Changing Land-Use from Pinus Elliottii to Eucalyptus Bentamii In

Louisiana State University LSU Digital Commons

LSU Doctoral Dissertations Graduate School

12-12-2018 Changing Land-Use from Pinus Elliottii to Eucalyptus Bentamii in Southwest Louisiana Affects Understory Vegetation Richness, Diversity, and Functional Diversity Patterns Andrea De Stefano Louisiana State University and Agricultural and Mechanical College, [email protected]

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_dissertations Part of the Biodiversity Commons, Forest Management Commons, and the Terrestrial and Aquatic Ecology Commons

Recommended Citation De Stefano, Andrea, "Changing Land-Use from Pinus Elliottii to Eucalyptus Bentamii in Southwest Louisiana Affects Understory Vegetation Richness, Diversity, and Functional Diversity Patterns" (2018). LSU Doctoral Dissertations. 4769. https://digitalcommons.lsu.edu/gradschool_dissertations/4769

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Doctoral Dissertations by an authorized graduate school editor of LSU Digital Commons. For more information, please [email protected]. CHANGING LAND-USE FROM PINUS ELLIOTTII TO EUCALYPTUS BENTAMII IN SOUTHWEST LOUISIANA AFFECTS UNDERSTORY VEGETATION RICHNESS, DIVERSITY, AND FUNCTIONAL DIVERSITY PATTERNS

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

The School of Renewable Natural Resources

by Andrea De Stefano B.S., Università Mediterranea di Reggio Calabria, Italy, 2005 M.S., The Pennsylvania State University, 2014 May 2019 ACKNOWLEDGMENTS

I would like to express my infinite love and gratitude to my family: my parents

Domenico and Elgidia and my brother Daniele. I would never have been able to accomplish anything in my life without your support.

I owe my deepest gratitude to my advisor, Dr. Michael Blazier, who has been an extraordinary mentor during the past years. I would also like to thank Drs. Thomas Dean,

Richard Keim, Christopher Comer, Shaun Tanger, and Thanos Geminis, who served as committee members, providing me their invaluable assistance and support. Very special thanks to Dr. Michael Kaller, Dr. Kyle Harms, Dr. Chris Reid, and Melinda Hughes for all the priceless help and knowledge.

I am extremely grateful to the National Council for Air and Stream Improvement, Inc.

(NCASI), WestRock, and International Paper for funding this project, as well as Hill Farm

Research Station personnel Michelle Moore, Robert Hane, and Kenny Kidd for their irreplaceable help in the field and laboratory. Thanks also to Rice Land and Lumber Company for the access to the sites, and Larson & McGowin, Inc., for logistic support and assistance in the field. Finally, my greater gratitude go my officemates Shannon Kidombo, Cory Garms, and

Lewis Peters for sharing the grad school experience with me.

ii TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... ii

LIST OF TABLES ...... v

LIST OF FIGURES ...... vii

ABSTRACT ...... viii

CHAPTER 1. BACKGROUND ...... 1 1.1. FOREST PLANTATIONS, BIODIVERSITY, AND FUNCTIONAL DIVERSITY ...... 1 1.2. EUCALYPTUS SILVICULTURE ...... 2 1.3. OBJECTIVES ...... 5 1.4. REFERENCES ...... 6

CHAPTER 2. GENERAL METHODS ...... 10 2.1. SITE DESCRIPTION ...... 10 2.2. EXPERIMENTAL DESIGN ...... 13 2.3. VEGETATION SURVEYS ...... 14 2.4. LEAF AREA INDEX AND TOTAL DAILY RADIATION UNDER CANOPY ...... 15 2.5. OVERSTORY CHARACTERISTICS ...... 16 2.6. SOIL SAMPLING AND ANALYSIS ...... 16 2.7. REFERENCES ...... 16

CHAPTER 3. UNDERSTORY VEGETATION RICHNESS AND DIVERSITY OF EUCALYPTUS BENTHAMII AND PINUS ELLIOTTII PLANTATIONS IN THE MID-SOUTH U.S...... 20 3.1. INTRODUCTION ...... 20 3.2. METHODS ...... 22 3.3. RESULTS ...... 26 3.4. DISCUSSION ...... 37 3.5 CONCLUSION ...... 44 3.6. REFERENCES ...... 45

CHAPTER 4. UNDERSTORY VEGETATION FUNCTIONAL TRAITS AND DIVERSITYOF EUCALYPTUS BENTHAMII AND PINUS ELLIOTTII PLANTATIONS IN THE MID-SOUTH U.S...... 52 4.1. INTRODUCTION ...... 52 4.2. METHODS ...... 55 4.3. RESULTS ...... 59 4.4. DISCUSSION ...... 63 4.5. CONCLUSIONS ...... 68

iii 4.6. REFERENCES ...... 69

CHAPTER 5. UNDERSTORY VEGETATION COMMUNITIES AS STRUCTURED BY OVERSTORY, CANOPY, AND SOILS IN EUCALYPTUS BENTHAMII AND PINUS ELLIOTTII PLANTATIONS IN SOUTHWESTERN LOUISIANA ...... 79 5.1. INTRODUCTION ...... 79 5.2. METHODS ...... 81 5.3. RESULTS ...... 86 5.3. DISCUSSION ...... 102 5.4. CONCLUSION ...... 106 5.5. REFERENCES ...... 108

CHAPTER 6. SUMMARY AND CONCLUSION ...... 113

APPENDIX A. UNDERSTORY SPECIES LIST ...... 116

APPENDIX B. DOMINANT UNDERSTORY SPECIES LIST ...... 124

VITA ...... 128

iv LIST OF TABLES

Table 3.1. Number of families, species, life cycle (%), and form (%) of understory vegetation in slash pine plantations planted in 2008 (S08) and 2013 (S13) and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana...... 27

Table 3.2. Test of fixed effects for species richness and Shannon index (H’). Main effects consisted of plantation type (plant) and year (yr) (n=180)...... 30

Table 3.4. Test of fixed effects for DBH, height, and basal area...... 32

Table 3.6. Multiple linear regression stepwise selection procedure, ANOVA table and parameter estimates for mean species richness and Shannon index (H’) (n=60)...... 35

Table 4.2. Multiple linear regression stepwise selection procedure, ANOVA table and parameter estimates for AGB and COV1-4 including all variables as explanatory factors (full models) (n=15)...... 63

Table 5.1 NMDS stress values for each k=1-5 dimension during 2014-2015...... 87

Table 5.2 PERMANOVA results based on NMDS scores for years 2014-2016. P-values relative to post-hoc contrasts between plantation types were adjusted using Bonferroni method...... 100

Table 5.3. Cosine of variables with NMDS axes, R2, P-value, and significance level for the vectors relative to the variables superimposed in the ordination...... 101

v LIST OF FIGURES

Fig. 2.1. Locations of Eucalyptus bethamii (E13), and Pinus elliottii planted in 2013 (S13) and 2008 (S08) sites...... 12

Fig. 2.2. Average and 30-year monthly temperature and precipitation during 2014 (a), 2015 (b) and 2016 (c) for the National Oceanic and Atmospheric Administration Leesville station, LA. . 13

Fig. 3.1. En example Species accumulation curve. The red lines represents the mean species richness (solid) and the 95% confidence intervals (dotted) generated through randomization, while the blue line represents an example of sampling...... 24

Fig. 3.4. Mean and standard error of the mean for species richness (a) and Shannon index in slash pine plantations planted in 2008 (S08) and 2013 (S13), and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana (n=60 plots)...... 30

Fig. 3.5. Leaf area index (LAI) (a) and total daily radiation under the canopy (b) in slash pine plantations planted in 2008 (S08) and 2013 (S13) and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana (n=60 plots) in 2016...... 32

Fig. 3.7. Canonical plot for slash pine plantations planted in 2008 (S08) and 2013 (S13) and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana...... 37

Fig. 4.1. Pearson correlation plot for functional richness (FRic), evenness (FEve), divergence (FDiv), CWMs of understory plant height (PLH), leaf area (LA), specific leaf area (SLA), leaf N concentration (LNC), understory aboveground biomass (AGB), Nudd’s board coverage scores (COV1-4), species richness (Rich), Shannon index (H’), leaf area index (LAI), total daily radiation under the canopy (RAD), diameter at breast high (DBH), and tree height (height)...... 61

Fig. 5.1. Scree plots (stresses vs. number of dimensions) related to NMDS ordination for for years 2014, 2015, and 2016...... 87

Fig. 5.2. NMDS Shepard diagrams for years 2014-2016 ...... 88

Fig. 5.3. Non-metric multidimensional scaling (NMDS) ordination of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R

vi prefix and red color), using Bray-Curtis dissimilarity. Black dots indicate site centroids, while standard deviation 95% ellipses outline position of plantation types. Superimposed vector indicate species richness...... 93

Fig. 5.5. Diameter at breast height (DBH), tree height (Height), and basal area (BA) fitted as smooth surfaces on the 2015 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity...... 95

Fig. 5.7. Diameter at breast height (DBH), tree height (Height), and basal area (BA) fitted as smooth surfaces on the 2016 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity...... 97

Fig. 5.8. Leaf area index (LAI) and total daily radiation under the canopy (RAD) fitted as smooth surfaces on the 2016 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity...... 98

Fig. 5.9. Soil pH, phosphorous (P), potassium (K), sulfur (S), copper (Cu), and sodium (Na) fitted as smooth surfaces on the 2016 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity ...... 99

vii ABSTRACT

In the Western Gulf region of the United States cold-tolerant eucalyptus have been explored as pulpwood feedstock. However, non-native plantations may alter understory species diversity, modifying environmental conditions and soil characteristics. Few studies have compared eucalyptus plantations with other ecosystems to understand the impacts of converting these land uses on understory vegetation in the United States. Three plantations were selected:

(1) slash pine (Pinus elliottii) established in 2008, (2) slash pine established in 2013, and (3) and

Camden white gum (Eucalyptus benthamii) established in 2013. The objectives of this study were to: (1) investigate potential changes in understory species over three years through analyses of rarefaction curves, species richness, and Shannon index, (2) understand potential differences between plantation types in understory univariate functional traits (plant height, leaf size, and specific leaf area, and leaf nitrogen concentrations) and multivariate diversity indices (functional richness, evenness, and divergence), (3) explore understory vegetation dynamics along gradients of overstory, canopy, and soil characteristics. Results indicated a decline in understory species richness over time, with Camden white gum in an intermediate condition between same-age slash pine (highest richness) and older slash pine (lowest richness). No differences in functional indices were detected, suggesting functional redundancy. Differences in functional traits were observed, indicating a taller understory in Camden white gum, with leaves richer in N and larger in area. Non-metric Multidimensional Scaling revealed similarities in understory species composition between Camden white gum and same-age slash pine sites in 2014 and 2015.

Changes caused by gradients of tree dimensions, canopy complexity, radiation availability, and soil characteristics triggered changes in Camden white gum understory. In 2016, understory of

Camden white gum sites was similar to older slash pine ones. Understory vegetation species composition moved towards species with shade tolerance adaptation, but the structural

viii characteristics of Camden white gum allowed species with full sun and partial shade requirements. This study indicated that large-scale land-use change to Eucalyptus benthamii will likely result in declines in understory species and ecosystem functionality in an area that has already experienced a reduction of floristically-rich native ecosystems, such as longleaf pine savannas and woodlands.

CHAPTER 1. BACKGROUND

1.1. FOREST PLANTATIONS, BIODIVERSITY, AND FUNCTIONAL DIVERSITY

Forest plantations can have either positive or negative impacts on plant richness and diversity at different levels (single-tree, stand, or landscape) in relation to the ecological context in which they are found (Brockerhoff et al., 2008; Carnus et al., 2006). Generally, plantations support fewer species than natural forests (Barlow et al., 2007a, 2007b, Brockerhoff et al., 2013,

2008; Carnus et al., 2006; Fork et al., 2015; Makino et al., 2007; Matthews et al., 2002).

However, forest plantations that utilize native tree species might promote the regeneration of native understory vegetation (Abiyu et al., 2011; Brockerhoff et al., 2008; Mason et al., 2005;

Parrotta, 1995; Stallings, 1991), providing important services like critical habitat for endangered species (Arrieta and Suárez, 2006; Brockerhoff et al., 2001; Pejchar et al., 2005), and corridors for wildlife (Hobbs et al., 2003; Lindenmayer and Hobbs, 2004). There are concerns about the suitability of exotic forest species as habitat for indigenous species (Hartley, 2002) and whether exotic species plantations can support species richness is dependent on numerous factors, such as land use of the site prior the establishment of the plantation, conservation and naturalistic values of the site, implementation of diversity management practices, and characteristics of the existent understory vegetation (Brockerhoff et al., 2013).

Changes in ecosystem processes and services might alter biodiversity as consequence of changes in functional diversity (Díaz et al., 2007). The relationships between the taxonomic diversity and productivity of ecosystems is a central topic in ecology (Johnson et al., 1996), and in the last two decades general consensus is growing on the importance of functional traits and their interactions in defining the multiple facets of biological diversity, intended as the variety of forms, functions, and organization of the living organisms and their interactions with the world that surround them, rather than the species number per se (Chapin III et al., 2000; Grime, 1997;

Huston, 1997; Johnson et al., 1996; Loreau, 2000; Schläpfer and Schmid, 1999; Tilman, 1999;

Tilman et al., 1997; Walker et al., 1999). Species richness is often considered a surrogate for functional richness as consequence of increasing niche space coverage as species richness increases in case of random or uniform occupation of niche space (Diaz and Cabido, 2001).

However, neither of these cases are common in nature, and occupation of niche space are likely connected to (1) strong convergence of different species into divergent functional types, or (2) strong separation in niche space among different genotypes, phenotypes, or ontogenetic stages within a single species, with a discrepancy in taxonomic and functional richness. Therefore, taxonomic richness is not always for the same as functional richness. A functional-based approach to biodiversity assessment takes into account the variation in responses to environmental changes and the relative effects on ecosystem processes (such as primary production, nutrient cycling, water dynamics), giving a holistic understanding of ecosystem services (goods and services provided by ecosystem processes to humans such as food, fiber, fuel, water provision, carbon sequestration, recreation) that would not be possible using an approach based exclusively on species richness (Diaz and Cabido, 2001).

1.2. EUCALYPTUS SILVICULTURE

The Eucalyptus genus (Eucalyptus spp.) includes more than 500 species native to

Australia and the bordering islands of Polynesia (González-Orozco et al., 2014), and it is mostly cultivated in large areas of Asia, South America, and Africa (Vose et al., 2015). With more than

70 naturalized species, eucalypts are probably the most widely planted trees in the world (Fork et al., 2015; Simberloff and Rejmánek, 2011) due to their prolific seed production, low incidence of disease, pest resistance, drought tolerance, high biomass productivity, and fast growth rate in soils of relatively low fertility (Booth, 2012; Wear et al., 2015). Eucalyptus plantations provide

several forest products, such as timber, poles, firewood, pulpwood, windbreaks, shelterbelts, ornamentals, essential oil, and honey (Forsyth et al., 2004; Lorentz and Minogue, 2015a, 2015b;

Ritter and Yost, 2009).

Eucalyptus was introduced to the southern United States as early as 1878, while extensive commercial plantations were established in Florida and Texas, respectively, in the 1950’s and

1960’s (Geary et al., 1983). In the early 1970’s, a eucalyptus research cooperative was formed in order to provide financial and research support to the US Department of Agriculture (USDA)

Forest Service; subsequently, four eucalyptus species (Eucalyptus grandis, Eucalyptus robusta,

Eucalyptus camaldulensis, and Eucalyptus tereticornis) were selected and commercial plantations were established in southern Florida (Kellison et al., 2013; Rockwood, 2012). In

1971, the Hardwood Research Cooperative at North Carolina State University (NCSU) initiated a systematic evaluation of species and sources to determine Eucalyptus suitability primarily for the Lower Coastal Plain of the South (Kellison et al., 2013). The Hardwood Cooperative pursued eucalyptus research and development until 1985, when a series of severe freezes occurred during winters of 1983-85 (Kellison et al., 2013. In Florida, eucalyptus planting and research were continued thanks to the efforts of the University of Florida and collaborators.

Environmental factors, in particular sensitivity to cold temperatures, have limited the diffusion of eucalyptus in the United States (Vose et al., 2015). However, field trials with genetically improved and freeze-tolerant eucalyptus varieties show the potential to expand commercial eucalyptus plantations in the zone defined by the USDA as “Plant Hardiness Zones

8b” (annual minimum temperature 9.4° C) (Gonzalez et al., 2011; Hinchee et al., 2011) and greater (Hinchee et al., 2011). This research renewed attention to the potential of eucalyptus to significantly boost hardwood fiber production in many areas of the southern United States

(Gonzalez et al., 2011; Vose et al., 2015). Among cold-resistant eucalyptus species, Eucalyptus

benthamii Maiden & Cambage raised interest for its suitability to produce biomass for industrial activities (Baccarin et al., 2015). Eucalyptus benthamii is native to Camden (western part of

Sidney), Australia, where it occurs in rows on alluvial plains on sand or loam over clay along the

Nepean River and its tributaries (Benson and McDougall, 1998; Benson, 1985). Due to its resistance to low temperatures and its high growth productivity, Eucalyptus benthamii (Camden white gum) plantations have been successfully established in diverse regions such as Brazil,

Argentina, Uruguay, Chile, China, South Africa, and United States (Baccarin et al., 2015; Lin et al., 2003; Shifflett et al., 2014).

In 2007, the MeadWestvaco company began an operational-scale experiment in cultivating eucalyptus plantations in southwestern Louisiana and southeastern Texas (Blazier et al., 2010). The eucalyptus plantations were desired as supplemental sources of hardwood fiber for their paper mill, but it was recognized that these plantations could have deleterious effects on non-timber forest attributes. From 2010-2016 around 8,093 ha (20,000 acres) of land were procured to lease for 10-year rotations. In Louisiana, eucalyptus plantations were predominantly on former slash pine plantations managed by Larson McGowin for the Rice University

Foundation. Although eucalyptus plantations have been in more southerly portions of the

Southeast, particularly Florida, for decades, the MeadWestvaco plantations were the first at operational scale in the Western Gulf region. Wear et al. (2015) recently forecast much of the southern portions of the Southeast (with the region in which the MeadWestvaco eucalyptus project was conducted being characterized as a high-productivity area) as being potentially planted in freeze-tolerant eucalyptus due to advances in eucalyptus breeding and its potential economic benefits relative to native species plantations. Due to their operational scale and their recent conversion from native species plantations, the MeadWestvaco eucalyptus plantation

project provided an opportunity to explore the impacts of eucalyptus plantation conversion on non-timber vegetation diversity, productivity, and structure.

1.3. OBJECTIVES

Understanding the consequences of exotic forest plantations on understory vegetation communities is a critical issue for forest managers to ensure wood and forest products supply while maintaining biodiversity and ecosystem functionalities. Large-scale conversion to

Eucalyptus benthamii on sites formerly occupied by slash pine plantations could have important implications on non-timber vegetation attributes. Stand, site, and vegetation factors that govern non-timber vegetation responses to conversion from slash pine to eucalyptus must be understood to better guide decision-making about these exotic plantations. Therefore, the specific objectives of this study are:

• To investigate changes in understory species communities in Camden white gum and

slash pine plantations over a period of three years in terms of total number and evenness

of species in relation to site factors that may govern these parameters (Chapter 3).

• To understand potential differences in understory vegetation functional diversity and

functional traits and their relationships to species richness and understory biomass and

structure between the two plantation types in relation to understory characteristics

(Chapter 4).

• To explore the understory vegetation dynamics in relation to gradients of overstory,

canopy, and soil characteristics (Chapter 5).

1.4. REFERENCES

Abiyu, A., Lemenih, M., Gratzer, G., Aerts, R., Teketay, D., Glatzel, G., 2011. Status of native woody species diversity and soil characteristics in an exclosure and in plantations of Eucalyptus globulus and Cupressus lusitanica in northern Ethiopia. Mt. Res. Dev. 31, 144– 152. https://doi.org/10.1659/MRD-JOURNAL-D-10-00116.1

Arrieta, S., Suárez, F., 2006. Scots pine (Pinus sylvestris L.) plantations contribute to the regeneration of holly (Ilex aquifolium L.) in mediterranean central Spain. Eur. J. For. Res. 125, 271–279. https://doi.org/10.1007/s10342-006-0121-y

Baccarin, F.J.B., Brondani, G.E., Almeida, L.V. de, Vieira, I.G., Oliveira, L.S. de, Almeida, M. de, 2015. Vegetative rescue and cloning of Eucalyptus benthamii selected adult trees. New For. 46, 465–483. https://doi.org/10.1007/s11056-015-9472-x

Barlow, J., Gardner, T.A., Araujo, I.S., Ávila-Pires, T.C., Bonaldo, A.B., Costa, J.E., Esposito, M.C., Ferreira, L. V, Hawes, J., Hernandez, M.I.M., Hoogmoed, M.S., Leite, R.N., Lo- Man-Hung, N.F., Malcolm, J.R., Martins, M.B., Mestre, L.A.M., Miranda-Santos, R., Nunes-Gutjahr, A.L., Overal, W.L., Parry, L., Peters, S.L., Ribeiro Junior, M.A., Silva, M.N.F. da, Motta, C. da S., Peres, C.A., 2007a. Quantifying the biodiversity value of tropical primary, secondary, and plantation forests. Proc. Natl. Acad. Sci. U. S. A. 104, 18555–18560. https://doi.org/10.1073/pnas.0703333104

Barlow, J., Gardner, T.A., Ferreira, L. V, Peres, C.A., 2007b. Litter fall and decomposition in primary, secondary and plantation forests in the Brazilian Amazon. For. Ecol. Manage. 247, 91–97. https://doi.org/10.1016/j.foreco.2007.04.017

Benson, D., McDougall, L., 1998. Ecology of Sydney plant species Part 6: Dicotyledon family Myrtaceae in Cunninghamia. 5 (4): 808-983. Natl. Herb. New South Wales, R. Bot. Gard. Sydney.

Benson, D.H., 1985. Aspects of the ecology of a rare tree species, Eucalyptus benthamii, at Bents Basin, Wallacia. Cunninghamia 1, 371–383.

Blazier, M.A., Taylor, E.L., Holley, A.G., 2010. Eucalyptus plantations: an emerging management option in the Southeast. Tree Farmer 29, 17–18.

Booth, T.H., 2012. Eucalypts and their potential for invasiveness particularly in frost-prone regions. Int. J. For. Res. 2012, Article ID 837165.

Brockerhoff, E.G., Ecroyd, C.E., Langer, E.R., 2001. Biodiversity in New Zealand plantation forests: policy trends, incentives, and the state of our knowledge. New Zeal. J. For. 46, 31– 37.

Brockerhoff, E.G., Jactel, H., Parrotta, J.A., Ferraz, S.F.B., 2013. Role of eucalypt and other planted forests in biodiversity conservation and the provision of biodiversity-related ecosystem services. For. Ecol. Manage. 301, 43–50. https://doi.org/10.1016/j.foreco.2012.09.018

Brockerhoff, E.G., Jactel, H., Parrotta, J.A., Quine, C.P., Sayer, J., 2008. Plantation forests and biodiversity: oxymoron or opportunity? Biodivers. Conserv. 17, 925–951. https://doi.org/10.1007/s10531-008-9380-x

Carnus, J.M., Parrotta, J., Brockerhoff, E., Arbez, M., Jactel, H., Kremer, A., Lamb, D., O’Hara, K., Walters, B., 2006. Planted forests and biodiversity. J. For. 104, 65–77.

Chapin III, F.S., Zavaleta, E.S., Eviner, V.T., Naylor, R.L., Vitousek, P.M., Reynolds, H.L., Hooper, D.U., Lavorel, S., Sala, O.E., Hobbie, S.E., Mack, M.C., Diaz, S., 2000. Consequences of changing biodiversity. Nature 405, 234.

Diaz, S., Cabido, M., 2001. Vive la difference: plant functional diversity matters to ecosystem processes. Trends Ecol. Evol. 16, 646–655. https://doi.org/10.1016/s0169-5347(01)02283-2

Díaz, S., Lavorel, S., Bello, F. de, Quétier, F., Grigulis, K., Robson, T.M., 2007. Incorporating plant functional diversity effects in ecosystem service assessments. Proc. Natl. Acad. Sci. U. S. A. 104, 20684–20689. https://doi.org/10.1073/pnas.0704716104

Fork, S., Woolfolk, A., Akhavan, A., Dyke, E. van, Murphy, S., Candiloro, B., Newberry, T., Schreibman, S., Salisbury, J., Wasson, K., 2015. Biodiversity effects and rates of spread of nonnative eucalypt woodlands in central California. Ecol. Appl. 25, 2306–2319. https://doi.org/10.1890/14-1943.1

Forsyth, G.G., Richardson, D.M., Brown, P.J., Wilgen, B.W. van, 2004. A rapid assessment of the invasive status of Eucalyptus species in two South African provinces. S. Afr. J. Sci. 100, 75–77.

Geary, T.F., Meskimen, G.F., Franklin, E.C., 1983. Growing eucalypts in Florida for industrial wood production. Gen. Tech. Report, Southeast. For. Exp. Station. USDA For. Serv. iii + 43 pp.

González-Orozco, C.E., Thornhill, A.H., Knerr, N., Laffan, S., Miller, J.T., 2014. Biogeographical regions and phytogeography of the eucalypts. Divers. Distrib. 20, 46–58. https://doi.org/10.1111/ddi.12129

Gonzalez, R., Treasure, T., Wright, J., Saloni, D., Phillips, R., Abt, R., Jameel, H., 2011. Exploring the potential of Eucalyptus for energy production in the Southern United States: financial analysis of delivered biomass. Part I. Biomass and Bioenergy 35, 755–766. https://doi.org/10.1016/j.biombioe.2010.10.011

Grime, J.P., 1997. Biodiversity and Ecosystem Function: The Debate Deepens. Science (80-. ). 277, 1260–1261.

Hartley, M.J., 2002. Rationale and methods for conserving biodiversity in plantation forests. For. Ecol. Manage. 155, 81–95. https://doi.org/10.1016/S0378-1127(01)00549-7

Hinchee, M., Zhang, C., Chang, S., Cunningham, Mi., Hammond, W., Nehra, N., 2011. Biotech Eucalyptus can sustainably address society’s need for wood: the example of freeze tolerant Eucalyptus in the southeastern U.S. BMC Proc. 5, I24–I24. https://doi.org/10.1186/1753-

6561-5-S7-I24

Hobbs, R., Catling, P.C., Wombey, J.C., Clayton, M., Atkins, L., Reid, A., 2003. Faunal use of bluegum (Eucalyptus globulus) plantations in southwestern Australia. Agrofor. Syst. 58, 195–212. https://doi.org/10.1023/A:1026073906512

Huston, A.M., 1997. Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversity. Oecologia 110, 449–460. https://doi.org/10.1007/s004420050180

Johnson, K.H., Vogt, K.A., Clark, H.J., Schmitz, O.J., Vogt, D.J., 1996. Biodiversity and the productivity and stability of ecosystems. Trends Ecol. Evol. 11, 372–377. https://doi.org/http://dx.doi.org/10.1016/0169-5347(96)10040-9

Kellison, R.C., Lea, R., Marsh, P., 2013. Introduction of Eucalyptus spp. into the United States with special emphasis on the southern United States. Int. J. For. Res. 2013, Article ID 189393.

Lin, M., Arnold, R., Li, B., Yang, M., 2003. Selection of cold-tolerant eucalypts for Hunan Province. Eucalypts Asia.

Lindenmayer, D.B., Hobbs, R.J., 2004. Fauna conservation in Australian plantation forests - a review. Biol. Conserv. 119, 151–168. https://doi.org/10.1016/j.biocon.2003.10.028

Loreau, M., 2000. Biodiversity and Ecosystem Functioning: Recent Theoretical Advances. Oikos 91, 3–17.

Lorentz, K.A., Minogue, P.J., 2015a. Exotic Eucalyptus plantations in the southeastern US: risk assessment, management and policy approaches. Biol. Invasions 17, 1581–1593. https://doi.org/10.1007/s10530-015-0844-0

Lorentz, K.A., Minogue, P.J., 2015b. Potential invasiveness for Eucalyptus species in Florida. Invasive Plant Sci. Manag. 8, 90–97. https://doi.org/10.1614/IPSM-D-14-00030.1

Makino, S., Goto, H., Hasegawa, M., Okabe, K., Tanaka, H., Inoue, T., Okochi, I., 2007. Degradation of longicorn beetle (Coleoptera, Cerambycidae, Disteniidae) fauna caused by conversion from broad-leaved to man-made conifer stands of Cryptomeria japonica (Taxodiaceae) in central Japan. Ecol. Res. 22, 372–381. https://doi.org/10.1007/s11284- 007-0359-y

Mason, N.W.H., Mouillot, D., Lee, W.G., Wilson, J.B., 2005. Functional richness, functional evenness and functional divergence: the primary components of functional diversity. Oikos 111, 112–118. https://doi.org/10.1111/j.0030-1299.2005.13886.x

Matthews, S., O’Connor, R., Plantinga, A.J., 2002. Quantifying the impacts on biodiversity of policies for carbon sequestration in forests. Ecol. Econ. 40, 71–87. https://doi.org/10.1016/S0921-8009(01)00269-5

Parrotta, J.A., 1995. Influence of overstory composition on understory colonization by native species in plantations on a degraded tropical site. J. Veg. Sci. 6, 627–636. https://doi.org/10.2307/3236433 8

Pejchar, L., Holl, K.D., Lockwood, J.L., 2005. Hawaiian honeycreeper home range size varies with habitat: implications for native Acacia koa forestry. Ecol. Appl. 15, 1053–1061. https://doi.org/10.1890/04-0577

Ritter, M., Yost, J., 2009. Diversity, reproduction, and potential for invasiveness of Eucalyptus in California. Madroño 56, 155–167. https://doi.org/10.3120/0024-9637-56.3.155

Rockwood, D.L., 2012. History and status of Eucalyptus improvement in Florida. Int. J. For. Res. 2012, Article ID 607879.

Schläpfer, F., Schmid, B., 1999. Ecosystem effects of biodiversity: a classification of hypotheses and exploration of empirical results. Ecol. Appl. 9, 893–912. https://doi.org/10.1890/1051- 0761(1999)009[0893:EEOBAC]2.0.CO;2

Shifflett, S.D., Hazel, D.W., Frederick, D.J., Nichols, E.G., 2014. Species Trials of Short Rotation Woody Crops on Two Wastewater Application Sites in North Carolina, USA. BioEnergy Res. 7, 157–173. https://doi.org/10.1007/s12155-013-9351-2

Simberloff, D., Rejmánek, M., 2011. Encyclopedia of Biological Invasions, 1st ed. University of California Press, Berkeley.

Stallings, J.R., 1991. The importance of understorey on wildlife in a Brazilian eucalypt plantation. Rev. Bras. Zool. 7, 267–276. https://doi.org/10.1590/S0101- 81751990000300008

Tilman, D., 1999. The ecological consequences of changes in biodiversity: a search for general principles. Ecology 80, 1455–1474. https://doi.org/10.1890/0012- 9658(1999)080[1455:TECOCI]2.0.CO;2

Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M., Siemann, E., 1997. The influence of functional diversity and composition on ecosystem processes. Science (80-. ). 277, 1300– 1302. https://doi.org/10.1126/science.277.5330.1300

Vose, J.M., Miniat, C.F., Sun, G., Caldwell, P. V, 2015. Potential implications for expansion of freeze-tolerant Eucalyptus plantations on water resources in the southern United States. For. Sci. 61, 509–521. https://doi.org/10.5849/forsci.14-087

Walker, B., Kinzig, A., Langridge, J., 1999. Plant Attribute Diversity, Resilience, and Ecosystem Function: The Nature and Significance of Dominant and Minor Species. Ecosystems 2, 95– 113.

Wear, D.N., Dixon IV, E., Abt, R.C., Singh, N., 2015. Projecting potential adoption of genetically engineered freeze-tolerant Eucalyptus in the United States. For. Sci. 61, 466– 480. https://doi.org/10.5849/forsci.14-089

CHAPTER 2. GENERAL METHODS

2.1. SITE DESCRIPTION

The study was conducted in forest plantations in the area surrounding the town of

Merryville, Louisiana, USA (30°45′14″N, 93°32′13″W), shown in Fig. 2.1. Topography of the stands was quite uniform, relatively flat, and with mean elevation of 47 m above sea level.

Average monthly minimum and maximum air temperatures for the study region are 12°C and

26°C respectively, and average annual temperature is 19°C. Annual rainfall reaches 1412 mm.

