Ecological Informatics最新文献

{"title":"Comparative study of daily streamflow prediction based on coupling SWAT+ with interpretable machine learning algorithms","authors":"Chen Cao ,&nbsp;Miaomiao Ying","doi":"10.1016/j.ecoinf.2025.103406","DOIUrl":"10.1016/j.ecoinf.2025.103406","url":null,"abstract":"<div><div>In recent years, climate change has substantially affected the global water cycle, leading to an increase in the frequency and intensity of extreme hydrological events. Developing more accurate and efficient hydrological models is therefore essential for flood prevention, drought mitigation, and sustainable water resources management. In this study, four machine learning (ML) algorithms were coupled with SWAT+ to simulate streamflow in the Mishui River Basin (MRB). Both SWAT+–derived hydrological variables and raw meteorological observations were used as input features for the ML models, aiming to improve predictive performance. Additionally, the SHAP (SHapley Additive exPlanations) method was employed to quantify the contribution of different features to model predictions. During the validation period (2020−2023), the SWAT-Informer model exhibited the best performance, achieving R² and NSE values of 0.91 and 0.89, respectively. In contrast, improvements in streamflow prediction using DeepState and Bi-LSTM were less pronounced, with a notable performance decline during the testing period, likely due to the complexity of their multi-layer architectures. SHAP analysis revealed that precipitation was the most influential feature, contributing 29.1 % to the predictions. Moreover, SWAT+–derived outputs accounted for 64.9 % of the predictive power, highlighting the substantial value of SWAT+ in providing informative features for the ML algorithms. Overall, the four SWAT-ML coupled models outperformed the standalone SWAT+ model in streamflow prediction, demonstrating the considerable potential of ML techniques to enhance the performance of conceptual hydrological models such as SWAT+. Furthermore, the application of the SHAP method improved the interpretability of the models, fostering greater understanding, trust, and transparency for both researchers and decision-makers.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103406"},"PeriodicalIF":7.3,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145018930","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":"Transferability of stream benthic macroinvertebrate distribution models to drought-related conditions","authors":"Graciela Medina-Madariaga ,&nbsp;Hong Hanh Nguyen ,&nbsp;Jens Kiesel ,&nbsp;Kristin Peters ,&nbsp;Christian K. Feld ,&nbsp;Sonja C. Jähnig ,&nbsp;Yusdiel Torres-Cambas","doi":"10.1016/j.ecoinf.2025.103395","DOIUrl":"10.1016/j.ecoinf.2025.103395","url":null,"abstract":"<div><div>Freshwater ecosystems, which include rivers and streams, are increasingly threatened by climate change-induced extreme events, such as droughts, which disrupt hydrological processes and biodiversity. Species distribution models (SDMs) are essential for predicting species responses to environmental change. However, the transferability of SDMs beyond the conditions under which they were trained, such as from drought-free to drought-influenced scenarios, remains limited. These drought-influenced conditions represent novel environmental conditions for the models, posing challenges for accurate predictions. This study investigated the transferability of SDMs for freshwater macroinvertebrates from drought-free to drought-influenced conditions in a central German catchment via four modeling techniques (generalized linear models (GLMs); spatial stream networks (SSNs); random forests (RF) and maximum entropy (MaxEnt)) and species tolerance scores to assess how these factors independently and jointly affect model transferability. The transferability is evaluated on the basis of the accuracy gap (AUC gap/TSS gap), which quantifies performance differences between drought-free and drought conditions. Our findings reveal a marked decline in model performance under drought-influenced conditions, highlighting the challenges of predicting species distributions in novel environments. SDM transferability varied by species tolerance, with tolerant species exhibiting lower transferability. Additionally, SSN and RF models demonstrated slightly greater transferability for specific tolerances, suggesting their potential for modeling species responses under hydrological stress. Our study underscores the limitations of conventional SDMs in capturing species responses to extreme hydrological events, such as droughts, and advocates for integrating ecologically relevant predictors (such as stream connectivity) and accounting for species traits in SDMs to increase predictive accuracy in novel environmental scenarios.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103395"},"PeriodicalIF":7.3,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144922613","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":"How to minimize the annotation effort in aerial wildlife surveys","authors":"Giacomo May ,&nbsp;Emanuele Dalsasso ,&nbsp;Alexandre Delplanque ,&nbsp;Benjamin Kellenberger ,&nbsp;Devis Tuia","doi":"10.1016/j.ecoinf.2025.103387","DOIUrl":"10.1016/j.ecoinf.2025.103387","url":null,"abstract":"<div><div>Aircraft-based monitoring of wildlife is a popular way among conservation practitioners to obtain animal population counts over large areas. Nowadays, these aerial censuses are becoming increasingly scalable due to the advent of drone technology, which is frequently combined with deep learning-based image recognition. Yet, the annotation burden associated with training deep learning