基于Stacking集成学习的日尺度海洋三维温度场重构-海洋预报

基于Stacking集成学习的日尺度海洋三维温度场重构

单位：1. 青岛科技大学, 山东青岛 266061;
2. 中国科学院海洋研究所, 山东青岛 266071;
3. 中国空间技术研究院, 北京 100086

分类号：P731.11

出版年·卷·期（页码）：2026··第二期（1-10）

摘要：

提出了一种基于Stacking集成学习的日尺度海洋三维温度场重构方法，旨在通过融合多源海面观测数据与历史温盐观测数据，提升温度场重构的精度与鲁棒性。本研究采用随机森林、线性回归、决策树、K近邻、多层感知器和梯度提升树等多种基模型，结合Optuna超参数优化技术，构建了Stacking集成学习模型。结果表明：Stacking模型在0～1 000 m深度范围内的深度平均均方根误差（RMSE）为0.840℃，显著优于单一基模型（最优基模型平均RMSE为0.992℃）。通过与黑潮延伸体区域浮标观测站和GLORYS再分析数据的对比，验证了重构温度场在时间序列平滑性和温度变化捕捉能力上的优越性。

In this paper, we propose a daily-scale ocean 3D temperature field reconstruction method based on Stacking integrated learning, aiming to improve the accuracy and robustness of the temperature field reconstruction by integrating multi-source sea surface observation data and historical temperature and salt observation data. The study adopts various base models such as random forest, linear regression, decision tree, K-nearest neighbor, multilayer perceptron and gradient boosting tree, and combines them with Optuna's hyperparameter optimization technique to construct the Stacking integrated learning model. The experimental results show that the average Root Mean Square Error(RMSE) of the Stacking model is 0.840 ℃ in the depth range of 0 to 1 000 m, which is significantly better than that of the single base model(the average RMSE of the optimal base model is 0.992 ℃). The superiority of the reconstructed temperature field in terms of time-series smoothing and the ability to capture temperature variations is verified by comparing with the Kuroshio Extension Observatory(KEO) and GLORYS reanalysis data.

