Light Detection And Ranging Metrics Research Articles

Developing kNN forest data imputation for Catalonia

The combined use of LiDAR (Light Detection And Ranging) scanning and field inventories can provide spatially continuous wall-to-wall information on forest characteristics. This information can be used in many ways in forest mapping, scenario analyses, and forest management planning. This study aimed to find the optimal way to obtain continuous forest data for Catalonia when using kNN imputation (kNN stands for “k nearest neighbors”). In this method, data are imputed to a certain location from k field-measured sample plots, which are the most similar to the location in terms of LiDAR metrics and topographic variables. Weighted multidimensional Euclidean distance was used as the similarity measure. The study tested two different methods to optimize the distance measure. The first method optimized, in the first step, the set of LiDAR and topographic variables used in the measure, as well as the transformations of these variables. The weights of the selected variables were optimized in the second step. The other method optimized the variable set as well as their transformations and weights in one single step. The two-step method that first finds the variables and their transformations and subsequently optimizes their weights resulted in the best imputation results. In the study area, the use of three to five nearest neighbors was recommended. Altitude and latitude turned out to be the most important variables when assessing the similarity of two locations of Catalan forests in the context of kNN data imputation. The optimal distance measure always included both LiDAR metrics and topographic variables. The study showed that the optimal similarity measure may be different for different regions. Therefore, it was suggested that kNN data imputation should always be started with the optimization of the measure that is used to select the k nearest neighbors.

Driving Factors and Spatial Distribution of Aboveground Biomass in the Managed Forest in the Terai Region of Nepal

Above-ground biomass (AGB) is affected by numerous factors, including topography, climate, land use, or tree/forest attributes. Investigating the distribution and driving factors of AGB within the managed forests in Nepal is crucial for developing effective strategies for climate change mitigation, and sustainable forest management and conservation. A total of 110 field plots (circular 0.02 ha plots with a 9 m radius), and airborne laser scanning (ALS)-light detection and ranging (LiDAR) data were collected in 2021. The random forest (RF) model was employed to predict the AGB at a 30 m × 30 m resolution based on 32 LiDAR metrics derived from ALS returns. The study assessed the relationships between the AGB distribution and nine independent variables using statistical techniques like the random forest model and partial dependence plots. Results showed that the mean value of the estimated AGB was 120 tons/ha, ranging from 0 to 446.42 tons/ha. AGB showed higher values in the northeast and southeast regions, gradually decreasing towards the northwest. Land use land cover, mean annual temperature, and mean annual precipitation were identified as the primary factors influencing the variability in AGB distribution, accounting for 64% of the variability. Elevation, slope, and distance from rivers were positively correlated with AGB, while proximity to roads had a negative correlation. The increase in precipitation and temperature contributed to the initial rise in AGB, but beyond a certain lag, these variables led to a decline in AGB. This study showed the efficiency of the random forest model and partial dependence plots in examining the relationship between the AGB and its driving factors within managed forests. The study highlights the importance of understanding the AGB driving factors and utilizing LiDAR data for informed decisions regarding the region’s sustainable forest management and climate change mitigation efforts.

Estimation of Above-Ground Forest Biomass in Nepal by the Use of Airborne LiDAR, and Forest Inventory Data

Forests play a significant role in sequestering carbon and regulating the global carbon and energy cycles. Accurately estimating forest biomass is crucial for understanding carbon stock and sequestration, forest degradation, and climate change mitigation. This study was conducted to estimate above-ground biomass (AGB) and compare the accuracy of the AGB estimating models using LiDAR (light detection and ranging) data and forest inventory data in the central Terai region of Nepal. Airborne LiDAR data were collected in 2021 and made available by Nepal Ban Nigam Limited, Government of Nepal. Thirty-two metrics derived from the laser-scanned LiDAR point cloud data were used as predictor variables (independent variables), while the AGB calculated from field data at the plot level served as the response variable (dependent variable). The predictor variables in this study were LiDAR-based height and canopy metrics. Two statistical methods, the stepwise linear regression (LR) and the random forest (RF) models, were used to estimate forest AGB. The output was an accurate map of AGB for each model. The RF method demonstrated better precision compared to the stepwise LR model, as the R2 metric increased from 0.65 to 0.85, while the RMSE values decreased correspondingly from 105.88 to 60.9 ton/ha. The estimated AGB density varies from 0 to 446 ton/ha among the sample plots. This study revealed that the height-based LiDAR metrics, such as height percentile or maximum height, can accurately and precisely predict AGB quantities in tropical forests. Consequently, we confidently assert that substantial potential exists to monitor AGB levels in forests effectively by employing airborne LiDAR technology in combination with field inventory data.

