ALBERT

Description: This paper reports on the behaviour of 10 experimental crown fires conducted between 1997 and 2000 during the International Crown Fire Modelling Experiment (ICFME) in Canada's Northwest Territories. The primary goal of ICFME was a replicated series of high-intensity crown fires designed to validate and improve existing theoretical and empirical models of crown fire behaviour. Fire behaviour characteristics were typical for fully developed boreal forest crown fires, with fires advancing at 1570 m/min, consuming significant quantities of fuel (2.85.5 kg/m2) and releasing vast amounts of thermal heat energy. The resulting flame fronts commonly extended 2540 m above the ground with head fire intensities up to 90 000 kW/m. Depth of burn ranged from 1.43.6 cm, representing a 25%65% reduction in the thickness of the forest floor layer. Most of the smaller diameter (

Description: Manually designing road networks for planning purposes is labour-intensive. As an alternative, we have developed a computer algorithm to generate road networks under a variety of assumptions related to road design standards. This method does not create an optimized road network, but rather mimics the procedure a professional might use when projecting roads by hand. Because many feasible road networks are possible, sensitivity analysis is required to choose the best ones. Such analysis gives forest planners additional information with which to assess the long-term consequences of road density and road standards common in forest management decisions. The procedures used to create road networks are presented in this paper, along with a sensitivity analysis of assumptions on total network length, percentage of landings connected, grades, and horizontal and vertical alignment for a case study. We also include a sensitivity analysis of spatial detail such as node density and link characteristics. Although the road network generation algorithm requires manipulation of many input parameters to create desired road networks, and variation between outputs is a concern, the method still offers considerable improvement over manual methods, especially for applications in strategic planning, and appears to be suitable for all types of topography and road standards.

Description: The relationship between charcoal production from fires and charcoal deposition in lakes is poorly understood, which limits the interpretation of sediment charcoal records. This calibration study assessed charcoal particle production, size, and transport during the International Crown Fire Modelling Experiment (ICFME) and compared fossil charcoal particle accumulation from 16 lakes in boreal forests of North America. Particle accumulation averaged 20.1 mm2·cm2 inside the ICFME fire; accumulation declined sharply outside the fire, with only 1% of the measured particles transported beyond 20 m from the burn edge. Fossil charcoal accumulation during the past 9000 years was much lower than observed deposition in traps located within the ICFME fire but similar to airborne deposition in traps located 1060 m from the burn edge. A higher fraction of large diameter particles (〉1 mm) was present in fossil charcoal accumulation from historical fires and charcoal peaks that exceeded background accumulation by 1.4 times, suggesting large particles are characteristic of nearby fires. On the basis of a charred-particle production of ~2% of the total fuel consumed by the ICFME fire, we estimate a potential long-term carbon sequestration of 58.2 ± 12 g C·m2 as charred particles from this fire stored in soils or lake sediments.

Description: We have developed extensions of traditional distance-dependent, spatial competition analyses that estimate the magnitude of the competitive effects of neighboring trees on target tree growth as a function of the species, size, and distance to neighboring trees. Our analyses also estimate inter- and intra-specific competition coefficients and explicitly partition the competitive effects of neighbors into the effects of shading versus crowding. We tested the method using data from forests of northern, interior British Columbia dominated by western hemlock (Tsuga heterophylla (Raf.) Sarg.) and western redcedar (Thuja plicata Donn ex D. Don). For both species, the most parsimonious regression models included terms for the effects of tree size, crowding, and shading and separate competitive effects of four different groups of competing species. The models explained 33%59% of the variation in radial growth of the two species. For both species, growth declined more steeply as a function of crowding than shading. There was striking asymmetry in the strength of interspecific competition between hemlock and redcedar, with crowding by hemlock having a strong per capita effect on redcedar, while crowding by redcedar had relatively little effect on the radial growth of hemlock.

Description: Crown symptoms and other aboveground variables were examined on 36 Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco.) (40209 years old), 46 white fir (Abies concolor (Gord. & Glend.) Lindl.) (36165 years old), and 97 ponderosa pine (Pinus ponderosa Dougl. ex Laws.) (64220 years old) trees in northern New Mexico and southern Colorado. Root systems of all trees were excavated to determine extent of root disease. Symptoms observed on infected trees, including reductions in height growth, changes in foliage characteristics, and crown dieback, worsened as the number of infected roots increased. Trees with aboveground symptoms had a significantly higher (p 〈 0.05) number of infected lateral roots than trees without symptoms. In mixed conifer stands on the Archuleta Mesa, Colo., four qualitative crown symptoms were used to accurately detect Armillaria ostoyae (Romagn.) Herink and (or) Heterobasidion annosum (Fr.) Bref. infection of Douglas-fir (21/22, or 95%) and white fir (19/28, or 68%). Similarly, 61% (48/79) of the A. ostoyae infected ponderosa pine trees on the Jemez site, N.M., were detected using the qualitative Thomson vigor rating system. Discriminate analysis, using more thorough variables and analysis, resulted in correct infection classifications of 82%, 85%, and 78% for Douglas-fir, white fir, and ponderosa pine, respectively, suggesting that aboveground variables are reasonable indicators of root disease.

Description: Stem, branch, needle, and total aboveground biomass were assessed for three 9- to 12-year-old white spruce (Picea glauca (Moench) Voss) plantations, each subjected to three different stand tending options at age 4 to 7. Biomass components were predicted from measures of stem diameter with coefficients of variation between 24% and 29%. Diameter at breast height (DBH) generally provided lower prediction precision than did the lower stem measures tested (coefficient of variation 〉 35%). The addition of tree height in models reduced the standard error of the estimates for stem and total biomass by an average of 48% and 8%, respectively, and compensated for different height/diameter ratios imposed on the spruce by the stand tending treatments. Needle and branch biomass models were invariant to the tending treatments and, consequently, to the addition of height as an independent variable. Predictions from existing published white spruce equations suggest that extrapolation to this study area would have led to adequate stem biomass estimation but to serious (〉55%) underestimates of branch, needle, and, correspondingly, total biomass. Slow self-pruning by plantation spruce, particularly before crown closure, is cited as a possible reason for these differences.

Description: The postfire regeneration dynamics of black spruce and jack pine were documented by a study of three successive cohorts (woody debris, snags, seedlings) within a large area burnt in 1989. The objectives of this study were (i) to describe how fire interval can influence the abundance of regenerating black spruce and jack pine and (ii) to model the future abundance trends of these two species for fire cycles of different lengths. The transition probabilities after fire were calculated for mixed stands of black spruce and jack pine for fire intervals of 47 and 67 years in well-drained sites and for fire intervals varying between 92 and 270 years in poorly drained sites. These probabilities were incorporated into a model of regeneration dynamics that took into account the drainage type, the regeneration potential, and the natural mortality rate of both species. After the 1989 fire, jack pine seedlings made up 55%82% of the regeneration in well-drained sites and 11%40% in poorly drained sites. Model simulations show that fire intervals 220 years lead to that of jack pine. The simulation results also suggest that jack pine could expand its populations under a fire cycle of 50 years or after short fire intervals during longer fire cycles. Thus, in the short term and medium term (i.e., ca.

Description: Saproxylic beetles associated with Populus coarse woody material were sampled from two age classes of fire-origin aspen stands in north-central Alberta, Canada. A combination of rearings from wood bolts and window traps attached to snags yielded 9571 beetles representing 257 taxa over the 3-year period (19931995). We investigated faunal variation across regions, stand ages (mature, 6090 years; old, 〉100 years), decay classes, wood types, and years in terms of species richness, abundance, and trophic differences. Although trophic structure was similar, faunal composition differed between the two study regions. Species richness and abundance were similar across stand ages; however, many species were collected exclusively or in great majority from old stands and from snags of large diameter, which suggested that truncation of stand age structure through widespread industrial harvest could have serious consequences for saproxylic assemblages. Beetle species richness increased with the level of wood decay, whereas the total catch of beetles tended to be higher in early stages of decay. Wood borer abundance tended to be higher in snags; however, total species richness was higher in logs. Our analyses suggest that (i) many beetle species in the Canadian boreal forest depend directly upon standing and fallen large-diameter woody material from Populus trees, (ii) variation in stage of decay is critical to beetle diversity, and (iii) provision for retention of representative old stands is critical to conservation of saproxylic communities.

Description: The effect of site properties and forest drainage on the dominant height development of Scots pine (Pinus sylvestris L.) stands in peatland sites was studied using data from permanent sample plots located in natural and drained sites in northern Finland. The Korf model was used to describe the height development of dominant trees in natural sites. The effect of drainage on height development was accounted for by a term giving a nonlinear height increase for drained sites as a function of the time elapsed since drainage. The variance component model was applied to account for the hierarchical data structure. Natural height development after 30 years of age at DBH was significantly slower in PF sites (sparsely forested pine fens) than in PS sites (genuine forested pine swamps). Within PF sites, there were further differences in relation to nutrient availability. Temperature sum explained the variation in the intercept. In PS sites, drainage resulted in a moderate increase in the maximum attainable height, while in PF sites, drainage improved site productivity by 80%85% in terms of the attainable height. The asymptote for drained stands was dependent on stand age at the time of drainage. Differences between the two major groups were assumed to be due to initial differences in site water regime.

Description: We evaluated the effectiveness of 4-allylanisole (4AA) as a protective treatment for loblolly pines threatened by the southern pine beetle, Dendroctonus frontalis Zimmermann. Three products were evaluated in combination with two methods that promoted attack of trees by D. frontalis. One method used attractive semiochemicals and the other decreased host resistance, both of which are important scenarios for implementing protective treatments of individual trees. Each method promoted mass attack of unprotected control trees, thus providing a statistically verifiable challenge to the candidate protectants. In trees with increased susceptibility, mortality ranged from 63% (untreated) to 77% (4AA applied in paintball formulation), and two products appeared to alter the relative composition of scolytid species that attacked at two heights; however, tree mortality was unaffected. In trees challenged with semiochemical attractants, mortality ranged from 54% (4AA released from vials) to 82% (untreated and paintball application of 4AA). Although 4AA consistently reduces catch of D. frontalis in traps, it was not efficacious for protecting individual loblolly pines over a period of 30 or 60 days in this experiment.

Description: If equilibrium is assumed in unmanaged forests, the volume of coarse woody debris (CWD), VCWD, may be calculated from (i) the volume of living trees, Vliving, (ii) average volume of a dead stem in relation to when it was alive, k, (iii) tree mortality rate, m, and (iv) residence time of CWD, t, by the equation VCWD = Vlivingkmt. We parameterized this equation with data from Norway spruce (Picea abies (L.) Karst.) dominated forests in Fennoscandia. The Vliving was assumed to be directly proportional to forest productivity. Tree mortality data were from the National Forest Inventory, while it was difficult to find quantitative data on k and t. The predicted amounts (74138 m3/ha, with larger amounts in the south) and size distribution (a negative exponential distribution of the number of stems) of CWD corresponded fairly well to averages from field inventories. By using a computer simulation program, the variability in tree mortality, density of living trees, and residence time of CWD were considered. In the simulations, the amount of CWD varied widely between 1-ha plots, especially for individual decay classes. Therefore, this model could be used to predict averages from larger landscapes unaffected by large disturbances, while no model can predict the amount of CWD at individual plots.

Description: The northeastern boreal forest of Quebec is characterized by a humid climate. Consequently, fires are less frequent and small-scale disturbances play an important role in forest dynamics. Natural mortality and nonfire disturbances such as insect outbreaks and windthrow lead to gap-driven processes. Changes in structure and species composition can result from gap dynamics. The objectives of this study were to characterize gaps and examine patterns of species replacement in gaps in old conifer stands. Line intersect sampling was used to sample stands dominated by balsam fir (Abies balsamea (L.) Mill.) and (or) black spruce (Picea mariana (Mill.) BSP). Results show that 54% of the forest was in expanded gaps and that canopy gaps are relatively small, since 87% of them were smaller than 100 m2. The majority (94%) of the openings were caused by the mortality of less than 10 gap makers. Replacement probabilities show self-replacement of A. balsamea in Abies stands and of P. mariana in Picea stands. However, in Abies-Picea stands, there seems to be a reciprocal replacement of the two species. These results provide knowledge of the disturbance dynamics of the region as a basis for development of silvicultural practices that preserve the structural components of older forest stands.

