Jump to content
  • Member Statistics

    17,586
    Total Members
    7,904
    Most Online
    LopezElliana
    Newest Member
    LopezElliana
    Joined

Jan 23-24 Storm Banter


ORH_wxman

Recommended Posts

  • Replies 956
  • Created
  • Last Reply

You would think right?

But more likely that's an artifact of BOX agreeing with WPC on the south coast, while thinking that the northern edge will be farther north than WPC thinks.

So there is a larger difference around the CT/MA/RI border versus the south coast. When you add the difference to WPC 90th percentile, you increase totals near the CT/MA/RI border, and don't change them much or at all on the south coast. So it looks like taint, but it's just forecast agreement.

This is my problem with these probability maps, because in actuality the greatest threat for higher totals is the south coast, not near ORH.

Why do you guys have to agree with WPC? Why aren't you able to totally disagree with them if you feel it's necessary?
Link to comment
Share on other sites

Interesting. So isn't the problem really that they forced their forecast to agree with wpc but only for part of their forecast area?

Or reading it again, I'm not even sure I understand.

 

Not even that they were forced into it, but just any agreement with WPC makes it seem like potential snow is lower in those areas. The same goes for the min map. Sometimes you'll see low amounts or zero where you actually expect the highest snow to fall. It seems BOX edited in this case to make it zero everywhere.

Link to comment
Share on other sites

Why do you guys have to agree with WPC? Why aren't you able to totally disagree with them if you feel it's necessary?

 

It's not that we're forced into it. It's just that BOX and WPC happen to agree on amounts for say S CT. But BOX thinks (or thought) snow will fall farther north than WPC does.

 

So the difference grid at GON is 0, at IJD it might be 3. The way those maps are created are to add the difference grid to the 10th and 90th percentiles generated by WPC's model blend. Basically adjusting their forecast to the local WFO's.

 

So if the WPC 90th percentile was 6 at GON and 4 at IJD, the difference grid leaves GON alone and bumps IJD up to 7. Making it looking like IJD has more potential for snow that reality.

 

It's just bad logic.

Link to comment
Share on other sites

This is the part that doesn't make sense to me.

 

It's a little deep to get into gory detail, but basically based off their model blend there is a distribution (like a bell curve for instance) based off model spread. We don't have that locally, so we use WPC distribution. Our forecast locally essentially bumps that curve up or down to get the probabilities.

 

But you've identified the problem. It works well for WPC's forecast, but not so much for ours. Either we need to create our own distribution, or just use WPC's snowfall to make this work correctly. I am not a fan of doing the latter.

Link to comment
Share on other sites

It's a little deep to get into gory detail, but basically based off their model blend there is a distribution (like a bell curve for instance) based off model spread. We don't have that locally, so we use WPC distribution. Our forecast locally essentially bumps that curve up or down to get the probabilities.

But you've identified the problem. It works well for WPC's forecast, but not so much for ours. Either we need to create our own distribution, or just use WPC's snowfall to make this work correctly. I am not a fan of doing the latter.

Thanks, the way you explained issue is what I thought it was. It's combining apples and oranges and calling it a sandwich.
Link to comment
Share on other sites

It's a little deep to get into gory detail, but basically based off their model blend there is a distribution (like a bell curve for instance) based off model spread. We don't have that locally, so we use WPC distribution. Our forecast locally essentially bumps that curve up or down to get the probabilities.

 

But you've identified the problem. It works well for WPC's forecast, but not so much for ours. Either we need to create our own distribution, or just use WPC's snowfall to make this work correctly. I am not a fan of doing the latter.

 

Or just toss the whole thing. 

 

Some of these min/max ranges people are tweeting out are absurd. 10th/90th percentiles for Philadelphia is 2" and 20". How is that of any use to the end user? 

Link to comment
Share on other sites

Or just toss the whole thing. 

 

Some of these min/max ranges people are tweeting out are absurd. 10th/90th percentiles for Philadelphia is 2" and 20". How is that of any use to the end user? 

 

It's definitely not meant for public consumption right now. It was supposed to be a tool to provide EMs and state level decision makers to convey uncertainty. Unfortunately it's being executed poorly, AND they made us post the link on our homepage so anybody can look at them now.

Link to comment
Share on other sites

We've gone through the same routine for three nights now..... Guidance looks marginally better at 00z.... Then is worse the next day

Seems better then marginal now however, but it is the nam... See what the big dogs have to say.. Feels like we've been following this for a month I'm beat lol

Link to comment
Share on other sites

If you SNEers pull this out of your arse I'm deleting the subforum. :devilsmiley:

My dream is that this somehow splits the CC Canal and everyone south of Nashua mixes/dry slots. JK lol. Seriously though this has been one of the most fun storms to track here with the ups and downs and drama. Of course it has helped knowing that you're out of the picture for anything right from the get go. If you do have to end up deleting the subforum, create a NNE only subforum, we're much more laid back.
Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...