Species Status Assessment Report for the Mountain Doll’S Daisy (Boltonia Montana)

Home , Amblystegium serpens, Augusta (plant), Boltonia, Boltonia asteroides, Boltonia decurrens, Boltonia montana

Species Status Assessment Report for the Mountain Doll’s Daisy (Boltonia montana)

February 2020 Version 1.0

U.S. Fish and Wildlife Service North Atlantic-Appalachian Region Hadley, MA

Primary Contributors

• Keith Hastie – North Atlantic-Appalachian Region, Hadley, Massachusetts; Lead Biologist • Ron Popowski – North Atlantic-Appalachian Region, New Jersey Field Office; Assist Biologist • Heidi Crowell – California-Great Basin Region, Sacramento Regional Office; Species Assessment Team Project Manager

Contributors & Agency Reviewers (underlined)

We greatly appreciate the assistance of our conservation partners, who provided helpful information and review of the draft report:

• Tom Brumbelow – U.S. Forest Service, George Washington and Jefferson National Forests • Mike Donahue – U.S. Forest Service, George Washington and Jefferson National Forests • Gary Fleming – Virginia Department of Conservation and Recreation, Division of Natural Heritage • Rachel Goad – Western Pennsylvania Conservancy, Pennsylvania Natural Heritage Program • Troy Morris – U.S. Forest Service, George Washington and Jefferson National Forests • John Townsend – Virginia Department of Conservation and Recreation, Division of Natural Heritage • Kathleen Walz – New Jersey Department of Environmental Protection, Natural Heritage Program

Peer Reviewers

We also thank the following experts for reviewing the draft report and providing helpful comments:

• Walter Bien, Ph.D. – Drexel University, Academy of Natural Sciences • Gerry Moore, Ph.D. – U.S. Department of Agriculture, Natural Resources Conservation Service • Nancy Van Alstine – Virginia Department of Conservation and Recreation, Division of Natural Heritage (Retired)

SUGGESTED CITATION: U.S. Fish and Wildlife Service (Service). February 2020. Species Status Assessment Report for the Mountain Doll’s Daisy (Boltonia montana), Version 1.0. Department of the Interior Region: North Atlantic-Appalachian. Hadley, Massachusetts.

iii

Table of Contents EXECUTIVE SUMMARY ...... vii CHAPTER 1 - INTRODUCTION ...... 9 1.1 Purpose of the SSA ...... 9 1.2 Petition History ...... 9 1.3. State Listing Status ...... 9 1.4 Methodology and Data ...... 10 1.4.1 Species (Resource) Needs ...... 10 1.4.2 Current Species Condition ...... 11 1.4.3. Future Species Condition ...... 11 CHAPTER 2 - SPECIES INFORMATION ...... 12 2.1 Taxonomy ...... 13 2.2 Physical Description ...... 15 2.3 Habitat ...... 17 2.4 Life History ...... 20 2.4.1 Reproduction ...... 20 2.4.2 Seasonality...... 23 2.5 Population Structure ...... 23 2.6 Species (Resource) Needs ...... 24 2.6.1 Appropriate Soil Conditions...... 24 2.6.2 Adequate Water ...... 25 2.6.3 Adequate Sunlight ...... 26 2.6.4 Pollination Services ...... 26 2.6.5 Seasonal Temperature Regime ...... 26 2.6.6 Habitat Connectivity ...... 27 CHAPTER 3 – HISTORICAL AND CURRENT DISTRIBUTION AND ABUNDANCE .. 27 3.1 Historical Context ...... 27 3.2 Historical Distribution and Abundance ...... 28 3.2.1 New Jersey ...... 29 3.2.2 Pennsylvania ...... 30 3.2.3 Virginia ...... 30 3.3 Current Distribution and Abundance ...... 32 3.3.1 New Jersey ...... 33 3.3.2 Virginia ...... 36

CHAPTER 4 – FACTORS AFFECTING THE SPECIES ...... 39 4.1 Stressors Affecting Viability ...... 39 4.2 Habitat Modification ...... 39 4.2.1 Development...... 40 4.2.2 Agriculture ...... 42 4.2.3 Off-road Vehicle Use...... 43 4.2.4 Altered Surface Hydrology ...... 45 4.2.5 Groundwater Withdrawals ...... 45 4.3. Effects of Climate Change ...... 45 4.3.1 Altered Precipitation Patterns...... 46 4.3.2 Altered Temperature Patterns ...... 48 4.4 Potential Catastrophic Events ...... 50 4.5 Conservation Measures ...... 50 4.5.1 New Jersey ...... 50 4.5.2 Virginia ...... 50 CHAPTER 5 – CURRENT CONDITION ...... 51 5.1 Assessment Methodology...... 51 5.2 Site Assessment Metrics ...... 52 5.2.1 Pond Size ...... 52 5.2.2 Hydrology ...... 52 5.2.3 Proximity ...... 53 5.2.4 Disturbance ...... 54 5.2.5 Landscape Setting ...... 54 5.2.6 Conservation Measures ...... 54 5.3 Metric Criteria and Current Site Scores ...... 54 5.4 Determination of Current Population Status ...... 57 5.4.1 Unknowns and Assumptions ...... 59 5.5 Current Condition Summary ...... 59 CHAPTER 6 – FUTURE CONDITIONS ...... 65 6.1 Scenario 1 ...... 65 6.1.1 Potential Future Viability, Scenario 1 ...... 65 6.2 Scenario 2 ...... 68 6.2.1 Potential Future Viability, Scenario 2 ...... 71 6.3 Scenario 3 ...... 73

6.3.1 Potential Future Viability, Scenario 3 ...... 75 6.4 Summary of Future Conditions ...... 77 CHAPTER 7 – OVERALL SYNTHESIS ...... 79 REFERENCES CITED ...... 84 APPENDIX A. Taxonomic information demonstrating inconsistent recognition of Boltonia montana in various databases and taxonomic systems...... 94 APPENDIX B. Location and general characteristics of all known Boltonia montana population sites. Population ID numbers correspond to numbers on the relevant map following each table...... 96 APPENDIX C. Years indicating when Boltonia montana occurrence data are available, for the 30-year period 1980 to 2009...... 100 APPENDIX D. Historical and current survey data for Boltonia montana. Gray shading indicates no survey data available...... 101

EXECUTIVE SUMMARY

This report summarizes the results of a species status assessment (SSA) conducted for the plant Boltonia montana, which is known variously as “mountain doll’s daisy,” “valley doll’s daisy,” or “Appalachian mountain boltonia.” This report is intended to provide the biological support for determining whether or not to propose to list the species as an endangered or threatened species, and if so, whether or not to propose designating critical habitat. This SSA report does not represent a decision by us (the U.S. Fish and Wildlife Service (Service)) whether or not to list the species under the Endangered Species Act (Act).

Boltonia montana is a perennial plant in the Asteraceae family that is known from Augusta County, Virginia; Sussex and Warren Counties, New Jersey; and Dauphin County, Pennsylvania, the latter regarded as a historical occurrence. The species is known primarily from isolated depressional ponds in the Valley and Ridge physiographic province, though historical specimens from Pennsylvania were collected from riverine areas.

Boltonia montana population structure has not been studied. For purposes of this SSA, we consider each pond where the species has been confirmed to be a “population.” Some of these populations are in close proximity to each other, potentially allowing for genetic connectivity; we refer to these groups as “population areas.” We are uncertain of any relationship between population areas themselves or between other individual isolated populations, but herein we refer to the collective groups of populations in either New Jersey or Virginia as a “metapopulation” even though a strict definition of the term may not apply. We assume there is no natural connectivity between these two metapopulations.

The best available information, including survey results from the Virginia Department of Conservation and Recreation (VADCR), the New Jersey Department of Environmental Protection (NJDEP), and the U.S. Forest Service (USFS), confirm the species is extant at 5 of 21 known population sites in New Jersey and 11 of 22 known sites in Virginia. Current survey data are lacking or inconclusive for many other population sites, but if the habitat is intact we presume the species is extant. We estimate that two populations in New Jersey and five populations in Virginia have been extirpated.

At the scale of the individual plant or population site, we identified soil, water, sunlight, pollinator services, and a suitable annual temperature regime as interrelated resource needs. At the metapopulation scale, it is likely the species requires some degree of habitat connectivity to maintain viability; however, there is significant uncertainty regarding the degree of connectivity that may be necessary between population sites. Seasonally fluctuating water levels with periods of inundation are a key requirement for the species to persist at a population site.

We also identified a variety of stressors in two broad categories, habitat modification and the effects of climate change, that are likely to affect the viability of Boltonia montana. While we generally discuss the stressors individually, many of them may act together additively or synergistically to affect B. montana resiliency. We also considered certain conservation measures that may affect the species. Boltonia montana is listed as state endangered by both New Jersey and Virginia, which affords the species some protections on public lands in Virginia

vii

and helps guide conservation decisions in New Jersey. In New Jersey, 50 percent (11 of 22) of the known B. montana population sites are on land managed by the NJDEP or a conservation non-governmental organization. In Virginia, about 32 percent (7 of 22) of the known populations are on USFS land designated a “special biological area” (SBA), which is managed specifically for the conservation of rare species and their supporting habitats. Additionally, the VADCR is actively pursuing additional protections for at least two privately-owned population sites.

Data sufficient to directly assess the condition of the various Boltonia montana populations are sparse; therefore, we developed an assessment model using habitat metrics to compare population sites with each other, both currently and under three plausible future scenarios. We then summarize the current and potential future condition of the species using the conservation 3Rs—resiliency, redundancy, and representation. We project the potential condition of the species in about the year 2050. We selected this 30-year timeframe because we assume that our modeled changes in land use, development, or conservation measures will manifest their effects on B. montana populations within that timeframe. Additionally, we assume that uncertainties surrounding our assumptions will increase past that point in time.

Under future scenario 1, we project no significant changes are made to the activities currently affecting the extant populations. Under scenario 2, we project that public and private land managers will implement various measures that result in the protection or enhancement of Boltonia montana habitat, where feasible. And under scenario 3, we assume no new conservation measures are implemented and habitat disturbances increase at most privately- owned B. montana populations.

The model results indicate changes in the resiliency of the species at the population and population area scale. Under scenarios 1 and 3, by 2050, between 3 and 11 populations are predicted to have lower resiliencies than the current condition; under scenario 3, the changes may result in the extirpation of several low resiliency populations, perhaps causing a loss of redundancy. Under scenario 2, we predict feasible conservation efforts would improve the condition of 22 populations, significantly improving the species resiliency in three population areas. Under all scenarios, the species is projected to remain extant in the New Jersey and Virginia metapopulations; therefore, its representation is not predicted to change from the current condition (though we note that the historical extirpation of the Pennsylvania metapopulation may have reduced the species representation).

viii

CHAPTER 1 - INTRODUCTION

1.1 Purpose of the SSA

This report summarizes the results of a species status assessment (SSA) conducted for the plant Boltonia montana, which is known variously as “mountain doll’s daisy,” “valley doll’s daisy,” or “Appalachian mountain boltonia.” The SSA report, the product of conducting a SSA, is a concise review of the species’ biology and factors (both negative and positive) influencing the species, an evaluation of its biological status, and an analysis of the species’ potential status under plausible future scenarios. The SSA report is intended to support all functions of the Endangered Species Program, including the development of listing rules, recovery plans, and 5- year reviews, should the species warrant listing as an endangered or threatened species under the Act. The SSA report is a living document and we may update it periodically as new information becomes available. This SSA report for Boltonia montana (version 1.0) is intended to provide the biological support for determining whether or not to propose to list the species as an endangered or threatened species and if so, whether or not to propose designating critical habitat. The process and this SSA report do not represent a decision by us (the U.S. Fish and Wildlife Service (Service)) whether or not to list the species under the Act. Instead, this SSA report provides a review of the best scientific and commercial information available strictly related to the biological status of B. montana. We will make a listing decision after reviewing this document and all relevant laws, regulations, and policies, and will announce the decision in the Federal Register at a later date.

1.2 Petition History

We were petitioned by the Center for Biological Diversity (CBD) and others to list “Doll’s daisy” (Boltonia montana) as an endangered or threatened species under the Endangered Species Act of 1973, as amended (Act), as a part of a 2010 petition to list 404 aquatic, riparian, and wetland species in the southeastern United States (CBD 2010, pp. 569–573). On September 27, 2011, we found that the petition presented substantial scientific or commercial information indicating that listing B. montana may be warranted (76 FR 59836); substantial findings were made for the other species in this same Federal Register notice, although analyses and findings for those other species are addressed separately.

1.3. State Listing Status

Boltonia montana is listed as an endangered plant species by the New Jersey Department of Environmental Protection (NJDEP); this designation is intended to facilitate the conservation and protection of listed plant species, but the governing statute does not afford listed species any specific protections (New Jersey Administrative Code (NJAC) 2013, entire). However, implementing rules for the New Jersey Freshwater Wetlands Protection Act and the Flood Hazard Area Control Act may afford protections to B. montana or its supporting habitat (NJAC 2019a, entire; NJAC 2019b, entire).

The Pennsylvania Department of Conservation and Natural Resources (PADCNR) does not include Boltonia montana on its list of extirpated, vulnerable, rare, threatened, or endangered

plants (Pennsylvania Bulletin 2018, pp. 7758–7778; Pennsylvania Natural Heritage Program 2019, entire). Therefore, while the species historically occurred in Pennsylvania (Townsend and Karaman-Castro 2006, pp. 877, 881; NatureServe 2018, entire), the Commonwealth of Pennsylvania does not afford B. montana any specific protections.

The Virginia Department of Conservation and Recreation (VADCR) lists Boltonia montana as an endangered species (Virginia Administrative Code (VAC) 2013, entire). Under the governing statute, it is unlawful to “dig, take, cut, process, or otherwise collect, remove, transport, possess, sell, offer for sale, or give away” B. montana occurring in the wild, “other than from such person’s own property” (VAC 2008, entire). In addition, the species’ wetland habitats may be afforded protections by provisions of the Virginia Water Protection Permit Program Regulation (VAC 2016, entire).

1.4 Methodology and Data

This report is a summary of the SSA analysis, which entails three iterative assessment stages: species (resource) needs, current species condition, and future species condition (Figure 1-1).

1.4.1 Species (Resource) Needs

The SSA includes a compilation of the best available biological information on the species and its ecological needs at the individual, population, and rangewide levels based on how environmental factors are understood to act on the species and its habitat.

● Individual level: These resource needs are those life history characteristics that influence the successful Figure 1-1. The three analysis completion of each life stage. In other words, these steps in a Species Status are survival and reproduction needs that make the Assessment (Service 2016, entire). species sensitive or resilient to particular natural or anthropogenic influences.

● Population level: These components describe the resources, circumstances, and demographics that most influence resiliency of the populations.

● Rangewide level: This is an exploration of what influences redundancy and representation for the species. This requires an examination of Boltonia montana’s evolutionary history and historical distribution to understand how the species functions across its range.

To assess the biological status of Boltonia montana across its range, we used the best available information, including peer-reviewed scientific literature and academic reports, and survey data provided by state and Federal agencies as well as non-governmental organizations. Additionally,

10 we consulted with several species experts who provided important information and comments on B. montana distribution, life history, and habitat.

Using the SSA framework (Service 2016, entire), we consider what a species needs to maintain viability by characterizing the biological status of the species in terms of its resiliency, redundancy, and representation (together the “3Rs”) (Shaffer et al. 2002, pp. 139–140; Wolf et al. 2015, entire; Smith et al. 2018, entire). For the purpose of this assessment, we generally define viability as the ability of the species to sustain populations in natural ecosystems within a biologically meaningful timeframe: in this case, 30 years. We selected this 30-year timeframe because we assume that our modeled changes in land use, development, or conservation measures will manifest their effects on B. montana populations within that timeframe.

1.4.2 Current Species Condition The SSA describes the current known condition of Boltonia montana’s habitat and demographics, and the probable explanations for past and ongoing changes in abundance and distribution within areas representative of the geographic, genetic, or life history variation across the species range.

We considered Boltonia montana’s historical and current distribution and identified factors that negatively and positively influence the species. Scale, intensity, and duration of threats were considered for their potential impacts on the species and habitat. The magnitude and scale of potential impacts to B. montana or its habitat by a given threat are described using a High/Moderate/Low category scale. Additional detail/discussion on how populations were evaluated and defining analysis units is described at the beginning of the Current Condition section below (Chapter 5).

1.4.3. Future Species Condition The SSA forecasts a species’ potential response to three plausible future scenarios. The result of this future condition modeling is a prediction of the species’ ability to sustain populations in the wild over time (viability) based on the best scientific understanding of current and future abundance, distribution, and habitat condition.

To examine the potential future condition of Boltonia montana, we developed three future scenarios based on projections of habitat changes and beneficial conservation actions. The range of what may happen in each scenario is described based on the current condition and how resilience, representation, and redundancy may change. We chose a 30-year time frame for our analysis based on the best available information regarding land use, development, or beneficial conservation measures expected to occur during this period. We also assume that uncertainties surrounding some necessary assumptions for this analysis will increase at that point in time. The scenarios considered the most probable threats with the potential to influence the species at the population or rangewide scales, including potential cumulative impacts if applicable.

For this assessment, we define viability as the ability of Boltonia montana to sustain resilient populations in the wild over time. Using the SSA framework (Figure 1-1, above), we consider what the species needs to maintain viability by characterizing its status in terms of its resiliency, redundancy, and representation (Wolf et al. 2015, entire; Service 2016, entire). 11

• Resiliency is assessed at the level of populations and reflects a species’ ability to withstand stochastic events (arising from random factors). We can measure resiliency based on metrics of population health, such as population size, if that information exists. Resilient populations are better able to withstand disturbances such as random fluctuations in reproductive rates and fecundity (demographic stochasticity), variations in rainfall (environmental stochasticity), and the effects of anthropogenic activities. • Redundancy is assessed having a sufficient number of populations for the species to withstand catastrophic events (such as a rare destructive natural event or episode involving many populations). Redundancy is about spreading the risk and can be measured through the duplication and distribution of populations across the range of the species. Generally, the greater the number of populations a species has distributed over a larger landscape, the better it can withstand catastrophic events. • Representation is assessed at the species level and characterizes the ability of a species to adapt to changing environmental conditions. Metrics that speak to a species’ adaptive potential, such as genetic and ecological variability, can be used to assess representation. Representation is directly correlated to a species’ ability to adapt to changes (natural or human caused) in its environment.

The decision whether to list a species is based not on a prediction of the most likely future for the species, but rather on an assessment of the species’ risk of extinction. Therefore, to inform this assessment of extinction risk, we describe the species’ current biological status and assess how this status may change in the future under a range of scenarios to account for the uncertainty of the species’ future. We evaluate the current biological status of the species by assessing the primary factors negatively and positively affecting the species to describe its current condition in terms of the 3Rs. We then evaluate the future biological status by describing a range of plausible future scenarios representing a range of conditions for the primary factors affecting the species and forecasting the most likely future condition for each scenario in terms of the 3Rs. As a matter of practicality, the full range of potential future scenarios and the range of potential future conditions for each potential scenario are too large to analyze and describe them individually. These scenarios do not include all possible futures but rather include specific plausible scenarios that represent examples from the continuous spectrum of possible futures. Additional detail/discussion on how populations were evaluated for the future condition analysis is described at the beginning of the Future Condition section below (Chapter 6).

CHAPTER 2 - SPECIES INFORMATION

Boltonia montana (variously “mountain doll’s daisy,” “valley doll’s daisy,” or “Appalachian mountain boltonia”) (Figure 2-1) is a perennial plant in the Asteraceae (alternately Compositae) family that is known from isolated sinkhole pond habitats and associated riparian areas in Augusta County, Virginia; Sussex and Warren Counties, New Jersey; and Dauphin County, Pennsylvania, the latter regarded as a historical occurrence (Townsend and Karaman-Castro 2006, pp. 874, 881).

Figure 2-1. Boltonia montana plant in flower (© DCR-DNH, Gary P. Fleming).

2.1 Taxonomy

The species currently accepted taxonomic classification is as follows:

● Phylum: Plantae ● Division Magnoliophyta (flowering plants) ● Class: Magnoliopsida (dicotyledons) ● Subclass: Asteridae ● Order: Asterales ● Family: Asteraceae ● Genus: Boltonia ● Species: montana

Linnaeus (1767a, p. 116; 1767b, p. 962) first described Matricaria asteroides from specimens collected in Pennsylvania by William Barthram [sic]. In 1788, the taxon was reassigned to the new genus Boltonia (L’Heritier de Brutelle 1788, p. 16). Both L’Heritier de Brutelle (1788, p. 16) and Buffon (1799, p. 38) refer to the specimens’ habitat as “stagnis” (i.e., pools or inundated sites) in Pennsylvania and Virginia. Members of the genus Boltonia are endemic to North America and, as currently understood, includes at least five described species (Karaman-Castro and Urbatsch 2006, p. 353; Nesom and Robinson 2007, p. 328). While some earlier works and existing taxonomic databases indicate there are species of Boltonia native to East Asia (e.g., B. lautureana), morphological and genetic analysis suggest there is no close relationship between North American Boltonia and the East Asian species now generally recognized as belonging in

either the genus Aster or Kalimeris (Neson 1994, p. 163; Gu and Hoch 1997, entire; Noyes and Rieseberg 1999, p. 409; Nesom 2000, pp. 266–267; Chen et al. 2011, pp. 585–587; Li et al. 2012, pp. 1,345–1,347).

Boltonia asteroides is the most widespread member of the genus, occurring primarily in wetland habitats throughout the central and eastern United States. Three varieties of this species are commonly recognized, including: B. a. var. asteroides, B. a. var. latisquama, and B. a. var. recognita (Karaman-Castro and Urbatsch 2006, pp. 355–356). We note here that prior to 1985, Boltonia decurrens, a federally listed species, was recognized variously as either B. a. var. decurrens, B. a. var. latisquama, or B. a. var. recognita (Schwegman and Nyboer 1985, entire). Additional revisions to the taxonomy of B. asteroides have since been proposed, with both B. a. var. glastifolia and B. montana being described as distinct from B. a. var. asteroides, the taxon they were each often ascribed to (Townsend and Karaman-Castro 2006, entire; Townsend 2013, entire).

Members of the Boltonia asteroides complex are differentiated primarily by their means of vegetative proliferation, their flower and seed morphology, and their geographic distribution and habitat types (Table 2-1) (Karaman-Castro and Urbatsch 2006, pp. 355–356; Townsend and Karaman-Castro 2006, entire; Townsend 2013, entire; Weakley 2015, p. 1,093). As far back as the 1980s and 90s, botanists in New Jersey and Virginia began to question (independently) the taxonomy of Boltonia occurring in isolated sinkhole ponds in their respective states (Townsend 2020, entire). Around 2001, researchers in Virginia noted that specimens identified as B. a. var. asteroides collected from isolated sinkhole ponds in Augusta County, Virginia had physical characteristics indicating they were not a good fit for that or other described taxa (Townsend and Karaman-Castro 2006, 873–874). A subsequent examination of additional live and herbarium specimens also identified as B. a. var. asteroides from similar isolated pond or riverine habitats in the ridge and valley physiographic province of New Jersey and Pennsylvania found the same unique physical characteristics as the specimens from Virginia (Townsend and Karaman-Castro 2006, pp. 873–874). Additional comparisons of the physical characteristics of the plants from these ridge and valley populations to specimens of B. a. var. asteroides and B. caroliniana, the two most similar taxa, found them all to be morphologically distinct. This differentiation was further supported by preliminary (and unpublished) genetic analyses that suggested the Virginia and New Jersey populations had only a distant relationship to B. a. var. asteroides, and a “sister relationship” with B. caroliniana (Townsend and Karaman-Castro 2006, p. 882). Based on the observed physical characteristics, the distinctive ecological habitats occupied, and by genetic evidence, the authors described the new species B. montana (Townsend and Karaman-Castro 2006, entire).

Table 2-1. Some distinguishing characters of Boltonia montana and related taxa (data from Weakley 2015, p. 1,093). B. asteroides B. montana B. caroliniana var. asteroides var. glastifolia var. latisquama var. recognita Narrow Oblanceolate to Linear-subulate to Linear-subulate to Spatulate to Oblanceolate to Phyllary Shape oblanceolate to oblong lanceolate lanceolate obovate-spatulate linear-oblanceolate linear Phyllary Width (mm) 0.4 - 0.9 0.2 - 0.5 0.7 - 0.9 0.7 - 0.9 2.5 - 6.0 1.0 - 2.5 Achene Wing Width (mm) 0 - 0.1 0 - 0.1 0.2 - 0.5 0.2 - 0.5 0.2 - 0.5 0.2 - 0.5 Pappus Length (mm) 0.15 0.15 0.8 - 1.1 0.8 - 1.1 0.3 - 1.8 0.3 - 1.8 Sinkhole ponds and Bottomlands, Marshes, ditches, Marshes and Habitats associated riparian ditches, roadsides, Riverbanks Ditches isolated ponds prairies areas and prairies NJ to FL, MS, LA; NJ, VA, and PA MD and PA along Midwest US, VA to SC; coastal coastal plain w/ Range/Geographic Regions (extirpated); ridge lower Susquehanna disjunct in NC and Midwest US plain and piedmont disjunct inland and valley R. VA occurrences

Boltonia montana does not appear in the Flora of North America (Karaman-Castro and Urbatsch 2006, entire) or the Integrated Taxonomic Information System (ITIS) (ITIS 2019, entire) and is classified as “ambiguous” in the World Flora Online (WFO 2020, entire). The species is “provisionally accepted” in NatureServe Explorer® (NatureServe 2018, entire) and “accepted” by the Biota of North America Program (BONAP) (Kartesz 2015, entire), the Global Compositae Checklist (TICA 2019, entire), and the draft Flora of the Southern and Mid-Atlantic States (Weakley 2015, pp. 1,093–1,094). The omission of B. montana from some floras and datasets, and its unsettled recognition in others, likely reflects latency between its initial description in 2006 and subsequent review by the (mostly volunteer) scientists and experts responsible for inputting and reviewing new data, and not necessarily its taxonomic validity.

