Remote Sensing in Ecology and Conservation最新文献

{"title":"Listening to the City: Evaluating BirdNET Reliability and Bias in Urban Bird Community Assessments","authors":"Valeria Ferrario, Riccardo Alba, Fabio Marcolin, Giordano Brambilla, Dan Chamberlain, Enrico Caprio","doi":"10.1002/rse2.70116","DOIUrl":"https://doi.org/10.1002/rse2.70116","url":null,"abstract":"BirdNET is increasingly used for passive acoustic monitoring (PAM), yet its reliability in urban environments that have high levels of anthropogenic noise remains poorly understood. We evaluated BirdNET's performance in two Italian cities, Turin and Florence, by analysing over 4300 10‐min audio recordings (approximately 723 h of audio in total) and manually validating up to 500 detections per species. In total, BirdNET detected 203 species in Florence and 192 in Turin, many of which were shared between the two cities. Using BirdNET Analyzer, we quantified true and false positives, characterized misclassification types, compared classifier performance between seasonally filtered and full‐year species lists, and estimated species‐specific confidence thresholds. BirdNET overestimated species richness, and a substantial proportion of detected species could not be confirmed during manual review: only 48 species in Turin and 65 in Florence were validated as genuinely present, corresponding to 58% and 52% of detected species being false detections, respectively. Applying seasonally filtered species lists improved precision relative to the full‐year list in both cities (by 11.2% in Turin and 8.2% in Florence). Most misclassifications resulted from biophonies (confusion with other bird species), followed by anthropophonies, such as traffic noise and human voices. Despite these errors, overall classification performance was good, with segment‐level precision reaching 0.704 in Turin and 0.780 in Florence. Optimal confidence thresholds varied widely among species and differed between cities, reflecting local noise conditions and community composition and highlighting the need for site‐specific calibration. Our results indicate that although BirdNET can effectively support urban biodiversity monitoring, its performance is highly context‐ and species‐dependent. Applying ecological filters and calibrating species‐specific confidence thresholds are essential to reduce misclassifications and improve the reliability of automated acoustic data in complex urban soundscapes.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"25 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149221448","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":"The Application of Autonomous Fixed‐Wing Drones in the Sub‐Antarctic Operating Beyond Visual Line of Sight to Enable Large Area, Multi‐Species Population Monitoring in a Changing Environment","authors":"N. Fenney, J. Coleman, A. Fox, P. N. Trathan, M. A. Collins, P. R. Hollyman","doi":"10.1002/rse2.70109","DOIUrl":"https://doi.org/10.1002/rse2.70109","url":null,"abstract":"Autonomous fixed‐wing drones operating beyond visual line of sight of the pilot were used to conduct the first aerial, multi species baseline reference survey at the sub‐Antarctic island of South Georgia. Long‐term population monitoring in this remote, challenging environment has historically proved difficult, with terrestrial monitoring generally limited to the immediate vicinity of two research stations, the only permanent human presence on the island. Methodologies capable of monitoring remote wildlife aggregations, and in challenging conditions, are critical for understanding the consequences of the recent rapid ecological and environmental changes at South Georgia. Here we utilised small, simple, but highly capable fixed‐wing drones, operating autonomously and beyond visual line of sight to overcome the difficulties faced by operating in the sub‐Antarctic, including with limited access and weather. The methodologies and workflows developed here represent a tool kit for conducting a large‐scale, long‐term wildlife monitoring programme in remote environments elsewhere around the world, where conventional terrestrial and aerial methods are not practical. Utilising these methods, we conducted a survey of key indicator species around South Georgia.