Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Machine learning applied to global scale species distribution models

  • Fuster-Alonso, Alba
  • Mestre-Tomás, Jorge
  • Baez, Jose Carlos
  • Pennino, Maria Grazia
  • Barber, Xavier
  • Bellido, Jose Maria
  • Conesa, David
  • Lopez-Quilez, Antonio
  • Steenbeek, Jeroen
  • Christensen, Villy
  • Coll, Marta
[EN] Species Distribution Models (SDMs) are widely used in ecology to analyze historical and future patterns of marine species distributions. Given the growing impact of climate change, predicting potential shifts in species ranges has become a key challenge. In this study, we apply Bayesian Additive Regression Trees (BART), a non-parametric machine learning algorithm, to estimate and forecast the global distribution of marine turtle species under different climate change scenarios. We model both individual species and their combined functional group, assess their historical and future habitat suitability, and examine the contribution of key environmental predictors. To evaluate BART's performance, we conduct a simulation study under two contrasting distributional scenarios: a cosmopolitan and a persistent species. We also test the sensitivity of BART to pseudo-absence data and compare its performance with MaxEnt and Generalized Additive Models (GAMs). Results indicate that BART performs slightly better overall, particularly under pseudo-absence settings, showing higher accuracy and more stable sensitivity and specificity. These findings highlight BART as a reliable alternative for long-term, global-scale species distribution modeling in marine systems., This study is a contribution to the project ProOceans (Ministerio de Ciencia e Innovacion, Proyectos de I + D + I (RETOS PID2020-118097RB-I00)). AFA received funding from the Spanish project ProOceans through an FPI grant (Ministerio de Ciencia e Innovacion, Proyectos de I + D + I (RETOSPID2020-118097RB-I00)). AFA, MC, JMT and JS acknowledge funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 817578 (TRIATLAS project) and No. 869300 (FutureMares). XB, DC and ALQ acknowledge support by the grant PID2022-136455NB-I00, funded by Ministerio de Ciencia, Innovacion y Universidades of Spain (MCIN/AEI/10.13039/501100011033/FEDER, UE) and the European Regional Development Fund. DC and MGP also acknowledge funding provided by the Conselleria de Educacion, Universidades y Empleo de la Generalitat Valenciana via the project CIAICO/2022/165. Authors also acknowledge institutional support of the "Severo Ochoa Center of Excellence" accreditation (CEX2024-001494-S funded by AEI 10.13039/501100011033). This project was supported by the Catalan government through the iMARES research group (2021 SGR 00435) at the Institute of Marine Sciences (ICM-CSIC) in Barcelona.This work was supported by the project SOSPEN (Spanish National Plan for Scientific and Technical Research and Innovation, 2021, PID2021-124831OA-I00).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

GLOSSA: A user-friendly R Shiny application for Bayesian machine learning analysis of marine species distribution

  • Mestre-Tomás, Jorge
  • Fuster-Alonso, Alba
  • Bellido, Jose M.
  • Coll, Marta
[EN] Species distribution models (SDMs) are one of the most common statistical methods to assess species occupancy and geographic distribution patterns. With the increasing complexity and availability of ecological data in the marine context, many methodological approaches have been developed to support SDM analysis. However, their application often requires expertise in data analysis, statistical modelling and programming, which limits their accessibility for broader use. Here we introduce GLOSSA, an open-source R package and Shiny application designed to make marine species distribution modelling more accessible. GLOSSA provides a user-friendly interface for fitting Bayesian Additive Regression Trees (BART) SDMs using species occurrence and environmental data. GLOSSA guides users through key SDM steps, including data uploading, filtering occurrence data, harmonizing environmental layers, generating pseudo-absences, tuning BART complexity, performing spatial and temporal block cross-validation, visualizing predictions and uncertainty and exporting configuration files to ensure reproducibility. We demonstrate the functionality of GLOSSA through three marine case studies, addressing a range of ecological scenarios at regional and global scales. Along with detailed documentation, examples and tutorials, GLOSSA provides an example of how an intuitive graphical interface can make species distribution modelling accessible to a broad audience., Funding Ministerio de Ciencia e Innovacion, Grant/Award Number: PID2020-118097RB-I00 and PID2021-124831OA-I00; Horizon Europe, Grant/Award Number: 101059877; Agencia de Gestio d'Ajuts Universitaris i de Recerca, Grant/Award Number: 2021 SGR 00435