Individual Tree-Level Monitoring of Pest Infestation Combining Airborne Thermal Imagery and Light Detection and Ranging

The infestation of pine shoot beetles (Tomicus spp.) in the forests of Southwestern China has inflicted serious ecological damages to the environment, causing significant economic losses. Therefore, accurate and practical approaches to detect pest infestation have become an urgent necessity to mitigate these harmful consequences. In this study, we explored the efficiency of thermal infrared (TIR) technology in capturing changes in canopy surface temperature (CST) and monitoring forest health at the scale of individual tree crowns. We combined data collected from TIR imagery and light detection and ranging (LiDAR) using unmanned airborne vehicles (UAVs) to estimate the shoot damage ratio (SDR), which is a representative parameter of the damage degree caused by forest infestation. We compared multiple machine learning methods for data analysis, including random forest (RF), partial least squares regression (PLSR), and support vector machine (SVM), to determine the optimal regression model for assessing SDR at the crown scale. Our findings showed that a combination of LiDAR metrics and CST presents the highest accuracy in estimating SDR using the RF model (R2 = 0.7914, RMSE = 15.5685). Our method enables the accurate remote monitoring of forest health and is expected to provide a novel approach for controlling pest infestation, minimizing the associated damages caused.

Stratification of vertical canopy structure to improve estimation of forest inventory attributes using airborne LiDAR data in a large subtropical region of China

Over the last two decades, airborne light detection and ranging (LiDAR) has been developed into an advanced tool for practical forest resource inventory monitoring over large areas. Nonetheless, improving the accuracy of forest inventory attribute estimations remains an ongoing challenge. This paper introduces a novel framework for estimating forest inventory attributes based on the stratification of vertical forest structures (VFS). According to the composition and spatial arrangement of the superior, middle, and inferior strata in the tree layer, the forest stand was classified into six distinct VFS classes. Subsequently, the multiplicative power models were established for the stratification-based estimations of the forest inventory attributes, including stand volume (VOL), basal area (BA), and above-ground biomass (AGB), by using a rule-based exhaustive combination approach, and their performances were comparatively analyzed. The result indicated that: compared to the accuracy (relative root mean squared error, rRMSE) of the species-based estimation, the weighted average rRMSE of stratumbased VOL, BA, and AGB estimations of four forest types (Chinese fir, Masson pine, eucalyptus, and broadleaf forests) decreased by 0.3%–7.3%, +3.6%–9.4%, and 0.7%–8.7%, respectively, and the accuracy was significantly improved after stratification. Even after clustering the VFS into two or three classes using cluster analysis, the accuracy of forest attribute estimations remained superior to that of the species-based estimations. Notably, the coefficients of variation for both forest attributes and LiDAR metrics experienced a substantial decrease, and their statistical relationship considerably strengthened within most strata post-VFS classification, which led to an improvement in the accuracy of the forest attribute estimations. The methodology presented in this paper provides a significant advance in improving the accuracy of forest inventory attributes for large areas using airborne LiDAR data.

Performance, effectiveness and computational efficiency of powerline extraction methods for quantifying ecosystem structure from light detection and ranging

