Jump to content
  • Member Statistics

    17,502
    Total Members
    7,904
    Most Online
    Weathernoob335
    Newest Member
    Weathernoob335
    Joined

Upstate/Eastern New York


Recommended Posts

Oh, look at that. The temperatures over performed today. Shocker.

Models need to be programmed to take into consideration what seems to be a rapidly warming climate.

Snow is melting one day after it falls again...still a couple inches though. Thankfully and hopefully we get a good little thump tomorrow to make it a super wintry weekend.

Link to comment
Share on other sites

12 minutes ago, TugHillMatt said:

Oh, look at that. The temperatures over performed today. Shocker.

Models need to be programmed to take into consideration what seems to be a rapidly warming climate.

Snow is melting one day after it falls again...still a couple inches though. Thankfully and hopefully we get a good little thump tomorrow to make it a super wintry weekend.

That’s my biggest beef with the models. They don’t learn. Daytime high temps are always too low and snowfall is always too high. As soon as AI gets integrated into these things we’ll see rapid improvement. We’ll also be really disappointed with accurate snowfall maps- cut them in half or so. 
These models readily lend themselves to AI corrections. It would only take a season or two. Tons of data is available. In fact, they could run them historically and have corrections built-in within hours. 
I wonder if there’s a way to monetize that because it’s going to be drastic. Probably already occurring in the commodities hedge fund computers. Big money in oil futures based on temp forecasts. I’ve always thought this forum could beat that market. 

Link to comment
Share on other sites

45 minutes ago, rochesterdave said:

That’s my biggest beef with the models. They don’t learn. Daytime high temps are always too low and snowfall is always too high. As soon as AI gets integrated into these things we’ll see rapid improvement. We’ll also be really disappointed with accurate snowfall maps- cut them in half or so. 
These models readily lend themselves to AI corrections. It would only take a season or two. Tons of data is available. In fact, they could run them historically and have corrections built-in within hours. 
I wonder if there’s a way to monetize that because it’s going to be drastic. Probably already occurring in the commodities hedge fund computers. Big money in oil futures based on temp forecasts. I’ve always thought this forum could beat that market. 

AI?  As we all are aware (I think), wx models are deterministic algos based on known physical properties, rules, and physical-mathematical relationships/laws...they are not sentient or machine learning beings (Replicants!).  Pretty sure I would  not want some sort of "learning bias" interjected...as it would invariably be highly dependent on a set of predetermined instructions (temp has been higher/lower than model output for the last x of y runs...so let's adjust...). Not sure where you would be going with this other than creating Skynet for weather?

Having said that, if u do create Skynet I'd like to get in on the ground floor like the Winklevoss twins did with Bitcoin. ;)

Link to comment
Share on other sites

21 minutes ago, Thinksnow18 said:

Yeah I saw it. I didn’t think anything could be worse than the last 3 weeks but the next 2 look just as boring. If we don’t get any snowfall we probably won’t see much if any at the lower elevations into February. At least the ground will freeze.

GFS is in suppression mode. We've seen it before many times over the years.   If other models join it for the mon/tues system, I'll take the GFS more seriously. Until then...I wouldn't get ratchet over the prospects next 2 weeks.

  • Like 2
Link to comment
Share on other sites

2 hours ago, Syrmax said:

AI?  As we all are aware (I think), wx models are deterministic algos based on known physical properties, rules, and physical-mathematical relationships/laws...they are not sentient or machine learning beings (Replicants!).  Pretty sure I would  not want some sort of "learning bias" interjected...as it would invariably be highly dependent on a set of predetermined instructions (temp has been higher/lower than model output for the last x of y runs...so let's adjust...). Not sure where you would be going with this other than creating Skynet for weather?

Having said that, if u do create Skynet I'd like to get in on the ground floor like the Winklevoss twins did with Bitcoin. ;)

See! You get it. I like your equation. Why wouldn’t you want them to self correct? We see these biases and they go untouched- they simply persist. 
Don’t get me started on Bitcoin. I’ll have to kick it to the banter section. “The monetary value is based on mining an equation that grows tougher over time”:axe:

  • Haha 1
Link to comment
Share on other sites

2 hours ago, Syrmax said:

Here's the KBGM Clown for the event.  Matt and I are mocked again. Why can't we at least be in the yellow snow? ;)

StormTotalSnowWeb.png

Ha! Yep! I'm RIGHT on the 6 inch line. If I sit here watching the radar and the band runs just west of us the entire event.......................................................................

  • Haha 1
Link to comment
Share on other sites

1 hour ago, rochesterdave said:

See! You get it. I like your equation. Why wouldn’t you want them to self correct? We see these biases and they go untouched- they simply persist. 
Don’t get me started on Bitcoin. I’ll have to kick it to the banter section. “The monetary value is based on mining an equation that grows tougher over time”:axe:

BTC.X ... It's an interesting banter topic...do we trust math more than Uncle Sam's full faith and credit...?  ;)

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...