The effect of physical and psychological environments on the users thermal perceptions of educational urban precincts

Accepted Manuscript The effect of physical and psychological environments on the users thermal perceptions of educational urban precincts Salman Shooshtarian, Ian Ridley PII:

S0360-1323(16)30517-0

DOI:

10.1016/j.buildenv.2016.12.022

Reference:

BAE 4752

To appear in:

Building and Environment

Received Date: 22 July 2016 Revised Date:

7 December 2016

Accepted Date: 14 December 2016

Please cite this article as: Shooshtarian S, Ridley I, The effect of physical and psychological environments on the users thermal perceptions of educational urban precincts, Building and Environment (2017), doi: 10.1016/j.buildenv.2016.12.022. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT THE EFFECT OF PHYSICAL AND PSYCHOLOGICAL ENVIRONMENTS ON THE USERS THERMAL PERCEPTIONS OF EDUCATIONAL URBAN PRECINCTS

Salman Shooshtarian School of Property, Construction and Project Management, RMIT University, Melbourne, Australia.

RI PT

Ian Ridley

AC C

EP

TE D

M AN U

SC

School of Energy and Environment, City University of Hong Kong, Kowloon, Hong Kong

ACCEPTED MANUSCRIPT THE EFFECT OF PHYSICAL AND PSYCHOLOGICAL ENVIRONMENTS ON THE USERS’ THERMAL PERCEPTIONS OF EDUCATIONAL URBAN PRECINCTS Abstract

TE D

M AN U

SC

RI PT

Assessment of outdoor thermal comfort requirements is of primary importance in development of sustainable outdoor environments including urban precincts. The full understanding of thermally comfortable conditions outdoors requires investigating other factors that are beyond the thermophysiological aspect of thermal comfort including those contextual factors that are contextually impacting people’s thermal judgment. Using a socio-ecological system model (SESM) as the research framework, this study aimed to investigate the role of physical environment factors (i.e. weather conditions, including solar radiation intensity, spatial features: sky clearness and aspect ratio, length of residence and time of exposure) and the psychological environment factors (i.e. users’ overall comfort and thermal preference, purpose and frequency of visit, seasonal change, thermal history, consideration of weather forecasts, character and features of place, and naturalness). The data used in this study were collected during three rounds of the field surveys consisting of on-site measurement and questionnaire surveys. The field surveys were conducted in three case study sites which were the premises of an educational precinct in Melbourne, Australia, from November 2014 to May 2015. The analytical results demonstrated a medium and low influence of physical and psychological environments on outdoor thermal sensations, respectively. The research outcome has shed light on those aspects of thermal comfort in outdoor spaces that are yet to be fully understood.

Keywords:

EP

Outdoor thermal comfort, contextual factors, climate sensitive design, environmental parameters, design descriptors, inter-seasonal change

1. Introduction

AC C

The occurrence of more intense urban heat island (UHI) effects together with evident effects of climate change including unusual climate conditions and frequent heatwaves has deteriorated people’s outdoor thermal environment (Coutts et al. 2007, Hatvani-Kovacs and Boland 2015). Hence, in recent years the number of studies assessing outdoor thermal comfort requirements has been on the rise; since this is an issue directly related to outdoor space users’ health and well-being. These studies have attempted to understand the various dimensions of thermal comfort conditions leading to the development of policies and standards being concerned with navigating efforts to create sustainable urban spaces such as urban precincts (Eliasson 2000, Wilson et al. 2007, Coutts et al. 2012). The common method to assess thermal comfort is based on the heat balance comfort models (Fanger 1970, Gagge et al. 1986) and deals with prediction of thermal comfort for a large number of people

1

ACCEPTED MANUSCRIPT using the collective effects of four environmental parameters (i.e. air and radiant temperature, relative humidity, wind speed) and two personal factors (i.e. level of clothing insulation and activity). Concurrent to measurements and prediction, people’s thermal response is also acquired to provide information on the real-world perceptions of thermal environment and to potentially compare or calibrate against the prediction. It is argued here that this method relies too heavily on the thermoregulation aspects of thermal comfort and overlooks the modifying role of other contextual

RI PT

factors in determination of thermal perceptions. This is of particular importance since according to ASHRAE 55 (2010) definition thermal comfort is a thermal satisfaction expressed by people’s mind conditions.

Therefore, it is imperative to take into account the various determinants of thermal perceptions in

SC

outdoor spaces with the view to providing a more explicit explanation for the dynamic of outdoor thermal comfort conditions. This study modified and employed a theoretical model that is Socio

M AN U

Ecological System Model (SESM) to address the research objectives. This model which is built on the “Ecological System Theory” (Bronfenbrenner 1992) has long been used to study the human attitude and behaviour in an ecosystem. This multi-cluster framework assumes that people are influenced through a set of environments that together with their personal characteristics create the knowledge of the reality (Bronfenbrenner and Evans 2000).

This model consisting of five

environments was used to explore the effect of various contextual factors on thermal perceptions;

TE D

these environments included individual, social, physical, and psychological, and policies and standards. The results of analysis on the first two environments are presented before in (Shooshtarian and Ridley 2016b). This study is concerned with investigation of the modifying role of factors being classified under two environments: (1) physical environment (i.e. weather conditions, including solar radiation intensity, spatial features: sky clearness and aspect ratio, length of residence and time of

EP

exposure) and (2) the psychological environment (i.e. users’ overall comfort and thermal preference, purpose and frequency of visit, seasonal change, thermal history, consideration of weather forecasts,

AC C

character and features of place, and naturalness). To shed light on the impact of the above-mentioned physical and psychological factors on outdoor thermal perceptions a summary of the relevant literature is presented below.

The role of spatial design in determination of thermal perceptions has been investigated in several studies (Djenane et al. 2008, Lin et al. 2010, Bourbia and Boucheriba 2010, Mahmoud 2011, Qaid and Ossen 2014). Two most frequently investigated features of urban design in relation to human thermal comfort are aspect ratio and level of sky clearness (Table 1). These two factors are primarily assessed and reported owing to their ease of estimation and relatively proven impact on urban microclimate (Oke 1982, Matzarakis et al. 2007, Bourbia and Boucheriba 2010). As indicated in previous studies these factors can alter the air movement pattern (wind speed), solar radiation intensity (mean radiant

2

ACCEPTED MANUSCRIPT temperature) and thus give variation to shadow pattern (air and surface temperature). Table 1 summarises the key findings of recent studies on the impact of these two design descriptors on the outdoor thermal perceptions.

Aspect ratio is the ratio between the average height of buildings and intermediate width of the Study

RI PT

Street. This feature influences the magnitude of both incoming and outgoing solar radiation, and wind pattern. Johansson (2006) argued that a higher aspect ratio improves the summertime comfort whereas it is a source of discomfort in wintertime. He also found a larger stability in thermal conditions in deep street canyons with higher aspect ratio compared to shallow street canyons. Djenane et al. (2008)

SC

demonstrated the dependence of the given thermal budget on the urban design, including H/W ratio. Furthermore, adjustment in H/W ratio was identified as a mitigation strategy in response to hot

M AN U

conditions (Bourbia and Boucheriba 2010).

Solar radiation, including short and long-wave fluxes received by outdoor users influences their energy balance. Sky clearness is used to present the level of shading in the open spaces encompassing buildings, trees, landscape and other urban structures which modify the visible horizon and incoming radiation (Oke 1982). Sky view factor (SVF) is defined as the fraction of free sky at the given location ranging from 0 to 1, indicating completely obstructed and completely vertically free space,

TE D

respectively (Oke 1982). As SVF controls the incidence of direct solar radiation, it can show a positive and negative impact in summertime and in wintertime, respectively, in regards to human thermal perceptions (Eliasson 1994).

In this line, He et al. (2015) explained further consequences of SVF impact as “…SVF being an

EP

indicator of urban canyon geometry affects the surface energy balance, local air circulation, and outdoor thermal comfort” (p. 285). In another study, the thermal conditions of eight urban

AC C

environments in a semi - arid climate of Algeria were investigated; and the built-up surfaces such as asphalt were noted as the heat traps required to be mitigated by the creation of obstacles so as to reduce the SVF level and thus surface and air temperature (Bourbia and Boucheriba 2010). In hot and arid climate of Cairo, Mahmoud (2011) observed the impact of SVF on thermal conditions in eight settings of an urban park. It is noteworthy to indicate that in some climate conditions while SVF brings thermal comfort in some seasons it can have the opposite effect in others. Lin et al. (2010) conducted an investigation to understand the long-term effect of SVF on thermal comfort in a University campus in Taiwan. The analytic results being based on predicted comfort showed that the number of hours of discomfort varied with the level of SVF, mostly comfortable in a shaded area in the summer and uncomfortable in winter times. What was an interesting finding was a similar discomfort hours in both seasons, indicating the role of obstacles in controlling the wind speed 3

ACCEPTED MANUSCRIPT and solar radiation. However, these results would have been more fulfilling if the researchers had considered concurrent thermal responses. The same trend was also observed in another study (Hwang and Lin 2011) where the effect of shading, produced by various SVF levels, on thermal comfort was found to be dissimilar in various seasons. In Curitiba, Brazil, one study considered using data from different sources: field measurement, survey and simulation to better characterize the urban design impact on outdoor thermal conditions and comfort (Krüger et al. 2011). Interestingly, the results

RI PT

pointed out that SVF is not crucial in determination of thermal comfort and other factors should be taken into consideration. To sum up, the major trend found in the literature shows the impact of urban design on both thermal conditions and perceptions. However, the level of impact on people needs to be investigated in relation to the contextual conditions, including prevailing microclimate, dominant

SC

usage pattern, type of users; and function and design of space.

M AN U

Length of residence (LoR) and time of exposure (ToE) to outdoor environmental parameters are linked to long and short-term physical (physiological) thermal adaptation (Brager and de Dear 1998). According to the concept of thermal adaptation, and acclimatization in particular people are expected to be gradually acclimatized to a local microclimate when they are repeatedly exposed to it (De dear and Brager 1998). This process is otherwise known as physiological thermal adaptation. Also, extended exposure to outdoor environmental stimulus induces the reflective physiological thermal

TE D

adaptation but in the shorter term (Humphreys and Nicol 1998).

Considering ToE in the assessment of outdoor thermal comfort is of particular importance for a number of reasons:

EP

(1) users of open spaces often stay for a short while and are thus transient users (Leech et al. 2002, Aljawabra and Nikolopoulou 2010) (2) and behavioural adjustments by people take place following a considerable time spent

AC C

outdoor (Ahmed et al. 2015);

(3) this factor was also categorized under psychological thermal adaptation by in that it changes thermal expectations which in turn affects thermal perceptions. In this line Krüger et al. (2015) indicated that “...stepping from thermal homogeneity to transient outdoor conditions

should create immediate responses that would then diminish with time of exposure” (p. 1). Höppe (2002) found that thermal steady state, being the basis of typical thermal comfort indices, is never reached after several hours, or may be reached after 30 minutes in cold and warm conditions, respectively. Thermal comfort under non-steady state conditions primarily deals with rapid microclimate transients and noticeable changes in microclimate conditions, level of activity and clothing insulation within the course of minutes (Katavoutas et al. 2015)

4

ACCEPTED MANUSCRIPT

Some studies have shown the significant relationship between LoR and thermal perceptions in outdoor spaces of various climate conditions, including Tropical (Chow et al. 2016) and Sub-tropical (Makaremi et al. 2012, Yin et al. 2012) and Temperate climates (Kenawy and Elkadi 2011). In the sub-tropical climate of Shanghai observed that in winter time people with prolonged residence were

RI PT

better adapted to the given thermal conditions. There are a large number of research studies wherein the duration of participants’ residence was registered and no analytical result was subsequently presented to establish a relationship between LoR and thermal perception (Oliveira and Andrade 2007, Pantavou et al. 2013). In other studies the importance of LoR was simply neglected where the participants with short length of residence were excluded from interviews (Krüger and Rossi 2011). In

SC

spite of obtaining a more uniform study population, excluding such people from study would result in having a sample size, which is not representing the real world conditions wherein open spaces are

M AN U

likely to be visited by different people.

Thermal history is long known to be a key psychological mechanism to the adaptation to indoor (Humphreys 1995) and recently in outdoor thermal conditions (Nikolopoulou and Steemers 2003, Knez et al. 2009) which theoretically influences thermal sensations. Past thermal history is often assumed to modify expectations and preferences for thermal conditions (De dear et al. 1997). The meaningful relationship between past thermal history and thermal sensations has been verified in

TE D

several studies (Aljawabra and Nikolopoulou 2010, Lin et al. 2011, Yin et al. 2012). Becoming aware of weather conditions an individual would experience during a day is an effective thermal adaptive behaviour. According to the concept of thermal adaptation, people tend to react to a given thermal environment in an attempt to better cope with environmental stressors. Many people

EP

tend to check forthcoming weather conditions prior to leaving an indoor or semi-indoor environment. Getting to know weather conditions outside, not only makes people to be mentally prepared to the

AC C

given thermal conditions (Yin et al. 2012, Leviston et al. 2015), but also encourages them to physically adjust their the level of clothing or activity for a better adaptation (Höppe 1999, Chun et al. 2008). In a report (Leviston et al. 2015) describing Australian attitudes to climate change it is indicated that if people have knowledge of what they are to perceive outside there is a greater likelihood of better adaptation to weather conditions. In a comparative study on the adaptive behaviour of two nations it was found that Japanese were more tolerant of weather conditions than their Korean counterparts simply because the former tended to check weather forecasts more often than latter (Chun et al. 2008). They also observed that Japanese’s clothing pattern was primarily influenced by weather forecasts than fashion. In this line, Höppe (1999) indicated that weather reports should also advice the audience on the kind of clothing before leaving home which would most likely provide better thermal comfort. Yin et al. (2012) argued that outdoor users foresee the type of weather 5

ACCEPTED MANUSCRIPT conditions they will face and it causes a change in their thermal expectations; which in turn will affect their thermal perceptions. Overall comfort and preference for a certain thermal condition (two elements of thermal perception) could potentially influence thermal sensations. The conceptual and functional characteristics of these elements were compared in the previous studies: in indoor, semi-indoor (Brager et al. 1993, Andamon

RI PT

2005, Zhang and Zhao 2008) and outdoor conditions (Chen et al. 2015, Pantavou et al. 2013, Shooshtarian and Ridley 2016a). According to key findings reported, some studies proved the influence of such elements on subjective thermal sensations (Cheng et al. 2012, Pantavou et al. 2013). There is some evidence, however, which present the opposite trend in certain circumstances such as seasides wherein people sensed thermal conditions outside specified comfort zones while they still

SC

preferred the current conditions identified as discomfort (de Freitas 1985, Höppe and Seidl 1991). Therefore, these two perception elements are theoretically linked to other than thermal factors,

M AN U

including psychological drivers, which depending on conditions may or may not influence thermal sensations.

Purpose and frequency of visit (FoV) to an open space could potentially affect people’s thermal perceptions in accordance to the conventional wisdom. It is assumed that people are gradually acclimatized to a local microclimate when they are repeatedly exposed to it. Reviewing the relevant literature, Johansson et al. (2014) argued that the purpose of visit to an open space evidently

TE D

influences thermal perceptions as those users passing through an space to reach another place may not be concerned with weather conditions, in the same way as when they are in a recreational spaces, wherein people may avoid visiting an space with poor comfort conditions. Intention and frequency of spatial use is somewhat linked to function of a space. In other words, function of a place including

EP

available facilities, design options, and level of accessibility partially determines intention and frequency of usage in open spaces (Zacharias et al. 2004, Knez et al. 2009). Several studies have reported that in public spaces with different functions people’s thermal evaluation and thus usage

AC C

pattern differed (Thorsson et al. 2007, Nikolopoulou and Lykoudis 2007, Spangenberg et al. 2008, Chen et al. 2015).

Despite the existence of a limited number of studies assessing the impact of these two factors on thermal perception (Nikolopoulou and Lykoudis 2007, Oliveira and Andrade 2007, Pantavou et al. 2013, Chow et al. 2016), a direct relationship was only rarely established (Pantavou et al. 2013, Lam et al. 2016, Middel et al. 2016). Instead, they tried to link usage pattern to the character of various spaces to show such relationship using indicators such as number of attendants, as a function of thermal perceptions (Nikolopoulou and Lykoudis 2007, Eliasson et al. 2007, Lai et al. 2014). Seasonal change profoundly dictates the way people evaluate thermal conditions outdoors. To better determine the possible effect of seasonal change on outdoor thermal perceptions a myriad of studies 6

ACCEPTED MANUSCRIPT have launched comparative evaluation of thermal comfort conditions in various seasons (Spagnolo and De Dear 2003, Lin et al. 2011, Pantavou et al. 2013, Huang et al. 2015, Shooshtarian et al. 2015, Middel et al. 2016). It is often assumed that change in people’s thermal expectations coincide with change in season. Moreover, the neuropsychological concept of perceptual alliesthesia (Cabanac 1971) allows researcher to explain the dynamic of change in people’s thermal perceptions in different seasons (Spagnolo and De Dear 2003). According to Alliesthesia “…a given stimulus can induce a

RI PT

pleasant or unpleasant sensation depending on the subject’s internal state” (Cabanac 1971, p 1107) and hence individuals long for a differing thermal conditions in different seasons.

Depending on their spatial features, open spaces can attract people for different reasons. Given the fact that each open space may bring a convenience to people, users theoretically can compromise their

SC

thermal judgement to take advantage of such comfort (Nikolopoulou 2004). Knez et al. (2009) put forward a model to understand the psychological mechanism of thermal perceptions in which the

M AN U

place character is regarded as a moderating factor in people’s thermal experience. The spatial features include aesthetic and visual characteristics of space, accessibility, facilities provided in place, type of users, opportunities to adapt to thermal conditions and suchlike that together shapes the character of place. The character of space is closely related to the function of place and sometimes they are interchangeably used (Alcoforado et al. 2009). The character of space is closely related to function of place and sometimes they are interchangeably used (Alcoforado et al. 2009).

TE D

Among others, spatial naturalness is one of the conveniences that a space can offer to users; this term is a twofold concept: on one hand, it is concerned with the extent to which a user can be connected to nature and on the other hand, it reflects people’s opinion on the level of spatial naturalness in open spaces. Nikolopoulou and Steemers (2003) suggested the naturalness as one of the sixfold factors

EP

forming mechanism for psychological thermal adaption in outdoor spaces. Reviewing the relevant literature the moderating role of these factors in variation of thermal

AC C

sensations has become apparent. Therefore, it is worthwhile to perform a research project through which the interactions between people’s thermal perceptions and these factors can be investigated. Therefore, in exploring such interactions in an urban educational precinct this research aimed to achieve the following objectives: (1) to examine the influence of the each factor listed above on outdoor thermal sensations, (2) to further analyse the overall relationship between physical and psychological environments and thermal sensations (3) to study the impact of inter-seasonal change on the modifying influence of the study factors, and (4) to assess the users’ specific requirements of thermal comfort in the context of educational urban precincts.

2. Material and Method 2.1 Case studies 7

ACCEPTED MANUSCRIPT The field studies were carried out in an educational urban precinct in Melbourne City Centre. Melbourne has an oceanic climate (Cfb) according to the most updated Köppen-Geiger climate classification (Peel et al. 2007). The study area is subject to a range of issues including UHI effects caused by surrounding high-rise buildings, densely urbanised and crowded spaces (Coutts et al. 2007, Chen et al. 2013). A number of studies assessing microclimate conditions of

Melbourne

demonstrated the existence of an UHI, approximating from 2 ºC to 4 ºC, with diurnal spikes as much

RI PT

as 7 ºC, depending on time of day, year and location (Morris and Simmonds 2000, Torok et al. 2001, Chen et al. 2013). Three open spaces were selected as the case studies, which are the premises of the RMIT University City Campus (37°48 S 144°57 E). These sites are also representative of typical urban spaces in the inner city of Melbourne. These sites are mostly frequented by university students

SC

and staff; and to a lesser extent by other users including employees working in the neighbourhood and tourists visiting Melbourne CBD. These three sites are primarily designed to serve as access paths to academic buildings in RUCC and to provide opportunities for students and staff to take rests between

M AN U

their breaks. Site 1: University Lawn which is used as a recreational space by university students and staff. The site is mostly used by the non-transient users where a university café is located nearby and an area of Astro turf that facilitates a longer stay outdoors. The compact design of University Lawn makes it representative of many recreational outdoor spaces in the inner city. This venue has a varied surface coverage; artificial grass, timber deck, water features and a natural green space. Site 2: Ellis Court is used as a thoroughfare and a main path to other parts of the campus. Also, its court, providing

TE D

shade and sitting areas, serves those wishing to stay longer outdoors. On average, the site is more frequented by passers-by to get to other buildings than the non-transient users. The site includes a range of urban elements and surface covers that collectively affect thermal conditions and resemble many urban precincts in the City of Melbourne. Site 3: RMIT A’Beckket Urban Square is a 2800

EP

m2 recreational space, which provides multi-functional courts for outdoor activities, green spaces, and shading features. Figure 1 shows the geographical location of these case studies in the heart of

AC C

Melbourne City Centre.

2.2 Field survey

The field surveys were carried out in three successive seasons to assess thermal comfort conditions across a wide range of microclimate conditions. Thus, three months of November (2014), February and May (2015) were selected as they best represented the study seasons (i.e. spring, summer and autumn); as these months occur at the end of each season, it is argued that the study took place when the participants’ thermal history and expectations for each season had become settled . The summary of monthly mean microclimate conditions of the case studies is presented in Table 2. The 8

ACCEPTED MANUSCRIPT microclimate data were acquired from a Bureau of Meteorology (BOM) Melbourne (Olympic Park) Station (ID: 086338), and was taken for the period of November 2014 to May 2015. The field surveys comprised of performing on-site measurements of microclimate parameters concurrent to administration of questionnaire surveys. The surveys carried out during the busiest times at each site (9:00 am to 5:00 pm) allowing a better reflection of interaction between wide range of people and study open spaces and recruitment of more participants. Table 3 specifies the date of field surveys

RI PT

during the three study seasons.

2.3 Survey population

SC

Included in the survey population were male participants (N=704, 66.4%) and female counterparts (N=355, 33.6%) who were mostly from a young age group of 18-30 (60.5%). The majority of

M AN U

participants were students (62.7%) who were mostly not born in Melbourne (65%). The respondents were present in the outdoor environments predominantly for a short time period (5-10 mins), in most cases less than 5 minutes (39%). The distinctive nature of a university campus accommodating people from different cultural and climate backgrounds provided an opportunity to better study contextual factors among a non-uniform target population; this in turn allowed the research findings to represent public open spaces located in global cities described by the diversity of its population.

