The reliability of aircraft control surfaces, constructed from thermoplastic materials, can be affected by impacts from airborne particles. Recognizing the exact position of such impacts is essential for correctly estimating the resulting damage. This research intended to address the issue by introducing an innovative structural health monitoring solution capable of autonomously detecting and localizing impacts using acoustic emission monitoring. The objective of this research is to investigate the application of AE for the localization of impacts on aircraft elevators using machine learning techniques, specifically regression algorithms. To achieve this goal, two algorithms, linear regression, and random forest, were employed for predicting the impact locations based on AE signals. The performance of each algorithm was validated on a thermoplastic composite aircraft elevator. Results indicated that both linear regression and random forest models show high accuracy in predicting the impact locations. The random forest model, with an R² value of 0.98616 and an RMSE of 0.6778, outperformed the linear regression model, which exhibited an R² value of 0.9361 and an RMSE of 1.4614.

IntroductionRubber trees are an important cash crop in Hainan Province; thus, monitoring sample plots of these trees provides important data for determining growth conditions. However, existing monitoring technology and rubber forest sample plot analysis methods are relatively simple and present widespread issues, such as limited monitoring equipment, transportation difficulties, and relatively poor three-dimensional visualization effects in complex environments. These limitations have complicated the development of rubber forest sample plot monitoring.MethodThis study developed a terrestrial photogrammetry system combined with 3D point-cloud reconstruction technology based on the structure from motion with multi-view stereo method and sample plot survey data. Deviation analyses and accuracy evaluations of sample plot information were performed in the study area for trees to explore the practical significance of this method for monitoring rubber forest sample plots. Furthermore, the relationship between the height of the first branch, diameter at breast height (DBH), and rubber tree volume was explored, and a rubber tree standard volume model was established.ResultsThe Bias, relative Bias, RMSE, and RRMSE of the height of the first branch measured by this method were −0.018 m, −0.371%, 0.562 m, and 11.573%, respectively. The Bias, relative Bias, RMSE, and RRMSE of DBH were −0.484 cm, −1.943%, −2.454 cm, and 9.859%, respectively, which proved that the method had high monitoring accuracy and met the monitoring requirements of rubber forest sample plots. The fitting results of rubber tree standard volume model had an R2 value of 0.541, and the estimated values of each parameter were 1.745, 0.115, and 0.714. The standard volume model accurately estimated the volume of rubber trees and forests using the first branch height and DBH.DiscussionThis study proposed an innovative planning scheme for a terrestrial photogrammetry system for 3D visual monitoring of rubber tree forests, thus providing a novel solution to issues observed in current sample plot monitoring practices. In the future, the application of terrestrial photogrammetry systems to monitor other types of forests will be explored.

International audience | Forest biomass plays a key role in our climate and constitutes a major component of the Earth carbon cycle. However, forests are facing increasingly challenging environmental conditions induced by climate change globally. In this context, monitoring the temporal evolution of the biomass on a global scale is fundamental. Ground-sampling trees regularly in global-scale is very difficult, not only because of the massive resource and time it requires, but also because of the inaccessibility of most forested areas. Space borne instruments can overcome some of these difficulties and are able to probe most of the Earth regularly. Each type of mission comes with its advantages and drawbacks. Optical instruments can map the land surfaces in great details but are heavily affected by atmospheric (clouds, haze, aerosols) and light conditions. Radars and passive microwave radiometers data often has relatively lower spatial-resolution, but they come with all-weather capabilities so their temporal-resolution is very high. In addition, their signal is able to penetrate the vegetation layer so they have particular advantages in soil moisture and biomass monitoring. To combine the strength of both optical and microwave instruments, time series of various indices, derived from both sensors exist and characterize different components of trees (water content, leaf, ...). To further the study of forest monitoring, ESA is planning to launch BIOMASS mission in 2024. It is the first P-band (435 MHz) radar in space that is dedicated to estimating forest biomass. The ESA OSMOSE project therefore aims to produce harmonized optical and passive microwave data cubes spanning more than 20 years. It also aims to create innovative algorithms to reprocess the existing indices and output a dataset dedicated to vegetation monitoring. By using complementary information from optical indices (LAI, FAPAR, NDVI) and biomass estimates derived from the passive microwave radiometers’ observations by the AMSRE/2 and SMOS missions, OSMOSE aims to provide a comprehensive dataset to apprehend in details the global evolution of forest biomass. It includes the development of new methodologies to derive the vegetation optical depth (VOD) which will be further processed to derive forest biomass. In parallel OSMOSE is also developing a method to estimate forest biomass directly from passive microwave brightness temperatures as well as canopy water content from optical sensors.After one year of study, we will present the OSMOSE project and the first associated initial results (e.g. new product datacubes). The next step will consist in evaluating the syngery of all these products with the ultimate goal of closely monitoring the status and evolution of forest biomass over the period 2002 - present.

International audience | Forest landscape restoration (FLR) commitments have been established in the past years to restore over 200 million hectares, as part of the global Bonn Challenge goal, mostly through the implementation of several different restorative practices in degraded lands, ranging from com-mercial tree monocultures to restoration plantings. The potential of such contrasting restorative practices to support biodiversity conservation and ecosystem services provision vary over space and time, making the monitoring of FLR programs an emerging challenge. Remote sensing techniques, together with innovative technologies for data acquisition, treatment, and analysis have proven to be strategic for planning and monitoring FLR, yet there are still important un-resolved questions. Here, we evaluated the potential of multispectral orbital images of the high spatial (5 m) and spectral (12 bands) resolution VEN mu S microsatellite, joint project of the Israeli Space Agency and CNES, to classify the spectral behavior and diversity of six tree cover classes (savanna woodlands, old-and second-growth semi-deciduous forests, young restoration plant-ings, and eucalyptus and pine tree monocultures) commonly found in FLR programs in tropical regions. We assessed how these tree cover classes located in a study landscape in southeastern Brazil differ according to their spectral response (winter and summer bands, and vegetation indices), canopy variability (textural features), seasonal behavior (delta layers -difference be-tween summer and winter vegetation indexes), and spectral diversity, and used these attributes as variables to the model. We used the Random Forest algorithm to generate the models and evaluate how the tree cover classes differ in the classification and how the variables supported the model. We achieved high values of global accuracy (91.9%) and "F1 score" (above 0.8) for all tree cover classes, in which second-growth forest presented the lowest accuracy. The textural layers, delta layers, and the spectral diversity layers were the most important attributes to discriminate among tree cover classes. We demonstrate here the potential of using VEN mu S or similar sensor images together with different image processing and machine learning algorithms to monitor FLR programs, allowing further remote sensing approaches and in-deep field assessments to advance evaluation of FLR benefits for nature and people. We demonstrated how the fusion of all these types of data and innovative approaches to data processing, can result in novel ways to assess restoration performance and open new avenues to upscale monitoring, bridging the gap between FLR expectations and achieved goals.

International audience | Forest landscape restoration (FLR) commitments have been established in the past years to restore over 200 million hectares, as part of the global Bonn Challenge goal, mostly through the implementation of several different restorative practices in degraded lands, ranging from com-mercial tree monocultures to restoration plantings. The potential of such contrasting restorative practices to support biodiversity conservation and ecosystem services provision vary over space and time, making the monitoring of FLR programs an emerging challenge. Remote sensing techniques, together with innovative technologies for data acquisition, treatment, and analysis have proven to be strategic for planning and monitoring FLR, yet there are still important un-resolved questions. Here, we evaluated the potential of multispectral orbital images of the high spatial (5 m) and spectral (12 bands) resolution VEN mu S microsatellite, joint project of the Israeli Space Agency and CNES, to classify the spectral behavior and diversity of six tree cover classes (savanna woodlands, old-and second-growth semi-deciduous forests, young restoration plant-ings, and eucalyptus and pine tree monocultures) commonly found in FLR programs in tropical regions. We assessed how these tree cover classes located in a study landscape in southeastern Brazil differ according to their spectral response (winter and summer bands, and vegetation indices), canopy variability (textural features), seasonal behavior (delta layers -difference be-tween summer and winter vegetation indexes), and spectral diversity, and used these attributes as variables to the model. We used the Random Forest algorithm to generate the models and evaluate how the tree cover classes differ in the classification and how the variables supported the model. We achieved high values of global accuracy (91.9%) and "F1 score" (above 0.8) for all tree cover classes, in which second-growth forest presented the lowest accuracy. The textural layers, delta layers, and the spectral diversity layers were the most important attributes to discriminate among tree cover classes. We demonstrate here the potential of using VEN mu S or similar sensor images together with different image processing and machine learning algorithms to monitor FLR programs, allowing further remote sensing approaches and in-deep field assessments to advance evaluation of FLR benefits for nature and people. We demonstrated how the fusion of all these types of data and innovative approaches to data processing, can result in novel ways to assess restoration performance and open new avenues to upscale monitoring, bridging the gap between FLR expectations and achieved goals.

