buildings

Article
Urban Microclimate, Outdoor Thermal Comfort, and
Socio-Economic Mapping: A Case Study of Philadelphia, PA
Farzad Hashemi 1, * , Ute Poerschke 1 , Lisa D. Iulo 1 and Guangqing Chi 2

1 Department of Architecture, The Pennsylvania State University, University Park, PA 16802, USA
2 Department of Agricultural Economics, Sociology, and Education, Population Research Institute, and Social
Science Research Institute, The Pennsylvania State University, University Park, PA 16802, USA
* Correspondence: fxh99@psu.edu

Abstract: Urban areas are often warmer than rural areas due to the phenomenon known as the
“urban heat island” (UHI) effect, which can cause discomfort for those engaging in outdoor activities
and can have a disproportionate impact on low-income communities, people of color, and the elderly.
The intensity of the UHI effect is influenced by a variety of factors, including urban morphology,
which can vary from one area to another. To investigate the relationship between outdoor thermal
comfort and urban morphology in different urban blocks with varying social vulnerability status,
this study developed a geographic information system (GIS)-based workflow that combined the
“local climate zone” (LCZ) classification system and an urban microclimate assessment tool called
ENVI-met. To demonstrate the effectiveness of this methodology, the study selected two different
urban blocks in Philadelphia, Pennsylvania–with high and low social vulnerability indices (SVI)–to
compare their microclimate conditions in association with urban morphological characteristics such
as green coverage area, sky view factor (SVF), albedo, and street height to width (H/W) ratio. The
results of the study showed that there was a strong correlation between tree and grass coverage and
outdoor air and mean radiant temperature during hot seasons and extremely hot days, which in
turn affected simulated predicted mean vote (PMV). The effects of greenery were more significant in
the block associated with a low SVI, where nearly 50% of the site was covered by trees and grass,
Citation: Hashemi, F.; Poerschke, U.; compared to only 0.02% of the other block associated with a high SVI. Furthermore, the investigation
Iulo, L.D.; Chi, G. Urban discovered that reduced SVF, along with increased albedo and H/W ratio, had a beneficial impact on
Microclimate, Outdoor Thermal
the microclimate at the pedestrian level within the two studied urban blocks. This study provided an
Comfort, and Socio-Economic
effective and easy-to-implement method for tackling the inequity issue of outdoor thermal comfort
Mapping: A Case Study of
and urban morphology at fine geographic scales.
Philadelphia, PA. Buildings 2023, 13,
1040. https://doi.org/10.3390/
Keywords: urban heat island; outdoor thermal comfort; predicted mean vote (PMV); geographic
buildings13041040
information system (GIS); local climate zones (LCZ); social vulnerability index (SVI)
Academic Editors: Peng Du and
Rahman Azari

Received: 15 March 2023

Revised: 9 April 2023 1. Introduction
Accepted: 12 April 2023 In 2017, human-induced warming was about 1 ◦ C above pre-industrial levels and
Published: 15 April 2023 increasing at a rate of 0.2 ◦ C per decade. If the current rate continues, there is a strong
possibility of a further rise of 1.5 ◦ C in global temperatures between 2030 and 2052 [1]. With
55% of the world’s population living in urban areas in 2017, the UN world urbanization
prospect [2] estimates the proportion of the population living in urban areas will increase
Copyright: © 2023 by the authors.
to 68% (i.e., 2 of 3 people will live in urban areas) by 2050. As a result of urbanization, most
Licensee MDPI, Basel, Switzerland.
cities worldwide have experienced a significant conversion of natural landscapes to built
This article is an open access article
environments that are associated with low vegetation index, the emergence of dark surfaces
distributed under the terms and
conditions of the Creative Commons
with low albedo, increased heat waste from both buildings and traffic, and reduced sky
Attribution (CC BY) license (https://
view [3,4]. As a result of these land use conversions, air temperatures in urban areas are
creativecommons.org/licenses/by/
typically higher than their surrounding rural areas, which is termed the “urban heat island”
4.0/). (UHI) effect [5,6], one of the most evident anthropogenic impacts on urban microclimates.

Buildings 2023, 13, 1040. https://doi.org/10.3390/buildings13041040 https://www.mdpi.com/journal/buildings

Buildings 2023, 13, 1040 2 of 19

Increases in urban air temperature have a severe impact on building energy uses and may
endanger life and increase mortality rates.
UHIs have been observed globally in cities regardless of their locations and size [7–10].
The UHI effect and its direct consequence (i.e., heat stress in cities) cause severe discom-
fort in outdoor activities, especially for groups of people who are known to be socially
vulnerable [11]. From 2008 to 2017, extreme heat was associated with higher all-cause
mortality in the contiguous U.S., with a greater increase noted among older adults, men,
and non-Hispanic Black individuals [12]. More than 3800 death were recorded to be caused
by heat-related illness (HRI), and almost 23,000 people were hospitalized for HRI between
2016–2020 across the U.S. [13]. Overall, heat island hotspots and mortality rates are found
to be greater in urban blocks with a socially disadvantaged population [14–16]. According
to a recent study by Hsu et al. [17], the mean and dispersion of summer daytime Surface
UHI intensity are worse for both people of color and the poor compared to white and
wealthier populations in nearly all major U.S. cities.
Due to the significant impacts of UHIs on human health and comfort, several studies
have been conducted to evaluate these impacts during past decades in cities located in
different climate zones. Field observation and numerical models are the two main methods
used in these studies to investigate the impacts of the induced temperature on comfort sen-
sations. A field study was carried out in outdoor urban spaces of Singapore with a focus on
the human thermal sensation of urban dwellers from August 2010 to May 2011 [18]. Solar
radiation showed the most significant effect on human thermal sensation, and to improve
outdoor thermal comfort, a combination of lower-density urban areas with higher building
height was found to be an effective solution. During the summers and winters of 2008 and
2009, a field survey was conducted in two hot and arid climate cities-Marrakech, Morocco
and Phoenix, Arizona. The survey included structured interviews using a standard ques-
tionnaire, observations of human activities, and microclimatic monitoring [19]. They found
that the socioeconomic status and cultural background of attendees showed a significant
impact on their thermal sensation and comfort expectation. Noticeable differences between
the two cultures in the type of outdoor activities were also found. Field measurement and
surveys were employed by several other studies in cities such as Rome, Italy [20], Wuhan
and Xi’an, China [21,22], Belo Horizonte, Brazil [23], and Tempe, Arizona [24], to name a
few.
Due to the differences in urban fabrics in the examined cities, as well as the overall
efforts required to conduct field observations/surveys, this method imposes substantial
limitations in terms of the applicability of results and the required time and cost to ac-
complish similar studies. Hence, numerical models have become of significant interest
to planners, architects, and environmental assessors to simulate urban microclimate and
perform comparative analyses based on different scenarios; these methods often use spatial
overlay and simulation methods to combine both physical/climatic and socioeconomic
data [25,26]. Urban microclimate modeling tools, especially simulation with Computational
Fluid Dynamics (CFD), are used both to simulate UHI intensities and to estimate how
effective it would be to apply different mitigation measures. The yearly number of studies
on the topic has increased rapidly since 1998 [27]. In recent years, many studies have been
conducted on modeling urban microclimates using tools such as OpenFOAM, FLUENT,
STAR-CCM+, PHOENICS, and ENVI-met. Among these CFD-based models, ENVI-met
was found to be the most popular due to a balance of sophistication, user-friendliness, and
lower computational costs [28]. From whole microclimate simulation at a neighborhood
scale to evaluation of hourly outdoor thermal comfort, ENVI-met has been validated by
several studies [29–36]. Forouzandeh [37] compared the ENVI-met V4 outputs with values
measured experimentally under various seasonal conditions inside a courtyard located
in Hannover, Germany. The ENVI-met model was proved to be accurate in predicting
microclimate variables in medium-narrow courtyards with low root mean square error
(RMSE) values for temperature, humidity, and wind speed. RMSE values ranged between
0.73 ◦ C for air temperature, 3.34% for humidity, and 0.01 m/s for wind speed.
Buildings 2023, 13, 1040 3 of 19

ENVI-met’s capacity to estimate Physiological Equivalent Temperature (PET), a ther-

mal index, was evaluated [38] by comparing it to field measurements for various points in
a street canyon in Port Said, Egypt, throughout the summer and winter seasons. Except for
the limited air velocity correlation, the results presented a strong agreement, particularly
with respect to the final results of the PET, with an index of agreement value ranging from
0.81 to 0.95.
Existing literature adopted urban blocks or master plan developments, normally only
one in each study, for urban microclimate and/or outdoor thermal sensations studies.
Assessment of microclimate conditions in multiple urban blocks of a city for a meaningful
comparison of impacts caused by urban parameters and the effectiveness of heat mitigation
efforts remains limited. In addition, the areas being studied are typically modeled with
added details for urban parameters and building properties, which often involves a time-
consuming and resource-intensive process of collecting real data, including manpower and
funding. To overcome these disadvantages, a GIS-based methodology was developed and
used in this study to first identify urban blocks with different socioeconomic statuses in U.S.
cities and then model urban microclimate inside the blocks creating a set of generic inputs
eliminating the need for field data collection. The proposed workflow is replicable and
can be applied by researchers and practitioners, from novices to experts, to better address
urban microclimate conditions in their designs and urban masterplans.

