International Journal of GEOMATE, Oct., 2016, Vol. 11, Issue 26, pp.

2541-2546
Geotec., Const. Mat. & Env., ISSN: 2186-2982(Print), 2186-2990(Online), Japan

THE COOLING EFFECT OF A MEDIUM SIZED PARK ON AN URBAN

ENVIRONMENT
Hayder Al-Gretawee12, Scott Rayburg1and Melissa Neave3
1
Faculty of Science, Engineering and Technology, Swinburne University of Technology,
Australia;2University of Karbala, Iraq;3School of Global, Urban and Social Studies, RMIT University,
Australia

ABSTRACT: The Urban Heat Island (UHI) effect can best be described as an increase in the temperature of
urban areas relative to their surroundings. This effect can exceed 5oC in places. This study investigates how
vegetation, in particular urban parklands, can be used to reduce the intensity of the UHI effect. To achieve this,
the study uses a ground based approach relying on high spatial and temporal resolution temperature
measurements using both a hand-held weather meter and a hand-held thermal laser-gun. The study focusses on
one medium sized park in Melbourne, Australia and samples air temperatures (at 5 cm and 1.5 m above the
ground) and land surface temperature profiles six times a day over one month starting within the park and
extending to approximately 1 km outside of the park. The study shows that the park has a significant cooling
effect for a distance of up to 860 m from its boundaries and that this is most significant in the early morning. The
study also shows that land surface temperatures are more sensitive to park cooling effects than are air
temperatures.

Keywords: UHI, Cooling Distance, Urban Parks, Air Temperature, Land Surface Temperature.

1. INTRODUCTION by tall buildings, and a comparatively high

concentration of greenhouses gases relative to
Global warming is of increasing concern around surrounding areas [8]-[9]-[10]. Temporal parameters
the world with global temperatures predicted to rise that effect UHI intensity include time of day and
by 2oC or more by 2100. These changes have been seasonality with UHI intensity being higher during
attributed to human activities, especially the release the night and in summer. Additionally, cloudy
of greenhouse gases into the atmosphere from the conditions can decrease UHI intensity [11]-[12]-
burning of fossil fuels [1]. A predicted outcome of [13]-[14] while clear skies and low wind velocities
this warming is that heat waves will become more tend to increase UHI intensity [15]-[16].
frequent and severe, particularly in the second half UHIs have many negative consequences.
of the 21st century. These heat waves, and the Increasing temperature discrepancies between urban
increased temperatures more generally, are (and will centers and their surroundings can result in serious
be) felt most keenly in cities. This is because cities social and environmental impacts. For example,
are already warmer than their surrounding UHIs lead to increases in energy use (particularly
environments due to the “Urban Heat Island” (UHI) cooling energy) that further increases urban
effect [2]-[3]. pollution through emissions and enhances
The UHI phenomenon can best be described as a weathering rates [17]. In addition, health issues have
difference in surface air temperature between urban been associated with an increase in temperature
areas and their surrounding non-urban landscapes [18]-[19]. These health issues range from relatively
[4]. The temperature difference that can be attributed minor impacts, such as heat rash, to potentially life
to the UHI effect varies from place to place threatening conditions such as heat stroke. Those
depending on a number of spatial (e.g., degree of with weakened immune systems and the young or
urbanization, amount of urban greenspace) and elderly are especially susceptible to heat related
temporal factors (e.g., seasonality, time of day), but illnesses [20].
is generally within the range of 0.4°C to 11°C To combat these problems, UHI mitigation
although it has been reported as high as 14°C in strategies have been extensively researched in recent
Moscow [5]-[6]. Many causes have been proposed years. One of the most promising ways to reduce
to explain this phenomenon, all of which are related UHI intensity is through the use of green areas
to urbanization [7] including: a high percentage of within or adjacent to urban centers [21]-[22]. These
low-albedo and impermeable surfaces, a reduction in can take the form of green roofs, street trees or urban
the cooling effect of shading as trees are replaced by parklands, with research in Taipei and Paris, for
buildings or roads, the urban canyon effect where example, showing that the temperature in urban
heat gets trapped and reradiated back to the ground parks is cooler than that of the surrounding areas,

