United States Department of Agriculture Terrestrial Ecosystems Assessment Supplemental Report Wayne National Forest

Forest Wayne National Forest Plan Service Forest Revision July 2020 Prepared By: Greg Nowacki (Forest Service - Eastern Region) Lisa Kluesner (Forest Service - Wayne National Forest) Randy Swaty (The Nature Conservancy) Jason Stevens (Forest Service - Eastern Region) Jarel Bartig (Forest Service - Wayne National Forest) Melissa Thomas-Van Gundy (Forest Service - Northern Research Station) Patricia Leopold (Forest Service - Northern Institute of Applied Climate Science) Cornelia Pinchot (Forest Service - Northern Research Station) Sierra Dawkins (Forest Service - Eastern Region)

Responsible Official: Forest Supervisor Carrie Gilbert

Cover Photo: A scenic vista in the Ironton Unit of the Wayne National Forest. USDA photo by Kyle Brooks

The use of trade or firm names in this publication is for reader information and does not imply endorsement by the U.S. Department of Agriculture of any product or service.

In accordance with Federal civil rights law and U.S. Department of Agriculture (USDA) civil rights regulations and policies, the USDA, its Agencies, offices, and employees, and institutions participating in or administering USDA programs are prohibited from discriminating based on race, color, national origin, religion, sex, gender identity (including gender expression), sexual orientation, disability, age, marital status, family/parental status, income derived from a public assistance program, political beliefs, or reprisal or retaliation for prior civil rights activity, in any program or activity conducted or funded by USDA (not all bases apply to all programs). Remedies and complaint filing deadlines vary by program or incident.

Persons with disabilities who require alternative means of communication for program information (e.g., Braille, large print, audiotape, American Sign Language, etc.) should contact the responsible Agency or USDA’s TARGET Center at (202) 720-2600 (voice and TTY) or contact USDA through the Federal Relay Service at (800) 877-8339. Additionally, program information may be made available in languages other than English.

To file a program discrimination complaint, complete the USDA Program Discrimination Complaint Form, AD-3027, found online at http://www.ascr.usda.gov/complaint_filing_cust.html and at any USDA office or write a letter addressed to USDA and provide in the letter all of the information requested in the form. To request a copy of the complaint form, call (866) 632-9992. Submit your completed form or letter to USDA by: (1) mail: U.S. Department of Agriculture, Office of the Assistant Secretary for Civil Rights, 1400 Independence Avenue, SW, Washington, DC. 20250-9410; (2) fax: (202) 690-7442; or (3) email: [email protected].

USDA is an equal opportunity provider, employer, and lender. Terrestrial Ecosystems Supplemental Report

Table of Contents Notable Changes between Draft & Final of this Supplemental Report ...... 1 Introduction ...... 1 Use of Ecological Units ...... 4 Use of the Central Appalachians Forest Ecosystems ...... 5 Use of the United States National Vegetation Classification ...... 5 Key Ecological Characteristics ...... 6 Southern Unglaciated Allegheny Plateau Section (221E) ...... 7 Geologic Origins and Ecological Conditions ...... 7 Fire History of Appalachian Ohio ...... 13 From Climatic Climax Theory to Disturbance Ecology ...... 15 Ecological Subsections of Southeastern Ohio ...... 16 Pre-Euro-American Settlement Vegetation ...... 17 Witness Trees as Pyroindicators ...... 21 Ecological Landtype Classes ...... 25 Dry Oak Forest Landtype Class ...... 26 Vegetation Composition and Structure ...... 27 Drivers ...... 28 Current Conditions, Stressors, and Connectivity ...... 28 Dry-Mesic Mixed Oak Hardwood Forest Landtype Class ...... 33 Vegetation Composition and Structure ...... 34 Drivers ...... 35 Current Conditions, Stressors, and Connectivity ...... 35 Rolling Bottomlands Mixed Hardwood Forest Landtype Class ...... 40 Vegetation Composition and Structure ...... 41 Drivers ...... 42 Current Conditions, Stressors, and Connectivity ...... 42 Rare & Novel Habitats ...... 48 Western Allegheny Plateau Oak Barrens ...... 49 Appalachian Low-Elevation Mixed Pine/Blue Ridge Blueberry Forest...... 49 Appalachian Chestnut Oak-Mixed Oak Forest ...... 49 Appalachian-Alleghenian Sandstone Dry Cliff Vegetation ...... 50 Northern Mixed Mesophytic Forest...... 50 Interior Low Plateau Beech-Maple Forest...... 51 Interior Plateaus Hemlock-Hardwood Forest ...... 51 River Birch-Sycamore Small River Floodplain Forest ...... 51 Skunk-cabbage Seepage Meadow ...... 51 Pin Oak Mixed Hardwood Depression Forest ...... 51 Midwest Buttonbush Shrub Swamp ...... 52 Reclaimed Grasslands ...... 52 Climate Change ...... 52

i | Page

Terrestrial Ecosystems Supplemental Report

Summary ...... 56 Glossary ...... 57 References ...... 58 List of Figures Figure 1. The Wayne National Forest in Ohio (one of the 15 National Forest System units in the Eastern Region) and 17-county study area (circled in red above) ...... 2 Figure 2. The 17-county study area (red outline) and Wayne National Forest proclamation boundary (yellow outline) superimposed on Ecological Subsections within the Southern Unglaciated Allegheny Plateau Section (black outline) ...... 3 Figure 3. Areal extent of 17-county study area with government ownership patterns mapped ...... 4 Figure 4. Ecological subsections overlaid on Ohio bedrock geology ...... 9 Figure 5. Glacial map of Ohio showing the influence of the glaciation ...... 10 Figure 6. Changes in major tree species (forest types) and vegetation drivers for southeastern Ohio climatic periods are designated by background colors: Early Holocene Warming (peach), Holocene Thermal Maximum (red), Neoglacial Cooling (blue), and Anthropocene (yellow) ...... 11 Figure 7. Area from which witness tree data were obtained from two Ohio land surveys conducted in the late 18th century, relative to ecological subsection boundaries of Section 221E ...... 21 Figure 8. Pyrophilic percentage map for units of the Wayne National Forest ...... 23 Figure 9. Subsections partitioned by pyrophilic percentage class. OVL = Ohio Valley Lowland (221Ec), East Hocking Plateau (221Ed), WHP = Western Hocking Plateau (221Ef) ...... 24 Figure 10. Landtype classes partitioned by pyrophilic percentage class ...... 24 Figure 11. Connectedness map of the 17-county study area and embedded Wayne National Forest ...... 26 Figure 12. Distribution of the Dry Oak landtype class across the 17-county study area ...... 27 Figure 13. State-and-transition model showing the natural range of variability for the Dry Oak Forest landtype class...... 28 Figure 14. Current land use acres (percentages) for the Dry Oak Forest landtype class within the 17-county study area ...... 30 Figure 15. Current land use acres (percentages) for the Dry Oak Forest landtype class within the Wayne National Forest proclamation boundary ...... 30 Figure 16. Vegetation class acres (percentages) for the Dry Oak Forest landtype class within the 17-county study area ...... 31 Figure 17. Vegetation class acres (percentages) for the Dry Oak Forest landtype class within the Wayne National Forest proclamation boundary ...... 31 Figure 18. Historic annual acres (percentages) burned for the Dry Oak Forest landtype class within the 17-county study area ...... 32 Figure 19. Historic annual acres (percentages) burned for the Dry Oak Forest landtype class within the Wayne National Forest proclamation boundary ...... 32 Figure 20. Connectedness class distribution for the Dry Oak Forest landtype class for the 17-county study area ...... 33 Figure 21. Distribution of the Dry-mesic Mixed Oak Hardwood Forest landtype class across the 17-county study area ...... 34 Figure 22. State-and-transition model showing the natural range of variability for Dry-Mesic Mixed Oak Hardwood Forest Landtype class ...... 35 Figure 23. Current land use acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the 17-county study area ...... 37 Figure 24. Current land use acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the Wayne National Forest proclamation boundary ...... 37 Figure 25. Succession class acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the 17-county study area ...... 38 Figure 26. Succession class acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the Wayne National Forest proclamation boundary ...... 38 Figure 27. Historic annual acres (percentages) burned for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the 17-county study area ...... 39

ii | Page

Terrestrial Ecosystems Supplemental Report

Figure 28. Historic annual acres (percentages) burned for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the Wayne National Forest proclamation boundary ...... 39 Figure 29. Connectedness class distribution for the Dry-Mesic Mixed Oak Hardwood Forest landtype class for the 17-county study area ...... 40 Figure 30. Distribution of the Rolling Bottomlands Mixed Hardwood Forest landtype class across the 17-county study area ...... 41 Figure 31. State-and-transition model showing the natural range of variability for the Rolling Bottomlands Mixed Hardwood Forest landtype class ...... 42 Figure 32. Current land use acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the 17-county study area ...... 44 Figure 33. Current land use acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the Wayne National Forest proclamation boundary ...... 45 Figure 34. Succession class acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the 17-county study area ...... 45 Figure 35. Succession class acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the Wayne National Forest proclamation boundary ...... 46 Figure 36. Historic annual acres (percentages) burned for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the 17-county study area ...... 46 Figure 37. Historic annual acres (percentages) burned for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the Wayne National Forest proclamation boundary ...... 47 Figure 38. Connectedness class distribution for the Rolling Bottomlands Mixed Hardwood Forest landtype class for the 17-county study area ...... 47 Figure 39. Ecological Landtype Classes, Forest Ecosystems, and Rare and Novel Habitats in the Wayne National Forest ...... 48 List of Tables Table 1. Map scale and polygon size of ecological units (adapted from Cleland et al. 1997) ...... 5 Table 2. Historical and major disturbance events in Ohio since the 16th century ...... 12 Table 3. Abundance of witness tree taxa by ecological subsection of Section 221E ...... 18 Table 4. Pre-Euro-American settlement vegetation based on witness-tree percentages in each ecological landtype class within the Ohio portion of Ecological Subsection 221Ec – Ohio Valley Lowland...... 19 Table 5. Pre-Euro-American settlement vegetation based on witness-tree percentages in each ecological landtype class within the Ohio portion of Ecological Subsection 221Ed – East Hocking Plateau ...... 19 Table 6. Pre-Euro-American settlement vegetation based on witness-tree percentages in each ecological landtype class within the Ohio portion of Ecological Subsection 221Ef – Western Hocking Plateau ...... 20

iii | Page

Terrestrial Ecosystems Supplemental Report

Notable Changes between Draft & Final of this Supplemental Report Ecosystem diversity across different spatial scales has been detailed. A middle scale tier was incorporated using the forest ecosystems as described in Central Appalachians Forest Ecosystem Vulnerability Assessment and Synthesis (Butler et al. 2015). Several additional rare habitats were identified and described to better encompass significant plant communities identified in the Wayne National Forest. Specific key ecosystem characteristics have been included for each of the three landtype classes. State-and-transition models have been included for each landtype class to provide a more thorough explanation of the natural range of variability under historic disturbance regimes. The modeling process behind LANDFIRE’s Biophysical Settings development has been explained. The Connectedness section has been split and incorporated into the respective key ecosystem characteristics. Introduction The Wayne National Forest is in the midst of revising the 2006 Wayne National Forest Land and Resource Management Plan (2006 Wayne Forest Plan) under the guidance of the 2012 National Forest System Land Management Planning Rule. The 2012 planning rule’s purpose is to guide the collaborative and science-based development, amendment, and revision of forest plans that promote the ecological integrity of National Forest System lands. The forest plan describes the strategic direction for the management of natural resources, while remaining adaptive and amendable as conditions change over time in order to remain relevant for their intended application period of 10 to 15 years.

The Terrestrial Ecosystems Supplement Report provides a characterization of ecosystems at multiple spatial scales across the 17-county study area, as well as identification of rare or unique high-value ecosystems. This supplemental report includes a description of key ecosystem characteristics represented across these ecosystems, their current status, and stressors that impact terrestrial ecosystem health. A baseline qualitative evaluation of ecological integrity is determined using the status of key ecological characteristics and perceived climate change resilience. This supplemental report—one of ten supplemental reports—summarizes the past and present ecological conditions (and trends therein) for a 17-county study area of southeastern Ohio (figure 1). This area was selected for the forest plan revision process for several reasons:

• Corresponds with the 2015-2017 Joint Chiefs’ Landscape Restoration Partnership, where the Forest Service and Natural Resources Conservation Service partnered with the Ohio Department of Natural Resources (ODNR) Division of Forestry and Ohio State University Extension to collaboratively manage oak ecosystems in southeastern Ohio. The partnership has since expanded to include ODNR Division of Wildlife and Central State University Extension and is formally working together under an interagency business model.

1 | Page

Terrestrial Ecosystems Supplemental Report

• Incorporates the ecological boundaries defined in the USDA Forest Service Section, Subsection and Landtype Descriptions for Southeastern Ohio General Technical Report (see figure 2; Iverson et al. 2019a). • Encompasses a larger area centered on the Wayne National Forest to account for broader ecological conditions and processes and cumulative impacts of threats and stressors, thus facilitating integrated restoration efforts across multiple land ownerships (see figure 3).

Figure 1. The Wayne National Forest in Ohio (one of the 15 National Forest System units in the Eastern Region) and 17-county study area (circled in red above) Note: Green outlines depict the proclamation boundaries for National Forest System units.

2 | Page

Terrestrial Ecosystems Supplemental Report

Figure 2. The 17-county study area (red outline) and Wayne National Forest proclamation boundary (yellow outline) superimposed on Ecological Subsections within the Southern Unglaciated Allegheny Plateau Section (black outline) Note: As depicted in USDA Forest Service Section, Subsection and Landtype Descriptions for Southeastern Ohio (Iverson et al. 2019a).

3 | Page

Terrestrial Ecosystems Supplemental Report

Figure 3. Areal extent of 17-county study area with government ownership patterns mapped

Use of Ecological Units Ecological units derived from the Forest Service’s National Hierarchical Framework of Ecological Units were employed in this assessment (Cleland et al. 1997). It is a nested, eight-tier hierarchical system that allows for the classification, mapping, and description of ecosystems from global (millions of square miles) to site level (<100 acres) (table 1). The primary purpose of ecological units is to delineate land areas at different levels of resolution that have similar capabilities and potentials for management and conservation. Depending on scale, ecological units are designed to exhibit similar patterns in vegetation, soils, hydrology, landform, climate, and important underlying processes such as nutrient and water cycling, succession, and natural and human disturbances. It allows for understanding ecological patterns and processes operating at different spatial scales and relationships among scales (ecological context from larger “umbrella” units; ecological specificity from smaller embedded units). The ecological basis and spatial flexibility of this multi-tiered classification and mapping system covers all planning needs by supplying scientifically defensible base layers from which forest plans can be created and subsequently implemented at the project level. This report follows the USDA Forest Service Section, Subsection, and Landtype Descriptions for Southeastern Ohio (Iverson et al. 2019) to describe and analyze the terrestrial ecosystems described hereafter.

