Bonus plot for Bluesky folks: Seems like AO is part of the story for this season. The teal dot is Mar 2025 (w/ prelim SPC report data), modestly AO+, which can be a subseasonal signal for increased tornado activity, i.e., top-left quadrant La Niña/AO+ does feature a lot of big tornado months...

How QBO affects weather/climate patterns can be confusing, but @jorgelgf.bsky.social tries to break down the main QBO teleconnection routes in this post. As usual, we had fun with graphics... here's an animated schematic:

Aaand an example of how the GEFS index looks for Spring 2013. Chiclet plots can be fun/pretty. ✨

GEFS has some skill into week 2 at times, particularly for CONUS-wide tornado outbreak activity, and particularly for winter and spring. Forecast skill in U.S. outbreak activity is most limited by forecast skill in conv. precip.

An outbreak index based on conv precip, SRH, and CAPE has seasonal cycle biases, which we can correct via a post-calibration scaling. This can extend skill a couple days in some cases!

GEFS has a systematic low-CAPE bias, which we are not sure has been documented before, especially in the context of how that might affect convective storm representation/skill.

New SC graphic just dropped. 😈

What would you want to know if you had 2000+ yrs of daily severe convective storm data? 🤔 I'm giving a talk at AGU in the Climate-Informed Risk Assessment for Extreme Events session on Tuesday at 3pm. Basically @mktippett.bsky.social and I are building a synthetic event set for tornado outbreaks 🌪

Hi weather and climate community - I am recruiting a PhD student for Fall 2025! Please see flyer and share widely 🙂

...and detected upward trends in winter-spring tornado outbreak activity during 1979-2021 period.

We can also more robustly assess climate signals in tornado outbreak activity. With both smoothed tornado reports and the index, we found that La Niña is linked to increased tornado outbreak activity in late winter/early spring in Tennessee River and Ohio Valley regions...

The stochastic aspect of the index means we can simulate a range of potential outcomes from a single large-scale environment, ideal for estimating risk from such sporadic/rare events.

Given a U.S. map of environmental predictors of convective precip, SRH, and CAPE, the index provides (1) a U.S. map of outbreak-level occurrence probabilities and (2) a probability distribution for total U.S. outbreak tornadoes (case example shown in fig).