In summary, Townsend and Karaman-Castro’s (2006, entire) monograph is the only complete taxonomic treatment of Boltonia montana and we are not aware of any scientific disagreement with their determination that it is a unique species. The current taxonomy of the B. asteroides complex, including the description of B. montana as a species, is based primarily on subtle differences in flower and seed morphology, reproductive habit, and geographical distribution (Schwegman and Nyboer 1985, entire; Anderson 1987, entire; Townsend and Karaman-Castro 2006, entire; Townsend 2013, entire). There are only limited genetic data available to help us understand the relationships of the various Boltonia taxa. Townsend and Karaman-Castro (2006, p. 882) reference unpublished analyses demonstrating some undescribed level of genetic differentiation between B. montana and the two taxa most closely resembling it: B. a. var. asteroides and B. caroliniana. For context, DeWoody et al. (2011, pp. 197–199) noted B. asteroides and B. decurrens, two co-occurring Boltonia species, were readily distinguished based on morphology but displayed relatively minor genetic differentiation. Therefore, based on the best available information, we conclude B. montana is a valid taxon under the Act.

Additional taxonomic information is detailed in Appendix A.

2.2 Physical Description

Boltonia montana is described as a perennial (i.e., living for 2 or more years) herb standing 1.2

15 to 15 decimeters (dm) (4.7 to 59.0 inches (in)) tall (Townsend and Karaman-Castro 2006, p. 874). Its habit is generally erect, although often the lower stems remain prostrate (lying flat on the ground) (Figure 2-2). Plant stems are glabrous (smooth), light to yellow-green in color, and ribbed with yellow or golden striations. The upper stem has a branching structure with compound racemes (clusters of flowers), often arranged to appear as a convex or flat-topped cluster (Figure 2-3).

Figure 2-2. Boltonia montana growing in a "lawn" of needle spikerush (Eleocharis acicularis) at Split Level Pond, Augusta County, Virginia (© DCR-DNH, Gary P. Fleming).

Figure 2-3. Branching structure of Boltonia montana plant from New Jersey (Kathleen S. Walz, NJDEP). 16

The disk flowers (the tiny florets that make up the “eye” of Asteraceae flower heads) are pale yellowish green to pale yellow white, and the ray flowers (the petals or modified leaves that surround the disk) are pale lavender to pinkish (Figure 2-4).

Figure 2-4. Boltonia montana flowers (© DCR-DNH, Gary P. Fleming).

The leaves are smooth, generally broad and rounded at the end, and taper to the base. They are dark green or blue green with whitish or reddish color at their base and have prominent, lighter colored midribs. Basal (lower) leaves are 1.0 to 17.5 centimeters (cm) (0.4 to 6.9 in) long and 0.1 to 2.0 cm (0.04 to 0.8 in) wide and can be arranged alternately or in a rosette (Townsend and Karaman-Castro 2006, p. 874. The cauline (upper) leaves are 2.7 to 11.0 cm (1.1 to 4.3 in) long and 0.5 to 1.9 cm (0.2 to 0.7 in) wide, though sometimes smaller leaves cluster at stem nodes (Townsend and Karaman-Castro 2006, p. 874). For a more thorough description of Boltonia montana, including specific diagnostic features, the reader is referred to Townsend and Karaman-Castro (2006, pp. 874–875, 882–884).

2.3 Habitat

Boltonia montana is known primarily from isolated depressional ponds subject to fluctuating water levels. In New Jersey, the species occurs in calcareous sinkhole ponds (“dolines”) in the Kittatinny Valley of the Appalachian Valley and Ridge physiographic province (Figure 2-5) (Walz et al. (2001, pp. 1-1–1-5). In Virginia, B. montana is known primarily from similar “Shenandoah Valley sinkhole ponds,” located along the western base of the Blue Ridge Mountains, also in the Valley and Ridge physiographic province (Fleming and Van Alstine 1999, pp. 67–69). The only information available regarding the species’ habitat in Pennsylvania indicates specimens were collected from riverine areas along the Susquahanna River (Townsend and Karaman-Castro 2006, 877, 881).

Figure 2-5. Profile of typical Boltonia montana pond in New Jersey (from Walz et al. 2001).

The bedrock geology underlying Boltonia montana sites in New Jersey and Virginia is primarily dolomite and limestone. In New Jersey, surface soils at Boltonia montana sites are described as calcareous silt loams to silty clays, and occasionally silt marl or peat, with pH ranging from 6.2 to 7.8 (lightly acidic to mildly alkaline) (Townsend and Karaman-Castro 2006, p. 880). In Virginia, site soils are described as infertile clays or clay loams with pH ranging from 3.9 to 4.0 (acidic), though early (1936) collections from the South River mention “sandy, wet soil” (Townsend and Karaman-Castro 2006, pp. 878, 881). In both the New Jersey and Virginia metapopulations, some sites had soils with elevated levels of aluminum, calcium, or magnesium (Townsend and Karaman-Castro 2006, pp. 878–881).

Seasonal hydrological fluctuations are characteristic of Boltonia montana sites. For example, between April 1998 and March 1999, water levels at nine sites in New Jersey varied from about 0.7 to 4.8 m (2.3 to 15.8 ft) (Walz, et al. 2001, pp. 5-1–5-8).

The size of Boltonia montana population sites in New Jersey range from about 0.1 to 8.0 hectares (ha) (0.3 to 19.8 acres (ac)), with the average size being about 2.6 ha (6.5 ac) (Appendix B). In Virginia, population sites ranged from about 0.1 to 5.8 ha (0.3 to 14.3 ac), with the average size being about 1.1 ha (2.7 ac) (Appendix B).

The community type occupied by Boltonia montana in New Jersey is the Boltonia asteroides var. asteroides – Symphyotrichum racemosum – Mentha arvensis Herbaceous Vegetation Association (Grossman et al. 1998, p. 22; Walz et al. 2001, pp. 6–10; NatureServe 2018, [www.natureserve.org/explorer]), a community type that is ranked G1G2 (Critically

Imperiled/Imperiled – a very high risk of extinction due to extreme rarity, very steep decline or other factors) by NatureServe. Characteristic species in this community type include: Asclepias incarnata, Cyperus strigosus, Carex cryptolepis, C. lurida, C. typhina, C. viridula, Cuphea viscosissima, Eleocharis palustris, Eragrostis frankii, Eupatorium perfoliatum, Iris versicolor, Dichanthelium acuminatum var. fasciculatum (=Panicum lanuginosum var. tennesseense), Panicum capillare, P. rigidulum var. pubescens (=Panicum longifolium), Polygonum amphibium, Sium suave, the green algae Chara spp., and the bryophytes Hypnum lindbergii, Campylium stellatum, and Amblystegium serpens (Walz et al. 2001, pp. 6-22–6-23). Species in this community type which are rare in New Jersey include: Boltonia montana, Carex bebbii, C. cryptolepis, C. haydenii, C. retrorsa, C. typhina, C. viridula, Eleocharis quadrangulata, and Eragrostis frankii (Townsend and Karaman-Castro 2006, pp. 880–881; NJDEP 2016, entire).

The community types occupied by Boltonia montana in Virginia are the Quercus palustris/Panicum rigidulum – Panicum verrucosum – Eleocharis acicularis Herbaceous Vegetation Association and the Cephalanthus occidentalis/Polygonum hydropiperoides – Glyceria acutiflora – Proserpinaca palustris Shrubland Association (Grossman et al. 1998, pp. 21–22; NatureServe 2018, [www.natureserve.org/explorer). The former community type is endemic to Virginia and assigned a rarity rank of G1G2 (NatureServe 2018). The latter type is somewhat more widespread but considered endemic to mid-Atlantic region. These communities are open-canopied except on pond edges, where the forest canopy can provide partial shading. Associated species in these communities include Quercus palustris, Nyssa sylvatica, and Cephalanthus occidentalis, Panicum rigidulum var. rigidulum, P. philadelphicum, P. verrucosum, Eleocharis acicularis, E. melanocarpa, Eriocaulon aquaticum, Lysimachia hybrida, Ludwigia palustris, Leersia oryzoides, Hypericum boreale, and the Federally threatened Helenium virginicum” (Figure 2-6) (Townsend and Karaman-Castro 2006, pp. 878, 880).

Figure 2-6. Boltonia montana and Virginia sneezeweed (Helenium virginicum) co-dominate the rims of Twin Pond, Augusta County, Virginia (© DCR-DNH, Gary P. Fleming).

2.4 Life History

Information specific to the life history of Boltonia montana is limited; however, Townsend and Karaman-Castro (2006, p. 877) indicate the closely related and well-studied B. decurrens (previously B. asteroides var. decurrens) provides a reasonable surrogate. The two species are similar in structure and habit, and both depend on periodic disturbance to persist at a population site. In the following paragraphs, where B. montana information is lacking or sparse, we assume B. decurrens is an appropriate surrogate; where available, observed similarities or differences between the species are discussed.

2.4.1 Reproduction

Boltonia montana is known from habitats with widely fluctuating hydrological conditions and has developed certain life history strategies and physical characteristics for persisting in these habitats. Like other members of the genus, B. montana can reproduce both sexually from achenes (i.e. a dry, one seeded fruit) and vegetatively by the production of “basal offsets,” the formation of advantageous root systems at stem nodes that become independent of the parent plant (Figure 2-7) (Townsend and Karaman-Castro 2006, p. 876; Townsend 2013, p. 5).

Figure 2-7. Reproductive strategies of Boltonia montana. Sexual reproduction pathway is at the top; vegetative or clonal pathway is at the bottom (no timescale inferred).

Individual Boltonia decurrens (and presumably B. montana) plants behave as perennials and can persist over multiple growing seasons by producing basal offsets that establish as independent rosettes before the parent plant flowers and dies (Schwegman and Nyboer 1985, p. 114). The new rosettes are genetically identical to the parent; therefore, under favorable conditions (e.g., freedom from competition, seasonal water level drawdowns) vegetative propagation can result in genetic individuals persisting at a site for years (Schwegman and Nyboer 1985, p. 114; Baskin and Baskin 2002, p. 17). Observations indicate that in established B. decurrens populations, vegetative reproduction is the dominant method of reproduction and seedling recruitment is rare (Smith and Keevin 1998, p. 70; Smith and Mettler 2002, p. 114). However, after consecutive years of inundation that inhibit vegetative reproduction in a population, seedling recruitment from the seedbank is likely the dominant method of reproduction. This strategy is suggested for B. montana by Townsend and Karaman-Castro (2006, p. 877) and has been documented in Helenium virginicum (Virginia sneezeweed), another rare member of the family Asteraceae that co-occurs with B. montana at 15 Virginia pond sites (Knox 1997, pp. 243–245; Townsend 2020b, p. 1).

The prevalence of either seedlings or vegetative clones at a population site varies over time in response to local site conditions (i.e. inundation) (Townsend and Karaman-Castro 2006, pp. 876– 877). At a site in New Jersey, following water level drawdown, well-developed Boltonia montana plants with flowers, achenes, and basal offsets were common at pond margins not recently inundated, while many rosettes occurred in recently inundated areas nearer the pond center. It is unclear if the rosettes originated from newly germinated seeds, seeds that had germinated earlier, or if they originated as basal offsets independent of a parent plant (or a combination of scenarios).

Boltonia decurrens is a prolific seed producer, with mature plants producing up to 50,000

achenes (Schwegman and Nyboer 1985, p. 114; Smith and Keevin 1998, pp. 73, 76). Because B. montana shares similar plant structure and flower morphology (i.e., the approximate number of seeds produced by each flower head), it is reasonable to conclude that B. montana also has the potential for high fecundity (Smith and Keevin 1998, p. 74; Townsend and Karaman-Castro 2006, p. 874).

Boltonia montana is likely highly out-crossing (or self-incompatible), meaning that individual plants do not readily self-pollinate and will sexually reproduce most successfully with other unrelated individuals. Self-incompatibility is relatively common in Asteraceae (Ferrer and Good-Avila 2007, p. 410) and reproductive studies of B. decurrens confirm that it is highly out- crossing (Tofari 2000, as cited in DeWoody et al. 2004, p. 607; DeWoody et al. 2004, p. 610). There have been no similar studies of B. montana, though we note the rare, co-occurring (in Virginia) H. virginicum also exhibits high self-incompatibility (Messmore and Knox 1997, p. 320). Boltonia decurrens is reported to rely on bees, flies, gnats, and wasps for fertilization (DeWoody et al. 2004, p. 604) and the similarity in flower morphology suggests B. montana is also likely to be reliant on these generalist insect pollinators for sexual reproduction.

The germination of Boltonia montana seeds is dependent on their exposure to sunlight and soil moisture (or water). In a laboratory setting, 40 percent (n=200) of fresh B. decurrens seeds placed on a moist sediment surface germinated during a 4-week period, while only one out of 1,000 seeds buried between 0.5 and 2.5 cm (0.2 and 1.0 in) germinated (Smith and Keevin 1998, pp. 74–76). This suggests that siltation or other disturbance that results in the burial of B. decurrens seeds may prevent their germination. Additionally, B. decurrens seeds floated in deionized water remained viable after 4 weeks, with 1.5 percent of the floated seeds germinating while still in the water (Smith and Keevin 1998, p. 76). At natural sites, B. montana seeds have also been observed to germinate while under water, and the rosettes of both species are known to survive inundation (Smith and Keevin 1998, p. 79; Townsend and Karaman-Castro 2006, p. 876). However, while B. decurrens rosettes may bolt while still submerged, B. montana rosettes have not been observed developing further while underwater (Smith and Keevin 1998, p. 79; Townsend and Karaman-Castro 2006, pp. 876–877).

It is unknown how long and under what conditions Boltonia montana seeds can remain viable in the environment. The seeds of B. decurrens were found to remain viable for about 1 year in moist, anoxic soil conditions (i.e., a soil environment that has no oxygen available) (Redmond 1993, cited in Mettler-Cherry et al. 2006, p. 344), but when covered with moist (but presumably not anoxic) soil and stored in an unheated greenhouse nearly all seeds (n=400) remained viable after 88 months (at least 7 years) (Baskin and Baskin 2002, pp. 19–20). Mettler-McClure (1997, entire) reported that B. decurrens seeds set in 1992 remained viable in the soil throughout an extended flooding event the following year. The 1993 flood prevented the species from producing seeds that year, however, in 1994, the species re-emerged with flowering populations at half of the study sites (n=14). The authors concluded that for the recovery of B. decurrens following site disturbance (e.g. flooding), a substantial seedbank was necessary (Mettler- McClure 1997, p. 6). Smith et al. (2005, p. 1040–1041) studied the germination of B. decurrens seeds recovered from soil cores at natural sites and found 0 to 6 percent germination rates. They noted that B. decurrens seeds have a permeable seed coat and hypothesized the seeds might be susceptible to damage from soil microbes, thus decreasing their viability. Townsend and

Karaman-Castro (2006, p. 877) reference new B. montana plants developing from the seedbank after pond water levels dropped; however they did not indicate how long the site had been inundated nor is there any information provided regarding the percentage of seeds that remained viable. Thus, B. montana does appear to form a seedbank, with seeds likely remaining viable in the environment for 1 to 7 years, depending on soil conditions. The available information also suggests that extended, multi-year periods of inundation could reduce or prevent seed germination or plant development and thus prove detrimental to local populations.

The presence or dominance of reproductive lifeform (i.e. vegetative rosettes or seedlings) may also vary in response to local environmental conditions (Mettler-McClure 1997, p. 6; Smith and Keevin 1998, pp. 70, 76–77; Smith and Mettler 2002, p. 114; Townsend and Karaman-Castro 2006, pp. 876–877). Following extended periods of inundation, a local population may be composed solely of new plants originating from the seedbank or perhaps a combination of new plants from the seedbank and vegetative rosettes that survive from the previous season’s adult plants. Conversely, following extended periods of drawdown, seedling establishment may be low and local populations may be dominated by vegetative rosettes or adult plants that originated from vegetative rosettes.

2.4.2 Seasonality

Boltonia flowering and achene production occur in the late summer and early fall, with B. montana typically flowering from August to mid-October (although during dry years or in dry micro habitats, some plants may flower in July) (Townsend and Karaman-Castro 2006, p. 876). Boltonia decurrens seeds exhibit an annual dormancy/non-dormancy cycle that allow them to germinate at any time from late March to late October (Baskin and Baskin 2002, p. 22). Depending on the timing of germination and local environmental conditions, the new B. decurrens plants may behave as:

• Winter annuals (seeds germinate in the fall, seedlings overwinter as rosettes, then bolt, flower, and senesce during the following season); • Summer annuals (seeds germinate in the spring and seedlings bolt, flower, and senesce during the same season); or • Biennials (seeds germinate in the spring and remain as rosettes until the following season when they bolt, flower and senesce).

(Smith and Keevin 1998, p. 70; Baskin and Baskin 2002, pp. 20–23; Smith and Mettler 2002, pp.113–114). Therefore, it is reasonable to conclude that Boltonia montana also exhibits this flexibility in germination and growth behavior.

2.5 Population Structure

The population structure of Boltonia montana has not been studied. The species is known primarily from isolated depressional ponds, which we consider a B. montana “population.” Some B. montana populations are in close proximity to each other and heavy precipitation events could conceivably cause overland flows sufficient to transport seeds between them (Townsend and Karaman-Castro 2006, p. 883; Townsend 2019, entire), but other sites are isolated by

23 significant distances or topography and no surface hydrological connection is possible. Researchers have proposed that mammals or waterfowl may occasionally transport Boltonia decurrens seeds between sites (DeWoody et al. 2004, p. 614); however, the degree to which this potential vector might affect the dispersal and genetic variability of B. montana is unknown. For the purposes of this SSA, we refer to geographically proximate groups of B. montana populations as “population areas” (see section 3.3, below).

Boltonia decurrens, which is associated with open, wet riverine floodplain habitats, appears to function as a metapopulation; the seeds of that species are transported by seasonal overland flooding and populations appear, disappear, and shift annually (at least historically, prior to flood control projects that modified the river hydrology) (Smith et al. 2005, pp. 1,037, 1,049). The distribution of the species in Virginia (including historical observations of the species in riparian areas) and local stream morphology suggests surface water flows occasionally disperse B. montana seeds; therefore, the population structure in Virginia may exhibit metapopulation characteristics. The species’ distribution and the regional geomorphology in New Jersey does not suggest a clear connection between most of those populations, therefore the overall population structure in New Jersey is less clear. Despite these uncertainties, for the purposes of this SSA we will refer to the collective groups of populations in New Jersey and Virginia each as a “metapopulation” even though a strict definition of the term may not apply.

2.6 Species (Resource) Needs

At the scale of the individual Boltonia montana plant or population site, we identified soil, water, sunlight, pollinator services, and a suitable annual temperature regime, as interrelated resource needs. At the metapopulation scale, it is likely the species requires some degree of habitat connectivity to maintain viability; however, there is significant uncertainty regarding the degree of connectivity that may be necessary between population sites. We assume there is no natural connectivity between the two extant metapopulations in New Jersey and Virginia.

2.6.1 Appropriate Soil Conditions

Like most terrestrial plants, Boltonia montana requires soil for physical support and as a source of nutrients and water (via root uptake). In New Jersey, soils at B. montana sites are described as calcareous silt loams to silty clays, and occasionally silt marl or peat, with pH ranging from 6.2 to 7.8 (lightly acidic to mildly alkaline) (Walz et al. 2001, p. 3-9; Townsend and Karaman- Castro 2006, p. 880). In Virginia, site soils are described as infertile clays or clay loams with pH ranging from 3.9 to 4.0 (acidic), though early (1936) collections from the South River mention “sandy, wet soil” (Townsend and Karaman-Castro 2006, pp. 878, 881). In both the New Jersey and Virginia metapopulations, some sites had soils with elevated levels of aluminum, arsenic, calcium, or magnesium (Walz et al. 2001, pp. 10-8–10-24, 10-63–10-73; Townsend and Karaman-Castro 2006, pp. 878–881). It is hypothesized that plants tolerating these adverse soil conditions (e.g. low pH, high concentrations of some metals) may be released from competitive pressure from other plant species that are less tolerant of such conditions (Knox 1997, p. 244; Townsend and Karaman-Castro 2006, pp. 878–881).

The species is classified as a “facultative wetland plant,” which means it is associated with

24 seasonally saturated “hydric” soils (Lichvar et al. 2012, p. 4; Lichvar et al. 2016, p. 25), though Townsend (2019, entire) reported B. montana specimens growing well under garden conditions. The related B. decurrens also grows well in garden soils, with “sandy loams” appearing to be the preferred soil type for growth under natural conditions (Mettler et al. 2001, p. 95; Baskin and Baskin 2002, p. 17).

Boltonia montana seeds are likely to germinate best on bare soils, though it is uncertain if a particular soil type is preferred. The seeds of B. decurrens require bare, saturated soils for germination and seedling establishment (Smith and Hunsley 2006, p. 529). In the laboratory, the species’ seeds were found to germinate on clean white quartz sand, “loamy sand,” and “silty clay,” with the highest germination rates reported on the sandy substrates (Smith et al. 1995, p. 393; Baskin and Baskin 2002, p. 17).

In summary, individual B. montana plants appear physiologically tolerant of a variety of soil conditions, however, the species’ tolerance of certain adverse soil conditions (e.g. low pH, saturated conditions, high concentrations of some metals) may serve to reduce competitive pressure from other plant species. Therefore, to persist at a natural site (i.e., uncultivated), adverse soils conditions, as described above, may be required by the species.

2.6.2 Adequate Water

There is little information on Boltonia montana’s specific water needs during its various life stages (i.e., dormant seed, seedling, rosette, adult plant). As discussed previously, soil moisture is required for B. decurrens seed germination and to support the basic physiological processes of rosettes and growing plants, and conversely, extended periods of saturated soil conditions may reduce the viability of the seedbank. Therefore, it is reasonable to conclude that B. montana plants likely require sufficient soil moisture during the growing season (spring through fall) to support normal growth and reproduction and periods of drier conditions sufficient to prevent anoxic soils that reduce the viability of dormant seeds.

Additionally, evidence from Boltonia decurrens suggests periodic site inundation is a disturbance mechanism likely required for B. montana to persist at a site. Studies of B. decurrens demonstrate that plants survived extended periods of inundation during the growing season and that following drawdown, its survival and growth was greater than co-occurring species (Stoecker et al. 1995, pp. 123–124; Smith and Moss 1998, p. 558). Other studies indicate that B. decurrens populations increased following flood events and that in areas without flooding, the species was outcompeted and disappeared from sites after 3 to 5 years (Smith et al. 1998, p. 197; Smith and Keevin 1998, p. 70). Given clear water allowing sufficient sunlight penetration, B. montana seeds can germinate and rosettes develop while submerged, but the plants do not appear to develop further until water levels drop; if the site remains inundated for successive years, the plants may die (Smith and Keevin 1998, p. 79; Townsend and Karaman- Castro 2006, p. 876). Therefore, seasonally fluctuating water levels that serve to limit competing vegetation are needed for B. montana to persist.

Periodic flood events may also serve to disperse Boltonia montana seeds, contributing to its genetic variability and capacity to colonize (or recolonize) other nearby ponds or riparian sites

25 within the metapopulation area. This appears to be a life-history strategy of B. decurrens, a species associated with riverine habitats subject to annual overland flooding (McClain et al. 1997, entire; Mettler-Cherry et al. 2006, pp. 341–347). While B. montana seed morphology implies a reduced level of seed flotation as compared to B. decurrens (Townsend and Karaman- Castro 2006, p. 883), the pattern of stream channels and population sites in Virginia suggests B. montana seeds may occasionally be dispersed by surface water flows. However, this is less certain in New Jersey, where the present-day geomorphology does not permit surface water flows between most population sites. Therefore, periodic high water events that cause overland flows from B. montana sites to other ponds (or suitable riparian habitats) may be needed for long-term metapopulation persistence.

2.6.3 Adequate Sunlight

Boltonia montana is associated with open-canopy community types, suggesting it is adapted to high sunlight conditions during the growing season (Walz et al. 2001, p. 6-22; Townsend and Karaman-Castro 2006, p. 878). Additional support to demonstrate that the species requires adequate sun is provided by studies of the related B. decurrens, which has increased seed germination, seedling survival, plant growth, and seed production under high light (full sun) conditions as compared to lower light (partially shaded) conditions (Smith et al. 1993, pp. 861– 862; Smith et al. 1998, p. 192; Smith and Keevin 1998, pp. 75, 79). Therefore, the best available information indicates B. montana needs adequate levels of sunlight and minimal shading during the growing season.

2.6.4 Pollination Services

The production and soil banking of fertilized seeds is one of the reproductive strategies Boltonia montana uses to help ensure its persistence at sites that are subject to highly variable water levels. Therefore, we assume B. montana is likely reliant on generalist insect pollinators such as bees, flies, gnats, and wasps.

2.6.5 Seasonal Temperature Regime

There is no information available on the specific temperature tolerances of Boltonia montana. However, the species’ range is within a temperate climatic region characterized by relatively mild winters and warm or hot summers (Kottek et al. 2006, entire) which suggests it is adapted to seasonal temperature changes. The lowest average temperatures the species is naturally exposed to occur in January (-9 to -5 degrees Celsius (⁰C) (23 to 29 degrees Fahrenheit (⁰F)), while the highest average temperatures occur in July (28 to 29 ⁰C (83 to 84 ⁰F)) (U.S. Climate Data 2019).

