Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Towards Improved Harmful Algal Bloom Forecasts: A Comparison of Symbolic Regression With Dome and Stream Learning Performance

  • Molares-Ulloa, Andrés
  • Rocruz, Elisabet
  • Rivero, Daniel
  • Padin, Xosé A.
  • Nolasco, Rita
  • Dubert, Jesús
  • Fernández-Blanco, Enrique
©2025 Elsevier B.V. All rights reserved. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/bync-nd/4.0/. This version of the article has been accepted for publication in Computers and Electronics in Agriculture. The Version of Record is available online at https://doi.org/10.1016/j.compag.2025.110112, [Abstract]: Diarrhetic Shellfish Poisoning (DSP) is a global health issue caused by shellfish contaminated with toxins from dinoflagellates, posing significant risks to public health and the shellfish industry. Harmful Algal Blooms (HABs), driven by toxin-producing algae like DSP, require effective monitoring and forecasting systems. Predicting HABs is challenging due to the time-series nature of the problem, influenced by historical seasonal patterns and recent anomalies from meteorological and oceanographic changes. Stream Learning shows promise for handling time-series problems with concept drifts but has yet to be validated for HAB prediction compared to Batch Learning. Limited historical data availability in oceanography highlights the importance of advanced tools like the CROCO ocean hydrodynamic model, which provides high-resolution temporal and spatial data. This study developed a machine learning workflow to predict toxic dinoflagellate (Dinophysis acuminata) cell counts, comparing seven algorithms across two learning paradigms. The CROCO model data addressed historical data gaps. The DoME model, with an average of 0.77 for 3-day-ahead predictions, proved the most effective and interpretable, underscoring the value of model explainability and rigorous comparison methodologies., The authors want to acknowledge the support from INTECMAR, who has provided part of the data for this work; and CESGA, who allowed the conduction of the tests in their installations. Thanks are also due for the financial support to CESAM by FCT/MCTES, Portugal (UIDP/50017/2020+UIDB/50017/2020+LA/P/0094/2020), through national funds, and the co-funding by the FEDER, Spain, within the PT2020 Partnership Agreement and Compete 2020. Funding for open access charge: Universidade da Coruña/CISUG, Spain. CITIC is funded by the Xunta de Galicia, Spain through the collaboration agreement between the Regional Ministry of Culture, Education, Vocational Training and Universities and the Galician universities to strengthen the research centres of the Galician University System (CIGUS). Grant PID2021-126289OA-I00 funded by MCIN/AEI/10.13039/501100011033, Spain and by ERDF A way of making Europe. Elisabet Rocruz was supported by the Portuguese Science and Technology Foundation (FCT) through PhD fellowship PD/BD/143085/2018. This research was funded by REDEIRA, Spain (Research, development and innovation of a Coastal Observation network: Ría de Arousa) project (Proyecto Estratégico Orientado a la Transición Ecológica y a la Transición Digital; Ref: TED2021-132188B-I00)., Portugal. Fundação para a Ciência e a Tecnologia; UIDP/50017/2020, Portugal. Fundação para a Ciência e a Tecnologia; UIDB/50017/2020, Portugal. Fundação para a Ciência e a Tecnologia; LA/P/0094/2020, Portugal. Fundação para a Ciência e a Tecnologia; PD/BD/143085/2018
RUC. Repositorio da Universidade da Coruña
Publicación Artículo científico (JournalArticle)

Advanced ocean modelling and machine learning to forecast Dinophysis acuminata blooms: A tool to support shellfish farming management

  • Rocruz, Elisabet
  • Molares-Ulloa, Andrés
  • Padin, Xosé A.
