In this paper, the authors measure vertical profiles of ozone and its precursors concentrations. The authors measure directly net ozone production rate P(O3)net at ground level, and discuss ozone concentration variations in terms of both photochemical ozone production and physical transportation using measured P(O3)net and ozone concentrations. In addition, they compare observed and modeled values for P(O3)net and discuss the vertical distribution of P(O3)net and ozone production regimes calculated from the model. The discussion on the ozone budget and its vertical distribution is very important to mitigate ozone pollution problems, so that I recommend this paper to be published in ACP. However, I found several concerns to be published in the present form, so the authors should perform appropriate revisions sufficiently.

Major comments:

Line 225: NO2 and NOx concentrations measured by commercially available NOx analyzer include NOz species such as PAN and HNO3. I think this is a large problem because NO2 and NOx are important ozone precursors. If this is no problem for the authors, they should prove that there is no problem. For example, an intercomparison of NO2 concentrations measured by the CAPS and chemiluminescence methods should be performed.

Fig. 5: Why are there significant P(O3)net (not zero) in the nighttime? What is the precision of P(O3)net measured by this instrument? This should be discussed. Since Ox concentrations derived from the reaction chamber and reference chamber are measured alternately by solenoid valves, large fluctuations in ambient Ox concentrations are expected to cause poor precision.

Figs. S2 and 6: For P(O3)net, there are cases where the model agrees with the observation and cases where it does not. Why? The authors should discuss in depth? For IOA, NMB, and NME, the authors state their values during the whole measurement period only. What about the values of these parameters for each episode? Episode I and III may be good, but are the other episodes adequately reproduced, as described in Lines 571-573? Also, I think the discussion on the accuracy also concern the accuracy of the discussion on the vertical profiles of ozone budgets and ozone production regime described in Figs. 7 and 9.

Other minor comments:

Line 61-63: The authors should explain ozone production regime in more detail.

Line 100: The authors should define OBM-MCM.

Sections 2.1 and 2.2.1: I think it would be easier for the readers to understand if the authors explain the details of the SZMGT and sampling method at SZMGT, using schematic diagrams in supplement.

Line 149: O3 + NO = NO2 → O3 + NO → NO2 (This is a chemical reaction, not an equation)

Section 2.2.3: What kinds of VOCs did the authors measure? Listed in Table S2? If so, the authors should refer to Table S2 in the text.

Line 233: at 424 nm → less than 424 nm?

Line 289: In order to investigated → In order to investigate

Fig. 1 and Table 1: How did the authors measure CO and TVOCs? And the authors should define TVOCs.

Line 363-365: Is this sentence made during the daytime?

Line 454: Sect. 3.3.1 → Sect. 3.1.1?

The manuscript of Zhou et al. reports novel information on the O₃ budget and chemistry at a monitoring site in the PRD region (China). This work takes advantage of ozone production rates measurements performed at the ground level during one month in Nov.-Dec. 2021 to infer the contributions of both photochemical and transport processes to the local ozone budget. In addition, the authors use a rich dataset of trace gas measurements performed at multiple heights (up to 335m) and 0-D box modeling to investigate the vertical distribution of P(O₃). The ozone formation chemistry is investigated at the different heights, highlighting that ozone production occurs in the VOC-limited and transition regimes most of the time.  The authors conclude that emission regulations focusing on the reduction of AVOCs and OVOCs should help reducing O₃ at this location.

While the manuscript is well structured, the writing needs to be revised before publication. Some suggestions are provided below but there are more instances in the manuscript where improvements are needed.  The methodology and the results seem scientifically sound and the authors provide novel information that will be of interest for the atmospheric community. I recommend publication after the authors have addressed the writing issues and the following comments:

Major comments:

• L172-178: This section is confusing. The authors indicate that a light-enhanced loss of O3 is corrected for but present an equation to compute an uptake coefficient for O3. It’s not clear how the correction is done. The authors should clarify how this uptake coefficient is derived and how it is considered when computing P(O3)net from Eq. 1. The amplitude of this correction should also be clearly stated. It would be useful to add time series of P(O3)net with and without correction in the supplementary material to show how this correction changes over time.
• L180-181: “The limit of detection (LOD) of the NPOPR detection system is 2.3 ppbv h-1 at the sampling air flow rate of 5 L min-1, which is obtained as three times the measurement error of P(O3)net.” – It’s not clear how the authors derive a LOD from the error associated to P(O3)net. This error will scale with P(O3)net. Please clarify.
• L194-195: “Therefore, we corrected the measured P(O3)net using the quantified P(O3)net in the reference chamber.” – How was P(O3)net quantified in the reference chamber?
• L223-225: The authors should address the specificity of their NO2 measurements since chemiluminescence instruments also detect some NOy species in the NO2 channel. Were O3, NO and NO2 corrected for O3+NOàNO2 in the sampling line? If so, please indicate the amplitude of this correction.
• Section 2.2.3: The authors should provide some details about the GC measurements. It is stated that an offline GC was used. How were the VOCs sampled? How were the sampled analyzed? How and how often were calibration and zeroing done on the GC instrument?
• Section 2.3: The authors should add a subsection to explain how Ozone Formation Potential (OFP) values are computed. OFP values are reported in Table 1 and Figure 3, and discuss in the result section.
• Section 2.3.1: The authors should indicate which chemical species were constrained in the box model and how they were constrained. It is not clear how the authors deal with ozone in the model. In section 3.2.2, the authors compare simulated ozone concentrations to field observations, which seems to indicate that measured ozone concentrations were not directly constrained in the model. However, if the O3 advected to the site was not constrained, the simulations would very likely not reproduce the measured ozone concentrations. How did the authors constrain the ozone transported to the observation site in their model?
• Figure 4: The authors should indicate how R(O3)tran is derived. Is it computed as d[O3]/dt – P(O3)net ? Please add error bars on the time series of d[O3]/dt, P(O3)net and R(O3)tran.
• L535: Dilution was constrained in the model using a species lifetime of 12h. How sensitive are the simulation results to this parameter? Please indicate how modelled P(O3) changes when the lifetime is varied from 6 to 24 h. How does it affect the main conclusions?
• L541-543: Some values are provided for NMB and NME without addressing what it means for the model performance. The authors should comment these values in the text.

