National Park Service U.S. Department of the Interior

Natural Resource Stewardship and Science

Fire History at New River Gorge National River Backus Mountain, Endless Wall, and Glade Creek (Babcock)

Natural Resource Report NPS/NERI/NRR—2019/1898

ON THIS PAGE Concord University geography students pose by a fire-scarred white oak (Quercus alba) sampled at the Backus Mountain site. From left: Justin Watkins (holding partial cross sections), Alysan Miller, Samuel Stockton, John Cline, and Jonathon Watkins. Credit: Thomas Saladyga

ON THE COVER Partial cross section of a live fire-scarred northern red oak (Quercus rubra) collected at the Backus Mountain site on September 7, 2018. This sample has an inner ring year of 1898 and dormant season fire scars dated to 1948 and 1992. Credit: Thomas Saladyga

Fire History at New River Gorge National River Backus Mountain, Endless Wall, and Glade Creek (Babcock)

Natural Resource Report NPS/NERI/NRR—2019/1898

Thomas Saladyga, Taylor Carter, John Cline, Maggie-Jean Cook, Kenneth Maiolo, Alysan Miller, Cody Perkins, Polina Proko’yeva, Samuel Stockton, Jonathon Watkins, and Justin Watkins

Environmental Geography Lab Department of Social Sciences Concord University P.O. Box 1000 Athens, WV 24712

March 2019

U.S. Department of the Interior National Park Service Natural Resource Stewardship and Science Fort Collins, Colorado

ii

The National Park Service, Natural Resource Stewardship and Science office in Fort Collins, Colorado, publishes a range of reports that address natural resource topics. These reports are of interest and applicability to a broad audience in the National Park Service and others in natural resource management, including scientists, conservation and environmental constituencies, and the public.

The Natural Resource Report Series is used to disseminate comprehensive information and analysis about natural resources and related topics concerning lands managed by the National Park Service. The series supports the advancement of science, informed decision-making, and the achievement of the National Park Service mission. The series also provides a forum for presenting more lengthy results that may not be accepted by publications with page limitations.

All manuscripts in the series receive the appropriate level of peer review to ensure that the information is scientifically credible, technically accurate, appropriately written for the intended audience, and designed and published in a professional manner.

This report received formal peer review by subject-matter experts who were not directly involved in the collection, analysis, or reporting of the data, and whose background and expertise put them on par technically and scientifically with the authors of the information.

Views, statements, findings, conclusions, recommendations, and data in this report do not necessarily reflect views and policies of the National Park Service, U.S. Department of the Interior. Mention of trade names or commercial products does not constitute endorsement or recommendation for use by the U.S. Government.

This report is available in digital format from the Eastern Rivers and Mountains Inventory & Monitoring Network website and the Natural Resource Publications Management website. If you have difficulty accessing information in this publication, particularly if using assistive technology, please email irma@nps.gov.

Please cite this publication as:

Saladyga, T., T. Carter, J. Cline, M. J. Cook, K. Maiolo, A. Miller, C. Perkins, P. Proko’yeva, S. Stockton, J. Watkins, and J. Watkins. 2019. Fire history at New River Gorge National River: Backus Mountain, Endless Wall, and Glade Creek (Babcock). Natural Resource Report NPS/NERI/NRR—2019/1898. National Park Service, Fort Collins, Colorado.

NPS 637/151064, March 2019

iii

Contents Page

Figures.................................................................................................................................................... v

Tables ..................................................................................................................................................... v

Photographs ............................................................................................................................................ v

Executive Summary ............................................................................................................................. vii

Acknowledgments ................................................................................................................................. ix

Introduction ............................................................................................................................................ 1

Methods .................................................................................................................................................. 3

Study Area ...................................................................................................................................... 3

Backus Mountain ....................................................................................................................... 3

Endless Wall .............................................................................................................................. 4

Glade Creek ............................................................................................................................... 4

Field Methods ................................................................................................................................. 4

Laboratory Methods ....................................................................................................................... 5

Fire Interval Analyses..................................................................................................................... 7

Climate-Fire Relationships ............................................................................................................. 7

Results .................................................................................................................................................... 9

Fire History Reconstructions .......................................................................................................... 9

Backus Mountain ....................................................................................................................... 9

Endless Wall ............................................................................................................................ 10

Glade Creek ............................................................................................................................. 10

Climate-Fire Relationships ........................................................................................................... 11

Discussion ............................................................................................................................................ 13

Conclusions .......................................................................................................................................... 15

Literature Cited .................................................................................................................................... 17

Appendix. Site Fire Years and Climate ............................................................................................... 21

v

Figures Page

Figure 1. Fire history site locations at New River Gorge National River. .........................................viii

Figure 2. Backus Mountain fire history site map. ................................................................................. 5

Figure 3. Site fire indices and number of samples for Backus Mountain, Endless Wall, and Glade Creek at New River Gorge National River. .......................................................................... 9

Figure 4. Interdecadal trends in fire activity, climate, and population density... ................................ 11

Figure 5. Results of Superposed Epoch Analysis (SEA) for 2-scar minimum fire events. ................ 12

Figure 6. Standardized Precipitation Evapotranspiration Index with 2-scar minimum fire events scaled to the number of sites recording fire. ............................................................................. 12

Tables Page

Table 1. Fire history site characteristics, including the species and number of trees used in each fire history reconstruction, location information, and recreational trail status at New River Gorge National River (NERI).............................................................................................. 4

Table 2. Site composite fire interval statistics for Backus Mountain, Endless Wall, and Glade Creek at New River Gorge National River (NERI). .................................................................. 10

Photographs Page

The rimrock pine-oak community at Endless Wall as viewed from the Chesapeake and Ohio Railroad tracks along the New River. Credit: Thomas Saladyga................................................. ix

