Local perception and adaptation strategies to landslide occurrence in the Kivu catchment of Rwanda

Nema, Ma-Lyse; Saley Mahaman, Bachir; Diedhiou, Arona; Mugabe, Assiel

doi:https://doi.org/10.5194/nhess-2023-47

Preprints

https://doi.org/10.5194/nhess-2023-47

Preprints

30 Mar 2023

| 30 Mar 2023

Status: a revised version of this preprint is currently under review for the journal NHESS.

Local perception and adaptation strategies to landslide occurrence in the Kivu catchment of Rwanda

Ma-Lyse Nema, Bachir Saley Mahaman, Arona Diedhiou, and Assiel Mugabe

Abstract. This study aims to assess local perception and adaptation strategies to landslide occurrences in the Kivu catchment of Rwanda. Qualitative, quantitative, and combined methods were applied for data collection to investigate how the locals perceived the types of landslides, origins, impacts, contributing variables to landslides, and adaptation strategies to landslides. The investigation conducted interviews with 384 residents from the six districts of the study area, key informant interviews and field observations. The findings showed that falls, flows, slides and spreads are the types of landslides frequently occurring in the study area. Heavy rain, steep slope, road construction, inappropriate agriculture practices, deforestation, earthquake, and mining were found to cause landslides with effects such as human fatalities, infrastructure damage, injuries, and property losses. Different measures are adopted for landslide risk reduction, including agroforestry, terracing, stormwater drainage systems, and relocating people from high-risk areas. Residents have a positive opinion of their community’s approach to managing landslides effectively. However, the findings revealed gaps in cooperation between the parties where Non-Governmental Organizations do not appear to be active participants during intervention activities for landslide management in the study area. Regression analysis has shown that deforestation, inappropriate agricultural practices, roads construction, earthquakes, and climate change are the key factors significantly contributing to landslide occurrences in the study area. Further research must be conducted on the subject using a variety of methodologies, notably those related to applied artificial intelligence in order to enrich the literature presently available on landslides in the Kivu catchment of Rwanda.

Received: 25 Mar 2023 – Discussion started: 30 Mar 2023

Download & links

Preprint (PDF, 1471 KB)

Supplement (315 KB)

Download & links

Ma-Lyse Nema et al.

Status: final response (author comments only)

RC1:
'Comment on nhess-2023-47', Anonymous Referee #1, 05 Jun 2023
Overall evaluation: Considering the major gaps in the manuscript, the manuscript needs major revisions to be moved to the next step.

Specific comments:

Abstract: to be revised as it starts with objective of the study without indicating the issue, the problematic or why the study is to be conducted

Introduction: Literature review presented in this chapter is general and did not consider the published papers on landslides in the study area and in the region of study. The author is suggested to improve the quality of this section through reading the following papers published in the study area: (a) Arthur Depicker et al. 2021: Historical dynamics of landslide risk from population and forest-cover changes in the Kivu Rift: https://doi.org/10.1038/s41893-021-00757-9, (b) Uwihirwe et al. 2022. Integration of observed and model-derived groundwater levels in landslide threshold models in Rwanda. https://doi.org/10.5194/nhess-22-1723-2022, (c) Maki Mateso et al. Characteristics and causes of natural and human-induced landslides in a tropical mountainous region: the rift flank west of Lake Kivu (Democratic Republic of the Congo). https://doi.org/10.5194/nhess-23-643-2023 and (d) Uwihirwe 2022. Potential of satellite-derived hydro-meteorological information for landslide initiation thresholds in Rwanda. https://doi.org/10.5194/nhess-22-3641-2022

The introduction ends without indicating the main objective and specific objectives of the manuscript.

Methodology: The study area is not well described (comments in the PDF document). Possible mistakes of the Geographic coordinates of the study area. The map presented in Figure 1 to be revised

The title of the manuscript is clear that the study was conducted in lake Kivu catchment on Rwanda side. However, data collection, sampling sites selection are based on administrative boundaries which are different from catchment boundaries. Therefore, it is suggested to revise the methodology, aggregate data at catchment level. Therefore, this will impact the presentation and discussion of results.