2541
 International Journal of GEOMATE, Oct., 2016, Vol. 11, Issue 26, pp. 2541-2546

especially at night [23]. Similarly, research on green and it contains two major attractions (the Royal
roofs show that they have been successful at cooling Exhibition Building and the Melbourne Museum)
the air adjacent to concrete roofs in Manchester, UK and other facilities such as playgrounds and tennis
[24]. Generally, this research indicates that parks courts.
and green roofs both contribute strongly to the The study involved collecting air (at ground level
reduction of surface air temperatures and that urban and 1.5 m above the ground) and surface
areas that contain these can be locally cooler than temperature data six times a day (one sampling run
areas that lack these features [25]-[26]-[27]. every 4 hours starting at 8 am) over 24 hours. Air
Although vegetation has been shown to be highly temperature data were collected using a Kestrel 4000
effective at reducing the intensity of the urban heat series weather and environmental meter while land
island effect, there is no generally agreed upon ideal surface temperatures were recorded using a
vegetation type, cover or distribution that is most Milwaukee laser temperature gun. Prior to using the
effective at mitigating it [28]. Likewise, there remain laser temperature gun, the emissivity of each surface
many unanswered questions about how best to type sampled in the study was determined (using the
achieve significant reductions in UHI intensity. thermocouple that comes with the Milwaukee
These include identifying an effective cooling area temperature gun) and this value was set in the device
(the distance over which cooling benefits are to ensure that actual surface temperatures were
observed) for parks and green roofs. recorded. During each 4 hourly sampling run, at
To help understand urban heat islands, least ten points were sampled at increasing distances
contemporary researchers often utilize remote (from within the park to 860 m away) from the
sensing data, as they provide a cost effective and park’s centre. To ensure that representative
time saving method for the analysis of land surface temperatures were recorded, the land surface
temperatures (LST) over wide areas and at different temperature (LST) data included the average of three
times of the year or day. Typically, such to four readings over a period of 2-3 minutes.
investigations involve the use of the thermal bands Likewise, the air temperature measurements were
of the Landsat series of satellites or the Moderate- averaged over a period of 4-5 minutes to ensure that
resolution Imaging Spectroradiometer (MODIS) the temperature had stabilized before the values
[29]. These approaches have the benefit of assessing were recorded.
temperatures over a wide area. However, remote
sensing approaches are limited by the resolution of 2.1 Sample Date Characteristics
the sensors and the limited times at which such
images are available (once per day for MODIS to Data were collected on 27-28/04/2015 and 30-
once per fortnight for Landsat). Moreover, the 31/05/2015, starting from 8 am and ending at 4 am
images are sensitive to cloud cover, which conceals the next day. The weather for the two sampling
the temperature signature. In contrast, ground based events was: (1) cloudy with a few showers at
approaches, although limited in spatial extent midnight on 27-28/04/2015; and (2) partially cloudy
compared to remote sensing, allow researchers to for the whole day on 30-31/05/2015. The winds
sample many times per day and at a very fine spatial varied between 1.1-16.1 km/h and 1.3-12.5 km/h for
resolution. the two days respectively (Table 1).
The aim of this study is to investigate the cooling
effect of a medium sized park in Melbourne, 3. RESULTS
Australia on its surrounding urban area. The study
adopts a ground based sampling approach and seeks 3.1 Air Temperature
to determine the effective cooling distance from the
boundary of the park. Such research will enhance Ambient air temperature was measured at two
our understanding of the cooling benefits of urban heights; one very close to the ground (5 cm above
parks and provide specific information about the size the ground) and the other at 1.5 m above the ground.
and spacing of such parks needed to achieve
widespread urban cooling benefits. 3.1.1 5 cm above the ground

2. STUDY AREA AND METHODOLOGY The cooling effect of the park at a distance of 5
cm above the ground extended out as far as 860 m
This research was undertaken in a medium-sized (for one measurement at 8 am on 30/05/2015) and
urban park (Carlton Gardens) near the central exhibited a maximum cooling impact of 6oC (at 8
business district (CBD) of Melbourne in Victoria, am on 27/04/2015) (Figs. 2(A) and 3(A)). Cooling
Australia (Fig. 1). Carlton Gardens is a World was most evident at this elevation at 8 am and 12 pm
Heritage Site located on the north-eastern edge of (for both sampling dates) with minimal or no
the CBD in the suburb of Carlton (37°48′22″S cooling exhibited at other times (ranging from no
144°58′13″E). The area of the park is 26 hectares effect to approximately 1.5oC) with the exception of

