Comment on esurf-2021-58 Anonymous Referee # 2 Referee comment on " Identification of typical eco-hydrological behaviours using InSAR allows landscape-scale mapping of peatland condition

That said, I have some reservations about the manuscript in its current form. The methods section is very difficult to follow and could benefit from some rewriting, possibly with some parts of the results section relocated there (see specific comment on L284-314 and 370-390 below). Additionally, throughout the paper, there seems to be an effort to use the technical terms for measured factors (e.g. multiannual average velocity). Given that these terms are defined (albeit quite late on in some cases), it would make it far more understandable to the reader in some cases to use the definitions in text rather than expecting the reader to constantly refer back (or in some cases forward) to the definitions. For example, L334-335 “Where bare peat is dominant, the most negative velocities occur.” Could be written as “Where bare peat was dominant, the greatest peat subsidence occurred.”

That said, I have some reservations about the manuscript in its current form. The methods section is very difficult to follow and could benefit from some rewriting, possibly with some parts of the results section relocated there (see specific comment on L284-314 and 370-390 below). Additionally, throughout the paper, there seems to be an effort to use the technical terms for measured factors (e.g. multiannual average velocity). Given that these terms are defined (albeit quite late on in some cases), it would make it far more understandable to the reader in some cases to use the definitions in text rather than expecting the reader to constantly refer back (or in some cases forward) to the definitions. For example, L334-335 "Where bare peat is dominant, the most negative velocities occur." Could be written as "Where bare peat was dominant, the greatest peat subsidence occurred." I also have two concerns relating to the methodology and interpretation of results: This study used four and a bit years of data (see below to comments on actual length) yet only really considered data from 3 years due to 2018 being classed as a drought year. If data had been taken over a different period (e.g. 2017 to 2021), I am not sure that 2018 would have been excluded quite so readily as May 2020 was hot and dry and April 2021 was very dry but cold. Therefore, particularly in the context of climate change and dry summers likely to become more prevalent in the UK and parts of Europe, I would like to see much greater consideration of the 2018 drought otherwise I have doubts that this method would prove particularly useful for regions where the climate is changing (and see more specific comments below). Whilst I understand that pool systems will (should) always have Sphagnum and are easy to identify, I question whether assessing the predictive accuracy using just one PFT and hydrology cluster is wise (L255-264 and 380-394). Surely this is the category that is most likely to be accurate anyway, given that pools will mainly be on similar topography and slope. Why was predictive accuracy not verified by ground-truthing, since field visits are mentioned elsewhere, across a range of clusters? Also, to claim that "[Sphagnum pools] correspond to the part of peatland systems most unequivocally associated with 'near-natural' ecohydrological condition" is possibly a dangerous assumption. Whilst this may be true for the bogs that you included in your study, it is not necessarily the case for all bogs across the world, nor even across the UK (there is quite a bit of peat formed from sedges for example). Relying on this assumption could seriously limit the applicability of your method of peatland mapping and condition assessment for global use, as you have indicated you think the method could be used for in both the abstract and discussion, and therefore I feel that further validation is necessary.

Specific comments:
Figure 1 -This figure would make much more sense if b-f were given later in the paper after they had been explained. The z axis (velocity) is unclear as to which end contains negative numbers and which positive. What exactly does "frequency" (inset) mean? Is this number of obtained measurements? For the explanation of coloured dots, more clarity is needed. Why are there two different colours for front and back? Front and back of what? The graph? Also, why do these graphs only use one year of data when there is over 4 years data available?
L62-65 and L95-97 -these lines repeat the same information and define (some of) the same acronyms. Better to combine and only have once.
L101-104 -whilst this may all be fully understandable and sufficient detail to someone using the software, to someone who has not, it feels as though there are lacking details/explanations. Are these default settings? If not, why were those thresholds chosen?
L104 -presumably Wick Airport is not underlain by peat? Worth clarifying.
L107-109 -this is not a full sentence and it is unclear where it is going. Consider revising.
L118 -it says 4-year 5-month data here but the dates given in L94 do not add to this. Likewise Figure 2 legend starts 12 th March 2015 to 1 st July 2019 but text suggests June not July. Which is correct?
L118-121 -this is the only time that PCA channels are mentioned. Why are 10 calculated? Did you use them for anything else? If not, why are they mentioned? L122-128 -whilst I can see from Figures 2 and S1 that 2018 is different from the rest due to the drought (although not that much if just looking at the R2 line in Figure 2), I feel it would be worth giving more consideration to the results given that include 2018. Unfortunately droughts such as in 2018 are likely to become more frequent under climate change (e.g. May 2020 was very dry and warm, and then April 2021 was incredibly dry throughout much of the UK but it also happened to be very cold) and therefore if you want the model to truly represent peatland condition and apply to restoration endpoints, would drought events not need to be considered within the approach? It is also interesting that the drier bog merely showed greater amplitude in displacement rather than being completely out of pattern as the wet bog was. Either it needs to be specified that one legend relates to horizontal and the other vertical lines or each line needs to be given a more distinct colour, preferably the latter since specifying lines as "grey" (L131) in the legend does not improve clarity. Figure S1 -the legend states that EOFs 1-4 and 7-8 have periodicities greater than 12 months but it looks as though 7-8 may be less. However, it is hard to tell as the quality of the graphs is poor and the axis labels need to be larger. Figure 4 get mentioned before Figure 3? Also, the relevance of Figure 1 to the exclusion of the irregular time series needs further explanation.

L147 -why does
L188-190 -this sounds like vegetation is separate from PFT. Please clarify. Table S5 -this needs further explanation. Why is "100% -5/5ths" by itself and seemingly different from "up to 5/5ths"? Figure 4 -the legend for b) states the thin/modified peat areas are hatched but they appear just to be grey (using legend from 4a). I'm a little lost as to the usefulness of 4c, particularly as forested land is the only thing that is clearly distinguishable and yet forested land was not classified on the peat condition map in 4a. Either there needs to be more explanation in the legend/text about it, or it could be removed.
L284-314 and L370-390 -these lines are not really results. They are mainly methods and should thus be in that section. However, they explain what you did far more clearly than the majority of the current methods section. I would suggest either beginning the methods with this and then continuing to fill in more details after or using these lines as the basis for the methods and expanding each step out from this more simple explanation. There are also some details which are given the methods section (e.g. about HCA) that are then repeated in the results. You need to decide which is which and avoid so much repetition, which should make it much clearer.
L319-329 -it is very unclear as to what "monthly frequencies" refers to. Is this when the maximum peat height occurs or the number of up and down oscillations? Or something else?
L343-344 -I disagree. Whilst the first year appears to show strong linear relationships, the other two are far less convincing. As a minimum, I would suggest a linear regression on these data and providing R 2 values in the text to back up this statement.
L348-350 -again, I disagree. To me, it looks like the middle of the three years shown had the greatest amplitude. Making sweeping statements across only three data points (effectively each year is a point here) is not advisable. Also, what does this mean? That some years are wetter than others?
L355-357 -this is the first time I have understood exactly where this bimodality is coming from and what it means. It might be worth explaining this earlier in the text. L102 and 105 -the (54) and (55) feel like they should be references but the reference list is not numbered.
L105 -define/give external reference to SNAPHU. L108 (and some other mentions of this paper) -why is this "L. Alshammari,"? Referencing should be consistent. Check reference software input/settings. L110 -"Each motion time series was processed as follows, to quantify the specific peatland surface motion metrics." Would read better as "Each motion time series was processed to quantify the specific peatland surface motion metrics as follows." L114 -Same for "R. Hyndman" about referencing.
L152 -"in in" at end of line.
Section 2.2 and 2.3 appear to be a continuation of one another so would make more sense combined into one section. Supplement 1.2 -this is not referred to in the main text; should it be? It also refers to Figure S3 but should to S2.
L182 -what does (64) refer to? Table S1 and S3 -define PFG. Also define all letter in PFG column as not all are obvious (e.g. LoA; G in brackets). Table S2 and S4 -what is the purpose of Rank Name/Name when there is highlighting to give ranks and Hydro Name (presumably final class name) as given? Needs more explanation in the legend.
L258 - Figure 4d is not a map. Did you mean 4a? Section 3.2 -there needs to be consistency in the way variables are referred to. In the first paragraph of this section, all variables are given as words. Why change to capitals for a variety of different lengths for the rest of the section, especially since most variables are one, or maximum two, words long? This makes it look very much like these parts were written by different authors who had not discussed the approach. Also, there is no need to constantly redefine abbreviations in brackets -if it is done well once, that is sufficient. L321 -"September and or October". Is it "and" or "or"?