Ecological Informatics最新文献_第3页

Framework for deep learning diagnosis of plant disorders in horticultural crops: From data collection tools to user-friendly web and mobile apps 园艺作物植物病害深度学习诊断框架：从数据收集工具到用户友好型网络和移动应用程序

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102900

Ma. Luisa Buchaillot , Jose A. Fernandez-Gallego , Henda Mahmoudi , Sumitha Thushar , Amna Abdulnoor Aljanaahi , Sherzod Kosimov , Zied Hammami , Ghazi Al Jabri , Alexandra La Cruz Puente , Alexi Akl , M. Isabel Trillas , Jose Luis Araus , Shawn C. Kefauver

Food security is a pressing global concern, particularly highlighted by the United Nations Sustainable Development Goal 2 (SDG 2), which focuses on enhancing the productivity and incomes of smallholder farmers. In the Middle East and North Africa (MENA) region, horticultural crops are increasingly threatened by pests and diseases, exacerbated by climate change. Local farmers often lack the necessary expertise to effectively manage these issues, resulting in significant reductions in both yield and quality of their crops. This study seeks to develop an accessible mobile crop diagnosis application. By utilizing machine learning and deep learning technologies, the app is designed to help MENA farmers quickly and accurately identify and treat crop disorders. We used Open Data Kit (ODK) to gather a large dataset of crop images required to train deep learning models. These models, built on open-source deep learning architectures, were designed to classify 21 different leaf disorders, including diseases, pests, and nutritional deficiencies. The system was implemented in both a web app and an Android mobile app. Our deep learning models demonstrated an overall accuracy of 94 % in diagnosing plant disorders. The app, Doctor Nabat, includes a decision support system that offers treatment options in the three primary languages spoken in the MENA region. Doctor Nabat is an effective and scalable tool for enhancing crop management in the MENA region, promoting food security by minimizing crop losses through improved pest and disease diagnosis and treatment strategies.

粮食安全是一个紧迫的全球问题，联合国可持续发展目标 2（SDG 2）尤其强调了这一点，其重点是提高小农的生产力和收入。在中东和北非（MENA）地区，园艺作物日益受到病虫害的威胁，而气候变化又加剧了这一威胁。当地农民往往缺乏有效管理这些问题所需的专业知识，导致作物产量和质量大幅下降。本研究旨在开发一款便于使用的移动作物诊断应用程序。通过利用机器学习和深度学习技术，该应用旨在帮助中东和北非地区的农民快速准确地识别和治疗作物病害。我们使用开放数据工具包（ODK）收集了训练深度学习模型所需的大量作物图像数据集。这些模型建立在开源深度学习架构上，旨在对 21 种不同的叶片病害进行分类，包括病害、虫害和营养缺乏症。该系统通过网络应用程序和安卓手机应用程序实施。我们的深度学习模型在诊断植物疾病方面的总体准确率达到 94%。这款名为 "Doctor Nabat "的应用程序包括一个决策支持系统，可提供中东和北非地区三种主要语言的治疗方案。Doctor Nabat 是加强中东和北非地区作物管理的有效且可扩展的工具，通过改进病虫害诊断和治疗策略，最大限度地减少作物损失，从而促进粮食安全。

{"title":"Framework for deep learning diagnosis of plant disorders in horticultural crops: From data collection tools to user-friendly web and mobile apps","authors":"Ma. Luisa Buchaillot , Jose A. Fernandez-Gallego , Henda Mahmoudi , Sumitha Thushar , Amna Abdulnoor Aljanaahi , Sherzod Kosimov , Zied Hammami , Ghazi Al Jabri , Alexandra La Cruz Puente , Alexi Akl , M. Isabel Trillas , Jose Luis Araus , Shawn C. Kefauver","doi":"10.1016/j.ecoinf.2024.102900","DOIUrl":"10.1016/j.ecoinf.2024.102900","url":null,"abstract":"<div><div>Food security is a pressing global concern, particularly highlighted by the United Nations Sustainable Development Goal 2 (SDG 2), which focuses on enhancing the productivity and incomes of smallholder farmers. In the Middle East and North Africa (MENA) region, horticultural crops are increasingly threatened by pests and diseases, exacerbated by climate change. Local farmers often lack the necessary expertise to effectively manage these issues, resulting in significant reductions in both yield and quality of their crops. This study seeks to develop an accessible mobile crop diagnosis application. By utilizing machine learning and deep learning technologies, the app is designed to help MENA farmers quickly and accurately identify and treat crop disorders. We used Open Data Kit (ODK) to gather a large dataset of crop images required to train deep learning models. These models, built on open-source deep learning architectures, were designed to classify 21 different leaf disorders, including diseases, pests, and nutritional deficiencies. The system was implemented in both a web app and an Android mobile app. Our deep learning models demonstrated an overall accuracy of 94 % in diagnosing plant disorders. The app, Doctor Nabat, includes a decision support system that offers treatment options in the three primary languages spoken in the MENA region. Doctor Nabat is an effective and scalable tool for enhancing crop management in the MENA region, promoting food security by minimizing crop losses through improved pest and disease diagnosis and treatment strategies.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102900"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142700686","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}

引用次数: 0

Automated classification of tree species using graph structure data and neural networks 利用图结构数据和神经网络自动进行树种分类

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102874

Hadi Yazdi , Kai Zhe Boey , Thomas Rötzer , Frank Petzold , Qiguan Shu , Ferdinand Ludwig

The classification of tree species in urban contexts is pivotal in assessing ecosystem services and fostering sustainable urban development. This paper explores using graph neural networks (GNNs) on graph structure data derived from quantitative structure models (QSMs) and tree structural measurement for appropriate species classification. The study addresses gaps in existing methods by integrating relationships between tree components, such as branches and cylinders, and considering the entire tree structure in a novel graph data format. The results demonstrate the efficacy of GNNs, particularly the Gated Graph Convolutional Network (GatedGCN), in appropriately classifying urban tree species. It gained an overall classification accuracy and weighted F1 score of 0.84. An analysis of confusion matrices revealed similarities in visual characteristics among several species, including A. platanoides and T. cordata, which pose significant challenges in accurately distinguishing between them. However, certain species, such as A. hippocastanum and P. nigra var. italica, have proved easier to classify than others. Furthermore, the results highlight the importance of relationships between different tree components in species recognition, such as the ratio between branch radius and parent branch radius, the factors often overlooked by previous methods. This underscores the novelty and effectiveness of the proposed approach in this study. Future research could explore integrating additional data sources, such as Leaf Area Density (LAD) calculated from LiDAR and hyperspectral data, to enhance classification accuracy. In conclusion, the evaluation results of the GatedGCN model demonstrated its effectiveness in classifying tree species using a novel data structure format derived from QSM tree characteristics. Advancing urban tree species classification through such methods can enhance future urban tree management using automated AI and robotics solutions.

