the system, it also leads to friction and inefficiencies in the forecasting system. In carrying out its legislated responsibilities, the NWS must make continuous weather predictions on all spatial and temporal scales. The resulting forecasts can be specialized, to some extent, by private companies for local and/or individualized applications. Inefficiencies result from the necessary duplication of some of the efforts to prepare these two types of forecasts. However, it must be recognized that to satisfy the different requirements and objectives of the three sectors, and thus achieve the economic and public benefits to the nation, the weather forecasting system as a whole cannot be perfectly efficient.
Similarly, some level of friction is inevitable in a three-sector system when activities overlap and advances in science and technology change what the sectors are capable of doing. An increasing role of the private sector in data collection, for example, could make it more difficult to maintain the open access to data upon which weather and climate applications depend. Advances in modeling and improvements in forecast accuracy, particularly extending forecast accuracy to seasonal and longer timescales, will open new business opportunities for predicting the effects of El Niño and other climate events on industries such as agriculture, energy, and insurance. And new methods of dissemination will reduce some business opportunities (e.g., making government data more accessible) and create others (e.g., personalized weather services via wireless devices).
A number of ideas have been proposed that putatively would increase the efficiency of the weather and climate information system and decrease friction among the sectors. Some in the private sector have suggested eliminating some weather services from the NWS or ending research grants, which they view as subsidies, by atmospheric science funding agencies (e.g., National Science Foundation, National Oceanic and Atmospheric Administration [NOAA], Department of Defense) to scientists creating value-added products.1 Some NWS officials and academic researchers have proposed that companies contribute their data without restrictions to the general pool. However, these suggestions have shortcomings (see Chapter 4), and none have proven to be practical to implement. A better solution is to use criteria such as those given in Box 4.3 to make case-by-case decisions on which sector should undertake a particular activity. Rigid boundaries between the sectors are hard to define, and any such definitions quickly become obsolete as technological capabilities and user demands evolve. Consequently, any precise and detailed division of responsibility that the