2542
 International Journal of GEOMATE, Oct., 2016, Vol. 11, Issue 26, pp. 2541-2546

the 4 am measurement on 27/04/2015 which For all sampling times and dates, the cooling effect
displayed 3.7oC of cooling. experienced at 1.5 m was lower than or equal to that
at 5 cm above the ground (Figs. 2(C) and 3(C)).
3.1.2 1.5 m above the ground
3.2 Land Surface Temperature (LST)
The cooling effect at a height of 1.5 m above the
ground displays a similar pattern to that of the near- Results for the land surface temperature
ground temperatures (Figs. 2(B) and 3(B)). Once measurements are presented in Figs. 2(D) and 3(D).
again, the maximum cooling effect at 1.5 m As with the air temperatures, the maximum cooling
extended to 860 m (at 8 am on 30/05/2015), effect extended as far as 860 m from the park’s
although the maximum cooling effect was lower at centre (at 8 am on 30/05/2015 when a maximum
4.3oC (at 8 am on 30/05/2015). Also similar to the cooling of 11.7oC was recorded). For both sampling
pattern observed for temperatures 5 cm above the dates the maximum cooling effect was observed at 8
ground, the maximum cooling effects were apparent am and 12 pm, but significant cooling (1oC or more)
only for the 8 am and 12 pm measurements (for both occurred for all observation dates and times. Thus,
sampling dates) with all other times (except the 4 am the land surface temperature exhibited greater levels
measurement on 27/04/2015 which displayed a of cooling than either of the air temperature
3.6oC cooling effect) showing minimal or no cooling. measurements.

Fig. 1 Location of Melbourne, Australia (A), location of the case study site Carlton Gardens (B), the vegetation
cover and surrounding landscape of Carlton Gardens (C).

Table 1 Specifications of the 6 sample periods for each date.

Date Period (P) Total travelling Wind speed Humidity Moisture

time (minute) (km/h) (%) (g/kg)
27-28/04/2015 P1 (8-10) am 94 2.9 67 7.55
P2 (12-14) pm 77 7.3 67.5 7.4
P3 (4-6) pm 53 4.8 65.5 7.1
P4 (8-10) pm 71 0.8 61.5 9.3
P5 (12-2) am 45 1.5 68 8.5
P6 (4-6) am 48 2.4 70 6.2
30-31/05/2015 P1 (8-10) am 89 5.4 61 5.7
P2 (12-14) pm 100 7.85 41.7 5.7
P3 (4-6) pm 67 7.3 50 6.3
P4 (8-10) pm 75 4.1 62.5 6.7
P5 (12-2) am 67 2.6 68.5 6.8
P6 (4-6) am 54 5 65 6.25

2543
 International Journal of GEOMATE, Oct., 2016, Vol. 11, Issue 26, pp. 2541-2546

A. B.

C. D.

Fig. 2 Temperature measurements for 6 periods on 27-28/04/2015. (A) 5 cm above the ground (B) 1.5 m above
the ground (C) the difference between the measurements at 5 cm and 1.5 m (D) land surface temperature.

A. B.

C. D.
Fig. 3 Temperature measurements for 6 periods on 30-31/05/2015.(A) 5 cm above the ground (B) 1.5 m above
the ground (C) the difference between the measurements at 5 cm and 1.5 m (D) land surface temperature.