Climate of the site is defined as humid subtropical subtype Cfa – humid temperate with uniform rainfall, under Köppen classification (Kottek et al., 2006). All climatic data were obtained through the National Oceanic and Atmospheric Administration website (http://www.noaa.gov/) for the Leesville, Louisiana station (8.5 m above sea level, 31°8′37″ N, 93°16′16″W) (Fig. 2.2).

Soils of the study area belong to Ultisol (73% of the sites) and Alfisol (27%) orders, and they are classified as Udults, Udalfs, Aqualfs, and Fluvents, according to USDA Soil Taxonomy (Soil

Survey Staff, 2014; USDA, 2002).

The study locations were within West Gulf Coastal Plain Wet Longleaf Pine Savanna and

Flatwoods ecotypes (Holcomb et al., 2015). These ecotypes are floristically-rich, herb-dominated wetlands that are naturally sparsely stocked with longleaf pine (Pinus palustris). More than 140 species of vascular plants have been observed in a 1,000-m2 area, and more than 40 species m-2 have been recorded in many longleaf pine communities, including a large number of endemic species (Peet and Allard, 1993). Pine savannas are characterized by high fluctuations of water tables, ranging from surface saturation/shallow flooding in late fall/winter/early spring to growing season drought. Soils are generally acidic, nutrient-poor, fine sandy loams and silt loams low in organic matter. Fires play a major role in controlling species and community structure. Without frequent fire, shrubs and trees will become predominant in these ecotypes, and

they eventually take over the herbaceous flora. Moreover, the establishment of pulp mills during the 1950s created high demand for smaller trees, which accelerated the conversion of naturally- regenerated longleaf pine forests into loblolly pine and slash pine plantations, resulting in a continuous decline of the area occupied by longleaf pine ecosystems (Brockway et al., 2005).

For Camden white gum silviculture in the area, Blazier et al. (2015) listed the following management operations for the first year: (1) pre-planting broadcast herbicide applications to suppress hardwood species, (2) pre-mechanical site preparation involving of shearing, raking and piling of standing trees and logging debris, (3) mechanical site preparation consisting of bedding and subsoil plow in case of root-restrictive soil texture, (4) manual planting of seedlings prior to the first freeze events of the year, (5) banded P fertilizer applications, and (6) pre- and post- emergence herbicide applications. During the second and the third years after planting, herbicide applications are applied if necessary to refine competing vegetation control, and banded NP and

NPK are applied during the second and third years respectively. Due to the nature of the present study, all herbicide applications were ceased after the first year. At mid-rotation (age 3), all eucalyptus plantations in this study were operationally fertilized via aerial broadcast application of urea and diammonium phosphate (57 kg ha-1 of DAP and 57 kg ha-1 urea).

Precipitation in 2014 was high relative to the 30-year (long-term) average in summer months, and drier in winter period (Fig 2.2). In 2015, precipitation was higher than the long-term average in the spring; lower in late summer and early fall, and higher in late fall. Precipitation patterns of 2016 were relatively wet in spring through summer (with the exception of July).

Temperature seemed to deviate less from the long-term average than precipitation. In 2014, temperatures were similar to those of the 30-year average, with the greatest deviation in January.

Temperatures in 2015 were often slightly above the 30-year average. In 2016, temperatures continued to be slightly above the long-term average for the region.

Fig. 2.1. Locations of Eucalyptus bethamii (E13), and Pinus elliottii planted in 2013 (S13) and 2008 (S08) sites.

Fig. 2.2. Average and 30-year monthly temperature and precipitation during 2014 (a), 2015 (b) and 2016 (c) for the National Oceanic and Atmospheric Administration Leesville station, LA.

2.2. EXPERIMENTAL DESIGN

We selected three different plantation types: (1) slash pine plantation established in 2008

(S08); slash pine plantation established in 2013 (S13); (3) and Camden white gum plantation established in 2013 (E13). For each plantation type, we identified five study stands (15 study sites in total) from among those available on lands owned by cooperating landowners. Within the region in which this study was conducted, Camden white gum plantations were often established on sites formerly managed as slash pine plantations, and slash pine plantations were a predominant land use type surrounding the Camden white gum plantations. Slash pine

plantations established in 2013 and 2008 were similar in age and height (at the beginning of the study), respectively, to the Camden white gum plantations.

At each study site, we established four 6 m × 50 m plots along a transect in the center of the stand with plots separated from each other and plantation edges by >50 m. This protocol was adapted from characteristics of Forest Inventory and Analysis and Plant Ecology Laboratory methods (Beals and Cottam, 1960; Curtis, 1959; Johnson et al., 2008). Each transect was designed to start at a random point and to stretch along a random azimuth that was restricted to keep the transect within the site (Messick, 2016). To reduce edge effect, transects were buffered by 50 meters from the site edges (Jensen and Finck, 2004; Kaiser and Lindell, 2007).

2.3. VEGETATION SURVEYS

The quadrat method (Kent and Coker, 1992) was used to conduct understory vegetation surveys within the 6 × 50 m plots. Ten vegetation sampling points each separated by 5 m were established along the center of each plot. A 1 m × 1 m frame was placed at each of the 10 sampling points per plot, and species present were recorded (Gotelli and Colwell, 2010). This method has been employed in studies of understory species richness and diversity of E. globulus plantations in Portugal (Carneiro et al., 2008, 2007). Two floristic surveys (one in spring and one in fall) were conducted each year from 2014 to 2016, and all the species were identified and recorded at each quadrat. These surveys began in spring 2014 because all herbicide treatments within the Camden white gum plantations had ceased in late fall 2013 consistent with operational silvicultural protocols for these plantations, allowing understory vegetation to become established in those plantations. Species were also classified according their growth form and life cycle duration as listed in the USDA PLANTS Database (https://plants.usda.gov/).

2.4. LEAF AREA INDEX AND TOTAL DAILY RADIATION UNDER CANOPY

Light availability and intensity have critical effects on understory vegetation composition and abundance influencing temperature and humidity of the site (Barbier et al., 2008; Bartemucci et al., 2006; Battles et al., 2001; Lemenih et al., 2004; Parrotta, 1995; Yirdaw and Luukkanen,

2004). Light radiation in the transmittance depends on the overstory characteristics, which can be synthetically described using leaf area index (Barbier et al., 2008). We measured leaf area index

(LAI) (Walter, 2009) using a hemispherical (fisheye) lens placed under the canopy (Anderson,

1964; Jonckheere et al., 2004), with a 180° field of view, such that all sky directions were simultaneously visible (Hill, 1924; Rich, 1990). Leaf area index was then mathematically derived from inversion of gap fraction (Walter, 2009), with LAI being proportional to the natural logarithm of gap fraction (Chianucci et al., 2015). WinSCANOPY software (Regent Instruments,

Ste-Foy, Quebec) was used to analyze hemispherical pictures and calculate LAI and total (direct

+ diffuse) daily radiation under the canopy.

Hemispherical photos were taken in August-September 2016 (a period of seasonal maximum leaf area) using a Nikon Coolpix 4500 camera equipped with Nikon FC-E8 fish-eye converter lens. The camera was placed on a levelled tripod at 147 cm of height from the ground.

Images were collected at maximum resolution (3,871,488 pixels total), pointing the levelled camera directly upwards at zenith. Five pictures per plot were taken in E13 and S08 stands during early morning or late afternoon to avoid direct sunlight. The picture sampling locations corresponded to five of the sampling locations used for vegetation sampling. In S13 stands, one picture per plot (at the vegetation sampling point at the center of the plot) was taken due to relatively uniform overstory and open-canopy conditions (which fosters direct sunlight in images until a narrow timeframe in early morning and late afternoon).

2.5. OVERSTORY CHARACTERISTICS

Tree within each plot, tree heights, and diameters were measured three times: February

2015, January 2016, and December 2016. Due to frequent rain and labor availability, part of the

2015 measures were taken in April 2015. Diameter at breast height (DBH) was measured on all trees with height >1.30 m and diameter >1.2 cm. DBH measurements were used to calculate basal area (m2 ha-1). Total tree height was measured using a height pole for S13 plots and a

Haglof Vertex IV hypsometer (Haglof Inc., Madison, MS, USA) for S08 and E13 plots.

2.6. SOIL SAMPLING AND ANALYSIS

Differences in understory diversity and composition are often a consequence of differences in topsoil (Barbier et al., 2008). Five composite soil samples per plot (2.5-cm diameter cores taken to a 30-cm depth) were collected in proximity of vegetation sampling quadrats (numbers 1, 3, 5, 7, and 9) during August 2016. The soil subsamples were pooled into one composite sample per plot, yielding four subsamples per site and five replicates per plantation type. Soil samples were immediately placed on ice packs and sent to the laboratory in a cooler. All soil samples were oven-dried (40°C), passed through a 2-mm sieve, and sent to the laboratory for chemical analysis, including pH, Mehlich 3 extractable nutrients (P, K, Ca, Mg,

Na, S, Cu, and Zn), total C, and total N (Allen and Schlesinger, 2004; Blazier et al., 2005; Van der Heijde et al., 2008; Zhao et al., 2014).

2.7. REFERENCES

Allen, A.S., Schlesinger, W.H., 2004. Nutrient limitations to soil microbial biomass and activity in loblolly pine forests. Soil Biol. Biochem. 36, 581–589. https://doi.org/10.1016/j.soilbio.2003.12.002

Anderson, M.C., 1964. Studies of the Woodland Light Climate: I. The Photographic Computation of Light Conditions. J. Ecol. 52, 27–41. https://doi.org/10.2307/2257780

Barbier, S., Gosselin, F., Balandier, P., 2008. Influence of tree species on understory vegetation diversity and mechanisms involved - A critical review for temperate and boreal forests. For. Ecol. Manage. 254, 1–15. https://doi.org/10.1016/j.foreco.2007.09.038

Bartemucci, P., Messier, C., Canham, C.D., 2006. Overstory influences on light attenuation patterns and understory plant community diversity and composition in southern boreal forests of Quebec. Can. J. For. Res. Can. Rech. For. 36, 2065–2079. https://doi.org/10.1139/x06-088

Battles, J.J., Shlisky, A.J., Barrett, R.H., Heald, R.C., Allen-Diaz, B.H., 2001. The effects of forest management on plant species diversity in a Sierran conifer forest. For. Ecol. Manage. 146, 211–222. https://doi.org/10.1016/s0378-1127(00)00463-1

Beals, E.W., Cottam, G., 1960. The Forest Vegetation of the Apostle Islands, Wisconsin. Ecology 41, 743–751. https://doi.org/10.2307/1931808

Blazier, M.A., Hennessey, T.C., Deng, S., 2005. Effects of Fertilization and Vegetation Control on Microbial Biomass Carbon and Dehydrogenase Activity in a Juvenile. For. Sci. 51, 449– 459.

Blazier, M.A., Johnson, J., Jernigan, P., Taylor, E., Tanger, S., 2015. Management guidelines for establishing eucalyptus plantations in Western Gulf states. Louisiana State Univ. Agric. Cent.

Blazier, M.A., Taylor, E.L., Holley, A.G., 2010. Eucalyptus plantations: an emerging management option in the Southeast. Tree Farmer 29, 17–18.

Brockway, D.G., Outcalt, K.W., Tomczak, D.J., Johnson, E.E., 2005. Restoration of Longleaf Pine Ecosystems, Gen. Tech. Rep. SRS-83. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 34 p

Carneiro, M., Fabião, A., Martins, M.C., Cerveira, C., Santos, C., Nogueira, C., Lousã, M., Hilário, L., Fabião, A., Abrantes, M., Madeira, M., 2007. Species richness and biomass of understory vegetation in a Eucalyptus globulus Labill. coppice as affected by slash management. Eur. J. For. Res. 126, 475–480. https://doi.org/10.1007/s10342-006-0143-5

Carneiro, M., Fabião, A., Martins, M.C., Fabião, A., Abrantes da Silva, M., Hilário, L., Lousã, M., Madeira, M., 2008. Effects of harrowing and fertilisation on understory vegetation and timber production of a Eucalyptus globulus Labill. plantation in Central Portugal. For. Ecol. Manage. 255, 591–597. https://doi.org/http://dx.doi.org/10.1016/j.foreco.2007.09.028

Chianucci, F., Macfarlane, C., Pisek, J., Cutini, A., Casa, R., 2015. Estimation of foliage clumping from the LAI-2000 Plant Canopy Analyzer: effect of view caps. Trees 29, 355– 366. https://doi.org/10.1007/s00468-014-1115-x

Curtis, J.T., 1959. The vegetation of Wisconsin. University of Wisconsin. University of Wisconsin Press, Madison, WI.

Gotelli, N.J., Colwell, R.K., 2010. Estimating species richness, Biological diversity: frontiers in

measurement and assessment. A. E. Magurran and B. J. McGill, eds. Oxford University Press, Oxford, UK.

Hill, R., 1924. A lens for whole sky photographs. Q. J. R. Meteorol. Soc. 50, 227–235. https://doi.org/10.1002/qj.49705021110

Holcomb, S.R., Bass, A.A., Reid, C.S., Seymour, M.A., Lorenz, N.F., Gregory, B.B., Javed, S.M., Balkum, K.F., 2015. Louisiana Wildlife Action Plan.

Jensen, W.E., Finck, E.J., 2004. Edge Effects on Nesting Dickcissels (Spiza americana) in Relation to Edge Type of Remnant Tallgrass Prairie in Kansas. Am. Midl. Nat. 151, 192– 199.

Johnson, S.E., Mudrak, E.L., Beever, E.A., Sanders, S., Waller, D.M., 2008. Comparing power among three sampling methods for monitoring forest vegetation. Can. J. For. Res. 38, 143– 156.

Jonckheere, I., Fleck, S., Nackaerts, K., Muys, B., Coppin, P., Weiss, M., Baret, F., 2004. Review of methods for in situ leaf area index determination: Part I. Theories, sensors and hemispherical photography. Agric. For. Meteorol. 121, 19–35. https://doi.org/http://dx.doi.org/10.1016/j.agrformet.2003.08.027

Kaiser, S.A., Lindell, C.A., 2007. Effects of Distance to Edge and Edge Type on Nestling Growth and Nest Survival in the Wood Thrush. Condor 109, 288–303.

Kent, M., Coker, P., 1992. Vegetation description and analysis: a practical approach. Belhaven Press, London, UK.

Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World Map of the Köppen-Geiger climate classification updated. Meteorol. Zeitschrift.

Lemenih, M., Gidyelew, T., Teketay, D., 2004. Effects of canopy cover and understory environment of tree plantations on richness, density and size of colonizing woody species in southern Ethiopia. For. Ecol. Manage. 194, 1–10. https://doi.org/10.1016/j.foreco.2004.01.050

Messick, E.J., 2016. Breeding season avian community composition and prey availability in eucalyptus and slash pine plantations of southwestern Louisiana. Stephen F. Austin State University.

Peet, R.K., Allard, D.J., 1993. Longleaf pine vegetation of the southern Atlantic and eastern Gulf Coast regions: a preliminary classification, in: Proceedings of the Tall Timbers Fire Ecology Conference. pp. 45–81.

Rich, P.M., 1990. Characterizing plant canopies with hemispherical photographs. Remote Sens. Rev. 5, 13–29. https://doi.org/10.1080/02757259009532119 18

Soil Survey Staff, 2014. Keys to Soil Taxonomy, 12th ed. USDA-Natural Resources Conservation Service, Washington, DC.

USDA, 2002. Soil survey of Beauregard Parish, Louisiana.

Van der Heijde, M.G.A., Bardgett, R.D., Van Straalen, N.M., 2008. The unseen majority: soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecol. Lett. 11, 296–310.

Walter, J.-M.N., 2009. CIMES-FISHEYE © A Package of Programs for the Assessment of Canopy and Solar Radiation Regimes through Hemispherical Photographs. Manual.

Yirdaw, E., Luukkanen, O., 2004. Photosynthetically active radiation transmittance of forest plantation canopies in the Ethiopian highlands. For. Ecol. Manage. 188, 17–24. https://doi.org/https://doi.org/10.1016/j.foreco.2003.07.024

Zhao, J., Wan, S., Zhang, C., Liu, Z., Zhou, L., Fu, S., 2014. Contributions of Understory and/or Overstory Vegetations to Soil Microbial PLFA and Nematode Diversities in Eucalyptus Monocultures. PLoS One 9, 1–8.

CHAPTER 3. UNDERSTORY VEGETATION RICHNESS AND DIVERSITY OF EUCALYPTUS BENTHAMII AND PINUS ELLIOTTII PLANTATIONS IN THE MID-SOUTH U.S.

3.1. INTRODUCTION

Eucalyptus (Eucalyptus spp.) plantations are common in Asia, Africa, and South America because of their fast growth, high productivity, and good adaptability to an array of site conditions (JianPing et al., 2015), but they have not been widely planted in the United States due to their sensitivity to cold temperatures (Kellison et al., 2013; Rockwood, 2012). Eucalyptus plantations can be successfully established in the southeastern U.S. by planting cold-tolerant species, such as Camden white gum (Eucalyptus benthamii (Maiden & Cambage)), which are able to thrive in the areas classified by the U.S. Department of Agriculture (USDA) as Plant

Hardiness Zones 8b (annual minimum temperature > 9.4° C) and greater (Hinchee et al., 2011).

In the southeastern U.S., pine (Pinus spp.) plantations are currently the most important source of softwood forest products with about 16 million ha of planted forests (Wear et al.,

2015). In the Western Gulf region cold-tolerant eucalyptus species have been explored as short- rotation pulpwood feedstock for paper mills that require hardwood. Some mills in the region have recurring difficulties obtaining local raw material during wet periods, when harvesting operations are restricted to protect soil and water quality. During such periods, mills must import feedstock material at high costs (Blazier et al., 2012).

The establishment of non-native forest plantations has led to questions about potential impacts on native species diversity (Brockerhoff et al., 2009). Lower species diversity in eucalyptus plantations relative to natural forests has been observed in numerous studies

(Brockerhoff et al., 2013; Stallings, 1991), considering both woody species (Abiyu et al., 2011;

Lemenih et al., 2004; Stephens and Wagner, 2007; Tyynelä, 2001; Zhao et al., 2014) and understory species (Fork et al., 2015; Hua-Feng et al., 2011; YuanGuang et al., 2010; Zhang et

al., 2014). On the other hand, eucalyptus plantations did not alter understory species richness when eucalyptus was established on barren lands (JianPing et al., 2015) when compared with native woodland (Bone et al., 1997). Eucalyptus plantations showed higher (Lemenih et al.,

2004; Michelsen et al., 1996) or similar (Geldenhuys, 1997) understory and woody species richness than other non-native species plantations. Both forest species and herbaceous species typical of savannas were encountered in eucalyptus plantations established in the Republic of the

Congo, suggesting that eucalyptus plantations acted as a connectivity corridor between natural forest patches or a surrogate habitat for native species (Loumeto and Huttel, 1997).

A review conducted by Bremer and Farley (2010) emphasized the effect of plantations on biodiversity in relation to land-use change. Plantations of native tree species were richer in species than in plantations of exotic tree species; management and structural factors rather than native/exotic dichotomy were identified as primary reasons for the differences in species richness of the planation types. In particular, Stephens and Wagner (2007) suggested that native plantations are on the average more similar in habitat structure to natural forests than are exotic plantations and therefore support a more diverse flora and fauna.

Overstory and canopy influence the habitat for understory by altering light, temperature and humidity regimes (Parrotta, 1993; Richardson et al., 1989). Moreover, soil nutrient availability and pH are influenced by overstory and may affect the presence and growth of understory species (Légaré et al., 2001). Therefore, it is important to understand the role of overstory in determining the species assemblage in the understory. Overstory characteristics can be described by variables well-known in forestry, such as diameter at breast height (DBH), basal area, and tree height, while canopy characteristics can be described through leaf area index (LAI) and total daily radiation under the canopy (RAD) previously illustrated in Chapter 2.

In Louisiana, slash pine plantations have shown less understory species diversity than native longleaf pine savannas and woodlands as a consequence of management practices that decreased habitat heterogeneity, such as high stocking (which leads to dense shading), intensive use of herbicides, and exclusion of prescribed fire (Holcomb et al., 2015). Few studies have compared vegetation communities in short-rotation forestry plantations with those in pine plantations in the United States, with very few cases regarding eucalyptus plantations. A study was established in southwestern Louisiana within operational-scale plantations to explore differences in vegetation richness between intensively-managed Eucalyptus benthamii and slash pine (Pinus elliottii) plantations.

Understory species comprise an important component of forest ecosystems, but they are studied less frequently and intensively compared to tree species. In this study, understory species richness and diversity were compared over three years between Camden white gum plantations established in 2013, same-age slash pine plantations, and older but similar-height slash pine plantations established in 2008. Specifically, understory species communities were compared between the three plantation types through rarefaction curves and two diversity indices (species richness and Shannon index). Multivariate techniques were also used to understand the differences among the plantation types in terms of richness and diversity indices, overstory biometrics, leaf area index, total daily radiation under the canopy, and soil characteristics.

3.2. METHODS

Information about site, experimental design, vegetation surveys, LAI and total daily radiation under the canopy, overstory metrics, and soil sampling were described in Chapter 2.

3.2.1 Species richness, diversity indices, and rarefaction curves

Understory vegetation data collected during the vegetation surveys were used to compute the following community diversity indices: species richness (S) and Shannon index (H’) (Jost,

2007, 2006; Lee-Ann and Martin, 2010; Magurran, 2004). While species richness represents just the number of species without any information about abundance, Shannon index considers both abundance and evenness of species present in the community, usually ranging from 1.3 to 3.5, with higher values indicating higher diversity (Magurran, 2004; Margalef, 1972). Shannon index was calculated as

* " ! = − % &'()&' '+,

where pi is the relative proportion of species i. Understory species richness was calculated on a plot basis, adding together all the species recorded in the 10 sampling quadrats within each plot.

Species richness and Shannon index were calculated using EstimateS 9.1.

An alternative way to estimate species richness is to construct species accumulation curves, which defines the cumulative number of species discovered in a community as function of sampling effort, defined as numbers of collected individuals or the cumulative number of samples (Colwell and Coddington, 1994). Species-area curves are widely used in botanical research (Magurran, 2004), where the x-axis is the number of samples and the y-axis is the accumulated number of species (Gotelli and Colwell, 2010). Since sampling order influences the accumulation curve shape, smooth curves (rarefaction curves; Fig 3.1) can be produced through a randomization process (Colwell et al., 2004; Gotelli and Colwell, 2010, 2001).

20 Number of species

0 0 20 40 60 80 100 Samples

Fig. 3.1. An example species accumulation curve. The red lines represent the mean species richness (solid) and the 95% confidence intervals (dotted) generated through randomization, while the blue line represents an example of sampling.

Sample-based rarefaction curves were generated for plant communities in the three plantation types (Colwell et al., 2004; Gotelli and Colwell, 2001) using EstimateS 9.1 (Colwell,

2013). To generate smooth rarefaction curves, the sample order was randomized 100 times. To evaluate whether species richness was different among plantation types at the level of α = 0.05, the 95% confidence intervals were checked for overlap (Colwell et al., 2012).

3.2.2. Statistical analysis: generalized linear mixed modes, generalized linear models, correlation analysis, and multiple regression analysis

Analyses of variance (ANOVA) at α=0.05 were conducted through generalized linear mixed models using the GLIMMIX procedure in SAS 9.4 (SAS Institute Inc., 2002–03) to test for differences in species richness, Shannon index, overstory metrics, LAI, and total daily radiation under the canopy, and soil characteristics. When ANOVA revealed significant effects, means separations were performed using Tukey’s honest significant difference test.

We determined Pearson’s correlation coefficients (r) to explore possible relationships between variables related to forest structural conditions (LAI, daily total radiation under canopy, tree DBH, height, basal area), plant community metrics (species richness and Shannon index), and soil characteristics (pH, C, N, etc.).

Multiple linear regression analysis was performed at α=0.05 to characterize the influence of overstory and soil parameters on understory species richness and diversity. The stepwise method of the REG procedure of SAS 9.4 was used for this analysis, using 0.15 and 0.05 as significance levels for entry and staying in the model, respectively. Species richness and

Shannon index were dependent variables for the regression equations. Overstory metrics, LAI, total daily radiation under canopy, pH, and soil nutrient concentrations were independent

(explanatory) variables in the models. Multicollinearity of independent variables was assessed using variance inflation factor, eigensystem analysis of correlation matrix, and condition index.

3.2.3. Statistical analysis: canonical discriminant analysis

To further explore the relationships between understory, overstory, and soil variables during 2016, stepwise discriminant analysis (SDA) – STEPDISC procedure in SAS – was used to reduce the dimension of the dataset. This analysis identifies the variables that contribute most to the discriminatory power of the model, as measured by Wilk’s lambda (Huberty and Olejnik,

2006). A significance threshold of a=0.05 was used for variables to enter and be retained in the discriminant function, and the corresponding partial R2 values were listed. Partial R2 characterizes the amount of variation accounted for by that variable in discriminating between groups; the larger the partial R2 value, the greater the contribution of the respective variable to discrimination between groups. The reduced dataset was then used in a discriminant canonical analysis (DCA), conducted using the candisc package (Friendly and Fox, 2017) in R version

3.5.1 (R Core Team, 2018), to derive canonical functions representing linear combinations of the variables selected through DCA that best summarize the differences among plantations. A canonical biplot was also produced. In a canonical biplot the axes (the canonical functions) are the two dimensions that provide maximum separation among the groups. The biplot shows how each observation is represented in terms of canonical functions and how each covariate contributes to the canonical functions. In the canonical plot, rays define the covariates, and their length and direction indicate the degree of association of the corresponding covariate with the canonical functions.

3.3. RESULTS

3.3.1. Families, species, life cycle, and form of understory vegetation

The number of families and species and the life cycle and form of understory vegetation observed throughout this study are reported in Table 3.1. Asteraceae contained the highest number of species, followed by Fabaceae and Poaceae. Common species present in all plantation types were little blue stem (Schizachyrium scoparium), southern dewberry (Rubus trivialis), cat greenbrier (Smilax glauca), swamp sunflower (Helianthus angustifolius), and several species of the genera Eupatorium, Solidago, Dichanthelium, Rhynchospora, Diodia and Rhexia.

Information about families, species, incidence, life form, and duration observed is reported in

Appendix A1.

The predominant form was forb/herb for all the plantation types, while trees and vines were more common in S08 and E13 than in S13, except for 2016 where S13 showed the highest number of tree species (Table 3.1). Shrubs and graminoids were present in all the plantation types as well. Perennial species were slightly more frequent in S08 than in the other plantation types, while annual and annual/biennial species had higher proportions in S13 and E13. Biennial species were not found in S08 during the three years, and they were not found in S13 and E13 during 2015 and 2016, respectively.

- 53 56 40 74 85 66 72 75 52 Species

Annual 1.80 7.10 5.00 3.77 10.47 6.56 8.33 5.33 11.54 Biennial 0.00 0.00 0.00 1.89 0.00 1.64 1.39 1.33 0.00 Cycle Annual/biennial 1.89 0.00 7.50 5.66 2.33 6.56 5.56 5.33 9.62 Perennial 96.23 92.86 87.50 88.68 87.21 85.25 84.72 88.00 78.85

Graminoid 13.21 16.07 12.50 18.92 17.44 13.11 13.89 12.00 9.62 Forb/herb 50.94 44.64 45.00 56.76 56.98 59.02 55.56 56.00 61.54 Form Shrub 16.98 19.64 20.00 14.86 16.28 14.75 15.28 16.00 15.38 Tree 5.66 7.14 5.00 1.35 2.33 6.56 5.56 5.33 0.00 Vine 13.21 12.50 17.50 8.11 6.98 6.56 9.72 10.67 13.46

3.3.2. Rarefaction curves

Rarefaction curves were different among and within plantations types in all years of the study. During 2014, understory species richness was greater in E13 and S13 than in S08 (Fig.

3.2a). In 2015 and 2016, species richness was highest in S13, intermediate in E13, and lowest in

S08 (Fig. 3.2b and c). Within each plantation type, species richness did not change from 2014 to

2015 in any plantation type, while it declined from 2015 to 2016 in E13 and S08, but not in S13

(Fig. 3.3).

100 (a) 100 (b)

80 80

60 60

40 40

S13 (A) S13 (A) 20 E13 (A) 20 E13 (B) S08 (B) S08 (C) 0 0 0 50 100 150 200 0 50 100 150 200

100 (c)

S13 (A) E13 (B) Number of species 20 S08 (C) 0 0 50 100 150 200 Quadrats

Fig. 3.2. Species accumulation curves for slash pine plantations planted in 2008 (S08) and 2013 (S13) and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana as observed in 2014 (a), 2015 (b), and 2016 (c). Dotted lines represent 95% confidence intervals, overlapping confidence intervals denote non-significant difference, and different letters indicate significant difference (a = 0.05).

100 (a) 100 (b)

80 80

60 60

40 2014 (A) 40 2015 (A) 2014 (A) 2016 (B) 20 20 2015 (A) 2016 (B) 0 0 0 50 100 150 200 0 50 100 150 200

100 (c)

40 2014 (AB) 2015 (A) 2016 (B)

Number of species 20

0 0 50 100 150 200 Quadrats

Fig. 3.3. Species accumulation curves in 2014, 2015, and 2016 for (a) Camden white gum plantations planted in 2013 (E13), (b) slash pine plantations planted in 2008 (S08), and (c) slash pine plantations planted in 2013 (S13) plantations. Dotted lines represent 95% confidence intervals, overlapping confidence intervals denote non-significant difference, and different letters indicate significant difference (a = 0.05).

3.3.3. Species richness and diversity indices

Understory species richness and Shannon index were significantly different among the three

plantation types and among years (

Table 3.2 and Fig. 3.4). Both species richness and Shannon index were greater in S13, intermediate in E13, and lower in S08 during 2014 and 2015. In 2015, they were similar in S13 and E13, but different from S08. Within plantation, species richness and Shannon index were similar in 2014 and 2015, then declined in 2016.

25 (a) 3.0 (b) Aa Aa Aa Aa 20 Aa Ba Aa Ba Ab Ab 2.5 15 Ca Ca Ba Bb Ba Bb H’

10 Cb

Number of species 2.0 Cb S13 S13 5 E13 E13 S08 S08

0 1.5

2014 2015 2016 2014 2015 2016 Fig. 3.4. Mean and standard error of the mean for species richness (a) and Shannon index (b) in slash pine plantations planted in 2008 (S08) and 2013 (S13), and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana (n=60 plots). Different capital letters indicate significant difference (P < 0.05) between plantations within the same year, while different lowercase letters indicate significant difference between years within the same plantation type.

Table 3.2. Test of fixed effects for species richness and Shannon index (H’). Main effects consisted of plantation type (plant) and year (yr) (n=180). Richness H’ Effect NDF DDF F P-value F P-value plant 2 36 34.61 < .0001 44.22 < .0001 yr 2 36 46.15 < .0001 49.07 < .0001 plant*yr 4 36 0.56 0.6921 0.83 0.5173 NDF = numerator degrees of freedom, DDF = denominator degrees of freedom.

3.3.4. Overstory, canopy, and soil characteristics

Diameter at breast height was significantly greater in S08 than E13 plantations over the three years (Table 3.3 and Table 3.4), and it increased significantly over time in both plantation types. Tree heights for S08 plantations were greater for E13 plantations in the first and second years of the study, but it was greater in E13 than for S08 in the third year. Height increased within plantations each year. Basal area of S08 plantations was significantly greater than for E13 plantations over the three years, and it significantly increased from 2014 to 2016 for both plantation types. Leaf area index and total daily radiation under the canopy both were significantly different among plantation types (S08 > E13 > S13 for LAI and S13 > E13 > S08 for radiation) in 2016 (Fig. 3.5).