Concord University geography students Polina Prokof’yeva and Maggie-Jean Cook (foreground) crossdating white oak (Quercus alba) tree core samples used to develop the Backus Mountain reference chronology. Credit: Thomas Saladyga ...................................................... 6

vii

Executive Summary Fire was historically an important ecosystem disturbance in the oak (Quercus spp. L.) forests of eastern North America. Fire exclusion in the 20th century via suppression policies and land use change, however, has altered species composition and forest structure (Nowacki and Abrams, 2008). Other interacting factors such as climate change, invasive species, and increased herbivory have compounded these effects, leading to reduced oak regeneration (McEwan et al., 2011; Knott et al., 2019). Regional land managers have recently begun to reintroduce fire to these landscapes, but there continues to be a need to place prescribed fire objectives into a historical context. Tree-ring, or dendroecological, reconstructions of fire can be used to inform these management objectives.

Dendroecological reconstructions of fire history use fire scars preserved in the annual growth rings of trees to develop precisely dated fire event chronologies. Previously published fire history reconstructions in the New River Gorge region of West Virginia, however, are restricted to two cliff edge or “rimrock” pine-oak communities at Endless Wall within New River Gorge National River (NERI) (Maxwell and Hicks, 2010) and at Glade Creek within adjacent Babcock State Park (Saladyga, 2017a) (see Figure 1).

The primary objective of this study was to provide resource managers at NERI with a historical perspective of wildfire in the dry oak forests at Backus Mountain and across the landscape in pine-oak forest communities. Specifically, our objectives included:

1. Reconstruct the fire history of Backus Mountain from fire-scarred oak trees.

2. Reanalyze previously published fire scar records from Endless Wall (Maxwell and Hicks,2010) and Glade Creek (Saladyga, 2017a) using methods that are consistent across studysites.

3. Assess the influence of climate (i.e., drought) on fire occurrence.

4. Evaluate trends in fire activity in the context of local land use history and deindustrialization.

Our major findings were the following:

1. Fire occurred every 3-4 years on average at Backus Mountain between 1958 and 1993 andevery 5-6 years at Endless Wall and Glade Creek between 1900 and 1970.

2. Fire activity was more closely linked to patterns of land use change (i.e., Park establishment)and regional deindustrialization, rather than to annual climate conditions.

3. Fire occurrence was generally asynchronous across sites. There was only one year (1953)when a fire scar was recorded by at least two trees at more than one site.

Future fire management objectives should take into account local variability in historical fire regimes at NERI as well as an increasingly wet and temperate regional climate. In addition, there is potential to incorporate local knowledge in fire management and forest restoration activities at Backus Mountain through targeted outreach programs, whereas locations with a heavy tourism footprint (i.e., Endless Wall and Glade Creek) and nearly 60 years of fire exclusion might present more logistical and perception barriers.

Figure 1. Fire history site locations (red triangles) at New River Gorge National River. Photo credits: Thomas Saladyga (Endless Wall and Glade Creek) and Jonathon Watkins (Backus Mountain).

ix

Acknowledgments We thank S. Perles and J. Perez for their insights regarding fire management and forest restoration activities at NERI and, specifically, at Backus Mountain. Funding for travel to the Backus Mountain study site was provided by Concord University. Supplies for processing wood samples were provided by the National Park Service.

The rimrock pine-oak community at Endless Wall as viewed from the Chesapeake and Ohio Railroad tracks along the New River. Credit: Thomas Saladyga.

1

Introduction Fire is an integral component to many plant communities across the Central Hardwood Forest Region in eastern North America (Hart and Buchanan, 2012). Twentieth century development, land use change, and active fire suppression, however, have altered fire regimes and caused shifts in species composition and forest structure. This “mesophication” of eastern oak (Quercus spp. L.) forests is characterized by reduced oak regeneration and increased presence of fire-intolerant species, notably red maple (Acer rubrum L.) (Nowacki and Abrams, 2008). In addition to fire exclusion, increased moisture availability, intensified herbivory, and invasive species have also contributed to these compositional and structural changes (McEwan et al., 2011; Knott et al., 2019). Regional land managers have recently begun to reintroduce fire to these landscapes, but there continues to be a need to place prescribed fire objectives into a historical context. Tree-ring derived, or dendroecological, fire history reconstructions provide this context for science-based management decisions.

Dendroecological reconstructions of fire history use fire scars preserved in the annual growth rings of trees to develop precisely dated fire event chronologies. These fire histories are used to characterize fire regimes (i.e., frequency, extent, and seasonality) and provide context for fire management objectives. Fire history reconstructions are generally sparse in eastern oak forests (Hart and Buchanan, 2012) and studies in the Appalachian Region are limited to the western edge of the Allegheny Plateau and the relatively drier Blue Ridge Mountains and Ridge and Valley physiographic provinces (Lafon et al., 2017). In southeastern Ohio and eastern Kentucky, for example, oak fire histories indicate that fire occurred frequently, every 2-12 years on average, from 1870 to 1950 (McEwan et al., 2007a). Similar patterns of fire activity have been documented at high elevation pine-oak sites in North Carolina and Tennessee (Flatley et al., 2013), Virginia (Aldrich et al., 2014), and West Virginia (Hessl et al., 2011), where reduced fire activity coincided with deindustrialization and the application of fire suppression policies in the early- to mid-20th century.

Previously published fire history reconstructions in the New River Gorge region of West Virginia are restricted to two cliff edge or “rimrock” pine-oak communities at Endless Wall within New River Gorge National River (NERI) (Maxwell and Hicks, 2010) and at Glade Creek within adjacent Babcock State Park (Saladyga, 2017a) (Figure 1). Prescribed fires have been conducted at Backus Mountain and in the Grandview area of NERI in recent years, but the development of future management strategies will require local ecological and historical context.

