the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Nov 2023
Solar FTIR measurements of NOx vertical distributions: Part II) Experiment-based scaling factors describing the diurnal increase of stratospheric NO2 and NO
Abstract. Long-term experimental stratospheric NO2 and NO partial columns measured by means of solar Fourier-transform infrared (FTIR) spectromertry at Zugspitze (47.42° N, 10.98° E, 2964 m a.s.l.), Germany were used to create a set of experiment-based monthly scaling factors (SFexp). The underlying data set is published in a companion paper (Nürnberg et al., 2023) comprising over 25 years of measurements depicting the diurnal variability of stratospheric NO2 and NO partial columns in dependence of local solar time (LST). In analogy to recently published simulation-based scaling factors by Strode et al. (2022), we created SFexp normalized to local solar noon for NO2 and NO for every month of the year as a function of solar zenith angle (SZA). Beside a boundary value problem at minimum SZA values originating in averaging over different times of the month, the obtained scaling factors SFexp(NO2) and SFexp(NO) in dependence of SZA represent very well the diurnal behavior already shown in model simulations and experiment in the literature. This behavior is a well pronounced increase of the NO2 and NO stratospheric partial colum with the time of the day and a flattening of this increase after noon. In addition to the discussion of SFexp, we validate the simulation-based scaling factors SFsim(NO2) (Strode et al., 2022) and present simulation-based scaling factors for NO SFsim(NO). The simulation-based scaling factors show an excellent agreement with our the experiment-based ones, i.e. for NO2 and NO the mean bias of the modulus between experiment and simulation over all SZA and months is only 0.02 %. We show, that recently used model simulations can describe very well the real behavior of nitrogen oxide (NOx) variability in the stratosphere. Furthermore, we conclude that ground-based FTIR measurements can be used for validation of the output of photochemistry models as well as creating experiment-based data sets describing the diurnal stratospheric NOx variability in dependence of SZA. This is a contribution to improved satellite validation and a better understanding of stratospheric photochemistry.
- Preprint
(1235 KB) - Metadata XML
-
Supplement
(543 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-1437', Anonymous Referee #2, 21 Nov 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1437/egusphere-2023-1437-RC1-supplement.pdf
-
RC2: 'Comment on egusphere-2023-1437', Anonymous Referee #1, 24 Nov 2023
Review of "Solar FTIR measurements of NOx vertical distributions: Part II) Experiment-based scaling factors describing the diurnal increase of stratospheric NO2 and NO" by P. Nürnberg et al.
This paper presents observation-derived scaling factors that relate NO2 and NO stratospheric columns at a given SZA to those at another, within the range of SZA accessible by FTIR measurements at Zugspitze, at a monthly resolution. These observational scaling factors are compared to scaling factors based on an atmosphere model with photochemistry. The results show good agreement over most of the SZA range.
The construction of these observation-derived scaling factors constitutes a valuable verification of the model-based scaling factors often used to account for different photo-chemical conditions when comparing (satellite) measurements obtained at different overpass times, and opens to possibility to do a similar conversion using purely observational factors (although the latitude-specificity probably limits this use to specific comparisons at Northern mid-latitudes).
The paper is clear, concise and to-the-point, but I do feel that several of the somewhat difficult steps in the process were handled in an unexpected/sub-optimal way:
1) As remarked by referee #2, the choice to normalize with a different SZA for each month is inconvenient. It's not a show stopper, but normalizing by a fixed SZA (at Zugspitze there must be one which is covered every single day of the year), would have made more sense to me.
2) Could you not have avoided the sampling issues (the bias due to only half of the month contributing before or after the 15th in spring/autumn) by normalizing the data for each day to a fixed SZA and then taking the monthly mean (instead of the reverse order as you do now)? In that way the seasonal variation in absolute NOx levels would have been taken out before the monthly averaging.
Some more justification for your approach should be provided. If there is no good justification on point 2 (and my reasoning is not flawed), I think it's worth the effort to redo the work.
There are a few other implicit points that deserve some discussion in my opinion:- Why the cut-off at 16km, which is well above the tropopause at mid latitudes? If justified in Part 1, please briefly repeat the justification here.
- I find the use of negative SZA between noon and sunset somewhat confusing. I think you could have dropped the "minus" in the graphs and just annotated left and right with "AM" and "PM". But ok, no real need to change this.
- The experimental scaling factors are limited to true daytime, excluding the twilight regimes at sunrise and sunset. This is a limitation inherent to the measurement technique, but I think it should be made explicit that the strong and fast photochemistry at sunrise and sunset is outside the scope of these experimental scaling factors.
- The poorer comparison to the modelled scaling factors at high SZA: to what extent does your FTIR retrieval take into account the wide range in photochemical regimes along the line-of-sight at these high SZA: high up in the atmosphere, the sun is already well above the horizon, so NO2 loss has been significant already, while lower down the atmosphere is still much darker and NO2 levels still higher. Is that taken into account in the FTIR retrieval, and if so, how? If already discussed in Part 1, please summarize here as well.
- Also related to the retrieval: Does the stratospheric temperature affect your retrievals (e.g., through the NOx cross sections) and so potentially the observed diurnal cycle? Please spend a few words on this.Minor comments:
- a spell check is needed. I counted several already in the abstract.
- abstract: mean bias -> mean value
- line 123: "month 15th"-> 15th day of the month? Confirmed by a somewhat redundant sentence a couple of lines further on.
- I still think "normed" should be "normalized".Citation: https://doi.org/10.5194/egusphere-2023-1437-RC2 -
AC1: 'Comment on egusphere-2023-1437', Pinchas Nürnberg, 05 Apr 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1437/egusphere-2023-1437-AC1-supplement.pdf
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 229 | 76 | 25 | 330 | 29 | 18 | 12 |
- HTML: 229
- PDF: 76
- XML: 25
- Total: 330
- Supplement: 29
- BibTeX: 18
- EndNote: 12
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1