypothesis, 1) plant populations are regulated by natural enemies, 2) natives are affected more by enemies than exotic species, and 3) reduction in enemy regula tion should lead to incr eased plant population growth (Kean and Crawley 2002). In order to demonstrate the natural enemies hypothesis, a study would have to show that native herbivores or pathogens reduce plant population sizes and growth rates and at the same time show those same plant species suffer little herbivory or diseases and have increased population size and growth rate in their introduced habitat. Clidemia hirta, a neotropical shrub native to Costa Rica, which is currently invading Hawaii, was used to test the natural enemies

PAGE 31

31 hypothesis with the use of bot h insecticides and fungicide s and it was observed that Clidemia survival in Costa Rica increased 41% when both tr eatments were used (figure 2-2)(DeWalt et al. 2004). Plant growth in Hawaii was unaffected by the fungicide suggesting that fungal pathogens only limit its growth in its native land (DeWalt et al. 2004). In this case, Clidemia escaped the suppression by fungal pathogens, which is why it was invasive in Hawaii. Escape from natural enemies provides for the logi c behind the use of biological control. If a plant is released from suppression by some sort of specialist herbivore or pathogen, then that specialist enemy could be used to regulate plan t populations in habitats where that plant was introduced. A field study was conducted to test the biologic al control of saltcedars ( Tamarix ), an Asian tree species invading riparian areas of the United states with no real insect threat, by the Asian leaf beetle Diorhabda elongata deserticola in which the saltcedars were caged in with beetles (Lewis et al. 2003). A 60 to 99% de foliation of the saltcedars was observed by the beetles as well as substantial dieback, mortalit y of young plants, and limited regrowth in the following growing season (Lewis et al. 2003). Many insects were observed to feed on the plant species in Asia, where it grows in isolated patc hes, leading to the coev olution of specialized insects which feed solely on it (Lewis el al. 2003 and sources therein). The successful invasion by Tamarix species in the United States provides fo r an example of natural enemies hypothesis because it has escaped the specialist herbivores from its native range. Evolution of Increased Competitive Ability Plants in the presence of specialist herb ivores and pathogens develop defenses and tolerances of these enemies in order to surviv e and proliferate in their native lands. Their relationship with their specialist enemies are ofte n coevolved in that over time the plant species have learned to designate their resources towards surviving in the presence of enemies. When these plant species escape their native enemies a nd are introduced to new habitats, the resources

PAGE 32

32 they have been utilizing for defense can be al located to growth and re production (Blossey and Notzold 1995; Hanfling and Kollmann 2002). This may be what makes these introduced plants invasive. The evolution of increased competitive ability (EICA) hypothesis takes what was stated in the natural enemies hypothe sis and takes it a step further. It states that only when plants escape from coevolved specialist enemies, they ar e able to gain advantage over other plants in their introduced community by usin g the resources that were previously used for defense for growth and reproduction (Blossey and Notzold 1995). It has been suggested that the liberation from herbivores, allows for the selection of genotypes in the new community with increased competitive abilities (Blossey and Notzold 1995). An efficient method of testing the EICA is to grow the particular invasive species from seed from both locations in a common garden or identical conditions wh ile excluding pests. Support for the hypothesis comes from observing the plants from the introduced habitat performing better than the native since they experien ce little herbivore pressure and have adapted to allocating their resources toward s growth. Purple loosestrife ( Lythrum salicaria L.) from two locations, one with herbivory, (L eselle, Switzerland, where it is native) and without herbivory (Ithaca, New York, where it was introduced) we re grown in identical conditions and it was observed that plants from the region that expe rienced little herbivor e pressure had greater vegetative growth (figure 2-3) (Blossy and Notz old 1995). The results could be explained by the fact that introduced plants had es caped pressures of herbivory and were able to allocate resources for growth. Reproductive Traits Some invasion success could be explained by th e reallocation of resources from defenses to reproduction, while some species may simply be good invaders because they have the ability to reproduce quickly and in great numbers. Severa l authors have pointed out that invasive plants

PAGE 33

33 tend to be r-strategists consider ing that they tend to invade disturbed habitats (Rejmanek 1989, Hobbs 1991). This seems to be the st rategy utilized by several species of Pinus in invading regions outside their natural ra nge throughout the world. Rejman ek and Richardson (1996) have identified the main reproductive ch aracteristics, which cause certain species of pine (particularly P. raidata, P. contorta, P. hal epensis, P. patula, P. pinaster ) to be more invasive. They include short juvenile period, short interv als between large seed crops and small seed mass. Some of the other characteristics they identified include large number of seeds produced, better dispersal, shorter chilling period needed to overcome dor mancy, high initial germinability, and higher relative growth rate (Rejmanek and Richardson 1996 and sources therein). It seems apparent that all these characteristics allow for quick efficient spread of the pines. Superior Competitor Specialized reproductive abilities allow for some species to establish earlier, faster, and in greater numbers and these abilitie s put invasives at a competitive advantage. Some species after establishment may be better competitors for re sources than native vegetation, which may make them more successful. Bakker and Wilson (2001) proposed that differences in competitive ability may determine what species invade new ar eas. In their field study, they demonstrated that an introduced C3 grass, Agropyron cristatum (L.) Gaertn had a strong er ability to resist competition than the native C4 grass Bouteloua gracilis (HBK.) lag. from native and introduced competitors. In this case, Agropyron was a superior competitor th an the native vegetation. Several other studies have demonstrated an invasives ability to outcompete native species for resources. In a study examin ing the invasion of a longleaf pine ( Pinus palustris ) savanna in southeastern United States by the exotic grass, Imperata cylindrica it was observed that the clonal expansion of th e grass was reduced when plots were fertilized with phosphorus suggesting that the invasive was a better competit or for the resource (Brewer and Cralle 2003).

PAGE 34

34 When photosynthetic measurements were take n of two exotic invasive species of Rubus and were compared to two native species, it was observe d that the exotic species had higher rates of photosynthesis than its native c ousins giving implications of why the exotics were better competitors. In addition to being a superior compet itor in an introduced habitat under normal conditions, some invasives may be better competitors in stressful situations like disturbance. Ruderals, highly competitive natives and stress-tole rant species, are expected in communities immediately after a disturbance event (Grime 1979) However, it has been proposed that certain invasives are much more adapted to disturbance and thus are better competitors in the presence of disturbance than natives simply because the na tives have not experienced as much disturbance over time (Gray 1879; Mack et al. 2000). This concept known as the disturbance hypothesis has received little attention since it was proposed in 1879 (Gray 1879), but would be fairly easy to test and should receive attention. Novel Weapons Hypothesis Over time, plants may have persisted in their native range by exuding chemicals, which help them deal with competition by inhibiting the activity of neighbors (allelopathy). Due to much exposure to these chemical exudates, the neighboring plants may ha ve evolved resistance and thus are not impeded by them, but when thes e allelopathic plants are introduced to new communities, they have weapons that the plants in the new community have never experienced before and as a result may suffer the inhibitory aff ects. The inhibition of plant growth allows the introduced species to be at a competitive ad vantage over its new nei ghbors, allowing it to become invasive. The novel weapons hypothesis st ates that biochemicals released by a plant, which are ineffective against na tive neighbors who have been exposed to them over time, are inhibitory to plants or soil microbes in a new community contributing to its invasiveness

PAGE 35

35 (Callaway and Ridenour 2004). The introduced speci es are invasive because they present their new competitors with weapons they have not been exposed to before. Because of its high affinity for adsorbi ng organic compounds, activated carbon is often used to test allelopathy. Ride nour and Callaway (2001) used activ ated carbon to test whether or not the noxious weed Centaurea maculosa used novel weapons (all elopathy) to hinder the growth of the native bunchgrass, Festuca idahoensis They observed that Festuca had reduced root and shoot growth in the presence of Centaurea in pure sand compared to sand mixed with activated carbon (figure 2-4), implying that the Centaurea uses biochemicals to gain competitive advantage. Another activated carbon st udy demonstrated the use of allelopathy by Centaurea diffusa which had stronger negative effects on Nort h American species than Eurasian species (Callaway and Aschehoug 2000). As allelopathy continues to work for a species, natural selection will favor its reproduction and growth over other intruding species, which are able to compete. This has been referred to as the allelopathic advantage ag ainst resident species (AARS) (Callaway and Ridenour 2004). Just as the sp ecies that reallocated resources towards growth and reproduction (the evolution of increased competitive ability), species having success with biochemical weapons may allocate more to the production of th ese chemicals, thus increasing their success. Support for the AARS could then be observe d by introduced species being even more allelopathic than source populat ions (Callaway and Ridenour 2004). Integrated Mechanisms Although several mechanisms for invasion have been proposed and demonstrated through examples, it is clear that many species utilize more than one mechanism to gain competitive dominance in their new habitats. Dana Blumen thal (2005) suggested th at there might be an interaction between the natural enemies hypothesi s and the fluctuating resource theory because

PAGE 36

36 fast growing, species, which require high resources tend to be susceptible to enemies. This is due to the fact that high-resource species tend to be nutritious and lack structural material and defensive chemicals, which encourage herbivory (Blumenthal 2005). In a new range, the invader escapes its enemies and encount ers a flush of resources, whic h may be the result from a disturbance. The mechanisms for in vasion, therefore, may be integrated. The rhizochemical dominance hypothesis integr ates several mechanisms in explaining the success of invasive plants (Collins and Jose, in press). Th is hypothesis attributes invasive success to allelopathy (novel w eapons) and alteration of so il chemical properties by the rhizosphere exudates of the invader, which in turn favors its own growth while inhibiting the growth of competing vegetation. These chemical alterations may include changes in soil pH, and nutrient levels and availability. Imperata cylindrica was shown to alter soil pH and decrease soil nitrate and potassium levels in invaded areas co mpared to noninvaded areas (Collins et al., in press). Imperata has also been shown to be allelopath ic suppressing the growth of crops like tomato and cucumber (Eussen 1979) having greater impacts at lower pH (Eussen and Wirjihardia 1973). In this case, it seems that Imperata increases the potency of its weapons by altering soil chemical properties. The genus Centaurea has been proposed to have invasive success by suppressing the growth of native species by phytot oxicity or by altering soil microbial activity leading to restrictions of nutri ent availability (LeJeune and Se astedt 2001). By root induced mechanisms, the genus may be gaining dominance. Conclusions The fact that several mechanisms for invasion have been proposed in recent years, (many of which were discussed in this chapter) and that basically no generalizati ons can be made about the nature of invasive plants indi cates that research in this area is still fairly new and needs much attention. No one has yet to explain invasion pa tterns across a large range of systems and this

PAGE 37

37 may be simply due to the fact that each inva sive species is unique and that invasions are unpredictable (Williamson 1999: Dietz and Edward s 2006 and references therein). It has recently been proposed, however, that the conf licts in invasion theory result from the examination of different parts of the invasion process and it should be recognized that the processes enabling a species to invade change over the course of th e invasion (Dietz and Edwards 2006). Consideration of these different phases may allow for some generalizations to be made.

PAGE 38

38 Figure 2-1. The balance between gross resour ce supply and resource upt ake (as denoted by the isocline, where gross resource supply = res ource uptake) represents a communitys barrier to invasion. A quick increase in resource supply (A B), a decline in resource uptake (A C) or a combination of the both (A D) lead to an increase in a communitys invasibility because the re source supply is not matched by the uptake from the community. Source: Davis et al (2000) Reproduced after permission from Blackwell Scientific Publishers.

PAGE 39

39 Figure 2-2. Survival of Clidemia hirta in A) Costa Rica (where its native) and B) Hawaii (introduced range) in four natural enemy escape treatments in understory and open habitats. Survival was much higher in Hawaii where Clidemia has escape herbivores and fungal pathogens which have reduced it s survival in its native land. Source: (DeWalt 2004) Reproduced after permission fr om the Ecological Society of America.

PAGE 40

40 Figure 2-3. Mean (+ SE) dry biomass and height of purpl e loosestrife from Ithaca, New York, USA (Black) and Lucelle, Switzerland (gre y) after the growing season grown in common gardens under identical conditions. Based on data from Blossey and Notzold 1995. 0 20 40 60 80 100 120 0 20 40 60 80 100 120 140 160 180 20019911992Dry biomass (g) Plant height (cm)

PAGE 41

41 Figure 2-4. Biomass of root s, shoots, and total of Centaurea and Festuca grown together in grown in pots with and without activated ca rbon, which has a high affinity for organic chemicals. Shared letters designating mean s that are not significantly different. Source: (Ridenour and Callaway 2001) Re produced after permission from SpringerVerlag.

PAGE 42

42 CHAPTER 3 ROLE OF SPECIES IDENTITY IN PLANT INVASIONS: EXPERIMENTAL TEST USING Imperata cylindrica Introduction Studies on the invasibility of co mmunities have focused either on invader characteristics or properties of the community being invaded. A co mmunity property that has been examined and debated heavily in regards to invasion is di versity. Ever since E lton (1958) proposed the diversity-invasibility hypothesis, which states that more divers e communities are more resistant to invasion, observational and experimental st udies have been conducted in a variety of communities with a plethora of invaders to de termine the relationship between diversity and invasion. A negative relationship, in support of Elton (Fargione a nd Tilman 2005; Milbau et al. 2005; Fargione et al. 2003; Ruijven et al. 2003 ; Dukes 2002; Kennedy et al. 2002; Naeem et al. 2000; Knops et al. 1999; Hooper 1998; Tilman 1997; Rejmanek 1989) a positive one, rejecting the hypothesis, (Smith et al. 2004; Stohlgren et al. 2003; Foster et al. 2002; Pysek et al. 2002; McKinney 2001; Lonsdale 1999; Stohlgren et al. 1999) and no relationship (Collins et al. 2006; Crawley et al. 1999) have been shown. The discrepancy amongst all these studies may be a matter of scale. Levine (2000) proposed that correlations between native and exo tic species will be positive at large scales and negative at small scales. Other factors that va ry with species richness including disturbance, biomass, propagule pressure, resident cover, and climate, which have not always been accounted for may explain the conflicting results (Von Holle 2005) Despite the variabi lity in these factors, scientists have proposed mechan isms for the relationship between diversity and invasion they observe. Rejection of the diversity-invasibility hypothesi s and the success of exotics in diverse systems have been given several explanations. The success of exotic colonization may be driven

PAGE 43

43 by the same factors that contri bute to high native richness including propagule pressure and resource availability (S tohlegren et al. 1999). Increased rates of plant turnover, that might result from increased native richness, lead to pulse s of resources that allow exotic species establishment (Stohlegren et al. 2003). Others contest that diverse systems are more likely to have a facilitator(s) or habitat forming species which make the conditio ns ideal for invasion (Stachowizc et al. 2002). For example, when the dominant C4 species were removed from a Kansas grassland, establishment of the invasive Melilotus was reduced because increased light availability altered microsite evapotranspira tion and soil moisture (S mith et al. 2004). Diverse systems may be unsatur ated and propagule limited (B runo et al. 2004), which may explain the positive relationship between native species richness and invasion. With an ample amount of unutilized resources, co mmunity assembly can occur. However, when the availability of resources decreases, invasi on becomes difficult. As the community becomes more diverse and more resources are used, the number of invade rs will plateau and the success of invasion will decrease, thus supporting Eltons hypothesis (T ilman 2004). In other words, invasion success will be low when the community is crowded and all resources are utilized. A crowding effect has been suggested as a mechanism of resistan ce. For example, Kenne dy et al. (2002) reported that crowding of up to 24 species in experime ntal grassland plots in Cedar Creek, Minnesota reduced the cover of 13 exo tics species by up to 98%. In a diverse community, the ability of multiple species to utilize different resources while coexisting in the same area (complementary effe ct) may prevent invasive success. Increased neighborhood richness of an European grassla nd enhanced complimentarity, which negatively affected exotic grass leaf-bla de length (Milbau et al. 2005). Species representing different functional groups tend to compliment each other becau se of the different roles they play in the

PAGE 44

44 community. For example, Hooper (1998) found that late season forbs ( Hemizonia luzulaefolia and Lessingia micradenia ) combined with perennial bunchgrasses ( Sitanion jubatum and Stipa pulchra ) resulted in increased relative growth yield. Lavorel et al. (1999) suggested that studies examining invasion success should not focus on species richness, but on the functional groups since functional diversity might play a larger role in invasion. E xperimental evidence exists to support this hypothesis. In a microcosm study usi ng eight different Califor nian grassland species of varying functionali ty, the success of Centaurea solstitialis was reduced with increased functional diversity, which utilized more resources (Dukes 2001). A single functional group may e nhance or deter invasion. Le gumes increased the biomass and net fecundity of Conyza while members of the Asteraceae family improved its survival (Prieur-Richard et al. 2000). With high nitrate consumption and root biomass production, C4 bunchgrasses were shown to be prevent establishmen t of invaders in prairie grasslands (Fargione et al. 2003; Fargione and Tilman 2005). Thus, a community with increased functional diversity may have the functional group that will determine th e success of an invade r (sampling effect). The sampling effect by a diverse system ma y not be limited to a functional group, but a particular species. Species identity may be the limiting factor to invader establishment and growth. For example, Bromus diandrus dominance prevented the establishment of the ruderal, Eschscholzia californica in a California winter annual grassland (Robinson et al. 1995). The conflicting results of the many studies that have examined the relationship between invasion and diversity have left many questions. One criticism, however, for several of these studies is that all new species to a community, wh ether they were native or exotic, were treated as invaders. The dilemma arises when not all the species that are addressed as invaders in these experiments are truly aggressive or invasive, providing for poor evidence of a communitys

PAGE 45

45 resistance to invasion. We test ed the validity of dive rsity-invasibility hypot hesis in a mesocosm experiment with Florida sandhill forest understory species a nd an aggressive rhizomatous invasive C4 grass, Imperata cylindrica which is considered a pe st in over 73 countries (MacDonald 2004). A perennial grass, I. cylindrica thrives in both undisturbed a nd disturbed areas in a wide variety of soil types (Jose et al. 2002). I. cylindrica spreads sexually, through prolific seed production (3000 1-millimeter long grains per plan t; Holm et al. 1977), and vegetatively through rhizomes, which are resistant to heat and breakag e. Rhizomes comprise over 60% of the plants biomass resulting in a low shoot/root ratio that a llows it to survive and thrive after burns or cuttings (Sajise 1976). Regeneration of new plan ts can occur from rhizome fragments weighing as little as 0.1 g (Ayeni and D uke 1985). The extensive growth of I. cylindrica results in dense mat of rhizomes in the soil weighing up to 40 Mg per hectare (Terrry et al. 1997) that can produce approximately 4.5 million shoots (Soerjani 1970). Because I. cylindrica occupies a lot of space belowground, it prevents the root grow th of germinating seeds. Lippincott (1997) showed that I. cylindrica replaced most understory species in a Florida sandhill savanna, greatly reducing plant diversity. Criticisms of small scale experiments examini ng diversitys role in invasion suggest that the observed trends reveal little about the processes and patterns th at occur in reality in larger more complex ecosystems (Stohlgren et al. 2003) However, isolated mesocosms may allow for the elimination of extrinsic factors, which ma y disrupt the potential for diversity to reduce invasion (Naeem et al. 2000). Collins et al. ( 2006) have reported no relationship between I. cylindrica invasion and native diversity in both distur bed and undisturbed forest ecosystems.

PAGE 46

46 These authors suggested that ex trinsic factors in the study might have prevented them from drawing a relationship. The primary objective of this study was to te st if the diversity-i nvasibility hypothesis proposed by Elton held true in controlled meso cosms. We hypothesized that as diversity increased from monocultures to more diverse communities I. cylindrica invasion rate would decrease. We also examined whether or not functi onal diversity or species identity were a factor in invasion success. Because I. cylindrica is primarily an underground competitor, we hypothesized that the species that competed be tter belowground would be more successful in reducing invasion and spread of I. cylindrica For this reason, we also examined species characteristics belowground including biomass, root length, root length de nsity and specific root length. Methods Experimental Design and Study Site A completely randomized block design consisting of eight blocks and ten treatments was used to test the diversity-invasibility hypothesis with the exotic invasive I. cylindrica and five common Florida sandhill understory species (based on Buchanan et al. 1999). The understory species included a shrub, Ilex glabra two grasses, Aristida stricta and Andropogon virginicus and two forbs, Chamaecrista fasciculata and Pityopsis graminifolia. The design allowed for the comparison of five levels of species richness (0, 1, 2, 3, and 5 species) and three functional groups, grasses, forbs and shrubs. The ten treatm ents used in this study (Table 3-1) included a control with no native species, five monocultures of the prev iously mentioned species, a treatment with a grass mix, a treatment with a forbs mix, a 3-species treatment with a representative species from each functional group ( I. glabra A. stricta and C fasciculata ) and a 5-species treatment with all of the native species. The I. glabra monoculture was treated as the

PAGE 47

47 shrub functional group treatment. The monocu lture treatment mesocosms were used to determine if particular species were effective in preventing inva sion. The three grass treatments, the three forbs treatments, and the shrub treatmen t were compared to identify a particularly resistant functional group. The study was conducted on a flat unshaded 3600 m2 field on the West Florida Research and Education Center Farm of the University of Florida, located in northwestern Florida, USA (30 77 N, 87 14 W). In June 2004, native species were purchased from a local nursery and planted in eighty mesocosms (117 liter galvanized iron cans). Six individual plants from 1.9 liter pots were planted in each mesoco sm, except in the 5-species treatment, which had only five individuals. To ensure establishment and surviv al of healthy mesocosm communities, the plants were planted in a soil matrix commonly used in horticultural practices fo r growing Florida native species. This media was 90% bark (50/50 ble nd of aged and semi-aged) by volume and 10% sand. To each can, 18-6-12 Osmocote fertilizer wa s applied at a rate of 150 g of N, 50 g of P and 100 g of K per can. Lime was also applie d to maintain the pH between 4.5 and 5. The mesocosm communities were watered daily wi th a programmed sprinkler system through October 2004 and weeded when undesired species a ppeared. Holes were drilled at the bottom of each mesocosm to ensure proper drainage. By November 2004, the annual Chamaecrista fasciculate had died, but was able to reestablish from seeds in the same mesocosms by January 2005. In May 2005, a square meter of I. cylindrica (including rhizomes) was dug up from a local infestation. A single frag ment of rhizome 5 cm long each with a single shoot less than 10 cm tall was cut from the sample and plan ted directly in the center of each mesocosm. The mesocosm were continually weeded after the introduction of I. cylindrica to ensure that only the original six species were present.

PAGE 48

48 Data Collection Starting in May 2005 and throughout the summer, the number of I. cylindrica shoots were counted and the percent cover of each species wa s determined by ocular assessment by the same individual biweekly for each mesocosm. In October, all of the aboveground biomass was harvested from all 80 mesocosms, separated by species, dried at 65 C for 48 hours and weighed. The belowground biomass from 4 blocks was harves ted and separated by sp ecies to a depth of 60 cm by 20 cm increments. After all the biomass was washed, a subsample was taken from each species from each mesocosm to measure root leng th. Root length was measured using the line intercept method developed by Newman (1966) and as modified by Tennant (1975). Root fragments were evenly distributed over a grid of lines 2 cm apart and the number of intersections between roots and grid lines were counted. R oot length was determined with the number of intersections and the size of th e grid units. The samples were used to quantify the total root length of each species and root length density (RLD). The roots were then dried at 65 C for 48 hours and weighed. The dry weights of roots were used to calculate specific root length (SRL). Statistical Analyses For all the measurements that were taken at the conclusion of the study (species biomass, root length, RLD, SRL, final % cover, and final I. cylindrica shoot number) a one-way analysis of variance (ANOVA) was used to detect treatme nt differences using th e PROC GLM procedure (SAS Institute, Cary, NC) within the framework of a complete ly randomized block design. Treatment means were declared significantly different at < 0.05. Tukeys HSD post-hoc test was used for mean separation. For the biweekly measurements of species cover and number of I. cylindrica shoots, a repeated measures ANOVA was used to correlate the repeated observations within treatments as well as to determin e treatment differences. The success of I. cylindrica as

PAGE 49

49 measured by cover and biomass was regressed wi th the independent variables, % cover and biomass of native species and with number of species (richness) us ing PROC REG in SAS. Results There were not any significan t changes in the na tive species cover th roughout the summer of 2005 after I. cylindrica invasion, in all the treatments except for the C. fasciculata treatment, which dropped from 63% in June to 27% in August (P = 0.0002). Throughout the summer the A. virginicus treatment had significantly more cover than all of the ot her monoculture treatments with a summer mean of 89% (P < 0.001). The A. stricta treatment had significantly greater cover than the remaining three monocu lture treatments (P < 0.001). Comparing the three functional group treatments, the grass mix maintained the highest level of cove r (summer mean 85%), which was significantly greater than the forbs mix and shrubs treatments at 37 and 39%, respectively (P < 0.001). There was no significant difference between the 3 and 5-species treatment covers, but both were signi ficantly less than the grass mix and Andropogon viriginicus treatment. At the end of the study, there was a significant difference in the cover of I. cylidnrica between all the treatments (P < 0.0001) (Table 32). There was not significant difference in the I. cylindrica cover when all the grass treatments were compared to all the forbs treatments and the shrub treatment (P = 0.05). By the end of August, there was a significant negative relationship between the mean % cover of native species and the cover of I. cylindrica (r2 = 0.59, P = 0.01) (Figure 3-1). At the end of the study, there was a significant difference in the native biomass amongst all the treatments (P = 0.035)(Table 3-2). Of all th e treatments, the grass treatment had the most overall native biomass, while the Pityopsis graminifolia treatment had the least. The final total biomass of all of the monoculture treatments was not significantly different. Of the three functional groups, the grass mix treatment had the most biomass, being significantly greater than

PAGE 50

50 forbs mix (P = 0.034). The shrubs treatment was not different than the grass mix or forbs mix treatments in total biomass. Nonlinear regressi on analysis revealed th at the total biomass of I. cylindrica at the time of harvest had a negative loga rithmic relationship with the treatment mean total native biomass (r2 = 0.70, P = 0.003) (Figure 3-2). The biomass of I. cylindrica by the time of harvest was hi ghest in control (84.8 g) and lowest in the grass treatment (1.2 g) (P = 0.0203). Of the treat ments with native species, the forbs mix treatment produced the most I. cylindrica biomass (60.3 g). Becau se of the variability in I. cylindrica biomass production there were no signifi cant differences between the remaining treatments. There was not a relationship betw een the number of species and the biomass of I. cylindrica (Figure 3-3). Of the three functional groups that were te sted, the grass mix treatment had the least number of I. cylindrica shoots and smallest amount of I. cylindrica cover at harvest (P < 0.0001), while having the greatest reduction of I. cylindrica cover from the control (Table 3-3). These results of the grass mix treatment were not signifi cantly differently than the treatment that had all three functional groups. The shrub and forbs mi x treatment were the same in terms of the number of shoots, biomass and cover of I. cylindrica These two also had much less reduction in cover and biomass of I. cylindrica compared to the grass mix, 3-species and 5-species treatments. Repeated measures analysis revealed that as early as June 15, a month after introduction, there were differences in I. cylindrica cover among the treatments (P < 0.001) (Figure 3-4); the grass mix treatment had significantly less cover than the control, I. glabra C. fasciculata forbs mix, and 3-species treatments, while the control and I. glabra treatments had more I. cylindrica cover than all of the grass containing treatments and the P. graminifolia treatment. By July 18,

PAGE 51

51 all the grass containing treat ments had significantly less I. cylindrica cover than all the other treatments. By the first of August, the I. cylindrica cover greatly increased in the A. stricta treatment and it no longer was grouped with the other grass treatments. By August 15th, there were three significantly different groups of treatments; the first group having the highest I. cylindrica cover included the control, the I. glabra and the A. stricta treatments, the second group included the C. fasciculata and the Pityopsis graminifolia treatments and the forbs mix treatment, the third group with the lowest I. cylindrica cover included the 3-species, 5-species, grass mix and A. virginicus treatments. Belowground in monoculture, the A. virginicus and A. stricta treatments had significantly greater root length than th e other species including I. cylindrica (P = 0.0017) (Table 3-4a). A. virginicus had the highest RLD at 1.97 cm cm-3, which was significantly greater (P = 0.0109) than C. fasciculata Pityopsis graminifolia and I. cylindrica, but not A. stricta and I. glabra A. virginicus had the highest SRL as well at 1184 cm g-1, significantly greater (P < 0.0001) than Ilex glabra, C. fasciculata Pityopsis graminifolia and I. cylindrica, but not A. stricta In mixed communities, A. virginicus had longer root length (P = 0.0012) and greater RLD (P = 0.0001) than all of the other species and I. cylindrica (Table 3-4b). With interspecific competition, A. virginicus maintained the greatest SRL (P < 0.0001), while I. cylindrica had the lowest. In monoculture, all the species had roots in the top 40 cm of soil (Table 3-5). Ilex glabra, A. virginicus, and I. cylindrica roots reached the bottom of thei r mesocosm (60 cm). In mixed communities, I. glabra, C. fasciculata, and P. graminifolia had their roots only in the top 20 cm of soil. Two and four percent of A. stricta and I. cylindrica roots respectively we re found at soil depth of 20 to 40 cm. Nineteen percent of the roots of A. virginicus occurred at this depth and six

PAGE 52

52 percent occurred in the bottom 20 cm. Only one percent of the I. cylindrica roots reached the bottom. The distribution of A. virginicus roots was relatively the sa me in monoculture and in mixed communities. Discussion Previous work, using I. cylindrica as a model invasive plant showed no relationship between community properties and resistance to invasion (Collins et al 2006). It had been suggested that properties of the exotic sp ecies, including its exte nsive rhizome network, tolerance of low fertility, and low light compensation point allowed its invasive success regardless of the communitys composition (Collins et al. 2006). However, extrinsic factors such as disturbance and varying soil fertility may have precluded a relationship between I. cylindrica invasion and community properties. In this controlled mesocosm experiment, most of the identifiable extrinsic factors were eliminated and still no relationship between diversity and invasion were observed (R2 = 0.12, p = 0.32). Increasing the cover of native species, within each mesocosm community, resulted in a decrease in the cover of I. cylindrica a negative relationship, which likely was due to a decrease in available light. Although I. cylindrica can survive in the shade [Ramsey et al. (2003) reported a light compensation point of 32 to 35 mol m-2s-1], it is best adapted for full sun conditions (Hubbard et al. 1944). Several stud ies have demonstrated that as light penetration increases so does invasion success (Knops et al. 1999; Milbau et al. 2005). Thomsen and DAntonio (2007) showed that in native Californian grass monocultu res, as light penetrati on decreased as a result of increased cover, the number of the European perennial grass, Holcus lanatus seedlings and culms decreased.

PAGE 53

53 The observed negative relationship between I. cylindrica and native cover could simply be the result of the short duration of the study. Data collection ceased at the end of August, four months after I. cylindrica was introduced. Had we pursued the study for multiple growing seasons, there was a possibility th at the trend would disappear. Patterson (1980) suggests that I. cylindrica can adapt to changes in light level through changes in specif ic leaf area and leaf area ratio, thus with time the invasive could adju st, survive and even maintain some cover in conditions of increased cover by native species. Field observations in ma ture forests, however, reveal that I. cylindrica generally is limited to forest edges and tree-fall light ga ps, where there is ample light penetration. It was also observed that I. cylindrica biomass decreased logarithmically with increasing native biomass. When the native species had ac cumulated a dry biomass of approximately 50 g, the biomass of I. cylindrica decreased dramatically. Because both native species biomass and cover were significantly related with I. cylindrica invasion, these results suggest that there may be a crowding effect. On small scales a crowding effect is commonly obse rved. Kennedy et al. (2002) demonstrated, over the course of two y ears in small experimental grassland plots, a greater than 90% reduction in inva der cover, which they attributed to an increase in neighbors and a rise in crowding index. Co mmunity saturation happens quickly at small scales (Brown and Peet 2003), which results in plants competing more directly for space and resources (Huston 1999). Limited space and resources reduce the proba bility of invasion. On the shorelines of Rhode Island, invasive plants were successful regard less of the level of di versity because of bare space (Bruno et al. 2004). With greater than 60% of I. cylindrica biomass occurring belowground (Sajise 1976), space, especially in a containerized mesocosm, would be a factor. It is likely that a crowding effect would occur.

PAGE 54

54 With increasing species richness, there is an increase the probability that a community has a particular functional group that aids the communitys resistance, a sampling effect (Pimm 1991). The grass group, including A. stricta and A. virginicus demonstrated the greatest resistance to invasion by reducing I cylindrica cover by 90% and its biomass by 98% compared to the control. When combined with the other functional groups, the gras ses performed nearly as well, reducing I. cylindrica biomass by 95%. This result sugg ests that as long as there are grasses within a communitys composition, some resistance to I. cylindrica will occur. Dukes (2001) tested the impacts of functional groups on the biomass of the invader Centaurea solstitialis and found that in combination, four functional groups reduced community invasibility, which he attributed to reductions in resource availability. The forbs mix and shrub treatments were less successful in resisting invasion. The shrubs treatment was not significantly differe nt from the control in terms of I. cylindrica cover. Of the functional groups that were teste d, the forbs mix treatment reduced I. cylindrica biomass the least (29%) compared to the contro l. The lack of resistance by these two treatments are likely due to the fact that both averaged the leas t amount of native cove r throughout the summer compared to all nine treatments that had na tive species, which suggests that increased light availability may have favored I. cylindrica. Light availability may not be the sole reason for I. cylindrica success. These functional groups may be facilita ting the growth of I. cylindrica The forbs mix group was composed of a legume ( C. fasciculata ) and an Asteraceae ( P. graminifolia ), two plant types that have been shown to facilitate invasion. In Mediterr anean old fields, it was shown that two Conyza species biomass and net fecundity increased w ith the presence of legumes, while Asteraceae favored its survival (Prieur-Richard et al. 2002). C. fasciculatas effect on nitrogen availability may play a

PAGE 55

55 role in I. cylindrica success. Alone, forbs could facilitate invasions, but their role as facilitators are reduced in functionally mixed communities. The presence of forbs in the functionally mixed communities may be the reason why the mixed comm unities were not as resistant as the grass only treatments. Based upon the biomass and cover of I. cylindrica at the end of the study (Table 3-3), it was clear that the grasses were the most resist ant functional group. A temporal examination of the mesocosms revealed that I. cylindrica spread was fastest in the I. glabra treatment over the course of the summer, with the cover increas ing by 16.4%. All the mesocosms containing forbs allowed I. cylindrica to spread at approximate ly same rate (increasing I. cylindrica cover by 7 to10%). The four best performing treatments, with I. cylindrica spreading the slowest, all contained grasses (the 3-speci es, 5-species, grasses, and Andropgon virginicus monoculture treatments). Community resistance thus appears to be a sampling effect with all treatments that contain grass being resistant. The results reveal, however, that not all grasses equally resist I. cylindrica invasion. For example, in the A. stricta monoculture treatment, I. cylindrica cover increased by 15.7% over the course of a month (from the middle of July to middle August; Figure 3-4). The sampling effect that we observed may be a matter of communities containing A. virginicus The three treatments that had A. virginicus had the strongest resistance to I. cylindrica and none of these had I. cylindrica cover increase by more than 1%. In fact, the A. virginicus monoculture treatment saw a decrease in I. cylindrica cover by 0.25%. This was mostly due to I. cylindrica mortality. Because the presence of A. virginicus dictated a communitys resistance to I. cylindrica, our work suggests that a communitys resistance to in vasion is a matter of species identity and not richness. As long as a communitys composition includes A. virginicus it should have some

