{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,14]],"date-time":"2026-04-14T23:34:43Z","timestamp":1776209683854,"version":"3.50.1"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,7,9]],"date-time":"2024-07-09T00:00:00Z","timestamp":1720483200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Aquac."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Hypoxia is defined as a critically low-oxygen condition of water, which, if prolonged, can be harmful to fish and many other aquatic species. In the context of ocean salmon fish farming, early detection of hypoxia events is critical for farm managers to mitigate these events to reduce fish stress, however in complex natural systems accurate forecasting tools are limited. The goal of this research is to use a machine learning approach to forecast oxygen concentration and predict hypoxia events in marine net-pen salmon farms.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>The developed model is based on gradient boosting and works in two stages. First, we apply auto-regression to build a forecasting model that predicts oxygen concentration levels within a cage. We take a global forecasting approach by building a model using the historical data provided by sensors at several marine fish farms located in eastern Canada. Then, the forecasts are transformed into binary probabilities that indicate the likelihood of a low-oxygen event. We leverage the cumulative distribution function to compute these probabilities.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results and discussion<\/jats:title><jats:p>We tested our model in a case study that included several cages across 14 fish farms. The experiments suggest that the model can detect future hypoxic events with a commercially acceptable false alarm rate. The resulting probabilistic predictions and oxygen concentration forecasts can help salmon farmers to prioritize resources, and reduce harm to crops.<\/jats:p><\/jats:sec>","DOI":"10.3389\/faquc.2024.1365123","type":"journal-article","created":{"date-parts":[[2024,7,9]],"date-time":"2024-07-09T04:57:59Z","timestamp":1720501079000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":7,"title":["Forecasting ocean hypoxia in salmonid fish farms"],"prefix":"10.3389","volume":"3","author":[{"given":"Vitor","family":"Cerqueira","sequence":"first","affiliation":[]},{"given":"Jo\u00e3o","family":"Pimentel","sequence":"additional","affiliation":[]},{"given":"Jennie","family":"Korus","sequence":"additional","affiliation":[]},{"given":"Francisco","family":"Bravo","sequence":"additional","affiliation":[]},{"given":"Joana","family":"Amorim","sequence":"additional","affiliation":[]},{"given":"Mariana","family":"Oliveira","sequence":"additional","affiliation":[]},{"given":"Andrew","family":"Swanson","sequence":"additional","affiliation":[]},{"given":"Ram\u00f3n","family":"Filgueira","sequence":"additional","affiliation":[]},{"given":"Jon","family":"Grant","sequence":"additional","affiliation":[]},{"given":"Luis","family":"Torgo","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2024,7,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2718","DOI":"10.1007\/s11227-023-05580-x","article-title":"A deep learning-enabled IoT framework for early hypoxia detection in aqua water using light weight spatially shared attention-LSTM network","volume":"80","author":"Arepalli","year":"2024","journal-title":"J. Supercomputing"},{"key":"B2","doi-asserted-by":"publisher","first-page":"112896","DOI":"10.1016\/j.eswa.2019.112896","article-title":"Forecasting across time series databases using recurrent neural networks on groups of similar series: A clustering approach","volume":"140","author":"Bandara","year":"2020","journal-title":"Expert Syst. Appl."},{"key":"B3","doi-asserted-by":"crossref","first-page":"62","DOI":"10.1007\/978-3-642-36318-4_3","article-title":"Machine learning strategies for time series forecasting","volume-title":"Business Intelligence: Second European Summer School, eBISS 2012, Brussels, Belgium, July 15-21, 2012, Tutorial Lectures 2","author":"Bontempi","year":"2013"},{"key":"B4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/2907070","article-title":"A survey of predictive modeling on imbalanced domains","volume":"49","author":"Bronco","year":"2016","journal-title":"ACM Computing Surveys (CSUR)"},{"key":"B5","doi-asserted-by":"publisher","first-page":"736143","DOI":"10.1016\/j.aquaculture.2020.736143","article-title":"Oceanographic processes control dissolved oxygen variability at a commercial atlantic salmon farm: application of a real-time sensor network","volume":"533","author":"Burke","year":"2021","journal-title":"Aquaculture"},{"key":"B6","article-title":"Exceedance probability forecasting via regression for significant wave height forecasting","author":"Cerqueira","year":"2022","journal-title":"arXiv"},{"key":"B7","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1007\/s10844-022-00713-9","article-title":"A case study comparing machine learning with statistical methods for time series forecasting: size matters","volume":"59","author":"Cerqueira","year":"2022","journal-title":"J. Intelligent Inf. Syst."},{"key":"B8","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","author":"Dem\u0161ar","year":"2006","journal-title":"J. Mach. Learn. Res."},{"key":"B9","first-page":"53","article-title":"Activity monitoring: Noticing interesting changes in behavior","author":"Fawcett","year":"1999"},{"key":"B10","doi-asserted-by":"publisher","first-page":"125","DOI":"10.1146\/annurev-statistics-062713-085831","article-title":"Probabilistic forecasting","volume":"1","author":"Gneiting","year":"2014","journal-title":"Annu. Rev. Stat Its Appl."},{"key":"B11","doi-asserted-by":"publisher","first-page":"107518","DOI":"10.1016\/j.knosys.2021.107518","article-title":"Ensembles of localised models for time series forecasting","volume":"233","author":"Godahewa","year":"2021","journal-title":"Knowledge-Based Syst."