Happy Monday! It’s inevitable as a meteorologist or weather forecaster in the United States to hear the phrase, “must be nice to be wrong 50% of the time and keep your job,” at some point in their career.
It can be a challenge to pin down what exactly Mother Nature will throw at us on a day-to-day basis, but forecast skill has improved greatly in the last 50 to 100 years as we’ve learned more about how the atmosphere works.
Which brings us to the coronavirus, which in addition to changing how Americans live their lives, has impacted the tools that forecasters use every day to predict the weather. To understand why this is happening, we have to explain how weather forecast models work.
You’ve probably heard us talk about “the European model” or “the American model” at times when we’re discussing the scenarios involved with an upcoming storm. Weather forecast models can be described by the area they forecast for (resolution) and how far out into the future they predict.
You’ve likely heard of the ECMWF and GFS, but Your Local Weather Authority also uses the NAM, HRRR and other forecast models to predict our weather each day in southwest Virginia.
So how do these models work? They run multiple times per day, taking current weather data (temperatures, cloud cover, precipitation, etc.) and plugging them into atmospheric equations. The equations get resolved by the model and produce an output of forecast data. Each model assimilates the data differently and uses different equations, hence the slight variations in model output.