Laboratory studies of Boltonia decurrens found that its seeds would germinate when simulated daytime temperatures were 15 ⁰C (59 ⁰F), although germination percentages increased markedly at higher temperatures (12 percent germination at 15 ⁰C (59 ⁰F) to 99 percent at 35 ⁰C (95 ⁰F) (Baskin and Baskin 2002, pp. 19–20). Notably, the researchers also found that germination percentages were positively correlated with the period of cold stratification (the exposure of dormant seeds to cool or cold temperatures). The germination percentages for seeds maintained

26 at 5 ⁰C (41 ⁰F) for 12 weeks ranged from 96 to 100 percent across all temperature regimes. This suggests that extended periods of cool or cold weather contributes to the persistence of B. montana populations.

2.6.6 Habitat Connectivity

As we discussed in section 2.5, the population structure of Boltonia montana has not been studied; however, the available information suggests that some populations in each metapopulation are (or were historically) connected by overland water flows or perhaps bird or animal vectors. Thus, some B. montana seeds could occasionally be dispersed to colonize or recolonize nearby habitats and contribute to the genetic variability of nearby populations. Additionally, the ability of insect pollinators to move between populations likely contributes to the genetic variability of the species (Cane and Love 2018, entire). The degree of connectivity needed to ensure persistence of the species within a particular metapopulation is uncertain. In the riverine-associated B decurrens, reduced habitat connectivity caused by flood control projects has led researchers to conclude fragmentation leaves that species more vulnerable to local declines and extirpations (Mettler-Cherry et al. 2006, pp. 342–347). Therefore, it is likely that within each B. montana metapopulation, some degree of population connectedness (characterized by geographic proximity, natural or semi-natural habitat between populations, and perhaps water drainage patterns) is beneficial to the species. Because of the long geographical distance between the New Jersey and Virginia metapopulations, we do not consider connectedness between these areas to be a species need.

CHAPTER 3 – HISTORICAL AND CURRENT DISTRIBUTION AND ABUNDANCE

3.1 Historical Context

For the purposes of this SSA, we determined the historical status of Boltonia montana (previously documented locally as B. asteroides) using all known survey data collected between 1864 and 2010 (Appendices C and D). We note that these data are temporally and geographically inconsistent (particularly for the period prior to 1980) and emphasize that all were collected after the European settlement of North America caused significant modification to natural vegetation communities (Bellemare et al. 2002, entire), possibly eliminating B. montana populations before they could be identified and recorded.

While many reports provide an estimate of Boltonia montana numbers at the time of the survey, a significant number simply indicate that the species was observed or that specimens were collected. Survey methodologies were rarely documented and we have no data regarding the level of survey effort across the species’ range (i.e., we have little consistent information on other potential sites that may have been surveyed but where it was not confirmed). Compounding these uncertainties is the adaptive life history of B. montana, whereby changing habitat conditions can lead to wide fluctuations in plant numbers at a site year-to-year (or even season-to-season) (Townsend and Karaman-Castro 2006, p. 876–877; Van Alstine 2019, pp. 10– –16). Observations of B. decurrens are illustrative. At one B. decurrens site, ten large flowering plants were observed one season, but in a subsequent “favorable” year when floodwaters receded early and precipitation was high, 20,000 flowering plants developed; conversely, a population of

18,000 plants declined to 125 flowering individuals following a late flood and low precipitation (Smith et al. 1995, p. 1,049). Therefore, even when population estimates are provided, quantitative comparisons between sites or across time should be interpreted with caution.

3.2 Historical Distribution and Abundance

The best available data indicate that Boltonia montana’s historical range was limited to three disjunct areas in New Jersey, Pennsylvania, and Virginia (Figure 3-1). As we discussed in section 2.5, for the purposes of this analysis we refer to these occurrence areas as metapopulations. The three metapopulations are widely separated from each other by between about 180 and 310 kilometers (km) (112 and 193 miles (mi)), with approximately 500 km (310 mi) between the southernmost and northernmost metapopulations. The species’ range and distribution in New Jersey and Virginia is confined to areas with an underlying carbonate bedrock; there is insufficient information to determine the geology of the Pennsylvania occurrences.

Figure 3-1. Historical distribution of Boltonia montana metapopulations (indicated in red). Note that the location of the Pennsylvania site is an approximation. Carbonate geological formations shaded tan (data from U.S. Geological Survey (USGS)).

3.2.1 New Jersey

In New Jersey, Boltonia montana is historically known from 21 population sites within an area of about 144 square kilometers (km2) (56 square miles (mi2)) near the town of Newton (Figure 3- 1; Appendices C and D).

In 1997, researchers conducted an ecological community study of sinkhole ponds within the sedimentary carbonate (e.g., limestone or dolomite) geological formations that appear to define the extent of the New Jersey Boltonia montana occurrences (Figure 3-2) (Walz et al. 2001, p. 1– 6). The study identified 85 sinkhole ponds in the region, 40 of which were considered relatively intact (i.e., undisturbed by development) and 45 which were determined to be degraded (i.e., farm ponds, destroyed by development, permanently flooded, or invaded by invasive plant species) (Walz 2020, p. 5). The researchers visited the 40 intact ponds and conducted detailed vegetation surveys at 22 sites; B. montana was confirmed at 14 sites (or about 35 percent of the intact ponds visited).

Five New Jersey Boltonia montana populations are on land owned or managed by the NJDEP (Swartswood State Park and White Lake Wildlife Management Area), six populations are on private property owned or managed by a conservation non-governmental organization (NGO, e.g., The Nature Conservancy (TNC)) (Muckshaw Ponds Preserve and Pleistocene Cave Sinkholes), and the remaining eleven populations are privately owned. The size of B. montana population sites in New Jersey range from about 0.1 to 8.0 ha (0.2 to 19.8 ac), with the average size being about 2.6 ha (6.5 ac). The combined area of all population sites in New Jersey is about 55 ha (137 ac) (Appendix B).

At two New Jersey sites, the historical data indicate Boltonia montana populations with greater than 10,000 individuals; one of these sites was estimated to have more than 710,000 plants during a survey in July of 1988 (NJDEP 2019a). Four sites had populations estimated to be in the thousands of plants, two sites had populations estimated to be in the hundreds, and six sites had populations estimated to be less than 100 plants (with three of these in the single digits). There are no estimated population numbers for the eight remaining New Jersey sites.

Figure 3-2. Geological setting and wetland distribution near Boltonia montana populations in New Jersey. Carbonate geological formation shaded tan; ponds and wetlands shaded blue; B. montana populations indicated by red diamonds (data from the Service’s National Wetland Inventory and USGS National Geologic Map database).

3.2.2 Pennsylvania

In Pennsylvania, the presumption of a historical Boltonia montana metapopulation is based on two museum specimens collected in July 1864 and August 1865 from the banks of the Susquehanna River near Dauphin, Pennsylvania (Townsend and Karaman-Castro 2006, pp. 877, 881). A comparison of these two specimens (identified as B. montana in 2006) to B. asteroides specimens collected from the same general area confirmed they were distinct from each other (Townsend 2019, entire). These are the only records of B. montana in the state; general surveys by the Pennsylvania Natural Heritage Program have failed to re-confirm the species (Goad 2020, p. 2). Because there is no other information available for the 1860s occurrences, including the specific location of the collections, numbers of plants present, or other habitat characteristics, we are unable to ascertain the historical distribution or abundance of the species in this presumed metapopulation.

3.2.3 Virginia

In Virginia, Boltonia montana is historically known from at least 19 individual sites within an area of about 34 km2 (13 mi2) in western Virginia, southwest of the town of Waynesboro (Figure 3-1; Appendices C and D). We note that three populations were discovered after 2010 and are

not defined as historical occurrences for purposes of this SSA. Similar to the New Jersey historical sites, all Virginia B. montana sites are located within sedimentary carbonate (e.g., limestone or dolomite) geological formations (Figure 3-3). More than 100 relatively undisturbed ponds and seepage wetlands in this area were surveyed for vegetation in the 1990s, but no other B. asteroides (as B. montana would have then been identified) occurrences were reported (Fleming and Van Alstine 1999, p. 69).

Figure 3-3. Geological setting and wetland distribution near Boltonia montana populations in Virginia. Carbonate geological formation shaded tan; ponds and wetlands shaded blue; B. montana populations indicated by red diamonds (data from Service’s National Wetland Inventory and USGS National Geologic Map database).

In Virginia, 7 Boltonia montana populations are on U.S. Forest Service land (George Washington and Jefferson National Forest) and the remaining 15 populations are on private property (this includes 3 populations discovered after 2010; two on private property and one on U.S. Forest Service land). The size of the Virginia B. montana population sites ranged from about 0.1 to 5.8 ha (0.3 to 14.3 ac), with the average size being about 1.1 ha (2.7 ac); the total area of all sites is about 23 ha (57 ac) (Appendix B).

In Virginia, seven historical sites had Boltonia montana populations estimated to be in the hundreds, while six had populations numbering less than 100 plants (two of these numbered in the single digits). Six historical sites in Virginia had no estimated population numbers available.

In summary, we reiterate that limitations in the available data and natural fluctuations in plant numbers introduce uncertainty into our understanding of the historical distribution and abundance of Boltonia montana. The historical data indicate B. montana occurred within a larger area in New Jersey than it did in Virginia. Relatedly, except for two sites, all of the Virginia population sites are in relatively close proximity to each other (generally within about 1 km (0.6 miles)). While some population sites in New Jersey are clustered together, in general the species appears to be more widely dispersed in this metapopulation. On average, the New Jersey sites are physically larger than the Virginia sites, and the data suggests the maximum population sizes of some New Jersey sites are also greater than the largest Virginia sites (Table 3-1). While the species was historically confirmed in Pennsylvania, there is no information available on the local population or habitat from which the specimens were collected.

Table 3-1. The 20 largest Boltonia montana population observations (historical and current observations). Additional survey details are provided in Appendix D. Maximum Est. Population Population Duck Pond 100,000+ Piggyback Pond 10,000+ Four Angle Pond 1,000s Little Frog Pond 1,000s NJ Spring Lake 1,000s Swallowhole Pond 1,000s Catfish Pond 100s Frog Pond 100s Huntsburg Sinkhole East 100s Lyndhurst Pond #18-15 1,000s Pond I 1,000s Twin Ponds #11, 12 1,000s Wood Duck Pond 1,000s Abshire Pond 100s VA Campbells Pond 100s Fence Line Ponds 100s Grove Farm Pond 100s Grove Farm Pond East 100s Route 664 Meadow East 100s Split Level Pond 100s

3.3 Current Distribution and Abundance

For the purposes of this report, data from 2010 to 2019 are used to assess the current condition of Boltonia montana (Appendix D). We selected this period because annual and seasonal pond water level variations largely determine the species’ population size (or detectability) at a site and because the survey data for the species are limited (see Appendix C). This period captures more than one annual observation for many sites, and at least one annual survey record for most others. Therefore, the 9-year period helps ensure sufficient data are available to ascertain B.

montana’s current condition in light of its natural habitat variability.

The best available information, including survey results from the Virginia Department of Conservation and Recreation’s Division of Natural Heritage (VADCR), the New Jersey Department of Environmental Protection’s Natural Heritage Program (NJDEP) and the U.S. Forest Service (USFS), indicate B. montana is extant within the New Jersey and Virginia metapopulations. There are no reports of B. montana from Pennsylvania since two specimens were collected near Dauphin in 1864 and 1865 (see Section 3.2 above). Therefore, we presume the species has been extirpated from Pennsylvania.

3.3.1 New Jersey

For this SSA, we group the Boltonia montana population sites within the New Jersey metapopulation by proximity, considering multiple populations within about 3 km (1.9 mi) of each other as a “population area” (Figure 3-4). Two of these population areas, Swartswood and Muckshaw, contain a relatively dense complex of population sites while two population areas, White Lake and Greendell, contain population sites that are more widely spaced. The two remaining population areas, Lake Grinnell and Huntsville, are single populations more isolated from the other known sites. The best available information indicates B. montana is extant in the Swartswood, Muckshaw, and White Lake population areas. We presume it remains extant within the Greendell and Huntsville population areas, but is extirpated at Lake Grinnell. However, because of limited survey data, there is significant uncertainty regarding the status of the species at these locations. This is especially so for the Lake Grinnell population area. There are only two survey records for this site. In 1887, the species was confirmed there but in 2019, it was not detected. Because conditions at the site may not be conducive to the species’ persistence there and because it has not been detected in more than 100 years, we presume this population is extirpated (the methodology behind this presumption is discussed further in section 5.4).

Figure 3-4. Distribution of all Boltonia montana populations in New Jersey. Individual population sites indicated by red circles; occurrences making up a population area circled in black.

In August of 2013, surveys in the Swartswood population area reported as many as 100 Boltonia montana plants at Frog Pond and up to 40 plants at Little Frog Pond; in August of 2016, more than 10,000 B. montana plants were observed at Duck Pond (NJDEP 2019a, p. 5).

In October and November 2019, botanists with the NJDEP surveyed or attempted to survey 21 of 22 historical Boltonia montana populations in New Jersey. Five sites were not visited because landowner permission was not obtained; therefore, sixteen historical B. montana populations were actually visited in 2019 (Walz 2019b, entire). Pond water levels were reportedly very high at most sites, which likely prevented the development of adult plants or the detection of submerged rosettes. However, the species was confirmed at 5 of 16 sites surveyed, with thousands of plants estimated at Duck Pond, about 250 plants at Catfish Pond, and 8, 1, and 17 plants reported at Frog Pond, Little Frog Pond, and Muckshaw Pond 2, respectively. A summary of current survey information for each known New Jersey population is provided in Table 3-2.

Based on the best available data, Boltonia montana’s distribution within the New Jersey metapopulation is similar to its historical condition. However, there is uncertainty regarding the species’ status at 16 previously documented population sites which have not been effectively surveyed during the current period because of high water levels or limited site access. In summary, the species is currently confirmed within three of six population areas and at 5 of 22 known B. montana populations in New Jersey.

Table 3-2. Summary of current (2010-19) survey data for Boltonia montana in New Jersey. Gray shading indicates no survey available. Location 2010 - 2019 Population Area Population 2016: Aug - 10,000+ plants abundant throughout pondshore and bottom of dry pond; 2nd year of dry pond conditions (NJDEP 2019). Duck Pond 2019: July - no plants observed, water level very high, above areas where plant has been observed in the past (NJDEP 2019). 2019: Oct - thousands of plants observed, very wet year, water levels high (NJDEP 2019) 2013: Aug - 50 to 100 plants in flower (NJDEP 2019). Frog Pond Swartswood 2019: Oct - 8 plant, very wet year, water levels high (NJDEP 2019). 2013: Aug - 20 to 40 plants in flower; overrun by Phalaris (NJDEP 2019). Little Frog Pond 2019: Oct - A single plant observed, very wet year, water levels high (NJDEP 2019). Spring Lake 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Swallowhole Pond Muckshaw Pond 1 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019).

Muckshaw Pond 2 2019: Oct - 17 plants observed, very wet year, water levels high (NJDEP 2019). Muckshaw Pond 3 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Muckshaw Muckshaw Pond 5 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Muckshaw Pond 6 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Muckshaw Pond 7 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Piggyback Pond 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). White Lake Catfish Pond 2019: Oct - 249 plants observed, very wet year, water levels high (NJDEP 2019). Shuster Ponds 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Huntsburg Sinkhole East Huntsburg Sinkhole West Greendell Johnsonburg Sinkhole Ponds Four-Angle Pond 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Pleistocene Cave Sinkhole 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019). Huntsville Turtle Sinkhole Ponds Lake Grinnell Lake Grinnell Ponds 2019: Oct - No plants observed, very wet year, water levels high (NJDEP 2019).

3.3.2 Virginia

For purposes of this SSA, we used site proximities and creek watershed boundaries to group the Virginia population sites into five population areas (Figure 3-5). Three of these population areas, Maple Flats, Campbells, and Lyndhurst, contain population sites located in relatively close proximity to each other. Boltonia montana is currently extant within these population areas. We note that the current mapped South River site is approximate; Townsend and Karaman-Castro (2006, p. 881) include a 1936 account describing Boltonia montana as “growing on bank of South River vicinity Lipscomb, abundant in Stuart’s Draft area around ponds and along South River.” Therefore, we are uncertain of the location or characteristics of this site (or sites). There are no other observations of B. montana in this general area; therefore, we consider it extirpated. Likewise, the Hattons population area (Peterson Pond) is presumed extirpated; the species was reported there in 1987 but has not been detected since.

Figure 3-5. Distribution of Boltonia montana populations in Virginia. Individual population sites indicated by red circles; populations making up a population area circled in black.

Since 2009 (i.e., during the current period for purposes of this SSA), three new population sites have been determined in Virginia. In the Maple Flats population area, Boltonia montana was first documented at Split Level Pond in 2011 and at Deep Pond in 2019. Pond “I” in the Lyndhurst population area was first documented in 2012.

In August and October 2018, botanists with the VADCR attempted to assess the status of 22 previously mapped Boltonia montana populations in Virginia (note that while VADCR identified

22 historical sites, we determined that three ponds are so closely associated they should be considered a single site, therefore we identified 19 historical population sites in Virginia) (Van Alstine 2019, entire). These surveys confirmed the species at eight of the sites, with five reportedly supporting “healthy” or “robust” populations (no explicit definition of these terms was provided, but plant numbers were estimated in the 100s to 1,000s). Two populations were described as having expanded slightly in area from earlier observations and three had fewer numbers of plants than indicated by previous surveys. The species was not confirmed at 10 sites; however, the author noted that because of limited access to privately owned sites, 3 could only be partially surveyed and 3 others were not surveyed at all. High water levels at several sites reportedly limited the emergence of B. montana (or perhaps its detectability).

In October 2019, the USFS surveyed nine ponds in the Maple Flats area (USFS 2019, entire). Five of these sites have historical records of the species and during this survey, it was reconfirmed at one (Twin Ponds #11, 12), with numbers estimated in the thousands (USFS 2019, entire). Four sites with no previous records of the species were also surveyed and a single B. montana plant was confirmed at one of these (Deep Pond) (USFS 2019, entire). A summary of current survey information for each known Virginia B. montana site is provided in Table 3-3.

Based on the best available data, Boltonia montana is extant within three Virginia population areas. However, the species may be extirpated from two population areas from which it has not been documented since 1936 and 1987. There is uncertainty regarding the species’ status at eight previously documented population sites because high water levels or limited access prevented effective surveys during the current period. In summary, the species is currently confirmed within three of five population areas and at 11 of 22 known B. montana populations.

Table 3-3. Summary of current (2010-19) survey data for B. montana in Virginia. Gray shading indicates no survey available. Location 2010 - 2019 Population Area Population 2015: water levels remain high in response to 2001 beaver dam (VADCR). Spring Pond 2018: no plants observed (Van Alstine 2019, p. 15). 2019: no plants observed, pond was full (USFS 2019). 2011: thousands of flowering plants (Van Alstine 2019, p. 15). 2015: thousands of flowering plants (Van Alstine 2019, p. 15). 2016: thousands of flowering plants (Van Alstine 2019, p. 15). Twin Ponds #11, 12 2017: abundant in western pond (Van Alstine 2019, p. 15). 2018: no plants observed, but site flooded to forest edge (Van Alstine 2019, p. 15). 2019: approx. 163 flowering plants and 2,000 - 3,000 rosettes in N. pond and 674 flowering plants and 400 rosettes in S. pond (USFS 2019). North Maple Flat Pond A 2018: no plants observed, but pond water levels never drew down and site was flooded (Van Alstine 2019, p. 14). North Maple Flat Pond B 2018: no plants observed, but pond water levels never drew down and site was flooded (Van Alstine 2019, p. 14). Maple Flats 2016: 12 sprawling plants in flower (Van Alstine 2019, p. 15). Oak Pond #13 2018: no plants observed, but little drawdown (Van Alstine 2019, p. 15). 2019: no plants observed (USFS 2019). 2018: no plants observed, but pond water levels never drew down and site was flooded (Van Alstine 2019, p. 14). Kennedy Mountain Meadow 2019: no plants observed, pond dry (USFS 2019). Deep Pond 2019: a single flowering plant observed (USFS 2019) 2011: hundreds of vigorous clumps throughout pond (Van Alstine 2019, p. 15). 2015: several hundred depauperate, sprawling plants (Van Alstine 2019, p. 15). Split Level Pond 2017: scattered plants, mostly matted/sprawling and fairly small; much less abundant and vigorous than 2011; (Van Alstine 2019, p. 15). 2018: 21 to 26 flowering plants (Van Alstine 2019, p. 15). 2019: no plants observed, pond was dry (USFS 2019). Campbells Pond 2018: Not surveyed, but aerial imagery indicates habitat likely present (Van Alstine 2019, p. 11) Campbells Grove Farm Pond 2018: Hundreds of flowering plants, no attempt to count rosettes (Van Alstine 2019, p. 10). Grove Farm Pond East 2018: Ten flowering stems and 1 bolted stem with no flowers (Van Alstine 2019, p. 10). 2017: observation of flowering plants (Van Alstine 2019, p. 11). Abshire Pond 2018: About 90-100 plants (Van Alstine 2019, p. 11). 2012: thousands of flowering plants (Van Alstine 2019, p. 12). I 2015: 1,200 stems (Van Alstine 2019, p. 12). 2018: thousands of flowering plants; rosettes present but difficult to assess numbers, occupied area has expanded (Van Alstine 2019, p. 12). Route 664 Meadow East 2018: no access provided to the majority of the area. In recently mowed area possibly a few rosettes (Van Alstine 2019, p. 13). Route 664 Meadow West 2018: No plants observed (Van Alstine 2019, p. 13). Lyndhurst Sherlynd Church 2018: Area recently mowed, no plants observed (Van Alstine 2019, p. 13). 2012: Hundreds of flowering plants (Van Alstine 2019, p. 14). Wood Duck Pond 2015: 4,100 to 5,100 stems (Van Alstine 2019, p. 14). 2018: Couple of hundred flowering plants (Van Alstine 2019, p. 14). Anderson Pond 2018: a single young flowering plant (Van Alstine 2019, p. 11). Fence Line Ponds (A, B, and South) 2018: pond B no plants or rosettes observed, but pond water levels very high; no access to ponds A and South (Van Alstine 2019, p. 12). Lyndhurst Pond #18-15 2018: thousands of flowering plants, population area slightly expanded (Van Alstine 2019, p. 13). South River South River Hattons Peterson Pond 2018: no plants observed, wetter than usual and no access to much of the property (Van Alstine 2019, p. 15).

CHAPTER 4 – FACTORS AFFECTING THE SPECIES

4.1 Stressors Affecting Viability

We identified a variety of stressors in two broad categories that are likely to affect the viability of Boltonia montana. Figure 4-1 is a conceptual model showing the relationships of the identified stressors to the habitat and life history needs of the species’ at the population (site) level (but note that many may also apply at the metapopulation level). While we generally discuss these stressors individually, many of them may act together additively or synergistically to affect B. montana resiliency.

Figure 4-1. Factors affecting the resiliency of individual Boltonia montana populations. Connecting lines indicate stressor-effect relationships, but not the potential magnitude of the relationships.

4.2 Habitat Modification

This broad category includes stressors that can physically disturb or kill Boltonia montana plants

or degrade or destroy the habitat required by the species. While most of the stressors identified here are associated with human activities, some natural processes such as beaver activity (causing semi-permanent inundation) or the natural accumulation of silt or vegetative debris (leading to community succession) may also alter the local habitat conditions. However, because the species evolved in concert with these naturally occurring phenomena, it is reasonable to conclude that it is adapted to these stressors (absent other human-induced stressors). The habitat modification category also includes several stressors that increase the potential for invasive plant species to be introduced into B. montana ponds, thereby increasing competition for resources. While there is evidence of past and ongoing habitat modification at many B. montana population sites, there have been no studies of the species’ response to these stressors at the individual, population, or metapopulation level.

4.2.1 Development

This stressor includes those activities associated with the direct physical disturbance, modification, or destruction of the habitats required by Boltonia montana. These include land- disturbing activities such as the filling of wetland areas for road or construction purposes; trash dumping; conversion of depressional areas for agriculture, silviculture, pasture, or turf; or the dredging or excavation of depressional areas in order to create deeper pond habitats for farm or recreational uses. The effects of such physical disturbances include the loss of suitable habitat, the destruction of individual B. montana plants, and the elimination of any viable seed bank (Van Alstine 2019, entire; Walz 2019a, pp. 5–6).

Other activities associated with development that do not necessarily involve disturbance of the soil or substrate are included in this category. One example is the mowing of depressional areas during the growing season, which may damage or kill adult Boltonia montana plants and reduce or eliminate sexual reproduction in the population (Van Alstine 2019, p. 13; Walz et al. 2001, p. 6-40) (Figure 4-2). Additionally, human development near B. montana population sites increases the likelihood that invasive plant species are introduced into B. montana sites (Walz 2019a, p. 5). For example, “creeping Jenny” (Lysimachia nummularia), a plant native to Asia and Europe was introduced into the United States as an ornamental groundcover; however, it readily spreads to moist habitats where it grows vigorously and can form a mat-like growth capable of excluding native plant species (Swearingen and Bargeron 2016, entire). Other invasive plant species identified as potential threats to B. montana include purple loosestrife (Lythrum spicata), Japanese stiltgrass (Microstigium vimineum), reed canary grass (Phalaris arundinacea), common reed (Phragmites australis), and Eurasian water-milfoil (Myriophyllum spicatum) (Johnson and Walz 2013, pp. 34, 39).

