Ecological Informatics最新文献_第4页

A2ANet: Real-time detection of floating marine debris using atrous convolution and channel attention A2ANet：利用亚光卷积和航道关注实时检测海洋浮物

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-21 DOI: 10.1016/j.ecoinf.2026.103620

Badiu Badams , Usman Ullah Sheikh , Norhaliza Abdul Wahab , Syed A.R. Abu Bakar , Muhammad I. Masud , Mohammed Khouj , Urooj Waheed , Zeeshan Ahmad Arfeen , Kam Meng Goh

Floating marine debris such as plastics, cans, and discarded packaging has become one of the most persistent threats to aquatic ecosystems and coastal sustainability. Detecting and tracking this debris in real time is vital for protecting biodiversity and guiding cleanup and policy actions. In this study, we introduce A2ANet, a lightweight deep learning framework that combines multi-scale atrous convolutions and Enhanced Channel Attention (ECA) to detect small, submerged, and visually ambiguous debris under challenging aquatic conditions. These mechanisms-expanding the receptive field and highlighting salient cues-reduce errors from reflections, glare, and clutter in aquatic scenes.

A2ANet was evaluated on two datasets: a newly developed six-class dataset (D_six) representing real-world river conditions, and the publicly available FloW-Img benchmark. The model achieved [email protected] of 0.841 and 0.892 on D_six and FloW-Img datasets, respectively, with inference time as low as 39 ms/image. Beyond detection performance, by enabling automated and frequent monitoring, A2ANet provides actionable insights for mapping pollution, tracking trends, and supporting ecosystem management. The framework offers a practical pathway toward intelligent aquatic observation systems aligned with Sustainable Development Goal 14: Life Below Water.

All code and datasets are openly available (see Data Availability).

漂浮的海洋垃圾，如塑料、易拉罐和废弃包装，已成为水生生态系统和沿海可持续性的最持久威胁之一。实时检测和跟踪这些碎片对于保护生物多样性和指导清理和政策行动至关重要。在这项研究中，我们介绍了A2ANet，这是一个轻量级的深度学习框架，结合了多尺度亚图卷积和增强通道注意（ECA），可以在具有挑战性的水中条件下检测小的、淹没的和视觉上模糊的碎片。这些机制——扩大接受域和突出突出的线索——减少了水中场景中反射、眩光和杂乱造成的错误。A2ANet在两个数据集上进行了评估：一个是新开发的代表真实河流状况的六类数据集（D_six），另一个是公开可用的FloW-Img基准。该模型在d_6和FloW-Img数据集上分别实现了[email protected] 0.841和0.892，推理时间低至39 ms/image。除了检测性能之外，通过实现自动化和频繁的监测，A2ANet还为绘制污染地图、跟踪趋势和支持生态系统管理提供了可操作的见解。该框架为实现符合可持续发展目标14：水下生命的智能水生观测系统提供了一条切实可行的途径。所有代码和数据集都是公开可用的（参见数据可用性）。

{"title":"A2ANet: Real-time detection of floating marine debris using atrous convolution and channel attention","authors":"Badiu Badams , Usman Ullah Sheikh , Norhaliza Abdul Wahab , Syed A.R. Abu Bakar , Muhammad I. Masud , Mohammed Khouj , Urooj Waheed , Zeeshan Ahmad Arfeen , Kam Meng Goh","doi":"10.1016/j.ecoinf.2026.103620","DOIUrl":"10.1016/j.ecoinf.2026.103620","url":null,"abstract":"<div><div>Floating marine debris such as plastics, cans, and discarded packaging has become one of the most persistent threats to aquatic ecosystems and coastal sustainability. Detecting and tracking this debris in real time is vital for protecting biodiversity and guiding cleanup and policy actions. In this study, we introduce A2ANet, a lightweight deep learning framework that combines multi-scale atrous convolutions and Enhanced Channel Attention (ECA) to detect small, submerged, and visually ambiguous debris under challenging aquatic conditions. These mechanisms-expanding the receptive field and highlighting salient cues-reduce errors from reflections, glare, and clutter in aquatic scenes.</div><div>A2ANet was evaluated on two datasets: a newly developed six-class dataset (D_six) representing real-world river conditions, and the publicly available FloW-Img benchmark. The model achieved [email protected] of 0.841 and 0.892 on D_six and FloW-Img datasets, respectively, with inference time as low as 39 ms/image. Beyond detection performance, by enabling automated and frequent monitoring, A2ANet provides actionable insights for mapping pollution, tracking trends, and supporting ecosystem management. The framework offers a practical pathway toward intelligent aquatic observation systems aligned with Sustainable Development Goal 14: Life Below Water.</div><div>All code and datasets are openly available (see Data Availability).</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103620"},"PeriodicalIF":7.3,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146025988","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}

引用次数: 0

Laboratory-based hyperspectral reflectance analysis for phytoplankton species identification 基于实验室的浮游植物物种鉴定的高光谱反射分析

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-21 DOI: 10.1016/j.ecoinf.2026.103626

R. Bentivogli , L. Pezzolesi , N. Caputo , B. Casarotto , S. Silvestri

Monitoring phytoplankton communities is essential for assessing ecosystem health and detecting harmful algal blooms (HABs). Hyperspectral imaging has emerged as a promising tool to discriminate among microalgal species based on their unique reflectance signatures. This study presents a laboratory spectral analysis of five phytoplankton species, including bloom-forming and toxin-producing taxa common in coastal waters. Reflectance spectra were measured at multiple cell concentrations and analyzed using two normalization approaches, second- and fourth-derivative transformations, and dimensionality reduction techniques including principal component analysis (PCA) and linear discriminant analysis (LDA).

Our results demonstrate that specific spectral features, particularly in the 470–500 nm and 620–680 nm ranges, enable species-level discrimination. PCA and LDA effectively enhanced separability by reducing spectral redundancy and emphasizing class features. We further applied linear spectral unmixing (LSU) to estimate fractional species abundances in synthetic mixtures. LSU performed well in simple mixtures but revealed limitations in complex communities, where nonlinear effects and spectral similarity reduced accuracy.

