Building Damage Information System (BDIS):

Master Thesis

submitted within the UNIGIS MSc. programme at the Centre for Geoinformatics (Z_GIS)

Salzburg University, Austria

under the provisions of UNIGIS joint study programme with Goa University, India

Building Damage Information System (BDIS):

Application of BDIS in Banda Aceh City - Indonesia After Sumatra Earthquake

by

Edy Irwansyah

523107

A thesis submitted in partial fulfilment of the requirements of the degree of

Master of Science (Geographical Information Science & Systems) – MSc (GISc) Advisors:

Dr. Shahnawaz and Dr. Mahender Kotha Jakarta, Indonesia, December 2008

2 

Abstract

A massive earthquake shook Sumatra Island on December 26^th, 2004 (magnitude 9.0 on the Richter scale). The epicenter was geographically located at 3.298^o N Latitude and 95.779^o E Longitude. As an impact of the earthquake, that around 19 percent of the approximately 820,000 building units in the affected districts suffered an average of about 50 percent damage while around 14 percent are in totally destroyed condition. In this study, high resolution IKONOS Imagery, aerial photograph and geographic information system (GIS) data applied for development of building damage information system (BDIS) in Banda Aceh City with specific location at Setui village.

Damage database created as a relational database management system (RDBMS) and based on the map, level of building damage in BAC can be classified in five class and continue re-classification with visual image interpretation based on European Microseismic Scale (EMS) classification

In This Study, age of each building calculated with formula: BCY – PR = BE, age of individual building calculated with constant value of resistance (CR = 0.1) with the aim to find number of earthquake resistance using formula: BE * CR = NER. Using value of NER, present building earthquake resistance can be calculated with following formula:

BER - NER = PER in where BCY = year of building construction, PR = present year, BE = age of building, CR = constant value of resistance, NER = number of earthquake resistance (in Richter scale), BER = building earthquake resistance (in Richter scale) and PER = present building earthquake resistance (in Richter scale). Calculation process conducted using GIS software and updated of the BDIS database. Result of calculation and selection of each building with earthquake resistance number more than 5 in Richter scale

From total 27,796 building unit in BAC, 35 percent or 9,842 units are classified as totally destruction (G5), 29 percent or 8,075 units on the slightly till moderate damage condition (G4-G2) and 36 percent or 9,879 another units just flooded upon tsunami. One floors building with number of people less than 10 dominating in Setui village reaching around 80 percent and generally experience damage at level G1 and G2, more than 35 percent of buildings have size between 101 and 200 m2 and around 88 percents consist of 2 – 5 rooms. Type of building usage predominated by non commercial building with more than 90 percents. The ages of buildings in Setui village are between 3 until 40 years. Building with commercial function and located along of primary street have age older than 30 years with size of building generally wider than others part and have level of low earthquake resistance (less than 5 in Richter Scale)

Three dimensional modeling for visualization and data sharing of building damage especially in Setui applied using Google Earth (GE) application. GE application have many excellences because can be done by broader user using online and also offline with small compressed KML/KMZ file.

Key words: Building damage; Building damage information system; Earthquake;

IKONOS imagery and aerial photograph; Google earth.

Science Pledge

By my signature below, I certify that my thesis is entirely the result of my own work. I have cited all sources I have used in my thesis and I have always indicated their origin.

Jakarta-Indonesia, December 26^th, 2008

(Place, Date) (Signature)

4 

Acknowledgements

First, I would like to give my sincere thank you to Spatial Information and Mapping Center (SIM-Center) of Bureau of reconstruction and rehabilitation of Nanggroe Aceh Darussalam and Nias (BRR- Aceh Nias) who provide IKONOS and aerial photograph data. This data was collected in the framework of cooperation between BRR and USAID/Indonesia tsunami reconstruction activities under Aceh Governance Development Program (AGDP) and Aceh-Nias Governance Enhancement Program (ANGEP).

Special thanks for Dr. Alit Merthayasa Director of Center for Local Governance Innovation and Dr. Kabul Sarwoto , Program Manager of AGDP and ANGEP who giving the opportunity to be involved in the program as a GIS expert and permission to use the data.

I would like to give my sincere thanks to my advisors Dr. Shahnawaz (Salzburg University) and Dr. Mahender Kotha (Goa University). Especially thanks to Dr Shahnawaz, I feel very happy discuss with you through online or offline which I am very lucky because I get guidance directly in your visit to Jakarta, Indonesia and in several previous meetings. Thanks too for Dr Mahender Kotha although we can not interact more in discussion but you give more knowledge through your module along the program.

Thank you for special beloved my wife Dyah Candrasari who always supports to achieve my ideals and my daughter Rania Ghaisani Puspawangi who have lost a lot of time together with me during the program. Special thanks to my parent and my parent in law without your support and trust, I would not have this wonderful experience in my life.

Thanks to all friends UNIGIS India 2006 student especially Salahudin Sayyed and Farook Nathire, It is great prides that can be meet you in this program and study together through online facility. Thanks for your helps.

Table of Contents

Abstract………..

Science Pledge………

Acknowledgments……….

Table of Contents………

List of Tables ……….

List of Figures and Maps………

1. Introduction ………...

1.1. Background ………

1.2. Objectives ………..

1.3. Area of Study………..

1.4. Physical Setting of Banda Aceh City……….

1.4.1. Morphology and Geology………

1.4.2. Climate and Rainfall …… ………..

1.4.3. Landuse Change and Infrastructure Inundated ………….. ……

1.4.3.1. Landuse in Banda Aceh ………

1.4.3.2. Impact of Tsunami to Landuse Changes………

1.4.3.3. Inundated City Infrastructure ………

1.5. Population……….………..

2. Earthquake, Tsunami and Damages ……….

2.1. Tectonic and Sumatra Fault System (SFS)……….

2.2. Sumatra Earthquake and Indian Ocean Tsunami (IOT) ………

2.2.1. Sumatra Earthquake ………..

2.2.2. IOT and Level of Inundation ……….

2.3. Infrastructure Damage in BAC………...

2.3.1. Infrastructure Damage ………

2.3.2. Building Damage Assessment ………

2.4. Remote Sensing and GIS for Damage Assessment ………

2.4.1. Building Extraction from High Resolution Image ……….

2.4.2. Remote Sensing and GIS for Damage Assessment ………

3. Methodology ………..

2 3 4 5 8 9 11 11 12 13 13 13 15 15 15 16 18 20 22 22 24 24 25 27 28 30 31 31 32 35

