A Bayesian Model for Inferring Prehistoric Temperatures from Natural Proxies

• Activity Number: 290
• Type: Topic Contributed
• Date/Time: Tuesday, August 3, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics and the Environment
• Author(s): Martin Tingley*+ and Peter Huybers and Konrad Hughern
• Companies: Statistical and Applied Mathematical Sciences Institute and Harvard University and Woods Hole Oceanographic Institution
• Keywords: Bayesian ; Climate Change ; Paleoclimate ; Environment ; Application ; Temperature

Abstract

The instrumental temperature record covers at most 150 years. Because a longer record is needed to characterize the natural variability of the climate system, it is necessary to call upon climate proxy data, which are noisy and sparsely distributed in space. Information about pre-historic temperatures can be derived from historical documents and from elements of the natural world sensitive to local temperature variations, such as tree rings, ice cores, and lake-floor sediment cores. We present a Bayesian hierarchical model for reconstructing the past evolution of a univariate climate field from proxy and instrumental observations that overlap in time, and we apply this model to a 600-year high-northern-latitude temperature proxy data set. Results indicate that both the magnitude and rate of warming in the late 20th Century are unprecedented with respect to the previous 600 years.