Ecological Informatics最新文献

{"title":"Ensemble-based forecasting of wildfire potentials using relative index in Gangwon Province, South Korea","authors":"Sang Yeob Kim ,&nbsp;Changhyun Jun ,&nbsp;Wooyoung Na","doi":"10.1016/j.ecoinf.2025.103021","DOIUrl":"10.1016/j.ecoinf.2025.103021","url":null,"abstract":"<div><div>Wildfire indices have been widely used to assess wildfire potential under varying climate conditions. However, their region-specific applicability remains limited due to inherent incompatibilities among various indices. This study proposes post-processing procedures for wildfire forecasting by applying statistical index-merging methods to enhance the utility of conventional wildfire indices in forested regions of South Korea. Accordingly, 126 wildfire cases from 2014 to 2023 are analyzed, and the performance of conventional indices is assessed individually and in combination using three merging methods: simple averaging, variance-covariance (VC), and triple collocation (TC). The forecasting capabilities of individual and merged indices are evaluated using hit/miss metrics, specifically the probability of detection and false alarm ratio. The results reveal that ensemble merging techniques can partially enhance the forecasting performance of conventional indices that are otherwise suboptimal for the target region. The forecasting performance of both individual and merged indices is higher in forested areas, highlighting vegetation as a significant factor in wildfires. Notably, the forecasting performance of the VC method, which incorporates inter-index correlations, is superior to that of TC, which does not account for these correlations. This robust performance is evident regardless of wildfire size or severity. Furthermore, the findings underscore the critical role of climate conditions in wildfire detection, highlighting the need to address the effects of climate change. Consequently, the application of statistical index-merging methods enhances wildfire forecasting accuracy in Gangwon Province, South Korea, surpassing that of conventional indices alone, despite their global reliability.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103021"},"PeriodicalIF":5.8,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101438","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 comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study","authors":"Guilherme Cassales ,&nbsp;Serajis Salekin ,&nbsp;Nick Lim ,&nbsp;Dean Meason ,&nbsp;Albert Bifet ,&nbsp;Bernhard Pfahringer ,&nbsp;Eibe Frank","doi":"10.1016/j.ecoinf.2025.103014","DOIUrl":"10.1016/j.ecoinf.2025.103014","url":null,"abstract":"<div><div>As a dominant terrestrial ecosystem, forests play a pivotal role, which is substantially challenged by climate extremes. At the same time, the practice of patient science to investigate and understand different intricate climate-driven phenomena is no longer an option. On the other hand, recent technological advancements enable scientists to simultaneously collect and analyse a large volume of complex data. High-resolution tree stem radius measurements and predictive simulation through machine learning algorithms offer powerful opportunities for understanding these dynamics. However, when these machine learning methods are applied without careful consideration of data quality, model biases, and other critical factors, their potential is often compromised. In this study, we aimed to evaluate four Deep Learning algorithms (namely CNN, LSTM, Transformer, and ETSFormer), using automatically measured and high temporal resolution tree stem radius data. We explore the complexities of handling voluminous and authentic datasets required by these algorithms. Initial experiments show that it is possible to achieve an MAE as small as 0.0026 mm on the full data. Furthermore, our study delves into the temporal resolution of data, demonstrating the feasibility of using reduced datasets without compromising accuracy levels. Our best result showed that a reduction of 97 % in collection events increases the MAE by only 6 % with the LSTM model, demonstrating that resource use optimisation can be achieved by slightly reducing the temporal resolution of data collection with marginal error increase. This also shows that LSTM can effectively capture longer-term and complex dependencies, which indicates promising future work with additional environmental data.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103014"},"PeriodicalIF":5.8,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097678","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":"Cross-scale observation of riparian vegetation: Testing the potential of satellite-UAV-Field integrated observations for large-scale herbaceous species","authors":"Weiwei Jiang ,&nbsp;Chenyu Li ,&nbsp;Henglin Xiao","doi":"10.1016/j.ecoinf.2025.103016","DOIUrl":"10.1016/j.ecoinf.2025.103016","url":null,"abstract":"<div><div>In response to the dual challenges of global change and human activities, riverine ecological management urgently requires a deep understanding of the large-scale ecological processes of dominant vegetation populations along riverbanks. However, current knowledge in this field remains inadequate. This is primarily due to the inherent conflict between pursuing large-scale coverage and acquiring fine-grained information in riverine vegetation observation, especially in complex environments where dammed rivers are predominantly covered by herbaceous plants. To address this issue, we innovatively propose a cross-scale information fusion framework that effectively integrates satellite, unmanned aerial vehicle (UAV), and ground observation data, successfully breaking through the bottleneck of large-scale herbaceous species information extraction along riverbanks. This framework employs Support Vector Machine (SVM) technology to eliminate non-vegetation interferences, adopts object-based rather than traditional pixel-based units for supervised classification, and incorporates rich spatial information beyond spectral data, significantly enhancing the accuracy of dominant species identification with F1 scores improved by 10 % to 50 %. This advancement not only validates the effectiveness of the cross-scale fusion method but also reveals qualitative consistency and quantitative differences (1 %–11 %) in vegetation population coverage assessments across different observation methods, providing new insights into understanding vegetation distribution heterogeneity. More importantly, the application potential of this framework extends far beyond riverine ecological management. It demonstrates broad application prospects in various fields such as forest monitoring, agricultural management, invasive and endangered species monitoring, and biodiversity research. For ecological managers and policymakers, this means that more precise large-scale species distribution and dynamic change data can be used for systematic planning and precise interventions at the regional scale.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103016"},"PeriodicalIF":5.8,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101497","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":"Efficient wildlife monitoring: Deep learning-based detection and counting of green turtles in coastal areas","authors":"Naoya Noguchi ,&nbsp;Hideaki Nishizawa ,&nbsp;Taro Shimizu ,&nbsp;Junichi Okuyama ,&nbsp;Shohei Kobayashi ,&nbsp;Kazuyuki Tokuda ,&nbsp;Hideyuki Tanaka ,&nbsp;Satomi Kondo","doi":"10.1016/j.ecoinf.2025.103009","DOIUrl":"10.1016/j.ecoinf.2025.103009","url":null,"abstract":"<div><div>Drones have recently been used to assess wildlife populations and their abundance. The automatic detection of target animals in drone footage enables efficient abundance estimation. However, accurately detecting animals remains challenging, especially in complex field environments. Moreover, automating the tracking of individuals across consecutive images and counting them along transect lines is necessary to apply drones to line-transect surveys. In this study, deep-learning-based You Look Only Once, Version 7 (YOLOv7) models were developed to automatically detect green turtles (<em>Chelonia mydas</em>) in Japanese coastal areas featuring coral reefs and seagrass beds. Drone footage yielded 103,296 annotated images of green turtles. The model was trained and validated using 78 % and 22 % of the images. The best model performances were 0.848, 0.853, and 0.922 for precision, recall, and mean average precision at the threshold of the intersection over union = 0.5, respectively. Then, the BoT-SORT object-tracking algorithm was implemented to track green turtles detected using the YOLOv7 model, and the counting of individuals was automated. When this automatic counting model was tested using eight drone footage clips, green turtles at the sea surface were successfully tracked and counted (<em>n</em> = 3/3); however, the performance in counting underwater green turtles was relatively poor (<em>n</em> = 27/59). The reduced performance might be attributable to accumulated errors in detecting green turtles while processing numerous images in the footage (approximately 60 fps). Nonetheless, relatively high precision was achieved by reducing false positives in complex coastal areas. The methods in this study should enhance the efficiency of long-term wildlife monitoring programs.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103009"},"PeriodicalIF":5.8,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101510","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":"Key traits in functional trait networks can identify the temporal structure of fish communities in coastal waters","authors":"Jianyu Zou ,&nbsp;Xiaozhuang Zhang ,&nbsp;Yupeng Ji ,&nbsp;Ying Xue ,&nbsp;Chongliang Zhang ,&nbsp;Yiping Ren ,&nbsp;Binduo Xu","doi":"10.1016/j.ecoinf.2025.103017","DOIUrl":"10.1016/j.ecoinf.2025.103017","url":null,"abstract":"<div><div>Traditional analyses of fish community structure based on species composition and abundance ignore many functional attributes of fish species. The composition of fish functional traits can reflect the attributes and functions of the fish community. However, little