Ecological Informatics最新文献

Monitoring the dynamics of coastal wetlands ecosystems in Brittany (France) using LANDSAT time series and machine learning 利用LANDSAT时间序列和机器学习监测布列塔尼（法国）沿海湿地生态系统的动态

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-07-03 DOI: 10.1016/j.ecoinf.2025.103303

Adrien Le Guillou , Simona Niculescu

{"title":"Monitoring the dynamics of coastal wetlands ecosystems in Brittany (France) using LANDSAT time series and machine learning","authors":"Adrien Le Guillou , Simona Niculescu","doi":"10.1016/j.ecoinf.2025.103303","DOIUrl":"10.1016/j.ecoinf.2025.103303","url":null,"abstract":"<div><div>Coastal wetlands protect against erosion, reduce flood risks, and maintain watercourses during periods of drought, which can mitigate global warming and its effects on humans. This study aims to analyse the spatio-temporal dynamics of coastal wetland ecosystems in relation to the main drivers of change urbanisation and coastal erosion in a specific region of Brittany (France) between 1990 and 2020. The study exploits the potential of satellite image time series (SITS), machine learning (ML), and Random Forest (RF) algorithms.</div><div>These algorithms enable the software to learn autonomously from multiple datasets, including Landsat 4/5 and 8 SITS archive images. The different elements or ecosystems within a dataset are classified into categories using the Corine Biotopes classification system and multi-scalar analyses. The study proposes two scales of analysis: at the scale of coastal wetlands throughout Brittany and at the local scale of two RAMSAR coastal wetlands (Mont-Saint-Michel Bay and Audierne Bay). The results revealed contrasting spatial patterns. The Audierne Bay has experienced significant urban expansion, with a 24% increase upstream, as well as coastal erosion reaching 1.63 m/year locally, with a retreat of approximately 50 m in the most affected areas during the period 1990–2020. The wetlands in this region are receding in parallel with the coastline and have slightly decreased in area over the last 30 years, with a reduction of 8%. In contrast, the Bay of Mont-Saint-Michel has seen an expansion of salt marshes. Using Landsat archives and an automated coastline detection method, we found that the area of salt marshes has increased by 36% over the last 30 years. In both nested spatial scale approaches, the proposed spatial methodology generates spatial statistics on the dynamics of key ecosystems at the designated time scale. The results provide useful information for stakeholders. These results highlight the diversity of coastal dynamics in Brittany.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103303"},"PeriodicalIF":5.8,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144548837","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

Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data 利用Sentinel-2遥感数据预测河网浊度的机器学习方法评估

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-07-03 DOI: 10.1016/j.ecoinf.2025.103313

Victor Oliveira Santos , Paulo Alexandre Costa Rocha , Jesse Van Griensven Thé , Bahram Gharabaghi

{"title":"Evaluation of machine learning methods for forecasting turbidity in river networks using Sentinel-2 remote sensing data","authors":"Victor Oliveira Santos , Paulo Alexandre Costa Rocha , Jesse Van Griensven Thé , Bahram Gharabaghi","doi":"10.1016/j.ecoinf.2025.103313","DOIUrl":"10.1016/j.ecoinf.2025.103313","url":null,"abstract":"<div><div>Turbidity is an important indicator of river water quality and of great interest to improve the data acquisition methods and efficiency of decision support systems for sustainable ecosystem management. However, river water quality monitoring stations are very expensive to operate and maintain and lack spatial coverage. Therefore, this study takes advantage of the vast spatial coverage of remote sensing datasets from satellites to provide a more efficient hybrid system with comprehensive coverage of both spatial and temporal changes in water quality across a vast river network. Spectral bands from Sentinel-2 were analyzed using machine learning algorithms, namely XGBoost, Random Forests, GMDH, Support Vector Regression, k-Nearest Neighbors and Least Absolute Shrinkage and Selection Operator to model turbidity, using data from twelve monitoring stations across the Mississippi River, USA. Results show that considering the individual monitoring stations, the ML algorithms for turbidity modeling were satisfactory at locations with a larger range and standard deviation of turbidity values, achieving a mean R<sup>2</sup> value of 59.5 %. Tree-based models were the best overall approach, often ranking as the best or second-best performing model. Using all the samples from the monitoring stations, the XGBoost provided a superior output for turbidity modeling, reaching R<sup>2</sup> equal to 75.7 %. This represents an improvement of over 16 % compared to the average metric value for the individual stations. A comprehensive comparison with the literature found that the models implemented using this study's methodology could provide competitive results, deeming it as an alternative for turbidity modeling from remote sensing data.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103313"},"PeriodicalIF":5.8,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144556717","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

Conformal taxonomic validation: A semi-automated validation framework for citizen science records 适形分类验证：公民科学记录的半自动验证框架

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-07-02 DOI: 10.1016/j.ecoinf.2025.103290

Matthieu de Castelbajac , Sandra Bringay , Arnaud Sallaberry , Maximilien Servajean , Clémence Epinoux , Juan Carlos Molinero , Delphine Bonnet

{"title":"Conformal taxonomic validation: A semi-automated validation framework for citizen science records","authors":"Matthieu de Castelbajac , Sandra Bringay , Arnaud Sallaberry , Maximilien Servajean , Clémence Epinoux , Juan Carlos Molinero , Delphine Bonnet","doi":"10.1016/j.ecoinf.2025.103290","DOIUrl":"10.1016/j.ecoinf.2025.103290","url":null,"abstract":"<div><div>Citizen science records are a valuable source of marine biodiversity data, especially where standardized sampling campaigns are limited in spatial or temporal scope. However, such records often contain biases and errors and typically require expert validation before they can reliably support scientific research. Validating large volumes of citizen science data remains an important challenge. In this paper, we present a semi-automated validation framework that combines a deep learning classifier with conformal prediction to generate sets of plausible taxonomic labels at multiple ranks, while providing rigorous control over prediction confidence. Extensive evaluation was carried out using 25,000 jellyfish records, both with and without prior validation, as well as against 800 expert-validated entries. Our results show that the method frequently produces singleton prediction sets that can be accepted automatically, offering a high-confidence and scalable solution for validating marine citizen science data.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103290"},"PeriodicalIF":5.8,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144548848","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

Linking morpho-taxonomy to ecosystem functions: Trait-based estimation of biomass and potential carbon budget in omnivore nematodes 形态分类与生态系统功能的联系：杂食线虫生物量和潜在碳收支的性状估计

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-30 DOI: 10.1016/j.ecoinf.2025.103309

Reza Ghaderi , Uffe N. Nielsen , Ramesha H. Jayaramaiah , Helen L. Hayden , Ji-zheng He

{"title":"Linking morpho-taxonomy to ecosystem functions: Trait-based estimation of biomass and potential carbon budget in omnivore nematodes","authors":"Reza Ghaderi , Uffe N. Nielsen , Ramesha H. Jayaramaiah , Helen L. Hayden , Ji-zheng He","doi":"10.1016/j.ecoinf.2025.103309","DOIUrl":"10.1016/j.ecoinf.2025.103309","url":null,"abstract":"<div><div>Omnivore nematodes within the order Dorylaimida are among the largest free-living soil-dwelling nematodes, suggesting a significant role in soil biomass and carbon cycling. However, their contribution to these soil processes remains underexplored. Estimating biomass based on nematode morphological traits provides a practical and reliable approach for assessing their contribution in carbon dynamics. This study provides estimated individual biomass and the daily carbon budget of dorylaimids, utilizing a database of taxon-specific body-size measurements sourced from publicly available literature. We calculated biomass and potential carbon budgets for 618 reported populations worldwide, encompassing 464 species, 127 genera, 47 subfamilies, and 19 families. Biomass estimates derived using body diameter as a sole predictor, based on two recently published formulae and two adjusted formulae developed in this study, were compared with Andrássy's original formula, which incorporates both body length and diameter. The adjusted formulae proposed in this study demonstrated a superior fit compared to the recently published models. Overall, we found an estimated average individual omnivore nematode biomass (fresh weight) of 3.33 μg for females and 3.55 μg for males, and the corresponding daily carbon budgets of 0.03903 μg and 0.04163 μg for females and males, respectively. The considerable variability in biomass data across the taxonomic ranks, highlight the need for robust taxonomic resolution for ecological studies. This study offers a comprehensive dataset and improved formulae for estimating biomass and potential carbon budget in omnivore nematodes, enhancing our understanding of their functional roles in carbon dynamics and other ecosystem processes.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103309"},"PeriodicalIF":5.8,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144548836","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

