EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Assessing the impact of mass movements on alpine trails and huts using EO data

Florian Albrecht1, Daniel Hölbling1, Lorena Abad1, Zahra Dabiri1, Gerald Reischenböck2, Gabriela Scheierl3, Tobias Hipp3, Hannes Resch4, and Gernot Resch4
Florian Albrecht et al.
  • 1Salzburg University, Interfaculty Department of Geoinformatics - Z_GIS, Salzburg, Austria (
  • 2MJP Ziviltechniker GmbH, Gmunden, Austria
  • 3Deutscher Alpenverein e.V., Munich, Germany
  • 4Österreichischer Touristenklub, Vienna, Austria

The alpine infrastructure of trails and huts is an essential asset for summer tourism in the Austrian Alps. Every year, around five million people use the trail network for hiking and other mountaineering activities. Mass movements such as shallow landslides, debris flows and rockfalls cause significant damages to the alpine infrastructure and may block access to certain mountain areas for weeks or even months. Such damages require repair and increased maintenance activity or even rerouting of trails. Climate change will exacerbate the problem as more frequent and severe mass movements can be expected. Therefore, the Alpine associations have to take natural hazards into account for their trail and hut management.

A promising opportunity for assessing the impact of natural hazards on alpine infrastructure arises through the new generation of Earth observation (EO) satellites of the European Copernicus programme. The high spatial and temporal resolution allows the detection of mass movements with an impact on trails and huts.

Therefore, we initiated the project MontEO (The impact of mass movements on alpine trails and huts assessed by EO data) to investigate the opportunities for EO-based mass movement mapping and hazard impact assessment for alpine infrastructure. We start with a user requirements analysis that describes the demand for consistent and appropriate information on mass movements for alpine infrastructure management. We perform interviews with the Alpine associations and other relevant stakeholders. They help us to identify significant mass movements, their impact on the alpine infrastructure, and the actions that trail keepers and hut facility managers take to deal with the impacts. Based on this, we assess the suitability of EO-derived mass movement information for alpine infrastructure management, and define requirements for its production and delivery.

Based on the user requirements, we develop a multi-scale approach and combine optical and synthetic aperture radar (SAR) satellite data (e.g. Sentinel-1/2, Pléiades) to comprehensively map mass movements and to detect mass movement hotspots. Further, we integrate the EO-based mapping results with ancillary data for landslide susceptibility mapping, and for modelling and simulating rockfalls and debris flows. Finally, we analyse the network of trails and huts in relation to the obtained mass movement information and thereby assess their impact on alpine infrastructure, i.e. identify the trails and huts that are (potentially) affected by mass movements.

We demonstrate the concept and methods for three study areas in the Austrian Alps: Großarl and Kleinarl Valley in Salzburg, Karwendel in Tyrol, and the Salzkammergut in central  Austria. For these areas, we will create EO-based mass movement inventory maps, hotspot maps, and hazard impact maps. We validate our results in close collaboration with the users and analyse their usefulness for alpine infrastructure maintenance and management. The outcomes of MontEO will contribute to improved maintenance efficiency and will lead to a safer alpine infrastructure with an increased value for hikers, the tourism industry and the society.

How to cite: Albrecht, F., Hölbling, D., Abad, L., Dabiri, Z., Reischenböck, G., Scheierl, G., Hipp, T., Resch, H., and Resch, G.: Assessing the impact of mass movements on alpine trails and huts using EO data, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-21325,, 2020

Display materials

Display file

Comments on the display material

AC: Author Comment | CC: Community Comment | Report abuse

Display material version 1 – uploaded on 04 May 2020
  • CC1: Comment on EGU2020-21325, Joshua Jones, 07 May 2020

    Hi Florian,

    Looks like this a super interesting project!

    I was just wondering if you could give any insight into how stakeholders in these study regions actually manage / mitigate the risk posed to trails by mass-movements? Are any hard/soft engineering strategies employed to increase the stability of the most at-risk trails? Or are at-risk paths simply closed and re-routed? 

    I ask this as we have made some preliminary observations of how mass-movements in Nepal affect the trekking infrastructure there (paper here if you're interested: ). It is clear that in Nepal there are no efforts made to mitigate landslide-risk, with trails simply re-dug through unstable deposits, likely increasing future risk. So I wonder if there are any simple low-cost management techniques used in the Alps that could potentially be applied here.



    • CC2: Reply to CC1, Joshua Jones, 07 May 2020

      Oops, sorry, paper link didn't appear...


    • AC1: Reply to CC1, Florian Albrecht, 07 May 2020

      Hello Josh,

      thank you for your interesting comment. My answer has to remain vague because we only just started the discussion with trail keepers in Austria.