城市环境中的树种分类对于评估生态系统服务和促进城市可持续发展至关重要。本文探讨了在定量结构模型（QSM）和树木结构测量得出的图形结构数据上使用图神经网络（GNN）进行适当的树种分类。该研究通过整合树枝和圆柱体等树木组成部分之间的关系，并以新颖的图数据格式考虑整个树木结构，弥补了现有方法的不足。研究结果证明了 GNN，特别是门控图卷积网络（GatedGCN）在对城市树种进行适当分类方面的功效。它的总体分类准确率和加权 F1 得分为 0.84。对混淆矩阵的分析表明，包括 A. platanoides 和 T. cordata 在内的多个树种的视觉特征具有相似性，这给准确区分它们带来了巨大挑战。不过，某些物种（如海马草和黑叶桉变种）被证明比其他物种更容易分类。此外，研究结果还强调了不同树体成分之间的关系在物种识别中的重要性，如树枝半径与母枝半径之间的比率，而这些因素往往被以前的方法所忽视。这凸显了本研究提出的方法的新颖性和有效性。未来的研究可以探索整合其他数据源，如通过激光雷达和高光谱数据计算的叶面积密度（LAD），以提高分类的准确性。总之，GatedGCN 模型的评估结果表明，该模型能有效地利用源自 QSM 树木特征的新型数据结构格式进行树种分类。通过这种方法推进城市树种分类，可以利用自动化人工智能和机器人解决方案加强未来的城市树木管理。

{"title":"Automated classification of tree species using graph structure data and neural networks","authors":"Hadi Yazdi , Kai Zhe Boey , Thomas Rötzer , Frank Petzold , Qiguan Shu , Ferdinand Ludwig","doi":"10.1016/j.ecoinf.2024.102874","DOIUrl":"10.1016/j.ecoinf.2024.102874","url":null,"abstract":"<div><div>The classification of tree species in urban contexts is pivotal in assessing ecosystem services and fostering sustainable urban development. This paper explores using graph neural networks (GNNs) on graph structure data derived from quantitative structure models (QSMs) and tree structural measurement for appropriate species classification. The study addresses gaps in existing methods by integrating relationships between tree components, such as branches and cylinders, and considering the entire tree structure in a novel graph data format. The results demonstrate the efficacy of GNNs, particularly the Gated Graph Convolutional Network (GatedGCN), in appropriately classifying urban tree species. It gained an overall classification accuracy and weighted F1 score of 0.84. An analysis of confusion matrices revealed similarities in visual characteristics among several species, including <em>A. platanoides</em> and <em>T. cordata</em>, which pose significant challenges in accurately distinguishing between them. However, certain species, such as <em>A. hippocastanum</em> and <em>P. nigra var. italica</em>, have proved easier to classify than others. Furthermore, the results highlight the importance of relationships between different tree components in species recognition, such as the ratio between branch radius and parent branch radius, the factors often overlooked by previous methods. This underscores the novelty and effectiveness of the proposed approach in this study. Future research could explore integrating additional data sources, such as Leaf Area Density (LAD) calculated from LiDAR and hyperspectral data, to enhance classification accuracy. In conclusion, the evaluation results of the GatedGCN model demonstrated its effectiveness in classifying tree species using a novel data structure format derived from QSM tree characteristics. Advancing urban tree species classification through such methods can enhance future urban tree management using automated AI and robotics solutions.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102874"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142700598","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}

引用次数: 0

Operationalising weather surveillance radar data for use in ecological research 将气象监测雷达数据应用于生态研究

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102901

Maryna Lukach , Thomas Dally , William Evans , Elizabeth J. Duncan , Lindsay Bennett , Freya I. Addison , William E. Kunin , Jason W. Chapman , Ryan R. Neely III , Christopher Hassall

Global biodiversity declines require a step change in monitoring frameworks to properly track and diagnose population trends. National weather surveillance radar (WSR) networks offer high spatial (ca. 1-10 km) and temporal (5–10 min) resolution data collected over regional and decadal scales, with well-supported infrastructure that holds great promise for the study of biodiversity. However, WSR datasets pose new challenges for ecologists due to their format, volume, and three-dimensional spatial structure. Here, we define a novel approach to the processing of WSR data to produce a product that can be used to interrogate trends in aerial biodiversity (abundance or diversity) at and across individual ground-level sites. From the full volume of WSR data collected approximately every six minutes we extract vertical columns of WSR observations above sites to compare against standardised nocturnal macro-moth monitoring data at ground level. The results show that there is strong agreement between the WSR-derived proxy of biodiversity in the air column and ground-level measurements of abundance and diversity in nocturnal moth communities. The columnar product operates on a biologically relevant scale with a diameter of 5 km, although column dimensions can easily be customised, and can be deployed at any site within a WSR's observable range. These findings have the potential to unlock past and present WSR observations for widespread application to existing and novel ecological questions and can be applied to weather radar networks around the world.

