One-year temporal distribution
One-year hourly observation data and daily average of air pollutants and meteorological elements were statistically analyzed, as shown in Fig. 1 and Fig. S1, the vertical axis of VOCs concentration was set logarithmic coordinates. The VOCs concentration mainly distributed within 102–103 with an annually average of 420.25 µg·m− 3 and ranged from 1.58 µg·m− 3 to 142299.83 µg·m− 3 as accidentally happened high concentration. High concentrations of VOCs are primarily observed in April and November, when moderate temperatures facilitate volatilization and atmospheric oxidation for removal. The concentration of SO2, NO2, PM2.5 and PM10 all showed relatively higher in winter and lower in summer, besides the low NO2 concentration in February due to the lower traffic of Spring Festival. The maximum hourly observed concentration of SO2, NO2, PM2.5 and PM10 reached 36.0 µg·m− 3, 120.0 µg·m− 3, 193.0 µg·m− 3 and 443.0 µg·m− 3. As to daily average concentration, SO2, NO2 and CO all met the national air quality standard. However, there were still some hours showed high particles concentration in winter and spring, exceeding the daily PM2.5 of 75 µg·m− 3, the PM10 of 150 µg·m− 3. The maximum daily concentration of PM2.5 and PM10 reached 129.5 µg·m− 3 and 180.3 µg·m− 3 in December, respectively. All daily average concentration of PM2.5 and PM10 in September and October met the air quality standard. The O3 concentration was further calculated by maximum concentration and maximum daily 8-h average (MDA8) concentration. The days that MDA8 concentration exceed the national standard of 160 µg·m− 3 mostly occurred in June and September, when owned the top two monthly average maximum and MDA8 concentration in 2021. The days with the highest maximum and MDA8 concentration of O3 reached 286.0 µg·m− 3 and 225.9 µg·m− 3 in June, respectively. Therefore, O3 is the key pollutant that should be take consideration in air quality guarantee during 2022 Hangzhou Asian Games in Deqing county.
The meteorological elements revealed typically subtropical monsoon climate of low temperature, high pressure in winter, and high temperature, low pressure, high relative humidity in summer. The highest temperature was 38.1℃ in July. The wind speed was mostly under 2 m·s− 1, while the highest wind speed reached 7.75 m·s− 1.
The concentrations of pollutants were further averagely calculated by each hour, as shown in Fig. S3. The hourly distribution of CO, SO2, PM2.5 and PM10 in a day were relatively stable, while that of NO2 showed a slight trough at 14:00. Moreover, the hourly distribution of VOCs in a day revealed two peaks at 11:00 and 17:00, with an extremely high concentration at 11:00 which partially due to plant emission of VOCs. Similar with other typical O3 daily distribution (Y. Yang et al., 2019; Zhaofeng Tan et al., 2018), the O3 concentration passed a peak and then decreased during daytime because of photochemical reactions. The O3 concentration reached the peak at 15:00 and reached 93 µg·m− 3 averagely.
O 3 formation sensitive factors
The formation of O3 can be affected by a lot of factors, including other ambient pollutants and meteorological elements. The Pearson correlations between each observed factors were calculated, as shown in Fig. 2. VOCs revealed almost no correlation to each other factors, as VOCs were consisted by amount of single VOC species from different sources with different chemical reactivity. SO2 had a negative correlation of -0.4087 to RH due to the reaction with H2O. NO2 had positive correlations to PM2.5 and PM10, since NO2 is one of the precursors of secondary particles. As to O3, the correlations to RH was the strongest, then Temp and NO2. Therefore, the scatter distribution of O3 to key factors were further analyzed. High O3 concentration mostly occurred at the meteorological conditions of high Temp above 15℃, and RH between 20–70%. Extremely high O3 concentration mostly occurred at stricter conditions of high Temp above 30℃, and RH between 30–60%. As to the key O3 precursors of VOCs and NO2, O3 acted almost no correlation to VOCs while strong correlation to NO2 of − 0.4196. It can be inferred that the O3 concentration was much more sensitive to NO2 than to VOCs, and the O3 concentration is likely to increase with the decrease of NO2 concentration. Moreover, high O3 concentration mostly occurred at the NO2 concentration ranging from 10 µg·m− 3 to 30 µg·m− 3.
Figure 2. Pearson correlation and scatter distribution of O 3 to sensitive factors
It has been proved in many studies that VOCs and NO2 affect dominantly on O3 formation among ambient pollutants(Ke Li et al., 2019; Xiao Lu et al., 2019; Vermeuel et al., 2019), but the results in Fig. 2 showed that VOCs had little correlation with O3 as a whole. Hence, further investigation was carried out with the daytime monitoring data during 7:00 to 18:00, as shown in Fig. S4, high O3 concentration mainly occurred at low NO2 concentration under 30 µg·m− 3 and distributed within the whole range of VOCs concentration. In addition, the ratio of VOCs and NO2 was calculated, as shown in Fig. S5, the ratio mostly distributed at 13.0 with an averagely ratio of 28.3, compared to other regions(Xue Yang et al., 2023; Shan et al., 2023; Qu et al., 2020), this ratio indicated the study area was most likely located at transition zone and also had some extremely high VOCs concentration conditions. According to the distribution between this ratio to NO2 and O3, as shown in Fig. S6, when NO2 concentration was lower enough, the ratios were higher than higher NO2 concentration conditions as a whole, and the ratio revealed obviously trend that decreased with the increasing of NO2 concentration. Interestingly, higher O3 concentration, higher than 200 µg·m− 3 for example, occurred at two ratio zones with low NO2 concentration, from 9.5 to 26.3 and from 99.8 to 251.6.
