First Legit Storm Potential of the Season Upon Us

[40/70 Benchmark]

2 minutes ago, ORH_wxman said:

Hopefully they can get another 6”+ event like last year. 

I honestly wonder if Mother Nature is John Henry....the timeline of the evolution fits perfectly.

[WeatherGeek2025]

2 minutes ago, 40/70 Benchmark said:

Just teasing, nothing personal

ok brother thanks i'm kind of a sensitive sally

[WxWatcher007]

14 minutes ago, MJO812 said:

Euro isnt even in the same ballpark as the other models so far. It doesnt even show a storm. Very weak and way OTS.

I don’t like definitive statements generally but this one almost certainly looks gone. It’s not that the ops “lost”, but the ensembles have moved en masse in the wrong direction as we close in and we’re relying on unproven AI to pull an epic short(ish) range coup. 

[40/70 Benchmark]

Just now, WxWatcher007 said:

I don’t like definitive statements generally but this one almost certainly looks gone. It’s not that the ops “lost”, but the ensembles have moved en masse in the wrong direction as we close in and we’re relying on unproven AI to pull an epic short(ish) range coup. 

Don't you have a LES band to chase, Don?

[Go Kart Mozart]

8 minutes ago, WeatherGeek2025 said:

okay be nice i didn't say anything disrespectful i was just giving my opinion

I have never met Ray (40/70), nor have I ever heard him speak.  But in my mind, he speaks with the Seth MacFarlane voice of Ted.  If Ted were a friend of yours, he would insult you but you would just laugh him off.  Or you might rip the stuffing out him...who knows.

[dryslot]

She gone.

[ORH_wxman]

8 minutes ago, 40/70 Benchmark said:

I honestly wonder if Mother Nature is John Henry....the timeline of the evolution fits perfectly.

Lots of “interest” these days but can never bring home the bacon. Whether it’s the models or FSG, same idea. 

[WxWatcher007]

Just now, 40/70 Benchmark said:

Don't you have a LES band to chase, Don?

Haha I’m in CT through early next week so I’m angry over the brown grass outside too.

I won’t belabor it anymore but I was bullish on the snowfalls we had in December and still think we can produce nicely in the last 1/3 of January so I’m not turning into TBlizz, but I’ve never liked the thread the needle nature of this setup. I was hopeful a compromise or moderate event was on the table but we couldn’t even keep that trend for a full day. 

[MegaMike]

In my opinion, there's too much trust in AI for weather prediction. I've mentioned this a few times, but

1. It was made operational recently
2. There's nothing wrong with current NWP excluding (a) it takes longer to run and (b) it requires a lot of resources vs. AI
3. There is no significant evidence the EC-AIFS/AIGFS outperforms NWP for sensible weather at the surface during inclement weather (please provide a source if I'm wrong).
4. For AI, nobody knows how forcing(x,y,z,t) is calculated (doesn't rely on traditional methods). ie... what is 1+1? Human = 1 + 1 == 2 ||| AI = :performs multi-dimensional math on 'n' fields: == 2. Do you trust that?
5. Given the initial state of the atmosphere is captured flawlessly, there's no guarantee AI will perform well.

AI is great when there is no known relationship/correlation between a predictor and many predictands. Weather is relatively predictable so I don't find AI useful unless the fields are bias-corrected then ingested back into data assimilation grids.

If the AIs outperform NWP for this, *** and it evaluates well ***, I'll take it a little more seriously.

Who knows... Maybe truncating/rendering certain fields may increase its accuracy for this one event <AND/OR> data assimilation is poor at the current location(s) where the disturbance(s) is/are, and AI could use historic events to predict this event with some level of accuracy. 

[40/70 Benchmark]

3 minutes ago, ORH_wxman said:

Lots of “interest” these days but can never bring home the bacon. Whether it’s the models or FSG, same idea. 

It started in 2018.....before that, it was two decades of blizzards and marquee FA signings being dished out like Halloween candy...now, every vort either cuts to the lakes, or is sheared until it reaches the Maritimes, while every large FA target sublimates into a broke-dick high AAV-short-term signing. All they do is sign clippers....even Suarez kind of a broke-dick/clipper on roids.

[WxWatcher007]

1 minute ago, 40/70 Benchmark said:

It started in 2018.....before that, it was two decades of blizzards and marquee FA signings being dished out like Halloween candy...now, every vort either cuts to the lakes, or is sheared until it reaches the Maritimes, while every large FA target sublimates into a broke-dick high AAV-short-term signing. All they do is sign clippers.

I do think that changes next week…the signal is there as we’ve all said for a good overrunning event…maybe late next week.

[40/70 Benchmark]

4 minutes ago, WxWatcher007 said:

I do think that changes next week…the signal is there as we’ve all said for a good overrunning event…maybe late next week.

