No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Simple GIS Model for Mapping Landslide Susceptibility
Abstract
Digital maps of geology, ground slope, and dormant landslides are combined statistically in a geographic information system (GIS) to identify sites of future landsliding over a broad area. The resulting index number, a continuous variable, predicts a range of susceptibility both within and between existing landslides. Spatial resolution of the index can be as fine as that of the slope map, and areal coverage is limited only by the extent of the input data. Susceptibility is defined for each geologic-map unit as the spatial frequency of the unit occupied by dormant landslides, adjusted locally by ground slope. Susceptibility of terrain between landslides is calculated for each one-degree slope interval as the percentage of grid cells that coincide with the failures. Susceptibility within landslides is the same percentage times the comparative frequency of recent failures within and outside the old landslides. We tested the model in an 872 km 2 urban area in California, using 120 geologic units, a 30-m digital elevation model, 6714 dormant landslide deposits, 1192 recent landslides, and ARC/INFO software. The method could generate a similar map for any area where the necessary digital-map data are available.
... Pada tahun 1988 Evans dan King [4] telah mempetakan sebuah daerah kerentanan gerakan tanah berdasarkan korelasi dari gerekan tanah yang sudah terjadi dengan kemiringan lereng serta kondisi geomorfologi. Berbagai penelitian juga telah membahas berbagai hal mengenai aplikasi GIS pada daerah rentan longsor [5], [6] dan [7]. ...
Landslide is one of the most serious natural disasters causing great losses in term of materials and lives. Digital maps of geology, ground slope, and dormant landslides are combined statistically in a geographic information system (GIS) to identify sites of future land sliding over a broad area. The case study area is at Timor Tengah Selatan (TTS) District, Nusa Tenggara Timur (NTT) Province, Indonesia. Landslide hazard potential and prediction model were assessed at regional scale 1:25.000 [1]. In this study, weighting and ranking of importance of factors to landslide occurrence are used to identification landslide potential areas. It is based on the observed relationship between each instability factor and the past landslide distribution. The obtained results allow to define the main factors causing land sliding as: slope rate, hydro-geological structure, surface weathering factors, distance to active faults and impact of human activities (land usage, plantation coverage etc). The degree of landslide hazard is expressed in relative term from very low to very high hazard level, and represents the expectation of future landslide occurrence based on the conditions of that particular area. It is obvious from the result map that the areas under high and very high hazard level are near the first and second stream orders of the study area. Finally, landslide hazard maps were produced. The result from this study represents differing hazard levels that show only the order of relative hazard at a particular site and not the absolute hazard.
... Jakobsen, 2003;Clark et al., 2004); 2) GIS used in slope analysis and (natural) hazard zonation and management (e.g. Dai and Lee, 2002;Zerger, 2002;Pike et al., 2003;van Westen et al., 2003;Gaspar et al., 2004;Otto and Dikau, 2004;Seijmonsbergen and de Graaff, 2006); and 3) automated or supervised landscape classification from remote sensing data often in combination with Digital Elevation Models (DEMs) (Brown et al., 1998;Giles and Franklin, 1998;Bocco et al., 2001;Bartsch et al., 2002;Plaza et al., 2004;van Asselen and Seijmonsbergen, 2006). Even though some attempts to construct applied or thematic GIS databases have incorporated a wide variety of data (e.g. ...
... Jakobsen, 2003;Clark et al., 2004); 2) GIS used in slope analysis and (natural) hazard zonation and management (e.g. Dai and Lee, 2002;Zerger, 2002;Pike et al., 2003;van Westen et al., 2003;Gaspar et al., 2004;Otto and Dikau, 2004;Seijmonsbergen and de Graaff, 2006); and 3) automated or supervised landscape classification from remote sensing data often in combination with Digital Elevation Models (DEMs) (Brown et al., 1998;Giles and Franklin, 1998;Bocco et al., 2001;Bartsch et al., 2002;Plaza et al., 2004;van Asselen and Seijmonsbergen, 2006). Even though some attempts to construct applied or thematic GIS databases have incorporated a wide variety of data (e.g. ...