4 | Page

Terrestrial Ecosystems Supplemental Report

Table 1. Map scale and polygon size of ecological units (adapted from Cleland et al. 1997) Ecological unit Map scale range General polygon size Domain 1:30,000,000 or smaller 1,000,000s of square miles Division 1:30,000,000 to 1:7,500,000 100,000s of square miles Province 1:15,000,000 to 1:5,000,000 10,000s of square miles Section 1:7,500,000 to 1:3,500,000 1,000s of square miles Subsection 1:3,500,000 to 1:250,000 10s to low 1,000s of square miles Landtype association 1:250,000 to 1:60,000 1,000s to 10,000s of acres Landtype 1:60,000 to 1:24,000 100s to 1,000s of acres Landtype phase 1:24,000 or larger <100 acres Note: Shaded rows denotes ecological units used in this supplemental report Use of the Central Appalachians Forest Ecosystems Forest ecosystems were described for a climate change vulnerability assessment of the Central Appalachians. An ecosystem is defined as a spatially explicit, relatively homogenous unit of the Earth that includes all interacting organisms and components of the abiotic environment within its boundaries (Society of American Foresters 2014). For the purposes of the assessment, “forest ecosystem” refers to a specific classification system defined by its species assemblage, its spatial distribution, and other distinct characteristics (Butler et al. 2015). Forest ecosystems occurring within the 17-county study area have been cross-walked to and nested under the landtype classes to provide an additional description of vegetative structure, composition, function, and stressors. Use of the United States National Vegetation Classification1 The United States National Vegetation Classification provides a common language for the effective management and conservation of plant communities in the United States. It is based on the Vegetation Classification Standard that enables Federal agencies to produce uniform statistics about vegetation resources across the Nation, facilitates interagency cooperation on vegetation management issues that transcend jurisdictional boundaries, and encourages non-Federal partners to utilize and contribute to a common system when working with their Federal partners. The classification system is hierarchical with floral assemblages called associations at the finest level. Rare and unique ecosystems found in southeast Ohio and within the proclamation boundaries of the Wayne National Forest are embedded within the ecological units (landtype classes) and below the forest ecosystems used in this assessment. These rare ecosystems are represented by special interest areas and research natural areas in the Wayne National Forest. The corresponding United States National Vegetation Classification community name that represents each of these habitat types is included with a brief description in the Rare and Novel Habitats section.

1 FGDC 2019

5 | Page

Terrestrial Ecosystems Supplemental Report

Key Ecological Characteristics Key ecological characteristics are measurable traits that characterize the health of an ecosystem. For this assessment, the following traits have been identified and discussed for each landtype class:

• Vegetation structure and composition pertains to how the vegetation is generally “shaped” and which species are most prevalent. For example, under historic disturbance regimes, mixed mesophytic hardwood forests are normally multi-canopied, dense forests composed of late-successional, shade-tolerant species driven by wind disturbance (Aubin et al. 2007; Fahey et al. 2015), whereas oak ecosystems have an array of conditions ranging from single-canopied forests to open savannas, lower tree densities, and composed of fire- tolerant species driven by recurrent fire (Lafon et al. 2017). Importantly, any ecosystem will have vegetation developmental stages resulting from natural disturbances that develop their own compositional and structural characteristics known as the natural range of variability. The expected compositional and structural characteristics and successional states under historic disturbance regimes are referred to as “reference conditions” (Rollins 2009). • Drivers are how an ecosystem would naturally work within its historic disturbance regime. Common disturbance agents in southeast Ohio include fire, flooding, insect and disease, drought, and windthrow (Mori 2011). This is linked to the aforementioned key ecological characteristic, as these processes create the patterns of structure and composition. • Stressors and their impacts are relative to, or contingent on, historic disturbance regimes. For instance, fire is a common and revitalizing force in pryogenic (fire-based) ecosystems, but is considered destabilizing in mixed mesophytic forests not prone to burn. This has management implications when applying prescribed burning as a silvicultural treatment for ecosystem restoration or forest regeneration. Likewise, responses to drought will differ among ecosystems, having a greater impact on mesophytic forests with drought-sensitive species than xerophytic systems (Elliott et al. 2015). This has implications in regards to climate change. Invasive diseases and pests—such as Dutch elm disease, chestnut blight, and emerald ash borer—indiscriminately kill host trees and therefore are definite stressors in many ecosystems, especially those harboring host trees as a primary component. Non- native invasive plant species can also disproportionately affect successional processes within ecosystems, as well as exacerbate the effects of other stressors when cumulatively evaluated. Climate and natural disturbances are still important drivers of forest succession; however, under current conditions they operate under different temporal scales, while continuing to interact with one another as well as with these new stressors. • Connectivity pertains to how ecosystems are spatially configured and how fragmented they are currently relative to the historic landscape. Connectivity was mapped by The Nature Conservancy’s Eastern Science Team (Anderson et al. 2014). The metric captures the degree to which regional landscapes, encompassing a variety of natural, semi-natural, and developed land cover types, will sustain ecological processes and are conducive to the movement of many types of organisms (McRae et al. 2012).

6 | Page

Terrestrial Ecosystems Supplemental Report

Southern Unglaciated Allegheny Plateau Section (221E)2 Geologic Origins and Ecological Conditions The study area falls within the west-central portion of the Southern Unglaciated Allegheny Plateau Section (figure 2). The section lies between the maximum ice floes of Wisconsinan and Illinoian glaciations to the northwest and the Appalachian Mountains to the southeast. This highly dissected plateau marks a transition from flat till plains to the northwest and the deformed rocks of the Appalachians (Camp 2006). Because it is unglaciated, much of its ecological character is directly attributed to the underlying geology and erosional forces.

The geologic layers that underpin this Section are the result of mountain-building processes that span a half billion years (Camp 2006; Van Driver 1990). Three distinct mountain-building orogenies directly affected this section, each event contributing sediments that solidified over time to form present-day bedrock layers. The first event took place some 475 million years ago when the Baltic (European) and Laurentia (North American) plates collided, forming the Taconic Highlands. During the Ordovician Period, shallow seas dominated much of Laurentia, giving rise to marine-based shales and limestones. Long after the Taconic Highlands eroded away, another Baltic-Laurentia plate collision took place about 375 million years ago, creating the Acadian Mountains. Westward-forming streams actively eroded this mountain range, forming the shales and sandstones that mark the Devonian and Mississippian Periods. The last event occurred some 318 million years ago during the Pennsylvanian Period, when northwestern Africa collided with North America and uplifted the Appalachians (Camp 2006). Again, the sediments were shed westward towards Ohio, forming a complex mix of depositional environments that eventually became the Pennsylvanian- and Permian-aged rocks of southeastern Ohio (Camp 2006).

Starting at the late Ordovician time, around 450 million years ago, rock layers in southeastern Ohio had begun to sink (as part of the western edge of the Appalachian Basin) and tilt eastward from the stable, westward-lying Cincinnati Arch (Camp 2006). Erosion was greatest along the crest of the arch, so the oldest rock (Ordovician) occurs on top of it, with parallel bands of increasingly younger rock radiating outward along its flanks. Thus, the rock layers that span the Allegheny Plateau are aligned from oldest (Ordovician) to youngest (Permian) from west to east, their long axes lying in a north-south orientation (Camp 2006) (figure 4).

Even though this portion of the Allegheny Plateau was not overridden by glacial ice, it was influenced by it in other ways, as shown in the Ohio glacial history map (figure 5). During the height of past glaciations, periglacial processes (geomorphic changes resulting from seasonal freeze-thaw cycles) dominated the frozen tundra that lined the glacier front. These processes, such as ice wedging and frost creep, resulted in the extensive mixing of surficial deposits. Meltwater rivers and streams gushed southward from vast ice sheets, depositing outwash sands and gravels along their course. These deposits are called valley trains, with the Scioto River

2 Text extracted and slightly modified from Iverson et al. (2019a).

7 | Page

Terrestrial Ecosystems Supplemental Report

serving as a primary example in southeastern Ohio (Camp 2006). Where lakes were impounded along certain segments of rivers, fine-textured, glacio-lacustrine sediments (silts and clays) were deposited. Lastly, fine windblown sediments called loess were deposited on surfaces immediately adjacent to glacier-fed rivers. The preponderance of unglaciated soils are old and highly weathered, fine-textured, and residual, being derived from bedrock decay through chemical and physical processes.

The Wisconsinan Ice Sheet reached its maximum extent roughly 20,000 to 25,000 years ago, retracting at an irregular pace thereafter. Plants migrated northward independent of one another based on differences in (1) refugia location, (2) starting times, and (3) rates of dispersal (Pielou 1991). As such, many plant communities that formed were temporary and novel (no modern analog). Plants constantly reassembled along their northward advance as climates continually fluctuated. According to Shane and Anderson (1993), sometime after 16,000 years before the present (BP), tundra on the Allegheny Plateau was replaced by spruce (Picea) woodlands and forests, which dominated until 13,000 years BP. Tamarack (Larix), fir (Abies), and pine (Pinus) occurred at low frequencies at that time. Spruce declined from 13,000 years to 11,000 years BP, being replaced by deciduous trees, principally oak (Quercus), ash (Fraxinus), and hophornbeam/musclewood (Ostrya/Carpinus). The release from megafaunal herbivory and concurrent increase of fire have been implicated in this pronounced increase in deciduous vegetation (Gill et al. 2012; Gill 2013). A sharp expansion of pine occurred at 11,000 years BP, corresponding to the short-lived Younger Dryas cooling event (Shuman et al. 2002), and ended when oak increased markedly at 9,500 years BP. Since then, oak has dominated the arboreal vegetation with lesser amounts of hickory, according to a local pollen record at Stages Pond in south-central Ohio (Shane et al. 2001). Palynological analyses at Patton Bog, within the proclamation boundary of the Wayne National Forest, revealed a high presence of prairie plants and charcoal over the last 3,000 years, which were linked to and best explained by American Indian activities (Abrams et al. 2014).

8 | Page

Terrestrial Ecosystems Supplemental Report

Figure 4. Ecological subsections overlaid on Ohio bedrock geology

9 | Page

Terrestrial Ecosystems Supplemental Report

Figure 5. Glacial map of Ohio showing the influence of the glaciation Note that southeastern Ohio is primarily unglaciated, hence soils are old and largely a product of bedrock weathering.

The antiquity of oak dominance over the Southern Unglaciated Allegheny Plateau is rather remarkable considering the climatic fluxes incurred over the past 10,000 years (figure 6) (Abrams 2002; Lorimer and White 2003). Oak is generally a warm-adapted genus (Nowacki and Abrams 2015), so its rise to prominence during the Early Holocene Warming and continued dominance throughout the Holocene Thermal Maximum (warmer, drier conditions) makes sense.

However, its persistence during Neoglacial Cooling (3,300 years BP onward) is anomalous and begs for an additional explanatory factor (Abrams and Nowacki 2015). That factor is most likely fire, as oak and associated plant species are highly adapted to fire (Abrams 1992 and 1996; Brose and Van Lear 2004; Hart and Buchanan 2012; Lorimer and White 2003; Varner et al. 2016) (figure 6). Most pre-European fires occurred during the dormant “non-lightning” season (at least in the southern Appalachians, where the most data are available), have limited relations to drought, and vary in frequency over cultural time periods. For these reasons, ignitions were most likely human-caused, generally increasing over time with human population growth (Abrams et al. 2014; Abrams and Nowacki 2008; Brose et al. 2013; Guyette et al. 2003, 2006; Nowacki and

10 | Page

Terrestrial Ecosystems Supplemental Report

Abrams 2008; Stambaugh et al. 2018) (figure 6). Upon Euro-American settlement (ca. 1800), oaks continued to be the most dominant tree species across the Southern Unglaciated Allegheny Plateau (Dyer 2001; Gordon 1969; Sears 1925). Major, multiyear droughts in the 16th and 17th centuries may have reinforced oak’s dominance in the region, as recorded in sediment cores (Bird et al. 2017) and the Public Land Survey record (Dyer and Hutchinson 2019).

Figure 6. Changes in major tree species (forest types) and vegetation drivers for southeastern Ohio climatic periods are designated by background colors: Early Holocene Warming (peach), Holocene Thermal Maximum (red), Neoglacial Cooling (blue), and Anthropocene (yellow) Note: Forest type changes based on Shane et al. 2001. Vegetation drivers for southeastern Ohio based on Abrams and Nowacki (2015) and Patton and Curran (2016). Timeline represents years before present.

Fire regimes intensified with Euro-American settlement of the Allegheny Plateau, causing an east-to-west “wave of fire” to roll over the landscape (Brose et al. 2013; Stambaugh et al. 2018). Most of today’s forests in this region regenerated either during this period of intense cutting and burning dating to the late 1800s and early 1900s, or thereafter, as farms and mines were abandoned (Brose et al. 2013; Gordon 1969; Hutchinson et al. 2008; McEwan et al. 2007). Oaks and light-seeded tree species (e.g., maples, tuliptree, black cherry, bigtooth aspen, ash) flourished under this disturbance regime, with oaks maintaining their dominance on cutover woodlots through vigorous stump sprouting and light-seeded species recolonizing abandoned fields and pastures (Dyer and Hutchinson 2019). However, with 20th-century fire suppression and natural forest succession, understories now principally consist of fire-sensitive, shade-tolerant trees due, in part, to mesophication (Hutchinson et al. 2008; Nowacki and Abrams 2008; Palus et al. 2018). The degree of mesophication is contingent on site conditions. Invasion by mesophytic trees is most aggressive and enduring on mesic, lower slope positions, and is more subdued on xeric, south-facing upper slopes and ridges (Iverson et al. 2017, 2018).

Southeastern Ohio was almost entirely wooded before Euro-American settlement (ca. 1800), with oaks and hickories dominating (Gordon 1969). White oak (Quercus alba L.) was particularly abundant, accounting for more than 30 percent of survey witness trees (Dyer 2001).

11 | Page

Terrestrial Ecosystems Supplemental Report

After Euro-American settlement, most forests were either cleared for agriculture or cut over for industrial uses (e.g., charcoal iron production, railroads) or mining (table 2). Surface and underground mining for coal has affected more than 568,000 acres in Ohio (Lorenz and Lal 2007), and much of that land remains unreclaimed (Mishra et al. 2012). Other mined land was excessively compacted during reclamation, preventing adequate tree establishment. The rate of forest loss was especially rapid from 1850 to 1880, when most of the counties in the study area went from being more than 60 percent forested to less than 35 percent forested (Leue 1886; Trautman 1977). With agricultural abandonment and the decline of industrial exploitation of forests in the first several decades of the 1900s, forest cover has returned to ca. 1850 levels in most counties, albeit compositionally and structurally different across much of the study area. Unlike some other areas in the central Appalachians, such as Pennsylvania (Stout et al. 2013), deer browsing pressure in southeastern Ohio is relatively low and does not appear to have a major impact on tree regeneration outcomes (Apsley and McCarthy 2004).

The vegetation today generally reflects land use, with more level and productive surfaces in agriculture and more rugged and less productive lands as forests. Forest overstories are still largely oak dominated, as they were before Euro-American settlement. Oak dominance as it currently stands is primarily the result of major Euro-American land disturbances (cutting and fire), along with other interacting factors (Drury and Runkle 2006; McEwan et al. 2011). Because landscape-level burning is no longer a factor, and moisture levels have been generally increasing in recent decades (Hayhoe et al. 2018), forests are rapidly undergoing mesophication and shade-tolerant, fire-sensitive trees are greatly increasing (Hutchinson et al. 2008; Iverson et al. 2017; Nowacki and Abrams 2008; Palus et al. 2018) at the expense of oaks and other fire- adapted species. Tree diversity remains quite high though, with 90 different tree species currently recorded by the Forest Service’s Forest Inventory and Analysis within the study area.

Table 2. Historical and major disturbance events in Ohio since the 16th century Year Event 1550–1640: Indigenous peoples depopulated due to pandemic. 1788: Euro-American settlement begins; Ohio 95 percent wooded. 1795: Treaty of Greenville accelerates Euro-American settlement. 1803: Ohio becomes a state. 1820s: Coal mining begins in Ohio. 1830–1900: Charcoal produced within Hanging Rock Iron Region. 1886: Ohio's coal production peaks. 1900s: Oil and natural gas begin to replace coal in Ohio. 1916: Ohio State Forest System established. 1923: Active fire suppression begins (Fire Control District established in southern Ohio). 1930s: Chestnut blight enters Ohio (see Figure 2 in Anagnostakis [1987]). 1933–1942: Millions of trees planted in Ohio by the Civilian Conservation Corps. 1934: Establishment of the Wayne National Forest 1938: Annual acreage burned by wildfires greatly reduced. Present day: Non-native invasive species make up 25 percent of Ohio's plants (3,000 species documented); wildfires no longer a consequential disturbance.

12 | Page

Terrestrial Ecosystems Supplemental Report

Fire History of Appalachian Ohio To what extent did American Indian burning influence the development of the oak-dominated landscape that was present in Appalachian Ohio ca. 1800? Written observations by early traders and explorers in southern Ohio describe open habitats and landscape burning by the Shawnee and other historic Tribes in some areas. Joseph Barker, noting conditions near Marietta, Ohio, in 1790, observed: “The Indians, by burning the Woods every Year, kept down the undergrowth and made good pasture for the deer and good hunting for himself” (Barker 1958). However, quantifying the extent and frequency of pre-Euro-American settlement fire is difficult as few direct sources of scientific evidence (e.g., fire-scarred trees that predate 1800) have been found and examined (Hutchinson et al. 2019).

