Landscape and Urban Planning 146 (2016) 29–42

Contents lists available at ScienceDirect

Landscape and Urban Planning
journal homepage: www.elsevier.com/locate/landurbplan

Research Paper

Hot playgrounds and children’s health: A multiscale analysis of
surface temperatures in Arizona, USA
Jennifer K. Vanos a,∗ , Ariane Middel b , Grant R. McKercher a , Evan R. Kuras c ,
Benjamin L. Ruddell d
a

Atmospheric Sciences Research Group, Department of Geosciences, Texas Tech University, Lubbock, TX 79409-1053, USA
School of Geographical Sciences and Urban Planning, Arizona State University, Tempe, AZ 85281, USA
c
Department of Environmental Conservation, University of Massachusetts at Amherst, Amherst, MA 01002, USA
d
Fulton Schools of Engineering, Arizona State University, Tempe, AZ 85281, USA
b

h i g h l i g h t s
•
•
•
•
•

A mismatch exists between remotely sensed and in situ urban surface temperatures (Ts ).
The hottest Ts in a Phoenix area neighborhood were found on playground surfaces.
Children are more vulnerable to the effects of heat stress and high Ts than adults.
Shade of any type is found effective in reducing Ts and improving thermal safety.
Data must be collected at the touch-scale for spatially accurate high Ts mitigation.

a r t i c l e

i n f o

Article history:
Received 2 July 2015
Received in revised form 23 October 2015
Accepted 27 October 2015
Available online 10 November 2015
Keywords:
Surface temperature
Urban climate
Children’s health
Playgrounds
Bioclimatic urban design
Geographic scale

a b s t r a c t
Objectives: To provide novel quantification and advanced measurements of surface temperatures (Ts ) in
playgrounds, employing multiple scales of data, and provide insight into hot-hazard mitigation techniques and designs for improved environmental and public health.
Methods: We conduct an analysis of Ts in two Metro-Phoenix playgrounds at three scales: neighborhood
(1 km resolution), microscale (6.8 m resolution), and touch-scale (1 cm resolution). Data were derived
from two sources: airborne remote sensing (neighborhood and microscale) and in situ (playground site)
infrared Ts (touch-scale). Metrics of surface-to-air temperature deltas (Ts–a ) and scale offsets (errors)
are introduced.
Results: Select in situ Ts in direct sunlight are shown to approach or surpass values likely to result in burns
to children at touch-scales much finer than Ts resolved by airborne remote sensing. Scale offsets based
on neighbourhood and microscale ground observations are 3.8 ◦ C and 7.3 ◦ C less than the Ts–a at the
1 cm touch-scale, respectively, and 6.6 ◦ C and 10.1 ◦ C lower than touch-scale playground equipment Ts ,
respectively. Hence, the coarser scales underestimate high Ts within playgrounds. Both natural (tree) and
artificial (shade sail) shade types are associated with significant reductions in Ts .
Conclusions: A scale mismatch exists based on differing methods of urban Ts measurement. The sub-meter
touch-scale is the spatial scale at which data must be collected and policies of urban landscape design and
health must be executed in order to mitigate high Ts in high-contact environments such as playgrounds.
Shade implementation is the most promising mitigation technique to reduce child burns, increase park
usability, and mitigate urban heating.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction

∗ Corresponding author.
E-mail addresses: jennifer.vanos@ttu.edu (J.K. Vanos), amiddel@asu.edu
(A. Middel), grant.mckercher@ttu.edu (G.R. McKercher), erkuras@gmail.com
(E.R. Kuras), bruddell@asu.edu (B.L. Ruddell).
http://dx.doi.org/10.1016/j.landurbplan.2015.10.007
0169-2046/© 2015 Elsevier B.V. All rights reserved.

A neighborhood’s thermal environment is complex. It is scale
dependent and influenced by numerous physical characteristics
such as weather, urban design, and the thermal properties of building and ground materials (Yaghoobian, Kleissl, & Krayenhoff, 2010).
This complexity results in processes that impact health at several

30

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

scales, yet urban heat-health research often employs coarse scales
collected from sparse standardized meteorological observations
(BASC, 2012; Kuras, Hondula, & Brown-Saracino, 2015), which gives
little evidence of linkages between urban form, air temperature,
and surface temperature at the human scale.
Microscale urban climate models and remotely sensed data (on
the order of 1 m2 to 1 km2 ) are often employed for heat stress analysis (e.g., Harlan, Declet-Barreto, Stefanov, & Petitti, 2013; Hondula
et al., 2012; Masson, Champeaux, Chauvin, Meriguet, & Lacaze,
2003; Mishra, Ganguly, Nijssen, & Lettenmaier, 2015; Stefanov,
Prashad, Eisinger, Brazel, & Harlan, 2004), with airborne remotely
sensed data providing surface temperatures at a resolution ranging
from 7 to 140 m (Stefanov et al., 2004). These scales may possess insufficient spatial resolution to determine certain personal
health impacts based on micro-environmental heat exposure. The
human-scale ranges from ∼1.0 m for personal interactions with the
proximate atmospheric and radiative microenvironment, to ∼1 cm
for touch-scale interactions. At this touch-scale, extreme temperatures can cause burns or damage to the skin, with a severity
that depends greatly on the object’s initial temperature, its material properties, and the contact thermal conductance (ISO 13732,
2010; Ungar & Stroud, 2010). In an outdoor urban environment,
touch-scale surface temperature is based on urban design, material, orientation, and sun angle (e.g., Ketterer & Matzarakis, 2014a;
Krayenhoff & Voogt, 2007). The touch-scale is on the order of 100×
to 10,000× finer than that which remotely sensed data and urban
climate models can currently provide. This variance therefore complicates spatially accurate design guidelines and policies.
The current study addresses the surface temperatures (ground
and equipment) in small urban public playgrounds frequented
by children aged 2–12 (ASTM F1487, 2011; CPSC, 2010). Urban
parks, childcare play spaces, and urban playgrounds are important
resources for urban sustainability, physical activity, and community health (Moore & Cosco, 2014; Vanos, 2015; Wolch et al., 2011).
In the absence of design measures to mitigate heating, surfaces
may become dangerously hot (e.g., Fig. 1), and playgrounds may
become microscale heat islands that enhance, rather than mitigate, the larger urban heat island effect (UHI) (e.g., Moogk-Soulis,
2010). On clear, warm days with direct solar radiation, exposure
to the mean radiant temperature (Tmrt )—the combination of all
short- and long-wave radiant fluxes (Thorsson, Honjo, Lindberg,

Eliasson, & Lim, 2007)—becomes the most significant agent of heat
gain to humans (Johansson, Thorsson, Emmanuel, & Krüger, 2014;
Kántor & Unger, 2011; Mayer & Höppe, 1987), as well as to surfaces. The open design of urban parks with high radiant heat loads
does not provide conducive spaces for safe physical activity and
thermal comfort compared to shaded areas on warm–hot days
(Koppe & Jendritzky, 2005; Matzarakis & Endler, 2010; Vanos, 2015;
Vanos, Warland, Gillespie, Slater, et al., 2012). However, Tmrt can be
reduced through design by (1) controlling incoming solar radiation
with shade, indirectly reducing surface temperatures; and (2) by
modifying surface materials to reduce surficial longwave radiation
flux, indirectly reducing surface and air temperatures.
Quantifying temperature exposures (surface and air) in urban
areas conserved for active use (i.e., parks and playgrounds) is
important for understanding health effects (McGeehin & Mirabelli,
2001), and to adapt to future urban heating effects (Harlan &
Ruddell, 2011; McCarthy & Sanderson, 2011), urban land use
change (Adachi et al., 2013), and increasing temperatures with projected climate change (Meehl & Tebaldi, 2004; Tebaldi, Hayhoe,
Arblaster, & Meehl, 2006). Most cities are not designed to ameliorate the effects of warming, although it is well known that this
is possible through evidence-based climate-responsive design of
urban spaces (Brown, Vanos, Kenny, & Lenzholzer, 2015; Erell,
Pearlmutter, & Williamson, 2012; Masson et al., 2014; Middel,
Chhetri, & Quay, 2015). Properly designed playgrounds contribute
to microscale cooling, serving as heat refuges through the summer
season (Cheng, Wei, Chen, Li, & Song, 2014; Chow, Pope, Martin, &
Brazel, 2011; Vanos, Warland, Gillespie, Slater, et al., 2012), as well
as comfortable and safe places for children to play and engage in
activities for improved health and well-being (Ciucci et al., 2013;
Moore & Cosco, 2014; Vanos, 2015; Wolch et al., 2011).
Children are uniquely sensitive and vulnerable to hot ambient environments as compared to adults (Balbus & Malina, 2009),
mainly due to their high ratios of both metabolism-to-surface
area, and surface-area-to-body-mass (Falk & Dotan, 2008; Wenger,
2003), resulting in thermoregulatory inferiority relative to adults
during physical activity in hot conditions (Falk & Dotan, 2008).
Such hot conditions result in air temperature becoming greater
than skin temperature, and convective heat (which is normally
a loss of heat from the skin’s surface) becomes a gain and thus
a quicker path to heat stress for children. In the same sense,

Fig. 1. Surface temperature images photographed in the study playgrounds using Infrared Thermography (IRT) (FLIR camera, TG165). (A) Slide and black/green rubber ground
surface in sun (71 ◦ C on slide; 82 ◦ C on rubber), and under shade sail (blue/green); (B) playground steps in sun, black powder-coated metal (58 ◦ C). Photos were taken at
1045 h LST. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

31

Table 1
Burn thresholds when skin is in contact for short periods of time (3 s, 5 s, 1 min) with hot surfaces made of materials commonly found within playgrounds. Thresholds of
materials with similar heat conductivity are combined to represent one value.
Material

Burn threshold (◦ C)

Material characteristics
Contact time

Metal
Coated metala

Stone material
Plasticb
Wood

3s

Uncoated
Lacquer coat: 100 ␮m
Powder: 90 ␮m
Enamel: 160 ␮m
Polyamid 11 or 12: 400 ␮m
Concrete, granite, asphalt
Polyamide, acrylglass, polytetrafluorethylene, duroplastic
Bare, low moisture

5s
◦

60 C
68 ◦ C
65 ◦ C
63 ◦ C
77 ◦ C
73 ◦ C
77 ◦ C
99 ◦ C

1 min
◦

57 C
61 ◦ C
60 ◦ C
59 ◦ C
70 ◦ C
60 ◦ C
74 ◦ C
93 ◦ C

51 ◦ C
51 ◦ C
51 ◦ C
51 ◦ C
51 ◦ C
56 ◦ C
60 ◦ C
60 ◦ C

Source: ISO 13732 (2010).
a
Polyurethane enamel-coated steel is used predominantly in the study site playgrounds for hold/touch surfaces, and powder coated steel for walking surfaces.
b
UV stabilized high-density polyethylene (HDPE) used in playgrounds is similar in material properties to polyamide.

