 |
 |
||||||||
|---|---|---|---|---|---|---|---|---|---|
|
|||||||||
 |
|||||||||
|
|---|
 |
 |
 |
 |
 |
|
 |
|
| Home > Results > Northern Hardwoods | |
|
The following document is also available for download.
Clickable Table of Contents I. Presettlement Vegetation and Disturbance
II. Postsettlement Vegetation and Historic Land Use
III. Management Challenges and Recommendations
IV. Conclusions Northern hardwood forests lie between the transition hardwoods of southern New England & central Pennsylvania and the spruce-fir forests at high elevations & latitudes in the Northeast (Lull 1968; Figure 1). Northern hardwoods occur in a climatic zone with a frost-free period of 120-150 days (Lull 1968), covering parts of northern Pennsylvania, New York, Vermont, New Hampshire, northwestern Massachusetts, western Maine, and extreme northeastern Maine. Northern hardwood forests are primarily composed of American beech (Fagus grandifolia), sugar maple (Acer saccharum), and yellow birch (Betula alleghaniensis). Less-shade-tolerant species also found in this region include paper birch (Betula papyrifera), gray birch (Betula populifolia), mountain paper birch (Betula papyrifera var. cordifolia), pin cherry (Prunus pensylvanica), white ash (Fraxinus americana), and red maple (Acer rubrum). Conifers associated with northern hardwoods include white pine (Pinus strobus), eastern hemlock (Tsuga canadensis), balsam fir (Abies balsamea) and red spruce (Picea rubens), the latter two making up the majority of trees in the more northern and higher-elevation spruce-fir zone. Conifers also dominate lowland and riparian forests in much of the northern hardwood zone. The northern hardwoods zone was located inland and was therefore less accessible to the earliest European settlers, but colonization was nonetheless well underway by the end of the 1700s. Northern hardwood forests were generally located on hilly terrain over glacial till, and were usually farmed after more desirable coastal and riverine areas (often transition hardwoods in southern portions of the zone) were already settled. Like the transition hardwood forests, however, large acreages of farmed land adjacent to settlements were abandoned in the mid 19 th century and have since reverted to forest. Nearly all northern hardwood forest has been subject to logging, which peaked with industrial-scale clearcutting around 1900. In addition to the logging itself, slash fires during this period of railroad logging also played a role in shaping the character of many of the second-growth northern hardwood forests in the Northeast today. Like the transition hardwoods, much of the forested area of northern hardwoods is comprised of single-cohort stands with an unbalanced range of stand ages. Much of the area has a history of overexploitation and degradation, though unlike areas to the south, it is not as fragmented or as closely associated with many large human population centers. However, concerns over rapid land turnover, parcelization, and development for vacation and second homes reflect emerging challenges. In this section, we review the presettlement composition of northern hardwoods, and how those forests were regulated and organized by natural disturbance. Secondly, we review the changes that European settlement brought to the northern hardwoods, including the effects of European land use on forest composition and the changes in disturbance regime that accompanied those land uses. Finally, we consider key management questions tied to how, and whether, a presettlement model can advance conservation and sustainable production goals.
I. Vegetation and Disturbance Before European Settlement
Much of what is known about northern hardwoods forests in the period before European settlement comes from records made by colonial surveyors and explorers. In the northern hardwood zone, virtually none of these records can be considered “pre-contact,” and the ecological effects of the European settlers often traveled before them in the form of disease, war, and social disruption of groups included in (or excluded from) trade. Very little research has been done to identify plant pollen in lake sediments in the northern hardwoods zone, a process that has been valuable in identifying presettlement vegetation in pine barrens and transition hardwoods. No studies of this sort were found during our literature review. Records of ‘witness trees’ along survey lines, on the other hand, seemed to be much more available for northern hardwoods than for any other forest type. Despite their limitations, they provide valuable information about forest composition and structure.
Colonial records of witness trees exist for much of the Northeast that is not adjacent to the coast. After seaside settlements were well established, colonization of inland areas by Europeans was governed by the British government, and later, by individual states. Settlement of “new” lands was by way of the proprietary town system, whereby unsettled (or depopulated) land was surveyed and divided up into ‘towns’ with individual lots, and then families were recruited to clear and farm those lots (Cook 1989). Surveyors were competent naturalists and took accurate notes of the witness trees that they used to mark lot and town boundaries (Cogbill 2000). Their surviving records, reviewed by McIntosh (1972), Abrams and Ruffner (1995), Cogbill (2000), Cogbill et al. (2002), and others, provide a biased but systematic sample of presettlement forests of the Northeastern United States (Cogbill 2000, Cogbill et al. 2002). Such archival documents include proprietors’ records, surveyors’ field books, manuscripts (Cogbill et al. 2002), ‘warrant’ maps (Abrams and Ruffner 1995) and published records of town land surveys before settlement (Cogbill et al. 2002). In northern New England and New York, surveyors consistently identified 49 species of trees, and 79% of all recorded witness trees were beech, spruces, maples, hemlock, and birches (Cogbill 2000). Beech (mean of 32%) was the most abundant, followed by spruce (14%), maple (12%), hemlock (12%), and birches (9%) (Cogbill 2000). In a study that ranged throughout all of New England except eastern Maine, Cogbill et al. (2002) found a distinct division separating ‘northern’ and ‘southern’ vegetation, which they termed the New England tension zone. This boundary is used in our study as the border between northern hardwoods and transition hardwoods in New England, and is also consistent with Westveld et al.’s (1956) and Lull’s (1968) published vegetation maps of the Northeast. North of this border, Cogbill et al. (2002) found a mixture of beech, maple, and birch. In Maine, New Hampshire, and Vermont, beech averaged 24%, maples 11.8%, and birches 9.7% of all witness trees. Other important species in these states were oaks, pines, hemlock, and spruces (Cogbill et al. 2002). In the Catskill Mountains of southern New York at about 1800 (prior to settlement), northern hardwoods existed at middle elevations (305-914 m), bounded by transition hardwoods at lower elevations and spruce-fir at higher elevations (McIntosh 1972). In the Adirondacks, the upper limit of northern hardwoods was 980 m (McGee et al. 1999). Species composition in the Catskills was gradational according to elevation; however, beech constituted 49.5% of all witness trees, followed by hemlock (20.3%), sugar maple (12.8%), and birch (7.3%). Surveyors categorized the Catskills as beech-maple-hemlock (42% of records), beech-birch-maple-hemlock (20%), beech-maple (9%), and beech-birch-maple (5%). Similarly, presettlement forest composition in north central Pennsylvania (1765-1798) was dependent on an elevational gradient. Abrams and Ruffner (1995) found that forests resembling northern hardwoods were mapped by surveyors on the Allegheny High Plateau, whereas warmer and drier elevations of the Allegheny Front were dominated by oaks of the central hardwoods. High Plateau forests were composed primarily of beech and hemlock (Abrams and Ruffner 1995), though other northern hardwoods associates such as maple and birch were not as common in these forests. Surveyors also took notes of evidence of major disturbances that they encountered in their travels. In northern New England and New York, evidence of fire was most often mentioned in survey notes, although wind disturbances were also quite common (Cogbill 2000). One or both of these disturbances were recorded in 21% of the town surveys reviewed by Cogbill (2000). Most researchers have agreed that fire was uncommon in northern hardwoods forests relative to other forest types in the Northeast, primarily using postsettlement fire ignition data. For example, Fahey and Reiners (1981) and Lorimer and White (2003) estimated the fire return interval in northern hardwoods to approach or exceed 1000 years. Their assessment was corroborated by Parshall and Foster (2002), who found that the abundance of presettlement charcoal preserved in lake sediments (indicating the frequency and severity of forest fires) was the lowest in northern hardwoods compared to all other northeastern forest types. The relative rarity of fire in northern hardwoods was attributed to its cooler climate, ample rains, and the higher water content of the soil and deciduous vegetation (Fahey and Reiners 1981, Parshall and Foster 2002). The source of ignition of presettlement forest fires in northeastern North America has been hotly debated. Native American-origin for fires has been both invoked (Day 1953, Fahey and Reiners 1981) and downplayed (Russell 1983, Parshall and Foster 2002). However, whether or not such fires were commonplace, the greatest impact of Native Americans was most certainly concentrated near major population centers, primarily on the coast and in major river valleys (Day 1953, McIntosh 1972, Parshall and Foster 2002). Such a distribution excludes much of the northern hardwoods region. Although Native American activities may have played an important role in fire ecology in other areas (i.e., pine barrens, central hardwoods, transition hardwoods), fires were more likely to be caused by lightning where human populations were low (Fahey and Reiners 1981).