Archaeological evidence unearthed from Archaic sites in Ohio suggest that fire was being applied by aboriginal cultures as early as 8,000 years BP to propagate nut trees and reduce wildfire danger, among other purposes (Lepper 2005). Nut crops, especially acorns, became an important subsistence crop by 6,000 years ago. Climate change and fire influenced subsistence strategies and affected cultural change. Toward the end of the Archaic Period, the Hopewell peoples, who lived 2,000 years ago along the lower Scioto River, Paint Creek, Muskingum River, Licking River, Kanawha River, and Ohio River, emerged as one of North America’s most advanced aboriginal traditions. A suite of domesticated annual plant species common throughout Hopewell archaeological sites is now referred to as the “Eastern Agricultural Complex.” This transformation of local native plant species into domesticated food crops represents one of only four such achievements worldwide (Smith 1989).

Seeman and Dancy (2000), using palynological analysis from occupied Hopewell sites along the lower Scioto River and Kanawha River, conclude that these places have been in a state of constant disturbance for the last 2,200 years. The introduction of maize (Zea mays L.) agriculture marked the beginning of the Late Prehistoric Period 1,100 years ago and increased the scale of agriculture and the use of fire as a management tool. The addition of maize as a supplement to the aboriginal diet also increased the human population throughout the period until its peak in 1550. An assessment of 18 sites across the southern unglaciated Allegheny Plateau from the Prehistoric Period provides archaeological evidence of wild plant and animal remains that indicate more than 90 percent of plants were fire dependent (D. Minney; The Nature Conservancy, retired; pers. comm.; Nov. 9, 2018). For oak-dominated ecosystems, an overall fire-return interval of 5 to 15 years promotes nearly every aspect of subsistence resources: hard mast, soft mast, and game animal (D. Minney; pers. comm.; Nov. 9, 2018). The fire-maintained plant communities that directly supported aborigines also promoted game abundance, which further supported human subsistence (Abrams and Nowacki 2008). Palynological evidence from Patton Bog (Athens County) showed increased charcoal deposition rates throughout the Late Prehistoric Period, indicative of increasing fire frequency and extent burned (Abrams et al. 2014). The peoples of the Late Prehistoric Period were the last intact aboriginal culture before European contact (table 2). They exerted an important effect on vegetation patterns in the study area until depopulation occurred from ca. 1550 to 1640.

There are more data showing that fire occurred regularly in southern Ohio after Euro-American settlement, particularly from the late 1800s to ca. 1930, as forests were regenerating in many

13 | Page

Terrestrial Ecosystems Supplemental Report

areas after having been cleared for agriculture or cut over for industrial purposes (e.g., railroads, charcoal for iron production) (Hutchinson et al. 2008; McEwan et al. 2007). In the First Annual Report of the Ohio Forestry Bureau for the year 1885 (Leue 1886), the Reverend J.G. Hall, describing conditions in Nile Township (Scioto County), wrote: “Forest fires are the greatest enemies our forests have. They seem to come with fully as much regularity as drouth [sic], and perhaps nearly as regularly as the seasons themselves. Forest fires also leave or make good pasturage, and it is suspected that cattle owners ‘fire the woods’ to secure the better pasture.” European settlers may have just adopted and continued American Indian ways of burning. Hall also stated that in 1885, “forest-fires covered more than half the township this spring.”

These and other written accounts suggest that the occurrence of fire in southern Ohio forests was common, at least in some areas, both before and after Euro-American settlement. However, to gain a better understanding of historical fire frequency, severity, and seasonality, quantitative fire history data are required. In our region, this information is best provided by dendrochronology studies of fire-scarred trees. Across the Appalachian Region, there have been a number of such studies in the last 15 years, which are reviewed in Lafon et al. (2017). Fires in this collection of studies were historically frequent prior to the fire suppression era, which started in the 1930s; mean fire return intervals were usually between 3 and 10 years. This was also the case for sites with deeper chronologies that included fire scars formed in the pre-Euro-American settlement era (ca. 1600s and 1700s). These Appalachian fire history studies, though not from our 17-county area, also show that the great majority of fires occurred in the dormant season (between annual rings, roughly October to April). These fires were largely human-caused as lightning is uncommon during the dormant season. Further, because many of the fire-scarred trees in these studies survived multiple fires, even in the xeric pine-dominated stands in the southern Appalachians, the authors conclude that fires were mostly low to moderate severity (Lafon et al. 2017).

A recent study in central Pennsylvania provides further evidence of long-term (350+ year) fire regimes in the Appalachians (Stambaugh et al. 2018). The authors dated and analyzed fire scars formed in living and dead pines at 12 sites. Similar to the findings of Lafon et al. (2017), low to moderate intensity dormant season fires occurred frequently until the fire suppression era. Although fires occurred regularly at most sites before Euro-American settlement, with mean fire return intervals ranging from 6 to 18 years, the authors also described an increased frequency of fire, a “wave of fire” at all sites that coincided with this settlement. During this era, mean fire return intervals were reduced to 3 to 7 years, until fire suppression policies went into effect in the early 20th century. There is evidence that this Euro-settlement-based “wave of fire” extended into southern Ohio, where mean fire intervals fell from 8.4 (pre-industrial period) to 2.7 years (industrial period) at McAtee Run (Hutchinson et al. 2019). Several studies have used fire scars formed in oaks in second-growth forests to quantify fire histories dating back to the late 1800s (Hutchinson et al. 2008; McEwan et al. 2007; Sutherland 1997). These studies, which cover 13 sites, show that low severity dormant season fires occurred frequently from ca. 1875 to 1930, with mean fire return intervals typically 5 to 10 years. In addition, Hutchinson et al. (2008) showed that the establishment and persistence of maples (red and sugar) began when fires ceased. Finally, using fire scar data, Guyette et al. (2012) modeled mean fire return intervals for

14 | Page

Terrestrial Ecosystems Supplemental Report

1650 to 1850 for the continental United States; for the Ohio portions of the subsections, the mean period between fires ranged from about 7 years to 14 years.

Old-growth forests are relatively rare throughout the eastern United States due to extensive European land practices and industrial logging (Nowacki and Trianosky 1993). This is indeed the case in southeast Ohio where most fire-history studies are temporally and spatially constrained, with fire scars only dating back to the 1600s and 1700s on sites where yellow pines (e.g., pitch pine [Pinus rigida Mill.], shortleaf pine [Pinus echinata Mill.], Table Mountain pine [Pinus pungens]) were a sizable component of the forest. These yellow pines are much better fire-scar recorders than hardwoods because fire scars are preserved with resin and stumps are durable (less prone to rot) and long-lasting. Using cross-dating techniques, pine stumps, snags, and logs that have been dead for many years can be used in fire history studies. Although witness trees indicate that yellow pines were not historically abundant (<2 percent of trees) in southern Ohio (Dyer 2001), there were areas where oak-pine stands were somewhat common (e.g., Jones 1945). Ongoing fire history research at a site in Shawnee State Forest (Scioto County) is using live and dead yellow pines to document a longer-term fire record for this area (Hutchinson et al. 2019). At this 1-km2 (250-acre) site, 34 fires were recorded from 1796 to 1941, and the mean fire return interval was 4.4 years. In the preindustrial period (before 1850), the mean fire return interval was 8.4 years and only 9 percent of trees were scarred in each fire. In the industrial period (1850 to 1930), fires became more frequent (mean fire return interval = 2.7 years) and a greater percentage of trees were scarred per fire (23 percent), suggesting greater fire intensity. After nearly 140 years of moderate to high frequency fire (ca. 1796 to 1930), only one fire, in 1941, was recorded in the fire suppression era (1930 to Present).

In 1923, fire control began in earnest across southern Ohio. Fire towers were erected in the 1920s and 1930s, and detailed statistics were kept to record the number of fires and area burned in the Fire Control District (Leete 1938). Soon after fire control policies were implemented in southern Ohio’s Forest Fire District in 1923, the acreage burned annually was greatly reduced (Leete 1938). Although wildfires, nearly all caused by people, have continued to occur throughout the suppression era, effective control has resulted in a landscape where fire has been essentially absent from most forest stands for more than 80 years. Fire is no longer the primary disturbance factor that drove community assemblage, maintenance, and dynamics over thousands of years (Abrams et al. 2014), allowing succession to shade-tolerant, maple-dominated forests even on drier south-facing slopes (Drury and Runkle 2006). From Climatic Climax Theory to Disturbance Ecology The Southern Unglaciated Allegheny Plateau Section falls within Braun’s (1950) Mixed Mesophytic Forest Region, whereby mixed mesophytic forests (beech, maple, basswood, tulip poplar) are considered the climatic climax—the endpoint of vegetation succession. Although it is true that mixed mesophytic communities probably represented late seral stages in certain areas within the Mixed Mesophytic Forest Region (specifically valley bottoms, mesic coves, and cool, north-facing slopes), most of southeastern Ohio was historically in a subclimax state, dominated by mixed oaks and maintained over time by recurrent fire. On upper slopes and ridgetops,

15 | Page

Terrestrial Ecosystems Supplemental Report

edaphic restrictions (shallow soils, high insolation, and low moisture index) were such that mixed mesophytic conditions were untenable due to dryness and infertility:

Four subdivisions are recognized in the Cumberland and Allegheny Plateaus section of the Mixed Mesophytic Forest Region, one being the Low Hills Belt. This is an area of low relief and relatively gentle slopes. It is an irregular band extending from southern Kentucky northeastward to about Pittsburgh, and is widest in southeastern Ohio and adjacent Kentucky and West Virginia. Here, there was a larger proportion of oak in the original mixed forests (Braun 1950).

Oak-dominated systems were inherently shaped by disturbance, not despite it. Although insightful for its time, Braun’s concept was based on Clement’s (1916, 1936) dated successional model founded on equilibrium and stability, whereby vegetation change unfolds in a predictable, stepwise manner towards a climax condition governed principally by climate, further modified by edaphics (soils and topography), while discounting natural disturbance processes. Under Braun’s concept, disturbance is viewed negatively as setting back succession instead of being embraced as an important and vital driver of vegetation pattern development and maintenance and plant diversity. Today, we know that ecosystems operate with a high degree of complexity and dynamism, hence a greater array of ecological phenomena must be considered (Pahl-Wostl 1995; Cook 1996). Increasingly, historical and disturbance ecology principles are being viewed as critical components to land management (Engstrom et al. 1999; Egan and Howell 2001; Wiens et al. 2012) and have led to the emulation of past disturbance regimes as a silvicultural, “nature- based” way to successfully regenerate forests and restore ecosystems (Seymour et al. 2002; Crow and Perera 2004; Long 2009). Thus making this link among composition, structure, and disturbance processes paves the way for sound, ecologically-defensible land management decisions as managing for a diversity of forest ages and sizes can better prepare a resource to tolerate significant disturbances such as insect and disease outbreaks as well as support a variety of plants and wildlife (Albright et al. 2018). Ecological Subsections of Southeastern Ohio There are three primary subsections that comprise the study area, delineated mainly on geology (figure 2). The Ohio Valley Lowland Subsection (221Ec) spans the Ohio River, landing squarely on Permian-Pennsylvanian stratigraphy (figure 4) composed of sedimentary rocks of shale, sandstone, siltstone, and mudstone (Slucher et al. 2006). True to its name, this subsection encompasses lower elevations surrounding the Ohio River. The topography is fairly steep and rugged, consisting of a highly dissected terrain that drains to the Ohio River from south-flowing streams in Ohio. The catchment collects fine-textured fluvial sediments, so soils tend to be heavier than that of surrounding subsections, with high clay content and low sand content. The rich, mesic tendencies of this subsection are reflected in the current dominance of mesophytic trees, namely sugar maple (Acer saccharum Marshall), white oak, red maple (Acer rubrum L.), and tuliptree (Liriodendron tulipifera L.).

The East Hocking Plateau Subsection (221Ed) conforms to the Upper Pennsylvanian stratigraphy that strikes northeast to southwest across southeastern Ohio (figure 4). Roughly 302 million to

16 | Page

Terrestrial Ecosystems Supplemental Report

307 million years in age, the bedrock is composed of shale, sandstone, siltstone, and mudstone (Slucher et al. 2006). Soils tend to be heavy, fertile, and high in pH (5.8) compared to the other subsections. Wetlands are rather infrequent and mainly palustrine. Although elevation grades from 1,300 feet in the northeast to 550 feet in the southwest, percent slope and topographic roughness are rather consistent throughout. Sixty-nine tree species are currently recorded by FIA; the most numerous species include sugar maple, black cherry (Prunus serotina Ehrh.), American elm (Ulmus americana L.), black locust (Robinia pseudoacacia L.), red maple, tuliptree, and white ash (Fraxinus americana L.).

The Western Hocking Plateau Subsection (221Ef) spans Lower Pennsylvanian and Mississippian stratigraphies of sandstone, siltstone, shale, conglomerate, and some limestone of principally marine origin (Slucher et al. 2006). Elevation is rather uniform, ranging mostly from 700 to 1,000 feet, and percent slope and topographic roughness are subdued compared to other subsections. Areas of rather high sand concentrations and soil permeability exist in this subsection, presumably due to the underlying sandstone. Wetlands are infrequent and largely palustrine. Today, tree diversity is very high, with 70 species. The most populous species are red maple, tuliptree, white oak, sugar maple, chestnut oak (Quercus montana Willd.), black oak (Quercus veluntina Lam.), sassafras (Sassafras albidum (Nutt.) Nees), and Virginia pine (Pinus virginiana Mill.). Pre-Euro-American Settlement Vegetation Witness trees from Public Land Surveys can be used to document past tree composition and underlying disturbance processes that help describe important reference conditions and guide silvicultural and restoration goals. Coverage by a witness-tree database created by Dyer and Hutchinson (2019) adequately characterizes pre-settlement vegetation for three Ecological Subsections of southeastern Ohio (figure 7). The Ohio Valley Lowland Subsection (221Ec) was overwhelmingly white oak (53%), following distantly by the black oak subgroup (14%) and hickory (8%) (Table 4). All of these are considered fire-tolerant or pyrophilic species/genera (Thomas-Van Gundy and Nowacki 2013; Nowacki and Abrams 2015). Fire-sensitive, mesophytic trees were infrequent and comprised mainly of American beech (Fagus grandifolia Ehrh.) (4%) and sugar maple (3%). Across landtype classes of increasing moisture and fertility (Dry Oak to Rolling Bottomlands), pyrophytes white oak, black oak subgroup, and hickory (Carya spp.) progressively decreased, whereas the principal mesophytes incrementally increased (beech: 1% to 9%; sugar maple <1% to 6%) (table 4). White oak dominated the East Hocking Plateau (221Ed) as well, though at a lower percentage (36%), followed by hickory (14%), beech (11%), and black oak subgroup (11%) (table 5). Mesophytes sugar maple and red maple each represented 5% of the witness trees. As before, white oak, hickory, and black oak subgroup generally decreased as landtype class moisture and fertility increased, whereas mesophytes beech, sugar maple, and red maple increased along this edaphic gradient (table 5). Pyrophytes white oak (35%), hickory (14%), and black oak subgroup (13%) were the most common trees in the Western Hocking Plateau (221Ef), followed by the mesophyte beech (10%) (table 6). Blackgum (Nyssa sylvatica Marshall) and red maple each comprised 4% of the witness trees. Across the landtype class gradient of increasing moisture and fertility, pyrophytes white oak and

17 | Page

Terrestrial Ecosystems Supplemental Report black oak decreased, mesophytes beech and maples increased, while hickory and blackgum showed maximum frequency in the middle (Dry-Mesic Mixed Oak Hardwood forest) (table 3).

