| 摘要: | 本研究目的為探討台北市公園綠地對於都市熱島的局地與遠距影響。利用移動觀測法，於94年夏季7-9月及95年冬季12-3月的白天AM10:00-PM2:00及夜晚PM9:00-AM1:00，對台北市主要街區之氣溫進行實地測量，並依中央氣象局之氣溫，作為都市內相對熱島強度之指標。環境因子部份則針對台北市116個公園及其周邊局地空間因子及微氣候因子，包括：遠距公園之面積與NDVI；局地空間內的公園面積比、NDVI、建地面積、樓地板面積、樓層數、交通量、溼度、風速與相對溫差進行統計分析。
研究結果顯示，夏季夜晚各測點對中央氣象局台北測站相對溫差之標準差最小，為0.40，冬季夜晚最大，為1.29。影響夏天相對溫差的因子主要為局地空間的NDVI及溼度，冬天則為同為微氣候因子的溼度及風速。
因為時間變異量的干擾，公園綠地對於周邊局地氣候的影響顯著，但遠距影響並不顯著，資料中可發現某些路段之相對溫差明顯地會隨著調查時間變化，進而衍生計量的誤差值。此結果說明，時間變異量對於相對熱島強度變化的影響大於公園對於相對熱島強度的影響，因此，未來在測量都市內相對熱島時，應須將時間變異量所造成的誤差值一併考量，將測量樣本在空間中分散，在同一時段內，取得代表不同地區之樣本，方能解決時間變異量所產生的誤差情況。
從樹狀分析得知，夏季白天當溼度大於48.95%或局地NDVI大於0.005時，相對溫差較低；夏季夜晚當遠距中心公園面積大於0.43公頃，或溼度大於68.35%，或交通量小於6.9輛*m/m2時，相對溫差較低；冬季白天當溼度大於52.73%或風速1.48m/s時，則有降溫的作用；冬季夜晚，當風速大於0.74m/s、或溼度大於75.53%、或局地的NDVI小於0.11時，皆能降低局地溫度。
綜合分析，得出遠距公園之NDVI大於0.146、面積大於0.43公頃或局地空間內的NDVI大於0.005，有利夏季降溫作用，同時若限制建築物高度在4層以下，及降低交通量則可有效地降低溫度，達到減緩熱島效應的作用。此分析量化結果將可作為未來台北市都市規劃、綠地計畫、土地利用的參考，降低能源的消耗，提昇環境品質，藉以有效的改善都市微氣候及實踐永續都市的概念。
The purpose of this thesis is to study how urban parks affect the urban microclimate at local to 1 km scales.
In July-September 2005 and in December 2005 to March 2006 during daytime and nighttime to record microclimate measurements, including temperature, humidity, and wind speed in the major urban district of Taipei city. The synchronous air temperatures recorded from the Central Weather Bureau Taipei Station were then subtracted from the field temperature measurements to produce a spatial map of relative heat island intensity of central Taipei city. The characteristics of 116 Taipei urban parks and their vicinity, including the area and NDVI of the parks, the green coverage, NDVI, building coverage, floor area, average number of floors, and traffic volume and the microclimatic conditions in their vicinity were analyzed using GIS and statistical procedures.
Results show that the temperature difference between the field measurements and the CWB Taipei station showed least variance at night in summer, is 0.40 and most variance at night in winter, is 1.29. Local NDVI and humidity are the major factors affecting the temperature differences in summer, whereas humidity and wind speed are the factors most associated with temperature differences in winter.
Temporal variance made up a large component in the variability of the temperature differences, and the temperature differences showed clear disparity among days. As a result of this bias, the long-distance effect of parks on the microclimate in its vicinity is unclear. In future studies, the temporal variance must be considered, and samples representing an extensive area should be measured within limited time.
From the tree-analysis, in daytime in summer, the temperature difference is lower when the humidity is greater than 48.95% or local NDVI is greater than 0.005. At night in summer, the temperature difference is lower when the area of the park is greater than 0.43 hectares, the humidity is greater than 68.35%, or the volume of traffic is smaller than 6.9 car-equivalent unit*m/m2. In daytime in winter, the temperature difference is lower when humidity is greater than 52.73% or wind speed is 1.48 m/s. At night in winter, the temperature difference is lower when wind speed is greater than 0.74 m/s, humidity is greater than 75.53%, or local NDVI is smaller than 0.11.
In conclusion, when a park has a NDVI greater than 0.146 or area larger than 0.43, or local NDVI is greater than 0.005, the temperature difference in its vicinity is lower in summer. At the local scale, limiting the height of buildings to 4 stories or less, or lowering the negative impacts of heavy traffic also mitigates the heat island effect. Such analytical results may be used in future urban plans for Taipei city in hopes for a sustainable future. |
