Ecological Informatics最新文献

{"title":"First-of-its-kind AI model for bioacoustic detection using a lightweight associative memory Hopfield neural network","authors":"Andrew Gascoyne,&nbsp;Wendy Lomas","doi":"10.1016/j.ecoinf.2025.103382","DOIUrl":"10.1016/j.ecoinf.2025.103382","url":null,"abstract":"<div><div>A growing issue within conservation bioacoustics is the laborious task of analysing the vast amount of data generated from the use of passive acoustic monitoring devices. In this paper, we present an alternative AI model which has the potential to help alleviate this problem. Our model formulation addresses the key issues encountered when using current AI models for bioacoustic analysis, namely: the limited training data available; the environmental impact, particularly in energy consumption and carbon footprint of training and implementing these models; and the associated hardware requirements. The model developed in this work uses associative memory via a transparent and explainable Hopfield neural network to store signals and detect similar signals which can then be used to classify species. Training is rapid (3<!--> <!-->milliseconds), as only one representative signal is required for each target sound within a dataset. The model is fast, taking only 5.4<!--> <!-->seconds to pre-process and classify all 10384 publicly available bat recordings, on a standard Apple MacBook Air. The model is also lightweight, i.e., has a small memory footprint of 144.09<!--> <!-->MB of RAM usage. Hence, the low computational demands make the model ideal for use on a variety of standard personal devices with potential for deployment in the field via edge-processing devices. It is also competitively accurate, with up to 86% precision on the labelled dataset used to evaluate the model. In fact, we could not find a single case of disagreement between model and manual identification via expert field guides. Although a dataset of bat echolocation calls was chosen to demonstrate this first-of-its-kind AI model, trained on only two representative echolocation calls, the model is not species specific. In conclusion, we propose an equitable AI model that has the potential to be a game changer for fast, lightweight, sustainable, transparent, explainable and accurate bioacoustic analysis.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103382"},"PeriodicalIF":7.3,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144865178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the spatiotemporal effects of human activity intensity on ecosystem health using VORS and spatial models: Evidence from the e Yellow River Deltaeconomic zone","authors":"Jie Wu ,&nbsp;Lei Cao","doi":"10.1016/j.ecoinf.2025.103377","DOIUrl":"10.1016/j.ecoinf.2025.103377","url":null,"abstract":"<div><div>Understanding how coastal human activities influence ecosystem health is critical for sustainable deltaic region management amid rapid industrialization and environmental change. This study evaluates ecosystem health dynamics in China's Yellow River Delta Economic Zone (YRDEZ) from 2005 to 2022 by integrating the Vigor–Organization–Resilience–Services (VORS) model with a Coastal Human Activity Intensity (CHAI) framework incorporating marine aquaculture, petroleum extraction, and port development impacts. The novelty lies in high-resolution (500m) coupling of coastal ecosystem performance with anthropogenic pressures using machine learning algorithms (Random Forest and XGBoost) alongside Geographic Detector and Geographically Weighted Regression to capture saltwater intrusion and sediment dynamics.Findings reveal intensifying negative spatial correlations between CHAI and Ecosystem Health Index (EHI), with coastal urban centers (Dongying, Binzhou) showing highest CHAI and lowest EHI values. Yellow River wetlands exhibit moderate EHI but extreme sensitivity to hydrological modifications. Machine learning models identified salinity gradient, sediment load, and petroleum extraction as dominant explanatory factors (73% variance) for coastal EHI distribution. Aquaculture zones display unique temporal patterns with initial ecosystem enhancement followed by cumulative degradation. A novel Coastal Vulnerability Index successfully predicts ecosystem thresholds with 89% accuracy.Results provide evidence-based guidance for integrated coastal zone management and Blue Economy development in deltaic regions facing industrial expansion and sea-level rise challenges.