Review Reports

Round 1

Reviewer 1 Report

The article by Bonnell et al. provides very interesting insight into potential biases in retrieval of Snow Water Equivalent (SWE) in wet snow conditions. SWE is a key input parameter for studies in hydrology, snow, water resources and biodiversity. The determination of SWE over large areas using satellite remote sensing remains a major challenge due to the many sources of uncertainty.

In this study the authors combine high mountain local measurements from ground penetrating radar (GPR) and in-situ measurements to compare simulated SWE and observation based one. The authors show that SWE biases are significant and can be as high as 40% if dry snow assumptions are used for permittivity when the snow is wet.

The article is very well written and the elements for a good understanding are given. A few elements remain to be clarified. In particular, the authors could expand the section on the implications for satellite SWE retrieval, as the expectations of the research and user community are growing and it is important to draw the attention of this community to the potential uncertainties of some of these products:

 

• I found it complicated to follow the flow of pre-processing for SWE and LWC calculations. It would be very useful to add a general workflow diagram, with the working assumptions at each step, to make the paper easier to read. 
• Implications for remote sensing of SWE: what would be the strategy using satellite data to reduce SWE uncertainties ? what would be the place of physically-based snow models ?
• Do you think it would be possible to develop bias correction schemes (fonction of LWC) to be applied to SWE estimates, and these bias corrections could then be adjusted near the measurement stations?
• The strong biases observed also concern rain-on-snow situations, don't they? Were there any such events during the measurement campaign?
• it would be useful to have elements on the spatial variability of SWE or LWC? the authors could possibly introduce some maps or other 2D diagrams ?
• How snow depth is retrieved from terrestrial laser scanning and how the retrievals are they evaluated ?

 

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

I think the authors carried out a significant amount of work to evaluate the spatiotemporal variations in liquid water content in a seasonal snowpack over Colorado, USA. I recommend the manuscript for publication after the following minor changes:

• The introduction starts are very clear and lead correctly to a certain problem statement. Can you explain more about synthetic SWE- retrievals with schematic diagrams? Also,  you need to discuss the dry snow density model clearly.
• Provide a list of remote sensing approaches that are used to measure SWE and their limitation? Then introduce your methodology for this study how did you apply it? What is the uniqueness of the proposed technique and its potential impacts, over other previous techniques?
• Can you explain hydrologic uncertainty? Also, explain uncertainty in terms of climate change.
• Your study area has complex terrain (e.g. study area surrounded by mountains and high elevation) which has a high dependency on local climate, topographic complexity. Can you provide climatic information for the selected study areas? You should provide a show Köppen–Geiger climatic zones on the map.
•  Can you provide a high impactful schematic diagram for the overall process?
• Your error analysis results are incomplete. Can you please add detailed error analysis results using random error (normalized centered root-mean-square error) and systematic error (mean relative error)?
• can you please explain more about the uncertainty of your proposed methods?
• Your study area is characterized by complex topography, you should compare your results with another complex terrain with the future recommendations.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript of Bonnell et al. presents a very thorough analysis of potential pitfalls and biases of snow characteristics based on remote sensing. This is a highly relevant topic within climate change research, as the authors rightly point out. Longterm changes in snow characteristics are an important parameter. Nevertheless, approximation techiques that map such parameters across large areas based on satellite sensors need to be carefully calibrated. The authors have produced an extremely detailed groundtruthing study that provides crucial data to improve such satellite-based interpretations in the future. Bonnell et al. found devations of up to 40% of a key parameter, highlighting the need for much better calibration and additional efforts to assure reproducibility of the desired snow parameters. The stability of the interpreted changes is important to avoid misrepresentations and misleading conclusions.

The text is very well written and large parts of it can also easily be understood by non-specialists. The abstract summarizes well the key points and also mentions the discrepancy transparently, which hopefully will motivate significant future research to address the issues. I notice only a few points which may be worth addressing:

 

Lines 40/41: For specialists, it is clear what snow-water equivalent means. For others, it might be useful to briefely define it again.

 

Line 41: SWE in the US has declined. But Northern Hemisphere (NH) snow cover has apparently increased: https://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=1

Is there a trend for NH SWE?

 

Summed up, I believe that this paper requires very little additional work and is suitable of being published.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf