Influence of Photochemical Loss of VOCs on Understanding Ozone 1 Formation Mechanism 2

Volatile organic compounds (VOCs) tend to be consumed by atmospheric oxidants, 19 resulting in substantial photochemical loss during transport. An observation-based model was 20 used to evaluate the influence of photochemical loss of VOCs on the sensitivity regime and 21 mechanisms of ozone formation. Our results showed that a VOC-limited regime based on 22 observed VOC concentrations shifted to a transition regime with a photochemical initial 23 concentration of VOCs (PIC-VOCs) in the morning. The net ozone formation rate was 24 underestimated by 3 ppb h -1 (~36 ppb day -1 ) based on the PIC-VOCs. The relative contribution 25 of the RO 2 path to ozone production based on the PIC-VOCs accordingly increased by 13.4%; 26 in particular, the contribution of alkene-derived RO 2 increased by approximately 10.2%. In 27 addition, the OH-HO 2 radical cycle was obviously accelerated by highly reactive alkenes after 28 accounting for photochemical loss of VOCs. The contribution of local photochemistry might 29 be underestimated for both local and regional ozone pollution if consumed VOCs are not 30 accounted for, and policymaking on ozone pollution prevention should focus on VOCs with a 31 high reactivity.

observed VOC concentrations shifted to a transition regime with a photochemical initial 23 concentration of VOCs (PIC-VOCs) in the morning. The net ozone formation rate was 24 underestimated by 3 ppb h -1 (~36 ppb day -1 ) based on the PIC-VOCs. The relative contribution 25 of the RO2 path to ozone production based on the PIC-VOCs accordingly increased by 13.4%; 26 in particular, the contribution of alkene-derived RO2 increased by approximately 10.2%. In 27 addition, the OH-HO2 radical cycle was obviously accelerated by highly reactive alkenes after 28 accounting for photochemical loss of VOCs. The contribution of local photochemistry might 29 be underestimated for both local and regional ozone pollution if consumed VOCs are not 30 accounted for, and policymaking on ozone pollution prevention should focus on VOCs with a 31 9 ki is the second-order reaction rate between compound i and OH radical; and [OH] and ∆t are 160 the concentration of OH radical and the photochemical ageing time, respectively. kX and kE are 161 rate constants for the reaction between OH radicals and ethylbenzene (7.0010 -12 cm 3 162 molecule -1 s -1 ) and xylene (1.8710 -11 cm 3 molecule -1 s -1 ) ( were well correlated ( Figure S4), which indicated that they were simultaneously emitted; 2) 171 they had different degradation rates in the atmosphere; and 3) the calculated PICs were in good 172 agreement with those calculated using other tracers, such as i-butene/propene ( Figure S5) 173 (Zhan et al., 2021). To test the relative constant emission ratio from different sources, we chose 174 benzene vs. acetylene and n-hexane vs. toluene as references, and the result is shown in Figure  175 S6. These ambient ratios could directly reflect their relative emission rates from sources 176 (Goldan et al., 2000;Jobson et al., 2004). The linear correlation coefficients (R 2 ) were generally 177 higher than 0.7, which were equal to that reported by Shao et al. (2011). To further test the 178 assumption that the emissions of xylene and ethylbenzene were constant throughout the day, 179 their potential sources were calculated using a source-receptor model (the potential source 180 contribution function, PSCF). As shown in Figure S7, xylene and ethylbenzene showed similar 181 distributions. In addition, the ratio of ethylbenzene/xylene at 5:00 and 6:00 was similar to that 182 during the daytime. These results indicated that the emissions of xylene and ethylbenzene were 183 constant throughout the day. The ratio of xylene to ethylbenzene and the OH exposure 184 concentration are shown in Figure S8. The results showed that the ratio of xylene to 185 ethylbenzene increased gradually (07:00~12:00), which is consistent with the trend of xylene 186 and ethylbenzene. The OH exposure was from 0.82 to 8.1×10 6 molecule cm -3 h, with a mean 187 daytime value of 4.3±1.9×10 6 molecules cm -3 h. Accordingly, the mean photochemical ages 188 were 1.7±0.9 h using the mean daytime (8:00-17:00 LT) OH concentrations (4.3±3.1×10 6 189 molecules cm -3 ) calculated based on JO1D using the method reported in our previous work 190 (Liu et al., 2020b;Liu et al., 2020c). This meant that VOCs would undergo obvious degradation 191 even during a short range of transport in the atmosphere. 192 It should be noted that the kOH of isoprene is 9.9810 -11 cm 3 molecule -1 s -1 at 298.15 K 193 (Atkinson and Arey, 2003), almost two orders of magnitude greater than other VOCs. The ratio 194 method assumes constant emissions for VOCs. However, the emission of isoprene greatly 195 depends on temperature and solar irradiation intensity (Zhang et al., 2021b). In addition to 196 accounting for photochemical loss, additional correction of daytime isoprene concentrations 197 was performed using the average diurnal flux of isoprene emissions ( Figure S9 (Zhang et al., 198 2021b). The emission of isoprene showed a clear unimodal curve, and the volume 199 concentration of isoprene was calculated based on the daily emission curve using Eq. (S1). 200 11