This paper presents the structure and contents of a standardised geomorphological GIS database that stores comprehensive scientific geomorphological data and constitutes the basis for processing and extracting spatial thematic data. The geodatabase contains spatial information on morphography/morphometry, hydrography, lithology, genesis, processes and age. A unique characteristic of the GIS geodatabase is that it is constructed in parallel with a new comprehensive geomorphological mapping system designed with GIS applications in mind. This close coupling enables easy digitalisation of the information from the geomorphological map into the GIS database for use in both scientific and practical applications. The selected platform, in which the geomorphological vector, raster and tabular data are stored, is the ESRI Personal geodatabase. Additional data such as an image of the original geomorphological map, DEMs or aerial orthographic images are also included in the database. The structure of the geomorphological database presented in this paper is exemplified for a study site around Liden, central Sweden.
Reichenbach et al. (Earth Sci Rev 180:60–91, 2018) provide an extremely valuable review of statistically-based landslide susceptibility modelling and mapping techniques. In their analysis, they describe an excessive interest in statistical experimentation, seemingly at the expense of focusing on developing reliable and useable outputs. Landslide susceptibility mapping has flourished in research circles, but has yet to become widely embraced by land use planners and those responsible for the site selection and route selection of engineering infrastructure. If landslide susceptibility mapping is to become recognised as a credible tool by planning and engineering practitioners, it requires robust and universally-applied guidelines, a focus on geological and geomorphological observation, and demonstrable reliability and usability. Short case histories are provided to illustrate some of the issues concerned, how they have been overcome in an applied sense, and some of the problems that still remain to be resolved. Recommendations are provided for a step-by-step approach to landslide susceptibility mapping, that emphasise the need to: (1) become fully-familiar with the terrain in the area of interest and the controls on slope stability; (2) derive a credible dataset for spatial analysis, combining desk study and field-derived data sources; (3) test and trial the output mapping; and (4) liaise with the intended map end-user over issues concerning scale, reliability, uncertainty and application constraints.
In this paper we presented an attempt to determine landslide susceptibility of slopes. The investigation was done for central part of Kamienne Mts. (Central Sudetes), where slopes are strongly affected by mass movements. We used DEM and basic and derivate land-surface parameters (slope, relief, topographic wetness index) to modeling in SAGA GIS and MicroDEM software. Using raster map algebra four usceptibility classes were obtained. Results of modeling were compared with elevations and colluvial deposits areas derived from geological and geomorphological maps. Used method showed a helpfulness of these land-surface parameters for landslide predicting. It emphasized also a problem of bedrock
structure influence on mass wasting distribution.
Slope instabilities in the central Southern Alps, New Zealand, are assessed in relation to their geological and topographic
distribution, with emphasis given to the spatial distribution of the most recent failures relative to zones of possible permafrost
degradation and glacial recession. Five hundred nine mostly late-Pleistocene- to Holocene-aged landslides have been identified,
affecting 2% of the study area. Rock avalanches were distinguished in the dataset, being the dominant failure type from Alpine
slopes about and east of the Main Divide of the Alps, while other landslide types occur more frequently at lower elevations
and from schist slopes closer to the Alpine Fault. The pre-1950 landslide record is incomplete, but mapped failures have prevailed
from slopes facing west–northwest, suggesting a structural control on slope failure distribution. Twenty rock avalanches and
large rockfalls are known to have fallen since 1950, predominating from extremely steep east–southeast facing slopes, mostly
from the hanging wall of the Main Divide Fault Zone. Nineteen occurred within 300 vertical metres above or below glacial ice;
13 have source areas within 300 vertical metres of the estimated lower permafrost boundary, where degrading permafrost is
expected. The prevalence of recent failures occurring from glacier-proximal slopes and from slopes near the lower permafrost
limit is demonstrably higher than from other slopes about the Main Divide. Many recent failures have been smaller than those
recorded pre-1950, and the influence of warming may be ephemeral and difficult to demonstrate relative to simultaneous effects
of weather, erosion, seismicity, and uplift along an active plate margin.
