Inertial-Convective Subrange Estimates of Thermal Variance Dissipation Rate from Moored Temperature Measurements
Inertial-Convective Subrange Estimates of Thermal Variance Dissipation Rate from Moored Temperature Measurements
复制标题
DOI：
10.1175/2010jtecho746.1
复制
发表时间：
2010-11
影响因子：
2.2
通讯作者：
Yanwei Zhang;J. Moum
中科院分区：
文献类型：
--
作者：
Yanwei Zhang;J. Moum
估计热方差耗散率x T的程序，通过缩放的惯性对流温度梯度谱的热敏电阻测量热带大气海洋（TAO）赤道系泊，由NOAA的国家数据浮标中心，演示。的inertialconventional subrange的wavenumbers/频率被污染的垂直运动所引起的泵的表面浮通过当地的垂直温度梯度的表面重力波。未受污染的信号可以通过去除与由表面重力波引起的信号相干的测量信号的部分来恢复，所述表面重力波必须被独立地测量。然后，通过将修正后的谱与惯性对流子范围（0.05，f，0.5 Hz）上的理论温度梯度谱拟合，获得x T的估计值;该估计值称为x T。在这里，x T计算超过120分钟的时间间隔，并与估计的x T确定缩放温度梯度谱在高波数（粘性对流和粘性扩散子范围）。大的差异高达20倍和未知的起源很少发生，特别是当背景电流和垂直温度梯度都很弱，但本文的结果表明，75%的数据对是在一个因素的3彼此。15分钟，30分钟，60分钟，120分钟的时间间隔的测试表明，这两种方法之间的差异几乎是随机的，无偏的，并小于从无关的实验在同一地点确定的自然变异的估计。由于惯性对流子范围占据了比通常用于湍流测量的频率范围更低的频率范围，因此存在更多常规测量x T的可能性。退化信号的评估（从原始测量重新采样）表明，这种测量的一个特别重要的组成部分是表面波感应信号的独立分辨率。
A procedure for estimating thermal variance dissipation rate x T by scaling the inertial-convective subrange of temperature gradient spectra from thermistor measurements on a Tropical Atmosphere Ocean (TAO) equatorial mooring, maintained by NOAA’s National Data Buoy Center, is demonstrated. The inertialconvective subrange of wavenumbers/frequencies is contaminated by the vertical motion induced by the pumping of the surfacefloat by surface gravity waves through the local vertical temperature gradient. The uncontaminated signal can be retrieved by removing the part of the measured signal that is coherent with the signal induced by surface gravity waves, which must be measured independently. An estimate of x T is then obtained by fitting corrected spectra to theoretical temperature gradient spectra over the inertial-convective subrange (0.05 , f , 0.5 Hz); this estimate is referred to as x T. Here x T was calculated over 120-min intervals and compared with estimates of x T determined by scaling temperature gradient spectra at high wavenumbers (viscous-convective and viscous-diffusive subranges). Large differencesup to a factor of 20 and of unknown origin occur infrequently, especially when both background currents and vertical temperature gradients are weak, but the results herein indicate that 75% of the data pairs are within a factor of 3 of each other. Tests on 15-, 30-, 60-, 120-min intervals demonstrate that differences between the two methods are nearly random, unbiased, and less than estimates of natural variability determined from unrelated experiments at the same location. Because the inertial-convective subrange occupies a lower-frequency range than is typically used for turbulence measurements, the potential for more routine measurements of x T exists. The evaluation of degraded signals (resampled from original measurements) indicates that a particularly important component of such a measurement is the independent resolution of the surface wave‐induced signal.