Remote Sensing in Ecology and Conservation最新文献

{"title":"ECOSTRESS‐derived semi‐arid forest temperature and evapotranspiration estimates demonstrate drought and thinning impacts","authors":"Temuulen Tsagaan Sankey, Thu Ya Kyaw, Julia Tatum, George W. Koch, Thomas Kolb, Rayni Lewis, Helen M. Poulos, Andrew M. Barton, Blase LaSala, Andrea Thode","doi":"10.1002/rse2.70026","DOIUrl":"https://doi.org/10.1002/rse2.70026","url":null,"abstract":"Southwestern US forests are experiencing increasing wildfire activity, and land managers are implementing large‐scale forest thinning treatments. We investigated semi‐arid ponderosa pine forest thinning treatment and regional drought impacts on ECOSTRESS land surface temperature (LST) and evapotranspiration (ET). Our study period at a northern Arizona study site included an average precipitation year, 2019, a regional drought period of 2020–2022, and a record winter snowfall year 2023. We examined ECOSTRESS LST and ET during spring seasons when the region experiences an annual dry period, and plant water stress is heightened. Our results indicate that ECOSTRESS LST data are sensitive to forest thinning, regional drought and their interaction. Consistent with high‐resolution UAV images, ECOSTRESS LST data indicate the thinned forest had significantly greater temperature across years, regardless of precipitation patterns. During drought, ECOSTRESS LST increased in both thinned and non‐thinned forests (by up to 10°C) and then declined in 2023. ECOSTRESS ET was similarly sensitive to forest thinning and regional drought. Consistent with <jats:italic>in situ</jats:italic> ET measurements, ECOSTRESS ET was significantly greater in the non‐thinned forest compared to the thinned forest. ECOSTRESS ET significantly decreased during drought in both forests. Our analysis of EMIT data indicates that EMIT trends are not consistent with ground‐based hyperspectral data that documented thinned forest moisture content is greater than that of the non‐thinned forest. While quality filtering reduces ECOSTRESS data temporal resolution, both ECOSTRESS LST and ET data can be used across large spatial extents to examine impacts of regional drought and management treatments in semi‐arid ponderosa pine forests.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"25 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144898131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating terrestrial and canopy laser scanning for comprehensive analysis of large old trees: Implications for single tree and biodiversity research","authors":"Barbara D'hont, Kim Calders, Alexandre Antonelli, Thomas Berg, Wout Cherlet, Karun Dayal, Olivia Jayne Fitzpatrick, Leonard Hambrecht, Maurice Leponce, Arko Lucieer, Olivier Pascal, Pasi Raumonen, Hans Verbeeck","doi":"10.1002/rse2.70021","DOIUrl":"https://doi.org/10.1002/rse2.70021","url":null,"abstract":"Large old trees provide multiple ecosystem services and contribute disproportionately to forest biomass and biodiversity. Yet their canopies remain among the least‐explored terrestrial habitats, despite their structural influence on key ecological processes such as light interception, moisture regulation, carbon storage and habitat formation. While terrestrial laser scanning (TLS) captures tree structure primarily from the ground, it struggles with occlusion and reduced precision in dense upper canopies, limiting information on fine‐scale branches and canopy vegetation. To address this, we introduce canopy laser scanning (CLS). We lifted a high‐end laser scanner into the canopy of six large, old trees by using scaffolding or climbers. Four trees are in diverse tropical rainforests in Colombia, Brazil and Peru and have large complex crowns with dense foliage. Two ‘giant’ trees stand out in Tasmania's wet, temperate eucalypt forests. Combining canopy and terrestrial scans resulted in a consistent high point cloud quality. The combined point clouds exhibited uniform point densities throughout the entire tree (downsampled to 1 cm), enabling a thorough examination of both the tree structure and its associated vegetation. Quantitative Structure Models (QSMs) showed, on average, a 20% increase (compared to TLS) in estimated branch volume and length, particularly concentrated in the upper crown region. We identified key epiphytic groups for a 5 × 5 × 5 m<jats:sup>3</jats:sup> subset of a tree. Our results show that CLS improves point cloud precision and reduces occlusion, enabling more accurate assessments of tree architecture and canopy biodiversity. Where feasible, this advancement creates new opportunities for 3D modelling of microhabitats, estimating aboveground carbon stocks, monitoring species and studying ecological dynamics.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"14 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144897861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantifying aboveground herbaceous biomass in grassy ecosystems: a comparison of field and high‐resolution UAV‐LiDAR approaches","authors":"Tyler C. Coverdale, Peter B. Boucher, Jenia Singh, Andrew B. Davies","doi":"10.1002/rse2.70023","DOIUrl":"https://doi.org/10.1002/rse2.70023","url":null,"abstract":"Grassy ecosystems cover &gt;25% of the world's land surface area. The abundance of herbaceous vegetation in these systems directly impacts a variety of ecological processes, including carbon sequestration, regulation of water and nutrient cycling, and support of grazing wildlife and livestock. Efforts to quantify herbaceous biomass, however, are often limited by a trade‐off between accuracy and spatial scale. Here, we describe a method for using Light Detection and Ranging (LiDAR) to estimate continuous aboveground biomass (AGB) at sub‐meter resolutions over large (10–10 000 ha) spatial scales. Across two African savanna ecosystems, we compared field‐ and LiDAR‐derived structural metrics—including measures of vegetation height and volume—with destructively harvested AGB by aligning our geospatial data with the location of harvested quadrats. Using this combination of approaches, we develop scaling equations to estimate spatially continuous herbaceous AGB over large areas. We demonstrate the utility of this method using a long‐term, large herbivore exclosure experiment as a case study and comprehensively compare common field‐ and LiDAR‐derived metrics for estimating herbaceous AGB. Our results indicate that UAV‐borne LiDAR provides comparable accuracy to standard field methods but over considerably larger areas. Nearly every measure of vegetation structure we quantified using LiDAR provided estimates of AGB that were comparable in accuracy (<jats:italic>R</jats:italic><jats:sup>2</jats:sup> &gt; 0.6) to the suite of common field methods we evaluated. However, marked differences between our two sites indicate that, for applications where accurate estimation of absolute biomass is a priority, site‐specific parameterization with destructive harvesting is necessary regardless of methodology. With the increasing availability of high‐resolution remote sensing data globally, our results indicate that many measures of herbaceous vegetation structure can be used to accurately compare AGB, even in the absence of complementary field data.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"15 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144792391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thermal drone observations capture fine‐scale population decline of short‐tailed shearwaters","authors":"Jacob Virtue, Darren Turner, Guy Williams, Stephanie Zeliadt, Arko Lucieer","doi":"10.1002/rse2.70020","DOIUrl":"https://doi.org/10.1002/rse2.70020","url":null,"abstract":"Monitoring seabird populations is increasingly urgent as numerous species become more vulnerable to climate change and urbanisation. Surveying burrow‐nesting seabirds is challenging due to their nocturnal behaviour, the inaccessibility of colonies, and the disturbance that monitoring poses to nesting sites. Traditional survey methods, which are manual transects conducted by researchers (~200 m), extrapolate this data to derive the population estimates of entire colonies. To enhance the accuracy beyond interpolated data, a survey method was developed using Unoccupied Aerial Systems (UAS) equipped with thermal sensors to survey short‐tailed shearwaters (<jats:italic>Ardenna tenuirostris</jats:italic>). Thermal imagery of breeding colonies was collected from 2019 to 2024, providing comprehensive coverage capturing all occupied burrows (chick presence) at each colony. Occupied burrow densities decreased from 0.28 to 0.18 burrows per m<jats:sup>2</jats:sup> over this period. Chick numbers decreased by 27% from 2019 (6129) to 2024 (4445). Burrow occupancy counts varied widely (0%–66%) with transect location, highlighting the advantages of using UAS‐mounted thermal sensors for providing spatially complete data. This indicates that counts are not uniform, highlighting the bias of using transect data to estimate chick production. A series of simulated transects were imposed over the thermal imagery to compare whole colony chick counts with extrapolated counts. Using data from this study, we estimated that the global breeding population of short‐tailed shearwaters is currently 13.5 million, which is approximately 41% less than the last reported global estimate in 1985 of 23 million. This study highlights the utility of emerging technology that addresses the challenges of studying species that are nocturnally active or in remote/inaccessible habitats.