Joshua Pringle

"That intelligent, well-meaning, young people are willing to call white black is a matter of concern." -Solomon Asch The striking thing about this work on group conformity is that 75% gave at least one incorrect answer, even when they later admitted that they knew it was wrong.

[Saturday, Feb 21 update] After poring over today's data, I made slightly more detailed updates to my snow map. The main uncertainties are: - how far east/west the snow gradient sets up in NJ & east PA - where to draw a >24" contour Forecast reasoning follows in this thread:

[Friday, Feb 20 update 🧵] I decided to give it a shot and draw an early first guess for a snow map. Keep in mind forecast uncertainty is still large, as highlighted on the map, and significant changes are still possible. Forecast reasoning in the thread below:

This is why meteorologists sometimes have a such a hard time forecasting because if the two leading models are near opposites to each other for a storm that is expected to hit in a mere 72 hours, what route/solution do you choose? The snow or no snow route?

[Thursday, Feb 19 update 🧵] This chain of quoted threads over the last few days below is an example of applying synoptic meteorology & probabilistic forecasting to avoid locking in an outcome most models showed & ruling out an outcome few/no models showed. For most of today, the GFS/GEFS suite was on an island of its own with a major Mid Atlantic/coastal snowstorm, with other models showing a much farther offshore and weaker storm. We currently appear to be in the midst of a correction amongst the rest of the guidance towards a scenario closer to the GFS. I'll explain below what changed over the last day, why I still suspect the GFS may be a bit too far west/amplified, and what the remaining uncertainties are.