Diagnosing relationships between mean state biases and El Niño shortwave feedback in CMIP5 models
Journal of Climate
American Meteorological Society
© 2017 American Meteorological Society
Reason for embargo
Currently under an indefinite embargo pending publication by American Meteorological Society of Version of Record. On publication of VoR, this AAM to be replaced with the VoR under a 6 month embargo
The rate of damping of tropical Pacific sea surface temperature anomalies (SSTAs) associated with El Niño events by surface shortwave heat fluxes has significant biases in current coupled climate models (Coupled Model Intercomparison Project Phase 5; CMIP5). Sixteen of 33 CMIP5 models have shortwave feedbacks that are weakly negative in comparison to observations, or even positive, resulting in a tendency of amplification of SSTAs. Two biases in the cloud response to El Niño SSTAs are identified and linked to significant mean state biases in CMIP5 models. First, cool mean SST and reduced precipitation are linked to comparatively less cloud formation in the eastern equatorial Pacific during El Niño events, driven by a weakened atmospheric ascent response. Second, a spurious reduction of cloud driven by anomalous surface relative humidity during El Niño events is present in models with more stable eastern Pacific mean atmospheric conditions, and more low cloud in the mean state. Both cloud response biases contribute to a weak negative shortwave feedback, or a positive shortwave feedback that amplifies El Niño SSTAs. Differences between shortwave feedback in the coupled models and the corresponding atmosphere-only models (AMIP) are also linked to mean state differences, consistent with the biases found between different coupled models. Shortwave feedback bias can still persist in AMIP, as a result of persisting weak shortwave responses to anomalous cloud and weak cloud responses to atmospheric ascent. This indicates the importance of bias in the atmosphere component to coupled model feedback and mean state biases.
This work was supported by the UK-China Research & Innovation Partnership Fund through the Met Office Climate Science for Service Partnership (CSSP) China as part of the Newton Fund. MC acknowledges additional support from the Natural Environment Research Council grant number NE/N018486/1. H-L Ren is supported by the China Meteorological Administration Special Public Welfare Research Fund (GYHY201506013) and the Project for Development of Key Techniques in Meteorological Forecasting Operation (YBGJXM201705).
This is the author accepted manuscript.
Awaiting full citation