The complexity of numerical climate models has created an enormous challenge when trying to understand and compare resulting datasets. There is often no easy way to determine the reason for differences other than interrogating numerical model code and documentation. While there are efforts underway to describe the numerical model output data (such as netCDF Climate and Forecast (CF) Metadata), there has been no organised effort to uniformly describe a numerical climate models.

The Centre for Global Atmospheric Modelling (CGAM) in the Department of Meteorology at the University of Reading, in collaboration with the British Atmospheric Data Centre (BADC), Climateprediction.net and participants of the Global Organisation for Earth System Science Portals (GO-ESSP) have begun developing a framework with which to describe numerical climate models.

The three-layer description of numerical model metadata is based on a framework of standards using XML. The first layer describes the numerical model, the second layer describes the numerical model component(s), and the third layer describes the numerical model experiment. The fourth layer - the resulting dataset - is being described by other metadata standards such as CF.

The development of a schema to describe a numerical climate model, its components, and the experiments run using those components will ultimately allow for vast variety of tools to better understand, and use, the models and the data derived from them.

This talk will describe the progress in defining the model metadata standard and in building tools that can exploit it.