基于浮标观测资料及再分析数据集的北欧海海气热通量特征研究

发布时间：2018-08-14 17:57

【摘要】：作为北大西洋暖流进入极区的唯一通道,北欧海是北半球亚极区海洋径向热量输送的最重要通道,在输运过程中海洋内部的热量以感热和潜热的方式源源不断向大气中释放,成为驱动北半球高纬度大气环流的重要一环。研究表明北欧海海域是北极涛动和北大西洋涛动的核心区域,海气湍流热交换过程一方面影响北欧海大气下界面的能量收支;另一方面海气湍流热交换的变化还会导致局地海表面的冷却,从而引起上层海洋热含量的异常变化,进而对北大西洋深层水的形成存在间接影响。北欧海海气热量输送的变化还与格陵兰海气旋活动存在密切关系,而格陵兰海气旋活动的路径和强度变化影响着北极大西洋扇区的液态水输送和海冰变化。因此,研究北欧海海气热通量对于研究整个北半球大尺度大气环流的变化有着重要贡献。同时,海气界面湍流热通量的变化对于海洋与大气环流模式的驱动,海气相互作用的研究,以及数值天气预报模式的评估和评价等都有重要意义。对北欧海海气界面热通量的研究需要借助再分析资料,然而,现有的海气通量数据集(包括卫星遥感反演数据和再分析数据)中的海气通量场都存在着不确定性。造成偏差的主要原因一是块体算法中对海表面粗糙度参量的估算不够精确;二是计算热通量所需的各海气界面要素存在测量误差。因此,充分利用浮标现场观测数据,改进海气热通量估算方法,以及对目前常用的相关数据产品进行验证和比较评估,显得十分迫切和重要。2012年我国首次在北欧海罗弗敦海盆(5°E,70°N)进行了23天连续的现场浮标观测,通过获得的数据,本文首先对获得的资料进行了分析和描述,并计算出海表湍流热通量,对海表面要素和热通量之间的关系以及观测期间热输送的短期变化特征进行了进一步分析:海气界面感热通量为-5.21W/m~2~41.15W/m2,潜热通量为-3.59W/m2~82.81W/m2,海洋向大气为正,说明罗弗敦海盆夏季热通量以海洋向大气传递热量为主。两者都呈现升高趋势,潜热通量平均是感热通量的3.4倍。海气温差是北欧海罗弗敦海盆夏季影响感热输送的主导因素;在中低风速状态下,海气湿度差是引起潜热通量变化的主导因素。功率谱分析显示感热通量存在15.4天的主要周期,而潜热通量存在15.4天及3天左右的高频周期。其次,本文利用获得的数据对四套再分析资料集(ERA-Interim、OAFlux、NCEP1及NCEP2)在高纬度海域的估算进行了比较分析和评估:四类再分析数据集对海表面10m风速、2m大气温度、2m空气比湿的估算与实测值相关性良好,从统计特征来看ERA-Interim和OAFlux的数据在此海域夏季期间要优于NCEP1和NCEP2.四类数据集和实测相差最大的是海表皮温。ERA-Interim对海表高频波动的变化更敏感。造成感热通量偏差的原因是对海表气温和皮温不同程度的高估或低估。ERA-Interim对感热的估算优于其它三者。四者对潜热的估算在潜热值较大时误差也增大,其中NCEP1和NCEP2相对ERA-Interim和OAFlux对实测数据有明显的高估,这些误差应该是来自对海表风速和海表比湿的估算偏差。OAFlux对潜热的估算优于其他三者。本文利用和实测数据吻合最好的ERA-Interim月均数据对北欧海的年内季节变化和长期时空变化特征进行了分析。结果表明北欧海作为全球大洋少数强海气耦合海域,感热和潜热释放全年都大于0(海洋向大气为正),三种热通量都在冬季达到最大值,而夏季最小,年内变化都呈单峰周期性变化。感热不同的季节分布呈现明显的纬向变化,潜热季节变化则更多的呈现经向变化。感热年内变化最大的是斯瓦尔巴岛西侧的暖流回流区;潜热变化最大的是西斯匹兹卑尔根流流经海域和冰岛西部、南部海域,潜热波动值仅为感热的一半。对净热通量多年逐月距平场的EOF第一模态方差贡献占48.25%,代表了北欧海距平值一致的空间变化,其中东格陵兰寒流海域是正负异常变化最明显的海域。时间序列年平均表明第一模态在逐渐由正异常转为负异常,从侧面也说明净热整体降低的趋势。第二模态和第三模态贡献率为12.5%和11.1%,空间分别呈现南北偶极子和东西跷跷板分布,其中第二模态有明显的四年左右的周期,第三模态的时间序列和NAO指数则有0.6的正相关,说明北大西洋涛动带来的气压变化对北欧海海气热输送有不小的影响。利用SODA再分析资料的海水垂向温度数据和ERA-Interim的气压数据对北欧海海气净热通量与混合层深度、混合层温度、海表气旋活动之间的关系进行了初步探讨,发现热通量与混合层深度存在一个月的超前相关,即海洋向大气释放(吸收)热量一个月后混合层会变深(浅),相关性达0.9,其中暖流区相关性更好;热通量与混合层温度存在两个月的超前相关,即海洋向大气释放(吸收)热量两个月后混合层会变冷(暖),相关性达到-0.9,而空间分布较一致。北欧海热通量与气旋活动指数在长期变化上存在0.62的相关性,说明两者关系密切,但气旋活动肯定也受到其他因素制约。
[Abstract]:As the only passage for the North Atlantic warm current to enter the polar region, the Nordic Sea is the most important passage for the radial heat transfer in the northern hemisphere's subpolar region. During the process of transport, the heat in the ocean is released into the atmosphere by sensible heat and latent heat, which is an important part of the atmospheric circulation in the northern hemisphere. Sea area is the core area of Arctic and North Atlantic Oscillation. On the one hand, the turbulent heat exchange process affects the energy budget at the lower atmosphere interface of the Nordic Sea; on the other hand, the change of turbulent heat exchange between sea and air will lead to the cooling of the local sea surface, thus causing the abnormal change of the heat content in the upper ocean, and then to the deep North Atlantic. There is an indirect effect on the formation of water. The change of sea-air heat transfer in the Nordic Sea is also closely related to the cyclone activity in the Greenland Sea. The path and intensity of the cyclone activity in the Greenland Sea affect the liquid water transport and sea ice change in the Arctic Atlantic sector. Therefore, the study of the sea-air heat flux in the Nordic Sea is of great significance to the study of the whole Northern Hemisphere. Meanwhile, the variation of turbulent heat flux at the air-sea interface is of great significance to the driving of ocean-atmosphere circulation models, the study of air-sea interaction, and the evaluation and evaluation of numerical weather prediction models. However, there are uncertainties in the current air-sea flux data sets (including satellite remote sensing inversion data and reanalysis data). One of the main reasons for the deviation is that the estimation of sea surface roughness parameters in the block algorithm is not accurate enough, and the other is that there are measurement errors in the air-sea interface elements needed to calculate the heat