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"119 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144763123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial distribution and drivers of aboveground forest biomass in Mexico using GEDI and national forest inventory data","authors":"José Luis Hernández‐Stefanoni, Luis A. Hernández‐Martínez, Juan Andres‐Mauricio, Víctor Alexis Peña‐Lara, Karina Elizabeth González‐Muñoz, Fernando Tun‐Dzul, Carlos A. Portillo‐Quintero, Eric Antonio Gamboa‐Blanco, Stephanie George‐Chacon","doi":"10.1002/rse2.70019","DOIUrl":"https://doi.org/10.1002/rse2.70019","url":null,"abstract":"Accurate assessment of forest aboveground biomass density (AGBD) is essential for understanding the role of vegetation in climate change mitigation and developing forest management and environmental policies at national and regional levels. The Global Ecosystem Dynamics Investigation (GEDI) uses full‐waveform LiDAR and provides a valuable tool for estimating AGBD. Calibrating GEDI biomass products with local field data is vital for improving model accuracy, as current estimates rely on global datasets. Additionally, evaluating key factors that influence biomass estimation is essential to refine GEDI‐based models. In this research, we calibrated linear models with field AGBD as the dependent variable and GEDI metrics as independent variables, and compared the performance against the GEDI L4A product across forest types. Additionally, we evaluated the effects of terrain slope, forest structural complexity, and forest type on the accuracy of the models. Finally, we mapped AGBD in Mexico by aggregating footprint‐level estimates with local models and compared it with the GEDI AGBD map (L4B product). Model validation showed <jats:italic>R</jats:italic><jats:sup>2</jats:sup> values from 0.35 to 0.46 across forest types, with most models having %RMSE below 52.0. Errors were 32.7 to 34.2% lower than GEDI L4A, highlighting a notable accuracy improvement. The total carbon stocks in Mexico estimated here are approximately 1.78 Gt, aligning closely with official FAO estimates, whereas GEDI estimates are 33.5% higher than the official estimate. Biomass estimation with GEDI is most accurate in areas with moderate slopes and low forest structural complexity. Coniferous and tropical forests showed the lowest errors in estimating AGBD with GEDI (46.7 and 47.3 of %RMSE, respectively) likely due to the widespread presence of uniformly structured coniferous trees and the moderate terrain slopes found in tropical forests. Our findings highlight the importance of calibrating local AGBD data with GEDI forest structure metrics to improve biomass estimations at the footprint and national levels.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"14 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144669664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Precipitation and temperature drive woody vegetation dynamics in the grasslands of sub‐Saharan Africa","authors":"Francesco D'Adamo, Rebecca Spake, James M. Bullock, Booker Ogutu, Jadunandan Dash, Felix Eigenbrod","doi":"10.1002/rse2.70018","DOIUrl":"https://doi.org/10.1002/rse2.70018","url":null,"abstract":"Identifying the drivers of ecosystem dynamics, and how responses vary spatially and temporally, is a critical challenge in the face of global change. Grasslands in sub‐Saharan Africa are vital ecosystems supporting biodiversity, carbon storage, and livelihoods through grazing. However, despite their importance, the processes driving change in these systems remain poorly understood, as cross‐scale interactions among drivers produce complex, context‐dependent dynamics that vary across space and time. This is particularly relevant for woody vegetation dynamics, which are often linked to degradation processes (e.g., woody encroachment), with consequences for biodiversity, forage availability, and fire regimes. Here, we used satellite data and structural equation models to investigate the effects of rainfall, temperature, fire, and population density on woody vegetation dynamics in four African grassland regions (the Sahel grasslands, Greater Karoo and Kalahari drylands, Southeast African subtropical grasslands, and Madagascar) during 1997–2016. Across all regions, rainfall was consistently positively correlated with increased woody vegetation, while higher temperatures were associated with decreased woody vegetation, suggesting that water availability promotes woody plant growth, whereas rising aridity limits it. Unexpectedly, fire had a negative effect on woody cover only in the Greater Karoo and Kalahari drylands, while in Madagascar, higher temperatures and greater population density reduced fire; yet these relationships did not translate into significant indirect effects on woody vegetation. These findings illustrate the complex ways by which environmental and anthropogenic drivers shape woody vegetation dynamics in grasslands across sub‐Saharan Africa. Compared to savannas, fire plays a weaker and more region‐specific role in grasslands, where its feedback with woody cover is less consistent. The opposing effects of rainfall and temperature may currently constrain woody expansion, but climate change could disrupt this balance and further weaken fire's limited regulatory role. These differences highlight the need for management strategies tailored to the distinct climate–vegetation dynamics of grassland systems.