Date of Completion

12-20-2013

Embargo Period

12-20-2013

Keywords

climate, variability, CMIP5

Major Advisor

Anji Seth

Associate Advisor

William Sillander

Associate Advisor

Carol Atkinson-Palumbo

Associate Advisor

William Ouimet

Field of Study

Geography

Degree

Doctor of Philosophy

Open Access

Campus Access

Abstract

Climate variability at the regional scale poses a challenge to short-term adap- tation and mitigation strategies. This study examined the ability of climate models from the CMIP 5 archive to capture the observed modes of climate vari- ability and their relationship with climate indicators in the Northeast United States (NE US). As year to year climate variability is largest in the boreal win- ter, this study focused mainly on observed and projected winter climate. The CMIP5 models were adequately able to simulate observed trends, the annual cycle, and synoptic boreal winter teleconnection patterns for temperature. For precipitation, the models did well capturing cold season trends, but overem- phasized the annual cycle. The models also did not do well simulating NE US winter teleconnection patterns from the NAM, PNA, and EP NINO. Future trends in climate indicators for the NE US suggest that will be warmer and wetter with more year to year variability, particularly for precipitation. In

creased variability in climate indicators suggest that there will be an increase in extreme events, which will be more dicult for society to adapt to than changes in seasonal or annual mean. Changes in teleconnection patterns for the 21st Century are projected under the high emission scenario (RCP8.5), but there is little model agreement in direction or magnitude of future changes. However, the CMIP5 models offer a better view into synoptic scale forcings than the CMIP3 models, and these results add to the collective understanding of potential changes to NE US climate.

COinS