全球生物多样性的减少要求监测框架发生重大变化，以正确跟踪和诊断种群趋势。国家气象监测雷达（WSR）网络提供高空间分辨率（约 1-10 公里）和时间分辨率（5-10 分钟）的数据，收集范围覆盖区域和十年尺度。然而，WSR 数据集因其格式、容量和三维空间结构而给生态学家带来了新的挑战。在此，我们定义了一种处理 WSR 数据的新方法，以生成一种产品，用于分析单个地面站点的空中生物多样性（丰度或多样性）趋势。从大约每六分钟采集一次的全量 WSR 数据中，我们提取了站点上方的垂直 WSR 观测数据列，与地面的标准化夜间大型蛾类监测数据进行比较。结果表明，WSR 得出的气柱生物多样性替代值与地面夜蛾群落的丰度和多样性测量值非常吻合。柱状产品的直径为 5 千米，在生物相关尺度上运行，但柱状产品的尺寸可以很容易地定制，并且可以部署在 WSR 可观测范围内的任何地点。这些发现有可能将过去和现在的 WSR 观测结果广泛应用于现有和新的生态问题，并可应用于世界各地的天气雷达网络。

{"title":"Operationalising weather surveillance radar data for use in ecological research","authors":"Maryna Lukach , Thomas Dally , William Evans , Elizabeth J. Duncan , Lindsay Bennett , Freya I. Addison , William E. Kunin , Jason W. Chapman , Ryan R. Neely III , Christopher Hassall","doi":"10.1016/j.ecoinf.2024.102901","DOIUrl":"10.1016/j.ecoinf.2024.102901","url":null,"abstract":"<div><div>Global biodiversity declines require a step change in monitoring frameworks to properly track and diagnose population trends. National weather surveillance radar (WSR) networks offer high spatial (ca. 1-10 km) and temporal (5–10 min) resolution data collected over regional and decadal scales, with well-supported infrastructure that holds great promise for the study of biodiversity. However, WSR datasets pose new challenges for ecologists due to their format, volume, and three-dimensional spatial structure. Here, we define a novel approach to the processing of WSR data to produce a product that can be used to interrogate trends in aerial biodiversity (abundance or diversity) at and across individual ground-level sites. From the full volume of WSR data collected approximately every six minutes we extract vertical columns of WSR observations above sites to compare against standardised nocturnal macro-moth monitoring data at ground level. The results show that there is strong agreement between the WSR-derived proxy of biodiversity in the air column and ground-level measurements of abundance and diversity in nocturnal moth communities. The columnar product operates on a biologically relevant scale with a diameter of 5 km, although column dimensions can easily be customised, and can be deployed at any site within a WSR's observable range. These findings have the potential to unlock past and present WSR observations for widespread application to existing and novel ecological questions and can be applied to weather radar networks around the world.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102901"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142700684","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}

引用次数: 0

PlantView: Integrating deep learning with 3D modeling for indoor plant augmentation PlantView：将深度学习与 3D 建模相结合，实现室内植物扩增

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102899

Sitara Afzal, Haseeb Ali Khan, Jong Weon Lee

Indoor plant recognition poses significant challenges due to the variability in lighting conditions, plant species, and growth stages. Despite the growing interest in applying deep learning techniques to plant data, there still needs to be more research focused on the automatic recognition of indoor plant species, highlighting the need for real-time, automated solutions. To address this gap, this study introduces a novel approach for real-time identification and visualization of indoor plants using a Convolutional Neural Network (CNN)-based model called PlantView, integrated with Augmented Reality (AR) for enhanced user interaction. The proposed PlantView model not only accurately classifies the plant species but also visualizes them in a 3D AR environment, allowing users to interact with virtual plant models seamlessly integrated into their real-world surroundings. We developed a custom dataset comprising over 28,000 images of 48 different plant species at various growth stages, captured under diverse lighting conditions and camera settings. Our proposed approach achieves an impressive accuracy of 98.20 %. To validate the effectiveness of PlantView model, we conduct extensive experiments and compared its performance against state-of-the-art methods, demonstrating its superior accuracy and processing speed. The results indicate that our method is not only highly effective for real-time indoor plant recognition but also offers practical applications for enhancing indoor plant care and visualization. This research offers a comprehensive solution for indoor plant enthusiasts and professionals, combining advanced computer vision techniques with immersive AR visualization to revolutionize the way indoor plants are identified, visualized, and integrated into living spaces.

由于光照条件、植物种类和生长阶段的多变性，室内植物识别面临着巨大挑战。尽管人们对将深度学习技术应用于植物数据的兴趣与日俱增，但仍需要更多的研究来关注室内植物物种的自动识别，这凸显了对实时、自动解决方案的需求。为了填补这一空白，本研究介绍了一种基于卷积神经网络（CNN）的室内植物实时识别和可视化模型--PlantView，该模型与增强现实（AR）技术相结合，可增强用户交互。所提出的 PlantView 模型不仅能准确地对植物种类进行分类，还能在三维 AR 环境中将其可视化，从而使用户能够与虚拟植物模型进行交互，并将其无缝集成到现实世界的环境中。我们开发了一个自定义数据集，其中包括在不同光照条件和相机设置下捕捉到的处于不同生长阶段的 48 种不同植物的 28,000 多张图像。我们提出的方法达到了令人印象深刻的 98.20% 的准确率。为了验证 PlantView 模型的有效性，我们进行了广泛的实验，并将其性能与最先进的方法进行了比较，证明了其卓越的准确性和处理速度。结果表明，我们的方法不仅在实时室内植物识别方面非常有效，而且在加强室内植物护理和可视化方面也有实际应用。这项研究为室内植物爱好者和专业人士提供了一个全面的解决方案，将先进的计算机视觉技术与身临其境的 AR 可视化技术相结合，彻底改变了室内植物的识别、可视化和融入生活空间的方式。

{"title":"PlantView: Integrating deep learning with 3D modeling for indoor plant augmentation","authors":"Sitara Afzal, Haseeb Ali Khan, Jong Weon Lee","doi":"10.1016/j.ecoinf.2024.102899","DOIUrl":"10.1016/j.ecoinf.2024.102899","url":null,"abstract":"<div><div>Indoor plant recognition poses significant challenges due to the variability in lighting conditions, plant species, and growth stages. Despite the growing interest in applying deep learning techniques to plant data, there still needs to be more research focused on the automatic recognition of indoor plant species, highlighting the need for real-time, automated solutions. To address this gap, this study introduces a novel approach for real-time identification and visualization of indoor plants using a Convolutional Neural Network (CNN)-based model called PlantView, integrated with Augmented Reality (AR) for enhanced user interaction. The proposed PlantView model not only accurately classifies the plant species but also visualizes them in a 3D AR environment, allowing users to interact with virtual plant models seamlessly integrated into their real-world surroundings. We developed a custom dataset comprising over 28,000 images of 48 different plant species at various growth stages, captured under diverse lighting conditions and camera settings. Our proposed approach achieves an impressive accuracy of 98.20 %. To validate the effectiveness of PlantView model, we conduct extensive experiments and compared its performance against state-of-the-art methods, demonstrating its superior accuracy and processing speed. The results indicate that our method is not only highly effective for real-time indoor plant recognition but also offers practical applications for enhancing indoor plant care and visualization. This research offers a comprehensive solution for indoor plant enthusiasts and professionals, combining advanced computer vision techniques with immersive AR visualization to revolutionize the way indoor plants are identified, visualized, and integrated into living spaces.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102899"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142701319","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}

