Change in potential snowfall phenology: past, present, and future in Chinese Tianshan mountainous region, Central Asia

Li, Xuemei; Liu, Xinyu; Zhao, Kaixin; Li, Lanhai

doi:https://doi.org/10.5194/tc-2022-244

Preprints

https://doi.org/10.5194/tc-2022-244

Preprints

08 Dec 2022

| 08 Dec 2022

Status: a revised version of this preprint was accepted for the journal TC and is expected to appear here in due course.

Change in potential snowfall phenology: past, present, and future in Chinese Tianshan mountainous region, Central Asia

Xuemei Li, Xinyu Liu, Kaixin Zhao, and Lanhai Li

Abstract. Rapid climate warming speeds up the solid-liquid water cycle and reduces the solid water storage in cold regions of the Earth. Snowfall is the most crucial input for the cryosphere. However, the potential snowfall phenology (PSP) variability has not been systematically and comprehensively studied. For this reason, we initially proposed three indicators, i.e., the start of potential snowfall season (SPSS), the end of potential snowfall season (EPSS), and the length of potential snowfall season (LPSS), to describe the characteristics of the PSP, then we explored the spatial-temporal variation of those three PSP indicators past, present, and future across the Chinese Tianshan mountainous region (CTMR) based on the observed daily air temperature from 26 meteorological stations during 1961–2017/2020 combined with 14 models data from CMIP6 (the Phase 6 of the Coupled Model Intercomparison Project) under four different scenarios (SSP126, SSP245, SSP370, and SSP585) during 2021–2100. It proved that the SPSS, EPSS, and LPSS could reproduce features of the PSP well across the study area. In the past and present, the potential snowfall season started on October 29th, ended on March 20th, and lasted for about four months and 23 days across the CTMR on average. The rate of advancing EPSS (−1.6 days/10a) was faster than that of postponing SPSS (1.1 days/10a) during 1961–2017/2020. It also found significant delaying by 2–13 days in the starting time, and advancing for 1–13 days in the ending time, respectively, which reduced 1–27 days for the LPSS. The potential snowfall season started later, ended later, and lasted longer in the north and center than in the south. Similar to the past and present, the SPSS, EPSS, and LPSS will vary under four emission scenarios during 2021–2100. Potential snowfall season will start later, end earlier, and last fewer days under the higher emission scenario. Under the highest emission scenario, SSP585, the starting time will be postponed by 41 days in 2100, while the ending time will be up to 23 days in advance in 2100, implying that it will cut down the length by 63 days (about two months), and the length of the potential snowfall season will only last two and a half months in 2100 under the SSP585 scenario. Spatially, the length of the potential snowfall season in the west and southwest of the CTMR will be compressed by more days because of the more delayed starting time and advanced ending time under all four scenarios. The results indicate that annual total snowfall will decrease, including amount and frequency, then reduce snow cover or mass, which finally feedback to the atmosphere in the form of more rapid warming for the lower reflectivity to solar radiation. Our research provides a new direction to capture the potential snowfall phenology in the alpine region and can be easily expanded to other snow-dominated areas worldwide.

Received: 03 Dec 2022 – Discussion started: 08 Dec 2022

Xuemei Li et al.

Status: closed

RC1:
'Comment on tc-2022-244', Anonymous Referee #1, 31 Dec 2022

Global warming speeds up the solid-liquid water cycle and change snowfall phenology. It was found that postponed snowfall occurrence and advanced snowfall ending took place in the Eurasian continent. Potential snowfall phenology can identify the possible onset, end, and duration of snowfall. To describe the characteristics of the potential snowfall phenology, this manuscript proposed three indicators, the start of potential snowfall season (SPSS), the end of potential snowfall season (EPSS), and the length of potential snowfall season (LPSS). Spatial-temporal variations of those three PSP indicators past, present, and future across the Chinese Tianshan mountainous region (CTMR) were explored. The research is sound and thorough. It provides a new direction to understand the potential snowfall phenology in the alpine region. Therefore, I recommend minor revision of this manuscript. Some specific comments are as follows:

Lines 35-37: please check the number of days cut down for the length, 63 or 64?

Lines 41-43: It suggested that “The results indicate that with constant snowfall intensity, annual total snowfall will decrease, including amount and frequency, .......”

Lines 50-105: The status, shortcomings and need of potential snowfall phenology studies is poorly described in the introduction.

Lines 95-100: The research from Jennings et al.(2018) only illustrated that RST was spatially heterogeneous and does not show that different methods of precipitation pattern separation yield different RST. Please check it.

Lines 128: The sentense “The frequency of rainfall increases while that of snowfall decreases. Besides, precipitation shifting from solid to liquid is obvious” is not clear enough, please check it.

In Figure 1: It is recommended that a general overview map could be added to Figure 1 to help the reader quickly identify the location of the study area.

In Table 1: It is suggested that a column could be added to Table 1 to indicate the duration of the data.

Line 230-232: It is recommended to use “advance / delay” instead of using “smaller or larger” to describe the change of LPSS and EPSS as much as possible.

Line 241-242: please check the slope of LPSS.

Line 371: More should be added here on the comparison of potential snowfall phenology with observed snowfall phenology.

line 377: “4.2 Spatial and temporal heterogeneity” should be changed to “4.2 Temporal heterogeneity”ã

line 408: Is there also an uncertainty error in the definition of PSP?