4. DISCUSSION

2544
 International Journal of GEOMATE, Oct., 2016, Vol. 11, Issue 26, pp. 2541-2546

A medium sized urban park in Melbourne, Urban parks have been demonstrated to have a
Victoria is shown to exert a considerable cooling significant cooling effect on surrounding urban areas.
impact on the surrounding city. This effect is most This effect is strongest for land surface temperatures
pronounced during the morning until around noon, (compared to air temperatures) and in the early
with the maximum observed cooling at 8 am and the morning and around noon. The cooling effect can be
second highest cooling effect observed at 12:00 pm felt as far as 860 m from the park boundary and can
for both sample dates. Although cooling was also exceed 11oC for land surface temperatures and 6oC
observed to occur at other times of the day it was for air temperatures. These results suggest that parks
less significant. These results contrast with those of represent a viable means of controlling urban
[16] who found that maximum cooling occurred at temperatures and that a strategic increase in
night although their study was conducted in the parklands could offset even the worst predicted
summer, not the winter as was this study. temperature rises associated with global warming
Interestingly, the cooling effect of the park in this (i.e., 2-5oC).
study is most pronounced for land surface
6. REFERENCES
temperatures (rather than air temperatures) which
exhibit a cooling influence from the park at all times [1] Chen, X-L, Zhao, H-M, Li, P-X & Yin, Z-Y
of the day. 2006, “Remote sensing image-based analysis
The distances over which these cooling effects of the relationship between urban heat island
were observed ranged from relatively short (100-200 and land use/cover changes”, Remote sensing
m) during the late afternoon and evening (4 pm and of environment, Vol. 104, no. 2, pp. 133-146.
8 pm) to as much as 860 m at 8 am in the morning, [2] Meehl, GA & Tebaldi, C 2004, “More intense,
which is similar to the results of [30] who showed more frequent, and longer lasting heat waves in
that park cooling effects extended as much as 840 m. the 21st century”, Science, Vol. 305, no. 5686,
This indicates that the park has a significant cooling pp. 994-997.
effect that extends well out into the surrounding city, [3] Kinney, PL, O'Neill, MS, Bell, ML & Schwartz,
J 2008, “Approaches for estimating effects of
but that this is not equally true at all times of the day.
climate change on heat-related deaths:
In terms of the three different temperature
challenges and opportunities”, Environmental
measurements (near-ground, above-ground and land Science and Policy, Vol. 11, no. 1, pp. 87-96.
surface), the land surface temperatures are affected [4] Bowler, DE, Buyung-Ali, L, Knight, TM &
the most by the presence of the park (up to 11oC) Pullin, AS 2010, “Urban greening to cool
followed by near-ground air temperatures (up to towns and cities: A systematic review of the
6oC) and then air temperatures at 1.5 m above the empirical evidence”, Landscape and urban
ground (up to 4.5oC). These results compare well planning, Vol. 97, no. 3, pp. 147-155.
with those of [31] who showed park cooling effects [5] Lokoshchenko, MA 2014, “Urban 'heat island' in
of 5-7oC and who also showed that land surface Moscow”, Urban Climate, Vol. 10, no. P3, pp.
temperatures were more sensitive to cooling than 550-562.
were air temperatures. These results mean that [6] Santamouris, M 2015, “Analyzing the heat island
studies that focus solely on measuring temperatures magnitude and characteristics in one hundred
at 1.5 m (which is common in the literature) are less Asian and Australian cities and regions”,
likely to see the true cooling potential of the park. Science of the Total Environment, Vol. 512-
However, as this height above the ground is the most 513, pp. 582-598.
significant for people in terms of their experience of [7] Kim, Y-H & Baik, J-J 2005, “Spatial and
urban heat, it also suggests that the park is less able temporal structure of the urban heat island in
to provide cooling for comfort than it is at providing Seoul”, Journal of Applied Meteorology, Vol.
overall cooling of the environment. 44, no. 5, pp. 591-605.
In addition, it should be noted that the much [8] Kleerekoper, L, van Esch, M & Salcedo, TB
higher levels of cooling displayed by land surfaces 2012, “How to make a city climate-proof,
addressing the urban heat island effect”,
means that the park could potentially reduce
Resources, Conservation and Recycling, Vol.
temperatures within buildings by cooling external
64, pp. 30-38.
surfaces. This may even reduce the need for artificial [9] Qiu, G-y, Li, H-y, Zhang, Q-t, Chen, W, Liang,
cooling (which also contributes to the UHI effect). X-j & LI, X-z 2013, “Effects of
Likewise, other urban surfaces such as cars, roads Evapotranspiration on Mitigation of Urban
and footpaths can be impacted upon, which may Temperature by Vegetation and Urban
help to mitigate against general increases in Agriculture”, Journal of Integrative Agriculture,
atmospheric temperature. Vol. 12, no. 8, pp. 1307-1315.
[10] Rosenfeld, AH, Akbari, H, Bretz, S, Fishman,
BL, Kurn, DM, Sailor, D & Taha, H 1995,
5. CONCLUSION “Mitigation of urban heat islands: materials,