Table 3.3. Mean DBH (cm), height (m), and basal area (m2 ha-1) and standard error of the mean (SEM) for slash pine plantations planted in 2008 (S08) and 2013 (S13) and in 2013 Camden white gum plantations (E13) in southwestern Louisiana. For each variable, different capital letters within a row indicate significant difference (P < 0.05) between plantations within the same year, while different lowercase letters indicate significant difference between years within the same plantation type. S08 E13 S13 Variable Measurement Mean SEM Obs. Mean SEM Obs. Mean SEM Obs. Feb/Apr 2015 8.69 (Ab) 0.08 965 4.10 (Bc) 0.06 1013 - - - DBH Jan 2016 9.33 (Aa) 0.09 1001 5.01(Bb) 0.08 1082 1.93 0.05 844 Dec 2016 10.31 (Aa) 0.10 968 7.31(Ba) 0.11 1023 3.92 0.05 849

Feb/Apr 2015 6.14 (Ac) 0.04 965 4.68 (Bb) 0.05 1013 0.10 0.001 813 Height Jan 2016 7.26 (Ab) 0.05 1001 5.45 (Bb) 0.07 1082 2.23 0.04 844 Dec 2016 8.93 (Ba) 0.05 968 9.79 (Aa) 0.12 1023 2.88 0.03 849

Feb/Apr 2015 10.35 (Ac) 1.10 20 2.69 (Bc) 0.25 20 - - - Basal area Jan 2016 12.36 (Ab) 1.19 20 4.63 (Bb) 0.46 20 0.49 0.08 20 Dec 2016 14.53 (Aa) 1.44 20 7.80 (Aa) 0.69 20 1.80 0.22 20 No DBH and basal area data were available for S13 during Feb-Apr 2015 because no trees had reached DBH at that time. S13 data related to DBH, height, and basal area were not included in the statistical analysis because not all trees in this plantation type had reached DBH within this study.

Table 3.4. Test of fixed effects for DBH, height, and basal area. DBH Height BA Effect NDF DDF F P-value F P-value F P-value plant 1 24 110.96 <.0001 3.59 0.0704 46.12 <.0001 yr 2 24 92.69 <.0001 149.67 <.0001 78.23 <.0001 plant*yr 2 24 13.01 0.0001 221.24 <.0001 1.09 0.3511 NDF = numerator degrees of freedom, DDF = denominator degrees of freedom

There were differences among plantation types in some of the soil parameters (Table

3.5). S13 stands were less acidic than S08 and S13. The S13 stands had greater Cu concentrations than S08. Higher concentrations of S were observed in S08 stands than in S13 and

E13 stands, and both slash pine plantation types were richer in Na and Zn than E13. The E13 stands had higher P concentrations than the other plantation types.

2.5 (a) A 10 (b) A

2.0 8 -1 B B C

1.5 day 6 -2 LAI 1.0 4

C MJ m 0.5 2

0.0 0 E13 S08 S13 E13 S08 S13

Table 3.5. Soil characteristics means and standard error of the mean (SEM) in slash pine plantations planted in 2008 (S08) and 2013 (S13) and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana (n=60 plots) in 2016. Means followed by different letters within each row indicate statistically significant difference (P-value < 0.05). E13 S08 S13 Variable Mean SEM Mean SEM Mean SEM C (%) 1.077 0.062 1.137 0.077 1.211 0.116 N (%) 0.041 0.003 0.048 0.003 0.048 0.003 Ca (µg g-1) 123.146 8.886 128.242 11.785 149.669 20.001 Cu (µg g-1) 0.523(AB) 0.033 0.450(B) 0.027 0.602(A) 0.037 Mg (µg g-1) 30.011 1.773 38.879 4.826 44.975 9.034 P (µg g-1) 7.221(A) 2.096 2.434(B) 0.214 2.298(B) 0.205 K (µg g-1) 26.027 1.252 20.297 2.775 19.703 1.541 Na (µg g-1) 14.551(B) 0.640 19.563(A) 0.860 19.044(A) 1.845 S (µg g-1) 7.876(B) 0.322 13.131(A) 1.139 9.827(B) 0.798 Zn (µg g-1) 0.746(B) 0.071 1.506(A) 0.250 1.532(A) 0.179 pH 4.765(B) 0.045 4.673(B) 0.062 5.235(A) 0.050

3.3.5. Correlation and multiple regression analyses

Species richness and Shannon index were positively correlated with total daily radiation under the canopy, pH, and Cu and negatively correlated with LAI, DBH, basal area, and tree height (Fig 3.6). Overstory variables (height, DBH, basal area) were positively related with LAI and negatively correlated with pH and total daily radiation under the canopy. Among soil variables, Ca was positively correlated with C, Mg, and K and Cu was positively correlated with pH. Positive correlations were also found between N and C, K and Mg, and Na and Zn.

Stepwise regression indicated significant influences of LAI (negative) and pH (positive) for species richness and Shannon richness index (Table 3.6). Potassium was a positive predictor for species richness, while tree height was a positive predictor for Shannon index. The variance explained by the model was 51% for both species richness and Shannon index. Multicollinearity was indicated by condition index values (>30) for each model except for Simpson inverse index, while tolerance and variance inflation factors failed to detect multicollinearity for any model, because their values were greater than 0.1 and less than 9, respectively (Belsley et al., 2005).

pH Mg Height Ca N C Cu Rich H' RAD Na S Zn BA LAI DBH P K 1 Ca

N 37 0.8 C 48 80

Cu 33 15 12 0.6 Rich 18 7 8 41 H' 21 9 10 42 91 0.4 RAD 16 9 11 33 64 58 0.2 pH 29 0 11 42 58 58 62 Mg 51 9 -7 32 13 18 17 21 0 Na 26 17 -1 21 -2 -3 12 1 73 S 13 15 -4 6 -27 -41 -21 -23 1 23 -0.2 Zn 5 13 7 5 5 -11 14 -5 15 39 30

Height -16 -5 -10 -31 -42 -34 -79 -63 -24 -27 -1 -26 -0.4 BA -14 -1 -9 -31 -49 -41 -74 -48 -12 -1 11 -6 66 LAI -11 -5 -12 -32 -64 -60 -94 -61 -9 -2 31 -1 75 77 -0.6 DBH -11 5 -5 -36 -57 -52 -84 -60 -17 -11 21 -9 84 83 87 -0.8 P -7 16 9 -2 1 3 -3 -23 -16 -23 -13 -5 20 -5 5 2 K 37 -2 6 21 16 18 -12 -7 35 -3 -13 -9 23 -1 9 8 12 -1

Fig. 3.6. Pearson correlation plot for species richness, Shannon index (H’), diameter at breast height (DBH), tree height, leaf area index (LAI), total daily radiation under the canopy (RAD), and soil variables. Upper diagonal denotes correlation magnitude, with correlations in blue (positive) and red (negative) tonalities. Lower diagonal denotes correlation coefficients (%).

Table 3.6. Multiple linear regression stepwise selection procedure, ANOVA table and parameter estimates for mean species richness and Shannon index (H’) (n=60). ANOVA table Parameter estimates 2 Model R Source DF F P-value Parameter Estimate SE P-value Tol. VIF CI

Richness 0.51 Model 3 19.73 <.0001 Intercept -5.766 7.787 0.462 0 0 1

Error 56 LAI -3.084 0.766 0.000 0.625 1.601 3.992

Total 59 pH 3.779 1.470 0.013 0.626 1.596 6.026

K 0.102 0.044 0.023 0.992 1.008 51.721

H’ 0.51 Model 3 19.21 <.0001 Intercept -0.145 0.715 0.8402 0 0 1

Error 56 LAI -0.357 0.081 <.0001 0.407 2.458 3.875

Total 59 height 0.043 0.015 0.006 0.393 2.546 8.155

pH 0.481 0.133 0.0006 0.560 1.785 55.378

SE = standard error, Tol. = tolerance, VIF = variance inflation factor, CI = condition index.

3.3.6. Canonical discriminant analysis

A total of 19 variables were initially included in the DCA. The stepwise discriminant analysis (SDA) procedure retained 6 variables (Table 3.7). The discriminant analysis yielded two significant functions (Can1 and Can2), which are the linear combination of the variables previously selected in the stepwise procedure. The total canonical structure identifies the degree of correlation, positive or negative, between the continuous variables and the two discriminant functions. Almost all the variables loaded on the first canonical axis, with LAI, height, and DBH negatively loaded, while species richness and pH positively loaded. Soil Zn loaded positively on the second axis.

A canonical biplot was produced, showing the canonical scores for the groups defined by the term as points and the canonical structure coefficients as vectors from the origin (Fig. 3.7 and

Table 3.7). The DCA offered an overview of the differences in the three plantation types: Can1 clearly discriminated the plantation types in terms of overstory complexity (more complex overstory on the left side of the axis) and higher floristic richness and soil pH levels (right side of

the axis). The second canonical axis discriminated in terms of soil Zn, which increases towards the positive side (upper) of the axis. S13 plots were clearly separated from S08 and E13 in the first canonical axis, while the second axis separated E13 from the slash pine plots (no overlapping in the 95% confidence level ellipses were detected).

Table 3.7. Variables retained (P ≤ 0.05) in the SDA. Total canonical structure indicates the correlation degree between the variables and the canonical functions, while canonical standardized coefficients represent the coefficients of the variables (standardized) in the linear combination. Tot. can. structure Can. std coefficients Step Variable Partial R2 Wilk’s l P-value Can1 Can2 Can1 Can2 1 LAI 0.90 0.097 <.0001 -0.97 0.05 -0.89 -0.01 2 Height 0.60 0.038 <.0001 -0.81 -0.47 -0.13 -2.05 3 DBH 0.73 0.011 <.0001 -0.91 0.09 -0.36 1.94 4 Richness 0.33 0.007 <.0001 0.74 -0.17 0.50 -0.13 5 pH 0.15 0.006 <.0001 0.73 0.15 0.48 0.06 6 Zn 0.13 0.005 <.0001 0.09 0.44 0.22 0.36

10 5

s08 + Zn s13 pH + DBHLAI 0 Rich Can2 (24.1%) Height

e13+ 5 − 10 −

−15 −10 −5 0 5 10 15

Can1 (75.9%)

Fig. 3.7. Canonical plot for slash pine plantations planted in 2008 (S08) and 2013 (S13) and in Camden white gum plantations planted in 2013 (E13) in southwestern Louisiana. Multivariate mean for each group is denoted by a plus (+) marker. A 95% confidence level (inner) ellipse is plotted for each mean. If two groups differ significantly, the confidence ellipses do not overlap. A 95% contour (outer) is plotted for each group, indicating the region that contains approximately 95% of the observations. To facilitate the interpretation, rays are enlarged by a scaling factor of 5. Rays denote DBH, tree height, LAI, species richness, soil pH, and zinc content.

3.4. DISCUSSION

In the three years of this study, the fast growth rate of Camden white gum was able to modify habitat conditions, drifting the understory composition towards a situation similar to older slash pine, which was described by more developed overstory. These findings support the idea of negative effects of plantation age on understory species diversity as a consequence of

growth rate of the plantation species used (Cusack and Montagnini, 2004; Hartley, 2002).

Overall understory species richness quantified by the rarefaction curves declined more rapidly in

Camden white gum than in slash pine plantations, which are more common for the region, over the three years of this study. During the study, E13 understory vegetation richness went from being similar to that of S13 (2014), to intermediate between S13 and S08 (2015, 2016). The rapid changes in tree dimensions in Camden white gum likely produced a faster change in understory conditions and consequently a faster decrease in species richness. The substantial increase in Camden white gum height and corresponding decline in richness was likely promoted by the operational fertilization in November 2015, which likely led to the higher soil P concentrations of Camden white gum plantations. This management facet of the Camden white gum plantations may indicate that the understory changes observed in this study could occur more rapidly if such plantations are treated with more aggressive fertility management, as observed in Portugal where fertilization and harrowing practices decreased the understory species diversity in E. globulus plantations over 5 years (Carneiro et al., 2008). The confounding of operational fertilization with Camden white gum management in this study may also lead to different levels of contrast in understory condition development in similar studies if another fertilized plantation type were compared with fertilized eucalyptus plantations.

It is noteworthy that plot-level richness analysis did not detect the shift in species richness in the Camden white gum plantations relative to slash pine plantation types consistent with that revealed by the rarefaction approach. Inconsistencies in species richness between the rarefaction and plot-level methods were observed in 2014 and 2015, and it is illustrative of the elusive nature of measuring species richness (May, 1988) and the relative sensitivity of rarefaction for detecting differences in richness (James and Rathbun, 1981). An inherent issue with species richness determination is that more species are recorded as sampling intensity

increases (Bunge and Fitzpatrick, 1993). Sampling equal-sized areas as done with plot-level richness determination ameliorates this issue, but as species sampling progresses fewer species are discovered. In addition, the order in which sampling units are added affects this accumulation of species detection, and variation in species detection accumulation arises from sampling error and from heterogeneity of sampling units (Colwell and Coddington, 1994). Rarefaction curves account for the increasing rarity of species detection during sampling through curve shape (with asymptotes substantiating sampling adequacy for richness characterization) and the arbitrary nature of species detection via randomization of sample accumulation order (Colwell, 2009;

Gotelli and Colwell, 2001). As such, rarefaction was likely more sensitive than plot-level richness at detecting differences in richness, and both together provided a comprehensive view of the change in species richness.

Species richness and Shannon index followed a similar pattern during the years and within plantation types. Camden white gum and same-age slash pine plantations may have been more conducive to higher understory species diversity than the older slash pine plantations because they had more recently had been clearcut-harvested and treated with vegetation suppression operations (burning with single applications of herbicide for slash pine, one year of near-complete vegetation suppression with herbicides for Camden white gum). After these harvesting and vegetation control operations, colonization of the sites with an array of species occurred. The older slash pine plantations had a longer time for canopy closure to affect understory species diversity. Camden white gum interestingly had similar understory species diversity (as measured by rarefaction) as same-age slash pine at the inception of this study, which was the year immediately following the cessation of operational herbicide treatments in those plantations; the slash pine plantations were treated with herbicides only once prior planting. The banded herbicide application in E13, done to constrain vegetation suppression

costs, likely fostered colonization of species between tree rows that expanded once herbicide applications ceased.

Richness (for rarefaction and plot-level methods) and Shannon index declined in all three plantation types in 2016 relative to the prior years. Timing of spring and fall floristic surveys was similar in all years, so species detection in the surveys was likely unaffected by survey timing.

The spring portion of the 2016 assessment may have been affected by precipitation patterns.

March 2016 had ~2.5 times greater precipitation than the long-term average for the region (Fig.

2.2), which led to standing water in the flatwoods sites of this study. This standing water could have reduced vegetation germination in spring 2016. In prior years of the study, deviations from long-term precipitation patterns were not of the magnitude observed during the spring sampling period of 2016.

As of the final year of this study, the plantation types were different primarily in understory species richness, overstory parameters (LAI, height, DBH), and soil pH and Zn, as evidenced by discriminant analysis results. Multiple regression indicated that LAI had a strong negative influence on understory vegetation richness and diversity, while soil pH had a positive influence on these parameters. Together these results suggest that LAI and soil pH had the strongest influences on understory richness and diversity in this study.

Leaf area index was the dominant factor among the overstory and soil variables observed in this study influencing species richness and diversity. It had strong correlation with and was a robust predictor of species richness and diversity indices. Under-canopy radiation was also correlated with species richness and Shannon index, but it may have been eliminated in stepwise regression due to its high (94%) correlation with LAI. Moreover, both LAI and radiation values are mathematically calculated from gap fraction derived from hemispherical canopy pictures.

Leaf area index has a strong effect on understory light regimes. Light regime is a key factor in

determining the microclimatic conditions of sites and associated plant communities (Parrotta,

1995; Richardson et al., 1989). Several elements influence overstory transmittance, such as spatial arrangement of leaves, leaf size, and optical properties of leaves, which can be approximatively described by variables such as canopy closure or LAI (Barbier et al., 2008). In

2016, Camden white gum height was greater than older slash pine, but its LAI was intermediate between same-age and older slash pine. Those findings suggested a canopy architecture of

Camden white gum with less-dense foliage, which allowed more light for understory and correspondingly higher richness and diversity than older slash pine. Prior studies observed relationships between LAI and understory similar to findings of this study. Lemenih et al. (2004) identified a correlation between canopy characteristics of plantation stands and understory species richness, with species richness decreasing from open-canopy to closed-canopy plantations. A decrease in understory richness with increasing overstory LAI and basal area was observed in Eucalyptus saligna, Flindersia brayleyana, and Fraxinus uhde plantations in Hawaii

(Harrington and Ewel, 1997). In a study conducted in six plantation types (Acacia mangium,

Schima superba, Eucalyptus citriodora, Eucalyptus exserta, mixed-coniferous, and mixed native species plantations) in South China, light was the most important environmental factor influencing composition and structure of understory vegetation (Duan et al., 2010). Light was positively correlated with Shannon diversity indices (r = 0.4, p = 0.002) in the Duan et al. (2010) study; similar levels of correlation (r = 0.53) were observed between diversity indices and parameters associated with light (LAI and under-canopy radiation) in this study, further confirming the idea that light was one of the most influencing factors in determining understory species composition.

Shade tolerant and partial shade tolerant species were observed more frequently in S08 and E13 stands than in S13 stands, indicating a difference in light regimes among the plantations

associated with different vegetation. For instance, Carolina jessamine (Gelsemium sempervirens) was observed in more than 41-46.5% of the sampling quadrats in S08, 5-7% of the quadrats of

E13, and nearly absent in S13 over the years observed. Similarly, yaupon holly (Ilex vomitoria), a perennial tree shrub, was observed in 6.5-11.5% of sampling quadrats in S08, 6.5-11% in E13, and 0-1.5% of S13 quadrats. The incidence of Japanese climbing fern (Lygodium japonicum), a non-native invasive, perennial, vine-like fern, was 8-10.5% in E13, 0.5-1.5% in S08, and 0-2% in

S13, rising potential concerns on the diffusion of invasive non-native species. On the other hand, annual and perennial forbs and herbs typical of open-pine savannas and flatwoods, such as

Agalinis filicaulis, Centaurium pulchellum, Lobelia puberula, Lobelia appendiculata, Polygala cruciata, Polygala cymosa, and Polygala mariana were nearly absent in S08 and E13 but present in S13, although with a low incidence rate over the three years. Based on E. benthamii having an intermediate level of light-shade preferring species over the course of this study, understory vegetation of eucalyptus sites may become more representative of plant communities in closed- canopy forests, similar to older slash pine, where shade-tolerant woody perennial vines and shrubs are common. However, eucalyptus stands are harvested in very short rotations (3-7 years), which might prevent the reaching of fully closed-canopy conditions, with important implications on understory vegetation succession. Successional processes in understory vegetation (i.e. shifts from heliophilic species towards shade-tolerant species due to forest canopy closure over time) normally occur in plantations (although thinning affects this progression), and this process is profoundly linked to several ecosystem services and functions, such as nutrient cycling and faunal habitats (Brockerhoff et al., 2009; Carnus et al., 2006). Vegetation successional patterns under the different harvesting patterns of Camden white gum (which are clearcut on 7-10 year intervals) and slash pine (which are thinned two or three times within 30- to 35-year rotations) is an unexplored issue for future research.

Soil pH positively affected understory species richness and diversity in this study. Similar to findings of our study, Eycott et al. (2006) reported higher species richness in recently clear- felled Pinus nigra and Pinus sylvestris stands, where soil pH was higher. Likewise, Hutchinson et al. (1999) reported a positive correlation between floristic richness and soil pH in oak forests in Ohio. Their results matched the condition of younger slash pine in this study, where both richness and pH were higher compared to older slash pine and eucalyptus stands. The higher pH of the S13 plantations was likely due to prescribed burning that preceded their establishment

(Messick et al., 2018); no prescribed burning was conducted prior to Camden white gum establishment. The S08 stands also had prescribed burning prior to their establishment, but the time interval between burning and the years observed in our study was longer than that for S13 plantations. Soil pH increases with heating (Certini, 2005); with temperatures > 450-500 °C, bases are released as a consequence of complete combustion of fuel (Arocena and Opio, 2003), and base saturation is enhanced (Macadam, 1987). Ulery et al. (1993) reported three-unit pH increases in topsoil immediately after burning in Q. engelmanii, P. ponderosa, and mixed conifer stands. Soil pH is a key factor in determining nutrient availability: macronutrients (N, P, K, Ca,

S, and Mg), and various micronutrient (Cu, Fe, Mo, and B) deficiencies occur for pH values lower than 5-5.5 (Havlin et al., 2014). I hypothesize that higher pH values in same-age slash pine increased the availability of nutrients for understory, promoting higher richness and diversity.

These findings highlight another facet of conversion to eucalyptus plantations on understory species richness; exclusion of fire from management in eucalyptus plantations can contribute to lower species richness.

Soil K was a significant positive predictor in the species richness regression model, which indicates that understory vegetation may benefit of higher soil K levels. Janssens et al.

(1998) determined that the number of species in grassland increased with increasing soil K

concentration up to 150-200 µg g-1 of soil (optimal levels for grass nutrition), which was tenfold higher than the K concentrations observed in this study. Higher soil K may have helped support the herbaceous species found in greater numbers in the S13 (62 genera/species) and E13 (65 genera/species) plantations than in the S08 (40 genera/species) plantations. However, Crawley et al. (2005) reported no significant effect of K on species richness in Parkland Grass experiment, and Huston (1980) pointed out that forests with the greatest species richness were observed on sites with the lowest nutrient values, expressed as total bases (Ca, K, Mg and Na).

3.5 CONCLUSION

For the region in which this study was conducted, the conversion of slash pine sites to

Camden white gum could lead to substantial species losses in an area where part of the native vegetation has been already compromised by the land-use change from longleaf savannas and woodlands to slash pine plantations. Species richness of Camden white gum understory decreased through the three years of this study to the point of becoming intermediate to same-age and similar-height slash pine plantations. Evidence suggests that those changes were triggered by rapid increases in Camden white gum tree dimensions, which increased LAI and reduced total daily radiation under the canopy. Fertilization of the Camden white gum plantations to promote tree growth likely accentuated this trend. Species richness was lower on acidic soils, and it was higher in less acidic same-age slash pine plantations that had been recently prescribed-burned.

These results of the study may suggest that converting land use from slash pine to Camden white gum is associated with a relative decline in understory species richness, and exclusion of prescribed fire preceding this land use conversion may have accentuated this change in understory species richness.

Further research is required to understand the effects of disturbances associated with

harvesting regimes of these plantations (with Camden white gum clearcut-harvested in 7- to 10- year rotations and slash pine thinned twice then clearcut-harvested in a 30-year rotation) on understory species richness and diversity. Besides, changes in species richness and diversity can have functional consequences on ecosystem functionality, services, and processes connected to the understory, such as nutrient cycling or net primary productivity. These issues also merit further research. At present, Eucalyptus benthamii plantations are a relatively small component of the region in which this study was conducted. Expanded establishment of Eucalyptus benthamii plantations on sites formerly occupied by slash pine plantations could have implications at the landscape scale. Combining biodiversity maintenance and wood production at different spatial scales is a critical issue for forest managers (Carnus et al., 2006), and further research is needed for a better understanding of implications at larger spatial and longer temporal scales.

3.6. REFERENCES

Anderson, M.C., 1964. Studies of the Woodland Light Climate: I. The Photographic Computation of Light Conditions. J. Ecol. 52, 27–41. https://doi.org/10.2307/2257780

Beals, E.W., Cottam, G., 1960. The Forest Vegetation of the Apostle Islands, Wisconsin. Ecology 41, 743–751. https://doi.org/10.2307/1931808

Belsley, D.A., Kuh, E., Welsch, R.E., 2005. Detecting and Assessing Collinearity, in: Belsley, D.A., Kuh, E., Welsch, R.E. (Eds.), Regression Diagnostics, Wiley Series in Probability and Statistics. New York, NY, USA. https://doi.org/doi:10.1002/0471725153.ch3

Blazier, M.A., Hennessey, T.C., Deng, S., 2005. Effects of Fertilization and Vegetation Control on Microbial Biomass Carbon and Dehydrogenase Activity in a Juvenile. For. Sci. 51, 449– 459.

Bone, R., Lawrence, M., Magombo, Z., 1997. The effect of a Eucalyptus camaldulensis (Dehn) plantation on native woodland recovery on Ulumba Mountain, southern Malawi. For. Ecol. Manage. 99, 83–99. https://doi.org/10.1016/S0378-1127(97)00196-5

Brockerhoff, E.G., Jactel, H., Parrotta, J.A., Quine, C.P., Sayer, J., Hawksworth, D.L., 2009. Plantation Forests and Biodiversity: Oxymoron or Opportunity?, Topics in Biodiversity and Conservation. Springer Netherlands, Dordrecht. https://doi.org/10.1007/978-90-481-2807-5

Bunge, J., Fitzpatrick, M., 1993. Estimating the Number of Species: A Review. J. Am. Stat. Assoc. 88, 364–373. https://doi.org/10.1080/01621459.1993.10594330

Carnus, J.M., Parrotta, J., Brockerhoff, E., Arbez, M., Jactel, H., Kremer, A., Lamb, D., O’Hara, K., Walters, B., 2006. Planted forests and biodiversity. J. For. 104, 65–77.

Colwell, R.K., 2013. EstimateS: Statistical estimation of species richness and shared species from samples. Version 9 and earlier. User’s Guide and application. Publ. http//purl.oclc.org/estimates.

Colwell, R.K., 2009. Biodiversity: concepts, patterns, and measurement. Princet. Guid. to Ecol. Princet. Univ. Press. Princet. 257–263.

Colwell, R.K., Chao, A., Gotelli, N.J., Lin, S.-Y., Mao, C.X., Chazdon, R.L., Longino, J.T., 2012. Models and estimators linking individual-based and sample-based rarefaction, extrapolation and comparison of assemblages. J. Plant Ecol. 5, 3–21. https://doi.org/10.1093/jpe/rtr044

Colwell, R.K., Coddington, J.A., 1994. Estimating terrestrial biodiversity through extrapolation. Philos. Trans. R. Soc. London B Biol. Sci. 345, 101–118.

Colwell, R.K., Mao, C.X., Chang, J., 2004. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology 85, 2717–2727.

Crawley, M.J., Johnston, A.E., Silvertown, J., Dodd, M., de Mazancourt, C., Heard, M.S., Henman, D.F., Edwards, G.R., 2005. Determinants of Species Richness in the Park Grass Experiment. Am. Nat. 165, 179–192. https://doi.org/10.1086/427270

Curtis, J.T., 1959. The vegetation of Wisconsin. University of Wisconsin. University of Wisconsin Press, Madison, WI.

Friendly, M., Fox, J., 2017. Candisc: Visualizing Generalized Canonical Discriminant and Canonical Correlation Analysis.

Geldenhuys, C.J., 1997. Native forest regeneration in pine and eucalypt plantations in Northern Province, South Africa. For. Ecol. Manage. 99, 101–115. https://doi.org/http://dx.doi.org/10.1016/S0378-1127(97)00197-7

Gotelli, N.J., Colwell, R.K., 2010. Estimating species richness, Biological diversity: frontiers in measurement and assessment. A. E. Magurran and B. J. McGill, eds. Oxford University Press, Oxford, UK.

Gotelli, N.J., Colwell, R.K., 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecol. Lett. 4, 379–391.

Harrington, R.A., Ewel, J.J., 1997. Invasibility of tree plantations by native and non-indigenous plant species in Hawaii. For. Ecol. Manage. 99, 153–162. 47

https://doi.org/https://doi.org/10.1016/S0378-1127(97)00201-6

Hill, R., 1924. A lens for whole sky photographs. Q. J. R. Meteorol. Soc. 50, 227–235. https://doi.org/10.1002/qj.49705021110

Hua-Feng, W., Lencinas, M. V, Friedman, C.R., Xiao-Ke, W., Jiang-Xiao, Q., 2011. Understory plant diversity assessment of Eucalyptus plantations over three vegetation types in Yunnan, China. New For. 42, 101–116. https://doi.org/10.1007/s11056-010-9240-x

Huberty, C.J., Olejnik, S., 2006. Applied MANOVA and discriminant analysis, Second. ed. John Wiley & Sons, New York, NY, USA.

Huston, M., 1980. Soil Nutrients and Tree Species Richness in Costa Rican Forests. J. Biogeogr. 7, 147–157. https://doi.org/10.2307/2844707

James, F.C., Rathbun, S., 1981. Rarefaction, Relative Abundance, and Diversity of Avian Communities. Auk 98, 785–800.

Janssens, F., Peeters, A., Tallowin, J.R.B., Bakker, J.P., Bekker, R.M., Fillat, F., Oomes, M.J.M., 1998. Relationship between soil chemical factors and grassland diversity. Plant Soil 202, 69–78. https://doi.org/10.1023/A:1004389614865

Jensen, W.E., Finck, E.J., 2004. Edge Effects on Nesting Dickcissels (Spiza americana) in Relation to Edge Type of Remnant Tallgrass Prairie in Kansas. Am. Midl. Nat. 151, 192– 199.

JianPing, W., HouBao, F., WenFei, L., GuoMin, H., JianFu, T., RuiJin, Z., Jing, H., ZhanFeng, L., 2015. Should exotic Eucalyptus be planted in subtropical China: insights from understory plant diversity in two contrasting Eucalyptus chronosequences. Environ. Manage. 56, 1244–1251.

Johnson, S.E., Mudrak, E.L., Beever, E.A., Sanders, S., Waller, D.M., 2008. Comparing power among three sampling methods for monitoring forest vegetation. Can. J. For. Res. 38, 143– 156.

Jost, L., 2007. Partitioning diversity into independent alpha and beta components. Ecology 88, 2427–2439. https://doi.org/10.1890/06-1736.1

Jost, L., 2006. Entropy and diversity. Oikos 113, 363–375. https://doi.org/10.1111/j.2006.0030- 1299.14714.x 48

Kaiser, S.A., Lindell, C.A., 2007. Effects of Distance to Edge and Edge Type on Nestling Growth and Nest Survival in the Wood Thrush. Condor 109, 288–303.

Kellison, R.C., Lea, R., Marsh, P., 2013. Introduction of Eucalyptus spp. into the United States with special emphasis on the southern United States. Int. J. For. Res. 2013, Article ID 189393.

Kent, M., Coker, P., 1992. Vegetation description and analysis: a practical approach. Belhaven Press, London, UK.

Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World Map of the Köppen-Geiger climate classification updated. Meteorol. Zeitschrift.

Lee-Ann, H., Martin, B., 2010. Surveying Natural Populations, Quantitative Tools for Assessing Biodiversity. Columbia University Press, Berlin, Boston. https://doi.org/10.7312/haye14620

Loumeto, J.J., Huttel, C., 1997. Understory vegetation in fast-growing tree plantations on savanna soils in Congo. For. Ecol. Manage. 99, 65–81. https://doi.org/http://dx.doi.org/10.1016/S0378-1127(97)00195-3

Magurran, A.E., 2004. Measuring biological diversity. Blackwell Pub.

Margalef, R., 1972. Homage to Evelyn Hutchinson, or why there is an upper limit to diversity. Connecticut Academy of Arts and Sciences.

May, R.M., 1988. How Many Species Are There on Earth? Science (80-. ). 241, 1441 LP-1449.

Messick, E.J., 2016. Breeding season avian community composition and prey availability in eucalyptus and slash pine plantations of southwestern Louisiana. Stephen F. Austin State University.

Michelsen, A., Lisanework, N., Friis, I., Holst, N., 1996. Comparisons of understorey vegetation and soil fertility in plantations and adjacent natural forests in the Ethiopian highlands. J. Appl. Ecol. 33, 627–642. https://doi.org/10.2307/2404991

Nagendra, H., 2002. Opposite trends in response for the Shannon and Simpson indices of landscape diversity. Appl. Geogr. 22, 175–186. https://doi.org/https://doi.org/10.1016/S0143-6228(02)00002-4

Peet, R.K., 1974. The measurement of species diversity. Annu. Rev. Ecol. Syst. 5, 285–307.

R Core Team, 2018. R: A language and environment for statistical computing.

Rich, P.M., 1990. Characterizing plant canopies with hemispherical photographs. Remote Sens. Rev. 5, 13–29. https://doi.org/10.1080/02757259009532119

Richardson, D.M., Macdonald, I.A.W., Forsyth, G.G., 1989. Reductions in Plant Species Richness under Stands of Alien Trees and Shrubs in the Fynbos Biome. South African For. J. 149, 1–8. https://doi.org/10.1080/00382167.1989.9628986

Rockwood, D.L., 2012. History and status of Eucalyptus improvement in Florida. Int. J. For. Res. 2012, Article ID 607879.

Simpson, E.H., 1949. Measurement of diversity. Nature.

Soil Survey Staff, 2014. Keys to Soil Taxonomy, 12th ed. USDA-Natural Resources Conservation Service, Washington, DC.

Stallings, J.R., 1991. The importance of understorey on wildlife in a Brazilian eucalypt plantation. Rev. Bras. Zool. 7, 267–276. https://doi.org/10.1590/S0101- 81751990000300008

Stephens, S.S., Wagner, M.R., 2007. Forest plantations and biodiversity: a fresh perspective. J. For. 105, 307–313.

