Ecological Informatics最新文献

{"title":"Autonomous data sampling for high-resolution spatiotemporal fish biomass estimates","authors":"Astrid A. Carlsen ,&nbsp;Michele Casini ,&nbsp;Francesco Masnadi ,&nbsp;Olof Olsson ,&nbsp;Aron Hejdström ,&nbsp;Jonas Hentati-Sundberg","doi":"10.1016/j.ecoinf.2024.102852","DOIUrl":"10.1016/j.ecoinf.2024.102852","url":null,"abstract":"<div><div>Many key ecological dynamics such as biomass distributions are only detectable on a fine spatiotemporal scale. Autonomous data collection with Unmanned Surface Vehicles (USV) creates new possibilities for cost efficient and high-resolution aquatic data sampling. However, the spatial coverage and sampling resolution remain uncertain due to the novelty of the technology. Further, there is no established method for analysing such fine-scale autocorrelated data without aggregation, potentially compromising data resolution. We here used a USV with an echosounder, a conductivity-temperature sensor and a flourometer to collect data from April–July 2019–2023 in a 60x80km area in the central Baltic Sea. The USV covered a total distance of 8000 nmi, over 42–81 days per year, with an average speed of 0.5 m/s. We combined the hydroacoustic data with publicly available oceanographic data from Copernicus Marine Service Information (CMSI) to describe seasonal distribution dynamics of a small pelagic fish community. Key oceanographic variables collected by the USV were correlated with CMSI estimates at daily/monthly resolution, respectively, to test for suitability to scale (Temperature 0.99/0.97; Salinity −0.77/−0.26; Chlorophyll-a 0.12/0.28). We investigated two approaches of Species Distribution Models (SDMs): generalized additive models (GAM) versus spatiotemporal generalized linear mixed effect models (GLMM). The GLMMs explained the observed data better than the GAMs (R² 0.31 and 0.20, respectively). The addition of environmental variables increased the explanatory capability of GAM and GLMM by 25 % and ∼ 3 %, respectively. Due to the high data resolution, we found significant amounts of positive autocorrelation (R: 0.05–0.30) across more than 50 sequential observations (&gt;6 hours). However, we found that diel patterns in fish detection strongly affected the abundance estimates due to vertically migrating species hiding in the ‘acoustic dead zone’ near the seabed. Such dynamics could only be estimated and corrected for in predictions on the high-resolution data, complicating the trade-off between autocorrelation and high-resolution for SDMs. We compared estimates and effect sizes/directions in identical SDMs on 2x2km/month aggregated (i.e non-autocorrelated) observations and non-aggregated (i.e. autocorrelated) observations, and found relatively little difference in spatiotemporal estimates (<em>r</em> = 0.80). For the first time, we predicted the distribution of a small pelagic fish community at a high spatial resolution, in an area essential to breeding top predators, opening up for new applications in ecological studies locally and globally.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142529246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multidimensional machine learning framework for LST reconstruction and climate variable analysis in forest fire occurrence","authors":"Hatef Dastour,&nbsp;Quazi K. Hassan","doi":"10.1016/j.ecoinf.2024.102849","DOIUrl":"10.1016/j.ecoinf.2024.102849","url":null,"abstract":"<div><div>Land Surface Temperature (LST) datasets play a crucial role in understanding the complex interplay between forest fires, climate variables, and vegetation dynamics. This study is divided into two primary parts: the first part investigates the predictive performance of a machine learning framework based on CatBoost and XGBoost models in estimating LST across different land cover classes in Alberta, Canada. On the test set, for LST-Day data, CatBoost and XGBoost achieved Median Absolute Errors (MedAE) of approximately 1.434 °C and 1.425 °C, respectively. For LST-Night data, also on the test set, the MedAE values were approximately 1.186 °C for CatBoost and 1.176 °C for XGBoost. The second part explores the intricate relationships between climatic variables—LST, precipitation, and relative humidity—forest fire occurrences, and vegetation dynamics in various subregions. The findings revealed complex interactions, with high LST, reduced precipitation, and humidity associated with increased forest fire activity and subsequent changes in vegetation patterns, particularly in the Central Mixedwood, Dry Mixedwood, and Montane subregions. A notable potential association was identified between high LST, reduced precipitation and humidity, and increased forest fire activity in these areas. These climate change impacts and fire events were found to influence ecological processes, altering species composition, reducing biodiversity, and potentially disrupting ecosystem services such as carbon sequestration and nutrient cycling. These insights are crucial for informing adaptive forest management strategies aimed at understanding and mitigating the cascading effects of climate change on fire regimes and vegetation dynamics in Alberta's diverse landscapes.