Use of Landsat Images for the Identification of Damage
Due to the 1999 Kocaeli, Turkey Earthquake
Miguel Estrada*, Fumio Yamazaki**
*Graduate Student, **Associate Professor, Institute of Industrial Science,
The University of Tokyo, Bw304, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505
Tel: +81 (3) 5452-6388 Fax: +81 (3) 5452-6389
E-mail: miguel@rattle.iis.u-tokyo.ac.jp, yamazaki@iis.u-tokyo.ac.jp
Masashi Matsuoka
Chief Research Engineer, Earthquake Disaster Mitigation Research Center,
RIKEN, 2465-1 Mikiyama Miki, Hyogo 673-0433, Japan
Tel: +81 (794) 83-6632 Fax: +81 (794) 83-6695
E-mail: matsuoka@miki.riken.go.jp
Key Words
The 1999 Kocaeli Earthquake, Building damage detection, subsidence, fire.
Abstract
In this study, remote sensing satellite images are used to identify the affected areas due to the 1999 Kocaeli, Turkey Earthquake with a magnitude of 7.4. The aim of this study is to identify the hard-hit areas in different levels of damage, sunk areas into the sea and areas affected by fire. The analysis has been carried out in two different cities, Gölcük and Adapazari, where these types of damage were seen. The data that have been used are optical remote sensing images taken by Landsat/TM satellite on March 27 and August 18, 1999, before and after the earthquake, respectively. For the detection of burned and sunk areas a spectral comparison of the Landsat/TM is carried out. In order to detect the different levels of damages in the urban areas, a comparison of the ratio between different bands of the pre- and post-event images is conducted. In addition, principal components analysis is conducted for each one of the sets of images and later compared. In order to calibrate the results of these comparisons the product images are geographically corrected and mapped, afterwards they are compared with the ground truth data that are in a GIS system.
Introduction
The identification of damage due to large earthquakes is a vital issue to grasp the level and extension of the hard-hit areas. The evaluation can be conducted through a field reconnaissance survey. Even though, the field survey has a high accuracy it requires a lot of resources and time. In the event of large natural disasters, like a destructive earthquake, a fast assessment of the damaged areas is required to send off rescue teams and help. Also the awareness of the extension of damage can help to outline recovery plans. Recently remote sensing technology has become a tool in damage identification after the occurrence of natural disasters like floods, landslides or earthquakes (e.g. Matsuoka and Yamazaki, 1999). The identification of damage from a large area gives vital information that authorities can use to plan rescue procedures as well as to draw a general idea of the magnitude of the damage. The location of different types of damage like fire outbreak, ground settlement and building damage using optical remote sensing data is considered in this study.
In order to identify the different kinds of damage, the comparison of optical satellite images taken before and after the earthquake is conducted. In this study the area around Gölcük city in Turkey has been focused. First, image-to-image registration was carried out to match the images. To detect the fire outbreak we compared the profiles along X-axis and Y-axis of the pre- and post-event images. For the detection of the sunken area the histograms of the infrared band are matched and then the result images are compared. For the detection of different levels of damage an analysis in the visible region is conducted as well as Principal Component Transformation (Yasuoka, 1990). Data for the 1999 Kocaeli Earthquake are: Date: August 17th, 1999. Time: 00:01:38.56 (UTC).
Mw = 7.4. Epicenter: Lat. = 40.639N, Long. = 29.830E (Fig. 1). Hypocentral depth: 17 km. More than 200,000 buildings were lightly to heavily damaged. More than 17,000 people were killed and almost 44,000 were injured.
Landsat/TM Image
The data in this research are remote sensing images from Landsat/TM satellite taken over the affected area due to the earthquake. The images have been taken on 27 of March 1999, pre-event image, and on 18 of August 1999, for the post-event image. The composite color images of the pre-event and post-event (751) are shown in Figure 2(a) and 2(b), respectively. The images cover an area of 185 km by 154 km. Before making comparison, these two sets of images have to be registered; it means the pixels in both images must represent the same geographic location. To make an image-to-image registration one of them is chosen as a master to which the other, known as the slave, is to be registered. In this study the master image is the pre-event image and the slave image is the post-event image. The image registration has been conducted with the total area of the images and defining 250 ground control points deployed on the extension of the images. Rotation, scaling and translation method was used for warping the slave image and nearest-neighbor method was used for resampling.
Figure 2. Composite color images (751). Raw images.
Identification of Damage Caused by Fire
To identify the areas affected by fire we compare the profiles of the pre- and post-event images. The profile represents the distribution of the digital number (DN) of certain band along a strip of the image. This strip can be taken over the X-axis or Y-axis. For this comparison the band 5 (mid infrared), band 6 (far infrared or thermal band) and band 7 (mid infrared) have been used. Figure 3 shows the profiles of the DN of band 6 along the X-axis and Y-axis. Figure 1. Map of Turkey and the epicenter of the earthquake.
Figure 3. Profiles of the pre- and post-event images of the band 6. Notice the peak values that
represent the area around Tüpras refinery.
It can be observed (Figure 3) that there is a pattern in the profile before the earthquake as well in the after one. But in the region between sample 100 and sample 200, in the X-axis, and between the line 500 and 600, in the Y-axis, there are peak values, which represent the high temperature. Since the coordinates of this area are known, we can identify it in the image. This area corresponds to the Tüpras refinery, which suffered from a very large fire. Figure 4 shows the composite color image (752) as well as an aerial photo of the area around of the refinery.
Figure 4. Color composite (752) images around Tüpras refinery.
