Is there an air temperature variation inside semi-enclosed spaces dependent on their size? /
The smaller the semi-enclosed space is the higher the air temperature will be. /
Candidate Number: 1400342 /
11/25/2013 /
Content page
P1- Introduction
P2- Methodology
P4-Judging our methodology: Limitations and Observations
P5- Results
P7- Analysis
P8- Conclusions
P9- References
P10- Appendices
List of Figures:
-Figure 1-Map of RHUL Campus, page 2
-Figure 2: Founders North and Founders West Quads entrances, page 4
- Figure 3: Table displaying results collected, page 5
-Figure 4: Air temperature change with Time, page 5
-Figure 5: Mean temperature against the size of the Semi-enclosed areas, page 6
-Figure 6: Air temperature against Data sample number, page 6
-Figure 7: Recorded Weather conditions at each site during the time air temperature was recorded, page 7
-Figure 8: Wedderburn’s residual heat, page 8
Introduction
This paper presents a study that aims to understand the influence that the built environment has on air temperature. These are known as urban micro-climatic variations (Peng et al 2012). This type of urban climatic control can be defined as theshifts in climatic conditions in smallareaswhich contrast with the larger climatic regime of the region. An example of this condition is found in Hong Kong; known to be an “urban heat island”, just like many other densely built and populated cities of the world (WONG, PPY et al). This is a climatic variation between a city and the directly surrounding area; however, there are also differences at a much smaller scale, such as the ones this paper proposes. These variations may include changes in wind strength or temperature caused by specific features of the urban morphology (Shishegar, 2013). Firstly, these can betypes of urban material which absorb more or less heat, for example fire brick will absorb and release more heat than concrete. Secondly,small distances between tall buildings can create wind tunnels, increasing wind speeds by “squeezing” air (McGill 1983). Other urban related micro-climatic variations areareas of absolute shade and areas of increased heat exposure through reflected sunlight (Smith-Spark, 2013). This paper studies air temperature in different sized semi-enclosed spaces, its purpose is to understand the relationship; if at all there is one, between these two variables. We understand a semi-enclosed space to be a partially sheltered area with no roof or/and with gaps between the walls. Our hypothesis is:
The smaller the semi-enclosed space is the higher the air temperature will be.
Whilst the null hypothesis is:
There is no correlation between the size of semi-enclosed spaces and air temperature.
Our hypothesis was based upon the basis that as buildings emit heat we can expect to find small semi-enclosed spaces to be warmer than bigger semi-enclosed spaces, because radiated heat from surrounding building’s walls is spread into a smaller area. In addition, the built environment around the semi-enclosed space may protect it from winds; therefore the heat will not readily leave the space through radiation as fast as in a larger or less sheltered area.
Methodology
The data collected consists in area and air temperaturefrom three different semi-enclosed spaces. This was doneroughly throughout a 2 hour period. To begin with, the temperature of the control point was recorded. This was an empty room inside the Queens Building (Site 1 as seen in Figure 1). The purpose of the control point is to assure there are no faults in the thermometer which could produce anomalies. This room was used because it was unattended, there were no people coming in or out and windows being opened or closed. These would add inputs or losses of heat making the room temperature unstable and inadequate toprovidean indication of the well-functioning of our equipment. Thereafter, we took one measurement on each site (Site 1, 2 and 3, figure 1) andmeasured the size in meters of the semi-enclosed areas. Width and length were recorded. To do this two 30 meter tapes were used. The size measurements for site one and two are quite accurate as they are straight lines similar to a square and a rectangle respectively. However Site three was more challenging as it has an irregular shape. In order to measure it we designed a rectangle which would cover most of the area and measured it. Therefore the real value for Site three will be larger than the recorded measurement. Once air temperature and size were measured we returned to the control point to reassure the well-functioning of the thermometer.
Figure 1-Map of RHUL Campus
A total of 6 air temperature measurements were recorded afterwards in two sets of three. In addition, these were done on a one minute interval to allow the thermometer to calibrate and add consistency. For consistency purposes we also had the same roles throughout the day, therefore body and ground distance to thermometer remained constant.Three measurements rather than one were taken in order to increase the accuracy of the results; mean values can be calculated andtheseare a more accurate representation ofreal air temperature. Moreover, taking three measurements also avoids including random inconsistencies from the thermometer which would be recognized as such when compared to the rest.
Measurements were not done at the same time; there was a lag of about 15 minutes between sites. Thisis the result of having to walk from one location to another prior to the measurement.After every cycle of measurements (that is Site one, followed by Site two, and followed by Site three) we returned to the control point and recorded air temperature there.
Our primary data was collected through the sampling method of ‘systematic sampling’. We defined a specific order to follow between sites, the number of measurements to take, and the intervals at which to take them. This means we were taking conscious samples from the population of possible measurements.The whole population is the temperature at all times, something impossible to measure given the circumstances; therefore we had to record a substantial number of measurements within the given time that would allow us to perceive the relationship between temperature and the size of semi-enclosed areas. We deemed this method appropriate because it provides an even coverage of the target population necessary for our study (Fei-Fei Kao et al 2011).
In order to select three semi-enclosed areas in Royal Holloway campus we used a system of rejection from limiting factors. These spaces had to be different sized for us to test our hypothesis. In addition, it was best to design a route which would be as short as possible to reduce diurnal changes in air temperature caused by cloud cover, sun exposure, wind and rain. These are very limiting factors leaving us with few options to choose from. Once we decided theavailable options we picked the best in terms of distance to each other, difference in size and level of en-closeness. Note here that we chose founders South Quad instead of Founders North Quad because the gate in the North quad is completely open whereas the gate in the opposite quad is half closed (Figure 2). This means the likelihood of a wind tunnel increasing the variability of our results is higher in the North Quad. In addition, the Statue at the center of the South Quad meant measurements could not be taken in the exact middle; they were however taken beside its west side, acting as further shelter from wind.
In order to select where inside the enclosed area would air temperature bemeasured it was decided, based on basic principles of heat transfer, that the best place to measure air temperature would be the center. This is because heat is radiated by buildings, therefore any position but the center would be influenced by the heat transferred directly from the nearest wall. However, the center will be the point where there will be a heat balance. The heat balance is the result of temperature gradients between objects or bodies and its surroundings evening out and reaching thermal equilibrium. (H.Lienhard IV, 1984). These heat transfer movements happen from regions of warmer air to regions of colder air (H.Lienhard IV, 1984). Thereby the center of the space will be the best place to take air temperature measurements.
Judging our methodology: Limitations and Improvements
The limitations of our methods were first and most importantly temporal; measurements could only be recorded in a short period of time (90 minutes). Secondly, shortage of equipment: -only one thermometer per group; disabling the group from splitting to increase data collection. Without these restrains it would have been possible to collect a larger data base and measure more sites. If this project was to be done again without limitations thermometers would be placed at each location to measure air temperature over a long period of time. This would increase the reliability of our conclusions by increasing the data base and by reducing the variables of diurnal temperature change experienced from walking from site to site.
Results
Statistic / Queens / Wedderburn / FoundersMean Temperature / 14,3 / 13,17 / 12,9
Minimum Temperature / 12,9 / 12,4 / 12,2
Maximum Temperature / 15,7 / 13,9 / 13,6
Range / 2,8 / 1,5 / 1,4
Standard Deviation / 0,98 / 0,56 / 0,62
Figure 3: Table displaying results collected
Figure 3 shows the statistical data of our experiment. Queens has the highest temperature recorded and the highest mean value.It also has the highest range and standard deviation. This shows that there is a greater variability in the temperature readings from this site as compared to the others.Wedderburn and Founders data looks alike; the values in Figure 2 do not differ from any more than 0.3 in any category. Nevertheless, Founders does appear to be slightly colder than Wedderburn.
Figure 4: Air temperature change with Time
Figure four shows how temperature readings changed throughout the day. Sharp changes show a move from one set of measurements to another. Queens definitely stands out as the warmest location. Wedderburn and Founders show similar results. All three sites show a decreasing trend in temperature throughout the day.
Figure 5: Mean temperature against the size of the Semi-enclosed areas
Figure 5 shows an inverse relationship between air temperature and size. Temperature decreases as size increases. The gradient between 3500 and 3700 meters squared (Wedderburn and Foudners) is greater than the gradient between 0 and 3500 meters squared (Queens and Weddeburn).
Figure 6: Air temperature against Data sample number
Figure 6 shows all the values collected for air temperature in all sites. Temperature is always higher in Queens, site 1. The middle and lowest temperaturevaries during the experiment between Wedderburn and Founders; however, the linear lines show that total average temperature is higher in Wedderburn. Alltogether, the linear lines show a general downwards trend in temperature in all locations.
Site / Weather 1st reading / Weather 2nd reading / Weather 3rd readingQueens / Slightly Windy, cloudy, grey sky. / Sunny, Low wind. / Grey sky, slight drizzle and Low wind.
Wedderburn / Partially Windy, grey sky, cloudy. / Fairly Windy, grey sky, Slight drizzle / Fairly windy, Darker Sky, Light Shower.
Founders / Slightly Windy, Grey sky, cloudy. / Slightly Windy, Slight Drizzle. / Low wind, Darker sky, Light shower.
Figure 7: Recorded Weather conditions at each site during the time air temperature was recorded.
Figure 7 describes the weather conditions at the time when air temperature was being measured. There is a general trend showing a worsening in weather conditions from cloudy skies towards the beginning to slight showers towards the end.
