中国科学院海洋研究所机构知识库

研究者们已经认识到在全球海洋中普遍存在着中尺度海气相互作用现象，并对其进行了广泛的研究。但是，目前的研究还未能完全理解在中纬度海区中尺度海面温度扰动（SST_meso）对大气的影响机制及气候效应。在本研究中，使用高分辨率的卫星观测资料和WRF（Weather Research and Forecasting Model）模式研究了黑潮–亲潮延伸体海区大气对SST_meso的响应。本研究的主要研究方法是进行WRF模式对比试验，其中控制试验（CTRL）由不包含SST_meso的低分辨率海面温度边界条件驱动，而扰动试验（PER）中海面温度边界条件额外添加了从高分辨率卫星观测资料中提取的SST_meso。其中SST_meso的提取采用了LOESS（locally weighted smoothing）技术，可有效分离中尺度扰动信号。通过计算CTRL与PER的模拟结果的差异，就能得到在SST_meso的作用下产生的大气响应场。这一对比试验方法属首次被应用于研究中纬度海区的大气。

对月平均的大气响应场与观测资料进行比较分析的结果显示，WRF 模式很好地模拟出了边界层内的大气对SST_meso的响应，并且响应信号可以影响到对流层的中下层。大气响应场的空间分布同卫星观测以及前人的模拟结果高度一致。线性回归分析显示WRF模式中海面上10 m处风速扰动和SST_meso之间具有合理的相关系数，但是耦合强度较弱，说明模式使用的大气边界层参数化方案仍有待改进。我们还对大气边界层内的动力响应现象进行了机制分析，结果显示研究海区内压力调整机制和垂向混合机制都在起作用。要判断哪种机制起主导作用，必须根据具体情况进行讨论。

我们进一步使用WRF模式对2005到2013年的模拟结果研究了中尺度海面温度场的气候效应。我们发现经过时间平均后研究海区的SST_meso主要表现为三处海面温度锋面。这些锋面对局地的气候态降水产生了显著的影响，在锋面的暖区（冷区）降水及其变率都有所增加（减少），其净效应是局地降水的增加。这些锋面的作用改善了PER中模拟得到的气候态降水的空间分布，使其与观测结果更加接近，说明了在大气模式中使用高分辨海面温度边界条件的必要性。我们进一步分析了在锋面处产生降水响应的物理机制。结果证明了在过去针对湾流海区的降雨带所提出的压力调整机制能够推广到本研究的海区中，并成功地解释了由三处锋面引起的气候态变量场的响应。我们还发现仅靠压力调整机制无法圆满地解释气候态降水响应的季节变化现象，因此需要考虑其他因素，例如背景场的环境条件以及SST_meso的热力作用对温带气旋活动的影响。

The ubiquitous mesoscale air–sea interaction in the global oceans has been recognized and extensively investigated. Currently, however, how mesoscale sea surface temperature (SST) perturbations (SST_meso) affect the atmosphere in mid-latitude regions as well as the related climatic effects have not been fully understood. In this study, high-resolution satellite observations and the WRF model (Weather Research and Forecasting Model) are used to investigate the atmospheric responses to SST_meso in the Kuroshio-Oyashio Extension (KOE) region. The main approach is to perform a pair of WRF simulations, in which the control run (CTRL) is forced by low-resolution SST fields (almost no mesoscale SST signals) and the perturbation run (PER) by superimposing additional SST_meso fields extracted from high-resolution satellite observations. To extract SST_meso, the LOESS (locally weighted smoothing) method is adopted, which can effectively isolate mesoscale signals. The atmospheric responses to SST_meso are isolated by calculating the differences between the output from CTRL and PER. It is the first time this experimental setup is used to study the atmosphere at the mid-latitude.

Our analysis of the monthly mean atmospheric responses and satellite observations shows that the WRF model reproduces the boundary layer atmospheric responses to SST_meso well, and the responses can affect the middle and lower troposphere. The spatial patterns of mesoscale atmospheric responses agree quite well with the satellite observations and other previous model results. Results of the linear regression between the 10-m wind perturbations and SST_meso indicate that in the WRF model, the high positive correlation between the two fields is well captured but the coupling strength is relatively weak, suggesting the boundary layer parameterization needs to be improved. An examination on the underlying mechanism suggests that in the KOE region the pressure adjustment mechanism and the vertical mixing mechanism are both at work. The dominant mechanism at work should be assessed based on specific environmental conditions.

The climatic effects of mesoscale SST fields are further investigated using model output from 2005 to 2013. The time-mean SST_meso in the KOE region is characterized by three SST fronts, which have a significant impact on local climatological precipitation. Climatological precipitation and its variability increase (decrease) on the warmer (colder) flank of the fronts and the net effect is a rectification on local precipitation. This effect improves the spatial distribution of the climatological precipitation simulated by PER and makes it agree better with observations, which indicates the important role of high-resolution SST boundary conditions in atmospheric modeling. The mechanism of the climatological precipitation response is further analyzed and the result shows that the pressure adjustment mechanism previously proposed to explain the rain band along the Gulf Stream is also valid in the KOE region and successfully explains several climatological features near the fronts. It is also noted that the pressure adjustment alone is not sufficient to explain the variability in the climatological precipitation response, and thus other factors such as the environmental conditions and the thermodynamic effect of SST_meso on extratropical cyclones need to be taken into consideration.