参考文献：

[1] PENG Q H, XIE S P, WANG D X, et al. Surface warming–induced global acceleration of upper ocean currents[J]. Science Advances, 2022, 8(16):eabj8394.
[2] CHEN X Y, TUNG K K. Global surface warming enhanced by weak Atlantic overturning circulation[J]. Nature, 2018, 559(7714):387-391.
[3] MEEHL G A, ARBLASTER J M, FASULLO J T, et al. Modelbased evidence of deep-ocean heat uptake during surfacetemperature hiatus periods[J]. Nature Climate Change, 2011, 1(7):360-364.
[4] XIANG L, XU Y S, SUN H W, et al. Reconstruction of interior velocity in the southern pacific ocean using satellite and Argo data[J]. IEEE Geoscience and Remote Sensing Letters, 2025, 22:1500405.
[5] YUE W H, XU Y S, XIANG L, et al. Prediction of 3-D ocean temperature based on self-attention and predictive RNN[J]. IEEE Geoscience and Remote Sensing Letters, 2024, 21:1501305.
[6] SU H, WU X B, YAN X H, et al. Estimation of subsurface temperature anomaly in the Indian Ocean during recent global surface warming hiatus from satellite measurements:a support vector machine approach[J]. Remote Sensing of Environment,2015, 160:63-71.
[7] 孙春健,张晓爽,张寅权,等.卫星遥感重构海洋次表层研究进展[J]. 海洋信息, 2018, 33(4):21-28.SUN C J, ZHANG X S, ZHANG Y Q, et al. Progress in reconstruction of ocean subsurface by satellite remote sensing data[J]. Marine Information, 2018, 33(4):21-28.
[8] LAPEYRE G, KLEIN P. Dynamics of the upper oceanic layers in terms of surface quasigeostrophy theory[J]. Journal of Physical Oceanography, 2006, 36(2):165-176.
[9] 王喜冬,韩桂军,李威,等.利用卫星观测海面信息反演三维温度场[J]. 热带海洋学报, 2011, 30(6):10-17.WANG X D, HAN G J, LI W, et al. Reconstruction of ocean temperature profile using satellite observations[J]. Journal of Tropical Oceanography, 2011, 30(6):10-17.
[10] 谢旭丹,王静,储小青,等.南海中尺度涡温盐异常三维结构[J]. 海洋学报, 2018, 40(4):1-14.XIE X D, WANG J, CHU X Q, et al. Three-dimensional thermohaline anomaly structures of mesoscale eddies in the South China Sea[J]. Haiyang Xuebao, 2018, 40(4):1-14.
[11] GAVART M, DE MEY P. Isopycnal EOFs in the Azores current region:a statistical tool for dynamical analysis and data assimilation[J]. Journal of Physical Oceanography, 1997, 27(10):2146-2157.
[12] PASCUAL A, GOMIS D. Use of surface data to estimate geostrophic transport[J]. Journal of Atmospheric and Oceanic Technology, 2003, 20(6):912-926.
[13] SUN C, WATTS D R. A circumpolar gravest empirical mode for the Southern Ocean hydrography[J]. Journal of Geophysical Research:Oceans, 2001, 106(C2):2833-2855.
[14] PARK J H, WATTS D R, TRACEY K L, et al. A multi-index GEM technique and its application to the southwestern Japan/East Sea[J]. Journal of Atmospheric and Oceanic Technology, 2005, 22(8):1282-1293.
[15] LU W F, SU H, YANG X, et al. Subsurface temperature estimation from remote sensing data using a clustering-neural network method[J]. Remote Sensing of Environment, 2019, 229:213-222.
[16] SU H, HUANG L J, LI W E, et al. Retrieving ocean subsurface temperature using a satellite-based geographically weighted regression model[J]. Journal of Geophysical Research:Oceans,2018, 123(8):5180-5193.
[17] 黎文娥,苏华,汪小钦,等.多源卫星观测的全球海洋次表层温度异常信息提取[J]. 遥感学报, 2017, 21(6):881-891.LI W E, SU H, WANG X Q, et al. Estimation of global subsurface temperature anomaly based on multisource satellite Observations[J]. Journal of Remote Sensing, 2017, 21(6):881-891.
[18] SU H, ZHANG F Y, TENG J C, et al. Reconstructing highresolution subsurface temperature of the global ocean using deep forest with combined remote sensing and in situ observations[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2024,218:389-404.
[19] PAVLYSHENKO B. Using stacking approaches for machine learning models[C] //2018 IEEE Second International Conference on Data Stream Mining&Processing(DSMP). Lviv, Ukraine:IEEE, 2018:255-258.
[20] SHAO W Z, HU Y Y, MIGLIACCIO M, et al. Machine learningbased algorithm for SAR wave parameters retrieval during a tropical cyclone[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17:15166-15177.
[21] LUO Y P, LIU Y, HUANG C Y, et al. An ensemble machine learning approach for sea ice monitoring using CFOSAT/SCAT data[J]. Remote Sensing, 2024, 16(17):3148.
[22] ZHU X T, GUO H W, HUANG J J, et al. An ensemble machine learning model for water quality estimation in coastal area based on remote sensing imagery[J]. Journal of Environmental Management, 2022, 323:116187.
[23] EMERY W J, BALDWIN D J, SCHLüSSEL P, et al. Accuracy of in situ sea surface temperatures used to calibrate infrared satellite measurements[J]. Journal of Geophysical Research:Oceans,2001, 106(C2):2387-2405.
[24] LV X Y, BAO S L, WANG H Z, et al. Quality assessment and comparison of two merged sea surface height products:ALT MUL and AVISO[J]. Acta Oceanologica Sinica, 2025, 44(2):114-124.
[25] RISER S C, FREELAND H J, ROEMMICH D, et al. Fifteen years of ocean observations with the global Argo array[J]. Nature Climate Change, 2016, 6(2):145-153.
[26] MIENYE I D, SUN Y X. A survey of ensemble learning:concepts, algorithms, applications, and prospects[J]. IEEE Access,2022, 10:99129-99149.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】