PAGE 56

56 resistance regardless of its richness or functional diversity. Species identity has also been shown to dictate a communitys resist ance to invasion in a perennia l grassland in the United Kingdom (Crawley et al. 1999), a Californian winter annual grassland (Robi nson et al. 1995), a Californian coastal prairie (Thomsen and DAntonio 2007), a Californian serpentine grassland (Hooper 1998), and amongst mycorrhizal communities in Hawaii (Stampe and Daehler 2003). The success of grasses and in particular A. virginicus against an exotic grass suggests that resistance comes from native species that are si milar in function as the exotic. The invasive nature of I. cylindrica, in general, has been attributed to its competitive interactions belowground with its extensive rhizome network. I. cylindrica retains a large amount of biomass belowground [40 Mg per hectare according to Terry et al. ( 1997)], which occupies large amount of physical space and surface area for nutrient absorption. Whether in monoculture or in mixed communities, A. viriginicus had the greatest root length, RLD, and SRL of all the native species that were tested. These values exceeded that of the I. cylindrica rhizomes. With the greatest root length, A. virginicus was the most competitive at accessing new areas in the soil profile for resources. With the highest RLD, A. virginicus occupied the greatest vo lume of soil as well. Given these two conditions, it can compete h eavily for belowground resources. Apparently, A. virginicus uses the same belowground competitive strategy that I. cylindrica uses as its invasion strategy. The idea that resident species resist species functionally similar to themselves has been shown in several studies (Fargione and Tilman 2005). Fargione et al. (2003) showed that species from the same functional guild as the invader had the greatest negative effect on the invader cover. In monocultural micr ocosms, in which the invader Centaurea was introduced, the most effective competitor was a similar summer-active annual forb (Dukes 2002). In an experimental

PAGE 57

57 grassland, the two species that most strongly predicted the composition in a community were similar in growth form and hi story (Crawley et al. 1999). Andropogon virginicus also proved to be a successful competitor with other species by displaying little root morphological plasticity between the monoculture and mixed community treatments. While the other native species s howed reduction in root length, RLD and SRL, A. virginicus maintained statistically similar values for both monoculture and mixed community treatments. Interspecific competition did not affect the morphology and distribution of A. virginicus roots. We also observed that A. virginicus maintained the same proportion of roots throughout the soil profile of the mesocosms, whil e interspecific competition limited the roots of other native species to the upperm ost part of the soil column. A. virginicus extensive roots and position in the soil profile s uggest that the species was dominant in the belowground environment as well. Comparison of A. virginicus to A. stricta in the treatments that contained both species showed that A. virginicus accumulated more biomass (Figure 35). In the grass treatment, 87% of the total biomass belonged to A. virginicus, dominating over A. stricta and in the 5-species treatment, the largest portion of the total biomass (57%) belonged to A. virginicus A. stricta was never the dominant species. Fargione and Tilman (2003) observed a simila r trend in grasslands, with C4 bunchgrasses dominating and occupying 50% of the biomass. It was these bunchgrasses that were involved in inhibiting invaders. Having such a dominant species in a communitys composition may benefit a community by increasing its resistance to invaders. In a California winter annual grassland, dominance by one species of Bromus reduced the invasibility of the native ruderal, Eschscholzia californica (Robinson et al. 1995). The same study demonstrated that in the absence of Bromus more invasions occurred. These works suggest that a

PAGE 58

58 communitys resistance to invasion depends on having a particular dominant species that can outcompete the invader. Our study demonstrates that a dominant native species can resist invasion through the same mechanis ms that make the introduced species invasive. Resistance, then, is a matter of species identity. One criticism of this work may be related to the fact that only a single introduction of I. cylindrica was made to each mesocosm and that the results do not account for propagule pressure. Several studies have indicated that propagule pressure lead to increased invader richness and success despite the le vels of richness of the native species (Tilman 1997; Lonsdale 1999; Levine 2000). Brown and Peet (2003) fo und greater success of invasion with high propagule pressure in riparian areas of high diversity. With I. cylindrica being such a prolific seeder, multiple introductions to a small-sc ale area are probable. However much of I. cylindrica spread is belowground and encroachment into new communities usually happens through single rhizome fragments or single seedlings, which then develops into circular patches. The result of this work has implications for restoration of areas that have been invaded by I. cylindrica Countless studies have been done exam ining methods of control, whether by mechanical, cultural, biological, or chemical means (Macdonald 2004). Jose et al. (2002) suggested that an integrated appr oach that utilizes all the availa ble methods such as mechanical, chemical and biological may be necessary to control I. cylindrica infestations. The results of this study suggest that the proposed integrated manageme nt could be taken a st ep further. Following herbicide application, A. virginicus could be considered as a possible cover crop in sandhill communities. By doing this, there is insurance of the return of native species to the infested areas as well as preventing fu ture infestations and spread of this troublesome weed.

PAGE 59

59 Conclusions This work demonstrates evidence of negative relationships between the biomass and cover of native species and I. cylindrica Grasses proved to be the most resistant functional group providing resistance alone and in mixed functiona l communities. Repeated measures analysis demonstrated that treatments including A. virginicus were the most resistant to invasion over time, suggesting that resistance is a matter of species identity and the diversity-invasibility hypothesis held true by means of the sampling effect. The success of A. virginicus can be attributed it having significantly greater root length, RLD and SRL than all of the native species and I. cylindrica in monocultures. The same trends were observed of A. virginicus in mixed communities. The root morphology characteri stics allow it to be a strong competitor belowground where I. cylindrica is most aggressive. The native gr ass is able to compete with the invasive by utilizing the same growth and compe titive strategies. Future work should explore using more functionally similar species to an invasive in testing co mmunity resistance. Bunchgrasses were used in this experiment, how ever, grasses that spread clonally by rhizomes should be tested with I. cylindrica which may result in even greater resistance to invasion. The implications of this work are that A. virginicus should be planted to prevent invasion and in restoration areas following treatments of I. cylindrica

PAGE 60

60 Table 3-1. Summary of the ten treatments used. Treatment Functional group(s) Richness Control 0 A. stricta monoculture Grass 1 A. viriginicus monoculture Grass 1 I. glabra monoculture Shrub 1 C. fasciculata monoculture Forb 1 P. graminifolia monoculture Forb 1 Grass mix A. stricta and A. viriginicus Grasses 2 Forbs mix C. fasciculata and P. graminifolia Forbs 2 3-Species A. stricta, I. glabra, and C. fasciculata Grass, Shrub, Forb 3 5-Species A. stricta, A. viriginicus, I. glabra, C. fasciculata, and P. graminifolia Grass, Shrub, Forb 5

PAGE 61

61 Table 3-2. Analysis of variance of measuremen ts taken at the end of the study between all treatments of by functional group. Source of variation df F P Native biomass All treatments 9 2.409 0.0352 I. cylindrica biomass All treatments 9 2.703 0.0203 By functional group 2 0.794 0.4636 I. cylindrica cover All treatments 9 4.965 <0.0001 By functional group 2 3.078 0.0544 I. cylindrica shoots All treatments 9 8.253 <0.0001 By functional group 2 8.4526 0.0007

PAGE 62

62 Table 3-3. Summary of the I. cylindrica shoots, cover and biomass in the functional group treatments as well as the reduction of these values from the control. Significantly different means are accompanied by different letters ( = 0.05). Number of I. cylindrica Shoots % I. cylindrica cover % I. cylindrica cover reduction I. cylindrica biomass I. cylindrica biomass reduction Control 18.1(0.9) a 29.6(4.7) a 0 84.8(29.8) 0 Forbs mix 7.9(2.2) b 12.5(4.1) bc 58 60.3(45.4) 29 Grass mix 1.1(0.3) c 3.0(0.3) c 90 1.2(0.03) 98 Shrubs 9.7(2.4) b 21.2(5.9) ab 28 31.9(8.6) 62 Forbs+grasses+shrubs 2.2(0.4) c 5.0(0.8) c 83 3.9(2.2) 95 df 4 4 4 F 25.82 11.56 3.17 P <0.0001 <0.0001 0.06

PAGE 63

63 Table 3-4. Summary of root length, root length density and specific root length means(SE) for the native species and Imperata cylindrica in a) monoculture and in b) mixed communities. Significantly different mean s are accompanied by different letters ( = 0.05). a) Species Root Length (cm) RLD (cm cm-3) SRL (cm/g) A. stricta 107362(65900) ab 1.17(0.75) ab 1084(280) ab I. glabra 68796(11228) b 0.55(0.05) ab 562(80) bc C. fasciculata 4544(1149) b 0.08(0.01) b 264(56) c P. graminifolia 7010(1820) b 0.12(0.02) b 772(153) bc A. virginicus 226800(54089) a 1.97(0.54) a 1841(244) a I. cylindrica 7631(4020) b 0.07(0.03) b 119(37) c p-value 0.0017 0.0109 <0.0001 b) Species Root Length (cm) RLD (cm cm-3) SRL (cm/g) A. stricta 23834(9570) b 0.23(0.104) b 1052(148) b I. glabra 1383(897) b 0.03(0.02) b 110(37) bc C. fasciculata 1768(529) b 0.03(0.013) b 300(173) bc P. graminifolia 1694(223) b 0.03(0.005) b 1087(241) b A. virginicus 212427 a 1.83(0.945) a 2691(715) a I. cylindrica 606(353) b 0.005(0.003) b 31(6) a p-value 0.0012 0.0001 <0.0001

PAGE 64

64 Table 3-5. Percentage of total roots accounted for each species by depth in the soil profile. In monoculture With competition Species 0-20 cm 20-40 cm 40-60 cm 0-20 cm 20-40 cm 40-60 cm A. stricta 87.5 12.5 0 98 2 0 I. glabra 59.5 28.75 11.75 100 0 0 C. fasciculata 98.25 1.75 0 100 0 0 P. graminifolia 91 9 0 100 0 0 A. virginicus 77.5 17 5.5 75 19 6 I. cylindrica 75 24 1 95 4 1

PAGE 65

65 Figure 3-1. Relationship between % cover of na tive understory species treatment means and mean % cover of Imperata cylindrica. y = -0.2539x + 25.067 R2 = 0.590 5 10 15 20 25 30 35 40 0102030405060708090100% Cover of native species % Cover of I. c y lindrica control A. stricta I. glabra C. fasciculata P. graminifolia A. virginicus grass mix forbs mix 3 species 5 species

PAGE 66

66 Figure 3-2. Relationship between native treatment total biomass means and biomass of Imperata cylindrica. Native biomass (g) I. c y lindrica biomass (g) y = -5.1572Ln(x) + 40.104 R2 = 0.70 p = 0.003 0 20 40 60 80 100 120 020040060080010001200 control A. stricta I. glabra C. fasciculata P. graminifolia A. virginicus grass mix forbs mix 3 species 5 species

PAGE 67

67 Figure 3-3. Relationship betw een species richness as represented by species number and Imperata cylindrica biomass. Number of species I. cylindrica biomass (g) 0 20 40 60 80 100 120 0123456 control A. stricta I. glabra C. fasciculata P. graminifolia A. virginicus grass mix forbs mix 3species

PAGE 68

68 Figure 3-4. Percent cover of Imperata cylindrica in each mesocosm for the ten treatments in the summer of 2005. 0 5 10 15 20 25 304-Jun14-Jun24-Jun4-Jul14-Jul24-Jul3-Aug13-Aug23-Aug% I. cylindrica Cover control A. stricta I. glabra C. fasciculata P. graminifolia A. virginicus grasses forbs 3 species 5 species

PAGE 69

69 Figure 3-5. Above and belowground biomass means of native species in all the gr ass containing treatments. -400 -300 -200 -100 0 100 200 300 400 500 A. virginicusA. strictaGrasses3 Species5 SpeciesBiomass (g) P. graminifolia roots I. glabra roots C. fasciculata roots A. stricta roots A. virginicus roots P. graminifolia I. glabra C. fasciculata A. stricta A. virginicus

PAGE 70

70 CHAPTER 4 IMPACTS OF Imperata cylindrica AN ALIEN INVASIVE GRASS, ON THE PRODUCTIVITY OF AN ESTABLISHING PINE FOREST Introduction Invasion by non-indigenous speci es can have major impacts on native ecosystems with both ecological and economic conseq uences (Mack et al 2000). Th e structure and function of an ecosystem can be altered by the presence of an exotic invasive (Vitousek et al. 1997) due to changes in system-level rates of resource suppl y, trophic structure, a nd disturbance regime (DAntonio and Vitousek 1992). Such changes are cl ear in forests, where i nvasive plants affect the establishment, growth and productivity of new seedlings, wh ich represent the future canopy species. Several studies have been conducted testi ng the impacts of invasive plants on forest regeneration and growth and the mechanisms by which they do so. In Hawaii, it was observed that the dominant native tree Metropsideros polymorpha was not able to establish a single seedling beneath the canopy of th e invasive nitrogen-fixing Myrica faya due to physical characteristics of the leaf litter (Walker and Vitousek 1991). Increasing biomass of Tradescantia fluminensis led to an exponential decrease in the speci es richness and abundance of native forest seedlings in New Zealand by rapidly reducing th e light availablity (Standish et al. 2001). Competition for light was also implied to be the mechanism by which Lonicera maackii was reducing survival and biomass of Acer saccharum in Ohio, USA (Gorchov and Trisel 2003). The shrub Rhamnus frangula reduced the growth and survival of Acer rubrum Acer saccharam Fraxinus Americana and Pinus strobus in New Hampshire, USA, which the authors suggested might be due to belowground competition from Rhamnus with its extensive shallow root system (Fagan and Peart 2004).

PAGE 71

71 A species that is making a significant impact on forests in Southeas tern United States is Imperata cylindrica (cogongrass), a perennial, rhizomatous grass thriving in both undisturbed and disturbed areas with soil types ranging from nutrient-poor, coarse sands to nutrient-rich, sandy loams (Jose et al. 2002). I. cylindrica spreads by both sexual and asexual mechanisms. I. cylindrica is a prolific producer of seeds, with as many as 3000 1-millimeter long grains per plant, which generally are dispersed within 15 m of the plant (Holm et al. 1977), but may be carried by wind up to 24 km over open country (Hubba rd et al. 1944). Established plants spread vegetatively through rhizomes, long tough, white underground stems with short internodes. They comprise over 60% of the plants biomass, re sulting in a high root/shoot ratio that allows it survive and thrive after fire or cuttings (Sajise 1976). Regene ration of new plants can occur from rhizome fragments weighing as little as 0.1 g (Ayeni and Duke 1985). The extensive growth of I. cylindrica results in dense mats of rhizomes in the soils forming dense monocultural patches with fresh weights ranging up to 40 Mg per hectare (Terrry et al. 1997), producing approximately 4.5 million shoots (Soerjani 1970). Imperata cylindrica can negatively affect forests in a va riety of ways. The density of the belowground rhizome network makes I. cylindrica a mechanical hindrance to growth of roots of native species. The rhizome tips are sharp; they may even penetrate the roots of native species leading to damage or mortality by infection (Eussen and Soerjani, 1975). I. cylindrica occupies a significant of space belowground, which may prevent root growth of germinating seeds. Lippincott (1997) showed that I. cylindrica replaced most understory species in a Florida sandhill savanna and greatly reduced th e diversity. The leaf blades of I. cylindrica have been observed to reach heights of 1.5 m under good mois ture and fertility conditions (Holm et al 1977), which suggests that I. cylindrica may compete for light on the fo rest floor especially with

PAGE 72

72 understory species and young tree se edlings. The dense carpet of leaf blades may prevent sun light from reaching the upper soil layer, elimin ating the opportunity for seedling germination. Allelopathy has been suggested to be another mechanism by which I. cylindrica gains dominance over native species. Several st udies have demonstrated the impacts of I. cylindrica extracts on the germination, growth and survival of crop plants (Hubbard et al. 1944; Soerjani, 1970; Eussen et al. 1976). It was shown that I. cylindrica suppressed tomato ( Solanum lycopersicum ) and cucumbers ( Cucumis sativus ) especially at low pH (Eussen and Wirjahardja 1973). Koger and Bryson (2004) de monstrated that extracts of I. cylindrica roots and foliage with concentrations as low as 0.5% inhibited germination and Cynodon dactylon (L.) and Lolium multiflorum Lam. by up to 62%. Imperata cylindrica has been shown to alter disturban ce regimes of forests. Lippincott (2000) suggests that fires from swards of I. cylindrica burn at high temperatures reaching 450 C and at greater heights. If fire s this hot persist long er than a few seconds in any given area, not only will the understory species die, but also the j uvenile trees. Mortality was even observed for longleaf pine juveniles ( Pinus palustris ), which normally are fire tolerant (Lippincott 2000). Because I. cylindrica allocates significant carbon belowground, it is able to recover quickly after fire, which is why Lippincott ( 2000) suggests that frequent inte nse fires can convert a pine savanna into a I. cylindrica dominated grassland. I. cylindrica is also favored by disturbances other than fire. King and Grace (2000) showed that I. cylindrica can germinate, survive and grow in wet pine savanna comm unities after several types of disturbance including mowing, tilling, and light gaps created by cu tting or natural stand mortality. All the evidence thus far has demonstrated that I. cylindrica does pose a threat to forests by altering trophic structure and di sturbance regime and by belowgr ound interactions. Most of the

PAGE 73

73 studies, however, have focused on the impacts of this invasive on crop or understory species and little on how it impacts the overstory canopy species. Pine forests throu ghout the southeastern United States are being impacted by I. cylindrica and little work has been done to quantify the impacts of the invasive on the pine themse lves. In this study, we examined how I. cylindrica impacts the survival, growth and produc tivity of young loblolly pine seedlings ( Pinus taeda ) up to three years after planting. We compared how the seedlings performed in the absence of competition, with competition from na tive species and with competition from I. cylindrica using growth parameters and physiological measurements. With pine seedlings growing in the absence of competition as a reference for their growth potential for our site, we hypothesized that I. cylindrica competition would have a much greater impact on seedling growth and productivity than competition from native species by reducin g their photosynthetic and growth rates. Methods Site Description This field study was conducted on an industria l plantation site in Santa Rosa County, Northwest Florida, U.S.A. (30 50N. 87 10W). The site was a 60 ha cutover area (site index = 24.4m), which quickly became infested with I. cylindrica after harvesting of the 17-year-old loblolly pine in early 2002. Th e climate in this area is temperat e with moderate winters and hot, humid summers. Total annual precipitation in 2003 was 1928 mm with the wettest month being June (418 mm) and the mean annual temperature was 19.4 C and with August being the hottest (27.3 C) (NOAA). The soils on the site were mapped a Lakeland series of a Typic Quartzipsamment (89% sand, 7.8% silt, and 3.4% cl ay). Native species on the site included Smilax rotundifolia, Smilax aspera, Rubus o ccidentalis, Ilex glabra, Andropogon virginicus, Asclepius veriegota, Carphephorus paniculatus, Ilex vomitoria, and Erechtites hierarifolia

PAGE 74

74 Cultural Treatments The experimental design consisted of 15 plot s (7.9 x 10.4 m) in which the following three treatments were replicated five times: VF : vegetation freemaintained by weekly hand weeding NC : native competitionnatural establishment and growth of native vegetation was allowed. The major native species on site included Smilax rotundifolia, Smilax aspera, Rubus occidentalis, Ilex glabra, Andropogon virginicus, Asclepius ve riegota, Carphephorus paniculatus, Ilex vomitoria, and Erechtites hierarifolia IC : I. cylindrica competitiona dense I. cylindrica monocultural patch fr om local seed or rhizome sources In fall 2002, a uniform patch of I. cylindrica of about 1 ha in size was selected for the five IC treatment plots. The VF and NC treatment pl ots were randomly established in an area 20 m away from the advancing front of the I. cylindrica patch. The 20 m buffer zone was considered appropriate since I. cylindrica spread had been estimated to be about 2 m per year at this site (Collins et al. 2006). All the plots, except for the IC treatment, were site prepared in October 2002 with a tank mix of imazapyr and triclopyr at a rate of 936.2 ml ha-1 in order to ensure that there was no I. cylindrica infestation. All treatment plots had a 3 meter buffer around them. Because of the close proximity of the plots (all were within an area approximately 2 ha), the soil conditions were assumed uniform before installm ent of the treatments. On March 6, 2003,1-yrold bareroot loblolly pine seedli ngs, purchased from a local nursery, were planted in four rows of eight seedlings in each plot (32 seedlings per plot ) with spacing of 1.1m x 2.0 m. Seedlings were fertilized with Ammonium Sulfat e fertilizer at the rate of 55kg N per ha in March 2003. Four randomly selected seedlings were fertilized with 15N labeled Ammonium Sulfate (5% enrichment) at the same rate for a comp anion project (Daneshg ar et al. 2007).

