Asplenium Scolopendrium Var. Americanum): Factors Affecting Successful Establishment of Transplants

SUNY College of Environmental Science and Forestry Digital Commons @ ESF

Dissertations and Theses

Fall 12-16-2017

Experimental Reintroduction of American Hart’s-Tongue Fern (Asplenium scolopendrium var. americanum): Factors Affecting Successful Establishment of Transplants

Michael Serviss [email protected]

Follow this and additional works at: https://digitalcommons.esf.edu/etds

Recommended Citation Serviss, Michael, "Experimental Reintroduction of American Hart’s-Tongue Fern (Asplenium scolopendrium var. americanum): Factors Affecting Successful Establishment of Transplants" (2017). Dissertations and Theses. 9. https://digitalcommons.esf.edu/etds/9

This Open Access Thesis is brought to you for free and open access by Digital Commons @ ESF. It has been accepted for inclusion in Dissertations and Theses by an authorized administrator of Digital Commons @ ESF. For more information, please contact [email protected], [email protected]. EXPERIMENTAL REINTRODUCTION OF AMERICAN HART’S-TONGUE FERN

(ASPLENIUM SCOLOPENDRIUM VAR. AMERICANUM): FACTORS AFFECTING

SUCCESSFUL ESTABLISHMENT OF TRANSPLANTS

Michael J. Serviss

A thesis submitted in partial fulfillment of the requirements for the Master of Science Degree State University of New York College of Environmental Science and Forestry Syracuse, New York December 2017

Department of Environmental and Forest Biology

Approved by: Danilo Fernando, Major Professor Theresa Selfa, Chair, Examining Committee Donald Leopold, Department Chair S. Scott Shannon, Dean, The Graduate School

Acknowledgements

I would like to sincerely thank all of those who made this work possible: my major professor, Dr. Danilo Fernando, whose mentorship helped to guide me through this project; my steering and defense committee members, Thomas Hughes, Drs. Donald Leopold, Robin

Kimmerer and Theresa Selfa, whose advice and feedback were invaluable; my colleagues,

Joshua Weber-Townsend, Shannon Fabiani, Patricia Shulenburg, Mat Bilz and the entire

FORCES “Fern Crew”; my field assistants, Kelsey West and Olivia Morris; the many (MANY) volunteers who donated their time to this project; and last, but not least, my family and friends who have provided support and encouragement (and sometimes delicious baked goods) every step of the way. Special thanks to my dog, the Dude, for abiding.

Funding for this project was provided by the US Fish and Wildlife Service through the

Great Lakes Restoration Initiative, the New York State Office of Parks, Recreation, and Historic

Preservation, the New York State Department of Environmental Conservation Section 6, the

Edna Bailey Sussman Foundation, and the Josiah Lowe and Hugh Wilcox Scholarship Fund.

iii

Table of Contents

List of Tables ...... vi List of Figures ...... vii Abstract ...... viii

Chapter 1: Introduction ...... 1 Pteridophyte Conservation ...... 1 Reintroduction ...... 2 Pteridophyte Reintroduction Success Stories ...... 6 Case Studies from the Global Re-introduction Series ...... 6 Marsilea quadrifolia ...... 7 Diellia pallida ...... 7 Woodsia ilvensis ...... 8 Study Species ...... 9 Varietal Status and Morphology ...... 9 Ecology and Distribution ...... 12 Legal Status ...... 16 Goals and Objectives ...... 17 Study Area ...... 18 Sentinel Basin ...... 18 Twin Basins ...... 19 Split Rock Unique Area ...... 19 Literature cited ...... 21

Chapter 2: Experimental Reintroduction of Asplenium scolopendrium var. americanum: Factors Affecting Survival and Growth ...... 25 Introduction ...... 25 History of Translocation and Reintroduction of AHTF in New York ...... 25 General Reintroduction Considerations ...... 26 Objectives ...... 31 Hypotheses ...... 32 Methods ...... 32 Propagation ...... 32 Acclimatization ...... 33 Reintroduction Site Selection ...... 34 Transplanting ...... 34 Monitoring and Data Collection ...... 36 Statistical Analyses ...... 38 Results ...... 40 Effects of Life Stage on Survival and Growth ...... 40 Ordinal Logistic Regression and Predictors of Transplant Survival ...... 50 Discussion ...... 54 Effects of Life Stage on Survival and Growth ...... 54 Ordinal Logistic Regression and Predictors of Transplant Survival ...... 58 Literature cited ...... 61

Chapter 3: Microclimate of Reintroduction Sites and Implications for Transplant Survival ...... 65 Introduction ...... 65 Methods ...... 69 Study Design...... 69 Data Analysis ...... 71 Results ...... 72 Discussion ...... 81 Literature cited ...... 85

Chapter 4: Conclusions and Recommendations ...... 89 Literature cited ...... 95

APPENDIX A ...... 97 APPENDIX B ...... 98 APPENDIX C ...... 102

Resume ...... 106

List of Tables

Table 2.1. Vigor ratings and respective descriptions for AHTF transplants ...... 38 Table 2.2. Summary of the number of initial transplants and confirmed surviving transplants over a 24-month monitoring period ...... 45 Table 2.3. One-way analysis of variance results for mean percent survival at 24 months ...... 47 Table 2.4. Mean percent survival (± standard error) of AHTF transplants over 20 months ...... 48 Table 2.5. Results of the Proportional Odds Model ordinal logistic regression analysis for AHTF transplants across three reintroduction sites...... 53 Table 4.1. Candidate sites for AHTF reintroduction and augmentation...... 91

List of Figures

Figure 1.1. A general model of integrated plant conservation...... 4 Figure 1.2. An individual mature sporophyte of Asplenium scolopendrium var. americanum .... 10 Figure 1.3. The stages of early development of Asplenium scolopendrium var. americanum. .... 12 Figure 1.4. A portion of one population of Asplenium scolopendrium var. americanum growing on a talus slope ...... 14 Figure 1.5. General locations of the three selected AHTF reintroduction sites ...... 20 Figure 2.1. AHTF sporelings in Twin Basins in July 2017 ...... 46 Figure 2.2. Raw percent survival of four transplant-types of AHTF ...... 47 Figure 2.3. Mean percent survival and standard error of four transplant-types of AHTF over 20 months...... 48 Figure 2.4. Mean number of fronds and standard error of AHTF transplants of three transplant- types over 24 months...... 49 Figure 2.5. Mean total frond length and standard error of AHTF transplants of three transplant- types over 24 months...... 49 Figure 3.1. A microclimate monitoring station mounted in Clark Reservation State Park ...... 70 Figure 3.2. Wind radar chart for Sentinel Basin lower slope from 2015-2017...... 77 Figure 3.3. Wind radar chart for Sentinel Basin mid-slope from 2015-2017...... 77 Figure 3.4. Wind radar chart for Sentinel Basin upper slope from 2015-2017...... 78 Figure 3.5. Wind radar chart for Split Rock lower slope from 2015-2017...... 78 Figure 3.6. Wind radar chart for Split Rock mid-slope from 2015-2017...... 79 Figure 3.7. Wind radar chart for Split Rock upper slope for 2015 ...... 79 Figure 3.8. Wind radar chart for Twin Basins lower slope from 2015-2016...... 80 Figure 3.9. Wind radar chart for Twin Basins mid-slope from 2015-2017...... 80

vii

Abstract

M. J. Serviss. Experimental Reintroduction of American Hart’s-Tongue Fern (Asplenium scolopendrium var. americanum): Factors Affecting Successful Establishment of Transplants, 108 pages, 6 tables, 19 figures, 2017. Conservation Biology style guide used.

This study analyzed biological and microclimatic factors affecting the post-reintroduction survival and growth of laboratory-propagated American hart’s-tongue ferns (Asplenium scolopendrium var. americanum, AHTF), a rare and threatened fern species with a fragmented distribution across the eastern United States and Canada. In total, 1,925 AHTF transplants, representing four life history stages (protonemata, gametophytes, sporelings, and immature sporophytes) were reintroduced into three sites determined to be ideal habitats within Onondaga County, New York. Factors that resulted in higher rates of survival included more rigorous acclimatization of transplants, better transplant vigor at the time of transplanting, and higher humidity at reintroduction sites. Analysis of wind speed and direction provided implications of low spore dispersal, and therefore low gene flow, between extant AHTF populations. Insights from this research contribute to a growing, yet still understudied, body of knowledge regarding fern species reintroductions and inform best practices for future AHTF reintroductions.

Key Words: fern reintroduction, life history stage, acclimatization, spore dispersal

M. J. Serviss Candidate for the degree of Master of Science, December 2017 Danilo D. Fernando, Ph.D. Department of Environmental and Forest Biology State University of New York College of Environmental Science and Forestry, Syracuse, New York

viii

Chapter 1: Introduction

Pteridophyte Conservation

Concerns in the field of plant conservation are growing as an increasing number of plant species face the risk of extinction across the globe. Brummitt et al. (2015) estimate that, according to IUCN red list standards, more than one out of every five plant species are threatened with extinction and nearly one third of plant species are near-threatened. The authors further estimate that 16% of all pteridophyte and lycophyte species are threatened and 6% are near-threatened, meaning that 22% are of elevated concern to conservation. These proportions are concerning as surveys of the conservation status and management actions regarding pteridophytes and their allies are considered by some to be lacking (Blackmore & Walter 2007;

Brummitt et al. 2016). Some research has also noted that there is limited literature with a focus on pteridophyte conservation compared with the volume of literature focused on the conservation of flowering plants (Given 1993) and that many urgent research needs regarding ferns and fern allies have yet to be addressed (Brummitt et al. 2016; Gasper et al. 2016).

While there appears to be a general lack of research focused on pteridophyte diversity and conservation, several recent studies have ventured to address these concerns. Research with a global focus on pteridophyte conservation has identified hotspots of fern endemism and diversity and suggested priority conservation management actions (See Given 1993; Kreft et al.

2010; Brummitt et al. 2016). An even larger volume of literature focuses on quantifying pteridophyte diversity on a more regional scale and often correlates this diversity with various environmental factors including climate, elevation, and habitat characteristics (See Young &

Leon 1989; Tuomisto & Poulsen 2000; Kessler 2001; Paciencia & Prado 2005; Kessler &

Lehnert 2009). Studies focusing on the conservation of individual species of pteridophytes are relatively more abundant than global or regional scale studies, mainly due to the research requirements of recovery plans that tend to focus on single species conservation.

Several threats contribute to the risk of extinction of pteridophyte species globally. The most pressing threats originate from sources related to human activity, including agriculture, climate change, invasive species, logging, urban development, and mining, among others.

Additional threats include natural disasters and vulnerability of the intrinsic biology of some pteridophytes (Brummitt et al. 2016; Corlett 2016). To address these threats, the consensus in conservation is that in situ methods are often the most appropriate and desirable from a management perspective (Kaye 2017). This is mainly because, in contrast to ex situ methods, more biologic and genetic diversity can be conserved in situ. Additionally, some issues surrounding ex situ conservation can more readily be avoided, including inbreeding, hybridization, artificial selection, and lack of space to represent the complete diversity of a species. However, the Global Strategy for Plant Conservation has set desired targets for the ex situ conservation of plant species to help fill in the gaps when in situ conservation is ineffective or not feasible. These targets include the ex situ conservation of at least 75% of threatened plant species with 20% being available for recovery and restoration projects (Corlett 2016).

Reintroduction

Reintroduction of threatened and endangered plants is a method of population and/or species recovery that is growing in popularity in the field of plant conservation. Reintroduction

2 refers to the physical placement of plants or plant materials from various sources into an area where they had once occurred historically, but are no longer found, for conservation purposes.

The term reintroduction also refers to population augmentation, where plants or plant materials are added to an existing population. Reintroduction can also be used in the context of introducing plants to an area where there is no documented historical occurrence. Lastly, reintroduction can occur in the context of translocation, where plants or plant materials are removed from an in situ area and placed elsewhere. In each of these cases, reintroduction generally occurs within the species known geographic range for the purpose of ensuring the persistence of a species (Allen

1994; Guerrant Jr. 2012).

The framework for integrated plant conservation (Figure 1.1) outlines the placement of reintroduction projects in the overall practice of plant conservation. Reintroduction acts as an intermediary between in situ and ex situ conservation methods. The crucial link between in situ and ex situ methodology helps to facilitate conservation goals and the persistence of rare, threatened and endangered plant species when in situ methods alone are not sufficient (Kaye

2017). The link between ex situ and in situ conservation has strengthened over the past few decades. Practitioners of both methods have allied with the aim of producing more desirable conservation outcomes, including the restoration of some of the most endangered plant species in the world. In the United States, this alliance is perhaps best embodied by the Missouri Botanical

Garden’s Center for Conservation and Sustainable Development (CCSD). The CCSD conducts research on experimental reintroduction of endangered plants with integration of critical reintroduction factors, including genetic, habitat, and microenvironmental considerations. The reintroduction programs at the CCSD are part of a growing connection between ex situ facilities such as botanical gardens and arboreta with in situ research projects. Botanical gardens and

3 arboreta have historically featured rare exotic plants but are increasingly becoming a critical part of integrated plant conservation and reintroduction. One example of the critical role that botanical gardens play in reintroduction projects is that of the fern Woodsia ilvensis in the United

Kingdom. The Royal Botanic Garden of Edinburgh developed propagation protocol for W. ilvensis that produced over 2,500 sporophytes, many of which were used in a successful reintroduction project outlined in the next section of this manuscript. Research on how to best improve plant collections at ex situ facilities so that they are applicable to reintroduction projects is ongoing (See Pritchard et al. 2010; Oldfield & Newton 2012; Cavender et al. 2015).

Figure 1.1. A general model of integrated plant conservation, including the role of reintroduction or augmentation as a bridge between in situ and ex situ conservation practices. Source: Kramer et al. 2011.

Reintroduction has been promoted as a means of recovering imperiled plant species by conservation organizations throughout the world. While this recovery strategy has gained in popularity over the past several decades, quantitating the effectiveness of plant reintroductions

4 has suffered from a lack of understanding of criteria that define success or failure (Godefroid et al. 2011; Maschinski & Haskins 2012). Allen (1994) reported that in a survey of 45 plant reintroductions in California, only four reintroductions were considered completely successful, where success was defined as the ability of a population to survive and reproduce. He also reported the results of a survey performed by the British Nature Conservancy Council in 1991 that stated only 22% of 145 plant reintroductions were considered successful. Criteria for success or failure was not defined for the 1991 survey. In each of these cases, specific objectives and goals (plant growth rates, reproduction rates, population growth, timeframe to determine success, etc.) were not defined to provide context for the criteria to determine relative success. A lack of specific objectives and definitions of success are common issues with many plant reintroduction projects.

Godefroid et al. (2011) first attempted to analyze the relative success of plant reintroductions to better understand the effectiveness of the practice and determine which approaches may lead to greater success in establishing viable plant populations. Their meta- analysis of 249 plant reintroductions found that reintroduction success is often over-reported in literature and generally based on short-term data that may not be sufficient to predict long-term outcomes. Furthermore, they found that probability of success increased when working in protected areas, using larger/more developed transplants, using a diverse mix of source population material, increasing number of transplants, and managing the transplant site before and after reintroduction.

Dalrymple et al. (2012) also conducted a meta-analysis of threatened plant reintroductions with the purpose of characterizing criteria important to understanding the relative success or failure of reintroductions. In a subgroup analysis of the use of single or multiple donor

5 populations for reintroduction they concluded that, in the case of reintroduction of juvenile plants, survival was higher when multiple donor populations were used. They went on to characterize the common causes of failure of plant reintroductions, which include drought shortly after transplanting, competition from invasive plant species, inherent biological vulnerability of the species, and reintroducing plants too early in their life stage. Areas that need improvement in plant reintroduction research were identified, including improved evaluation of extant populations, identification of threats to extant populations, more detailed monitoring of transplants, and more long-term monitoring of transplants after reintroduction.

Pteridophyte Reintroduction Success Stories

Although there are very few examples of pteridophyte reintroduction in the literature, and even fewer that were performed in the context of an organized scientific experiment, some potential successes in this niche area of research have been identified.

Case Studies from the Global Re-introduction Series

The IUCN’s Reintroduction Specialist Group releases volumes of reports of reintroduction case studies with summaries for dozens of projects in each volume. A review of all five volumes of the Global Re-introduction Series, released between 2008 and 2016, contained a total of three pteridophyte reintroduction case studies. The authors of each case study were asked to classify the success of each project as highly successful, successful, partially successful, or failed. Reasons were also provided to support the determination of relative success or failure, however, there is often no statistical or scientific data to support these claims within the document itself. Therefore, an effort was made to locate any peer-reviewed articles published

6 in scientific journals that relate to each project. Brief summaries of those case studies and corresponding peer-reviewed literature are provided below.

Marsilea quadrifolia

The reintroduction of M. quadrifolia (Marsileaceae), the only example of the reintroduction of an endangered aquatic pteridophyte, was conducted in the Po River plain in

Italy between 2010 and 2012. Goals of the project included establishing viable populations in the reintroduction area and improving methods of ex situ propagation. The authors of the case study reported the reintroduction as a success, citing that transplant cover was at 50% of its starting level after one year (Orsenigo et al. 2016). They also report that, while some of the sites saw complete mortality of transplants after three years, many sub-populations were still persistent.

No data that quantify the overall percent cover or survival of transplants after three years were reported.

While a peer-reviewed article for the reintroduction of M. quadrifolia in the Po River plain could not be located, a peer-reviewed article for a previous reintroduction of M. quadrifolia in the Ebro Delta in Spain (Estrelles et al. 2001) was found. The authors of this article report that

34% of juvenile transplants took root and developed. There was no indication of how much time had passed between transplanting and surveying to obtain this percentage. While reporting of conclusive data in each case is lacking, the end results are promising for the recovery of M. quadrifolia in some areas of Italy and Spain.

Diellia pallida

Diellia pallida (Aspleniaceae) was listed as endangered by the USFWS in 1994. This fern, endemic to the island of Kaua’i, consists of only 13 mature individuals across three sites.

Reintroduction of transplants propagated in ex situ facilities began in 2007 with 30 individuals transplanted into each of seven experimental sites. Aguraiuja (2010) reports that survival was highest at two sites, providing critical habitat requirement information for transplants. The author further reports that survival rates and microhabitat conditions of the first transplanting have been analyzed and will aid in future reintroduction efforts, although no specific survival or microhabitat data was provided. The case study concludes that the reintroduction has been partially successful, citing that a better understanding of habitat requirements has been achieved and that a large scale, landscape level recovery of the species is now possible (Aguraiuja 2010).

A peer-reviewed article related to this reintroduction project could not be located.

Woodsia ilvensis

Perhaps the best currently documented pteridophyte reintroduction project is that of W. ilvensis (Woodsiaceae), an historically rare arctic-alpine fern that was reintroduced in both the

United Kingdom and Estonia. In Estonia, W. ilvensis was designated as extirpated in 2005 as it was last observed in the wild in 1977. A small-scale project to test the possibility of reintroduction in Estonia began in 1996, with 32 sporophytes transplanted between 1996 and

2001. Maintenance of the transplants after reintroduction was extremely limited. Despite limited maintenance, a total of 13 transplants were reported to have survived in 2011 (10 years after transplanting), showing that W. ilvensis has promising potential for recovery (Aguraiuja 2011).

Woodsia ilvensis was only known to exist at 14 sites in England, Scotland, and Wales when a separate reintroduction was performed between 1999 and 2004 in the United Kingdom.

A total of 299 sporophytes were transplanted across four sites, two each in England and

Scotland, respectively. After six years of monitoring, percent survival was as high as 88% at one

8 site in England and 49% at one site in Scotland (McHaffie 2006). Mean number of fronds and mean length of longest fronds increased or remained relatively stable at most of the sites after six years. Furthermore, the proportion of surviving transplants producing spores was documented at

68% to 91% after six years across the sites. The impressive survival, growth, and transplant spore production rates offer perhaps the best recorded success of a pteridophyte reintroduction project to date.

Study Species

Varietal Status and Morphology

Asplenium scolopendrium var. americanum (Fern.) Kartesz & Gandhi (syn. Phyllitis scolopendrium var. americana; American hart's-tongue fern; AHTF; Aspleniaceae) is a rare

North American evergreen fern. It was first distinguished from its European counterpart,

Asplenium scolopendrium var. scolopendrium (European hart’s-tongue fern; EHTF), as a unique variety by Fernald (1935) who noted several significant morphological differences between the mature sporophytes. Britton (1953) provided further evidence of distinction between the varieties when he discovered that AHTF is tetraploid compared to the diploid var. scolopendrium, although, only two AHTF sporophytes from Ontario, Canada were analyzed in this report. Other reports of A. scolopendrium include var. lindenii (Hooker), native to Mexico and Hispaniola

(Wagner et al. 1993), and a report from Japan (Sato 1982). However, the varietal status of reports of A. scolopendrium in Mexico, Hispaniola, and Japan has not yet been confirmed.

Sporophytes of AHTF consist of a cluster of simple, strap-like fronds with an auriculate or cordate base emanating from a short, stout rhizome covered in cinnamon-colored scales

(Figure 1.2) (Lellinger 1985). Fronds can grow up to 65 cm in length and 7.5 cm in width with a stipe that ranges from 3 to 20 cm in length with cinnamon-colored scales (DD Fernando, unpublished data). Linear sori are slightly diagonal in relation to the rachis along the abaxial surface of the blade and may span more than half way down the blade, originating from the acute distal tip of the frond. The appearance of the sori on the blade is thought to resemble the many legs of a centipede, as the specific epithet scolopendrium, of Greek origin for “centipede,” suggests. The sporophyte produces the bulk of its new fronds at the beginning of the growing season between early May and mid-June, although in New York, new frond development may continue into the autumn season (personal observation).

Figure 1.2. An individual mature sporophyte of Asplenium scolopendrium var. americanum in September 2017 with fertile fronds developed during the current growing season (erect) and fertile fronds from the previous growing season (flattened). Photo: Michael Serviss.

