Reducing landslide risk is a complex task, which requires knowledge of various environmental and social characteristics, the use of acceptable technical solutions and the collaboration of several actors. Scientists are among them, though usually limited attention is paid to understanding their constraints for participation in long-term landslide risk reduction projects. This article summarizes expert involvement in the twelve years long landslide investigations in the Rampac Grande community and illustrates that the voluntary commitment to the Sendai Partnerships for landslide disaster risk reduction could motivate research institutions to assume responsibility for community-centred landslide risk reduction. It also points out the possible negative effects of the expert–community contacts disruption due to the COVID-19 pandemic restrictions as one aspect increasing the community’s vulnerability to landslides. The study site represents a socially, culturally and geologically complex environment which limits applicable technical solutions for landslide risk reduction and demands a high level of community participation in all landslide risk reduction steps. Landslide surface movement monitoring and slope stability calculations show that the studied slopes are very close to failure. The detailed hazard assessment was combined with field investigations of household vulnerabilities to assess risk in the zone around the 2009 catastrophic landslide. Results show that high vulnerability, rather than very high hazard, is responsible for assigning houses to the high-risk classes. This finding points out the preferable direction of landslide risk reduction efforts.
Research Article
Landslide hazard assessment and risk reduction in the rural community of Rampac Grande, Cordillera Negra, Peru
https://doi.org/10.21203/rs.3.rs-3115964/v1
This work is licensed under a CC BY 4.0 License
Journal Publication
published 22 Dec, 2023
Read the published version in Environmental Earth Sciences →
You are reading this latest preprint version
landslide risk reduction
slope stability calculation
landslide monitoring
hazard
vulnerability
Landslides cause considerable damage to infrastructure and an increasing number of deaths globally (Froude & Petley, 2018), while mountains are well-recognized hotspots of their occurrence (Nadim et al., 2013). This calls for ongoing efforts for effective risk assessment and management preventing landslide disasters in the mountain regions. Landslide risk reduction (LRR) is a complex task involving measures preventing landslide initiations or the intensity of their harmful impact on the protected elements containing also social innovations aimed to reduce exposure or vulnerability (Hostettler et al., 2019). Moreover, it has been proved many times that the implementation of effective risk reduction measures requires the participation and collaboration of several actors including local inhabitants (Hostettler et al., 2019; Maes et al., 2018; Muñoz et al. 2016). Otherwise, the desired risk reduction measures may not be accepted and effectively implemented by their recipients among which belong also to remote mountain communities (Maskrey, 1989). Even when the risk reduction measures are successfully implemented, ever-changing social (e.g., local power shifts, the dominance of lay over scientific knowledge, urban development), environmental (e.g., changing climate or rock strength properties) or technical (e.g., the construction life cycle of mitigation measures) conditions may again increase the risk even to levels prior the risk reduction effort (Evans, et al., 2009). Therefore, risk reduction management should be applied as an ongoing process capable of identifying and reflecting social (including risk perception; Haller, 2010) and environmental changes which drive the risk (Neal, 1997). This is highly challenging for all involved actors (e.g., communities, and research institutes), which participation the landslide risk management needs to be supported by a national legal framework. All this has been acknowledged by the International Strategy for Disaster Reduction – International Consortium on Landslides (ISDR-ICL) Sendai Partnerships for reducing landslide disaster risk as one of the voluntary commitments contributing to the goals set during the Third United Nations World Conference on Disaster Risk Reduction (WCDRR, Sassa, 2015). The Partnerships encourage to reduction of vulnerabilities to landslides through research of the signatory institutions producing results, which otherwise would not be achieved (e.g., Stember et al., 2017; Vilímek et al., 2021).
This article documents the long-term effort to design, implement and maintain effective measures to increase resilience to landslides in the small rural community of Rampac Grande in the Cordillera Negra Mountains, Peru. It presents approaches to improve local landslide hazard and risk assessment and evaluates the effectiveness of the previously adopted risk reduction measures four years after the termination of the LRR project. It provides an overview of the community participation and perception of the performed mitigation measures as well as management of the project by external expert organizations.
1.1 Study area
The study area is located in the Cordillera Negra, on the NW slopes of the intermontane basin drained by the Santa River to the Pacific Ocean (Fig. 1). The graben structure of the valley originated during Early Pliocene and Holocene deforming Mesozoic and Tertiary sediments and volcanic rock infill. The Mesozoic basement rocks are represented by limestone, shale, mudstone, claystone, sandstone, and conglomerate (INGEMMET, 1995; Weston, 2008) covering the majority of the study area (Carhuaz and Parihuanca-Chulec-Pariatambo Formations in Fig. 1; Vargas et al. under review; Weston, 2008). The east side of the graben valley is bounded by the Cordillera Blanca normal fault with long-term vertical displacements near the study site smaller than 3 mm/y and no historical earthquakes recorded (Macharé et al., 2003; Košťák et al., 2002). Despite that, the site is located in a region with very high seismic hazards (Heras and Tavera, 2002) with earthquakes being capable of triggering thousands of landslides during a single event (Plafker et al., 1971).
The slopes around the Rampac Grande village are densely covered (26%) by different types of landslides (Fig. 1) varying from deep to shallow slides and debris flows. A number of these landslides were assigned as active during geomorphological and InSAR-based inventory mapping with the latter showing surface movement rates ranging from 10 to > 100 mm/a (Strozzi et al., 2018). In some cases, detailed deformation histories were studied showing ongoing movement activity or significant reactivation following the above-average precipitations during the 2008/2009 rainy season. A number of active landslides located on the banks of the Santa River or its tributaries point out that stream erosion is an important factor contributing to landslide development (Fig. 1). Debris flows occurrence is largely constrained to stream channels which are well acknowledged by the local community. One of the streams within its territory is called “Lloclla” (Fig. 1), which means “Frequent debris flows”, reflecting their frequent occurrences (personal communication).
Landslide occurrence in the study area is closely related to the annual precipitation pattern dominated by the rainy season (October to April) with average total precipitation of 556 mm for the city of Huaráz (26 km SE from the study site, Klimeš & Vilímek, 2011). The annual average temperature varies between 11°C and 16°C depending mainly on the altitude (Kasser et al., 1990) while diurnal amplitudes are much higher possibly reaching 40°C (Yoshikawa et al., 2020).
The Rampac Grande community has 1,036 inhabitants (according to the local health centre, 2016) with roughly half of them (48%) dwelling near the Santa River and the rest living in the area around the 2009 landslide (Fig. 1). The inhabitants predominantly work in agriculture or as unqualified labour. More than 60% of their houses have access to electricity and water. The entire community is accessible via unpaved, but well-maintained road crossing the renovated (in 2017) bridge over the Santa River. Public lighting covers a large part of the settlement.