Beyond classification, LSU enables quantitative assessment of species contributions, providing a valuable complement to PCA and LDA for ecological interpretation and bloom dynamics investigation. This integrated approach lays the foundation for future development of operational tools that combine spectral unmixing and machine learning for automated HAB detection. The combined use of hyperspectral reflectance data and computational methods supports scalable, real-time monitoring of phytoplankton diversity and abundance, with strong potential for deployment in early-warning systems and coastal observatories.

监测浮游植物群落对评估生态系统健康和发现有害藻华（HABs）至关重要。基于微藻独特的反射率特征，高光谱成像已经成为一种有前途的工具来区分微藻物种。本文介绍了五种浮游植物的实验室光谱分析，包括在沿海水域常见的开花和产生毒素的分类群。在多种细胞浓度下测量反射光谱，并使用两种归一化方法、二阶导数和四阶导数变换以及主成分分析（PCA）和线性判别分析（LDA）等降维技术进行分析。我们的研究结果表明，特定的光谱特征，特别是在470-500 nm和620-680 nm范围内，可以实现物种水平的区分。PCA和LDA通过减少谱冗余和强调类特征，有效地增强了可分性。我们进一步应用线性光谱分解（LSU）来估计合成混合物中的分数物种丰度。LSU在简单混合物中表现良好，但在复杂群落中显示出局限性，其中非线性效应和光谱相似性降低了精度。除了分类之外，LSU还可以定量评估物种贡献，为PCA和LDA的生态解释和花华动态调查提供了有价值的补充。这种集成方法为未来开发将光谱分解和机器学习相结合的操作工具奠定了基础，以实现自动化HAB检测。结合使用高光谱反射数据和计算方法，支持对浮游植物多样性和丰度进行可扩展的实时监测，具有在预警系统和沿海观测站部署的强大潜力。

{"title":"Laboratory-based hyperspectral reflectance analysis for phytoplankton species identification","authors":"R. Bentivogli , L. Pezzolesi , N. Caputo , B. Casarotto , S. Silvestri","doi":"10.1016/j.ecoinf.2026.103626","DOIUrl":"10.1016/j.ecoinf.2026.103626","url":null,"abstract":"<div><div>Monitoring phytoplankton communities is essential for assessing ecosystem health and detecting harmful algal blooms (HABs). Hyperspectral imaging has emerged as a promising tool to discriminate among microalgal species based on their unique reflectance signatures. This study presents a laboratory spectral analysis of five phytoplankton species, including bloom-forming and toxin-producing taxa common in coastal waters. Reflectance spectra were measured at multiple cell concentrations and analyzed using two normalization approaches, second- and fourth-derivative transformations, and dimensionality reduction techniques including principal component analysis (PCA) and linear discriminant analysis (LDA).</div><div>Our results demonstrate that specific spectral features, particularly in the 470–500 nm and 620–680 nm ranges, enable species-level discrimination. PCA and LDA effectively enhanced separability by reducing spectral redundancy and emphasizing class features. We further applied linear spectral unmixing (LSU) to estimate fractional species abundances in synthetic mixtures. LSU performed well in simple mixtures but revealed limitations in complex communities, where nonlinear effects and spectral similarity reduced accuracy.</div><div>Beyond classification, LSU enables quantitative assessment of species contributions, providing a valuable complement to PCA and LDA for ecological interpretation and bloom dynamics investigation. This integrated approach lays the foundation for future development of operational tools that combine spectral unmixing and machine learning for automated HAB detection. The combined use of hyperspectral reflectance data and computational methods supports scalable, real-time monitoring of phytoplankton diversity and abundance, with strong potential for deployment in early-warning systems and coastal observatories.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103626"},"PeriodicalIF":7.3,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146025990","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}

引用次数: 0

A comparative study of ensemble and non-ensemble machine learning methods for predicting river pollution index 集成与非集成机器学习方法在河流污染指数预测中的比较研究

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-21 DOI: 10.1016/j.ecoinf.2026.103617

Luisa S.R. Nogueira , Mariana A.S. de Carvalho , Berilo de O. Santos , Roland Yonaba , Apoorva Bamal , Md Galal Uddin , Matteo Bodini , Leonardo Goliatt

Accurate prediction of river water quality is fundamental to environmental sustainability and public health, particularly amid increasing freshwater scarcity. This study develops a robust Machine Learning (ML) framework to forecast the River Pollution Index (RPI) using a comprehensive 36-year national dataset from Taiwan’s Environmental Protection Administration, covering over 500 monitoring stations. We conducted a systematic comparison of ensemble methods (CatBoost, XGBoost, NGBoost) and non-ensemble benchmarks (SVM, ElasticNet, and 1D CNN). Hyperparameters were optimized via Bayesian optimization, and statistical significance was ensured by evaluating model stability using a suite of complementary indicators (RMSE, MAE, R

^{2}

, A10 index) across 30 independent experimental runs. The results demonstrated the consistent superiority of ensemble models over non-ensemble counterparts. Among them, CatBoost achieved the highest accuracy and stability (RMSE

\approx

0.85, MAE

\approx

0.61, R

^{2}

= 0.78), reducing prediction error by approximately 20% relative to SVM and ElasticNet. These findings highlight the capacity of ensemble learning techniques to capture complex, non-linear interactions inherent in water quality data. The study makes two principal contributions: (1) the systematic implementation, optimization, and comparison of ensemble and non-ensemble ML models for river pollution prediction on a long-term national dataset; and (2) the identification of ensemble-based methods, particularly CatBoost, as robust and data-driven tools to enhance RPI forecasting and to support informed decision-making in sustainable water resource management.

