A transition from CMIP3 to CMIP5 for climate information providers: the case of surface temperature over eastern North America

• Published in: Climatic change 2013-09, Vol.120 (1-2), p.197-210
• Main Authors: Markovic, Marko, de Elía, Ramón, Frigon, Anne, Matthews, H. Damon
• Format: Article
• Language: English
• Subjects: Americas; Atmospheric Sciences; Climate; Climate change; Climate Change/Climate Change Impacts; Climate science; Climatology. Bioclimatology. Climate change; Datasets; Earth and Environmental Science; Earth Sciences; Earth, ocean, space; Emissions; Exact sciences and technology; Experiments; External geophysics; Global temperatures; Global warming; Greenhouse effect; Mathematical models; Meteorology; Radiative forcing; Ratios; Software; Surface temperature; Temperature; Uncertainty
• ISSN: 0165-0009; 1573-1480
• DOI: 10.1007/s10584-013-0782-8

The release of new data constituting the Coupled Model Intercomparison Project—Phase 5 (CMIP5) database is an important event in both climate science and climate services issues. Although users’ eagerness for a fast transition from CMIP3 to CMIP5 is expected, this change implies some challenges for climate information providers. The main reason is that the two sets of experiments were performed in different ways regarding radiative forcing and hence continuity between both datasets is partially lost. The objective of this research is to evaluate a metric that is independent of the amount and the evolution of radiative forcing, hence facilitating comparison between the two sets for surface temperature over eastern North America. The link between CMIP3 and CMIP5 data sets is explored spatially and locally (using the ratio of local to global temperatures) through the use of regional warming patterns, a relationship between the grid-box and the global mean temperature change for a certain time frame. Here, we show that local to global ratios are effective tools in making climate change information between the two sets comparable. As a response to the global mean temperature change, both CMIP experiments show very similar warming patterns, trends, and climate change uncertainty for both winter and summer. Sensitivity of the models to radiative forcing is not assessed. Real inter-model differences remain the largest source of uncertainty when calculating warming patterns as well as spatially-based patterns for the pattern scaling approach. This relationship between the datasets, which may escape users when they are provided with a single radiative forcing pathway, needs to be stressed by climate information providers.