1.2 Rampac Grande landslide and the community
April 25, 2009, Rampac Grande earth slide – earth flow was the reactivation of older landslides on the slope with an inclination between 15° and 25° formed by weathered argillite rocks and tuffs. The landslide material was classified as fine gravelly very coarse silt sediment with high content of secondary carbonates (20–57%) and in some cases also of smectite clay mineral (10% or 29%; Klimeš & Vilímek, 2011). The landslide H/L ratio of 0.444 (Fahrböschung of 24°) does not represent highly mobile failure (Scheidegger, 1973) though its main movement phase was represented by very rapid flow responsible for 5 victims. The repeated expert-based hazard assessment (Vilímek et al., 2016; Klimeš & Vilímek, 2011) correctly predicted the accommodation of the 2009 landslide through partial sliding below its scarp and accumulating new debris in its distal part. Furthermore, the observations identified ongoing erosion unloading toes of two partial landslides identified after the 2009 event along its major scarp (Fig. 2) possibly lowering their stability. Initial results of the movement monitoring of these two landslides suggested ongoing creep and further stressed their persisting, possibly high hazard (Klimeš et al., 2019).
The unexpectedly fast failure of the 2009 landslide upset the community seriously limiting its development for several years (2009–2017) as it was regarded as highly landslide-prone territory by the local administration and other state agencies (Klimeš et al., 2019). During the first years after the catastrophic event, the community was largely left without relevant support for the recovery process due to a lack of expert knowledge at regional administration offices and missing or inappropriate communication of gained scientific results produced by external expert groups (both national and international). Therefore, the community applied its own measures to reduce landslide risk, which included housing relocation to safer parts of their territory and limitation of external visitors (Klimeš et al., 2019). Although these measures were at least partly effective and supported by expert opinion (cf. housing relocation), other traditional landslide risk mitigation measures strongly contrasted with the expert recommendations. They were revealed during interviews with the local inhabitants, who repeatedly claimed, that they were waiting until the mountain will “heal” to re-settle even the most hazardous sites around and below the 2009 landslide. The “healing” process referred to a gradual process of vegetation growth and surface erosion gradually masking morphological evidence of the 2009 landslide providing the appearance of a normal landscape state (Vilímek et al., 2016).
1.3 Response of the scientific community to the Rampac Grande landslide
These developments motivated participative landslide risk reduction project executed by the affected community, regional administration, national research institute INAIGEM (Instituto Nacional de Investigación en Glaciares y Ecosistemas de Montaña) and foreign experts (Vargas et al. under review) with the last two actors initiating it. The project (2016–2017) identified and implemented landslide risk reduction measures (e.g., hazard zonation, movement monitoring of the most hazardous landslides), which were accepted and maintained by the community and in a short-term (one year after the LRR project termination, Klimeš et al., 2019) allowed the development of the irrigation infrastructure, which otherwise would not be possible due to high perceived landslide risk resulting into suspension of state funding of the village development (Klimeš et al., 2019). Nevertheless, the long-term effect of the LRR project is unsure especially with respect to the changing political (e.g., changes in community and local administration power structures) and environmental conditions (e.g., precipitation pattern, agricultural practices related to field irrigation) as well as future participation of external experts (e.g., INAIGEM).
2.1 Landslide movement monitoring
Reliable landslide movement velocity monitoring can provide crucial information for the prediction of landslide accelerations possibly resulting in failure (Federico et al., 2012). Nevertheless, obtaining such data requires high measurement precision, regular reading frequency and permanent control over all accuracy parameters of the monitoring process. Achieving such conditions at the study site is very difficult due to the natural as well as social environment. The latter is very sensitive to any permanent installations on the land belonging to the rural communities. It may be considered as a disruption of local power or spiritual relations causing refusal of such installations through violent action damaging the monitoring equipment (Huggel et al., 2020). Therefore, we decided to apply a less technically demanding monitoring solution of a portable tape extensometer (Klimeš, 2018), which requires field installation of only inexpensive permanent posts (Fig. 2). All field works were conducted in close collaboration with the community, which also accepted the obligation to guard the measurement points (Klimeš et al., 2019). On the other hand, this technique provides less precise measurements further biased by usually irregular reading intervals, measurement sensitivity to external factors (e.g., temperature, wind) and skills of the operator. Despite that, previous works clearly evidenced its applicability for landslide monitoring (Rechberger et al., 2021).
Experts from the INAIGEM research institute were trained in the correct measurement techniques (Geocon, 2016) minimizing the environmental disturbances by performing the readings during early mornings when the wind is usually calmer. The thermal expansion of the stainless-steel tape was corrected using the following formula:
$$l={l_0}.(1+\alpha (t - t'))$$
3.1
l – corrected length; l0 – measured length; α – coefficient of thermal expansion for steel tape equal to 11.5 10− 6; t – measured temperature, t' – calibration temperature (20°C). It was not possible to avoid irregular reading intervals especially due to the very limited field works possibilities during the 2020–2021 COVID pandemic. These shortcomings allowed us to interpret the results for the monitored period (2016–2020) only as 12-month moving averages considering 6 preceding and 6 following readings combining only readings from points located within the most hazardous parts of the partial landslides (cf. red points 8 to 10 on the west and 7´and 8´ on the east profile, Fig. 2).
2.2 Slope stability calculations
Calculations of the slope stability under different environmental scenarios aimed to specify how close to failure (cf. threshold value of factor of safety, Fs) the monitored slopes are (Fig. 2) and what are the governing slope stability conditions. The appropriate slope profiles for calculations were defined based on field mapping supported by the DEM derived from the 2016 drone flights and soil mechanical properties ascertained through analysis of two disturbed soil samples (for location see Fig. 2). The DEM was prepared from field photographs taken by INAIGEM using DJI FC550 camera mounted on UAV carrier and processed in AgiSoft without using ground control points. The set of photographs was taken with 70% overlap from an average altitude of 120 m. The dense point cloud was used to create a raster model with a 5 m horizontal resolution. The landslide material is represented by soils described as sandy silt with gravel or gravel clay whose critical strength was determined by direct shear tests of two reconstituted samples reaching 31.5° and 34°respectively. Their residual strength was estimated using empirical relationships with the particle size distribution (Roháč et al., 2020; Roháč, 2021) obtaining a range of 19.4° − 20.8°. Slope stability calculations were performed in FINE Geo 05 software using the Sarma and Bishop method, applied to the engineering geological model (Parry et al., 2018; Novotný 2014). The presence of superficial hard crust (Vilímek et al., 2016) on the slope surface along the east extensometric profile was not considered in the model since its thickness is only about 0.4 m and it is continuously fractured by the post-2009 landslide tension cracks (Fig. 2).
