Cooling effects of water body in hot and humid climate

COOLING EFFECTS OF WATER BODY IN HOT AND HUMID CLIMATE ANDRITA DYAH SINTA NINDYANI (BACHELOR OF ARCHITECTURE, GADJAH MADA UNIVERSITY) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE (BUILDING) DEPARTMENT OF BUILDING NATIONAL UNIVERSITY OF SINGAPORE 2012 DECLARATION I hereby declare that this thesis is my original work and it has been written by me in its entirety. I have duly acknowledged all the sources of information which have been used in the thesis. This thesis has also not been submitted for any degree in any university previously. _________________________ Andrita Dyah Sinta Nindyani 7 December 2012 ii ACKNOWLEDGEMENTS “I can do all things through Christ who strengthen me” (Phillipians 4:13). Thank you dear Father Jesus Christ for my life. Thank you for the faith. I am able to finish this thesis through You who gives me strength. I love you. I owe an enormous debt of gratitude to Professor Wong Nyuk Hien, whose depth of knowledge and immense wisdom greatly aided my scholarly development. He provided a patient and critical eye on my ideas, analyses and writing – I am very thankful for that, and more. Many thanks are also due for Steve Kardinal Jusuf, for his comments and critiques while I was planning and writing up my research for this thesis. During my time in NUS, I received support, encouragement and assistance of various kinds from very many friends and colleagues in graduate school. To me, you guys have been fantastic sounding boards for ideas. Unfortunately, naming all of you is an impossible task; but do know that I would have never been able to complete this demanding and timeconsuming thesis without your help. I have to specifically mention that Nedyomukti Imam Syafii, Erna Tan, Rosita Samsudin, Religiana Hendarti, Enrica Rinintya, Leni Sagita Supriadi, and Bayu Aditiya Firmansyah were all directly involved in assisting my research at one point or another, for which I am deeply grateful. The final words of gratitude are for the people who know me best, and who have stuck with me throughout my academic accomplishments in a foreign land. To my dear parent, Papi iii Nindyo Suwarno, and Mami Dewi Rindjani – thank you for all the pray, support, help, patience, and believe despite being thousands miles away. I am gratefully proud to be your daughter. I love you both. Lastly, to my Husband Antonius Aditiyo Wibisono and Son Dominik Jonathan Pratama – thank you for keeping me grounded, and for giving me a reason to believe. I love you both. iv TABLE OF CONTENTS DECLARATION…………..………………………………………………………………..ii ACKNOWLEDGMENTS...………………………………………………………………..iii TABLE OF CONTENTS ....................................................................................................... v EXECUTIVE SUMMARY................................................................................................. viii LIST OF TABLES ................................................................................................................. x LIST OF FIGURES .............................................................................................................. xi CHAPTER 1 INTRODUCTION ........................................................................................... 1 1.1 Background ............................................................................................................. 1 1.2 Research Questions ................................................................................................. 4 1.3 Research Objectives ................................................................................................ 4 1.4 Scope of the Study and General Methodology ........................................................ 4 1.5 Contributions of the Study ...................................................................................... 7 1.6 Organization of the Study........................................................................................ 7 CHAPTER 2 LITERATURE REVIEW ................................................................................ 9 2.1 Climate of Singapore ............................................................................................... 9 2.2 Solar Radiation ...................................................................................................... 10 2.3 The Hydrologic Cycle (Water Cycle): Evaporation .............................................. 11 2.4 Wind ...................................................................................................................... 12 2.5 Water Bodies and Their Effect on Air Temperatures in Sub-Tropic Areas .......... 14 2.6 Water Facilities and Their Effect on the Surrounding Microclimate .................... 16 2.6.1 Water Facilities and Their Effect in Sub-Tropic Areas ................................. 16 2.6.2 Water Facilities and Their Effect in Singapore .............................................. 18 2.6.3 Water Facilities and Their Effect through Simulation Study ......................... 20 2.7 Green Area Studies in Singapore .......................................................................... 21 2.8 ENVI-met Simulation............................................................................................ 23 2.9 Validation and Sensitive Analysis of ENVI-met Simulation ................................ 24 v 2.10 Simulation Limitations .......................................................................................... 24 2.11 Knowledge Gap ..................................................................................................... 25 2.12 Hypothesis Development ...................................................................................... 27 CHAPTER 3 RESEARCH METHODOLOGY .................................................................. 29 3.1 Research Design .................................................................................................... 29 3.2 Selection of Water Bodies ..................................................................................... 29 3.2.1 Kallang River ................................................................................................. 29 3.2.2 Sungei Api-api River...................................................................................... 30 3.2.3 Marina Bay ..................................................................................................... 31 3.2.4 Bedok Reservoir ............................................................................................. 32 3.3 Instruments Used ................................................................................................... 33 3.4 Data selection ........................................................................................................ 35 3.5 Method of Data Collection .................................................................................... 37 3.6 Data Processing ..................................................................................................... 37 3.7 Method of Analysis ............................................................................................... 38 3.8 Point Location of Experiment ............................................................................... 39 3.8.1 Kallang River ................................................................................................. 39 3.8.2 Sungei Api-api River...................................................................................... 39 3.8.3 Marina Bay Promenade and Marina Bay Promontory ................................... 40 3.8.4 Bedok Reservoir ............................................................................................. 41 3.9 ENVI-met Simulation............................................................................................ 42 3.9.1 Simulation Procedure in ENVI-met ............................................................... 43 3.9.2 Simulation Boundary Condition .................................................................... 44 CHAPTER 4 OBJECTIVE DATA ANALYSIS ................................................................. 45 4.1 A Comparison of Four Measurements in Regard to the Distance ......................... 45 4.1.1 Away from the Water Bodies (Clear Days) ................................................... 46 4.1.2 Away from the Water Bodies (Cloudy Days) ................................................ 49 4.1.3 Along the Water Bodies (Clear Days) ........................................................... 50 4.1.4 Distance Effect with Polynomial Regression ................................................. 52 vi 4.2 4.1.4.1 Kallang River and Sungei Api-api River ................................................. 52 4.1.4.2 Marina Bay and Bedok Reservoir ............................................................ 54 Solar Radiation Effect on Water Bodies Cooling Performance ............................ 56 4.2.1 Kallang River ................................................................................................. 56 4.2.2 Sungei Api-api River...................................................................................... 59 4.2.3 Bedok Reservoir ............................................................................................. 63 4.2.4 Marina Bay ..................................................................................................... 65 4.2.5 Solar Radiation Effect with Linear Regression .............................................. 69 4.3 Overall Field Measurement Findings .................................................................... 70 4.3.1 Additional Findings ........................................................................................ 71 CHAPTER 5 ENVI-met SIMULATION ANAYLSIS ........................................................ 75 5.1 Simulation Validation............................................................................................ 75 5.2 Scenario 1: Kallang River Real Condition ............................................................ 76 5.3 Scenario 2: Kallang River with a Wider Waterway .............................................. 84 5.4 Scenario 3: Kallang River with Wind Speed of 2m/s ........................................... 89 5.5 Scenario 4: Kallang River with All Grass Covered Microclimate ........................ 95 5.6 Scenario 5: Kallang River with All Pavement Covered Microclimate ................. 99 5.7 Overall Simulation Findings ............................................................................... 105 CHAPTER 6 SUMMARY AND CONCLUSION ............................................................ 108 6.1 Summary and Conclusion ................................................................................... 108 6.2 Limitations of the Study ...................................................................................... 112 6.3 Recommendations for Future Work .................................................................... 113 BIBILOGRAPHY .............................................................................................................. 115 APPENDIX A .................................................................................................................... 122 vii EXECUTIVE SUMMARY Water bodies are known to be about the best absorbers of radiation, yet they exhibit very little thermal response. A study in sub-tropic regions found a difference of approximately 3–5°C in air temperature between the river and the city area. The water bodies of the river operate as a cooling source on the microclimate of the surrounding area. Air temperature near or over bodies of water is much different from that over land due to differences in the way water heats and cools. In Singapore, water features within an urban area have a positive effect on the microclimate of the surrounding areas when natural cooling from the evaporative process is needed on a hot sunny day. The increased availability of water usually enhances evaporation, and the associated uptake of latent heat provides a daytime cooling effect. Many other researchers have argued that evaporative cooling from water bodies or water features is one of the most efficient ways to ensure the passive cooling of building and urban spaces. However, evaporative cooling might not work optimally in a hot, humid tropical country as it has high relative humidity. Research on the mitigation plans has been widespread, demonstrating that having more greenery is an efficient measure for curbing urban heat islands. Although studies have also mentioned that having more water surfaces could improve the urban heat island effect, this possibility has received comparatively less attention. In this research, four water bodies in Singapore, the Kallang River, Sungei Api-Api River, Marina Bay, and Bedok Reservoir, were looked into as an effective measurement of the waterways’ evaporative cooling performance fort the surrounding microclimate, especially viii in hot and humid climates. Air temperature, relative humidity, wind velocity, and solar radiation are continuously measured for the collective data and analyzed for the clearest day of the measurement. A field measurement was conducted for five months, from May to September 2010, for Kallang and Sungei Api-api Rivers and for another five months, February to June 2012, for Marina Bay and Bedok Reservoir. Based on solar radiation performance during the daytime, in order to observe the extent of its cooling effect from the waterway, there are measurement points for each location. Some points of measurement are located along the waterway while other points are located moving away from the waterway. In order to study the water bodies’ cooling effect unable to be investigated using the field measurements, an ENVI-met simulation study was conducted. Five scenarios are simulated to investigate the temperature profile of each scenario. A parametric study, which cross references the results obtained from the investigation and boundary data, was performed to observe the impact of the cooling effect for each scenario. These research results support the hypothesis that the four water bodies in Singapore and the simulation study have similar results. On a clear day with enough solar radiation, temperatures in surrounding area increase by 0.15 to 0.20oC approximately 20–30 meters away from the water bodies. ix LIST OF TABLES Table 1: Data selection ......................................................................................................... 36 Table 2: Basic input parameters to the ENVI-met model .................................................... 44 x LIST OF FIGURES Figure 1.1: Research Methodologies...................................................................................... 6 Figure 2.1: Illustration of the hydrologic cycle .................................................................... 11 Figure 3.1: Kallang Basin .................................................................................................... 30 Figure 3.2: Sungei Api-api River ......................................................................................... 31 Figure 3.3: Marina Bay ........................................................................................................ 31 Figure 3.4: Bedok Reservoir ................................................................................................ 32 Figure 3.5: Weather station .................................................................................................. 34 Figure 3.6: Hobo Data Logger ............................................................................................. 35 Figure 3.7: Points located along the Kallang waterway....................................................... 39 Figure 3.8: Points located along the Sungei Api-api waterway ........................................... 39 Figure 3.9: Points located along the Marina Bay Promenade points MP1 – MP5 .............. 40 Figure 3.10: Points located along the Marina Bay Promenade points MP6 – MP8 ............ 40 Figure 3.11: Points located away the Marina Bay Promontory ........................................... 41 Figure 3.12: Points located away the Bedok Reservoir ....................................................... 41 Figure 4.1: Comparison of daytime average temperature (clear days) at points at a distance from the four water bodies .............................................................................. 46 Figure 4.2: Comparison of daytime average temperature (cloudy days) at points at a distance from the four water bodies ................................................................ 50 Figure 4.3: Comparison of daytime average temperature (clear days) at points along the water bodies ................................................................................................... 51 Figure 4.4: Correlation between temperature reduction and distance at Kallang ................ 53 Figure 4.5: Correlation between temperature reduction and distance at Sungei Api-api .... 53 Figure 4.6: Correlation between temperature reduction and distance at Marina Bay .......... 54 Figure 4.7: Correlation between temperature reduction and distance at Bedok Reservoir .. 55 Figure 4.8: Comparison of diurnal average temperatures (clear days) at points away from the Kallang waterway ..................................................................................... 56 Figure 4.9: Diurnal average temperatures (clear days) at points along the Kallang waterway ........................................................................................................................ 56 xi Figure 4.10: Comparison of diurnal average temperatures (cloudy days) at points away from the Kallang waterway............................................................................. 57 Figure 4.11: Comparison of diurnal average temperature (clear days) at points along the Kallang waterway with solar radiation ........................................................... 58 Figure 4.12: Comparison of diurnal average temperatures (clear days) at points away from the Sungei Api-api River ................................................................................ 60 Figure 4.13: Comparison of diurnal average temperatures (cloudy days) at points away from the Sungei Api-api River........................................................................ 60 Figure 4.14: Diurnal average temperatures (clear days) at points along the Sungei Api-api waterway and the beach .................................................................................. 61 Figure 4.15: Comparison of diurnal average temperature (clear days) at points along the Sungei Api-api waterway with solar radiation ............................................... 62 Figure 4.16: Comparison of diurnal average temperature (clear days) at points along the Bedok Reservoir with solar radiation ............................................................. 63 Figure 4.17: Diurnal average temperatures (cloudy days) at points away from the Bedok Reservoir ......................................................................................................... 64 Figure 4.18: Comparison of diurnal average temperatures (clear days) at points away from the Marina Bay................................................................................................ 65 Figure 4.19: Comparison of diurnal average temperatures (clear days) at points along Marina Bay ..................................................................................................... 66 Figure 4.20: Diurnal average temperatures (cloudy days) at points away from Marina Bay ........................................................................................................................ 67 Figure 4.21: Comparison of diurnal average temperature (clear days) at points along Marina Bay with solar radiation .................................................................................. 68 Figure 4.22: Correlation between temperature reduction and solar radiation ...................... 70 Figure 4.23: Comparison between temperature and relative humidity along the waterway 72 Figure 4.24: Comparison between temperature and relative humidity away from the waterway ......................................................................................................... 73 Figure 4.25: Correlation between avverage relative humidity and average temperature on clear days in Kallang....................................................................................... 73 xii Figure 5.1: Comparison of diurnal average temperature of field measurement and simulation on Kallang River ........................................................................... 75 Figure 5.2: Temperature profile of Kallang River at 10:00 a.m .......................................... 78 Figure 5.3: Temperature profile in Kallang River at 1:00 p.m ............................................ 79 Figure 5.4: Temperature profile in Kallang River at 5:00 p.m ............................................ 81 Figure 5.5: Comparison of diurnal average temperatures of Kallang River at three times . 82 Figure 5.6: Temperature profile of Kallang River with a wider river width at 10:00 a.m ... 85 Figure 5.7: Temperature profile of Kallang River with a wider river width at 1:00 p.m .... 86 Figure 5.8: Temperature profile of Kallang River with a wider river width at 5:00 p.m .... 87 Figure 5.9: Comparison of diurnal average temperature of real condition and wider river on Kallang ............................................................................................................ 88 Figure 5.10: Temperature profile of Kallang River with wind speed 2m/s at 10:00 a.m .... 90 Figure 5.11: Temperature profile of Kallang River with wind speed 2m/s at 1:00 p.m ...... 91 Figure 5.12: Temperature profile of Kallang River with wind speed 2m/s at 5:00 p.m ...... 93 Figure 5.13: Comparison of diurnal average temperature of Kallang real condition with 2m/s wind speed.............................................................................................. 94 Figure 5.14: Temperature profile of Kallang River with all grass surrounding cover at 10:00 a.m................................................................................................................... 96 Figure 5.15: Temperature profile of Kallang River with all grass surrounding cover at 1:00 p.m .................................................................................................................. 97 Figure 5.16: Temperature profile of Kallang River with all grass surrounding cover at 5:00 p.m .................................................................................................................. 