Written by Every time I tell someone that I’m a meteorologist, I ALWAYS get the response “So you can be wrong most of the time and still keep your job?” I swear to you, EVERY time. I laugh it off but it annoys me.
The accuracy of weather forecasting has improved immensely over the past 20 years. Meteorologists would use Velcro icons and dry erase boards to deliver their 5-day. They had rudimentary computer modeling systems that were good at best. Anything past 4 days was a guess, mostly based on the climate (average weather for that day).
Nowadays, with advances in computer modeling like the creation of WRF (Weather Research and Forecasting model), our forecasts have improved a ton. For a forecast of 1-3 days, a good meteorologist can forecast with an 80% accuracy. Days 4 and 5, it decreases to about 65-70%. After day 5, the accuracy is at 50%. After day 7, the ‘forecast’ is really nothing more than climate averages.
There are some things to remember about forecasting too. Even though we’ve made great advances in computer modeling, there is still many improvements that can be made. The way computer models are initialized is by collecting data from radiosonde (weather balloons). The data that is collected is then placed on a map. The surface of the Earth is then giving gridpoints. Data and equations are plopped on each gridpoint and the model is allowed to run. (Meteorologist reading this — I know this is a water-downed version of everything). While we can improve the data going into the models (and there have been major improvements), models will never be perfect. It takes a meteorologist to interpret what the model is putting out.
Another challenge we have is forecasting for an extended area. Forecasting for New Haven is different than forecasting for Hartford. In the same respect, forecasting for DC is different than Ocean City, MD. However, at the end of each weathercast when you see the 7-Day Forecast, the icons/temperatures you see are averages of the entire state or region. Snow in Hartford may be rain in New Haven. Sunny skies in Ocean City may be rain in DC. The reason meteorologists say “50% chance of rain” is partly because some of the viewing area will see rain, and some won’t.
As I’ve written before, the difference between a good meteorologist and a great one is the ability to learn from his or her mistakes. Everyone busts on a forecast from time to time, but to truly be good, you must learn from it.
Forecast for tonight? Dark.