准确预测河流水质对环境可持续性和公众健康至关重要，特别是在淡水日益稀缺的情况下。本研究开发了一个强大的机器学习（ML）框架来预测河流污染指数（RPI），该框架使用了台湾环境保护署36年的综合全国数据集，涵盖了500多个监测站。我们对集成方法（CatBoost、XGBoost、NGBoost）和非集成基准（SVM、ElasticNet和1D CNN）进行了系统比较。通过贝叶斯优化对超参数进行优化，并通过30次独立实验运行使用一套互补指标（RMSE， MAE, R2， A10指数）评估模型稳定性来确保统计显著性。结果表明，集合模型相对于非集合模型具有一致的优越性。其中，CatBoost的准确率和稳定性最高（RMSE≈0.85,MAE≈0.61,R2 = 0.78），相对于SVM和ElasticNet的预测误差降低了约20%。这些发现突出了集成学习技术捕获水质数据中固有的复杂、非线性相互作用的能力。本研究的主要贡献有两方面：(1)系统实施、优化和比较了基于长期国家数据集的河流污染集成和非集成ML模型；(2)确定基于集合的方法，特别是CatBoost，作为增强RPI预测和支持可持续水资源管理的明智决策的强大数据驱动工具。

{"title":"A comparative study of ensemble and non-ensemble machine learning methods for predicting river pollution index","authors":"Luisa S.R. Nogueira , Mariana A.S. de Carvalho , Berilo de O. Santos , Roland Yonaba , Apoorva Bamal , Md Galal Uddin , Matteo Bodini , Leonardo Goliatt","doi":"10.1016/j.ecoinf.2026.103617","DOIUrl":"10.1016/j.ecoinf.2026.103617","url":null,"abstract":"<div><div>Accurate prediction of river water quality is fundamental to environmental sustainability and public health, particularly amid increasing freshwater scarcity. This study develops a robust Machine Learning (ML) framework to forecast the River Pollution Index (RPI) using a comprehensive 36-year national dataset from Taiwan’s Environmental Protection Administration, covering over 500 monitoring stations. We conducted a systematic comparison of ensemble methods (CatBoost, XGBoost, NGBoost) and non-ensemble benchmarks (SVM, ElasticNet, and 1D CNN). Hyperparameters were optimized via Bayesian optimization, and statistical significance was ensured by evaluating model stability using a suite of complementary indicators (RMSE, MAE, R<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span>, A10 index) across 30 independent experimental runs. The results demonstrated the consistent superiority of ensemble models over non-ensemble counterparts. Among them, CatBoost achieved the highest accuracy and stability (RMSE <span><math><mo>≈</mo></math></span> 0.85, MAE <span><math><mo>≈</mo></math></span> 0.61, R<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span> = 0.78), reducing prediction error by approximately 20% relative to SVM and ElasticNet. These findings highlight the capacity of ensemble learning techniques to capture complex, non-linear interactions inherent in water quality data. The study makes two principal contributions: (1) the systematic implementation, optimization, and comparison of ensemble and non-ensemble ML models for river pollution prediction on a long-term national dataset; and (2) the identification of ensemble-based methods, particularly CatBoost, as robust and data-driven tools to enhance RPI forecasting and to support informed decision-making in sustainable water resource management.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103617"},"PeriodicalIF":7.3,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146025989","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}

引用次数: 0

Deep learning models used to detect fish movement over resistivity counters 通过电阻率计数器检测鱼类运动的深度学习模型

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-21 DOI: 10.1016/j.ecoinf.2026.103606

Sophie A.M. Elliott , Keerthan Boraiah , Chun Kee Tham , William R.C. Beaumont , Paul Elsmere , Luke Scott , Adrian Fewings

Diadromous fish are one of the most threatened groups of fish species, being subject to pressures from freshwater, estuarine and marine environments. Of these fish, Atlantic salmon is the most economically important and increasingly threatened. To assess salmonid (Atlantic salmon and sea trout) stocks, resistivity counters have been widely used. However, verification of data from the counters can be challenging due to miscounts, misidentification and biases in human verification of fish counts.

We applied deep learning models to identify diadromous fish using continuous electrical resistivity data from resistivity fish counters. Our models were tested on three rivers (Frome, Fowey and Test in the South and South-West of England) and compared with a minimum of one year's manually validated data.

We detected fish signals from background noise with an F1-score of 99%, large from small fish (≥30 cm) with a precision of 95%, and an increase of >38% small and large fish waveforms. The F1-score for salmonids was 92%, and a significantly greater proportion (>173%) of upstream-moving large salmonids (≥30 cm) were detected compared to manual methods.

To date, abundance estimates for resistivity counters have only been applied to salmonids because of labour-intensive waveform identification. Using deep learning methods, we quantified salmonids and other diadromous fish with varying accuracies. Our method can be applied to resistivity counters to detect diadromous fish globally, reducing human bias and improving detection accuracy.

二恶鱼是最受威胁的鱼类之一，受到淡水、河口和海洋环境的压力。在这些鱼类中，大西洋鲑鱼是经济上最重要的，而且日益受到威胁。为了评估鲑鱼（大西洋鲑鱼和海鳟鱼）种群，电阻率计数器已被广泛使用。然而，由于计数错误、错误识别和人类对鱼类计数的偏差，对计数器数据的验证可能具有挑战性。我们应用深度学习模型，利用电阻率鱼计数器的连续电阻率数据来识别二项式鱼。我们的模型在三条河流（英格兰南部和西南部的Frome、Fowey和Test）上进行了测试，并与至少一年的人工验证数据进行了比较。我们从背景噪声中检测到的鱼信号f1得分为99%，从小鱼（≥30 cm）中检测到的鱼信号精度为95%，大小鱼波形提高了>；38%。鲑鱼的f1得分为92%，与人工方法相比，检测到上游移动的大型鲑鱼（≥30 cm）的比例显著提高（173%）。迄今为止，电阻率计数器的丰度估计只应用于鲑科鱼类，因为需要进行劳力密集的波形识别。使用深度学习方法，我们以不同的精度量化了鲑鱼和其他二恶鱼。该方法可应用于电阻率计数器在全球范围内检测二恶鱼，减少人为偏差，提高检测精度。

