Understanding the earthquake multi-hazard and risk using remote sensing and geographic information system in Pleret, Yogyakarta, Indonesia.

Abstract

Situated on the three junctions of active tectonic plates, Indonesia archipelago has the high probability of the disaster triggered by natural hazards (DTNH) such as earthquake, tsunami, landslide, flood, volcano eruption, and forest fires. Among the DTNH, earthquake, and tsunami caused the highest number of casualties and the highest level of economic losses. Several great earthquakes hit Indonesia and claimed more fatality than any other disasters. Furthermore, the earthquake event can directly generate the other secondary hazards which cause the worst impact on the local community. As one of the developing countries, Indonesia typically has the high population density, poor condition in economy, politic, and education which potentially increase the vulnerability level to natural hazard and local community risk.

Almost all regions in Indonesia, including Pleret situated about 10 km southeast of Yogyakarta, Java Island, are prone to earthquakes. Located approximately 200 km in the north of Sunda Megathrust, this area was often struck by big earthquake greater than 5 Richter scale. The most recent earthquake hit Yogyakarta was at 5.55 am local time, 27 May 2006. This earthquake was located inland 25 km southeast part of Yogyakarta with the focal depth only 35 km. This earthquake badly affected Yogyakarta and surrounding area, including Pleret.

Recently, twelve years after the earthquake, the population of Pleret has increased significantly. The population growth of Pleret Sub-District reached 1.29% and 1.59% in 1990 to 2000 and 2000 to 2010, respectively. As the population increases, the property and housing demand are growing up as well. However, due to lack of the spatial planning in Pleret, the local people have been building new houses without proper planning to reduce the effect of seismic shaking. Furthermore, some buildings are located on unstable slopes or in the more vulnerable areas.

The multi-hazard and risk study need to be conducted in Pleret. The dense population and the increasing of the improper new buildings in the middle part of Pleret which has a complex geological structure (fault, fold, and joint) can amplify the effect of ground motion. Additionally, the west part of Pleret is covered by the dense material from Mount Merapi which potentially amplifies the ground motion and prone to liquefaction. Lastly, the eastern part of Pleret is also prone to the coseismic landslide as this has relatively more unstable slope due to the extensive erosion and extreme relief.

The main objective of this Ph.D. research is to assess the multi-hazard and risk in Pleret Sub-District and surrounding areas using the remote sensing and geographic information system (GIS). The combination approach between the remote sensing data acquisition technique, fieldwork, structure from motion based on the GIS platform were applied to conduct this study. This study is divided into five major parts, namely: the outcrop study, the coseismic landslide, the liquefaction, the vulnerability level of local community, and the multi-hazard analysis.

The unchartered faults in the study area were resulted from the outcrop study by using the fieldwork observation and structure from motion method. The detail of geological information was also obtained by using the remote sensing technique and the field observation. Furthermore, the relatively prone areas to the earthquake hazard due to their proximity to the faults can be identified. The coseismic landslide map was produced to support the multi-hazard and risk analysis. The coseismic landslide map was generated by following the Grade 2 analysis of Mora and Vahrson, 1994 due to the absence of the landslide data inventory. This map was derived from the relief, lithology, soil humidity, rainfall intensity, and earthquake intensity.

The liquefaction map was generated by using the combination of secondary data and primary data analysis. The ground shear-strain analysis was used to anlyse the liquefaction probability since the CPT and SPT data is unavailable in the study area. Furthermore, the detailed land use data of the study area were generated by applying the Anderson land use and land cover classification through the interpretation of Quickbird imagery. This data was used to support the multi-vulnerability study in this study. The general equation of building collapse probability was also produced and was applied in the other similar areas. Through the statistical analysis, the best combination of house feature (building structure, roof material, lithological characteristics, and the distance from the earthquake epicentre) could be identified.

Lastly, the multi-hazard and risk analysis were generated by combining the multi-hazard and multi-vulnerability. The result showed that the middle part of the study area consisted of South part of Wonokromo and Pleret Villages, the north part of Segoroyoso Village, and the middle part of Bawuran and Wonolelo Villages. The fact showed that these areas were more vulnerable to earthquake and any other related secondary hazards than other areas in the study area.