Observation-based model simulation 201
A box model based on the Master Chemical Mechanism (MCM3.3.1) and the Regional 202 Atmospheric Chemical Mechanism (RACM2) was used in this study. The MCM3.3.1 was used 203 to understand the instantaneous ozone formation process, and the RACM2 was used to depict 204 the ozone isopleth due to its high computational efficiency (Sect. 2.4). Table S1  and halohydrocarbons (7)). The model was validated using the observed and simulated O3 209 concentrations, which showed good consistency, as shown in Figure S10. The slope and 210 correlation coefficients were 0.9 and 0.8, respectively ( Figure S11, respectively, indicating the 211 validity of the model simulation. It is worth mentioning that the results of model simulation 212 can sometimes be overestimated or underestimated to some extent, which has also been 213 reported by previous studies (Zong et al., 2018;Zhang et al., 2020), but this did not affect our 214 simulations of the ozone formation process and mechanisms because we constrained the ozone 215 concentration during our simulations. 216 The ozone formation rate P(O3) can be quantified by the oxidation rate of NO to NO2 by 217 peroxyl radicals (Tan et   interval of 60 minutes, which was enough for NOx, OH, HO2, and RO2 to reach a steady state 244 because the typical relaxation time of the chemical system is 5-10 minutes in summer (Tan et 245 al., 2018). However, all the species and parameters were input at a 5 min interval by data 246 interpolation to reduce simulation inconsistencies and large distortions of meteorological 247 parameters at longer time intervals (Tan et al., 2018). The ozone production rate was calculated 248 as described in Sect. 2.3. It is worth mentioning that the average survey data were selected as 249 the baseline scenario in simulating the EKMA curve in this study. 250

Influence of photochemical loss of VOCs on the O3 formation sensitivity regime 278
The sensitivity of O3 formation is analysed using the isopleth diagram generated from the 279 EKMA model, which is widely used to qualitatively study O3-NOx-VOCs sensitivity. As and ozone production shifted from VOC-limited to NOx-limited conditions from morning to 290 afternoon, which was consistent with the mean diurnal profiles (Figure 1). This was similar to 291 the data reported in Wangdu (Tan et al., 2018). As expected, ozone production shifted from a 292 VOC-limited regime (the observed VOCs) to a transition regime based on the PIC-VOCs in 293 the morning. Ozone production clearly moved further to a NOx-limited regime in the afternoon 294 after the photochemically consumed NOx and VOCs had been accounted for (Figure 2). 295 Because the average photochemical ageing time was only 1.7±0.9 h, these results indicated 296 that the O3 formation mechanism might typically be misdiagnosed, which misleads mitigation 297 black and blue colours represent the observed and corrected statuses, respectively. 304

