the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Nov 2025
Ozone Risk to Forests and Crops under Drought Modulation: A 15 years Flux-Based and Economic Loss Assessment for Saxony, Germany
Abstract. Tropospheric ozone (O3) at ground level is a phytotoxic pollutant that affects vegetation and reduces crop productivity, with implications for forest ecosystems, agriculture, and food security. The present study presents a 15-year assessment (2006–2020) of O3 risk in Saxony, Germany, using the stomatal flux-based metric PODySPEC. POD1SPEC was applied to forests (spruce and beech) and grasslands, while POD6SPEC was used for croplands (wheat). Risk estimations were conducted under two scenarios: a worst-case, assuming unrestricted irrigation and a best-case incorporating modelled soil water content (SWC). Given Saxony’s extensive forest cover and the sensitivity of high-elevation ecosystems, a detailed forest evaluation was performed. POD1SPEC in spruce and beech frequently exceeded critical levels, with values up to 70 % higher at mountain than rural sites. While stomatal O3 uptake declined in dry years at rural sites, likely due to drought-induced closure, forests at mountain sites sustained O3 stomatal uptake even during prolonged droughts, reflecting drought tolerance. The number of dry days, used as a proxy for drought duration, helped explain these contrasting responses. Grasslands were also consistently in the high-risk zone, with POD1SPEC exceeding critical levels throughout the time series. Under worst-case assumptions, potential reductions reached ~9 % for above-ground biomass and ~16 % for flower numbers, with impacts about 20 % higher at mountain than rural sites. These findings suggest that meteorological conditions strongly modulate O3 uptake in grassland systems. For wheat, estimations under worst-case conditions indicate yield reductions of up to 14 % at mountain sites and 7 % at rural sites, corresponding to average annual economic losses of about €66 million and €34 million, respectively, based on the 2016–2020 producer price. Assuming similar losses under the 2025 wheat price, the economic loss increases by 13 %. These results highlight the importance of site-specific, flux-based O3 risk assessments for guiding air quality and land-use policies in Saxony. More broadly, the approach offers a framework for evaluating O3 impacts in other mountain regions where agriculture is essential and adaptive capacity is limited.
- Preprint
(1338 KB) - Metadata XML
-
Supplement
(616 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2025-5542', Anonymous Referee #1, 04 Dec 2025
- AC1: 'Reply on RC1', Hartmut Herrmann, 06 Mar 2026
-
RC2: 'Comment on egusphere-2025-5542', Anonymous Referee #2, 17 Jan 2026
Engelhardt et al. present an ozone risk assessment for forests and cropland for the German state of Saxony, based on in-situ ozone observations in cropland and mountainous regions. They consider exposure- and dose-based ozone impact metrics, and for the latter, they examine the impact of soil moisture. The choice of impact metric can affect the risk assessment, and therefore this topic is relevant for scientific and policy purposes related to air quality, but this manuscript unfortunately does not offer a large contribution to this topic. The methodology is not clearly described, the manuscript is poorly structured making it difficult to distill the key messages, and the presentation of the results is insufficient. Therefore, I cannot recommend publication of this article. Below, I list some key shortcomings that the authors may wish to address going forward.
Major issues
- Ozone flux calculation in the Methods section: the soil moisture function and the soil moisture models, key components of this study, are merely introduced by referring to another document. This document is not open-access and is written in German, so this is not accessible for the (English-speaking) reader. Therefore, this information must be included in this manuscript. Additionally, the authors acknowledge that the Jarvis (1976) formulation of stomatal conductance does not include the effect of SWC (line 215-216), but it is unclear how they apply this model to study the impact of soil moisture on ozone fluxes.
- Role of drought: the authors correctly identify that soil moisture may modulate ozone fluxes in the case of drought, and that ozone risk may be reduced in case of drought. However, droughts also affect vegetation productivity and crop yield. In turn, droughts may affect surface ozone concentrations by reducing stomatal uptake (e.g. Lin et al., 2020) and by changing emissions of biogenic VOCs (e.g. Peñuelas & Staudt, 2010). I would encourage the authors to consider the risks of drought and ozone jointly in their analysis.
