The current modernization and restructuring of the National Weather Service (NWS) to date has not involved the rehabilitation of the upper-air observing network over North America. The authors discuss the need for a thorough evaluation of the current upper-air observing system that examines both the old (current) rawinsonde network and newer remote sensing networks. The key concern is that the meteorological community has given insufficient attention to the scientific questions underlying the choices of observing systems as related to the needs of operational forecasting. The authors argue that the principal basis for a scientific and objective evaluation of observing strategies is the ability of a particular observing system (or component of a system) to improve numerical weather prediction in the modernized NWS. Given this starting point, a number of possible new (and old) observing systems and observing strategies are discussed that offer the possibility for much improved initialization of mesoscale numerical forecast models. Perhaps the key shift in observation strategy suggested is the establishment of many additional radiosonde and pilot balloon sounding stations that rely on contract observers rather than the NWS workforce. However, a comprehensive and satisfactory upgrading of the upper-air observing system will be possible only after more substantial participation of the scientific community in the planning and implementation processes.

This content is only available as a PDF.