6  3.1. Data Availability and Sources ………

3.2. Data Extraction Using Pre and Post Tsunami Imagery ……….

3.2.1. Preparation of Pre and Post Tsunami Imagery ………..

3.2.2. IKONOS Pan Sharpening ……….

3.2.3. Data Extraction Through Visual Image Interpretation ………….

3.3. Map Preparation ……….

3.3.1. Preparation of Inundation Map………. …………..

3.3.2. Preparation of Topography Map … ………

3.3.3. Preparation of Elevation Map ……….

3.3.4. Preparation of Infrastructure Map and Attribute Data……….

3.4. Development of Building Damage Database. ………..

3.4.1. Data Integration in GIS …... ………..

3.4.2. Development of Damage Database ……….. ………

3.5. Classification of Building Damage ……….. ………

3.5.1. Damage Map Construction (Grid Method)……….

3.5.2. Classification of Building Damage ….. .……….

3.6. Development of Setui Building Damage Information System………

3.6.1. Data Availability and Pre Processing ………...

3.6.2. Setui Building Damage Database ………

3.6.3. Building Earthquake Resistance (BER)………

4. Result …….. ……….

4.1. Building Damage in BAC ……….

4.1.1. Building Damages and Distance from Coastline……….

4.1.2. Building Size Classification……… ………

4.1.3. Building Damage in Inundation Area ……….

4.2. Setui Building Damage Information System………

4.2.1. Number of Floor and Level of Damage ……….………

4.2.2. Building Size and Level of Damage ……….……….

4.2.3. Number of Room and Level of Damage ……….……….

4.2.4. Building Usage and Level of Damage ……….………

4.2.5. Number of People and Level of Damage ………

4.3. Building Earthquake Resistance ………..

4.3.1. Spatial Distribution of Building Age ………..

35 37 37 37 38 41 41 41 42 42 44 44 46 46 46 48 49 49 50 51 54 54 54 56 59 61 61 62 64 66 68 69 69

4.3.2. Building Earthquake Resistance ……….

4.4. Spatial Pattern and Cost of Building Damage ……….

4.4.1. Distribution of Building Damage in BAC………

4.4.2. Spatial Pattern of Building Damage ………

4.4.3. Cost of Building Damage ………

5. Damage Data Sharing Through Internet ………

5.1. Technology and Application of Google Earth (GE)……….

5.2. Conversion of Building Data to KML Format ………

5.3. Visualization and Modeling of Building Damage Data ………..

5.4. Google Earth Data Sharing ………..

6. Conclusion ……….

7. References ……….

70 72 72 73 76 79 79 80 81 83 84 85

8 

List of Tables

Table 1.1 Landuse Year 2005 of Band Aceh City ………

Table 1.2. Landuse Change in BAC as Impact of Tsunami………..

Table 1.3 Number of Infrastructure Damage in BAC………

Table 1.4 Area and Population in Kota Banda Aceh year 2004 and 2005 ………

Table 2.1 Summary of Damage and Losses to Infrastructures ……….

Table 3.1 Data Availability and Sources ………..

Table 3.2 Comparison of EMS, 1998 and JICA Study Team Classification ………

Table 4.1 Averages of Building Distance from Coastline in Each Level of Damage …..

Table 4.2 Number of Building in Each Different Building Size and the Level of Damage Table 4.3 Number of Building in Different Inundation Area and the Level of Damage…

Table 4.4 Number of Building Unit and Level of Damage in Each Number of Building Floor ………

Table 4.5 Number of Building Unit and Level of Damage in Each Building Size ………

Table 4.6 Number of Room in Each Building Unit and Level of Damage ………

Table 4.7 Number of Building Based on Type of Usage and Level of Damage …………

Table 4.8 Number of Building in Each Class of Number of People and the Level of Damage ………..

Table 4.9 Number of Building Unit Based on Type of Usage and The Level of Damage.

Table 4.10 Damage Cost of the G5 and G4 Damage Level ………..

Table 4.11 Damage Cost of the G3 Damage Level ………

Table 4.12 Damage Cost of the G2 Damage Level ………

15 16 18 21 28 35 48 55 57 59

62 63 65 66

69 72 77 77 78

List of Figures and Maps

Figure 1.1 Map of Banda Aceh City and Surrounding ………...

Figure 1.2 Landuse of BAC Year 2003 (Pre Tsunami) ………

Figure 1.3. Landuse of BAC Year 2005 (Post Tsunami) ………

Figure 1.4 Inundated City Infrastructures ………..

Figure 1.5 Inundated Residential Area ………..

Figure 2.1 Position of Indonesia Island in Eurasian Plate ……….

Figure 2.2 Seismo-tectonic of Sumatra Island ………..

Figure 2.3 Epicenter of Sumatra Earthquake ……….

Figure 2.4 Area Affected by the Indian Ocean Tsunami (IOT) ………

Figure 2.5. Store Building with the Dashed Line (red color) on the Wall ………

Figure 2.6. Graphs of Tsunami Heights Measured by ITS Team ………..  

Figure 2.7 Infrastructure Damages, Transportation (a) and Communication (b)

Infrastructure ………..

Figure 2.8 Building Damage in BAC ……….

Figure 3.1. Flowchart of Methodology Used in This Study ………..

Figure 3.2. Delineation of Tsunami Inundated Using Post Tsunami Aerial Photograph..

Figure 3.3 Results of Building Boundary Digitations from Aerial Photograph And Building Map in Shapefile Format ………

Figure 3.4 High Inundation Map (in Meter) was created using SPline Interpolation Method ………

Figure 3.5 Classifications and Distribution of Topographic Zone ………

Figure 3.6 Distribution of Class Elevation (in meter) ………...

Figure 3.7 Edit Tool Extension in Arcview Software for Polygon Checking and Editing Figure 3.8 Red Color Nodes Showing Error in Polygon and Clean Data After Editing ...

Figure 3.9 Attribute Classes in Building Damage Database. Link between Object and Attribute in Database marked by color ………..

Figure 3.10 Overlay Between 100m Grid Polygon and Location of Damage Assessment for Damage Map Construction ………..

Figure 3.11 Location of Setui village in BAC ………

Figure 3.12 Showing Database of Building Damage in Setui Village and Result of

14 17 17 20 20 22 23 25 26 27 27

30 31 36 40

41

43 43 43 45 45

47

47 49

10  Classification Based on Building Size ………

Figure 3.13 Attribute Classes and Buildings Map in BDIS Database ………

Figure 4.1 Building Distance from Coastline in Each Level of Damage ………...

Figure 4.2 Correlation Between Building Size and Building Damage ………..

Figure 4.3 Number of Building Unit in Each Inundation Area and the Damage Level ….

Figure 5.1 Object Visualization Which Done in Parallel with File Conversion Process Using shp2kml tools ………...

Figure 5.2 Visualization (color and attribute) and 3D modeling of Building Damage Using Google Earth ………

List of Maps

Map 1 Spatial Distribution of Each Class of Building Damage in BAC ………

Map 2 Distribution of Building in BAC Based on Building Size ………..

Map 3 Distribution of Building Damage in Each Level of Inundation ……….

Map 4 Distribution of Number of Building Floor in Each of Damage Level ………

Map 5 Distribution of Building Size in Each of Damage Level ………

Map 6 Distribution of Number of Room in Each Building and Level of Damage ………

Map 7 Distribution of Building Usage and Level of Damage ………

Map 8 Distributions of Class Number of People and Level of Damage ………

Map 9 Distributions of Class Building Age in Setui Village ……….

Map 10 Distributions of Class Building Earthquake Resistance in Setui Village ……….

Map 11 Class Damage Map is Showing Boundary and Spatial Patterns of Each Class … 51 53 54 58 60

81

82

56 58 60 62 64 65 67 68 70 71 75  

1. INTRODUCTION 1.1. Background

A massive earthquake shook Sumatra Island on December 26^th, 2004, registering a magnitude 9.0 on the Richter scale. The epicenter was located in Indian Ocean, 250 kilometers from Banda Aceh City (BAC), the capital of Nanggroe Aceh Darussalam (NAD) Province and geographically at 3.298^o N Latitude and 95.779^o E Longitude. This earthquake was recorded as the fifth strongest over last 100 years and the worst in 40 years (Mehdiyev, et al, 2005). It was followed by a big Tsunami known as Indian Ocean Tsunami (IOT) which caused death of over 270,000 people in 11 countries across Asia and Africa. In Asia, the fatalities mostly happened in Indonesia, Sri Lanka, India and Maldives (Haque. et. al, 2006). The human impact in Indonesia due the earthquake and the Tsunami has been massive, and larger than any other country. As reported in January 2005, 110,229 people were accounted as dead, 12,132 people as missing and 703,518 as displaced (BAPPENAS, 2005).

The worst affected areas were BAC, the capital of the NAD Province, many buildings and infrastructure were collapsed or failed, especially transportation and communication facility. Researches estimated that about 8,000 km³ ocean water temporarily invaded land with the horizontal extent of coastal inundation from a few meter to as far as 3 kms (Haque. et. al, 2006) . In this city, the damages were centered within a 3.2-6.4 kilometer zone along the inundation area. Tsunami swept debris and sea water into buildings up to 5 kilometers inland, crushing them and damaging roads and bridges, telecommunications, drinking water supplies, electricity, crops, irrigation channels, fishery infrastructure, food and fuel station etc.

BAPPENAS (2005) estimated that total damages and losses from this disaster in Indonesia was USD 4.45 billion which was dominated by the damage of transportation and irrigation facilities, flood control and coastal protection infrastructure, energy, water and sanitation and communications.

As an impact of earthquake, BAPPENAS (2005) estimated that around 19 percent of the approximately 820,000 building units in the affected districts suffered an average of

12  about 50 percent damage while around 14 percent are in totally destroyed condition.

BAC, Sabang city, district of Aceh Jaya, Aceh Besar, Aceh Jaya was bearing the brunt of the disaster with damages of over 80 percent of their housing stocks. Houses in the affected building units, semi modern and modern units appear relatively suffered more than traditional ones.

After occurrence of disaster, devastated area in the city will be changed enormously, availability of building damage information system (BDIS) is very important for fast damage and loss calculation after the event. Boundary of individual building can be extracted from help of high resolution imagery and aerial photograph and various attribute data can be developed in one system in the geographical information system database (geodatabase).

Usage of remote sensing for disaster management especially for individual building extraction have been studied many times and applied in various parts of the world. This has been shown by many reports and researches regarding this particular study and according to Vu, et.al, 2004, recent researches in damage detection have employed remote sensing for quick building damage detection and quick damage assessment. These studies are focusing in developing building damage information system (BDIS) using high resolution imagery, aerial photograph and GIS data. The result of analysis was attempted to conversion and sharing through internet technology using Google Earth application.

1.2. Objectives

The general objective of this study was to apply remote sensing and geographic information system (GIS) data for development of building damage information system (BDIS) in Banda Aceh City and specific location at Setui village with Sumatra earthquake event as case study.

Aims of this study were:

1. To extract individual building information using post earthquake aerial photograph

2. To classify level of building damage based on European Microseismic Scale (EMS) classification

3. To develop building damage database and Setui building damage information system (Setui BDIS)

4. To assess building damage at BAC and specific location at Setui village.

5. To identify spatial pattern of building damage

6. To develop three dimensions (3D) of Setui building damage model in Google Earth and sharing through internet.

1.3. Study Area

The study area covers approximately 61 Km2 in the administrative area of BAC, which is located in northwest of Indonesia, approximately from 5^o 28’ 52” N to 5^o 36’ 37” N and from 95^o 15’ 47” E to 95^o 25’ 23” E. The average altitude of 0.8 meters above sea level with lowest altitude around 0.2 meter and the highest altitude 1.8 meters above sea level.

By being the capital, BAC is the center of commerce, education and culture in Nanggroe Aceh Darussalam (NAD) Province. There are nine sub-districts in this city, which are Meuraxa, Baiturrahman, Kuta Alam, Ulee Kareng, Jaya Baru, Banda Raya, Lueng Bata, Syiah Kuala and Kuta Raja. From capital city of Indonesia-Jakarta, this city is reachable by air transportation approximately in 2, 5 hour or land transportation through Sumatra inland as well as sea transportation. Figure 1.1 shows location of BAC as area study.

1.4. Physical Setting of BAC 1.4.1. Morphology & Geology.

Morphology of BAC and its surroundings is consisted of coastal plain morphological unit with slope about 0–3 percent and the elevation around 0–3 meters above sea level.

Morphological units are dominated by river and coastal alluvium deposition.

14  According with Bennett, et. al, 1981, geology of BAC is dominated by undifferentiated Holocene alluvium consisted of gravel, sand and mud above unconformity to Meulaboh formation which is composed from semi unconsolidated sands and gravel. Using geotechnical method, Culshaw.M.G, et.all, 1979 classified lithology unit in BAC into six classes and they are clay, clayey sand, gravel, sand, silt and swamp. Near from shoreline, lithology is dominated by sand and much more swamps toward inland.