architectures remains a problem especially for commonly used bounding box detection models. Point-based density estimation- and localization models are cheaper to train, and often work better when the aerial imagery is recorded at an oblique angle. Beyond this, though, there currently is little consensus about which strategy to use for what kind of data. In this work, we address this knowledge gap and evaluate modifications to a state-of-the-art detection model (<span>YOLOv8</span>) that minimize labeling efforts by enabling it to work on point-annotated images. We study the effect of these adjustments on detection accuracy and extensively compare them to a localization architecture on four datasets consisting of nadir and oblique images. The goal of this paper is to offer wildlife conservationists practical advice on which of the recently proposed deep learning architectures to use given the properties of their images, as well as on the data properties that will maximize model performance independently of the architecture. We find that counting accuracy can largely be maintained at reduced annotation effort, that object detection technology outperforms the localization approach on nadir images, and that it shows competitive performance in the oblique setting. The images used to obtain the results presented in this paper can be found on <span><span>Zenodo</span><svg><path></path></svg></span> for all publicly available datasets, as well as all code necessary to reproduce our results was uploaded to <span><span>GitHub</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103387"},"PeriodicalIF":7.3,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144865179","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":"Spatiotemporal dynamics of the normalized difference vegetation index and its multidimensional drivers in a rapidly urbanizing coastal city: A case study of Lianyungang, China (2000−2023)","authors":"Xue Li ,&nbsp;Haihong He ,&nbsp;Dewei Wang ,&nbsp;Yiming Sun ,&nbsp;Yichen Qin ,&nbsp;Ke Wang ,&nbsp;Yu Han ,&nbsp;Jiabao Tang ,&nbsp;Wenli Qiao","doi":"10.1016/j.ecoinf.2025.103397","DOIUrl":"10.1016/j.ecoinf.2025.103397","url":null,"abstract":"<div><div>Coastal cities, as critical intersections of ecological integrity and human development, face escalating challenges from urbanization and climate change. This study investigates the spatiotemporal dynamics of the Normalized Difference Vegetation Index (NDVI) and its multidimensional drivers in Lianyungang, China (2000–2023), using multi-source remote sensing data and statistical methods such as Sen's slope and the Hurst exponent. Key findings reveal that NDVI exhibited significant growth (0.26 %/yr, <em>p</em> &lt; 0.01), with medium–high coverage areas (NDVI 0.5–0.7) expanding from 36.69 % to 63.29 %. BEAST identified two critical changepoints in 2003 (probability: 21.41 %) and 2013 (probability: 23.96 %), delineating three phases. The early period (2000−2003) exhibited stable growth, with 62.3 % of areas exhibiting positive NDVI trends and widespread sustainability (weak + strong sustainability: 100 %). However, rapid urbanization during 2004–2013 triggered vegetation degradation, as evidenced by negative NDVI trends in 68.6 % of areas and a surge in anti-persistence (strong + weak anti-sustainability: 23.5 %). Post-2013, policy interventions facilitated recovery, with 56.0 % of areas regaining positive NDVI trends and strong sustainability increasing to 17.3 %. Geodetector revealed that anthropogenic drivers dominated NDVI dynamics. Land use remained the strongest factor (q = 0.306 in 2000 and 0.228 in 2020), while nighttime light (NTL) showed the sharpest increase (q = 0.082 to 0.194). Conversely, precipitation's explanatory power declined markedly (q = 0.060 to 0.028). This study provides a scientific basis for ecological management in coastal cities, highlighting the role of policy in balancing urbanization and sustainability.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103397"},"PeriodicalIF":7.3,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144893266","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":"AntPi: A Raspberry Pi based edge–cloud system for real-time ant species detection using YOLO","authors":"Lorenzo Palazzetti ,&nbsp;Daniele Giannetti ,&nbsp;Antonio Verolino ,&nbsp;Donato A. Grasso ,&nbsp;Cristina M. Pinotti ,&nbsp;Francesco Betti Sorbelli","doi":"10.1016/j.ecoinf.2025.103383","DOIUrl":"10.1016/j.ecoinf.2025.103383","url":null,"abstract":"<div><div>Ant detection is essential for ecological research, offering insights into biodiversity, habitat health, and environmental change. Traditional detection techniques rely on manual sampling methods, which are labor-intensive and time-consuming. Recent advances in autonomous, vision based systems show promise for insect monitoring, yet no dedicated, field ready solution exists for ant identification. In this work, we present <span>AntPi</span>, a deep learning based system for real-time detection and classification on a Linux development board. To the best of our knowledge, the system is trained on the first dedicated dataset for arboricolous ants, comprising five species and one morphotype, sourced from citizen science contributions and direct field captures. Our approach employs the “You Only Look Once” (YOLO) framework for efficient object detection, augmented with environmental sensors to enable correlation between climatic variables and ant activity. To evaluate performance and robustness, we compare <span>AntPi</span> with an alternative configuration, including controlled experiments using background-only images with artificial ant-like noise, and introduce a novel robustness indicator to assess reliability under realistic conditions. Experimental results demonstrate strong detection performance and confirm the feasibility of automated, in-field ant monitoring.