- The added value of the statistical assessment of the drivers of the ozone flux is unclear. The main conclusions merely seem to confirm pre-existing knowledge on the drivers of ozone uptake by vegetation. As an example, the authors write: “Number of dry days as a key modulator of O3 uptake” (l 575). This is incorrect, it is not the number of dry days that modulates ozone uptake, but rather the meteorological/environmental variables that affect stomatal conductance (in the Jarvis formulation: vapor pressure deficit and soil water content).
- Presentation of results
- The graphical abstract unclear. I am unable to distill the key findings of the paper. What do the arrows and their colors indicate?
- The different subsections in the Methods section are not well linked, making it difficult to understand how the different analysis steps relate to each other. As an example, the link between the use of the dose-response function and the economic impact analysis is unclear.
- The resolution of the figures is poor. This can be easily improved by saving the figures at a higher resolution.
- At various points in the manuscript, the text is highly redundant. Please consider writing more to the point.
Minor issues
Line 173: what does 'good' mean?
Line 204-205: y is not defined. Do you mean Y?
Line 215-216: Abbreviations for the meteorological variables are not defined. How do you convert these meteorological variables to their equivalent values at canopy height?
Equation 2: Please remove the units from the equations, and use more descriptive variable names (e.g. to replace 'potential maximum reduction rate').
Line 247: 'best-case' and 'worst-case' scenarios do not provide any information on what these scenarios actually mean. Please use more descriptive variable names.
Section 2.6: why don't you use soil moisture data from observations or from re-analysis datasets?
Equation 5: this is not how an equation should be written.
Line 336-338: where can I see the evidence for this?
Line 356-357: where can I see the evidence for this?
Line 357-359: this is a reference to a result from another study. I don't think this belongs in the Results section.
Line 388: "the observed greater biomass reduction in these areas". What type of biomass observations were used here?
Line 390-392: again, this is a reference to a result from another study. I don't think this belongs in the Results section.
Caption of Table 2: "maximum potential reduction rate". Maximum reduction of what variable?
Line 407-409: where can I see the evidence for this?
Line 777-778: I don't see any results for crop yield loss in Figure 7.
Line 778-779: It's unclear how you arrive at this CPL estimate.
Line 780: "pre-industrial times". Was the crop yield the same in pre-industrial times as in the present? I think you refer to a scenario in which crop yield is the same as in the present, but ozone mixing ratios are kept at their pre-industrial values. It's important to be specific!
Line 803-804: No information in the paper on how this is calculated.References
Jarvis, P. G. (1976). The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 273(927), 593–610. https://doi.org/10.1098/rstb.1976.0035.
Lin, M., Horowitz, L. W., Xie, Y., Paulot, F., Malyshev, S., Shevliakova, E., Finco, A., Gerosa, G., Kubistin, D., & Pilegaard, K. (2020). Vegetation feedbacks during drought exacerbate ozone air pollution extremes in Europe. Nature Climate Change, 10(5), 444–451. https://doi.org/10.1038/s41558-020-0743-y.
Peñuelas, J., & Staudt, M. (2010). BVOCs and global change. Trends in Plant Science, 15(3), 133–144. https://doi.org/10.1016/j.tplants.2009.12.005.
Citation: https://doi.org/10.5194/egusphere-2025-5542-RC2 - AC2: 'Reply on RC2', Hartmut Herrmann, 06 Mar 2026
Status: closed
-
RC1: 'Comment on egusphere-2025-5542', Anonymous Referee #1, 04 Dec 2025
Authors took advantage of long dataset from various O3 concentration measurements in stations in Germany, some of them being in higher altitudes in mountainous regions. They calculated AOT40 and PODy for wheat and two forest tree species. In addition, they calculated the potential reduction of above ground biomass and corresponding economic loss. The paper is written in bad and complicated English with vague sentences bearing no or little information. The text is highly redundant. In addition, the methodology is highly unclear and sometimes authors are writing contradictory information. Moreover, PODy calculation would highly benefit from calculation based on flux measurements, rather than only concentration with subsequent modelling, otherwise causing large uncertainties. Due to methodology flaws and bad non-scientifically written text, I recommend to reject the paper.
line 37: spruce does not have deeper rooting
line 64-65. The sentence seems abrupt.
line 66: check dots or commas in the sentence
line 71: what kind of receptors?
line 126: in line 11 you have Easter Germany and here eastern Germany. Please be consistent.
