This is a well written paper on projected hydrologic change in Alpine catchments through a conceptual rainfall-runoff model. I appreciated the fact that many different runoff signatures are discussed (actually the title could state that not only extremes are considered) and the results are compared to the existing literature. Also, the non trivial differences obtained comparing RCP4.5 and RCP8.5 is interesting and point out how mountain areas are peculiar. CMIP5 climate projections are used here, which will be soon become obsolete given the CMIP6 simulations already available (the Authors should also discuss this limitation). I also would have expected a more critical questioning on how well this modelling approach can capture runoff changes (e.g. based on the discussions in Duethmann et al., 2020, HESS). There are two subsections dedicated to that which could maybe be expanded. Regarding the attribution of runoff changes to the causes (precipitation, evapotranspiration and snowmelt), a more quantitative analysis would have made the paper even more interesting. These are just some minor suggestions/comments that I hope can help to improve the discussions in this paper that, overall, I liked reading. Some detailed comments follow.

Line 85: what could be the bias introduced by scaling the runoff data, rather than the precipitation data, for consistency with Budyko? Since the Authors are interested in percental changes, wouldn't this choice exacerbate the changes in magnitude for runoff in those catchments?

Line 104: what is the advantage of transferring the precipitation and temperature data from the EUROCODEX pixels to the ground stations and not working with the pixels themselves?

Line 115: is Table 1 the right table?

Line 170: has the glacier extent in the Pitztal a linear decrease till 2100?

Page 20: the Gailtal catchment seems special, in terms of annual maxima. Apart from the double peak season, what could explain the largest change in magnitude for RCP8.5?

Line 346: one interesting reference on observed trends in evapotranspiration in Austria and possible causes is Duethmann and Blöschl (2018, HESS).

Section 4: the section contains a lot of attribution statements. Are these statements confirmed by simulations, e.g., increase in precipitation keeping temperature unchanged and vice-versa? Or is it just an expert interpretation of the simulations with all variables changing?

Section 4.7: this section could be linked to the hypotheses for potential causes of the divergence between observed and simulated discharge changes in Duethmann et al. (2020, HESS).

References:

Duethmann, D. and Blöschl, G.: Why has catchment evaporation increased in the past 40 years? A data-based study in Austria, Hydrol. Earth Syst. Sci., 22, 5143–5158, https://doi.org/10.5194/hess-22-5143-2018, 2018.

Duethmann, D., Blöschl, G., and Parajka, J.: Why does a conceptual hydrological model fail to correctly predict discharge changes in response to climate change?, Hydrol. Earth Syst. Sci., 24, 3493–3511, https://doi.org/10.5194/hess-24-3493-2020, 2020.

General comments

The paper investigates the influence of climate change on timing and magnitude of future runoff extremes (maxima and minima) in six Alpine catchments in Austria using a process-based semi-distributed hydrological model and projections from 14 different regional climate and global climate model combinations for two future emission scenarios, namely RCP 4.5 and RCP 8.5. The selected catchments differ in elevation ranges as well as in hydrological regimes. The influence of climate change was quantified by comparison of the projected modelled runoff for the period 2071-2100 to the runoff of the reference period 1981-2010. The main aim of the study was testing the hypothesis that future changes in magnitude and timing of annual runoff extremes differ between high- and low-elevation Alpine catchments. The paper is in the scope of the journal. It is well written and structured. The data seem to be of appropriate quality. References are up to date and appropriate. Below there are only some minor suggestions that could further improve the manuscript.