Citation: https://doi.org/10.5194/tc-2022-244-RC1
- AC1: 'Reply on RC1', Xuemei Li, 17 Mar 2023
  
  Thanks for your recognition of our work. Your specific comments could help us to improve our manuscript greatly. We will revise our manuscript seriously according to your specific comments.
  Some specific comments and replies are as follows:
  1 Lines 35-37: please check the number of days cut down for the length, 63 or 64?
  Description on the modification: Thank you very much for your advice. We have checked the number as suggested.
  2 Lines 41-43: It suggested that “The results indicate that with constant snowfall intensity, annual total snowfall will decrease, including amount and frequency, .......”
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  3 Lines 50-105: The status, shortcomings and need of potential snowfall phenology studies is poorly described in the introduction.
  Description on the modification: we will show more details about status, shortcomings, need, or significance of potential snowfall phenology in the revised manuscript.
  4 Lines 95-100: The research from Jennings et al. (2018) only illustrated that RST was spatially heterogeneous and does not show that different methods of precipitation pattern separation yield different RST. Please check it.
  Description on the modification: we will check it and revise the related sentences.
  5 Lines 128: The sentence “The frequency of rainfall increases while that of snowfall decreases. Besides, precipitation shifting from solid to liquid is obvious” is not clear enough, please check it.
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  6 In Figure 1: It is recommended that a general overview map could be added to Figure 1 to help the reader quickly identify the location of the study area.
  Description on the modification: Thank you very much for your good suggestion. We will revise Figure 1 as suggested.
  7 In Table 1: It is suggested that a column could be added to Table 1 to indicate the duration of the data.
  Description on the modification: Thank you very much for your advice. We have added the duration of the data in Table 1 as suggested.
  8 Line 230-232: It is recommended to use “advance / delay” instead of using “smaller or larger” to describe the change of LPSS and EPSS as much as possible.
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  9 Line 241-242: please check the slope of LPSS.
  Description on the modification: Thank you very much for your advice. We checked the slope of LPSS. It should be -2.7 days/10a. We changed the text and replace the corresponding figure.
  10 Line 371: More should be added here on the comparison of potential snowfall phenology with observed snowfall phenology.
  Description on the modification: Thank you very much for your good suggestion. We will collect more observed data to validate the performance of potential snowfall phenology.
  11 line 377: “4.2 Spatial and temporal heterogeneity” should be changed to “4.2 Temporal heterogeneity”.‚
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  
  Citation: https://doi.org/10.5194/tc-2022-244-AC1
RC2:
'Comment on tc-2022-244', Anonymous Referee #2, 09 Feb 2023

General comments:
The manuscript by Li et al. mainly investigated the changes in potential snowfall phenology for the past, present, and future periods in Tianshan, China. They defined three potential snowfall season metrics based on temperature data. Although the definition of potential snowfall phenology is interesting, the hydrological meaning of these metrics is questionable. The methodology of the study was not clearly described. The results lack a strict validation. The credibility of the study needs to be further improved.
Specific comments:
1. Introduction. The authors failed to well justify the motivation of the study. For example, what is the significance of predicting "potential snowfall phenology"? The start of potential snowfall season (SPSS) does not mean there is a snowfall. Even there may be no any snow during an entire "potential snowfall season". Thus, it may have no any effect on the water and energy balance of a region. In my opinion, the named metrics of "potential snow phenology" here only reflect the fluctuations of temperature, and they have limited hydrological significance.
2. The methodology is quite unclear. For example, the authors should provide more details of the calculation process of RST, as it is critical for this study. What data were you used to calculate RST? Did you validate the accuracy of the RST results?
If RST is calculated based on a long-term probability statistic of snowfall/rainfall, why is it reasonable to calculate RST at an annual scale?
Besides, precipitation phase partitioning is challenging in technique, as temperature humidity, and pressure jointly determine whether precipitation falls as snow/rain (Jennings et al., 2018). If the authors cannot prove the robustness and high accuracy of the RST calculation method for this region, I do not think the results of potential snow phenology are credible.
Jennings, K.S., Winchell, T.S., Livneh, B. et al. Spatial variation of the rain–snow temperature threshold across the Northern Hemisphere. Nat Commun 9, 1148 (2018). https://doi.org/10.1038/s41467-018-03629-7
3. L176-181. The authors indicated that they interpolated the model data to stations and then applied a bias correction method to improve the results. However, the interpolated results shown in the maps (Fig. 3, 5, & 7-10) still show large spatial biases. For example, many metrics show obvious circular changes around some stations. These maps do not well reflect the real spatial variations of these variables. The big errors of the results further reduce the value of the study.
4. L430-432. The estimated changes in potential snowfall season metrics fall in big variation ranges (e.g., 1-27 days). Is it induced by the large uncertainties of the prediction models or different changes among the stations? If it is because of the former, are these results that have so large uncertainties really meaningful?

Technical Points:
1. Table 2. Please change resolution to degree x degree.
2. Fig. 2. Please add values for the y-axis.
3. Fig. 3. Why are Fig. 3b and 3c have the same spatial distribution? Please recheck your data and results.
I would suggest classifying the metrics of SPSS, EPSS, and LPSS into a number of categories, instead of using continuous color bars, which reduces the readability of the figures. Besides, please add units for the legends.
4. Fig. 5, 7-10. Same to Fig. 3, it would be better to classify the metrics of SPSS, EPSS, and LPSS into a number of categories from low to high. Please also add units for the legends.
What are the significance levels of the trends? Are these trends significant?
5. Fig. 6 & 11. The confidence intervals and error bars should be added into these figures.

Citation: https://doi.org/10.5194/tc-2022-244-RC2
- AC2: 'Reply on RC2', Xuemei Li, 17 Mar 2023
  
  Thanks for your good comments and suggestion, in particular, for your recognition of defining potential snowfall phenology. Snowfall is a solid phase of precipitation affects the ecological environment and hydrological processes in mountainous areas and an important water resource (Barnett et al., 2005; Jonas et al., 2008; Bai et al., 2019). One of the most prominent impacts of climate warming has been a shift from snow to rain in temperate and cold regions across the globe (Knowlesetal., 2006; Trenberth, 2011, Jennings and Molotch, 2019). In this context, the snowfall season will inevitably be cut down. Inspired by definitions of vegetation phenology and snow cover phenology (Lu et al., 2006; Piao et al., 2008; Da Silva et al., 2015; Dahlin et al., 2015; Thackeray et al. 2016; Wang et al. 2016; Li et al., 2017; Zhang et al., 2022), we defined potential snowfall phenology and employed three indicators, start of potential snowfall season (SPSS), the end of potential snowfall season (EPSS), and the length of potential snowfall season (LPSS), to identify the possible onset, end, and duration of snowfall. The advancing or delaying of SPSS means potential snowfall comes earlier or later in late-autumn or early-winter, which affect accumulation and storing of solid water resource as snow cover. Likewise, the advancing or delaying of EPSS indicates potential snowfall ends earlier or later in late-winter or early-spring, which is likely to influence the snow-melting, snow albedo, and runoff yield and concentration in mountainous areas such as the Chinese Tianshan mountainous region (CTMR). Above all, the potential snowfall phenology has hydrological significance in snow-dominated region. The the motivation and methodology of the study were not clearly described, we will revise it in the revised manuscript. Besides, in 4.1 Performance of PSP indicators, our work showed the potential snowfall season was able to cover the observed one, which partially validated our results. we will add more detailed validation in the revised manuscript. Related references are as followed:
  Da, Silva, A., Valcu, M., Kempenaers, B.: Light pollution alters the phenology of dawn and dusk singing in common European songbirds, Phil. Trans. R. Soc. B., 370, 1-9, https://doi.org/10.1098/rstb.2014.0126, 2015.
  Dahlin, K., Fisher, R., and Lawrence, P.: Environmental drivers of drought deciduous phenology in the Community Land Model, Biogeosciences, 12, 5061-5074, https://doi.org/10.5194/bg-12-5061-2015, 2015.
  Lu, P., Yu, Q., Liu, J., Lee, X.: Advance of tree-flowering dates in response to urban climate change, Agr. Forest Meteorol., 138, 120-131, https://doi.org/10.1016/j.agrformet.2006.04.002, 2006.
  Piao, S., Ciais, P., Friedlingstein, P., Peylin, P., Reichstein, M., Luyssaert, S., Margolis, H., Fang, J., Barr, A., Chen, A., Grelle, A., Hollinger, D., Laurila, T., Lindroth, A., Richardson, A., Vesala, T.: Net carbon dioxide losses of northern ecosystems in response to autumn warming, Nature, 451, 49-52, https://doi.org/10.1038/nature06444, 2008.
  Wang, X., Wang, S., Hang, Y., Peng, Y.: Snow phenology variability in the Qinghai-Tibetan Plateau and its response to climate change during 2002-2012, J. Geo-Infor. Sci., 18, 1573-1579, 2016. (in Chinese with English abstract)
  Li, X., Zhou, Y., Asrar, G. R., Lin, M.: Characterizing spatiotemporal dynamics in phenology of urban ecosystems based on Landsat data, Sci. Total. Environ., 605, 721-734, https://doi.org/10.1016/j.scitotenv.2017.06.245, 2017.
  Thackeray, S., Henrys, P., Hemming, D., Bell, J., Botham, M., Burthe, S., Helaouet, P., Johns, D., Johns, D., Jones, I., Leech, D., Mackay, E., Massimino, D., Atkinson, S., Bacon, P., Brereton, T., Carvalho, L., Clutton-Brock, T., Duck, C., Edwards, M., Elliott, J., Hall, S., Harrington, R., Pearce-Higgins, J., Høye, T., Kruuk, L., Pemberton, J., Sparks, T., Thompson, P., White, I., Winfield, I., Wanless, S.: Phenological sensitivity to climate across taxa and trophic levels, Nature, 535, 241-245, https://doi.org/10.1038/nature18608, 2016.
  Zhang, B., Li, X., Li, C., Nyiransengiyumva, C., Qin, Q.: Alpine vegetation responses to snow phenology in the Chinese Tianshan mountainous region, J. Mt. Sci-Engl., 19, 1307-1323, https://doi.org/10.1007/s11629-021-7133-4, 2022.
  Knowles, N., Dettinger, M., and Cayan, D.: Trends in snowfall versus rainfall in the western United States, J. Climate, 19, 4545-4559, https://doi.org/10.1175/JCLI3850.1, 2006.
  Trenberth, K., E.: Changes in precipitation with climate change, Clim. Res., 47, 123-128, https://doi.org/10.3354/cr00953, 2011.
  Barnett, T., Adam, J., and Lettenmaier, D.: Potential impacts of a warming climate on water availability in snow-dominated regions, Nature, 438, 303-309, https://doi.org/10.1038/nature04141, 2005.
  Jonas, T., Rixen, C., Sturm, M., Stoeckli, V.: How alpine plant growth is linked to snow cover and climate variability, J. Geophys. Res., 113, 377, G03013, https://doi.org/10.1029/2007JG000680, 2008.
  Specific comments and replies:
  1 The authors failed to well justify the motivation of the study. For example, what is the significance of predicting "potential snowfall phenology"? The start of potential snowfall season (SPSS) does not mean there is a snowfall. Even there may be no any snow during an entire "potential snowfall season". Thus, it may have no any effect on the water and energy balance of a region. In my opinion, the named metrics of "potential snow phenology" here only reflect the fluctuations of temperature, and they have limited hydrological significance.
  Description on the modification: Thanks very much for your good comments. We did fail to show the significance of predicting "potential snowfall phenology". In fact, the start of potential snowfall season (SPSS) means possible onset of snowfall and snowfall is not guaranteed. We don’t think no any snowfall occur during an entire "potential snowfall season" in the snow-dominated region such as the CTMR. Just like potential evapotransporation (PET), it can reflect the energy required to evaporate water, effective wind that can carry water vapor from the surface to the lower atmosphere and other factors, can be a comprehensive reflection of a region's evaporation capacity (https://baike.so.com/doc/4035668-4233412.html). We believe potential snowfall season can reflect comprehensively intra-annual fluctuation of air temperature, timing allocation and capacity of snowfall, as well as water and energy balance in a region. Because if potential snowfall season becomes shorter, the potential rainfall season will expand. Potential water and energy needed by snowfall will change accordingly. We'll take it all into consideration if we have the chance to revise.
