Severe Weather Forecasting Secrets: Pt. 1: The Maps

Want to be the "backyard weather forecaster" who warns your neighbors when storms are coming to town, a day or more in advance? You won't hear this secret from a lot of meteorologists, but it's not as hard as it looks -- with a few internet resources and a few moments of time, you, the blog reader, can be reasonably accurate forecasting thunderstorms in your state for the next several days out.* You may even be able to pinpoint the time of arrival of storms and project the regions of your state that will get them. Amaze your friends and family! Here's how.

The first thing you should do is check AccuWeather.com Severe Weather Expert Henry Margusity's blog for a forecast map (and watch his video for more info). His maps are really simple to read, usually for the next one to two days, and typically issued an hour either side of 9 a.m., but because he is a blogger there is no guaranteed schedule. Here's his map for today:

His maps get very specific as to the type of lightning, hail, wind and tornadoes projected with the storms. It's so simple, I don't even need to explain it, unlike the next resource which you can use as a second opinion, and which looks further out:

The SPC Maps. NOAA's Storm Prediction Center has dozens of severe weather meteorologists drawing "Convective Outlook" maps showing the chance of severe weather over the United States for the next eight days, issued several times per day at set times. Because they are a little hard for the layperson to understand, I've broken down a tutorial for them into several sections.

Colors Show You Risks: The color scheme is very simple for Days 1-3:

General Risk / See Text: There will be a chance of thunderstorms.Slight Risk: There will be a chance of severe thunderstorms.Moderate/High Risk: There will probably be severe thunderstorms, likely with large hail and possible tornadoes.

After "Day 3" they have a different color scheme, which you can see below the four- to eight-day map - essentially they use one color for each day to outline the "at least General, maybe more" area.

Watch Your Areas: All you need to do is look to see whether or not your town is in one of their risk areas. This can get tricky because they don't color in the risk areas - the area is "to the right of the arrowhead when standing behind it." See below for an example of today's risk - actual at left, what you should visualize at right. This will just take some practice, you'll get used to it.

Each risk must be contained within another -- you can't have a Moderate outside of a Slight, for example. These lines aren't exact - the weather never is, so even if you're close to a risk area, be prepared for what it could contain. And the SPC will tell you that these forecasts may not cover rogue thunderstorms or weak tornadoes; there have been plenty of small tornadoes in General and Slight risks, for example.

Keep Checking Back: Do check back multiple times per day - these maps are issued at 12Z - 7 or 8 a.m. Eastern Time), again at 13Z and once more at 16Z (lunchtime). For whatever reason, the SPC is reticent to issue a Moderate or High Risk beyond Day 1, often they upgrade part of a Slight to a Moderate the morning of. If they have issued anything besides a Slight on Day 2 or 3 - look out!

Watch Your Times:Make sure you watch the timing and don't assume this is an "afternoon" forecast - the "days" for SPC forecasts are split in the early morning (12Z - 7 or 8 a.m. Eastern). Today's a great example -- there is a Slight Risk for "today" in northern Pennsylvania and one for "tomorrow" in southern Pennsylvania. What does this really mean?

When you look at the WRF model (see Part 2 tomorrow) you'll find out that a line of storms will pass through the state from north to south starting this evening and ending tomorrow morning. Despite the "Slight" tomorrow there may not even be any storms in the afternoon - it may have only referred to the storms passing through in the morning.

Are the SPC's forecasts always accurate? My opinion based on years of observing them is that they rarely underwarn but do sometimes overwarn here in Pennsylvania (which is fine because here we're talking about making you aware about the possibility of severe weather, not nailing a forecast).

FACEBOOK TIP: When posting SPC's web pages on Facebook, you may notice that it always posts the wrong version of the map as a thumbnail. This is because Facebook stores ("caches") the maps on their server and everyone always posts the same URL for today's forecast. Instead, put in today's date in the form at the bottom of the main outlooks page to get "permalinks" to each forecast SPC has issued today. Click through to one of those (the date should be in the URL) and when you post those to Facebook, they will have the correct thumbnail map.

So now you know when there is a chance of storms - but how will you know exactly when and where? You can read the text below the SPC maps but you'll need this special decoder ring to understand them. Sometimes they provide information on when and where storms will develop and move; sometimes they don't. Once storms start popping up, or even a few hours before, you will be able to get more information through the SPC's Mesoscale Discussions or Watches but then we're getting into "nowcasting" which may end up being "Part 3" of this series - I want this one to stick to forecasting storms 12-48 hours out.

To get a better picture on timing and location, tune in to "Severe Weather Forecasting Secrets: Part 2: The Models" tomorrow, where I'll tell you where to see a simulated picture of what the radar scope will look like up to 48 hours in advance!

*Meteorological Disclaimer: This advice is for the U.S. only and shouldn't be used for protection of property or life; for that, use a forecast from the National Weather Service who has hundreds of meteorologists looking at the weather in your area. It's true that if you went through Calculus IV in college to get your meteorology degree and can read the most complicated weather maps, you'll be able to make a very accurate forecast, say 90-percent accuracy with a couple hours of work, but what I'm arguing here is that you can forecast with probably 80-percent accuracy in a few minutes.

Report a Typo