Hopefully it's a higher upside deal like PD II, and not some lame, Jan-Feb 2014 rendition. 

[ORH_wxman]

EPS actually looked slightly better vs 00z…that isn’t saying much though. But it gets 1-2” to south shore area and Cape. 

[weatherwiz]

8 minutes ago, MegaMike said:

In my opinion, there's too much trust in AI for weather prediction. I've mentioned this a few times, but

1. It was made operational recently
2. There's nothing wrong with current NWP excluding (a) it takes longer to run and (b) it requires a lot of resources vs. AI
3. There is no significant evidence the EC-AIFS/AIGFS outperforms NWP for sensible weather at the surface during inclement weather (please provide a source if I'm wrong).
4. For AI, nobody knows how forcing(x,y,z,t) is calculated (doesn't rely on traditional methods). ie... what is 1+1? Human = 1 + 1 == 2 ||| AI = :performs multi-dimensional math on 'n' fields: == 2. Do you trust that?
5. Given the initial state of the atmosphere is captured flawlessly, there's no guarantee AI will perform well.

AI is great when there is no known relationship/correlation between a predictor and many predictands. Weather is relatively predictable so I don't find AI useful unless the fields are bias-corrected then ingested back into data assimilation grids.

If the AIs outperform NWP for this, *** and it evaluates well ***, I'll take it a little more seriously.

Who knows... Maybe truncating/rendering certain fields may increase its accuracy for this one event <AND/OR> data assimilation is poor at the current location(s) where the disturbance(s) is/are, and AI could use historic events to predict this event with some level of accuracy. 

This should be pinned at the top of the board  

As I've also mentioned before, AI can probably be very useful in the nowcasting department or short-term (6-12 hours) but beyond that...a very long ways to go

[40/70 Benchmark]

1 hour ago, ORH_wxman said:

EPS actually looked slightly better vs 00z…that isn’t saying much though. But it gets 1-2” to south shore area and Cape. 

Sweet, lets bury Jordan's box on Belicheck's beach side bungalow in ACK, and keep bare ground in Methuen.

[CoastalWx]

Even the euro AI ensembles got more bullish. This is ridiculous lol. Gotta think they crumble soon.

[CoastalWx]

16 minutes ago, MegaMike said:

In my opinion, there's too much trust in AI for weather prediction. I've mentioned this a few times, but

1. It was made operational recently
2. There's nothing wrong with current NWP excluding (a) it takes longer to run and (b) it requires a lot of resources vs. AI
3. There is no significant evidence the EC-AIFS/AIGFS outperforms NWP for sensible weather at the surface during inclement weather (please provide a source if I'm wrong).
4. For AI, nobody knows how forcing(x,y,z,t) is calculated (doesn't rely on traditional methods). ie... what is 1+1? Human = 1 + 1 == 2 ||| AI = :performs multi-dimensional math on 'n' fields: == 2. Do you trust that?
5. Given the initial state of the atmosphere is captured flawlessly, there's no guarantee AI will perform well.

AI is great when there is no known relationship/correlation between a predictor and many predictands. Weather is relatively predictable so I don't find AI useful unless the fields are bias-corrected then ingested back into data assimilation grids.

If the AIs outperform NWP for this, *** and it evaluates well ***, I'll take it a little more seriously.

Who knows... Maybe truncating/rendering certain fields may increase its accuracy for this one event <AND/OR> data assimilation is poor at the current location(s) where the disturbance(s) is/are, and AI could use historic events to predict this event with some level of accuracy. 

The only thing I heard is that the euro AI ensemble seems to have the best scores in terms of 500 MB. But you and I both know that doesn’t necessarily translate to better and more accurate forecast for sensible weather.

[Sey-Mour Snow]

Just now, CoastalWx said:

The only thing I heard is that the euro AI ensemble seems to have the best scores in terms of 500 MB. But you and I both know that doesn’t necessarily translate to better and more accurate forecast for sensible weather.

I mean the EURO op and EURO AIFS look nothing alike for Sunday at 500….

[Layman]

3 minutes ago, weatherwiz said:

This should be pinned at the top of the board  

As I've also mentioned before, AI can probably be very useful in the nowcasting department or short-term (6-12 hours) but beyond that...a very long ways to go

Since this storm is gone...I don't necessarily feel like I'm derailing the thread by asking:

Do you have experience or knowledge about how the AI models are attempting to assimilate data and/or generate output?  My very limited understanding is that they're supposed to "learn" but what are they learning from?  Historical data?  Current data?  Compared to other model outputs?  Or, comparing after the fact, what actually happened?

Curious about all this but not even sure where to look for a quality resource that may explain some of these things.

