Nighttime Cool Skin Effect Observed from Infrared SST Autonomous Radiometer (ISAR) and Depth Temperatures
Nighttime Cool Skin Effect Observed from Infrared SST Autonomous Radiometer (ISAR) and Depth Temperatures
复制标题
从红外 SST 自主辐射计 (ISAR) 和深度温度观察到的夜间凉爽皮肤效应
DOI：
10.1175/jtech-d-19-0161.1
复制
发表时间：
2020
影响因子：
2.2
通讯作者：
A. Babanin
中科院分区：
文献类型：
--
作者：
Haifeng Zhang;H. Beggs;A. Ignatov;A. Babanin
利用103天的船载红外SST自主辐射计（ISAR）SST skin与澳大利亚附近海域~7.1-9.9-m深度的取水口SST depth的匹配，研究了夜间海洋冷表皮信号ΔT [定义为表皮海表温度（SST skin）减去深度SST（SST depth）]。在数据分析之前，对ISAR SST皮肤数据进行严格的质量控制，并仔细地将可能的昼夜变暖污染降至最低。ΔT的统计分布及其对风速、热通量等的依赖性，与之前的发现一致。总平均ΔT值为−0.23 K。观察到冷皮肤信号的幅度在午夜之后增加，并且在黎明附近发现最冷的皮肤偏移（平均值为-0.36 K）。观察到ΔT对SST条件的依赖性。当净热通量的方向是从大气到海洋时，直接暖肤事件被发现，这更有可能发生在高纬度，当空气非常潮湿和比SST更温暖时。此外，还对几种冷肤模型进行了验证：一种广泛使用的物理模型表现最好，可以捕捉到大多数趋肤效应的趋势和细节;经验模型只反映了观测到的ΔT值的基本特征。如果用户不能应用物理模型（例如，算法复杂性或缺失输入），则可以使用2002年研究中提出的形式的经验参数化。然而，我们建议使用一组新的参数，在本研究中计算，基于更具代表性的数据集，并进行更严格的质量控制。
The nighttime ocean cool skin signal ΔT [defined as skin sea surface temperature (SSTskin) minus depth SST (SSTdepth)] is investigated using 103 days of matchups between shipborne Infrared SST Autonomous Radiometer (ISAR) SSTskin and water intake SSTdepth at ~7.1–9.9-m depths, in oceans around Australia. Before data analysis, strict quality control of ISAR SSTskin data is conducted and possible diurnal warming contamination is carefully minimized. The statistical distribution of ΔT, and its dependencies on wind speed, heat flux, etc., are consistent with previous findings. The overall average ΔT value is −0.23 K. It is observed that the magnitude of the cool skin signal increases after midnight and a coolest skin offset (with an average value of −0.36 K) is found at around dawn. The dependency of ΔT on SST conditions is observed. Direct warm skin events are discovered when the net heat flux direction is from the atmosphere to the ocean, which is more likely to occur at high latitudes when the air is very humid and warmer than the SST. In addition, several cool skin models are validated: one widely used physical model performs best and can capture most skin-effect trends and details; the empirical models only reflect the basic features of the observed ΔT values. If the user cannot apply the physical model (due to, e.g., the algorithm complexity or missing inputs), then the empirical parameterization in the form proposed in a 2002 study can be used. However, we recommend using a new set of parameters, calculated in this study, based on much more representative dataset, and with more rigorous quality control.