  • Nolasco, Rita
  • Rivero, Daniel
  • Fernández-Blanco, Enrique
  • Pazos, Yolanda
  • Dubert, Jesús
Data related to cell counts of D. acuminata is available upon request to the Instituto Tecnolóxico para o Control do Medio Mariño de Galicia (INTECMAR). The remaining dataset used in this study is available on Zenodo ( https://doi.org/10.5281/zenodo.17143117) and the code used to reproduce our machine learning models results is accessible on GitHub ( https://github.com/AndresMolares/hab_featureSelection_study)., [Abstract]: The presence of toxin-producers phytoplankton is a natural phenomenon that threatens marine ecosystems, endangers human health, and causes significant economic losses in shellfish harvesting. The Galician Rías Baixas (NW Spain) are one of the main mussels producing regions worldwide and the leading producer in Europe. Annual occurrence of Dinophysis acuminata, responsible for diarrhetic shellfish poisoning toxins, lead to a ban on the mussel harvesting for several months, each year. To help mitigate these impacts, this study explores the prediction of D. acuminata cells densities 3-days ahead in the outer and inner parts of three of the Rías Baixas (Arousa, Pontevedra and Vigo), testing three local machine learning (ML) models: Artificial Neural Network (ANN), Random Forest (RF) and Support Vector Machine (SVM). Local ML models were selected to account for the differences in occurrence and variability of D. acuminata densities across the different parts of each Ría. These ML models were assessed by (1) reducing the number of features through a feature selection approach to identify the most relevant ones, (2) exploring different sets of features and (3) comparing models trained with 7 and 30 days of past information. The dataset combined daily hydrodynamic and biological features, from 2013 to 2019, obtained from a high-resolution 3D hydrodynamic model (CROCO), and in-situ observations. Our results show that RF provided the best predictive performance. Increasing the number of days of past information did not significantly improve results, as similar averaged R2 values were obtained for 7 and 30 days: 0.75 for Ría de Arousa, 0.72 for Ría de Pontevedra, and 0.67 for Ría de Vigo. Feature selection process showed that, as expected, previous cells densities of D. acuminata were essential for capturing bloom timing and amplitude. Also, the temperature, salinity, and the vertical and meridional components of current velocity were key predictors at outer stations of the Ría de Pontevedra and Vigo, where more features were required. In contrast, for the other stations, good predictions were achieved using only five features. This study represents one of the first attempts to predict D. acuminata in the Rías Baixas using local ML models. Our findings highlight the need for local approaches, as bloom dynamics vary between Rías and within different parts of each Ría. We also demonstrate the value of hydrodynamic model outputs to train ML models and compensate for the lack of long-term, spatially extensive in-situ data., The authors would like to thank INTECMAR for creating the dataset related to the cell count of Dinophysis acuminata and CESGA, who allowed the run of the simulations in their installations. Thanks are also due for the financial support to CESAM by FCT/MCTES (UIDP/50017/2020+UIDB/50017/2020+LA/P/0094/2020), through national funds. ER was supported by the Portuguese Science and Technology Foundation (FCT) through PhD fellowship PD/BD/143085/2018, within the scope of the National Strategic Reference Framework (NSRF) and the Human Potential Operational Programme (POPH), co-financed by the European Fund and national funds from the Ministry of Science, Technology and Higher Education (MC-TES). ER has also received funding through DATAMARE, from Galicia Marine Science programme, which forms part of the Complementary Science Plans for Marine Science of Ministerio de Ciencia, Innovación and Universidades included in the