Forecasting for a Remote Island: A Class Exercise

A Class Exercise

View More View Less
© Get Permissions Rent on DeepDyve
Restricted access

Students enrolled in a satellite meteorology course at North Carolina State University, Raleigh, recently had an unusual opportunity to apply their forecast skills to predict wind and weather conditions for a remote site in the Southern Hemisphere. For about 40 days starting in early February 2001, students used satellite and model guidance to develop forecasts to support a research team stationed on Bouvet Island (54°26′S, 3°24′E). Internet products together with current output from NCEP's Aviation (AVN) model supported the activity. Wind forecasts were of particular interest to the Bouvet team because violent winds often developed unexpectedly and posed a safety hazard.

Results were encouraging in that 24-h wind speed forecasts showed reasonable reliability over a wide range of wind speeds. Forecasts for 48 h showed only marginal skill, however. Two critical events were well forecasted—the major February storm with wind speeds of over 120 kt and a brief calm period following several days of strong winds in early March. The latter forecast proved instrumental in recovering the research team.

Marine, Earth and Atmospheric Sciences Department, North Carolina State University, Raleigh, North Carolina

CORRESPONDING AUTHOR: Allen J. Riordan, Marine, Earth and Atmospheric Sciences Department, Box 8208, North Carolina State University, Raleigh, NC 27695-8208, E-mail: al_riordan@ncsu.edu

Students enrolled in a satellite meteorology course at North Carolina State University, Raleigh, recently had an unusual opportunity to apply their forecast skills to predict wind and weather conditions for a remote site in the Southern Hemisphere. For about 40 days starting in early February 2001, students used satellite and model guidance to develop forecasts to support a research team stationed on Bouvet Island (54°26′S, 3°24′E). Internet products together with current output from NCEP's Aviation (AVN) model supported the activity. Wind forecasts were of particular interest to the Bouvet team because violent winds often developed unexpectedly and posed a safety hazard.

Results were encouraging in that 24-h wind speed forecasts showed reasonable reliability over a wide range of wind speeds. Forecasts for 48 h showed only marginal skill, however. Two critical events were well forecasted—the major February storm with wind speeds of over 120 kt and a brief calm period following several days of strong winds in early March. The latter forecast proved instrumental in recovering the research team.

Marine, Earth and Atmospheric Sciences Department, North Carolina State University, Raleigh, North Carolina

CORRESPONDING AUTHOR: Allen J. Riordan, Marine, Earth and Atmospheric Sciences Department, Box 8208, North Carolina State University, Raleigh, NC 27695-8208, E-mail: al_riordan@ncsu.edu
Save