The slope stability was calculated under different water conditions and seismicity as well as under changed slope morphology reflecting the slope toe unloading due to water erosion. The soil wetness conditions were characterized by the pore pressure coefficient ru which represents the portion of water above the slip surface (e.g., ru=0.1 represents water saturation of 20–25% of the material above the slip surface). The effects of earthquakes were introduced as horizontal seismic acceleration coefficient kh, which is the ratio of horizontal acceleration to the gravitational acceleration constant. To assess realistic values of this coefficient we estimated the peak ground accelerations (PGA) at the study site for two historical earthquakes. The PGA of the Mw 7.9, May 31st 1970 Peru earthquake, which caused serious damage and triggered thousands of landslides in the region around the study site (Plafker et al., 1971), was estimated to reach 0.21 to 0.39 g of PGA (USGS 2020a). The shaking effects of the Mw 8.0, May 26th 2019 Loreto earthquake (Miranda et al., 2019) were characterized by 0.02 g PGA (USGS 2020b). Determining respective horizontal seismic acceleration coefficient kh with the estimated PGA values is a complex task (Jibson, 2012). Therefore, we applied a conservative approach considering the kh values to be equal to the estimated PGA values. The kh values used in the slope stability calculations (c.f., 0.05/0.1) are thus realistic with respect to historical events affecting the study site.
2.3 Landslide hazard assessment
The article focuses on the detailed hazard assessment of the two monitored landslides (Fig. 2) considering the failure of their entire estimated volumes. Their hazard was determined by the surface landslide movement rates with the threshold value of 20 mm y− 1 used to define low and medium hazard classes (Raetzo et al., 2002). The calculated Fs and estimated volumes of the landslides were used to further specify their hazard in terms of probability and run-out. The Fs was used as a relative measure of landslide occurrence probability assuming that the slope closer to the threshold stability value under dry conditions, has a higher probability to fail due to meteorological triggers. The volumes of the monitored landslides were calculated using UAV-derived DEM and raster elevation model representing expert-defined sliding planes characterizing maximum expected landslide depths. The volumes were compared with the volume assessment and known run-out of the 2013 minor landslide from the main scarp area (Fig. 2; Vilímek et al., 2016) to hypothesize the possible run-out distances of the monitored landslides. The typical pattern of the empirical relationship between travel angle and run-out distance (Crominas 1996) was applied.
2.4 Landslide risk assessment
The landslide risk assessment focused on the immediate vicinity of the 2009 landslide (Fig. 1) combining newly performed vulnerability assessment with the already existing hazard zoning (Klimeš et al., 2019). It generally follows the approach described in the EVAR manual (Assessment of Risks Caused by Natural Phenomena) used by all governmental agencies in Peru to assess and evaluate risk caused by various natural phenomena (CENEPRED, 2014). This manual suggests a set of parameters to assess different aspects of social or economic vulnerabilities providing quantitative weights of each descriptor used to quantify the vulnerability and risk levels. The vulnerability data was collected in the field (2019) describing social and economic vulnerabilities of each household to landslides (Table 1) and they were combined to define four vulnerability classes (Table 2). Each vulnerability class is defined by descriptors representing very high to low vulnerability.
The hazard zoning used for the risk assessment is largely based on previous geomorphological mapping (e.g., landslide identification, description of their activity) and expert assessment of areas possibly affected by landslides in the future (Klimeš et al., 2019).
| Exposure | Fragility | Resilience | ||
|---|---|---|---|---|
| Vulnerability descriptors | Social | Uninhabited houses were excluded from the analysis | Age groups | Beliefs of the population regarding the occurrence of the landslide |
| Degree of house occupation | Family response to landslides | |||
| Participation in trainings, emergency drills | ||||
| Economic | Use of the house | Technical conditions of the house (degree of preservation) | The economic activity of inhabitants | |
| Access to social programs | ||||
| Ownership of other houses |
| Risk | Hazard | Vulnerability |
|---|---|---|
| Very high | Active landslides; gullies | Age groups 0–5, 61–70 years; house occupation 75–100%; the belief that no more landslides will occur; unaware of safe places; no training or emergency drills are available; buildings are used as homes and businesses; buildings are in a poor state of conservation; low income and no access to social programs; they have only one house in R. G. |
| High | Temporarily inactive landslides; small landslides and debris flows | Age groups 12–17, 51–60 years; house occupation 50–75%; the belief that landslides could occur or that nothing can be done about the landslides; some family members know the safe places; participate in training and emergency drills when are asked to; buildings are used as homes and warehouses; buildings are in poor state of conservation; occasional economic income and access to some social programs; they live in rented house in R. G. or other place |
| Medium | High or medium slopes where landslides may occur | Age groups 30–50 years; house occupation 10–50%; belief that landslides will occur, but nothing can be done about them; family members know the safe places; they are interested to participate in training and emergency drills; buildings are used warehouses; buildings are in regular or good state of conservation; economic income from up to 2 activities and have access to more than one social program; live mostly elsewhere |
| Low | Flat areas with a very low probability of landslides or debris flow occurrences | Age groups 18–29 years; house occupation 0–10%; the belief that the local administration should execute mitigation measures, but fails to do so; family members know the safe places and are well organized; they are interested to participate in training and emergency drills; they do not use the buildings; buildings are in good to a very good state of conservation; they have permanent economic income; mostly live elsewhere |
The risk classes were defined by combining hazard and vulnerability according to the CENEPRED (2014) matrix. Each class is defined by a qualitative description of the respective hazard and vulnerability level (Table 2).
2.5 Evaluation of the effectiveness of the adopted mitigation measures and participation of experts on the LRR
Evaluation of how effective and useful the adopted LRR measures are from a mid- and long-term perspective is very important (Pedoth et al., 2019). Therefore, we performed a mid-term (c.f., four years after the LRR project completion) impact assessment on the community. The ex-post assessment of the implemented measures was performed following Mercer et al. (2009) to better understand community perspectives on the success or failure of individual measures. The evaluation criteria include sustainability (i.e., Could be the measure maintained or useful in a changing environment?); cost (i.e., Does the measure make use of available resources?); equitability (i.e., Will some community members benefit more from the measures?); and stability (i.e., Does the measure cause a minimum level of disruption in the ecology, social structure and livelihoods of the community?). The structured interviews with the community gatekeeper (cf. community president serving during the first year − 2016 - of the LRR project) and personal observations during the field visits in June and November 2021 were used to assess the impacts of the performed LRR measures. The 2021 visits allowed us to re-establish contact with the community, which was limited due to mobility restrictions caused by the COVID-19 pandemic (personal communication).
The landslide research of the Rampac Grande 2009 landslide was performed by a dedicated group of national and foreign scientists and while ongoing attention is paid to the approaches and methods for community participation, only limited attention has been dedicated to understanding motivations and institutional constraints of the expert’s participation in the long-term LRR. This is despite the fact, that experts contribute with key knowledge and technology to successfully reduce landslide risk and at the same time they are exposed to specific institutional settings and a number of funding schemes, which not always encourage long-term involvement in LRR. To evaluate issues the description of character, timing, duration and funding of the involvement of different groups of experts at the Rampac Grande site was prepared.
