Papers by Melissa Thomas-Van Gundy
Ecological applications, Feb 13, 2024
Fire ecology, Jan 30, 2024
Forest Service Research Data Archive
Vita. Abstract. Thesis (M.S.)--State University of New York. College of Environmental Science and... more Vita. Abstract. Thesis (M.S.)--State University of New York. College of Environmental Science and Forestry, Syracuse, 1992. Includes bibliographical references (leaves 38-40). Microfilm of typescript. s
Natural Areas Journal, Jan 19, 2022
ABSTRACT Twenty years of monitoring data are assessed for trends in abundance of running buffalo ... more ABSTRACT Twenty years of monitoring data are assessed for trends in abundance of running buffalo clover (Trifolium stoloniferum), an endangered species found on the Fernow Experimental Forest in northeastern West Virginia, where the clover is found under various forest management intensities. Monitored locations were grouped for analysis into sites with common disturbance histories. Of the 13 sites, about half (46%, six sites) were found to have statistically significant trends between numbers of rooted crowns and time. Three out of eight sites with recent and sustained periodic disturbance showed a significant association between time and running buffalo clover counts and the trend is positive. Of the sites with no management or no recent harvest activity, three were found to have negative trends of running buffalo clover counts over time, although at only one site was this statistically significant. This analysis of monitoring data gives some support for previous findings that the number of running buffalo clover rooted crowns is tied to disturbance on the skid roads and increased light conditions created by periodic harvesting.
Forest Service Research Data Archive, Jun 21, 2021
Forest Service Research Data Archive
Forests
Red spruce (Picea rubens Sarg.) was a prized timber species in West Virginia during the era of re... more Red spruce (Picea rubens Sarg.) was a prized timber species in West Virginia during the era of resource exploitation in the late 1800s and early 1900s. As a result, central Appalachian red spruce comprise a much smaller component of high-elevation stand composition and a greatly constricted presence across the region. Widespread restoration efforts are underway to re-establish red spruce across this landscape. However, without benchmarks to gauge growth rates and stand developmental patterns, it is unclear whether these efforts are successful. Our goal was to develop reference curves predicting centile height growth for understory red spruce (≤7.6 m) across the region. We reconstructed the height growth patterns of over 250 randomly selected red spruce seedlings and saplings from 22 high-elevation stands in West Virginia. We also harvested 24 mature red spruce from the same stands to develop juvenile growth curves up to 7.6 m to compare understory growth rates of historical to conte...
This publication was prepared in part or in whole by employees of the Federal Government on offic... more This publication was prepared in part or in whole by employees of the Federal Government on official time and therefore cannot be copyrighted.
The American chestnut (Castanea dentata) was a keystone species that was decimated by nonnative d... more The American chestnut (Castanea dentata) was a keystone species that was decimated by nonnative diseases, most notably a fungus (Cryphonectria parasitica) that causes chestnut blight disease, during the early 20 th century in eastern North America. Breeding for a blight-resistant tree began over 100 years ago, and a backcross breeding approach that incorporated blight-resistant genes from Chinese chestnut (C. mollissima) was initiated in the 1980s. Field trials to test pure American chestnuts and hybrid trees from different breeding generations were established from 2009 to 2017.
Journal of Forestry, Sep 17, 2014
Active restoration of threatened or endangered species habitat may seem in conflict with the prov... more Active restoration of threatened or endangered species habitat may seem in conflict with the provisions of the Endangered Species Act because of the prohibition of "take," which can include habitat modification as well as death or harm to individuals. Risk-averse managers may choose to forego active management in known or presumed endangered species habitat to avoid killing an individual or harming critical habitat. We used the landscape-scale model LANDIS-II to simulate red spruce (Picea rubens Sarg.)-dominated forest response to restoration actions for 100 years. Restoration strategies differed in management of the potential habitat for the recently delisted Virginia northern flying squirrel (Glaucomys sabrinus fuscus). Model simulations show that active management with protections for existing red spruce stands resulted in a greater area dominated by red spruce than did larger, areawide protections. However, protecting larger areas of potential habitat resulted in an increase in red spruce in areas of low to moderate probability of occurrence for the flying squirrel, potentially increasing the area's suitability for this species.
Journal of Forestry, Jul 26, 2015
Cabwaylingo State Forest in southern West Virginia has experienced numerous anthropogenic wildfir... more Cabwaylingo State Forest in southern West Virginia has experienced numerous anthropogenic wildfires over the past 36 years. In this case study, we assessed the relationship between fire frequency and recency and stand composition and structure, with emphasis on oak and its competitors. Frequent and recent fire was significantly correlated with reduced red maple overstory stem density and basal area. Overstory oak density did not significantly vary with either fire frequency or recency. Total overstory basal area was greatest in areas of either no fire or nonrecent fire. Oak sapling density was significantly greater with high frequency and recent fire. Red maple sapling densities were greatest when fires were infrequent and recent, and red maple seedlings were greatest in no fire and low-frequency nonrecent fire areas. Our results suggest that recurring fire can enhance the development of large oak advanced reproduction. However, frequent fires without a sufficient fire-free interval could prevent the recruitment of oaks into the overstory.
