>This post illustrates an observation I’ve made a number of times while investigating other subjects, so I thought I’d post it. It shows the relative strengths of the polar SST anomalies.
Using the NCDC’s Optimum Interpolation (OI.v2) SST Anomaly dataset, which begins in November 1981, the Arctic Ocean shows a significant positive linear trend. Refer to Figure 1. Over the same period, the Southern Ocean shows a negative linear trend that’s much less significant. The latitudes I’ve used for the Arctic Ocean are 65N-90N, while the Southern Ocean latitudes are 90S-60S.
According to Wikipedia, the area of the Arctic Ocean is 14,056,000 sq. km. The Southern Ocean, on the other hand, has an area of 20,327,000 sq. km or about 45% larger than the Arctic Ocean. Sea ice also varies in the polar oceans, as is well known.
Taking those factors into consideration, which plays a greater role in global SST anomaly trends, the rise in the Arctic Ocean SST anomalies, or the drop in the Southern Ocean SST anomalies?
The easiest way to illustrate it is to compare global SST anomalies with and without the polar oceans. (Or someone other than me could go through all of the calculations to factor in area, monthly sea ice extent, etc., using polar ocean data.) If the rise in the SST anomalies of the Arctic Ocean is having the greater impact on global SST anomalies than the drop in the Southern Ocean data, then the Global SST anomaly dataset including the polar oceans would have a higher trend than the a global dataset without the polar oceans. Refer to Figure 2. The opposite takes place.
While the two linear trends are close, the drop in the Southern Ocean SST anomalies has the greater effect. That is, adding the polar oceans to the data caused the global SST anomaly trend to decrease.
Someone will note that the latitudes of the Arctic and Southern Ocean datasets are not the same. The Southern Ocean is generally defined as the ocean south of 60S. But the Arctic Ocean borders aren’t defined by one latitude. Does this skew the results? Since the NOAA NOMADS system requires latitudes and longitudes as inputs, I can’t customize the Arctic Ocean borders. But I can extend the latitudinal range to 60N-90N. Same results. The linear trend of the dataset without the polar data has the higher linear trend, though the gap is closing.
SST anomaly data is available through the NOAA NOMADS website:http://nomad1.ncep.noaa.gov/cgi-bin/pdisp_sst.sh