Xixi WEN1,2
Corresponding author: Wansuo DUAN,duanws@lasg.iap.ac.cn;
1.The State Key Laboratory of Numerical Modelling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China2.University of Chinese Academy of Sciences, Beijing 100049, China
Manuscript received: 2018-07-02
Manuscript revised: 2018-10-16
Manuscript accepted: 2018-10-29
Abstract:Using the Regional Ocean Modeling System, this study investigates the simulation uncertainties in the current velocity in the low-latitude North Pacific where the Kuroshio originates [i.e., the beginning of the Kuroshio (BK)]. The results show that the simulation uncertainties largely reflect the contributions of wind stress forcing errors, especially zonal wind stress errors, rather than initial or boundary errors. Using the idea of a nonlinear forcing singular vector, two types of zonal wind stress errors (but sharing one EOF mode) are identified from error samples derived from reanalysis data as having the potential to yield large simulation uncertainties. The type-1 error possesses a pattern with positive anomalies covering the two zonal bands of 0°-15°N and 25°-40°N in the Pacific Ocean, with negative anomalies appearing between these two bands; while the type-2 error is almost opposite to the type-1 error. The simulation uncertainties induced by the type-1 and -2 errors consist of both large-scale circulation errors controlled by a mechanism similar to the Sverdrup relation and mesoscale eddy-like errors generated by baroclinic instability. The type-1 and -2 errors suggest two areas: one is located between the western boundary and the meridional 130°E along 15°-20°N, and the other is located between 140°-150°E and along 15°-20°N. The reduction of errors over these two areas can greatly improve the simulation accuracy of the current velocity at BK. These two areas represent sensitive areas for targeted observations associated with the simulation of the current velocity at BK.
Keywords: Kuroshio,
nonlinear forcing singular vector,
targeted observation,
摘要:该研究使用著名的区域海洋模式ROMS(Regional Ocean Modeling System), 对源区黑潮流速的模拟不确定性进行了解剖式研究. 结果表明, 在驱动模式运行的初始条件, 边界条件, 以及风应力外强迫中, 风应力强迫误差, 特别是纬向风应力误差常常导致源区黑潮流速具有更大的模拟误差. 为了揭示对源区黑潮模拟不确定性具有最大影响的纬向风应力误差, 本研究基于历史分析资料, 利用集合的办法, 识别了对源区黑潮具有最大影响的非线性强迫奇异向量(nonlinear forcing singular vector; NFSV)型-纬向风应力误差. 该误差可分为两种类型, 即type-1和type-2误差. type-1误差位于0°–15°N, 25°–40°N两个纬度带内, 且表现为横跨太平洋海盆的正风应力误差, 而在这两个纬度带之间的区域, type-1误差则表现为横跨太平洋海盆的负风应力误差; type-2误差与type-1型结构相同, 但符号几乎相反. type-1和type-2误差对源区黑潮流速模拟的影响, 一方面通过改变大尺度海洋环流影响源区黑潮的模拟; 另一方面则诱发海洋中尺度涡, 并促使其西移而影响源区黑潮. 根据type-1和type-2误差, 并通过敏感性实验, 确定了能够有效改善源区黑潮模拟效果的目标观测敏感区, 即位于15°–20°N之间, 呈东西分布, 且包含两个区域: 一个区域位于西边界与130°E之间; 另一个则位于135°–150°E之间.
关键词:源区黑潮,
模拟不确定性,
非线性强迫奇异向量,
目标观测
