Figure 1 is a graph of North and South Indian Ocean SST anomalies. From 1935 to 1941, the North Indian Ocean SST increased more than 0.5 deg C, then dropped back down to previous levels. And those are based on temperatures that have been smoothed with a 37-month filter. The North Indian is relatively small in size, so its reaction isn’t dampened by area. Still, the South Indian, which is not minimized by large land masses to the same extent as the North, rose more than 0.5 deg C from 1930 to 1941, only to drop 0.35 deg in the next six years.
Also note the oscillation in both signals. Check the timing of the troughs; they’re almost spaced perfectly every 10+ years. That’s not unusual in the Southern Hemisphere. I’ll have to do a post comparing this and other areas to solar cycles. It could simply be that noise from other sources hides the strength of the solar signal.
Some of the drop in temperature from the 1940s through 1960s will be reduced when the powers that be correct an earlier adjustment for the change in sampling method at that time. It won’t eliminate it all. Also note that there are other Southern Hemisphere ocean segments that have the same blip. I’ll point them out during discussions.
Sea Surface Temperature Data is Smith and Reynolds Extended Reconstructed SST (ERSST.v2) available through the NOAA National Operational Model Archive & Distribution System (NOMADS).