{"title":"Deep learning models used to detect fish movement over resistivity counters","authors":"Sophie A.M. Elliott , Keerthan Boraiah , Chun Kee Tham , William R.C. Beaumont , Paul Elsmere , Luke Scott , Adrian Fewings","doi":"10.1016/j.ecoinf.2026.103606","DOIUrl":"10.1016/j.ecoinf.2026.103606","url":null,"abstract":"<div><div>Diadromous fish are one of the most threatened groups of fish species, being subject to pressures from freshwater, estuarine and marine environments. Of these fish, Atlantic salmon is the most economically important and increasingly threatened. To assess salmonid (Atlantic salmon and sea trout) stocks, resistivity counters have been widely used. However, verification of data from the counters can be challenging due to miscounts, misidentification and biases in human verification of fish counts.</div><div>We applied deep learning models to identify diadromous fish using continuous electrical resistivity data from resistivity fish counters. Our models were tested on three rivers (Frome, Fowey and Test in the South and South-West of England) and compared with a minimum of one year's manually validated data.</div><div>We detected fish signals from background noise with an F1-score of 99%, large from small fish (≥30 cm) with a precision of 95%, and an increase of >38% small and large fish waveforms. The F1-score for salmonids was 92%, and a significantly greater proportion (>173%) of upstream-moving large salmonids (≥30 cm) were detected compared to manual methods.</div><div>To date, abundance estimates for resistivity counters have only been applied to salmonids because of labour-intensive waveform identification. Using deep learning methods, we quantified salmonids and other diadromous fish with varying accuracies. Our method can be applied to resistivity counters to detect diadromous fish globally, reducing human bias and improving detection accuracy.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103606"},"PeriodicalIF":7.3,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174412","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}

引用次数: 0

A critical review of statistical, signal processing and machine learning methods for continuous and high-frequency water quality data improvement 对持续和高频水质数据改进的统计、信号处理和机器学习方法的重要回顾

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-19 DOI: 10.1016/j.ecoinf.2026.103619

A.T. Badrudeen , D. Sahoo , C.B. Sawyer , J.W. Pike , R.D. Harmel

In the digital water world, high-frequency water quality monitoring from sensors is crucial for capturing rapid changes, especially during storm events or discharge fluctuations, in which important signals can occur at sub-hourly intervals. These signals are represented in a time series and can sometimes be irregular, noisy, and prone to missing values or errors due to buried conditions, sediment interference, and signal loss. The fine resolution of reporting also increases the risk of sensor errors and data loss, necessitating effective correction methods to ensure the accuracy and usability of the data. This literature review investigates the current state of time series data correction and denoising techniques in water quality monitoring. A systematic review of peer-reviewed studies was conducted to identify commonly applied methods, evaluate their effectiveness, and assess their adaptability to high-frequency, nonlinear, and non-stationary water quality datasets. The study explored techniques, including statistical methods such as moving averages, median filtering, Savitzky-Golay smoothing, wavelet transforms, and Kalman Filter, as well as machine learning models such as random forest, support vector machine and gradient boosting. While many of these methods are well established in other fields, this review collates evidence of their application and adaptation to water resources. This review serves as a comprehensive resource for researchers and water resource practitioners to implement appropriate denoising and correction techniques for continuous and high-frequency monitoring data. It highlights the potential of both statistical, signal processing, and machine learning-based methods to support accurate analysis, decision-making, and long-term water quality monitoring, management, and modeling.

在数字水世界中，来自传感器的高频水质监测对于捕捉快速变化至关重要，特别是在风暴事件或流量波动期间，其中重要信号可能以小时为间隔出现。这些信号以时间序列表示，有时可能是不规则的，有噪声的，并且由于埋藏条件，沉积物干扰和信号丢失而容易丢失值或误差。报告的精细分辨率也增加了传感器误差和数据丢失的风险，需要有效的校正方法来确保数据的准确性和可用性。本文综述了水质监测中时间序列数据校正和去噪技术的现状。对同行评议的研究进行了系统的回顾，以确定常用的方法，评估其有效性，并评估其对高频、非线性和非平稳水质数据集的适应性。该研究探索了一些技术，包括移动平均线、中值滤波、Savitzky-Golay平滑、小波变换和卡尔曼滤波等统计方法，以及随机森林、支持向量机和梯度增强等机器学习模型。虽然其中许多方法在其他领域已经建立，但本综述整理了它们在水资源中的应用和适应的证据。这篇综述为研究人员和水资源从业者实施适当的去噪和校正技术对连续和高频监测数据提供了全面的资源。它强调了统计、信号处理和基于机器学习的方法的潜力，以支持准确的分析、决策和长期水质监测、管理和建模。

{"title":"A critical review of statistical, signal processing and machine learning methods for continuous and high-frequency water quality data improvement","authors":"A.T. Badrudeen , D. Sahoo , C.B. Sawyer , J.W. Pike , R.D. Harmel","doi":"10.1016/j.ecoinf.2026.103619","DOIUrl":"10.1016/j.ecoinf.2026.103619","url":null,"abstract":"<div><div>In the digital water world, high-frequency water quality monitoring from sensors is crucial for capturing rapid changes, especially during storm events or discharge fluctuations, in which important signals can occur at sub-hourly intervals. These signals are represented in a time series and can sometimes be irregular, noisy, and prone to missing values or errors due to buried conditions, sediment interference, and signal loss. The fine resolution of reporting also increases the risk of sensor errors and data loss, necessitating effective correction methods to ensure the accuracy and usability of the data. This literature review investigates the current state of time series data correction and denoising techniques in water quality monitoring. A systematic review of peer-reviewed studies was conducted to identify commonly applied methods, evaluate their effectiveness, and assess their adaptability to high-frequency, nonlinear, and non-stationary water quality datasets. The study explored techniques, including statistical methods such as moving averages, median filtering, Savitzky-Golay smoothing, wavelet transforms, and Kalman Filter, as well as machine learning models such as random forest, support vector machine and gradient boosting. While many of these methods are well established in other fields, this review collates evidence of their application and adaptation to water resources. This review serves as a comprehensive resource for researchers and water resource practitioners to implement appropriate denoising and correction techniques for continuous and high-frequency monitoring data. It highlights the potential of both statistical, signal processing, and machine learning-based methods to support accurate analysis, decision-making, and long-term water quality monitoring, management, and modeling.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103619"},"PeriodicalIF":7.3,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173997","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}