2. Methodology and Case Study

This study investigates outdoor thermal comfort levels in two Philadelphia, PA neigh-
borhoods, one with a high social vulnerability index (SVI) and one with a low index. In
2021, Philadelphia was the city with the highest poverty rate (with 22.8% of people who live
below the poverty level) among the United States’ most populated cities [39]. Moreover,
since 2010, Philadelphia experienced a variety of extreme weather events, including the
snowiest winter, the two warmest summers, the wettest day, and the two wettest years on
record [40]. Philadelphia’s architectural landscape is mainly composed of narrow, attached
rowhouses that have a one-to-four-story structure. The city experienced rapid growth
during the early 20th century resulting in the construction of numerous high-rise buildings
and skyscrapers. The city’s street grid layout was established in the late 17th century, and
its urban structure has undergone several changes. Urban renewal projects in the 1950s and
1960s led to the demolition of several historic buildings and the construction of modernist
structures [41].
According to the U.S. Department of Energy’s Building America Program, Philadel-
phia is located in a Mixed-Humid region, which is characterized by receiving over 20 inches
(50 cm) of yearly rainfall, having around 5400 heating degrees days (based on a temperature
of 65 ◦ F), and experiencing an average monthly outdoor temperature below 45 ◦ F (7 ◦ C)
during winter [42]. The proposed methodology in this research consists of two primary
steps, which are outlined below.

2.1. Step 1: Exploring Social Vulnerability Index to Identify Neighborhoods

Using U.S. Census data, the SVI is gathered by the Agency for Toxic Substances &
Disease Registry (ATSDR) at the Centers for Disease Control and Prevention (CDC) to
identify the United States communities that will likely need support before, during, and
after a public health emergency [43]. Typical census tracts incorporate 1500 to 8000 people,
with an optimum size of 4000 people [44], and they are designed to be demographically
homogeneous. The CDC/ATSDR SVI ranks U.S. census tracts according to 15 individual
variables, i.e., poverty, unemployment, income, education, age (65 or older and 17 or
younger), disability, single-parent households, minority, language (speak English less
than well), housing unit structure, mobile homes, crowding, lack of vehicle, and group
quarter. These 15 variables are further categorized into four related themes including;
1-socioeconomic status, 2-household competition and disability, 3-minority status and
language, and 4-housing type and transportation.
 Buildings
Buildings2023,
2023,13,
13,x xFOR
FORPEER
PEERREVIEW
REVIEW 4 4ofof1919

than
thanwell),
well),housing
housingunit
unitstructure,
structure,mobile
mobilehomes,
homes,crowding,
crowding,lack lackofofvehicle,
vehicle,and
andgroup
group
quarter. These 15 variables are further categorized into four related themes
quarter. These 15 variables are further categorized into four related themes including; including; 1-1-
Buildings 2023, 13, 1040 4 of 19
socioeconomic
socioeconomicstatus,
status,2-household
2-householdcompetition
competitionand anddisability,
disability,3-minority
3-minoritystatus
statusand
andlan-
lan-
guage, and 4-housing type and transportation.
guage, and 4-housing type and transportation.
The
TheCDC
CDCSVI SVIdata
data(2018,
(2018,the
thelatest
latestversion
versionatatthe
thetime
timeofofconducting
conductingthisthisresearch)
research)
wereThe CDC SVI data (2018, the vulnerability
latest versionof atevery
the time of conducting this research)
were used to determine the social vulnerability of every census tract in Philadelphia.Fig-
used to determine the social census tract in Philadelphia. Fig-
were
ure used to the
determine the social vulnerability of every censusthemestract in Philadelphia.
ure11shows
shows thetasks
taskstotorank
rankthe
thetracts
tractsfor
forthe
thefour
fourvulnerability
vulnerability themesand andthe
theoverall
overall
Figure 1 shows the tasks
vulnerability to rank the tracts for the four vulnerability themes and the overall
vulnerabilityofofthe
thetracts
tractsasasthe
thelast
lasttask.
task.
vulnerability of the tracts as the last task.

Figure
Figure 1. The workflow
workflow forranking
ranking Philadelphia’scensus
census tract according to their SVI, Source:
Figure1.1.The
The workflowfor
for rankingPhiladelphia’s
Philadelphia’s censustract
tractaccording
accordingtototheir
theirSVI,
SVI,Source:
Source:Au-
Au-
Authors.
thors.
thors.
Census
Census tracts ininPhiladelphia are compared to one another in order to map and analyze
Censustracts
tracts inPhiladelphia
Philadelphiaare arecompared
comparedtotoone oneanother
anotherininorder
ordertotomap
mapandandana-
ana-
relative
lyze vulnerability
relative as
vulnerability a consequence
as a of
consequence the four
of aforementioned
the four themes
aforementioned called
themes the overall
called
lyze relativeindex
vulnerability vulnerability
across as acity.
the consequence
Tract of the four
rankings aforementioned
are based themes
on percentiles. calledthe
Percentilethe
overall
overall vulnerability
vulnerability index
index across
across the
thecity.
city.Tract
Tract rankings
rankings are
arebased
based on
onpercentiles.
percentiles. Percen-
Percen-
ranking values range from 0 (dark green) to 1 (dark red), with higher values indicating
tile
tileranking
ranking values
valuesrange
range from
from 0 (dark
(darkgreen) toto11(dark red), with higher values indicating
greater vulnerability (Figure 2). 0Tracts green)
for which (dark
there is red),
no SVIwith higher
data values
available areindicating
shown
greater vulnerability (Figure
(Figure 2). Tracts for which there is no SVI data available areshown
2). Tracts for which there is no SVI data available are
asgreater vulnerability
black and white stripes. shown
asasblack
blackand
andwhite
whitestripes.
stripes.

Figure
Figure 2.2.2.
Figure Maps
Maps
Maps show
show
show the
the
the range
range
range ofof
of vulnerability
vulnerability
vulnerability inin
in Philadelphia, PA
Philadelphia,PA
Philadelphia, PA asas
as aaconsequence
aconsequence
consequence ofof
of the
the
the four
four
four
themes
themes (overall ranking).
(overall ranking).
themes (overall ranking).

According to the CDC SVI data, census tracts in Philadelphia show significant differ-
ences in 15 socioeconomic characteristics, such as poverty, lack of vehicle access, housing
type, education, living alone, and disability across the city. Tracts located in northwestern
 Buildings 2023, 13, x FOR PEER REVIEW 5 of 19

Buildings 2023, 13, 1040 5 of 19

According to the CDC SVI data, census tracts in Philadelphia show significant differ-
ences in 15 socioeconomic characteristics, such as poverty, lack of vehicle access, housing
type, education, living alone, and disability across the city. Tracts located in northwestern
Philadelphia
Philadelphiaare areassociated
associatedwith witha alow
low percentile ranking
percentile ranking forfor
SVI, 0.10.1
SVI, to to
0.2,0.2,
while tracts
while in
tracts
the south, central, and a big part of the north sides show a very high percentile
in the south, central, and a big part of the north sides show a very high percentile ranking, ranking,
more
morethan
than0.80.8indicating
indicatinggreater
greatervulnerability
vulnerability(Figure
(Figure 2).
2). Moreover,
Moreover, Figure
Figure 33 shows
shows aa high
high
density
density of tree canopy in neighborhoods with a low vulnerable population on
of tree canopy in neighborhoods with a low vulnerable population and, theon
and, other
the
hand,
other ahand,
low density of trees
a low density ofin tracts
trees with awith
in tracts higher vulnerability
a higher index.
vulnerability Accordingly,
index. Accordingly,the
process of selecting
the process two urban
of selecting blocks
two urban in Philadelphia,
blocks one with
in Philadelphia, one awith
lowalevel
low of vulnerability
level of vulner-
and a high concentration of tree canopy and another with a high level of
ability and a high concentration of tree canopy and another with a high level of vulnera-vulnerability and
ability
low concentration of tree coverage, was achieved by overlaying the SVI
and a low concentration of tree coverage, was achieved by overlaying the SVI and and tree canopy
coverage
tree canopyfrom GIS shapefiles.
coverage from GIS shapefiles.

Figure3.3.Tree
Figure Treecanopy
canopycoverage
coverageininPhiladelphia,
Philadelphia,PA.
PA.