ABSTRACT National and regional data products of the ecosystem structure derived from airborne laser scanning (ALS) surveys with Light Detection And Ranging (LiDAR) technology are essential for ecology, biodiversity, and ecosystem monitoring. However, noises like powerlines often remain, hindering the accurate measurement of 3D ecosystem structures from LiDAR. Currently, there is a lack of studies assessing powerline noise removal in the context of generating data products of ecosystem structures from ALS point clouds. Here, we assessed the (1) performance and accuracy, (2) effectiveness, and (3) time efficiency and execution time of three powerline extraction methods (i.e. two point-based methods based on deep learning and eigenvalue decomposition, respectively, and one hybrid method) for removing powerline noise when quantifying 3D ecosystem structures in landscapes with varying canopy heights and vegetation openness. Twenty-five LiDAR metrics representing three key dimensions of the ecosystem structure (i.e. vegetation height, cover, and vertical variability) across 10 study areas in the Netherlands were used for our assessment. The deep learning method had the best performance and showed the highest accuracy of powerline removal across various landscape types (average F 1 score = 96%), closely followed by the hybrid method (average F 1 score = 95%). In contrast, the accuracy of the eigenvalue decomposition method was lower (average F 1 score = 82%) and depended on landscape context and vegetation composition (e.g. the F 1 score decreased from 96% to 63% when the average canopy height increased across landscapes). Powerline noise removal had the highest effectiveness (i.e. generating LiDAR metrics closest to those derived from manually labeled ground truth data) for LiDAR metrics capturing height and cover of low- and high-vegetation layers. Time efficiency (processed points per second) was highest for the eigenvalue decomposition method, yet the hybrid method reduced the execution time by > 50% compared to the deep learning method (ranging from 20% to 89% in study areas with different landscape composition). Based on our findings, we recommend the hybrid method for upscaling powerline removal on multi-terabyte ALS datasets to a regional or national extent because of its high accuracy and computational efficiency. Remaining misclassifications in LiDAR metrics could be further minimized by improving the training dataset for deep learning models (e.g. including various shapes of transmission towers from different datasets). Our findings provide novel insights into the performance of different powerline extraction methods, how their effectiveness varies for improving vegetation metrics and mapping the 3D ecosystem structure from LiDAR, and their computational efficiency for upscaling powerline removal in multi-terabyte ALS datasets to a national extent.

Assessment of Oak Groves Conservation Statuses in Natura 2000 Sacs with Single Photon Lidar and Sentinel-2 Data

Among the main objectives of Natura 2000 Network sites management plans is monitoring their conservation status under a reasonable cost and with high temporal frequency. The aim of this study is to assess the ability of single-photon light detection and ranging (LiDAR) technology (14 points per m2) and Sentinel-2 data to classify the conservation status of oak forests in four special areas of conservation in Navarra Province (Spain) that comprise three habitats. To capture the variability of conservation status within the three habitats, we first performed a random stratified sampling based on conservation status measured in the field, canopy cover, and terrain slope and height. Thereafter, we compared two metric selection approaches, namely Kruskal–Wallis and Dunn tests, and two machine learning classification methods, random forest (RF) and support vector machine (SVM), to classify the conservation statuses using LiDAR and Sentinel-2 data. The best-fit classification model, which included only LiDAR metrics, was obtained using the random forest method, with an overall classification accuracy after validation of 83.01%, 75.51%, and 88.25% for Quercus robur (9160), Quercus pyrenaica (9230), and Quercus faginea (9240) habitats, respectively. The models include three to six LiDAR metrics, with the structural diversity indices (LiDAR height evenness index, LHEI, and LiDAR height diversity index, LHDI) and canopy cover (FCC) being the most relevant ones. The inclusion of the NDVI index from the Sentinel-2 image did not improve the classification accuracy significantly. This approach demonstrates its value for classifying and subsequently mapping conservation statuses in oak groves and other Natura 2000 Network habitat sites at a regional scale, which could serve for more effective monitoring and management of high biodiversity habitats.

Fusion of LiDAR and Multispectral Data for Aboveground Biomass Estimation in Mountain Grassland

Grassland aboveground biomass (AGB) is an important indicator for studying the change in grassland ecological quality and carbon cycle. The rapid development of high-resolution remote sensing and unmanned aerial vehicles (UAV) provides a new opportunity for accurate estimation of grassland AGB on the plot scale. In this study, the mountain grassland was taken as the research object. Using UAV Light Detection and Ranging (LiDAR) data and multispectral satellite images, the influence of topographic correction methods on AGB estimation was compared and a series of LiDAR metrics and vegetation indices were extracted. On this basis, a comprehensive indicator, the vegetation index-height-intensity model (VHI), was proposed to estimate AGB quickly. The results show that: (1) Among the four topographic correction methods, the Teillet regression has the best effect, and can effectively improve the accuracy of AGB estimation in mountain grassland. The correlation between corrected ratio vegetation index and AGB was the highest (correlation coefficient: 0.682). (2) Among the height and intensity metrics, median height and max intensity yielded the higher accuracy in estimating AGB, with Root Mean Square Error (RMSE) of 322 g/m2 and 333 g/m2, respectively. (3) The VHI integrated spectrum and LiDAR information, and its accuracy for AGB estimation for mountain grassland, was obviously better than other indicators, with an RMSE of 272 g/m2. We also found that the accuracy of VHI in univariate models was comparable to that of complex multivariate models such as stepwise regression, support vector machine, and random forest. This study provides a new approach for estimating grassland AGB with multi-source data. As a simple and effective indicator, VHI has shown strong application potential for grassland AGB estimating in mountainous areas, and can be further applied to grassland carbon cycle research and fine management.

