Thesis (Ph. D.)--Oregon State University, 2007. / Printout. Includes bibliographical references (leaves - ). Also available on the World Wide Web.

<div>Prescribed fire is one of the most useful tools available to forest managers attempting to maintain oak-hickory forests in the Central Hardwood Region. Prescribed fire can be useful in promoting regeneration of desirable species groups like oak (Quercus spp.) and hickory (Carya spp.) by preparing the seedbed, managing competition, and creating canopy gaps. The use of prescribed fire has been limited by concerns regarding the effect of the practice on standing timber. A perception of strong negative effects to tree-quality and tree-value from fire originated largely from sometimes deleterious effects of wildfire on timber. Less research exists demonstrating the potential effects of controlled, prescribed burning on timber quality and value. Furthermore, most research that exists focuses on individual tree characteristics, and is often focused on a relatively small geographic areas.</div><div><br></div><div>I conducted a regional study on the effects of prescribed fire on timber quality across a gradient of the Central Hardwood Region, ranging from the Missouri Ozarks to the Appalachian foothills. I studied 139 stands in selected prescribed fire units and control sites in Mark Twain National Forest (MO), Hoosier National Forest (IN), Wayne National Forest (OH) and Daniel Boone National Forest (KY). Selected stands were dominated by hardwoods species and had variable prescribed fire histories, ranging from 0 to 6 prescribed fires. </div><div><br></div><div>Measurements were taken concurrently across this plot network for two studies. First, we assessed the estimated effect of prescribed fire on stumpage value, and secondly, we assessed wounding patterns and effects of prescribed fire on tree-quality. Loss in estimated stumpage value from prescribed fire averaged approximately 4.2% across all measured stands. Estimated loss in stumpage value varied significantly by the number of prescribed fires in the last 30 years, with increasing numbers of prescribed fires leading to higher estimated losses in stumpage value. Further, stands in Mark Twain National Forest exhibited higher estimated loss in stumpage value, exceeding 10% on average. Stands in Hoosier, Wayne, and Daniel Boone National Forest only rarely exceed 5% losses in estimated stumpage value, and averaged less than 3%. </div><div><br></div><div>Approximately 25% of trees had at least one wound associated with prescribed fire across all study sites, while approximately 5% of trees experienced a reduction in tree quality (as measured by United States Forest Service tree grade) from prescribed fire. Both the rate of wounding and rate of tree grade reduction increased with increasing numbers of prescribed fires. Stands in the western portion of the Central Hardwood Region (Hoosier and Mark Twain National Forest) exhibiting higher rates of wounding from fire compared to eastern sites (Wayne and Daniel Boone National Forest.)</div><div><br></div><div>Effects of wounding varied significantly by type of wound. Catfaces accounted for far more volume loss and reduction in tree grade than any other wound type. Alternatively, some wound types, like seams and bark slough, caused minimal tree-quality and tree-volume effects. Effects also varied by species, with higher wounding effects on sugar maple and red oak, and relatively low effects on white oak and yellow-poplar. </div><div><br></div>

A stratified sampling method was used to study the effects of two prescribed fires on a 250-acre section at the northwest end of Fork Ridge, Hoosier National Forest, in the spring of 1993 and 1995. An unburned area at the southeast end of Fork Ridge, adjacent to the burned area, and with similar forest communities, was used as a control area. Three growing seasons after the last fire, the burned area exhibited noticeable differences in understory vegetation. Herbaceous species diversity and richness, total herb cover on mesic sites, and mean percent cover and relative frequency for mesic-site, shade-tolerant species were greater on the burned area than on the unburned area. Dry-site, shade-intolerant tree seedlings including scarlet oak (Quercus Coccina), chestnut oak (Quercus prinus), sassafras (Sassafras albi dum), and tulip poplar (Liriodendron tulipifera) had higher relative frequency in the burned area, while shade-tolerant flowering dogwood (Cornus Florida) had lower relative frequency. / Department of Biology