PAGE 75

75 Growth and Gas Exchange Measurements The root collar diameter (RCD) and height were measured for every seedling at planting and re-measured at the end of the growing season in 2003, 2004 and 2005. These values were used to calculate stem volume index (SVI; RCD2*height). Seedling survival was quantified during each measurement period. In December 2003 and June 2005, four seedlings from each plot were harvested. The needles of all the seedlings harvested in December 2003 were scanned with a LI 3100 leaf area meter (LiCor Inc., Linc oln, NE, USA) to estimate total lead area. The roots, leaves and stems of all the harvested seedlings were separated, dried at 65 C for 72 hours and weighed. The dried foliage of the Decem ber 2003 harvest was ground using a Wiley Mill to pass through a 1 mm screen. Foliar N concentratio n was determined using an isotope ratio mass spectrometer (UC Davis Stable Isotope Labor atory) since we al so quantified foliar 15N of the same samples for the companion stu dy (Daneshgar et al., 2007). Light saturated net photosynthesis (Amax), stomatal conductance (gs), and internal leaf CO2 concentration (Ci) of four pine seedlings per plot, were measured using a portable open leaf gas exchange system (LI-6400, LiCor Inc., Lincol n, NE) with a photosynthesis photon flux density (PPFD) of 1600 mol m s and a flow rate of 500 m/s of CO2 with a reference concentration of 370ppm. Measurements were taken between the 1000 and 1400 hours, monthly from July 2003 through October 2003. The uppermos t fully developed needles were measured. Approximately three fascicles (9 needles) were utilized for each measurement. The gas exchange measurements were recalculated using the total surface area of the needles, which was calculated based on the assumption that each fascicle was a cylinder (Madgwick 1964).

PAGE 76

76 Water Potential Measurements At a companion study site adjacent to the study plots, tensiometers (Soil Measurement Systems, Tuscon, AZ) were used to measure water potential in three identi cal treatments. Four tensiometers were placed in the ground in each tr eatment (12 total), two at a depth of 30 cm and two at 60 cm The tensiometers were re filled with water after each measurement. Statistical Analysis Seedling productivity and growth parameters (height, RCD, SVI, biomass, specific leaf area, and light saturated photosynt hesis) were compared for th e three treatments. One-way analysis of variance (ANOVA) was used to detect differences in the means with PROC GLM of the SAS statistical software pack age. (SAS Institute, Cary, NC 1999). Differences were declared significant at < 0.05. Using SAS, Levenes test for hom ogeneity among variances was used to determine which pairwise post hoc comparis on method should be used. For homogenous variances, Duncans post hoc test was used; fo r heterogeneous variances, Dunnetts T3 test (Dunnett 1980) was used. Results Survival and Growth At the end of the study, approximately 27 months after planting, only 26% of the seedlings growing in the IC treatment survived compared to 51 and 61% su rvival observed in the NC and VF treatments, respectively (Table 4-1) At planting, the seed ling mean root collar diameter was 4.23 mm and grew to 31.5, 13.4, and 5.9 mm for the VF, NC, and IC treatments, respectively (p<0.0001) (Figure 4-1a) by the end of the study. The root collar diameter differed twenty months after planting (p< 0.0001). Mean seedling height at the start of the study was 27.7 cm and grew to 136.6, 75.7, and 50.8 cm by the end of the study for the VF, NC, and IC treatments, respectively (p<0.0001) (Figure 4-1b). The seedling heights for the IC and NC

PAGE 77

77 treatments did not differ after the first growing season, but di d so by the end of the second growing season (p<0.0001). Mean SVI for th e seedlings in all treatments was 4.95 cm3 at planting and rose to 1455.8, 158.6, and 21.8 cm3 for the VF, NC, and IC treatments, respectively (p<0.0001) (Figure 4-1c). The SVI did not diffe r between the IC and NC treatments until the second growing season (F=44.63, p<0.0001). Above and Belowground Biomass After one full growing season, pine seedlings growing in the VF treatment had a mean foliar, stem and root biomass that differed from the other two treatments (p<0.0001) (Figure 42). Foliar and stem biomass did not differ between the NC and IC treatments. Only the root biomass differed between the two (p<0.0001). By the end of the third growing season, all treatments showed significant differences in foliar (p<0.0001), stem (p<0.0001) and root biomass (p<0.0001) (Figure 4-2). Seedlings in the VF treatment had th e highest biomass (612+ 79.9 g) followed by NC (83.3 + 12.8 g) and IC (14.0 + 3.8 g) treatments, with significant differences (F=59.4, p<0.0001) detected among them. Gas Exchange and Leaf Characteristics There was significant difference in Amax between the treatments throughout the summer of 2003 (Figure 4-3). The Amax summer means were 3.97, 5.07 and 5.85 mol m-2s-1 for the IC, NC and VF treatments, respectively (p=0.0006). Th e seedlings growing in the IC treatment maintained the lowest levels of Amax each month and always differed significantly from the seedlings growing in the absence of competition (Figure 4-4). Stomatal conductance of the pine seedlings in the VF treatment was higher than the other treatments, but was only signifi cantly higher than the IC trea tment (p = 0.0157)(Figure 4-3). There was no significant differences in the su mmer means of the three treatments for pine seedlings internal CO2.

PAGE 78

78 After one growing season, the s eedlings in the VF treatment had significantly greater total leaf area and specific leaf area (SLA) than the other two treatments (p<0.0001) (Table 4-2). There was no difference in leaf area and SLA be tween the NC and IC treatment seedlings. The pine seedlings in the IC treatment had the lowest nitrogen concentration (p= 0.0004) compared to the other two treatments, which we re not significantly different. Discussion Competition from native species and I. cylindrica prevented the pine seedlings from reaching the full growth potential th at was observed in the VF treatment. It was evident, as early as nine months after planting, that competition lim ited the growth of loblolly pine seedlings. The seedlings growing in the presence of competiti on had reduced height, diameter and mean SVI compared to the seedlings in the VF treatment. The mean SVI of the NC treatment was 27% of the VF treatment, while in the IC treatment it was only 7%. Nine months after planting, the seedlings growing in NC and IC treatments r eached only 21 and 11% the total biomass of the seedlings grown in the VF treatment, respective ly. Competition reduced the root biomass the most, with the NC and IC pine seedlings gr owing to only 18 and 7% of VF treatment, respectively. These trends in growth continued through the end of the study. Height, RCD, and SVI for both NC and IC treatments were all less than half of what was observed in the VF treatment. The seedlings growing in the NC treatment had a total biomass that was 13% of their full potential, while in the I. cylindrica it was 2%. The greatest difference at this point was observed in the foliage biomass. Interspecific competition impacts on loblolly pine seedlings has been well documented in experiments using weed control to create competition free treatments (Britt et al. 1990, Cain 1991, M iller et al. 1991, Morris et al. 1993, Martin and Jokela 2004) and these results support previous findings that the growth of seedlings are heavily impacted by competing vegetation.

PAGE 79

79 Because all seedlings experience a reduction in growth due to competition, the impacts of I. cylindrica on the pine seedlings are better seen when compared to the impacts of native vegetation. Nine months after planting, the IC treatment seedli ngs had a smaller mean root collar diameter (44% of NC) a nd smaller mean SVI (27% of NC). The total biomass of the seedlings grown in the IC treatment was 54% of what was measured in the NC treatment. For the growth parameters that were measured, th e difference between the two treatments nine months after planting was not significant (with th e exception of the root biomass and RCD). All the growth parameter measurements became significantly different between the two treatments by the end of the second growing season and the diffe rences increased with time. By the end of the third growing season, the seedlings in the IC treatment had smaller diameters (44% of NC), lower SVI (15% of NC), were shorter in height (71% of NC) and had a much lower total biomass (18% of NC). Concerning the biomass, the foliag e biomass of the seedlings in the IC treatment was most reduced compared to the NC treatment (12% of NC). Competition, as demonstrated by both the NC and IC treatments, impacted the rates of photosynthesis of the young pines with in the first year after planti ng. For the first three months of measurement, the seedlings growing free of competition maintained the highest rates of Amax. This conflicts with the findings of Green et al (1991) who observed that competition control had no effect on 4-year-old loblolly pine photosynthesis and Munger et al (2003) who reported a decrease in loblolly pine light-saturated photosynthesis with competition control. In October, pine seedlings in NC trea tment exhibited their highest rate of Amax, a rate that was higher than the VF treatment. The IC treatm ent pine seedlings also showed increased level of photosynthesis compared to how they had been performing all season. This significant increase in Amax in the two treatments may be explaine d by the decrease in temperature (from

PAGE 80

80 26.7 C in the summer to 20.1 C in October) or by dieback of competing vegetation. Because several of the competing species in the NC treatment were annua ls, it is possible that their mortality at the end of the seas on would result in an increase in light, nutrients, and moisture, which would favor an increase in photosynthesis of the remaining species. As temperature declines, I. cylindrica like most grasses, shifts its a llocation of nutrients and biomass belowground, resulting in browning and death of the aboveground biomass. This may favor competing species that are stil l capable of photosynthe sizing such as the pine seedlings. The results of the gas exchange measurements demonstrated that I. cylindrica competition impacts the pine seedlings physio logical function greater than NC competition. Every month, the seedlings in the IC treatment had lower rates of light-saturated photos ynthesis than other two treatments. The lower rates of photosynthesis were matched by lower stomatal conductance (gs), suggesting that photosynthetic rate of the pine seedlings may be limited by greater stomatal limitation brought about by I. cylindrica competition. The stomatal limitations may have been due to water limitation caused by I. cylindrica invasion. At the companion study site, with the same three treatments, it was observed that I.cylindrica reduced water availability leading to the lowest soil water potential of the three treatments at multiple depths throughout the summer (Figure 4-5). The water stress was more severe at shallowe r depths, where most of the belowground biomass of I. cylindrica occurs (Holm et al. 1977). Water limitation may play a role in the d ecreased photosynthetic capacity of pine seedlings, however, a stronger case may be made for nutrient deficienci es that result from I. cylindrica invasion. I. cylindrica was shown to decrease soil n itrate and potassium levels in invaded compared to non-invaded pine flatwood s (Collins and Jose, 2007). The invasive was shown to be more competitive for phosphorus than native pine-savanna species in the southern

PAGE 81

81 U.S. (Brewer and Cralle 2003) implying that phos phorus levels in the soil may drop with the presence of this invasive species. The same study also demonstrated that the extent of I. cylindrica invasion was negatively corre lated to the number of legumes present (Brewer and Cralle 2003), which would lead to the assumpti on that species that are capable of obtaining nitrogen from other sources (fixation) are able to compete with the grass. I. cylindrica thus, may be efficient at gathering nitr ogen hindering competing species from absorbing nitrogen. Analysis of the foliar nitrogen concentration reveal ed that seedlings growin g in IC treatment had significantly lower levels of nitrogen. Because a large proportion of nitrogen in the leaves occur in photosynthetic enzymes, reduced nitrogen co ncentrations in the IC pine seedlings may account for the reduced levels of photosynthesis. Several authors have demonstrated that capacity for photosynthesis correlat es with leaf nitrogen concen tration (Field and Mooney 1986, Reich et al. 1999, Henderson and Jose 2005), which was also observed in this study across the three treatments (Figure 4-6). The reduced levels of foliar nitrogen may also have contributed to the reduced total leaf area and sp ecific leaf area (SLA) that were observed of the IC treatment seedlings. Reduced SLA and leaf area imply re duced light capturing ab ility and productivity, which explains why the biomass of the seedlings was reduced. The decrease in photosynthesis in pines in the IC treatment, whether it was due to decreased nutrient or water stress, does indicat e belowground stress caused by the invasive. It was demonstrated that Liquidambar styraciflua had reduced leaf photos ynthetic capacity not only from aboveground competition with vines Lonicera japonica and Parthenocissus quinquefolia but from belowground competition as we ll (Dillenburg et al. 1995). After a full growing season, the roots of the s eedlings grown in the IC treatmen t were the most affected part of the seedling compared to the NC treatment. This suggests that in the presence of I. cylindrica

PAGE 82

82 emerging juvenile trees deal with the greatest competitive stress belowground. I. cylindrica retains more of its biomass belowground (over 60% according to Sajise, 1976) which has been reported to be 5 to 10 times that of native understory belowground biomass in southeastern forests (Ramsey et al., 2003). This clearly indicates that the in tensity of resource competition belowground between I. cylindrica and pine seedlings could be far greater than that between pine seedlings and native vegetation. Species that main tain high levels of density either above or belowground decrease the growth of competing trees. After one season, loblolly pine growing with Andropogon virginicus showed decreases in SVI with increasing density; 4 individuals/m2 reduced SVI by 60% compared to a co mpetition control and 16 individuals/m2 reduced SVI by an additional 22% (Pe rry et al. 1993). I. cylindrica maintained greater than 90% cover of the ground in all the plots during the course of this st udy suggesting that its high density, both above and below, could be a factor in the reduced growth of pine seedlings. Conclusions Only recently, has attention been br ought to the potential impact of I. cylindrica invasion on establishing forests. Though forest managers have acknowledged the ne gative effect of this species on new plantations (Jose et al 2002), this is the first work to demonstrate the impacts of I. cylindrica on establishing pines. Compared to native vegetation, competition from this alien grass leads to half the survival of pine seedlings after three years. Pine seedlings competing with I. cylindrica were significantly smaller in RCD, height and biomass than those competing with native species. We believe that I. cylindrica reduces the productivity of young pines by altering conditions (water and nutrient availability) belo wground where its presence may be felt the most. We show some evidence here of I. cylindrica altering soil water availa bility and foliar nitrogen which causes reduction in photosynthetic capacity. Detailed examination of its impacts on

PAGE 83

83 belowground resources is essential to fully understand the mechanisms responsible for the observed reduction in growth. Table 4-1. Percent survival (%) of total loblolly pines seedlings planted in each treatments in March 2003. Growing Seasons Treatment* 1 2 3 VF 63 62 61 NC 62 57 51 IC 57 51 26 *Treatments: VFvegetation free (no competition), NCnative competiton, ICI. cylindrica competition

PAGE 84

84 Table 4-2. Mean (SE) leaf area, specific leaf area (SLA) and % foliar nitrogen of 9-month-old loblolly pine seedlings for different treatme nts Different letters represent significant differences in means ( = 0.05) Treatment Total leaf area (mm2) SLA (mm2/g) %N VF 3450.8(279.6) a 50.63(2.4) a 1.71(0.03) a NC 631.1(223.8) b 14.58(4.6) b 1.57(0.09) a IC 265.8(33.1) b 7.06(0.8) b 1.10(0.1) b P-value <0.0001 <0.0001 0.0004

PAGE 85

85 Figure 4-1. Pine seedling root collar diameter (RCD), height and stem volume index (SVI) means(SE) for the three treatments from planting through the end of the study. 2 6 10 14 18 22 26 30 34 VF NC IC 10 30 50 70 90 110 130 150 0 200 400 600 800 1000 1200 1400 1600 0 10 20 27 Months after p lantin g RCD (mm) Seedlin g hei g ht ( cm ) SVI(cm 3 )

PAGE 86

86 Figure 4-2. Mean biomass(SE) of the pine seed ling foliage, stems, and roots for the three treatments after one growing season ( 2003) and three growing seasons (2005). Different letters represent signi ficant differences in means ( = 0.05) 20030 10 20 30 40 IC NC VF 20050 50 100 150 200 250 300 350 400 FoliarStemRootBiomass (g) Biomass (g) a a a a a b a a a b b b b b c c c c