The primary objective of this study was to provide resource managers at NERI with a historical perspective of wildfire in the dry oak forests at Backus Mountain and across the landscape in pine-oak forest communities. Specifically, our objectives included:

1. Reconstruct the fire history of Backus Mountain from fire-scarred oak trees. 2. Reanalyze previously published fire scar records from Endless Wall (Maxwell and Hicks,

2010) and Glade Creek (Saladyga, 2017a) using methods that are consistent across study sites.

3. Assess the influence of climate (i.e., drought) on fire occurrence. 4. Evaluate trends in fire activity in the context of local land use history and deindustrialization.

3

Methods Study Area New River Gorge National River (NERI) is located in southern West Virginia on the Allegheny Plateau in the Central Appalachian Mountains. Its boundary stretches from Hawks Nest Lake north of US 19 in Fayette County to the Bluestone Dam south of Interstate 64 in Summers County (Figure 1). Established in 1978, the park encompasses approximately 30,000 hectares of mixed mesophytic forest along 85 km of the New River and contains numerous historical sites, including the remnants of more than 50 company-owned coal towns (NPS, 2018). Commercial coal mining and associated industries in the New River Gorge region increased steadily after the completion of the Chesapeake and Ohio Railroad in 1873. By the mid-20th century, however, most mines in the gorge ceased operation. Consequently, Fayette County, for example, lost 25% of its population between 1950 and 1960 (US Census Bureau, 2018), while poverty and unemployment rates in the region have since been consistently above the national average (ARC, 2018). Today, tourism and service industries are the primary driver of local economies with more than 1 million visitors to NERI spending nearly $50 million annually (Cullinane et al., 2018).

Plant communities at NERI are diverse, reflecting the complex topography and dissected stream valleys of the region. In the uplands, dry oak and pine forests dominate the ridges, southern facing slopes, and cliff edges, while hardwoods such as maple (Acer spp. L.), American beech (Fagus grandifolia Ehrh.), and eastern hemlock (Tsuga canadensis (L.) Carrière) occupy cool, moist northern facing slopes (Vanderhorst et al., 2007). There are also Appalachian cove forests, floodplain forests with sycamore (Platanus occidentalis L.) and river birch (Betula nigra L.), and Appalachian flatrock communities along the New River (Vanderhorst et al., 2007). The regional climate is humid continental, with an average annual temperature of 9.7°C and monthly temperatures ranging from 20.4°C in July to -2.0°C in January (1981-2010; NOAA, 2018). Annual precipitation is 124.4 cm, ranging from 13.3 cm in July to 8.2 cm in October (1981-2010; NOAA, 2018). The climate across West Virginia and the region in general has become increasingly wet and temperate since the 1960s (Kutta and Hubbart, 2018).

Backus Mountain The Backus Mountain site is located approximately 5 km southeast of the community of Prince in Fayette County. It is a flat to gently sloping southern facing site comprised primarily of oak species, but with scattered red maple and tulip poplar (Liriodendron tulipifera L.) in the canopy. In the 1870s, the New River Coal Company began operating multiple coal mines along Laurel Creek, a tributary of the New River that flows along the north flank of Backus Mountain (Peters and Carden, 1926). These mines, owned later by various coal companies, ceased operations by the 1950s, with the exception of the Layland Mine, which produced coal until the 1980s. Coal was mined at a loss on the south side of the mountain and mining operations ceased there in 1911 (Peters and Carden, 1926). Today, Backus Mountain is relatively undeveloped for tourism compared to the other two fire history sites, having only one primitive campground with eight campsites and no recreational trail system (Table 1).

Table 1. Fire history site characteristics, including the species and number of trees used in each fire history reconstruction, location information, and recreational trail status at New River Gorge National River (NERI).

Site Species (n) Latitude Longitude Elevation

(m) Recreational

trails

Backus Mtn. Quercus alba (5) Quercus rubra (3) Quercus velutina (3)

37°50’55” N 81°00’10” W 770 No

Endless Wall Pinus virginiana (21) 38°03’15” N 81°03’32” W 580 Yes

Glade Creek Pinus rigida (4) Pinus virginiana (2)

37°59’20” N 81°56’54” W 740 Yes

Endless Wall The Endless Wall site is located approximately 4 km east of the town of Fayetteville in Fayette County. The site is described as a rimrock pine community situated on southern facing slopes at the edge of extensive sandstone cliffs. The canopy is characterized by Virginia pine (P. virginiana Mill.), black gum (Nyssa sylvatica Marsh.), sourwood (Oxydendrum arboreum (L.) DC.), red maple, and oak (Maxwell and Hicks, 2010). The nearby Nuttallburg Mine ceased operations in 1958 and the property was acquired by the National Park Service in 1998. The Endless Wall continues to be one of the most visited areas at NERI, with easy access from US 19 and the town of Fayetteville.

Glade Creek The Glade Creek site is located approximately 2 km south of the community of Clifftop in Fayette County. It is situated along southern facing slopes above Glade Creek within Babcock State Park, which was established in 1934 during the Civilian Conservation Corps. (CCC) era. The forest type is similar to the Endless Wall site, but with a higher concentration of pitch pine (Pinus rigida Mill.) (Saladyga, 2017a). Industrial activities began in the area in 1873 when the Longdale Iron Company opened mines at nearby Sewell. A decade later, mines were opened along Manns Creek near Clifftop when the coal was exhausted at Sewell (Peters and Carden, 1926). Coal processing at Sewell and the mines at Clifftop ceased operations in 1956. Today, Babcock State Park offers a variety of recreational activities in close proximity to NERI.