children will cool off more quickly if a cooler space is provided,
such as a shaded area. Children’s sweating rates are also lower
than that of adults, thus they rely mainly on dry heat dissipation
through convection (Falk & Dotan, 2008), which is not effective in
extreme heat. Lastly, children are more susceptible to burn injuries
on playground equipment due to more sensitive skin (Oliveria,
Saraiya, Geller, Heneghan, & Jorgensen, 2006), slower reflexes to
remove their hand from what may be causing the burn (ISO 13732,
2010), and the lack of experience in the causes and effect of
burns (Ford, Moriarty, Riches, & Walker, 2011). This lack of experience results in differences in the perception of heat between
children and adults, where expectations and experience, perception, personal preference, and feedback mechanisms related to the
thermal environment (e.g., Knez, Thorsson, Eliasson, & Lindberg,
2009; Metje, Sterling, & Baker, 2008; Parsons, 2003) are more
limited in children, affecting the interpretation of their physical state and subsequent adaptive behavior to avoid heat stress
(Vanos, 2015).
Due to many of the above-mentioned cumulative factors,
children experience burns on playgrounds even in moderate midlatitude climates (Ford et al., 2011; Strong, Tahir, & Verma, 2007).
Burns can occur in 3 s on solid metal surfaces with temperatures
≥60 ◦ C (ASTM 1055, 2014; CPSC, 2010; ISO 13732, 2010; Ungar &
Stroud, 2010); for plastics or nonmetallic coatings that are more
insulating, higher threshold temperatures are set (e.g., ≥77 ◦ C for
plastic ISO 13732 (2010), see Table 1). Guidelines in the United
States for public playground safety are put forth by the U.S. Consumer Product Safety Commission (CPSC) (CPSC, 2010) and contain
suggestions and information for the use of materials and coatings,
as well as the orientation of objects with respect to sun angles; however, in situ quantitative evidence is lacking to adequately guide
designers on the implementation and material of equipment based
on bioclimatic design principles. Although the guidelines are set
by the CPSC to minimize injury to children, a small number of
burns continue to be reported, e.g., 14 severe burns in American
playgrounds reported between 2001 and 2008 (Ford et al., 2011;
O’Brien, 2009).
Accordingly, the goal of this study is to provide novel
quantitative insight into the multiscale issues of the thermal
microenvironment in two playgrounds in Gilbert, Arizona. We
focus on touch-scale surface temperature variations in comparison with remotely sensed surface temperatures (6.8 m resolution)
and the real-world applicability of such data based on scale. The
role of shade (natural and artificial) in altering surface temperatures to mitigate burn risk is also quantified at the touch-scale.
We address how overlooking sub-grid scale variation in airborne
remotely sensed surface temperatures in playgrounds may result in
inferior evidence for planning, decision-making, and the establishment of equipment safety standards related to playground burns
and heat stress in children.

2. Methods
2.1. Study site
Within the Master Planned community of Power Ranch
(33.27040, −111.69527) in Gilbert, Arizona, USA, two playgrounds
(denoted as the North and South Playground) comprised the in situ
sites in the current study (Fig. 2). This community is in the southeast valley of metropolitan Phoenix, just on the edge of Phoenix’s
large UHI effect (Chow, Brennan, & Brazel, 2012). Phoenix is the
U.S. metropolitan area with the highest summer temperatures
(National Oceanic and Atmospheric Administration, 2012), averaging 74 days per year with daily maximum air temperatures at or
above 40 ◦ C. These high temperatures are shown to be associated
with the highest risk for heat-related morbidity (Petitti, Hondula,
Yang, Harlan, & Chowell, 2015), which can be particularly dangerous for the most socially vulnerable, low income, and isolated
populations (Harlan et al., 2013). Moreover, the desert Southwest
is warming faster than most US regions and is projected to warm at
a high rate in the coming decades (Karl, Melillo, & Peterson, 2009;
Walsh et al., 2014), with the temperature change in the hottest
days projected to increase by approximately 5.5–8.5 ◦ C under a high
emissions scenario (Luber et al., 2014).
The Power Ranch community features parks, playgrounds, walking trails, ponds, splash pads, swimming pools, athletic fields, and
meeting spaces. A dozen adjacent neighborhoods comprise Power
Ranch, each organized around a central park (which contains the
North Playground) where residents walk dogs, exercise, and congregate. Two elementary schools within the community host over
1400 students combined (HUSD, 2015a,b). 13.2% of the residential
population is less than 5 years of age and 27.9% is between the ages
of 5 and 17. The median age of Power Ranch residents is 27.4 years
and the median household income is $75,642 (U.S. Census Bureau).
The community is predominantly white (91.0%), with 4.2% Black or
African American ethnicity, 4.7% Asian, 20.6% Hispanic or Latino,
and 2.9% identifying as ‘other’.
The locations of the two playgrounds are displayed in Fig. 2,
with examples of the playgrounds and equipment provided in the
Supplemental material. Each playground contains numerous pieces
of common playground equipment and 2–3 benches. The North
playground spans a rectangular area of 497 m2 , and its surface cover
is predominantly sand and concrete, with some black/green soft
rubber. A swimming pool and basketball court are nearby. The
NASA MODIS/ASTER Airborne Simulator (MASTER, Hook, Myers,
Thome, Fitzgerald, & Kahle, 2001) overflight took place on July 13,
2011 between 1100 and 1330 h, collecting land surface temperatures (LST) at a resolution of 6.8 m (see Section 2.2.1 for MASTER
details). On this date, there was no shade sail present in the North
playground; however, in 2013 when the in situ data collection took
place, a large shade sail (12 m × 17 m) was installed in the North

32

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

Fig. 2. (A) Location of the state of Arizona (green shading) in the USA, and the Phoenix Metropolitan Area (PMA) (star with PHX label); (B) map of east PMA (with the city
center to the west), identifying the location of the study site of Gilbert containing the Power Ranch Neighborhood in yellow; (C) Visible overhead image of the Power Ranch
neighborhood with playground location in red boxes; (D) Thermal Infrared MASTER super spectral image of Power Ranch neighborhood obtained midday (1100–1330 h) on
July 13, 2011.

playground. The addition of the shade sail between the two study
years provides a unique opportunity for a natural experiment to
assess the improvement in the thermal environment created by
the investment in the sail. Surfaces that are adequately shaded
will generally demonstrate little-to-no difference from air temperature during the daytime. Within the 497 m2 rectangular area of the
North playground, no trees are present, however three ash trees
to the southwest of the area provide shade in the late afternoon
hours. The common tree types found within the 25 m radius from
the center of the North playground are the southwestern urban tree
species of Shamel Ash, Palo Verde, Western Cedar, Willow Acacia, and Mesquite. Hence, in the area outside of the playground
area, tree canopy provides some shade, however minimal to the
playground itself.
The South playground is circular, 402 m2 in area, with sand
as the playing surface. A basketball court is nearby, and there is
no shade sail present. Within the circular playground itself, two
Cottonwood trees are found with canopy covering ∼5% of the
play area. The common tree types found within the 25 m radius
of the playground’s center include Southern Live Oak, Western
Cedar, Cottonwood, and Shamel Ash. The two Cedar trees and the
Cottonwood tree to the south and southwest of the playground
provide shade in the late afternoon hours. For both parks, the trees

are approximately 15–20 years of age, with a mature height, are
well-maintained, and in good health. Finally, trees were in full leaf
with average shading coefficients ranging from 0.65 (Mesquite) to
0.85 (Cottonwood, Palo Verde).
2.2. Data acquisition
2.2.1. Airborne data
Superspectral data acquired by the NASA MODIS/ASTER Airborne Simulator (MASTER) (Hook et al., 2001) were used to
investigate LST for the full Power Ranch neighborhood, including
the two playground study sites. MASTER is an aircraft mounted
multispectral sensor system that provides the same observations
as NASA’s MODIS and ASTER satellites, but at higher resolution
owing to the much lower altitude of airborne observations (e.g.,
6.8 m resolution for the Thermal Infrared (TIR) product) (Hook
et al., 2001). This TIR data captures near-peak surface temperatures
(Ts ) with a higher spatial and spectral resolution compared to many
sensor systems (e.g., Stefanov et al., 2004; Van Der Meer & de
Jong, 2001). The MASTER data used in this study were acquired
midday (1100–1330 h LST) on July 13, 2011 (Jenerette et al., 2015).
The geo-referenced LST data are further extracted by each parcel
polygon (obtained from cadastral GIS vector layer) to represent the

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

33

Fig. 3. Diurnal variations in air temperature, dew point temperature, and solar radiation for the two dates of comparisons (July 13, 2011 and September 15, 2014).

parcel mean LST (see Fig. 2). Further details regarding the MASTER
overflight and processing of the data to obtain LST can be found
in Jenerette et al. (2015). The retrieved resolution (6.8 m) of the
Ts data from the MASTER sensor allow for microclimate-related
analysis and also allow for comparison with touch-scale in situ site
playground measurements detailed below.
2.2.2. Playground site data
On September 15, 2014, in situ ground-based microclimate data
were collected within each playground from 1200 h to 1400 h LST
using research-grade instruments obtained from Campbell Scientific (Logan, Utah). Air temperature (Ta ) was measured using high
accuracy fine-wire chrome–constantan (Type ‘E’) thermocouples
at 1 m and 2 m height. A secondary measurement of Ta and relative humidity was taken using an HC-S3 sensor probe at 1 m and
2 m height. Ta collected at a 2 m height was used for analyses. Total
incoming solar radiation was measured using an SC300 silicon photovoltaic pyranometer. All above-listed instruments were securely
mounted on a mobile golf cart. For September 2011, air and dew
point temperatures and solar radiation were collected using a scientific grade WeatherHawk signature meteorological station, 3 m
height over grass, located approximately 100 m from the center
of Power Ranch with comparable measurements calibrated to the
same standard as the mobile station. This was used for comparative purposes with the in situ field day collected at 2 m over grass
on September 2014, as displayed in Fig. 3.
Two types of in situ infrared thermometry (IRT) Ts measurements were made: (1) hand IRT measurements taken at 1 cm
perpendicular to the surface of interest using a DeltaTrak ThermoTrace Handheld infrared thermometer (Pleasanton, CA); (2)
cart-mounted measurements (front of cart mounted 30 cm from
surface) with SI-111 Precision Infrared Radiometers (Campbell Scientific, Logan, Utah). Emissivity corrections were made for surfaces
with low emissivity, as the IRTs assume an emissivity equal to
0.95. Surface temperature images were also taken using Infrared
Thermography (FLIR camera, TG165) (e.g., Fig. 1).
Selection of surfaces for hand IRT was based on an intersection
of the following two criteria: (1) objects presenting high quality

sampling opportunities featuring both full direct perpendicular sun
angle and full shade; (2) objects that had a high likelihood to be
walked on, sat on, or held for ≥3 s (such as the ground, steps, slides,
benches, handrails, and seats). Full shade and direct perpendicular sun exposure were chosen to represent the extreme bounding
conditions inclusive of all types of solar angles and partial shading.
Surface solar exposure was of three classes: full direct perpendicular sun, natural shade (i.e., tree), and artificial shade (e.g., shade
sail, pavilion, building shadow).
2.2.3. Multiscale metrics and temporal analyses
In order to standardize the Ts for comparative purposes between
scales, we calculate the relationship of the difference (‘delta’)
between the surface temperature and 2 m air temperature (Ts ).
This delta method allows for a relatively unbiased comparison
of the Ta between the three scales of measurement (touch-,
micro-, and neighborhood scales) using the Ta values that are
in approximate equilibrium with respect to time and sun/shade.
This is in contrast to using only the magnitude of Ts between
two differing surfaces in sun and/or shade that do not have
standard Ts values at the different scales for comparison. This
is particularly important for the playground equipment, where
the touch-scale deltas between Ta and Ts are hypothesized to
increase as we approach maximum Ta (mid-afternoon). However,
this delta is expected to be lesser at the 1 km scale (neighborhood) at mid-afternoon, where due to the coarse-scale averaging,
the measurements of Ts will be closer to that of Ta (hence,
Ta 1km ∼
= Ts 1km ).
A second metric is also employed to compare deltas by scale,
denoted as the ‘scale offset’. The scale offset represents the difference of the T̄s–a between scales. We first assess the mean
differences within the scale making use of the delta method above
(T̄s–a ), and secondly we calculate the differences in the delta
between scales, which gives the scale offset (representing an error).
This offset provides dual benefits for interpretation of results in
the current study as it relates both T̄s–a and T̄s at the different
scales of observations, thus demonstrating how spatial uncertainty
is influenced by the context of measurement.

34

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

Fig. 4. Frequency histograms of surface temperatures in (A) Full Power Ranch neighborhood, including parks and playgrounds, and (B) study site playgrounds and their
immediate surroundings within the Power Ranch neighborhood.

Further interpretation employs the sky view factor (SVF) for
each location where a touch-scale Ts measurement was taken.
Fisheye photos of the sky were taken at each observational stop
at a height of 1.5 m. The SVFs were calculated from these photos using Rayman 2.1 (Matzarakis, Rutz, & Mayer, 2010). All hand
IRT readings were completed in less than 10 min in the time periods surrounding 1230 h LST; however, the cart-mounted Ts were
obtained in a neighborhood transect during a 2 h sampling period.
To account for the warming of the air and surfaces that takes place
during each transect, observations were normalized to a fixed point
in time (time detrended), assuming a linear increase in Ts over the
2 h period. An empirical warming rate was calculated using a linear
regression based on the observations. Using the regression equation, observations were normalized to 1230 h LST for comparison
to playground hand IRT readings.
Although time detrending has the benefit of making the data
points comparable across time, comparing derived measurements
(e.g., T̄s–a and scale offsets) using a time detrending function that
is based on all samples is unfavorable as different surface types
have different heating/cooling rates in the sun/shade. For example,
if a heating rate for sun-exposed asphalt is applied to shaded grass,
then the function will overestimate the heating correction for grass
Ts . During sampling in the current study, all sun-exposed surfaces
were collected within the first 15 min of analysis, thus heating was

minimal. Therefore, this is merely an issue with shaded surfaces
that were sampled 30–60 min later. However, the sun-exposed surfaces heat at a much faster rate introducing bias if using the all
sample (‘global’) regression function for shaded surfaces (i.e., the
regression over-corrects for heating that took place on sun exposed
surfaces but not on shaded surfaces). As this error produced by the
regression function increases over time, we employed the Ts data
in their original state, even with slight heat/cooling variability, and
accept a slight warming of the shaded surfaces (1 ◦ C for grass and
2 ◦ C for concrete over the transect period) that may not be comparable across time. This error is only present in the tree analysis
and does not affect the playground surface Ts . In order to calculate an average-rate correction factor for the time detrending for
each surface type and each exposure separately, further in situ data
are required, which will build upon the foundation provided by the
current study.
3. Results
3.1. Daily microclimate variations
Fig. 3 displays the daily patterns of air and dew point temperature, and incoming solar radiation on July 13, 2011 and Sept
15, 2014. Both days were classified as ‘Dry Tropical’ weather

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

35

Table 2
In situ surface temperature measurements taken at site playgrounds, September 15, 2014. Temperatures measured (1200–1300 h LST). Corresponding materials, thermal
conductivities, and average albedos for each surface also listed (ASTM C1057, 2012; Campbell & Norman, 1998; Environmental Protection Agency, 2014; ISO 13732, 2010;
Oke, 1988; Paul, Pal, Ghosh, & Chakraborty, 2004).
Equipment

Thermal cond (W m−1 K−1 )

Albedo mean (range)

Playground surface temperatures
Ts (◦ C) Sun

Ts (◦ C) Shade (tree)

Ts (◦ C) Shade (sail)

Seating use children
Slidea (beige)
Bouncy spring ridera seat (green)
Bouncy spring ridera seat (red)
Bouncy spring ridera seat (yell/blue)
Slidea (green)

0.25–0.51
0.25–0.51
0.25–0.51
0.25–0.51
0.25–0.51

0.18 (0.15–0.20)
0.18 (0.15–0.20)
0.18 (0.15–0.20)
0.18 (0.15–0.20)
0.18 (0.15–0.20)

63.9
55.0
57.2
46.7
71.7

40.6
44.4
45.0
–
43.9

–
–
–
37.8
–

Walking use (foot contact)
Asphalt
Sand
Steps and Walkwayb (black)
Basketball surface asphalt (green paint)
Basketball surface asphalt (white paint)
Rubber soft ground surface (blk/green)
Splash padc (white/blue)
Concrete
Grass
Pool deckd (light gray)
Crushed granite

0.75
0.15–0.25
45.20
0.75
0.75
0.29
0.29–0.62
1.00–2.50
1.10
1.00–1.80
0.20–0.90

0.13 (0.05–0.20)
0.33 (0.25–0.40)
0.08 (0.02–0.14)
0.28 (0.05–0.50)
0.38 (0.25–0.50)
0.10 (0.05–0.15)
0.38 (0.30–0.45)
0.38 (0.30–0.45)
0.25
0.38 (0.30–0.45)
0.57 (0.14–1.00)

53.7–60.7
59.4–62.8
60.0
60.0
51.7
87.2
42.2
49.5–56.0
38.1–40.1
52.2
54.2

32.9–40.2
32.8–33.9
39.4
–
–
42.2
27.8
31.8–40.5
30.6–43.2
–
35.0–42.2

–
33.9–35.3
41.1
–
–
46.7
–
35.2
–
–
–

Holding (hand contact)
Balance rail (red)e
Bouncy spring rider handlee (green)
Bouncy spring rider handlee (red)
Posts of playground structuree (purple)
Posts of playground structuree (green)
Boardwalk railinga (brown)
Railinge (blue)
Railinge (brown)
Railinge (red)
Railinge (green)
Post of shade structuree (beige)
Steering wheele (brown)
Basketball postb (gray)

0.22
0.22
0.22
0.22
0.22
0.22
0.22
0.22
0.22
0.22
0.22
0.22
45.20

0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.20 (0.10–0.30)
0.08 (0.02–0.14)

62.2
55.0
43.9
51.1
51.7
68.3
48.9
61.1
55.0
–
46.7
48.3
47.8

–
–
–
38.3
–
–
40.0
–
–
40.0
–
39.4
–

–
–
–
38.9
–
–
–
–
–
42.2
37.8
–
–

0.22
0.42–0.51
0.25–0.51

0.20 (0.10–0.30)
0.18 (0.15–0.20)
0.18 (0.15–0.20)

50.0
63.9
48.3

40.0
–
–

–
–
–

Parent or child seating
Bench (metal, rubber coated)
Bencha
Miniature stoolsa (brown)
a
b
c
d
e

Molded plastic (UV stabilized high density polyethylene (HDPE)).
Powder-coated steel.
Rubber pad, slightly moist.
Light colored, textured, slip-resistant concrete.
Polyurethane enamel-coated steel.

types (Sheridan, 2002, 2015) with clear skies. There was little
difference in the Ta between the two days, with means (ranges)
of 32.6 ◦ C (25.8–40.2 ◦ C) and 33.6 ◦ C (26.4–41.6 ◦ C), respectively.
Hence, the maximum 2 m Ta was 40.2 ◦ C on July 13, 2011 and
41.6 ◦ C on September 15, 2014. This ∼1 ◦ C average difference is
insignificant, and minor given that it is the rough estimate of
error assumed in the mean temperature delta (T̄s–a ) caused
by meteorology on the given day. The average daily incoming
solar radiation was 502 W m−2 on July 13, 2011 and 519 W m−2
on Sept 14, 2014, with maximum radiation intensities reaching 829 W m−2 and 874 W m−2 at 1230 h LST for both days,
respectively.
3.2. Surface temperature variation
This study quantifies the Ts and Ts−a of two neighborhood
playgrounds at three spatial scales of measurement: 1 km neighborhood scale, 6.8 m microscale, and 1 cm touch-scale. We present
results at the microscale, followed by a touch-scale analysis, and
finally an integration of the datasets that address the distributions
of Ts at each scale incorporating the delta and scale offset temperatures.