Forest restoration would be greatly helped by understanding just what forests looked like a centu... more Forest restoration would be greatly helped by understanding just what forests looked like a century or more ago. One source of information on early forests is found in old deeds or surveys, where boundary corners were described by noting nearby trees known as witness trees. This paper describes the creation and analysis of a database of witness trees from original metes and bounds surveys of what became the Monongahela National Forest in West Virginia. We include an estimate of positional error from the conversion of paper maps to digital format. The final database contains 15,589 corners and 22,328 trees of 49 species from deeds dating from 1752 to 1899. White oak was the most frequent witness tree, followed by sugar maple, American beech, and American chestnut, and distribution patterns were recognizable across the study area. In early forests of the study area, magnolia, sugar maple, and black cherry were found on high-elevation ridges. Red spruce, hemlock, birch, and American beech were found on high-elevation toe slopes. Basswood was found in high-elevation coves, and red oak was associated with bench landforms at high elevations. At moderate elevations American chestnut and chestnut oak were associated with ridges, white pine and yellow pine occurred on benches, and an unknown species called spruce-pine was found on valley landforms. Blackgum was associated with toe slopes on low elevations, and black walnut was found on low-elevation benches. Low-elevation valleys contained white oak, elm, and sycamore. An important finding from this analysis is that some associations between species and environmental variables differed based on the ecological setting. Indicator kriging, using presence-absence data, resulted in probability of occurrence maps for selected species. We estimate that white oak covered 26 percent of the study area, sugar maple 19 percent, American chestnut 3 percent, and red spruce 2 percent.
Journal of The Torrey Botanical Society, Dec 14, 2021
Forest Service Research Data Archive, Jun 21, 2021
ABSTRACT Approximately 2600 km of bison and elk trails have been mapped in West Virginia using hi... more ABSTRACT Approximately 2600 km of bison and elk trails have been mapped in West Virginia using historical records, including county histories and travelers' accounts with published dates between 1876 and 1977. Most points of interest associated with these trails document European bison or elk hunting events. We explore some potential uses of the digital trails data and discuss the limitations of the data and need for further refinements. This map of major bison trails (available for download at https://doi.org/10.2737/RDS-2021-0055) is an important resource for telling a more complete story of the history of forests and woodlands of West Virginia.
Two prescribed fires and reductions in mid and overstory canopies by herbicide may be starting to... more Two prescribed fires and reductions in mid and overstory canopies by herbicide may be starting to reverse the mesophication trend in oak-dominated stands. Before treatment, the sapling layer was about 360 stems per acre and of that, approximately 20 percent was striped maple. The relative abundances of several mesic species increased after treatment (birch and red maple); however, the relative abundance of seedlings of other mesic species showed a temporary reduction after one fire; yellow-poplar seedling abundance dropped after two fires. The relative abundance of non-oak tree seedlings categorized as dry-mesic to xeric (e.g., hickory, sassafras) did increase after the two prescribed fires. Prescribed fire and mid-or overstory reduction treatments occurred at the same time in our study, with little to no existing advanced oak regeneration available to take advantage of favorable conditions created by the treatments. However, our results suggested fire reduced low shade (striped maple) but also shows the disadvantage of creating post-disturbance competitors for oaks (increases in birch and red maple). Two levels of canopy reduction by herbicide treatment conferred no immediate advantage to oak seedlings when combined with fire. For the oak seedlings 2 years post fire, the manipulation of the canopy and midstory by herbicide has not resulted in any benefit over fire alone. We expect that the species composition and structure of the seedling and sapling layers will show greater differences at 5 or 10 years postfire. In this analysis of almost immediate post-treatment effects, there has not been enough time for oak saplings to develop.