KeywordsLandslide inventory–Rock avalanche–Glacial change–Permafrost–Southern Alps–New Zealand
Landslides and associated types of slope failure such as accelerated soil and rock creep have become a major geologic hazard in the San Francisco Bay region. As increasing development of hillside areas has taken place since the mid-1940's, the cost of damage from slope failures have steadily increased. For the entire San Francisco Bay region, more than $25 million of damage was caused by landslides during the rainy season of 1968-69 and more than $10 million in 1972-73. These losses can be greatly reduced by: 1) using geologic information to recognize, evaluate, and map those areas and slopes that are potentially unstable, and 2) applying this information in planning, designing, and organizing the use of hillside areas. For this report, we have prepared the first standardized relative slope stability maps (scale l:125 000) of the entire San Francisco Bay region, and we discuss the implications and uses of these maps in the regional land-use planning process. -from Authors
In the Summit Ridge area, the 1989 Loma Prieta earthquake generated many anomalously large, complex landslides that differed from other landslides triggered by this or other, comparable historical earthquakes. The area in which these large landslides occurred was immediately southwest of the San Andreas fault and contained abundant other coseismic ground-deformation features. The landslides were recognized and differentiated from other coseismic effects by detailed mapping and analysis of the geomorphology of the study area. In addiition to mapping and analysis, investigations included postearthquake monitoring of movement and ground-water levels; determination of subsurface conditons through drilling, trenching and a survey of earthquake-related damage to water wells; and analytical slope-stability modeling. Distinctive characteristics of the landslides in the Summit Ridge area included their large sizes: The landslides were as much as 980 m long and as much as 1220 m wide, and had surface areas as great as 85 ha. Maximum landslide depths ranged from 27 to possibly more than 100 m, and estimated landslide volumes were as great as 27 million m3. Other distinctive features of the landslides included irregular shapes and boundaries, defined by discontinous sets of surface cracks, scarps, and compressional features. Commonly, main scarps and heads were well developed and marked by conspicuous, relatively continuous scarps and cracks; flanks were moderately developed and delineated by discontinuous, locally echelon cracks; and toes were poorly developed, with local compressional features present on some landslides but absent on others. Measured landslide displacements, which ranged from 25 to 244 cm, were small relative to landslide lengths, and the incomplete development of landslide boundaries probably resulted from these comparatively small displacements. Material in the landslides was heterogeneous colluvium and bedrock consisting of poorly to moderately cemented sandstone, mudstone, siltstone and shale, which were locally sheared, faulted and folded. The largest landslides occurred where folding was most intense and faults were most numerous. Virtually all the earthquake-induced landslides exhibited geomorphic indications of previous landslide movement. Postearthquake movement and ground-water levels were monitored through spring 1991. Rainfall during that 2-year period was below normal, except for a short time in February and March 1991, when renewed displacement and associated ground cracking occurred along and adjacent to the main scarps of three landslides. This activity, which was the only significant landslide movement recorded during the 2 years of monitoring, was probably due to local upslope migration (retrogressive failure) of the landslide heads. Measured ground-water levels during the 1991 period of high rainfall rose as much as 6.1 m. These large, complex landslides occurred only where ground cracks produced by other coseismic processes were also abundant. This correlation indicates that the two types of features are mechanistically related. Many of the coseismic ground cracks were almost certainly produced either directly by tectonic deformation or by the effects of seismic shaking on preexisting planes of weakness. Once these ground cracks formed, they provided loci for downslope movement and landslide formation as shaking continued. This mechanism implies that landslides of this type are specifically generated by seismic shaking, a conclusion supported by both slope-stability modeling and by the historical record of landslides in the study area.