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103383"},"PeriodicalIF":7.3,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144886455","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":"Declining net carbon sequestration of west-central Indian ecosystem in response to frequently occurring drought: Inference from satellite measurements and modeling","authors":"Aparnna Ravi ,&nbsp;Dhanyalekshmi Pillai ,&nbsp;Monish Vijay Deshpande","doi":"10.1016/j.ecoinf.2025.103386","DOIUrl":"10.1016/j.ecoinf.2025.103386","url":null,"abstract":"<div><div>Extreme weather events significantly impact vegetation dynamics, with droughts becoming increasingly frequent and adversely affecting plant growth and carbon sequestration. Understanding vegetation responses to such events is essential for effective climate change mitigation, as it directly influences the capacity of ecosystems to absorb Carbon Dioxide (CO₂). This study investigates the impact of the 2016 drought in west-central India on vegetation productivity and ecosystem carbon exchange. We utilize satellite-based reflectance products, including Vegetation Indices (VIs) and Solar-Induced chlorophyll Fluorescence (SIF), alongside carbon flux estimates from vegetation models such as a subset of the TRENDY ensemble, Vegetation Photosynthesis and Respiration Model (VPRM), and FLUXNET-X. VIs and SIF were analyzed to identify the vegetation responses to drought and to assess model performance in the absence of extensive ground-based observations. Our findings indicate that the region’s carbon uptake capacity declined by 30 Tg C season⁻¹ due to the 2016 drought. The region persisted as a net carbon source annually, with 75% of the source contribution from the drought period. However, ecosystem respiration remained largely unaffected. The drought significantly suppressed vegetation growth, with deciduous vegetation experiencing the most severe impact. When croplands recovered more quickly, shrublands showed a slower recovery, and deciduous vegetation exhibited the longest delay in post-drought recovery. The vegetation models employed generated varied ecosystem carbon fluxes, but most of them showed similar levels of responses to the drought. The uncertainties in these models emphasize the need for better representation of ecological processes and high-density observations in this region. The outcome of this study can help in making important policy decisions at the national level, especially for managing ecosystems and maintaining carbon storage during extreme weather events.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103386"},"PeriodicalIF":7.3,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144878478","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":"Estimating chlorophyll content in tea leaves using spectral reflectance and deep learning methods","authors":"Yuta Tsuchiya ,&nbsp;Yuhei Hirono ,&nbsp;Rei Sonobe","doi":"10.1016/j.ecoinf.2025.103399","DOIUrl":"10.1016/j.ecoinf.2025.103399","url":null,"abstract":"<div><div>Accurate estimation of chlorophyll content in tea leaves is essential for evaluating plant health, managing fertilization, and optimizing harvest timing in precision agriculture. This study investigates the use of hyperspectral reflectance data (400–850 nm, 5 nm intervals; 91 bands) to estimate chlorophyll content in tea leaves (<em>Camellia sinensis</em>) using three deep learning models: a one-dimensional convolutional neural network (1D–CNN) tailored for spectral regression, a vision transformer (ViT) adapted for one-dimensional inputs, and a self-supervised learning (SSL) model with regression. The key innovation of this study is the introduction of a self-supervised learning framework specifically adapted for spectral data, in which an autoencoder is first trained on unlabeled spectra to learn compact and noise-tolerant representations. These pretrained features are then used in a downstream regression task to predict chlorophyll content, allowing effective use of limited labeled data. To our knowledge, this is the first application of SSL in chlorophyll estimation using high–resolution leaf–level spectral measurements. Among the three models, the SSL approach achieved the highest accuracy, with a root mean square error (RMSE) of 3.33 μg/cm², outperforming both the 1D–CNN (5.05 μg/cm²) and ViT (4.28 μg/cm²). These findings demonstrate that SSL is particularly effective for capturing subtle spectral patterns and improving prediction performance, especially when labeled data are scarce. This study highlights the potential of combining hyperspectral sensing with advanced representation learning to non–destructively monitor chlorophyll dynamics in tea cultivation, supporting more sustainable and data–driven agricultural practices.