Tyynelä, T.M., 2001. Species diversity in Eucalyptus camaldulensis woodlots and miombo woodland in Northeastern Zimbabwe. New For. 22, 239–257. https://doi.org/10.1023/A:1015616010976

USDA, 2002. Soil survey of Beauregard Parish, Louisiana.

Walter, J.-M.N., 2009. CIMES-FISHEYE © A Package of Programs for the Assessment of Canopy and Solar Radiation Regimes through Hemispherical Photographs. Manual.

Whittaker, R.H., 1972. Evolution and Measurement of Species Diversity. Taxon 21, 213–251. https://doi.org/10.2307/1218190

YuanGuang, W., Duo, Y., Fang, C., ShiRong, L., HongWen, L., 2010. The changes of understory plant diversity in continuous cropping system of Eucalyptus plantations, South China. J. For. Res. 15, 252–258. https://doi.org/10.1007/s10310-010-0179-8

Zhang, D., Zhang, J., Yang, W., Wu, F., Huang, Y., 2014. Plant and soil seed bank diversity across a range of ages of Eucalyptus grandis plantations afforested on arable lands. Plant

Soil 376, 307–325. https://doi.org/10.1007/s11104-013-1954-z

CHAPTER 4. UNDERSTORY VEGETATION FUNCTIONAL TRAITS AND DIVERSITYOF EUCALYPTUS BENTHAMII AND PINUS ELLIOTTII PLANTATIONS IN THE MID-SOUTH U.S.

4.1. INTRODUCTION

In the southeastern U.S. cold-tolerant eucalyptus species have been explored as short- rotation pulpwood feedstock for paper mills that require hardwood fiber during wet periods, when harvesting operations are restricted to protect soil and water quality (Blazier et al., 2012).

In the Western Gulf region, Camden white gum (Eucalyptus benthamii (Maiden & Cambage)) plantations have been established as a possible alternative source of hardwood fiber. These plantations have been established on sites where slash pine (Pinus elliottii) has been historically managed.

The establishment of non-native forest plantations such as eucalyptus has prompted studies on the potential impacts of this land use conversion on native species richness and diversity (Brockerhoff et al., 2009), with contrasting results. Eucalyptus plantations showed lower species diversity than natural forests (Brockerhoff et al., 2013; Stallings, 1991), considering both woody species (Abiyu et al., 2011; Lemenih et al., 2004; Stephens and Wagner,

2007; Tyynelä, 2001; Zhao et al., 2014) and understory species (Fork et al., 2015; Hua-Feng et al., 2011; YuanGuang et al., 2010; Zhang et al., 2014). On the other hand, eucalyptus plantations did not alter the understory species composition when established on barren lands (JianPing et al., 2015) and in comparison with native woodland (Bone et al., 1997), even showing higher

(Lemenih et al., 2004; Michelsen et al., 1996) or similar (Geldenhuys, 1997) understory and woody species richness than other non-native species plantations. Understory vegetation accounts for about 4.7% of total aboveground biomass in the United States (Smith et al., 2013), influencing significant ecosystem properties – defined as an attribute that characterizes the ecosystem in terms of size, biodiversity, stability, organization, functions, and processes (Bastian

et al., 2012) such as net primary productivity and nutrient cycling (Gilliam, 2007; Mallik, 2003;

Moore et al., 2007; Yarie, 1980) as well as wildlife diversity and abundance (Pardini et al., 2005;

Russell et al., 2017). Ecosystems services (benefits that humans obtain from ecosystems to support their survival and quality of life) are dependent on ecosystem properties. Functional properties of an ecosystem can be estimated using functional traits, which are defined by Crisp and Cook (2012) as “morphological, physiological, or phenological heritable features measurable at individual level, from the cell to the whole organism, without references to the environment or any other level of organization”.

The linkage between plant species diversity and ecosystem functionality is a controversial topic (Diaz and Cabido, 2001). Although there is consensus that changes in species richness and abundance have direct impacts on ecosystem structure in terms of community dynamics, causing a change in normal ecosystem functioning (Balvanera et al., 2006; Cardinale et al., 2011, 2006; Hooper et al., 2005; Isbell et al., 2011; Loreau, 2000), higher levels of biodiversity might not always correspond to superior ecosystem in terms of functions and services provided (Givnish, 1994; Grime, 1997).

Ecosystem properties and services can be better understood via the plant functional traits that influence them (Díaz et al., 2007; Díaz et al., 2006; Hooper et al., 2005; Pla et al., 2011).

According to the mass-ratio hypothesis, the most abundant species in systems have the strongest effects on ecosystem properties (Grime, 1998). The mass-ratio hypothesis states that dominant and some subordinate species are the determinants of ecosystem functions and properties because of their greater participation in processes such as photosynthesis, nutrient cycling, and hydrologic cycles. Moreover, a functional diversity approach can also be used as an indirect measure of resilience to changes or disturbance factors and for assessing levels redundancy in terms of ecosystem properties, functionality, and services.

One of the ecosystem properties of understory involved in numerous ecosystem services

(carbon sequestration, wildlife habitats, nutrient cycling) is total aboveground biomass (AGB), comprised of live understory shrubs, forbs, graminoids, and seedlings, excluding dead components of the understory such as woody debris and litter (Hudak et al., 2012).

The relationship between functional traits and ecosystem properties has been studied for temperate grasslands in relation to productivity (Schwartz et al., 2000; Spehn et al., 2005; Tilman et al., 1996), and for forests in relation to carbon sequestration (Ruiz‐Jaen and Potvin, 2011), different management practices (Maeshiro et al., 2013), and land-use change to plantations

(Bergès et al., 2017).

There is a general lack of knowledge about the effects of land-use change to plantations on ecosystem properties and services in southern United States, and there is a specific need to address the effects of the transition from slash pine to E. benthamii on understory vegetation communities and ecosystem properties and services in southwestern Louisiana. In fact, only a few studies have been conducted in this area (see Chapter 3), which compared species richness and diversity of understory vegetation communities in eucalyptus plantations with those in native forest ecosystems in the United States, without addressing changes in functional traits nor in functional diversity. The main objective of this study was to explore understory vegetation functional diversity and functional traits in intensively-managed E. benthamii and slash pine plantations in southwestern Louisiana. Specifically, we wanted to address the following research questions: (1) How do understory functional traits and diversity vary among plantation types? (2)

Are understory functional diversity indices and functional traits good predictors for understory

AGB and understory vertical structure?

4.2. METHODS

For details about site descriptions, experimental design, vegetation surveys, and overstory metrics refer to Chapter 2.

4.2.1. Functional diversity indices and traits

To calculate functional diversity indices we used both multi- and single-trait approaches, as described by Pla et al. (2011). The single-trait approach concerns the quantification of selected traits weighted to the relative species abundance to evaluate the contribution of each species on each trait. We used community weighted mean to represent the expected functional value of a single trait (Garnier et al., 2004). For the multi-trait approach we selected the framework proposed by Villéger et al. (2008), which included three functional diversity components: (1) functional richness (FRic), (2) functional evenness (FEve), and (3) functional divergence (FDiv).

Functional richness indicates the trait functional space filled by a species assemblage, functional evenness represents the spacing regularity between species and trait functional space, while functional divergence quantifies the spread of trait values along the range of trait functional space (Mason et al., 2005; Pla et al., 2011). Functional evenness ranges between 0 and 1, approaching zero when functional distances among species are less-even or when abundances are less-evenly distributed among species. Functional divergence ranges between 0 and 1, approaching zero when highly-abundant species are distributed close to the center of gravity of the volume occupied and approaching one when highly-abundant species are distributed far away from the center of gravity.

Understory functional traits were measured for the dominant species, defined as the species that collectively make up about 80% of cumulative relative site-abundance (Garnier et al., 2004; Pakeman, 2011; Pérez-Harguindeguy et al., 2013). In this study dominant species were

determined from the vegetation survey conducted in 2016, and functional traits were measured on reproductively mature, healthy-looking individuals of dominant understory species from each of the 15 sites (5 per plantation type) during July-August 2016, following the sampling guidelines described by Pérez-Harguindeguy et al. (2013). Specifically, we measured plant height (PLH, cm), leaf area (LA, cm2), specific leaf area (SLA, mm2 m g−1), and leaf nitrogen content (LNC, %). These traits were selected according to their relationship with aboveground biomass and vertical structure. These traits were found to be correlated with primary productivity through growth rate (plant height, leaf dry weight, leaf N concentration), C fixation (leaf area,

SLA, leaf N concentration), and instantaneous photosynthetic rate (leaf area, SLA, leaf N concentration) (Da S. Pontes et al., 2007; Garnier et al., 2004; Lavorel and Garnier, 2002).

We measured plant height on 10 individuals per site for each dominant species, while leaf traits were measured on 5 replicates (leaves) from 5 individuals per site for each dominant species. LA was measured by scanning the leaves and analyzing the images with Image J

(Abràmoff et al., 2004; Rasband, 2005), SLA was calculated using leaf area divided by leaf dry mass (air-dried at 20°C for 72 hours). The air-dried leaves were then ground and sent to the LSU

AG Center Soil Testing & Plant Analysis Lab

(http://www.lsuagcenter.com/portals/our_offices/departments/spess/servicelabs/soil_testing_lab) for measuring the LNC using a LECO CN dry combustion analyzer (Dumas method).

4.2.2. Understory vertical structure and aboveground biomass

Understory herbaceous vegetation vertical structure was measured using a profile board similar to that described by Nudds (1977). The board was about 2.00 m high, composed of two separable 1.00 m sections that were about 30.53 cm wide, and it was made of 1.00 cm-thick plywood. It was marked in alternate colors, blue and red, at 50 cm intervals. Metal spikes

attached to the lower end of the board were placed to hold the board upright in the ground. In

September 2016 the board was set in the ground and read at a distance of 15 m in a randomly- chosen direction from the board. The proportion of each 50-cm interval covered by vegetation

(COV1 corresponded to 0-50 cm, COV2 to 51-100 cm, COV3 to 101-150 cm, and COV4 to 151-

200 cm) was noted as a single digit "density score", which corresponded to the mean value of a range of quintiles (i.e., 1 corresponded to 0-20%, 2 to the 21-40%, and so on). The board was placed in the mid-section of each plot (one reading per plot); the vertical structure sampling point corresponded to the central vegetation survey sampling point for each plot.

In tandem with the September 2016 vertical structure sampling, understory biomass was collected. Understory herbaceous vegetation biomass was clipped within a 1-m2 quadrat and collected in five quadrats per plot; these biomass sampling points were the odd-numbered sampling points of the vegetation survey. Biomass clippings were oven-dried (104° C) until constant weight, then the weight was recorded.

4.2.3. Data analysis

Functional diversity indices calculations were described extensively by Pla et al. (2011) and Mason et al. (2005). For functional indices calculation, traits values were standardized to have zero mean and unit variance to avoid scale effect assigning more importance to traits with higher variance values (Pla et al., 2011). Community weighted mean (CWM) was calculated for each functional trait through the following equation:

-./ = % 1'2' '+,

th where S is the total number of species, wi is the relative abundance of the i species, and xi is the trait value of the ith species in the assemblage (Díaz et al., 2007; Garnier et al., 2004; Pla et al.,

2011). Relative abundances of the species were listed in Appendix B. FDiversity software (Di

Rienzo et al., 2008) was used to calculate functional diversity indices and CWM values of each trait.

Understory vegetation data collected during the vegetation surveys were used to compute species richness and Shannon index (H’) (Jost, 2007, 2006; Lee-Ann and Martin, 2010;

Magurran, 2004). The analyses of these community diversity indices were reported in Chapter 3; their values (averaged by site to be on equivalent basis of the functional diversity sampling) are included in this component of the study to explore relationships between them and functional diversity parameters. To test for differences in CWMs of functional traits, functional diversity indices, understory AGB and vertical structure, overstory metrics, LAI, and total daily radiation under the canopy, analyses of variance (ANOVA) at α=0.05 were conducted using the GLM procedure in SAS 9.4 (SAS Institute Inc., 2002–03); means separations were performed using

Tukey’s honest significant difference test when ANOVA revealed significant effects. To explore possible relationships between understory community functional traits, species composition

(species richness, Shannon richness index), and variables related to forest structural conditions

(LAI, daily total radiation under canopy, tree DBH, height, basal area, and stem density), we determined Pearson’s correlation coefficients (r) (CORR procedure SAS 9.4).

Multiple linear regression analysis (REG procedure of SAS 9.4) was performed at α=0.05 to explore the influence of understory functional traits, functional and diversity indices, and the other variables (overstory metrics, LAI, and daily radiation under the canopy) on AGB and

Nudd’s board coverage scores. Multicollinearity of independent variables was assessed using variance inflation factor, eigensystem analysis of correlation matrix, and condition index.

4.3. RESULTS

4.3.1. Understory functional diversity indices, traits, vertical structure and aboveground biomass

No differences were detected in FRic, FEve, and FDiv, while CWMs of plant height, leaf area, and leaf nitrogen concentration differed among plantation types (Table 4.1). CWM of plant height was greater in E13 than in S13, but similar to that of S08 plantations. CMW of leaf area was greater in E13 and S08 rather than in S13, while CWM of leaf N concentration was greatest in E13, intermediate in S08, and smallest in S13.

Differences among plantation types were found in understory aboveground biomass and vertical structure (Table 4.1). Higher levels of aboveground biomass were measured in S13 and

E13 plantation types than in S08 plantations. S13 and E13 plantations had higher vegetation density scores than S08 in the third interval (COV3, 101-150 cm), while no differences were detected for the other layers. Differences in species richness, diversity indices, overstory characteristics, LAI, and total daily radiation under the canopy were previously reported in

Chapter 3.

Table 4.1. Site-based means and standard errors of them mean (SEM) for functional diversity indices (FRich, FEve, FDiv), CWMs of functional traits, understory aboveground biomass (AGB), and Nudd’s board coverage intervals (COV1-COV4) in 2008 (S08) and 2013 (S13) slash pine plantations, and in Camden white gum plantations planted in 2013 (E13) observed in southwestern Louisiana in 2016 (n=15). E13 S08 S13 Variable F P-value Mean SEM Mean SEM Mean SEM FRich 0.07 0.05 0.06 0.03 0.10 0.08 0.15 0.8621 FEve 0.77 0.03 0.65 0.08 0.73 0.04 1.21 0.3329 FDiv 0.78 0.02 0.68 0.05 0.78 0.03 2.73 0.1052 CWM PLH (cm) 81.63(A) 8.52 66.02(AB) 3.56 54.54(B) 3.51 5.69 0.0183 CWM LA (cm2) 13.56(A) 1.44 9.98(A) 1.06 4.68(B) 0.52 17.17 0.0003 CWM SLA (mm2 mg-1) 22.41 1.52 27.28 10.04 13.95 0.67 1.32 0.3035 CWM LNC (%), 2.20(A) 0.13 1.85(B) 0.06 1.37(C) 0.05 22.85 <.0001 AGB (Mg ha-1) 0.53(A) 0.06 0.21(B) 0.03 0.68(A) 0.10 11.71 0.0015 COV1 4 0.01 3 0.24 4 0.20 2.80 0.1005 COV2 3 0.37 2 0.32 3 0.45 2.82 0.0992 COV3 3(A) 0.20 1(B) 0.24 2(B) 0.40 6.62 0.0116 COV4 2 0.32 1 0.20 1 0.20 3.56 0.0613 *Model DF = 2, Error DF = 12, and Total DF = 14.

4.3.2. Correlation analysis and multiple linear regression

Correlations were observed among several of the parameters (Fig. 4.1). Functional evenness was negatively correlated with CWM of specific leaf area, while no correlations were detected for functional richness and functional divergence with any of the other variables of the dataset. CMW of specific leaf area was negatively correlated with Shannon index, CMW of leaf area was positively correlated with CMW of leaf N concentration and tree height and negatively correlated with total daily radiation under the canopy. CWM of leaf N concentration was negatively correlated with total daily radiation under the canopy and positively correlated with

DBH and tree height. Aboveground understory biomass was positively correlated with total daily radiation under the canopy, species richness, and Shannon index and negatively correlated with basal area and LAI. Nudd’s board coverage scores were positively correlated between each

other, and the third and the fourth layers were positively correlated with CMW of plant height and leaf area.

Height CWM CWMSLA CWMLA LNC BA LAI DBH FRic CWM COV3PLH COV4FEve FDiv AGB RAD Rich H' COV1COV2 1 CWM SLA CWM LA 50 0.8 CWM LNC 37 91 Height 34 87 92 0.6 BA 22 40 46 69

LAI 46 63 57 79 84 0.4 DBH 42 62 67 83 91 91

FRic -5 5 -2 -5 0 -14 -5 0.2 CWM PLH 7 46 47 50 20 32 26 18 COV3 20 57 54 34 -19 6 -2 41 53 0 COV4 -6 56 50 38 -7 13 8 48 60 82 FEve -78 -15 -4 -1 -2 -24 -22 -17 7 -10 5 -0.2 FDiv -48 -5 4 -10 -46 -38 -40 -32 -13 6 19 53 AGB -22 -36 -27 -50 -76 -75 -74 -10 -1 27 0 21 44 -0.4 RAD -43 -67 -66 -85 -84 -98 -91 14 -41 -11 -16 19 34 74 -0.6 Rich -50 -44 -40 -57 -72 -84 -78 27 -33 19 5 41 49 72 82 H' -70 -40 -33 -47 -60 -77 -71 23 -23 14 11 64 60 64 75 95 -0.8 COV1 24 33 21 -3 -51 -22 -35 14 0 66 46 -28 29 37 20 30 19 COV2 -16 6 0 -21 -56 -31 -47 22 27 73 57 4 27 57 30 45 41 64 -1 Fig. 4.1. Pearson correlation plot for functional richness (FRic), evenness (FEve), divergence (FDiv), CWMs of understory plant height (PLH), leaf area (LA), specific leaf area (SLA), leaf N concentration (LNC), understory aboveground biomass (AGB), Nudd’s board coverage scores (COV1-4), species richness (Rich), Shannon index (H’), leaf area index (LAI), total daily radiation under the canopy (RAD), diameter at breast high (DBH), and tree height (height). Upper diagonal denotes correlation magnitude, with correlations in blue (positive) and red (negative) tonalities. Lower diagonal denotes correlation coefficients (%).

Stepwise multiple regression model details are reported in Table 4.2. Basal area was the only significant predictor for the aboveground biomass model, having a negative effect; the model explained 58% of the variance in the dataset. The density of the first Nudd’s board coverage layer did not contain any predictor (intercept only model). The model for the density of the second Nudd’s board coverage layer specified basal area as a predictor, with a negative effect; the model explained 31% of the variance in the dataset. In the model for the density of the third Nudd’s board coverage layer, CWMs of plant height and leaf area (both positive effect), functional evenness (negative effect), and species richness (positive effect) were identified as predictors; the model explained 82% of the variance in the dataset. Lastly, CMW of plant height was selected as independent variable in the model predicting the density of the fourth Nudd’s board coverage layer, with a positive effect; the proportion of the variance explained by the model was 35%. Condition index, tolerance, and variance inflation factors failed to detect multicollinearity for any model, because their values were less than 30, greater than 0.1, and less than 9, respectively (Belsley et al., 2005).

Table 4.2. Multiple linear regression stepwise selection procedure, ANOVA table and parameter estimates for AGB and COV1-4 including all variables as explanatory factors (full models) (n=15). ANOVA table Parameter estimates Model R2 Step Source DF F P-value Parameter Estimate SE P-value Tol. VIF CI AGB 0.58 1 Model 1 17.96 0.0010 Intercept 0.722 0.073 <.0001 . 0 1 Error 13 BA -0.031 0.737 0.001 1 1 3.061 Total 14

COV1* ......

COV2 0.31 1 Model 2 6.30 0.0261 Intercept 4.440 0.363 <.0001 . 0 1 2 Error 13 BA -0.088 0.036 0.0311 1 1 3.061 Total 14

COV3 0.82 1 Model 4 11.66 0.0009 Intercept -1.546 0.903 0.1180 . 0 1 2 Error 10 CWM PLH 0.029 0.008 0.0064 0.726 1.378 4.897 3 Total 14 CWM LA 0.124 0.032 0.0031 0.695 1.439 10.089 FEve -2.600 1.505 0.0334 0.784 1.276 14.488 Rich 0.191 0.040 0.0008 0.648 1.543 20.010

COV4 0.35 1 Model 1 7.16 0.0191 Intercept -0.093 0.599 0.8785 . 0 1 Error 13 CWM PLH 0.023 0.009 0.0191 1 1 8.586 Total 14 R2 = model R2, Cum. R2 = cumulative R2, SE = standard error, Tol. = tolerance, VIF = variance inflation factor, CI = condition index. The asterisk denotes an intercept only model.

4.4. DISCUSSION

Functional diversity results revealed similarities in (1) the amount (functional richness),

(2) regularities of spacing and evenness of species distribution between species in the traits space

(functional evenness), and (3) the dispersion of trait values along the range of trait space

(functional divergence). Together these results suggest that understory communities were relatively resilient in the trait space (a synthesis of understory plant height, leaf area, specific leaf area, and leaf N) in response to the land-use change of slash pine plantations to Camden white gum within the timeframe observed in this study. Moreover, the decline in taxonomic species richness and evenness observed in Chapter 3 was not coupled with a decrease in functional

richness. A positive relationship between taxonomic and functional richness was shown by

Villéger et al. (2008), using data generated by artificial communities. Further evidence of the direct relationships (both linear and non-linear) between taxonomic and functional richness were found in riparian and upland habitats in Ontario (Biswas and Mallik, 2011, 2010), in Mongolian rangelands, (Sasaki et al., 2009), and in agricultural lands of United States and Costa Rica (Flynn et al., 2009).

The strength of the association between taxonomic and functional richness is expected to decay with the number of species, becoming asymptotical due to functional redundancy (Petchey and Gaston, 2002). The lack of linear relationship between functional richness and species richness in this study may suggest a possible case of functional redundancy, indicating different species occupying similar functional niches for the trait space defined by functional richness.

Therefore, species with similar traits likely supported similar functions in the different plantation types of this study.

Functional redundancy promotes a degree of resilience, defined as the ability of a community to return to a previous state of stability in terms of ecosystem properties after changing due to a disturbance, which is different from resistance, or the ability of a community to avoid the change (Carpenter et al., 2001; Harrison, 1979). Similar functional richness values in this study indicated a resilience to the land-use change from slash pine to eucalyptus, as can be deduced from the sharing of particular species across the plantation types, among the dominant species. In particular, the three plantations shared little blue stem (S. scoparium), woolly rosette grass (D. scabrisculum), Virginia buttonweed (D. virginiana), southern dewberry (R. trivialis), roundleaf thoroughwort (E. rotundifolium), swamp sunflower (H. angustifolius), and cat greenbrier (S. glauca) (see Appendix B). These species seemed to adopt a generalist strategy, which led to an overlap in the functional niches and saturation of the functional trait space

(Rosenfeld, 2002), as highlighted by the lack of differences in functional diversity indices among the plantations.

Differences in understory plant height, leaf area, and leaf nitrogen content suggested different adaptation strategies in response to changes in the environments of the plantations.

Taller understory with larger leaf area was observed in Camden white gum and older slash pine, where light conditions were lower than in same-age slash pine. These habitat conditions may have led to taller understory in response to competition for radiation with overstory and larger leaves to facilitate the radial harvesting of radiation. Leaf size can range from values less than 1 mm2 to more than 2 m2 (data retrieved from TRY traits database, accessed on 21 October 2018: http://www.try-db.org), and various explanations have been provided to explain this conspicuous variation. Leaf size can be the product of adaptation to environmental conditions to maintain optimal levels of leaf temperature and photosynthetic and water-use efficiencies (Givnish and

Vermeij, 1976). Therefore, a small leaf area denotes an adaptation to higher temperatures and higher levels of solar radiation, or nutrient deficiency (Niinemets et al., 2007; Westoby, 1998); same-age slash pine, which was characterized by higher total daily radiation than the other plantation types, also had lowest leaf area.

Taller understory, larger leaves, higher concentration of foliar N, and greater understory aboveground biomass in Camden white gum suggested a fast-growing strategy (Chai et al., 2015;

Ivanova et al., 2018; Spasojevic and Suding, 2012). However, operational fertilization of the eucalyptus sites also likely increased these traits. Generally fertilization has a positive effect on understory vegetation, increasing its biomass except when the overstory is too dense to allow a response (Binkley and Fisher, 2013). Increases of leaf N content in the understory as a consequence of fertilization were similarly reported in red pine (Rainey et al., 1999), Eucalyptus saligna (Frew et al., 2013) plantations, and Douglas-fir stands (Footen et al., 2009).

Despite similar values of vegetation densities observed in most of the height increments,

Camden white gum understory was found denser in the 101-150 cm height layer. This trend may be due to the combined effects of competition with trees for light, taller understory with larger leaves, intermediate light conditions, and higher levels of aboveground biomass of the Camden white gum plantations. In addition, fertilization likely affected understory vertical density; several authors reported increased understory coverage subsequent to fertilizer applications

(Haywood, 2012; VanderSchaaf, 2008). Nitrogen addition, in concurrence with light availability, increased the coverage and the dominance of Rubus spp. in the herbaceous layer in a 40 year-old stand in Central Appalachia, USA (Walter et al., 2016). Similar conditions were observed in the eucalyptus sites in this study, where vegetation had higher leaf N and southern dewberry (Rubus trivialis) was one of the dominant species.

Functional traits and diversity indices failed to explain understory aboveground biomass, which was predicted by overstory basal area, with a negative relationship. In contrast, several studies have found plant functional traits directly related with aboveground biomass and aboveground net primary productivity. In grasslands, a positive relationship was found between aboveground biomass and specific leaf area, leaf dry matter content, plant height, and specific root area (Chanteloup and Bonis, 2013; Roscher et al., 2013; Zhang et al., 2017). In pastures,

Mason et al. (2016) found ryegrass pastures productivity to be positively predicted by specific leaf area and leaf thickness and negatively by leaf dry matter content; tall fescue productivity was positively predicted by leaf dry matter content, and negatively by specific leaf area and leaf thickness. In meadows, leaf N concentration and specific leaf area were positive predictors of aboveground biomass (Doležal et al., 2011; Li et al., 2018). Interestingly, in a study in forests

Finegan et al. (2015) found understory aboveground biomass rate to be positively predicted specific leaf area of the understory and by tree height. The finding of Finegan et al. (2015) that

tree height was an understory biomass predictor is similar to the finding in this study that basal area was significant predictor of understory biomass. It is likely that in forests, and particularly in plantations, that tree traits can be linked with the mass-ratio hypothesis because it accounts for the greatest proportion of the overall biomass, acting as dominant determinants of ecosystem properties, such as understory vegetation aboveground biomass and vertical structure.

The importance of canopy characteristics as factors influencing the understory herbaceous layer in forests because of its influences on light regimes was emphasized by Bratton

(1976). Forest understory vegetation biomass decreased as a consequence of reduced light availability as overstory canopy coverage and leaf area increased (Del Rio and Berg, 1979; Hale,

2003). The relationship between basal area and canopy coverage was explored in several cases

(Buckley et al., 1999; Cade, 1997; Halpern and Lutz, 2013; Jameson, 1967; Mitchell and

Popovich, 1997; Parker, 2014), indicating a positive relationship between these parameters.

Those studies were in line with this study, where overstory basal area was a significant negative predictor for understory aboveground biomass and vegetation density at 51-100 cm layer

(COV2). Similar findings were observed in different eucalyptus woodlands of Central

Queensland (Australia), where herbaceous layer biomass was negatively correlated with basal area of the overstory (Scanlan and Burrows, 1990). Neither LAI nor total daily radiation under the canopy were found as predictors for aboveground biomass or density scores in this study, despite their importance in determining growth and characteristics of understory vegetation communities. Their exclusion as predictors may be due to the fact that both LAI and radiation are derived from hemispherical photography, and they may lack of species-specific conversion factors.

Some of the functional traits were important determinants of understory vertical structure, as evidenced by the regression models. The role of species richness and CWMs for plant height

and leaf area may have been related to competition with overstory, foliar adaptation to light regimes, and the effects of fertilization in eucalyptus plantations. Foliage height diversity and percent vegetation cover were important characteristics in determining avian species diversity

(Karr and Roth, 1971), which increased as vertical heterogeneity increased (Rotenberry and

Wiens, 1980). Understory vegetation vertical structure also influenced arthropod abundance, diversity, abundance, and size distribution (de Souza and de Souza Módena, 2004), since it is related to prey abundance, availability of refuges from predators and favorable microclimate conditions (Gunnarsson, 1990; Halaj et al., 1998). In a study conducted on the same sites, similar levels of family richness, biodiversity, and biomass were observed in the three plantations for terrestrial arthropods, while arboreal arthropods were least abundant in eucalyptus sites. Those findings indicated that insectivorous fauna may be able to feed effectively in eucalyptus plantations, at least during the early stages of rotation (Messick et al., 2018), The higher vegetation density found in Camden white gum in the layer comprised between 101-150 cm might increase their habitability and food source opportunities for insectivorous fauna.

4.5. CONCLUSIONS

This study was conducted in an area where natural habitats and native vegetation have been already modified by the land-use change from longleaf savannas and woodlands to slash pine plantations. The conversion of slash pine sites to Camden white gum did not reveal substantial changes in ecosystem functional attributes (functional indices and traits) and properties (understory aboveground biomass and vertical structure), maintaining the trend initiated by slash pine plantations.

Differences were found in some understory vegetation traits (understory plant height, leaf area, and leaf N concentration), but no differences were found in functional indices (richness,

evenness, and divergence) between Camden white gum and slash pine plantations. These results together indicated resilience to the land-use change from slash pine to Camden white gum. The functional indices and traits observed in this study were selected because of their potential to govern understory aboveground biomass and vertical structure, and the lack of differences in functional indices suggested functional redundancy of the dominant understory species of these plantations in determining selected understory characteristics.

In particular, understory aboveground biomass was controlled by basal area rather than by any of the functional indices or traits observed. The dominant influence of overstory on understory biomass was likely due to this comparison occurring between plantations rather than mixed-age and/or mixed-tree species forests. On the other hand, vegetation vertical structure was controlled by understory plant height, leaf area, and functional evenness. The Camden white gum plantations were relatively vertically heterogeneous, which could influence the animal fauna they could support. However, further study over longer time periods, possibly with supplementary traits relative to complementary understory characteristics, will be necessary to more comprehensively understand whether these plantation types have longer-term similarity in functional diversity associated with understory biomass productivity.

4.6. REFERENCES

Abràmoff, M.D., Magalhães, P.J., Ram, S.J., 2004. Image processing with ImageJ. Biophotonics Int. 11, 36–42.

Balvanera, P., Pfisterer, A.B., Buchmann, N., He, J., Nakashizuka, T., Raffaelli, D., Schmid, B., 2006. Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecol. Lett. 9, 1146–1156. https://doi.org/10.1111/j.1461-0248.2006.00963.x

Bastian, O., Haase, D., Grunewald, K., 2012. Ecosystem properties, potentials and services – The

EPPS conceptual framework and an urban application example. Ecol. Indic. 21, 7–16. https://doi.org/https://doi.org/10.1016/j.ecolind.2011.03.014

Bergès, L., Feiss, T., Avon, C., Martin, H., Rochel, X., Dauffy ‐ Richard, E., Cordonnier, T., Dupouey, J.‐ L., Wulf, M., 2017. Response of understorey plant communities and traits to past land use and coniferous plantation. Appl. Veg. Sci. 20, 468–481.

Binkley, D., Fisher, R.F., 2013. Nutrition Management, in: Ecology and Management of Forest Soils, Wiley Online Books. John Wiley & Sons, pp. 254–275. https://doi.org/doi:10.1002/9781118422342.ch14

Biswas, S.R., Mallik, A.U., 2011. Species diversity and functional diversity relationship varies with disturbance intensity. Ecosphere 2, 1–10. https://doi.org/10.1890/ES10-00206.1

Biswas, S.R., Mallik, A.U., 2010. Disturbance effects on species diversity and functional diversity in riparian and upland plant communities. Ecology 91, 28–35. https://doi.org/10.1890/08-0887.1

Bratton, S.P., 1976. Resource Division in an Understory Herb Community: Responses to Temporal and Microtopographic Gradients. Am. Nat. 110, 679–693. https://doi.org/10.1086/283097

Buckley, D.S., Isebrands, J.G., Sharik, T.L., 1999. Practical field methods of estimating canopy cover, PAR, and LAI in Michigan oak and pine stands. North. J. Appl. For. 16, 25–32.