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142441572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Post-fire vegetation dynamic patterns and drivers in Greater Hinggan Mountains: Insights from long-term remote sensing data analysis","authors":"Bohan Jiang ,&nbsp;Wei Chen ,&nbsp;Yuan Zou ,&nbsp;Chunying Wu ,&nbsp;Ziyi Wu ,&nbsp;Xuechun Kang ,&nbsp;Haiting Xiao ,&nbsp;Tetsuro Sakai","doi":"10.1016/j.ecoinf.2024.102850","DOIUrl":"10.1016/j.ecoinf.2024.102850","url":null,"abstract":"<div><div>Fire has become a major disturbing factor in boreal forests, and giant forest disturbances play a vital role in regulating the climate under global warming. Therefore, it is essential to investigate the spatiotemporal patterns and main drivers of post-fire vegetation recovery for forest ecological research and post-fire recovery management. However, previous studies have focused on the post-fire forest change within the entire fire perimeter, lacking separate analysis and comparison of the burned zone (BZ) and unburned zone (UNBZ). Here, we propose the utilization of Moderate Resolution Imaging Spectroradiometer land cover type and vegetation index data to monitor vegetation dynamics and explore its drivers after the most serious forest fire in the history of P.R. China in the Greater Hinggan Mountains (GHM). The temporal and spatial patterns of vegetation recovery in the BZ/UNBZ in the GHM were analyzed using the Sen &amp; Mann-Kendall method, Hurst index and coefficient of variation, and their driving mechanisms were explored using GeoDetector and geographically weighted regression. The results showed that there were significant differences in the spatial distribution and fluctuation of vegetation between the BZ and UNBZ, and that the BZ exhibited higher productivity and vigor. Vegetation recovery was influenced by different dominant factors and changed over time, in which land surface temperature and precipitation dominated all the time, whereas topographic relief and elevation had a more significant contribution to vegetation recovery in the BZ and UNBZ, respectively. This study provides a scientific basis for the protection and management of vegetation in disturbed forested areas, particularly after fires.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rapid diagnosis of the geospatial distribution of intertidal macroalgae using large-scale UAVs","authors":"Andrea Martínez-Movilla ,&nbsp;Juan Luis Rodríguez-Somoza ,&nbsp;Marta Román ,&nbsp;Celia Olabarria ,&nbsp;Joaquín Martínez-Sánchez","doi":"10.1016/j.ecoinf.2024.102845","DOIUrl":"10.1016/j.ecoinf.2024.102845","url":null,"abstract":"<div><div>Macroalgae have been used as indicators of the health of coastal ecosystems, they function as sinks of CO<span><math><msub><mrow></mrow><mn>2</mn></msub></math></span> and are essential contributors to primary production. With the increase in anthropogenic activities, it is crucial to assess the impact of such activities on these ecosystems. As traditional surveying techniques, although accurate, are time-consuming and their area coverage is limited, novel techniques are required to monitor the coverage and diversity of intertidal macroalgae. We propose a methodology using the free-source Semi-Automatic Classification Plugin from QGIS to use UAV and multispectral cameras for the spatiotemporal monitoring of intertidal macroalgae. We also compared the performance of six classifiers: Minimum Distance (MD), Maximum Likelihood (ML), Spectral Angle Mapping (SAM), Multi-Layer Perceptron (MLP), Random Forest (RF) and Support Vector Machine (SVM), for three types of macroalgae classification: general, taxonomical groups and species. As proof of concept, an intertidal rocky shore in a marine protected area (NW Spain) was studied for four months. RF and SVM achieved similar results, with both being recommended for the general (OA<span><math><msub><mrow></mrow><mi>SVM</mi></msub></math></span> = 97.4<span><math><mo>±</mo></math></span>1.7 and OA<span><math><msub><mrow></mrow><mi>RF</mi></msub></math></span> = 98.3<span><math><mo>±</mo></math></span>1.7) and taxonomical groups (OA<span><math><msub><mrow></mrow><mi>SVM</mi></msub></math></span> = 91.6<span><math><mo>±</mo></math></span>1.9 and OA<span><math><msub><mrow></mrow><mi>RF</mi></msub></math></span> = 