Decomposing observations of high-frequency radar-derived surface currents by their forcing mechanisms: Decomposition techniques and spatial structures of decomposed surface currents
Decomposing observations of high-frequency radar-derived surface currents by their forcing mechanisms: Decomposition techniques and spatial structures of decomposed surface currents
复制标题
通过其强迫机制对高频雷达产生的表面电流进行分解观测：分解技术和分解表面电流的空间结构
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
E. Terrill
中科院分区：
文献类型：
--
作者：
S. Kim;B. Cornuelle;E. Terrill
[1]从部署在圣地亚哥南部的高频雷达网络的表面电流观测分解，根据其驱动力：纯潮汐及其邻近的带外能量，当地风，和低频。几个叠加的海洋响应是由于复杂的海底地形和相对较弱的风关闭南部圣地亚哥，而不是沿海地区的环流可以解释为一个占主导地位的强迫机制。这需要应用统计分解方法。用调和分析法计算了与纯潮相干的表层流。局部风驱动的表面电流估计观测到的风对观测到的表面电流的回归。采用加权最小二乘拟合方法对脱风和退潮表层流进行滤波，假设噪声为白色，信号带为低频带（小于0.4周/天）和近潮峰（周日（K1）和半日（M2）频率）。分解的表面流的每一部分的空间和时间的变化进行了研究海洋响应的驱动力。此外，各个分量的空间相关性表现出不同的去相关长度尺度的高斯和指数形状。
[1] Surface current observations from a high-frequency radar network deployed in southern San Diego are decomposed according to their driving forces: pure tides and their neighboring off-band energy, local winds, and low frequency. Several superposed ocean responses are present as a result of the complicated bottom topography and relatively weak winds off southern San Diego, as opposed to coastal regions where circulation can be explained by a dominant forcing mechanism. This necessitates an application of a statistical decomposition approach. Surface currents coherent with pure tides are calculated using harmonic analysis. Locally wind-driven surface currents are estimated by regression of observed winds on observed surface currents. The dewinded and detided surface currents are filtered by weighted least-squares fitting assuming white noise and three colored signal bands: low-frequency band (less than 0.4 cycles per day) and near-tidal peaks at the diurnal (K1) and semidiurnal (M2) frequencies. The spatial and temporal variability of each part of the decomposed surface currents is investigated in terms of ocean response to the driving forces. In addition, the spatial correlations of individual components exhibit Gaussian and exponential shapes with varying decorrelation length scales.