Theoutcome
The outcomeof ofthis
thisstage
stage involved
involved the
the selection
selection of
of two
two tracts
tracts out
out of
of 359
359 tracts
tracts in
Philadelphia. One
Philadelphia. One tract
tract was
was chosen
chosen from
from the
the tracts
tracts with
with the
the lowest
lowest vulnerability
vulnerability index
(STFID:42101021700,
(STFID: 42101021700,see seeFigure
Figure2)2)and
andhigh
hightree
treecoverage,
coverage, while
while the
the other
other was
was selected
selected
from the tracts with the highest vulnerability index (STFID: 42101004101, see
from the tracts with the highest vulnerability index (STFID: 42101004101, see Figure 2) and Figure 2)
and
the the lowest
lowest tree coverage
tree coverage to proceed
to proceed with with the step:
the next next step: microclimate
microclimate modeling.
modeling. TheseThese
two
two tracts
tracts and their
and their snapshots
snapshots are illustrated
are illustrated in Figure
in Figure 4. 4.
Buildings 2023,13,
Buildings2023, 13,1040
x FOR PEER REVIEW 66 of
of19
19

Figure 4. Two selected tracts for microclimate simulation; top left with low vulnerability index and
Figure 4. Two selected tracts for microclimate simulation; top left with low vulnerability index and
bottom right with high vulnerability index, the purple highlights show the exact boundary of each
bottom right with high vulnerability index, the purple highlights show the exact boundary of each
census tract.
census tract.
2.2.Step
2.2. Step2:2:Microclimate
MicroclimateandandOutdoor
OutdoorThermal
ThermalComfort
ComfortSimulation
Simulation
Thephysical
The physical basis
basis and
and spatial/temporal
spatial/temporalresolution resolution ofofurban
urbanmicroclimate
microclimate models
models dif-
fer greatly. ENVI-met [45], one of the most widely used tools for
differ greatly. ENVI-met [45], one of the most widely used tools for simulating urban simulating urban micro-
climates, can be
microclimates, canused to analyze
be used to analyzethe the
thermal
thermal comfort
comfortregime
regimewithin
withinthethestreet
street canyon
canyonat
atfine
fineresolutions,
resolutions,downdownto 0.5 × 0.5×m.
to 0.5 0.5The
m.user-friendly
The user-friendlytool aims
toolataims
replicating the primary
at replicating the
atmospheric
primary processes
atmospheric that affect
processes theaffect
that microclimate on a physical
the microclimate on abasis (i.e., the
physical basis fundamen-
(i.e., the
tal laws of fluid
fundamental lawsdynamics and thermodynamics).
of fluid dynamics and thermodynamics).BioMet isBioMet
a plug-in
is a for ENVI-met
plug-in for ENVI-that
calculates
met human thermal
that calculates human comfort
thermal indices
comfortfrom ENVI-met
indices model output
from ENVI-met model files. ENVI-met
output files.
BIO-met can
ENVI-met be used
BIO-met to be
can assess
usedtheto effects
assess of thethe presence
effects of theof presence
vegetation ofin an urban space,
vegetation in an
local shading
urban on shading
space, local the person, theperson,
on the position theofposition
the person
of theinperson
relation
in to buildings,
relation and the
to buildings,
albedo
and of paving
the albedo materials.
of paving materials.
ENVI-metBioMet
ENVI-met BioMetisisable
abletotopredict
predictoutdoor
outdoorthermal
thermalcomfort
comfortininterms
termsofofindices
indicessuchsuch
as a predicted mean vote (PMV), physiological equivalent temperature
as a predicted mean vote (PMV), physiological equivalent temperature (PET), universal (PET), universal
thermal
thermalclimate
climateindex
index(UTCI),
(UTCI),and andstandard
standardeffective
effectivetemperature
temperature(SET).(SET).In Inthis
thisresearch,
research,
the
thePMV
PMVindex
indexwas
wasadopted
adoptedto toevaluate
evaluatethe theoutdoor
outdoorthermal
thermalcomfort
comfortatatthethestreet
streetlevel
levelofof
the
theselected
selectedtracts
tractsininPhiladelphia.
Philadelphia.The ThePMVPMVisisaathermal
thermalindex
indexdeveloped
developedby byFanger
Fanger[46][46]
totoassess
assessindoor
indoorthermal
thermalcomfort.
comfort.The ThePMV PMVcalculated
calculatedby byENVI-met
ENVI-methas hasbeen
beenadapted
adaptedfor for
Buildings 2023,13,
Buildings2023, 13,1040
x FOR PEER REVIEW 7 of
of 19
19

outdoor
outdoorconditions,
conditions,including
includingsolar
solarradiation
radiationandandwindwindspeed.
speed. To Toperform
performaasimulation
simulation
through
throughENVI-met
ENVI-metBioMet,
BioMet,four
fourenvironmental
environmentalfactors,
factors,i.e.,
i.e.,air
airtemperature
temperature(T(Ta ), a),mean
mean
radiant temperature (MRT), wind speed (UV), and the specific humidity
radiant temperature (MRT), wind speed (UV), and the specific humidity (q), were ex- (q), were extracted
from the from
tracted ENVI-met outputs. Moreover,
the ENVI-met two personal
outputs. Moreover, twofactors,
personal i.e.,factors,
metabolici.e.,rate (ISO 8996)
metabolic rate
and clothing index (ISO 9920) for four seasons were extracted from
(ISO 8996) and clothing index (ISO 9920) for four seasons were extracted from the the literature [47]litera-
and
were
ture incorporated
[47] and wereinto the BIO-met
incorporated intomodel.
the BIO-met model.
To
Tosimulate
simulatethethePMVPMVindex
indexfor forthe
theselected
selectedtracts,
tracts,two
twohigh-resolution
high-resolution3-D 3-Dmodels
models
were
were created in ENVI-met (Figure 5), with green color representing grass coverageand
created in ENVI-met (Figure 5), with green color representing grass coverage and
trees
treesinineach
eachmodel.
model. AsAsa main
a maininput to ENVI-met,
input the geometrical
to ENVI-met, the geometrical data of theof
data models were
the models
obtained from Philadelphia
were obtained GIS Shapefile,
from Philadelphia open-source
GIS Shapefile, data provided
open-source data providedby Pennsylvania
by Pennsyl-
Spatial Data Access (PASDA) [48].
vania Spatial Data Access (PASDA) [48].

Figure5.5.The
Figure Theexact
exactboundary
boundaryofof selected
selected blocks
blocks within
within thethe census
census tracts
tracts (left(left images)
images) and and ENVI-
ENVI-met
met models created for each block (right images).
models created for each block (right images).

Inorder
In ordertotoconstruct
constructthree-dimensional
three-dimensionalmodelsmodelsofofneighborhoods,
neighborhoods,Grasshopper
Grasshopper3-D 3-D
andMeerkat,
and Meerkat,whichwhichare
arebased
basedon onRhinoceros
Rhinocerossoftware,
software,were
wereutilized.
utilized.The
Theprocess
processof ofusing
using
thesetools
these toolstotocreate
create urban
urban 3-D3-D models
models is explained
is explained in detail
in detail in [49,50].
in [49,50]. The The resulting
resulting raw
raw 3-D
3-D geometries were then imported into Dragonfly, a plug-in created [51] for
geometries were then imported into Dragonfly, a plug-in created [51] for use in Grasshopper, use in Grass-
hopper,
which whichthe
enables enables the of
creation creation of 3-Dusing
3-D models models using Envi-met
Envi-met (INX). To(INX). To incorporate
incorporate specific
specific urban characteristics, such as construction materials, land cover,
urban characteristics, such as construction materials, land cover, albedo, and anthropogenicalbedo, and an-
thropogenic
heat generatedheat generated
by human by human
activities, activities,
the two the two
areas under areas under
investigation investigation
were were
classified based
classified
on the localbased
climateonzone
the local
(LCZ) climate
system.zone
The(LCZ)
LCZs system.
categorizeThe LCZs categorize
landscapes based on landscapes
a typical
basedofonvalues
range a typical range ofcover,
for surface values for surface
urban cover,
structure, urban
fabrics structure,
(radiative andfabrics
thermal (radiative and
properties
thermal
of properties
construction of construction
materials), materials),
and anthropogenic andflux
heat anthropogenic
[52]. The LCZ heat flux [52]. The
classification LCZ
scheme
Buildings 2023, 13, x FOR PEER REVIEW 8 of 19
Buildings 2023, 13, 1040 8 of 19

classification scheme consists of 10 built and 7 land cover types, and each of the 17 basic
consists
types of 10 builtwith
is associated and typical
7 land cover types, and
value ranges for aeach ofkey
set of the urban
17 basic types is associated
parameters. The defi-
with typical
nition value ranges
and physical for aof
properties setall
ofLCZs
key urban parameters.
are provided in 17The definition
illustrative and physical
datasheets. Ac-
properties of
cordingly, thealltract
LCZs are STFID
with provided in 17 illustrative
of 42101021700 (topdatasheets.
images in Accordingly,
Figure 4, simplythe tract with
Block 1,
STFID of 42101021700 (top images in Figure 4, simply Block 1, hereafter)
hereafter) was classified as LCZ6: Open low-rise, where buildings are detached and small was classified as
LCZ6:
size Open
with low-rise,
1–3 stories where
tall. On thebuildings
other hand,are detached and the
the tract with small size of
STFID with 1–3 stories(bot-
42101004101 tall.
On the other hand, the tract with the STFID of 42101004101 (bottom
tom images in Figure 4, simply Block 2, hereafter) was classified as LCZ 3: Compact low- images in Figure 4,
simply
rise Block 2,
in which hereafter)
buildings arewas classified
attached as LCZ spaced
or closely 3: Compact
withlow-rise in which
1–3 stories buildings
tall. Land coverare
in
attached
LCZ 3 is or closelypaved
mostly spacedorwith 1–3 stories LCZ6
hard-packed; tall. Land cover in
is mostly LCZ 3 isbymostly
covered paved
scattered or hard-
trees and
packed; LCZ6
abundant is mostly
plants (Figurecovered by scattered
6). In order treesthe
to balance andneed
abundant plants (Figure
for accurate simulation6). Inresults
order
to balance the need for accurate simulation results with the computational
with the computational time required, the 3-D models were constructed for only a portion time required,
thethe
of 3-D modelstracts.
selected were constructed
Specifically, for theonly a portion
models of the selected
were created to covertracts.
an areaSpecifically,
with a mini- the
models were created to cover an area with a minimum diameter
mum diameter of 400 m (or a radius of 200 m), which is the minimum size specified in the of 400 m (or a radius of
200 m), which
LCZ description. is the minimum size specified in the LCZ description.

