Forest Management Applications Research Articles

A Novel Drone Design Based on a Reconfigurable Unmanned Aerial Vehicle for Wildfire Management

Our study introduces a new approach, leveraging robotics technology and remote sensing for multifaceted applications in forest and wildfire management. Presented in this paper is PULSAR, an innovative UAV with reconfigurable capabilities, able of operating as a quadcopter, a co-axial quadcopter, and a standalone octocopter. Tailored to diverse operational requirements, PULSAR accommodates multiple payloads, showcasing its adaptability and versatility. This paper meticulously details material selection and design methods, encompassing both initial and detailed design, while the electronics design section seamlessly integrates essential avionic components. The 3D drone layout design, accomplished using SOLIDWORKS, enhances understanding by showcasing all three different configurations of PULSAR’s structure. Serving a dual purpose, this study highlights UAV applications in forest and wildfire management, particularly in detailed forest mapping, edge computing, and cartographic product generation, as well as detection and tracking of elements, illustrating how a UAV can be a valuable tool. Following the analysis of applications, this paper presents the selection and integration of payloads onto the UAV. Simultaneously, each of the three distinct UAV configurations is matched with a specific forest application, ensuring optimal performance and efficiency. Lastly, computational validation of the UAV’s main components’ structural integrity is achieved through finite element analysis (FEA), affirming the absence of issues regarding stress and displacement. In conclusion, this research underscores the efficacy of PULSAR, marking a significant leap forward in applying robotics technology for wildfire science.

Enhancing unmanned aerial vehicle capabilities: integrating YOLO algorithms for diverse industrial applications

The integration of UAVs with advanced deep learning algorithms, particularly the You Only Look Once models, has opened new horizons in various industries. This paper explores the transformative impact of YOLO-based systems across diverse sectors, including agriculture, forest fire detection, ecology, marine science, target detection, and UAV navigation. We delve into the specific applications of different YOLO models, ranging from YOLOv3 to the lightweight YOLOv8, highlighting their unique contributions to enhancing UAV functionalities. In agriculture, UAVs equipped with YOLO algorithms have revolutionized disease detection, crop monitoring, and weed management, contributing to sustainable farming practices. The application in forest fire management showcases the capability of these systems in real-time fire localization and analysis. In ecological and marine sciences, the use of YOLO models has significantly improved wildlife monitoring, environmental surveillance, and resource management. Target detection studies reveal the efficacy of YOLO models in processing complex UAV imagery for accurate and efficient object recognition. Moreover, advancements in UAV navigation, through YOLO-based visual landing recognition and operation in challenging environments, underscore the versatility and efficiency of these integrated systems. This comprehensive analysis demonstrates the profound impact of YOLO-based UAV technologies in various fields, underscoring their potential for future innovations and applications.

Estimating tree species composition from airborne laser scanning data using point-based deep learning models

Airborne laser scanning (ALS) data is increasingly being used for accurate estimations of forest inventory attributes, such as tree height and volume. However, determining tree species information, an important attribute required by inventories for a range of forest management applications, is challenging to extract from ALS data. This is due to complex species assemblages and vertical structures in certain forest environments, and a lack of spectral information for most ALS sensors. Our objective was to estimate tree species compositions in a Canadian boreal forest environment using ALS data and point-based deep learning techniques. To accomplish this, we utilised existing polygonal forest resource information to develop a large comprehensive dataset of tree species compositions across a 630,000 ha forest management area. We then used an adapted PointAugment generative adversarial network (GAN) coupled with the dynamic graph convolutional neural network to estimate tree species proportions. This innovative deep learning approach resulted in an overall R2adj of 0.61 for all tree species proportions. A weighted F1 score of 63% was observed when classifying the leading species of a plot, 66% for leading genus, and 85% when distinguishing between coniferous and deciduous dominated plots. Furthermore, the PointAugment GAN successfully generated augmented point clouds of forest plots which can alleviate issues associated with limited training samples for use in deep learning. This approach demonstrates the capability of point-based deep learning techniques to accurately estimate tree species compositions from ALS point clouds within an expansive forested region, characterized by a complex management history and a diversity of tree species. The source code along with the pretrained models are available at https://github.com/Brent-Murray/point-dl/tree/main/Pytorch/models/PointAugment.

Artificial intelligence and machine learning applications in forest management and biodiversity conservation

The recent progress in data science, along with the transformation in digital and satellite technology, has enhanced the capacity for artificial intelligence (AI) applications in the forestry and wildlife domains. Nevertheless, the swift proliferation of developmental projects, agricultural, and urban areas pose a significant threat to biodiversity on a global scale. Hence, the integration of emerging technologies such as AI in the fields of forests and biodiversity might facilitate the efficient surveillance, administration, and preservation of biodiversity and forest resources. The objective of this paper is to present a comprehensive review of how AI and machine learning (ML) algorithms are utilized in the forestry sector and biodiversity conservation worldwide. Furthermore, this research examines the difficulties encountered while implementing AI technology in the fields of forestry and biodiversity. Enhancing the availability of extensive data pertaining to forests and biodiversity, along with the utilization of cloud computing and digital and satellite technology, can facilitate the wider acceptance and implementation of AI technology. The findings of this study would inspire forest officials, scientists, researchers, and conservationists to investigate the potential of AI technology for the purposes of forest management and biodiversity conservation.

Plugging the Gaps in the Global PhenoCam Monitoring of Forests—The Need for a PhenoCam Network across Indian Forests

Our understanding of the impact of climate change on forests is constrained by a lack of long-term phenological monitoring. It is generally carried out via (1) ground observations, (2) satellite-based remote sensing, and (3) near-surface remote sensing (e.g., PhenoCams, unmanned aerial vehicles, etc.). Ground-based observations are limited by space, time, funds, and human observer bias. Satellite-based phenological monitoring does not carry these limitations; however, it is generally associated with larger uncertainties due to atmospheric noise, land cover mixing, and the modifiable area unit problem. In this context, near-surface remote sensing technologies, e.g., PhenoCam, emerge as a promising alternative complementing ground and satellite-based observations. Ground-based phenological observations generally record the following key parameters: leaves (bud stage, mature, abscission), flowers (bud stage, anthesis, abscission), and fruit (bud stage, maturation, and abscission). This review suggests that most of these nine parameters can be recorded using PhenoCam with >90% accuracy. Currently, Phenocameras are situated in the US, Europe, and East Asia, with a stark paucity over Africa, South America, Central, South-East, and South Asia. There is a need to expand PhenoCam monitoring in underrepresented regions, especially in the tropics, to better understand global forest dynamics as well as the impact of global change on forest ecosystems. Here, we spotlight India and discuss the need for a new PhenoCam network covering the diversity of Indian forests and its possible applications in forest management at a local level.

UAV4TREE: DEEP LEARNING-BASED SYSTEM FOR AUTOMATIC CLASSIFICATION OF TREE SPECIES USING RGB OPTICAL IMAGES OBTAINED BY AN UNMANNED AERIAL VEHICLE

Abstract. Automated tree classification from unmanned aerial vehicle (UAV) images is a challenging task with several applications in forest management and conservation. In this study, we propose UAV4Tree a Deep Learning based system that automatically classifies RGB optical images obtained by the UAV. In particular, we explore the use of augmented datasets and various deep learning models, including ResNet, DenseNet, InceptionV3, and Vision Transformer, for the classification of tree images obtained from UAVs. Our experiments show that the use of an augmented dataset can significantly improve the accuracy of the classification by approximately 10 points compared to the use of a non-augmented dataset. We also found that fine-tuning and the introduction of dropout were essential for improving the generalization ability of the models on the augmented dataset. Furthermore, the use of Super Resolution Generative Adversarial Network (SR-GAN) in the original dataset allowed us to increase the performance of some models. Our findings provide valuable insights into the use of deep learning models for automated tree classification from UAV imagery, which has significant implications for sustainable forest management and conservation.

Assessment of the Full Density Curve in Stand Density Management Diagrams for Hinoki (Chamaecyparis obtusa) in Kyushu Island, Japan: Implications for Forest Management

Accurate forest information on tree species, stand age, tree density, and stand volume is required to ensure effective forest management practices. In Japan, forest information is consolidated in forest yield tables and stand density management diagrams (SDMDs) that are specifically designed for major forest plantation species. In this study, we analyzed whether the current full density curve in the SDMD of Hinoki (Chamaecyparis obtusa) plantation stands in Kyushu Island aligns with the characteristics of the existing stands. Data from 18 Hinoki forests were used to measure tree heights, diameters, and densities. Equations were developed to establish relationships between stand factors, and various curves were derived for average height, competition ratio, full density, yield ratio, average diameter, and natural mortality. The results showed that the current full density curve in the SDMD for Hinoki plantation stands did not completely align with the characteristics of the existing Hinoki plantation stands in Kyushu Island. Thus, the full density curve in the SDMD for the Hinoki stands in this region should be significantly adjusted. These results can enhance forest management in Japan and advance SDMD modeling and its application in forest planning and management.

Tree Stem Detection and Crown Delineation in a Structurally Diverse Deciduous Forest Combining Leaf-On and Leaf-Off UAV-SfM Data

Accurate detection and delineation of individual trees and their crowns in dense forest environments are essential for forest management and ecological applications. This study explores the potential of combining leaf-off and leaf-on structure from motion (SfM) data products from unoccupied aerial vehicles (UAVs) equipped with RGB cameras. The main objective was to develop a reliable method for precise tree stem detection and crown delineation in dense deciduous forests, demonstrated at a structurally diverse old-growth forest in the Hainich National Park, Germany. Stem positions were extracted from the leaf-off point cloud by a clustering algorithm. The accuracy of the derived stem co-ordinates and the overall UAV-SfM point cloud were assessed separately, considering different tree types. Extracted tree stems were used as markers for individual tree crown delineation (ITCD) through a region growing algorithm on the leaf-on data. Stem positioning showed high precision values (0.867). Including leaf-off stem positions enhanced the crown delineation, but crown delineations in dense forest canopies remain challenging. Both the number of stems and crowns were underestimated, suggesting that the number of overstory trees in dense forests tends to be higher than commonly estimated in remote sensing approaches. In general, UAV-SfM point clouds prove to be a cost-effective and accurate alternative to LiDAR data for tree stem detection. The combined datasets provide valuable insights into forest structure, enabling a more comprehensive understanding of the canopy, stems, and forest floor, thus facilitating more reliable forest parameter extraction.

Advances in remote sensing for sustainable forest management: monitoring and protecting natural resources

Remote sensing has emerged as a powerful tool for the monitoring and management of forests, contributing to the sustainable utilization and protection of natural resources. This paper presents a review of recent advances in remote sensing techniques and technologies for forest management, highlighting their role in monitoring and protecting forests. The integration of remote sensing with other geospatial methods enhances the accuracy and efficiency of data acquisition, aiding in the assessment of forest cover, biomass estimation, disturbance detection, and biodiversity monitoring. Furthermore, the potential of remote sensing for supporting decision-making processes in sustainable forest management is explored, emphasizing its versatility, cost-effectiveness, and ability to provide invaluable insights at both local and global scales. The paper concludes by discussing current challenges and future opportunities in remote sensing applications for sustainable forest management.