[TauntonBlizzard2013]

7 minutes ago, ORH_wxman said:

EPS actually looked slightly better vs 00z…that isn’t saying much though. But it gets 1-2” to south shore area and Cape. 

I totally expect this type of outcome. It’s going to thread the needle and become the exact thing people don’t want. A middling 1.5” that has grass showing and is just enough to make the roads awful, but not really do anything else.

This will be justttttttt close enough to piss people off.

[Go Kart Mozart]

5 minutes ago, weatherwiz said:

This should be pinned at the top of the board  

As I've also mentioned before, AI can probably be very useful in the nowcasting department or short-term (6-12 hours) but beyond that...a very long ways to go

There are many online articles and verification charts that show just the opposite, with AI outperforming beyond 4 days...at least at h500.  It does have lots of flaws, such as black swan events with minimal historical corollaries.  It will probably not be good for something like December 92, for instance.

[SouthCoastMA]

Just now, TauntonBlizzard2013 said:

I totally expect this type of outcome. It’s going to thread the needle and become the exact thing people don’t want. A middling 1.5” that has grass showing and is just enough to make the roads awful, but not really do anything else.

This will be justttttttt close enough to piss people off.

Lets just get it close enough for some flakes during gametime

[CoastalWx]

2 minutes ago, Sey-Mour Snow said:

I mean the EURO op and EURO AIFS look nothing alike for Sunday at 500….

I hope I’m wrong, but I feel like we toss the AI stuff.

[Chrisrotary12]

Saturday > Sunday at this point. 

[Sey-Mour Snow]

6 minutes ago, CoastalWx said:

I hope I’m wrong, but I feel like we toss the AI stuff.

I am not taking any sides , I give up, just kind of wait and see approach 

[CoastalWx]

Just now, Sey-Mour Snow said:

I am not taking any sides , I give up, just kind of wait and see approach 

Saturday is white rain here so I need Sunday night. But I’m sure the weed whacker awaits instead. Hopefully 2-3 for you there. 

[Sey-Mour Snow]

Fact of the matter is. This period was never highlighted as snowy on ensembles, long wave pattern was there, but we knew it was a long shot, and as unscientific as it is, the ensembles were never snowy. This weekend is gravy, the money pattern is Jan 22-Feb 2nd. Could be an epic stretch.. Ensembles are honking… yes we have heard this before, but we should have 2 shots at big qpf producers over a cold dome. 

[weatherwiz]

2 minutes ago, Layman said:

Since this storm is gone...I don't necessarily feel like I'm derailing the thread by asking:

Do you have experience or knowledge about how the AI models are attempting to assimilate data and/or generate output?  My very limited understanding is that they're supposed to "learn" but what are they learning from?  Historical data?  Current data?  Compared to other model outputs?  Or, comparing after the fact, what actually happened?

Curious about all this but not even sure where to look for a quality resource that may explain some of these things.

As stated the idea of they're supposed to "learn" is totally overblown. Traditional models already have some AI built into them and already do this to an extent. From what I understand (and this may not apply equally to AI models) is

AI assists with the initializing scheme whereas it combs through ingested data and will "remove" what it believes to be bad data or an outlier based on a slew of historical information. The goal here is, or the idea is, this will lead to a more accurate initialization which is important because once you move forward in time you start to introduce error and that error becomes compounded over time...that is why forecast models (OP) are generally useless beyond D7-10 and can even be relatively useless past D5 if there is alot going on. Error also occurs because of rounding and approximations, especially approximations. 

AI models are built on a wealth of historical data where it runs and looks for similarities, both to the initialized field and then forecasts based on how these similarities evolved in the past. 

The challenges in all of this is, there is still a lot we don't understand about weather, particularly when it comes down to processes which occur during storm evolution and it becomes even more of a challenge because for forecast models to ingest this data we have to be able to parameterize it. 

Just now, Go Kart Mozart said:

There are many online articles and verification charts that show just the opposite, with AI outperforming beyond 4 days...at least at h500.  It does have lots of flaws, such as black swan events with minimal historical corollaries.  It will probably not be good for something like December 92, for instance.

There is much more to this then just verifying a specific level or variable and even that leads to a lot of questions. Probably in a tame weather pattern that is not hostile, AI will outperform but what good is that or what value is that really adding? 

[WinterWolf]

27 minutes ago, CoastalWx said:

Saturday is white rain here so I need Sunday night. But I’m sure the weed whacker awaits instead. Hopefully 2-3 for you there. 

We’ll take it and add to the season total.   

[WinterWolf]

1 hour ago, Supernovice said:

Not trying to break balls but this is a fundamental misunderstanding of ai as it pertains to weather models. 

Oh ok. Was told just the opposite. But whatever.  
 

Then the AI should fold soon.