According to the thermometers in the USHCN database, 1931 was the third hottest year on record in the US – after 1921 and 1934.
USHCN then adjusts 1931 temperatures downwards by a remarkable 2.51 degrees, before they publish it. The year 1931 had 708,317 valid thermometer readings, but somehow Tom Karl and James Hansen have recently determined that this huge body of data averaged 2.51 degrees too high. (This is the same James Hansen who can precisely calculate the temperature in the high Arctic, without a single thermometer reading north of 80N.)
So what else was going on in 1931?