A regret theory-based multi-granularity three-way decision model with incomplete T-spherical fuzzy information and its application in forest fire management

Forest fires are an abrupt and highly destructive meteorological disaster that can occur in all regions of the world, resulting in significant ecological, economic and social losses. Moreover, the causes of forest fire disasters are usually complex, involving several uncertain factors such as temperature, relative humidity, wind speed and rainfall. All of those pose the greatest challenge to the study of forest fire management (FRM). In order to efficiently explore FRM via valid intelligent decision-making techniques, a novel model of regret theory (RT)-based multi-granularity (MG) three-way decisions (TWD) in incomplete T-spherical fuzzy (T-SF) environments has been constructed, where incomplete T-spherical fuzzy sets (T-SFSs) have been employed to describe diverse types of uncertain information in FRM, and RT-based MG TWD is conducive to analyzing multi-source T-SF information via reducing decision risks and modeling bounded rationality owned by decision-makers (DMs). Specifically, the concept of MG T-SF incomplete information systems (IISs) has been first constructed for information depictions of FRM. Then, MG T-SF IISs have been processed via the presented T-SF similarity principles for developing adjustable MG T-SF probabilistic rough sets (PRSs). Afterwards, an RT-based MG TWD approach has been built with the support of adjustable MG T-SF PRSs. Finally, a real-world FRM case analysis has been performed by using the built RT-based MG TWD approach, and extensive comparative and experimental analyses have been performed to validate the practicability of the presented methodology. To sum up, the presented methodology has simultaneously incorporated MG T-SF IISs, MG TWD and RT to model various uncertainties, valid information fusion processes and bounded rationality for FRM, which serves as a valid intelligent decision-making technique in processing incomplete and imprecise multi-source information with plentiful decision risks and regret emotions.

Pre-fire measures of boreal forest structure and composition inform interpretation of post-fire spectral recovery rates

Wildfire is a key driver of forest dynamics in boreal forests; however, the annual area burned in boreal forests is highly variable, with increasing wildfire activity documented over the past half century. Post-fire recovery has important implications for carbon balance, and for the wide range of ecosystem goods and services provisioned by the boreal forest. Monitoring post-fire recovery is challenging given the vast and often inaccessible areas impacted, coupled with the marked variability in post-fire conditions and recovery processes. Field assessments of recovery are constrained in space and time, whereas remotely sensed data are spatially explicit, cover large areas, and can retrospectively provide assessments of pre- and post-fire conditions and establish spectral recovery baselines. However, there is a need to link spectral measures of recovery to manifestations of post-fire forest recovery on the ground. Understanding how different forest characteristics influence spectral recovery is key to the successful application of spectral recovery metrics for forest management applications in different forest environments. Herein, using a synthesis of plot data for the North American boreal forest, we assess the influence of pre- and post-fire field-measured characteristics on spectral recovery rates, with a focus on stem density and composition in Canadian boreal forests. Plots that experienced rapid spectral recovery (as measured using Landsat-derived NBR time series data) were associated with a transition from conifer to broadleaf or were broadleaf pre-fire. Plots that experienced a decrease in stem density took significantly longer to spectrally recover than plots that experienced no change in stem density. Plots that had not yet spectrally recovered by the end of the time series were associated with higher elevations, drier sites, greater pre-fire basal area, and had the greatest change magnitude. Our results emphasize that recovery is a process that is highly variable and that knowledge of pre-fire condition is important for characterizing and interpreting measures of post-fire spectral recovery for forest management applications. Remotely sensed time series data are uniquely able to provide information on both pre- and post-fire condition. Our analysis of pre- and post-fire field measures provided novel insights regarding the influence of forest composition and changes in stem density on measures of post-fire spectral recovery in boreal forests.

A Modified Hooke–Jeeves Algorithm in Two-Player Nonlinear Static Game: With an Application in Forest Management

Strategy considerations for optimal outcomes play the key roles in every decision-making problems. Game theory in itself preserves the sustainability to draw some decision-making conclusions. Different forms of game exist in literature. This article is concerned with two-player nonlinear static game. We consider here the solution of the game problems through the algorithmic structures, with some modification, of Hooke–Jeeves (HJ) algorithm which is a tool to encounter unconstrained optimization. Basically, we lay emphasis on the modification of the considered function which is to be optimized. Hence, the name of the proposed algorithm is modified HJ algorithm. Numerous articles present different solutions of various real-life problems using HJ algorithm. Forest management problem is considered here under modified HJ algorithm to get solution. This article concludes with some comparative analysis with respect to some different methods, for example, analytical method and graphical method. Our proposed methodology gives better results than the results obtained in analytical method and graphical method with respect to optimal strategy sets and optimal payoff values.