3.2.1. Remote sensing data: 6.8 m scale
The TIR-based Ts product from the MASTER airborne data allows
for Ts estimation of objects at a resolution of 6.8 m or greater, such
as roads, buildings, houses, backyards, street medians, swimming
pools, and basketball courts. A total of 2,731,883 pixels representing Ts were obtained from the MASTER airborne data for the Power
Ranch neighborhood between 1100 h and 1330 h LST on July 13,
2011. The average 6.8 m pixel Ts within the entirety of the neighborhood was 48.8 ◦ C (SD = 6.89 ◦ C), ranging from 15.8 ◦ C (representing
water) to 71.4 ◦ C (dark asphalt and roofs). Data from both playgrounds were combined to form a subset of the full neighborhood
data. The playground 6.8 m Ts was 45.3 ◦ C (SD = 5.89 ◦ C), ranging
from 32.8 ◦ C (moist pool deck) to 60.7 ◦ C (sand, asphalt).
The distributions of these temperatures (full neighborhood and
a subset of the two playgrounds combined) are displayed in Fig. 4.
The probability distribution of the full neighborhood is skewed
to the right (skew = −0.45) signifying the bulk of the temperatures greater than the median, while the playground distribution is
skewed to the left (skew = +0.52). The most common 6.8 m Ts (the
mode) is 54.9 ◦ C in the neighborhood (which is above the neighborhood mean), and 44.3 ◦ C in the playgrounds (which is below the
playground mean). This 6.8 m resolution misses human scale and

36

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

touch-scale details that are important for thermal health, such as
sun versus shade, that were observed in the 1 cm Ts data collected at
touch-scales with the IR thermometers. This 6.8 m MASTER product
is also not able to ‘see’ surface temperatures under shade structures or trees and misses the shade effects entirely, yielding tree
canopy and roof top temperatures rather than the actual surface
temperatures that are experienced by neighborhood residents.
3.2.2. In situ: 1 cm touch scale
The range of the 1 cm touch-scale Ts is compared across objects
grouped into three categories of use: ‘sitting’, ‘walking’, and ‘holding’, as listed in Table 2. Wide ranges in the Ts are present between
various playground features and ground cover in both open sun
and shade (natural and artificial). The maximum 1 cm touch-scale
Ts was 87.2 ◦ C, measured on the soft rubber ground (black/green in
color) installed in place of a natural ground surface. In the shade,
this rubber surface’s Ts decreases to 42.2 ◦ C (tree shade) and 46.7 ◦ C
(shade sail), representing a Ts of 45.0 ◦ C and 40.5 ◦ C, respectively.
Playground walking surfaces bore an average Ts of 56.0 ◦ C in the
sun, 36.9 ◦ C in the shade of trees, and 39.3 ◦ C under a shade sail.
This yields a mean Ts of −19.1 ◦ C in tree shade, and −16.7 ◦ C for
artificial shade by the shade sail. Five of the ground surfaces measured with hand IRT (rubber surface, sand, steps, walkway, and the
basketball courts) surpass their material specific threshold for burn
based on a 5 s contact time (Table 1); however, natural or artificial
shade brings these five surfaces to safe values for touch.
The 1 cm scale Ts of playground equipment sat on by children
range from 46.7 ◦ C to 71.7 ◦ C in the sun. The plastic bench intended
for seating (Ts = 63.9 ◦ C) surpasses the 1 min 60 ◦ C threshold for
plastic, while slide temperatures also presented high Ts values of
71.7 ◦ C and 63.9 ◦ C for the green and beige slides, respectively (e.g.,
see Fig. 1). However, under shade trees the Ts of these sitting surfaces decrease by 18.5 ◦ C on average, and thus are brought into a
safe range for sitting. Lastly, for objects that are held by children, the
Ts ranges from 43.9 ◦ C (spring rider handle) to 68.3 ◦ C (boardwalk
railing). The majority of these “holding” objects were present in
full sun, which resulted in few sun versus shade comparisons of Ts
on the given holding objects; however, playground structure posts
were commonly found in both sun and shade, with Ts decreasing
by an average of 11.3 ◦ C in the shade.
3.2.3. Surface-to-air temperature variations by scale
It is evidenced above that shade has the ability to substantially
reduce the Ts of both artificial and natural surfaces by intercepting shortwave radiation. The SVF (inverse of the species shading
coefficient) is used here as a general interpretive framework for all
types of shade, including shade sails and artificial shade. In addition, in order to standardize the Ts for comparative purposes, the
Ts–a is employed and further interpreted using the SVF (Fig. 5).
Surfaces that are adequately shaded (low SVF) generally demonstrate no difference from Ta during early afternoon, while those
with high SVFs have a high proportion of the surface exposed, thus
an increased Ts is found. This is a new type of multiscale model for
Ts–a as a function of surface type and SVF, and in this implementation the model can be used to relate 6.8 m remotely sensed Ts to
touch-scale and (partially) shaded Ts in early afternoon. The same
principles apply to other times of day, seasons, and locations, but
the model’s coefficients need to be estimated separately for those
applications.
Within Fig. 5, we further see that the Ts–a values increase with
SVF, with a combined moderate Pearson’s correlation of r = 0.64.
Linear regressions are presented to show the positive relationships
of Ts–a over each surface type with SVF; however these are not statistically significant due to low sample numbers, and are presented
for visual purposes only. The T̄s–a values also vary by surface type
in both sun and shade (Table 3), with natural grass demonstrating

Table 3
Mean differences between surface and air temperatures (T̄s–a ) in open sun or tree
shade cover based on surface type within playgrounds.
T̄s−a (◦ C) 1 cm scale
Shade

Sun

Asphalt
Concrete
Grass
Rock
Sand

0.1
−0.7
−5.5
−1.3
−1.6

22.3
14.7
3.5
16.7
23.0

Average

−1.81

16.04

Table 4
Offset values representing the difference in the mean surface-to-air temperature
deltas (T̄s–a ) in the playground of (A) the 1 cm touch-scale deltas versus 6.8 m sand
deltas, and (B) the 1 cm touch scale delta versus the 1 km delta. Offsets represent
potential Ts errors when using the 6.8 m and 1 km scales, respectively, for playground
equipment analysis.
Burn threshold (◦ C)

(A) Playground offset of
1 cm to 6.8 m scale (◦ C)

(B) Playground offset of
1 cm to 1 km scale (◦ C)

Sitting – child
Walking
Holding
Sitting – adult and child

13.6
9.9
8.0
8.8

10.1
6.4
4.5
5.3

Average

10.1

6.6

the smallest delta when shaded (T̄s–a = −5.5 ◦ C), thus resulting
in Ts much lower than Ta . This T̄s–a is 3.7 ◦ C cooler than the average delta in the shade for all surfaces combined (T̄s–a = −1.8 ◦ C).
It is important to note here that irrigation practices and practitioners are consistent between the study periods. The Power Ranch
Community Association’s landscapers use a computer-controlled
irrigation system that waters the grass every night, and trees every
several days, during the summer season. Root zone soil moisture is
kept above deficit levels, and by early afternoon any residual surface moisture has evaporated. This should result in nearly constant
evapotranspiration rates and Ts during afternoon hours between
the two study periods.
Shaded asphalt demonstrates the least deviation from Ta
(T̄s–a = 0.1 ◦ C). In the open sun, the asphalt and sand result in
the greatest deviation from Ta (22.3 and 23.0 ◦ C, respectively),
much higher than the average for all surfaces in the sun combined
(T̄s–a = 16.0 ◦ C). These stark contrasts in surface-to-air temperature differences based on surface type and SVF highlight the
importance of surface type when modeling the effect of sun and
shade on Ts. We also find that the type of tree providing shade
does not result in significantly different surface-to-air temperature deltas, demonstrating that the SVF has little variation between
the tree specifies in the neighborhoods during the summertime,
and that each tree provided enough shade to provide a large Ts
reduction.
We further assess the scale offsets (or errors) associated with
measurements made at differing scales by examining the difference
in the average T̄s–a at each scale and by condition and/or surface
type, which are depicted by the horizontal solid lines in Fig. 5. The
scale offsets of the 1 cm as compared to either 6.8 m or 1 km scale
observations are calculated by taking the difference of the lines representing the T̄s–a values (select offsets are presented in Table 4).
When considering the full playground equipment in the sun, the
mean T̄s–a sun is 19.6 ◦ C (horizontal orange line) at the touchscale, while the 1 km sun value T̄s–a mean is 8.7 ◦ C (horizontal
magenta line), giving an offset between the two of 10.8 ◦ C. For all
sun exposed surfaces, T̄s–a of 16.0 ◦ C is found, yet if we were to
consider only the overall 6.8 m playground T̄s–a (8.7 ◦ C) or neighborhood T̄s–a (12.2 ◦ C), we would underestimate the Ts found at

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

37

Fig. 5. Variation in mean surface-to-air temperature deltas (T̄s–a ) at the 1 cm touch-scale due to change in sky view factor (SVF) for each surface type (asphalt, concrete,
grass, rock, sand). Sloped regression lines represent linear relationship of Ts–a with SVF for each surface type, with regression equations shown. Solid horizontal lines
represent the T̄s–a at various scales and surfaces: 1 km (neighborhood), PG6.8m (playground T̄s at 6.8 m scale), Sand6.8m (T̄s of sand at 6.8 m scale), Sun1cm (T̄s in sun at 1 cm
scale), Shade1cm (T̄s in shade at 1 cm scale), Eq Sun1cm (T̄s of PG equipment in sun at 1 cm scale). Scale offsets (presented in Table 4) are calculated from differences in the
horizontal lines compared to T̄s–a at the 1 cm scale.

the touch-scale measurements within the playgrounds. Therefore,
for all playground surfaces, there is an average offset of −3.8 ◦ C if
using the 6.8 m scale, and −7.3 ◦ C using the 1 km scale, both representing a scale measurement error of underestimations in actual Ts
measured at the 1 cm scale. These underestimations increase when
examining playground equipment (Table 4), with offsets of −6.6 ◦ C
at the 6.8 m scale, and −10.1 ◦ C at the 1 km scale as compared to
the 1 cm scale. Hence, the coarser scales underestimate high Ts and
Ts–a found in the playgrounds.
Comparisons of the distributions of the touch-scale Ts with the
overall neighborhood distribution of 6.8 m Ts (smoothed linear histogram as in Fig. 4b) are presented in Fig. 6. The handheld-IRT
touch-scale Ts data, including playground equipment temperatures
(Table 2), are presented as sun exposed (plot A), and in shade (plot
B) for nine surface types (both ground and equipment surfaces).
The sun exposed surfaces display a Ts of 54.8 ◦ C (vertical solid gray
line, plot A), while the mean in the shade for all surfaces is 38.4 ◦ C
(vertical solid black line, plot C). This 54.8 ◦ C touch-scale sun value
is 9.5 ◦ C above the remotely sensed playground Ts of 45.3 ◦ C (dashed
turquoise line, plot C) at the 6.8 m scale, and 6.0 ◦ C above the T̄s of
the full neighborhood (48.8 ◦ C; vertical solid green line). Furthermore, select touch-scale temperature values recorded in the sun
(rubber and some plastic surfaces) are outside of the range of the
distribution of the 6.8 m MASTER data, as seen by comparing Ts
data points in panel A to the distribution curve in panel C in Fig. 6.

This again displays the inability of coarse resolution observations
to discern touch-scale Ts .
These scale discrepancies clearly demonstrate that the hottest
touch-scale temperatures (exceeding the material-specific thresholds; Table 1) cannot be resolved with the 6.8 m MASTER data.
For example, when examining temperature pixels in the circular
South playground, the remote sensing platform detects predominantly the average temperature of the sand cover (∼55–60 ◦ C)
rather than the hotter equipment within these pixels, e.g., slides
(63.9 and 71.7 ◦ C), and railings (62.2–68.0 ◦ C). Hence, remotely
sensed data contains information about the mean and baseline Ts
to judge the microscale distribution of Ts in the neighborhood, yet
can have valuable use at a finer scale if an offset model is applied
to estimate the touch-scale Ts using airborne remotely sensed
microscale Ts . However, material-specific (e.g., for various types of
playground equipment) and shade-specific offsets (e.g., by SVF or
similar metric) are needed to accurately model these touch-scale
variations.
4. Discussion
4.1. Playground design in hot climates
Contrasting surface temperatures are found between MASTER overflight observations (6.8 m resolution) and in situ infrared

38

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

Fig. 6. Observed mean surface temperatures (T̄s ) at the touch-scale in sun (A) and shade (B) obtained from in situ Ts measurements in the playgrounds, with solid vertical
lines representing the 1 cm touch-scale T̄s in sun and shade for the respective plot. Plot C presents the full neighborhood scale Ts smoothed distribution from Fig. 4b at the
1 km neighborhood scale from MASTER 6.8 m grid temperatures. Vertical lines in plot C represent T̄s for the 1 km neighborhood scale (T̄s of the 6.8 m grid temperatures), and
the T̄s at the 6.8 m scale for water, grass, playgrounds, roof, and sand/asphalt.

thermometer data (‘touch-scale’ of centimeters), where the hottest
touch points of Ts found in the playgrounds are outside the
distribution of the MASTER 6.8 m T̄s observations and the 1 km
neighborhood mean. The differences highlight the essential discrepancy that exists when working with urban T̄s observations
at differing scales, especially when influential material types only
exist at a much finer scale on the order of centimeters. This study
identifies the sub-meter “touch” scale of approximately 1 cm as
the spatial scale at which surface temperatures pose a risk to children on playgrounds, and more generally of people engaged in
any kind of outdoor activity in the sun. Hence, the touch-scale
is the spatial scale at which data must be collected and policies
of landscape design must be executed to mitigate high Ts . This
is especially important in high-contact environments like playgrounds. A method to address the scale mismatch between airborne
remote sensing data and touch-scale in situ data does not currently exist. Research in this area is at the forefront of temperature
sensing technologies at fine scales, and is needed to better understand cross-scalar relationships and provide modeling capabilities
to link the various scales of Ts and temperature scale offsets in conjunction with biophysical and built environmental features. The
offset temperatures at each scale allow for comparisons between
the average of the probability distribution of Ts at various spatial
scales and/or shading conditions, for both upscaling and downscaling applications in remote sensing and climate modeling. These
offset temperatures can be empirically observed for each time of
day, season, material type, shading factor, and location, as the basis
of improved modeling and remote sensing techniques.
Urban climate research completed in the Phoenix metropolitan area demonstrates that air and surface temperatures differ
widely across area neighborhoods, with the greatest differences
during summertime heat waves (e.g., Chow et al., 2012; Harlan,
Brazel, Prashad, Stefanov, & Larsen, 2006; Hartz, Prashad, Hedquist,
Golden, & Brazel, 2006; Jenerette, Harlan, Stefanov, & Martin, 2011;
Middel et al., 2015). However, we demonstrate that within urban
neighborhoods and small playgrounds, these surface temperatures
vary significantly—and with implications for human thermal comfort and safety—at spatial scales as fine as 1 cm. Although satellite
remote sensing data are more applicable for mesoscale urban climate modeling than mesoscale climate models with ≥ 1 km scale
resolutions (e.g., Masson et al., 2003; Mishra et al., 2015), these

remote sensing data are critically limited for human- and touchscale design, missing crucial information related to touch-scale
extreme Ts exposures, mean radiant temperatures, and effects of
shade. These limitations are impactful especially for children, but
also other people exposed to the outdoor environment during partially shaded daytime conditions, including outdoor workers and
those engaged in sports or other recreational activities. However,
we have demonstrated that coarser scale satellite remote sensing data does provide valuable information serving as a basis for
estimating the fine-scale distribution around the average surface
temperature, T̄s , when accompanied by an offset model to estimate
the touch-scale Ts .
This scale mismatch is a consequential issue because the scale
of a remotely sensed observation may result in misleading conclusions at the most important scales for urban design if inter-scalar
bias exists. Here, the use of 6.8 m scale Ts is biased in such a fashion,
and would dramatically underestimate the touch-scale temperature if used for urban design and decision-making related to
individual-scale temperature exposure. This methodological issue
results in a spatial incongruence problem due to applying areabased attributes of size, shape, and spatial means (e.g., remote
sensing-based) that are insufficient for point-based (human- or
touch-scale) temperature exposure (Kwan, 2012). This represents a
Modifiable Areal Unit Problem (MAUP) (Houston, 2014; Openshaw
& Openshaw, 1984), where spatial uncertainty is influenced by
the context of measurement, and is a critical concern when using
research to link science and policy. The application of data from
the wrong scale results in a further issue of temporal uncertainty
related to the duration of time spent in certain condition, which
may not be accurate given the location or areal size of a temperature
observation (e.g., Kuras et al., 2015). Such spatiotemporal uncertainties may result in unsuitable decisions concerning park design
and materials for use under warm and sunny conditions, or may
complicate future work with higher-resolution urban climate models that require accurate microscale information to assess adaption
and UHI mitigation strategies.
The cooling effect of vegetation is documented in Phoenix,
where small (<20 ha), highly vegetated neighborhood parks create ‘cool islands’ when situated in the midst of densely urbanized
areas (Chow et al., 2011; Declet-Barreto, Brazel, Martin, Chow, &
Harlan, 2013; Jenerette et al., 2007; Middel, Häb, Brazel, Martin,

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

& Guhathakurta, 2014). Vegetation is also found more effective at
cooling hotter neighborhoods (Jenerette et al., 2015). However, this
cooling effect as studied costs a substantial amount of water to
achieve—an expensive proposition in a desert city. Further bioclimatic design in hot and sunny climates must employ shade
as a means of effectively reducing the mean radiant temperature
imposed on a human at the hottest time of the day, and thus overall heat stress (Ali-Toudert & Mayer, 2007; Matzarakis & Endler,
2010; Shashua-Bar, Pearlmutter, & Erell, 2011). We show that shade
generates a Ts that is similar or lower than the reference temperature, Ta , reversing the general trend where Ts > Ta . Although the
Tmrt is reduced by the shade sail—which effectively reduces experienced temperature—the comparisons to Ta reductions within
related studies in Phoenix and other cities from vegetation is only
valid when examining results under the shade of trees, as there is
no evaporative cooling effect from vegetation under the shade sail.
However, this study concurs with studies finding Ts reduction from
many types of shade (e.g., Antoniadis, Katsoulas, Papanastasiou,
Christidou, & Kittas, 2015; Ketterer & Matzarakis, 2014b; Lin &
Lin, 2010), where all shade, whether tree or artificial (e.g., mesh
or shade sails), is found effective in reducing the Ts in the hot
and arid climate to safe levels. This would in turn reduce thermal discomfort from sensible heat fluxes as found in related shade
studies (e.g., Antoniadis et al., 2015; Shashua-Bar et al., 2011).
Further microclimate factors influencing heat stress and thermal
discomfort—in addition to exchanges of solar radiation and sensible heat (via convection) discussed here—include evaporation and
longwave emittance (Kenny, Warland, Brown, & Gillespie, 2009;
Vanos, Warland, Kenny, & Gillespie, 2010), with metabolic activity
increasing core and skin temperatures due to increased heat generation (Vanos, Warland, Gillespie, & Kenny, 2012; Yao, Lian, Liu, &
Shen, 2007).
A study examining the energy budgets and/or the air cooling
potential of the differing types of shade can provide valuable information within a hot, arid city that can address the co-benefits of
water use for vegetation and evaporative cooling versus the cost
for water use at different times of the year. The effect of shade on
Ts is comparable to the effect of moistening a surface with water,
and shade can be utilized over any surface or urban space, not just
vegetated surfaces and specially designed water features. Artificial
shade is a very effective and flexible means of reducing Ts and Tmrt ,
and of improving thermal comfort and thermal safety in spaces
used by people during a specific time of day, without the need for
water use. This shade should employ sound design principles to
target the locations, times of day, and times of year when people
will be using a specific space while it is too hot; the optimal type
of shade might be different for a bus stop as compared with a playground. In Phoenix and similar cities, shade should be provided for
all playgrounds during the warmer seasons to avoid unsafe touchscale Ts . As an added benefit for air conditioned buildings, shade on
exterior building walls will reduce wall Ts , and shade on windows
will reduce shortwave radiative loading, both reducing energy costs
of air conditioning. Our findings and data also speak directly to such
building energy considerations. In a very hot climate like Phoenix,
carefully designed shade has the helpful characteristic of lowering thermal loads and Ts during the daytime year-round, where
daytime reductions in Ts may be more effective in reducing health
risks than nighttime (Jenerette et al., 2015). In a colder climate, it
might be important to avoid shading the microenvironment during
the day during winter. Our findings and design recommendations
should be interpreted accordingly, and further research conducted
in those other urban geographies and seasonalities.
The state of Arizona leads the U.S. in both deaths from heat
exposure (Centers for Disease Control and Prevention, 2006) and
extreme heat events representative of severe outdoor thermal
risks to urban dwellers (Grossman-Clarke, Zehnder, Loridan, &

39

Grimmond, 2010). The Phoenix metropolitan area is among the
hottest urban regions in the world, with heat amplified by one of the
strongest UHIs worldwide (Chow et al., 2012). The Town of Gilbert
and the neighborhood of Power Ranch possess an upper socioeconomic demographic with modern parks such as the study site
playgrounds that are subject to advanced planning and safety standards. Therefore this neighborhood is typical of the global extreme
both in terms of a hot urban climate and of recent best practices
of neighborhood adaptation to urban climate, including relatively
safe playground materials and the use of targeted shade. The findings of this study are even more important in communities that do
not yet follow modern playground safety standards. Results from
this extreme circumstance will therefore be increasingly relevant
globally and in the future as outdoor urban environments warm
and planning practices modernize.