89.2<span><math><mo>±</mo></math></span>4.5). SVM and ML were found to be more suitable for species classification (OA<span><math><msub><mrow></mrow><mi>SVM</mi></msub></math></span> = 77.4<span><math><mo>±</mo></math></span>11.4 and OA<span><math><msub><mrow></mrow><mi>ML</mi></msub></math></span> = 74.2<span><math><mo>±</mo></math></span>9.7). SAM and MLP provided the least performant species classifiers because of the overlap in the macroalgae spectral signatures. The plugin showed limitations when tuning the input parameters of the MLP classifier and did not let to add a validation dataset. Additionally, we present an open-access GIS web application, Alganat 2000 GIS web, to facilitate the monitoring and management of coastal areas. We conclude that the proposed methodology using the SVM or ML classifiers is an effective tool for assessing intertidal macroalgal assemblages. Its easy and rapid implementation is beneficial for researchers who are not very familiar with coding and machine learning frameworks and reduces the time and cost of fieldwork. As future work, we propose the combination of the multispectral bands with topographic and spectral indices and to research the application of deep learning models to the classification of intertidal macroalgae.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting potential biomass production by geospatial modelling: The case study of citrus in a Mediterranean area","authors":"G.A. Catalano,&nbsp;P.R. D'Urso,&nbsp;C. Arcidiacono","doi":"10.1016/j.ecoinf.2024.102848","DOIUrl":"10.1016/j.ecoinf.2024.102848","url":null,"abstract":"<div><div>Residual biomass from agricultural production and processing, such as citrus pulp and olive pomace, is an important resource for energy production. In particular, this is the case in regions where transformation industries are concentrated.</div><div>Current biomass estimates often focus on actual production data, that may not fully capture the biomass potential across all suitable cultivation areas. To bridge this gap, the study predicts the overall potential biomass available for energy production, taking into account the total area suitable for citrus cultivation.</div><div>The research is focused on the study of citrus species in the province of Syracuse, Sicily, Italy. The methodology combines Geographic Information System (GIS) tools, for data interpolation and map overlays, with Software for Assisted Habitat Modelling (SAHM) for local level simulations.</div><div>The results of the different models showed accurate and spatially coherent predictions, with AUC values ranging from 0.85 to 0.90, and highest potentialities in the northern and eastern regions of the study area. The results highlighted potential citrus cultivation on 47,706 ha and estimated 184,340 t of biomass, 16,461,520.82 Nm³ of biogas, and 8110 t of digestate. The results of the study identified potential areas for both increasing biomass production and optimising the distribution of digestate, thus demonstrating the utility of these thematic maps as a decision support tool for land management. The simulations and the methodology applied in this study indicated potential economic and environmental benefits to be gained from sustainable by-product management. This approach facilitates the optimisation of decision-making processes for land planning, thereby contributing to the broader objective of improving resource efficiency and sustainability in the agricultural production and processing sectors.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142440947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Functional data analysis to describe and classify southern resident killer whale calls","authors":"Paul Nguyen Hong Duc ,&nbsp;David A. Campbell ,&nbsp;Michael Dowd ,&nbsp;Ruth Joy","doi":"10.1016/j.ecoinf.2024.102841","DOIUrl":"10.1016/j.ecoinf.2024.102841","url":null,"abstract":"<div><div>The Southern Resident Killer Whale (SRKW) is an endangered population of whales found in the northeast Pacific. They have a vocal dialect unique from other killer whales, having a repertoire of distinct stereotyped calls. A framework for distinguishing SRKW call types using the frequency traces of the amplitude ridges from their spectrograms (termed frequency ridges) is proposed. The first step is the extraction of these ridges of SRKW calls using an Sequential Monte Carlo approach. Next, they are converted into functional data using B-spline functions. They are analyzed with a functional principal component (FPC) analysis to characterise the intrinsic variability of frequency ridges within a call type. The FPCs are able to capture the general patterns in the frequency ridges of the different SRKW call types. The FPCs are also used as the basis for call classification. Using a cross-validation procedure to assess the robustness of the classification, this framework proves to be successful for classification with some call types having an F1-score <span><math><mo>≥</mo><mn>80</mn><mo>%</mo></math></span>, but other calls less well discriminated. On balance, this approach showed reasonable performance given the small sample size available, and provides a useful contribution towards the development of a universal tool for call classification.