Catastrophic Disturbance Research Articles

Vegetation composition, structure, and biomass of two unpolluted watersheds in the Cordillera de Piuchué, Chiloé Island, Chile

Research in pristine forests provides a necessary reference of energy and nutrient cycling in absence of anthropogenic influence. Therefore two unpolluted watersheds in the Cordillera de Piuchue of southern Chile (42 ◦ 30 � S) were chosen for detailed ecosystem analysis. The goals of this study were to quantify the distribution of the living biomass in the research watershed and to document topographic gradients in the vegetation. Across a small spatial gradient from ravine bottom to ridgetop (approximately 60 m in elevation and < 300 m in length) in the Cordillera de Piuchue watersheds, there were significant shifts in vegetation composition, structure, and biomass. Based on sampling in 18, 100 m 2 plots, we identified three distinct community associations: a Fitzroya forest at the bottom of the watershed, a mid-slope Pilgerodendron-Tepualia transition zone, and a ridgetop moorland community. The Fitzroya forest was dominated by a cohort of approximately 400 year-old Fitzroya cupressoides trees. Both tree basal area (138 m 2 /ha) and total live biomass (656 Mg ha −1 ) reached a maximum in this vegetation type. The Pilgerodendron-Tepualia forest consisted of smaller, shorter, and younger trees with dominance shared by Pilgerodendron uviferum, Tepualia stipularis, and to a lesser extent, F. cupressoides. Basal area and biomass were half that of the Fitzroya forest (69.5 m 2 ha −1 ; 350 Mg ha −1 ) but tree density was 65% greater. The moorland can best be described as an open community of mosses and cushion plants that included low stature individuals of P. uviferum, F. cupressoides ,a ndT. stipularis. The size and age structure of F. cupressoides in the bottomland forest suggest that the current cohort of adult trees was established following a catastrophic disturbance and that F. cupressoides is unable to regenerate under its own canopy. In contrast, the size structure of the tree populations in the Pilgerodendron-Tepualia zone indicates that all the constituent tree species, except F. cupressoides ,a re able to reproduce at least at some microsites in the understory. The watershed-level means for live biomass were 306 Mg ha −1 of aboveground tree biomass, 25 Mg ha −1 of large root biomass (diameter ≥ 1 cm), and 46 Mg ha −1

Extinction Risk in Successional Landscapes Subject to Catastrophic Disturbances

We explore the thesis that stochasticity in successional-disturbance systems can be an agent of species extinction. The analysis uses a simple model of patch dynamics for seral stages in an idealized landscape; each seral stage is assumed to support a specialist biota. The landscape as a whole is characterized by a mean patch birth rate, mean patch size, and mean lifetime for each patch type. Stochasticity takes three forms: (1) patch stochasticity is randomness in the birth times and sizes of individual patches, (2) landscape stochasticity is variation in the annual means of birth rate and size, and (3) turnover mode is whether a patch is eliminated by disturbance or by successional change. Analytical and numerical analyses of the model suggest that landscape stochasticity is the most important agent. Landscape stochasticity increases the extinction risk to species by increasing the risk that the habitat will fluctuate to zero, by reducing the mean abundance of species, and by increasing the variance in species abundance. The highest risk was found to occur in species that inhabit patches with short lifetimes. The results of this general model suggest an important mechanism by which climate change threatens biodiversity: an increase in the frequency of extreme climate events will probably cause pulses of disturbance during some time periods; these in turn would cause wider fluctuations in annual disturbance rates and thus increase the overall level of landscape stochasticity. However, the model also suggests that humans can manipulate landscape stochasticity to reduce risk. In particular, if managed disturbances were more evenly distributed in time, attrition of the regional biota might be prevented. Other work on the connection between patch dynamics and extinction risk assumes the absence of landscape stochasticity and thus overlooks an important component of risk to biodiversity.

WildCountry

REGARDLESS of the merits and values of individual national parks and nature reserves, Australia's conservation reserves do not ensure the survival of the continent's biota. There are many reasons for this. Reserves, even the largest, are too small and vulnerable to broad area disturbance. Consider that, in January 2003, fires burnt more than two-thirds of Kosciuszko National Park, which at 690 000 ha is the largest park in New South Wales and one of the largest in Australia. This shows how even the largest conservation reserves are at risk of catastrophic disturbance. The much smaller Nadgee Nature Reserve (21 000 ha) in southeastern New South Wales has burnt almost in its entirety twice in the 35 years I have worked there. The Nadgee fires and those in Kosciuszko were started by lightning and were the result of prolonged drought, events common across the continent. When small size is coupled with isolation, the long-term survival of populations and the exchange of propagules within the reserve system becomes problematical. Small size and isolation do not leave much scope for plants and animals to adapt to long-term climate change, either through dispersal or by evolution. Even reserving 10 or 15% of land for nature conservation, as recommended by some international conservation agencies, will be inadequate; a target of 30% would have better ecological credentials, but even this could prove inadequate unless the nature conservation reserve system was designed to allow for long-term evolutionary change, which it is not (see Archer 2002; Recher 2002a,b).

A spatially explicit stochastic model to simulate boreal forest cover transitions: general structure and properties

We describe a spatially explicit simulation model of large-scale forest cover transition in the North American boreal biome. This model is a time-dependent Markov chain with discrete states corresponding to dominant tree species in the forest canopy. To parameterize the model, we used three temporal variables, extracted from published data from field studies: period of persistence of a given cover type, the time interval of forest cover replacement by another, and the time of complete replacement. Environmental domains based on climate, soil moisture regime, and soil nutrient status stratified all forest cover transitions. Probability matrices of forest cover transition were derived for 15 discrete states at 20-year intervals for each of the environmental domains. Five spatial databases from boreal North America were used as input to the model: forest cover composition, time since last forest cover change, climate zone, soil moisture regime, and soil nutrient status. The model output consists of spatially explicit prediction of (a) time since last disturbance, (b) transition age, (c) forest cover composition, and (d) canopy age in 20-year time steps at 1 ha resolution. As a case study, we simulated forest cover transition in a 3.7 million ha area in boreal Canada. These 200-year simulations show that spatio-temporal transition of forest cover type is significant even in the absence of catastrophic disturbances.

Forest dynamics in Westland, New Zealand: the importance of large, infrequent earthquake‐induced disturbance

Summary 1 Large, infrequent disturbances can exert a dominant influence on the structure and dynamics of forested landscapes and in Westland, New Zealand there is evidence of at least three massive earthquakes within the last 650 years. We reconstructed the history of forest disturbance in two study areas, totalling 1412 ha, to quantify the role of such disturbance in structuring the conifer/hardwood forests. 2 We divided the study area into different land-surface types, aged trees on each land surface and identified cohorts of trees established in response to past disturbance. The type of disturbance (tree fall or erosion/sedimentation event) that initiated cohort establishment was determined from the type of land surface and other physical evidence. We also dated abrupt growth releases or suppressions in tree rings to aid disturbance history reconstruction. 3 Erosion and sedimentation events dominated the disturbance regime, affecting 86% of the study area in the last 650 years. Catchment-wide episodes of forest cohort-initiating disturbances were centred around 1820‐30, 1710‐20, 1610‐20 and 1460 ad . Of the 51 cohorts identified in the study area, 47 were initiated during one of these episodes, when disturbance by erosion or sedimentation affected from 10‐50% of the study area. Consequently, over 80% of the forested area currently comprises simple, first generation cohorts of trees established after catastrophic disturbance. Only 14% of the study area is more complex, all-aged forest. 4 Three disturbance episodes coincide with the three most recent Alpine fault earthquakes ( c . 1717, 1630 and 1460 ad ), while one coincides with earthquakes recorded to the south of the study catchment in 1826 ad . Age structures from throughout Westland show that extensive, similar-aged, post-earthquake cohorts of trees are a feature of the region, suggesting that infrequent, massive earthquakes are the dominant coarse-scale disturbance agent, triggering episodes of major erosion and sedimentation and leaving a strong imprint on the forest structure.

Forest dynamics in Westland, New Zealand: the importance of large, infrequent earthquake-induced disturbance

1 Large, infrequent disturbances can exert a dominant influence on the structure and dynamics of forested landscapes and in Westland, New Zealand there is evidence of at least three massive earthquakes within the last 650 years. We reconstructed the history of forest disturbance in two study areas, totalling 1412 ha, to quantify the role of such disturbance in structuring the conifer/hardwood forests. 2 We divided the study area into different land-surface types, aged trees on each land surface and identified cohorts of trees established in response to past disturbance. The type of disturbance (tree fall or erosion/sedimentation event) that initiated cohort establishment was determined from the type of land surface and other physical evidence. We also dated abrupt growth releases or suppressions in tree rings to aid disturbance history reconstruction. 3 Erosion and sedimentation events dominated the disturbance regime, affecting 86% of the study area in the last 650 years. Catchment-wide episodes of forest cohort-initiating disturbances were centred around 1820–30, 1710–20, 1610–20 and 1460 ad. Of the 51 cohorts identified in the study area, 47 were initiated during one of these episodes, when disturbance by erosion or sedimentation affected from 10–50% of the study area. Consequently, over 80% of the forested area currently comprises simple, first generation cohorts of trees established after catastrophic disturbance. Only 14% of the study area is more complex, all-aged forest. 4 Three disturbance episodes coincide with the three most recent Alpine fault earthquakes (c. 1717, 1630 and 1460 ad), while one coincides with earthquakes recorded to the south of the study catchment in 1826 ad. Age structures from throughout Westland show that extensive, similar-aged, post-earthquake cohorts of trees are a feature of the region, suggesting that infrequent, massive earthquakes are the dominant coarse-scale disturbance agent, triggering episodes of major erosion and sedimentation and leaving a strong imprint on the forest structure.

Size distribution dynamics for a marine sessile organism with space‐limitation in growth and recruitment: application to a coral population

Summary Corals are clonal organisms and show a plastic growth. We study a partial differential equation model for the dynamics of size distribution of corals and predict the trajectory of recovery after a catastrophic disturbance, such as the recent bleaching that killed most corals in southern Japan. We assume that the mean growth rate of colony size, measured in projected area, is a linear function of colony size, and that the variance in growth rate is proportional to the size, which is consistent with the growth data of a coral Acropora hyacinthus Dana 1846. The model incorporates the space‐limitation in colony growth and recruitment. The growth rate and recruitment rate are proportional to the fraction of free space within the local habitat. In many corals, including A. hyacinthus, recruitment occurs in a short period once a year. However, our model illustrates that the colony size distribution does not show distinguishable cohorts since the observed variance in growth rate is large. The model with discrete settlement and a large variance in growth rate results in size distributions that are very well approximated by an explicitly soluble model with constant recruitment and no growth variance. When mortality is low, the dynamics of size distribution show two different phases in the recovery process. In the first phase, size distribution is determined by recruitment and growth, and can be predicted well by the case without mortality. After free space is depleted, recruitment and growth slow down and become balanced with mortality. The equilibrium size distribution is controlled by all the three processes. Both for the transient and the equilibrium size distribution, the average colony size increases with growth rate but decreases with recruitment rate. Strongly skewed colony‐size distributions in which small size classes have the largest numbers are generated for a wide range of parameters by the space‐limitation in growth, even without partial death of colonies.

Predators increase the risk of catastrophic extinction of prey populations.

There has been considerable research on both top-down effects and on disturbances in ecological communities; however, the interaction between the two, when the disturbance is catastrophic, has rarely been examined. Predators may increase the probability of prey extinction resulting from a catastrophic disturbance both by reducing prey population size and by changing ecological traits of prey individuals such as habitat characteristics in a way that increases the vulnerability of prey species to extinction. We show that a major hurricane in the Bahamas led to the extinction of lizard populations on most islands onto which a predator had been experimentally introduced, whereas no populations became extinct on control islands. Before the hurricane, the predator had reduced prey populations to about half of those on control islands. Two months after the hurricane, we found only recently hatched individuals--apparently lizards survived the inundating storm surge only as eggs. On predator-introduction islands, those hatchling populations were a smaller fraction of pre-hurricane populations than on control islands. Egg survival allowed rapid recovery of prey populations to pre-hurricane levels on all control islands but on only a third of predator-introduction islands--the other two-thirds lost their prey populations. Thus climatic disturbance compounded by predation brought prey populations to extinction.

Growth and development of the thinning canopy in a post-hurricane tropical rain forest in Nicaragua

Abstract The effect of catastrophic disturbance on forest structure was studied for 10 years subsequent to the landfall of Hurricane Joan in 1988. Four sites within the damage area and one control site outside of the area were established in the early 1990s and positions and sizes (DBH and crown height) were measured annually. Results of the study are reported in the context of the debate between equilibrium versus non-equilibrium models of tropical forest dynamics. Equilibrium models imply some sort of niche segregation on a species to species basis, whereas non-equilibrium models emphasise chance and history in the development of canopies. In either case, the details of the dynamics of building and thinning after a disturbance are key to resolve the debate. Physical aspects of the developing canopy began with a severely damaged forest with effectively no canopy at all. Two distinct canopies had developed 4 or 5 years after the disturbance; (1) a larger but diffuse canopy made up of the tress that had survived the hurricane standing and had begun resprouting high in their damaged crowns and (2) a ‘thinning canopy’ of very densely packed crowns of small individual trees, resulting from the growth of surviving saplings and seedlings and trees that had sprouted near to the ground. Competitive thinning has begun to occur in this dense lower canopy. Since the disturbance, the increase in number of species has been great, ranging from 134 to 208% over what would be expected in a mature forest. Mortality patterns in the thinning canopy suggest that trees are not dying at random and thus supports a niche-based model of forest regeneration. Qualitative analysis of the thinning canopy also suggests that some species have come to occupy the upper position in the canopy (and are thus likely to be competitive dominants) while other species have come to occupy the lower position (and are thus likely to be competitively suppressed). A qualitative model is proposed illustrating where, during the process of regeneration, various authors have emphasized either a niche (equilibrium) or non-niche (non-equilibrium) model. It is suggested that in the building phase of the post-disturbance forest a non-niche interpretation is most appropriate while in the thinning phase a niche interpretation may be warranted.

Simulating Catastrophic and Individualistic Large Woody Debris Recruitment for a Small Riparian System

Surprisingly little research has been done to partition the contribution of catastrophic disturbance from that of small-scale individualistic mortality events on riparian large woody debris (LWD) recruitment. This study compared the impact of both processes on recruitment through simulation of several catastrophic disturbances (a spruce beetle outbreak, a moderately intense fire, and a clearcut) and undisturbed (individualistic mortality only) old growth for a small headwater stream in the Intermountain West of the United States. All scenarios progressed through a two-stage process, with the Forest Vegetation Simulator growth and yield model controlling forest dynamics and a postprocessor (CWD, version 1.2) predicting riparian LWD recruitment. Projections indicate that individualistic-only conditions delivered 2.5 m3 LWD·100 m reach−1·10-yr cycle−1; while the spruce beetle-, fire-, and clearcut-affected stands averaged 2.9, 3.2, and 1.5 m3 LWD·100 m reach−1·cycle−1, respectively. Stands impacted by natural catastrophic disturbance significantly (P < 0.05) increased cumulative (300 yr) LWD recruitment over the individualistic-only scenario, whereas clear-cutting significantly decreased total delivery. In-stream LWD loads, relatively stable in undisturbed riparian zones, fluctuated sharply under catastrophic disturbance. Peak channel loads associated with natural perturbation occurred ∼30 yr after the event while debris volumes under clear-cutting immediately declined. The postevent recruitment and in-stream LWD stocks of all disturbance scenarios eventually fell below undisturbed conditions, requiring decades to recover historical volumes. Catastrophic disturbances induced such steep oscillations in riparian LWD load that the systems experiencing frequent large-scale perturbations never achieved a long-term steady state, as some have postulated. Because of the inflation in cumulative LWD delivery, it may prove advantageous to encourage (or imitate) some catastrophic disturbance in forests along streams noticeably depauperate of LWD.

SIMULATING CATASTROPHIC AND INDIVIDUALISTIC LARGE WOODY DEBRIS RECRUITMENT FOR A SMALL RIPARIAN SYSTEM

Surprisingly little research has been done to partition the contribution of catastrophic disturbance from that of small-scale individualistic mortality events on riparian large woody debris (LWD) recruitment. This study compared the impact of both processes on recruitment through simulation of several catastrophic disturbances (a spruce beetle outbreak, a moderately intense fire, and a clearcut) and undisturbed (individualistic mortality only) old growth for a small headwater stream in the Intermountain West of the United States. All scenarios progressed through a two-stage process, with the Forest Vegetation Simulator growth and yield model controlling forest dynamics and a postprocessor (CWD, version 1.2) predicting riparian LWD recruitment. Projections indicate that individualistic-only conditions delivered 2.5 m3 LWD·100 m reach−1·10-yr cycle−1; while the spruce beetle-, fire-, and clearcut-affected stands averaged 2.9, 3.2, and 1.5 m3 LWD·100 m reach−1·cycle−1, respectively. Stands impacted by natural catastrophic disturbance significantly (P < 0.05) increased cumulative (300 yr) LWD recruitment over the individualistic-only scenario, whereas clear-cutting significantly decreased total delivery. In-stream LWD loads, relatively stable in undisturbed riparian zones, fluctuated sharply under catastrophic disturbance. Peak channel loads associated with natural perturbation occurred ∼30 yr after the event while debris volumes under clear-cutting immediately declined. The postevent recruitment and in-stream LWD stocks of all disturbance scenarios eventually fell below undisturbed conditions, requiring decades to recover historical volumes. Catastrophic disturbances induced such steep oscillations in riparian LWD load that the systems experiencing frequent large-scale perturbations never achieved a long-term steady state, as some have postulated. Because of the inflation in cumulative LWD delivery, it may prove advantageous to encourage (or imitate) some catastrophic disturbance in forests along streams noticeably depauperate of LWD.

A gravel lobe deposit in the prodelta of the Doumsan fan delta (Miocene), SE Korea

Abstract The Doumsan fan-delta system in the Pohang Basin (Miocene), SE Korea comprises tripartite components of Gilbert-type topset, foreset, and bottomset environments with an extended prodelta in the deeper part (a few hundred metres deep). The present study documents sedimentologic features and origin of a gravel body (here interpreted as a gravel lobe) formed in the prodelta region of the Doumsan fan delta. The gravel lobe lies on sandy mud deposits and is capped by a thick massive sand bed. It extends for more than 1.5 km with a height of up to 30 m and shows a narrow tongue-like geometry. Eight sedimentary facies have been distinguished to describe characteristic features of the gravel lobe and the associated deposits. Of these, three types of gravelly sedimentary facies are important with regard to volumetric contribution and depositional processes: (1) crudely stratified pebble-grade conglomerate; (2) disorganized, clast-rich pebble(-to-cobble)-grade conglomerate; and (3) matrix-rich, bimodal cobble-grade conglomerate. The former two types dominate the central part of the lobe where they are not accompanied by sand beds, whereas the latter, as subordinate units, is prevalent in the fringe which otherwise is dominated by thick sandy mud deposits. The stacked successions of crudely stratified pebble-grade conglomerate are representative of the active aggradational phases of the gravel lobe, whereas the occurrence of channels within the lobe reflects that the gravel lobe prograded under the influence of subaqueous channel systems. The gravel lobe resulted from catastrophic disturbance (slumping) on the foreset region that further caused the development of channel systems, promoting efficient transport of gravelly sediments. This type of deposit may represent an important additional category of low-efficiency subaqueous fans.

Chapter 11. Inherent Disturbance Regimes

Summary A science-based ecosystem management approach requires valid reference points to assess the long-term maintenance of forest systems. Historical range of variability (HRV) in vegetation patterns has served as the initial reference point and has support in the coarse-filter approach to conserving biodiversity (Hunter 1991). The HRV becomes less useful as a reference with increasing human disturbance on the landscape, public aversion to treatments to restore historical states, and continued public expectations for non-historical conditions. The inherent disturbance regime (IDR) is put forward as an alternative or to be used in combination with HRV as a suitable reference point for evaluating long-term maintenance of human-altered ecosystems. The IDR is defined by the types of disturbance; their frequency, intensity, and extent in turn defines the vegetation composition and structure supported over time. Although public expectations for vegetation conditions may differ from historical, these altered states may also be supported by IDR and have long-term maintenance potential. The use of a disturbance reference (IDR) rather than a condition reference (HRV) appears more compatible with the dynamic nature of forests. As an accurate description of historical conditions and disturbance regimes is difficult we recommend that both references be used reiteratively to better refine the other. Increased discontinuity between public expectations for forest conditions and the ability of those conditions to be supported by disturbance regimes may lead to catastrophic disturbance events and a decline in long-term site nutrient capital and biodiversity.

Effects of topography on rainforest tree community structure and diversity in American Samoa, and implications for frugivore and nectarivore populations

Summary Aim We investigated the spatial variation of rainforest tree community structure and composition to determine if forest structure and diversity varied as a function of topography; and in turn if this could influence patterns of habitat use by native forest birds and pteropodid bats. Location The study was undertaken on the island of Tutuila, American Samoa, located in the South Pacific Ocean. Methods All trees ≥10 cm diameter were censused in sixty 200 m2 plots in ridge, slope and valley forest across the island of Tutuila. ResultsForest structure varied significantly across topographical space. Ridge forest was shortest and had the highest stem densities, and valley forest was tallest with the fewest stems per unit area. Species richness was highest on ridges, and slope and valley forest were more similar in composition with each other than they were with ridge forest. Of the fifty‐two tree species encountered in the plots, nine showed a statistical affiliation to one of the three topographical positions. Main conclusionsWe explain patterns of forest structure and diversity in the context of chronic and catastrophic disturbances. Higher stem densities in ridge forest suggested a higher degree of disturbance on ridges, and this was supported by the fact that the height/diameter ratio of the forest was lowest on ridges, which indicated wind‐cropping. We hypothesize the potential effects of topographical variation and known phenological patterns on wildlife abundances. We predict that flowering episodes of ridge‐affiliated, bird‐visited species (particularly Syzygium inophylloides (A. Gray) C. Muell.; Myrtaceae) will concentrate honeyeater densities on ridges, and that fruiting of the tree Canarium vitiense A. Gray (Burseraceae) could localize populations of the Pacific pigeon (Ducula pacifica). Overall (i.e. net) bat foraging patterns are unlikely to be affected by either flowering or fruiting events. Most of the tree species on Tutuila are generalist in their demographic patterns, and the island is depauperate in wildlife fauna; the evolutionary and conservation implications are discussed. We conclude with the argument that conservation of vertebrate species is essential to maintain the current generalist demographic patterns of Samoan trees.

Colonization of new habitats by benthic foraminifera: a review

Colonization of new habitats, which have been established as a result of a catastrophic disturbance of the environment, is one of the characteristic repetitive events throughout the Phanerozoic. In recent years, much attention has been paid to investigations focusing on biological recovery of benthic habitats severely disturbed by human activity. In order to improve our environmental and stratigraphical interpretations of such events, we need a more thorough understanding of the processes involved in colonization by one of the most abundant and useful fossil groups, the benthic foraminifera. The present review focuses on processes governing benthic foraminiferal dispersion and colonization patterns in modern environments. For benthic foraminifera, the only active dispersal mechanism is through self-locomotion on or within the sediment and this is considered to be efficient over short distances only. Several passive dispersal methods have been suggested but two seem to be of more general importance. These are dispersal through release and transport of embryonic juveniles and passive suspension and transport of various growth stages. Both are probably important for most benthic foraminifera but the former is likely to be the main mechanism for attached, tubular and larger foraminifera, which are not easily entrained at a later life stage. The latter seems to be a more important dispersion mechanism for benthic foraminifera than previously realized. The colonization rate of soft-bottom substrates is closely related to the hydraulic regime in, and the transit time from, the source area inhabited by species capable of colonizing the new habitat (as long as food and other environmental characteristics are not limiting factors). The transit time depends on the speed of the transporting medium and the distance from the source area. There seems to be two end-processes which can operate during the colonization, depending on whether physically induced or biological processes are allowed to dominate. They are characterized by different colonization patterns. In high energy environments (bottom current velocities often >20 cm/s), a short transit time may cause the major components of the nearest ambient seafloor assemblages to colonize the new habitat within days. In this case the colonization is simply through a physical transfer of parts of the source community to the new habitat, allowing no time for pioneer, opportunistic assemblages to develop. In low energy environments (bottom current velocities generally <10 cm/s), the transit time is long for most species. Here, colonization follows the classic metazoan successional pattern with an initial, high abundance pioneer assemblage strongly dominated by small opportunists followed by development of assemblages with increasing numbers of specialized species and recovery can take from one to several years. Initial lack of food (e.g., volcanic ash) or `hostile' substrate properties (e.g., recently reoxygenated or severely contaminated sediments) may delay colonization by months or even years. Small, infaunal species (both calcareous and agglutinated) are among the first and most successful colonizers of soft bottom habitats from shallow waters to the deep sea. Throchospiral agglutinated taxa are among the most abundant colonizers on deep sea hard substrates.

Stand dynamics of Dacrydium cupressinum dominated forest on glacial terraces, south Westland, New Zealand

Stand structure, disturbance histories, and patterns of tree replacement of three conifer-dominated stands growing on glacial terraces, differing in age, were investigated. Regeneration patterns of the conifers Dacrydium cupressinum, Manoao colensoi, Prumnopitys ferruginea and Phyllocladus alpinus were inferred from age and size (diameter) distributions, the spatial distribution of tree ages, and tree locations. Disturbance histories and patterns of tree replacement were reconstructed from age distributions, the spatial distribution of tree ages, and tree locations. Disturbance histories were related to soil drainage which was inferred from the depth of rusting on mild steel rods inserted across each stand. Stands differed in structure and composition, reflecting different patterns of disturbance. Manoao colensoi and Dacrydium cupressinum were established in response to catastrophic disturbance on extremely poorly drained soils on the oldest terrace. On less poorly drained soils, Dacrydium cupressinum, Prumnopitys ferruginea and Phyllocladus alpinus were established in response to smaller canopy gap formation from intermittent windthrow and synchronous canopy collapse. A secular model of stand dynamics is envisaged across the glacial terraces, reflecting an even long-term soil development, where the chance of catastrophic disturbance will increase as soils become increasingly poorly drained over millennia. The implications of these findings for sustainable forest management are discussed.

On the causes of gradients in tropical tree diversity

1 The number of woody species in tropical forests tends to increase with precipitation, forest stature, soil fertility, rate of canopy turnover and time since catastrophic disturbance, and decrease with seasonality, latitude, altitude, and diameter at breast height (d.b.h.). 2 A model is presented to account for these trends. Novel hypotheses include how increased rainfall and substrate fertility, and decreased seasonality, might (i) increase attacks by natural enemies, and thus the overall level of density-dependent plant mortality; (ii) increase shade tolerance, canopy turnover, and stem density of the species-rich understorey; and (iii) increase reliance on relatively sedentary forest-interior birds for seed dispersal, fostering high rates of speciation in understorey genera. 3 High rainfall and low seasonality in the tropics favour two key groups of natural plant enemies – insects and fungi – that are directly responsible for promoting high rates of density-dependent plant mortality. Lower rainfall, greater seasonality, soil infertility, or unfavourable rooting conditions favour greater allocation to anti-herbivore defences, and thus lead to lower rates of such mortality and thence to lower tree diversity. The increased number of individuals on rainier sites is a minor contributor to increased tree diversity, accounting for only about 17% of the 8.3-fold increase with rainfall in the lowland Neotropics. 4 Predictions of the model are consistent with many ecological patterns of variation in tropical tree diversity within regions, and may help explain the decrease in tree diversity with elevation and the accompanying decrease in horizontal patchiness (within-habitat β diversity). 5 Random drift over evolutionary time in the relative effectiveness of density-dependent control of individual tree species by specialized natural enemies may better account for the observed distribution of tropical tree abundance than a random walk of species abundance through ecological time.

PATTERNS OF ICE ACCUMULATION AND FOREST DISTURBANCE DURING TWO ICE STORMS IN SOUTHWESTERN VIRGINIA

Major ice storms deposit heavy loads of freezing rain on trees, causing catastrophic forest disturbance in eastern North America. During 1994, two ice storms struck southwestern Virginia within a three-week period. Heavy ice deposition occurred during each storm when warm, moist air was advected over a shallow subfreezing layer at the surface. Cold-air damming helped trap the cold surface air, as is typical for ice storms in parts of the Appalachians and Piedmont. Heaviest forest damage occurred on mountain slopes facing south and east. Ice damage in these topographic positions removed approximately 30% to 60% of total basal area and created large canopy openings. Several processes may have contributed to topographic variations in ice accretion, including orographic effects on rainfall and influences of wind on twig surface and raindrop thermodynamics. [Key words: ice storms, freezing rain, forest disturbance, Virginia, Appalachian Mountains.]

Catastrophic disturbance in the presettlement forests of the Upper Peninsula of Michigan

Catastrophic disturbance in the presettlement forests of the Upper Peninsula of Michigan

Catastrophic disturbance in the presettlement forests of the Upper Peninsula of Michigan

The General Land Office (GLO) survey notes (1840-1856) were used to examine the interaction among natural disturbance, vegetation type, and topography in the presettlement forests of the Luce District, an ecological unit of approximately 902 000 ha in the Upper Peninsula of Michigan, U.S.A. The surveyors recorded 104 fire and 126 windthrow incidences covering 3.1 and 2.8% of the total length of the surveyed lines, respectively. The rotation periods over the entire landscape were 480 years for fire and 541 years for windthrow, but these varied with vegetation type and topographic position. Fire occurred more frequently on southerly aspects and at elevations where pinelands were concentrated. The density of windthrow events increased with elevation and slope, with the highest occurrence on westerly aspects. Based on the estimated rotation periods, we calculated that 7.5, 24.4, and 68.1% of the presettlement forest were in the stand initiation, stem exclusion, and old forest (including both understory reinitiation and old growth) stages, respectively. Pinelands and mixed conifers were the major components in both the stand initiation (34.5 and 31.1%) and the stem exclusion stage (20.9 and 39.8%), while mixed conifers (39.3%) and northern hardwoods (34.7%) were the major old-forest cover types. The diverse mosaic of various successional stages generated by natural disturbance suggests a "shifting-mosaic" landscape in this region.