Doppler Lidar-Driven CFD Simulation for High-Density Urban Areas: Validation with Near-Ground Measurements in Singapore

Assessing urban ventilation plays an increasingly important role in supporting urban planning/design, especially for high-density cities suffering from poor wind conditions and an intensive urban heat island effect. Many design and technical guidelines have been developed to assess the impact of building clusters on outdoor natural ventilation and provide mitigation strategies (Franke, 2006; Tominaga, 2008; Ng, 2009; BCA, 2012).

However, most technical guidelines have limitations in dealing with input wind profiles for numerical simulations. The current ways to estimate wind profiles in the urban boundary layer, e.g., regional scale wind tunnel, weather research and forecasting (WRF) model, and empirical models, are not accurate enough to run CFD simulation, and often lead to significant errors (He, et al., 2022 (a) and (b)).

To address this issue, this study applied Doppler LiDAR technology to measure the incoming wind profiles approaching high density urban areas and used these measured wind profiles as boundary conditions to drive CFD simulation. Hourly-averaged wind profiles were applied, and the CFD simulation model covered the high-density and highly heterogeneous urban areas in Singapore, as shown in Figure 1. Ultra-sonic wind anemometers were deployed to collect near-ground wind velocity data for validation. As shown in Figure 2, the validation results indicate that the Doppler LiDAR driven CFD simulation is accurate, with only very minor deviation compared with real measurement, and much more accurate than other traditional ways, e.g., WRF, Logarithmic law, and power law.

Figure 1. Test bed at downtown areas in Singapore, for LiDAR boundary layer wind measurement, Ultra-sonic near ground wind measurement, and CFD simulations.

Figure 2. Cross-comparison between real near-ground measurements and CFD simulation results driven by different incoming wind profiles (e.g., LiDAR, Power Law, WRF, Logarithmic Law).