3.1 Extensometric measurements
Despite the care dedicated to the field measurements (e.g. training of the staff, selection of days with convenient weather), the displacement readings show high variability including several outliers, which were removed from the dataset. These are likely to be systematic errors resulting from variable weather conditions (strong wind) or caused by human factors (different operators). Nevertheless, the results show large similarities among both monitored landslides with total displacement of 31 mm (west-red) and 38 mm (east-blue, Figs. 2, 3) respectively giving average annual displacements of 7.7 mm and 9.5 mm. The east landslide is characterized by variable, mostly reversible movements, which from Marz 2019 changed to steady increasing displacements reaching 26 mm, representing an average annual displacement of about 13 mm. On the other hand, the west landslide went through a contraction phase (by about 10 mm) followed by displacement stagnation from December 2018. The acceleration on the west profile started in January 2020 with a total displacement of 24 mm. These accelerations observed on both landslides with different onset dates suggest an increasing hazard of possible sudden landslide movements.
Despite these differences, the general displacements identified by the extensometric measurements have similar trends suggesting, that their dynamics depend on more general, slope-wide conditions similar to both monitored sites. The described differences in the displacement dynamics of the monitored landslides (inset on Fig. 3) are probably caused by their contrasting structures. The east extensometric profile documents displacement of the landslide separated by narrow (up to 0.4 m) tension crack with sub-meter total vertical and horizontal offset. On the other hand, the terminal points of the west profile are located in the partly failed material (the partial landslide scarp between points 7 and 9 is about 8 m high, Fig. 3) with a number of tension cracks (between points 9 and 10 on Fig. 3), which are parallel with the side scarp of the 2009 landslide.
3.2 Slope stability calculations
Slope stability calculations showed very low Fs values when using residual strengths for the landslide soils. Such low values (between 0.58–0.62) cannot be achieved on temporarily stable slopes as those under investigation and are attributed to inappropriate values of the estimated residual strength for the studied slopes. It can be explained by the relatively low clay fraction of the soil (15% − 20%) and low plasticity index (Ip = 10,4–12,3), which may result in real values of residual strength being close to the critical values (Skempton 1986; Chandler 1984). Therefore, we considered the critical strength (Table 3) to be a realistic representation of the slope material properties as both slopes gained Fs above the threshold value using these strength parameters. When considering the critical strength values without cohesion forces, the slopes barely exceed the Fs threshold values. The west slope reached Fs of 1.01 only under the most stable environmental scenario (dry slope without seismic shaking with φcr = 31.4°, Table 3). On the other hand, the east slope was stable even when seismicity (kh=0.05) or water saturation (20%-25%) was considered (Table 3).
Based on the observed erosion of the west slope toe, we calculated its stability when removing part of its support. Results suggested that the toe erosion could place the slope into an unstable position (Fs = 0.94 with φcr = 31.4°) under conditions (dry slope without seismic shaking), which otherwise would result in Fs above the threshold.
Table 3 Calculation slope models of the two monitored landslides (Fig. 2) and results of the Fs calculations under different environmental slope stability scenarios (GWT – ground water table, ru - pore pressure coefficient, horizontal seismic acceleration coefficient kh) with cohesion set to 0 (*critical shear strength determined by the direct shear box tests).
| Shear strength φcr [°] | Dry slope | Dry slope | Dry slope | GWT on terrain surface | ru = 0.1 | ru = 0.1 | ru = 0.1 | |
| + | + | + | + | + | + | + | ||
| no seismicity | seismicity kh=0.1 | seismicity kh = 0.05 | no seismicity | no seismicity | seismicity kh = 0.1 | seismicity kh = 0.05 | ||
| East | 31.4* | 1.08 | 0.88 | 0.97 | 0.56 | 0.94 | 0.76 | 0.84 |
| 34* | 1.19 | 0.97 | 1.07 | 0.61 | 1.03 | 0.83 | 0.93 | |
| West | 31.4* | 1.01 | 0.82 | 0.91 | 0.52 | 0.87 | 0.70 | 0.78 |
| 34* | 1.11 | 0.90 | 1.00 | 0.57 | 0.96 | 0.77 | 0.86 | |
3.3 Landslide hazard and risk assessment
Landslide characteristics indicating hazard levels of the two monitored landslides as well as for the 2013 partial slide (Fig. 2) are in Table 4. The observed annual surface movements during the acceleration phase assigned the west, more disrupted landslide to the medium hazard class while the east landslide is classified to low hazard class. The observed creep movement along with slope stability close to the threshold Fs value even for dry slope (west landslide, Table 4) evidences a high probability of landslide initiation, which depends on the external (e.g., rainfalls) forces. Almost double the volume of the west landslide as compared with the historical 2013 landslide, suggests its potential for longer run-out as compared to the historical event in the case when its entire suggested volume will be mobilized at once. This is reflected in the hazard zoning (Fig. 4a), which assigns a high hazard class to the zone around the mapped accumulation of the 2009 and 2013 landslides.
| West/red slide | East/blue slide | 2013 slide | |
|---|---|---|---|
| Mean annual velocity (mm·y− 1) | > 24 | > 13 | NA |
| Hazard class | medium | low | NA |
| Fs under dry/dry with seismic effects#/full sat. conditions | 1.01/0.77/0.49 | 1.08/0.88/0.56 | 1.08/0.87/0.55 |
| Relative probability class | high | high | high |
| Max. depth (m) | 13 | 7 | 4* |
| Area (m2) | 3620 | 4658* | |
| Volume (m3) | 36000 | 8000 | 19000 |
| Run-out (m, expert assessment) | > 890 | < 890 | 890* |
| *Field observation; #seismic acceleration coefficient kh=0.1 | |||
The vulnerability assessment showed that 43% of houses around the 2009 landslide had been abandoned or are used less than 10% of the year, which significantly reduced the exposure of the community. None of the remaining houses was classified into a very high vulnerability class, less than 20% of houses are highly vulnerable, while the rest has the medium vulnerability. The differences between high and medium vulnerability classes are in economic income, access to social programs, knowledge of safe places and interest to participate in training and emergency drills as well as the state of the conservation of the buildings and their use (Table 2). The resulting landslide risk closely follows the vulnerability assessment with only four houses moving from medium vulnerability class to high-risk class where is 25% of the houses while the rest is in the medium-risk class (Fig. 4b).