引用次数: 0

Supporting fire behavior modelling with canopy base height and canopy bulk density estimates using airborne and spaceborne lidar 支持使用机载和星载激光雷达进行树冠底部高度和树冠体积密度估算的火灾行为建模

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-17 DOI: 10.1016/j.ecoinf.2026.103612

Adrian Pascual , Juan Guerra-Hernández , Brigite Botequim , Eduardo González-Ferreiro

The next-generation of fire behavior models must integrate 3D forest structural metrics to better explain fire spread, risk and severity. Canopy base height (CBH) and canopy bulk density (CBD) can be calibrated using lidar data collocated over field plots. Where no airborne lidar scanning data (ALS) exist, GEDI spaceborne lidar can provide 25-m predictions of CBH and CBD contingent to ALS-calibrated workflows. Our research presents GEDI footprint-level estimates of CBH and CBD for Mediterranean forests and builds upon collocated ALS-GEDI crossovers. Our accuracies are based on Random Forests classification: the R2 values for CBH and CBD were < 0.4 for all plant functional types evaluated. The classification of vertical continuity was satisfactory to inform on fire-prone conditions at the GEDI footprint level. Predictions for CBD were aggregated to produce a regional baseline maps, one at 1-km resolution. The usability of these coarse-scale aggregations of fuel estimates is limited because of resolution, presence of gaps and high heterogeneity of forest fuels within small steps. To inform about this heterogeneity and change estimates over time, we predict CBH and CBD over adjacent GEDI tracks collected 5-years apart (2019/24). These change estimates are relevant to show the high variability of the forest fuels that compromises the ability to depict change adding the issue of exact collocation between GEDI measurements that are adjacent, not collocated and therefore not repeated. We discuss methodological differences between our approach and recent studies on mapping fuel baselines and their approach to inform on dynamics.

下一代火灾行为模型必须集成三维森林结构指标，以更好地解释火灾的蔓延、风险和严重程度。冠层基高（CBH）和冠层容重（CBD）可以通过在野外地块上配置的激光雷达数据进行校准。在没有机载激光雷达扫描数据（ALS）的情况下，GEDI星载激光雷达可以根据ALS校准的工作流程提供25米的CBH和CBD预测。我们的研究提出了地中海森林的CBD和CBD的GEDI足迹水平估计，并建立在ALS-GEDI交叉的基础上。我们的准确性基于随机森林分类：在所有评估的植物功能类型中，CBH和CBD的R2值为0.4。垂直连续性的分类是令人满意的通知火灾易发条件在GEDI足迹水平。对CBD的预测被汇总成一个区域基线地图，分辨率为1公里。这些燃料估计的粗略集合的可用性是有限的，因为在小的步骤内森林燃料的分辨率、存在差距和高度异质性。为了了解这种异质性和随时间变化的估计，我们预测了间隔5年（2019/24）收集的相邻GEDI轨道上的CBH和CBD。这些变化估计与显示森林燃料的高度可变性有关，这损害了描述变化的能力，并增加了GEDI测量之间精确搭配的问题，这些测量是相邻的，不是搭配的，因此不会重复。我们讨论了我们的方法和最近关于绘制燃料基线及其动态信息方法的研究之间的方法差异。

{"title":"Supporting fire behavior modelling with canopy base height and canopy bulk density estimates using airborne and spaceborne lidar","authors":"Adrian Pascual , Juan Guerra-Hernández , Brigite Botequim , Eduardo González-Ferreiro","doi":"10.1016/j.ecoinf.2026.103612","DOIUrl":"10.1016/j.ecoinf.2026.103612","url":null,"abstract":"<div><div>The next-generation of fire behavior models must integrate 3D forest structural metrics to better explain fire spread, risk and severity. Canopy base height (CBH) and canopy bulk density (CBD) can be calibrated using lidar data collocated over field plots. Where no airborne lidar scanning data (ALS) exist, GEDI spaceborne lidar can provide 25-m predictions of CBH and CBD contingent to ALS-calibrated workflows. Our research presents GEDI footprint-level estimates of CBH and CBD for Mediterranean forests and builds upon collocated ALS-GEDI crossovers. Our accuracies are based on Random Forests classification: the R2 values for CBH and CBD were < 0.4 for all plant functional types evaluated. The classification of vertical continuity was satisfactory to inform on fire-prone conditions at the GEDI footprint level. Predictions for CBD were aggregated to produce a regional baseline maps, one at 1-km resolution. The usability of these coarse-scale aggregations of fuel estimates is limited because of resolution, presence of gaps and high heterogeneity of forest fuels within small steps. To inform about this heterogeneity and change estimates over time, we predict CBH and CBD over adjacent GEDI tracks collected 5-years apart (2019/24). These change estimates are relevant to show the high variability of the forest fuels that compromises the ability to depict change adding the issue of exact collocation between GEDI measurements that are adjacent, not collocated and therefore not repeated. We discuss methodological differences between our approach and recent studies on mapping fuel baselines and their approach to inform on dynamics.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103612"},"PeriodicalIF":7.3,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146025991","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}

引用次数: 0

Effective indicators to enable robust decision making in the Murray-Darling Basin 有效的指标，使强大的决策在默里-达令盆地

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-17 DOI: 10.1016/j.ecoinf.2026.103613

Georgia K. Dwyer, Galen Holt, Rebecca E. Lester

Decision making in natural resource management is increasingly challenged by the complexity, uncertainty and volume of information available. This leads to information overload and reduces decision quality. Here, we aim to reduce the degree of redundant information in a highly complex model used to manage environmental outcomes in the Murray-Darling Basin, Australia's largest river system. We identified extremely high levels of redundancy with 6–8 hydrological indicators able to explain 98–100% of the variation in the full suite of >2500. Many sets of 6–8 had similar explanatory power, providing flexibility for managers to tailor the subset used, and sets were produced for individual management targets (e.g. native fish, waterbirds) and for individual catchments. These subsets maintained the ability to distinguish between scenarios of plausible future climates and different policy settings, while traditional aggregation techniques did not (particularly for policy settings). Aggregations using key flow categories (e.g. overbank, low flows) were also largely able to distinguish among scenarios. This study is novel in its use of decision science to reduce information overload associated with a commonly used ecological model, illustrating the value of explicitly considering whether complexity in such a model is necessary. Such assessments are rarely undertaken. We illustrated how subsets of complex models provide simple, parsimonious approaches for credible, salient and legitimate decision making under uncertainty, and increase the likelihood of good evidence-based decision making in water management.