LiDAR Point Clouds Usage for Mapping the Vegetation Cover of the “Fryderyk” Mine Repository

The paper investigates the usage of LiDAR (light detection and ranging) data for the automation of mapping vegetation with respect to the evaluation of the ecological succession process. The study was performed for the repository of the “Fryderyk” mine (southern Poland). The post-flotation area analyzed is a unique refuge habitat—Natura2000, PLH240008—where a forest succession has occurred for several dozen years. Airborne laser scanning (ALS) point clouds were used for deriving detailed information about the morphometry of the spoil heap and about the secondary forest succession process—mainly vegetation parameters i.e., height and canopy cover. The area of the spoil heap is irregular with a flat top and steep slopes above 20°. Analyses of ALS point clouds (2011 and 2019), confirmed progression in the forest succession process, and land cover changes especially in wooded or bushed areas. Precise vegetation parameters (3D LiDAR metrics) were calculated and provided the following parameters: mean value of vegetation height as 6.84 m (2011) and 8.41 m (2019), and canopy cover as 30.0% (2011) and 42.0% (2019). Changes in vegetation volume (3D area) were shown: 2011—310,558 m3, 2019—325,266 m3, vegetation removal—85,136 m3, increasing ecological succession—99,880 m3.

Mapping Dominant Boreal Tree Species Groups by Combining Area-Based and Individual Tree Crown LiDAR Metrics with Sentinel-2 Data

Airborne light detection and ranging (LiDAR) data are increasingly used to inform sustainable forest management practices. Information about species composition is needed for a range of applications; however, commonly used area-based summaries of LiDAR data are limited to accurately differentiate tree species. The objective of this study was to map dominant species groups across a large (>580,000 ha) boreal forest by combining area-based and individual tree metrics derived from single photon LiDAR data with multispectral information derived from Sentinel-2 imagery. Classification of the forest into jack pine (Pinus banksiana), black spruce (Picea mariana), mixed conifer, mixedwood, and hardwood species groups resulted in an overall accuracy of 67.8% compared with field data. A more generalized classification into softwood, hardwood, and mixedwood had an overall accuracy of 83.2%. The reflectance in the red edge region of the electromagnetic spectrum (λ = 740 nm), the area and volume of tree crowns, and the cumulative distribution of LiDAR returns in the canopy were particularly important variables to discriminate between species groups. Wall-to-wall predictions of species groups based on remotely sensed data—as derived herein—could provide a spatially-detailed alternative to stand-level species composition information traditionally derived from photo-interpreted aerial imagery.

Requirements for Automotive LiDAR Systems.

Light detection and ranging (LiDAR) are fundamental sensors that help driving tasks for autonomous driving at various levels. Commercially available systems come in different specialized design schemes and involve plenty of specifications. In the literature, there are insufficient representations of the technical requirements for LiDAR systems in the automotive context, such as range, detection quality, resolving power, field of view, and eye safety. For this reason, the requirements above require to be derived based on ADAS functions. The requirements for various key LiDAR metrics, including detection range, field of view, angular resolution, and laser safety, are analyzed in this paper. LiDAR systems are available with various radiation patterns that significantly impact on detection range. Therefore, the detection range under various radiation patterns is firstly investigated in this paper. Based on ADAS functions, the required detection range and field of view for LiDAR systems are examined, taking into account various travel speeds to avoid collision and the coverage of the entire lane width. Furthermore, the angular resolution limits are obtained utilizing the KITTI dataset and exemplary 3D detection algorithms. Finally, the maximum detection ranges for the different radiation patterns are compared under the consideration of derived requirements and laser safety.