flux. Therefore, it is very urgent and important to make full use of the buoy field data to improve the estimation method of air-sea heat flux and to verify and compare the commonly used data products. In this paper, the data obtained are analyzed and described, and the turbulent heat flux on the sea surface is calculated. The relationship between sea surface elements and heat flux and the short-term variation characteristics of heat transfer during the observation period are further analyzed. The sensible heat flux at the sea-air interface is - 5.21W/m~2~41.15W/m 2, and the latent heat flux is - 3.59W/m~2~82.81W/m 2. The ocean-to-atmosphere heat flux is positive, indicating that the ocean-to-atmosphere heat flux is the main heat flux in the Lofton Basin in summer. Both of them show an upward trend, and the latent heat flux is 3.4 times of the sensible heat flux on average. Power spectrum analysis shows that sensible heat flux has a main period of 15.4 days, while latent heat flux has a high frequency period of 15.4 days and 3 days. The data of ERA-Interim and OAFlux are better than those of NCEP1 and NCEP2. The difference between ERA-Interim and NCEP2 is the largest. ERA-Interim estimates sensible heat better than the other three. Four estimates of latent heat also increase when the latent heat value is large. NCEP1 and NCEP2 overestimate the measured data compared with ERA-Interim and OAFlux. These errors should be attributed to the estimation bias of sea surface wind speed and specific humidity.OAFlux is superior to the other three methods in estimating latent heat.This paper analyses the seasonal and long-term spatial and temporal variations of the Nordic Sea using ERA-Interim monthly mean data with the best agreement with the measured data.The results show that the Nordic Sea is less of a global ocean. The sensible heat and latent heat release in several strong air-sea coupled waters are greater than 0 (the ocean is positive to the atmosphere) all year round, and the three heat fluxes reach the maximum in winter, while the summer is the smallest, and the annual variation is a single peak periodic change. The largest internal variation is in the warm current recirculation zone on the western side of Svalbard Island, while the greatest change in latent heat is in the west of Iceland and the western part of the western part of the West Spitsbergen Current. The latent heat fluctuation is only half of the sensible heat in the southern part of the sea. The annual average of time series shows that the first mode is gradually changing from positive to negative anomaly, and the overall decrease trend of net heat is also shown from the side. The contribution rates of the second mode and the third mode are 12.5% and 11.1%, respectively. The space shows the North-South dipole and the east-south dipole respectively. Western seesaw distribution, in which the second mode has an obvious period of about four years, the third mode time series and NAO index have a positive correlation of 0.6, indicating that the North Atlantic Ocean Oscillation brings about a change in atmospheric pressure in the Nordic Sea air-sea heat transfer has a small impact. The relationship between the air-sea net heat flux and mixing layer depth, mixing layer temperature and sea surface cyclone activity in Northern Europe is preliminarily discussed. It is found that there is a one-month advance correlation between the heat flux and mixing layer depth, that is, the mixing layer will deepen (shallow) after one month when the ocean releases (absorbs) heat to the atmosphere, and the correlation reaches 0.9, in which the warm current phase is formed. There is a two-month advance correlation between heat flux and mixing layer temperature, that is, the mixing layer will become cold (warm) two months after the ocean releases (absorbs) heat to the atmosphere, the correlation reaches - 0.9, and the spatial distribution is more consistent. Cyclonic activity is also restricted by other factors.
【学位授予单位】：上海海洋大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P715.2;P732.6

【参考文献】