TE D

The participants were asked to indicate their length of residence (LoR) in Melbourne, which was then categorised into four period ranges: “less than one month to 12 months” (N=102), “1 to 3 years” (N=71), “3 to 10 years” (N=140) and “10 years and above” (N=500). The four temporal categories of ToE initially reported by participants were re-categorised into two groups: “below 5 min” (N=414)

EP

and “above 5 min” (N=606). The users who stayed outdoor “below 5 mins” were considered as transient users and those who stayed above that were non-transient users.

AC C

2.4 Meteorological measurements and thermal comfort index

The protocol for the measurement including the measuring range and the accuracy of the instruments is in 100% compliance with ASHRAE 55 (2010) and ISO 7730 (2006). During the field surveys four environmental variables: air temperature (Ta), wind velocity (Va), relative humidity (RH) and globe temperature (Tg) were measured that together with the two personal factors (level of activity and clothing insulation) were used to predict conditions of thermal comfort for a large number of people in outdoor environment. As the intensity of solar radiation intensity (SR) was not directly considered in the calculation of PET, its values were also monitored concurrent to field surveys using a Pyranometer. A mobile weather station (Testo 480 IAQ Pro Measurement Kit) was used to monitor 9

ACCEPTED MANUSCRIPT the four environmental variables mentioned above. This kit consisted of the following probes: TESTO IAQ probe 0632 (Ta and RH), TESTO Globe thermometer 0602 (Tg), and TESTO COMFORT 0628 (Va). The measuring interval was one minute and the TESTO 480 data logger and H21-002-HOBO Micro Station logged the measurements. To predict outdoor thermal sensations this study used the Physiological Equivalent Temperature

RI PT

(PET)(Mayer and Höppe 1987) as the thermal comfort index. PET has been increasingly used in the recent studies on outdoor thermal comfort assessment; allowing comparative evaluation of thermal comfort requirements within various contexts. Rayman Software Package 1.2 was used to calculate PET values assuming the constant values for the level of activity (80 W) across all seasons and the averaged seasonal value of clothing insulation reported by the participants (spring= 0.55,

M AN U

SC

summer=0.41 and autumn= 0.81 Clo).

2.5 The questionnaire

The questionnaire was structured according to the universal thermal comfort standards of ISO 10551 (1995), (ISO 7730 2006) and ASHRAE 55 (2010) and was phrased in a simple and concise language to be comprehensible to all respondents particularly international students. The questionnaire elicited

TE D

information in three categories: the participants’ demographic details, their thermal responses, and spatial usage related factors. From the scales on thermal responses obtained this study only used ‘thermal sensation votes’ (TSV). Thermal sensation refers to sensory unconscious notification of environmental stimulation. Thermal sensation scale consisted of seven-point (“cold” (− 3), “cool

AC C

(3)).

EP

” (−2), “slightly cool” (− 1), “neutral” (0), “slightly warm” (1), “warm” (2) and “hot”

During the surveys, the researcher also recorded some information to reduce the time of participation and keep the rejection rate to a minimum. This information included: participants’ characteristics (e.g. gender, position, ensemble of clothing pattern, and their accessories), exact location of survey, type of usage (e.g. companionship and posture), and pre-awareness of weather conditions. On average, the questionnaire survey took less than five minutes to complete and participants had been briefed prior to the survey about the objectives of the study. The RMIT University Ethics and Human Research Committee approved the protocol used for the field survey.

2.6 Design descriptors of space

10

ACCEPTED MANUSCRIPT Two design descriptors of sky clearness and aspect ratio were studied to examine the role of spatial design on microclimate and thermal sensations. The sky clearness or SVF was quantified by calculations of the ratio between obstacles and total vertical horizon using 180° fish-eye images. The images were taken using Canon EOS 6D SLR which was fitted with Canon EF 8–15 mm f/4 L Fisheye USM. The SVF percentages were calculated through Rayman Software Package V.2.1 (Matzarakis et al. 2007). In each site five fish-eye images were taken which represented the vertical

RI PT

conditions of sky view of the point centred each of fivefold sub- areas; Table 4 displays three images taken in each site and associated SVF percentage.

For the purpose of better analyses and even distribution of thermal responses, the SVF values were categorised into four groups: “18-26%”, “26-34%”, “34-42%”, and “42-50%”. The aspect ratio was

SC

also calculated for those sub-areas wherein the largest number of thermal responses in each site was collected. In this way, the number of participants assigned to each category of aspect ratio was kept

M AN U

balanced. As a result, three levels of aspect ratio were defined as: “1.80” (N=285), “2.00” (N=295), and “2.82” (N=234). These levels represent the ratio between mean of surrounding buildings heights to width of the main street (Figure 2).

2.7 Data processing and analysis

TE D

In total 1059 questionnaires were collected across the three sites and three seasons; 36 questionnaires were excluded due to missing concurrent measurements. Statistical analysis was conducted to evaluate the interaction among the factors clustered under two SESM environments (physical and psychological), thermal conditions and thermal sensations of study urban precinct. Data obtained from

EP

the field surveys was screened and analysed using Microsoft Excel Spreadsheets V. 2010 and SPSS V.22. Where applicable, data groups were re-categorised to ensure a better comparison among study groups and to achieve even distributions of TSVs. The basis of analysis was on the aggregated dataset

AC C

(combined data for all seasons), however, the separate seasonal datasets were also used to further understand the dynamics of thermal comfort conditions in each season. In addition to descriptive analysis, inferential analysis was employed to determine the strength of correlation between predictors and response with association to given thermal conditions. Hence, ordinal logistic regression model (Norušis 2012), linear regression model, Pearson (for continuous dependent variable) and Spearman’s rank (for categorical dependent variables) correlation tests were used. The ordinal logistic regression (OLR) applied in a three-step process: (1) separately to each predictor of TSV clustered under the two SESM environments; (2) collectively to all of those factors found to be statistically significant in each environment during entire period and (3): seasonally to each predictor variable. The graphs and tables were plotted using Excel Spreadsheets. 11

ACCEPTED MANUSCRIPT 3. Results

3.1 Characteristics of thermal sensations

RI PT

During the study period people’s TSV were recorded over the entire range of the 7-point ASHRAE scale. As for aggregated data the largest percentage of TSVs (60%) were cast for the three central categories (+1, 0,-1). This trend was repeated in spring (64.4%) and summer (63.2%). However, in autumn the votes were more skewed towards the cool side of the scale with 88.5% of the responses varied from slightly cool (-1) to cold (-3). Figure 3 gives the characteristics of participants’ thermal

M AN U

SC

responses in different seasons.

3.2 Environmental parameters and thermal sensations

The role of contextual factors clustered under psychological environment on people’s thermal sensation was investigated using ordinal logistic regression model (OLRM). This environment

TE D

consisted of “environmental parameters”, “design descriptors”, “time of exposure”, and “length of residence in Melbourne”. The collective (in the form of PET values) and individual effect of environmental parameters on outdoor thermal sensations was respectively analysed using OLRM and Spearman’s correlation test, respectively. The results showed users’ thermal sensations were

EP

significantly impacted by four environmental variables as well as the intensity of solar radiations (Table 5). As such, environmental parameters were included in the overall regression model conducted for physical environment (Table 6). Table 5 and 6 present the summary findings of OLRM

AC C

for each and overall impact on thermal sensations in physical environment.

Meteorological data was also analysed using the Spearman’ correlation test to further understand the extent of the correlation between thermal sensations and each of environmental parameters (Table 7). From the test parameters, throughout the study period, Ta (r=0.72, P<0.01) and Tg (r=0.70, P<0.01) had the strongest correlation; wind speed and relative humidity were found to be in negative relationship with thermal sensations. Furthermore, it was found that the extent of influence of environmental parameters changed with season, for instance, unlike spring and summer, in autumn the wind speed values had a statistically significant effect on thermal sensations (r=-0.27, P<0.01). In

12

ACCEPTED MANUSCRIPT addition, the correlation between temperatures (Ta, Tg, Tmrt) and thermal sensations was more apparent in spring and summer compared to autumn (Table 7).

3.3 Design descriptors of space, thermal conditions and thermal sensations

RI PT

The results of ordinal regression among TSV, thermal conditions and SVF groups showed that SVF was not statistically linked to variation of TSV (Table 5). However, the closeness of its P-value to the significance level of 0.05 allows for inclusion of this variable in the overall OLRM for the physical environment (Table 6). To explore the effect of SVF on local microclimate, the correlation between the SVF and the environmental parameters was tested. Table 8 summarised the resultant statistics for

SC

Pearson’s correlation in different seasons and aggregated dataset. The SVF correlation with environmental parameters varied among the seasons; however a large correlation was found between

M AN U

Ta and SVF over the study seasons ranging from r=0.20 to r=0.48 respectively in autumn in summer. Furthermore, the effect of more obstructed sky on reduction of Va (r=-0.17, P<0.01) and RH (r=-0.24, P<0.01) values was found statistically significant but in the negative direction. The effect of SVF conditions on sunlight regarding its both indictor (Tg and SR) was found to be more evident in summer and autumn (Table 8).

The relationship between aspect ratio and environmental parameters was investigated and tabulated in

TE D

Table 8. Among the environmental parameters RH (P<0.01, r= 0.31) was found to have the largest correlation throughout the study period, followed by air temperature and solar radiation (P<0.01, r=.20). The correlation test showed that there was a negative but significant relationship in the cases of Ta, Tg and SR; however, the extent of relationship varied with season, for instance, in autumn, the

EP

correlation between environmental parameters and aspect ratio was the most apparent.

AC C

3.4 Length of residence, time of exposure and thermal sensations

The findings of OLRM analysis indicated that ToE is a significant determinant of users’ outdoor thermal sensations over the course of study (Table 5); hence, it was included in the overall regression model (Table 6). The overall ordinal logistic regression model of physical environment revealed that according to Nagelkerke’s classification of Pseudo- r squared (Delhey and Newton 2002) physicalrelated contextual factors had a medium influence (psudo-r2: 54.1) on variation of outdoor thermal sensations in the context of study.

3.5 Purpose and frequency of visit to spaces and thermal sensations

13

ACCEPTED MANUSCRIPT The modifying role of factors clustered under the SESM psychological environment in variation of TSV under various microclimate conditions, were investigated using ordinal regression model (Table 9). Included in the psychological environment were: “purpose and frequency of visit”, “overall comfort”, “thermal preference”, “seasonal change”, “thermal history”, “features of place”, “character of place”, and “the level of naturalness”. The relationships between TSVs, purpose (PoV) and frequency of visit (FoV) were examined for the period of study (Table 9). The initial five levels of

RI PT

visit frequency were re-categorised into three groups as: “daily to several times a week”, “few times a week to few times a month” and “rarely to first time”. Almost half of the survey users (N=598) had a pattern usage of “daily to few times a day” visits. The findings of regression analysis showed that

SC

FoV had an insignificant relationship with TSV variation throughout the study period (P>0.05).

As shown in Figure 4, on average, more than 50% of PoV in all study spaces belonged to “having

M AN U

break/resting” and “getting fresh air/change of environment”. However, this varied across the study sites; while having break/resting seemed to be the primary PoV in sites 1 (29.6% of total use) and 3 (37.2% of total use), “passage to another place” (28.3% of total use) alongside “having break/resting” (26.4% of total use) were most common in the use of site 2. At site 3, which included sports facilities, 11.1% of use was accounted for by “playing”, while only a small number of people used this space as a means to go to another place (3.0 %). Accommodating a café and many places to sit, site 1 had the

TE D

highest percentage of users to have their launch/snack (20.6%), while this type of usage only accounted for 11.60% and 12.59% of total use in sites 2 and 3, respectively (Figure 4). The results of OLRM on the relationship between TSV and PoV showed no meaningful interaction for the

EP

aggregated data (Table 9).