2545
 International Journal of GEOMATE, Oct., 2016, Vol. 11, Issue 26, pp. 2541-2546

utility programs, updates”, Energy and warming”, Landscape Ecology, Vol. 28, no. 5,
Buildings, Vol. 22, no. 3, pp. 255-265. pp. 959-978.
[11] Figuerola, PI & Mazzeo, NA 1998, “Urban-
rural temperature differences in Buenos Aires”,
International Journal of Climatology, Vol. 18, [23] Chang, C-R, Li, M-H & Chang, S-D 2007, “A
no. 15, pp. 1709-1723. preliminary study on the local cool-island
[12] Magee, N, Curtis, J & Wendler, G 1999, “The intensity of Taipei city parks”, Landscape and
urban heat island effect at Fairbanks, Alaska”, Urban Planning, Vol. 80, no. 4, pp. 386-395.
Theoretical and Applied Climatology, Vol. 64, [24] Speak, A, Rothwell, J, Lindley, S & Smith, C
no. 1-2, pp. 39-47. 2013, “Reduction of the urban cooling effects
[13] Morris, C & Simmonds, I 2000, “Associations of an intensive green roof due to vegetation
between varying magnitudes of the urban heat damage”, Urban Climate, Vol. 3, pp. 40-55.
island and the synoptic climatology in [25] Li, D, Bou-Zeid, E & Oppenheimer, M 2014,
Melbourne, Australia”, International Journal of “The effectiveness of cool and green roofs as
Climatology, Vol. 20, no. 15, pp. 1931-1954. urban heat island mitigation strategies”,
[14] Unger, J, Sümeghy, Z & Zoboki, J 2001, Environmental Research Letters, Vol. 9, no. 5.
“Temperature cross-section features in an [26] Lopes, A, Alves, E, Alcoforado, MJ & Machete,
urban area”, Atmospheric Research, Vol. 58, R 2013, “Lisbon Urban Heat Island Updated:
no. 2, pp. 117-127. New Highlights about the Relationships
[15] Pal, S, Xueref-Remy, I, Ammoura, L, Chazette, between Thermal Patterns and Wind Regimes”,
P, Gibert, F, Royer, P, Dieudonné, E, Dupont, Advances in Meteorology, Vol. 2013.
JC, Haeffelin, M, Lac, C, Lopez, M, Morille, Y [27] Norton, BA, Coutts, AM, Livesley, SJ, Harris,
& Ravetta, F 2012, “Spatio-temporal variability RJ, Hunter, AM & Williams, NSG 2015,
of the atmospheric boundary layer depth over “Planning for cooler cities: A framework to
the Paris agglomeration: An assessment of the prioritise green infrastructure to mitigate high
impact of the urban heat island intensity”, temperatures in urban landscapes”, Landscape
Atmospheric Environ., Vol. 63, pp. 261-275. and Urban Planning, Vol. 134, pp. 127-138.
[16] Vardoulakis, E, Karamanis, D, Fotiadi, A & [28] Wong, NH & Yu, C 2005, “Study of green
Mihalakakou, G 2013, “The urban heat island areas and urban heat island in a tropical city”,
effect in a small Mediterranean city of high Habitat International, Vol. 29, no. 3, pp. 547-
summer temperatures and cooling energy 558.
demands”, Solar Energy, Vol. 94, pp. 128-144. [29] Batra, N, Islam, S, Venturini, V, Bisht, G &
[17]Tumanov, S, Stan-Sion, A, Lupu, A, Soci, C & Jiang, L 2006, “Estimation and comparison of
Oprea, C 1999, “Influences of the city of evapotranspiration from MODIS and AVHRR
Bucharest on weather and climate parameters”, sensors for clear sky days over the Southern
Atmospheric Environment, Vol. 33, no. 24, pp. Great Plains”, Remote Sensing of Environment,
4173-4183. Vol. 103, no. 1, pp. 1-15.
[18] Bai, L, Ding, G, Gu, S, Bi, P, Su, B, Qin, D, Xu, [30] Lin, W, Yu, T, Chang, X, Wu, W & Zhang, Y
G & Liu, Q 2014, “The effects of summer 2015, 'Calculating cooling extents of green
temperature and heat waves on heat-related parks using remote sensing: Method and test',
illness in a coastal city of China, 2011-2013”, Landscape and Urban Planning, Vol. 134, pp.
Environ. Research, Vol. 132, pp. 212-219. 66-75.
[19] Williams, S, Nitschke, M, Sullivan, T, Tucker, [31] Spronken-Smith, RA & Oke, TR 1998, 'The
GR, Weinstein, P, Pisaniello, DL, Parton, KA thermal regime of urban parks in two cities
& Bi, P 2012, “Heat and health in Adelaide, with different summer climates', International
South Australia: Assessment of heat thresholds Journal of Remote Sensing, Vol. 19, no. 11, pp.
and temperature relationships”, Science of the 2085-2104.
Total Environment, Vol. 414, pp. 126-133.
[20] MMWR 1996, “Heat-Wave-Related Mortality - International Journal of GEOMATE, Oct., 2016,
- Milwaukee, Wisconsin”, July 1995viewed. Vol. 11, Issue 26, pp. 2541-2546.
[21]Hamada, S & Ohta, T 2010, “Seasonal MS No. 5183 received on June 12, 2015 and
variations in the cooling effect of urban green reviewed under GEOMATE publication policies.
areas on surrounding urban areas”, Urban Copyright © 2016, Int. J. of GEOMATE. All
forestry & urban greening, Vol. 9, no. 1, pp. rights reserved, including the making of copies
15-24. unless permission is obtained from the copyright
[22] Myint, SW, Wentz, EA, Brazel, AJ & proprietors. Pertinent discussion including
Quattrochi, DA 2013, “The impact of distinct authors’ closure, if any, will be published in Oct.
anthropogenic and vegetation features on urban 2017 if the discussion is received by April 2017.
Corresponding Author: Scott Rayburg

2546