the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Oct 2025
Quantifying and addressing the uncertainties in tropospheric ozone and OH in a global chemistry transport model
Abstract. The major physical and chemical processes governing the abundance of atmospheric oxidants such as ozone and hydroxyl radicals (OH) are largely understood, but quantitative assessment of their importance in different environments remains challenging. Atmospheric chemistry transport models allow exploration of these processes on a global scale, but weaknesses in process representation in these models introduces uncertainty, and model intercomparisons show considerable diversity even in representing current atmospheric composition. Formal constraint of models with atmospheric observations is needed to provide more critical insight into the causes of model weaknesses. In this study we perform a global sensitivity analysis on a chemistry transport model using Gaussian process emulation and identify the processes contributing most to uncertainty in tropospheric ozone and OH. We then explore the use of atmospheric measurements to calibrate the model and identify weaknesses in process representation and understanding. We find that the largest uncertainties are associated with photochemical kinetic data and with factors governing photolysis rates and surface deposition. Calibration constrains the uncertainty in key processes, informing model development and improving comparisons with observations, but we show that it is also valuable in identifying structural errors in models. We show that surface ozone measurements alone provide insufficient constraint, and we highlight the importance of applying a broad range of different observational metrics. While this study is exploratory in nature, focussing on a limited number of constraints, we clearly demonstrate the value of rigorous calibration for providing important new insight into key processes and their representation in atmospheric models.
- Preprint
(1052 KB) - Metadata XML
-
Supplement
(504 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-4534', Anonymous Referee #1, 15 Oct 2025
-
RC2: 'Comment on egusphere-2025-4534', Anonymous Referee #2, 09 Nov 2025
The manuscript demonstrates the potential for an uncertainty quantification framework built on Gaussian process emulation to identify specific processes most in need of constraining. An impressively large amount of work underlies the results, first identifying key sensitivities with brute force model simulations (60 processes), then developing and training emulators from a subset of 36 processes to target global tropospheric ozone, methane lifetime, surface and 500 hPa ozone sampled at 60 10x10 degree gridded observations. While the processes emphasized here have been pointed out in prior work as driving differences across models or between observations and models, the novelty is the ranking among individual parameters within the categories of emissions, deposition, meteorology, photolysis, and chemistry. The manuscript will be a useful contribution though some clarification noted below would help readers better interpret the findings. Importantly, the authors are providing the perturbation simulations and emulator codes in an archive, which makes them useful to benchmark future work.
General comments.
It’s tempting to interpret the dominant sensitivities as causal, but is it possible that inter-correlations among the 36 tested processes in the training confound a clean interpretation? What assumptions are made regarding temporal and spatial correlations especially considering variations in ozone lifetime in space and time? What are the ‘attractive statistical properties’ in line 79? Is there an assumption that ozone is locally controlled, and if so, where is that likely to work best? Transport seems like a critical missing piece.
Is the main point that the approach is generalizable but the findings model-specific, or are sensitivities identified here applicable to the real atmosphere or other models? For example, the study design maps out which errors would matter most in which region given the base state distribution. So if that base state distribution were to change dramatically, so too would the rankings identified here?
Specific Comments.How are NOx and HOx reservoirs handled by the emulator? Murray et al. (2021) showed strong cross-model sensitivity of global OH to the treatment of NOy and how much emitted carbon is oxidized all the way to CO2, which is understandably beyond the processes studied here and discussed in lines 485-489 but could be acknowledged. Where does this model mechanism fall in terms of chemical complexity? In particular, understanding of isoprene chemistry has advanced rapidly in recent years (e.g., Bates and Jacob, 2019); are those updates in this model?
Line 44. Observational constraints are being used to investigate tropospheric oxidation processes, most notably around field campaigns and satellite data. Some examples: Nicely et al., 2016; Bourgeois et al.,2021; Pimlott et al., 2022; Anderson et al., 2023; Baublitz et al., 2023; Guo et al., 2023, Mirrezaei et al., 2025 and others.