引用次数: 0

Efficient dual-stream neural networks: A modeling approach for inferring wild mammal behavior from video data 高效双流神经网络：从视频数据推断野生哺乳动物行为的建模方法

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102902

Ao Xu , Zhenjie Hou , Jiuzhen Liang , Xing Li , Xinwen Zhou , Hongbo Geng

Monitoring animal behavior is crucial for protecting ecosystems, maintaining ecological balance, and improving animal welfare. By utilizing various monitoring devices, a wealth of behavioral data can be collected, which machine learning techniques can then analyze to identify specific behaviors. Artificial neural networks are particularly important in movement ecology. However, current research on animal behavior recognition faces several limitations. Many existing datasets are limited by homogeneous species categories, simplistic environmental conditions, restricted video perspectives, and a lack of alignment with the complexity of real-world environments. Consequently, there is significant room for improvement in the robustness and generalization of automatic animal behavior recognition models. To address these challenges, this paper introduces a diverse dataset of wild mammal behaviors. This dataset includes a wide variety of typical wild mammal species, providing a foundation for enhancing the generality and robustness of recognition models. The videos in this dataset capture different environmental contexts where wild mammals reside, across various times of day. Based on this dataset, a novel and highly efficient wild mammal behavior recognition model, EDNN, is proposed. The EDNN model integrates both temporal and spatial scales and achieves an average recognition accuracy of 79.17 % for basic locomotion behaviors, with a Top-1 accuracy of 81.37 % and a Top-5 accuracy of 98.04 %. These results demonstrate the feasibility of automating animal behavior recognition using large datasets collected from modern monitoring devices. The EDNN model is highly effective for behavior recognition and can be readily applied across diverse species and scenarios. It efficiently processes various video data and contributes to a deeper understanding of the movement ecology of species that are challenging to observe.

监测动物行为对于保护生态系统、维持生态平衡和改善动物福利至关重要。通过利用各种监测设备，可以收集到大量行为数据，然后利用机器学习技术对这些数据进行分析，从而识别特定行为。人工神经网络在运动生态学中尤为重要。然而，目前有关动物行为识别的研究面临着一些限制。许多现有的数据集受到物种类别单一、环境条件简单化、视频视角受限以及与真实世界环境复杂性不符等因素的限制。因此，动物行为自动识别模型的鲁棒性和通用性还有很大的改进空间。为了应对这些挑战，本文引入了一个多样化的野生哺乳动物行为数据集。该数据集包括各种典型的野生哺乳动物物种，为提高识别模型的通用性和鲁棒性奠定了基础。该数据集中的视频捕捉了野生哺乳动物在一天中不同时间段的不同环境背景。在此数据集的基础上，提出了一种新型、高效的野生哺乳动物行为识别模型 EDNN。EDNN 模型整合了时间和空间尺度，对基本运动行为的平均识别准确率为 79.17%，Top-1 准确率为 81.37%，Top-5 准确率为 98.04%。这些结果证明了利用从现代监测设备收集的大型数据集自动识别动物行为的可行性。EDNN 模型对行为识别非常有效，可随时应用于不同物种和场景。它能有效地处理各种视频数据，有助于加深对难以观察到的物种的运动生态学的理解。

{"title":"Efficient dual-stream neural networks: A modeling approach for inferring wild mammal behavior from video data","authors":"Ao Xu , Zhenjie Hou , Jiuzhen Liang , Xing Li , Xinwen Zhou , Hongbo Geng","doi":"10.1016/j.ecoinf.2024.102902","DOIUrl":"10.1016/j.ecoinf.2024.102902","url":null,"abstract":"<div><div>Monitoring animal behavior is crucial for protecting ecosystems, maintaining ecological balance, and improving animal welfare. By utilizing various monitoring devices, a wealth of behavioral data can be collected, which machine learning techniques can then analyze to identify specific behaviors. Artificial neural networks are particularly important in movement ecology. However, current research on animal behavior recognition faces several limitations. Many existing datasets are limited by homogeneous species categories, simplistic environmental conditions, restricted video perspectives, and a lack of alignment with the complexity of real-world environments. Consequently, there is significant room for improvement in the robustness and generalization of automatic animal behavior recognition models. To address these challenges, this paper introduces a diverse dataset of wild mammal behaviors. This dataset includes a wide variety of typical wild mammal species, providing a foundation for enhancing the generality and robustness of recognition models. The videos in this dataset capture different environmental contexts where wild mammals reside, across various times of day. Based on this dataset, a novel and highly efficient wild mammal behavior recognition model, EDNN, is proposed. The EDNN model integrates both temporal and spatial scales and achieves an average recognition accuracy of 79.17 % for basic locomotion behaviors, with a Top-1 accuracy of 81.37 % and a Top-5 accuracy of 98.04 %. These results demonstrate the feasibility of automating animal behavior recognition using large datasets collected from modern monitoring devices. The EDNN model is highly effective for behavior recognition and can be readily applied across diverse species and scenarios. It efficiently processes various video data and contributes to a deeper understanding of the movement ecology of species that are challenging to observe.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102902"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142700685","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}

引用次数: 0

Lake surface water temperature in China from 2001 to 2021 based on GEE and HANTS 基于 GEE 和 HANTS 的 2001 至 2021 年中国湖泊地表水温度

