The PCMDI Metrics Package: Objectives and Capabilities

The PCMDI Metrics Package (PMP) provides a diverse suite of summary statistics that objectively gauge how well climate model simulations agree with observations.  It is built on the python-based UV-CDAT.  Climatological comparisons for a range of observables are available from regional to large scales for the simulated atmosphere, ocean, sea ice and land surface.   Variability statistics are also provided for leading modes of interannual variability along with higher frequency composites of monsoon onset, decay, and the diurnal cycle of precipitation. The PMP has been applied to all generations of models contributing to the Coupled Model Intercomparison Project (CMIP), and is designed to rapidly deliver results from new simulations that are made publically available for research.  Because most climate modeling groups now have a workflow branch that parallels the CMIP data structure, interested modeling groups can implement the PMP as part of their analysis workflow to complement their existing model evaluation methods. This enables modeling groups to routinely contrast their simulations within a multi-model context, facilitating the assessment of model strengths and weaknesses during the model development process. In this way, the PMP will allow modeling groups to benefit from the perspective and insights the multi- model CMIP ensemble provides, and enhancing the value of CMIP within the model development cycle.  In this presentation we describe the overall design and use of the PMP and explain how analysts can contribute.   Specific examples from selected PMP variability analysis will be discussed in a separate presentation.