PAGE 87

87 Figure 4-3. Mean light sa turated photosynthesis (Amax), stomatal conductance (gs), and internal leaf CO2 concentration (Ci) summer means(SE) for pine seedlings in the three treatments. Different letters represen t significant differences in means ( = 0.05). 0 0.05 0.1 0.15 0.2 0.25gs (molm-2s-1) ab b a 0 1 2 3 4 5 6 7 IC NC VF 225 230 235 240 245 250Amax ( molm-2s-1) Ci ( molmol-1) a b c

PAGE 88

88 Figure 4-4. Monthly light saturated photosynthesis (Amax) of the pine seedlings in the VF, NC, and IC treatments. 3 4 5 6 7 JulyAugustSeptemberOctoberAmax ( molm-2s-1) IC NC VF

PAGE 89

89 Figure 4-5. Soil water potential of the three tr eatments at two soil depths (30 and 60 cm). -100 -80 -60 -40 -20 0 IC NC VF -60 -50 -40 -30 -20 -10 0 150200250Soil water potential (mbars) Soil water potential (mbars) Day of year 30 cm 60 cm

PAGE 90

90 Figure 4-6. Relationship between leaf nitroge n concentration and light saturated net photosynthesis (Amax). Each point represents a pl ot mean (DiamondsVF, squaresNC, trianglesIC). y = 0.2329x + 1.602 R2 = 0.66 p = 0.0002 1 2 3 4 5 6 7 8101214161820A max ( mol m-2s-1 ) Leaf nitrogen concentration (mg g-1)

PAGE 91

91 CHAPTER 5 Imperata cylindrica AN ALIEN INVASIVE GRASS, MAINTAINS CONTROL OVER N AVAILABILITY IN AN ESTA BLISHING PINE FOREST Introduction It is well known that exotic invaders threaten the biodive rsity and stability of native ecosystems (Wilcove et al. 1998; Mack et al. 200 0). The changes in ecosystems brought about by biological invasion include, but are not limited to, alteratio ns in trophic structure and disturbance regime, and in system-level ra tes of resource supply (Vitousek 1990). The implications of plant invasion on soil resource suppl y, in particular soil nutrients, have received much attention in the recent past. Ehrenfeld ( 2003) suggested that cha nges in nutrient cycling associated with plant introductions might result from invasive species induced changes in soil physical and microbiological properties. This may be preceded by changes in species dominance and/or plant functional groups. New species in an ecosystem that do not cause any of these changes are not likely to cause sh ifts in nutrient cycling. The e ffects of exotic plant invasions depend on how different the new species character istics are from the native resident species (Chapin et al. 1996; Ehrenfeld 2003), The traits that make a particular species invasive may also lead to impacts on nutrient cycling. Invasive species that achieve success by utilizing resources not being taken up by the local species (Elton 1958, Levine and DAntonio 1999, Mack et al 2000), alter nut rient cycling by capturing untapped nutrients and redistributin g them through litter decomposition. According to Davis et al. (2000), the inva sibility of a plant community is based upon a balance between resource uptake and gross resource supply and as long as these two are equivalent, the community should be resistant to invasion. Puls es in nutrients, from outside sources, promote invasion and so nutrient increases would be ut ilized and retained in the ecosystem by the invasive plants.

PAGE 92

92 Some exotics are invasive due to higher photosyn thetic rates and specifi c leaf area that lead to higher growth rates (Baruch and Goldstein 1999; McDowell 2002) which could be driven by increased uptake of nutrients. Some invasive pl ants maintain a higher nitrogen (N) concentration in their tissues (Vitousek et al. 1987; Vitousek and Walker 19 89; Witkowski 1991; Ashton et al. 2005) indicating increased uptake. Increases in N in plant tiss ues imply improved litter quality leading to increased rates of decomposition followed by more uptake. Invasive plants can affect nutrient availability simply th rough their litter inputs and can increase cycling through accelerated rates of decomposition and uptake (Allis on and Vitousek 2004; Ashton et al. 2005). Exotic grass invasions can alter nutrient cycling directly or indirectly. Grasses can modify the microclimate of the soil by filling in all available space and thereby preventing radiation from reaching and drying the soil surface (H ughes et al. 1991; DAntonio et al. 1998). Conversion of woodlands to grassl ands or changes in species domi nance of grasslands can have major implications for nutrient cycling due to changes in litter type and quality and decompositions rates. The conversion of Hawaiia n woodlands to grassla nds by invasion lead to 3.4 times greater N cycling rates (Mack and D Antonio 2003). Grass invasions alter fire regimes, sometimes even enhancing them (D Antonio and Vitousek 1992), which can reduce nutrient availability through litte r loss by burning and volatili zation of nutrient containing compounds. Because grasses generally have sh allow roots, they may reduce the nutrient availability of the uppermost soil layers. When they form dense root systems with significant biomass belowground, they can retain increased amounts of nutrients making them unavailable to other species. In a study comparing grasses, Tilman and Wedin (1991) demonstrated that grasses that allocate more carbon to roots reduce soil N the greatest.

PAGE 93

93 The changes in nutrient cycling caused by exotic grasses can endanger young tree seedlings in a regenerating forest. Weed contro l experiments have been conducted on many crop tree species throughout the world, showing that reduction in comp eting vegetation promoted the growth of the tree species (Britt et al. 1990, Cain 1991, Miller et al. 1991, Morris et al. 1993, Martin and Jokela 2004). Supplementing nutrients through fertilization can have similar effects as vegetation control, suggesting that competitio n with other vegetation for nutrients plays a major role in limiting tree growth. Exotic specie s that are better competitors for nutrients than natives are more likely then to be greater constraints on competing tree species. An exotic grass species that threatens emergi ng pine forests in the southeastern United States is Imperata cylindrica a C4 rhizomatous perennial grass introduced from Asia. Although I. cylindrica can have leaf blades of up to 1.5m ta ll in conditions of good soil moisture and fertility (Holm et al. 1977), the majority of its biomass occurs belowground (Ramsey et al. 2003). It has a low shoot-to-roo t/rhizome ratio, with greater th an 60% of the total biomass occurring as rhizomes that are resistant to he at and breakage (Sajise 1976). Their rhizomes, which generally occur in the top 15 cm in clay soils, can reach depths of one meter or more (Holm et al. 1977). With its rhizomes, I. cylindrica can reproduce asexually from fragments as small as 0.1g (Ayeni and Duke 1985). Sexual produc tion also occurs by way of seed production. I. cylindrica is a prolific seeder producing as many as 3000 seeds per plant (Holm et al. 1977). Whether by seed or by rhizome, I. cylindrica can invade a variety of ecosystems from xeric uplands to shaded mesic sites (Jose et al. 2002), tolerating a variety of soil conditions, mostly favoring acidic soils (Wilcut et al. 1988). Its invasion and spread lead to changes in the functioning of an ecosystem by altering soil ch emistry, nutrient avai lability, hydrology and disturbance regimes. Collins and Jose (2007) showed that I. cylindrica decreased soil nitrate and

PAGE 94

94 potassium levels and altered soil pH in invade d pine flatwoods. It was demonstrated that I. cylindrica was more competitive for phosphorus than native pine-savanna species (Brewer and Cralle 2003). In Florida sandhills, I. cylindrica reduced soil moisture by 50% in the upper 30 cm of soil (Lippincott 1997). I. cylindrica increased fire temperatures a nd heights in a longleaf pine forest (Lippincott 2000). Whether I. cylindrica is successful in outcompeting native vegetation and young pines by manipulating nutrient cyc ling and availability re mains unexplored. Using 15N nitrogen isotope as a tracer, we compared how I. cylindrica and native vegetation competed for N, while examining which exerted a greater competitive stress on emerging one-year-old Pinus taeda seedlings. With its greater amount of belowground biomass, we hypothesized that I. cylindrica was a better competitor than native vegetation for N available in the soil. We also hypothesized that P. taeda growth would be impacted more severely by I. cylindrica than native vegetation as a result of the higher degree of belowground competition. Methods Site Description and Experimental Design The study was conducted in 2003 in Santa Rosa County in Northwest Florida, U.S.A. (30 50,N. 87 10,W). The region has temperate climat e with moderate winters and hot, humid summers. The mean temperature for this region in 2003 was 19.4 C with August being the hottest month (27.3 C) (Figure 5-1) (NOAA). Total annu al precipitation was 192.8 cm with the wettest month being June (41.8 cm). The soils on the site were mapped a lakeland series of a Typic Quartzipsamments (89% sand, 7.8% silt, a nd 3.4% clay) and were acidic (pH of 4.8) and had a mean observed soil temperature of 29 C during the growing season. The study area became infested with I. cylindrica when a 17-year-old P. taeda plantation was harvested the

PAGE 95

95 previous year. A design was used consisting of 15 plots (7.9 x 10.4 m) of the following three treatments replicated five times: Vegetation free (VF): plots were hand weeded weekly in order to represent the full growth potential in the ab sence of other vegetation. Native competition (NC): Establishment and growth of native vegetation from natural sources was allowed. The major na tive species on site included Smilax rotundifolia, Smilax aspera, Rubus occidentalis, Il ex glabra, Andropogon virginicus, Asclepius veriegota, Carphephorus pani culatus, Ilex vomitoria, and Erechtites hierarifolia Imperata cylindrica competition (IC): Established dense m onocultural patches of I. cylindrica from local seed or rhizome sources. In fall 2002, a uniform patch of I. cylindrica of about 1 ha in size wa s selected for the five IC treatment plots. The VF a nd NC treatment plots were random ly established in an area 20 m away from the advancing front of the I. cylindrica patch. The 20 m buffer zone was considered appropriate since I. cylindrica spread had been estimated to be about 2 m per year at this site (Collins et al. 2006). All the plots, except for the IC treatment, were site prepared in October 2002 with a tank mix of imazapyr and triclopyr at a rate of 936.2 ml ha-1 in order to ensure that there was no I. cylindrica infestation. All treatment plots had a 3 meter buffer around them. Because of the close proximity of the plots (all were within an area approximately 2 ha), the soil conditions were assumed uniform before installmen t of the treatments. In each plot, 32 1-yearold bareroot P. taeda seedlings, purchased from a commerci al nursery and graded for uniform size, were planted on March 6th in four rows with spacing of approximately 1 x 2m, corresponding to 3900 trees/ha. Fertilizer Application In order to trace the movement of N in the three treatments, four P. taeda seedlings were randomly selected from each plot (totaling 60 tr ees for the study) and fertilized on 27 May, 2003

PAGE 96

96 with 5% 15N enriched Ammonium Sulfat e in a 1 meter square around each seedlings at a rate equivalent to 55kg of N per ha. Sampling Methods Initial data collection began in May 2003. Monthly percent cover was assessed of each plot for the duration of the st udy by placing a square meter quadra t randomly in four locations within each plot and estimating the mean percenta ge vegetative cover. The means of the quadrat measurements were scaled up for each plot. Above and below ground biomass of the co mpeting vegetation was collected monthly (June 2003 to October 2003) from each plot. Aboveground I. cylindrica and native vegetation was clipped at ground level from two quadrats (0.5 m2). Live and dead foliage was separated and all samples were dried at 65 oC for 72 hours and weighed. Below ground biomass was collected using soil augers from the same plots used for aboveground biomass harvesting. Three soil cores were taken within each quadrat up to a depth of 18 cm. Roots a nd rhizomes were washed, dried at 65 oC for 72 hours and weighed. In December 2003, the 15N-fertilized loblolly seedlings were harvested. All competing vegetation, surrounding the seedlings within the en riched area was harvested as well. The foliage, stems and roots were separated for all harvested species and were dried at 65 C for 72 hours. All dried plant tissues were ground using a Wiley Mill to pass through a 1 mm screen. Cross-contamination of the 15N plant material was prevented by thoroughly cleaning the grinder between each sample with a vacuum and ethanol. On the same days the pine seedlings were ha rvested, four soil cores up to a depth of 90 cm were extracted from each 15N micro-plot using a soil auger, w ith a 3.5 cm diameter. The cores

PAGE 97

97 were divided into 30 cm increments to give 3 soil depths. Soil samp les were air-dried and subsamples were fine-ground with a mortar and pestle. Total N and 15N concentration of plant materials a nd soil were determined on an Isotope Ratio Mass Spectrometer (UC Davis Stable Isot ope Laboratory). We calculated %N derived from fertilizer, %utilization of fertilizer N, and %N recovery in soil using the data. A measure of the relative amount of N the pi ne seedlings obtained from the soil and from applied fertilizer, the percentage of plant N derived from fertilizer (%NDF), was calculated using the following formula (Wienhold et al. 1995): NDF (%) = 100* (a b)/(c d), where a = atom % 15N abundance in loblolly seedli ng needles, stems or roots; b = atom % 15N abundance in control loblolly seed ling needles, stems, or roots; c = atom % 15N abundance of fertiliz er applied; and d = natural atom % 15N abundance. The percentage utlization of fertilizer N (%UFN), was calculated for P. taeda seedlings as follows (Wienhold et al. 1995, Barber et al. 1996): UFN (%) = (%NDF* S)/R, where %NDF = percentage of plant N derived from fertilizer; S = kg N ha-1 in P. taeda seedling needles, stems, and roots; and R = kg N ha-1 applied. Percentage recovery of 15N fertilizer in soil (RFNsoil), a representation of the applied 15N remaining in soil, was determined using the equation as follows (De Mattos, 2000): RFNsoil (%) = 100* ((a c)/(b c)*(Np/Nf), where a = atom % 15N abundance in fertilized soil materiual; b = atom % 15N abundance in labeled fertilizer; c = atom % 15N abundance in non-fertilized so il (average background level); Np = total N of soil sample (in g); and Nf = total amount of 15N applied to the soil as labeled fertilizer (g). Statistical Analyses One-way analysis of variance (ANOVA) was used to detect differences in the means with the PROC GLM procedure of the SAS statistical software package. (SAS Institute, Cary, NC

PAGE 98

98 1999). Differences were declared significant at < 0.05. Using SAS, Levenes test for homogeneity among variances was used to dete rmine which pairwise post hoc comparison method to use. For homogenous variances, Duncans post hoc test was used; for heterogeneous variances, Dunnetts T3 test (Dunnett 1980) was used (SPSS ver. 9, SPSS Inc.). Results The aboveground biomass of the competing vegeta tion in the native and IC treatment plots were not significantly different throughout the summer months (Table 5-1). The summer mean of aboveground biomass for competing vegetation fo r both treatments were nearly identical at 239.7 g/m2 for the IC treatment plots and 234.8 g/m2 for the NC treatment plots. However, the IC plots maintained significantly greater percen t cover than the NC treatment plots each month (Table 5-1). The IC treatment plots had greatest cover in July (97.2%), while the NC treatment plots peaked in August (88.7%). Both treatm ents had their highest amount of biomass in September. The belowground biomass was significantly higher for the IC treatment compared to the NC treatment. The IC treatment plots consiste ntly maintained significantly higher belowground biomass than the NC treatment plots each month (Table 5-1). After one full growing season, the pine seedlings growing in the VF treatment had a mean total dry biomass of 90.4 + 9.5 g, which was significantly grea ter (p<0.0001) than what was observed in the other two treatments. The mean to tal dry biomass of the pi ne seedlings grown in the NC treatment (19.3 + 3.7 g) was not significantly differe nt from the IC treatment pine seedling biomass (10.4 + 1.0 g). The pine seedlings in the IC treatment pine s eedlings had the lowest amount of nitrogen in their biomass, while the VF seedlings had the grea test (Figure 5-2). The N content of the foliage

PAGE 99

99 in the VF and NC treatment seedlings were signif icantly greater than IC treatment pine seedlings (p<0.0001). The amount of N in the seedling stems in the IC treatment did not differ from NC treatment, but both were significantly lower than the VF treatment (p<0.0001). All three treatments differed in the amount of N in the pi ne seedling roots (p<0.0001). The amount of N in the competing vegetation was greater in the NC treatment plots than the IC treatment plots both above and belowground (p<0.001). The pine seedlings growing in the VF treatmen t had the highest N content per hectare. The N content of the pine seedling foliage was 2.05 kg N/ha in the VF treatment, which was significantly greater (p<0.0001) than both of the other treatments (Table 5-2). The NC treatment and IC treatment did not differ in pine seedling foliage N content. All three treatments were significantly different in the N content of pine seedling st ems (p<0.0001) and roots (p<0.0001), with the IC treatment pine seedlings consistently being the lowest and th e VF treatment being the highest. The seedling pine roots of the I. cylindrica treatment plots had the lowest overall N content (0.04 kg N/ha). The N content of th e native competing foliage (24.3 kg N/ha) was significantly greater (p<0.0001) than the N content of the I. cylindrica foliage (10.14 kg N/ha); however, the reversed wa s observed belowground; I. cylindrica had stored 32.9 kg N/ ha, while it was only 14.7 kg N/ha in NC treatment belo wground. This was significantly different (p<0.0001). The N derived from fertilizer (NDF) in the VF treatment was significantly lower than the other treatments in the pine foliage (p<0.0001) pine stem (p=0.0143), an d pine root (p=0.0011) (Table 5-3). In the pine foliage, stems and roots, the NDF was not significantly different between the NC and IC treatments. The greatest NDF was observed in the pine foliage of the IC treatment (11.3 %), while the lowest NDF was in th e pine roots of the VF treatment (4.2%). The