4.2. Health and safety standards in playgrounds related to
temperature
Urban-dwelling children are routinely identified as a vulnerable population in environmental health review literature related
to heat-health and climate change (e.g., Ebi & Paulson, 2010;
English et al., 2009; Perera, 2008). The current findings and ensuing discussion provide important avenues for mitigating heat and
burn exposures to children and the associated negative health
consequences in playgrounds, as well as mechanisms for adapting the urban environment to manage increasing temperatures
and population growth. Although many days in Phoenix now
and in the future exhibit extreme surface and air temperatures,
shade cover can result in a wider temporal window within which
a child can safely play without the risk of being burned by a
surface.
The CPSC works with ASTM International (ASTM) (formerly
American Society for Testing and Materials) to ensure playground
safety via equipment standards. The current standard in use is
the ASTM F1487-Standard Consumer Safety Performance Specification for Playground Equipment for Public Use (ASTM F1487, 2011).
Although its main purpose is to reduce life-threatening and debilitating injuries, many of the standards are employed voluntarily
rather than mandated, which can result in vague and inadequate
procedures for playground safety. In general, few if any mandatory
standards exist for surface temperature modification or reduction,
although guidelines are available. The full outline and processes
for creating mandatory rules and guidelines are outlined in Earls
(2011) and Ford et al. (2011). Ford et al. (2011) reported results from
the In-Depth-Injury Reports (IDIs) issued by the CSPC1 (O’Brien,
2009) revealing 14 severe burns in children occurring on various playground surfaces (2001–2008), with nine being 2nd or 3rd
degree burns.
Materials with high heat capacities and/or thermal conductivities (e.g., black rubber, concrete, asphalt, artificial turf, steel,
aluminum) transfer heat quickly to the skin, and thus have a higher
potential to cause extensive burns. Results here show that in direct
sunlight, many of these materials can reach temperatures that are
above the thresholds for burn injury to a child (CPSC, 2010; Ford
et al., 2011; ISO 13732, 2010). O’Brien (2009) reported playground
slides to be the second most common equipment for burns (after
walking surfaces), with slides contributing to 29% of cases report by
the IDIs from 2001 to 2008. Although safer than metal, the plastic
slides in the study parks reached T̄s of 67.8 ◦ C in direct sunlight.

1
IDIs are part of the National Electronic Injury Surveillance System (NEISS) that
is run by CPSC to report injuries in the US (http://www.cpsc.gov/en/Research–
Statistics/NEISS-Injury-Data/).

40

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

Walking and standing surfaces made of rubber, concrete,
asphalt, and metal exposed to the sun were also found to reach temperatures approaching or above burn threshold temperatures (see
Tables 1 and 2), thus increasing the likelihood of burns to the feet
which can occur when children are playing and easily remove/lose
footwear and run or fall onto a scalding surface (e.g., O’Brien, 2009).
Based on CPSC reported data from 2001 to 2011, Ford et al. (2011)
reported 21 of 38 burn cases occurring on walking surfaces, including rubber mats, concrete, and asphalt. The application of various
shock-absorbing ground materials (sand, wood chips, pea gravel,
rubber) in playgrounds is based on reducing the risk of injury due
to falling (CPSC, 2010), yet these surfaces commonly reach the highest Ts in the sun. However, if these surfaces are already in place or
have a mandatory use, then our results show that providing shade
cover in any form can reduce Ts to a safe level. Shade on “walking”
type materials that cover large areas occupied by people will also
mitigate Tmrt in that space, providing a substantial thermal comfort
benefit that extends beyond reduction in the risk of burns.
Although greater attention has been given to other causes of
serious playground injuries in equipment design (Branson et al.,
2012; Mack, Hudson, & Thompson, 1997; Spiegal, Gill, Harbottle, &
Ball, 2014), the likely underreporting of heat-related injuries in children (e.g., Yeo, 2004) and lack of mandatory rules warrant greater
attention to the issue. According to the ASTM, further analyses similar to ASTM C1057 (2012) is required using a thermesthesiometer,
an instrument specific for burn hazard measurement (Marzetta,
1974; Ungar & Stroud, 2010), or employing a mathematical model.
This paper’s methods are the beginning of such a mathematical
model. The urban design community should carefully consider
whether evidence-based touch-scale temperature and shade standards should be adopted for playgrounds and other spaces occupied
by people during hot daytime hours. Based on results of this study,
we recommend that the adoption of new standards be considered.
Playgrounds serve the purpose of eliciting physical activity,
while providing spaces that are safe, thermally comfortable, functional, accessible, and healthy. These services are of paramount
importance in maintaining child health and well-being (Ciucci et al.,
2013; Fjørtoft, 2004; Wolch et al., 2011). The solution to reducing playground heat illness and surface burns on extreme heat
days is not to prohibit play, but to encourage safe play, invoke
awareness, and/or limit play to the morning or evening hours.
Quality play settings result in increased activity and improved
health (e.g., lessen obesity and diabetes) (Barnes, 2011) and social
networking (Bessesen, 2008; Sallis, Floyd, Rodríguez, & Saelens,
2012; Wolch et al., 2011), increased cognitive and motor development (Boldemann et al., 2006; Brussoni, Olsen, Pike, & Sleet, 2012;
Ciucci et al., 2011, 2013; Wenner, 2009), and improved behavior
(Bessesen, 2008; Ciucci et al., 2013; Lagacé-Séguin & d’Entremont,
2005) and learning (Taylor & Kuo, 2009). Quantifying both the
T̄s–a over various surfaces and the scale offsets of these deltas can
produce accurate and vital information for reducing playground Ts
through bioclimatic design. Such evidence-based design is applicable for public health agencies or local community organizations and
warranted in order to achieve thermally safe playgrounds during
the warm season and/or in warm climates.
4.3. Future work
This is a foundational study supporting an increased effort
to create viable methods for applying remotely sensed data at
finer environmental scales in order to advance public health
and urban design practice (e.g., White-Newsome et al., 2013).
Promising future opportunities exist in this research domain
to generate more evidence-based research largely centered on
child-specific physiological responses and behavior in order to
optimally incorporate effective bioclimatic design in outdoor

playgrounds (e.g., mandating a certain amount of shade cover
conserved/implemented in schoolyards). Findings from this study
lay the groundwork for future research to explore the systematic
relationships between in situ touch-scale surface temperatures,
remotely sensed microscale Ts , and those represented by urban
climate models. Future work is needed to develop a robust empirical mathematical model of scale offsets in T̄s–a to link remotely
sensed data at a given scale (e.g., 6.8 m) with 1 cm touch scale surface temperature data based on material, color, and biophysical
data (i.e., solar angle, transmissivity, microclimate, shade type, etc.),
and to extend this database and modeling work to other times of
day, seasons, locations, and climates. This will provide vital information to the ASTM, CPSC, and urban landscape architects to make
decisions at the correct scale.
4.4. Conclusions
We presented surface temperature data at multiple scales
within two playgrounds during hot, dry weather conditions in
Gilbert, AZ. High surface temperatures of playground equipment
and ground surfaces exposed to the sun were routinely found
from the in situ touch-scale (1 cm) measurements, which were not
detected by the airborne MASTER sensor (6.8 m resolution). Select
touch-scale temperatures approached or surpassed thresholds of
burn temperatures to children. Temperature deltas representing
surface-to-air differences were highest for playground equipment
and ground surface touch-scale measurements in sun (T̄s−a =
19.6 and 15.9 ◦ C, respectively). Scale offsets (errors due to the scale
of measurement) systematically underestimated the Ts and Ts–a
when using micro-scale and neighborhood scale Ts data rather than
touch-scale. This Ts underestimation error at the coarser scales can
cause spatial uncertainty in linking urban climate research to policy, yet may be valuable for use at a finer scale if an offset model is
applied to account for this error.
Shade from both natural (tree) and artificial (shade sail)
sources—no matter the type—reduced touch-scale playground Ts
to safe levels, with a T̄s–a of–24.3 ◦ C in the shade as compared
to the sun for all surfaces. The relatively new installation of the
large shade sail in the North Playground of Power Ranch demonstrates the effectiveness of shade at reducing the potential for
burns, providing empirical evidence that in hot climates or during warm seasons, the temporal window within which a child may
be burned on a surface is completely diminished through the use
of shade. This is a reduction not detected by the overhead-angle of
airborne remotely sensed data. Thus, coarser scale data or models
(such as urban climate models) cannot resolve Ts at the proper scale
for playground design. This discrepancy can lead to inaccuracies in
decision-making and applications related to health-outcome analysis, establishment of playground safety burn standards (reduce
risk of heat illness and injuries), urban park design guidelines and
policies, and urban cooling adaptation strategies and UHI mitigation. Finally, our results strongly motivate new touch-scale urban
temperature research, and the provision that bioclimatic principles
and effective green space should have a central position in landscape architecture, urban sustainability, and urban planning policy
for outdoor environments, particularly parks and children’s play
places.
Acknowledgements
The authors would like to acknowledge the Power Ranch Community Association’s (PRCA) support of the data collection and
David Hondula of ASU for aid in manuscript review and insightful
discussions. The collection of the MASTER data was jointly funded
by the NSF grants GEO-0816168, BCS-1026865. Work was also