2) Reconstruction of Vegetation and Disturbance from Contemporary Stands
Most of what we know of the dynamics and natural disturbance of northern hardwoods comes from contemporary stands. Catastrophic stand-replacing disturbances are rare in northern hardwoods ecosystems (> 380 years, Boose et al. 2001), and their effects will be outlined later. However, because of the infrequency of large-scale disturbances, presettlement northern hardwood landscapes were thought to be dominated by stands in advanced developmental stages; those stands, in turn, were dominated by shade-tolerant species such as beech and sugar maple (Woods 1984, 2000; Lorimer 1989). Abiotic factors associated with elevation seem to control the competitive abilities of associated shade-tolerant conifers; hemlocks are more abundant below 500-550 m, whereas red spruce is more competitive above 500-550 m in the White Mountains (Leak 1987). Species of early-medium shade tolerance such as birches are thought to have persisted as gap-fugitives in the periods between stand-replacing disturbances, as they cannot persist in a shady forest matrix. Runkle (1981) defines a gap as the ground area under a canopy opening extending to the bases of the canopy trees surrounding the canopy opening. Most authors do not differentiate between Runkle’s (1981) definition and the area of the canopy opening itself; as a consequence, descriptions in this review generally follow opening size. Gaps can be caused by the death of one to many adjacent trees in the canopy; natural causes include one or more interacting factors such as old age, insects and fungi, physical damage, or windthrow. Northern hardwood species respond differently to the newfound availability of light and other resources in gaps (Leak 1972, Leak et al. 1987, Canham 1988, McClure et al. 2000). Trees that were already established in the understory and released from suppression at the time the gap was created are termed advanced regeneration. Trees likely to assume this strategy of regenerating in a gap are shade-tolerant enough to persist in the understory for many years. Examples of such species in northern hardwoods are beech, sugar maple, hemlock, and red spruce (Runkle 1981; Canham 1988; Bormann and Likens 1994, p. 107; McClure et al. 2000). Within this group, species have different strategies of securing light and resources that stem from their physiology (Canham 1989). Beech is extremely shade-tolerant but shows only a modest response to increased light in gaps (Canham 1988). Able to survive for an average of 11.1 years (max 30 years) in a suppressed state (McClure et al. 2000), consistent slow growth of beech in low light conditions of no or small gaps favors its increase in forests with low rates of gap formation (Canham 1988, Woods 2000). The ability to root sucker aggressively also contributes to the capacity of beech to dominate stands with low gap size and frequency. Sugar maple saplings, on the other hand, may only survive an average of 2.3 years in the understory (McClure et al. 2000), but are capable of increasing the amount of light intercepted by their leaves in response to a gap by altering their leaf architecture, incurring no significant additional metabolic costs (Canham 1988). Sugar maple can increase its canopy presence more than beech following periods of higher gap formation because of its higher growth rate (Leak 1987, Canham 1988). Research in the western parts of the Northeast has produced limited evidence for reciprocal replacement of beech and sugar maple in each other’s gaps (Woods 1979, Runkle 1981). Shade-tolerant understory species such as striped maple (Acerpensylvanicum), common in northern hardwoods, are gap-specialists that use the temporary increase in light to radically increase in size and produce seeds (Hibbs 1979), but then die back when the gap closes over them. Unfortunately, scientific studies as well as management plans often fail to distinguish between advance regeneration that establishes in the shade of an existing overstory, and shade-tolerants of the same cohort as higher strata in a stratified mixture. While this may seem an esoteric point, it can lead to confused inferences about stand demographics and age structures. Serious errors can occur in predicting undisturbed stand development, as well as the response to silvicultural intervention. The problem may be particularly vexing when diameter classes are confounded with age classes. Unlike shade-tolerants with an advanced regeneration capability, most shade-intolerant species must arrive in a gap as a seed, germinate, and outgrow the established shade-tolerant species (Bormann and Likens 1994, p. 107; McClure et al. 2000). Both paper and yellow birches face this challenge. A study in New Hampshire found that only 21% of yellow birch stems in forest gaps were present as advanced regeneration (versus >70% of stems for beech and sugar maple; McClure et al. 2000). The relative success of each life history is largely controlled by gap size; smaller gaps favor more shade-tolerant regeneration and larger gaps shade-intolerant regeneration (Leak et al. 1987, McClure and Lee 1993, Woods 2000). Brewer and Merritt (1978) showed that regeneration in windthrow gaps of only one to a few trees were unlikely to regenerate species other than sugar maple and beech in places where these species had appreciable numbers of replacement trees already established. Larger disturbances (forming gaps of more than a few trees) were required to maintain diversity of shade-intolerant trees across the landscape in their study. This suggests that if intolerant trees (such as the birches) were indeed “gap-fugitives” between infrequent stand replacement events, large gaps would be a frequent characteristic of many stands. Within gaps large enough to potentially allow many kinds of species to establish, microsite characteristics such as rotting logs, exposed mineral soil, and moisture conditions are likely to influence the regeneration success of different species (Beatty and Stone 1986, Nakashizuka 2001). The small windblown seeds of yellow birch, for example, are often successful on bare mineral soil of tip-up mounds (Carlton and Bazzaz 1998) or on rotting logs or stumps, all often present in gaps (Moore and Vankat 1986). Heavy leaf litter or the absence of soil disturbance may preclude the regeneration of this species because it has a particular regeneration niche (sensu Nakashizuka 2001) requiring soil disturbance. A third regeneration strategy, employed primarily by pin cherry (Prunus pensylvanica) is to store dormant seeds in the soil until a light gap is created overhead, which then triggers germination and growth (Marks 1974, Bormann and Likens 1994). Pin cherry is shade-intolerant, but its fast growth rate makes it a successful opportunist in larger gaps and after stand-clearing disturbances such as fires and clearcutting. In forest gaps, pin cherry is transient, eventually being overtopped by other species that began growth at the same time, but that did not initially grow as fast. This reversal of stratum layers during stand development, followed by the elimination of a less-tolerant stratum, is a classic example of dynamic stratification (Ashton and Ducey 1996). In northern hardwoods, the creation and appropriation of various sizes of gaps creates a patchwork of developmental stages throughout the forest, with the main matrix being shade-tolerant species. In the White Mountains of New Hampshire, smaller gaps and older gaps were associated with hemlock, whereas larger and newer gaps were associated with paper birch, striped maple, pin cherry, and red maple (McClure and Lee 1993). Yellow birch, red maple, and striped maple were less abundant in the old-growth forest matrix than in gaps, whereas hemlock, beech, and sugar maple showed the opposite trend. Gap age, gap size, and location within the gap all explained variation in species abundance and community structure, though gap age was the most influential (McClure and Lee 1993). Gap size was important to pin cherry and paper birch regeneration, because both needed larger gaps to successfully regenerate. Location within a gap was important to most shade-intolerant species, because regeneration success increased towards the center. Finally, the strong predictive ability of gap age on species composition indicated a consistent pattern of serial dominance as gaps in the forest canopy were reclaimed. In the hardwood forests studied by Lorimer (1989) and reviewed by Seymour et al. (2002), common gap sizes ranged from one to several trees, encompassing mean areas of less than 25 m 2 to about 125 m 2. Larger but less frequent blowdowns ranged from less than 1 to more than 3000 hectares (Lorimer 1989, Seymour et al. 2002). In old-growth hardwoods, gap formation was more pronounced, with gap sizes averaging 280-375 m 2, affecting 1% of the forest area per year (Runkle 1982, Lorimer 1989). Across the landscape, gaps closed at the same rate (1%) by upgrowth of saplings. Larger gaps in older forests provide more opportunity for capture by saplings, while smaller gaps in younger forests are often closed by the growth of neighboring trees. The mean forest area in gaps at any one time in northern hardwoods old-growth in Pennsylvania and New York was 5.8% (Runkle 1982). Lorimer’s (1989) estimate that gaps would cover less than 10% of a steady-state northern hardwood forest agreed with observations by (Runkle (1982). Lorimer (1989) calculated that 12% of forest area would be occupied by gap saplings (0-10 cm), 18% by pole-sized trees (11-25 cm), 24% by mature trees (26-45 cm), and 46% by large trees ( >46 cm dbh) in steady-state northern hardwoods. About 80% of mature hardwood forest area would be operating in a gap-dynamic state, while the remaining 20% would be first-generation recovery from catastrophic disturbances, based on a return interval of 1000 years (Lorimer 1989). Lorimer and White (2003) provided a more detailed estimate for northern hardwoods, based on catastrophic fire and wind return intervals of 1000 years (500 years combined): 3% seedling/sapling (1-15 years), 3% small pole (15-30 years), 6% large pole (30-60 years), 8% mature even-aged (60-100 years), 10% old even-aged (100-150 years), 30% transitional uneven-aged (150-300 years), and 40% old uneven-aged (300+ years). The shorter the interval between catastrophic disturbances, the smaller the percent of forest that would be in an uneven-aged, gap-dynamic state. Although the existence of a truly steady-state forest in the past can be questioned, it serves as a useful conceptual benchmark. Disturbances of intermediate intensity, such as ice storms, do not cause the catastrophic mortality that hurricanes and forest fires do; however, ice storms can increase the rate of gap formation by breaking branches, taking down scattered trees, and accelerating the rate of decline of injured trees. Ice storms can cause heavy partial damage and mortality, affecting stands across entire regions. Most northern hardwoods are prone to moderate to heavy ice storms because of their higher elevational and/or latitudinal setting (Rhoads et al. 2002). Lorimer and White (2003) reported that ice storms damage less than 50% of trees in northern hardwoods on average, but inflict heavy or severe crown loss on 21% of trees. Twelve percent of trees (those severely damaged) were judged unlikely to survive following a severe ice storm, thus creating actual gaps. Smaller openings, such as those formed from partially damaged crowns, were subsequently filled by lateral branch expansion (Runkle 1982, Rhoads et al. 2002). Ice storms damaged or killed older trees more often than younger ones because they had larger crowns, which could catch more freezing rain. Larger trees were also more likely to have been weakened by decay with age (Rhoads et al. 2002). Beech bark disease significantly weakened beech trees, making them more susceptible to crown breakage from ice buildup, which can increase effective crown weights by 100 times (Rhoads et al. 2002). Ice storm damage also increased the severity of the disease, which in turn also interfered with the ability of beech to heal from ice damage. Gaps caused by tree mortality at a variety of spatial and temporal scales are important to the herbaceous and understory diversity of the northern hardwood zone. Moore and Vankat (1986) found that herbaceous species richness and percent cover were largely dependent on the presence of gaps. The greater cover values of herbaceous species in gaps probably reflected the increases in photosynthetically active radiation (PAR) and moisture in gaps. Richness was related to heterogeneity of nutrients and soil characteristics across gap microsites, such as tip-ups and pits, but declined with subsequent succession in gaps as light availability declined (Moore and Vankat 1986). Moore and Vankat’s (1986) findings about richness in gap succession mirror the downward trend in species richness documented during old-field succession in transition hardwoods by Howard and Lee (2003). Alternatively, experimental gap creation of small gaps (< 150 m 2) on the Allegheny Plateau found that species richness of the herb layer did not change appreciably over the first four years following gap creation (Collins and Pickett 1987, 1988). Their studies suggested that larger gaps than 150 m 2 may be necessary to increase diversity, that soil disturbance such as tip-ups may be necessary (trees were sawn down in their studies), or that plant response to the environment may take longer than four years. Goldblum (1997) found that for understory plants in a New York northern hardwoods forest, there were no gap specialists among understory herbs; rather populations already present increased when light and other resources were made available in gaps, and then returned to low levels when gaps closed. Like Collins and Pickett (1987), Goldblum (1997) did not find richness differences in gaps. Taken together, these studies indicate that the size and origin of gaps is important to species response, and that a variety of gap sizes may be required to maintain diversity at the landscape scale.