Results presentation and discussion: Results are presented but not deeply discussed.

Conclusion/recommendation: haphard conclusion
Citation: https://doi.org/10.5194/nhess-2023-47-RC1
- AC1: 'Reply on RC1', Ma-Lyse Nema, 16 Jun 2023
  
  Thank you for your thoughtful remarks; we will take them into account when we revise this paper. Every component of this document will be specifically improved in accordance.
  Please allow me to provide clarifications regarding the map and sampling:
  Despite the fact that the provided map was based on administrative boundaries, the data collection and sampling locations were based on catchment, which was not completely delimited due to shapefile issues, but taking into account what a catchment was in the sense of its hydrological network diverting into the Kivu Lake (I was familiar with the area because it is where I was born). Since each of the six districts is part of that catchment area, the sampling was done taking into account the sites that were located in the areas that were being drained. Therefore, as long as the data were gathered using the proper approach, there are no restrictions on their trustworthiness.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-AC1
CC1:
'Comment on nhess-2023-47', Blaise Mafuko Nyandwi, 10 Jun 2023

Assessing the perceived risk of hazards generally involves evaluating two aspects: severity and likelihood of occurrence. However, in this article it is not clear whether these concepts have been assessed or not. Can you explain how these two aspects have been taken into account in assessing the perception of risk?
In your article you try to explain the perception of lay people on the basis of the local geological context, i.e. volcanism and earthquakes, but this doesn't seem to me to be well explained (see line 200 onwards). Is it possible to clarify this on the basis of the supported literature?

Citation: https://doi.org/10.5194/nhess-2023-47-CC1
- AC2:
  'Reply on CC1', Ma-Lyse Nema, 21 Jun 2023
  
  Thank you for your good and valuable remarks on this article.
  I completely agree with you that none of the two aspects (severity and likelihood of occurrence) was clarified nor talked about in this paper. Thankfully, we have examined them, and the actual data and results are available, allowing them to be presented appropriately and discussed with support from credible literature just like they were proposed by other researchers on this forum. Additionally, throughout the revision, perception of local community based on the regional geological background will also be explained appropriately with reference to literature.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-AC2
  - CC5: 'Reply on AC2', Blaise Mafuko Nyandwi, 21 Jun 2023
    
    Good
    Maybe this reference could be useful for you
    https://nhess.copernicus.org/articles/23/933/2023/nhess-23-933-2023.html
    
    Citation: https://doi.org/10.5194/nhess-2023-47-CC5
RC2:
'Comment on nhess-2023-47', Anonymous Referee #2, 14 Jun 2023

This study talks about the landslide occurrences in the Kivu catchment of Rwanda, with the aspect of local perception and adaptation strategies.
I was wondering what is the real contribution of this paper in such kind of journal. As there is a big amount of (statistical) information. Since there are many references to the adaptation strategies to landslide occurrences, it seems very strange that no modeling or other analysis is being explained.
There is no introduction indicating clearly the objectives of the paper. Description of the study area seems to be even very short (no mentions about thickness, soil type, inclination and so on are done). Moreover, in my opinion, most of the total percentages reported in the document are completely useless for finding any adaptation strategies for landslide occurrence. Are they really relevant?
The (statistical) results presented seem to make this manuscript more likely a technical note.

Citation: https://doi.org/10.5194/nhess-2023-47-RC2
- CC4: 'Reply on RC2', Assiel MUGABE, 20 Jun 2023
  
