Friday, June 12, 2009

Weather Forecasting is Not Fast Food

Weather forecasters are often the butt of jokes in our culture. "What other job allows you to be wrong so often...?" "I just shoveled 6 inches of partly cloudy out of my driveway..." These and other little zingers suggest that meteorologists are most often wrong, and that there are no consequences for this other than perhaps a red face on the next nightly local news program.

What I find most interesting about this representation of a highly skilled and technically demanding field of a fairly new science is how wrong it is, and how little most people actually know about what goes on to create a weather forecast. I think this is so, because most people approach meteorology with the same mindset that they approach pulling up to the drive-thru of their favorite local fast-food restaurant. "I'd like the number 2 value meal please, with a chocolate shake...", and just seconds later, their food is in their hands, exactly as they ordered it. The teenager serving them just had to push the right two or three buttons, and voila, there's the food. Little thought, and no analysis was needed.

Forecasting the weather is not quite so easy. You friendly local meteorologist does not show up at his or her office, push a button and dish out your forecast to you, with a side of fries, if you'd like. It is rare, at least in the middle-atlantic states where I live, that a forecast is arrived at quite so easily. Sure, much of the data available to the forecaster is a mouse-click away, courtesy of massive computer models run numerous times a day. But the science behind these models, and the analysis done by the forecasters in reviewing all the data available to them, produces a high-quality forecast which, more often then not, is actually right, and is not the same as pushing the cheeseburger button, instead of the fish-sandwich button.

Most people don't recognize that the studies one must undertake in order to become a meteorologist are extremely demanding. Frequently, while in college, non-meteorology majors will take an Intro to Meteorology class, and think, "Hey that was easy, I can do that!" What they don't actually realize is that for meteorology majors, that class is just the beginning of a sequence of courses that few students actually make it through, which include many semesters of high level calculus, physics, thermodynamics, and so on, and so on. There is precious little time actually spent observing and reviewing the weather these students love. Instead, hours and days are spent memorizing formulas and learning mathemetical equations that, at their best, numb the mind and make one wonder whether it is all worth it.

Eventually, those gifted and lucky enough to make it through graduate with their degree and proceed to try to enter what is at best, a tiny field. There are very few openings, especially for one without prior experience. Sure, there are jobs for those with military experience, or for those who after suffering through what it took to graduate with a bachelor's degree, were masochistic enough to pursue their Master's degree. But there is no corner forecaster in each town, looking to hire the latest crop of new graduates. Truly, only the best and brightest find their way into the field of weather forecasting.

Once in their field, the studies these men and women have done is put to the test. Computer model data is not exactly the same as a polished, accurate, ready-made forecast. What do you do when one model predicts dry weather and another suggests torrential downpours? How does one judge whether the 500-1000 mb thickness is right in suggesting snow, when normally, the storm track predicted would suggest rain? Should a flood watch be issued - precipatable water values are very high right now, but streamflows are low in the forecast area.

All of this is not to say that weather forecasting is too difficult, and can't be done. This is a science that has made great progress in just the past 10 or 20 years. Weather forecasting models of today make the earliest models look foolish. We can predict with accuracy once only dreamed of. And when mistakes are made, they are regrettable but unfortunately, expected given the complex science. Consider the intricacy of the atmosphere - it is never exactly the same on any one day as it was on any other day. And yet we are able to take volumes of data and, on most days, turn out an accurate forecast. I think that is worthy of a great deal of respect.

So the next time you read your forecast, remember that the men and women behind it have worked very hard to make it as accurate as possible, using the skill and knowledge honed over years of study and experience. Perhaps you can't get it with fries and a soda, but you can get it with the knowledge that it is probably right.

No comments:

Post a Comment