98 Figure 5.17: Temperature profile of Kallang River with all pavement surrounding cover at 10:00 a.m ...................................................................................................... 100 Figure 5.18: Temperature profile of Kallang River with all pavement surrounding cover at 1:00 p.m ........................................................................................................ 101 Figure 5.19: Temperature profile of Kallang River with all pavement surrounding cover at 5:00 p.m ........................................................................................................ 102 Figure 5.20: Comparison of diurnal average temperature of Kallang River with different surrounding cover at 10:00 a.m .................................................................... 103 xiii Figure 5.21: Comparison of diurnal average temperature of Kallang River with different surrounding cover at 1:00 p.m ...................................................................... 104 Figure 5.22: Comparison of diurnal average temperature of Kallang River with different surrounding cover at 5:00 p.m ...................................................................... 105 xiv CHAPTER 1 1.1 INTRODUCTION Background Urban heat islands (UHIs) have created serious environmental problems all over the world, where differential heating is registered in urban areas unlike rural surroundings. The concept of UHI, proposed by Manley (1958), stated that—when a city has expanded enough to change the properties of the underlying surface to suffer from serious air pollution and release substantial waste heat—the urban temperature is considerably higher compared to the rural system, thereby generating the thermal isolated island. The occurrence of UHI can be attributed to mainly manmade surroundings; however, the amount of heat released is dependable on the prevailing natural environmental conditions (Memon et al., 2008). The phenomenon of the UHI effect is has affected the long-term trends of mean temperature and the rise of both day- and night-time temperatures (Memon et al., 2008). According to Santamouris et al. (2001), the heat island’s intensity can result in up to 10 K temperature differences between dense urban areas and the surrounding rural zones. The negative effect in especially hot and humid countries has resulted in an increase of energy consumption associated with the need for air conditioning (Ca et al., 1998), elevation in ground-level ozone (Rosenfeld et al., 1998), deterioration in the quality of living environments with a significant increase of pollutant emissions and smog production (Santamouris and Mihalakakou, 2000), decrease in human comfort, and even an increase in mortality rates (Changnon et al., 1996; Ichinose et al., 2008). 1 Many studies have shown that the mitigation of UHI measures, such as increasing the quantity of vegetation cover, might alter the severe impact of urbanization. The vegetation provides a cooling effect, mainly through its shadowing effect and evapotranspiration process. The process is basically a natural mechanism in which heat is removed by changing the heat from sensible heat to latent heat. A similar process happens over water bodies with the help of solar radiation. This process is known as evaporative cooling. When solar radiation from the sun reaches the water’s surface, the water will vaporize and remove the heat, thereby cooling the surrounding air. Water features within an urban area have a positive effect on the microclimate of the surrounding areas when natural cooling from the evaporative process is needed on hot sunny days. The increased availability of water usually enhances evaporation, and the related uptake of latent heat provides an additional daytime cooling effect. The air temperature near or over bodies of water is much different from that over land due to differences in the way water heats and cools. Water bodies are noted to be about the best absorbers of radiation; on the other hand, they exhibit very little thermal response. Many other researchers have argued that evaporative cooling from water bodies or water features is one of the most efficient ways to passively cool building and urban spaces. However, evaporative cooling might not work optimally in a hot, humid tropical country with high relative humidity. The lack of response can be attributed to four characteristics (Oke, 1987): 2  Penetration: as water allows short-wave radiation transmission to considerable depths, energy absorption is diffused through a large volume.  Mixing: the existence of convection and mass transport by fluid motion also permits the heat gains/losses to be spread throughout a large volume.  Evaporation: unlimited water availability provides an efficient latent heat sink, and evaporative cooling tends to destabilize the surface layer and further enhance mixing.  Thermal capacity: the thermal capacity of water is exceptionally large; therefore, it requires about three times as much heat to raise a unit volume of water through the same temperature interval as most soil. These properties make the surface temperature of water bodies cooler than that over the land. A cooler surface results in a cooler air temperature above the water. A study by Murakawa (1990) showed a difference of approximately 3–5°C in air temperature between the river and the city area in Japan. The water bodies of the river operate as a cooling source on the microclimate of the surrounding area. Many other researchers have argued that evaporative cooling from water bodies or water features is one of the most efficient ways to provide passive cooling for building and urban spaces (Krüger, 2008; Adebayo, 1991). Thus, the current study examines this evaporative cooling performance of water bodies for the surrounding microclimate of Singapore. Ambient air temperatures are measured to make a clear distinction of the influence of cooling from the water bodies horizontally. 3 1.2 Research Question 1. How is solar radiation affecting the cooling effect of water bodies in the tropics, especially Singapore? 2. How do water bodies in hot and humid climates extend the cooling effect to the surrounding area in terms of distance? 3. How does the surrounding area impact the cooling effect performance of water bodies, as determined using a simulation model? 1.3 Research Objectives The objectives of the study are as follows: 1. To determine the cooling effect and benefit of water bodies to their surrounding microclimate in hot and humid climates (field measurement study). 2. To determine the possible impact of water bodies’ cooling effect on the air temperature on a hot sunny day (field measurement study). 3. To identify different types of surrounding areas near the water bodies in order to investigate their impacts on the water bodies’ cooling effect (parametric study using an ENVI-met simulation). 1.4 Scope of the Study and General Methodology The scope of this research is focused on: 1. Water bodies’ cooling effect performance in terms of distance in several locations in Singapore based on solar radiation performance; and 4 2. The use of field measurement method and ENVI-met simulations to study the effect of a water body’s surrounding area cover on the cooling performance on a hot sunny day; but 3. No mean radiant temperature (MRT) reading was used as this research is not focused on human thermal comfort. Figure 1.1 shows the general methodology of this study. The process started with a preliminary literature review, focusing on UHI, the water cycle, and the variables that affect the water’s evaporation process to create a cooling performance on its surroundings. Water bodies in sub-tropic areas seem to have a significant cooling effect amount. On the other hand, Singapore, with its hot and humid climate, seeks to achieve the same effect. Some researchers in Singapore have measured water facilities such as water walls, fountains, and ponds. They found that evaporation from the water facilities could help reduce the surrounding heat. Hence, the aim of the study is to determine the cooling effect of water bodies (rivers, bays, and reservoirs) on a larger scale in Singapore. Field measurements and an ENVI-met simulation were conducted to determine the water bodies’ cooling effect performance. 5 Preliminary literature review Formulation of research problem In-depth literature review Design of research Field Measurement study Simulation study with ENVI-met Data measurement and collection Modeling and running simulation Data analysis using EXCEL Extract data for ENVImet using LEONARDO Research findings Simulation findings Recommendations Thesis report Figure 1.1 Research Methodologies 6 Generate hypothesis 1.5 Contribution of the Study These findings could be used to understand the extent of water bodies’ cooling effect on the surroundings area in hot and humid climates based on solar radiation performance. In addition, through simulation study, the findings could be used to determine how the surroundings near the water bodies affect the water bodies’ cooling effect performance. 1.6 Organization of Study This thesis paper will be organized as follows: Chapter 1 provides an introduction to the background and rationale of this study. A brief outline of the research paper is provided. Chapter 2 presents an extensive literature review on past research and papers done in relation to the cooling effect of water features in Singapore (a hot and humid climate). The literature on the water evaporation process and factors that affect it to produce water bodies’ cooling effect, such as solar radiation and wind speed, are also described. Chapter 3 introduces the research methodology used in this study. An account of the measurement instruments deployed and the parameters measured will be included in the chapter. It also covers the methodology of the ENVI-met simulation program to conduct simulations in water bodies’ areas in Singapore regions. Chapter 4 delivers the analyses of water bodies’ cooling effect based on solar radiation performance using data collected from four field measurements in Singapore’s water bodies’ areas. It further discusses the distance effect based on clear days at the study cases. 7 Chapter 5 produces the simulation outcome of the water bodies by conducting an ENVImet simulation on the Kallang River in Singapore to analyze how much the water cooling effect might vary on a hot sunny day in the region as it refers to real conditions on site. Chapter 6 puts forward a summary of the issues, in which the conclusions from the analyses will be established. It also discusses the limitations of this study and suggests recommendations for future works. 8 CHAPTER 2 2.1 LITERATURE REVIEW Climate of Singapore Based on National Environmental Agency (NEA) Meteorological Services, Singapore, with an area of approximately 710.2 km2, is located in the tropics surrounded by sea (lies at 15 meters sea level) and has fairly high humidity. The climate is characterized by a uniformity of temperature, pressure, high relative humidity (RH), and heavy rainfall. Singapore is influenced by the sea based on its geographical location and maritime exposure. The sea breeze is a steady wind that blows inland during the day from the sea, carrying humidity. Singapore has a stable climate condition, with temperatures varying from 22°C to 34°C and average RH of 85%–90% in the morning and 55%– 60% during the daytime. On a rainy day, the RH could easily reach 100%. June and July are the hottest months of the year, and November and December are considered as the rainy season. Generally, due to Singapore's geographic location (i.e., close to the equator), Singapore has warm weather conditions with relatively high RH. Located between latitudes 1 degree 09’N and 1 degree 29’N and longitudes 103 degree 36’E and 104 degree 25’E, Singapore can be classified as having a hot humid climate. Uniform high temperatures, humidity, and rainfall throughout the year characterize the climate. The diurnal temperature variations are small, ranging from 23 to 26°C and 31 to 34°C. The mean annual temperature was 27.4oC between 1982 and 2001. Relative humidity is generally high and, although it invariably exceeds 90% in the early hours of the morning just before sunrise, it frequently falls to 60% during the afternoons 9 when there is no rain. During prolonged heavy rains, relative humidity often reaches 100%. The mean annual relative humidity is 84.2%. There are two main seasons in Singapore, the northeast (NE) monsoon and the southwest (SW) monsoon seasons. The NE monsoon occurs between November and early March, with the prevailing wind blowing from the north to northeast. Meanwhile, the SW monsoon occurs between June and September, with the prevailing wind blowing from the south to southwest. Two short inter-monsoon periods lasting two months separate the main seasons. There is no clear distinct wet or dry season as rainfall occurs throughout the year. However, the NE monsoon season is considered to be wet weather as the wind is generally cool and brings frequent spells of wet weather, accounting for about 48% of the total annual rainfall. On the other hand, the SW monsoon wind brings about 36% of the total annual rainfall. 2.2 Solar Radiation Solar radiation is a radiant energy derived from thermonuclear processes occurring in the sun. Solar radiation controls the climate conditions on earth and varies at different latitudes (Kiehl, 1992). This is mainly due to differences in solar radiation that reaches the surface. The distribution of solar radiation is generally larger at low altitudes and much less nearer the poles. This imbalance in the net radiation for the surface and the atmosphere system as a whole (positive in lower altitudes and negative in higher altitudes) combined with the effect of the earth’s rotation on its axis produces the dynamic circulation system of the atmosphere (Henderson-Sellers and McGuffie, 1987). As previously mentioned, the most influential factor for determining the climatic conditions on earth is solar radiation. According to an earth energy budget by Schneider (1992), 45% 10 of incoming solar radiation is absorbed by the surface of the planet, 25% by the atmosphere, and 23% is reflected by the atmosphere. 2.3 The Hydrologic Cycle (Water Cycle): Evaporation Figure 2.1 Illustration of the hydrologic cycle. Source: National Weather Service, NOAA (www.srh.noaa.gov/jetstream//atmos/hydrocycle_max.htm) The hydrologic cycle, or the water cycle, as seen in figure 2.1 is the continuous cycling of water through the atmosphere, ocean, land surface, and biosphere. The hydrologic cycle is mainly driven by energy from the sun. The major reservoirs of water on Earth include oceans, lakes, rivers, wetlands, land, and seas. Water can move through several different processes—namely, evaporation, precipitation, melting, and running downhill in rivers or underground. Evaporation is the process by which water gradually changes from a liquid to a gas or vapor at a significant volume. Evaporation is the primary pathway that water 11 moves from the liquid state back into the water cycle as atmospheric water vapor. Studies have shown that the oceans, seas, lakes, and rivers provide nearly 90% of the moisture in the atmosphere via evaporation. A lower wind speed is one factor that could decrease the amount of evaporation from the earth’s surface. Evaporation requires heat for the water to transition from a liquid to a gas. Once in the atmosphere, the water vapor rises and condenses (the process of changing from water vapor to liquid water and releasing heat). The wind can blow this cloud over land, and the water can precipitate as rain or snow. The water might then run over the Earth’s surface into a river or lake or seep into the ground to become groundwater. From the lake, river, or groundwater, the water could flow into the ocean again. At any point in this process, the water can take a different path. Much of the water that evaporates from the ocean precipitates right back into the ocean. Evaporation from the oceans is the primary mechanism supporting the surface-to-atmosphere portion of the water cycle. After all, the large surface area of the oceans (over 70% of the Earth's surface is covered by the oceans) provides the opportunity for large-scale evaporation to occur. On a global scale, the amount of water evaporating is about the same as the amount of water delivered to the Earth as precipitation. 2.4 Wind Wind is an overlooked resource and plays an important role in ameliorating the urban environment. At a fundamental level, wind behavior addresses issues concerning pedestrian 12 comfort in urban areas in general with respect to how people feel while relaxing or walking in the present or future urban environment. On the other hand, the wind behaviors can be observed through the identification of wind patterns that illustrate local circulation. Such an approach has some relevance in different contexts. Taseiko (2008) and Oke (1987), examining air quality studies, found that some wind patterns can transport pollutants/heat into an urban environment, while others can bring clean and cooler air. In the sustainable design context, understanding wind patterns is mainly important for modeling the efficiency and performance of the natural ventilation design (Croome and Roberts, 1980). In addition, the highest intensity of wind over an area usually presents some advantage directions and, for this reason, wind pattern classifications can help understand which flow direction has the possibility to risk or improve microclimate conditions. Another factor of the wind is that, in the afternoon, the wind moves faster than at night. In Singapore, surface winds generally blow from north or northeast during the NE monsoon season (December to March) and from south or southeast during the SW monsoon season (June to September). Mean wind speeds are usually light, below 20 km/hr, although mean wind speeds of up to 40 km/hr can occur during a NE monsoon surge. Winds during the inter-monsoon months are mostly light and variable. The characteristics of the wind flow pattern at low levels in the urban environment are influenced more by the local geometry, such as street geometry, trees, and building height distribution and less affected by the characteristics of the flow in the upper layer (Ricciardelli, 2006). 13 2.5 Water Bodies and Their Effect on Air Temperature in Sub-Tropic Areas Santamouris and Asimakopoulos (1996) pointed out that a space area can be cooled by passive evaporation, a process where evaporation occurs naturally from standing or moving water (such as basins or fountains). According to Munn et al. (1969) and Naot et al. (1991), the evaporation and transfer of sensible heat result in lowering air temperature at the water bodies. Evaporation decreases air temperature due to the latent heat of absorption and increase in specific humidity. Meanwhile, the transfer of sensible heat between air and the underlying water (water is cooler than air) also reduces air temperature, especially under hot weather conditions. In sub-tropic regions, Huang et al. (2008) conducted a fieldwork study in Nanjing on the effect of different types of ground cover on temperature. The results revealed that lawn, water areas, and woods or the shade from tress have the potential to decrease air temperatures between 0.2 and 2.9°C when compared to bare concrete cover. In terms of the potential cooling effect brought about by the different ground covers, usually lawns and woods or shade from trees contribute significant cooling effects during both days and nights. In addition, water bodies were shown to be cooler than concrete areas in the day, but the study also showed the probability of these large water areas contributing heat at night and, in some instances, being even warmer than the concrete surfaces by about 0.4 to 1.0oC. Other literature focused on the restoration of Cheong-Gye Stream in Seoul to determine the evaporative cooling effect of the water surfaces. The stream, running 5.8 km eastward through a central region of Seoul, was expected to help mitigate Seoul’s thermal stress and change hydrology as well as street-level wind fields (Kim et al., 2009). From the 13 points studied along the stream, restoration reflected a near-surface average temperature decrease 14 of 0.4°C over the stream area, with the largest local temperature drop being 0.9°C. Near the middle of the stream, the restoration’s mitigation effects on the UHI were quantified as 0.13°C (Kim et al., 2009). Althought Kim et al. (2008) acknowledged that the decline in temperature cannot be fully attributed to the effect of stream due to distinct weather conditions before and after the restoration, it was mentioned that the Cheong-Gye stream was principally responsible for the temperature distribution along a sheet traversing the stream, with temperatures decreasing as one moves southwards along the stream. In addition, following the restoration, the ratio of sensible heat flux to net irradiative flux dramatically decreased from 0.63 to 0.18 in the daytime (Kim et al., 2008), which could be explained by the absorptive of the surfaces as well as the high heat capacity of the water. However, before and after the stream restoration (2003–2007), it was found that the restoration affected the local environment, resulting in changes in sensible heat flux and temperature mitigation (Ichinose et al., 2009). Other studies have found that city park characteristics have strong impacts on the urban cooling island intensity (Cao et al., 2010; Chang et al., 2007). Recent research in China revealed that the environmental temperature can be reduced by narrow built-up land and rounded shapes in water and green landscapes (Shi, Deng, Wang, Luo, & Qiu, 2011). Consistent with these results, the urban water body area and geometry were also found to significantly influence the urban cooling island (UCI) effects in Beijing. The area of the water body had the greatest effects on variations in the UCI intensity. However, the UCI efficiency had a significantly negative correlation with water body area. This means that, given the same total area of water bodies, more small water bodies can offer more beneficial effects. The water body’s geometry had a negative impact on UCI intensity and 15 efficiency, meaning that square or round water bodies can intensify their cooling effects. The water body’s location and surrounding built-up land were important for the cooling island effects. Dense built-up areas substantially increase the land surface temperature (LST) around water bodies, resulting in higher cooling intensity and efficiency. Moreover, water bodies with the same characteristics in different spatial locations had varying abilities to cool the surrounding thermal environments (Ranhao Sun and Liding Chen, 2012). 