Figure
Figure 6. LCZ
LCZ description,
description, data
data adapted
adapted from
from [52],
[52], aerial
aerial images
images from
from Google
Google Earth
Earth (Philadelphia,
(Philadelphia,
PA).
PA).

Table
Table 11 summarizes
summarizesthe thedetails
detailsofofeach
eachmodel
model forfor
geometrical
geometrical data
dataandandurban
urbanconfig-
con-
uration.
figuration. Although
Althoughthe the
buildings
buildingsin the
in two models
the two havehave
models almost the same
almost heightheight
the same (6 to 7(6m),
to
other properties
7 m), other suchsuch
properties as greenery
as greeneryarea, sitesite
area, coverage
coverage byby
built
builtelements,
elements,and and pavement
pavement
coverage
coverage are aresubstantially
substantiallydifferent
differentbetween
betweenthe thetwo models.
two models.Additionally,
Additionally, thethe
canyon
canyonas-
aspect
pect ratio,
ratio, height-to-width(H/W),
height-to-width (H/W), forfor streets
streets locatedwithin
located withinBlock
Block11isisconsistently
consistently0.25;
0.25;
ratio varies
this ratio varies for
for streets
streets in different
different directions within Block 2. Specifically,
Specifically, thethe H/W ratio
H/W ratio
Block 22 ranges
in Block ranges from
from 0.32
0.32 for
for primary
primary East-West
East-West streets to 0.8 for for secondary
secondary streets
streets in the
the
North-South and East-West directions.
In total, 180 h of medium forcing and parallel microclimate simulations were con-
ducted
Table 1. for
Modelthedimensions
two models. andEach model wasforsubject
site properties to six different weather scenarios, with
two blocks.
four scenarios for each of the four seasons (winter, spring, summer, and fall), as well as two
Site Properties
scenarios for extreme hot and cold days. The simulations were conducted separately for
Model Average Site Cover- Grass Cov- Pavement Road Cov-
Census Block each scenario,
Number using
of weather data obtained from the National Oceanic and Atmospheric
Dimensions Site Area Building age erage Coverage erage H/W Ratio
Administration
Buildings (NOAA) for the year 2021, which included hourly measurements of air
Height Ratio Ratio Ratio Ratio
temperature, relative humidity, wind speed, and wind direction and are listed in Table 2.
152 × 184 × 30
To represent each season, average data for a month within each season (February, May,
dx = 2 m 81,182
Block 1 July, 183 were 6selected.
and October) m 0.21literature
The 0.48 0.03
suggests beginning the0.28
simulation0.25 process
dy = 2 m (m2)
dz = 2 m during the early hours of the morning, specifically after sunrise, as this time is characterized
by stable atmospheric conditions and minimal impact from the warming effects of the sun
Primary E-W:
132 × 137 × 30 on the urban environment [33,53]. Therefore, the simulation for each case was 0.32 started at
dx = 2 m 06:00
53,026a.m. and lasted for 24 h to accurately reproduce the characteristics of the selected
Primary N-S:day.
Block 2 302 7m 0.42 0.02 0.18 0.22
dy = 2 m (m2) 0.5
dz = 2 m Secondary N-
S, E-W: 0.8
Buildings 2023, 13, 1040 9 of 19

Table 1. Model dimensions and site properties for two blocks.

Site Properties
Census Model Number Average Site Grass Pavement Road
Block Dimensions H/W
Site Area of Building Coverage Coverage Coverage Coverage
Ratio
Buildings Height Ratio Ratio Ratio Ratio
152 × 184
× 30
81,182
Block 1 dx = 2 m 183 6m 0.21 0.48 0.03 0.28 0.25
(m2 )
dy = 2 m
dz = 2 m
Primary
132 × 137 E-W: 0.32
× 30 Primary
53,026
Block 2 dx = 2 m 302 7m 0.42 0.02 0.18 0.22 N-S: 0.5
(m2 )
dy = 2 m Secondary
dz = 2 m N-S, E-W:
0.8

Table 2. Climatology data for 6 defined cases, source: NOAA.

Average
Wind
Cases Tmax (◦ C) Tmin (◦ C) TAVG (◦ C) Wind Speed
Direction
(m/s)
Winter Case 4.24 −1.84 1.23 4.35 NW
Coldest Day
0 −9.99 −4.99 7.07 NW
(2 February 2021)
Summer Case 30 21.6 25.98 3.34 SW
Hottest Day
35.52 24.42 29.99 3.52 SW
(7 July 2021)
Fall Case
17 7.8 12.47 3.7 SE
(24 October 2021)
Spring Case
20.1 8.48 14.13 4.11 NW
(12 May 2021)

3. Results and Discussion

To determine the climate variables and outdoor thermal comfort in various seasons,
fine-level (2 m × 2 m) microclimate and the PMV index were simulated using the ENVI-met
and BIO-met models, respectively. As it has been mentioned, 24 h microclimate simulation
for six scenarios (winter, spring, summer, fall, extreme hot, and extreme cold) was run for
the two models. It means, for example, the PMV inside one model was simulated 144 times
to cover 24 h of the six scenarios. After conducting simulations for all six scenarios, the
ENVI-met Leonardo tool was employed to visualize the results.
Figure 7 depicts the mean radiant temperature (MRT) simulated at a height of 1.5 m
above the ground level at noon for the two blocks. MRT is considered the most important
meteorological factor for evaluating outdoor thermal sensation under sunny conditions,
regardless of the comfort index used [54]. The MRT range for both models is almost
identical, with values exceeding 60 ◦ C for streets that are fully exposed to solar radiation
and constructed with asphalt. However, the effects of shading and evapotranspiration
from trees result in a range of MRTs between 47 ◦ C and 51 ◦ C for areas below or close to
trees (yellow to green areas), leading to an average MRT of 54.5 ◦ C at street level in Block 1,
while a very low number of trees in Block 2 led to an average simulated MRT of 61.5 ◦ C.
It should be noted that the average values are calculated for the entire street surfaces,
including areas under trees and other surfaces made of asphalt and concrete pavement.
Buildings 2023, 13, 1040 10 of 19

The influence of grass coverage on MRTs was negligible, with grassy areas exhibiting MRTs
that were at most 2 ◦ C lower than other surfaces in the two blocks. Additionally, the
Buildings 2023, 13, x FOR PEER REVIEW 10 of 19
simulations revealed a minimum MRT of 35.41 ◦ C and 38.47 ◦ C and a maximum of 66.45 ◦ C
◦
and 67.12 C for Block 1 and Block 2, respectively.

Figure7.7.Simulated
Figure SimulatedMRT
MRTat
at noon
noon and
and 1.5
1.5 m
m above the ground level,
level, summer
summer case.
case.

Figure
Figure88provides
provides aa visual
visual comparison
comparison of of surface temperatures for
surface temperatures forthe
thesummer
summercase case
atatnoon. It should be noted that the dark blue areas represent the footprint
noon. It should be noted that the dark blue areas represent the footprint of buildings of buildings
and
andareare excluded
excluded from the surface
from the surfacetemperature
temperatureanalysis.
analysis.TheseThese areas
areas areare
notnot reflective
reflective of
of the surface material or vegetation cover and, therefore, would not
the surface material or vegetation cover and, therefore, would not provide accurate infor-provide accurate
information on surface
mation on surface temperatures.
temperatures. However,
However, the maps
the maps showshow significant
significant differences
differences in sur-in
surface temperatures across various materials, such as asphalt, concrete
face temperatures across various materials, such as asphalt, concrete pavement, and grass pavement, and
grass
areas. Among these materials, the largest temperature difference is observed in asphalt,in
areas. Among these materials, the largest temperature difference is observed
asphalt,
where thewhere the temperature
temperature exceedsexceeds 47 ◦ C (purple
47 °C (purple and redand red areas).
areas). In contrast,
In contrast, the lowest
the lowest tem-
temperature of 23 ◦ C (blue areas) is recorded on the surface right below trees. Additionally,
perature of 23 °C (blue areas) is recorded on the surface right below trees. Additionally,
the ◦ C to 38 ◦ C
thesurface
surfacetemperature
temperature of of grass
grass areas
areas was
was simulated
simulated at at aa range
range between
between 34.5
34.5 °C to 38 °C
(yellow areas). Accordingly, Block 1, with almost 48% covered by
(yellow areas). Accordingly, Block 1, with almost 48% covered by grass, benefits from grass, benefits from
shading
shading and cooling effects that mitigate heat buildup on surfaces compared to Block 2,2,
and cooling effects that mitigate heat buildup on surfaces compared to Block
with
withonly
only0.02%
0.02%grass
grasscoverage
coverage but
but 40%
40% covered
covered by by impervious
impervious surfaces
surfacessuch
suchas asasphalt
asphalt
and concrete pavement.
and concrete pavement.
The output files from ENVI-met, along with two personal factors, i.e., metabolism rate
and clothing index for each season were incorporated into ENVI-met BioMet to simulate
the PMV indices. The PMV model is based on heat balance principles and connects six
crucial elements of thermal comfort to the average human response on a specified scale.
Normally, the PMV scale is defined on a seven-point thermal sensation scale ranging from
cold (−3) to neutral or comfort (0) to hot (+3).
In Figure 9, the output of the ENVI-met model is shown as PMV index maps for the
summer scenario at four different hours of the day: 10:00 am, 02:00 pm, 06:00 pm, and
10:00 pm. Maps in Figure 9 serve as examples to demonstrate the variations of comfort
index across the models’ area at different hours, and findings are explained in the following
sections.
PMV values at 10:00 am, summer
According to the results of the simulations, the PMV values at street level and grass
areas that were fully exposed to the sun were above 3, which indicates a hot environment
and is represented by red to pink colors in the output maps of both models. For areas
near or directly below trees, PMV values were recorded in a range from slightly warm
(PMV = 1.5) to warm (PMV = 2.5), as indicated by the blue to yellow colors in the maps. For
Figure 8. Simulated Surface temperature at noon, summer case.