Contribution of VOC species to O3 production 305
The time series of simulated OH, HO2, and RO2 concentrations were used to calculate the 306 P(O3) and L(O3). The diurnally averaged P(O3) and L(O3) are shown in Figure 3. Ozone 307 formation can be divided into processes related to RO2+NO and HO2+NO (Sect. 2.3). 308 According to their VOC precursors, peroxyl radical groups were divided into alkane-derived 309 (ALKAP), alkene-derived (ALKEP), aromatic-derived (AROMP), isoprene-derived (ISOP), 310 oxygenated-VOC-derived (OVOCP), and halohydrocarbon-derived (HALOP) RO2 and HO2P. 311 The ozone destruction processes included the reaction between O3 and HOx (O3D1), the 312 reaction between O1D and H2O (O3D2), the reaction between O3 and alkenes (O3D3), and the 313 reaction between NO2 and OH (O3D4). 314 Based on the observed VOCs (or PIC-VOCs), a fast O3 production rate was observed at 315 14:00 (or 13:00), with a diurnal maximum value of 16.1 (or 25.6) ppb h -1 (Figure 3a and 3b), 316 while the peak destruction rate was 6.4 (or 8.6) ppb h -1 at 15:00 (or 13:00) (Figure 3c and 3d). 317 The average daytime P(O3) from 07:00 to 19:00 based on the initial concentrations of VOCs 318 was 4.03.1 ppb h -1 higher than that based on the measured VOCs concentrations (Figure 3b). 319 At the same time, the F(O3) from 07:00 to 19:00 based on the initial concentrations of VOCs 320 was also 3.02.1 ppb h -1 higher than the measured counterpart ( Figure S12). Thus, the net O3 321 production could be accumulatively underestimated by ~36 ppb day -1 from 07:00 to 19:00 if 322 the consumption of VOCs was not considered. This meant that the contribution of the local 323 formation of O3 could be underestimated using the directly measured VOCs concentrations. It 324 should be pointed out that it is better to compare O3 production with the true metric for O3 325 production. However, it is impossible to directly measure the true metric for O3 production in 326 the atmosphere at the present time to know how well the method presented here corrects for 327 that underestimation. In addition, the ozone concentrations must be constrained when to directly compare the ozone production based on PIC-VOCs with that using measured VOCs 330 18 concentrations. Therefore, we alternatively compared the integrated net ozone production rates 331 rather than ozone production or concentrations between the two scenarios. An upwind O3 and 332 VOCs measurement combined with a trajectory analysis might provide an approach for 333 checking the accuracy of our results. Alternatively, conducting a transient O3 production rate 334 analysis after subtracting the transport of O3 with a regional model and/or satellite observation 335 might be another option. Unfortunately, neither the upwind measurement nor the regional 336 model simulation was available at the time of our study. To further check the accuracy of our 337 results, we chose August 4 th as a test case to explore the influence of the transport of ozone on 338 a downwind site based on the trajectory analysis. As shown in Figure S13, the mean ozone 339 concentration of the downwind site (national monitoring station, NMS) was 27.6±21.9 ppb 340 day -1 higher than that of the observation site (OS), which was slightly less than the difference 341 (~36 ppb day -1 ) between PIC-VOCs and observed VOCs and indirectly rationalized our results. 342 The HO2 path contributed 64.8% to the total ozone formation on average, which was 343 slightly higher than the reported value (57.0%) in Wangdu (Tan et al., 2018), whereas the RO2 344 path, in which aromatics (9.4%), alkenes (8.4%), isoprene (7.8%), alkanes (4.7%), OVOCs 345 (4.3%) and halohydrocarbons (0.6%) were the main contributors, contributed to the remaining 346 part. For the PIC-VOCs, the dominant path of O3 production (51.7%) was still the HO2 path, 347 followed by the RO2 path related to alkenes (14.7%), aromatics (12.8%), and isoprene (11.7%). 348 The relative contribution of the RO2 path to P(O3) increased by 13.4% compared with the 349 measured VOCs, particularly alkene-derived RO2, which increased by 10.2%. As shown in 350 Figure 3c and 3d, the destruction of total oxidants was dominated by the reaction between O3 351 and alkenes (O3D3) in the morning. It gradually shifted to the reaction between NO2 and OH 352 (O3D4) from 11:00 to 16:00 and the photolysis of O3 followed by a reaction with water (O3D2) 353 from 12:00 to 15:00 because O3 concentration increased while NO2 decreased (Figure 3c). 354 Figure  on the PIC-VOCs. For example, the reaction between ozone and alkenes based on initial VOC 383 concentrations (percentages inside the brackets) contributed more to OH (from 7% to 21%) 384 and HO2 radical production (from 6% to 12%), while photolysis of HONO and HCHO 385 contributed less to the production of OH (from 76% to 60%) and HO2 radicals (from 44% to 386 40%), respectively. Other radical sources were consistent between the two scenarios. 387 Interestingly, the average formation rates of OH, HO2 and RO2 radicals derived from the PIC-388 VOCs were obviously higher than those from the observed VOCs. In particular, the oxidation 389 of NO by RO2 and HO2 increased by 1.6 and 1.3 ppb h -1 , respectively. The enhanced oxidation 390 rate of NO was equal to the increase in the average F(O3) in the analysis process above. This 391 meant that the radical propagation of OH-RO2-HO2 sped up in the case of PIC-VOCs, 392 subsequently accelerating the chemical loop of NO-NO2-O3. For the radical sinks and equilibria 393 related to HNO4, RONO2 and PAN, the values were basically comparable between the two 394 scenarios. In addition, the O3 formation from the RO2 path increased by 4.1% (from 39.5% to 395 43.6%) in the simulation using the PIC-VOCs compared with the observed VOCs. The above 396 budget analysis explained the observed increases in F(O3) (~3 ppb h -1 ), which were mainly 397 driven by the reaction of missed reactive VOCs, such as alkenes, with O3. 398 VOCs under daytime conditions (07:00 to 19:00 LT). The green, black, red and yellow boxes 401 denote the sources of radicals, radical sinks, radical propagation, and racial equilibrium, 402 respectively. The numbers or percentages outside and inside the brackets are the average 403 formation rates (ppb h -1 ) or relative contributions to a specific radical of the corresponding 404 reaction path based on observed VOCs and PIC-VOCs, respectively. 405