Geology structure in this area is controlled by two major faults located in northeast and southwest of this city continues from Semangka bay through central part of Sumatra Island

Figure 1.1 Map of Banda Aceh City and Surrounding (Sources: http://www.google.com and BRR Aceh-Nias, 2005)

1.4.2. Climate and Rainfall

In general, cities in NAD Province have Monsoon’s climate pattern which can be seen by the commutation of climates that happened every six month between dry season and the rain season. Statistics of Indonesia (BPS), 2005, stated that the monthly temperature average is 26.6 ^o C with the minimum temperature is 26.4 ^o C and the maximum temperature is 28 ^o C. The average of atmospheric pressure in this city and surrounding is 1,010.9 mb with the humidity around 76 till 89 percents. The highest rainfall in this city is 639 mm/month with the average of rainy days is from 6 to 21 days in December. While the lowest rainfall is between 33 mm until 291 mm with the average of rainy days is amount 2 days in March.

1.4.3. Landuse Change and Infrastructure Inundated 1.4.3.1. Landuse in BAC

Landuse in BAC covering 6,100 hectare land dominated by residential & activity place in which reach 42 percents of the total land area and it is followed by 6.7 percents of waste land and 5,6 percent of rice field. Others 45.7 percent land uses are consist of bush, dry field, forest/mangrove, grass field, cemetery, plantation and sand dunes. Spatial distribution of each type of landuse can be seen in Figure 1.2 and the wide in Table 1.1.

Table 1.1 Landuse Year 2005 of Band Aceh City

No Type of Landuse Wide in Hectares

1 Bush 23.3

2 Dry Field 159.8

3 Forest and Mangrove 141.8

4 Grass Field 104.3

5 Cemetery 8.1

6 Plantation 134.9

7 Res. and Activity Place 2,583.3

8 Rice Field 342.7

9 Sand Dune 24.9

10 Wasteland 411.9

16  1.4.3.2. Impact of Tsunami to Landuse Changes

Landuse change analysis has been conducted using pre tsunami IKONOS Imagery and post tsunami aerial photograph. The analysis result showing extremely changes only in two years that is covering 1,606 hectares area as impact of tsunami inundation. The changes of various types of landuse become water bodies that are reaching more than 1,232 hectares from totals of 6,100 hectares area of BAC. Change of fisheries area become water body is the biggest change that is reaching more than 58 percentages of total changes. Spatially new water body in inland located relatively close from coastline (See Map 2). Landuse change in BAC in 2003 – 2005 periods can be seen in Table 1.2.

Table 1.2. Landuse Change in BAC as Impact of Tsunami

No Type of Landuse (Pre Tsunami)

Type of Landuse (Post Tsunami)

Landuse Change (in Hectares)

1 Fishery Water Body 934.0

2 Fishery Wasteland 76.2

3 Forest and Mangrove Water Body 72.5

4 Rice Field Water Body 71.8

5 Residential and Activity Place Water Body 62.9

6 Sand Water Body 55.8

7 Plantation Water Body 50.3

8 Forest and Mangrove Wasteland 48.2

9 Fishery Forest and Mangrove 29.5

10 Residential and Activity Place Wasteland 91.3

11 Others Others 113.7

TOTAL 1,606

Figure 1.2 Landuse of BAC Year 2003 (Pre Tsunami)

Figure 1.3. Landuse of BAC Year 2005 (Post Tsunami)

18  1.4.3.3. Inundated City Infrastructures

Assessment in this study conducted for city infrastructure such as road and water network, residential area, bridges, health facility and education facility (schools) using each thematic map and map of inundated/un inundated area.

Overlay between map of inundated/un-inundated and road networks showing that more than 335 km roads in BAC is inundate at occurrence of tsunami and more than 203 km other are un inundated. Roads which are experience inundated, showing damage at various levels from only suffused till experience very hard damage even losing after the tsunami occurrence. Road with very hard damage condition generally located near from coastline such as roads in Ueleeleu (Coastal area in BAC). Spatial distribution and long of inundated road network are shown in Table 1.3 and Figure 1.4.

Table 1.3 Number of Infrastructure Damage in BAC

No Type Infrastructures Inundated Un Inundated 1 Roads Network (in kilometers) 355,412 203,904 2 Water Network (in kilometers) 7.0 33.9 3 Residential (in hectare) 1,526 1,057

4 Number of Bridges 81 72

5 Number of Health Facility 11 9

6 Number of School 133 56

Elementary School 72 36 Junior High School 28 10

High School 33 10

The same overlay between map of inundated/un-inundated and water networks showing that only 7 km water network in BAC is inundate at occurrence of tsunami and more than 33 km other are un inundated. Water network which the installation location relatively far from coastline, cause this infrastructure not many experiencing damage. The networks are impacted by Tsunami, generally experience slightly damage and easy for repairing.

Spatial distribution and long of inundated water network are shown in Table 4.2 and Figure 1.4

From the total of 2,583 hectares residential area in BAC, more than 60 percent are inundated on the tsunami event. Residential located near from coastline such as government residential complex in Ueleeleu, buildings generally in heavy to very heavy damage (Grade 5) condition. Distribution and wide of residential in inundated area are showing in Figure 1.5 and Table 1.3.

The same overlay using bridge map showings that from totally 153 bridges in BAC, only 72 save from water inundation and the rest experience damage at various level. Bridges with heavy damage generally happened at small bridge connecting location and relatively near from coastline. Large size bridges connecting main road in BAC generally save from damage or only experience slightly structural damage. Distribution of each bridge in the city is showing in Figure 1.4 and the number of inundated bridge in Table 1.3.

Table 1.3 is showing number health facility in where from 20 health facility including hospital and public health service, more than 50 percent experience inundated. Heavy damage health infrastructure was experienced by one hospital government property which located in sub district Baiturrahman in the downtown of BAC. Damage of building infrastructure not even is tsunami impact however more because as an impact of big earthquake happened before. Spatial distribution of health facility in BAC is showing in Figure 1.4.

Assessment of education facility is one important thing in this study. As impact of earthquake and tsunami about 70 percent schools from different education level is inundated on the tsunami event (See Table 4.2) and spatially distribution of school spread over around the city (Figure 1.4). Like a health facility, damage of education facility generally not even is tsunami impact however more because as an impact of big earthquake happened before.

