Machine learning techniques for marine heatwave prediction: a comprehensive review

Abstract

Marine ecosystems and coastal economies are seriously threatened by marine heatwaves (MHWs), which are defined as extended periods of abnormally high sea surface temperatures (SSTs). Accurate and early MHW forecasting has become essential because climate change has increased the frequency and severity of such phenomena. In this review, we examine the application of traditional machine learning (ML) and deep learning (DL) methods for MHW detection and prediction. Specifically, we investigate the algorithms (neural networks, ensemble methods, and hybrid architectures) as well as the input variables, datasets, and evaluation metrics employed. Additionally, we review previous studies conducted on different ocean basins to highlight regional patterns and model transferability. Furthermore, we identify the emerging trends in DL, such as the use of explainable artificial intelligence and physics-guided learning for MHW prediction, and outline key challenges and limitations. Finally, we discuss future directions for improving the accuracy, generalization, and interpretability of MHW forecasting systems.

Keywords

Marine heatwave / Sea surface temperature / Machine learning / Deep learning

References

[1] AtlasR, HoffmanRN, ArdizzoneJ, LeidnerSM, JusemJC, SmithDK, et al.. A cross-calibrated, multiplatform ocean surface wind velocity product for meteorological and oceanographic applications. Bull Amer Meteorol Soc, 2011, 92(2): 157-174.

[2] BakerAC, GlynnPW, RieglB. Climate change and coral reef bleaching: an ecological assessment of long-term impacts, recovery trends and future outlook. Estuar Coast Shelf Sci, 2008, 80(4): 435-471.

[3] BanzonV, SmithTM, SteeleM, HuangBY, ZhangHM. Improved estimation of proxy sea surface temperature in the Arctic. J Atmos Ocean Technol, 2020, 37(2): 341-349.

[4] BoninoG, GalimbertiG, MasinaS, McAdamR, ClementiE. Machine learning methods to predict sea surface temperature and marine heatwave occurrence: a case study of the Mediterranean Sea. Ocean Sci, 2024, 20(2): 417-432.

[5] CartonJA, GieseBS. A reanalysis of ocean climate using simple ocean data assimilation (SODA). Mon Weather Rev, 2008, 136(8): 2999-3017.

[6] CMEMS (2017) Sea surface height anomalies (SSHA) data. https://doi.org/10.48670/moi-00149

[7] Coles S (2001) Classical extreme value theory and models. In: An Introduction to statistical modeling of extreme values. Springer, London, pp 45–73. https://doi.org/10.1007/978-1-4471-3675-0_3

[8] ECMWF (2022) ECMWF sub-seasonal to seasonal (S2S) prediction project dataset-2022 version. https://www.ecmwf.int/en/research/projects/s2s. Accessed 1 Mar 2025

[9] FordyceAJ, AinsworthTA, HeronSF, LeggatW. Marine heatwave hotspots in coral reef environments: physical drivers, ecophysiological outcomes, and impact upon structural complexity. Front Mar Sci, 2019, 6498.

[10] FrölicherTL, LaufkötterC. Emerging risks from marine heat waves. Nature Commun, 2018, 91650.

[11] GarrabouJ, Gómez-GrasD, MedranoA, CerranoC, PontiM, SchlegelR, et al.. Marine heatwaves drive recurrent mass mortalities in the Mediterranean Sea. Glob Change Biol, 2022, 28(19): 5708-5725.

[12] GiamalakiK, BeaulieuC, ProchaskaJX. Assessing predictability of marine heatwaves with random forests. Geophys Res Lett, 2022, 4923. e2022GL099069

[13] HeJJ, YinSL, ChenXY, YinB, HuangXQ. An informer-based prediction model for extensive spatiotemporal prediction of sea surface temperature and marine heatwave in Bohai Sea. J Mar Syst, 2025, 247. 104037

[14] HeQ, ZhuZH, ZhaoDF, SongW, HuangDM. An interpretable deep learning approach for detecting marine heatwaves patterns. Appl Sci, 2024, 142601.

[15] HerringSC, ChristidisN, HoellA, StottPA. Explaining extreme events of 2020 from a climate perspective. Bull Amer Meteorol Soc, 2022, 103(3): S1-S129.

