The Geophysical Statistics Program (GSP) at the National Center for Atmospheric Research (NCAR) makes contributions to atmospheric and oceanic science through the development and application of new statistical methodology. The project supports a critical mass of statistics visitors at NCAR, trains young researchers, and involves close collaboration with resident NCAR geophysical scientists. The continuation of this project expands statistical research to the entire spectrum of scientific activity at NCAR and in so doing, establishes a permanent statistics group at this research center. The richness of geophysical measurements pose substantive research problems in statistical science and continue to challenge the boundaries of statistics. Some broad areas of research include modeling and inference for spatial and spatial/temporal processes, statistical design, inverse problems, statistical computing, and functional data analysis. These statistical topics have many direct applications to the geophysical sciences, including the detection of climate change, improvement of numerical models, prediction of severe weather, and design of field experiments. The primary purpose of the Geophysical Statistics Program (GSP) at the National Center for Atmospheric Research (NCAR) is to promote collaboration between the statistical and geophysical sciences. NCAR, base-funded by the National Science Foundation, is a focal point for basic research on the atmosphere and has strong links to universities throughout North America. The complexity of problems in the atmospheric and oceanic sciences (e.g., large numbers of variables and massive data sets) indicates that statistics could make important contributions. There is also a broader benefit to statistics and mathematical sciences. New statistical methods, motivated by geophysical problems, can be transferred to other areas which consider spatiotemporal data and the output of numerical models. Besides these benefits to the scientific fields involved, this collaborative research would result in potential benefits to society. For instance, it could lead to improved weather and climate forecasts, as well as to more reliable quantification of uncertainty in global climate change studies. Such information should be of substantial value to decision makers and policy analysts.
