The authors mapped the open water body in the PPR using SAR images and high-resolution DTM data. Time series of water body maps were obtained and the dynamics of open water bodies were analyzed. Generally, this manuscript is well prepared with good structures and figures. However, I have some major concerns about the current work, as:

1) The validation (or accuracy evaluation) of the method. Currently, the authors evaluate the accuracy from a viewpoint of classification, using user accuracy and produce accuracy for the pixels selected. However, as the authors analyzed in section 3.2, what the hydrologist cares about is the number and area of the water body. So, the authors are suggested to validate (evaluate) their method by comparing the area and number estimated from SAR with those from NAIP.

2)L332-345, the authors explained the different performances of different polarization combinations. However, this work can be done more physically by introducing the Radar functions. Actually, the authors can start from the radar functions and then build their algorithm on the basis of microwave radiative transfer theory.

3) the abstract is too long, please shorten it.

4) some figures, like figure 5, are suggested to add scales and the compass. Since not all readers are familiar with UTM(zone 14)

5) The authors are suggested to pay attention to the usage of abbreviations, for example, please define GRASS and dual-pol.

 

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 

Review

This article is novel in its approach to coping with the difficult problem of separating surface water from (possibly wet) vegetated land in a small catchment of 2770 sq.km in the Prairie Pothole Region in North America.  The data to make the scheme feasible was obtained from the orbiting satellite Sentinal-1A, at a resolution of about 20m, sampled during the summer and autumn months, so were not blanketed by snow.  The authors developed the tools to obtain valid images to work from.

The text is very well laid out and informative – there are only a few places that need repair and I am glad to say that I read every word so that [perhaps not to the authors’ taste] I made some very minor alterations, where I deemed necessary.  The math is well set out, but I would like the equations inset from the margins; however that’s a minor issue of layout. The Figures are good and easy to understand.  Regretfully some of them are placed up to 3 pages away from their first mention, which is irritating. There is however the odd glitch, like Fig. 9, where the caption and label are in in hectares, and the image legends are in sq m!!  Fortunately, I could not find any others.

I judge that this paper is worth publishing after repair, so recommend minor corrections.  I am returning the version of the paper that I have marked up, attached to this review.  In addition, I will import the more substantial of my remarks to be listed below my signature, which is my wont.

Geoff Pegram

18 August 2021

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 

Comments, line-by-line from the article, followed by # and my suggestions

Abstract. The North American Prairie Pothole Region (PPR) represents a large system of wetlands with great importance for biodiversity, water storage and flood management. Knowledge of seasonal and inter-annual surface water dynamics in the PPR is important for understanding the functionality of these wetland ecosystems and the changing degree of hydrologic connectivity between them.

#  Try and make this more interesting so that, although it is informative as an introduction, please make your abstract (or intro) more descriptive - Wiki's description is a good springboard – your paper is not overlong:

“The Prairie Pothole Region (PPR) is an expansive area of the northern Great Plains that contains thousands of shallow wetlands known as potholes. These potholes are the result of glacier activity in the Wisconsin glaciation, which ended about 10,000 years ago. The decaying ice sheet left behind depressions formed by the uneven deposition of till in ground moraines. These depressions are called potholes, glacial potholes, kettles, or kettle lakes. They fill with water in the spring, creating wetlands, which range in duration from temporary to semi-permanent. The region covers an area of about 800,000 sq. km and expands across three Canadian provinces (Saskatchewan, Manitoba, and Alberta) and five U.S. states (Minnesota, Iowa, North and South Dakota, and Montana). The hydrology of the wetlands is variable, which results in long term productivity and biodiversity. The PPR is a prime spot during breeding and nesting season for millions of migrating waterfowl. [Wikipedia]”

27…..  due to the rather low temporal resolution of 12 days over the PPR.

#  You make this point later on in the text (lines 97, 155, 451 & 473) but the reader is left unsure as to whether this data is sampled 12 days apart, or averaged over that interval

90 attention, as co and cross-polarised data

# Unsure what 'co' means.  Aha! found it on the web - does it make sense? :- "Co-polarization is the antenna's radiation in your desired directions. Whereas cross-polarization is the antenna's radiation in the unwanted directions, i.e the cross-polar is basically considered as a dissipation in antenna radiation."

184: composites of the images are shown in Appendix A (Fig. A1).

# Fig. A1 is lonely in an appendix, but informative - I recommend your replanting it about here; there's enough room for it in this medium sized paper

225: ……   D = p2|μ1 −μ2| _2 1 +_2 2 ,.…….. (1)

# For easier readability, please indent your 7 equations at both ends

313  water extent (Table 1).