2.6 Water Facilities and Their Effect on the Surrounding Microclimate Land cover change can have a significant impact on climate (Sagan et al., 1979; Myers, 1992). On the other hand, the use of water features in a city offers an alternative to vegetation as a method of alleviating high urban temperatures by increasing the latent heat flux from the surface to provide cooler air (Smith and Levermore, 2008). 2.6.1 Water Facilities and Their Effect in Sub Tropic Area In Osaka City, N. Nishimura et al. (1998) conducted a field measurement on existing water facilities, including a pond, waterfall, and spray fountain in a park located in an urban area. The temperature decline effects of waterfalls and spray-type water facilities in urban areas were measured. The results indicated that, even when the spray facility was not in operation, the air temperature was 1o to 2oK lower than the average air temperature in the park. This suggests considerable a potential cooling effect brought about by the water pond and waterfall, which could be amplified with the spray facility function. The presence of water facilities can bring about a positive impact on the microclimate as it was also observed from the results indicating that the closer the measuring points are to the 16 water features, the greater the effect of temperature fall. On the other hand, humidity was noted to be higher when the water sprays were operated, indicating the possibility of thermal discomfort for people. N. Nishimura et al. (1998) analyzed the results and noticed that the temperature fall area induced by the evaporation of water from the spray and water fall operation at the water feature in focus spreads out to a distance of nearly 35 m away from the location. The studies demonstrated that the degree of temperature decline is dependent on the types of water facilities, and water spray facilities were shown to be effective for providing a cooling environment. In a study in Japan, He and Hoyano (2008) examined surface temperature distributions in the mid-day temperatures of the walls with water and found a 2–7oC reduction to nearambient air temperature; the walls were several degrees cooler than those without water. The decreased surface temperature is due to the evaporative cooling effect as the film of water evaporates. In addition, the mean radiant temperature (MRT) near walls with water was slightly lower compared to those without water, with a maximum reduction of approximately 6°C. Furthermore, the heat island potential (HIP) was found to have decreased in areas with watering as compared to values attained from those without watering. Another development in Tokyo, Japan (2003), revived by The Tokyo Metropolitan Government at four different places, was the "Sidewalk Sprinkling Campaign in Tokyo" movement. The results revealed an average decrease of 1°C in temperature before and after sprinkling water as measured by researchers and elementary school pupils (Japan for Sustainability, 2003). In addition, a simulation study by the Public Works Research Institute in Tokyo, Japan, also revealed that the temperature at noon could be reduced a 17 maximum of 2 to 2.5°C, with the assumption that water would be sprinkled concurrently over an area of about 265 square kilometers, using one liter of water per square meter (Yagi, 2009). Furthermore, to enhance the amount of evaporative cooling from the sprinkling of water, some studies (Gartland, 2008; Asawa et al., 2000; Yamagata et al., 2008) explored the use of water retentive and water permeable pavement. Together with this material, water stored in the pavement can evaporate slowly and lower the temperature by vaporization. It was assessed that sprinkling-reclaimed wastewater decreased the road surface temperature by 8°C during the daytime and 3°C at night as measured using thermographs. The lowered temperatures were found to be equal to those on planting zones, and the effect continued overnight due to the use of the material (Yamagata et al., 2008). These results suggest that the cooling effect was provided by the sprinkling of water. In addition, sprinkling on waterretentive pavement is able to adequately mitigate the UHI phenomenon. 2.6.2 Water Facilities and Their Effect in Singapore In hot and arid climates, raising the humidity of the air has brought welcome relief in traditional architecture through different devices, such as fountains, pools, or just splashing the floor of the courtyard with water several times a day. However, as Singapore is humid, it might not find the same benefits. Choo (2008) studied three different water features within a garden area in Biopolis, Singapore. The study concluded that weak relationships existed between air temperature and water features and justified that the water wall had the largest cooling potential. The studies justify the cooling potential of water features, with 18 Jusuf et al. (2009) and Choo (2008) further analyzing their effect in hot and humid Singapore. Field measurements conducted by Jusuf et al. (2009) on a water wall at One North Park, Singapore, further substantiate the cooling benefits brought about by the wall based on air temperatures measured at nearly 1.7-1.8oK cooler than the surroundings. These results showed that the temperature drop is induced by the evaporation of water from the water wall. It was asserted that the presence of the water-wall at One North Park improves the thermal environment by cooling the air via the cooler air temperature near the water wall, resulting in a lower air temperature for the nearby environment (Jusuf et al., 2009). Both studies on the water wall were conducted in the park, where the cooling effect could be enhanced with evapotranspiration from the greenery in the surrounding area. The evaporative effect of water features located in the vicinity of buildings, with comparatively lesser greenery than the park, was not considered. In addition, water fountains that have water sprinkling in the air with an increased surface area, thereby increasing the evaporation rate leading to an added cooling effect, should be considered as a feature just as capable of cooling the air as a water wall (Energy and Resources Institute et al., 2004). In addition to being able to reduce heat load, as previously mentioned, the sprinkling of water into the air can clean dust particles from the air. Hui (2009) found that “moving” water features displayed the potential cooling in both the day and evening. Water fountains showed a higher capability of reducing air temperature than the water wall, with temperature variations of 4.0oC in the day and 1.3oC in the evening. However, the “still” water feature (i.e., water pond) recorded a higher temperature than the reference point in both the day and evening. This result conflicted with the existing 19 literature as ponds are believed to have a cooling potential. The disparity could be attributed to the large surface area and depth of the pond, which acts as a heat sink during the day and releasing the absorbed heat in the night. The high heat capacity of water was another underlying reason for the water being warmer in the evening. The results generally inform the cooling potential of “moving” water features, with the water fountain and water wall producing a better cooling effect and thermal comfort acceptance. Perhaps an integration of both these water facilities could better facilitate the improvement of the thermal environment. People will continue to use the space under some tolerable level of discomfort, even if it lacks their preferred environmental diversity. The extent of this discomfort can only be assessed by considering people’s expectations, preferences, and acceptability thresholds. 2.6.3 Water Facilities and Their Effect through Simulation Study Kinouchi and Yoshitani (2001) did a simulation that modeled the urban environment in central Tokyo to project the cooling impact brought about by the employment of roof vegetation and the increase in water surfaces by 2015. Their results revealed that the maximum reduction in air temperature is estimated to be 0.5°C should the area of water surface increase by twofold. Comparatively, this reduced surface air temperature per unit area increase of water surface is greater than the double of that of roof vegetation, indicating that water surfaces might be substantially more effective in alleviating UHI than roof vegetation. 20 A simulation from another study comparing the surface temperatures of a water pond and asphalt revealed that, when the maximum surface temperatures were recorded (33.2°C for the water surface and 58.3°C for the asphalt surface), large differences from 4°C at 6:00 a.m. to 25°C at 1:00 p.m. were registered between the two surfaces (Robitu et al., 2003). This difference is due to water evaporation, as well as the materials properties of the two surfaces. Asphalt surfaces have very low reflectivity with an absorptive rate of 0.9, meaning it absorbs almost all the solar radiation to which it is exposed (Santamouris and Asimakopoulos, 1996; Robitu et al., 2003 ). On the other hand, the water surface has a lower value, with an absorptive rate of 0.7 (Bolz, 1973; Robitu et al., 2003). This could be the underlying reason for the water bodies giving out a substantial amount of heat at night as they absorb a sizeable amount of solar radiation in the day. However, water bodies (rivers, lakes, and ponds) do have the potential to cool the urban atmosphere as much as vegetation does (He & Hoyano, 2008). 2.7 Green Area Studies in Singapore In 2004, Wong and Yu studied green areas and an UHI for a tropical city. They observed a maximum difference of 4.01oC between the well-planted area and the CBD area. In 2005, they also observed that UHI mitigation measures had largely concentrated on the employment of primarily plants and green spaces to encourage evapotranspiration so as to curb the rising air temperatures in the urban areas. Chen and Wong (2006) observed that large urban parks could extend the positive effect to the surrounding built environment. Through the field measurement, the built environment, 21 which is close to park, has a lower temperature at an average 1.3oC. Thus, the more parks are built in an urban area, the lower the urban temperature will be. The air temperatures measured within parks have a strong relationship with the density of plants, as plants with higher leaf area indexes (LAIs) might cause a lower ambient temperature. The ENVI-met simulation indicated that a park has a significant cooling effect on surroundings during both the day and night. Wong and Jusuf (2007) conducted an ENVI-met simulation and observed that the ambient temperature of the NUS Master Plan 2005 could increase by about 1oC when it is completed, due to the reduction of the greenery rate from 55.1% in the current condition to 52.31%. The bare pavement between buildings without any greenery also contributes to the increase of ambient temperature. In 2008, Rajagopalan and Wong also studied on the microclimatic modelling of Singapore’s urban thermal environment to mitigate the UHI. The study verified the existence of the UHI effect in the present context of Singapore. The central building district (CBD) area showed the highest temperatures. The maximum temperature difference of 4oC was observed between the vegetated area and the CBD area. However, despite the promising measure to improve microclimatic conditions in urban areas through greenery contribution, the benefits of other alternatives should be considered, such as efficient water surface arrangement (Ichinose et al., 2008). This alternative, which has not received much attention compared to vegetation, points to the use of the evaporative effect of water as an alternative to cool the environment. Water ponds favoring the evaporative cooling were identified as one of the potential mitigations for UHI (Nishimura et al., 1998; Givoni and La Roche, 2000). 22 2.8 ENVI-met Simulation Environmental modeling has been a major component of the scientific approach in understanding and solving problems in complex environmental settings at the meso-scale simulations of climate change (e.g., Jacob, 2008). With resolutions of several kilometers, these simulations only display the climate within the city, which creates its own distinct urban climate very coarsely. Due to their specific albedo, roughness length, and soil sealing, cities create their own micro climate, mostly referred to as the UHI effect (Grimmond, 2006). As regional climate models predict heat waves to occur more often and are more intensive and longer lasting (e.g., Meehl and Tebaldi, 2004), it is necessary to study the effect of cities on heat waves in order to identify possible countermeasures. The environmental modeling system serves a purpose and has certain characteristics that are useful in understanding the micro-scale climatic behavior of building structures and landscape elements in the environment for this study. Beck et al. (….) offers three objectives for constructing and evaluating environmental models that are useful in understanding the micro-scale climatic behavior of building structures and landscape elements in the environment for this study: 1. Prediction of future behavior under various courses of action—namely, in the service of informing a decision (project development/evaluation, impact analysis building design regulation, planning regulations) 2. Identification of those constituent mechanisms of behavior that are crucial to the generation of a given pattern of future behavior but insufficiently secure in their 23 theoretical and empirical basis—namely, in designing the collection of further observations (future planning, development/redevelopment decisions) 3. Reconciliation of the observations of the past behavior with the set of concepts embodied in the model—namely, in the modification of theory and in explaining why a particular input disturbance of the system gives rise to a particular output response (the impact of different structural modifications in the environment) 2.9 Validation and Sensitive Analysis of ENVI-met Simulation ENVI-met software is a three-dimensional non-hydrostatic model for the simulation of surface–plant–air interactions in the urban environment. It has been widely used in the computer-aided design and evaluation of various urban planning cases (Bruse, 1998). However, in order to ensure that this software can be applied to the future sensitive analysis and other studies in a hot and humid climate, a validation assessment was carried out. Zhuolun et al. (2009) conducted a validation assessment of both the iterative and grid convergence. Thus, the ENVI-met results were compared to the previously discussed experimental data, demonstrating that, within the uncertainties of experimental data (more or less 0.7°C in air temperature, more or less 5% in relative humidity), the simulation results can meet the measured data of most spots. 2.10 Simulation Limitations ENVI-met has certain limitations. The tools to create the urban environment are limited to buildings, soils, water area, pavement materials, and trees or other vegetation. The albedo and thermal resistance of the building surfaces are constant and cannot be varied 24 (Emmanuel and Fernando, 2007). There are no tools to create any other objects, such as shade structures independent of the building blocks. Another significant limitation is that the building blocks have no thermal mass and only a single constant indoor temperature. ENVI-met cannot simulate water turbulence mixing so the use of water strategies is limited to still water bodies. Therefore, ENVI-met is unable to simulate fountains or water spray types of systems. Water bodies are inputted as a type of soil, and the processes are limited to the transmission and absorption of shortwave radiation (Bruse, 2007). 2.11 Knowledge Gap The literature indicate that UHI mitigation has been studied. It has been the most widely applied mitigation measure for achieving extensive energy savings through the temperature reduction of an area (Konopacki and Akbari, 2002). Water features have not received much attention compared to vegetation in tropical areas whereas the evaporative effect of water is seen as an alternative for cooling the environment. Some researchers have argued the need to aid in the cooling of the water bodies more than in the cooling effect produced by the greenery. Past studies have found that, in sub-tropical areas, water bodies can provide a significant cooling effect by lowering the ambient temperature by 4oC compared to areas without water bodies. In addition, water ponds favoring the evaporative cooling were identified as one of the potential mitigations for UHI (Nishimura et al., 1998; Givoni and La Roche, 2000). Yet Ken-Ichi (1991) and Givoni (1991) mentioned that evaporative cooling is arguably one of the most efficient ways of passive cooling for buildings and urban spaces in hot regions. Based on the 25 literature review, water bodies’ cooling effect might work better at a low temperature and in low humidity, as in the sub-tropical climate. In Singapore, some researchers have only studied the cooling effect of the water feature, showing a significant temperature reduction near water facility areas. This might suggest that, in Singapore, water can be one of the potential cooling factors on its surrounding environment. However, Singapore might not find the same result due to its humid conditions. In a very humid environment, the water does not evaporate very fast at all. In hot and arid climates, raising the humidity of the air has brought welcome relief through water bodies’ water cooling effect. Hence, the aim of the thesis is to first establish a relationship between water bodies’ cooling effect on a larger scale—namely, rivers, reservoirs, and bays—with the surrounding air temperature on a hot sunny day. Second, it evaluates the effectiveness of the different types of surrounding areas in helping the water bodies reduce heat from the thermal environment around it. Whilst evaporative cooling has been one of the most effective ways of passive cooling for architecture and urban spaces in hot regions since ancient times, it is more effective in hot and dry regions in terms of total amount of cooling, as the increase in humidity gives additional comfort. However, it can be equally effective in hot and humid regions in terms of the enhanced level in a thermal environment compared to severe summer conditions (Kimura, 1991). 26 2.12 Hypothesis Development Based on the experiments that have been conducted (further analyzing its effect) in a hot and humid climate like Singapore, which justifies the cooling potential of water features, Choo (2008), Jusuf et al. (2009), and Hui (2009) all looked at artificial water features on a smaller scale. In addition, studies of sub-tropical areas have demonstrated significant environmental cooling. Hence, this study focused on water bodies in Singapore, which has a larger scope of water area. Water bodies provide an efficient channel for evaporative cooling in tropical climates, which might suggest that a greater possible cooling effect level on the outdoor air temperature and could extend performance in tropical areas where it is hot and humid. This study will help further substantiate the existence of the water bodies in the design of the surrounding microclimate in outdoor places. Based on literature review, it has been found that: 1. Evaporative cooling is hypothesized to be especially well suited for climates where the air is hot and humidity is low. In highly humid climates, evaporative cooling might have a limited thermal comfort benefit beyond the increased ventilation and air movement it provides; 2. The evaporation process primarily depends on solar radiation performance; 3. Water features have a cooling effect on the surrounding environment regarding the distance to the outdoor air temperature compared to that in the absence of water features; and 4. The surrounding condition affects the performance of water bodies to produce the cooling effect to the surrounding air temperature. 27 The hypothesis of the study is that water bodies’ cooling effect stemming from the evaporation process could provide a cooling effect for the surrounding. It can lead to a reduction in the temperature of the surroundings. Dependent variables: 1. Air temperature 2. Distance from the area of the water bodies 3. Wind speed 4. Size/width of the water bodies 5. Surrounding area near the water bodies Independent variable: 1. Solar radiation 28 CHAPTER 3 3.1 RESEARCH METHODOLOGY Research Design The research strategy consists of collecting primary and secondary data. Primary data will be obtained from the results acquired from a field experiment and simulation conducted based on the ENVI-met results. Field-based research allows for a highly randomized sampling of subjects in changing urban conditions, thereby widening the scope of results for better understanding. Relevant information attained from books, journal articles, publications, press reports, and conference articles, together with internet websites, will be used as a secondary source of data. 3.2 Selection of Water Bodies A five-month period of field measurement was conducted in four water bodies in Singapore: Kallang River, Sungei Api-api River, Marina Bay, and Bedok Reservoir. 3.2.1 Kallang River The Kallang River is an urban planning area and subdivision located in the southeastern part of Singapore. The first measurement location is located within the vicinity of the Kallang MRT Station (north site of the MRT, along the Kallang Park Connector). The Kallang Park Connector is part of the Park Connector Network managed by the National Parks Board. The measurement location is also near the Kallang riverside park, which is famous for its water sports and serves the urban planning area of Kallang. 29 The measurement was conducted in an open space along the Kallang River, between Boon Keng Road and Kallang Road. The Kallang River’s open space is surrounded by a common green area and a few trees. Dense trees are located near the Kallang MRT (Mass Rapid Transport). Figure 3.1 Kallang Basin (Source: Author’s photo) 3.2.2 Sungei Api-api River A 5-hectare patch of mature mangrove forest was preserved during reclamation and development by maintaining tidal inundation: A rivulet was dug to connect the patch with Sungei Tampines. With its massive scale and mature shady trees surrounding it, the 14hectare marine pond is undoubtedly the main focal point of Pasir Ris Town Park (National Park, 2009). The park can be said to be located in a relatively rural area in the eastern part of Singapore. Wong (2000) stated that Pasir Ris registered a lower temperature compared to the densely populated or industrialized areas, which recorded a relatively higher temperature. Complications can be realized by studying this water feature with reference to its location. 