Specific comments 

1. Data: The authors should explain the reason of using Thornthwhaite method (one of the simplest methods) for daily evaporation estimation since the choice of the method affects the results.
2. Data: Authors could give at least a basic description of both applied scenarios RCP4.5 and RCP8.5 (e.g. moderately optimistic scenario RCP4.5, worst-case climate change scenarios RCP8.5).
3. Methods: Please, elaborate the decision for the selection of the period 2071-2100 and not the period 2041-2070 for the investigation of the influence of climate change on future runoff extremes. The uncertainty of the results for such a distant future (2071-2100) is much greater and therefore confidence in the results is much lower.
4. Results, Figure 5 caption: It could be useful for readers to add colours to the explanation in the brackets: “Red and grey lines represent the mean flow regime….. “
5. Results: Section 3.3.2: I would suggest to change the order of the words in the title (timing and magnitude) to follow the structure of the section. The same goes to some other sections (i.e. 3.3.3, 4.3, 4.4)
6. Discussion, lines 339-340: The statement that the increase in projected future precipitation compared to the past is in contrast to other climate projections for Austria is not elaborated enough. What are the possible reasons? Did other authors (e.g. Stanzel and Nachtnebel, 2010) use the same scenarios for the same catchments/regions?
7. Discussion: In my opinion the section 4.5 is redundant. The societal impacts were not evaluated in the research; therefore, the discussion about the topic is not relevant.
8. Discussion: Section 4.6 (Climate model uncertainty) should be a part of the next section Uncertainty & limitations, where all other uncertainties are discussed.
9. Discussion, line 495: “….in the largest increases in magnitudes”. Of what? Annual minimum and maximum flows?
10. Discussion, line 500: It would be useful for readers to define the model 10 more precisely.

Technical corrections 

1. Page 12, Figure 5 caption: One bracket is missing after shaded area.
2. Page 20, line 310: “is” is repeated twice. One should be deleted.

General comments

This study investigates the projected evolution of hydrometeorological aspects in small mountainous catchments located in Austria, using a classic modelling chain: emission scenario / GCM / RCM / bias correction / hydrological model. This paper is very pleasant to read and I congratulate the authors for this fine work. I appreciated the detailed and concise description of the data and methods. The results are clearly presented and described. I hope that the following comments will help to improve the discussion of the results.

Areal meteorological inputs

There are several limitations concerning meteorological forcings. This is briefly discussed at l. 502-208 but it could be discussed earlier in the text. First, the typical problem with hydrological applications in mountainous areas is that weather stations are mostly located in plains, typically below 1000 m, while most of the area covered by the catchments is above. In addition to the fact that point measurements in space can misrepresent areal values, the problem is that there is generally a strong relationship between precipitation (and temperature of course) and altitude (see section 3.2 in Ménégoz et al., 2020), these altitudinal gradients being also dependent on the meteorological situations (Gottardi et al., 2012). Reanalysis datasets provided on a regular grid usually take these gradients into account, and the same kind of gradients could be applied to your interpolated data. Without this kind of corrections, I do not see how a correct water balance can be obtained. Could the authors comment on that point?

Bias-correction

It is very briefly mentioned at l. 104 that the climate simulations are bias-corrected using scaled distribution mapping. I would appreciate more details about the method proposed by Switanek et al. (2017) and applied in this study. For example, what is the distribution applied to the positive observed precipitation values? Is it a gamma distribution? It is not clear to me what we can expect concerning the correction of extreme values either. Looking at Figure 6, I was puzzled by the mismatch between observed and monthly runoff when climate simulations are used as inputs. It is acknowledged at l. 216 that there could be an “underestimation of temperature in these catchments in the climate simulations”. I understand that the bias-correction is not performing very well then, is that correct? If it is the case, I think it should be discussed in more depth.

Climate model uncertainty

Section 4.6, dedicated to climate model uncertainty, could be improved. First, as indicated in Table 2 of the manuscript, different GCM / RCM combinations are used in EURO-CORDEX. However, at l. 493-495, it seems that these pairs of climate models are considered as different models (e.g. “model 10”). It must be understood that the different GCMs and RCMs have their own structure, parametrization and, as a consequence, effects on the simulated variables. It is well described in papers dedicated to the partitioning of the different uncertainties (Déqué et al., 2012, Christensen and Kjellström, 2020). The study by Evin et al., 2021 clearly shows the individual effects of each GCM and RCM on the mean seasonal changes of precipitation and temperature in EURO-CORDEX ensembles (my apologies for citing my own work).

Other uncertainties

In section 4.7, other types of uncertainties could be discussed. The hydrological model can have a huge impact and the bias-correction / downscaling methods can also have an important influence (Lafaysse et al., 2014).