  I appreciate you for taking part in this discussion and offering wise suggestions that will help us revise this manuscript.
  -This paper fits under the NHESS scope “The design, implementation, and critical evaluation of mitigation and adaptation strategies to reduce the impact of hazardous natural events on human-made structures and infrastructure, to reduce vulnerability and to increase resilience of individuals and societies”, and contributes to the NHESS in the form of fresh data on how local communities, policymakers, interact to set strategies for natural disaster risk reduction in a catchment that was not previously studied on that subject .
  -There is no way to compare this manuscript to a technical report. This research work is supported by clear questions, a sound methodology, and useful findings that add to the body of knowledge. However, just as in other scientific works, there may be certain gaps. As a result, everyone's contribution to this platform is essential to helping the author and co-authors transform the "half-cooked status" of their work into a "tasty product." As evidenced by the fact that this study was accepted for review in this journal and given access to this network of researchers, it provides strong discussion points that can be used to fix any flaws and meet publication standards for NHESS.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-CC4
- AC3: 'Reply on RC2', Ma-Lyse Nema, 07 Aug 2023
  
  I acknowledge you for participating in this discussion and providing insightful comments that will aid in the revision of this manuscript
  This study contributes to the NHESS by providing new information on how local communities and policymakers interact to establish strategies for reducing natural disaster in a study area that hasn't been previously researched on the topic. The study falls under the NHESS scope "The design, implementation, and critical evaluation of mitigation and adaptation strategies to reduce the impact of hazardous natural events on human-made structures and infrastructure, to reduce vulnerability and to increase resilience of individuals and societies." However, some gaps were highlighted including the introduction which misses some relevant literature and objectives, description of the study area with complete information as well as strong literatures on the study area to support the discussion of the findings. It is evident that we will take them into account in revising the manuscript to take it to NHESS standards. In addition to that, we will improve the way of presenting the results by adding all related missing findings as well as clarifying adaptation strategies for landslide occurrence, and this is possible because we have already a rich database.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-AC3
RC3:
'Comment on nhess-2023-47', Ugur Ozturk, 14 Jun 2023
Referring to the interviews, the authors evaluate the local perception and adaptation strategies to landslides along Kivu Lake in Rwanda. They provided valuable information regarding how locals see the landslide threat and what should be done. As suggested by Gill et al. (2021)*, "understanding and listening to stakeholders" is a crucial aspect of addressing natural hazard threats adequately and building sustainable and resilient communities; the study makes an important contribution not only relevant to their research area but also for the global landslide risk community.
In places, the tone of the manuscript misinforms the reader through people's perceptions to sound like reality. Authors should pay attention to keeping the perceptions and facts separate. Below I provide all my major comments that need to be addressed to elevate the quality of the research work to the NHESS standards. I also provided several minor concerns in my attached document. Please refer to them for my line-by-line suggestions.
The introduction inefficiently leads to research questions and the manuscript's goals. Maybe more than half of the information is irrelevant to the study goals. I recommend trimming this section considerably.

A major relevance of this study is the collected data (e.g., questionaries and interviews); hence I strongly encourage the authors to provide the collected data as supplementary information. Providing the data will not only increase the credibility of the current work but also will attract more attention to the manuscript.

At the end of the results section, section authors highlight the main landslide-controlling factors from the locals' perspective. They should extend this section and compare the perceptions to reality (referring to literature, I provided several options in my detailed comments) in a dedicated discussion section. I believe this is one of the most exciting outcomes of their study.

In a dedicated discussion section, authors should also highlight the relevance of their observations/results in the broader context. What new knowledge have they provided for the global landslide community? What are the main limitations of the landslide research in the study area? And what should be done next?

Authors should consider repeating their statistical assessment for different education groups (or considering other relevant divisions such as age) to show a contrast between them, if there is any. They should also link these perceptions to reality in the dedicated discussion section.

