In order to utilize kudzu for rooftop gardening, appropriate cultivation methods of Kudzu were examined. The temperature of the concrete surface covered with potted kudzu was 5℃ lower than that of a non-shaded surface, and the temperature with hip-hydroponics cultivated kudzu was 10℃ lower. This result indicates that improvement in the thermal environment provided by kudzu is affected by the cultivation method. Among three plants species tested, the conductance and evapotranspiration rate were highest for hydroponics cultivated sweet potato, followed by hip hydroponics cultivated kudzu and then potted kudzu, indicating that the improvement in the thermal environment of concrete surface temperatures provided by plants is affected by the amount of evapotranspiration. It was also suggested that kudzu plants in Japan are diversed in the conductance, evapotranspiration rate and photosynthesis rate, which are related to ability to improve thermal environment.

A field experiment was performed in and surrounding built-up areas of the wooded area consisting of Yoyogi-park and Meiji-shrine in Tokyo on 1 December 2012 to examine the cool-spot phenomenon in early winter. The air temperature difference between the park and the surrounding built-up area was 1.8~2.9°C. The height of stable boundary layer reached 25m above ground level in the park during calm conditions. Air temperatures increased consistantly from the center of the park toward to the park boundary, but air temperature differences between inside the park and the surrounding built-up areas differed by location. For example, the air temperature difference in the east side of the park was 2.3°C, but it was 1.8°C in the west side. The largest air temperature difference was 2.9°C around Shibuya-station. The spatial distribution of the heights of buildings around the park was inhomogeneous to north, south, east, and west. For example, the average number of building floors in the south side of the park was higher than the west side. There was a positive correlation between the average number of building floors and air temperature difference from each of the built-up areas around the park. These results indicate the difficulty of selecting location points for obtaining the average air temperature difference between the park interior and surroundings.