The output files from ENVI-met, along with two personal factors, i.e., metabolism rate
and clothing index for each season were incorporated into ENVI-met BioMet to simulate
Buildings 2023, 13, 1040 11 of 19

Buildings 2023, 13, x FOR PEER REVIEW 10 of 19

areas located 2 to 4 m to the north or northwest of the building geometries, the PMV values
were recorded at the lowest value, below 2, which represents a slightly warm environment.
However, no spot was found within the two models that had a PMV value lower than 1,
which is the comfort range.
PMV values at 02:00 pm, summer
Based on the ENVI-met output maps at 2:00 pm, it can be observed that the entire
areas of the two blocks showed PMV values higher than 2.5, which indicates a warm
environment. These PMV values were recorded in a few spots located under trees or in
areas 2 to 4 m to the north or northeast of the buildings, as shown by light to dark blue
colors in the maps. However, spots located on grass areas, asphalt, and concrete pavement
all showed PMV values higher than 4, which represents a very hot environment. This
indicates that the shading effects of the trees were able to provide some relief from the
heat, but other areas that were fully exposed to solar radiation experienced extremely hot
conditions.
PMV values at 06:00 pm, summer
During the simulation at 6:00 pm, there were still no spots found in both Block 1 or
Block 2 with PMV values within the comfort range. However, spots that were affected by
trees or located at a distance of 2 to 10 m on the east side of the buildings showed PMV
Figure between
values 7. Simulated MRT
1 and 2,atindicating
noon and 1.5 m abovewarm
a slightly the ground level, summer
environment. Basedcase.
on the results, the
shading effect of trees and buildings had a significant impact on reducing the PMV index
comparedFigureto8spots
provides
on thea visual
street comparison
level, whichofhad surface
PMVtemperatures
values higherfor the 3,
than summer casea
indicating
at noon. It
hot environment. should be noted that the dark blue areas represent the footprint of buildings
and Upon
are excluded
comparingfromthe themaps
surface temperature
of the two blocks, analysis. These areas
it was observed thatare not 1,
Block reflective
which hadof
the surface material or vegetation cover and, therefore, would not
more trees, had more areas with blue color (light to dark), indicating a lower average of provide accurate infor-
mation
PMV on surface
values comparedtemperatures.
to Block 2.However, the maps show significant differences in sur-
face temperatures across various materials, such as asphalt, concrete pavement, and grass
PMV values at 10:00 pm, summer
areas. Among these materials, the largest temperature difference is observed in asphalt,
where Thethe
simulation
temperature results for 10:00
exceeds 47 °C pm indicate
(purple andthat
rednearly allInspots
areas). within
contrast, theBlocks
lowest1 tem-
and 2
have PMVofvalues
perature 23 °C ranging
(blue areas) between −0.5 toon+0.5,
is recorded the which
surfaceisright
within the trees.
below comfort range. This
Additionally,
suggests
the surface thattemperature
trees and buildings
of grasshaveareasno significant
was simulated impact on thebetween
at a range PMV values
34.5 °Cafter
to sunset.
38 °C
It(yellow
shouldareas).
be noted that this result
Accordingly, Block is observed regardless
1, with almost of the location
48% covered of the
by grass, spot within
benefits from
the two blocks,
shading whether
and cooling it is that
effects influenced
mitigatebyheat
trees or buildings.
buildup Overall,
on surfaces the PMV
compared to values
Block 2,at
10:00
with pm onlysuggest that the
0.02% grass thermal
coverage butcomfort conditions
40% covered have improved
by impervious significantly
surfaces within
such as asphalt
the
andurban environment
concrete pavement.of the two blocks.

Figure8.8.Simulated
Figure Simulated Surface
Surface temperature
temperature at
at noon,
noon, summer
summer case.
case.

The output files from ENVI-met, along with two personal factors, i.e., metabolism rate
and clothing index for each season were incorporated into ENVI-met BioMet to simulate
 cold (−3) to neutral or comfort (0) to hot (+3).
In Figure 9, the output of the ENVI-met model is shown as PMV index maps for th
summer scenario at four different hours of the day: 10:00 am, 02:00 pm, 06:00 pm, an
10:00 pm. Maps in Figure 9 serve as examples to demonstrate the variations of comfor
Buildings 2023, 13, 1040
index across the models’ area at different hours, and findings are explained12inof the
19
follow
ing sections.

Figure 9. Simulated PMV inside the two models during the summer case for every 4 h starting at
10 am and ending by 10 pm.
Buildings 2023, 13, 1040 13 of 19

The findings from the summer scenario and the other five weather scenarios demon-
strate a strong correlation between tree and grass coverage and outdoor air temperature,
mean radiant temperature, and surface temperature during hot seasons and extreme heat
events. Areas with a higher concentration of tree canopies experienced a maximum reduc-
tion of 1.5 ◦ C in outdoor air temperatures and up to 31 ◦ C in mean radiant temperatures
during the daytime of summer and extremely hot conditions. The effects of greenery were
more significant in Block 1 (associated with a low SVI), where nearly 50% of the site was
covered by trees and grass, compared to only 0.02% of Block 2 (associated with a high
SVI). However, the benefits of tree and grass coverage were limited to their immediate
Buildings 2023, 13, x FOR PEER REVIEW 13
surroundings, with areas without green coverage exhibiting significantly higher mean
radiant temperatures.
Furthermore, the results showed a strong relationship between PMV and MRT, with
a Pearson correlation Furthermore, the results
analysis indicating showed a strong
a significant relationship
correlation between
of 0.987. PMV and MRT,
A significant
a Pearson correlation analysis indicating a significant
correlation of 0.859 was also found between air temperature and PMV. The study correlation of 0.987.
also A signifi
correlation of 0.859 was also found between air temperature
found that a higher tree canopy cover in Block 1 was associated with lower outdoor heat and PMV. The study
found that a higher tree canopy cover in Block 1 was
exposure in that area, while more impervious surfaces such as streets and pavements in outdoor
associated with lower
exposure
Block 2 were linked in that
to higher area,
heat while more
exposure impervious
in that area. surfaces such as streets and pavemen
To obtain a more accurate representation of the simulatedthat
Block 2 were linked to higher heat exposure in area. variables and com-
climate
fort index in the two models, the study included nine specific pointssimulated
To obtain a more accurate representation of the for closer climate variables and c
investigation,
fort index in the two models, the study included nine
as shown in Figure 10. These points were strategically placed in primary and secondary specific points for closer inves
tion,surfaces,
streets, on asphalt as shownsidewalks,
in Figure 10. andThese points
under were
trees, strategically
with the streetplaced in primary
direction either and sec
ary streets, on asphalt surfaces, sidewalks, and under
running from North to South or East to West. The nine specific points were named as trees, with the street direction ei
follows: running from North to South or East to West. The nine specific points were name
follows:
• North to South primary street, on asphalt (N-S)PA
• East to West primary
North street,
to Southonprimary
asphalt street,
(E-W)PA on asphalt (N-S)PA
•  East to West primary street,
North to South secondary street, on asphalt (N-S)SA on asphalt (E-W)PA
•  North to South secondary
East to West secondary street, on asphalt (E-W)SA street, on asphalt (N-S)SA
•  East to West secondary street,
North to South primary street, on pavement (N-S)PP on asphalt (E-W)SA
•  North to South primary
East to West primary street, on pavement (E-W)PP street, on pavement (N-S)PP
•  East
North to South secondary street, on pavement (N-S)SP (E-W)PP
to West primary street, on pavement
 North to South secondary street, on pavement (N-S)SP
• East to West secondary street, on pavement (E-W)SP
 East to West secondary street, on pavement (E-W)SP
• Under Tress (TRS)
 Under Tress (TRS)

Figure
Figure 10. Location 10. Location
of nine of nine
points inside thepoints inside the two models.
two models.

