An objective method to rank the importance of the factors predisposing to landslides with the GIS methodology: application to an area of the Apennines (Valnerina; Perugia, Italy)

https://doi.org/10.1016/S0013-7952(01)00087-4Get rights and content

Abstract

The present work aims at contributing to further our knowledge on the predisposing factors to landslides as well as proposing a method that allows us to weigh up, as objectively as possible, the influence that the various factors have on landslides in order to construct a realistic map of potential landslide hazards. In order to assess the differential incidence of very predisposing factor to landslides, maps of the various factors considered were drawn up with the aid of IDRISI software (1997). The factors were as follows: the distance from faults, parallelism between the fractures and the landslide scarps, land use, lithology, distance from the streams, orientation and steepness of slopes, orientation of layers compared to the slope. The global analysis of the different incidence of the factors analyzed on the landslides present in the study area was carried out by comparing the respective maps with that of the scarps; in this way the number of pixels forming the scarps which fell into the various classes of the maps was calculated. The result thus obtained can form the basis of an objective assignation of the different ratings to be attributed to the various factors under consideration. In fact, the analysis carried out in this study has shown that the factors act differently and, for every factor, only some of the classes considered have marked importance.

Introduction

Over the past few decades, awareness has grown regarding the need for greater harmony between the development of anthropical activity and the evolution of the natural environment. For this reason, scientists have shown an ever increasing interest in the problems concerning the hazards of natural phenomena and the risks they pose for anthropical structures (Varnes and Commission on Landslides and Other Mass Movements IAEG, 1984).

In the literature numerous examples of methods have been proposed to deal with the prediction of hazards resulting from natural phenomena, and with particular reference to the problem of landslides, hazard maps of varying complexity and reliability can be constructed (van Westen, 1993). In fact, there are some maps, such as those providing the distribution and characteristics of landslides, which are quite easy to construct and which only report active and dormant landslides recognisable in the territory Guzzetti and Cardinali, 1989, Guzzetti and Cardinali, 1990, Wieczorek, 1984, whereas there are others of greater complexity based on deterministic hazard analysis such as the study of safety factors Genevois et al., 1987, Hammond et al., 1992 or on statistical analysis such as multivariate analysis of the various factors which characterise unstable areas Carrara, 1983, Carrara, 1988, Carrara et al., 1990, Clerici and Dall'Olio, 1995, Guzzetti, 1993. An intermediate category may be considered the maps of potential hazards, or also susceptibility to landslide (Brabb, 1984), which examine the main predisposing factors to landslides and subdivide the territory into areas of variable hazards according to the incidence of these factors Anbalagan, 1992, Gee, 1992, Stevenson, 1977, Turrini and Visintainer, 1998. The greatest drawback of these maps is the subjectivity characterising the phases in which ratings are assigned to the various factors taken into consideration (van Westen, 1993).

The present paper aims at contributing to further our knowledge on the incidence of the predisposing factors to landslides as well as proposing a method that allows us to weigh up, as objectively as possible, the influence that the various factors have on landslides in order to construct a realistic map of potential landslide hazards.

The study area is located in Southeast Umbria, East of Spoleto (Perugia). It covers some 18 km2 to the west of the Nera River (Fig. 1).

Cretaceous to Miocene basin-fill sediments are exposed within the area, showing an increase in terrigenous material towards the top. Two thrust faults and a series of NNW–SSE striking folds can be traced throughout the area (Donati, 1997). Following this compressive tectonism, the region was uplifted and direct faults, striking transversally, developed.

In response to the active tectonism, a high relief geomorphology, ranging in altitude from 360–1460 m above sea level with deeply incised gorges developed.

Where softer marly–clayey facies are exposed, a gentler morphology has developed allowing use of the land for agricultural purposes. Calcareous lithologies however, have led to the development of very rugged terrain. The former lithologies are greatly affected by large landslides, mostly dormant at present. The whole area affected by mass movements, measured from the scarp to the body toe, represents almost 11.5% of the total area. Although there are few active landslides, they do, however, represent a serious threat to human activity and the fairly numerous infrastructures. The most frequent type of landslides are earth rotational slides and earth flows; at certain points landslide events are so widespread that the various landslide heaps are difficult to discern and the area is found to be generally unstable.

Section snippets

Work method

This work was based on a fairly detailed field survey to a scale of 1:5000; with the aid of photointerpretation of aerial views of scale 1:13,000, a geological–geomorpohological map was obtained.

The causative factors of the landslides considered were chosen after a careful bibliographical review of those most frequently cited by the authors and which, in theory, greatly affect the predisposition to instability in the territory Abramson, 1996, Brunsden, 1996, Rib and Liang, 1978, Van Westen, 1993

Steepness map and slope orientation map

The construction of these maps was practically automatic beginning from the Digital Elevation Model (D.E.M.), which, in its turn, was automatically constructed by digitalizing the contour lines with contour intervals of 25 m and intermediate contour lines and quoted points where the main contour lines were too wide. The classification used for the two maps is shown in Table 1, Table 2. In the same tables, the results of the analyses carried out are also collected.

As far as Table 1 is concerned,

Results and discussion

The overall analysis of the different incidences of the factors analysed on the landslides present in the study area was carried out through a series of crosstabulations [CROSSTAB] between the respective maps and that of the scarps. In this way, it was possible to reveal the number of pixels forming the scarps which fell into the various classes of the different maps and, consequently, to calculate the percentage of area of the different classes affected by landslides (normalised values).

Conclusions

The study has demonstrated that, at least for the study area, the factors considered should, in theory, have had a strong influence on the formation of landslides, but in reality they had a different incidence. Furthermore, with the proposed method it was also possible to demonstrate that, for each factor considered, only some of the classes had a good influence, whereas others exercised a less important influence.

For the study area, in particular, the most important predisposing factor is,

Acknowledgements

The authors wish to express their great appreciation to the Umbria Region for their help in the field survey and the important data provided, and to the anonymous referee.

References (29)

  • R. Anbalagan

    Terrain evaluation and zonation mapping in mountainous terrain

    Engineering Geology

    (1992)
  • M.C. Turrini et al.

    Proposal of a method to define areas of landslide hazard and application to an area of the Dolomites, Italy

    Engineering Geology

    (1998)
  • L.W. Abramson

    General slope stability concepts

  • E.E. Brabb
  • D. Brunsden

    Mass movement, the research frontier and beyond: a geomorphological approach

    Geomorphology

    (1996)
  • A. Carrara

    Multivariate models for landslide hazard evaluation

    Mathematical Geology

    (1983)
  • A. Carrara
  • A. Carrara et al.
  • A. Clerici et al.

    La realizzazione di una Carta della stabilità potenziale dei versanti mediante tecniche di Analisi Statistica Multivariata e un Sistema d'Informazione Geografica

    Geologia Tecnica e Ambientale

    (1995)
  • A. Clerici et al.

    La realizzazione di una carta della stabilità dei versanti mediante l'uso di un Sistema di Informazione Geografica

    Geologia Tecnica e Ambientale

    (1993)
  • J.C. Davis

    Statistics and Data Analysis in Geology

    (1986)
  • Donati, L., 1997. Studio della franosità dell'area Meggiano—Cerreto di Spoleto (Valnerina, Umbria) mediante la...
  • J.R. Eastman

    Idrisi for Windows User's guide Vol. 2.0

    (1997)
  • M.D. Gee
  • Cited by (234)

    View all citing articles on Scopus
    View full text