3) Observations on Stand-Replacing Disturbances
Although gaps of a variety of sizes are common in northern hardwood stands, and particularly in those in advanced development, northern hardwood stands occasionally were subjected to natural stand-replacing disturbances such as fire and windstorms. The former (addressed in Part 1) probably had return frequencies approaching or exceeding 1000 years in northern hardwoods (Fahey and Reiners 1981, Lorimer and White 2003). Other than those resulting from hurricanes, stand replacing disturbances due to tornadoes (Peterson and Pickett 1995) or other catastrophic windstorms (e.g., Canham et al. 2001) are rare enough so that frequencies and extent in northern hardwoods are difficult to estimate precisely. Northern hardwoods closer to coastal areas probably were disturbed more frequently than that by hurricanes (Woods 1984, Boose et al. 2001, Lorimer and White 2003). In a study of 67 hurricanes over the last 400 years in New England, Boose et al. (2001) characterized the effect that these may have had on forest vegetation. Hurricanes and tropical storms generated disturbance in forest ecosystems hundreds of kilometers inland, though the frequency of damage decreased as distance from the coast increased. Northern hardwoods grow where F2 (Fujita scale) hurricane disturbances occur on average more than 380 years apart (Boose et al. 2001); transition hardwoods close to the coast experience F2 hurricanes more frequently. F2 hurricanes potentially can flatten large areas of forests; for example, the 1938 hurricane destroyed 250,000 hectares of timber across New England (Foster 1988). However, lesser hurricanes and tropical storms may enhance the production of gaps in northern hardwoods by toppling older, weaker trees and causing crown damage similar to that described for ice storms (Boose et al. 2001). The potential effect on stands dominated by such trees, and on landscapes dominated by such stands, must remain speculative. In between intervals of catastrophic disturbance, early-successional species probably persisted in moderate-sized openings (large gaps formed by the death of many trees; greater than 500 m 2 according to Lorimer 1989) (Lorimer and White 2003). Large disturbances such as hurricanes and fires subsequently reduced the frequency of gaps in affected areas for many decades by removing most of the older, larger trees (Lorimer 1989) and allowing for the development of an even-aged cohort that persisted for 150-300 years before returning to an uneven-aged state (Lorimer and White 2003). This pattern of development closely parallels the standard stand development model (Oliver and Larson 1996), with a stem exclusion period lasting a century or more and repeated episodes of understory reinitiation leading to a gradual transition to multicohort dynamics. According to Merrens and Peart (1992), 1938 Hurricane-devastated northern hardwood stands in the White Mountains were dominated initially by pin cherry, a pattern also observed following clear cutting of northern hardwoods (Bormann and Likens 1994, Heitzman and Nyland 1994). Pin cherry was overtopped and eliminated by 50 years following the disturbance, with dominance passing to beech, sugar maple, and yellow birch (Merrens and Peart 1992). Although shade-tolerant species had reasserted dominance by 50 years post-disturbance, the hurricane had long-lasting effects on the character of the vegetation: relative to sheltered areas that remained relatively unaffected by the hurricane, low- to medium-tolerant species such as white ash increased, and growth rates of all species increased, with greater sustained increases for increasingly shade-intolerant species (Merrens and Peart 1992). The legacy of devastating hurricanes on forest composition and dynamics can last hundreds of years (Henry and Swan 1974, Foster 1988, Merrens and Peart 1992) until the last remnants of single-cohort stands have made the transition to multi-cohort dynamics.
II. Postsettlement Vegetation and Historic Land Use
European colonization of the Northeastern United States has had tremendous impacts on the forested landscape. In areas that were historically northern hardwoods, these impacts included the effects of land clearing for agriculture and subsequent agricultural abandonment by the mid-1800s, the effects of industrial-scale logging around 1900, and the effects of human-introduced non-native pathogens.
1) Agriculture and Abandonment
Early European settlers were primarily farmers, and the forest of the Northeast was a barrier to be cleared away. By 1850, more than 60% of forests over the entire New England region had been cleared for agriculture (Cogbill et al. 2002), and the remainder were heavily exploited for building materials and firewood (Foster 1999). By the mid-1800s, coincident with the industrial revolution in the cities and the opening of midwestern states by canal and railroad, agricultural land in the Northeast was abandoned (Foster 1999, Howard and Lee 2002). Much of the present forest land in New England is second-growth that re-established on abandoned pastures or cultivated land (Hannah 1999). Farm abandonment led to the formation of large areas of old-field white pine, birch, and maple (Whitney and Foster 1988), white pine having especially high densities in the transition hardwoods zone (see Part 2). Present-day forests, though revegetated for more than 100 years, still reflect the legacies of previous land use, and have not returned to prior presettlement composition or dynamics (Foster et al. 2003). Formerly plowed areas retain mixing of soil horizons to a depth of 10-30 cm that can persist for hundreds of years (Foster et al. 2003). Ecologists in Europe have reported that agricultural effects on soil, even from more than 1700 years before present, still explain major variation in forest composition and dynamics (Dupouey et al. 2002). The ground surface was also homogenized by agricultural use; tip-up mounds and pits normally associated with windthrow in natural forest systems are lacking due to plowing, or simply the long-term absence of large trees susceptible to windthrow (Foster et al. 2003). Single-cohort stands have not yet had time to break up. As a result of the absence of large trees, these forests also lack large old trees, large dead snags, and large coarse woody debris (Foster et al. 2003), which provide habitats for a diversity of woodland creatures , including insects (Chandler 1987, 1991; Chandler and Peck 1992), fungi (Vujanovic and Brisson 2002) , and lichens (Selva 1987, 2003), in natural systems. It is not inherently the single-cohort nature of the stands, but the fact that there was no previous stand to leave these structural legacies, that makes the legacies rare. Small invertebrates (Foster et al. 2003) and clonal understory plants (Bellemare et al. 2002) can be especially slow to recolonize large areas that had been extensively disturbed by agriculture. Foster et al. (2003) found that exposure to a few natural disturbances such as hurricanes, fires, or major windthrows are not enough to ‘undo’ the legacy of human impact on forested ecosystems. In addition to physical characteristics of forests, composition is also highly dependent on agricultural history. Because a large percentage of land that is now forested was abandoned between 1850 and 1900, a large percentage of present-day forests are in the same stages of succession (Litvaitis 1993, Hall et al. 2002). As a result, long-lived, late-successional species such as beech, sugar maple, hemlock, and yellow birch have declined compared to presettlement times, and shorter-lived, early- to mid-successional species such as red maple, paper birch, poplars (Populus spp.), cherries (Prunus spp.), white pine, and white ash have increased (Hall et al. 2002). In Massachusetts, red maple increased from 7% to 27% relative density from presettlement to present (Hall et al. 2002). (For a complete review of the effects of agriculture on postsettlement forest composition and dynamics, see Part 2.) The result of this ‘successional wave’ is a homogenization of landscape-scale age structure in the Northeast (Hall et al. 2002), that today favors mid-successional animal species. Currently, populations of early- and late-successional animal species are low and many are declining (Litvaitis 1993, Foster et al. 2002). (For a review of the effects of postsettlement land use on animal populations, see Part 2.)
2) Forest Harvesting, High Grading, and Fire
The northern hardwoods forests have been and still are a tremendous resource for the forest harvesting industry. Today many silvicultural systems are employed on many scales, and many provide the basis for a sustainable and sustaining forest. However, forest harvesting from early settlement through at least the first part of the 20 th century followed patterns of questionable sustainability. A combination of indiscriminate and uncontrolled clearing, with partial cuts that high-graded the biggest and best trees from vast acreages, provided a devil’s brew for the northern hardwood forest. While some authors have suggested large-scale disturbances were relatively rare in northern hardwoods prior to European settlement (Lorimer 1989, Seymour et al. 2002, Lorimer and White 2003), northern hardwoods were heavily logged during the late 1800s and early 1900s ( Belcher 1981, Schwarz et al. 2001). This large-scale logging of previously inaccessible areas was made possible by custom-built logging railroads, and affected forests from the White Mountains of New England (McClure and Lee 1993, Schwarz et al. 2001) to the Alleghenies of northern Pennsylvania (Abrams and Ruffner 1995). Extensive slash fires, often attributed to cinders from steam-powered locomotives, often raged across recently-logged landscapes (Fahey and Reiners 1981, Abrams and Ruffner 1995). Such scenes motivated the first forest conservation movements in the Northeast, and led among other things to the passage of the Weeks Act and the establishment of the eastern National Forests. Turn-of-the-century logging operations focused on virgin red spruce (Schwarz et al. 2001), sugar maple, yellow birch, and white pine, which were sawn into boards or used to construct high quality hardwood furniture. Beech, being a low-quality hardwood, was seldom logged in interior forests. However, beech may have been more heavily utilized close to rail lines, as many locomotives were wood-fired. Hemlock was extensively cut for the leather tanning industry, which used only its bark (McIntosh 1972, Abrams and Ruffner 1995). As a result of extensive clearcutting and burning during the period from 1870-1930, northern hardwood landscapes are, like their transition hardwood counterparts, often dominated by middle-aged stands in an unbalanced distribution. These stands, in turn, are often stratified single-cohort mixtures that lack the fine-grained horizontal gap structure associated with more advanced developmental stages. Following previous high-grading, many of these forests exist in a degraded state. For example, Westveld et al. (1956) reported that many northern hardwood forests lacked diversity and much currently merchantable timber. With the passage of time, merchantable timber has developed. However, the erroneous belief that many of these forests (which, being stratified single-cohort mixtures, have complex vertical structure and a reverse-J diameter distribution) are uneven-aged, may have led to further unintentional high-grading of some stands. By contrast, in managed stands that have not been high-graded, continuing improvement in volume and grade is possible (Leak and Sendak 2002). Secondary forest development following clearcutting in northern hardwoods is strongly influenced by soil characteristics (Leak 1991).Typically, a young stand in the stem exclusion stage will be dominated by fast-growing, relatively shade-intolerant species. On fine till, the early dominants include yellow birch, pin cherry, and paper birch, while red maple, paper birch, pin cherry, and yellow birch dominate on sandy till (Leak 1991). In Vermont, Hannah (1999) described canopy domination by red maple, paper birch, aspen, white pine, and beech following major stand-replacing disturbances. The relative abundances of these early dominants then decrease over time, as shade-tolerant species from lower strata increase and eventually dominate the stand (Leak 1991). Leak (1991) groups the species inhabiting the late-developmental northern hardwoods community into three major classes: 1) Dominating climax species are the most shade-tolerant, but can also grow in a wide range of microsite conditions. On either fine or sandy glacial till, beech fulfills this role, dominating the understory and growing into the canopy, possibly even in full shade. 2) Stable climax species are also shade tolerant but maintain a constant and abundant (more than 15%) presence in the forest. Sugar maple is an example of a stable climax species on fine till soils, but it is not as successful on sandy till. 3) Minor climax species are tolerant or somewhat tolerant of shade. They are able to maintain a small but constant presence in the forest at less than 15%. Examples of minor climax species are hemlock and red maple on fine till and hemlock on sandy till (Leak 1991). Early-successional species can persist at low levels within a late-developmental forest by occupying wet or shallow-soiled sites and also regenerating in areas of natural disturbance. Paper birch, yellow birch, and red maple can persist in northern hardwoods (Leak 1991). Paper birch, however, requires larger canopy gaps than yellow birch to successfully regenerate (Leak 1991). Pin cherry, a temporary successional species, completely dies out and is present only as buried seed, awaiting the next major disturbance. The pattern of development described above applies to present-day clearcuts in the northern hardwoods region. In the late 19 th century and early 20 th century, however, succession was often modified by fire. In the Alleghenies of northern Pennsylvania, clearcutting coupled with recurring fires was responsible for the elimination of fire-intolerant beech and hemlock regeneration, as well as the elimination of most white pine (including seed source) from the region (Abrams and Ruffner 1995). Consequently, the forest now has elevated percentages of red maple and red oak, both of which can stump-sprout following fire, and shows no evidence of returning to its beech-dominated presettlement condition (Abrams and Ruffner 1995). A similar conclusion was reached by McIntosh (1972) about the Catskill Mountains of New York, though the role of slash fires is unclear. In most portions of the northern hardwood zone, fire-intolerants such as beech and hemlock are relatively common; however, it is likely that burning did modify local forest regeneration patterns. Parshall and Foster (2002) recorded increases in the deposition of charcoal in lakes in all forest regions in New England after European settlement (denoted in lake sediment cores by increases in agricultural and grass pollen from land clearance). Whether by logging or fire or a combination of both, beech and hemlock pollen in northern hardwoods lake sediments declined over the same period as white pine increased (Parshall and Foster 2002). A survey of twentieth century fires in Maine and New Hampshire by Fahey and Reiners (1981) revealed that most fires started in cutover areas where dry fuel was most abundant.
