Stochasticity and Organization of Tropical Convection through a Multicloud Lattice Model

Despite the many recent improvements in the numerical representation of atmospheric convection, state of the art climate models still present serious biases in mean tropical rainfall and climate variability associated with organized tropical convection. The main problem lies in the inadequacy of the deterministic response of the underlying convection parameterization schemes to the large scale atmospheric state, restricting the development of convective variability in the models. In order to overcome this limitation, Khouider et al. (2010) proposed a stochastic lattice multicloud model (SMCM) to represent the sub-grid variability in GCM's due to tropical convection. The SMCM tracks the evolution of the area fractions of three cloud types: congestus, deep convective, and stratiform, based on the occupation of the lattice sites by one of the cloud types or clear sky and dynamical interactions with each other and with the environment. By design the SMCM can be easily coupled to existing (deterministic) parameterization and used as a stochastic trigger for convection. Moreover, the evolving stochastic area fractions can also be used to inform the radiation scheme. In order to make the SMCM compatible with conventional spectral mass-flux convection schemes (e.g. Zhang and McFarlane 1995, ZM), we expand the SMCM modeling framework to represent the dynamical interactions of shallow, congestus, and deep convective cloud types to constrain the vertical structure distribution of the plume ensemble. As a first step toward the implementation of the SMCM in the ZM scheme, here we test the stochastic model against observations. Namely, we choose two geographic points, one in the western Pacific and one in Indian ocean and find the model parameters that best produce the stochasticity and organization of convection in observations. One key question is to determine to what extend the model parameters are universal throughout the tropics and extensions of the current study to further stations is warranted.

References:

Khouider B, Biello J, Majda AJ (2010) A stochastic multicloud model for tropical convection. Communications in Mathematical Sciences 8(1):187–216 Zhang, G. J. and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Center general circulation model. Atmos. Ocean, 33 (3), 407–446.