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"157 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149172903","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":"High‐Resolution Satellite Imagery Reveals Landscape‐Scale Benthic Habitat Changes in a Turbid Coral Reef","authors":"Thais Andrade Galvão de Medeiros,Maria Laura Zoffoli,Gabriel Moiano,Vittorio Brando,Milton Kampel","doi":"10.1002/rse2.70112","DOIUrl":"https://doi.org/10.1002/rse2.70112","url":null,"abstract":"ABSTRACT Coral reefs are under increasing threat from local and global stressors, resulting in changes to the composition and spatial structure of benthic habitats. This study uses high‐spatial resolution satellite imagery and landscape metrics to analyse changes to the shallow, turbid reef at Sebastião Gomes on the Abrolhos Bank in the Southwestern Atlantic Ocean between 2010 and 2024. Benthic habitats were classified using a Random Forest model, which was trained using in situ observations and the following predictors: reflectance bands, spectral indices, texture metrics and satellite‐derived bathymetry from WorldView‐2 and WorldView‐3 imagery. This produced an overall classification accuracy of 91.9% (2010) and 88.3% (2024), although these values should be interpreted in the light of the temporal limitations of the dataset. Change detection analyses revealed a 77.1% decline in coral‐dominated habitats over 14 years, accompanied by a 25.4% increase in macroalgae and a 173.4% increase in zoanthids. This suggests a shift away from coral dominance. Landscape metrics revealed a spatial reorganisation of the reef. Coral habitats showed reductions in class area and largest patch size, as well as increased fragmentation and greater spatial isolation. In contrast, algal habitats expanded in a manner consistent with patch coalescence and spatial consolidation. This resulted in increasingly homogeneous benthic landscapes. Zoanthids displayed intermediate behaviour, expanding into newly available substrates without forming extensive, continuous patches. By integrating classification, change detection and landscape metrics, this study provides a comprehensive, spatially explicit assessment of reef structural change. The results highlight not only the loss of coral dominance, but also the reconfiguration of spatial complexity, which are critical for reef resilience and recovery after disturbance. This framework demonstrates the value of combining remote sensing and landscape ecology to monitor and interpret environmental change in turbid coral reef systems, supporting the development of spatially informed conservation and management strategies.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"25 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149091358","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":"Few Annotations, High Accuracy: Transfer Learning and Data Augmentation Improve Passive Acoustic Monitoring of the Vulnerable African Manatee","authors":"Lucas Dubus, Auguste Verdier, Nina Giotto, Grace Mbemba, Gabriel Michelin, Baptiste Mulot, Stéphanie Manel, David Mouillot","doi":"10.1002/rse2.70110","DOIUrl":"https://doi.org/10.1002/rse2.70110","url":null,"abstract":"Monitoring elusive and threatened species remains a major challenge in ecology and conservation, particularly in turbid aquatic environments. The African manatee ( <jats:styled-content style=\"fixed-case\"> <jats:italic>Trichechus senegalensis</jats:italic> </jats:styled-content> ), listed as Vulnerable by the IUCN, remains difficult to survey using visual methods, limiting reliable assessments of its distribution and population status. Passive acoustic monitoring (PAM) offers a promising alternative monitoring tool, but the large volumes of audio data generated require efficient automated detection of vocalizations. Here, we evaluate and compare three automated approaches for detecting African manatee vocalizations and occurrences across 52 sites over 13 months in the Conkouati‐Douli National Park (Republic of the Congo): (i) a widely used traditional machine learning recognizer implemented in Kaleidoscope, and two deep learning models, (ii) a convolutional neural network (CNN) that learns from local patterns in a visual representation of sounds, and (iii) a transformer model that incorporates attention layers to capture broader acoustic context. We further assess the influence of training dataset size, transfer learning, and data augmentation on model performance under realistic, data‐limited conditions. Both deep learning models substantially outperformed the traditional approach, achieving F1‐scores of 0.89 compared with 0.49 for Kaleidoscope, primarily due to much higher recall while maintaining high precision. Remarkably, deep learning models reach a high performance using only 10% of the annotated training dataset (≈300 vocalizations), demonstrating the effectiveness of transfer learning and data augmentation for bioacoustics applications with limited annotation effort. This ability to reliably detect true occurrences is particularly critical for monitoring threatened and elusive species, for which missed detections can lead to underestimation of habitat occupancy, spatial distribution, and population status. By minimizing false absences, deep learning‐based acoustic recognizers provide more reliable occurrence data for conservation decision‐making and long‐term monitoring even in data‐poor conditions.