Prince William Sound Field Experiment

weather forecasting
in the sound

weather research and forecasting

Go to AOOS Wind data (WRF)

Also available from AOOS:

Go to AEFF WRF data

Ocean Modeling

Principal Investigator: Peter Olsson,
University of Alaska Anchorage

It's 3 a.m. and a fisherman needs to know what the weather will be like for Prince William Sound over the next 48-hour fishing window.

Mariners need solid weather forecasts—especially when conditions turn marginal. 

The Sound has a dense network of observation platforms, with more than 20 weather stations within an area of 100 square kilometers. Because of this dense network, AOOS Prince William Sound Observing System forecasts can be better scaled for the needs of local mariners. AOOS weather stations provide more surface observations that tell us how well the models are performing, and also provide real-time measurements of actual weather conditions.

Weather models: WRF

Using these data, AEFF operates the Weather Research and Forecasting (WRF) model and the North American Mesoscale Weather Research and Forecasting (NAM-WRF) model to make forecasts. These models provide much finer resolution than the current National Weather Service model. Where the NWS now only has forecasts for areas of about 12 square km, the models developed by AEFF allow for forecasts of areas as small as 4 square km.

Next-level modeling: WRF to RAMS

"The finer resolution allows us to capture topographic effects that are not in the NWS simulations," Olsson said. The result are used as the basis for a Regional Atmospheric Modeling System (RAMS), a system for predicting the weather that is much more reliable than anyone has seen in the Sound.

To produce RAMS, the AEFF enters the data from the many different buoyed and land-based weather stations around the sound into a supercomputer made up of 13 smaller computers, all working together to create a mathematical model of the behavior of the Sound. Using this model, the AEFF can predict, with reasonable accuracy, the temperature, pressure, precipitation, winds, cloudiness, radiation, and many other elements of any area around the Sound. Using the data points, the mathematical model "fills in the gaps" between the measurement stations around the Sound, creating a complete picture of the Sound at any given moment.