Field Methods Fieldwork at the Backus Mountain site was conducted on September 7-8, 2018. We cut a partial cross section from a total of 17 fire-scarred oak trees spread across an approximately 1.5 km linear study area (Figure 2) using well-established methods (e.g., Arno and Sneck, 1977; Speer, 2010). Canopy dominant/co-dominant oak trees were targeted for sampling based on the visible presence of a basal scar. Each partial cross section was cut at a height that ranged between 0.5 m and 1 m above ground level depending on the height of the basal scar and degree of heart rot. There were no fallen logs, snags (standing dead trees), or stumps available for sampling at the site. Each tree was identified to species and its geographic coordinates and diameter at breast height (DBH) were recorded. We also collected two tree cores approximately 1 m above ground level from 10 live,

5

unscarred white oaks (Q. alba L.) using an increment borer in order to develop a local reference chronology. The DBH and geographic coordinates of each of these trees was recorded. Similar field methods were used at the Endless Wall and Glade Creek sites in 2005 and 2015, respectively, and are described in previous publications (Maxwell and Hicks, 2010; Saladyga, 2017a).

Figure 2. Backus Mountain fire history site map. Tree cores collected from nine white oaks (Quercus alba) were used to develop a local reference chronology. Partial cross sections cut from 11 fire-scarred oak trees were crossdated against the reference chronology. Six partial cross sections and two cores (collected from one tree) could not be dated due to substantial heart rot.

Laboratory Methods Backus Mountain tree cores and partial cross sections were air dried and then sanded until cells were clearly visible under magnification (Speer, 2010). We visually crossdated (Schweingruber et al., 1990) and then scanned tree core samples using a high-resolution flatbed scanner (3200 dpi). We then used the image analysis software CooRecorder 8.1 (Larsson, 2016) to measure annual rings to the nearest 0.01 mm. The dating of each measurement series was quality checked and the reference chronology was finalized using the program COFECHA (Grissino-Mayer, 2001a). We then dated the fire-scarred cross sections against this local reference chronology using skeleton plots (Schweingruber et al., 1990). Each fire scar was assigned to a calendar year and, if possible, the season of occurrence based on its relative position to annual growth rings. Seasonal categories include: earlywood (E), latewood (L), dormant (D), and undetermined (U). Fire scars occurring between annual rings (i.e., dormant season) were assigned to the year prior to callus response in order to remain consistent with published regional fire histories (Hessl et al. 2011; Maxwell and Hicks, 2010; Saladyga, 2017a). Therefore, a dormant season fire dated to 1964 could have occurred in fall 1964 or spring 1965. Small, less than 1 cm, compartmentalized wounds in oak samples that did not

correspond to wound years in nearby trees were marked as “injuries” and left out of subsequent analyses. Therefore, the results presented in this report are likely an underestimation of fire activity at Backus Mountain (see McEwan et al., 2007b).

Concord University geography students Polina Prokof’yeva and Maggie-Jean Cook (foreground) crossdating white oak (Quercus alba) tree core samples used to develop the Backus Mountain reference chronology. Credit: Thomas Saladyga

7

Fire Interval Analyses We entered Backus Mountain fire scar data in FHX2 format (Grissino-Mayer, 2001b) and downloaded Endless Wall fire scar data from the International Multiproxy Paleofire Database (IMPD, 2018). Glade Creek fire scar data was obtained from the first author of this report. We then calculated composite fire interval statistics for each site using Fire History Analysis and Exploration System (FHAES) software (Brewer et al., 2016). Annual site fire indices, which provide a measure of relative fire extent, were calculated by dividing the number of fire-scarred trees per year by the number of recorder (or live) trees in that year (Taylor et al., 2016; Saladyga and Standlee, 2018). Fire interval statistics were calculated using “filtered” fire event years, which included site fire index values ≥ 0.100 when there was a minimum of four recorder trees. Filtering fire scar data minimizes potential misinterpretations of site fire histories due to changes in sample size over time and is useful for comparing sites with different sample sizes and sample area (Falk et al., 2011).

Climate-Fire Relationships We assessed interannual climate-fire relationships using Superposed Epoch Analysis (SEA) (e.g., Silver et al., 2013). For this study, we tested for statistically significant departures from mean Palmer Drought Severity Index (PDSI; Palmer, 1965) and Standardized Precipitation Evapotranspiration Index (SPEI; Vicente-Serrano et al., 2010) within a five-year window lagged three years before, during, and one year after fire years when at least two trees at a site were scarred. We hypothesized that these fire years would have the strongest relationship with climate if one existed. The climate data used in the analysis included instrumental PDSI for West Virginia Climate Division 4 (NOAA, 2018) and data for SPEI grid point -81.25, 37.75 (Vicente-Serrano et al., 2010). Unlike PDSI, which is autoregressive and has a fixed temporal scale of 9-12 months, SPEI can be calculated for multiple timescales (1-48 months). The annual time series used in our analysis display similar trends during the common, overlapping period (1901-2015; r = 0.752, p < 0.001).

9

Results Fire History Reconstructions Backus Mountain We crossdated 11 partial cross sections cut from fire scarred white oak (n = 5), black oak (Q. velutina Lam.; n = 3), and northern red oak (Q. rubra L.; n = 3) (Table 1; Figure 2). The diameter of these trees ranged from 40.1 cm to 69.3 cm, with an average of 53.1 cm. We could not crossdate six of the samples due to substantial heart rot and only one crossdated sample contained pith. Crossdated samples recorded a total of 26 dormant season fire scars representing 18 unique fire years between 1948 and 2014 (Figure 3). Mean fire interval was 3.5 years for filtered fire events (Table 2) and there were five years between 1982 and 1993 when two or more trees recorded a fire scar. Longer term, or interdecadal, trends indicate that fires were most extensive during the 1980s-early 1990s at Backus Mountain (Figure 4). This time period immediately followed the establishment of NERI in 1978.