Lines 62-65. Is the sensitivity of CH4 to H2 at present levels correct, and if so, is there a soil sink for H2 in the model?
Section starting line 72. How long was the model spun up? That comes later but might fit here.
Table 1. Consider adding the base case values for parameters to make it easier for readers to contextualize the findings. Is isoprene the only non-anthropogenic VOC emitted?
Line 138. Deposition is likely a structural issue; for example see a community review paper suggesting most global models do not include dry deposition processes consistent with latest understanding from the field and lab (Clifton et al., 2020).
Lines 237-238. Recommended guidance here?
Line 250. ACCMIP models all used the same anthropogenic emissions
Line 264. Does the larger error in the ozone burden relative to surface reflect error accumulation and propagation given the longer ozone lifetime away from the surface?
Lines 280-284. It makes sense that uncertainty in a particular process should matter most where that process dominates. Could that be illustrated?
Figure 3. It would be interesting to include some selected free troposphere locations.
Line 306. Do the pdfs from the independent chains end up overlapping? Do we have to worry about identifying local minima during the training?
Lines 435. Are these scalings provided in the zenodo repository?
Lines 505-510. Is some of this implicitly accounted for by training separately for each month?
Line 513 or elsewhere: consider noting that emulation should be a simpler problem for short-lived OH ?
Technical Corrections
Line 161. precursors --> reservoirs ?
Line 230. 1536 (64 locations x 12 months x 2 layers)? Either way, please briefly explain where this value is coming from.
Line 475. refined --> confirmed?
References
Anderson, D. C., Duncan, B. N., Nicely, J. M., Liu, J., Strode, S. A., and Follette-Cook, M. B.: Technical note: Constraining the hydroxyl (OH) radical in the tropics with satellite observations of its drivers – first steps toward assessing the feasibility of a global observation strategy, Atmos. Chem. Phys., 23, 6319–6338, https://doi.org/10.5194/acp-23-6319-2023, 2023.
Bates, K. H. and Jacob, D. J.: A new model mechanism for atmospheric oxidation of isoprene: global effects on oxidants, nitrogen oxides, organic products, and secondary organic aerosol, Atmos. Chem. Phys., 19, 9613–9640, https://doi.org/10.5194/acp-19-9613-2019, 2019.
Baublitz et al., An observation-based, reduced-form model for oxidation in the remote marine troposphere, Proc. Natl. Acad. Sci. U.S.A. 120 (34) e2209735120, https://doi.org/10.1073/pnas.2209735120 (2023).
Bourgeois et al., Large contribution of biomass burning emissions to ozone throughout the global remote troposphere, Proc. Natl. Acad. Sci. U.S.A. 118 (52) e2109628118, https://doi.org/10.1073/pnas.2109628118 (2021).
Clifton, O. E. et al. (2020). Dry deposition of ozone over land: processes, measurement, and modeling. Reviews of Geophysics, 58, e2019RG000670.
Guo et al., Heterogeneity and chemical reactivity of the remote troposphere defined by aircraft measurements – corrected, Atmos. Chem. Phys., 23, 99–117, https://doi.org/10.5194/acp-23-99-2023, 2023.
Murray et al., Large uncertainties in global hydroxyl projections tied to fate of reactive nitrogen and carbon, Proc. Natl. Acad. Sci. U.S.A. 118 (43) e2115204118, https://doi.org/10.1073/pnas.2115204118 (2021).
Mirrezaei et al. Toward Realistic Prognostic Modeling Of The Methane Chemical Loss. ESS Open Archive . July 19, 2025. DOI: 10.22541/essoar.175288325.58779694/v1
Nicely, J. M., et al. (2016), An observationally constrained evaluation of the oxidative capacity in the tropical western Pacific troposphere, J. Geophys. Res. Atmos., 121, 7461–7488, doi:10.1002/ 2016JD025067.
