Upscale feedbacks through microphysics fields at nesting domains of the MM5 model

Multi-scale interactions and associated physical processes are common features of the atmospheric flow. Their spatial and temporal behaviour presents a challenge to understand and predict the state of the flow at finer and broader scales. The objective of this study is to focus on capabilities of various parameterisation schemes in the MM5 model to simulate interactions and feedback system at nesting domains. Two-way nesting allows us to track the occurrence and consequence of affecting smaller scale processes explicitly produced at a finer resolution domain to larger scales at a coarser resolution. The quantitative estimations of model errors in two-level nesting domains are compared against the similar ones in a single-level domain. Various combinations of parameterisation schemes for cumulus, PBL, moisture and radiation are used to identify the one that provides the least difference between the model state and reanalysis ERA40 that is considered as the reference state. Basic attributes of model errors are measured to determine spatial structures, vertical profiles and synoptical patterns. The results show that feedbacks from finer to larger scales usually lead to better behaviour in the simulated state. However, this is mainly true for the atmospheric properties characterized by smooth patterns with large scale structure functions such as geopotential and temperature. On the contrary, the humidity model error in the nesting mode is sensitive to the choice of a parameterisation scheme. In some cases feedbacks from finer scale processes simulated in the nesting domains toward larger scales lead to an increase in the model error. This effect is especially remarkable for humidity fields in the middle troposphere.