Search

A realistic representation of processes that are not resolved by the model grid is one of the key challenges in Earth-system modelling. In particular, the non-linear nature of processes involved makes a representation of the link between the atmosphere and the land surface difficult. This is especially so when the land surface is horizontally strongly heterogeneous. In the majority of present day Earth system models two strategies are pursued to couple the land surface and the atmosphere. In the first approach, surface heterogeneity is not explicitly accounted for, instead effective parameters are used to represent the entirety of the land surface in a model's grid box (parameter-aggregation). In the second approach, subgrid-scale variability at the surface is explicitly represented, but it is assumed that the blending height is located below the lowest atmospheric model level (simple flux-aggregation). Thus, in both approaches the state of the atmosphere is treated as being horizontally homogeneous within a given grid box. Based upon the blending height concept, an approach is proposed that allows for a land-surface-atmosphere coupling in which horizontal heterogeneity is considered not only at the surface, but also within the lowest layers of the atmosphere (the VERTEX scheme). Below the blending height, the scheme refines the turbulent mixing process with respect to atmospheric subgrid fractions, which correspond to different surface features. These subgrid fractions are not treated independently of each other, since an explicit horizontal component is integrated into the turbulent mixing process. The scheme was implemented into the JSBACH model, the land component of the Max Planck Institute for Meteorology's Earth-system model, when coupled to the atmospheric general circulation model ECHAM. The single-column version of the Earth system model is used in two example cases in order to demonstrate how the effects of surface heterogeneity are transferred into the atmosphere, influencing local stability and the turbulent mixing process. Furthermore, a simple flux-aggregation scheme was implemented into the JSBACH model. By comparing single-column simulations utilizing the VERTEX scheme and the simple flux-aggregation scheme, it can be shown that the horizontal disaggregation of the turbulent mixing process has a substantial impact on the simulated mean state of a grid box. Here, the differences between simulations with the two schemes may, in certain cases, be even larger than the differences between simulations with the simple flux-aggregation scheme and simulations in which surface heterogeneity is not explicitly accounted for (i.e., a parameter-aggregation scheme). [ABSTRACT FROM AUTHOR]

The increase in built surfaces constitutes the main reason for the formation of the Urban Heat Island (UHI), that is a metropolitan area significantly warmer than its surrounding rural areas. The urban heat islands and other urban-induced climate feedbacks may amplify heat stress and urban flooding under climate change and therefore to predict them correctly has become essential. Currently in the COSMO model, cities are represented by natural land surfaces with an increased surface roughness length and a reduced vegetation cover, but this approach is unable to correctly reproduce the UHI effect. By increasing the model resolution, a representation of the main physical processes that characterize the urban local meteorology should be addressed, in order to better forecast temperature, moisture and precipitation in urban environments. Within the COSMO Consortium a bulk parameterization scheme (TERRA_URB or TU) has been developed. It parametrizes the effects of buildings, streets and other man-made impervious surfaces on energy, moist and momentum exchanges between the surface and atmosphere, and additionally accounts for the anthropogenic heat flux as a heat source from the surface to the atmosphere. TU implements an impervious water-storage parameterization, and the Semi-empirical Urban canopy parametrization (SURY) that translates 3D urban canopy into bulk parameters. This paper presents evaluation results of the TU scheme in high-resolution simulations with a recent COSMO model version for selected European cities, namely Turin, Naples and Moscow. The key conclusion of the work is that the TU scheme in the COSMO model reasonably reproduces UHI effect and improves air temperature forecasts for all the investigated urban areas, despite each city has very different morphological characteristics. Our results highlight potential benefits of a new turbulence scheme and the representation of skin-layer temperature (for vegetation) in the model performance. Our model framework provides perspectives for enhancing urban climate modelling, although further investigations in improving model parametrizations, calibration and the use of more realistic urban canopy parameters are needed. [ABSTRACT FROM AUTHOR]