An assessment on the use of Terra ASTER L3A data in landslide susceptibility mapping

https://doi.org/10.1016/j.jag.2011.08.005Get rights and content

Abstract

The main purpose of the present study is to evaluate the potential use of Terra ASTER data—the L3A DEM and its derivatives in landslide susceptibility mapping. For the purpose, an appropriate application site from the Western Black Sea region of Turkey—the Kelemen catchment area was selected. During the analyses, a two-stage comparative evaluation was carried out. In the first stage, the differences between the DEMs obtained from Terra ASTER L3A data and the conventional topographic data; and their first and second derivatives were investigated. Subsequently, different susceptibility maps were produced by using the DEMs and the topographic attributes obtained from both source of data in addition to the spectral information acquired from satellite sensor. According to the results of the comparative evaluations, a strong correlation between Terra ASTER L3A DEM and the conventional topographic data was obtained. However, depending on the increment of the degree of the derivative, an evident decrease in the spatial correlations was observed. On the contrary, the final model performance, prediction capacity, and the spatial performance statistics for the landslide susceptibility maps produced by using both source of data were found as very high and close to each other.

Highlights

► To minimize the data production load for landslide susceptibility maps, the potential use of Terra ASTER L3A data is investigated. ► A strong correlation between Terra ASTER L3A DEM and the reference DEM was obtained (r = 0.99; zASTER L3A  zReference HGK). ► Depending on the increment of the degree of the derivative, the spatial susceptibility differences between the maps produced from different data sources increase and the spatial correlations decrease. ► Among the topographic derivatives obtained from Terra ASTER L3A DEM, only the first ones could be used effectively in landslide susceptibility mapping—in case the acquisition quality of Terra ASTER imagery is high.

Introduction

Landslide susceptibility maps are of great importance for landslide hazard mitigation efforts throughout the World. For this reason, preparation of landslide susceptibility maps has been one of the most crucial subjects among the international geomorphology and engineering geology communities for the last decade. It is evident that the reliability of a landslide susceptibility map depends on not only the technique employed but also the data used. Therefore, in producing a regional-scaled landslide susceptibility map, substantial efforts are spent to collect well-documented landslide inventory, geological and topographic data of an area to be studied. These efforts obviously require reasonably lots of time and cost. To minimize this problem, remotely sensed data have been used extensively, because the relevant products constitute the most important sources when working in large areas and difficult field conditions such as mountainous regions. For the purpose, Terra ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) satellite data has been used in landslide evaluations in the last decade in particular (Crowley et al., 2003, Liu et al., 2004, Fourniadis et al., 2007, Santini et al., 2009). Crowley et al. (2003) investigated the utility of airborne and spaceborne remote sensing systems and digital elevation data for mapping hydrothermally altered rocks and other volcanic features that may contribute to potential debris flow hazards. As the result of the study performed by Crowley et al. (2003), comparisons of the imagery and digital elevation data sets acquired from both airborne and spaceborne sensors demonstrate the feasibility of remotely sensed data in evaluation of altered rock masses that constitute potential volcanic debris flow source areas. Liu et al. (2004) perform a regional assessment of landslide hazard in the Three Gorges area, China, based on Terra ASTER imagery data, including a stereo image-derived Digital Elevation Model (DEM) and multispectral reflective and thermal imagery, in combination with limited field investigation. According to Liu et al. (2004), the results of their work have demonstrated that Terra ASTER imagery constitutes a useful source of topographic and spectral information for regional landslide hazard mapping. The study regarding use of Terra ASTER data in producing landslide susceptibility map was also published by Fourniadis et al. (2007). The objectives of the research published by Fourniadis et al. (2007) are to develop a methodology for regional scale assessment of landslide hazard using Terra ASTER data, and to produce a landslide hazard map of the Wushan–Badong region in the Three Gorges in China. The results obtained by Fourniadis et al. (2007) indicate adequate correlations between the defined hazard classes and field confirmed slope failures, and represent the potential benefits of Terra ASTER imagery in landslide hazard assessment at regional scales. Another critical evaluation was done by Santini et al. (2009). Santini et al. (2009) emphasized that the choice of the most appropriate DEM processing procedure affects the implementation of countermeasures against landslides. The researchers performed a comparative study to investigate the influence of different terrain analysis procedures on the results of the slope stability model SHALSTAB (Shallow Landslide Stability) using remotely Terra ASTER DEMs.

