The basic unit of the forest-tundra landscape is a toposequence extending from a wet, forested valley to a xeric, deforested hilltop; the contact zone between these two environments being called a subarctic tree line. Dendrochronological analysis of living, dead, and subfossil black spruce, and radiocarbon dating of peat samples were used to reconstruct the dynamics of a subarctic tree line since its post-fire origin about 1000 y ago. Fire is not the sole disturbance to have influenced the dynamics of the toposequence. A regional-scale flooding event ca. 1120 AD killed many black spruce trees, growth of permafrost during the Little Ice Age, and its subsequent degradation in the 20^th century, also had major consequences. The climate was favourable to black spruce growth between ca. 300 and 1100 AD, as evidenced by large growth rings and tree growth forms. Ring widths then decreased markedly between the 12^th and 19^th centuries and trees were replaced by stunted growth forms. Although climate warming during the 20^th century resulted in increased ring widths, black spruces have still not produced tree growth forms, a necessary condition for viable seed production and eventual re-colonization of deforested hilltops.

Fire regimes in the boreal forest are dominated by crown fires that burn over large areas. However, these fires rarely burn forest stands entirely and with the same intensity throughout, resulting in a mosaic of vegetation burnt to varying degrees of severity. The objectives of this study were to: (1) assess regeneration of tree species six or seven years after fire in relation to crown fire severity in non-salvaged jack pine and black spruce stands, and (2) assess establishment preferences of seedlings on the different types of germination beds created by fire. Logistic regressions indicated that seed trees abundance influenced regeneration success. The relationship between fire severity at the crown level and regeneration success was not significant, although seedling recruitment appeared limited in areas where severity was light to moderate. Poisson regressions showed that seedlings preferentially establish on mineral soil, and that woody debris seem to be a good substrate for germination and survival. Seedlings establish more frequently and grow better where thickness of residual organic matter is lowest. Crown fire severity, combined with severity at ground level may therefore be a good indicator of regeneration success in coniferous stands.

The effects of fire severity and initial post-fire tree composition on long-term stand structural development were investigated in the Picea mariana–feathermoss bioclimatic domain of northwestern Québec. Paleoecological methods were used to categorize the severity of the last fire (high or low) and initial tree composition (Picea mariana versus Pinus banksiana). Changes in stand structure were evaluated by quantifying stand structural attributes along three chronosequences. Except for accelerating stand break-up, the post-fire presence of P. banksiana (which is eventually replaced by P. mariana) had little effect on stand structural development. Fire severity had significant effects on the evolution of stand structural attributes, with low severity fires being particularly detrimental for stand productivity. Stands colonizing low severity fires were characterized by low post-fire tree recruitment and growth and remained open throughout succession. In contrast, after high severity fires, dense productive stands were rapidly established regardless of tree composition and gradually became open as succession proceeded. These results suggest that in the prolonged absence of fire, the different stand structural development pathways gradually converge regardless of fire severity or initial composition. We argue that stand structural diversity within the coniferous boreal forest is a result of the severity of the last fire and of processes operating at the stand scale in the absence of fire.

Alberta's boreal mixedwood forest has seen intensifying industrial activity in the past several decades, largely from logging and petroleum extraction. At the same time, populations of North American beaver (Castor canadensis) have been recovering from past near-extirpation. We conducted detailed field surveys of six beaver dam sites on low-order streams in northeastern Alberta and examined a 50-y chronosequence of air photos at each site to quantify beavers' effects on riparian forests. Beaver activity increased the width and diversity of riparian zones along first- and second-order streams. Over the 50-y time sequence, dam number increased considerably and beaver activity converted narrow, entirely lotic habitats to a mix of lentic and lotic. Current forestry operating ground rules in Alberta require 30- to 60-m unharvested buffer strips on permanent streams. Around dams, beaver felling removed most or all Populus trees within 30–40 m of the pond edge. The abundance of dams and their tendency to be built in chains altered vegetation structure over long stretches of riparian corridors. Beavers thus could be removing forest cover from entire buffer strips in direct conflict with forest management objectives. We argue that beavers may be the primary disturbance agent structuring riparian zones on low-order streams in the study area and that unharvested riparian buffer strips should be much wider than currently prescribed in order both to provide beaver habitat and to ensure appropriate protection of riparian habitats.