Recovery, Transformation and Resilience Plan (PRTR-C17.I1), funded through Xunta de Galicia with NextGenerationEU and the European Maritime Fisheries and Aquaculture Funds. ER and RN were supported by MITECO programme for the Spanish Recovery, Transformation and Resilience Plan (European Union Recovery and Resilience Mechanism established by Regulation (EU) 2020/2094), funded by the European Union -NexGenerationEU-. CITIC is funded by the Xunta de Galicia through the collaboration agreement between the Regional Ministry of Culture, Education, Vocational Training and Universities and the Galician universities to strengthen the research centres of the Galician University System (CIGUS). Grant PID2021-126289OA-I00 funded by MCIN/AEI/10.13039/501100011033 and by ERDF A way of making Europe. This work was also partially supported by the Xunta de Galicia and the ERDF Funds A way of making Europe with grant (Ref. ED431C 2022/46). This work was supported by MAGIC project (PID2024-156623OB-C22) funded by MICIU/AEI/10.13039/501100011033 and by FEDER, UE; and by REDEIRA project (TED2021-132188B-I00) funded by MICIU/AEI/10.13039/501100011033 and by Unión Europea NextGenerationEU/PRTR . Part of the funding for this work also comes from the consolidation and structuring funds for Competitive Research Units of the GAIN-Xunta de Galicia, modality A: Competitive Reference Groups (IN607A2025-05). We thank to the editor and reviewers for their constructive comments and rapid response, which greatly helped us improve the manuscript., Portugal. Fundação para a Ciência e a Tecnologia; UIDP/50017/2020, Portugal. Fundação para a Ciência e a Tecnologia; UIDB/50017/2020, Portugal. Fundação para a Ciência e a Tecnologia; LA/P/0094/2020, Portugal. Fundação para a Ciência e a Tecnologia; PD/BD/143085/2018, Xunta de Galicia; ED431C 2022/46, Xunta de Galicia; IN607A2025-05, https://zenodo.org/records/17143117, https://github.com/AndresMolares/hab_featureSelection_study
DIGITAL.CSIC. Repositorio Institucional del CSIC
Datos de investigación Set de datos (Dataset)

Daily maximum, minimum, range and deviation of time series of pH and oxygen concentration in the water column and benthic compartment in the Spanish littoral

  • Hendriks, Iris E.
  • Iuculano, Francesca
  • Vaquer-Sunyer, Raquel
  • Mayol, Elvira
  • González-Dávila, Melchor
  • Santana-Casiano, Juana Magdalena
  • Velo, Antón
  • Padín, X. A.
  • Pérez, Fiz F.
  • Flecha, Susana
  • Ricart, Aurora M.
  • Steckbauer, Alexandra
  • Duarte, Carlos M.
[Description of methods used for collection/generation of data] Water column: Water column data were acquired at 1 m depth in the Bay of Palma, at 4 m depth in Cabrera PN and at 12 m in CanOA buoy by using a SAMI-pH (Sunburst Sensors LCC) sensors attached to a mooring line. The pH sensors were measuring pH, in the total scale (pH??), hourly, since December 2018 in the Bay of Palma and since October 2019 in Cabrera, and every three hours since March 2020 in the CanOA buoy. Monthly maintenance of the sensors including data download and surface cleaning for Palma Bay and Cabrera PN and every two-three months for Gran Canaria. Since the SAMI sensor failed due to bubbles in the tubing, the pH was calculated in the total scale (pHT) with alkalinity determined from salinity and pCO2 variables for CanOA buoy data. Data were acquired at 0.8 m depth in the Ria de Arousa using a prototype instrument called SURCOM from pH Honeywell Durafet III ISFET technology. The pH sensor was measuring pH, in the total scale (pH??), every 10-5 minutes since July 2020. Monthly maintenance and cleaning of the complete mooring was performed. Temperature and conductivity data for the Cabrera PN and Palma Bay mooring line were obtained starting October 2019 and December respectively, with a CT SBE37 (SBE37SMP-ODO-RS232 Sea-Bird Scientific ©). Additionally, dissolved oxygen (DO) data from a SBE 63(Sea-Bird Scientific ©) sensor attached to the CT in Cabrera and Palma Bay were used. Sea surface temperature and salinity