Debris flow susceptibility mapping using frequency ratio and seed cells, in a portion of a mountain international route, Dry Central Andes of Argentina
Graphical abstract
Introduction
Road networks play a vital role in maintaining a functioning modern society. Natural hazards, such as debris flows, can significantly impact the flow of vehicles and cargo along road networks. Unlike traffic disruptions that result from accidents or maintenance activities, natural disasters are capable of destroying large numbers of roads and usually cover vast areas (Bíl et al., 2015). The vulnerability of transport infrastructures exposed to geological hazards has attracted much attention recently. To assess these issues quantitatively, operational measures are needed. Transportation agencies can use operational measures to prioritize road maintenance as well as to avoid causing unnecessary disturbances in the planning of roadwork (Jenelius et al., 2006).
Mountainous environments are highly unstable. An increasing intensity of human land use on mountainous terrain has led to an increase in natural hazards (Kellerer-Pirklbauer, 2002). Furthermore, road construction on a hillslope increases the slope above the road and stress on the back of the slope. Thus, a hillslope that was in stable condition before the road was constructed may become unstable because of water ingress (Regmi et al., 2014).
Dry Central Andes of Argentina is affected by various natural geologic processes. Nevertheless, debris flows, mainly triggered by heavy rainfall or by rapid snowmelt, are the most common type of landslide processes and occur frequently in the Central Andes, causing damages and economic losses in the region. Debris flow susceptibility maps yield valuable information for planners and engineers who implement land use strategies, not only in the design stages of road infrastructure, but also in the stages of hazard mitigation to which the that infrastructure are exposed. Since the occurrence of small slide on the head of the basin has a great potential to initiate debris flows, the undisturbed morphological conditions before slides occurred can be inferred from landforms in the vicinity of the slide polygons. Süzen and Doyuran, 2004a, Süzen and Doyuran, 2004b, Nefeslioglu et al., 2008 used an approach called “seed cells” in generating decision rules of landslide occurrence. Seed cells are located in a buffer zone along the crown and flanks of each landslide.
Many recent studies using GIS to evaluate debris flow susceptibility have applied probabilistic models, such as frequency ratios and seed cells (Achour et al., 2018, Bai et al., 2010, Che et al., 2012, Dholakia et al., 2015, Esper Angillieri, 2013, Esper Angillieri and Perucca, 2014, Wang et al., 2011).
Seventeen potential debris flow-prone basins have been identified on a segment of route N° 150 in the Agua Negra river basin, Argentina (Esper Angillieri and Fernandez, 2017), potentially exposing, this sector, to economic and life losses. This study analyzes six different scenarios to debris flow susceptibility assessment. Although, each scenario and model, has been applied successfully to resolve real problems with others specific studies, these methods have never been applied or compared for debris flow susceptibility modeling in the Andes of Argentina.
Therefore, the current study presents debris flow susceptibility maps using a widely-accepted model, a statistical method (frequency ratio model), and evaluates and compares the performance of these six models of debris flow susceptibility assessment for an important segment of a route on the Dry Andes of San Juan in the Agua Negra river basin, Argentina.
Section snippets
Study area
The Quebrada de Agua Negra basin (Fig. 1) is located in a segment of the Frontal Cordillera, in the Iglesia Department, 206 km from San Juan City, San Juan Province, Argentina. The area includes a section of 150 Road in the Quebrada de Agua Negra, which provides access to the International Agua Negra Pass (Argentina – Chile) and future international tunnel. The drainage basin's area is approximately 420 km2. Basin elevations range from 2634 m to 6180 m, with a mean value of 4490 m, while slopes
Material and methods
Fieldwork; high resolution satellite imagery (SPOT 5 with a 2.5 m spatial resolution) from Google EarthTM; orthorectified basemap satellite imagery from ArcGIS online (Esri, 2014) digital topographic data (Aster GDEM) and GIS technology (ESRI's ArcGis 10.3) were used to produce the debris flow inventory map (Fig. 1). Three mapping methodologies were selected to compile the inventory: all flow track (landslide detachment zone, main flow track, and the accumulation zone), flow track without the
Evaluation of inventory map and debris flows-related factors
Fifty-nine debris flows were identified in the study area by evaluating satellite imagery (Fig. 1). They covered an area of 12.52 km2 and encompassed 2.97% of the study area (420.79 km2). Only 47 debris flows were used for susceptibility analysis; the rest were reserved for verification purposes.
To calculate frequency ratios, values for each of the 15 debris flow-related factors were input into raster grids with 30-m resolution.
Table 1 shows the class debris flows-related factors and frequency
Conclusions
This paper presented an approach for identifying the most vulnerable points to debris flows susceptibility along a significant mountain road. Six models — never before compared in the literature — were used to assess debris flow susceptibility in the study area. A debris flow inventory map and 15 maps of debris flow-related factors were applied to simulate the models. Findings indicated that the classes of debris flows-related factors most sensitive to the occurrence of debris flows in the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors acknowledge to anonymous reviewer and the funding received from CONICET (Argentinean National Council of Scientific and Technological Research); CICITCA (Secretary of Science and Technology UNSJ) to support this research. And CIGEOBIO by providing funds for ArcGIS 10.3 software license.