3.6 Overall comfort, thermal preference and thermal sensations

AC C

The interaction between the two perceptual indicators of thermal perceptions: overall comfort and thermal preference and TSV was studied under various thermal conditions. Thermal perceptions is a general term used to describe human thermal judgement in given thermal conditions and consists of three thermal scales mentioned above and thermal acceptance. Considering overall comfort in different seasons it was found that a large number of people (77.7%) in spring perceived the surrounding environment to be comfortable indicated by votes from “just right to very comfortable”, compared to other seasons when the percentage of votes for comfortable conditions were 45.9% and 37.8% for summer and autumn, respectively (Figure 5). In terms of thermal preference scale, the

14

ACCEPTED MANUSCRIPT number of votes requesting “no change”1 in the current thermal conditions was 54%.8 in spring and 51.3% in summer. The results of OLRM showed that the two indicators of thermal responses were statistically related to variation of TSV for aggregated data (Table 9) and therefore they were included in the overall

3.7 Weather forecast, thermal history and thermal sensations

RI PT

regression model of psychological environment (Table 10).

The participants’ thermal history (thermal experience) were also acquired using a multiple-choice

SC

question; the results were as follows: “indoor non-ventilated space” (N=98), “indoor-conditioned space” (N=536), “outdoor under shade” (N=200) and “outdoor exposed to sunlight” (N=185). The percentage of participants who have checked the weather forecast before leaving home (56.2%), was

M AN U

more than those who did not (43.8%). The OLRM applied to these two factors and the outcome was indicative of non-significant relationship in the period of study (Table 9); consequently, they were not taken into account for the overall regression model of psychological environment.

TE D

3.8 Place character, spatial features, naturalness and thermal sensations

The character and features of a place can encourage people to use the outdoor environment. This study asked the participants’ views about the most attractive features of the place through the following options: “plants and exposure to nature”, “an environment with a better ambient

EP

conditions”, “beauty of the place compared to other environments”, and “convenient access and closeness to my school/workplace”. Furthermore, the respondents’ opinion about establishment of

AC C

new natural green spaces was sought, 75% of people advocated (or did not disagree with) the establishment of more natural green spaces. The analytical results for these three factors could not provide a significant effect on TSV, except for sub-factor of “an environment with better ambient conditions” (ß=-0.25, P<0.05) as a feature that attracts people to visit the study open spaces (Table 9).

3.9 Seasonal change and thermal sensations The total number of thermal responses in the study seasons was 368, 413, and 240 in spring, summer, and autumn, respectively. As displayed in Figure 3 the pattern of survey users’ thermal sensations This study associated thermal sensations to the scale developed for the preference of an air temperature and not to other parameters of thermal conditions including (e.g. RH, Va and Tmrt).

1

15

ACCEPTED MANUSCRIPT varied with seasons. As indicated in 3.1 while the thermal votes in autumn were skewed towards the left side of scale (cool conditions), in spring and particularly in summer they were more inclined to its warm side. Change in thermal sensations with season was investigated using OLRM and it was found that seasonal change could significantly affect participants’ thermal sensations (Table 9); hence, this factor was also included in the overall regression model developed for the psychological environment.

RI PT

To further understand the level of association between the seasonal users’ thermal sensations, thermal conditions, and the characteristics thereof, a simple linear regression was fitted for the mean TSV for

seasons as follows: Equation 1: (Spring)

= .

(R2=0.86, P<0.001)

− .

= .

Equation 3: (Autumn)

= .

(R2=0.86, P<0.001)

− . − .



(R2=0.85, P<0.001)

M AN U

Equation 2: (Summer)

SC

each 2°C degree PET interval (Figure 6). As a result, three equations were developed for the study

By solving for zero in the equations above the seasonal neutral temperatures were determined to be 19.4 °C, 20.47°C and 25.10°C, for respectively spring, summer and autumn. Separate analysis was conducted to further understand the extent of contextual factors effect on the

TE D

relationship between thermal conditions and TSV in different seasons. The results revealed an interesting pattern of interaction between the study factors and TSV with respect to inter-seasonal change (Table 11). In the case of weather conditions (PET), the survey population in autumn were more affected (ß=0.28, P<0.01) than in summer (ß=0.23, P<0.05) and spring conditions (ß=0.20,

EP

P<0.01). While not significantly interacting with thermal sensations in summer, SVF showed a significant but positive and negative relationship with TSV in spring and autumn, respectively;

AC C

comparing these two later seasons, people’s TSV within the third level of SVF (34-42%) was found to be more influenced in spring (ß=-1.06, P<0.01) than in autumn (ß=0.97, P<0.05). The H/W ratio, was found to be significant in spring (ß=-0.89, P<0.01), but not in the other seasons. LoR was not a determinant factor in the aggregated data, however, in the seasonal analysis it was shown that this factor was statistically influential in spring (ß=1.05, P<0.05) and summer (ß=-0.98, P<0.01), with roughly similar weight of effect within the first level (>1-3 months of residence). Unlike in the aggregated data, ToE status was not a meaningful modifier of TSV variation in each season (P>0.05). As per PoV while the findings for aggregated data failed to demonstrate any meaningful relationship, “playing” in spring was found to be a determinant of TSV (ß=-0.89, P<0.05) (Table 11). Except for 16

ACCEPTED MANUSCRIPT “an environment with better ambient conditions” in summer time (ß=-0.43, P<0.05) which was also found to be influential factor within the aggregated data previously (ß=-0.25, P<0.05) none of other choices for place features were a significant factor in TSV. The findings of seasonal examination of TSV pattern in the study sites also indicated that the character of place was a determinant of thermal sensations in spring.

RI PT

Seasonal analysis was also applied to other people’s thermal responses (i.e. thermal preference and overall comfort) (Table 11). While having no impact on thermal sensations in summer (P>0.05), thermal preference and overall comfort contributed to variation of TSV in autumn. In spring, only thermal preference was found as a determinant of thermal sensations. The findings also suggested that people’s thermal preference played a more significant role on thermal sensations in autumn (P<0.05,

SC

ß=1.93) than in springtime (P<0.01, ß=1.65). The results of seasonal analysis for FoV, thermal history and the level of naturalness resembled the findings from their aggregated data and had no statistically

M AN U

significant influence.

4. Discussion:

4.1 Effect of design feature on microclimate conditions

TE D

Using the aggregated data, this study proved the impact of outdoor design features on the local microclimate conditions (Table 8). This finding corroborates the results of prior studies emphasising the importance of incorporating climate sensitive design principles in development plans of sustainable open spaces. It is also noteworthy to understand the design features influence was not the

EP

same in different seasons, necessitating the need to consider strategies to provide users with comfortable conditions throughout the year. For instance, as indicated by the results (Table 8), the strong correlation between H/W and wind speed suggests the use of wind shelters protecting people

AC C

from exposure to strong winds in widely open spaces in cool seasons with prevailing strong winds. In addition, according to the findings of correlation between SVF and Tg, construction of shading devices are also can be suggested as an effective design option. There exist a number of design strategies contributing to building a sustainable open space with thermally comfortable conditions for potential users: •

Provision of shades by means of trees or other man-made structures such as device shades, asymmetry, galleries and overhanging facades to reduce both air and surface temperature (Lin et al. 2011, Mahmoud 2011);

•

Use of cool urban surfaces also known as smart material surfaces to reduce surface temperature and UHI (Pomerantz et al. 2000); 17

ACCEPTED MANUSCRIPT •

Establishment of green spaces (i.e. on and above ground) improves thermal conditions by means of evapotranspiration, shading, photosynthesis and trapping long wave radiations (Kleerekoper et al. 2012);

•

Placing water features (Xu et al. 2010) and mist fans (Farnham et al. 2015) to cool down ambient air temperature for long term and temporarily, respectively.

•

Funnelling the air movement to direct air circulation into a tunnel-form structure and thus

RI PT

reduce convectional heat in hot conditions (Kato and Hiyama 2012, Niu et al. 2015, Hsieh and Huang 2016) •

Using urban heaters and wind shelters in small pockets of a space such as catering establishments allowing them to service for extended hours during the cool seasons

Simulations tools can be used to obtain an insight into the dynamic performance of the above-

SC

•

mentioned options for improving outdoor thermal comfort. Simulation tools such as ENVImet (Bruse and Fleer 1998), SOLWEIG (Lindberg et al. 2008) and Rayman Pro. (Matzarakis

M AN U

et al. 2007)allow space managers to visualise the quality of improvement in thermal conditions prior to their implementation.

TE D

4.2 Effect of physical environment on thermal sensations

The analysis revealed that physical parameters: SVF and ToE were the modifying factors of TSV over the study period (Table 5). The findings in the physical environment also indicated the strength of association as 54.1 % representing a medium influence on variation of TSV. Different types of temperatures (Ta, Tg and Tmrt) showed the largest correlation with users’ TSV (Table 7), which is in

EP

good agreement with the results of prior studies reporting the large effect of these parameters on thermal sensations (Lin et al. 2011, Lin et al. 2012, Villadiego and Velay-Dabat 2014, Chen et al.

AC C

2015, Middel et al. 2016). Besides its impact on thermal conditions, SVF could also determine people’s thermal evaluation and even usage pattern in different microclimate conditions. Therefore, this finding can raise the awareness of the importance of this factor in the design of a space among designers and planners and for its implication on the success of outdoor space. The significance of ToE in the determination of TSV implies that there is a need to spend a minimum time outdoors before people adapt to the surrounding thermal conditions (Höppe 2002). ToE is also associated to the psychological concept of Alliesthesia wherein the resultant immediate response to a given environmental stimulus creating a pleasure of displeasure is diminished with time of exposure (Krüger et al. 2015). In the case of people who spent only a short time at the study sites, they did not spend the minimum time needed to offset the resultant pleasure/displeasure; this per se explains the

18

ACCEPTED MANUSCRIPT statistical difference found among participants’ TSV with various time spent outdoor. Furthermore, the steady-state based indices tend to overestimate thermal comfort conditions when the human body is exposed to the outdoor microclimate conditions for only a short time and steady stated conditions cannot be achieved (Höppe 2002) To address the overestimation by steady state indices some efforts including the development of non-

RI PT

steady state (dynamic) indices (Fiala et al. 2001, Zolfaghari and Maerefat 2011) have been put forward to account for thermal comfort conditions during short visits (transient conditions). Alternatively, some researchers tended to exclude those participants who spent shorter than a certain time (e.g. under 5 mins) or keep participants outside for a set time (Ahmed 2003, Xi et al. 2012). However, this procedure in some ways is not realistic as in the real-world situation the length of stay

SC

varies considerably.

From the practical perspective since most of outdoor users are transient users (Aljawabra and

M AN U

Nikolopoulou 2010, Leech et al. 2002) providing thermally acceptable conditions is a challenging task. Therefore, it is suggested that spatial managers consider strategies to extend the length of visits including providing opportunities and facilities with users to enjoy their time outdoors. For instance, as indicated by a number of participants in this study, providing electric power in open spaces encourages people to stay longer than usual, as they are able to use their devices. Similarly there exists a number of strategies to maximise time spent outdoors in relation to a space including

TE D

diversifying the function, promoting use of the space, organising events, development of small spaces (pocket spaces) with multi-character, providing opportunities for outdoor activities, holding outdoor classes particularly in educational precincts, and a choice of comfortable sitting areas with desirable

EP

views, just to name a few.