20   

Figure 1.4 Inundated City Infrastructures Figure 1.5 Inundated Residential Area  

1.5. Population

BPS-Social welfare survey year 2005 indicated that population in BAC were 177,744 which spread over in nine districts on the city. Male residents reach amount to 89.880 and female of 87.864 with sex ratio equal to 102. This ratio means for every 100 woman resident there are counted 102 men. Different condition showed in 2004 where population in this city reached 239.146 with sex ratio equal to 97 (See Table 1.4). Big depopulation is an indicator that occurrence of earthquake and tsunami on December 26^th, 2004 have snatched tens of thousand peoples especially in BAC. Density of BAC in year 2005 is 2,512 people/km² which have been decreased from condition of 2004 where the density reaching 3,920 people/km²

Table 1.4 Area and Population in Kota Banda Aceh year 2004 and 2005

Indicator 2004 2005

Area (km²) 61 61

Population 239,146 177,744

-Men 117,611 89,880

-Women 121,535 87,864

Sex Ratio

(Number of men per 100 women)

97 102 Productive age 166,876 130,831

Unproductive age 72,269 43,510 Dependency Ratio

(Number unproductive people Per 100 productive people)

43.3 33.25

Population Density 3,920 2,512

Sources: BPS- Banda Aceh in Figure 2004, BPS-Social Welfare Statistic 2005 and BPS-Population of Kota Banda Aceh 2005

Based on population by sub-district and specific age group in BAC from BPS (2005), productive age (15-64 year) reach 75 percent, young age (0-14 year) is 22.53 percent and population with 65 years old and older is only 2.41 percent. From the data above, number of dependency ratio is showing the existence of decrease from 43.3 in 2004 become only 33.25 in 2005. Dependency ratio 33.25 means every 100 productive people will account 33 peoples from group of unproductive age. Detail comparison of unproductive and productive people in BAC can be seen in Table 1.4 above.

22  2. EARTHQUAKE, TSUNAMI AND DAMAGES

2.1 Tectonics and Sumatra Fault System (SFS)

Indonesia islands are formed of the southeastern extremity of the Eurasian lithospheric plate (see Figure 2.1),

Figure 2.1 Position of Indonesia Island in Eurasian Plate (Source: USGS, 2000)

It is bounded by the northward-moving Indo-Australian plate and the westward-moving Pacific and Philippine plates and it is certainly one of the most complex active tectonic zones on the earth. The rate of subduction in some locations is around 0.6 cm per year like in the West Java Trench at 0°S 97°E and 0.49 cm per year like in the East Java Trench at 12°S 120°E. Result of GEODYSSEA showing different result where GEODYSSEA defined a 'Sunda' Block, - that includes Borneo, the Malay Peninsula and Indochina -, moves east relative to Eurasian at 0.7-1.0 cm per year (Milsom, 2003).

According with Milsom (2003), tectonic of Sumatra continental island processes are controlled by three major fault systems. The most obvious of which is the subduction thrust which outcrops in the Sunda Trench. The trench curves very little in the 800 km between Enggano and Nias. In this subduction zone, the Indo-Australia plate moves

toward the northeast at a rate of about 6 cm/year relative to the Euro-Asian plate in oblique convergence at the Sunda trench. The oblique motion is partitioned into thrust- faulting, which occurs on the plate-interface and which involves slip directed perpendicular to the trench, and strike-slip faulting, which occurs several hundred kilometers to the east of the trench and involves slip directed parallel to the trench (Natawidjaja, 2005). Figure 2.2 shows seismo tectonic of Sumatra and view of Sumatra Fault in the western side of BAC.

Figure 2.2 Seismo-tectonic of Sumatra Island (Source: Natawidjaja, 2005)

Natawidjaja, 2005, wrote in his dissertation that Sumatran faults as product of regional tectonic was passes through the entire island. The fault is divided into three segments,

24  southern, central and northern. The fault is thrust type with a dextral form. Sumatra Fault System (SFS) probably constructed from the Middle Miocene age and the opening of the Andaman Sea, although the relative motions of the major plates have changed little since the Middle Eocene. The SFS runs the length of the Barisan Mountains in the middle part of Sumatra Island which is of uplifted basement blocks granitic intrusions. Tertiary sediments located at the top of this formation.

2.2. Sumatra Earthquake and Indian Ocean Tsunami (IOT) 2.2.1. Sumatra Earthquake

Sumatra earthquake with magnitude 9.3 on the Richter Scale occurred some 30 km beneath in the Indian Ocean Floor with total rupture area about 250,000 km2 and a length about 1,200 km. The rupture started in the south and propagated to the northwest up to the Andaman and Nicobar Island during seven-minute duration (Haque, et. al, 2005).

Different scale has been reported by the USGS that which measured the earthquake at 9.0 on the Richter scale, making it one of the largest recorded. The epicenter was some 150 kilometers south of Meulaboh City (Aceh Barat District) and about 250 kilometers from Banda Aceh, the capital of NAD province (See Figure 2.3). The earthquake happened is approximately shallow earthquake of 30 km in subduction zone of the Indian Ocean.

According to USGS (2005), the earthquake, where one tectonic plate subducts beneath another. The interface between the two plates results in a large fault, termed a megathrust, that can be traced along an arc parallel to the Sunda Trench from Myanmar to Java and occurred along the boundary of the India and Burma plates. The India plate moves an average of 6 centimeters per year in relation to the Burma plate in northeast, and generates strike-slip faulting several hundred kilometers east of the Sunda Trench.

Many strong aftershocks followed the initial earthquake. USGS analyses of their characteristics indicate that nearly 1200 kilometers of the plate boundary fractured and slipped, with a likely width of more than 100 kilometers and a displacement of about 15 meters.

Figure 2.3 Epicenter of Sumatra Earthquake (Source: USGS, 2005)

2.2.2. Indian Ocean Tsunami (IOT) and Level of Inundation

The Sumatra earthquake generated a large tsunami that traveled rapidly throughout the Indian Ocean, striking coastal areas in many countries with catastrophic results in Indonesia, Thailand, Sri Lanka, India and Bangladesh, as well as other Maldives and African countries (See Figure 2.4). More than 150,000 people died with many more still missing, while infrastructure, productive activities and the natural environment were either destroyed or damaged (BAPPENAS, 2005). According to Haque, et.all, 2006, IOT is not only propagated throughout the Indian Ocean but also reached the pacific and

26  Atlantic Ocean, although it was not destructive in those two Ocean. Close to the epicenter, the tsunami amplitude did it reach maximum value of up to 30 m or sporadically even more in Banda Aceh Area. Tsunami amplitudes could have reached up to 20 m in Thailand, up to 11 m in Sri Lanka, 7 m on the Tamil Nadu coast of India, 8 m in Andaman and Nicobar Island and around 4 m in Maldives, Kerala Coast of India and Somalia.