},{"key":"B12","doi-asserted-by":"publisher","first-page":"900","DOI":"10.3390\/rs15040900","article-title":"A novel deep learning model for mining nonlinear dynamics in lake surface water temperature prediction","volume":"15","author":"Hai","year":"2023","journal-title":"Remote Sens."},{"key":"B13","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1016\/j.ijforecast.2019.05.008","article-title":"Criteria for classifying forecasting methods","volume":"36","author":"Januschowski","year":"2020","journal-title":"Int. J. Forecasting"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1341","DOI":"10.1109\/TNN.2011.2162110","article-title":"Comprehensive review of neural network-based prediction intervals and new advances","volume":"22","author":"Khosravi","year":"2011","journal-title":"IEEE Trans. Neural Networks"},{"key":"B15","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1016\/0304-4076(92)90104-Y","article-title":"Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root","volume":"54","author":"Kwiatkowski","year":"1992","journal-title":"J. econometrics"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1751","DOI":"10.1016\/j.eng.2020.10.023","article-title":"A hybrid neural network model for marine dissolved oxygen concentrations time-series forecasting based on multi-factor analysis and a multi-model ensemble","volume":"7","author":"Liu","year":"2021","journal-title":"Engineering"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1346","DOI":"10.1016\/j.ijforecast.2021.11.013","article-title":"M5 accuracy competition: Results, findings, and conclusions","volume":"38","author":"Makridakis","year":"2022","journal-title":"Int. J. Forecasting"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.aquaculture.2010.11.020","article-title":"Environmental drivers of atlantic salmon behaviour in sea-cages: a review","volume":"311","author":"Oppedal","year":"2011","journal-title":"Aquaculture"},{"key":"B19","doi-asserted-by":"publisher","first-page":"101480","DOI":"10.1016\/j.ecoinf.2021.101480","article-title":"Interpretable machine learning to forecast hypoxia in a lagoon","volume":"66","author":"Politicos","year":"2021","journal-title":"Ecol. Inf."},{"key":"B20","unstructured":"RemenM.\n          NorwayInstitute of Biology, University of BergenThe oxygen requirement of atlantic salmon (salmo salar l.) in the on-growing phase in sea cages2012"},{"key":"B21","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1037\/1082-989X.7.2.147","article-title":"Missing data: our view of the state of the art","volume":"7","author":"Schafer","year":"2002","journal-title":"psychol. Methods"},{"key":"B22","doi-asserted-by":"publisher","first-page":"75","DOI":"10.1016\/j.ijforecast.2019.03.017","article-title":"A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting","volume":"36","author":"Smyl","year":"2020","journal-title":"Int. J. Forecasting"},{"key":"B23","doi-asserted-by":"publisher","first-page":"2","DOI":"10.1111\/j.1753-5131.2008.01002.x","article-title":"Global aquaculture and its role in sustainable development","volume":"1","author":"Subasinghe","year":"2009","journal-title":"Rev. aquaculture"},{"key":"B24","doi-asserted-by":"publisher","first-page":"7067","DOI":"10.1016\/j.eswa.2012.01.039","article-title":"A review and comparison of strategies for multi-step ahead time series forecasting based on the nn5 forecasting competition","volume":"39","author":"Taieb","year":"2012","journal-title":"Expert Syst. Appl."},{"key":"B25","doi-asserted-by":"publisher","first-page":"1069","DOI":"10.1111\/rssa.12176","article-title":"Using auto-regressive logit models to forecast the exceedance probability for financial risk management","volume":"179","author":"Taylor","year":"2016","journal-title":"J. R. Stat. Society: Ser. A (Statistics Society)"},{"key":"B26","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1007\/s10618-005-0039-x","article-title":"Characteristic-based clustering for time series data","volume":"13","author":"Wang","year":"2006","journal-title":"Data Min. knowledge Discovery"},{"key":"B27","first-page":"359","article-title":"Learning to predict rare events in event sequences","volume":"98","author":"Weiss","year":"1998","journal-title":"KDD."},{"key":"B28","doi-asserted-by":"publisher","first-page":"817","DOI":"10.2307\/1912934","article-title":"A heteroscedasticity-consistent covariance matrix estimator and a direct test for heteroscedasticity","author":"White","year":"1980","journal-title":"Econometrica: J. Econometric Soc."},{"key":"B29","volume-title":"Macquarie Harbour Oxygen Process model (FRDC 2016-067): CSIRO Final Report.","author":"Wild-Allen","year":"2020"},{"key":"B30","first-page":"17","article-title":"Effects of fish farming on macroinvertebrates. comparison of three localities suffering from hypoxia","volume-title":"Water Effluent and Quality, with Special Emphasis on Finfish and Shrimp Aquaculture","author":"Yokoyama","year":"1997"},{"key":"B31","doi-asserted-by":"publisher","first-page":"118518","DOI":"10.1016\/j.watres.2022.118518","article-title":"A physics-informed statistical learning framework for forecasting local suspended sediment concentrations in marine environment","volume":"218","author":"Zhang","year":"2022","journal-title":"Water Res."},{"key":"B32","doi-asserted-by":"publisher","first-page":"102385","DOI":"10.1016\/j.ecoinf.2023.102385","article-title":"Predictions of runoff and sediment discharge at the lower yellow river delta using basin irrigation data","volume":"78","author":"Zhao","year":"2023","journal-title":"Ecol. Inf."}],"container-title":["Frontiers in Aquaculture"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/faquc.2024.1365123\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,7,9]],"date-time":"2024-07-09T04:58:08Z","timestamp":1720501088000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/faquc.2024.1365123\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7,9]]},"references-count":32,"alternative-id":["10.3389\/faquc.2024.1365123"],"URL":"https:\/\/doi.org\/10.3389\/faquc.2024.1365123","relation":{},"ISSN":["2813-5334"],"issn-type":[{"value":"2813-5334","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,7,9]]},"article-number":"1365123"}}