30 Figure 3.2 Sungei Api-api river (Source: Author’s photo) 3.2.3 Marina Bay Figure 3.3 Marina Bay (Source: Author’s photo) Marina Bay is a bay near the Central Building District (CBD) area in the southern part of Singapore. The area is a 360-hectare development designed to seamlessly extend Singapore's downtown district and further support the city-state's continuing growth as a major business and financial hub in Asia. Reclaimed from the sea since the 1970s to provide room for the long-term expansion of the city, Marina Bay is planned to seamlessly extend from the existing CBD at Raffles Place. The development parcels at Marina Bay are based on an urban grid pattern and extended from the existing city grid network to ensure good connectivity. This grid framework has been developed to allow for the flexible amalgamation or subdivision of land parcels into 31 plots of different sizes, including larger land parcels to cater to buildings with large floor plates and offer maximum flexibility and efficiency for financial institutions. The 3.5-km long waterfront promenade linking the string of attractions at the Marina Centre, Collyer Quay, and Bayfront areas was completed in 2010. There are experiential strolls along the promenade around the bay. The promenade forms part of the 11.7-km waterfront route around Marina Reservoir, linking Gardens by the Bay, Marina Barrage, and the new Sports Hub. Walking along the Bayfront and into the Mist Walk brings down the ambient temperature, making a stroll along this stretch a breeze in the equatorial heat. 3.2.4 Bedok Reservoir Figure 3.4 Bedok Reservoir (Source : Wikipedia) Bedok Reservoir was constructed under the Sungei Seletar/Bedok Water Scheme, completed in 1986. The scheme involved the damming of Sungei Seletar to form a reservoir (Lower Seletar Reservoir), the creation of Bedok Reservoir from a former sand quarry between 1966 and 1972, and the construction of Bedok Waterworks. The earth excavated was used for East Coast Reclamation in the 1970s (HDB Annual Reports). Its unique feature was the construction of nine storm water collection stations to tap the storm 32 runoffs of the surrounding urbanized catchments. Eight of these collection stations are ponds in the new towns Yishun, Tampines, Bedok, and Yan Kit. 3.3 Instruments Used The major instruments used in the measurement were the HOBO data logger and weather station. The HOBO was used to measure relative humidity and ambient temperature for every particular location designated in the measurement. A weather station was used to measure wind speed and solar radiation for a reference measurement. To prevent the interference of solar radiation and obtain accurate ambient air temperature and relative humidity, every sensor was protected in a white painted wooden box with ventilation holes on both sides. These boxes were secured at a height of 2 m on lamp posts or trees nearby. The data loggers were pre-programmed to record the data continuously every 10 min. Localized weather data are good reference for the measurement. Therefore, a HOBO weather station was employed to monitor the weather condition. It was set up on a rooftop of HDB near to the Kallang River. The sensors were installed at the edge of the rooftop to avoid any possible influence from either the surrounding buildings or landscapes. Six weather parameters were measured:  Ambient temperature  Relative humidity  Wind velocity  Wind direction  Rain fall 33  Solar radiation 1. Weather station Figure 3.5 Weather station The weather station was used as the reference point. The weather data were logged into the HOBO Weather Station H21-001 with the following specifications: - Memory: 512K non-volatile data storage - Logging interval: 1 sec to 18 hours, user-selectable - Operating range: -20oC to 50oC - 10 sensor inputs, expandable to 15 with optional 1-to-2 sensor adapters - Built-in weatherproof communication port The station measured the following parameters: - Solar radiation (W/m2) - Ambient air temperature (C) - Relative humidity (%) - Wind direction (Degree) - Wind speed (m/s) 34 - Rainfall (mm) 2. H08-003-02 HOBO temperature / RH data logger and solar cover H08-003-02 was the main instrument used to measure the air temperature across the sites installed at the height of 2 meters above the ground. The specifications are as follows: - Temperature measurement range: -20C to +70C - Temperature accuracy: +- 0.2 at 21C - RH measurement range: 0-95% RH non-condensing - RH accuracy: +-3% RH; +-4% RH in condensing environments Figure 3.6 Hobo Data Logger 3.4 Data Selection The selected data for the study were chosen by analyzing climate data from the weather station. As can be seen from Table 1, eight typical hot day (clear days) and 12 typical cloudy/rainy day (cloudy days) for the Kallang study case area as well as nine clear days 35 and 13 cloudy days for the Sungei Api-api study case area were investigated to have a better understanding of the cooling effect of the water way. Table 1 Data selection Clear days Kallang waterway Cloudy/ rainy days 23rd and 24th of May; 1st and 3rd of June; 11th, 19thand 29th of July; 24th of August 2010 Sungei Api-api and 25th of June; 17th, 18th, 22nd and 26th of July; 7th, 10th and 20th of August 2010 23rd of May; 31st of May; 8th, 10th and 25th of 11th, 15th,25th and 29th of July; June; 8th, 16th, 18th, 22nd and 26th 12th, 25th and 28th of August; of July; 20th and 27th of August; 9th of September 2010 Marina Bay 19th, 22nd and 31st of May; 22nd 8th and 10th of September 2010 8th, 16th, 22nd , 23rd of of March ; 2nd, 3rd, 5th, 9th, 14th, 19th, 20th, 1st, 13th, 18th, 20th, 23rd, 26th, 27th 28th of March ; 19th, 21st, 24th of of April ; 3rd, 19th, 21st, 22nd of April ; 7th, 17th of May 2012 May 2012 Bedok Reservoir 23rd, 25th, 27th of March ; 1st, 13th, 6th, 11th, 19th, 21st of April ; 4th, 18th, 23rd, 27th of April ; 9th, 10th, 7th, 15th, 17th, 27th of May 2012 16th, 22nd, 30th of May 2012 36 3.5 Method of Data Collection The readings were collected between 7:30 a.m. and 7:30 p.m. The two time intervals were chosen so that the effect of water bodies can be examined during the noon period. The experiment was conducted over five months in each location during the stipulated time for data collection. Two measurement locations were conducted from May to September 2010 and the rest from February to June 2012. Ten-point measurement concurrent readings were taken during the time frame and at five points along the water bodies; the remaining readings were taken at a location 30m in each direction. One point was chosen as the reference point to determine if there are any differences in air temperature due to the presence of the water bodies. Readings of the environmental variables were recorded at ten-minute intervals. The reference points chosen were closely characteristic of the area of water bodies. Every 20 to 30 m away from the water bodies, tests were conducted at an average height of 2 m from the ground. This height was chosen for recording the air temperature as it is what individuals experience. Temperature readings were only recorded when the wind speed is below 2 m/s in order to eliminate and minimize the cooling effect brought about by the wind. 3.6 Data Processing The result from this experiment is expected to reveal the benefit of water bodies’ cooling effect on the surrounding air temperature for both clear and cloudy days. Therefore, the following steps were conducted before the analysis: 37 - Data selection: The data from rainy days are not included in the analysis. - Data classification: The data are classified according to clear and cloudy days. 3.7 Method of Analysis The analyses only consider the data during the day time on clear and cloudy days. A clear day is defined as an irradiance level above 900 W/m2 and a cloudy day is defined as an irradiance level of below 500 W/m2. The parameters for the collection of objective data include the air temperature, wind velocity, and solar radiation. The analyses, using Microsoft Office Excel software, were performed to calculate whether the samples come from the normal population. Furthermore, linear and polynomial regressions are applied to analyze the correlation of the variables. The dependent variables are air temperature and solar radiation performance. The independent variables are distance and variances in the surrounding microclimate near the water bodies. Correlation and regression analyses are conducted in order to determine to what extent water bodies have a beneficial cooling effect on the reduction of surrounding air temperature. Air temperature helps to register the impact the presence of the water bodies can bring to the microclimate (Boussoualim, 2000). Wind influences the air temperature by bringing about convective heat loss and evaporation. It can also help remove the humid air to improve comfort as well. This study also used rainfall data records to determine hot sunny days within the measurement period. Solar radiation influences individuals’ visual 38 perception and tactile senses. It is characterized by the received, emitted, and reflected light. Its effect is perceived independently of air temperature. 3.8 Point Location of Experiment The measuring points were chosen in order to determine the relative extent of the influence of the cooling effect from the nearby water bodies. Five to six measurement points along the waterway and five to six other points lined up away from the waterway were included. 3.8.1 Kallang River Figure 3.7 Points located along the Kallang waterway 3.8.2 Sungei Api-api River Figure 3.8 Points located along the Sungei Api-api waterway 39 3.8.3 Marina Bay Promenade and Marina Bay Promontory Figure 3.9 Points located along the Marina Bay Promenade, points MP1 – MP5 Figure 3.10 Points located along the Marina Bay Promenade, points MP6 – MP8 40 Figure 3.11 Points located away the Marina Bay Promontory 3.8.4 Bedok Reservoir Figure 3.12 Points located away the Bedok Reservoir 41 3.9 ENVI-met Simulation ENVI-met is a three-dimensional non-hydrostatic model for the simulation of surface– plant–air interactions within urban environments. It is designed for a micro scale with a typical horizontal resolution of 0.5–10 m and a typical time frame of 24 to 48 hours. This resolution allows for analyzing small-scale interactions among individual buildings, surfaces, and plants. In this research, ENVI-met was employed to compare the temperature profile conditions in Kallang, which cannot be observed in fieldwork study. The wider the river, the higher the wind speed, changing surrounding cover with or without vegetation (replaced with pavement cover) in the surrounding environment. The basic aim of this study is to evaluate the practical applications of the model and generate an innovative framework for its use. The research discusses the detailed constructs of a simulation modeling procedure in order to see how the water bodies’ cooling effect from the scenarios vary and to help urban planners in considering their open spaces, especially in water bodies’ areas, and planning strategies in advance. The methodology will follow the model’s own procedures, from establishing an information base to developing criteria for the design of the micro-scale (and macro-scale) environments. The initial boundary parameters were set according to real site conditions in the Kallang waterway, where the field measurement was conducted. Based on the preliminary analyses of weather data obtained from the field measurement, a clear sunny day was chosen: 29 July 2010. The air temperature, wind speed, and relative humidity settings are based on the 42 mean values of the weather data. The wind direction was set from the southwest to the northeast. Eighteen hours of simulation time were employed to allow the model to spin up and obtain the optimum data. Setting sections were determined based on the material properties of the water and grass park. The temperature was set similar to the outdoor air temperature in order to fully investigate the cooling effects of the water bodies. 3.9.1 Simulation Procedure in ENVI-met ENVI-met will be used to make a simulation based on the actual conditions in the Kallang River area for a comparison with existing measurement results. The scenarios with the water body area in Singapore, with a temperature profile of a normal summer day in Singapore (i.e., temperatures up to 30°C), representing a clear day in Singapore, were used for this study. Based on the situation in the Kallang waterway, the experiment replaced the actual width of the river to determine if an additional cooling effect occurs if the width of the river is enlarged. In order to identify the wind velocity effect on the water bodies’ cooling effect, the experiment enhanced the wind speed to be two times faster than the actual speed (1 m/s). The boundary conditions were further divided into more surrounding scenario profiles. Two more surrounding scenarios were used to simulate river environs with grass only and a river surrounded with pavement, regardless of the real condition, where the river is surrounded with a 3-meter foot path along the waterway, grass area, and some trees. 43 For those simulations, the 29th of July 2010 was selected as the simulation day due to its maximum solar radiation from the field measurement in the Kallang and Sungei Api-api Rivers, Singapore. For each case, westerly winds at 3 m/s 10 m above ground level were assumed to determine the effect of wind speed in the cooling performance. 3.9.2 Simulation Boundary Condition A hot day was chosen for the analysis because small increases in comfort deliver the greatest health benefits in thermally stressful conditions. Table 2 Basic input parameters to the ENVI-met model Location Tropic climate area (Singapore) Date, time of simulation 29th July 2010; 1800 simulation hours Initial wind 1 m/s and 2 m/s at 10 m above the ground from 350 deg. (slightly southwest to northwest) Boundary conditions Temperature (at 2500m) = 293K Specific humidity (at 2500m) = 3g/Kg Grid size 100 x 100 x 20; X-Y grid spacing, 6 m; Z grid spacing, 6 m Plants Trees: 10 to 15 m high, dense foliage, deciduous Grass: 5-cm grass covering Surfaces/soil profiles Pavement profile: 60 cm depth, then loam down to 2m. Main road cover: asphalt Soil initial conditions Temperature (-2m) = 290K 44 CHAPTER 4 OBJECTIVE DATA ANALYSIS This chapter discusses the general cooling capacity of four bodies of water in Singapore and investigates the capability of the different types of surrounding microclimates, which might improve the cooling effect of the water through a simulation study. In order to analyze the effect of water bodies on air temperature in a hot and humid climate, all the data collected were normalized and compared with the reference point stipulated at approximately 20–30 m away from the water bodies. Data collected from the experiments were processed in a Microsoft Excel worksheet and represented with graphs. The correlation tests revealed a positive relationship between air temperature and solar radiation; the air temperature increased with regard to distance. Wind was found to have a minimal effect on air temperature as data collection was restricted to when wind velocity was 1 m/s and below; hence, for the analysis, wind velocity as an influencing factor was eliminated. Temperature fluctuations can be attributed to humidity and/or solar radiation, as humidity has a significant impact on air temperature. 4.1 A Comparison of Four Measurements in Regards to the Distance In order to investigate the evaporative cooling effect of the four waterways, the average temperatures from different points were obtained and compared. The average daytime diurnal temperature was determined from a clear day, with an assumption of about 12 hours of daytime, from 7:30 a.m. to 7:30 p.m. Figure 4.1 shows a comparison of daytime average temperature (clear days) of points among the four water bodies investigated in this research. Each point in the simulation result represents one point in the field measurement study in four measurement locations. R represents point K8 along the Kallang waterway, A6 along 45 the Sungei Api-api waterway, point MP 9 along the Marina Bay, and point BC 9 along the Bedok Reservoir. P1 represents point K1 in Kallang, A1 in Sungei Api-api, MP 11 in Marina Bay, and BC 10 in Bedok. P2 represents point K2 in Kallang, A2 in Sungei Apiapi, MP 12 in Marina Bay, and BC 11 in Bedok. P3 represents point K3 in Kallang, A3 in Sungei Api-api, MP 13 in Marina Bay, and BC 12 in Bedok. 4.1.1 Away from the Water Bodies (Clear Days) Figure 4.1 Comparison of daytime average temperature (clear days) at points at a distance from the four water bodies Figure 4.1 shows that generally in the four waterways the average temperature near the water bodies in a tropic area is quite high, at about 28.6oC to 30.3oC. As can be seen, the four waterways showed an increase at point P1, which is located 20 to 30 m away from the water bodies. This suggests that, during the day, the closer the measurement is to the 46 waterway, the cooler the temperature of the area is and that temperature variations are relative to the distance. The diurnal difference in average temperature from the reference point and P1 was found in range from 0.2oC up to 0.4°C. Data collected in Kallang revealed that air temperature increases the further one moves from the reference points; hence, one can expect the water bodies to bring about a possible cooling effect in hot and humid climates, with enough solar radiation for the evaporation process. Furthermore, point P2 in Kallang, Sungei Api-api, and Bedok showed an increasing average temperature, whereas in Marina it showed a decreasing value. Air temperature are expected to be cooler closer to waterways; as such, the average temperature profile at the Marina Bay study case area seems to have unexpected trends. The result from most of the points in this case show an inconsistency compared to the average temperature profile at points near the other three waterways, although points P2 and P3 at Marina Bay are closer to the CBD area. These findings suggest that, at Marina Bay, the surrounding condition are likely to have more impact on the sensors’ readings, which cover up the cooling effect from the waterway. Point P3, located about 30 m beyond P2 (about 60 m away from the water bodies) also shows an increasing value for the three waterways, compared to P2. In Sungei Api-api River, the first three points are likely the only points that show a fairly evaporative cooling effect from the waterway. There is a slight increase in average temperature at the points further from the waterway. Looking at point R (a6-in Sungei Api-api) as the closest point to the waterway, the temperature difference at the two nearest points from the waterway is only 0.20˚C. 47 The Marina Bay case study shows a similar trend to the prior case studies: The further one moves from the water bodies, the higher the temperature increase becomes, although the temperature increase in Marina Bay overall is quite small. Figure 4.1 shows that Marina Bay’s diurnal average temperature recorded near water bodies (point R) experienced a cooling effect by an average air temperature of up to 0.15oC lower than further points (point MP11). This insight suggests that the cooling effect occurred along the Marina Bay, although some factors (e.g., high-rise buildings and other materials) might have affected the data-recording process. Point MP9, the closest measurement point to the waterway, shows 29.30oC as the average temperature; point MP11, which is 20 meters from the border of the waterway, showed an average temperature of 29.40oC; and at point MP12, the average temperature decreased to 29.20oC—lower than average temperature at MP9. Prior findings have suggested that a possible cooling effect caused by the wind near the road could have affected the temperature decrease at point MP12. At MP13, the average temperature decreased to 29.03oC—lower than point MP9, which is closer to water bodies. The possible cause for this is the high-rise buildings, which create wind tunnels that allow winds to blow at high velocity, thereby resulting in lower average temperatures. The results from Marina Bay differ from the other case studies, which affirmed the cooling potential of the water bodies. In Bedok Reservoir also shows a similar trend with prior case studies. The measurement record shows that the microclimate near the water bodies (Point BC9) experienced a cooling effect, with an average 0.25oC lower air temperature than the further point (Point BC10). This indicates that the environment is cooler near the water bodies. The temperature at Point BC10 increased by 0.25oC compared to Point BC9, which is also around 0.25oC 48 lower than Point BC11. The average temperature difference between the points is approximated to be 0.25–0.30oC for every point further from the reservoir where maximum solar radiation occurs. The temperature decreases at Point BC12; the possible cause could be the vegetation coverage nearby as shading and cooling effects from the vegetation cause temperature decreases lower than the other measurement points. The temperature difference for the Kallang River was about 0.3–0.4oC. For the Sungei Apiapi River, it was about 0.2oC. For Marina Bay, it is only up to 0.15oC, and Bedok is about 0.25–0.30oC for every 30-meter span away from the waterway. These findings suggest that evaporative cooling from water bodies could produce cooler air for the surrounding area, based on the distance, in a hot and humid climate. The next immediate points (point P4 and P5) show a decreasing value in all locations. This suggests that this measurement was most likely already influenced by the surrounding conditions. The pavement and mature trees nearby could be one possible reason for the inconsistent profile. On the contrary, the possible cause of the disparity for the Marina Bay study is its close proximity to a CBD (Commercial Building District) area in Marina Bay. 4.1.2 Away From the Water Bodies (Cloudy Days) Compared to the cooling capacity on clear days, on cloudy days (figure 4.2) the temperature profile for the four bodies of water show a fairly constant trend and do not indicate temperature changes at each measurement point during the day. There is no evaporative cooling effect found during these particular times. The temperature profile was shown to be unpredictable, with some points near the waterway (point R) being warmer 49 than the nearby point P1 while others where the evaporative cooling effect should be evident remaining constant. This suggests that the water bodies did not provide a cooling effect in terms of distance. In addition, producing the evaporative cooling effect from the water bodies seems to depend on the solar radiation exposure. On a rainy or cloudy day, the solar radiation is relatively low compared to clear conditions during daytime hours. Figure 4.2 Comparison of daytime average temperature (cloudy days) at points at a distance from the four water bodies 4.1.3 Along the Water Bodies (Clear Days) In addition to examining the cooling capacity at points away from the water bodies, the field measurement also studied points along the water bodies, which were shown to have a similar temperature profile during the day as the water bodies. 