Almost all documented debris flows triggered by the storm of January 3-5, 1982, in the San Francisco Bay region developed from shallow slides (soil slips). These soil slip/debris flows ranged from small events that traveled about 10m, to soil slip/debris torrents that traveled down major canyons for more than a kilometer. Elements common to these phenomena were their typical occurrence during intense storm rainfall, flow as slurry, generally rapid movement, and sudden impact. -from Authors
Landslide inventory maps are generally prepared by interpreting the geomorphic expression of landsliding on aerial photos, topographic maps, or on the ground. Distinctive landslide geomorphology allows the recognition and mapping of landslides, although there are always landslides that have very subtle expression and are not identified. The difficulties of mapping landslides based on their geomorphic expression are amplified in heavily forested terrain. The ground surface is obscured by tree cover on aerial photographs, and landslide-related features are often hidden. This limitation affects not only aerial photo interpretation, but also interpretation of topographic maps, which are based on aerial photographs. We compared five maps showing landslides in the Laurel Quadrangle in the Santa Cruz Mountains, California. These include a geologic map, a map prepared for the county based on interpretation of aerial photographs, a map prepared by us based on aerial photographs and compilation of previous work, a map of features interpreted from the U.S. Geological Survey 7.5-minute topographic map, and a detailed field-based landslide map. Comparison of these maps shows that the geologic map identifies few landslides, but most landslides on the geologic map are also shown on the other maps. The two maps based mainly on aerial photo interpretation tend to show the larger slides, but there is only about 60 percent correspondence of landslide areas between the two. Comparing the reconnaissance techniques with the much more detailed field mapping shows that the reconnaisance maps emphasize the large slides of bedrock and identify a lower percentage of shallow debris slides and debris flows.
This paper presents and analyses the findings of a landslide inventory survey undertaken in deeply dissected upland valleys in layered Carboniferous lithology, Rhondda Valleys, South Wales. This area has one of the highest recorded concentrations of landslides in the UK.
The paper analyses the terrain factors identified at landslide sites (lithology, slope angle, geographical setting) with factors denning the landslide (such as type/depth of movement, age, activity, material, areal extent, geometry). Using these data, a composite profile of probable landslide sites is developed. This provides a valuable tool in further quantifying likely landslide hazard at sites, giving an indication of the potential of, for example, first-time failures or whether degraded relict landslides may be present.
Discussion is provided on the findings, particularly landslide locations within the valley and their age and current state of activity.
This study is concerned with (1) defining the terrain parameters and attributes that are important to landsliding in the Amahata River basin near Tokyo, Japan and (2) mapping landslide susceptibility from these characteristics. After aerial photographic interpretation and field surveys, ten terrain parameters, which were thought to be important to landsliding, were measured at each landslide and at the intersections of a 250-meter grid. Failure rate analysis and quantification scaling type II suggest that slope gradient, aspect, slope plan form, break of slope, elevation, and vegetation are the most critical site characteristics.
Landslide-susceptibility mapping was then accomplished by combining the three most basic terrain parameters: slope gradient, aspect, and slope plan form. First, stable and unstable categories were separated; then each category was divided into two classes to give four susceptibility classes: high, moderate, low, and least. These classes represent a qualitative index of the likelihood of a landslide occurring during heavy rainfall.
To investigate the linkage between erosion process and channel network extent, we develop two simple erosion threshold theories driven by a steady state runoff model that are used in the digital terrain model TOPOG to predict the pattern of channelization. TOPOG divides the land surface into elements defined by topographic contours and flow lines, which can be classified as divergent, convergent and planar elements. The calibration parameter for the runoff model is determined using empirical evidence that the divergent elements which comprise the ridges in our study area do not experience saturation overland flow, where as the convergent elements in the valleys do during significant runoff events. A threshold theory for shallow landsliding predicts a pattern of instability consistent with the distribution of landslide scars in our 1.2 km2 study site and confirms the interpretation, based on field observations, that indicate the steeper channel heads to be at least partially controlled by slope instability. Most sites of predicted and observed slope instability do not, however, support a channel head, hence landslide instability alone is not sufficient for channelization. In contrast, most elements predicted to be eroded by saturation overland flow coincided with the observed location of the channel network. In addition, areas of predicted downslope decrease in relative sediment transport capacity were found to correspond to locations where channels became discontinuous. -from Authors
Hillsides bordering the Santa Clara ("Silicon") Valley are prone to
earthquake- and rainfall-triggered landslides. To narrow the
uncertainty surrounding the location of future slope movement, we
estimate the relative likelihood of landsliding at 30-m resolution from
regional data sets. While small areas are evaluated accurately from
detailed observations on material properties, topography, and hydrology,
resource limitations force a regional, statistical, approach to mapping
large areas. This is the first county-size susceptibility map prepared
from the Arc/Info GIS model developed by Pike and others (2001) in the
Oakland, CA, area (USGS MF-2385). Their index of susceptibility is the
spatial frequency of prior slope failure for each one-degree slope
interval in each geologic-map unit, obtained by combining a geologic map
(a proxy for rock or soil strength) with a landslide-inventory map and a
map of slope gradient. Areas in existing landslides are multiplied by
an observationally derived constant, 1.33, to reflect their higher
susceptibility. Santa Clara County's 170 geologic units occupy about
3,440,000 30-m grid cells. The geology is from three digital USGS maps
(OF 97-710, 98-348, 98-795). The thousands of existing landslides were
digitized from 13 inventories--six by the California Geological Survey,
five by USGS, two by private consultants. These source maps vary in
scale (mostly 1:24,000 or 1:12,000), date (1970-98), attribution of
failures (most are large rock and debris slides, rock slumps, and earth
flows; few are debris flows), completeness, and detail. The map of
slope gradient (OF 98-766) was computed from a 30-m digital elevation
model (OF 98-625). Severity of prior landsliding throughout the county
strongly reflects geology--from a mean spatial frequency of zero, e.g.
