Thank you for your comments concerning our manuscript entitled “Dual-frequency radar observations of snowmelt processes on Antarctic perennial sea ice by CFOSCAT and ASCAT”. We will incorporate the citation into the paper as you suggested.

Thank you for your comments concerning our manuscript entitled “Dual-frequency radar observations of snowmelt processes on Antarctic perennial sea ice by CFOSCAT and ASCAT”. These insightful comments have been invaluable in helping us identify areas for improvement within our manuscript. We have carefully studied your suggestions and would like to provide the following response. Please kindly note that all the figures and the table referenced are included in the supplementary material for your review

1. Line 125. The authors used the SIR-enhanced ASCAT data with a spatial resolution of 4.45 km but resampled it onto a 12.5 km grid to match up with CFOSCAT data. Why not directly use the original 12.5 km ASCAT data (e.g. provided by IFREMER)?

Response: Thanks for your comments. Regarding the choice of using SIR-enhanced ASCAT data rather than other 12.5km ASCAT data such as those provide by IFREMER, this decision was primarily based on the following considerations:

On the one hand, this work is an extension of Arndt & Haas (2019). To maintain consistency and allow for better comparison with the study of Arndt & Haas (2019), we used the same ASCAT backscatter coefficient product, i.e., the SIR-enhanced ASCAT product. On the other hand, the spatial coverage of the SIR-enhanced ASCAT data is higher than that of the IFREMER ASCAT data (e.g., Fig. S1), indicating the former is expected to provide more valid observations.

We also retrieved the PMO and SMO based on IFREMER ASCAT data for 12 perennial ice study sites and compared them with those derived from SIR-enhanced ASCAT data. Figure S2a illustrates the SMO for the 12 study sites from 2007/2008 to 2021/2022 using two ASCAT data sets. It shows that the two are mostly consistent, with an absolute difference of 8 days between the two and 76% of the differences within 5 days. There are two main reasons for the difference between the two. First, there are some missing data in the IFREMER ASCAT which leads to retrieval errors of PMO or SMO (e.g., Fig. S2b). Second, in areas where snowmelt is not significant, the increase in the backscatter coefficient caused by snowmelt is inconsistent between the two data sets, leading to differences in the retrieved PMO and SMO between the two (e.g., Fig. S2c). By examining the results of all the study sites, we found that the missing IFREMER ASCAT data was the main reason for the difference between the two. In addition, missing IFREMER ASCAT data resulted in reduced retrieval rates, e.g., the SMO in Location 1 from 2014/2015-2015/2016 cannot be retrieved by IFREMER ASCAT data (Fig. S2a).

In general, even though IFREMER ASCAT data can be used directly without spatial resampling, we believe that using SIR-enhanced ASCAT data is still a better choice. This is based on two key considerations: firstly, the use of SIR-enhanced ASCAT data ensures consistency between this study and Arndt & Haas (2019); and secondly, the SIR-enhanced ASCAT data can provide more reliable observations of snowmelt onsets owing to their higher spatial coverage.

We will add some description of this issue to our manuscript.

2. Line 206-211. The differences between the local backscatter maximum and the preceding local backscatter minimum are the key points to identify the PMO or SMO. Are there any statistics on the interval days of minimum and following maximum for the identified PMO or SMO? Figure 2 shows that the interval may be 10-15 days for PMO and SMO. I suggest adding some statistics and discussion of the interval days. Perhaps this interval length could also represent the strength of the PMO and SMO events, apart from the amplitude of the backscatter increase.

Response: Thank you for your constructive feedback on our manuscript. We have made some discussion based on your suggestion about the time interval between the local minimum and the following maximum when PMO and SMO are detected.

Due to the drier and colder air in the south, the snowmelt intensity in the southern regions is weaker compared to the north. Therefore, we use latitude as an indicator of snowmelt intensity and give the variation of interval days with latitude as show in Fig. S3. It can be seen that the intervals mostly vary from 6 days at low latitudes to 18 days at high latitudes. Except for the high interval at Location 4 caused by the data gap of CFOSCAT data (Fig. S3b) and the low interval at Location 10 caused by the difficulty in detecting PMO and SMO (Fig. S3a-c), the interval generally shows a decreasing trend with decreasing latitude. The interval at Location 11 also sometimes appears shorter than that at Location 12 (Fig. S3d), which is caused by the rapid change in backscatter of several pixels around Location 11 (as shown in Fig. S4, where backscatter coefficients change rapidly during occurrences of SMO). The above discussion suggests that time intervals can reflect snowmelt intensity to certain extent, but rapid snowmelt processes can affect the use of the interval to determine the snowmelt intensity (e.g., Fig. S4).

In distinguishing the strength between PMO and SMO, the interval shows some potential, as indicated in Table S1, where the PMO interval is consistently shorter than the SMO interval. However, as illustrated in Fig. S4, SMO interval may sometimes be much shorter than PMO interval, while the backscatter rise amplitude remains a good indicator for distinguishing the strength between PMO and SMO.

We also found that these intervals do not differ much between the two scatterometers. For all regions, the average PMO interval between ASCAT and CFOSCAT is comparable, with both around 11 days, while the average SMO interval for both is about 12 days (Table S1). In Fig. 5 of the manuscript, we observe that differences in the amplitude of the backscatter rise between ASCAT and CFOSCAT are significant at each study sites, indicating the backscatter rise amplitude is more robust in reflecting the differences between the two scatterometers. Moreover, if latitude is used as an effective indicator of snowmelt intensity, the latitudinal dependence of backscatter rise is more significant than that of interval on latitude, indicating that backscatter rise is a more effective indicator of snowmelt intensity. In general, the backscatter rise amplitude is more suitable for determining PMO and SMO, and exploring the differences between the two scatterometers. However, using intervals to characterize the snowmelt intensity provides us with a new perspective to study snowmelt based on remote sensing. In the future, we may further study the snowmelt processes on Antarctic sea ice based on the "interval" parameter.

