Justin McGinley
EF1
Ok from reading some posts it looks like the 20th or the 21st is going to get active. So i am wondering what do I need to look at from twister data for southeast nebraska? And also how do i read a gfs sounding map?
-
While Stormtrack has discontinued its hosting of SpotterNetwork support on the forums, keep in mind that support for SpotterNetwork issues is available by emailing [email protected].
Twister Data
- Thread starter Justin McGinley
- Start date
Dave Marshall
Nothing like drinking from a firehose eh?
The models are finicky, enigmatic creatures. A bit like women, only women are more predictable and usually make more sense.
It isn't really as simple as just looking at a couple maps and drawing a bullseye. Using the models is an ongoing process of changing and refining your forecast. It has more to do with timing, model cohesion, and run-to-run evolution than it does with what the current run says is going to happen. You also can get numerous bits of information from a single panel. For example, starting a week out or so, my usual method is reviewing the following panels, with what I'm looking for in paranthesis. I'm usually comparing the 4 most recent runs, and I'm looking at panels for 18 hours or so surrounding the potential event:
Surface Temperature (highs, gradient, frontal position & timing)
Wind - Surface to 300mb (jet stream position, wind direction, speed, convergence/divergence, etc)
Dew Point - All levels
Theta-e All levels
Bulk Shear all levels
Vorticity - All levels
Point soundings for potential chase areas
Being that all this is done with the GFS, making any specific calls on system intensity/timing is madness. This is just to give me an idea of what *potential* is there.
Once it's inside 84 hours, I start comparing the NAM and GFS to eachother, so I've roughly doubled the number of panels I'm reviewing. I also start including new panels:
Lifted Index
Helicity 1km & 3km
Precipitable Water
Once I'm inside 48 hours, I start looking at a couple others that are notoriously unreliable even inside that 48 hour window:
Cape
CIN
Simulated Composite Reflectivity
EHI
Storm Motion
Then, starting late the night before, I use the RUC in conjunction as well, so all three models, roughly the same panels, comparing and contrasting what they say. It isn't hard to learn, but it isn't something you just get a quick primer on and you're off and running. It takes some practice to learn where the weaknesses and strengths are for each model. It also isn't really a quick process to create a refined forecast. Each day closer to the event usually adds more time to what I spend looking at the models to draw up my plan. I'd say the night before, on the 00z runs, I probably spend close to 90 minutes reviewing everything and drawing a preliminary bullseye. What sucks is, when you wake up and the 12z RUC and NAM are out, your bullseye always changes and you get to do it all again.
As for reading soundings & hodographs, there are a dozen different websites that you can google for that will give you some great pointers. CoMet comes to mind, as do a few others. Personally, if I was just chasing solo and not trying to coordinate multiple people, I would never forecast beyond about 72 hours, and I'd rely mostly on the soundings. I'd say that particular skill is more important than reading the model panels themselves.
Every one here is going to say every one else is wrong in how they forecast. It's very much a subjective thing shaped by experience and intent.
The models are finicky, enigmatic creatures. A bit like women, only women are more predictable and usually make more sense.
It isn't really as simple as just looking at a couple maps and drawing a bullseye. Using the models is an ongoing process of changing and refining your forecast. It has more to do with timing, model cohesion, and run-to-run evolution than it does with what the current run says is going to happen. You also can get numerous bits of information from a single panel. For example, starting a week out or so, my usual method is reviewing the following panels, with what I'm looking for in paranthesis. I'm usually comparing the 4 most recent runs, and I'm looking at panels for 18 hours or so surrounding the potential event:
Surface Temperature (highs, gradient, frontal position & timing)
Wind - Surface to 300mb (jet stream position, wind direction, speed, convergence/divergence, etc)
Dew Point - All levels
Theta-e All levels
Bulk Shear all levels
Vorticity - All levels
Point soundings for potential chase areas
Being that all this is done with the GFS, making any specific calls on system intensity/timing is madness. This is just to give me an idea of what *potential* is there.
Once it's inside 84 hours, I start comparing the NAM and GFS to eachother, so I've roughly doubled the number of panels I'm reviewing. I also start including new panels:
Lifted Index
Helicity 1km & 3km
Precipitable Water
Once I'm inside 48 hours, I start looking at a couple others that are notoriously unreliable even inside that 48 hour window:
Cape
CIN
Simulated Composite Reflectivity
EHI
Storm Motion
Then, starting late the night before, I use the RUC in conjunction as well, so all three models, roughly the same panels, comparing and contrasting what they say. It isn't hard to learn, but it isn't something you just get a quick primer on and you're off and running. It takes some practice to learn where the weaknesses and strengths are for each model. It also isn't really a quick process to create a refined forecast. Each day closer to the event usually adds more time to what I spend looking at the models to draw up my plan. I'd say the night before, on the 00z runs, I probably spend close to 90 minutes reviewing everything and drawing a preliminary bullseye. What sucks is, when you wake up and the 12z RUC and NAM are out, your bullseye always changes and you get to do it all again.