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103377"},"PeriodicalIF":7.3,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144920127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From pixels to objects: Integrated indicators for balancing sustainable management in protected areas","authors":"Reza Peykanpour Fard,&nbsp;Alireza Soffianian,&nbsp;Mohsen Ahmadi,&nbsp;Saeid Pourmanafi","doi":"10.1016/j.ecoinf.2025.103371","DOIUrl":"10.1016/j.ecoinf.2025.103371","url":null,"abstract":"<div><div>Balancing human activities with environmental protection is a critical challenge in managing Protected Areas (PAs). Spatial zoning serves as the cornerstone of PA management and plays a crucial role in harmonizing conservation with development. This study introduces an integrated land-use management strategy that supports conservation, rehabilitation, tourism, and multilateral objectives. The approach combines pixel-based optimization with AI-enhanced object-based evaluation. Using GIS, we compiled and mapped a comprehensive set of biological, physical, infrastructural, and socio-economic criteria. Two pixel-based methods, Weighted Linear Combination (WLC) and Ordered Weighted Averaging (OWA), were applied to assess land-use suitability. A decision space was established using the Multi-Objective Land Allocation (MOLA) method. Object-based land allocation was subsequently performed using a suite of artificial intelligence models, including Boosted Regression Trees (BRT), Artificial Neural Networks (ANN), Classification and Regression Trees (CART), eXtreme Gradient Boosting (XGBoost), and Random Forest (RF). The results showed that the OWA method outperformed WLC in land suitability analysis. When integrated into the object-based approach, the RF model demonstrated the highest allocation performance, followed by ANN. Notably, RF and MOLA models showed the highest spatial agreement. This integrative framework underscores the potential of combining advanced AI-driven object-based methods with conventional pixel-based techniques to strengthen land management in PAs. The findings offer actionable insights for sustainable spatial planning in protected areas where land-use goals may be diverse and competing.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103371"},"PeriodicalIF":7.3,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144886457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating data from unmanned aerial vehicles and Sentinel-2 with PROSAIL-5D-driven machine learning for fuel moisture content estimation in agroecosystems","authors":"Jinlong Liu ,&nbsp;Jia Jin ,&nbsp;Jing Huang ,&nbsp;Mengjuan Wu ,&nbsp;Shaozheng Hao ,&nbsp;Haoyi Jia ,&nbsp;Tengda Qin ,&nbsp;Yuqing Huang ,&nbsp;Dan Chen ,&nbsp;Nathsuda Pumijumnong","doi":"10.1016/j.ecoinf.2025.103389","DOIUrl":"10.1016/j.ecoinf.2025.103389","url":null,"abstract":"<div><div>Fuel moisture content (FMC) is a critical ecological indicator for evaluating vegetation water status and ecosystem resilience, particularly in agricultural ecosystems. This study presents an advanced framework integrating multi-source remote sensing data fusion, physically based modeling, and machine learning to enable high-resolution and high-precision FMC estimation. An additive wavelet transform (AWT) was employed to fuse unmanned aerial vehicle (UAV) multispectral imagery with Sentinel-2 data, generating enhanced spatial-spectral reflectance composites while retaining key shortwave infrared bands essential for moisture analysis. To address the challenge of sparse ground observations, a calibrated PROSAIL-5D radiative transfer model was used to simulate diverse spectral responses, augmenting the training dataset. A genetic algorithm-optimized backpropagation neural network was then applied to assess the effectiveness of the fused remote sensing data and PROSAIL-5D simulation in improving FMC retrieval accuracy. The results indicate: (1) The AWT fusion method effectively integrates UAV and Sentinel-2 data, improving spatial and spectral consistency with field measurements. (2) Calibration of the PROSAIL-5D model significantly improves the retrieval accuracy of equivalent water thickness (Cw, R² = 0.847) and dry matter content (Cm, R² = 0.735), both key parameters for FMC calculation. (3) Incorporating 70 % of the measured spectral data (UAV and fused Sentinel-2) into the PROSAIL-5D simulated dataset enhanced FMC estimation accuracy (R² = 0.765), representing a 133.94 % improvement compared with using UAV data alone. This study demonstrates the potential of data fusion and physically based modeling for enhancing vegetation moisture monitoring in agroecosystems. This approach contributes to ecological informatics by offering a scalable, transferable solution for remote sensing-based analysis of ecosystem water status.