Large-eddy simulation resolving buildings and tree crowns for urban micrometeorology
Keigo Matsuda (JAMSTEC)
Séminaire du LMD à l’ENS.
Urban high temperatures due to the combined influence of global warming and urban heat islands increase the risk of heat stroke. Greenery is one of possible countermeasures for mitigating the heat environments since the transpiration and shading effect of trees can reduce the air temperature and the radiative heat flux. In order to formulate effective measures, it is important to estimate the influence of the greenery on the heat stroke risk considering micrometeorological processes. In this study, we have developed a tree-crown resolving large-eddy simulation (LES) model that is coupled with three-dimensional radiative transfer (3DRT) model. The multiscale atmosphere-ocean coupled model, MSSG (Multi-Scale Simulator for the Geoenvironment), is used for performing the micrometeorology LES. The MSSG is capable of performing multiscale simulations from the global and mesoscales to micrometeorological scales, resolving the topography, building shapes, and tree crowns of several meters. The 3DRT model is implemented in the MSSG so that the 3DRT is calculated repeatedly during the time integration of the LES.
The developed model is applied to the analysis of the heat environment in an actual urban area around the Tokyo Bay area, covering 8 km × 8 km with 5-m grid mesh, in order to confirm its feasibility. The results show that the wet-bulb globe temperature (WBGT), which is an indicator of the heat stroke risk, is predicted in a sufficiently high accuracy to evaluate the influence of tree crowns on the heat environment. In addition, by comparing with a case without the greenery in the Tokyo Bay area, we have confirmed that the greenery increases the low WBGT areas in major pedestrian spaces by a factor of 3.4. The model is then applied to the heat environment analysis in the Kumagaya Sports & Culture Park, where field observation was carried out in 2016. For the simulation, the multiscale downscaling technique is used to obtain reliable results. The finest domain covers 3 km × 3 km with 2-m grid mesh. The comparison with the field observation data confirms that the present model can predict the local difference of the wind speed, air temperature, and WBGT. These results indicate that the present model can predict the greenery effect on the urban heat environment.
ENS – salle Lien Hua – E350 • 24, rue Lhomond 75005 PARIS