Synthetic Aperture Radar Monitoring of Snow in a Reindeer-Grazing Landscape

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper presented a method to observe start of snow melt and end of snow cover based on Sentinel-1 SAR satellite observations. The method is based on backscattering behaviour of VV and VH polarizations. The study was conducted in northern Sweden in 2017-2020 in a reindeer herding district. Automated weather station is located in the area, which was used for comparison with closest satellite pixel. Additionally, snow melt properties were analyzed for different landcover classes. The study resulted that Sentinel-1 SAR tool is useful for estimating the snow melt properties.

The paper presented novel method to derive the start day of snow melt and the end day of snow cover. The parameters are relevant for several purposes, including hydrological applications, in addition to the mentioned vegetation growth important for reindeer. Overall, the paper was well and clearly written. I have only minor comments.

Specific comments:

Figures 2, 6 & 7: Marking for AWS could be better visible from the background.

Line 138: "Data from an automatic weather station (AWS) was provided from [5, 40]." You could define in the text where AWS data was available.

Lines 148, 241, 298: Change S-1 to Sentinel-1 in the headings

Line 155: What are IW2 and IW3?

Lines 252-278: The chapter is a bit difficult to read. Consider to add a table for percentage values.

Line 301: You could consider to use data only from the expected range so that August would not be included, or does it show any melt date then?

Line 371 and 444: Consider to change first S-1 to Sentinel-1 in the beginning of discussion and conclusion chapters.

Line 375: Consider more simple heading such as "Variations in backscattering and spring snowmelt"

Line 407: Remove "And" 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Snow cover is an important indicator of global climate change, and its changes exhibit complex correlations with climate. Changes in snow cover also have a significant impact on ecosystems. The response patterns of different ecosystem elements to changes in snow cover characteristics are complex and diverse, which may affect the phenology of vegetation, growth season length, productivity, as well as the quantity and quality of litter. Studying the impact of snow cover changes on ecosystems can help us better protect and manage ecosystem resources.

The authors have used Sentinel-1 Synthetic Aperture Radar (SAR) to determine the onset of surface melt and the end of the snow cover  in northern Sweden during 2017-2020.   it is found that the method authors is suitable for investigating snowmelt.  VH data provide a more reliable overview of SOSM and EOS than VV data.

It's an interesting work. but there is still room for improvement.

1. the research  time period is too short to make better understand of the impact of changes in snow cover and surface melt timing on vegetation. Please expand the research period to obtain more reliable research results.

2. Only explained the phenomenon of snow melting changes during the research period, lacking sufficient mechanism analysis and explanation for the differences in changes

3.The author compared the changes in snow melting under different land cover types. However, land cover is related to factors such as altitude and terrain, and further in-depth analysis of these factors should be conducted to explain the differences in snow melting and annual changes

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper utilizes Sentinel-1 synthetic aperture radar (SAR) data spanning from 2017 to 2020 to examine the onset and cessation times of surface melting in specific regions of northern Sweden.    Comparative analysis was conducted with field measurements obtained from an automatic weather station situated in Laevás sámi.    The findings reveal minimal disparities between the analytical outcomes derived from SAR data and the ground truth measurements.    The methodologies employed, including the minimum backscatter and threshold techniques, offer valuable supplements to global snowmelt monitoring efforts.    Nevertheless, the experimental comparison within the paper is notably deficient, as outlined below:

 

1.   The article's data selection is limited to just four years, rendering it somewhat arbitrary.    To bolster the credibility of the methodologies delineated in the article, a broader range of sample years should be incorporated.

 

2.   The examination solely focuses on data from a singular region, thereby failing to demonstrate the universal applicability of the proposed method for global snowmelt monitoring.    To establish the efficacy of the proposed method across diverse geographical contexts, additional sample regions should be included.

 

3.   The comparison section of the experiment exclusively evaluates the method introduced in this paper, neglecting comparisons with alternative approaches.    This omission undermines the ability to underscore the effectiveness of the proposed method vis-à-vis existing techniques. Therefore, incorporating comparative analyses with other methodologies is essential to contextualize the method's effectiveness.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

This article uses Sentinel-1 Synthetic Aperture Radar (SAR) data from 2017 to 2021 to study the start and end times of surface melting in specific areas of northern Sweden. The analysis results were compared with on-site measurements from an automatic weather station located in Laevás sámi. The results indicate that the difference between the two is relatively small. The minimum backscatter and threshold methods used in this article are beneficial supplements to global snowmelt monitoring. Compared with the initial draft, this manuscript has added data from 2021 and further refined the comparison of experimental results, making the comparison more detailed and enhancing the persuasiveness of the article. The language and content of the article have been reasonably revised and become more fluent. But there are still some details that need to be revised in the article. For example, is Table 2 in the fourth section missing a description of the 2021 data?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf