General comments

The introduction is not tailored to the research question but rather puts in various issues about urban trees, which are unfortunately sometimes wrongly cited (see also specific comments). The ways how cooling and shading might be affected by heat and drought (stomata, leaf senescence, hydraulic failure) are not properly described, nor are direct (heat and drought) and indirect impacts (nutrition, air pollution impacts) or impacts that may mitigate damaging influences (CO2, irrigation) properly differentiated. Also, hypotheses are not clearly specified and H1 and H3 seem to be the same anyway. Transpiration will be increased (vpd) or decreased (stomatal control), or both (in which cases)? Drought decreases stomatal conductance (old news) or photosynthetic capacity (and then impacts stomata)?

The methodology is problematic, trees were of different species at each site, so neither species differences nor site differences can be evaluated. Furthermore, trees varied considerable in height and age, and measurements were taken at different heights and also the stratification within the crown was not homogeneously done. In addition, some sites were irrigated while others were not. It is not clear, how the ‘normalization’ by vpd is actually done and if this is in accordance with theoretical considerations regarding the vpd response. In addition, it is not clear, how much drought stress was actually present at the different sites since only percentages of water are given without indication of how much water is available in absolute or relative (absolute in relation to total water holding capacity) terms. Overall, since vpd is not the only influence, which is necessary to consider in order to compare the different site and species responses, a more complex approach seems to be necessary in order to differentiate between site and species impacts. Perhaps, this means using a model that describes gas exchange based on climatic as well as soil conditions.

Despite the very different boundary conditions and species and the few samples for each condition, the discussion tries to differentiated between heat and drought effects although both influences were simultaneously occurring and can hardly be distinguished. This leads to counter-intuitive results (such as a restricting (significant?) role of soil moisture for forest gas exchange in the wet period) and very disputable conclusions (e.g. “severe weather events did not alter the stomatal action”). The potential benefit of this study, which is the different response of species that a) do not close stomata, b) close stomata, c) are damaged by non-stomatal effects, could not be addressed due to the various degree of stress and different boundary conditions. Therefore, the conclusion is that species and site differences are causing the variation – which is probably true but could not really be shown given that neither species could be compared on different sites nor site influences could be evaluated using the same species. Accordingly, also the impacts that different species might have on their environment under heatwave conditions are highly speculative and are more based on literature than on measurements presented here.

Specific comments (and some technical corrections)

L1: I guess the role in offsetting CO2 emission is actually not important. If anything, consider pollution deposition effects. (see also L27)

L5: how many trees?

L10ff: Do you say that street trees reduced their transpiration during a heatwave (relative to other sites) but not during dry periods? Can you actually differentiate between the two kinds of stress?

L15ff: Here you say that there is no effect on stomata during heatwaves - also for street trees. How does this fit to the indication of street trees having reduced their transpiration?

L17-20: In one sentence you say that drought limited transpiration in forests and in the following it is stated that drought was so mild that it could not affect stomata control. What did I miss here??

L28/29: regulating energy balance and cooling the surrounding is actually the same.

L29: pollution deposition, not ‘infiltration’

L32/33: Sorry, but none of the indicated references are saying that urban trees are important to mitigate the global GHG. In the contrary!

L35: human disturbances? Do you mean damages due to traffic or pruning by urban managers?

L43-45: I cannot derive this conclusion from Bussotti et al.. Did I miss something?

L65: What’s the difference between research question 3 and 1/2? Do you want to address the impacts in combination?

L60ff: Since the research capacity is limited, I think it is logical to assume that not all urban trees but a selection based of the most representative species is targeted. Please elaborate the text to make it more specific.

L71-75: Shift to site description.

L80: sparse tree cover! Roadside single trees (not a plantation)!

L118: do you mean damages by pedestrians?

L134: ‘normalizing’ flux density by dividing by vpd seems strange to me since velocity will likely have an S-shape response of vpd. Is this a common strategy (please cite) or may it be that you first would need to derive the dependency between both and then normalize by applying the respective function? Or did you do that as it seems later on that you fitted curves to this relationship?

L175ff: I don’t understand this. What is the control period to define heatwaves? The long-term (30year) average for the respective season/month/specific days? The heatwave duration was more than one month? Can you provi de explicit temperatures?

L183: What do you mean with a ‘further’ separation of periods … based on soil moisture? What is this give an absolute value that you indicate here – an arbitrary value between field capacity and wilting point? What is the water content relative to field capacity at this point?

L185: Are the hypotheses (if better defined) not better evaluated with transpiration than with sapflux density? Or by Water use efficiency?

L202: I would rather say site climate instead of microclimate (which is normally used for microsites such as canopy layers).

L206: Soil moisture at what time of the year? Or do you mean maximum water content?

L211: Do you mean sapflow rate instead water use? (Water use would be expressed per m-2 ground area).

Figure 3: Strange that there is no increase in soil moisture at the 1^st of July despite considerable rainfall. Any explanation for this?

Table 2: Do you mean average soil moisture within the indicated period. Please use average (relative) available soil moisture instead.

L236: Still unclear how a normalization was done (see also comment to L134).

L282: vpd is due to impervious cover? Do you mean it is due to the increased temperatures that are caused by the impervious cover? Or are you assuming that vpd could be lower because of soil evaporation?

L289ff: Again, it is difficult to comprehend, what ‘water use’ means. If it is defined on the sapwood area, stem size indeed plays a major role (as indicated), but is this also the case if the water consumption/transpiration on a ground area basis would be calculated? I guess the latter is more important in the context of comparing site conditions.

L291ff: Similarly, water availability is only meaningful if it is the water storage capacity minus the water bound by the wilting point. The definition is however, not clear.

L298ff: needs references. For Betula isohydry, I would recommend to consider Zapater et a. 2013. Tilia, however, seems also to be fairly isohydric (Leuschner et al. 2019), so I am not sure about the logic of the reasoning here.

L301ff: What do you want to tell here? That Tilia may use more or less water than other trees? Or that site conditions might influence the transpiration rate of Tilia? But this is generally true for any species.

L323: replace ‘cover’ by ‘represent’ or similar (of course, one day does not cover the whole period)

L324ff: This is a conclusion that is contrary to the statement that Js/VPD declined during drought. Please provide an explanation for your conclusion. I can also not see where you see a similarity to Gillner et al. despite the very general fact that transpiration is mostly higher in summer. If this is about species comparison, the whole paragraph needs to be directed towards this.

L333ff: You are probably indicating that the drought was not sufficient to cause damages to photosynthesis (non-stomatal effects, see e.g. Gourlez de la Motte et al. 2020), since you already discussed that a stomatal effect should have taken place. In addition, the stomatal control may be differently strong, depending on the isohydry or anisohydry of the species. The literature references should support your respective message and not only indicate similar results under possibly similar conditions.

L351: delete ‘, suggesting that … at these sites’ (redundant)

L389-391: not shown in this paper

L391-393: this cannot be derived from this study

L393/4: wishful thinking, not a conclusion

Mentioned references

Gourlez de la Motte L, Beauclaire Q, Heinesch B, Cuntz M, Foltýnová L, Sigut L, Manca G, Ballarin I, Vincke C, Roland M, et al. 2020. Non-stomatal processes reduce gross primary productivity in temperate forest ecosystems during severe edaphic drought. Philosophical Transactions of The Royal Society B Biological Sciences 375(1810): 20190527.

Leuschner C, Wedde P, Lübbe T. 2019. The relation between pressure–volume curve traits and stomatal regulation of water potential in five temperate broadleaf tree species. Annals of Forest Science 76(2): 60.

Zapater M, Bréda N, Bonal D, Pardonnet S, Granier A. 2013. Differential response to soil drought among co-occurring broad-leaved tree species growing in a 15- to 25-year-old mixed stand. Annals of Forest Science 70(1): 31-39.

This paper addresses the issue of green infrastructure effect on ecosystem services like carbon sequestration, reports on the influence in water cycle processes like transpiration, and overall, provide insights on trees contributing to human well-being. The issue of urban trees improving such ecosystem services is fundamental to support the pertinence of nature-based solutions (NbS) Scientific evidence for these innovative approaches (NbS) are still scarce. The authors provide valuable evidence on this line of research.

Park and street sites have the same tree (species), but forest and orchard have two different tree species. This give a better condition to compare in a fair way the first two sites but could be questionable the comparison of the four sites together. In line 348 to 350, the authors state “The difference in saturation levels may be species-specific or caused by differences in the availability of soil moisture as the Orchard site had higher soil moisture than the Forest site during the heatwave and dry period". How can we separate these effects from the whole variables analyzed? Could the authors provide a section in discussion where it is addressed something like limitations and recommended improvements in this type of methodologies to cope with such result?

Also, it will contribute to clarify the context if the authors add a line within the descriptive table 1, the main  drought-resistant features of trees analyzed as their defense strategies (escaping, avoiding or tolerating the loss of water) if there is something noticeable.

There are differences in soil type between sites too, although all of them contains sand in some proportion, specifically speaking the authors are not comparing the exact soil conditions. The data provided on soil is reduced to the names of soil types. To provide more accurate information about the potential effects of soil type on the analysis, it would be desirable to also have a short description of the soil class (taxonomy) specially in terms of water retention properties.

In Figure 4, for the sake of clarity, it is important to explain here why there is no data the month 8 for street site at panel b. Figures should be interpreted as stand-alone piece in the paper, or self-explained.

In table 3, please explain here why there is no pre-heatwave in street, and why is no data for no heatwave for orchard site. Such information also affects figure 5.