*2.3. Quality Assurance and Control*

The SUMMA canister was cleaned 3 times with high-purity nitrogen with an automatic de-tanker (Entech 3100) before each sampling and pumped to a vacuum state, so that the pressure in the tank was less than 50 millitorr, and was ready for later. A cleaned SUMMA canister was injected with high-purity nitrogen as a laboratory blank, and the laboratory blank test was performed before the analysis of each batch of samples.

In the sample analysis process, 1 parallel sample was analyzed for every 10 samples measured, and the relative deviation in the target in the parallel sample was ≤30%. The correlation coefficients of the target compounds in the standard curve were >0.995. The retention time of the internal target in the sample deviated from the retention time of the internal standard in the continuous or recently drawn calibration curve of the day by no more than 20 s. Every 24 h, we analyzed the middle concentration point of the standard curve (1 × 10<sup>−</sup>9); the measurement result was ≤30% of the initial mass concentration value; otherwise, the cause should be found or redrawn. The results showed no contamination during sample handling and collection, as assured by the quality assurance and control (QC/QA) procedures.

## *2.4. Analytical Methods*

## 2.4.1. Ozone Formation Potentials (OFP)

Ozone formation potentials can be used to evaluate the potential of VOCs emissions participating in the reaction to generate ozone, and they provided some guidance for the formulation of VOCs control measures. In this study, the maximum incremental reactivity method (MIR) was used to determine the contribution of active components and key species in VOCs to O3 production [21,22], calculated as in Equation (1):

$$\rm{OFP\_i} = \rm{VOCs\_i} \times \rm{MIR\_i} \tag{1}$$

where OFPi represents the amount of ozone generation potential of species i in μg·m<sup>−</sup>3; VOCsi represents the mass concentration of species i in μg·m<sup>−</sup>3; and MIRi represents the MIR coefficient of species i [23].

#### 2.4.2. HYSPLIT Model

The HYSPLIT model(https://ready.arl.noaa.gov/HYSPLIT\_traj.php, accessed on 1 July 2021) is a comprehensive model system developed by the National Oceanic and Atmospheric Centre (NOAA) and the Australian Meteorological Agency (BOM). It can be used to calculate and analyze processes such as airflow movement, sedimentation, air pollutant transport, and diffusion trajectory [24]. At present, it has been widely used to study the transmission routes and source analysis of air pollutants [25–28].

In this paper, TrajStat follow-up software [29] was used to analyze and study the backward trajectory of air masses in Jinghong. As the 500 m height wind field accurately reflects the average flow field characteristics of the boundary layer [27], the simulated height was chosen as 500 m. Jinghong (100◦47 38" E, 22◦00 07" N) was the simulated receiving point, 8:00 (Beijing time) every day was the pushback start time, and the 72 h backward trajectory of the receiving point from July 2016 to June 2017 was calculated, to reflect the characteristics of the airflow.

#### 2.4.3. Health Risk Assessment

In order to study the potential harm of VOCs to human health in Jinghong City, it is necessary to assess the health risk of VOCs. This study adopted a new health risk assessment method (EPA-540-R-070-002) proposed by the U.S. EPA in 2009 for inhaled route pollutants in specific places. The calculation formula is as follows:

$$\text{EC} = \frac{\text{CA} \times \text{ET} \times \text{EF} \times \text{ED}}{\text{AT}} \tag{2}$$

$$\text{HQ} = \frac{\text{EC}}{(\text{RFC} \times 1000)} \tag{3}$$

$$\text{IR} = \text{EC} \times \text{IUR} \tag{4}$$

$$\text{HI} = \sum \text{HQ}\_{\text{i}} \tag{5}$$

where EC is the exposure concentration in units of μg·m<sup>−</sup>3; CA is the ambient concentration of VOCs in μg·m<sup>−</sup>3; ET is the exposure time in h·d<sup>−</sup>1, with a value of 24; EF is the exposure frequency in d·a<sup>−</sup>1, with a value of 365; ED is the exposure time in a, with a value of 70; AT is the average time in h, with a value of 70 × 365 × 24; HQ is the noncarcinogenic risk hazard quotient value; RfC is the reference concentration in μg·m<sup>−</sup>3; R is the lifetime risk value of carcinogenicity, IUR is the inhalation risk in μg·m<sup>−</sup>3; and HI is the hazard index. The RfC and IUR values are from the reference [30].

#### 2.4.4. Data Sources

This study used daily averaged and hourly data on the concentrations of six atmospheric pollutants (SO2, NO2, CO, O3, PM2.5, and PM10) at two environmental monitoring stations in Jinghong from July 2016 to June 2017. Contaminant data were from the National Environmental Monitoring Center of China (https://air.cnemc.cn:18007/, accessed on 1 July 2021). The data used in the backward trajectory model were 2016–2017 Global Data Assimilation System (GDAS) data provided by the NCEP (National Center for Environmental Prediction). Meteorological data were derived from the National Meteorological Science Data Sharing Service Platform (http://data.cma.cn/, accessed on 1 July 2021).