Climate projections

In the discussion, I think it could be interesting to indicate that CMIP6 simulations are now available but cannot be used for this kind of applications considering that GCM outputs are particularly misrepresented in mountainous areas (I must contradict reviewer #1 here). CMIP6 simulations will probably be downscaled dynamically in the next few years and RCMs represent a real added-value in these areas (Rummukainen 2016). In addition, a few RCMs are now able to represent convective processes and are expected to improve the representation of the precipitation in future climate projections (e.g. CNRM-AROME, Fumière et al., 2020), in particular the “localized convective high-intensity summer rainstorms” indicated at l. 505.

Minor comments

- Abstract: l. 5: I would add “two emission scenarios:” before RCP 4.5 and RCP 8.5 for the reader who does not necessarily know these scenarios.

- Abstract: l. 15: “Minimum annual runoff…” I guess this result is still obtained with RCP 8.5, is that correct?

- Figure 3: I suggest adding a reference to Table 2 in order to remind the meaning of the different objective functions.

- l. 235: missing space after “year.”

References

Christensen, Ole B., and Erik Kjellström. 2020. “Partitioning Uncertainty Components of Mean Climate and Climate Change in a Large Ensemble of European Regional Climate Model Projections.” Climate Dynamics, April. https://doi.org/10.1007/s00382-020-05229-y.

Déqué, M., S. Somot, E. Sanchez-Gomez, C. M. Goodess, D. Jacob, G. Lenderink, and O. B. Christensen. 2012. “The Spread amongst ENSEMBLES Regional Scenarios: Regional Climate Models, Driving General Circulation Models and Interannual Variability.” Climate Dynamics 38 (5): 951–64. https://doi.org/10.1007/s00382-011-1053-x.

Evin, Guillaume, Samuel Somot, and Benoit Hingray. 2021. “Balanced Estimate and Uncertainty Assessment of European Climate Change Using the Large EURO-CORDEX Regional Climate Model Ensemble.” Earth System Dynamics Discussions, March, 1–40. https://doi.org/10.5194/esd-2021-8.

Fumière, Quentin, Michel Déqué, Olivier Nuissier, Samuel Somot, Antoinette Alias, Cécile Caillaud, Olivier Laurantin, and Yann Seity. 2020. “Extreme Rainfall in Mediterranean France during the Fall: Added Value of the CNRM-AROME Convection-Permitting Regional Climate Model.” Climate Dynamics 55 (1): 77–91. https://doi.org/10.1007/s00382-019-04898-8.

Gottardi, Frédéric, Charles Obled, Joël Gailhard, and Emmanuel Paquet. 2012. “Statistical Reanalysis of Precipitation Fields Based on Ground Network Data and Weather Patterns: Application over French Mountains.” Journal of Hydrology 432–433 (April): 154–67. https://doi.org/10.1016/j.jhydrol.2012.02.014.

Lafaysse, M., B. Hingray, A. Mezghani, J. Gailhard, and L. Terray. 2014. “Internal Variability and Model Uncertainty Components in Future Hydrometeorological Projections: The Alpine Durance Basin.” Water Resources Research 50 (4): 3317–41. https://doi.org/10.1002/2013WR014897.

Ménégoz, Martin, Evgenia Valla, Nicolas C. Jourdain, Juliette Blanchet, Julien Beaumet, Bruno Wilhelm, Hubert Gallée, Xavier Fettweis, Samuel Morin, and Sandrine Anquetin. 2020. “Contrasting Seasonal Changes in Total and Intense Precipitation in the European Alps from 1903 to 2010.” Hydrology and Earth System Sciences 24 (11): 5355–77. https://doi.org/10.5194/hess-24-5355-2020.

Rummukainen, M. (2016). Added value in regional climate modeling. WIREs Climate Change, 7(1):145–159.

Switanek, Matthew B., Peter A. Troch, Christopher L. Castro, Armin Leuprecht, Hsin-I. Chang, Rajarshi Mukherjee, and Eleonora M. C. Demaria. 2017. “Scaled Distribution Mapping: A Bias Correction Method That Preserves Raw Climate Model Projected Changes.” Hydrology and Earth System Sciences 21 (6): 2649–66. https://doi.org/10.5194/hess-21-2649-2017