Attributing the Inter-model spread in CMIP5 Seasonal Cloud Amount
Abstract
CMIP5 models predict rapid Arctic climate change between 4 - 14 K by 2100 and simulate large inter-model spread in cloud amount (CA). Arctic clouds have a crucial impact on climate variability and change by modulating surface energy fluxes, temperature, and sea ice conditions, and thus indirectly surface turbulent fluxes and boundary-layer stability. Clouds and radiative fluxes are also influenced by surface conditions such as sea ice, open water, melt ponds, and snow. Despite the importance of clouds, contemporary climate models strongly disagree on CA and its seasonality. The goal of this analysis is to unravel the contributions to the large model bias, and attribute model differences to physical processes. Reanalysis data from MERRA2 are used to assess the fidelity of CMIP5 model simulations of atmospheric conditions and observations from C3M provide the observed vertically-resolved seasonal CA. We find that models fall into one of two robust, well-defined groups: one group simulating maximum CA in winter driven by large-scale dynamics, and one group simulating maximum CA during summer with too large a seasonal amplitude in CA. Next, a compositing scheme is used to bin CA by various cloud-influencing factors (CIF) to explain the distinct behavior of the two model groups. The CIF chosen each play a crucial role in cloud formation and include: 500-hPa vertical velocity and lower tropospheric stability (representative of different atmospheric states), surface latent/sensible heat and sea ice concentration (representative of surface-atmosphere coupling), profiles of ice and liquid water content (representative of cloud microphysics), and the vertical profiles of temperature and relative humidity. The covariance between CIF is also explored. Finally, a decomposition methodology is used to attribute inter-model differences in CA to a model's sensitivity to atmospheric state and the frequency of occurrence of various atmospheric states.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.A51P2433B
- Keywords:
-
- 3311 Clouds and aerosols;
- ATMOSPHERIC PROCESSESDE: 3339 Ocean/atmosphere interactions;
- ATMOSPHERIC PROCESSESDE: 0738 Ice;
- CRYOSPHEREDE: 0750 Sea ice;
- CRYOSPHERE