海量数据驱动下的高分辨率海洋数值模式发展与展望
|
摘要
海洋数值模式是定量描述海洋物理现象及其变化的数值模型,也是海洋与气候研究、预测的核心工具。随着海洋观测的不断投入与积累、对海洋认识的不断深入,特别是在高性能计算技术的支撑下,海洋数值模式已有了长足进步,正朝着高分辨率和多物理过程的方向发展。随着分辨率的提高、物理过程的细化,海洋数值模式的发展面临着多个方面的挑战。当前,海洋数据数量和种类不断增多,同时超级计算机、高性能计算和深度学习等技术的快速发展,为海洋数值模式的突破提供了机遇与挑战。本研究回顾了海洋数值模式的发展现状,梳理和分析了其发展中遇到的大规模高效并行计算和参数优化这两个关键问题,探讨和展望了当前海量数据驱动下海洋数值模式的发展趋势。提出计算负载均衡、计算与I/O重叠的并行流水线设计以及降低全局交换的算法改进是当前突破高分辨率海洋模式大规模高效并行效率的关键。从海洋科学、高性能计算以及深度学习深度交叉融合的角度,提出了实现海洋科学与深度学习相结合的6个途径,在此基础上,探讨了基于深度学习的参数化优化可能实现的途径。
The ocean model is the key tool for ocean and climate research and prediction, which is a numerical model for the quantitative description of marine phenomena based on the physical laws. With the continuous marine investment, accumulation of ocean observation data, deeply understanding of the ocean process, and high-performance computing technology development, ocean numerical models have made great progress. Now, the main stream of ocean model development focuses on higher resolution and more accurate parameterization of unresolved physical processes. With the finer resolution and the more physical process, the development of ocean numerical models faces several challenges. The increasing ocean data, rapid development of high-performance computing and neural network depth learning technologies provide an opportunity for the breakthrough of ocean numerical models. This paper mainly reviewed the history and research status of the ocean numerical models, clarified and analyzed the two bottlenecks, performance of large-scale parallel computing and physical parameterization scheme, of ocean numerical models' development and applications. Then, from the perspective of marine science, high-performance computing and deep learning integration, we proposed the six approaches of deep learning and ocean science convegence, the future development and trend of ocean numerical models driven by the mass data.
The ocean model is the key tool for ocean and climate research and prediction, which is a numerical model for the quantitative description of marine phenomena based on the physical laws. With the continuous marine investment, accumulation of ocean observation data, deeply understanding of the ocean process, and high-performance computing technology development, ocean numerical models have made great progress. Now, the main stream of ocean model development focuses on higher resolution and more accurate parameterization of unresolved physical processes. With the finer resolution and the more physical process, the development of ocean numerical models faces several challenges. The increasing ocean data, rapid development of high-performance computing and neural network depth learning technologies provide an opportunity for the breakthrough of ocean numerical models. This paper mainly reviewed the history and research status of the ocean numerical models, clarified and analyzed the two bottlenecks, performance of large-scale parallel computing and physical parameterization scheme, of ocean numerical models' development and applications. Then, from the perspective of marine science, high-performance computing and deep learning integration, we proposed the six approaches of deep learning and ocean science convegence, the future development and trend of ocean numerical models driven by the mass data.
引文
[1] ZHANG X H,LIANG X Z.A numerical world ocean general circulation model[J].Advances in Atmospheric Sciences,1989,6(1):43-61.
[2] ZHAO Y L,ZHANG M,SI G Y.Review and perspectives of study on the thermohaline circulation[J].Journal of PLA University of Science and Technology,2006,7(3):281-290.赵艳玲,张铭,司广宇.海洋环流模式研究回顾与展望.解放军理工大学学报(自然科学版),2006,7(3):281-290.
[3] ZHAO W,LEI X Y,CHEN D X,et al.Porting and application of global eddy-resolving parallel ocean model POP to SW supercomputer[J].Computer Applications and Software,2014,31(5):42-45.赵伟,雷晓燕,陈德训,等.全球涡分辨率并行海洋模式POP在神威蓝光上的移植和应用[J].计算机应用与软件,2014,31(5):42-45.
