微氣候尺度高齡農民熱風險之時空分析—以雲林縣五塊社區為例

Abstract

近年來，高溫事件日益頻繁，高齡農業勞動者在面對極端氣候所帶來的熱風險逐年升高。本研究以雲林縣五塊社區為研究場域，並以高齡農民為目標對象，透過氣象觀測資料與高解析地表空拍資訊，進行微氣候尺度下的熱風險評估。使用資料包括逐時氣溫、風速、相對濕度、全天空日射量等參數，並利用實際拍攝之數值地表模型結合建物資料計算天空視野因子等型態參數，以推估平均輻射溫度(Tmrt)並進一步計算生理等效溫度(PET)作為熱壓力指標，並且根據暴露時間計算累積PET壓力，評估一段時間內戶外勞動者所受的熱壓力多寡。 研究亦根據農民實際的工作路徑與時段，提取路徑上取不同時間段的PET數值，分析農民於不同時段、不同位置所遭受之熱壓力變化。結果顯示，農民在間移動路線中，08時至16時之PET值普遍達40至44°C，17時更來到43至46°C；計算日間累積PET壓力同樣顯示為高強度持續熱壓力環境。實地定點測量結果顯示中午時段在無遮蔽農田區測得氣溫高達38.5°C、WBGT達34.2°C，屬於極高風險級距；而在有遮蔭的區域，WBGT僅為29.4至30.8°C，顯示環境條件對熱壓力有調節效果；聚落區域因建物遮蔽造成SVF降低，使得模型推估的PET較實測之WBGT值低，同時實際測量的氣溫值較氣象署觀測資料高，因此在使用氣象測站資料模擬評估該區域時，將有低估熱風險的情況發生。敏感度測試中也顯示，PET對於太陽輻射以及風速較為敏感，因此易低估低SVF情況的實際熱感受，必要時須配合實地測量或其他熱指標進行熱風險評估。 本研究證實，在微氣候尺度下的熱風險同樣具高度時間與空間變異性，建議未來在高齡農民的風險評估上，可結合高解析資料、移動量測與地景因子進行整合分析，作為農業熱危害預警平台與作業調適政策之依據。

In recent years, extreme heat events have become more frequent, increasing heat risks for elderly agricultural workers. This study focused on elderly farmers in five communities in Yunlin County, using meteorological data and high-resolution aerial imagery for a microclimate-scale heat risk assessment. Hourly temperature, wind speed, humidity, and solar radiation were combined with digital surface models and building data to calculate sky view factor and mean radiant temperature. These were used to estimate physiological equivalent temperature and cumulative PET load to evaluate heat stress over time. PET values along farmers’ actual work routes ranged from 40–44°C between 8 AM and 4 PM, rising to 43–46°C at 5 PM. Cumulative diurnal PET load indicated sustained high heat stress. Field measurements showed midday exposed farmland temperatures up to 38.5°C and WBGT of 34.2°C, indicating very high risk, while shaded areas had lower WBGT (29.4–30.8°C). Shading reduced SVF, causing modeled PET to underestimate heat stress compared to WBGT. Sensitivity analysis revealed PET is strongly influenced by solar radiation and wind speed, leading to underestimation in low SVF conditions. Therefore, field measurements should complement heat risk assessments when possible. This study confirms that microclimate heat risk varies significantly across time and space. Future assessments should integrate high-resolution data, mobile measurements, and landscape factors to support heat hazard early warning systems and adaptation policies for elderly farmers.