AHTF is homosporous, giving rise to both unisexual and bisexual gametophytes. Spore dispersal for first season fronds generally starts in late August, although older mature fronds may release spores throughout the year (personal observation). Upon spore germination, early protonemal development follows the Vittaria-type and subsequent prothallial development is consistent with the Aspidium-type as outlined by Nayar and Kaur (1971) and confirmed by Testo and Watkins (2011). A single proximal rhizoid generally develops concurrently with the initial protonemal cell (Testo & Watkins 2011). The prothallial plate typically forms when protonemata are four to eight cells long (Nayar & Kaur 1971; Testo & Watkins 2011) and can begin forming within 14 days of sowing spores in soil under semi-in vitro conditions (Figure 1.3A) (personal observation). Gametophytes may begin to transition into the heart-shaped stage of development within 30 days (Figure 1.3B) and begin to form archegonia (Figure 1.3C) and antheridia (Figure

1.3D) between 45 and 60 days after sowing (personal observation). Sporophyte production may begin to occur between 90 and 120 days (Figure 1.3E), although sporophytes may be produced several months or more after maturation of the gametophyte (personal observation).

A B C

D E

Figure 1.3. The stages of early development of Asplenium scolopendrium var. americanum. A. Protonema at 14 days after sowing (400X). B. Early heart-shaped stage at 30 days after sowing (100X). C. Archegonia at 60 days after sowing (400X). D. Antheridia at 60 days after sowing (400X). E. Mature gametophyte with young sporeling at 120 days after sowing (10X). Photos: Michael Serviss.

Ecology and Distribution

Unlike EHTF, which is widespread throughout Europe and very common in the United

Kingdom and Ireland, distribution of AHTF in North America is limited to typically small, disjointed populations in the United States and Ontario, Canada. The most recent official review of the fern’s distribution in the United States by the U.S. Fish and Wildlife Service (USFWS) indicates that the fern has one extant population each in Alabama and Tennessee, eleven in

Michigan and sixteen in New York (USFWS 2012). I personally confirmed the persistence of the last known extant population in Tennessee in March 2016, although no mature sporophytes were located during a survey of the population by myself and a botanist from the Tennessee

Department of Environment and Conservation. Fifteen total plants were located during the survey, including 12 sporelings (fronds < 2.5 cm in length) and three immature sporophytes

(fronds > 2.5 cm in length, no visible sori). The persistence of the last extant population in

Alabama was also confirmed in March 2016 by USFWS biologist John Wiley (personal communication). There is great concern for AHTF populations in New York that consist of fewer than 25 total plants as of the last census in 2011-2012, including Evergreen Lake 1 (21),

Glacier Lake (19), Perryville Falls (16), Munnsville 1 (10), and Rams Gulch (8). Historical census data indicates a trend of decreasing population size over time for most of these occurrences (Brumbelow 2014). The persistence of populations at Evergreen Lake 1, Evergreen

Lake 2, and Ram’s Gulch were all confirmed in June 2017 by Drs. Leopold and Fernando of

SUNY-ESF, Joshua Weber-Townsend of the Fernando lab at SUNY-ESF, and myself.

Some habitat features are consistent throughout the range of AHTF. Populations are always found in predominantly mesic deciduous forest types with high occurrence of dolomitic limestone outcroppings, contributing to elevated calcium and magnesium concentrations in the soils (Futyma 1980; Cinquemani Kuehn & Leopold 1993). Differences in AHTF habitat, however, tend to align regionally. In southern Tennessee and northern Alabama, the southernmost portion of AHTF’s range, the fern is associated with cool, shady cave entrances and limestone sinkholes (Short 1979; personal observation). In Michigan, AHTF is associated with steeply sloped boulder fields where the bedrock is exposed at the surface under hardwood canopies and is almost exclusively found growing from cracks in or from the bases of moss covered boulders (Futyma 1980). In Ontario, Canada, AHTF is closely associated with the rocky outcroppings of the Niagara escarpment, specifically on the talus, but also regularly occurs in more level rocky woodlands, cliff edges, and in the seams of rocky outcroppings (Soper 1954).

In New York, AHTF typically occurs on the steep slopes of glacial ravines and plunge basins

13 along a portion of the Niagara escarpment in Onondaga and Madison counties (Cinquemani

Kuehn & Leopold 1993).

AHTF habitat associations have been particularly well studied in New York. Cinquemani

Kuehn and Leopold (1993) report that the steep talus of glacial ravines and plunge basins on which AHTF almost exclusively grows are typically north to northeast in aspect with a mean slope of about 60º. Mature sporophytes can be found just outside the top of the talus slopes and into the basin bottoms in some cases, although most individuals occur at mid-slope positions

(Figure 1.4). Soil pH is circumneutral and mean organic matter is around 55%. Sporelings are positively correlated with higher light intensity and moss cover. These correlations suggest that

AHTF sporelings and mosses may be responding independently to favorable high moisture conditions rather than mosses facilitating sporeling survival and growth. Mature sporophytes have no significant correlation with light intensity and are negatively associated with moss cover.

It is thought that rock crevices, either bare or humus filled, provide the preferred microsite conditions for AHTF sporophytes as over 80% of all mature sporophytes sampled occurred within such crevices (Cinquemani Kuehn & Leopold 1993).

Figure 1.4. A portion of one population of Asplenium scolopendrium var. americanum growing on a talus slope in September 2017 at Clark Reservation State Park, Jamesville, NY. Photo: Michael Serviss.

Climatic and other environmental sensitivities have been reported as potentially limiting factors to AHTF population vigor and growth. It has long been observed that the plunge basins in which AHTF occur in central New York are cooler in temperature than the surrounding areas

(Maxon 1900; Faust 1960; Brumbelow 2014). Early observations of temperature and humidity along the slope of a plunge basin in central New York, where AHTF was noted as present, reported that temperature dropped and humidity increased from the rim of the basin to the basin bottom (Petry 1918). Petry also reported that AHTF was found at only one of four sampling areas, approximately 50 feet down from the rim of the basin. More recent analyses into the relationship between AHTF distribution and vigor in central New York have further detailed the climatic sensitivity of the species. Kelsall et al. (2004) reported that there are positive correlations between mature AHTF vigor with spring precipitation and cold days, respectively.

The authors also report negative correlations with very wet and very dry months. The negative correlation between AHTF vigor and very dry months helps to support the assertion that AHTF is intolerant to drought as has long been suspected (Cinquemani et al. 1988). Brumbelow (2014) found that annual mean temperature was positively correlated with the presence of mature AHTF and that annual mean daily minimum temperature was positively correlated with the presence of all AHTF sporophyte life stages (sporeling, immature, and mature). He concluded that some microclimate factors are a limiting factor for AHTF distribution, in part. The findings of

Brumbelow (2014) help to explain why AHTF sporophytes are extremely rarely distributed outside of glacial ravines and plunge basins in central New York (personal observation).

Legal Status

At the federal level, AHTF has been listed as threatened by the USFWS since 1989 under the Endangered Species Act of 1973. The goal of the original 1993 recovery plan for AHTF is to recover the species sufficiently enough to delist the species from the Endangered Species Act of

1973 (USFWS 1993). The main criterion for delisting the species is the protection of at least fifteen self-sustaining populations in the United States. These fifteen populations include two each in Alabama and Tennessee, four in Michigan, and seven in New York. While the recovery plan is unclear as to what constitutes a self-sustaining population or to what level the populations should be protected, many recovery actions are outlined to achieve this goal. These actions include protecting the populations (presumably by protecting the habitat), regular census of populations, conducting biological studies to fill knowledge gaps about AHTF genetics and reproductive biology, implementing management set forth by the plan, and reestablishing populations. In New York State, the New York Natural Heritage Program ranks AHTF as S2, meaning that it is imperiled due to rarity as demonstrated by a limited number of populations (6-

20) or vulnerable to extirpation due to biological or anthropogenic factors.

Goals and Objectives

The main goal of this research is to examine the possibility of successful reintroduction of AHTF in accordance with the recovery plan for the species. Reintroduction may be an important recovery method of AHTF because, throughout the range of AHTF, many populations suffer from very small population sizes that are susceptible to local extinction due to stochastic events (Cinquemani et al. 1988). Most populations of AHTF are severely isolated and exhibit depauperate genetic diversity that may contribute to inbreeding depression (Fernando et al. 2015;

Weber-Townsend 2017). These populations may benefit from augmentation with a mix of genetically diverse transplants (Fernando et al. 2015). Numerous AHTF populations face the threat of competition from invasive plant species which may render their critical habitat unsuitable (Brumbelow 2014). In this case, reintroduction of ex situ propagated individuals into more suitable habitat may be the best option to ensure that the number of self-sustaining populations does not fall below critical recovery levels cited in the recovery plan for the species.

Several objectives are investigated herein to achieve these overall goals:

1. Identify at least three suitable reintroduction sites with habitat characteristics that are

associated with AHTF occurrences following Cinquemani Kuehn and Leopold (1993).

2. Transplant at least 500 laboratory-propagated AHTF of five transplant-types at each

reintroduction site.

3. Determine the effect of transplant-type, vigor, and acclimatization method on the survival

and growth of transplants within reintroduction sites.

4. Characterize the microclimate conditions at reintroduction sites, including temperature,

relative humidity, solar radiation, and wind speed and direction.

Study Area

The three selected reintroduction sites are within Onondaga County in the State of New

York (Figure 1.5). Two of the reintroduction sites are within Clark Reservation State Park in

Jamesville, NY and one is within the New York State Department of Environmental

Conservation (NYS-DEC) Split Rock Unique Area in the Town of Onondaga, NY. The disclosure of the exact location of each AHTF reintroduction site is prohibited due to the protections provided by the legal status of the species at the New York State and federal levels.

Clark Reservation State Park is managed by the New York State Office of Parks,

Recreation and Historic Preservation (OPRHP) in Jamesville, Onondaga County. The 153 ha park offers great potential for reintroduction, introduction, and augmentation as it allows for the best possibility of the facilitation of gene flow between AHTF populations due to the density of the six extant populations within the park. Two sites within Clark Reservation State Park

(Sentinel Basin and Twin Basins) were chosen for the transplanting of laboratory propagated

AHTF.

Sentinel Basin

Located along the southeastern boundary of Clark Reservation State Park, the Sentinel

Basin reintroduction site hosts one of the largest historical populations of AHTF throughout its

U.S. range (Cinquemani Kuehn & Leopold 1992; Brumbelow 2014). Habitat characteristics of the site are typical of that of AHTF in New York, consisting of a north to northeast facing slope along a glacial plunge basin with a mostly hardwood canopy. The genetic significance of the extant population was a key factor in the selection of this site. This population represents the most genetically diverse of eight populations sampled in central New York by Fernando et al.

(2015) and of 18 populations sampled range-wide by Weber-Townsend (2017). Sentinel Basin also represents the second largest population in New York by mature individuals (779) and total individuals (1191) (Brumbelow 2014). Another significant factor in the selection of this site was the abundance of available transplants produced through spore culture of individuals sourced from Sentinel Basin.

Twin Basins

The Twin Basins reintroduction site is located just to the northeast of Glacier Lake at

Clark Reservation State Park. No known AHTF occurrence has been recorded at this site throughout the long history of surveys for AHTF within the park. This site consists of two glacial plunge basins separated by a raised natural barrier that runs roughly north to south. The steep north facing slope of the western basin was chosen as an ideal transplant site due to its similarity to typical AHTF habitat and its central location among AHTF populations within the park. The site is located approximately 320 m from the nearest AHTF population at Lower Basin and is within 800 m of the Sentinel Basin population, the farthest population within the state park. All four additional AHTF populations within the park are within 500 m of this site, including populations at Long Ravine, Grand Canyon, Pulpit Rock, and Glacier Lake.

Split Rock Unique Area

The NYS-DEC Split Rock Unique Area is located in the Town of Onondaga, Onondaga

County. The rich and varied history of the site, along with availability of suitable habitat, were significant factors in the selection for augmentation of its extant population. AHTF was first discovered in North America by Frederick Pursh on July 20, 1807 at this location. It was thought that the population at this site was extirpated after a clearing of the forest in the area for a

19 limestone quarry until it was rediscovered by the Syracuse Botanical Club on September 30,

1879 (Rust 1879). Faust (1960) recounts that, since its rediscovery, trees had been cut in the area, quarrying operations had damaged or destroyed many of the rocks that shelter the population, and a massive explosion from a WWI munitions operation occurred nearby. Today, large portions of the 13 ha unique area remain intact, although, evidence of past industrial activity and dumping can be observed in areas close to the AHTF population. Furthermore, a vast covering of invasive species, including Lonicera spp., Alliaria petiolata and Cynanchum rossicum threaten large areas of the site and are currently unmanaged.

Figure 1.5. General locations of the three selected AHTF reintroduction sites. Sites at Twin Basins and Sentinel Basin were selected at Clark Reservation State Park (bottom inset) and one site was selected at Split Rock Unique Area (top inset).

Literature cited

Allen WH. 1994. Reintroduction of endangered plants. BioScience 44(2): 65-68.

Aguraiuja R, 2010. Reinforcement of the populations of critically endangered endemic fern Diellia pallida, Kaua’i, Hawaiian Islands, USA. Global Re-introduction Perspectives: Additional case-studies from around the globe p. 319. Aguraiuja R. 2011. Reintroduction of the endangered fern species Woodsia ilvensis to Estonia: a long-term pilot study. Biodiversity and Conservation 20(2): 391-400. Blackmore S, Walter K. 2007. Assessing the conservation status of pteridophytes, a challenge for the Global Strategy for Plant Conservation. Fern Gazette 18(2): 71-76. Britton DM. 1953. Chromosome studies on ferns. American Journal of Botany 40(8): 575-583.

Brumbelow TR. 2014. Population and microclimate studies of the American hart's-tongue fern (Asplenium scolopendrium var. americanum (fern.) Kartesz & Gandhi) in central New York. Thesis, State University of New York College of Environmental Science and Forestry. Brummitt NA, Bachman SP, Griffiths-Lee J, Lutz M, Moat JF, Farjon A, Donaldson JS, Hilton- Taylor C, Meagher TR, Albuquerque S, Aletrari E. 2015. Green plants in the red: A baseline global assessment for the IUCN sampled Red List Index for plants. PloS One 10(8): e0135152. doi:10.1371/journal.pone.0135152. Brummitt N, Aletrari E, Syfert MM, Mulligan M. 2016. Where are threatened ferns found? Global conservation priorities for pteridophytes. Journal of Systematics and Evolution 54(6): 604-616. Cavender N, Westwood M, Bechtoldt C, Donnelly G, Oldfield S, Gardner M, Rae D, McNamara W. 2015. Strengthening the conservation value of ex situ tree collections. Oryx 49(3): 416-424. Cinquemani DM, Faust ME, Leopold DJ. 1988. Periodic censuses (1916-1986) of Phyllitis scolopendrium var. americana in central New York State. American Fern Journal 78(2): 37-43.

Cinquemani Kuehn DM, Leopold DJ. 1992. Long-term demography of Phyllitis scolopendrium (L.) Newm. var. americana Fern. in central New York. Bulletin of the Torrey Botanical Club 119(1): 65-76.

Cinquemani Kuehn DM, Leopold DJ. 1993. Habitat characteristics associated with Phyllitis scolopendrium (L.) Newm. var. americana Fern.(Aspleniaceae) in central New York. Bulletin of the Torrey Botanical Club 120(3): 310-318.

Corlett RT. 2016. Plant diversity in a changing world: status, trends, and conservation needs. Plant Diversity 38(1): 10-16.

Dalrymple SE, Banks E, Stewart GB, Pullin AS. 2012. A meta-analysis of threatened plant reintroductions from across the globe. In Plant reintroduction in a changing climate (pp. 31-50). Island Press/Center for Resource Economics, Washington, DC. Estrelles E, Ibars AM, Iranzo J, Morales F. 2001. Recuperación y reintroducción de Marsilea quadrifolia L. en los arrozales del delta del Ebro (Tarragona, España). Botanica Complutensis 25: 251-259.

Faust ME. 1960. Survival of hart's-tongue fern in central New York. American Fern Journal 50(1): 55-62.

Fernald ML. 1936. Critical plants of the upper Great Lakes region of Ontario and Michigan. Indian Forester 62(4): 233-233. Fernando DD, Discenza JJ, Bouchard JR, Leopold DJ. 2015. Genetic analysis of the threatened American hart's-tongue fern (Asplenium scolopendrium var. americanum [Fernald] Kartesz and Gandhi): Insights into its mating system and implications for conservation. Biochemical Systematics and Ecology 62(2), 25-35.

Futyma RP. 1980. The distribution and ecology of Phyllitis scolopendrium in Michigan. American Fern Journal 70(3): 81-87. Gasper ALD, Eisenlohr PV, Salino A. 2016. Improving collection efforts to avoid loss of biodiversity: lessons from comprehensive sampling of lycophytes and ferns in the subtropical Atlantic Forest. Acta Botanica Brasilica 30(2): 166-175.

Given DR. 1993. Changing aspects of endemism and endangerment in Pteridophyta. Journal of Biogeography 20(3): 293-302. Godefroid S, Piazza C, Rossi G, Buord S, Stevens AD, Aguraiuja R, Cowell C, Weekley CW, Vogg G, Iriondo JM, Johnson I. 2011. How successful are plant species reintroductions?. Biological Conservation 144(2): 672-682.

Guerrant Jr EO. 2012. Characterizing two decades of rare plant reintroductions. In Plant Reintroduction in a Changing Climate (pp. 9-29). Island Press/Center for Resource Economics, Washington, DC.

Kaye TN. 2017. Working toward the same goal: reintroducing rare plants in support of wild population conservation. Douglasia 41: 4-8. Kelsall N, Hazard C, Leopold DJ. 2004. Influence of climate factors on demographic changes in the New York populations of the federally-listed Phyllitis scolopendrium (L.) Newm. var. americana. Journal of the Torrey Botanical Society 131(2): 161-168.

Kessler M. 2001. Pteridophyte species richness in Andean forests in Bolivia. Biodiversity & Conservation 10(9): 1473-1495.

Kessler M, Lehnert M. 2009. Do ridge habitats contribute to pteridophyte diversity in tropical montane forests? A case study from southeastern Ecuador. Journal of Plant Research 122(4): 421-428. Kreft H, Jetz W, Mutke J, Barthlott W. 2010. Contrasting environmental and regional effects on global Pteridophyte and seed plant diversity. Ecography 33(2): 408-419. Lellinger DB. 1985. A field manual of the ferns and fern-allies of the United States and Canada (Volume 3). 389 pages. Smithsonian Institution Press, Washington, DC. Maschinski J, Haskins KE. 2012. Introduction. In Plant reintroduction in a changing climate: promises and perils. (pp. 1-6) Island Press/Center for Resource Economics, Washington, DC. Maxon WR. 1900. On the occurrence of the hart's-tongue in America. Fernwort Papers 30-46. McHaffie H. 2006. A reintroduction programme for Woodsia ilvensis (L.) R. Br. in Britain. Botanical Journal of Scotland 58(1): 75-80. Nayar BK, Kaur S. 1971. Gametophytes of homosporous ferns. The Botanical Review 37(3): 295-396. Oldfield S, Newton AC. 2012. Integrated conservation of tree species by botanic gardens: A reference manual. 54 pages. Botanic Gardens Conservation International, Richmond, Virginia. Orsenigo S, Cauzzi P, Gentili R, Rossi G, Abeli T. 2016. Re-introduction of the four leaf clover in the agricultural context of the Po River Plain, Italy. Global Re-introduction Perspectives: 2016. Case-studies from around the globe pp. 238-241. Paciencia MLB, Prado J. 2005. Effects of forest fragmentation on pteridophyte diversity in a tropical rain forest in Brazil. Plant Ecology 180(1): 87-104. Petry LC. 1918. Studies on the vegetation of New York State-II. The vegetation of a glacial plunge basin and its relation to temperature. Bulletin of the Torrey Botanical Club 45(5): 203- 210. Pritchard DJ, Fa JE, Oldfield S, Harrop SR. 2012. Bring the captive closer to the wild: redefining the role of ex situ conservation. Oryx 46(1): 18-23. Rust MO. 1879. Pursh's station for Scolopendrium rediscovered by the Syracuse Botanical Club. Bulletin of the Torrey Botanical Club 6(57): 345-347. Sato T. 1982. Phenology and wintering capacity of sporophytes and gametophytes of ferns native to northern Japan. Oecologia 55(1): 53-61. Short JW. 1979. Phyllitis scolopendrium newly discovered in Alabama. American Fern Journal 69(2): 47-48. Soper JH. 1954. The hart's-tongue fern in Ontario. American Fern Journal 44(4): 129-147.

Testo WL, Watkins Jr JE. 2011. Comparative development and gametophyte morphology of the hart's-tongue fern, Asplenium scolopendrium L. The Journal of the Torrey Botanical Society 138(4): 400-408. Tuomisto H, Poulsen AD. 2000. Pteridophyte diversity and species composition in four Amazonian rain forests. Journal of Vegetation Science 11(3): 383-396. Wagner WH, Moran RC, Werth CR. 1993. Aspleniaceae Newman: spleenwort family. Flora of North America North of Mexico 2: 228-245. USFWS (United States Fish and Wildlife Service). 1993. Recovery plan for American hart's- tongue fern (Asplenium scolopendrium L. var. americanum [Fernald] Kartezs and Gandhi [Synonym: Phyllitis scolopendrium (L.) Newman var. americana Fernald]). U.S. Fish and Wildlife Service, Ashville, NC.

USFWS (United States Fish and Wildlife Service). 2012. American Hart's-tongue fern (Asplenium scolopendrium var. americanum) 5 year review: summary and evaluation. U.S. Fish and Wildlife Service Ecological Services Field Office Cookeville, TN.

Weber-Townsend JR. 2017. Contributions of genetic data to the conservation and management of the threatened American hart's-tongue fern (Asplenium Scolopendrium var. americanum). Thesis, State University of New York College of Environmental Science and Forestry.