By selecting
By selecting points points
in different in different
locations, locations,
the study wasthe study
able was able
to gather datatoon
gather
the data on
varying effects of different surface types and orientations, which
varying effects of different surface types and orientations, which allowed for a moreallowed for a more
anced analysis of the simulated climate variables and their impact on human com
Figures 11 and 12 illustrate hourly MRTs simulated at each point inside Block 1 and 2
ing the summer case, respectively. The simulation was programmed to take measurem
every hour, starting at 6:00 am and ending at 6:00 am, 24 h overall.
Buildings 2023, 13, 1040 14 of 19

nuanced analysis of the simulated climate variables and their impact on human comfort.
Figures 11 and 12 illustrate hourly MRTs simulated at each point inside Block 1 and 2 during
the summer case, respectively. The simulation was programmed to take measurements
Buildings 2023, 13, x FOR PEER REVIEW 14 of 19
Buildings 2023, 13, x FOR PEER REVIEW
every hour, starting at 6:00 am and ending at 6:00 am, 24 h overall. 14 of 19

Figure 11. MRTs during summer case at nine points inside Block 1.
Figure 11. MRTs during summer case at nine points inside Block 1.
Figure 11. MRTs during summer case at nine points inside Block 1.

Figure 12. MRTs during summer case at nine points inside Block 2.
Figure 12. MRTs during summer case at nine points inside Block 2.
Figure 12. MRTs during summer case at nine points inside Block 2.
The points located under trees showed lower mean radiant temperatures (MRTs)
The points
throughout
The points located compared
the daytime
located under trees
under trees showed
to showed lower
other points.
lower mean
For radiant
further
mean temperatures
investigation
radiant temperatures of the(MRTs)
impact
(MRTs)
throughout the daytime
of other factors
throughout the daytime compared
on MRT compared
and PMV, to to other
theother points.
otherpoints. For further
eight points
For further investigation
wereinvestigation
selected at placesof the
of thewithout
impact
impact
of
of other
tree factors
and factors
other greenery oneffects.
on MRT and
MRT and PMV,
ThePMV,
urban the
the other eight
eightparameters,
morphology
other points were
points wereincluding
selected at
selected atSky
places
View
places without
Factor
without
tree and greenery effects.
(SVF), Albedo, andeffects. The urban morphology parameters, including
H/W ratio, were recorded at each point and analyzed for their effect Sky View Factor
(SVF),
on MRT
(SVF), Albedo,
and PMV
Albedo, and(Table
and H/W ratio,
H/W ratio, wereparameters
3). These
were recordedatat
recorded eachpoint
describe
each point
theand and analyzed
physical forfor their
characteristics
analyzed their effect
of
effect ur-
on
on
banMRT
MRT andand
PMV PMV
environments, (Table
(Table 3). These
and3).they
These parameters
canparameters
affect describe
describe
various aspects theofthe
thephysical
physical
urban characteristics
characteristics
environment, of ur-
ofinclud-
urban
ban environments,
environments,
ing microclimate and andand
they they
humancan can affect
affect
comfort. various
various aspects
aspects of the
of the urban
urban environment,
environment, includ-
including
ing microclimate
microclimate
In Block and and
1, sixhumanhuman
points comfort.
comfort.
showed the approximately same amount of MRT during the
In Block
daytime, while1,thesixother
points showed
three points, the approximately
including (N-S)PA, same amount
(N-S)SP, and of
TRS,MRT during
showed the
lower
daytime,
MRT values while the other
at certain three
times points, 7:00
between including
am and (N-S)PA,
7:00 pm. (N-S)SP, and TRS,
For instance, theshowed lower
point located
MRT
on thevalues
primary at certain
asphalttimes
in thebetween 7:00 am
North-South and 7:00(N-S)PA,
direction, pm. For showed
instance,a the
valuepoint located
of 45 °C at
on
7:00the primary
am, 37 °C at asphalt
8:00 am,in the
40 °CNorth-South
at 9:00 am,direction,
and 45 °C(N-S)PA,
at 10:00 am.showed a value
For the rest ofof the
45 °C at
day,
7:00
MRTam, 37 °C at
recorded at8:00
thisam,
point 40 was
°C atsimilar
9:00 am,to and
that 45of °C at 10:00
other points,am.which
For thewas rest of the65day,
above °C.
MRT
The MRTrecorded
at the at thislocated
point point wason thesimilar to that
pavement of other
surface points,
of the whichstreet
secondary was above 65 °C.
in the North-
The
SouthMRT at the point
direction, located
(N-S)SP, on the pavement
fluctuated between 65 surface of the
°C at 9:00 amsecondary
and 50 °Cstreet
at noon,in the North-
returning
South
to the direction,
same amount (N-S)SP, fluctuated
as other pointsbetween 65pm.
after 3:00 °C atThe
9:00point
am and 50 °Cunder
located at noon, returning
a tree (TRS)
to
hadthe sameMRT
lower amount as otherthe
throughout points afterwith
daytime, 3:00 apm. The point
maximum located
value of 20 under
°C at 9:00a tree
am(TRS)
com-
had lower MRT throughout
pared to the other points. the daytime, with a maximum value of 20 °C at 9:00 am com-
Buildings 2023, 13, 1040 15 of 19

Table 3. SVF, Albedo, and H/W ratio at each point inside two blocks.

Sky View Factor

Blocks Surface Albedo H/W Ratio
(SVF)
Block 1 0.84 0.2 0.25
(E-W)PA
Block 2 0.78 0.2 0.32
Block 1 0.75 0.35 0.25
(E-W)PP
Block 2 0.68 0.35 0.32
Block 1 0.84 0.2 0.25
(E-W)SA
Block 2 0.62 0.2 0.8
Block 1 0.81 0.35 0.25
(E-W)SP
Block 2 0.52 0.35 0.8
Block 1 0.61 0.2 0.25
(N-S)PA
Block 2 0.70 0.2 0.5
Block 1 0.72 0.35 0.25
(N-S)PP
Block 2 0.52 0.35 0.5
Block 1 0.66 0.2 0.25
(N-S)SA
Block 2 0.61 0.2 0.8
Block 1 0.45 0.35 0.25
(N-S)SP
Block 2 0.51 0.35 0.8

In Block 1, six points showed the approximately same amount of MRT during the
daytime, while the other three points, including (N-S)PA, (N-S)SP, and TRS, showed lower
MRT values at certain times between 7:00 am and 7:00 pm. For instance, the point located
on the primary asphalt in the North-South direction, (N-S)PA, showed a value of 45 ◦ C
at 7:00 am, 37 ◦ C at 8:00 am, 40 ◦ C at 9:00 am, and 45 ◦ C at 10:00 am. For the rest of
the day, MRT recorded at this point was similar to that of other points, which was above
65 ◦ C. The MRT at the point located on the pavement surface of the secondary street in the
North-South direction, (N-S)SP, fluctuated between 65 ◦ C at 9:00 am and 50 ◦ C at noon,
returning to the same amount as other points after 3:00 pm. The point located under a tree
(TRS) had lower MRT throughout the daytime, with a maximum value of 20 ◦ C at 9:00 am
compared to the other points.
In Block 2, MRT recorded at two points was lower than other points, including the
TRS point, during the morning. These two points were both in the North-South direction
and located on both asphalt and pavement surfaces in a secondary street. These two points
were well-shaded by the surrounding building, especially buildings on the right side that
blocked east sun radiation.
Although the H/W ratio was constant (0.25 for all the points) inside Block 1, SVF and
albedo showed a strong effect on MRTs and PMV. A Pearson correlation of 0.61 was found
between SVF and MRTs inside Block 1, while it was found up to 0.75 inside Block 2. It means
that points with higher SVF have more exposure to heat, resulting in higher MRTs and PMV.
Conversely, the urban morphology parameter of albedo displayed a negative correlation
of −0.60 with MRTs inside Block 1 and −0.55 inside Block 2, indicating a significant yet
inverse relationship. Thus, it can be inferred that higher albedo values correspond to lower
MRTs. In Model 2, where different streets had varying H/W ratios, the coefficient with
MRTs was −0.39, suggesting that this parameter also has a notable impact on MRT values.