AC C

4.3 Effect of psychological environment on thermal sensations

Psychological aspect of comfort is considered as a vital element to formation of outdoor users’ thermal perceptions (Nikolopoulou and Steemers 2003). Many studies have found the evidence on thermal adaptation among their study population and some of them have identified the mechanisms explaining the quality of adaptation occurrence including, physiological (De dear et al. 1997), psychological (Nikolopoulou and Steemers 2003, Knez et al. 2009) and behavioural (Cheng et al. 2009, Wu et al. 2015). Analysis of the psychological environment involved the investigation of the modifying role of nine factors on outdoor TSV (Table 9), which showed that out of these nine study factors only four including one sub-factor were statistically related to the variation of thermal sensations using aggregated data: overall comfort, thermal preference, seasonal change and the feature 19

ACCEPTED MANUSCRIPT place of “an environment with better ambient conditions”. However, accounting for these factors in the OLRM, the models ability to predict thermal responses improved by only 5.6%, indicating a low influence.

The evident relationship between overall comfort and thermal sensations was already reported in

RI PT

studies assessing outdoor thermal comfort (Cheng et al. 2012, Pantavou et al. 2013). Overall comfort may not be only related to weather conditions but to psychological factors by moderating thermal expectations. There is also a debate around the appropriateness of application of overall comfort instead of thermal sensation scale as it may better represent people’s satisfaction with thermal

SC

environment (Humphreys and Hancock 2007, Salata et al. 2016). Accordingly, the application of benchmark resulted from thermal sensation scale including comfort range and thermal neutrality that are enshrined in universal thermal comfort standards (ISO 7730 2006, ASHRAE 55 2010) is being

M AN U

questioned (Lai et al. 2014, Huang et al. 2016, Shooshtarian and Ridley 2016a). Hence , some have put forward methods to advance the accuracy of these results and to enhance the applicability of recommendations thereof (Zolfaghari and Maerefat 2011).

“An environment with a better ambient condition” was found to modify the relationship between TSV and thermal conditions (Table 9). This finding can be linked to the fact that users who were walking in outdoor from surrounding buildings in quest of experiencing a different environment

TE D

tended to consider outdoor conditions a better environment; this tendency in turn modified their expectations and therefore they indicated a different pattern of thermal sensations, particularly in summertime.

EP

Seasonal change is a critical factor in determination of human thermal perceptions. Besides impacting thermal sensations due to change in thermal conditions, seasonal change may psychologically affect people’s thermal judgement in three ways: by changing (1) people’s thermal expectations from

AC C

seasonal weather conditions (2) people’s thermal preference through the psychological concept of Alliesthesia in which people long for an opposite thermal conditions in response to current undesirable microclimates and (3) the influence pattern of other determinants of TSV. The effect of latter on study participants’ TSV is thoroughly discussed below and the findings were in good agreement with those of prior studies (Lin 2009, Cheng et al. 2012, Pantavou et al. 2013).

4.4 Effect of inter-seasonal change on the extent of influence of thermal sensations modifying factors

20

ACCEPTED MANUSCRIPT The further investigation of modifying effects of study factors on variation of TSV revealed a disparity between the results of aggregated dataset and seasonal datasets (Table 11). It was also noted that a statistically significant relationship between the study factors and TSV over the entire course of study did not similarly occur in all seasons. For instance, “playing” (purpose of visit) was a statistically influential factor on the variation of TSV in spring, whereas it was found to be insignificant using the aggregated dataset. This disparity was observed in the cases of SVF, H/W

RI PT

ratio, LoR, ToE, thermal preference, overall comfort, and character of place (Table 11) in which their influence on TSV changed with season. In summary, it can be stated that the requirements of thermal comfort differed, in both direction and magnitude, during different seasons. For instance, LoR and SVF were negatively related to TSV variation in spring and summer, but positively related in autumn

change in the two successive seasons of spring and summer.

SC

(Table 11). Furthermore, in the case of thermal preference the extent of influence showed a noticeable

M AN U

By scrutinizing the trend of change in interaction between study factors; and thermal conditions and responses in each season, it can be noted that inter-seasonal change not only influenced thermal conditions and thus thermal sensations (Figure 3) but also indirectly impacted those factors that were related to individuals’ physiology, psychology, emotions, expectations, adaptive strategies taken and the quality of weather assessment which determine their thermal perceptions. While it is possible to measure the impact of inter-seasonal change on the level of influence of certain factors on TSV (e.g. SVF and H/W), it may be quite difficult to identify the mechanism of such impact on that of others

TE D

(e.g. overall comfort). The reason for this may be attributed to other drivers or barriers that are not measurable. It is also a difficult task to grasp the full characteristics and performance of an individual’s psychology such as emotions and feelings (Johansson et al. 2014). For the same reasons mentioned above, in addition to reporting averaged annual thermal comfort requirements several

AC C

EP

studies have indicated comfort requirements for each season separately.

4.5 Evidence for occurrence of thermal adaptation

The occurrence of thermal adaptation in this study proved using the analytical findings on some contextual factors. As indicated by De dear and Brager (1998) thermal adaptation may take place in three forms: physiological, psychological and behavioural adaptation; using the data collected in research work, this study could provide evidence for occurrence of the two first forms. In the physical environment, ToE and LoR were the indicators of physiological thermal adaptation. The former occurred throughout the year (Table 5) and the latter only in summer and spring (Table 11), suggesting a short and long-term acclimatization to local thermal conditions, respectively. 21

ACCEPTED MANUSCRIPT The evidence for occurrence of psychological thermal adaptation was demonstrated by matching the findings with some of concepts of psychological adaptation introduced by Nikolopoulou and Steemers (2003) including thermal expectation, environmental stimulation and time of exposure. Seasonal change, overall comfort and thermal preference impacted thermal expectations throughout the year and therefore are linked to “thermal expectation” (Table 10), one feature of place (i.e. an ambient with better environment) in summertime is related to environmental stimulation, and ToE is associated

RI PT

with psychological concept of “time of exposure” despite being clustered under the physical environment in this study (Table 5). Also, varying seasonal neutral temperature indicated thermal adaptation in relation to seasonal outdoor thermal environment; these temperatures were also largely different from those reported in studies conducted in various climate conditions (Mahmoud 2011,

M AN U

acclimatization to local climate among the study population.

SC

Yahia and Johansson 2013, Salata et al. 2016, Kántor et al. 2016), verifying the existence of

5. Conclusions

Understanding different dimensions of outdoor thermal comfort conditions underpins the policies, standards, and decision-making processes being involved in creation of sustainable and successful

TE D

urban spaces. In order to fully understand the bigger picture of thermal comfort conditions in outdoor spaces it is necessary to investigate the role of contextual factors along with thermal factors in creation of users’ thermal perception. Therefore, this study aimed to explore the impact of such factors on determination of outdoor users’ thermal sensations using SESM as a research framework

EP

and the data extracted from the 1059 questionnaires that were collected in three successive seasons in an educational precinct in Melbourne (2014-2015). Specifically this paper discussed the role of factors clustered under the two SESM environments: physical and psychological. To achieve the

AC C

study aim the following objectives were set: (a) exploration of collective effect of physical and psychological factors on prediction of outdoor thermal sensations, (b) studying the impact of interseasonal change on study factors influence, and (c) assessment of outdoor users’ specific requirements of thermal comfort in the context of educational urban precincts. The findings in physical environment suggested that weather conditions, SVF and ToE were the major predictors of thermal sensations under various thermal conditions. In psychological environment out of eight study factors only three (i.e. seasonal change, overall comfort, thermal preference) and one sub-factor (place character: an environment with better ambient conditions) were found to be in statistical relationship with TSV variation throughout the year). As a result and according to the strength of association these two environments showed respectively a medium and low influence on outdoor thermal sensations. The results also suggested that the pattern of influence is dependent on study season and quite 22

ACCEPTED MANUSCRIPT different conditions may govern the quality and quantity of their modifying effect. This research also provides evidence of the occurrence of thermal adaptation to local climate conditions among outdoor users. Lastly, the research findings have shed light on those aspects of thermal comfort in outdoor spaces that are yet to be fully understood. It is expected that findings provide motivations for new development plans of urban spaces to consider the principles of climate sensitive designs to ensure

RI PT

developing a comfortable conditions outdoor.

Acknowledgement:

M AN U

SC

Authors would like to thank the RMIT Property Services for coordinating installation of measuring devices and facilitating conduction of field surveys in the University City Campus, Melbourne. Authors also would like to appreciate Dr. Hassan Doosti advice on statistical parts of this work.

AC C

EP

TE D

Ahmed, AQ, Ossen, DR, Jamei, E, Manaf, NA, Said, I & Ahmad, MH 2015, 'Urban surface temperature behaviour and heat island effect in a tropical planned city', Theoretical and Applied Climatology, 119(3-4), 493-514. Ahmed, KS 2003, 'Comfort in urban spaces: defining the boundaries of outdoor thermal comfort for the tropical urban environments', Energy and Buildings, 35(1), 103-110. Alcoforado, M-J, Andrade, H, Lopes, A & Vasconcelos, J 2009, 'Application of climatic guidelines to urban planning: the example of Lisbon (Portugal)', Landscape and Urban Planning, 90(1), 5665. Aljawabra, F & Nikolopoulou, M 2010, 'Influence of hot arid climate on the use of outdoor urban spaces and thermal comfort: Do cultural and social backgrounds matter?', Intelligent Buildings International, 2(3), 198-217. Andamon, MM 2005 'Building Climatology and Thermal Comfort'. PhD, University of Adelaide, Adelaide, 282 p. ASHRAE 55 2010. Thermal environmental conditions for human occupancy. Standard 55-2010. Atlanta, GA: American Society of Heating, Refrigerating and Air-conditioning Engineers. Bourbia, F & Boucheriba, F 2010, 'Impact of street design on urban microclimate for semi arid climate (Constantine)', Renewable Energy, 35(2), 343-347. Brager, G, Fountain, M, Benton, C, Arens, EA & Bauman, F 1993, 'A comparison of methods for assessing thermal sensation and acceptability in the field'. Brager, GS & de Dear, RJ 1998, 'Thermal adaptation in the built environment: a literature review', Energy and Buildings, 27(1), 83-96. Bronfenbrenner, U 1992 Ecological systems theory, Jessica Kingsley Publishers, p. Bronfenbrenner, U & Evans, GW 2000, 'Developmental Science in the 21st Century: Emerging Questions, Theoretical Models, Research Designs and Empirical Findings', Social Development, 9(1), 115-125. Bruse, M & Fleer, H 1998, 'Simulating surface–plant–air interactions inside urban environments with a three dimensional numerical model', Environmental Modelling & Software, 13(3–4), 373384. Cabanac, M 1971, 'Physiological role of pleasure', Science, 173(4002), 1103-1107. Chen, D, Wang, X, Khoo, Y, Thatcher, M, Lin, B, Ren, Z, Wang, C-H & Barnett, G 2013. Assessment of Urban Heat Island and Mitigation by Urban Green Coverage. In: KHARE, A & BECKMAN, T (eds.) Mitigating Climate Change. Springer Berlin Heidelberg, 247-257. 23