# Too far ahead on page 16

Figure 5. Map of predicted p(W). Scales are in UTM (zone 14) coordinates.

# Please tell us a bit more about this interesting figure in the caption which should be expanded. The caption should redefine the acronyms and symbols (p(W) and UTM).

This suggestion applies to all figure captions. It helps the first quick scan for the tentative reader.  I had to make a list of acronyms before I started reading through critically, as I can't retain them in all my head!

329: … HAND….

# How do you measure HAND, and to what precision?  I see that you mention it in subsection 2.2.2 Topographical data, but you don't elaborate.

Figure 7. Backscatter in a) VV and b) VH polarisation and derived water bodies on 12 October 2019

# That's pretty smart

365 Fig.8a

# Please rearrange Figs 8 & 9 closer to first mention - they are up to 2 pages distant - I have to split the document to follow text.  Not good if it's a printed copy ...

398: Fig. 10 

# 3 pages ahead ... and I have to expand the figure to 300% to find the tiny yellow patches; can't you take a small clip and park it on the empty space like I've done?

Fig. 9

# The caption and label are in hectares, the image legends are in sq m!! That’s confusing - please fix

447-460

# That last paragraph is a good summary

475 … programme

# there is a choice here – what about ‘program’?

'In American English, program is the correct spelling. In Australian and Canadian English, program is the more common spelling. In British English, programme is the preferred spelling, although program is often used in computing contexts.' [Grammarly]

476  Author contributions

# I like this list of the contributions of the team.  Nice job.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

The manuscript entitled “Monitoring surface water dynamics in the Prairie Pothole region using Dual-Polarized Sentinel-1 SAR time series” by Schlaffer et al. developed a new approach for classifying open water extent dynamic. The manuscript is novel and well organized. Most sections are informative, which can also lead to a long description and thus need to be significantly improved before acceptance. In the revised version, authors need to revise the writing style of this manuscript.  

1. In this study, authors used many data sources to achieve their goals and these data are not in the same spatial resolution. For instance, the spatial resolution of Sentinel-1 data, the digital terrain model, and land cover are 20 m, 1m, and 30 m, respectively. To upscale the resolution from lower to higher is fine, while to downscale from lower to higher resolution can introduce uncertainty in data analysis, particularly for this study. In this study, land cover data (CDL) in 2015, which scaled from 30 m to 10 m, are used as reference data. Such a downscale can introduce potential uncertainty and should be discussed in the manuscript.
2. In this manuscript, for Eqn. (7), the author only included the regression model. However, for data transparency, authors should list the regression parameter in the supplementary material.    3.For section 3.2 surface water dynamic, the authors described their results based on the temporal dynamics, i.e., only describe the results change with time, which helps understand model accuracy. At the same time, the authors compared surface water dynamic with wet or dry conditions. In sum, this section is too long and not well-organized. I suggest authors can describe the temporal dynamics in the first section while discussing how surface dynamics change with hydrological conditions (wet or dry conditions). In this way, the readers can easily get the ideas that you want to express.
3. Research title is not specific. The current title can mislead readers that authors had mapped surface water dynamics in the whole prairie pothole region. Actually, this is a case study. Please revise your research title.
4. The abstract and introduction are informative, but they are too long. For instance, the authors discussed the many key points, I can easily lose while I am reading the paper. I strongly suggest authors shorten the abstract and introduction. In addition, the conclusion section is well-written. Please follow this style to revise your abstract.
5. In figures 8 and 9, the authors only compared the water body with wet or dry seasons. I recommend authors can plot the hydrography on figures 8 and 9. Authors can compare the relationship between flood and drought conditions. In this way, the relationship between hydrology and water bodies can be quantitively assessed. Authors can relate the hydrology data to the surface water body.
6. Line 89, please list the references here.
7. Here are some thoughts for authors’ reference. 1) In this study, authors used a high-resolution digital terrain model (1m). For the topography, there are many topographic indices, for instance, topographic wetness index (TWI) and others, which can represent the topographic information in detail. In my thoughts, the appearance of surface water is highly related to topography information. I suggest authors can try to relate water body location with TWI or other indices. This may help authors to validate your results. 2) Authors extracted the water body from Sentinel-1 data. To date, Google Earth has spatial resolution data. In authors’ data, authors validated the results using the same data source but for different locations. The validation is good. However, this data is not easy to obtain. Can you try to validate the data through Google Earth. I know this is beyond the scope of this study, but authors can try this in future studies.