The influence of stand age and site conditions on the structure of coniferous stands was studied in the boreal forest of Québec's Côte-Nord, a region with low fire recurrence. Stand diameter diversity was measured in 2202 forest inventory plots in black spruce (Picea mariana), balsam fir (Abies balsamea), and mixed stands using the Shannon-Wiener diversity index. A relative productivity index was developed based on the relationship between height and age of dominant trees. A stepwise regression analysis indicated that this productivity index best explains stand structure variation in all composition types, while stand age seems to influence structure more at the beginning of stand development. The results suggest that productive stands become uneven-sized earlier than unproductive stands and also maintain a greater diameter diversity. These contrasting structural dynamics may be explained by (i) a higher growth rate in richer stands that likely induces earlier senescence and thus an earlier passage to an uneven-sized structure and (ii) a restricted maximum tree diameter in poor stands caused by a scarcity of resources, which in turn reduces the diameter diversity of these stands, even after their break-up time.

The first prerequisite for old-growth forest conservation or restoration is an adequate understanding of the structure and development of old, unmanaged forests. In order to describe the structure and development of old Picea abies-dominated forests we selected a study area from the Paanajärvi wilderness in northwestern Russia. Forest structural characteristics (tree height, diameter, position, age, etc.) were measured in 20 study plots in a sampling grid located north of Paanajärvi National Park. A chronosequence of different forest ages was used to study the late successional development of the forests. The volume of living trees was highest in the youngest age class (110–140 y), whereas that of dead wood increased with forest age. Betula and other deciduous trees occurred as a mixture in all forest age classes, but their proportion of the total volume decreased considerably with forest age. Forest floor microhabitat distribution was most diverse in the oldest age class, where the number of saplings was also highest. Our results show that from 110 to 300 y of age the northern Picea-dominated forests undergo significant structural and compositional changes. In conservation and management of old boreal forests, care should be taken to preserve the full range of this old-growth structural habitat diversity.

Species composition and number of species in ground vegetation after windthrow varies depending on damage severity and management actions after a storm event. In this paper we seek to determine the changes in species composition depending on the severity of storm damage. The vegetation response was studied by comparing areas with different levels of windthrow damage (undamaged, partly damaged, totally damaged, and totally damaged and harvested) in Norway spruce–dominated forests in Estonia. A total of 108 herbal and shrub species were found in all areas. Species number was lowest in the control areas and highest in the harvested plots. The number of species that responded to disturbance severity, when compared to the control plots, increased during the second year of the two-year survey period. The groups of species responding to different degrees of storm severity and silvicultural treatments were identified. Early invasion after a storm was observed in the case of several species, such as Epilobium angustifolium, Rubus idaeus, and Ranunculus repens. All the ferns that responded to damage extent seemed to be most favoured in totally damaged areas. Rubus saxatilis showed the clearest response trend, with higher coverage in response to increasing wind severity.

The spatial patterns of Scots pine recruitment and ground layer vegetation are heterogeneous in dry pine forests. The goal of our 1994–2004 work in the dry pine forests of Russian Karelia was to find the reasons for this heterogeneity at the meso-and fine levels of scale. Twenty-seven sample plots (30 × 30 m) with different tree age structures, stand densities, and times since fire were examined. In each sample plot the tree layer, saplings, and ground layer were described in detail using a coordinate grid. Needle, bark, and branch debris was also measured. Detrended canonical correspondence analysis (DCCA) and analysis of variance (ANOVA) confirmed the conclusion of earlier studies that time since fire and stand structure determine the formation of recruitment and ground layers at the patch level. Ripley's K-function analysis showed that the spatial pattern of recruitment was clustered in both even-aged and uneven-aged forests. The bivariate analysis showed a repulsion effect between mature trees and recruitment in even-aged forests, and attraction between mature trees and saplings in uneven-aged forests. Analyses of tree layer, recruitment, and ground layer spatial patterns in the sample plots proved that the spatial pattern of regeneration and its intensity at the fine scale were related to the ground layer structure. At this scale, ground layer community structure was generally determined by local canopy projection and the amount of needle, branch, and bark debris.