50 Figure 4.3 Comparison of daytime average temperature (clear days) of points along the water bodies Figure 4.3 shows that the average temperatures along each of the three measurements are in a similar range. Some points have higher temperatures than other measurement points, which might be due to the surrounding materials near the points. The average temperature in Sungei Api-api shows the lowest value of 27.5oC on a hot day. This temperature might have been affected by the nearby sea. The sea breezes could cause cooler air along the river. In Kallang, the average temperature was 28.7oC, 1.2oC higher than Sungei Api-api, while Marina Bay shows an average temperature of 29.3oC, which is 0.6oC higher than in Kallang. This difference in average temperature suggests that the width of the river as well as the surrounding cover can affect the temperature conditions near bodies of water. 51 4.1.4 Distance Effect with Polynomial Regression 4.1.4.1 Kallang River and Sungei Api-api River The polynomial regression approach is used to examine the extent to which combinations of two predictor variables related to an outcome variable, particularly in the case when the discrepancy (difference) between the two predictor variables is a central consideration (Linda et al., 2010). In this research the independent variable is solar radiation, and the dependent variables are temperature and distance. These variables are used to determine to what extent they can enable bodies of water to cool their surroundings. The data suggest that this approach has the potential to address the research objective related to reducing temperatures at a distance from water bodies. Three typical hot days were chosen (23rd of May, 11th and 29th of July 2010) from the list of clear days at Kallang and Sungei Api-api Rivers to examine the distance effect. The point nearest the waterways in both study case areas (points k8 and point a6) were used as the reference point to find the temperature reduction, as shown in figure 4.4 and figure 4.5. For the regression at both the Kallang River (figure 4.4) and the Sungei Api-api River (figure 4.5), a simple mathematical calculation was used to suggest that a reduction of the evaporative cooling impact occurs in the range of 0.2oC to 0.3oC for every 30 to 35 meters from the waterway. This finding, however, only applies during the day, with clear, hot, sunny conditions. The polynomial regression is shown in figure 4.4. In Kallang, the maximum cooling effect extends about 65–70 m away from the body of water. After that point, the cooling effect 52 decreases in value. In addition, the water cooling effect can further spread the cooling to approximately 120 m away. The Kallang profile had very few trees and was covered mostly with grass near the water. This environment might have affected the water cooling performance, suggesting that water bodies near grass-covered areas with several trees can allow for cooling performance to a significant distance. Figure 4.4 Correlation between temperature reduction and distance at Kallang Figure 4.5 Correlation between temperature reduction and distance in Sungei Api-api 53 For the Sungei Api-api River, the regression trend shown in figure 4.5 seems similar to that of Kallang. The maximum cooling effect from the waterway can reach about 70 m. The farther the cooling effect reaches is 135 m, which is further than at Kallang. Sungei Api-api is located near the sea, which might affect the river’s cooling performance as the sea breeze could create cooler air in the surroundings. 4.1.4.2 Marina Bay and Bedok Reservoir Using the same method as in Kallang and Sungei Api-api, four typical hot days (23rd of March, 18th of April, 23rd of April, and 22nd of May 2012) were chosen from the list of clear days at Marina Bay and Bedok Reservoir study case areas for the distance effect discussion. The point nearest the waterway in both study case areas (point MP 9 and point BC 9) was used as the reference point to find the temperature reduction. Figure 4.6 Correlation between temperature reduction and distance at Marina Bay Using a simple mathematical calculation, figure 4.6 and figure 4.7 indicate a reduction of about 0.1˚C to 0.3˚C in evaporative cooling as far as 35 m from Marina Bay and Bedok 54 Reservoir. However, this finding only applies during daytime and in clear conditions (i.e., solar radiation performance reaches above 900 W/m2). As shown in figure 4.7, in Bedok, the difference in air temperature reaches 0.25–0.35oC. The area around Bedok Reservoir is similar to that of Pasir Ris Park (Sungei Api-api), but it is not affected by a sea breeze like in Sungei Api-api. This cooling performance suggests that it can support the previous (i.e., 2010) measurement results in nearby greenery areas, meaning the cooling effect might have a better performance. Interestingly, Marina Bay also showed an increase for the temperature profile of the second point away from the waterway; however, it is a small value, with a temperature difference of less than 0.1°C compared to the reference point, as shown in figure 4.6. Figure 4.7 Correlation between temperature reduction and distance at Bedok Reservoir 55 4.2 Solar Radiation Effect on Water Bodies Cooling Performance 4.2.1 Kallang River Further investigation is needed to see the water bodies’ cooling performance due to the evaporation process based on the presence of solar radiation on a hot day. In the water cycle, as water evaporates, it cools the surrounding air temperature. Thus, if water bodies are exposed to solar radiation, the evaporation process could produce a cooling effect on the air. Figure 4.8: Comparison of diurnal average temperatures (clear days) at points away from the Kallang waterway Diurnal Average Temperature (clear day) 34.0 Temperature (°C) 33.0 32.0 31.0 30.0 k6 29.0 k7 28.0 k8 27.0 k9 26.0 k10 25.0 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Hour Figure 4.9: Diurnal average temperatures (clear days) at points along the Kallang waterway 56 Figure 4.8 shows that the temperature difference could easily reach 1.8°C at the hottest hour of the day, with point k3 as the warmest point (32.5°C) and point k1 as the coolest point (30.7°C). However, after point k3, the temperature profile shows a significant declining trend. One possible reason for this is that these points are influenced by the nearby trees and may not actually experience the cooling effect from the waterway. The average temperature at the points near the waterway, on the other hand, shows similar result. With a fairly constant trend, the average temperature difference is up to 0.1°C, as shown in figure 4.9. For a comparison with diurnal average temperature on clear days, figure 4.10 shows the diurnal average temperature on cloudy days. As can be seen, the diurnal average temperature shows is about 27.5oC. There is no difference among the points, which should increase the diurnal average temperature the further away from the water bodies, as shown in figure 4.8. On cloudy days, the humidity is higher. In addition, with less heat absorbed by the water vapor, the atmosphere cools rapidly. Thus, without the presence of solar radiation, water bodies do not produce a cooling effect due to the evaporation process. Temperature (°C) Diurnal Average Temperature (cloudy days) 34 32 30 28 26 24 ave near K1 K2 K3 0.00 3.00 6.00 9.00 12.00 15.00 18.00 21.00 K4 K5 Hour Figure 4.10: Comparison of diurnal average temperatures (cloudy days) at points away from the Kallang waterway 57 Diurnal Average Temperature (clear days) 34.0 33.0 Temperature (°C) 32.0 31.0 30.0 ave near 29.0 k1 28.0 k2 27.0 k3 26.0 k4 k5 25.0 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Hour Diurnal Solar Radiation (clear days) solar radiation (W/m²) 900.0 750.0 600.0 450.0 300.0 ave SR 150.0 0.0 0:00 3:00 6:00 9:00 12:00 Hour 15:00 18:00 21:00 Figure 4.11: Comparison of diurnal average temperature (clear days) at points along the Kallang waterway with solar radiation Further, as can be seen in figure 4.11, in order to investigate temperature profiles at the Kallang study case area, the diurnal average temperature for points near and away from the waterway are compared with the average diurnal solar radiation from the nearby weather station. The figure shows that, using the nearby points (points k1 – k4) as a reference, the meeting point of the diurnal graph of point k1 with other points can be said to be the start and end of the cooling effect of the waterway due to the intensity of the solar radiation. By 58 looking at the diurnal solar radiation figure, the cooling effect is likely start at around 8:30 a.m., when the solar radiation reaches around 150-200 W/m², and end at around 6:30 p.m., when the solar radiation is less than 100 W/m². 4.2.2 Sungei Api-api River Looking further at the diurnal average temperature of points away from the waterway (figure 4.12), points a3, a4, and a5 are found to be warmer compared to points a1, a2, and a6 during most of the daytime. As mentioned earlier, the former points are closer to the hard surface around the center of the park, which may influence the HOBO reading and cover the effect of the evaporative cooling from the waterway. Using point a6 (the closest point to the waterway, with relatively less influence from the beach and sea breezes) as the reference point, points a1 and a2 tend to have warmer average temperatures during the early part of the day and become slightly cooler toward the end of the day. The morning solar radiation seems to have a greater impact on the evaporative cooling of the waterway to start giving the cooling effect to its surrounding. Further analysis of the extent of evaporative cooling with solar radiation performance is described later. The hottest point during the day was found at point a3, with a maximum average temperature of 32°C. During the night, when there is no solar radiation, most of the points are found within the same range of average temperature (25°C–26°C). 59 Diurnal Average Temperature (clear days) 33.0 32.0 Temperature (°C) 31.0 30.0 29.0 a6 28.0 a1 27.0 a2 26.0 a3 a4 25.0 a5 24.0 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Hour Figure 4.12: Comparison of diurnal average temperatures (clear days) at points away the Sungei Api-api River Temperature (°C) Diurnal Average Temperature (cloudy days) 34 32 a6 30 a1 28 a2 26 a3 24 0.00 3.00 6.00 9.00 12.00 15.00 18.00 Hour 21.00 a4 a5 Figure 4.13: Comparison of diurnal average temperatures (cloudy days) at points away from the Sungei Apiapi River From figure 4.13, the Sungei Api-api River’s diurnal average temperature on cloudy days shows a similar trend as in Kallang (figure 4.10). The average temperature shows 26 oC. There is no indication of increasing diurnal average value at Sungei Api-api on clear days (figure 4.12). This result supports the previous finding from Kallang: Without the presence of solar radiation, water bodies’ cooling effect does not occur. 60 Diurnal Average Temperature (clear days) 33.0 32.0 Temperature (°C) 31.0 30.0 29.0 a6 28.0 a7 27.0 a8 26.0 a9 25.0 a10 24.0 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Hour Figure 4.14: Diurnal average temperatures (clear days) at points along the Sungei Api-api waterway and the beach As can be seen in figure 4.14, a similar trend occurs at points a8, a9, and a10, with an average of 32.5oC. On the other hand, point a7 shows a lower average temperature, about 31oC, than the same reference point (a6). The range in differences between these points shows an inconsistent value, where the results were expected to have similar average temperatures at all points. Points a7 to a10, however, are closer to the beach. It is believed that they are affected by sea breezes. These findings suggest that in the Sungei Api-api River area, the surrounding conditions are likely to have more impact on the sensor readings, covering up the cooling effect from the waterway. 61 Diurnal Average Temperature (clear days) 33.0 32.0 Temperature (°C) 31.0 30.0 29.0 a6 28.0 27.0 a1 26.0 25.0 a2 24.0 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Hour Diurnal Solar Radiation (clear day) solar radiation (W/m²) 900.0 750.0 600.0 450.0 300.0 ave SR 150.0 0.0 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Hour Figure 4.15: Comparison of diurnal average temperature (clear days) at points along the Sungei Api-api waterway with solar radiation Figure 4.15 shows the relationships between the potential evaporative cooling from the waterway and solar radiation. Using a similar analysis methods as with the Kallang waterway, the figure suggests that the evaporative cooling in the Sungei Api-api area started at around 8:30 a.m., when the solar radiation reached about 150 W/m², and ended when the solar radiation was lower than 150 W/m² at 6:30 p.m.. This finding further 62 supports the earlier finding that evaporative cooling seems to depend on the availability of solar radiation. 4.2.3 Bedok Reservoir Diurnal Average Temperature (clear days) Temperature ((oC) 35 33 31 BC 9 29 bc 10 27 BC 11 BC 12 25 1.0 3.0 5.0 7.0 9.0 11.0 13.0 15.0 17.0 19.0 21.0 23.0 Hour solar radiation (W/m²) Diurnal Solar Radiation (clear day) 900 750 600 450 SR 300 150 0 1.00 4.00 7.00 10.00 13.00 16.00 19.00 22.00 Hour Figure 4.16: Comparison of diurnal average temperature (clear days) at points along the Bedok Reservoir with solar radiation 63 Similar to the prior results, Bedok Reservoir shows an increasing value from the diurnal average temperature to points away from the reservoir on clear days (figure 4.16). As can be seen, the cooling effect happens at about 8:30 a.m., when solar radiation is about 150– 200 W/m2. The air temperature in the Bedok Reservoir area gradually decreases after reaching its peak at 3:00 p.m., with the average temperature decreasing by 0.25oC at point BC 10 and 0.20oC at BC 11. The amount of decrease, however, diminishes as it neared 6:00 p.m. This decrease is related to the dissipated heat stored during daylight. Water generally reflects an amount of solar radiation, but at the same time, it absorbs a substantial amount as well (Robitu et al., 2003). Temperature (°C) Diurnal Average Temperature (cloudy days) 33 32 31 30 29 28 27 26 25 24 BC 9 BC 10 BC 11 BC 12 1.00 6.00 11.00 16.00 21.00 Hour Figure 4.17: Diurnal average temperatures (cloudy days) at points away from the Bedok Reservoir During cloudy weather, the results from the Bedok Reservoir show a similar trend as Kallang and Sungei Api-api measurements. As shown in figure 4.17, on a cloudy day, the area has a low amount of solar radiation exposure and no cooling effect, resulting in no 64 differences in the diurnal average temperature between the points. This also suggests that, in the absence of solar radiation, the evaporation is non-existent. This phenomenon seemed to be in fair agreement with the conclusions of Huang et al. (2008). 4.2.4 Marina Bay Figure 4.18: Comparison of diurnal average temperatures (clear days) at points away from the Marina Bay Figure 4.18 shows the relationship between potential evaporative cooling from the waterway and solar radiation. A similar trend occurred at Marina Bay waterway. Using the same analysis methods as with the Kallang waterway, the data indicate that the evaporative cooling at Marina Bay starts at around 8:30 a.m., when the solar radiation reaches about 150 W/m², and ends when the solar radiation falls below 150 W/m² at about 6:30 p.m. This insight supports the prior finding that the evaporative cooling seems to depend on the availability of solar radiation. Marina Bay showed some distinct differences from other 65 cases. At the first point (MP 10) away from the reference (MP 9), the temperature increased by 0.15oC. At further points, it significantly decreased. This difference likely stems from the location in the CBD (Central Building District) area, which might affect the performance of the water bodies’ cooling effect. Further investigation, shown in figure 4.19, indicates inconsistencies in the diurnal average temperature along the waterway, ranging from 31.5oC to 33oC (average of 32oC). The range seems greater than at Kallang and Sungei Api-api, where similar diurnal average temperatures were recorded at points along the water bodies. This suggests that, in the Marina Bay case study, the surrounding conditions (i.e., pavement and building materials) affect the water bodies’ cooling effect. Temperature (oC) Diurnal Average Temperature (clear days) 34 33 32 31 30 29 28 27 26 00:00 MP 1 MP 2 MP 3 MP 4 MP 8 00:00 00:00 00:00 00:00 Hour Figure 4.19: Comparison of diurnal average temperatures (clear days) at points along the Marina Bay 66 Temperature (°C) Diurnal Average Temperature (cloudy days) 33 32 31 30 29 28 27 26 25 24 MP 9 MP 11 MP 12 MP 13 1.00 4.00 7.00 10.00 13.00 16.00 19.00 22.00 Hour Figure 4.20: Diurnal average temperatures (cloudy days) at points away from Marina Bay Unlike diurnal average temperatures on clear days at points away from the water bodies, on cloudy days (figure 4.20) the trend seems to be a lower average temperature and a similar diurnal temperature at each point. In addition, the area has a low amount of solar radiation exposure. This result supports the previous findings that, without the presence of solar radiation, there is no occurrence of water bodies’ cooling effect. 67 solar radiation (W/m²) Diurnal Solar Radiation (clear day) 900 750 600 450 SR 300 150 0 1.00 4.00 7.00 10.00 13.00 16.00 19.00 22.00 Hour Figure 4.21: Comparison of diurnal average temperature (clear days) at points along Marina Bay with solar radiation Figure 4.21 shows the relationship between the potential evaporative cooling from the Marina waterway and solar radiation performance. The cooling effect starts around 8:30 a.m., with 100–150 W/m2 of solar radiation. Solar radiation peaks at 3:00 p.m., with radiation reaching 900–950 W/m2 and a temperature of 32oC. Evaporative cooling ends 68 when the solar radiation falls below 150 W/m2 at 6:30 p.m. This finding is similar to the findings in 2010 from the Kallang and Sungei api-api waterway, indicating that evaporative cooling depends on the availability of solar radiation. Research has revealed that, in contrast to a shallow pond, a larger one will register higher temperatures (Sugawara et al., 2009), suggesting that the water bodies in Marina Bay should have a high temperature given its substantial breadth. In addition, Marina Bay might be somewhat ineffective in cooling the environment given the high heat capacity of its surroundings. Further, it could be a concern as the air temperature near the water could eventually be the same or even higher than the surrounding air temperature in nearby urban areas. This result supports Takahashi’s (2004) finding that increasing urban temperatures by diminishing green areas, increasing building density, and changing the street surface coating materials can lead to overheating due to the absorption of solar radiation on dark surfaces and the surrounding environment. 4.2.5 Solar Radiation Effect with Linear Regression Four typical hot days (23rd of March, 18th of April, 23rd of April, and 22nd of May 2012) and four cloudy days (19th April, 21st April, 7th May, and 17th May) were chosen from the list of clear days at Marina Bay and Bedok Reservoir to examine the correlation between solar radiation and temperature differences. 69 Solar Radiation and Temperature difference Temperature difference (oC) 2 1.5 y = 0.0009x + 0.1819 R² = 0.4758 1 cloudy day clear day 0.5 Linear (cloudy day) y = 0.0003x + 0.0983 R² = 0.36 Linear (clear day) 0 -200 0 -0.5 200 400 600 800 1000 1200 Solar Radiation Figure 4.22 Correlation between temperature reduction and solar radiation It can be seen in figure 4.22, the correlation between solar radiation and temperature reduction shows relative agreement between the experimentally measured data, suggesting a relatively good estimation given the good correlation between solar radiation and water bodies’ cooling effect on their surroundings. 4.3 Overall Field Measurement Findings Based on the data collected and analyzed from the four field measurements, water bodies can cool the air near the water bodies, lowering air temperatures in nearby environments. The presence of water bodies obviously improves the thermal environment on hot sunny days by cooling the air over a distance. The temperature decreases about 0.20–0.40oC for every 30 meters moved away from the water bodies. However, the cooling effect is limited 70 to its immediate surroundings. The regression results suggest that the water bodies’ cooling effect could extend to approximately 100 meters in its surroundings. Furthermore, the field measurements indicated that water begins to evaporate at about 9:30 to 10:00 a.m. as solar radiation increases. The highest temperature of the day occurred between 1:00 and 3:00 p.m. caused by the highest solar radiation of the day. The temperature subsequently dropped due to decreased solar radiation. The nearer to the water bodies, the cooler the ambient air temperature was. The air temperatures near the water bodies were found to be 1.7-1.8°K cooler than the further points. This result shows that the temperature drop is induced by the evaporation of water from the water bodies. However, the fourth study also found that solar radiation plays an important role. Evaporation is always associated with the heat transfer between water and air, meaning solar radiation is essential in the process as it provides the energy needed to change liquid water into water vapor. The results support the hypothesis established in the beginning of the paper as it is especially prevalent that water bodies provided a significant cooling effect in the day with enough solar radiation performance. 