is known about which functional traits can reflect the temporal structure of fish communities. In this study, we employed trait network analysis, principal coordinates analysis, cluster analysis, and Spearman correlation analysis to construct a functional trait network, identify key traits, examine interannual changes in functional trait composition, and explore changes in the temporal structure and functions of fish communities in Haizhou Bay and its adjacent waters, Yellow Sea, China, from 2013 to 2022 based on species data collected from bottom trawl surveys. The composition of functional traits has changed significantly over the past decade, indicating significant alterations in the attributes and functions of the fish community within Haizhou Bay. All 95 strong correlations in the functional trait network were positive, suggesting that competitive interaction was very weak in the entire community assembly. The temporal structure of the fish community was determined based on 10 key traits, and it was consistent with that identified via species-based analysis. The highly significant positive correlations between the similarity matrices of the key traits and the species similarity matrix indicated that interannual differences in key traits could well represent discrepancies in the fish community structure in different years. Our study revealed that key traits in the functional trait network could be used to identify the temporal structure of fish communities and revealed differences in the functions of different temporal components.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103017"},"PeriodicalIF":5.8,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097689","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":"Study on efficient recognition and accurate localization method of waste plastic bottles based on deep learning","authors":"Shilong Xie,&nbsp;Hu Wu,&nbsp;Wenjie Mao,&nbsp;Xianlong Chu,&nbsp;Yixing Meng,&nbsp;Xianhai Yang","doi":"10.1016/j.ecoinf.2025.103020","DOIUrl":"10.1016/j.ecoinf.2025.103020","url":null,"abstract":"<div><div>As a vital component of ecologically sustainable development, the effective recovery and reuse of waste plastic bottles is essential for environmental protection and resource recycling. Given the varying recycling values of plastic bottles based on their colors, precise sorting and recycling are particularly important. Traditional manual sorting methods face challenges such as low efficiency and high costs. In contrast, machine vision-based image recognition technology offers a more efficient solution for classifying and recovering waste plastic bottles, with classification recognition and target positioning being critical technologies for the optimal use of ecological resources. This study introduces a deep learning approach for identifying and locating waste plastic bottles, utilizing the reversible column network (RevCol) as the backbone to prevent information loss. A lightweight combined decoupling head is designed to minimize computational load while enhancing accuracy. The Weighted Intersection over Union version 3 (WIoU v3) loss function is incorporated to improve detection performance. By leveraging depth information from an infrared camera alongside RGB image mapping, the method achieves recognition and three-dimensional positioning. Experimental results indicate that the proposed model outperforms traditional models, with a 36.39 % reduction in parameters and a 50.62 % decrease in computational requirements, while accuracy and recall rates improve by 4.56 % and 12.14 %, respectively. Additionally, mAP50 and mAP50–95 values increase by 5.86 % and 3.89 %, and the recognition speed reaches 62 FPS, a 51.22 % improvement, meeting real-time detection needs. Experiments conducted on a deep learning-UR5 robot platform demonstrate high recognition accuracy and sorting success rates in actual waste plastic bottle sorting scenarios. The promotion and implementation of this method will significantly enhance the recycling of waste plastic resources and contribute to the protection and sustainable development of the ecological environment.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103020"},"PeriodicalIF":5.8,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101498","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":"Simulation of thermal stratification and water temperature dynamics in the Joumine reservoir (Tunisia)","authors":"Haifa Madyouni ,&nbsp;Pol Magermans ,&nbsp;Sihem Benabdallah ,&nbsp;Romdhane Mohamed Saleh ,&nbsp;Hamadi Habaieb ,&nbsp;Jean François Deliege","doi":"10.1016/j.ecoinf.2025.103012","DOIUrl":"10.1016/j.ecoinf.2025.103012","url":null,"abstract":"<div><div>Thermocline stratification has become increasingly important under climate change conditions, impacting the water bodies' quality, by changing the epilimnion thickness, particularly the biological quality related to phytoplankton communities. Advanced modeling techniques based on the new Derived EOLE Joumine Model (DEOLE-J) and metaheuristic approaches were used to model thermocline stratification