flat-lying Pleistocene alluvial fan and fluvial deposits, to 100%, e.g.
a mudstone member of the Oligocene-Eocene San Lorenzo Formation. Values
of the susceptibility index, ranging from zero (1,000,000 cells largely
on the valley floor) to 1.33 (4200 cells in the most hazardous terrain),
are not randomly distributed. Among large areas of highest
susceptibility (>0.60) are the hills flanking Santa Clara Valley just
E of the cities of San Jose, Milpitas, and Morgan Hill, as well as some
terrain in the Santa Cruz Mountains.
We have developed a rapid, flexible method for predicting potentially unstable landscapes, in the form of earthflows and rotational landslides, using a set of land-surface morphology parameters and related process calculations. We used an eastern Ohio drainage basin as a trial area, having calculated the land-surface morphology parameters from an elevation data set for the basin. The parameters, along with the calculations of gravitational stress on the soil mass and solar irradiance at an equinox and geologic-column information, were combined in a multiple-factor analysis to identify potentially unstable areas. Combining the factors of gravitational stress, surface curvature, solar irradiance, and geologic features eliminated 55, 30, 5, and 2 percent, respectively, of the total land area when combined in the sequence presented. Field investigations found that earthflow and landslide remnants in the trial drainage basin were confined to those areas the model had determined to be potentially unstable.
(C) Williams & Wilkins 1983. All Rights Reserved.
A method of preparing a detailed landslide-inventory map has been developed which provides the engineering geologist with
the basic information for evaluating and reducing landslide hazards or risk on a regional or community level. For each landslide,
the map depicts (1) state of activity, (2) certainty of identification, (3) dominant type of slope movement, (4) primary direction
of movement, (5) estimated thickness of material involved in land-sliding, and (6) date(s) of known activity. This information
is developed from interpreting aerial photographs and examining landslide features in the field. Although preparing detailed
landslide-inventory maps involves considerably more time and effort than landslide reconnaissance mapping, these maps are
directly useable by planners and decisionmakers as a basis for requiring site-specific investigations prior to development
or adopting land-use regulations.
Landslide susceptibility in San Mateo County
- E E Brabb
- E H Pampeyan
- M G Bonilla
Brabb, E. E., Pampeyan, E. H. and Bonilla, M. G. (1972) Landslide susceptibility in San Mateo County, California: U.S. Geol. Survey Misc. Field Studies Map 360
Preliminary photo-interpretation maps of landslide and other surficial deposits of 56 7.5-minute quadrangles, Alameda, Contra Costa, and Santa Clara Counties
- T H Nilsen
Nilsen, T. H. (1975) Preliminary photo-interpretation maps of landslide and other surficial deposits of 56 7.5-minute quadrangles, Alameda, Contra Costa, and Santa Clara Counties, California: U.S. Geol. Survey Open-File Report 75-277
Natural conditions that control landsliding in the San Francisco Bay region—an analysis based on data from the 1968–69 and 1972–73 rainy seasons: U
- T H Nilsen
- F A Taylor
- R M Dean
Nilsen, T. H., Taylor, F. A. and Dean, R. M. (1976) Natural conditions that control landsliding in the San Francisco Bay region—an analysis based on data from the 1968–69 and 1972–73 rainy seasons: U.S. Geol. Survey Bull., 1424, 35 pp
Earth flows-morphology, mobilization, and movement: U
- D K Keefer
- A M Johnson
Keefer, D. K. and Johnson, A. M. (1983) Earth flows-morphology, mobilization, and movement: U.S. Geol. Survey, Prof. Paper 1264, 56 pp
Geologic map of the Oakland Metropolitan Area
- R W Graymer
Graymer, R. W. (2000) Geologic map of the Oakland Metropolitan Area, California: U.S. Geol.