Stronger response of vegetation photosynthesis to climate change than greenness in pan-Arctic region: First evidence from SIF satellite observations 泛北极地区植被光合作用对气候变化的响应强于绿化率：来自SIF卫星观测的首个证据

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-30 DOI: 10.1016/j.ecoinf.2025.103312

Mihang Jiang , Liangyun Liu , Xinjie Liu , Chu Zou

{"title":"Stronger response of vegetation photosynthesis to climate change than greenness in pan-Arctic region: First evidence from SIF satellite observations","authors":"Mihang Jiang , Liangyun Liu , Xinjie Liu , Chu Zou","doi":"10.1016/j.ecoinf.2025.103312","DOIUrl":"10.1016/j.ecoinf.2025.103312","url":null,"abstract":"<div><div>The pan-Arctic terrestrial ecosystems are highly vulnerable to climate change. However, critical uncertainties remain regarding the relationship between vegetation photosynthesis and greenness and their climate sensitivities under accelerating climate change. Here, we conducted the first investigation of their climate responses by synergistically analyzing solar-induced chlorophyll fluorescence (SIF), a direct indicator of photosynthesis from the GOME-2A satellite, and normalized difference vegetation index (NDVI) from MODIS observations during the summers of 2007–2021. Our results demonstrate that both photosynthesis and greenness exhibited increasing trends, where 70.05 % of the vegetation pixels showed a consistent change, but with significant differences in growth magnitudes. Specifically, the growth magnitudes of SIF and NDVI were 6.09 % and 3.31 %, respectively, with the increase in SIF being approximately twice that of NDVI. Mechanistically, the accelerated rise in atmospheric CO<sub>2</sub> concentration (Δ33.27 ppm), and climate warming jointly enhanced the apparent SIFyield by 4.82 %, thereby directly amplifying SIF's stronger climate sensitivity. Structural equation modeling further quantified this disparity, revealing that SIF responded to climatic factors about 1.5 times stronger than NDVI. This study provides the first evidence from satellite SIF observations that vegetation photosynthesis exhibits larger growth magnitudes and more vigorous responses to climate change than greenness in pan-Arctic ecosystems, suggesting a new perspective on the traditional NDVI-centered framework in climate impact assessment. By integrating vegetation structural and functional dynamics, our findings provide critical physiological benchmarks to refine carbon-climate feedback projections.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103312"},"PeriodicalIF":5.8,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535529","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

Exploring innovative assessment and driving mechanisms for achieving land degradation neutrality in rocky desertification areas: A case study of Yunnan–Guangxi–Guizhou, China 石漠化地区土地退化中性的创新评价与驱动机制探索——以云桂黔为例

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-29 DOI: 10.1016/j.ecoinf.2025.103310

Weihua Liao, Yifang Wei, Zhiyan Wei

{"title":"Exploring innovative assessment and driving mechanisms for achieving land degradation neutrality in rocky desertification areas: A case study of Yunnan–Guangxi–Guizhou, China","authors":"Weihua Liao, Yifang Wei, Zhiyan Wei","doi":"10.1016/j.ecoinf.2025.103310","DOIUrl":"10.1016/j.ecoinf.2025.103310","url":null,"abstract":"<div><div>Rocky desertification (RD), a manifestation of land degradation in humid and semi-humid zones, plays a pivotal role in pursuing the global goal of land degradation neutrality (LDN). However, the definition of desertification as outlined by the United Nations Convention to Combat Desertification is confined to arid and semi-arid territories, which may lead to neglect of RD emergence and rehabilitation within karst regions. To address this, the current study focused on the three most severely RD-affected provinces in China (Yunnan, Guangxi, and Guizhou) and developed a specialized LDN-RD assessment framework for RD areas to monitor the spatiotemporal dynamics of LDN. Furthermore, by employing a gradient boosting machine and Shapley Additive exPlanations values, this study investigated the influence of environmental factors and human endeavors on achievement of the LDN goal. Overall, the research findings indicated that: (1) from 2001 to 2020, Yunnan, Guangxi, and Guizhou provinces in China achieved the LDN target, with particularly notable performances in Guizhou and Guangxi; (2) environmental factors were the key determinants in achieving the LDN goal in RD areas, with nighttime light, low temperature, and water scarcity limiting achievement of LDN in degraded regions, and; (3) the introduction of an RD index enhanced the accuracy of identifying regional land-degradation phenomena. Therefore, we recommend global promotion of this new assessment framework in RD to support implementation of the LDN initiative. In summary, the full utilization and coordination of environmental factors in RD areas are highly important for accelerating achievement of the LDN goal.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103310"},"PeriodicalIF":5.8,"publicationDate":"2025-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535530","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