Table 3. Abundance of witness tree taxa by ecological subsection of Section 221E 221Ec – Ohio 221Ed – East 221Ef – Western Valley Lowland Hocking Plateau Hocking Plateau Pyrophilic or (Percentage of (Percentage of (Percentage of Taxon* Pyrophobic total, by taxon) total, by taxon) total, by taxon) White oak Pyrophilic 52.8 35.7 35.1 Hickory Pyrophilic 7.7 14.2 14.1 Black oak† Pyrophilic 13.6 10.6 13.4 Beech Pyrophobic 4.1 10.8 9.5 Sugar maple Pyrophobic 2.7 4.6 3.3 Red maple Pyrophobic 1.4 4.6 3.5 Chestnut oak Pyrophilic 4.3 3.0 3.3 Tulip-poplar Pyrophilic 1.9 3.1 3.1 Blackgum Pyrophilic 1.1 1.7 4.3 Ash Pyrophobic 1.4 3.1 1.1 Elm Pyrophobic 1.3 1.4 1.2 Silver maple Pyrophobic 1.2 1.4 1.6 Pine Pyrophilic 3.2 1.3 0.9 Chestnut Pyrophilic -- 0.2 2.5 Buckeye Pyrophobic 1.6 1.7 0.3 Sycamore Pyrophobic 0.4 0.7 0.5 Butternut Pyrophobic 0.1 0.6 0.5 Black walnut Pyrophobic 0.3 0.6 0.3 Basswood Pyrophobic 0.1 0.5 0.3 Locust Pyrophilic 0.1 0.0 0.1 Oak-other Pyrophilic 0.1 0.1 0.3 Black cherry Pyrophobic 0.1 0.1 0.2 Sassafras Pyrophobic -- 0.1 0.2 Mulberry Pyrophobic 0.2 0.0 0.1 Aspen Pyrophilic 0.1 -- 0.1 Boxelder Pyrophobic 0.3 -- 0.0 Birch Pyrophobic -- 0.0 0.1 Sourwood Pyrophilic -- -- 0.1 Hackberry Pyrophobic ------Cottonwood Pyrophobic ------Willow Pyrophobic ------*Taxa listed in order of descending abundance (>1 percent) within the ecological subsection. †Quercus velutina and probably also Q. rubra, Q. coccinea, and other species of the red oak group.

18 | Page

Terrestrial Ecosystems Supplemental Report

Table 4. Pre-Euro-American settlement vegetation based on witness-tree percentages in each ecological landtype class within the Ohio portion of Ecological Subsection 221Ec – Ohio Valley Lowland Dry Oak Rolling Bottomlands Pyrophilic Number of forest Dry-Mesic Mixed Oak Mixed Hardwood or witness (percent in Hardwood forest forest Taxon* Pyrophobic trees landtype) (percent in landtype) (percent in landtype) White oak Pyrophilic 771 56.3 55.0 47.2 Black oak† Pyrophilic 198 17.1 13.9 9.3 Hickory Pyrophilic 112 8.2 9.1 5.9 Chestnut oak Pyrophilic 63 6.2 5.8 1.0 Beech Pyrophobic 60 0.9 2.4 9.1 Pine Pyrophilic 46 6.7 0.7 1.2 Sugar maple Pyrophobic 40 0.2 2.9 5.5 Tulip-poplar Pyrophilic 27 1.8 1.9 1.8 Buckeye Pyrophobic 23 0.4 1.2 3.3 Red maple Pyrophobic 20 0.2 1.7 2.4 Ash Pyrophobic 20 0.2 1.4 2.6 Elm Pyrophobic 19 0.2 0.7 3.0 Silver maple Pyrophobic 17 0.7 0.2 2.4 Blackgum Pyrophilic 16 0.7 1.2 1.4 *Taxa listed in order of descending abundance (>1 percent) within the ecological subsection. †Quercus velutina and probably also Q. rubra, Q. coccinea, and other species of the red oak group.

Table 5. Pre-Euro-American settlement vegetation based on witness-tree percentages in each ecological landtype class within the Ohio portion of Ecological Subsection 221Ed – East Hocking Plateau Dry Oak Rolling Bottomlands Pyrophilic Number of forest Dry-Mesic Mixed Oak Mixed Hardwood or witness (percent in Hardwood forest forest Taxon* Pyrophobic trees landtype) (percent in landtype) (percent in landtype) White oak Pyrophilic 1,076 39.4 36.5 30.5 Hickory Pyrophilic 427 14.2 16.0 12.6 Beech Pyrophobic 326 8.0 10.2 14.8 Black oak† Pyrophilic 320 16.9 7.4 5.8 Sugar maple Pyrophobic 139 2.3 5.1 7.1 Red maple Pyrophobic 138 3.0 4.8 6.3 Tulip-poplar Pyrophilic 93 2.3 4.4 2.9 Ash Pyrophobic 92 2.4 3.8 3.1 Chestnut oak Pyrophilic 90 4.5 3.4 0.8 Blackgum Pyrophilic 50 0.9 2.4 1.9 Buckeye Pyrophobic 50 0.3 1.6 3.3 Elm Pyrophobic 42 0.7 0.6 3.0 Silver maple Pyrophobic 41 1.3 1.2 1.7 Pine Pyrophilic 40 2.4 0.2 0.9 *Taxa listed in order of abundance (>1 percent) within the ecological subsection. †Quercus velutina and probably also Q. rubra, Q. coccinea, and other species of the red oak group.

19 | Page

Terrestrial Ecosystems Supplemental Report

Table 6. Pre-Euro-American settlement vegetation based on witness-tree percentages in each ecological landtype class within the Ohio portion of Ecological Subsection 221Ef – Western Hocking Plateau Dry-Mesic Mixed Rolling Oak Hardwood Bottomlands Mixed Pyrophilic Number of Dry Oak forest forest Hardwood forest or witness (percent in (percent in (percent in Taxon* Pyrophobic trees landtype) landtype) landtype) White oak Pyrophilic 1,215 35.8 35.3 33.8 Hickory Pyrophilic 489 13.7 17.9 11.2 Black oak† Pyrophilic 465 19.6 10.2 7.5 Beech Pyrophobic 329 8.5 8.4 12.1 Blackgum Pyrophilic 150 3.4 5.5 4.5 Red maple Pyrophobic 121 2.2 2.6 6.3 Sugar maple Pyrophobic 114 1.6 3.9 5.1 Chestnut oak Pyrophilic 114 4.8 2.8 1.6 Tulip-poplar Pyrophilic 107 3.4 3.7 2.1 Chestnut Pyrophilic 86 2.8 3.3 1.3 Silver maple Pyrophobic 56 1.5 1.1 2.3 Elm Pyrophobic 42 0.3 0.7 3.0 Ash Pyrophobic 38 0.5 1.3 1.8 *Taxa listed in order of abundance (>1 percent) within the ecological subsection. †Quercus velutina and probably also Q. rubra, Q. coccinea, and other species of the red oak group.

20 | Page

Terrestrial Ecosystems Supplemental Report

Figure 7. Area from which witness tree data were obtained from two Ohio land surveys conducted in the late 18th century, relative to ecological subsection boundaries of Section 221E

Witness Trees as Pyroindicators Further analysis was conducted by dividing witness trees into two classes—pyrophiles and pyrophobes—based on synecological and autecological traits to produce a pyrophilic percentage map for the Wayne National Forest using the methods of Thomas-Van Gundy and Nowacki (2013) (database obtained through Dr. Todd Hutchison, Forest Service, Northern Research Station). The resulting map illustrates that a high degree of pyrophilicity once existed across all units in pre-Euro-American settlement times (figure 8). With little observable spatial variation, ecological units were used to investigate and discern possible patterns and underlining ecological relationships at multiple scales. First, the pyrophilic percentage map was clipped to subsection boundaries and distributional histograms created to assess if pyrophilic percentages differed among these coarse-scale ecological units (figure 9). All three Subsections had similar negatively skewed distributions (right-leaning curves), indicative of comparable (high) fire settings at this spatial scale (Thomas-Van Gundy and Nowacki 2016). Fine-scale landtype classes (defined in following section) were subsequently used as analysis units and a similar distributional pattern occurred among all three (figure 10). Perhaps this is not unexpected given the combined dominance of oak, hickory, and other pyrophilic trees across the landscape (table 4, table 5, and

21 | Page

Terrestrial Ecosystems Supplemental Report

table 6) and the strong spatial autocorrelation of fire (i.e., areas adjacent to active fire have a high probability of burning too). Alternatively, perhaps the pyrophilic percentage map is at a resolution too coarse to capture landtype class-level characteristics. In any case, to maintain such a large component of pyrogenic trees across all settings, fire seemingly was a landscape-level phenomenon, burning regularly across most of southeastern Ohio and embedded topographic positions. Only the pervasiveness of pre-Euro-American settlement fire can adequately explain oak (and associated pyrophiles) dominance throughout, especially on more rich and mesic locations, such as the Rolling Bottomlands Mixed Hardwood Forest, where competition from shade-tolerant mesophytes is otherwise overwhelming (Iverson et al. 2018). Due to their desirability by humans (level land for habitation, fertility and arability for food production, close proximity to drinking water and river transportation, etc.), American Indians undoubtedly bolstered fire in the Rolling Bottomlands (Stambaugh and Guyette 2008).

22 | Page

Terrestrial Ecosystems Supplemental Report

Figure 8. Pyrophilic percentage map for units of the Wayne National Forest Note: No witness trees were recorded south of the Ohio River (outside the state of Ohio), hence pyrophilic percentages drop precipitously along the state boundary adjacent to the Marietta Unit, an artifact of no data.

23 | Page

Terrestrial Ecosystems Supplemental Report

35

30

25

20

% of Subsection WHP 15 EHP OVL 10

5

0 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 Pyrophilic Percentage Class

Figure 9. Subsections partitioned by pyrophilic percentage class. OVL = Ohio Valley Lowland (221Ec), East Hocking Plateau (221Ed), WHP = Western Hocking Plateau (221Ef)

30

25

20

Dry Oak Forest 15 Dry-mesic Mixed Oak Hardwood

% of of Landtype % Forest 10 Rolling Bottomland Mixed Hardwood Forest 5

0

Figure 10. Landtype classes partitioned by pyrophilic percentage class

24 | Page

Terrestrial Ecosystems Supplemental Report

Ecological Landtype Classes Three ecological landtype classes have been delineated and described for the 17-county study area by Iverson et al. (2019a) and include Dry Oak Forest, Dry-Mesic Mixed Oak Forest, and Rolling Bottomlands Mixed Hardwood Forest. Key ecosystem characteristics (vegetation composition and structure, drivers, stressors, and connectivity) are described for each landtype class below. Integrity ratings have also been qualified to inform forest plan development.

LANDFIRE (Landscape Fire and Resource Management Planning Tools) was used to document current land use and model vegetation trends and historical fire regimes. LANDFIRE is a shared program between the wildland fire management programs of the U.S. Department of Agriculture Forest Service and U.S. Department of the Interior, providing landscape scale geo-spatial products to support cross-boundary planning, management, and operations. This multi-partner program produces consistent, comprehensive, geospatial data and databases that describe vegetation, wildland fuel, and fire regimes across the United States and insular areas. LANDFIRE is a cornerstone of a fully integrated national data information framework, developing and improving vegetation and fuels data products based on the best available authoritative data and science in an all lands landscape conservation approach based on inter- organizational collaboration and cooperation. LANDFIRE refers to historic ecosystems as “Biophysical Settings.” Using an expert-based development process to identify deterministic and probabilistic model parameters, quantitative state-and-transition models that describe pre- settlement ecosystem structure and function are created for every Biophysical Setting. These models are used to estimate the natural range of variability for each Biophysical Setting, which are then used to help evaluate ecosystem health through a departure metric called Vegetation Departure. These models can be examined and modified through ST-Sim, a generalized stochastic space/time based modeling framework.

To assess connectivity across the study area, we used the “Local Connectedness” dataset3 developed by The Nature Conservancy’s Eastern Science Team (Anderson et al. 2016). Connectivity estimates how easily organisms can move and access suitable habitat based on the arrangement of roads, industrial agriculture, development, and other human structures. In general each “cell,” or 30m square of land, is assigned a “resistance” value based on the degree it alters flow (e.g., of an organism) arriving from an adjacent cell. A “resistance grid” was developed from these assignments and modified land cover datasets. Further, Local Connectedness is calculated for a 28km2 circle around each point to assess the surrounding context. Areas closest to the point are weighted higher (figure 11).

3 This dataset and resultant maps are not designed to assess how easily any one particular species might move— requirements for all species of potential interest are too variable to capture in one dataset. Rather, the local connectedness metric helps to assess long-term movement potential in general, with potential movement in response to climate change in mind (i.e., not seasonal migrations for example).

25 | Page

Terrestrial Ecosystems Supplemental Report

Figure 11. Connectedness map of the 17-county study area and embedded Wayne National Forest

Dry Oak Forest Landtype Class This oak-dominated landtype class occurs on rugged upland positions, primarily ridgetops and southwest-facing upper slopes, and is distributed throughout southeast Ohio comprising 40% of the 17-county study area (figure 12). Here, available water storage capacity is relatively low, with a low integrated moisture index score. The inherent dryness of these sites promote xerophytic tree species. The dry oak and oak/pine forest and woodland forest ecosystem is encompassed within this landtype class.

26 | Page

Terrestrial Ecosystems Supplemental Report

Figure 12. Distribution of the Dry Oak landtype class across the 17-county study area Note: The higher resolution inset map provides better spatial depiction of this landtype class across the landscape. Vegetation Composition and Structure This landtype class was typically dominated by white oak, black oak, northern red oak (Quercus rubra L.), scarlet oak (Quercus coccinea Münchh.), and chestnut oak, with small inclusions of Virginia pine, shortleaf pine, and pitch pine (Pinus rigida Mill.) (Iverson et al. 2019a; LANDFIRE 2019a). Ericaceous shrubs are common in the understory and include such characteristic species as Blue Ridge blueberry (Vaccinium angustifolium Aiton) and mountain laurel (Kalmia latifolia L.) (Butler et al. 2015; LANDFIRE 2019a). Four vegetation classes have been quantified that describe the natural range of variability of the pre-Euro-American settlement landscape: early development (Class A), mid-development 1 – open (B), late-development 1 – open (C), and late-development 2 – closed (D) (LANDFIRE 2019a). Composition and structure grade from an herbaceous or savanna community with limited trees maintained by frequent fire to a late succession open oak forest (figure 13).

27 | Page

Terrestrial Ecosystems Supplemental Report

Figure 13. State-and-transition model showing the natural range of variability for the Dry Oak Forest landtype class.

Drivers The predominant fire regime for this landtype class is Fire Regime I, characterized by low- severity surface fires. Historically, Indigenous fires accounted for more than 95% of the ignitions over these landscapes. Vegetation types varied based on fire frequency and intensity. This type is naturally dominated by stable, uneven-aged forests. Most oaks are long-lived with typical age of mortality ranging from 200-400 years. Scarlet and black oaks are shorter-lived with typical ages being approximately 50-100 years, while white oaks can live as long as 600 years. A mixed pine component exists on poor soils on ridgetops. Extreme wind or ice storms occasionally create larger canopy openings (LANDFIRE 2019a). Current Conditions, Stressors, and Connectivity Forests currently dominate this landtype class within the study area (76%; figure 14) and the Wayne National Forest (82%; figure 15). Agriculture is the next common land use, more so over the study area (18%) than the proclamation boundary (9%). Current forests have uncharacteristically high canopy closure when compared with the more open conditions of the

28 | Page

Terrestrial Ecosystems Supplemental Report

past (Figure 16), especially within the Wayne (47%; Figure 17). The historic open oak woodlands now have much greater tree densities resulting in less—or even absent—oak regeneration in forest understories, low diversity of grasses and forbs, and significantly reduced soft mast shrubs. Early successional stages (e.g., regenerating oaks, herbaceous vegetation) are also absent from current conditions. This is the result of 80+ years of fire suppression, mesophication, and densification, impacting both species composition and structure. Red maple has been strongly associated as an invader of this landtype class on lower pH soils (Radcliffe et al. 2020). It is estimated that 10% of this landtype class would have historically burned annually, burning over 200,000 acres of the study area (figure 18) and 37,000 acres within the Wayne (figure 19). In contrast, few acres burn today in the study area and approximately 2,000 acres are burned annually within the Wayne National Forest across all landtype classes.