3.4 Evaluation of the mid-term effectivity of the LRR measures
Interviews with the gatekeeper revealed that the community avoids new constructions in the very high-hazard zones (e.g., water tank construction, Klimeš et al., 2019) and if possible, relocates important infrastructure (cf., electric poles) into safer places. The 2021 expert´s visits also confirmed that all hazard zoning maps and landslide warning or evacuation routes signs within the Rampac Grande community are well maintained. On the other hand, extensometric measurements in June 2021 (half a year after the previous reading) found that two of the measuring posts on the east profile (points 1 and 2) were damaged, thus further monitoring will require their reparation. The exact reasons for the monitoring point’s destruction are not clear and were not known to the gatekeeper. Nevertheless, we assume, that this action reflects the opposition of about half of the community to landslide monitoring, as revealed in the 2021 interview.
| Implemented LRR measures | Effectivity criteria | |||
|---|---|---|---|---|
| Sustainability | Cost | Equitability | Stability | |
| Hazard zoning map | yes | yes | yes | rather yes - depends on power distribution within the community |
| Warning, evacuation signs | rather yes - signs are durable, but hazard maps require replacement (according to the state in 11/2021) | yes | yes | yes |
| Extensometric monitoring | rather yes - if protected from vandalism | rather yes - depends on the availability of INAIGEM institutional funding | no - only those within potential run-out of the landslides benefit | no - it has the potential to disrupt the social environment, it requires ongoing external intervention |
Evaluation of the effectivity of the implemented mitigation measures clearly identified that extensometric monitoring is the least effective measure to be implemented (Table 5). Both of the criteria which failed (equitability, stability) were considered as an issue already during the project implementation. They were regularly discussed with the community and its representatives to explain monitoring importance for hazard assessment and risk reduction. It can be speculated that the serious limitation of the communication between experts and the community due to the COVID-19 pandemic failed to keep the community agreement to this measure at the required level possibly contributing to its destruction between 12/2020 and 6/2021 (Fig. 5).
3.5 Expert´s collaboration on the Rampac Grande LRR
The involvement of different groups of external experts in the LRR at the Rampac Grande is documented in Fig. 5. The investigations of the foreign researchers (cf. Charles University Prague, CUP; The Czech Academy of Sciences, CAS) were supported by institutional funding or related research project and were largely driven by personal interest. Knowledge gathered during this phase of the expert involvement was used to prepare the LRR project led by the Peruvian research institute (INAIGEM - Instituto Nacional de Investigación en Glaciares y Ecosistemas de Montaña) and financed by the Czech Development Agency. Only this project (2016–2017) mobilized the necessary resources, social science experts and community participation to identify and implement effective measures to improve community resilience. The foreign experts (cf., CAS) were also involved, but funded by internal resources of their institution. This team of external experts continue investigations even after the 2016–2017 LRR project termination (e.g., landslide movement monitoring, field check of the implemented measures and collection of data necessary for slope stability calculation, Fig. 5). All these activities were supported by internal funding of respective institutions (INAIGEM, CAS). The vulnerability assessment was done by a group of students of the regional university (UNASAM - Universidad Nacional Santiago Antúnez de Mayolo, Huaráz), which included an expert participating in the 2016–2017 LRR project and who was personally motivated to continue research in this community.
The hazard assessment of the monitored landslides (Table 4) represents the worst-case scenario considering mobilization of the entire landslide volumes in a single event and ignoring the effects of additional slope stabilizing factors. They are most likely responsible for the fact that the partial landslides despite the very low calculated Fs values remained without major displacements since their origin in 2009 withstanding the number of rainy seasons during which soil saturation or groundwater levels most likely increased. It strongly suggests that the additional stabilization forces (e.g., suction and cementation; Bommer et al., 2002; Rotisciani et al., 2021) result in higher effective strengths, which were not included in the calculations. The key process is probably the soil suction, which reliable description requires detailed in-situ monitoring (Pirone et al., 2015). A possible explanation for its elevated values brings the results of volumetric water content monitoring performed on the west landslide from October 2016 to May 2017 representing the rainy season (for location see Fig. 2). It showed only partial saturation of the soils with several fast cycles (e.g., within 5 hours) of wetting and drying at the depth of 0.3 m, which was almost simultaneously reflected also at the depth of 0.6 m, but with a water content of one order smaller than the one observed during the peaks near the surface. It illustrates the difficult soil saturation by rainfalls, which along with possibly high surface run-off could result in lasting suction contributing to the slope stability. This observation further confirms the conclusion of Klimeš & Vilímek (2011), that the 2009 landslide occurred due to the extreme rainy period when deep-underground water inflow from the slope above the landslide scarp could also significantly contribute to soil saturation above the sliding surface. Another important factor of the temporary stability could be material cementation which is evidenced by the high carbonate content (20–37% of the total sample weight) of the 2009 landslide material (Vilímek et al., 2009). Quantitative assessment of the cementation effects on the slope stability is speculative, but considering laboratory results for the mixture of volcanic soils with lime and cement (Davila et al., 2020), we suggest that a 10% increase the cohesion could represent a conservative cementation effect on the slope stability helping the slope remain without major landslide movement even during rainy seasons.
The new information about landslide movements and slope stability conditions shows that the studied slopes are very close to failure (i.e., high failure probability), which initiation depends on hydrological conditions (e.g., groundwater table fluctuation) with largely unknown temporal dynamics. It can be approximated by assessing the landslide return period from relevant historical events recorded near the study site. They are represented by five landslides since 1966 (cf. 1966, 1970 two events, 2009 in Klimeš & Vilímek, 2011 and 2017 – personal communication), which results in 11 years return period. It is considered a “frequent” landslide occurrence by FOEN (2016) and underlines the high degree of the hazard, which landslides represent for the Rampac Grande community.
When evaluating the results of the risk assessment, it is important to consider that the spatial distribution of occupied and abandoned houses around the 2009 landslide (Fig. 5) reflects the LRR measure adopted by the community before the 2016–2017 project without the participation of external actors (Klimeš et al., 2019). It is represented mainly by housing relocation to safer places within the community territory, which agrees with the finding of Figueiredo et al. (2019), that selection of safe places to live in the main adaptation strategy of rural communities related to ancestral knowledge. The majority of all houses are located in the medium hazard zone with only seven in the high hazard zone. Despite that, 25% of the inhabited houses are still under high landslide risk even after the 2016–2017 LRR project. The project contributed to risk reduction by, lowering the vulnerability by disseminating scientific knowledge about the landslide hazard and providing guidelines for correct responses to possible future landslide occurrences (e.g., marking escape routes in case of a major event). Nevertheless, the damage to the extensometric poles may indicate the process of vulnerability increase through a gradual change in the community's perception of the implemented LRR measures with a possible shift from the expert-based measures towards more traditional practices (e.g., the “healing” mountain principle, see Chap. 1.2). This could be related to the power shift within the community increasing sensitivity to permanent installations, considering them as disruption of local natural and religious conditions (compare with Muñoz et al., 2016). We can only speculate that disruption of contact between Peruvian experts performing the measurements and the community, during the COVID-19 pandemic, played a negative role in this process, which may result in the landslide risk increase in the long term (Evans et al., 2009). This further stress the well-known fact that risk management is an ongoing process requiring repeated risk assessments. The focus should be on possible vulnerability changes and the effectiveness of the implemented mitigation measures as these are directly and relatively easily affected by human decisions and actions. The retrospective mid-term effectivity evaluation (Table 5) suggests that measures where some of the sustainability criteria were not fulfilled are highly prone to fail in the future, when the involvement of the external actors (e.g., research groups) may decline. It is worth to consider responsibility and technology transfer to more permanent actors (e.g., the Carhuaz Provincial Municipality) to continue with the LRR activities (e.g., monitoring). Such transfer requires a lot of effort and time, but if it succeeds, it is highly effective (Vilímek et al., 2020).