Using the Error-in-Variable Simultaneous Equations Approach to Construct Compatible Estimation Models of Forest Inventory Attributes Based on Airborne LiDAR

Airborne LiDAR has been extensively used for estimating and mapping forest attributes at various scales. However, most models have been developed separately and independently without considering the intrinsic mathematical relationships and correlations among the estimates, which results in the mathematical and biophysical incompatibility of the estimates. In this paper, using the measurement error model approach, the error-in-variable simultaneous equation (SEq) for airborne LiDAR-assisted estimations of four forest attributes (stand volume, V; basal area, G; mean stand height, H; and diameter at breast height, D) for four forest types (Chinese fir, pine, eucalyptus, and broad-leaved forest) is developed and compared to the independence models (IMs). The results indicated that both the SEqs and IMs performed well, and the rRMSEs of the SEqs were slightly larger than those of the IMs, while the increases in rRMSE were less than 2% for the SEqs. There were statistically significant differences (α = 0.05) in the means of the estimates between SEqs and IMs, even though their average differences were less than ±1.0% for most attributes. There were no statistically significant differences in the mean estimates between SEqs, except for the estimates of the D and G of the eucalyptus forest. The SEqs with H and G as the endogenous variables (EVs) to estimate V performed slightly better than other SEqs in the fir, pine, and broad-leaved forests. The SEq that used D, H, and V as the EVs for estimating G was best in the eucalyptus forests. The SEq ensures the definite mathematical relationship among the estimates of forest attributes is maintained, which is consistent with forest measurement principles and therefore facilitates forest resource management applications, which is an issue that needs to be addressed for airborne LIDAR forest parameter estimation.

Toward the Genetic Improvement of Drought Tolerance in Conifers: An Integrated Approach

The constant rise in the global temperature and unpredictable shifts in precipitation patterns are two of the main effects of climate change. Therefore, there is an increasing amount of interest in the identification of tree species, provenances and genotypes capable of withstanding more arid conditions and tolerating drought stress. In the present review, we focus our attention on generally more susceptible conifers and describe the different strategies that plants adopt to respond to drought stress. We describe the main approaches taken in studies of conifer adaptations to low water availability, the advantages and limitations of each, and the main results obtained with each of these approaches in the recent years. Then we discuss how the increasing amount of morphological, physiological and genetic data may find practical applications in forest management, and in particular in next-generation breeding programs. Finally, we provide some recommendations for future research. In particular, we suggest extending future studies to a broader selection of species and genera, increasing the number of studies on adult plants, in particular those on gene expression, and distinguishing between the different types of drought stress that a tree can withstand during its life cycle. The integration of data coming from different disciplines and approaches will be a key factor to increasing our knowledge about a trait as complex as drought resistance.

Managing forest carbon and landscape capacities

Widespread impacts of a warming planet are fuelling climate change mitigation efforts world-wide. Decision makers are turning to forests, the largest terrestrial primary producer, as a nature-based contribution to mitigation efforts. Resource-based economies, however, have yet to include carbon (C) in their resource planning, slowing the implementation of these important measures for atmospheric greenhouse gas reduction. The realisation of forest mitigation potential depends greatly on our ability to integrate C-sequestration practices in our forest management applications. This requires robust C-estimates, an understanding of the natural potential for a specific landscape to sequester C, the current state of the landscape relative to this potential, and the evaluation of management practices as a tool to sequester forest C in the midst of all the other values forests offer humans. Discrepancies between models used in management decisions and C estimation are the first hurdle impeding the application of forest-based mitigation strategies. Here, we combine forest disturbance and management models with a well-established C model on an open-source simulation platform. We then use the modelling system to produce C estimates of the natural C-holding capacity (potential) and two management scenarios for a study area in BC, Canada. Our simulations provide an essential metric if forests are to be managed for C-sequestration: the natural landscape C-holding capacity. Our simulations also point to a decreasing trend in simulated C on the study area over time and to a bias of the current C-levels compared to the landscape C-holding capacity (477 vs 405.5 MtC). Our explanations for this bias may provide an avenue for improved current C-state estimates. We provide a framework and the information needed for the implementation of nature-based solutions using forests for climate change mitigation. This study is a step towards modelling systems that can unify scientifically based forest management and informed C-management.

