基于GODAS资料的南海上层海洋热状态气候变化特征分析

为了深入了解南海上层海洋热力状态的变化规律,利用1980—2015年共36年GODAS月平均海温资料,将5~366 m的垂直平均海温表征南海地区海洋上层的热含量,分析了南海海洋上层热状态的水平和垂直分布特征以及季节和年际变化特征。结果表明：年平均南海热含量水平分布表现为东高西低的形势,垂直纬向平均分布表现为暖水厚度和温跃层深度东厚（深）西薄（浅）,垂直经向平均表现为暖水厚度南厚北薄,温跃层深度中间浅两边深;南海地区海温变化幅度在75~200 m处最大,不同深度海温距平均具有明显的年际和年代际变化特征;南海区域平均热含量在秋季最高,春夏次之,冬季最低,其年际变化明显,且在1998年之后出现明显的突变,由负值转变为正值,表现出明显的增温趋势;热含量季节EOF主模态空间分布形势表现为东高西低的特征,对应的时间序列在20世纪90年代末存在年代际转折,由主要为负值转化为主要为正值,表现在空间分布上,则为南海地区热含量由西高东低型转化为东高西低型。

Upper Ocean Thermal State in the South China Sea: Analysis of Climate Change Characteristics Based on GODAS Dataset

To understand the change rules of the upper ocean thermal state in the South China Sea (SCS), the vertical mean sea temperature of 5-366 m was characterized by the thermal content (HC) of the upper ocean in the South China Sea, the horizontal and vertical distribution characteristics and seasonal and interannual variability of upper thermal state in the South China Sea was analyzed based on the monthly subsurface sea temperature data of GODAS from 1980 to 2015. The results showed that: the horizontal distribution of annual mean HC in the SCS was characterized by high in east and low in west, the vertical zonal mean distribution was characterized by warm pool thickness and thermocline depth was thick (deep) in east and thin (shallow) in west; the vertical meridional mean performance was that the warm pool thickness was thick in the south and thin in the north, and the depth of the thermocline was shallower in the middle and deep on both sides; the biggest change of the sea temperature was kept in the subsurface (75-200 m) over the SCS, and the anomaly of the temperature in different depths had obvious interannual and interdecadal variability; the average HC in the SCS region was the highest in autumn, followed by that in spring and summer, and the lowest was in winter and the interannual variability of HC was obvious, and there was a significant mutation after 1998, which changed from negative to positive, showing an obviously warming trend; the spatial distribution of the first mode of S-EOF in the HC revealed high in east and low in west, the corresponding time series had interdecadal transitions in the late 1990s, which were mainly negative values transformed into positive values, in terms of spatial distribution, the HC was transformed from the west high and east low type to the east high and west low type.