Acute Oak Decline was confirmed at 22 sites across Croatian pedunculate oak forests; among 2,580 trees assessed at confirmed sites, 21.0
Drone-borne LiDAR can measure forest structure quickly and accurately. In a poplar plantation, tree height was retrieved reliably under almost any flight setting, whereas trunk diameter proved far more demanding: accurate values required flying in winter (leafless trees), at low altitude, with triple returns and generous plot margins, and along the planting rows. These practical guidelines make drone-based forest monitoring more reliable and reproducible. Accurate tree height and trunk diameter measurements underpin forest inventory, biomass estimation, and sustainable management. UAV-borne LiDAR is a promising tool for tree-level inventory, yet how flight parameters and acquisition context affect the accuracy of derived structural metrics remains insufficiently documented. We quantified how UAV flight altitude, speed, plot margins, number of returns, flight plan geometry, and seasonal conditions influence canopy height and trunk diameter estimation. Twenty-two UAV flights were conducted over a managed 0.23-ha poplar plantation in southwestern France (2024–2025) using a DJI Matrice 300 with a Zenmuse L1 sensor and different varied parameters. Tree-level metrics were compared with field measurements, terrestrial and airborne LiDAR, and satellite-derived canopy height maps. Canopy height was retrieved robustly (best RMSE = 0.33 m, best R2 = 0.99 versus terrestrial LiDAR), requiring roughly 1000 pts·m⁻2 and altitudes below 60 m above the canopy. Trunk diameter was far more sensitive: accurate retrieval (best RMSE = 3.2 cm; best R2 = 0.90) required leaf-off conditions, low altitude, triple returns, generous margins, and flight lines aligned with planting rows. UAV LiDAR can replace labor-intensive field measurements for canopy height under almost any flight setting, but reliable trunk diameter requires leaf-off, low-altitude flights aligned with planting rows; notably, generous plot margins allow flying higher and faster without loss of accuracy, reducing acquisition time and data volume.
Pine weevil feeding on bark reduced root carbohydrate concentrations in Scots pine seedlings, while no changes were detected in needle phenolics or non-volatile metabolomic profiles after seven days. The large pine weevil (Hylobius abietis L.) is a major bark-feeding pest of conifer seedlings in Europe. Pine weevil feeding has been shown to induce systemic volatile emissions from undamaged shoots, raising the question of whether bark feeding also affects whole-plant carbon allocation and systemic chemical responses. We investigated how pine weevil feeding affects organ-specific compounds from primary and secondary metabolism in Scots pine (Pinus sylvestris L.) seedlings and whether these responses differed between two local provenances. Soluble sugars were quantified in roots and needles to assess organ-specific changes in carbohydrate status. Phenolic compounds and non-volatile metabolomic profiles were analyzed in needles to evaluate systemic chemical responses after seven days of pine weevil feeding. Stem damage was quantified, and treatment and provenance effects were evaluated using linear mixed-effects models. Pine weevil feeding significantly reduced soluble carbohydrate concentrations in roots but not in needles. No treatment effect was detected for total needle phenolic concentrations, although ferulic acid decreased following pine weevil feeding. Non-volatile metabolomic profiles showed no clear separation between damaged and control seedlings. Pine weevil feeding decreased carbohydrate levels in roots but did not induce detectable changes in secondary metabolites in the two tested provenances.
Alpine or mountainous forest genomics research is rapidly moving from classic population genetic studies toward exploring population environmental adaptation to address climate vulnerabilities. For alpine conifers in particular, future progress in sustainable management relies heavily on overcoming the current bottleneck in the availability of genomic and experimental resources through targeted, interdisciplinary research efforts. Mountainous forests are among the most vulnerable ecosystems to climate change. Over the past two decades, rapid advances in high-throughput sequencing have transformed research on mountainous tree species, enabling a transition from marker-based population genetic studies to genome-wide analyses of demography and adaptation. Here, we present a quantitative survey of how the field of mountainous or alpine forest genomics has evolved and provide future perspectives. We conducted a bibliometric analysis of 662 research articles published between 2000 and 2024, focusing on seven of the most important high-elevation European conifer species. Topic modeling and co-occurrence network analyses were used to identify major research topics and their dynamics in time. Moreover, members of the Alpine Forest Genomics Network (AForGeN, IUFRO Working Party 2.04.11) were asked in a survey to indicate future priorities and challenges of the field. Our results show that publication activity has increased substantially after 2010 and that Picea abies (59
Integrating ecological process knowledge into machine learning substantially improves the robustness and extrapolation of forest carbon stock predictions. The knowledge-guided model maintained strong predictive performance over 10–15-year horizons, when conventional machine-learning performance declined markedly, providing a promising framework for long-term forest carbon assessment under climate change. Accurate quantification of forest carbon stocks and their dynamics is essential for understanding ecosystem carbon sequestration potential. However, traditional machine learning models often exhibit limited extrapolation capability in long-term forest carbon predictions because they rely primarily on statistical relationships learned from historical observations and lack explicit representation of ecological processes. This study aimed to develop a knowledge-guided machine learning (KGML) framework that integrates ecological process knowledge with data-driven algorithms to improve the accuracy, stability, and long-term generalization of forest carbon stock predictions. The study focused on natural forests in Heilongjiang Province, China, defined as stands classified as having a natural origin in forest inventory records. Forest fixed plot observations from 1976 to 2015 were combined with historical and projected climate data and atmospheric CO2 concentration data. The CO2 fertilization effect was incorporated into an optimized forest carbon sequestration (FCS) process model, and its outputs and process-based constraints were subsequently integrated with machine learning algorithms to construct the KGML framework. Model performance was compared with that of standalone machine learning and process-based models across forest types and prediction horizons. Incorporating the CO2 fertilization effect improved the predictive performance of the FCS model, increasing the coefficient of determination (R2) from approximately 0.85 to above 0.90. The KGML model consistently outperformed the standalone machine learning and process-based models across the evaluated forest types. Its advantage was particularly evident for prediction horizons of 10 years or longer, over which the performance of conventional machine learning models declined substantially. Scenario simulations indicated that natural forests in Heilongjiang Province had the greatest projected carbon sequestration potential under SSP126. Depending on model configuration and forest type, projected carbon stocks in 2060 ranged from 1371.37 ± 422.04 to 1442.38 ± 316.77 Tg. Integrating ecological process knowledge with machine learning can improve the accuracy and long-term extrapolation capability of forest carbon stock predictions. The proposed KGML framework provides a potentially transferable approach for regional forest carbon assessment. Nevertheless, the projections remain subject to uncertainty arising from climate scenarios, process-model assumptions, forest inventory coverage, and the representation of disturbances and management effects.
