Ecological Informatics最新文献

{"title":"A dissimilarity-adaptive cross-validation method for evaluating geospatial machine learning predictions with clustered samples","authors":"Yanwen Wang ,&nbsp;Mahdi Khodadadzadeh ,&nbsp;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}

{"title":"Convolutional neural networks and vision transformers for Plankton Classification","authors":"Loris Nanni ,&nbsp;Alessandra Lumini ,&nbsp;Leonardo Barcellona ,&nbsp;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}

{"title":"Urban buzz or urban bust? Beekeeping challenges, suitability, and survival insights in Montreal, Canada","authors":"Navid Mahdizadeh Gharakhanlou ,&nbsp;Julien Vadnais ,&nbsp;Liliana Perez ,&nbsp;Nico Coallier","doi":"10.1016/j.ecoinf.2025.103296","DOIUrl":"10.1016/j.ecoinf.2025.103296","url":null,"abstract":"<div><div>The rising interest in urban beekeeping underscores the need to investigate whether urban habitats are sustainable for managed honeybee populations. This study, conducted on the island of Montreal, Canada, aimed to i) assess honeybee colony survival within an urban environment, ii) determine the primary drivers affecting honeybee colony survival, and iii) explore the potential of urban areas to support beekeeping activities. This study applied two distinct survival analysis methods, namely random survival forests (RSF) and time-dependent Cox models, incorporating both static and dynamic geospatial variables including normalized difference vegetation index (NDVI), digital elevation model (DEM), percentages of urban areas and water, floral source diversity, road density, proximity to roads, surrounding hive count, ozone (O₃) concentration, fine particulate matter (PM2.5) levels, maximum temperature, and precipitation. To reflect typical honeybee foraging ranges, two buffer distances (1 km and 3 km) were analyzed, and model performance was assessed using the concordance index (C-index) and integrated Brier score (IBS). For the 1 km buffer, the RSF model achieved a C-index of 0.90 (training) and 0.82 (test) with IBS scores of 0.06 and 0.10, outperforming the Cox model, which showed a C-index of 0.56 (both training and test) and IBS values of 0.19 and 0.18. At 3 km, RSF further improved (C-index: 0.93 (training) and 0.84 (test); IBS: 0.05 (training) and 0.08 (test)), while the Cox model remained lower (C-index: 0.58 (training) and 0.60 (test); IBS: 0.19 (training) and 0.18 (test)). These results confirm RSF's superior performance and suggest that broader spatial context may enhance prediction accuracy. Additionally, our findings revealed that the surrounding hive count was the strongest predictor of beehive survival in both buffer scenarios. At 1 km, road proximity and elevation (i.e., DEM) followed in importance, while at 3 km, elevation and vegetation density (i.e., NDVI) were more influential. A primary outcome of this study was the generation of spatially explicit beehive habitat suitability maps for Montreal. Averaged over 2017–2021, these maps showed that large portions of the island are favorable for urban beekeeping, with 30.94 % of land classified as highly suitable and 38.28 % as moderately suitable, demonstrating strong potential for sustainable apiculture in urban environments. This study contributes to providing insights into urban planning and managed honeybee conservation through suitability mapping and predictor analysis.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103296"},"PeriodicalIF":5.8,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144500870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A framework for predicting zoonotic hosts using pseudo-absences: the case of Echinococcus multilocularis","authors":"Andrea Simoncini ,&nbsp;Dimitri Giunchi ,&nbsp;Marta Marcucci ,&nbsp;Alessandro Massolo","doi":"10.1016/j.ecoinf.2025.103295","DOIUrl":"10.1016/j.ecoinf.2025.103295","url":null,"abstract":"<div><div>Identifying the host range of zoonotic parasites is challenging due to limited data and sampling biases. In particular, while more information exists for susceptible hosts, data on resistant species is extremely scant. <em>Echinococcus multilocularis</em> (Leuckart, 1863) (Cestoda: Taeniidae) is the causative agent of alveolar echinococcosis, one of the most significant food-borne zoonoses worldwide. Using data on susceptibility and competence of Holarctic cricetid and murid rodents, key intermediate hosts for <em>E. multilocularis</em>, we developed models to predict the likelihood of infection for any rodent species in the Holarctic. These models incorporated morphological and ecological characteristics and employed two approaches: Generalized Linear Models (GLM) and Presence-Unlabeled Learning (PU-L), a machine learning technique. To train the models, we defined pseudo-absences based on the bias in research effort. We compared the two algorithms