Description: The effects of seasonal prescribed fire on the belowground ectomycorrhizal community and live fine root biomass were investigated before, 1 year after, and 2 years after prescribed underburning. Ectomycorrhizas were sampled from four replications of three treatments (fall underburning, spring underburning, and a nonburned control) in a randomized complete block design. Samples were separated in two subsamples representing the upper 5 cm and lower 5 cm of a soil core. Molecular tools were used to distinguish 140 restriction fragment length polymorphism (RFLP) species of fungi directly from the ectomycorrhizas. Prior to underburning, the number of RFLP species and amount of live root biomass were similar among treatment units and between upper and lower core samples. Fall underburning largely removed live root biomass to a depth of 10 cm and significantly reduced ectomycorrhizal species richness compared with spring underburning and the nonburned control for at least 2 years. RFLP species richness and live root biomass following spring underburning were generally similar to the nonburned treatment. The successful reintroduction of fire to the ecosystem to retain high species diversity of ectomycorrhizal fungi and achieve the desired future condition of large-tree ponderosa pine retention with low fuel loads may require more than underburning in a single season.

Description: Several xylophagous insect species have adapted to recurrent fires in boreal forests and use high-quality habitats created by these disturbances. To characterize the xylophagous insect assemblages of fire-killed black spruce and their patterns of substratum use, eighty-four 40 cm long bole segments were cut in 2000 and 2001 according to tree diameter, segment height, and fire severity criteria in a 1999 burn in the Grands-Jardins provincial park, Quebec, Canada. The segments were suspended in rearing cages, and neonates were collected until November 2001. The cerambycid Mono chamus scutellatus (Say) and the scolytids Dryocoetes affaber (Mann.) and Polygraphus rufipennis (Kirby) were the most common beetles collected. For all common taxa, more neonates emerged from larger-diameter trees. Few neonates emerged from the upper parts of the trees, and none of the species were specialist of the upper parts of the tree. Fire severity had a drastic effect, and heavily charred trees yielded very few insects. The effect of fire severity on insect colonization density varies widely among tree species. This effect may be linked to varying bark thickness and to bark's insulating potential against water loss during the fire. The host's vigor before its death, measured from growth rings of the last 10 years, had a positive effect on cerambycid emergence, but no effect on scolytids.

Description: We studied the effect of tree species and fertilization on the forest floor microbial community of 15-year-old regenerating forests. We sampled F and H forest floor layers of plots planted to Thuja plicata (Donn ex D. Don.) or Tsuga heterophylla (Raf.) Sarg. on N-poor and N-rich sites, with and without fertilizer treatments. Microbial community composition was assessed using phospholipid fatty acid analysis and by enumerating populations of culturable bacteria and fungi. Potential microbial functioning was assessed using community-level physiological profiling. There was little differentiation of community-level physiological profiles of F and H layers and few differences among the treatments. Total microbial biomass was greater in the F than H layer, and the two layers had distinct phospholipid fatty acid profiles. Site effects were detected mainly in the residual H layer, and tree species effects were seen mainly in the F layer, which has developed since harvesting. The effect of fertilization depended on site and tree species, with very little response in cedar plots, and the greatest effects in hemlock plots, coinciding with the greater growth response of hemlock. These results indicate that differences in plant growth rates, rather than direct effects of fertilization, influenced the microbial communities.

Description: This paper develops methods for estimating leakage from forest-based carbon projects that seek to reduce carbon emissions from timber harvesting in tropical forests. A theoretical framework is presented in which a specific country, in this case Bolivia, is treated as a supplier to the global timber market. Leakage is measured, over a 30- to 50-year time period, as the difference in net national carbon emissions from timber harvesting between the baseline case and a scenario in which some of the land is removed from the concession base. Estimates of timber leakage are made for several different assumptions about future global sequestration policies, capital constraints, demand elasticity, and deadwood decomposition rates. The results suggest that leakage could range from 5% to 42% without discounting carbon, and from 2% to 38% when carbon is discounted. Demand elasticity and wood decomposition rates have the largest effects on the leakage calculation. Leakage is lowest when demand is more elastic and wood decomposition rates are faster, and vice-versa when these conditions are reversed. Leakage appears to be sensitive to capital constraints only when project benefits are measured over a shorter time period.

Description: Advances in technology for in vitro propagation and genetic transformation have accelerated the development of genetically engineered trees during the past 15 years. At least 33 species of transformed forest trees have been successfully regenerated to date. Targeted traits include herbicide tolerance, pest resistance, abiotic stress tolerance, modified fiber quality and quantity, and altered growth and reproductive development. Commercial potential has been demonstrated in the field for a few traits, in particular herbicide tolerance, insect resistance, and altered lignin content. Now that commercial implementation is feasible, at least for the few genotypes that can be efficiently transformed and propagated, environmental concerns have become the main obstacle to public acceptance and regulatory approval. Ecological risks associated with commercial release range from transgene escape and introgression into wild gene pools to the impact of transgene products on other organisms and ecosystem processes. Evaluation of those risks is confounded by the long life span of trees, and by limitations of extrapolating results from small-scale studies to larger-scale plantations. Issues that are central to safe deployment can only be addressed by permitting medium- to large-scale release of transgenic trees over a full rotation. Current regulations restricting field releases of all transgenes in both time and space need to be replaced with regulations that recognize different levels of risk (as determined by the origin of the transgene, its impact on reproductive fitness, and nontarget impacts) and assign a commensurate level of confinement. The next step in determining acceptability of transgene technology for forest tree improvement is the unconfined release of constructs that pose little risk in terms of gene escape and nontarget impacts, such as lignin-altered poplar or pine, to permit evaluation of ecological risks and environmental or agronomic benefits at relevant scales.

Description: Root cohesion is an important parameter governing slope stability in steep forested terrain. Forest harvesting impacts root cohesion, and although the temporal effects have been noted, this dynamic parameter is often assumed to be spatially uniform. A model was developed to simulate the variation in root cohesion on a hillslope resulting from various forest management treatments. The model combines physical data on the horizontal rooting distribution of Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco var. menziesii) together with a temporal relation of root cohesion decay. Harvesting methods examined include clear-cutting, single-tree selection cutting, and strip-cutting. Model outputs are analysed qualitatively for regions of root cohesion minima and quantitatively for the average root cohesion within the simulated hillslope. A selection cutting simulation maintained the highest average root cohesion value, decreasing to only 81% of the preharvest condition. In contrast, the minimum root cohesion following clear-cutting declined to 38% of the preharvest value. Selection and strip-cutting scenarios resulted in smaller areas of reduced root cohesion that were adjacent to areas with high root cohesion. Such partial cutting methods shorten the period of reduced root cohesion following timber harvesting compared with clear-cutting.

Description: A process-based model (MAIDEN, modeling and analysis in dendroecology) is compared with statistical response functions to analyze the climategrowth relationship for two different species (sessile oak (Quercus petraea (Matt.) Liebl.) and Aleppo pine (Pinus halepensis Mill.)) growing in contrasting climates (temperate and Mediterranean). Growth data were obtained using dendroecological sampling procedures. Results obtained with both methods are consistent, but the MAIDEN model provides a more explanatory analysis than classical statistical tools. Analyses with MAIDEN indicate that oak stomatal conductance is sensitive to atmospheric vapor deficit, whereas precipitation and soil water deficit are not limiting factors in the study area. In contrast, the soil water regime is the major controlling factor of bole growth variability for Aleppo pine. The model shows that remobilization of the carbohydrate reserves in spring is of major importance for leaf and root expansion, and it affects the subsequent bole increment of oak. Because the amount of carbon that can be mobilized in spring is linked to carbon stored the previous year, these processes explain how the growth of oak trees is linked in this way from one year to the next. In contrast, the MAIDEN model shows that carbohydrate-reserve dynamics, phenology, and bole growth are less connected for Aleppo pine trees.

Description: We investigated impacts of harvesting on soil disturbance and vegetation in the aspen cover type of northern Minnesota, United States. The soil disturbance (resistance to penetration) and understory vegetation were characterized for 19 sites on five 60-m2 plots placed along a disturbance gradient: landings (high harvesting traffic), skid trails (intermediate harvesting traffic), and areas off skid trails (low to no harvesting traffic). Penetration levels were quite variable, but they still indicated that within-site responses to disturbance patterns created by clear-cut harvesting were not uniform. In general, soil disturbance and understory species composition within landings were similar to those with skid trails. The soil disturbance and vegetation composition of these two levels differed from those of the low-disturbance plots (off skid trails), indicating that removing trees alone did not affect vegetation composition as much as creating an established skid trail, regardless of harvest timing. However, sites with more variable species composition (winter-harvested sites) and lower disturbance levels were less altered than sites with likely lower initial diversity (summer-harvested sites). The results suggest that it is important for recovery of understory plant communities to not only limit the amount and level of disturbances but also consider the spatial layout of harvesting, thus maintaining a spatially connected network of remnant forest patches large enough to contain interior forest species.

Description: This study examines the short-term impact of forest soil compaction and organic matter removal on soil mesofauna, in general, and oribatid mite species, in particular. Both soil compaction and organic matter removal reduced the density of soil mesofauna. Stem-only harvesting reduced total mesofauna densities by 20% relative to uncut forest values. A combination of whole-tree harvest and forest floor removal with heavy soil compaction significantly reduced total soil mesofauna densities by 93% relative to the uncut forest control. Removal of the forest floor represents a substantial loss of habitat for most soil mesofauna. The forest floor apparently buffered the mineral soil by limiting both the impact of soil compaction and fluctuations in soil temperature and moisture. The relative abundance of Prostigmata and Mesostigmata increased with treatment severity, whereas that of Oribatida decreased. Species richness of the oribatid mite fauna was reduced as the severity of treatments increased. The number of rare oribatid species (those representing

Description: An individual-tree heightdiameter model was developed for stone pine (Pinus pinea L.) in Spain. Five biparametric nonlinear equations were fitted and evaluated based on a data set consisting of 8614 trees from 455 plots located in the four most important regions where the species occurs in Spain. Because of the problem of high correlation among observations taken from the same sampling unit, a mixed-model approach, including random coefficients, is proposed. Several stand variables, such as density, dominant height, or diametric distribution percentiles, were included in the model as covariates to explain among plot variability. To determine interregional variability among the regions studied, regional effects were included in the model using fixed dummy variables. Two models, one for inland regions and one for coastal regions, were found to be sufficient to explain regional variability in the heightdiameter relationship for the species in Spain. Mixed models allow predictive role in two ways: a typical response, including only fixed effects, and a calibrated response, where random effects are predicted and included in the model from the prior measurement of the height in a subsample of trees. Different alternatives were tested to determine optimum subsample size. Measurement of the height of the 20% largest trees in the plot has been shown to be a useful approach.

Description: Most studies of postfire tree recruitment have occurred in severely burned portions, despite the fact that partial burning is common. In this study we examined regeneration following a 1997 fire in the boreal forest of Quebec. A model of postfire recruitment was elaborated using parameters such as the proportion of trees killed (severity), the proportions of postfire seedbed types and their associated juvenile survivorship, the available seed supply, the available bud supply (for Populus tremuloides Michx.), and the granivory rate. All three species had peak recruitment in the first or second summer, and the recruitment episode was essentially finished after the third year. Mineral soil and surviving Sphagnum were the best seedbeds for both conifer species. Seedbed frequency was essentially independent of crown fire severity except for surviving Sphagnum, which was concentrated primarily where severity was light. Conifer fecundity was much lower in the lightly burned stands, a result we attribute to a higher granivory rate. The fecundity (seedlings/basal area for the conifers or suckers/basal area for Populus) in the severe sites was typical of the few other North American studies of postfire recruitment, where the published data permit us to make the comparison.

Description: The construction of diameter-distribution models sometimes calls for the simultaneous prediction of population parameters from hierarchical data. Appropriate data for such models have characteristics that should be preserved or accommodated: clustering and contemporaneous correlations. Fitting techniques for such data must allow for these characteristics. Using a case study, I compare two techniques seemingly-unrelated regression (SUR) and principal components analysis (PCA) whilst using mixed-effects models. I adapt and apply a metric that focuses on volume prediction, which is a key application for diameter distributions. The results suggest that using mixed-effects models provides useful insights into environmental variation, and that SUR is more convenient and produces a slightly better fit than PCA. Both techniques are acceptable with regard to regression assumptions.