4.3.1 Additional Findings As already explained in the previous paragraph, solar radiation plays an important role in water bodies’ cooling effect performance through evaporation. It affects the temperature reduction over a distance. The drop in temperature shown in the results could be further analyzed in correlation to relative humidity, which plays an important role in temperature reduction. Solar radiation, which causes water evaporation, might result in increased humidity in the area. A higher humidity level contributes to a lower temperature. This 71 possibility might be used to assess the impact of what already appears to be very small cooling margins in this research. The correlation between relative humidity and temperature has the potential to fundamentally alter the conclusion that there is a small but significant benefit. The average relative humidity graph has been analyzed from selection clear days (Table 1) in Kallang and compared to the average temperature on clear days in Kallang along and away from measurement points. Water bodies’ evaporation during the day could increase the relative humidity amount in nearby areas. As can be seen in figure 4.23 and figure 4.24, the increase of relative humidity could produce a decrease of temperature near the water bodies and vice versa. Average temperature Comparison of average day time day temperature and relative humidity along the waterway 31 80 30.5 78 30 76 29.5 74 29 72 28.5 70 28 68 K6 K7 K8 K9 K10 Location Avg temp. tools along the river Avg humidity along the river Figure 4.23 Comparison between temperature and relative humidity along the waterway 72 Average temperature Comparison of average day time day temperature and relative humidity away the waterway 31 80 30.5 75 30 29.5 70 29 65 28.5 28 60 K8 K1 K2 K3 K4 K5 Location Avg temp. tools away the river Avg humidity away the river Figure 4.24 Comparison between temperature and relative humidity away from the waterway Figure 4.25 Correlation between average relative humidity and average temperature on clear days in Kallang Figure 4.25 shows that temperature and relative humidity have a negative correlation. This suggests that relative humidity affects water’s cooling effect performance. However, this research is most likely not focused on outdoor thermal comfort; hence, relative humidity has not been investigated in depth in this research. 73 CHAPTER 5 ENVI-MET SIMULATION ANAYLSIS Operative temperature is the air temperature equivalent affected by radiation and convection in the actual environment. It is technically defined as the uniform temperature of an imaginary enclosure in which an individual exchanges the latent heat by radiation and convection, as occurs in the actual environment (Turner, 2004). This chapter presents a simulation analysis in order to study the water bodies’ cooling effect unable to be investigated in the field measurement. Five scenarios are simulated to investigate the temperature profile of each scenario. Therefore, a parametric study, which is the crossreferences between the results obtained from the investigation and boundary data, is performed to observe the impact of the cooling effect for each scenario. The numerical simulations have been carried out using ENVI-met version 3.1 Beta2 (Bruse and Fleer, 1998). The field measurement area has been transformed in a 100 x 60 x 20-grid model grid with a resolution of 6 m x 6 m x 6 m, resulting in total area of 480 m x 240 m along the horizontal axis. This investigation examines the effect of the water bodies’ cooling process on the surrounding environment when exposed to different conditions. In addition, the study investigates to what extent the surrounding environment affects the cooling effect performance of the water bodies. According to the previous field measurement findings of four different regions near water bodies in Singapore, the temperature increases the further one moves from the water bodies, thereby proving the cooling effect from water bodies, although the reduction of temperature is minor. The water cooling effect occurs when the weather is clear and the solar radiation intensity is 900–1000 W/m2. Therefore, the 74 boundary condition of the simulation study, which was conducted on 29th July 2010, is based on the measurements from the Kallang waterway in Singapore. 5.1 Simulation Validation Figure 5.1 Comparison of diurnal average temperature of field measurement and simulation on Kallang River Figure 5.1 shows a comparison of diurnal temperatures on 29 July 2010 between field measurement results and simulation results. R represents point K1 in the field measurement study and point P5 in the simulation; both are reference points. A1 to A5 references K2 to K6 in the Kallang field measurement and P6 to P10 in the simulation study. Generally, as can be seen in figure 5.1, a similar trend occurs for both profiles. A1 shows a higher average temperature compared to R, which is about 30 m from the waterway. A2, located 75 60 m from the water, shows a higher average temperature than A1. A3 is the maximum temperature profile that could be reached in both studies. Thus, from R to A3, which is moving away from the waterway, shows the water bodies’ cooling effect on the surrounding area. However, both studies show a decline in A4 and A5, which might happen because green areas have their own cooling performance from trees’ shading; thus, greenery might have a cooling effect as significant as water bodies. A similar trend between both studies shows that the ENVI-met simulation could represent almost all spots of field measurement. Thus, the ENVI-met simulation could help conduct a parametric study in this research. The detailed study of each scenario is explained in the following sections. 5.2 Scenario 1: Kallang River Real Condition Figure 5.2 represents the condition at 10:00 a.m, which indicates the initial phase of the cooling effect from the water bodies. Generally, the simulation results show a similar trend as the field measurement. The evaporation of the water bodies has begun, thus the cooling effect is produced and extended into the surrounding area. The temperature profile indicates the water’s initial phase of cooling effect, ranging between 27oC and 28oC at this time. Points P1 and P2 indicate the temperature profile before reaching the water bodies’ area. P1 seems the warmest point of the temperature profile: 27.80oC in the morning. Even closer to the water bodies is P2 (20 m from the water bodies), which shows a lower temperature at about 27.5oC. Thus, the temperature decreases closer to the water bodies (point P3), and the heat from the surrounding area can be reduced in line with the presence of the water bodies. P3 shows a lower profile compared to P2 and a 0.20oC temperature difference at 27.30oC. As the wind blows from the southwest to the northeast, it brings some surrounding heat into 76 the water bodies. P4 represents the other outer line of the river, which represents the temperature of the cooling performance of the water bodies due to the evaporation process. The temperature is the lowest in the surrounding area, at about 27.10oC at this time. P5 to P10 represent the same point as Kallang field measurements (point K1 to K6), indicating the water’s cooling effect through the simulation. P5 (point K1 in the field measurement study) indicates the same temperature as the water bodies (P4). P6 (representing point K2) is 30 m further away from the water bodies and has a temperature of 27.25oC, 0.15oC higher than point P5. As can be seen in the section view, the temperature starts to increase from point P7. P7 to P10 are all at about 27.30oC, which is slightly higher than P6. This suggests that, in the morning, the evaporation of the water bodies has just begun, with the solar radiation values ranging from 150 W/m2 to 250 W/m2; thus, the cooling effect has not yet extending far into the surrounding still. The simulation result shows a similar temperature profile as the field measurement findings. 77 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 Figure 5.2 Temperature profile of Kallang River at 10.00 a.m. The section viewed in figure 5.2 also shows that the heat extends 10 m above the ground in the morning. Buildings and roads might produce significant heat in the surroundings. As can be seen, the cooling of the water starts to have an effect on the surroundings at point P5, as indicated by the lower height of the heat produced near the water bodies at about 4 m above the ground, which increases every 30 meters moving away from the water bodies, shown by points P6 to P9. 78 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.3 Temperature profile of Kallang River at 1.00 p.m. Figure 5.3 represents the temperature profile condition of the Kallang River at 1:00 p.m. The 1:00 p.m. profile was chosen because it represents the highest temperature value in the field measurement study, when the solar radiation reaches its highest value (above 900 W/m2). Generally speaking, the simulation result indicates that the evaporation of the water bodies generates a cooling effect, which is then carried by the wind into the surroundings. The temperature profile indicates the initial phase of cooling effect performance of the 79 water bodies, and the temperature shows a high range between 29oC to 30oC at this time. In the afternoon, P1 and P2 indicates the highest temperature profile for surroundings at about 30oC, suggesting that buildings and roads produce a lot of heat due to their high value of absorption of solar radiation. The section view also shows that the heat is reaching twice as high as in the morning: It is about 20 m above the ground. At point P8, as in the field measurement, the cooling effect from the water bodies seems lower. Nearby vegetation may be the reason for this result. Points P8 to P10, as in the field measurement study, interestingly show a similar trend in lower temperatures. Again, nearby vegetation might be the cause of this. This point leads to an assumption that water bodies give a similar amount of cooling effect compared to that produced by vegetation during clear weather. Figure 5.4 shows the temperature profile condition of Kallang River in the evening time. Namely, 5:00 p.m. was chosen as it represents the crossing diurnal temperature profile when the water bodies’ cooling effect has stopped. The temperature profile indicates the highest value of the initial phase of the cooling effect performance compared to morning and afternoon time profiles. The temperature shows a range between 29.5oC and 30oC at this time. As can be seen, points P1 and P2 are still the warmest points in the day (about 30oC). The section view shows that the heat is still reaching about 10 m above the ground, as in the afternoon, suggesting that buildings and roads might be releasing their heat from the solar radiation absorbed during the daytime. P3 shows a higher temperature value compared to P3 in the afternoon, suggesting that water bodies are releasing the heat they absorbed from the solar radiation during the day. Point P3, however, shows a lower temperature profile (29.80oC) compared to point P2; the temperature difference is about 0.30oC. This suggests that the air temperature near the water bodies tends to be lower than 80 the surroundings. Points P4 and P5 as references are 29.5oC, which is the highest air temperature in the morning and afternoon profiles of the day. P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.4 Temperature profile of Kallang River at 5:00 p.m. The section view shows that heat released from the rural area at 5: 00 p.m. (figure 5.4), is higher than at 1:00 p.m. (figure 5.3). Points P1, P2, and P3 indicate that the evening temperatures are about 0.7oC–1oC higher than the afternoon ones. In the evening, rural areas are releasing their heat absorbed during the day. The cooling of the water from 81 evaporation is starting to lower due to the minimum presence of the solar radiation. P4 shows the temperature of 29.5oC, which is 0.7oC higher than the afternoon time at the same point, indicating that water bodies are also releasing their heat from the solar radiation gathered during the day. Thus, the cooling effect is continues at this time profile, as shown by points P6 (29.65oC) and P7 (29.75oC), whose temperatures are still increasing. The temperature profile starts to lower at point P7, 1 point closer to the water bodies, compared to the afternoon (figure 5.3), which starts to lower the temperature at point P8. This suggests that, in the evening, water bodies spread their cooling effect to about 60 m away, about 30–50 m shorter than in the afternoon. Figure 5.5 Comparison of diurnal average temperature of Kallang River at three times Figure 5.5 shows the comparison of diurnal temperatures in Kallang for different times in one day. Based on the three modeling profiles, the amount of the cooling effect can be observed from the temperature difference between water bodies and the surroundings. 82 Generally, the trend of the graph shows that the temperature of the air entering water bodies is higher than the air leaving water bodies. The evaporation from the water bodies carries the cooling effect from the water bodies and spreads it to the surrounding area. As can be seen, the morning generally has lower temperatures (average of 27.5oC), with a presence of about 200–250 W/m2 solar radiation recorded in the field measurement for this time. In the afternoon, the average temperature is 29.30oC, almost a 2oC difference from the morning. At this time, solar radiation reaches more than 900 W/m2. This suggests that solar radiation plays an important role in the evaporation process that produces water bodies’ cooling effect. In the evening time, the average temperature is slightly higher than in the afternoon, with a lower solar radiation presence (250–300 W/m2) at this time. This suggests that the cooling effect will not be produced without enough solar radiation performance. On the other hand, a higher temperature profile at this time suggests that pavement material and also the water bodies themselves are releasing heat from the solar radiation absorbed during the day. The result of this scenario provides important insights related to the solar radiation and cooling performance of the water bodies from field measurement results. The three simulation profiles have shown that the trend of cooling performance of the water bodies depends on the amount of the solar radiation, which causes evaporation and produces the cooling effect. 83 5.3 Scenario 2: Kallang River with a Wider Waterway This scenario is based on the three profiles in scenario 1, with a modification for a wider waterway. This scenario tries to unravel the relationship between the size of water bodies and the amount of the cooling effect. The waterway used in this scenario is twice as wide as the original one. Basically, this scenario profile seems to have a similar pattern of the spreading of the cooling effect compared to Kallang’s real condition scenario using the original width of the Kallang River. The average temperature is lower for the wider river; however, the distance that the water cooling effect spreads seems lower compared to the original river width. As can be seen in figure 5.6, in the morning, the water bodies seem to produce their evaporation process, yet the cooling effect is not occurring at a distance from the water area. In tropical areas, the humidity level in the morning is quite high; therefore, when the water area is bigger, it produces more humidity, and the surrounding temperature tends to be lower. As in the previous scenario, points P1 and P2 also indicate a higher temperature profile. The temperature decreases as it nears the water bodies (point P3). This suggests that the heat from the surrounding area can be reduced in the presence of water bodies. The different pattern compared to the real condition scenario is that, in the morning, with a wider river, the water bodies seem to have a minimum spreading of the cooling effect. The area near the water is cooler than at about 30 meters away (P5 up to point P6), which averages 27oC. Between points P6 and P10, the surrounding area seems to have a similar temperature profile. Given Singapore’s high humidity level in the morning, adding water areas could produce more humidity and cooler surrounding temperatures. 84 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.6 Temperature profile of Kallang River with a wider river width at 10.00 a.m. Figure 5.7 shows in the 1:00 p.m. profile with the wider river scenario compared to real condition in the same time profile; a similar trend profile emerges in terms of the spread of the water cooling effect. The difference is in the average temperature profile as the wider river is 0.5oC lower than the real condition. The cooling effect extends only 50 meters, about 30 meters shorter compared to the real condition in the same timestamp. 85 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.7 Temperature profile of Kallang River with a wider river width at 1:00 p.m. In figure 5.7, P4 represents the air temperature above the water (29.08oC). In the immediate point (P5), the temperature increases by 0.15oC, which supports Sugawara et al.’s (2009) claim that, in contrast to a shallow pond, a larger one registers higher temperatures. In addition, in a tropical area, a larger water area could add to the humidity level of the surroundings. Point P7 shows a decreasing average temperature, suggesting no significant cooling effect spread by water bodies. Yet it could also signal that a larger water area might lower the average temperature because water bodies produce more evaporative cooling. In 86 addition, in the tropics, a larger water body area could produce more humidity, which cools temperatures. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 Figure 5.8 Temperature profile of Kallang River with a wider river width at 5:00 p.m. In figure 5.8, as in the real condition scenario, the average temperature is higher in the evening than in the afternoon. In this case, the wider river seems to produce a significant heat release after absorbing solar radiation during the day. The average temperature 87 (29.60oC) is about 0.60oC higher than in the afternoon. Thus, in general, a larger river could absorb a bigger amount of solar radiation exposure during the day and release a significant amount of heat in the evening time. On the other hand, a larger water area could provide a better cooling environment by increasing the humidity level in the evening time, as can be seen that releasing heat from surrounding vegetation produces a cooler environment. Figure 5.9 Comparison of diurnal average temperature of real condition and wider river on Kallang According to figure 5.9, a wider river creates more evaporative water, thereby leading to cooler average air temperatures. The average temperature of the wider Kallang could lower the surrounding air temperature by about 1oC compared to the actual river size. This suggests that, in hot and humid climates, more water could provide more evaporation and higher humidity levels, resulting in cooler air in the surrounding area. Therefore, a bigger area of water bodies (related to the bigger amount of evaporated water) increases the cooling effect from the water bodies. 88 5.4 Scenario 3 : Kallang River with Wind Speed of 2m/s The added parameter applied in this scenario model is the wind speed. As a lower wind speed decreases the amount of evaporation from the earth’s surface, in this scenario, the velocity of the wind increases to 2 m/s—twice as fast as the average velocity in the field measurement (about 1 m/s). In the morning (figure 5.10), generally the average temperature for the Kallang wind speed scenario is very low: on average, 25.20oC, which is 3oC lower than the real condition scenario. It is assumed that the evaporation process is lower due to the lower solar radiation exposure, yet the velocity of the wind itself could produce a cooler environment. As shown in points P5 to P10, the cooling of the water bodies is also assumed to have a minimum effect on the surroundings in the morning. As the wind speed blows from the southwest to the northeast, a higher wind speed in this model brings much of the rural areas’ heat from the southwest to the water bodies, shown in point P3 in each time profile in this scenario. P4 represents the water bodies, showing a temperature of 25oC. As mentioned before, in Singapore, the humidity level is very high, thus the temperature is low. Increased wind speed could reduce air temperatures to create a much lower temperature profile. This might be the reason for the average temperature at the reference point in the morning with a higher wind speed being a very low value compared to all scenarios. 89 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.10 Temperature profile of Kallang River with wind speed 2m/s at 10.00 a.m. In the afternoon (figure 5.11), generally the average temperature for the Kallang wind speed scenario increases to an average of 28.10oC, up to 3oC higher than the morning profile. This suggests that, in the afternoon, the surface material’s absorption during this time increases the average temperature, even when a higher wind speed occurs. 90 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.11 Temperature profile of Kallang River with wind speed 2m/s at 1:00 p.m. In addition, wind velocity influences the water bodies’ cooling effect when enough solar energy is provided, as shown in figure 5.11. Points P1 and P2 in this scenario also have a higher temperature, as they are influenced more by the urban area. P3 has a lower temperature. P4 as the reference point is 28oC in the afternoon with a higher wind speed. Compared to the real condition, it is about 1oC lower at the same time profile. Thus, with the same amount of solar radiation, the air temperature could be lower with a higher velocity of wind. In addition, with a higher wind speed, the cooling effect from the water 91 bodies extends to about 60 m away, 20–30 m less than the real condition in figure 5.2, which extends to 90 m from the water bodies. In addition, in the afternoon, with a high value of solar radiation, a wind speed of 2 m/s did not contribute much more to water bodies’ cooling performance than the 1 m/s wind speed. On the other hand, with a higher wind speed, trees’ shade produced more cool air. Points P5 and P6 showed insignificant increasing values compared to the water bodies’ air temperature. The temperature difference was up to 0.5oC, leading to the assumption that the wind velocity has less influence on water bodies’ cooling effect compared to other profiles. In addition, nearby trees could help cool the air (from the trees’ shade), as the wind blows faster. In some open spaces (lawn cover), the average temperature increased by up to 0.20oC. Thus, higher wind speeds could also help water bodies spread their cooling effect. At point P7, the average temperature decreased to 28.05oC; this continued at point P8, which had the same average temperature as the water’s surface. As previously suggested, with a higher wind speed, the temperature under the trees is producing significantly cooler air. Further, as with prior findings, the water bodies are assumed to have the same cooling effect as vegetation. He and Hoyano (2008) found that water bodies have the potential to cool the urban atmosphere as much as vegetation does. Nonetheless, a higher wind speed in this case helped spread the evaporation cooling from the water bodies to the surrounding areas. 