3) Introduced Diseases: Beech Bark Disease
In the 20 th century, northern hardwood forests have come under assault from a variety of non-native predators and pathogens that threaten to permanently alter species composition. For example, the hemlock wooly adelgid, an insect that is currently defoliating hemlock trees in the transition hardwoods zone (see Part 2) may eventually impact hemlock in northern hardwoods forests as it moves north and west at 30 km per year; it currently does not appear to be limited by climate at the edges of its range (Orwig et al. 2002). Beech bark disease has probably created the largest impact of any pathogen throughout the range of northern hardwoods, as beech was the most abundant species in northern hardwood forests in presettlement times and is also the most shade-tolerant. Beech bark disease was imported into Halifax, Nova Scotia, on ornamental European beech trees around 1890. Since then the disease has spread into the United States through Maine as far south as Tennessee and as far west as Michigan (Latty et al. 2003). In the Adirondacks of New York, the disease reduced vigor and increased mortality to the extent that beech nut production of the forest has declined 37%. This decline has dire consequences for the populations of mammals, birds, and other organisms that depend on beech mast (Latty et al. 2003). It also negatively impacts tree grade, and as a result restricts silvicultural options in heavily infested stands. The disease is transported from tree to tree by a scale insect, the woolly beech scale (Cyptococcus fagisuga), as it feeds on the bark. The disease is caused by two ascomycete fungi species: the introduced species Nectria coccinea var. faginata and the native Nectria galligena. Latty et al. (2003) documented that the severity of the fungal disease was related to the degree of scale infestation, and that scale infestation depended on a number of interplaying factors. Larger trees have more bark area, which can support larger scale populations. Secondly, larger, older trees in old-growth forests have a higher nitrogen concentration in their bark, making them more nutritious and increasing the population growth rate of scale insects. Lastly, trees in forests that have been under attack for a longer period of time are more diseased; for example, northern hardwoods in Maine were more affected than forests in New York (reached in 1950s; Runkle 1990), which were more affected than forests in Michigan (reached recently; Latty et al. 2003). With the decline of beech, other tree species have increased in importance in the northern hardwoods forests of New York. Runkle (1990) documented that small stems (< 10 cm in diameter) of sugar maple, red maple, and red oak had reduced mortality rates and larger stems had a higher growth rate, indicating an increase in available light and other resources in the forest understory with the decline of overstory beech. Eastern hemlock, already at 60% density in the Niles Huyck Preserve ( New York) was the primary beneficiary, filling openings left by dead and dying beech trees (Runkle 1990). In other areas of the Northeast, such as in New Hampshire, root suckers from beech trees appeared to increase in density and vigor following the demise of overstory beech (Rhoads et al. 2002). Beech stems less than 6-10 cm dbh appeared to have a lower incidence of the disease (Runkle 1990, Rhoads et al. 2002, Latty et al. 2003), probably for the reasons discussed in the preceding paragraph. While in New York mortality was primarily by disease and senescence rather than by physical disturbance (Runkle 1990), in the White Mountains of New Hampshire beech bark disease interacted with elevation and climate to become more deadly. These northern hardwoods, by virtue of their more northerly location and mountainous location, were more prone to freezing rain than southern New York. Rhoads et al. (2002) found that beech bark disease made infected beech trees more prone to damage by ice storms in this area. The disease hampered beech’s normal load bearing capacity by reducing bark elasticity, allowing more snapping of ice-loaded branches on diseased trees than on disease-free trees (Rhoads et al. 2002). The fungus also interfered with the bark’s ability to repair wounds following an ice storm, creating a feedback that increased the severity of the disease. Two years following a 1998 ice storm, beech bark disease was more advanced on trees that had been ice damaged (Rhoads et al. 2002).
III. Management Challenges and Recommendations
1) Features Affecting Biodiversity
Species richness among vertebrates in northern hardwood forests follows a bimodal distribution, with diverse assemblages in regenerating stands (< 10 years post disturbance), dropping off in pole stands, and then increasing to a maximum in mature and over mature stands (DeGraaf et al. 1992). Because much of the northern hardwood type is characterized by mid- successional stands, considerable attention has focused on assuring an adequate representation of the habitat features that are found in young and old stands (e.g., Tubbs et al. 1987, DeGraaf and Yamasaki 2003). Early-successional habitats offer a variety of herbaceous ground cover and fruit-bearing shrubs (e.g., Howard and Lee 2002). In northern hardwood forests, tenure among vertebrates in early-successional habitats varies and it may be extremely short for some species. For example, olive-sided flycatchers (Contopus cooperi) and eastern bluebirds (Sialia sialis) may colonize a site 1-2 years after a disturbance but abandon it after only 2-3 breeding seasons (DeGraaf and Yamasaki 2003). In addition to this very discrete period of suitability, both species rely on relatively large canopy gaps (Costello et al. 2000) that may be associated with beaver (Castor canadensis) flowages or other disturbance that kills many trees. As a result, these species do not exploit openings created by single or small group-selection harvests. In the Northeast, mammals tend to occupy a greater diversity of habitats in comparison to breeding birds. Over 85% of the 60 species endemic to the region use various combinations of forest types and seral stages (Scanlon 1992). However, nearly all mammals in this region use early-successional habitats, and about 20 have shown a preference for such habitats (Fuller and DeStefano 2003). Several species are indeed tightly associated with young forests and their abundance is directly dependent on the dense understory vegetation characteristic of regenerating stands. Litvaitis (2001) considered lagomorphs as such a group of “early-successional obligates”. The importance of this group is based on their role as a major prey of a number of carnivores, and changes in lagomorph abundance result in functional and numeric responses by their predators (Keith et al. 1977, Litvaitis 1993). In northern hardwood forests, snowshoe hares (Lepus americanus) are essentially the only lagomorph present (DeGraaf and Yamasaki 2001). The recent designation of lynx (Lynx canadensis) as a threatened species in the Northeast (Nordstrom et al. 2000) has highlighted the importance of snowshoes in this region. Because the demography of lynx is closely associated with the abundance of hares, considerable attention is now being direct toward maintaining adequate hare habitat to assure lynx viability (Ruggerio et al. 2000). From the above descriptions, it is clear that young seral stages are essential in northern hardwood forests. However, vertebrate species richness in mature, over mature, and stands with all age classes, exceeds that found in early-successional stands (DeGraaf et al. 1992). One of the important features of mature stands is the prevalence of decay in standing trees and woody debris. In northern hardwood forests, 41 species of birds and mammals nest, den, roost, or forage for insects in trees with bole cavities (Tubbs et al. 1987). Although many cavity-nesting species forage on the boles and limbs of sound trees regardless of size, most nest in relatively large trees (> 45 cm DBH) with suitable decay (DeGraaf and Shigo 1985). Although there is information on the relationship between northern hardwood forest development and the biodiversity of vertebrates, little is known about forest development and other groups of organisms. Organisms that require large, living trees or coarse woody detritus may be more abundant and diverse in older stands experiencing some mortality of canopy trees. Indeed, some groups of beetles, such as the Pselaphidae (Chandler 1987) and Leiodidae appear to be more abundant in old-growth forests as compared to younger managed forests (Chandler and Peck 1992). At least one Leiodid beetle has been cited as a possible indicator of old-growth northern hardwoods (Chandler and Peck 1992) and the species richness of beetles that feed on fungi under bark was higher in an old-growth stand than in a 40 year-old managed stand (Chandler 1991). The species richness of Calicioid lichens and fungi, which are often epiphytic on the bark of large, slow-growing trees, increases over time, with older stands supporting more rare species (Selva 1987, 1994, 2003). Maintaining species richness in these insect and fungal groups may depend on the protection of some existing old-growth northern hardwoods and management of some older stands to allow for a sufficient number of large trees and coarse woody debris.
2) Contemporary Issues and Considerations
Today’s northern hardwood forests are different from presettlement northern hardwoods in a number of ways, even though the tree species that make up this community are largely the same. Today’s forests are much younger; large-scale clearing and burning around the turn of the 20 th century ensured that in many areas the oldest second-growth trees are no more than about 100 years old. Agricultural fields and pasture abandoned in the 1850s may support forests no more than 150 years old. Third-growth forests, which were logged and regenerated again in the twentieth century, are even younger. As a result, much of the range of northern hardwoods is dominated by middle-aged stands. While these stands are often stratified, even-aged mixtures with complex vertical structure, the lack of horizontal structure both within many stands and across many landscapes presents some challenges. While sheer volume of snags or downed logs is rarely a challenge, the absence of large standing trees, large snags and large downed logs may represent a habitat and biodiversity issue. In northern hardwoods, certain groups of fungi, lichens, and insects are dependent on either large living trees or coarse woody debris, and their abundance and diversity tend to be higher in old-growth stands ( Chandler 1987, Selva 1987, Vujanovic and Brisson 2002). The 14.6 million hectares of northern hardwood forest in the Northeast have lower percentages of later-successional beech and sugar maple, and higher densities of low- to mid-tolerant red maple, paper birch, and white ash than presettlement forests (McIntosh 1972, Seymour 1989, Abrams and Ruffner 1995, Hall et al. 2002). Reduced landscape-scale habitat heterogeneity, and modified stand composition and structure, are both results of a “mid-successional wave” within the northern hardwood zone.
Because most northern hardwood forests are not as fragmented, parcelized, or as heavily associated with urban centers as transition or central hardwoods, more management options are theoretically possible in this forest type. While grade is often an issue, the resource has recovered to merchantable size across the landscape, with 58% of stands in the sawtimber size class over a decade ago (Hornbeck and Leak 1992). A range of silvicultural techniques and tools exist to carry out silvicultural objectives (Mattson 1988). At the stand scale, silvicultural systems, regeneration methods, and intermediate treatments must reflect the objectives of the landowner and respect the needs of society. Forestry as a profession has long recognized the dual goals of sustainable yield and the enduring maintenance of biodiversity. Within the northern hardwood zone, a key question is whether forest management can maintain critical forest functions (productivity, nutrient cycling, water cycling) and the full array of native organisms present in presettlement forests, while helping to counter the loss of landscape-scale diversity in age and stand structure produced by more than 200 years of prior land use. Recently, some authors have argued that silviculture designed to mimic natural disturbance may be the most effective method of protecting plants, animals, and natural ecosystem processes while extracting commodities from forests (Mitchell et al. 2002; Lorimer and White 2003). Depending on ones’ perspective, this assertion may be either tautological or an untested hypothesis. In either case, the debate is but a new chapter in a long dialogue. Proponents of a new approach have asserted that forest management designed to mimic natural processes requires specific scientific knowledge of stand development and a comprehensive understanding of how natural and artificial disturbances affected forests (Franklin et al. 2002). The failure to adopt a natural disturbance paradigm, some suggest, is due to the lack of specific, quantitative guidelines to design and manipulate natural patterns (Seymour et al. 2002). In the following section, we briefly review recent literature on even-aged and uneven-aged management, and ask whether a “presettlement model” demanding a conversion from even-aged to uneven-aged management is well-founded, would achieve ecological or social objectives, or is feasible. We do not seek to dictate choices but rather to frame them.