Cade, B.S., 1997. Comparison of Tree Basal Area and Canopy Cover in Habitat Models: Subalpine Forest. J. Wildl. Manage. 61, 326–335. https://doi.org/10.2307/3802588

Cardinale, B.J., Matulich, K.L., Hooper, D.U., Byrnes, J.E., Duffy, E., Gamfeldt, L., Balvanera, P., O’Connor, M.I., Gonzalez, A., 2011. The functional role of producer diversity in ecosystems. Am. J. Bot. 98, 572–592. https://doi.org/10.3732/ajb.1000364

Cardinale, B.J., Srivastava, D.S., Emmett Duffy, J., Wright, J.P., Downing, A.L., Sankaran, M., Jouseau, C., 2006. Effects of biodiversity on the functioning of trophic groups and ecosystems. Nature.

Carpenter, S., Walker, B., Anderies, J.M., Abel, N., 2001. From Metaphor to Measurement: Resilience of What to What? Ecosystems 4, 765–781. https://doi.org/10.1007/s10021-001- 0045-9

Chai, Y., Liu, X., Yue, M., Guo, J., Wang, M., Wan, P., Zhang, X., Zhang, C., 2015. Leaf traits in dominant species from different secondary successional stages of deciduous forest on the Loess Plateau of northern China. Appl. Veg. Sci. 18, 50–63. https://doi.org/10.1111/avsc.12123

Chanteloup, P., Bonis, A., 2013. Functional diversity in root and above-ground traits in a fertile grassland shows a detrimental effect on productivity. Basic Appl. Ecol. 14, 208–216. https://doi.org/https://doi.org/10.1016/j.baae.2013.01.002

Crisp, M.D., Cook, L.G., 2012. Phylogenetic niche conservatism: what are the underlying evolutionary and ecological causes? New Phytol. 196, 681–694. https://doi.org/10.1111/j.1469-8137.2012.04298.x

Da S. Pontes, L., Soussana, J. ‐F., Loualult, F., Andueza, D., Carrère, P., 2007. Leaf traits affect the above‐ground productivity and quality of pasture grasses. Funct. Ecol. 21, 844–853. https://doi.org/10.1111/j.1365-2435.2007.01316.x de Souza, A.L.T., de Souza Módena, É., 2004. Distribution of Spiders on Different Types of Inflorescences in the Brazilian Pantanal. J. Arachnol. 32, 345–348.

Del Rio, E., Berg, A., 1979. Specific Leaf Area of Douglas-fir Reproduction As Affected by Light and Needle Age. For. Sci. 25, 183–186.

Di Rienzo, J.A., Casanoves, F., Pla, L., 2008. FDiversity.

Diaz, S., Cabido, M., 2001. Vive la difference: plant functional diversity matters to ecosystem processes. Trends Ecol. Evol. 16, 646–655. https://doi.org/10.1016/s0169-5347(01)02283-2

Doležal, J., Mašková, Z., Lepš, J., Steinbachová, D., de Bello, F., Klimešová, J., Tackenberg, O., Zemek, F., Květ, J., 2011. Positive long-term effect of mulching on species and functional trait diversity in a nutrient-poor mountain meadow in Central Europe. Agric. Ecosyst. Environ. 145, 10–28. https://doi.org/https://doi.org/10.1016/j.agee.2011.01.010

Finegan, B., Peña‐Claros, M., Oliveira, A., Ascarrunz, N., Bret‐Harte, M.S., Carreño‐Rocabado, 71

G., Casanoves, F., Díaz, S., Velepucha, P.E., Fernandez, F., Licona, J.C., Lorenzo, L., Negret, B.S., Vaz, M., Poorter, L., 2015. Does functional trait diversity predict above‐ ground biomass and productivity of tropical forests? Testing three alternative hypotheses. J. Ecol. 103, 191–201. https://doi.org/10.1111/1365-2745.12346

Flynn, D.F.B., Gogol‐Prokurat, M., Nogeire, T., Molinari, N., Richers, B.T., Lin, B.B., Simpson, N., Mayfield, M.M., DeClerck, F., 2009. Loss of functional diversity under land use intensification across multiple taxa. Ecol. Lett. 12, 22–33. https://doi.org/10.1111/j.1461- 0248.2008.01255.x

Footen, P.W., Harrison, R.B., Strahm, B.D., 2009. Long-term effects of nitrogen fertilization on the productivity of subsequent stands of Douglas-fir in the Pacific Northwest. For. Ecol. Manage. 258, 2194–2198. https://doi.org/https://doi.org/10.1016/j.foreco.2009.02.033

Frew, A., Nielsen, U.N., Riegler, M., Johnson, S.N., 2013. Do eucalypt plantation management practices create understory reservoirs of scarab beetle pests in the soil? For. Ecol. Manage. 306, 275–280. https://doi.org/https://doi.org/10.1016/j.foreco.2013.06.051

Garnier, E., Cortez, J., Billès, G., Navas, M.-L., Roumet, C., Debussche, M., Laurent, G., Blanchard, A., Aubry, D., Bellmann, A., Neill, C., Toussaint, J.-P., 2004. Plant Functional Markers Capture Ecosystem Properties during Secondary Succession. Ecology 85, 2630– 2637.

Gilliam, F.S., 2007. The Ecological Significance of the Herbaceous Layer in Temperate Forest Ecosystems. Bioscience 57, 845. https://doi.org/10.1641/B571007

Givnish, T., Vermeij, G.J., 1976. Sizes and Shapes of Liane Leaves, American Naturalist. https://doi.org/10.1086/283101

Givnish, T.J., 1994. Does diversity beget stability? Nature 371, 113.

Grime, J.P., 1997. Biodiversity and Ecosystem Function: The Debate Deepens. Science (80-. ). 277, 1260–1261.

Grossman, J.J., Cavender‐Bares, J., Hobbie, S.E., Reich, P.B., Montgomery, R.A., 2017. Species richness and traits predict overyielding in stem growth in an early‐successional tree diversity experiment. Ecology 98, 2601–2614. https://doi.org/10.1002/ecy.1958

Gunnarsson, B., 1990. Vegetation Structure and the Abundance and Size Distribution of Spruce- Living Spiders. J. Anim. Ecol. 59, 743–752. https://doi.org/10.2307/4892

Halaj, J., Ross, D.W., Moldenke, A.R., 1998. Habitat Structure and Prey Availability as Predictors of the Abundance and Community Organization of Spiders in Western Oregon Forest Canopies. J. Arachnol. 26, 203–220.

Hale, S.E., 2003. The effect of thinning intensity on the below-canopy light environment in a Sitka spruce plantation. For. Ecol. Manage. 179, 341–349. https://doi.org/10.1016/S0378- 1127(02)00540-6

Halpern, C.B., Lutz, J.A., 2013. Canopy closure exerts weak controls on understory dynamics: a 30-year study of overstory-understory interactions. Ecol. Monogr. 83, 221–237.

Harrison, G.W., 1979. Stability under Environmental Stress: Resistance, Resilience, Persistence, and Variability. Am. Nat. 113, 659–669.

Haywood, J.D., 2012. Pine Straw Harvesting, Fire, and Fertilization Affect Understory Vegetation within a Louisiana Longleaf Pine Stand. South. J. Appl. For. 36, 130–135.

Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., Lodge, D.M., Loreau, M., Naeem, S., Schmid, B., Setälä, H., Symstad, A.J., Vandermeer, J., Wardle, D.A., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol. Monogr. 75, 3–35. https://doi.org/10.1890/04-0922

Hudak, A.T., Strand, E.K., Vierling, L.A., Byrne, J.C., Eitel, J.U.H., Martinuzzi, S., Falkowski, M.J., 2012. Quantifying aboveground forest carbon pools and fluxes from repeat LiDAR surveys. Remote Sens. Environ. 123, 25–40. https://doi.org/https://doi.org/10.1016/j.rse.2012.02.023

Isbell, F., Calcagno, V., Hector, A., Connolly, J., Harpole, W.S., Reich, P.B., Scherer-Lorenzen, M., Schmid, B., Tilman, D., van Ruijven, J., Weigelt, A., Wilsey, B.J., Zavaleta, E.S., Loreau, M., 2011. High plant diversity is needed to maintain ecosystem services. Nature 477, 199–202.

Ivanova, L.A., Zolotareva, N. V, Ronzhina, D.A., Podgaevskaya, E.N., Migalina, S. V, Ivanov, L.A., 2018. Leaf functional traits of abundant species predict productivity in three temperate herbaceous communities along an environmental gradient. Flora 239, 11–19. https://doi.org/https://doi.org/10.1016/j.flora.2017.11.005

Jameson, D.A., 1967. The Relationship of Tree Overstory and Herbaceous Understory Vegetation. J. Range Manag. 20, 247–249. https://doi.org/10.2307/3896260

Jost, L., 2007. Partitioning diversity into independent alpha and beta components. Ecology 88, 2427–2439. https://doi.org/10.1890/06-1736.1

Jost, L., 2006. Entropy and diversity. Oikos 113, 363–375. https://doi.org/10.1111/j.2006.0030- 1299.14714.x

Karr, J.R., Roth, R.R., 1971. Vegetation Structure and Avian Diversity in Several New World Areas. Am. Nat. 105, 423–435. https://doi.org/10.1086/282735

Lavorel, S., Garnier, E., 2002. Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the Holy Grail. Funct. Ecol. 16, 545–556. https://doi.org/10.1046/j.1365-2435.2002.00664.x

Lee-Ann, H., Martin, B., 2010. Surveying Natural Populations, Quantitative Tools for Assessing Biodiversity. Columbia University Press, Berlin, Boston. https://doi.org/10.7312/haye14620

Li, H., Yu, K., Xu, D., Li, W., Tondrob, D., Du, G., 2018. Structural, compositional and trait differences between the mature and the swamp meadow communities. J. Plant Ecol. 11, 158–167.

Loreau, M., 2000. Biodiversity and Ecosystem Functioning: Recent Theoretical Advances. Oikos 91, 3–17.

Maeshiro, R., Kusumoto, B., Fujii, S., Shiono, T., Kubota, Y., 2013. Using tree functional diversity to evaluate management impacts in a subtropical forest. Ecosphere 4, 1–17. https://doi.org/10.1890/ES13-00125.1

Magurran, A.E., 2004. Measuring biological diversity. Blackwell Pub.

Mallik, A.U., 2003. Conifer Regeneration Problems in Boreal and Temperate Forests with Ericaceous Understory: Role of Disturbance, Seedbed Limitation, and Keytsone Species Change. CRC. Crit. Rev. Plant Sci. 22, 341–366. https://doi.org/10.1080/713610860

Mason, N.W.H., Orwin, K., Lambie, S., Woodward, S.L., McCready, T., Mudge, P., 2016. Leaf economics spectrum–productivity relationships in intensively grazed pastures depend on dominant species identity. Ecol. Evol. 6, 3079–3091. https://doi.org/10.1002/ece3.1964

Messick, E.J., Comer, C.E., Blazier, M.A., Wigley, T.B., 2018. Arthropod abundance, diversity, and biomass during bird breeding season in Camden white gum and slash pine plantations in the southeastern United States. For. Sci. fxy004-fxy004.

Mitchell, J.E., Popovich, S.J., 1997. Effectiveness of basal area for estimating canopy cover of ponderosa pine. For. Ecol. Manage. 95, 45–51. https://doi.org/https://doi.org/10.1016/S0378-1127(97)00002-9

Moore, P.T., Van Miegroet, H., Nicholas, N.S., 2007. Relative role of understory and overstory in carbon and nitrogen cycling in a southern Appalachian spruce–fir forestAES Publication 7863. Utah Agricultural Experiment Station, Utah State University, Logan, Utah. Can. J. For. Res. 37, 2689–2700. https://doi.org/10.1139/X07-115

Niinemets, Ü., Portsmuth, A., Tena, D., Tobias, M., Matesanz, S., Valladares, F., 2007. Do we Underestimate the Importance of Leaf Size in Plant Economics? Disproportional Scaling of Support Costs Within the Spectrum of Leaf Physiognomy. Ann. Bot. 100, 283–303. https://doi.org/10.1093/aob/mcm107

Nudds, T.D., 1977. Quantifying the Vegetative Structure of Wildlife Cover. Wildl. Soc. Bull. 5, 113–117.

Pakeman, R.J., 2011. Functional diversity indices reveal the impacts of land use intensification on plant community assembly. J. Ecol. 99, 1143–1151. https://doi.org/10.1111/j.1365- 2745.2011.01853.x

Pardini, R., de Souza, S.M., Braga-Neto, R., Metzger, J.P., 2005. The role of forest structure, fragment size and corridors in maintaining small mammal abundance and diversity in an Atlantic forest landscape. Biol. Conserv. 124, 253–266. https://doi.org/https://doi.org/10.1016/j.biocon.2005.01.033

Parker, W.C., 2014. The relationship of stand structure with canopy transmittance: Simple models and practical methods for managing understory light conditions in eastern white pine (Pinus strobus L.)-dominated forests. For. Chron. 90, 489–497. https://doi.org/10.5558/tfc2014-099

Pérez-Harguindeguy, N., Díaz, S., Garnier, E., Lavorel, S., Poorter, H., Jaureguiberry, P., Bret- Harte, M.S., Cornwell, W.K., Craine, J.M., Gurvich, D.E., 2013. New handbook for standardised measurement of plant functional traits worldwide. Aust. J. Bot. 61, 167–234.

Petchey, O.L., Gaston, K.J., 2002. Functional diversity (FD), species richness and community composition. Ecol. Lett. 5, 402–411. https://doi.org/10.1046/j.1461-0248.2002.00339.x

Pla, L., Casanoves, F., Di Rienzo, J., 2011. Quantifying functional biodiversity. Springer Science & Business Media.

Prado‐Junior, J.A., Schiavini, I., Vale, V.S., Arantes, C.S., Sande, M.T., Lohbeck, M., Poorter, L., 2016. Conservative species drive biomass productivity in tropical dry forests. J. Ecol. 104, 817–827. https://doi.org/10.1111/1365-2745.12543

Rainey, S.M., Nadelhoffer, K.J., Silver, W.L., Downs, M.R., 1999. Effects of chronic nitrogen additions on understory species in a red pine plantation. Ecol. Appl. 9, 949–957. https://doi.org/10.1890/1051-0761(1999)009[0949:EOCNAO]2.0.CO;2

Rasband, W.S., 2005. ImageJ (computer program). US Natl. Institutes Heal. Bethesda, MD, USA.

Roscher, C., Schumacher, J., Lipowsky, A., Gubsch, M., Weigelt, A., Pompe, S., Kolle, O., Buchmann, N., Schmid, B., Schulze, E.-D., 2013. A functional trait-based approach to understand community assembly and diversity–productivity relationships over 7 years in experimental grasslands. Perspect. Plant Ecol. Evol. Syst. 15, 139–149. https://doi.org/http://dx.doi.org/10.1016/j.ppees.2013.02.004

Rosenfeld, J.S., 2002. Functional Redundancy in Ecology and Conservation. Oikos 98, 156–162.

Rotenberry, J.T., Wiens, J.A., 1980. Habitat Structure, Patchiness, and Avian Communities in North American Steppe Vegetation: A Multivariate Analysis. Ecology 61, 1228–1250. https://doi.org/10.2307/1936840

Ruiz‐Jaen, M.C., Potvin, C., 2011. Can we predict carbon stocks in tropical ecosystems from tree diversity? Comparing species and functional diversity in a plantation and a natural forest. New Phytol. 189, 978–987. https://doi.org/10.1111/j.1469-8137.2010.03501.x

Russell, M.B., Woodall, C.W., Potter, K.M., Walters, B.F., Domke, G.M., Oswalt, C.M., 2017. Interactions between white-tailed deer density and the composition of forest understories in the northern United States. For. Ecol. Manage. 384, 26–33. https://doi.org/https://doi.org/10.1016/j.foreco.2016.10.038

Sasaki, T., Okubo, S., Okayasu, T., Jamsran, U., Ohkuro, T., Takeuchi, K., 2009. Two‐phase functional redundancy in plant communities along a grazing gradient in Mongolian rangelands. Ecology 90, 2598–2608. https://doi.org/10.1890/08-1850.1

Scanlan, J.C., Burrows, W.H., 1990. Woody overstorey impact on herbaceous understorey in Eucalyptus spp. communities in central Queensland. Aust. J. Ecol. 15, 191–197. https://doi.org/10.1111/j.1442-9993.1990.tb01527.x

Schwartz, M.W., Brigham, C.A., Hoeksema, J.D., Lyons, K.G., Mills, M.H., van Mantgem, P.J., 2000. Linking biodiversity to ecosystem function: implications for conservation ecology. Oecologia 122, 297–305. https://doi.org/10.1007/s004420050035

Smith, J.E., Heath, L.S., Hoover, C.M., 2013. Carbon factors and models for forest carbon estimates for the 2005–2011 National Greenhouse Gas Inventories of the United States. For. Ecol. Manage. 307, 7–19. https://doi.org/https://doi.org/10.1016/j.foreco.2013.06.061

Spasojevic, M.J., Suding, K.N., 2012. Inferring community assembly mechanisms from functional diversity patterns: the importance of multiple assembly processes. J. Ecol. 100, 652–661. https://doi.org/10.1111/j.1365-2745.2011.01945.x

Spehn, E.M., Hector, A., Joshi, J., Scherer-Lorenzen, M., Schmid, B., Bazeley-White, E.,

Beierkuhnlein, C., Caldeira, M.C., Diemer, M., Dimitrakopoulos, P.G., Finn, J.A., Freitas, H., Giller, P.S., Good, J., Harris, R., Hogberg, P., Huss-Danell, K., Jumpponen, A., Koricheva, J., Leadley, P.W., Loreau, M., Minns, A., Mulder, C.P.H., O’Donovan, G., Otway, S.J., Palmborg, C., Pereira, J.S., Pfisterer, A.B., Prinz, A., Read, D.J., Schulze, E.D., Siamantziouras, A.S.D., Terry, A.C., Troumbis, A.Y., Woodward, F.I., Yachi, S., Lawton, J.H., 2005. Ecosystem effects of biodiversity manipulations in European grasslands. Ecol. Monogr. 75, 37–63. https://doi.org/10.1890/03-4101

Stallings, J.R., 1991. The importance of understorey on wildlife in a Brazilian eucalypt plantation. Rev. Bras. Zool. 7, 267–276. https://doi.org/10.1590/S0101- 81751990000300008

Stephens, S.S., Wagner, M.R., 2007. Forest plantations and biodiversity: a fresh perspective. J. For. 105, 307–313.

Tilman, D., Wedin, D., Knops, J., 1996. Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379, 718.

Tyynelä, T.M., 2001. Species diversity in Eucalyptus camaldulensis woodlots and miombo woodland in Northeastern Zimbabwe. New For. 22, 239–257. https://doi.org/10.1023/A:1015616010976

VanderSchaaf, C.L., 2008. Estimating understory vegetation response to multi-nutrient fertilization in Douglas-fir and ponderosa pine stands. J. For. Res. 13, 43–51. https://doi.org/10.1007/s10310-007-0050-8

Villéger, S., Mason, N.W.H., Mouillot, D., 2008. New multidimensional functional diversity indices for a multifaceted framework in functional ecology. Ecology 89, 2290–2301. https://doi.org/10.1890/07-1206.1

Walter, C., Raiff, D., Burnham, M., Gilliam, F., Adams, M., Peterjohn, W., 2016. Nitrogen fertilization interacts with light to increase Rubus spp. cover in a temperate forest. Plant Ecol. 217, 421–430.

Westoby, M., 1998. A leaf-height-seed (LHS) plant ecology strategy scheme. Plant Soil 199, 213–227. https://doi.org/10.1023/A:1004327224729

Yarie, J., 1980. The Role of Understory Vegetation in the Nutrient Cycle of Forested Ecosystems in the Mountain Hemlock Biogeoclimatic Zone. Ecology 61, 1498–1514. https://doi.org/10.2307/1939057

Zhang, D., Zhang, J., Yang, W., Wu, F., Huang, Y., 2014. Plant and soil seed bank diversity across a range of ages of Eucalyptus grandis plantations afforested on arable lands. Plant Soil 376, 307–325. https://doi.org/10.1007/s11104-013-1954-z

Zhang, Q., Buyantuev, A., Li, F.Y., Jiang, L., Niu, J., Ding, Y., Kang, S., Ma, W., 2017. Functional dominance rather than taxonomic diversity and functional diversity mainly affects community aboveground biomass in the Inner Mongolia grassland. Ecol. Evol. 7, 1605–1615. https://doi.org/10.1002/ece3.2778

CHAPTER 5. UNDERSTORY VEGETATION COMMUNITIES AS STRUCTURED BY OVERSTORY, CANOPY, AND SOILS IN EUCALYPTUS BENTHAMII AND PINUS ELLIOTTII PLANTATIONS IN SOUTHWESTERN LOUISIANA

5.1. INTRODUCTION

The diffusion of forestry plantations has raised concerns in terms of biodiversity, becoming a major issue between foresters and conservationists (Calviño-Cancela et al., 2012). In particular, the ability of plantations to develop ecological characteristics comparable to natural ecosystems has been one of the main research topics in forestry (Aubin et al., 2008). Eucalyptus plantations may increase local biodiversity by offering regeneration sites for native plant species from adjacent ecosystems (Brockerhoff et al., 2008). In a study conducted on degraded Ethiopian highlands, about 83% of woody species found in the adjacent natural forests were also found in the E. globulus plantation understory (Yirdaw and Luukkanen, 2003). E. urophylla x E. grandis plantations did not alter the understory richness when the plantations were established on clearcut Chinese fir (JianPing et al., 2015), E. camaldulensis plantations established on degraded land in Malawi had similar herbaceous species composition relative to nearby woodlands (Bone et al.,

1997), and E. globulus, E. grandis, and E. saligna plantations established in Ethiopian highlands had similar species richness and composition when compared to local natural forests (Michelsen et al., 1996). In South Africa, a large number of native forest species were found in the understory of E. grandis stands in the proximity of the forest edge (Geldenhuys, 1997). In

Congo, E. alba × E. urophylla plantations were generally intermediate between savanna and forest, indicating plantation age, density, and proximity to a natural forest stand as factors that favor increasing floristic diversity and development towards native secondary forest (Loumeto and Huttel, 1997).

In southwestern Louisiana and southeastern Texas, Camden white gum (Eucalyptus benthamii) was established on over 4,000 ha (Hart et al., 2016) on sites formerly occupied by

loblolly (Pinus taeda) and slash pine (Pinus elliottii) plantations (Blazier et al., 2012). Camden white gum plantations rely on intensive management practices that are often detrimental for native understory species conservation, such as heavy mechanical and chemical site preparation, pre- and post-emergence, and yearly herbicide applications, multiple fertilization, and clear-cut harvesting often with short rotations.

Generally, non-native plantations support less species compared to the natural ecosystem they have replaced (Tyynelä, 2001). Native species plantations seemed to mimic natural forest conditions, so they are able to support more biodiversity compared to non-native plantations

(Aubin et al., 2008). Therefore, the understory habitat provided by slash pine plantations in southwestern Louisiana could be more ecologically similar to that provided by native longleaf pine savannas and woodlands, suggesting a potential negative effect for understory species in the land-use change towards eucalyptus. Moreover, the short rotation length in Camden white gum

(as little as 5-10 years for compared to 25-35 for slash pine – see Chapter 1), may negatively influence the establishment of native understory species, since the length of rotation has been broadly advocated to promote native species diversity in plantations (Humphrey, 2005).

To evaluate the impact of Camden white gum on understory vegetation, it is necessary to understand its impacts on the understory habitat, specifically on light availability and soil characteristics (Barbier et al., 2008). Understory light is closely related to the canopy structure, and its influence on air temperature and humidity makes it a synthetic factor grouping microclimatic variations (Barkman, 1992). Canopy properties, such as spatial arrangement of the leaves (Horn, 1971) and leaf size (Barkman, 1992), influence light transmittance to understory; those properties can be approximately described using leaf area index (LAI). Differences in topsoil characteristics can trigger differences in understory species composition and diversity

(Barbier et al., 2008). Different species have different requirements in terms of soil nutrient

contents and pH (Wherry, 1927), and understory species composition is often used as an indicator of site acidity (Barbier et al., 2008). With regards to soil reaction, it is commonly accepted that topsoil in conifer stands is more acidic than in hardwoods, but this generalization is sometimes wrong in case of eucalyptus species, which were found to lower soil pH (Rhoades and

Binkley, 1996; Temesgen et al., 2016)

Although the understory vegetation was analyzed in terms of species composition, functional traits, and diversity in the previous chapters, the relationships between understory vegetation communities, overstory, canopy, and soil variables have not been studied in detail.

Without such knowledge, the impacts of converting slash pine to eucalyptus on biodiversity and ecosystem services cannot be fully understood, and management guidelines for sustainable plantations cannot be properly developed. The main goal of this study was, thus, to explore the relationships between understory vegetation communities of Camden white gum and slash pine plantations and the physical and biological gradients related to overstory, canopy, and soil characteristics. Specifically, we aimed to answer the following research questions: (1) How were understory vegetation communities organized in Camden white gum, same-age, and older slash pine plantations? (2) How is the composition of understory vegetation controlled by overstory dimensions, canopy complexity, and soil characteristics?

5.2. METHODS

Information about study locations, experimental design, floristic surveys, and soil and overstory sampling are provided in Chapter 2, while respective data analysis and results are reported in Chapter 3.

5.2.1. Data analysis

To analyze changes in understory species composition in relation to plantation types, we performed nonmetric multidimensional scaling (NMDS) using Bray-Curtis measure of dissimilarity (Bray and Curtis, 1957) for years 2014-2016. NMDS is a suitable method for analyzing ecological data (Kenkel and Orlóci, 1986; McCune et al., 2002; Minchin, 1987a,

1987b), and when associated with Bray-Curtis dissimilarity has been determined to be a robust and effective ordination compared to other methods, such as principal components analysis or correspondence analysis, especially in case of lack of linear or symmetric unimodal fitting of species responses to ecological gradients (Minchin, 1987a, 1987b). Bray-Curtis dissimilarity is a non-metric method for measuring distance between samples that is well-suited for ecological data collected at different sampling locations (Legendre and Legendre, 2012a). The Bray-Curtis dissimilarity is given by the following equation:

2-'5 4-'5 = 1 − 8' + 85

Where i and j represent two sites, S represents the species abundance, and Cij is the sum of only the lesser abundance for each species found in both site. Legendre and Legendre (2012b) defined

NMDS as an unconstrained method based only on vegetation dissimilarities, representing data in a few dimensions so that distances in the ordination reflect the similarities between sampling sites. NMDS focuses on the representation of the objects in a small and specified number of dimensions (usually 2-3) rather than the exact preservation of the distances among objects.

Therefore, the goal is to plot dissimilar objects far apart in the ordination space and similar objects close to one another, preserving the ordering relationships among objects.

A useful way to evaluate the suitability of an NMDS is provided by the Shepard diagram, which compares for a given distance matrix Dh,i the fitted distances :;,' to the empirical 82

= distances Dh,i and regresses :;,' on Dh,i, giving the forecasted distances :;,'. Two type of regressions are considered, linear and monotone (nonparametric). A monotone regression can be considered as linear regression executed after monotonic transformation that maximizes the linear relationship between Dh,i and :;,', fitted by least squares. An objective function (stress) is also used to assess the goodness of fit of NMDS. Stress is defined by the following equation:

F ∑ D: − := E 8>?@AA = B ;,' ;,' ;,' ̅ F ∑;,'D:;,' − :;,'E

̅ Where :;,' are the mean distance values. Stress values around 0.1 are generally considered good, while values less than 0.05 are considered excellent (Clarke, 1993). Increasing the number of dimensions typically leads to lower values of stress and better fits. However, it can be very difficult to interpret if stress values are more than 2 or 3 due to the unconstrained nature of axes.

For this reason, both stress value and number of interpretable dimensions were considered in selecting the number of dimensions.

The vegan package (Oksanen et al., 2018) in R version 3.5.1 (R Core Team, 2018) was used to compute NMDS through the metaMDS() function, using stress values and Shepard plots of the ordination distances against original dissimilarities to find a convergent solution

(Haeussler et al., 2017). The function metaMDS() performs NMDS according to the criteria and recommendations specified by Minchin (1987a, 1987b), correcting the indeterminacy of scaling and orientation of axes through (1) moving the origin to the average of the axes, and (2) rotating the configuration using principal components so the variance of points is maximized on first dimension (Oksanen et al., 2018). The rotation assures that the first NMDS axis lies along the direction of maximum scatter among the objects, minimizing their overlapping and improving

the visualization. NMDS is suited to gradient analysis when the patterns in ordination corroborate existing knowledge or a hypothesis. A gradient analysis investigates relationships between ordination and environmental variables by correlating the ordination scores with those variables. Through gradient analysis it is often possible to interpret ordination using ecological variables. One of the most commonly used methods such an analysis is to fit environmental vectors onto ordination. The fitted arrows foster identification of (1) the direction of the gradient, defined as the direction of most rapid change in the environmental variable and represented by the direction pointed by the arrow of the environmental vector; and (2) the strength of the gradient, which is given by the length of the arrow (proportional to the correlation between ordination and environmental variable).

The function “envfit()” of the vegan package was used to fit environmental vectors related to overstory (DBH, tree height, and basal area), canopy (LAI and total daily radiation under the canopy), and soil (nutrients and pH – see Chapter 3) onto an ordination. However, overstory and soil variables were only available for 2016, which constrained their analyses for the last year of the study. Permutations were used to assess the relationships between ordination and environmental variables described in the previous Chapters. The degree of fit (R2) between environmental variables and NMDS axes were determined by 999 permutations generated by repeatedly shuffling the actual data (Oksanen et al., 2018). Furthermore, linearity of the assemblage–environment relationships was investigated by adding thin-plate surfaces fitted onto the NMDS ordination using penalized splines (Wood 2003) with the ordisurf() function of the vegan package.

In the case of two NMDS axes, envfit() fits the following regression model (ter Braak,

1995):

H = I,2, + IF2F + J

th where y is the environmental variable, bi is the slope coefficient relative to the i NMDS axis, xi represents the ith NMDS axis scores, and ε is the unexplained variance. The model has no intercept because both y and xi are centered prior to fitting the model. The permutation test permutes the values of the response y into random order. The randomized response data are then predicted through the fitted regression model and the R2 between the randomized response and the fitted values from the model is computed. The R2 value is recorded and then the procedure is repeated with a different random permutation for a predetermined number of times. Under the null hypothesis of no relationship between the ordination axes scores and the environmental variable, the observed R2 value should be a common value among the permuted R2 values. If however the observed R2 is extreme relative to the permutation distribution of R2, then it is unlikely that the null hypothesis is true. The proportion of times a randomized R2 from the distribution is equal to or greater than the observed R2 gives the permutation P-value.

Contrary to other ordination techniques based on eigenvector methods, such as principal component analysis, NMDS computation does not maximize the variability associated with individual axes of the ordination, making axes arbitrary, and plots may arbitrarily be rotated, centered, or inverted (Legendre and Legendre, 2012b). Moreover, is not directly possible to estimate a loading of environmental variables on NMDS axes, but it is possible to estimate slope coefficients between each variables and dimensions, and these coefficients can be interpreted as loadings. However, direct comparisons between NMDS axes are not possible because of their independence.

Permutational multivariate analysis of variance (PERMANOVA) was used to fit linear models between each environmental variables and NMDS axes through the adonis() function of the vegan package (999 permutations). Anderson (2017) defined PERMANOVA as a semiparametric method of geometric partitioning of multivariate variation in the space of a

selected measure of dissimilarity, according to a given ANOVA design, with p-values calculated using appropriate distribution-free permutation techniques. PERMANOVA allows tests and estimation of sizes of main effects, interaction terms, hierarchical structures, and random components in mixed models while preserving statistical characteristics of rank-based nonparametric multivariate methods, such as the ability to base the analysis on a dissimilarity measure of choice and distribution-free inferences achieved by permutations, with no assumption of multivariate normality. Exchangeability of permutable units under a true null hypothesis is the only assumption required (Fisher, 1935; Kempthorne, 1966); inferences are distribution-free, but a linear model is applied to the dissimilarity space. PERMANOVA produces pseudo-F statistics to test the null hypothesis of no differences in the centroids among groups given any differences in within-group dispersions. PERMANOVA generates partial R2 coefficients for each of the factors included in the model, and they are obtained by dividing the sum of square of each factor by total sum of squares. Differences in NMDS scores (sites x scores matrix) were also tested using PERMANOVA run with 999 permutations. In case of significant differences, post hoc pairwise tests were performed to identify significant differences among the plantation types. Post hoc significance tests were adjusted using the Bonferroni correction control for type I error rate

(α = 0.05).