自然资源管理决策日益受到可获得信息的复杂性、不确定性和数量的挑战。这会导致信息过载，降低决策质量。在这里，我们的目标是在一个高度复杂的模型中减少冗余信息的程度，该模型用于管理澳大利亚最大的河流系统墨累-达令盆地的环境结果。我们通过6-8个水文指标确定了极高的冗余水平，这些指标能够解释整套>；2500中98-100%的变化。许多6-8组具有类似的解释力，为管理人员量身定制所使用的子集提供了灵活性，并且为个别管理目标（例如本地鱼类，水鸟）和个别集水区制作了组。这些子集保持了区分似是而非的未来气候情景和不同政策设置的能力，而传统的聚合技术则没有（特别是对于政策设置）。使用关键流量类别（例如，过岸、低流量）的聚合在很大程度上也能够区分不同的场景。这项研究的新颖之处在于它使用决策科学来减少与常用生态模型相关的信息过载，说明了明确考虑这种模型的复杂性是否必要的价值。很少进行这种评估。我们说明了复杂模型的子集如何在不确定的情况下为可信、突出和合理的决策提供简单、简约的方法，并增加了水管理中良好的循证决策的可能性。

{"title":"Effective indicators to enable robust decision making in the Murray-Darling Basin","authors":"Georgia K. Dwyer, Galen Holt, Rebecca E. Lester","doi":"10.1016/j.ecoinf.2026.103613","DOIUrl":"10.1016/j.ecoinf.2026.103613","url":null,"abstract":"<div><div>Decision making in natural resource management is increasingly challenged by the complexity, uncertainty and volume of information available. This leads to information overload and reduces decision quality. Here, we aim to reduce the degree of redundant information in a highly complex model used to manage environmental outcomes in the Murray-Darling Basin, Australia's largest river system. We identified extremely high levels of redundancy with 6–8 hydrological indicators able to explain 98–100% of the variation in the full suite of >2500. Many sets of 6–8 had similar explanatory power, providing flexibility for managers to tailor the subset used, and sets were produced for individual management targets (e.g. native fish, waterbirds) and for individual catchments. These subsets maintained the ability to distinguish between scenarios of plausible future climates and different policy settings, while traditional aggregation techniques did not (particularly for policy settings). Aggregations using key flow categories (e.g. overbank, low flows) were also largely able to distinguish among scenarios. This study is novel in its use of decision science to reduce information overload associated with a commonly used ecological model, illustrating the value of explicitly considering whether complexity in such a model is necessary. Such assessments are rarely undertaken. We illustrated how subsets of complex models provide simple, parsimonious approaches for credible, salient and legitimate decision making under uncertainty, and increase the likelihood of good evidence-based decision making in water management.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103613"},"PeriodicalIF":7.3,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080650","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}

引用次数: 0

A prototype coupled modeling approach for predicting harmful algal blooms: A case study in Chile 预测有害藻华的原型耦合建模方法：智利案例研究

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-16 DOI: 10.1016/j.ecoinf.2026.103615

Ishara Uhanie Perera , So Fujiyoshi , Daiki Kumakura , Carolina Medel , Kyoko Yarimizu , Osvaldo Artal , Pablo Reche , Oscar Espinoza-González , Leonardo Guzman , Felipe Tucca , Alexander Jaramillo-Torres , Jacquelinne J. Acuña , Milko A. Jorquera , Shinji Nakaoka , Satoshi Nagai , Fumito Maruyama

Predicting harmful algal blooms (HABs) remains a major challenge for coastal management and aquaculture. This study compares three forecasting approaches developed under the Monitoring of Algae in Chile (MACH) project: a particle dispersion model, an LSTM neural network, and an empirical dynamic model (EDM) to evaluate their ability to forecast bloom events. Consequently, we applied the EDM to forecast two Pseudo-nitzschia species groups using data collected from Metri, Quellón, and Melinka in southern Chile. The results showed that the genus Ceratium and Leptocylindrus were commonly associated with both Pseudo-nitzschia species groups, and the best prediction by causal species was obtained for the P. seriata group, with a correlation coefficient of 0.733 (P < 0.0001) between observed and predicted values. This case study demonstrated that species interactions can be used to predict specific HAB species; however, the prediction performance may vary depending on location and species. This study provides one of the first applications of EDM for HAB forecasting using causal species in a real-world monitoring context, demonstrating the potential of hybrid modeling frameworks to improve early warning systems and mitigate aquaculture losses.

预测有害藻华（HABs）仍然是沿海管理和水产养殖的主要挑战。本研究比较了在智利藻类监测（MACH）项目下开发的三种预测方法：粒子分散模型、LSTM神经网络和经验动态模型（EDM），以评估它们预测水华事件的能力。因此，我们利用智利南部的Metri、Quellón和Melinka收集的数据，应用EDM预测了两个伪nitzschia物种群。结果表明，Ceratium属和lepto圆柱属与伪尼齐亚属和伪尼齐亚属均有共同关联，其中seratiata属的因果种预测效果最好，观测值与预测值的相关系数为0.733 （P < 0.0001）。该案例研究表明，物种间相互作用可用于预测特定的赤潮物种；然而，预测效果可能因地点和物种而异。该研究提供了EDM在实际监测环境中利用因果物种进行赤潮预测的首批应用之一，展示了混合建模框架在改善预警系统和减轻水产养殖损失方面的潜力。

{"title":"A prototype coupled modeling approach for predicting harmful algal blooms: A case study in Chile","authors":"Ishara Uhanie Perera , So Fujiyoshi , Daiki Kumakura , Carolina Medel , Kyoko Yarimizu , Osvaldo Artal , Pablo Reche , Oscar Espinoza-González , Leonardo Guzman , Felipe Tucca , Alexander Jaramillo-Torres , Jacquelinne J. Acuña , Milko A. Jorquera , Shinji Nakaoka , Satoshi Nagai , Fumito Maruyama","doi":"10.1016/j.ecoinf.2026.103615","DOIUrl":"10.1016/j.ecoinf.2026.103615","url":null,"abstract":"<div><div>Predicting harmful algal blooms (HABs) remains a major challenge for coastal management and aquaculture. This study compares three forecasting approaches developed under the Monitoring of Algae in Chile (MACH) project: a particle dispersion model, an LSTM neural network, and an empirical dynamic model (EDM) to evaluate their ability to forecast bloom events. Consequently, we applied the EDM to forecast two <em>Pseudo-nitzschia</em> species groups using data collected from Metri, Quellón, and Melinka in southern Chile. The results showed that the genus <em>Ceratium</em> and <em>Leptocylindrus</em> were commonly associated with both <em>Pseudo-nitzschia</em> species groups, and the best prediction by causal species was obtained for the <em>P. seriata</em> group, with a correlation coefficient of 0.733 (<em>P</em> < 0.0001) between observed and predicted values. This case study demonstrated that species interactions can be used to predict specific HAB species; however, the prediction performance may vary depending on location and species. This study provides one of the first applications of EDM for HAB forecasting using causal species in a real-world monitoring context, demonstrating the potential of hybrid modeling frameworks to improve early warning systems and mitigate aquaculture losses.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103615"},"PeriodicalIF":7.3,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174091","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}

引用次数: 0

Informer-based cross-site transfer learning for water demand forecasting via domain adaptation and meta-learning 基于信息者的跨站点迁移学习：基于领域适应和元学习的需水量预测

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-16 DOI: 10.1016/j.ecoinf.2025.103593

Nevena Ranković , Dragica Ranković

Accurate forecasting of water demand is essential for reliable and efficient operation of distribution networks. However, existing forecasting approaches are usually restricted to single-site closed settings and fail under distributional shifts across pumping stations, leaving the gap of cross-site generalization unresolved. This study addresses the problem using operational data collected from Supervisory Control and Data Acquisition (SCADA) systems provided by the Public Utility Company for Water Treatment Valjevo, which operates the Kolubara water supply region in Serbia. We extend the Informer architecture with statistical alignment (CORAL, MMD), adversarial adaptation (DANN), and meta-learning (MAML, Reptile) to explicitly handle zero, few, and full shot transfer scenarios under covariate shift. The proposed framework reduces mean and peak errors across forecasting horizons, improving resilience where operations are most vulnerable to demand surges, and thus carries direct social relevance for water security. Results show that statistical/adversarial alignment enables effective zero-shot transfer, while meta-learning supports rapid adaptation with only 24–72 h of labeled data, consistently outperforming classical and deep-learning baselines. The evaluation, conducted on multivariate SCADA data, was checked using standard accuracy and peak-sensitive metrics. Generally, the study establishes cross-site transfer learning as an engineering solution for water utilities, offering a deployable pipeline that adapts to new pumping stations with minimal calibration while reducing operational risks and costs and grounding its value in both methodological innovation and empirical validation.

准确的水量预测对配电网的可靠、高效运行至关重要。然而，现有的预测方法通常局限于单站点封闭设置，并且在泵站之间的分布移位情况下失败，使得跨站点泛化的差距无法解决。本研究使用从Valjevo水处理公用事业公司提供的监控和数据采集（SCADA）系统收集的运行数据来解决这个问题，该公司经营塞尔维亚的Kolubara供水区。我们用统计对齐（CORAL， MMD）、对抗适应（DANN）和元学习（MAML, Reptile）扩展了Informer架构，以显式地处理协变量移位下的零、少和全射击转移场景。拟议的框架减少了预测范围内的平均误差和峰值误差，提高了最容易受到需求激增影响的地区的恢复能力，因此对水安全具有直接的社会意义。结果表明，统计/对抗对齐可以实现有效的零射击迁移，而元学习支持快速适应，只需24-72小时的标记数据，始终优于经典和深度学习基线。对多变量SCADA数据进行评估，使用标准精度和峰敏感指标进行检查。总的来说，该研究建立了跨站点迁移学习作为水务公司的工程解决方案，提供了一种可部署的管道，以最小的校准适应新的泵站，同时降低了操作风险和成本，并将其价值建立在方法创新和经验验证上。

{"title":"Informer-based cross-site transfer learning for water demand forecasting via domain adaptation and meta-learning","authors":"Nevena Ranković , Dragica Ranković","doi":"10.1016/j.ecoinf.2025.103593","DOIUrl":"10.1016/j.ecoinf.2025.103593","url":null,"abstract":"<div><div>Accurate forecasting of water demand is essential for reliable and efficient operation of distribution networks. However, existing forecasting approaches are usually restricted to single-site closed settings and fail under distributional shifts across pumping stations, leaving the gap of cross-site generalization unresolved. This study addresses the problem using operational data collected from Supervisory Control and Data Acquisition (SCADA) systems provided by the Public Utility Company for Water Treatment Valjevo, which operates the Kolubara water supply region in Serbia. We extend the Informer architecture with statistical alignment (CORAL, MMD), adversarial adaptation (DANN), and meta-learning (MAML, Reptile) to explicitly handle <em>zero</em>, <em>few</em>, and <em>full</em> shot transfer scenarios under covariate shift. The proposed framework reduces mean and peak errors across forecasting horizons, improving resilience where operations are most vulnerable to demand surges, and thus carries direct social relevance for water security. Results show that statistical/adversarial alignment enables effective <em>zero</em>-shot transfer, while meta-learning supports rapid adaptation with only 24–72 h of labeled data, consistently outperforming classical and deep-learning baselines. The evaluation, conducted on multivariate SCADA data, was checked using standard accuracy and peak-sensitive metrics. Generally, the study establishes cross-site transfer learning as an engineering solution for water utilities, offering a deployable pipeline that adapts to new pumping stations with minimal calibration while reducing operational risks and costs and grounding its value in both methodological innovation and empirical validation.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"93 ","pages":"Article 103593"},"PeriodicalIF":7.3,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146022404","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}

引用次数: 0

AI-driven analysis of fish reproductive behaviour 鱼类繁殖行为的人工智能驱动分析

IF 7.3 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2026-01-14 DOI: 10.1016/j.ecoinf.2026.103614

Aurora González-Vidal , Simona Caruso , Fabio Badalamenti , Jesús E. Argente-Garcia , Trevor J. Willis , Antonio F. Skarmeta

Understanding animal behaviour is essential for wildlife management and conservation. However, studying the intricate behaviours of marine fish presents unique challenges, including their mobility, size, and the difficulty of maintaining continuous visibility in dynamic underwater environments. Traditional methods, such as manual video analysis and direct observation, are time-consuming, labour-intensive, and often limited in scope. While AI has been applied to aquaculture and species identification, its use in decoding complex mating behaviours of wild fish – particularly nest-building species like the grey wrasse (Symphodus cinereus) – remains underexplored. This study bridges this gap by leveraging advancements in Artificial Intelligence (AI) and computer vision to automate the analysis of S. cinereus reproductive behaviours. By labelling image datasets and using the YOLOv9 algorithm, we developed a robust system to detect nest maintenance, courtship, egg care, and territorial defence behaviours during mating seasons. Observations revealed intricate male behavioural patterns, including nest building, courtship displays, and parental care, with high detection accuracy for key objects (e.g., nests, females) and varied performance across behaviour categories. Our work highlights the potential of AI-driven tools to overcome the constraints of manual methods, providing scalable, precise methodologies for ethological research. Beyond S. cinereus, this framework can be adapted to study other nest-building fish (e.g., damselfish, cichlids), supporting conservation efforts through large-scale behavioural monitoring and guiding sustainable ecotourism practices. By integrating AI and behavioural ecology, this study advances both scientific understanding and practical applications for marine ecosystem management.

了解动物行为对野生动物的管理和保护至关重要。然而，研究海洋鱼类复杂的行为提出了独特的挑战，包括它们的机动性，大小，以及在动态水下环境中保持持续能见度的困难。传统的方法，如手工视频分析和直接观察，耗时、劳动密集，而且范围往往有限。虽然人工智能已应用于水产养殖和物种鉴定，但它在解码野生鱼类——特别是灰濑鱼（Symphodus cinereus）等筑巢物种——复杂交配行为方面的应用仍未得到充分探索。这项研究通过利用人工智能（AI）和计算机视觉的进步来自动化分析S. cinereus的生殖行为，从而弥补了这一差距。通过标记图像数据集并使用YOLOv9算法，我们开发了一个强大的系统来检测交配季节的巢维护、求偶、蛋护理和领土防御行为。观察揭示了复杂的雄性行为模式，包括筑巢、求爱和亲代照顾，对关键物体（如巢穴、雌性）的检测准确率很高，并且在行为类别中表现各异。我们的工作强调了人工智能驱动工具克服人工方法限制的潜力，为动物行为学研究提供了可扩展的、精确的方法。除了S. cinereus，这一框架还可以适用于研究其他筑巢鱼类（如小雀鲷、慈鲷），通过大规模行为监测支持保护工作，并指导可持续生态旅游实践。本研究将人工智能与行为生态学相结合，促进了对海洋生态系统管理的科学认识和实际应用。

{"title":"AI-driven analysis of fish reproductive behaviour","authors":"Aurora González-Vidal , Simona Caruso , Fabio Badalamenti , Jesús E. Argente-Garcia , Trevor J. Willis , Antonio F. Skarmeta","doi":"10.1016/j.ecoinf.2026.103614","DOIUrl":"10.1016/j.ecoinf.2026.103614","url":null,"abstract":"<div><div>Understanding animal behaviour is essential for wildlife management and conservation. However, studying the intricate behaviours of marine fish presents unique challenges, including their mobility, size, and the difficulty of maintaining continuous visibility in dynamic underwater environments. Traditional methods, such as manual video analysis and direct observation, are time-consuming, labour-intensive, and often limited in scope. While AI has been applied to aquaculture and species identification, its use in decoding complex mating behaviours of wild fish – particularly nest-building species like the grey wrasse (<em>Symphodus cinereus</em>) – remains underexplored. This study bridges this gap by leveraging advancements in Artificial Intelligence (AI) and computer vision to automate the analysis of <em>S. cinereus</em> reproductive behaviours. By labelling image datasets and using the YOLOv9 algorithm, we developed a robust system to detect nest maintenance, courtship, egg care, and territorial defence behaviours during mating seasons. Observations revealed intricate male behavioural patterns, including nest building, courtship displays, and parental care, with high detection accuracy for key objects (e.g., nests, females) and varied performance across behaviour categories. Our work highlights the potential of AI-driven tools to overcome the constraints of manual methods, providing scalable, precise methodologies for ethological research. Beyond <em>S. cinereus</em>, this framework can be adapted to study other nest-building fish (e.g., damselfish, cichlids), supporting conservation efforts through large-scale behavioural monitoring and guiding sustainable ecotourism practices. By integrating AI and behavioural ecology, this study advances both scientific understanding and practical applications for marine ecosystem management.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"94 ","pages":"Article 103614"},"PeriodicalIF":7.3,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146001725","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}

引用次数: 0