data for the CanOA buoy were obtained hourly starting March 2020, with a SBE 37-SI/SIP Thermosalinometer MicroCAT sensor (Sea-Bird Scientific ©). Despite the fact that the oxygen sensor was installed on the buoy, oxygen data have not been used due to their low stability and biofouling problems. Salinity data for the Ria Arousa station was obtained starting July 2020 every 10 minutes through the INTECMAR website (http://www.intecmar.gal/Plataformas/platInfo.aspx?idEst=15001) with a CTD model SBE 37-SIP. No oxygen data was provided for this dataset. Multiparametric sensors: Multiparametric sensors data were acquired by sensors located within meadow (see 5) at a height approx. 0.20 from the sediment. As multiparametric sensors a Hydrolab HL4, Hydrolab DSX5, YSI EXO2, Eureka Manta¨, SeapHOx and SAMI-pH were used. Additionally, a custom build array was used courtesy by Dr. Tommy Moore. This array was built to monitor pH variation within a 1m3 area within marine vegetation, the custom built multi-sensor array consisted of an autonomous data logger (Datataker DT85) in a water-tight housing with 16 pre-amplified pH electrodes (Omega, PHE-1304-NB) attached by 5 m long cables, for more information see (Krause-Jensen et al. 2015). The sensors were configured in-situ in a three-dimensional array on the metal frame with 1m between each side (replica), with 4 sensors at 0.1 m from the bottom, 4 sensors at 0.2 m, 4 sensors just underneath the canopy and 4 above the canopy.Ê All pH sensors were calibrated, allowing ample time between each reading for the sensors to stabilize, using NIST buffers of pHNBS 4.0, 7.0 and 10.0. Krause-Jensen D, Duarte CM, Hendriks IE, Meire L, Blicher ME, Marbˆ N, Sejr MK. 2015. Macroalgae contribute to nested mosaics of pH variability in a subarctic fjord. Biogeosciences 12:4895-4911. Data from SURCOM-pH system were acquired by a Honeywell Durafet III sensor with custom electronics described in Velo and Padin (2022). The sensor was located at about 0.8m depth on the same wire as INTECMAR CTDs. Raw data was transmitted every 10.5 min in near-real-time to IIM servers through LoRA/SIGFOX network, where itÕs calibrated by discrete spectrophotometric samples taken by CETMAR and analysed by PEACHESEA CT service at IIM-CSIC., [Methods for processing the data] Raw data was recorded with a frequency between 5 and 30 minutes for multiparametric sensors and 60 to 180 minutes for water column data. Minimum, maximum and average values for all variables were calculated by extrapolating the averages from 24 hours of measurements starting from midnight (local time), unless specified otherwise. Ranges over 24 hours of pH and DO (where data available) were calculated as the differences between maximum and minimum values within the 24 hours period., This research was funded by the European Union Ð NextGeneration EU Ð as part of the MITECO program for the Spanish Recovery, Transformation and Resilience Plan (Recovery and Resilience Facility of the European Union established by the Regulation (EU) 2020/2094), and was entrusted to CSIC, AZTI, SOCIB, and the universities of Vigo and Cadiz. Funding for the BOATS time series: CYCLE project PID2021-123723OB-C21 ÒComplex DYnamics of CoastaL Ecosystems: Resilience to Climate Change of the Spanish Ministery of Science, innovation and Universities. Project: SAMI -Adquisici—n de sensor oceanogr‡fico de medida de pH (AAEE111/2017), Govern de les Illes Balears. Project: Servicios ecosistŽmicos; Sumideros de carbono y oxigenaci—n del agua como incentivo para conservaci—n de praderas de Posidonia oceanica (Posi-COIN), Fundaci—n BBVA. Funding for the Cortegada system comes from REDEIRA(TED2021-132188B-I00) project funded by MICIU/AEI/10.13039/501100011033 and Uni—n Europea NextGenerationEU/PRTR. Fundings for CanOA-CanBIO Project from Gobierno de Canarias and Fundaci—n Loro Parque 2019-2027. SumaEco project RTI2018-095441-BC21 Sostenibilidad de ecosistemas marinos costeros en el contexto del cambio global en el Mar