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103389"},"PeriodicalIF":7.3,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144878479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Locating and profiling city street trees using Baidu street view images for carbon storage evaluation","authors":"Zhuobai Dong ,&nbsp;Yingcai Su ,&nbsp;Yuru Zhang ,&nbsp;Lifang Wang ,&nbsp;Shujun Yuan ,&nbsp;Baoyi Zhang","doi":"10.1016/j.ecoinf.2025.103394","DOIUrl":"10.1016/j.ecoinf.2025.103394","url":null,"abstract":"<div><div>Traditional methods for estimating the carbon storage of street trees involve manual sampling, which incurs substantial human, material, and temporal costs in establishing a city-wide comprehensive inventory of street trees. In this study, we propose a multi-task convolutional neural network called STV-MNet to identify individual- level and city-wide street trees from Baidu street view images (BSVIs). We measured the structural and locational information of the identified trees using cylindrical projection and MonoDepth depth estimation network. STV-MNet achieved a mean intersection over union (mIoU) of 0.733 and a mean average precision of 0.881 at IoU 50 % (mAP50) in individual tree identification, outperforming DeepLab v3+ (mIoU of 0.641) and YOLO v3 (mAP50 of 0.767). Validation with street-measured data demonstrates that our method produces more precise estimations for both tree height and breast diameter, with the root mean square error (RMSE) of 0.09 m and the normalized RMSE of 0.005 m for tree height and the RMSE of 0.01 m and the normalized RMSE of 0.016 m for diameter at breast height (DBH). The location prediction of street trees achieves a minimum error of 0.67 m and an average error of 7.37 m. Using the biomass carbon storage equation, we calculated the carbon storage of individual street trees in Changsha City, Hunan Province, China. The results indicate that the total carbon storage of 333,717 street trees in urban areas of Changsha City is 1.64 × 10⁵ tons, and the annual carbon sequestration capacity across the urban areas is 8014.57 tons. In certain areas, street tree resources have enabled the achievement of carbon neutrality in road transportation. This study presents a novel approach to managing urban street tree carbon storage, leveraging STV-MNet for automatic carbon storage estimates, and demonstrates high practical significance in low-cost and city-wide street tree carbon storage estimation.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103394"},"PeriodicalIF":7.3,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Local-scale analysis of projected climate change impact on Arabica coffee distribution in selected districts of southwestern Ethiopia: Are the future production areas commercially viable?","authors":"Melkamu Mamuye ,&nbsp;Caleb Gallemore ,&nbsp;Ng'winamila Kasongi ,&nbsp;Kristjan Jespersen ,&nbsp;Gezahegn Berecha","doi":"10.1016/j.ecoinf.2025.103392","DOIUrl":"10.1016/j.ecoinf.2025.103392","url":null,"abstract":"<div><div>Climate change is reshaping the geographies of coffee production globally, impacting the livelihoods of coffee farmers and the international coffee market. A local-scale understanding of these shifts is essential for designing effective adaptation and policy planning. This study assessed the local-scale (district-level) impact of projected climate change on coffee area suitability and how future production geographies intersect with the forest cover in five major coffee-growing districts of southwestern Ethiopia. The study models coffee distribution using an ensemble of three machine-learning algorithms (Maxent, SVM, and RF) to predict suitable areas presently and in the 2030s, 2050s, 2070s, and 2090s under SSP2–4.5 and SSP5–8.5 scenarios. The models perform well in predicting suitable areas with an AUC value of &gt;0.96 for Ale, Goma, Gera, and Yayu and &gt; 0.86 for Limu Seka. Rainfall and temperature variables are the most important factors for predicting coffee area suitability. Under the SSP2–4.5 scenario, the study predicts an overall increase in suitable areas in Ale (+19 %), Gera (+41 %), Goma (+4 %), Limu Seka (+124 %), and Yayu (+21 %) at the end of the century, while most current production areas remain suitable. In the SSP5–8.5 scenario, however, we expect suitable areas to increase in Ale (+16 %), Gera (+52 %), Limu Seka (+71 %), and losses in Goma (−0.5 %), and Yayu (−47 %). Problematically, projected suitable coffee production sites overlap by 25 % to 90 % with areas currently designated as forest under the Global Forest Cover 2020 map, potentially placing production in those areas off limits for export to the European Union under the provisions of the EUDR 2023/1115 regulation. We therefore conclude that many areas in the region that could become newly suitable for coffee production may not be commercially viable. The heterogeneity of primary local drivers of coffee suitability means that micro-scale spatial analyses of climate change impacts on coffee production could provide valuable insights for other regions in planning targeted and effective climate adaptation strategies.