Review for Okamoto et al., 2022

Okamoto et al. present in their paper an analysis of ozone plumes in the lowermost troposphere, and follow those plumes through Europe using a new satellite product IASI+GOME2. They use a chemical reanalysis TCR-2 to investigate the chemical composition and attribute sources to the plumes.

The topic of the paper is important and certainly within the scope of ACP. However, the authors need to revise the structure and logic of their manuscript (see general points). Also, it is important to motivate the usage of TCR-2 data, which often is not in agreement with validated measurements, such as IASI+GOME2 ozone, or IASI CO. In this form, the conclusions drawn from TCR-2 are not very sound. I encourage the authors to address my points below in a major revision of the manuscript.

General points:

- Overall the structure could be improved to help the reader to understand the contents: When introducing a new figure, only parts of it are sometimes mentioned in the text and others not, while references to later figures are mixed in the same part of the manuscript. I would suggest to first think, which point should be made with a certain figure, then compile the figure with all the information necessary to make the point, and introduce the whole figure at once, in order to make the point then in the text. Later in the discussion, information from different figures can be put together, but in the beginning, it would be preferable to introduce the figures in the same order as they appear in the manuscript.

- It is not clear to me, why the validation approach of the satellite data with ground based in situ measurements is performed on such a coarse spatio-temporal grid.

- Section 3 should be reorganized: First the observed plumes should be introduced, then trajectories connected to these plumes should be discussed. Figure 7 shows the IASI+GOME2 measurements the first time in this paper, but these measurements are motivating the study after all. So, I would highly recommend to start with this figure and explain it first. Then either explain the meteorological situation, trajectories or chemical reanalysis results. Or maybe even start with a figure containing several panels of IASI+GOME2 measurements for different days? This would illustrate nicely the temporal evolution of the plume(s), and demonstrate the strength of the observations, which are the base of this work.

- It is not well motivated, why the TCR-2 model is useful to explain the chemical situation. For temporal and horizontal comparisons in ozone between IASI+GOME2 and TCR-2, there are large differences (see Figures 4a/b, 7a/b, 9a/b, 10a/b, 11a/b), which are not explained sufficiently enough here. Sometimes, plumes are marked in the satellite data by boxes, and these plumes are not reproduced by TCR-2. Further, the only comparison of CO between TCR-2 and IASI shows problems in TCR-2. So how can the authors draw conclusions for origin of the pollution, which are only based on this model data, which seems not to match the observations?

- For some figures, fonts (on color bars or axes) are very small.

Specific points:

- Line 13: Define "IASI" and "GOME2"

- Line 38: "The main sources of tropospheric ozone ...": This is a very long sentence, and it is not quite clear, if "in situ photochemical production through oxidation of CO and CH4" is a main source, or if these two parts should be considered separately. Please try to make this sentence shorter and easier to read.

- Line 101: "Ozone concentrations from the surface to 3 km of altitude (a.s.l.) are provided as volume mixing ratio in ppb (parts per billion)": So, this is rather an average value for the ozone concentration below 3 km altitude?

- Line 104 and following: What kind of measurements are these surface observations of ozone? Are these in-situ measurements? Which kind of measurement technique has been used? Some background information would be useful here.

- Line 119: Could you please motivate the time period for the comparison?

- Line 118 and following: I did not really understand, which quantities are compared here: Do the authors compare averages over the whole time period? Or are data compared in a finer temporal resolution? Which horizontal resolution of the satellite data is compared to ground measurements? Please indicate a bit more in detail the settings of these comparisons.

- Line 123: Please define "p".

- Line 127: Are these numbers given in ppb considered to be an average bias for the given regions? How are these biases calculated from the correlations shown in Fig. 2?

- Line 130: "but they may reflect differences in the spatio-temporal representativity of surface in situ measurements and satellite data at daily scale and this particular event.": So maybe, it would make more sense to make this correlation based on a finer temporal-spatial resolution? Why did the authors choose a different approach?

- Line 162: Please define the acronym MIROC-CHASER.

