Professor Paul Thursday - A Look At Regional Forecast Models

There are various types of forecast models that assist meteorologists in the challenges with making an accurate weather forecast. During the previous segment, some of the most used global forecast models were analyzed. One of the biggest benefits of global models is that they look at the weather all over the world. This is important because major weather systems that are over one area of the world can have significant impacts on the weather thousands of miles away. For example, a major typhoon that is over southeast Asia can produce major disruptions in how the jet stream winds set up over the US. Meteorologists also utilize a type of forecast model known as a regional model. These models focus on the weather in one "region" of the world. 

The resolution of regional forecast models is much greater than that of global models. Basically, forecast models divide the area they are looking at into separate cells. This is comparable to how computer programmers look at pixels when making graphics. Global forecast models divide the world into much larger cells than regional models. The European global model splits the world up into cells that have an area of 9 km. This is the best resolution of all the global models. 

The US has developed some of the most used regional forecast models over the last few decades. One of the most useful is the RAP (Rapid Refresh) model. A forecast model that comes from the RAP is called the HRRR (High Resolution Rapid Refresh) model. This forecast model has a resolution of 3 km. This higher resolution gives the forecast model an increased ability to look at smaller scale weather features, such as thunderstorm cells. They also look at many more layers of the atmosphere from top to bottom. This gives these forecast models a superior ability to predict the severity of thunderstorm activity. They are also better able to pinpoint where and when storms will develop on any given day. During the winter, they can have advantages in predicting small-scale heavy snow bands as well. 

Since regional forecast models are at such a high resolution, they are not able to forecast the weather for more than a few days out. This is because the computing power needed is so much greater for a high resolution model.

Other regional forecast models that are run daily in the US are the NAM (North American Mesoscale) model and the WRF (Weather Research and Forecasting) model.