Geophysical Statistics Project

The mission of the Geophysical Statistics Project (GSP) is to encourage the application and further development of statistical analysis to the problems faced in the Earth sciences.

The Geophysical Statistics Project pursues the innovative application and development of statistical methodology to address problems faced in the Earth sciences. A complementary activity is to generalize specific problems in the geophysical sciences to broad based statistical research.

Some major research areas of GSP are:

  • Extension of statistical methodology for spatial processes and space/time processes.

  • Application of modern regression and model selection to the analysis of geophysical data.

  • Deriving a statistical basis for forecasting including the assimilation of observational data with numerical models.

  • Modeling complicated physical processes through Bayesian hierarchical models.

  • Understanding physical processes through the use of dynamical systems and nonlinear time series.

In addition to such statistical research, the GSP also seeks to provide statistical expertise concerning the application of traditional statistical ideas for researchers at NCAR.

The GSP attempts to serve as a bridge between the atmospheric-oceanographic and the statistical-probabilistic research communities. The Project sponsors visits by statisticians at various levels, as well as postdoctoral positions.

GSP is an institution-wide effort at the National Center for Atmospheric Research (NCAR) and is funded by the Division of Mathematical Sciences of the National Science Foundation.

GSP is administratively housed in the Institute for Mathematics Applied to Geosciences (IMAGe) of NCAR at the Mesa Laboratory in beautiful Boulder, Colorado.