4. Conclusions
This research presented a method that integrates a climate classification system with a
3-D microclimate model to investigate the correlation between outdoor urban climate and
the socioeconomic status of urban blocks, using two urban blocks in Philadelphia, Pennsyl-
Buildings 2023, 13, 1040 16 of 19

vania, USA, as the study case. The study involved creating and simulating microclimate
models in two urban blocks, which were extracted from two census tracts with varying
vulnerability indices. The results demonstrated that urban morphological factors such as
vegetated areas, sky view factor (SVF), albedo, and height-to-width (H/W) ratio have an
influence on air temperature, mean radiant temperature (MRT), and predicted mean vote
(PMV).
The study revealed that trees have a significant cooling effect on outdoor air tempera-
ture, MRT, and PMV index during the summer day scenario. These findings are consistent
with previous research [55–57] that highlighted the beneficial impacts of urban vegetation
and roadside trees on the microclimate by reducing surface temperature and MRT, increas-
ing evapotranspiration, providing shading, and reducing PMV. However, the study also
found that the benefits of trees are limited to their immediate surroundings, with areas
lacking green coverage exhibiting significantly higher MRTs. Consistent with the finding
from [56], this study shows the effect of trees on air temperature decreases with increasing
distance from the tree canopy.
The study also revealed that the impact of the H/W ratio and surface albedo on
outdoor thermal comfort is significant, with higher H/W ratios and albedo leading to
lower MRT and greater comfort levels, which is consistent with findings from [53,58,59].
Furthermore, the results presented in this research align with those of a study conducted
by [60], which also indicated that factors related to urban morphology, such as building
density and orientation, play a notable role in influencing microclimate parameters and
thermal comfort.
In conclusion, it is recommended that the utilization of taller and denser buildings
with higher albedo pavement surfaces should be considered to enhance thermal comfort
conditions, particularly in urban blocks with limited vegetation coverage. The proposed
workflow provides evidence to support incorporating urban climate knowledge into the
planning and design of healthful and resilient urban environments. Overall, the study
highlights the importance of considering urban morphology and vegetation coverage in
the design of sustainable and livable urban environments.
This research could be extended in four directions. First, the research involved a 24-h
simulation cycle, while it is recommended for future studies to extend the timeframe to
48-h or 72-h cycles to obtain a more comprehensive understanding. Second, the ENVI-
met models were supplemented with generic data from the LCZ datasheet for building
typology and properties. However, to improve accuracy, genuine information on building
materials, A/C types, green roofs and walls, and blue (water) surfaces, if available within
the studied blocks, can be added to the models. Third, the research solely concentrated
on outdoor weather variables and comfort levels. However, researchers in this era are
interested in indoor comfort, which can be investigated in future works. Last, the research
concentrated on Philadelphia, PA, from the U.S. Northeast region. However, to gain a better
understanding of the effects of urban properties on outdoor climate variables and thermal
comfort inside urban blocks with varying social vulnerability indexes, this research should
be expanded to incorporate more cities from other climate zones, including hot and dry or
extremely cold.

Author Contributions: Conceptualization, F.H., U.P., L.D.I. and G.C.; methodology, F.H. and U.P.
and G.C.; software, F.H.; writing—original draft preparation, F.H.; review and editing, U.P., L.D.I.
and G.C. All authors have read and agreed to the published version of the manuscript.
Funding: This research was supported in part by the National Science Foundation (Award #2207436),
the USDA National Institute of Food and Agriculture and Multistate Research Project (#PEN04623;
Accession #1013257), the Eunice Kennedy Shriver National Institute of Child Health and Human
Development (Award #P2C HD041025), and the 2021 Institute of Energy and the Environment (IEE)
seed grant program at the Pennsylvania State University.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Buildings 2023, 13, 1040 17 of 19

Acknowledgments: The authors are grateful for the support from the Hamer Center for Community
Design, a research arm of Stuckeman School, and the Department of Architecture at the Pennsylvania
State University. Additionally, the authors acknowledge the contributions of Yosef Bodovski, a GIS
research analyst at Penn State’s Computational and Spatial Analysis (CSA) Core, for his aid with the
GIS maps.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Masson-Delmotte, T.W.; Zhai, P.; Pörtner, H.-O.; Roberts, D.; Skea, J.; Shukla, P.R.; Pirani, A.; Moufouma-Okia, W.; Péan, C.;
Pidcock, R.; et al. Global Warming of 1.5 ◦ C. An IPCC Special Report on the Impacts of Global Warming of 1.5 ◦ C above
Pre-industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global
Response to the Threat of Climate Change. 2018. Available online: https://www.ipcc.ch/sr15/ (accessed on 17 February 2022).
2. P. D. United Nations, Department of Economic & Social Affairs. World Urbanization Prospects: The 2018 Revision; United Nations:
New York, NY, USA, 2019; Available online: https://population.un.org/wup/publications/Files/WUP2018-Report.pdf (accessed
on 13 March 2022).
3. Taha, H. Urban climates and heat islands: Albedo, evapotranspiration, and anthropogenic heat. Energy Build. 1997, 25, 99–103.
[CrossRef]
4. Unger, J. Connection between urban heat island and sky view factor approximated by a software tool on a 3D urban database. Int.
J. Environ. Pollut. 2009, 36, 59. [CrossRef]
5. Landsberg, H.E. Urban energy fluxes. In The Urban Climate; University of Maryland: College Park, MD, USA, 1981; pp. 53–82.
6. Oke, T.R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 1982, 108, 1–24. [CrossRef]
7. Taha, H. Characterization of Urban Heat and Exacerbation: Development of a Heat Island Index for California. Climate 2017, 5, 59.
[CrossRef]
8. Santamouris, M. Science of the Total Environment Analyzing the heat island magnitude and characteristics in one hundred Asian
and Australian cities and regions. Sci. Total Environ. 2015, 512–513, 582–598. [CrossRef]
9. Peng, S.; Piao, S.; Ciais, P.; Friedlingstein, P.; Ottle, C.; Bréon, F.-M.; Nan, H.; Zhou, L.; Myneni, R.B. Surface Urban Heat Island
Across 419 Global Big Cities. Environ. Sci. Technol. 2011, 46, 696–703. [CrossRef] [PubMed]
10. Estoque, R.C.; Murayama, Y.; Myint, S.W. Science of the Total Environment Effects of landscape composition and pattern on
land surface temperature: An urban heat island study in the megacities of Southeast Asia. Sci. Total. Environ. 2017, 577, 349–359.
[CrossRef] [PubMed]
11. Ho, H.C.; Knudby, A.; Chi, G.; Aminipouri, M.; Lai, D.Y.-F. Spatiotemporal analysis of regional socio-economic vulnerability
change associated with heat risks in Canada. Appl. Geogr. 2018, 95, 61–70. [CrossRef] [PubMed]
12. Khatana, S.A.M.; Werner, R.M.; Groeneveld, P.W. Association of Extreme Heat with All-Cause Mortality in the Contiguous US,
2008–2017. JAMA Netw. Open 2022, 5, e2212957. [CrossRef]
13. Centers for Disease Control and Prevention. National Environmental Public Health Tracking Network. 2023. Available online:
www.cdc.gov/ephtracking (accessed on 10 May 2022).
14. Anderson, G.B.; Bell, M.L. Heat Waves in the United States: Mortality Risk during Heat Waves and Effect Modification by Heat
Wave Characteristics in 43 U.S. Communities. Environ. Health Perspect. 2011, 119, 210–218. [CrossRef]
15. Heaton, M.J.; Sain, S.R.; Greasby, T.A.; Uejio, C.K.; Hayden, M.H.; Monaghan, A.J.; Boehnert, J.; Sampson, K.; Banerjee, D.; Nepal,
V.; et al. Characterizing urban vulnerability to heat stress using a spatially varying coefficient model. Spat. Spatio-Temporal
Epidemiol. 2014, 8, 23–33. [CrossRef] [PubMed]
16. Madrigano, J.; Ito, K.; Johnson, S.; Kinney, P.L.; Matte, T. A Case-Only Study of Vulnerability to Heat Wave—Related Mortality.
Environ. Health Perspect. 2015, 123, 672–678. [CrossRef] [PubMed]
17. Hsu, A.; Manya, D. Intensity across Major US Cities. Nat. Commun. 2021, 12, 2721. [CrossRef] [PubMed]
18. Yang, W.; Wong, N.H.; Jusuf, S.K. Thermal comfort in outdoor urban spaces in Singapore. Build. Environ. 2013, 59, 426–435.
[CrossRef]
19. Aljawabra, F.; Nikolopoulou, M. Thermal comfort in urban spaces: A cross-cultural study in the hot arid climate. Int. J. Biometeorol.
2018, 62, 1901–1909. [CrossRef]
20. Salata, F.; Golasi, I.; Vollaro, R.D.L.; Vollaro, A.D.L. Outdoor thermal comfort in the Mediterranean area. A transversal study in
Rome, Italy. Build. Environ. 2016, 96, 46–61. [CrossRef]
21. Huang, J.; Zhou, C.; Zhuo, Y.; Xu, L.; Jiang, Y. Outdoor thermal environments and activities in open space: An experiment study
in humid subtropical climates. Build. Environ. 2016, 103, 238–249. [CrossRef]
22. Xu, M.; Hong, B.; Mi, J.; Yan, S. Outdoor thermal comfort in an urban park during winter in cold regions of China. Sustain. Cities
Soc. 2018, 43, 208–220. [CrossRef]
23. Hirashima, S.Q.D.S.; de Assis, E.S.; Nikolopoulou, M. Daytime thermal comfort in urban spaces: A field study in Brazil. Build.
Environ. 2016, 107, 245–253. [CrossRef]
24. Middel, A.; Selover, N.; Hagen, B.; Chhetri, N. Impact of shade on outdoor thermal comfort—A seasonal field study in Tempe,
Arizona. Int. J. Biometeorol. 2016, 60, 1849–1861. [CrossRef]
Buildings 2023, 13, 1040 18 of 19