Survey, Misc. Field Studies Map 2342; http://geopubs.wr.usgs.gov/mapmf/mf2342/
Earthquake-induced landslide zones: In Seismic hazard zone report for the Oakland East 7.5- minute quadrangle
- R I Wilson
- M O Wiegers
- T P Mccrink
- W D Haydon
- J R Mcmillan
Wilson, R. I., Wiegers, M. O., McCrink, T. P., Haydon, W. D. and McMillan, J. R. (2002)
Earthquake-induced landslide zones: In Seismic hazard zone report for the Oakland East 7.5-
minute quadrangle, Alameda County, California, Seismic Hazard Zone Report 080, California
Geological Survey, pp. 25–43 and Plate 2.1.
Slope maps of the San Francisco Bay region, California; a digital database: U.S. Geol. Survey Open-file Report 98-766
- S E Graham
- R J Pike
Graham, S. E. and Pike, R. J. (1998) Slope maps of the San Francisco Bay region, California; a digital
database: U.S. Geol. Survey Open-file Report 98-766, 17 pp.; http:wrgis.wr.usgs.gov/open-file/
of98-766/index.html
Landslide hazard mapping in Tuscany, Italy—an example of automatic evaluation
- F Brunori
- N Casagli
- S Fiaschi
- C A Garzonio
- S Moretti
Brunori, F., Casagli, N., Fiaschi, S., Garzonio, C. A. and Moretti, S. (1996) Landslide hazard
mapping in Tuscany, Italy—an example of automatic evaluation: In Slaymaker, O. (ed.),
Geomorphic Hazards, pp. 56–67. Wiley, Chichester.
Map and map database of susceptibility to slope failure by sliding and earthflow in the Oakland area
- R J Pike
- R W Graymer
- S Roberts
- N B Kalman
- S Sobieszczyk
Pike, R. J., Graymer, R. W., Roberts, S., Kalman, N. B. and Sobieszczyk, S. (2001) Map and map
database of susceptibility to slope failure by sliding and earthflow in the Oakland area,
California: U.S. Geol. Survey, Misc. Field Studies Map 2385, 37 pp.; http://geopubs.wr.usgs.gov/
map-mf/mf2385/
Map showing locations of damaging landslides in Alameda County, California, resulting from 1997–98 El Niño rainstorms: U.S. Geol. Survey, Misc. Field Studies Map 2325-B
- J A Coe
- J W Godt
- D Brien
- N Houdre
Coe, J. A., Godt, J. W., Brien, D. and Houdre, N. (1999) Map showing locations of damaging
landslides in Alameda County, California, resulting from 1997–98 El Niño rainstorms: U.S.
Geol. Survey, Misc. Field Studies Map 2325-B; http://pubs.usgs.gov/mf/1999/mf-2325-b/
Landslides and cities—an unwanted partnership
- R J Pike
- D G Howell
- R W Graymer
Pike, R. J., Howell, D. G. and Graymer, R. W. (2003) Landslides and cities—an unwanted
partnership: In G. Heiken, R. Fakundiny and J. F. Sutter (eds.), Earth Science in the City—A
Reader, American Geophys. Union Special Publication 56, pp. 187–254.
Map showing areas of relative amounts of landslides in California
- D H Radbruch
- K C Crowther
Radbruch, D. H. and Crowther, K. C. (1970) Map showing areas of relative amounts of landslides
in California: U.S. Geol. Survey, Open-file Report OF 70-270, 36 pp.