基於土地利用迴歸之機器學型模型分析新冠肺炎三級警戒政策對臺灣工業城市細懸浮微粒之影響

Impacts of the Level 3 Alert Brought by COVID-19 on Fine Particulate Matter of an Industrial City in Taiwan Using a Land-Use Based Machine Learning Model

10.6574/JPRS.202409_29(3).0003

蘇均珺(Jun-Jun Su)；翁佩詒(Pei-Yi Wong)；曾于庭(Yu-Ting Zeng)；李佳禾(Chia-Ho Lee)；吳治達(Chih-Da Wu)

空氣污染 ； 細懸浮微粒物（PM_(2.5)） ； 基於土地利用的機器學習模型 ； COVID-19 ； 三級警戒 ； Air Pollution ； Fine Particulate Matter (PM_(2.5)) ； Land-Use Based Machine Learning Model ； COVID-19 ； Level 3 Alert

航測及遙測學刊

29卷3期（2024 / 09 / 01）

165 - 176

繁體中文;英文

COVID-19疫情對全球帶來巨大衝擊，臺灣政府於2021年5月19日宣布三級警戒，限制民眾活動。本研究旨在評估警戒期間臺灣工業城市PM_(2.5)濃度之變化。本研究以代表性的工業城市高雄市為研究區域，蒐集1994至2020年的空污觀測數據和地理變量，利用土地利用迴歸和逐步變量選擇建立模型、選取重要變數，再使用不同機器學習演算法建立模型，其中結果以Random Forest（RF）演算法的模型表現最佳，R^2達0.95；推估成果顯示封鎖期間空氣品質改善，高雄市平均PM_(2.5)濃度為18.1μg/m^3，低於警戒前19.9 μg/m^3。Paired t-test結果顯示差異達到統計顯著水準（p值<0.001），各土地利用區域（居住區、工業區、街道和綠地）亦呈現一致結果。

The COVID-19 epidemic has brought significant changes to human activities worldwide, including in Taiwan. On May 19, 2021, the government announced a level 3 alert to restrict public movement. This study aims to assess the impact of the lockdown policy on PM_(2.5) concentrations in Taiwan's industrial city, Kaohsiung. Daily PM_(2.5) observations and geographic data from 1994 to 2020 were collected. A land-use regression model, combined with stepwise variable selection, was used to identify important factors affecting PM_(2.5) variability. These predictors were used to develop machine learning models with algorithms such as Random Forest (RF), which showed the best performance with an R^2 of 0.95. Paired t-tests indicated that PM_(2.5) levels were significantly lower during the alert (18.1 μg/m^3) compared to before (19.9 μg/m^3), with consistent results across residential, industrial, street, and green areas (p < 0.001).