and selected GLM as the most effective, using it to map potentially susceptible rodent species across phylogeny and geographic space. Predictions identified several potentially unreported hosts, suggesting that the current understanding of <em>E. multilocularis</em> host distribution may underestimate the true risk. The predicted richness of intermediate hosts peaked in Central-Eastern Europe, Western North America and Central Asia, while the ratio of predicted hosts to total rodent richness was highest in the northern latitudes and the Tibetan Plateau. The average temperature in the geographic range and range size emerged as the strongest predictors of host susceptibility. The workflow demonstrates promise for application to other host-parasite systems with unknown host ranges.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103295"},"PeriodicalIF":5.8,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144480467","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":"Composite likelihood inference for analysis of individual animal identification data","authors":"Xueli Xu ,&nbsp;Xiaoyue Zhang ,&nbsp;Hal Whitehead ,&nbsp;Dehan Kong ,&nbsp;Ximing Xu","doi":"10.1016/j.ecoinf.2025.103298","DOIUrl":"10.1016/j.ecoinf.2025.103298","url":null,"abstract":"<div><div>Individual identification data collection is a common practice in animal behaviour, movement ecology, and conservation biology. While likelihood analysis is widely employed for ecological insights, the complexity of individual identification data, characterized by numerous interdependent individuals and identification times, makes direct likelihood calculation challenging. To address this, we introduce a composite likelihood inference framework. We establish the consistency and asymptotic normality of maximum composite likelihood estimators within this framework. Furthermore, we develop a composite likelihood-based information criterion for model selection, capable of handling complex individual identification data. Our approach is demonstrated through extensive simulations and applied to the northern bottlenose whale population in the Gully, Nova Scotia. This study provides a statistically rigorous framework for individual animal identification models, with potential applications extending beyond whale populations.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103298"},"PeriodicalIF":5.8,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144500871","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 Gonipterus defoliation levels using multispectral unmanned aerial vehicle (UAV) data in Eucalyptus plantations","authors":"Phumlani Nzuza ,&nbsp;Michelle L. Schröder ,&nbsp;Rene J. Heim ,&nbsp;Louis Daniels ,&nbsp;Bernard Slippers ,&nbsp;Brett P. Hurley ,&nbsp;IIaria Germishuizen ,&nbsp;Benice Sivparsad ,&nbsp;Jolanda Roux ,&nbsp;Wouter. H Maes","doi":"10.1016/j.ecoinf.2025.103301","DOIUrl":"10.1016/j.ecoinf.2025.103301","url":null,"abstract":"<div><div>Invasive insect pest <em>Gonipterus</em> sp. n. 2 (Coleoptera: Curculionidae) threatens <em>Eucalyptus</em> species, causing defoliation and yield loss through adult and larval feeding. Early detection is important for early intervention to prevent pest outbreaks. As conventional insect pest monitoring methods are time-consuming and spatially restrictive, this study assessed the potential of UAV monitoring. Multispectral imagery was obtained with Unmanned Aerial Vehicles (UAVs) across six different stands of young <em>Eucalyptus dunnii</em> with varying levels of <em>Gonipterus</em> sp. n. 2 infestations. Some stands were revisited, a total of 9 datasets were covered. Reference damage levels were obtained through visual assessments of (<em>n</em> = 89–100) trees at each site. Across sites, a decrease in canopy reflectance in both the visual and the near-infrared domains with increasing damage levels was consistently observed. Several vegetation indices showed consistent patterns, but none showed site independence. XGBoost, Support Vector Machine and Random Forest (RF) were used to predict damage levels using five input spectral data types. XGBoost performed best, closely followed by RF. Both models consistently selected very similar features. The best-performing models included reflectance, vegetation indices and grey-level co-occurrence matrix data. When data from 10 different wavelengths were used, the highest classification accuracy was 92 % across all sites in classifying defoliation levels. With a classical 5-band multispectral camera, accuracy was 88 %, but distinguishing medium damage from low remained challenging. However, the method was less reliable when trained and validated on separate fields. This study highlights the potential of multi-site datasets in increasing the model's generalization, using UAV based multispectral imagery to assess <em>Gonipterus</em> sp. n. 2 damage and demonstrating reliable upscaling from individual tree assessments to stand scale. However, it also recognises the difficulty of generating a robust model that performs well on untrained sites.