If you track back to the 2016 paper, Fig. 6, you find similar graphs (not as pretty, not interactive) with this explanation: "Frequency of occurrence (vertical

If you track back to the 2016 paper, Fig. 6, you find similar graphs (not as pretty, not interactive) with this explanation: "Frequency of occurrence (vertical axis) of local seasonal mean temperature anomalies in unit of local standard deviation of Northern Hemisphere land areas." It takes some time to unwrap the words there. The "anomaly" is the difference of an observation at a particular place and time from a seasonal and/or regional mean value. As you probably know, anomalies are pervasively used in geosciences. So in essence he's standardizing into what in statistics would be called a z-score, z = (x - mu) / sigma where mu is the regional/temporal average, and sigma is the local standard deviation. I guess the above is probably the reason the NYT graphic did not explain it fully.