InSAR-Informed In-Situ Monitoring for Deep-Seated Landslides: Insights from El Forn (Andorra)
Abstract. Monitoring deep-seated landslides via borehole instrumentation can be an expensive and labor-intensive task. This work focuses on assessing the fidelity of Interferometric Synthetic Aperture Radar (InSAR) as it relates to subsurface ground motion monitoring, as well as understanding uncertainty in modeling active landslide displacement for the case study of the in-situ monitored El Forn deep-seated landslide in Canillo, Andorra. We used the available Sentinel-1 data to create a velocity map from deformation time series from 2019–2021. We compared the performances of InSAR data from the recently launched European Ground Motion Service (EGMS) platform and the ASF On Demand InSAR processing tools in a time series comparison of displacement in the direction of landslide motion with in-situ borehole-based measurements from 2019–2021, suggesting that ground motion detected through InSAR can be used in tandem with field monitoring to provide optimal information with minimum in-situ deployment. While identification of active landslides may be possible via the use of the high-accuracy data processed through the EGMS platform, the intents and purposes of this work are in assessment of InSAR as a monitoring tool. Based on that, geospatial interpolation with statistical analysis was conducted to better understand the necessary number of in-situ observations needed to lower error on a remote-sensing recreation of ground motion over the entirety of a landslide, suggesting between 20–25 total observations provides the optimal normalized root mean squared error for an ordinarily-kriged model of the El Forn landslide surface.
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-212', Anonymous Referee #1, 25 Mar 2024
The paper focuses on the assessment of the capability of Interferometric Synthetic Aperture Radar (InSAR) for the monitoring of surface ground motion using two different platforms:i) the European Ground Motion Service (EGMS) platform and ii) the ASF On Demand InSAR processing tools. InSAR are also used for understanding uncertainty in modelling active landslide displacement. The case study is the instrumented El Forn deep-seated landslide in Canillo, Andorra. The purposes of this paper concern also the evaluation of InSAR as a monitoring tool and as a statistical analysis to understand the necessary number of in situ observations to reduce the error.
I have read the paper and I have some recommendations. While the topic of the paper is interesting (for instance optimization of in situ sensors), I found the content and the structure of the manuscript, in the current form, weak and with the necessity of deep modifications.
In its current form, the article needs strong modifications and major revisions.
Suggestions
I recommend including a more detailed geomorphological, geological and stratigraphical description of the area interested by the El Forn landslide. For a complete characterization of a landslide through InSAR data, it is essential to investigate the geologic context in more detail to understand the conditions under which it is developing.
I recommend extending this InSAR analysis to the entire landslide body and not just the Cal Ponet-Cal Borronet lobe sector. Because chapter 2.1 describes the presence of 12 scattered boreholes in the landslide body for monitoring that should be exploited as a real opportunity for comparison with the InSAR data. The greatest strength of satellite interferometry is the ability to monitor large areas, here authors have focused only on a very small sector of a very large landslide, missing the most important information provided by the InSAR data.
I recommend expanding the monitoring period of InSAR data, as the abstract specifies that Sentinel-1 data processed for 2019-2021 has been exploited, while Chapter 2.2 explains that interferograms from only a narrow time period between June and November 2019 were used. Again, this choice comes at the expense of one of the major strengths of the InSAR data, namely the possibility of providing long time series. Instead, focusing the analysis only to a 6-month time period and on a narrow area of the landslide appears as a serious limitation in the study. When analysing the behaviour of a landslide, it is a good practice to expand the analysis of the time series as much as possible in order to know as much information as possible.
I strongly recommend adding a chapter discussing the results before the conclusions. A chapter of discussion is essential for the explanation of the results and to understand applicability, advantages, and limitations of the proposed approach.
About the figures I suggest:
- Figure 1: a geographical localization is missing. North arrow between the large-scale image of the landslide and the focus are not in the same direction. I recommend pointing the north arrow upward. In addition, for the purpose of characterizing the study area please add the location of all 12 boreholes.
- Figure 2: the deformation map on the left has neither north arrow nor scale, also in the legend there is no explanation of what the coloured circles are. Moreover, the figure on the right is just a screen captured by EGMS: I recommend downloading the data and reshaping the figure (you can use the EGMS-stream application to download the data).
