都市的快速發展和氣候變遷的影響下，都市熱島現象已無法忽視，高溫化環境使得都市發展和都市活動都受到挑戰。本研究基於物聯網和智慧城市概念，建構並妥善利用都市尺度監測數據，針對城市大數據對兩項戶外熱舒適生理等效溫度(PET)和綜合溫度熱指標(WBGT)進行分析，探討都市氣候的極端現象，並以大數據進行機器學習訓練分析都市規劃和都市熱環境之關係。
監測方面，本研究以MySQL建構都市熱環境數據資料庫，並透過WebGIS以網頁即時呈現都市熱環境。同時，以地理資訊系統和空間內插針對台南都市核心區的都市活動和都市熱環境進行分析。接著，將土地利用資料網格化，建構並訓練極限梯度提升模型(eXtreme Gradient Boosting, XGBoost)，以土地利用進行台南都市核心區的夏季日均溫預測。
研究結果顯示，台南都市核心區夏季各個月份的日間月均溫的趨勢相近，皆靠內陸區域，整體夏季日間季均溫則以東區作為主要集中區域向東邊內陸區域擴散。台南都市核心區建案和景點分布的情形較為不同，建案相較更散佈於內陸區域，南區和安南區偏少，而多位於東區、永康區、北區一帶，其中大型建案現場，分布於安平區與中西區交界、東區和永康區等地帶。景點則多集中在中西區、東區等都市中心位置。曝曬在全天空日照的情況下PET分布和空氣溫度分布相近，WBGT則較為均勻，沿海地區仍高達30°C(WBGT)，兩者比起空氣溫度都更能凸顯台南都市區域的都市熱環境情形。
XGBoost模型若納入地理位置資訊(距海遠近、經度、緯度)針對夏季日均溫的預測有極高的解釋力，R平方可高達0.94，所有季節及時間的模型準確度都能達到0.85以上，顯示地理位置影響台南都市核心區溫度分布較多，土地使用的影響被大幅弱化。僅以土地使用作為樣本建構之夏季日均溫預測模型，其R平方最高為弱相關性的0.32，土地使用的影響層面相較有限，其中建築用地影響最多。
以土地使用決策模式進行台南都市核心區夏季日均溫的假設農業用地轉用作建築用地的預測，則安平區漁港部分和安南區大片魚塭的區域明顯升溫。行政區域中土地使用組成的改變，代表都市規畫和建築興建的趨勢確實對都市熱環境有所影響。

Under the influence of rapid urban development and climate change, urban heat island cannot be ignored. A high-temperature urban environment makes thermal comfort a problem. Based on the concept of the IoT and smart cities, this research constructs and makes proper use of urban-scale monitoring data, and analyzes PET and WBGT based on urban big data and uses big data for machine learning training to analyze the relationship between urban planning and urban thermal environment.
This research uses MySQL to build a database of urban thermal environment data and presents the urban thermal environment in real-time through WebGIS. It uses GIS and spatial interpolation to analyze the urban activities and urban thermal environment in Tainan City.
The results of the study show that the situation of construction sites in Tainan City and the distribution of scenic spots are quite different. Construction sites are more scattered in the inland area. PET and WBGT can highlight the urban thermal environment of the Tainan more than the air temperature.
XGBoost model incorporates geographic location information to predict the summer average daily temperature, the R-squared can be as high as 0.94. The summer daily average temperature prediction model constructed only with land use has the highest R-squared of 0.32, which is weakly correlated. The impact of land use is relatively limited, of which construction land has the most impact.Using the land-use decision-making model to predict the summer average daily temperature in Tainan City while the agricultural land converses to construction land, Anping District and Annan District have significantly warmed up. Changes in the composition of land use have caused differences in urban planning and building construction, and have an impact on the urban thermal environment.

1. Allam, Z., &Newman, P. (2018). Redefining the Smart City: Culture, Metabolism and Governance. Smart Cities, 1(1), 4–25. https://doi.org/10.3390/smartcities1010002
2. Bonafoni, S., Baldinelli, G., &Verducci, P. (2017). Sustainable strategies for smart cities: Analysis of the town development effect on surface urban heat island through remote sensing methodologies. Sustainable Cities and Society, 29, 211–218. https://doi.org/10.1016/j.scs.2016.11.005
3. Chen, T., &Guestrin, C. (2016a). XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 13-17-Augu, 785–794. https://doi.org/10.1145/2939672.2939785
4. Chen, T., &Guestrin, C. (2016b). XGBoost: A Scalable Tree Boosting System. The Journal of the Association of Physicians of India, 42(8), 665.
