Transcript Geospatial Information for Landslide Hazard Assessment in Lombok

Geospatial Information for
Landslide Hazard Assessment
in Lombok, Indonesia
Yukni Arifianti (CVGHM, GAI)
Akbar Cita (CGS, GAI)
Heryadi Rachmat (GM, GAI)
Matthias Dorn (BGR)
Malaysia Geospatial Forum
Melaka, 6 - 7 March 2012
Outline
I.
II.
III.
IV.
V.
VI.
Introduction
Landslides in Indonesia
Landslides in Lombok
Geospatial data
Process analysis
Discussion
I. Introduction
• Landslide is the down-slope movement of the rock, soil and/or vegetation
due to gravity in a process that can be abrupt collapse or in slow gradation
slides.
• It is significant because the distribution spread across the country and the
most intensive of occurrence.
• It can occur simultaneously with other geological hazards, such as
earthquake and volcanic eruptions
• The main objective is to develop and test practical hazard assessment
methodologies in Lombok Island, especially for landslide.
CONTROLLING FACTORS
RAINFALL
VEGETATION
LANDSLIDE
GEOLOGICAL CONDITION
MORPHOLOGY/SLOPE
EARTHQUAKE
•
•
1.
2.
3.
4.
5.
GEORISK PROJECT
The Geological Agency of Indonesia (GAI), Ministry of Energy and Mineral
Resources (MEMR) and BGR (German Federal Institute for Geoscience and
Natural Resources) set up a joint project to support the implementation of
the results of georisk assessment methodologies in national, provincial
and local georisk mitigation strategies.
Partners and cooperation:
GAI
Ministry of Home Affairs (MoHA)
Indonesian National Board for Disaster Management (BNPB)
Provincial and Local Government
GIZ with “Decentralization and Good Local Governance”
II. Landslides Hazards in Indonesia
• Indonesia is one of the most prone country to geological hazards in the
world: Volcanic Eruptions, Earthquake, Tsunami, LANDSLIDES
• Developing country: need more spaces for infrastructures, industries,
energy sectors, settlement
• There are 199 landslides occurrence in 2010, killed 470 people (CVGHM)
Geological condition of Indonesia
(CVGHM, GAI)
LANDSLIDE EVENTS IN INDONESIA (1990 – 2009)
(CVGHM, GAI)
Spatial Planning in Landslides Prone Area
(CEG, GAI)
III. Landslides Hazards in Lombok
• Lombok is an island in West Nusa Tenggara Province of Indonesia with an
area of 5.435 km2 with 3.098.480 inhabitants
• Located at 115° 45’ – 116° 45’ E and 8° 00’ – 9° 00’ S
• It is as an internationally alternative eco-tourism destination after Bali.
• Field checking was carried out in 2011 and found 4 new landslides from 13 possible
landslides that were detected by ASTER.
• The investigation classified 147 landslides are of natural origin and 47 landslides are
man-made triggered.
• The type of landslide are mainly rotational and translational.
Unconsolidated pumice
Translational @ Krujuk
Mining area
Man-made landslide
(Heryadi Rachmat, 2011)
Landslide Susceptibility Assessment (Technical Workflow)
IV. Geospatial data
Data sources
Landslide point
•Landslide points from GAI and local government.
• Additional landslide inventory using ASTER images and Google Earth.
Landslide factors
• Lithology from GAI
• Updated lineament map using the 1995 Landsat TM image.
• Land cover data derived from 2004 and 2009 Landsat ETM+ images.
• Topographic features: slope angle, slope aspect, flow direction, elevation
derived from ASTER GDEM data.
• Vegetation Indices; Normalized Difference Vegetation Index (NDVI) derived
from ASTER and MODIS images.
Landslide’s Geo-factors
Landslide
points
V. Process analysis
• Image processing was done with the ENVI 4.8 software. Other data
layers listed above were processed and added to a Geo-database with
Arc GIS 10 software.
• A statistical approach using probability and weighting methodology
was applied to extract standardized weights for the different layers
• The spatial data analysis and integration for landslide assessment can
be divided into three steps (Teerarungsigul, et.al, 2011):
1. To overlay the landslide distribution with factors
2. To perform the probability analysis and weighting.
3. To produce the landslide susceptibility map
1. To overlay the landslide distribution with factors
2. To perform the probability analysis and weighting
• Reliability probability (RP) was calculated by the percentage area of factors
corresponding to landslides
Σ % number of landslide
RP =
Σ % area in the same classes
• Accountability probability (AP) was calculated by the total landslides population
accounted for each factor.
AP =
Σ % number of landslides in classes having a pobability ratio ≥ 1
Σ % landslide point over the entire study area
Probability analysis and weighting
Factor
RP
W(RP)
AP
W(AP)
Weight
Slope
3.86
22
76.65%
14
18
Lineament
1.81
14
79.70%
19
17
NDVI
1.18
3
91.96%
22
13
Elevation
2.61
19
54.27%
6
13
Lithology
1.23
6
79.40%
17
12
Flow direction
1.39
11
63.82%
8
10
Aspect
1.24
8
69.54%
11
10
Land-use
2.15
17
44.72%
3
10
3. To produce the landslide susceptibility map
VI. Discussion
• The analysis resulted in slope, lineament and NDVI as the most influence
factors. The slope factor ranges from 45 to 50 while dense forest is the landuse class that has a high probability of landslides. Landslides in Lombok
mostly occurred on the mountainous and hilly areas.
• Lithology and lineament was examined to find out how geological factor
affect the landslides. The analysis demonstrates that lineament is more
affected. It creates weak zone that vulnerable to landslides. This shape built
by volcanic and fault activities. The presence of pumice as unconsolidated
rock proceeds as a weak layer for landslides movement.
• Geospatial Technology of Geographic Information System (GIS) and Remote
Sensing can ease landslide hazard assessment. It provides a high efficiency
and optimizes time resources. However a field work needs to be done to
validate the analysis.
THANK YOU…
[email protected]
[email protected]