Even-aged management has traditionally focused on regenerating, tending, and harvesting stands that are dominated by a single cohort through a repeatable cycle. Sustainability stems from, and requires, the successful regeneration of stands and the maintenance of the productive capacity of the site. An even flow of goods, benefits, and services (on ownerships in which that is an objective or constraint) arises from the deliberate scheduling and regulation of the rate of harvest. One major benefit to using even-aged management to harvest and regenerate forests is economic: even-aged cutting is cheaper (Smith 1988, Tubbs 1988). Administration and supervision of the cutting operation are less complex, more wood can be harvested from the site at one time, and the costs of building roads and log landings are lower because they require little or no long-term upkeep. This can ease labor constraints in typical cash-strapped natural resource management entities, be they governmental, for-profit, or non-profit. It can also lead to reduced landscape-scale impacts from the engineered infrastructure (especially roads and landings). Major even-aged regeneration methods include the clearcut, seed-tree methods, and a wide variety of shelterwood approaches. Since the 1960s, both clearcutting and shelterwood cutting have been used to regenerate northern hardwoods (Wang and Nyland 1996). Clearcutting produces an even-aged cohort of new growth, with the new tree canopy closing 10-15 years following the logging event (Wang and Nyland 1996). By 10-15 years after clearcutting in New York, Wang and Nyland (1996) observed that herbaceous forage had declined due to shading, and by 17-20 years, height to the base of the live canopy was 3-4 m off the ground. No new species of trees arrived on the plots following canopy closure at about 10-15 years; change in canopy composition thereafter was based on the elimination of short-lived and shade-intolerant species. Shelterwood cutting was found to produce better rates of seedling establishment in northern hardwoods, but this did not necessarily translate into more wood produced long-term (Nyland et al. 2000). Nyland et al. (2000) compared three northern hardwood clearcuts in upstate New York, ranging in size from 3 to 26 hectares, with comparable sized northern hardwood shelterwood cuts in the Adirondacks. Sites logged with the shelterwood system had higher maximum stem densities than the clearcut sites at 10 years post-cut, but stems grew more slowly. By 25 years the clearcuts had slightly more average basal area per hectare than the shelterwood cuts (26 vs. 22 m 2 per ha; Nyland et al. 2000). Ray et al. (1999) concluded after a 26-year study that the shelterwood technique could successfully regenerate northern hardwood species following cutting of old-growth northern hardwood stands, providing certain other control treatments also accompanied the logging. Sites were thinned to 35-65% canopy cover on the first cut, but dense beech advanced regeneration was previously controlled with herbicides, and deer browsing of new seedlings was controlled by hunting. Most new stems (arriving as seeds) were established by five years after the first cut, and, like Nyland et al.’s (2000) study, canopy closure occurred by 10 years. Between 10 and 26 years, the new stand was in the process of self-thinning. The regeneration method used greatly affects which species will dominate the regenerating northern hardwood forest. Leak et al. (1987) summarized the general trends expected in northern hardwoods with each silvicultural technique: Clearcutting favored shade-intolerant birch regeneration. Open shelterwood (residual crown of 30 to 50%) favored mid-tolerant yellow birch and red maple regeneration. Dense shelterwood (residual crown of 80%) favored shade-tolerant sugar maple and beech regeneration.
Even-aged management is not a particularly popular method of forest harvesting with the public at large, primarily because it creates an aesthetically displeasing landscape (Wang and Nyland 1996). The long-standing alternative, uneven-aged management, has a number of putative advantages beyond aesthetic ones, including protection of wildlife habitat, reduction of soil erosion, protection of permanent seed sources, and management of pests (Smith 1988, Tubbs 1988). However, uneven-aged management may lack some of the efficiencies of even-aged approaches, and it is harder to regenerate shade-intolerant commercially valuable species in limited canopy openings (Wang and Nyland 1996). Furthermore, some bird species do not find suitable habitat in stands managed using uneven-aged approaches (Costello et al. 2000). According to Leak and Filip (1975), there was little difference in harvest-cutting costs among single-tree selection, group selection, or patch selection, and three uneven-aged methods that could be used to produce a range of gap sizes. However, harvesting technology has changed considerably in the last three decades, so revisiting this question might be fruitful. Recently, some have argued that uneven-aged management is the most similar to presettlement disturbance regimes in northern hardwoods (Seymour et al. 2002, Lorimer and White 2003). In particular, large clearcuts emulate large catastrophic disturbances. While most studies are reconstructive and rely on imperfect evidence, most agree that the frequency of such events in a given northern hardwood stand was at most once per several centuries (Fahey and Reiners 1981, Boose et al. 2001, Lorimer and White 2003). Uneven-aged management, using a within-stand area regulation approach regenerating 0.7-1.3% of the stand area per year, might approximate natural canopy turnover (Lorimer 1989, Seymour et al. 2002). Seymour et al. (2002) developed an approach that is innovative, though not without its challenges. Basically, Seymour et al. (2002) collated estimates of scale and frequency for disturbances generating complete mortality within a group of trees. They pooled studies from both the northern hardwood and the Acadian mixed-conifer forests, and plotted scale against frequency on a log-log chart. The scale-frequency pairs were grouped, surrounded by ellipses, and then bounded by eye with a straight-line relationship. Seymour et al. (2002) suggested that an index of distance from this straight line (which, in terms of raw scale and frequency, translates to a power function) could be used to judge the conformance of silvicultural strategies to “natural” disturbance regimes. Their results suggest that a stand-replacing disturbance affecting 20 ha of contiguous forest would only occur in the same spot on average every 347 years. This largely agrees with the studies cited above, because Seymour et al. (2002) depended on those same studies. By contrast, common even-aged systems might allow a clearcut on that area of forest using a 50-100 year rotation. The Seymour et al. (2002) approach is interesting, and suggests quantitative guidance, but it presents methodological challenges. First, any difficulties in interpretation of survey data, or reconstructive studies on highly selected old-growth relicts translate directly into the results; the incorporation of multiple studies does not cancel their similar biases. Seymour et al. (2002) acknowledge the highly selected nature of the study sites as a potential limitation. Second, surrounding data with ellipses by eye, and setting a line against those ellipses by eye, does not produce an objective result with known or knowable error properties. One could easily imagine two research groups using the same data, and coming to different conclusions; on a log-log scale, the results could be dramatic. However, Seymour et al. (2002) do not plot or present the raw data, so it is even more difficult to ascertain the latitude for subjective influence. Third, decisions about which data points to include can dramatically impact the results. For example, Seymour et al. (2002) exclude an 80,000 ha fire referenced by Lorimer (1977), on the grounds that it was probably human-caused, in addition to other unnamed “anomalous events.” Excluding this large fire was consistent with Hunter’s (1996) reference point of “natural” as lacking human influence, but it also provides evidence, regardless of ignition source, of a forest type and structure just before large-scale settlement that could and did sustain a large-scale crown fire. Including just that one data point would completely upset the results of the study. Should one quibble, for example, with the premise that the complete lack of human influence represents a reasonable reference point, the entire exercise could be thrown into doubt. Finally, scales of several to tens of hectares – precisely the scales about which a decision between even-aged and uneven-aged methods are likely to be made -- are poorly represented in the data. Seymour et al. (2002) consider the possibility that this gap is real, i.e. that such disturbance sizes are rare, in which case interpolation by a line would be inaccurate and would overestimate frequency. They do not address the possibility that the gap is caused by site selection bias and consequent censoring of the data, in which case a linear interpolation could result in an underestimate of the frequency. These challenges, taken together, do not necessarily invalidate the specific results of the Seymour et al. (2002) study, but they do suggest that if the approach is to live up to its promise, then serious attention to methodology (e.g. Lorimer 1985) is warranted. If we accept the concept that regenerating 0.7-1.3% of the landscape per year approximates natural canopy turnover (Lorimer 1989, Seymour et al. 2002), an implied maximum tree age within a managed stand will be 70 to 140 years, perhaps somewhat longer than conventional rotation lengths in even-aged silviculture. The irrepressible law of compound interest dictates that such an approach has very high financial costs. In private lands subject to rapid ownership turnover, the accumulated growing stock becomes a tempting source of capital to retire initial investment costs. These factors suggest that such an approach, while biologically sustainable, is unlikely to actually be sustained on many lands in the northern hardwood zone. If the maximum tree age is reduced, the return frequency must be increased; but this may push the management regime out of the acceptable zone as defined by Seymour et al. (2002). Uneven-aged management, in and of itself, does not necessarily conform to a “presettlement” model. There are a number of challenges to successful uneven-aged forest management in northern hardwoods. Beech bark disease has significantly lowered the value of American beech as a commercial species by killing most of the larger individuals (Bohn and Nyland 2003). The death of adult stems leads to prolific sprouting from beech root systems, causing high understory beech densities that compete with regeneration of more valuable species such as sugar maple (Bohn and Nyland 2003). In cases where understory beech densities are high before logging, some site preparation may be necessary. An established uneven-aged stand must continuously produce not only the same volume of wood that is extracted per unit time, but also a consistent species and grade mix, in order to create sustainable yield. Species composition can be regulated in part by gap size (Leak et al. 1987): group selection with groups and patches up to 0.81 hectares (2 acres) favors initial dominance by birch regeneration, while single-tree and small group selection favors sugar maple and beech regeneration. The forest must have enough trees in the smaller size classes (pole) to grow into the larger size classes (sawtimber) over time, but not be so dense that growth is hampered (Leak and Solomon 1975, Solomon 1977). While most approaches to structure and harvest regulation in uneven-aged stands have focused on classic “q ratio” approaches (Leak 2003), other approaches may provide more flexibility in describing stand structures (Liu et al. 2002).
6) Should Stands Be Converted to Uneven-Aged?