Sugarcane Biomass Prediction with Multi-Mode Remote Sensing Data Using Deep Archetypal Analysis and Integrated Learning

The use of multi-mode remote sensing data for biomass prediction is of potential value to aid planting management and yield maximization. In this study, an advanced biomass estimation approach for sugarcane fields is proposed based on multi-source remote sensing data. Since feature interpretability in agricultural data mining is significant, a feature extraction method of deep archetypal analysis (DAA) that has good model interpretability is introduced and aided by principal component analysis (PCA) for feature mining from the multi-mode multispectral and light detection and ranging (LiDAR) remote sensing data pertaining to sugarcane. In addition, an integrated regression model integrating random forest regression, support vector regression, K-nearest neighbor regression and deep network regression is developed after feature extraction by DAA to precisely predict biomass of sugarcane. In this study, the biomass prediction performance achieved using the proposed integrated learning approach is found to be predominantly better than that achieved by using conventional linear methods in all the time periods of plant growth. Of more significance, according to model interpretability of DAA, only a small set of informative features maintaining their physical meanings (four informative spectral indices and four key LiDAR metrics) can be extracted which eliminates the redundancy of multi-mode data and plays a vital role in accurate biomass prediction. Therefore, the findings in this study provide hands-on experience to planters with indications of the key or informative spectral or LiDAR metrics relevant to the biomass to adjust the corresponding planting management design.

Laserfarm – A high-throughput workflow for generating geospatial data products of ecosystem structure from airborne laser scanning point clouds

Quantifying ecosystem structure is of key importance for ecology, conservation, restoration, and biodiversity monitoring because the diversity, geographic distribution and abundance of animals, plants and other organisms is tightly linked to the physical structure of vegetation and associated microclimates. Light Detection And Ranging (LiDAR) — an active remote sensing technique — can provide detailed and high resolution information on ecosystem structure because the laser pulse emitted from the sensor and its subsequent return signal from the vegetation (leaves, branches, stems) delivers three-dimensional point clouds from which metrics of vegetation structure (e.g. ecosystem height, cover, and structural complexity) can be derived. However, processing 3D LiDAR point clouds into geospatial data products of ecosystem structure remains challenging across broad spatial extents due to the large volume of national or regional point cloud datasets (typically multiple terabytes consisting of hundreds of billions of points). Here, we present a high-throughput workflow called ‘Laserfarm’ enabling the efficient, scalable and distributed processing of multi-terabyte LiDAR point clouds from national and regional airborne laser scanning (ALS) surveys into geospatial data products of ecosystem structure. Laserfarm is a free and open-source, end-to-end workflow which contains modular pipelines for the re-tiling, normalization, feature extraction and rasterization of point cloud information from ALS and other LiDAR surveys. The workflow is designed with horizontal scalability and can be deployed with distributed computing on different infrastructures, e.g. a cluster of virtual machines. We demonstrate the Laserfarm workflow by processing a country-wide multi-terabyte ALS dataset of the Netherlands (covering ∼34,000 km2 with ∼700 billion points and ∼ 16 TB uncompressed LiDAR point clouds) into 25 raster layers at 10 m resolution capturing ecosystem height, cover and structural complexity at a national extent. The Laserfarm workflow, implemented in Python and available as Jupyter Notebooks, is applicable to other LiDAR datasets and enables users to execute automated pipelines for generating consistent and reproducible geospatial data products of ecosystems structure from massive amounts of LiDAR point clouds on distributed computing infrastructures, including cloud computing environments. We provide information on workflow performance (including total CPU times, total wall-time estimates and average CPU times for single files and LiDAR metrics) and discuss how the Laserfarm workflow can be scaled to other LiDAR datasets and computing environments, including remote cloud infrastructures. The Laserfarm workflow allows a broad user community to process massive amounts of LiDAR point clouds for mapping vegetation structure, e.g. for applications in ecology, biodiversity monitoring and ecosystem restoration.

