On estimating climate sensitivity from 20th century climate records: the impact of forcing uncertainty

The climate records of the 20th century provide an opportunity to estimate the climate system properties controlling the transient response of the climate system to external forcings. Of these properties, climate sensitivity is the most well known and was the subject of the Charney Report (NRC, 1979). Since 1979, several significant papers have appeared that use the surface, upper-air, and oceanic temperature records (some starting from the 19th century) to place likelihood constraints on the value of climate sensitivity. We will present the latest results using these records to construct a joint probability distribution for climate sensitivity, rate of heat uptake by the deep ocean and anthropogenic aerosol forcing. We also discuss sensitivity of obtained estimated to different 20th century forcings. The inclusion of the entire set of known forcings (particularly, volcanic aerosol forcing in addition to the anthropogenic forcings) is a critical factor in estimating the climate system properties. This forcing uncertainty and the poor estimates of the deep-ocean temperature records (mainly below 700m) are the key research issues needed to improve our estimates of climate sensitivity and other climate system properties. We also note that only "effective" climate sensitivity can be estimated from a transient record as the active feedbacks would be different from those at equilibrium. This implies an additional uncertainty in estimates of "equilibrium" climate sensitivity not often considered.