Land Use Regression Models for BTEX Volatile Organic Compounds in a Middle Eastern Megacity: Tehran Study of Exposure Prediction for Environmental Health Research (Tehran SEPEHR)

Abstract

Background/Aim: Land use regression (LUR) models have not been applied, to date, to volatile organic compounds (VOCs) in highly polluted megacities. We aimed to develop LUR models for benzene, toluene, ethylbenzene, p-xylene, m-xylene, o-xylene (BTEX), and total BTEX in Tehran megacity, Iran. Methods: We advanced LUR models for BTEX and total BTEX using measurement based estimates of annual means at 179 selected sites. In total, 520 potential predictors were used in the Tehran Study of Exposure Prediction for Environmental Health Research (Tehran SEPEHR). The annual high-resolution emission inventory of VOCs and meteorological estimates from the Weather Research and Forecasting (WRF) model (temperature, humidity, and wind speed) were also evaluated as predictors. Results: The annual median (25th–75th percentile) for benzene, the most carcinogenic of the BTEX species, was 7.8 (6.3–9.9) µg/m3. The final models with R2 values ranging from 0.64 for p-xylene to 0.70 for benzene were mainly driven by traffic-related variables but distance to sewage treatment plants was present in all models indicating a major local source of BTEX VOCs in the ambient air of megacities not used in any previous study. WRF-based variables and emission inventory did not explain long-term spatial variability of BTEX VOCs in Tehran. Overall, about 83% of Tehran’s surface had predicted benzene concentrations above air quality standard of 5 µg/m3 set by European Union with maximum values up to 29 µg/m3. Conclusions: This is the largest LUR study to estimate fine-scale annual mean of all BTEX species in a megacity. These estimates could be used for health effects studies, urban planning, air quality management, and monitoring of evidence-based policy making.