Figure 4-2. Aerial imagery dated October 3, 2017 indicating mowing at the Sherlynd Church population site (imagery from Google Earth, accessed December 23, 2019).

The best available information indicates that most Boltonia montana ponds were historically subject to disturbance caused by land conversion to agriculture, road construction, and other activities. At many populations, development pressure continues to occur; for example, surveys and aerial imagery indicate six Virginia population sites have been recently modified or disturbed (Grove Farm Pond East, Campbells Pond, Abshire Pond, Route 664 Meadow Ponds East and West, and Peterson Pond), two sites have been mowed (Route 664 Meadow East and Sherlynd Church), and non-native species have been planted near two sites (Pond I and Wood Duck Pond) (Van Alstine 2019, pp. 10–16). In New Jersey, a survey report (Walz 2019b, entire) indicates trash and debris is washing into Catfish Pond from an adjacent road and an invasive plant (creeping Jenny) has invaded four B. montana ponds (Duck Pond, Frog Pond, Muckshaw Pond 2, and Catfish Pond). Aerial imagery indicates Swallowhole Pond was mowed when water levels were low in 2016.

It is reasonable to assume that development pressure is tied to human population levels; therefore, we consider human population an indicator of how this stressor has affected the species in the past and how it will continue to affect Boltonia montana into the future. Census data indicate the human population in the New Jersey counties where the species is found increased significantly between about 1950 and 2010, but is projected to decline up to 2050 (Figure 4-3). In Virginia, human population increased modestly during the same period and is projected to continue increasing up to 2050.

350

300 Augusta, VA Sussex, NJ

250 Warren, NJ

200

150

Persons per100 Square Mile

0 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Ye a r Figure 4-3. Historical and projected human population density in counties occupied by Boltonia montana (data from U.S. Census 2019, entire; ProximityOne 2019).

4.2.2 Agriculture

We identified several activities associated with agricultural land use (other than the direct conversion of habitat, discussed above) that may be detrimental to Boltonia montana. These stressors include livestock pasturing, exposure to agricultural chemicals, and excessive groundwater withdrawals (which we discuss separately below).

Livestock (e.g., cattle, sheep) have the potential to affect Boltonia montana or its habitat in a number of ways. Hooved animals treading in depressional ponds can physically damage or kill adult plants or rosettes. Additionally, treading in wetlands causes soil disturbance that may reduce seed bank viability and increase water turbidity, thereby reducing the growth or survival of submerged B. montana rosettes (Morris and Reich 2013, p. 6). The deposition of livestock urine and feces in the wetland or adjacent riparian area can introduce excessive nutrients, leading to algal blooms, reduction in water clarity, and changes in plant community structure (Morris and Reich 2013, pp. 17–18). We have no information on the susceptibility of B. montana to livestock herbivory, but if the species is desirable then adult plants or rosettes could be damaged or killed by feeding livestock. Research indicates that in some ecological settings herbivory can prove beneficial to some species by limiting competition from other species (Morris and Reich 2013, pp. 20–23); however, we are unaware of potential benefits of livestock herbivory to B. montana or its sinkhole pond habitats. Finally, livestock fur or feces may transport the seeds of invasive species (Morris and Reich 2013, p. 24), which if established may increase competition with B. montana (see section 4.2.1).

In the 1940s, synthetic herbicides (chemicals designed to control undesirable plants or “weeds”) were introduced and are now used widely for agricultural, commercial, and residential purposes (Aspelin 2003, Ch.2 pp. 8–9, Part 4, entire; Fernandez-Cornejo et al. 2014, entire). However, data indicate non-target plant species or community structures can be harmed by herbicide spray drift, runoff, or soil accumulation (Kleijn and Snoeijing 1997, pp. 1,418–1,423; Gove et al. 2007, pp. 377–383; Boutin et al. 2013, pp. 298–304; Saunders and Pezeshki 2015, pp. 470–473). A model developed to assess the effects of three herbicides to Boltonia decurrens determined that herbicide spray drift and runoff could affect the survival and growth of exposed plants, and that herbicide spray drift could cause population-level effects (Schmolke et al. 2018, pp. 1,552– 1,554). Therefore, the cultivation of herbicide-treated crops in areas adjacent to B. montana sites increases the likelihood that plants or rosettes are exposed and affected, with potential population-level effects.

Recent surveys and aerial imagery indicate recent livestock grazing at two Virginia sites (Grove Farm Pond and Lyndhurst Pond #18-15) and crop production or haying at or adjacent to six other sites (Campbells Pond, Route 664 Meadows East and West, Pond I, Peterson Pond, and Abshire Pond)(Van Alstine 2019, pp. 10–16). In New Jersey, aerial imagery indicates agricultural cultivation near Swallowhole Pond and Huntsburg Sinkhole West. Because many privately owned B. montana population sites are located in a mixed low density rural residential/ agricultural setting, it is likely agriculture will remain a stressor to some populations.

4.2.3 Off-road Vehicle Use

The operation of off-road vehicles (ORVs) at Boltonia montana sites can damage or kill adult plants and rosettes, disturb soils and increase water turbidity thus reducing the growth and survival of submerged rosettes, disturb the seedbank, and damage or kill overwintering rosettes (Ouren et al. 2007, pp. 25–26; Switalski and Jones 2012, pp. 16–17). Additionally, ORVs are a potential vector of invasive plant species into B. montana ponds (Ouren et al. 2007, pp. 11–12). The operation of ORVs on the George Washington and Jefferson National Forest is prohibited except on designated trails (USFS 2014, pp. 4-17; 4-85– 4-86); there are no authorized ORV trails in the vicinity of the B. montana population areas on Forest Service lands (USFS 2015, maps 26 and 27). The NJDEP generally prohibits the operation of ORVs on lands they manage, including Swartswood State Park and the White Lake Wildlife Management Area (Piggyback Pond) (NJDEP 2002, entire). Other than having landowner permission, there are no restrictions on ORV use at privately owned B. montana sites.

Recreational ORV operators often seek out wet, muddy areas to ride in, and aerial imagery (Figures 4-4a and 4-4b) and survey reports indicate recent ORV use at several Virginia B. montana sites (Anderson Pond, Lyndhurst Pond #18–15, Maple Flats North Pond B, Oak Pond #13, and Twin Pond) (USFS 2019, p. 16; Van Alstine 2019, pp. 10–16). While we presume ORV disturbance damages or kills growing B. montana plants or disturbs the seed bank, there have been no studies confirming the degree to which ORV use affects B. montana populations.

Figure 4-4a. Aerial imagery dated October 3, 2017 showing ORV tracks within North Maple Flats Pond B. The U.S. Forest Service does not authorize ORV use in this area (imagery from Google Earth, accessed December 23, 2019).

Figure 4.4b. Aerial imagery dated October 20, 2013 showing ORV tracks within the privately- owned Lyndhurst Pond #18-15 (imagery from Google Earth, accessed December 23, 2019).

4.2.4 Altered Surface Hydrology

Surface modifications such as the construction of berms, levees, ditches, swales, or land grading can change patterns of overland water flow, possibly affecting water levels and soil moisture at some Boltonia montana sites (Jackson et al. 2014, pp. 58–60; McCauley et al. 2015, pp. 8–12). Because the species relies on seasonally fluctuating water levels to maintain suitable habitat conditions and limit competition, such changes to surface water inflows or discharges could be detrimental to individual populations. Survey reports from Virginia indicate several B. montana sites have had their hydrology modified to varying degrees (Van Alstine 2019, pp. 10–16). Aerial imagery indicates several New Jersey population sites (Huntsburg West, Turtle Sinkhole Ponds, Frog Pond, and Lake Grinnell Ponds) are adjacent to roads or railroads that likely altered the local hydrology; however, there is no information available documenting effects to B. montana at those sites.

4.2.5 Groundwater Withdrawals

Boltonia montana relies on fluctuating water levels to reduce competing vegetation at occurrence sites. The best available information suggests water levels in most B. montana ponds is closely tied to local groundwater fluctuations; however, this linkage has only been studied at several sites in New Jersey (Walz et al. 2001, pp. 5-3–5-8). Therefore, excessive groundwater withdrawals that lower the water table near occupied B. montana sites could affect the conditions required by the species (Walz 2019a, p. 5). Ongoing changes in the climate (discussed below) may also alter groundwater conditions. For example, extended drought conditions could simultaneously decrease aquifer recharge rates and increase the demand for groundwater pumping for agricultural and municipal purposes. In both Augusta County, Virginia, and Sussex County, New Jersey, groundwater makes up more that 90 percent of the water used (Maynard 2012, p. 26; Sussex County New Jersey 2019, p. 15). In Augusta County, Virginia, total water demand is predicted to increase from about 5.9 million gallons per day (gpd) in 2017 to about 9.5 million gpd in 2037 (Central Shenandoah Planning District Commission 2011, pp. 5-1–5-3); similar data are unavailable for New Jersey. There are no studies directly linking excessive groundwater withdrawals to B. montana population sites; however, it is reasonable to conclude that the species could be negatively affected by this stressor.

4.3. Effects of Climate Change

The available climate records indicate that since the late 1800s, precipitation and temperature patterns within the range of Boltonia montana have changed (National Oceanic and Atmospheric Administration (NOAA) 2019, entire). In general, temperatures have warmed (e.g., increased mean annual temperatures, increased frequency of extreme hot events, decreased number of frost days) and precipitation has increased (including the frequency of heavy precipitation events) (Romero-Lankao et al. 2014, p. 1,452). These changes can affect wetland hydrology and community structure (Burkett and Kusler 2000, entire); however, models suggest significant response variability by wetland type and landscape position (Wardrop et al., 2019, entire). There have been no studies of the response of B. montana or its habitat to the observed changes in climate, though an assessment of rare plants in New Jersey ranked the species as “moderately

vulnerable” to continued changes in climate (Ring, et al. 2013, p. 28). The report defines moderately vulnerable as “abundance and/or range extent within geographical area assessed likely to decrease by 2050” (Ring, et al. 2013, p. 6). Another report on rare plant communities in New Jersey concluded the species’ habitat (calcareous sinkhole ponds) are “highly vulnerable” to climate change, though the degree of vulnerability is not defined (Johnson and Walz 2013, pp. 43, 45).

4.3.1 Altered Precipitation Patterns

As discussed in section 2.6.2, Boltonia montana is adapted to sites with seasonally fluctuating water levels. Therefore, shifts in the magnitude and seasonality of within-year precipitation events, as well as changes in the frequency of extreme events such as extended or multi-year flood or drought conditions, could make conditions less suitable for the species, or perhaps make conditions more suitable for competing species. For example, extended (i.e., multi-year) periods of inundation could prevent seasonal growth and seed production in B. montana and allow for the establishment of a floating vegetation mat that shades submerged rosettes. If the viability of the seedbank is insufficient or degraded by extended inundation, the local population may not reestablish when water levels eventually recede. Conversely, extended periods of drought could allow for the establishment of tree and shrub species otherwise inhibited by high water levels. The encroachment of woody vegetation (trees and shrubs) into the sinkhole ponds would increase site shading, which is detrimental to B. montana growth (section 2.6.3).

Between 1895 and 2019, average annual precipitation in Sussex County, New Jersey increased by 8.4 millimeters (mm) (0.33 inches (in)) per decade; in Augusta County, Virginia, annual precipitation increased 6.6 mm (0.26 in) per decade (NOAA 2019) (Figure 4-5). While there is uncertainty regarding the potential effects of these changes on the Boltonia montana, because the species has a very limited geographical distribution within relatively small, isolated wetland ponds and because it is reliant on seasonally fluctuating water levels, it is reasonable to conclude changes in precipitation patterns may negatively affect the species.

Figure 4-5. Average annual precipitation in Boltonia montana counties 1895 to 2019 (NOAA 2019).

4.3.2 Altered Temperature Patterns

Little information exists regarding the species’ temperature tolerances, but it is reasonable to conclude it requires a seasonally variable temperature regime typical of its range (section 2.6.5). Periods of cool or cold weather during the winter likely increase the germination rates of Boltonia montana seeds later in the year (or succeeding years) as water levels recede. High germination rates may be particularly important for the species to persist at a site when all of the vegetative plants and rosettes are killed by high water levels, as periodically happens. Additionally, high temperatures during periods of drought can increase evapotranspiration and decrease soil moisture, reducing B. montana plant vigor and seed production, thereby decreasing a population’s reproductive output, as is suggested in B. decurrens (Smith et al. 1993, entire).

Between 1895 and 2019, the average annual temperature in Sussex County, New Jersey increased 0.17 ⁰C (0.3 ⁰F) per decade; in Augusta County, Virginia average annual temperature increased 0.06 ⁰C (0.1 ⁰F) per decade (NOAA 2019) (Figure 4-6). While there is uncertainty regarding the potential effects of these temperature changes on Boltonia montana, because the species has a very limited geographical distribution, a complicated life history whereby high seed production and germination rates may be critical for it to persist at a site, and a limited capacity to shift its range in response to changing conditions, we reasonably conclude increases in temperatures may negatively affect the species.

Figure 4-6. Average annual temperature in Boltonia montana counties 1895 to 2019 (NOAA 2019).

4.4 Potential Catastrophic Events

Along with the stressors described above, unforeseen catastrophic events could affect individual Boltonia montana populations or perhaps population areas. One example would be the accidental release of chemicals, petroleum, or other substance that kills or damages growing plants or contaminates soils and damages seeds or prevents their germination. Populations adjacent to roadways, rail lines, or pipelines are likely vulnerable to such events. Other potential catastrophic events include the introduction of an exotic insect or pathogen that affects Boltonia montana or perhaps a wildfire that destroys growing plants or the seedbank. However, because we cannot predict the likelihood of these catastrophic events occurring or the magnitude of the effect to the species, we do not include catastrophic events in our analysis.

4.5 Conservation Measures

4.5.1 New Jersey

The State of New Jersey lists Boltonia montana as an endangered plant species (NJAC 2013, p. 5); however, the governing statute does not afford any protections to such listed species.

The New Jersey Department of Environmental Protection (NJDEP) contributed to the development of integrated management guidelines for four rare New Jersey habitats and associated endangered plant species, which includes Boltonia montana (Johnson and Walz 2013, entire). These guidelines identify threats to B. montana habitats (i.e. calcareous sinkhole ponds) and include conservation strategies and actions that can be implemented with state and NGO conservation partners to help avoid negatively affecting the species (Johnson and Walz 2013, pp. 33–45).

The NJDEP also manages Swartswood State Park and the White Lake Wildlife Management Area (NJDEP 2018, entire; NJDEP 2019b, entire). Four ponds supporting Boltonia montana are located in the park and are designated as “Swartswood Natural Area” which is managed for “the preservation of rare species and plant communities in sinkhole pond ecosystems” (NJAC 2010, p. 25).

With the exception of one residential inholding, the Muckshaw Ponds population area is owned and managed as natural habitat by TNC. Specific conservation measures include reforesting buffer areas between adjacent agricultural areas and sensitive plant communities (TNC 2019, entire). The Pleistocene Cave Sinkhole is owned and managed by the Ridge and Valley Conservancy (Walz 2020, p. 6).

4.5.2 Virginia

Boltonia montana is listed as an endangered species by the Commonwealth of Virginia (VAC 2013, p. 1). Under the governing law, it is unlawful to “dig, take, cut, process, or otherwise collect, remove, transport, possess, sell, offer for sale, or give away” B. montana occurring in the wild, “other than from such person’s own property” (VAC 2008, pp. 2–3).

The VADCR is actively working to acquire or develop conservation partnerships to protect or enhance the species’ habitat at two privately-owned Boltonia montana sites; however, these conservation activities have not been finalized at the time of this SSA (Townsend 2020a, p. 3).

The Augusta County Comprehensive Plan (Augusta County 2007, entire) is a guide for land use decision making, establishing long-term goals and policy guidelines for the county. The Comprehensive Plan recognizes the pond habitats that support Boltonia montana (described as Montaine Depression Wetlands) as sites of special interest that are threatened by development, hydrologic alterations, ORVs, and trash dumping, but does not recommend any specific conservation measures (Augusta County 2007, p. 156). Therefore, while the County appears to support the conservation of B. montana habitats, we are unaware of any policies or projects advancing this goal.

The USFS (George Washington and Jefferson National Forest) has designated 7,200 ha (17,793 ac) of forestland surrounding the Maple Flats population sites as a Special Biological Area (SBA) (USFS 2014, p. 4-59). The SBA is managed for: (1) protection of threatened, endangered, sensitive, or locally rare species from human taking or human-caused detrimental habitat changes; (2) stable or increasing populations of threatened, endangered, sensitive, or locally rare species; and (3) functioning ecosystems (USFS 2014, p. 4-53). Some specific SBA management measures that may benefit Boltonia montana include the control of invasive species, prohibition of timber production, cooperation with the state NHP in adjusting the SBA, ensuring human recreation is compatible with the protection of rare or sensitive habitats, and limiting new trail or road construction (USFS 2014, pp. 4-53–4-57).

In summary, Boltonia montana is listed as state endangered by both New Jersey and Virginia. This status affords the species some protections on public lands in Virginia and helps guide conservation decisions in New Jersey. In New Jersey, 50 percent (11 of 22) of the known B. montana population sites are on land managed by the NJDEP or a conservation NGO, which affords them protection from development. In Virginia, about 32 percent (7 of 22) of the known populations are on USFS land designated an SBA, which is managed specifically for the conservation of rare species and their supporting habitats. Additionally, the Commonwealth of Virginia is actively pursuing additional protections for at least two privately-owned population sites. In total, about 57 percent (25 of 44) of the known B. montana populations currently occur on privately-owned property with no specific protections or conservation measures in place.

CHAPTER 5 – CURRENT CONDITION

5.1 Assessment Methodology

Data sufficient to directly assess the resiliency of the various Boltonia montana populations are sparse; therefore, we developed an assessment model using surrogate metrics to compare population sites with each other, both currently (as described below) and under plausible future scenarios (Chapter 6). We identified six metrics with which to assess the condition of each population site. These metrics were selected because: (1) each can be related, directly or indirectly, to factors influencing B. montana; (2) the data informing the metric are available and generally consistent across the range of the species; and (3) reasonable change (or lack of

change) in each metric can be projected under different future scenarios (see sections 6.1–6.3).

As discussed in section 4.3, the best available information indicates temperature and precipitation patterns within the range of Boltonia montana have changed since the late 1800s and are predicted to continue changing in the future. However, there is significant uncertainty regarding the species’ response to these changes. Therefore, we do not directly assess the potential effects of climate change to the population sites, but instead consider their relative condition (as estimated using the methodology discussed below) as an indicator of their capacity to persist in light of changing climate conditions.

5.2 Site Assessment Metrics

5.2.1 Pond Size

It is reasonable to infer that, given otherwise suitable conditions, larger sites can provide more overall habitat area for Boltonia montana population establishment and growth. Additionally, larger sites are likely to possess a wider variety or gradient of habitat features (e.g., sloping shoreline habitat, areas of sun/shade) than smaller sites. Together, these factors may promote larger B. montana populations and increase the population’s capacity to withstand demographic or environmental stochasticity. We determined site size by delineating the wetland area using aerial imagery (ESRI, Google Earth) and data from the National Wetland Inventory (NWI) (appendix B).

5.2.2 Hydrology

Variable water levels (seasonally and annually fluctuating) are required for Boltonia montana to persist naturally at a site (see section 2.6.2, above). Walz et al. (2001, pp. 5-1–5-8) monitored hydrological fluctuations at nine B. montana sites in New Jersey and found annual pond water level changes ranged from 0.7 to 4.8 m (2.3 to 15.8 ft). We are unaware of other detailed hydrological records for B. montana sites. Therefore, we assessed hydrological variability using time-series aerial imagery from Google Earth, which is available at various points in time back to the early 1990s, and information from survey reports, when available. Year-to-year (or season- to-season) changes in water level, as reported in survey reports or visible on the imagery (e.g., USFS 2019, entire; Van Alstine 2019, pp. 10, 14–15; Figure 5-1), are assumed to indicate that hydrological conditions were likely suitable for B. montana persistence (assuming no other degradation). The presence of dams or levies (including beaver dams), artificial channels, or ditches, are assumed to be evidence that site hydrology may be altered and conditions necessary for B. montana persistence could be compromised.

Figure 5-1. Aerial imagery showing water level fluctuations at Duck Pond, New Jersey. The image on the left is from August 27, 2016, and indicates no standing water at the site; the image on the right is dated May 23, 2018, and shows the site completely inundated (images from Google Earth, accessed December 17, 2019).

5.2.3 Proximity

It is hypothesized that the close proximity of Boltonia montana sites may facilitate seed transport between populations, thereby allowing for site colonization or recolonization as environmental or demographic conditions change (Townsend and Karaman-Castro 2006, pp. 877, 883; Johnson and Walz 2013, p. 31; Walz 2019a, p. 4; Townsend 2019, entire). The historical distribution of the species in Virginia and in some population areas in New Jersey support this hypothesis. Additionally, close proximity may promote genetic exchange (via pollen transport) between populations (Cane and Love 2018, entire). As discussed in section 2.4.1, B. montana is largely out-crossing (self-incompatible) and likely relies on insects for cross pollination. There is little information regarding the flight distances or effectiveness of generalist insect pollinators, but a study of honeybees (Apis mellifera) found that bees from a colony in a natural setting (eastern temperate forest) regularly foraged several kilometers from their nest (median distance of 1,650 m (5,413 ft) (Visscher and Seeley 1982, pp. 1,796–1,799). We determined proximity by measuring the straight-line distance between a site and its nearest neighbor site.

5.2.4 Disturbance

Land disturbing activities, such as dredging or filling, mowing or agricultural cultivation, livestock treading, or ORV operation can physically destroy Boltonia montana plants, the seedbank, or the habitat conditions required by the species. Additionally, disturbance increases the likelihood of invasive species introductions, which can alter the community structure and compete with B. montana for resources (see sections 4.2.1, 4.2.2, and 4.2.3, above). We used information from site survey reports and aerial imagery to determine historical and ongoing disturbance.

5.2.5 Landscape Setting

We reasonably assume that open sinkhole pond sites surrounded by forested areas are likely to maintain habitat conditions conducive to Boltonia montana persistence, including naturally fluctuating hydrologic conditions (section 2.6.2) and a close association with native plant and pollinator communities (section 2.6.4). Additionally, a forested buffer is likely to isolate B. montana population sites from other potential stressors, such as direct disturbance, exposure to herbicides, or the introduction of invasive species. Increasing degrees of residential, commercial, or agricultural development within the immediate watershed of a population site is likely to increase the risk of these stressors to B. montana. We determined the general landscape setting for each site using aerial imagery (e.g., ESRI, Google Earth).

5.2.6 Conservation Measures

Public land managers or private landowners may implement conservation measures that benefit Boltonia montana. Conservation measures may range from those designed to directly protect or improve the habitat used by B. montana (or in Virginia the co-occurring Helenium virginicum), or measures designed to protect otherwise undeveloped areas for low-impact recreational activities. We used formal management plans, map data layers, survey reports, organization mission statements, and information from species experts to determine or infer conservation measures (USFS 2014, pp. 1–6; USGS 2018, entire; Townsend 2019, entire; Townsend 2020a, entire; Walz 2019b, entire; Walz 2020, entire)

5.3 Metric Criteria and Current Site Scores

For each metric, we developed criteria to estimate how each is likely to affect the habitat supporting Boltonia montana. For each B. montana population site, the metrics are individually scored (-1, 0, or 1) based on the criteria in Table 5-1. These scores are summed to produce a site characteristics score (Table 5-2), which is our surrogate measure of B. montana resiliency. We describe site characteristic scores greater than two as “high,” zero to two as “moderate,” and less than zero as “low,” based on the quartile distribution of the observed scores. This systematic estimate of each site’s physical condition and its potential to support and maintain a viable B. montana population allows us to compare population sites currently and under plausible future scenarios (Chapter 6).

Table 5-1. Criteria for scoring individual Boltonia montana site metrics. Pond Size Disturbance

No reported or visable evidence of physical Large (greater than 5 Ha (12.4 ac)). 1 1 disturbance or invasive plant species.

Reported or visable evidence of historical Medium (0.5 to 5 Ha (1.2 to 12.4 ac)). 0 0 disturbance or invasive plant species.

Reported or visable evidence of current Small (less than 0.5 Ha (1.2 ac)). -1 -1 disturbance or invasive plant species present.

Hydrology Landscape Setting

Watershed forested with little or no Water levels appear to fluctuate naturally. 1 1 development.

Watershed includes residential, commercial, or Water levels appear to fluctuate with altered 0 agricultural development, but the pond site is 0 periodicity and/or magnitude. buffered. Watershed mostly residential, commercial, or Water levels do not appear to fluctuate. -1 agricultural development and the pond site is not -1 buffered.

Proximity Conservation Measures

Population located within about 0.5 km of more 1 Area managed for land or habitat conservation. 1 than one other population.

Population located within about 0.5 km of one Unmanaged natural area or area managed for 0 0 other population. outdoor recreational use.

Unmanaged residential, commercial, or No other populations within about 0.5 km. -1 -1 agricultural area.