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Chen, L, Wen, Y, Zhang, L & Xiang, W-N 2015, 'Studies of thermal comfort and space use in an urban park square in cool and cold seasons in Shanghai', Building and Environment, 94, Part 2, 644653. Cheng, M-J, Lo, J-H & Li, C 2009, 'A field study of thermal comfort for the campus outdoor environment', J. Archit, 69, 1-16. Cheng, V, Ng, E, Chan, C & Givoni, B 2012, 'Outdoor thermal comfort study in a sub-tropical climate: a longitudinal study based in Hong Kong', International journal of biometeorology, 56(1), 4356. Chow, WT, Akbar, SNABA, Heng, SL & Roth, M 2016, 'Assessment of measured and perceived microclimates within a tropical urban forest', Urban Forestry & Urban Greening, 16, 62-75. Chun, C, Kwok, A, Mitamura, T, Miwa, N & Tamura, A 2008, 'Thermal diary: Connecting temperature history to indoor comfort', Building and Environment, 43(5), 877-885. Coutts, AM, Beringer, J & Tapper, NJ 2007, 'Impact of increasing urban density on local climate: spatial and temporal variations in the surface energy balance in Melbourne, Australia', Journal of Applied Meteorology and Climatology, 46(4), 477-493. Coutts, AM, Nigel, T, Beringer, J, Loughnan, M & Demuzere, M 2012, '“Watering our cities: The capacity for Water Sensitive Urban Design to support urban cooling and improve human thermal comfort in the Australian context', Progress in Physical Geography, 37(1), 2-28. De dear, R, Brager, G & Cooper, D 1997, ASHRAE RP-884 Final Report: Developing an Adaptive Model of Thermal Comfort and Preference., Macquarie Research Limited, Macquaire University and Center for Environmental Design Research, University of California Sydney, Australia and Berkely CA, USA. 297 p. De dear, R & Brager, GS 1998, 'Developing an adaptive model of thermal comfort and preference'. de Freitas, CR 1985, 'Assessment of human bioclimate based on thermal response', International Journal of Biometeorology, 29(2), 97-119. Djenane, M, Abdallah, F, Mohamed, B & Marjorie, M 2008. Microclimatic behaviour of urban forms in hot dry regions: towards a definition of adapted indicators. In: KENNY, P, V. BROPHY, J.O. LEWIS (ed.) PLEA 2008: the 25th Conference on Passive and Low Energy Architecture. Dublin. Eliasson, I 1994, 'Urban-suburban-rural air temperature differences related to street geometry', Physical geography, 15(1), 1-22. Eliasson, I 2000, 'The use of climate knowledge in urban planning', Landscape and urban planning, 48(1), 31-44. Eliasson, I, Knez, I, Westerberg, U, Thorsson, S & Lindberg, F 2007, 'Climate and behaviour in a Nordic city', Landscape and Urban Planning, 82(1–2), 72-84. Fanger, OP 1970 Thermal comfort. Analysis and applications in environmental engineering, Danish Technical Press, Copenhagen, 244 p. Farnham, C, Emura, K & Mizuno, T 2015, 'Evaluation of cooling effects: outdoor water mist fan', Building Research & Information, 43(3), 334-345. Fiala, D, Lomas, KJ & Stohrer, M 2001, 'Computer prediction of human thermoregulatory and temperature responses to a wide range of environmental conditions', International Journal of Biometeorology, 45(3), 143-159. Gagge, AP, Fobelets, AP & Berglund, LG 1986, 'A standard predictive index of human response to the thermal environment', ASHRAE Transactions, 92(2B), 709-731. Hatvani-Kovacs, G & Boland, J 2015, 'Retrofitting precincts for heatwave resilience: Challenges and barriers in Australian context', Challenges, 6(1), 3-25. He, X, Miao, S, Shen, S, Li, J, Zhang, B, Zhang, Z & Chen, X 2015, 'Influence of sky view factor on outdoor thermal environment and physiological equivalent temperature', International journal of biometeorology, 59(3), 285-297. Höppe, P 1999, 'The physiological equivalent temperature–a universal index for the biometeorological assessment of the thermal environment', International journal of Biometeorology, 43(2), 71-75. 24

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Höppe, P 2002, 'Different aspects of assessing indoor and outdoor thermal comfort', Energy and Buildings, 34(6), 661-665. Höppe, PR & Seidl, HA 1991, 'Problems in the assessment of the bioclimate for vacationists at the seaside', International Journal of Biometeorology, 35(2), 107-110. Hsieh, C-M & Huang, H-C 2016, 'Mitigating urban heat islands: A method to identify potential wind corridor for cooling and ventilation', Computers, Environment and Urban Systems, 57, 130143. Huang, J, Zhou, C, Zhuo, Y, Xu, L & Jiang, Y 2016, 'Outdoor thermal environments and activities in open space: An experiment study in humid subtropical climates', Building and Environment, 103, 238-249. Huang, K-T, Lin, T-P & Lien, H-C 2015, 'Investigating Thermal Comfort and User Behaviors in Outdoor Spaces: A Seasonal and Spatial Perspective', Advances in Meteorology, 2015, 1. Humphreys, M 1995, 'Thermal comfort temperatures and the habits of Hobbits', Standards for Thermal Comfort: Indoor Air Temperature Standards for the 21st Century, 3-13. Humphreys, MA & Hancock, M 2007, 'Do people like to feel ‘neutral’?: Exploring the variation of the desired thermal sensation on the ASHRAE scale', Energy and Buildings, 39(7), 867-874. Humphreys, MA & Nicol, JF 1998, 'Understanding the adaptive approach to thermal comfort', ASHRAE transactions, (104 (1b)), 991-1004. Hwang, R-L & Lin, T-P 2011, 'Thermal Comfort Requirements for Occupants of Semi-Outdoor and Outdoor Environments in Hot-Humid Regions', Architectural Science Review, 50(4), 357-364. Hwang, R-L, Lin, T-P & Matzarakis, A 2011, 'Seasonal effects of urban street shading on long-term outdoor thermal comfort', Building and Environment, 46(4), 863-870. ISO 7730 2006. Moderate Thermal Environments- Determination of the PMV and PPD Indices and Specifications of the Conditions for Thermal Comfort.: Geneva: International Organization for Standardization (ISO). ISO 10551 1995. Ergonomics of the thermal environment—assessment of the influence of the thermal environment using subjective judgement scales. Geneva, CH: International Organization for Standardization, Geneva. Johansson, E 2006, 'Influence of urban geometry on outdoor thermal comfort in a hot dry climate: A study in Fez, Morocco', Building and Environment, 41(10), 1326-1338. Johansson, E, Thorsson, S, Emmanuel, R & Krüger, E 2014, 'Instruments and methods in outdoor thermal comfort studies–The need for standardization', Urban Climate, 10(2), 346-366. Kántor, N, Kovács, A & Takács, Á 2016, 'Seasonal differences in the subjective assessment of outdoor thermal conditions and the impact of analysis techniques on the obtained results', International journal of biometeorology, 1-21. Katavoutas, G, Flocas, HA & Matzarakis, A 2015, 'Dynamic modeling of human thermal comfort after the transition from an indoor to an outdoor hot environment', International journal of biometeorology, 59(2), 205-216. Kato, S & Hiyama, K 2012 Ventilating cities: air-flow criteria for healthy and comfortable urban living, Springer Science & Business Media, p. Kenawy, I & Elkadi, H 2011. THERMAL COMFORT ADAPTATION IN OUTDOOR PLACES. Proceedings of the 2011 International Conference of the Association of Architecture Schools of Australasia, 18-21 Sep. Melbourne, Australia: Deakin University. Kleerekoper, L, van Esch, M & Salcedo, TB 2012, 'How to make a city climate-proof, addressing the urban heat island effect', Resources, Conservation and Recycling, 64, 30-38. 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. Krüger, E, CA Tamura , M Schweiker, A Wagner & Bröde, P 2015. Short-term acclimatization effects in an outdoor comfort study. ICUC9: the 9th International Conference on Urban Climate

25

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

jointly with 12th Symposium on the Urban Environment. France: The International Association for Urban Climate Krüger, EL, Minella, FO & Rasia, F 2011, 'Impact of urban geometry on outdoor thermal comfort and air quality from field measurements in Curitiba, Brazil', Building and Environment, 46(3), 621-634. Krüger, EL & Rossi, FA 2011, 'Effect of personal and microclimatic variables on observed thermal sensation from a field study in southern Brazil', Building and environment, 46(3), 690-697. Lai, D, Zhou, C, Huang, J, Jiang, Y, Long, Z & Chen, Q 2014, 'Outdoor space quality: A field study in an urban residential community in central China', Energy and Buildings, 68, 713-720. Lam, CKC, Loughnan, M & Tapper, N 2016, 'Visitors’ perception of thermal comfort during extreme heat events at the Royal Botanic Garden Melbourne', International journal of biometeorology, 1-16. Leech, JA, Nelson, WC, Burnett, RT, Aaron, S & Raizenne, ME 2002, 'It's about time: A comparison of Canadian and American time–activity patterns', Journal of Exposure Analysis & Environmental Epidemiology, 12(6), 427-432. Leviston, Z, Greenhill, M & Walker, I 2015 Australian Attitudes to Climate Change: 2010-2014, CSIRO, p. Lin, T-P 2009, 'Thermal perception, adaptation and attendance in a public square in hot and humid regions', Building and Environment, 44(10), 2017-2026. Lin, T-P, Matzarakis, A & Hwang, R-L 2010, 'Shading effect on long-term outdoor thermal comfort', Building and Environment, 45(1), 213-221. Lin, T-P, Tsai, K-T, Hwang, R-L & Matzarakis, A 2012, 'Quantification of the effect of thermal indices and sky view factor on park attendance', Landscape and Urban Planning, 107(2), 137-146. Lin, TP, de Dear, R & Hwang, RL 2011, 'Effect of thermal adaptation on seasonal outdoor thermal comfort', International Journal of Climatology, 31(2), 302-312. Lindberg, F, Holmer, B & Thorsson, S 2008, 'SOLWEIG 1.0–Modelling spatial variations of 3D radiant fluxes and mean radiant temperature in complex urban settings', International Journal of Biometeorology, 52(7), 697-713. Mahmoud, AHA 2011, 'Analysis of the microclimatic and human comfort conditions in an urban park in hot and arid regions', Building and Environment, 46(12), 2641-2656. Makaremi, N, Salleh, E, Jaafar, MZ & GhaffarianHoseini, A 2012, 'Thermal comfort conditions of shaded outdoor spaces in hot and humid climate of Malaysia', Building and environment, 48, 7-14. Matzarakis, A, Rutz, F & Mayer, H 2007, 'Modelling radiation fluxes in simple and complex environments—application of the RayMan model', International Journal of Biometeorology, 51(4), 323-334. Mayer, H & Höppe, P 1987, 'Thermal comfort of man in different urban environments', Theoretical and Applied Climatology, 38(1), 43-49. Middel, A, Selover, N, Hagen, B & Chhetri, N 2016, 'Impact of shade on outdoor thermal comfort—a seasonal field study in Tempe, Arizona', International journal of biometeorology, 1-13. Morris, C & Simmonds, I 2000, 'Associations between varying magnitudes of the urban heat island and the synoptic climatology in Melbourne, Australia', International Journal of Climatology, 20(15), 1931-1954. Nikolopoulou, M 2004, 'Outdoor comfort', Environmental diversity in architecture, 101. Nikolopoulou, M & Lykoudis, S 2007, 'Use of outdoor spaces and microclimate in a Mediterranean urban area', Building and Environment, 42(10), 3691-3707. Nikolopoulou, M & Steemers, K 2003, 'Thermal comfort and psychological adaptation as a guide for designing urban spaces', Energy and Buildings, 35(1), 95-101. Niu, J, Liu, J, Lee, T-c, Lin, ZJ, Mak, C, Tse, K-T, Tang, B-s & Kwok, KC 2015, 'A new method to assess spatial variations of outdoor thermal comfort: onsite monitoring results and implications for precinct planning', Building and Environment, 91, 263-270. 26