- Figure 4: it is not clear why the representation of the landslide is now rotated 90°. I recommend defining a direction for the landslide representation and using it for all the figures in the paper.
- Figure 5: the colour scale of the ordinary kriging results of various random samples are always different, this makes it so that an immediate visual comparison between the figures is not possible.
Citation: https://doi.org/10.5194/egusphere-2024-212-RC1 - AC1: 'Reply on RC1', Rachael Lau, 25 Apr 2024
-
RC2: 'Comment on egusphere-2024-212', Anonymous Referee #2, 28 Mar 2024
The paper presents an interesting work on the use of InSAR data for monitoring deep landslides and defining the minimum number of observations useful for ensuring a valid model of landslide evolution. Once data have been correlated with field measurements, the Authors focused on how they can be used to quantify the overall uncertainty in planning borehole placement and implementing in-situ monitoring.
The case study analyzed, located in the Andorran Pyrenees, is significant both in terms of geomorphological evolution and natural hazard, thus well suited for the purpose of the work. In the introduction section, the work is well presented and contextualized in the available literature; the purpose and scientific validity of the work are adequately pointed out. The methods and methodological flow are clear and well-explained. In my opinion, results are valid but should be better discussed to highlight the strengths of the work. In detail, the following points are suggested to be developed for publication.
- Although it is not the main focus of the paper, I believe that additional morphological and geological data of the study area should be provided. If available, the Authors should state in more detail the geological framework and the material involved in the landslide process, the geometry of the body, and the type of activity (e.g., periods of greatest acceleration). Additionally, if possible, it would be interesting to see the temporal evolution of some of the monitored data (e.g., displacements or piezometric level) and perhaps relate them to external factors such as rainfall. I think that providing a more extensive characterization of the landslide would allow an easier understanding of the data presented in the subsequent paragraphs, as well as the choices made; also, it will improve the quality of the paper. If the Authors do not consider it necessary, they may still better mention previous works carried out on the landslide body exploring these issues.
- In Section 3.2, I would suggest a better discussion of the choice of using a 6-month time interval to compare in situ observations and InSAR data. What are the reasons for this choice that do not allow for the analysis of annual trends? It could be the presence of snow cover or lack of monitoring data, either way, it needs to be specified and discussed in the text.
- Reading the paper, I found the lack of a general discussion highlighting the significance of the work, especially in terms of the use and applicability of the method. I think a "discussion section" could help to better emphasize the validity of the results obtained as well as the achievement of the goals. In my opinion, some scientific issues and questions that could be addressed in this section would be: What are the main advantages and limitations over other landslide monitoring methods? Can this method be applied in remote areas where there are no active monitoring systems? How can the method be improved and applied to implement an in situ monitoring network?
- Figures should be better developed to help reading and understanding of the work. I would like to suggest these modifications to the Authors:
- Figure 1: The orientation of the satellite images is not very clear; it would be better to point the north arrow upward to maintain the same visual angle in all the sketches presented. I find placing the EGMS point in this figure to be not very intuitive; I would move it to other subsequent figures and insert here the location of the 12 boreholes mentioned in the text. Also, a small legend with the meaning of the red and blue lines and geographic framing should be given. If there are any significant points such as villages or mountain peaks, I would suggest highlighting them in this introductory figure and use them as reference points in subsequent plots.
- Figure 2: unclear. I think it would be necessary to add reference points and represent the same observation area to have a better qualitative comparison of the two methods. If possible, I would implement the quality of the legend of the velocity values. Also, it is not clear what the blue line in Figure 2b means; a small legend would be needed. Are the red and green lines in Figure 2a the same as the blue and red lines in Figure 1? If so, maybe better to keep the same colors. What do the dots in Figure 2a represent?
- Figure 5: for clarity of representation, I would suggest always keeping the north arrow upward (same as Figures 1 and 2); if possible, add reference points (e.g., location of villages) and the graphical scale of representation (even just in the first plot in the upper left corner). The legend has no units and does not indicate what kind of values are plotted.
Citation: https://doi.org/10.5194/egusphere-2024-212-RC2 - AC2: 'Reply on RC2', Rachael Lau, 25 Apr 2024
Viewed
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 135 | 0 | 0 | 135 | 0 | 0 |
- HTML: 135
- PDF: 0
- XML: 0
- Total: 135
- BibTeX: 0
- EndNote: 0
Viewed (geographical distribution)
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1