Geoscience Reference
In-Depth Information
1 Introduction
Over the last 30 years, landslides had caused thousands of fatalities and damages,
costing of almost 10 billion USD in Asia (Chu et al.
2009
). This statistics shows
that landslide modeling as one of the most highly sought topics among the
researchers. Landslides in Malaysia mostly occur during the heavy monsoon
rainfall season. Also, anthropogenic factors such as deforestation and unplanned
developmental works play important roles in initiation of landslides.
According to Varnes (
1984
), ''past is a key for the future'' thus, landslide
inventory play the main role in spatial and temporal prediction of landslides.
Therefore, in recent years many statistical approaches have been developed to
measure the spatial correlation between landslide location as a dependent factor
and its conditioning factors (Akgun et al.
2011
,
2012
; Devkota et al.
2013
;
Pourghasemi et al.
2012a
,
b
,
2013a
,
b
,
c
; Pradhan
2013
; Pradhan et al.
2010a
,
b
,
2011
,
2012
; Tien Bui et al.
2012a
,
b
,
c
,
d
,
2013
; Zare et al.
2013
).
Spatial pattern in landslide inventory plays a vital role in predictive analysis.
Generally, landslides are frequently distributed in cluster pattern groups both in
space and time than a disperse pattern (Jarman
2006
). Cluster pattern can be
described as, high density of events occurring in specific location than other
locations. Moreover, the random simulation test of data distribution should reject
the hypothesis of independency among the events. In an earlier chapter, Keeper
(
1984
) concluded that landslides triggered by earthquakes, have more tendencies
to occur at well-defined location around the epicenter. Some chapters mentioned
that, the intensity of geological events in some areas, are not constant along the
area, then their cluster pattern cannot be considered as an evident (Bai et al.
2011
;
Oh and Lee
2011
).
Many statistical approaches have been used for spatial pattern analysis,
Ripley's K-function K(r) (Ripley
1976
), Poisson model (Zuo et al.
2009
). In this
chapter, a second-order statistics Nearest Neighbor Index (NNI) (Clark and Evans
1954
), was used to test the spatial nature pattern of landslides events in Kuala
Lumpur and surrounding areas. NNI method uses a ratio between two distances i.e.
nearest neighbor distance and mean random nearest neighbor distance that is
expected basis by chance. It is worth to mention that, in this article we used point
features representation, because it is widely used in landslide modeling and it ease
the analysis process simpler in a stochastic way (Stoyan
2006
).
2 Study Area
Kuala Lumpur and vicinity areas, plays a major role in economic and social
development in Malaysia. During the monsoon, the area receive high amount of
precipitation that weakness the slopes stability (Pradhan
2011
; Pradhan and Lee
2007
). The study area is enclosed geographically between 21560 and 31200N
