1
Urban-Scale Building Air Change Rate Estimation Using Corrected Wind Speeds and
Three-Zone Building Modeling
Yasemin Usta
1
, W. Stuart Dols
2
, Guglielmina Mutani
1*
1
Department of Energy, Politecnico di Torino, Torino 10129, Italy
2
National Institute of Standards and Technology, Gaithersburg, MD, 20899, USA
Email: guglielmina.mutani@polito.it
Presneted at 10th AIGE/IIETA International Conference and 20th AIGE 2025 Conference June 9–11, 2025 in Carpi, Modena, Italy
ABSTRACT
This study presents a scalable, time-efficient methodology for estimating building-specific air change rates by incorporating local
urban morphology and wind conditions — addressing the limitations of constant air change rate commonly used in urban energy
models. The presented methodology is used to develop a QGIS-based plugin (Quantum Geographic Information System) to
automate the integration of aerodynamic parameters—specifically, roughness length (z
0
)—derived from the Urban Multi-scale
Environmental Predictor, with a simplified three-zone lumped-parameter model and multizone airflow simulations using CONTAM.
Using a 5 × 5m analysis grid and a 1m resolution 3D built environment model, the plugin calculates façade-specific wind-speed
modifiers across 30° directional intervals. These modifiers are then applied in a three-zone CONTAM model to calculate building
air change rates with hourly weather conditions. The resulting air change rates were incorporated into an hourly energy consumption
model for space heating and validated against measured energy-use from residential buildings in Turin, Italy. Results show that
replacing constant air change rates with dynamic, site-specific estimates reduced the Mean Absolute Percentage Error by 11 % to
20 %. In particular, the error was reduced from 68% ± 4.1% to 48% ± 2.5% in December and from 49% ± 3.3% to 32% ± 2% in
January, corresponding to relative error reductions of approximately 30 % and 34 %. The proposed method demonstrates improved
accuracy in simulating building energy consumption taking into account the influence of an accurate shape of the urban environment
on air change rates; moreover, it offers a robust, automated framework for urban-scale assessments of ventilation, infiltration, and
energy performance.
Keywords: Air Change Rate; Building Infiltration; Urban Roughness; UMEP; Multizone Airflow; Energy Modeling;
QGIS Plugin
1. INTRODUCTION
Accurate estimation of building infiltration is essential for
reliable energy performance assessments. However, most
simulation tools mainly use constant air change rates (ACRs)
based on construction period or archetype categories,
overlooking both temporal variability and site-specific factors
like urban roughness and local climate. This simplification can
lead to significant errors in predicted infiltration rates and thus
energy use and cost [1].
In urban environments, wind speed distributions are
significantly affected by the presence and configuration of
buildings, resulting in notable variations in wind velocities
between inter-building spaces, urban canyons, and rooftop
levels compared to those observed in open, unobstructed areas.
[2]. Context-specific Computational Fluid Dynamics (CFD)
simulations can capture these effects, but their high
computational requirements make them impractical for large-
scale analyses [3, 4]. Instead, logarithmic wind-profile
corrections adjust reference wind speeds at façade heights
using two main aerodynamic parameters: displacement height
(z
d
) and roughness length (z
0
).
Empirical methods, such as Kanda et al.’s study [5] of over
100 large-eddy simulations in Tokyo and Nagoya, link z
d
and
z
0
to five geometric descriptors: mean and maximum building
heights, plan area index (PAI), frontal area index (FAI), and
building-height variability. This approach provides a practical
way to estimate aerodynamic parameters with acceptable
accuracy for urban-scale analyses. Alternatively, the Urban
Multi-scale Environmental Predictor (UMEP) [6, 7] plugin for
QGIS (Quantum Geographic Information System) automates
the derivation of these parameters across complex urban areas.
UMEP uses high-resolution 3D environment descriptions such
as digital surface models (DSMs)—which capture terrain plus
all above-ground features—and digital elevation models
(DEMs)—which represent the bare-earth surface—along with
land-cover data, and building characteristics to compute
associated morphometrics. These DSMs/DEMs, are typically
derived from LiDAR or satellite sources [8], and can be
imported directly into QGIS via GDAL tools or plugins [9].
UMEP then applies one of six roughness schemes (including
Kanda’s) to calculate z
d
and z
0
for each wind-direction interval.
To apply wind corrections efficiently at the urban scale,
building geometry must be simplified. In 2024, Santantonio et
al. [10] introduced a three-zone lumped-parameter model that
represents each building as two heated zones: upper and lower
apartments and an unheated shaft, enabling urban-scale
simulations of combined wind-driven and buoyancy-driven
infiltrations using corrected wind-speed inputs. Similarly,
multizone tools like CONTAM can perform pressure-driven
airflow simulations with leakage and weather inputs, but they
depend on user-defined external boundary conditions that are
typically input with limited consideration of the surrounding
environment [11].
Given the need for site- and time-specific ACRs, a
simplified workflow is essential. This study presents a QGIS-
based methodology for wind-speed correction using UMEP-
derived aerodynamic parameters. The workflow automates the
generation of building-specific wind-speed modifiers and
integrates them into a three-zone CONTAM model.
