Snow Cover Temporal Dynamic Using MODIS Product, and Its Relationship with Precipitation and Temperature in the Tropical Andean Glaciers in the Alto Santa Sub-Basin (Peru)
, Saul Sardón 1, Fredy Calizaya 4, Osmar Cuentas 5, José Cahuana 6, Carmen Mindani 7 and Walquer Huacani 8Round 1
Reviewer 1 Report
This is an interesting paper and the author nicely explains the changing pattern of the snow-covered area using MODIS and Landsat Data using Google Earth Engine. The author also established and examined the consistency of the historical series of climatological variables of precipitation and daily temperature is an essential step in understanding the impacts of climate change on tropical glaciers. From this point of view, this paper has had great importance in recent times. However, there are some observations that need to be adequately addressed (see below).
· Line 88, change “spatial resolution” to “temporal resolution.
· Section 2.3, it is necessary to add the process of how cloud cover was removed from the MODIS images, other than winter. Also, how were atmospheric and radiometric corrections performed? What is the positional accuracy of the images? And how was it measured/checked (accuracy level)? Need to be cleared.
· It is better to provide climatological data used characteristics in a table in section 2.4. Such as, in the table, Name of station location, location (lat, long), time of data available etc.
· Figure 4, Arrange figures Year wise (In 1st row 2003 2010 2016 for MODIS classified maps and in the second row 2003 2010 2016 for Landsat Classified maps)
· Figure 5, X Y information is not clear, increase resolution for more visibility.
· Fig 9, in the x-axis delete ‘(daily classification)’ and in fig caption delete the word ‘analysis’ after trend.
· Lines 384-387, No information was found in the paper regarding snow depth data and validation.
· Line 388, is it Fig 8 or Fig 9?
· In the discussion section, if the author adds an analysis/relationship between the rate of temperature increase and the rate of snow-covered area decrease and also a relationship between precipitation and snow cover area, it would be better to understand the impotence of temperature and rainfall as major factors.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors consisted of processing 6,578 MODIS images and generating 18-year time series using the Platform Google Earth Engine (GEE), and analyzed the dynamics of the snow cover area (SCA) of the Tropical Andean Glaciers in the Alto Santa Sub-Basin. The results showed that SCA decreased from 649 km2 to 311.6 km2 between 2000-2017 represents a retreat of 41%, it is interesting and useful. There are some comments:
1. Lines 107-108:“The climate is humid mainly and;”？and line 116:“in the lower part of the basin. [20,21].”
2. Whether the tourism affects the retreat behavior of SCA?
3. The authors should provide the average snow cover area from 2000 to 2017, so that readers can intuitively found the trend (In Fig.3).
4. I think Fig. 9 should give a mean value, and not maximum SCA. The results are not representative. For example, if we analyze the snow cover area (SCA) at the alto Santa sub-basin from 2001 – 2015, the conclusion is not valid.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
Trend and Snow Cover Dynamic Using MODIS Product, Precipitation and Temperature in the Tropical Andean Glaciers in the Alto Santa Sub-Basin (Peru)
This paper uses the Google Earth Engine Platform (GEE) MODIS image processing to generate an 18-year time series to study the dynamics of snow cover in the Upper Santa River Sub basin located in the tropical Andes. The normalized differential snow index (NDSI) of spectral characteristics, daily precipitation, and temperature were analyzed and validated. Research shows that during the period 2000-2017, the snow cover area decreased from 649 km2 to 311.6 km2 and these changes are related to climate.
Here are some suggestions for the authors:
1. Pictures need to be renumbered. Two figure 4 appear in the text.
2. Figure 4 and Figure 5 lack clarity, please improve the resolution of the images.
3. Line 209, What does ‘likely’ mean? Please accurately describe the statistical method used by the authors, rather than using such uncertain words.
4. In the discussion in Section 4, there were too many references and too few comments on the research results in Section 3 of this study. The author should provide a more detailed explanation of the conclusive images in the article
5. In Section 3.5, precipitation and temperature are separately analyzed for standardization with SCA, But a single factor cannot represent climate.The author can consider supplementing the correlation between multiple factors and SCA.
6. The English writing should be improved. For example, lines 321-333 are too long and difficult to read and the same problem occurs in multiple sentences in the text.
My opinion is major revision. Overall, the article is logically rigorous and the results of the experiment are clear. However, some minor issues still need to be improved.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 4 Report
Summary
Glaciers are important components of earth system but vulnerable to climate change. About 71% of tropical glaciers are in Peru, with most in the Cordillera Blanca (study region), but the long-term temporal dynamics of snow cover and its relationship with temperature and precipitation in the studied region remain underexplored. This study used multi-year MODIS snow cover data and site observed climate data to analyze the temporal dynamics of snow cover area in Peru. The results highlight the importance of climate change on snow cover changes. The study is important and could be published after addressing my following concerns listed as bellow.
Specific comments
1) The title seems to have grammar issue. I suggest to revise it as “Temporal dynamics of snow cover area and its relationship with local climate in the Tropical Andean glaciers of Peru”
2) For the introduction, Line 65-85, I understand it’s important to analyze the glacier in the tropical Andes, but the differences between this study and previous studies are not clear. Please clarify, what’s the major differences between this study and previous studies? That’s related to the innovation of this study and is very important.
3) Line 59: “GTA” or “GCA”?
4) Line 79: it’s unclear what Landsat images are used for here?