Increased drought conditions, especially during the growing season, may increase susceptibility to insect pests and diseases. Non-native invasive species, including tree-of-heaven (Ailanthus altissima (Mill.) Swingle), Japanese stiltgrass (Microstegium vimineum (Trin.) A. Camus), multiflora rose (Rosa multiflora Thunb.), bush honeysuckle (Lonicera spp.), autumn olive (Eleaegnus umbellata Thunb.), and Japanese barberry (Berberis thunbergii DC.) often outcompete native herbs and shrubs in this class, and are also likely to benefit from warmer temperatures and increased disturbance. Invasive shrubs in the understory may provide additional ladder fuels and increase fire intensity where wildfires occur (Butler et al. 2015).

The study area includes a wide variety of land uses from urban to protected, environmentally sensitive areas, which results in a range of connectedness values. The highest connectedness levels occur in areas managed for conservation such as within National Forest System or Ohio Department of Natural Resource lands. Connectedness distributions formed a bell-shaped curve pronouncedly cresting at the “average” class (figure 20). The flanks of the bell were pretty much evenly distributed among “above average” and “below average” classes for the Dry Oak Forest Landtype class.

This ecosystem has low-moderate integrity based on a variety of factors:

• Moderate-High Vegetation Departure (value of 65 on a scale of 0 to 100, with low values being closest to reference conditions) with a substantial overrepresentation of closed canopy succession classes and underrepresentation of more open classes; • Severe alteration of fire regimes, leading to closed canopy succession classes, and concomitant loss of native biodiversity; • Average to Slightly Above Average Connectedness; • Moderate to High climate change adaptive capacity; • Low vulnerability to climate change; • Represents the best “zone of investment” for regenerating and supporting oak of all landtype classes (Iverson et al. 2018).

29 | Page

Terrestrial Ecosystems Supplemental Report

Figure 14. Current land use acres (percentages) for the Dry Oak Forest landtype class within the 17- county study area

Figure 15. Current land use acres (percentages) for the Dry Oak Forest landtype class within the Wayne National Forest proclamation boundary

30 | Page

Terrestrial Ecosystems Supplemental Report

Figure 16. Vegetation class acres (percentages) for the Dry Oak Forest landtype class within the 17- county study area

Figure 17. Vegetation class acres (percentages) for the Dry Oak Forest landtype class within the Wayne National Forest proclamation boundary

31 | Page

Terrestrial Ecosystems Supplemental Report

Figure 18. Historic annual acres (percentages) burned for the Dry Oak Forest landtype class within the 17-county study area

Figure 19. Historic annual acres (percentages) burned for the Dry Oak Forest landtype class within the Wayne National Forest proclamation boundary

32 | Page

Terrestrial Ecosystems Supplemental Report

Figure 20. Connectedness class distribution for the Dry Oak Forest landtype class for the 17-county study area

Dry-Mesic Mixed Oak Hardwood Forest Landtype Class This landtype class is also well distributed across uplands throughout southeast Ohio, comprising 29% of the 17-county study area (figure 21). Here too, forests currently dominate, though at a lower level with the study area (68%; figure 23) compared to the Wayne National Forest (82%; figure 24). The Dry-Mesic Mixed Oak Hardwood Forest landtype class occurs primarily on midslope positions on northeasterly aspects. The integrated moisture index score for this landtype class is intermediate, indicating a dry-mesic setting. The upper and midslopes of the dry/mesic oak forest ecosystems are encompassed within this landtype class.

33 | Page

Terrestrial Ecosystems Supplemental Report

Figure 21. Distribution of the Dry-mesic Mixed Oak Hardwood Forest landtype class across the 17- county study area Note: The higher resolution inset map provides better spatial depiction of this landtype class across the landscape. Vegetation Composition and Structure This landtype class is dominated by oaks, especially white oak (LANDFIRE 2019b). Varying amounts of hickory, blackgum, and American chestnut (Castanea dentata (Marshall) Borkh.) demonstrated a positive association with this class (Iverson et al. 2019a). Subcanopies and shrub layers are usually well-developed by American witchhazel (Hamamelis virginiana L.), flowering dogwood (Cornus florida L.), and hophornbeam (Ostrya virginiana (Mill.) K. Koch) (LANDFIRE 2019b). Five vegetation classes have been quantified that describe the natural range of variability of the pre-Euro-American settlement landscape: early development (A), mid- development 1 – open (B), mid-development 2 – open (C), mid-development 3 – closed (D), and late-development 1 – closed (E). Composition grades from an herbaceous community with limited trees maintained by frequent fire (every 1-5 years) to a late succession oak forest (figure 22).

34 | Page

Terrestrial Ecosystems Supplemental Report

Figure 22. State-and-transition model showing the natural range of variability for Dry-Mesic Mixed Oak Hardwood Forest Landtype class

Drivers This landtype class is predominantly Fire Regime I, characterized by low-severity surface fires. Historically, Indigenous fires accounted for over 95% of the ignitions over these landscapes. These fire-driven open conditions increased food-producing plants and aided hunting large ungulates. Vegetation types varied based on fire frequency and intensity. Ice damage, periodic insect defoliation, windstorms, and historically the extinct passenger pigeon likely contributed to increased canopy openings that facilitated light penetration to the forest floor, and ultimately, greater possibility of germination and recruitment of oaks (LANDFIRE 2019b). Current Conditions, Stressors, and Connectivity Agriculture was a common land use over the study area (21%; figure 23), but less so within the Wayne (9%; figure 24). Historically, oak-dominated ecosystems were the principal cover type of the study area (57%; figure 25) and the Wayne (61%; figure 26). Similar to the Dry Oak Forest landtype class, mesophication and densification processes are again evident as today’s systems have moved to closed-canopy oak forests and mixed mesophytic forests across the study areas (33% and 29%, respectively). This shift is even starker within the Wayne National Forest, with 40% and 37% of the land in closed-canopy oak forests and mixed mesophytic forests. American beech has been found to be strongly associated with the landscape positions associated with this landtype class (Radcliffe et al. 2020). It is estimated that 14% of the land would historically burn

35 | Page

Terrestrial Ecosystems Supplemental Report

annually, affecting 220,000 acres of the study area (figure 27) and 34,000 acres over the Wayne National Forest (figure 28). However, these number are greatly reduced today.

Increased drought risk, especially during the growing season, may increase susceptibility to insect pests and diseases. Non-native invasive species, such as tree-of-heaven, Japanese stiltgrass, and garlic mustard (Alliaria petiolata (M. Bieb.) Cavara & Grande), which often outcompete native herbs and shrubs, are expected to do well in warmer temperatures. Low- severity late-season drought generally favors oak species, although severe drought may hinder regeneration, or combine with other stressors to make individuals more susceptible to mortality or reduced productivity (Butler et al. 2015). This type has largely converted to closed-canopy forests progressively increasing in mesophytic species. As these systems become increasingly mesophytic, the ability to get fire back on the landscape becomes increasing difficult (LANDFIRE 2019b).

Similar to the Dry Oak Forest landtype class, the flanks of the connectivity bell were pretty much evenly distributed among “above average” and “below average” classes for the Dry-Mesic Mixed Oak Hardwood Forest landtype class (figure 29).

This ecosystem has low-moderate integrity based on a variety of factors:

• High Vegetation Departure (value of 72 on a scale of 0 to 100, with low values being closest to reference conditions) driven primarily by a substantial underrepresentation of the open succession classes and a concomitant overrepresentation of closed canopy succession classes; • Vegetation Departure largely driven by altered fire regimes that would have historically led to dominance by the open canopy succession classes; • Average to Slightly Above Average Connectedness; • High adaptive capacity due to drought tolerance and adaptations to fire; • Low-moderate vulnerability due to high number of species and occurrence on wide range of habitats; • Represents a relatively poor “zone of investment” for regenerating and supporting oak (Iverson et al. 2018).

36 | Page

Terrestrial Ecosystems Supplemental Report

Figure 23. Current land use acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the 17-county study area

Figure 24. Current land use acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the Wayne National Forest proclamation boundary

37 | Page

Terrestrial Ecosystems Supplemental Report

Figure 25. Succession class acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the 17-county study area

Figure 26. Succession class acres (percentages) for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the Wayne National Forest proclamation boundary

38 | Page

Terrestrial Ecosystems Supplemental Report

Figure 27. Historic annual acres (percentages) burned for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the 17-county study area

Figure 28. Historic annual acres (percentages) burned for the Dry-Mesic Mixed Oak Hardwood Forest landtype class within the Wayne National Forest proclamation boundary

39 | Page

Terrestrial Ecosystems Supplemental Report

Figure 29. Connectedness class distribution for the Dry-Mesic Mixed Oak Hardwood Forest landtype class for the 17-county study area

Rolling Bottomlands Mixed Hardwood Forest Landtype Class This landtype class forms a web-like pattern, being concentrated within riparian corridors throughout the study area (figure 30). It occurs on mesic-to-wet, lower topographic positions, comprising 31% of the 17-county study area. They encompass broad-to-narrow valley floors and surrounding toeslopes. The lower slopes of the dry/mesic oak forest, mixed mesophytic and cove forest, small stream riparian forest, and large stream floodplain and riparian forest ecosystems are encompassed within this landtype class.

40 | Page

Terrestrial Ecosystems Supplemental Report

Figure 30. Distribution of the Rolling Bottomlands Mixed Hardwood Forest landtype class across the 17-county study area Note: The higher resolution inset map provides better spatial depiction of this landtype class across the landscape. Vegetation Composition and Structure This landtype class was generally dominated by oaks and hickories, encompassing 50-63% of the witness trees. White oak was the primary oak species recorded, but species from the red oak group were also noted. However, mesic species such as beech, elm, ash, buckeye (Aesculus spp.), and sugar, red, and silver maples (Acer saccharinum L.) demonstrate strong site affinity for this type and encompassed 26-42% of the witness trees (Iverson et al. 2019a). The shrub and understory layers ranged widely from an open oak forest community, to a shade-tolerant mesophytic forest herbaceous community, to a flood-adapted, hydrophytic community. No single LANDFIRE Biophysical Setting matched this landtype class, but the same expert-based development process was used to create an ST-Sim state-and-transition model. Four vegetation classes have been quantified that describe the natural range of variability of the pre-settlement landscape: early development (A), mid-development 1 – open (B), late-development 1 – open (C), and late-development 2 – closed (D). Vegetation dynamics range from an open herbaceous/shrub community to a closed mixed mesophytic forest (figure 31).

41 | Page

Terrestrial Ecosystems Supplemental Report

Figure 31. State-and-transition model showing the natural range of variability for the Rolling Bottomlands Mixed Hardwood Forest landtype class

Drivers This type is predominantly Fire Regime I, characterized by low-severity surface fires. Historically, Indigenous fires accounted for over 95% of the ignitions over these landscapes. These fire-driven open conditions increased food-producing landscapes and aided hunting large ungulates. The embedded floodplain ecosystems are produced and maintained by active hydrologic and geomorphic processes such as channel meandering, sedimentation, and erosion (Gregory et al. 1991) caused by natural hydrological variation (Richter and Richter 2000). Regeneration of the early floodplain vegetation is dependent on flooding, scouring, and movement of river channels, which creates bare, moist soil needed for seedling establishment. Oxbow and slough development also influence the floodplain system and create variability in plant community composition. Vegetation types vary based on disturbance frequency and intensity. Current Conditions, Stressors, and Connectivity The desirability of this landtype class for agriculture and development is clearly evident, especially across the study area (31%, 16%; figure 32) compared to the Wayne National Forest

42 | Page

Terrestrial Ecosystems Supplemental Report

(16%, 14%; figure 33). Accordingly, forests here currently comprise a lower proportion of the land compared to the other two landtype classes (52% of the study area; 69% of the Wayne). The high degree of land conversion makes past-to-current vegetation comparisons difficult (figure 34 and figure 35). In any case, there has been a distinct shift away from historic oak woodlands and forests and an uptick in the regenerating tree class. In addition, the floodplain component of this landtype class has experienced significant decoupling from the streams and rivers resulting in deep entrenchment, which has virtually eliminated the active hydrologic and geomorphic processes of floodplain habitat development. Although historically this young regenerating tree class was dominated by oaks, it is today represented by a variety of trees, mainly mesophytes (American beech, American basswood [Tilia americana L.], sugar maple, yellow buckeye) (Radcliffe et al. 2020), from 10 to 25 meters with canopy closure of greater than 70%. It is estimated that 7% of this landtype class historically burned annually, resulting in >110,000 acres burned over the study area (figure 36) and >15,000 acres over the Wayne (figure 37). However, these number have been greatly reduced in today’s forests.

Increased drought risk, especially during the growing season, may increase susceptibility to insect pests and diseases. Low-severity late-season drought generally favors oak species, although severe drought may hinder regeneration, or combine with other stressors to make individuals more susceptible to mortality or reduced productivity. Non-native invasive species, such as tree-of-heaven, Japanese stiltgrass, garlic mustard, and bush honeysuckle, which often outcompete native herbs and shrubs, are expected to do well in warmer temperatures. Increases in invasive species in the mesic coves could increase fire fuels leading to potentially more intense fires and therefore negatively impact the fire-intolerant communities. Increased flashiness followed by dry periods could cause amplification of the current hydrologic cycle, potentially increasing the spread and establishment of current and newly introduced invasive species in the small stream riparian areas. In the floodplains, increases in storm intensity and flooding events have the potential to increase soil erosion and sedimentation, and compound anthropogenic stressors such as agricultural runoff and industrial pollution (Butler et al. 2015).

Connectivity skewness was evident for this landtype class, with higher values found for “below average” classes. This makes sense as this type is more heavily affected by agriculture and development (figure 38) than the other two.

This ecosystem has low integrity based on a variety of factors:

• Moderate-High Vegetation Departure (value of 61 on a scale of 0 to 100, with low values being closest to reference conditions) with an overrepresentation of the mid-seral, closed canopy succession class and an underrepresentation of mixed mature, closed canopy succession class; • Historic Vegetation Departure patterns likely driven by logging in this ecosystem which results in a younger forest on average. Current patterns reflect pasture and cropland in these areas; • Low adaptive capacity due to highly restricted habitat and disconnection of this ecosystem from water bodies as a result of roads and other infrastructure;

43 | Page

Terrestrial Ecosystems Supplemental Report

• High vulnerability due to projected altered hydrology and invasive species; • Represents a relatively poor “zone of investment” for regenerating and supporting oak (Iverson et al. 2018).

Figure 32. Current land use acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the 17-county study area

44 | Page

Terrestrial Ecosystems Supplemental Report

Figure 33. Current land use acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the Wayne National Forest proclamation boundary

Figure 34. Succession class acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the 17-county study area

45 | Page

Terrestrial Ecosystems Supplemental Report

32%

4%

18% 35% 9% 40% 35% 22% 1% 3%

Figure 35. Succession class acres (percentages) for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the Wayne National Forest proclamation boundary

Figure 36. Historic annual acres (percentages) burned for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the 17-county study area

46 | Page

Terrestrial Ecosystems Supplemental Report

Figure 37. Historic annual acres (percentages) burned for the Rolling Bottomlands Mixed Hardwood Forest landtype class within the Wayne National Forest proclamation boundary

Figure 38. Connectedness class distribution for the Rolling Bottomlands Mixed Hardwood Forest landtype class for the 17-county study area

47 | Page

Terrestrial Ecosystems Supplemental Report

Rare & Novel Habitats Biodiversity can be measured across different spatial scales. Landscape diversity is best measured through the diversity of habitats and associated species within a landscape, also known as gamma diversity. Eleven rare habitats and one novel habitat, embedded within the landtype classes, have been identified in the Wayne (figure 39). These habitats provide critical microsites that bolster diversity of genes, species, and ecological processes, and enhance ecosystem health and resilience.

