The observing system is in place, the models are ready, the applications are beginning to be appreciated: the next four years will see a flourish of activity in bringing the field of short term (up to one year) climate prediction to fruition.
As the TOGA program winds down, the international CLIVAR (Climate Variability and Predictability) Program and its U.S. contribution, GOALS (Global Ocean Atmosphere Land System) will begin. Both have seasonal-to-interannual prediction as their foci and both are committed to perfecting and expanding the prediction capabilities through programs of observing, modeling, and predicting.
One of the basic issues the national and international communities will have to deal with is the proper institutional form that these regular and systematic prediction activities will take. In a recent proposal commissioned by the Inter-governmental TOGA Board [ Moura, 1992], an International Research Institute for climate prediction (IRI) was proposed to perfect the science of climate prediction; to make the best possible predictions using the best possible prediction systems; and to interact with those countries that need such predictions in a partnership mode such that their applications needs are understood and properly dealt with. The US agreed to lead the effort to begin such an Institution and it was recognized that such an Institution, as well as the national efforts currently underway at the National Meteorological Center, can serve as a national and international focus for prediction and applications research. The next four years will see whether or not such a concept can indeed be implemented.
Acknowledgments. We are grateful for the critical reviews by the following scientists: M. Cane, D. Neelin, C. Penland, P. Sardeshmukh, S.-P. Xie, and an anonymous reviewer. Editorial comments by R. Pielke helped make the paper more readable. This work was supported by grants from the NOAA Office of Global Change Programs.