PAGE 100

100 I. cylindrica had significantly higher NDF than the native vegetation both above (p=0.0016) and belowground (p<0.0001) (Table 5-3). The total fertilizer N use efficiency (UFN) was significantly higher in the VF treatment than the other treatments in the pine seedli ng foliage (p<0.0001), stems (p<0.0001), and roots (p<0.0001) (Table 5-4). The pine se edlings in the other treatments we re never half as efficient as the ones growing in the VF treatment. The UFN was significantly higher in the pine stem and pine root of the NC treatment than the IC trea tment. There was no significant difference in the UFN of the competing native and I. cylindrica foliage; however, the I. cylindrica roots/rhizomes were significantly more efficient (7.68 %) than the native ro ots (0.99%; p<0.0001). The percent recovery of 15N in the soil (RFN) at the e nd of the growing season was significantly less in the VF treatme nt plots than the other treatment s at depths of 0-30 cm (8.2%; p=0.0037) and 30-60 cm (1.4%; p=0.0032) (Figure 5-3) The RFN of the NC and IC treatments never differed at any depth. There wa s no difference in the recovery of 15N in the soil between any of the treatments at a de pth greater than 60 cm. Using the treatment means of the N derive d from fertilizer in pines and competing vegetation as well as the 15N recovered from the soil, an estimate was made of the total percentage of applied fertilizer that was accoun ted for in the vegetation and soil. Approximately 81.5 % of the applied fertilizer N was accounted for in the IC treatment, while 62.2% and 24.7% was accounted for in the NC and VF trea tments respectively (Figure 5-4). Discussion Throughout the growing season, the I. cylindrica had significantly greater cover than the native vegetation despite that aboveground bioma ss in both treatments was the same. The evenspread I. cylindrica shoot biomass matched the patchy woody tissues of the shrubs in the NC treatment. The conversion from mixed species to a grass, by invasion, makes the forest floor

PAGE 101

101 more homogenous, filling more space and reduc ing radiation transmitta nce to the soil surface (Hughes et al. 1991; DAntonio et al. 1998). This has implications for the microclimate conditions of the invaded forest floor. The surface soil moisture of the I. cylindrica invaded plots was the wettest of the three treatments in July and August (Figure 5-5). Similarly, Ashton et al. (2005) showed that mixe d deciduous forest sites invaded with exotic woody species were wetter than uninvaded sites. The changes in microclimate brought on by invasion might affect the cycling of nutrients and the productivity of other species. In the conversion of Hawaiian woodland to grassland, by invading exotic gra sses following fire, alterations in microclimate were observed, which had implications for N mineralization (Mack and DAntonio 2003). Belowground, the biomass of the I. cylindrica was far greater than native vegetation. Each month, the I. cylindrica mean belowground biomass was at leas t 5 times greater than the native mean belowground biomass. The greater total biomass of I. cylindrica suggests that the invasive is far more productive than native species. The increased productivity mi ght be attributed to being more competitive for nutrients, which would have negative implications for the establishing P. taeda seedlings. Young P. taeda seedlings growing in the IC treatment grew only to 11% of their potent ial after one growing season, which was 54% of the biomass that was observed for the pine seedlings growing in the NC treatment. I. cylindrica reduced the growth of the young pine seedlings just as the shrub Rhamnus frangula reduced the growth and survival of Acer rubrum Acer saccharam Fraxinus Americana L. and Pinus strobus in New Hampshire, USA, which the authors suggest may be due to belowground competition from the Rhamnus with its extensive shallow root system (Fagan and Peart 2004). The N content in the P. taeda seedling tissues was affected by treatment. The NC treatment prevented the P. taeda from maintaining VF treatment N concentrations in their stems

PAGE 102

102 and roots. The pine seedlings in the IC plots never matched th e N content of the VF treatment seedlings in their foliage, stems, and roots. The pine seedlings growi ng in the IC plots had reduced levels of N in tissues compared to the NC treatment in both the foliage and roots. This suggests that I. cylindrica is more competitive for N than native species causing reduced levels of N in the P. taeda seedlings. Analysis of the N content of I. cylindrica above and belowground revealed, however, that I. cylindrica had less than half the N con centration of native species in both foliage and roots. On an ecosystem scale, native species stored 2.4 times the amount of N per hectare than I. cylindrica in foliage. On a site with native species, 61% of the to tal N occurs aboveground in the foliage, which drops to 23% after invasion of I. cylindrica. Belowground, I. cylindrica maintains much more N per hectare than native species Even though the N concentration of the belowground tissues is less than native species, th e fact that there is si gnificantly greater amount of I. cylindrica biomass belowground permits for more N stored belowground. In a study, comparing five grasses growing along a N gradie nt, Tilman and Wedin (1991) showed that the two grasses, Andropogon gerardi and Schizachyrium scoparium that reduced soil solution N the greatest had lower tissue N. These two species also had higher root allocation (Tilman and Wedin 1991) similar to the way I. cylindrica allocated biomass belowground. After invasion by I. cylindrica, the percentage of total belowground N on site increased from 37 to 76%. The invasion of I. cylindrica into an emerging pine fore st leads to a shift in where majority of the N is stored, from above to belowground. The low levels of N concentration in I. cylindrica compared to native species suggest that I. cylindrica may be productive at lower levels of soil N. This is contradictory to several studies, which have shown that invasives generally have higher concentrations of N in their tissues,

PAGE 103

103 which leads to higher litter quality and decomposition rates (Vitousek and Walker 1989; Ehrenfeld 2001; Allison and Vitousek 2004; Ashton et al. 2005). Increased decomposition rates act as a feedback mechanism for promoting the grow th of an invasive and it is the rapid cycling of nutrients that favors an inva sives success. Funk (2005) sugge sts that in invaded plots of Hawaiian montane forest, the herbaceous Hedychium gardnerianum factors in a tight N cycle with high N resorption from senescing Hedychium leaves combined with lower net nitrification rates. At our study site, a majority of the N belo wground that occurred prior to fertilization, must had been tightly retained in the underground networ k of the exotic rhizome in a rapid cycle of root/rhizome turnover and uptake. Controlli ng movement of N belowground may be a primary mechanism by which I. cylindrica maintains dominance after establishment. The NDF of I. cylindrica roots was more than three times the NDF of native species roots and the NDF of the I. cylindrica foliage was more than twice than that of the native foliage implying that I. cylidnrica is aggressive at attaining all sources of N. Not only was the I. cylindrica taking up more fertilizer N, it was also us ing the ammonium sulfate more efficiently than the native species (total UFN of 10.4% for I. cylindrica compared to 3.7% for the native species). The native species and I. cylindrica were utilizing the fertilizer N in their foliage equally, but I. cylindrica was much more efficient in the ro ots (7.68% compared to 0.99% by the native species). This demonstrates that I. cylindrica is both better in ta king up and utilizing N. This suggests that I. cylindrica maintains tight control of N cycling in infested communities. With I. cylindrica capable of taking up most of availa ble N, competing species are forced to acquire N from other sources. This explains why the foliage and roots of the pine seedlings growing in the IC plots had th e highest percentages of NDF of the three treatments. These seedlings took up more of the ammonium sulfate because they were more N limited. The pine

PAGE 104

104 seedlings in all three treatments were poor at utilizing the ammoni um sulfate efficiently. Total UFN for foliage, stem and roots for all three trea tments never exceeded 1%. However, there was a significant difference among treat ments with the seedlings in the VF treatment being the most efficient. The foliage, stems, and roots UFN of the pine seedlings in the IC treatment, were significantly less than what was observed in the NC treatment indicating that I. cylindrica applied more competitive stress. With increases in competitive stress came reduction in pine seedling growth and fertilizer use efficiency, wh ich has been observed in crops such as cotton (Allen et al. 2004) and maize (Jose et al. 2000b; Wienhold et al. 1995). When quantifying the amount of applied ammoni um sulfate remaining in the soil at the end of the growing season, approximately the same quantity was recovered from both the NC and IC treatments regardless of depth. Up to a soil depth of 60 cm, less than half of what was recovered in these two treatment s was recovered in the VF treatment. Loss of the ammonium sulfate in the VF treatment likely is due to leaching. Loss of fer tilizer in this treatment might explain why approximately 25% of the 15N applied was accounted for in the VF treatment. The greatest amount of 15N was accounted for in the IC treatment nearly 20% more than in the NC treatment. A large perc entage of the applied 15N (81.5%) was traced in the IC treatment. In I. cylindrica invaded plots little N was lost s uggesting tight retention of N. Overall, our results suggest that I. cylindrica is more competitive than native vegetation in an emerging pine forest when it comes to acquiri ng and using N. Species competing with this invasive will experience greater levels of stress than from nati ve species resulting in decreased growth and nutrient use efficiency. Although I. cylindrica maintains lower concentrations of N in its tissues, it is able to retain greater amounts of N th an native vegetation on an ecosystem scale because of much greater amount of biomass. N is tightly retained in invaded areas, with

PAGE 105

105 most of the N being stored belowgro und. Invasion of a native ecosystem by I. cylindrica results a shift in where bulk of the N occurs, fr om above to belowground (Figure 5-6). The shift in N pools resulting from invasion ha s implications for the role fire plays on N availability. On a non-invaded site, the large percentage of N occurring aboveground could be lost in a burn. I. cylindrica invasion represents si gnificant change in location of biomass and N, thus a burn of the aboveground foliage would have li ttle effect on the amount of N. Mack et al. (2001) found that burning had li ttle effect on the total ecosys tem N of Hawaiian woodlands because >95% of the ecosystem N occurred in the soil and only high-intensity burns would result in significant loss of N from the soil pool There have been suggestions that I. cylindrica will burn at greater intensity with higher maximum temperatures a nd heights than native species (Lippincott 2000), but there is no ev idence to suggest that an intense burn would lead to losses of N belowground. Despite the fact that N is re tained belowground on a site by I. cylindrica potentially even in the face of disturbance, it does not seem to be available to other species. Tightly controlling the availability of N may be a mechanism by which I. cylindrica is able to suppr ess the growth of P. taeda seedlings. It has been demonstrated that the pine seedlings ha ve reduced rates of photosynthesis in invaded areas (Cha pter 4), which may be due to reduced N availability. Other than controlling N availability, I. cylindrica may have other mechanisms by which it is able to achieve dominance and reduce the productivity of competing species. The way it is able to manipulate other nutrient cycles should be further explored. Brew er and Cralle (2003) showed that I. cylindrica is a better competitor for phosphorus than native species in a longleaf pine savanna based upon percent ground cover of the co mpeting species. Collins and Jose (2007) demonstrated that I. cylindrica reduces potassium and nitrate in invaded forestlands in the

PAGE 106

106 Florida panhandle. Work should also be done examining I. cylindrica s role in water relations. In chapter 4, we showed evidence that I. cylindrica decreases soil water potential, but this effect on other species has not be examined. Allelopath y has been suggested as a potential mechanism for I. cylindrica to gain dominance over other species (Eussen 1979; Casini et al. 1998; Koger and Bryson 2003), but work has only been done e xploring its effects on crop species. Further research is needed to explore these competitive interactions in detail.

PAGE 107

107 Table 5-1. Summary of competing vegeta tion means (SE) for the native and I. cylindrica treatments. P values for the T-test are pr ovided. Means were cons idered significantly different at = 0.05 level (marked with asterisk). June July August Sept. Overall mean I. cylindrica aboveground biomass (g/m2) 208.0(16.8) 194.9(17.8) 185.8(24.1) 369.9(46.7) 239.7(21.5) Native aboveground biomass (g/m2) 152.3(18.0) 226.2(23.2) 253.2(19.7) 307.7(78.4) 234.8(21.9) P value 0.0533 0.3167 0.0623 0.5145 0.8790 I. cylindrica belowground biomass (g/m2) 902.0(99.9) 595.4(128.6) 927.9(149.4) 2043.7(78.4) 1117.3(94.3) Native belowground biomass (g/m2) 127.8(30.3) 42.6(13.4) 177.0(24.1) 223.6(37.0) 143.7(17.2) P value <0.0001* 0.0027* 0.0011* <0.0001* <0.0001* I. cylindrica % cover 91.3(4.6) 97.2(1.0) 96.5(0.9) 96.4(1.1) 95.4(1.2) Native % cover 70.1(3.2) 80.9(4.3) 88.7(3.6) 87.7(3.3) 81.9(1.8) P value 0.0006* 0.0007* 0.0457* 0.0176* <0.0001*

PAGE 108

108 Table 5-2. Mean nitrogen content (kg/ha) in pine foliage, stem, a nd roots as well as foliage and roots of competing vegetation. Different letters represent sign ificantly different means. Treatment* Pine foliage Pine stem Pine root Competing foliage Competing roots/rhizomes VF 2.05(0.22)a 1.10(0.13)a 1.06(0.13)a NC 0.40(0.09)b 0.19(0.01)b 0.19( 0.03)b 24.35(3.39)a 14.74(1.45)b IC 0.17(0.02)b 0.09(0.01)c 0.04(0.01)c 10.14(0.28)b 32.95(2.23)a p-value <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 *Treatments: VFvegetation free (no competition), NCnative competiton, ICI. cylindrica competition

PAGE 109

109 Table 5-3. Mean percentage of N derived from fertilizer (%NDF) for pine foliage, stem, and roots as well as foliage and roots of comp eting vegetation. Different letters represent significantly different means. Treatment Pine foliage Pine stem Pine root Competing foliage Competing roots/rhizomes VF 4.57(0.33)b 5.1(0.51)b 4.2(0.29)b NC 8.4(1.26)a 8.5(1.28)a 6.6(0.82)a 6.8(1.89)b 3.9(0.69)b IC 11.3(1.12)a 7.7(0.83)ab 7.6(0.76)a 14.7(1.18)a 13.2(1.17)a p-value <0.0001 0.0143 0.0011 0.0016 <0.0001

PAGE 110

110 Table 5-4. Mean percentage utiliz ation of fertilizer N (% UFN) fo r pine foliage, stem, and roots as well as foliage and roots of competi ng vegetation. Different letters represent significantly different means. Treatment Pine foliage Pine stem Pine root Competing foliage Competing roots/rhizomes VF 0.16(0.02)a 0.093(0.010)a 0.074(0.007)a NC 0.054(0.01)b 0.028(0.004)b0.020(0.03)b 2.7(0.9) 0.99(0.11)b IC 0.03(0.004)b 0.012(0.002)c 0.007(0.002)c 2.7(0.23) 7.68(0.68)a p-value <0.0001 <0.0001 <0.0001 0.986 <0.0001

PAGE 111

111 Figure 5-1. Summary of mean mont hly temperature (bars) and tota l monthly precipitation (line) for the study site in 2003. 0 5 10 15 20 25 30 Jan.Feb.Mar.Apr.MayJun.Jul.Aug.Sep.Oct.Nov.Dec.Temperature (C)0 5 10 15 20 25 30 35 40 45Precipitation (cm)

PAGE 112

112 Figure 5-2. Amount of nitrogen in ti ssues (% of total biomass)(SE) of P. taeda seedlings and competing vegetation for the three treatments. Different letters re present a significant difference in % nitrogen. For all five comparisons p<0.0001. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 pine foliagepine stempine rootcompeting foliagecompeting rootsNitrogen in tissue (%) VF NC IC a a b a b b a b c a b a b

PAGE 113

113 Figure 5-3. Percentage of 15N recovery in soil at end of grow ing season in the VF (triangles), NC (squares), and IC (diamonds) treatments. 0 5 10 15 20 25 0306090Depth (cm)

PAGE 114

114 Figure 5-4. Percent total of N recovered at the end of st udy for the three treatments. 0 10 20 30 40 50 60 70 80 90 100 ICNCVF% Total N Soil (up to 90cm) Competin g ve g etation Pine seedling

PAGE 115

115 Figure 5-5. Monthly mean soil moisture of th e three treatments for the summer of 2003. 10 11 12 13 14 15 16 17 18 19MayJuneJulyAugustsoil moisture (volumetric %) IC NC VF

PAGE 116

116 Figure 5-6. Demonstration of the percent of to tal nitrogen occurring in a young emerging pine forest before and after invasion by I. cylindrica ; Nitrogen in P. taeda seedlings in white, competing vegetation foliage in black and competing vegetation roots/rhizomes in grey. Native 37% (10.1 kg N/ha) 61% (24.3 kg N/ha) 2% (0.8 k g N/ha) Imperata cylindrica76% (32.9 kg N/ha) 23% (14.7 kg N/ha) 1% (0.3 k g N/ha)

PAGE 117

117 CHAPTER 6 SUMMARY AND CONCLUSIONS The overall objective of this study was to examine the impacts of Imperata cylindrica on southeastern forests, specifically examining th e role of its invasion on productivity, nutrient cycling and understory diversity of pine forests. The first objec tive was to review the literature for all the proposed mechanisms fo r invasion success as well as so me of the theories that have been proposed on what makes communities susceptible to invasion (Chapter 2). With an understanding of community susc eptibility, the next objective was to test the diversityinvasibility hypothesis, propos ed by Elton (1958), in a mesocosm experiment with I. cylindrica and native pine forest understory species fo cusing on the importance of species richness, functional diversity and species identity (Chapt er 3). Next, using measurements of size, biomass, and gas exchange, the impacts of I. cylindrica on the productivity of establishing pine seedlings in comparison with native species were determined (Chapter 4). The last objective was to quantify the competitive usage of nitrogen (N) between I. cylindrica pine seedlings and native vegetation using stable isot opes (Chapter 5). The following are the major findings. A review of the literature re vealed that in general, there are two avenues for discussions of invasion mechanisms. The first focuses on the community characteristics that lead to invasion, while the other specifica lly addresses invader traits. Factors that play a role in community susceptibility to invasion include su fficient resource use, em pty niches, diversity, invasion facilitators, biotic resistance, and novel enemies. Traits that make a plant species invasive include increased allocation to reprodu ction, better reproductive mechanisms, superior competition for resources, and allelopathy. I nvasion may also occur as the result of a combination of the community and invader prop erties mentioned previously. The array of

