Overview

Statistical emulators for climate model output, among which pattern scaling has been perhaps the most popular thus far, are techniques for generating (low- dimensional or fully spatial) projections of future climate using a statistical model designed to reproduce results that would be expected from a projection with a full global climate model. These emulators in general, and pattern scaling in particular, are expected to play an important role in a new process underway by the climate change research community to produce integrated scenarios of future climate and societal change. These scenarios will underpin research by the integrated assessment and impact modeling communities on options for mitigating or adapting to climate change, as well as on estimating impacts that may occur. The research community would like to explore a large number of scenarios, but projecting climate change in each case with a large, computationally expensive climate model is infeasible. At the same time, the option of using emulators as a practical and credible alternative to these models is open to question.

This workshop will have three main goals:

  • Assess the current state of climate model emulator science
  • Assess to what extent current approaches can meet the needs of integrated assessment and impact modelers for climate change information
  • Identify and prioritize research directions so that these statistical methods can better meet the needs of applied research in the future.

The workshop will be held on April 23-25 at the National Center for Atmospheric Research and will bring together statistical, climate, integrated assessment and impact modeling communities.

Home


Backgrounder


Bibliography


Schedule


Download Schedule (pdf)


Talks (pdf)


Poster Session


Speakers


Participants


Request participation/ Submit an abstract


Local Information


Download Announcement (pdf)