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103392"},"PeriodicalIF":7.3,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144932452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Plant leaf classification using the multiscale entropy of curvature and feature aggregation","authors":"Raphael G. Pinheiro ,&nbsp;José G.F. Lopes ,&nbsp;Marcelo M.S. Souza ,&nbsp;Fátima N.S. Medeiros","doi":"10.1016/j.ecoinf.2025.103373","DOIUrl":"10.1016/j.ecoinf.2025.103373","url":null,"abstract":"<div><div>This paper presents a methodology for classifying plant leaves on the basis of handcrafted features derived from the multiscale entropy of curvature and texture, as well as deep features obtained from convolutional neural networks (CNNs). We propose three object descriptors on the basis of the multiscale entropy of curvature. These object descriptors rely on the differential entropy of the probability distributions of multiscale curvatures to create a coarse-to-fine representation of the shape contour. Thus, we present a descriptor that aggregates the multiscale entropy of curvature, bending energy of curvature, and texture features to improve feature extraction of object signatures and subtle texture details of leaf images. The texture descriptor combines the statistics of the local binary pattern and gray-level co-occurrence matrix. We compare our handcrafted descriptors with deep features from various CNNs in multiclass classification using the random forest classifier, replacing the fully connected layer of the CNNs with this classifier. The experiments were conducted on four public leaf datasets: Plantscan, MED117, Flavia, and Swedish. The results of the F1-score and accuracy metrics, which exceed 99.50%, validate the aggregation strategy and show that it is competitive and powerful. The results also confirm that the proposed strategy outperformed six different sets of deep features according to the F1-score and accuracy. Moreover, the handcrafted descriptors achieved better results with 40 features than LeNet’s 50 features. The qualitative analysis of the multidimensional data visualization results prove that combining different shape features and texture details improved the description of the leaf images, as it provided better intraclass compactness and interclass separation in these datasets.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103373"},"PeriodicalIF":7.3,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144830212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The influence of pH and temperature on benthic chlorophyll-a: Insights from SHAP-XGBoost and random forest models","authors":"Sangar Khan ,&nbsp;Noël P.D. Juvigny-Khenafou ,&nbsp;Tatenda Dalu ,&nbsp;Paul J. Milham ,&nbsp;Yasir Hamid ,&nbsp;Kamel Mohamed Eltohamy ,&nbsp;Habib Ullah ,&nbsp;Bahman Jabbarian Amiri ,&nbsp;Hao Chen ,&nbsp;Naicheng Wu","doi":"10.1016/j.ecoinf.2025.103355","DOIUrl":"10.1016/j.ecoinf.2025.103355","url":null,"abstract":"<div><div>Biological threats to river health relate to algal biomass, for which benthic chlorophyll–<em>a</em> (chl–<em>a</em>) is an indicator; consequently, predicting chl–<em>a</em> helps understand ecosystem dynamics. There is little information on machine learning predictive models of benthic chl–<em>a</em> and input parameters in lotic ecosystems, and to fill this gap, we predict benthic chl–<em>a</em> levels in China's Thousand Islands Lake (TIL) watershed using machine learning algorithms. Water samples for nutrient and metal analysis were collected across 147 sites in the TIL catchment. We employed Random Forest (RF), eXtreme gradient boosting (XGBoost) and SHAP-enhanced eXtreme gradient boosting (SHAP XGBoost) models, alongside Support Vector Regression (SVR), to predict chl–<em>a</em> levels in diverse reaches and identify the key determinants. The XGBoost outperformed the RF model in the test, training and validation datasets. In the SHAP XGBoost, pH was the most important characteristic, followed by mean average temperature (AT). The SVR demonstrated that AT is vital for the upper and middle catchment reaches, while pH is more important in the lower reaches. In partial dependence plots, the chl–<em>a</em> concentration depended highly on pH and AT. High pH and AT released P from stream colloids, lowered colloid adsorption, increasing chl–<em>a</em> concentration. We concluded that the SHAP XGBoost model could be used to identify the key determinants of chl–<em>a</em> from chemical and physical variables in the lotic system.