Dynamique spatio-temporelle de la forêt de mangrove dans la province de Trat en Thaïlande

In the United Nations 2021–2030 ecosystem restoration programme, coastal ecosystems such as mangroves are listed as a priority for biodiversity restoration. Therefore, understanding mangrove species diversity and changes over time are essential to predict ecosystem health, viability and resilience to changing climatic and human pressures. However, when considering future conservation ambitions and policies for mangroves, it is also crucial to understand the effects of conservation interventions. To address these concerns, we needed to develop reliable inventory methods for mangrove forests, spatialised predictions of biodiversity and good practices for using Earth observation data. In this study, we investigated the gaps in knowledge concerning the spatial organisation, intertidal zones and the recent history of mangroves in Thailand's Trat province. We investigated the impacts on forest diversification of environmental parameters, such as topography, and of human interventions such as stand rehabilitation or plantations. We were able to integrate historical multi-satellite data, current ecological data and micro-topographic measurements to establish the status and describe the spatial organisation of the mangrove forests in the Province of Trat. Using the method described in this study, we were able to overcome the technical limitations of monitoring protocols and thus develop a powerful decision-support system to assess the recovery period of mangrove forests, their structural growth and the species composition of plantations and natural native stands over three decades. Our study also identifies the main influencing factors that compromise the quality of Earth observation data, and proposes specific best practices for monitoring mangrove ecosystems. In addition, we developed the ARMA tool (Automatic Regrowth Monitoring Algorithm) and summarized functional indicators (secondary succession) by type of stand. ARMA can identify the years of planting, recovery period, age and structural development of rehabilitated mangroves compared to adjacent natural and naturally regenerating mangroves. We believe that our study makes a significant contribution to research on mangrove biodiversity, as it has several potential applications for forest restoration planning and management. It can therefore be a useful tool to measure and assess biodiversity and thereby improve ecosystem-based mangrove forest management.

Studies of plant hydraulics and water relations in Mexican environments: adaptation, physiology, and applications

Given the outstanding global progress of research on the hydraulic pathway in plants, and its important role as an indicator of plant mortality risk, we reviewed: (1) the adaptive basis of hydraulic traits and their importance for overall plant function; (2) the number of primary scientific articles on plant hydraulics that have been produced in Mexico in the last 40 years, (3) research related to specific environments in Mexico, and (4) the possible applications of plant hydraulics to natural resource management. Our systematic review included 83 articles. The number of publications per year steadily increased over time, reaching its maximum in 2021. Veracruz and Yucatán are the states where the majority of scientific articles on plant hydraulics have been produced, but for most states less than two publications on this subject appeared in ca. 40 years, and none was found for Oaxaca and Chiapas, the most biodiverse states. In plant hydraulics, the most studied biome in Mexico was the tropical dry forest, followed by crops; trees were the most studied growth-form, followed by herbaceous crops and epiphytes. We point to the need of enhancing research in the interface between plant hydraulic function and remote sensing, as well as developing applications in adaptive forest management and ecological restoration. We hope that this review may ignite a national collaborative effort to quantify critical traits that could inform the hydraulic functioning of Mexican ecosystems, particularly in the underrepresented and highly diverse states of Mexico.