[4] QIAO F L.Ocean models system and the surface wave-circulation coupled theory[J].Frontier Science,2007,3:81-86.乔方利.海洋动力系统数值模式体系及海浪-环流耦合理论[J].前沿科学,2007,3:81-86.
[5] SHU Q,QIAO F L,SONG Z Y,et al.A comparison of two global ocean-ice coupled models with different horizontal resolutions[J].Acta Oceanoglogica Sinica,2013,32(8):1-11.
[6] DONGARRA J,BECKMAN P,MOORE T.The International exascale software project roadmap[J].International Journal of High Performance Computing Applications,2011,25(1):3-60.
[7] YU Z L.Review of progress on artificial intelligence[J].Journal of Nanjing University of Information Science and Technology (Natural Science Edition),2017,9(3):297-304.俞祝良.人工智能技术发展概述[J].南京信息工程大学学报(自然科学版),2017,9(3):297-304.
[8] ZHANG R,LI W P,MO T.Review of deep learning[J].Information and Control,2018,47(4):385-397,410.张荣,李伟平,莫同.深度学习研究综述[J].信息与控制,2018,47(4):385-397,410.
[9] SARKISYAN A S.The role of the pure drift advection of density in the dynamics of the wind-driven currents of a baroclinic ocean[J].Izvestiya Akademii Nauk SSSR Seriya Geologicheskaya,1961,9:1396-1407.
[10] SARKISYAN A S.On the dynamics of occurrence of wind-driven currents in a baroclinic ocean[J].Okeanologiya,1962,2(3):393-409.
[11] GORMATYUK Y K,SARKISYAN A S.The results of calculations of currents in the north Atlantic according to the four-level model[J].Izvestiva Akademii Nauk SSSR,Fizika Atmosfery i Okeana,1965,1(3):313-326.
[12] HOLLAND W R,HIRSCHMAN A D.A numerical calculation of the circulation in the north Atlantic ocean[J].Journal of Physical Oceanography,1972,2:336-354.
[13] DEMIN Y L,SARKISYAN A S.Calculation of equatorial currents[J].Journal of Marine Research,1977,35:339-356.
[14] GILL A E,BRYAN K.Effects of geometry on the circulation of a three-dimensional southern-hemisphere ocean model[J].Deep Sea Research,1971,18:685-721.
[15] BRYAN K,COX M.A numerical investigation of the oceanic general circulation[J].Tellus,1967,19(1):163-174.
[16] BRYAN K.A numerical method for the study of the circulation of the world ocean[J].Journal of Computational Physics,1969,4(3):347-376.
[17] GRIFFIES S M,STOUFFER R J,ADCROFT A J,et al.A historical introduction to MOM[EB/OL].[2018-03-12]https://www.gfdl.noaa.gov/wp-content/uploads/files/model_development/ocean/mom_history_v15.09.05.pdf.
[18] SMITH R D,MALONE R C,DUKOWICZ J K.Parallel ocean general circulation modeling[J].Physica D:Nonlinear Phenomena,1992,60(1-4):38-61.
[19] BLECK R.An oceanic general circulation model framed in hybrid isopycnic-cartesian coordinates[J].Ocean Modelling,2002,4(1):55-88
[20] ADCROFT A J,HILL C N,MARSHALL J C.A new treatment of the Coriolis terms in C-Grid Models at both high and low Resolutions[J].Monthly Weather Review,1999,127(127):1928.
[21] MADEC G.THE NEMO TEAM.Nemo Ocean Engine v3.4[EB/OL].[2018-03-12] http://www.nemo-ocean.eu/27.
[22] LIU H L,LIN P F,YU Y Q,et al.The baseline evaluation of LASG/IAP climate system ocean model (LICOM) version 2[J].Acta Meteorologica Sinica,2012,26(3):318-329.
[23] BLUMBERG A F,MELLOR G L.A description of a three-dimensional coastal ocean circulation model[C]//HEAPS N S.Three-dimensional coastal ocean models,Coastal and Estuarine Series volume 4.Washington D C,USA:American Geophysical Union (AGU),2013:1-16.
[24] SHCHEPETKIN A F,MCWILLIAMS J C.The regional oceanic modeling system (ROMS):a split-explicit,free-surface,topography-following-coordinate oceanic model[J].Ocean Modelling,2005,9(4):347-404.
[25] CHEN C S,LIU H,BEARDSLEY R C.An unstructured grid,finite-volume,three-dimensional,primitive equations ocean model:application to coastal ocean and estuaries[J].Journal of Atmospheric & Oceanic Technology,2003,20(20):159-186.
[26] IPCC.Climate Change 2001:The scientific basis[M].Cambridge,USA:Cambridge University Press,2001:476-479.
[27] SMEDSTAD O,HURLBURT H,METZGER E,et al.An operational real-time eddy-resolving 1/16°global ocean nowcast/forecast system[J].Journal of Marine Systems,2003,40-41(3):341-361.
[28] MASUMOTO Y.A fifty-year eddy-resolving simulation of the world ocean:preliminary outcomes of OFES (OGCM for the Earth Simulator)[J].Journal of Earth Simulator,2004(1):35-56.
[29] MALTRUD M,MCCLEAN J.An eddy resolving global 1/10° ocean simulation[J].Ocean Modelling,2005,8(1-2):31-54.
[30] YU Y Q,LIU H L,LIN P F.A quasi-global (1/10)° eddy-resolving ocean general circulation model and its preliminary results[J].Chinese Science Bulletin,2012,57(3):3908-3916.
[31] MELLOR G L,YAMADA T.Development of a turbulence closure model for geophysical fluid problems[J].Reviews of Geophysics,1982,20(4):851-875.
[32] PACANOWSKI R C,PHILANDER S G H.Parameterization of vertical mixing in numerical models of tropical oceans[J].Journal of Physical Oceanography,1981,11(11):1443-1451.
[33] LARGE W G,MCWILLIAMS J C,DONEY S C.Oceanic vertical mixing:a review and a model with a nonlocal boundary layer parameterization[J].Reviews of Geophysics,1994,32(4):363-403.
[34] SMAGORINSKY J.General circulation experiments with the primitive equations:I.the basic experiment[J].Monthly Weather Review,1963,91:99-164.
[35] GENT P R,MCWILLIAMS J C.Isopycnal mixing in ocean circulation models[J].Journal of Physical Oceanography,1990,20(20):150-160.
[36] QIAO F L,YUAN Y L,YANG Y Z,et al.Wave-induced mixing in the upper ocean:distribution and application to a global ocean circulation model[J].Geophysical Research Letters,2004,31(11):293-317.
[37] QIAO F L,YUAN Y L,EZER T,et al.A three-dimensional surface wave-ocean circulation coupled model and its initial testing[J].Ocean Dynamics,2010,60(5):1339-1355.
[38] WANG X W,PENG S Q,LIU Z Y,et al.Tidal mixing in the south China sea:an estimate based on the internal tide energetics[J].Journal of Physical Oceanography,2016,46(1):107-124.
[39] YANG X D,SONG Z Y,ZHOU S.et al.Code modernization optimization of MASNUM wave model[J].Advances in Marine Science,2017,35(4):473-482.杨晓丹,宋振亚,周姗,等.MASNUM海浪模式的代码现代化优化[J].海洋科学进展,2017,35(4):473-482.
[40] FU H H,LIAO J J,YANG J Z,et al.The Sunway TaihuLight supercomputer:system and applications[J].Science China:Information Sciences,2006,59(7):072001.1-072001.16.
[41] ZHAO W,SONG Z Y,QIAO F L,et al.High efficient parallel numerical surface wave model based on an irregular quasi-rectangular domain decomposition scheme[J].Science China:Earth Science,2014,57(8):1869-1878.
[42] QIAO F L,ZHAO W,YIN X Q,et al.A highly effective global surface wave numerical simulation with ultra-high resolution[C]// SC16:International Conference for High Performance Computing,Networking,Storage and Analysis.Piscataway N J,USA:IEEE Press,2016.
[43] HU Y,HUANG X M,BAKER A H,et al.Improving the scalability of the ocean barotropic solver in the community earth system model[C]//SC15:International Conference for High Performance Computing,Networking,Storage and Analysis.New York,USA:IEEE Press,2015.