5.3. RESULTS

5.3.1. Understory dynamics

The number of dimensions selected was k=3 for all the three years of the analysis (Table

5.1) with stress values less than 0.01 for all the three years of the analysis (Fig. 5.1). The Number of dimension did not compromise the interpretability of NMDS plots, maintaining a good value of fit, as can be noted by the stress value. Shepard plots with R2 values for the linear and non-

linear (monotone) regressions are shown in Fig. 5.2. To facilitate the interpretability, objects of

NMDS (species and sites) were plotted on two dimensions (axes 1 and 2 of NMDS).

Table 5.1 NMDS stress values for each k=1-5 dimension during 2014-2015. k Stress (2014) Stress (2015) Stress (2016) 5 0.031 0.037 0.037 4 0.054 0.053 0.049 3 0.090 0.077 0.069 2 0.168 0.117 0.104 1 0.288 0.289 0.226

2014 2015 0.30 0.30 0.25 0.25 0.20 0.20 0.15 0.15 Stress Stress 0.10 0.10 0.05 0.05 0.00 0.00

1 2 3 4 5 6 7 1 2 3 4 5 6 7

Dimensions Dimensions

2016 0.30 0.25 0.20 0.15 Stress 0.10 0.05 0.00

1 2 3 4 5 6 7 Dimensions Fig. 5.1. Scree plots (stresses vs. number of dimensions) related to NMDS ordination for years 2014, 2015, and 2016.

2014 2015

2 2

1.4 Non−metric fit, R = 0.992 Non−metric fit, R = 0.994 1.2

2 Linear fit, R2 = 0.925 Linear fit, R = 0.953 1.2 1.0 1.0 0.8 0.8 0.6 0.6 Ordination Distance Ordination Distance 0.4 0.4 0.2 0.2 0.5 0.6 0.7 0.8 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85

Observed Dissimilarity Observed Dissimilarity

2016

Non−metric fit, R2 = 0.995

2 1.5 Linear fit, R = 0.967 1.0 Ordination Distance 0.5

0.5 0.6 0.7 0.8 0.9

Observed Dissimilarity

Fig. 5.2. NMDS Shepard diagrams for years 2014-2016.

Results of PERMANOVA (Table 5.2) detected a difference among the plantations in

NMDS scores for all three years. Specifically, NMDS scores of Camden white gum sites were different from those of older slash pine but similar to those same-age slash pine in 2014 and

2015. Differences in NMDS scores were found among all three plantation types in 2016.

NMDS ordination plots are shown in Fig. 5.3. In 2014, species found only within confidence ellipses for older slash pine sites included tree species (Magnolia virginiana,

Liquidambar styraciflua), shurbs (Morella cerifera, Ilex opaca), vines (Vitis rotundifolia,

Campsis radicans, Gelsemium sempervirens). Camden white gum confidence ellipses included mostly forbs (Eupatorium serotinum, Centrosema virginiamun, Rhynchosia tomentosa,

Gamochaeta purpurea, Eupatorium rotundifolius), shrubs (Rhus copallina, Vaccinum elliottii,

Hypericum galioides), graminoids (Setaria parviflora, Fuirena breviseta). Same-age slash pine confidence ellipses shared the totality of all its species with Camden white gum, including a large number of forbs (Solidago rugosa, Solidago altissima, Eupatorium perfoliatum, Centaurum pulchellum, Helianthus angustifolius), some graminoids (Rhyncospora caduca), few shrubs

(Hypericum stans, Vaccinium arboreum). Older slash pine and Camden white gum shared shrubs

(Ilex vomitoria, Symplocos tinctoria, Rubus trivilias, Callicarpa americana), vines (Smilax glauca, Lygodium japonicum), and graminoids (Schizachyrium scoparium, Panicum verrucosum). Older slash pine sites, close in the ordination to shrubs, vines, and tree saplings, were distant from same-age slash pine typical of closed-canopy environments, which understory was close to forbs and herbs common of open environments. Camden white gum position was intermediate, totally encompassing same-age slash pine, and partially overlapping older slash pine.

In 2015, shrubs (Morella cerifera, Ilex opaca, Symplocos tinctoria, Rubus trivialis,

Vaccinium elliotti, Ilex vomitoria), vines (Gelsemium sempervirens, Vitis rotundifolia, Smilax glauca), trees (Liquidambar styraciflua), forbs (Eupatorium resinosum, Diodia virginiana,

Lespedeza capitata), and graminoids (Panicum verrucosum, Rhyncospora inexpansa) were contained exclusively within older slash pine 95% confidence ellipses. Confidence ellipses of same-age slash pine and Camden white gum partially overlapped, even though showing few share species (Helianthus angustifolius, Campsis radicans, Xyris stricta). Camden white gum confidence ellipses included forbs (Scutellaria integrifolia, Solidago altissma, Conyza canadensis, Euphorbia corollata, Pteridium aquilinum, Centrosema virginianum, Monarda punctata, Eupatorium perfoliatum, Tephrosia Onobrychoides) graminoids (Dichanthelim aciculare, Dichanthelium scabrisculum), shrubs (Hypericum stans, Baccharis halimifolia,

Vaccinium arboreum). Same-age slash pine ellipses included mostly forbs (Eutamia leptocephala, Lobelia appendiculata), and graminoids (Juncus polycephalus). The distance patterns of the two slash pine plantations from the understory vegetation were similar to those observed the previous year. Camden white gum sites were separated now from slash older slash pine.

In 2016, slash pine confidence ellipses included shrubs (Morella cerifera, Ilex opaca,

Symplocos tinctoria, Vaccinium elliotti, Ilex vomitoria, Rubus trivilias), vines (Smilax glauca,

Toxicodendron radicans, Gelsemium semprevires, Vitis rotundifolia), graminoids (Panicum verrucosum), trees (Liquidambar styraciflua) and few forbs (Diodia virginiana). Camden white gum confidence ellipses included forbs (Centella erecta, Conyza canadensis, Eupatorium perfoliatum, Lobelia puberula, Lobelia appendiculata, Monarda punctata) and shrubs

(Callicarpa americana, Hypericum stans). Same-age slash pine confidence ellipses included several forbs (Tephrosia onobrychoides, Diodia teres, Linum medium var. texanum, Xyris ambigua, Euthamia leptpcephala, Solidago altissima, Solidago rigosa, Polygola mariana,

Helianthus angustigolius, Rhexia lutea). The three plantation types appeared separated in 2016, with same-age slash pine placed distant from older slash pine and eucalyptus sites.

In 2015, the superimposition of overstory variables on ordination space as vectors and surfaces (Table 5.3, Fig. 5.4, and Fig. 5.5) showed the placement of older slash pine sites on gradients of greater tree dimensions (DBH, height, and basal area), while same-age slash pine sites were placed along gradients of smaller tree dimensions. The position of Camden white gum sites was intermediate between the two slash pine plantations. DBH, height, and basal area

PERMANOVA models showed positive first NMDS axis slope coefficients, with R2 values of

0.79, 0.62, and 0.68 respectively (Table 5.4). The second NMDS axis showed a negative slope for DBH (R2=0.09) and height (R2=0.24) PERMANOVA models.

In 2016, eucalyptus sites were placed on gradients of increasing tree height, while older slash pine sites along gradients of greater tree diameter and basal area (Table 5.3, Fig. 5.6, and

Fig. 5.7). The position of same-age slash pines was opposite to Camden white gum in terms of height, and opposite to older slash pine in terms of DBH and basal area. PERMANOVA models for DBH, height, and basal area indicated negative slopes for the first NMDS axis and R2 values of 0.76, 0.91, and 0.60 respectively (Table 5.4). The second NMDS axis slopes were negative for

DBH (R2=0.16) and basal area (R2=0.13) PERMANOVA models.

The two slash pine plantations were placed in almost antipodal position in relation to LAI and total daily radiation under the canopy (Table 5.3, Fig. 5.6, and Fig. 5.8), with older slash pine sites associated with gradients of high crown development and same-age slash pine sites with gradients of high levels of solar radiation available in the understory. The position of

Camden white gum sites was similar to those of older slash pine. Slopes of the first NMDS axis were negative for LAI (R2=0.81) and positive for radiation (R2=0.90) PERMANOVA models

(Table 5.4). The second NMDS axis slopes were negative for LAI (R2=0.11) and positive for radiation (R2=0.05) PERMANOVA models.

Among soil characteristics, P, K, Cu, Na, S, and pH had a significant correlation with ordination space (Table 5.3, Fig. 5.6, and Fig. 5.9). Same-age slash pine sites were placed along gradients of higher levels of pH and Cu, while Camden white gum and older slash pine sites were showed gradients of increased soil acidity and lower soil Cu concentrations. Camden white gum sites were on gradients of soil rich in P and K, and gradients of lower S and Na. On the other hand, slash pine sites of both ages were placed along gradients of soil richer in Na and S and poor in P and K. The first NMDS axis showed a positive slope for pH (R2=0.79) and Cu

(R2=0.43) PERMANOVA models, negative for P (R2=0.04) model (Table 5.4). On the other hand, the second NMDS slopes were positive for P, K, and Cu PERMANOVA models (R2=0.29,

0.31, 0.18 respectively), and negative for S and Na (R2=0.52 and 0.28 respectively) models

(Table 5.4).

2014 2015 1.5 1.5

E.sang

1.0 Lycopodiella 1.0 L.styra L.acid C.sagiH.arteP.quinS.rotuE.sang G.semp M.ceriS08_3X.ambi. T.diveC.virg S08_36R.mari A.parv C.cana R.alifM.virg I.opac V.elli M.punc E.compP.caro D.virg V.rotuC.radi R.copa 0.5 0.5 R9C.amerR6R.tomeR8 S.tinc R.luteV.arbo J.bracH.fasc S08_32 R.luteR.trivP.pectC.cren I.opac M.ceriC.erec E.perfL.japoG.purpE.sero L.tenu S08_1S.glauS.herbL.piloS.parv P.proc S08_17D.tereI.vomiT.radi E.leucoI.vomiL.japo E.sero M.striR11 S08_17 A.aure V.rotu L.capiR.triv C.amerR6 C.erec S08_36L.styra S.inteE.coro S.scopT.radiP.verr S13_17 G.semp C.cren P.aquiL.capiV.elli R.cadu H.fasc P.verrS.glau E.capi C.virg Richness A.rubr S08_1 P.aqui Richness 0.0 E.comp H.arte C.echi 0.0 H.stan R.copaD.acic J.brac S.scopT.onobE.rotuS.altiM.acumL.appe D.scabE.rotuT.onobS13_F.brevi S.tinc D.scabR.gracD.acic R.tome E.resi B.haliR.mariS13_2E.leptR.granP.sang NMDS2 S13_26H.stan D.tere NMDS2 C.radiH.angJ.poly S.rugoE.capiA.filiH.hypeP.plicM.acum E.leuco S13_17S.rugoL.medi S13_2P.mariR.reco S08_3D.virg S13_9S.grac S.dumoR11H.angV.arboC.pulc P.procR8 S13_R10S.dumoP.plicP.mari A.parv R9E.lept R.gran H.punc A.obovBidensC.pulc C.sagi P.gramH.galiP.caro S.umbe S08_32R.inex S13_26X.striR.reco S13_9S.altiOxalisC.horriS.inte L.piloP.pect R.cadu P.tenu 0.5 N.sylv 0.5 R.plum P.crucP.digiF.brevi E.perf C.escu − C.canaA.obov − Sabatia R10G.purp

B.hali 1.0 1.0 − − 1.5 1.5 − −

−2 −1 0 1 2 −2 −1 0 1 2 2016 NMDS1 NMDS1 1.5 1.0 X.stri.obs R.grac S.altiN.sylvH.hype P.mariP.caroA.parvV.arboM.striP.sero R.caduS13_2H.anguD.acicM.acumOxalisX.ambi T.onobS.rugoJ.polyE.lept R.reco P.pect S13_6L.medi.texP.proc S13_26S13_9D.tere R.lute

0.5 S.dumo E.resi L.pilo S13_17 M.ceri E.leuc S08_3 P.aquiC.radiR.hirt S08_32D.virg L.styr R.copa E.rotu Q.nigrV.rotuS.tinc V.elli E.capi G.sempS.glau T.radiS.scop Richness 0.0 S08_36S08_1 A.fili NMDS2 S08_17 B.hali P.verru I.opacR.trivI.vomiC.erecL.appeH.stanC.virg R6 L.pube 0.5 C.amerR9 E.perf − E.hieraC.canaG.purpR11R10 L.japoLactuca E.compC.horr M.puncR8 C.conoD.scop T.perf 1.0 − 1.5 −

−2 −1 0 1 2

Fig. 5.3. Non-metric multidimensionalNMDS1 scaling (NMDS) ordination of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity. Black dots indicate site centroids, while standard deviation 95% ellipses outline position of plantation types. Superimposed vectors indicate species richness.

1.5 1.0

E.sang C.sagiH.arteP.quinS.rotu Height A.parv T.dive C.virg C.cana V.elli M.punc E.comp P.caro

0.5 DBH R8 BA R9R.copaC.amerR6R.tome V.arbo H.fasc R.lute E.perf J.brac I.opac M.ceri C.erec L.japoG.purpE.sero S08_17D.tere L.tenu P.proc I.vomiT.radi R11 R.triv M.stri V.rotu L.capi S08_36 S.inte L.styra C.cren E.coro G.semp S.glau A.rubr P.verr S08_1 E.capi P.aqui Richness 0.0 S.scop T.onobH.stanM.acum E.rotuS.alti L.appe NMDS2 S.tinc D.scabR.grac D.acicS13_2 E.resi B.haliR.mari E.lept R.gran P.sang C.radiH.angJ.poly S08_3 E.leuco S13_17S13_9S.rugoL.medi D.virg S13_R10 S.grac S.dumoP.plicP.mariBidens H.punc S08_32 S13_26 A.obovC.pulc R.inex X.stri R.reco P.tenu L.piloP.pect R.cadu 0.5 P.digi C.escu R.plum P.cruc F.brevi − Sabatia 1.0 − 1.5 −

−2 −1 0 1 2

NMDS1

Fig. 5.4. Non-metric multidimensional scaling (NMDS) ordination of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity for year 2015. Black dots indicate site centroids, while standard deviation 95% ellipses outline position of plantation types. Superimposed vectors indicate species richness, diameter at breast height (DBH), tree height (Height), and basal area (BA). 94

DBH Tree Height

E.sang E.sang C.sagiH.arteP.quinS.rotu C.sagiH.arteP.quinS.rotu

A.parv T.dive C.virg Height A.parv T.dive C.virg C.cana C.cana V.elli M.punc V.elli M.punc E.comp P.caro E.comp P.caro

0.5 R8 0.5 R8 R9R.copaC.amerR6R.tome R9R.copaC.amerR6R.tome DBH V.arbo H.fasc V.arbo H.fasc J.brac J.brac R.lute E.perf R.lute E.perf I.opac M.ceri C.erec L.japoG.purpE.sero I.opac M.ceri C.erec L.japoG.purpE.sero S08_17D.tere L.tenu S08_17D.tere L.tenu P.proc I.vomiT.radi R11 P.proc I.vomiT.radi R11 R.triv M.stri R.triv M.stri V.rotu L.capi V.rotu L.capi S08_36 S.inte S08_36 S.inte L.styra C.cren E.coro L.styra C.cren E.coro G.semp G.semp S.glau S.glau A.rubr P.verr S08_1 E.capi Richness A.rubr P.verr S08_1 E.capi Richness

0.0 P.aqui 0.0 P.aqui H.stan H.stan S.scop T.onob M.acumL.appe S.scop T.onob M.acumL.appe

E.rotuS.alti NMDS2 E.rotuS.alti S.tinc D.scabR.grac D.acicS13_2 S.tinc D.scabR.grac D.acicS13_2 E.resi B.haliR.mari E.lept R.gran P.sang E.resi B.haliR.mari E.lept R.gran P.sang C.radiH.angJ.poly C.radiH.angJ.poly S08_3 E.leuco S13_17S13_9S.rugoL.medi S08_3 E.leuco S13_17S13_9S.rugoL.medi D.virg S13_R10 S.grac D.virg S13_R10 S.grac S.dumoP.plicP.mariBidens S.dumoP.plicP.mariBidens H.punc S08_32 S13_26 A.obovC.pulc H.punc S08_32 S13_26 A.obovC.pulc R.inex X.stri R.reco R.inex X.stri R.reco P.tenu P.tenu L.piloP.pect R.cadu L.piloP.pect R.cadu 0.5 P.digi 0.5 P.digi P.cruc F.brevi P.cruc F.brevi

− C.escu R.plum − C.escu R.plum Sabatia Sabatia

−1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 Basal Area NMDS1 NMDS1 E.sang C.sagiH.arteP.quinS.rotu

A.parv T.dive C.virg C.cana V.elli M.punc E.comp P.caro

0.5 R8 R9R.copaC.amerR6R.tome BA V.arbo H.fasc J.brac R.lute E.perf I.opac M.ceri C.erec L.japoG.purpE.sero S08_17D.tere L.tenu P.proc I.vomiT.radi R11 R.triv M.stri V.rotu L.capi S08_36 S.inte L.styra C.cren E.coro G.semp S.glau A.rubr P.verr S08_1 E.capi Richness

0.0 P.aqui S.scop T.onobH.stan M.acum E.rotuS.alti L.appe S.tinc D.scabR.grac D.acicS13_2 E.resi B.haliR.mari E.lept R.gran P.sang C.radiH.angJ.poly S08_3 E.leuco S13_17S13_9S.rugoL.medi D.virg S13_R10 S.grac S.dumoP.plicP.mariBidens H.punc S08_32 S13_26 A.obovC.pulc R.inex X.stri R.reco P.tenu L.piloP.pect R.cadu 0.5 P.digi P.cruc F.brevi

− C.escu R.plum Sabatia

−1.0 −0.5 0.0 0.5 1.0 Fig. 5.5. Diameter at breast heightNMDS1 (DBH), tree height (Height), and basal area (BA) fitted as smooth surfaces on the 2015 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray- Curtis dissimilarity.

1.5 1.0

X.stri.obs R.grac H.hype S.alti V.arboN.sylvM.striP.sero P.mariP.caroA.parv Na S13_2D.acicM.acumOxalisX.ambi R.caduH.anguS.rugo E.lept S T.onob J.poly R.reco P.pect S13_6 L.medi.tex S13_9 R.lute P.procpH RAD S13_26D.tere 0.5 S.dumo S13_17 E.resi L.pilo M.ceri E.leuc R.hirt S08_3 P.aquiC.radi S08_32D.virg L.styr R.copa E.rotu Q.nigr V.rotuS.tinc V.elli E.capi Cu T.radi S.scop Richness

0.0 S08_36G.sempS.glau S08_1 NMDS2 BA A.fili DBH S08_17 B.hali P.verru D.scabI.vomi H.stan LAI I.opac R.triv C.erec L.appe C.virg R6 L.pube 0.5 C.amer E.perf

− R9 E.hieraC.cana R10 G.purpR11 L.japoLactuca E.comp K R8 C.horrP M.punc C.cono Height D.scop

1.0 T.perf − 1.5 −

−2 −1 0 1 2

NMDS1

Fig. 5.6. Non-metric multidimensional scaling (NMDS) ordination of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray-Curtis dissimilarity for year 2016. Black dots indicate site centroids, while standard deviation 95% ellipses outline position of plantation types. Superimposed vectors indicate species richness (black), overstory variables (black), canopy variables (purple), and soil variables (orange). 96

DBH Tree Height 1.0 1.0

X.stri.obs X.stri.obs R.grac R.grac H.hype H.hype S.altiN.sylvM.stri S.altiN.sylvM.stri P.mariP.caroA.parvV.arboP.sero P.mariP.caroA.parvV.arboP.sero S13_2D.acicM.acumOxalisX.ambi S13_2D.acicM.acumOxalisX.ambi R.caduH.anguS.rugo E.lept R.caduH.anguS.rugo E.lept T.onob J.poly T.onob J.poly R.reco P.pect L.medi.tex R.reco P.pect L.medi.tex S13_9S13_6 P.proc S13_9S13_6 P.proc S13_26D.tere R.lute S13_26D.tere R.lute 0.5 S.dumo S13_17 0.5 S.dumo S13_17 E.resi L.pilo E.resi L.pilo M.ceri E.leuc R.hirt M.ceri E.leuc R.hirt S08_3 P.aquiC.radi S08_3 P.aquiC.radi S08_32D.virg S08_32D.virg L.styr L.styr R.copa E.rotu R.copa E.rotu Q.nigr S.tinc E.capi Q.nigr S.tinc E.capi V.rotu T.radiV.elliS.scop V.rotu T.radiV.elliS.scop G.semp Richness G.semp Richness 0.0 S08_36S.glau 0.0 S08_36S.glau S08_1 S08_1 NMDS2 A.fili NMDS2 A.fili DBH S08_17 B.hali S08_17 B.hali P.verruI.vomi P.verruI.vomi I.opacR.trivD.scab C.erec L.appeH.stan C.virg I.opacR.trivD.scab C.erec L.appeH.stan C.virg R6 L.pube R6 L.pube 0.5 C.amer R9 E.perf 0.5 C.amer R9 E.perf − E.hieraC.cana − E.hieraC.cana G.purpR11R10 G.purpR11R10 L.japoLactuca L.japoLactuca E.compC.horr E.compC.horr M.puncR8 M.puncR8 C.conoD.scop C.conoD.scop Height

1.0 T.perf 1.0 T.perf − −

−1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 −1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 Basal Area NMDS1 NMDS1 1.0

X.stri.obs R.grac H.hype S.altiN.sylvM.stri P.mariP.caroA.parvV.arboP.sero S13_2D.acicM.acumOxalisX.ambi R.caduH.anguS.rugo E.lept T.onob J.poly R.reco P.pect L.medi.tex S13_9S13_6 P.proc S13_26D.tere R.lute 0.5 S.dumo S13_17 E.resi L.pilo M.ceri E.leuc R.hirt S08_3 P.aquiC.radi S08_32D.virg L.styr R.copa E.rotu Q.nigr S.tinc E.capi V.rotu T.radiV.elliS.scop G.semp Richness 0.0 S08_36S.glau S08_1 NMDS2 A.fili BA S08_17 B.hali P.verruI.vomi I.opacR.trivD.scab C.erec L.appeH.stan C.virg R6 L.pube 0.5 C.amer R9 E.perf − E.hieraC.cana G.purpR11R10 L.japoLactuca E.compC.horr M.puncR8 C.conoD.scop

1.0 T.perf − −1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 Fig. 5.7. Diameter at breast heightNMDS1 (DBH), tree height (Height), and basal area (BA) fitted as smooth surfaces on the 2016 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray- Curtis dissimilarity.

LAI Total Daily Radiation Under The Canopy 1.0 1.0

X.stri.obs X.stri.obs R.grac R.grac H.hype H.hype S.altiN.sylvM.stri S.altiN.sylvM.stri P.mariP.caroA.parvV.arboP.sero P.mariP.caroA.parvV.arboP.sero S13_2D.acicM.acumOxalisX.ambi S13_2D.acicM.acumOxalisX.ambi R.caduH.anguS.rugo E.lept R.caduH.anguS.rugo E.lept T.onob J.poly T.onob J.poly R.reco P.pect L.medi.tex R.reco P.pect L.medi.tex S13_9S13_6 P.proc S13_9S13_6 P.proc RAD S13_26D.tere R.lute S13_26D.tere R.lute 0.5 S.dumo S13_17 0.5 S.dumo S13_17 E.resi L.pilo E.resi L.pilo M.ceri E.leuc R.hirt M.ceri E.leuc R.hirt S08_3 P.aquiC.radi S08_3 P.aquiC.radi S08_32D.virg S08_32D.virg L.styr L.styr R.copa E.rotu R.copa E.rotu Q.nigr S.tinc E.capi Q.nigr S.tinc E.capi V.rotu T.radiV.elliS.scop V.rotu T.radiV.elliS.scop G.semp Richness G.semp Richness 0.0 S08_36S.glau 0.0 S08_36S.glau S08_1 S08_1 NMDS2 A.fili NMDS2 A.fili S08_17 B.hali S08_17 B.hali P.verruI.vomi P.verruI.vomi LAI I.opacR.trivD.scab C.erec L.appeH.stan C.virg I.opacR.trivD.scab C.erec L.appeH.stan C.virg R6 L.pube R6 L.pube 0.5 C.amer R9 E.perf 0.5 C.amer R9 E.perf − E.hieraC.cana − E.hieraC.cana G.purpR11R10 G.purpR11R10 L.japoLactuca L.japoLactuca E.compC.horr E.compC.horr M.puncR8 M.puncR8 C.conoD.scop C.conoD.scop

1.0 T.perf 1.0 T.perf − −

−1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 −1.5 −1.0 −0.5 0.0 0.5 1.0 1.5

Fig. 5.8. Leaf area index (LAINMDS1) and total daily radiation under the canopy (RAD) fitted asNMDS1 smooth surfaces on the 2016 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray- Curtis dissimilarity.

Soil pH Soil P Soil K X.stri.obs X.stri.obs X.stri.obs R.grac R.grac R.grac H.hype H.hype H.hype S.alti S.alti S.alti N.sylvM.striP.sero N.sylvM.striP.sero N.sylvM.striP.sero P.mariP.caroA.parvV.arbo P.mariP.caroA.parvV.arbo P.mariP.caroA.parvV.arbo S13_2D.acicM.acumOxalisX.ambi S13_2D.acicM.acumOxalisX.ambi S13_2D.acicM.acumOxalisX.ambi R.caduH.anguS.rugo E.lept R.caduH.anguS.rugo E.lept R.caduH.anguS.rugo E.lept T.onob J.poly T.onob J.poly T.onob J.poly R.reco P.pect S13_6L.medi.tex R.reco P.pect S13_6L.medi.tex R.reco P.pect S13_6L.medi.tex S13_26S13_9 R.lute P.proc S13_26S13_9 R.lute P.proc S13_26S13_9 R.lute P.proc 0.5 D.tere pH 0.5 D.tere 0.5 D.tere S.dumo S13_17 S.dumo S13_17 S.dumo S13_17 E.resi L.pilo E.resi L.pilo E.resi L.pilo M.ceri E.leuc R.hirt M.ceri E.leuc R.hirt M.ceri E.leuc R.hirt S08_3 P.aquiC.radi S08_3 P.aquiC.radi S08_3 P.aquiC.radi S08_32D.virg S08_32D.virg S08_32D.virg L.styr L.styr L.styr R.copa E.rotu R.copa E.rotu R.copa E.rotu Q.nigr S.tinc E.capi Q.nigr S.tinc E.capi Q.nigr S.tinc E.capi V.rotu T.radiV.elliS.scop Richness V.rotu T.radiV.elliS.scop Richness V.rotu T.radiV.elliS.scop Richness 0.0 S08_36G.sempS.glau 0.0 S08_36G.sempS.glau 0.0 S08_36G.sempS.glau S08_1 S08_1 S08_1 NMDS2 NMDS2 A.fili A.fili A.fili S08_17 B.hali S08_17 B.hali S08_17 B.hali P.verruD.scabI.vomi P.verruD.scabI.vomi P.verruD.scabI.vomi I.opacR.triv C.erec L.appeH.stan C.virg I.opacR.triv C.erec L.appeH.stan C.virg I.opacR.triv C.erec L.appeH.stan C.virg R6 L.pube R6 L.pube R6 L.pube 0.5 C.amer E.perf 0.5 C.amer E.perf 0.5 C.amer E.perf − R9 − R9 − R9 E.hieraC.cana E.hieraC.cana E.hieraC.cana G.purpR11R10 G.purpR11R10 G.purpR11R10 L.japoLactuca L.japoLactuca P L.japoLactuca K R8 E.compC.horr R8 E.compC.horr R8 E.compC.horr M.punc M.punc M.punc C.conoD.scop C.conoD.scop C.conoD.scop

−1.0 −0.5 T.perf 0.0 0.5 1.0 −1.0 −0.5 T.perf 0.0 0.5 1.0 −1.0 −0.5 T.perf 0.0 0.5 1.0 Soil S Soil Cu Soil Na X.stri.obs X.stri.obs X.stri.obs NMDS1 R.grac NMDS1 R.grac NMDS1 R.grac H.hype H.hype H.hype S.alti S.alti S.alti N.sylvM.striP.sero N.sylvM.striP.sero N.sylvM.striP.sero P.mariP.caroA.parvV.arbo P.mariP.caroA.parvV.arbo P.mariP.caroA.parvV.arbo S13_2D.acicM.acumOxalisX.ambi S13_2D.acicM.acumOxalisX.ambi S13_2D.acicM.acumOxalisX.ambi R.caduH.anguS.rugo E.lept R.caduH.anguS.rugo E.lept Na R.caduH.anguS.rugo E.lept T.onob J.poly T.onob J.poly T.onob J.poly S R.reco P.pect S13_6L.medi.tex R.reco P.pect S13_6L.medi.tex R.reco P.pect S13_6L.medi.tex S13_26S13_9 R.lute P.proc S13_26S13_9 R.lute P.proc S13_26S13_9 R.lute P.proc

0.5 D.tere 0.5 D.tere 0.5 D.tere S.dumo S13_17 S.dumo S13_17 S.dumo S13_17 E.resi L.pilo E.resi L.pilo E.resi L.pilo M.ceri E.leuc R.hirt M.ceri E.leuc R.hirt M.ceri E.leuc R.hirt S08_3 P.aquiC.radi S08_3 P.aquiC.radi S08_3 P.aquiC.radi S08_32D.virg S08_32D.virg S08_32D.virg L.styr L.styr L.styr R.copa E.rotu R.copa E.rotu R.copa E.rotu Q.nigr S.tinc E.capi Q.nigr S.tinc E.capi Q.nigr S.tinc E.capi V.rotu T.radiV.elliS.scop Richness V.rotu T.radiV.elliS.scop Cu Richness V.rotu T.radiV.elliS.scop Richness 0.0 S08_36G.sempS.glau 0.0 S08_36G.sempS.glau 0.0 S08_36G.sempS.glau S08_1 S08_1 S08_1 NMDS2 NMDS2 A.fili A.fili A.fili S08_17 B.hali S08_17 B.hali S08_17 B.hali P.verruD.scabI.vomi P.verruD.scabI.vomi P.verruD.scabI.vomi I.opacR.triv C.erec L.appeH.stan C.virg I.opacR.triv C.erec L.appeH.stan C.virg I.opacR.triv C.erec L.appeH.stan C.virg R6 L.pube R6 L.pube R6 L.pube 0.5 C.amer E.perf 0.5 C.amer E.perf 0.5 C.amer E.perf

− R9 − R9 − R9 E.hieraC.cana E.hieraC.cana E.hieraC.cana G.purpR11R10 G.purpR11R10 G.purpR11R10 L.japoLactuca L.japoLactuca L.japoLactuca R8 E.compC.horr R8 E.compC.horr R8 E.compC.horr M.punc M.punc M.punc C.conoD.scop C.conoD.scop C.conoD.scop

−1.0 −0.5 T.perf 0.0 0.5 1.0 −1.0 −0.5 T.perf 0.0 0.5 1.0 −1.0 −0.5 T.perf 0.0 0.5 1.0

Fig. 5.9. Soil pH, phosphorousNMDS1 (P), potassium (K), sulfur (S),NMDS1 copper (Cu), and sodium (Na) fitted as smoothNMDS1 surfaces on the 2016 NMDS ordinations of understory species composition (indicated in gray) of plantation sites (same-height slash pine sites = S08 prefix and blue color; same-age slash pine sites = S13 prefix and green color; Camden white gum sites = R prefix and red color), using Bray- Curtis dissimilarity

Table 5.2 PERMANOVA results based on NMDS scores for years 2014-2016. P-values relative to post-hoc contrasts between plantation types were adjusted using Bonferroni method. Year Model Pseudo-F R2 P-value Adj. P-value Sign. level 2014 Overall 8.827 0.60 0.002 - *** E13 vs. S08 10.207 0.56 0.01 0.03 * E13 vs. S13 3.829 0.32 0.038 0.114 S13 vs. S08 14.743 0.64 0.011 0.033 *

2015 Overall 11.95 0.67 0.001 - *** E13 vs. S08 9.637 0.55 0.007 0.021 * E13 vs. S13 9.11 0.53 0.017 0.051 S13 vs. S08 18.295 0.70 0.01 0.03 *

2016 Overall 29.068 0.83 0.001 - *** E13 vs. S08 19.478 0.71 0.005 0.015 * E13 vs. S13 35.159 0.81 0.011 0.033 * S13 vs. S08 31.947 0.80 0.011 0.033 *

100

Table 5.3. Cosine of variables with NMDS axes, R2, P-value, and significance level for the vectors relative to the variables superimposed in the ordination. Cosine of the angle to NMDS axes represented the inclination of the vectors in the ordination space (not the correlation with NMDS axes). The value of R2 represent the correlation of the (external) variable and the ordination scores projected onto the arrow (not the axes), while the P-value and significance levels are relative the permutation test described in the methods. Year Variable Cos. with NMDS R2 P-value Sign. level Axis 1 Axis 2 2014 Species richness 1 0.0 0.66 0.002 **

2015 Species richness 1 0.0 0.61 0.003 ** DBH -0.92 0.92 0.80 0.001 *** Height -0.79 0.38 0.87 0.001 *** Basal area -0.91 0.61 0.75 0.001 ***

2016 Species richness 1 0.0 0.76 0.001 *** DBH -0.61 -0.17 0.89 0.001 *** Height -0.89 -0.79 0.92 0.001 *** Basal area -0.98 -0.17 0.69 0.001 *** LAI -0.97 -0.26 0.89 0.001 *** RAD 0.91 0.42 0.91 0.001 *** Soil P 0.22 -0.97 0.42 0.005 ** Soil K 0.33 -0.94 0.43 0.029 * Soil Cu 1.00 0.03 0.55 0.005 ** Soil Na -0.49 0.87 0.40 0.028 * Soil S -0.73 0.68 0.52 0.011 * Soil pH 0.89 0.46 0.80 0.001 ***

101

Table 5.4. Intercepts, slope coefficients, R2 values, P-values, and significancy relative to PERMANOVA linear models of environmental variables (dependent variables) and NMDS axes (independent variables) for years 2015 and 2016. Year Variable Intercept NMDS1 NMDS2 slope R2 P-value Sig. slope R2 P-value Sig. 2015 DBH 5.521 6.647 0.79 0.001 *** -2.649 0.09 0.016 * Height 4.895 4.471 0.62 0.001 *** -3.418 0.24 0.002 ** Basal area 5.820 10.914 0.68 0.001 *** -4.814 0.09 0.058

2016 DBH 7.269 -4.512 0.76 0.001 *** -2.889 0.16 0.002 ** Height 7.265 -5.728 0.91 0.001 *** 0.976 0.01 0.175 BA 8.035 -8.157 0.60 0.001 *** -5.417 0.13 0.036 * LAI 1.053 -1.016 0.81 0.001 *** -0.517 0.11 0.005 ** RAD 5.563 3.328 0.90 0.001 *** 1.122 0.05 0.005 ** pH 4.891 0.424 0.79 0.002 ** 0.011 0.03 0.206 P 3,985 -2.540 0.16 0.04 * 4.738 0.29 0.008 ** K 22.009 -2.998 0.13 0.115 6.526 0.31 0.035 * S 10.278 -0.373 0.00 0.75 -5.749 0.52 0.002 ** Na 17.719 1.548 0.05 0.374 -4.909 0.28 0.036 * Cu 0.525 0.102 0.43 0.001 *** 0.092 0.18 0.046 *

5.3. DISCUSSION

The position of plantation sites in the NMDS ordination space changed over time, to the point that Camden white gum sites were positioned closer to older slash pine sites and distant from same-age slash pine at the end of the study. These changes in position were associated with changes in the distances of the sites from certain understory species. In the first two years, eucalyptus sites had an intermediate position between the two slash pine sites. These results suggested a landscape matrix in which Camden white gum sites were embedded and from which understory species propagules were obtained, supporting the role of plantations as catalysts for regeneration, conservation and/or restoration of native species described by several authors

(Brockerhoff et al., 2008, 2001; Carnus et al., 2006; Cusack and Montagnini, 2004; Hartley,

2002). Parrotta (1995), detected the presence of natural and naturalized understory species in young exotic species plantations (Eucalyptus robusta, Leucaena leucocephala, and Casuarina

102

equisetifolia), and similar results were observed in different Eucalyptus spp. plantations understory, where the understory was similar to those of adjacent natural forests (Bone et al.,

1997; Geldenhuys, 1997; Michelsen et al., 1996; Yirdaw and Luukkanen, 2003).