Analysis
Overall, all evidence suggests Queens is the warmest site. Being the smallest semi-enclosed space this supports our hypothesis. However,Queens also has the highest range and standard deviation. This shows that there is a greater variability in the temperature readings from this site as compared to the others. This could be the result of changes in diurnal temperature due to cloud cover, wind and sunlight exposure. This is supported when comparing Figure 6 to Figure 5; the highest recorded temperatures coincide with the best weather conditions, “Sunny, low wind”. Nevertheless there is some confidencein supporting our hypothesis because the lowest value in Queens is higher than the lowest value in site 1 and 2, even when weather conditions were similar in each site.
Moreover, if we take into consideration Figure 4: Temperature readings matched with time, Queens is always above Founders and Wedderburn. It is important to reiterate the importance of weather variability. On strong wind conditions the ´Urban Heat Island´ effect is reduced, whereas with low wind conditions it is enhanced (McClatchey 2005). Therefore, residual heat is more present with low wind speeds. In our sites, strong winds can reduce the accumulation of residual heat inside semi-enclosed spaces. Winds induce vigorous mixing which spreads any impact of heat residues towards greater depths of the atmosphere (McClatchey 2005).
Looking at the data, in particular Figure 3 and 4, we can see remarkable similarities in temperature between Wedderburn and Founders. Taking into account their similar size (Figure 4),our hypothesis could be supported by the data, as similarly sized semi-enclosed spaces are expected to have similar air temperature records. However, bearing in mindthat Wedderburn is less enclosed than Founders and its real size is slightly larger it would be nothing but logical for this site to be colder, but the results suggest otherwise. This could be explained by the shape of the semi-enclosed spaces, Founders is much wider than Wedderburn, where the total area is relatively large but the average width is not [Founders width at any point: 52.1 meters, Wedderburnmaximum width: 42.9 meters, Figure 8]. This means opposite walls are closer to each other in Wedderburn and further apart in Founders. Thereforeheat accumulation in the center should be more pronounced than in Founders, where the walls at opposite ends are further apart. This could explain why Wedderburn is warmer.
In addition, temperature inside these semi-enclosed spaces will depend to some extent on temperature inside the building and the material which the building is made of. These variables have not been studied in this research project and if they had been more accurate conclusions could be made.
Conclusions
The data collected supports to some extent the hypothesis that the smaller the semi-enclosed space is the higher the air temperature will be. Air temperature does appear to be related to size, the smallest space has the highest temperature and the larger has the lowest temperature. However, not having recorded more measurements and not having recorded them at the same time in each site as well as not having measured temperature inside buildingsmeans that conclusions cannot be very confident. In addition, due to distance between walls, the shape of the enclosed space has to have an effect upon its temperature so total area may not be critical. This escapes our hypothesis as it doesn’t include this variable. Moreover, air temperature at the site relies heavily on diurnal temperature changes due to shifts in cloud cover and positioning. For this reason, if air temperature at each site was measured at times where the weather was not the same, and there was for example more or less sun shining, data would directly be influenced by this variable. This again escapes our hypothesis and deems to weaken its support.
On that account, it can be concluded that there is a strong rejection for the null hypothesis however, that does not go hand by hand with a strong support for the hypothesis. There is enough supporting evidence to hint the relationship between sizes of semi-enclosed spaces and their temperature. Nonetheless, there is not enough data collected to support its exact influence which could place size as the key factor for temperature in semi-enclosed areas.
References
-Douglas C. McGill (1983) With skyscrapers, a windy day is windier, The New York Times. [Online at
-Fei-Fei Kao, Ching-Ho Leu, Chien-HaoKo (2011), Remainder Markov systematic sampling, Journal of Statistical Planning and Inference, Volume 141, Issue 11, November 2011, Pages 3595-3604, ISSN 0378-3758
-John H.Lienhard IV/ John H. Lienhard V (2008) A Heat Transfer TextBook, The Third Edition, Published by Phlogiston Press.
-John McClatchey ‘Regional and local climates’, in Joseph Holden (2005) An introduction to physical geography and the environment
-Laura Smith-Spark, 2013, Reflected light from London skyscraper melts car, [Online at
-NastaranShishegar (2013)Street Design and Urban Microclimate: Analyzing the Effects of Street Geometry and Orientation on Airflow and Solar Access in Urban Canyons, Journal of Clean Energy Technologies, Vol.1(1), p.52
-S. S. Peng, S. L. Piao, P. Ciais, P. Friedlingstein, C. Ottle, F. M. Breon, H. J. Nan, L. M. Zhou and R. B. Myneni(2012)Surface Urban Heat Island Across 419 Global Big Cities,
Environmental Science & Technology, 2012, Vol.46(2), pp.696-703[Peer Reviewed Journal]
Appendices
Site / Width(meters) / Length(meters) / Total Area(M ²)Queens / 13 / 23 / 299
Wedderburn / 42.9 / 83.2 / 3569
Founders / 52.1 / 70.9 / 3694