What models do you look at...

[Dustin Janise]

When you are making a Forecast, what Models do you look at, and what are you looking for?

Also, does the the 3 or 6 hour Precipitation Models play any role in your Forecast?

I myself look at the F5Data GFS Models: SBCAPE, SBCIN, Lifted Index, Temp, and Dew Point. What other models can I look at to be a better forecaster?

 

[rdale]

I think you're tackling this from the wrong perspective... The goal isn't to figure out what model has the best dew point forecast, but what the dew point is and how that relates to severe weather. Learn how to analyze 700mb temps and find the cap breaking point. Read 850mb winds. Once you have that, then worry about the other fields. Knowing that CAPE = 3000 and LI = -5 isn't "making a forecast" but just "reading the grid." There's no value in that.

 

[Dustin Janise]

Learn how to analyze 700mb temps and find the cap breaking point. Read 850mb winds. Once you have that, then worry about the other fields.

This is not my strong suit, so how do you recommend I Learn about the more "crucial" info to better forecast severe weather?

 

[Jason Foster]

and don't necessary believe the models either...or just bullseyes on those products.

But I do like to use the Deep Layer Convergence a bit. One product I avoid, regardless of whether it's GFS, NAM, RUC is the simulated radar things (like the one in twisterdata). Sure...it's kinda guidance, but to me it about as bad as a WxWeenie saying where to go chase.

 

[Greg Blumberg]

Dustin,

The models are mainly guidance used to compliment your forecast. They're like several additional forecasters you have at your disposal, but like other humans they have flaws that you need to be aware of. They're useful if you either reject or accept their reasoning based on your knowledge of the atmosphere. Reading the models without thinking critically about what they're saying doesn't help you understand more about forecasting. Today's bust in Oklahoma was a perfect example of the importance of not following the models blindly.

Most forecasting follows something called the forecast funnel, which is a method of forecasting used by many. To follow this method, you start at analyzing the hemispheric features (like longwaves and blocking patterns), then synoptic features (things like highs, lows, troughs, ridges, and fronts), down to the mesoscale (things like storms.) It's a way of assessing meteorological phenomenon by scale and helps you gain a full understanding of how the atmosphere is structured. After knowing the atmospheric situation now, you can start to look at how atmospheric features will change into the future. This is often how I attack forecasting problems. I identify the features and then think about the Problem of the Day (POTD). The POTD could range between what type of storms will form (this was a problem today in Oklahoma), when the storms will form, how much rain will a place get, etc. In most situations there could be more than one problem.

After identifying the POTD, a forecaster will use his own experience, knowledge of the atmosphere's 'rules', and the forecast models to try to solve that problem.

There are several things you can do to improve your forecasting. The first (and sometimes not most obvious) is practice. Achieving a feat like forecasting the weather isn't easy and you have to build experience. It takes being wrong and learning from your mistakes. Something else you can do (and are already doing!) is ask questions. Also, it's good to read things about it when you can, but be sure to try to apply what you've read. Be patient (this is something I had trouble with)! It takes time.

I highly recommend http://www.theweatherprediction.com/ for beginners. I also started reading forecast discussions to become familiar with the language and terms. If a forecaster identified something on a map, I'd try to pick it out too. Tim Vasquez's books are also helpful too: http://weathergraphics.com. Also, the University of Illinois has a great website that can get you started: http://ww2010.atmos.uiuc.edu/(Gh)/guides/home.rxml

Hope this helps and good luck!

 

[Dustin Janise]

Dustin,

The models are mainly guidance used to compliment your forecast. They're like several additional forecasters you have at your disposal, but like other humans they have flaws that you need to be aware of. They're useful if you either reject or accept their reasoning based on your knowledge of the atmosphere. Reading the models without thinking critically about what they're saying doesn't help you understand more about forecasting. Today's bust in Oklahoma was a perfect example of the importance of not following the models blindly.

Most forecasting follows something called the forecast funnel, which is a method of forecasting used by many. To follow this method, you start at analyzing the hemispheric features (like longwaves and blocking patterns), then synoptic features (things like highs, lows, troughs, ridges, and fronts), down to the mesoscale (things like storms.) It's a way of assessing meteorological phenomenon by scale and helps you gain a full understanding of how the atmosphere is structured. After knowing the atmospheric situation now, you can start to look at how atmospheric features will change into the future. This is often how I attack forecasting problems. I identify the features and then think about the Problem of the Day (POTD). The POTD could range between what type of storms will form (this was a problem today in Oklahoma), when the storms will form, how much rain will a place get, etc. In most situations there could be more than one problem.

After identifying the POTD, a forecaster will use his own experience, knowledge of the atmosphere's 'rules', and the forecast models to try to solve that problem.

There are several things you can do to improve your forecasting. The first (and sometimes not most obvious) is practice. Achieving a feat like forecasting the weather isn't easy and you have to build experience. It takes being wrong and learning from your mistakes. Something else you can do (and are already doing!) is ask questions. Also, it's good to read things about it when you can, but be sure to try to apply what you've read. Be patient (this is something I had trouble with)! It takes time.

I highly recommend http://www.theweatherprediction.com/ for beginners. I also started reading forecast discussions to become familiar with the language and terms. If a forecaster identified something on a map, I'd try to pick it out too. Tim Vasquez's books are also helpful too: http://weathergraphics.com. Also, the University of Illinois has a great website that can get you started: http://ww2010.atmos.uiuc.edu/(Gh)/guides/home.rxml

Hope this helps and good luck!

Thanks for the reply!