Synolakis (2005) wrote in his paper, the essential elements of the database of post- tsunami surveys include measurements of run-up, inundation, and flow depth of tsunami water. Inundation is the maximum horizontal penetration of waves in the direction normal to inland during the flooding. Characterizing water penetration in the affected area can be based on a watermark, such as a line of debris on building or vegetation.

Figure 2.5 shows water mark around 2.5 meters high in store building in Banda Aceh.

Figure 2.4 Area Affected by the Indian Ocean Tsunami (IOT) (Source: Overseas AID-AusAid, 2005)

Field survey conducted by Earthquake Research Institute (ERI), University of Tokyo, Japan in 2005, has showed variation of tsunami inundation in BAC and West Coast of that city. In the Center of BAC, Tsunami has been inundated between 3.5 to 12 meters and reaches 35 meters in west coast (ERI, 2005). International Tsunami Survey Team (ITST) studied the effects of the December 26 tsunami on Indonesia's island of Sumatra and reported wave heights of 20 to 30 meters at the island's northwest and found evidence that wave heights may have ranged from 15 to 30 meters along at least a 100 kilometers stretch of the northwest coast. According to USGS, 2005, these wave heights were higher than those predicted by computer models made soon after the earthquake and tsunami (See Figure 2.6)

Figure 2.5. Store Building with the Dashed Line (red color) on the Wall

Indicates Water Mark Left by Tsunami  (Source: YIPD/CLGI, 2005) 

Figure 2.6. Graphs of Tsunami Heights Measured by ITS Team

In Banda Aceh and along the west coast (Source: USGS, 2005)

2.3. Infrastructure Damage in BAC

Result of Infrastructure damage and losses assessment has been reported by BAPPENAS, 2005. The assessment was carrying on under the standard internationally accepted methodology developed by the United Nations Economic Commission for Latin America and the Caribbean (BAPPENAS, 2005). This methodology using conceptual basis is a

28  stock/flow analysis that evaluates effects on physical assets that will have to be repaired, restored or replaced or discounted in the future and flows that will not be produced until the asset is repaired or rebuilt (ECLAC, 2003)

2.3.1. Infrastructure Damage

BAPPENAS, 2005 reported damage and losses to infrastructure caused by earthquake and tsunami in NAD Province totaled USD 891 million and were dominated by the damage to transportation or 61 percent of total impact and followed by irrigation, flood control and coastal protection around 25 percent, 7.7 percent in energy, 3.4 percent in water and sanitation and the rest 2.5 percent in communication.

Table 2.1 Summary of Damage and Losses to Infrastructures Infrastructure Damages Losses Total

(Percentage)

Transport 3,632 1,352 61

Water and Sanitation 247 29 3,4

Energy 631 1 7.7

Communication 176 27 2.5

Flood Control Irrigation 1,230 829 25

TOTAL 5,915 2,239 100

       Source: BAPPENAS, 2005

The damage in all provincial transportation facility was dominated by roads and land transport, reaching USD 369 million or 94 percent. The Impacted facilities in the affected coastal areas were including about 316 kilometers or 10 percent of the national and provincial road network, 1,900 kilometers of local roads and more than 400 bridges, about 30,000 vehicles or 7.2 percent from the registered vehicle fleet. Figure 2.7a are shown bridge and road damage in BAC after disaster. Public port infrastructure suffered moderate damage is totaling USD 25.7 million or five percent over 14 facilities in Aceh and 5 in North Sumatra, about 70 percent overall sustaining heavy damage. Airports suffered moderate earthquake-related damages to runways and terminal buildings, estimated to cost around USD 1.8 million or 1 percent. Losses for transport due to increased costs and lost revenues are estimated to be USD 141 million over five years.

Reported by the provincial urban water association providers damage and losses totaling USD 10.5 million, mainly to pipelines and lost about 15 percent of production capacity overall. However, the damage was acute in the most affected urban zones in the north and west part of that affected area. The most substantial damage appears to be to small-scale providers and rural water supply, totaling USD 18.5 million, where it is estimated that almost all wells in the affected coastal region have been contaminate by salt and sandy clay water . Urban sanitation, comprising entirely on-site facilities, suffered damage to the treatment servicing equipment, valued at USD 978,260. Total damage to water and sanitation was valued at USD 30 million.

The energy sector damage was reaching USD 68.5 million, the majority of it to the distribution networks in both electric power and petroleum fuel supplies. Electric power supply suffered light damage to generation capacity, no damage to the transmission network but substantial damage to the distribution networks in the affected area. Total damage is estimated an around USD 54.3 million, with negligible operational losses. The state-owned petroleum fuel supply suffered substantial damage to fuel depots, where storage facilities were damaged and some fuel lost, mostly on the west coast, with a total damage of USD 14.2 million.

In communications and telecommunications sectors, suffered severe damage in the affected areas, primarily to fixed connection services where 40 percent of connections were broken, and to transceiver facilities for cellular phones, with total damage of around USD 18.2 million and postal services facilities suffered damages of USD 956,521. Figure 2.7b are shown damage of mobile communication facility in this city.

Flood control and irrigation infrastructure damage was substantial, totaling USD 133.7 million. About 45 percent of the damage cost was sustained on the flood control and coastal protection infrastructure (USD 67.8), and 53 percent on major irrigation facilities of which about a third was earthquake-related.

30 

(a) (b)

Figure 2.7 Infrastructure Damages, Transportation (a) and Communication (b) Infrastructure (Source: USGS, 2005 and YIPD/CLGI, 2005)

2.3.2. Building Damage Assessment

Building damage assessment especially in BAC, has been done and reported by The BAPPENAS (2005). BAPPENAS, 2005 release estimation that about 19 percent of the approximately 820,000 building units (about 151,600 units) in the affected districts suffered an average of about 50 percent damage while about 14 percent or around 127,300 building completely destroyed. The damages were centered within a 3.2-6.4 kilometers zone along the coast; Kota Banda Aceh, Aceh Jaya, Aceh Besar, Kota Sabang and Aceh Jaya bearing the brunt of the disaster with damages of over 80 percent of their housing. Building estimated that NAD Province and North Sumatra Province suffered damages and losses totaling about 13.4 trillion rupiah. In an effort to show how different income groups were affected, BAPPENAS attempt to classify houses as modern, semi- modern and traditional similar with BPS categorization in the 2000 census. In terms of numbers, semi-modern and modern units appear to have suffered proportionately more than the traditional ones. The reasons for this is not very clear; perhaps it could be because of the relatively richer residents were living closer to the sea.