  2 The methodology is quite unclear. For example, the authors should provide more details of the calculation process of RST, as it is critical for this study. What data were you used to calculate RST? Did you validate the accuracy of the RST results?If RST is calculated based on a long-term probability statistic of snowfall/rainfall, why is it reasonable to calculate RST at an annual scale? Besides, precipitation phase partitioning is challenging in technique, as temperature humidity, and pressure jointly determine whether precipitation falls as snow/rain (Jennings et al., 2018). If the authors cannot prove the robustness and high accuracy of the RST calculation method for this region, I do not think the results of potential snow phenology are credible. Jennings, K.S., Winchell, T.S., Livneh, B. et al. Spatial variation of the rain-snow temperature threshold across the Northern Hemisphere. Nat Commun 9, 1148 (2018). https://doi.org/10.1038/s41467-018-03629-7
  Description on the modification: thanks for your good suggestion. We did not provide more details of the calculation process of RST for it is the preliminary work from our team and we cited it in the submitted manuscript (Zhang et al, 2017). In the China, including the CTMR, after 1980, the precipitation phase is not labelled (Ding et al., 2014). Visual observer reports of daily precipitation phase are available from 26 meteorological stations across the CTMR during 1950s-1979 (number of records = 237115). Whereas, records from only 20 meteorological stations across the CTMR were used in the work from Jennings et al. (2018), and total number of records was 15535. Although values of rain-snow threshold (RST) or proportions are not included in the observational data, we used the daily precipitation phase data to calculate RSTs based on the frequency intersection method and the probability guarantee method (Zhang et al, 2017). Exactly, precipitation phase partitioning is challenging in technique for temperature, humidity, and pressure jointly determining whether precipitation falls as snow/rain (Jennings et al., 2018). simulation of RSTs were not involved in our work, so precipitation phase partitioning did not affect robustness and high accuracy of our RSTs. Related references are as followed:
  Ding, B., Yang, K., Qin, J., Wang, L., Chen, Y. and He, X.: The dependence of precipitation types on surface elevation and meteorological conditions and its parameterization. J. Hydro., 513, 154-163, https://doi.org/10.1016/j.jhydrol.2014.03.038, 2014.
  Zhang, X., Li, X., Gao, P., Li, Q., and Tang, H.: Separation of precipitation forms based on different methods in Tianshan Mountainous Area, Northwest China, J. Glaciol. Geocryol., 39, 235-244, 2017 (in Chinese with English abstract).
  Jennings, K. S., Winchell, T. S., Livneh, B., Molotch N. P.: Spatial variation of the rain–snow temperature threshold across the Northern Hemisphere, Nat. Commun., 9, 1–9, https://doi.org/10.1038/s41467-018-03629-7, 2018.
  3 L176-181. The authors indicated that they interpolated the model data to stations and then applied a bias correction method to improve the results. However, the interpolated results shown in the maps (Fig. 3, 5, & 7-10) still show large spatial biases. For example, many metrics show obvious circular changes around some stations. These maps do not well reflect the real spatial variations of these variables. The big errors of the results further reduce the value of the study.
  Description on the modification: Thank you very much for your good suggestion. We will check the bias correction method and then revise maps in the revised manuscript.
  4 L430-432. The estimated changes in potential snowfall season metrics fall in big variation ranges (e.g., 1-27 days). Is it induced by the large uncertainties of the prediction models or different changes among the stations? If it is because of the former, are these results that have so large uncertainties really meaningful?
  Description on the modification: Thank you very much for your good suggestion. The variation ranges (e.g., 1-27 days) during the observed period (1961-2017/2020) were big for different warming rates across the CTMR. It was calculated based on the observed data from meteorological stations had no connection with the prediction models.
  Technical Points:
  1 Table 2. Please change resolution to degree x degree.
  Description on the modification: Thank you very much for your good suggestion. We will revise Table 2 as suggested in the revised manuscript.
  2 Fig. 2. Please add values for the y-axis.
  Description on the modification: Thank you very much for your good suggestion. We will revise Fig. 2 as suggested in the revised manuscript.
  3 Fig. 3. Why are Fig. 3b and 3c have the same spatial distribution? Please recheck your data and results. I would suggest classifying the metrics of SPSS, EPSS, and LPSS into a number of categories, instead of using continuous color bars, which reduces the readability of the figures. Besides, please add units for the legends.
  Description on the modification: Thank you very much for your good suggestion. We will revise all figures as suggested in the revised manuscript.
  4 Fig. 5, 7-10. Same to Fig. 3, it would be better to classify the metrics of SPSS, EPSS, and LPSS into a number of categories from low to high. Please also add units for the legends. What are the significance levels of the trends? Are these trends significant?
  Description on the modification: Thank you very much for your good suggestion. The significance level is 0.05 and these trends are significant.We will revise all figures including Fig. 5, 7-10 as suggested and provide details about significance levels in the revised manuscript.
  5 Fig. 6 & 11. The confidence intervals and error bars should be added into these figures.
  Description on the modification: Thank you very much for your good suggestion. We will revise Fig. 6 & 11 as suggested in the revised manuscript.
  
  Citation: https://doi.org/10.5194/tc-2022-244-AC2

Status: closed

RC1:
'Comment on tc-2022-244', Anonymous Referee #1, 31 Dec 2022

Global warming speeds up the solid-liquid water cycle and change snowfall phenology. It was found that postponed snowfall occurrence and advanced snowfall ending took place in the Eurasian continent. Potential snowfall phenology can identify the possible onset, end, and duration of snowfall. To describe the characteristics of the potential snowfall phenology, this manuscript proposed three indicators, the start of potential snowfall season (SPSS), the end of potential snowfall season (EPSS), and the length of potential snowfall season (LPSS). Spatial-temporal variations of those three PSP indicators past, present, and future across the Chinese Tianshan mountainous region (CTMR) were explored. The research is sound and thorough. It provides a new direction to understand the potential snowfall phenology in the alpine region. Therefore, I recommend minor revision of this manuscript. Some specific comments are as follows:

Lines 35-37: please check the number of days cut down for the length, 63 or 64?

Lines 41-43: It suggested that “The results indicate that with constant snowfall intensity, annual total snowfall will decrease, including amount and frequency, .......”

Lines 50-105: The status, shortcomings and need of potential snowfall phenology studies is poorly described in the introduction.

Lines 95-100: The research from Jennings et al.(2018) only illustrated that RST was spatially heterogeneous and does not show that different methods of precipitation pattern separation yield different RST. Please check it.

Lines 128: The sentense “The frequency of rainfall increases while that of snowfall decreases. Besides, precipitation shifting from solid to liquid is obvious” is not clear enough, please check it.

In Figure 1: It is recommended that a general overview map could be added to Figure 1 to help the reader quickly identify the location of the study area.

In Table 1: It is suggested that a column could be added to Table 1 to indicate the duration of the data.

Line 230-232: It is recommended to use “advance / delay” instead of using “smaller or larger” to describe the change of LPSS and EPSS as much as possible.

Line 241-242: please check the slope of LPSS.

Line 371: More should be added here on the comparison of potential snowfall phenology with observed snowfall phenology.

line 377: “4.2 Spatial and temporal heterogeneity” should be changed to “4.2 Temporal heterogeneity”ã

line 408: Is there also an uncertainty error in the definition of PSP?

Citation: https://doi.org/10.5194/tc-2022-244-RC1
- AC1: 'Reply on RC1', Xuemei Li, 17 Mar 2023
  
  Thanks for your recognition of our work. Your specific comments could help us to improve our manuscript greatly. We will revise our manuscript seriously according to your specific comments.
  Some specific comments and replies are as follows:
  1 Lines 35-37: please check the number of days cut down for the length, 63 or 64?
  Description on the modification: Thank you very much for your advice. We have checked the number as suggested.
  2 Lines 41-43: It suggested that “The results indicate that with constant snowfall intensity, annual total snowfall will decrease, including amount and frequency, .......”
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  3 Lines 50-105: The status, shortcomings and need of potential snowfall phenology studies is poorly described in the introduction.
  Description on the modification: we will show more details about status, shortcomings, need, or significance of potential snowfall phenology in the revised manuscript.
  4 Lines 95-100: The research from Jennings et al. (2018) only illustrated that RST was spatially heterogeneous and does not show that different methods of precipitation pattern separation yield different RST. Please check it.
  Description on the modification: we will check it and revise the related sentences.
  5 Lines 128: The sentence “The frequency of rainfall increases while that of snowfall decreases. Besides, precipitation shifting from solid to liquid is obvious” is not clear enough, please check it.
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  6 In Figure 1: It is recommended that a general overview map could be added to Figure 1 to help the reader quickly identify the location of the study area.
  Description on the modification: Thank you very much for your good suggestion. We will revise Figure 1 as suggested.
  7 In Table 1: It is suggested that a column could be added to Table 1 to indicate the duration of the data.
  Description on the modification: Thank you very much for your advice. We have added the duration of the data in Table 1 as suggested.
  8 Line 230-232: It is recommended to use “advance / delay” instead of using “smaller or larger” to describe the change of LPSS and EPSS as much as possible.
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  9 Line 241-242: please check the slope of LPSS.
  Description on the modification: Thank you very much for your advice. We checked the slope of LPSS. It should be -2.7 days/10a. We changed the text and replace the corresponding figure.
  10 Line 371: More should be added here on the comparison of potential snowfall phenology with observed snowfall phenology.
  Description on the modification: Thank you very much for your good suggestion. We will collect more observed data to validate the performance of potential snowfall phenology.
  11 line 377: “4.2 Spatial and temporal heterogeneity” should be changed to “4.2 Temporal heterogeneity”.‚
  Description on the modification: Thank you very much for your advice. We have revised the sentence as suggested.
  