5. Chen, Y. C. (n.d.). 臺灣都市氣候地圖之研究 Urban Climate Map in Taiwan.
6. dosSantos, R. S. (2020). Estimating spatio-temporal air temperature in London (UK) using machine learning and earth observation satellite data. International Journal of Applied Earth Observation and Geoinformation, 88(October 2018), 102066. https://doi.org/10.1016/j.jag.2020.102066
7. Farkas, G. (2017). Applicability of open-source web mapping libraries for building massive Web GIS clients. Journal of Geographical Systems, 19(3), 273–295. https://doi.org/10.1007/s10109-017-0248-z
8. Gobakis, K., Kolokotsa, D., Synnefa, A., Saliari, M., Giannopoulou, K., &Santamouris, M. (2011). Development of a model for urban heat island prediction using neural network techniques. Sustainable Cities and Society, 1(2), 104–115. https://doi.org/10.1016/j.scs.2011.05.001
9. Ho, M. C., Huang, K. T., &Wang, J. C. (2013). The Development and research on Hourly Typical Meteorological Years (TMY3) for Building Energy Simulation Analysis of Taiwan. Architecture and Building Research Institute Ministry of the Interior Research Project Report, 1–83.
10. Howard. (1966). The Climate of London. The Geographical Journal, 132(1), 84. https://doi.org/10.2307/1793062
11. Hsieh, C. M., Aramaki, T., &Hanaki, K. (2007). Estimation of heat rejection based on the air conditioner use time and its mitigation from buildings in Taipei City. Building and Environment, 42(9), 3125–3137. https://doi.org/10.1016/j.buildenv.2006.07.029
12. Hwang, R. L., Lin, T. P., &Matzarakis, A. (2011). Seasonal effects of urban street shading on long-term outdoor thermal comfort. Building and Environment, 46(4), 863–870. https://doi.org/10.1016/j.buildenv.2010.10.017
13. Ioannou, L. G., Tsoutsoubi, L., Samoutis, G., Bogataj, L. K., Kenny, G. P., Nybo, L., Kjellstrom, T., &Flouris, A. D. (2017). Time-motion analysis as a novel approach for evaluating the impact of environmental heat exposure on labor loss in agriculture workers. Temperature, 4(3), 330–340. https://doi.org/10.1080/23328940.2017.1338210
14. IPCC. (2014). Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. In Ipcc.
15. Kimura, F., &Takahashi, S. (1991). The effects of land-use and anthropogenic heating on the surface temperature in the Tokyo Metropolitan area: A numerical experiment. Atmospheric Environment. Part B, Urban Atmosphere, 25(2), 155–164. https://doi.org/10.1016/0957-1272(91)90050-O
16. Lin, T. P. (2009). Thermal perception, adaptation and attendance in a public square in hot and humid regions. Building and Environment. https://doi.org/10.1016/j.buildenv.2009.02.004
17. Lin, T. P., &Matzarakis, A. (2008). Tourism climate and thermal comfort in Sun Moon Lake, Taiwan. International Journal of Biometeorology, 52(4), 281–290. https://doi.org/10.1007/s00484-007-0122-7
18. Logan, T. M., Zaitchik, B., Guikema, S., &Nisbet, A. (2020). Night and day: The influence and relative importance of urban characteristics on remotely sensed land surface temperature. Remote Sensing of Environment, 247(June 2019), 111861. https://doi.org/10.1016/j.rse.2020.111861