[16] Hersbach H, Bell B, Berrisford P, Biavati G, Horányi A, Muñoz Sabater J et al (2023) ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). https://doi.org/10.24381/cds.adbb2d47

[17] HersbachH, BellB, BerrisfordP, HiraharaS, HorányiA, Muñoz-SabaterJ, et al.. The ERA5 global reanalysis. Q J R Meteorol Soc, 2020, 146(730): 1999-2049.

[18] Hersbach H, de Rosnay P, Bell B, Schepers D, Simmons A, Soci C et al (2018) Operational global reanalysis: progress, future directions and synergies with NWP. In: ERA Report Series. pp 1–65. https://doi.org/10.21957/tkic6g3wm

[19] HobdayAJ, AlexanderLV, PerkinsSE, SmaleDA, StraubSC, OliverEC, et al.. A hierarchical approach to defining marine heatwaves. Prog Oceanogr, 2016, 141: 227-238.

[20] HobdayAJ, OliverEC, GuptaAS, BenthuysenJA, BurrowsMT, DonatMG, et al.. Categorizing and naming marine heatwaves. Oceanography, 2018, 31(2): 162-173.

[21] HochreiterS, SchmidhuberJ. Long short-term memory. Neural Comput, 1997, 9(8): 1735-1780.

[22] HughesTP, KerryJT, Álvarez-NoriegaM, Álvarez-RomeroJG, AndersonKD, BairdAH, et al.. Global warming and recurrent mass bleaching of corals. Nature, 2017, 543(7645): 373-377.

[23] JacoxMG, AlexanderMA, BogradSJ, ScottJD. Thermal displacement by marine heatwaves. Nature, 2020, 584(7819): 82-86.

[24] Kalnay E, Kanamitsu M, Kistler R, Collins W, Deaven D, Gandin L et al (2018) The NCEP/NCAR 40-year reanalysis project. In: Renewable energy. Routledge, pp Vol1_146–Vol1_194

[25] Leadbetter MR, Lindgren G, Rootzen H (2012) Extremes and related properties of random sequences and processes. Springer Science & Business Media

[26] Lellouche JM, Le Galloudec O, Greiner E, Garric G, Regnier C, Drevillon M et al (2018) The copernicus marine environment monitoring service global ocean 1/12$^{\circ }$ physical reanalysis GLORYS12V1: description and quality assessment. In: EGU General Assembly Conference Abstracts. Copernicus Meetings, p 19806. https://doi.org/10.48670/moi-00021

[27] LiXF, LiuB, ZhengG, RenYB, ZhangSS, LiuYJ, et al.. Deep-learning-based information mining from ocean remote-sensing imagery. Natl Sci Rev, 2020, 7(10): 1584-1605.

[28] LiuG, HeronSF, EakinCM, Muller-KargerFE, Vega-RodriguezM, GuildLS, et al.. Reef-scale thermal stress monitoring of coral ecosystems: new 5-km global products from NOAA coral reef watch. Remote Sens, 2014, 6(11): 11579-11606.

[29] MerchantCJ, EmburyO, BulginCE, BlockT, CorlettGK, FiedlerE, et al.. Satellite-based time-series of sea-surface temperature since 1981 for climate applications. Sci Data, 2019, 61223.

[30] MiCR, YiA, XueJY, DongCM, ShanHX. Forecasting summer marine heatwaves in the South China Sea using explainable machine learning models. Deep Sea Res Part I-Oceanogr Res Pap, 2025, 218. 104457

[31] Ning D, Vetrova V, Delaux S, Tappenden R, Bryan KR, Koh YS (2025) A study on monthly marine heatwave forecasts in New Zealand: an investigation of imbalanced regression loss functions with neural network models. Preprint at arXiv:2502.13495

[32] Ning D, Vetrova V, Koh YS, Bryan KR (2024) Advancing marine heatwave forecasts: an integrated deep learning approach. Preprint at arXiv:2412.04475

[33] NohKM, LimHG, KugJS. Global chlorophyll responses to marine heatwaves in satellite ocean color. Environ Res Lett, 2022, 176. 064034

[34] OliverECJ, BenthuysenJA, DarmarakiS, DonatMG, HobdayAJ, HolbrookNJ, et al.. Marine heatwaves. Annu Rev Mar Sci, 2021, 13(1): 313-342.

[35] OliverECJ, WotherspoonSJ, ChamberlainMA, HolbrookNJ. Projected Tasman Sea extremes in sea surface temperature through the twenty-first century. J Clim, 2014, 27(5): 1980-1998.