(a) (b) Figure 2.8 Building Damage in BAC, 0 – 50 Percent Loss(a) and Totally Loss Building (b)

(Source: YIPD/CLGI, 2005)

2.4. Remote Sensing and GIS for Building Damage Assessment 2.4.1. Building Extraction from High Resolution Image

Study about building extraction using high resolution images has been done by Lhomme, et.al, 2004, Sohn & Dowman, 2003 and Sahin, et.al, 2004.

Lhomme’s paper focuses on building extraction from high resolution image in urban areas and original building detection approach is proposed based only on discrimination by ratio of variance (DRV) features. This parameter is tested to extract buildings centroid from a panchromatic IKONOS image, theory and methodology is studied first and continues to attempt in the small area on Sherbrooke City, Canada using panchromatic IKONOS image. This study results shows that the DRV is an interesting method for building detection, although the methodology still needs additional developments. There are major problem is the high commissions errors, which could be reduced by using additional spatial and spectral information.

Sohn & Dowman, 2003 studied a system for automatically detecting building objects and delineating boundaries from IKONOS images and LIDAR data. He was released a new idea to combine complementary nature of intensity images and high-quality of 3D information to solve problems associated with building detection and building description. The conclusion from his study is IKONOS images can be used in

32  topographic mapping at large scale in a combination of LIDAR data. The current system is limited to delineating polygonal shapes buildings with flat roofs and he suggest a further development study to reconstruct 3D roof structures based on the ground plans extracted by the current techniques.

Different from above researches, Sahin, et.all, 2004 published their own research about developments use of a high resolution panchromatic KVR-1000 image for extraction of man-made structures and he attempted this changes in Zonguldak city. In his study, manual on-screen digitizing and the automatic object oriented image analysis methods have been compared. By manual manner, building and road details that are available or not available could be derived. Sahin,et.all, 2004, has written although the effective pixel size of KVR-1000 orthoimage is about 2 pixel, experience and function of operator are the main factors on the success rate. Sahin research concern was not the geometric accuracy of the classification but attempted the accuracy potential of from KVR-1000 image digitized vector maps. The study comes to the conclusion that pixel size does not appoint the map scale of end product to be extracted from the satellite images such as KVR-1000.

2.4.2. Remote Sensing and GIS for Damage Assessment

Application of various resolution of Remote Sensing for damage assessment (earthquake or tsunami) has been applied of various researchers including Vu, et.all, 2004, Adams, 2004, Sumer and Turker, 2004, Miura, et.all, 2005 and Mehdiyev, et.all, 2005.

Vu, et.all (2004) studied about the advantage of shadow appearance in standard imagery produced from QuickBird for damage detection in urban areas. In a very complex scene of an urban area acquired from very high resolution satellite-based optical sensors, fortunately, the buildings tend to align in some dominant directions in a small area and posses geometric regularity. Vu proposed an automatic shadow analysis, which compared the extracted shadows from pre event and post event high resolution imageries to find out the cue for damage detection.

In this paper, he presents only the core processing, which is fully automatic where must be some pre-processing steps, which depend on the in-hand high resolution satellite imagery, before this automatic processing can be employed. Two QuickBird scenes over the city of Boumerdes are employed in this study. The first one was about one year before the event and the second one was two days after the event. The results from his analysis prove that shadow-based information could be used as a potential information for automated detection of building damage.

Adams, 2004 presented his research paper in which demonstrate the use of high resolution Quickbird and moderate resolution SPOT and ERS satellite imagery for determine the location and severity of post-earthquake building collapse. Damage detection algorithms have been developed and become based on the comparative analysis of a multi-temporal sequence of optical or SAR images before and after the event. Adam Research is carried in the Marmara city Turkey, for which 10 meter panchromatic resolution of SPOT 4 and 20 meter resolution of SAR ERS coverage are used.

Preliminary results are also presented for the May 21st, 2003 Boumerdes earthquake in northern Algeria, where 60 cm resolution of Quickbird imagery are used. In this Boumerdes location, the change detection algorithms offer a quick-look region-wide damage assessment, providing the focus for more detailed visual inspection of building damage on a per structure basis. Adams in his conclusion said these techniques could be guide the work of field reconnaissance teams, support the prioritization of relief efforts, direct search and rescue teams to victims and facilitate loss estimation

Sumer and Turker, 2004 studied the collapsed buildings due to earthquake using post- event aerial images and development of watershed segmentation algorithm. The objective of this study was to detect the collapsed buildings based on the analysis of the cast shadows. The building boundaries were available and stored in a GIS as vector polygons.

The building polygons were utilized to perform assessments in a building specific manner. The approach was implemented in a selected urban area of Golcuk City Turkey.

The shadow regions were detected using a watershed segmentation algorithm. This methodology was followed by measuring the agreement between the shadow producing edges of the buildings and the corresponding shadows based on the percentage of the

34  shadow pixels. In they study area, 284 buildings analyzed, 229 were correctly labeled as collapsed or un-collapsed with accuracy of 80 percent. The conclusion results prove that the collapsed buildings caused by the earthquake can be successfully detected from post- event aerial images.

Miura, et.all, 2005 specifically research of tsunami damage in the eastern part of Sri Lanka due to the 2004 Sumatra Earthquake using High-Resolution Satellite Images.

Visual detection of building damage is applied to pre- and post-event images in Sri Lanka that were severely damaged due to the 2004 Sumatra earthquake. In their study; buildings are classified according to the damage level where the classification shows good agreement with the actual damage. Damaged buildings are Distribution of damage building investigate using GPS and continue overlay with inundation area. Result shows about 10 percent of buildings are classified into severely damaged buildings and the buildings are mostly concentrated in the eastern coastal line. The damage is distributed in inland area within 1km distance from coast and severest damage is observed in northern area in which width of land is narrow.

Study about damage detection and assessment as an impact of tsunami using high resolution satellite, GIS and GPS has done by Mehdiyev, et.all in Galle City-Southern part of Srilanka. High resolution IKONOS and Quickbird imagery used for determine the location of post-earthquake building collapse. The conducted study shows that high- resolution satellite images can provide the level of information that needed to identify most damaged areas after disaster and to distinguish totally and partially collapsed or not collapsed building. Their study shows that visual observation with support of standard GIS and image processing enable for damage identification and mapping could be done very rapidly. Others particular conclusion from their study is the information can be integrated into GIS and transfer via communication facility directly to the affected area and the results could help the public and for better coordination in the emergency operations.

3. METHODOLOGY

Methodology chapter consist of some sub-chapter interacting that are about data availability and sources, data extraction, map preparation, development of building damage database and classification of building damage. Sub chapter data availability and data source informs reader concerning data type which applied in this study and also the specification, year and source of data. Second sub chapter explain about data extraction from pre tsunami imagery and post tsunami aerial photograph and following by map preparation in the next sub chapter. Sub chapter fourth explained about development of building damage database including data integration and development database its selves.