  Citation: https://doi.org/10.5194/tc-2022-244-AC1
RC2:
'Comment on tc-2022-244', Anonymous Referee #2, 09 Feb 2023

General comments:
The manuscript by Li et al. mainly investigated the changes in potential snowfall phenology for the past, present, and future periods in Tianshan, China. They defined three potential snowfall season metrics based on temperature data. Although the definition of potential snowfall phenology is interesting, the hydrological meaning of these metrics is questionable. The methodology of the study was not clearly described. The results lack a strict validation. The credibility of the study needs to be further improved.
Specific comments:
1. Introduction. The authors failed to well justify the motivation of the study. For example, what is the significance of predicting "potential snowfall phenology"? The start of potential snowfall season (SPSS) does not mean there is a snowfall. Even there may be no any snow during an entire "potential snowfall season". Thus, it may have no any effect on the water and energy balance of a region. In my opinion, the named metrics of "potential snow phenology" here only reflect the fluctuations of temperature, and they have limited hydrological significance.
2. The methodology is quite unclear. For example, the authors should provide more details of the calculation process of RST, as it is critical for this study. What data were you used to calculate RST? Did you validate the accuracy of the RST results?
If RST is calculated based on a long-term probability statistic of snowfall/rainfall, why is it reasonable to calculate RST at an annual scale?
Besides, precipitation phase partitioning is challenging in technique, as temperature humidity, and pressure jointly determine whether precipitation falls as snow/rain (Jennings et al., 2018). If the authors cannot prove the robustness and high accuracy of the RST calculation method for this region, I do not think the results of potential snow phenology are credible.
Jennings, K.S., Winchell, T.S., Livneh, B. et al. Spatial variation of the rain–snow temperature threshold across the Northern Hemisphere. Nat Commun 9, 1148 (2018). https://doi.org/10.1038/s41467-018-03629-7
3. L176-181. The authors indicated that they interpolated the model data to stations and then applied a bias correction method to improve the results. However, the interpolated results shown in the maps (Fig. 3, 5, & 7-10) still show large spatial biases. For example, many metrics show obvious circular changes around some stations. These maps do not well reflect the real spatial variations of these variables. The big errors of the results further reduce the value of the study.
4. L430-432. The estimated changes in potential snowfall season metrics fall in big variation ranges (e.g., 1-27 days). Is it induced by the large uncertainties of the prediction models or different changes among the stations? If it is because of the former, are these results that have so large uncertainties really meaningful?

Technical Points:
1. Table 2. Please change resolution to degree x degree.
2. Fig. 2. Please add values for the y-axis.
3. Fig. 3. Why are Fig. 3b and 3c have the same spatial distribution? Please recheck your data and results.
I would suggest classifying the metrics of SPSS, EPSS, and LPSS into a number of categories, instead of using continuous color bars, which reduces the readability of the figures. Besides, please add units for the legends.
4. Fig. 5, 7-10. Same to Fig. 3, it would be better to classify the metrics of SPSS, EPSS, and LPSS into a number of categories from low to high. Please also add units for the legends.
What are the significance levels of the trends? Are these trends significant?
5. Fig. 6 & 11. The confidence intervals and error bars should be added into these figures.

Citation: https://doi.org/10.5194/tc-2022-244-RC2
- AC2: 'Reply on RC2', Xuemei Li, 17 Mar 2023
  