Minor comments:

• L68-71: The authors should provide a brief summary of what is known about ozone formation in the PRD region.
• L121-122: What is the air residence time in the sampling lines?
• L158-168: This section is not necessary here. Please just indicate that pulse experiments were performed to quantify the residence time in the chambers and cite the paper of Hao et al. (2023).
• L208-210: Please indicate the frequency and duration of zero measurements.
• L212: Please provide the E/N value.
• L233: “the photolysis of NO2 at 424 nm” – please provide a range of wavelength instead of a unique wavelength.
• L550: Please remove ”and in Indiana in the United States (~ 30 ppbv h-1 in spring) (Sklaveniti et al., 2018)”. Sklaveniti et al. did not measure ambient P(O3) but investigated the sensitivity of P(O3) to NO additions in the instrument.
• L570: Please rephrase. It’s not clear what is meant by “underestimate the NOx limited regime”
• L793-794: “The maximum estimated error of modelled P(O3)net ranged from 22-45 % during different episodes and non-episodes.”. This has not been discussed in the manuscript before the general conclusion. The authors should discuss this point in more details the manuscript. How is the 22-45% error estimated?

Edits:

• L32: “photochemical reactions play a dominate role” should read “photochemical reactions playing a major role”
• L56: “Tropospheric ozone (O3), which have adverse effects on ecosystems” should read “Tropospheric ozone (O3), which has adverse effects on ecosystems”
• L58: “important factor resulting severe regional air pollution” should read “important factor resulting in severe regional air pollution”
• L59: “Tropospheric O3 mainly comes from the external transport from the stratosphere” should read “Tropospheric O3 mainly comes from stratospheric intrusions”
• L87-88: “thus largely hindered our in depth understanding of” should read “thus largely hindering our understanding of”
• L90: “observation system based on the Shenzhen Meteorological Gradient Tower” should read “observation system located on the Shenzhen Meteorological Gradient Tower”
• L92: “To diagnose the P(O3)net and O3 formation” should read “To diagnose the net ozone production rate, P(O3)net, and O3 formation”
• L94: “with the Master Chemical Mechanism (MCM v3.3.1).” should read “with the Master Chemical Mechanism (MCM v3.3.1), referred to as OBM-MCM in the following.”
• L104: “while acknowledging potential biases associated modelling.” Should read “while acknowledging potential biases associated to the modelling. »
• L172: “[O3] represents the difference » should read “d[O3] represents the difference”
• L 173 & L184: “(Hao et al., 2013) » should read “(Hao et al., 2023).”
• L198: “mass spectrometry » should read « mass spectrometer”
• L241-247: Please rephrase. This sentence does not have a conjugated verb.
• L268: “P(O3)net denotes the net photochemical O3 production rate (ppbv h-1)” should read “P(Ox)net denotes the net photochemical O3 production rate (ppbv h-1)”
• L297: “different kinds of VOCs groups together to investigated their influence to the gradient P(O3)net change with heights in Sect. 3.2.3.” should read “different kinds of VOC groups together to investigate their influence on the vertical gradient of P(O3)net in Sect. 3.2.3.”
• L454: “As concluded in Sect. 3.3.1” should read “As concluded in Sect. 3.1.1”
• L458: “As the dry deposition are usually contribute” should read “As dry deposition usually contributes”
• Legend Figure 5: “R(O3)net” should read “R(O3)tran”
• L504: “concentration became stable, suggests that the photochemical reaction competed against physical transport and jointly affect O3 concentration change” should read “concentration became stable, suggesting that the photochemical reaction competed against physical transport and jointly affected O3 concentration change”
• L 506: “the O3 concentration decreases due to the diffuse of photochemically formed O3” should read “the O3 concentration decreases due to the transport of photochemically formed O3”
• L514: “with O3 diffuse to » should read “with O3 transport to”
• L558: “presence of missing RO2 under high NO conditions” should read “underestimation of RO2 under high NO conditions”
• L569-570: “OVOCs photolysis (Wang et al., 2022) in modelling approach, may result in the underestimation of RO2, thus underestimate the modelled P(O3)net” should read “OVOCs photolysis (Wang et al., 2022) in modelling approaches, may result in the underestimation of RO2, thus underestimating the modelled P(O3)net”
• L700: “heights, indicates the similar photochemical O3 formation regime” should read “heights, indicating a similar photochemical O3 formation regime”
• L708: “during polluted episode І, both reduce VOCs and NOx” should read “during polluted episode І, reducing both VOCs and NOx”
• L724: “which located in the » should read “which is located in the”
• L731: “ROx radicals cycle reactions involved Nox” should read “ROx radicals cycle reactions involving NOx”
• L754: “Given that NOx has a significant titration effect on ozone” should read “Given that NO has a significant titration effect on ozone”
• L766: “with local photochemical reactions play a dominate role” should read “with local photochemical reactions playing a key role”
• L771: “the measurement period, indicated the” should read “the measurement period, indicating the”
• L774: “differences of measured and modelled P(O3)net” should read “differences between measured and modelled P(O3)net”