Which exerts a greater impact on ecosystem resilience: Cropland expansion or urban expansion? Insights from a spatiotemporal analysis 耕地扩张和城市扩张，哪个对生态系统恢复力的影响更大？来自时空分析的见解

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-29 DOI: 10.1016/j.ecoinf.2025.103314

Jinhua Wu , Hongwei Wang , Can Wang , Xin Huang , Zhenggang Wang , Chi Zhang , Bei Chen , Yilinuer Yiming , Chunshan Zhou

{"title":"Which exerts a greater impact on ecosystem resilience: Cropland expansion or urban expansion? Insights from a spatiotemporal analysis","authors":"Jinhua Wu , Hongwei Wang , Can Wang , Xin Huang , Zhenggang Wang , Chi Zhang , Bei Chen , Yilinuer Yiming , Chunshan Zhou","doi":"10.1016/j.ecoinf.2025.103314","DOIUrl":"10.1016/j.ecoinf.2025.103314","url":null,"abstract":"<div><div>Ecosystem resilience (ER) is pivotal for regional ecological security and sustainable ecosystem development. Both cropland expansion and urban expansion impact ER; however, the comparative magnitude and spatial patterns of their effects remain underexplored. This study investigates the spatiotemporal evolution and distribution characteristics of ER in the Northern Slope Economic Belt of the Tianshan Mountains (NSEBTM) from 2000 to 2020, using Fragstats and ArcGIS. A combination of the Geo-informatic Tupu method, Zonal Statistics, and Standard Deviation Ellipse analysis was employed to assess and compare the positive (ecological adaptive expansion, EAE) and negative (ecological trade-off expansion, ETE) impacts of cropland and urban expansion on ER. ER levels in the NSEBTM remained low throughout the study period, with a turning point occurring in 2010, followed by gradual improvement. Spatially, ER exhibited a “northwest high, southeast low” distribution pattern, with high-value regions clustering in the northwest and low-value regions dispersing in the southeast. Both cropland and urban expansion exerted dual impacts on ER, with ETE effects dominating in the central and western areas and EAE effects prevailing in the eastern areas. Notably, cropland expansion had broader and more intense negative impacts on ER than urban expansion, with cropland changes affecting larger areas and exhibiting more pronounced ecological trade-offs. These findings highlight the need for sustainable land-use strategies that reconcile ecological preservation with the demands of agricultural and urban development.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103314"},"PeriodicalIF":5.8,"publicationDate":"2025-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144518934","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

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-28 DOI: 10.1016/j.ecoinf.2025.103287

Yanwen Wang , Mahdi Khodadadzadeh , Raúl Zurita-Milla

{"title":"A dissimilarity-adaptive cross-validation method for evaluating geospatial machine learning predictions with clustered samples","authors":"Yanwen Wang , Mahdi Khodadadzadeh , Raúl Zurita-Milla","doi":"10.1016/j.ecoinf.2025.103287","DOIUrl":"10.1016/j.ecoinf.2025.103287","url":null,"abstract":"<div><div>Spatially clustered samples are prevalent in geospatial machine learning (ML) predictions, especially in ecological mapping. Since densely sampled regions in the prediction area are overrepresented, leading to dissimilarities in the data distribution between samples and predictions and thus posing a noticeable challenge for the evaluation of geospatial ML predictions. Neither random nor spatial cross-validation (CV) methods can consistently yield accurate evaluations: Random CV overestimates prediction performance when clustering is high, while spatial CV underestimates it when clustering is low. To tackle this challenge, we propose a novel “adaptive” evaluation method called dissimilarity-adaptive cross-validation (DA-CV), which is based on the data feature space. DA-CV categorizes the prediction locations as “similar” and “different” groups according to the dissimilarity between their covariates and those of the sampled locations. DA-CV applies random CV to evaluate “similar” locations and spatial CV to evaluate “different” ones. The final evaluation metric is obtained through a weighted average of the two. To test DA-CV, we conducted a series of experiments on synthetic species abundance and real above ground biomass datasets, where the clustering degree was gradually changed, and we also compared DA-CV with current CV methods (RDM-CV, SP-CV, and kNNDM) in the experiments. Results showed that DA-CV provided the most accurate evaluations in 85% of scenarios. DA-CV effectively overcomes the common limitations of random and spatial CV methods, such as only considering a part of predictions in the evaluation. This means that DA-CV can provide accurate evaluations for most situations of clustered samples. The success of DA-CV confirms that considering feature space information is an effective way to improve the evaluation of geospatial ML predictions.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103287"},"PeriodicalIF":5.8,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144548835","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