  Thanks for your good comments and suggestion, in particular, for your recognition of defining potential snowfall phenology. Snowfall is a solid phase of precipitation affects the ecological environment and hydrological processes in mountainous areas and an important water resource (Barnett et al., 2005; Jonas et al., 2008; Bai et al., 2019). One of the most prominent impacts of climate warming has been a shift from snow to rain in temperate and cold regions across the globe (Knowlesetal., 2006; Trenberth, 2011, Jennings and Molotch, 2019). In this context, the snowfall season will inevitably be cut down. Inspired by definitions of vegetation phenology and snow cover phenology (Lu et al., 2006; Piao et al., 2008; Da Silva et al., 2015; Dahlin et al., 2015; Thackeray et al. 2016; Wang et al. 2016; Li et al., 2017; Zhang et al., 2022), we defined potential snowfall phenology and employed three indicators, start of potential snowfall season (SPSS), the end of potential snowfall season (EPSS), and the length of potential snowfall season (LPSS), to identify the possible onset, end, and duration of snowfall. The advancing or delaying of SPSS means potential snowfall comes earlier or later in late-autumn or early-winter, which affect accumulation and storing of solid water resource as snow cover. Likewise, the advancing or delaying of EPSS indicates potential snowfall ends earlier or later in late-winter or early-spring, which is likely to influence the snow-melting, snow albedo, and runoff yield and concentration in mountainous areas such as the Chinese Tianshan mountainous region (CTMR). Above all, the potential snowfall phenology has hydrological significance in snow-dominated region. The the motivation and methodology of the study were not clearly described, we will revise it in the revised manuscript. Besides, in 4.1 Performance of PSP indicators, our work showed the potential snowfall season was able to cover the observed one, which partially validated our results. we will add more detailed validation in the revised manuscript. Related references are as followed:
  Da, Silva, A., Valcu, M., Kempenaers, B.: Light pollution alters the phenology of dawn and dusk singing in common European songbirds, Phil. Trans. R. Soc. B., 370, 1-9, https://doi.org/10.1098/rstb.2014.0126, 2015.
  Dahlin, K., Fisher, R., and Lawrence, P.: Environmental drivers of drought deciduous phenology in the Community Land Model, Biogeosciences, 12, 5061-5074, https://doi.org/10.5194/bg-12-5061-2015, 2015.
  Lu, P., Yu, Q., Liu, J., Lee, X.: Advance of tree-flowering dates in response to urban climate change, Agr. Forest Meteorol., 138, 120-131, https://doi.org/10.1016/j.agrformet.2006.04.002, 2006.
  Piao, S., Ciais, P., Friedlingstein, P., Peylin, P., Reichstein, M., Luyssaert, S., Margolis, H., Fang, J., Barr, A., Chen, A., Grelle, A., Hollinger, D., Laurila, T., Lindroth, A., Richardson, A., Vesala, T.: Net carbon dioxide losses of northern ecosystems in response to autumn warming, Nature, 451, 49-52, https://doi.org/10.1038/nature06444, 2008.
  Wang, X., Wang, S., Hang, Y., Peng, Y.: Snow phenology variability in the Qinghai-Tibetan Plateau and its response to climate change during 2002-2012, J. Geo-Infor. Sci., 18, 1573-1579, 2016. (in Chinese with English abstract)
  Li, X., Zhou, Y., Asrar, G. R., Lin, M.: Characterizing spatiotemporal dynamics in phenology of urban ecosystems based on Landsat data, Sci. Total. Environ., 605, 721-734, https://doi.org/10.1016/j.scitotenv.2017.06.245, 2017.
  Thackeray, S., Henrys, P., Hemming, D., Bell, J., Botham, M., Burthe, S., Helaouet, P., Johns, D., Johns, D., Jones, I., Leech, D., Mackay, E., Massimino, D., Atkinson, S., Bacon, P., Brereton, T., Carvalho, L., Clutton-Brock, T., Duck, C., Edwards, M., Elliott, J., Hall, S., Harrington, R., Pearce-Higgins, J., Høye, T., Kruuk, L., Pemberton, J., Sparks, T., Thompson, P., White, I., Winfield, I., Wanless, S.: Phenological sensitivity to climate across taxa and trophic levels, Nature, 535, 241-245, https://doi.org/10.1038/nature18608, 2016.
  Zhang, B., Li, X., Li, C., Nyiransengiyumva, C., Qin, Q.: Alpine vegetation responses to snow phenology in the Chinese Tianshan mountainous region, J. Mt. Sci-Engl., 19, 1307-1323, https://doi.org/10.1007/s11629-021-7133-4, 2022.
  Knowles, N., Dettinger, M., and Cayan, D.: Trends in snowfall versus rainfall in the western United States, J. Climate, 19, 4545-4559, https://doi.org/10.1175/JCLI3850.1, 2006.
  Trenberth, K., E.: Changes in precipitation with climate change, Clim. Res., 47, 123-128, https://doi.org/10.3354/cr00953, 2011.
  Barnett, T., Adam, J., and Lettenmaier, D.: Potential impacts of a warming climate on water availability in snow-dominated regions, Nature, 438, 303-309, https://doi.org/10.1038/nature04141, 2005.
  Jonas, T., Rixen, C., Sturm, M., Stoeckli, V.: How alpine plant growth is linked to snow cover and climate variability, J. Geophys. Res., 113, 377, G03013, https://doi.org/10.1029/2007JG000680, 2008.
  Specific comments and replies:
  1 The authors failed to well justify the motivation of the study. For example, what is the significance of predicting "potential snowfall phenology"? The start of potential snowfall season (SPSS) does not mean there is a snowfall. Even there may be no any snow during an entire "potential snowfall season". Thus, it may have no any effect on the water and energy balance of a region. In my opinion, the named metrics of "potential snow phenology" here only reflect the fluctuations of temperature, and they have limited hydrological significance.
  Description on the modification: Thanks very much for your good comments. We did fail to show the significance of predicting "potential snowfall phenology". In fact, the start of potential snowfall season (SPSS) means possible onset of snowfall and snowfall is not guaranteed. We don’t think no any snowfall occur during an entire "potential snowfall season" in the snow-dominated region such as the CTMR. Just like potential evapotransporation (PET), it can reflect the energy required to evaporate water, effective wind that can carry water vapor from the surface to the lower atmosphere and other factors, can be a comprehensive reflection of a region's evaporation capacity (https://baike.so.com/doc/4035668-4233412.html). We believe potential snowfall season can reflect comprehensively intra-annual fluctuation of air temperature, timing allocation and capacity of snowfall, as well as water and energy balance in a region. Because if potential snowfall season becomes shorter, the potential rainfall season will expand. Potential water and energy needed by snowfall will change accordingly. We'll take it all into consideration if we have the chance to revise.
  2 The methodology is quite unclear. For example, the authors should provide more details of the calculation process of RST, as it is critical for this study. What data were you used to calculate RST? Did you validate the accuracy of the RST results?If RST is calculated based on a long-term probability statistic of snowfall/rainfall, why is it reasonable to calculate RST at an annual scale? Besides, precipitation phase partitioning is challenging in technique, as temperature humidity, and pressure jointly determine whether precipitation falls as snow/rain (Jennings et al., 2018). If the authors cannot prove the robustness and high accuracy of the RST calculation method for this region, I do not think the results of potential snow phenology are credible. Jennings, K.S., Winchell, T.S., Livneh, B. et al. Spatial variation of the rain-snow temperature threshold across the Northern Hemisphere. Nat Commun 9, 1148 (2018). https://doi.org/10.1038/s41467-018-03629-7
  Description on the modification: thanks for your good suggestion. We did not provide more details of the calculation process of RST for it is the preliminary work from our team and we cited it in the submitted manuscript (Zhang et al, 2017). In the China, including the CTMR, after 1980, the precipitation phase is not labelled (Ding et al., 2014). Visual observer reports of daily precipitation phase are available from 26 meteorological stations across the CTMR during 1950s-1979 (number of records = 237115). Whereas, records from only 20 meteorological stations across the CTMR were used in the work from Jennings et al. (2018), and total number of records was 15535. Although values of rain-snow threshold (RST) or proportions are not included in the observational data, we used the daily precipitation phase data to calculate RSTs based on the frequency intersection method and the probability guarantee method (Zhang et al, 2017). Exactly, precipitation phase partitioning is challenging in technique for temperature, humidity, and pressure jointly determining whether precipitation falls as snow/rain (Jennings et al., 2018). simulation of RSTs were not involved in our work, so precipitation phase partitioning did not affect robustness and high accuracy of our RSTs. Related references are as followed:
  Ding, B., Yang, K., Qin, J., Wang, L., Chen, Y. and He, X.: The dependence of precipitation types on surface elevation and meteorological conditions and its parameterization. J. Hydro., 513, 154-163, https://doi.org/10.1016/j.jhydrol.2014.03.038, 2014.
  Zhang, X., Li, X., Gao, P., Li, Q., and Tang, H.: Separation of precipitation forms based on different methods in Tianshan Mountainous Area, Northwest China, J. Glaciol. Geocryol., 39, 235-244, 2017 (in Chinese with English abstract).
  Jennings, K. S., Winchell, T. S., Livneh, B., Molotch N. P.: Spatial variation of the rain–snow temperature threshold across the Northern Hemisphere, Nat. Commun., 9, 1–9, https://doi.org/10.1038/s41467-018-03629-7, 2018.
  3 L176-181. The authors indicated that they interpolated the model data to stations and then applied a bias correction method to improve the results. However, the interpolated results shown in the maps (Fig. 3, 5, & 7-10) still show large spatial biases. For example, many metrics show obvious circular changes around some stations. These maps do not well reflect the real spatial variations of these variables. The big errors of the results further reduce the value of the study.
  Description on the modification: Thank you very much for your good suggestion. We will check the bias correction method and then revise maps in the revised manuscript.
  4 L430-432. The estimated changes in potential snowfall season metrics fall in big variation ranges (e.g., 1-27 days). Is it induced by the large uncertainties of the prediction models or different changes among the stations? If it is because of the former, are these results that have so large uncertainties really meaningful?
  Description on the modification: Thank you very much for your good suggestion. The variation ranges (e.g., 1-27 days) during the observed period (1961-2017/2020) were big for different warming rates across the CTMR. It was calculated based on the observed data from meteorological stations had no connection with the prediction models.
  Technical Points:
  1 Table 2. Please change resolution to degree x degree.
  Description on the modification: Thank you very much for your good suggestion. We will revise Table 2 as suggested in the revised manuscript.
  2 Fig. 2. Please add values for the y-axis.
  Description on the modification: Thank you very much for your good suggestion. We will revise Fig. 2 as suggested in the revised manuscript.
  3 Fig. 3. Why are Fig. 3b and 3c have the same spatial distribution? Please recheck your data and results. I would suggest classifying the metrics of SPSS, EPSS, and LPSS into a number of categories, instead of using continuous color bars, which reduces the readability of the figures. Besides, please add units for the legends.
  Description on the modification: Thank you very much for your good suggestion. We will revise all figures as suggested in the revised manuscript.
  4 Fig. 5, 7-10. Same to Fig. 3, it would be better to classify the metrics of SPSS, EPSS, and LPSS into a number of categories from low to high. Please also add units for the legends. What are the significance levels of the trends? Are these trends significant?
  Description on the modification: Thank you very much for your good suggestion. The significance level is 0.05 and these trends are significant.We will revise all figures including Fig. 5, 7-10 as suggested and provide details about significance levels in the revised manuscript.
  5 Fig. 6 & 11. The confidence intervals and error bars should be added into these figures.
  Description on the modification: Thank you very much for your good suggestion. We will revise Fig. 6 & 11 as suggested in the revised manuscript.
  
