Probabilistic Machine Learning Estimation of Ocean Mixed Layer Depth From Dense Satellite and Sparse In Situ Observations.
Probabilistic Machine Learning Estimation of Ocean Mixed Layer Depth From Dense Satellite and Sparse In Situ Observations.
复制标题
DOI：
10.1029/2021ms002474
复制
发表时间：
2021-12
影响因子：
6.8
通讯作者：
Whitt, Daniel B.
中科院分区：
文献类型：
--
作者：
Foster, Dallas;Gagne, David John, II;Whitt, Daniel B.
海洋混合层在广泛的时间尺度上，在上层海洋与大气之间的耦合中起着重要作用。因此，对海洋混合层变异性的估计对于大气 - 海洋预测和分析非常重要。现场阿尔戈剖面数据覆盖范围的不断扩大，使得对与季节气候偏差相关的混合层深度（MLD）变异性的分析越来越准确。然而，采样率不足以完全解析次季节（<90天）的MLD变异性。然而，许多基于多变量观测的分析都包含了隐含建模的次季节MLD变异性。一种分析方法是对现场数据进行最优插值，但通过利用回归或变分方法利用海面数据可以改进内部分析。在此，我们在两个区域（中纬度南印度洋和赤道东太平洋）的试点研究中展示了机器学习方法以及卫星海表温度、盐度和高度如何有助于MLD估计。我们构建了多种机器学习架构，以生成具有不确定性估计的每周1/2°网格MLD异常场（相对于月气候态）。我们测试了多种传统和概率机器学习技术，以比较准确性和概率校准。我们通过将其应用于海洋模型模拟来验证我们的方法。我们发现，与仅对阿尔戈观测进行最优插值相比，通过机器学习模型纳入海表数据可提高时空MLD变异性估计的性能。这些初步结果是将机器学习应用于MLD预测的有希望的第一步。
同时使用海洋剖面和海面数据的机器学习模型比仅使用剖面的模型能更好地改进混合层深度变异性估计
模型性能取决于空间位置和次季节方差的强度
概率参数技术比采样或集合方法能更好地捕捉不确定性
The ocean mixed layer plays an important role in the coupling between the upper ocean and atmosphere across a wide range of time scales. Estimation of the variability of the ocean mixed layer is therefore important for atmosphere‐ocean prediction and analysis. The increasing coverage of in situ Argo profile data allows for an increasingly accurate analysis of the mixed layer depth (MLD) variability associated with deviations from the seasonal climatology. However, sampling rates are not sufficient to fully resolve subseasonal ( day) MLD variability. Yet, many multivariate observations‐based analyses include implicit modeled subseasonal MLD variability. One analysis method is optimal interpolation of in situ data, but the interior analysis can be improved by leveraging surface data with regression or variational approaches. Here, we demonstrate how machine learning methods and satellite sea surface temperature, salinity, and height facilitate MLD estimation in a pilot study of two regions: the mid‐latitude southern Indian and the eastern equatorial Pacific Oceans. We construct multiple machine learning architectures to produce weekly 1/2° gridded MLD anomaly fields (relative to a monthly climatology) with uncertainty estimates. We test multiple traditional and probabilistic machine learning techniques to compare both accuracy and probabilistic calibration. We validate our methodology by applying it to ocean model simulations. We find that incorporating sea surface data through a machine learning model improves the performance of spatiotemporal MLD variability estimation compared to optimal interpolation of Argo observations alone. These preliminary results are a promising first step for the application of machine learning to MLD prediction. Machine learning models using both ocean profile and surface data improve mixed layer depth variability estimates over models using profiles alone Model performance is dependent on spatial location and strength of the sub‐seasonal variance Probabilistic parametric techniques capture uncertainty better than sampling or ensemble approaches