Figure 39. Ecological Landtype Classes, Forest Ecosystems, and Rare and Novel Habitats in the Wayne National Forest

48 | Page

Terrestrial Ecosystems Supplemental Report

Western Allegheny Plateau Oak Barrens4 This barren habitat occurs on moderate to steep hillside slopes and more level to gently rolling ridgetops with southern or southwestern exposure. Bedrock is often exposed at the surface, and seepage zones from porous bedrock may be present. In the absence of fire, tree density and diversity increase while herbaceous cover and diversity decrease. Prior to European contact, bison were also thought to be important ecological factors. Oak barrens are small, often only one to twenty acres in size, occurring scattered throughout the seral class for woodlands that is today greatly reduced. Remnant examples of this habitat type are represented at Buffalo Beats Research Natural Area. Other examples include Bluegrass Ridge, Handley Branch, and Fradd Hollow Special Interest Areas. Appalachian Low-Elevation Mixed Pine/Blue Ridge Blueberry Forest5 This habitat develops on narrow ridges and steep slopes and positions with high solar exposure. Virginia pine is most often the dominant canopy species, but pitch pine, short-leaf pine, and dry- site oak species are significant canopy associates. Shortleaf pine and pitch pine were collectively a minor species across southeast Ohio, comprising <2% of the pre-settlement forest (Dyer and Hutchinson 2019). Pine is even rarer today due to the cutting of larger seed trees and post- logging consumption of regenerating trees by wildfire (Nowacki and Abrams 1992). Lick Branch Special Interest Area supports an example of this habitat type. Appalachian Chestnut Oak-Mixed Oak Forest6 This habitat occurs on dry to subxeric upper slopes and narrow ridgetops. Chestnut oak and scarlet oak are the dominant trees in the stands, with white oak a common co-dominant. Black oak and northern red oak may be just as common in this habitat within the study area. The American chestnut was historically noted in this habitat, but now only occur as resprouts. Soils are typically shallow and occur over non-calcareous sandstone, conglomerate, or shale. Examples of this habitat can be found at Felter Ridge, Thompson Cemetery Woods, and Young’s Branch Special Interest Areas.

American chestnut was one of the slowest species to extend its range northward following the retreat of the Laurentide ice sheet (Davis 1983). Chestnut arrived in western Tennessee around 15,000 year ago, and expanded slowly up the Appalachian Mountains chain, reaching Ohio around 5,000 years ago and arriving in Maine—the northern extent of its range—only about 2,000 years ago (Davis 1983). Before the arrival of blight (caused by Cryphonectria parasitica), American chestnut was a dominant tree of eastern forests, with a range spanning from central Alabama to southern Maine, extending to the west to Tennessee, Kentucky, Southern Indiana, Ohio, and southern Ontario (Russell 1987). In Ohio, chestnut was generally confined to sandstone soils of the unglaciated Allegheny Plateau, though small populations were found on

4 Sneddon et al. 2003 5 Patterson 2010 6 Faber-Langendoen et al. 2002

49 | Page

Terrestrial Ecosystems Supplemental Report

the Lake Plains in the northwestern part of the state (Braun 1961; Sears 1925). In the state, chestnut was most abundant on ridges, where it grew with oak (primarily scarlet, black, white, and chestnut), hickory, and pine (Heffner 1939; Sears 1925), though witness tree data suggests the species was likely a minor component of Ohio’s pre-Euro-American settlement forests (Dyer and Hutchinson 2019). Since the arrival of blight in the late 19th century, most mature chestnuts have been killed and the species persists primarily as understory sprouts originating from blight- killed trees and saplings. Chestnut sprouts occur sporadically in the Wayne National Forest, but have not been mapped.

The American Chestnut Foundation has pursued the development of a blight-tolerant hybrid American chestnut since the early 1980s (Hebard 2006; Steiner et al. 2017). Their main approach has been to introgress alleles that confer blight tolerance from Chinese chestnut (C. mollissima) into American chestnut, followed by repeated backcrossing to American chestnut (Steiner et al. 2017). In anticipation of the availability of blight-tolerant hybrid chestnuts, multiple studies have evaluated silvicultural methods for reintroducing the species to managed forests (Clark et al. 2012; McCament and McCarthy 2005; Rhoades et al. 2009; Pinchot et al. 2017), with a particular focus on the restoration of chestnut to National Forest System lands (Clark et al. 2014). Silvicultural prescriptions to facilitate the establishment of chestnut founder populations will likely vary regionally, with differing ecological conditions and management objectives and strategies. Appalachian-Alleghenian Sandstone Dry Cliff Vegetation7 This outcrop community occurs on steep to vertical exposures of sandstone bedrock and rockhouses. Plants are mostly restricted to rock shelves, cracks, and crevices. Mosses and lichens are common inhabitants, but sparse woody and herbaceous species are present. Examples of this habitat have been noted at Lick Branch, Minnow Hollow, Waterfall Cove, and Young’s Branch Special Interest Areas. Northern Mixed Mesophytic Forest8 This mesic habitat occurs in narrow valley positions, ravines, and headwaters. The tree canopy is closed and contains a variety of tree species including sugar maple, American beech, ash, tuliptree, black cherry, white oak, and northern red oak. Trees indicative of this community type include yellow buckeye and American basswood, with cucumber tree (Magnolia acuminata (L.) L.) occurring locally. Examples of this habitat are represented by Caulley Creek, Eels Run, Fly Gorge, Little Storms Creek, Minnow Hollow, Morgan Sisters Woods, Rockcamp Run, Rocky Fork Gorge, Waterfall Cover, and Young’s Branch Special Interest Area, and by Reas Run Research Natural Area.

7 Sneddon and Gawler 2010 8 Faber-Langendoen et al. 2001

50 | Page

Terrestrial Ecosystems Supplemental Report

Interior Low Plateau Beech-Maple Forest9 While vegetationally very similar to the Northern Mixed Mesophytic Forest, this habitat type is clearly dominated by American beech and sugar maple. It occurs on unglaciated alluvial terraces and mesic slopes on maturely dissected plateaus with moderately well-drained, moist, rich, deep soils. A regionally significant example of this habitat can be found at Deadhorse Run Special Interest Area. Interior Plateaus Hemlock-Hardwood Forest10 This habitat occurs on dry-mesic to mesic valley slopes. Soils are typically acidic due to the sandstone or shale parent material. In southeast Ohio, eastern hemlock (Tsuga canadensis (L.) Carrière) is at the western part of its range, and is confined to cool valleys, north and east slopes, and ravines. Dominant canopy species include eastern hemlock, sugar maple, and American beech. Other canopy associates include white oak, northern red oak, and hickory. The understory is composed of various shrubs such as American witch-hazel, maple-leaf viburnum, and mountain laurel. Examples of this habitat are represented at Dismal Creek, Rockcamp Run, Rocky Fork Gorge, and Witten Run Special Interest Areas. River Birch-Sycamore Small River Floodplain Forest11 This floodplain habitat occurs along streams and small riverbanks, and in swampy forests. The sandy alluvial soils are deep, moist, and well-drained. River birch (Betula nigra L.) and American sycamore (Platanus occidentalis L.) are the typical dominants, but canopy associates include species from adjacent bottomland and toeslope mesic communities (oak, elm, ash, maple). Examples of this habitat can be found at Lick Branch and Sardis Wetlands Special Interest Areas. Skunk-cabbage Seepage Meadow12 These wetlands develop around spring heads and broader areas of groundwater discharge resulting in a shallow peat layer. Skunk cabbage (Symplocarpus foetidus (L.) Salisb. ex W.P.C. Barton) is the dominant species, but other forbs and ferns can be present. Minnow Hollow Special Interest Area supports an example of this habitat type. Pin Oak Mixed Hardwood Depression Forest13 This wet forest habitat occurs on poorly drained depressions with a canopy dominated by pin oak (Quercus ellipsoidalis E.J. Hill) and red maple, but also contains at least 25% other oak species. Some occurrences of this forest community type in Ohio are associated with clay deposits left behind by the glacial Teays River across Jackson and Gallia counties. The communities in Ohio

9 Guetersloh and Faber-Langendoen 2001 10 Faber-Langendoen 2018 11 Landaal and Gawler 2008 12 Faber-Langendoen and Sneddon 2013 13 Faber-Langendoen and Sneddon 2001

51 | Page

Terrestrial Ecosystems Supplemental Report

have more river birch and sugar maple. Examples include Paynes Crossing and Sardis Wetlands Special Interest Areas. Midwest Buttonbush Shrub Swamp14 These swamps develop in shallow water depressions, oxbow ponds, and backwater sloughs of floodplains from the Teays River outwash. Common buttonbush (Cephalanthus occidentalis L.) is the dominant species, comprising as much as 90% of the well-developed shrub layer. While this habitat type can develop from anthropogenic influences (e.g., impoundments or other hydrologic alterations), Lick Branch Special Interest Area supports an example of a naturally- developed buttonbush shrub swamp maintained by beavers. Reclaimed Grasslands This habitat constitutes a novel habitat type and is the result of the reclamation of abandoned coalmine lands. The soils are acidic and nutrient-poor, resulting in little potential for forest development. Grasses and leguminous forbs have been planted to facilitate a grassland community able to tolerate the site conditions. These habitats have proven useful to a variety of grassland-dependent birds that have experienced population declines as a result of conversion of their native habitats. Climate Change So what does this all mean in the face of changing climate? Understanding potential impacts is an important first step to sustaining healthy forests in the face of changing conditions. An assessment evaluating the vulnerability of forest ecosystems in the Central Appalachians of Ohio, West Virginia, and Maryland was completed for a range of future climates (Butler et al. 2015), and is summarized below.

Indeed the climate is changing. Since the turn of the last century (1901 to 2011), daily low temperatures have changed more than daily high or average temperatures. Daily lows have warmed the most during summer and fall months. Both daily highs and lows increased in April and November. Extremely hot days have become more frequent, while extremely cold days have decreased. Daily low temperatures increased by 1o F annually, 1.6o F in summer and 1.4o F in fall. Daily high temperatures increased by 3.2o F in April and decreased by 2o F in September and October. With warmer temperatures in April and November, the growing season length has increased, resulting in observed changes in the timing of biological activity such as plant phenology and bird migration. The Central Appalachians is receiving 8 percent more precipitation, particularly in the fall. Extreme rain events of 3 inches or greater have become more frequent, while light rain events have decreased.

Downscaled global climate models can help illustrate how climate may change in the future given changes in greenhouse gas emissions. In this assessment, climate projections for two climate models are reported under two contrasting greenhouse gas emissions scenarios over the

14 Faber-Langendoen 1998

52 | Page

Terrestrial Ecosystems Supplemental Report

next century compared to the average over the last 30 years of the 20th century. Results show that temperatures will increase. All global climate models project that temperatures will increase in the Central Appalachians. Models project an increase in temperature over the next century across all seasons by 2 to 9o F. Studies from across the Midwest and Northeast consistently project 20 to 30 more hot days per year by the end of the century, and the frequency of multi-day heat waves is also projected to increase by 3 to 6 days in southeastern Ohio.

Precipitation will continue to change as well. Precipitation is projected to increase in winter and spring by 1 to 2 inches for the two seasons combined. There is a difference in model projections for later in the growing season, but evidence seems to indicate there may be a decrease in precipitation in either summer or fall, depending on scenario. Even if the total annual amount of precipitation does not change substantially, some evidence suggests it may occur as heavier rain events interspersed among relatively drier periods.

Forests will experience both direct and indirect impacts from a changing climate. Two global climate models, three forest impact models, hundreds of scientific papers, and forest managers’ expertise were combined to assess the effects of climate change on regional forest ecosystems. Based on this information, the following impacts will occur in the Central Appalachians and are described here.

Soil moisture patterns will change, with drier soil conditions in summer and fall. Due to potential decreases in summer and fall precipitation and increases in winter and spring precipitation, it is likely that soil moisture regimes will also shift. Longer growing seasons and warmer temperatures may also result in greater evapo-transpiration and lower soil-water availability later in the growing season. Oak is generally a warm-adapted genus (Nowacki and Abrams 2015), and its historic increase to dominance across the eastern United States during the early Holocene Warming into the Holocene Thermal Maximum (warmer, drier conditions) suggests oak- dominated forest systems will be able to respond well to a warming climate. Oaks are adapted to drought conditions with their thicker leaves and deeper root systems, and can continue photosynthesis when other tree species are not able.

National and global studies agree that wildfire risk will increase in the region. Fire is expected to accelerate changes in forest composition, promoting changes in species faster than temperature or moisture availability. Fire is documented as a landscape-level disturbance in southeast Ohio in both prehistoric and historic times, and it influenced the development of the oak-dominated forests that persist on the landscape today (Iverson et al. 2019a). Increased fire would bode well for the recovery of oak-dominated ecosystems in the future, although a tipping point might have been reached whereby the heavy presence of mesophytic species have changed environmental conditions (shaded, cool, moist understories; wet, compact litter beds unreceptive to fire) to such a degree as to possibly temper the occurrence of future fire (Nowacki and Abrams 2008).

Tree species’ responses will be driven by local conditions. Tree species’ responses to two climate change scenarios at the end of the century are presented below for three forest impact models for species in southeast Ohio. The Climate Change Tree Atlas modeled changes in suitable habitat for 72 species. LINKAGES predicted establishment probability and growth for 21 species. LANDIS simulated changes in productivity for 17 tree species. A complete set of model results

53 | Page

Terrestrial Ecosystems Supplemental Report

can be found in Appendix 4 of the Central Appalachians Forest Ecosystem Vulnerability Assessment and Synthesis (Butler et al. 2015). Model results project an increase in suitable habitat and growth for many species with ranges largely south of the region, including shortleaf pine, post oak, and blackjack oak. Oaks as a group are projected to have an increase in suitable habitat and volume (basal area). Model results project a decrease in habitat suitability for northern species such as eastern hemlock, which is currently limited to specific landscape positions where conditions are cool and moist. Where microhabitats remain cool and moist, this species may continue to find refugia, but its presence on the landscape may become rare. For other northern species, like sugar maple, populations may be able to persist in southern refugia if new competitors from the south are unable to colonize or other species with more suitable habitat do not out-compete.

All three models show that the range of possible future climates will likely benefit tuliptree, white oak, and loblolly pine (Pinus taeda L.) (currently a plantation species). For these species, model results suggest that establishment success, productivity, and suitable habitat will increase. All three models show that the range of possible future climates will likely stress American beech, eastern hemlock, eastern white pine (Pinus strobus L.) (plantation species), northern red oak, scarlet oak, and sugar maple. For these species, model results suggest that establishment success and productivity will face greater risk of decreasing with higher temperatures, and that there will be less suitable habitat available.

Six species were modeled by LINKAGES and Tree Atlas, but not LANDIS. American elm, black cherry, and flowering dogwood are expected to lose suitable habitat, but not establishment probability, suggesting that current marginal habitat may disappear, but that the species will do well where suitable habitat remains. Blackgum is expected to increase moderately, while post oak and shortleaf pine are expected to increase. Fifty-one tree species were modeled by only the Tree Atlas and a subset of results is presented here. Habitat decreases under both climate scenarios for the following species: bigtooth (Populus grandidentata Michx.) and quaking aspens (Populus tremuloides Michx.), black locust (Robinia pseudoacacia L.), red pine (Pinus resinosa Aiton) (plantation species), silver maple, and sweet birch (Betula lenta L.). Habitat increases under both climate scenarios for the following species: bitternut hickory, black, blackjack and chinquapin oaks, cucumber tree, eastern redbud (Cercis canadensis L.), green ash (Fraxinus pennsylvanica Marshall), common hackberry (Celtis occidentalis L.), shagbark hickory, sourwood (Oxydendrum arboreum (L.) DC.), sweetgum (Liquidambar styraciflua L.), and Virginia pine.

Models are unable to include every threat and interaction that may cause a species to do better or worse than predicted, and threats may change over time. Invasive plants, pests, and pathogens are expected to increase or become more damaging, especially when trees are stressed by other disturbances, heat, or drought. A warming climate is allowing some invasive plant species, insect pests, and pathogens to survive further north than they have previously. Threats such as the southern pine beetle, oak decline, and many invasive species, such as tree of heaven and bush honeysuckles, may increase in the future. Introductions of nonnative species from other ecosystems are expected to occur again, and the next threat is currently unknown.