DeepBioFusion: Multi-modal deep learning based above ground biomass estimation using SAR and optical satellite images DeepBioFusion：基于多模态深度学习的地面生物量估算，使用SAR和光学卫星图像

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-26 DOI: 10.1016/j.ecoinf.2025.103277

Abdul Hanan , Mehak Khan , Nieves Fernandez-Anez , Reza Arghandeh

{"title":"DeepBioFusion: Multi-modal deep learning based above ground biomass estimation using SAR and optical satellite images","authors":"Abdul Hanan , Mehak Khan , Nieves Fernandez-Anez , Reza Arghandeh","doi":"10.1016/j.ecoinf.2025.103277","DOIUrl":"10.1016/j.ecoinf.2025.103277","url":null,"abstract":"<div><div>Accurate estimation of forest above-ground biomass (AGB) is essential for ecosystem conservation, sustainable forest management, and mitigating climate change and wildfire risks. Traditional methods, such as manual field surveys, are labor-intensive and limited in scope. This study presents DeepBioFusion, a multi-modal deep learning framework that first estimates AGB for validation as ground truth generation by using LiDAR-derived tree heights and a Tree Species map, employing allometry equations to relate tree height to Diameter at Breast Height (DBH). After this initial estimation, the framework is trained to predict AGB using high-resolution optical imagery and multiple bands of Synthetic Aperture Radar (SAR), including X, C, and L bands. The use of SAR bands enables improved canopy penetration, particularly in dense and cloud-covered forests. DeepBioFusion leverages the complementary strengths of SAR and optical data to enhance the accuracy of biomass predictions. Benchmarking against models like ResNet50 and Transformer, the proposed model demonstrates superior performance in AGB estimation across diverse forest environments. This study offers a scalable, cutting-edge approach to biomass monitoring, advancing efforts in climate change mitigation and sustainable forest management.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103277"},"PeriodicalIF":5.8,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144522078","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

Convolutional neural networks and vision transformers for Plankton Classification 用于浮游生物分类的卷积神经网络和视觉变压器

IF 5.8 2区环境科学与生态学

Ecological Informatics Pub Date : 2025-06-25 DOI: 10.1016/j.ecoinf.2025.103272

Loris Nanni , Alessandra Lumini , Leonardo Barcellona , Stefano Ghidoni

{"title":"Convolutional neural networks and vision transformers for Plankton Classification","authors":"Loris Nanni , Alessandra Lumini , Leonardo Barcellona , Stefano Ghidoni","doi":"10.1016/j.ecoinf.2025.103272","DOIUrl":"10.1016/j.ecoinf.2025.103272","url":null,"abstract":"<div><div>In this paper, we present a study on plankton classification for automated underwater ecosystems monitoring. The study considers the creation of ensembles combining different Convolutional Neural Network (CNN) models and transformer architectures to understand whether different optimization algorithms can result in more robust and efficient classification across various plankton datasets. Tests involved different variants of the Adam optimizer and multiple learning rate variation strategies applied to several CNN architectures, building an ensemble of classifiers. Such ensembles were tested together with transformer-based models in a detailed comparative analysis considering feature extraction efficiency, computational cost, and robustness to species imbalances. The study highlights the performance of individual nets and ensembles on multiple plankton datasets, and discusses the potential for generalizing this approach to broader aquatic ecosystems. Experiments demonstrate that combining diverse neural network models in a heterogeneous ensemble significantly improves performance with respect to other state-of-the-art approaches across all the problems considered. Final results show that the ensemble-based approach achieves a remarkable accuracy improvement over individual CNN models and over standalone Vision Transformers.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103272"},"PeriodicalIF":5.8,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144500889","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