Understory species composition changed relatively rapidly within Camden white gum plantations, suggesting that they could not support similar species as slash pine plantations for the same duration. In the first two years of the study, Camden white gum sites were closely placed to same-age slash pine with similar understory species composition, which indicated that

Camden white gum plantations could support species typical of open conditions found in same- age slash pine. During the third year, the changes in Camden white gum sites altered the understory vegetation composition toward shade-tolerant species found in older slash pine sites.

The quick variation in distance of Camden white gum sites during the third year raised concerns related to the fast-growing nature of eucalyptus and its influence on understory vegetation dynamics. Although variations in species composition related to stand age and characteristics normally occur in slash pine plantations, the same changes were occurring earlier in Camden white gum. Faster changes in habitat conditions are often detrimental to the conservation of specialists, which are particularly sensitive to land-use changes (Brockerhoff et al., 2003). For instance, in this study several beak-sedge species (Rhynchospora sp.) with full sun requirements/partial shade light requirements progressively moved away from eucalyptus sites while staying close to same-age slash pine ones. The relatively short rotation of eucalyptus plantations (often less than 9 years) could be also particularly discriminating against old-forest succession species (Alrababah et al., 2007; Buscardo et al., 2008; Richardson and van Wilgen,

1986).

Increased overstory and canopy complexity affected understory vegetation composition.

Several shade-tolerant or intermediate shade-tolerant species were placed on increasing gradients

103

of DBH, basal area, and LAI, such as red maple (Acer rubrum), American holly (Ilex opaca), water oak (Quercus nigra), sweetgum (Liquidamabar stryraciflya), sweetleaf (Symplocos tinctoria), poison-ivy (Toxicodendron radicans), wax myrtle (Morella cerifea), French mulberry

(Callicarpa americana). These species were located in the ordination space in the proximity of older slash pine sites, which was characterized by higher LAI values, greater DBH and basal area, and lower radiation intensity in the understory. On the other hand, species typical of open fields, clearings, prairies, woodlands and savannas, such as St. Andrew’s Cross (Hypericum hypericoides), comb-leaf mermaidweed (Proserpinaca pectinata), slender beaksedge

(Rhynchospora gracilenta), stiff yellow flax (Linum medium var. Texanum), pineland yelloweyed grass (Xyris stricta var. obscura), and rice button aster (Symphyotrichum dumosum) were located proximal to same-age slash pine (which had higher light availability). Camden white gum sites had intermediate (2015) or taller (2016) overstory as well as intermediate LAI and radiation, which allowed the growth of species that can thrive in both light and partial-shade environments. These levels of radiation transmittance allowed species with mixed-light requirements, such as eastern baccharis (Baccharis halimifolia), butterfly pea (Centrosema virginianum), common boneset (Eupatorium perfoliatum), spoonleaf purple everlasting

(Gamochaeta purpurea), and Canadian horseweed (Conyza canadensis).

Light transmittance, defined as quantity of light that reaches the forest floor through canopy, is largely determined by tree species composition, stand density, stand structure, and canopy patterns (Barbier et al., 2008). A study compared the transmittance values of two broadleaf (Eucalyptus globulus and Grevillea robusta) and two conifer (Cupressus lusitanica and Juniperus procera) plantations, showing that conifers had relatively denser crowns and shorter, clearer boles resulting in higher LAI values (Yirdaw and Luukkanen, 2004). On the other hand, broadleaf plantations have relatively open crowns, lower LAI values, and higher

104

transmittance percentages. Findings of this study are in line with results of those studies, with

Camden white gum having an overstory that was taller but with less-dense foliage than older (but same-height at the beginning of the study) slash pine.

Differences in canopy light transmittance was also reflected by the herbaceous ground cover. A study carried out in different plantation types showed that herbaceous ground cover was greater under broadleaf plantations (Grevillea robusta) than conifer plantations (Pinus patula,

Juniperus procera, and Cupressus lusitanica) (Yirdaw, 2001). Herbaceous vegetation levels were greater in broadleaf plantations, and in this study Camden white gum sites were similarly characterized by the proximity of more herbs, forbs, and grasses than older slash pine. According to Pohjonen (1989), tall eucalypt plantations typically have a dense undergrowth because their relatively sparse and usually pendulous leaves allows light to pass through them.

Species that thrive in acidic and slightly acidic soils were located in the proximity of older slash pine centroids, as shown by the pH gradient. Examples of these species are Carolina jessamine (Gelsemium sempervirens), Elliott’s huckleberry (Vaccinium elliottii), American holly

(Ilex opaca), sweetgum (Liquidambar styraciflua), wax myrtle (Morella cerifera), and southern dewberry (Rubus trivilias). It is well-known that acidic soils support less species than neutral and slightly alkaline ones (Ellenberg, 1988), and the importance of soil reactions in determining understory species richness and diversity was already discussed in Chapter 3. Conifers generally produce more acidic soils than broadleaves on similar sites (Fisher and Binkley, 2012), which explains the lower pH levels observed in older slash pine sites. However, eucalyptus plantations can increase soil acidity over time. Eucalyptus saligna plantations decreased soil pH from 5.9 to

5.0 in eight years compared to N-fixing Albizia spp. plantations (Rhoades and Binkley, 1996).

Moreover, a meta-analysis indicated that afforestation with pine and eucalyptus species led to

105

more acidic and nutrient-deﬁcient mineral soils (Berthrong et al., 2009), supporting the idea of potential increased acidity and nutrient-depleted soil in Camden white gum sites over time.

The positions of slash pine sites along gradients of increased soil S and Na, and Camden white gum sites on increased K, and same-age slash pine on increased Cu could be due to nutrient cycling trends of these plantations. The S gradient could be the effect of the deposition of sulfate–S due to enhanced mineralization processes, which was studied for conifers (Davis,

1995, 1994; Ross et al., 2009). The gradient of Na and K seemed to be an effect of the variation in soil pH, with their availability increasing with increasing soil alkalinity. The operational fertilization received by eucalyptus sites is the likely the reason of P gradient.

It is interesting how copper gradients had the same orientation of species richness gradient, and it might be connected to plant micronutrient requirements. The sites of this study spanned across locations with a majority of soils of the heavily leached, acidic, low-fertility

Ultisols order (USDA, 2002). Copper deficiencies can occur in soils developed from sands, sandstones, and acid igneous rocks (Jarvis, 1981), which constitute common parent materials of the study sites. Van Lear and Smith (1972) reported copper and micronutrients deficiencies and the necessity of supplementary fertilization prior to slash pine establishment in several areas of the Western Gulf flatwoods in which this study was conducted. It is possible that on the soils in this study Cu was limiting to species richness irrespective of plantation type.

5.4. CONCLUSION

Understory species communities of eucalyptus plantations changed over the years observed, and they were influenced by changes in habitats conditions driven by overstory, canopy and soil characteristics. In particular, canopy architecture and soil conditions appeared to be the major drivers of understory community assemblages. At the beginning of the study, and

106

over the second year, the understory species supported by Camden white gum sites were similar to those of same-age slash pine. Those species included a variety of forbs, herbs, and shrubs adapted to thrive in full light or partial shade light condition whereas vegetation of older slash pine sites was characterized by shrubs, tree saplings, and vines shade or intermediate shade tolerant. Changes in Camden white gum tree dimensions (diameter at the breast height, basal area, and total tree height) and the canopy architecture (leaf area index) triggered by the fast growth of the species altered the light availability for the understory. By the third year (2016),

Camden white gum understory composition was closer to that of older slash pine. However, particular overstory and canopy conditions (taller trees with intermediate LAI among plantation types) allowed the presence of forbs and herbs typical of open environments in Camden white gum plantations.

Soil characteristics, in particular pH, determined the understory communities in the plantations, influencing the edaphic characteristics and nutrient availability. Greater acidity of soils in older slash pine sites corresponded to a more acidophilus understory vegetation, while

Camden white gum sites were placed along gradients of increasing acidity, intermediate between the two slash pine sites. A concern about soil acidification in eucalyptus plantations and its effect on understory is raised because a decline in soil pH could contribute to nutrient deficiencies or toxic conditions.

Routine management operations, such as eucalyptus fertilization and pre-planting burning in slash pine site, could have affected soil reactions, nutrient availability, and tree growth; these variables influenced, in turn, understory species composition. Moreover, this study was conducted in only the first half of a Camden white gum rotation and the initial years of a slash pine rotation, and it lacks information about mid-, late-, and in-between rotations. The different management regimes adopted in slash pine (i.e. thinning, clear-cut, regeneration by re-planting)

107

and eucalyptus (herbicide applications and regeneration via coppicing) could have important long-term effects on plantation characteristics and understory communities that were beyond the scope of this study.

In conclusion, land-use change from slash pine to Camden white gum seemed to accelerate changes in overstory structure and soil characteristics, which in turn influenced the understory microclimatic conditions and consequently species assemblages. The adoption of

Eucalyptus benthamii plantations on a large scale will likely result in a substantial impoverishment in terms of understory species in the area, which has already experienced a decline of floristic richer native ecosystems on private lands, such as longleaf pine savannas and woodlands.

5.5. REFERENCES

Alrababah, M.A., Alhamad, M.A., Suwaileh, A., Al‐Gharaibeh, M., 2007. Biodiversity of semi‐ arid Mediterranean grasslands: Impact of grazing and afforestation. Appl. Veg. Sci. 10, 257–264. https://doi.org/10.1111/j.1654-109X.2007.tb00524.x Anderson, M.J., 2017. Permutational Multivariate Analysis of Variance (PERMANOVA). Wiley StatsRef Stat. Ref. Online, Major Reference Works. https://doi.org/doi:10.1002/9781118445112.stat07841 Aubin, I., Messier, C., Bouchard, A., 2008. Can plantations develop understory biological and physical attributes of naturally regenerated forests? Biol. Conserv. 141, 2461–2476. https://doi.org/https://doi.org/10.1016/j.biocon.2008.07.007 Barbier, S., Gosselin, F., Balandier, P., 2008. Influence of tree species on understory vegetation diversity and mechanisms involved - A critical review for temperate and boreal forests. For. Ecol. Manage. 254, 1–15. https://doi.org/10.1016/j.foreco.2007.09.038 Barkman, J.J., 1992. Canopies and microclimate of tree species mixtures, in: Cannell, M.G.R., Malcolm, D.C., Robertson, P.A. (Eds.), Canopies and Microclimate of Tree Species Mixtures. Blackwell Scientific Publications, Oxford, UK, pp. 181–188. Berthrong, S.T., Jobbágy, E.G., Jackson, R.B., 2009. A global meta-analysis of soil exchangeable cations, pH, carbon, and nitrogen with afforestation. Ecol. Appl. 19, 2228– 2241. https://doi.org/10.1890/08-1730.1

108

Blazier, M.A., Johnson, J., Taylor, E.L., Osbon, B., 2012. Herbicide site preparation and release options for eucalyptus plantation establishment in the western gulf. Butnor, John R., ed. 2012. Proc. 16th Bienn. South. Silvic. Res. Conf. e-Gen. Tech. Rep. SRS-156. Asheville, NC U.S. Dep. Agric. For. Serv. South. Res. Station. 19-23. Bone, R., Lawrence, M., Magombo, Z., 1997. The effect of a Eucalyptus camaldulensis (Dehn) plantation on native woodland recovery on Ulumba Mountain, southern Malawi. For. Ecol. Manage. 99, 83–99. https://doi.org/10.1016/S0378-1127(97)00196-5 Bray, J.R., Curtis, J.T., 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol. Monogr. 27, 325–349. Brockerhoff, E.G., Ecroyd, C.E., Langer, E.R., 2001. Biodiversity in New Zealand plantation forests: policy trends, incentives, and the state of our knowledge. New Zeal. J. For. 46, 31– 37. Brockerhoff, E.G., Ecroyd, C.E., Leckie, A.C., Kimberley, M.O., 2003. Diversity and succession of adventive and indigenous vascular understorey plants in Pinus radiata plantation forests in New Zealand. For. Ecol. Manage. 185, 307–326. https://doi.org/10.1016/S0378- 1127(03)00227-5 Brockerhoff, E.G., Jactel, H., Parrotta, J.A., Quine, C.P., Sayer, J., 2008. Plantation forests and biodiversity: oxymoron or opportunity? Biodivers. Conserv. 17, 925–951. https://doi.org/10.1007/s10531-008-9380-x Buscardo, E., Smith, G.F., Kelly, D.L., Freitas, H., Iremonger, S., Mitchell, F.J.G., O’Donoghue, S., McKee, A.-M., 2008. The early effects of afforestation on biodiversity of grasslands in Ireland. Biodivers. Conserv. 17, 1057–1072. https://doi.org/10.1007/s10531-007-9275-2 Calviño-Cancela, M., Rubido-Bará, M., van Etten, E.J.B., 2012. Do eucalypt plantations provide habitat for native forest biodiversity? For. Ecol. Manage. 270, 153–162. https://doi.org/https://doi.org/10.1016/j.foreco.2012.01.019 Carnus, J.M., Parrotta, J., Brockerhoff, E., Arbez, M., Jactel, H., Kremer, A., Lamb, D., O’Hara, K., Walters, B., 2006. Planted forests and biodiversity. J. For. 104, 65–77. Clarke, K.R., 1993. Non-parametric multivariate analyses of changes in community structure. Aust. J. Ecol. 18, 117–143. https://doi.org/10.1111/j.1442-9993.1993.tb00438.x Cusack, D., Montagnini, F., 2004. The role of native species plantations in recovery of understory woody diversity in degraded pasturelands of Costa Rica. For. Ecol. Manage. 188, 1–15. https://doi.org/10.1016/S0378-1127(03)00302-5 Davis, M.R., 1995. Influence of radiata pine seedlings on chemical properties of some New Zealand montane grassland soils. Plant Soil 176, 255–262. https://doi.org/10.1007/BF00011789

109

Davis, M.R., 1994. Topsoil properties under tussock grassland and adjoining pine forest in Otago, New Zealand. New Zeal. J. Agric. Res. 37, 465–469. https://doi.org/10.1080/00288233.1994.9513085 Ellenberg, H., 1988. Vegetation Ecology of Central Europe, th. ed. Wiley-Blackwell, Cambridge, UK. https://doi.org/10.1002/fedr.19901010711 Fisher, R.A., 1935. The design of experiments., The design of experiments. Oliver & Boyd, Oxford, England. Fisher, R.F., Binkley, D., 2012. Ecology and Management of Forest Soils. John Wiley & Sons, Ltd, Chichester, UK. https://doi.org/10.1002/9781118422342 Geldenhuys, C.J., 1997. Native forest regeneration in pine and eucalypt plantations in Northern Province, South Africa. For. Ecol. Manage. 99, 101–115. Haeussler, S., Kaffanke, T., Boateng, J.O., McClarnon, J., Bedford, L., 2017. Site preparation severity influences lodgepole pine plant community composition, diversity, and succession over 25 years. Can. J. For. Res. 47, 1659–1671. Hart, P.W., Johnson, J., Paim, R., 2016. Status update on development of a eucalyptus plantation program in the southeastern United States and higher elevations of southern Brazil. Tappi J. 15, 148–155. Hartley, M.J., 2002. Rationale and methods for conserving biodiversity in plantation forests. For. Ecol. Manage. 155, 81–95. https://doi.org/10.1016/S0378-1127(01)00549-7 Horn, H.S., 1971. The adaptive geometry of trees. Princeton University Press, Princeton, NJ. Humphrey, J.W., 2005. Benefits to biodiversity from developing old-growth conditions in British upland spruce plantations: a review and recommendations. For. An Int. J. For. Res. 78, 33– 53. Jarvis, S.C., 1981. Copper concentrations in plants and their relationship to soil properties, in: Copper in Soils and Plants. JianPing, W., HouBao, F., WenFei, L., GuoMin, H., JianFu, T., RuiJin, Z., Jing, H., ZhanFeng, L., 2015. Should exotic Eucalyptus be planted in subtropical China: insights from understory plant diversity in two contrasting Eucalyptus chronosequences. Environ. Manage. 56, 1244–1251. Kempthorne, O., 1966. Some aspects of experimental inference. J. Am. Stat. Assoc. 61, 11–34. Kenkel, N.C., Orlóci, L., 1986. Applying Metric and Nonmetric Multidimensional Scaling to Ecological Studies: Some New Results. Ecology 67, 919–928. https://doi.org/10.2307/1939814 Legendre, P., Legendre, L., 2012a. Chapter 7 - Ecological resemblance, in: Legendre, P.,

110

Legendre, L.B.T.-D. in E.M. (Eds.), Numerical Ecology. Elsevier, pp. 265–335. https://doi.org/https://doi.org/10.1016/B978-0-444-53868-0.50007-1 Legendre, P., Legendre, L., 2012b. Chapter 9 - Ordination in reduced space, in: Legendre, P., Legendre, L.B.T.-D. in E.M. (Eds.), Numerical Ecology. Elsevier, pp. 425–520. https://doi.org/https://doi.org/10.1016/B978-0-444-53868-0.50009-5 Loumeto, J.J., Huttel, C., 1997. Understory vegetation in fast-growing tree plantations on savanna soils in Congo. For. Ecol. Manage. 99, 65–81. https://doi.org/10.1016/s0378- 1127(97)00195-3 McCune, B., Grace, J.B., Urban, D.L., 2002. Analysis of ecological communities. MjM software design Gleneden Beach, OR. Michelsen, A., Lisanework, N., Friis, I., Holst, N., 1996. Comparisons of understorey vegetation and soil fertility in plantations and adjacent natural forests in the Ethiopian highlands. J. Appl. Ecol. 33, 627–642. https://doi.org/10.2307/2404991 Minchin, P.R., 1987a. Simulation of multidimensional community patterns: towards a comprehensive model. Vegetatio 71, 145–156. https://doi.org/10.1007/BF00039167 Minchin, P.R., 1987b. An evaluation of the relative robustness of techniques for ecological ordination BT - Theory and models in vegetation science: Proceedings of Symposium, Uppsala, July 8–13, 1985, in: Prentice, I.C., van der Maarel, E. (Eds.), . Springer Netherlands, Dordrecht, pp. 89–107. https://doi.org/10.1007/978-94-009-4061-1_9 Oksanen, J., Blanchet, F.G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P.R., O’Hara, R.B., Simpson, G.L., Solymos, P., Stevens, M.H.H., Wagner, H., 2018. Vegan: Community Ecology Package. Parrotta, J.A., 1995. Influence of overstory composition on understory colonization by native species in plantations on a degraded tropical site. J. Veg. Sci. 6, 627–636. https://doi.org/10.2307/3236433 Pohjonen, V., 1989. Establishment of fuelwood plantations in Ethiopia. Silva Carelica 14, 1–388. R Core Team, 2018. R: A language and environment for statistical computing. Rhoades, C., Binkley, D., 1996. Factors influencing decline in soil pH in Hawaiian Eucalyptus and Albizia plantations. For. Ecol. Manage. 80, 47–56. https://doi.org/https://doi.org/10.1016/0378-1127(95)03646-6 Richardson, D.M., van Wilgen, B.W., 1986. Effects of thirty-five years of afforestation with Pinus radiata on the composition of mesic mountain fynbos near Stellenbosch. South African J. Bot. 52, 309–315. https://doi.org/https://doi.org/10.1016/S0254-6299(16)31526-5 Ross, C.W., Watt, M.S., Parfitt, R.L., Simcock, R., Dando, J., Coker, G., Clinton, P.W., Davis, M.R., 2009. Soil quality relationships with tree growth in exotic forests in New Zealand.

111

For. Ecol. Manage. 258, 2326–2334. https://doi.org/https://doi.org/10.1016/j.foreco.2009.05.026 Temesgen, D., Gonzálo, J., Turrión, M.B., 2016. Effects of short‐rotation Eucalyptus plantations on soil quality attributes in highly acidic soils of the central highlands of Ethiopia. Soil Use Manag. 32, 210–219. https://doi.org/10.1111/sum.12257 ter Braak, C.J.F., 1995. Ordination, in: Data Analysis in Community and Landscape Ecology. Cambridge University Press, pp. 91–274. Tyynelä, T.M., 2001. Species diversity in Eucalyptus camaldulensis woodlots and miombo woodland in Northeastern Zimbabwe. New For. 22, 239–257. https://doi.org/10.1023/A:1015616010976 USDA, 2002. Soil survey of Beauregard Parish, Louisiana. Van Lear, D.H., Smith, W.H., 1972. Relationships between macro- and micronutrient nutrition of slash pine on three coastal plain soils. Plant Soil 36, 331–347. https://doi.org/10.1007/BF01373488 Wherry, E.T., 1927. Divergent Soil Reaction Preferences of Related Plants. Ecology 8, 197–206. https://doi.org/10.2307/1928959 Yirdaw, E., 2001. Diversity of naturally-regenerated native woody species in forest plantations in the Ethiopian highlands. New For. 22, 159–177. https://doi.org/10.1023/A:1015629327039 Yirdaw, E., Luukkanen, O., 2004. Photosynthetically active radiation transmittance of forest plantation canopies in the Ethiopian highlands. For. Ecol. Manage. 188, 17–24. https://doi.org/https://doi.org/10.1016/j.foreco.2003.07.024 Yirdaw, E., Luukkanen, O., 2003. Indigenous woody species diversity in Eucalyptus globulus Labill. ssp. globulus plantations in the Ethiopian highlands. Biodivers. Conserv. 12, 567– 582. https://doi.org/10.1023/A:1022483700992

112

CHAPTER 6. SUMMARY AND CONCLUSION

The main objective of this dissertation was to assess the understory vegetation composition and functionality triggered by the land-use change from slash pine (Pinus elliottii) plantations to Camden white gum (Eucalytus benthamii) plantations over a period of three years within the early rotation of both species. Specifically, the present dissertation has analyzed understory species in terms of (1) number of species, Shannon index (evenness), and accumulation of species over sampling areas, (2) potential factors influencing its composition and patterns, such as overstory conditions, and soil characteristics, (3) functional morphological and stoichiometric traits and functional diversity indices, (4) impact of functional traits and diversity indices on understory aboveground biomass and vertical structure, (5) similarities, dynamics, and responses to overstory, canopy, and soil gradients. Three different plantation types were investigated: (1) Camden white gum established in 2013, which represents the novelty treatment (2) slash pine established in 2013, which represents the business-as-usual scenario, (3) slash pine plantation established in 2008, selected to match the increase in eucalyptus tree size due to eucalyptus’s fast-growth rate.

A decline in number of species was observed in the three plantations. Initially, Camden white gum was similar to same-age slash pine in understory species, with higher number of species than older slash pine. During the second and the third years, eucalyptus plantation occupied an intermediate position among the two slash pine plantations in understory species richness. Among all the variables investigated, leaf area index, soil pH and K, and tree height were the most important factors influencing understory species numbers and Shannon index, pointing out the major influences of overstory and soil resources in understory species richness.

Certain functional traits, such as plant height, leaf area, and leaf N were different between the plantation types, but no differences were revealed in functional diversity indices (functional

113

richness, evenness, and divergence). The similarity in functional richness and the lack of linear relationship between functional and taxonomic richness seemed to suggest functional redundancy, which causes a resilience in the land-use change from slash pine to Camden white gum plantation, allowing understory communities to return to a previous state of stability in terms of functional traits, without evident changes in ecosystem properties (understory aboveground biomass and vertical structure). Understory aboveground biomass was predicted from overstory characteristics rather than its functional traits and functional diversity indices.

These findings identify the overstory as the true determinant of the ecosystem properties of understory biomass and vertical structure because of its greater contribution in terms of biomass.

However, understory functional traits contributed to determining understory vertical structure.

Understory species communities of eucalyptus and slash pine plantations were influenced by changes in habitat conditions driven by gradients of overstory and soil characteristics. During

2014 and 2015, similar understory communities were observed in Camden white gum and same- age slash pine sites in terms of forbs, herbs, and shrubs adapted to thrive in full-light or partial- shade light conditions. In contrast, understory of older slash pine sites was characterized by shade- or intermediate-shade tolerant species. Changes in Camden white gum overstory size and its canopy architecture reduced light availability for the understory, and during 2016 its understory composition was closer to older slash pine. However, configuration of foliage and tree height of Camden white gum plantations allowed the presence of forbs and herbs typical of open environments. Among soil characteristics, pH gradients determined the understory communities in the plantations. Acidophilus understory vegetation was found in acidic soils of older slash pine sites. Soil pH appeared to influence also nutrient availability for the understory.

However, this study concerns only the initial stages of the rotation and lacks important information related to the different stages of rotations and management operations that will occur

114

later in the rotations, particularly thinning (only in slash pine) and clearcutting (done at shorter intervals in Camden white gum). Future research can build upon the work of this project by exploring understory species diversity and the overstory, edaphic, and climate conditions that govern diversity over a longer period that includes harvesting patterns of the plantation types.

Comparisons encompassing eucalyptus plantations relative to native ecosystem conditions as well as business-as-usual land uses will also provide a more comprehensive assessment of land conversion to eucalyptus plantations on non-crop vegetation diversity, structure, and productivity.

From a management standpoint, the result of this dissertation discourage the land-use change from slash pine to non-native and fast-growing Camden white gum. Louisiana already experienced declines in longleaf pine habitats in the region in which this study occurred. These floristically-rich, herb-dominated wetlands were converted to loblolly and slash pine plantations, which rely on intensive management activities detrimental for natural understory communities.

The adoption of fast-growing eucalyptus on these sites formerly occupied by longleaf pine will probably accelerate the deterioration of natural habitats and eventually reduce open-condition species in favor of shade-tolerant species, overturning the longleaf habitat conservation efforts already put in place by governmental agencies and conservation groups.