[44] KOHONEN T.A simple paradigm for the self-organized formation of structured feature maps[M]//Competition and Cooperation in Neural Nets.Springer Berlin Heidelberg,1982.
[45] HOPFIELD J J.Neural networks and physical systems with emergent collective computational abilities[J].Proceedings of the National Academy of Sciences,1982,79(8):2554-2558.
[46] LIPPMANN R,BECKMAN P.Adaptive neural net preprocessing for signal detection in non-Gaussian noise [C]//TOURETZKY D S.Advances in Neural Information Processing Systems 1.San Francisco C A,USA:Morgan Kaufmann Publishers Inc,1989:124-132.
[47] LECUN Y,BENGIO Y,Hinton G.Deep learning[J].Nature,2015,521(7553):436-444.
[48] GAO S,ZHAO P,PAN B,et al.A nowcasting model for the prediction of typhoon tracks based on a long short term memory neural network[J].Acta Oceanologica Sinica,2018,37(5):8-12.
[49] DEO M C,NAIDU C S.Real time wave forecasting using neural networks[J].Ocean Engineering,1998,26(3):191-203
[50] MANDAL S,PRABAHARAN N.Ocean wave forecasting using recurrent neural networks[J].Ocean Engineering,2006,33(10):1401-1410.
[51] ZHANG W,LEUNG Y,CHAN J C.The analysis of tropical cyclone tracks in the western north Pacific through data mining.Part I:tropical cyclone recurvature[J].Journal of Applied Meteorology and Climatology,2013,52:1394-1416.
[52] PARK M,KIM M,LEE M,et al.Detection of tropical cyclone genesis via quantitative satellite ocean surface wind pattern and intensity analyses using decision trees[J].Remote Sensing of Environment,2016,183:205-214.
[53] ZHANG T,XIE F,XUE W,et al.Quantification and optimization of parameter uncertainty in the grid-point atmospheric model GAMIL2[J].Chinese Journal of Geophysics,2016,59(2):465-475.张涛,谢丰,薛巍,等.格点大气环流模式GAMIL2参数不确定性的量化分析与优化[J].地球物理学报,2016,59(2):465-475.
[54] JIANG G-Q,XU J,WEI J.A deep learning algorithm of neural network for the parameterization of tyhpoon-ocean feedback in typhoon forecast models.Geophysical Research Letters,2018,45(8):3706-3716.
[55] BOLTON T,ZANNA L.Applications of deep learning to ocean data inference and subgrid parameterization[J].Journal of Advances in Modeling Earth Systems,2018,11.DOI:10.1029/2018MS001472.
[56] MEN X L,JIAO R L,WANG D,et al.A temperature correction method for multi-model ensemble forecast in North China based on machine learning[J].Climatic and Environmental Research,2019,24(1):116-124.门晓磊,焦瑞莉,王鼎,等.基于机器学习的华北气温多模式集合预报的订正方法[J].气候与环境研究,2019,24(1):116-124.
[57] ZHU J C,HU S,ARCUCCI R,et al.Model error correction in data assimilation by integrating neural networks[J].Big Data Ming and Analytics,2019,2(5):83-91.
[58] BRENOWITZ N D,BRETHERTON C S.Prognostic validation of a neural network unified physics parameterization[J].Geophysical Research Letter,45(12):6289-6298.
[59] WILLIS M J,VON STOSCH M.Simultaneous parameter identification and discrimination of the nonparametric structure of hybrid semi-parametric models[J].Computers & Chemical Engineering,104:366-376.
[60] JAMES S C,ZHANG Y,O’DONNCHA F.A machine learning framework to forecast wave conditions[J].Coastal Engineering,2018,137:1-10.
[61] REICHSTEIN M,CAMPS-VALLS G,STEVENS B,et al.Deep learning and process understanding for data-driven earth system science[J].Nature,2019,566:195-204.
[2] ZHAO Y L,ZHANG M,SI G Y.Review and perspectives of study on the thermohaline circulation[J].Journal of PLA University of Science and Technology,2006,7(3):281-290.赵艳玲,张铭,司广宇.海洋环流模式研究回顾与展望.解放军理工大学学报(自然科学版),2006,7(3):281-290.