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

funded by the NSF grant EF-1049251, and a grant to Arizona State
University by the Salt River Project. Any opinions, findings, and conclusions or recommendations expressed in this material are those
of the author(s) and do not necessarily reflect those of the National
Science Foundation. Benjamin Ruddell discloses membership on
the Board of Directors of the PRCA, which proposed construction of
shade structures at neighborhood playgrounds.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.landurbplan.
2015.10.007.
References
Adachi, S. A., Hara, M., Takahashi, H. G., Ma, X., Yoshikane, T., & Kimura, F. (2013).
Changes in urban climate due to future land-use changes based on population
changes in the Nagoya region. pp. 455. AGU fall meeting abstracts (Vol. 1)
Ali-Toudert, F., & Mayer, H. (2007). Effects of asymmetry, galleries, overhanging
facades and vegetation on thermal comfort in urban street canyons. Solar
Energy, 81(6), 742–754.
Antoniadis, D., Katsoulas, N., Papanastasiou, D., Christidou, V., & Kittas, C. (2015).
Evaluation of thermal perception in schoolyards under Mediterranean climate
conditions. International Journal of Biometeorology, 1–16.
ASTM 1055. (2014). Standard guide for heated system surface conditions that produce
contact burn injuries. West Conshohocken, PA. http://www.astm.org/
Standards/C1055.htm
ASTM C1057. (2012). Standard practice for determination of skin contact temperature
from heated surfaces using a mathematical model and thermesthesiometer. West
Conshohocken, PA. http://dx.doi.org/10.1520/C1057
ASTM F1487. (2011). Standard consumer safety performance specification for
playground equipment for public use. No. F1487-07ae1. West Conshohocken, PA.
http://dx.doi.org/10.1520/F1487-11
Balbus, J. M., & Malina, C. (2009). Identifying vulnerable subpopulations for climate
change health effects in the United States. Journal of Occupational and
Environmental Medicine, 51(1), 33–37.
Barnes, M. (2011). White house task force on childhood obesity report to the
president: One year progress report. Washington, DC. http://www.letsmove.gov/
sites/letsmove.gov/files/Obesity update report.pdf
Bessesen, D. H. (2008). Update on obesity. Journal of Clinical Endocrinology &
Metabolism, 93(6), 2027–2034.
Board on Atmospheric Sciences and Climate (BASC) Summer Community
Workshop. (2012). Urban meteorology: Forecasting, monitoring, and meeting
users’ needs (15th ed.). Washington, DC: National Research Council (U.S.), The
National Academies Press. http://www.nap.edu/openbook.php?record
id=13328
Boldemann, C., Blennow, M., Dal, H., Mårtensson, F., Raustorp, A., Yuen, K., et al.
(2006). Impact of preschool environment upon children’s physical activity and
sun exposure. Preventive Medicine, 42(4), 301–308.
Branson, L. J., Latter, J., Currie, G. R., Nettel-Aguirre, A., Embree, T., & Hagel, B. E.
(2012). The effect of surface and season on playground injury rates. Paediatrics
& Child Health, 17(9), 485.
Brown, R. D., Vanos, J. K., Kenny, N. A., & Lenzholzer, S. (2015). Designing urban
parks that ameliorate the effects of climate change. Landscape and Urban
Planning, http://dx.doi.org/10.1016/j.landurbplan.2015.02.006
Brussoni, M., Olsen, L. L., Pike, I., & Sleet, D. A. (2012). Risky play and children’s
safety: Balancing priorities for optimal child development. International Journal
of Environmental Research and Public Health, 9(9), 3134–3148.
Campbell, G. S., & Norman, J. M. (1998). An introduction to environmental biophysics
(2nd ed.). New York: Springer.
Centers for Disease Control and Prevention. (2006). Heat-related deaths—United
States, 1999–2003. Morbidity and Mortality Weekly Report, 55, 796–798.
Cheng, X., Wei, B., Chen, G., Li, J., & Song, C. (2014). Influence of park size and its
surrounding urban landscape patterns on the park cooling effect. Journal of
Urban Planning and Development, http://dx.doi.org/10.1061/(ASCE)UP.19435444.0000256
Chow, W. T. L., Brennan, D., & Brazel, A. J. (2012). Urban heat island research in
Phoenix, Arizona: Theoretical contributions and policy applications. Bulletin of
the American Meteorological Society, 93(4), 517–530.
Chow, W. T. L., Pope, R. L., Martin, C. A., & Brazel, A. J. (2011). Observing and
modeling the nocturnal park cool island of an arid city: Horizontal and vertical
impacts. Theoretical and Applied Climatology, 103(1–2), 197–211.
Ciucci, E., Calussi, P., Menesini, E., Mattei, A., Petralli, M., & Orlandini, S. (2011).
Weather daily variation in winter and its effect on behavior and affective states
in day-care children. International Journal of Biometeorology, 55(3), 327–337.
Ciucci, E., Calussi, P., Menesini, E., Mattei, A., Petralli, M., & Orlandini, S. (2013).
Seasonal variation, weather and behavior in day-care children: A multilevel
approach. International Journal of Biometeorology, 57(6), 845–856.
CPSC. (2010). Handbook for public playground safety. Bethesda, MD. http://www.
cpsc.gov/PageFiles/122149/325.pdf

41

Declet-Barreto, J., Brazel, A. J., Martin, C. A., Chow, W. T. L., & Harlan, S. L. (2013).
Creating the park cool island in an inner-city neighborhood: Heat mitigation
strategy for Phoenix, AZ. Urban Ecosystems, 16(3), 617–635.
Earls, A. (2011). The CPSC-ASTM collaboration. ASTM Standardization News, 39(1),
32–35.
Ebi, K. L., & Paulson, J. A. (2010). Climate change and child health in the United
States. Current Problems in Pediatric and Adolescent Health Care, 40(1), 2–18.
English, P. B., Sinclair, A. H., Ross, Z., Anderson, H., Boothe, V., Davis, C., et al. (2009).
Environmental health indicators of climate change for the United States:
Findings from the State Environmental Health Indicator Collaborative.
Environmental Health Perspectives, 117(11), 1673–1681.
Environmental Protection Agency. (2014). Cooling summertime temperatures..
http://www.epa.gov/heatislands/resources/pdf/HIRIbrochure.pdf
Erell, E., Pearlmutter, D., & Williamson, T. (2012). Urban microclimate: Designing the
spaces between buildings. Routledge.
Falk, B., & Dotan, R. (2008). Children’s thermoregulation during exercise in the heat
– A revisit. Applied Physiology, Nutrition, and Metabolism, 33(2), 420–427.
Fjørtoft, I. (2004). Landscape as playscape: The effects of natural environments on
children’s play and motor development. Children Youth and Environments,
14(2), 21–44.
Ford, G., Moriarty, A., Riches, D., & Walker, S. (2011). Playground equipment:
Classification & burn analysis. Bethesda, MD. http://www.wpi.edu/Pubs/Eproject/Available/E-project-122111-202154/unrestricted/WPI Final Report
IQP-CPSC.pdf
Grossman-Clarke, S., Zehnder, J. A., Loridan, T., & Grimmond, C. S. B. (2010).
Contribution of land use changes to near-surface air temperatures during
recent summer extreme heat events in the Phoenix metropolitan area. Journal
of Applied Meteorology and Climatology, 49(8), 1649–1664.
Harlan, S. L., Brazel, A. J., Prashad, L., Stefanov, W. L., & Larsen, L. (2006).
Neighborhood microclimates and vulnerability to heat stress. Social Science
and Medicine, 63(11), 2847–2863.
Harlan, S. L., Declet-Barreto, J. H., Stefanov, W. L., & Petitti, D. B. (2013).
Neighborhood effects on heat deaths: Social and environmental predictors of
vulnerability in Maricopa County, Arizona. Environmental Health Perspectives,
121(2), 197.
Harlan, S. L., & Ruddell, D. M. (2011). Climate change and health in cities: Impacts
of heat and air pollution and potential co-benefits from mitigation and
adaptation. Current Opinion in Environmental Sustainability, 3(3), 126–
134.
Hartz, D. A., Prashad, L., Hedquist, B. C., Golden, J., & Brazel, A. J. (2006). Linking
satellite images and hand-held infrared thermography to observed
neighborhood climate conditions. Remote Sensing of Environment, 104(2),
190–200.
Hondula, D. M., Davis, R. E., Leisten, M. J., Saha, M. V., Veazey, L. M., & Wegner, C. R.
(2012). Fine-scale spatial variability of heat-related mortality in Philadelphia
County, USA, from 1983–2008: A case-series analysis. Environmental Health,
11(1), 1–11.
Hook, S. J., Myers, J. J., Thome, K. J., Fitzgerald, M., & Kahle, A. B. (2001). The
MODIS/ASTER airborne simulator (MASTER)—A new instrument for earth
science studies. Remote Sensing of Environment, 76(1), 93–102.
Houston, D. (2014). Implications of the modifiable areal unit problem for assessing
built environment correlates of moderate and vigorous physical activity.
Applied Geography, 50, 40–47.
HUSD. (2015a). Higley United School District. Welcome Centennial Elementary.
http://www.husd.org/Page/4646
HUSD. (2015b). Higley United School District. Welcome, Power Ranch Elementary.
http://www.husd.org/Domain/1525
ISO 13732. (2010). ISO 13732-3: Ergonomics of the thermal environment. Methods for
the assessment of human responses to contact with surfaces. Hot surfaces. Geneva,
Switzerland: Vernier.
Jenerette, G. D., Harlan, S. L., Brazel, A., Jones, N., Larsen, L., & Stefanov, W. L. (2007).
Regional relationships between surface temperature, vegetation, and human
settlement in a rapidly urbanizing ecosystem. Landscape Ecology, 22(3),
353–365. http://dx.doi.org/10.1007/s10980-006-9032-z
Jenerette, G. D., Harlan, S. L., Buyantuev, A., Stefanov, W. L., Declet-Barreto, J.,
Ruddell, B. L., et al. (2015). Micro scale urban surface temperatures are related
to land cover features and residential heat-related health impacts in Phoenix,
AZ, USA. Landscape Ecology, http://dx.doi.org/10.1007/s10980-015-0284-3
Jenerette, G. D., Harlan, S. L., Stefanov, W. L., & Martin, C. A. (2011). Ecosystem
services and urban heat riskscape moderation: Water, green spaces, and social
inequality in Phoenix, USA. Ecological Applications, 21(7), 2637–2651.
Johansson, E., Thorsson, S., Emmanuel, R., & Krüger, E. (2014). Instruments and
methods in outdoor thermal comfort studies – The need for standardization.
Urban Climate, 10, 346–366.
Kántor, N., & Unger, J. (2011). The most problematic variable in the course of
human-biometeorological comfort assessment—The mean radiant
temperature. Central European Journal of Geosciences, 3(1), 90–100.
Karl, T. R., Melillo, J. M., & Peterson, T. C. (2009). Global climate change impacts in the
United States. Cambridge University Press.
Kenny, N. A., Warland, J. S., Brown, R. D., & Gillespie, T. G. (2009). Part A: Assessing
the performance of the COMFA outdoor thermal comfort model on subjects
performing physical activity. International Journal of Biometeorology, 53(5),
415–428.
Ketterer, C., & Matzarakis, A. (2014a). Human-biometeorological assessment of
heat stress reduction by replanning measures in Stuttgart, Germany. Landscape
and Urban Planning, 122, 78–88.