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103301"},"PeriodicalIF":5.8,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144548834","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":"Detecting stress parameters in dromedary camels using computer vision","authors":"Hiba Moideen ,&nbsp;Manar Abu Talib ,&nbsp;Nabil Mansour ,&nbsp;Shaher Bano Mirza ,&nbsp;Ali Bou Nassif ,&nbsp;Simon Zerisenay Ghebremeskel ,&nbsp;Fouad Lamghari ,&nbsp;Yaman Afadar ,&nbsp;Takua Mokhamed","doi":"10.1016/j.ecoinf.2025.103292","DOIUrl":"10.1016/j.ecoinf.2025.103292","url":null,"abstract":"<div><div>Dromedary camels exhibit behavioral responses influenced by both physiological conditions and environmental factors. Poor health, physical or emotional, can manifest as behavioral abnormalities. This study aims to build a video-based stress detection model by analyzing camel behavior under different conditions. Camels from Marmoom Farm, UAE, were observed over eight days: six days included interventions such as blood collection and/or intensive training, and two days followed their typical routine. Video footage was captured from three cameras positioned around the enclosures and pens. Using the YOLOv8 architecture, we developed a model to classify normal behaviors - “standing”, “sitting”, “sleeping” and stress-related behaviors - “distressed sitting”, “moving around uncontrollably”, “pulling on rope”. The model obtained a precision of 0.971, recall of 0.959, mAP50 of 0.985, and mAP50–95 of 0.924. Four camels were closely monitored to analyze correlations between behavioral stress indicators and activities such as blood sampling, race training, and environmental conditions. Results indicate that while high-intensity training often induces stress, individual endurance levels and external factors like weather also significantly influence stress responses. This study presents a novel, automated method for early stress detection in camels, contributing to improved animal welfare and farm management practices.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103292"},"PeriodicalIF":5.8,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144490222","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 generative AI and climate modeling for urban heat island mitigation","authors":"Mo Wang ,&nbsp;Ziheng Xiong ,&nbsp;Shiqi Zhou ,&nbsp;Jiayu Zhao ,&nbsp;Chuanhao Sun ,&nbsp;Yuankai Wang ,&nbsp;Lie Wang ,&nbsp;Soon Keat Tan","doi":"10.1016/j.ecoinf.2025.103284","DOIUrl":"10.1016/j.ecoinf.2025.103284","url":null,"abstract":"<div><div>Conventional urban heat island (UHI) studies often rely on static urban morphology inputs and oversimplified design variables, limiting their ability to support dynamic, climate-responsive urban planning. To address this gap, this study proposes a novel framework that integrates a hybrid generative adversarial network (GAN) with the Urban Weather Generator (UWG) for high-fidelity 3D urban form generation and microclimate simulation. The proposed GAN architecture combines the geometric accuracy of Pix2pix with the style refinement capability of CycleGAN, achieving improved morphologicalrealism (SSIM = 0.754, R² = 0.834 against ground-truth data) and resolving key distortions that impede microclimate analysis. Applied Shenzhen Bay Super Headquarters as a case study, ten urban development plans were generated and evaluated for their thermal performance. Results revealed that plans exceeding a facade-to-site ratio of 5.0 and footprint density of 0.30 showed intensified nocturnal heat retention, with Plan V exhibiting a + 2.3 °C nighttime temperature increase. In contrast, Plan I, with lower morphological density, achieved a 1.8 °C reduction, demonstrating superior heat dissipation. These insights provide actionable guidelines for climate-responsive urban planning, advocating for lower-density layouts with optimized facade exposure and increased vegetative cover. The proposed framework offers a robust tool for planners and policymakers to assess and design urban forms that enhance climate resilience while reducing UHI intensity.