2.2 data set or dataset?
line 167: it would be useful to add here instruments that measure O3 concentration, which height above ground, length of inlet lines and how frequently they were calibrated.
line 170. check dots within the sentence
line 173: what should that mean that data were good? Be specific
line 175: this is confusing. If that dataset was not used, why are mentioning that here? Why it is in Table 1, when later you do not use that? Please if you did not used this dataset, then delete that from Figure 1 and do not mention that anywhere in the manuscript.
line 184-187: no need to mention that you tested other ways, if they are not present here. Although it might have taken you some portion of time, do not mention that here.
line 189: calculation of calculation--please rephrase
line 212: Firstly, you calculated stomatal O3 flux from concentration and later you summed to calculate PODy - so why here in methods you did not describe the process in this way? Please write the methods as you were processing the data.
line 214: O3 load means O3 concentration?
line 215: please add equations here
line 217: Does it mean that you did not measure those variables at canopy height? It might differ a lot when measured at different height. How did you convert those variables?
line 221: should be specimen
section 2.3 make it shorter. You have here too many redundant passages and sentences are too long an vague.
line 230: table 2 - what should that mean?
line 232: this is in contradiction to what you have stated above (line 208), that the critical level is defined as the plant´s detoxification level. Moreover, how did you estimate the preindustrial stomatal ozone flux?
line 237-238: please be consistent. Exposure means concentration? What should then mean O"3 exposure situation concerning the exceedance of the CLPODySPEC" how should be concentration being concerning the flux? It is very unclear here.
line 240: "is denoted as Ref10𝑃𝑂𝐷𝑌𝑆𝑃𝐸𝐶" - this is very critical how you calculated possible stomatal O3 flux during preindustrial era. However, here you do not describe that at all! This is a major shortcoming.
line 240: "The percentage effect caused by O3" - you mean concentration or flux?
line 242: why you mix here comparison to 1980? Without any interconnection to the comparison before preindustrial era.
line 243-245: This is completely unclear. You should write more clearly and describe abbreviations at their first use - as in line 242.
line 255: before in the methods you wrote, that you did measure SWC, in here you write that you did model that. So what is true?
2.5 most of that is redundant
276-278: write better English - too complicated sentence
Eq 5. this is no equation. There must be physical variables.
286: write better English
lines 293-296: how exactly did you calculate this? Please ass equation. Could you add comparison to eddy covariance measured data?
line 305-309: so you did a pearson correlation test or PCA? Here you mix both together. very unclear what you have done.
line 316: Abbreviation should be defined at its 1st use, not again and again. Please check through the manuscript.
- do not describe here what is in next sections! You should write shortly, not writing redundantly!
In here I stop reading, the above written issues are too major for this paper to be considered for publication.
Citation: https://doi.org/10.5194/egusphere-2025-5542-RC1 - AC1: 'Reply on RC1', Hartmut Herrmann, 06 Mar 2026
-
RC2: 'Comment on egusphere-2025-5542', Anonymous Referee #2, 17 Jan 2026
Engelhardt et al. present an ozone risk assessment for forests and cropland for the German state of Saxony, based on in-situ ozone observations in cropland and mountainous regions. They consider exposure- and dose-based ozone impact metrics, and for the latter, they examine the impact of soil moisture. The choice of impact metric can affect the risk assessment, and therefore this topic is relevant for scientific and policy purposes related to air quality, but this manuscript unfortunately does not offer a large contribution to this topic. The methodology is not clearly described, the manuscript is poorly structured making it difficult to distill the key messages, and the presentation of the results is insufficient. Therefore, I cannot recommend publication of this article. Below, I list some key shortcomings that the authors may wish to address going forward.