* Gill, J. C., Taylor, F. E., Duncan, M. J., Mohadjer, S., Budimir, M., Mdala, H., and Bukachi, V.: Invited perspectives: Building sustainable and resilient communities – recommended actions for natural hazard scientists, Nat. Hazards Earth Syst. Sci., 21, 187–202, https://doi.org/10.5194/nhess-21-187-2021, 2021.
Citation: https://doi.org/10.5194/nhess-2023-47-RC3
- CC6: 'Reply on RC3', Ma-Lyse Nema, 03 Jul 2023
  
  I appreciate your thoughtful suggestions and contributions to this effort, Ugur. Once more, I would want to thank you for the references. We are taking them into account to raise the caliber of this manuscript.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-CC6
- AC4: 'Reply on RC3', Ma-Lyse Nema, 07 Aug 2023
  
  Dear Ugur, thank you for your bright comments and contributions to our project. I want to express my gratitude again for the recommendations. They are being considered in order to improve the quality of this manuscript.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-AC4
CC2:
'Comment on nhess-2023-47', Olivier Dewitte, 15 Jun 2023
The reviewers have pointed out some concern about the lack of reference to recent literature on landslide studies in the region of interest; which leads to several issues in the analysis/discussion. Here I want to provide more information on the landslides in the region, hopefully it will help the review process.

Landslide triggering factors:
Analysis of hundreds of dated landslide events in the region over a period of more than 20 years shows that no landslide events could be triggered by earthquakes (Monsieurs et al., 2018; Dewitte et al., 2021). This does not discard the role of earthquakes in triggering landslides in the region but instead reminds us that the return period of earthquakes with a magnitude large enough to trigger slope instabilities can be much longer than a few decades (Delvaux et al., 2017); and hence not been seen by people. Their potential impact, rather localised compared to that of climatic drivers, can be inexistent during a narrow time window of observation (Dewitte et al., 2021; Depicker et al., 2021b). In it interesting to mention that a new landslide, when large enough, will always be accompany by a local shaking of the ground that can be perceived by people living in its close neighbourhood. Although such a landslide is not necessarily triggered by an earthquake, the perception of the people could go into that direction. This has been reported in a case study in the city of Bukavu located close to the study area (Mugaruka Bibentyo et al., 2017).
On the other hand, the link between rainfall seasonality and the triggering of landslides can be demonstrated at the regional level (Monsieurs et al., 2018; Dewitte et al., 2021).
Landslides can also occur due to rock weathering and regolith formation (Dille et al., 2019). In other words, the long-term evolution of these preconditioning drivers alone can explain that a slope can also fail without any apparent trigger. This implies that the many landslides that occur in isolation of other events must be interpreted with care in terms of origin.

Landslide and environmental changes:
With respect to the changing climate, the last IPCC report (IPCC, 2021) clearly points to the lack of information in this region (and in many other parts of the African continent in general) and hence to the difficulty to detect trends in the climate. This, added to the lack of time series of landslide events leads to the impossibility to currently draw any conclusion about the role of climate change on the occurrence of landslides in Rwanda.
On the other hand, combining information from landslide inventory (thousands of landslide mapped from Google Earth imagery) and forest cover change (from existing remote sensing products from Landsat and Sentinel-2), the role of deforestation on the increase of landslide occurrence could be demonstrated, the increase lasting about 10-15 years (Depicker et al., 2021a). Then, using this information, with respect to historical forest change (derived from historical aerial photographs), about 40 landslide susceptibility scenarios were reconstructed, which allowed to assess long term changing hazard and risk in the region (Depicker et al., 2021b).
The role of road and mining activities on landslide occurrence in Rwanda is also evident (Dewitte et al., 2021). That said, it has not yet been quantified in the same way as for the other side of the lake (Maki Mateso et al., 2023).
The role of agricultural terraces is worth to be mentioned and research on the topic is currently being done. First results show that, in some cases, terracing leads to more landslides in the region (Sibomana et al., 2023).

Regional landslide susceptibility:
Relying on the landside inventories, a robust and reliable regional-scale landslide susceptibility model that covers the study area has been produced (Depicker et al., 2020). This model, such as the study by Dewitte et al. (2021), shows that the Virunga Volcanic Province is a region where the potential of landsliding is relatively low (less seismic activity, less weathering, forest cover).

Overall, this literature on the landslide in Rwanda, show that the discussion proposed in this manuscript is not correct in many points.