The research institutes, which took a major part in the LRR in the Rampac Grande (cf. CUP, CAS, INAIGEM) dedicated considerable financial and time resources coming from their internal institutional sources. It was made possible by the institutional internal processes and regulations which give certain freedom in the allocation of resources (time, financial) according to the research interests of individual researchers. In the case of the Czech institutions, this effort was further motivated by the fulfilment of their voluntary commitment to lower landslide risk within the scope of the Sendai Partnerships (Vilímek et al., 2021). The Peruvian researchers acknowledged their responsibility to society and interest to collaborate with foreign partners as important motivations for the joint research. Nevertheless, the latter was the mutual motivation for both, national and international experts materialized in scientific papers preparation (Fig. 5). Despite the favourable conditions at the institutional level, the 2016–2017 LRR project was performed only thanks to the external financing of the specific funding scheme. This external financial support was crucial for the participation of the community and implementation of effective LRR measures which lowered its vulnerability and allowed the community development otherwise seriously restricted after the 2009 landslide.
The above-described context of the twelve years long LRR efforts in the Rampac Grande community (Fig. 5) follows closely the “ability to assist principle” originally defined as one of the possible principles to bring climatic justice (AAP, Huggel et al., 2016). It suggests that the responsibility for solving the negative impacts of climate change is more linked to economic, technological and logistic capacities, than e.g., to legal liability. Similarly, in the case of the Rampac Grande, a large portion of the responsibility to reduce the long-term negative impacts of the 2009 landslide was assumed by a group of experts who possessed economic, technological, knowledge and logistic capacities to solve the problem. At the same time, the entity legally responsible for risk reduction, the Carhuaz Province Municipality (Klimeš et al., 2019) significantly contributed to the 2016–2017 project implementation. We assume that the AAP is largely contained in the Sendai Partnerships for landslide disaster risk reduction (Sassa, 2015), which is, therefore, an effective tool to strengthen institutional responsibility for LRR at any place where it represents thread to communities.
Site-specific hazard assessment of two partial slides of the 2009 catastrophic compound landslide in the Rampac Grande community was performed. Results of the surface deformations monitoring and the calculation of the local factor of safety along with detailed morphology of the landslides allowed us to define their hazard. It shows medium or low hazard levels of the studied landslides under present environmental conditions showing that the slope stability is governed by so far unknown soil properties (e.g., suction, cementation) and fluctuation of underground water levels. The detailed hazard assessment was combined with field investigations of household vulnerabilities into risk assessment in the zone around the 2009 catastrophic landslide where the majority of the houses are under medium landslide risk (75%). Nevertheless, despite more than a decade of landslide risk reduction effort by various national and foreign expert groups, a quarter of houses is still subject to high landslide risk. It is due to high vulnerability, driven by social and economic conditions which improvement is beyond the applied landslide risk reduction measures. The high-vulnerability drives are relatively lower economic income, limited access to social programs, worst knowledge of safe places, limited interest to participate in training events and technical conditions of the houses.
Mid-term (e.g., after four years from implementation) evaluation of the effectivity of the mitigation measures showed that one of them (cf., extensometric monitoring of landslide movements) is no longer accepted by the community. No clear explanation of this development was obtained through the interviews with the community representative, but it may indicate the process of vulnerability increase due to the shift of the community's perception of the implemented LRR measures from preferring the expert-based measures to more traditional practices (e.g., the “healing” mountain principle). Many internal as well as external social processes could be related to this development, but we suggest that disruption of contacts between Peruvian experts performing the regular on-site measurements and the community, caused at least partly by the COVID-19 pandemic restrictions, could play a significant negative role in this process. This finding illustrates that the risk reduction process is an ongoing effort driven largely by the changing social conditions, which directly and sometimes very quickly affect the vulnerabilities of the exposed society.
Described context of the twelve years long landslide risk reduction efforts illustrate in detail the institutional settings and motivations of the expert’s involvement in it. In this case, it shows the importance of both, favourable institutional conditions and personal expert motivations as well as effective external funding schemes to implement effective long-term risk reduction measures. This specific case history also shows that the voluntary commitment to the Sendai Partnerships for landslide disaster risk reduction could motivate research institutions to assume responsibility for community-centered landslide risk reduction.
Acknowledgement
We would like to thank to Dr. Mariana Pirone (University of Naples Federico II, Italy) and Dr. Bohuslav Kuřík (Charles University Prague, Czech Republic) for discussions on slope stability and anthropological aspects of the research. We also acknowledge the ongoing support of our research activities by the Czech Embassy in Lima, Peru. The research works were supported by the long-term conceptual development research organization (RVO: 67985891) project and by the Strategic Research Plan of the Czech Geological Survey (DKRVO/ČGS 2018-2022, topic 5).
The preparation of the manuscript was not directly supported by any specific grant proposal. Funding used to perform field work is acknowledged in the Acknowledgement section and it includes especially internal institutional funding.
The authors have no relevant financial or non-financial interests to disclose.
- Aaron J, Hungr O, Stark T, Baghdady A (2017) Oso, Washington, Landslide of March 22, 2014: Dynamic Analysis. J Geotech Geoenv Eng 10.1061/(ASCE)GT.1943-5606.0001748, 143:9, (05017005)
- Bommer JJ, Rolo R, Mitroulia A, Berdousi P (2002) Geotechnical properties and seismic slope stability of volcanic soils. 12th European Conference on Earthquake Engineering, 9-13 September 2002, London, United Kingdom, Paper Reference 695
- CENEPRED (2014) Manual for the evaluation of risks caused by natural phenomena (in Spanish). CENEPRED, Lima
- Chandler RJ (1984) Recent European experience of landslides in over-consolidated clays and soft rocks (State-of-the-Art Lecture). In Proc. 4th Int. Symp. Landslides Toronto (Vol. 1, pp. 61–81)
- Davila C, Vera R, Pacheco L, Duran G (2020) Evaluation of the geotechnical behavior of a volcanic soil wall with additions of lime and cement against landslides. IOP Conf Ser: Mater Sci Eng 758, 012083. doi:10.1088/1757-899X/758/1/012083
- Evans SG, Bishop NF, Fidel Smoll L, Valderrama Murillo P, Delaney KB, Oliver-Smith A (2009) A re-examination of the mechanism and human impact of catastrophic mass flows originating on Nevado Huascarán, Cordillera Blanca, Peru in 1962 and 1970. Eng Geol 108, 96-118
- Federico A, Popescu M, Elia G, Fidelibus C, Interno G, Murianni A (2012) Prediction of time to slope failure: a general framework. Env Earth Sci 66:245-256. DOI: 10.1007/s12665-011-1231-5ns
- FOEN (2016) Protection against Mass Movement Hazards. Guideline for the integrated hazard management of landslides, rockfall and hillslope debris flows. Federal Office for the Environment, Bern. The environment in practice no. 1608
- Figueiredo AR, Simões JC, Menegat R, Strauss S, Rodrigues BB (2019) Perceptions of and adaptation to climate change in the Cordillera Blanca, Peru. DOI: http://dx.doi.org/10.14393/SN-v31-2019-45623
- Froude MJ, Petley DN (2018) Global fatal landslide occurrence from 2004 to 2016. Nat Hazards Earth Syst Sci 18, 2161-2181. https://doi.org/10.5194/nhess-18-2161-2018
- Geocon (2015) Instruction manual Model 1610 Tape Extensometer. Geocon
- Haller A (2010) Yungay: recent tendencies and spatial perceptions in an Andean risk zone. Espacio y Desarrollo 22:65-75.
- Heras H, Tavera H (2002) Localzación de áreas probables a ser afectadas por grandes sismos en el borde oeste de Perú: estimación a partir de períodos de retorno local basado en la distribución de valores de “b”. Bol Soc Geológica Perú 93:7-16
- Hostettler S, Jöhr A, Montes C, D’Acunzi A (2019) Community-based landslide risk reduction: a review of a Red Cross soil bioengineering for resilience program in Honduras. Landslides 16:1779-1791. https://doi.org/10.1007/s10346-019-01161-3
- Huggel C, Wallimann-Helmer I, Stone F, Wolfgang C (2016) Reconciling justice and attribution research to advance climate policy. Nat Clim Chang. doi: 10.1038/NCLIMATE3104
- Huggel C, Cochachin A, Drenkhan F, Fluixá-Sanmartín J, Frey H, García Hernández J, Jurt C, Muñoz R, Price K, Vicuña L (2020) Glacier Lake 513, Peru: lessons for early warning service development. WMO Bulletin 69(1):45-52
- INGEMMET (1995) Carta Geologica del Peru, 19-h Carhuaz, M 1:100 000. INGEMMET, Lima
- Iverson RM, et al. (2015) Landslide mobility and hazards: implications of the 2014 Oso disaster. Earth Planet Sci Lett 412: 197-208. https://doi.org/10.1016/j.epsl.2014.12.020
- Jakob M, Hungr O (2005) Debris Flow Hazard and Related Phenomena. Springer, Berlin
- Jibson RW (2012) Models of triggering of landslides during earthquakes. In: Clague JJ, Stead D (eds), Cambridge University Press, Cambridge, pp 196-206
- Jurt C, Burga MD, Vicuña L, Huggel C, Orlove B (2015) Local perceptions in climate change debates: insights from case studies in the Alps and the Andes. Clim Chang 133:511–523. https://doi.org/10.1007/s10584-015-1529-5
- Košťák B, Vilímek V, Zapata ML (2002) Registration of microdisplacements at a Cordillera Blanca fault scarp. Acta Montana IRSM AS CR, 19:61-74
- Kaser G, Ames A, Zamora M (1990) Glacier fluctuations and climate in the Cordillera Blanca, Perú. Ann Glaciol, 14:136-140
- Keaton JR, Wartman J, Anderson S, Benoit J, de LaChapelle J, Gilbert R, Montgomery DR (2014) The 22 March 2014 Oso landslide, Snohomish County, Washington. Rep. No. GEER- 036. Alexandria, VA: Geotechnical Extreme Events Reconnaissance Association.
- Klimeš J, Vilímek V (2011) A catastrophic landslide near Rampac Grande in the Cordillera Negra, northern Peru. Landslides 8:309-320
- Klimeš J (2018) Extensometer. In: Bobrowsky PT, Marker B (eds) Encyclopedia of Engineering Geology. Encyclopedia of Earth Sciences Series. Springer, Cham, pp. 323-324
- Klimeš J, Rosario AM, Vargas R, Raška P, Vicuña L, Jurt C (2019) Community participation in landslide risk reduction: a case history from Central Andes, Peru. Landslides 16:1763-1777. https://doi.org/10.1007/s10346-019-01203-w
- Maes J, Mertens K, Jacobs L et al. (2018) Social multi-criteria evaluation to identify appropriate disaster risk reduction measures: application to landslides in the Rwenzori Mountains, Uganda. Landslides. https://doi.org/10.1007/s10346-018-1030-0
- Macharé J, Fenton CH, Machette MN, Lavenu A, Costa C, Dart RL (2003) Database and map of Quaternary faults and folds in Perú and its offshore region. Open File Report 03-451, USGS
- Maskrey A (1989) Disaster mitigation: a community based approach (development guidelines). Oxfam, Oxford
- Mercer J, Kelman I, Suchet‐pearson S, Lloyd K (2009) Integrating indigenous and scientific knowledge bases for disaster risk reduction in Papua New Guinea, Geografiska Annaler: Series B, Human Geog, 91:2, 157-183. DOI: 10.1111/j.1468-0467.2009.00312.x
- Muñoz AR, Gonzales C, Price K, Rosario A, Huggel C, Frey H, García J, Cochachin A, Portocarrero C, Mesa L (2016) Managing glacier related risks disaster in the Chucchún catchment, Cordillera Blanca, Peru. In: Salzmann N, Huggel C, Nussbaumer S, Ziervogel G (eds) Climate change adaptation strategies — an upstream-downstream perspective, Springer, Cham, 59-78 https://doi.org/10.1007/978-3-319-40773-9_4
- Nadim F, Jaedicke C, Smebye H, Kalsnes B (2013) Assessment of Global Landslide Hazard Hotspots. In: Sassa K, Rouhban B, Briceño S, McSaveney M, He B (eds) Landslides: Global Risk Preparedness. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-22087-6_4
- Neal DM (1997) Reconsidering the phases of disaster. Int J Mass Emerg Disasters, 15:239-264.