American Midland Naturalist, 2020
Bearing-tree data were used to calculate an index, pyrophilic percentage, depicting the importanc... more Bearing-tree data were used to calculate an index, pyrophilic percentage, depicting the importance of fire before Euro-American settlement on three landscapes, two within the Prairie Peninsula and one outside the region. Based on functional traits, bearing trees were classified as either pyrophilic or pyrophobic, applied to Public Land Survey points, and the pyrophilic percentage was calculated for each point. Kriging was applied to this point database to create a continuous surface of pyrophilic percentages. Regression analysis was used to relate this surface to environmental factors. Regression models created separately for each study area explained 38 to 53% of the variation in pyrophilic percentage. A positive association between pyrophilic percentage and distance to water and summer potential evapotranspiration was consistent across all study sites. The consistently high values and spatial patterns of pyrophilic percentage revealed fire-dominated landscapes interspersed with patches of pyrophobic vegetation. The restriction of pyrophobic areas to the leeside (east) of waterbodies indicated these served as firebreaks in a fire-swept landscape. Lake Michigan must have had a profound effect on pre-Euro-American settlement fire environments, serving as a massive physical firebreak while casting a moist maritime climate eastward. In southern and southwestern Illinois, the Mississippi River and associated tributaries along with an increase in topographic complexity also served as firebreaks, with pyrophobic forests restricted to riparian zones which progressively graded to pyrophilic vegetation on surrounding uplands. Our analysis expands the use of Public Land Survey data by converting bearing trees into a
On the George Washington and Jefferson National Forests, managers have used prescribed fire to cr... more On the George Washington and Jefferson National Forests, managers have used prescribed fire to create and maintain early-successional and open forest conditions across large areas. We used a landscape-scale and image-based approach to assess the extent that prescribed fires, including repeated fires, have created these forest conditions and put the results in context of the new George Washington National Forest management plan. At the landscape level, early-successional forest made up an average of 5 percent of burn unit area after one burn, 9 percent after two fires, 17 percent after three fires, and 14 percent after four fires. On average across all burn unit acreage, open forest made up 5 percent of the area after one burn, 7 percent after two burns, 9 percent after three, and 8 percent after four fires. The forest plan desired condition of 12 percent of the area in early-successional forest was met after three or four fires and was exceeded in some individual burn units. It is harder to achieve open-forest than early-successional conditions using prescribed fire alone. We also examined possible drivers of canopy gap creation in these forests. Vegetation type and heat load index, a topographic-based measure of solar radiation received by a site, were important predictors of where canopy gaps formed after prescribed fire.
Journal of The Torrey Botanical Society, Oct 1, 2012
Forest ecosystems will be affected directly and indirectly by a changing climate over the 21st ce... more Forest ecosystems will be affected directly and indirectly by a changing climate over the 21st century. This assessment evaluates the vulnerability of 11 forest ecosystems in the Mid-Atlantic region (Pennsylvania, New Jersey, Delaware, eastern Maryland, and southern New York) under a range of future climates. We synthesized and summarized information on the contemporary landscape, provided information on past climate trends, and described a range of projected future climates. This information was used to parameterize and run multiple forest impact models, which provided a range of potential tree responses to climate. Finally, we brought these results before two multidisciplinary panels of scientists and land managers familiar with the forests of this region to assess ecosystem vulnerability through a formal consensus-based expert elicitation process. Each chapter of this assessment builds on the previous chapter. The description of the contemporary landscape presents major forest trends and stressors currently threatening forests in the Mid-Atlantic region and defines the forest communities being assessed. The background information in Chapter 2 summarizes climate data analysis and climate models. Analysis of climate records in Chapter 3 indicates that average temperatures and total precipitation in the region have increased. Downscaled climate models in Chapter 4 project potential increases in temperature in every season, but projections for precipitation indicate slight increases in winter and spring, and high variability in summer and fall projections, depending on the scenario. Potential impacts on forests in Chapter 5 were identified by incorporating the future climate projections into three forest impact models (DISTRIB, LINKAGES, and LANDIS PRO). These models project declines in growth and suitable habitat for many mesic species, including American beech, eastern hemlock, eastern white pine, red spruce, and sugar maple. Species that tolerate hotter, drier conditions are projected to persist or increase, including black oak, northern red oak, pignut hickory, sweetgum, and white oak. Climate impacts related to topics such as wildfire, invasive species, and forest pests were not included in the forest impact models, but were summarized from published literature. In Chapter 6, we assessed vulnerability for 11 forest communities in the Mid-Atlantic region. Twenty-six science and management experts from across the region considered vulnerability in terms of the potential impacts on a forest ecosystem and the adaptive capacity of the ecosystem. The montane spruce-fir and lowland conifer forest communities were determined to be the most vulnerable ecosystems in the interior portion of the Mid-Atlantic region. Maritime and tidal swamp forest communities were determined to be the most vulnerable ecosystems in the coastal plain portion of the region. The woodland, glade, and barrens forest community was perceived as less vulnerable to projected changes in climate. Forest ecosystem vulnerabilities are expected to affect other forest-dependent topics such as wildlife management, timber production, and recreation. Information on these and other topics is summarized in Chapter 7. Cover Photo Allegheny Reservoir. This 25-mile-long lake touches nearly 100 miles of forested shoreline within the Allegheny National Forest boundaries, and provides both recreation opportunities and municipal water. Photo by USDA Forest Service, Eastern Region, via flickr.com.
Uploads
Papers by Melissa Thomas-Van Gundy