Major issues
- Ozone flux calculation in the Methods section: the soil moisture function and the soil moisture models, key components of this study, are merely introduced by referring to another document. This document is not open-access and is written in German, so this is not accessible for the (English-speaking) reader. Therefore, this information must be included in this manuscript. Additionally, the authors acknowledge that the Jarvis (1976) formulation of stomatal conductance does not include the effect of SWC (line 215-216), but it is unclear how they apply this model to study the impact of soil moisture on ozone fluxes.
- Role of drought: the authors correctly identify that soil moisture may modulate ozone fluxes in the case of drought, and that ozone risk may be reduced in case of drought. However, droughts also affect vegetation productivity and crop yield. In turn, droughts may affect surface ozone concentrations by reducing stomatal uptake (e.g. Lin et al., 2020) and by changing emissions of biogenic VOCs (e.g. Peñuelas & Staudt, 2010). I would encourage the authors to consider the risks of drought and ozone jointly in their analysis.
- The added value of the statistical assessment of the drivers of the ozone flux is unclear. The main conclusions merely seem to confirm pre-existing knowledge on the drivers of ozone uptake by vegetation. As an example, the authors write: “Number of dry days as a key modulator of O3 uptake” (l 575). This is incorrect, it is not the number of dry days that modulates ozone uptake, but rather the meteorological/environmental variables that affect stomatal conductance (in the Jarvis formulation: vapor pressure deficit and soil water content).
- Presentation of results
- The graphical abstract unclear. I am unable to distill the key findings of the paper. What do the arrows and their colors indicate?
- The different subsections in the Methods section are not well linked, making it difficult to understand how the different analysis steps relate to each other. As an example, the link between the use of the dose-response function and the economic impact analysis is unclear.
- The resolution of the figures is poor. This can be easily improved by saving the figures at a higher resolution.
- At various points in the manuscript, the text is highly redundant. Please consider writing more to the point.
Minor issues
Line 173: what does 'good' mean?
Line 204-205: y is not defined. Do you mean Y?
Line 215-216: Abbreviations for the meteorological variables are not defined. How do you convert these meteorological variables to their equivalent values at canopy height?
Equation 2: Please remove the units from the equations, and use more descriptive variable names (e.g. to replace 'potential maximum reduction rate').
Line 247: 'best-case' and 'worst-case' scenarios do not provide any information on what these scenarios actually mean. Please use more descriptive variable names.
Section 2.6: why don't you use soil moisture data from observations or from re-analysis datasets?
Equation 5: this is not how an equation should be written.
Line 336-338: where can I see the evidence for this?
Line 356-357: where can I see the evidence for this?
Line 357-359: this is a reference to a result from another study. I don't think this belongs in the Results section.
Line 388: "the observed greater biomass reduction in these areas". What type of biomass observations were used here?
Line 390-392: again, this is a reference to a result from another study. I don't think this belongs in the Results section.
Caption of Table 2: "maximum potential reduction rate". Maximum reduction of what variable?
Line 407-409: where can I see the evidence for this?
Line 777-778: I don't see any results for crop yield loss in Figure 7.
Line 778-779: It's unclear how you arrive at this CPL estimate.
Line 780: "pre-industrial times". Was the crop yield the same in pre-industrial times as in the present? I think you refer to a scenario in which crop yield is the same as in the present, but ozone mixing ratios are kept at their pre-industrial values. It's important to be specific!
Line 803-804: No information in the paper on how this is calculated.References
Jarvis, P. G. (1976). The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 273(927), 593–610. https://doi.org/10.1098/rstb.1976.0035.
Lin, M., Horowitz, L. W., Xie, Y., Paulot, F., Malyshev, S., Shevliakova, E., Finco, A., Gerosa, G., Kubistin, D., & Pilegaard, K. (2020). Vegetation feedbacks during drought exacerbate ozone air pollution extremes in Europe. Nature Climate Change, 10(5), 444–451. https://doi.org/10.1038/s41558-020-0743-y.