In addition, the ratio of VOCs and NO2 during corresponding period in 2021 was also calculated and revealed a similar distribution as that in the whole year of 2021, as shown in Fig. S7. The ratio mostly distributed at 14.8 with an averagely ratio of 21.5, when NO2 concentration was low, the ratios were higher, and the ratio revealed obviously trend that decreased with the increasing of NO2 concentration to about 200 µg·m− 3, which was higher than that distributed in the whole year of 2021. Higher O3 concentration occurred at two ratio zones with low NO2 concentration similarly.
Thus, further investigation on O3 to VOCs concentration at different levels of NO2 concentration was analyzed, as shown in Fig. 3, generally, the O3 concentration increased with the increasing of VOCs concentration in different degrees. When NO2 concentration was lower than 3 µg·m− 3, the distribution of O3 to VOCs concentration revealed highest R2 of 0.3068 and highest correlation of 0.5539 by Pearson correlation. Indicating that O3 concentration was obviously affect by VOCs at low NO2 concentration. Then, the R2 and Pearson correlation sharply decreased with the increasing of NO2 concentration, meaning that the VOCs concentration had weaker effect on O3 formation. When NO2 concentration was higher than 5 µg·m− 3, the distribution of O3 revealed almost no relation to VOCs concentration. Thus, according to the distribution among VOCs, NO2, the ratio of VOCs and NO2, and O3, when NO2 concentration was lower enough, the atmospheric condition belonged to transition zone, while NO2 concentration was higher, the atmospheric condition belonged to NO2-sensitive zone.
The O 3 formation during corresponding period
As the O3 was the key pollutant that should be take consideration in air quality guarantee for the Hangzhou Asian Games during 23, Sep. to 8 Oct. in 2020–2022, the O3 formation in the corresponding period were further investigated. As shown in Fig. S2, there were high O3 pollution days in the 3 years. The highest hourly concentration increased from 202 µg·m− 3 to 239 µg·m− 3, and to 265 µg·m− 3 by year. While the highest MDA8 concentrations were 165.5 µg·m− 3, 205.6 µg·m− 3 and 204.8 µg·m− 3 with an average concentration of 113.4 µg·m− 3, 137.3 µg·m− 3 and 115.1 µg·m− 3 in 2020–2022, indicating the year 2021 owned the worst O3 pollution. Considering the important affection from Temp and RH, the probability density distribution of Temp and RH were further analysis, as shown in Fig. 4. In the corresponding period, Temp was mostly within 20–23℃, but there was still probability of hot days with Temp higher than 30℃. The RH revealed a wide distribution during the corresponding period which also covered the range of 30–60% for extremely high O3 concentration. Therefore, based on the historical distribution of Temp and RH, high O3 concentration will likely occur without addition control strategies in this period.
NO2 concentration was another factor that the O3 concentration strongly correlated to. Interestingly, there was an evident decrease in O3 concentration before October 1st, coinciding with the extended national holiday period characterized by heavy traffic and increased NO2 emissions. Thus, the timely concentrations between NO2 and O3 were compared, as shown in Fig. 5. The concentration of NO2 revealed improvement with different degree before 1 Oct., but the decrease resulted in different expression in the change of O3 concentration. In 2021, the concentration of NO2 had a more stable increase and the O3 concentration also increased. Inversely in 2020 and 2022, the concentration of NO2 changed sharply while the O3 concentration decreased with NO2 increasing. Thus, it can be inferred that the NO2 concentration had important effect on O3 concentration, and the affection might at the transfer region between positive and negative. Therefore, the NO2 concentration should be the most crucial factor to be reasonably controlled for short-time and also long-term air quality improvement.
The reactivity of NO 2 to O 3
To investigate the response of O3 concentration change to the change of NO2 concentration, the reactivity of NO2 to O3 was further investigated by IR and RIR. For the whole year of 2021, as shown in the first panel of Fig. 6, the absolute IR of NO2 trended to be larger when NO2 concentration was within 10–30 µg·m− 3, and it expressed similar probability whether IR was positive or negative. Indicating the change in NO2 concentration at relatively low level of NO2 concentration was more likely to cause obvious fluctuation in O3 concentration. As to RIR, since higher NO2 concentration leaded to smaller change in O3 concentration, the absolute RIR of NO2 trended to be larger when NO2 concentration was within 20–80 µg·m− 3, while it also expressed similar probability whether RIR was positive or negative, according to the second panel of Fig. 6. When NO2 concentration was within 10–20 µg·m− 3, the absolute IR trended to some extremely high value. While the absolute RIR trended to some extremely high value when NO2 concentration was within 40–50 µg·m− 3. When NO2 concentration was higher than 80 µg·m− 3, the IR and RIR both trended to 0, indicating the oversaturation of ambient NO2 concentration, and O3 formation might transfer to VOCs sensitive.
The IR and RIR of NO2 to O3 concentration showed similar rules during the corresponding period in 2020–2022, as shown in the last two lines of Fig. 6, the absolute IR of NO2 trended to be larger when NO2 concentration was lower than 20 µg·m− 3, and it showed similar probability whether IR was positive or negative. On the contrary, the absolute RIR of NO2 trended to be larger when NO2 concentration was higher than 20 µg·m− 3, while it also expressed similar probability whether RIR was positive or negative. Thus, the O3 concentration responded complex to the change of NO2 concentration and obvious fluctuation in O3 concentration might occur. The local O3 pollution control is still facing severe challenges.