Robert K. Peet

Factors Influencing Succession: Lessons from Large, Infrequent Natural Disturbances

ABSTRACT Disturbance events vary in intensity, size, and frequency, but few opportunities exist to study those that are extreme on more than one of these gradients. This article characterizes successional processes that occur following infrequent disturbance events that are exceptional in their great intensity or large size. The spatial variability in disturbance intensity within large, infrequent disturbances (LIDs) often leads to a heterogeneous pattern of surviving organisms. These surviving organisms dictate much of the initial successional pattern on large disturbances where the opportunities for seeds to disperse into the middle of the disturbance are limited. The traditional distinction between primary and secondary succession is insufficient to capture the tremendous variability in succession following LIDs. Disturbance size influences succession where long-distance colonization by propagules is important. Observations from LIDs suggest the following interrelated hypotheses about trends in succession with increasing distance from seed sources when disturbanceintensity is high: (a) initial densities of organisms will be lower; (b) nucleation processes, in which recovering patches serve as foci for additional colonization and expand spatially, will be more important; (c) competitive sorting will be less important relative to chance arrival in determination of community composition, and (d) community composition will be initially less predictable; and (e) the rate of recovery of community composition will be slower. Prediction of succession following LIDs without considering contingencies such as the abundance, types, and spatial distribution of residuals, and distance to seed sources is likely to be unsuccessful for large portions of the landscape. Abundance and spatial arrangement of survivors and arrival patterns of propagules may be the pivotal factors determining how succession differs between intense disturbances of large and small extent.

Plant succession : theory and prediction

Prologue. The nature of vegetation dynamics D.C. Glenn-Lewin, E. van der Maarel. Establishment, colonization and persistence A.G. van der Valk. Community structure and ecosystem properties R.K. Peet. Regeneration dynamics T.T. Veblen. From population dynamics to community dynamics: modeling succession as a species replacement process R. van Hulst. Statistical models of succession M.B. Usher. Individual-based models of forest succession D.L. Urban, H.H. Shugart. Climate change and long-term vegetation dynamics I.C. Prentice. Epilogue.

Forest vegetation of the Colorado Front Range - Composition and dynamics

The forest vegetation of the northern Colorado Front Range was studied using a combination of gradient analysis and classification methods. A graphical model of forest composition based on elevation and topographic-moisture gradients was constructed using 305 0.1 ha samples. To derive the topographic moisture gradient, stands were stratified into eight 200 m elevation belts, and then ordinated by correspondence analysis using understory (<1 m) data. Each of the resultant gradients was scaled against a standard site moisture scalar derived from incident solar radiation and topographic position. Except for krummholz sites, the vegetation defined gradients fit the moisture scalar closely. Once scaled, these gradients were stacked vertically, sandwich-style, to create the graphical representation shown in Figure 5. Gradient analysis and ordination (direct and indirect gradient analysis of Whittaker, 1967) are frequently viewed as alternative approaches for analysis of vegetation. With gradient analysis the axes are readily interpretable, but stand placement is often difficult and at times questionable. Ordination defines an optimal arrangement for species and/or stands, but axis interpretation is often impossible. With the present combination of methods, the interpretability of gradient analysis complements the precision of placement obtained with ordination. Forest vegetation was classified by dividing the gradient model into eight series and 29 types on the basis of similar successional trends in canopy dominants. On dry, low-elevation sites above 1 700 m Pinus ponderosa woodlands dominate. With increasing elevation or site moisture, tree density increases and Pinus ponderosa, Pseudotsuga forests prevail. At middle elevations on mesic sites forests of mixed composition occur. Pinus contorta forests dominate at middle elevations over much of the central position of the moisture gradient, though these are primarily post-fire forests. With protection from fire only a small percentage of sites retain dominance by Pinus contorta. Over the lower portion of its range Pinus contorta is succeeded by Pseudotsuga, while at higher elevations Abies lasiocarpa and Picea engelmannii can eventually achieve dominance. At high elevations on all except the driest sites Picea engelmannii and Abies lasiocarpa are exclusive dominants, both after disturbance and in climax forests. Pinus flexilis dominates on the driest high-elevation sites. Above 3 500 m forests are replaced by alpine tundra, often with a transitional krummholz zone. Structure and post-fire development were examined in the context of the gradient-based classification scheme. Three generalized types of forest development were recognized as reference points in a continuum of developmental patterns varying with both elevation and soil moisture. On favorable, middle-elevation sites, trees become established rapidly after disturbance. Rapid growth results in severe overcrowding and competitive elimination of reproduction. As a consequence bell-shaped diameter distributions develop. Diversity and productivity appear to drop while biomass remains roughly constant. Following decades or even centuries of stagnation, the forests eventually breakup through mortality of the canopy trees, thereby allowing regeneration to resume. During this period of renewed regeneration, biomass, diversity, and productivity all show dramatic changes in response to the changing population structure (Fig. 9). This type of forest development can be found in forests dominated by Picea engelmannii and Abies lasiocarpa, Pinus contorta, Pseudotsuga menzeisii, Pinus flexilis or Populus tremuloides. On highest elevation forest sites or at middle elevations on the very driest sites reestablishment rates are greatly reduced. These forests dominated by Picea and Abies or Pinus flexilis gradually approach predisturbance levels of biomass, diversity and productivity, while regeneration remains at a roughly constant level. At lower elevations in the Pinus ponderosa woodlands, regeneration appears episodic, reflecting variation in seed rain and favorable conditions for seedling growth. Here, inter-tree competition is relatively unimportant and diameter distributions show irregular humps resulting from periodic recruitment. A few species pairs presented consistent problems and their treatment as single species was necessary. Garex rossii and C. brevipes were lumped as Carex rossii. Rosa woodsii and R. acicularis were lumped as Rosa sp. Cirsium scopulorum and C. coloradense were lumped as Cirsium coloradense. Extreme forms of Arnica cordifolia and A. latifolia are easily distinguishable, but as these species intergrade and hybridize extensively, they have been lumped as Arnica cordifolia. The native bluegrass, Poa agassizensis, was lumped with Poa paratensis. Solidago missouriensis includes some S. canadensis.