References (58)
- et al.
The application of GIS-based logistic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan
Geomorphology
(2005) - et al.
GIS-based logistic regression for landslide susceptibility mapping of the Zhongxian segment in the Three Gorges area, China
Geomorphology
(2010) - et al.
Evaluating road network damage caused by natural disasters in the Czech Republic between 1997 and 2010
Transp. Res. Part A Policy Pract.
(2015) - et al.
Landslide susceptibility assessment in Limbe (SW Cameroon): A field calibrated seed cell and information value method
Catena
(2012) Application of logistic regression and frequency ratio in the spatial distribution of debris-rockslides: Precordillera of San Juan, Argentina
Quat. Int.
(2015)- et al.
A large and active debris-rockslide in the Central Andes of Argentina (30.26°S): Morphometry and triggering mechanisms
Quat. Int.
(2015) - et al.
Different landslide sampling strategies in a grid-based bi-variate statistical susceptibility model
Geomorphology
(2016) - et al.
Road network vulnerability: Identifying important links and exposed regions
Transp. Res. A
(2006) - et al.
An assessment on the use of logistic regression and artificial neural networks with different sampling strategies for the preparation of landslide susceptibility maps
Eng. Geol.
(2008) - et al.
Data driven bivariate landslide susceptibility assessment using geographical information systems: A method and application to Asarsuyu catchment
Turkey. Eng. Geol.
(2004)
Landslide susceptibility assessment by bivariate methods at large scales: Application to a complex mountainous environment
Geomorphology
How can statistical models help to determine driving factors of landslides?
Ecol. Modell.
Landslide susceptibility mapping: A comparison of logistic regression and neural networks methods in a medium scale study, Hendek region (Turkey)
Eng. Geol.
GIS-based spatial prediction of debris flows using logistic regression and frequency ratio models for Zêzere River basin and its surrounding area, Northwest Covilhã
Portugal. Arab. J. Geosci.
Comparing heuristic landslide hazard assessment techniques using GIS in the Tirajana basin, Gran Canaria Island, Spain
Int. J. Appl. Earth Obs. Geoinf.
Landslide susceptibility analysis in the Hoa Binh province of Vietnam using statistical index and logistic regression
Nat. Hazards
Integrated application of the analytic hierarchy process and the geographic information system for flood risk assessment and flood plain management in Taiwan
Nat. Hazards
Probabilistic prediction models for landslide hazard mapping
Photogramm. Eng. Remote Sens.
System for Automated Geoscientific Analyses (SAGA) v. 2.1.4
Geosci. Model Dev.
Factors selection in landslide susceptibility modelling on large scale following the gis matrix method: Application to the river Beiro basin (Spain)
Nat. Hazards Earth Syst. Sci.
Susceptibility analysis of landslide in Chittagong City Corporation Area
Bangladesh. Int. J. Environ.
Landslide Susceptibility Assessment at a Part of Uttarakhand Himalaya, India using GIS – based Statistical Approach of Frequency Ratio Method
Int. J. Eng. Res.
Application of logistic regression for landslide susceptibility zoning of Cekmece Area, Istanbul
Turkey. Environ. Geol.
Susceptibilidad a la ocurrencia de flujos de detritos, mediante sig, en una porción del Departamento de Iglesia, San Juan
Rev. la Asoc. Geol. Argentina
Geomorphological of a portion of the Andes and Pre-Andes of San Juan, Argentina
J. Maps
Morphometric analysis of river basins using GIS and remote sensing of an Andean section of Route 150, Argentina. A comparison between manual and automated delineation of basins
Rev. Mex. Ciencias Geológicas
Geomorphology and morphometry of the de La Flecha river basin, San Juan, Argentina
Environ. Earth Sci.
Cited by (27)
Extreme climate and tectonic controls on the generation of a large-scale, low-frequency debris flow
2022, CatenaCitation Excerpt :A higher FI indicates that the freezing degree of the soil in winter and the thickness of the frozen soil layer is less than that in a typical year and that the barrier effect of the soil layer on seepage is reduced (Iwata et al., 2011). This allows large amounts of snowmelt infiltration to trigger landslides in the watershed and replenish the soil mass of the debris flow (Esper Angillieri, 2020; Patton et al., 2019; Schneuwly-Bollschweiler and Stoffel, 2012). In addition, the freeze-thaw index (FTI), defined as the sum of the TI and FI, was 6294.1, which was the smallest in 2014 since the records began.
Economic Risk Assessment of Future Debris Flows by Machine Learning Method
2024, International Journal of Disaster Risk ScienceModelling of debris-flow susceptibility and propagation: a case study from Northwest Himalaya
2024, Journal of Mountain Science