[3] ZHAO W,LEI X Y,CHEN D X,et al.Porting and application of global eddy-resolving parallel ocean model POP to SW supercomputer[J].Computer Applications and Software,2014,31(5):42-45.赵伟,雷晓燕,陈德训,等.全球涡分辨率并行海洋模式POP在神威蓝光上的移植和应用[J].计算机应用与软件,2014,31(5):42-45.
[4] QIAO F L.Ocean models system and the surface wave-circulation coupled theory[J].Frontier Science,2007,3:81-86.乔方利.海洋动力系统数值模式体系及海浪-环流耦合理论[J].前沿科学,2007,3:81-86.
[5] SHU Q,QIAO F L,SONG Z Y,et al.A comparison of two global ocean-ice coupled models with different horizontal resolutions[J].Acta Oceanoglogica Sinica,2013,32(8):1-11.
[6] DONGARRA J,BECKMAN P,MOORE T.The International exascale software project roadmap[J].International Journal of High Performance Computing Applications,2011,25(1):3-60.
[7] YU Z L.Review of progress on artificial intelligence[J].Journal of Nanjing University of Information Science and Technology (Natural Science Edition),2017,9(3):297-304.俞祝良.人工智能技术发展概述[J].南京信息工程大学学报(自然科学版),2017,9(3):297-304.
[8] ZHANG R,LI W P,MO T.Review of deep learning[J].Information and Control,2018,47(4):385-397,410.张荣,李伟平,莫同.深度学习研究综述[J].信息与控制,2018,47(4):385-397,410.
[9] SARKISYAN A S.The role of the pure drift advection of density in the dynamics of the wind-driven currents of a baroclinic ocean[J].Izvestiya Akademii Nauk SSSR Seriya Geologicheskaya,1961,9:1396-1407.
[10] SARKISYAN A S.On the dynamics of occurrence of wind-driven currents in a baroclinic ocean[J].Okeanologiya,1962,2(3):393-409.
[11] GORMATYUK Y K,SARKISYAN A S.The results of calculations of currents in the north Atlantic according to the four-level model[J].Izvestiva Akademii Nauk SSSR,Fizika Atmosfery i Okeana,1965,1(3):313-326.
[12] HOLLAND W R,HIRSCHMAN A D.A numerical calculation of the circulation in the north Atlantic ocean[J].Journal of Physical Oceanography,1972,2:336-354.
[13] DEMIN Y L,SARKISYAN A S.Calculation of equatorial currents[J].Journal of Marine Research,1977,35:339-356.
[14] GILL A E,BRYAN K.Effects of geometry on the circulation of a three-dimensional southern-hemisphere ocean model[J].Deep Sea Research,1971,18:685-721.
[15] BRYAN K,COX M.A numerical investigation of the oceanic general circulation[J].Tellus,1967,19(1):163-174.
[16] BRYAN K.A numerical method for the study of the circulation of the world ocean[J].Journal of Computational Physics,1969,4(3):347-376.
[17] GRIFFIES S M,STOUFFER R J,ADCROFT A J,et al.A historical introduction to MOM[EB/OL].[2018-03-12]https://www.gfdl.noaa.gov/wp-content/uploads/files/model_development/ocean/mom_history_v15.09.05.pdf.
[18] SMITH R D,MALONE R C,DUKOWICZ J K.Parallel ocean general circulation modeling[J].Physica D:Nonlinear Phenomena,1992,60(1-4):38-61.
[19] BLECK R.An oceanic general circulation model framed in hybrid isopycnic-cartesian coordinates[J].Ocean Modelling,2002,4(1):55-88
[20] ADCROFT A J,HILL C N,MARSHALL J C.A new treatment of the Coriolis terms in C-Grid Models at both high and low Resolutions[J].Monthly Weather Review,1999,127(127):1928.
[21] MADEC G.THE NEMO TEAM.Nemo Ocean Engine v3.4[EB/OL].[2018-03-12] http://www.nemo-ocean.eu/27.
[22] LIU H L,LIN P F,YU Y Q,et al.The baseline evaluation of LASG/IAP climate system ocean model (LICOM) version 2[J].Acta Meteorologica Sinica,2012,26(3):318-329.
