HD(CP)2 (2nd phase): Flux heterogeneity and boundary layer circulations
|Leitung:||Siegfried Raasch, Björn Maronga|
This project is a sub-project within the module S4 "Land surface heterogeneity" of the 2nd phase of the BMBF joint project "HD(CP)² - High definition clouds and precipitation for advancing climate prediction".
In nature, atmospheric state variables are modified by surface flux variations at all scales, which are induced by dynamic (vegetation growth, land use management) surface heterogeneities, and which feedback on larger scale atmospheric boundary layer (ABL) fluxes. ABL circulations driven by these heterogeneities modify the surface fluxes resulting in two-way feedbacks and impacts on even extended circulations. Sub grid-scale heterogeneity of land surfaces in climate models is usually tackled by the tile approach, which assumes a homogeneous atmosphere interacting with a heterogeneous land surface. In this approximation, the atmosphere is solely driven by grid-averaged fluxes at the surface. The work package will quantify the effect of climate model sub-grid flux variability including induced boundary circulations within the scale of the climate model grid cell and their feedbacks on the fluxes on the uncertainty of regional climate predictions. In contrast to the other work packages, the main intention is to explicitly resolve the near-surface processes both in order to gain a better physical understanding, and to quantify errors in the existing ICON surface parameterizations and suggest improvements.
While the LES focus in phase 1 of HD(CP)² was on a spatial (ICON patch size) scale of 100 m in order to resolve the grey zone scales, surface layer processes take place on even smaller scales and their explicit simulation requires grid spacings down to 10 or even 1 m. Only this scale allows e.g. the envisaged proof of the validity of Monin-Obukhov similarity theory (Maronga, 2014), explicit simulations of nocturnal boundary layer processes (Beare et al., 2006), or effects of sub-patch size surface heterogeneities on the boundary layer turbulence structure and transport (Heinze et al., 2015). Results of these simulations will help to improve the respective ICON-GCM parameterizations. The central tool of the work package will be the LES models PALM and ICON-LES. PALM will be used in addition to ICON-LES since most of the intended studies require a very fine grid spacing more than two orders of magnitude below the envisaged ICON-LES resolution of 100 m. This grid spacing will allow for resolving even smallest-scale heterogeneities of size of a few decametres. A comparison between PALM with ICON-LES runs will allow for identifying weaknesses of current ABL parameterizations regarding the effect of surface heterogeneity in ICON-NWP/GCM. Moreover, a comparison of PALM will allow for a rather direct evaluation of the ICON-LES, which has only been used for simple and very idealized simulations during the 1st phase of HD(CP)². Unlike previous LES simulations on the effect of surface heterogeneity, PALM will make use of a fully coupled land surface model (LSM) that has been recently implemented. This work package will imply comparative runs of PALM with ICON-LES at its envisaged resolution of 100 m in order to determine the extent to which climate prediction uncertainty depends on the knowledge and accurate parameterisation of small-scale land surface heterogeneity. The final goal of the proposed research actions is to contribute to the design of appropriate parameterizations for ICON-NWP/GCM.