Young KR, León B. 1989. Pteridophyte species diversity in the central Peruvian Amazon: importance of edaphic specialization. Brittonia 41(4): 388-395.

Chapter 2: Experimental Reintroduction of Asplenium scolopendrium var. americanum:

Factors Affecting Survival and Growth

Introduction

History of Translocation and Reintroduction of AHTF in New York

The documented history of translocation and reintroduction of the American hart’s- tongue fern (AHTF) extends as far back as the year 1900, when Dr. Homer House translocated

AHTF from Chittenango Falls to a ravine near Munnsville, NY (Hunter 1924; Faust 1960).

These transplants persisted until at least 1920, but could not be relocated in a survey in 1946

(Faust 1960). In the mid-1920’s, about 700 AHTF were translocated into Clark Reservation State

Park after being rescued from quarry operations around the area of White Lake and Green Pond

(“Scolopendrium Pond” to some) that has since been quarried (House et al. 1926; Faust 1960).

Unfortunately, many transplants were washed out by heavy rains shortly after translocation

(Faust 1960) and details on the survival of the AHTF transplants could not be found.

Ralph Benedict, of the Brooklyn Botanic Garden, started propagating AHTF shortly after the loss of the populations of AHTF due to the quarry operations around White Lake and Green

Pond (Benedict 1927). He distributed many immature plants across the continental United States with the intent of naturalizing AHTF over a broader geographic range. In 1929 he reported that, of the few updates he received, most of the introduced AHTF did not survive longer than one winter season (Benedict 1929). However, one batch was reported as surviving in a home garden in Merion, PA. No other updates on the fate of these transplants could be located.

No further documented attempts at translocation or reintroduction of AHTF were noted again until 1992, when researchers at SUNY-ESF in Syracuse, NY planned to augment populations of AHTF at Clark Reservation State Park and Split Rock Unique Area. According to the research proposal, 50-100 sporelings were propagated at an off-site facility and were transplanted in May 1993 (Moore & Leopold 1992). Shortly after transplanting at Split Rock

Unique Area, a passer-by removed flagging critical to locating and tracking the augmentation site, causing the project to ultimately be abandoned as the transplants could not be relocated (DJ

Leopold, personal communication). The unfortunate conclusion to this project left researchers interested in the feasibility of reintroduction of AHTF with little more than anecdotal information regarding the survival and growth of transplants over the past one hundred years.

General Reintroduction Considerations

Reintroduction of rare and threatened plants has become a common and increasingly well-studied method of conserving plant diversity and recovering imperiled species. The decision to execute a plant reintroduction project, however, requires justification and careful planning. In the case of AHTF, reintroduction is a recovery objective of the species management plan

(USFWS 1993) and recommended as a means of recovery of AHTF in the most recent review of the status of the overall project (USFWS 2012). Despite a great deal of work to recover the species, including in situ conservation, microclimate analysis of habitat, periodic censuses, ex situ propagation, and genetic analyses, reintroduction of AHTF has yet to be definitively addressed. A major factor in the decision to reintroduce AHTF is that populations are isolated across its rather fragmented range and often exhibit dramatic fluctuations in population size that have historically reduced individual populations by as much as 93% in New York (Cinquemani et al. 1988). AHTF populations are typically lacking in genetic diversity throughout the species’

26 range and there is evidence of very low gene flow, if any, between populations (Fernando et al.

2015; Weber-Townsend 2017). Therefore, reintroduction of AHTF in strategic geographic areas with proximity to one or more extant populations may be used to promote genetic connectivity.

If placed very close to an extant population, reintroduction may not only promote genetic diversity, but also augment small population sizes and potentially buffer against dramatic decreases in population size. Finally, reintroduction of AHTF can, theoretically, serve to establish new populations where in situ conservation of extant populations is not practical. Some current populations are heavily impacted by the spread of unmanaged invasive plant species and may be impacted by their proximity to developed lands (Brumbelow 2014). Protection of these populations with in situ conservation methods may not be practical or effective.

Site selection is among the critical considerations in the reintroduction of many rare and threatened plant species (Fiedler & Laven 1996; Guerrant Jr. & Kaye 2007; Kaye 2008; Guerrant

Jr. 2012). Maschinski et al. (2012) states that habitat requirements for rare plant species are not likely to be common, and therefore selection of suitable microhabitat with appropriate niche characteristics is recommended. Reintroduction with the consideration of different, but appropriate, niche habitats was explored in the reintroduction of the locally extinct fern Woodsia ilvensis in Estonia (Aguraiuja 2011). While only 40 transplants of W. ilvensis were deployed in this study, it was found that reintroduction of the species was successful on stone walls and not on boulders. In a separate reintroduction of W. ilvensis in the United Kingdom (UK), sites where the fern was historically documented but had since been thought extirpated were targeted because soil samples from the site contained W. ilvensis spores and subsequently promoted spore germination (McHaffie 2006). The historical occurrence of W. ilvensis, coupled with evidence of appropriate soils for the reproductive needs of the species at the UK sites, indicated that the

27 habitat at the sites was likely to support W. ilvensis transplants. Larger-scale considerations regarding site selection are equally important as niche or microsite scale considerations. The use of reintroduction sites within protected areas, for example, is likely to increase successful establishment of transplants (Godefroid et al. 2011). Species distribution modeling has also predicted that large-scale loss of critical habitat attributed to climate change, inclusive of potential reintroduction sites, could lead to the extinction of over half of the threatened and endangered plant species on the Colorado Plateau (Krause & Pennington 2012). The potential effects of climate change on the distribution of a species further highlights the need to carefully identify and select reintroduction sites based on the ecological requirements of the target species, especially for climatically sensitive species such as AHTF.

Genetic, source population, and demographic considerations are also of great importance to the planning of a plant reintroduction. Reintroduction should aim to maintain or increase genetic diversity, and therefore evolutionary potential, within a species as well as promote gene flow between populations (Weeks et al. 2011). Ideally, transplants should exhibit similar ecotype characteristics and similar or higher levels of genetic diversity as the source populations (Kaye

2008; Menges 2008). Analysis of genetic diversity of a species can aid in the determination of which source populations to include in a reintroduction and are often more informative than selecting source populations based on population size alone (Gravuer et al. 2005). The geographic distance of the source populations in relation to the reintroduction sites are also important considerations because the negative effects of outbreeding depression are of concern when non-locally adapted genotypes are used in reintroduction or augmentation projects

(Hufford & Manzer 2003; McKay et al. 2005). Generally, the selection of source populations closer to reintroduction sites is desirable in reducing the risk of transplanting genotypes that are

28 not well-adapted to the reintroduction site (McKay et al. 2005). These issues can likely be avoided by using a mix of plant material from many source populations of locally adapted genotypes (McKay et al. 2005). Using a mix of transplants from different source populations has also shown to greatly increase the probability of success of plant reintroductions, especially when those populations are stable or increasing in population size (Godefroid et al. 2011).

Recent genetic analyses of eight populations of AHTF in New York report that they are differentiated enough to be considered distinct populations (Fernando et al. 2015; Weber-

Townsend 2017). However, there is little evidence that non-locally adapted genotypes exist due to the similarity of the habitat between populations; which occur exclusively on the steep talus of glacial ravines and plunge basins in New York. Furthermore, it is suggested that AHTF transplants produced by spore culture from nearby populations with higher levels of genetic diversity be used in reintroduction projects to improve fitness and reproductive success of transplants (Fernando et al. 2015).

The decision to use certain life stages of transplants in plant reintroduction is another critical factor in achieving success. However, few plant reintroductions have examined life history stage as a variable since 78% of projects were determined to have used a single life stage in a recent characterization of reintroductions (Guerrant Jr. 2012). Several studies have suggested that the use of more advanced life stages leads to higher survival success in plant reintroductions

(Drayton & Primack 2000; Jusaitis et al. 2004; Maschinski & Duquesnel 2007; Godefroid et al.

2011; Albrecht & Maschinski 2012; Dalrymple et al. 2012). Using more advanced life stages is beneficial to the goal of successfully establishing transplants and achieving quicker recruitment of new generations for species that take many years to reach reproductive maturity. In the case of fern reintroductions, the use of life stages preceding that of sporophytes is not well known and

29 little evidence exists that gametophyte, protonema, and spore life stages have been utilized in any fern reintroductions. However, the use of gametophytes as transplant material has been suggested as an area of interest in at least one previous fern reintroduction project (Aguraiuja 2011). In

2015, a pilot study for the AHTF reintroduction project assessed the possibility of using AHTF spores as transplant material with unsuccessful results (Fernando & Serviss, unpublished results).

While using more advanced life stages may lead to greater success, there is also increased cost and difficulty associated with years of propagation and maintenance, particularly for rare and threatened plants. If less advanced transplants can be successfully established, significant amounts of time, money, and resources can be saved and diverted to other important conservation tasks related to species recovery.

Acclimatization is defined as the habituation of a plant’s physiological response to environmental conditions (Begon et al. 1990) and constitutes another critical, yet understudied, factor in plant reintroduction. While there are many studies that analyzed methods of acclimatization of in vitro micropropagated plants of economic importance to controlled lab or greenhouse conditions (see Preece & Sutter 1991; Pospóšilová et al. 1999; Dewir et al. 2015), there is very little research that focused on acclimatization of plants grown in controlled laboratory or greenhouse conditions to less forgiving environmental conditions prior to reintroduction (see Brito et al. 2009; Yong et al. 2015). Plants propagated in vitro tend to have issues with stomatal functionality (Preece & West 2006; Khan et al. 2009) and thinner cuticles

(Wetzstein & Sommer 1982; Brutti et al. 2002) compared to greenhouse and field acclimatized plants. The anatomical differences between greenhouse acclimatized plants and field acclimatized plants make it potentially more difficult to adapt to environmental conditions at the reintroduction site. In AHTF, excessive waviness of the margins, thinness, and a dull light green

30 to gray color of the fronds of plants propagated in a controlled lab environment might be indicative of insufficient cuticle and palisade tissue formation that could negatively impact survival post-reintroduction (DD Fernando, personal communication). The acclimatization of transplants of AHTF to more natural environmental conditions is an area of research that warrants investigation.

Objectives

The first objective of this chapter is to identify at least three reintroduction sites with habitat characteristics that are associated with AHTF occurrences following Cinquemani Kuehn and Leopold (1993). Reintroduction sites should be located as such to promote genetic connectivity between extant AHTF populations and/or promote increased genetic diversity within an extant population via augmentation according to recommendations for the species

(Fernando et al. 2015). Additionally, reintroduction sites should exist within the boundaries of protected areas where land use management promotes the protection of each site, including associated habitat characteristics. The historical importance of the site, as it relates to AHTF, should also be considered. The second objective is to transplant at least 500 laboratory- propagated AHTF of various life stages at each reintroduction site. The third objective is to determine the effects of life stage, vigor at time of transplanting, and acclimatization method on the survival and growth of transplants.

Hypotheses

1. More advanced life stages will exhibit significantly higher percent survival and growth

than less advanced life stages.

2. Transplants treated with both indoor and outdoor acclimatization methods will exhibit

significantly higher percent survival than those treated with only indoor acclimatization

methods.

3. Transplants with better vigor at the time of transplanting will be significantly more likely

to survive longer than those with less vigor.

4. Transplants of larger size, as measured by number of fronds and frond length, will be

significantly more likely to survive longer than smaller transplants.

Methods

Propagation

All AHTF transplants were propagated through spore culture (Fernando, unpublished protocol). AHTF spores were collected, with permits, from six different extant populations between Madison and Onondaga counties, NY, including four populations at Clark Reservation

State Park (Jamesville, NY), one at Chittenango Falls State Park (Cazenovia, NY), and one at

Split Rock Unique Area (Onondaga, NY). Spores were germinated in Petri dishes half filled with autoclaved potting soil and exposed to a continuous light regime under fluorescent growing lights at approximately 23-25 °C. They were given one to two mL of distilled water per week applied with a pipette at the edges of the dish so as not to submerge the germinating spores in the soil. Upon maturation of gametophytes, the light regime was changed to 16 hours of light and

32 eight hours of dark per day. The Petri dishes were then flooded with distilled water to facilitate the fertilization of eggs and discourage intragametophytic selfing. The cover of each Petri dish was removed after sporelings were approximately one cm tall. The dishes were placed into larger growth chambers and distilled water was applied as needed to keep the soil moist. Within one to two months after sporeling production, sporelings were transplanted individually into small pots and allowed to develop into immature sporophytes in a growth chamber. Protonema, gametophyte, and sporeling life stages were propagated in the same manner as immature sporophyte life stages, although the timing of their propagation was such that they were transplanted at 20 days (protonemata), 90 days (gametophytes), and 6 to 12 months (sporelings) after sowing spores, respectively.

Acclimatization

123 potted immature sporophytes were placed in a cold room during the winter months

(December to March) set to 4 °C and they were subjected to a light regime of six hours of light and eighteen hours of dark to simulate outdoor environmental conditions. All immature sporophytes were exposed to these conditions for one to two winter seasons prior to field transplanting. The 123 immature sporophytes treated with only indoor acclimatization were labeled as IA sporophytes. 102 immature sporophytes that received indoor acclimatization for one winter season similar to the above conditions were placed in the ground, still in pots, at a plot near the reintroduction sites at Clark Reservation State Park in April 2014. The plants were watered on an as needed basis during the growing season. The plants were grown under these conditions until they were transplanted into their respective reintroduction sites in July 2015. The

102 immature sporophytes treated with both indoor and outdoor acclimatization were labeled as

IOA sporophytes. The purpose of exposing IOA sporophytes to the natural outdoor conditions

33 was to test the effect of more rigorous acclimatization on survival and growth. Conditions at the outdoor acclimatization plot were more similar to conditions within the natural habitat of AHTF than lab or greenhouse conditions can simulate. Therefore, physiological characteristics of IOA sporophytes after acclimatization, such as frond and cuticle thickness, were likely to be better adapted to the reintroduction sites upon transplanting than less rigorously acclimatized transplants.

Reintroduction Site Selection

Reintroduction sites were identified with the assistance of an Element Distribution Model

(EDM) of suitable AHTF habitat in New York State (New York Natural Heritage Program

2012). Potential reintroduction sites outside of protected areas were identified by the EDM, but were not considered for the purposes of this project. The historic importance of the site and the likelihood of augmenting AHTF genetic diversity within a site and/or facilitating gene flow between extant populations were all considered in site selections. Candidate reintroduction sites within protected areas were visited in Spring 2015 to verify the quality and similarity of the sites to known AHTF habitat characteristics following Cinquemani Kuehn and Leopold (1993).

Ultimately, three reintroduction sites were selected and include two sites at Clark Reservation

State Park in Jamesville, NY and one site at Split Rock Unique Area in Onondaga, NY.

Descriptions of these sites were outlined in the first chapter of this thesis.

Transplanting

Transplanting of AHTF into the reintroduction sites occurred from July 10-14, 2015.

Suitable transplanting microsites were identified at each site by observation of several criteria including ground surface stability, soil depths (> 5 cm), presence of limestone cobble with at

34 least some moss cover, non-excessive leaf litter, and non-excessive crowding from native vegetation. Nine plots were chosen at each of the three sites based on availability of suitable microsites. The plots were irregular in size and shape in accordance with the highly variable terrain within each site and were used mainly as an organizational tool to help rediscover transplants during monitoring and data collection. Removal of invasive plant species was performed throughout each reintroduction site prior to transplanting.

Transplants represent five transplant-type categories based on life stage and, for immature sporophytes, acclimatization method: protonema (~20 days old), gametophytes (~90 days old, predominantly exhibiting the heart-shaped stage of development and at or nearing maturity), sporelings (sporophytes between 6 and 12 months old), immature sporophytes with indoor acclimatization only (sporophytes between 2 and 4 years old, IA sporophytes), and immature sporophytes with both indoor and outdoor acclimatization (sporophytes between 3 and

5 years old, IOA sporophytes). An effort was made to distribute each transplant-type evenly between plots, although a record keeping error influenced an uneven distribution of IOA sporophytes between reintroduction sites and plots. An effort was also made to equally distribute transplants of varying sizes and vigor ratings across all plots.

Protonemata growing on top of soil in Petri dishes were transplanted by scooping the soil out using a spatula and gently placing the soil with protonemata into a small hole in the ground that was prepared beforehand. The edges of each clump of soil were patted down to stabilize the soil as much as possible without disturbing the protonemata. Gametophytes were transplanted in a similar manner to protonemata, making effort to not disturb the attachment of the gametophyte’s rhizoids to their soil substrate. Gametophytes were transplanted as clumps, which consisted of three to five gametophytes growing close together, to promote potential fertilization

35 and sporeling production. As few as one clump to as many as 36 clumps were transplanted into each microsite. Sporelings, IA, and IOA sporophytes were removed from their pots with as much potting soil as possible, the roots were slightly loosened in the potting soil and then the transplants, along with potting soil, were placed into holes in the ground dug just prior to transplanting. A small plastic plant marker with a unique identification number was placed next to each transplant. The plastic plant markers were later replaced with an engraved metal tag affixed to a 30 cm tall PVC pipe for all remaining surviving plants in August 2017.

Post-transplanting treatment of plots included watering on an as-needed basis for no longer than two weeks after the transplant date. Watering was applied via backpack sprayers to all transplants at each site. Re-growth or establishment of any additional invasive plants was managed by removing the invasive plants in June of each growing season post-transplanting. No additional post-transplanting treatments were performed.

Monitoring and Data Collection

One day prior to transplanting, the number of fronds and length of each frond (from base of the stipe to tip of the blade) were measured and recorded for each transplant. A vigor rating was also recorded for each transplant on a scale of one to five, with one representing very poor vigor and five representing excellent vigor. Vigor rating was based largely on the appearance and color of the fronds with minimal consideration of the number and size of the fronds as transplants of all sizes varied in vigor (Table 2.1). A census of all transplants at each site was conducted every October and July post-transplanting for a total of four censuses between

October 2015 and July 2017. Survival and growth of protonemata were monitored by observing the presence or absence of their growth into more advanced stages of gametophytic development

36 that can be seen without the aid of a microscope or magnifying lens. Survival, number of fronds, frond lengths, and vigor rating were recorded during each census. A transplant was considered living upon observation of any area(s) of green tissue on any remaining frond(s). A transplant was considered dead if an absence of any area(s) of green tissue on any remaining frond(s) was observed, or, if no remaining plant tissue of any kind was observed. If a transplant could not be rediscovered, it was noted as such after a fifteen minute search of each plot for “missing” transplants. If a “missing” transplant was rediscovered in a subsequent census and observed to be living, it was updated as living for each previous census during which it was “missing”. If a transplant was never rediscovered, it was subsequently removed from all statistical analyses.

Table 2.1. Vigor ratings and respective descriptions for AHTF transplants used in experimental reintroductions.

Vigor rating Description based on frond appearance Very poor -most fronds shriveled, excessively wavy at margins -necrosis present on many fronds -dull grey to very dull green in color -fronds mostly flattened to substrate surface Poor -many fronds shriveled, excessively wavy at margins -necrosis present on some fronds -very dull green in color -some fronds flattened to substrate surface Average -a few fronds shriveled, excessively wavy at margins -no necrosis present -light green in color -most fronds upright Good -no fronds shriveled or excessively wavy at margins -no necrosis present -darker green in color, but lack waxy appearance -all fronds upright Excellent -no fronds shriveled or excessively wavy at margins -no necrosis present -darker green in color with waxy appearance -all fronds upright

Statistical Analyses

Raw percent survival at each reintroduction site was calculated by dividing the number of confirmed surviving transplants by the number of rediscovered identification tags at each site; as all transplants and/or identification tags could not be rediscovered (see Appendix A). Transplants and/or identification tags that could not be rediscovered were eliminated from the calculation of raw percent survival since the survival or death of these transplants could not be confirmed. The protonemata transplant-type was not included in any calculation of percent survival because it was too difficult to determine if they survived in a microscopic filamentous stage without

38 disturbing their attachment to the soil and therefore increasing their risk of mortality. However, an attempt was made to quantify the survival of protonemata as determined by observation of presence or absence of their growth into more advanced stages of gametophytic development that can be seen without the aid of a microscope or magnifying lens (Table 2.2).

One-way analysis of variance (ANOVA) comparing mean percent survival of each transplant-type (except protonema) was performed after each census period using Minitab®

Statistical Software (version 17.1.0). Post hoc tests with Tukey’s honest significant difference

(Tukey’s HSD) were performed to further identify the significance of transplant-type on the mean survival of transplants. Mean percent survival of each transplant-type was calculated as the summation of raw percent survival at each reintroduction site divided by the total number of reintroduction sites.

Growth of transplants over time was characterized by measures around the central tendency of transplants, including the mean and standard error of the number of fronds and total frond length, for each frond-bearing transplant-type. Mean total frond length was calculated as the summation of the total frond length per transplant divided by the number of transplants.

Frond-bearing transplant-types that were living as of July 2017 were the only transplants included in the calculation of mean number of fronds and mean total frond length for any time period.

A proportional odds model (POM) ordinal logistic regression analysis (McCullagh 1980)

(inclusive of only frond-bearing transplant-types) was performed to identify significant predictors of longer-term survival of transplants using Minitab Statistical Software. The POM of ordinal logistic regression is a useful analysis when considering a response variable with ordered

39 categorical outcomes (Brant 1990). In this analysis, the ordinal response variable is survival over time, which was categorized in a succession over time from less desirable to more desirable durations of survival: 1) Survived less than 4 months; 2) Survived up to 4 months; 3) Survived up to 12 months; 4) Survived up to 16 months; 5) Survived up to 24 months. The model was optimized by starting with a broad model of potentially significant predictors followed by the elimination of insignificant predictors until all insignificant predictors were removed. The model was further optimized using a stepwise model fitting test to confirm the inclusion of all significant predictors in the model. Transplants and/or identification tags that could not be rediscovered at any census were eliminated from the model as it could not be determined with certainty if or how long they survived.