Description: The problem we consider is annual harvesting planning from the perspective of Swedish forest companies. The main decisions deal with which areas to harvest during an annual period so that the wood-processing facilities receive the required amount of assortments. Each area has a specific size and composition of assortments, and the choice of harvesting areas affects the production level of different assortments. We need to decide which harvest team to use for each area, considering that each team has different skills, home base, and production capacities. Also, the weather and road conditions vary during the year. Some roads cannot be used during certain time periods and others should be avoided. The road maintenance cost varies during the year. Also, some areas cannot be harvested during certain periods. Overall decisions about transportation and storage are also included. In this paper, we develop a mixed integer programming model for the problem. There are binary variables associated with harvesting, allocation of teams, and road-opening decisions. The other decisions are represented by continuous variables. We solve this problem directly with CPLEX 8.1 within a practical solution time limit. Computational results from a major Swedish forest company are presented.

Description: In this study we used dendrochronology to reconstruct the history of eastern spruce budworm (Choristoneura fumiferana (Clem.)) outbreaks over the last 450 years in the Bas-Saint-Laurent region of southeastern Quebec. In total, 260 tree cores were sampled from 204 beams in seven historic buildings and 12 trees in a virgin forest stand. Eight previously documented outbreaks (19751992, 19471958, 19141923, 18681882, 18321845, 18051812, 17521776, 17101724) and three presumed previous outbreaks (16781690, 16421648, 15771600) were identified based on periods of growth reduction. Of these 11 confirmed or presumed outbreaks, six were documented for the first time in eastern Quebec. Such data suggest that outbreak frequency has remained quite stable, with a mean interval of about 40 years between the midpoint of successive outbreaks since the mid-16th century. In addition, together with previous studies, our results indicate a strong spatial synchrony of spruce budworm outbreaks across central and eastern Quebec during the last 300 years. Consequently, our study does not support the hypothesis that spruce budworm outbreak frequency and synchrony increased during the 20th century.

Description: At the landscape scale, one of the key indicators of sustainable forest management is the age-class distribution of stands, since it provides a coarse synopsis of habitat potential, structural complexity, and stand volume, and it is directly modified by timber extraction and wildfire. To explore the consequences of several landscape-scale boreal forest management strategies on age-class structure in the Mauricie region of Quebec, we used spatially explicit simulation modelling. Our study investigated three different harvesting strategies (the one currently practiced and two different strategies to maintain late seral stands) and interactions between fire and harvesting on stand age-class distribution. We found that the legacy of initial forested age structure and its spatial configuration can pose short- (

Description: Short rotation forestry (SRF) is regarded as a silvicultural practice employing high-density plantations of fast-growing tree species on fertile land with a rotation period of fewer than 1012 years. I address the challenges and possibilities of SRF applications under the circumstances of a boreal climate, today as well as after anticipated climate change. The implications of a pronounced winter season for the performance of biomass crops are discussed. Poplars, aspens, and willows are superior in boreal SRF because of their fast growth rate in combination with good cold hardiness. These trees can enrich the coniferous forests of boreal regions and increase biodiversity in open agricultural landscapes of the boreal zone. Further, SRF plantations can serve as tools for the amelioration of environmental problems at local (e.g., phytoremediation) and global (e.g., increased greenhouse effect) scales. The biomass yields achieved in boreal SRF and the appropriate production systems appear do not appear to be principally different from warmer regions, but there are some differences with respect to the importance of fertilization, appropriate spacing, and rotation length. The major barriers for a rapid development of SRF appear not to be climatic, technical, or environmental constraints in many boreal regions.

Description: Replicated circular openings ranging in size from 0.1 to 1 ha were cleared in 1996 at Blodgett Forest Research Station, California, and planted with seedlings of six native species. After 5 years of postharvest growth, heights were measured and analyzed according to species, opening size, and location within opening. The sequence of mean height from tallest to shortest, according to species, was as follows: giant sequoia (Sequoiadendron giganteum (Lindl.) Buchholz) 〉 incense-cedar (Calocedrus decurrens (Torr.) Florin) 〉 Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco var. menziesii) ≈ ponderosa pine (Pinus ponderosa Dougl. ex Laws.) 〉 sugar pine (Pinus lambertiana Dougl.) ≈ white fir (Abies concolor (Gord. & Glend.) Lindl.). To describe the influence of openings size on seedling height, we use an information-theoretic approach to select from competing models that predicted fifth-year height from group selection opening size. Asymptotic fits (modeled with MichaelisMenton curves) were selected for giant sequoia, ponderosa pine, sugar pine, and incense-cedar. Quadratic fits were selected for white fir and Douglas-fir. Linear models predicting increasing growth with opening size were consistently ruled out for all species. Although a marked depression in seedling-height growth occurred along the edges within the openings, mean annual radial increment of the 90-year-old border trees surrounding the openings increased by 30%, compared with other canopy trees in the forested matrix between openings.

Description: This paper describes the Zone Allocation Model (ZAM) that uses the simulated annealing algorithm to create forest management zones. ZAM partitions the landscape into the Timber, Habitat, and Old Growth zones by allocating small land tiles into contiguous areas. The zone allocation process is guided by landscape-level targets and size and shape objectives. An ecological representation objective proportionally distributes all ecosystem types into each of the three zones. Priority objectives control allocation of identified lands that are targeted for specific zones. All objectives are combined within an objective function, with a penalty-weighting system specifying relative importance of each objective. The ZAM model found 1.7%4.4% of theoretical optimum scores from small to large problems, respectively. A demonstration on a 1.2 × 106ha landscape from coastal British Columbia illustrates the iterative exploration of compromises between objectives that leads to informed zone allocation decisions.

Description: A multi-objective optimization model was created for medium-term forest development planning for an integrated forest products company located in the East Kootenay area of British Columbia, Canada. First, a set of sustainable forest management criteria and indicators were developed based on information that could be collected from regional geographic information sytem (GIS) databases and potential outputs from the model. Next, a new forest development planning unit was created (stewardship unit) in which adjacent forest polygons with similar indicator attributes were aggregated. The planning model was designed to determine appropriate harvest levels and management treatments on each stewardship unit to satisfy objectives determined in a participatory process. The mathematical model uses a fuzzy MAXMIN approach, where each indicator represents an objective in the model. Indicators are valued in the model using targets, thresholds, and triggers (called the 3-T approach). A case study is used to demonstrate the use of the model in a sustainable forest development planning context. The results of the case study show that the planning area is highly sensitive to visual quality, old-growth, and community watershed indicators. The paper concludes with a sensitivity analysis that determines the relative opportunity cost of various sustainable forest management indicators on company profits, employment, and tax revenues.

Description: We synthesize current information on input, accumulation, and decay of coarse woody debris (CWD) compared with other aboveground litter to assess the role of CWD in the nutrient cycles of northern coniferous forests. CWD contributes between 3% and 73% of aboveground litter input, but

Description: We compare rates of decay of foliar litters of British Columbia tree species in two field studies, and assess which initial litter chemistry parameters best predict the decay rates. Nutrient concentrations, tannins, and carbon fractions (based on proximate analysis and nuclear magnetic resonance spectroscopy) were measured in fresh litter of 14 tree species in one experiment and seven species in a second experiment. Each study was replicated in a different site in order to assess the transferability of results. Broadleaf litters decayed faster than needle litters only during the first year; thereafter, they decayed slower. Lignin concentration was a good predictor of mass loss only during the first year and only in one of the two experiments, which may have resulted from all foliar litters having high lignin concentrations (〉170 mg·kg1). Litter chemistry effects on first-year decay were consistent and transferable among sites. None of the initial litter chemistry parameters were good predictors of mass remaining after 4 or 5 years, because mass loss of most litters was similar by this time. The convergence in mass losses of litters after 45 years despite initial differences indicates that decomposition estimates extrapolated from early rates or initial chemistry may not accurately predict long-term decay.

Description: A Gaussian mixture model (GMM) is used to classify Forest Inventory and Analysis (FIA) plots into six ecological habitats in the northeastern USA. The GMM approach captures intra-class variation by modeling each habitat class as a mixture of subclasses of Gaussian distributions. The classification is achieved based on the appropriate posterior probability. The GMM classifier outperforms a traditional statistical method (i.e., linear discriminant analysis or LDA), and produces similar overall accuracy rates to a commonly used neural network model (i.e., multi-layer perceptrons or MLP). For the classifications of individual ecological habitats, however, MLP produces better (or same) producers' classification accuracies for five of the six ecological habitats than does GMM. But the GMM's accuracy rates are more consistent (92%97%) across the six ecological habitats than those of the MLP model (82%99%). This study shows that GMM offers an attractive alternative for modeling the complex stand structure and relationships between variables in mixed-species forest stands.

Description: The effects of initial leaf litter chemistry on first-year decomposition rates were studied for 16 common Pacific Northwest conifers, hardwoods, and shrubs at the H.J. Andrews Experimental Forest in western Oregon. Leaf litters were analyzed for C, N, P, K, Ca, Mg, proximate organic fractions (nonpolar, polar, acid-hydrolyzable extractives, acid-hydrolyzable lignin, and acid-unhydrolyzable residue, previously termed "Klason lignin"), and biochemical components (total phenolics, reactive polyphenols, water-soluble carbohydrates, water-soluble proanthocyanidins, and water- and acid-unhydrolyzable proanthocyanidins). By including measurements of reactive and residual phenolic fractions and acid-hydrolyzable lignin, these analytical methods improve upon traditional proximate leaf litter analyses. Significant differences in litter chemistries and decomposition rates were found between species. For all species combined, the 1-year decay rate (k) values had highly significant correlations (P 〈 0.001) with 30 out of the 36 initial chemistry variables tested in this study. The three highest correlations were with acid-unhydrolyzable proanthocyanidins, lignocellulose index, and acid-unhydrolyzable residue (r = 0.83, 0.81, 0.80, respectively, with P 〈 0.0001 and n = 339). We found that no single litter chemistry variable was a universal predictor of the 1-year k value for each of the individual 16 species studied, though phenolic components were more frequent significant (P 〈 0.001) predictors of decomposition rate.

Description: Demographic data from a 15-year outbreak of the spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae), in a boreal mixedwood forest in Ontario, Canada, are used to interpret stand-level ecological disturbance in terms of susceptibility and vulnerability (mortality) of balsam fir (Abies balsamea (L.) Mill.), white spruce (Picea glauca (Moench) Voss), and black spruce (Picea mariana (Mill.) BSP). All three host-tree species are highly susceptible for oviposition by the spruce budworm and all are suitable for completion of the budworm life cycle. Host-related differences in susceptibility arise from the degree of synchrony between spruce budworm phenology during the feeding stages and host-tree phenology. Spruce budworm density was highest on white spruce throughout the budworm's life cycle and over the course of the outbreak, but more rapid flushing and growth of current-year buds in white spruce reduced damage relative to that on balsam fir. Conversely, later flushing of current-year buds on black spruce led to a reduction in budworm density early in the season and a corresponding reduction in defoliation. The combination of high budworm densities and severe defoliation caused mortality first on balsam fir. By the end of the outbreak, 89% of the balsam fir component 〉10 cm DBH was eliminated compared with 49% of the white spruce in the same size class. The lower susceptibility of black spruce resulted in survival of all but the smallest size classes of that species. Nonhost species such as trembling aspen (Populus tremuloides Michx.) nearly doubled their basal area during the outbreak. The results link processes inherent in the insecthost relationship with the population ecology of the insect and the disturbance ecology of the forest.

Description: We examined the growth of white spruce planted underneath a 33-year-old stand of trembling aspen following thinning and fertilization. Thinning the stand to 2000 stems·ha1 resulted in increased light reaching spruce seedlings and significant increases in height and diameter growth of seedlings over that observed in either the unspaced or 3000 stems·ha1 treatments. Fertilization (applied 3 years prior to planting) stimulated development of understory vegetation and did not benefit the planted spruce. While growth was improved by thinning, the benefits of the thinning treatment to establishment of white spruce appeared to be small, and amounted to increasing seedling height from 62 cm (unthinned) to 73 cm (thinned to 1000 or 2000 stems·ha1) at the end of 5 years. Results indicated that when stands are thinned for the purpose of accelerating growth rates of underplanted spruce, residual basal areas above 25 m2·ha1 should be retained to suppress growth of understory vegetation. Comparison of height at age 5 for seedlings planted under untended stands at Fort Nelson with three sites near Dawson Creek indicated no differences between locations, when height at the time of planting was included as a covariate.

