U.S. Dept. of Commerce / NOAA / OAR / PMEL / Publications
The mid-twentieth-century warm event in the Arctic is an interesting phenomenon of relevance to current climate change issues. The simulated results by AOGCMs add to the understanding of this event. We are encouraged that all model ensembles for IPCC AR4 portray upward trends to various degrees for the Arctic in the last two decades of the twentieth century. Five models reproduce somewhat reasonable amplitudes compared to the midcentury event and have comparable variance to arctic temperature observations: (CCSM3, CRISO-Mk3.0, INM-CM3.0, ECHO-G, and PCM). Three other models (ECHAM5/MPI-OM, GFDL-CM2.0, and GFDL-CM2.1) also reproduced reasonable magnitudes compared to the midcentury warm event, even though the intrinsic variability in the control runs is small. However, all of these models do not have the sustained duration of the observed mid-century event. Four additional models cannot be excluded based on the variance test of their control runs (CGCM3.1-T47, CGCM3.1-T63, CNRM-CM3, and UKMO-HadCM3), but they fail to reproduce the required magnitude of midcentury warm anomalies in their single realization in the 20C3M simulations. We consider that the eight models [GISS-AOM, GISSEH, GISS-ER, IPSL-CM4, MIROC3.2(hires), MIROC3.2(medres), MRI-CGCM2.3.2, and FGOALSg1.0] that did not pass both criteria (magnitude in 20C3M simulation and control runs variance) do not have enough intrinsic decadal variability to produce a reasonable magnitude for arctic warm anomalies. Passing our criteria is not a complete acceptance of the models for climate projections in the Arctic, only that they should be given priority in assessments of projected change relative to natural variability.
The random timing of the midcentury warm anomalies in the model 20C3M simulations together with the similarity of midcentury events in the 20C3M simulations to the control runs (with neither natural nor anthropogenic external forcing), and the qualitative difference in the behavior of their time series in the early and end of the twentieth century, are evidence that the midcentury Arctic warming event in the observational data was due to different causes from those of the late twentieth century. The intrinsic variability of the atmosphere together with the feedbacks between the atmosphere and other components of the climate system (e.g., sea ice, ocean, and land processes) are likely responsible for the observed warm anomalies in the mid-century, as also noted by Bengtsson et al. (2004).
Finally, in IPCC TAR, the ACIA report, and other documents, the projections from the climate models are often the averages from all of the models and their ensemble members. We suggest that the projection of the future climate should be based on a subgroup of models that perform reasonable simulations of the past based on fixed criteria. Here, eight models have serious limitations for the near-term Arctic climate predictions (20–50 yr), because of the lack of the potential interplay of anthropogenic contributions and intrinsic variability. On the other hand, five models show promise in this aspect, and another seven might be further considered with reservation. Our results are a step toward constraining the currently scattered projections of the Arctic climate (see, e.g., Symon et al. 2005).
Acknowledgments. We acknowledge the international modeling groups for providing their results for analysis, the PCMDI for collecting and archiving the model data, the JSC/CLIVAR Working Group on Coupled Modelling (WGCM) and their Coupled Model Intercomparison Project (CMIP) and Climate Simulation Panel for organizing the model data analysis activity, and the IPCC WG1 TSU for technical support. The IPCC data archive at LLNL is supported by the Office of Science, U.S. Department of Energy. We thank three anonymous reviewers for their thorough review and suggestion in the review process, which helped us keep good focus on the discussion. This research is supported by the NOAA/CMEP Project of Office of Global Programs and the NOAA/Arctic Research Program. Kattsov and Pavlova were supported by the NSF via the IARC (Subaward UAF05-0074 of OPP-0327664). Zhang was supported by Japan Agency for Marine-Earth Science and Technology. Preparation of this manuscript was supported by the NOAA/Arctic Research Office. This publication is partially completed through the Joint Institute for the Study of the Atmosphere and Ocean (JISAO) under NOAA Cooperative Agreement NA17RJ1232.
Return to previous section or go to References