I came up w/ list of subseasonal to seasonal (S2S) analogs for this year's tornado season in late April into early May & this is what the 1°x1° tornado track density composite grid looks like over the CONUS using
@NWSSPC's database that goes back to 1950.

Note like my plot from earlier this year, I use a 30-year sliding base period anomaly here relative to April 20 - May 5th in the year to account for long-term non-stationary behavior in tornado data (due to better detection & increased spotting in recent years).

These analogs hint at a fairly classic late April into early May this year with potential for above average tornado activity over climatologically favored areas of the Great Plains, including in/around Oklahoma.

The mean anomaly is plotted on the left & the mean of the analogs is on the right.

🌪️
I used a combination of many things in this forecast:

Years from the seasonal forecast I showed earlier this year were given a little weight, as well as global SSTa, West-Central Gulf SSTa, Global 200mb zonal wind anomalies, tropical OLR anomalies, forecast CONUS precipitation anomalies, forecast & observed 500mb pattern in/around N America, SLP over N America, & 200mb velocity potential anomalies (to detect subseasonal variability from things like Kelvin Waves & the MJO).

I came up w/ list of analogs for this year's tornado season & this is what the 1°x1° tornado track density composite grid looks like over the CONUS using @NWSSPC's database that goes back to 1950.

Note I used a 30-year sliding base period anomaly here to account for long-term non-stationary behavior in tornado data (due to better detection & increased spotting in recent years).

Although tornadoes will still certainly occur over the Great Plains, this analog composite is highlighting an *anomalous* eastward shift in tornado activity this year, with the epicenter of anomalously positive tornado activity closer to the mid-Mississippi Valley.

🌪️ What really catches my eye w/ this anomaly composite is how closely it resembles the long-term trends in favorable tornado environments (left figure from @gensiniwx & Brooks (2018)).

This suggests that this year may reinforce some of the long-term observed trends in tornado activity over the CONUS.

While there are many reasons this winter is starting out so cold in the Eastern US this year…

Imo, the -EPO/+TNH pattern in early Dec is largely being driven by low frequency variability (not the MJO) from the Indo-Pacific Warm Pool.

Namely, the Indo-Pacific Warm Pool is rectifying long-term observed changes since the early-mid 20th century that conflict with most climate model forecasts: the Indian & West Pacific oceans warming at a faster rate than the East-Central Pacific (in part as a response to AGW), which changes the downstream waveguide over the N Pacific & N America to resemble the +TNH pattern.
For example, the North Indian Ocean used to be cooler during the La Nina phase of ENSO & this would help maintain a canonical ENSO -PNA pattern downstream. Warming the Indian Ocean (like this year) adds enough +U to the North Pacific Jet thru convective feedbacks to cause the anticyclonic wave breaking to occur closer to the Gulf of Alaska & US West Coast instead of the Aleutians.

This sort of thing is happening more frequently than it used to due to long-term Indo-Pacific warming trends that are outpacing much of the equatorial Pacific.

In trying to figure out what’s driving the strong -EPO/+TNH (Alaska ridge - Hudson Bay trough) pattern over the next few weeks & if it means anything for this coming winter…

…I ended up going down a deep literature rabbit hole 🐰 🕳️ & discovered the ABNA (Asia - Bering Sea - North America) teleconnection pattern.

Attached here are some relevant literature pieces and the negative phase of the ABNA pattern, which features a trough over East Asia, ridge in Alaska, & trough in Eastern North America.

iopscience.iop.org/article/10.108…

journals.ametsoc.org/view/journals/…

journals.ametsoc.org/view/journals/…

journals.ametsoc.org/view/journals/… What I found pretty interesting here in one of these pieces of literature was that Maritime Continent SSTa in Nov seemed to be a good predictor for the development of the negative ABNA pattern in winter, which delivers anomalously cold air into N America.

The DJF ABNA index mean regression onto the preceding Nov SSTa (left) certainly looks a lot like this November so far (right)

journals.ametsoc.org/view/journals/…

My overall take on Hurricane #Helene is:

While the forecasts from the NHC, WPC, & local NWS WFOs certainly couldn't have been much better overall, how that forecast was presented, communicated, and then acted upon by both the public & decision-makers was far from sufficient to put it lightly.

This on top of the other socioeconomic issues (w/ the hardest hit areas also being generally isolated & poor) has greatly exacerbated the disaster that's ongoing in/around western NC... (1/n) It's easy to point fingers & blame local, state, federal governments, etc. for an apparent lack of or a slow response, but it's a lot harder to look in the mirror & ask yourself what WE can do as meteorologists to better to communicate this information & get these folks to act more accordingly next time... (2/n)

An often overlooked element in seasonal hurricane forecasts is how the tropical mid-upper troposphere will respond to an increasingly warmer background climate. The influence of climate change in this hurricane season's abnormally quiet start thus far through late August (relative to expectations) is very likely non-negligible in this case.

While sea surface temperatures have been record warm lately, the upper troposphere has been warming at an even faster rate (!), as shown in the quoted tweet below.

This long-term, increasingly stable trend in the tropical troposphere is also consistent w/ what's expected from climate change.

The increasing stabilization of the tropical troposphere is due to moist adiabatic adjustment in the tropics. Moist adiabatic adjustment causes the deep-layer temperature profile to stabilize in the tropics, w/ the upper troposphere warming faster than the lower-mid troposphere. The combination of more marginal lapse rates & greater saturation vapor pressure deficits also makes dry/stable air intrusions more effective at inhibiting tropical convection in a warmer mean climate (see attached figure).

Simply put, there's a lot more to seasonal hurricane forecasting than record high SST anomalies.

For more, see:







journals.ametsoc.org/view/journals/…
agupubs.onlinelibrary.wiley.com/doi/full/10.10…
nature.com/articles/s4146…
nature.com/articles/ngeo1… There's been a lot of discussion (myself included) about how far north this year's easterly waves have been lately & how this may be inhibiting TC development.

In that context, it's hard not to notice how this summer has been the hottest one on record for much of the Sahara & how it's hydrostatically coupled w/ record low SLP anomalies in NE Africa.

These lower SLP anomalies are indicative of a stronger Saharan Heat Low, which is helping to draw the low-level ITCZ further north, esp over east-central Africa. The end result is what's likely going to be the wettest summer on record for much of northern Chad, southern Libya, & southern Egypt via ERA-5.

Hurricane #Ian is starting to undergo extratropical transition tonight on its way to the Carolinas. Notice the big dry slot sneaking in from the SW and frontal boundary extending from #Ian's south into the Bahamas... Fwiw, one of the simplest methods I use to discern b/t tropical & extratropical cyclones is to locate the subtropical jet & assume everything poleward/north of that is extratropical, while to the south = tropical. #Ian is in the middle rn => "hybrid" warm core/subtropical cyclone

In the coming days, you'll see many Velocity Potential & Madden Julian Oscillation (MJO) plots on #wxtwitter.

Here, I show what they often entail for Atlantic tropical cyclone (TC) activity.

Notice the Atlantic is most active (🔴) when the MJO (🟢) reaches Indonesia (phase 4) Upper left: (5x5°) daily IBTrACS TC track density anomaly vs 40-yr sliding base pd w/ 2 wk-centered climo (+/-7 days)

Lower left: (1x1°) ERA-5 daily 200mb Velocity Potential Anomalies (VP200a) (1979-2020 climo)

Right: MJO phase diagram w/ shading corresponding to spatial VP200a