Charles B. Halpern

Early Successional Patterns of Forest Species: Interactions of Life History Traits and Disturbance

Patterns of abundance were examined for vascular plant species during 21 yr of succession in two clear-cut and burned Pseudotsuga forests in the western Cascade Range of Oregon. A majority of forest understory species persisted through disturbance. Most colonizing species established within 2 yr after burning. Individualistic species re- sponses were described by a series of broadly overlapping, unimodal curves of constancy and canopy cover, differing in time of initiation, duration, and magnitude. Thus, early successional change was characterized by gradual shifts in the abundance of generally persistent species. Eleven population patterns (species groups) were identified. Interactions of life history traits and disturbance explain the temporal trends of the most common species. Within the groups of invading species, the timing of initial establishment, as well as the timing and magnitude of peak abundance were related to the origin of propagules, phenological traits, potential for vegetative expansion, and temporal and spatial variation in disturbance. Abundance patterns of invading species were also influenced by stochastic and historical factors. Contrasting responses of species between sites reflected differences in histories of logging and slash burning. Within the groups of residual species, temporal patterns of abundance reflected initial species distributions, resistance to logging and burning distur- bance, mode of reproduction, morphological traits, and spatial variation in disturbance intensity. These observations suggest that early secondary succession in Pseudotsuga forests has a deterministic component, founded in the life history traits of the available species, and

EARLY SUCCESSIONAL PATHWAYS AND THE RESISTANCE AND RESILIENCE OF FOREST COMMUNITIES

Vegetation changes were studied for 21 yr in two clearcut logged and slash- burned Pseudotsuga forests in the western Cascade Range of Oregon. Detrended corre- spondence analysis (DCA) was used to examine the successional relationships among six understory communities exposed to a gradient of disturbance intensity. Euclidean distances between pre- and postdisturbance samples in ordination space were used to compare community resistance to disturbance and long-term recovery, or resilience. Ordination through time for plant communities revealed a common pattern of rapid floristic change away from predisturbance composition, followed by gradual, unidirectional return. Early, but transient, convergence of successional pathways was common among mesic- and dry- site communities, reflecting the broad distribution of colonizers and the floristic similarity of predisturbance understories. Distinct sequences were observed on moist sites, reflecting more unique residual and colonizing floras. Ordinations also revealed increasing compo- sitional change with disturbance intensity. Successional sequences were dominated by residual species on relatively undisturbed sites and by alternate suites of invading species on moderately disturbed and burned sites. Variation in the response gradient between watersheds reflected the modifying influence of local environment, stand history, and chance in succession. Resistance and resilience varied little among plant communities but were generally lowest for the depauperate Coptis community and greatest for the compositionally and structurally diverse Polystichum and Rhododendron-Gaultheria types. Both measures were strongly influenced by disturbance intensity. The stability of Pseudotsuga understories derives from the moderate tolerance of initial understory dominants to burning and in their ability to subsequently perennate from subterranean structures. Variation in the long- term response of communities reflects complex interactions between species life history, disturbance intensity, and chance, suggesting that both deterministic and stochastic factors must be considered in evaluating community stability and response to disturbance.