Mediterr‡neoÓ. Funding for Portlligat, Delta de l'Ebre and Menorca, Sa Nitja time series came from "la Caixa" Foundation (fellowship Junior Leader LCF/BQ/PI23/11970014)Ó and project CNS2023-145492 funded by MCIN/AEI/10.13039/ 501100011033 y European Union ÒNextGenerationEUÓ/PRTR., This work contributes to the Institut de Ciències del Mar "Severo Ochoa Centre of Excellence" accreditation CEX2024-001494-S funded by AEI 10.13039/501100011033 of the Spanish Ministry of Science and Innovation., File List: H4.2_database_pH_T_Watercolumn.csv H4.2_database_pH_T_sensors.csv readme.txt, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Otros (Other)

Dinámica de Dinophysis spp. y cierres de bateas por DSP bajo variabilidad ambiental modelada y observada por satélite en las Rías Baixas (costa de Galicia, 2013–2019), Dinámica de Dinophysis spp. e peches de bateas por DSP baixo variabilidade ambiental modelada e observada por satélite nas Rías Baixas (costa de Galicia, 2013–2019), Dynamics of Dinophysis spp. and mussel raft closures due to DSP under modeled and satellite-observed environmental variability in the Rías Baixas (NW Iberian Peninsula, 2013–2019)

  • Padín, X. A.
  • Rodríguez Cruz, Elisabet
  • Nolasco, Rita
  • Cacabelos, Jesús
  • Longa, Ángeles
  • Pazos, Yolanda
  • Otero, Pablo
  • Babarro, José M. F.
  • Dubert, Jesús
  • Velo, Antón
14:08 minutos de duración, calidad 3840x2160px (4k), relación de aspecto 16:9, en color, sin sonido, bitrate 2kbps, Este vídeo muestra la evolución diaria de los contajes de Dinophysis spp. y los cierres de los polígonos de cultivo de mejillón por toxinas lipofílicas asociados a estos dinoflagelados publicados por INTECMAR en las Rías Baixas entre el 2013 y el 2019. El vídeo también incluye mapas superficiales de variables ambientales procesadas específicamente para este estudio: temperatura, salinidad y corrientes (estimadas unificando los modelos hidrodinámicos Regional Ocean Modeling System, ROMS y Coastal and Regional Ocean COmmunity model, CROCO), y clorofila a del producto Atlantic Ocean Colour BGC L4 del Servicio de Monitorización del Medio Marino de Copernicus, índice de afloramiento y viento procedentes de www.indicedeafloramiento.ieo.es; el sumatorio de la descarga fluvial de los ríos Ulla y Umia calculado por Meteogalicia; y un indicador de disponibilidad de alimento resultado del producto de la velocidad de la corriente y la clorofila. Para mejorar la legibilidad de los datos, la descarga fluvial se muestra en una escala logarítmica que permite apreciar tanto los caudales bajos como las grandes riadas. El viento se representa mediante una rosa de los vientos dinámica en coordenadas polares y el índice de afloramiento utiliza una escala simétrica personalizada para resaltar tanto los eventos de upwelling como de downwelling. Finalmente, la disponibilidad de alimento se visualiza a través de un campo de vectores o flechas que indica la dirección y fuerza de la corriente sobre los mapas de clorofila. Esta animación permite visualizar la evolución en el tiempo y las relaciones entre las condiciones del entorno marino y atmosférico. El vídeo no solo ayuda a detectar patrones estacionales y eventos climáticos extremos, sino que ofrece una perspectiva unificada de cómo los factores físicos y biológicos, junto con las decisiones de gestión, condicionan el día a día del cultivo de mejillón en las Rías Baixas. En la web del proyecto REDEIRA (www.redeira.iim.csic.es), el vídeo está disponible en una versión interactiva, que permite seleccionar eventos concretos de la línea temporal, así como ajustar la velocidad de reproducción y su resolución, Este vídeo fue financiado por el proyecto REDEIRA (Proyecto Estratégico Orientado a la Transición Ecológica y a la Transición Digital; Ref: TED2021-132188B-I00), Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Transboundary pathways of Dinophysis acuta in the Iberian upwelling system using biophysical modeling

  • Rodríguez Cruz, Elisabet
  • Nolasco, Rita
  • Padín, X. A.