19. Lundberg, S. M., &Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 2017-Decem(Section 2), 4766–4775.
20. Ma, J., Ding, Y., Cheng, J. C. P., Jiang, F., Tan, Y., Gan, V. J. L., &Wan, Z. (2020). Identification of high impact factors of air quality on a national scale using big data and machine learning techniques. Journal of Cleaner Production, 244, 118955. https://doi.org/10.1016/j.jclepro.2019.118955
21. MIKAMI, T., YAMATO, H., &Extended-METROS Research Group. (2011). High-resolution Temperature Observations Using Extended-METROS in the Tokyo Metropolitan Area and Their Urban Climatological Significance. In Journal of Geography (Chigaku Zasshi) (Vol. 120, Issue 2, pp. 317–324). https://doi.org/10.5026/jgeography.120.317
22. Muller, C. L., Chapman, L., Grimmond, C. S. B., Young, D. T., &Cai, X. (2013). Sensors and the city: A review of urban meteorological networks. International Journal of Climatology, 33(7), 1585–1600. https://doi.org/10.1002/joc.3678
23. Nam, T., &Pardo, T. A. (2011). Conceptualizing smart city with dimensions of technology, people, and institutions. ACM International Conference Proceeding Series, 282–291. https://doi.org/10.1145/2037556.2037602
24. Oke, T. R. (1982). The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society, 108(455), 1–24. https://doi.org/10.1002/qj.49710845502
25. Osborne, P. E., &Alvares-Sanches, T. (2019). Quantifying how landscape composition and configuration affect urban land surface temperatures using machine learning and neutral landscapes. Computers, Environment and Urban Systems, 76(August 2018), 80–90. https://doi.org/10.1016/j.compenvurbsys.2019.04.003
26. Oudin Åström, D., Forsberg, B., Ebi, K. L., &Rocklöv, J. (2013). Attributing mortality from extreme temperatures to climate change in Stockholm, Sweden. Nature Climate Change, 3(12), 1050–1054. https://doi.org/10.1038/nclimate2022
27. Potchter, O., Cohen, P., Lin, T. P., &Matzarakis, A. (2018). Outdoor human thermal perception in various climates: A comprehensive review of approaches, methods and quantification. Science of the Total Environment, 631–632, 390–406. https://doi.org/10.1016/j.scitotenv.2018.02.276
28. Roberts, H., Sadler, J., &Chapman, L. (2019). The value of Twitter data for determining the emotional responses of people to urban green spaces: A case study and critical evaluation. Urban Studies, 56(4), 818–835. https://doi.org/10.1177/0042098017748544
29. Sarrat, C., Lemonsu, A., Masson, V., &Guedalia, D. (2006). Impact of urban heat island on regional atmospheric pollution. Atmospheric Environment, 40(10), 1743–1758. https://doi.org/10.1016/j.atmosenv.2005.11.037
30. Vulova, S., Meier, F., Fenner, D., Nouri, H., &Kleinschmit, B. (2020). Summer Nights in Berlin, Germany: Modeling Air Temperature Spatially with Remote Sensing, Crowdsourced Weather Data, and Machine Learning. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 13, 5074–5087. https://doi.org/10.1109/JSTARS.2020.3019696
31. Warren, E. L., Young, D. T., Chapman, L., Muller, C., Grimmond, C. S. B., &Cai, X. M. (2016). The Birmingham Urban Climate Laboratory-A high density, urban meteorological dataset, from 2012-2014. Scientific Data, 3, 1–8. https://doi.org/10.1038/sdata.2016.38
32. Webb, A., Waseda, T., Fujimoto, W., Horiuchi, K., Kiyomatsu, K., Matsuda, K., Miyazawa, Y., Varlamov, S., &Yoshikawa, J. (2016). A High-Resolution, Wave and Current Resource Assessment of Japan: The Web GIS Dataset. http://arxiv.org/abs/1607.02251
33. Yu, S. Y., Matzarakis, A., &Lin, T. P. (2020). A study of the thermal environment and air quality in hot–humid regions during running events in southern taiwan. Atmosphere, 11(10). https://doi.org/10.3390/atmos11101101
34. ヒートアイランド対策大綱. (2013). Ministry of the Environment Government of Japan.
35. 國土資訊系統土地使用分區 資料標準制度規劃建置作業 結案報告書. (n.d.).
36. 小野雅司, &登内道彦. (2014). Estimation of wet-bulb globe temperature using generally measured meteorological indices. In Japanese Journal of Biometeorology (Vol. 50, Issue 4, pp. 147–157). https://doi.org/10.11227/seikisho.50.147
37. 鄭清萬張金堅. (2013). 環境熱急症. 56(9), 476–481.
38. 韓文珮. (2018). 以神經網路統計分析推估都市熱環境舒適度研究.
39. 高氣溫戶外作業勞工熱危害預防指引. (2019).
40. 王禹方. (2020). 應用高密度地面氣溫測量網(HiSAN)探討三維熱環境特徵及預測都市熱舒適分布