Table 5-2. Relative current condition of Boltonia montana population sites. Green shading indicates the parameter is conducive to the species, yellow shading indicates an intermediate status, and red shading indicates the parameter is not necessarily conducive to B. montana. Gray shading indicates the parameter is unknown or undeterminable. Site Characteristics Site Population Area Population Size Hydrolo Proximit Disturba Landsca Conserv Score Duck Pond 1 1 1 -1 1 1 4 Frog Pond 0 1 1 -1 1 1 3 Swartswood Little Frog Pond -1 1 1 -1 1 1 2 Spring Lake 1 1 1 1 0 1 5 Swallowhole Pond 0 0 -1 -1 0 -1 -3 Muckshaw Pond 1 0 1 1 1 1 1 5 Muckshaw Pond 2 0 1 1 -1 1 1 3 Muckshaw Pond 3 0 1 1 1 1 0 4 Muckshaw Muckshaw Pond 5 0 1 1 1 1 1 5 Muckshaw Pond 6 0 1 1 1 1 1 5 Muckshaw Pond 7 -1 1 1 -1 1 1 2 Piggyback Pond 0 1 -1 1 1 1 3 New Jersey New White Lake Catfish Pond 1 1 -1 0 0 -1 0 Shuster Ponds 1 0 -1 -1 0 0 -1 Huntsburg Sinkhole East 0 1 0 1 0 -1 1 Huntsburg Sinkhole West 0 1 0 1 0 -1 1 Greendell Johnsonburg Ridge Sinkholes -1 1 -1 1 1 0 1 Four-Angle Pond 0 1 0 1 0 -1 1 Pleistocene Cave Sinkhole 0 1 0 1 0 1 3 Huntsville Turtle Sinkhole Ponds 1 1 -1 1 -1 -1 0 Lake Grinnell Lake Grinnell Ponds 0 1 -1 -1 0 0 -1 PA Susquehanna R. Susquehanna River UNK UNK UNK UNK UNK UNK UNK Spring Pond 0 -1 1 1 1 1 3 Twin Ponds #11, 12 0 1 1 -1 1 1 3 North Maple Flat Pond A -1 0 0 1 1 1 2 North Maple Flat Pond B 0 0 1 -1 1 1 2 Maple Flats Oak Pond #13 -1 1 1 -1 1 1 2 Kennedy Mountain Meadow 0 1 1 1 1 0 4 Deep Pond 0 1 1 1 1 1 5 Split Level Pond 1 1 1 1 1 1 6 Campbells Pond 0 1 -1 -1 -1 -1 -3 Campbells Grove Farm Pond 0 1 0 -1 -1 -1 -2 Grove Farm Pond East -1 1 0 0 0 -1 -1 Abshire Pond 0 1 1 -1 -1 -1 -1 Virginia I -1 1 1 1 -1 -1 0 Route 664 Meadow East -1 1 1 -1 -1 -1 -2 Route 664 Meadow West -1 1 1 -1 -1 -1 -2 Lyndhurst Sherlynd Church -1 0 0 -1 -1 -1 -4 Wood Duck Pond 0 1 1 1 0 -1 2 Anderson Pond -1 0 0 -1 0 -1 -3 Fence Line Ponds (A, B, and 0 1 0 1 0 0 2 Lyndhurst Pond #18-15 0 1 -1 -1 -1 -1 -3 South River South River UNK UNK UNK UNK UNK UNK UNK Hattons Peterson Pond -1 1 -1 -1 -1 -1 -4

5.4 Determination of Current Population Status

Because of sparse or inconclusive survey information, there is uncertainty regarding Boltonia montana’s continued existence at some sites. To help resolve this uncertainty in a consistent and systematic manner, we developed criteria that relies on the available survey data and the site condition scores to categorize B. montana populations as: extant, presumed extant, or presumed extirpated (Table 5-3).

Table 5-3. Criteria for estimating the current Boltonia montana population status (extant or extirpated) at each population site. “Unk” indicates no record of the site being surveyed during the period, “ND” indicates the site was surveyed but B. montana was not detected.

Extant B. montana detected during the current period (2010-19).

Unk or ND during the current period, but detected during Presumed Extant the 2000-09 period. Presumed Unk or ND since 2000 and Site Characteristic Score <0 (or Extirpated Unk).

A summary of the survey histories for each population site and our estimates of Boltonia montana’s current population status (e.g., extant or presumed extirpated) are at Table 5-4. We acknowledge the results of this process are uncertain; however, in the absence of more conclusive data, we consider the results reasonable based on the available information.

Table 5-4. Decadal survey history and estimated current status (e.g., extant or presumed extirpated) of Boltonia montana populations. Survey Data (by time period) Pop. Status Population Area Population Pre- Extant or 1990-99 2000-09 2010-19 1990 Extirp? Duck Pond 710,000 10,000+ 10,000+ Frog Pond 100s UNK 10s Swartswood Little Frog Pond 1,000s UNK 10s Spring Lake 1,000s UNK ND Swallowhole Pond UNK 1,000s UNK Muckshaw Pond 1 UNK 1s UNK ND Muckshaw Pond 2 1s UNK 10s Muckshaw Pond 3 10s UNK ND Muckshaw Muckshaw Pond 5 ND Muckshaw Pond 6 UNK UNK 10s ND Muckshaw Pond 7 ND Piggyback Pond UNK 1,000s 10,000+ ND New Jersey New White Lake Catfish Pond ND UNK 100s Shuster Ponds ND UNK ND Huntsburg Sinkhole East 100s UNK UNK Huntsburg Sinkhole West 10s UNK UNK Greendell Johnsonburg Ridge Sinkholes UNK UNK UNK Four-Angle Pond UNK 1,000s ND Pleistocene Cave Sinkhole UNK 1s UNK ND Huntsville Turtle Sinkhole Ponds UNK UNK UNK Lake Grinnell Lake Grinnell Ponds UNK UNK ND PA Susquehanna R. Susquehanna River UNK UNK UNK Spring Pond UNK ND Twin Ponds #11, 12 UNK 1,000s North Maple Flat Pond A UNK 10s 1s ND North Maple Flat Pond B UNK UNK 1s ND Maple Flats Oak Pond #13 UNK 1s 10s 10s Kennedy Mountain Meadow 10s ND Deep Pond UNK UNK UNK 1s Split Level Pond UNK UNK UNK 100s Campbells Pond 100s 100s UNK Campbells Grove Farm Pond 10s 100s 100s Grove Farm Pond East UNK UNK 100s 10s Abshire Pond 100s 100s 100s Virginia I UNK UNK UNK 1,000s Route 664 Meadow East UNK 100s UNK UNK Route 664 Meadow West UNK 1s UNK ND Lyndhurst Sherlynd Church UNK UNK ND Wood Duck Pond UNK UNK 1,000s Anderson Pond UNK 1s ND 1s Fence Line Ponds (A, B, and UNK 100s 100s UNK Lyndhurst Pond #18-15 100s 1,000s South River South River UNK UNK UNK Hattons Peterson Pond ND ND ND

5.4.1 Unknowns and Assumptions

As we discussed in sections 2.5, 3.1, 3.3, and 5.4, Boltonia montana plant numbers can fluctuate widely year-to-year and survey data are generally sparse; therefore, there is uncertainty regarding the species’ status, population size, or trends at the various population sites. Additionally, while the best available information suggests a link between the species’ needs and the identified stressors, there are no scientific studies documenting the magnitude or significance of the stressor effects to the species. This uncertainty is especially concerning in regards to the effects of climate change, which has the potential to affect the species across its range.

The best available information is not sufficient for determining potential trends in Boltonia montana abundance. For many population sites, the available survey data are limited to “presence/absence,” and where population estimates are provided, the data are infrequent and generally incomparable because survey methodologies were not documented. Therefore, it is unknown if B. montana population numbers are changing over time at the various population sites. Relatedly, in the absence of current survey data for some populations, we assume that if a historically known population site maintains habitat conditions conducive to the species, the population is presumed extant. If this assumption is incorrect, we may be overestimating the current condition of the species; conversely, if B. montana populations exist at sites that have not been surveyed, we may be underestimating the species’ current condition.

There is also uncertainty surrounding the potential effects some stressors may have already had on the species. This may result in our not knowing or in underestimating the importance of some species needs. For example, while the species is generally known from isolated sinkhole ponds, it is possible that the various populations were originally more connected on the landscape than the historical distribution information suggests. The negative effects of genetic isolation and demographic or environmental stochasticity (a risk inherent to small, isolated populations) may manifest over many generations (perhaps decades) and the best available information may not be sufficient to detect such changes, even if they are currently occurring. Therefore, we may be underestimating the significance of habitat fragmentation as an ongoing threat to the species.

5.5 Current Condition Summary

We used the conservation 3Rs—resiliency, redundancy, and representation—to summarize the current condition of Boltonia montana. As we discussed in section 5.3, the site characteristic scores are an indirect measure of population resiliency. We also estimated the resiliency of each population area based on the survey data and condition of the individual population sites. Specifically, we considered the site characteristic scores for the extant populations within each population area, the total number and size of extant populations in each area (i.e., redundancy within the population area), and other factors such as observed population size, specific local stressors, and the certainty of the available survey data. We assessed the species’ redundancy and representation based on the distribution of the species across its range.

Resiliency

In New Jersey, Boltonia montana is extant in three population areas, presumed extant in two

59 others, and presumed extirpated from one population area (Figure 5-1, Table 5-5). In the Swartswood population area, the species is extant at three sites and presumed extant at two others. The current site characteristic scores for these populations indicate their resiliencies range from low to high (-3 to 5; mean 2). Four populations are in close proximity to each other within the Swartswood State Park and while physical disturbance is not apparent, an invasive plant species occurs at three population sites. One of these sites has population estimates (both historical and current) of greater than 10,000 individual plants (the largest reported for any B. montana site). We consider the Swartswood population area to currently have high resiliency.

Figure 5-1. Current resiliency of Boltonia montana populations in New Jersey. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Table 5-5. Current resiliency of Boltonia montana population areas. Factors considered include total extant populations; the number of populations in each resiliency category (high, moderate, low); and the average of the condition scores for the extant populations within each population area. Total Extant Pop. Resiliencies Avg. Pop. Area Population Area Pops. Pops. High Mod Low Cond. Resiliency Swartswood 5 5 3 1 1 2 High Muckshaw 6 6 5 1 0 4 High White Lake 3 2 1 1 0 2 Mod NJ Greendell 5 5 1 4 0 2 Mod Huntsville 1 1 0 1 0 0 Low Lake Grinnell 1 0 Extirpated Maple Flats 8 8 5 3 0 3 High Campbells 3 3 0 1 2 -2 Low VA Lyndhurst 9 6 0 3 3 -1 Mod South River 1 0 Extirpated Hattons 1 0 Extirpated Total NJ 21 19 10 8 1 Total VA 22 17 5 7 5 Grand Total 43 36 15 15 6

Boltonia montana is currently extant at one population site in the Muckshaw population area and presumed extant at five others. The current site scores for the extant (or presumed extant) populations indicate their conditions range from moderate to high (2 to 5; mean 4). These sites are located in close proximity to each other within an area managed by TNC, a conservation organization (though one site is part of a residential inholding). With the exception of one small pond where the dumping of vegetation debris is occurring and one population site where an invasive plant species was noted, these populations are generally undisturbed and within a protected forested landscape. Therefore, we consider the Muckshaw population area to currently have high resiliency.

The White Lake population area consists of three isolated Boltonia montana population sites; the species is extant at one site, presumed extant at one site, and presumed extirpated at the third site. The current site characteristic scores for the two extant (or presumed extant) populations indicate their conditions range from moderate to high (0 to 3; mean 2). At the site where the species is confirmed extant, the population was estimated to number in the hundreds of plants; however, habitat disturbance from sediment and debris from an adjacent roadway, as well as the presence of an invasive plant species, was noted. The site where the species is presumed extant is located on land managed by the NJDEP and is relatively free from disturbance. While B. montana was not confirmed at this site during the current period, in 2005 more than 10,000 plants were estimated there. Because there are only two extant (or presumed extant) isolated population sites in this population area, we consider the resiliency of the White Lake population area to be moderate.

Boltonia montana is presumed extant in the Greendell population area; however, the survey data for these populations are sparse. Only two population sites were surveyed during the current period, and the species was not detected at either, possibly because of very high water levels.

The current site characteristic scores for the five extant (or presumed extant) populations indicate four are moderate and one is high (1 to 3, mean 2). Most surveys of this population area occurred in the 1990s, with individual population sites ranging from the tens to hundreds of individual plants. Even though most of these sites are located in a mixed, low-density residential/agricultural landscape, and all are privately owned, they do not show evidence of historical or ongoing disturbance. In light of the sparse survey data from this population area and because the sites are mostly isolated from each other, we consider the current resiliency of the Greendell population area to be moderate.

The Huntsville population area consists of a single isolated population site. The current site score for this site—Turtle Sinkhole Ponds—is moderate (0) and the species is presumed to be extant. This site was last surveyed in 1988, when Boltonia montana was observed but no population size estimated. The site appears relatively undisturbed but is on private property and is bounded on two sides by a road and residences. Because this population site is geographically isolated from other known B. montana populations, and because of significant uncertainty regarding the species’ past or current status there, we consider it to be a low resiliency population area.

In the Virginia metapopulation, the species is extant within three of five known population areas (Figure 5-2, Table 5-5). In the Maple Flats population area, Boltonia montana is extant at four sites and presumed extant at four others. The current site scores for the extant (or presumed extant) populations indicate their conditions range from moderate to high (2 to 6; mean 3). At most of these sites the populations are small, with only a few to about 20 plants ever observed; two sites are reported to have plants numbering in the hundreds or thousands. Seven of the Maple Flats populations are within the USFS special biological area, which is managed primarily for the conservation of the rare sinkhole habitats used by the species. The sites are generally undisturbed, although there is evidence of illegal ORV use at three sites. These sites are in relatively close proximity to each other, in a forested landscape setting, and appear relatively undisturbed (other than the ORV use mentioned). Therefore, we consider the current resiliency of this metapopulation to be high.

Figure 5-2. Current resiliency of Boltonia montana populations in Virginia. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Of the three populations in the Campbells population area, two are extant and one is presumed extant. The current site scores for the extant populations indicate their conditions are low (-3 to - 1; mean -2). Survey results indicate that these population sites may support hundreds of individual Boltonia montana plants, but they are in a low density agricultural and residential setting and one site is reportedly subject to recent livestock grazing. Therefore, we consider B. montana in this population area to currently have low resiliency.

In the Lyndhurst population area, five population sites are extant, one is presumed extant, and three are presumed extirpated. The current site scores for the extant (or presumed extant) populations indicate their conditions range from low to moderate (-3 to 0; mean -1). The area is mostly in a low-density agricultural and residential setting with disturbances such as mowing or livestock grazing noted at three of the six extant sites. However, three sites have reported population numbers in the thousands of plants and two have numbers in the hundreds. Many of the Boltonia montana populations are proximate to other populations. Therefore, even though the landscape setting and evidence of disturbance suggests low resiliency in this population area, the observation of five sites with relatively large numbers of plants leads us to conclude that it currently has moderate resiliency.

In summary, 15 Boltonia montana populations (10 in New Jersey, 5 in Virginia) are estimated to have high resiliency, 15 populations (8 in New Jersey, 7 in Virginia) have moderate resiliency, and 6 have low resiliency (1 in New Jersey, 5 in Virginia). At least seven populations (two in New Jersey, five in Virginia) are presumed extirpated. Based on these data, we conclude three population areas (two in New Jersey, one in Virginia) currently have high resiliency, three

63 population areas (two in New Jersey, one in Virginia) have moderate resiliency, and two population areas (one in New Jersey, one in Virginia) have low resiliency. Three population areas (one in New Jersey, two in Virginia) are presumed extirpated.

Redundancy

Boltonia montana is extant in two geographically isolated metapopulations, one in New Jersey and one in Virginia; a Pennsylvania metapopulation is presumed historically extirpated. Within the New Jersey metapopulation, the species is extant (or presumed extant) at 19 of 21 individual population sites (5 of 6 population areas) (Table 5-7). In Virginia, B. montana is extant (or presumed extant) at 17 of 22 population sites (3 of 5 population areas). However, as discussed above, the resiliency or population status of many of these sites is uncertain. Therefore, we consider B. montana to currently maintain moderate to high redundancy in the New Jersey metapopulation and moderate redundancy in the Virginia metapopulation.

Table 5-7. Current redundancy of Boltonia montana (does not include the Pennsylvania metapopulation, which is presumed historically extirpated). Meta Populations Population Areas pop Total Extant Percent Total Extant Percent NJ 21 19 90% 6 5 83% VA 22 17 77% 5 3 60% All 43 36 84% 11 8 73%

Representation

There are no available studies on the genetics or other characteristics with which to determine variation between Boltonia montana populations or metapopulations. The three metapopulations (two extant, one extirpated) are geographically isolated from each other by between about 180 and 310 km (112 and 193 mi) (approximately 500 km (310 mi) between the extant New Jersey and Virginia metapopulations) and as discussed in sections 2.3, 2.6.1, and 2.6.5, have certain differing habitat characteristics (e.g., community associates, soil chemistry, temperature fluctuations). Therefore, we assess the species’ representation based on its status in each metapopulation. Because B. montana currently maintains representation in two of three known metapopulations (at the northern and southern extremes of its range) we conclude that it currently has appropriate representation.

In summary, we estimate that Boltonia montana currently has high resiliency in three population areas (two in New Jersey, one in Virginia). These population areas have five to eight extant (or presumed extant) B. montana populations in relatively close proximity to each other within protected, generally undisturbed, forested landscapes. Three population areas (two in New Jersey, one in Virginia) have moderate resiliency. These population areas have two to six extant (or presumed extant) populations in landscapes ranging from relatively undisturbed forest to disturbed agricultural settings. Two population areas (one in New Jersey, one in Virginia) have low resiliency. The low-resiliency New Jersey population area has a single presumed extant population and the low-resiliency Virginia population area has three extant (or presumed extant) 64

populations in a disturbed agricultural setting. Three population areas (one in New Jersey, two in Virginia) are presumed extirpated.

The species is currently extant (or presumed extant) at 84 percent (36 of 43) of the known population sites spread between two isolated metapopulations. There is uncertainty regarding the status of many of these sites; therefore, we consider Boltonia montana’s current redundancy to be moderate. There is little information available with which to determine the species’ variability across its range, but because it is extant in two widely separated metapopulations with differing habitat characteristics, we conclude its representation is appropriate.

CHAPTER 6 – FUTURE CONDITIONS

We used the general assessment framework described above to model the potential future condition of Boltonia montana under three plausible scenarios. Based on the best available information and reasonable assumptions, we developed three potential future scenarios and manipulated various site metrics in the model to estimate the potential future resiliency of each population and population area.

We emphasize that our scenario modeling is not an attempt to predict the future or identify the likely future or futures. It is an attempt to provide a range of plausible alternatives with which we can explore the range of risk to the species. Because of significant uncertainties regarding B. montana’s potential response to climate change, our model relies on plausible changes to site metrics associated with more localized human activities. We project the potential condition of the species in about the year 2050. We selected this 30-year timeframe because we assume that our modeled changes in land use, development, or conservation measures will manifest their effects on B. montana populations within that timeframe. Additionally, we assume that uncertainties surrounding our assumptions will increase past that point in time.

6.1 Scenario 1

Under this scenario, we project no significant changes are made to the activities currently affecting the extant Boltonia montana populations. We consider extant populations currently with high or moderate resiliency (Tables 5-2 and 5-4) to be at low risk of extirpation within the next 30 years. The exception to this assumption is at populations where invasive plant species currently occur. Under this scenario, we assume the growth and spread of invasive plants will increase competitive pressure on these B. montana populations and lower their resiliency over the next 30 years. Populations currently with low site characteristic scores are considered to be at risk of decline (and perhaps extirpation) because of continued disturbances and environmental or demographic stochasticity.

6.1.1 Potential Future Viability, Scenario 1

Resiliency

Based on the plausible assumptions under scenario 1, it is likely that by 2050, invasive species competition will continue to affect several Boltonia montana populations in the Swartswood and

Muckshaw population areas in New Jersey; however, our model does not project a significant loss of resiliency in these areas. Physical disturbances are projected to continue affecting Virginia populations in the Campbells and Lyndhurst population areas, but again, under this scenario their resiliencies are not predicted to change significantly from the current condition (Figures 6-1 and 6-2, Table 6-1).

Figure 6-1. Resiliency of Boltonia montana populations in New Jersey under scenario 1. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Figure 6-2. Resiliency of Boltonia montana populations in Virginia under scenario 1. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Table 6-1. Resiliency of Boltonia montana population areas under scenario 1. Total Extant Pop. Resiliencies Avg. Pop. Area Population Area Pops. Pops. High Mod Low Cond. Resiliency Swartswood 5 5 3 1 1 2 High Muckshaw 6 6 4 2 0 3 High White Lake 3 2 1 1 0 2 Mod NJ Greendell 5 5 1 4 0 2 Mod Huntsville 1 1 0 1 0 0 Low Lake Grinnell 1 0 Extirpated Maple Flats 8 8 5 3 0 3 High Campbells 3 3 0 1 2 -2 Low VA Lyndhurst 9 6 0 3 3 -1 Mod South River 1 0 Extirpated Hattons 1 0 Extirpated Total NJ 21 19 9 9 1 Total VA 22 17 5 7 5 Grand Total 43 36 14 16 6

Redundancy

Under the modeled assumptions of scenario 1, we do not project a change in Boltonia montana redundancy from the current condition; the species will maintain moderate to high redundancy in the New Jersey metapopulation and moderate redundancy in the Virginia metapopulation.

Representation

Under scenario 1, we do not project that either the New Jersey or Virginia metapopulations will be extirpated; therefore, Boltonia montana representation, as we describe it, would remain appropriate.

6.2 Scenario 2

Under this scenario, we project that public and private land managers will implement various measures that result in the protection or enhancement of Boltonia montana habitat, where feasible. The types of activities we envision include invasive species control or eradication, removal of disturbances such as livestock grazing, mowing or agricultural cultivation, trash dumping, and ORV use. These changes could be implemented as part of formal management plans, land acquisition, conservation easements, conservation partnerships, or increased enforcement of existing policies.

For example, under scenario 2, we project the NJDEP may implement invasive species control projects at three Boltonia montana population sites on park property where invasive plant species have been reported. Likewise, TNC may implement similar measures at one of the Muckshaw ponds on their preserve. These activities are plausible and if implemented would improve the resiliency of four B. montana populations.

Other measures that would increase the resiliency of certain Boltonia montana populations include the institution of site protections and mitigation of ongoing disturbances. Under scenario 2, we project that the VADCR, perhaps working with other agencies or conservation organizations, may successfully negotiate protections for seven extant population sites in Virginia (at the time of this report, the VADCR is pursuing the acquisition of two privately- owned B. montana sites intended to be designated “Natural Area Preserves” (Townsend 2020a, p. 3)). Under voluntary conservation plans, landowners could mitigate ongoing disturbances by eliminating livestock or ORV access to B. montana ponds, cease mowing or cultivating population sites, and where possible increase forested buffers around sites. Additionally, two privately-owned B. montana sites are within or adjacent to lands already managed for the species’ conservation, one in the Muckshaw population area in New Jersey and one adjacent to the Maple Flats population area in Virginia. Under this scenario, we project that TNC and the USFS, perhaps in concert with other agencies or conservation organizations, could acquire these two properties. This would increase the number of B. montana population sites being actively managed for the species’ protection.

Additionally, under scenario 2 we project the USFS could implement measures to reduce or eliminate unauthorized ORV operation at Boltonia montana population sites in the Maple Flats population area in Virginia. Feasible measures might include placing physical barriers on access trails, posting signage warning of potential penalties for unauthorized ORV operation, and increased patrols by law enforcement. These types of conservation measures could reduce habitat disturbance at three population sites where there is evidence of ongoing ORV damage, and reduce the risk of such damage at four other population sites in the area.

To model the potential effects of such changes, we manipulated three related assessment metrics (disturbance, ownership, and conservation measures) at Boltonia montana population sites where we concluded conservation measures could be feasibly implemented (Table 6-2). We then assess the potential future condition of the species under the 3Rs.

The results modeled under scenario 2 represent perhaps the best possible outcome for the species. Under this scenario, we assume agencies and organizations have the resources necessary to implement these efforts, landowners are willing to make land use decisions supportive of Boltonia montana conservation, and the proposed conservation measure actually increase the species’ resiliency. While this scenario is plausible, there is uncertainty in these assumptions.