- Line 167; "32 vertical levels from the surface to 4.4 hPa.": How many of these layers are in the region of interest of this study (lowermost troposphere below 3 km)?

- Line 182: Why are you not smoothing by averaging kernel? Are the vertical resolutions of IASI+GOME2 and TCR-2 comparable?

- Line 188: Suggestion to rephrase: "Cuesta et al. (2018) used the same data product jointly with IASI+GOME2."

- Line 242: Suggestion to rephrase: "... daily mean windspeeds lower than i) 3.2 m s–1 at 10 m and ii) lower than and 13.0 m s-1 at 500 hPa, and iii) daily total precipitation less than 1.0 mm."

- Lines 250 and following: "The major ozone outbreak travelling across Europe is formed by three ozone plumes originating from the Iberian Peninsula, Western Europe and North America. Three-day forward trajectories from HYSPLIT depict the pathway of these ozone plumes (Fig. 3).": In Fig. 3, foreward trajectories are shown, which originate in northern Germany, central France and the Iberian Peninsula. How are these three plumes (named "Norhternmost plume", "Middle latitude plume", and "Southermost plume") connected with the plumes originating from the Iberian Peninsula, Western Europe and North America, which are mentioned in the text? The authors should tell the reader, how they know, where the plume comes from (if not from the plots shown), and further should be consistent with the naming of the plumes.

- Line 251: Figure 3 is poorly introduced and it is not clear at this point of the paper, why this figure is needed at all.

- Line 252 and following: The definition of the "plumes" is not very clear. Are these plumes connected with the trajectories shown in Fig. 3? Later on, there are references to the plume definitions, but they use figures, which have not been introduced at this point.

- Line 256: What is considered to be a "high concentration" of ozone? Do the authors use a threshold here? What is the threshold?

- Line 257: I suggest to provide the threshold for informing the population also in ppb, since this unit has been used throughout the paper so far. Further, all of the shown measurement and simulation results are well below 90 ppb, so I would not agree to the formulation "often near 90 ppb)

- Figure 4: I miss a discussion, which compares the measured ozone to the TCR-2 simulated ozone. For me it looks like that there are considerable differences between panels a and b, and that it should be well motivated, why it is still useful to look into other TCR-2 species, if the simulated ozone is backed so poorly by the measurements.

- Line 294: Suggestion: "As typical for Iberian summer, ..."

- Figure 5b/c: I suggest to mark the latitude/longitude (respectively), where the other cross section is located.

- Figure 5e: This panel is not mentioned in the text. The authors should introduce it in the text or remove this panel, if it is not important for the manuscript.

- Figure 6, caption: typo: blown -> brown

- Figure 6b: y-axis label "Percentage" is not very informative and should be replaced (e.g. by "heatwave extension/air stagnation")

- Line 341: "... lack of sensitivity (or spatial coverage) in the satellite data (particularly over the ocean).": Is anything like such a lack of sensitivity known to be typical for the IASI+GOME2 data? Please give a reference here. Further, how much does the mismatch of observed and simulated ozone affect the following interpretation, which is only based on model data? What about mismatches between IASI+GOME2 and TCR-2 ozone above land (e.g. north-west Africa, or Turkey)?

- Line 384: "In addition, this outbreak may also be affected by downward transport from middle troposphere": I don't think that this has been explained in Section 3.1.1. I also cannot see such an event in Figures 5b/c

- Line 401 following: "It is worth noting...": Again, this sounds like there is a problem with the TCR-2 data. Why is it still useful to look into this data in the remaining of this manuscript?

- Figure 9 (and similar figures): It would be helpful to repeat the colored boxes, which mark the plumes also for other species than ozone. This would help to identify the air masses associated with the plumes.

- Figure 9g/h: Please state in the caption the differences between those two panels.

- Line 413: "However, this ozone-enriched air masses are co-located with clearly higher CO, NO2 and HCHO concentrations ...": The co-located CO plume simulated by TCR-2 is not visible in IASI measurements, as shown in Fig. 9d. In fact, this panel is not mentioned at all in the text, but it seems to be very important, since it may highlight a problem in the TCR-2 data.