Results

Effects of Life Stage on Survival and Growth

A total of 1,925 AHTF, representing five different transplant-types, were transplanted into three reintroduction sites within Onondaga County, New York and censused for survival over a 24-month period (Table 2.2). No growth of protonemata into the gametophyte life stage was observed during the 24-month monitoring period. The gametophyte transplant-type represented 68% of the total number of initial transplants across all sites. Gametophyte mortality

(96%) was extremely high in the first four months following transplanting. Gametophyte mortality was 100% at two sites, Sentinel Basin and Split Rock, after 12 and 16 months, respectively. At Twin Basins, a total of 7 clumps of gametophytes survived at a single microsite and have produced up to 26 sporelings after 24 months (Figure 2.1). Sporelings suffered a 62%

40 reduction in individuals after 4 months and represented the second highest rate of mortality among life stages after 24 months (97%). The rate of mortality of IA sporophytes was mostly consistent with that of sporelings, being reduced by 63% after 4 months and 93% after 24 months. IOA sporophytes consistently had the lowest rate of mortality among all transplant- types. IOA sporophytes were reduced by 40% after 4 months and 77% after 24 months, which is

17% lower than the next lowest rate of mortality (IA sporophytes) after 24 months.

After 24 months, the Twin Basins reintroduction site exhibited the highest raw percent survival for every transplant-type with exception of sporelings; which was lowest at Twin Basins

(1.25%) (Figure 2.2). The Sentinel Basin and Split Rock reintroduction sites exhibited sporeling survival rates of 3.95% and 3.90%, respectively. At each reintroduction site, IOA sporophytes exhibited the highest raw percent survival compared to all other life stages (Sentinel Basin,

30.77%; Split Rock, 13.73%; Twin Basins, 37.03%). IA sporophyte raw percent survival ranged from 2.78% (Sentinel Basin) to 10.81% (Twin Basins). Gametophytes exhibited 0.00% survival at Sentinel Basin and Split Rock, respectively, and 2.24% survival at Twin Basins.

At each census period, every transplant-type exhibited a higher mean percent survival than the transplant-type preceding it in terms of life stage, except for sporelings at 4 months after transplanting which exhibited a higher mean percent survival than the IA sporophyte transplant- type (Table 2.4; Figure 2.2). One-way ANOVA at 24 months after transplanting indicated a significant difference in mean percent survival between transplant-types (P = 0.004) (Table 2.3) and a Tukey’s HSD post hoc test indicated that IOA sporophytes consistently had a significantly higher mean percent survival than all other transplant-types (Table 2.4). Four months after transplanting, gametophytes exhibited a significantly lower mean percent survival (3.64% ± SE

1.71) than all other transplant-types. Sporelings (38.33% ± SE 4.53) and IA sporophytes (36.53%

± SE 6.77) exhibited a similar mean percent survival, with each transplant-type exhibiting a significantly higher mean percent survival than gametophytes. IOA sporophytes (60.79% ± SE

1.75) exhibited a significantly higher mean percent survival compared to all other transplant- types after 4 months.

In July 2016, 12 months after transplanting, all transplant-types exhibited a lower mean percent survival compared with October 2015. The mean percent survival of sporelings (5.92 ±

SE 1.73) and IA sporophytes (8.51 ± SE 3.98) were reduced drastically compared to October

2015. From July 2016 until the final census in July 2017, the gametophyte, sporeling, and IA sporophyte transplant-types exhibited mean percent survival rates that were not significantly different. The IOA sporophyte life stage, however, exhibited a significantly higher mean percent survival than all other life stages from the time of the first census in October2015 until the final census in July 2017. Following the census in July 2016, a notable “leveling off” of mean percent survival was observed, with reduction of mean percent survival for all life stages slowing dramatically during this time.

A total of 6 transplants reached maturity over the 24-month monitoring period as indicated by the presence of sori on the underside of the frond(s). At Twin Basins, 3 IOA sporophytes and 1 IA sporophyte reached maturity. Two IOA sporophytes reached maturity at

Split Rock, while no transplants of any type reached maturity at Sentinel Basin. Most of the transplants that reached maturity did so between 16 and 24 months, while a single transplant at

Twin Basins attained maturity within 12 months after transplanting.

The mean number of fronds and standard error was calculated for each frond-bearing transplant-type at each census period (Figure 2.3). Sporelings exhibited the lowest mean number of fronds (5.86 ± SE 1.60) among transplant-types at the time of transplanting. Four months after

42 transplanting, mean number of fronds for sporelings decreased slightly (4.17 ± SE 1.33) before an increase was observed in July 2016 (6.29 ± SE 1.69). Mean number of fronds has remained near constant for sporelings from July 2016 until the final census period in July 2017. IA sporophytes exhibited the highest mean number of fronds (13.86 ± SE 2.82) at the time of transplanting. The most dramatic decrease in mean number of fronds for IA sporophytes was observed 4 months after transplanting with the mean falling to 8.33 (± SE 1.12) fronds in

October 2015. Mean number of fronds was further reduced 12 months after transplanting (6.17 ±

SE 1.80) before increasing in October 2016 (7.50 ± 1.84) and remaining stable in July 2017 (7.29

± SE 2.74). IOA sporophytes exhibited an initial mean number of fronds of 9.19 (± SE 1.35) in

July 2015. Similar to IA sporophytes, a decrease in mean number of fronds was observed after 4 months (8.00 ± SE 0.87) and 12 months (5.40 ± SE 0.74) before increases at 16 months (7.00 ±

SE 1.04) and 24 months (8.57 ± SE 1.24) were observed. A general trend of a decrease in mean number of fronds over the first 4 to 12 months after transplanting, followed by a recovery and stabilization from 12 to 24 months after transplanting, was observed consistently between transplant-types.

The mean total frond length and standard error was calculated for each frond-bearing transplant-type at each census period (Figure 2.4). Sporelings exhibited the lowest mean total frond length (13.79 cm ± SE 8.34) of all transplant-types at initial transplanting. A decrease in mean total frond length was observed 4 months after transplanting (6.25 cm ± SE 2.53) followed by an increase at 12 months (19.58 cm ± SE 6.98) and slow but stable growth until 24 months

(21.57 cm ± 8.09) after transplanting. IA sporophytes exhibited the highest mean total frond length (86.20 cm ± SE 23.10) at initial transplanting. A dramatic decrease in mean total frond length was observed at 4 months after transplanting (25.86 cm ± SE 6.70) followed by steady

43 growth until 24 months after transplanting (45.60 cm ± 15.40). Mean total frond length of IOA sporophytes was 20.00 cm (± SE 4.06) at initial transplanting and remained relatively stable until an increase at 24 months after transplanting (33.30 cm ± SE 11.40). Similar to the trends observed for mean number of fronds, mean total frond length tended to decrease at 4 months after transplanting for all transplant-types. At 12 months after transplanting and beyond, a trend of stable and increasing growth is observed until the end of the 24-month monitoring period for each life stage.

Table 2.2. Summary of the number of initial transplants (July 2015) and confirmed surviving transplants over a 24-month monitoring period by transplant-type and reintroduction site.

Number of surviving transplants Transplant-type | Site Jul 2015 Oct 2015 Jul 2016 Oct 2016 Jul 2017 Protonemata (dishes) Sentinel Basin 42 0 0 0 0 Twin Basins 43 0 0 0 0 Split Rock 42 0 0 0 0 Total 127 0 0 0 0 Gametophytes (clumps) Sentinel Basin 493 20 0 0 0 Twin Basins 317 10 9 7 7 Split Rock 495 2 2 0 0 Total 1305 32 11 7 7 Sporelings Sentinel Basin 93 21 3 3 3 Twin Basins 88 36 6 4 1 Split Rock 87 28 5 3 3 Total 268 85 14 10 7 IA Sporophytes1 Sentinel Basin 42 10 2 1 1 Twin Basins 44 19 5 4 4 Split Rock 37 9 2 2 2 Total 123 38 9 7 7 IOA Sporophytes2 Sentinel Basin 17 9 4 4 4 Twin Basins 30 35 17 14 14 Split Rock 55 30 10 10 7 Total 102 74 31 28 25 Total (all transplant-types) Sentinel Basin 687 60 9 8 8 Twin Basins 522 100 37 29 26 Split Rock 716 69 19 15 12 Grand Total (all transplant-types and sites) 1,925 229 65 52 46

1 IA (Indoor acclimatized) sporophytes were subjected to only indoor acclimatization prior to transplanting. 2 IOA (Indoor/outdoor acclimatized) sporophytes were subjected to both indoor and outdoor acclimatization prior to transplanting.

Figure 2.1. AHTF sporelings in Twin Basins in July 2017 produced by the only surviving gametophyte transplants across three reintroduction sites. Initial sporeling production was recorded in October 2015. Photo: Michael Serviss.

Figure 2.2. Raw percent survival of four transplant-types of AHTF by reintroduction site as of July 2017.

Table 2.3. One-way analysis of variance results for mean percent survival at 24 months versus transplant-type at α= 0.05.

Source df SS MS F P Transplant-type 3 1316.5 438.82 10.60 0.004 Error 8 331.2 41.41 Total 11 1647.7

Table 2.4. Mean percent survival (± standard error) of AHTF transplants over 20 months (N=3) and corresponding grouping with Tukey’s HSD pairwise comparisons (α= 0.05). Values with different letters within each census period indicate significant difference from each other.

Mean % survival Transplant-type Oct 2015 July 2016 Oct 2016 July 2017 Gametophyte 3.64 (±1.71)c 0.96 (±1.44)b 0.75 (±0.75)b 0.75 (±0.75)b Sporeling 38.33 (±4.53)b 5.92 (±1.73)b 4.26 (±0.34)b 3.03 (±0.89)b IA sporophyte 36.53 (±6.77)b 8.51 (±3.98)b 6.68 (±2.32)b 6.68 (±2.32)b IOA sporophyte 60.79 (±1.75)a 31.61 (±6.84)a 29.14 (±5.09)a 27.18 (±6.96)a

Figure 2.3. Mean percent survival and standard error (N= 3) of four transplant-types of AHTF over 20 months.

Figure 2.4. Mean number of fronds and standard error of AHTF transplants of three transplant- types over 24 months.

Figure 2.5. Mean total frond length (cm) and standard error of AHTF transplants of three transplant-types over 24 months.

Ordinal Logistic Regression and Predictors of Transplant Survival

A proportional odds model (POM) ordinal logistic regression analysis (inclusive of frond-bearing transplant-types) was performed to determine the predictive effect of several categorical and continuous variables on the odds of survival of transplants over the period of 24 months (Table 2.5). Optimization of the model indicated that significant predictors of transplant survival (α = 0.10) include transplant-type (P < 0.001), vigor rating in July 2015 (P < 0.001), number of fronds in July 2015 (P = 0.074), mean frond length in July 2015 (P < 0.001), and transplant plot (P = 0.008). Optimization determined that the factors of reintroduction site (P >

0.10) and source population (P > 0.10) were not significant predictors of the survival of transplants over 24 months. The validity of the model was assessed with measures of association between the response variable and predicted probabilities with Somers’ D (0.62), Goodman-

Kruskal Gamma (0.62), and Kendall’s Tau-a (0.33).

The significant effect of transplant-type on the survival of transplants was highlighted with the results of multiple one-way ANOVA and Tukey’s HSD post hoc tests over the course of

24 months in the previous section. The effect of transplant-type was further quantified in the

POM ordinal logistic regression analysis. When using IOA sporophytes as the reference level in the model, the odds of sporelings dying within the first twelve months were 6.41 (95% CI, 3.48 to 11.81) times that of IOA sporophytes (P < 0.001). The odds of IA sporophytes dying within the first twelve months were 14.87 (95% CI, 6.70 to 32.98) times that of IOA sporophytes (P <

0.001); more than double the odds of sporelings. When using IA sporophytes as the reference level in the model, the odds that IA sporophytes would survive longer than twelve months were

43% (95% CI, 0.21 to 0.90) that of sporelings (P = 0.024) and only 7% (95% CI, 0.03 to 0.15) that of IOA sporophytes (P < 0.001).

Vigor rating at the time of transplanting was also a significant predictor of transplant survival in the model. When using “average” vigor as the reference level in the model, the odds that a transplant of “very poor” vigor would die prior to twelve months were 5.03 (95% CI, 1.29 to 19.57) times that of transplants of “average” vigor (P = 0.020). The odds of dying before twelve months for transplants rated as “poor” were improved to 2.49 (95% CI, 1.11 to 5.58) times that of “average” transplants (P = 0.027). As vigor rating increased, the odds that an

“average” transplant survived beyond twelve months were 43% (95% CI, 0.23 to 0.80) that of a transplant in “good” vigor (P = 0.007) and 32% (95% CI, 0.15 to 0.67) that of a transplant of

“excellent” vigor (P = 0.003). When using a vigor rating of “excellent” as the reference level, it was found that transplants of “very poor” (P < 0.001) and “poor” (P < 0.001) vigor were 15.62

(95% CI, 3.52 to 69.32) and 7.72 (95% CI, 2.96 to 20.13) times more likely to die before twelve months than transplants of “excellent” vigor, respectively. Furthermore, transplants of “average” vigor were 3.10 (95% CI, 1.49 to 6.48) times more likely to die before twelve months than transplants of “excellent” vigor (P = 0.003). There was no significant difference in the odds of survival past twelve months between transplants in “good” vigor and those in “excellent” vigor

(P = 0.410).

Predictors regarding the size of transplants at the time of transplanting, specifically the number of fronds and the mean frond length, were identified as significant predictors during model optimization. This predictive effect held true for mean frond length as an increase in mean frond length (per cm) corresponded with a 66% (95% CI, 0.56 to 0.79) increase in the odds that a transplant will survive beyond twelve months (P < 0.001). While mean frond length was a

51 significant predictor of survival, the effect of mean frond length is likely relatively small compared to other predictors as indicated by a coefficient that is closer to zero than most other predictors (Coefficient = -.411023). Number of fronds at the time of transplanting was a less significant predictor than mean frond length (P = 0.074). Additionally, the effect of number of fronds is limited in the model as evidenced by a coefficient that is extremely close to zero

(Coefficient = -0.0572257). Nevertheless, for each additional frond, the odds that a transplant is likely to die before twelve months is reduced by 6% (95% CI, 0.89 to 1.01).

The effect of transplant plot was also determined as a significant predictor of survival.

When compared with plot # SR9 (the plot with the most significant effect on survival beyond twelve months), transplants within an immediately adjacent plot (SR2) were 5.83 (95% CI, 1.01 to 33.66) times more likely to die within twelve months (P = 0.049). The plot with the most significant departure in results from plot # SR9 was plot # SB5. Transplants in plot # SB5 were

56.05 (95% CI, 5.83 to 538.91) times more likely to die before twelve months than transplants within plot # SR9 (P < 0.001). In total, transplants in 9 of 26 transplant plots had significantly lower odds of survival beyond twelve months when compared to plot # SR9.

Table 2.5. Results of the Proportional Odds Model ordinal logistic regression analysis for AHTF transplants across three reintroduction sites.

95% Confidence interval Predictor variable (categorical reference level) Coefficient P-value Odds ratio Lower Upper Transplant-type (IOA sporophyte) Sporeling 1.85856 < 0.001 6.41 3.48 11.81 IA sporophyte 2.69902 < 0.001 14.87 6.70 32.98 Transplant-type (IA sporophyte) Sporeling -0.840474 0.024 0.43 0.21 0.90 IOA sporophyte -2.69903 < 0.001 0.07 0.03 0.15 Vigor rating in July 2015 (Average) Very poor 1.61591 0.020 5.03 1.29 19.57 Poor 0.910902 0.027 2.49 1.11 5.58 Good -0.842066 0.007 0.43 0.23 0.80 Excellent -1.13284 0.003 0.32 0.15 0.67 Vigor rating in July 2015 (Excellent) Very poor 2.74838 < 0.001 15.62 3.52 69.32 Poor 2.04338 < 0.001 7.72 2.96 20.13 Average 1.13248 0.003 3.10 1.49 6.48 Good 0.290414 0.410 1.34 0.67 2.67 Number of fronds in July 2015 -0.0572257 0.074 0.94 0.89 1.01 Mean frond length in July 2015 -0.411023 < 0.001 0.66 0.56 0.79 Transplant plot (Plot# SR9) SR8 2.32129 0.003 10.19 2.18 47.56 SB3 1.59811 0.060 4.94 0.94 26.13 TB1 2.03294 0.011 7.64 1.60 36.39 TB7 0.351707 0.654 1.42 0.31 6.60 TB4 0.17667 0.819 1.19 0.26 5.42 SB5 4.02633 < 0.001 56.05 5.83 538.91 SB8 1.005 0.231 2.73 0.53 14.14 SR3 0.900267 0.273 2.46 0.49 12.31 TB8 1.54469 0.044 4.69 1.04 21.09 SB9 1.74888 0.087 5.75 0.78 42.47 SB6 0.404736 0.633 1.50 0.28 7.90 SR7 1.14000 0.137 3.16 0.69 14.36 SB2 3.11853 0.003 22.61 2.99 170.77 SR5 0.0355733 0.960 1.04 0.26 4.17 SB7 1.31595 0.127 3.73 0.69 20.21 TB9 1.2359 0.142 3.44 0.66 17.91 TB6 2.28291 0.009 9.81 1.77 54.21 SR6 1.46575 0.052 4.33 0.99 18.99 TB3 0.881202 0.257 2.41 0.53 11.07 SR4 1.96285 0.017 7.12 1.42 35.67 TB2 1.40551 0.089 4.08 0.81 20.59 TB5 0.296218 0.694 1.34 0.31 5.89 SR1 1.41481 0.076 4.12 0.86 19.61 SR2 1.7631 0.049 5.83 1.01 33.66 SB1 1.18708 0.178 3.28 0.58 18.44 SB4 2.23451 0.007 9.34 1.86 46.92

Discussion

Effects of Life Stage on Survival and Growth

Previous research on rare and threatened plant reintroductions often conclude that transplant life stage has a significant effect on the survival and growth of transplants (Drayton &

Primack 2000; Jusaitis et al. 2004; Maschinski & Duquesnel 2007; Godefroid et al. 2011;

Albrecht & Maschinski 2012; Dalrymple et al. 2012). In general, more advanced life stages tend to exhibit higher rates of survival and growth than less advanced life stages. However, previous research has consistently reached these conclusions based on reintroductions of seed-bearing plants. The results of this experiment on AHTF are inconclusive regarding the hypothesis that transplants of more advanced life stages will exhibit higher percent survival than transplants of less advanced life stages. IA and IOA sporophytes, the most advanced life stages, consistently exhibited a higher mean percent survival than less advanced life stages over the duration of the

24-month monitoring period. However, the difference in mean percent survival between IA sporophytes, sporelings, and gametophytes was not observed to be significant between 12 and 24 months after transplanting. The lack of significance suggests that, since IA sporophytes, sporelings, and gametophytes were not treated with outdoor acclimatization prior to reintroduction, the most significant effect on survival was the more rigorous outdoor acclimatization treatment received by IOA sporophytes prior to reintroduction. IOA sporophytes exhibited significantly higher survival compared with all other life stages, with a mean percent survival of 27.18% (± SE 6.96) after 24 months. It should be noted that the results of the one- way ANOVA and Tukey’s HSD tests were based on a small sample size (N=3), although the significant differences in survival between IOA sporophytes and other transplant-types are evident in practice when considering the number of individuals that have survived from each

54 transplant-type after 24 months. Despite representing the smallest number of individuals (102) at initial transplanting, IOA sporophytes have the highest number of surviving individuals (25) across all reintroduction sites after 24 months compared with IA sporophytes (7), sporelings (7), and gametophytes (7), respectively. IOA sporophytes also had the highest number of surviving individuals within each reintroduction site when compared to all other life stages. At the opposite end of the life stage spectrum, the use of protonemata as a potential reintroduction transplant life stage was unsuccessful (Table 2.2). Protonemata appear more sensitive to drying than all other life stages used in this experiment and received no acclimatization prior to transplanting, which helps to explain this result.

Differences in physiology at the time of transplanting between transplants that did not receive outdoor acclimatization and those that did could explain the significantly better performance of IOA sporophytes. While this experiment did not test for differences in leaf anatomy or physiology between transplants treated with different acclimatization methods at the time of transplanting, previous microscopy research has found that the leaves of field acclimatized plants have more defined palisade and spongy parenchyma layers, higher cell densities, and thicker epidermal and cuticle layers (Wetzstein & Sommer 1982; Pospóšilová et al. 1999). These anatomical differences improve the ability of leaves to buffer against the effects of desiccation, including lowering transpirational water loss post-reintroduction. At least two other studies have reported increased survival of lab propagated plants after transplanting into field conditions via the use of acclimatization of transplants in increasingly rigorous phases.

Perhaps the most successful documented fern reintroduction project noted that transplants of the rare and threatened Woodsia ilvensis propagated from spore were acclimatized in a shade tunnel at the Royal Botanic garden Edinburgh prior to reintroduction (McHaffie 2006). Multi-phased

55 acclimatization methods, including a succession of growth chamber, greenhouse, open greenhouse, and field acclimatization phases, resulted in the successful (95% reported survival) reintroduction of the endangered wild olive Olea maderensis (Brito et al. 2009).

The recruitment of new plants by a reintroduced population is typically the long-term goal and benchmark of success in plant reintroduction (Godefroid et al. 2011). A total of six transplants, including five IOA sporophytes and one IA sporophyte, reached maturity (based on the appearance of sori) during the 24-month monitoring period. Four transplants (3 IOA sporophytes, 1 IA sporophyte) reached maturity at Twin Basins, while the other two transplants

(IOA sporophytes) reached maturity at Split Rock. The recruitment of new AHTF individuals produced by transplants has yet to be observed, although, it is very early to expect any noticeable growth of new plants as most mature transplants started producing sori within 12 months of the end of the monitoring period. Maturation of transplants is a critical step in the process of promoting genetic connectivity between AHTF populations, particularly since most mature transplants are within Twin Basins. The Twin Basins site was strategically selected largely because it is between 320 and 800 m of six populations of AHTF within Clark Reservation State

Park. Additionally, the maturation of transplants at Split Rock increases the potential to augment the genetic diversity of the extant population at that site, which is less than 50 m from the reintroduction site.