Peñuelas, J., & Staudt, M. (2010). BVOCs and global change. Trends in Plant Science, 15(3), 133–144. https://doi.org/10.1016/j.tplants.2009.12.005.
Citation: https://doi.org/10.5194/egusphere-2025-5542-RC2 - AC2: 'Reply on RC2', Hartmut Herrmann, 06 Mar 2026
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 288 | 161 | 27 | 476 | 49 | 19 | 19 |
- HTML: 288
- PDF: 161
- XML: 27
- Total: 476
- Supplement: 49
- BibTeX: 19
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Authors took advantage of long dataset from various O3 concentration measurements in stations in Germany, some of them being in higher altitudes in mountainous regions. They calculated AOT40 and PODy for wheat and two forest tree species. In addition, they calculated the potential reduction of above ground biomass and corresponding economic loss. The paper is written in bad and complicated English with vague sentences bearing no or little information. The text is highly redundant. In addition, the methodology is highly unclear and sometimes authors are writing contradictory information. Moreover, PODy calculation would highly benefit from calculation based on flux measurements, rather than only concentration with subsequent modelling, otherwise causing large uncertainties. Due to methodology flaws and bad non-scientifically written text, I recommend to reject the paper.
line 37: spruce does not have deeper rooting
line 64-65. The sentence seems abrupt.
line 66: check dots or commas in the sentence
line 71: what kind of receptors?
line 126: in line 11 you have Easter Germany and here eastern Germany. Please be consistent.
2.2 data set or dataset?
line 167: it would be useful to add here instruments that measure O3 concentration, which height above ground, length of inlet lines and how frequently they were calibrated.
line 170. check dots within the sentence
line 173: what should that mean that data were good? Be specific
line 175: this is confusing. If that dataset was not used, why are mentioning that here? Why it is in Table 1, when later you do not use that? Please if you did not used this dataset, then delete that from Figure 1 and do not mention that anywhere in the manuscript.
line 184-187: no need to mention that you tested other ways, if they are not present here. Although it might have taken you some portion of time, do not mention that here.
line 189: calculation of calculation--please rephrase
line 212: Firstly, you calculated stomatal O3 flux from concentration and later you summed to calculate PODy - so why here in methods you did not describe the process in this way? Please write the methods as you were processing the data.
line 214: O3 load means O3 concentration?
line 215: please add equations here
line 217: Does it mean that you did not measure those variables at canopy height? It might differ a lot when measured at different height. How did you convert those variables?
line 221: should be specimen
section 2.3 make it shorter. You have here too many redundant passages and sentences are too long an vague.
line 230: table 2 - what should that mean?
line 232: this is in contradiction to what you have stated above (line 208), that the critical level is defined as the plant´s detoxification level. Moreover, how did you estimate the preindustrial stomatal ozone flux?
line 237-238: please be consistent. Exposure means concentration? What should then mean O"3 exposure situation concerning the exceedance of the CLPODySPEC" how should be concentration being concerning the flux? It is very unclear here.
line 240: "is denoted as Ref10𝑃𝑂𝐷𝑌𝑆𝑃𝐸𝐶" - this is very critical how you calculated possible stomatal O3 flux during preindustrial era. However, here you do not describe that at all! This is a major shortcoming.
line 240: "The percentage effect caused by O3" - you mean concentration or flux?
line 242: why you mix here comparison to 1980? Without any interconnection to the comparison before preindustrial era.
line 243-245: This is completely unclear. You should write more clearly and describe abbreviations at their first use - as in line 242.
line 255: before in the methods you wrote, that you did measure SWC, in here you write that you did model that. So what is true?
2.5 most of that is redundant
276-278: write better English - too complicated sentence
Eq 5. this is no equation. There must be physical variables.
286: write better English
lines 293-296: how exactly did you calculate this? Please ass equation. Could you add comparison to eddy covariance measured data?
line 305-309: so you did a pearson correlation test or PCA? Here you mix both together. very unclear what you have done.
line 316: Abbreviation should be defined at its 1st use, not again and again. Please check through the manuscript.
In here I stop reading, the above written issues are too major for this paper to be considered for publication.