To quantify the differences in winter thermal environment and air quality and to clarify the main factors influencing PM2.5 concentrations in cold regions, providing references for regional heating supply design and urban planning. In this study, pedestrian-level thermal environmental parameters and PM2.5 concentration were measured and compared across different urban functional zones (UFZs). Additionally, multiple linear regression (MLR), principal component analysis (PCA), and principal component regression (PCR) were employed to analyze the main controlling factors of PM2.5 and air temperature. The findings reveal that regional microclimate temperatures differ significantly, with variations of 2.68–4.31 °C compared to typical MET data. Notably, the Sky View Factor (SVF) emerged as the dominant influence on temperature variations, while PM2.5 concentrations were primarily driven by a combination of ENV (BD, SVF, GnPR) and MET factors (Ta, RH, TSr). The PCR model demonstrated superior predictive accuracy for PM2.5 concentrations (Adjusted R-squared = 0.78) compared to the MLR model (Adjusted R-squared = 0.63). This study not only deepens the understanding of ENV-MET interactions in cold regions, but also provides important recommendations for optimizing urban planning and heating strategies to improve air quality and thermal comfort.

中文翻译：

---

气象要素与 PM2.5 的空间异质性：寒冷城市 PM2.5 的环境-气象联合效应


量化冬季热环境和空气质量的差异，明确影响寒冷地区PM2.5浓度的主要因素，为区域供热设计和城市规划提供参考。在这项研究中，测量了行人层面的热环境参数和 PM2.5 浓度，并比较了不同城市功能区 （UFZ） 的热环境参数和浓度。此外，采用多元线性回归 （MLR） 、主成分分析 （PCA） 和主成分回归 （PCR） 分析了 PM2.5 和气温的主控因素。研究结果表明，区域小气候温度差异很大，与典型的 MET 数据相比，变化为 2.68-4.31 °C。值得注意的是，天空视野因子 （SVF） 成为对温度变化的主要影响，而 PM2.5 浓度主要由 ENV （BD、SVF、GnPR） 和 MET 因子 （Ta、RH、TSr） 共同驱动。与 MLR 模型 （调整后的 R 平方 = 0.63） 相比，PCR 模型对 PM2.5 浓度 （调整后的 R 平方 = 0.78） 的预测准确性更高。这项研究不仅加深了对寒冷地区 ENV-MET 相互作用的理解，还为优化城市规划和供暖策略以改善空气质量和热舒适度提供了重要建议。