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102903

Song Song , Jinxin Yang , Linjie Liu , Gale Bai , Jie Zhou , Deirdre McKay

Warming of lakes' surface water leads to accelerated loss of biodiversity and eco-environmental collapse of aquatic systems. Changes in lack surface water temperature (LSWT) are a crucial indicator of lake warming. LSWT growth potentially leads to a higher greenhouse gas emissions and deterioration of the ecological environment within lake systems. However, the magnitude of these changes remains uncertain due to data limitations, particularly for small lakes (1–5 km²). Small lakes will experience increasing perturbation with accelerating climate change and our methods demonstrate how the impacts of changes in lakes can be accurately measured and monitored. Our study assessed the spatial and temporal patterns of LSWT in China from 2001 to 2021. We utilized Google Earth Engine (GEE) and the Harmonic Analysis of Time Series (HANTS) algorithm to reconstruct LSWT series and detect spatiotemporal dynamics. The innovative connection of GEE and HANTS provides powerful tool for LSWT analysis. Our results show LSWT increased at a rate of 0.24 °C per decade, albeit with notable spatial and temporal variations. The nighttime rate of increase was greater than the daytime rate of increase. However, there was an abrupt change in daytime LSWT in approximately 2010 and this occurred earlier than an abrupt change in nighttime LSWT. Geographically, the lakes in the Eastern Plain zone exhibited the most significant LSWT warming trend. The majority of lakes warmed more rapidly between 2011 and 2021 as compared to 2001 to 2010. We found a concurrent and pronounced increase in the frequency of algal bloom occurrences after 2010. Our results demonstrate how GEE and HANTS can deliver the continued monitoring and assessment of LSWT trends needed to inform management strategies aimed at mitigating potential negative impacts of climate change on lake ecosystems, both locally and globally. Building on this method, future research should explore the underlying mechanisms driving LSWT trends and their long-term impacts on lake health.

湖泊表层水变暖会导致生物多样性加速丧失和水生系统生态环境崩溃。湖泊表层水温（LSWT）的变化是湖泊变暖的一个重要指标。缺水表层水温的增长可能会导致温室气体排放量增加和湖泊系统生态环境恶化。然而，由于数据的局限性，这些变化的幅度仍不确定，尤其是小湖泊（1-5 平方公里）。随着气候变化的加速，小型湖泊将受到越来越多的干扰，我们的方法展示了如何精确测量和监测湖泊变化的影响。我们的研究评估了 2001 年至 2021 年中国湖泊水量变化的时空模式。我们利用谷歌地球引擎（GEE）和时间序列谐波分析（HANTS）算法来重建LSWT序列并检测时空动态。GEE 和 HANTS 的创新连接为 LSWT 分析提供了强大的工具。我们的研究结果表明，尽管存在明显的时空变化，LSWT 的上升速率为每十年 0.24 °C。夜间的上升率大于白天的上升率。然而，大约在 2010 年，白天的 LSWT 突然发生了变化，这要早于夜间 LSWT 的突然变化。从地理位置上看，东部平原区的湖泊表现出最明显的整周最低温度变暖趋势。与 2001 年至 2010 年相比，大多数湖泊在 2011 年至 2021 年期间升温更快。我们发现，2010 年之后，藻华发生的频率也同时明显增加。我们的研究结果表明了 GEE 和 HANTS 如何能够对 LSWT 趋势进行持续监测和评估，从而为旨在减轻气候变化对当地和全球湖泊生态系统的潜在负面影响的管理策略提供依据。在此方法的基础上，未来的研究应探索驱动 LSWT 趋势的潜在机制及其对湖泊健康的长期影响。

{"title":"Lake surface water temperature in China from 2001 to 2021 based on GEE and HANTS","authors":"Song Song , Jinxin Yang , Linjie Liu , Gale Bai , Jie Zhou , Deirdre McKay","doi":"10.1016/j.ecoinf.2024.102903","DOIUrl":"10.1016/j.ecoinf.2024.102903","url":null,"abstract":"<div><div>Warming of lakes' surface water leads to accelerated loss of biodiversity and eco-environmental collapse of aquatic systems. Changes in lack surface water temperature (LSWT) are a crucial indicator of lake warming. LSWT growth potentially leads to a higher greenhouse gas emissions and deterioration of the ecological environment within lake systems. However, the magnitude of these changes remains uncertain due to data limitations, particularly for small lakes (1–5 km<sup><strong>2</strong></sup>). Small lakes will experience increasing perturbation with accelerating climate change and our methods demonstrate how the impacts of changes in lakes can be accurately measured and monitored. Our study assessed the spatial and temporal patterns of LSWT in China from 2001 to 2021. We utilized Google Earth Engine (GEE) and the Harmonic Analysis of Time Series (HANTS) algorithm to reconstruct LSWT series and detect spatiotemporal dynamics. The innovative connection of GEE and HANTS provides powerful tool for LSWT analysis. Our results show LSWT increased at a rate of 0.24 °C per decade, albeit with notable spatial and temporal variations. The nighttime rate of increase was greater than the daytime rate of increase. However, there was an abrupt change in daytime LSWT in approximately 2010 and this occurred earlier than an abrupt change in nighttime LSWT. Geographically, the lakes in the Eastern Plain zone exhibited the most significant LSWT warming trend. The majority of lakes warmed more rapidly between 2011 and 2021 as compared to 2001 to 2010. We found a concurrent and pronounced increase in the frequency of algal bloom occurrences after 2010. Our results demonstrate how GEE and HANTS can deliver the continued monitoring and assessment of LSWT trends needed to inform management strategies aimed at mitigating potential negative impacts of climate change on lake ecosystems, both locally and globally. Building on this method, future research should explore the underlying mechanisms driving LSWT trends and their long-term impacts on lake health.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102903"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142701320","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}

引用次数: 0

Comparison of global and zonal modeling strategies - A case study of soil organic matter and C:N ratio mapping in Altay, Xinjiang, China 全球和分区建模策略的比较--中国新疆阿勒泰地区土壤有机质和碳氮比绘图案例研究

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-17 DOI: 10.1016/j.ecoinf.2024.102882

Hongwu Liang , Guli Japaer , Tao Yu , Liancheng Zhang , Bojian Chen , Kaixiong Lin , Tongwei Ju , Yongyu Zhao , Ting Pei , Yimuranzi Aizizi

Digital soil mapping (DSM) based on remote sensing is the dominant method for soil monitoring. Currently, the global modeling strategy (GMS) is used in most soil mapping studies. In the GMS, it is assumed that the relationship between soil and the landscape is the same throughout a region. However, the soil–landscape relationship varies in different geographic zones, such as among different land cover types. In the zonal modeling strategy (ZMS), a region is divided into multiple geographic zones on the basis of zoning rules, and each geographic zone is modeled individually, to fully capture the soil–landscape relationships within different zones. This study was conducted in Altay, Xinjiang, China. The soil organic matter (SOM) content and C:N ratio were mapped on the basis of the GMS and the ZMS to compare the performance differences between the two strategies. The ZMS mapping results exhibited better spatial heterogeneity across different land cover types. Moreover, the ZMS mapping results displayed lower uncertainty and were closer to the observed values than were the GMS results, which included more outliers. Overall, we recommend the ZMS. The accuracy validation results indicated that the accuracy of the ZMS is not necessarily higher than that of the GMS in some zones, but the overall accuracy is similar. Combining similar zones for modeling can improve the accuracy of the ZMS, surpassing that of the GMS. Moreover, the importance of synthetic aperture radar (SAR) data was analyzed. The results revealed that SAR data are highly important for mapping the SOM of bare land and cropland and the C:N ratio of bare land and forest. SAR data may provide soil nutrient information indirectly from moisture levels; therefore, we believe that SAR data have great potential for soil nutrient mapping.

基于遥感技术的数字土壤制图（DSM）是土壤监测的主要方法。目前，大多数土壤制图研究都采用全球建模策略（GMS）。在全球建模策略中，假定土壤与景观之间的关系在整个区域都是相同的。然而，在不同的地理区域，如不同的土地覆被类型，土壤与景观的关系是不同的。在分区建模策略（ZMS）中，根据分区规则将一个区域划分为多个地理分区，并对每个地理分区进行单独建模，以全面反映不同分区内的土壤-景观关系。本研究在中国新疆阿勒泰地区进行。在 GMS 和 ZMS 的基础上绘制了土壤有机质（SOM）含量和 C:N 比值图，以比较两种策略的性能差异。ZMS 测绘结果在不同土地覆被类型中表现出更好的空间异质性。此外，与包含更多异常值的 GMS 结果相比，ZMS 绘图结果显示出更低的不确定性，更接近观测值。总体而言，我们推荐使用 ZMS。精度验证结果表明，在某些区域，区域监测系统的精度并不一定高于全球监测系统，但总体精度相似。合并相似区域进行建模可以提高 ZMS 的精度，从而超过 GMS。此外，还分析了合成孔径雷达（SAR）数据的重要性。结果表明，合成孔径雷达数据对于绘制裸地和耕地的 SOM 以及裸地和森林的 C:N 比率图非常重要。合成孔径雷达数据可通过湿度间接提供土壤养分信息；因此，我们认为合成孔径雷达数据在绘制土壤养分图方面具有巨大潜力。

{"title":"Comparison of global and zonal modeling strategies - A case study of soil organic matter and C:N ratio mapping in Altay, Xinjiang, China","authors":"Hongwu Liang , Guli Japaer , Tao Yu , Liancheng Zhang , Bojian Chen , Kaixiong Lin , Tongwei Ju , Yongyu Zhao , Ting Pei , Yimuranzi Aizizi","doi":"10.1016/j.ecoinf.2024.102882","DOIUrl":"10.1016/j.ecoinf.2024.102882","url":null,"abstract":"<div><div>Digital soil mapping (DSM) based on remote sensing is the dominant method for soil monitoring. Currently, the global modeling strategy (GMS) is used in most soil mapping studies. In the GMS, it is assumed that the relationship between soil and the landscape is the same throughout a region. However, the soil–landscape relationship varies in different geographic zones, such as among different land cover types. In the zonal modeling strategy (ZMS), a region is divided into multiple geographic zones on the basis of zoning rules, and each geographic zone is modeled individually, to fully capture the soil–landscape relationships within different zones. This study was conducted in Altay, Xinjiang, China. The soil organic matter (SOM) content and C:N ratio were mapped on the basis of the GMS and the ZMS to compare the performance differences between the two strategies. The ZMS mapping results exhibited better spatial heterogeneity across different land cover types. Moreover, the ZMS mapping results displayed lower uncertainty and were closer to the observed values than were the GMS results, which included more outliers. Overall, we recommend the ZMS. The accuracy validation results indicated that the accuracy of the ZMS is not necessarily higher than that of the GMS in some zones, but the overall accuracy is similar. Combining similar zones for modeling can improve the accuracy of the ZMS, surpassing that of the GMS. Moreover, the importance of synthetic aperture radar (SAR) data was analyzed. The results revealed that SAR data are highly important for mapping the SOM of bare land and cropland and the C:N ratio of bare land and forest. SAR data may provide soil nutrient information indirectly from moisture levels; therefore, we believe that SAR data have great potential for soil nutrient mapping.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102882"},"PeriodicalIF":5.8,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142700597","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}

引用次数: 0

Dynamical systems-inspired machine learning methods for drought prediction 动态系统启发的干旱预测机器学习方法

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-16 DOI: 10.1016/j.ecoinf.2024.102889

Andrew Watford , Chris T. Bauch , Madhur Anand

Drought is a naturally occurring phenomenon that affects millions of people and results in billions of dollars in damages each year, with impacts expected to worsen due to climate change. At the same time, definitions of drought are nebulous, and extant quantitative drought indicators suffer from short prediction horizons. One such indicator is the Normalized Vegetation Difference Index (NDVI), which measures photosynthetic activity, making it a strong proxy for vegetation density. Recent studies have identified chaotic attractors in satellite-derived NDVI time-series, suggesting a dynamical systems framework may be helpful for time-series prediction of NDVI. In this study, we compare the performance of a mechanistic model and two physics-informed machine learning methods (Sparse Identification of Nonlinear Dynamics [SINDy] and reservoir computing) on the prediction of NDVI time-series data in drought-prone regions of Kenya. We find that SINDy, a sparse polynomial modelling architecture, narrowly outperforms the other two methods with the use of precipitation data, while also retaining some of the interpretability of the mechanistic model. We also find that none of the methods perform as well in the regions in which the chaotic NDVI attractors were originally identified. We conclude by proposing more sophisticated extensions to the methods presented here, both with and without the availability of precipitation data, that draw on the existing dynamical systems and machine learning literature to enable better quantitative predictions of key drought indicators.