Description: Composition, structure, and diversity of vascular and nonvascular plant communities was compared 3 years after wildfire and clear-cutting in mesic trembling aspen (Populus tremuloides Michx.) forests of the southern Canadian boreal forest. We examined mean response to disturbance and variability around the mean across four to five spatial scales. Four 1997 wildfires were located near Timmins, Ontario, and ten 19961997 clearcuts were located adjacent to the wildfires. We randomly located plots within mesic, aspen-dominated stands selected to minimize predisturbance environmental differences. Correspondence analysis separated wildfire and clearcut samples based on community composition: wildfires had more aspen suckers, Diervilla lonicera Mill., and pioneering mosses; clearcuts had more under story tall shrubs, forbs, bryophytes, and lichens. Live tree basal area averaged 1.7 m2/ha in wildfires and 1.8 m2/ha in clearcuts (p = 0.59), and understory community structure (the horizontal and vertical distribution of live and dead plant biomass) was not markedly different. Clearcuts had higher species richness with greater variance than wildfires across all spatial scales tested, but differences in beta and structural diversity varied with spatial scale. Generally, clearcutwildfire differences were more evident and wildfire variability greater at larger analytical scales, suggesting that plant biodiversity monitoring should emphasize cumulative effects across landscapes and regions.

Description: This study assessed growth, condition, and mortality of residual trees one decade after harvest across three silvicultural treatments in thirty 85- to 125-year-old Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) stands in the Oregon Coast Range foothills. Group-selection cuts had 33% of the entire stand volume extracted as patches approximately 0.20.8 ha in size; two-story regeneration harvests had 75% of the volume extracted, and 2030 residual trees/ha were left; clearcuts had all trees removed, except for 1.2 trees/ha. One decade after harvest, tree basal area, diameter, and height growth, and crown width and fullness did not differ between silvicultural treatments. In contrast, live crown ratio was largest in clearcuts (0.74), and the proportion of trees with epicormic branching was highest in two-story stands (35%). Overall, 45% of trees had more basal area growth in the decade after harvest than in the previous decade. Residual green trees in clearcuts and group-selection stands experienced the highest and lowest percentage mortality, respectively (30.6% vs. 0.2%). Our results reflect little differentiation in the characteristics of trees growing under three silvicultural conditions one decade after harvest. However, percent residual green tree mortality increased with increasing harvest intensity.

Description: A structured expert-judgement elicitation technique was used to develop probability distributions for fireline production rates for Ontario's three- and four-person initial-attack crews for seven common fuel types and two distinct levels of fire intensity (i.e., low, 500 kW/m; moderate, 1500 kW/m). A total of 141 crew leaders provided 900 estimates of the minimum, maximum, and most likely (mode) time to construct 610 m (2000 ft) of fireline. This information was used to estimate parameters for beta probability distributions for each individual and scenario. Analysis of variance (ANOVA) of the beta-distribution parameters (α and β) and the three time estimates indicated that fuel type, intensity, crew size, and crew-leader experience all have a statistically significant (p 〈 0.05) influence on estimated crew productivity. The 28 scenario-specific and 7 aggregated distributions and expected values can be used in many operational fire-management activities (e.g., presuppression planning, initial-attack dispatching, initial-fire assessments) and incorporated into initial-attack containment models. These results also provide baseline data on crew productivity that can be used in larger strategic analyses to gauge the benefits of new fire-suppression equipment and techniques for the entire fire-management program.

Description: Individual tree mortality models for western white pine (Pinus monticola Dougl. ex D. Don), Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco), grand fir (Abies grandis (Dougl. ex D. Don) Lindl.), western redcedar (Thuja plicata Donn ex. D. Don), western hemlock (Tsuga heterophylla (Raf.) Sarg.), and western larch (Larix occidentalis Nutt.) were developed using data from permanent research plots located throughout the inland Northwest. The proposed linear logistic models included the following independent variables: diameter at breast height, basal area in larger trees, basal area per hectare, individual tree diameter increment, and a set of dummy variables for the six bedrock types, which were granite, basalt, metasedimentary, sedimentary, mixed glacial till, and deep deposit. The results show that rock type significantly affects individual tree mortality for western white pine, Douglas-fir, and western redcedar, while grand fir, western hemlock, and western larch were not affected. Western white pine and Douglas-fir growing on meta sedimentary rocks exhibited greater mortality probabilities than on other rocks. Mortality probabilities for western hemlock were low across all rock types, including "nutrient-poor" rocks like metasedimentary types.

Description: Plant growth-promoting rhizobacteria promote plant growth and induce biocontrol, but are affected by soil type, water stress, microbial competition, and environmental conditions. One unexplored factor is the interaction of rhizobacteria-inoculated plants exposed to ozone. Loblolly pine (Pinus taeda L.) seeds were inoculated with either Bacillus subtilis (Ehrenberg) Cohn or Paenibacillus macerans (Schardinger) Ash. In field trials, 4-week-old seedlings were exposed for 12 weeks to carbon-filtered (CF ≈ 12 ppb), 1× (≈46 ppb), or 2× (≈97 ppb) ozone for 12 h·d1 in open-top chambers (OTC) in 1998 and 1999. In three greenhouse trials, 5-week-old seedlings were exposed to ozone at 0× (≈8 ppb), 1× (≈105 ppb), 2× (≈199 ppb), and 3× (≈298 ppb) for 4 h·d1, 5 d·week1 for 8 weeks in continuously stirred tank reactors (CSTR). In both the CSTRs and the OTCs, ozone-exposed seedlings exhibited 20%50% less biomass and more foliar injury as compared with nonexposed seedlings. In CSTRs, at the 3× exposure, B. subtilis-inoculated seedlings had 12% less foliar injury than noninoculated seedlings. Foliar injury was 65% less for B. subtilis-treated seedlings in 1998, and root surface area, total root length, and root diameter was 25%35% greater when seedlings were exposed to 2× ozone in the OTCs. This is the first report of rhizobacteria protecting seedlings against the negative effects of ozone exposure.

Description: Forests in the New Jersey pine barrens are frequently prescribed burned to reduce fuel loads and risk of wildfire. To acquire baseline data for effects of prescribed burns on ectomycorrhizal diversity and nutrient uptake, field studies were undertaken in two upland pineoak forests in the New Jersey pine barrens subjected to different burn regimes. Ectomycorrhizal diversity was assessed by extraction of roots from soil cores and separation according to morphological characters. Nutrient availability to plant roots was measured using root bioassays. Relative to unburned plots, plots exposed to a fire at Greenwood, where burning was more frequent, had decreased total abundance of ectomycorrhizal tips, richness of ectomycorrhizal types, and Simpson's diversity in the L and F horizons, but increased Simpson's diversity in the deeper A horizon. At Lebanon, under a less frequent burn regime, richness was lower in whole cores and in the A horizon of burned versus unburned plots. Nitrogen, phosphorus, and potassium uptake by root bioassays indicated higher field availability of nutrients to roots in burned plots than roots in unburned plots, indicating a fertilization effect of the fire. Prescribed burning primarily impacted ectomycorrhizal community structure in the L and F horizons at these sites. Changes in function of ectomycorrhizae (nutrient uptake) in response to the burns was associated with decreased ectomycorrhizal diversity.

Description: We used a 72-year chronosequence to study the loss and recovery of ecosystem C pools following stand-replacing wildfire in Michigan, USA, jack pine (Pinus banksiana Lamb.) forests. We quantified the amount of C stored in aboveground plant biomass, standing dead timber, downed dead wood, surface organic soil, and mineral soil in 11 jack pine stands that had burned between 1 and 72 years previously. Total ecosystem C ranged from a low of 59 Mg C·ha1 in the 4-year-old stand to 110 Mg C·ha1 in the 72-year-old stand. Changes in total ecosystem C across the chronosequence conformed to theoretical predictions, in which C stocks declined initially as decomposition of dead wood and forest-floor C exceeded production by regenerating vegetation, and then increased asymptotically with the development of a new stand of jack pine. This pattern was well described by the following "gamma" function: total ecosystem C (Mg·ha1) = 112.2 39.6 × age0.351 × exp(0.053 × age01.039); mean-corrected R2 = 0.976. Using the first derivative of this parameterized gamma function, we estimated that jack pine stands function as a weak source of C to the atmosphere for only ca. 6 years following wildfire, and reach a maximum net ecosystem productivity of 1.6 Mg C·ha1·year1 by year 16. We attribute the rapid transition from carbon source to carbon sink in these ecosystems to two factors: (i) stand-replacing wildfires in these xeric forests leave behind little respirable substrate in surface organic horizons, and (ii) jack pine is able to rapidly reestablish following wildfires via serotinous cones. Jack pine stands remained net sinks for C across the chronosequence; however, net ecosystem productivity had declined to 0.12 C ha1·year1 by year 72. Carbon sequestration by mature jack pine ecosystems was driven primarily by continued growth of overstory jack pine, not by accumulation of detrital C.

Description: The frequency of mature hybrids, including post-F1 individuals, between loblolly (Pinus taeda L.) and shortleaf pine (Pinus echinata Mill.), detectable with a codominant nuclear marker, was studied in a sympatric population from central Arkansas. The direction of introgression was also examined. The marker revealed 10 putative hybrids from the 80 trees sampled. PCR-RFLP analysis of their rbcL gene showed two of the putative hybrids (HL) share loblolly pine chloroplast DNA, and eight (HS) share the shortleaf pine chloroplast DNA. The two putative HL hybrids were morphologically similar to loblolly pine, and the eight putative HS hybrids were morphologically similar to shortleaf pine. Utilizing microsatellite data, Nei's measure of genetic identity showed the putative HL hybrids to be similar to loblolly pine, and the putative HS hybrids as being similar to shortleaf pine. An inferred tree of the individuals, using simple sequence repeat data and the neighbor-joining method, also suggested that some of the putative hybrids were not F1 individuals. Principle component analysis of the morphological characters groups the HL trees with loblolly pine and the HS trees with shortleaf pine. These results suggest bidirectional introgression occurred within the study population, and some of the putative hybrids were likely derived from early-generation backcross(es) with either shortleaf or loblolly pine.

Description: The search for a stand density that maximizes total volume growth has continued since the beginning of forestry without producing a definite answer. One of the reasons is that the effect of density on growth is not always separated from those of tree size and age. Such a separation is not easy when the relationship between density and growth is expressed as a graph (Langsaeter's curve). This study develops a simple model that accounts for each main growth predictor individually. It allows one to calculate the density that maximizes volume growth at any given moment (current annual increment of volume). Just as the maximum Reineke's index, this density optimum does not change with age. Fitting the model to long-term data confirms the obvious: because a complete crown closure intercepts more light than a broken canopy, the densest stands produce maximum volume growth. Thus, the current optimal density is equal to maximum density. What is less obvious is that maximum current stand volume growth does not sum up to maximum stand volume. In addition to density, stand volume depends on average tree size, which is larger in less dense stands. The high current density that maximizes volume growth also minimizes diameter increment, which eventually reduces average diameter and stand volume. When density is kept at a stationary (over time) level by thinning, maximum volume is produced by stands with substantially lower density than the current optimum. even higher volume can probably be obtained when density changes with age. The challenge is to find this optimum path of density.

Description: We use two rate-process models to describe cell mortality at elevated temperatures as a means of understanding vascular cambium cell death during surface fires. In the models, cell death is caused by irreversible damage to cellular molecules that occurs at rates that increase exponentially with temperature. The models differ in whether cells show cumulative effects of heating. The temperature dependencies of the models' rate parameters were estimated from cell-count data after exposing live-bark tissues from four Canadian Rocky Mountain tree species to a range of fixed temperatures in a water bath. Based on both models, lodgepole pine's (Pinus contorta Dougl. ex Loud.) growing season vascular cambium cells experienced lower mortality rates at elevated temperatures than those of aspen (Populus tremuloides Michx.), Engelmann spruce (Picea engelmannii Parry ex Engelm.), and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco). Growing and dormant season differences were marginal. With reservations for lodgepole pine, both models predicted cell survival after exposures to rising and falling temperatures such as would be experienced by live tissues during fires. A simulation involving conduction heat transfer from flames and vascular cambium cell mortality suggests that differences among species in thermal tolerance are small compared with the effects of bark thickness. Although stem vascular cambium cell mortality was complete when tissues reached 60 °C during simulated surface fires, it may not be warranted to apply the 60 °C threshold to other tissues exposed to contrasting temperature regimes during fires.

Description: Fuel moisture content (FMC) estimation is a critical part of any fire danger rating system, since fuel water status is determinant in fire ignition and fire propagation. However, FMC alone does not provide a comprehensive assessment of fire danger, since other factors related to fire ignition (lightning, human factors) or propagation (wind, slope) also need to be taken into account. The problem in integrating all these factors is finding a common scale of danger rating that will make it possible to derive synthetic indices. This paper reviews the importance of FMC in fire ignition and fire propagation, as well as the most common methods of estimating FMC values. A simple method to convert FMC values to danger ratings is proposed, based on computing ignition potential from thresholds of moisture of extinction adapted to each fuel. The method has been tested for the Madrid region (central Spain), where a fire danger assessment system has been built. All the variables related to fire danger were integrated into a dedicated geographic information system and information provided to fire managers through a web mapping server.