Populations of Betula pendula and Betula pubescens in felling stands subject to different brushing regimes were studied in the southern taiga forest (Tsentralno-Lesnoi Biosphere Zapovednik, Russia). The stands are 14- and 20 y old and are situated in the Oxalis type of spruce forest. The number of birch saplings, the height and ontogenetic stage (reflecting the biological age) of individual saplings in the community were determined in ten 5- × 10-m plots, five in a 14-y-old stand that had undergone a single brushing event and five in a 20-y-old stand that had undergone three (regular) cleanings. In the 14-y-old stand, birches were more numerous (the total number of birch saplings was 10,800.0 ± 1365.3 stems·ha⁻¹), saplings were higher (B. pendula: 2.59 ± 0.08 m; B. pubescens: 1.80 ± 0.12 m), and populations were more mature (most B. pendula saplings had changed to tree stage v from bush stage im). The total number of saplings in the 20-y-old stand was 1120.0 ± 338.2 stems·ha⁻¹, the mean height of B. pendula was 2.47 ± 0.45 m, the mean height of B. pubescens was 1.45 ± 0.14 m, and saplings in immature stages dominated. Natural forest regeneration was dominated by B. pendula, which was almost twice as abundant as B. pubescens in the 20-y-old stand, and three times more abundant in the 14-y-old stand.

Remnants of old forests left on the landscape following forest harvesting, especially corridors, provide benefits of connectivity and facilitation of movement or dispersal, which may be hindered by the presence of edges. Our objective was to determine the extent of edge influence on forest structure in these forest remnants in black spruce boreal forest. We sampled canopy cover and the density of trees, snags, and logs along clearcut edge–forest gradients in large forest patches, cutblock separators, and riparian buffers. The distance-of-edge influence was determined by comparing values at different distances from the edge to values in interior forest using randomization tests. Forest remnants had lower live tree density and canopy cover and higher mortality and windthrow than interior forest. Distance-of-edge influence on forest structure extended 10–30 m from the edge, and was slightly more extensive into cutblock separators where two edges are in close proximity, but was less extensive in riparian buffers, possibly due to the presence of stable internal edges near the stream. Because of edge influence, structure near the edges of forest remnants and across narrow corridors is modified; wider corridors would be required to provide a core habitat of interior forest conditions.

The last decade of innovation in forest management in Canada is reviewed. Institutions such as the Sustainable Forest Management Network and Canada's Model Forest Program have attempted to develop a better understanding of ecological disturbance patterns and processes. Additional research has explored socio-economic dimensions of sustainable forestry, such as ways to incorporate the aspirations of indigenous peoples, build community capacity, and facilitate forest certification. The most promising innovations tend to have both environmental benefits (sustaining non-timber values) and economic benefits (reducing costs and sustaining future timber values), making their implementation more likely. Some on-the-ground examples of “win-win” solutions at stand and landscape levels in Canada's boreal forests include: patch retention in conjunction with the creation of large cutblocks; protection of advance regeneration during timber harvesting; promotion and prediction of natural regeneration; various approaches to mixedwood (broadleaf and conifer) management; avoidance of unnecessary brush control; extended rotations and selection management for some tree species and stand types; promoting the flow of fibre to its highest value uses; and zoning in support of intensive silviculture, thereby potentially reducing harvesting pressures from lands with high conservation value. More closely emulating natural patterns of forest disturbance and forest recovery can help sustain biodiversity and ecosystem services, but may not generate all values desired from managed forests. Further research is needed to calibrate indicators of ecological sustainability. Institutional and policy innovation must also be evaluated in the context of adaptive management to improve the effectiveness of forestry practices and nurture the social license for the utilization and management of public forests.

Local species richness in highly fragmented landscapes is expected to be affected by characteristics of both the considered patch and its surrounding matrix, especially in areas dominated by human activities. However, the proximity of other suitable habitat patches may also be determinant for the conservation of biodiversity by enhancing dispersal and thus, “rescue effects” and re-colonization. We investigated breeding bird communities on 63 patches suitable for wildlife in the suburbs of Paris, France, in 2003. After exploring the spatial structure of the data, we tested hypotheses relating local species richness to patch characteristics (size and vegetation), urbanization intensity around patches, and patch isolation relative to the nearest natural remnant. Spatial dependence in the data was considered in the regression analyses because patch distribution may affect local species richness through ecological processes and non-independence between patches situated nearby can pose statistical problems. Fifty-seven regular breeding bird species were recorded over the entire study area. Local bird species richness, patch size and vegetation descriptors were not spatially autocorrelated. Conversely, urbanization intensity around the patches and patch isolation were spatially autocorrelated. Species richness was mainly related to patch size and to the diversity of trees and shrubs, while neither herbaceous plants nor patch distance to the nearest large forest remnant had significant effects. Moreover, we found that the degree of urbanization around patches did not provide a strong predictor of bird richness, which is more newsworthy. Overall, this study stresses once more the complexity of factors and processes responsible of community patterns in urban landscapes. It also highlights the need to explicitly consider spatial issues when investigating these patterns.