115

APPENDIX A. UNDERSTORY SPECIES LIST

A1. Understory species found in E. benthamii plantation (E13)

Abundance Family Scientific name Form Duration 2014 2015 2016 Anacardiaceae Rhus copallinum 2.00% 5.00% 0.50% Shrub Perennial Anacardiaceae Toxicodendron radicans 8.00% 9.00% 3.50% Vine Perennial Anacardiaceae Toxicodendron diversilobum 0.00% 2.00% 0.00% Shrub Perennial Annonaceae Asimina parviflora 1.00% 1.50% 0.00% Tree Perennial Apiaceae Centella crenata 2.50% 1.00% 0.00% Forb/herb Perennial Apiaceae Centella erecta 1.50% 5.00% 4.50% Forb/herb Perennial Aquifoliaceae Ilex opaca 0.00% 0.50% 1.00% Shrub Perennial Aquifoliaceae Ilex vomitoria 6.50% 10.00% 11.00% Shrub Perennial Asteraceae Baccharis halimifolia 4.50% 3.00% 6.50% Shrub Perennial Asteraceae Chrysopsis mariana 0.50% 0.00% 0.00% Forb/herb Perennial Asteraceae Cirsium horridulum 2.50% 0.00% 4.00% Forb/herb Annual/biennial Asteraceae Conoclinium coelestinum 0.00% 0.00% 0.50% Forb/herb Perennial Asteraceae Conyza canadensis 62.50% 8.50% 16.50% Forb/herb Annual/biennial Asteraceae Diaperia candida 0.00% 0.00% 0.50% Forb/herb Annual Asteraceae Echinacea sanguinea 0.00% 0.50% 0.50% Forb/herb Perennial Asteraceae Erechtites hieraciifolius 0.00% 0.00% 0.50% Forb/herb Annual Asteraceae Eupatorium capillifolium 39.00% 44.50% 7.50% Forb/herb Perennial Asteraceae Eupatorium leucolepsis 9.00% 7.00% 2.00% Forb/herb Perennial Asteraceae Eupatorium compositifolium 4.50% 15.00% 49.00% Forb/herb Perennial Asteraceae Eupatorium perfoliatum 4.50% 8.00% 4.50% Forb/herb Perennial Asteraceae Eupatorium resinosum 0.00% 1.00% 0.00% Forb/herb Perennial Asteraceae Eupatorium rotundifolium 12.50% 25.00% 11.00% Forb/herb Perennial Asteraceae Eupatorium serotinum 4.50% 2.00% 1.00% Forb/herb Perennial Asteraceae Euthamia leptocephala 1.50% 2.00% 0.00% Forb/herb Perennial Asteraceae Gaillardia aestivalis 0.00% 0.50% 0.00% Forb/herb Perennial Asteraceae Gamochaeta purpurea 44.00% 38.00% 39.50% Forb/herb Annual/biennial Asteraceae Helianthus angustifolius 94.00% 23.00% 2.50% Forb/herb Perennial Asteraceae Hymenopappus artemisiifolius 0.50% 1.50% 0.00% Forb/herb Biennial Asteraceae Lactuca sp. 0.00% 0.00% 2.50% Forb/herb Annual/biennial Asteraceae Pityopsis graminifolia 8.50% 0.00% 0.00% Forb/herb Perennial Asteraceae Pyrrhopappus carolinianus 1.50% 5.00% 0.00% Forb/herb Annual/biennial Asteraceae Rudbeckia grandiflora 0.00% 1.00% 2.50% Forb/herb Perennial Asteraceae Rudbeckia hirta 0.00% 0.00% 0.50% Forb/herb Perennial Asteraceae Solidago altissima 28.50% 2.00% 0.00% Forb/herb Perennial Asteraceae Solidago rugosa 5.50% 1.00% 0.00% Forb/herb Perennial Asteraceae Symphyotrichum dumosum 2.00% 2.50% 5.50% Forb/herb Perennial Bignogniaceae Bignonia capreolata 0.00% 0.00% 0.50% Vine Perennial Bignogniaceae Campsis radicans 1.50% 2.50% 0.50% Vine Perennial Campanulaceae Lobelia appendiculata 0.00% 1.00% 1.00% Forb/herb Annual/biennial Campanulaceae Lobelia puberula 0.50% 0.00% 3.00% Forb/herb Perennial Campanulaceae Triodanis perfoliata 0.00% 0.00% 0.50% Forb/herb Annual Cistaceae Lechea tenuifolia 0.00% 0.50% 0.00% Forb/herb Perennial Cyperaceae Cyperus echinatus 1.00% 0.00% 0.00% Graminoid Perennial

116

Cyperaceae Cyperus esculentus 0.50% 0.00% 0.00% Graminoid Perennial Cyperaceae Rhynchospora caduca 0.00% 0.50% 0.00% Graminoid Perennial Cyperaceae Rhynchospora gracilenta 0.00% 2.00% 0.00% Graminoid Perennial Cyperaceae Rhynchospora recognita 1.00% 0.00% 0.00% Graminoid Perennial Dennstaedtiaceae Pteridium aquilinum 8.00% 7.00% 6.00% Forb/herb Perennial Ericaceae Vaccinium arboreum 1.00% 3.50% 0.00% Shrub Perennial Ericaceae Vaccinium elliottii 1.50% 0.00% 0.00% Shrub Perennial Euphorbiaceae Euphorbia corollata 0.00% 4.50% 0.00% Forb/herb Perennial Fabaceae Centrosema virginianum 10.50% 13.00% 1.00% Forb/herb Perennial Fabaceae Crotalaria sagittalis 0.00% 1.00% 0.00% Forb/herb Annual Fabaceae Lespedeza capitata 3.00% 0.00% 0.00% Forb/herb Perennial Fabaceae Mimosa strigillosa 1.00% 2.00% 0.00% Forb/herb Perennial Fabaceae Rhynchosia tomentosa 0.00% 1.50% 0.00% Forb/herb Perennial Fabaceae Sesbania herbacea 4.00% 0.00% 0.00% Forb/herb Annual Fabaceae Strophostyles umbellata 0.50% 0.00% 0.00% Forb/herb Perennial Fabaceae Tephrosia onobrychoides 5.00% 8.50% 0.00% Forb/herb Perennial Fagaceae Quercus marilandica 0.00% 1.00% 0.00% Tree Perennial Gelsemiaceae Gelsemium sempervirens 5.00% 4.50% 7.00% Vine Perennial Gentianiaceae Centaurium pulchellum 2.00% 0.00% 0.00% Forb/herb Annual Hamamelidaceae Liquidambar styraciflua 0.50% 1.00% 0.00% Tree Perennial Hypericaceae Hypericum fasciculatum 0.00% 1.50% 1.00% Shrub Perennial Hypericaceae Hypericum galioides 1.50% 0.00% 0.00% Shrub Perennial Hypericaceae Hypericum hypericoides 5.00% 2.00% 0.00% Shrub Perennial Hypericaceae Hypericum stans 5.50% 4.50% 2.50% Shrub Perennial Iridaceae Sisyrinchium angustifolium 0.00% 0.00% 0.50% Forb/herb Perennial Juncaceae Juncus brachicarpus 2.00% 1.00% 0.00% Graminoid Perennial Juncaceae Juncus polycephalus 0.00% 3.00% 0.00% Graminoid Perennial Lamiaceae Monarda punctata 0.00% 6.50% 2.50% Forb/herb Perennial Lamiaceae Physostegia digitalis 0.00% 1.00% 0.00% Forb/herb Perennial Lamiaceae Pycnanthemum tenuifolium 0.00% 1.00% 0.00% Forb/herb Perennial Lamiaceae Scutellaria integrifolia 6.50% 5.00% 0.00% Forb/herb Perennial Linaceae Linum medium var. texanum 1.00% 0.00% 0.00% Forb/herb Annual Lygodiaceae Lygodium japonicum 8.00% 10.50% 10.00% Vine Perennial Melastomataceae Rhexia lutea 3.50% 3.50% 0.50% Forb/herb Perennial Melastomataceae Rhexia mariana 1.50% 15.50% 0.00% Forb/herb Perennial Myricaceae Morella cerifera 0.50% 4.50% 0.00% Shrub Perennial Nyssaceae Nyssa sylvatica 1.00% 1.50% 0.00% Tree Perennial Onagraceae Ludwigia pilosa 1.00% 0.00% 0.00% Forb/herb Perennial Orobranchaceae Agalinis filicaulis 0.00% 0.00% 1.00% Forb/herb Annual Oxalidaceae Desmondia rotundifolia 0.00% 2.00% 0.00% Oxalidaceae Oxalis sp. 2.00% 0.00% 1.00% Forb/herb Perennial Pinaceae Pinus elliottii 0.50% 0.00% 0.00% Tree Perennial Poaceae Schizachyrium scoparium 79.00% 82.50% 83.00% Graminoid Perennial Poaceae Dichanthelium aciculare 20.50% 52.50% 0.50% Graminoid Perennial Poaceae Dichanthelium scabrisculum 64.50% 58.00% 41.50% Graminoid Perennial Poaceae Dichantelium scoparium 0.00% 0.00% 2.00% Graminoid Perennial Poaceae Panicum verrucosum 3.50% 5.50% 17.00% Graminoid Perennial Poaceae Paspalum plicatum 2.00% 3.00% 0.00% Graminoid Perennial Poaceae Setaria parviflora 0.50% 0.00% 0.00% Graminoid Perennial

117

Polygalaceae Polygala mariana 0.00% 1.00% 0.00% Forb/herb Annual Rosaceae Rubus trivialis 41.50% 47.50% 63.50% Vine Perennial Rubiaceae Diodia virginiana 7.50% 1.00% 0.50% Forb/herb Annual Rubiaceae Diodia teres 0.50% 5.00% 0.50% Forb/herb Annual Scrophulariaceae Mecardonia acuminata 1.50% 6.00% 0.00% Forb/herb Perennial Smilaceae Smilax glauca 54.00% 36.50% 29.50% Vine Perennial Symplocaceae Symplocos tinctoria 2.00% 2.50% 0.50% Shrub Perennial Tetrachondraceae Polypremum procumbens 9.50% 0.00% 0.00% Forb/herb Annual Verbenaceae Callicarpa americana 6.00% 8.50% 3.00% Shrub Perennial Verbenaceae Phyla sp. 0.50% 0.00% 0.00% Forb/herb Perennial Violaceae Viola primulifolia 0.00% 0.00% 0.50% Forb/herb Perennial Vitaceae Vitis rotundifolia 1.00% 0.50% 0.00% Vine Perennial Vitaceae Partenocissus quinquefolia 0.00% 0.50% 0.00% Vine Perennial Xiridaceae Xyris stricta var. obscura 0.00% 1.00% 0.00% Forb/herb Perennial

118

A2. Understory species found in P. elliottii plantation established in 2013 (S13)

Abundance Family Scientific name Form Duration 2014 2015 2016 Acanthaceae Ruellia humilis 0.00% 0.50% 0.00% Forb/herb Perennial Anacardiaceae Rhus copallinum 1.00% 1.50% 1.00% Shrub Perennial

Anacardiaceae Toxicodendron radicans 1.50% 1.00% 2.00% Vine Perennial

Annonaceae Asimina parviflora 0.00% 0.00% 1.00% Tree Perennial Apiaceae Centella crenata 6.00% 0.00% 0.00% Forb/herb Perennial

Apiaceae Centella erecta 0.00% 2.50% 2.00% Forb/herb Perennial

Aquifoliaceae Ilex opaca 0.00% 0.50% 0.00% Shrub Perennial Aquifoliaceae Ilex vomitoria 0.00% 1.50% 1.50% Shrub Perennial

Asclepiadaceae Asclepias obovata 2.50% 0.50% 0.00% Forb/herb Perennial

Asteraceae Baccharis halimifolia 2.00% 5.50% 6.00% Shrub Perennial Asteraceae Conyza canadensis 19.50% 0.00% 1.50% Forb/herb Annual/biennial

Asteraceae Eupatorium capillifolium 41.50% 47.00% 16.00% Forb/herb Perennial

Asteraceae Eupatorium compositifolium 3.00% 0.00% 0.50% Forb/herb Perennial Asteraceae Eupatorium leucolepsis 11.00% 16.50% 13.50% Forb/herb Perennial

Asteraceae Eupatorium perfoliatum 2.00% 0.50% 1.50% Forb/herb Perennial

Asteraceae Eupatorium rotundifolium 45.00% 37.50% 21.00% Forb/herb Perennial Asteraceae Eupatorium serotinum 5.50% 1.00% 0.00% Forb/herb Perennial

Asteraceae Euthamia leptocephala 4.50% 7.00% 1.50% Forb/herb Perennial

Asteraceae Gaillardia aestivalis 0.00% 0.50% 0.00% Forb/herb Perennial Asteraceae Gamochaeta purpurea 7.50% 19.00% 2.50% Forb/herb Annual/biennial

Asteraceae Helianthus angustifolius 77.00% 72.00% 63.50% Forb/herb Perennial

Asteraceae Hymenopappus artemisiifolius 1.50% 0.00% 0.00% Forb/herb Biennial Asteraceae Liatris acidota 0.00% 0.50% 0.00% Forb/herb Perennial

Asteraceae Pityopsis graminifolia 0.50% 0.00% 0.00% Forb/herb Perennial

Asteraceae Pyrrhopappus carolinianus 0.00% 3.00% 7.50% Forb/herb Annual/biennial

Asteraceae Rudbeckia grandiflora 1.00% 1.00% 0.00% Forb/herb Perennial

Asteraceae Rudbeckia hirta 2.00% 4.50% 3.00% Forb/herb Perennial

Asteraceae Silphium gracile 0.00% 1.50% 0.50% Forb/herb Perennial

Asteraceae Solidago altissima 20.50% 10.00% 1.50% Forb/herb Perennial

Asteraceae Solidago rugosa 3.50% 8.50% 21.00% Forb/herb Perennial

Asteraceae Symphyotrichum dumosum 5.50% 13.50% 32.50% Forb/herb Perennial

Asteraceae Pluchea rosea 0.00% 2.00% 0.00% Forb/herb Perennial

Asteraceae Bidens sp. 0.00% 0.50% 0.00% Forb/herb Perennial

Bignogniaceae Campsis radicans 2.50% 4.00% 5.00% Vine Perennial

Campanulaceae Lobelia appendiculata 0.50% 0.00% 0.50% Forb/herb Annual/biennial

Campanulaceae Lobelia puberula 0.00% 0.50% 1.50% Forb/herb Perennial

119

Cistaceae Lechea tenuifolia 0.50% 0.00% 0.00% Forb/herb Perennial

Cyperaceae Cyperus echinatus 2.00% 0.50% 0.00% Graminoid Perennial

Cyperaceae Fuirena breviseta 1.50% 2.00% 0.50% Graminoid Perennial

Cyperaceae Rhynchospora caduca 12.00% 22.00% 14.50% Graminoid Perennial

Cyperaceae Rhynchospora gracilenta 0.50% 2.00% 2.50% Graminoid Perennial

Cyperaceae Rhynchospora inexpansa 0.50% 0.50% 0.00% Graminoid Perennial

Cyperaceae Rhynchospora recognita 16.00% 4.50% 4.50% Graminoid Perennial

Cyperaceae Rhynchospora plumosa 0.00% 1.50% 0.00% Graminoid Perennial

Dennstaedtiaceae Pteridium aquilinum 19.00% 23.50% 22.50% Forb/herb Perennial

Ericaceae Vaccinium arboreum 6.00% 0.50% 2.50% Shrub Perennial

Ericaceae Vaccinium elliottii 0.50% 0.00% 0.50% Shrub Perennial

Euphorbiaceae Euphorbia corollata 0.00% 2.50% 0.00% Forb/herb Perennial

Fabaceae Centrosema virginianum 14.00% 1.00% 0.50% Forb/herb Perennial Fabaceae Crotalaria sagittalis 5.00% 0.00% 0.00% Forb/herb Annual

Fabaceae Lespedeza capitata 14.00% 2.50% 0.50% Forb/herb Perennial

Fabaceae Mimosa strigillosa 0.00% 2.00% 2.50% Forb/herb Perennial Fabaceae Rhynchosia tomentosa 1.50% 0.50% 0.00% Forb/herb Perennial

Fabaceae Tephrosia onobrychoides 17.50% 15.00% 7.50% Forb/herb Perennial

Fagaceae Quercus marilandica 0.00% 0.00% 1.50% Tree Perennial Gelsemiaceae Gelsemium sempervirens 0.00% 0.50% 0.00% Vine Perennial

Gentianaceae Centaurium pulchellum 5.50% 2.50% 0.00% Forb/herb Annual

Gentianaceae Sabatia sp. 0.00% 0.50% 0.00% Forb/herb Annual Haloragaceae Proserpinaca pectinata 2.00% 1.00% 5.00% Forb/herb Perennial

Hamamelidaceae Liquidambar styraciflua 0.50% 0.50% 0.50% Tree Perennial

Hypericaceae Hypericum fasciculatum 1.00% 1.00% 0.00% Shrub Perennial Hypericaceae Hypericum galioides 1.00% 1.00% 0.00% Shrub Perennial

Hypericaceae Hypericum hypericoides 13.50% 1.00% 2.00% Shrub Perennial

Hypericaceae Hypericum stans 24.50% 12.50% 1.00% Shrub Perennial Juncaceae Juncus brachicarpus 5.50% 0.50% 0.00% Graminoid Perennial

Juncaceae Juncus polycephalus 0.50% 8.50% 0.00% Graminoid Perennial

Lamiaceae Monarda punctata 0.00% 1.00% 0.00% Forb/herb Perennial

Lamiaceae Pycnanthemum tenuifolium 0.50% 1.00% 0.00% Forb/herb Perennial

Lamiaceae Scutellaria integrifolia 2.50% 1.50% 1.50% Forb/herb Perennial

Liliaceae Aletris aurea 0.50% 1.00% 0.00% Forb/herb Perennial

Linaceae Linum medium var. texanum 0.00% 2.50% 3.50% Forb/herb Annual

Lycopodiaceae Lycopodiella sp. 0.50% 0.00% 0.00% Forb/herb Perennial

Lygodiaceae Lygodium japonicum 1.00% 2.00% 0.00% Vine Perennial

Melastomataceae Rhexia alifanus 0.00% 0.50% 0.00% Forb/herb Perennial

Melastomataceae Rhexia lutea 1.50% 1.00% 5.00% Forb/herb Perennial

Melastomataceae Rhexia mariana 7.50% 26.00% 1.50% Forb/herb Perennial

120

Myricaceae Morella cerifera 0.00% 1.00% 0.50% Shrub Perennial

Nyssaceae Nyssa sylvatica 0.00% 0.00% 5.50% Tree Perennial

Onagraceae Ludwigia pilosa 4.50% 1.00% 1.00% Forb/herb Perennial

Orobanchaceae Agalinis filicaulis 0.50% 1.00% 0.50% Forb/herb Annual

Oxalidaceae Oxalis sp. 0.00% 1.00% 9.50% Forb/herb Perennial

Poaceae Schizachyrium scoparium 96.50% 72.50% 64.50% Graminoid Perennial

Poaceae Dichanthelium aciculare 68.50% 76.00% 65.50% Graminoid Perennial

Poaceae Dichanthelium scabrisculum 79.50% 75.00% 1.00% Graminoid Perennial

Poaceae Panicum verrucosum 8.50% 3.50% 2.50% Graminoid Perennial

Poaceae Paspalum plicatum 6.00% 7.00% 0.00% Graminoid Perennial

Poaceae Setaria parviflora 0.50% 0.50% 0.00% Graminoid Perennial

Polygalaceae Polygala cruciata 0.00% 0.50% 0.00% Forb/herb Annual

Polygalaceae Polygala cymosa 0.00% 0.00% 2.50% Forb/herb Biennial Polygalaceae Polygala mariana 31.50% 17.00% 0.00% Forb/herb Annual

Polygalaceae Polygala sanguinea 0.00% 0.50% 0.00% Forb/herb Annual

Rosaceae Prunus serotina 0.00% 11.00% 0.00% Tree Perennial Rosaceae Rubus trivialis 25.00% 28.00% 32.00% Vine Perennial

Rubiaceae Diodia virginiana 17.50% 5.50% 0.00% Forb/herb Annual

Rubiaceae Diodia teres 1.00% 1.50% 3.00% Forb/herb Annual Scrophulariaceae Mecardonia acuminata 26.00% 13.50% 1.00% Forb/herb Perennial

Smilaceae Smilax glauca 29.50% 26.50% 30.00% Vine Perennial

Symplocaceae Symplocos tinctoria 5.50% 4.50% 0.00% Shrub Perennial Tetrachondraceae Polypremum procumbens 4.00% 0.00% 1.00% Forb/herb Annual

Verbenaceae Callicarpa americana 1.50% 3.00% 0.00% Shrub Perennial

Vitaceae Vitis rotundifolia 0.50% 0.00% 0.00% Vine Perennial Xiridaceae Xyris ambigua 1.00% 0.00% 2.00% Forb/herb Perennial

Xiridaceae Xyris stricta var. obscura 0.00% 4.00% 0.50% Forb/herb Perennial

Unknown Unknown 0.00% 0.00% 1.50% - - Unknown Unknown 0.00% 0.00% 0.50% - -

Unknown Unknown 0.00% 0.00% 1.00% - -

121

A3. Understory species found in P. elliottii plantation established in 2008 (S08)

Abundance Family Scientific name Form Duration 2014 2015 2016

Aceraceae Acer rubrum 0.00% 1.50% 0.00% Tree Perennial Anacardiaceae Rhus copallinum 1.00% 1.50% 0.50% Shrub Perennial

Anacardiaceae Toxicodendron radicans 6.00% 8.00% 5.50% Vine Perennial

Annonaceae Asimina parviflora 0.50% 1.50% 0.00% Tree Perennial Apiaceae Centella crenata 3.50% 0.50% 0.00% Forb/herb Perennial

Apiaceae Centella erecta 0.50% 2.50% 1.00% Forb/herb Perennial

Aquifoliaceae Ilex opaca 1.50% 2.00% 1.50% Shrub Perennial Aquifoliaceae Ilex vomitoria 11.50% 11.50% 6.50% Shrub Perennial

Asteraceae Baccharis halimifolia 0.00% 1.50% 0.00% Shrub Perennial

Asteraceae Conyza canadensis 0.00% 0.00% 1.00% Forb/herb Annual/biennial Asteraceae Echinacea sanguinea 0.50% 0.00% 0.00% Forb/herb Perennial

Asteraceae Eupatorium capillifolium 2.00% 2.00% 0.00% Forb/herb Perennial

Asteraceae Eupatorium leucolepsis 13.50% 29.50% 5.50% Forb/herb Perennial Asteraceae Eupatorium compositifolium 0.00% 0.50% 0.50% Forb/herb Perennial

Asteraceae Eupatorium resinosum 0.00% 7.50% 2.50% Forb/herb Perennial

Asteraceae Eupatorium rotundifolium 16.50% 11.50% 0.50% Forb/herb Perennial Asteraceae Eupatorium serotinum 1.50% 0.00% 0.00% Forb/herb Perennial

Asteraceae Gamochaeta purpurea 0.50% 0.00% 5.50% Forb/herb Annual/biennial

Asteraceae Helianthus angustifolius 16.50% 7.50% 0.00% Forb/herb Perennial Asteraceae Lactuca sp. 0.00% 0.00% 0.50% Forb/herb Annual/biennial

Asteraceae Liatris acidota 1.00% 0.00% 0.00% Forb/herb Perennial

Asteraceae Rudbeckia hirta 0.00% 0.00% 0.50% Forb/herb Perennial Asteraceae Solidago altissima 2.50% 0.50% 0.00% Forb/herb Perennial

Asteraceae Solidago rugosa 0.50% 0.00% 1.00% Forb/herb Perennial

Asteraceae Symphyotrichum dumosum 1.00% 0.50% 2.50% Forb/herb Perennial

Bignogniaceae Campsis radicans 2.00% 1.50% 2.00% Vine Perennial

Cyperaceae Rhynchospora caduca 0.50% 5.50% 0.00% Graminoid Perennial

Cyperaceae Rhynchospora gracilenta 0.00% 0.50% 0.00% Graminoid Perennial

Cyperaceae Rhynchospora inexpansa 0.00% 0.50% 0.00% Graminoid Perennial

Cyperaceae Rhynchospora plumosa 0.00% 0.50% 0.00% Graminoid Perennial

Cyperaceae Rhynchospora recognita 0.50% 0.00% 2.00% Graminoid Perennial

Cyperaceae Cyperus esculentus 0.00% 0.50% 0.00% Graminoid Perennial

Dennstaedtiaceae Pteridium aquilinum 9.50% 9.50% 7.00% Forb/herb Perennial

Ericaceae Vaccinium arboreum 1.00% 0.50% 0.00% Shrub Perennial

Ericaceae Vaccinium elliottii 0.00% 0.50% 1.00% Shrub Perennial

Euphorbiaceae Euphorbia corollata 0.00% 0.50% 0.00% Forb/herb Perennial

122

Fabaceae Centrosema virginianum 5.50% 0.00% 0.00% Forb/herb Perennial

Fabaceae Lespedeza capitata 8.00% 3.50% 0.00% Forb/herb Perennial

Fabaceae Mimosa strigillosa 0.00% 2.00% 0.00% Forb/herb Perennial

Fabaceae Rhynchosia tomentosa 1.00% 0.00% 0.00% Forb/herb Perennial

Fabaceae Tephrosia onobrychoides 6.50% 5.00% 0.50% Forb/herb Perennial

Fagaceae Quercus nigra 0.00% 1.00% 0.50% Tree Perennial

Gelsemiaceae Gelsemium sempervirens 46.50% 41.00% 41.50% Vine Perennial

Gentianaceae Sabatia sp. 0.00% 0.50% 0.00% Forb/herb Annual

Haloragaceae Proserpinaca pectinata 1.00% 3.50% 2.00% Forb/herb Perennial

Hamamelidaceae Liquidambar styraciflua 5.00% 2.00% 5.00% Tree Perennial

Hypericaceae Hypericum hypericoids 0.50% 0.00% 0.00% Shrub Perennial

Hypericaceae Hypericum stans 1.50% 2.00% 0.50% Shrub Perennial

Hypericaceae Hypericum punctatum 0.00% 0.50% 0.00% Shrub Perennial Juncaceae Juncus brachicarpus 0.50% 0.00% 0.00% Graminoid Perennial

Lamiaceae Scutellaria integrifolia 0.50% 0.50% 0.00% Forb/herb Perennial

Lycopodiaceae Lycopodiella sp. 1.50% 0.00% 0.00% Forb/herb Perennial Lygodiaceae Lygodium japonicum 1.50% 0.50% 0.50% Vine Perennial

Magnoliaceae Magnolia virginiana 4.50% 0.00% 0.00% Tree Perennial

Melastomataceae Rhexia alifanus 0.50% 0.00% 0.00% Forb/herb Perennial Melastomataceae Rhexia lutea 7.00% 1.00% 0.00% Forb/herb Perennial

Melastomataceae Rhexia mariana 18.00% 5.50% 0.00% Forb/herb Perennial

Myricaceae Myrica cerifera 1.50% 2.00% 1.00% Shrub Perennial Onagraceae Ludwigia pilosa 1.00% 7.00% 3.00% Forb/herb Perennial

Orobanchaceae Agalinis filicaulis 0.00% 0.00% 1.00% Forb/herb Annual

Oxalidaceae Oxalis sp. 0.00% 0.00% 0.50% Forb/herb Perennial Poaceae Schizachyrium scoparium 94.50% 61.00% 62.00% Graminoid Perennial

Poaceae Dichanthelium aciculare 16.00% 5.50% 1.50% Graminoid Perennial

Poaceae Dichanthelium scabrisculum 34.00% 33.50% 30.00% Graminoid Perennial Poaceae Panicum verrucosum 4.50% 34.00% 16.00% Graminoid Perennial

Rosaceae Rubus trivialis 47.00% 46.00% 38.50% Vine Perennial

Rubiaceae Diodia virginiana 24.00% 17.00% 1.50% Forb/herb Annual

Rubiaceae Diodia teres 0.00% 7.50% 0.00% Forb/herb Annual

Smilaceae Smilax glauca 62.00% 54.50% 44.50% Vine Perennial

Symplocaceae Symplocos tinctoria 5.00% 6.50% 7.00% Shrub Perennial

Tetrachondraceae Polypremum procumbens 0.00% 0.50% 0.00% Forb/herb Annual

Verbenaceae Callicarpa americana 2.50% 1.00% 1.00% Shrub Perennial

Vitaceae Vitis rotundifolia 3.00% 4.00% 2.00% Vine Perennial

Xiridaceae Xyris ambigua 0.50% 0.00% 0.00% Forb/herb Perennial

Xiridaceae Xyris stricta var. obscura 0.00% 0.50% 0.00% Forb/herb Perennial

123

APPENDIX B. DOMINANT UNDERSTORY SPECIES LIST

Plantation Site Species Relative abundance (wi) E13 R6 D. virginiana 0.03 E13 R6 E. capillifolium 0.05 E13 R6 E. leucolepsis 0.12 E13 R6 L. japonicum 0.06 E13 R6 R. trivialis 0.22 E13 R6 S. glauca 0.12 E13 R6 S. scoparium 0.2

E13 R8 D. scabrisculum 0.13 E13 R8 E. capillifolium 0.08 E13 R8 R. trivialis 0.26 E13 R8 S. glauca 0.12 E13 R8 S. scoparium 0.24

E13 R9 B. halimifolia 0.08 E13 R9 D. virginiana 0.07 E13 R9 E. capillifolium 0.07 E13 R9 E. perfoliatum 0.07 E13 R9 R. trivialis 0.19 E13 R9 S. altissima 0.07 E13 R9 S. glauca 0.12 E13 R9 S. scoparium 0.21

E13 R10 B. halimifolia 0.11 E13 R10 E. capillifolium 0.13 E13 R10 E. leucolepsis 0.1 E13 R10 E. rotundifolius 0.13 E13 R10 H. angustifolius 0.09 E13 R10 S. scoparium 0.34

E13 R11 B. halimifolia 0.04 E13 R11 C. americana 0.04 E13 R11 D. scabrisculum 0.18 E13 R11 E. capillifolium 0.03 E13 R11 E. perfoliatum 0.06 E13 R11 R. trivialis 0.15

124

E13 R11 S. glauca 0.09 E13 R11 S. rugosa 0.04 E13 R11 S. scoparium 0.18

S13 S13-2 D. aciculare 0.1 S13 S13-2 D. scabrisculum 0.14 S13 S13-2 E. capillifolium 0.04 S13 S13-2 E. rotundifolius 0.04 S13 S13-2 H. angustifolius 0.15 S13 S13-2 P. aquilinum 0.11 S13 S13-2 R. recognita 0.04 S13 S13-2 S. rugosa 0.05 S13 S13-2 S. scoparium 0.16 S13 S13-2 T. onobrychoides 0.03

S13 S13-6 D. aciculare 0.12 S13 S13-6 D. scabrisculum 0.14 S13 S13-6 E. capillifolium 0.04 S13 S13-6 E. leptocephala 0.03 S13 S13-6 E. rotundifolius 0.05 S13 S13-6 H. angustifolius 0.02 S13 S13-6 P. aquilinum 0.05 S13 S13-6 R. gracilenta 0.02 S13 S13-6 R. mariana 0.03 S13 S13-6 R. trivialis 0.08 S13 S13-6 S. altissima 0.03 S13 S13-6 S. glauca 0.14 S13 S13-6 S. scoparium 0.15

S13 S13-9 D. aciculare 0.17 S13 S13-9 D. scabrisculum 0.13 S13 S13-9 E. capillifolium 0.04 S13 S13-9 E. leptocephala 0.03 S13 S13-9 E. rotundifolius 0.06 S13 S13-9 H. angustifolius 0.13 S13 S13-9 P. carolinianus 0.04 S13 S13-9 S. altissima 0.03 S13 S13-9 S. glauca 0.05 S13 S13-9 S. scoparium 0.17

125

S13 S13-17 C. erecta 0.03 S13 S13-17 D. aciculare 0.07 S13 S13-17 D. scabrisculum 0.08 S13 S13-17 D. virginiana 0.03 S13 S13-17 E. capillifolium 0.06 S13 S13-17 E. leucolepsis 0.03 S13 S13-17 E. rotundifolius 0.06 S13 S13-17 H. angustifolius 0.1 S13 S13-17 R. caduca 0.05 S13 S13-17 R. mariana 0.03 S13 S13-17 R. trivialis 0.06 S13 S13-17 S. altissima 0.02 S13 S13-17 S. glauca 0.07 S13 S13-17 S. scoparium 0.16

S13 S13-26 D. aciculare 0.13 S13 S13-26 D. scabrisculum 0.11 S13 S13-26 E. rotundifolius 0.04 S13 S13-26 H. angustifolius 0.17 S13 S13-26 H. hypericoides 0.03 S13 S13-26 P. plicatum 0.04 S13 S13-26 P. verrucosum 0.03 S13 S13-26 S. glauca 0.05 S13 S13-26 S. scoparium 0.18 S13 S13-26 T. onobrychoides 0.05

S08 S08-1 D. aciculare 0.08 S08 S08-1 E. rotundifolius 0.01 S08 S08-1 G. sempervirens 0.23 S08 S08-1 I. vomitoria 0.05 S08 S08-1 P. aquilinum 0.04 S08 S08-1 R. mariana 0.08 S08 S08-1 R. trivialis 0.13 S08 S08-1 S. glauca 0.12 S08 S08-1 S. scoparium 0.13

S08 S08-3 D. scabrisculum 0.07 S08 S08-3 G. sempervirens 0.01 S08 S08-3 R. mariana 0.01 S08 S08-3 R. trivialis 0.09

126

S08 S08-3 S. glauca 0.23 S08 S08-3 S. scoparium 0.41

S08 S08-17 E. rotundifolius 0.03 S08 S08-17 G. sempervirens 0.08 S08 S08-17 S. glauca 0.08 S08 S08-17 R. trivialis 0.17 S08 S08-17 D. scabrisculum 0.12 S08 S08-17 I. vomitoria 0.11 S08 S08-17 H. angustifolius 0.04 S08 S08-17 S. scoparium 0.17

S08 S08-32 D. scabrisculum 0.19 S08 S08-32 D. virginiana 0.05 S08 S08-32 E. leucolepsis 0.06 S08 S08-32 G. sempervirens 0.18 S08 S08-32 R. trivialis 0.05 S08 S08-32 S. glauca 0.17 S08 S08-32 S. scoparium 0.24

S08 S08-36 D. aciculare 0.05 S08 S08-36 D. scabrisculum 0.09 S08 S08-36 D. virginiana 0.04 S08 S08-36 R. lutea 0.04 S08 S08-36 R. mariana 0.04 S08 S08-36 R. trivialis 0.21 S08 S08-36 S. glauca 0.22 S08 S08-36 S. scoparium 0.19

127

VITA

Andrea De Stefano is a native of Reggio Calabria, Italy. He attended the Università

Mediterranea di Reggio Calabria from Decemeber 1999 to December 2005, where he earned his

Laurea Magistrale (a second cycle degree equivalent to a master's degree) in Forest and

Environmental Sciences. In August 2012, Andrea joined the Pennsylvania State University, where he received his master’s degree in Forest Resources August 2014, with thesis research focused on soil organic carbon stocks and carbon biomass in agroforestry systems. After graduating, Andrea began his doctoral program at Louisiana State University in August 2014 under the supervision of Dr. Michael Blazier. He will receive his doctoral degree in May 2019.

128