Table 6-2. Relative condition of Boltonia montana population sites under scenario 2. Numbers in bold indicate a plausible change from the current condition. Green shading indicates the parameter is predicted to be conducive to the species, yellow shading indicates an intermediate status, and red shading indicates the parameter is predicted to not be conducive to B. montana. Cross-hatching indicates the population is presumed extirpated. Site Characteristics Population Area Population Size Hydrolo Proximit Disturba Landsca Conserv Score Duck Pond 1 1 1 0 1 1 5 Frog Pond 0 1 1 0 1 1 4 Swartswood Little Frog Pond -1 1 1 0 1 1 3 Spring Lake 1 1 1 1 0 1 5 Swallowhole Pond 0 0 -1 -1 0 -1 -3 Muckshaw Pond 1 0 1 1 1 1 1 5 Muckshaw Pond 2 0 1 1 0 1 1 4 Muckshaw Pond 3 0 1 1 1 1 1 5 Muckshaw Muckshaw Pond 5 0 1 1 1 1 1 5 Muckshaw Pond 6 0 1 1 1 1 1 5 Muckshaw Pond 7 -1 1 1 -1 1 1 2 Piggyback Pond 0 1 -1 1 1 1 3 New Jersey New White Lake Catfish Pond 1 1 -1 0 0 -1 0 Shuster Ponds 1 0 -1 -1 0 0 -1 Huntsburg Sinkhole East 0 1 0 1 0 1 3 Huntsburg Sinkhole West 0 1 0 1 0 1 3 Greendell Johnsonburg Ridge Sinkholes -1 1 -1 1 1 1 2 Four-Angle Pond 0 1 0 1 0 1 3 Pleistocene Cave Sinkhole 0 1 0 1 0 1 3 Huntsville Turtle Sinkhole Ponds 1 1 -1 1 -1 -1 0 Lake Grinnell Lake Grinnell Ponds 0 1 -1 -1 0 1 0 PA Susquehanna R. Susquehanna River UNK UNK UNK UNK UNK UNK UNK Spring Pond 0 -1 1 1 1 1 3 Twin Ponds #11, 12 0 1 1 0 1 1 4 North Maple Flat Pond A -1 1 0 1 1 1 3 North Maple Flat Pond B 0 1 1 0 1 1 4 Maple Flats Oak Pond #13 -1 1 1 0 1 1 3 Kennedy Mountain Meadow 0 1 1 1 1 1 5 Deep Pond 0 1 1 1 1 1 5 Split Level Pond 1 1 1 1 1 1 6 Campbells Pond 0 1 -1 0 -1 1 0 Campbells Grove Farm Pond 0 1 0 0 -1 1 1 Grove Farm Pond East -1 1 0 0 0 1 1 Abshire Pond 0 1 1 0 -1 1 2 Virginia I -1 1 1 1 -1 1 2 Route 664 Meadow East -1 1 1 -1 -1 -1 -2 Route 664 Meadow West -1 1 1 -1 -1 -1 -2 Lyndhurst Sherlynd Church -1 0 0 -1 -1 -1 -4 Wood Duck Pond 0 1 1 1 0 1 4 Anderson Pond -1 0 0 0 0 1 0 Fence Line Ponds (A, B, and 0 1 0 1 0 1 3 Lyndhurst Pond #18-15 0 1 -1 0 -1 1 0 South River South River UNK UNK UNK UNK UNK UNK UNK Hattons Peterson Pond -1 1 -1 -1 -1 -1 -4

6.2.1 Potential Future Viability, Scenario 2

Resiliency

Based on the reasonable assumptions under scenario 2, in New Jersey, invasive species control projects in the Swartswood and Muckshaw population areas are projected to increase the resiliencies of five populations and the implementation of conservation land management in the Greendell population area is projected to increase the resiliencies of three populations. In Virginia, hydrological improvements, removal of disturbances, and conservation land management are projected to increase the resiliencies of 14 populations in the Maple Flats, Campbells, and Lyndhurst population areas (Figures 6-3 and 6-4, Table 6-3).

Figure 6-3. Resiliency of Boltonia montana populations in New Jersey under scenario 2. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Figure 6-4. Resiliency of Boltonia montana populations in Virginia under scenario 2. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Table 6-3. Resiliency of Boltonia montana population areas under scenario 2. Total Extant Pop. Resiliencies Avg. Pop. Area Population Area Pops. Pops. High Mod Low Cond. Resiliency Swartswood 5 5 4 0 1 3 High Muckshaw 6 6 5 1 0 4 High White Lake 3 2 1 1 0 2 Mod NJ Greendell 5 5 4 1 0 3 High Huntsville 1 1 0 1 0 0 Low Lake Grinnell 1 0 Extirpated Maple Flats 8 8 8 0 0 4 High Campbells 3 3 0 3 0 1 Mod VA Lyndhurst 9 6 2 4 0 2 High South River 1 0 Extirpated Hattons 1 0 Extirpated Total NJ 21 19 14 4 1 Total VA 22 17 10 7 0 Grand Total 43 36 24 11 1

Therefore, under scenario 2, three population areas in New Jersey are projected to have high resiliency, one moderate resiliency, and one low resiliency. In Virginia, two population areas are projected to have high resiliency and one moderate resiliency.

Redundancy

Under scenario 2, we do not project that any Boltonia montana populations will be extirpated; therefore, the species will maintain moderate to high redundancy in the New Jersey metapopulation and moderate redundancy in the Virginia metapopulation.

Representation

Under scenario 2, we do not project that either the New Jersey or Virginia metapopulations will be extirpated; therefore, Boltonia montana representation, as we describe it, would remain appropriate.

6.3 Scenario 3

Under this scenario, we assume no new conservation measures are implemented and habitat disturbances increase at most privately-owned Boltonia montana populations. These disturbances result from plausible increases in residential development, agricultural intensity (e.g., livestock grazing, crop production), forest clearing, and other disturbances associated with human development. The growth plans for both Sussex County, New Jersey and Augusta County, Virginia have stated goals of attracting new businesses, increasing tourism, and supporting and enhancing agricultural production while also maintaining the rural character and protecting natural areas in each county (Augusta County 2007, entire; Sussex County 2014, entire). Under scenario 3, we project that the counties are successful in achieving the former goals, but the latter goals are insufficient for protecting or limiting disturbances at certain B. montana populations. Additionally, under this scenario, we project that by 2050, invasive plant species spread from ponds where they have already been identified into adjacent sites, increasing competitive pressure on these populations.

To model the potential effects of such changes, we manipulated three related assessment metrics (disturbance, landscape setting, and conservation measures) at Boltonia montana population sites where we concluded the stressors were likely to increase (Table 6-4). We then assess the potential future condition of the species under the 3Rs.

The results modeled under scenario 3 represent perhaps the worst possible outcome for the species. Under this scenario, we assume no new conservation measures are implemented and landowners make land use decisions detrimental to Boltonia montana conservation. While this scenario is plausible, there is uncertainty in these assumptions.

Table 6-4. Relative condition of Boltonia montana population sites under scenario 3. Numbers in bold indicate a plausible change from the current condition. Green shading indicates the parameter is predicted to be conducive to the species, yellow shading indicates an intermediate status, and red shading indicates the parameter is predicted to not be conducive to B. montana. Cross-hatching indicates the population is presumed extirpated. Site Characteristics Population Area Population Size Hydrolo Proximit Disturba Landsca Conserv Score Duck Pond 1 1 1 -1 1 1 4 Frog Pond 0 1 1 -1 1 1 3 Swartswood Little Frog Pond -1 1 1 -1 1 1 2 Spring Lake 1 1 1 -1 0 1 3 Swallowhole Pond 0 0 -1 -1 0 -1 -3 Muckshaw Pond 1 0 1 1 -1 1 1 3 Muckshaw Pond 2 0 1 1 -1 1 1 3 Muckshaw Pond 3 0 1 1 -1 1 0 2 Muckshaw Muckshaw Pond 5 0 1 1 1 1 1 5 Muckshaw Pond 6 0 1 1 1 1 1 5 Muckshaw Pond 7 -1 1 1 -1 1 1 2 Piggyback Pond 0 1 -1 1 1 1 3 New Jersey New White Lake Catfish Pond 1 1 -1 0 0 -1 0 Shuster Ponds 1 0 -1 -1 0 0 -1 Huntsburg Sinkhole East 0 1 0 -1 0 -1 -1 Huntsburg Sinkhole West 0 1 0 -1 0 -1 -1 Greendell Johnsonburg Ridge Sinkholes -1 1 -1 1 1 0 1 Four-Angle Pond 0 1 0 1 0 -1 1 Pleistocene Cave Sinkhole 0 1 0 1 0 1 3 Huntsville Turtle Sinkhole Ponds 1 1 -1 1 -1 -1 0 Lake Grinnell Lake Grinnell Ponds 0 1 -1 -1 0 0 -1 PA Susquehanna R. Susquehanna River UNK UNK UNK UNK UNK UNK UNK Spring Pond 0 -1 1 1 1 1 3 Twin Ponds #11, 12 0 1 1 -1 1 1 3 North Maple Flat Pond A -1 0 0 1 1 1 2 North Maple Flat Pond B 0 0 1 -1 1 1 2 Maple Flats Oak Pond #13 -1 1 1 -1 1 1 2 Kennedy Mountain Meadow 0 1 1 -1 1 0 2 Deep Pond 0 1 1 1 1 1 5 Split Level Pond 1 1 1 1 1 1 6 Campbells Pond 0 1 -1 -1 -1 -1 -3 Campbells Grove Farm Pond 0 1 0 -1 -1 -1 -2 Grove Farm Pond East -1 1 0 -1 0 -1 -2 Abshire Pond 0 1 1 -1 -1 -1 -1 Virginia I -1 1 1 -1 -1 -1 -2 Route 664 Meadow East -1 1 1 -1 -1 -1 -2 Route 664 Meadow West -1 1 1 -1 -1 -1 -2 Lyndhurst Sherlynd Church -1 0 0 -1 -1 -1 -4 Wood Duck Pond 0 1 1 -1 0 -1 0 Anderson Pond -1 0 0 -1 0 -1 -3 Fence Line Ponds (A, B, and 0 1 0 -1 0 -1 -1 Lyndhurst Pond #18-15 0 1 -1 -1 -1 -1 -3 South River South River UNK UNK UNK UNK UNK UNK UNK Hattons Peterson Pond -1 1 -1 -1 -1 -1 -4

6.3.1 Potential Future Viability, Scenario 3

Resiliency

Under this scenario, in New Jersey, the spread of invasive species and increased habitat disturbances are projected to lower the resiliencies of two populations in the Swartswood and Muckshaw population areas and increased habitat disturbance is projected to lower the resiliencies of three populations in the Greendell population area. In Virginia, increasing land- use intensity and habitat disturbances are projected to lower the resiliencies of five populations. We project this will significantly lower the resiliency of the Greendell, Campbells, and Lyndhurst population areas (Figures 6-5 and 6-6, Table 6-5).

Figure 6-5. Resiliency of Boltonia montana populations in New Jersey under scenario 3. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Figure 6-6. Resiliency of Boltonia montana populations in Virginia under scenario 3. Green indicates high resiliency, yellow is moderate, and red is low resiliency. Gray indicates the population is presumed extirpated.

Table 6-5. Resiliency of Boltonia montana population areas under scenario 3. Total Extant Pop. Resiliencies Avg. Pop. Area Population Area Pops. Pops. High Mod Low Cond. Resiliency Swartswood 5 5 3 1 1 2 High Muckshaw 6 6 4 2 0 3 High White Lake 3 2 1 1 0 2 Mod NJ Greendell 5 5 1 2 2 1 Low Huntsville 1 1 0 1 0 0 Low Lake Grinnell 1 0 Extirpated Maple Flats 8 8 4 4 0 3 High Campbells 3 3 0 0 3 -2 Low VA Lyndhurst 9 6 0 1 5 -2 Low South River 1 0 Extirpated Hattons 1 0 Extirpated Total NJ 21 19 9 7 3 Total VA 22 17 4 5 8 Grand Total 43 36 13 12 11

Redundancy

While our model does not explicitly predict population extirpations, it is possible that the modeled increases in land-use intensity and habitat disturbance could cause populations with low resiliencies to disappear. In New Jersey, Boltonia montana resiliency and within-population area redundancy suggests that while several populations have low resiliencies and may be at risk of

76 extirpation, their potential loss would not cause the extirpation of the population area. However, in Virginia, four populations in the Lyndhurst and Campbells population areas are projected to have lower resiliencies under scenario 3. Because this includes all three of the populations in the Campbells population area, the entire analytical unit may be at risk of extirpation. Table 6-6 presents the effects of these potential extirpations on the species’ redundancy in New Jersey and Virginia. Therefore, under scenario 3 we project B. montana will maintain moderate to high redundancy in the New Jersey metapopulation but perhaps decrease from moderate to low redundancy in the Virginia metapopulation.

Table 6-6. Potential redundancy of Boltonia montana under scenario 3. Meta Populations Population Areas pop Total Extant Percent Total Extant Percent NJ 21 17 81% 6 5 83% VA 22 14 64% 5 2 40% All 43 31 72% 11 7 64%

Representation

Under scenario 3, we do not project that either the New Jersey or Virginia metapopulations would be extirpated; therefore, Boltonia montana representation, as we describe it, would remain appropriate.

6.4 Summary of Future Conditions

Under the three modeled future scenarios, we predict various plausible changes in the stressors, habitat condition, and conservation measures affecting Boltonia montana. These changes are predicted to affect the resiliency of the species at the population (Tables 6-7 and 6-8) and population area (Table 6-9) scale. Under scenario 1, by 2050, three populations are predicted to have lower resiliencies than the current condition; however, the population areas are predicted to maintain similar resiliencies as currently estimated. Under scenario 2, we predict feasible conservation efforts would improve the condition of 22 populations, significantly improving the species resiliency in one New Jersey population area and two Virginia population areas. Under scenario 3, the modeled changes are predicted to result in 11 populations having lower resiliency. Additionally, several low-resiliency populations are at risk of extirpation, possibly resulting in the loss of a Virginia population area. Under all scenarios, the species would remain extant in the New Jersey and Virginia metapopulations; therefore, its representation is not predicted to change from the current condition (though we note that the historical extirpation of the Pennsylvania metapopulation may have reduced the species representation).

Table 6-7. Relative condition of Boltonia montana population sites currently and under three potential future scenarios. Green shading indicates the parameter is predicted to be conducive to the species, yellow shading indicates an intermediate status, and red shading indicates the parameter is predicted to not be conducive to B. montana. Gray shading and cross-hatching indicates the population is presumed extirpated. Future Scenarios Population Area Population Current 1 2 3 Duck Pond 4 4 5 4 Frog Pond 3 3 4 3 Swartswood Little Frog Pond 2 2 3 2 Spring Lake 5 3 5 3 Swallowhole Pond -3 -3 -3 -3 Muckshaw Pond 1 5 3 5 3 Muckshaw Pond 2 3 3 4 3 Muckshaw Pond 3 4 2 5 2 Muckshaw Muckshaw Pond 5 5 5 5 5 Muckshaw Pond 6 5 5 5 5 Muckshaw Pond 7 2 2 2 2 Piggyback Pond 3 3 3 3 New Jersey New White Lake Catfish Pond 0 0 0 0 Shuster Ponds -1 -1 -1 -1 Huntsburg Sinkhole East 1 1 3 -1 Huntsburg Sinkhole West 1 1 3 -1 Greendell Johnsonburg Ridge Sinkholes 2 2 2 1 Four-Angle Pond 1 1 3 1 Pleistocene Cave Sinkhole 3 3 3 3 Huntsville Turtle Sinkhole Ponds 0 0 0 0 Lake Grinnell Lake Grinnell Ponds -1 -1 0 -1 PA Susquehanna R. Susquehanna River UNK UNK UNK UNK Spring Pond 3 3 3 3 Twin Ponds #11, 12 3 3 4 3 North Maple Flat Pond A 2 2 3 2 North Maple Flat Pond B 2 2 4 2 Maple Flats Oak Pond #13 2 2 3 2 Kennedy Mountain Meadow 4 4 5 2 Deep Pond 5 5 5 5 Split Level Pond 6 6 6 6 Campbells Pond -3 -3 0 -3 Campbells Grove Farm Pond -2 -2 1 -2 Grove Farm Pond East -1 -1 1 -2 Abshire Pond -1 -1 2 -1 Virginia I 0 0 2 -2 Route 664 Meadow East -2 -2 -2 -2 Route 664 Meadow West -2 -2 -2 -2 Lyndhurst Sherlynd Church -4 -4 -4 -4 Wood Duck Pond 2 2 4 0 Anderson Pond -3 -3 0 -3 Fence Line Ponds (A, B, and South) 2 2 3 -1 Lyndhurst Pond #18-15 -3 -3 0 -3 South River South River UNK UNK UNK UNK Hattons Peterson Pond -4 -4 -4 -4

Table 6-8. Summary of current and projected Boltonia montana population resiliency. NJ Population Resiliency VA Population Resiliency High Mod Low Extirp. High Mod Low Extirp. Current 5 13 1 n/a 4 7 5 n/a Scenario 1 4 14 1 0 4 7 5 0 Scenario 2 8 10 1 0 7 9 0 0 Scenario 3 3 13 3 0 4 5 5 2

Table 6-9. Summary of current and projected Boltonia montana population area resiliency. Curr. Future Scenario Population Area Cond. 1 2 3 Swartswood High High High High Muckshaw High High High High White Lake Mod Mod Mod Mod NJ Greendell Mod Mod High Low Huntsville Low Low Low Low Lake Grinnell Extirp. Extirp. Extirp. Extirp. Maple Flats High High High High Campbells Low Low Mod Low VA Lyndhurst Mod Mod High Low South River Extirp. Extirp. Extirp. Extirp. Hattons Extirp. Extirp. Extirp. Extirp.

CHAPTER 7 – OVERALL SYNTHESIS

The goal of this assessment is to describe the current and potential future conditions of Boltonia montana in terms of resiliency, redundancy, and representation by using the best available commercial and scientific information. To capture the uncertainty associated with the degree and extent of potential future risks and their effects on the species’ needs, we assessed potential future conditions using three plausible scenarios. These scenarios were based on a variety of negative and positive influences on the species across its current range, allowing us to predict potential changes in habitat used by the species. The results of our analysis describe a range of possible conditions of B. montana populations.

Historical Range and Abundance

The best available data indicate that Boltonia montana’s historical range was limited to three disjunct areas in New Jersey, Pennsylvania, and Virginia. In New Jersey, the species is historically known from 21 population sites within an area of about 144 km2 (56 mi2). Five New Jersey B. montana populations are on land owned or managed by the NJDEP, 6 populations are on private property owned or managed by a conservation NGO, and the remaining 11 populations are privately owned. The size of B. montana population sites in New Jersey range from about 0.1 to 8.0 ha (0.2 to 19.8 ac), with the average size being about 2.6 ha (6.5 ac). The 79 combined area of all population sites in New Jersey is about 55 ha (137 ac). Two populations are reported to be greater than 10,000 individual plants, four sites have populations estimated in the thousands of plants, two sites have populations estimated in the hundreds, and six sites have populations estimated at less than 100 plants (with three of these in the single digits). There are no estimated population numbers for eight New Jersey sites.

In Virginia, Boltonia montana is known from 22 individual sites, encompassing a total range of about 34 km2 (13 mi2) (three of these sites were first reported after 2010). Seven of these populations are on USFS land and the remaining 15 populations are on private property. The size of the Virginia B. montana population sites range from about 0.1 to 5.8 ha (0.2 to 14.3 ac), with the average size being about 1.0 ha (2.4 ac); the total footprint of all sites is about 23 ha (56 ac) within the 34 km2 (13 mi2) range. Seven Virginia populations are estimated to number in the hundreds of plants, six have populations numbering less than 100 plants (two of these numbered in the single digits), and six sites in Virginia have no estimated population numbers available.

In Pennsylvania, the presumption of a historical Boltonia montana metapopulation is based on two museum specimens collected in July 1864 and August 1865 from the banks of the Susquehanna River near Dauphin, Pennsylvania. These are the only records of B. montana in the state and because there is no other information available for these records, we are unable to ascertain the historical distribution or abundance of the species in this presumed metapopulation.

Limitations in the available data and natural fluctuations in plant numbers introduce uncertainty into our understanding of the historical distribution and abundance of Boltonia montana. The species occurred within a larger area in New Jersey than in Virginia and on average, the New Jersey sites are physically larger than the Virginia sites. Similarly, the data suggest the maximum population sizes of some New Jersey sites are also greater than the largest Virginia sites. Except for two sites, all of the Virginia population sites are in relatively close proximity to each other (generally within about 1 km (0.6 miles)). While some population sites in New Jersey are clustered together, in general the species appears to be more widely dispersed in this metapopulation. While the species was historically confirmed in Pennsylvania, there is no information available on the local population or habitat from which the specimens were collected.

Current Viability Summary

Using the best available information and our modeled assumptions, 15 Boltonia montana populations currently have high resiliency: 10 in New Jersey and 5 in Virginia. These high resiliency populations occur within three population areas. Fifteen populations currently have moderate resiliency: eight in New Jersey and seven in Virginia. Populations with moderate resiliency occur within most extant population areas. There is one population with low resiliency in New Jersey and five in Virginia. These occur within four population areas.

Within the New Jersey metapopulation, Boltonia montana is extant (or presumed extant) at 19 of 21 individual population sites (5 of 6 population areas); in Virginia, the species is extant (or presumed extant) at 17 of 22 population sites (3 of 5 population areas). Therefore, we conclude the species has moderate to high redundancy in New Jersey and moderate redundancy in

Virginia. Because the species is extant in two geographically isolated metapopulations, but extirpated from Pennsylvania, we consider B. montana to currently maintain moderate redundancy across its range.

Future Condition Scenarios

We modeled the potential future condition of Boltonia montana under three plausible scenarios. We use the results to describe the species’ condition using the 3Rs in about the year 2050. We selected this 30-year timeframe because we assume that our modeled changes in land use, development, or conservation measures will manifest their effects on B. montana populations within that timeframe.

Under scenario 1, we project no significant changes are made to the activities currently affecting the extant Boltonia montana population sites. We consider extant population sites currently with high or moderate site condition scores to have high resiliencies and be at low risk of extirpation within the next 30 years. The exception to this assumption is at population sites where invasive plant species currently occur. Under this scenario, we assume the growth and spread of invasive plants will increase competitive pressure on these populations and lower their resiliency over the next 30 years.

The results of the scenario 1 model are that Boltonia montana will maintain three high-resiliency population areas (two in New Jersey, one in Virginia), three moderate-resiliency population areas (two in New Jersey, one in Virginia), and two low-resiliency population areas (one in each state). While it is possible some small populations with low resiliency could be extirpated, our analysis indicates that the species would maintain redundancy within each metapopulation and across its range. Because we do not project that either the New Jersey or Virginia metapopulations will be extirpated, the species’ representation would not change from the current condition.

Under scenario 2, we project that public and private land managers will implement various measures that result in the protection or enhancement of Boltonia montana habitat, where feasible. The results modeled under this scenario represent perhaps the best possible outcome for the species. Under this scenario, we assume agencies and organizations have the resources necessary to implement these efforts, landowners are willing to make land use decisions supportive of B. montana conservation, and the proposed conservation measure actually increase the species’ resiliency.

The results of the scenario 2 model project five high-resiliency population areas (three in New Jersey, two in Virginia), two moderate-resiliency population areas (one in each state), and one low-resiliency population area in New Jersey. We do not project that any Boltonia montana populations will be extirpated; therefore, the species would maintain redundancy within each metapopulation and across its range. Because we do not project that either the New Jersey or Virginia metapopulations will be extirpated, the species’ representation would not change from the current condition.

Under scenario 3, we assume no new conservation measures are implemented and habitat disturbances increase at most privately-owned Boltonia montana populations. These

disturbances result from plausible increases in residential development, agricultural intensity (e.g., livestock grazing, crop production), forest clearing, and other disturbances associated with human development. Additionally, under this scenario, we project that by 2050, invasive plant species spread from ponds where they have already been identified into adjacent sites, increasing competitive pressure on these populations. The results modeled under this scenario represent perhaps the worst possible outcome for the species.

The results of the scenario 3 model indicate three population areas (two in New Jersey, one in Virginia) would maintain high resiliency, one New Jersey population area would maintain moderate resiliency, and four population areas (two in each state) would have low resiliency. Additionally, we project several low-resiliency populations could be extirpated, perhaps resulting in the loss of a population area in Virginia. Therefore, we conclude the species would have low redundancy in Virginia and maintain moderate redundancy in New Jersey. Because we do not project that either the New Jersey or Virginia metapopulations would be extirpated under this scenario, the species’ representation would not change from the current condition.

Overall Summary

Boltonia montana occurs in certain isolated sinkhole ponds that have widely fluctuating water levels. The species’ reproductive strategies are adapted to these variable conditions and population numbers can vary greatly from year to year. The best available information indicates B. montana’s distribution has historically been limited to three disjunct metapopulations in New Jersey, Pennsylvania, and Virginia, with the Pennsylvania metapopulation currently presumed to be extirpated. There are 21 and 22 populations in the New Jersey and Virginia metapopulations, respectively; however, sparse or inconclusive survey data make it uncertain if the species is extant or extirpated at many of these sites. Likewise, there is insufficient data to determine potential trends in B. montana abundance.

Because survey data are sparse, we developed an assessment model using habitat metrics to estimate the condition of each population and compare them with each other, both currently and under plausible future scenarios. There is significant uncertainty regarding the species’ potential response to changing temperature and precipitation patterns. Therefore, we do not directly assess the potential effects of climate change to the population sites but instead consider their relative condition as an indicator of their capacity to persist in light of changing climate conditions.

Under the three future scenarios, we predict various plausible changes in the stressors, habitat condition, and conservation measures affecting Boltonia montana. These changes in turn are predicted to affect the resiliency of the species at the population and population area scale. Under scenarios 1 and 3, by 2050, between 3 and 11 populations are predicted to have lower resiliencies than the current condition, and under scenario 3, the changes may result in the extirpation of several low resiliency populations, perhaps causing a loss of redundancy. Under scenario 2, we predict feasible conservation efforts would improve the condition of 22 populations, significantly improving the species resiliency in three population areas. Under all scenarios, the species would remain extant in the New Jersey and Virginia metapopulations; therefore, its representation is not predicted to change from the current condition (though we

82 note that the historical extirpation of the Pennsylvania metapopulation may have reduced the species representation).

REFERENCES CITED

Anderson, L.C. 1987. Boltonia apalachicolensis (Asteraceae): a new species from Florida. Systemic Botany Vol. 12, No. 1.

Aspelin, A.L. 2003. Pesticide use in the United States: trends during the 20th century. Center for Integrated Pest Management, North Carolina State University, Raleigh, North Carolina.

Augusta County. 2007. Augusta County comprehensive plan update 2007-2027, Volume 1; April 25, 2007, as amended January 28, 2009. Augusta County, Virginia

Baskin, C.C. and J.M. Baskin. 2002. Achene germination ecology of the federally threatened floodplain endemic Boltonia decurrens (Asteraceae). American Midland Naturalist 147(1).

