Bulgin Claire E, Merchant Christopher J, Ferreira David
University of Reading, Department of Meteorology, Reading, RG6 6AL, UK.
National Centre for Earth Observation, Leicester, UK.
Sci Rep. 2020 May 14;10(1):7986. doi: 10.1038/s41598-020-64785-9.
Quantifying global trends and variability in sea surface temperature (SST) is of fundamental importance to understanding changes in the Earth's climate. One approach to observing SST is via remote sensing. Here we use a 37-year gap-filled, daily-mean analysis of satellite SSTs to quantify SST trends, variability and persistence between 1981-2018. The global mean warming trend is 0.09 K per decade globally, with 95% of local trends being between -0.1 K and + 0.35 K. Excluding perennial sea-ice regions, the mean warming trend is 0.11 K per decade. After removing the long-term trend we calculate the SST power spectra over different time periods. The maximum variance in the SST power spectra in the equatorial Pacific is 1.9 K on 1-5 year timescales, dominated by ENSO processes. In western boundary currents characterised by an intense mesoscale activity, SST power on sub-annual timescales dominates, with a maximum variance of 4.9 K. Persistence timescales tend to be shorter in the summer hemisphere due to the shallower mixed layer. The median short-term persistence length is 11-14 days, found over 71-79% of the global ocean area, with seasonal variations. The mean global correlation between monthly SST anomalies with a three-month time-lag is 0.35, with statistically significant correlations over 54.0% of the global oceans, and notably in the northern and equatorial Pacific, and the sub-polar gyre south of Greenland. At six months, the mean global SST anomaly correlation falls to 0.18. The satellite data record enables the detailed characterisation of temporal changes in SST over almost four decades.
量化全球海表温度（SST）的趋势和变率对于理解地球气候变化至关重要。观测海表温度的一种方法是通过遥感。在此，我们利用37年填补间隙的卫星海表温度日均值分析，来量化1981年至2018年间的海表温度趋势、变率和持续性。全球平均变暖趋势为每十年0.09开尔文，95%的局部趋势在-0.1开尔文至+0.35开尔文之间。排除常年海冰区域后，平均变暖趋势为每十年0.11开尔文。去除长期趋势后，我们计算了不同时间段的海表温度功率谱。赤道太平洋海表温度功率谱在1 - 5年时间尺度上的最大方差为1.9开尔文，主要由厄尔尼诺 - 南方涛动（ENSO）过程主导。在以强烈中尺度活动为特征的西边界流中，亚年时间尺度上的海表温度功率占主导，最大方差为4.9开尔文。由于混合层较浅，夏季半球的持续性时间尺度往往较短。短期持续性长度的中位数为11 - 14天，在全球海洋面积的71% - 79%范围内出现，存在季节变化。月海表温度异常之间三个月时间滞后的全球平均相关性为0.35，在全球海洋面积的54.0%以上存在统计显著相关性，特别是在北太平洋和赤道太平洋以及格陵兰岛以南的亚极地环流区。在六个月时，全球海表温度异常平均相关性降至0.18。卫星数据记录能够详细描述近四十年来海表温度的时间变化。