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"35 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149086130","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":"Repurposing Bird‐Based Acoustic Models for Amphibian Surveys: Strengths and Limitations in Real‐World Deployments","authors":"Vasylyna Strus, Yurii Strus, Nazar Smirnov, Anna Fedorova, Nataliia Suriadna, Mykola Drohvalenko","doi":"10.1002/rse2.70111","DOIUrl":"https://doi.org/10.1002/rse2.70111","url":null,"abstract":"Traditional biodiversity monitoring is often hindered by hazardous conditions and restricted human access. To bridge critical data gaps in these areas, we developed AnuraSearch (implemented as a frozen BirdNET v2.4 feature extractor coupled with a newly trained amphibian‐specific classification head). The 11 included species represent the majority of acoustically active Ukrainian anurans with available open‐access recordings in Xeno‐canto and <jats:ext-link xmlns:xlink=\"http://www.w3.org/1999/xlink\" xlink:href=\"http://observation.org\">Observation.org</jats:ext-link> . The dataset comprised 15,213 3‐s audio clips partitioned into training (80%; 12,170 clips) and validation (20%; 3043 clips) subsets. We evaluated the model via a two‐stage process. On an external test set of 4389 manually annotated 3‐s clips from independent high‐quality, predominantly single‐species recordings from open‐access archives—Xeno‐canto and <jats:ext-link xmlns:xlink=\"http://www.w3.org/1999/xlink\" xlink:href=\"http://observation.org\">Observation.org—(Stage</jats:ext-link> 1), AnuraSearch achieved macro‐averaged Precision = 0.994 and Recall = 0.62 (weighted Recall = 0.73). On a second external test set of 8686 expert‐verified field clips from 22 autonomous recording units deployed across diverse amphibian habitats within the Strict Nature Reserve Roztochya (Stage 2), macro‐Precision remained high (0.69) but macro‐Recall fell to 0.23, revealing species‐specific domain shift. Performance particularly declined for species with low‐amplitude calls or underwater vocalisations, highlighting the physical limitations of airborne monitoring for cryptic taxa. We applied the model to over 18,000 h of scheduled audio recordings (2 h around sunrise + 2 h around sunset + 5‐min clips every 30 min over the remaining 24‐h cycle), successfully reconstructing phenological and diel activity patterns without constant human presence. For acoustically cryptic and domain‐shifted species, AnuraSearch functions as a high‐precision clip‐prioritisation filter rather than a tool for reliable absence inference; reported confidence scores are not yet calibrated measures of detection reliability. Our results demonstrate that repurposing avian bioacoustic models via transfer learning is a viable, resource‐efficient strategy for maintaining ecological oversight.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"171 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149086236","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":"RETRACTION : Eigenfeature‐Enhanced Deep Learning: Advancing Tree Species Classification in Mixed Conifer Forests with Lidar","authors":"","doi":"10.1002/rse2.70108","DOIUrl":"https://doi.org/10.1002/rse2.70108","url":null,"abstract":"<jats:mixed-citation publication-type=\"journal\"> RETRACTION : <jats:string-name> <jats:given-names>R. C.</jats:given-names> <jats:surname>Blackburn</jats:surname> </jats:string-name> , <jats:string-name> <jats:given-names>R.</jats:given-names> <jats:surname>Buscaglia</jats:surname> </jats:string-name> , <jats:string-name> <jats:given-names>A. J. Sánchez</jats:given-names> <jats:surname>Meador</jats:surname> </jats:string-name> , <jats:string-name> <jats:given-names>M. M.</jats:given-names> <jats:surname>Moore</jats:surname> </jats:string-name> , <jats:string-name> <jats:given-names>T.</jats:given-names> <jats:surname>Sankey</jats:surname> </jats:string-name> , and <jats:string-name> <jats:given-names>S. E.</jats:given-names> <jats:surname>Sesnie</jats:surname> </jats:string-name> , “,” <jats:source>Remote Sensing in Ecology and Conservation</jats:source> <jats:volume>11</jats:volume> , no. ( <jats:year>2025</jats:year> ): <jats:fpage>670</jats:fpage> ‐ <jats:lpage>685</jats:lpage> , <jats:ext-link xmlns:xlink=\"http://www.w3.org/1999/xlink\" xlink:href=\"https://doi.org/10.1002/rse2.70014\">https://doi.org/10.1002/rse2.70014</jats:ext-link> . </jats:mixed-citation> The above article, published online on 09 June 2025 in Wiley Online Library ( <jats:ext-link xmlns:xlink=\"http://www.w3.org/1999/xlink\" xlink:href=\"https://onlinelibrary.wiley.com\">onlinelibrary.wiley.com</jats:ext-link> ), has been retracted by agreement between the authors; the journal Editor‐in‐Chief, Nathalie Pettorelli; Zoological Society of London; and John Wiley &amp; Sons Ltd. The authors contacted the journal and disclosed that errors in the training, validation, and testing datasets, as well as unintended data leakage, led to the publication of inflated performance metrics in the article. The editors have determined that the results and conclusions presented in the paper are compromised due to these errors, and thus the article must be retracted. The authors have been invited to submit a revised version of the article for further review by the journal. Any future publication of a revised version of the article will include a link to the retracted version of the article.