  Citation: https://doi.org/10.5194/tc-2022-244-AC2

Xuemei Li et al.

Viewed

Total article views: 471 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
296	160	15	471	3	6

HTML: 296
PDF: 160
XML: 15
Total: 471
BibTeX: 3
EndNote: 6

Views and downloads (calculated since 08 Dec 2022)

Month	HTML	PDF	XML	Total
Dec 2022	150	29	4	183
Jan 2023	38	74	2	114
Feb 2023	50	21	2	73
Mar 2023	33	17	4	54
Apr 2023	16	11	1	28
May 2023	9	8	2	19

Cumulative views and downloads (calculated since 08 Dec 2022)

Month	HTML	PDF	XML	Total
Dec 2022	150	29	4	183
Jan 2023	38	74	2	114
Feb 2023	50	21	2	73
Mar 2023	33	17	4	54
Apr 2023	16	11	1	28
May 2023	9	8	2	19

Viewed (geographical distribution)

Total article views: 459 (including HTML, PDF, and XML) Thereof 459 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 31 May 2023

Short summary

Quantifying change in potential snowfall phenology (PSP) is an important area of research for understanding of regional climate change present and future. However, few studies have focused on the PSP and its change in alpine mountainous region. We initially proposed three indicators to describe characteristics of the PSP. It provides a new direction to capture the potential snowfall penology in alpine mountainous region and can be easily expanded to other snow-dominated regions.


Total:	0
HTML:	0
PDF:	0
XML:	0