- Line 442: "This ozone plume shows rather moderate concentrations ...": Please give an example or a range for "moderate concentrations".

- Line 470: "Significant enhancements of CO are seen ...": In Fig. 9, it was shown that TCR-2 CO may considerably disagree with IASI measurements. How is performing TCR-2 in this region in comparison to IASI?

- Line 539: "... show high concentrations of ozone precursors suggesting a significant impact of anthropogenic emissions ...": This statement is too general here: It depends on the specific precursor to state anthropogenic origin. I think in this case, the authors want to refer to enhanced CO levels for this plume. However, this enhanced CO in TCR-2 was shown to be not very robust compared to direct IASI measurements.

- Line 542: "The satellite approach highlights significant photochemical production of ozone ...": How do the satellite measurements allow for an attribution of ozone production to photochemical production? I understood it more like the difference between IASI+GOME2 and TCR-2 suggested photochemical production?

- Line 545: The beginning of this paragraph is rather repetitive from the beginning of this section.

Overview

The paper deals with the impact of the various sources of ozone precursors on the evolution of ozone concentrations during a high ozone episode over Europe in July 2017. I think that the paper presents an interesting analysis of data originating from various sources such as in-situ and satellite measurements as well as modelling. In my opinion, the manuscript is generally scientifically sound, and it deserves publication in ACP after considering the recommendations listed below. 

General comments

I would suggest that, in addition to the direct photochemical ozone production at the surface and the boundary layer, which is quite properly presented and interpreted, the role and the influence of upper tropospheric ozone (generally at higher levels) to the lower tropospheric/boundary layer ozone should be more considered, especially under prevailing anticyclonic conditions, when it is reported that large scale subsidence movements might occur, especially at the edge of the anticyclones and at the interface with low pressure systems, given also the fact that the IASI+GOME2 satellite measurements are most sensitive at 2-3 km height, which is generally the free-tropospheric level. According to relatively recent publications the variability of free tropospheric ozone, especially over the Central and Eastern Mediterranean basin during summer (Kalabokas et al., 2013; Doche at al., 2014; Zanis et al., 2014; Akritidis et al., 2016; Gaudel et al., 2018) could be better understood, if the variability of synoptic meteorological conditions, affecting especially vertical ozone transport are considered. These characteristics are also detected in Central Europe during the warm period of the year by analyzing also IASI and IASI+GOME2 satellite measurements, although they seem to be more enhanced during spring months (Kalabokas et al., 2017; Kalabokas et al., 2020).

So, I would suggest examining also the IASI satellite upper tropospheric ozone and eventually the corresponding TCR-2 simulations as well as the charts of geopotential height and the omega vertical velocity at least, during the peak phase of the ozone episode (16-21 July 2017) and for the 700 and 500 hPa pressure levels. It seems that during this period, significant and extended downward movements are observed starting from the west and moving eastwards, following the evolution of episode, which might explain better the observed phenomena.

 

Specific comments

Page 16, line 345-347: I think that this factor should be more stressed by examining the daily omega vertical velocity measurements over the troposphere, as also noted above. 

Page 20, lines 401-404: As noted in general comments, the plotting of omega vertical velocity charts would be helpful for the better assessment of these observations.

Page 22, lines 424-427: As indicated earlier, the described eastward movement of air masses seem to be also accompanied with significant downward transport throughout the troposphere, so the influence of the upper layers on ozone concentrations, have to be also investigated.

Page 25, lines 481-486: It should be added that the persistent summer northerlies, known as Etesians, in the boundary layer over the Aegean are also associated with strong large-scale subsidence in the lower troposphere occurring simultaneously, thus significantly influencing surface ozone concentrations, as it is suggested form relevant publications based on vertical and satellite measurements over the area (Kalabokas et al., 2007; Kalabokas et al., 2013; Doche at al., 2014; Zanis et al., 2014). 

Page 29, lines 542-544: Based also on the above remarks this discrepancy could be due to the effect of large scale tropospheric subsidence, which is very frequently observed over this region in July.