Abstract 3D laser scanning is a cornerstone of modern forest and ecological research, allowing detailed insight into forest structures and dynamics. Although point cloud processing and noise management are crucial steps in the exploitation of LiDAR data, traditional denoising methods are usually based on the assumption of a Gaussian noise distribution, whereas many factors (including structural complexity of forest environments) lead to the opposite By overlooking the complexity of noise distribution within point clouds, these methods often fail to distinguish noise correctly, resulting in a loss of relevant information and introducing potential biases. This is particularly problematic in areas with smaller stems and branches, where their pronounced curvature leads to light diffraction, complicating reliable detection. To address these issues, we propose a new noise compression method designed to better distinguish noise from information‐bearing points within MLS‐point clouds. By self‐optimizing its parameters against ground‐truth data, our method significantly enhances the accuracy and reliability of MLS‐derived metrics while preserving essential structural details. Compared to previous denoising methods, our approach minimizes data loss and ensures a more accurate representation of forest structures across diverse conditions (including small stems and branches), widening the potential contributions of 3D mapping to other ecological spheres.

For several decades, the proportion of American beech (Fagus grandifolia) has increased in sugar maple (Acer saccharum) stands in North America, particularly in the understory, where it can hinder the regeneration of other species. Although this dominance is well documented, the underlying mechanisms by which it develops remain poorly understood. We aimed to determine whether the spatial relationships between beech saplings, mature beech trees, and canopy openness, previously observed among stands using discontinuous plot-based sampling, also applies within stands using continuous sampling of full stands. Our objective was to better understand how this dominance develops, as it is uncertain whether these relationships hold at finer spatial scales within stands, i. e., whether beech dominance within a stand is spatially related to canopy gaps and/or mature beech density. We created 3D-maps of 11 1-ha stands in Quebec, Canada using mobile laser scanning (MLS) technology and developed a new method for analyzing regeneration which allowed us to map a total of 8455 trees and 30,498 saplings and to investigate the relationship between beech saplings and mature beech trees as well as canopy openness at various spatial scales. Contrary to our expectations, we found very little to no relationship between beech saplings and canopy openness, and a highly variable relationship with mature beech trees among sites. Our findings underscore the importance of spatial scale in regeneration studies and suggest that beech proliferation may be difficult to reverse, as sexual reproduction appears to play a more important role in beech regeneration than previously thought.

The formation of a single species, recalcitrant understory vegetation layer can limit tree regeneration and, in the long term, modify the composition of forests. Few studies have investigated how recalcitrant vegetation influences competition for water resources although the formation of a dense understory is likely to modify the forest water balance. In eastern North American hardwood forests, the development of a dense understory layer of American beech (Fagus grandifolia) has been observed in stands dominated by sugar maple (Acer saccharum), a phenomenon that shares many characteristics associated with recalcitrant vegetation. Given that water availability is generally negatively correlated with stand density, we hypothesized that the formation of a dense understory beech layer increased competition for water resources, thus leading to reduced water use by sugar maple trees in beech-dominated stands. Using thermal dissipation sensors, we measured sap flux density (Fd) of two sugar maple trees at three beech-dominated sites and three control sites. During the growing season, Fd of sugar maple trees was significantly larger at beech-dominated sites compared to control sites, indicating a greater rate of water use by sugar maples in stands with a dense understory beech layer. We provide two hypotheses to explain our results at the tree scale: (i) reduced cover by forest floor vegetation could limit transpiration by this layer, thus allowing increased water availability to supply transpiration by overstory trees, or (ii) increased tree transpiration rate could be a mechanism to satisfy nutrient requirements in beech-dominated stands often associated with lower soil fertility.

Under climate change, forests are expected to experience drier conditions that may increase tree mortality. Silvicultural treatments, such as thinning, have been proposed to reduce moisture competition and to improve forest resistance to drought events. Most studies have investigated the effectiveness of thinning under semi-arid conditions, while little information is available regarding temperate forest responses, together with the residual basal area (BA) that is required to reap the benefits of these treatments. This research aims to understand how the residual BA influences transpiration in mixed temperate forest stands that are dominated by red maple (Acer rubrum) in southeastern Canada. We monitored the sap flux density (Fd) with thermal dissipation-type sensors for 18 red maples spread across three sites that were each thinned to obtain a gradient of residual BA (20, 12.5, 6 m2/ha). The study was conducted during the first growing season following treatment. Low residual BA plots (6 m2/ha) incurred drier atmospheric conditions as shown by a greater vapor pressure deficit (VPD) compared to high residual BA plots (20 m2/ha), although we observed considerable variability between sites. At the tree scale, Fd increased with residual BA, with the most pronounced differences under dry atmospheric conditions: when daily VPD exceeded 1.1 kPa, mean Fd in high residual BA plots was respectively 20 % and 75 % greater than in medium (12.5 m2/ha) and low residual BA plots. At the stand level, we simulated total transpiration considering the stand as only made of red maples. The transpiration in medium and low residual BA plots amounted to 41 % and 79 % of transpiration simulated in the high residual BA plot. Overall, this work highlighted broad variation in response to residual BA treatments, emphasizing the need to better model forest water budgets, and partitioning overstory and understory evapotranspiration to make more adequate residual BA prescriptions in temperate forests.

