Multi-Scale Interaction In Idealized Walker Cell Simulated With Sparse Space And Time Super-Parameterization

Interactions between large-scale flow and convection are a crucial aspect of the tropical atmosphere. However, the wide discrepancy in spatial and temporal scales involved makes such interactions impractically expensive to resolve numerically. Recently, the Sparse Space and Time-Super Parameterization (SSTSP; Xing et al. 2009) algorithm has been proposed to address this issue and has been shown to capture the propagation of a squall line. In the SSTSP framework, the embedded cloud-resolving model (CRM) is used to simulate convection on a domain whose area is small in comparison to the resolution of the large-scale model and for a time-period that is short when compared to the time-step in the global model. To compensate, extrapolation of the dynamics of the small scales over a larger domain and longer time scales is required. In principle, SSTSP offers significant computational benefits, but it is uncertain if it leads to correct numerical solutions when applied to a variety of convective situations in the tropics.  

Here, the accuracy of the SSTSP algorithm, implemented within the EULAG model (Smolarkiewicz and Margolin 1997; Prusa et al. 2008), is investigated in the context of an idealized Walker cell, simulated over a 40000-km domain and over 300 days. Simulation using the EULAG model in a CRM configuration provide benchmark for simulations with the SSTSP algorithm. CRM simulation of the Walker cell exhibit low-frequency variability (~20 days) tied to large-scale fluctuations of water vapor and zonal wind. The low-frequency pattern in the CRM is coupled with a change in time of mesoscale convective organization. The SSTSP algorithm is shown to be successful at reproducing low-frequency variability for relatively large values of spatial and temporal accelerations. The ability of SSTSP to capture the coupling of convection with the large-scale and its fluctuation is demonstrated, in particular in the context of the distribution of precipitation, properties of convective systems and their propagation characteristics.