Description: This paper presents data on early postfire tree regeneration. The data were obtained from repeated observations of recently burned forest stands along the Yukon British Columbia border and in interior Alaska. Postfire measurements of tree density were made periodically for 2030 years, providing direct observations of early establishment patterns in boreal forest. Recruitment rates of the dominant tree species in both study areas were highest in the first 5 years after fire, and additional net establishment was not observed after 10 years. The postfire population of spruce (Picea mariana (Mill.) BSP and Picea glauca (Moench) Voss s.l.) remained constant after the first decade in the two study areas. Populations of aspen (Populus tremuloides Michx.) and lodgepole pine (Pinus contorta Dougl. ex Loud. var. latifolia Engelm.) both declined after 10 years in mixed-species stands along the Yukon British Columbia border. Mortality rates of aspen and pine were positively correlated with their initial densities, indicating that thinning occurred as a density-dependent process. At all sites, measurements of stand density and composition made early were highly correlated with those made late in the monitoring period, indicating that patterns of stand structure initiated within a few years after fire are maintained through subsequent decades of stand development.

Description: The effects of partial and clear-cut harvesting on abundance and community composition of forest floor microarthropods and oribatid mites were investigated in conifer and deciduous stands of the mixedwood boreal forest. Soil samples from clearcuts, strip-cut corridors in a partial cut, green-tree retention patches in a partial cut, and uncut control sites were collected 2.5 years after harvest and microarthropods were extracted, enumerated, and identified. Densities of microarthropod suborders were lower in the strip-cut corridors of the deciduous stands and significantly lower in the suborder Oribatida (oribatid mites). Changes in microarthropod community composition, decreased relative abundance of prostigmatid mites, and increased relative abundance of mesostigmatid mites were observed in corridor and clear-cut treatments. Lowered abundances and changes in community composition are likely due to compaction of the forest floor during harvesting. Selected oribatid mite species showed significantly lower abundances in clearcuts than in uncut sites, but diversity indices for oribatid mites were generally not significantly different between uncut sites and clearcuts. Changes in oribatid mite communities following harvesting were thus more quantitative (absolute abundance) than qualitative (diversity, composition), and as a result, use of oribatid mites as biological indicators of disturbance is limited because of the lack of changes in species composition.

Description: This study compared productivity in mixed-species plantations of Eucalyptus globulus ssp. pseudoglobulus (Naudin ex Maiden) Kirkpatr. and Acacia mearnsii de Wild with pure stands of each species and investigated how this might be explained by canopy stratification between species and changes in leaf characteristics of eucalypts. Investigations were carried out at a trial using the replacement series design, which consisted of the following combinations: 100% eucalypts (100%E), 75% eucalypts + 25% acacia (75%E:25%A), 50% eucalypts + 50% acacia (50%E:50%A), 25% eucalypts + 75% acacia (25%E:75%A), and 100% acacia (100%A). At 9.5 years, stem volume and biomass were highest in 50%E:50%A treatments. Canopy stratification occurred in all mixtures, with acacias in the lower and eucalypts in the upper canopy stratum. This and the increasing canopy light interception with increasing proportion of acacia in the mixture indicated that A. mearnsii is substantially more shade tolerant than E. globulus. Midcanopy foliage of E. globulus in the 50%E:50%A mixture had higher foliage nitrogen (N) but lower phosphorus (P) concentrations and lower light-saturated net photosynthesis rates (Amax) than those in the 100%E treatment. In addition, similar relationships between eucalypt crown volume and stem biomass across treatments indicated that eucalypt crowns were not more efficient in mixture. Our study indicates that the productivity gains in these mixtures may be partially attributable to aboveground niche separation between species.

Description: To evaluate the potential use of tree-ring data as a proxy for fire activity at the scale of a large boreal region, we analyzed a set of regional tree-ring chronologies of Siberian larch (Larix sibirica L.), a spatially implicit annual fire record, and monthly climate data for the Komi Republic for the period 19501990. In most years, annually burned area was below 0.001% of the republic's forested area and reached up to 0.7% during fire-prone years. Principal components (PC) of summer aridity resolved 64.2% of the annual variation in the number of fires, 12.2% in the average fire size, and 59.2% in the annually burned area. In turn, tree-ring PCs explained 65.2% of variation in fire-related weather PCs. Dendrochronological reconstruction of the annual number of fires and of the log-transformed annually burned area predicted 27.0% and 40.1% of the high-frequency variance of these variables, respectively. Coefficient of efficiency, a measure of reconstruction usefulness, reached 0.081 (number of fires) and 0.315 (annual area burned), supporting the obtained index as a realistic proxy for regional fire activity. Decadal variation in coefficient of efficiency values suggested improved monitoring accuracy since 1960 and more effective fire suppression during the last studied decade (19801990).

Description: Variable plot sampling has been widely used for many years. It was recognized, early in its application, that the process of getting stand volume could be divided into two components, counting trees to get basal area per unit area and measuring trees to get volume/basal area ratios (VBARs). It was further recognized that these two components had different amounts of variation and therefore should be sampled at different intensities. The fact that basal area per unit area is almost always more variable than the VBARs of individual trees has led to the widespread practice of counting trees on all plots and subsampling trees for VBAR measurements, typically by measuring all the trees on every third or fourth plot. This article presents an alternative, the "big BAF method," which uses a larger basal-area-factor angle gauge to do a second sweep of each plot to select the trees to be measured for VBAR. This procedure spreads the tree measurements throughout the stand and is thus more statistically efficient. The method is simple to apply, requires no additional computations, and is easy to audit. Two case-study examples are used to demonstrate the method.

Description: A ground-based scanning lidar (light detection and ranging) system was evaluated to assess its potential utility for tree-level forest mensuration data extraction. Ground-based-lidar and field-mensuration data were collected for two forest plots: one located within a red pine (Pinus resinosa Ait.) plantation and another in a mixed deciduous stand dominated by sugar maple (Acer saccharum Marsh.). Five lidar point cloud scans were collected from different vantage points for each plot over a 6-h period on 5 July 2002 using an Optech Inc. ILRIS-3D laser imager. Field- validation data were collected manually over several days during the same time period. Parameters that were measured in the field or derived from manual field measures included (i) stem location, (ii) tree height, (iii) stem diameter at breast height (DBH), (iv) stem density, and (v) timber volume. These measures were then compared with those derived from the ILRIS-3D data (i.e., the lidar point cloud data). It was found that all parameters could be measured or derived from the data collected by the ground-based lidar system. There was a slight systematic underestimation of mean tree height resulting from canopy shadow effects and suboptimal scan sampling distribution. Timber volume estimates for both plots were within 7% of manually derived estimates. Tree height and DBH parameters have the potential for objective measurement or derivation with little manual intervention. However, locating and counting trees within the lidar point cloud, particularly in the multitiered deciduous plot, required the assistance of field-validation data and some subjective interpretation. Overall, ground-based lidar demonstrates promise for objective and consistent forest metric assessment, but work is needed to refine and develop automatic feature identification and data extraction techniques.

Description: To analyze the effect of landscape structure (viz. amount of wetlands) on the past forest fire regime in boreal Sweden, we reconstructed detailed fire histories by cross-dating fire scars in living and dead Scots pine (Pinus sylvestris L.) in two different landscape types: mire-free landscapes with a low proportion (1%2%) of mires and mire-rich landscapes with a high proportion (21%33%) of mires. Two localities were selected and at each one, adjacent mire-free and mire-rich areas of 256601 ha were sampled. Over the studied 650-year period, the two landscape types differed primarily in the fire intervals and sizes of fires. In the mire-rich landscapes, fires had frequently stopped against mire elements. The net effect was significantly longer fire intervals in the mire-rich than in the mire-free landscape (on average, 32 versus 56 years). The mire-rich areas also had a tail of very long fire intervals lacking in the mire-free areas (maximal interval 292 years). We conclude that mires can have a profound effect on both spatial and temporal patterns of forest fires in the boreal forest, but only when they are effective fuel breaks (i.e., they are wet enough) at the time the fires burn and if they truly dissect the nonmire portion of the forest landscape.

Description: The International Crown Fire Modelling Experiment (ICFME), carried out between 1995 and 2001 in Canada's Northwest Territories, involved 18 experimental high-intensity crown fires, with more than 100 participants representing 30 organizations from 14 countries. ICFME has provided valuable new data and insights into the nature and characteristics of crowning forest fires, which will assist in addressing fire management problems and opportunities affecting both people and ecosystems in future decades. ICFME evolved as the result of a number of converging issues: the recognition that the US and Canada could not continue separate approaches to fire behaviour model development, the opening of Russia to the western world, increased communication, and the formation of international associations to facilitate collaboration. While the initial impetus for ICFME was the desire to improve the physical modeling of crown fire propagation and spread, the project also created the opportunity to examine many other aspects and impacts of crown fires. This special issue of the Canadian Journal of Forest Research devoted to ICFME is intended to summarize most of the major research results from the project.

Description: Indicators of forest health used in previous studies have focused on crown variables analyzed individually at the tree level by summarizing over all species. This approach has the virtue of simplicity but does not account for the three-dimensional attributes of a tree crown, the multivariate nature of the crown variables, or variability among species. To alleviate these difficulties, we define composite crown indicators based on geometric principles to better quantify the entire tree crown. These include crown volume, crown surface area, and crown production efficiency. These indicators were then standardized to a mean of 0 and variance of 1 to enable direct comparison among species. Residualized indicators, which can also be standardized, were defined as the deviation from a regression model that adjusted for tree and plot conditions. Distributional properties were examined for the three composite crown indicators and their standardized-residualized counterparts for 6167 trees from 250 permanent plots distributed across Virginia, Georgia, and Alabama. Comparisons between the composite crown indicators and their associated standardized residual indicators revealed that only two or three plots were jointly classified as poor by both when thresholds were set at the lower 5 percentiles of statistical distributions. In contrast, 19-21 other plots were classified differently, emphasizing that different aspects of crown condition are being summarized when the raw values are adjusted and standardized. Generally, crown volume and crown surface area behaved similarly, while crown production efficiency was substantially different.

Description: Estimating the wind force or drag acting on tree crowns is central to understanding both the chronic effects of wind and the calculation of critical wind speed in windthrow prediction models. The classical drag equation is problematic for porous, flexible tree crowns whose frontal area declines as wind speeds increase and branches streamline. Juvenile crowns of three morphologically different conifers, western redcedar (Thuja plicata Donn ex D. Don), western hemlock (Tsuga heterophylla (Raf.) Sarg.), and lodgepole pine (Pinus contorta Dougl. ex Loud.), were exposed to wind speeds from 4 to 20 m/s in a wind tunnel. At 20 m/s, streamlining reduced the frontal area by 54% for redcedar, 39% for hemlock, and 36% for lodgepole pine. Crown drag coefficients calculated using frontal area in still air varied with wind speed. At 20 m/s, they were 0.22, 0.47, and 0.47 for these species, respectively. Drag was proportional to the product of mass and wind speed and also to the product of wind speed squared and wind speed specific frontal area. Lodgepole pine and redcedar had lower drag per unit of branch mass than did hemlock. Removing branches by pruning had little effect on drag per unit branch mass.

Description: We propose a simple method that uses aerial photographs to characterize the impacts of a spruce budworm (Choristoneura fumiferana (Clem.)) outbreak on canopy structure. Using aerial photographs taken at the beginning (1972) and after (1994) a major spruce budworm outbreak (19701987), we evaluated the change in canopy openness that occurred during the period of the outbreak, in five compositionally different stands. We compared canopy openness evaluated by photointerpretation with two independent field techniques and found a high degree of similarity between methods. Interpretation of the 1972 photographs (prior to the outbreak) shows that regardless of composition, four of our five analysed stands had about the same degree of mean canopy openness (17%20%). Following the outbreak, openness increased in all stands except for the hardwood-dominated stand. The highest increase in openness (from 18% to 45%) occurred in the stand with the highest conifer content. Thematic maps and spatial analysis techniques were used to describe canopy openness distribution. Openness was low and uniformly distributed before the outbreak, whereas after the outbreak, the various degrees of openness had a patchy distribution in most stands. Furthermore, patch size increased with conifer content. Using the amount of increase in canopy openness and its specific distribution within stands, we propose guidelines for the development of silvicultural practices that mimic spruce budworm disturbances in boreal mixedwoods.