In situ O3 formation process 406
In addition to chemical processes, which can be simulated using the OBM-MCM model, 407 transport processes, including horizontal, vertical transportation and dry deposition processes 408 respectively.) 420 The O3 budget analysis was performed during an O3 pollution episode (Aug. 1 st ). Figure  421 5 shows the simulated local ozone formation process based on the measured and PIC-VOCs. 422

24
The hourly variation in O3 concentrations from 19:00 to 6:00 the next day was dominated by 423 regional transportation without O3 formation, while local photochemical O3 formation could 424 explain all or part of the O3 concentration change during the time window from 07:00 to 19:00. 425 The d(O3)/dt shows an increase from 07:00 to 15:00 LT. However, d(O3)/dt sharply changed to 426 negative values at 16:00, which was consistent with diurnal O3 (the O3 peaks at 15:00) in Figure  427  rate based on the PIC-VOCs was slightly higher than that based on the measured VOCs, while 436 both rates were within a range of 2.0~6.5 ppb h -1 . From 12:00 to 17:00, the O3 formation rate 437 based on the PIC-VOCs and the observed concentration of VOCs greatly increased due to 438 active photochemistry. 439 As shown in Figure 5, the increased O3 concentration was larger than the local O3 440 photochemical production from 07:00 to 12:00 (R(O3) was positive). This was mainly because,