Estimation of biomass burning emissions by integrating ICESat-2, Landsat 8, and Sentinel-1 data

Anthropogenic carbon emissions directly contribute to global warming, which has induced severe environmental concerns like extreme droughts and devastating fires. To evaluate the effects of fires on carbon cycling and climate change, it is crucial to accurately estimate the amount of carbon released during fires. The Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) mission offers spaceborne light detection and ranging (LiDAR) measurements of global forests, providing excellent opportunities for monitoring forest dynamics and carbon emissions. The ICESat-2 Land and vegetation height product (ATL08), Landsat 8 data, and Sentinel-1 data were integrated to estimate biomass burning emissions with a three-phase hierarchy model. In Phase 1, the least absolute shrinkage and selection operator (LASSO) regression was used to establish the relationship between ATL08 LiDAR metrics and reference aboveground biomass (AGB), R2 0.77 and RMSE 56.67 Mg ha−1. In Phase 2, the LASSO regression predicted AGB for all ATL08 segments. A Random Forest Regression model was trained with Landsat 8 reflectance data and derived vegetation indices, Sentinel-1 data, and terrain variables as predictors and with the LASSO predicted AGB as the response variable, R2 0.71 and RMSE 45.91 Mg ha−1. Wall-to-wall maps of pre-fire and post-fire AGB were produced with the Random Forest model. In Phase 3, the difference between the pre-fire and the post-fire AGB was converted to carbon emissions. The estimated biomass burning emissions of the 2018 Carr fire group, the 2018 Camp fire, and the 2019 Walker fire in California are 3.54 (± 0.067) Tg C, 1.66 (± 0.041) Tg C, and 0.45 (± 0.022) Tg C, respectively. Forests have the highest biomass burning emissions with an average emission of 30.09 Mg ha−1, approximately twice and four times the average emissions from shrubs and grasses. This study highlights the merit of integrating spaceborne remote sensing data, including LiDAR data, optical data, and radar data, for estimating biomass burning emissions, opening an avenue for accurate forest biomass and carbon dynamics monitoring, as well as climate change mitigation.

Estimation of volume resources for planted forests using an advanced LiDAR and hyperspectral remote sensing

Planted forests play a key role in alleviating the stress of harvesting woods and carbon emission on natural forests. Accurate estimation of forest volume resources in planted forests is crucial for balancing relationships between ecological benefits from timber production and ecological benefits from biomass energy. In this study, we explored the use of spectral indices (SIs) and wavelet features (WFs) derived from hyperspectral imaging (HSI) data, as well as Light Detection And Ranging (LiDAR) metrics, with three multivariate regression methods for estimating volume resources in subtropical forests. The results showed that using combined LiDAR and HSI metrics generally outperformed LiDAR-only models. HSI metrics exhibited stronger relationships with volume in lower volume plots (volume < 320 m3/ha) but had saturation problem in higher volume plots, which could be well alleviated by LiDAR metrics. Specifically, coupling WFs and SIs as HSI metrics can further reinforce the synergetic use of LiDAR and HSI metrics, especially with the approach of backward elimination. In comparing three regression methods, Gaussian Processes Regression models (CV-R2 = 0.87, rRMSE = 16.12% for the best model) mostly outperformed Partial Linear Squares Regression and Random Forest regression models. The common important LiDAR metrics for three regression methods were height-related (H50) and density-related (D3 and D9) metrics, while the common optimal spectral metrics (WF615, 2, WF685, 5 and REIP) were related to the chlorophyll absorption features. These findings demonstrated the significance of the wavelet approach in strengthening the synergetic use of LiDAR and HSI metrics for enhancing forest parameter estimations.

Forest canopy stratification based on fused, imbalanced and collinear LiDAR and Sentinel-2 metrics