As discussed above, following a “mid-successional wave,” the age and structure distribution of modern northern hardwood landscapes does not mirror the likely distribution in a presettlement landscape. Some authors (e.g. Franklin et al. 2002) have argued that foresters should eschew even-aged management to mirror natural processes, whereas others (e.g. Seymour et al. 2002) have suggested that current stand boundaries should be ignored as a more fine-grained harvesting approach is imposed on the landscape. This would, in effect, convert large portions of the landscape from even-aged (or single-cohort) stands to uneven-aged (or multicohort) stands over time. To convert an even-aged stand into an uneven-aged one requires many decades (Lorimer 1989, Nyland 2003). Silviculture must increase the rate of canopy break-up rather than waiting for the trees in the initial even-aged cohort to reach senescence. The resulting forest should be similar in structural character to presettlement forests except that the maximum diameter and basal area will be smaller (Lorimer 1989), especially if the gap frequency and maximum tree age are at or beyond the short end of the natural range. During the transition period, one must successfully partition the cut of older trees so that some remain, create gaps large enough that new cohorts establish at regular intervals, maintain species diversity (especially of shade-intolerant species), safeguard the health and vigor of old trees, sustain seed production until younger cohorts mature, protect all age classes from injury, and regulate the number and interspersion of trees in each cohort as multi-aged stand develops over as much as a century (Nyland 2003). Once established, a selection system can be sustainable on the same forested site (Nyland 2003). Many businesses are theoretically sustainable as going concerns, but fail during startup or periods of transition. Likewise, the transition between two potentially sustainable silvicultural systems holds many pitfalls for sustainability. Nyland (2003) provides some of the most specific guidelines available for converting even-aged stands to uneven-aged stands. He suggests conversion by one of two methods: 1) regularly scheduled uniform partial cuts similar to heavy thinning or light shelterwood to establish new seedlings – for eventual single tree selection, or 2) periodic patch cutting (1-2 tree heights in diameter) with thinning to establish clusters of seedlings, eventually supporting group selection. The first would favor more shade-tolerant species, and the second would allow more regeneration of shade-intolerant species. A mix of these two strategies would increase spatial heterogeneity, an important characteristic of natural uneven-aged forests. Success in conversion depends largely on the success of sustaining viable regeneration of seedlings and saplings over many decades. For this to occur, the stand must have freedom from intense herbivory, lack of interference by undesirable woody or herbaceous plants, and protection from fire and drought (Nyland 2003). The overstory trees must produce seeds at the appropriate thinning intervals. The eventual forest should have 3-4 age classes with consistent intervals between them and each occupying a similar amount of space (Nyland 2003). This allows sustainable cuttings that perpetuate the system. Although focused primarily on the transition hardwood zone, Kelty et al. (2003) provide a review that highlights many of the issues also encountered in conversions in the northern hardwood zone.
7) Stand and Landscape Legacies
Three major structural features recur as being relatively common in the pre-European northern forest landscape, and relatively uncommon today: large live trees, large snags, and large downed coarse woody material. These three features are linked by the simple premise that it is impossible to have large snags or large downed woody material in stands that have not had large live trees. Because of the land use history in the northern hardwood zone, and its concomitant “mid-successional wave,” relatively few stands have large live trees today. Those areas that were cleared for farmland or pasture may not have seen large live trees for two centuries, while those with a long history of partial harvest have often seen such trees (whether merchantable or cull) systematically removed. In extreme cases, stands with depauperate species composition can be “reset” through clearcutting or other aggressive silviculture; but no such quick fix exists for large live trees or their standing or fallen remains. It is certain that, in comparison to modern old-growth stands, current managed stands (whether even- or uneven-aged) have lower abundances of snags and downed logs, both in total and when examining large size classes. For example, several studies in northern hardwoods and in comparable central Appalachian forests have found snag basal area in old-growth stands to range from approximately 4 to 8 m 2/ha (Carbonneau 1986, McCarthy and Bailey 1994, Goodburn and Lorimer 1998, McGee et al. 1999), while a survey of managed stands in the White Mountains of New Hampshire found snag basal areas from 0.3 to 4 m 2/ha (Jordan 2003). Total volume of downed coarse woody material in old-growth stands has ranged from 60-160 m 3/ha (Carbonneau 1986, Goodburn and Lorimer 1998, McGee et al. 1999, while volumes in managed stands have often been in the range of 15-65 m 3/ha (Gore and Patterson 1986, McCarthy and Bailey 1994, Goodburn and Lorimer 1998, McGee et al. 1999, Kenefic and Nyland 2000). While comparison across studies is not straightforward, due in part to differing lower diameter limits for snags and coarse woody material, the general pattern is fairly clear, and is consistent when results are examined at a regional scale (Heath and Chojnacky 2001). To put these numbers in perspective, let us suppose for the moment that a typical managed stand has 40 m 3/ha of wood in downed logs, and another 12 m 3/ha in snags (2 m 2/ha basal area, with a tree height of 18m and a form factor of 0.33). Suppose that a target old-growth level (following McGee et al. 1999) includes 100 m 3/ha in downed logs and 36 m 3/ha in snags (6 m 2/ha with similar height and form). The difference in total dead wood volume is 136-52=84 m 3/ha. A reasonable total wood production rate on a moderate to good site in northern New England might be 4.2 m 3/ha/yr (Leak et al. 1987). Thus, the target level could be achieved within 20 years, but only if all biological production were converted to dead wood, implying zero net increment of live volume (merchantable or otherwise), and if existing dead wood did not decay in the meantime. More realistically, we might expect that reaching the dead wood volume target would require several decades even if no harvesting occurred. If we add to the total volume target, a size-distribution goal that 50% of the material should have a diameter of 25cm or greater, and 10% should have a diameter of 50cm or greater (e.g. McGee et al. 1999), the situation becomes even more challenging. Because of taper in the log, a typical northern hardwood tree must be 64cm in diameter at breast height to produce a 5m long log of downed wood with a diameter >50cm. Relatively few managed northern hardwood stands appear to contain such dead trees (Gore and Patterson 1986, Jordan 2003), and live trees of that size are also comparatively rare in most managed stands. Meeting the combined volume and diameter target would require 2 standing trees at least 64 cm DBH, and 8 downed logs per hectare of comparable size. To be fully useful as wildlife habitat, the trees would also need to develop hollow interiors before falling. Thinning treatments might be used to accelerate the development of large-diameter trees (and eventually snags) but many decades would still be required to meet these targets in most northern hardwood stands. Therefore, even if these structural targets are accepted, we believe they should be viewed as long-term goals and not short-term standards. At least three potential approaches exist to move toward the kinds of structural goals suggested:
Passive approach
This approach recognizes high-value structural attributes (such as large snags and large downed logs) and stands with high structural values (those with large tree sizes and other legacies with “old-growth” character) and seeks to protect them. Protection of snags often runs afoul of OSHA guidelines in stands being logged (see, e.g., NHFSSWT 1997), but protection from crushing and other logging damage may be important for downed logs (McCarthy and Bailey 1994). Protecting entire stands has some economic cost and provides little protection against the possibility – perhaps rare but always present – that protected stands will be subject to major disturbance. A passive protective approach requires relatively low investment, and while it provides little hope for a rapid approach to aggressive structural targets such as those outlined above, it might also provide some improvement over what the land-use history of the northern hardwood zone has provided in many stands. Still, these structural legacies (large live trees, large snags, and large downed logs) would be present primarily as a landscape-scale feature in a few stands, and not as a common feature in many or even most stands.
Long-rotation approach
It might also be possible to provide large live trees, large snags, and large downed logs through extended rotations in even-aged systems, or through slow cycles with large maximum tree sizes in uneven-aged systems. (In an even-aged system, the abundance of these legacies might fluctuate through time for a given stand, while in an uneven-aged system the goal would be for an eventual approach to a near-steady state within a given stand; the long-term average through time and space might be similar in a well-regulated landscape.) This might require rotation ages (perhaps 120-200 years) in even-aged systems, though (as another result of the land-use history in this zone) there is virtually no direct data on the yield or structural development of northern hardwoods to these ages. In an uneven-aged system, gap creation rates would almost certainly be toward or beyond the low end of the turnover rate identified by Lorimer (1989). Again, there is a general lack of data on northern hardwood stands managed with such a system. Both approaches would entail the accumulation of large amounts of growing stock (capital)on the landscape, with little promise of improved production (total allowable cut, expressed in terms of cubic volume, would probably decline relative to current common practice, though grade distribution might improve with a shift of some production toward larger tree sizes). Either approach would likely entail significant economic sacrifices and might be wholly unadoptable except, perhaps, on large public ownerships.
Active Retention and Creation of Legacies
McGee et al. (1999) suggest that aggregated green tree retention might be used to maintain and improve the abundance of snags and downed logs in stands managed on shorter rotations. We believe there is much to be commended in this approach, but know of relatively little field data and no comprehensive descriptive or prescriptive modeling analysis to support precise recommendations on green-tree retention in northern hardwood systems. For example, the susceptibility of leave trees or islands to windthrow remains largely unexplored in this type, as has the long-term influence of leave trees or islands on stand dynamics. The size or number of retention patches required to achieve or maintain target snag or downed log levels might be addressed at least in a preliminary fashion through modeling. It is not clear how to maintain reserve trees or patches operationally if the stand matrix is to be managed using a fine-grained selection approach, as suggested by Seymour et al. (2002): trees and patches designated for reservation might not stand out from the rest of the stand, especially after successive stand entries, and could easily be harvested inadvertently. In short, while green-tree retention seems appealing, it also appears to be an area with significant research needs and some operational challenges.
Finally, we must ask whether maintenance of these legacies is required in every stand, or whether having them as significant landscape features is required. Certainly if the goal is to recreate the “look and feel” of what many studies describe for the pre-European northern hardwood forest, or if the goal is to develop structures within a certain pattern and scale of disturbance in the hopes that (perhaps unspecified) ecological patterns and processes might be better sustained, then large trees, snags, and downed logs are likely to be important features and should be present in many stands. However, if the goal is to maintain populations of a range of plant and animal species on the landscape, the jury is out on whether such a comprehensive approach is needed.
Despite their limitations, analyses of early survey records and reconstructions in old modern stands point to a northern hardwood zone once characterized by frequent small-scale disturbance, and relatively infrequent large-scale stand-replacing disturbance. Both kinds of disturbance, however, appear to have been important to the within-stand and across-landscape diversity of structures and species. If we accept the basic calculations of Lorimer and White (2003), then approximately 30% of the pre-European northern forest landscape would have been in stand developmental stages (stand initiation and stem exclusion) best produced now by even-aged or single-cohort silvicultural systems, 40% in developmental stages best approximated by multicohort systems with gap sizes ranging from small groups to small patches (conforming to the demographic definition of old-growth in Oliver and Larson 1996), and 30% in transitional stages between the two (including understory reinitiation and transitional multicohort stages). A diversity of silvicultural systems appears necessary to maintain a managed forest landscape with such proportions. However, it may be useful to modify existing even-aged systems to better incorporate structural legacies, through some combination of passive maintenance, green-tree retention, and longer rotations depending on land tenure and landowner objectives. We must recognize as well that uneven-aged systems do not necessarily provide those same structural legacies even if they are designed to match pre-European disturbance scale-frequency combinations. Conservation of biodiversity and sustainable forest production have both been significantly impacted by land-use history within the northern hardwood zone, and both will continue to be impacted has land-use history unfolds. A “mid-successional wave” has left a deficit in both early- and late-successional stands and structures, but heavy cutting (often associated with changes in land ownership) may be ameliorating the early-successional deficit in portions of the northern hardwood zone. Successful redevelopment of structures associated with older stands and landscapes will almost certainly require modifications and additions to both even- and uneven-aged silvicultural approaches in this region, but the general lack of older stands (and especially older managed stands) in the region has provided few opportunities for research to anticipate these needs. We expect that practice and research must evolve in tandem, that some approaches will be tried and found in error, and that the redevelopment of the northern hardwood landscape will be a long-term proposition, not a quick and easy fix.