Figure 3. Site fire indices (bars) and number of samples (lines) for Backus Mountain, Endless Wall, and Glade Creek at New River Gorge National River (NERI). Solid bars indicate site fire index values ≥ 0.100 when there was a minimum of four recorder (or live) trees (solid lines). These years represent “filtered” fire events.

Table 2. Site composite fire interval statistics for Backus Mountain, Endless Wall, and Glade Creek at New River Gorge National River (NERI). Data are presented for filtered* fire event chronologies.

Fire interval statistics Backus Mtn. Endless Wall Glade Creek

Analysis period 1951-2017 1909-2005 1898-2012

Earliest fire 1958 1914 1900

Latest fire 1993 1970 1968

Total intervals 10 9 13

Mean fire interval 3.5 6.2 5.2

Standard deviation 3.3 7.0 3.2

Median fire interval 2.5 4.0 4.0

Maximum fire interval 11.0 20.0 13.0

Minimum fire interval 1.0 1.0 2.0

* Site fire index values ≥ 0.100 and a minimum of four recorder (or live) trees

Endless Wall Samples collected at the Endless Wall site recorded a total of 53 dormant season fire scars representing 25 unique fire years between 1914 and 2002 (Figure 3). Mean fire interval was 6.2 years for filtered fire events (Table 2) and there were nine years between 1934 and 1970 when two or more trees recorded a fire scar. Fire frequency and extent peaked in the mid-20th century and rapidly declined following the cessation of area mining operations (Figure 4). Fire has been essentially absent from the site since NERI was established in 1978, with only a single tree recording fire in subsequent decades.

Glade Creek Samples collected at the Glade Creek site recorded a total of 21 dormant season fire scars representing 17 unique fire years between 1887 and 1968 (Figure 3). Mean fire interval was 5.2 years for filtered fire events (Table 2) and there were three years between 1929 and 1953 when two or more trees recorded a fire scar. Fire activity peaked in the 1920s-1930s and gradually declined following the establishment of Babcock State Park in 1934 (Figure 4). Fire has been absent at the site since deindustrialization in the 1950s-1960s.

11

Figure 4. Interdecadal trends in fire activity, climate, and population density: (a) Site fire indices (filtered) and Standardized Precipitation Evapotranspiration Index (SPEI; grid point -81.25, 37.75); (b) Fayette County population density (US Census Bureau, 2018). Normalized values (z-scores) are smoothed for presentation using an adjustable B-spline (de Boor, 2001). Population density for 2016 is an estimated value. Relevant historical events are noted for reference.

Climate-Fire Relationships Superposed Epoch Analysis (SEA) results indicate that climate conditions were not uniquely dry in the three years preceding, during, or one year after 2-scar minimum fire events (n = 16; Figure 5). There were no statistically significant departures from mean SPEI during the five-year analysis window. For PDSI, there was a statistically signification positive departure from the mean (i.e., wet) three years before a fire event. This result, however, likely has no meaningful relationship to fire occurrence. The similar results for both drought indices suggest weak, or near absent, climate-fire relationships, at least at the annual scale. Fires occurred under a variety of annual moisture conditions (Figure 6), although the climate was relatively drier during the mid-20th century compared to the decades before and after (Figure 4).

Figure 5. Results of Superposed Epoch Analysis (SEA) for 2-scar minimum fire events (n = 16). The vertical bars represent departure from mean (a) Standardized Precipitation Evapotranspiration Index (SPEI; grid point -81.25, 37.75) and (b) Palmer Drought Severity Index (PDSI; West Virginia Climate Division 4). Both variables are lagged three years before and one year after a fire year. Dashed lines indicate 95% confidence intervals.

Figure 6. Standardized Precipitation Evapotranspiration Index (SPEI; grid point -81.25, 37.75) with 2-scar minimum fire events (n = 16) scaled to the number of sites recording fire.

13

Discussion Fire regimes at NERI were variable across sites during the last 70 to 120 years. Regardless of location, however, fire activity was more closely linked to patterns of land use change (i.e., Park establishment) and regional deindustrialization, rather than to annual climate conditions. These findings, including weak climate-fire relationships, are consistent with other regional fire histories (e.g., Flatley et al., 2013, McEwan et al., 2007a; Silver et al., 2013). Our conclusions, however, are limited by small sample size, particularly at Glade Creek, and few samples that pre-date the early 1900s at all sites. Nevertheless, our results indicate that fire occurred every 5-6 years on average at the two pine-oak sites between 1900 and 1970, while fire was slightly more frequent at Backus Mountain, occurring every 3-4 years on average between 1958 and 1993. There was only one year (1953) when a fire scar was recorded by at least two trees at more than one site (see Figure 6), suggesting that fire occurrence was generally asynchronous across sites.

The Backus Mountain fire history is unique compared to other oak fire scar records from the Central Appalachian Mountains. For example, recent (post-1950) fires are absent or rare in records from the Ridge and Valley physiographic province in West Virginia (Hessl et al., 2011) and Maryland (Shumway et al., 2001). The oak fire scar record at these sites extends to the late-19th century and early 17th century, respectively, while the absence of fire scars after 1960 at both sites coincided with changes in land use, population loss, and fire suppression by state and federal agencies. Similarly, fire scar records from oak sites on the Appalachian Plateau in southeastern Ohio and eastern Kentucky show that fire was rare after 1930 and nearly absent after 1950 (McEwan et al., 2007a; McEwan et al., 2014). Contrary to these findings, fires at Backus Mountain continued to occur decades after mid-20th century deindustrialization, with fires being most extensive during the 15 years after the establishment of NERI (Figure 4). These late 20th century wildfires at Backus Mountain might be attributed to the relative remoteness of the area and proximity to local residents with a tradition of fire use for informal forest management (e.g., Coughlan, 2016). Alternatively, or additionally, these fires could have been caused by arson carried out in response to land acquisition by the federal government (Kuhlken, 1999). Informal resource management and incendiary fires are both well-documented in the anthracite coal region of Pennsylvania (Saladyga, 2017b; Saladyga and Standlee, 2018), which has a similar industrial history and socioeconomic context as the New River Gorge region. Prior to 1950, we can only assume that fire was a frequent disturbance at Backus Mountain due to the substantial amount of heart rot found in samples and the fact that oaks are prone to decay following injury (Smith and Sutherland, 2001).