- Novotný J (2014) Engineering Geological Models – Some Examples of Use for Landslide Assessments, 12th IAEG Congress Torino 2014, Lollino G et al. (eds), Engineering Geology for Society and Territory – Volume 7, doi: 10.1007/978-3-319-09303-1_2, Springer International Publishing Switzerland 2014, pp 11-15
- Parry S, Baynes F, Novotný J (2018) Conceptual engineering geological models. in IAEG/AEG annual meeting proceedings. Editors Shakoor A, Kato K. (San Francisco, California San Francisco: Springer). doi:10.1007/978-3-319-93142-5
- Pedoth L, Taylor R, Kofler C, Stawinoga AE, Forrester J, Matin N, Schneiderbauer S (2019) The Role of Risk Perception and Community Networks in Preparing for and Responding to Landslides, A Dolomite case study. Deeming H, Fordham M, Kuhlicke C, Pedoth L, Schneiderbauer S, Shreve C (eds): Framing Community Disaster Resilience: Resources, Capacities, Learning, and Action. John Wiley & Sons Ltd., pp 197-219
- Pirone M, Papa R, Nicotera MV, Urciuoli G (2015) Soil water balance in an unsaturated pyroclastic slope for evaluation of soil hydraulic behaviour and boundary conditions. J Hydro 528:63-83. http://dx.doi.org/10.1016/j.jhydrol.2015.06.005
- Plafker G, Ericksen GE, Concha JF (1971) Geological aspects of the May 31, 1970, Peru earthquake. Bull Seismol Soc Am 61:543-578
- Pollock W, Wartman J (2020) Human vulnerability to landslides. Geo Health, 4. DOI:10.1029/2020GH000287
- Rechberger C, Fey C, Zangerl C (2021) Structural characterisation, internal deformation, and kinematics of an active deep-seated rock slide in a valley glacier retreat area. Eng Geol, 286: 106048
- Raetzo H, Lateltin O, Bollinger D, Tripet JP (2002) Hazard assessment in Switzerland - codes of practice for mass movements. Bull Eng Geol Environ 61: 263-268. https://doi.org/10.1007/s10064-002-0163-4
- Roháč J, Scaringi G, Boháč J, Kycl P, Najser J (2020) Revisiting strength concepts and correlations with soil index properties: insights from the Dobkovičky landslide in Czech Republic. Landslides 17:597-614
- Roháč J (2021) Experimental investigation of shear strength of soils from the Dobkovičky landslide area and development of correlations between soil strength and their physical properties, PHD Thesis, Charles University, Faculty of Science
- Rotisciani GM, Desideri A, Amorosi A (2021) Unsaturated structured soils: constitutive modelling and stability analyses. Acta Geotech, 16:3355–3380. https://doi.org/10.1007/s11440-021-01313-7
- Sassa K (2015) ISDR-ICL Sendai partnerships 2015–2025 for global promotion of understanding and reducing landslide disaster risk. Landslides 12, 631-640. https:// doi.org/10.1007/s10346-015-0586-1
- Scheidegger A (1973) On the prediction of the reach and velocity of catastrophic landslides. Rock Mech 5:231-236
- Schuster RL, Highland ML (2001) Socieconomic and environmental impacts of landslides in the Western Hemisphere. Open File Report 01-0276, USGS
- Skempton AW (1985) Residual strength of clays in landslides, folded strata and the laboratory. Géotechnique 35:3-18
- Stark TD, Baghdady A, Hungr O, Aaron J (2017) Case study: Oso landslide on 22 March 2014—Material properties and failure mechanism. J Geotech Geoenv Eng 10.1061/(ASCE)GT.1943-5606.0001615, (05017001)
- Stemberk J, Vilímek V, Klimeš J, Blahůt J, Hartvich F, Balek J (2017) Landslide Hazard and Risk Management (WCoE 2014–2017). In: Sassa K, Mikoš M, Yin Y (eds) Advancing Culture of Living with Landslides. WLF 2017. Springer, Cham, 373-377. doi: 10.1007/978-3-319-59469-9_32
- Strozzi T, Klimeš J, Frey H, Caduff R, Huggel C, Wegmüller U, Rapre AC (2018) Satellite SAR interferometry for the improved assessment of the state of activity of landslides: A case study from the Cordilleras of Peru. Remot Sens Env 217: 111-125. https://doi.org/10.1016/j.rse.2018.08.014
- USGS, Earthquake Hazard Program, 2020a, https://earthquake.usgs.gov/earthquakes/eventpage/iscgem796163/shakemap/pga, [accessed 27.8.2021]
- USGS, Earthquake Hazard Program, 2020b, https://earthquake.usgs.gov/earthquakes/eventpage/us60003sc0/executive, [accessed 27.8.2021]
- Vilímek V, Zapata ML, Stemberk J (2000) Slope movements in Callejón de Huaylas, Peru. Acta Univer Carolinae Geograph, XXXV, Supplementum: 39-51
- Vilímek V, Klimeš J, Vlčko J, Carreño R (2006) Catastrophic debris flows near Machu Picchu village (Aguas Calientes), Peru. Env Geol 50:1041-1052
- Vilímek V, Hanzlík J, Sládek I, Šandov M, Santillán N (2012) The share of landslides in the occurrence of natural hazards and the significance of el Niño in the Cordillera Blanca and Cordillera Negra Mountains, Peru. In: Sassa K, Rouhban B, Briceno S, McSeveny M, He B (eds) Landslides: Global Risk Preparedness. Springer, 133–148. https://doi.org/10.1007/978-3-642-22087-6_9
- Vilímek V, Klimeš J, Torres MZ (2016) Reassessment of the development and hazard of the Rampac Grande landslide, Cordillera Negra, Peru. Geoenv Disas, 3:5. doi:10.1186/s40677-016-0039-8
- Vilímek V, Klimeš J, Tito Mamani RV, Bastante AJ, Astete FV, Champi PZM (2020) Contribution of the collaborative effort of the Czech WCoE to landslide risk reduction at the Machupicchu World Heritage Site, Peru. Landslides 17: 2683-2688. https://doi.org/10.1007/s10346-020-01509-0
- Vilímek V, Klimeš J, Stemberk J, Burda J, Kycl P, Blahůt J (2021) Complex Geomorphological and Engineering Geological Research of Landslides with Adverse Societal Impacts. In Sassa K, Mikoš M, Sassa S, Bobrowsky PT, Takara K, Dang K (eds) Understanding and Reducing Landslide Disaster Risk. WLF 2020. ICL Contribution to Landslide Disaster Risk Reduction. Volume 1: Sendai Partnerships and Kyoto Landslide Commitment. Springer, Cham, 275-280. https://doi.org/10.1007/978-3-030-60196-6_20
- Weston R (2008) Preliminary summary of the Pallcamachay Property, Ancash, Peru. Report Strait Gold Corp. [accessed 29.4.2010]
- Yoshikawa K, Úbeda J, Masías P, Pari W, Apaza F, Vasquez P, Ccallata B, Concha R, Luna G, Iparraguirre J, Ramos I, De la Cruz R, Pellitero R, Bonshoms, M (2020) Current thermal state of permafrost in the southern Peruvian Andes and potential impact from El Niño–Southern Oscillation (ENSO). Permafrost Periglac Process, 1-12. https://doi.org/10.1002/ppp.2064
No competing interests reported.