Land Surface Heterogeneity

Project S4 will assess and reduce the uncertainties of climate models caused by non-resolved subsurface processes, surface heterogeneity and its feedback to the atmospheric boundary layer (ABL) including cloud development and convection initiation.
Using the HD(CP)² framework, the project will analyse the true impact of the land including sub-surface, vegetation and anthropogenic structures including its variability and change on the regional climate of central Europe with a focus on cloud and precipitation development, intensity, and distribution. Questions that will be assessed are:

  • What is its impact on cloud and precipitation development?
  • What is the effect of catchment-scale circulations, which evolve in structured terrain, on clouds and precipitation and on the local and regional climate including wind extremes and fog development?
  • How different was our climate before humankind started to dominate the landscape, and what part did related land surface changes ranging from deforestation to water management and urban development actually play in this?

The scientists in S4 will answer these questions following an extended validation of the ICON-LEM with state of the art land models that include a consistent representation of the terrestrial water and energy cycle. Finally, appropriate methods will be developed to bring the findings back to the ICON-NWP/GCM model development.

The figure shows the structure of the S4 project and how the individual work packages are related.

Flux heterogeneity and boundary layer circulations

In addition to different land surface properties, cloud shadows significantly modify surface fluxes. Credit is given to Fabian Hoffmann for taking this photo of clouds and their shadows over the Gobi desert.

In work package 1 it will be quantified to what extend the Monin-Obukhov similarity-theory can be applied over heterogeneous terrain. For this, high-resolution large-eddy simulations (LES) with grid spacings of 10m down to 1m will be used. This allows to explicitly resolve the surface layer instead of parametrizing it. Unlike previous studies, a land-surface model (LSM) will be employed together with the LES. Furthermore, it will be accounted for flux heterogeneities induced by cloud shadows. Thereafter, it is planned to investigate the impact of terrain-induced circulations on the nocturnal boundary layer. The starting point will be in examining spacial requirements for LSMs which are coupled to LES.

The influence of the soil-vegetation continuum on land-surface-atmosphere feedback

Work package 2 addresses the consistent simulation of water and energy fluxes at the land surface focusing on agricultural landscapes and their interaction with the evolution and the thermodynamic structure of the atmospheric boundary layer (ABL). About 50% of the land surface in Germany and Europe is determined by agricultural activities, which highlights the importance of its appropriate representation in land surface models.The performance of the land surface model of ICON_LEM, ICON_NWP (ICON in numerical weather prediction mode), and TERRA will be evaluated and improvements with respect to the simulation of soil moisture and surface flux partitioning will be suggested. It is also investigated whether an advanced parametrization of vegetation dynamics on the daily and annual time scales can be incorporated in the TERRA model. This is preferably extracted from JSBACH (the land surface model of the ICON-General Circulation Model).
In this work package, the coupling between surface fluxes due to heterogeneities and their extensions into the ABL are studied and the performance of turbulence parametrization evaluated. Studies are performed in strong cooperation with work package 1 and with studies on ABL clouds in project S2. Moreover, findings also feed into the upscaling methodology developed in work package 6.

Catchment-scale circulations

Work package 3 will assess how land-surface heterogeneity impacts on convection initiation, cloud and precipitation development. Land surface and subsurface flow, soil moisture redistribution and ground water flow, feedback between soil moisture and land surface heterogeneity, and their feedback with circulation pattern in catchment-scale will be studied. To do this, ICON_LEM will be coupled to CLM-ParFlow, which is the advanced land surface model, via the OASIS3-MCT coupler. ICON_LEM simulations will be compared to observed data sets, to understand to what extent land surface heterogeneity-induced circulation impacts on clouds and precipitation development and initiation of extreme events.

The figure is showing flashes over Germany and joining areas for a certain period. This gives an indication of hot spot regions of convection and will help to choose areas of investigation.

Impact of ABL Heterogeneity/Troposhere Coupling on Convection

Convection initiation and evolution partially result from the subtle interplay between local variations in convection-related parameters with the atmoshperic boundary layer (ABL) and mid-tropospheric conditions and flow. These variations may originate from horizontal surface flux gradients caused by for instance land use and/or root zone soil moisture gradients and the orography. A key role in the process chain from land-surface variations, ABL heterogeneity and convection is played by ABL-troposphere coupling - a process which occurs on different scales.
Work package 4 will provide a better understanding of this coupling, which is important for a better estimation of the occurrence and strength of convection and its influence on heavy precipitation events. This includes the quantification of ABL troposphere coupling and the connection between coupling and regions of convection initiation and intensity.

In our work we will answer the following questions:

  • How can we quantify the ABL-mid-troposphere coupling in ICON_LEM?
  • How important is an adequate representation of land surface heterogeneity to simulate this coupling?
  • What is the impact of model resolution on simulating the coupling?

Effects of cities on regional climate and their role in extreme event evolution

Mean nocturnal (20–24 h) urban heat island of Hamburg in summer, simulated by the model METRAS on a grid spacing of 250 m. Image taken from Schlünzen K. H. and Linde M. (2014): Berichte aus den KLIMZUG-NORD Modellgebieten, Band 4.

Cities and settlements in general have shaped the earth surface and created heterogeneities to a degree easily visible even from space. By their sheer extent, cities constitute the largest and most dynamic land surface heterogeneities and modify daily cycles of radiation balance, surface flux partitioning, and the hydrology including the effects of water/waste water management. Cities significantly penetrate the atmospheric boundary layer, thereby generating extended obstacles to atmospheric flow and the vertically extended heat and matter sources and sinks.
Besides affecting regional climate by generating feedback on clouds, precipitation formation, surface flux, and radiation budget in their surroundings, urban areas are themselves vulnerable to extreme events, e.g. flooding of urban drainage systems.

In the standard land surface model TERRA, as it is currently implemented into ICON_LEM, urban areas are parametrized in terms of a bulk scheme as a natural land surface category, with an increased surface roughness length, and reduced vegetation cover. However, this approach is not sufficient to account for all characteristics of urban areas that influence the atmosphere, like the vertical extent of cities, and the trapping of heat within building materials ("urban heat island") at night time (see Figure).
Within work package 5, the uncertainty caused be the current bulk representation of urban areas is quantified and modeling capability improved by the implementation of an urban canopy model into ICON_LEM. Furthermore, is it investigated how the feedback of urban areas on local and regional climate due to triggering of convection and extreme event initiation can be quantified. Another research question is the influence of the feedback of a future warmer climate with increased urbanization.

Detecting and Parametrizing effects of land heterogeneity

Land heterogeneity influences cloud and precipitation initiation and evolution, on the regional and on the climate scale.This is due to its strong local influence on partitioning of fluxes of carbon, water and energy. While computation challenges still exist to explicitly represent this heterogeneity in regional and climate models with finer grid resolution, its effects on boundary layer evolution needs to be better understood and parametrized for coarser grid resolutions currently being used. This heterogeneity is ubiquitous even when grids are moved to resolutions less than 1 km, such these parametrizations will always be essential.

Work package 6 investigates the effects of the non-resolved land surface heterogeneity in the ICON-NWP and ICON-GCM simulations over multiple observation sites. We hypothesize the existence of a dynamic linkage between the turbulent mixing length scales in the PBL schemes and the land surface heterogeneity, which could implicitly improve the land surface heterogeneity effects on the larger NWP/GCM scales. Using the ICON modeling platform over the multiple observation sites, a benchmark database will be generated and used to explore and develop the existence of such linkages.

Horizontal cross-section of vertical velocity at a certain height (3 hrs after start of simulation) for convectively induced rolls. The figures shows ensembles of idealized simulations of COSMO coupled with the land-surface model CLM for varying horizontal grid resolution (x-direction) and turbulent master length scales (y-direction).