Forest landscape restoration (FLR) commitments have been established in the past years to restore over 200 million hectares, as part of the global Bonn Challenge goal, mostly through the implementation of several different restorative practices in degraded lands, ranging from commercial tree monocultures to restoration plantings. The potential of such contrasting restorative practices to support biodiversity conservation and ecosystem services provision vary over space and time, making the monitoring of FLR programs an emerging challenge. Remote sensing techniques, together with innovative technologies for data acquisition, treatment, and analysis have proven to be strategic for planning and monitoring FLR, yet there are still important unresolved questions. Here, we evaluated the potential of multispectral orbital images of the high spatial (5 m) and spectral (12 bands) resolution VENμS microsatellite, joint project of the Israeli Space Agency and CNES, to classify the spectral behavior and diversity of six tree cover classes (savanna woodlands, old- and second-growth semi-deciduous forests, young restoration plantings, and eucalyptus and pine tree monocultures) commonly found in FLR programs in tropical regions. We assessed how these tree cover classes located in a study landscape in southeastern Brazil differ according to their spectral response (winter and summer bands, and vegetation indices), canopy variability (textural features), seasonal behavior (delta layers - difference between summer and winter vegetation indexes), and spectral diversity, and used these attributes as variables to the model. We used the Random Forest algorithm to generate the models and evaluate how the tree cover classes differ in the classification and how the variables supported the model. We achieved high values of global accuracy (91.9%) and “F1 score” (above 0.8) for all tree cover classes, in which second-growth forest presented the lowest accuracy. The textural layers, delta layers, and the spectral diversity layers were the most important attributes to discriminate among tree cover classes. We demonstrate here the potential of using VENμS or similar sensor images together with different image processing and machine learning algorithms to monitor FLR programs, allowing further remote sensing approaches and in-deep field assessments to advance evaluation of FLR benefits for nature and people. We demonstrated how the fusion of all these types of data and innovative approaches to data processing, can result in novel ways to assess restoration performance and open new avenues to upscale monitoring, bridging the gap between FLR expectations and achieved goals.

Most European forests are used for timber production. Given the limited extent of unmanaged (and especially primary) forests, it is essential to include commercial forests in the conservation of forest biodiversity. In order to develop ecologically sustainable forest management practices, it is important to understand the management impacts on forest-dwelling organisms. Experiments allow testing the effects of alternative management strategies, and monitoring of multiple taxa informs us on the response range across forest-dwelling organisms. To provide a representative picture of the currently available information, metadata on 28 multi-taxa forest management experiments were collected from 14 European countries. We demonstrate the potential of compiling these experiments in a single network to upscale results from the local to continental level and indicate directions for future research. Among the different forest types, temperate deciduous beech and oak-dominated forests are the best represented in the multi-taxa manage ment experiments. Of all the experimental treatments, innovative ways of traditional manage ment techniques (e.g., gap cutting and thinning) and conservation-oriented interventions (e.g., microhabitat enrichment) provide the best opportunity for large-scale analyses. Regarding the organism groups, woody regeneration, herbs, fungi, beetles, bryophytes, birds and lichens offer the largest potential for addressing management–biodiversity relationships at the European level. We identified knowledge gaps regarding boreal, hemiboreal and broadleaved evergreen forests, the treatments of large herbivore exclusion, prescribed burning and forest floor or water ma nipulations, and the monitoring of soil-dwelling organisms and some vertebrate classes, e.g., amphibians, reptiles and mammals. To improve multi-site comparisons, design of future experi ments should be fitted to the set-up of the ongoing projects and standardised biodiversity sam pling is suggested. However, the network described here opens the way to learn lessons on the impact on forest biodiversity of different management techniques at the continental level, and thus, supports biodiversity conservation in managed forests. | publishedVersion

The forestry sector in Italy and throughout Europe is going through a critical period due to ongoing natural and anthropological processes, such as climate change and the abandonment of rural areas. These processes lead to a constant fragmentation of properties in small forest parcels, with direct impacts on management capacity. In this framework, new sustainable forest management methods are being tested and are shown to be good practices to oppose the decline of forest ecosystems. Their innovative aspects concern the introduction of a form of shared and circular economy, where management is built on the process, rather than on the product. Their technical activities are based on precision forestry systems and digitalization. The new approach takes into consideration the fact that the woods are an asset available to the whole community, in terms of benefits and protection. Forest Sharing^® is an example of the application of shared forest management systems, due to which the owner user benefits from several services and opportunities, such as the advanced monitoring platform and the access to investment funds. After eighteen months of activity, the first results of the application of the new management systems can already be seen. Many aspects need further development, such as case studies concerning the enhancement due to forest certification and new recreational activities. Shared forest management systems have the potential to increase the level of knowledge and awareness of citizens about environmental and territorial issues.