Institute for Risk Assessment Sciences, Utrecht University, 3584 CM Utrecht, Netherlands.
Professorship of Big Geospatial Data Management, Technical University of Munich, 85521 Ottobrunn, Germany.
Environ Sci Technol. 2024 Aug 13;58(32):14372-14383. doi: 10.1021/acs.est.4c06144. Epub 2024 Jul 31.
Addressing the challenge of mapping hyperlocal air pollution in areas without local monitoring, we evaluated unsupervised transfer learning-based land-use regression (LUR) models developed using mobile monitoring data from other cities: CORrelation ALignment (Coral) and its inverse distance-weighted modification (IDW_Coral). These models mitigated domain shifts and transferred patterns learned from mobile air quality monitoring campaigns in Copenhagen and Rotterdam to estimate annual average air pollution levels in Amsterdam (50m road segments) without involving any Amsterdam measurements in model development. For nitrogen dioxide (NO), IDW_Coral outperformed Copenhagen and Rotterdam LUR models directly applied to Amsterdam, achieving MAE (4.47 μg/m) and RMSE (5.36 μg/m) comparable to a locally fitted LUR model (AMS_SLR) developed using Amsterdam mobile measurements collected for 160 days. IDW_Coral yielded an of 0.35, similar to that of the AMS_SLR based on 20 collection days, suggesting a minimum requirement of 20-day mobile monitoring to capture city-specific insights. For ultrafine particles (UFP), IDW_Coral's citywide predictions strongly correlated with previously published mixed-effect models fitted with 160-day Amsterdam measurements (Pearson correlation of 0.71 for UFP and 0.72 for NO). IDW_Coral demands no direct measurements in the target area, showcasing its potential for large-scale applications and offering significant economic efficiencies in executing mobile monitoring campaigns.
为了解决在没有本地监测的地区绘制局地空气污染地图的挑战，我们评估了基于无监督迁移学习的土地利用回归（LUR）模型，这些模型使用来自其他城市的移动监测数据开发：CORrelation ALignment（Coral）及其逆距离加权修正（IDW_Coral）。这些模型减轻了领域转移，并将从哥本哈根和鹿特丹的移动空气质量监测活动中学到的模式转移到阿姆斯特丹（50m 道路段），而无需在模型开发中涉及任何阿姆斯特丹的测量值。对于二氧化氮（NO），IDW_Coral 优于直接应用于阿姆斯特丹的哥本哈根和鹿特丹 LUR 模型，实现 MAE（4.47μg/m）和 RMSE（5.36μg/m）与使用在 160 天内收集的阿姆斯特丹移动测量值开发的本地拟合 LUR 模型（AMS_SLR）相当。IDW_Coral 的 为 0.35，与基于 20 个采集日的 AMS_SLR 相似，表明需要至少 20 天的移动监测才能捕获城市特有的见解。对于超细颗粒（UFP），IDW_Coral 的全市范围预测与以前发表的使用 160 天阿姆斯特丹测量值拟合的混合效应模型强烈相关（UFP 的 Pearson 相关系数为 0.71，NO 为 0.72）。IDW_Coral 不需要在目标区域进行直接测量，展示了其在大规模应用中的潜力，并在执行移动监测活动中提供了显著的经济效益。
