Column 14: A day in the life (originally posted January 13, 2006)

I was asked to write a column about a "day in the life" of a meteorologist.
Good topic idea! For the purposes of this column, I'll write about a typical 12-hour day shift, as there are obviously aberrations. And boy, can they be crazy. (And usually fun, too!)

I was asked to write a column about a "day in the life" of a meteorologist.
Thanks, Rain! Good topic idea! For the purposes of this column, I'll write about a typical 12-hour day shift, as there are obviously aberrations. And boy, can they be crazy. (And usually fun, too!)

Here's a tour of our office!

Before I walk in the door, I make sure to have my fuel at hand--for me it's an extra large from Timmy's.

The shift change briefing is meant to be an overview of the goings on, areas of concern, and potential problems. The outgoing shift usually gives a 5- to 10-minute description of the overall meteorological pattern and any diversions from this pattern that might not be obvious--an unexpected wind, a temperature higher than expected, things like that.

After the briefing but before starting anything else, I have to make sure the current forecasts aren't "broken". That is, that no unforecast weather is occurring. This is pretty rare, but once in a while it happens. If an amendment is needed, that's when I do this.

Then I sit down and roll up my sleeves. Literally.

As you might know if you've read any of my columns, I'm a giant proponent of analysis, diagnosis, and prognosis. This isn't a trivial set of tasks.

I print out the unanalyzed upper air observations plotted on a map of most of North America, at 250, 500, 700, 850, and 925 mb. On all of them I contour the heights, which are analogous to pressure. On the 250 mb chart, in addition, I analyze the jet streams and streaks. On 500 and 700 mb, I analyze the temperatures. At 850 and 925 mb, I analyze the wet bulb potential temperature--this is a much better field to analyze at this level for frontal analysis than simply temperature. After those are done, then, the surface analysis follows.

Once I'm done all the chart analyses, I bring up the individual soundings around the region and nearby, looking for features that will give me clues to possible weather for the day. For example, in the summer, I'll look for potential instability. In the winter I'll look for freezing rain potential and snowmaking layers.

That's the analysis done.

Next, I look at my satellite images and RADAR composites. I compare those against my upper chart analyses to try explaining to myself and my colleagues why the weather is doing what it is, have a discussion with them about all that. This is really where the University training comes in--you have to be able to instantly recall meteorological theory and have all sorts of conceptual models at the tip of your brain. Usually, the situation is atypical, so you have to construct a hybrid conceptual model of what the atmosphere is doing that day. The RADAR and satellite imagery will also sometimes fine tune the analysis. Once I've finished justifying why the weather is doing what it's doing, the diagnosis is done. And now it's time to move onto prognosis.

Prognosis is something on which entire University courses are given. I use all sorts of techniques I won't go into here to forecast the future evolution of the different elements of weather--cloud, precipitation, wind, and temperature. One thing I will point out, however, is that in the first 12 hours of my forecast period, I ignore the forecast models.

Say what?

Yes, I ignore the models. There are 2 main reasons I do this: first off, it has been shown that, in the first 12 hours of its valid period, a forecast model is horrible. Secondly, even when a forecast model is past its first 12 hours, it still does fairly poorly (in non-benign situations) in the lowest 1 km or so of the atmosphere. Shock of shocks, that's where most of us live! (Although many of my friends would say I always have my head in the clouds, but I digress.)

Okay, so I've done my analysis, diagnosis, and prognosis. It has taken me anywhere between 1 (relatively easy day) and 2 and a half hours (complex day). What comes next? I have to put out forecasts--the morning updates.

Putting out the forecasts is the easiest part of the job, because all of the "work" (i.e. the meteorology) is done. The software we use allows me to modify every important weather element (sky condition, precipitation, POP, wind, temperature, etc.) and then at the press of a button ... it produces the worded forecast! Awesome thing, this technology!

Once the morning updates are done, it's time for lunch. A half hour to 45 minutes is good, and then I come back to start on my midday analysis.

It's only a surface analysis this time, because the upper air observations only come twice a day. But it's still important, especially if satellite and RADAR have indicated something that I hadn't seen with the morning analyses.

Anyhow, the afternoon is pretty much the same--I redraw the mental picture of what the state of the atmosphere is, and put together the forecasts for the afternoon update.

After the afternoon update is done, it's time to sit back and maintain a "weather watch". That's code for checking to make sure my forecasts stay valid.

In the summer, of course, if there are storms brewing, I'll be involved with that--either with interpreting RADAR, calling weather watchers, or issuing watches and warnings. And there are other things to do once in a while--a radio or TV interview, but mostly the day is spent in our insular society of an office.

This rounds out the rest of the day until the night shift comes in. I give them their briefing and my day is done!

I hope this has given you a glimpse into what we do. If you have any questions about what we do, don't hesitate to ask!