This manuscript investigates the impacts of extreme precipitation events on seasonal historical and future SMB in the Eastern Canadian Arctic and Greenland using simulations from the RACMO and VR-CESM models. Comparing extreme precipitation days with non-extreme precipitation days, one of the main findings is that historically extreme precipitation days consistently leads to a higher SMB, both in the cold and warm seasons, while for the future this relationship only persists for the cold season. For the warm season, the extreme precipitation results in a less positive/more variable SMB and its contribution becomes more prominent. Further because of a shift towards more rainfall, extreme precipitation days increasingly coincide with mass loss, particularly in SW Greenland and Baffin Island.

The manuscript is well structured and pleasant to read. The content of the manuscript is very interesting and fits well in the scope of the journal. However, the manuscript needs some improvements here and there. I have added several comments/suggestions that may help the authors to improve their manuscript. 

• Title:
  • Although the title itself is good, it can use a bit more improvement to reflect the content of the manuscript better. Personally, I was thinking about: “Modelling the Impacts of Extreme Precipitation Events on Seasonal Historical and Future Surface Mass Balance in the Eastern Canadian Arctic and Greenland”

• Section 1: Introduction:
  • L75-80: The authors mention an example of how an extreme snowfall event in the Swiss Alps can impact albedo and surface melt for several days, but can they also give an example of heavy snowfall events and related impacts in Greenland and or the Eastern Canadian Arctic? In my opinion, these kinds of examples contribute better to the (geographical) settings described in this manuscript.

• Section 2: Data and Methodology:
  • L123-124: The sentence “The snow cover … (Lawrence et al., 2019)” is redundant as the authors already explain the snow model in CLM5 in L133-134. Therefore, the sentence here can be removed.
  • L124-L126: This sentence needs to be rephrased a bit as elevation downscaling is only applied over glacierized land units and not over the entire grid cell. Also, it is relevant to mention (in my opinion) that CLM uses a subdivision scheme that heterogeneously subdivides grid cells into several land units to account of the heterogeneity of the land surface. This kind of information is maybe less relevant for the GrIS itself but could be relevant for interpreting results in the ablation zones and/or partially glaciated grid cells.
  • L133: Please add 10 m water equivalent (w.e.) or 10 m w.e.
  • L155: What is the main reason for defining extreme precipitation in 2 different ways? Please explain in the manuscript.
  • L157-158: Could the authors elaborate more on the meaning of “total daily precipitation volume over all grid cells”? Is it the areal mean of daily precipitation expressed in mm/day (i.e. averaged over all grid cells)?
  • L160: I assume that the extreme threshold would be the 95^th percentile of total daily precipitation at subregional level?
  • L166: Is the window of +/- 15 days centered around each extreme precipitation day? If so, please indicate that in the manuscript. And why do the authors choose for a window of +/- 15 days. Does the calculation of anomalies in this way not increase the risk of including values (for SMB and its components) for non-extreme days and therefore cause a mixed/noisy signal?
  • L171-173: This sentence is not clear. So, if I understand well the future IQR increases are statistically significant if the real IQRdiff is greater than the IQRdiff in 975 of the 1000 tests (or maybe in other words the 97.5^th percentile of all IQRdiff values)? And why 975 tests are chosen as the threshold?

• Section 3: Extreme Precipitation:
  • L191-193: I guess also SE Greenland forms an exception in the winter months but then in an opposite way with RACMO/ERA5 extreme precipitation amounts being lower than VR-CESM extreme precipitation amounts.
  • Figure 2 shows the mean monthly accumulation per grid cell for the defined domains. Does the mean monthly accumulation based on extreme precipitation at subregional-level as defined in L157-158 look different?

• Section 4: SMB Response to Extreme Precipitation:
  • L218: “…we first consider…” instead of “…we consider…” (I assume the authors will consider mean seasonal SMB from VR-CESM simulations in a later stage).
  • Figure 4; L218-226: To what extent differs the simulated/projected SMB in VR-CESM from those simulated/projected in RACMO? Please explain in the manuscript. Further I suggest the authors to merge Figure 4 and Figure S1 into one Figure. To do this I recommend removing the FUT fields as the FUT-HIST fields already show the differences in the future relative to the historical period. Once the FUT fields are removed it is possible to make one figure with 2x4 panels where the rightmost panels show the results for VR-CESM, and the leftmost panels show the results for RACMO (or other way around). In my opinion showing the VR-CESM SMB fields in Figure 4 is also important for understanding the results shown in the figures that follow.
  • L239-241: Both RACMO and VR-CESM experience an increase in temperature. Then why VR-CESM shows a more pronounced increase in extreme precipitation days than RACMO? Is it because VR-CESM experiences a stronger increase in temperature which leads to stronger increases in atmospheric moisture content? Or is the difference explained by other phenomena related to the atmosphere or maybe even sea ice? For example, what happens with the zonal wind patterns (for example at 850 hPa) during the cold season? Could it be possible that future zonal wind patterns (i.e. related to the jet stream) display a stronger poleward displacement in VR-CESM than in RACMO and therefore explain the SMB increases in NO Greenland?
  • L262: Could the authors elaborate more on what they mean with: “the difference between the SMB on extreme and non-extreme days shifts in many subregions as well”? Are the authors trying to say that the trendlines for future and historical simulations have the same slope but another intercept (i.e. lower intercept)?
  • Figure 5 and 6: In my opinion it could have an added value to insert trendlines in the figure so readers can more easily derive the ratio/relation between extreme day SMB and non-extreme day SMB for both models and periods.
  • L262-264: I would suggest rephrasing the sentence to: “Only SE, CW, NW, and NE Greenland continue to show a more positive SMB on extreme precipitation days than the SMB on non-extreme precipitation days (in the same year)”. I think the latter part is a little redundant.
  • L274, L287: “Table 1 and 2…” instead of “Table 2 and Table 1…”
  • L275, Table 1 and 2: What is statistically significant according to the authors? Is that statistically significant at the 95% confidence interval? And how is the statistical significance of changes/differences defined/calculated? Via a student’s t-test?
  • L288: Please remove the text “Figure 6”.
  • L298-299: Are the days with positive SMB (SMB+) referring to all days with positive SMB (SMB+ all) or to extreme precipitation days with positive SMB (SMB+ ex)? To avoid confusion, I recommend the authors to use the abbreviations as defined in Section 2.2.
  • Figures 7 and 8: Similar to Figure 4 I strongly suggest removing the FUT fields and instead adding the VR-CESM fields for HIST and FUT-HIST as these also include the key findings that are described in the manuscript.
  • L311-313: Could the reduction in SE Greenland be somehow related to the increased SMB on extreme precipitation days in NO Greenland? As mentioned in one of my earlier comments I can imagine that a poleward displacement of zonal wind patterns could be responsible for these changes, that is a decrease in extra-tropical cyclones/moisture inflow in SE Greenland, but an increase in moisture inflow in NO Greenland.
  • L329: Please remove the text “Figure 8”.
  • L335-336: Please remove the text “Figure 9”. Also, could the authors add (SMB+ ex) between brackets after “extreme precipitation days” for clarification? I presume it is SMB+ ex and not SMB+ all.
  • L336: Although I can understand the relation between positive SMB and positive temperature anomalies in the Greenland area, readers could question about why positive temperature anomalies coincide with a positive SMB as they could expect the SMB to be negative (positive temperature anomalies  more melt  negative SMB). Therefore, it could be useful to explain in more detail how the authors interpret the coincidence of positive SMB and positive temperature anomalies (or refer to Section 5.1 where the authors explain these findings in more detail).
  • Figures 9-11 and S5-S9: Please change SMB+ (or SMB-) to SMB+ ex (SMB- ex). Also, I presume the window of +/-15 days used for the calculation of the anomalies is centered around the extreme precipitation day?
  • L343 and L356: “for RACMO and VR-CESM…” instead of “from RACMO and VR-CESM…”.
  • L355: SMB- ex instead of SMB-?
  • L356: I presume Fig. 8 is referring to Fig. 8f?
  • L363: “…large runoff increases…” instead of “…runoff large increases…”.
  • Figures 9-11 and S5-S9: In my opinion it could also have an added value to add the rainfall and snowfall fields associated with positive and negative SMB extreme precipitation days. That also supports the findings described in L373-376.

• Section 5: Discussion & Limitations:
  • L430-432: I don’t entirely get the point of the authors as the authors already use +/- 15-day anomalies for SMB and its components for an extreme precipitation day. In that way I can imagine that the effects of extreme precipitation beyond the day itself are already included?

• Data availability:
  • I miss a data availability statement in this manuscript. Could the authors include one?

• Supplementary Data:
  • Could the authors prevent overlap between text and panels as shown in Figures S5-S6 and S8-S9?

Summary

In this study, the authors analyze the historical and future contributions of extreme precipitation events to the surface mass balance (SMB) of ice masses in Greenland and the eastern Canadian Arctic Archipelago. They use two distinct sets of SSP5-8.5 simulations (CESM downscaled with RACMO and its multi-layer snow module; CESM-VR with the CLM5 land model). They find 

that in the historical period, extreme precipitation days are associated with increases in SMB during both the cold and warm season, but in future warming scenarios, mass loss becomes more common on extreme precipitation days due to an increase in the proportion of liquid precipitation, particularly in some favored areas such as southwest Greenland and Baffin Island.

Overall, the paper is well written with clear explanations and figures. I have several comments that are primarily requests for the authors to better explain and contextualize their findings, especially with regard to the physical processes that cause melt on extreme precipitation days. Provided these comments are addressed, overall I think this study will provide an important and novel contribution to projections of the future evolution of the Arctic cryosphere.

Main comments

- I think the discussion about the physical mechanisms that lead to reduced SMB on extreme negative SMB days ("SMB -ex") days needs to be sharpened and clarified. If I'm interpreting the results correctly, the study calculates changes in SMB on these days without considering whether melt is directly caused by rainfall or whether it's primarily the result of atmospheric warming that tends to occur on the same days as extreme precipitation. A full surface energy balance analysis that would be required to definitely answer this question is likely beyond the scope of this study. However, to avoid confusion I feel that this uncertainty about the physical drivers of melt should be addressed in more detail prior to the brief statement in L426–429. I note that prior studies (e.g. Doyle et al. 2015 ; Fausto et al. 2016a,b; Box et al. 2023) have found that the rain heat flux (i.e. the direct effect of rain on ice ablation) is relatively small (< 20%) even during extreme rainfall events, at least in the historical period.

- It would be helpful to have more explanation about the somewhat counterintuitive changes in temperature anomalies during extreme SMB days in the future. Do the authors have an explanation for why the temperature anomalies during positive SMB JJAS extreme precipitation days are greater in the historical than the future simulation (Fig. 9–10, L357–360)? Is this because the temperature is already nearer to the freezing point in the future scenario and there is less overall variance in warm season temperature? And the finding of negative temperature anomalies in some low-lying and coastal areas on the future positive SMB JJAS extreme precipitation days (Fig. 10, L345–346) seems counterintuitive as well – do the authors have an explanation for this? To me it looks like the temperature anomalies are negative in the ablation zone and positive everywhere else on these days.

- In addition to the increase in atmospheric moisture with warming explained by the Clausius-Clapeyron relationship, is it possible that the increased contribution of extreme precipitation days to positive SMB in northern Greenland and northern Ellesmere Island has some contribution from Arctic sea ice decline? See L237–241, L305–307, L470–474.

- For the Greenland Ice Sheet, the study areas merge elevation zones of the ice sheet that have distinct characteristics in the current climate, from the low-elevation ablation zone up through the percolation zone and high-elevation accumulation zone. I think it would be a really helpful addition to the paper to provide a few analyses of how these results vary with elevation in the past and future simulations. For example, the warm season panels of Fig. 2 (described in L225–226) imply a huge expansion of the annual ablation zone into higher elevations, and it would be interesting to know the elevation range of the historical and future ablation zone. In Fig. 8 (L325–326), it would be interesting to know the elevation ranges of the areas over which extreme precipitation days make substantial negative contributions to JJAS SMB in the historical and future simulations.