Knowledge about the forest canopy stratification is of essential importance for forest management and planning. Collecting structural information (e.g. natural regeneration) still depends on cost and labour intensive forest inventories with a coarse spatio-temporal resolution. Remote sensing partly overcomes these limitations and particularly active sensors of type light detection and ranging (LiDAR) have proven their great potential of separating forest strata. The applicability of LiDAR metrics for the differentiation of the spruce dominated forest strata in Central Germany has not been tested yet. Additionally, studying the potential of Sentinel-2 metrics for the classification of forest strata is lacking too.In this study, we investigated the capabilities of six different classification approaches for the differentiation of five forest strata that are typical for the study region. Reference data were derived from forest inventory measurements surveyed on a dense 200 × 200 m grid. The six classification approaches were trained with fused and un-fused LiDAR and Sentinel-2 inferred metrics. The classification results were compared using the overall mean accuracy, sensitivity and specificity via receivers operating characteristics of multi-class problems. We were interested in the classification abilities of Sentinel-2 metrics due to the obvious advantages of Sentinel-2 based metrics (free of charge, high spatio-temporal coverage). We assumed that the canopy structure determines the reflection on stand level and thus might facilitate the classification of different canopy strata. Beforehand, it was important to examine the influence of distinctly imbalanced and collinear reference data on the classification results.We found that the Random Forest classifier most accurately separated the five forest strata with a mean overall accuracy of 83.3% (Kappa = 76.2%). These values were achieved from balanced training data and the classification capability was confirmed by classification results from an independent test data set. Fused predictors of active (LiDAR) and passive (Sentinel-2) remote sensing revealed no substantial improvement in the classification accuracy due to the dominant role of LiDAR metrics. Herein, we identified that especially the height variability, top height, portion of LiDAR-returns between 2 m and 10 m and the standard deviation of the return number between the 25th and 50th height percentile, predominately contributed to the classification accuracy. Classification results purely based on Sentinel-2 metrics revealed a rather small overall mean accuracy of 54.7%. The metrics (e.g. median, variance, entropy) were derived from Sentinel-2 indices, covering the visible and near to short infrared spectrum. Variable importance computations unraveled a detectable but minor contribution of MSI, TCG, NDVI to the classification result. Finally, our data driven observations illustrated serious drawbacks associated to data imbalance, collinearity and autocorrelation and presented practical guidance to cope with these issues.

LiDAR metrics predict suitable forest foraging areas of endangered Mouse-eared bats (Myotis myotis)

Habitat shift caused by human impact on vegetation structure poses a great threat to species which are specialized on unique habitats. Single layered beech forests, the main foraging habitat of Greater Mouse-eared Bats (Myotis myotis), are threatened by recent changes in forest structure. After this species suffered considerable population losses until the 1970s, their roosts in buildings are strictly protected. However, some populations are still declining. Thus, the spatial identification of suitable foraging habitat would be essential to ensure conservation policy. The aim of this study was (a) to verify the relevance of forest structural variables for the activity of M. myotis and (b) to evaluate the potential of LiDAR (Light Detection and Ranging) in predicting suitable foraging habitat of the species. We systematically sampled bat activity in forests close to 18 maternity roosts in Switzerland and applied a generalized linear mixed model (GLMM) to fit the activity data to forest structure variables recorded in the field and derived from LiDAR. We found that suitable forest foraging habitat is defined by single layered forest, dense canopy, no shrub layer and a free flight space. Most importantly, this key foraging habitat can be well predicted by airborne LiDAR data. This allows for the first time to create nationwide prediction maps of potential foraging habitats of this species to inform conservation management. This method has a special significance for endangered species with large spatial use, whose key resources are hard to identify and widely distributed across the landscape.

Pre-fire aboveground biomass, estimated from LiDAR, spectral and field inventory data, as a major driver of burn severity in maritime pine (Pinus pinaster) ecosystems

The characterization of surface and canopy fuel loadings in fire-prone pine ecosystems is critical for understanding fire behavior and anticipating the most harmful ecological effects of fire. Nevertheless, the joint consideration of both overstory and understory strata in burn severity assessments is often dismissed. The aim of this work was to assess the role of total, overstory and understory pre-fire aboveground biomass (AGB), estimated by means of airborne Light Detection and Ranging (LiDAR) and Landsat data, as drivers of burn severity in a megafire occurred in a pine ecosystem dominated by Pinus pinaster Ait. in the western Mediterranean Basin. Total and overstory AGB were more accurately estimated ( R 2 equal to 0.72 and 0.68, respectively) from LiDAR and spectral data than understory AGB ( R 2 ​= ​0.26). Density and height percentile LiDAR metrics for several strata were found to be important predictors of AGB. Burn severity responded markedly and non-linearly to total ( R 2 ​= ​0.60) and overstory ( R 2 ​= ​0.53) AGB, whereas the relationship with understory AGB was weaker ( R 2 ​= ​0.21). Nevertheless, the overstory plus understory AGB contribution led to the highest ability to predict burn severity (RMSE ​= ​122.46 in dNBR scale), instead of the joint consideration as total AGB (RMSE ​= ​158.41). This study novelty evaluated the potential of pre-fire AGB, as a vegetation biophysical property derived from LiDAR, spectral and field plot inventory data, for predicting burn severity, separating the contribution of the fuel loads in the understory and overstory strata in Pinus pinaster stands. The evidenced relationships between burn severity and pre-fire AGB distribution in Pinus pinaster stands would allow the implementation of threshold criteria to support decision making in fuel treatments designed to minimize crown fire hazard.

