Charles W. Lafon

THE CONTEMPORARY FIRE REGIME OF THE CENTRAL APPALACHIAN MOUNTAINS AND ITS RELATION TO CLIMATE

This paper uses records of wildland fire to investigate the contemporary fire regime on federal lands in the central Appalachian Mountains of Virginia and West Virginia. During the study period (1970-2003), 1557 anthropogenic fires and 344 natural fires occurred on these lands. Most were small, low-intensity burns. However, fires of moderate to high intensity also occurred, and because of their larger sizes they were responsible for most of the area burned. Fire size also differed between natural and anthropogenic fires (median size 1.2 ha vs. 0.4 ha). A few of the anthropogenic fires were quite large, however (up to 6484 ha), whereas the largest natural fire measured only 1188 ha. Anthropogenic fires burned more area than natural fires and consequently they had a shorter fire cycle (1196 years for anthropogenic fires, 6138 years for natural fires). These fire cycles appear to be much longer than in the past, prior to fire suppression. Nonetheless, despite suppression efforts, a substantial amount of fire activity occurred during the study period when conditions were sufficiently dry. The dry conditions of spring and fall were especially favorable for burning. Moreover, on an interannual level, drought had a strong influence on the amount of fire activity.

Factors structuring the treeline ecotone in Fennoscandia

Many hypotheses have been put forward to explain the structure and position of alpine treelines. The spatial complexity of the ecotone, ranging from sharp boundaries to networks of tree patches within a heath matrix, may explain why no consensus has been reached. In this paper, we discuss factors from abiotic disturbances to herbivory that may help understand the spatial structure of the alpine treeline ecotone in Fennoscandia. The ecotone is dominated by mountain birch (Betula pubescens ssp. tortuosa), and may show a wide range of spatial structures. We discuss the influence of topography, seed limitations, seedling establishment, growth limitations, abiotic disturbances and herbivory as structuring factors. All of these factors may operate, but their relative importance in space and time is unknown. There is a basic difference between factors that prevent the establishment of trees, and thus act on early life history stages, and factors that thin out a previously dense forest, and thus act on adult trees. Mortality caused directly or indirectly by geometrid moths may belong to the latter category. We suggest that seedling and sapling mortality is more important than seed limitation for the establishment of new individuals in the treeline ecotone. Important mortality factors may be abiotic disturbances, competition (or allelopathy) from field layer plants and herbivory. The relative role of these factors needs to be examined further.

Fire history, related to climate and land use in three southern Appalachian landscapes in the eastern United States

Fire-maintained ecosystems and associated species are becoming increasingly rare in the southern Appalachian Mountains because of fire suppression policies implemented in the early 20th century. Restoration of these communities through prescribed fire has been hindered by a lack of information on historical fire regimes. To characterize past fire regimes, we collected and absolutely dated the tree rings on cross sections from 242 fire-scarred trees at three different sites in the southern Appalachian Mountains of Tennessee and North Carolina. Our objectives were to (1) characterize the historical frequency of fire in southern Appalachian mixed pine-oak forests, (2) assess the impact of interannual climatic variability on the historical occurrence of fire, and (3) determine whether changes in human culture and land use altered the frequency of fire. Results demonstrate that fires burned frequently at all three sites for at least two centuries prior to the implementation of fire suppression and prevention in the early to mid 20th century. Composite mean fire return intervals were 2-4 yr, and point mean fire return intervals were 9-13 yr. Area-wide fires that burned across multiple stands occurred at 6-13-yr intervals. The majority of fires were recorded during the dormant season. Fire occurrence exhibited little relationship with reconstructed annual drought conditions. Also, fire activity did not change markedly during the transition from Native American to Euro-American settlement or during the period of industrial logging at the start of the 20th century. Fire activity declined significantly, however, during the fire suppression period, with a nearly complete absence of fire during recent decades. The characterization of past fire regimes should provide managers with specific targets for restoration of fire-associated communities in the southern Appalachian Mountains. The fire chronologies reported here are among the longest tree-ring reconstructions of fire history compiled for the eastern United States and support the hypothesis that frequent burning has played a long and important role in the development of forests in the southern Appalachian Mountains.