      They do maintenance on trails every year and are aware of the quality of their trail installations (small bridges, handrails etc.). They get notified of damages over the summer. And in general, they notice if installations have to be repaired/replaced more often. So this is how they get aware of susceptible trail segments.

      For some trails, they do stabilising installations. However, they told me of trails especially below glaciers that they had to give up and re-route. Obviously, it depends on the case.

      The Alpine clubs that are partners in our project joined because they are interested in getting a more comprehensive approach for managing the impact of mass movements.

      Interesting project that you are working on in Nepal! I definitely am interested your paper. Can you forward it tom me (best via email We probably can learn from each other’s work. We can keep in touch about the evolution of the research of each of us.

      Kind regards,


      • CC4: Reply to AC1, Joshua Jones, 08 May 2020

        Hi Florian,

        Thanks for your answer, sounds like they at least have some people responsible for management, which is more than they seem to have in Langtang, Nepal!

        Definitely keen to keep in touch about your work. I'll send an email now with a working link to our paper. 



  • AC2: Summarizing comments from session chat, Florian Albrecht, 08 May 2020

    Dear community members,

    this comment summarizes the session chat from Thursday, 7th of May. 

    Recently, my co-authors and me started the project MontEO. Therein, we investigate the opportunities for EO-based mass movement mapping and hazard impact assessment for alpine infrastructure. We discuss the ways of usage in management and maintenance of trails together with trail keepers of the Alpine clubs. This allows us to refine the concept behind MontEO, presented in the upper middle of our poster. We develop a multi-scale approach and combine optical and SAR satellite data (e.g. Sentinel-1/2, Pléiades) to comprehensively map mass movements and to identify (potentially) affected trails and huts for an assessment of the impact on alpine infrastructure. This process is complemented with fieldwork. Our approach is depicted in the lower middle of our poster. We demonstrate the concept and methods for four study areas in the Austrian Alps of Salzburg, Tyrol, and Upper Austria, and validation our results in a workshop with trail keepers.

    More information about our project is available on 


    The following questions were answered:

    Question 1: Which is the data fusion procedure between optical and SAR data? Which is the methodology to process optical and SAR data?

    Answer 1: We will use object-based image analysis classifcation that happens on optical data and the radiometric variables from SAR; with radiometric variables I mean coherence loss, for example.

    Questions 2, 4 and 5 about the relation between EO data resolution and size of detectable landslides:

    • Question 2: What is the size of the investigated landslides? I know well the situation of the mountain roads / trails landslides, in few sites they are small and you can recognize only with very high-res topography (e.g. lidar). An approach with Pléiades is enough?
    • Question 4: Is S-2 not too coarse?
    • Question 5: I was in general thinking of the capabilities of 10 m resolution - how large has the smallest landslide to be for detection?

    Answer for questions 2, 4, and 5: True, the resolution from Sentinel-2 is rather limited. However, we can detect landslides of a certain size. Pléiades helps us to get more detail in hot spot areas. We plan to do additional field investigations to get a more complete picture.

    Additional clarification for chat answer: Thank you for the question 5. We do not have a “precise” answer yet. However, we expect to be able to give the answer at the end of our project. Then we also will find out how valuable landslide information derived with our approach is for Alpine infrastructure management, i.e. whether it is a problem if we missed landslides that are only detectable with Lidar.

    Question 3: Do you aim to identify events that occurred, or also provide early-warning prior to events?

    Answer 3: No, no early warning is envisioned.

    Question 6: Where have you found the baseline layer with all the alpine trails?

    Answer 6: Open Streetmap (OSM) contains data that is used by Alpine clubs to manage their trails. In addition, we get trail data from the Alpine clubs.


    Kind regards,


  • CC3: Comment on EGU2020-21325, Doris Hermle, 08 May 2020

    Hi Florian, a really nice approach!
    I am interested in two things:
    i) the combination of the different multispectral images - how were they referenced to each other? And did you use the S-2 images always from the same orbit? How was there positional accuracy/quality?

    ii) what is the smallest possible area to detect the landslides or did you solely use S-1 for detection?


    Thank you,

    • AC3: Reply to CC3, Florian Albrecht, 15 May 2020

      Hello Doris,

      thank you for your comment!
      Concerning your first question: 
      As our project has just started, we have not implmented all of our approach yet. Currently we are selecting the S-2 images for analysis. I expect that we need to use S-2 images from the same orbit because the available geometric correction for S-2 images causes limitations of positional accuracy.

      concerning your second question:
      In our approach, we use various EO sources. We will use additional images of higher resolution than S-2 and we will do field mapping with the objective to arrive at a high completeness of mapping results. The trail keepers of the Alpine clubs will care more about the quality of the final result.
      So, I cannot answer your question now, but I think, with all the other sources, we are well-posed to find an answer which minimum size landslides need to have to be visible in S-2 images.

      Best regards,