Pimlott, et al. Investigating the global OH radical distribution using steady-state approximations and satellite data, Atmos. Chem. Phys., 22, 10467–10488, https://doi.org/10.5194/acp-22-10467-2022, 2022.
Citation: https://doi.org/10.5194/egusphere-2025-4534-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 256 | 49 | 9 | 314 | 38 | 8 | 7 |
- HTML: 256
- PDF: 49
- XML: 9
- Total: 314
- Supplement: 38
- BibTeX: 8
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of egusphere-2025-4534
Wild & Ryan
“Quantifying and addressing the uncertainties in tropospheric ozone and OH in a global chemistry transport model”
This thoughtful and novel study by Wild and Ryan centers on using a chemistry-transport model to understand how the uncertainties in our current knowledge of the physics and chemistry in the model translate into uncertainties in the model generated tropospheric O3 and OH (the major sink of CH4). The current state of understanding is readily seen in the recent model intercomparison projects (MIPs), where the budgets for O3 and CH4 show wide spreads. So, yes, this is an important area of study for both air quality and climate.
The authors perform an intense series of objective sensitivity simulations with the CTM involving 60 some identified uncertainties and reduce this to 36 for more serious study. This is the most thorough such analysis that this reviewer has seen. The authors then use “Gaussian process emulators” to further study the full range of coupled uncertainties in order to map out cost functions for matching some key large-scale observational constrains (surface and 500 hPa O3, total tropospheric O3 burden, CH4 lifetime).
This work is a valuable contribution to the community’s understanding of the how and why our models do not always give us what we want. The only major problem is the denseness of the manuscript and the difficulty in keeping track of the specific uncertainties. I have the following suggestions for minor rewrites.
Starting with Table 1. This table provides the core data and assumptions used here. The problem is that the reader has little chance of remembering all the 3 letter abbreviations later in the paper. Can the authors come up with a reasonable, abbreviated table for the 36 key uncertainties that could be included in the legend of some of the later figures? Shorthand examples:
fir BB all
iso biogenic C5H8
hna Henry’s HNO3
xfm X-section HCHO
Some of the paragraphs are very dense and difficult to follow. Would it be possible to segment these, maybe with bulleted points to give us the sequence of the logic?
L5: “even when we all agree on the physical or chemical rate constants” Yes, indeed, the diversity is distressing.
L39: But these studies did not look at photochemistry since they did not do cross sections!
L112: Yes, it is good you are dealing with uncertainties in the numbers we often assume are fixed.
L160: Thanks for doing the cross sections.
L163: is it good to see a clear definition of you uncertainty ranges. You might also note than when you report an uncertainty factor, it is boeh / and x, effectively you are treating the internal range as lon-normal??
L172 of the 2-sigma (95% range) uncertainty. Your uncertainty prob distrib is not stopped at 2 sigma.
Fig. 1: I really like Fig. 1 and would expect near perfect symmetry in the up/down perturbations, but items like VOC emissions are asymmetric, presumably because you did log normal with a large factor. You should comment on that.
L255: I am a bit worried about the apportionments based on log-normal ranges/
Fig 2: This really needs a legend with a somewhat expanded explanation for the 3 letter designations.
Fig 3: ditto
Fig.7: Ditto. This paddle figure is challenging for me, but OK
L325ff: You are worried about the vertical resolution in comparing surface O3, but I think the observations are MDA8 or daytime and so a thin nocturnal BL is not a problem. I would be more worried about the horizontal resolution because you cannot resolve urban centers.
L341ff: This is an interesting statement about structural errors. We all certainly have them, But, do they apply in a similar way to all models/
Fig 7: What is the color code here? More important, why does everything seem to be pushed away from the prior?
L411: Nice case study, but too bad it did not work out better.
Fig 8: Again, why is everything (except ddf & ddg) pushed far off the prior? Some discussion.
L476ff: Why not do trop O3 burden as NH and SH, since there are distinct differences with separate causes presumably
L482: Note that errors in J-NO2 would do the same thing?
L492: Great, major conclusion!