Remote sensed data for automatic detection of land-use changes due to human activity in support to landslide studies

Cristina Tarantino (), Palma Blonda () and Guido Pasquariello ()

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, 2007, vol. 41, issue 1, 245-267

Abstract: Slope instability studies appear to recognize a number of potential superficial slide-producing agents, which may be directly detected and monitored with Earth Observation (EO) data. The main objective of this work is to use conventional EO data and automatic techniques for providing land-use change maps useful in landslide prevention. The idea is to use the detection of changes in areas already involved in landslide events as a precursory sign of variations in the equilibrium status of the slope, independently from other natural triggering events, such as rain and seismic events. Attention is focused on man-induced surface changes, such as deforestation, urban expansion and construction of artificial structures. A historical set of 20 multi-temporal Landsat TM images, covering the period 1987–2000, was analyzed using a supervised change detection technique on a test site affected by slope instability phenomena located in the Abruzzo region in Southern Italy. A change image is obtained by comparing year-specific thematic map pairs. It contains useful information not only on the place where a transition occurred, but also on the specific classes involved in the transitions between two different years. The full set of change images is used to extract class-conditional transition probabilities, to evaluate variations in specific class distribution and the total number of changed pixels in time. Four classes and their transitions were considered in the analysis: (1) arboreous land, (2) agricultural land, (3) barren land, and (4) artificial structures. The quantitative analysis of the class-joint transition probability values of some specific class-transitions that may worsen slope stability showed that in an area prone to landslides the probability of landslide re-activation or first activation is higher where changes have occurred. Although based on a limited number of known events, such a result encourages extensive experimentation of the proposed technique on better documented landslide test sites. Copyright Springer Science+Business Media B.V. 2007

Keywords: Supervised change detection; Neural networks; Natural hazard; Land-use change; Landslides (search for similar items in EconPapers)
Date: 2007
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://hdl.handle.net/10.1007/s11069-006-9041-x (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:nathaz:v:41:y:2007:i:1:p:245-267

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11069

DOI: 10.1007/s11069-006-9041-x

Access Statistics for this article

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards is currently edited by Thomas Glade, Tad S. Murty and Vladimír Schenk

More articles in Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards from Springer, International Society for the Prevention and Mitigation of Natural Hazards
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().