Multipurpose GIS Portal for Forest Management, Research, and Education

The main objective of this research was to develop a web-based geographic information system (GIS) based on a detailed analysis of user preferences from the perspective of forest research, management and education. An anonymous questionnaire was used to elicit user preferences for a hardware platform and evaluations of web-mapping applications, geographic data, and GIS tools. Mobile GIS was used slightly more often than desktop GIS. Web-mapping applications that provide information to the public and the present research results were rated higher than the forest management application. Orthophotos for general purposes and thematic layers such as forest stand maps, soils, protected areas, cadastre, and forest roads were preferred over highly specialized layers. Tools for data searching, map printing, measuring, and drawing on digital maps were rated higher than tools for online map editing and geographic analysis. The analysis of user preferences was used to design a new multipurpose GIS portal for the University Forest Enterprise. The GIS portal was designed with a three-tier architecture on top of the software library for managing user access, working interactively with digital maps, and managing web map applications. The web map applications focus on tools and geographic information not available elsewhere, specifically timber harvest and logistics, research plots, and hunting game management.

A Comparison of the Performances of Unmanned-Aerial-Vehicle (UAV) and Terrestrial Laser Scanning for Forest Plot Canopy Cover Estimation in Pinus massoniana Forests

Canopy cover is an important indicator and commonly used in forest management applications. Unmanned-Aerial-Vehicle (UAV)—Borne Laser Scanning (ULS) has drawn increasing attention as a new alternative source for forest field inventory due to its spatial resolution comparable to that of Terrestrial Laser Scanning (TLS). In this study, the performance of plot canopy cover estimations from ULS and TLS is investigated. The experiment was conducted in 16 plots from two Pinus massoniana forests with different stand conditions in Guangxi, China. Both the Canopy Height Model (CHM)-based and Individual Tree Delineation (ITD)-based methods were used to estimate the canopy cover. The influence of CHM pixel sizes on the estimations was also analyzed. Our results demonstrated that the accuracies of ULS (R2: 0.992–0.996, RMSE: 0.591–0.820%) were better than those of TLS (R2: 0.541–0.846, RMSE: 3.642–6.297%) when compared against the reference. The average difference between the ULS and TLS estimations was 6.91%, and the disagreement increased as the forest complexity increased. The reasonable CHM pixel sizes for the canopy cover estimations were 0.07–1.2 m for ULS and 0.07–1.5 m for TLS. This study can provide useful information for the selection of data sources and estimation methods in plot canopy cover mapping.

High-resolution analysis of observed thermal growing season variability over northern Europe

Strong historical and predicted future warming over high-latitudes prompt significant effects on agricultural and forest ecosystems. Thus, there is an urgent need for spatially-detailed information of current thermal growing season (GS) conditions and their past changes. Here, we deployed a large network of weather stations, high-resolution geospatial environmental data and semi-parametric regression to model the spatial variation in multiple GS variables (i.e. beginning, end, length, degree day sum [GDDS, base temperature + 5 °C]) and their intra-annual variability and temporal trends in respect to geographical location, topography, water and forest cover, and urban land use variables over northern Europe. Our analyses revealed substantial spatial variability in average GS conditions (1990–2019) and consistent temporal trends (1950–2019). We showed that there have been significant changes in thermal GS towards earlier beginnings (on average 15 days over the study period), increased length (23 days) and GDDS (287 °C days). By using a spatial interpolation of weather station data to a regular grid we predicted current GS conditions at high resolution (100 m × 100 m) and with high accuracy (correlation ≥ 0.92 between observed and predicted mean GS values), whereas spatial variation in temporal trends and interannual variability were more demanding to predict. The spatial variation in GS variables was mostly driven by latitudinal and elevational gradients, albeit they were constrained by local scale variables. The proximity of sea and lakes, and high forest cover suppressed temporal trends and inter-annual variability potentially indicating local climate buffering. The produced high-resolution datasets showcased the diversity in thermal GS conditions and impacts of climate change over northern Europe. They are valuable in various forest management and ecosystem applications, and in adaptation to climate change.