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Navigating uncertainty in carbon efficiency: A global assessment across income groups","authors":"Ziyao Li ,&nbsp;Sangmok Kang","doi":"10.1016/j.ecoinf.2024.102837","DOIUrl":"10.1016/j.ecoinf.2024.102837","url":null,"abstract":"<div><div>This study evaluates the carbon efficiency of 163 countries between 1992 and 2019, focusing on the relationship between economic growth and emission reductions. By using a novel approach that integrates Stochastic Metafrontier Analysis with Bayesian inference, the study robustly analyzes data variability and uncertainty. The results highlight significant differences in carbon efficiency across income groups. High-income countries (G1) show a technology gap uncertainty of 0.118, while low-income countries (G4) have a slightly higher uncertainty at 0.133, indicating challenges in technology transfer for both groups. Middle-income countries (G2), with the lowest uncertainty at 0.045, demonstrate a strong capacity to adopt advanced technologies and improve carbon efficiency. The study also identifies critical factors influencing carbon efficiency uncertainty, such as urbanization, forest area, and foreign direct investment. Urbanization affects these groups differently: it raises uncertainty in G4 by 0.0107 but reduces it in G1 by −0.0069, reflecting varying stages of urban development. These findings suggest the need for targeted policies to improve technology transfer, optimize urbanization, and enhance sustainable resource use, thereby facilitating a more effective shift to a low-carbon economy and reducing carbon efficiency uncertainties.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142432080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SooSim, a landscape model for assessing mire habitat degradation and restoration","authors":"Asko Lõhmus ,&nbsp;Raido Kont ,&nbsp;Triin Kaasiku ,&nbsp;Marko Kohv ,&nbsp;Tauri Arumäe ,&nbsp;Ants Kaasik","doi":"10.1016/j.ecoinf.2024.102844","DOIUrl":"10.1016/j.ecoinf.2024.102844","url":null,"abstract":"<div><div>Open mires constitute a characteristic part of boreal natural landscapes, which is under various cumulative anthropogenic pressures. As a response, remaining mires are increasingly protected, and their degradation is addressed by ecological restoration (mostly drainage closure). To evaluate alternative environmental policies, there is a necessity for high-resolution landscape simulation models to assess future dynamics of mires under different management scenarios. We present such a model, <em>SooSim</em>, its R-script, and derivation and validation of its key parameters. <em>SooSim</em> iterates mire types and woody encroachment dynamics within 25 × 25 m grid at 1-year intervals. Management interventions (restoration; ditch renovation) are sequentially introduced based on priority rules in locations delineated prior to simulation. We applied <em>SooSim</em> to three management scenarios, compared with natural succession, until 2050 in Estonia. The ‘current’ (2022) database comprised &gt;3.8 M mire pixels and &gt; 7 M peatland-forest pixels (sparse-cover ones considered for mire restoration). The model parameterization, based on Lidar data, revealed rapid ongoing woody encroachment across all mire types, with significant positive feedback. The simulations revealed that, even in scenarios with intensive restoration (2500 ha annually), open mire conditions are reduced by &gt;10 % until 2050, while few mire types lose &gt;1 % in area. Ditch renovations mostly reduced restoration perspectives in currently forested peatlands. Thus, <em>SooSim</em> explicitly depicts a decision-making dilemma where mire restoration is time-sensitive but also uncertain. To address this and related land-use dilemmas in the environmental policy, landscape models such as <em>SooSim</em> have further importance as visualization tools to explain complex processes to a wide range of stakeholders.