The main purpose of the present study is to investigate the potential use of Terra ASTER L3A data in landslide susceptibility mapping in medium scale. For the purpose, a two-stage comparative evaluation was carried out. In the first stage, the differences between the DEMs obtained from Terra ASTER L3A data and the conventional methods, and their first and second derivatives were investigated. Spatial difference and correlation analyses were taken into account in this stage. Subsequently, different landslide susceptibility maps were produced by using the spectral information acquired from satellite sensor, and the DEMs and the derivatives obtained from both Terra ASTER L3A data and conventional topographic data. Similar comparison procedures, spatial difference and correlation analyses were also applied for the resultant susceptibility maps in the second stage of the study. As the study area, a landslide-prone region from the Western Black Sea region of Turkey, the Kelemen catchment area was selected (Fig. 1).

Section snippets

Study area

The Kelemen catchment area is located in the Ulus basin, which is the largest sedimentary basin of the Western Pontids (Tuysuz, 1999) (Fig. 2). In geological and tectonic framework, the catchment is located in the Istanbul–Zonguldak zone of the Western Pontids (Yilmaz et al., 1997). These sedimentary units having large areal extent in the region are grouped in the name of Ulus formation by Tuysuz (1999). The Early–Late Cretaceous aged Ulus formation (Tuysuz, 1999) covers 95% of the whole

Data

Topographic parameters such as altitude, slope gradient, slope aspect, slope curvatures constitute one of the fundamental factor groups in landslide susceptibility evaluations. All of these parameters are derived from elevation models. Commonly, 1:25,000 scaled elevation models in digital format are employed when producing regional scaled landslide susceptibility maps. In some countries such as Turkey, it is possible to find 1:25,000 scaled elevation models in digital format. Otherwise, an

Landslide susceptibility mapping

Landslide researchers have considered different techniques such as simple overlay (Gupta and Joshi, 1990, Pachauri and Pant, 1992, Gokceoglu and Aksoy, 1996, Turrini and Visintainer, 1998, Donati and Turrini, 2002, Ayenew and Barbieri, 2005), statistical analysis (Carrara et al., 1991, Guzzetti et al., 1999, Baeza and Corominas, 2001, Lee and Min, 2001, Ercanoglu et al., 2004, Suzen and Doyuran, 2004a, Suzen and Doyuran, 2004b, Ayalew and Yamagishi, 2005, Gokceoglu et al., 2005, Duman et al.,

Discussions and conclusions

The main purpose of this research is to investigate the potential use of Terra ASTER L3A data – the L3A DEM and its derivatives in landslide susceptibility mapping. For the purpose, an appropriate application site – the Kelemen catchment area was selected. Since, all of the catchment area locates in unique geological unit—Ulus formation, landslide conditioning factors raised from the geological attributes were extracted by implementing Terra ASTER L3A spectral data—SWIR and TIR channels.

Acknowledgements

This study was performed as a part of the joint research project executed by the General Directorate of Mineral Research and Exploration (MTA), Geotechnos Co., Ltd. Company, and Earth Remote Sensing Data Analysis Center (ERSDAC).

References (82)

  • I.G. Fourniadis et al.

    Landslide hazard assessment in the Three Gorges area, China, using ASTER imagery: Wushan–Badong

    Geomorphology

    (2007)
  • C. Gokceoglu et al.

    Landslide susceptibility mapping of the slopes in the residual soils of the Mengen region (Turkey) by deterministic stability analyses and image processing techniques

    Engineering Geology

    (1996)
  • C. Gokceoglu et al.

    The 17 March 2005 Kuzulu landslide (Sivas, Turkey) and landslide-susceptibility map of its near vicinity

    Engineering Geology

    (2005)
  • H. Gomez et al.

    Assessment of shallow landslide susceptibility using artificial neural networks in Jabonosa River Basin, Venezuela

    Engineering Geology

    (2005)
  • R.P. Gupta et al.

    Landslide hazard zoning using the GIS approach—a case study from the Ramganga Catchment, Himalayas

    Engineering Geology

    (1990)
  • F. Guzzetti et al.

    Landslide hazard evaluation: a review of current techniques and their application in a multi–scale study, Central Italy

    Geomorphology

    (1999)
  • A. Hirano et al.

    Mapping from ASTER stereo image data: DEM validation and accuracy assessment

    ISPRS Journal of Photogrammetry and Remote Sensing

    (2003)
  • D. Kawabata et al.

    Landslide susceptibility mapping using geological data, a DEM from ASTER images and an Artificial Neural Network (ANN)

    Geomorphology

    (2009)
  • S. Lee et al.

    Determination and application of the weights for landslide susceptibility mapping using an artificial neural network

    Engineering Geology

    (2004)
  • J.G. Liu et al.