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103284"},"PeriodicalIF":5.8,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144480468","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":"Spatio-temporal heterogeneity and topic evolution trends of public carbon neutrality attention in China","authors":"Lifang Fu ,&nbsp;Changjin Ma","doi":"10.1016/j.ecoinf.2025.103274","DOIUrl":"10.1016/j.ecoinf.2025.103274","url":null,"abstract":"<div><div>This study examines the spatiotemporal evolution and regional disparities in public attention to carbon neutrality under the ”dual carbo” goals to inform more effective policy design. Departing from traditional single-dimensional approaches, it introduces an interdisciplinary analytical framework – spatiotemporal measurement, sentiment analysis, and topic evolution – to capture dynamic shifts in public discourse on carbon neutrality in China, based on 119,000 Sina Weibo posts (2018–2023). The study makes the following key contributions: (1) It applies Dagum Gini coefficient decomposition and kernel density estimation to identify regional attention patterns, revealing higher attention in central regions, lower levels in the west, and evident ”multi-polarization” within regions; (2) It develops a CNN-BiLSTM-Attention model for sentiment classification, demonstrating that the emotional polarity of topics such as ”low-carbon lifestyle” closely aligns with policy promulgation frequency; (3) It employs the VSTC clustering model to examine topic evolution, identifying four major thematic trajectories: individual environmental behavior, green economy, global governance, and sustainable development. These reflect a progression from micro-level personal actions to macro-level policies and industrial practices. Overall, this study provides a solid quantitative basis for optimizing carbon neutrality policies in China.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103274"},"PeriodicalIF":5.8,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144500888","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 effect of collinearity between observed and model derived training variables on estuarine algal species distribution models","authors":"Dante M.L. Horemans ,&nbsp;Jennifer C. Lin ,&nbsp;Marjorie A.M. Friedrichs ,&nbsp;Pierre St-Laurent ,&nbsp;Raleigh R. Hood ,&nbsp;Christopher W. Brown","doi":"10.1016/j.ecoinf.2025.103225","DOIUrl":"10.1016/j.ecoinf.2025.103225","url":null,"abstract":"<div><div>Forecasts of organism distributions in time and space are needed to mitigate risks associated with changes in environmental conditions. These forecasts are often generated using correlative species distribution models (SDMs) that relate environmental variables to species presence or abundance. Biological complexity makes the construction of SDMs challenging because the collinearity between the environmental variables used to train the SDM may increase model parameter uncertainty. To analyze the effect of collinearity on SDMs, we (1) train SDMs for seven estuarine algal species commonly observed in the Chesapeake Bay (U.S.A.) using different levels of collinearity in the training information, (2) identify the environmental predictors, and (3) study their association with species presence using two statistical techniques (generalized linear models and regression trees). The novelty of our contribution is that our analysis uses both environmental <em>in situ</em> observations and environmental information generated by a mechanistic model. The environmental variables show strong collinearities in both the <em>in situ</em> observations (32 out of the total of 165 correlations) and mechanistic model output (12 out of the total of 120 correlations). To determine how collinearity between these variables affect our SDM results, we remove environmental variables that surpass a specific correlation threshold. We find that using these two different types of training information (i.e., observed vs. modeled) affects (1) the optimal set of predictors, (2) the associations between environmental variables and algal presence, and (3) the model’s predictive skill. Water temperature is generally selected as an important predictor. Strong positive or negative associations between environmental variables and algal presence are not substantially impacted by the type of training information used. Although removing collinearities may result in the detection of new important predictors, it may also result in a slight decrease (<span><math><mo>∼</mo></math></span> 5 %) of the SDM prediction skill, depending on the species of interest and type of training information. Our findings suggest that the main environmental predictors rely on both the species characteristics and training information type used in SDM construction, and highlight the challenge of interpreting the associations between environmental conditions and species presence predicted by these SDMs. These insights help us to better understand the environmental conditions of importance to these algal species and hence optimize monitoring efforts by revealing which <em>in situ</em> observations are vital to accurately forecast blooms of these estuarine algae.</div></div>","PeriodicalId":51024,"journal":{"name":"Ecological Informatics","volume":"90 ","pages":"Article 103225"},"PeriodicalIF":5.8,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144330063","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}