Competition and Tree Death

can flatten all the trees in their paths. More subtle events of synchronous tree mortality, such as those caused by defoliating insects or fungal diseases, may attract attention because of the ghostly skeletons that remain. But these dramatic forms of tree mortality are the exception; most tree deaths go unnoticed. At best only a few of the many thousands of seedlings produced by a typical mature tree can survive and grow to achieve canopy position and reproductive status. Most die as a direct or indirect consequence of failure to compete successfully for light, water, or soil nutrients. This form of tree death has come to be called natural thinning, or self-thinning when applied to a relatively even-aged population of a single species. These terms contrast the process with artificial thinning, where less vigorous or unwanted trees are selectively removed by man. Although individually the deaths attributable to natural

Succession: A Population Process

Recent critical reviews suggest the need for a reductionistic approach to the study of secondary plant succession. We propose viewing succession as the result of the underlying plant population dynamics. This approach is being developed using nearly 50 years of permanent sample plot records. After initial establishment Pinus taeda shows an exponential depletion with stands of various densities conforming to the reciprocal yield relationship. Uneven-aged hardwoods also show exponential depletion. Canopy disturbance can enhance the establishment process, though severe disturbance and the consequent abundant regeneration can lead again to dense, even-aged stands with low levels of establishment. These results suggest a general pattern of forest development wherein establishment is initially important, but is quickly replaced by mortality as the dominant process when the dense, even-sized stand starts to thin. Eventually, failing additional disturbance, natural mortality will again open the canopy allowing development of a balance between establishment, and mortality.

DIVERSITY AND INVASIBILITY OF SOUTHERN APPALACHIAN PLANT COMMUNITIES

We propose that the relationship between diversity and community invasibility depends on the degree to which community composition is driven by immigration processes. When immigration is enhanced by high propagule pressure or low-intensity disturbance, the relationship between diversity and exotic species invasion should be positive. Only when such immigration processes are limited should competitive interactions lead to a negative correlation between diversity and invasibility. Moreover, competition should be more apparent at smaller scales where individual plants compete directly for space; thus, diversity and invasibility are more likely to be negatively correlated at small spatial scales. We tested these predictions by comparing exotic and native species diversity of vascular plants across five spatial scales in riparian and upland plant communities in the southern Appalachians. We found a positive relationship between species diversity and exotic invasion in riparian areas at large scales (100 m2), w...

