【作者】 李丕学；

【导师】 张占海；

【作者基本信息】 中国海洋大学 ， 物理海洋， 2009， 博士

【摘要】 北冰洋是地球系统的一部分,它直接影响着全球尺度的大气环流、海洋环流和气候变异,因而成为全球气候变化的驱动器之一。近30年来北冰洋海冰的快速减少最为引人注目,而入海径流是北冰洋重要的淡水来源。入海径流的变化直接影响到北冰洋盐跃层的生长和消退,进而影响海冰的热力过程和海洋温盐环流。而且由北冰洋的海冰输出又直接影响北大西洋北部深层对流的强度,进而影响全球海洋温盐环流,对气候变化具有显著影响。在前人研究的基础上,本论文1)对径流的季节和年代际变化规律、影响径流的关键气候因子、径流与海冰的关系三个问题进行了资料分析;2)利用全球海冰-海洋耦合模式探索了入海径流对北冰洋上层海水结构、海水-海冰间的通量交换、海冰分布和北大西洋深层水生成的影响及可能机制;3)初步建立高分辨率的北冰洋区域海冰-海洋耦合模式模拟气候态北冰洋海冰和海洋的状态,并进一步研究入海径流对北冰洋海冰-海洋的影响。首先,通过对径流资料的分析发现:第一,北极地区的径流表现出剧烈的季节变化,并且近30年的年际变化明显:最大的四条河流在冬季、春季径流量增加明显,在夏季径流则减少明显,年平均径流量变化趋势与秋季变化趋势完全相同,反应了融冰期提前,本属于夏季径流的冰雪融水叠加到了冬春季的流量中。第二,利用多元线性回归分析表明:春季,影响入海径流变化的主要因子是气温;夏季,影响Ob和Yenisei河径流量变化的因子依次是积雪覆盖面积和降水,而对于Lena河和Mack河,降水是最主要的影响因子;与欧亚大陆的三条河流不同,大气环流对Mackenzie河起着不可忽略的影响。第三,分析了径流与海冰的关系:入海径流与海冰相关的显著区域一般出现在河口区和海冰边缘区域。春季,四条河流在河口附近为负相关,夏季,在春季负相关区域外侧为正相关。格陵兰寒流区域在春季与Ob河正相关,夏季与Lena河正相关,秋季与Ob、Mackenzie河负相关,冬季与Lena、Ob河负相关。其次,通过数值实验研究了径流的季节变化对海冰海洋的影响,与资料分析结果相似,对海冰的影响主要集中在海冰边缘区,东格陵兰寒流是主要的海冰输出流系,径流对此处海冰的影响代表了对海冰输出的影响,当冬春季径流量增加时,东格陵兰寒流等地的海冰密集度和冰厚增加,而在斯瓦尔巴岛与新地岛之间(大西洋暖水和北冰洋表层水的锋面区)的海冰密集度和冰厚则减少。而夏季径流减少同样能引起斯瓦尔巴岛与新地岛之间的海冰密集度和冰厚减少。径流的季节变化对盐度的影响最剧烈的地区在河口附近,河口区盐度的变化和月径流量有很好的相关性,当径流量小于年平均流量时,盐度为正值,反之为负值,盐度随径流量反相变化,并且有一到两个月滞后。淡水输入的多少是河口区盐度变化的最直接、最主要的原因。但对河口区的温度影响并不明显。值得注意的是格陵兰岛南侧下沉流的区域,当冬春季径流减少,夏秋季径流增大时,温度和盐度同时降低,下沉流有增强的趋势,而近30年来,径流在冬春季明显增加,在夏季明显减少,因此减弱了下沉流,阻碍了大西洋深层水的生成。最后,建立了北冰洋区域冰海耦合模式,利用与全球模式完全一样的初始场和强迫场,来模拟北冰洋的温盐结构和海冰情况,并与观测资料和全球模拟结果对比。结果表明:模拟的海冰基本体现了12个月的海冰形态和变化规律,在冬季春季模拟效果比全球模式要好,尤其在格陵兰海和巴伦支海地区,而在拉布拉多海和巴芬湾地区则不如全球模式。在夏秋季节,区域模式对楚科奇海的海冰模拟偏少。模拟的温盐也基本体现了表层还是低温低盐,次表层水高温高盐的特征,但在表层盐度的模式上存在一定的不足,北冰洋中心盐度偏高1psu,冷盐跃层的模拟优于全球模式,与观测结果吻合较好。更多还原

【Abstract】 The Arctic can be viewed as an integrated system, characterized by intimate couplings between it atmosphere, ocean and sea ice, linked in turn to the larger global system. River inflow is the major part of the Arctic freshwater budget, and of special importance in the Arctic Ocean. This paper discusses the seasonal and decadal variation, dominant climate factors of the arctic runoff, and association between the arctic runoff and sea ice; then using a coupled ice-ocean model studies the impacts on the ocean and sea ice; finally sets up arctic regional model to simulate the climate mean sate of the Arctic Ocean and sea ice.The results suggest that the arctic runoff exhibits an increasing trend for spring and winter and a decreasing trend in summer. The surface air temperature is the dominant factor in influencing for all the four rivers in spring. In summer, snow cover is the most important factor for the Ob and Yenisei rivers, while precipitation is the most important factor for Lena and Mackenzie. For Mackenzie, atmospheric circulation does play an important role for all the seasons, which is not the case for the Eurasian rivers. We further discuss the relationships between the Arctic runoff and sea ice. It shows significant relation in estuaries and sea ice edge, such as east Greenland current, Siberian coast and so on. It is noteworthy that Lena impacts the sea ice along east Greenland remarkably, negatively correlated in spring and winter, positively correlated in summer and autumn.The model results are similar to the observation results. The influence to sea ice is in its edge, such as east Greenland current and the sea area between Svalbard and Novaya Zemlya, an oceanic front. The sea ice concentration and thickness increase along east Greenland, decrease between Svalbard and Novaya Zemlya as the runoff rises in spring and winter. The summer and autumn situation are similar to the spring when the summer runoff reduces. The main affected zone to salinity is estuarine area. Relations between salinity and runoff are obvious anti-phase changed. When runoff increase in spring and winter, decrease in summer and autumn, the downwelling in the north Atlantic becomes weaken.The regional model and global model have merits and demerits. Regional model sea ice results are better in winter, while global model results are better in summer. For the salinity and temperature simulations, regional model is better in Atlantic layer and global model in surface.更多还原