We manipulated nutrient availability and competing vegetation to investigate the phenotypic plasticity of leaf morphology and nutrient content of saplings and seedlings of the evergreen holm oak Quercus ilex subsp. ballota in response to nutrient pulses. Different factorial experimental designs were established with N and/or P pulse fertilization as factors in a field experiment in Catalonia (northeast Spain) and with P fertilization, soil type, and competitive situation as factors in a pot experiment. In the field, P fertilization decreased leaf mass per area (LMA) 9–13% in the 3 y following the fertilization application. No significant effect was observed from variations in N supply in any leaf cohort. In the pot experiment, P fertilization increased the leaf size, pre-dawn leaf water content, and root growth and decreased the LMA. P fertilization strongly increased P leaf concentrations, as in the field experiment. Neighbour competition induced increases in LMA in the second leaf cohort and in leaf density and size in both leaf cohorts; it also reduced root growth and tended to increase P leaf concentrations. Soil type had a significant effect on individual leaf area, leaf mass, and LMA, the latter being higher in siliceous soils than in calcareous and artificial soils. The pattern of change in leaf thickness across the different soil types was the opposite of that for leaf P concentrations, P contents per unit of leaf area, and leaf density. These results point to a morphological leaf plasticity in response to P availability, competitive situation, and soil type. The relations between the changes in leaf morphology, leaf nutrient content, root growth, and pre-dawn leaf water content changed depending on the factor that induced them. However, leaf morphological traits changed in combination with other plant traits such as root growth and nutrient contents in order to balance resource availability and production capacity under different environmental conditions, such as changing degrees of neighbour competition or water availability. This plasticity of sclerophyllous leaf traits in response to nutrient pulses makes Quercus ilex subsp. ballota well suited to adapt to the unpredictable changes and stress situations likely to result from current and immediate future climate and other changes in the Mediterranean region.

We investigated regional diatom species distribution, local diversity, and spatial variation in body size from 105 boreal stream sites in Finland. Stream diatom body size distributions ranged from a slightly right-skewed pattern at the largest study extent, comprising the whole country, to symmetrical distributions within watersheds. Compared to earlier findings for larger organisms, diatom body size distributions were more symmetrical, implying either narrower study extent or fundamental differences between unicellular and multicellular organisms. There was a weak positive correlation (r = 0.239; P < 0.05) between latitude and average maximum cell size in stream diatom assemblages. Multiple linear regression showed that average maximum cell size was best accounted for by a model incorporating moss cover and latitude. There was also a weak (r = 0.290; P < 0.01; n = 105) negative correlation between local species richness and average cell size corroborating the earlier findings for other organism groups.

In recent decades, high-altitude balsam fir (Abies balsamea) forests of northern New England and adjacent areas have been fragmented by canopy openings associated with several stand disturbances. To document the causes of extensive tree dieback and mortality and the source of canopy gaps, we studied the tree growth patterns of live balsam fir from five stands in four sites in the northeastern United States (NUS): Whiteface Mtn. in New York state and Mt. Moosilauke (two stands), Mt. Blue, and Mt. Lafayette in New Hampshire. Additionally, tree-ring patterns of dead fir were documented at two sites, Whiteface Mtn. and Mt. Mégantic (southern Québec). Organic horizons from four of the five NUS stands were analyzed for identification of insect remains and plant macrofossils in 11 and seven soil cores, respectively. Four of the five balsam fir tree-ring chronologies from the NUS stands showed a major growth decline in the 1970s, which was also apparent in the two chronologies from dead trees at Whiteface Mtn. and Mt. Mégantic. At the NUS stands, spruce budworm (Choristoneura fumiferana) head capsules were found in all 11 soil cores from forest and open stands. In spite of large variations in the number of head capsules among levels (range: 1–34; mean: 9.5 ± 8.8 for 71 levels analyzed), their occurrence at all the levels analyzed indicates that the insect had sustained activity through time. Macrofossil remains from the organic horizons included plant assemblages similar to the aboveground forest vegetation. Rubus idaeus, an intolerant species, indicated that gap conditions prevailed sometime in the past. Our tree-ring and macrofossil data along with insect surveys from northern New England provide evidence for the major role played by spruce budworm in tree dieback and mortality during infestations and subsequent gap formation in high-altitude fir forests. The impact of other stand disturbance factors is discussed.