Search Results (211)

Forest density is a key parameter in forestry research, and its variation can significantly impact ecosystems. Saihanba, as a focal site for afforestation and restoration, offers an ideal case for monitoring these dynamics. In this study, we compared three machine learning algorithms—Random Forest, Support Vector Regression, and XGBoost—using Landsat surface reflectance data together with the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI), and reference tree densities derived from LiDAR individual tree segmentation. The best-performing algorithm, XGBoost (R² = 0.65, RMSE = 174 trees ha⁻¹), was then applied to generate a long-term forest density dataset for Saihanba at five-year intervals, covering the period from 1988 to 2023. Results revealed distinct differences among tree species, with larch achieving the highest accuracy (R² = 0.65, RMSE = 161 trees ha⁻¹), whereas spruce had larger prediction errors (RMSE = 201 trees ha⁻¹) despite a relatively high R² (0.70). Incorporating 30 m slope data revealed that moderate slopes (5–30°) favored faster forest recovery. From 1988 to 2023, average forest density rose from 521 to 628 trees ha⁻¹—a 20.6% increase—demonstrating the effectiveness of restoration and providing a transferable framework for large-scale ecological monitoring. Full article

Accurate estimates of operational speed are crucial for modeling skidding productivity and planning efficient timber extraction. This study provides an event-level characterization of operational speeds in timber skidding operations in Romania, comparing cable skidders and farm tractors. Unlike most previous studies, which are based on limited datasets, this research uses a large, diverse dataset obtained through GNSS tracking over 98 field days at 14 sites, supplemented by synchronized video recordings. A total of 1.74 million seconds of data were collected, with 1.20 million seconds retained for analysis after data quality filtering. Descriptive statistics and Mann–Whitney U tests revealed significant differences in speed. For cable skidders, median speeds ranged from 1.6 km/h during maneuvering at the pre-skidding site to 5.0 km/h during unloaded driving to the pre-skidding site. For farm tractors, median speeds ranged from 2.2 km/h during maneuvering on the forest road to 6.0 km/h when driving unloaded to the pre-skidding site. The highest speeds were observed during unloaded driving, while the lowest occurred during maneuvering. Surprisingly, farm tractors outperformed cable skidders in some operational events due to more favorable terrain. The findings document GNSS-derived speed as a sufficiently reliable proxy for machine performance assessment and provide robust data for predictive modeling, operational planning, and equipment selection in forestry. Full article

Climate change significantly affects hydrological processes in forest ecosystems, particularly in sensitive coastal areas such as the Croatan National Forest (CNF) in North Carolina. Accurate projections of future water yield are essential for managing agriculture, forestry, and natural ecosystems. This study investigates the potential impacts of climate change on water yield using a combination of statistical downscaling and machine learning. Two downscaling methods, a Statistical DownScaling Model (SDSM) and Multivariate Adaptive Constructed Analogs (MACA), were evaluated, with the SDSM providing superior performance for local climate conditions. To improve precipitation input accuracy, twenty ensemble scenarios were generated using the SDSM, and various machine learning algorithms were applied to identify the optimal ensemble. Among these, the Extreme Gradient Boosting (XGBoost) algorithm exhibited the lowest error and was selected for producing high-quality precipitation time series. This methodology is integrated into the MIDAS (Machine Learning-Based Integration of Downscaled Projections for Accurate Simulation) approach, which leverages machine learning to enhance climate input precision and reduce uncertainty in hydrological modeling. Water yield was simulated over the period 1961–2060, combining observed and projected climate data to capture both historical trends and future changes. The results show that combining statistical downscaling with machine learning algorithms can help improve the accuracy of water yield projections under climate change and be useful for water resource planning, forest management, and climate adaptation. Full article