CONVERGENCE DURING SECONDARY FOREST SUCCESSION

(1) Successional convergence in community composition was examined in terms of three questions: (i) for a given site is there a continuous shift in composition toward that characteristic of climax? (ii) does variation in community composition along an environmental gradient increase or decrease with succession? (iii) to what extent is the species composition along a successional gradient determined by site characteristics rather than by chance factors? Hypotheses regarding the nature of successional convergence are reviewed in terms of these questions. (2) These hypotheses were tested using data for tree (>0.5 cm dbh) and herb (all foliage 80-yr-old pines, and uneven-aged hardwoods). For each age-group first axis ordination scores (detrended correspondence analysis) were highly correlated most consistently with soil pH as compared with other soil and site variables. This correlation was greatest in the intermediate-age (40-60and 60-80-yr-old) pines and in the hardwood stands, and lowest in old pine stands. The distribution of species in the ordination space was most similar between the intermediate-age pine stands and the hardwoods; in contrast, young and old pine age classes were dissimilar to hardwoods. Species distributions in relation to the soil pH gradient were also most similar between intermediate-age pines and hardwoods. Thus, there is a shift in species composition toward that characteristic of climax, but it is probably not monotonic. (3) Beta-diversity was highest in the hardwoods and lowest among the pines. Community differentiation along gradients increased in this chronosequence. (4) Predictability of soil pH, based on stand species composition, was highest in the intermediate-age pines and in the hardwoods. Habitat breadth in relation to soil pH decreased steadily with successional age. These results are consistent with the hypothesis that the role of chance factors (e.g. site history and seed rain) decreases with successional age.

Changes in chalk‐grassland structure and species richness resulting from selective nutrient additions

A series of fertilization experiments was carried out over a 5-yr period in a chalk grassland in Limburg (The Netherlands) as part of a study of the maintenance of species richness in species-rich grasslands. Phosphorus and nitrogen were shown to be the most limiting nutrients. Addition of both elements doubled above-ground production, and species rich- ness dropped ca. 50 % in 0.01-m 2 subplots, relative to controls. However, neither the above-ground production nor plant growth-forms were sufficient to explain the observed changes in species richness. Small-scale structural heterogeneity of the vegetation is probably critical for maintaining high levels of richness. Historically, high nitrogen, low phosphorus condi- tions were rarely encountered in the Dutch landscape and few species appear adapted to these conditions. Among the chalk grassland species, Brachypodium pinnatum seems well adapted to these conditions, where it dominates and excludes most other species. A detailed understanding of the small-scale processes responsible for maintenance of species richness is critically important in efforts to maintain the biodiversity of natural ecosystems.

Composition and species diversity of pine-wiregrass savannas of the Green Swamp, North Carolina

Fire-maintained, species-rich pine-wiregrass savannas in the Green Swamp, North Carolina were sampled over their natural range of environmental conditions and fire frequencies. Species composition, species richness, diversity (Exp H′, 1/C), and aboveground production were documented and fertilization experiments conducted to assess possible mechanisms for the maintenance of high species diversity in these communities. Although savanna composition varies continuously, DECORANA ordination and TWINSPAN classification of 21 sites facilitated recognition of 3 community types: dry, mesic, and wet savannas. These savannas are remarkably species-rich with up to 42 species/0.25 m2 and 84 species/625 m2. Maximum richness occurred on mesic, annually burned sites. Aboveground production, reported as peak standing crop, was only 293 g · m−2 on a frequently burned mesic savanna but was significantly higher (375 g · m−2) on an infrequently burned mesic site. Production values from fertilized high and low fire frequency sites were equivalent. Monthly harvest samples showed that savanna biomass composition by species groups did not vary seasonally, but within groups the relative importance of species showed clear phenological progressions. The variation in species richness with fire frequency is consistent with non-equilibrium theories of species diversity, while phenological variation in production among similar species and the changing species composition across the moisture gradient suggest the importance of equilibrium processes for maintenance of savanna diversity.

FOREST VEGETATION OF THE COLORADO FRONT RANGE: PATTERNS OF SPECIES DIVERSITY*

Plant species diversity patterns of the Rocky Mountain forests were found to be at variance with patterns reported from other regions. The most centrally located forests in terms of elevation, site moisture and successional status were found to have the lowest diversity. In contrast, the peripheral and environmentally more severe sites were found to have relatively high diversity. In particular, the forest-grassland transition and the low elevation riparian forests have species diversity values as high as any yet reported from western North America. When diversity was examined in terms of variation across elevation or moisture gradients, varying results were obtained due to the interaction of these factors. The failure of previous studies to converge on generalizations about plant diversity reflects, in part, the failure of most investigators to view diversity in a regional context of variation across several interacting gradients. Diversity was seen to vary inversely with the degree of development of the forest canopy. The interaction of different components of the forest community is one reason for the failure of general patterns of plant species diversity to emerge from previous studies. A potentially rich herb community can be greatly suppressed by a single species tree stratum. Among the most successful work to date on species diversity is that on birds, a distinct albeit large and functional group. It is unlikely that similar success could have been achieved through work on all animal species simultaneously. This suggests the need to examine plant species diversity, not in terms of total diversity, but in terms of component functional groups, perhaps guilds, growing under similar microclimatic conditions and subject to similar competitive pressures.