发展土地利用回归、扩散和混合模型，以预测巴塞罗那的室外空气污染暴露情况。

Development of land use regression, dispersion, and hybrid models for prediction of outdoor air pollution exposure in Barcelona.

机构信息

Barcelona Institute for Global Health (ISGlobal), Barcelona, Spain; Universitat Pompeu Fabra (UPF), Barcelona, Spain; CIBER Epidemiología y Salud Pública (CIBERESP), Madrid, Spain.

出版信息

Sci Total Environ. 2024 Dec 1;954:176632. doi: 10.1016/j.scitotenv.2024.176632. Epub 2024 Oct 2.

DOI:10.1016/j.scitotenv.2024.176632

PMID:39362534

Abstract

BACKGROUND

Air pollution is the leading environmental risk factor for health. Assessing outdoor air pollution exposure with detailed spatial and temporal variability in urban areas is crucial for evaluating its health effects.

AIM

We developed and compared Land Use Regression (LUR), dispersion (DM), and hybrid (HM) models to estimate outdoor concentrations for NO, PM, black carbon (BC), and PM (Fe, Cu, Zn) in Barcelona.

METHODS

Two monitoring campaigns were conducted. In the first, NO concentrations were measured twice at 984 home addresses and in the second, NO, PM, and BC were measured four times at 34 points across Barcelona. LUR and DM were constructed using conventional techniques, while HM was developed using Random Forest (RF). Model performance was evaluated using leave-one-out cross-validation (LOOCV) and 10-fold cross-validation (10-CV) for LUR and HM, and by comparing DM and LUR estimates with routine monitoring stations. NO levels estimated by all models were externally validated using the home monitoring campaign. Agreement between models was assessed using Spearman correlation (rs) and Bland-Altman (BA) plots.

RESULTS

Models showed moderate to good performance. LUR exhibited R of 0.62 (NO), 0.45 (PM), 0.83 (BC), and 0.85 to 0.89 (PM). DM model comparison showed R values of 0.39 (NO), 0.26 (PM), and 0.65 (BC). HM models had higher R 0.64 (NO), 0.66 (PM), 0.86 (BC), and 0.44 to 0.70 (PM). Validation for NO showed R values of 0.56 (LUR), 0.44 (DM), and 0.64 (HM). Correlations between models varied from -0.38 to 0.92 for long-term exposure, and - 0.23 to 0.94 for short-term exposure. BA plots showed good agreement between models, especially for NO and BC.

CONCLUSIONS

Our models varied substantially, with some models performing better in validation samples (NO and BC). Future health studies should use the most accurate methods to minimize bias from exposure measurement error.

摘要

背景

空气污染是影响健康的主要环境风险因素。在城市地区，评估具有详细时空变化的室外空气污染暴露情况对于评估其健康影响至关重要。

目的

我们开发并比较了土地利用回归（LUR）、扩散（DM）和混合（HM）模型，以估计巴塞罗那的室外 NO、PM、黑碳（BC）和 PM（Fe、Cu、Zn）浓度。

方法

进行了两次监测活动。在第一次活动中，在 984 个家庭地址处两次测量了 NO 浓度，在第二次活动中，在巴塞罗那的 34 个地点四次测量了 NO、PM 和 BC 浓度。LUR 和 DM 是使用常规技术构建的，而 HM 是使用随机森林（RF）开发的。使用留一法交叉验证（LOOCV）和 10 折交叉验证（10-CV）评估 LUR 和 HM 的模型性能，并将 DM 和 LUR 估计值与常规监测站进行比较。使用家庭监测活动对所有模型估计的 NO 水平进行外部验证。使用 Spearman 相关系数（rs）和 Bland-Altman（BA）图评估模型之间的一致性。

结果

模型表现出中等至良好的性能。LUR 表现出 0.62（NO）、0.45（PM）、0.83（BC）和 0.85 至 0.89（PM）的 R 值。DM 模型比较显示 R 值为 0.39（NO）、0.26（PM）和 0.65（BC）。HM 模型的 R 值更高，分别为 0.64（NO）、0.66（PM）、0.86（BC）和 0.44 至 0.70（PM）。NO 的验证显示 R 值为 0.56（LUR）、0.44（DM）和 0.64（HM）。模型之间的相关性从长期暴露的-0.38 到 0.92，以及短期暴露的-0.23 到 0.94。BA 图显示模型之间具有良好的一致性，特别是对于 NO 和 BC。