Cloudy forecast for weather satellite data.

IN THE NEWS FOCUS STORY “WEATHER forecasts slowly clearing up” (9 November, p. 734), R. Kerr nicely summarizes how the growing improvement in prediction is coming from a focus on “more computer power, the assimilation of radar observations, and more physically realistic models.” He emphasizes that better assimilation of satellite data is a key element of improved forecasting. Unfortunately, these potential improvements will have little effect on forecasts if the basic data set from the existing polar-orbiting weather satellite system is not available. In a report issued in June 2012 (1), the U.S. Government Accountability Offi ce noted that “data from this system is the predominant input to numerical weather prediction models” and warned that “there will likely be a gap in satellite data lasting 17 to 53 months” when the National Polar-orbiting Operational Environmental Satellite System (NPOESS) Preparatory Project satellite ceases operations and NOAA’s new satellite system (the Joint Polar Satellite System) launches. The report also notes that there are “potential satellite data gaps in DOD [Department of Defense] and European polar satellite programs which provide supplementary information to NOAA forecasts.” These gaps are a grave problem and would seriously degrade weather forecasts. Therefore, the agencies responsible for weather forecasting—NOAA, DOD, and NASA—should make fi lling the polar satellite data gaps the fi rst priority in order to ensure that future forecasts are as good as possible. D. JAMES BAKER