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"13 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144629804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Drone photogrammetry reveals contrasting body conditions of dugongs across the Indo‐Pacific","authors":"Camille Goudalier, David Mouillot, Léa Bernagou, Taha Boksmati, Caulvyn Bristol, Harry Clark, Sekar M.C. Herandarudewi, Régis Hocdé, Anna Koester, Ashlie J. McIvor, Dhivya Nair, Muhammad Rizki Nandika, Louisa Ponnampalam, Achmad Sahri, Evan Trotzuk, Nur Abidah Zaaba, Laura Mannocci","doi":"10.1002/rse2.70016","DOIUrl":"https://doi.org/10.1002/rse2.70016","url":null,"abstract":"The monitoring of body condition, reflecting the state of individuals' energetic reserves, can provide early warning signals of population decline, facilitating prompt conservation actions. However, environmental and anthropogenic drivers of body condition are poorly known for rare and elusive marine mammal species over their entire ranges. We assessed the global patterns and drivers of body condition for the endangered dugong (<jats:italic>Dugong dugon</jats:italic>) across its Indo‐Pacific range. To do so, we applied the body condition index (BCI) developed for the related manatee based on the ratio of umbilical girth (approximated as maximum width times π), to straight body length measured in drone images. To cover the entire dugong's range, we took advantage of drone footage published on social media. Combined with footage from scientific surveys, social media footage provided body condition estimates for 272 individual dugongs across 18 countries. Despite small sample sizes relative to local population sizes, we found that dugong BCI was better, that is, individuals were ‘plumper’, in New Caledonia, the United Arab Emirates, Australia and Qatar where populations are the largest globally. Dugong BCI was comparatively poorer in countries hosting very small dugong populations such as Mozambique, suggesting a link between body condition and population size. Using statistical models, we then investigated potential environmental and anthropogenic drivers of dugong BCI, while controlling for seasonal and individual effects. The BCI decreased with human gravity, a variable integrating human pressures on tropical reefs, but increased with GDP per capita, indicating that economic wealth positively affects dugong energetic state. The BCI also showed a dome‐shaped relationship with marine protected area coverage, suggesting that extensive spatial protection is not sufficient to maintain dugongs in good state. Our study provides the first assessment of dugong body condition through drone photogrammetry, underlining the value of this non‐invasive, fast and low‐cost approach for monitoring elusive marine mammals.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"644 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144341173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interannual spectral consistency and spatial uncertainties in UAV‐based detection of boreal and subarctic mire plant communities","authors":"Franziska Wolff, Tiina H. M. Kolari, Aleksi Räsänen, Teemu Tahvanainen, Pasi Korpelainen, Miguel Villoslada, Mariana Verdonen, Eliisa Lotsari, Yuwen Pang, Timo Kumpula","doi":"10.1002/rse2.70017","DOIUrl":"https://doi.org/10.1002/rse2.70017","url":null,"abstract":"Unoccupied Aerial Vehicle (UAV) imagery is widely used for detailed vegetation modeling and ecosystem monitoring in peatlands. Despite high‐resolution data, the spatial complexity and heterogeneity of vegetation, along with temporal fluctuations in spectral reflectance, complicate the assessment of spatial patterns in these ecosystems. We used interannual multispectral UAV data, collected at the same time of the year, from two aapa and two palsa mires in Finland. We applied Random Forest classification to map plant communities and assessed spectral, temporal and spatial consistency, class relationships and area estimates. Further, we used the class membership probabilities from the classification to derive a secondary classification map, representing the second most likely class label per‐pixel and an alternative map to account for spatial uncertainty in area estimates. The accuracies of the primary classifications varied between 66 and 85%. The best results were achieved using interannual data, improving accuracy by up to 14%‐points when compared to single‐year imagery, particularly benefiting classes with lower accuracies. Spectral and temporal inconsistencies in the UAV data collected in different years led to variations in the classifications, notably for the <jats:italic>Rubus chamaemorus</jats:italic> community in palsa mires, likely due to weather fluctuations and phenology. The transformations from primary to secondary classifications in areas of high uncertainty aligned well with the class relationships in the confusion matrix, supporting the model's reliability. Confidence interval‐based adjusted estimates aligned largely with unadjusted area estimates of the alternative map. Our findings support incorporating class membership probabilities and alternative maps to capture spatially explicit uncertainty, especially when spatial variability is high or key plant communities are involved. Our presented approach is particularly beneficial for upscaling ecological processes, such as carbon fluxes, where spatial variability is driven by plant community distribution and where informed decision‐making requires detailed spatial assessments.