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"27 1 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148915619","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":"One Size Does Not Fit All—Ignoring Spatial Variation in Occupancy and Detectability Underestimates the Required Number of Camera Trap Sites","authors":"Pradeep Raj Joshi, Andreas Zedrosser, Rahel Sollmann, Nuria Selva, Simone Santoro","doi":"10.1002/rse2.70103","DOIUrl":"https://doi.org/10.1002/rse2.70103","url":null,"abstract":"Occupancy models applied to camera trap data are widely used in wildlife monitoring, yet recommendations on the number of survey sites often assume homogeneous conditions and simple model structures. These assumptions are rarely met, and violations can yield biased or imprecise estimates. We used a simulation‐based power analysis to evaluate site number requirements for accurate occupancy estimation under three scenarios of spatial variation in occupancy (𝜓) and detection ( <jats:italic>p</jats:italic> ): (i) no variation (benchmark), (ii) simple variation (one covariate per process) and (iii) complex variation (multiple covariates, interactions and categorical predictors). Across combinations of baseline occupancy (𝜓 = 0.20, 0.50, 0.80), cumulative detection probability ( <jats:italic>p</jats:italic> * = 0.65, 0.80, 0.95) and varying degrees of spatial variability, we fitted single‐season occupancy models to simulated datasets of 10–1000 sites. We then identified the minimum number of sites required to achieve root‐mean‐squared error (RMSE) ≤ 0.10 in occupancy estimates. Sample size requirements varied widely across scenarios. Cumulative detection probability strongly influenced required sample sizes, while occupancy mainly mattered when detection was low. With no variation, accurate estimates required 10–40 sites. With simple variation, spatial requirements ranged from 20–120 sites regardless of the degree of spatial variation. Complex variation imposed the highest costs: Models with 2, 3, 4 and 8 parameters per component (𝜓 and <jats:italic>p</jats:italic> ) required 40–600 sites. Our results highlight that there is no one‐size‐fits‐all number of sites, as requirements differ by an order of magnitude across realistic scenarios. When occupancy and detection vary across space, especially with multiple or categorical predictors, more sites are needed, and sampling effort scales with model complexity. Logistic constraints may force a trade‐off between complex models whose inference is unreliable due to insufficient numbers of sites and simpler, potentially unrealistic models. We recommend simulation‐based power analyses to navigate this trade‐off.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"35 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902499","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":"Reliability of Manual Counting for Thermal Census of Bats Within Human‐Made Structures","authors":"Sarah K. Lagattuta, Janna M. E. Freeman, Nicole R. Gardner, Christine K. Johnson","doi":"10.1002/rse2.70102","DOIUrl":"https://doi.org/10.1002/rse2.70102","url":null,"abstract":"Traditional methods for censusing wild bat populations are often prone to bias and can be impractical for complex roosts within human‐made structures. Thermal infrared cameras have increasingly provided a remote sensing solution, but resulting bat counts from thermal video are typically generated by human observers for which reliability remains poorly understood. Because these manual counts serve both as population estimates and as benchmarks for validating novel computer vision censusing methods, assessing their potential biases is essential. This study evaluated the inter‐observer consistency of manual bat counts from thermal videos across three Mexican free‐tailed bat ( <jats:styled-content style=\"fixed-case\"> <jats:italic>Tadarida brasiliensis</jats:italic> </jats:styled-content> ) roosts in human‐made structures. We outlined methods for collecting thermal footage, training human counters ( <jats:italic>n</jats:italic> = 4), and assessing count consistency. We assessed interrater agreement and reliability through Bland–Altman plots, supplemented by intraclass correlation coefficients (ICC). One counter exhibited consistent undercounting, significantly reducing agreement ( <jats:italic>p</jats:italic> &lt; 0.001) and reliability ( <jats:italic>p</jats:italic> = 0.001). Correlation heatmap analysis indicated that high abundances of insects, birds, and other non‐bat flying objects increased count variability within the study group ( <jats:italic>p</jats:italic> &lt; 0.001). These findings demonstrate the importance of applying standardized statistical assessments of observer consistency before using manual counts as census estimates. Ensuring reliability is particularly critical as observer‐based thermal counts increasingly serve as validation datasets for computer vision approaches to bat censusing.