The Relative Impact of SST Pattern Compared to Model Physics on Climate Sensitivity
Abstract
In this study we attempt to understand the differences in effective climate sensitivity (eff-CS) between a GFDL CMIP5 class model (AM3) and a new GFDL CMIP6 class model (AM4). To achieve this we subject both atmospheric models to sea surface temperature (SST) and sea ice patterns that are representative of those seen at the time of CO2 doubling following a percent per year increase (i.e. a TCR pattern) and those seen once full radiative equilibrium is achieved (i.e. an ECS pattern). We have performed detailed offline Partial Radiative Perturbation (PRP) calculations so to allow us to isolate which geographic regions and atmospheric components are responsible for the eff-CS obtained by each model.
When given the TCR SST pattern AM4 has a much lower eff-CS than AM3 (2.0 K vs. 3.2 K). We mainly attribute this to the ability of AM3 to maintain both a positive longwave and shortwave cloud feedback in the tropics, while in AM4 the longwave and shortwave cloud feedbacks cancel resulting in a near-zero net cloud feedback. Both models show a strong dependence of eff-CS on the SST pattern. When subjected to the ECS pattern, AM3 eff-CS is seen to increase by 60% relative to the TCR pattern from 3.2 K to 5.0 K. Similarly, the AM4 eff-CS increases by 70% from 2.0 K to 3.4 K. In the tropics the GCMs achieve this change in eff-CS through very different means: AM3 exhibits a positive shift in lapse rate feedback, but very little shift in cloud feedback, while in AM4 the change in eff-CS is almost solely due to change in shortwave cloud feedback. However, over the Southern Ocean, both models show very similar behavior of a large increase in the shortwave cloud feedback when given the ECS compared the TCR pattern. We show how an understanding of the interplay between the model physics parameterizations and SST pattern can help explain the differences and similarities of the two model responses.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.A21H2789P
- Keywords:
-
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSESDE: 3310 Clouds and cloud feedbacks;
- ATMOSPHERIC PROCESSESDE: 3337 Global climate models;
- ATMOSPHERIC PROCESSESDE: 1620 Climate dynamics;
- GLOBAL CHANGE