The reconstruction of trees from point clouds that were acquired with terrestrial LiDAR scanning (TLS) may become a significant breakthrough in the study and modelling of tree development. Here, we develop an efficient method and a tool based on extensive modifications to the skeletal extraction method that was first introduced by Verroust and Lazarus in 2000. PypeTree, a user-friendly and open-source visual modelling environment, incorporates a number of improvements into the original skeletal extraction technique, making it better adapted to tackle the challenge of tree perennial tissue reconstruction. Within PypeTree, we also introduce the idea of using semi-supervised adjustment tools to address methodological challenges that are associated with imperfect point cloud datasets and which further improve reconstruction accuracy. The performance of these automatic and semi-supervised approaches was tested with the help of synthetic models and subsequently validated on real trees. Accuracy of automatic reconstruction greatly varied in terms of axis detection because small (length < 3.5 cm) branches were difficult to detect. However, as small branches account for little in terms of total skeleton length, mean reconstruction error for cumulated skeleton length only reached 5.1% and 1.8% with automatic or semi-supervised reconstruction, respectively. In some cases, using the supervised tools, a perfect reconstruction of the perennial tissue could be achieved.

BACKGROUND AND AIMS To meet the increasing need for rapid and non-destructive extraction of canopy traits, two methods were used and compared with regard to their accuracy in estimating 2-D and 3-D parameters of a hybrid poplar sapling. METHODS The first method consisted of the analysis of high definition photographs in Tree Analyser (TA) software (PIAF-INRA/Kasetsart University). TA allowed the extraction of individual traits using a space carving approach. The second method utilized 3-D point clouds acquired from terrestrial light detection and ranging (T-LiDAR) scans. T-LiDAR scans were performed on trees without leaves to reconstruct the lignified structure of the sapling. From this skeleton, foliage was added using simple modelling rules extrapolated from field measurements. Validation of the estimated dimension and the accuracy of reconstruction was then achieved by comparison with an empirical data set. KEY RESULTS TA was found to be slightly less precise than T-LiDAR for estimating tree height, canopy height and mean canopy diameter, but for 2-D traits both methods were, however, fully satisfactory. TA tended to over-estimate total leaf area (error up to 50 %), but better estimates were obtained by reducing the size of the voxels used for calculations. In contrast, T-LiDAR estimated total leaf area with an error of <6 %. Finally, both methods led to an over-estimation of canopy volume. With respect to this trait, T-LiDAR (14·5 % deviation) greatly surpassed the accuracy of TA (up to 50 % deviation), even if the voxels used were reduced in size. CONCLUSIONS Taking into account their magnitude of data acquisition and analysis and their accuracy in trait estimations, both methods showed contrasting potential future uses. Specifically, T-LiDAR is a particularly promising tool for investigating the development of large perennial plants, by itself or in association with plant modelling.

Concerns over decreases in soil nitrogen reserves and productivity following the removal of logging residues (windrowing, shearblading and piling) have been raised by numerous researchers. Medium-term impacts of this practice on soil N reserves and availability and on indices of organic matter quality were assessed for balsam fire (Abies balsamea (L.) Mill.), white birch (Betula paperyfera Marsh.) and white spruce (Picea glauca (Moench) Voss) stands growing on dry to fresh clayey sites in northwestern Quebec, Canada. Unharvested control stands, whole-tree harvested cutovers and windrowed sites were compared. Fifteen years following harvesting and windrowing, forest floor Kjeldahl N concentrations and content and forest floor in situ net N mineralization rates (undisturbed closed top cores incubation) were affected by harvesting but not by windrowing. No differences in mineralization constant, potentially mineralizable N and cumulative mineralized N (526 day incubation period) were found between treatments, suggesting that treatment differences in field net N mineralization rates were the result of interactions between residual ecosystem structures such as forest floor, coarse woody debris and vegetation and meteorological conditions. If these trends persist over time, it could signal that, while whole-tree harvesting does not have a direct effect on soil organic matter quality, long-term impacts on N dynamics could result from changes in ecosystem structures. Slash removal following whole-tree harvesting did not have any additional negative impact.

A sustainable forest management system requires that a balance must be reached between ecosystem nutrient losses and gains in the course of a rotation. In order to determine the influence of stand characteristics (species composition, density, site potential productivity), method of forest harvesting (stem-only versus whole-tree) as well as rotation length on nutrient losses induced by biomass harvesting, a geochemical balance (nutrient inputs minus outputs) was computed from published information and forest inventory databases for the southern portion of the boreal forest of Quebec. Losses were compared with potential nutrient gains that varied according to soil types. We provided a tool for assessing the risk of having a negative nutrient budget (outputs>inputs) that forest managers can use with information that is already available to them. This exercise was conducted for five commercial tree species, namely paper birch (Betula papyrifera Marsh.), aspen (Populus tremuloides Michx.), balsam fir (Abies balsamea (L.) Mill.), jack pine (Pinus banksiana Lamb.), and black spruce (Picea mariana (Mill.) B.S.P.), grouped into four site indexes and three stand density classes. Strong differences in nutrient exportation in biomass appeared between stands of different compositions as well as between stands of the same species but of different classes of productivity or density. The most nutrient-demanding tree species were trembling aspen and balsam fir. As expected, whole-tree harvesting caused a greater drain on nutrient reserves than stem-only harvesting, but this effect varied strongly with tree species and was greatest for balsam fir and lowest for jack pine. Harvesting the forest before or after the age of financial maturity, which might be desirable under some circumstances, generally created a lesser nutrient drain but this was at the expense of biomass production. Aspen was an exception to this rule showing a greater nutrient drain for stands harvested prior to the age of financial maturity. Implications for the development of indicators of sustainable forestry and for future research are discussed.