Identification of Significative LiDAR Metrics and Comparison of Machine Learning Approaches for Estimating Stand and Diversity Variables in Heterogeneous Brazilian Atlantic Forest

Data collection and estimation of variables that describe the structure of tropical forests, diversity, and richness of tree species are challenging tasks. Light detection and ranging (LiDAR) is a powerful technique due to its ability to penetrate small openings and cracks in the forest canopy, enabling the collection of structural information in complex forests. Our objective was to identify the most significant LiDAR metrics and machine learning techniques to estimate the stand and diversity variables in a disturbed heterogeneous tropical forest. Data were collected in a remnant of the Brazilian Atlantic Forest with different successional stages. LiDAR metrics were used in three types of transformation: (i) raw data (untransformed), (ii) correlation analysis, and (iii) principal component analysis (PCA). These transformations were tested with four machine learning techniques: (i) artificial neural network (ANN), ordinary least squares (OLS), random forests (RF), and support vector machine (SVM) with different configurations resulting in 27 combinations. The best technique was determined based on the lowest RMSE (%) and corrected Akaike information criterion (AICc), and bias (%) values close to zero. The output forest variables were mean diameter at breast height (MDBH), quadratic mean diameter (QMD), basal area (BA), density (DEN), number of tree species (NTS), as well as Shannon–Waver (H’) and Simpson’s diversity indices (D). The best input data were the new variables obtained from the PCA, and the best modeling method was ANN with two hidden layers for the variables MDBH, QMD, BA, and DEN while for NTS, H’and D, the ANN with three hidden layers were the best methods. For MDBH, QMD, H’and D, the RMSE was 5.2–10% with a bias between −1.7% and 3.6%. The BA, DEN, and NTS were the most difficult variables to estimate, due to their complexity in tropical forests; the RMSE was 16.2–27.6% and the bias between −12.4% and −0.24%. The results showed that it is possible to estimate the stand and diversity variables in heterogeneous forests with LiDAR data.

A machine learning algorithm to detect pine wilt disease using UAV-based hyperspectral imagery and LiDAR data at the tree level

Pine wilt disease (PWD) is a global destructive threat to forests, having caused extreme damage in China. Therefore, the establishment of an effective method to accurately monitor and map the infection stage by PWD is imperative. Unmanned aerial vehicle (UAV)-based hyperspectral imaging (HI) and light detection and ranging (LiDAR) technique is an effective approach for forest health monitoring. However, few previous studies have used airborne HI and LiDAR to detect PWD and compared the capability for predicting PWD infection stage at the tree level. In this paper, PWD infection was divided into five stages (green, early, middle, heavy, and grey), and HI and LiDAR data were integrated to detect PWD. We estimated the power of the hyperspectral method (HI data only), LiDAR (LiDAR data only), and their combination (HI plus LiDAR data) to predict the infection stages of PWD using the random forest (RF) algorithm. We obtained the following results: (1) The classification accuracies of HI (OA: 66.86%, Kappa: 0.57) were higher than those of LiDAR (OA: 45.56%, Kappa: 0.27) for predicting PWD infection stages, and their combination had the best accuracies (OA: 73.96%, Kappa: 0.66); (2) LiDAR data had higher ability for dead tree identification than HI data; and (3) The combined use of HI and LiDAR data for estimation of PWD infection stages showed that LiDAR metrics (e.g., crown volume) were essential in the classification model, although the variables derived from HI data contributed more than those extracted from LiDAR. Therefore, we proposed a new approach combining the merits of HI and LiDAR data to precisely predict PWD infection stages at the tree level, allowing better PWD monitoring and control. The approach could also be employed for mapping and monitoring other forest disturbance issues.