[36] OliverECJ, WotherspoonSJ, HolbrookNJ. Estimating extremes from global ocean and climate models: a bayesian hierarchical model approach. Prog Oceanogr, 2014, 122: 77-91.

[37] ParasyrisA, MethenitiV, KampanisN, DarmarakiS. Marine heatwaves in the Mediterranean Sea: a convolutional neural network study for extreme event prediction. Ocean Sci, 2025, 21(3): 897-912.

[38] PerkinsSE, AlexanderLV. On the measurement of heat waves. J Clim, 2013, 26(13): 4500-4517.

[39] ReichsteinM, Camps-VallsG, StevensB, JungM, DenzlerJ, CarvalhaisN, et al.. Deep learning and process understanding for data-driven earth system science. Nature, 2019, 566(7743): 195-204.

[40] Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. In: Navab N et al (eds) Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015. Lecture notes in computer science, vol 9351. Springer, Cham, pp 234–241. https://doi.org/10.1007/978-3-319-24574-4_28

[41] RutiPM, SomotS, GiorgiF, DuboisC, FlaounasE, ObermannA, et al.. Med-CORDEX initiative for Mediterranean climate studies. Bull Amer Meteorol Soc, 2016, 97(7): 1187-1208.

[42] Shi XJ, Chen ZR, Wang H, Yeung DY, Wong WK, Woo WC (2015) Convolutional LSTM network: a machine learning approach for precipitation nowcasting. In: 29th Annual Conference on Neural Information Processing Systems (NIPS). NIPS, pp 1–15

[43] ShuR, WuH, GaoY, XuFH, GouRJ, XiongW, et al.. Advanced forecasts of global extreme marine heatwaves through a physics-guided data-driven approach. Environ Res Lett, 2025, 204. 044030

[44] SmaleDA, WernbergT, OliverECJ, ThomsenM, HarveyBP, StraubSC, et al.. Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat Clim Chang, 2019, 9(4): 306-312.

[45] SmithKE, BurrowsMT, HobdayAJ, KingNG, MoorePJ, Sen GuptaA, et al.. Biological impacts of marine heatwaves. Annu Rev Mar Sci, 2023, 15(1): 119-145.

[46] Smith KE, Burrows MT, Hobday AJ, Sen Gupta A, Moore PJ, Thomsen M et al (2021) Socioeconomic impacts of marine heatwaves: global issues and opportunities. Science 374(6566):eabj3593. https://doi.org/10.1126/science.abj3593

[47] SunD, JingZ, LiuHL. Deep learning improves sub-seasonal marine heatwave forecast. Environ Res Lett, 2024, 196. 064035

[48] SunWJ, ZhouSY, YangJS, GaoXQ, JiJL, DongCM. Artificial intelligence forecasting of marine heatwaves in the South China Sea using a combined U-Net and ConvLSTM system. Remote Sens, 2023, 15164068.

[49] TaylorJ, FengM. A deep learning model for forecasting global monthly mean sea surface temperature anomalies. Front Clim, 2022, 4. 932932

[50] Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN et al (2017) Attention is all you need. In: 31st Annual Conference on Neural Information Processing Systems (NIPS). NIPS, pp 1–15

[51] VitartF, ArdilouzeC, BonetA, BrookshawA, ChenM, CodoreanC, et al.. The subseasonal to seasonal (S2S) prediction project database. Bull Amer Meteorol Soc, 2017, 98(1): 163-173.

[52] WeiL, GuanL. Seven-day sea surface temperature prediction using a 3DConv-LSTM model. Front Mar Sci, 2022, 9. 905848

[53] WernbergT, BennettS, BabcockRC, de BettigniesT, CureK, DepczynskiM, et al.. Climate-driven regime shift of a temperate marine ecosystem. Science, 2016, 353(6295): 169-172.

[54] XieBW, QiJF, YangSG, SunGM, FengZK, YinBS, et al.. Sea surface temperature and marine heat wave predictions in the South China Sea: a 3D U-net deep learning model integrating multi-source data. Atmosphere, 2024, 15186.

[55] Zhou HY, Zhang SH, Peng JQ, Zhang S, Li JX, Xiong H et al (2021) Informer: beyond efficient transformer for long sequence time-series forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence. AAAI, pp 11106–11115. https://doi.org/10.1609/aaai.v35i12.17325

RIGHTS & PERMISSIONS

The Author(s)