    Landslide hazard assessment in the Three Gorges area of the Yangtze river using ASTER imagery: Zigui–Badong

    Geomorphology

    (2004)
  • A. Nandi et al.

    A GIS-based landslide susceptibility evaluation using bivariate and multivariate statistical analyses

    Engineering Geology

    (2010)
  • H.A. Nefeslioglu et al.

    Landslide susceptibility mapping for a part of tectonic Kelkit Valley (Eastern Black Sea region of Turkey)

    Geomorphology

    (2008)
  • H.A. Nefeslioglu 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

    Engineering Geology

    (2008)
  • F. Ocakoglu et al.

    Dynamics of a complex mass movement triggered by heavy rainfall: a case study from NW Turkey

    Geomorphology

    (2002)
  • A.K. Pachauri et al.

    Landslide hazard mapping based on geological attributes

    Engineering Geology

    (1992)
  • B. Pradhan et al.

    Landslide susceptibility assessment and factor effect analysis: backpropagation artificial neural networks and their comparison with frequency ratio and bivariate logistic regression modeling

    Environmental Modelling & Software

    (2010)
  • B. Pradhan et al.

    A GIS-based back-propagation neural network model and its cross–application and validation for landslide susceptibility analyses

    Computers, Environment and Urban Systems

    (2010)
  • L.C. Rowan et al.

    Lithologic mapping in the Mountain Pass, California area using Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data

    Remote Sensing of Environment

    (2003)
  • M. Santini et al.

    Pre-processing algorithms and landslide modelling on remotely sensed DEMs

    Geomorphology

    (2009)
  • E.A Sezer et al.

    Manifestation of an adaptive neuro-fuzzy model on landslide susceptibility mapping: Klang valley, Malaysia

    Expert Systems with Applications

    (2011)
  • H. Sonmez et al.

    Estimation of rock modulus: for intact rocks with an artificial neural network and for rock masses with a new empirical equation

    International Journal of Rock Mechanics and Mining Sciences

    (2006)
  • M.L. Suzen et al.

    Data driven bivariate landslide susceptibility assessment using Geographical Information Systems: a method and application to Asarsuyu Catchment, Turkey

    Engineering Geology

    (2004)
  • M.H. Tangestani

    A comparative study of Dempster–Shafer and fuzzy models for landslide susceptibility mapping using a GIS: an experience from Zagros Mountains, SW Iran

    Journal of Asian Earth Sciences

    (2009)
  • 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)
  • M.H. Vahidnia et al.

    A GIS-based neuro-fuzzy procedure for integrating knowledge and data in landslide susceptibility mapping

    Computers & Geosciences

    (2010)
  • W. Wang et al.

    Landslides susceptibility mapping in Guizhou province based on fuzzy theory

    Mining Science and Technology (China)

    (2009)
  • Y. Yamaguchi et al.

    ASTER instrument characterization and operation scenario

    Advances in Space Research

    (1999)
  • I. Yilmaz

    Landslide susceptibility mapping using frequency ratio, logistic regression, artificial neural networks and their comparison: a case study from Kat landslides (Tokat–Turkey)

    Computers & Geosciences

    (2009)
  • M. Abrams

    The advanced spaceborne thermal emission and reflection radiometer (ASTER): data products for the high spatial resolution imager on NASA's terra platform

    International Journal of Remote Sensing

    (2000)
  • Akgun, A., Sezer, E.A., Nefeslioglu, H.A., Gokceoglu, C., Pradhan, B., in press. An easy-to-use MATLAB program...
  • C. Baeza et al.

    Assessment of shallow landslide susceptibility by means of multivariate statistical techniques

    Earth Surface Processes and Landforms

    (2001)
  • Cited by (44)

    • Integrating object-based and pixel-based segmentation for building footprint extraction from satellite images

      2023, Journal of King Saud University - Computer and Information Sciences
    • Landslide susceptibility mapping using an automatic sampling algorithm based on two level random sampling

      2019, Computers and Geosciences
      Citation Excerpt :

      Gokceoglu et al. (2010) preferred 85% of whole data sets for training samples. Nefeslioglu et al. (2008), Oh and Pradhan (2011), San (2014), Nefeslioglu et al. (2012) and Ada and San (2018) used 80% of a training data set of a whole sample data set. Song et al. (2012) and (Chen et al., 2017) performed their studies using 70% of whole samples as training data.

    • CO-SEISMIC LANDSLIDE BASED VALIDATION OF SUSCEPTIBILITY MAPPING AFTER KAHRAMANMARAS EARTHQUAKES (FEB 6, 2023) IN AMANOS MOUNTAINS

      2023, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
    View all citing articles on Scopus
    View full text