Availability of the data:
Several of these mapped landslide inventories used in the cited literature are available online (in total, about 4000 landslides mapped in Rwanda):

https://zenodo.org/record/5027004       shallow landslide inventory (with their date of occurrence) from Depicker et al. (2021a)

https://zenodo.org/record/7198322       landslides events from Deijns et al. (2022)

https://maps.nccs.nasa.gov/arcgis/apps/MapAndAppGallery/index.html?appid=574f26408683485799d02e857e5d9521 The data from Monsieurs et al. (2018) have be inserted and made available in that global catalogue by NASA.

The regional landslide susceptibility model (Depicker et al., 2020) can be found here online: https://zenodo.org/record/5524161

Reference list (* = work that concerns Rwanda directly )
*Deijns, A.A.J., Dewitte, O., Thiery, W., D’Oreye, N., Malet, J.-P., Kervyn, F., 2022. Timing landslide and flash flood events from SAR satellite: a new method illustrated in African cloud-covered tropical environments. Natural Hazards and Earth System Sciences 22, 3679–3700. doi:https://doi.org/10.5194/nhess-22-3679-2022
*Delvaux, D., Mulumba, J.-L., Sebagenzi, M.N.S., Fiama Bondo, S., Kervyn, F., Havenith, H.-B., 2017. Seismic hazard assessment of the Kivu rift segment based on a new sismotectonic zonation model (western branch, East African Rift system). Journal of African Earth Sciences 134, 831–855. doi:10.1016/j.jafrearsci.2016.10.004

*Depicker, A., Govers, G., Jacobs, L., Campforts, B., Uwihirwe, J., Dewitte, O., 2021a. Interactions between deforestation, landscape rejuvenation, and shallow landslides in the North Tanganyika-Kivu rift region, Africa. Earth Surface Dynamics 9, 445–462. doi:10.5194/esurf-9-445-2021
*Depicker, A., Jacobs, L., Delvaux, D., Havenith, H.-B., Maki Mateso, J.-C., Govers, G., Dewitte, O., 2020. The added value of a regional landslide susceptibility assessment: the western branch of the East African Rift. Geomorphology 353, 106886. doi:10.1016/j.geomorph.2019.106886
*Depicker, A., Jacobs, L., Mboga, N., Smets, B., Van Rompaey, A., Lennert, M., Wolff, E., Kervyn, F., Michellier, C., Dewitte, O., Govers, G., 2021b. Historical dynamics of landslide risk from population and forest-cover changes in the Kivu Rift. Nature Sustainability 4, 965–974. doi:10.1038/s41893-021-00757-9
*Dewitte, O., Dille, A., Depicker, A., Kubwimana, D., Maki Mateso, J.C., Mugaruka Bibentyo, T., Uwihirwe, J., Monsieurs, E., 2021. Constraining landslide timing in a data-scarce context: from recent to very old processes in the tropical environment of the North Tanganyika-Kivu Rift region. Landslides 18, 161–177. doi:10.1007/s10346-020-01452-0
Dille, A., Kervyn, F., Mugaruka Bibentyo, T., Delvaux, D., Bamulezi Ganza, G., Ilombe Mawe, G., Kalikone Buzera, C., Safari Makito, E., Moeyersons, J., Monsieurs, E., Nzolang, C., Smets, B., Kervyn, M., Dewitte, O., 2019. Causes and triggers of deep-seated hillslope instability in the tropics – insights from a 60-year record of Ikoma landslide (DR Congo). Geomorphology 345, 106835. doi:10.1016/j.geomorph.2019.106835
IPCC, 2021: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change[Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, In press, doi:10.1017/9781009157896.
Maki Mateso, J., Bielders, C.L., Monsieurs, E., Depicker, A., Smets, B., Tambala, T., Bagalwa Mateso, L., Dewitte, O., 2023. Natural and human-induced landslides in a tropical mountainous region: the Rift flanks west of Lake Kivu (DR Congo). Natural Hazards and Earth System Sciences 23, 643–666. doi:https://doi.org/10.5194/nhess-23-643-2023
Mugaruka Bibentyo, T., Kulimushi Matabaro, S., Muhindo, S.W., Dewitte, O., 2017. Anatomy of Nyakavogo landslide (Bukavu, DR Congo): interplay between natural and anthropogenic factors. Geo-Eco-Trop 41, 249–262.
*Monsieurs, E., Jacobs, L., Michellier, C., Basimike Tchangaboba, J., Ganza, G.B., Kervyn, F., Maki Mateso, J.-C., Mugaruka Bibentyo, T., Kalikone Buzera, C., Nahimana, L., Ndayisenga, A., Nkurunziza, P., Thiery, W., Demoulin, A., Kervyn, M., Dewitte, O., 2018. Landslide inventory for hazard assessment in a data-poor context: a regional-scale approach in a tropical African environment. Landslides 15, 2195–2209. doi:10.1007/s10346-018-1008-y
*Sibomana, P., Vanmaercke, M., Nahayo, D., Depicker, A., Tychon, B., Hubert, A., Rukundo, E. and Dewitte, O., 2023. Potential effect of agricultural terraces on landslide occurrence: the tropical mountains of Rwanda (No. EGU23-9126). Copernicus Meetings.
Citation: https://doi.org/10.5194/nhess-2023-47-CC2
- CC3: 'Reply on CC2', Assiel MUGABE, 16 Jun 2023
  