Other comments

- L1: I suggest changing the title to incorporate the fact that the impacts of both historical and future extreme precipitation events are studied in this paper, and/or to emphasize that the key takeaway is that the impacts of extreme precipitation events are expected to qualitatively change in the future. Maybe "Modeling the impacts of extreme historical and future precipitation events...", or "Changes in future impacts of extreme precipitation events on surface mass balance in the eastern Canadian Arctic and Greenland".

- L28–39: I suggest revising the topic sentence of this paragraph to state that Arctic land ice is inclusive of both the Greenland Ice Sheet and the ice caps and glaciers of the eastern Canadian Arctic. As written, the eastern Canadian Arctic Archipelago is introduced rather abruptly near the end of the paragraph.

- L46: Be specific that increased water vapour holding capacity is due to climate warming

- L53–62: Nice explanation of potential changes in firn structure and response to precipitation due to climate warming

- L150–152: I like the idea to use four-month seasons (JJAS warm season, DJFM cold season)

- L196–198: The result that VR-CESM exhibits little change in extreme precipitation SE Greenland in any month appears to be consistent with the previous study by this lead author (Loeb et al., 2024), which found that VR-CESM projects decreases in extreme precipitation in SE Greenland. However this study shows that the RACMO-downscaled CESM simulation projects an extreme in warm season precipitation. Does this suggest that the results of the prior study were specific to the VR-CESM model?

- Figs. 9–11 and associated text: I'm a little confused about the relationship between modeled melt, runoff, and refreezing in these results. For example, does Fig. 10 show that on future positive SMB JJAS extreme precipitation days, there is anomalously large amounts of melt along the margins of the Greenland Ice Sheet, but less runoff and refreezing in most of these same areas? Or does this show that the amount of meltwater produced is below normal on those days? And if the latter, how is there anomalously large runoff in the lower elevations of SE Greenland despite there being less melt than normal? It would be helpful to give more detail about the physical processes represented by each term and the relationship between them, including whether a positive/negative value of each variable represents a positive or negative contribution to SMB.

Technical corrections

- L22: Add comma after "future"

- L67: The word "cause" is grammatically incorrect here. Should this be "causing" or "and cause"?

- L288: Typo with "Figure 6" inserted at the beginning of a sentence with no space

    - Please check this elsewhere, e.g. L329, 335–336

References

- Box, J. E., Nielsen, K. P., Yang, X., Niwano, M., Wehrlé, A., Van As, D., Fettweis, X., Køltzow, M. A. Ø., Palmason, B., Fausto, R. S., Van Den Broeke, M. R., Huai, B., Ahlstrøm, A. P., Langley, K., Dachauer, A., & Noël, B. (2023). Greenland ice sheet rainfall climatology, extremes and atmospheric river rapids. _Meteorological Applications_, _30_(4), e2134. https://doi.org/10.1002/met.2134

- Doyle, S. H., Hubbard, A., van de Wal, R. S. W., Box, J. E., van As, D., Scharrer, K., Meierbachtol, T. W., Smeets, P. C. J. P., Harper, J. T., Johansson, E., Mottram, R. H., Mikkelsen, A. B., Wilhelms, F., Patton, H., Christoffersen, P., & Hubbard, B. (2015). Amplified melt and flow of the Greenland ice sheet driven by late-summer cyclonic rainfall. _Nature Geoscience_, _8_(8), 647–653. https://doi.org/10.1038/ngeo2482

- Fausto, R. S., As, D., Box, J. E., Colgan, W., Langen, P. L., & Mottram, R. H. (2016). The implication of nonradiative energy fluxes dominating Greenland ice sheet exceptional ablation area surface melt in 2012. _Geophysical Research Letters_, _43_(6), 2649–2658. https://doi.org/10.1002/2016GL067720

- Fausto, R. S., van As, D., Box, J. E., Colgan, W., & Langen, P. L. (2016). Quantifying the Surface Energy Fluxes in South Greenland during the 2012 High Melt Episodes Using In-situ Observations. _Frontiers in Earth Science_, _4_. https://doi.org/10.3389/feart.2016.00082