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103355"},"PeriodicalIF":7.3,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144865176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic mapping of dissolved oxygen in freshwater aquaculture ponds using UAV multispectral imagery","authors":"Xingyu Liu ,&nbsp;Yancang Wang ,&nbsp;Xiaohe Gu ,&nbsp;Mengjie Li ,&nbsp;Wenxu Lv ,&nbsp;Xuqing Li ,&nbsp;Ruiyin Tang ,&nbsp;Guangxin Chen ,&nbsp;Baoyuan Zhang ,&nbsp;Shuaifei Liu ,&nbsp;Fajian Zong ,&nbsp;Yongkun Ji ,&nbsp;Xiaolong Yu ,&nbsp;Tianen Chen","doi":"10.1016/j.ecoinf.2025.103388","DOIUrl":"10.1016/j.ecoinf.2025.103388","url":null,"abstract":"<div><div>Dissolved oxygen (DO) is an important indicator of the water health of the freshwater aquaculture pond. However, since DO is a non-photosensitive parameter, it is difficult to directly inverse using UAV imaging technology. We proposed an estimation method of DO based on UAV multispectral data and machine learning algorithms. The method utilizes chlorophyll-a (Chl-a) and spectral indices as input features to accurately estimate DO content in water bodies. UAV images were collected in six periods at two aquaculture ponds. Machine learning algorithms were applied to map Chl-a concentration in each aquaculture pond, and a DO estimation model was developed through the relationship between Chl-a, spectral index and DO. The model was validated using measured samples, and the spatial and temporal variations in DO at the two freshwater aquaculture ponds were analyzed. The findings demonstrated that the model exhibited suboptimal performance when solely utilising spectral index. However, the incorporation of Chl-a as an input feature resulted in a substantial enhancement in model performance, in comparison to the utilisation of only spectral index. The RF model performed well during both training and testing phases at the first freshwater aquaculture pond, achieving R² = 0.87, RMSE = 1.785 mg/L, and MAE = 1.512 mg/L for the testing set. Concurrently, the validation in the other two periods(GC - August and October 2023 and PK-April and May 2024) further confirmed the model's generalization ability, with R² = 0.84, RMSE = 2.245 mg/L, and MAE = 1.251 mg/L. Similarly, the model showed robust performance at the second freshwater aquaculture pond, achieving R² = 0.85, RMSE = 3.743 mg/L, and MAE = 2.730 mg/L. UAV multispectral imaging technology combined with this method can efficiently and accurately capture the spatial and temporal distribution of DO in freshwater aquaculture pond, supporting aquaculture management.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103388"},"PeriodicalIF":7.3,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144830213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Four decades of spatio-temporal trends in Miankaleh Wetland´s water body and vegetation as revealed by remote sensing time series","authors":"Nima Arij ,&nbsp;Hooman Latifi ,&nbsp;Arvin Fakhri ,&nbsp;Rohollah Esmaili","doi":"10.1016/j.ecoinf.2025.103374","DOIUrl":"10.1016/j.ecoinf.2025.103374","url":null,"abstract":"<div><div>Coastal wetlands offer essential ecosystem services, but are increasingly threatened by anthropogenic activities and climate change. These disrupt regional patterns, necessitating time series analyses to inform their long-term trends. Remote sensing provides cost-effective alternatives to demanding traditional wetland monitoring. Here, we employed a 40-year time series of Landsat data, supplemented by Sentinel-1 SAR imagery and Sentinel-2 multispectral data for enhanced recent-period analysis, and applied non-parametric trend analysis to examine changes in water bodies, vegetation, and climatic conditions in Miankaleh peninsula, encompassing an extensive Ramsar site in Iran. We utilized spectral indices and random forest classification to derive the area of water bodies and vegetation, followed by identifying significant trends using various trend analysis methods: Mann-Kendall (MK), Modified Mann-Kendall (MMK), Sequential Mann-Kendall (SeqMK), Seasonal Mann-Kendall (SMK), Sen’s Slope (SS), and LOcally Estimated Scatterplot Smoothing (LOESS). Findings showed a significant reduction in water area (30,700 ha, SS = -1.074) and an increase in vegetation cover (31,400 ha, SS = 1.365) from baseline levels. Among climatic factors, groundwater levels (SS = -0.214) and evaporation (SS = -0.312) were most influential on the wetland. The MMK, accounting for data autocorrelation, provided more accurate results compared to MK, while SeqMK detected important trend change points that were mostly missed by MMK. LOESS visualized local, nonlinear changes and identify subtle trend shifts. The results underscore significant ecological shifts, particularly the reduction of water bodies, which threaten the wetland's functionality. We provide general and case-specific considerations on the sole and complementary application of non-parametric trend analysis approaches, expanding insights into ecological processes in coastal wetlands with broader implications for similar ecosystems.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"91 ","pages":"Article 103374"},"PeriodicalIF":7.3,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144831239","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}