Terpeast

Just graded my 23-24 winter outlook:

Grading my 23-24 winter outlook vs actual results Temperatures: Decent match spatially, but off on the magnitude of the warmth. It's not just us being really warm, it's everywhere. Including Canada. Grade: B- Precipitation: Good match spatially and magnitude, a little bit off on the precip maxes though both midwest and east coast. I was highly confident of a wet winter, and that did happen. Grade: A- Snowfall: Truth is, I had at least one big storm baked into my snowfall outlook even through I cited only a 15-30% chance of a HECS. Well, that didn't happen obviously, so my snowfall guess was shot to hell. Grade: F Overall: I think my methodology and use of MEI led to an OK outlook, but was a little overconfident on blocking. We did get blocking, but each episode was either too brief, or occurred in the wrong places. Unfortunately, the MEI is no more, and I will have to either use RONI or a different method altogether for next year's outlook. Overall grade: C

Yeah, I’d have to compare my conus temp/precip maps in my outlook vs actual, should be a decent match. What went wrong was my snowfall outlook had baked in at least one big storm. Which didn’t happen.

Low of 27

Low of 29. No frost because its so dry

Smoke in ashburn. Where are the fires?

Early peak and decay may give us a modoki nina, worst case scenario for the east if you like cold and snow.

Repeat of last years canada wildfires?

Thanks, that gives me a clue to which analogs would make a good starting set and add others based on QBO and PDO (and of course, adjust each warmer by a couple degrees...)

Quite the turnaround. >60% nino to >60% nina within 3 months. Has such a transition happened that quickly before?

Still a warm signal. Even if only +2 based on older analogs (pre 2016 super nino), one would need to adjust to today’s climate. Like +4 or something like that.

That’s not wrong

Yeah, I was disappointed as well. Even though I busted way too high on my snowfall outlook, looking back at how I based my CONUS temp and precip maps on the MEI, I think it actually proved an useful tool that led to fairly successful verification at least with the distribution of temp/precip anomalies. Enough to make me think that I can use it again in future outlooks... then they take it away.

Wow, that is news. I guess I won't be using the MEI for future outlooks then. Might have to go with RONI or use my own custom dataset or something. Bring it on. I'm not expecting a winter next year imby, so might as well go ape with this nina and hopefully the strong trades push those cooler waters all the way into the maritime content / west pacific warm pool while we're at it.

Saw a few flurries yesterday afternoon and lots of wind this morning

Next winter the sfc temp anomaly may look pretty much like this map if the pacific doesn't change and we get a +qbo nina. Could be historically warm and snowless for the immediate east coast up to coastal SNE. At some point, we've gotta get a dip that breaks the climo trend. Don't know when that's gonna be.

It’s very likely too late for a lowland snow, but I’m interested in the AIFS test case, too.

And yet he understands more about the climate and its trends than most mets and atmos scientists do.

I’m not sure if it will be even as long as one paragraph!

10 days. Last 3 are warm, and that’s going to verify for sure. Maybe even warmer Crap, i was using metric (been overseas for too long) here’s the fahrenheit version lol

As a rule of thumb, if I don’t see cold/BN temperatures in all of the last 3 days within my apple weather app, we’re not seeing cold anytime soon. Following this ^^ rule would have been much more accurate than models.

We may hit 70 next week

Of course, I’ll be out of the country most of April. So it’ll be a fun month for everyone here.

The upside was much higher, yes… but in the back of my mind, my greatest fear was a warmer version of a 1972-73, which was the only match if you looked at the PDO and ENSO. There was always a non-zero chance for a ratter, and to be fair, most of our subforum did avoid a rat. We just fell short of climo and way short of the upside, where our expectations were.

Used a linear trend from the 60s to date, and then applied a warmer temp correction on the 2006 snowfall to today’s climate based on that trend. One one hand, 2006 wasn’t that long ago so the correction may not have been more than a degree or so. Otoh, linear trend by definition doesn’t account for nonlinearity nor any acceleration in warming. So like I said, I’d take the under on my own numbers. Could just have been 3” of slop instead of a foot+