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"15 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144341174","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Consistent and scalable monitoring of birds and habitats along a coffee production intensity gradient","authors":"Marius Somveille, Joe Grainger‐Hull, Nicole Ferguson, Sarab S. Sethi, Fernando González‐García, Valentine Chassagnon, Cansu Oktem, Mathias Disney, Gustavo López Bautista, John Vandermeer, Ivette Perfecto","doi":"10.1002/rse2.70015","DOIUrl":"https://doi.org/10.1002/rse2.70015","url":null,"abstract":"Land use change associated with agricultural intensification is a leading driver of biodiversity loss in the tropics. To evaluate the habitat–biodiversity relationship in production systems of tropical agricultural commodities, birds are commonly used as indicators. However, a consistent and reliable methodological approach for monitoring tropical avian communities and habitat quality in a way that is scalable is largely lacking. In this study, we examined whether the automated analysis of audio data collected by passive acoustic monitoring, together with the analysis of remote sensing data, can be used to efficiently monitor avian biodiversity along the gradient of habitat degradation associated with the intensification of coffee production. Coffee is an important crop produced in tropical forested regions, whose production is expanding and intensifying, and coffee production systems form a gradient of ecological complexity ranging from forest‐like shaded polyculture to dense sun‐exposed monoculture. We used LiDAR technology to survey the habitat, together with autonomous recording units and a vocalization classifier to assess bird community composition in a coffee landscape comprising a shade‐grown coffee farm, a sun coffee farm and a forest remnant, located in southern Mexico. We found that LiDAR can capture relevant variation in vegetation across the habitat gradient in coffee systems, specifically matching the generally observed pattern that the intensification of coffee production is associated with a decrease in vegetation density and complexity. We also found that bioacoustics can capture known functional signatures of avian communities across this habitat degradation gradient. Thus, we show that these technologies can be used in a robust way to monitor how biodiversity responds to land use intensification in the tropics. A major advantage of this approach is that it has the potential to be deployed cost‐effectively at large scales to help design and certify biodiversity‐friendly productive landscapes.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"16 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144337521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Eigenfeature‐enhanced deep learning: advancing tree species classification in mixed conifer forests with lidar","authors":"Ryan C. Blackburn, Robert Buscaglia, Andrew J. Sánchez Meador, Margaret M. Moore, Temuulen Sankey, Steven E. Sesnie","doi":"10.1002/rse2.70014","DOIUrl":"https://doi.org/10.1002/rse2.70014","url":null,"abstract":"Accurately classifying tree species using remotely sensed data remains a significant challenge, yet it is essential for forest monitoring and understanding ecosystem dynamics over large spatial extents. While light detection and ranging (lidar) has shown promise for species classification, its accuracy typically decreases in complex forests or with lower lidar point densities. Recent advancements in lidar processing and machine learning offer new opportunities to leverage previously unavailable structural information. In this study, we present an automated machine learning pipeline that reduces practitioner burden by utilizing canonical deep learning and improved input layers through the derivation of eigenfeatures. These eigenfeatures were used as inputs for a 2D convolutional neural network (CNN) to classify seven tree species in the Mogollon Rim Ranger District of the Coconino National Forest, AZ, US. We compared eigenfeature images derived from unoccupied aerial vehicle laser scanning (UAV‐LS) and airborne laser scanning (ALS) individual tree segmentation algorithms against raw intensity and colorless control images. Remarkably, mean overall accuracies for classifying seven species reached 94.8% for ALS and 93.4% for UAV‐LS. White image types underperformed for both ALS and UAV‐LS compared to eigenfeature images, while ALS and UAV‐LS image types showed marginal differences in model performance. These results demonstrate that lower point density ALS data can achieve high classification accuracy when paired with eigenfeatures in an automated pipeline. This study advances the field by addressing species classification at scales ranging from individual trees to landscapes, offering a scalable and efficient approach for understanding tree composition in complex forests.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"47 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144252281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}