25. Chi, G.Q.; Ho, H.C. Population stress: A spatiotemporal analysis of population change and land development at the county level
in the contiguous United States, 2001–2011. Land Use Policy 2018, 70, 128–137. [CrossRef]
26. Chi, J.; Zhu, G. Spatial Regression Models for the Social Sciences; SAGE Publications: Thousand Oaks, CA, USA, 2019.
27. Toparlar, Y.; Blocken, B.; Maiheu, B.; van Heijst, G. A review on the CFD analysis of urban microclimate. Renew. Sustain. Energy
Rev. 2017, 80, 1613–1640. [CrossRef]
28. Crank, P.J.; Sailor, D.J.; Ban-Weiss, G.; Taleghani, M. Evaluating the ENVI-met microscale model for suitability in analysis of
targeted urban heat mitigation strategies. Urban Clim. 2018, 26, 188–197. [CrossRef]
29. Ozkeresteci, I.; Crewe, K.; Brazel, A.J.; Bruse, M. Use and evaluation of the ENVI-met model for environmental design and
planning. An Experiment on Lienar Parks. In Proceedings of the 21st International Cartographic Conference (ICC), Durban,
South Africa, 10–16 August 2003; pp. 10–16.
30. Rodríguez Algeciras, J.A.; Gómez Consuegra, L.; Matzarakis, A. Spatial-temporal study on the effects of urban street configura-
tions on human thermal comfort in the world heritage city of Camagüey-Cuba. Build. Environ. 2016, 101, 85–101. [CrossRef]
31. Sinsel, T. Advancements and Applications of the Microclimate Model ENVI-Met; Johannes Gutenberg-Universitat Mainz: Mainz,
Germany, 2022. [CrossRef]
32. Ayyad, Y.N.; Sharples, S. Envi-MET validation and sensitivity analysis using field measurements in a hot arid climate. IOP Conf.
Series Earth Environ. Sci. 2019, 329, 012040. [CrossRef]
33. Acero, J.A.; Arrizabalaga, J. Evaluating the performance of ENVI-met model in diurnal cycles for different meteorological
conditions. Theor. Appl. Clim. 2018, 131, 455–469. [CrossRef]
34. Acero, J.A.; Herranz-Pascual, K. A comparison of thermal comfort conditions in four urban spaces by means of measurements
and modelling techniques. Build. Environ. 2015, 93, 245–257. [CrossRef]
35. Alsaad, H.; Hartmann, M.; Hilbel, R.; Voelker, C. ENVI-met validation data accompanied with simulation data of the impact of
facade greening on the urban microclimate. Data Brief 2022, 42, 108200. [CrossRef]
36. Morakinyo, T.E.; Lai, A.; Lau, K.K.-L.; Ng, E. Thermal benefits of vertical greening in a high-density city: Case study of Hong
Kong. Urban For. Urban Green. 2019, 37, 42–55. [CrossRef]
37. Forouzandeh, A. Numerical modeling validation for the microclimate thermal condition of semi-closed courtyard spaces between
buildings. Sustain. Cities Soc. 2018, 36, 327–345. [CrossRef]
38. Elraouf, R.A.; Elmokadem, A.; Megahed, N.; Eleinen, O.A.; Eltarabily, S. Evaluating urban outdoor thermal comfort: A validation
of ENVI-met simulation through field measurement. J. Build. Perform. Simul. 2022, 15, 268–286. [CrossRef]
39. US Census Bureau. 2022. Available online: https://www.census.gov/quickfacts/fact/table/US/PST045222 (accessed on 23
January 2023).
40. Mayor’s Office of Sustainability and ICF International. Growing Stronger: Toward a Climate-Ready Philadelphia. 2015. Available
online: https://www.phila.gov/documents/growing-stronger-toward-a-climate-ready-philadelphia/ (accessed on 1 March
2023).
41. Schade, R.S. Philadelphia Rowhouse Manual: A Practical Guide for Homeowners; The City of Philadelphia: Philadelphia, PA, USA,
2008. Available online: https://www.phila.gov/media/20190521124726/Philadelphia_Rowhouse_Manual.pdf (accessed on 18
July 2021).
42. Baechler, P.M.; Michael, C.; Williamson, J.L.; Gilbride, T.L.; Cole, P.C.H.; Marye, G.L. Building America Best Practices Series:
Volume 7.1: Guide to Determining Climate Regions by County. 2010. Available online: https://www.osti.gov/biblio/1068658
(accessed on 22 July 2021).
43. Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry/Geospatial Research, Analysis,
and Services Program. 2018. Available online: https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.
html (accessed on 15 June 2021).
44. U.S. Census Bureau. 2009. Available online: https://www.census.gov/library/publications/2008/compendia/statab/128ed.
html (accessed on 21 August 2021).
45. Bruse, M.; Fleer, H. Simulating surface–plant–air interactions inside urban environments with a three dimensional numerical
model. Environ. Model. Softw. 1998, 13, 373–384. [CrossRef]
46. Shaw, E.W. Thermal Comfort: Analysis and applications in environmental engineering, by P. O. Fanger. 244 pp. DANISH
TECHNICAL PRESS. Copenhagen, Denmark, 1970. Danish Kr. 76, 50. R. Soc. Health J. 1972, 92, 164. [CrossRef]
47. Havenith, G.; Holmér, I.; Parsons, K. Personal factors in thermal comfort assessment: Clothing properties and metabolic heat
production. Energy Build. 2002, 34, 581–591. [CrossRef]
48. Pennsylvania Spatial Data Access PASDA. The Pennsylvania State University. 2016. Available online: http://www.pasda.psu.
edu/ (accessed on 15 August 2021).
49. Hashemi, F.; Marmur, B.; Passe, U.; Thompson, J. Developing a workflow to integrate tree inventory data into urban energy
models. In SimAUD 2018 Symposium on Simulation for Architecture and Urban Design; Simulation Series; TU Delft: Delft, The
Netherlands, 2018; Volume 50.
50. Hashemi, F.; Iulo, L.D.; Poerschke, U. A novel approach for investigating canopy heat island effects on building energy
performance: A case study of Center City of Philadelphia, PA. In Proceedings of the AIA/ACSA Intersections Research
Conference: CARBON, Fall Conference Proceedings, Washington, DC, USA, 30 September – 2 October 2020; pp. 197–203.
[CrossRef]
Buildings 2023, 13, 1040 19 of 19

51. Roudsari, M.S.; Pak, M.; Smith, A.; Gill, G. Ladybug: A Parametric Environmental Plugin For Grasshopper To Help Designers
Create An Environmentally-conscious Design. In Proceedings of the Building Simulation, Chambéry, France, 26–28 August 2013;
pp. 3128–3135. [CrossRef]
52. Stewart, I.D.; Oke, T.R. Local Climate Zones for Urban Temperature Studies. Bull. Am. Meteorol. Soc. 2012, 93, 1879–1900.
[CrossRef]
53. Ali-Toudert, F.; Mayer, H. Numerical study on the effects of aspect ratio and orientation of an urban street canyon on outdoor
thermal comfort in hot and dry climate. Build. Environ. 2006, 41, 94–108. [CrossRef]
54. de Abreu-Harbich, L.V.; Labaki, L.C.; Matzarakis, A. Effect of tree planting design and tree species on human thermal comfort in
the tropics. Landsc. Urban Plan. 2015, 138, 99–109. [CrossRef]
55. Gatto, E.; Buccolieri, R.; Aarrevaara, E.; Ippolito, F.; Emmanuel, R.; Perronace, L.; Santiago, J.L. Impact of Urban Vegetation
on Outdoor Thermal Comfort: Comparison between a Mediterranean City (Lecce, Italy) and a Northern European City (Lahti,
Finland). Forests 2020, 11, 228. [CrossRef]
56. Zaki, S.A.; Toh, H.J.; Yakub, F.; Saudi, A.S.M.; Ardila-Rey, J.A.; Muhammad-Sukki, F. Effects of Roadside Trees and Road
Orientation on Thermal Environment in a Tropical City. Sustainability 2020, 12, 1053. [CrossRef]
57. Bernabé, A.; Bernard, J.; Musy, M.; Andrieu, H.; Bocher, E.; Calmet, I.; Kéravec, P.; Rosant, J.-M. Radiative and heat storage
properties of the urban fabric derived from analysis of surface forms. Urban Clim. 2015, 12, 205–218. [CrossRef]
58. Elraouf, R.A.; Elmokadem, A.; Megahed, N.; Eleinen, O.A.; Eltarabily, S. The impact of urban geometry on outdoor thermal
comfort in a hot-humid climate. Build. Environ. 2022, 225, 109632. [CrossRef]
59. Zhang, Y.; Du, X.; Shi, Y. Effects of street canyon design on pedestrian thermal comfort in the hot-humid area of China. Int. J.
Biometeorol. 2017, 61, 1421–1432. [CrossRef] [PubMed]
60. Zaki, S.; Azid, N.; Shahidan, M.; Hassan, M.; Daud, M.; Abu Bakar, N.; Salim, S.A.Z.S.; Yakub, F. Analysis of Urban Morphological
Effect on the Microclimate of the Urban Residential Area of Kampung Baru in Kuala Lumpur Using a Geospatial Approach.
Sustainability 2020, 12, 7301. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.