East Coast Storm Midweek; Challenges in Numerical Weather Prediction
By now, I’m sure you’ve heard numerous weather forecasters talking about the potential of a major coastal storm headed for the Northeast this week. Forecasts continue to fluctuate as weather models are updated and as of today, it’s still unclear exactly what will happen. There are a few different solutions this morning from various forecasting models, but each takes the storm in a slightly different direction. See below:
Each of these forecasts show the storm, but in slightly different positions and with variations in intensity. A few tenths of an inch on a weather map translates to dozens of miles in real life, so these minor differences will drastically impact what happens in places like Rhode Island and southeastern Massachusetts. [Maps courtesy UCAR/NCAR and WeatherBell]
So why all the uncertainty? It all hinges on how weather models work. Weather models take current observations and plug them into complex math equations to produce a forecast. As the model continues to chug along, spitting out numbers further and further into the future, small errors grow larger and larger.
[To be more specific: Weather models use systems of differential equations based on the laws of physics, fluid motion, and chemistry, and use a coordinate system which divides the planet into a 3D grid. Winds, heat transfer, solar radiation, relative humidity, and surface hydrology are calculated within each grid cell, and the interactions with neighboring cells are used to calculate atmospheric properties in the future. Read more: http://en.wikipedia.org/wiki/Numerical_weather_prediction]
For example, the plots below show lines of constant pressure in the atmosphere (think of it like circling areas of high pressure and low pressure) . Notice that all of the model variations are in fairly good agreement with the first prediction time. Not bad!
Watch what happens when we chug along to the storm that forms for Wednesday:
The circles off the coast of New England represent our storm system. Look how different some of the placements are . . . this happens when the small disagreements at the starting point get amplified with each time-step. We’re fairly confident that high pressure will take over the eastern half of the country, but smaller features like our coastal storm have more uncertainty. This propagation of errors is often likened to the “Butterfly effect“, where one small variation at the start of a time period can greatly alter the outcome later in time.
Just for fun, this is what it looks like when you push the time to next Sunday (March 30th). Yikes!
So you might say weather forecasts created 7 days in advance starts to look like kindergarten art projects. We can make educated guesses about the temperature in a 7 day forecast, but watching weather systems develop that far out is very hard to do.
If you are like me, and love to get more information on science topics like this, you can read about Numerical Weather Prediction and the challenges we face when predicting long-range weather systems.
We’ll keep you posted as the models come into agreement in the next few days concerning this storm here at WeatherNation. Thanks for getting all the way to the end! And since you read the whole article, here’s a bonus funny cat photo (this IS the internet, after all).
Have a great week! -Meteorologist Miranda Hilgers (@mhilgersWNTV)