RESEARCH

Stochastic Representation of the Convective Entrainment in the Atmosphere

Model simulations are highly sensitive to the formulation of the atmospheric mixing process or entrainment in the deep convective parameterizations used in their atmospheric component. Simulations using stochastic entrainment outperformed default model simulations, as inferred from multiple metrics associated with the global and regional regions.

Uncertainty quantification (UQ) in weather and climate models is required to assess the sensitivity of their outputs to various parameterization schemes and thereby improve their consistency with observations. We present an efficient UQ and Bayesian inference for the cloud parameters of the Atmosphere Model using surrogate models based on a polynomial chaos expansion. The inferred parameters suggest improvements in the global Climate Earth System Model (CESM2) simulations of the tropics and sub-tropics.

We carry out systematic analysis of biases over global as well as the regional regions to find the source of biases in GCMs. A clustering algorithm is used to group the CMIP5 models based on the degree of similarity in the global bias patterns. AMIP5 models were analyzed to conclude if the biases were primarily due to the atmospheric component or due to the oceanic component of individual models. Further analysis is carried in individual component/parameterizations to trace the source of bias in a particular group of models.