54 | Page

Terrestrial Ecosystems Supplemental Report

Using the information from the Central Appalachians Forest Ecosystem Vulnerability Assessment and Synthesis (Butler et al. 2015) summarized above, a Forest Adaptation Planning Workshop for oak forests was conducted as part of the USDA Joint Chiefs’ 2015-2017 project in Ohio. Some potential common climate change impacts on oak forests, both positive and negative are listed below.

Positive impacts include:

• Shade-tolerant tree species are not expected to compete as well on drier “oak” sites. • Many oak, hickory, and pine species are expected to fare well or better under the range of potential future climates. Southern red oak occurs in the southernmost counties of Ohio and is projected to gain new suitable habitat in southern Ohio under both climate scenarios. • Some exceptions include chestnut oak, which is predicted to do well under the low change scenario but may decline under the high scenario due to predicted problems in regeneration. Northern red oak is currently not regenerating well and is expected to lose species establishment probability. Negative impacts include:

• Increases in extreme weather (e.g., wind, ice, rain) can create more canopy gaps that could release mesophytic understory trees, thus accelerating succession (Abrams and Nowacki 1992). Also might benefit nonnative species that often outcompete native species. • Droughty conditions are likely to result in higher susceptibility of oaks to certain pests and pathogens, collectively referred to as oak decline. • Reduced soil moisture in summer and fall are expected to shift burn windows for prescribed fire, increase dry fuels, and increase fire severity, especially where there is an abundance of fuels. • Increased heavy rain events may result in increased erosion and runoff on steep slopes. • Destabilization of soils may contribute to increased windthrow. • Increased precipitation in winter and spring may hinder silvicultural operations that encourage oak regeneration and restoration. Climate change impacts will vary across southeast Ohio, depending on site exposure and adaptability. Examples of characteristics that make systems more adaptable to climate changes include high species diversity, topographic variability, landscape connectivity, and ecological integrity. The high diversity of landforms and microclimates in southeastern Ohio may make it more resilient to change by allowing species to persist in pockets of refugia (persistent suitable habitat). Southeastern Ohio is also rich in forest types, most of which are classified as oak- dominated. Forest systems have a rich diversity of tree species with more than 90 species of trees (Prasad et al. 2007). Tree species richness, a high degree of tree canopy co-dominance, and oak- dominated forests are key attributes that can help forest systems in southeast Ohio adapt well to future climate change. Management practices, human activity, natural disturbances, and forest

55 | Page

Terrestrial Ecosystems Supplemental Report

succession currently influence forest growth and productivity and will continue to affect forests even as effects of climate change become amplified through the end of the century. Summary Oak-dominated ecosystems have been the hallmark of southeastern Ohio, being sustained by frequent, low-intensity surface fires (Abrams 2002). These systems were acutely affected by European disturbances starting in the late 1700s, with intensifying and overlapping waves of disturbance that included industrial logging, burning, mining, land clearing, and agricultural conversion (Sutherland 1997; Stambaugh et al. 2018). By the time the Forest Service began acquiring lands to form the Wayne National Forest (1930s; Conrad 1997), the land was akin to a moonscape—a complex mosaic of excessively cutover, burned, mined, and abandoned agricultural fields and pasturelands. Overtime, mainly through passive management, forests were allowed to redevelop on National Forest System lands (Radcliffe 2019).

Oaks continue to dominate today (Albright et al. 2018), perhaps even bolstered by recent past disturbances (loss of American chestnut by blight and pine through excessive cutting/burning at the turn of the 20th Century; Nowacki and Abrams 1992). However, the system now operates under a vastly different set of disturbance conditions. Indirectly, through long-term fire suppression, natural succession processes have initiated and steered these systems toward shade- tolerant, fire-sensitive mesophytic trees (Hutchinson et al. 2008). In a cascading fashion, former open systems converted to closed-canopy, second-growth forests with dense mesophytic (maple, beech) understories that cast a deep shade preventing oaks to adequately regenerate and recruit (Palus et al. 2018, Radcliffe 2019). Excessive shading has had a devastating effect on understory diversity, all but eliminating most heliophytic (sun loving) ground flora associated with former open oak systems. Ironically, this depauperate condition missing the rich array of pyrogenic plants has been casted as an indicator of the lack of fire in the distant past (Matlack 2013). Although current forests have a fair degree of connectedness, it is hard to interpret what that really means as the compositions (increased mesophyte representation), structures (closed, multi- tiered canopies), and disturbance processes (little to no fire) have greatly departed from historic conditions. Surely, plant and animal species that favor mesophytic trees and stand conditions have benefitted at the expense of those that evolved and flourished within the prominent open oak ecosystems of the past. Without combined silvicultural treatments of thinning and burning (Vander Yacht et al. 2017, 2018; Schweitzer et al. 2019), this trend will only escalate, sapping the oak resource and its many dependent wildlife species (McShea and Healy 2002). Southeast Ohio is not alone in this regard, merely a microcosm of what is happening across the eastern United States (Nowacki and Abrams 2008, 2015). The future is somewhat uncertain as a changing climate may further alter successional trajectories through drought and enhanced fire-weather conditions (Iverson et al. 2019b).

56 | Page

Terrestrial Ecosystems Supplemental Report

Glossary Autoecological: Physiological traits of a species that helps them live in a certain environment.

Connectedness: Measure of the resistance (or lack of) to successful dispersal based on anthropogenic changes in land use. For example, highly urban areas have high resistance to movement/low connectedness.

Fire-tolerant/-adapted: Plants possessing traits that allow them to survive repeated surface burning, such as thick back, high rot resistance and rapid wound closure after bole injury, and the ability to resprout if top-killed.

Fire-dependent: Plants that require fire for their life cycle.

Forest: Closed-canopy treed communities with shaded, light limited understories.

Heliophytic: Plants requiring high light conditions for their survival; sun-loving plants.

Mesophytic/mesophyte: Plants preferring moist (mesic), often fertile, environments.

Old-growth forest: Forests in later developmental stages relative to their life cycle and community type; see Nowacki and Trianosky (1993) and Tyrrell et al. (1998) for specifics.

Pyrogenic: Trees or systems that have evolved in a fire environment.

Pyrophilic/pyrophyte: Plants that proliferate in a fire environment through fire adaptation and/or tolerance.

Pyrophobic: Plants that do not do well in a fire environment; fire avoiders.

Succession classes: Different developmental stages that occurs overtime resulting from compositional and structural changes.

Synecological: The relationship of various groups of organisms to their common environment.

Woodland: Open treed communities that allow light penetration down to the forest floor (high light conditions) fostering an abundance of ground cover (shrubs, grasses, forbs).

Xerophytic: Plants that proliferate in dry (xeric), often infertile, environments.

57 | Page

Terrestrial Ecosystems Supplemental Report

References Abrams MD. 1992. Fire and the development of oak forests. Bioscience. 42(5):346-353.

Abrams MD, Nowacki GJ. 1992. Historical variation in fire, oak recruitment, and post-logging accelerated succession in central Pennsylvania. Bulletin of the Torrey Botanical Club. 119(1):19-28.

Abrams MD. 1996. Distribution, historical development and ecophysiological attributes of oak species in the eastern United States. Annals of Forest Science. 53(2-3):487-512.

Abrams MD. 2002. The postglacial history of oak forests in eastern North America. In: McShea W, Healy W, editors. Oak forest ecosystems: ecology and management for wildlife. Baltimore (MD): Johns Hopkins University Press. Pg. 34-45.

Abrams MD, Nowacki GJ. 2008. Native Americans as active and passive promoters of mast and fruit trees in the eastern USA. The Holocene. 18(7):1123-1137.

Abrams EM, Freter A, Stefanova V. 2014. Environmental change since the Woodland Period in the mid-Ohio Valley: Results from Patton Bog sediment core palynological analyses. Midcontinental Journal of Archaeology. 39(2):163-178.

Abrams MD, Nowacki GJ. 2015. Exploring the Early Anthropocene burning hypothesis and climate-fire anomalies for the eastern U.S. Journal of Sustainable Forestry. 34(1-2):30-48.

Albright TA, Butler BJ, Crocker SJ, Kurtz CM, Lister TW, McWilliams WH, Miles PD, Morin RS, Nelson MD, Riemann R, Smith JE. 2018. Ohio Forests 2016. Newtown Square (PA): USDA Forest Service, Northern Research Station. NRS-RB-118.

Anagnostakis SL. 1987. Chestnut blight: the classical problem of an introduced pathogen. Mycologia. 79(1):23-37.

Anderson MG, Clark M, Sheldon AO. 2014. Estimating climate resilience for conservation across geophysical settings. Conservation Biology. 28(4):959-970.

Anderson MG, Barnett A, Clark M, Ferree C, Sheldon AO, Prince J. 2016. Resilient Sites for Terrestrial Conservation in Eastern North America. Boston (MA): The Nature Conservancy, Eastern Conservation Science.

Apsley D, McCarthy BC. 2004. White-tailed deer herbivory on forest regeneration following fire and thinning treatments in southern Ohio mixed oak forests. In: Yaussy DA, Hix DM, Long RP, Goebel PC, editors. Proceedings, 14th central hardwood forest conference. Newtown Square (PA): USDA Forest Service, Northeastern Research Station. NE-GTR- 316

Aubin I, Gachet S, Messier C, Bouchard A. 2007. How resilient are northern hardwood forests to human disturbance? An evaluation using a plant functional group approach. Ecoscience. 14(2):259-271.

58 | Page

Terrestrial Ecosystems Supplemental Report

Barker J. 1958. Recollections of the first settlement of Ohio. [First published 1790, Blazier GJ, editor]. Marietta (OH): Marietta College.

Bird BW, Wilson JJ, Gilhooly WP III, Steinman BA, Stamps L. 2017. Midcontinental Native American population dynamics and late Holocene hydroclimate extremes. Scientific Reports. 7:41628.

Braun LE. 1950. Deciduous Forests of Eastern North America. London (United Kingdom): The Free Press, A Division of Macmillan Publishing Co, Inc. New York and Collier Macmillan Publishers.

Braun LE. 1961. The woody plants of Ohio. Columbus (OH): Ohio State University Press.

Brose P, Van Lear D. 2004. Survival of hardwood regeneration during prescribed fires: the importance of root development and root collar location. In: Spetich M, editor. Upland oak ecology symposium: history, current conditions, and sustainability. Asheville (NC): USDA Forest Service, Southern Research Station. Pg. 123-127.

Brose PH, Dey DC, Phillips RJ, Waldrop TA. 2013. A meta-analysis of the fire-oak hypothesis: Does prescribed burning promote oak reproduction in eastern North America? Forest Science. 59(3):322-334.

Butler PR, Iverson L, Thompson FR, Brandt L, Handler S, Janowiak M, Shannon PD, Swanston C, Karriker K, Bartig J, et al. 2015. Central Appalachians Forest Ecosystem Vulnerability Assessment and Synthesis: A Report from the Central Appalachians Climate Change Response Framework Project. USDA Forest Service, Northern Research Station. NRS- GTR-146.

Camp MJ. 2006. Roadside geology of Ohio. Missoula (MT): Mountain Press Publishing Company.

Clark S, McNab H, Loftis D, Zarnoch S. 2012. American chestnut growth and survival five years after planting in two silvicultural treatments in the southern Appalachians, USA. Forests. 3(4):1017-1033.

Clark SL, Schlarbaum SE, Pinchot CC, Anagnostakis SL, Saunders MR, Thomas-Van Gundy M, Schaberg PG, McKenna J, Bard J, Berrang P, et al. 2014. Reintroduction of American chestnut in the national forest system. Journal of Forestry. 112(5):502-512.

Cleland DT, Avers PE, McNab WH, Jensen ME, Bailey RG, King T, Russell WE. 1997. National Hierarchical Framework of Ecological Units. In: Boyce MS, Haney A, editors. Ecosystem management: applications for sustainable forest and wildlife resources. New Haven (CT): Yale University Press. Pg. 181-200.

Clements FE. 1916. Plant succession: An analysis of the development of vegetation. Washington D.C. (United States): Carnegie Institute Publication.

Clements FE. 1936. Nature and structure of the climax. Journal of Ecology. 24:252-284.

59 | Page

Terrestrial Ecosystems Supplemental Report

Conrad DE. 1997. The land we cared for: A history of the Forest Service’s Eastern Region. Carbondale (IL): American Resources Group, Ltd.

Cook JE. 1996. Implications of modern successional theory for habitat typing: A review. Forest Science. 42(1):67-75.

Crow TR, Perera AH. 2004. Emulating natural landscape disturbance in forest management-an introduction. Landscape Ecology. 19(3):231-233.

Davis MB. 1983. Quaternary history of deciduous forests of eastern North America and Europe. Annals of the Missouri Botanical Garden. 1:550-63.

Drury SA, Runkle JR. 2006. Forest vegetation change in southeast Ohio: Do older forests serve as useful models for predicting the successional trajectory of future forests? Forest Ecology and Management. 223(1-3):200-210.

Dyer JM. 2001. Using witness trees to assess forest change in southeastern Ohio. Canadian Journal of Forest Research. 31(10):1708-1718.

Dyer JM, Hutchinson T. 2019. Topography and soils-based mapping reveals fine-scale compositional shifts over two centuries within a central Appalachian landscape. Forest Ecology and Management. 433:33-42.

Egan D, Howell EA. 2001. The historical ecology handbook: A restorationist’s guide to reference ecosystems. Washington D.C. (United States): Island Press.

Elliott KJ, Miniat CF, Pederson N, Laseter SH. 2015. Forest tree growth response to hydroclimate variability in the southern Appalachians. Global Change Biology. 21(12):4627-4641.

Engstrom RT, Gilbert S, Hunter ML Jr, Merriwether D, Nowacki GJ, Spencer P. 1999. Practical applications of disturbance ecology to natural resource management. In: Szaro RC, Johnson NC, Sexton WT, Malk AJ, editors. Ecological stewardship: A common reference for ecosystem management, Volume II. Oxford (United Kingdom): Elsevier Science, Ltd.

Faber-Langendoen, D. 1998. Midwest Buttonbush Shrub Swamp. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Faber-Langendoen, D. 2018. Interior Plateaus Hemlock-Hardwood Forest. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Faber-Langendoen, Menard S, and Pyne M. 2002. Appalachian Chestnut Oak-Mixed Oak Forest. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

60 | Page

Terrestrial Ecosystems Supplemental Report

Faber-Langendoen D, Sneddon L. 2001. Pin Oak Mixed Hardwood Depression Forest. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Faber-Langendoen D, Sneddon LA. 2013. Skunk-cabbage Seepage Meadow. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Faber-Langendoen D, Sneddon L, Pyne M, White R, Gawler SC. 2001. Northern Mixed Mesophytic Forest. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Fahey RT, Fotis AT, Woods KD. 2015. Quantifying canopy complexity and effects on productivity and resilience in late‐successional hemlock-hardwood forests. Ecological Applications. 25(3):834-847.

Federal Geographic Data Committee [FGDC]. 2019. United States National Vegetation Classification Database. Washington D.C. (United States): Federal Geographic Data Committee, Vegetation Subcommittee.

Gill JL, Williams JW, Jackson ST, Donnelly JP, Schellinger GC. 2012. Climatic and megaherbivory controls on late-glacial vegetation dynamics: a new, high-resolution, multi-proxy record from Silver Lake, Ohio. Quaternary Science Reviews. 34:66-80.

Gill JL. 2013. Ecological impacts of the late Quaternary megaherbivore extinctions. New Phytologist. 201(4):1163-1169.

Gordon RB. 1969. The natural vegetation of Ohio in pioneer days. Bulletin of the Ohio Biological Survey. 3(2):1-113.

Gregory, SV, FJ Swanson, McKee A, Cummins KW. 1991. An ecosystem perspective of riparian zones. Bioscience. 41:540-551.

