The objectives of this project included to maintain and operate the EPA visibility monitoring network (VNM) in Banqiao, Xitun, and Xiaogang stations, develop the VNM database, audit the date quality assurance/ quality control, analyze the VNM optical monitoring instrument data, and interpret the monitoring data between EPA air quality monitoring stations and VNM.
The simultaneous in-situ measurements from three LPV-4 visibility transmissometers were regarded as the same for a 95 % confidence interval with ANOVA statistical analysis. There was also no statistically significant difference among three NGN-2A integrating nephelometer simultaneous in-situ measurements. The extinction and scattering coefficient data is above 94 % completeness. Further, the monitoring data is transferred to the CDX database of the EPA every hour. Furthermore, the difference between the instrumental monitored value, extinction coefficient of LPV-4, and the artificial observation value, visibility visual range (VR), is minilized by applying the statistic modified convert coefficient.
In the application of the absorbance coefficient, the three-dimensional indicators of CO/NOx, AAE and BC/PM2.5 are developed to distinguish the characteristics of biomass combustion and traffic source pollution. Under the same PM2.5 concentration, the extinction coefficient was proportional to the relative humidity (RH). In addition, the extinction mass efficiency also increased with the RH at the same extinction coefficient. It showed that PM2.5 affects visibility after deliquescence in high-humidity environment. In addition, this project distinguished the characteristics and occurrence frequency by PM2.5 and RH at different visibility condition. The results indicated that the hour frequency of VR < 10 km under the environment of RH 75 % and PM2.5 35 μg m-3 at Xiaogang site was higher than other sites. At Xiaogang site, the aerosol liquid water content (ALWC) in the low humidity and PM2.5 environment, simulated by the ISORROPIA model, was correlated well with the nitrate than with the sulfate. The concentrations of nitrate and ALWC were much higher at Xiaogang site than that at other two sites. Visibility degradation at Xiaogang site was strongly related to nitrate deliquescence.
In the feasibility assessment of the visibility distribution in Taiwan based on the existing measurement data, the multivariate regression method combined with the zero-order correlation and standardized regression coefficients determined the influence of pollutants and established the regression relationship between the air quality data and the extinction coefficient. The results show that the estimated extinction coefficient has a similar trend with the measured extinction coefficient, which can be used to simulate variations in visibility. The estimated extinction coefficient regressed by substituting the PM2.5 chemical compositions into the revised IMPROVE equation was well correlation with the measured value. It indicated that this method can use the PM2.5 chemical composition to estimate the visibility, and evaluate the effectiveness of reducing any chemical composition for visibility improvement.
The positive matrix factorization (PMF) was used at three sites for source apportionment of pollution. Then the extinction coefficient was reconstructed from PMF factors modified with RH to understand the source contributions of the visibility degradation in the field atmospheric environment. In the serious pollution period of winter and spring, the highest contribution factor of the reconstructed extinction coefficient at three sites was the nitrate modified with RH. The results highlight the importance of nitrate pollution source control in improving visibility policies.