Bellemare, J., G. Motzkin, and D.R. Foster. 2002. Legacies of the agricultural past in the forested present: an assessment of historical land-use effects on rich mesic forests. Journal of Biogeography Vol. 29.

Boutin, C., B. Strandberg, D. Carpenter, S.K. Mathiassen, and P.J. Thomas. 2013. Herbicide impact on non-target plant reproduction: what are the toxicological and ecological implications? Environmental Pollution Vol. 185.

Buffon 1799. Histoire naturelle, generale et particuliere, des plantes.

Burkett, V. and J. Kusler. 2000. Climate change: potential impacts and interactions in wetlands of the United States. Journal of the American Water Resources Association Vol. 36, No. 2.

Cane, J.H. and B. Love. 2018. Pollen carryover between sequential foraging trips by a solitary bee: implications for distant outcrossing. Journal of Pollination Ecology Vol. 24, No. 15.

Center for Conservation Biology (CBD). 2010. Petition to list 404 aquatic, riparian and wetland species from the southeastern United States as threatened or endangered under the Endangered Species Act. Sent to Mr. Ken Salazar, Secretary of the Interior April 20, 2010.

Central Shenandoah Planning District Commission. 2011. Upper Shenandoah water supply plan, August 2011.

Chen, Y.L., Y.S. Chen, L. Brouillet, and J.C. Semple. 2011. Chapter 8 Tribe Astereae in Flora of China Volume 20-21 (Asteraceae), Z.Y. Wu P.H. Raven, and D.Y. Hong, eds. Science Press and Missouri Botanical Garden Press.

DeWoody, J., J.D. Nason, and M. Smith. 2004. Inferring demographic processes from the

genetic structure of a metapopulation of Boltonia decurrens (Asteraceae). Conservation Genetics 5.

DeWoody, J., J.D. Nason, and M. Smith. 2011. Hybridization between the threatened herb Boltonia decurrens (Asteraceae) and its widespread congener, B. asteroides. Botany Vol. 89.

Fernandez-Cornejo, J., R. Nehring, C. Osteen, S. Wechsler, A. Martin, and A. Vialou. 2014. Pesticide use in U.S. agriculture: 21 selected crops, 1960-2008, EIB-124. U.S. Department of Agriculture, Economic Research Service, May 2014.

Ferrer, M.M. and S.V. Good-Avila. 2007. Macrophylogenetic analyses of the gain and loss of self-incompatibility in the Asteraceae. New Phytologist Vol. 173.

Flann, C. (ed). 2009. Global Compositae Checklist; Accessed November 27, 2019.

Fleming, G.P. and N.E. Van Alstine. 1999. Plant communities and floristic features of sinkhole ponds and seepage wetlands in southeastern Augusta county, Virginia. Banisteria, No. 13, 1999.

Flora of North America (FNA). 2020. (http://beta.floranorthamerica.org; Accessed February 12, 2020).

Goad, R. 2020. Pennsylvania Department of Conservation and Natural Resources comments on draft SSA. Received via electronic mail from Greg Podniesinski, Pennsylvania Department of Conservation and Natural Resources February 14, 2020.

Gove, B., S.A. Power, G.P. Buckley, and J. Ghazoul. 2007. Effects of herbicide spray drift and fertilizer overspread on selected species of woodland ground flora: comparison between short-term and long-term impact assessments and field surveys. Journal of Applied Ecology Vol. 44.

Grossman, D. H., D. Faber-Langendoen, A. S. Weakley, M. Anderson, P. Bourgeron, R. Crawford, K. Goodin, S. Landaal, K. Metzler, K. D. Patterson, M. Pyne, M. Reid, and L. Sneddon. 1998. International classification of ecological communities: terrestrial vegetation of the United States. Volume I. The National Vegetation Classification System: development, status, and applications. The Nature Conservancy, Arlington, Virginia, USA.

Gu, H. and P.C. Hoch. 1997. Systemics of Kalimeris (Asteraceae: Astereae). Annals of the Missouri Botanical Garden. Vol. 84: 762-814.

Integrated Taxonomic Information System (ITIS) 2019. (https://www.itis.gov; Accessed: December 10, 2019).

Jackson, C.R., J.A. Thompson, and R.K. Kolka. 2014. Wetland soils, hydrology, and geomorphology. In: Batzer, D. and Sharitz, R., eds. Ecology of freshwater and estuarine 85

wetlands. Berkeley, CA: University of California Press: Chapter 2

Jackson, P.W. and J.S. Miller. 2015. Developing a world flora online – a 2020 challenge to the world’s botanists from the international community. Rodriguesia Vol. 66, No. 4.

Johnson, E.A. and K.S. Walz. 2013. Integrated management guidelines for four habitats and associated State endangered plants and wildlife species of greatest conservation need in the skylands and pinelands landscape conservation zones of the New Jersey State Wildlife Action Plan. American Museum of Natural History, Center for Biodiversity and Conservation and New Jersey Department of Environmental Protection, Natural Heritage Program, for NatureServe, Arlington, VA.

Karaman-Castro, V. and L.E. Urbatsch. 2006. Boltonia. In: Flora of North America, Vol. 20: Mangoliophyta: Asteraceae (in part). Edited by Flora of North America Editorial Committee, Oxford University Press, New York. pp. 353–357 (http://beta.floranorthamerica.org/Boltonia, Accessed January 21, 2020).

Kartesz, J.T. 2015. The Biota of North America Program (BONAP). Taxonomic Data Center. (http://www.bonap.net/tdc; Accessed December 10, 2019). Chapel Hill, N.C.

Kleijn, D. and G.I.J. Snoeijing. 1997. Field boundary vegetation and the effects of agrochemical drift: botanical change caused by low levels of herbicide and fertilizer. Journal of Applied Ecology Vol. 34.

Knox, J.S. 1997. A nine year demographic study of Helenium virginicum (Asteraceae), a narrow endemic seasonal wetland plant. Journal of the Torrey Botanical Society Vol. 124, No. 3.

Kottek, M., J. Grieser, C. Beck, B. Rudolf, and F. Rubel. 2006. World map of the Koppen- Geiger climate classification updated. Meteorologische Zeitschrift, Vol. 15, No. 3.

L’Heritier de Brutelle, C.L. 1788. Sertum Anglicum, seu, plantae rariores quae in hortis Juxta Londinum imprimis in horto Regio Kewensi excoluntur, ab anno 1786 ad annum 1787 observata.

Li, W.P., F.S. Yang, T. Jivkova, and G.S. Yen. 2012. Phylogenetic relationships and generic delimitation of Eurasian Aster (Asteraceae: Astereae) inferred from ITS, ETS, and trnL-F sequence data. Annals of Botany. Vol. 109, 1341-1357.

Lichvar, R.W., N.C. Melvin, M.L. Butterwick, and W.N. Kirchner. 2012. National wetland plant list indicator rating definitions. ERDC/CRREL TN-12-1. U.S. Army Corps of Engineers, Wetland Regulatory Assistance Program. Washington, DC.

Lichvar, R.W., D.L. Banks, W.N. Kirchner, and N.C. Melvin. 2016. The National Wetland Plant List: 2016. Phytoneuron 2016-30: 1-17.

Linnaeus 1767a. Mantissa plantarum. Generum editionis VI; specierum editionis II.

Linnaeus 1767b. Systema naturae.

Maynard, J.P. 2012. Water use in the Shenandoah Valley of Virginia 1982-2010, Technical bulletin 12-02. Virginia Department of Environmental Quality, Office of Water Supply.

McCauley, L.A., M.J. Anteau, M.P. van der Burg, and M.T. Wiltermuth. 2015. Land use and wetland drainage affect water levels and dynamics of remaining wetlands. Ecosphere Vol. 6, No. 6.

McClain, W.E., R.D. McClain, and J.E. Ebinger. 1997. Flora of temporary sand ponds in Cass and Mason Counties, Illinois. Castanea 62(2).

Messmore, N.A. and J.S. Knox. 1997. The breeding system of the narrow endemic, Helenium virginicum (Asteraceae). Journal of the Torrey Botanical Society Vol. 124, No. 4.

Mettler, P.A., M. Smith, and K. Victory. 2001. The effects of nutrient pulsing on the threatened, floodplain species, Boltonia decurrens. Plant Ecology 155.

Mettler-Cherry, P.A., M. Smith, and T.M. Keevin. 2006. Habitat characterization and geospatial metapopulation dynamics of threatened floodplain species Boltonia decurrens using a GIS. Wetlands Vol. 26, No. 2.

Mettler-McClure, P.A. 1997. Determination of seed bank composition for sites of an endangered floodplain species, Boltonia decurrens. Project #97043, final report submitted to: The Illinois Department of Natural Resources, Division of Natural Heritage July 15, 1997.

Morris, K. and P. Reich. 2013. Understanding the relationship between grazing and wetland condition. Arthur Rylah Institute for Environmental Research Technical Report Series No. 252. Department of Environment and Primary Industry, Heidelberg, Victoria.

National Oceanic and Atmospheric Administration (NOAA). 2019. National Centers for Environmental information, Climate at a Glance: County Time Series (https://www.ncdc.noaa.gov; Accessed October 21, 2019).

NatureServe. 2018. NatureServe Explorer: An online encyclopedia of life, Version 7.1. NatureServe, Arlington, Virginia. (http://explorer.natureserve.org; Accessed December 10, 2019).

Nesom, G. 1994. Review of the taxonomy of Aster sensu lato (Asteraceae: Astereae), emphasizing the new world species. Phytologia Vol. 77, No. 3.

Nesom, G. 2000. New subtribes for North American Astereae (Asteraceae). SIDA, Contributions to Botany Vol. 19 No. 2.

Nesom, G. and H. Robinson. 2007. Tribe Astereae. In: The families and genera of vascular plants, Volume VIII, flowering plants – eudicots, asterales. J.W. Kadereit and C. Jeffrey, eds. Springer-Verlag, Berlin.

New Jersey Administrative Code (NJAC). 2010. NJAC 7:5A Natural Areas and the Natural Areas System Rules, last amended June 21, 2010.

NJAC. 2013. NJAC 7:5C Endangered plant species program rules, last amended September 3, 2013.

NJAC. 2019a. NJAC 7:7A Freshwater wetlands protection act rules, last amended July 15, 2019.

NJAC. 2019b. NJAC 7:13 Flood hazard area control act rules, last amended July 15, 2019.

New Jersey Department of Environmental Protection (NJDEP). 2002. Policy directive 2002-001, off-road vehicle use. (https://www.nj.gov/dep/commissioner/policy/orvpolicy.htm; accessed December 13, 2019).

NJDEP. 2016. List of endangered plant species and plant species of concern. NJDEP, Division of Parks and Forestry. Trenton, New Jersey; (https://www.nj.gov/dep/parksandforests/natural/heritage/spplant.html; Accessed November 27, 2018).

NJDEP. 2018. White Lake Wildlife Management Area map (https://www.nj.gov/dep/fgw/pdf/wmamaps/white_lake.pdf; accessed December 23, 2019).

NJDEP. 2019a. Boltonia montana occurrence data table. Submitted by Roman Senyk, NJ Department of Environmental Protection to Ron Popowski, US Fish and Wildlife Service via electronic mail September 9, 2019.

NJDEP. 2019b. Swartswood State Park website. (https://www.state.nj.us/dep/parksandforests/parks/swartswood.html; accessed December 31, 2019).

Noyes, R.D. and L.H. Rieseberg. 1999. ITS sequence data support a single origin for North American Astereae (Asteraceae) and reflect deep geographic divisions in Aster S.L. American Journal of Botany Vol. 86, No. 3.

Ouren, D.S., C. Haas, C.P. Melcher, S.C. Stewart, P.D. Ponds, N.R. Sexton, L. Burris, T. Fancher, and Z.H. Bowen. 2007. Environmental effects of off-highway vehicles on Bureau of Land Management lands: a literature synthesis, annotated bibliographies, extensive bibliographies, and internet resources. U.S. Geological Survey, Open-file report 2007-1353.

Pennsylvania Bulletin. 2018. Pennsylvania Bulletin Volume 48, Number 51. December 22, 2018. Harrisburg, PA.

Pennsylvania Natural Heritage Program. 2020. Environmental review list, last updated January 2020. http://www.naturalheritage.state.pa.us/Species.aspx; accessed January 21, 2020.

ProximityOne. 2019. Website (http://proximityone.com/demographics2060.htm; accessed November 7, 2019).

Ring, R.M., E.A. Spencer, and K.S. Walz. 2013. Vulnerability of 70 plant species of greatest conservation need to climate change in New Jersey. New York Natural Heritage Program, Albany, NY and New Jersey Natural Heritage Program, Department of Environmental Protection, Trenton, NJ.

Romero-Lankao, P., J.B. Smith, D.J. Davidson, N.S. Diffenbaugh, P.L. Kinney, P. Kirshen, P. Kovacs, and L. Villers Ruiz. 2014. North America. In: Climate Change 2014: impacts, adaptation, and vulnerability. Part B: regional aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Barros, V.R., C.B. Field, D.J. Dokken, M.D. Mastrandrea, K.J. Mach, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White (eds.). Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Saunders, L.E. and R. Pezeshki. 2015. Glyphosate in runoff waters and in the root-zone: a review. Toxics Vol. 3.

Schmolke, A., R. Brain, P. Thorbek, D. Perkins, and V. Forbes. 2018. Assessing and mitigating simulated population-level effects of 3 herbicides to a threatened plant: application of a species-specific population model of Boltonia decurrens. Environmental Toxicology and Chemistry, Vol. 37, No. 6.

Schwegman, J.E. and R.W. Nyboer. 1985. The taxonomic and population status of Boltonia decurrens (Torr. & Gray) Wood. Castanea Vol. 50, No. 2.

[Service] U.S. Fish and Wildlife Service. 2016. USFWS Species Status Assessment Framework: an integrated analytical framework for conservation. Version 3.4 dated August 2016.

[Service] U.S. Fish and Wildlife Service. 2019. National Wetlands Inventory website. U.S. Department of the Interior, Fish and Wildlife Service, Washington, D.C. (http://www.fws.gov/wetlands/; accessed: August 22, 2019.

Shaffer, M., L.H. Watchman, W.J. Snape, and I.K. Latchis. 2002. Population viability analysis and conservation policy, in Population Viability Analysis, Beissinger and

McCullough, eds. University of Chicago Press.

Smith, D.R., N.L. Allan, C.P. McGowan, J.A. Szymanski, S.R. Oetker, and H.M. Bell. 2018. Development of a species status assessment process for decisions under the U.S. Endangered Species Act. Journal of Fish and Wildlife Management Vol. 9, Issue 1.

Smith, M., T. Brandt, and J. Stone. 1995. Effect of soil texture and microtopography on germination and seedling growth in Boltonia decurrens (Asteraceae), a threatened floodplain species. Wetlands Vol. 15, No. 4.

Smith, M., H. Caswell, and P. Mettler-Cherry. 2005. Stochastic flood and precipitation regimes and the population dynamics of a threatened floodplain plant. Ecological Applications 15(3).

Smith, M. and E. Hunsley. 2006. Effects of achene pattern, achene density and competition on germination in Boltonia decurrens, a threatened plant of the Illinois River valley. Journal of the Torrey Botanical Society 133(4).

Smith, M. and T.M. Keevin. 1998. Achene morphology, production and germination, and potential for water dispersal in Boltonia decurrens (decurrent false aster), a threatened floodplain species. Rhodora, Vol. 100, No. 901.

Smith, M., T. Keevin, P. Mettler-McClure, and R. Barkau. 1998. Effect of the flood of 1993 on Boltonia decurrens, a rare floodplain plant. Regilated Rivers: Research & Management (14).

Smith, M. and P. Mettler. 2002. The role of the flood pulse in maintaining Boltonia decurrens, a fugitive plant species of the Illinois River floodplain: a case history of a threatened species. In: Flood pulsing in wetlands: restoring the natural hydrological balance, Chapter 4, pp. 109–144. Edited by B.A. Middleton. John Wiley & Sons, Inc

Smith, M. and J.S. Moss. 1998. An experimental investigation, using stomatal conductance and fluorescence, of the flood sensitivity of Boltonia decurrens and its competitors. Journal of Applied Ecology 35.

Smith, M., Y. Wu, and O. Green. 1993. Effect of light and water-stress on photosynthesis and biomass production in Boltonia decurrens (Asteraceae), a threatened species. American Journal of Botany Vol. 80, No. 8.

Stoecker, M.A., M. Smith, and E.D. Melton. 1995. Survival and aerenchyma development under flooded conditions of Boltonia decurrens, a threatened floodplain species and Conyza canadensis, a widely distributed competitor. American Midland Naturalist 134.

Sussex County New Jersey. 2014. Sussex County strategic growth plan update. Report submitted to Sussex County Administrative Center, Newton, NJ, November 2014.

Sussex County New Jersey. 2019. Natural resources inventory. https://www.sussex.nj.us/cn/webpage.cfm?TPID=6105; Accessed November 26, 2019.

Swearingen, J. and C. Bargeron. 2016. Invasive Plant Atlas of the United States. University of Georgia Center for Invasive Species and Ecosystem Health. (http://www.invasiveplantatlas.org/; Accessed December 31, 2019).

Switalski, A. and A. Jones. 2012. Off-road vehicle best management practices for forestlands: a review of scientific literature and guidance for managers. Journal of Conservation Planning Vol. 8.

The Global Compositae Checklist (TICA) 2019. International Compositae Alliance (https://compositae.landcareresearch.co.nz/Default.aspx; Accessed: December 10, 2019).

The Nature Conservancy (TNC). 2019. Muckshaw Ponds Preserve website. https://www.nature.org/en-us/get-involved/how-to-help/places-we-protect/muckshaw- ponds-preserve; Accessed: December 31, 2019.

Townsend, J.F. 2013. Recognition of Boltonia asteroides var. glastifolia and reestablishment of a more restrictive taxonomic concept for Boltonia asteroides var. asteroides. Phytoneuron 2013-9.

Townsend, J.F. 2018. Natural heritage resources of Virginia: rare plants. Natural Heritage Technical Report 18-11. Virginia Department of Conservation and Recreation, Division of Natural Heritage. Richmond, Virginia.

Townsend, J.F. 2019. Personal communication between John Townsend, VA Natural Heritage Program and Keith Hastie, US Fish and Wildlife Service. Telephone call December 5, 2019.

Townsend, J.F. 2020a. Virginia Department of Conservation and Recreation comments on draft SSA. Received via electronic mail February 12, 2020.

Townsend, J.F. 2020b. Personal communication between John Townsend, VA Natural Heritage Program and Keith Hastie, US Fish and Wildlife Service. Electronic mail February 20, 2020.

Townsend, J.F. and V. Karaman-Castro. 2006. A new species of Boltonia (Asteraceae) from the ridge and valley physiographic province, U.S.A. SIDA, Contributions to Botany, Vol. 22, No. 2.

U.S. Census. 2019. Website (https://www.census.gov/population/www/censusdata/cencounts/ and https://www.census.gov/library/publications/2011/compendia/usa-counties- 2011.html#LND; accessed November 7, 2019).

U.S. Climate Data, Version 2.3. 2019. Website (https://www.usclimatedata.com/; accessed December 12, 2019).

[USFS] U.S. Forest Service. 2014. George Washington National Forest revised land and resource management plan, R8-MB 143 A. U.S. Department of Agriculture.

[USFS] U.S. Forest Service. 2015. George Washington and Jefferson National Forests motor vehicle use map 2015, FS-873-808. U.S. Department of Agriculture.

[USFS] U.S. Forest Service. 2019. Boltonia montana survey data from Maple Flats, Virginia. Submitted by Mike Donahue, US Forest Service to Keith Hastie, US Fish and Wildlife Service October 4, 2019 via electronic mail.

[USGS] U.S. Geological Survey. 2018. Gap Analysis Project, Protected Areas Database of the United States (PADUS) version 2.0.

Van Alstine, N.E. 2019. A 2018 status survey for Boltonia montana Townsend & Karaman- Castro (valley doll’s-daisy) in Augusta County, Virginia. Natural Heritage Technical Report 19-02, prepared for the Virginia Department of Agriculture and Consumer Services, Richmond, Virginia.

Virginia Administrative Code (VAC). 2008. Endangered plant and insect species. Code of Virginia Title 3.2, Chapter 10 § 3.2-1003.

VAC. 2013. Listing of endangered and threatened plant and insect species. Code of Virginia Title 2, Chapter 320.

VAC. 2016. Rules for modification, revocation and reissuance, extension, transfer, and termination of VWP individual permits. Code of Virginia Title 9, 9VAC25-210-180.

Visscher, P.K. and T.D. Seeley. 1982. Foraging strategy of honeybee colonies in a temperate deciduous forest. Ecology Vol. 63 No. 6.

Walz, K.S. 2019a. Rapid data assessment questionnaire for Boltonia montana. Submitted by Kathleen Walz, NJ Natural Heritage Program to Ron Popowski US Fish and Wildlife Service February 4, 2019 via electronic mail.

Walz, K.S. 2019b. Draft summary of 2019 Boltonia montana surveys in New Jersey. Submitted by Kathleen Walz, NJ Natural Heritage Program to Keith Hastie, US Fish and Wildlife Service December 19, 2019 via electronic mail.

Walz, K.S. 2020. New Jersey Department of Environmental Protection comments on draft SSA. Received via electronic mail February 16, 2020.

Walz, K.S., R.J. Canace, J. Boyle, R. Witte, M.S. Serfes, W. Honachefsky, J. Kurtz, R. Dutko. 2001. Identification and protection of reference wetland natural communities in

New Jersey: calcareous sinkhole ponds of the Kittatinny Valley. New Jersey Department of Environmental Protection, Division of Parks and Forestry, Office of Natural Lands Management, Natural Heritage Program, Trenton, NJ.

Wardrop, D.H., A.T. Hamilton, M.Q. Nassry, J.M. West, and A.J. Britson. 2019. Assessing the relative vulnerabilities of Mid-Atlantic freshwater wetlands to projected hydrologic changes. Ecosphere Vol. 10, No. 2.

Weakley, A.S. 2015. Flora of the southern and mid-Atlantic states (working draft of 21 May 2015). University of North Carolina Herbarium, North Carolina Botanical Garden, Chapel Hill, NC.

Wolf, S., B. Hartl, C. Carroll, C. Maile, and D.N. Greenwald. 2015. Beyond PVA: why recovery under the Endangered Species Act is more than population viability. Bioscience.

World Flora Online (WFO). 2020. World flora online, published on the internet (http://www.worldfloraonline.org/; Accessed January 20, 2020).

APPENDIX A. Taxonomic information demonstrating inconsistent recognition of Boltonia montana in various databases and taxonomic systems.

The Flora of North America (FNA) is a comprehensive account of more than 20,000 plant species native to, or naturalized in, North America (north of Mexico) and is generally regarded as the taxonomic authority for the vascular flora of North America (FNA 2019, website). In the FNA, taxonomic treatments are written and reviewed by experts from the systematic botanical community and are based on original observations of living and herbarium specimens supplemented by a crucial review of the literature. Each treatment provides scientific and common names, pertinent synonymy, taxonomic descriptions, identification keys, distribution maps, summaries of habitat and geographic ranges, and other relevant biological information.

In 2006, the FNA published its most recent taxonomic review of the genus Boltonia, accepting five species: B. apalachicolensis, B. asteroides, B. caroliniana, B. decurrens, and B. diffusa (Karaman-Castro and Urbatsch 2006, pp. 353–357). Additionally, the FNA recognized the three previously mentioned varieties of B. asteroides (and two varieties of B. diffusa). While B. montana was not referenced by name, the editors did note that populations of B. asteroides var. asteroides restricted to sinkhole habitats in New Jersey and Virginia were physically distinct from populations of the variety collected elsewhere and that the taxonomic status of these populations warranted further study (Karaman-Castro and Urbatsch 2006, p. 355). Presumably after the FNA treatment went to print in 2006, Townsend and Karaman-Castro (2006, entire) published their taxonomic treatment describing these populations as B. montana. As of 2019, the FNA has yet to publish a revised taxonomic treatment of the genus Boltonia, therefore B. montana remains unclassified by this authority.

The World Flora Online (WFO) (and its direct precursor The Plant List, Ver. 1.1; http://www.theplantlist.org) is a working list of the world’s plant species compiled and maintained by a consortium of botanical institutions (Jackson and Miller 2015, entire). The WFO incorporates information from existing floras and databases including the FNA and The International Compositae Alliance (TICA, discussed below), published research, and input from experienced botanists and taxonomic specialists. Currently, the WFO lists five species of Boltonia from North America (the same species listed in the FNA, see above), and one species, B. lautureana, from East Asia. However, as discussed previously, the East Asian species is now understood as not belonging to the genus Boltonia. The WFO acknowledges B. montana (referencing Townsend and Karaman-Castro (2006) and TICA) but lists its taxonomic status as “unresolved,” meaning that it is not yet possible to assign a status of either “accepted” or “synonym” (WFO 2020).

The draft Flora of the Southern and Mid-Atlantic States treats approximately 7,000 taxa based on herbarium specimens, the scientific literature (published and gray), and consultation with a network of botanists and taxonomic experts (Weakley 2015, p. 6). This flora recognizes a total of “about 6–7” species of Boltonia; however, only the five that are native to the southeast and mid-Atlantic region are treated, including B. montana (Weakley 2015, pp. 1,093–1,094). John Townsend, the first author of the 2006 taxonomic description of B. montana, is co-contributor to this treatment.

The Global Compositae Checklist, which is maintained by TICA, is an online database of taxonomic information specific to the Compositae (alternatively Asteraceae) family (TICA 2019, entire). The database is a compilation of many contributed datasets and is continually edited and updated by botanists and taxonomists with expertise in this large plant family. The Checklist recognizes the five North American Boltonia species listed above and, referencing Townsend and Karaman-Castro (2006), lists B. montana as “accepted” (TICA 2019, entire).