干旱是一种自然发生的现象，每年影响数百万人，造成数十亿美元的损失，预计其影响将因气候变化而加剧。同时，干旱的定义模糊不清，现有的定量干旱指标也存在预测范围短的问题。其中一个指标是归一化植被差异指数（NDVI），它测量光合活动，是植被密度的有力代表。最近的研究发现了卫星获取的 NDVI 时间序列中的混沌吸引子，这表明动力系统框架可能有助于 NDVI 的时间序列预测。在本研究中，我们比较了力学模型和两种物理信息机器学习方法（非线性动力学稀疏识别 [SINDy] 和水库计算）在肯尼亚干旱易发地区预测 NDVI 时间序列数据方面的性能。我们发现，SINDy 是一种稀疏多项式建模结构，在使用降水数据时以微弱优势胜过其他两种方法，同时还保留了一些机理模型的可解释性。我们还发现，在最初识别出混沌 NDVI 吸引子的区域，没有一种方法表现得那么好。最后，我们利用现有的动力系统和机器学习文献，对本文介绍的方法提出了更复杂的扩展建议，包括使用降水数据和不使用降水数据，以便更好地对关键干旱指标进行定量预测。

{"title":"Dynamical systems-inspired machine learning methods for drought prediction","authors":"Andrew Watford , Chris T. Bauch , Madhur Anand","doi":"10.1016/j.ecoinf.2024.102889","DOIUrl":"10.1016/j.ecoinf.2024.102889","url":null,"abstract":"<div><div>Drought is a naturally occurring phenomenon that affects millions of people and results in billions of dollars in damages each year, with impacts expected to worsen due to climate change. At the same time, definitions of drought are nebulous, and extant quantitative drought indicators suffer from short prediction horizons. One such indicator is the Normalized Vegetation Difference Index (NDVI), which measures photosynthetic activity, making it a strong proxy for vegetation density. Recent studies have identified chaotic attractors in satellite-derived NDVI time-series, suggesting a dynamical systems framework may be helpful for time-series prediction of NDVI. In this study, we compare the performance of a mechanistic model and two physics-informed machine learning methods (Sparse Identification of Nonlinear Dynamics [SINDy] and reservoir computing) on the prediction of NDVI time-series data in drought-prone regions of Kenya. We find that SINDy, a sparse polynomial modelling architecture, narrowly outperforms the other two methods with the use of precipitation data, while also retaining some of the interpretability of the mechanistic model. We also find that none of the methods perform as well in the regions in which the chaotic NDVI attractors were originally identified. We conclude by proposing more sophisticated extensions to the methods presented here, both with and without the availability of precipitation data, that draw on the existing dynamical systems and machine learning literature to enable better quantitative predictions of key drought indicators.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102889"},"PeriodicalIF":5.8,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142701324","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}

引用次数: 0

DEBEcoMod: A dynamic energy budget R tool to predict life-history traits of marine organisms across time and space DEBEcoMod：用于预测海洋生物跨时空生命史特征的动态能量预算 R 工具

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-16 DOI: 10.1016/j.ecoinf.2024.102897

A. Giacoletti , M. Bosch-Belmar , G. Di Bona , M.C. Mangano , B. Stechele , G. Sarà

DEBEcoMod is an open-source R script designed to apply Dynamic Energy Budget (DEB) theory to predict life-history traits of marine organisms under various environmental and anthropogenic stressors. It presents a novel approach to overcoming the computational and scale limitations of previous DEB applications, enabling the generation of spatially explicit outputs. DEBEcoMod is intended to predict traits such as maximum size, reproductive output, and life-history traits across different temporal and spatial scales. It utilises parameters from the AddMyPet database for various species and environmental time series to simulate the past, present, and future performance of organisms. The tool also includes a module for spatio-temporal representation, producing clear and accessible maps for stakeholders. The document highlights DEBEcoMod's application in invasion biology, marine spatial planning, integrated multi-trophic aquaculture, and marine ecology, drawing on published examples of spatial applications to demonstrate its versatility and potential in ecological research and adaptive management. Furthermore, the code has been cross-validated with the official DEBtool to ensure its accuracy and reliability. DEBEcoMod is available for download on GitHub, enhancing its accessibility and utility for a wide range of ecological and conservation applications.

DEBEcoMod 是一个开源 R 脚本，旨在应用动态能量预算（DEB）理论预测各种环境和人为压力下海洋生物的生命史特征。它提出了一种新方法来克服以往 DEB 应用在计算和规模上的局限性，从而能够生成空间明确的输出结果。DEBEcoMod 用于预测不同时空尺度下的最大体型、繁殖输出和生命史特征。它利用 AddMyPet 数据库中不同物种和环境时间序列的参数来模拟生物在过去、现在和未来的表现。该工具还包括一个时空表示模块，可为利益相关者制作清晰易懂的地图。文件重点介绍了 DEBEcoMod 在入侵生物学、海洋空间规划、综合多营养水产养殖和海洋生态学方面的应用，并借鉴了已发表的空间应用实例，以展示其在生态研究和适应性管理方面的多功能性和潜力。此外，代码还与官方 DEBtool 进行了交叉验证，以确保其准确性和可靠性。DEBEcoMod 可在 GitHub 上下载，从而提高了其在广泛的生态和保护应用中的可访问性和实用性。

{"title":"DEBEcoMod: A dynamic energy budget R tool to predict life-history traits of marine organisms across time and space","authors":"A. Giacoletti , M. Bosch-Belmar , G. Di Bona , M.C. Mangano , B. Stechele , G. Sarà","doi":"10.1016/j.ecoinf.2024.102897","DOIUrl":"10.1016/j.ecoinf.2024.102897","url":null,"abstract":"<div><div>DEBEcoMod is an open-source R script designed to apply Dynamic Energy Budget (DEB) theory to predict life-history traits of marine organisms under various environmental and anthropogenic stressors. It presents a novel approach to overcoming the computational and scale limitations of previous DEB applications, enabling the generation of spatially explicit outputs. DEBEcoMod is intended to predict traits such as maximum size, reproductive output, and life-history traits across different temporal and spatial scales. It utilises parameters from the AddMyPet database for various species and environmental time series to simulate the past, present, and future performance of organisms. The tool also includes a module for spatio-temporal representation, producing clear and accessible maps for stakeholders. The document highlights DEBEcoMod's application in invasion biology, marine spatial planning, integrated multi-trophic aquaculture, and marine ecology, drawing on published examples of spatial applications to demonstrate its versatility and potential in ecological research and adaptive management. Furthermore, the code has been cross-validated with the official DEBtool to ensure its accuracy and reliability. DEBEcoMod is available for download on GitHub, enhancing its accessibility and utility for a wide range of ecological and conservation applications.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102897"},"PeriodicalIF":5.8,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142701331","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}

