Author's personal copy International Journal of Applied Earth Observation and Geoinformation 12 (2010) 466–476 Contents lists available at ScienceDirect International Journal of Applied Earth Observation and Geoinformation journal homepage: www.elsevier.com/locate/jag Satellite-based damage mapping following the 2006 Indonesia earthquake—How accurate was it? Norman Kerle * Faculty of Geo-information Science and Earth Observation, Department of Earth Systems Analysis, Twente University, Hengelosestraat 99, P.O. Box 6, 7500 AA Enschede, The Netherlands article info Article history: Received 14 July 2009 Accepted 27 July 2010 Keywords: International Charter Disaster Damage maps Bantul abstract The Yogyakarta area in Indonesia suffered a devastating earthquake on 27 May 2006. There was an imme- diate international response, and the International Charter “Space and Major Disasters” was activated, leading to a rapid production of image-based damage maps and other assistance. Most of the acquired images were processed by UNOSAT and DLR-ZKI, while substantial damage mapping also occurred on the ground. This paper assesses the accuracy and completeness of the damage maps produced based on Char- ter data, using ground damage information collected during an extensive survey by Yogyakarta’s Gadjah Mada University in the weeks following the earthquake and that has recently become available. More than 54,000 buildings or their remains were surveyed, resulting in an exceptional validation database. The UNOSAT damage maps outlining clusters of severe damage are very accurate, while earlier, more detailed results underestimated damage and missed larger areas. Damage maps produced by DLR-ZKI, using a damage-grid approach, were found to underestimate the extent and severity of the devastation. Both mapping results also suffer from limited image coverage and extensive cloud contamination. The ground mapping gives a more accurate picture of the extent of the damage, but also illustrates the chal- lenge of mapping a vast area. The paper concludes with a discussion on ways to improve Charter-based damage maps by integration of local knowledge, and to create a wider impact through generation of customised mapping products using web map services. © 2010 Elsevier B.V. All rights reserved. 1. Introduction A major earthquake (M w = 6.3) struck Eastern Java in Indone- sia on 27 May 2006 (designated as GLIDE: EQ-2006-000064-IDN). With an epicenter near the major Javanese city of Yogyakarta in the densely populated Bantul district (>1600 people per km 2 ; BAPPENAS, 2006), approximately 5800 people died, an estimated 140,000 houses were destroyed, and a further 260,000 damaged (Koseki et al., 2007). The affected area had not experienced an earthquake of comparable magnitude in over 100 years, and was thus ill prepared. Simple brick buildings, the principal housing type in the affected area, could not withstand the motion and readily collapsed (Werkle and Zahn, 2008). The established dis- aster management hierarchy at national, provincial and district levels was activated and reacted rapidly, efforts that were sup- ported by the International Charter “Space and Major Disasters” (www.disasterscharter.org) that had seen a growing number of annual activations since 2000. By the time of the Bantul earthquake nearly 100 response operations had made image-based damage mapping a routine process. The imagery acquired was predom- * Tel.: +31 53 4874476; fax: +31 53 4876335. E-mail address: kerle@itc.nl. inantly processed by the German Space Agency (DLR)’s Centre for Crisis Information (ZKI) and the UN Institute for Training and Research (UNITAR)’s Operational Satellite Applications Programme (UNOSAT), with both organisations producing a variety of different damage maps. In addition, a major field survey campaign to map structural damage was launched by the University of Gadjah Mada (UGM) in Yogyakarta, leading to an extensive database of affected buildings in parts of the disaster area. The data that include detailed information on building type, function and construction material have recently become available, thus the principal purpose of this paper is an evaluation of the accuracy and completeness of the image-based damage maps. Given the inherent difficulty of detect- ing and quantifying structural damage in image data (e.g. Arciniegas et al., 2007; Saito et al., 2004) even in high spatial resolution Ikonos or Quickbird scenes, exacerbated by the time pressure to deliver these maps rapidly, the purpose of this paper is not to catalogue image analysis mistakes. It is clear that an experienced image ana- lyst with sufficient time, and especially in retrospect, will be able to produce damage maps that exceed in accuracy what can be done following a Charter activation, where the focus by necessity has to be on rapid results. Hence, the aim here is to understand where the mapping was accurate, where and why it failed, and how different mapping approaches performed but also may serve different user groups. 0303-2434/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jag.2010.07.004