42

J.K. Vanos et al. / Landscape and Urban Planning 146 (2016) 29–42

Ketterer, C., & Matzarakis, A. (2014b). Human-biometeorological assessment of the
urban heat island in a city with complex topography – The case of Stuttgart,
Germany. Urban Climate, 10, 573–584.
Knez, I., Thorsson, S., Eliasson, I., & Lindberg, F. (2009). Psychological mechanisms
in outdoor place and weather assessment: Towards a conceptual model.
International Journal of Biometeorology, 53(1), 101–111.
Koppe, C., & Jendritzky, G. (2005). Inclusion of short-term adaptation to thermal
stresses in a heat load warning procedure. Meteorologische Zeitschrift, 14(2),
271–278.
Krayenhoff, E. S., & Voogt, J. A. (2007). A microscale three-dimensional urban
energy balance model for studying surface temperatures. Boundary-Layer
Meteorology, 123(3), 433–461.
Kuras, E. R., Hondula, D. M., & Brown-Saracino, J. (2015). Heterogeneity in
individually experienced temperatures (IETs) within an urban neighborhood:
Insights from a new approach to measuring heat exposure. International
Journal of Biometeorology, 1–10.
Kwan, M.-P. (2012). The uncertain geographic context problem. Annals of the
Association of American Geographers, 102(5), 958–968.
Lagacé-Séguin, D. G., & d’Entremont, M. L. (2005). Weathering the preschool
environment: Affect moderates the relations between meteorology and
preschool behaviors. Early Child Development and Care, 175(5), 379–394.
Lin, B.-S., & Lin, Y.-J. (2010). Cooling effect of shade trees with different
characteristics in a subtropical urban park. HortScience, 45(1), 83–86.
Luber, G., Knowlton, K., Balbus, J., Frumkin, H., Hayden, M., Hess, J., et al. (2014).
Human health. In G. W. Melillo, J. M. Richmond, & T. C. Yohe (Eds.), U.S. Global
Change Research Program (pp. 220–256). http://dx.doi.org/10.7930/J0PN93H5
Mack, M. G., Hudson, S., & Thompson, D. (1997). A descriptive analysis of children’s
playground injuries in the United States 1990–4. Injury Prevention, 3(2),
100–103.
Marzetta, L. A. (1974). Engineering and construction manual for an instrument to
make burn hazard measurements in consumer products. Washington, DC.
Masson, V., Champeaux, J.-L., Chauvin, F., Meriguet, C., & Lacaze, R. (2003). A global
database of land surface parameters at 1-km resolution in meteorological and
climate models. Journal of Climate, 16(9), 1261–1282.
Masson, V., Marchadier, C., Adolphe, L., Aguejdad, R., Avner, P., Bonhomme, M.,
et al. (2014). Adapting cities to climate change: A systemic modelling
approach. Urban Climate, 10, 407–429.
Matzarakis, A., & Endler, C. (2010). Climate change and thermal bioclimate in
cities: Impacts and options for adaptation in Freiburg, Germany. International
Journal of Biometeorology, 54(4), 479–483.
Matzarakis, A., Rutz, F., & Mayer, H. (2010). Modelling radiation fluxes in simple
and complex environments: Basics of the RayMan model. International Journal
of Biometeorology, 54(2), 131–139.
Mayer, H., & Höppe, P. (1987). Thermal comfort of man in different urban
environments. Theoretical and Applied Climatology, 38, 43–49.
McCarthy, M. P., & Sanderson, M. G. (2011). Urban heat islands: Sensitivity of urban
temperatures to climate change and heat release in four European Cities. Cities
and Climate Change, 175.
McGeehin, M. A., & Mirabelli, M. (2001). The potential impacts of climate
variability and change on temperature-related morbidity and mortality in the
United States. Environmental Health Perspectives, 109(Suppl 2), 185–189.
Meehl, G. A., & Tebaldi, C. (2004). More intense, more frequent, and longer lasting
heat waves in the 21st century. Science, 305(5686), 994–997.
Metje, N., Sterling, M., & Baker, C. J. (2008). Pedestrian comfort using clothing
values and body temperatures. Journal of Wind Engineering & Industrial
Aerodynamics, 96(4), 412–435.
Middel, A., Chhetri, N., & Quay, R. (2015). Urban forestry and cool roofs:
Assessment of heat mitigation strategies in Phoenix residential neighborhoods.
Urban Forestry & Urban Greening, 14(1), 178–186.
Middel, A., Häb, K., Brazel, A. J., Martin, C. A., & Guhathakurta, S. (2014). Impact of
urban form and design on mid-afternoon microclimate in Phoenix Local
Climate Zones. Landscape and Urban Planning, 122, 16–28.
Mishra, V., Ganguly, A. R., Nijssen, B., & Lettenmaier, D. P. (2015). Changes in
observed climate extremes in global urban areas. Environmental Research
Letters, 10(2), 24005.
Moogk-Soulis, C. (2010). Schoolyard heat islands: A case in Waterloo, Ontario. In
5th Canadian urban forest conference (pp. 24–27).
Moore, R., & Cosco, N. (2014). Growing up green: Naturalization as a health
promotion strategy in early childhood outdoor learning environments.
Children Youth and Environments, 24(2), 168–191.
National Oceanic and Atmospheric Administration. (2012). Comparative climatic
data for the United States through 2012. Asheville, NC. http://www1.ncdc.noaa.
gov/pub/data/ccd-data/CCD-2012.pdf#sthash.UpnVUfyR.dpuf
O’Brien, C. (2009). Injuries and investigated deaths associated with playground
equipment, 2001–2008. Bethesda, MD. http://www.cpsc.gov/library/foia/
foia10/os/playground.pdf
Oke, T. R. (1988). The urban energy balance. Progress in Physical Geography, 12, 471.
Oliveria, S. A., Saraiya, M., Geller, A. C., Heneghan, M. K., & Jorgensen, C. (2006). Sun
exposure and risk of melanoma. Archives of Disease in Childhood, 91(2),
131–138.
Openshaw, S., & Openshaw, S. (1984). The modifiable areal unit problem. In Geo
Abstracts University of East Anglia.

Parsons, K. C. (2003). Human Thermal Environments: The effects of hot, moderate and
cold environments on human health, comfort and performance (2nd ed.). New
York, NY: Taylor and Francis.
Paul, K. C., Pal, A. K., Ghosh, A. K., & Chakraborty, N. R. (2004). Thermal
measurements of coating films used for surface insulation and protection.
Surface Coatings International Part B: Coatings Transactions, 87(2), 137–141.
Perera, F. P. (2008). Children are likely to suffer most from our fossil fuel addiction.
Environmental Health Perspectives, 116(8), 987–990.
Petitti, D., Hondula, D. M., Yang, S., Harlan, S. L., & Chowell, G. (2015). Multiple
trigger points for quantifying heat-health impacts: New evidence from a hot
climate. Environmental Health Perspectives, http://dx.doi.org/10.1289/ehp.
1409119
Sallis, J. F., Floyd, M. F., Rodríguez, D. A., & Saelens, B. E. (2012). Role of built
environments in physical activity, obesity, and cardiovascular disease.
Circulation, 125(5), 729–737.
Shashua-Bar, L., Pearlmutter, D., & Erell, E. (2011). The influence of trees and grass
on outdoor thermal comfort in a hot-arid environment. International Journal of
Climatology, 31(10), 1498–1506.
Sheridan, S. C. (2002). The redevelopment of a weather-type classification scheme
for North America. International Journal of Climatology, 22(1), 51–68.
Sheridan, S. C. (2015). Synoptic scale classification.. http://sheridan.geog.kent.edu/
ssc.html
Spiegal, B., Gill, T. R., Harbottle, H., & Ball, D. J. (2014). Children’s play space and
safety management rethinking the role of play equipment standards. SAGE
Open, 4(1), 2158244014522075.
Stefanov, W. L., Prashad, L., Eisinger, C., Brazel, A., & Harlan, S. L. (2004).
Investigation of human modifications of landscape and climate in the Phoenix
Arizona Metropolitan area using MASTER data. Pan, 2, 2–8.
Strong, D., Tahir, A., & Verma, S. (2007). Not fun in the sun: Playground safety in a
heatwave. Emergency Medicine Journal, 24(9), 2. http://dx.doi.org/10.1136/emj.
2006.041657
Taylor, A. F., & Kuo, F. E. (2009). Children with attention deficits concentrate better
after walk in the park. Journal of Attention Disorders, 12(5), 402–409.
Tebaldi, C., Hayhoe, K., Arblaster, J. M., & Meehl, G. A. (2006). Going to the
extremes: An intercomparison of model-simulated historical and future
changes in extreme events. Climatic Change, 79, 185–211.
Thorsson, S., Honjo, T., Lindberg, F., Eliasson, I., & Lim, E.-M. (2007). Thermal
comfort and outdoor activity in Japanese urban public places. Environment and
Behavior, 39(5), 660–684.
U.S. Census Bureau. (2015). 2009–2013 5-year American Community Survey. U.S.
Bureau of the Census. http://www.census.gov/acs/www/
Ungar, E., & Stroud, K. (2010). A new approach to defining human touch
temperature standards. In 40th International conference on environmental
systems Barcelona, Spain.
Van Der Meer, F., & de Jong, S. M. (2001). . Imaging spectrometry: Basic principles
and prospective applications (Vol. 1) Springer Science & Business Media.
Vanos, J. (2015). Children’s health and vulnerability in outdoor microclimates: A
comprehensive review. Environment International, 76, 1–15.
Vanos, J. K., Warland, J. S., Gillespie, T. J., Slater, G. A., Brown, R. D., & Kenny, N. A.
(2012). Human energy budget modeling in urban parks in Toronto, ON and
applications to emergency heat stress preparedness. Journal of Applied
Meteorology & Climatology, 51(9), 1639–1653.
Vanos, J. K., Warland, J. S., Kenny, N. A., & Gillespie, T. J. (2010). Review of the
physiology of human thermal comfort while exercising in urban landscapes
and implications for bioclimatic design. International Journal of Biometeorology,
54(4), 319–334.
Vanos, J., Warland, J., Gillespie, T., & Kenny, N. (2012). Thermal comfort modelling
of body temperature and psychological variations of a human exercising in an
outdoor environment. International Journal of Biometeorology, 56(1), 21–32.
Walsh, J., Wuebbles, D., Hayhoe, K., Kossin, J., Kunkel, K., Stephens, G., et al. (2014).
Climate change impacts in the United States. In J. M. Melillo, T. C. Richmond, &
G. W. Yohe (Eds.), The third national climate assessment (pp. 19–67). US Global
Change Research Program.
Wenger. (2003). The regulation of body temperature. In G. Rhoades, & R. Tanner
(Eds.), Medical physiology (2nd ed., Vol. 1, pp. 527–550). Boston: Little Brown.
Wenner, M. (2009). The serious need for play. Scientific America. http://www.
scientificamerican.com/article/the-serious-need-for-play/
White-Newsome, J. L., Brines, S. J., Brown, D. G., Dvonch, J. T., Gronlund, C. J., Zhang,
K., et al. (2013). Validating satellite-derived land surface temperature with
in situ measurements: A public health perspective. Environmental Health
Perspectives, 121(8), 925.
Wolch, J., Jerrett, M., Reynolds, K., McConnell, R., Chang, R., Dahmann, N., et al.
(2011). Childhood obesity and proximity to urban parks and recreational
resources: A longitudinal cohort study. Health & Place, 17(1), 207–214.
Yaghoobian, N., Kleissl, J., & Krayenhoff, E. S. (2010). Modeling the thermal effects
of artificial turf on the urban environment. Journal of Applied Meteorology and
Climatology, 49(3), 332–345.
Yao, Y., Lian, Z., Liu, W., & Shen, Q. (2007). Experimental study on skin temperature
and thermal comfort of the human body in a recumbent posture under
uniform thermal environments. Indoor and Built Environment, 16(6), 505.
Yeo, T. P. (2004). Heat stroke: A comprehensive review. ACN Clinical Issues:
Advanced Practice in Acute and Critical Care, 2, 280–293.