We will add some discussion about “interval” in the manuscript.

3. Figure 4. I noticed some years when PMO occurred later than SMO, e.g., site 3 in Nov/Dec 2011 and site 7 in Nov/Dec 2013 and 2016, which seems abnormal. Please explain it.

Response: Thanks for your comment. In this study, the PMO and SMO for each study site were obtained by averaging 9 pixels around the site, as we have demonstrated in Section 2.2. At some pixels around the site, PMO can be detected but SMO cannot be detected, resulting in a PMO value being larger than the SMO value after averaging the results from 9 pixels. Figure S5 presents an example. We will add an explanation regarding this issue in the manuscript.

4. Figure 13 and Line 522. It seems to be a declining trend of SMO for all regions after 2015. I suggest the authors add two more quantitative trend analyses for the period 1992-2015 and 2015-2022, respectively. A comparison of the two trends could add more highlights to this paper.

Response: Thanks a lot for your suggestion. According to your suggestion, we have fitted the trends of SMO for 1992-2015 and 2015-2021, respectively, as shown in Fig. S6.

From 1992/1993 to 2021/2022, SMO dates across all regions show a positive trend, i.e. occur increasingly late (Fig. S6). However, there are large regional differences particularly in the BS/AS region where the trend is negative, i.e. SMO occurs earlier. In the RS region, there is no significant change in SMO.

Investigating the trends from 1991 to 2015 and from 2015 to 2021 separately, for the Antarctic as a whole, the mean date of the SMO was slightly delayed before 2015 (0.63 days/year), and slightly advanced after 2015 (-0.89 days/year) when the sea ice extent around the Antarctic began to decrease strongly (Fig. S6). For the four sub-regions, the mean SMO date exhibited a consistent delay trend from 1992 to 2015, with the delay being most pronounced in the SWS region at 0.9 days per year. From 1992 to 2015, the sea ice extent in the BS/AS region exhibited a decreasing trend, whereas the WS and RS regions experienced an increasing trend (Parkinson, 2019). After 2015, the sea ice extent in the four regions tended to decrease, with the largest ice loss being observed in the WS region (Jena et al., 2022). As illustrated in Fig. S6, the SMO change trend pattern after 2015 showed large regional differences. Specifically, the NWS and BS/AS regions experienced an earlier SMO trend, while the SWS and RS regions exhibited a later SMO trend. In the Arctic, studies have shown that melt onset is strongly correlated with changes in sea ice extent (Petty et al., 2017). In the Antarctic, however, the mechanisms influencing changes in sea ice extent are more complex (Parkinson, 2019). In this study, we only found relatively consistent changes between SMO and SIE for the Antarctic as a whole. For each sub-region, we observed a relatively weak correlation between SMO and sea ice extent changes. The weak correlation between SMO and Antarctic sea ice extent has also been emphasized in the previous studies (e.g., Stammerjohn et al., 2008; Arndt & Haas, 2019). Instead, regional differences in the SMO change patterns are more prominent. This also reminds us that further understanding of the relationship between SMO and sea ice extent changes requires comprehensive consideration of ocean/atmospheric factors influencing regional sea ice changes, such as the sea surface temperature and atmospheric circulation patterns (Yu et al., 2024).

We will include this discussion in the manuscript.

References:

Arndt, S. and Haas, C.: Spatiotemporal variability and decadal trends of snowmelt processes on Antarctic sea ice observed by satellite scatterometers, The Cryosphere, 13(7), 1943-1958, https://doi.org/10.5194/tc-13-1943-2019, 2019.

Jena, B., Bajish, C. C., Turner, J., Ravichandran, M., Anilkumar, N., & Kshitija, S. (2022). Record low sea ice extent in the Weddell Sea, Antarctica in April/May 2019 driven by intense and explosive polar cyclones. Npj Climate and Atmospheric Science, 5(1), 19.

Parkinson, C. L. (2019). A 40-y record reveals gradual Antarctic sea ice increases followed by decreases at rates far exceeding the rates seen in the Arctic. Proceedings of the National Academy of Sciences, 116(29), 14414-14423.

Petty, A. A., Schröder, D., Stroeve, J. C., Markus, T., Miller, J., Kurtz, N. T., Feltham D. L., & Flocco, D. (2017). Skillful spring forecasts of September Arctic sea ice extent using passive microwave sea ice observations. Earth's Future, 5(2), 254-263.

Stammerjohn, S. E., Martinson, D. G., Smith, R. C., Yuan, X., & Rind, D.: Trends in Antarctic annual sea ice retreat and advance and their relation to ENSO and Southern Annular Mode variability, J. Geophys. Res., 113, C03S90, https://doi.org/10.1029/2007JC004269, 2008.

Yu, L., Zhong, S., Sui, C., & Sun, B. (2024). Sea surface temperature anomalies related to the Antarctic sea ice extent variability in the past four decades. Theoretical and Applied Climatology, 155(3), 2415-2426.

Minor comments

P: page, L: line

P4, L139. “2-m ERA5 air temperature” should be “ERA5 2-m air temperature”.

P14, L373. “backscatter at ASCAT” should be “backscatter for ASCAT”.

P20, L539. “differences between” should be “difference between”.

Response: Thanks for your suggestion. We will make the necessary corrections to address the writing errors.