Description: A height–diameter (HD) model for Norway spruce (Picea abies (L.) Karst.) was estimated from longitudinal data. The Korf growth curve was used as the HD curve. Firstly, HD curves for each stand at each measurement time were fitted, and the trends in the parameters of the HD curve were modeled. Secondly, the trends were included in the HD model to estimate the whole model at once. To take the hierarchy of the data into account, a mixed-model approach was used. This makes it possible to calibrate the model for a new stand at a given point in time using sample tree height(s). The heights may be from different points in time and need not be from the point in time being predicted. The trends in the parameters of the HD curve were not estimated as a function of stand age but as a function of the median diameter of basal area weighted diameter distribution (dGm). This approach was chosen because the stand ages may differ substantially among stands with similar current growth patterns. This is true especially with shade-tolerant tree species, which can regenerate and survive for several years beneath the dominant canopy layer and start rapid growth later. The growth patterns in stands with a given dGm, on the other hand, seem not to vary much. This finding indicates that the growth pattern of a stand does not depend on stand age but on mean tree size in the stand.

Description: A wildfire area in a boreal forest landscape dominated by Pinus sylvestris L., in the Vienansalo wilderness area in eastern Fennoscandia, was examined for its spatial characteristics and fire history. The boundaries of the 360-ha fire that occurred in 1969 were mapped, and the vegetation types of burnt and unburnt areas were inventoried. Fire history was investigated in 40 study plots, and fire scars, tree ages, and charcoal in peat or soil were used for evidence of past fires. The complex shape of the 1969 fire and the detected small-scale variation in past fire frequencies were concordant with the existing small-scale variation in site moisture and vegetation characteristics in the area. Moist depressions, swamps, and more fertile forest patches dominated by Picea abies (L.) Karst. often did not burn when the nearby dryish forest type did. There was also temporal variability in fire frequency. An abrupt increase in the number of fires occurred in the late 17th century. In the mid-19th century, both the number of fires and the annually burnt area in the region decreased. Our results show that in the examined wildfire area, there has been considerable and consistent small-scale spatial variation in fire frequency and that historical fire regime evidently has been strongly affected by human activity.

Description: In the southeastern boreal forest of Canada, the presence of mixed stands of black spruce (Picea mariana (Mill.) BSP) and trembling aspen (Populus tremuloides Michx.) growing in similar abiotic conditions offers the opportunity to study the influence of aspen on stand volume and spruce growth. A regression analysis performed on field data from the ministère des Ressources naturelles du Québec showed a significant relationship between the relative basal area of aspen (aspen relative basal area was determined by the ratio of aspen basal area to total basal area of the stand) and the total stand merchantable volume after accounting for stand density. However, the relationship between total black spruce volume and relative basal area of aspen was not significant, implying that the volume gain was, in fact, aspen fibre. The positive effects of aspen on black spruce DBH and height were only present when the proportion of aspen in the stand ranged between 0% and 41% of the total stand basal area. These results suggest that aspen uses a different niche than black spruce. Furthermore, the significant increase in black spruce dominant height along the aspen gradient suggests that aspen enhances soil fertility by its influence on nutrient availability. The management of mixed stands, which make up an important proportion on the landscape, offers an example as to how commercial management of the forest can be in agreement with ecosystem management.

Description: Fast-growing hybrid poplar (Populus spp.) have potential as a short-rotation coppice crop grown for biomass energy. This work identifies traits for fast growth studied in an American interspecific pedigree derived from Populus trichocarpa Torr. & A. Gray × Populus deltoides Marsh. grown in the United Kingdom for the first time. The biomass yield after the first coppice rotation was estimated to range from 0.04 to 23.68 oven-dried t·ha1·year1. This great range suggests that genotypes from this pedigree may be used to understand the genetic basis of high yield in short-rotation coppice, which would be advantageous for informing breeding programs for biomass crops. Relationships between stem, leaf, cell traits, and biomass yield were investigated. Partial least-squares analysis was used to order the traits by importance. The traits most influential on biomass were maximum stem height throughout the growing season, basal diameter, number of stems, and number of sylleptic branches, which showed high heritability, indicating excellent potential for breeding programs. The leaf traits, leaf area, number of leaves on the leading stem, and plastochron index were also associated with an increase in biomass, leading to a better understanding of this trait.

Description: The importance of large woody debris (LWD) in forested stream ecosystems is well documented. However, little is known about LWD in northern boreal forest streams. We investigated the abundance, characteristics, and function of LWD in 13 small tributary streams of the upper Yukon River basin, Yukon Territory, Canada. LWD abundance was similar to values reported from temperate regions, whereas LWD size and total volume were well below values for the Pacific Northwest. LWD formed 28% of the pools, which provide important habitat for juvenile chinook salmon (Oncorhynchus tshawytscha Walbaum). The median diameter of pool-forming pieces was 17 cm, and ring counts on fallen riparian trees indicated that pool-forming pieces were likely 70200 years old when downed. Juvenile chinook salmon density was correlated with LWD abundance in our study reaches. We conclude that despite differences in climate and forest type, LWD in Yukon streams and LWD in temperate regions appear to perform a similar function in creating fish habitat. Resource managers should consider the relatively slow tree growth and thus potentially long recovery times following human disturbances in these watersheds.

Description: This study examines the ability of high-density laser scanning to produce single-tree estimates in mixed stands of heterogeneous structure. Individual trees were detected from a constructed digital canopy height model by locating local maxima of the height values. The reference material comprised accurately measured field data for 10 mapped sample plots containing Norway spruce (Picea abies (L.) Karst.), Scots pine (Pinus sylvestris L.), and different birches. To verify the accuracy of height measurements of single trees in more detail, the height of 29 Scots pine trees and their annual shoots of the last few years was carefully measured with a tacheometer and a glass fibre rod. The considered variables were the proportion of detected trees and tree height. As more than 80% of the dominant trees were detected, the results indicated that laser scanning can accurately describe the trees of the dominant tree layer. Because of the dense understorey tree layer in most of the sample plots, about 40% of all trees were detected. On the plot level, the stand structure affected the accuracy of the results considerably. The scanning-based tree height was most accurate for Norway spruce and least accurate for birches. The height of the separately measured 29 Scots pine trees was obtained with an accuracy of ±50 cm or better.

Description: The renewed interest in the use of fast-growing tree species is accompanied by concerns about the adverse effects that these trees may have on soil. Four Norway spruce (Picea abies (L.) Karst.) provenance trials in Quebec were used to test the hypothesis that a more vigorous growth would not occur at the expense of marginalizing available nutrient pools. On these sites, the provenance showing the greatest overall productivity (high treatment) and the one showing the lowest productivity (low treatment) were studied. The divergence in total aboveground nutrient contents between the high and low treatments was high in all sites (i.e., 161%209%). Increased nutrient immobilization in trees did not cause any significant soil depletion of available base cations or total N at any site. Moreover, exchangeable Ca concentrations, cation-exchange capacity, and exchangeable Ca pools in the forest floor were significantly higher in the high treatment. It is concluded that in the short term, increased nutrient immobilization in trees does not create an apparent depletion of available base cations, perhaps because of a stimulation of soil mineral weathering and (or) a better retention of nutrients by the trees. Also, an effort to simulate mineral weathering using PROFILE showed the need for model improvement for applications at the plot level.

Description: American chestnut (Castanea dentata (Marsh.) Borkh.) was once a dominant tree in eastern deciduous forests of North America and is now endangered in Canada, primarily because of the introduction of a fungal pathogen (Cryphonectria parasitica (Murrill) Barr) causing chestnut blight. A recovery plan is being developed, and more accurate information on the status of this species in its native range in southern Ontario is needed to assist in the restoration and management of this species. We conducted a 2-year inventory of the distribution and size of trees throughout southern Ontario, and characterized the habitat and incidence of blight. In total, 682 trees were sampled, 601 of which were not planted. Individuals were statistically most likely to occur in deciduous forest habitats with high canopy cover (〉50%), gentle slopes (0°10°), and acidic (pH 46), sandy (〉75%) soils. Most trees were small (80% were

Description: This study presents spatially and temporally resolved measurements of air temperatures and radiant energy fluxes in a boreal forest crown fire. Measurements were collected 3.1, 6.2, 9.2, 12.3, and 13.8 m above the ground surface. Peak air temperatures exceeded 1330 °C, and maximum radiant energy fluxes occurred in the upper third of the forest stand and reached 290 kW·m2. Average radiant flux from the flames across all experiments was found to be approximately 200 kW·m2. Measured temperatures showed some variation with vertical height in the canopy. Equivalent radiometric temperatures calculated from radiant heat flux measurements exceeded thermocouple-based temperatures for all but the 10-m height, indicating that fire intensity estimates based on thermocouple measurements alone may result in underestimation of actual radiant intensity. The data indicate that the radiative energy penetration distance is significantly longer in the forest canopy than in the lower levels of the forest stand.

Description: We compared the belowground biomass (BGB)/aboveground biomass (AGB) ratio and the vertical root distribution of 40-year-old Pinus radiata D. Don fertilized with 0 or 90 kg P·ha1 at planting. Root biomass was determined by a combination of coring (fine roots, ϕ 〈 2 mm; small roots, 2 ≤ ϕ 〈 15 mm) and excavation (coarse roots, ϕ ≥ 5 mm). Stand-level AGB and coarse root biomass (CRB) were estimated with the use of allometric relations. After 40 years, AGB and CRB of P-fertilized trees were 4.5 times those of unfertilized trees, indicating that CRB scaled isometrically with AGB independently of P supply. By contrast, P fertilization increased the fine and small root biomass (FSRB) pool by only 50%. As a result, the scaling of FSRB to AGB was dependent on P supply. The differential response of the FSRB to P fertilization caused the overall BGB/AGB ratio to decrease from 0.29 in control plots to 0.20 in P-fertilized plots. Phosphorus fertilization also altered the vertical distribution of fine root biomass (FRB). For example, the proportion of FRB in the top 15 cm increased from 41% to 52% with P fertilization. Collectively, the results showed that P added early in the growth phase had a persistent effect on the BGB/AGB ratio in P. radiata. This was primarily brought about by altered biomass partitioning to the nutrient-acquiring FSRB pool.

Description: Persistence of the root rot pathogen Heterobasidion annosum (Fr.) Bref. s.s. on infested areas and its transfer to a forest regeneration was studied in three forest sites in eastern Lithuania. The sites represented H. annosum disease centres in Pinus sylvestris L. stands, which were clear-felled and replanted with Betula pendula Roth 25 years prior to our study. Fungal isolation from trees and stumps on each site was performed on both replanted B. pendula and surrounding P. sylvestris from the previous generation. Low productivity of B. pendula stands (45.076.1 m3·ha1), high mortality rates, and comparatively low vigor of trees (measured as crown densities) indicated a strong impact of root rot. Based on somatic incompatibility tests, we detected large spreading areas of clones of H. annosum (up to 48 m across) and old (35- to 40-year-old) clonal individuals. Territorial clones covered areas that encompassed both previous stands of P. sylvestris and current stands of B. pendula. Our results showed that H. annosum is able to persist in root systems of diseased trees for decades and readily attack birch replanted on infested sites. In addition, a total of 83 fungal species (out of 398 isolates) was found as a result of sampling 508 B. pendula, 49 P. sylvestris, 21 Juniperus communis L., and 1 Salix cinerea L. trees.

Description: Model validation is an important part of model development. It is performed to increase the credibility and gain sufficient confidence about a model. This paper evaluated the usefulness of 10 statistical tests, five parametric and five nonparametric, in validating forest biometric models. The five parametric tests are the paired t test, the Χ2 test, the separate t test, the simultaneous F test, and the novel test. The five nonparametric tests are the Brown-Mood test, the KolmogorovSmirnov test, the modified KolmogorovSmirnov test, the sign test, and the Wilcoxon signed-rank test. Nine benchmark data sets were selected to evaluate the behavior of these tests in model validation; three were collected from Alberta and six were published elsewhere. It was shown that the usefulness of statistical tests in model validation is very limited. None of the tests seems to be generic enough to work well across a wide range of models and data. Each model passed one or more tests, but not all of them. Because of this, caution should be exercised when choosing a statistical test or several tests together to try to validate a model. It is important to reduce and remove any potential personal bias in selecting a favorite test, which can influence the outcome of the results.