Fire scar records from the Endless Wall and Glade Creek sites follow the dominant 20th century trend in fire activity observed across the Central Appalachian Mountains. Unlike the oak forest at Backus Mountain, fire has been essentially absent from these pine-oak sites since the 1950s-1960s. This timing of fire cessation generally coincides with fire scar records from pine-oak sites in Virginia (Aldrich et al., 2014; Silver et al., 2013) and West Virginia (Hessl et al., 2011). The establishment of Babcock State Park in 1934 marked the beginning of a gradual decline in fire activity at Glade Creek, while subsequent mine closures and associated population loss in the 1950s resulted in a further reduction in fire activity at Endless Wall and Glade Creek. The establishment of NERI in 1978

ushered in a new era of tourism and federal land management, which excluded fire from these landscapes via reduced ignitions and active fire suppression.

Future fire management objectives should take into account these local variations in historical fire regimes at NERI. For example, the areas around the Endless Wall and Glade Creek sites have been developed for tourism to a much greater degree than at Backus Mountain. This development has resulted in the nearly complete exclusion of fire from these pine-oak sites for almost 60 years. Therefore, it will be necessary to consider not only the potential ecological effects of reintroducing fire to these landscapes [e.g., proliferation of exotic species (Alba et al., 2015) and negative impacts on oak regeneration (Fredericksen and Frank, 2014], but also tourist perceptions and impacts. Local residents, on the other hand, could have an active stakeholder role in fire management and forest restoration activities at Backus Mountain and similar locations at NERI where fire is a more recent human (and ecological) memory. As such, fire management at Backus Mountain has potential to engage local residents as a living component of the fire regime, possibly through targeted outreach programs, whereas locations with a heavy tourism footprint might present more logistical and perception barriers.

15

Conclusions In this study, we used samples cut from fire-scarred oak trees to develop an approximately 70-year fire event chronology for Backus Mountain. We also reanalyzed previously published fire scar records for two nearby pine-oak sites (Figure 1) in an effort to provide one cohesive document that synthesizes current knowledge of 20th-21st century fire occurrence and extent at NERI. Our results highlight variability in fire regimes, both in time and space. In general, fire activity closely followed trends in land use and deindustrialization, but the timing of these relationships differed across sites. At Backus Mountain, fire occurred every 3-4 years between 1958 and 1993 and was most extensive during the decade and a half following the establishment of NERI. Substantial heart rot in samples suggests that fire was a frequent disturbance at Backus Mountain prior to 1950. Fire was relatively less frequent at Endless Wall and Glade Creek between 1900 and 1970 and trends in fire occurrence and extent were closely linked to mid-20th century deindustrialization. At these sites, fire has been nearly absent from the landscape since the 1960s-1970s. Fire managers at NERI should consider this local variability in historical fire regimes as well as an increasingly wet and temperate regional climate when developing fire management objectives. In addition, fire managers might also consider the role local communities and individual residents can have in informing these objectives.

17

Literature Cited Alba, C., H. Skálová, K. F. McGregor, C. D'antonio, and P. Pyšek. 2015. Native and exotic plant

species respond differently to wildfire and prescribed fire as revealed by meta‐analysis. Journal

of Vegetation Science 26:102–113.

Aldrich, S. R., C. W. Lafon, H. D. Grissino‐Mayer, and G. G. DeWeese. 2014. Fire history and its relations with land use and climate over three centuries in the central Appalachian Mountains, USA. Journal of Biogeography 41(11):2093–2104.

Appalachian Regional Commission (ARC). 2018. Research, Maps, and Data. Available at: https://www.arc.gov/research/DataReports.asp (accessed 04 November 2018).

Arno, S. F. and K. M. Sneck. 1977. A method for determining fire history in coniferous forests in the mountain west. Gen. Tech. Rep. INT-GTR-42. Ogden, UT: US Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station, pp. 1–27.

Brewer, P. W., M. E. Velásquez, E. K. Sutherland, and D. A. Falk. 2016. Fire History Analysis and Exploration System (FHAES) version 2.0.2. Available at: http://www.fhaes.org (accessed 10 October 2018).

Coughlan, M. R. 2016. Wildland arson as clandestine resource management: A space–time permutation analysis and classification of informal fire management regimes in Georgia, USA. Environmental Management 57(5):1077–1087.

Cullinane Thomas, C., L. Koontz, and E. Cornachione. 2018. 2017 National Park visitor spending effects: Economic contributions to local communities, states, and the nation. Natural Resource Report NPS/NRSS/EQD/NRR—2018/1616. National Park Service, Fort Collins, Colorado.

de Boor, C. 2001. A practical guide to splines. Springer, New York.

Falk, D. A., E. K. Heyerdahl, P. M. Brown, C. Farris, P. Z. Fulé, D. McKenzie, T. W. Swetnam, A. H. Taylor, and M. L. Van Horne. 2011. Multi‐scale controls of h istorical forest‐fire regimes:

New insights from fire‐scar networks. Fronti ers in Ecology and the Environment 9(8):446–454.