  • Moita, Maria Teresa
  • Sobrinho-Gonçalves, André
  • Pazos, Yolanda
  • Oliveira, Paulo B.
  • Otero, Pablo
  • Pérez, Fiz F.
  • Dubert, Jesús
16 pages, 10 figures, 2 tables.-- Data will be made available on request, Harmful algal blooms (HABs) pose major environmental, socio-economic, and health risks, mainly due to toxins produced by microalgae such as Dinophysis acuminata and D. acuta, responsible for diarrhetic shellfish poisoning. These blooms are particularly problematic in shellfish production areas, including the Galician Rías Baixas (NW Spain), at the northern limit of the Canary Current Upwelling System. This region, one of the world’s leading mussel producers, suffers recurrent economic losses from HABs. Blooms of D. acuta typically occur from late summer to autumn, and when preceded by D. acuminata, harvesting closures are extended. To investigate the role of alongshore transport in D. acuta dynamics, we applied a biophysical modeling approach combining a high-resolution hydrodynamic model (CROCO) with a lagrangian model (ROMSPath), covering August–November of 2015, 2017, and 2019. Monitoring programs in Portugal and Spain confirmed D. acuta presence in Aveiro (Portugal) and the Rías Baixas (Spain), with cells also detected at intermediate stations in 2015 and 2019, but not in 2017. Lagrangian simulations, validated with monitoring data, revealed poleward transport in 2015 and 2019, driven by alternating wind relaxation and southerly winds that promoted a coastal current extending from the shoreline to the 50 m isobath. In contrast, persistent northerly winds in 2017 prevented the onset of this current, consistent with the absence of cells at intermediate locations. The integration of monitoring and biophysical modeling improves understanding of HAB transport pathways across transnational aquaculture areas, underscoring the need for international cooperation to mitigate their impacts on the shellfish industry, Thanks for the financial support to CESAM by FCT/MCTES (UIDP/50017/2020+UIDB/50017/2020+LA/P/0094/2020), through national funds, and the co-funding by the FEDER, within the PT2020 Partnership Agreement and Compete 2020. Elisabet Rocruz was supported by: the Portuguese Science and Technology Foundation (FCT) through PhD fellowship PD/BD/143085/2018; MITECO programme for the Spanish Recovery, Transformation and Resilience Plan (European Union Recovery and Resilience Mechanism established by Regulation (EU) 2020/2094), funded by the European Union -NexGenerationEU-; and Dixital Marine Data Platform (DATAMARE) of the Galician Marine Sciences Programme, which forms part of the Complementary Science Plans of the Ministry of Science, Innovation and Universities included in the Transformation and Resilience Recovery Plan (PRTR-C17.I1). This work is also a contribution to REDEIRA (Research, development and innovation of a Coastal Observation network: Ría de Arousa) project (Proyecto Estratégico Orientado a la Transición Ecológica a la Transición Digital; Ref: TED2021-132188B-I00) and MAGIC (Modelling and Analysis of phytoplankton Growth in the rías: understanding productIvity and harmful algae Changes) project (Ref: PID2024-156623OB-C22), both supported by MICIU/AEI/10.13039/501100011033/ . Part of the funding for this work also comes from the consolidation and structuring funds for Competitive Research Units of the GAIN-Xunta de Galicia, modality A: Competitive Reference Groups (IN607A2025-05). Funding for open access charge: Consejo Superior de Investigaciones Científicas (CSIC), Peer reviewed