Description: This study compares the effects of full-tree versus cut-to-length forest harvesting methods on tree regeneration in jack pine (Pinus banksiana Lamb.), mixedwood (Picea glauca (Moench) Voss Populus tremuloides Michx. Abies balsamea (L.) Mill.), and black spruce (Picea mariana (Mill.) BSP) sites in southeastern Manitoba, Canada. We surveyed tree regeneration densities, disturbance characteristics, and understorey vegetation in replicated control and harvested plots in each site type preharvest (1993) and 1 and 3 years postharvest (1994, 1996). In jack pine sites, the full-tree harvest method promoted regeneration of Pinus banksiana through increased disturbance of soil and the moss layer, and decreased slash deposition relative to the cut-to-length method. Conversely, in mixedwood sites the cut-to-length method resulted in less damage to advance regeneration and proved better at promoting postharvest regeneration of Abies balsamea and Picea glauca relative to the full-tree method. In black spruce sites, there were few differences in the impact of the two harvesting methods on regeneration of Picea mariana, which increased in frequency and density after both types of harvesting.

Description: Several multi-criteria decision support methods have been introduced to sustainable management of natural resources, but different methods suit different planning situations. One way to support decision-making is to apply voting theory. In this study, a multi-criteria decision-support method based on voting theory, called multicriteria approval (MA), is applied to wood supply chain management in a forest area owned by the state of Finland. The area is called Leikko and is located in the rural municipality of Pieksämäki. MA seems to have some promising features in relation to participatory decision support. The most essential advantages are its ease and comprehensibility. MA is also able to deal with ordinal and imprecise information. Since the method does not demand much preference information from interest groups, the inquiries may be conducted using the Internet. In the case study, nine timber-harvesting alternatives were devised for the forest area. The study involved seven interest groups, whose representatives defined seven criteria by which the alternatives were compared. The purpose was to find a consensus or compromise solution for a practical harvesting schedule. Two different versions of MA were tested and compared from the participatory decision-support aspect. Usability and ease of method, the comprehensibility of the inquiries, and the congruence of the results were examined.

Description: The effects of six fertilization treatments on tree height and incidence of attack by the white pine weevil, Pissodes strobi (Peck), on interior spruce (a hybrid, Picea glauca (Moench) Voss × Picea engelmannii Parry ex Engelm.) were explored in a field study. In a corresponding laboratory study, changes in constitutive and traumatic resin canal defences in response to fertilization were measured. Incidence of weevil attack increased with fertilization intensity. This trend was explained by increased resources available for weevil feeding (adults and larvae) as a result of increased leader size and bark thickness, as well as by an observed weakening in the tree's constitutive resin canal defences. The ability of interior spruce to produce a traumatic resin response was not influenced by fertilization. Although incidence of weevil attack was greatest in trees from the most intense fertilization treatments, height losses due to weevil attack were not as great as height gains due to fertilization treatment. Therefore, we concluded that fertilization is a feasible option for increasing productivity of interior spruce plantations, particularly if other weevil control alternatives are implemented.

Description: Wildlandurban fire destruction depends on homes igniting and thus requires an examination of the ignition requirements. A physicaltheoretical model, based on severe case conditions and ideal heat transfer characteristics, estimated wood wall ignition occurrence from flame radiation heating and piloted ignition requirements. Crown fire experiments provided an opportunity for assessing model reliability. The crown fire experiments were specifically instrumented with wood wall sections and heat flux sensors to investigate direct flame heating leading to home ignition during wildland fires. The experimental results indicated that the flame radiation model overestimated the structure-to-flame distance that would result in wood wall ignition. Wall sections that ignited during the experimental crown fires did not sustain flaming after crown fire burnout. The experiments also revealed that the forest canopy attenuated the flame radiation as the crown fire spread within the forest plot. Ignition modeling and the associated crown fire experiments described the flame-to-structure distance scale associated with flame heating related to wall ignition.

Description: A new one-dimensional heat conduction model for predicting stem heating during fires is presented. The model makes use of moisture- and temperature-dependent thermal properties for layers of bark and wood. The thermal aspects of the processes of bark swelling, desiccation, and devolatilization are treated in an approximate fashion. An energy balance reveals that simulation with a heat flux input boundary condition requires that these phenomena be accounted for. Previous models have used temperaturetime boundary conditions, which prevents them from being used in conjunction with fire behavior models. This model uses a fluxtime profile for its boundary condition, making it possible to eventually couple it to fire behavior models. The model was developed and validated with laboratory experiments on Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) samples. It is intended that this model be used in conjunction with fire behavior and cell mortality models to make predictions of stem heating related mortality before prescribed burns.

Description: Sudden dieback and deterioration of mature aspen stands is commonly observed throughout North America. This dieback process has tremendous ecological and economic importance, yet remains poorly understood. This paper summarizes our understanding of aspen dieback in North America, identifies potential processes that contribute to reduced vigour and dieback of aspen stands, and examines the scales (stand, ecosite, regional) at which these processes operate. Many factors including pathogens, nutrition, or successional changes may be involved in the decline of aspen vigour and thereby contribute to the dieback process. However, insect defoliation, drought, and thawfreeze events appear to be the most likely factors initiating dieback in mature aspen stands. Further study is clearly needed to elucidate the mechanisms and landscape patterns of dieback. Information needs related to identifying processes and modeling landscape patterns of dieback are indicated.

Description: We used dendroecological techniques to analyze the temporal pattern in diameter growth following selection harvests in stands dominated by Acer saccharum Marsh. in central Ontario and examined differences in growth responses related to tree size, damage, and orientation relative to canopy gaps. While dendroecological studies have commonly assumed that trees show immediate growth responses to gap creation (i.e., within 12 years), we found that the growth enhancement in A. saccharum was gradual and did not reach a peak until 35 years following gap creation. Trees of intermediate size showed the largest proportional growth increases after gap creation, with the largest responses observed in trees on the north side of gaps. Trees with visible damage to the crown or bole had significantly lower preharvest basal area increments than trees with little or no damage, but showed greater proportional growth responses to gap creation. Both the long observed time delay in tree growth response to canopy opening and the variability in response relative to tree size and damage have important implications for attempts to reconstruct disturbance history using dendroecological methods and to sustainable forest management under selection system silviculture.

Description: Turnover rates of needle and branch biomass, number of needle cohorts, and needle-shed dynamics were modelled for Norway spruce (Picea abies (L.) Karst.) in southern Finland. Biomass turnover rates, vertical distribution, and biomass of the branches were modelled simultaneously. The rate of needle turnover was determined from needle-shed dynamics. The potential litterfall of branches was modelled by combining the vertical distribution of branch biomass and the annual change in height of the crown base. The mean annual turnover rates for needle and branch biomass are 0.10 and 0.0125, respectively. At the age of 5.5 years, 50% of the needles in the needle cohort have been shed. In addition, at the age of 12 years, all needles of the needle cohort have been shed. Turnover of branch biomass was dependent on stand density and tree size. The modelled rates of biomass turnover agreed with measurements of needle and branch litterfall. Many process- or inventory-based models use a single turnover rate for branch litterfall based on literature, and some of the models are fully ignoring the litterfall of branches. Species-specific turnover rates or dynamic litterfall models should be applied when carbon flows in forest stands are modelled.

Description: Ecophysiological and dendroecological data from a temperate sessile oak (Quercus petraea (Matt.) Liebl.) stand in Belgium were used to develop and parameterize a dendroecological process-based model. The purpose of this model is to serve as a tool for exploring the relationship between climate variability and tree growth based on dendro ecological data. When parameterized, the model was able to correctly simulate measurements of bud-burst date, through fall (r2 = 0.95), soil water content (r2 = 0.81), transpiration (r2 = 0.80), and ring-width series from 1960 to 1999 (r2 = 0.46). Model sensitivity analysis showed that atmospheric vapor pressure deficit is the major controlling factor of transpiration in this type of ecosystem. The model shows that bole increment is principally controlled by temperature because it affects the phenological process of bud burst and thus the growing season length. Precipitation variability does not affect variation of transpiration rate and bole increment because calculated soil water stress is negligible during the simulation period. Discrepancies between observed and simulated bole increment may be a consequence of stand density variations and worm defoliation in the spring. The MAIDEN model is particularly suited for dendreocological analysis because it takes simple species, site condition, and climatic variables as input.

Description: A short-term retrospective test trial was carried out using 90 open-pollinated families representing 30 provenances of black spruce (Picea mariana (Mill.) BSP) from Quebec. Seedlings were transplanted on three sites along a latitudinal gradient, and eleven growth and phenological traits were measured during the second and the third growing seasons. Analyses of variance indicated for most of the traits significant differences among provenances and families-within-provenances. Principal component analysis was used to summarize the variation observed among provenances into two principal components, which accounted for 79% of the total variation for all traits. Regression models developed to relate each trait and the principal component scores to geoclimatic variables explained between 55% and 86% of the variation observed among provenances. Variation in growth traits and phenological traits appeared to be related to geoclimatic factors. The models were validated using data from a range-wide provenance test, and relative risks associated with seed source transfer were estimated. The R2 values between the transfer risk and the provenance heights ranged from 0.02 to 0.58, whereas they were slightly lower for diameters. On average, the relative risks varied from 36% to 67%. Individual provenance values ranged from 4% to 94%. A geographic information system tool was designed to assist the forest managers in making seed transfer decisions.

Description: During the early establishment phase, outplanted white spruce (Picea glauca (Moench) Voss) and jack pine (Pinus banksiana Lamb.) seedlings are vulnerable to lethargic growth or mortality because of interspecific competition for soil nutrients, particularly N. Accurately quantifying the degree of N competition is essential for supporting effective vegetation management (VM) decisions. This study evaluated the use of in situ burials of ion-exchange membrane (IEM; Plant Root SimulatorTM-probes) for quantifying differences in soil N supply rate between different VM treatments and the relationship of this N availability index to early growth of conifer seedlings at four boreal forest sites. At most sites, the effect of noncrop N uptake on soil N availability was apparent, with smaller NH4+-N, NO3-N, and total dissolved inorganic N (DIN) supply rates in control plots than in VM plots. Total DIN supply rate was correlated (R2 = 0.60 to 0.73, P 〈 0.01) with seedling height, root-collar diameter, and stem volume growth. Ammonium-N supply rate was better correlated than NO3-N supply rate with conifer seedling growth, which is in agreement with preferential NH4+-N uptake by conifer species. The results of this study support the use of in situ burials of IEM for measuring soil N availability during the early establishment phase.

Description: Microsatellites or simple sequence repeats (SSRs) are highly polymorphic, codominant markers that have great value for the construction of genetic maps, comparative mapping, population genetic surveys, and paternity analyses. Here, we report the development and testing of a set of SSR markers derived from shotgun sequencing from Populus trichocarpa Torr. & A. Gray, a nonenriched genomic DNA library, and bacterial artificial chromosomes. Approximately 23% of the 1536 genomic clones and 48% of the 768 bacterial artificial chromosome subclones contained an SSR. Of the sequences containing an SSR, 72.4% contained a dinucleotide, 19.5% a trinucleotide, and 8.1% a tetranucleotide repeat unit; 26.6% of the sequences contained multiple SSR motifs in a complex or compound repeat structures. A survey of the genome sequence database revealed very similar proportional distribution, indicating that our limited rapid, shallow sequencing effort is representative of genome-wide patterns. In total, 492 primer pairs were designed and these yielded 77 markers that were mapped in an F2 pedigree, including 26 that were sufficiently informative to be included in a Populus framework map. SSRs with GC-rich motifs mapped at a significantly higher frequency than expected, although AT-rich SSRs accounted for the majority of mapped markers due to their higher representation in the genome. SSR markers developed from P. trichocarpa showed high utility throughout the genus, with amplification rates in excess of 70% for all Populus species tested. Finally, at least 30% of the markers amplified in several willow species, suggesting that some of these SSRs will be transferable across genera.

Description: Two nonspatial approaches for modeling tree crown recession (ΔHCB) were evaluated by using 5341 observations from Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco). The first approach applies a static height-to-crown-base (HCB) equation at the start and end of the growth period and uses the difference in these predictions as an estimate of ΔHCB. This allometric method can be applied to species lacking ΔHCB data from permanent plots. The incremental method directly predicts ΔHCB from an equation developed from adequate permanent plot data. Two allometric and six incremental equation forms were examined. Also examined were three approaches for determining the end-of-growth-period tree and plot attributes used by the allometric method. Although the allometric method can produce unbiased estimates of ΔHCB, the best allometric equation forms explained about one-half of the variation explained by the best incremental equation form. The two best incremental equation forms were modifications of a nonlinear logistic equation form previously developed for Douglas-fir. The modifications included using measured stand age (BHA) or predicted tree growth effective age (GEA) instead of measured tree age. The best equation form used BHA, which limits its application to modeling data collected from just even-aged stands. The equation form using GEA could be applied to modeling data sets from both even- and uneven-aged stands.

