AR hosts workshop on predictability and chaos in the geosciences by Robert Henson, NCAR Information and Education Outreach Program workshop that featured presentations on predictability and chaos in the geosciences drew 35 scientists from Canada, England, France, the Netherlands, and the United States to NCAR in early September. Sponsored by NCARs Geophysical Turbulence Program (GTP), the workshop examined issues of predictability both within the international community of geophysical turbulence research and in the broader domain of earth system science. Joseph Tribbia (NCAR Climate and Global Dynamics Division, CGD)along with Jackson Herring and Robert Kerr (NCAR Mesoscale and Microscale Meteorology Division, MMM) and Peter Fox (NCAR High Altitude Observatory, HAO)convened the workshop, with the assistance of Hope Hamilton (MMM). The atmospheric sciences have long been at the forefront in using chaos theory and examining related constraints on the prediction of natural systems. However, such issues span the geosciences, including physical, chemical, and biological systems. The symposiums goal was to find common ground among geoscience specialties and to encourage collaboration through communal discussion of (a) the chaotic behavior of geophysical systems and (b) related methods of analysis and information use. NCARs GTP includes scientists from three divisions: CGD, MMM, and HAO. Collaborative efforts extend to more than 30 national and international institutions. A major goal of GTP is to develop new methods and tools for representing turbulence effects on fluid motion in the atmosphere of the earth and sun. Among the topics covered in the symposium were the following. Does a Statistical Mechanic in the Ocean Circulation Contribute Enhanced Climate Predictability? (Greg Holloway, Institute of Ocean Sciences, Sidney, British Columbia, Canada). Coupled ocean-atmospheric models, while accounting for oceans interaction with the atmosphere, do not account for the interaction of ocean eddies with the sea floor. Holloway depicted the latter in a hybrid ocean model by assessing the departure of modeled ocean flow from a maximum-entropy configuration. Predictability and Sensitivity of Isolated Weather Systems. (Douglas Lilly, University of Oklahoma). Original analyses of the chaotic limits to predictability were estimated using closely spaced, interacting energetic elements. A more complex problem is the predictability of isolated or widely spaced intense elements, such as supercell thunderstorms and hurricanes. The stability of these systems structures may fluctuate in ways that are cyclic or secular as well as chaotic. Predicting Earthquakes: What Can We Learn from Models? (Jean Carlson, University of California, Santa Barbara). Algorithms based on pattern recognition have shown recent promise as a means for making intermediate-range earthquake forecasts. Carlson showed quantitative assessments of the success of such a strategy. Chaotic Modulation of the Solar Cycle. (Nigel Weiss, University of Cambridge, England). The suns 22-year magnetic cycle is aperiodic, butas with other starsit has not been observed long enough to determine if its pattern is deterministic or stochastic. Models of the dynamo process (the source of both solar and stellar magnetic fields) feature chaotic oscillations. Models of nonlinear dynamos yield chaotic behavior with intermittent cyclic activity, mimicking the behavior of the sun. A volume of abstracts and discussions from the symposium will be published this fall. For more information, see page 15. uQ[2YOB(q|fWP lgH5gٓfZOHR's)\\#MHg߆PX*( DMOV(?-|ay8