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"6 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148902402","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":"Remote Sensing‐Based Identification of Habitat Loss for Herbivorous Waterbirds Driven by Wetland Inundation Alterations","authors":"Dongshuo Lu,Haobo Zhang,Peng Zhang,Xiaoning Liu,Liwei Chang,Lei Cheng,Jianbo Chang","doi":"10.1002/rse2.70107","DOIUrl":"https://doi.org/10.1002/rse2.70107","url":null,"abstract":"ABSTRACT Altered inundation regimes are reshaping floodplain wetlands and degrading wintering habitats for herbivorous waterbirds, yet long‐term, spatially explicit assessments that link hydrological change to habitat dynamics remain limited. Here, we developed a remote sensing‐based deep learning framework to reconstruct wintering habitat dynamics of herbivorous waterbirds in Dongting Lake from 2000 to 2021. The framework integrates multi‐source environmental predictors, receptive‐field optimization for spatial–context recognition, and Shapley Additive Explanations (SHAP) analysis for ecological interpretation. Results revealed marked habitat decline over the study period, with total habitat area decreasing by nearly half. Habitat loss was highly uneven, with the most severe declines occurring in South (−77.3%) and West (−89.2%) Dongting Lake, while East Dongting Lake remained comparatively stable. Landscape metrics further indicated progressive habitat contraction, increased isolation, and structural simplification, reflecting weakened spatial connectivity and reduced ecological resilience. Among all predictors, water recession timing emerged as the most influential predictor of habitat suitability, with recession in mid‐to‐late October associated with the highest habitat probability. A comparison with the NDVI time series showed that this recession window coincided with higher vegetation productivity. These findings demonstrate that wintering habitat dynamics are strongly associated with hydrological timing, particularly water recession phenology, rather than static water extent alone. The proposed framework provides a transferable approach for identifying hydrological controls on waterbird habitats and offers a scientific basis for adaptive wetland management along migratory flyways.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"16 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148895359","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":"Synthetic Imagery Improves Ecological Classification When Real Data Are Scarce, Not Direct Substitutes","authors":"Thomas A. Lake,Brennen T. Farrell,Ross K. Meentemeyer,Chris M. Jones","doi":"10.1002/rse2.70105","DOIUrl":"https://doi.org/10.1002/rse2.70105","url":null,"abstract":"ABSTRACT The growing availability of ecological imagery creates new opportunities to monitor biodiversity and ecosystem change at broad scales. Yet image classification models remain constrained by uneven sampling, limited labeled data, and differences between training and deployment imagery. Synthetic images generated with generative methods may help address these limitations, but their value across image domains remains poorly tested. We evaluated whether synthetic imagery improves genus‐level classification for 20 North American tree genera using iNaturalist community‐science photographs and Auto Arborist street‐view images. We generated domain‐specific synthetic images with Stable Diffusion v1.5 and trained ResNet‐50 classifiers on controlled mixtures of real and synthetic data. We tested synthetic replacement, supplementation under data scarcity, conventional augmentation alternatives, and transfer between image domains. Replacing real images with synthetic images reduced performance on held‐out real test sets, although declines differed between domains. Adding synthetic images improved classification when real classifier‐training data were scarce, but gains diminished as real data increased. In cross‐domain tests, synthetic supplementation improved some scarce‐data baselines but did not close the transfer gap or replace real‐image training sets. Synthetic images were therefore most useful as supplements rather than direct substitutes. For ecological applications, generated images may support model development for poorly sampled taxa, but representative real images remain essential for robust model generalization.","PeriodicalId":21132,"journal":{"name":"Remote Sensing in Ecology and Conservation","volume":"35 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148895357","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}