CLARREO and Climate Scalar Prediction

Stephen Leroy, Harvard University

Abstract

Because of the persistently large uncertainties in inter-decadal climate prediction, the uncertainties being due to model error, it has become necessary to test climate models with well chosen datasets. For this reason the NRC Decadal Survey has chosen CLARREO (Climate Absolute Radiance and Refractivity Observatory) as one of its four top tier missions. Among other possible uses of the data, trends in long timeseries of CLARREO data can be used to estimate climate feedbacks. The Anderson Group has been heavily involved in the development of the science case for CLARREO in addition to its experimental work. Leroy et al. (2008a) presents a brief overview of that science case and will serve as the focus of discussion. Leroy et al. (2008c) presents generalized scalar prediction, a statistical method which combines the data from timeseries of arbitrary data and output of a perturbed physics ensemble of runs of a climate model to produce accurate and highly precise predictions for climate change on multi-decadal timescales.

Leroy, S.S., J.A. Dykema, P.J. Gero, and J.G. Anderson, 2008a: Testing climate models using infrared spectra and GNSS radio occulation. To be published in Occulations for Probing Atmosphere and Climate III, A. Steiner, G. Kirchengast, B. Pirscher, U. Foelsche (Eds.), Springer, Berlin.