92 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.12 Temperature profile of Kallang River with wind speed 2m/s at 5:00 p.m. In the evening (figure 5.12), the heat from the urban area (points P1 and P2, on average 28.85oC) could reach out over the water bodies (point P3) and bring the heat further until over the investigated area (point P5). Point P4, the reference point, shows a very low exposure of evaporative cooling from the water bodies. Almost all the air temperature above the water bodies is covered with the heat brought by the higher wind speeds from the urban area. This result supports Taseiko (2008) and Oke (1987), who found that some wind 93 patterns can transport pollutants/heat into an urban environment in their air quality studies. In addition, at this time, there is lower exposure to solar radiation. Furthermore, based on the simulation results in the wind speed scenario profiles, a higher velocity of wind causes lower diurnal average temperatures. On the other hand, the cooling performance extend a similar distance as water bodies’ cooling effect at a lower wind speed (given the same amount of solar radiation). The section view (figure 5.12) shows that, in the evening, the heat release reaches about 18 meters above the ground, about 5 meters higher than in the afternoon (section view-figure 5.11), which reaches up to 14 meters. On average, compared to the real condition case (figures 5.3 and 5.4), which only reaches about 10 meters above the ground, the releasing heat in the wind speed scenario is higher, suggesting that a higher wind speed might also lift the heat from the earth’s surface higher. Figure 5.13 Comparison of diurnal average temperature of Kallang real condition with 2m/s wind speed 94 Figure 5.13 shows a comparison of Kallang’s real condition and Kallang with a higher wind speed. Generally, the average temperature differs by almost 1.5oC between the two scenarios. In both results, the average temperature at all points shows a similar trend. The average temperature at P1 and P2 also shows a higher value, which decreases after reaching the water bodies (P4). The 2 m/s wind speed profile shows a constant trend (P5 to P10), compared to the actual wind speed model, which has an increasing value moving away from the water bodies. Thus, wind speed might also affect the surrounding covers near the water bodies (in this case, a vegetation cover). However, the higher wind speed could lower the surrounding air temperature near the water area. According to the literature review, a lower wind speed is one factor that could decrease the amount of evaporation from the earth’s surface. Thus, a higher wind speed could help the evaporation process over water bodies. 5.5 Kallang River with All Grass Covered Microclimate In this scenario, the simulation model is based on the real condition in Scenario 1 with the surrounding area near the waterway being replaced by an all grass cover. Similar conditions in the timestamp are applied in this scenario conducted at 10:00 a.m., 1:00 p.m., and 5:00 p.m. The scenario results suggest that the water bodies’ cooling effect could be enhanced with evapotranspiration from the greenery in the surrounding area; hence, the average temperature shows a lower value. Furthermore, surrounding areas with grass vegetation have a similar temperature as the temperature above the waterway. 95 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.14 Temperature profile of Kallang River with all grass surrounding cover at 10.00 a.m. In the morning (figure 5.14), the average temperature is generally 26.05o–26.65oC, which is quite low. The waterway has begun the evaporation process when the solar radiation is 150 W/m2, as recorded in the field measurement. Before reaching the water bodies, the temperature shows a higher value in the surrounding area, averaging 26.50oC (points P1 and P2); the temperature immediately drops after reaching the water bodies (26.30oC, point P3), and the lowest temperature is at point P4 (26oC), which is believed to represent the air 96 temperature above the water bodies. P5, as the reference point, already shows an increasing air temperature, at about 0.10oC higher compared to air temperature above the water bodies. As can be seen in figure 5.14, the air temperature profile shows a higher value at each point (point P6 to P10), increasing approximately 0.15oC for every span of 30 meters. P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.15 Temperature profile of Kallang River with all grass surrounding cover at 1:00 p.m. In the afternoon (figure 5.15), with solar radiation present, the water bodies are in the evaporation process and the temperature above the waterway is about 29.65oC (point P4). 97 In this case, interestingly it seems the water bodies spread a very low cooling effect to the grass area. P5 shows a 0.20o–0.30oC difference from P4, which only 5 m away from the waterway. Further, P6 also has minimal differences with the previous point, as the temperature is 29.95oC. From point P8, the temperature continues to decrease until the end of the investigated area, at 29.80oC. Thus, there is no significant temperature difference between the grass covered surrounding and the waterway. This result suggests that water bodies’ cooling effect has a low impact on the grass covered area since the area itself already provides a similar cooling effect compared to that of the surrounding. P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.16 Temperature profile of Kallang River with all grass surrounding cover at 5:00 p.m. 98 In the evening, figure 5.16 shows that evaporation of the waterway is decreasing and the temperature above the waterway is increasing (to almost 30oC) because of heat dissipation from the solar radiation exposure during daylight hours. As can be seen, generally the temperature of the grass cover is similar to that of the water bodies, suggesting that the grass area is not experiencing much heat release compared to a pavement covered surrounding area. 5.6 Kallang Waterway with All Pavement Covered Microclimate The last scenario is still based on the real condition by using similar timestamp profiles for the simulation model. The surrounding water way area is modified by the pavement surface. Generally, water bodies with pavement surroundings have an average temperature above the water surface of 29oC. The wind carries the cooling effect to the pavement area about 65 meters further from the border of the waterway (point P4), a shorter distance than the grass area cover and real condition scenarios. From point P8 in each timestamp, with average temperature of 31oC, it is believed that the average temperature already represents the pavement air temperature. In the morning (figure 5.17), this scenario generally shows a lower temperature profile— about 26oC on average, which is about 1o–2oC lower than the real condition scenario. This suggests that the pavement is still absorbing a little solar radiation that may not produce a significant heat amount in the morning. The morning humidity level in the tropical area might also create a cooler air temperature. 99 P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.17 Temperature profile of Kallang River with all pavement surrounding cover at 10.00 a.m. In the afternoon (figure 5.18), generally, the greenery on the west side of the waterway (P1 and P2) shows a lower temperature compared to the pavement area at points P6 to P10. In this case, P3 shows the same temperature as P4 (about 30oC), whereas in other scenarios P3 had a higher temperature than P4. P5, the reference point, has an increasing temperature about 0.20oC 5 meters away from the water bodies. P6, P7, and P8 also have significant temperature increases by about 0.20oC for every 30 meters beyond the water. This result suggests that the heat released from the pavement is more dominant than the cooling effect, 100 shown in points P8 to P10, which have a maximum 31.40oC temperature profile. In addition, the heat dissipation from the pavement during daylight seemed to cause a significant temperature increase. P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 Figure 5.18 Temperature profile of Kallang River with all pavement surrounding cover at 1:00 p.m. Generally, in the 5:00 p.m. profile (figure 5.19), the cooling effect of the waterway is relatively longer compared to the afternoon (figure 5.18), with an average temperature of 101 29.90o C. This occurrence is caused by heat dissipation from the pavement during the day. The pavement and its surrounding area have a high temperature due to the nature of pavement, being highly heat absorbent and demonstrating heat dissipation. The average temperature above the waterway surface is also relatively high (30.80 o C), as shown in point P4. This suggests that the waterway is beginning to dissipate heat; hence, evaporation has stopped. P1 P2 P3 P4P5 P6 P7 P8 P9 P10 Figure 5.19 Temperature profile of Kallang River with all pavement surrounding cover at 5:00 p.m. 102 Figure 5.20 Comparison of diurnal average temperature of Kallang River with different surrounding cover at 10.00 a.m. Figure 5.20 shows that, in the morning, the all grass cover scenario has a lower temperature compared to Kallang’s real condition. In tropical areas, the morning humidity level is quite high; the presence of greenery (lawn) increases the humidity level. Thus, the temperature will be significantly lower. Compared to the all pavement cover scenario, the average morning temperature is quite similar to the all grass cover scenario, indicating that not much heat absorption has occurred from the solar radiation of the surrounding area (building materials, road; in the field measurement 10:00 a.m. profile, solar radiation was about 200–250 W/m2). However, Kallang’s real condition scenario has an average lower temperature, compared to the field measurement. Figures 5.21 and 5.22 generally show that the real condition is the lowest temperature profile, compared to the all grass cover and all pavement cover. In the afternoon, solar radiation has a significant value (more than 900 W/m2). This suggests that, with the 103 presence of a high solar radiation value, the surrounding humidity will be lower, and the water area could have a maximum evaporation process to cool the surrounding area. Furthermore, the all grass cover has a bit higher temperature (on average 1oC higher) than the real condition, suggesting that in the real condition, with the presence of trees, the shade from trees will produce cooler air than the lawn area. Figure 5.21 Comparison of diurnal average temperature of Kallang River with different surrounding cover at 1:00 p.m. The average temperature of the 5:00 p.m. profile (figure 5.22) shows a higher diurnal average temperature value in all scenarios, suggesting that all material cover at this time produces a significant release of the heat gathered from solar radiation during the day. This result supports Huang’s (2008) fieldwork in Nanjing, which showed that water bodies were registered to be cooler than concrete areas in the day. Tree shade might have more potential to decrease air temperature compared to bare concrete or lawn areas. 104 Comparison of Kallang diurnal temperature 33 32 Temperature 31 30 29 all grass 28 all pavement 27 kallang condition 26 25 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 Location Figure 5.22 Comparison of diurnal average temperature of Kallang River with different surrounding cover at 5:00 p.m. 5.7 Overall Simulation Findings Evaporative cooling from the water bodies is suitable for areas with a high air temperature and low air humidity. In a high air humidity climate, evaporative cooling might be less applicable if it is not supported with an increased air flow. The benefits of such a climate model are useful in the urban planning of extreme desert or humid climates. Based on the five simulation scenarios in the Kallang waterway case, some insights related to the water evaporation can be discovered. Water evaporation depends on several factors, which are air humidity, air and water temperature, surface area of the water, the velocity of the wind over the water, and the amount of solar radiation exposure. Water bodies absorb the heat from their surroundings when they evaporate and produce a cooling effect. Then, the water molecules evaporate and leave the surface of the water 105 bodies. The field measurement results indicated that the cooling effect seems to depend on solar radiation to produce energy performance to produce evaporation. The heat from solar radiation exposure increases the energy absorbed in the water and speeds up the evaporation process. From the simulation experiments, the boundary condition attached to solar radiation conditions were sufficient for producing a cooling effect that is 900 W/m2. The wind increases the amount of evaporation as it carries the released vapor from the water bodies to reduce humidity. The occurrence of wind is caused by differences in air pressure, sending air to the lower pressure area. Increasing the wind speed increases the cooling effect from the water bodies, cooling the environment for a longer distance. However, wind speeds in Singapore are very limited, which might affect the extension of cooling performance benefits. This result suggests that wind speed is also an important aspect for the cooling effect of water bodies. Other findings obtained from the simulations showed that the surrounding area near the water bodies seems to affect the performance of water bodies in their efforts to cool the environment. The measurement result is enhanced if the surrounding area is covered by dense trees, as the cooling effect is influenced by evapotranspiration produced by the greenery in terms of distance. Interestingly, the result from Scenario 3, the wider the river width, did not result in a significantly greater cooling effect. The aforementioned insights lead to the following conclusions:  Water bodies contribute to the cooling effect performance in an environment, although the amount is not considerably high. 106  The surrounding areas of the water bodies affect the cooling effect performance of the water bodies.  The wind increases the cooling effect performance of the water bodies.  A wider river does not significantly increase the water cooling effect performance to its surrounding area in terms of distance.  Vegetation or solid materials (asphalt/pavements) in the surrounding area affect the cooling effect performance of the water bodies. 107 CHAPTER 6 6.1 SUMMARY AND CONCLUSION Summary and Conclusion In this study, experiments were conducted to identify any air temperature differences in Singapore’s water bodies on a clear day using a reference point stipulated on average every 30 meters away and along the water bodies. In order to define and develop the source of methodology, the objectives were establishing in Chapter 1, including observational parametric studies to evaluate the trends of cooling extended by the water bodies that should be defined by the simulation using ENVI-met experimental models. The main conclusions drawn from this thesis are detailed under each of those objectives in the following subsections. Objective 1: To determine the cooling effect and benefit of water bodies on their surrounding microclimate in hot and humid climates. A review of the current literature in Chapter 2 identified that water features (water wall, fountain) in Singapore have a cooling effect on their surroundings—namely, about 0.5oC at 100 meters from water area. The field measurement and analysis data in Chapter 4 considered the effect of typical activities on water bodies (river, reservoir) in Singapore. The main conclusions drawn from these studies generally indicate that the waterways in the four locations clearly improve the thermal environment by cooling the air over an extended distance. The cooling effect, however, was limited, and the temperature drop was found to 108 be relative to the distance from the waterway. The studies found that the air temperature increases an average of 0.2°C for every 30 meters moved away from the water bodies, to a maximum of about 100 meters, until it is the same average temperature as the surrounding environment. Thus, a water body could reduce the surrounding area’s temperature on a hot sunny day. The high humidity climate in Singapore and the low wind conditions might be one of the possible reasons for the increased temperatures. Kallang River showed a higher capability to reduce air temperature than the other areas studied, with a 0.3oC temperature variation every 35 meters from the water bodies. The regression analysis showed that the water bodies reached the maximum cooling effect at about 125 meters away. Marina Bay, however, recorded a higher temperature than the reference point during the day. This conflicted with the literature review results that indicated that water bodies were believed to have a cooling potential. The disparity could be attributed to the large surface area and the location of the bay chosen, which gathered heat during the day from the nearby CBD area. This result supports the work of Sagan et al. (1979) and Myers (1992), who found that land cover change has a significant impact on climate. Objective 2: To determine the possible impact of water bodies’ cooling effect on the air temperature on a hot sunny day. According to the literature review, evaporation from water occurs when molecules of the water have a higher amount of energy than is needed and turn into as by colliding with 109 other molecules. Molecules turn into a gas when the temperature is nearing the boiling point with help from the solar radiation absorbed by the water. This is known as evaporation. In this study, experiments were conducted to identify any air temperature differences at the water bodies in comparison to a reference point stipulated at 20 to 30 meters away from the waterways. The results revealed that, in all experiments, the presence of water bodies in Singapore displayed potential cooling effects on the surrounding areas on a clear day with a presence of solar radiation. The cooling performance from the water bodies obviously benefits in lowering the surrounding air temperature of the nearby environment in a hot and humid climate. However, the cooling effect is limited to the waterway’s immediate surroundings, compared to the sub-tropical areas, where the temperature reduction could reach about 4oC in the presence of water bodies. The results generally revealed that water bodies in a hot and humid climate produce a significant cooling effect to the surrounding area on a hot sunny day. Looking at the diurnal profile of the field measurement in four locations, the point of reference near the water bodies showed a greater likelihood of higher air temperature than other reference points further from the water bodies. This result shows that the temperature drop is induced by the evaporation of water from the water bodies. However, the fourth study also found that solar radiation plays an important role in providing a better temperature profile. Solar radiation provides the energy needed to change liquid into water vapour in the evaporation process. Evaporation is always associated with the heat transfer between water and air. Consequently, solar radiation is essential in the process, as it provides the energy needed to change liquid water into water vapor. 110 Objective 3: To identify different types of surrounding area near the water bodies in order to investigate their impacts on the water bodies’ cooling effect. Air temperature near water bodies is different from that over the land due to different properties. Water is a transparent medium, and land is opaque. Water allows short-wave radiation transmission to a considerable depth, leaving the surface layers cooler. A cooler water surface results in cooler air temperatures above it. Normally thought of as a landscaping area, potential cooling effects of water bodies are often neglected. Hence, the simulation study tried to find the extent of cooling benefits from vast waterways in the tropical climate of Singapore. The ENVI-met simulation results generally related to the field measurement results, although the formed indicated a modest cooling effect. Water bodies are notably able to influence the surrounding thermal environment. The difference in air temperature due to water evaporation through enough solar radiation input during the day is evident. Water evaporation produces a cooling effect in the surrounding area near water bodies. Water bodies with all grass surroundings produce the best results in terms of the cooling effect. However, the simulation also suggested that there might be other parameters or factors that are likely to have a stronger influence than water bodies and nearby vegetation. All these findings suggest that water bodies in Singapore should be considered as a source for improving the microclimatic conditions in outdoor spaces. 111 6.2 Limitations of the Study This research was completed with limitations and constraints, especially when conducting the field experiment. Due to the limited studies on the impact of water bodies on a hot and humid microclimate, the existing literature was limited in terms of information and inference from relevant papers and reports. The experiments were conducted with readings recorded from only four areas surrounding water bodies in Singapore: Kallang and Pasir Ris Rivers in 2010 and Marina Bay and Bedok Reservoir in 2012. Pasir Ris is located near the sea, and Marina Bay is located near a CBD area; therefore, they might not reflect the conclusive results of the potential cooling effects of water bodies in tropical areas. The experiment was conducted with readings recorded from only several points of the water bodies, meaning the results might not reflect conclusive findings of the potential cooling effect by the water facilities. Instead, the precision of the data would have been improved with temperatures taken from numerous water body locations in Singapore, especially if the scale of the bodies of water were large. In addition, the types of materials near each water body were not taken into account, although they might affect the rate of absorption of the solar radiation, which will in turn affect the recorded air temperature. In addition, the types of measurements used at each water body were taken nearly to the greenery area, which might affect the rate of absorption of the solar radiation and, in turn, affect the air temperature recorded. The cooling effect of the greenery area could also be considered in the recordings. 