Description: The use of near-infrared (NIR) spectroscopy to discriminate between uninfested seeds of Picea abies (L.) Karst and seeds infested with Plemeliella abietina Seitn (Hymenoptera, Torymidae) larva is sensitive to seed origin and year of collection. Five seed lots collected during different years from Sweden, Finland, and Belarus were used in this study. Initially, seeds were classified as infested or uninfested with X-radiography, and then, NIR spectra from single seeds were collected with a NIR spectrometer from 1100 to 2498 nm with a resolution of 2 nm. Discriminant models were derived by partial least squares regression using raw and orthogonal signal corrected spectra (OSC). The resulting OSC model developed on a pooled data set was more robust than the raw model and resulted in 100% classification accuracy. Once irrelevant spectral variations were removed by using OSC pretreatment, single-lot calibration models resulted in similar classification rates for the new samples irrespective of origin and year of collection. Dis criminant analyses performed with selected NIR absorption bands also gave nearly 100% classification rate for new samples. The origin of spectral differences between infested and uninfested seeds was attributed to storage lipids and proteins that were completely depleted in the former by the feeding larva.

Description: Trends and periodicities in summer drought severity are investigated on a network of Canadian Drought Code (CDC) monthly average indices extending from central Quebec to western Manitoba and covering the instrumental period 19131998. The relationship and coherency between CDC indices and oceanatmosphere circulation patterns are also examined. Trend analyses indicate that drought severity is unchanged in eastern and central Canada. Composite analyses indicate that for most of the corridor, severe drought seasons occur with a combination of positive 500-hPa geopotential height anomalies centered over the Gulf of Alaska and over the Baffin Bay. Additional severe drought seasons develop across the corridor in the presence of positive height anomalies located over or upstream of the affected regions. According to spectral analyses, the North Atlantic and the North Pacific circulation patterns modulate the drought variability at the decadal scale. Our results lead us to conclude that climate warming and the increases in the amount and frequency of precipitation in eastern Canada during the last century had no significant impact on summer drought severity. It is unlikely that linear climate change contributed to the change in the boreal forest dynamics observed over the past 150 years.

Description: This study extends the economic literature on forest stand management by applying a process-based, rather than empirical, stand growth model. The economics of timber production is investigated using a distance-independent, individual tree process model specified for pure Scots pine (Pinus sylvestris L.) stands. Stem taper and crown morphology information are used for bucking the harvested trees into several roundwood categories according to quality and dimension requirements applied in the Finnish timber markets. Explicit inclusion of causality and timber quality in stand-level economic optimization generates a set of new results. Economic optimization decreases biomass production but increases roundwood production, compared with undisturbed stands. Optimal rotation length is insensitive to changes in the rate of interest beyond 4% owing to nonmonotonic value growth. Better quality attributes and higher productivity in resource use are partial reasons for favoring lower canopy trees in optimal thinnings. The first thinnings are light, irrespective of the rate of interest, because of their favorable feedback effects on the quality of residual trees. Production of the highest-grade roundwood is rational only at rates of interest lower than those prevailing in the capital markets. An example of two optima representing distinct timber management strategies is shown.

Description: Cryopreservation of a core collection of 444 elm (Ulmus spp.) clones in liquid nitrogen was carried out by two laboratories participating in a European project of conservation of elm genetic resources. Plant material, collected in nine European countries, represented a large sample of the genetic diversity within three European elm species and their hybrids. The cryopreservation technique used in both laboratories involved the stepwise freezing of cryotubes containing dormant buds. Comparisons with fresh buds showed that the cryopreservation treatment had no negative effect on the viability and regrowth potential of frozen buds. Tests on a random sample of 26 clones showed that direct regrowth of cryopreserved buds (i.e., through propagation by microcuttings) of Ulmus minor and Ulmus laevis was possible; conversely, Ulmus glabra could only be regrown through micrografting. Most thawed explants from all 26 clones survived through the whole cultivation phase and were successfully transferred to the field. A calculation of costs indicates that cryopreservation of elm buds is economically competitive to field clonal archives.

Description: Eucalyptus nitens (Deane & Maiden) Maiden (shining gum) is widely grown for kraft pulp production in many cool temperate regions of the world. Improving the kraft pulp yield of this species is important for increasing plantation profitability, but traditional assessment is slow and expensive. Cellulose content, which is strongly correlated with pulp yield, has been used as an alternative in tree breeding programs. However, a direct measure of cellulose content still relies on wet chemistry, limiting the number of samples that can be processed and the subsequent gains that can be made in a tree breeding program. An indirect method such as near infrared (NIR) spectroscopy provides a large increase in the numbers of samples that can be analysed. In this study, the genetic gains in cellulose content of E. nitens were compared using cellulose content, determined using wet chemistry and predicted by NIR calibrations based on different sampling intensities. Genetic gains based on NIR-predicted cellulose content were high, and a large proportion of the gain was achievable using a direct measure of cellulose. Calibrations were robust and generally could be reliably used across sites. NIR-predicted cellulose is highly heritable, with heritabilities comparable to or better than direct measures of cellulose.

Description: A greenhouse experiment was conducted to investigate the effects of P and S fertilization on the growth and nutrition of two aspens, a clonal aspen (Populus tremuloides Michx.) and a hybrid aspen (a mixture of Populus tremu loides × P. tremula L. and Populus tremula × P. tremuloides), grown in a Gray Luvisol fertilized with N. Phosphorus application significantly increased basal diameter (BD) increment and most other growth parameters for both clonal and hybrid aspens, whereas S application increased BD increment only for the hybrid aspen. Diameter increment, leaf area, and seedling biomass were generally increased by P application, regardless of S levels, but they responded to S application only if P was also applied. Phosphorus application significantly increased foliar P concentrations and foliar N and P contents and their uptake for the clonal aspen. For the hybrid aspen, P application increased foliar N, P, and S concentrations, contents, and uptake. Sulfur application significantly increased foliar S concentration, content, and uptake for the hybrid aspen but not foliar N concentration, content, and uptake for the clonal aspen. Except for height increment, growth parameters were significantly related to foliar percent P and N/P ratios and percent S and N/S ratios. It can be concluded that both P and S were deficient, but P was more limiting after potential N limitation was removed.

Description: Beechnuts from two seedlots were pretreated, without medium, at a controlled moisture content (MC) of 30% before or after 1-year storage at 7 °C. Seeds treated with the two methods were germinated on substrates at decreasing osmotic potential down to 1.2 MPa. A moderate stress of 0.2 MPa caused a slight but significant decrease in germination percentage. Each further increment in water stress produced additional significant decreases in germination capacity. At 1.2 MPa, germination was almost prevented. Seeds pretreated before storage showed lower germination percentage and speed at all osmotic potentials, but this result was due to a marked effect of seed initial MC, rather than a lower resistance to water deficit. In fact, MC of beechnuts pretreated before storage was 8%, whereas beechnuts pretreated after storage started germination tests under water stress with an initial MC of 30%. Moreover, seeds at lower initial MC need more time to imbibe before the germination process can start. Seeds with higher initial MC were probably able to cope better with water deficit, at least during the 30-day germination test in the laboratory. A second experiment carried out on beechnuts treated only before storage but made to have different initial MC seemed to confirm this conclusion. Ungerminated seeds were not damaged, as revealed by a tetrazolium test performed at the end of each germination test. Advantages in nursery practice shown by dry, nondormant beechnuts (pretreated before storage) are discussed in relation to the possibility of sowing when water availability in the soil is not a limiting factor.

Description: The recommended method for N fertilization to winter wheat (Triticum aestivum L.) on the Canadian prairies has been to broadcast ammonium nitrate (AN) during early spring. In the Chinook region of southern Alberta, considerable interest exists in alternative formulations (particularly urea), times of application and placements. To determine the effect of alternative N fertilizer practices on winter wheat in southern Alberta, two field experiments were conducted over 2 consecutive years (1998-1999 and 1999-2000) at three locations. In the first experiment, fall applications of urea or coated urea, seed-placed or banded, were compared to the standard practice of spring-broadcast AN. At five of six sites, there was no difference between fall-banded urea and coated urea in plant stand, grain yield or protein concentrations when compared to spring-broadcast AN. In 1998-1999, fall-banded urea reduced grain yield by 13% at the site in the Brown soil. Seed-placed N was only safe for urea at 30 kg N ha-1 and for coated urea at rates up to 60 kg N ha-1. In the second experiment, urea and coated urea were broadcast in spring for comparison with AN. Coated urea was ineffective in dry years due to poor N release. Urea was equally effective as AN in this study, possibly due to the cool, dry conditions at the time of application and the relatively low surface soil pH levels at these reduced tillage sites. Further research will be required to confirm the effectiveness of this practice for this region. Key words: Ammonium nitrate, urea, coated urea, nitrogen fertilizer placement, nitrogen timing, grain protein

Description: The nitrogen (N) cycle is second in importance only to the carbon (C) cycle in natural ecosystems. However, the origin and nature of approximately one-third of total soil N remains poorly understood. We used N-XANES (X-ray absorption near edge structure) spectroscopy, a sensitive, non-destructive synchrotron-based analytical technique, to determine the organic N structures in selected Gleysolic soils of Saskatchewan. Initial results indicate the presence of amide structures and heterocyclic N compounds, which are significant in understanding the transformation of organic N moieties in the soil, both in terms of sources of bio-available N and long-term storage. Key words: Unknown N, amides, heterocyclic N compounds, wetland soils, N-XANES spectroscopy

Description: A study was initiated in 1996 in the Black Soil zone in east-central Saskatchewan to examine soil and crop response to application of feedlot cattle manure at different application rates, frequencies and incorporation timing in a sandy loam and loam soil. Three rates of feedlot cattle manure (approx. 100, 200 and 400 kg total N ha-1) were applied annually and under reduced frequency application regimes. Canola (Brassica napus, L.), spring wheat (Triticum aestivum, L.), hulless barley (Hordeum vulgare, L.) and canola were seeded in spring of 1997, 1998, 1999 and 2000, respectively. Pre-seeding available N (0–60 cm) increased with application rates. Annual application resulted in a linear increase in grain yield with application rates but had no effect on grain N concentration. Cumulative N use efficiency was low (7–10%) with no significant difference among treatments. Single application showed significant residual fertility benefit in the second year but not in the third year except at the high rate. Incorporation timing of feedlot cattle manure had no impact on soil or crop performance. Low availability of N in feedlot cattle manure over the short-term suggests the need for high application rates or addition of supplemental N fertilizer in order to meet crop N requirements. Key words: Feedlot cattle manure, N availability, manure management, N use efficiency

Description: Accurate in situ determination of soil water content is important in many fields of agricultural, environmental, hydrological, and engineering sciences. As numerous soil water content sensors are available on the market today, the knowledge of their performance will aid users in the selection of appropriate sensors. The objectives of this study were to evaluate five soil water sensors in the laboratory and to determine if laboratory calibration is appropriate for the field. In this study, the performances of five sensors, including the Profile Probe™ (PP), ThetaProbe™ , Watermark™, Aqua-Tel™, and Aquaterr™ were compared in the laboratory. The PP and ThetaProbe™ were more accurate than the other soil water sensors, reproducing soil water content using factory recommended parameters. However, when PP was installed on a loamy sand in the field, the same soil that was used for the laboratory evaluation, it overestimated field soil water, especially at depth. Another laboratory experiment showed that soil water content readings from the PP were strongly influenced by soil bulk density. The higher the soil bulk density, the greater was the overestimation of soil water content. Two regression parameters, a0 and a1, which are used to convert the apparent dielectric constant to volumetric water content, were found to increase linearly with the soil bulk density in the range of 1.2 to 1.6 Mg m-3. Finally, the PP was calibrated in the field and a good calibration function was obtained with an r2 of 0.87 and RMSE of 2.7%. The values of a0 and a1 obtained in the field were different from factory recommended parameters (a0 = 2.4 versus 1.6 while a1 = 12.5 versus 8.4) and were independent of soil depth, bulk density, and texture. As such, individual field calibration will be necessary to obtain precise and accurate measurement of soil water content with this instrument. Key words: Soil water content, Profile Probe, calibration, soil water content sensor