A Comparison of the CMIP6 midHolocene and lig127k Simulations in CESM2

• Published in: Paleoceanography and Paleoclimatology
• Main Authors: Robert A. Tomas, Sarah L. Shafer, Gunter R. Leguy, Esther C. Brady, Natalie M. Mahowald, Peter Lawrence, Aleah Sommers, Patrick J. Bartlein, Bette L. Otto-Bliesner, Samuel Albani, Erik Kluzek, Mathew Rothstein
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2020
• Subjects: NSF; National Science Foundation; Division of Atmospheric and Geospace Sciences; Directorate for Geosciences; Paleontology; Atmospheric Science; Oceanography; Arctic; Drake Passage; Sea ice
• Online Access: https://www.openaccessrepository.it/record/92003; https://doi.org/10.1029/2020pa003957

AbstractResults are presented and compared for the Community Earth System Model version 2 (CESM2) simulations of the middle Holocene (MH, 6 ka) and Last Interglacial (LIG, 127 ka). These simulations are designated as Tier 1 experiments (midHolocene and lig127k) for the Coupled Model Intercomparison Project phase 6 (CMIP6) and the Paleoclimate Modeling Intercomparison Project phase 4 (PMIP4). They use the low‐top, standard 1° version of CESM2 contributing to CMIP6 DECK, historical, and future projection simulations, and to other modeling intercomparison projects. The midHolocene and lig127k provide the opportunity to examine the responses in CESM2 to the orbitally induced changes in the seasonal and latitudinal distribution of insolation. The insolation anomalies result in summer warming over the Northern Hemisphere continents, reduced Arctic summer minimum sea ice, and increased areal extent of the North African monsoon. The Arctic remains warm throughout the year. These changes are greater in the lig127k than midHolocene simulation. Other notable changes are reduction of the Niño3.4 variability and Drake Passage transport and a small increase in the Atlantic Meridional Overturning Circulation from the piControl to midHolocene to lig127k simulation. Comparisons to paleo‐data and to simulations from previous model versions are discussed. Possible reasons for mismatches with the paleo‐observations are proposed, including missing processes in CESM2, simplifications in the CMIP6 protocols for these experiments, and dating and calibration uncertainties in the data reconstructions.