PAGE 118

118 hypotheses suggests that a single th eory explaining invasion does not exist and that much work needs to be done to understand invasion mechanisms. Eltons diversity-invas ibility hypothesis, spec ifically the role of species richness, functional diversity and species identity of native species we re tested with I. cylindrica in mesocosms. A randomized block design consisting of eight blocks and ten treatments with five common Florida sandhill understo ry species including a shrub, Ilex glabra two grasses, Aristida beyrichiana and Andropogon virginicus and two forbs, Chamaecrista fasciculata and Pityopsis graminifolia was used The treatments included a control, five monocultures, a grass mix treatment, a forb mix treatment, and a 3-specie s treatment with of a ll three functional groups (consisting of I. glabra A. beyrichiana and C. fasciculate ), and a 5-species treatment with all of the native species. There were five levels of dive rsity, (0, 1, 2, 3 and 5 species). After the native communities were allowed a year to establish, I. cylindrica was introduced at the start of the summer of 2005. Number of shoots and cover of I. cylindrica and cover of the native species were recorded biweekly through August. At the end of the st udy, the plants were harvested, dried and weighed. Prior to drying of the belowg round biomass, root lengt h, root length density (RLD) and specific root length (SRL) were dete rmined. In August, there was a significant negative linear relations hip between the cover of native species and I. cylindrica (r2 = 0.5925, p = 0.0092). There was a negative logarithmic rela tionship between the biomass of the native species and I. cylindrica (r2 = 0.6986, p = 0.0026). There was no relationship between the number of native species and I. cylindrica biomass suggesting that th e diversity-invasibility hypothesis does not explain invasion success of I. cylindrica Grasses proved to be the most resistant functional group provi ding resistance alone and in mixed functional communities. Repeated measures analysis dem onstrated that treatments including Andropogon virginicus were

PAGE 119

119 the most resistant to invasion over time (p < 0.001) suggesting that resistance is a matter of species identity. The success of Andropogon virginicus can be attributed to the fact that it had significantly greater root lengt h (p = 0.0017), RLD (p = 0.0109) and SRL (p < 0.001) than all of the native species and I. cylindrica in monocultures. The same trends were observed of A. virginicus in mixed communities. The root morphology characteristics allowed it to be a great competitor belowground where I. cylindrica was most aggressive. The results suggest that A. virginicus could be used in restor ation of infested ecosystems following chemical and mechanical control of I. cylindrica In Santa Rosa County, Florida, a 27-month study was conducted to compare the impacts of I. cylindrica and native vegetation competition on the productivity of loblolly pine ( Pinus taeda ) seedlings. In March 2003, one-year-old pine seedlings were planted in five plot replications of three treatments: 1) vegetati on free (VF) 2) native (NC) competition and 3) I. cylindrica (IC) competition. At the end of the study, only 26% of the IC seedlings survived, half of what was observed in NC. Nine months af ter planting, the IC seed lings had significantly smaller root collar diameters than the NC seedlings (p<0.000 1) and by November 2004, the heights and stem volume index were significantly less as we ll (p<0.0001). After one full growing season, the NC and IC pine seedlings ha d 21 and 11.5% of the total biomass of the VF seedlings, respectively. The NC and IC seed lings differed significantly in root biomass (p<0.0001). After 27 months, the IC pine seedling total biomass was 2.4% of the VF seedlings and 18% of the NC seedlings (p<0.0001). The greatest difference was in the pine needle biomass with the IC being only 11% of the NC pi ne needle biomass. In the summer of 2003, the IC pine seedlings maintained the lowest levels of light saturated net photosynthesis, which was matched by the lowest levels of stomatal c onductances. These results may be explained by

PAGE 120

120 reduced amounts of foliar nitrogen and so me water stress that result from I. cylindrica competition. The pines in the IC treatment had th e lowest total foliar surface area and the lowest specific leaf area, which may explain the redu ced productivity. Evidence from this study suggests that I. cylindrica competition significantly reduces the productivity and growth of loblolly pine seedlings comp ared to native vegetation. In the next study, 15N-labeled Ammonium Sulfate was used to compare how loblolly pine ( Pinus taeda ) seedlings compete for N in the presence of I. cylindrica and native vegetation competition in Santa Rosa County, Florida. I. cylindrica competition led to smaller pine seedlings with significantly less N content in the pine foliage and roots than those in the native treatment. Competition from I. cylindrica for N contributed to the pine seedlings taking up a greater percentage of the applied fertilizer than the seedlings co mpeting with native vegetation. However, because of their reduced growth they we re less efficient in utilizing the fertilizer. The belowground biomass of the invasive on average wa s seven times higher than the native species. Despite its lower N concentration in foliage and roots, I. cylindrica retained significantly more N per hectare. While the native sp ecies retained more N aboveground, I. cylindrica held significantly more belowground, thus invasion by this grass would l ead to a shift of N pools from above to belowground. I. cylindrica was more competitive than the native species at attaining N because in both its foliage and roots, significan tly more of the applied fertilizer N was found. The roots of I. cylindrica were seven times more efficient in utilizing the fertilizer. The fact that we were able to account for 81.5% of the applied fertilizer in the I. cylindrica plots compared to 62.2% in the native treatment suggests that I. cylindrica tightly retains most of the available N on site making it a particularly good invader.

PAGE 121

121 The results of all these studies indicate the I. cylindrica poses a serious threat to the establishment and productivity of young pine fo rests. This work shows evidence that I.cylindrica is competitive for N and alters N pools by retaining large amounts of N belowground. I. cylindrica s role in N cycling and cycling of nutrients still needs to be explored. Preliminary evidence, in this work, suggests that I. cylindrica may cause water stress. Work should be done exploring how the invasive co mpetes for water and how water its presence affects sap flow of other species. It was demonstrated that community resistance to this invasive plant lies in the hand of speci fic native species, which use similar mechanisms to gain dominance. Native species more similar to I. cylindrica need to be tested to see if they will be more successful at resisting invasion. Bunchgr asses proved to be strong competitors, but it is possible that rhizomatous species may be more su ccessful. Use of these native species should be considered in the integrated pest management strategies against I. cylindrica

PAGE 122

122 APPENDIX MESOCOSM STUDY EXTRAS Table A-1. Final mean % cover for each species in the ten treatments. Treatment % cover Control 0 A. stricta monoculture 66.25 A. virginicus monoculture 89.12 I. glabra monoculture 42.5 C. fasciculata monoculture 26.87 P. graminifolia monoculture 40.62 Grass mix: 1. A. stricta 11.25 2. A. virginicus 76.87 Total 88.12 Forbs mix: 1. C. fasciculata 33.75 2. P. graminifolia 8.12 Total 41.87 3-Species 1. A. stricta 16.25 2. I. glabra 0.12 3. C. fasciculata 33.75 Total 50.12 5-Species 1. A. stricta 10.62 2. A. virginicus 0.5 3. I. glabra 17.86 4. C. fasciculata 15.75 5. P. graminifolia 23.75 Total 68.48

PAGE 123

123 LIST OF REFERENCES Agrawal, A.A., and Kotanen, P.M. 2003. Herbiv ores and the success of exotic plants: a phylogenetically controlled experiment. Ecology Letters 6 : 712-715. Allen, S.C., Jose, S., Nair, P.K.R., Brecke, B.J. and Ramsey, C.L. 2004. Competition for 15Nlabeled fertilizer in a pecan ( Carya illinoensis K.Koch)-cotton ( Gossypium hirsutum L.) alley cropping system in the southern United States. Plant and Soil 263 : 151-164. Allison, S.D. and Vitousek, P.M. 2004. Rapid nutrient cy cling in leaf litter form invasive plants in Hawaii. Oecologia 141 : 612-619. Ashton, I.W., Hyatt, L.A., Howe, K.M., Gurevitc h, J. and Lerdau, M.T. 2005. Invasive species accelerate decomposition and litter nitrogen loss in a mixed deciduous forest. Ecological Applications 15 : 1263-1272. Ayeni, A.O. and Duke, W.B. 1985. The influe nce of rhizome features on subsequent regenerative capacity in speargrass ( Imperata cylindrica (L.) Beauv.). Agricultural Ecosystems and Environment 13 : 309-317. Bakker, J. and Wilson, S. 2001. Competitive abilitie s of introduced and native grasses. Plant Ecology 157 : 117-125. Barber, L.D., Joern, B.C., Volence, J.J. and Cunningham, S.M. 1996. Supplemental nitrogen effects on alfalfa regrowth and ni trogen mobilization from roots. Crop Science 36 : 12171223. Baruch, Z., and Goldstein, G. 1999. Leaf construc tion cost, nutrient concentration, and net CO2 assimilation of native and i nvasive species in Hawaii. Oecologia 121 : 183-192. Blossey, B. and Notzold, R. 1995. Evolution of increased competitive ability in invasive nonindigenous plants: a hypothesis. Journal of Ecology 83 : 887-889. Blumenthal, D. 2005. Interrelated causes of plant invasions. Science 310 : 243-244. Boonitte, A., and Ritdhit, P. 1984. Allelopathic effects of some weeds on mungbean plants ( Vigna radiata ) Proc.1st Tropical Weed Conf., Hat Yai, Songkhia, Thailand Vol. 2 pp. 401-406 Brewer, J.S. and Cralle, S.P. 2003. Phosphorus addition reduces invasion of longleaf pine savanna (Southeastern USA) by a non-indigenous grass ( Imperata cylindrica ). Plant Ecology 167 : 237-245. Britt, J.R., Zutter, B.R., Mitchell, R.J., Gjer stad, D.H., and Dickson, J.F. 1990. Influence of herbaceous interferences on growth and biom ass partitioning in planted loblolly pine ( Pinus taeda ). Weed Science 38 :497-503.

PAGE 124

124 Brook, R.M. 1989. Review of literature on Imperata cylindrica (L.) Raeuschel with particular reference to South East Asia. Trop. Pest Management 35 : 12-25. Brown, R.L. and R.K. Peet. Diversity and i nvasibility of southern Appalachian plant communities. 2003. Ecology 84 : 32-39. Bruno, J.F., Kennedy, C.W., Rand, T.A., and Gran t, M. 2004. Landscape-scale patterns of biological invasions in s horeline plant communities. Oikos 107 :531-540. Bryson, C. T., and R. Carter. 1993. Cogongrass, Imperata cylindrica in the United States. Weed Technology 7 :1005-1009. Buchanan, K.S. 1999. A comparison of techniques for restoring sandhill understory species on abandoned agricultural land (Florida). Ecological Restoration 17 : 235-238. Cain, M.D. 1991. The influence of woody and herbaceous competition on early growth of naturally regenerated loblo lly and shortleaf pines. Southern Journal of Applied Forestry 15 : 179-185. Callaway, R.M. and Aschehoug, E.T. 2000. Invasive pl ants versus their new and old neighbors: a mechanism for exotic invasion. Science 290 : 521-523. Callaway, R.M., Thelen, G.C., Rodriguez, A. a nd Holben, W.E. 2004. Soil biota and exotic plant invasion. Nature 427 : 731-733. Callaway, R.M. and Ridenour, W.M. 2004. Novel w eapons: invasive success and the evolution of increased competitive ability. Frontiers in Ecology and Environment 2 (8): 436-443. Casini, P., V. Vecchio and I. Tamantiti. 1998. Allelopathic interference of switchgrass and cogongrass: Germination and early development of rice. Tropical Agriculture 75 : 445451. Chapin, F.S. III, Reynolds, H., DAntonio, C.M., and Eckhart, V. 1996. The functional role of species in terrestrial ecosystems. Pages 403-430 in Pearcy, R.W., Ehleringer, J., Mooney, H.A., and Rundel, P.W. editors. Plant physiological ecology: field methods and instrumentation. Chapman & Hall, New York, New York, USA. Chikoye, D. and Ekeleme, F. 2001. Cogongrass suppression by intercropping cover crops on corn/cassava systems. Weed Science 49 : 658-667. Colautti, R.I., A. Ricciardi, I.A. Grigorovich and H.J. MacIsaac. 2004. Do es the enemy release hypothesis predict invasion success? Ecology Letters 7 : 721-733. Collins, A.R., Jose, S. and Daneshgar,P. ( in press ) Elton's hypothesis revis ited: an experimental test using cogongrass. Biological Invasions

PAGE 125

125 Collins, A.R. and Jose, S. (In review) Cogongrass invasion alters soil chemical properties of natural and planted forestlands. Crawley, M. J., S. L. Brown, M. S. Heard, a nd G. R. Edwards. 1999. Invasion-resistance in experimental grassland communities: species richness or species identity? Ecology Letters 2 :140-148. D'Antonio, C. M., and P. M. Vitousek. 1992. Bi ological invasions by exotic grasses, the grass/fire cycle, and global change. Annual Review of Ecological Systems 23 :63-87. DAntonio, C.M., Hughes, R.F., Mack, M., Hitchcock, D. and Vitousek, P.M. 1998. The response of native species to removal of i nvasive exotic grasses in seasonally dry Hawaiian woodland. Journal of Vegetation Science 9 : 699-712. Davis, M. A., J. P. Grime, and K. Thompson. 2000. Fluctuating resources in plant communities: a general theory of invasibility. Journal of Ecology 88 :528-534. De Mattos, D. 2000. Citrus response curves to N, P, and K fertilization and N uptake dynamics. Ph.D. Dissertation, University of Florida, Gainesville, Fl, USA. DeWalt, S.J., Denslow, J.S. and Ickles, K. 2004. Natural-enemies release facilitates habitat expansion of invasi ve tropical shrub Clidemia Hirta Ecology 85 (2): 471-483. Dillenburg, L.R., Teramura, A.H., Forseth, I.N ., and Whigham, D.F. 1995. Photosynthetic and biomass allocation responses of Liquidambar styraciflua (Hamamelidaceae) to vine competition. American Journal of Botany 82 (4): 454-461. Dozier, H., J. F. Gaffney, S. K. McDonald, E. R. R. L. Johnson, and D. G. Shilling. 1998. Cogongrass in the United States: Hist ory, Ecology, Impacts and Management. Weed Technology 12 : 737-743. Dukes, J.S. 2001. Biodiversity and inva sibility in grassland microcosms. Oecolgia 126 : 563-568 Dukes, J. S. 2002. Species composition and diversity affect grassland susc eptibility and response to invasion. Ecological Applications 12 : 602-617. Dunnett, C.W. 1980. Pairwise multiple comparisons in the unequal variance case. Journal of the American Statistical Association 75 : 796-800. Ehrenfeld, J.G. 2003. Effects of Exotic Plant I nvasions on Soil Nutrient Cycling Processes. Ecosystems 6 : 503-523 Ehrenfeld, J.G., P. Kourtev, W. Huang. 2001. Change s in soil functions following invasions of exotic understory plants in deciduous forests. Ecological Applications 11: 1287-1300

PAGE 126

126 Ehrenfeld, J.G. 2001. Plant-soil interactions. Pages 689-709 In Levin, S. editor. Encyclopedia of biodiversity. Academic Pre ss. San Diego, California. Elliott, K.J. and White, A.S. 1989. Competitive eff ects of various grasses and forbs on ponderosa pine seedlings. Forest Science 33 : 356-366. Elton, C. S. 1958. The Ecology of Invasions by Animals and Plants. T. Methuen and Co., London. Eussen, J.H.H. 1979. Some competition experiments with alang-alang [ Imperata cylindrica (L.) Beauv.] in replacement series. Oecologia 40 : 351-356. Eussen, J.H.H., and Wirjahardja, S. 1973. Studies of an alang-alang, Imperata cylindrica (L.) Beauv. vegetation. Biotropica Bull. No.6 Fagan, M.E. and Peart, D.R. 2004. Impact of the invasive shrub glossy buckthorn ( Rhamnus frangula L.) on juvenile recruitment by canopy trees. Forest Ecology and Management 194 : 95-107. Fargione, J., Brown, C.S., and Tilman, D. 2003. Community assembly and invasion: an experimental test of neutral vs. niche process. Proceedings of the National Academy of Sciences 100 : 8916-8920. Fargione, J. and Tilman, D. 2005. Diversit y decreases invasion via both sampling and complementarity effects. Ecology Letters 8 : 604-611. Field C., H.A. Mooney. 1986. The photosynthesis-nitroge n relationship in wild plants. In: T. J. Givinish., ed On the Economy of Plant Form and Function. Cambridge University Press. New York 717 pp. Foster, B.L., V.H. Smith, T.L. Dickson, T. Hildebrand. 2002. Invasibility and compositional stability in a grassland community: relations hips to diversity and extrinsic factors. Oilkos 99 : 300-307 Funk, J.L. 2005. Hedychium gardnerianum invasion into Hawaiian montane rainforest: interactions among litter quality, decompositi on rate, and soil nitrogen availability. Biogeochemistry 76 : 441-451. Gorchov, D.L and Trisel, D.E. 2003. Compe titive effects of the invasive shrub, Lonicera maackii (Rubr.) Herder (Caprifoliaceae), on the growth and survival of native tree seedlings. Plant Ecology 166 ; 13-24. Gray, A. 1879. The predominance of pertinacity of weeds. American Journal of Science and the Arts 118 : 161-167. Grime, J.P. 1979. Plant strategies and vegetation processes. Wiley, New York.

PAGE 127

127 Green, T.H., Mitchell, R.J., Paliwal, K.C., Path re, U.V., Zutter, B.R., and Gjerstad, D.H. 1991. Effects of herbaceous weeds on fourth year wa ter relations and gas ex change of loblolly pine. Weed Technology 5 : 753-758. Hanfling, B. and Kollman, J. 2002. An evolutiona ry perspective of biological invasions. TRENDS in Ecology and Evolution 17 (12): 545-546. Henderson D.E. and Jose S. 2005. Producti on physiology of three fast-growing hardwood species along a soil resource gradient. Tree Physiology 25 :1487-1494. Hierro, J.L., Maron, J.L. and Callaway, R.M. 2005. A biogeographical approach to plant invasions: the importance of studying exo tics in their introduced and native range. Journal of Ecology 93 : 5-15. Higgins, S.I., Richardson, D.M., Cowling, R.M. and Trinder-Smith, T.H. 1999. Predicting the landscape-scale distribution of alien plants and their threat to plant diversity. Conservation Biology 13 : 303-313. Hobbs, R.J. 1989. The nature and effects of distur bance relative to inva sions. Pages 389-405 in Drake, J.A., Mooney, H.A., di Castri, F., Gr oves, R.H., Kruger, F.J., Rejmanek, M., and Williamson, M, editors. Biological invasi ons: a global perspective. John Wiley, Chichester, UK. Holm, L. G., D. L. Plucknett, J. V. Pancho, and J. P. Herberger. 1977. The World's Worst Weeds: Distribution and Biology. Univer sity Press of Hawaii, Honolulu, Hawaii, USA. Hooper, D. U. 1998. The role of compleme ntarity and competition in ecosystem responses to variati on in plant diversity. Ecology 68 :704-719. Hooper, D.U, and P.M. Vitousek. 1998 Effects of plant composition and diversity of nutrient cycling. Ecological Monographs 68 :121-149 Hubbard, C.E., D. Brown, A.P. Gray, and R.O. Whyte. 1944. Imperata cylindrica : Taxonomy, Distribution, Economic Significance and Control. Imperial Agricultural Bureau Joint Publication 7 Hughes, R., Vitousek, P.M., and Tunison, T. 1991. Alien grass invasion and fire in the seasonal submontane zone of Hawaii. Ecology 72 : 743-746. Huston, M.A. 1999. Local processes and regional pa tterns: appropriate scal es for understanding variation in the diversity of plants and animals. Oikos 86 : 393. Jose, S., J. Cox, D. L. Miller, D. G. Shilli ng, and S. Merritt. 2002. Alien plant invasions: The story of cogongrass in southeastern Florida. Journal of Forestry 100 :41-44.