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Norušis, MJ 2012. Ordinal Regression In: NORUŠIS, MJ (ed.) IBM SPSS Statistics 19 Guide to Data Analysis. Addison Wesley, 69-89. Oke, TR 1982, 'The energetic basis of the urban heat island', Quarterly Journal of the Royal Meteorological Society, 108(455), 1-24. Oliveira, S & Andrade, H 2007, 'An initial assessment of the bioclimatic comfort in an outdoor public space in Lisbon', International journal of biometeorology, 52(1), 69-84. Pantavou, K, Theoharatos, G, Santamouris, M & Asimakopoulos, D 2013, 'Outdoor thermal sensation of pedestrians in a Mediterranean climate and a comparison with UTCI', Building and Environment, 66(0), 82-95. Peel, MC, Finlayson, BL & McMahon, TA 2007, 'Updated world map of the Köppen-Geiger climate classification', Hydrology and Earth System Sciences, 11(5), 1633-1644. Pomerantz, M, Brian, P, Hashem, A & Sheng-Chieh, C 2000, The effect of pavements' temperatures on air temperatures in large cities, Berkeley, CA. p. Qaid, A & Ossen, DR 2014, 'Effect of asymmetrical street aspect ratios on microclimates in hot, humid regions', International journal of biometeorology, 1-21. Salata, F, Golasi, I, de Lieto Vollaro, R & de Lieto Vollaro, A 2016, 'Outdoor thermal comfort in the Mediterranean area. A transversal study in Rome, Italy', Building and Environment, 96, 4661. Shooshtarian, S, Iyer-Raniga, U, Andamon, MM & Ridley, I 2015. Thermal perceptions and microclimates of educational urban precincts in two different seasons in Melbourne. In: R.H., C & STEPHAN, A (eds.) Living and Learning: Research for a Better Built Environment: the 49th International Conference of the Architectural Science Association. Melbourne, Australia Architectural Science Association, 1194-1202. Shooshtarian, S & Ridley, I 2016a, 'Determination of acceptable thermal range in outdoor built environments by various methods', Smart and Sustainable Built Environment, 5(4), 352-371. Shooshtarian, S & Ridley, I 2016b, 'The effect of individual and social environments on the users thermal perceptions of educational urban precincts', Sustainable Cities and Society (26), 119113. Spagnolo, J & De Dear, R 2003, 'A field study of thermal comfort in outdoor and semi-outdoor environments in subtropical Sydney Australia', Building and Environment, 38(5), 721-738. Spangenberg, J, Paula Shinzato, Johansson, E & Duarte, D 2008, 'Simulation of the influence of vegetation on microclimate and thermal comfort in the city of São Paulo', Journal of Brazilian Society of Urban Forest, 3(2), 1-19. Thorsson, S, Lindberg, F, Eliasson, I & Holmer, B 2007, 'Different methods for estimating the mean radiant temperature in an outdoor urban setting', International Journal of Climatology, 27(14), 1983-1993. Torok, S, J., Morris, CJG, Skinner, C & Plummer, N 2001, 'Urban heat island features of southeast Australian towns', Australian Meteorological Magazine, 50, 1-13. Villadiego, K & Velay-Dabat, MA 2014, 'Outdoor thermal comfort in a hot and humid climate of Colombia: a field study in Barranquilla', Building and Environment, 75, 142-152. Wilson, E, Nicol, F, Nanayakkara, L & Ueberjahn-Tritta, A 2007, 'Public Urban Open Space and Human Thermal Comfort: The Implications of Alternative Climate Change and Socio-economic Scenarios', Journal of Environmental Policy & Planning, 10(1), 31-45. Wu, C-F, Hsieh, Y-F & Ou, S-J 2015, 'Thermal Adaptation Methods of Urban Plaza Users in Asia’s HotHumid Regions: A Taiwan Case Study', International journal of environmental research and public health, 12(10), 13560-13586. Xi, T, Li, Q, Mochida, A & Meng, Q 2012, 'Study on the outdoor thermal environment and thermal comfort around campus clusters in subtropical urban areas', Building and Environment, 52(0), 162-170. Xu, J, Wei, Q, Huang, X, Zhu, X & Li, G 2010, 'Evaluation of human thermal comfort near urban waterbody during summer', Building and Environment, 45(4), 1072-1080. 27

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Yahia, MW & Johansson, E 2013, 'Evaluating the behaviour of different thermal indices by investigating various outdoor urban environments in the hot dry city of Damascus, Syria', International journal of biometeorology, 57(4), 615-630. Yin, J, Zheng, Y, Wu, R, Tan, J, Ye, D & Wang, W 2012, 'An analysis of influential factors on outdoor thermal comfort in summer', International journal of biometeorology, 56(5), 941-948. Zacharias, J, Stathopoulos, T & Wu, H 2004, 'Spatial Behavior in San Francisco’s Plazas The Effects of Microclimate, Other People, and Environmental Design', Environment and behavior, 36(5), 638-658. Zhang, Y & Zhao, R 2008, 'Overall thermal sensation, acceptability and comfort', Building and environment, 43(1), 44-50. Zolfaghari, A & Maerefat, M 2011, 'A new predictive index for evaluating both thermal sensation and thermal response of the human body', Building and Environment, 46(4), 855-862.

28

ACCEPTED MANUSCRIPT

Table 1. The summary of studies investigated the thermal conditions in relation to urban design. H/W ratio

Asymmetrical streets provide better thermal comfort via enhancing wind flow and blocking solar radiation. Aspect ratios between 0.8 and 2 ensure noticeable reduction in air and surface temperature in tropical regions.

Qaid and (2014)

Ossen

Cairo

SVF, albedo

Due to their shading and sheltering effect against intense solar radiation and wind patterns, tree-planted areas can provide better thermal comfort conditions in urban parks located in arid regions, respectively in summer and winter

Mahmoud (2011)

Huwei Township

SVF

Highly shaded outdoor spaces, featuring lower SVF values, are more beneficial in hot and humid climates in summer, spring and autumn.

Hwang et al. (2011)

Curitiba

SVF, street orientation, wind flow

While the similar comfort conditions were observed on days with low temperature in all study locations, on hot days areas with lower SVF provided better thermal comfort

Huwei Township

SVF

Barely shaded areas (high SVF percentages) had longer hours of discomfort in summer and more; whereas densely shaded-areas were more dis-comfortable in wintertime.

Lin et al. (2010)

Constantine City

SVF, H/W, orientation

street

With some few exceptions SVF and H/W ratio values had respectively positive and negative correlation with air and surface temperature

Bourbia and Boucheriba (2010)

Beni Isguen city

SVF, H/W ratio, plot ratio

The dependency of thermal behaviour on both solar exposure and magnitudes of wind velocity

Djenane (2008)

Fez

H/W ratio, SVF

Spaces with higher percentages of SVF had comparatively higher nocturnal air temperature due to release of heat stored by surfaces during the day caused by incoming solar radiations.

Johansson (2006)

EP

TE D

M AN U

SC

RI PT

Putrajaya

AC C

Table 2. Summary of the monthly mean climate conditions of the study sites. November December January February Ambient temperature (ºC) 17.2 18.6 20.3 20.9

Krüger et al. (2011)

et

al.

March 17.6

April 14.6

May 13.2

Maximum ambient temperature (ºC) Minimum ambient temperature (ºC) Relative humidity (%)

17.9

19.3

21

21.5

18.2

15.1

13.7

16.6

18.0

19.7

20.3

17

14.1

12.7

61.9

68.2

64.5

37.6

63.3

73.1

71.1

Wind velocity (m s-1)

2.8

2.8

3.0

2.6

2.5

2.2

3

Monthly mean global exposure (M.J m-2)

22.2

24.2

24.6

21.4

16.7

11.3

7.6

Source: Bureau of Meteorology (BOM) Olympic park station (2014-2015).

Table 3. The date of field surveys in different seasons. Season

1

2

3

ACCEPTED MANUSCRIPT Site 03.11.2014, 06.11.2014 21.11.2014

05.11.2017, 07.11.2014 25.11.2014

10.11.2014, 11.11.2014 26.11.2014

Summer

09.2.2015, 12.02.2015 18.02.2015

10.02.2015, 13.02.2015 20.02.2015

11.02.2015, 16.02.2015 23.02.2015

Autumn

05.05.2015, 08.05.205 15.05.2015

06.05.2015, 12.05.2015 18.05.2015

07.05.2015, 11.05.2015 20.05.2015

RI PT

Spring

Table 4 Fish-eye images taken in the three study sites and the associated SVF values.

M AN U

SC

Site 1

Sky view factor: 45.2% Horizon limitation: 54.8

Sky view factor: 45.8% Horizon limitation: 54.2%

TE D

Site 2

Sky view factor: 36.6% Horizon limitation: 63.4%

Sky view factor: 20.6% Horizon limitation: 79.3%

AC C

EP

Site 3

Sky view factor: 33.6% Horizon limitation: 66.4%

Sky view factor: 29.50% Horizon limitation: 70.5%

Sky view factor: 36.6% Horizon limitation: 63.3%

Sky view factor: 45.3% Horizon limitation: 54.7%

Sky view factor: 31.4% Horizon limitation: 68.5%

Table 5. The summary of the logistic regression model for the physical environment.

Thermal sensation vote Threshold

Est.

Std. error

sig

N

Cold= -3 Cool= -2 Slightly cool= -1 Neutral= 0 Slightly warm= +1 Warm= +2 Hot= +3 Location

1.392 2.994 4.997 6.026 7.464 10.126

.237 .213 .237 .257 .291 .379

.00 .00 .00 .00 .00 .00

45 102 262 157 200 48

PET Solar radiation intensity

0.269 0.001

.011 .000

.00 .00

1023 1023

Sky view factor (SVF) SVF=18-26% SVF=26-34% SVF=34-42% SVF=42-50%

-.278 -.183 -.372 0a

.354 .169 .189 .

.432 .276 .050 .

RI PT

ACCEPTED MANUSCRIPT

.162 .170 .

.417 .085 .

-.301 0a

.117 .

.010 .

Length of residence in Melbourne (LoR) >1-3 months (1) >3-12 months (2) >1-3 years (3) 3-10 years (4) >10 years (5)

-.039 .050 -.203 -.045 0a

SC

.131 -.292 0

285 295 234 414 606

M AN U

Aspect Ratio (H/W)* SITE 1(1.80) SITE 2 (2.00) SITE 3 (2.82) Time of exposure (ToE) Below 5 mins Above 5 mins

33 566 274 148

.198 .232 .175 .155 .

.844 .828 .245 .772 .

102 71 140 192 500

TE D

Link function: Logit. 0a: This parameter is set to zero because it is redundant (SPSS Software Package 22 assumes the last category as the reference index and sets it to zero to be compared with other categories and therefore the statistics will not be shown for the reference category).

Table 6. The summary of the overall logistic regression model for the physical environment.

AC C

Threshold Cold= -3 Cool= -2 Slightly cool= -1 Neutral= 0 Slightly warm= +1 Warm= +2 Location

Est.

Std. error

sig

1.074 2.703 4.795 5.893 7.375 9.995

.285 .264 .284 .301 .331 .408

.00 .00 .00 .00 .00 .00

EP

Thermal sensation vote

PET Solar radiation intensity Sky view factor (SVF) SVF=18-26% SVF=26-34% SVF=34-42% SVF=42-50% ToE Below 5 mins Above 5 mins

.249 .001

.012 .00

0.00 0.00

.017 -.095 -.192 0

.359 .170 .192 .

.962 .575 .317 .

-.306 0

.118 .

.009 .

Pseudo-R2

51.6 53.3 53.4

54.1

ACCEPTED MANUSCRIPT

Table 7. The summary of Spearman’s correlation analysis for relationship between the environmental parameters and TSV.

