The research investigates the potential of adapting vernacular urban morphologies known for their inherently efficient microclimates in novel urban development. Out of the 20 case studies under investigation, 5 matched the conditions of our site in Portland (Csb, "cool-dry summer" under the Koppen-Geiger climate classification scheme): Casbah of Algiers, the Greek island of Hydra; and Alberobello, Anticollo Corrado, and Positano in Italy. The design principles of the aforementioned sites, street condition, void condition (creation of plazas in the grid by removing building volumes from the grid) and volumetric roughness (expressed by rotation of the building masses around their axes and by asymmetrical walls and facades) were all applied iteratively to the building volumes of our site, and tested for climatic performance through UTCI and CFD analysis. The results were then compared to each other to determine how such operations on an urban development affect environmental performance within the grid.
I performed UTCI analysis of the city grid after applying each design principle in an iterative manner. By focusing on the two most important aspects of UTCI -sky heat exchange and wind speed within the city grid, I narrowed down the simulations to sky heat transfer benchmarking with Grasshopper Ladybug's radiation analysis suite (Fig. 3), and wind speed in the various geometries through ANSYS CFD analysis (Fig. 5). Using climatic data for Portland and the temperature and wind speed values at important locations within the development (plazas and key streets), I translated the simulated values into UTCI. I also translated the wind speed values into a unitless vorticity magnitude, which was compared against the UTCI to determine the wind effects on average UTCI (Table 1). Findings showed that the iterative operations on the urban morphology decresed the average UTCI values and placed it within the "no heat stress" category, when compared to the initial geometry.
According to Portland TMY data, summer months are characterized by the following averages: air temperature (Ta) at 19.21°C, mean radiant temperature (Tmrt) at 35.32°C, and relative humidity at 66.92%. and air speed at 3.4 m/s. The simulation was ran in a x=2090.6m, y=776.6m and z=65.0m chamber with resolutions dx = dy = dz = 1.98m in space. These maps of UTCI describe the thermal diversity driven only by the presence of direct sun or shade on pedestrians. These maps neither account for spatial differences in surface temperature across the urban area nor account for spatial differences in wind speed.
In order to account for the effect of wind speed on outdoor comfort, ANSYS Fluent was used to produce a second set of spatial maps. Portland TMY data determined the summer average wind speed and direction: 3.4m/s at NNW 112.5°, according to a true north-based azimuth. The simulation was ran in a x=2090.6m, y=776.6m and z=65.0m chamber with resolutions dx = dy = dz = 1.98m in space. From the CFD analyses, the wind speed for select points within each of the four plazas across all morphology iterations was chosen to determine the most relevant locations to the site in question. Since the CFD simulations were run for the center of gravity of an adult human (1.1m), the velocities were “back-converted” to meteorological height (10m) in order to calculate UTCI values (Jendritzky et al., 2007). After multiplying the simulation velocities by 1.5, the meteorological height wind speeds were then used to calculate UTCI values.
This methodology provides a computationally inexpensive and reliable approach in determining UTCI in relatively complex city grids, and showcases the importance of ancient vernacular urban morphologies in modern city development.
Data for the UTCI calculations and the Grasshopper components was sourced from: https://klimaat.ca/epw/
Andrew Heid, Christopher Purpura, Theo Dimitrasopoulos.
Rhinoceros, Grasshopper, LadyBug suite, AutoCAD, V-Ray, CFD OpenStudio, ANSYS CFD.