TREE MORTALITY DURING EARLY FOREST DEVELOPMENT: A LONG-TERM STUDY OF RATES, CAUSES, AND CONSEQUENCES

Tree mortality is a critical but understudied process in coniferous forest development. Current successional models assume that mortality during early forest development is dominated by density-dependent processes, but few long-term studies exist to test this assumption. We examined changes in forest structure and patterns of tree mortality 14-38 years (1979-2001) after clear-cut logging of two experimental watersheds in the western Cascade Range of Oregon, USA. We sampled 193 permanent plots (250 m 2 ) six times generating 75 126 data records and 7146 incidents of mortality. Mean density peaked at .3000 stems/ha (� 1.4 m tall) after 22-25 years; bole biomass increased continuously to .100 Mg/ha. At final sampling, stem density varied by two orders of magnitude and biomass by a factor of 10 among sample plots. Suppression mortality occurred in .80% of plots and was .2.5 times as frequent as mechanical damage (uprooting, stem snap, and crushing). However, biomass lost to mortality via mechanical damage was nearly four times that lost to suppression, a result of episodic storms that created windthrow patches, with some plots losing 30-50% of biomass. Total annual mortality increased from 1.0% to 5.3% of stems over the study period and was highly variable among species. Although mortality rates were highest for sprouting hardwoods (reaching 9.7% in Cornus nuttallii), biomass of most hardwood species increased through canopy closure as dominant stems achieved large sizes. Shade-tolerant conifers (Tsuga heterophylla and Thuja plicata), typically assumed to be absent or to play a minor role in early forest development, accounted for 26% of stems after 38 years. In regression tree models, environmental attributes of plots had limited ability to predict mortality. Instead, stem density prior to canopy closure was the strongest predictor of cumulative mortality (either suppression or mechanical damage). Our long-term studies suggest that current models of early forest development are overly simplistic, particularly in their treatment of mortality. Although suppression was the dominant demographic process, mechanical damage yielded greater loss of biomass and greater structural heterogeneity through creation of windthrow gaps. Thus, gap-forming processes that operate late in succession and contribute to structural complexity in old-growth forests can also occur early in stand development.

POTENTIAL RESPONSE OF PACIFIC NORTHWESTERN FORESTS TO CLIMATIC CHANGE, EFFECTS OF STAND AGE AND INITIAL COMPOSITION

We used an individual-based forest simulator (a gap model) to assess the potential effects of anthropogenic climatic change on conifer forests of the Pacific Northwestern United States. Steady-state simulations suggested that forest zones could be shifted on the order of 500–1000 m in elevation, which could lead to the local extirpation of some high-altitude species. For low-elevation sites, species which currently are more abundant hundreds of kilometers to the south would be favored under greenhouse scenarios. Simulations of transient responses suggested that forest stands could show complex responses depending on initial species composition, stand age and canopy development, and the magnitude and duration of climatic warming. Assumptions about species response to temperature, which are crucial to the models behaviors, were evaluated using data on species temperature limits inferred from regional distributions. The high level of within-species variability in these data, and other confounding factors influencing species distributions, argue against over-interpreting simulations. We suggest how we might resolve critical uncertainties with further research.

Effects of environment and grazing disturbance on tree establishment in meadows of the central Cascade Range, Oregon, USA

Within the last century there has been widespread establishment of trees in mountain meadows of the Pacific Northwest. We reconstructed patterns of tree invasion at 17 meadow sites in the central Cascade Range of Oregon, USA - sites representing diverse physical environments and vegeta- tion types and experiencing different histories of recent an- thropogenic disturbance (sheep grazing). Spatial distributions and age structures of invasive tree populations were analysed with respect to climatic records and grazing history. Patterns of establishment varied considerably among meadows, re- flecting strong differences in environment and grazing his- tory. In montane hydric meadows, tree establishment was spatially clumped beneath large old trees and on elevated microsites; however the timing of invasion differed between sites with stable versus fluctuating water tables. In upland mesic/dry montane meadows, timing of invasion corresponded with cessation of sheep grazing (early 1940s) and the onset of wetter summers (mid 1940s). In the subalpine zone, climate and aspect interacted to produce contrasting histories of inva- sion on north- and south-facing slopes. Establishment on north-facing slopes, concentrated in heath-shrub communities, coincided with regional warming (ca. 1920-1945) when snowpacks were lighter and melted earlier. Recruitment of trees onto south-facing slopes occurred later, when conditions were wetter (1945-1985). In many environments, the spatial distribution of recruitment suggests that once trees have established, autogenic factors become increasingly important as individual trees or groups of trees alter the physical or biotic conditions that once inhibited establishment. Knowledge of the factors that influence invasion, and of their varying importance across gradients in environment and vegetation, is critical to predicting future changes in these dynamic systems.

EDGE‐RELATED RESPONSES OF UNDERSTORY PLANTS TO AGGREGATED RETENTION HARVEST IN THE PACIFIC NORTHWEST

Aggregated retention of overstory trees is now a standard component of timber harvest prescriptions on federal lands in the Pacific Northwest. Patches of remnant forest retained during harvest are thought to enhance the structural and biological diversity of managed forests, but the extent to which they maintain components of the original understory or promote recovery in adjacent harvest areas has not been tested. We examined short-term (1- and 2-yr) responses of understory plants to disturbance and creation of edges in structural retention harvest units at two sites in the western Cascade Range of Washington. Pre- and post-treatment abundance of vascular plants was measured in four (two at each site) 1-ha aggregates (patches of intact forest) and in surrounding harvest areas along sixteen 81 m long transects placed perpendicular to the edges of these aggregates. Two years after treatment, aggregates had gained an average of two forest species (vs. a loss of two in adjacent areas of harvest) and less than one early-seral species (vs. a gain of nine in adjacent areas of harvest). Aggregates supported populations of late-seral species that disappeared from or declined substantially in harvested areas. However, aggregates showed edge-related changes in plant abundance: one third of common understory herbs declined significantly in cover toward the edge, and changes in community composition were distinctly higher withi n5mo f theedge than in the aggregate center. Early-seral species established infre- quently within the aggregates, and only within 10 m of the edge. Herbaceous species generally showed larger declines in abundance with proximity to edge than did shrubs, with declines becoming more prominent over time. Our results suggest that, over short time frames, forest aggregates of one or more hectares may play an important role in maintaining plant species richness and composition in forests managed for timber harvest. Assessing the longer term stability of forest aggregates and the degree to which they influence recovery in adjacent areas of harvest will require continued observation.

The structure and dynamics of Abies magnifica forests in the southern Cascade Range, USA

Abies magnifica (Red fir) forests in the Cascade Range and Sierra Nevada of California are composed of groups, or patches, of even-sized individuals that form structurally complex stands. Patches may be even-aged, resulting from synchronous post-disturbance establishment, or multi-aged, reflecting continuous recruitment of seedlings moderately tolerant of shade. We analyze the population structure (i.e. age, size, and spatial patterning) of A. magnifica, and associated A. concolor, White fir, and reconstruct the disturbance history of two mature to old-growth A. magnifica forests in order to determine the relationship between disturbance and forest structure. Within both stands examined, the distributions of A. magnifica seedlings, saplings, and small understory trees were clumped, with clump sizes corresponding to the area of canopy gaps. Gaps were created by frequent wildfire (mean fire return interval of 41 yr) and by windstorms. Severe fire initiated mass establishment of Abies magnifica, whereas gaps created by windthrow released already established individuals. Low inten- sity fire stimulated little recruitment, but Red fir established continuously during fire-free intervals. Thus, the complex age and structures of Red fir forests reflect both episodic and continuous recruitment, as determined by the type and severity of natural disturbance.

Disturbance and post-harvest ground conditions in a structural retention experiment

Federal guidelines now specify retention of live trees on cutting units within the range of the northern spotted owl, but to date there have been no formal tests of the ecological or silvicultural implications of these guidelines, nor of possible tradeoffs associated with different spatial patterns of retention. We examined disturbance patterns and post-harvest ground conditions in mature forests of western Oregon and Washington in a factorial experiment that compared both level (15 vs. 40%) and pattern (dispersed vs. aggregated) of live-tree retention. Ground cover associated with logging disturbance (slash, disturbed soil, and intact forest floor), slash depth, and the difference (pre- vs. post-logging) in coarse woody debris (decay class I) were quantified on four treatment units (two factors, two levels) at six locations (blocks). Significantly greater disturbance was evident at lower levels of retention (15%) and on aggregated cutting units, but even greater differences were evident among blocks, reflecting the predominant influence of initial site differences and yarding methods (ground-based, suspension cable, or helicopter). Short-term responses of vegetation to varying levels and patterns of structural retention are likely to be shaped by levels of ground disturbance. Thus, we expect that block- and treatment-level effects will have important consequences for tree regeneration and understory composition in these forests. Conventional strategies used to minimize soil disturbance and to manage slash accumulation during clearcut logging should be applicable to variable retention systems.

