|This paper is investigating the effect of ambient ozone on a mature Belgian pine forest, using a fairly simple semi-empirical/semi-mechanistic modelling approach that focusses on assessing the daily and annual effect of ozone on GPP. Studies of ozone effects on mature forest trees are rare, which makes this study to stand out and makes it worth publishing! The simplicity of the modelling approach limits the general explanatory power and significance of the results, which should be more critically discussed. Important aspects of the experiment description and methodology are either vague or missing and should be added; the analysis of results could be deeper in parts. Specifically, the following issues should be addressed before this paper could be accepted for publication:|
1. GPP is the key parameter of this paper and should be introduced in more detail. Why did the authors choose GPP as the effect parameter, how is it linked to tree/forest growth, timber production, C sequestration etc.? Why is the effect of ozone on GPP of (ecological/economic) interest? Is there evidence in the literature that ozone affects (tree) GPP?
2. Due to the importance of GPP in this study, the measurement of NEE in this forest stand and the derivation of GPP from measured NEE have to be described in much more detail, at least in the appendix (the sentence “…(NEE) measured with the eddy covariance technique with the instrumentation and following the standard data quality procedures as explained in (Carrara et al., 2003;Carrara et al., 2004;Gielen et al., 2013).” is not sufficient and too vague (e.g., which instrumentation?)). Also, for which time period was NEE measured/GPP derived? For the entire growing season of all 14 years? If so, why don’t the authors show in the results section scatterplots of GPP (as derived from measured NEE) against Fst or POD1?
3. For the derivation of LAImax across all years, measurements were also taken from the year 2003 (weather-wise an extreme year that unfortunately had to be excluded from this study – it would have been very interesting to see if it had created an outlier in figures 8 (E,F) and 9): How sure are the authors that this 2003 measurement was representative, given that European forests suffered immensely during that year from heat and drought, which might have had an effect on LAI?
4. It would have been interesting to do a year-to-year comparison of ozone effects on GPP to analyse whether the timing/onset/length of ozone episodes (e.g. late/mid/end of season) had any effect on GPP; the authors could for example add the year to data points in Fig. 9 and relate outlier years in these figures to ozone concentration anomalies in these years?
5. The description of the used ANN is very short, please add detail how the final model was derived. Is there any particular reason why the authors used ANNs for the analysis (could the authors maybe justify with reference to other biological studies that also used ANNs?)? One main strength of ANNs is their power to disentangle strong and weak input-output relationships, so one would have expected to at least see a ranking of parameters that define GPP. For ANN model runs with the full dataset (i.e. no exclusion of high ozone GPP days), the authors should have also included ozone as input to see whether it has any explanatory power on GPP.
6. Line 204f.: “Under the assumption that O3-induced 205 GPP reduction is most likely to occur during and shortly after days of high stomatal O3 fluxes, ..”, Line 239ff. “now not only excluding the days with the highest stomatal O3 fluxes from the dataset for model training but also the two following days.” And Line 255f. “we excluded the days with a peak of stomatal O3 flux plus the six following days”. The reasoning behind these timeframes is rather vague and these are big assumptions that need scientific discussion: Is there experimental/statistical evidence that high ozone episodes only have an effect on GPP for up to two days after the episode? Why exactly six days to look at carry-over effects? Surely timing will depend on the intensity and length of the ozone episode? How do the authors distinguish between “high” and low ozone episode? Where did they draw the line (please show how many high and low ozone episodes there were in each year)? The authors call the latter episodes “supposedly O3-damage free”, which is a very bold statement. This is one of the weakest point of the methodology that should be thoroughly addressed in the revision!
7. It is surprising that the authors chose the Jarvis-type stomatal conductance model rather than the Ball-Berry photosynthesis-based stomatal conductance model, which would have offered a more direct link between ozone uptake and photosynthesis/GPP. Please add a few sentences on this type of model to discussion section.
8. The authors should discuss long-term effects of ozone on forest/tree productivity (e.g. over the entire length of the 14 year study period), which presumably can’t be captured by the approach chosen?
9. Related to 8: Please more clearly summarise the limitations of your study/approach.
10. In Appendix A the authors introduce the stomatal model, including fphen, which is then set to 1 in the model runs, right? So maybe remove detail in appendix? Also, how was the start and end of growing season identified? Given the change in weather, one would have expected a variation of these two parameters from year-to-year.
11. Line 340f.: Please add reference for other Scots pine forests
12. Some of the references seem to be rather outdated, e.g. line 25 (IPCC, 2001). Please check manuscript for more recent ones.
13. Please check reference list: some references mentioned in main text are missing (e.g. CLRTAP, 2015; Mills et al., 2011), some authors misspelled (e.g. Büker rather than Buker)
14. Please check paragraphs starting on line 48 and 204 for spelling. Also, check entire manuscript for BG criteria for correct referencing, e.g. line 124 “explained in (Carrara et al., 2003;Carrara et al., 2004;Gielen et al., 2013)” should be “explained in Carrara et al. (2003, 2004) and Gielen et al. (2013)”. Make sure to leave space after “;” when listing references.
The reviewer would like to encourage the authors to address these issues and resubmit the paper, as it will definitely be of interest to the respective research community!