Spatial Patterns of Fire Occurrence in the Central Appalachian Mountains and Implications for wildland Fire Management

We investigated spatial variations in the incidence of anthropogenic and natural (lightning-ignited) fire in the central Appalachian Mountains of Virginia and West Virginia using a record of wildland fires that occurred on federal lands between 1970 and 2003. A consideration of spatial variability in wildland fire is important for allocating fire-suppression resources and for informing resource managers who use naturally ignited wildland fires or prescribed fires in ecological restoration efforts and fuel reduction treatments. The central Appalachian region contains three physiographic provinces with distinct climate, terrain, and vegetation characteristics. Comparing ignition density, maximum fire size, and fire cycle among the three provinces indicated that the Appalachian Plateau—the westernmost province and also the highest, coolest, and wettest of the three—was the least fire-prone environment. The Blue Ridge province along the eastern edge of the region was most fire prone. The Ridge and Valley province, which occupies the center of the region, generally was intermediate in fire characteristics, despite having the driest climate and the greatest extent of flammable pine (Pinus L.)-and oak (Quercus L.)-dominated forests. At a finer spatial scale, fire activity varied topographically in all three provinces: ignition density declined with increasing elevation, but showed weaker, less consistent relationships to aspect. Spatial variations in the importance of natural fires may be of particular interest to federal resource managers who are developing plans for permitting natural fires to burn to restore fire-associated ecosystems. The Blue Ridge appears to be a particularly favorable environment for natural ignitions.

Climatic and topographic controls on patterns of fire in the southern and central Appalachian Mountains, USA

Climate and topography are two important controls on spatial patterns of fire disturbance in forests globally, via their influence on fuel moisture and fuel production. To assess the influences of climate and topography on fire disturbance patterns in a temperate forest region, we analyzed the mapped perimeters of fires that burned during 1930–2003 in two national parks in the eastern United States. These were Great Smoky Mountains National Park (GSMNP) in the southern Appalachian Mountains and Shenandoah National Park (SNP) in the central Appalachian Mountains. We conducted GIS analyses to assess trends in area burned under differing climatic conditions and across topographic gradients (elevation, slope position, and aspect). We developed a Classification and Regression Tree model in order to further explore the interactions between topography, climate, and fire. The results demonstrate that climate is a strong driver of both spatial and temporal patterns of wildfire. Fire was most prevalent in the drier SNP than the wetter GSMNP, and during drought years in both parks. Topography also influenced fire occurrence, with relatively dry south-facing aspects, ridges, and lower elevations burning most frequently. However, the strength of topographic trends varied according to the climatic context. Weaker topographic trends emerged in the drier SNP than GSMNP, and during low-PDSI (dry) years than high-PDSI (wet) years in both parks. The apparent influence of climate on the spatial patterning of fire suggests a more general concept, that disturbance-prone landscapes exhibit weaker fine-scale spatial patterning of disturbance than do less disturbance-prone landscapes.

Effects of Ice Storms, Southern Pine Beetle Infestation, and Fire on Table Mountain Pine Forests of Southwestern Virginia

Forests dominated by Table Mountain pine (Pinus pungens) cloak xeric sites in the Appalachian Mountains. Frequent fire and other disturbances probably maintained open pine stands in the past, but fire-exclusion promoted increased stand density and hardwood invasion. In stands we sampled, most Table Mountain pines belonged to a cohort that originated in the early 1900s, probably following a severe burn. Recent disturbances changed the stands considerably. Ice storms reduced the basal area of one stand by about 23%, lowering the abundance of pines more than hardwoods. Subsequently, a southern pine beetle infestation removed 30% of the remaining pine basal area. These disturbances stimulated little pine regeneration. In a neighboring stand disturbed by wildfire as well as the ice storms and pine beetles, the abundance of both hardwoods and pines declined dramatically. Basal area loss in the stand was 45%. The burn top-killed many of the hardwood trees and shrubs, which comprised a dense understory, and stimulated abundant Table Mountain pine regeneration. Our results suggest that without fire, long-term maintenance of Table Mountain pine is improbable. Periodic disturbances by ice storms and pine beetles are hastening the loss of pine stands from most portions of the landscape.

