Review of: 

Climatic control on seasonal variations of glacier surface velocity

By 

Nanni et al. for The Cryosphere, (Manuscript No.:  egusphere-2022-1035)

General Comment

Nanni et al. developed a processing protocol for the detection and assessment of short-term glacier velocity changes from satellite EO and applied the method in an interesting study on seasonal variations of glacier velocity in the Pamir mountains (2013-2020) and its connections with local climate. The primary source data for generating high spatial- and temporal-resolution time series of ice velocity are NASA/USGS Landsat 8 and Copernicus Sentinel-2 optical satellite imagery, which are processed using the open source COSI-corr software. The post-processing steps include an extensive filtering procedure, calibration and central flowline velocity extraction. In this way the authors analyzed tens of glaciers with different characteristics and generated detailed spatiotemporal velocity plots for investigating seasonal variations in ice velocity. 

Most of the glaciers exhibited an annual recurring speed-up in the spring/summer and roughly half a lesser pronounced additional speed-up in autumn/winter, separated by slower velocities in between. Looking in more detail the spatiotemporal pattern revealed an upstream migration over time for the spring/summer speed up and a downstream migration for the autumn/winter speed up in most cases. Using the long-term averaged (1933-1995) air temperature dataset from a nearby station the authors find a good correlation of the evolution (timing and location/elevation) of the speed-up events with the migration of the isotherms, while other factors seem to have lesser pronounced or no relation (slope, orientation, velocity). The authors further provide a convincing discussion on the role of the efficiency of the subglacial hydrologic system, varying during the year, as a main driver for the observed accelerations.

The topic of this paper, seasonal variations of glacier surface velocity and its climatological controls, is very interesting and relevant, in particular thanks also to recent advances in modern computing technology and increasing availability of satellite EO data. This paper by Nanni et al. is a well written, illustrated and referenced manuscript and a valuable and original contribution of interest for the glaciology and wider community. The authors give a good motivation for their work, a detailed description of their methods and results and provide a set of interesting observation and a thorough discussion. The outcome provides new insights on seasonal variability of glaciers and environmental drivers in particular relevant for glacier and climate research. Some specific comments, corrections and suggestions for improvements are provided below.

Specific Comments:

Pg  2 – Ln 25: of the same order noise: of the same order as the noise

Pg  2 – Ln 26: which limits: complicating

Pg 2 – Ln 30: 10-day glacier velocity changes over 7 years: I know what is meant, but found this notation somewhat confusing. Later on it is also mentioned that 20-day time steps are used. Consider rephrasing, also elsewhere.

Pg 2 – Ln 42: compared to noise: compared to the noise 

Pg 2 – Ln 43: velocity changes over 48 glaciers: velocity changes for 48 glaciers

Pg 2 – Ln 45: Both result from changes in meltwater input: Wording too strong, ‘both appear to result’

Pg 3 – Ln 50: future changes: future change

Pg 3 – Ln 58: poor spatial coverage: poor spatial and temporal coverage

Pg 3 – Ln 63:  Consider adding: T. Strozzi, A. Luckman, T. Murray, U. Wegmuller and C. L. Werner, "Glacier motion estimation using SAR offset-tracking procedures," in IEEE Transactions on Geoscience and Remote Sensing, vol. 40, no. 11, pp. 2384-2391, Nov. 2002, doi: 10.1109/TGRS.2002.805079.

Pg 3 – Ln 72-73: The references seem to be provided as examples of ice velocity derived from images with a large time interval (annual to multi-year time periods) to support the previous sentence “The time interval … of the measurement.”, but note that in at least some of these references (e.g. Rignot) this is not the case. In general for SAR imagery a shorter time span is advantageous due to better coherence.

Pg 4 – Ln 93-98: New generations…: Worth mentioning here is also the crucial role of systematic acquisition planning.

Pg 5 – Ln 112: recovered: retrieved

Pg 5 – Ln 117: meteorologic: climatic (as in your title) 

Pg 8 – Ln 153:  for the time periods above: mention the time period here again.

Pg 8 – Ln 163:  ESA Sentinel-2: Copernicus Sentinel-2

Pg 9 – Ln 187:  we used the quality assessment band: it is not mentioned how this band is used, presumably as to mask the flagged pixels. Please explain.

Pg 12 – Ln 244:  for each pixel of all measured velocities: for each pixel and of all measured velocities

Pg 13 – Ln 257-273: Was there any longer term velocity trend for any of the glaciers and did you remove this? If not, could this affect your results?

Pg 14 – Ln 289: changes in the velocity: changes in the velocity relative to the multi-annual mean

Pg 14 – Ln 290: spring 2013 to winter 2019-2020: if I read these graphs correctly, they seem to run from spring 2013 to winter 2020-2021 (end of 2020). Please check.

Pg 14 – Ln 292-293: Such changes are higher than those described in the study of Lambrecht et al. (2014): How much higher?

Pg 14 – Ln 299-300: likely because of the combination of Landsat 8 and Sentinel-2 data: it would perhaps be interesting to mention if you found any differences in measured velocities between L8 and S2 for the same time period.

Pg 17 – Ln 348: H765Glacier: H765 Glacier

Pg 19 – Ln 378: 20 glaciers: Please check, in Figure 6 and several other places in the manuscript you mention 24.

Pg 19 – Ln 386: created: created

Pg 23 – Ln 458: Figure 8c: Describe figures in the correct order (first 8a & 8b).

Pg 23/24 – Ln 464/465: there is no clear influence of the glacier velocity on the downglacier migration rate: 1) add reference to figure 8e; 2) maybe it is a trick of the eye, but to me there appears to be a negative correlation.

Pg 28 – Ln 538: describes: follows

Pg 30 – Ln 586: (-0.05 m.d-1, -18.25 m.year-1): why negative if you talk about accelerations?

Pg 32 – Ln 643/646: the links do not work

Figures:

The figures and additional video supplement are nice and very informative, some minor comments below:

Fig 1: Caption: “mountain range area”: “mountain range”

Fig 1: Caption: “m.d-1” : “m d-1” also elsewhere in manuscript.

Fig 1: Caption: linear scale going from black (no displacement) to white (fast displacement): the scale seems to go from yellowish white to blueish white, but not from black to white.

Fig 1: Caption: background elevation: background shaded relief

Fig 1: Caption: '(c): (c)

Figure 3: 3b) Add label for vertical axis

Figure 3: Consider making c & d the same size. 

Figure 4: The color scale mentions m.d-1, I assume this must be %

Figure 4: Add label to y-axes

Figure 4: Some numbers are partly obscured (see for example 4d x-axis)

Figure 4: 4f) W731 seems to show up to 5/6 speed-up events, are these real, could you elaborate?

Figure 5: Add label to y-axes

Figure 6: n=24: see previous point regarding comment on Pg 19-Ln 378.