引用次数: 0

Spatio-temporal evaluation of MODIS temperature vegetation dryness index in the Middle East 中东地区 MODIS 温度植被干燥指数的时空评估

IF 5.8 2区环境科学与生态学 Q1 ECOLOGY

Ecological Informatics

Pub Date : 2024-11-13 DOI: 10.1016/j.ecoinf.2024.102894

Younes Khosravi , Saeid Homayouni , Taha B.M.J. Ouarda

Drought, a recurring meteorological event, can potentially cause devastating consequences for human populations, and its attributes vary significantly across diverse geographic areas. Therefore, recognizing drought events is paramount for strategically planning and managing water resource systems. In this study, the Temperature Vegetation Dryness Index (TVDI), derived using Moderate-Resolution Imaging Spectroradiometer (MODIS) data spanning from 2003 to 2022 in the Middle East, was used as the foundation for both trend and spectral analyses. To assess TVDI trends, the Mann-Kendall test and Sen's slope estimator were utilized, and harmonic analysis was conducted for spectral analyses. These methods were applied to a dataset comprising 258,087 pixels within the specified region, covering various time scales, including monthly and seasonal analyses. The monthly analyses indicated significant growth in March and April, with September showing the least significant increase, suggesting stability or decline. Geographically, upward trends were predominant in the northern Middle East, including Turkey, Syria, Iraq, western Iran, and eastern Jordan. Significant downward trends were observed in the southern Middle East during the warmer months. Seasonal assessments showed no significant TVDI trends in winter, but upward trends in the south, west, and northwest were identified during spring. The annual trend map indicates a long-term declining trend in TVDI for most regions within specific latitudes, particularly those below 34 degrees. The results of harmonic analysis revealed the presence of multiple cycles at a 95 % confidence level. Notably, there was a heightened prevalence of significant sinusoidal cycles, especially the 2–3-year cycles. This cycle was widespread in countries such as Iran, Oman, Yemen, and Turkey, as well as in the southern regions of Saudi Arabia and Egypt.

干旱是一种经常发生的气象事件，可能会对人类造成毁灭性的后果，其属性在不同的地理区域有很大差异。因此，识别干旱事件对于水资源系统的战略规划和管理至关重要。在本研究中，利用中东地区 2003 年至 2022 年的中分辨率成像光谱仪（MODIS）数据得出的温度植被干燥指数（TVDI），作为趋势和光谱分析的基础。为评估 TVDI 趋势，使用了 Mann-Kendall 检验和 Sen 的斜率估计器，并对光谱分析进行了谐波分析。这些方法适用于指定区域内由 258,087 个像素组成的数据集，涵盖各种时间尺度，包括月度和季节分析。月度分析表明，3 月和 4 月出现了显著增长，而 9 月的增长幅度最小，表明趋于稳定或下降。从地域上看，中东北部（包括土耳其、叙利亚、伊拉克、伊朗西部和约旦东部）主要呈上升趋势。在温暖的月份，中东南部出现了明显的下降趋势。季节性评估显示，冬季 TVDI 没有明显趋势，但春季南部、西部和西北部有上升趋势。年度趋势图显示，在特定纬度的大多数地区，特别是 34 度以下的地区，TVDI 呈长期下降趋势。谐波分析结果显示，在 95% 的置信度下存在多个周期。值得注意的是，显著的正弦波周期，尤其是 2-3 年周期更为普遍。这种周期在伊朗、阿曼、也门和土耳其等国以及沙特阿拉伯和埃及南部地区普遍存在。

{"title":"Spatio-temporal evaluation of MODIS temperature vegetation dryness index in the Middle East","authors":"Younes Khosravi , Saeid Homayouni , Taha B.M.J. Ouarda","doi":"10.1016/j.ecoinf.2024.102894","DOIUrl":"10.1016/j.ecoinf.2024.102894","url":null,"abstract":"<div><div>Drought, a recurring meteorological event, can potentially cause devastating consequences for human populations, and its attributes vary significantly across diverse geographic areas. Therefore, recognizing drought events is paramount for strategically planning and managing water resource systems. In this study, the Temperature Vegetation Dryness Index (TVDI), derived using Moderate-Resolution Imaging Spectroradiometer (MODIS) data spanning from 2003 to 2022 in the Middle East, was used as the foundation for both trend and spectral analyses. To assess TVDI trends, the Mann-Kendall test and Sen's slope estimator were utilized, and harmonic analysis was conducted for spectral analyses. These methods were applied to a dataset comprising 258,087 pixels within the specified region, covering various time scales, including monthly and seasonal analyses. The monthly analyses indicated significant growth in March and April, with September showing the least significant increase, suggesting stability or decline. Geographically, upward trends were predominant in the northern Middle East, including Turkey, Syria, Iraq, western Iran, and eastern Jordan. Significant downward trends were observed in the southern Middle East during the warmer months. Seasonal assessments showed no significant TVDI trends in winter, but upward trends in the south, west, and northwest were identified during spring. The annual trend map indicates a long-term declining trend in TVDI for most regions within specific latitudes, particularly those below 34 degrees. The results of harmonic analysis revealed the presence of multiple cycles at a 95 % confidence level. Notably, there was a heightened prevalence of significant sinusoidal cycles, especially the 2–3-year cycles. This cycle was widespread in countries such as Iran, Oman, Yemen, and Turkey, as well as in the southern regions of Saudi Arabia and Egypt.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"84 ","pages":"Article 102894"},"PeriodicalIF":5.8,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142700603","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}

引用次数: 0