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103399"},"PeriodicalIF":7.3,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144886456","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":"A two-stage model for enhanced mango leaf disease detection using an innovative handcrafted spatial feature extraction method and knowledge distillation process","authors":"Mohammad Manzurul Islam ,&nbsp;Mst. Nasrat Jahan Niva ,&nbsp;Abdullahi Chowdhury ,&nbsp;Saleh Masum ,&nbsp;Rifat Ara Shams ,&nbsp;Taskeed Jabid ,&nbsp;Md. Sawkat Ali ,&nbsp;Md. Mostofa Kamal Rasel ,&nbsp;Muhammad Firoz Mridha","doi":"10.1016/j.ecoinf.2025.103365","DOIUrl":"10.1016/j.ecoinf.2025.103365","url":null,"abstract":"<div><div>The economic stability of many countries is closely tied to agriculture, where crop quality and yield are heavily affected by weather conditions and disease control. Unpredictable climate patterns and plant diseases remain major challenges, emphasizing the need for more reliable disease detection methods. Traditionally, plant disease identification has relied on visual examination, a method that is often inaccurate. To address this, our study proposes a two-stage model for improved disease detection in mango leaves. In the first stage, we implement an innovative, block-based feature extraction technique using Local Directional Patterns (LDP) and Local Directional Pattern variance (LDPv) on a comprehensive dataset, MangoLeafBD, consisting of 4000 images, achieving satisfactory results in terms of detection accuracy, sensitivity, specificity, and false negative rate. In the second stage, we introduce a Knowledge Distillation (KD) process to further enhance model performance by transferring knowledge from a larger teacher model to a smaller student model. Our results demonstrate significant advancement, with the KD-enhanced model achieving an improvement in detection accuracy from 89.2% to 95.6%, sensitivity from 7.8% to 4.1%, and specificity from 71.2% to 88.9% for Anthracnose disease. Similar improvements were observed in detecting other diseases, making our approach a more robust and efficient solution for mango plant disease detection.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103365"},"PeriodicalIF":7.3,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144893264","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":"A high resolution spatial modelling framework for landscape-level, strategic conservation planning","authors":"T. Foxley ,&nbsp;P. Lintott ,&nbsp;S. Stonehouse ,&nbsp;J. Flannigan ,&nbsp;E.L. Stone","doi":"10.1016/j.ecoinf.2025.103363","DOIUrl":"10.1016/j.ecoinf.2025.103363","url":null,"abstract":"<div><div>Development is needed to improve living standards globally but poses a threat to many species through habitat loss and fragmentation. There is often a legal requirement to ensure new development does not negatively impact protected species and the habitats they depend on, however planners are unable to make informed decisions without a detailed understanding of how species use the landscape. The aim of this study was to develop a spatial modelling framework for protecting biodiversity in the planning process. Using habitat suitability and landscape connectivity modelling we aimed to produce high resolution mapping outputs that can inform development decisions. We illustrate our approach with a detailed case study of a species of conservation concern, the greater horseshoe bat (<em>Rhinolophus ferrumequinum</em>), in Somerset, UK. We gathered fine scale data on <em>R. ferrumequinum</em> habitat use with GPS telemetry, mapped habitat using a high resolution, satellite derived land classification, and built a detailed vegetation map with LIDAR. With these data we built models of habitat suitability and landscape connectivity, validated model predictions with an independent dataset, and generated a number of high resolution maps. We present a detailed case study to explore how different mapping outputs can guide development decisions. We propose that robust tools such as integrated spatial modelling should be central to the planning process; our framework can act as a template for implementing this.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103363"},"PeriodicalIF":7.3,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144878480","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":"Whale Vision: A tool for identifying sperm whales and other cetaceans by their flank or fluke","authors":"Sammie Fuller ,&nbsp;Silvia Maggi ,&nbsp;Barbara Mussi ,&nbsp;Theodore Kypraios ,&nbsp;Michael P. Pound","doi":"10.1016/j.ecoinf.2025.103384","DOIUrl":"10.1016/j.ecoinf.2025.103384","url":null,"abstract":"<div><div>The Mediterranean sperm whale (<em>Physeter macrocephalus</em>) is classified as endangered, and effective conservation strategies require accurate knowledge of individual whales’ movements and behaviours. However, non-invasive techniques to track individuals require photographic identification of single subjects, a time-consuming process that requires manual curation. This study presents machine learning tools developed to automate the identification of individual sperm whales from photographic data. While fluke images are traditionally used for cetacean identification, this research extends deep learning-based identification to include flank images, a novel approach for sperm whales. Two Residual Neural Network models were trained using a contrastive learning process to distinguish individuals from either fluke or flank images, representing each whale as a point in a 128-dimensional latent space for fast re-identification. Evaluation on the Oceanomare Delphis dataset demonstrated identification accuracies of 81.2% for fluke images and, notably, 76.5% for flank images, highlighting the effectiveness and potential of flank-based identification. A user-friendly interface was developed to facilitate practical application, enabling both the identification of known individuals and the incorporation of new subjects. The methods and tools presented are adaptable to other cetacean species, offering a scalable and non-invasive solution to support conservation efforts.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103384"},"PeriodicalIF":7.3,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144904657","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}