5) Line 87-88: revise “with daily spatial resolution” as “with daily temporal resolution and 500m spatial resolution”
6) Fig. 2, Daily precipitation and temperature should be organized in parallel instead of sequentially connected.
7) Fig. 2: how is Landsat used to validate the MODIS? What’s the validation performance? Please clarify it in the Methods section.
8) Fig. 3: How did you get the total area according to the NDSI index (Eq.1)? Please clarify it in the methods section. I assumed there should be a classification based on NDSI threshold and the classification accuracy should be given or the used method should be cited.
9) Line 161: “06 meteorological stations”?
10) Line 163-164: “02 meteorological stations”?
11) Line 169-170: “double mass analysis, completing missing data, extension, and validation” those items were first mentioned but without any brief introduction. Readers can hardly understand what these items specifically represent.
12) Line 186: “VAR”? what does it represent? Clarify in the main text
13) Remove line 249-252 if you did not show these results
14) Fig. 6: the unit of precipitation is “mm/day” or “mm/s”
15) Line 293: “using four three methods” to “using three methods”, also DCCA first mentioned but without full name. Additionally, citations are required for each method used.
16) Line 296-298, remove those sentences, since the results cannot judge what’s accurate and acceptable.
17) Table 2, are those correlation results all significant? Please clarify the p-value for each result
18) Fig. 8 and Fig.9 are in the main text but without any analysis. Please briefly analyze Fig.8-9 in section 3.5
19) Section 4.2, I understand that correlation-based methods are most commonly used methods, but correlation-based methods can be problematic for confounding impacts and causality inference therefore is more powerful. In addition, this study mainly analyzed the impacts of temperature and precipitation on snow cover, actually, other factors including solar radiation, surface albedo changes, and vapor pressure deficit (VPD) could also affect snow cover. The following papers list related studies including causality inference, and the impacts of solar radiation and VPD on land surface energy flux partitioning and surface temperature, please briefly discuss those factors in section 4.2.
Runge, J., Nowack, P., Kretschmer, M., Flaxman, S., & Sejdinovic, D. (2019). Detecting and quantifying causal associations in large nonlinear time series datasets. Science advances, 5(11), eaau4996.
Li, F., Zhu, Q., Riley, W. J., Yuan, K., Wu, H., & Gui, Z. (2022). Wetter California projected by CMIP6 models with observational constraints under a high GHG emission scenario. Earth's Future, 10(4), e2022EF002694.
Yuan, K., Zhu, Q., Riley, W. J., Li, F., & Wu, H. (2022). Understanding and reducing the uncertainties of land surface energy flux partitioning within CMIP6 land models. Agricultural and Forest Meteorology, 319, 108920.
Yuan, K., Zhu, Q., Zheng, S., Zhao, L., Chen, M., Riley, W. J., ... & Chen, L. (2021). Deforestation reshapes land-surface energy-flux partitioning. Environmental Research Letters, 16(2), 024014.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Trend and Snow Cover Dynamic Using MODIS Product, Pre- 2 cipitation and Temperature in the Tropical Andean Glaciers in 3 the Alto Santa Sub-Basin (Peru)
This manuscript described an analysis method for trend and snow cover dynamic in 18 years. Overall, the article is logically rigorous and the results of the experiment are clear. However, some minor issues still need to be improved.
Here are some suggestions for the authors:
1. In abstract: ‘Spectral signature Normalized Difference Snow Index (NDSI) and daily precipitation and temperature were analyzed and validated.’ It should be indicated on what method the analysis is based.
2. I am interested in whether this paper can predict the trend of snow and ice.
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
Reviewer 4 Report
Summary
Thanks for the authors’ efforts for addressing some of my concerns, but some of my concerns are still not addressed. Here I reclaim why they are important and it should be easy to address those problems.
Specific comments
- The title still seems to have grammar issue. It could be revised as “Snow Cover Temporal Dynamic Using MODIS Product, and its Relationship with Local Climate in the Tropical Andean Glaciers in the Alto Santa Sub-Basin (Peru)” instead of “Temporal and Snow Cover Dynamic Using MODIS Product, and its Relationship with Local Climate in the Tropical Andean Glaciers in the Alto Santa Sub-Basin (Peru)”
- Section 4.2, I understand that correlation-based methods are most commonly used methods, but correlation-based methods can be problematic for confounding impacts and causality inference therefore is more powerful. In addition, this study mainly analyzed the impacts of temperature and precipitation on snow cover, actually, other factors including solar radiation, surface albedo changes, and vapor pressure deficit (VPD) could also affect snow cover. The following papers list related studies including causality inference, and the impacts of solar radiation and VPD on land surface energy flux partitioning and surface temperature, please briefly discuss those factors in section 4.2.
I totally understand that there is data limitation in your country Peru, and I do not mean that you need to add any additional data and experiments to your current research. I mean that we need to at least clarify or briefly discuss the limitation of this study, which is very common and could be important for the readers to know that so that they can do additional research based on your discussion or limitations. Make sense?
Runge et al. "Detecting and quantifying causal associations in large nonlinear time series datasets." Science advances 5.11 (2019): eaau4996.
Li et al. "Wetter California projected by CMIP6 models with observational constraints under a high GHG emission scenario." Earth's Future 10.4 (2022): e2022EF002694.
Yuan et al. "Understanding and reducing the uncertainties of land surface energy flux partitioning within CMIP6 land models." Agricultural and Forest Meteorology 319 (2022): 108920.
Author Response
"Please see the attachment."
Author Response File: Author Response.pdf