  Thank you Olivier for helpful your important contribution in this discussion. It's true that lack of relevant information on recent studies carried out in the study area have affected the results discussion in our work.When we were working on discussion part, we found out the potential existing papers which could help us were covering the whole area of Kivu ( side of Rwanda and Congo) ; so we wondered the information from those sources were not puting particular emphasis on our study area( Kivu catchment of Rwanda) considering that Kivu lake could be a barrier that may have an implication on differences in geologic and hydroclimatic aspects on both sides. I am sure the recent documents you provided will help us to make a good review in order to bring the manuscript up to the good standard.
  Another information to provide is that: This manuscript comes from the first study among three which are being conducted by the author(Miss Ma-Lyse Nema) for her PhD thesis entitled "Characterization, modelling and risk evaluation of landslides occurrences in the Kivu catchment of Rwanda". The other two studies will be based on landslides modeling ( inventory, factors). So, this is a good experience to figure out on this platform and exchanging experience with other scientists.
  
  Citation: https://doi.org/10.5194/nhess-2023-47-CC3

Ma-Lyse Nema et al.

Supplement

https://doi.org/10.5194/nhess-2023-47-supplement

Ma-Lyse Nema et al.

Viewed

Total article views: 804 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
612	154	38	804	32	11	9

HTML: 612
PDF: 154
XML: 38
Total: 804
Supplement: 32
BibTeX: 11
EndNote: 9

Views and downloads (calculated since 30 Mar 2023)

Month	HTML	PDF	XML	Total
Mar 2023	57	8	4	69
Apr 2023	160	31	0	191
May 2023	64	17	4	85
Jun 2023	195	41	22	258
Jul 2023	60	16	3	79
Aug 2023	47	24	5	76
Sep 2023	29	17	0	46

Cumulative views and downloads (calculated since 30 Mar 2023)

Month	HTML	PDF	XML	Total
Mar 2023	57	8	4	69
Apr 2023	160	31	0	191
May 2023	64	17	4	85
Jun 2023	195	41	22	258
Jul 2023	60	16	3	79
Aug 2023	47	24	5	76
Sep 2023	29	17	0	46

Viewed (geographical distribution)

Total article views: 782 (including HTML, PDF, and XML) Thereof 782 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 28 Sep 2023

Download

Preprint (1471 KB)
Metadata XML

Short summary

My early experience inspired me to write this paper because I was always curious about the reasons behind the frequent landslides that occurred in the area where I was born. Now, my dream has come true because this study was centered on the same region, same people, and because I discovered the causes and preventative measures for landslides in my area. I hope that when establishing policies for disaster management in the study area, decision-makers will take these results into consideration.


Total:	0
HTML:	0
PDF:	0
XML:	0