Growth of transplants post-reintroduction, in terms of mean number of fronds, followed a similar trend for each frond-bearing transplant-type. Each transplant-type showed a decrease in mean number of fronds within 4 months after transplanting, likely due to difficulty acclimating to ambient conditions at each reintroduction site. Some loss during this period could also be due to natural senescence, as fronds may not tend to follow normal seasonal growth patterns

56 observed in wild plants due to differences between greenhouse, field (experimental), and wild environmental conditions. Sporelings recovered from the initial decrease within 12 months, but little increase in mean number of fronds between 12 and 24 months was observed. IA and IOA sporophytes exhibited a decrease in mean number of fronds at 4 and 12 months, respectively.

Each transplant-type then exhibited an increase in mean number of fronds at 16 and 24 months, respectively. However, neither IA nor IOA sporophytes achieved recovery to the same mean number of fronds exhibited at the time of transplanting. Despite only sporelings fully recovering in terms of mean number of fronds, the projection of mean number of fronds appears to be increasing for each transplant-type toward recovery and continued growth.

The continued production of new fronds is promising and it also influenced the measure of mean total frond length. Mean total frond length followed a similar pattern to mean number of fronds for each transplant-type, exhibiting a decrease at 4 months after transplanting followed by increases thereafter. Sporeling and IA sporophytes exhibited a higher decrease in mean total frond length than IOA sporophytes at 4 months after transplanting. This decrease could be because the larger fronds of sporelings and IA sporophytes tended to die at higher rates than smaller fronds, resulting in a reduced mean total frond length, especially for IA sporophytes.

Between 12 and 24 months, mean total frond length tended to increase for each transplant-type.

This trend is likely because the loss of fronds soon after transplanting gave way to production of new fronds in the spring and summer of 2016. Since AHTF is an evergreen fern with fronds that typically persist for more than one growing season, new fronds of any length increased the mean total frond length for a transplant because most living fronds from the previous field season were typically still present and were included in the analysis. Dead fronds from the previous field seasons were noted, but not included in the calculation of mean number of fronds or mean total

57 frond length because they were no longer contributing to the overall growth of each plant.

Evidence of herbivory of some fronds, mainly from slugs, was observed at each reintroduction site and likely contributed to death or inhibition of growth of at least some fronds. Additionally, the decrease in sample size over time as transplants died contributed to the inability to perform rigorous statistical hypothesis tests that could help quantify the effect of transplant-type on growth of transplants. The results of the effect of transplant-type on the overall growth of transplants are inconclusive. However, each transplant-type exhibited a trend of growth over time. This trend is promising for the use of each transplant-type in reintroduction. The effect of the size of transplants at the time of transplanting on the odds of survival over time are addressed in the next section.

Ordinal Logistic Regression and Predictors of Transplant Survival

The effect of transplant-type consistently favored the survival of IOA sporophytes over all other transplant-types. On the other hand, results of Multiple Tukey’s HSD pairwise comparisons indicated that there were no significant differences between mean percent survival of sporeling and IA sporophyte transplant-types. Regression analysis, however, showed interesting results regarding the odds of survival between sporelings and IA sporophytes. Despite exhibiting a lower mean percent survival compared with IA sporophytes at all census periods between 12 and 24 months, it appears that the odds that sporelings will survive longer than 12 months were 57% greater than that of IA sporophytes. As the model incorporates observations of both survival and death in an ordinal manner (in the context of time) as opposed to simply comparing mean percent survival at fixed time intervals for significance, it provides more accurate depictions of the likelihood of survival over time for each transplant-type. It should be noted that the coefficient calculated for sporelings in this analysis (- 0.840474) is more than three

58 times weaker than the coefficient comparing IOA sporophytes to IA sporophytes (- 2.69903).

Both coefficients indicate a likelihood of longer survival as indicated by the negative coefficient, however, the effect of the sporeling transplant-type is much less powerful than the effect of the

IOA sporophyte transplant-type when the size of the coefficients is incorporated into the interpretation. The interpretation of the regression analysis does not support the hypothesis that transplants of more advanced life stages are more likely to survive than transplants of less advanced life stages. However, the analysis strongly supports the hypothesis that transplants treated with more rigorous acclimatization are more likely to survive than transplants that received less rigorous acclimatization. There is clear evidence that more rigorous acclimatization is the most influential factor on the survival of reintroduced AHTF, rather than life stage.

One factor that is often overlooked in plant reintroduction is that, regardless of other factors, not all transplants are equal in terms of vigor at the time of transplanting. Size is not an accurate indicator of vigor as many larger transplants used in this experiment appeared to be of lesser vigor at the time of transplanting, and vice versa. Furthermore, differences in vigor at the time of transplanting were likely to influence the odds of survival and potential for growth of transplants of any transplant-type or size. Significant differences in survival between transplants with different vigor ratings were confirmed in model optimization when vigor rating was determined to be a significant predictor of survival over time. While the rating of vigor in this experiment was largely based on the appearance of the transplants as interpreted by practitioners, the results support the hypothesis that transplants of better vigor are more likely to survive longer than those of less vigor. This conclusion is supported by the significant results of the model, which indicated that even transplants with “average” vigor at the time of transplanting were 3 times more likely to die within 12 months than those of “excellent” vigor.

Size of transplants at the time of transplanting, as measured by number of fronds and mean frond length, showed a less impactful predictive power on prospects of survival over time when compared to the effects of transplant-type, nature of acclimatization, and vigor.

Nevertheless, there was a weak effect of mean frond length which suggests that as mean frond length at the time of transplanting increases, the odds of survival beyond 12 months also increase, even if the effect is very small. An even weaker effect of number of fronds was also present. As the number of fronds increased at the time of transplanting, the odds of surviving beyond 12 months also minimally increased. It is likely that the effect of both factors is not significant, as discussed previously, the size of a transplant is not an accurate proxy indicator of vigor, which had a much more significant effect on survival.

The effect of transplant plot on the odds of survival of a transplant was slightly less significant than the categorical predictors of transplant-type and vigor, but still had significant predictive influence on the survival of transplants. At least one previous fern reintroduction project concluded that a difference in niche habitat significantly influenced establishment of transplants (Aguraiuja 2011). While specific attributes of plot and/or microsite levels were not quantified in this experiment, including microclimate, edaphic, and biotic variables, field notes recorded at each census period can help describe some issues related to potentially excessive mortality at some of the plots. Many plots that showed significant departures from plot # SR9

(the best plot predictor of survival beyond 12 months) were observed to have issues of excessive erosion, rock slides, dryness of soil, heavy leaf litter, excessive moss growth over microsites, or evidence of soil heave likely caused by frost or ice. Some plots that did not have significant departures from plot # SR9 also had some of these issues, although the degree to which the issues affected the entire area of each plot was not quantified. It is likely that a combination of

60 the aforementioned issues related to mortality greatly influenced the odds of survival within each plot, however, it is not possible to analyze these variables with the data collected at each census.

Literature cited

Albrecht MA, Maschinski J. 2012. Influence of founder population size, propagule stages, and life history on the survival of reintroduced plant populations. In Plant reintroduction in a changing climate (pp. 171-188). Island Press/Center for Resource Economics, Washington, DC.

Aguraiuja R. 2011. Reintroduction of the endangered fern species Woodsia ilvensis to Estonia: a long-term pilot study. Biodiversity and conservation 20(2): 391-400.

Begon M, Harper JL, Townsend CR. 1986. Ecology: Individuals, populations and communities. 1068 pages. Blackwell Scientific Publications, Cambridge, Massachusetts.

Benedict RC. 1929. How have your hart's tongue plants succeeded? American Fern Journal 19(3): 108-109. Benedict RC. 1927. Is the distribution and naturalization of the hart's tongue scientifically defensible?. American Fern Journal 17(1): 19-23. Brant R. 1990. Assessing proportionality in the proportional odds model for ordinal logistic regression. Biometrics 46(4): 1171-1178.

Brito G, Costa A, Coelho C, Santos C. 2009. Large-scale field acclimatization of Olea maderensis micropropagated plants: morphological and physiological survey. Trees 23(5): 1019- 1031.

Cinquemani DM, Faust M E, Leopold D J. 1988. Periodic censuses (1916-1986) of Phyllitis scolopendrium var. americana in central New York State. American Fern Journal 78(2): 37-43.

Dewir YH, El-Mahrouk ME, Murthy HN, Paek KY. 2015. Micropropagation of Cattleya: Improved in vitro rooting and acclimatization. Horticulture, Environment, and Biotechnology 56(1): 89-93.

Drayton B, Primack RB. 2000. Rates of success in the reintroduction by four methods of several perennial plant species in eastern Massachusetts. Rhodora 102(911): 299-331.

Faust ME. 1960. Survival of hart's-tongue fern in central New York. American Fern Journal 50(1): 55-62.

Fernando DD, Discenza JJ, Bouchard JR, Leopold DJ. 2015. Genetic analysis of the threatened American hart's-tongue fern (Asplenium scolopendrium var. americanum [Fernald] Kartesz and Gandhi): Insights into its mating system and implications for conservation. Biochemical Systematics and Ecology 62(2): 25-35.

Fiedler PL, Laven RD. 1996. Selecting reintroduction sites. Restoring diversity (pp. 157-169). Washington, DC. Island Press.

Godefroid S, Piazza C, Rossi G, Buord S, Stevens AD, Aguraiuja R, Cowell C, Weekley CW, Vogg G, Iriondo JM, Johnson I. 2011. How successful are plant species reintroductions?. Biological Conservation 144(2): 672-682. Gravuer K, von Wettberg E, Schmitt J. 2005. Population differentiation and genetic variation inform translocation decisions for Liatris scariosa var. novae-angliae, a rare New England grassland perennial. Biological Conservation 124(2): 155-167.

Guerrant Jr EO. 2012. Characterizing two decades of rare plant reintroductions. In Plant Reintroduction in a Changing Climate (pp. 9-29). Island Press/Center for Resource Economics, Washington, DC.

Guerrant Jr EO, Kaye TN. 2007. Reintroduction of rare and endangered plants: common factors, questions and approaches. Australian Journal of Botany 55(3): 362-370.

House HD, Ransier HE, Benedict RC. 1926. Saving the hart's-tongue. American Fern Journal 16(2): 33-44.

Hufford KM, Mazer SJ. 2003. Plant ecotypes: genetic differentiation in the age of ecological restoration. Trends in Ecology & Evolution 18(3): 147-155.

Hunter MR. 1924. Notes on Scolopendrium vulgare Sm. American Fern Journal 14(4): 102-104.

Jusaitis M, Polomka L, Sorensen B. 2004. Habitat specificity, seed germination and experimental translocation of the endangered herb Brachycome muelleri (Asteraceae). Biological Conservation 116(2): 251-266.

Kaye TN. 2008. Vital steps toward success of endangered plant reintroductions. Native Plants Journal 9(3): 313-322.

Khan PSSV, Devi HS, Kishor RK, Rao BN. 2009. Micropropagation and some acclimatization characteristics of Centella asiatica (Linn.) Urban. Indian Journal of Plant Physiology 14(4): 353- 359.

Krause C, Pennington D. 2012. Strategic decisions in conservation: Using species distribution modeling to match ecological requirements to available habitat. In Plant Reintroduction in a Changing Climate (pp. 131-153). Island Press/Center for Resource Economics, Washington, DC.

Maschinski J, Duquesnel J. 2007. Successful reintroductions of the endangered long-lived Sargent’s cherry palm, Pseudophoenix sargentii, in the Florida Keys. Biological Conservation 134(1): 122-129.

Maschinski J, Falk DA, Wright SJ, Possley J, Roncal J, Wendelberger KS. 2012. Optimal locations for plant reintroductions in a changing world. In Plant reintroduction in a changing climate (pp. 109-129). Island Press/Center for Resource Economics, Washington, DC.

McCullagh P. 1980. Regression models for ordinal data (with discussion). J. Roy. Statist. Soc., B 42: 109-142.

McHaffie H. 2006. A reintroduction programme for Woodsia ilvensis (L.) R. Br. in Britain. Botanical Journal of Scotland 58(1): 75-80. McKay JK, Christian CE, Harrison S, Rice KJ. 2005. “How local is local?”—a review of practical and conceptual issues in the genetics of restoration. Restoration Ecology 13(3): 432- 440.

Menges ES. 2008. Restoration demography and genetics of plants: when is a translocation successful?. Australian Journal of Botany 56(3): 187-196.

Moore SG, Leopold DJ. 1992. Introduction of the Hart’s-tongue fern in central New York. Unpublished report. State University of New York College of Environmental Science and Forestry.

New York Natural Heritage Program, New York State Department of Environmental Conservation. 2012. Element Distribution Models. Albany, New York.

Pospóšilová J, Tichá I, Kadleček P, Haisel D, Plzáková Š. 1999. Acclimatization of micropropagated plants to ex vitro conditions. Biologia Plantarum 42(4): 481-497.

Preece JE, Sutter EG. 1991. Acclimatization of micropropagated plants to the greenhouse and field. Micropropagation: Technology and Application (pp. 71-93). Springer, Dordrecht, Netherlands.

Preece JE, West TP. 2006. Greenhouse growth and acclimatization of encapsulated Hibiscus moscheutos nodal segments. Plant Cell, Tissue and Organ Culture 87(2): 127-138.

USFWS (United States Fish and Wildlife Service). 1993. Recovery plan for American hart's- tongue fern (Asplenium scolopendrium L. var americanum [Fernald] Kartezs and Gandhi [Synonym: Phyllitis scolopendrium (L.) Newman var. americana Fernald]). U.S. Fish and Wildlife Service, Ashville, NC.

USFWS (United States Fish and Wildlife Service). 2012. American hart's-tongue fern (Asplenium scolopendrium var. americanum) 5 year review: summary and evaluation. U.S. Fish and Wildlife Service Ecological Services Field Office Cookeville, TN.

Weber-Townsend JR. 2017. Contributions of Genetic Data to the Conservation and Management of the Threatened American Hart's-Tongue Fern (Asplenium Scolopendrium var. americanum). Thesis, State University of New York College of Environmental Science and Forestry.

Weeks AR, Sgro CM, Young AG, Frankham R, Mitchell NJ, Miller KA, Byrne M, Coates DJ, Eldridge MD, Sunnucks P, Breed MF. 2011. Assessing the benefits and risks of translocations in changing environments: a genetic perspective. Evolutionary Applications 4(6): 709-725.

Wetzstein HY, Sommer HE. 1982. Leaf anatomy of tissue-cultured Liquidambar styraciflua (Hamamelidaceae) during acclimatization. American Journal of Botany 69(10): 1579-1586.

Yong YS, Yong WTL, Thien VY, Ng SE, Anton A, Yassir S. 2015. Acclimatization of micropropagated Kappaphycus alvarezii (Doty) Doty ex Silva (Rhodophyta, Solieriaceae) in outdoor nursery system. Journal of Applied Phycology 27(1): 413-419.

Chapter 3: Microclimate of Reintroduction Sites and Implications for Transplant Survival

Introduction

Ferns generally lack the coevolved dispersal vectors that influence the distribution of seed and pollen-bearing plants, therefore, climate and habitat preferences are considered the main constraints on fern species distribution (Barrington 1993). In particular, water availability and susceptibility to drought among most ferns are factors that tend to limit species distributions

(Quirk & Chambers 1981; Marquez et al. 1997; Karst et al. 2005; Bickford & Laffan 2006).

Increased periods of drought may present a severe problem for ferns, as the changing climate of the earth is expected to increase drought and tree mortality within forests across the globe, even for habitats that are not currently considered water-limited (Allen et al. 2010). Additionally, most ferns tend to be intolerant of wide fluctuations in light regimes, humidity, and wind speeds (Page

2002). Ferns that are distributed primarily within mesic habitats, such as mesic forests in central

New York, could be impacted significantly by drought because they are far less tolerant to desiccation than ferns that are distributed primarily in xeric habitats (Kessler & Siorak 2007).

While precipitation is projected to increase throughout the northeastern United States over the next century according to current models, drought events induced by higher temperatures and subsequent increases in evapotranspiration are also projected to increase throughout the region

(Wehner et al. 2017). Species with limited ranges also tend to be more vulnerable to changes in climate than those that are broadly distributed, which imparts an increased risk of extinction for species with limited ranges (Thuiller et al. 2005; Schwartz et al. 2006). While plant species have historically adapted to a changing climate (Davis & Shaw 2001), there are concerns regarding

65 genetic constraints on the adaptive potential of species in the face of a more rapidly changing climate (Etterson & Shaw 2001).

The impact of climate alone, of course, is not the only factor that influences the distribution of most ferns. Guo et al. (2003) found that, regarding life history, evergreen species and species that only reproduce sexually had smaller geographic ranges than seasonally green species and those with sexual and asexual strategies, respectively. Life history stage, particularly that of the gametophyte stage, also appears to be a factor in the distribution of some species.

Nutrient availability has been shown to significantly influence gametophyte mortality, growth, sex determination, and sporophyte production in Dryopteris filix-mas (Korpelainen 1994).

Regeneration niche, defined as the requirements necessary for a high chance of establishment of a new generation as determined by the processes and characters of a given area (Grubb 1977), is a critical factor in the distribution of highly specialized species, including many ferns.

Differential availability of a regeneration niche with higher humidity between forest types was found to significantly increase the number of gametophytes and sporophytes produced by spores in several fern species (Flinn 2007). In a fern reintroduction experiment examining aspects of regeneration niche, flooding and shading within microsites were found to be significant factors in the establishment of the endangered Hawaiian endemic, Marsillea villosa (Chau & Reyes 2013).

Wind, as it relates to spore dispersal, is another critical factor in the distribution of fern species. Anemochory, or wind dispersal, is the primary mechanism of dispersal of spores for most ferns. Long distance dispersal, at the scale of thousands of kilometers, is possible for pteridophytes because of the small, lightweight character of the spores and their ability to remain viable in the extreme conditions of the upper atmosphere (Page 2002). However, spore dispersal may be limited to the scale of tens of meters or less under conditions of low wind or when

66 physical barriers, such as dense forests, inhibit the movement of fern spores (Raynor et al. 1976).

Lack of gene flow in AHTF, as well as their occurrence within well protected glacial plunge basins and ravines, suggest that wind dispersal of spores between AHTF populations is limited at best (Fernando et al. 2015). Recent research suggests that alternative mechanisms of fern spore dispersal, including endozoochory and epizoochory, are potentially more common and widespread than is currently understood (Sugita et al. 2013; Barbe et al. 2016; Boch et al. 2016).

The relationship between climatic factors and the occurrence and demography of AHTF at various spatial and temporal scales has been the focus of several studies in recent decades.

Common microclimatic factors that influence the occurrence of AHTF throughout central New

York include a preference for habitats that are typically lower in temperature, but with fewer freezing days, than adjacent areas, as well as more limited extreme fluctuations in temperature

(Brumbelow 2014). The occurrence of AHTF in well protected glacial plunge basins and ravines likely helps to buffer against the effects of extreme fluctuations in climate, including freezing, which has been found to be damaging to mature sporophytes even for the non-threatened closest relative of AHTF, the European hart’s-tongue fern (Asplenium scolopendrium var. scolopendrium) (Bremer & Jongejans 2010). Higher humidity within AHTF habitats has also been documented in contrast to adjacent areas (Petry 1918). AHTF’s preference for higher levels of moisture, and sensitivity to drought, have been highlighted by decreases in population size following drought events, especially for populations less protected from wind desiccation

(Cinquemani et al. 1988; Cinquemani Kuehn & Leopold 1992). AHTF gametophytes may be especially susceptible to desiccation and an inability to recover following even mild drying events (Testo & Watkins 2013). The number of very wet months in a year has also been found to

67 negatively impact AHTF populations, possibly due to competition from the enhanced growth of neighboring understory herbaceous plants (Kelsall et al. 2004).

While there is extensive knowledge regarding the effect of climatic factors on extant populations of AHTF, at least in central New York, it is unclear what the effect may be on propagated AHTF post-reintroduction. It is also unclear if any differences in microclimate between reintroduction sites may be able to account for differences in survival between sites or transplant-types. Generally, reintroduction sites that are ecologically similar to that of the source population are recommended for rare and threatened plant reintroductions (Lawrence & Kaye

2009). The reintroduction sites selected for this experiment were deemed highly suitable by a habitat model created specifically for AHTF in New York (New York Natural Heritage Program,

2012) that incorporated thousands of data points related to dozens of environmental variables associated with the species known habitat characteristics. Field visits to each site further verified the similarity of habitat characteristics between reintroduction and extant AHTF sites.

Furthermore, two of the selected sites, Sentinel Basin and Split Rock, are within 50 m of extant

AHTF populations, so extreme differences in microclimate between extant and reintroduced populations at these sites are unlikely.

The goal of this study is to describe the general microclimate aspects of each reintroduction site into which AHTF were transplanted for this experiment. Any extreme differences in temperature, humidity, solar radiation, and wind speed and direction may help explain any observed differences in survival of AHTF transplants at each site, slope position, or in relation to each transplant-type. However, extreme differences in microclimate between the reintroduction sites are not expected, therefore, this study aims to provide baseline information regarding the microclimate at each site and to validate the selection criteria for each

68 reintroduction site. Baseline microclimate information is particularly important for the Twin

Basins reintroduction site because the microclimate of the site has not been previously studied due to the absence of an extant AHTF population within the basins. Wind speed and direction data have never been collected within AHTF habitats in central New York and could prove particularly useful in determining the extent to which wind is influencing AHTF spore dispersal and desiccation of sporophytes and gametophytes. Information on these microclimate aspects may be useful in the identification and selection of future AHTF reintroduction sites.

Methods

Study Design

Three microclimate monitoring stations were deployed at each of three reintroduction sites. Each station was equipped with one Davis Cup Anemometer (Davis Instruments Corp., adapted for use by Decagon Devices, Inc.) positioned approximately one meter above the ground, one solar radiation sensor (Apogee Instruments Inc., model SP-110) positioned approximately 75 cm above the ground, one temperature and humidity sensor (Decagon Devices,

Inc., model VP-3) positioned approximately 50 cm above the ground, and one data logger

(Decagon Devices Inc., model EM-50) (Figure 3.1). Davis Cup Anemometers were calibrated to true north in accordance with manufacturer specifications. Solar radiation sensor cables were positioned pointing to true north in accordance with manufacturer specifications. Each sensor and data logger were affixed to a PVC pipe that was staked into the ground with supports that helped to maintain the level surface of each sensor. The stations were arranged along a single transect running from the upper slope at each reintroduction site to the lower slope of each site

(not the bottom of the slope, but aligned with the lowest plots) at approximately 10 m distance from each other. The transect was placed approximately through the center of each reintroduction site. Sensors recorded and logged data every 15 min over the entire course of the

24-month monitoring period (except for January to March 2016 and intermittently when sensors failed or required repair) for a total of 96 measurements per day. The data were downloaded in the field periodically, at which time batteries were checked and sensors were inspected for damage and cleaned in accordance with manufacturer specifications.

Figure 3.1. A microclimate monitoring station mounted in Clark Reservation State Park, Jamesville, NY. Photo: Michael Serviss.

Data Analysis

For each day, the mean, minimum, and maximum values were calculated or recorded for the variables of air temperature (ºC), relative humidity (%), and solar radiation (Watts/m2) using

Excel 2016 (Microsoft Corporation). Daily statistics were then used to create seasonal profiles for each station by calculating the mean daily, mean daily minimum, and mean daily maximum values and standard deviations of each variable. Seasons were delineated in accordance with traditionally recognized seasonality including Winter (January to March), Spring (April to June),

Summer (July to September), and Autumn (October to December). Each season included data beginning with the first day of the first month included in the season and ending with the last day of the last month, whenever possible.

Wind speed and direction were analyzed separately from all other climate variables. The frequency of direction of wind, divided into sixteen distinct directions with a 22.5º allotment for each direction (360º scale), was calculated for each direction. Each wind event (an observation of wind with both a speed and a direction) was tallied by direction and calculated into a proportion

(representing frequency) of all wind events for each direction over the entire monitoring period using Excel 2016 (Microsoft Corporation). Additionally, mean wind speed (m/s) and mean gust speed (m/s) were calculated for each direction using wind events over the entire monitoring period, where gust speed was defined as the strongest observed wind during each wind event. All calculations of direction frequency, mean wind speed, and mean gust speed were performed on all available data for each station spanning the entire monitoring period except for data between

November 15th and March 15th of each year, as the likelihood that snow would cover transplants was greatest during these periods. Calculations of wind direction frequency, mean wind speed, and mean wind gusts were plotted concurrently for each station in a radar chart to aid in the

71 visualization of wind speed and direction recorded at each station. The radar charts, or wind rose diagrams as they are often referred to, are commonly used graphical representations of wind speed and direction in the meteorological and wind energy industries. The lower slope station at the Twin Basins site was omitted from the analysis due to corruption of most of the wind data collected over the monitoring period.

Results

A full microclimate profile of each season for which data was recorded, inclusive of air temperature, relative humidity, and solar radiation means and standard deviations for each slope position at each site, is included in Appendix B.

With respect to slope position and season, the difference in mean daily temperature was consistently within 1 ºC across each reintroduction site without exception. Similarly, the difference between the mean daily minimum temperature across all sites was never more than 1

ºC with respect to slope position and season. However, there were differences of larger than 1 ºC in mean daily maximum temperature during two seasons, Spring 2016 and Winter 2017. During

Spring 2016, a difference of 1.17 ºC between the lower slopes of Sentinel Basin (19.38 ºC) and

Twin Basins (18.21 ºC) was observed. During Winter 2017, the Sentinel Basin mid-slope position exhibited a mean daily maximum of 4.51 ºC, which was 1.34 ºC higher than the observed mean at Split Rock (3.17 ºC) and 1.44 ºC higher than the observed mean at Twin

Basins (3.07 ºC). Mean daily temperature, mean daily minimum temperature, and mean daily maximum temperature generally tended to increase with increasing slope position across all sites

72 and seasons. The difference in temperature between lower, mid, and upper slope positions, though, was rarely more than 1 ºC.

Mean daily relative humidity tended to vary more noticeably than temperature across sites with respect to season and slope position. For lower slope positions, differences between sites of 4.0% or more were observed in Summer 2015, Autumn 2016, and Winter 2017 for stations with full data sets. With respect to lower slope position, the Sentinel Basin site consistently exhibited the highest mean daily relative humidity compared to all other sites from

Summer 2015 until Summer 2016. After Summer 2016, the Twin Basins site then consistently exhibited the highest relative humidity across all sites at the lower slope position. With respect to upper and mid-slope positions, none of the sites exhibited a consistently higher or lower relative humidity compared to other sites across all seasons. The largest observed difference in mean daily relative humidity between sites, with respect to season and slope position, was a difference of 5.3 % between the upper slope positions of Split Rock (86.2 %) and Twin Basins (80.9 %) during Summer 2015.

Mean daily minimum relative humidity exhibited larger divergence between sites with respect to slope position and season. The largest such divergence was a difference of 8.3% between the lower slope positions of Twin Basins (80.7%) and Split Rock (72.4%) during

Autumn 2016. Mean daily minimum relative humidity rarely exhibited differences of more than

3.0% between sites with respect to mid-slope positions during each season. Divergence of less than 3.0% was also the trend for upper slope positions, except for a difference of 5.8% between the Split Rock (64.7%) and Twin Basins (58.9%) during Summer 2015.

Mean daily maximum relative humidity rarely exhibited differences of greater than 3.0% across all sites with respect to season and slope position. The largest divergence from this trend was observed in Autumn 2016, when Split Rock (94.7%) had a 6.4% lower mean than Sentinel

Basin (101.1%) and a 7.2% lower mean than Twin Basins (101.9%) with respect to the lower slope position. Mean daily, mean daily minimum, and mean daily maximum relative humidity generally tended to decrease when moving from lower to upper slope positions across all sites.

The largest difference for any relative humidity variable, with respect to site and season, was a difference of 9.7% in mean daily minimum relative humidity between lower (73.1%) and upper

(63.4%) slope positions at Twin Basins during Summer 2017.

Mean daily and mean daily maximum solar radiation varied widely between sites, seasons, and occasionally slope positions. In general, the lowest mean daily and mean daily maximum solar radiation was observed in the summer and autumn seasons across all sites. The lowest mean daily solar radiation was observed at the upper slope position of Twin Basins (4.44

W/m2) during Autumn 2016. The highest mean daily solar radiation during that same season was observed at the mid-slope of Split Rock (7.91 W/m2). The highest observations of mean daily solar radiation were during spring seasons, generally ranging from 20 to 45 W/m2 during 2016 and 2017. The highest observed mean daily solar radiation was at the lower slope position of

Sentinel Basin in Spring 2016 (48.31 W/m2). When comparing all slope positions at each site with respect to season, stations at Sentinel Basin had the highest mean daily solar radiation of any site during seven of the eight seasons that span the monitoring period. Sentinel Basin also had the highest mean daily maximum solar radiation of any site during seven of the eight monitored seasons. The highest observed mean daily maximum solar radiation was at the lower slope position of Sentinel Basin during Spring 2016 (417.10 W/m2). In contrast, the lowest

74 observed mean daily maximum solar radiation during the same season was at the upper slope position of Twin Basins (143.00 W/m2). The lowest observed mean daily maximum solar radiation was at the upper slope position of Twin Basins during Autumn 2016 (29.30 W/m2).

During the same season, the highest mean daily maximum solar radiation was observed at the mid-slope position of Sentinel Basin (109.90 W/m2).

The full set of data tables accompanying each wind radar figure presented in this section are available in Appendix C. Mean wind speed, gust speed, and frequency of direction varied widely across sites. Mean wind and gust speeds tended to be low overall but were highest at Split

Rock, which saw mean wind speeds consistently well over 0.12 m/s regardless of direction. In contrast, Sentinel Basin and Twin Basins rarely saw mean wind speeds over 0.12 m/s and often speeds were below 0.10 m/s. Mean wind speed at Split Rock never fell below 0.10 m/s for any direction. Additionally, Split Rock received the highest mean gust speeds and received wind more consistently from the same relative direction (west to west-southwest).

At Sentinel Basin, the most frequent direction of the wind originated from easterly directions, spanning from northeast to southeast more frequently than other directions. Each station exhibited a slightly different direction from which the wind most frequently originated, though, with the lower slope position most frequently receiving wind from the southeast (14.7%)

(Figure 3.1), the mid-slope from the northeast (15.9%) (Figure 3.2), and the upper slope from the east-northeast (23.8%) (Figure 3.3). The highest mean wind speed at Sentinel Basin was observed to originate from the north-northeast at 0.20 m/s as observed at the lower slope position. This slope position and direction also received the highest mean gust speed (0.52 m/s) compared to all other directions and slope positions.

Across all slope positions at Split Rock, the most frequent direction of the wind originated from the west to southwest. At the lower slope position, the most frequent wind direction originated from the southwest (18.1%) (Figure 3.4). The most frequent wind direction at the mid and upper slope positions originated from the west-southwest at 17.8% (Figure 3.5) and 17.1% (Figure 3.6), respectively. The highest mean wind speed at Split Rock was observed at the lower slope position at 0.38 m/s, originating from the southwest. The highest mean gust speed at Split Rock (0.79 m/s) was observed at the same slope position and originating from the same direction as the highest mean wind speed.

At Twin Basins, the most frequent direction of the wind varied between the lower and mid-slope positions (the only two slope positions included in this analysis). The lower slope position most frequently received wind originating from the east (18.1%) (Figure 3.7) while the mid-slope position most frequently received wind originating from the west-southwest (11.2%)

(Figure 3.8). The highest mean wind speed at Twin Basins was 0.14 m/s at the mid-slope position, originating from both the north-northeast and east-southeast, respectively. The highest mean gust speed originated from the north-northeast at 0.40 m/s, also at the mid-slope position.

N NNW 0.6 NNE 0.5 NW NE 0.4 0.3 WNW ENE 0.2 0.1 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.2. Wind radar chart for Sentinel Basin lower slope from 2015-2017.

N NNW 0.35 NNE 0.3 NW 0.25 NE 0.2 WNW 0.15 ENE 0.1 0.05 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.3. Wind radar chart for Sentinel Basin mid-slope from 2015-2017.

N NNW 0.6 NNE 0.5 NW NE 0.4 0.3 WNW ENE 0.2 0.1 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.4. Wind radar chart for Sentinel Basin upper slope from 2015-2017.

N NNW 0.8 NNE 0.7 NW 0.6 NE 0.5 0.4 WNW 0.3 ENE 0.2 0.1 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.5. Wind radar chart for Split Rock lower slope from 2015-2017.

N NNW 0.8 NNE 0.7 NW 0.6 NE 0.5 0.4 WNW 0.3 ENE 0.2 0.1 W 0 E

WSW ESE

SW SE

SSW SSE

Frequency Average wind speedS (m/s) Average gust speed (m/s)

Figure 3.6. Wind radar chart for Split Rock mid-slope from 2015-2017.

N NNW 0.8 NNE 0.7 NW 0.6 NE 0.5 0.4 WNW 0.3 ENE 0.2 0.1 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.7. Wind radar chart for Split Rock upper slope for 2015 (2016-2017 data missing).

N NNW 0.35 NNE 0.3 NW 0.25 NE 0.2 WNW 0.15 ENE 0.1 0.05 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.8. Wind radar chart for Twin Basins lower slope from 2015-2016 (2017 data missing).

N NNW 0.45 NNE 0.4 NW 0.35 NE 0.3 0.25 WNW 0.2 ENE 0.15 0.1 0.05 W 0 E

WSW ESE

SW SE

SSW SSE S Frequency Average wind speed (m/s) Average gust speed (m/s)

Figure 3.9. Wind radar chart for Twin Basins mid-slope from 2015-2017.

Discussion

There were no significant differences in temperature observed across all sites, seasons, and slope positions at AHTF reintroduction sites. With respect to season and slope position, temperature factors of mean daily, mean daily minimum, and mean daily maximum were consistently within 1ºC of each other across all sites. The minimal divergence in temperature across sites suggests that reintroduction sites all experienced similar temperature means and ranges, therefore, temperature alone is not likely to explain any differences in survival between reintroduction sites and transplant-types. Similarities in temperature between the reintroduction sites also validate the criteria used to select each site, especially for Twin Basins, which included similarity of habitat to known AHTF habitat characteristics. Temperature did tend to increase with increasing slope positions, which is consistent with previous findings (Petry 1918;

Brumbelow 2014). However, Brumbelow (2014), who placed sensors closer to the ground (2.5 cm), reported some significant differences in temperature variables between upper - lower plots and middle – lower plots, respectively. The positions of sensors in relation to slope position in this study, which did not span the entirety of any slope included in the study, are likely to align more similarly with the upper and mid-slope positions utilized by Brumbelow (2014), for which no significant differences in temperature were reported.

Relative humidity tended to vary more widely across sites than temperature, although, few clear trends were observed. One notable trend was that Twin Basins tended to have a higher relative humidity and a greater difference between lower and upper slope positions compared to the Sentinel Basin and Split Rock reintroduction sites. The higher humidity and difference in humidity between slope positions is likely explained by the topography and aspect of Twin

Basins, which is the deepest from top to bottom of all sites and faces roughly north-northeast,

81 with slopes and cliff faces protecting it to some degree on three sides. Fractures in the bedrock and the existence of caves at the bottom of the basin might also help to explain this difference.

Trapped water and ice are protected from light and wind and evidently release more moisture into Twin Basins than the other sites. Interestingly, Twin Basins, the only site where AHTF has never been documented historically, exhibited the highest percent survival of all AHTF transplant-types except sporelings. More sporophytes reached maturity at Twin Basins than at any other site and the only gametophytes to survive and produce sporelings were at Twin Basins.

These results are consistent with previous findings that AHTF sporophytes (Cinquemani et al.

1988; Cinquemani Kuehn & Leopold 1992) and gametophytes (Testo & Watkins 2013) respond positively to higher levels of moisture.

Sentinel Basin generally exhibited the highest solar radiation across all reintroduction sites. The higher solar radiation is likely due to the aspect of the reintroduction site, which spans from northeast to east facing and is less protected from light throughout most of the day than the other sites. The bottom of Sentinel Basin is also relatively clear of any significant canopy, allowing more light to reach the slope. It was observed at each census that the talus of Sentinel

Basin was drier than that of other sites, which is consistent with the higher solar radiation received at the site. Dryness could be a contributing factor to the rapid mortality of gametophyte transplants at Sentinel Basin, which experienced 100% mortality at least 4 months before Split

Rock. Previous research has shown that AHTF gametophytes are extremely sensitive to even mild drying events and have limited ability to recover following a drying event (Testo &

Watkins 2013). The site that received the lowest solar radiation, Twin Basins, was also the site with the highest mean percent survival of IA and IOA sporophytes and contained the only surviving gametophyte transplants. It is likely that the lower solar radiation at Twin Basins,

82 combined with the higher relative humidity and lower winds, buffered against excessive drying at the site and contributed to higher rates of survival of AHTF transplants.

Wind speed across all sites was extremely limited, which is consistent with previous descriptions of extant AHTF habitats (Cinquemani et al. 1988). Over a similar period of time, the mean wind speed for the nearest archived weather station (3.58 m/s, Hancock International

Airport in Syracuse, NY) (The Weather Company, LLC), was over 9 times greater than the highest mean wind speed (0.38 m/s, Split Rock lower slope) recorded at any AHTF reintroduction site. Similarly, mean gust speed (10.73 m/s) at Hancock International Airport was over 13 times greater than the highest mean gust speed (0.79 m/s, Split Rock lower slope) recorded at any AHTF reintroduction site. The generally low wind within the heavily protected talus supports recent genetic analyses of AHTF, which suggest that spore dispersal between populations is limited at best, citing evidence of lack of gene flow between populations throughout the range of AHTF (Fernando et al. 2015; Weber-Townsend 2017). However, evidence exists that spores are at least minimally dispersed at short distances outside of the heavily protected talus. Sporophytes of AHTF are occasionally distributed on the forest floor above the talus at tens of meters distance from the main population, indicating that spores disperse outside of the direct habitat on occasion (personal observation). However, the exact mode of dispersal for AHTF found outside of the talus is not known and may not be attributable to wind.

The lack of spore dispersal by AHTF inferred by low wind speeds likely explains why, in part, colonization of new sites by AHTF has not been documented even when suitable habitat is nearby, as seems to be the case for Twin Basins. Low wind speeds also have important implications for spore dispersal by AHTF transplants at reintroduction sites. Several AHTF

83 transplants have reached maturity, as indicated by the production of sori, at 2 of the 3 reintroduction sites. At Split Rock, winds originating from the west-southwest and southwest are the strongest and most prevalent, which roughly aligns with the west to east orientation of the ravine and is consistent with the prevailing winds in central New York. Given the prevailing wind at the site, spores from the extant population at Split Rock are probably more likely to be carried into the reintroduction site, which is less than 50 m due east of the extant population. The possibility of spore dispersal from the extant population at Split Rock into the reintroduction site indicates that cross-breeding between AHTF transplants and extant AHTF sporophytes will be possible. However, to encourage gene flow into the extant population at Split Rock, it would be beneficial to augment AHTF into sites along the talus to the west of the population. Candidate sites to the west of the extant AHTF population at Split Rock were evaluated for this reintroduction project, but the strong presence of currently unmanaged invasive species spanning the talus to the west of the population influenced the selection of the reintroduction site to the east. Twin Basins, which had lower wind speeds than Split Rock, had more complicated direction frequencies between slope positions. One possible explanation for the observed differences in wind direction between stations could have to do with the physical attributes of

Twin Basins. As previously discussed, Twin Basins is deeper than the other reintroduction sites and traps water and ice at the bottom of the basin. Research on wind dynamics in similar talus systems suggests that a chimney effect, caused by the difference in temperature between areas inside and outside of the talus, may be complicating the dynamics of wind flow in the system

(Delaloye & Lambiel 2005). The chimney effect may cause air to flow upward from the bottom of the basin and interact with winds flowing into the basin in an unpredictable manner. The dynamics of wind flow within talus and plunge basin systems are still poorly understood,

84 therefore, it is difficult to say with any certainty how this may affect spore dispersal in Twin

Basins.

Additional mechanisms of spore dispersal in AHTF are also possible. Slugs are common occupants of AHTF habitats and are herbivores of mature AHTF fronds (DD Fernando, personal communication). Recent research has shown that endozoochory of European hart’s-tongue fern spores by slugs resulted in about 90% germination success 81 days after passing through the slug digestive track (Boch et al. 2016). Unfortunately, the authors did not focus on dispersal distance as a factor in this study. Epizoochory may also be a possible mechanism of dispersal for AHTF.

Slugs were found to be potential short distance (less than 20 cm) dispersers of asexual “brood branches” of the moss Dicranum flagellare, dislodging and transporting the branches from the parent plant in their slime trail (Kimmerer & Young 1995). Additionally, small (Barbe et al.

2016) and large (Sugita et al. 2013) forest-dwelling mammals are capable of dispersing fern spores in their fur. Given the low wind speeds within reintroduction sites and extant AHTF populations, it is likely that the importance of other mechanisms of spore dispersal of AHTF have been underestimated. However, spore dispersal by zoochory is more likely to occur within

AHTF populations and not between them.

Literature cited

Allen CD, Macalady AK, Chenchouni H, Bachelet D, McDowell N, Vennetier M, Kitzberger T, Rigling A, Breshears DD, Hogg ET, Gonzalez P. 2010. A global overview of drought and heat- induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259(4): 660-684.

Barbé M, Chavel ÉE, Fenton NJ, Imbeau L, Mazerolle MJ, Drapeau P, Bergeron Y. 2016. Dispersal of bryophytes and ferns is facilitated by small mammals in the boreal forest. Ecoscience 23(3-4): 67-76.

Barrington DS. 1993. Ecological and historical factors in fern biogeography. Journal of Biogeography 20(3): 275-279.

Bickford SA, Laffan SW. 2006. Multi‐extent analysis of the relationship between Pteridophyte species richness and climate. Global Ecology and Biogeography 15(6): 588-601.

Boch S, Berlinger M, Prati D, Fischer M. 2016. Is fern endozoochory widespread among fern- eating herbivores?. Plant Ecology 217(1): 13-20.

Bremer P, Jongejans E. 2010. Frost and forest stand effects on the population dynamics of Asplenium scolopendrium. Population Ecology 52(1): 211.

Cinquemani DM, Faust M E, Leopold D J. 1988. Periodic censuses (1916-1986) of Phyllitis scolopendrium var. americana in central New York State. American Fern Journal 78(2): 37-43.

Cinquemani Kuehn DMC, Leopold DJ. 1992. Long-term demography of Phyllitis scolopendrium (L.) Newm. var. americana Fern. in central New York. Bulletin of the Torrey Botanical Club 119(1): 65-76.

Davis MB, Shaw RG. 2001. Range shifts and adaptive responses to Quaternary climate change. Science 292(5517): 673-679.

Delaloye R, Lambiel C. 2005. Evidence of winter ascending air circulation throughout talus slopes and rock glaciers situated in the lower belt of alpine discontinuous permafrost (Swiss Alps). Norsk Geografisk Tidsskrift-Norwegian Journal of Geography 59(2): 194-203.

Etterson JR, Shaw RG. 2001. Constraint to adaptive evolution in response to global warming. Science 294(5540): 151-154.

Flinn KM. 2007. Microsite‐limited recruitment controls fern colonization of post‐agricultural forests. Ecology 88(12): 3103-3114.