Guetersloh, M, Faber-Langendoen D. 2001. Interior Low Plateau Beech-Maple Forest. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Guyette RP, Dey DC, Stambaugh MC. 2003. Fire and human history of a barren-forest mosaic in southern Indiana. The American Midland Naturalist. 149(1):21-34.

Guyette RP, Dey DC, Stambaugh MC, Muzika RM. 2006. Fire scars reveal variability and dynamics of eastern fire regimes. In: Dickinson MB, editor. Fire in eastern oak forests: delivering science to land managers, proceedings of a conference. Newtown Square (PA): USDA Forest Service, Northern Research Station. NRS-GTR-P-1.

Guyette RP, Stambaugh MC, Dey DC, Muzika RM. 2012. Predicting fire frequency with chemistry and climate. Ecosystems. 15(2):322-335.

61 | Page

Terrestrial Ecosystems Supplemental Report

Hart J, Buchanan M. 2012. History of fire in eastern oak forests and implications for restoration. In: Dey DC, Stambaugh MC, Clark SL, Schweitzer CJ, editors. Proceedings of the 4th fire in eastern oak forests conference. Newtown Square (PA): USDA Forest Service, Northern Research Station. NRS-GTR-102.

Hayhoe K, Wuebbles DJ, Easterling DR, Fahey DW, Doherty S, Kossin J, Sweet W, Vose R, Wehner M. 2018. Our changing climate. In: Reidmiller DR, Avery CW, Easterling DR, Kunkel KE, Lewis KLM, Maycock TK, Stewart BC, editors. Impacts, risks, and adaptation in the United States: Fourth National Climate Assessment, Vol. II. Washington D.C. (United States): U.S. Global Change Research Program.

Hebard FV. 2006. The backcross breeding program of the American Chestnut Foundation. Journal of the American Chestnut Foundation. 19:55-77.

Heffner GA. 1939. Vegetation survey of an area in central Ohio at the edge of the Alleghany Plateau [Doctoral dissertation]. Columbus (OH): Ohio State University.

Hutchinson TF, Long RP, Ford R, Sutherland EK. 2008. Fire history and the establishment of oaks and maples in second-growth forests. Canadian Journal of Forest Research. 38(5):1184-1196.

Hutchinson TF, Stambaugh MC, Marschall JM, Guyette RP. 2019. Historical fire in the Appalachian Plateau of Ohio and Kentucky, USA, from remnant yellow pines. Fire Ecology. 15(1):33.

Iverson LR, Hutchinson TF, Peters MP, Yaussy DA. 2017. Long-term response of oak-hickory regeneration to partial harvest and repeated fires: influence of light and moisture. Ecosphere. 8(1):e01642.

Iverson LR, Peters MP, Bartig JL, Rebbeck J, Hutchinson TF, Matthews SN, Stout S. 2018. Spatial modeling and inventories for prioritizing investment into oak-hickory restoration. Forest Ecology and Management. 424:355-366.

Iverson LR, Bartig JL, Nowacki GJ, Peters MP, Dyer JM, Hutchinson TF, Matthews SN, Adams BT. 2019a. USDA Forest Service Section Subsection, and Landtype Descriptions for Southeastern Ohio. Madison (WI): USDA Forest Service, Northern Research Station. NRS-RM-10.

Iverson LR, Peters MP, Matthews SN, Prasad A, Hutchinson T, Bartig J, Rebbeck J, Yaussy D, Stout S, Nowacki GJ. 2019b. Adapting oak management in an age of ongoing mesophication but warming climate. In: Clark SL, Schweitzer CJ, editors. Oak symposium: Sustaining oak forests in the 21st Century through science-based management. Asheville (NC): USDA Forest Service, Southern Research Station. SRS- GTR-237.

Jones CH. 1945. Vegetation of the Adams-Scioto quadrangle. Columbus (OH): Ohio State University.

62 | Page

Terrestrial Ecosystems Supplemental Report

Lafon CW, Naito A, Grissino-Mayer HD, Horn S, Waldrop TA. 2017. Fire history of the Appalachian Region: a review and synthesis. Asheville (NC): USDA Forest Service, Southern Research Station. SRS-GTR-219.

Landaal, S, Gawler SC. 2008. River Birch-Sycamore Small River Floodplain Forest. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Leete B. 1938. Forest fires in Ohio 1923 to 1935. Wooster (OH): Ohio Agricultural Experiment Station. Bulletin 598.

LANDFIRE. 2019a. Biophysical Setting Model 6213170 Allegheny-Cumberland Dry Oak Forest and Woodland. In: LANDFIRE National Vegetation Dynamics Model. Washington D.C. (United States): USDA Forest Service and U.S. Department of Interior.

LANDFIRE. 2019b. Biophysical Setting Model 6213100 North-Central Interior Dry-Mesic Oak Forest and Woodland. In: LANDFIRE National Vegetation Dynamics Model. Washington D.C. (United States): USDA Forest Service and U.S. Department of Interior.

Lepper BT. 2005. Ohio archaeology: an illustrated chronicle of Ohio’s ancient American Indian cultures. Wilmington (OH): Orange Frazer Press.

Leue A. 1886. First annual report of the Ohio State Forestry Bureau. Columbus (OH): Ohio Agricultural Experiment Station.

Long JN. 2009. Emulating natural disturbance regimes as a basis for forest management: a North American view. Forest Ecology and Management. 257(9):1868-1873.

Lorenz K, Lal R. 2007. Stabilization of organic carbon in chemically separated pools in reclaimed coal mine soils in Ohio. Geoderma. 141(3-4):294-301.

Lorimer CG, White AS. 2003. Scale and frequency of natural disturbances in the northeastern US: implications for early successional forest habitats and regional age distributions. Forest Ecology and Management. 185(1-2):41-64.

Matlack GR. 2013. Reassessment of the use of fire as a management tool in deciduous forests of eastern North America. Conservation Biology. 27(5):916-926.

McCament CL, McCarthy BC. 2005. Two-year response of American chestnut (Castanea dentata) seedlings to shelterwood harvesting and fire in a mixed-oak forest ecosystem. Canadian Journal of Forest Research. 35(3):740-749.

McEwan RW, Dyer JM, Pederson N. 2011. Multiple interacting ecosystem drivers: toward an encompassing hypothesis of oak forest dynamics across eastern North America. Ecography. 34(2):244-256.

63 | Page

Terrestrial Ecosystems Supplemental Report

McEwan RW, Hutchinson TF, Long RP, Ford DR, McCarthy BC. 2007. Temporal and spatial patterns in fire occurrence during the establishment of mixed-oak forests in eastern North America. Journal of Vegetation Science. 18(5):655-664.

McRae BH, Hall SA, Beier P, Theobald DM, 2012. Where to restore ecological connectivity? Detecting barriers and quantifying restoration benefits. PloS One. 7(12):e52604.

McShea WJ, Healy WM. 2002. Oak forest ecosystems: Ecology and management for wildlife. Baltimore (MD): Johns Hopkins University Press.

Mishra SK, Hitzhusen FJ, Sohngen BL, Guldmann JM. 2012. Costs of abandoned coal mine reclamation and associated recreation benefits in Ohio. Journal of Environmental Management. 100(suppl. C):52-58.

Mori AS. 2011. Ecosystem management based on natural disturbances: hierarchical context and non‐equilibrium paradigm. Journal of Applied Ecology. 48(2):280-292.

Nowacki GJ, Abrams MD. 1992. Community, edaphic, and historical analysis of mixed oak forests of the Ridge and Valley Province in central Pennsylvania. Canadian Journal of Forest Research. 22(6):790-800.

Nowacki GJ, Trianosky P.A. 1993. Literature on old-growth forests of eastern North America. Natural Areas Journal. 13(2):87-107.

Nowacki GJ, Abrams MD. 2008. The demise of fire and “mesophication” of forests in the eastern United States. Bioscience. 58(2):123-138.

Nowacki GJ, Abrams MD. 2015. Is climate an important driver of post-European vegetation change in the eastern United States? Global Change Biology. 21(1):314-334.

Pahl-Wostl C. 1995. The dynamic nature of ecosystems: Chaos and order entwined. New York City (NY): John Wiley & Sons.

Palus JD, Goebel PC, Hix DM, Matthews SN. 2018. Structural and compositional shifts in forests undergoing mesophication in the Wayne National Forest, southeastern Ohio. Forest Ecology and Management. 430:413-420.

Patterson K, White R, Gawler SC. 2001. Appalachian Low-Elevation Mixed Pine/Blue Ridge Blueberry Forest. Washington D.C. (United States): Federal Geographic Data Committee.

Patton PE, Curran S. 2016. Archaic Period domesticated plants in the Mid-Ohio Valley: archaeobotanical remains from the County Home Site (33at40), southeastern Ohio. Midcontinental Journal of Archaeology. 41(2):127-158.

Pielou E. 1991. After the ice age: the return of life to glaciated North America. Chicago (IL): University of Chicago Press.

64 | Page

Terrestrial Ecosystems Supplemental Report

Pinchot CC, Schlarbaum SE, Clark SL, Saxton AM, Sharp AM, Schweitzer CJ, Hebard FV. 2017. Growth, survival, and competitive ability of chestnut (Castanea Mill.) seedlings planted across a gradient of light levels. New Forests. 48(4):491-512.

Prasad AM, Iverson LR, Matthews S, Peters M. 2007-ongoing. A climate change atlas for 134 forest tree species of the eastern United States [database, updated at least annually]. Delaware (OH): USDA Forest Service, Northern Research Station. https://www.fs.fed.us/nrs/atlas/.

Radcliffe DC. 2019. Topographic, edaphic, and stand structural factors associated with oak and hickory mortality and maple and beech regeneration in mature forests of Appalachian Ohio [Doctoral dissertation]. Columbus (OH): Ohio State University.

Radcliffe DC, Matthews SN, Hix DM. 2020. Beyond oak regeneration: modelling mesophytic sapling density drivers along topographic, edaphic, and stand-structural gradients in mature oak-dominated forests of Appalachian Ohio. Canadian Journal of Forest Research [Online article, not in print yet].

Richter BD, Richter HE. 2000. Prescribing flood regimes to sustain riparian ecosystems along meandering rivers. Conservation Biology 14:1467-1478.

Rhoades C, Loftis D, Lewis J, Clark S. 2009. The influence of silvicultural treatments and site conditions on American chestnut (Castanea dentata) seedling establishment in eastern Kentucky, USA. Forest Ecology and Management. 258(7):1211-1218.

Rollins MG. 2009. LANDFIRE: a nationally consistent vegetation, wildland fire, and fuel assessment. International Journal of Wildland Fire. 18(3):235-249.

Russell EW. 1987. Pre-blight distribution of Castanea dentata (Marsh.) Borkh. Bulletin of the Torrey Botanical Club. 1:183-90.

Schweitzer CJ, Dey DC, Wang Y. 2019. White Oak (Quercus alba) Response to Thinning and Prescribed Fire in Northcentral Alabama Mixed Pine-Hardwood Forests. Forest Science. 65(6):758-766.

Sears PB. 1925. The natural vegetation of Ohio. I: A map of the virgin forest. Ohio Journal of Science. 25(3):139-149.

Seeman M, Dancy W. 2000. The Late-Woodland Period in southern Ohio: basic issues and prospects. In: McElrath DL, Fortier AC, editors. Late Woodland societies: tradition and transformation across the midcontinent. Lincoln (NE): University of Nebraska Press.

Seymour RS, White AS, Philip GD. 2002. Natural disturbance regimes in northeastern North America—evaluating silvicultural systems using natural scales and frequencies. Forest Ecology and Management. 155(1-3):357-367.

Shands WE, Healy RG. 1977. The lands nobody wanted. Washington D.C. (United States): The Conversation Foundation.

65 | Page

Terrestrial Ecosystems Supplemental Report

Shane L, Snyder G, Anderson K. 2001. Holocene vegetation and climate changes in the Ohio Region. In: Prufer OH, Pedde SE, Meindl RS, editors. Archaic transitions in Ohio and Kentucky prehistory. Kent (OH): Kent State University.

Shane LCK, Anderson KH. 1993. Intensity, gradients and reversals in late glacial environmental change in east-central North America. Quaternary Science Reviews. 12(5):307-320.

Slucher E, Swinford E, Larsen GE. 2006. Bedrock geologic map of Ohio. Map BG-1, ver. 6.0 [scale 1:500,000]. Columbus (OH): Ohio Department of Natural Resources, Division of Geological Survey.

Smith B. 1989. Origins of agriculture in eastern North America. Science. 246(4937):1566-1571.

Sneddon LA, Gawler SC. 2010. Appalachian-Alleghenian Sandstone Dry Cliff Vegetation. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Sneddon L, Minney D, Gardner R. 2003.Western Allegheny Plateau Oak Barrens. United States National Vegetation Classification. Washington D.C. (United States): Federal Geographic Data Committee.

Society of American Foresters. 2014. The dictionary of forestry. Bethesda (MD): Society of American Foresters.

Stambaugh MC, Guyette RP. 2008. Predicting spatio-temporal variability in fire return intervals using a topographic roughness index. Forest Ecology and Management. 254:463-473.

Stambaugh MC, Marschall JM, Abadir ER, Jones BC, Brose PH, Dey DC, Guyette RP. 2018. Wave of fire: an anthropogenic signal in historical fire regimes across central Pennsylvania, USA. Ecosphere. 9(5):e02222.

Steiner KC, Westbrook JW, Hebard FV, Georgi LL, Powell WA, Fitzsimmons SF. 2017. Rescue of American chestnut with extraspecific genes following its destruction by a naturalized pathogen. New Forests. 48(2):317-336.

Stout SL, Royo AA, deCalesta DS, McAleese K, Finley JC. 2013. The Kinzua Quality Deer Cooperative: Can adaptive management and local stakeholder engagement sustain reduced impact of ungulate browsers in forest systems? Boreal Environment Research. 18(suppl. A):50-64.

Sutherland EK. 1997. History of fire in a southern Ohio second-growth mixed-oak forest. In: Pallardy SG, Cecich RA, Garrett HG, Johnson PS, editors. Proceedings of the 11th central hardwood forest conference. St. Paul (MN): USDA Forest Service, North Central Forest Experiment Station. NRS-GTR-188.

Thomas-Van Gundy MA, Nowacki GJ. 2013. The use of witness trees as pyro-indicators for mapping past fire conditions. Forest Ecology and Management. 304(2013):333-344.

66 | Page

Terrestrial Ecosystems Supplemental Report

Thomas-Van Gundy MA, Nowacki GJ. 2016. Landscape-fire relationships inferred from bearing trees in Minnesota. Madison (WI): USDA Forest Service, Northern Research Station. NRS-GTR-160.

Trautman M. 1977. The Ohio country from 1750 to 1977—a naturalist’s view. Biological Notes (Ohio Biological Survey), No. 10. Columbus (OH): Ohio State University.

Tyrrell LE, Nowacki GJ, Buckley DS, Nauertz EA, Niese JN, Rollinger JL, Zasada JC. 1998. Information about old growth for selected forest type groups in the eastern United States. St. Paul (MN): USDA Forest Service, North Central Forest Experimental Station. NC- GTR-197.

Vander Yacht AL, Barrioz SA, Keyser PD, Harper CA, Buckley DS, Buehler DA, Applegate RD. 2017. Vegetation response to canopy disturbance and season of burn during oak woodland and savanna restoration in Tennessee. Forest Ecology and Management. 390:187-202.

Vander Yacht AL, Keyser PD, Barrioz SA, Kwit C, Stambaugh MC, Clatterbuck WK, Simon DM. 2018. Reversing mesophication effects on understory woody vegetation in Mid- Southern oak forests. Forest Science. 65(3):289-303.

Van Driver B. 1990. Roadside geology of Pennsylvania. Missoula (MT): Mountain Press Publishing Co.

Varner JM, Arthur MA, Clark SL, Dey DC, Hart JL, Schweitzer CJ. 2016. Fire in eastern North American oak ecosystems: filling the gaps. Fire Ecology. 12(2):1-6.

Wiens JA, Hayward GD, Safford HD, Giffen CM. 2012. Historical environmental variation in conservation and natural resource management. Chichester (United Kingdom): John Wiley & Sons.

67 | Page