<p>  </p> <p>Alterations to the historic fire regime have contributed to widespread regeneration failure in <em>Quercus</em> L. (oak) forests of the eastern United States. Composition has shifted from <em>Quercus</em> and other fire-adapted species to dominance by mesophytic species. While land managers often focus efforts on restoring <em>Quercus</em> regeneration, the herbaceous layer experiences reduced cover and diversity of herb and graminoid species resulting from the increased woody stem density in fire-suppressed forests. Declining abundance of <em>Quercus</em> species and diversity in the herbaceous layer reduce the overall habitat quality and ecosystem functions provided by the forest. A combination of overstory harvests and prescribed burning are often conducted to restore the plant community in <em>Quercus</em> forests affected by mesophication. Initiated in 2010, our study on the Hoosier National Forest in Indiana conducted shelterwood and midstory (mechanical, chemical, or none) harvests followed by prescribed burning on a less productive site, while leaving a more productive site unburned. Our objective was to evaluate the survival and competitive response of <em>Quercus</em> spp. within the regeneration layer and whether diversity and cover increased in the herbaceous layer following treatments. Using nested circular plots, we measured seedling survival and resprout response, in addition to regeneration density before and after treatments. We measured the percent cover of herbaceous-layer species within quadrats and calculated species richness, evenness, and diversity. Using multiple mixed-effects models, ANOVA, and NMDS ordination, we evaluated woody species regeneration and herbaceous-layer composition before and after treatments. Post-treatment, monitored <em>Quercus</em> spp. seedlings at the burned site displayed greater survival (> 94%) and resprouting (> 92% of monitored stems), which exceeded most competing species, including<em> Acer</em> spp. (~ 59% survival and resprouting) and <em>Fraxinus americana </em>(72% survival and resprouting). <em>Q. alba</em> seedling (< 3.8 cm DBH) densities doubled after burning; it was one of the most abundant species (9,864 stems ha-1) at the burned site. NMDS ordination indicated a clear shift in regeneration species composition with the burn driving a shift away from mesophytic species towards greater importance of <em>Quercus</em> species. Additionally, our burned site had significantly increased herbaceous-layer richness, Shannon diversity index, and total cover compared to pre-treatment. Percent cover increased across all plant functional groups within the herbaceous layer, with trees/shrubs exhibiting the greatest increase. Herbaceous-layer composition at the burned site significantly shifted toward greater importance of graminoids and herbs post-treatment. Post-treatment, the unburned site contained fewer, and less competitive, <em>Quercus</em> seedlings compared to non-<em>Quercus</em> competitors and displayed no significant compositional shifts in seedling species composition post-harvest. Our unburned site exhibited significant, but minor, increases in herbaceous-layer richness, evenness, diversity, and total cover. Herbaceous-layer composition at the unburned site was significantly different post-treatment, shifting towards greater importance of vines, trees/shrubs, and herbs. The more-productive unburned site would likely require multiple burns to produce adequate competitive <em>Quercus</em> seedlings to perpetuate dominance in the developing stand. Burning would also likely result in greater increases in herbaceous-layer diversity compared to harvesting alone. Conversely, the shelterwood, followed by a single burn, on the less productive site had a more substantial effect on the herbaceous layer, and likely produced an adequate density of <em>Quercus</em> reproduction to ensure future dominance by the genus.</p>

This thesis examines the Schultz Fire as a case study to explain the complex history of fire suppression management in America’s forests, and to gain further understanding of how management practices have affected the increase in fire severity levels and how forests respond to such a disturbance. The thesis objectives were: (1) to analyze the causes of the fire severity of the Schultz Fire, especially: topography, fuels, or weather; (2); to examine the possible correlation between fire severity and tree density; (3) to investigate whether post-fire species richness was related to fire severity two years after the Schultz Fire; (4) to investigate whether post-fire plant species richness, plant cover, and tree regeneration was related to fire severity two years after the Schultz Fire; and (5) to interlink and convey how these factors relate to the history of fire management and policy and public perception. The history of fire related policy and management has significantly changed the dynamics of America's national parks and forests. Understanding the larger context of this history, both of national fire management and of the effects of language and perception on policy and public reaction, is part of understanding the Schultz Fire as a whole. Based on modeling, high winds combined with the presence of high surface fuel load were the main causes of the Schultz Fire's high fire severity levels. As fire severity increased there was a statistically significant increase in species richness. Severity level had little variation on percentage of cover by plants. No statistically significant relationship between tree density and fire severity levels was found. These findings underline the need for fuel treatments in southwest Ponderosa Pine forests, and effective cooperation between communities, managers, and ecologists. The Schultz Fire serves as an example in understanding the intricacies of how history affects the present and future of fire management. How fire has been managed and portrayed in the past has left an indelible mark on how fire is presently viewed. Without a clear understanding of the history of fire management and the role of fire in the ecology, future policies towards fire will be unable to account for and manage for the diversity of ecosystems and fires effects on those ecosystems across the United States. / Graduation date: 2013

<p dir="ltr">To examine how forest management affects terrestrial salamanders, this dissertation: (1) examines the effects of timber harvesting strategies on salamanders; (2) examines the effects of prescribed fire for oak regeneration on salamander populations; and (3) explores the influence of artificial cover object (ACO) wood type, size and shape, and placement on salamander monitoring results. These projects were conducted at the Hardwood Ecosystem Experiment (HEE) and Martell Experimental Forest in Indiana. Long-term salamander monitoring data from the HEE were used to examine the effects of clearcuts, shelterwoods, and patch cuts on salamander captures collected up to eleven years post-harvest and were analyzed with a before-after-control-impact (BACI) design. Clearcuts and patch cuts had negative effects on salamanders 4-6 years post-harvest, which coincided with a drought; however, preparatory and establishment shelterwood harvests showed no effects on salamander captures, suggesting that retaining canopy cover may protect salamanders from compound disturbances such as drought. Also at the HEE, capture-recapture techniques were used to examine salamander population estimates before and after fire. Only two of three fires affected salamander populations. In the short term, prescribed fire effects on salamanders may be weak and intermittent and microclimate may have a greater effect on populations, although the longer-term effects of fire remain unknown. At Martell Experimental Forest, salamander numbers were compared beneath ACOs of different wood types, sizes and shapes, and grid arrays of different spacings. Pine ACOs were preferred over ash, while several small ACOs yielded equal salamander numbers to one large ACO of equal total area. High ACO density may increase capture probability but reduce the area sampled by each ACO, while lower density ACO grids may cover a larger area with the same sampling effort and produce more comparable results, but with less precision; choice of ACO experimental design will therefore require careful consideration of management goals. This dissertation also suggests strategies to support salamander populations as guidelines for managers to consider in management planning.</p>