Grubb PJ. 1977. The maintenance of species‐richness in plant communities: the importance of the regeneration niche. Biological reviews 52(1): 107-145.

Guo Q, Kato M, Ricklefs RE. 2003. Life history, diversity and distribution: a study of Japanese pteridophytes. Ecography 26(2): 129-138.

Karst J, Gilbert B, Lechowicz MJ. 2005. Fern community assembly: the roles of chance and the environment at local and intermediate scales. Ecology 86(9): 2473-2486.

Kelsall N, Hazard C, Leopold DJ. 2004. Influence of climate factors on demographic changes in the New York populations of the federally-listed Phyllitis scolopendrium (L.) Newm. var. americana. Journal of the Torrey Botanical Society 131(2): 161-168.

Kessler M, Siorak Y. 2007. Desiccation and rehydration experiments on leaves of 43 pteridophyte species. American Fern Journal 97(4): 175-185.

Kimmerer RW, Young CC. 1995. The role of slugs in dispersal of the asexual propagules of Dicranum flagellare. Bryologist 98(1): 149-153.

Korpelainen H. 1994. Growth, sex determination and reproduction of Dryopteris filix‐mas (L.) Schott gametophytes under varying nutritional conditions. Botanical Journal of the Linnean Society 114(4): 357-366.

Lawrence BA, Kaye TN. 2011. Reintroduction of Castilleja levisecta: effects of ecological similarity, source population genetics, and habitat quality. Restoration Ecology 19(2): 166-176.

Marquez AL, Real R, Vargas JM, Salvo AE. 1997. On identifying common distribution patterns and their causal factors: a probabilistic method applied to Pteridophytes in the Iberian Peninsula. Journal of Biogeography 24(5): 613-631.

New York Natural Heritage Program, New York State Department of Environmental Conservation. 2012. Element Distribution Models. Albany, New York.

Page CN. 2002. Ecological strategies in fern evolution: a neopteridological overview. Review of Paleobotany and Palynology 119(1): 1-33.

Petry LC. 1918. Studies on the vegetation of New York State-II. The vegetation of a glacial plunge basin and its relation to temperature. Bulletin of the Torrey Botanical Club 45(5): 203- 210.

Quirk H, Chambers TC. 1981. Drought tolerance in Cheilanthes with special reference to the gametophyte. Fern Gazette 12(3): 121-129.

Raynor GS, Ogden EC, Hayes JV. 1976. Dispersion of fern spores into and within a forest. Rhodora 78(815): 473-487.

Schwartz MW, Iverson LR, Prasad AM, Matthews SN, O'Connor RJ. 2006. Predicting extinctions as a result of climate change. Ecology 87(7): 1611-1615.

Sugita N, Ootsuki R, Fujita T, Murakami N, Ueda K. 2013. Possible spore dispersal of a bird- nest fern Asplenium setoi by Bonin flying foxes Pteropus pselaphon. Mammal Study 38(3): 225- 229.

Testo WL, Watkins JE. 2013. Understanding mechanisms of rarity in pteridophytes: Competition and climate change threaten the rare fern Asplenium scolopendrium var. americanum (Aspleniaceae). American Journal of Botany 100(11): 2261-2270.

Thuiller W, Lavorel S, Araújo MB. 2005. Niche properties and geographical extent as predictors of species sensitivity to climate change. Global Ecology and Biogeography 14(4): 347-357.

Wehner MF, Arnold JR, Knutson T, Kunkel KE, LeGrande AN. 2017. Droughts, floods, and wildfires. In Climate Science Special Report: Fourth National Climate Assessment, Volume I, (pp. 231-256) Wuebbles DJ, Fahey DW, Hibbard KA, Dokken DJ, Stewart BC, TK Maycock TK, editors. U.S. Global Change Research Program, Washington, DC, doi: 10.7930/J0CJ8BNN.

Chapter 4: Conclusions and Recommendations

The American hart’s-tongue fern (Asplenium scolopendrium var. americanum, AHTF) is a climatically sensitive, rare, and threatened plant with genetically isolated populations spread over a large and fragmented geographic range. The extinction of some populations, and the remaining threat of extinction for others, have been a matter of concern for over 100 years. In accordance with the management plan for the species and based on recommendations from previous AHTF research, the goal of this project was to examine the possibility of successful reintroduction of AHTF and inform best practices for future AHTF reintroduction efforts.

Several objectives were central to achieving the overall goal of the project:

1. Identify at least three suitable reintroduction sites with habitat characteristics that are

associated with AHTF occurrences following Cinquemani Kuehn and Leopold (1993).

2. Transplant at least 500 laboratory-propagated AHTF of five transplant-types at each

reintroduction site.

3. Determine the effect of transplant-type, vigor, and acclimatization method on the survival

and growth of transplants within reintroduction sites.

4. Characterize the microclimate conditions at reintroduction sites, including temperature,

relative humidity, solar radiation, and wind speed and direction.

Several reintroduction sites were identified and evaluated prior to transplanting.

Ultimately, three reintroduction sites were chosen after considering aspects of habitat similarity, potential for genetic connectivity with extant populations, land use management, and historical importance. The three sites selected for this project were deemed suitable to the overall goal of

89 the project, although, many other potential reintroduction sites were considered and may be important to future management projects (Table 4.1). Candidate sites were identified using an

Element Distribution Model (EDM) of suitable AHTF habitat in New York State (New York

Natural Heritage Program 2012) and with historical records of AHTF occurrence. While a recommendation was made for each evaluated site, it should be noted that land owner permission and proper permits should be obtained prior to any further transplanting. Furthermore, visits to each evaluated site spanned from 2015-2017 and recommendations were made based on the conditions of the site at the time of evaluation. Further evaluations of suitability of each site should be conducted prior to any transplanting as the conditions of each site may have changed since the initial evaluation.

A total of 1,925 AHTF, representing five different transplant-types, were transplanted across three reintroduction sites in Onondaga County, New York. Forty six transplants were living at the end of a 24-month monitoring period, 6 of which reached maturity between 12 and

24 months after transplanting. Vigor and acclimatization method of transplants were found to be significant factors influencing the survival of AHTF transplants across reintroduction sites.

Generally, plants with better vigor at the time of transplanting were more likely to survive than those of less vigor. Sporophytes that were treated with both indoor and outdoor acclimatization methods (IOA sporophytes) had a significantly higher mean percent survival than those treated with less rigorous forms of acclimatization.

Table 4.1. Candidate sites for AHTF reintroduction and augmentation.

Evaluated Site name Location Land owner in field Recommended Justification Clark Clark Reservation High invasive species cover, limited Reservation State Park, limestone cobble, dryness, predominant (Eastern satellite) Jamesville, NY NYS OPRHP Yes No coniferous canopy Clark Reservation Large areas of suitable habitat, few East Sentinel State Park, invasive species, genetic significance Basin Jamesville, NY NYS OPRHP Yes Yes of extant population, protected area

Chittenango Moderate areas of suitable habitat, no Chips Trail Falls State invasive species, extant population AHTF Park, with low genetic diversity and small Population Cazenovia, NY NYS OPRHP Yes Yes population size, protected area Chittenango Moderate areas of suitable habitat, no Horseshoe Gorge Falls State invasive species, extant population AHTF Park, with low genetic diversity and medium Population Cazenovia, NY NYS OPRHP Yes Yes population size, protected area Chittenango Falls Colony Falls State Hazardous conditions (unstable talus), AHTF Park, public trail directly below site, high Population Cazenovia, NY NYS OPRHP Yes No invasive species cover nearby Oxbow Falls Moderate areas of suitable habitat, few County Park Madison invasive species (but high cover Oxbow Falls Canastota, NY County Parks Yes Yes nearby), protected area Mt. Hope City of Oneida Dryness, lack of limestone cobble (no Reservoir Oneida, NY Parks Yes No talus), clay-like soils, shallow slopes Canastota- Cazenovia State New York Trailway Canastota, NY State No n/a n/a

Whiskey Hollow Memphis, NY Public/Private No n/a n/a Perryville Falls Large areas of suitable habitat, few AHTF invasive species, historically small Population Sullivan, NY Private Yes Yes extant population Evergreen Lake 1 Moderate areas of suitable habitat, few AHTF invasive species, historically small Population Manlius, NY Private Yes Yes extant population Evergreen Lake 2 AHTF Population Manlius, NY Private Yes No High invasive species cover CNY Land Large areas of suitable habitat, few Three Falls Trust/Town of invasive species, geographic proximity Woods Manlius, NY Manlius Yes Yes to extant populations, protected area

The use of life stages preceding that of the immature sporophyte life stage have not been previously studied in regard to fern reintroduction projects. The use of protonema and gametophyte life stages of AHTF in the reintroduction project described in this thesis represent novel methods in the field of fern reintroduction. Furthermore, previously documented fern reintroduction projects lack analyses comparing the survival and growth of transplants of multiple life stages. Protonemata were unsuccessful at establishing across all reintroduction sites.

There was no significant difference in mean percent survival between gametophytes, sporelings, and IA sporophytes (treated with only indoor acclimatization) at 12 months after transplanting and beyond. However, some gametophytes at the Twin Basins reintroduction site successfully produced numerous sporelings, indicating that there is at least some potential for the use of gametophytes in future AHTF reintroduction projects. Across all reintroduction sites, 7 individuals each of the sporeling and IA sporophyte transplant-types survived the duration of the

24-month monitoring period. Survival of less rigorously acclimatized transplants suggests that reintroduction of these transplant-types may be possible, although, increased care of transplants post-reintroduction may be necessary to achieve greater survival. Many previous plant reintroduction projects and meta-analyses report that use of more advanced life stages significantly improves the likelihood of successful establishment of transplants (Drayton &

Primack 2000; Jusaitis et al. 2004; Maschinski & Duquesnel 2007; Godefroid et al. 2011;

Albrecht & Maschinski 2012; Dalrymple et al. 2012). The results of this research project did not support the hypothesis that more advanced life stages would exhibit higher survival than less advanced life stages. Therefore, it is possible that the application of results that are largely based on knowledge of seed plant reintroductions may not be entirely applicable to AHTF or fern reintroductions in general.

The use of both indoor and outdoor methods of acclimatization was successful overall, exhibiting a mean percent survival of IOA sporophytes of 27.18 % (±6.96). Acclimatization methods should be further improved in future AHTF reintroduction attempts. Earlier acclimatization of laboratory-propagated AHTF to greenhouse conditions, beginning at the sporeling stage, should be explored in future research projects. Exposing laboratory-propagated sporelings to wider daily ranges of temperature, humidity, and simulated wind events that are consistent with those reported in this and previous AHTF climate studies is recommended as a pre-treatment to outdoor acclimatization. Earlier acclimatization that better simulates ambient conditions of AHTF habitats could help to mitigate some of the physio-anatomical issues associated with mortality of transplants, including insufficient cuticle formation that contributes to transpirational water loss. The addition of a transitional acclimatization period just prior to outdoor acclimatization is also recommended. A transitional period, perhaps through the use of a shade tunnel, will further help propagated AHTF adjust to ambient environmental conditions prior to full outdoor acclimatization. Practitioners of a successful reintroduction of the endangered fern Woodsia ilvensis used a shade tunnel to acclimatize transplants prior to reintroduction (McHaffie 2006). The practitioners of the W. ilvensis reintroduction also partnered with a local botanic garden for propagation of transplants. Future AHTF reintroduction projects should consider partnering with a botanic garden to assist with larger-scale propagation of AHTF sporophytes for reintroduction. Botanic gardens offer additional expertise and facilities for threatened plant propagation and promote public awareness and involvement in plant reintroduction projects. AHTF has historically been propagated at the Brooklyn Botanic Garden

(BBG), where interpretive displays featuring AHTF are viewed by nearly a million visitors each

93 year. Larger-scale propagation of AHTF at the BBG or similar facility should be further explored as an additional means to produce high quality AHTF sporophytes for reintroduction.

Microclimate conditions at each reintroduction site were generally similar, however, a few differences between sites helped provide possible explanations some minor differences in survival between sites. Higher wind speeds at Split Rock and higher solar radiation at Sentinel

Basin might explain the 100% mortality of gametophytes observed at these sites. It is possible that desiccation from solar radiation at the outer edges of the talus at Sentinel Basin inhibits the germination of spores, which may help to explain why the extant population is densely clumped near the center of the talus and few individuals are distributed at the outer edges. A situation similar to that of Sentinel Basin may also be occurring at Split Rock, as otherwise suitable talus on either side of the population remains unoccupied. Higher humidity and lower wind at Twin

Basins might explain the higher mean percent survival of most transplant-types at that site.

Interestingly, no AHTF have been historically documented at Twin Basins, which performed slightly better as a site than Split Rock and Sentinel Basin. Very low wind speeds were recorded across all reintroduction sites, which is consistent with previous observations (Cinquemani

Kuehn & Leopold 1993). Low wind seems like an inhibiting factor for spore dispersal and achieving the goal of promoting genetic connectivity between AHTF populations. Interestingly, the possibility of additional mechanisms of spore dispersal exists for AHTF, including endozoochory by slugs. However, slug endozoochory is likely to occur on a very limited spatial scale.

In situ conservation of AHTF should remain the priority management goal for the species, including invasive species management and continued monitoring of population demography through periodic censuses. This research project, however, has demonstrated that

94 successful reintroduction of AHTF is possible in the short-term and may be a useful tool in recovering the species throughout its range. While recruitment of new generations of AHTF produced by the surviving transplants was not observed during the time frame of this study, the maturation of several transplants shows the potential for recruitment of future generations. The surviving AHTF transplants, including potential new recruitment of individuals, should continue to be monitored at least once a year. Ferns from the three reintroduction sites should also be augmented with additional AHTF transplants over time to enhance the results of this initial reintroduction.

Literature cited

Brito G, Costa A, Coelho C, Santos C. 2009. Large-scale field acclimatization of Olea maderensis micropropagated plants: morphological and physiological survey. Trees 23(5): 1019- 1031.

Drayton B, Primack RB. 2000. Rates of success in the reintroduction by four methods of several perennial plant species in eastern Massachusetts. Rhodora 102(911): 299-331.

McHaffie H. 2006. A reintroduction programme for Woodsia ilvensis (L.) R. Br. in Britain. Botanical Journal of Scotland 58(1): 75-80. New York Natural Heritage Program, New York State Department of Environmental Conservation. 2012. Element Distribution Models. Albany, New York.

APPENDIX A: Census Transplant Rediscovery

Table A.1. Summary of number of rediscovered AHTF transplant identification (ID) tags by transplant-type and site.

Number of initial Number of ID tags rediscovered (%) transplants Transplant-type | Site Jul 2015 Oct 2015 Jul 2016 Oct 2016 Jul 2017 Sporelings Sentinel Basin 93 71 (76.3) 76 (81.7) 76 (81.7) 76 (81.7) Twin Basins 88 79 (89.8) 81 (92.0) 81 (92.0) 80 (90.9) Split Rock 87 71 (81.6) 78 (89.7) 77 (88.5) 77 (88.5)

IA Sporophytes Sentinel Basin 42 35 (83.3) 36 (85.7) 36 (85.7) 36 (85.7) Twin Basins 44 38 (86.4) 37 (84.1) 37 (84.1) 37 (84.1) Split Rock 37 29 (78.4) 31 (83.8) 31 (83.8) 31 (83.8)

IOA Sporophytes Sentinel Basin 17 14 (82.4) 13 (76.5) 13 (76.5) 13 (76.5) Twin Basins 30 27 (90.0) 27 (90.0) 27 (90.0) 27 (90.0) Split Rock 55 51 (92.7) 51 (92.7) 51 (92.7) 51 (92.7)

All transplant-types Sentinel Basin 152 120 (78.9) 125 (82.2) 125 (82.2) 125 (82.2) Twin Basins 162 144 (88.9) 145 (89.5) 145 (89.5) 144 (88.9) Split Rock 179 151 (84.4) 160 (89.4) 159 (88.8) 159 (88.8)

APPENDIX B: Seasonal Climate Profiles

Mean Daily Max Mean Daily 156.30 (126.20) 96.10(86.92) 68.88(50.53) n/a 55.86(26.14) 60.47(30.30) 52.13(21.28) 48.92(20.25) 36.17(13.47)

1 (24.32) 1

62.73 (28.67) 62.73 Max Mean Daily 153.11 (83.80) 95.06(67.40) 61.19(35.68) 33.34(13.27) 42.30(18.18) 74.12(36.04) 94.99(81.39) 48.6

December2015.

Solar Radiation Solar(Watts/m2) Radiation

September2015.

Mean Daily 13.33(8.63) 7.13 (4.13) 6.65 (3.90) n/a 8.57 (4.17) 8.76 (4.68) 8.12 (3.39) 7.74 (3.03) 5.83 (2.35)

Solar Radiation Solar(Watts/m2) Radiation

8.77(3.09) Mean Daily 14.74(5.45) 8.74 (3.60) 6.56 (2.50) 6.41 (3.53) 7.31 (2.72) 10.96(3.74) 10.30(3.53) 7.98 (2.62)

(0.060)

Mean Daily Max Mean Daily 1.010(0.065) 0.990(0.074) 0.975(0.078) 1.009(0.057) 1.008(0.067) 0.994(0.067) 1.011 1.002(0.060) 0.969(0.068)

014 (0.037) 014

1.027 (0.029) 1.027 Max Mean Daily 1.030(0.034) 1.004(0.046) 0.997(0.047) 1.033(0.025) 1.013(0.038) 1. 1.001(0.039) 0.975(0.047)

115)

Relative Humidity Relative

Mean Daily Min Mean Daily 0.705(0.150) 0.676(0.161) 0.663(0.149) 0.701(0.130) 0.673(0.140) 0.664(0.138) 0.701(0.139) 0.699(0.139) 0.642(0.146)

Relative Humidity Relative

0.665 (0. 0.665 Min Mean Daily 0.724(0.124) 0.667(0.120) 0.633(0.121) 0.687(0.110) 0.646(0.112) 0.651(0.117) 0.647(0.118) 0.588(0.117)

Mean Daily 0.886(0.096) 0.855(0.104) 0.834(0.105) 0.877(0.084) 0.859(0.095) 0.848(0.097) 0.874(0.093) 0.867(0.093) 0.823(0.098)

0.883 (0.064) 0.883 Mean Daily 0.912(0.068) 0.865(0.075) 0.842(0.078) 0.900(0.060) 0.861(0.069) 0.865(0.069) 0.852(0.073) 0.809(0.075)

Mean Daily Max Mean Daily 11.74(5.31) 12.09(5.31) 11.63(5.18) 10.93(5.18) 11.12(5.35) 11.36(5.34) 11.07(5.14) 11.11(5.10) 11.59(5.26)

23.61 (3.29) 23.61 Max Mean Daily 23.44(3.23) 23.70(3.33) 23.80(3.33) 23.27(3.18) 23.87(3.38) 23.76(3.25) 23.79(3.27) 24.55(3.45)

)

Temperature (°C) Temperature

Mean Daily Min Mean Daily 3.60 (4.45) 4.04 (4.46) 3.99 (4.56) 3.33 (4.49) 3.27 (4.70) 3.71 (4.69) 3.83 (4.39) 3.99 (4.44) 4.30 (4.49)

Temperature (°C) Temperature

15.02 (3.53) 15.02 Min Mean Daily 15.19(3.25 15.35(3.36) 15.27(3.34) 14.94(3.34) 15.51(3.53) 15.33(3.20) 15.60(3.19) 15.72(3.29)

(4.45)

Mean Daily 7.42 7.82 (4.53) 7.80 (4.56) 7.01 (4.56) 7.12 (4.72) 7.42 (4.73) 7.33 (4.47) 7.42 (4.47) 7.83 (4.56)

19.09 (3.19) 19.09 Mean Daily 18.96(2.82) 19.39(3.01) 19.44(3.02) 18.84(3.05) 19.50(3.24) 19.38(2.91) 19.54(2.90) 19.95(3.04)

Mid Mid Mid

2015

Upper Upper Upper

Lower Lower Lower

Upper Upper Upper

Lower Lower Lower

Split Rock Split Autumn 2015Autumn Slope Site | Sentinel Basin Basins Twin

Summer Slope Site | Sentinel Basin Split Rock Basins Twin

Table B.2. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by slope position slope for Seasonal climate reintroduction (SD) mean of by sites October AHTF of B.2. profile Table Table B.1. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by slope position slope for Seasonal climate reintroduction (SD) mean of by sites July AHTF of B.1. profile Table

n/a Max Mean Daily 417.10 (239.60) 361.30 (242.50) 190.10 (197.10) n/a 256.00 (232.20) 286.30 (177.10) 246.60 (208.00) 143.00 (113.50)

Max Mean Daily 139.60 (102.90) 152.55 (89.86) 75.35(47.98) n/a 60.71(40.29) 80.86(58.43) 104.11 (67.10) 87.63(49.98) 42.13(12.43)

Solar Radiation Solar(Watts/m2) Radiation

June2016.