Flatley, W. T., C. W. Lafon, H. D. Grissino-Mayer, and L. B. LaForest. 2013. Fire history, related to climate and land use in three southern Appalachian landscapes in the eastern United States. Ecological Applications 23(6):1250–1266.

Fredericksen, T. S. and M. Frank. 2014. Effects of prescribed burning on tree regeneration in a Southwestern Virginia mixed oak forest. International Journal of Applied Forestry 1:1–8.

Grissino-Mayer, H. D. 2001a. Evaluating crossdating accuracy: A manual and tutorial for the computer program COFECHA. Tree-ring research 57(2):205–221.

Grissino-Mayer, H. D. 2001b. FHX2-software for analyzing temporal and spatial patterns in fire regimes from tree rings. Tree-Ring Research 57(1):115–124.

Hart, J. L. and M. L. Buchanan. 2012. History of fire in eastern oak forests and implications for restoration. In: Dey, D. C., M. C. Stambaugh, S. L. Clark, and C. J. Schweitzer, eds. Proceedings of the 4th fire in eastern oak forests conference; 2011 May 17-19; Springfield, MO. Gen. Tech. Rep. NRS-P-102. Newtown Square, PA: US Department of Agriculture, Forest Service, Northern Research Station, pp. 34–51.

Hessl, A. E., T. Saladyga, T. Schuler, P. Clark, and J. Wixom. 2011. Fire history from three species on a central Appalachian ridgetop. Canadian Journal of Forest Research 41(10):2031–2039.

International Multiproxy Paleofire Database (IMPD). 2018. Available at: https://www.ncdc.noaa.gov/paleo-search/study/8472 (Accessed 10 October 2018).

Knott, J. A., J. M. Desprez, C. M. Oswalt, and S. Fei. 2019. Shifts in forest composition in the eastern United States. Forest Ecology and Management 433:176–183.

Kuhlken, R. 1999. Settin ‘the Woods on Fire: Rural Incendiarism as Protest. Geographical Review 89(3):343–363.

Kutta, E. and J. Hubbart, J. 2018. Changing climatic averages and variance: implications for mesophication at the eastern edge of North America’s eastern deciduous forest. Forests 9(10):605.

Lafon, C. W., A. T. Naito, H. D. Grissino-Mayer, S. P. Horn, and T. A. Waldrop. 2017. Fire history of the Appalachian region: a review and synthesis. General Technical Report SRS-219. Asheville, NC: USDA Forest Service, Southern Research Station.

Larsson, L. Å. C. 2016. CDendro Programs of the CooRecorder/CDendro Package Version 8.1.

Maxwell, R. S. and R. R. Hicks. 2010. Fire history of a rimrock pine forest at New River Gorge National River, West Virginia. Natural Areas Journal 30(3):305–311.

McEwan, R. W., T. F. Hutchinson, R. P. Long, R. D. Ford, and B. C. McCarthy. 2007a. 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.

McEwan, R. W., T. F. Hutchinson, R. D. Ford, and B. C. McCarthy. 2007b. An experimental evaluation of fire history reconstruction using dendrochronology in white oak (Quercus alba). Canadian Journal of Forest Research 37(4):806–816.

McEwan, R. W., J. M. Dyer, and N. Pederson. 2011. Multiple interacting ecosystem drivers: toward an encompassing hypothesis of oak forest dynamics across eastern North America. Ecography 34(2):244–256.

19

McEwan, R. W., N. Pederson, A. Cooper, J. Taylor, R. Watts, and A. Hruska. 2014. Fire and gap dynamics over 300 years in an old‐growth temperate for est. Applied Vegetation Science 17(2):312–322.

National Oceanographic and Atmospheric Association (NOAA). 2018. Divisional climate datasets, West Virginia Climate Division 4. Available at: https://www.ncdc.noaa.gov/cag/divisional/time-series (accessed 05 December 2018).

National Park Service (NPS). 2018. New River Gorge National River website. Available at: https://www.nps.gov/neri (accessed 11 September 2018).

Nowacki, G. J. and M. D. Abrams. 2008. The demise of fire and “mesophication” of forests in the eastern United States. AIBS Bulletin 58(2):123–138.

Palmer, W. C. 1965. Meteorological Drought. Office of Climatology, Research Paper 45. Weather Bureau, Washington, DC.

Peters, J. T. and H. B. Carden. 1926. History of Fayette County, West Virginia. Jarrett Print Company.

Saladyga, T. 2017a. Forest disturbance history from ‘legacy’pitch pine (Pinus rigida) at the New River Gorge, West Virginia. Natural Areas Journal 37(1):49–57.

Saladyga, T. 2017b. Reconstructing a cultural fire regime in the Pennsylvania Anthracite Region. Physical Geography 38(5):404–422.

Saladyga, T. and A. Standlee. 2018. Historical drivers and contemporary perceptions of wildfire in a post-industrial rural landscape. Fire 1(2):33.

Shumway, D. L., M. D. Abrams, and C. M. Ruffner. 2001. A 400-year history of fire and oak recruitment in an old-growth oak forest in western Maryland, USA. Canadian Journal of Forest Research 31(8):1437–1443.

Schweingruber, F. H., D. Eckstein, F. Serre-Bachet, and O. U. Bräker. 1990. Identification, presentation and interpretation of event years and pointer years in dendrochronology. Dendrochronologia 8:9–38.

Silver, E. J., J. H. Speer, M. Kaye, N. J. Reo, L. F. Howard, A. K. Anning, S. W. Wood, and H. M. Wilbur. 2013. Fire history and age structure of an oakpine forest on Price Mountain, Virginia, USA. Natural Areas Journal 33(4):440–446.

Smith, K.T. and E.K. Sutherland. 2001. Terminology and biology of fire scars in selected central hardwoods. Tree-Ring Research 57(2):141–147.