Evaluating the restoration quality of university outdoor spaces is often constrained by subjective surveys and manual assessment, limiting scalability and objectivity. This study addresses this gap by applying explainable machine learning to predict restorative quality from campus imagery, enabling large-scale, data-driven evaluation and capturing complex nonlinear relationships that traditional methods may overlook. Using Fujian Agriculture and Forestry University as a case study, this study extracted road network data, generated 297 coordinates at 50-m intervals, and collected 1197 images. Surveys were conducted to obtain restorative quality scores. The Mask2Former model was used to extract landscape features, and decision tree algorithms (RF, XGBoost, GBR) were selected based on MAE, MSE, and EVS metrics. The combination of optimal algorithms and SHAP was employed to predict restoration quality and identify key features. This research also used a multivariate linear regression model to identify features with significant statistical impact but lower features importance ranking. Finally, the study also analyzed heterogeneity in scores for three restoration indicators and five campus zones using k-means clustering. Empirical results show that natural elements like vegetation and water positively affect psychological perception, while structural components like walls and fences have negative or nonlinear effects. On this basis, this study proposes spatial optimization strategies for different campus areas, offering a foundation for creating high-quality outdoor environments with restorative and social functions. Full article

In the context of the “Dual Carbon” goals and ecological civilization development, enhancing forestry ecological total factor productivity (FETFP) has become vital for advancing green development and environmental governance. Confronted with tightening resource constraints and pressure to transform traditional growth models, whether digital intelligence integration can effectively empower improvements in FETFP requires in-depth empirical validation. Based on publicly available panel data from 30 Chinese provinces spanning 2012 to 2022, this study constructs an index system for measuring digital intelligence integration and FETFP. Using the Double Machine Learning (DML) framework, the study empirically identifies the impact of digital intelligence development on FETFP and explores its internal mechanisms. The key results show that (1) digital intelligence integration significantly enhances FETFP. For every unit increase in digital and intelligent integration, FETFP rises by an average of 19.97%; (2) mechanism analysis reveals that digital intelligence improves FETFP by optimizing the forestry industrial structure, promoting green technological innovation, and amplifying the synergistic effects of fiscal support; (3) and heterogeneity analysis suggests that the positive impact of digital intelligence integration is more pronounced in regions with higher environmental expenditures and stronger green finance support. Accordingly, this study proposes several policy recommendations, including accelerating digital infrastructure development, strengthening foundational digital intelligence capabilities, enhancing support for green innovation, leveraging the ecological multiplier effects of digital transformation, tailoring digital strategies to local conditions, and improving the precision of regional environmental governance. The findings provide robust empirical evidence for improving FETFP in developing and developed economies. Full article

Globally, the accelerating aging of the rural labor force is profoundly impacting the economic resilience of the labor-intensive forestry sector. However, the intrinsic connection between the two has not been fully understood and requires further exploration. As the most populous nation globally and a top producer, trader, and consumer of forest products, China stands out as a perfect case study for this issue. Based on this, this study utilizes panel data from 30 provinces in China from 2012 to 2022 and employs a dual machine learning model to empirically examine the impact and mechanisms of rural labor force aging on forestry economic resilience from the perspective of production factors. The findings indicate: (1) overall, the increase in rural labor force aging significantly inhibits forestry economic resilience; (2) rural labor force aging enhances forestry economic resilience by promoting large-scale forest land management, driving forestry technological innovation, and increasing government capital investment; it also inhibits forestry economic resilience by reducing educational human capital and health human capital; (3) the rural force aging exerts a marked adverse effect on the resilience of the forestry economy in the eastern and central regions, major grain-producing areas, and major grain-consuming areas. Based on this, this study proposes policy recommendations in three areas: building a flexible and diversified labor supply and replacement system, exploring a “scale and technology” integration path suited to national conditions, and implementing differentiated regional strategies. The aim is to provide a reference for government departments in formulating strategies to enhance the resilience of the forestry economy in the era of aging. Full article

In the era of climate change, the strategic importance of forestry products for sustainable development is increasingly recognized. Amid a global resurgence of industrial policy aimed at addressing environmental challenges, this study investigates the impact of China’s central and provincial green industrial policies on the export competitiveness of wood-processing enterprises. Utilizing firm-level data from the China Industrial Enterprise Database and China Customs Export Database (2000–2013), we apply a double machine learning (DML) approach and construct a heterogeneous competitiveness model to evaluate policy effects along two dimensions: export quantity (volume and intensity) and export quality (product complexity and consumer-perceived quality). Our findings reveal a clear dichotomy in policy outcomes. While industrial policies have significantly improved export product complexity—reflecting China’s comparative advantage in labor-intensive production—they have had limited or even negative effects on export volume, intensity, and product quality. This suggests that current policy frameworks disproportionately reward horizontal innovation (product diversification) while neglecting vertical upgrading (quality enhancement), thereby hindering comprehensive export performance gains. Those results highlight the need for more balanced and targeted policy design. By aligning industrial policy instruments with both complexity and quality objectives, policymakers can better support the sustainable transformation of China’s forestry sector and enhance its competitiveness in global value chains. Full article