112 Data from the ENVI-met simulation experiments were gathered in the hopes that the results would provide a better conclusion in assessing the effects on the water bodies’ environments in different scenarios and conditions around the water bodies. The goal was to compare the results related to temperature increases among the three scenarios of the simulated water area to determine which features provide a greater cooling effect. However, these experiments relied on limited information because the water body in this experiment was not moving water, although in reality it is a river (i.e., moving water). 6.3 Recommendations for Future Work Based on this work, in order to assess urban environments, urban heat islands, and climate changes in Singapore, thereby developing general guidelines for urban development based on water bodies’ potential benefits, the following areas were identified for further research to enlarge the scope of this work as well as identify opportunities for linking urban design and climate. 1. The accuracy of the data will increase with temperature comparisons if readings are taken from a greater variety of water bodies in Singapore to determine if the cooling effect of water is applicable in a hot climate with a high humidity profile. 2. Qualitative or quantitative surveys for the area around the water bodies should be conducted. Having determined the cooling capacity of the water bodies, in the future, research needs to examine humans’ perceptions of its existence in terms of thermal sensation, comfort level, humidity, and acceptance of the environmental condition around it. Thermal sensation depends on the skin temperature while 113 thermal comfort is dependent on the physiological state of an individual. This aspect is important as the people will be the ones using such facilities; understanding the thermal comfort condition in relation to the presence of a body of water will further enhance facility planning. 114 BIBLIOGRAPHY Asawa, T., Hoyano, A., Yamamura, S., Asano, K., Matsunaga, T., & Shimizu, K. (2000). Passive methods for creating good thermal environments in outdoor space. In K. Steemers, & S. Yannas, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp. 477-482). London: James & James (Science Publishers) Ltd. Boussoualim, A. (2000). Towards a method of characterisation of the link between climate, urban morphology and user's behaviour in external public space. In S. Y. Koen Steemers, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp. 459-464). London: James and James (Science Publishers) Ltd. Brager, G. S., & de Dear, R. J. (1998). Thermal adaptation in the built environment: a literature review. Energy and Buildings , 27 (1), 83-96. Bruse, M., H. Fleer, 1998: Simulating surface-plant-air interactions inside urban environments with a three dimensional numerical model. Environ. Modell. Softw. 13, 373384. Bruse, Michael, ENVI-met website (n.d.) retrieved on September 16, 2007 from: http://www.envimet.com/ Ca, V. T., Asaeda, T., & Abu, E. M. (1998). Reductions in air conditioning energy caused by a nearby park. Energy and Buildings , 29 (1), 83-92. Changnon, S. A., Kunkel, K. E., & Reinke, B. C. (1996). Impacts and responses to the 1995 heatwave: A call to action. Bulletin of the American Meteorological Society , 77 (7), 14971505. Chen Yu and Wong Nyuk Hien. (2006). Thermal benefits of city parks. Energy and Buildings, 38, 105-120. Chen, X.-L., Zhao, H.-M., Li, P.-X., & Yin, Z.-Y. (2006). Remote sensing image-based analysis of the relationship betweenurban heat island and land use/cover changes. Remote Sensing of Environment , 104 (2), 133-146. Choo, S.E (2008). Influence of Sky View Factor and Water Features on Air Temperature. Unpublished undergraduate dissertation, School of Design and Environment, National University of Singapore. E.L. Krüger, D. Pearlmutter, The effect of urban evaporation on building energy demand in an arid environment, Energy and buildings (2008) 115 Emmanuel R, Fernando HJS, Urban heat islands in humid and arid climates: role of urban form and thermal properties in Colombo, Sri Lanka and Phoenix, USA, Climate Research, Volume 34, Issue 3, September 2007, pages 241-251. Energy and Resources Institute, Institut Catalá d'Energia, Asia Urbs Programme. (2004). Sustainable Building Design Manual: Sustainable building design practices. New Delhi: TERI Press. ENVI Software. http://www.ittvis.com/envi/index.asp Forwood, B., Tadepalli, S., & Hayman, S. (2000). Modelling thermal adaptation and thermal preferences in outdoor urban environments. In S. Y. Koen Steemers, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp. 577-580). London: James and James (Science Publishers) Ltd. Fowler, T. (2008, July 1). Urban Heat Islands: the other type of island in the Japanese Archipelago. Retrieved September 18, 2009, from Heso Magazine: http://hesomagazine.com/2008/07/urban-heat-islands-the-other-type-of-island-in-thejapanese-archipelago/ Gartland, L. (2008). Heat islands: understanding and mitigating heat in urban areas. London: Earthscan. Grimmond, C.S.B., 2006: Progress in measuring and observing the urban atmosphere. Theor. Appl. Climatol. 84, 3-22. Givoni, B. (1991). Impact of planted areas on urban environmental quality: A review.Atmospheric Environment. Part B. Urban Atmosphere , 25 (3), 289-299. Givoni, B., & La Roche, P. (2000). Indirect Evaporative Cooling with an Outdoor Pond. In S. Y. Koen Steemers, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp. 310-311). London: James & Janmes (Science Publishers) Ltd. Guerra, J., Molina, J., & Cejudo, J. (1991). Thermal Performance of Water Ponds; Modelling and Cooling Applications. In S. Álvarez, Architecture and urban space: proceedings of the Ninth International PLEA Conference, Seville, Spain (pp. 625-630). Kluwer Academic Publishers. Henderson-Sellers, A. and McGuffie, K. (1987). A Climate Model Primer. Wiley, New York, pp.217. He, J., & Hoyano, A. (2008). A numerical simulation method for analyzing the thermal improvement effect of super-hydrophilic photocatalyst-coated building surfaces with water film on the urban/built environment. Energy and Buildings , 40 (6), 968-978. 116 Hiroshi, N., Tatsuo, O., Haruhisa, S., & Kazukimi, S. (2005). Research on the cooling effect by the water film formation on the surface of an outer coated with the superhydrophilic photocatalyst. Journal of Environmental Engineering , 596, 109-115. Hitoshi, T., & Kuniaki, A. (2006). Actual survey of thermal environment in building for experimental use that employs photocatalyst-based exterior materials. Journal of Environmental Engineering , 606, 31-36. Honjo, T., & Takakura, T. (1990-1991). Simulation of thermal effects of urban green areas on their surrounding areas. Energy and Buildings , 15 (3-4), 443-446. Huang, L., Lia, J., Zhao, D., & Zhu, J. (2008). A fieldwork study on the diurnal changes of urban microclimate in four types of ground cover and urban heat island of Nanjing, China. Building and Environment , 43 (1), 7–17. Ichinose, T., Lee, L.-T., Matsumoto, F., Shiraki, Y., & Harada, I. (2009). Mechanism of Mitigation of Atmospheric Environment by a Large Restoration of Inner-city River. Ichinose, T., Matsumoto, F., & Kataoka, K. (2008). Counteracting Urban Heat Islands in Japan. In P. Droege, Urban energy transition: from fossil fuels to renewable power (pp.365-379). Amsterdam: Elsevier. Ichinose, T., Shimodozono, K., & Hanaki, K. (1999). Impact of anthropogenic heat on urban climate in Tokyo. Atmospheric Environment , 33 (24-25), 3897-3909. Jacob, D., H. Göttel, S. Kotlarski, P. Lorenz, K. Sieck, 2008: Klimaauswirkungen und Anpassung in Deutschland – Phase 1: Erstellung regionaler Klimaszenarien für Deutschland, Climate Change 11/08, http://www.umweltdaten.de/publikationen/fpdfl/3513.pdf Japan for Sustainability. (2003, September 27). Japan for Sustainability. Retrieved October 15, 2009, from Let's Cool Down Tokyo! -- Edo-Period Sprinkling Campaign: http://www.japanfs.org/en/pages/025368.html Jusuf, S.K., Wong, N.H., 2nd PALENC Conference and 28th AIVC Conference on Building Low Energy Cooling and Advanced Ventilation Technologies in the 21st Century, September 2007, Crete island, Greece. Jusuf, S.K., Wong, N.H., Hagen, E., Anggoro, R., Yan, H., 2007. The influence of land use on the urban heat island in Singapore. HabitatInternational 31, 232–242. Jusuf, S. K., Hien, W. N., & Syafii, N. I. (2009). Influence of Water Features on Temperature Condition in Hot Humid Climate. Bangkok, Thailand. 117 Ken-Ichi, K. (1991). Evaporative cooling effects in hot and humid urban spaces. In Alvarez, & Servando, Proceedings of Architecture and Urban Space, Ninth International PLEA Conference (pp. 631-636). Sevile, Spain: Kluwer Academic Publishers. Kiehl, J.T. (1992). Atmospheric general circulation modeling. In: K.E. Trenberth (Editor), Climate System Modelling. Cambridge University Press, Cambridge, pp. 319-370. Kim, K. R., Kwon, T. H., Kim, Y.-H., Koo, H.-J., Choi, B.-C., & Choi, C.-Y. (2009). Restoration of an Inner-City Stream and Its Impact on Air Temperature and Humidity Based on Long-Term Monitoring Data. Advances in Atmospheric Sciences , 26 (2), 283– 292. Kim, Y.-H., Ryoo, S.-B., Baik, J.-J., Park, I.-S., Koo, H.-J., & Nam, J.-C. (2008). Does the restoration of an inner-city stream in Seoul affect local thermal environment? Theoretical and Applied Climatology, Volume 92, Numbers 3-4 / May, 2008 , 92 (3-4), 239–248. Kimura, K.-I. (1991). Evaporative Cooling Effects in Hot and Humid Urban Spaces. In S. Álvarez, Architecture and urban space: proceedings of the Ninth International PLEA Conference, Seville, Spain (pp. 631-636). Netherlands: Kluwer Academic Publishers. Kinouchi T, Yoshitani J (2001, December) Simulation of the urban heat island in Tokyo with future possible increases of anthropogenic heat, vegetation cover and water surface. Presented at the Proceedings of the 3rd international symposium on environmental hydraulics, Tempe, Arizona, USA. Konopacki S, Akbari H, 2002. Energy savings for heat island reduction strategies in Chicago and Houston (including updates for Baton Rouge, Sacramento, and Salt Lake City). Draft Final Report, LBNL-49638, University of California,Berkeley Li, R. M., & Roth, M. (2009). Spatial Variation of the Canopy- Level Urban Heat Island in Singapore. Yokohama, Japan. Manley, G. (1958). On the frequency of snowfall in metropolitan England. Quarterly Journal of the Royal Meteorological Society , 84 (359), 70-72. M.B. Beck et al., 1993. Construction and Evaluation of Models in Environmental Systems in Modeling and Change in Environmental Systems, pub John Wiley & Sons Ltd., (ed. A.J. Jakeman et al.). Meehl, G.A., C. Tebaldi, 2004: More intense, more frequent, and longer lasting heat waves in the 21st century. Science 305, 994-997. Memon, R. A., Leung, D. Y., & Liu, C. (2008). A review on the generation, determination and mitigation of Urban Heat Island. Journal of Environmental Sciences , 20 (1), 120–128. 118 MEWR. (2006). The Singapore green plan 2012. Ministry of Environment and Water Resources. Singapore. Mochida, A. (2009). Japanese Experience of technology and policy for urban heat-island mitigating. Proceedings of International Conference on Urban Heat-Island Mitigation. Munn RE, Hirt MS, Findray BF (1969) A climatological study of the urban temperature anomaly in the lakeshore environment at Toronto. J Appl Meteor 8: 411–422 Myers, N. 1992. The Primary Source: Tropical Forests and Our Future, 2nd edition. W.W. Norton, New York, 416 pp.Takahashi, K., Yoshida, H., Tanaka, Y. Aotake, N. and Wang, F. (2004). Measurement of Thermal Environment in Kyoto City and Its Prediction by CFD Simulation. Energy and Buildings, 16, 771-779. National Park. (2009). Visitor Info. Retrieved August 20, 2009, from Pasir Ris Town Park: http://www.nparks.gov.sg/cms/index.php?option=com_visitorsguide&task=parks&id=27&I temid=73 Nikolopoulou, M., & Steemers, K. (2000). Thermal Comfort and Psychological adaptation as a Guide for Designing Uraban Spaces. In S. Y. Koen Steemers, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp.565-570). London: James & James (Science Publishers) Ltd. Nishimura, N., Nomura, T., Iyota, H., & Kimoto, S. (1998). Novel water facilities for creaton of comfortable urban micrometeorology. Solar Energy , 64 (4-6), 197-207. Priyadarsini, Rajagopalan; N.H. Wong; K.W.D. Cheong (March, 2008). Microclimatic modeling of the urban thermal environment of Singapore to mitigate urban heat island. Solar Energy, 82 (2008) 727–745. Ray E. Bolz, G. L. (1973). CRC handbook of tables for applied engineering science. UK: CRC Press. Robitu, M., Inard, C., Musy, M., & Groleau, D. (2003). Energy Study of Water Ponds and its Influence on Building Energy Consumption., (pp. 1123-1130). Eindhoven, Netherlands. Robitu, M., Musy, M., Inard, C., & Groleau, D. (2006). Modeling the influence of vegetation and water pond on urban microclimate. Solar Energy , 80 (4), 435-447. Rosenfeld, A., Akbari, H., & Romm, J. J. (1998). Cool communities: Strategies for heat island mitigation and smog reduction. Energy and Buildings , 28 (1), 51-62. Sagan, C., Toon, O. B. and Pollack, J.B. 1979. Antropogenic albedi changes and the earth’s climate. Science, 206: 1363-1368. 119 Santamouris, M., & Asimakopoulos, D. (1996). Passive cooling of buildings. London: Earthscan. Santamouris, M., & Mihalakakou, G. (2000). Modeling the heat island phenomenon in urban areas using neural networks. In K. Steemers, & S. Yannas, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp.431-435). Earthscan. Santamouris, M., & Mihalakakou, G. (2000). Modeling the heat island phenomenon in urban areas using neural networks. In S. Y. Koen Steemers, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp.431-435). London: Earthscan. Santamouris, M., Papanikolaou, N., Livada, I., Koronakis, I., Georgakis, C., Argiriou, A.,et al. (2001). On the impact of urban climate on the energy consumption of buildings. Solar Energy , 70 (3), 201-216. Sharples, S., & Heidari, S. (2000). A Thermal Comfort Survey of Courtyard housing in Iran. In S. Y. Koen Steemers, Architecture, city, environment: proceedings of PLEA 2000 : July 2000, Cambridge, United Kingdom (pp. 224-228). London: James & James (Science Publishers) Ltd. http://www.marina-bay.sg Shashua-Bar, L., & Hoffman, M. (2000). Vegetation as a climatic component in the design of an urban street An empirical model for predicting the cooling effect of urban green areas with trees. Energy and Buildings , 31 (3), 221-235. S.H. Lam, (2009), Influence of Different Water Feature on Thermal Comfort and Air Temperature. Unpublished undergraduate dissertation, School of Design and Environment, National University of Singapore. Shanock, Linda; Baran, Benjamin; Gentry, William; Pattison, Stacy; Heggestad, Eric (December, 2010). Polynomial Regression with Response Surface Analysis: A Powerful Approach for Examining Moderation and Overcoming Limitations of Difference Scores. Journal of Business and Psychology, Volume 25, Number 4, pp. 543-554(12). S. Murakawa, T. Sekine and K. Narita, Study of the effects of river on thermal environment in an urban area, Energy and buildings 15-16 (1991) 993-1001 Singapore Statistics, D. o. (2009). Key annual indicators. Retrieved October 1, 2009, from Statistics Singapore: http://www.singstat.gov.sg/stats/keyind.html#popnarea Smith, C., & Levermore, G. (2008). Designing urban spaces and buildings to improve sustainability and quality of life in a warmer world. Energy Policy (36), 4558–4562. 120 Sugawara, H., Narita, K.-i., & Kim, M. S. (2009). Cooling Effect By Urban River . Yokohama, Japan. Sun, Ranhao; Chen, Liding (2012). How can urban water bodies be designed for climate adaptation? Landscape and Urban Planning, 105 (27– 33). T. R. Oke, Boundary climate layer 2nd ed., 1987, Routledge, London and New York Turner, W. C. (2004). Energy management handbook. CRC Press. Wong, N. H. (2000). COVER STORY: Urban Heat Island Effect: Sinking the Heat. Retrieved December 1, 2009, from Innovation: The magazine of research and technology: http://www.innovationmagazine.com/innovation/volumes/v9n1/contents.shtml Wong, N. H. (2009). Innovation: The Magazine of Research and Technology. Urban Heat Island Effect: Sinking the Heat , 9 (1). Wong, N. H., & Yu, C. (2005). Study of green areas and urban heat island in a tropical city. Habitat International , 29 (3), 547-558. Yagi, K. (2009, July). [Newsletter] Sprinkling Water to Cool Down Summer's Heat. Retrieved September 15, 2009, from Japan Design for Sustainability Mail Magazine: http://www.japanfs.org/en/mailmagazine/newsletter/pages/029260.html Yagi, K. (2009, July). The Mission Uchimizu Campaign as Social Design. Japan for Sustainability Newsletter (83). Yamagata, H., Minamiyama, M., Nasu, M., & Yoshizawa, M. (2008). Heat island mitigation using water retentive pavement sprinkled with reclaimed wastewater. Water science and technology , 57, 763-771. Yamamoto, Y. (2006, January). Measures to Mitigate Urban Heat Island. Science and Technology Trends, Quarterly Review (18). Y.R. Adebayo, “Heat island” in a humid tropical city and its relationship with potential evaporation, Theoretical and applied climatology 47 (1991) 137-147 Zeng, Y., Qiu, X. F., Gu, L. H., He, Y. J., & Wang, K. F. (2009). The urban heat island in Nanjing. Quaternary International , 208 (1-2), 38-43. ‘Climate of Singapore’ Source: Based on metrological data on 1982-2001, National Environmental Agency, Singapore. 121 APPENDIX A STATEMENT OF PERMISSION I declare that I have obtained explicit permission from my co-authors for including some figure from one peer-reviewed scientific journal manuscripts as chapters in this thesis (Chapter 4). They are : • Wong Nyuk Hien; • Tan Chun Liang, Terrence and; • Steve Kardinal Jusuf. 122 [...]... distinction of the influence of cooling from the water bodies horizontally 3 1.2 Research Question 1 How is solar radiation affecting the cooling effect of water bodies in the tropics, especially Singapore? 2 How do water bodies in hot and humid climates extend the cooling effect to the surrounding area in terms of distance? 3 How does the surrounding area impact the cooling effect performance of water. .. sink during the day and releasing the absorbed heat in the night The high heat capacity of water was another underlying reason for the water being warmer in the evening The results generally inform the cooling potential of “moving” water features, with the water fountain and water wall producing a better cooling effect and thermal comfort acceptance Perhaps an integration of both these water facilities... effect of water features located in the vicinity of buildings, with comparatively lesser greenery than the park, was not considered In addition, water fountains that have water sprinkling in the air with an increased surface area, thereby increasing the evaporation rate leading to an added cooling effect, should be considered as a feature just as capable of cooling the air as a water wall (Energy and. .. background and rationale of this study A brief outline of the research paper is provided Chapter 2 presents an extensive literature review on past research and papers done in relation to the cooling effect of water features in Singapore (a hot and humid climate) The literature on the water evaporation process and factors that affect it to produce water bodies’ cooling effect, such as solar radiation and wind... standing or moving water (such as basins or fountains) According to Munn et al (1969) and Naot et al (1991), the evaporation and transfer of sensible heat result in lowering air temperature at the water bodies Evaporation decreases air temperature due to the latent heat of absorption and increase in specific humidity Meanwhile, the transfer of sensible heat between air and the underlying water (water. .. research in China revealed that the environmental temperature can be reduced by narrow built-up land and rounded shapes in water and green landscapes (Shi, Deng, Wang, Luo, & Qiu, 2011) Consistent with these results, the urban water body area and geometry were also found to significantly influence the urban cooling island (UCI) effects in Beijing The area of the water body had the greatest effects on... air (Smith and Levermore, 2008) 2.6.1 Water Facilities and Their Effect in Sub Tropic Area In Osaka City, N Nishimura et al (1998) conducted a field measurement on existing water facilities, including a pond, waterfall, and spray fountain in a park located in an urban area The temperature decline effects of waterfalls and spray-type water facilities in urban areas were measured The results indicated... performance of water bodies, as determined using a simulation model? 1.3 Research Objectives The objectives of the study are as follows: 1 To determine the cooling effect and benefit of water bodies to their surrounding microclimate in hot and humid climates (field measurement study) 2 To determine the possible impact of water bodies’ cooling effect on the air temperature on a hot sunny day (field measurement... the water cycle, and the variables that affect the water s evaporation process to create a cooling performance on its surroundings Water bodies in sub-tropic areas seem to have a significant cooling effect amount On the other hand, Singapore, with its hot and humid climate, seeks to achieve the same effect Some researchers in Singapore have measured water facilities such as water walls, fountains, and. .. extent of water bodies’ cooling effect on the surroundings area in hot and humid climates based on solar radiation performance In addition, through simulation study, the findings could be used to determine how the surroundings near the water bodies affect the water bodies’ cooling effect performance 1.6 Organization of Study This thesis paper will be organized as follows: Chapter 1 provides an introduction ... How water bodies in hot and humid climates extend the cooling effect to the surrounding area in terms of distance? How does the surrounding area impact the cooling effect performance of water. .. being warmer in the evening The results generally inform the cooling potential of “moving” water features, with the water fountain and water wall producing a better cooling effect and thermal... to the cooling effect of water features in Singapore (a hot and humid climate) The literature on the water evaporation process and factors that affect it to produce water bodies’ cooling effect,