[23] BLUMBERG A F,MELLOR G L.A description of a three-dimensional coastal ocean circulation model[C]//HEAPS N S.Three-dimensional coastal ocean models,Coastal and Estuarine Series volume 4.Washington D C,USA:American Geophysical Union (AGU),2013:1-16.
[24] SHCHEPETKIN A F,MCWILLIAMS J C.The regional oceanic modeling system (ROMS):a split-explicit,free-surface,topography-following-coordinate oceanic model[J].Ocean Modelling,2005,9(4):347-404.
[25] CHEN C S,LIU H,BEARDSLEY R C.An unstructured grid,finite-volume,three-dimensional,primitive equations ocean model:application to coastal ocean and estuaries[J].Journal of Atmospheric & Oceanic Technology,2003,20(20):159-186.
[26] IPCC.Climate Change 2001:The scientific basis[M].Cambridge,USA:Cambridge University Press,2001:476-479.
[27] SMEDSTAD O,HURLBURT H,METZGER E,et al.An operational real-time eddy-resolving 1/16°global ocean nowcast/forecast system[J].Journal of Marine Systems,2003,40-41(3):341-361.
[28] MASUMOTO Y.A fifty-year eddy-resolving simulation of the world ocean:preliminary outcomes of OFES (OGCM for the Earth Simulator)[J].Journal of Earth Simulator,2004(1):35-56.
[29] MALTRUD M,MCCLEAN J.An eddy resolving global 1/10° ocean simulation[J].Ocean Modelling,2005,8(1-2):31-54.
[30] YU Y Q,LIU H L,LIN P F.A quasi-global (1/10)° eddy-resolving ocean general circulation model and its preliminary results[J].Chinese Science Bulletin,2012,57(3):3908-3916.
[31] MELLOR G L,YAMADA T.Development of a turbulence closure model for geophysical fluid problems[J].Reviews of Geophysics,1982,20(4):851-875.
[32] PACANOWSKI R C,PHILANDER S G H.Parameterization of vertical mixing in numerical models of tropical oceans[J].Journal of Physical Oceanography,1981,11(11):1443-1451.
[33] LARGE W G,MCWILLIAMS J C,DONEY S C.Oceanic vertical mixing:a review and a model with a nonlocal boundary layer parameterization[J].Reviews of Geophysics,1994,32(4):363-403.
[34] SMAGORINSKY J.General circulation experiments with the primitive equations:I.the basic experiment[J].Monthly Weather Review,1963,91:99-164.
[35] GENT P R,MCWILLIAMS J C.Isopycnal mixing in ocean circulation models[J].Journal of Physical Oceanography,1990,20(20):150-160.
[36] QIAO F L,YUAN Y L,YANG Y Z,et al.Wave-induced mixing in the upper ocean:distribution and application to a global ocean circulation model[J].Geophysical Research Letters,2004,31(11):293-317.
[37] QIAO F L,YUAN Y L,EZER T,et al.A three-dimensional surface wave-ocean circulation coupled model and its initial testing[J].Ocean Dynamics,2010,60(5):1339-1355.
[38] WANG X W,PENG S Q,LIU Z Y,et al.Tidal mixing in the south China sea:an estimate based on the internal tide energetics[J].Journal of Physical Oceanography,2016,46(1):107-124.
[39] YANG X D,SONG Z Y,ZHOU S.et al.Code modernization optimization of MASNUM wave model[J].Advances in Marine Science,2017,35(4):473-482.杨晓丹,宋振亚,周姗,等.MASNUM海浪模式的代码现代化优化[J].海洋科学进展,2017,35(4):473-482.
[40] FU H H,LIAO J J,YANG J Z,et al.The Sunway TaihuLight supercomputer:system and applications[J].Science China:Information Sciences,2006,59(7):072001.1-072001.16.
[41] ZHAO W,SONG Z Y,QIAO F L,et al.High efficient parallel numerical surface wave model based on an irregular quasi-rectangular domain decomposition scheme[J].Science China:Earth Science,2014,57(8):1869-1878.
[42] QIAO F L,ZHAO W,YIN X Q,et al.A highly effective global surface wave numerical simulation with ultra-high resolution[C]// SC16:International Conference for High Performance Computing,Networking,Storage and Analysis.Piscataway N J,USA:IEEE Press,2016.