In this study, we comprehensively analyzed material consumption (fuel, hydraulic oil, lubricants, and AdBlue fluid) and estimated carbon dioxide emissions during logging operations. This study was carried out in the northeastern part of Poland. Four harvesters and four forwarders representing two manufacturers (John Deere-Deere & Co., Moline, USA, and Komatsu Forest AB, Umeå, Sweden) were analyzed to compare their operational efficiency and constructional influences on overall operating costs. Due to differences in engine emission standards, approximate greenhouse gas emissions were estimated. The results indicate that harvesters equipped with Stage V engines have lower fuel consumption, while large forwarders use more consumables than small ones per hour and cubic meter of harvested and extracted timber. A strong positive correlation was observed between total machine time and fuel consumption (r = 0.81), as well as between machine time and total volume of timber harvested (r = 0.72). Older and larger machines showed about 40% higher combustion per unit of wood processed. Newer machines meeting higher emission standards (Stage V) generally achieved lower CO₂ and other GHG emissions compared to older models. Machines with Stage V engines emitted about 2.07 kg CO₂ per processing of 1 m³ of wood, while machines with older engine types emitted as much as 4.35 kg CO₂ per 1 m³—roughly half as much. These differences are even more pronounced in the context of nitrogen oxide (NO_x) emissions: the estimated NO_x emissions for the older engine types were as high as ~85 g per m³, while those for Stage V engines were only about 5 g per m³ of harvested wood. Continuing the study would need to expand the number of machines analyzed, as well as acquire more detailed performance data on individual operators. A tool that could make this possible would be fleet monitoring services offered by the manufacturers of the surveyed harvesters and forwards, such as Smart Forestry or Timber Manager. Full article

Broader use of multioperational machines in forestry requires efficient methods for determining various timber parameters. Here, we present a novel approach to model the bark thickness (BT) as a function of stem diameter. Stem diameter (D) is any diameter measured along the bole, not a specific one. The following four regression models were tested: marginal model (MM; reference), classical nonlinear regression with independent residuals (M1), nonlinear regression with residuals correlated within a single tree (M2), and nonlinear regression with the correlation of residuals and random components, taking into account random changes between the trees (M3). Empirical data consisted of larch (Larix sp. Mill.) BT measurements carried out at two sites in northern Poland. Relative root square mean error (RMSE%) and adjusted R-squared (R2adj) served to compare the fitted models. Model fit was tested for each tree separately, and all trees were combined. Of the analysed models, M3 turned out to be the best fit for both the individual tree and all tree levels. The fit of the regression function M3 for SITE1 (50-year-old, pure stand located in northern Poland) was 87.44% (R2adj), and for SITE2 (63-year-old, pure stand situated in the north of Poland) it was 80.6%. Taking into account the values of RMSE%, at the individual tree level the M3 model fit at location SITE1 was closest to the MM, while at SITE2 it was better than the MM. For the most comprehensive regression model, M3, it was checked how the error of the bark thickness estimate varied with stem diameter at different heights (from the base of the trees to the top). In general, the model’s accuracy increased with greater tree height. Full article

Accurate biomass estimation is important for forest management and climate change mitigation. This study evaluates the potential of using LiDAR (Light Detection and Ranging) data, acquired through Unmanned Aerial Vehicles (UAVs), for estimating above-ground and total biomass in Eucalyptus globulus and Pinus pinaster stands in central and northern Portugal. The acquired LiDAR point clouds were processed to extract structural metrics such as canopy height, crown area, canopy density, and volume. A multistep variable selection procedure was applied to reduce collinearity and select the most informative predictors. Multiple linear regression (MLR) models were developed and validated using field inventory data. Random Forest (RF) models were also tested for E. globulus, enabling a comparative evaluation between parametric and machine learning regression models. The results show that the 25th height percentile, canopy cover density at two meters, and height variance demonstrated an accurate biomass estimation for E. globulus, with coefficients of determination (R²) varying between 0.86 for MLR and 0.90 for RF. Although RF demonstrated a similar predictive performance, MLR presented advantages in terms of interpretability and computational efficiency. For P. pinaster, only MLR was applied due to the limited number of field data, yet R² exceeded 0.80. Although absolute errors were higher for Pinus pinaster due to greater biomass variability, relative performance remained consistent across species. The results demonstrate the feasibility and efficiency of UAV LiDAR point cloud data for stand-level biomass estimation, providing simple and effective models for biomass estimation in these two species. Full article

This study advances vegetation and forest mapping in temperate mixed forests by integrating airborne hyperspectral imagery (HSI) and light detection and ranging (LiDAR) data, overcoming the limitations of conventional multispectral imaging. Employing a Digital Canopy Height Model (DCHM) derived from LiDAR, our approach integrates these structural metrics with hyperspectral spectral information, alongside detailed remote sensing data extraction. Through machine learning-based clustering, which combines both structural and spectral features, we successfully classified eight specific tree species, community boundaries, identified dominant species, and quantified their abundance, contributing to precise vegetation and forest type mapping based on predominant species and detailed attributes such as diameter at breast height, age, and canopy density. Field validation indicated the methodology’s high mapping precision, achieving overall accuracies of approximately 98.0% for individual species identification and 93.1% for community-level mapping. Demonstrating robust performance compared to conventional methods, this novel approach offers a valuable foundation for National Forest Ecology Inventory development and significantly enhances ecological research and forest management practices by providing new insights for improving our understanding and management of forest ecosystems and various forestry applications. Full article

Principal component analysis (PCA) is a widely applied multivariate statistical technique across scientific disciplines, with forestry being one of its most dynamic areas of use. Its primary strength lies in reducing data dimensionality and classifying parameters within complex ecological datasets. This study provides the first comprehensive bibliometric and literature review focused exclusively on PCA applications in forestry. A total of 96 articles published between 1993 and 2024 were analyzed using the Web of Science database and visualized using VOSviewer software, version 1.6.20. The bibliometric analysis revealed that the most active scientific fields were environmental sciences, forestry, and engineering, and the most frequently published journals were Forests and Sustainability. Contributions came from 198 authors across 44 countries, with China, Spain, and Brazil identified as leading contributors. PCA has been employed in a wide range of forestry applications, including species classification, biomass modeling, environmental impact assessment, and forest structure analysis. It is increasingly used to support decision-making in forest management, biodiversity conservation, and habitat evaluation. In recent years, emerging research has demonstrated innovative integrations of PCA with advanced technologies such as hyperspectral imaging, LiDAR, unmanned aerial vehicles (UAVs), and remote sensing platforms. These integrations have led to substantial improvements in forest fire detection, disease monitoring, and species discrimination. Furthermore, PCA has been combined with other analytical methods and machine learning models—including Lasso regression, support vector machines, and deep learning algorithms—resulting in enhanced data classification, feature extraction, and ecological modeling accuracy. These hybrid approaches underscore PCA’s adaptability and relevance in addressing contemporary challenges in forestry research. By systematically mapping the evolution, distribution, and methodological innovations associated with PCA, this study fills a critical gap in the literature. It offers a foundational reference for researchers and practitioners, highlighting both current trends and future directions for leveraging PCA in forest science and environmental monitoring. Full article

The use of Google Earth Engine (GEE), a cloud-based computing platform, in spatio-temporal evaluation studies has increased rapidly in natural sciences such as forestry. In this study, Sentinel-2 satellite imagery and Shuttle Radar Topography Mission (SRTM) elevation data and image classification algorithms based on two machine learning techniques were examined. Random Forest (RF) and Classification and Regression Trees (CART) were used to classify land use and land cover (LULC) in western Oregon (USA). To classify the LULC from the spectral bands of satellite images, a composition consisting of vegetation difference indices NDVI, NDWI, EVI, and BSI, and a digital elevation model (DEM) were used. The study area was selected due to a diversity of land cover types including research forest, botanical gardens, recreation area, and agricultural lands covered with diverse plant species. Five land classes (forest, agriculture, soil, water, and settlement) were delineated for LULC classification testing. Different spatial points (totaling 75, 150, 300, and 2500) were used as training and test data. The most successful model performance was RF, with an accuracy of 98% and a kappa value of 0.97, while the accuracy and kappa values for CART were 95% and 0.94, respectively. The accuracy of the generated LULC maps was evaluated using 500 independent reference points, in addition to the training and testing datasets. Based on this assessment, the RF classifier that included elevation data achieved an overall accuracy of 92% and a kappa coefficient of 0.90. The combination of vegetation difference indices with elevation data was successful in determining the areas where clear-cutting occurred in the forest. Our results present a promising technique for the detection of forests and forest openings, which was helpful in identifying clear-cut sites. In addition, the GEE and RF classifier can help identify and map storm damage, wind damage, insect defoliation, fire, and management activities in forest areas. Full article

Selecting the most suitable machines to use for wood recovery is essential for computing the operating costs of the whole forestry machinery chain (FMC). Nevertheless, a generalized approach for selecting the most suitable FMC and quantifying the corresponding economic performances for wood recovery (i.e., harvesting and long-distance transport) is still missing. The primary aim of this study is to describe a decision support model, called FOREstry MAchinery chain selection (“FOREMA v1”), which is able to (i) select the most feasible FMC and (ii) calculate the costs (such as EUR∙h⁻¹; EUR∙t⁻¹ of dry matter, DM) of each operation (OP) comprising the FMC. The model is made up of three different modules (Ms): machinery chain selection (M1), machinery chain organization (M2), and cost calculation (M3). In M1, feasible FMCs are defined according to seven technical parameters that characterize the forest area. For each FMC, FOREMA v1 defines the sequence of OPs and the types of machines that can potentially be used. Once the characteristics of the area in which wood recovery occurs are processed, the user selects the types of machines to use according to the model’s suggestions. In M2 and M3, the user is supported in organizing the FMC (e.g., calculation of the required time, working productivity, and so on) and computing the operating costs. The secondary aim of this study is to discuss a case study focused on chips production for energy generation, providing empirical evidence on how FOREMA v1 works. The proposed model provides a systematic approach for the selection and optimization of the most suitable FMC to adopt for biomass recovery, thus supporting decision-making processes. The results showed that felling had the lowest cost per unit of time (63.7 EUR·h⁻¹) but the highest cost per unit of mass (35.4 EUR·t DM⁻¹) due to its longer working time and lower productivity. Loading and long-distance transport incurred the highest costs both per unit of time (223.5 EUR·h⁻¹) and per unit of mass (29.4 EUR·t DM⁻¹), attributed to the use of medium–small-sized trailers coupled with tractors operating at low speeds, leading to a high number of cycles. For the entire FMC the costs were equal to 147.3 EUR·h⁻¹ and 101.1 EUR·t DM⁻¹. Full article

In the context of global climate change and the deepening of ecological civilization construction, forestry, as an ecological security barrier and green economic engine, faces many challenges to the enhancement of its ecological total factor productivity in the traditional development model. As a new type of production factor, the data factor provides a new path to crack the bottleneck of forestry eco-efficiency improvement. Based on China’s provincial annual panel data from 2014 to 2022, this study systematically examines the impact and mechanism of data factors on forestry ecological total factor productivity by using the SBM-GML model and dual machine learning model. It was found that data factors have a significant contribution to forestry ecological total factor productivity, a conclusion that passes a series of robustness tests and endogeneity tests. The analysis of the mechanism shows that the data factor enhances the total factor productivity of forestry ecology mainly through three paths: promoting the progress of forestry technology and promoting the rationalization and advanced structure of the forestry industry. Further analysis showed that the promotional effect of data elements is more obvious in regions with a high level of green finance development, high intensity of environmental regulation, and strong financial autonomy. It is recommended to systematically promote the in-depth application of data elements in forestry, build a data element-driven innovation system for the whole chain of forestry, and implement regionally differentiated data element-enabling strategies. Full article