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated long-term monitoring of stereotypical movement in polar bears under human care using machine learning","authors":"Matthias Zuerl ,&nbsp;Philip Stoll ,&nbsp;Ingrid Brehm ,&nbsp;Jonas Sueskind ,&nbsp;René Raab ,&nbsp;Jan Petermann ,&nbsp;Dario Zanca ,&nbsp;Ralph Simon ,&nbsp;Lorenzo von Fersen ,&nbsp;Bjoern Eskofier","doi":"10.1016/j.ecoinf.2024.102840","DOIUrl":"10.1016/j.ecoinf.2024.102840","url":null,"abstract":"<div><div>The welfare of animals under human care is often assessed by observing behaviours indicative of stress or discomfort, such as stereotypical behaviour (SB), which often shows as repetitive, invariant pacing. Traditional behaviour monitoring methods, however, are labour-intensive and subject to observer bias. Our study presents an innovative automated approach utilising computer vision and machine learning to non-invasively detect and analyse SB in managed populations, exemplified by a longitudinal study of two polar bears. We designed an animal tracking framework to localise and identify individual animals in the enclosure. After determining their position on the enclosure map via homographic transformation, we refined the resulting trajectories using a particle filter. Finally, we classified the trajectory patterns as SB or normal behaviour using a lightweight random forest approach with an accuracy of 94.9 %. The system not only allows for continuous, objective monitoring of animal behaviours but also provides insights into seasonal variations in SB, illustrating its potential for improving animal welfare in zoological settings. Ultimately, we analysed 607 days for the occurrence of SB, allowing us to discuss seasonal patterns of SB in both the male and female polar bear monitored. This work advances the field of animal welfare research by introducing a scalable, efficient method for the long-term, automated detection and monitoring of stereotypical behaviour, paving the way for its application across various settings and species that can be continuously monitored with cameras. We made the code publicly available at <span><span><span>https://github.com/team-vera/stereotypy-detector</span></span><svg><path></path></svg></span>.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Collectively advancing deep learning for animal detection in drone imagery: Successes, challenges, and research gaps","authors":"Daniel Axford ,&nbsp;Ferdous Sohel ,&nbsp;Mathew A Vanderklift ,&nbsp;Amanda J Hodgson","doi":"10.1016/j.ecoinf.2024.102842","DOIUrl":"10.1016/j.ecoinf.2024.102842","url":null,"abstract":"<div><div>Drones have emerged as a powerful tool in animal detection, significantly advancing wildlife monitoring, conservation, and management by capturing high-resolution, real-time imagery over areas often inaccessible or challenging for human observers to reach. However, manual analysis of drone imagery for animal detection is labour-intensive and time-consuming. The application of deep learning methods, particularly convolutional neural networks, in automating animal detection from drone imagery has the potential to revolutionise wildlife monitoring, conservation, and management protocols.</div><div>This review provides a comprehensive overview of the increasing use and prospects of deep learning in animal detection using drone imagery. It explores successful applications of deep learning for animal detection, localisation, recognition, and their combinations. The paper also discusses the challenges, limitations, and future research directions of this field. A key message from this review is the need for representative training data covering the various scenarios in which target animals may appear, image annotation difficulties, and the comparability of DL model results across studies. Many studies have focused on single species, locations, or images with a high density of common target species. Assessments of models are potentially biased from using a single test set; many studies lack metrics to evaluate model efficiency, feasibility, and generalizability, and there are uncertainties regarding the optimal number of training images and required ground sample distance (GSD) for different animal detection tasks in drone imagery.</div><div>The potential applications of deep learning in wildlife monitoring, conservation, and ecological research using drone imagery are substantial. By enhancing the accuracy and efficiency of animal detection in imagery, this technology could contribute to the understanding and protecting animal populations. To expand the applicability of deep learning to diverse species, environments, and spatial scales, researchers should create standardised benchmark datasets and prioritise open collaboration and data sharing, which would aid in addressing the current challenges.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":null,"pages":null},"PeriodicalIF":5.8,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}