[43] HU Y,HUANG X M,BAKER A H,et al.Improving the scalability of the ocean barotropic solver in the community earth system model[C]//SC15:International Conference for High Performance Computing,Networking,Storage and Analysis.New York,USA:IEEE Press,2015.
[44] KOHONEN T.A simple paradigm for the self-organized formation of structured feature maps[M]//Competition and Cooperation in Neural Nets.Springer Berlin Heidelberg,1982.
[45] HOPFIELD J J.Neural networks and physical systems with emergent collective computational abilities[J].Proceedings of the National Academy of Sciences,1982,79(8):2554-2558.
[46] LIPPMANN R,BECKMAN P.Adaptive neural net preprocessing for signal detection in non-Gaussian noise [C]//TOURETZKY D S.Advances in Neural Information Processing Systems 1.San Francisco C A,USA:Morgan Kaufmann Publishers Inc,1989:124-132.
[47] LECUN Y,BENGIO Y,Hinton G.Deep learning[J].Nature,2015,521(7553):436-444.
[48] GAO S,ZHAO P,PAN B,et al.A nowcasting model for the prediction of typhoon tracks based on a long short term memory neural network[J].Acta Oceanologica Sinica,2018,37(5):8-12.
[49] DEO M C,NAIDU C S.Real time wave forecasting using neural networks[J].Ocean Engineering,1998,26(3):191-203
[50] MANDAL S,PRABAHARAN N.Ocean wave forecasting using recurrent neural networks[J].Ocean Engineering,2006,33(10):1401-1410.
[51] ZHANG W,LEUNG Y,CHAN J C.The analysis of tropical cyclone tracks in the western north Pacific through data mining.Part I:tropical cyclone recurvature[J].Journal of Applied Meteorology and Climatology,2013,52:1394-1416.
[52] PARK M,KIM M,LEE M,et al.Detection of tropical cyclone genesis via quantitative satellite ocean surface wind pattern and intensity analyses using decision trees[J].Remote Sensing of Environment,2016,183:205-214.
[53] ZHANG T,XIE F,XUE W,et al.Quantification and optimization of parameter uncertainty in the grid-point atmospheric model GAMIL2[J].Chinese Journal of Geophysics,2016,59(2):465-475.张涛,谢丰,薛巍,等.格点大气环流模式GAMIL2参数不确定性的量化分析与优化[J].地球物理学报,2016,59(2):465-475.
[54] JIANG G-Q,XU J,WEI J.A deep learning algorithm of neural network for the parameterization of tyhpoon-ocean feedback in typhoon forecast models.Geophysical Research Letters,2018,45(8):3706-3716.
[55] BOLTON T,ZANNA L.Applications of deep learning to ocean data inference and subgrid parameterization[J].Journal of Advances in Modeling Earth Systems,2018,11.DOI:10.1029/2018MS001472.
[56] MEN X L,JIAO R L,WANG D,et al.A temperature correction method for multi-model ensemble forecast in North China based on machine learning[J].Climatic and Environmental Research,2019,24(1):116-124.门晓磊,焦瑞莉,王鼎,等.基于机器学习的华北气温多模式集合预报的订正方法[J].气候与环境研究,2019,24(1):116-124.
[57] ZHU J C,HU S,ARCUCCI R,et al.Model error correction in data assimilation by integrating neural networks[J].Big Data Ming and Analytics,2019,2(5):83-91.
[58] BRENOWITZ N D,BRETHERTON C S.Prognostic validation of a neural network unified physics parameterization[J].Geophysical Research Letter,45(12):6289-6298.
[59] WILLIS M J,VON STOSCH M.Simultaneous parameter identification and discrimination of the nonparametric structure of hybrid semi-parametric models[J].Computers & Chemical Engineering,104:366-376.
[60] JAMES S C,ZHANG Y,O’DONNCHA F.A machine learning framework to forecast wave conditions[J].Coastal Engineering,2018,137:1-10.
[61] REICHSTEIN M,CAMPS-VALLS G,STEVENS B,et al.Deep learning and process understanding for data-driven earth system science[J].Nature,2019,566:195-204.