in the Joumine reservoir in the North of Tunisia. Relative Water Column Stability (RWCS) and thermocline parameters such as thermocline depth and strength index (TSI) were used to assess the water temperature profile and the impact of the thermocline on the phytoplankton community distribution and abundance. Monthly samplings were conducted at eight gauging stations from May 2021 to August 2021. Water samples were collected to measure physical and biological parameters. Joumine's thermal stratification can be divided into three periods: Mixing, Formative, and Stable. During the Mixing period, TSI and air temperature had a significant negative correlation. Similarly, significant negative correlations were observed between TSI, air temperature, and RWCS during the Formative period. Our results reveal that weaker stratification in spring is primarily driven by increased inflow discharge, while summer stratification intensifies, creating sharp thermal gradients. The model successfully captures seasonal thermocline fluctuations and shows that wind speed plays a critical role in regulating vertical mixing. However, moderate wind speeds typical of the Joumine region have limited impact on the deeper layers of the reservoir, particularly during summer. A comparison of model estimates and measured data indicates a bias due to distant meteorological stations and the exclusion of horizontal fluxes, such as water withdrawal and throughflow. Despite these limitations, the DEOLE-J provides valuable insights into the thermal dynamics of reservoirs, showing that prolonged stratification periods reduce vertical mixing and nutrient circulation, potentially degrading water quality. These findings have significant implications for water quality management, particularly in the context of climate change, where extended stratification periods are expected leading to exacerbate water quality issues. Future research should explore two-dimensional models to enhance temperature estimation accuracy and include horizontal fluxes. Keywords: DEOLE-J, Thermal stratification, Joumine reservoir, Thermocline strength, Thermocline depth, phytoplankton distribution, water temperature.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103012"},"PeriodicalIF":5.8,"publicationDate":"2025-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143102084","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":"Optimized SVR model for predicting dissolved oxygen levels using wavelet denoising and variable reduction: Taking the Minjiang River estuary as an example","authors":"Peng Zhang ,&nbsp;Xinyang Liu ,&nbsp;Huiru Zhang ,&nbsp;Chengchun Shi ,&nbsp;Gangfu Song ,&nbsp;Lei Tang ,&nbsp;Ruihua Li","doi":"10.1016/j.ecoinf.2025.103007","DOIUrl":"10.1016/j.ecoinf.2025.103007","url":null,"abstract":"<div><div>Adequate dissolved oxygen (DO) is critical for the maintenance of aquatic ecosystems. However, predicting DO levels in regions with complex hydrological variations remains challenging. This study presents a novel DO prediction model using the Minjiang River estuary as an example by integrating advanced machine learning techniques. Key influencing factors were identified using the Maximum Information Coefficient (MIC) and noise was reduced using Wavelet Denoising (WD). Support Vector Regression (SVR) parameters were optimized using Particle Swarm Optimization (PSO), culminating in an optimized WD-MIC-PSO-SVR model for DO prediction. The results showed that the MIC effectively identified the key influencing factors of DO. Compared with the unoptimized SVR model, the proposed model achieved higher accuracy, R² and NSE reached 0.91 and 0.83, respectively, while the MAE and RMSE values were reduced by 67 % and 44 %, respectively, affirming its applicability for real-time DO prediction. This study contributes to water environment protection by providing an effective solution for DO modeling in regions with substantial hydrological changes. The integrated WD-MIC data processing method shows promising potential in reducing model errors and lowering water monitoring costs by focusing on highly correlated variables.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103007"},"PeriodicalIF":5.8,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101603","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":"Modeling the effects of climate change scenarios on the potential distribution of Vespa crabro Linnaeus, 1758 (Hymenoptera: Vespidae) in a Mediterranean biodiversity hotspot","authors":"Erika Bazzato ,&nbsp;Arturo Cocco ,&nbsp;Emanuele Salaris ,&nbsp;Ignazio Floris ,&nbsp;Alberto Satta ,&nbsp;Michelina Pusceddu","doi":"10.1016/j.ecoinf.2025.103006","DOIUrl":"10.1016/j.ecoinf.2025.103006","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Climate change poses unprecedented challenges to ecosystems and species, particularly in biodiversity hotspots like the European-Mediterranean regions. The ecological consequences are not easily discernible. Although the influence of climate on plants and vertebrates has been extensively studied, its impact on alien insects, especially social wasps, remains underexplored. To address this gap, this study identifies climatically suitable habitats for &lt;em&gt;Vespa crabro&lt;/em&gt; under current conditions, projects its potential future distribution, and assesses potential range shifts driven by climate change to guide monitoring programs and management measures. We focused on Sardinia, a Mediterranean island with a heterogeneous morphological, geological, and climatic pattern, where &lt;em&gt;V. crabro&lt;/em&gt; was accidentally introduced.