Abrams, M.D. and C.M. Ruffner. 1995. Physiographic analysis of witness-tree distribution (1765-1798) and present forest cover through north central Pennsylvania. Can J. For. Res. 25:659-668 Ashton, P.M.S. and M.J. Ducey. 1996. The development of mixed species plantations as successional analogues to natural forests. pp. 113-126 in D.B. Landis and T.D. South, tech. coords. National Proceedings, Forest and Conservation Nursery Associations. U.S.D.A. Forest Service, Gen. Tech. Rep. PNW-369. Beatty, S.W. and E.L. Stone. 1986. The variety of soil microsites created by treefalls. Can J. For. Res. 16:539-548 Belcher, C. F. 1981 . Logging railroads of the White Mountains . Revised Edition. Appalachian Mountain Club, Boston , Massachusetts . 237 pp. Bellemare, J., G. Motzkin, and D.R. Foster. 2002. Legacies of the agricultural past in the forested present: An assessment of historical land-use effects on Rich Mesic Forests. J. Biogeog. 29:1401-1420 Bohn, K.K. and R.D. Nyland. 2003. Forecasting development of understory American beech after partial cutting in uneven-aged northern hardwood stands. For. Ecol. & Mgmt. 180:453-461 Boose, E.R., K.E. Chamberlin, and D.R. Foster. 2001. Landscape and regional impacts of hurricanes in New England. Ecol. Mon. 71(1):27-48 Bormann, F. H. and G.E. Likens. 1994. Pattern and process in a forested ecosystem: disturbance, development and the steady state based on the Hubbard-Brook Ecosystem Study. Springer-Verlag. Brewer, R. and P.G. Merritt. 1978. Windthrow and tree replacement in a climax beech-maple forest. Oikos 30:149-152 Canham, C.D. 1988. Growth and canopy archetecture of shade-tolerant trees: response to canopy gaps. Ecology 69:786-795 Canham, C.D. 1989. Different responses to gaps among shade-tolerant species. Ecology 70:548-550 Canham, C.D., M.J. Papaik, and E.F. Latty. 2001. Interspecific variability in susceptibility to windthrow as a function of tree sizeand storm severity for northern temperate tree species. Can. J. For. Res. 31(1):1-10 Carbonneau, L.E. 1986. Old-growth forest stands in New Hampshire: A preliminary investigation. M. S. Thesis, Univ. of New Hampshire, Durham NH. Carlton, G.C. and F.A. Bazzaz. 1998. Regeneration of three sympatric birch species on experimental hurricane blowdown microsites. Ecol. Mon. 68:99-120 Chandler , D. S. 1987. Species richness and abundance of Pselaphidae in old-growth and 40-year-old forests in New Hampshire . Can. J. Zool. 65 (3): 608-615. Chandler , D. S. 1991. Comparison of some slime-mold and fungus feeding beetles (Coleoptera, Eucinetoidea, Cucujoidea) in an old-growth and 40-year-old forest in New Hampshire . Coleopterists Bulletin 45 (3): 239-256 SEP 1991 Chandler , D. S. and S. B. Peck. 1992. Diversity and seasonality of Leiodid beetles (Coleoptera, Leiodidae) in an old-growth and a 40-year-old forest in New Hampshire . Environmental Entomology (6): 1283-1293 Cogbill, C.V. 2000. Vegetation of the presettlement forests of northern New England and New York. Rhodora 102(911):250-276 Cogbill, C.V., J. Burk, and G. Motzkin. 2002. The forests of presettlement New England, USA: spatial and compositional patterns based on town proprietor surveys. J. Biogeog. 29:1279-1304 Collins, B.S. and S.T.A. Pickett. 1987. Influence of canopy opening on the environment and herb layer in a northern hardwoods forest. Vegetatio 70:3-10 Collins, B.S. and S.T.A. Pickett. 1988. Demographic responses of herb layer species to experimental canopy gaps in a northern hardwoods forest. J. Ecol. 76:437-450 Cook, E.M Jr. 1989. Ossipee, New Hampshire, 1785-1985: a history. Peter E. Randall Publisher, Portsmouth, NH. Costello, C.A., M. Yamasaki, P.J. Perkins, W.B. Leak, and C.D. Neefus. 2000. Songbird response to group selection harvests and clearcuts in a New Hampshire northern hardwood forest. For. Ecol. & Mgmt. 127:41-54 Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest. Ecology 34:329-346 DeGraaf, R.M. and A.L. Shigo. 1985. Managing cavity trees for wildlife in the Northeast. General Technical Report NE-101 Broomall, PA. USDA Forest Service, Northeastern Forest Experiment Station. DeGraaf, R.M., M. Yamasaki, W.B. Leak, and J.W. Lanier. 1992. New England wildlife: management of forested habitats. General Technical Report NE-144, USDA Forest Service, Northeastern Forest Experiment Station, Radnor, PA. DeGraaf, R. M., and M. Yamasaki. 2001. New England wildlife: habitat, natural history, and distribution. University Press of New England, Hanover, NH. DeGraaf, R.M. and M. Yamasaki. 2003. Options for managing early-successional forest and shrubland bird habitats in the northeastern United States. For. Ecol & Mgmt. 185:179-191. Dupouey, J.L., E. Dambrine, J.D. Laffite, and C. Moares. 2002. Irreversible Impact of past land use on forest soils and biodiversity. Ecology 83(11):2978-2984 Fahey, T.J. and W.A. Reiners. 1981. Fire in the forests of Maine and New Hampshire. Bull. Torrey Bot. Club 108(3):362-373 Foster, D.R. 1988. Species and stand response to catastrophic wind in central New England, U.S.A.. J. Ecol. 76:135-151 Foster, D.R. 1999. Thoreau's country: journey through a transformed landscape. Harvard University Press, Cambridge, MA. 270 pp. Foster, D.R., G. Motzkin, D. Bernardos, and J. Cardoza. 2002. Wildlife dynamics in the changing New England landscape. J. Biogeog. 29:1337-1358 Foster, D.R., F. Swanson, J. Aber, I. Burke, N. Brokaw, D. Tilman, and A. Knapp. 2003. The importance of land-use legacies to ecology and conservation. Biosci. 53(1):77-88 Franklin, J.F., T.A. Spies, R. Van Pelt, A.B. Carey, D.A. Thornburgh, D.R. Berg, D.B. Lindenmayer, M.E. Harmon, W.S. Keeton, D.C. Shaw, K. Bible, and J. Chen. 2002. Disturbances and structural development of natural forest ecosystems with silvicultural implications, using Douglas-fir forests as an example. For. Ecol. & Mgmt. 155:399-423 Fuller, T.K., and S. DeStefano. 2003. Relative importance of early-successional forests and shrubland habitats to mammals in the northeastern Untied States. For. Ecol. & Mgmt. 185:75-79. Goldblum, D. 1997. The effects of treefall gaps on understory vegetation in New York State. J. Veg. Sci. 8:125-132 Goodburn, J.M. and C.G. Lorimer. 1998. Cavity trees and coarse woody debris in old-growth and managed northern hardwood forests in Wisconsin and Michigan. Can. J. For. Res. 28: 427-438. Gore, J.A. and W.A. Patterson III. 1986. Mass of downed wood in northern hardwood forests in New Hampshire: potential effects of forest management. Can. J. For. Res. 16: 335-339. Hall, B., G. Motzkin, D.R. Foster, M. Syfert, and J. Burk. 2002. Three hundred years of forest and land-use change in Massachusetts, USA. J. Biogeog. 29:1319-1336 Hannah, P. 1999. Species composition and dynamics in two hardwood stands in Vermont: A disturbance history. For. Ecol. & Mgmt. 120:105-116 Heath, L.S. and D.C. Chojnacky. 2001. Down dead wood statistics for Maine timberlands, 1995. U.S.D.A. Forest Service, Res. Bull. NE-150. Heitzman, E. and R.D. Nyland. 1994. Influences of pin cherry (Prunus pensylvanica L. f.) on growth and development of young even-aged northern hardwoods. For. Ecol. & Mgmt. 67:39-48 Hibbs, 1979. The age structure of a striped maple population. Can. J. For. Res. 9:504-508. Henry, J.D. and J.M.A. Swan. 1974. Reconstructing forest history from live and dead plant material - an approach to the study of forest succession in southwest New Hampshire. Ecology 55:772-783 Hornbeck, J.W. and W.B. Leak. 1992. Ecology and management of northern hardwood forests in New England. General Technical Report NE-159 USDA, Forest Service, Northeastern Forest Experiment Station. 41 pp. Howard, L.F. and T.D. Lee. 2002. Upland old-field succession in southeastern New Hampshire. J. Tor. Bot. Soc. 129:60-76 Howard, L.F. and T.D. Lee. 2003. Temporal patterns of vascular plant diversity in southeastern New Hampshire Forests. For. Ecol. & Mgmt. 185:5-20 Hunter, Jr., M.L. 1996. Benchmarks for managing ecosystems: are human activities natural? Conservation Biology 10: 695-697. Jordan, G.J. 2003. Quantitative assessment of coarse woody debris in a managed northern hardwood forest. M.S. Thesis, University of New Hampshire, Durham, NH. Keith, L. B. , A. W. Todd, C. J. Brand, R. S. Adamcik, and D. H. Rusch. 1977. An analysis of predation during cyclic fluctuations of snowshoe hares. Proceedings of the International Congress of Game Biologists 13:151-175. Kelty, M.J., D.B. Kittredge Jr., T. Kyker-Snowman, and A.D. Leighton. 2003. The conversion of even-aged stands to uneven-aged structure in southern New England. North. J. Appl. For. 20: 109-116. Kenefic, L.S. and R.D. Nyland. 2000. Habitat diversity in uneven-aged northern hardwood stands: a case study. U.S.D.A. Forest Service, Res. Paper NE-714. Latty, E.F., C.D. Canham, and P.L. Marks. 2003. Beech bark disease in northern hardwood forests: The importance of nitrogen dynamics and forest history for disease severity. Can J. For. Res. 33(2):257-268 Leak, W.B. 1972. Competitive exclusion in forest trees. Nature 236:461-463 Leak, W.B. 1987. Fifty years of compositional change in deciduous and coniferous forest types in New Hampshire. Can J. For. Res. 17:388-393 Leak, W.B. 1991. Secondary forest succession in New Hampshire, USA. For. Ecol. & Mgmt. 43:69-86 Leak, W.B. 2003. Best density and structure for uneven-aged northern hardwood management in New England. North. J. Appl. For. 20: 43-44. Leak, W.B. and S.M. Filip. 1975. Uneven-aged management of northern hardwoods in New England. USDA Forest Service Research Paper NE-332 USDA, Forest Service, Northeastern Forest Experiment Station, Upper Darby, PA. 15 pp. Leak, W.B. and P.S. Sendak. 