Quantifying and characterizing the impacts of PM2.5 and humidity on atmospheric visibility in 182 Chinese cities: A nationwide time-series study.

Ambient particulate matter less than 2.5 μm (PM 2.5) and humidity have been considered as two leading determinants of atmospheric visibility. However, quantitative evidence of their independent and interactive effects on visibility is rare at multi-city level. Based on the data of 190 monitoring stations from 182 Chinese cities, the generalized linear model combined with nonlinear smoother was constructed to estimate the station-specific effects of PM 2.5 and relative humidity on visibility. Then, multivariate meta-analysis was used to pool the station-specific estimates. Finally, attributable visibility loss respectively due to PM 2.5 and humidity were calculated. In general, both PM 2.5 and relative humidity were detected to have a significant non-linear negative influence on visibility. Per 1 μg/m3 increase in PM 2.5 was associated with a visibility reduction of 0.232 km and 0.030 km under the level of PM 2.5 below or above 68 μg/m3, respectively. The visibility declined by 0.105 km, 0.254 km, and 0.569 km per 1% increase in relative humidity under low (<58%), middle (58%–85%), and high (>85%) humidity, respectively. Significant interactive effects of PM 2.5 and humidity were observed. The impacts of PM 2.5 on visibility were significantly amplified in days with moderate humidity. However, extremely high humidity could largely attenuate the PM 2.5 -visibility association. The influence of extremely high humidity (like 99%) also consistently declined as the PM 2.5 concentration elevated. From the national perspective, PM 2.5 accounted for a higher proportion of average daily visibility loss than humidity [9.75 km (95%CI: 7.94, 11.57) vs. 7.23 km (95%CI: 5.02, 9.45)]. The relatively mild polluted southern China suffered from similar PM 2.5 -attributable visibility loss with northern China. The findings could provide important information for the prediction and improvement of atmospheric visibility. [Display omitted] • Both PM 2.5 and relative humidity had a negative non-linear effect on visibility. • Humidity between 60 and 80% significantly amplified the impact of PM 2.5 on visibility. • PM 2.5 substantially attenuated the effect of extremely high humidity on visibility. • Daily average visibility loss of 10 km and 7 km were attributed to PM 2.5 and humidity. • Visibility in less polluted but more humid areas also suffered greatly from PM 2.5. [ABSTRACT FROM AUTHOR]