Canopy closure exerts weak controls on understory dynamics: a 30‐year study of overstory–understory interactions

Stem exclusion and understory reinitiation are commonly described, but poorly understood, stages of forest development. It is assumed that overstory trees exert strong controls on understory herbs and shrubs during the transition from open- to closed-canopy forests, but long-term observations of this process are rare. We use long-term data from 188 plots to explore patterns and correlates of variation in understory richness and abundance 15–45 years after clear-cut logging and burning of two experimental watersheds in western Oregon, USA. We test whether variation in the temporal dynamics of plots can be explained by topoedaphic factors that influence resource availability (insolation and soil moisture), variation in the pace and intensity of overstory development, or characteristics of the vegetation prior to canopy closure. Changes in forest structure were substantial over the study period; canopy cover increased fourfold, stem density by 75%, and bole biomass by two orders of magnitude, although trends were highly variable among individual plots. In contrast, understory richness, foliar cover, and biomass declined only 30–40%, driven by loss of early-seral colonists, not residual forest species. Canopy closure occurred earlier on north aspects but declines in understory biomass, reflecting loss of colonizing shrubs (without concomitant increases in forest shrubs), were limited to south aspects. In contrast, variation in effective soil moisture had little influence on the pace of decline. Temporal trends were highly asynchronous among plots: nearly 50% of plots experienced some form of decline, but >35% showed no discernible trend. Declines were more likely in plots with greater tree influence before or at peak overstory development, but also in plots with greater understory development prior to canopy closure. Quantile regression models indicated weak relationships between understory biomass and overstory structure at most points in time. Our long-term data support a model of understory dynamics in which characteristics of the pre-closure vegetation are as important as overstory structure in determining the timing and nature of decline. Long-term studies are critical for elucidating patterns and processes that cannot be inferred from short-term experiments or space-for-time substitutions.

Physiognomic development of Pseudotsuga forests in relation to initial structure and disturbance intensity

Physiognomic patterns may vary significantly during succession despite a tendency for larger-growth forms to gradually replace smaller ones. Development of understory structure was observed for 25 yr after harvest of Pseudotsuga forests on two sites in the western Cas- cade Range, Oregon. We examine the influences of disturbance intensity and initial vegetation structure on the origin, direction, and rate of physiognomic change. Broad-scale changes in vegetation structure differed between sites. On Watershed 1, herbs dominated for 11 yr, after which shrubs became co-dominant. In contrast, Watershed 3 never exhibited a distinct, transitional shrub phase - herbs dominated for 18 yr, after which trees assumed co-dominance. The pattern and rate of physiognomic succession also varied among pre-disturbance plant communities and with disturbance intensity. Differences among communities largely corresponded with initial vegeta- tion structure, reflecting the disturbance tolerance of forest herbs and shrubs. Canopy closure occurred most rapidly in the initially depauperate, but tree-dominated Coptis community. Along the disturbance gradient, shifts from herb to shrub dominance occurred earlier on burned than on unburned sites due to rapid development of invading shrubs, whose germination and establish- ment were stimulated by fire. However, subsequent transitions to tree dominance showed no clear relationship with disturbance intensity. These long-term trends suggest that pre-disturbance community structure and disturbance intensity are major determinants of physiognomic succession, but that their effects may be modified by historical or stochastic fac- tors such as limited seed availability or local fluctuations in weather.