Speer, J. H. 2010. Fundamentals of tree-ring research. University of Arizona Press.

Taylor, A. H., V. Trouet, C. N. Skinner, and S. Stephens. 2016. Socioecological transitions trigger fire regime shifts and modulate fire–climate interactions in the Sierra Nevada, USA, 1600–2015 CE. Proceedings of the National Academy of Sciences 113(48):13684–13689.

United States Census Bureau. 2018. Fayette County, West Virginia QuickFacts. Available at: https://www.census.gov/quickfacts (accessed 01 December 2018).

Vanderhorst, J. P., J. Jeuck, and S. C. Gawler. 2007. Vegetation classification and mapping of New River Gorge National River, West Virginia. Technical Report NPS/NER/NRTR—2007/092. U.S. Department of the Interior, National Park Service, Northeast Region. Philadelphia, PA.

Vicente-Serrano, S. M., S. Beguería, and J. I. López-Moreno. 2010. A multiscalar drought index sensitive to global warming: the standardized precipitation evapotranspiration index. Journal of Climate 23(7):1696–1718. Available at: http://spei.csic.es/database.html (accessed 12 December 2018).

21

Appendix. Site Fire Years and Climate

Table A1. Backus Mountain fire scar years, including number of trees scarred, number of recorder (or live) trees, proportion of recorder trees scarred (Site fire index), Standardized Precipitation Evapotranspiration Index (SPEI; grid point -81.25, 37.75), and Palmer Drought Severity Index (PDSI; West Virginia Climate Division 4).

Year Trees scarred Recorder trees Site fire index SPEI PDSI

1948 1 3 0.333 0.175 1.039

1958 1 5 0.200 0.178 0.623

1959 1 5 0.200 -0.175 -1.923

1962 1 6 0.167 0.336 0.521

1963 1 6 0.167 -0.507 -0.930

1974 1 7 0.143 0.057 2.196

1980 1 10 0.100 -0.003 1.173

1982 2 10 0.200 0.097 0.218

1983 3 10 0.300 -0.157 -0.278

1988 1 11 0.091 -0.563 -2.123

1989 2 11 0.182 0.433 1.917

1992 2 11 0.182 -0.146 -0.012

1993 4 11 0.364 0.043 -0.253

1994 1 11 0.091 0.226 1.703

1999 1 11 0.091 -0.426 -3.469

2005 1 11 0.091 -0.123 -0.922

2009 1 11 0.091 0.409 0.594

2014 1 11 0.091 0.298 -0.402

Table A2. Endless Wall fire scar years, including number of trees scarred, number of recorder (or live) trees, proportion of recorder trees scarred (Site fire index), Standardized Precipitation Evapotranspiration Index (SPEI; grid point -81.25, 37.75), and Palmer Drought Severity Index (PDSI; West Virginia Climate Division 4).

Year Trees scarred Recorder trees Site fire index SPEI PDSI

1914 1 7 0.143 0.020 -0.428

1922 1 11 0.091 0.238 0.088

1930 1 17 0.059 -1.009 -4.138

1934 2 17 0.118 0.061 -0.584

1938 4 20 0.200 0.028 -0.632

1939 2 20 0.100 -0.372 -0.545

1940 1 20 0.050 0.013 0.195

1941 2 20 0.100 -0.758 -2.594

1945 1 20 0.050 0.168 0.283

1946 6 20 0.300 -0.497 -1.953

1948 1 20 0.050 0.175 1.039

1952 3 20 0.150 -0.298 -0.743

1953 5 20 0.250 -0.605 -2.169

1954 3 20 0.150 0.005 0.336

1956 1 20 0.050 0.245 1.873

1958 1 20 0.050 0.178 0.623

1963 1 21 0.048 -0.507 -0.930

1965 1 21 0.048 -0.400 -1.563

1970 10 21 0.476 -0.458 -0.309

1976 1 21 0.048 -0.270 -0.925

1983 1 20 0.050 -0.157 -0.278

1989 1 18 0.056 0.433 1.917

1993 1 16 0.063 0.043 -0.253

1998 1 13 0.077 0.091 -0.123

2002 1 12 0.083 0.193 0.335

23

Table A3. Glade Creek fire scar years, including number of trees scarred, number of recorder (or live) trees, proportion of recorder trees scarred (Site fire index), Standardized Precipitation Evapotranspiration Index (SPEI; grid point -81.25, 37.75), and Palmer Drought Severity Index (PDSI; West Virginia Climate Division 4).

Year Trees scarred Recorder trees Site fire index SPEI PDSI

1887 1 2 0.500 – –

1890 1 3 0.333 – –

1894 1 3 0.333 – –

1900 1 4 0.250 – -3.021

1907 1 4 0.250 0.481 3.012

1910 1 4 0.250 -0.168 -0.153

1918 1 4 0.250 0.690 1.553

1926 1 4 0.250 0.225 0.843

1929 3 4 0.750 0.318 1.265

1932 1 4 0.250 0.192 0.663

1939 1 5 0.200 -0.372 -0.545

1942 2 6 0.333 0.115 -0.160

1946 1 6 0.167 -0.497 -1.953

1950 1 6 0.167 0.219 2.276

1953 2 6 0.333 -0.605 -2.169

1966 1 6 0.167 -0.185 -2.287

1968 1 6 0.167 -0.388 -1.588

The Department of the Interior protects and manages the nation’s natural resources and cultural heritage; provides scientific and other information about those resources; and honors its special responsibilities to American Indians, Alaska Natives, and affiliated Island Communities.

NPS 637/151064, March 2019

National Park Service U.S. Department of the Interior

Natural Resource Stewardship and Science 1201 Oakridge Drive, Suite 150 Fort Collins, CO 80525

EXPERIENCE YOUR AMERICA TM