As for reading soundings & hodographs, there are a dozen different websites that you can google for that will give you some great pointers. CoMet comes to mind, as do a few others. Personally, if I was just chasing solo and not trying to coordinate multiple people, I would never forecast beyond about 72 hours, and I'd rely mostly on the soundings. I'd say that particular skill is more important than reading the model panels themselves.
Every one here is going to say every one else is wrong in how they forecast. It's very much a subjective thing shaped by experience and intent.
Justin McGinley
EF1
Dave no offense but i have no idea what you were talking about. The only thing i got was the dew point that is because i know you need moisture for storms. Sorry about that dave I should of made it more clear in my first post.
Dave Marshall
No offense taken, everybody is a newbie at some point.
I'd highly recommend you hit this website http://www.meted.ucar.edu/ and spend some time going through a few modules. There really isn't any way to use the models on Twister Data without having a solid understanding of what each value is telling you.
In the list with the dark blue background, you should definitely go through the Convective Weather modules, and the Mesoscale Met modules. The NWP (Modeling) unit also has some great info, but a lot of it is more advanced stuff. I'm sure some other folks can provide other links to good resources for learning terminology and the like.
Honestly, I would say as a beginner your biggest sources of information should be the NWS field office forecast discussions and the SPC outlooks. I chased for 15 years before I started using the models. Even now I really mostly use them so I can coordinate days off for my team members and the like.
I'd highly recommend you hit this website http://www.meted.ucar.edu/ and spend some time going through a few modules. There really isn't any way to use the models on Twister Data without having a solid understanding of what each value is telling you.
In the list with the dark blue background, you should definitely go through the Convective Weather modules, and the Mesoscale Met modules. The NWP (Modeling) unit also has some great info, but a lot of it is more advanced stuff. I'm sure some other folks can provide other links to good resources for learning terminology and the like.
Honestly, I would say as a beginner your biggest sources of information should be the NWS field office forecast discussions and the SPC outlooks. I chased for 15 years before I started using the models. Even now I really mostly use them so I can coordinate days off for my team members and the like.
Jason Boggs
Justin, I promise I'm not trying to sound mean, but it sounds like you need to do a lot more reading and a lot less posting. Go to the educational section to learn more about storms and how to forecast.
Also, try to read up on the site weather prediction education
It's on my education site which is here-chaser education links
Again, I'm not trying to be rude with what I said, but take time to read and learn and you'll be well on your way.
Also, try to read up on the site weather prediction education
It's on my education site which is here-chaser education links
Again, I'm not trying to be rude with what I said, but take time to read and learn and you'll be well on your way.
Justin McGinley
EF1
Ok thanks for the links. Jason its alright i have been looking for places to read models anyway. And dave i do go by the forecast discussion which i read daily by the way and the spc outlooks.
Dave Marshall
For anybody who's getting started on models and doesn't understand why you never *ever* put confidence on a single run (or even 2 or 3 runs) of the GFS beyond a couple days out, compare the 18z 24April (next Saturday) forecast surface temperatures for Ohio and Pennsylvania on the last 6 runs (starting with yesterday's 06z run.
It started out very consistent, laying a strong cold front West to East across southern OH and PA at 18z Saturday. Then on the 18z model run yesterday, it picked up a new trend and brought the warm sector *way* north, into Ontario, and had a short wave moving through the area on a more SW/NE track Satruday night. Now, after about 3 runs of that warmer solution, it has moved back to almost exactly the same solution it started with. Forecast high temps for Saturday in the town I live in (Meadville, PA) have swung almost 25 degrees through 5 runs, and that's only 6 days out.
There was little agreement the last couple days between the ECMWF and GFS. The ECMWF has been bringing that warm front through the area fairly consistently, but the GFS had ignored that trend. When the GFS suddenly snapped into line with the ECMWF, the NWS CLE office bit on the new trend and changed the zone forecast this morning to reflect a high of 68F (raising their forecast high by almost 10 degrees) and a chance of thunderstorms in the area. I would expect if this return to the colder solution hangs on for the next 3-4 runs, they'll either sit on that forecast for a day or two and hope for better consensus, or they'll change the forecast back to the colder side.
It started out very consistent, laying a strong cold front West to East across southern OH and PA at 18z Saturday. Then on the 18z model run yesterday, it picked up a new trend and brought the warm sector *way* north, into Ontario, and had a short wave moving through the area on a more SW/NE track Satruday night. Now, after about 3 runs of that warmer solution, it has moved back to almost exactly the same solution it started with. Forecast high temps for Saturday in the town I live in (Meadville, PA) have swung almost 25 degrees through 5 runs, and that's only 6 days out.
There was little agreement the last couple days between the ECMWF and GFS. The ECMWF has been bringing that warm front through the area fairly consistently, but the GFS had ignored that trend. When the GFS suddenly snapped into line with the ECMWF, the NWS CLE office bit on the new trend and changed the zone forecast this morning to reflect a high of 68F (raising their forecast high by almost 10 degrees) and a chance of thunderstorms in the area. I would expect if this return to the colder solution hangs on for the next 3-4 runs, they'll either sit on that forecast for a day or two and hope for better consensus, or they'll change the forecast back to the colder side.
Last edited by a moderator:
