GISdevelopment.net --> Application --> Environment



Introduction
The urban heat island refers to the elevation of urban temperatures over those in surrounding rural areas, the difference generally being greater at night than during the day For people living and working in cities, both day- and night-time air temperatures are important, particularly in tropical and sub-tropical cities where the climate may often exceed the threshold of human comfort for large parts of the year. Heat island studies are additionally significant due to the perceived relationship between air pollution and heat islands. A number of satellite-based studies using thermal infra-red imagery have been carried out since satellite data provide a dense grid of almost instantaneous temperature measurements over a city and can permit visualisation of spatial relationships between temperature patterns and urban land use and infrastructural features. Satellite-based studies can therefore provide recommendations for building design and landscaping of urban developments, to minimise heat accumulation and retention by urban surfaces, and thus the accumulation of warm, polluted and stale air in the city. This study is a multi-spectral and multi-sensor evaluation of thermal infra-red imagery for microclimate monitoring in two densely-built high rise towns in the New Territories of Hong Kong.

Objectives and study area
The main objectives of the research are to supplement the existing body of research on satellite-derived urban heat islands by examining the enhanced spatial and spectral resolution of LANDSAT 7's Enhanced Thematic Mapper+ (ETM+) sensor as well as the multispectral thermal wavebands from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). The fact that the ASTER image studied is a night-time image can also provide a meaningful comparison of satellite-derived surface temperatures of urban surfaces between day and night. The towns, Tuen Mun and Yuen Long are located in the western New Territories, Hong Kong and were selected because the night-time ASTER image is cloud-free only over this part of Hong Kong. Tuen Mun, to be discussed in this paper is coastal, located in a narrow, steep-sided valley with grass and tree-covered slopes (Figure 1). Remnants of the original village settlements are located between the modern, high-rise town in the valley, and the mountain slopes.

Methods
The ASTER and LANDSAT images were obtained in the relatively warm part of the year (Table 1) September and October, 2001. Mean daily air temperatures on 17^th September and 6^th October were similar (29° and 27° respectively) suggesting similar climatic conditions on the two image dates thus permitting the day (9.15am) and night (9.40pm) images to represent a continuum of relative air and surface temperatures.


Due to the different types of thermal image data available (Table 1), different processing techniques were applied for derivation of the required parameter; Surface (or Kinetic) Temperature (Ts).

1. LANDSAT ETM+


The low gain thermal band (band 6:1) of ETM+ was converted to Ts in four steps, as follows:


1. Digital Number (DN) to Spectral Radiance (Ll)


Conversion of the image DN values to spectral radiance is carried out using the gain and offset values given in the image header file (Eq. (i)). Thus


L_l = ((LMAX-LMIN)/(QCALMAX-QCALMIN)) * (QCAL-QCALMIN) + LMIN -------------(i)

Where:
QCALMIN = 1, QCALMAX = 255, and QCAL = Digital Number
LMIN and LMAX = spectral radiance for band 6:1 at DN 0 and 255


2. Spectral Radiance to Black Body Temperature


The ETM+ thermal band data can be converted from spectral radiance Black Body Temperature (BBT) which assumes surface emissivity=1 (Eq. (ii))


T = K2 / ln(K1/ L_l+ 1) -------------------(ii)

Where:
T = Effective at-satellite temperature in Kelvin
K1 = Calibration constant 1 (W.m2.sr-1)(666.09)
K2 = Calibration constant 2 in K (1282.7)
L_l = Spectral radiance in (W.m2.sr-1)


3. Emissivity Correction


The visible wavelength bands of ETM+ image were classified into three main land cover classes: vegetation, non-vegetation and water using a supervised classification. Corrections for emissivity differences were carried out by land cover type by ratioing the BBT image with the classified image in which the pixel values for the land cover class were replaced with the corresponding emissivity value. Thus the emissivity corrected surface temperature (Ts) is derived by Equation (iii). (Artis and Carnahan 1982).


Ts = T / [1 + (lT / a) ln e] ---------------------(iii)

l = Wavelength of emitted radiance,
a = hc/K (1.438 ´ 10-2 mK),
h = Planck's constant (6.26 ´ 10-34 J.sec),
c = velocity of light (2.998 ´ 108 m/sec),
K = Stefan Bolzmann's Constant (1.38 ´ 10-23 J/K).


LANDSAT daytime image (Figure 3)

* High rise areas are relatively cool ("37°C) if buildings are closely spaced (Figure 5)
* Densely-built low rise areas are relatively warm ("39-40°) (Figure 6)
* Car parks and open spaces are the hottest areas (Tuen Mun bus station forecourt "45°C)
* Aspect of buildings in relation to sun elevation and azimuth has significant influence on Ts (Figure 7)

The emissivity correction procedure has the effect of increasing the spatial resolution of the thermal data due to fusion with the land cover map based on the 30 metre pixels of ETM+ visible wavebands.


4. Atmospheric correction


The Ts image was atmospherically corrected by comparison with Sea Surface Temperature data from the Hong Kong Observatory, which appeared to be stable both spatially and temporally over several days (at 29°C) at the image date. The addition of 10 DN values before conversion to radiance had the effect of raising Ts values by 5°C, resulting in Ts values for the sea surface, of 29°C, and all other image values were adjusted accordingly. The above procedures resulted in the majority of image Ts values between 34 and 45°C. Since air temperature at the image time (Figure 2) was 28.7°C and had been rising since 7am, and surface heating usually precedes increases in air temperature, usually exceeding it by several degrees, then these values appear acceptable.



Figure 2. Air temperatures for 24-hours on image dates (source: Hong Kong Observatory)


2. Aster


An ASTER image of Kinetic Temperature in °C (equivalent to Surface Temperature) was available. This image is a higher level product from EDC, derived from data collected in the five ASTER thermal wavebands (Table 1) and corrected for emissivity and atmospheric effects using image ratios and external data (Gillespie et al., 1996; Schmugge et al., 1995). Most image values for the study area ranged from 22-28°C, and this compares well with Sea Surface Temperature in-situ data of 27°C at Waglan Island (image value also 27°C at this point) and air temperature at the image time (9.40pm), of 26.6°C, which had been falling since 4pm (Figure 2). Such a situation, with Ts values on land lower than air and sea surface temperatures, is expected, especially on cloud-free nights due to long-wave radiant heat flux. The images were not de-striped, as the difference between adjacent scan lines was less than 0.3°for this 16-bit data.



Figure 3. ETM+ image of Surface Temperature (Ts) with building outlines.


3. Geometric correction, Overlay and Visualisation


The ASTER and LANDSAT Ts images were geometrically corrected to the Hong Kong 1980 Grid using the polynomial method and approximately eight ground control points resulting in accuracies of better than 0.3 pixel. Image Ts data were overlaid with road and building outlines in vector format (Figures 3 and 4) in order to examine micro- and meso-scale climatic patterns in and surrounding the two urban areas studied, and to compare the day and night situation.

Results
Both air and sea surface temperatures were 2°C cooler on 6th October 2001 (ASTER image) than on 17th September (ETM+ image) (Table 1). However, the ASTER night-time Ts values were found to be much (approximately 12°C) cooler than for daytime ETM+ (Figures 3 and 4). The range of Ts values were also smaller: 6° for ASTER, as opposed to 11°C for ETM+. Most spatial variations in Ts on both images can be explained by differences in urban morphology, and are illustrated in sections 5.1 and 5.2 and Figs 3-9.

1. Micro-scale Ts patterns related to urban morphology


LANDSAT daytime image (Figure 3)

* High rise areas are relatively cool ("37°C) if buildings are closely spaced (Figure 5)
* Densely-built low rise areas are relatively warm ("39-40°) (Figure 6)
* Car parks and open spaces are the hottest areas (Tuen Mun bus station forecourt "45°C)
* Aspect of buildings in relation to sun elevation and azimuth has significant influence on Ts (Figure 7)











2. ASTER night-time image (Figure 4)



• Densely-built high rise areas are relatively warm, especially near the coast ("28°C)



• Densely built low-rise areas are relatively cool ("25°C)



• Indigenous settlements (villages) are very cool (<22°C)



• Car parks and open spaces are relatively cool ("25°C) except near the coast ("28°C) (Figure 8)



• Areas of rural grassland surrounding the towns are the coolest areas ("18°C)


3. Meso-scale Ts distributions related to local climate


Vegetated mountain slopes appear to have a cooling effect on urban temperatures but only in urban areas immediately adjacent to them. A corridor of cool surfaces extending northwards from the coast along the river is interrupted by a shopping complex built over the river, thus blocking any potential funnelling effect of winds and sea breezes during the daytime, for dispersal of warm, stale and polluted air.

Discussion
The UHI effect is readily visible on images of both day and night, but the satellite-derived UHI based on Ts, is more pronounced during the day, as observed here. Densely-built high rise areas having a high building mass are relatively cool on daytime images due to the lag in heating of building materials, but the reverse is true at night. The daytime UHI is most intense in open, non-built areas, as well as in low rise developments which heat rapidly due to a larger proportion of their surface (roof) exposed to direct solar radiation. The daytime ETM+ image, after processing, is shown to be of adequate spatial resolution to show the effects of aspect and shadow due to building spacing and orientation, and both ASTER and ETM+ indicate the meso-scale climatic effects of vegetated open spaces and waterbodies in ameliorating heating especially during the daytime. The ASTER image additionally suggests that proximity to large waterbodies such as the sea may actually increase the temperatures of urban areas at night, even in open areas such as car parks where night-time heat loss would be expected to be rapid.

References

1. Artis, D.A., and W.H. Carnahan, 1982, Survey of emissivity variability in thermography of urban areas, Remote Sens. Environment, 12, 313-329.



2. Gillespie, A.R., Rokugawa, S., Hook, S., Matsunaga, T. and Kahle, A.B., 1996, Temperature/emissvity separation algorithm theoretical basis document, version 2.1. NASA/GSFC, Greenbelt, MD.



3. Nichol, J.E., 1996, High resolution surface temperature patterns related to urban morphology in a tropical city: a satellite-based study. Journal of Applied Meteorology, 35(1), 135-146.



4. Nichol, J.E., 1998, Visualisation of urban surface temperatures derived from satellite images. International Journal of Remote Sensing, 19(9), 1639-1649.



5. Schmugge, T., Hook, S.J., Kahle, A.B., 1995, TIMS observations of surface emissivity in Hapex-Sahel. IEEE Tr.Quant. Rem. Sens. Sci. Appl., V.III, Firenze, Italy, 2224-6.