TB2 Urban Environment 3

Akashi Mochida, Hiroshi Yoshino, Tatsuaki Iwata, Yuichi Tabata, Rong Wan, Jiaping Liu, Zhiwei Wang, Cheng‐Hsin Chang, Jin-Shian Lin, Chii‐Ming Cheng, Yuh‐Yi Lin, Tsuyoshi Nozu, Tetsuro Tamura, Yasuo Okuda, Satoshi Sanada, Hai Jin, Wei Yang, Xinyang Jin, Dailin Chen, Thomas R. Scott, David BanksWind 2006 · DOI

Introduction

As urban density increases and buildings grow taller, the phenomenon of "wind canyons" and dangerous street-level gusts has become a significant concern for pedestrian safety and comfort. While planting trees is a widely adopted strategy to mitigate these effects, urban planners require precise computational tools to predict their impact accurately. This paper, originally presented at the Fourth International Symposium on Computational Wind Engineering (CWE2006), explores the optimization of mathematical canopy models used in Computational Fluid Dynamics (CFD) to simulate how trees alter urban wind patterns.

Computational Modeling of Tree Canopies

Modeling individual leaves and branches in a large-scale urban simulation is computationally prohibitive. Instead, researchers employ "canopy models" that treat tree clusters as porous zones with specific mathematical properties. The standard approach utilizes the $k-\epsilon$ turbulence model, modified with additional terms to account for the physical presence of the trees.

Configuration of tree and inflow condition for the CFD simulation.

These modifications include:

  • Momentum Equation: A drag term to simulate how the tree slows wind velocity.
  • Turbulence ($k$) and Dissipation ($\epsilon$) Equations: Terms to represent the generation of turbulence within the canopy and the subsequent breakdown of that energy.

The study categorized existing models into four types (A through D), focusing specifically on optimizing two critical turbulence coefficients: $C_{p\epsilon1}$ and $C_{p\epsilon2}$. These coefficients control the time scale of energy dissipation, which is vital for capturing the complex wake patterns behind vegetation.

Validation and Methodology

To ensure physical reality, the numerical simulations were validated against high-quality field data from Kurotani et al. (2002). This dataset provided measurements of wind velocity and turbulent energy ($k$) in the wake of real pine trees in Izumo, Japan. By systematically varying the $C_{p\epsilon1}$ and $C_{p\epsilon2}$ coefficients in "Type B" and "Type C" models, the researchers sought the optimal mathematical configuration to match the observed field data.

Results and Discussion

The research identified that the "Type C" model, which includes an additional "sink term" to account for energy loss within dense canopies, was superior to other formulations. A key finding was that the standard coefficient values used in many previous studies were insufficient for capturing turbulence accurately.

Comparison of vertical profiles of turbulent energy (k) behind the tree for the Type C model.

The "Aha" moment occurred when the coefficient $C_{p\epsilon1}$ was increased to 1.8 (up from the standard 1.5). This adjustment allowed the model to align closely with the complex wake patterns observed in nature. Specifically, the Type C model using $C_{p\epsilon1} = 1.8$ and $C_{p\epsilon2} = 1.5$ provided the best overall agreement with field measurements for both wind velocity and turbulent energy. While velocity profiles could be reproduced by both Type B and Type C models when $C_{p\epsilon1}$ was in the 1.8–2.0 range, the Type C model was uniquely effective at capturing the turbulence energy critical for pedestrian comfort assessments.

Conclusion and Future Work

This study demonstrates that fine-tuning turbulence coefficients and incorporating sink terms significantly improves the accuracy of tree canopy models in urban CFD. However, limitations remain. The current work focused on 2D calculations and specific pine tree species. Future research should expand these models to 3D environments to account for wind flowing around tree clusters and investigate species-specific leaf area densities. Additionally, incorporating thermal effects like shading and evapotranspiration would provide a more holistic view of the urban microclimate.

Summary updated

BibTeX
@inproceedings{mochida2006tb,
  author = {Akashi Mochida and Hiroshi Yoshino and Tatsuaki Iwata and Yuichi Tabata and Rong Wan and Jiaping Liu and Zhiwei Wang and Cheng‐Hsin Chang and Jin-Shian Lin and Chii‐Ming Cheng and Yuh‐Yi Lin and Tsuyoshi Nozu and Tetsuro Tamura and Yasuo Okuda and Satoshi Sanada and Hai Jin and Wei Yang and Xinyang Jin and Dailin Chen and Thomas R. Scott and David Banks},
  title = {{TB2 Urban Environment 3}},
  booktitle = {Wind},
  year = {2006},
  doi = {10.5359/jawe.2006.561},
}
Figure