PAGE 128

128 Keane, R.M. and Crawley, M.J. 2002. Exotic plan t invasions and the enemy release hypothesis. TRENDS in Ecology and Evolution 17 : 164-170. Kennedy, T. A., S. Naeem, K. M. Howe, J. M. H. Knops, D. Tilman, and P. Reich. 2002. Biodiversity as a barrie r to ecological invasion. Nature 417 :636-638. Knops, J. M. H., D. Tilman, N. M. Haddad, S. Naeem C. E. Mitchell, J. H aarstad, M. E. Ritchie, K. M. Howe, P. B. Reich, E. Siemann, and J. Groth. 1999. Effects of plant species richness on invasion dynamics, disease outbr eaks, insect abundances, and diversity. Ecology Letters 2 :286-293. Koger, C.H. and Bryson, C.T. 2003. Effects of cogongrass ( Imperata cylindrica ) residues on bermudagrass ( Cynodon dactylon ) and Italian ryegrass ( Lolium multiflorum ). Proceedings Southern Weed Science Society 56 : 341 Lambers, H., Chapin III, F.S., and Pons, T.L. editors. 1998. Plant Physiological Ecology. Pages 2-3. Springer-Verlag, New York. Lavorel, S., A.-H. Prieur-Richar d, and K. Grigulis. 1999. Invasibi lity and diversity of plant communities: from patterns to process. Diversity and Distributions 5 :41-49. LeJeune, K.D. and Seastedt, T.R. 2001. Centaurea species: the forb that won the west. Conservation biology 15 (6): 1568-1574. Levine, J.M. 2000. Species divers ity and biological invasions : relating local process to community pattern. Science 288 : 852-854 Levine J.M., C. M. D'Antonio. 1999. Elton revisited: a review of evidence linking diversity and invasibility. Oikos 87 : 15-26. Lewis, P.A., DeLoach, C.J., Knutson, A.E., Tr acy, J.L. and Robbins, T.O. 2003. Biology of Diorhaba elongata deserticola (Coleoptera: Chrysomelidae), an Asian leaf beetle for biological control of saltcedars ( Tamarix spp.) in the United States. Biological Control 27 : 101-116. Lippincott, C. L. 1997. Ecological consequences on Imperata cylindrica (cogongrass) invasion in Florida sandhill. Ph.D. Univer sity of Florida, Gainesville. Lippincott, C. L. 2000. Effects of Imperata cylindrica (cogongrass) invasion on fire regime in Florida sandhill. Natural Areas Journal 20 :140-149. Lonsdale, W.M. 1999. Global patterns of plant invasions and the con cept of invasibility. Ecology 80 : 1522-1536

PAGE 129

129 MacDonald, G.E. 2004. Cogongrass ( Imperata cylindrica )Biology, Ecology and Management. Critical Reviews in Plant Sciences 23 :367-380 McDowell, S.C.L. 2002. Photosynthe tic characteristics of invasive and noninvasive species of Rubus (Rosaceae). American Journal of Botany 89 :1431-1438. McKinney, M.L. 2001. Effects of human population, ar ea, and time on nan-native plant and fish diversity in the United States. Biological Conservation 100 : 243-252. Mack, M.C. and DAntonio, C.M. 2003. Exotic gr asses alter controls over soil nitrogen dynamics in a Hawaiian woodland. Ecological Applications 13 : 154-166. Mack, M.C., DAntonio, C.M., and Ley, R.E. 2001. Alteration of ecosystem nitrogen dynamics by exotic plants: a case study of C4 grasses in Hawaii. Ecological Applications 11 : 13231335. Mack, R. N., D. Simberloff, W. M. Lonsdale H. Evans, M. Clout, and F. A. Bazzaz. 2000. Biotic Invasions: causes, epidemiol ogy, global consequences, and control. Ecological Applications 10 :689-710. Marler, M.J., Zabinksi, C.A., Callaway, R.M. 1999. Mycorrhizae indirectly enhance competitive effects of an invasive forb on a native bunchgrass. Ecology 80 : 1180-1186. Martin, T.A. and Jokela, E.J. 2004. Stand developm ent and production dynamics of loblolly pine under a range of cultural treatments in north-central Florida USA. Forest Ecology and Management 192 : 39-58. Maron, J.L and Vila, M. 2001. When do herbivores affect plant invasion? Evidence for the natural enemies and bitotic resistance hypotheses. Oikos 95 : 361-373. Milbau, A., Nijs, I., De Raedemaeck er, F. et al. 2005. Invasion in grassland gaps: the role of neighbourhood richness, light availability and species complementarity during two successive years. Functional Ecology 19 : 27-37. Miller, J.H. 2000. Refining rates and treatment sequences for cogongrass ( Imperata cylindrica ) control with imazapyr and glyphosate. Proceedings of the Southern Weed Science Society 53 : 131-132. Miller, J.H., Zutter, B.R. Zedaker, S.M., Ed wards, M.B., Haywood, J.D., and Newbold, R.A. 1991. A regional study on the influences of woody and herbaceous competition on early loblolly pine growth. Southern Journal of Applied Forestry 15 : 169-179. Morris, L.A., Moss, S.A., and Garbett, W.S. 1993. Competitive interferen ce between selected herbaceous and woody plants and Pinus taeda L. during two growing seasons following planting. Forest Science 39 : 166-187.

PAGE 130

130 Munger, G.T. Will, R.E. and Borders, B.E. Ef fects of competition control and annual nitrogen fertilization on gas exch ange of different-aged Pinus taeda Canadian Journal of Forest Research 33 : 1076-1083. Naeem, S. and Wright, J.P. 2003. Disentangling biodiversity effects on ecosystem functioning: deriving solutions to a seemingly insurmountable problem. Ecological Letters 6 : 567-579 Naeem, S., J. M. H. Knops, D. Tilman, K. M. Howe, T. Kennedy, and S. Gale. 2000. Plant diversity increases resistance to invasion in th e absence of co varyi ng extrinsic factors. Oikos 91 :97-108. Newman, E.I. 1966. A method for estimating th e total length of root in a sample. Journal of Applied Ecology 3 : 139-145. Parker, J.D., Burkepile, D.E. and Hay, M.E. 2006. Opposing effects of native and exotic herbivores on plant invasions. Science 311 : 1459-1461. Parker, M.A. 2001. Mutualism as a constraint on invasion success for legumes ans rhizobia. Diversity and Distributions 7 : 125-136. Patterson, D.T. 1980. Shading effects on growth a nd partitioning of plant biomass in cogongrass ( Imperata cylidnrica ) from shaded and exposed habitats. Weed Science 28 : 735-740. Perry, M.A. Mitchell, R.J., and Zutter, B.R., Glover, G.R. and Gjerst ad, D.H. 1993. Competitive responses of loblolly pine to gradients in loblolly pine, sweet gum, and broomsedge densities. Canadian Journal of Forest Research 23 : 2049-2058. Pimm, S. L. 1991. The Balance of Nature? Univer sity of Chicago Press, Chicago, Illinois. Prieur-Richard, A. H., and S. Lavorel. 2000. I nvasions: the perspectiv e of diverse plant communities. Austral Ecology 25 : 1-7. Prieur-Richard, A. H., S. Lavorel, K. Gri gulis, and A. Dos Santos. 2000. Plant community diversity and invisibility by exotics: invasion of Me diterranean old fields by Conyza bonariensis and Conyza canadensis Ecological Letters 3 : 412-422. Pysek, P., Jarosik, V., and Kucera, T. 2002. Patterns of invasion in temperate nature reserves. Biological Conservation 104 : 13-24. Ramsey, C.L., S. Jose, D.L. Miller, J. Cox, K.M. Portier, D.G. Shilling, S. Merritt. 2003. Cogongrass [ Imperata cylindrica (L.) Beauv.] response to herbicide and disking on a cutover site in a mid-rotation pi ne plantation in Southern USA. Forest Ecology and Management 179 : 195-209. Reich, P.B., D.S. Ellsworth, M. B. Walters, J. Vose, C. Gresham, J. Volin, W. Bowman. 1999. Generality of leaf traits relati onships: a test across six biomes. Ecology 80 :1955-1969.

PAGE 131

131 Reinhart, K.O. and Callaway, R.M. 2006. Soil biota and invasive plants. New Phytologist 170 : 445-457. Rejmanek, M. 1989. Invasibility of plant communities. Pages 369-388 in Drake, J.A., Mooney, H.A., Castri, F.di, Groves, R.H., Kruger, F. J., Rejamnek, M. and Wiliamson, M. editors. 1989. Biological Invasions. A global perspective. John Wiley and Sons, Chichester, England. Rejmanek, M. 1996. Species richness and resistance to invasions. In Biodiversity and Ecosystem Processes in Tropical Forests, Eds. Oria ns, G.H., Dirzo, R., Cushman, J.H. Berlin: Springer, pp. 153-172. Rejmanek, M. and Richardson, D.M. 1996. What at tributes make some plant species more invasive. Ecology 77 (6): 1655-1661. Ridenour, W.M. and Callaway, R.M. 2001. The relative importance of allelopathy in interference: the effect of an i nvasive weed on a native bunchgrass. Oecologia 126 : 444450. Robinson, G. R., J. F. Quinn, and M. L. Stan ton. 1995. Invasibility of experimental habitat islands in a California winter annual grassland. Ecology 76 :786-794. Roche Jr, B.F. Roche, C.T. and Chapman, R. C. 1994. Impacts of grassland habitat on yellow starthistle ( Centaurea solstitialis L.) invasion. Northwest Science 68 : 86-96. Rose, M. and Hermanutz, L. 2004. Are boreal ecosys tems susceptible to a lien plant invasion? Evidence from protected areas. Oecologia 139 : 467-477. Ruijven, J., G. B. De Deyn, and F. Berendse. 2003. Diversity reduces invasibility in experimental plant communities: the role of plant species. Ecology Letters 6 SAS Institute. 2002. SAS users guide, 9th edn. SAS Institute, Cary, N.C. Sajise, P. E., and J. S. Lales. 1975. Allelopathy in a mixture of cogon ( Imperata cylindrica) and Stylosanthes guyanensis. The Phillipine Journal of Biology 2 :155-164. Sajise, P.E. 1976. Evaluation of cogongrass [ Imperata cylindrica (L.) Beauv.] as a seral stage in Philippine vegetational succession. 1. The cogonal seral stage and plant succession. 2. Autoecological studies on cogon. Dissertati on Abstracts Interna tional B. (1973) 30403041. Weed Abstracts, No. 1339. Shilling, D. G., T. A. Beckwick, J. F. Gaffney, S. K. McDonald, C. A. Chase, and E. R. R. L. Johnson. 1997. Ecology, physiology, and management of Cogongrass ( Imperata cylindrica ). Florida Institute of Phos phate Research, Bartow, FL.

PAGE 132

132 Simberloff, D. and Von Holle, B. 1999. Pos itive interactions of nonindigenous species: invasional meltdown? Biological Invasions 1 : 21-32. Smith, M.D. and Knapp, A.K. 1999. Exotic plan t species in a C4-dominated grassland: invasibility, disturbance and community structure. Oecologia 120 : 605-612. Smith, M., Wilcox, J., Kelly, T. et al. 2004. Dominan ce not richness determines invasibility of tallgrass prairie. Oikos 106 : 253-262. Soerjani, M. 1970. Alang-alang Imperata cylindrica (L.) Beav., pattern of gr owth as related to its problem of control. Biological Tropical Bulletin 1: 88-96. Stachowicz, J.J., Fried, H., Whitlach, R.B., et al. 2002. Reconciling pattern and process in marine bioinvasions: how impor tant is diversity in determ ining community invasibility? Ecology 83 : 2575-2590. Stampe, E.D. and Daehler, C.C. 2003. Mycorrhizal species idientity a ffects plant community structure and invasi on: a microcosm study. Oikos 100 : 362-372. Standish, R.J., Robertson, A.W. and Williiams, P.A. 2001. The impact of an invasive weed Tradescantia fluminensis on native forest regeneration. Journal of Applied Ecology 38 : 1253-1263. Stohlgren, T.J., G.W.Chong, M.A. Kalkhan, L. D. Schell. 1997. Rapid assessment of plant diversity patterns: A methodology for landscapes. Environmental monitoring and assessment 48 : 25-43 Stohlgren, T.J., Barrett, D.T., Kartesz, J.T. 2003. The rich get richer: patterns of plant invasions in the United States. Frontiers in Ecology and the Environment 1 (1): 1114. Stohlgren, T.J., D. Binkley, G.W. Chong, M.A. Kalkhan, L.D. Schell, K.A. Bull, Y. Otsuki, G. Newman, M. Bashkin, and Y. Son. 1999. Exotic plant species invade hot spots of native plant diversity. Ecological Monographs 69 : 25-46 Tennant, D. 1975. A test of a modified line in tersect method for estimating root length. Journal of Ecology 63 : 995-1001. Terry, P.J., G. Adjers, I.O. Akobundo, A.U. An oka, M.E. Drilling, S. Tjitrosemito, M. Utomo. 1997. Herbicides and mechanical control of Imperata cylindrica as a first step in grassland rehabilitation. Agroforesty Systems 36 : 151-179 Thomsen, M.A. and D'Antonio, C.M. 2007. Mechanisms of resistance to i nvasion in a California grassland: the roles of competitor identit y, resource availability, and environmental gradients. Oikos 116 :17-30.

PAGE 133

133 Tilman, D. and Wedin, D. 1991. Plant traits and re source reduction for five grasses growing on a nitrogen gradient. Ecology 72 : 685-700. Tilman, D. 1997. Community invasibility, recr uitment limitation, and grassland biodiversity. Ecology 78 :81-92. Tilman, D. 2004. Niche tradeoffs, neutrality, and community structur e: A stochastic theory of resource competition, invasion, and community assembly. Proceedings of the National Academy of Sciences 101, no 30. Vitousek, P.M., Walker, L.R., Whiteaker, L. D., Mueller-Dombois, D. and Matson, P.A. 1987. Biological invasion by Myrica faya alters ecosystem development in Hawaii. Science 238 : 802-804. Vitousek, P.M. and Walker, L.R. 1989. Biological invasion by Myrica faya in Hawaii: plant demography, nitrogen fixation, ecosystem effects. Ecological Monographs 59 : 247-265 Vitousek, P.M. 1990. Biological invasions and ecos ystem processes: towards and interfration of population biology and ecosystem studies. Oikos 57 : 7-13. Von Holle, B. 2005. Biotic resistance to invader establsihment of a southern Appalachian plant community is determiend by environmental conditions. Journal of Ecology 93 : 16-26. Wienhold B.J., trooien, T.P. and Reichman, G. A. 1995. Yield and nitroge n use efficiency of irrifated corn in ther northern great plains. Agronomy Journal 87 : 842-846. Wilcove, D. S., and L. Y. Chen. 1998. Management costs for endangered species. Cons. Biol. 12 :1405-1407. Wilcove, D.S., Rothstein, D., Dubuw, J., Phillips, A., and Loso, E. 1998. Quantifying threats to imperiled species in the United States. Bioscience 48 : 607-615. Wilcut, J. W., R. R. Dute, B. Truelove, and D. E. Davis. 1988. Factors limiting the distributions of Cogongrass ( Imperata cylindrica ), and Torpedograss ( Panicum repens) Weed Science 36 :49-55. Willard, T.R. 1988. Biology, Ecology, and Management of Cogongrass [ Imperata cylindrica (L.) Beauv.]. Ph.D. dissertation, Universi ty of Florida, Gainesville, Fl. Williamson, M. 1996. Biological Invasions. Chapman & Hall, London. Williamson, M. and Fitter, A. 1996. The varying success of invaders. Ecology 77 (6): 1661-1666. Wiser, S.K., Allen, R.B., Clinton, P.W. and Plat t, K.H. 1998. Community structure and forest invasion by an exotic herb over 23 years. Ecology 79 : 2071-2081.

PAGE 134

134 Witkowski E.T.F. 1991. Effects of invasive a lien acacias on nutrient cy cling in the coastal lowlands of the Cape fynbos. Journal of Applied Ecological 28 : 1-15. Wolf, J.J., S.W. Betty and T.R. Seastedt. 2004. Soil characteristics of Rocky Mountain National Park grasslands invaded by Melilotus officinalis and M. Alba Journal of Biogeography 31 : 415-424.

PAGE 135

135 BIOGRAPHICAL SKETCH Pedram Patrick Daneshgar was born in Akron, Ohio on May 17th, 1979. He was raised in Wilmington, Delaware where he spent a majority of his time outside. His curiosity about all living things led him to study biol ogy at the University of Delawa re, (Newark) where he received his Bachelor of Arts in May 2001. At Saint Jose phs University in Philadelphia, Pennsylvania, he received a Master of Scien ce in biology in May 2003. It was th ere with his mentorship from Dr. Greg Ettl that his interest in ecology grew. Under the adviso rship of Dr. Shibu Jose, Pedram earned his PhD from the School of Forest Reso urces and Conservation at the University of Florida in Gainesville, Florida, studying the im pacts of invasive species. After earning his doctoral degree he attained a pos t-doctoral position studying the ecology of invasive species.