Solar Radiation (Watts/m2) SolarRadiation

September2016.

n/a Mean Daily 48.31(36.62) 38.35(29.07) 21.64(16.49) n/a 43.04(36.69) 35.24(23.63) 34.95(24.30) 20.30(11.46)

Mean Daily 12.86(5.31) 13.98(6.04) 8.53 (3.81) n/a 9.07 (3.07) 10.09(3.53) 11.02(4.45) 10.87(3.84) 9.42 (3.23)

n/a Max Mean Daily 0.959(0.067) 0.936(0.077) 0.930(0.076) n/a 0.931(0.079) 0.952(0.078) 0.936(0.091) 0.912(0.086)

Max Mean Daily 1.002(0.044) 0.983(0.056) 0.972(0.057) 0.939(0.038) 0.990(0.057) 0.965(0.060) 0.990(0.058) 0.980(0.062) 0.931(0.053)

positionfor March

59)

167)

(0.125)

Relative Humidity Relative

n/a Min Mean Daily 0.469(0.176) 0.475(0.172) 0.463(0.161) n/a 0.440(0.1 0.497(0.174) 0.467(0. 0.453(0.157)

Min Mean Daily 0.656(0.145) 0.636(0.136) 0.590(0.126) 0.625(0.119) 0.621(0.127) 0.590(0.127) 0.664(0.141) 0.603(0.131) 0.570

n/a Mean Daily 0.731(0.124) 0.711(0.129) 0.699(0.123) n/a 0.689(0.128) 0.731(0.134) 0.704(0.139) 0.684(0.133)

Mean Daily 0.862(0.089) 0.832(0.093) 0.804(0.089) 0.814(0.069) 0.840(0.084) 0.804(0.088) 0.851(0.096) 0.816(0.095) 0.777(0.086)

(3.41)

2 (3.71)2

Max Mean Daily 24.81(3.52) 24.79(3.36) 24.96(3.51) 24.49(3.64) 24.7 25.15(3.66) 24.34 25.12(3.59) 25.41(3.55)

n/a Max Mean Daily 19.38(7.87) 18.98(7.50) 18.44(7.55) n/a 18.34(7.96) 18.21(7.61) 18.34(7.83) 18.25(7.85)

)

Temperature (°C) Temperature

Min Mean Daily 15.86(3.88) 16.05(4.00) 15.91(4.09) 15.56(4.01) 15.60(4.11) 16.26(4.02) 16.36(3.66) 16.27(3.91) 16.61(3.85)

n/a Min Mean Daily 7.27 (6.90) 7.50 (6.92) 7.35 (6.94 n/a 7.32 (7.36) 7.71 (6.87) 7.76 (7.00) 8.05 (6.95)

(3.37)

Mean Daily 19.99(3.18) 20.40( 3.25) 20.45 19.84(3.49) 20.00(3.55) 20.61(3.47) 20.27(3.19) 20.59(3.33) 20.88(3.28)

n/a Mean Daily 12.93(6.64) 13.27(6.77) 13.01(6.93) n/a 12.94(7.28) 12.92(6.96) 13.10(7.13) 13.26(7.11)

Mid Mid Mid

Upper Upper Upper

Lower Lower Lower

Upper Upper Upper

Lower Lower Lower

Spring 2016Spring Slope Site | Sentinel Basin Split Rock Basins Twin

Site | Slope | Site Summer 2016Summer Sentinel Basin Split Rock Basins Twin

Table B.3. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by slope slope Seasonal climate reintroduction (SD) mean of by sites AHTF of B.3. profile Table Table B.4. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by slope position slope for Seasonal climate reintroduction (SD) mean of by sites July AHTF of B.4. profile Table

(29.85) 49.14 Mean Daily Max Mean Daily 71.57(66.99) 109.90 (114.20) 62.91(57.60) n/a 53.52(31.66) 39.02(22.24) 43.99(26.26) 29.30(16.12)

Mean Daily Max Mean Daily 141.60 (129.90) 131.50 (122.20) 123.10 (111.80) n/a 158.40 (123.40) 138.10 (100.90) n/a n/a 44.70(30.42)

98)

.75)

Solar Radiation Solar(Watts/m2) Radiation

Solar Radiation (Watts/m2) SolarRadiation

December 2016. December

March 2017. March

6.94(4.50) Mean Daily 6.84 (5.24) 8.89 (8.54) 5.82 (4.63) n/a 7.91 (4.89) 5.57 (3.64) 6.26 (3. 4.44 (2.81)

Mean Daily 16.68(14.43) 11.29(9.98) 12.78(9.88) n/a 25.69(25.69) 23.10(17.59) n/a n/a 7.70 (6

(0.05 0.960 Mean Daily Max Mean Daily 1.010(0.045) 1.002(0.053) 0.992(0.063) 0.946(0.044) 0.999(0.054) 1.018(0.038) 0.999(0.055) 0.966(0.055)

Mean Daily Max Mean Daily 0.971(0.051) 0.952(0.051) 0.962(0.061) 0.931(0.045) 0.969(0.050) 0.934(0.048) 0.996(0.037) 0.947(0.045) 0.957(0.046)

ope position ope for January

Relative Humidity Relative

(0.135) 0.704 Mean Daily Min Mean Daily 0.764(0.157) 0.755(0.161) 0.713(0.154) 0.724(0.125) 0.724(0.140) 0.806(0.128) 0.749(0.144) 0.721(0.145)

Relative Humidity Relative

Mean Daily Min Mean Daily 0.704(0.163) 0.673(0.157) 0.669(0.156) 0.692(0.149) 0.668(0.161) 0.653(0.153) 0.766(0.135) 0.703(0.144) 0.683(0.155)

(0.092) 0.851 Mean Daily 0.910(0.090) 0.899(0.095) 0.871(0.105) 0.854(0.078) 0.885(0.097) 0.931(0.077) 0.892(0.095) 0.862(0.095)

Mean Daily 0.852(0.103) 0.830(0.099) 0.823(0.108) 0.823(0.091) 0.833(0.105) 0.808(0.096) 0.893(0.084) 0.838(0.090) 0.832(0.100)

8.83(7.23) Mean Daily Max Mean Daily 8.61 (7.06) 9.11 (7.31) 8.97 (7.21) 8.46 (7.11) 8.56 (7.21) 8.66 (6.74) 8.62 (7.11) 8.96 (7.10)

Mean Daily Max Mean Daily 3.54 (6.23) 4.51 (6.73) 3.67 (6.42) 2.76 (6.49) 3.16 (6.67) 3.46 (6.68) 2.89 (5.69) 3.07 (6.11) 3.43 (6.08)

5.06)

Temperature (°C) Temperature .80 (6.09)

Temperature (°C) Temperature

1.52(6.55) Mean Daily Min Mean Daily 1.34 (6.35) 1 1.65 (6.27) 1.27 (6.2 1.09 (6.52) 2.20 (6.09) 1.58 (6.28) 2.01 (6.24)

4.49 (5.55) 4.03 (5.79) 4.19 (5.66) 4.65 (5.66) 4.80 (5.8 4.56 (5.84) 3.88 ( 4.23 (5.42) 3.91 (5.38)

Mean Daily Min Mean Daily ------

5.14(6.54) Mean Daily 4.95 (6.32) 5.36 (6.31) 5.33 (6.45) 4.86 (6.35) 4.79 (6.53) 5.40 (6.21) 5.10 (6.45) 5.45 (6.42)

0.45 (5.47) 0.13 (5.75) 0.91 (5.65) 0.81 (5.85) 0.53 (5.83) 0.48 (5.08) 0.55 (5.48) 0.194 (5.47)

Mean Daily - 0.00 (5.81) ------

Mid Mid Mid

Upper Upper Upper

Lower Lower Lower

Upper Upper Upper

Lower Lower Lower

Basin

Site | Slope | Site Basins Twin Winter 2017Winter Sentinel Split Rock

Autumn 2016 Autumn Site | Slope Site | Sentinel Basin Split Rock Basins Twin

Table B.6. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by sl Seasonal climate reintroduction (SD) mean of by sites AHTF of B.6. profile Table Table B.5. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by slope position for October for position slope by sites reintroduction AHTF of (SD) mean of profile climate Seasonal B.5. Table 100

00 00 (159.70)

71.35 (40.20) 71.35 Mean Daily Max Mean Daily 162.50 (143.30) n/a n/a 71.77(35.80) 67.63(40.58) n/a n/a 41.40(14.22)

196.80 (169.90) 196.80 Mean Daily Max Mean Daily 383.20 (238.70) n/a 159. n/a 211.30 (174.40) n/a n/a 121.51 (93.59)

Solar Radiation Solar(Watts/m2) Radiation

1.58 (4.35)1.58

June2017.

7.84(3.44) Mean Daily 14.12(6.96) n/a n/a 1 9.51 (3.63) n/a n/a 8.78 (3.01)

September2017.

33.67 (29.62) 33.67 Mean Daily 43.91(32.97) n/a 18.43(14.25) n/a 37.35(29.89) n/a n/a 18.72(9.85)

0.943 (0.027) 0.943 Mean Daily Max Mean Daily 0.934(0.030) 0.932(0.029) 0.915(0.012) 0.955(0.021) 0.930(0.028) 0.961(0.021) 0.919(0.024) 0.929(0.025)

0.915 (0.061 0.915 Mean Daily Max Mean Daily 0.942(0.043) 0.927(0.049) 0.943(0.050) 0.904(0.056) 0.952(0.055) 0.963(0.041) 0.919(0.055) 0.932(0.054)

(0.114)

es by slope position for April for position slope by es

Relative Humidity Relative

0.652 Mean Daily Min Mean Daily 0.685(0.105) 0.673(0.107) 0.678(0.088) 0.687(0.109) 0.663(0.106) 0.730(0.114) 0.657(0.109) 0.634(0.124)

Relative Humidity Relative

0.585 (0.177) 0.585 Mean Daily Min Mean Daily 0.608(0.189) 0.612(0.186) 0.610(0.176) 0.565(0.178) 0.607(0.188) 0.666(0.183) 0.608(0.167) 0.609(0.181)

0.831 (0.060) 0.831 Mean Daily 0.840(0.056) 0.831(0.056) 0.829(0.041) 0.860(0.050) 0.828(0.054) 0.876(0.052) 0.817(0.055) 0.818(0.062)

0.767 (0.116) 0.767 Mean Daily 0.800(0.107) 0.785(0.112) 0.789(0.115) 0.752(0.114) 0.801(0.120) 0.834(0.106) 0.776(0.109) 0.784(0.118)

23.08 (3.66) 23.08 Mean Daily Max Mean Daily 23.13(3.65) 23.06(3.60) 21.89(3.43) 22.85(3.63) 23.06(3.71) 22.58(3.47) 23.20(3.71) 23.50(3.73)

18.85 (6.25) 18.85 Mean Daily Max Mean Daily 19.42(6.39) 19.22(6.15) 18.60(5.96) 16.36(6.24) 18.69(6.20) 18.37(5.83) 18.71(6.01) 18.59(5.96)

rature (°C)

Temperature (°C) Temperature

Tempe

14.62 (3.50) 14.62 Mean Daily Min Mean Daily 14.80(3.27) 14.78(3.44) 13.77(3.52) 14.28(3.52) 14.75(3.54) 14.95(3.16) 14.80(3.39) 15.09(3.35)

9.15(5.74) Mean Daily Min Mean Daily 8.96 (5.53) 9.26 (5.61) 9.19 (5.70) 5.77 (4.57) 8.70 (5.66) 9.18 (5.41) 9.20 (5.49) 9.55 (5.52)

85 (5.64) 85

18.83 (3.12) 18.83 Mean Daily 18.56(2.92) 18.87(3.04) 17.67(3.09) 18.40(3.17) 18.77(3.20) 18.68(2.95) 18.92(3.10) 19.20(3.09)

13. Mean Daily Mean Daily 13.82(5.29) 14.14(5.37) 13.96(5.51) 10.99(4.89) 13.61(5.56) 13.68(5.30) 13.90(5.46) 14.08(5.46)

Mid Mid Mid

Upper Upper Upper

Lower Lower Lower

Upper Upper Upper

Lower Lower Lower

ummer 2017ummer

Spring 2017Spring Slope Site | Sentinel Basin Split Rock Basins Twin S Slope Site | Sentinel Basin Split Rock Basins Twin

Table B.7. Seasonal climate profile of mean (SD) of AHTF reintroduction sit reintroduction AHTF of (SD) mean of profile climate Seasonal B.7. Table Table B.8. Seasonal climate profile of mean (SD) of AHTF reintroduction sites by slope position slope for Seasonal climate reintroduction (SD) mean of by sites July AHTF of B.8. profile Table 101

APPENDIX C: Wind Radar Data

Table C.1. Wind frequency and speed data for Sentinel Basin lower slope from 2015-2017.

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 3.2% 0.14 0.39 NNE 4.0% 0.20 0.52 NE 7.7% 0.17 0.47 ENE 11.6% 0.15 0.43 E 13.7% 0.16 0.44 ESE 13.2% 0.13 0.37 SE 14.7% 0.14 0.37 SSE 11.9% 0.13 0.35 S 6.7% 0.13 0.34 SSW 3.6% 0.11 0.31 SW 2.0% 0.10 0.31 WSW 2.1% 0.08 0.27 W 1.6% 0.07 0.25 WNW 1.2% 0.10 0.31 NW 1.5% 0.11 0.34 NNW 1.3% 0.16 0.46

Table C.2. Wind frequency and speed data for Sentinel Basin mid-slope from 2015-2017.

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 9.4% 0.10 0.29 NNE 11.3% 0.10 0.31 NE 15.9% 0.12 0.33 ENE 12.3% 0.12 0.32 E 7.2% 0.11 0.32 ESE 3.2% 0.11 0.29 SE 2.1% 0.14 0.33 SSE 2.8% 0.15 0.33 S 3.5% 0.11 0.28 SSW 7.9% 0.11 0.29 SW 3.4% 0.07 0.22 WSW 2.5% 0.06 0.22 W 3.4% 0.05 0.20 WNW 4.5% 0.05 0.20 NW 5.0% 0.08 0.25 NNW 5.7% 0.09 0.28

102

Table C.3. Wind frequency and speed data for Sentinel Basin upper slope from 2015-2017.

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 2.6% 0.08 0.31 NNE 3.8% 0.04 0.20 NE 19.2% 0.09 0.33 ENE 23.8% 0.12 0.42 E 3.4% 0.15 0.50 ESE 2.4% 0.06 0.26 SE 3.3% 0.03 0.18 SSE 6.0% 0.02 0.14 S 6.1% 0.03 0.17 SSW 6.6% 0.04 0.20 SW 3.8% 0.05 0.25 WSW 5.7% 0.07 0.30 W 7.3% 0.09 0.35 WNW 2.5% 0.05 0.24 NW 1.9% 0.03 0.20 NNW 1.6% 0.03 0.18

Table C.4. Wind frequency and speed data for Split Rock lower slope from 2015-2017.

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 5.8% 0.26 0.60 NNE 4.3% 0.18 0.46 NE 4.6% 0.17 0.43 ENE 4.5% 0.15 0.39 E 4.6% 0.12 0.35 ESE 3.2% 0.16 0.42 SE 4.6% 0.16 0.44 SSE 6.1% 0.16 0.45 S 5.4% 0.24 0.60 SSW 9.2% 0.28 0.64 SW 18.1% 0.38 0.79 WSW 15.0% 0.37 0.77 W 5.7% 0.30 0.68 WNW 3.2% 0.23 0.56 NW 2.8% 0.18 0.46 NNW 2.8% 0.16 0.43

103

Table C.5. Wind frequency and speed data for Split Rock mid-slope from 2015-2017.

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 4.4% 0.26 0.58 NNE 2.9% 0.15 0.39 NE 3.5% 0.12 0.34 ENE 2.7% 0.13 0.35 E 2.7% 0.19 0.44 ESE 3.3% 0.22 0.49 SE 3.4% 0.18 0.44 SSE 3.3% 0.16 0.42 S 2.9% 0.26 0.61 SSW 6.0% 0.24 0.55 SW 12.2% 0.28 0.62 WSW 17.8% 0.35 0.74 W 15.5% 0.33 0.70 WNW 7.9% 0.18 0.45 NW 7.2% 0.25 0.57 NNW 4.5% 0.18 0.45

Table C.6. Wind frequency and speed data for Split Rock upper slope for 2015 (2016-2017 data missing).

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 1.7% 0.20 0.46 NNE 2.3% 0.08 0.26 NE 2.0% 0.13 0.35 ENE 1.4% 0.17 0.44 E 3.0% 0.11 0.33 ESE 2.9% 0.16 0.43 SE 4.8% 0.13 0.38 SSE 3.9% 0.17 0.42 S 5.5% 0.14 0.36 SSW 7.5% 0.12 0.32 SW 15.1% 0.15 0.36 WSW 17.1% 0.18 0.42 W 13.5% 0.29 0.61 WNW 11.0% 0.30 0.64 NW 5.6% 0.37 0.76 NNW 2.5% 0.23 0.54

104

Table C.7. Wind frequency and speed data for Twin Basins mid-slope from 2015-2017.

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 2.7% 0.12 0.35 NNE 1.1% 0.14 0.40 NE 1.2% 0.12 0.34 ENE 1.2% 0.11 0.31 E 1.5% 0.13 0.36 ESE 3.5% 0.14 0.39 SE 8.7% 0.12 0.37 SSE 9.9% 0.12 0.36 S 10.8% 0.12 0.36 SSW 9.1% 0.12 0.36 SW 10.7% 0.12 0.34 WSW 11.2% 0.11 0.33 W 8.3% 0.12 0.33 WNW 8.4% 0.11 0.34 NW 8.0% 0.11 0.35 NNW 3.7% 0.12 0.37

Table C.8. Wind frequency and speed data for Twin Basins lower slope from 2015-2016 (2017 data missing).

Direction Frequency Average wind speed (m/s) Average gust speed (m/s) N 6.1% 0.09 0.29 NNE 5.5% 0.10 0.30 NE 8.4% 0.10 0.29 ENE 9.7% 0.11 0.31 E 18.1% 0.09 0.29 ESE 14.1% 0.11 0.33 SE 13.8% 0.10 0.30 SSE 4.3% 0.09 0.29 S 2.2% 0.09 0.29 SSW 1.3% 0.07 0.25 SW 1.1% 0.07 0.25 WSW 1.0% 0.09 0.27 W 2.8% 0.05 0.21 WNW 3.0% 0.06 0.22 NW 4.9% 0.05 0.20 NNW 3.8% 0.07 0.24

105

Resume Michael Serviss 701 Northcliffe Road #5, Syracuse, NY 13206 Phone: 315-272-7598 ♦ E-mail: [email protected]

EDUCATION

SUNY College of Environmental Science and Forestry M.S. in Conservation Biology (August 2014 – December 2017) GPA: 3.550

B.S. in Conservation Biology (Spring 2014) GPA: 3.802

Mohawk Valley Community College A.S. in Liberal Arts (December 2010) GPA: 3.26

Math & Science Coursework (August 2011 – May 2012) GPA: 3.34

University of North Carolina at Charlotte B.A. classes in Liberal Arts (August 2003 – June 2004) GPA: 3.222

WORK EXPERIENCE

Conservation Project Coordinator NYS Office of Parks, Recreation and Historic Preservation (May 2015 - Present)

Seasonal. Leading and supervising a team of employees and volunteers in invasive species management at Clark Reservation and Chittenango Falls State Parks. Creating and organizing GIS maps and data management systems. Providing recommendations of management actions including herbicide application, mechanical removal, and preventative measures. Managing natural resource conservation projects. Leading guided tours and interpretive programs. Regularly participating in inter-departmental meetings and collaborations on a variety of conservation and stewardship projects.

Operations Supervisor Adirondack Scenic Railroad, Utica, NY (May 2013 – May 2015)

Supervised five employees in repairing and maintaining the Adirondack Scenic Railroad’s fleet of vintage passenger coaches. Organized and supervised special events. Coordinated volunteer based service and restoration projects.

Note: Worked an average of 25 hours per week while attending college

106

Assistant Utica Station Manager (September 2012 – May 2013) Interim Utica Station Manager (July 2012 –September 2012) Ticket Agent & Customer Service Representative (Aug. 2011 – June 2012) Adirondack Scenic Railroad, Utica, NY

Customer service, file management, and tour planning. Handled local and regional corporate marketing. Updated and maintained website and online sales systems. Coordinated charter trains and regularly communicated with the maintenance department and other station locations.

Note: Worked an average of 25 hours per week while attending college

Quality Assurance Representative ConMed Corporation, Utica, NY (Jan. 2010 –Aug. 2011)

Assisted customers with after-market quality concerns. Maintained and created Excel databases of quality-control records. Facilitated return shipments of goods. Prepared data, graphs, and reports for presentations.

TEACHING EXPERIENCE

Graduate Teaching Assistant SUNY College of Environmental Science and Forestry (August 2014 – June 2017)

Lead two lab sections per semester. Disseminated course materials, organized lab activities, graded tests, quizzes and term projects. Coordinated student learning and student support outside of scheduled lab time. Assisted professors in organizing class materials, proctoring exams, and creating curriculum based activities and lectures. Courses taught: Diversity of Plants, Mycology, Plant Anatomy and Development, General Biology II.

RESEARCH EXPERIENCE

M.S. Thesis Research SUNY College of Environmental Science and Forestry (August 2014 – December 2017)

Supervised an experimental reintroduction of Asplenium scolopendrium var. americanum (American hart’s-tongue fern) including fern propagation, greenhouse and field work supervision, GIS habitat analysis, and data collection and analysis.

Research Assistant SUNY College of Environmental Science and Forestry (August 2016 – December 2016)

Identifying and evaluating critical habitat for the rare and threatened fern Asplenium scolopendrium var. americanum, transplanting lab-propagated plants, monitoring and data collection related to transplants

107

RELATED EXPERIENCE

Environmental Education & Recreation Intern Adirondack Scenic Railroad (May 2013 – October 2013)

Created interpretive programs related to the Adirondack Railway Preservation Society’s organizational goals of promoting environmental education and conservation. Coordinated rail, trail, and outdoor recreation opportunities through research of flora and fauna along rail corridor. Created and designed informational trip-guides and interpretive posters.

Education Department Docent, Animal Handler/Interpreter Utica Zoo (June 2010- May 2015)

Assisted zookeepers with animal care and handling used in educational programming and zoo-mobiles. Lead interpretive programs for zoo members and special groups. Participated in continuing training and education. Maintained zoo facilities.

Mohawk Valley Young Professionals Founding Member (June 2010- January 2012)

Helped found and participated in MVYP, a group aiming to connect young professionals with each other and other organizations, businesses, and opportunities in the Mohawk Valley through networking events, professional development, and community involvement and advocacy.

AWARDS

Josiah Lowe and Hugh Wilcox Scholarship Edna Bailey Sussman Foundation Fellowship SUNY ESF, Dean’s List SUNY ESF, President’s List Mohawk Valley Community College, Dean’s List University of North Carolina at Charlotte, Dean’s List

108