TSV (Correlation coefficient, r) Pooled .72** .70** ns -.03 -.34** .65** .46**

Ta Tg Va RH Tmrt SR

Spring .60** .48** ns -.09 -.47** .34** .26**

Summer .60** .63** ns -.09 -.46** .58** .34**

Autumn .41** .43** -.27** -.14** .40* ns .08

RI PT

Physical environment

Level of significance: ** significant at the 0.01 level , *significant at the 0.05 level, ns: non- significant

Table 8. Results of correlation between design descriptors and the environmental parameters.

Ta

Tg

RH

Spring Summer Autumn Pooled

.33** .48** .20** .16**

.05 ns .40** .28** .15**

Spring Summer Autumn Pooled

-.36** -.08 -.32** -.18**

Sky view factor

Va

SR

-.23** -.17** -.26** -.17**

.03* .10* .18** .13**

.07ns -.09 .52** .09**

-.12** -.26** -.37** -.20**

SC

Season

M AN U

-.27** -.20** -.20** -.24**

Aspect ratio

-.18** -.18** -.52** -.20**

.47** .26** .58** .31**

Level of significance: ** significant at the 0.01 level, *significant at the 0.05 level, ns: non- significant

Threshold Cold= -3 Cool= -2 Slightly cool= -1 Neutral= 0 Slightly warm= +1 Warm= +2 Location PET

Est.

Std. error

sig

N

1.392 2.994 4.997 6.026 7.464 10.126

.237 .213 .237 .257 .291 .379

.00 .00 .00 .00 .00 .00

45 102 262 157 200 207

-.080 -.143 0

.172 .177 .

.643 .421 .

497 379 147

-.184 0

.136 .

.252 .

524 497

.057 0

.138 .

.654

745 276

.897 1.198

.227 .

0 -.053

.162 .

EP

Thermal sensation vote

TE D

Table 9. The summary of the logistic regression model for the psychological environment.

AC C

Frequency of visit (FoV) Daily to several time a week =1 Few times a week to few times a month =2 Rarely to first time =3

Purpose of visit (PoV) Having break/resting/change of environment a (chosen) b (not-chosen) Getting fresh air a (chosen) b (not-chosen) Playing a (chosen) b (not-chosen) Passage to another place a (chosen) b (not-chosen)

. .265 . .964

947 74 4 790

ACCEPTED MANUSCRIPT Having launch/snack a (chosen) b (not-chosen) Read/write a (chosen) b (not-chosen) Meeting/waiting for someone a (chosen) b (not-chosen)

0 -.086

.144 .

. .676 .

231 788

0 .309

.251 .

.174 .

233 964

0 .021

.173 .

.742

57 886

1.509 1.325 0

.200 .213 .

.00 .00 .

368 413 240

1.102 .993 0

.185 .145 .

.00 .00 .

.023 -1.047 -.775 -.258 -.138 -.349 0

.321 .253 .199 .181 .219 .171 .

.942 .00 .00 .154 .528 .041 .

Thermal history Indoor, non-ventilated= Indoor, conditioned= Outdoor, under shade= Outdoor, under sun=

1 2 3 4

-.315 -.149 -.199 0

.229 .156 .187 .

.170 .340 .289 .

98 536 200 185

Place character Site 1 Site 2 Site 3

1 2 3

.136 -.277 0

.012 .143 .147

.341 .059 .

315 366 340

-.104 0

.118

.378

429 592

-.250 0

.123

.042

689 332

.022 0

.120

.854

.075 0

.115

.067 0

.133 .

Naturalness Advocating/not disagreeing with more natural green spaces= 1 Disagreeing more natural green spaces= 2

211 486 322

SC

38 72 147 197 105 248 214

M AN U

TE D

AC C

EP

Feature of place Plants and exposure to nature a (chosen) b (not-chosen) An environment with a better ambient conditions a(chosen) b (not-chosen) Beauty of the place compared to the other environments a (chosen) b (not-chosen) Convenient access and closeness to my school/workplace a (chosen) b (not-chosen)

RI PT

Seasonal change Spring= 1 Summer= 2 Autumn= 3 Thermal preference Cooler= 1 No change= 2 Warmer= 3 Overall comfort Very uncomfortable= 1 Moderately uncomfortable= 2 Slightly uncomfortable= 3 Just right= 4 Slightly comfortable= 5 Moderately comfortable= 6 Very comfortable= 7

647 374

.514

.614 .

534 487

763 253

Table 10. The summary of the overall logistic regression model for the psychological environment

Thermal sensation vote Threshold

Est.

Std. error

sig

Pseudo-R2

ACCEPTED MANUSCRIPT Location

0.22

.013

.00

1.248 .946 0

.209 .221 .

.00 .00 .

1.030 .827 0

.193 .150 .

.00 .00 .

.296 -.643 -.293 -.063 .150 -.198 0

.333 .266 .212 .188 .223 .173 .

.374 .015 .166 .736 .501 .254 .

51.6

PET Seasonal change Spring= 1 Summer= 2 Autumn= 3 Thermal preference Cooler= 1 No change= 2 Warmer= 3 Overall comfort Very uncomfortable= 1 Moderately uncomfortable= 2 Slightly uncomfortable= 3 Just right= 4 Slightly comfortable= 5 Moderately comfortable= 6 Very comfortable= 7 Feature of place An environment with a better ambient conditions a(chosen) b (not-chosen)

54.6

RI PT

56.5

SC

57.0

57.2

.123 .

.067 .

M AN U

-.226 0

Table 11. Summary of statistics of inter-seasonal influence of study factors on thermal sensation.

Season

Spring

levels

LoR

>1-3 months= >3-12 months= >1-3 years= 3-10 years= >10 years=

ToE

Below 5 mins Above 5 mins

TE D

H/W

SITE 1: (1.80) SITE 2: (2.00) SITE 3: (2.82)

1 2 3 4

EP

SVF

SVF=18-26%= SVF=26-34%= SVF=34-42%= SVF=42-50%=

Sign.

Physical Environment .202 .00

AC C

PET

Continuous

Est.

Summer

FoV

Daily to several time a week = Few times a week to few times a month= Rarely to first time=

1 2 3 4 5

Autumn

Est.

Sign.

Est.

Sign.

.237

.019

.287

.00

-.272 -.422 -1.062 0

.668 .101 .00 .

-.276 -.151 .148 0

.611 .626 .684 .

.647 .388 .907 0

.402 .310 .029 .

.208 -.891 0

.441 .001 . .030 .009 .770 .046 .

.418 .136 . .007 .449 .252 .487 .

-.392 .297 0

1.053 -.879 -.106 -.465 0

.206 .487 0 -.986 -.321 .578 -.168 0

.718 .746 .253 .358 0

.367 .418 . .449 .094 .706 .226 .

-.364 0

.064 .

-.226 0

.222 .

.703 .

-.092 0

Psychological Environment 1

.028

.920

-.325

.228

-.201

.387

2 3

.166 0

.574 .

-.368 0

.194 .

-.269 0

.254 .

ACCEPTED MANUSCRIPT -.228 0

.303

.094 0

.738

.

-.037 0

.867 .

.247 0

.268 .

-.324 0

.278 .

-1.304 0

.029 .

-.259 0

.421 .

-.308 0

.449 .

.032 0

.911 .

-.155 0

.551 .

-.394 0

.230 .

-.165 0

.518 .

-.139 0

.536 .

.703 .

-.174 0

.707 .

.564 0

.119 0 -.187 0

-.237 0

.468 .

.072 0

.317 -.660 0

.196 .007 .

-.360 0

RI PT

.051 .

.178 . .789 .

-.017 0

.

.681 . .958 .

-.333 .418 0

.347 .158 .

.066 .

.022 0

.907 .

.166 0

.502 .

-.093 0

.651 .

-.436 0

.025 .

-.327 0

.200 .

-.315 0

.110 .

.102 0

.593 .

.330 0

.199 .

.149 0

.438 .

-.038 0

.836 .

-.074 0

.762 .

Cooler= No change= Warmer=

1 2 3

1.655 1.560 0

.00 .00 .

.241 .047 0

.428 .870 .

1.930 .481 0

.023 .074 .

Overall comfort

Very uncomfortable= Moderately uncomfortable= Slightly uncomfortable= Just right= Slightly comfortable= Moderately comfortable= Very comfortable=

1 2 3 4 5 6 7

n/a .597 -.697 -.291 -.404 -.292 0

n/a .526 .072 .328 .230 .209 .

.640 .531 .059 .129 .479 -.090 0

.117 .168 .858 .668 .204 .767 .

-.170 -2.032 -.725 .209 .313 .053 0

.833 .001 .162 .691 .587 .920 .

Indoor, non-ventilated= Indoor, conditioned= Outdoor, under shade= Outdoor, under sun=

1 2 3 4

-.308 -.086 -.013 0

.374 .726 .969 .

-.387 -.066 .005 0

.330 .786 .986 .

-.218 -.108 -.107 0

.658 .771 .785 .

Advocating more natural green spaces= 1 Disagreeing more natural green spaces= 2

.218 0

.334 .

.260 0

.198 .

-.317 0

.266 .

AC C

EP

TE D

Plants and exposure to nature a (chosen) b (not-chosen) An environment with a better ambient conditions a(chosen) b (not-chosen) Beauty of the place compared to the other environments a (chosen) b (not-chosen) Convenient access and closeness to my school/workplace a (chosen) b (not-chosen)

M AN U

.779 .517 .

Place feature

.063 .177 0

Tpref

Place character

Site 1 Site 2 Site 3

-.450 0

SC

PoV

Having break/resting/change of environment a (chosen) b (not-chosen) Getting fresh air a (chosen) b (not-chosen) Playing a (chosen) b (not-chosen) Passage to another place a (chosen) b (not-chosen) Having launch/snack a (chosen) b (not-chosen) Read/write a (chosen) b (not-chosen) Meeting/waiting for someone a (chosen) b (not-chosen)

Thermal history Naturalness

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Figure 1. The geographical location of study educational urban precinct. Source: Google earth Pro 2016

AC C

EP

TE D

Site 1 Aspect Ratio: 1.80

Site 2 Aspect Ratio: 2.00

ACCEPTED MANUSCRIPT

RI PT

Site 3 Aspect Ratio: 2.80

39.5

Spring

40

25.0 26.9

1.4

2.5

TE D

0

2.4

AC C

EP

Figure 3. The characteristics of people’s thermal sensation in three study seasons.

Figure 4. The characteristics of the purpose of visit across the study sites.

7.3 4.9

5.2

10 5

19.6

7.6

15

0.0

20

17.7

17.8

25

Autumn

15.5

25.5

23.4 25.9

31.2

30

M AN U

Summer

35

0.5 0.2

Frequency of thermal sensation votes (%)

45

SC

Figure 2. Study sites and the associated aspect ratio.

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

Figure 5. Pattern of the participants’ overall comfort (left) and thermal preference (right) in the various seasons.

Figure 6. Association of the users’ thermal sensation to environmental parameters expressed in PET in various seasons.

ACCEPTED MANUSCRIPT Highlights for this research: 1. Using socio-ecological system model, this study has investigated and reported the role of thermal and non-thermal factors in creation of thermal sensation for outdoor users of educational precincts under psychological and physical environments. 2. The results also suggested that the pattern of influence is dependent on study season and quite different conditions may govern the quality and quantity of their modifying effect.

AC C

EP

TE D

M AN U

SC

RI PT

3. Study has provided the evidence for occurrence of thermal adaptation among survey population analysing effect of psychological and physical factors on people’s thermal perceptions.