2002. Changes in species, grade, and structure over 48 years in a managed New England northern hardwood stand. North. J. Appl. For. 19: 25-27. Leak, W.B. and D.S. Solomon. 1975. Influence of residual stand density on regeneration of northern hardwoods. USDA Forest Service Research Paper NE-310 USDA, Forest Service, Northeastern Forest Experiment Station, Upper Darby, PA Leak, W.B., D.S. Solomon, and P.S. DeBald. 1987. Silvicultural guide for northern hardwood types in the Northeast (revised). Research Paper NE-603 USDA, Forest Service, Northeastern Forest Experiment Station Litvaitis, J.A. 1993. Response of early-successional vertebrates to historic changes in land use. Cons. Biol. 7:866-873 Litvaitis, J.A. 2001. Importance of early-successional habitats to mammals in eastern forests. Wildlife Society Bulletin 29:466-473. Liu, C., L. Zhang, C.J. Davis, D.S. Solomon, and J.H. Gove. 2002. A finite mixture model for characterizing the diameter distributions of mixed-species forest stands. Forest Science 48:653-661. Lorimer, C.G. 1977. The presettlement forest and natural disturbance cycle of northeastern Maine. Ecology 58: 139-148. Lorimer, C.G. 1985. Methodological considerations in the analysis of forest disturbance. Can. J. For. Res. 15: 200-213. Lorimer, C.G. 1989. Relative effects of small and large disturbances on temperate hardwood forest structure. Ecology 70:565-567 Lorimer, C.G. and A.S. White. 2003. Scale and frequency of natural disturbances in the Northeastern United States: implications for early-successional forest habitat and regional age distributions. For. Ecol. & Mgmt. 185:41-64 Lull, H.W. 1968. A forest atlas of the northeast. USDA, Forest Service, Northeastern Forest Experiment Station, Upper Darby, PA. 46 pp. Marks, P.L. 1974. The role of pin cherry (Prunus pensylvanica L.)in the maintenance of stability of northern hardwood ecosystems. Ecol. Monogr. 44:73-88 Mattson, J.A. 1988. Technological alternatives - what’s available to accomplish silvicultural objectives? IN: Martin, C.W., C.T. Smith, and L.M. Tritton (eds.). New perspectives on silvicultural management of northern hardwoods: proceedings of the 1988 symposium on the conflicting consequences of practicing northern hardwood silviculture. General Technical Report NE-124. USDA, Forest Service, Northeastern Forest Experiment Station McCarthy, B.C. and R.R. Bailey. 1994. Distribution and abundance of coarse woody debris in a managed forest landscape of the central Appalachians. Can. J. For. Res. 24: 1317-1329. McClure, J.W. and T.D. Lee. 1993. Small-scale disturbance in a northern hardwoods forest: effects on tree species abundance and distribution. Can J. For. Res. 23:1347-1360 McClure, J.W., T.D. Lee, and W.B. Leak. 2000. Gap capture in northern hardwoods: patterns of establishment and height growth in four species. For. Ecol. & Mgmt. 127:181-189 McGee, G.G., D.J. Leopold, and R.D. Nyland. 1999. Structural characteristics of old-growth, maturing, and partially cut northern hardwood forests. Ecol. App. 9(4):1316-1329 McIntosh, R.P. 1972. Forests of the Catskill Mountains, New York. Ecol. Mon. 42(2):143-161 Merrens, E.J. and D.R. Peart. 1992. Effects of Hurricane Damage on Individual Growth and Stand Structure in a Hardwood Forest in New Hampshire, USA. J. Ecol. 80(4):787-795 Mitchell, R.J., B.J. Palik, and M. L. Hunter, Jr. 2002. Natural disturbance as a guide to silviculture. For. Ecol. & Mgmt. 115:315-317 Moore, M.R. and J.L. Vankat. 1986. Responses of the herb layer to gap dynamics of a mature beech-maple forest. Am. Nat. 115:336-347 Nakashizuka, Tohru. 2001. Species coexistence in temperate, mixed deciduous forests. Trends Ecol. Evol. 16:205-210 NHFSSWT ( New Hampshire Forest Sustainability Standards Work Team). 1997. Good forestry in the Granite State. Concord, NH: New Hampshire Div. of Forests and Lands, and Soc. for the Protection of New Hampshire Forests. Nordstrom, L., A. Vandehey, and J. Mizzi. 2000. Endangered and threatened wildlife and plants; determination of threatened status for the contiguous U.S. distinct population segment of the Canda lynx and related rule. Federal Register 65:16051-16086. Nyland, R.D. 2003. Even- to uneven-aged: The challenges of conversion. For. Ecol. & Mgmt. 172:291-300 Nyland, R.D., D.G. Ray, R.D. Yanai, R.D. Briggs, L. Zhang, R.J. Cymbala, and M.J. Twery. 2000. Early cohort development following even-aged reproduction method cuttings in New York northern hardwoods. Can J. For. Res. 30(1):67-75 Oliver, C.D. and B.C. Larson. 1996. Forest Stand Dynamics. Wiley, New York. 520 pp. Orwig, D.A., D.R. Foster, D.L. Mausel. 2002. Landscape patterns of hemlock decline in New England due to the introduced hemlock woolly adelgid. J. Biogeog. 29:1475-1488 Parshall, T. and D.R. Foster. 2002. Fire on the New England landscape: regional and temporal variation, cultural and environmental controls. J. Biogeog. 29:1305-1317 Peterson, C. J. and S. T. A. Pickett. 1995. Forest reorganization – A case study in an old-growth forest catastrophic blowdown. Ecology 76(3):763-774 Ray, D.G., R.D. Nyland, and R.D. Yanai. 1999. Patterns of early cohort development following shelterwood cutting in three Adirondack northern hardwood stands. For. Ecol. & Mgmt. 119:1-11 Rhoads, A.G., S.P. Hamburg, T.J. Fahey, T.G. Siccama, E.N. Hane, J. Battles, C. Cogbill, J. Randall, and G. Wilson. 2002. Effects of an intense ice storm on the structure of a northern hardwood forest. Can J. For. Res. 32(10):1763-1775 Ruggerio, L.F., K.B. Aubry, S.W. Buskirk, G.M. Koehler, C.J. Krebs, K.S. McKelvey, and J.R. Squires, editors. 2000. Ecology and conservation of lynx in the United States. University Press of Colorado, Boulder. Runkle, J.R. 1981. Gap regeneration in some old-growth forests of the eastern United States. Ecology 62(4):1041-1051 Runkle, J.R. 1982. Patterns of disturbance in some old-growth mesic forests of eastern North America. Ecology 63(5):1533-1546 Runkle, J.R. 1990. Eight years of change in an old Tsuga canadensis woods affected by beech bark disease. Bull. Torrey Bot. Club 117(4):409-419 Russell, E.W.B. 1983. Indian-set fires in the forests of the northeastern United States. Ecology 64:78-88 Scanlon, J.J. 1992. Managing forests to enhance wildlife diversity in Massachusetts. Northeast Wildlife 49:1-9. Schwarz, P.A., T.J. Fahey, C.W. Martin, T.G. Siccama, and A. Bailey. 2001. Structure and composition of three northern hardwood-conifer forests with differing disturbance histories. For. Ecol. & Mgmt. 144:201-212 Selva, S. B. 1987. Lichens as indicators of old-growth forests in northern Maine. Amer. J. Bot. 74(5):601-601 Selva, S. B. 1994. Lichen diversity and stand continuity in the northern hardwoods and spruce-fir forests of northern New England and western New Brunswick . Bryologist 97(4):424-429 Selva, S. B. 2003. Using calicioid lichens and fungi to assess ecological continuity in the Acadian Forest Ecoregion of the Canadian Maritimes. For. Chron. 79(3):550-558 Seymour, R.S., A.S. White, and P.G. deMaynadier. 2002. Natural disturbance regimes in northeastern North America - evaluating silvicultural systems using natural scales and frequencies. For. Ecol. & Mgmt. 155:357-367 Smith, D.M. 1988. Even-aged management: when is it appropriate and what does it reveal about stand development? IN: Martin, C.W., C.T. Smith, and L.M. Tritton (eds.). New perspectives on silvicultural management of northern hardwoods: proceedings of the 1988 symposium on the conflicting consequences of practicing northern hardwood silviculture. General Technical Report NE-124. USDA, Forest Service, Northeastern Forest Experiment Station Solomon, D.S. 1977. The influence of stand density and structure on growth of northern hardwoods in New England. USDA Forest Service Research Paper NE-362 USDA, Forest Service, Northeastern Forest Experiment Station, Upper Darby, PA. 13 pp. Tubbs, C. 1988. Uneven-age management: when do conditions require this approach? IN: Martin, C.W., C.T. Smith, and L.M. Tritton (eds.). New perspectives on silvicultural management of northern hardwoods: proceedings of the 1988 symposium on the conflicting consequences of practicing northern hardwood silviculture. General Technical Report NE-124. USDA, Forest Service, Northeastern Forest Experiment Station Tubbs, C.H., R.M. DeGraaf, M. Yamasaki, and W.M. Healy. 1987. Guide to wildlife tree management in New England northern hardwoods. General Technical Report NE-118 USDA, Forest Service, Northeastern Forest Experiment Station, Broomall, PA. Vujanovic V. and J. Brisson. 2002. Microfungal biodiversity on Fagus grandifolia in an old -growth forest of Eastern North-America . Phyton-Annales Rei Botanicae 42(2):315-327 Wang, Z. and Nyland, R.D. 1996. Changes in the condition and species composition of developing even-aged northern hardwood stands in Central New York. North. J. Appl. For. 13(4):189-194 Westveld, M., R.I. Ashman, H.I. Baldwin, R.P. Holdsworth, R.S. Johnson, J.H. Lambert, H.J. Lutz, L. Swain, and M. Standish. 1956. Natural forest vegetation zones of New England. J. For. 54(5):332-338 Whitney, G.G. and D.R. Foster. 1988. Overstory composition and age as determinants of the understory flora of woods of central New England. J. Ecol. 76:867-876 Woods, K.D. 1979. Reciprocal replacement and the maintenance of codominance in a beech-maple forest. Oikos 33:31-39 Woods, K.D. 1984. Patterns of tree replacement: canopy effects on understory pattern in hemlock-northern hardwood forests. Vegetatio 56:87-107 Woods, K.D. 2000. Long-term change and spatial pattern in a late-successional hemlock-northern hardwood forest. J. Ecol. 88(2):267-282
|
Copyright © University of New Hampshire, Department of Natural Resources
James Hall, 56 College Road, Durham, NH 03824
Last modified February 28, 2005