The Integrated Taxonomic Information System (ITIS) is a joint project of multiple United States, Canadian, and Mexican governmental agencies, NatureServe, the Smithsonian National Museum of Natural History, and other organizations and taxonomists. ITIS relies on taxonomic data submitted by the world scientific community, which is reviewed by ITIS data stewards and other cooperating specialists in the systematics community. The ITIS database accepts the five previously mentioned North American Boltonia species, but does not acknowledge B. montana (ITIS 2019, website).

The Biota of North America Program (BONAP) is a collaborative effort to document the entire vascular flora of the North American continent, north of Mexico (Kartesz 2015, entire). Since 1980, information from the published scientific literature, natural heritage programs (NHPs), vouchered specimens, individual contributors, and other sources have been incorporated into a database of North American Vascular Flora, the Taxonomic Data Center (TDC). BONAP accepts the five species of Boltonia listed in the FNA along with B. montana; however, it does not provide a specific reference to support the taxon’s recognition.

NatureServe Explorer® is a database of detailed conservation information on more than 70,000 plants, animals, and habitats of the United States and Canada (NatureServe 2018, entire). The data are collected and compiled by a network of state NHPs and other botanists, zoologists, ecologists, and information specialists. NatureServe Explorer® accepts the taxonomy of the five previously mentioned North American Boltonia species and, referencing Townsend and Karaman-Castro (2006, entire), lists the taxonomic status of B. montana as “provisionally accepted,” meaning that it has yet to be formally addressed in NatureServe’s standard classification (NatureServe 2018, website).

APPENDIX B. Location and general characteristics of all known Boltonia montana population sites. Population ID numbers correspond to numbers on the relevant map following each table.

Location Area Land Proximity ID State County Landscape Setting Population Area Population Ha Ac Owner (km)

1 Duck Pond NJ Sussex 8.0 19.8 NJDEP 0.45 Pond surrounded by woodlands.

2 Frog Pond NJ Sussex 0.6 1.5 NJDEP 0.05 Pond surrounded by woodlands: road along northwest shore.

3 Swartswood Little Frog Pond NJ Sussex 0.3 0.7 NJDEP 0.05 Pond surrounded by woodlands.

4 Spring Lake NJ Sussex 6.0 14.8 NJDEP 0.28 Pond surrounded by woodlands; some residences to the east.

5 Swallowhole Pond NJ Sussex 0.5 1.3 Private 1.30 Pond in woodland setting with pasture/cultivated fields on southern edge.

6 Muckshaw Pond 1 NJ Sussex 0.6 1.5 NGO 0.18 Pond surrounded by woodlands.

7 Muckshaw Pond 2 NJ Sussex 1.2 3.0 NGO 0.17 Pond surrounded by woodlands.

8 Muckshaw Pond 3 NJ Sussex 1.7 4.2 Private 0.15 Pond surrounded by woodlands; residence on northeast edge. Muckshaw 9 Muckshaw Pond 5 NJ Sussex 0.8 2.0 NGO 0.04 Pond surrounded by woodlands.

10 Muckshaw Pond 6 NJ Sussex 1.8 4.4 NGO 0.04 Pond surrounded by woodlands.

11 Muckshaw Pond 7 NJ Sussex 0.1 0.2 NGO 0.05 Pond in woodland setting; agricultural fields to the south.

12 Piggyback Pond NJ Warren 1.4 3.5 NJDEP 1.40 Pond in woodland setting; agricultural fields to the east.

13 White Lake Catfish Pond NJ Sussex 6.2 15.3 Private 1.25 Pond in woodland setting; road and several residences at northern edge.

Adjacent ponds surrounded by woodlands; mowed residential yard at northern 14 Shuster Ponds NJ Warren 8.0 19.8 Private 1.25 end; commercial springwater bottling. Huntsburg Sinkhole 15 NJ Sussex 3.6 8.9 Private 0.09 Pond in woodland/ agricultural setting. East Huntsburg Sinkhole 16 NJ Sussex 0.5 1.2 Private 0.09 Pond in woodland/ agricultural setting. West Johnsonburg Sinkhole 17 Greendell NJ Warren 0.2 0.5 Private 2.25 Pond surrounded by woodlands; thick canopy cover. Ponds Sussex/ 18 Four-Angle Pond NJ 1.7 4.2 Private 0.50 Pond surrounded by woodland; several residences on eastern edge. Warren Pleistocene Cave 19 NJ Warren 2.2 5.4 Private 0.50 Pond in woodland setting. Sinkhole Adjacent ponds in mixed ag and residential area; narrow wooded buffer; 20 Huntsville Turtle Sinkhole Ponds NJ Sussex 6.3 15.6 Private 4.70 bordered by roads to the north and west; railroad to the south and east. Adjacent ponds in wooded area; railroad along eastern edge, some residential 21 Lake Grinnell Lake Grinnell Ponds NJ Sussex 3.6 8.9 Private 13.50 to the west.

Location Area Land Proximity ID State County Landscape Setting Population Area Population Ha Ac Owner (km)

1 Spring Pond VA Augusta 3.4 8.4 USFS 0.15 Pond surrounded by forest (USFS Special Biological Area).

2 Twin Ponds #11, 12 VA Augusta 0.8 2.0 USFS 0.15 Pond surrounded by forest (USFS Special Biological Area).

North Maple Flat 3 VA Augusta 0.4 1.0 USFS 0.03 Pond surrounded by forest (USFS Special Biological Area). Pond A North Maple Flat 4 VA Augusta 1.7 4.2 USFS 0.03 Pond surrounded by forest (USFS Special Biological Area). Pond B Maple Flats 5 Oak Pond #13 VA Augusta 0.3 0.7 USFS 0.15 Pond surrounded by forest (USFS Special Biological Area).

Kennedy Mountain 6 VA Augusta 1.2 3.0 Private 0.25 Pond surrounded by forest. Meadow

7 Deep Pond VA Augusta 0.5 1.2 USFS 0.25 Pond surrounded by forest (USFS Special Biological Area).

8 Split Level Pond VA Augusta 5.8 14.3 USFS 0.55 Pond surrounded by forest (USFS Special Biological Area).

9 Campbells Pond VA Augusta 1.3 3.2 Private 0.65 Modified pond surrounded by a cultivated ag field.

10 Campbells Grove Farm Pond VA Augusta 0.8 2.0 Private 0.35 Modified pond surrounded by grazed pasture.

Grove Farm Pond 11 VA Augusta 0.1 0.2 Private 0.35 Modified pond surrounded by pasture and woodlands. East

12 Abshire Pond VA Augusta 0.7 1.7 Private 0.35 Pond adjacent to residences and surrounded by pasture and woodlands.

13 I VA Augusta 0.4 1.0 Private 0.50 Pond surrounded by pasture and woodlands.

Route 664 Meadow 14 VA Augusta 0.3 0.7 Private 0.13 Linear drainage swale surrounded by mowed pasture and residences. East Route 664 Meadow 15 VA Augusta 0.2 0.5 Private 0.13 Modified pond surrounded by mowed pasture and residences. West

16 Lyndhurst Sherlynd Church VA Augusta 0.2 0.5 Private 0.35 Modified pond adjacent to church, residences, and an ag field.

17 Wood Duck Pond VA Augusta 1.0 2.5 Private 0.50 Pond adjacent to pasture, surrounded by woodlands.

18 Anderson Pond VA Augusta 0.1 0.2 Private 0.60 Small pond in wooded area adjacent to residences.

Fence Line Ponds (A, 19 VA Augusta 0.7 1.7 Private 0.60 Small pond complex surrounded by woodlands. B, and South) Lyndhurst Pond #18- 20 VA Augusta 2.3 5.7 Private 0.95 Adjacent to residences and grazed pasture. 15

21 South River South River VA Augusta Unk Unk Private Unk Unknown

22 Hattons Peterson Pond VA Augusta 0.3 0.7 Private 6.50 Small, disturbed pond adjacent to residences and pasture.

APPENDIX C. Years indicating when Boltonia montana occurrence data are available, for the 30-year period 1980 to 2009.

**N=80; not shown are 13 occurrence reports from the 115-year period 1864 to 1979. **Green shading indicates the species was confirmed during that particular year and an estimated population size was provided. Green shading with cross- hatching indicates the species was confirmed but no estimate of plant numbers was provided. Red shading indicates the site was surveyed but B. montana was not reported. Gray shading indicates no survey information available. Historical Period Current Period 1980-89 1990-99 2000-09 2010-19 Population Area Population 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

Duck Pond 1 1 1 1 1 1

Frog Pond 1 1 1

Little Frog Pond 1 1 1 Swartswood Spring Lake 1 1 Swartswood Lake

Swallowhole Pond 1 1

Muckshaw Pond 1 1

Muckshaw Pond 2 1 1

Muckshaw Pond 3 1 1 Muckshaw Muckshaw Pond 5 1 1 1 1

Muckshaw Pond 6 1 1

Muckshaw Pond 7 1 1 1

1 1 New Jersey New Piggyback Pond

White Lake Catfish Pond 1

Shuster Ponds 1

Huntsburg Sinkhole East 1 1 1

Huntsburg Sinkhole West 1 1 1

Greendell Johnsonburg Ridge Sinkholes 1

Four-Angle Pond 1 1

Pleistocene Cave Sinkhole 1

Huntsville Turtle Sinkhole Ponds 1 Lake Grinnell Lake Grinnell Ponds PA Susquehanna R. Susquehanna River

Spring Pond 1 1 1

Twin Ponds #11, 12 1 1 1 1 1 1 1

North Maple Flat Pond A 1 1 1

North Maple Flat Pond B 1 1 Maple Flats Oak Pond #13 1 1 1 1 1

Kennedy Mountain Meadow 1 1 1 1 1

Deep Pond 1

Split Level Pond 1 1 1 1 1

Cambells Pond 1 1 1 1 1

Cambells Grove Farm Pond 1 1 1 1

Grove Farm Pond East 1 1

Abshire Pond 1 1 1 1 1 1 Virginia

I 1 1 1

Route 664 Meadow East 1 1

Route 664 Meadow West 1 1

Lyndhurst Sherlynd Church 1 1

Wood Duck Pond 1 1 1 1

Anderson Pond 1 1 1

Fence Line Ponds (A, B, and South) 1 1 1 1

Lyndhurst Pond #18-15 1 1 1 1 1 South River South River Hattons Peterson Pond 1 1 1 1 1

100

APPENDIX D. Historical and current survey data for Boltonia montana. Gray shading indicates no survey data available.

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 2016: Aug - 10,000+ plants abundant 1988: July - plants ringing the length 1992: Sept - not as abundant as 1988 2007: July - pond filled with water, throughout pondshore and bottom of of the pond, some immature achenes (NJDEP 2019). few plants observed (NJDEP 2019). dry pond; 2nd year of dry pond present; population estimated at conditions (NJDEP 2019). 710,000 plants (NJDEP 2019). 1995: Sept - abundant on muddy 2007: Oct - pond dry except for small 2019: July - no plants observed, bottom of nearly dry limestone pool in the central portion; abundant water level very high, above areas Duck Pond NJ Sussex sinkhole pond in 3 distinct zones rosettes but few mature plants where plant has been observed in the (NJDEP 2019). (NJDEP 2019). past (NJDEP 2019). 1988: Aug - plants in full flower 2008: Sept - a few flowering plants, 1997: Sept - 10,000+ plants; 2019: Oct - thousands of plants (NJDEP 2019). not as abundant as previous years; abundant in upper pondshore around observed, very wet year, water levels pond drawdown was late (NJDEP periphery (NJDEP 2019). high (NJDEP 2019) 2019). 2013: Aug - 50 to 100 plants in flower 1997: subpopulation observed at Frog 1981: Sept - species observed (NJDEP 2019). Frog Pond NJ Sussex Pond, 200 to 300 plants (NJ NHP (NJDEP 2019). 2018). 2019: Oct - 8 plant, very wet year, Swartswood water levels high (NJDEP 2019). 2013: Aug - 20 to 40 plants in flower; 1997: Sept - thousands of plants overrun by Phalaris (NJDEP 2019). 1988: Aug - species observed Little Frog Pond NJ Sussex observed scattered throughout Little (NJDEP 2019). 2019: Oct - A single plant observed, Frog Pond (NJ NHP 2018). very wet year, water levels high (NJDEP 2019). 1988: Aug - thousands of plants 2007: Aug - lower numbers than in 2019: Oct - No plants observed, very Spring Lake NJ Sussex forming nearly solid stands (NJ NHP 1988, which was a drier year (NJ wet year, water levels high (NJDEP 2018). NHP 2018). 2019). 1997: Aug - thousands of plants, in flower, vigorous, dominant in sinkhole pond (NJDEP 2019). 2005: large population in disturbed Swallowhole Pond NJ Sussex 1997: Oct - hundreds of plants in half sinkhole pond habitat (NJ NHP of pond, other half plowed by 2018). landowner; plants in flower, in fruit, robust (NJ NHP 2018).

101

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 2019: Oct - No plants observed, very 1997: several plants in flower at Muckshaw Pond 1 NJ Sussex wet year, water levels high (NJDEP north end of pond (NJDEP 2019). 2019). 2019: Oct - 17 plants observed, very 1988: Sept - species observed 1997: several plant in flower at pond Muckshaw Pond 2 NJ Sussex wet year, water levels high (NJDEP (NJDEP 2019). center (NJDEP 2019). 2019). 2019: Oct - No plants observed, very 1988: Sept - species observed Muckshaw Pond 3 NJ Sussex 1997: 34 plants in flower, vigorous, at wet year, water levels high (NJDEP (NJDEP 2019). north and south ends (NJDEP 2019). 2019). 2000: plants mostly at north end and scattered in the middle of pond 2019: Oct - No plants observed, very Muckshaw 1988: Sept - species observed 1997: species observed (NJDEP Muckshaw Pond 5 NJ Sussex (NJDEP 2019). wet year, water levels high (NJDEP (NJDEP 2019). 2019). 2007: species observed (NJDEP 2019). 2019). 2000: a few plants in flower and fruit around north periphery and south 2019: Oct - No plants observed, very Muckshaw Pond 6 NJ Sussex end, vigorous (NJDEP 2019). wet year, water levels high (NJDEP 2007: species observed (NJDEP 2019). 2019). 2019: Oct - No plants observed, very 1988: Sept - species observed 1997: species observed (NJDEP 2000: weedy pond with B. montana Muckshaw Pond 7 NJ Sussex wet year, water levels high (NJDEP (NJDEP 2019). 2019). covering 30% (NJDEP 2019). 2019). 1997: Oct - thousands of plants, in 2005: Oct - 10,000+ plants, in flower 2019: Oct - No plants observed, very leaf and flower, abundant (35%) in Piggyback Pond NJ Warren and leaf, abundant in upper rocky wet year, water levels high (NJDEP upper pond shore and scattered (1- shore (NJDEP 2019). 2019). 5%) in lower pond shore (NJDEP 2019: Oct - 249 plants observed, very 1932: Sept - specimen collected from 1997: suitable habitat but failed to Catfish Pond NJ Sussex wet year, water levels high (NJDEP White Lake moist border of pond (NJDEP 2019). locate during survey (NJDEP 2019). 2019). 1920: July - specimen collected; 2019: Oct - No plants observed, very muddy calcareous shores and flats 1997: suitable habitat but failed to Shuster Ponds NJ Warren wet year, water levels high (NJDEP (Townsend and Karaman-Castro locate during survey (NJDEP 2019). 2019). 2006, p. 881).

102

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 1992: 100 - 200 plants mostly along Huntsburg Sinkhole 1985: Aug - species noted (NJDEP eastern shore (NJDEP 2019). NJ Sussex East 2019). 1997: 28 flowering plants along SE shore (NJDEP 2019).

1992: about 50 plants (NJDEP 2019). Huntsburg Sinkhole 1985: Aug - species noted (NJDEP NJ Sussex West 2019). 1997: not detected (NJDEP 2019).

1981: Oct - specimen collected from Johnsonburg Sinkhole Greendell NJ Warren muddy shore of limestone sink pond; Ponds "filled whole pond" (NJDEP 2019).

1997: Oct - 1,500-2,000 plants, 2009: Apr - vigorous population but 2019: Oct - No plants observed, very Sussex/ flowering and fruiting, vigorous, Four-Angle Pond NJ habitat moderately affected by wet year, water levels high (NJDEP Warren scattered throughout limestone housing development (NJDEP 2019). 2019). sinkhole pond (NJDEP 2019). 1997: a few plants scattered along 2019: Oct - No plants observed, very Pleistocene Cave edge of mote, in leaf and flower; NJ Warren wet year, water levels high (NJDEP Sinkhole small population in good sinkhole 2019). pond habitat (NJ NHP 2018). 1910: Sept - specimen collected from Turtle Sinkhole low swale (NJDEP 2019). Huntsville NJ Sussex Ponds 1988: Sept - species observed (NJDEP 2019). 1887: Sept - specimen collected near 2019: Oct - No plants observed, very Lake Grinnell Lake Grinnell Ponds NJ Sussex Lake Grinnell (Townsend and wet year, water levels high (NJDEP Karaman-Castro 2006, p. 881). 2019). 1864, 1865: banks of Susquehanna Susquehanna Susquehanna River PA Dauphin River (Townsend and Karaman- River Castro 2006, p. 881).

103

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 1964: adjacent to Spring Pond, 2015: water levels remain high in 2001: water levels increased b/c of bottom of dried up pond (Townsend response to 2001 beaver dam beaver dam (VADCR). and Karaman-Castro 2006, p. 881). (VADCR). Spring Pond VA Augusta 1970: around the margins and 2018: no plants observed (Van drawdown areas of Spring Pond 2002: species observed (Van Alstine Alstine 2019, p. 15). (Townsend and Karaman-Castro 2019, p. 15). 2019: no plants observed, pond was 2006, p. 881). full (USFS 2019). 2011: thousands of flowering plants (Van Alstine 2019, p. 15). 1970: common in drying portions of 2015: thousands of flowering plants pond (Townsend and Karaman- (Van Alstine 2019, p. 15). Castro 2006, p. 881). 2016: thousands of flowering plants (Van Alstine 2019, p. 15). 1991: in dried muddy bed of 2017: abundant in western pond (Van intermittent sinkhole pond Twin Ponds #11, 12 VA Augusta Alstine 2019, p. 15). (Townsend and Karaman-Castro 2006, p. 881). 2018: no plants observed, but site 1971: common in shallow water of flooded to forest edge (Van Alstine 2019, p. 15). Maple Flats open edge (Townsend and Karaman- Castro 2006, p. 881). 2019: approximately 163 flowering plants and 2,000 - 3,000 rosettes in N. pond and 674 flowering plants and 400 rosettes in S. pond (USFS 2019). 2018: no plants observed, but pond North Maple Flat 1992: 15 - 20 plants (Van Alstine 2008: two flowering clumps (Van water levels never drew down and VA Augusta Pond A 2019, p. 14). Alstine 2019, p. 14). site was flooded (Van Alstine 2019, p. 14). 2018: no plants observed, but pond North Maple Flat 2008: two flowering clumps (Van water levels never drew down and VA Augusta Pond B Alstine 2019, p. 14). site was flooded (Van Alstine 2019, p. 14). 2016: 12 sprawling plants in flower (Van Alstine 2019, p. 15). 1991: two low plants in flower (Van 2006: 11 plants (Van Alstine 2019, p. 2018: no plants observed, but little Oak Pond #13 VA Augusta Alstine 2019, p. 15). 15). drawdown (Van Alstine 2019, p. 15). 2019: no plants observed (USFS 2019).

104

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 1991: species observed (Van Alstine 2018: no plants observed, but pond 2019, p. 14). water levels never drew down and Kennedy Mountain 1937: specimen collected (Townsend 2001: trace observed (Van Alstine site was flooded (Van Alstine 2019, p. VA Augusta Meadow and Karaman-Castro 2006, p. 881). 1995: 10 - 20 plants (Van Alstine 2019, p. 14). 14). 2019, p. 14). 2019: no plants observed, pond dry (USFS 2019). 2019: a single flowering plant Deep Pond VA Augusta observed (USFS 2019) 2011: hundreds of vigorous clumps throughout pond (Van Alstine 2019, Maple Flats p. 15). (cont) 2015: several hundred depauperate, sprawling plants (Van Alstine 2019, p. 15). 2017: scattered plants, mostly Split Level Pond VA Augusta matted/sprawling and fairly small; much less abundant and vigorous than 2011; (Van Alstine 2019, p. 15).

2018: 21 to 26 flowering plants (Van Alstine 2019, p. 15). 2019: no plants observed, pond was dry (USFS 2019). 1986: species noted (Van Alstine 1990: species noted (Van Alstine 2019, p. 11). 2019, p. 11). 2008: abundant (at least a few 2018: Not surveyed, but aerial Cambells Pond VA Augusta hundred) flowering plants (Van imagery indicates habitat likely 1987: species noted (Van Alstine 1995: 500+ plants (Van Alstine 2019, Alstine 2019, p. 11) present (Van Alstine 2019, p. 11) 2019, p. 11). p. 11). 2008: many hundreds of flowering 2018: Hundreds of flowering plants, Cambells 1987: species noted (Van Alstine 1995: observed 11-50 plants (Van Grove Farm Pond VA Augusta plants and even more rosettes (Van no attempt to count rosettes (Van 2019, p. 10). Alstine 2019, p. 10). Alstine 2019, p. 10). Alstine 2019, p. 10). 2008: hundreds of flowering plants 2018: Ten flowering stems and 1 Grove Farm Pond VA Augusta and rosettes (Van Alstine 2019, p. bolted stem with no flowers (Van East 10). Alstine 2019, p. 10).

105

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 1990: 100+ flowering plants with 2006: 500+ plants observed (Van 2017: observation of flowering plants some rosettes (Van Alstine 2019, p. Alstine 2019, p. 11). (Van Alstine 2019, p. 11). 1987: species noted (Van Alstine 11). Abshire Pond VA Augusta 2019, p. 11). 2006: observed about 40 flowering 2018: About 90-100 plants observed 1995: 500+ plants observed; (Van plants, no attempt to count rosettes through binoculars (Van Alstine 2019, Alstine 2019, p. 11). (Van Alstine 2019, p. 11). p. 11). 2012: thousands of flowering plants (Van Alstine 2019, p. 12). 2015: 1,200 stems (Van Alstine 2019, I VA Augusta p. 12). 2018: thousands of flowering plants; rosettes present but difficult to assess numbers. Population has expanded (Van Alstine 2019, p. 12). 2018: no access provided to the majority of the area. In an area Route 664 Meadow 1990: hundreds of plants, most VA Augusta recently mowed no flowering plants East flowering (Van Alstine 2019, p. 13). Lyndhurst were observed, but possibly a few rosettes (Van Alstine 2019, p. 13). Route 664 Meadow 1990: a few plants observed (Van 2018: No plants observed (Van VA Augusta West Alstine 2019, p. 13). Alstine 2019, p. 13). 2018: Area recently mowed, no 1987: species noted (Van Alstine Sherlynd Church VA Augusta plants observed (Van Alstine 2019, p. 2019, p. 13). 13). 2012: Hundreds of flowering plants (Van Alstine 2019, p. 14). 1984: plants scattered in and around 2015: 4,100 to 5,100 stems (Van Wood Duck Pond VA Augusta edge of pond (Van Alstine 2019, p. Alstine 2019, p. 14). 14). 2018: Couple of hundred flowering plants (Van Alstine 2019, p. 14). 2008: not observed though area was 1995: 1 - 10 plants (Van Alstine 2019, in severe drought, some dried plants 2018: a single young flowering plant Anderson Pond VA Augusta p. 11). may have been B. montana (Van (Van Alstine 2019, p. 11). Alstine 2019, p. 11).

106

Location Historical Current State County Population Area Population Pre-1990 1990 - 1999 2000 - 2009 2010 - 2019 2008: pond A about 450 flowering 1995: pond A 11 - 500 plants (Van plants and 42 rosettes (Van Alstine Alstine 2019, p. 12). 2019, p. 12). 2018: pond B no plants or rosettes observed, but pond water levels very Fence Line Ponds (A, 2008: pond B 10 bolted plants (Van VA Augusta high and may not have drawn down B, and South) Alstine 2019, p. 12). 1997: pond A 11 - 500 plants (Van during the year; no access to ponds A Alstine 2019, p. 12). 2008: pond South 10 bolted plants and South (Van Alstine 2019, p. 12). Lyndhurst and 12 rosettes (Van Alstine 2019, p. (cont) 12). 1990: hundreds of plants (stems and rosettes) scattered around pond (Van 2006: Plants and rosettes noted (no 2018: thousands of flowering plants, Lyndhurst Pond #18- 1987: species observed (Van Alstine Alstine 2019, p. 13). VA Augusta count attempted) (Van Alstine 2019, population area slightly expanded 15 2019, p. 13). 1995: 500+ plants; 2006: Plants and p. 13). (Van Alstine 2019, p. 13). rosettes noted (no count attempted) (Van Alstine 2019, p. 13). 1936: bank of South River vicinity of Lipscomb, abundant in Stuart's Draft South River South River VA Augusta area around ponds and along South River (Townsend and Karaman- Castro 2006, p. 881). 1990: not observed (Van Alstine 2018: no plants observed, wetter 1987: species mentioned (Van Alstine 2019, p. 16). 2008: not observed (Van Alstine than usual and no access to much of Hattons Peterson Pond VA Augusta 2019, p. 16). 1995: not observed (Van Alstine 2019, p. 16). the property (Van Alstine 2019, p. 15). 2019, p. 16).

107