Simulating the reciprocal interaction of forest landscape structure and southern pine beetle herbivory using LANDIS

The reciprocal interaction of landscape structure and ecological processes is a cornerstone of modern landscape ecology. We use a simulation model to show how landscape structure and herbivory interact to influence outbreaks of southern pine beetle (Dendroctonus frontalis Zimmermann) in a landscape representative of the southern Appalachian Mountains, USA. We use LANDIS and its biological disturbance agent module to simulate the effects of landscape composition (proportion of landscape in host area) and host aggregation on the size and severity of insect outbreaks and the persistence of the host species, Table Mountain Pine (Pinus pungens Lamb.). We find that landscape composition is less important in the modeled landscapes than host aggregation in structuring the severity of insect outbreaks. Also, simulated southern pine beetle outbreaks over time tend to decrease the aggregation of host species on the landscape by fragmenting large patches into smaller ones, thereby reducing the severity of future outbreaks. Persistence of Table Mountain pine decreases throughout all simulations regardless of landscape structure. The results of this study indicate that when considering alternative restoration strategies for insect-affected landscapes, it is necessary to consider the patterns of hosts on the landscape as well as the landscape composition.

Forest disturbance by ice storms in Quercus forests of the southern Appalachian Mountains, USA

Abstract In 1994, two ice storms (major freezing rain events) disturbed forests on east- and southeast-facing slopes of the southern Appalachian Mountains, Virginia, USA. This study investigates impacts of the disturbance on Quercus forests, the predominant vegetation type in the Appalachians. Forests were sampled at six sites on two neighbouring mountains during 1997 and 1998. The two storms killed a significant number of trees, reducing basal area and density by 28.7% and 29.2%, respectively, in the transects sampled for this study. The storms also removed an estimated 43.7% of the canopy cover. Uprooting of trees disturbed 3.5% of the soil surface, on average. The ice storms eliminated some minor tree species from the sampled transects, reducing species richness; however, evenness increased. Both abiotic and biotic factors (mountain, slope, tree size, and species) influenced patterns of tree damage and mortality. Differences among species in susceptibility to ice storm damage seem particularly important for forest dynamics. The dominant Quercus prinus, along with Acer rubrum and Carya species, exhibited low mortality, whereas Pinus rigida had exceptionally high mortality. More generally, mortality was low in shade-tolerant and moderately shade-tolerant species, and high in those intolerant of shade. Periodic ice storms may contribute to the successional replacement of Pinus rigida and other susceptible species by less susceptible species.

FIRE HISTORY OF A TEMPERATE FOREST WITH AN ENDEMIC FIRE-DEPENDENT HERB

A dendroecological fire history study was conducted for The Nature Conservancys Narrows Preserve on Peters Mountain, Virginia, where the predominant vegetation is oak (Quercus L.)-dominated forest containing some other hardwoods and pines (Pinus L.). The site encompasses all the known habitat of the endangered and endemic Peters Mountain mallow (Iliamna corei Sherff.), a perennial herb that requires fire for seed germination and habitat maintenance. Fire scars from 73 pines indicate frequent burning in the past (Weibull median composite fire interval = 2.2 years), primarily during the dormant season. Fire frequency exhibited little temporal variability from the beginning of the fire chronology in 1794 until the 1940s, despite changing land uses. However, the incidence of fire declined subsequently with the advent of effective fire protection measures. Ageing trees near the mallow population indicates that the fire-tolerant chestnut oak (Quercus montana Willd.) recruited relatively continuously under frequent fire, but that other species were established primarily during the fire protection era. The decline in burning appears to have permitted an increase in tree density that likely inhibits the growth and recruitment of mallow plants. Our results suggest that reintroducing frequent fire would be an appropriate technique for managing the mallows and the greater Peters Mountain landscape.

Relationships of Fire and Precipitation Regimes in Temperate Forests of the Eastern United States

AbstractFire affects virtually all terrestrial ecosystems but occurs more commonly in some than in others. This paper investigates how climate, specifically the moisture regime, influences the flammability of different landscapes in the eastern United States. A previous study of spatial differences in fire regimes across the central Appalachian Mountains suggested that intra-annual precipitation variability influences fire occurrence more strongly than does total annual precipitation. The results presented here support that conclusion. The relationship of fire occurrence to moisture regime is also considered for the entire eastern United States. To do so, mean annual wildfire density and mean annual area burned were calculated for 34 national forests and parks representing the major vegetation and climatic conditions throughout the eastern forests. The relationship between fire activity and two climate variables was analyzed: mean annual moisture balance [precipitation P − potential evapotranspiration (PE...