The effect of building form and sky view factor on daytime land surface temperature in residential street canyons, Seoul, Korea

Abstract

Urbanization causes significant urban climate change, especially with increasing temperatures. This is called the Urban Heat Island (UHI) effect and affects human health and the quality of life. Changes in urban geometry are one of the key factors causing the UHI. Building structures change the urban canyon form that, in turn, changes the thermal condition in the urban canyon. Therefore, a quantitative analysis of the effect of building structures on the thermal conditions in an urban canyon is very important for urban planning. A commonly used indicator to describe the urban geometry is the sky view factor (SVF). This indicator, often denoted by , indicates the ratio of the radiation received (or emitted) by a planar surface from the sky to the radiation emitted (or received) from the entire hemispheric radiating environment. With its important role in radiation balance schemes, the SVF has been widely used by climatologists to investigate the relationships between urban geometry and thermal conditions. Many previous studies using photographic methods use a fish-eye lens to take onsite photographs that project the hemispheric environment onto a circular plane. However, this method is limited as direct sunlight or different cloud types can cause problems in image processing. The photographic method is used to extract some points by taking a picture at specific points. However, some points do not represent the correct values for the site, as SVF can have different values in the same urban canyon because of the distance from the buildings. Because of these limitations, software methods have been developed as computer performance has rapidly increased and digital mapping techniques have become prominent. Recently, software methods have been frequently used in the analysis of the urban thermal condition. The software methods increase processing speeds, while the accuracy of the method depends on the resolution of the raster database in the digital elevation model (DEM). For high accuracy, there needs to be high resolution images of the buildings and a topography database. These methods offer rapid ways of calculating the continuous SVF for large areas based on comprehensive analyses, and studies using this method have increased recently. The collection of temperature data in previous studies has relied on onsite surveys using thermometers to understand the thermal condition. However, since the microclimate has complex characteristics and is affected by many factors, the collection of representative temperatures is challenging in limited sample sites. Therefore, land surface temperatures (LST) obtained by remote sensing, which has been used by many previous studies, has advantages for analyzing the relative thermal condition in large areas simultaneously. With these advantages, some recent studies have analyzed the correlation between the simulated SVF and LST for understanding the thermal condition. However, there are limitations. First, the LST and the temperature pattern characteristic by land use are not controlled and can be very different. Second, since Landsat 8 has a 30 m LST resolution, the measurement of the thermal condition of building outdoor space and the effect of the surface temperature controlled by the roof (building) coverage ratio need to be considered. This is also missing. Third, the distance to a thermal reduction component is an important factor, because mountains, rivers, and green space have a cooling effect. In addition, different types of land cover have unique thermal characteristics. These factors have been disregarded in previous studies. The building arrangement effect on the thermal condition has also been disregarded. Both of these effects are included in this study. The goal of this study is to develop a quantitative measurement for the relationship among building forms, the shape of the urban canyon, and the thermal condition in the urban canyon using a suitable method for large areas considering the limitations of previous studies. Therefore, this study analyzes the correlation among LST obtined by remote sensing, SVF obtained by SVF simulation and building heights by roof (building) coverage ratio group. I analyzed results the mechanisms in previous studies and the green space in residential areas using the normalized difference vegetation index (NDVI) especially for the study site, which is a residential area where there is land use control. I especially focused on residential areas, since the thermal comfort of residential areas directly effects vulnerable people, such as children and the elderly. In this analysis, I attempted to evaluate the components directly affecting thermal reduction, such as mountains, the Hangang River, streams, and green space. In addition, there were efforts to find an organic relationship among building height, the SVF in outdoor space, and the LST by each similar roof (building) coverage ratio group to control the problem. Because there is a limitation in the study caused by LST resolution. To sum up the results of this study, first, low rise buildings, such as detached houses and multi-family housing, result in a high SVF condition, which can be explained by the formation of an urban canyon. A high SVF results in a high LST caused by the increased net radiation near the ground because of increased direct solar radiation. In contrast, high-rise buildings, such as apartments or tower type apartments, result in low SVF in the urban canyon and a low LST environment. Second, there are sections with different SVF values even with similar heights for high rise building, caused by the arrangement of the high rise flat-type apartments. The extracted area with an SVF value under 0.2 in the simulated data is an enclosed type arrangement with shapes such as a C or O. An open type arrangement with an L or I shape has SVF values mostly over 0.2. As the SVF decreases, the LST increases in the zone with SVF values under 0.2, such as in the enclosed arrangements. The enclosed arrangement form of a flat-type apartment has a high LST, and these areas have a poor radiant cooling ability at night caused by the low SVF. Therefore, enclosed type high rise buildings should be avoided in urban planning. This pattern with an SVF under 0.2 is unique. Therefore, I focused on this section and to develop the reasons for these results. The first hypothesis is based on previous studies. An area with a small SVF value leads to a decline in long wave radiation, resulting in increasing counter radiation in the nighttime. Therefore, the limited cooling of the surface in the nighttime influenced the daytime surface. The Landsat 8 LST on site is 11 a.m. At this time, the sun's elevation is not high enough to heat the surface in the low SVF area. Therefore, there is not enough direct solar radiation on the surface and the nighttime pattern still is present in the daytime in the area with the SVF under 0.2. The second hypothesis is that the area with the SVF value under 0.2 has lower green space than the area with SVF over 0.2. In a previous study, Moon (2011) found a pattern showing that the courtyard of enclosed arrangement flat-type apartments has more parking lot and lower green space than the other types. Therefore, the analysis between NDVI and SVF focused on the area with the SVF value under 0.2. The results show that NDVI and SVF have a positive relationship at all SVF values. In sections with the SVF under 0.2, the NDVI ratio increase ratio is higher than in the other sections. In sections with the SVF under 0.2, as SVF increased, NDVI increased while LST decreased, indicating NDVI affects the decrease in LST. In contrast, in sections with SVF values over 0.2, as SVF increased, NDVI increased, and LST increased. This pattern means the mechanism of increasing direct solar radiation is affected by an increase in NDVI. The goal of this study is to find quantitative correlations among building forms, the urban canyon, and the thermal environment using simulated SVF based on high resolution data and remote sensing. This study can use the basic data from urban planning or building architecture. Ultimately, this study can contribute to sustainable development.