Urban coolisland effects and their quantitative evaluations in accordance with areal size

为了估计城市绿地与土地覆盖之间的温度关系，我们于2006年夏天在东京的11个公园（皇宫、明治神宫和代代木公园、新宿御花园、shizen - kyiquen、小石川植物园、小石川口乐园、柴巴公园、丰山公园、陆见园、新宿区关南公园、南港区龙川亲王纪念公园）连续观测了长期温度。结果得到如下结果：1)在野外观测的基础上，获得了东京主要绿地公园的温度数据。城市绿地对气候的影响在热条件下变得清晰。例如，夏季绿地室内外温差最多在5℃左右。2)绿地室内外温差由绿地总面积或绿地森林面积的对数曲线近似表示。该研究为近年来探讨城市绿地热效应的问题SLOSS （Single Large Or few Small）提供了理论依据。此外，本研究还提出了一个基于森林面积大小和速率的新指数，该指数能够解释绿地内外的温差。3)基于高时间分辨率观测数据的多元回归分析，揭示了地表覆盖对温差贡献率大小的时间序列变化规律。

In order to estimate the temperature relationship between the green space in the city and its land cover, long term temperature was observed continuously in the 11 parks (Imperial Palace, Meiji Jingu and Yoyogi park, Shinjuku Imperial Gardens, Shizen-Kyoikuen, Koishikawa Botanical Garden, Koishikawa Kourakuen Garden, Shiba park, Toyama park, Rikugien Garden, Shinjuku ward Kansenen park, Minato ward Prince Arisugawa Memorial park) in Tokyo in the summer of 2006.As a result, the following results were obtained.1) Temperature data in the major green parks in Tokyo were obtained based on the field observations. The climatic effects of green spaces in the city for thermal conditions became clear. For example, the temperature difference between the inside and outside of the green spaces in summer season was at most about 5 degree Celsius.2) The temperature difference between the inside and outside of the green space was approximated by the curve of the logarithm of the gross area of the green space or the forest area of the green space. This study could contribute the SLOSS (Single Large Or Several Small) problem, which is a problem in discussing the thermal effect of the green space in the city in recent years. Moreover, this research proposed a new index based on the size and the rate of the forest area, and that was able to explain the temperature difference between the inside and outside of the green spaces.3) The time-series behavior of the magnitude of the land cover's contributing rate of the temperature difference was shown by the multiple regression analysis based on the higher time resolution observation in this study.