Sub chapter fifth systematically informing the classification process since construction of grid in IKONOS until classification method has been used. Last sub chapter concerning development of building damage information system (BDIS) in specific location of Setui village in southwest of BAC. Steps and data used in this study could be seen graphically in methodology flowchart in Figure 3.1.

3.1. Data Availability and Sources

Aerial photograph, IKONOS Imagery as well as various scale GIS dataset has been used in this study. Detail information and data sources are given in Table 3.1 below:

Table 3.1 Data Availability and Sources

Data Type Scale/

Resolution

Year Source

Damage Field Assessment Data

2005 JICA Study Team IKONOS Imagery 2 Meter 2003 CRISP Singapore Aerial Photograph < 1 Meters 2005 BRR-Aceh Nias Base Map 1 : 2,000 2005 BRR-Aceh Nias Infrastructure 1 : 25,000 2005 BRR-Aceh Nias Level Inundation - 2005 ITST/ERI, Tokyo Univ.

Building Map 1 : 2,000 2005 BRR-Aceh Nias Administration of Banda

Aceh City

1 : 50,000 2003 Updated 2005

BPS, updated by BRR- Aceh Nias

Topographic Zoning 1 : 5,000 2005 JICA Study Team

36  Figure 3.1. Flowchart of Methodology Used in This Study

Damage classification (Grid Method)

Pre-processed pre- tsunami image (IKONOS Imagery)

Pre-processed post- tsunami image (Aerial Photograph)

Preparation of topo, inundation, elevation

and lithology map 100 m Grid damage

construction

Damage Field Assessment Data

Delineate building object

Building map

Building damage classification based on JICA Study Team

Classification

Comparison between pre and post tsunami

Imagery Pre-tsunami image

(Ikonos Imagery)

Post-tsunami image (Aerial Photograph)

Visual detection for damage classification

validation

Validation result and conversion to EMS,

1998 Classification

Classification of Building Damage

Create Building Damage Database

BUILDING DAMAGE INFORMATION SYSTEM

(BDIS)

3.2. Data Extraction Using Pre and Post Tsunami Imagery 3.2.1 Preparation of Pre and Post Tsunami Imagery

Regarding with the aim of data extraction for change detection and damage estimation, pre tsunami IKONOS imagery observed in January, 2003 and post tsunami aerial photograph observed in June 2005 cover the severely damaged area were used in this study.

Datasets were received from different sources on different spatial resolution, processing levels, reference, and projections. Pre Tsunami IKONOS images were in WGS84 with geographic projection and the aerial photograph was in similar reference with local UTM projection. It was decided to convert all datasets to local projection, as this facilitates comparison with all existing GIS data.

Some technical procedure has been conducted in this study. At first IKONOS images were converted to Indonesia local coordinates using conversion parameters provided by Indonesia National Coordination Board for Survey and Mapping (Bakosurtanal) and Arcview software with Image Analyst extension was used based on procedure as have been explained above

3.2.2 IKONOS Pan Sharpening

In order to increase quality and spatial resolution of multi-spectral satellite imagery fusion was carried out on the IKONOS dataset. Both panchromatic and multi-spectral images have been ortho-rectified, in order to reduce geometry errors inherent with topography and imagery.

Ortho-rectify process for the IKONOS data was using Rational Polynomial Coefficients (RPC) sensor model in where the process combines several sets of input data to place each pixel in the correct ground location. The offset between mean sea level and the gravitational potential surface known as the geoid is required so the elevation can be correctly interpreted. The resulting ortho-image is accurate to real world coordinates and the IKONOS dataset itself contain this RPC information. After ortho-rectification

38  process, data fusion was attempted using various pan-sharpening techniques and it was found that color-normalized “Brovey” algorithm as better fused result for those images.

3.2.3 Data Extraction through Visual Image Interpretation - Pre Tsunami IKONOS Data Extraction

Data extraction from remote sensing imagery involves the identification of various targets in an image. The extraction process can be done variously either through automatic/semi automatic and also manually through visual interpretation. Recognizing targets is the key to visual interpretation and information extraction. Observing the differences between targets and their backgrounds involves comparing different targets based on the visual elements of image interpretation such as shape, size, pattern, tone, texture, shadow, site and association

Shape refers to the general form, structure, or outline of individual objects. Shape can be a very distinctive clue for interpretation. Straight edge shapes typically represent urban or agricultural (field) targets, while natural features, such as fishpond edges, are generally irregular in shape and the boundary with residential area very clearly seen.

Size of objects in an image is a function of scale. It is important to assess the size of a target relative to other objects in a imagery, as well as the absolute size, to aid in the interpretation of that target. The large buildings such as factories or warehouses would suggest commercial property, whereas small buildings would indicate residential use.

Pattern refers to the spatial arrangement of visibly discernible objects. Typically an orderly repetition of similar tones and textures will produce a distinctive and ultimately recognizable pattern. Orchards with evenly spaced trees and urban streets with regularly spaced houses are good examples of pattern

Tone refers to the brightness or color of objects in an image. Tone is the fundamental element for distinguishing between different targets or features. A variation in tone also allows the elements of shape, texture, and pattern of objects to be distinguished

Texture refers to the arrangement and frequency of tonal variation in particular areas of an image. Smooth textures are most often the result of uniform, even surfaces, such as asphalt, or river. A target with a rough surface and irregular structure, such as a tree canopy or a rough surface and regular structure such as building

Shadow is will helpful in interpretation as it may provide an idea of the profile and relative height of a target or targets which may make identification easier. Shadows may have possibility reduce interpretation in their area of influence, since targets within shadows are much less discernible from their surroundings.

Site refers to topographic or geographic location and is a particular important aid in identification of vegetation types. Certain tree species would be expected to occur on well-drained upland site, whereas other tree species would be expected to occur on poorly-drained lowland site.

Association refers to the occurrences of certain features relationship between other recognizable objects or features in proximity to the target of interest. The identification of features that one would expect to associate with other features may provide information to facilitate identification. Residential areas would be associated with schools and sport facility.

- Post Tsunami Aerial Photograph Data Extraction

Information and data which obtained from post tsunami especially are inundation boundary and building data. Extraction of the inundation area conducted with the helps of less than 1 meter spatial resolution of the aerial photograph. Boundary between inundated and un-inundated area recognized visually based on tone of the image. Tone is referring to the different level of brightness or color of objects, the boundary digitized to obtain spatial data in shapefile format. Figure 3.2 Showing extraction of boundary of inundated and un-inundated area.