Career In Meteorology

Why is Meteorology so inaccurate so often?

On Tuesday, the forecast for today in my area was supposed to be 83. It is currently already 87 degrees outside at 9am, and the forecast has changed to 108. I do live in Los Angeles and we are notorious for having crazy weather, but it just seems like they are very often wrong when predicting weather. Why can we go to Mars, clone mammals and create cutting edge communications technology, but we haven't mastered the art of deciding whether or not to wear a coat the next day?

Public Comments

1. One problem is weather guys have way too much faith in models. The models are not being developed in a manner that will ever lead to one that will predict much better until the people making the model realize the Earth is being warmed by energy flowing from the interior of the planet. So far all the experts in NOAA&NASA and elsewhere deny the Earth is being heated by geothermal heat. Since the geothermal flux is in fact the dominate force in weather and climate details its just hopeless to think a good model can develop with the exclusion of the geothermal flux.
2. for a single forecast it takes over 100 equations to make it pretty accurate, so if one thing goes on it could be off by a li'l to a lot. and weather changes a lot, too; like a sudden wind change could blow the cold/ warm front the other way or even the storm clouds and it could take a while to acquire the info from the satellites
3. If we cannot accurately measure atmospheric conditions at every single point in the atmosphere every single second...how can we be expect to accurately predict it? The atmosphere is so vast and chaotic, and we simply cannot know what's going on at every point...which makes it impossible to predict it with 100% accuracy. I like to use this analogy: I put a rubber ducky in a river...let's say it's 1/2 mile wide. I tell you what direction and speed the water is flowing at maybe 20 points within the river. Do you think you'd be able to tell me with 100% accuracy exactly where that rubber ducky will be in the river 5 minutes later? Even with the most sophisticated computer model, it will be difficult to predict the location of that rubber ducky with so little information given about the river. There's simply too many microscale turbulent features (i.e. eddies) within the river to make it possible to predict accurately. The same holds true in the atmosphere...just on a much larger scale. I do find it hard to believe the forecast for your area went from being 87 degrees one day, to 108 the other. It's hard to be that far off. The trouble with listening to the weather on the radio or seeing the forecast on TV...is that you don't always know exactly what point the forecast is for. It is entirely possible the forecast for 87 you saw, was for the beach, which will be cooler being right near the water...while when you heard the forecast for 108, it was an inland location that is typically much hotter. Neither forecaster may have made clear exactly where the forecast was for. And the weather can be highly variable depending on your exact location, especially out in LA where you have the coast, valleys, and mountains, all in a relatively small area.
4. Hey C.S. I'm no meterologist, but I studied to be one for a year. (It was too much for me between my music, mock trial, radio/tv, regular school responsibilities, job, and girlfriend.) Matt has a great answer for you. I'd only add to it that the smallest change can have a huge effect down the line. One storm may run over a particularly cool land mass and lose energy. Or if teh day is just 10% cloudier, that prevents some radiant heating, which may affect clouds, temperatures, pressures, and fronts "down the line." Finally, if you look, I think you'd find that weather predictions are actually pretty good (up to, say, 48-60 hours in advance) considering all of the things the weather guy / meterologist (and yes there's a HUGE difference) / computer models have to take into consideration. Here's a "freakonomics" blog post about it. The poster is generally pretty negative, but I think if you're no more than 3 degrees off 3 days out, that's pretty dang good. http://freakonomics.blogs.nytimes.com/2008/04/21/how-valid-are-tv-weather-forecasts/