&lt;/div&gt;&lt;div&gt;We used 316 verified citizen science occurrences, high-resolution bioclimatic variables (40 × 40 m) specifically developed for the island, and two future climate and socio-economic scenarios for two temporal horizons (2040 and 2060) to model climatic suitability using an ensemble framework with three machine learning algorithms: Artificial Neural Networks (ANN), Generalized Boosting Model (GBM), and Random Forest (RF). To ensure reliable predictions, we addressed several technical challenges, including correcting for sampling biases and spatial autocorrelation. The individual models were weighted based on spatial cross-validation performance and combined to obtain the ensemble model.&lt;/div&gt;&lt;div&gt;Performance varied among 150 individual models (3 algorithms × 10 replicates × 5 folds), depending on the algorithms, replicates, and subsets selected for training and testing. RF demonstrated the highest predictive performance, outperforming ANN and GBM. The ensemble model achieved even higher predictive accuracy with Receiver Operating Characteristics (ROC) = 0.95 ± 0.02 and True Skill Statistic (TSS) = 0.78 ± 0.06.&lt;/div&gt;&lt;div&gt;Key factors influencing &lt;em&gt;V. crabro&lt;/em&gt; distribution included Annual Mean Temperature, Maximum Temperature of Warmest Month, Temperature Annual Range, Precipitation of Driest Month, and Precipitation Seasonality. Currently, climatically suitable habitats are predominantly in the northern part of the island, in most coastal areas, and in specific inland regions, especially those near or inside mountainous areas. Future projections indicate a distribution range contraction by the 2040s and 2060s, primarily driven by extreme variability in precipitation and rising temperatures approaching the species' thermal tolerance limits.&lt;/div&gt;&lt;div&gt;Our study demonstrates the value of integrating citizen science data, high-resolution climate data, and advanced modeling techniques to understand and manage alien species in the context of climate change. It highlights the importance of fine-scale studies to complement broader analyses, providing deeper insight into the impacts of climate change on species distribut","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103006"},"PeriodicalIF":5.8,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101495","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":"Harmonizing remote sensing and ground data for forest aboveground biomass estimation","authors":"Ying Su ,&nbsp;Zhifeng Wu ,&nbsp;Xiaoman Zheng ,&nbsp;Yue Qiu ,&nbsp;Zhuo Ma ,&nbsp;Yin Ren ,&nbsp;Yanfeng Bai","doi":"10.1016/j.ecoinf.2025.103002","DOIUrl":"10.1016/j.ecoinf.2025.103002","url":null,"abstract":"<div><div>Accurate aboveground biomass (AGB) estimation is crucial for evaluating management and conservation policy of forests. However, the complexity of forest ecosystems and the diversity of geography bring great challenges to traditional biomass estimation methods. This study aims to develop an optimized AGB estimation framework that integrates heterogeneous data sources (i.e., ground survey data, National Forest Continuous Inventory (NFCI) data, and both active and passive remote sensing data) to enhance estimation accuracy and address the needs of future satellite missions and forest monitoring efforts. Using Longyan City, Fujian Province, China, as a case study, we construct a machine learning-based AGB estimation framework and generate high-resolution AGB spatial distribution maps through stepwise variable selection, hyperparameter optimization, and incremental integration of data sources. The effectiveness of this approach was demonstrated by a 0.67 increase in the correlation coefficient <em>R</em>², a 43.57 % reduction in the root mean square error (RMSE), and a 68.00 % reduction in the mean square error (MSE) achieved through the optimal combination of data sources. The optimization framework not only significantly improves AGB estimation accuracy but also facilitates the identification of key areas for afforestation through the generated spatial distribution map, offering a scientific foundation for targeted forest management and ecological restoration. This study highlights the potential of combining heterogeneous data sources with machine learning techniques, providing a scalable solution for future forest monitoring tasks.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"86 ","pages":"Article 103002"},"PeriodicalIF":5.8,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143101602","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}