| GISdevelopment.net ---> Application ---> Natural Hazard Management |
MICRO-LEVEL VULNERABILITY ASSESSMENT AND COPING MECHANISM RELATED TO FLOODS IN URBAN AREAS
Mone Iye Cornelia
E-mail:mone_iye@yahoo.com
Marschiavelli1, Pramono Hadi2, Michael K. McCall3, Nanette Kingma3
1 Staff of Center for Marine Resources Surveys, National Coordinating Agency for Survey and Mapping (BAKOSURTANAL).
2 Faculty of Geography, Gadjah Mada University, Yogyakarta, Indonesia.
3 International Institute for Geo-Information Science and Earth Observation, Enschede, The Netherlands
Abstract
Flooding has become a serious problem in Jakarta. During 2007, Kampung Melayu was the worst hit by the floods. Community have different perceptions on disaster and develop different efforts to overcome floods. Therefore, local government and relevant institution should consider this situation and transform this information into valuable input in developing and implementing response plans.
This research tries to explore the vulnerability as well as the capacity for flood management based on people’s perception. Data gathered from a communitybased field data collection were processed using Geographical Information Systems (GIS) in order to illustrate the flood event in 2007 and its vulnerability in Kampung Melayu. In addition, information about existing coping mechanisms based on people’s perception also discussed in this research.
The result shows that the ability of people to cope with the flooding is linked with the capacity of the people itself. The capability of people to deal with flooding was influenced by several indicators based on their socio-economic characteristics. Several coping strategies employed by the local people and local government, however, not enough to cope with flooding in the study area. Local government and people in this area should cooperate to reduce the impact of flooding using the knowledge of how to address disasters effectively through people’s participation in the process. All information may represent the situation against the flood and can be used as a tool to improve the capacities of local people and minimize the flood risk.
1. Introduction
Annual flooding is one of environmental issues in the Jakarta area due to the worsening river management both in the upland and lowland. It is occur during rainy season, triggered by heavy seasonal rain in December, January, February and March. Flood in Jakarta has been recognized since the Dutch occupation era in Indonesia. Historical record illustrates that some enormous floods have occurred that killed some people and destroyed properties i.e., year 1699, 1714, and 1854. While in the last few decades, the flooding occurred in 1918, 1942, 1996, 2002 and 2007; caused some damages and some people were killed and lost. The flood on February 2007 was spread in almost 60% of Jakarta areas and its surroundings.
Flood in Jakarta is affected by several other factors, such as: morphological (approximately 40% areas of Jakarta Province is lowland area, some of them even below sea level), hydro-meteorological aspects (heavy rainfall and high sea tide during rainy seasons), land use change, decreasing the flow cross-section of the Ciliwung River because of garbage along the river and illegal settlement in the riverbanks, and also because of socio-cultural aspect, i.e: weak policy implementation, rapid urbanization, solid waste dumping and management. Vulnerability assessments are necessary in order to reduce the impact of the next flooding event in Jakarta. Government already conducted many surveys to measure and assess flood damage, but the vulnerability assessment at the micro-level (based on community data), including the coping mechanism of the community hasn’t been done yet. Therefore, this research is addressed to identify certain elements at risk, assess the vulnerability of each of the element at risk, and assess the coping mechanism/capacities of the community. The result of this research can be used as a valuable input for local governments for making appropriate actions, policies and programs in the context of flood hazard management in this study are in order to reduce the risk of the flood hazard and to apply it in urban areas of Kampung Melayu, Jakarta, Indonesia.
2. The Study Area
Jakarta, located on the northwest coast of the island of Java has an area of 661.52 km². The study area of this research is Kelurahan Kampung Melayu in Jatinegara District, East Jakarta, located along the Ciliwung River (Figure 1). It divided into 8 Rukun Warga (RW) and 114 Rukun Tetangga (RT). Geographically, Kampung Melayu located in the non-coastal area (± 15 Km from shoreline) and relatively flat. The altitude of Kampung Melayu is between 9 until 18 above the sea level. Kampung Melayu with area 0.48 km2 has high density of people.
Figure 1: Study area of the research
3. Research Method
There are several methods that already applied in order to assess vulnerability at different scales: national, mega city and local scale (Villagrán, 2006). In order to achieve the main objective of this research, therefore the vulnerability assessment at community level is the most appropriate. This research is focused on identifying and classifying certain elements at risk which would be affected by flood in study area, assessing the vulnerability of the elements at risk, and identifying the current coping capacities and coping strategies. Sampling method used during the fieldwork is non-proportional stratified random sampling. Overall, there were 83 point sample designed for field survey distributed in eight RWs.
Building Inventory
Building inventory aimed to gain building information for database and to verify the respondents’ answer during interview. The activities in building inventory were measuring the height of floor from street and above the surface and observing the physical aspect of building such as wall material, floor material, roof material, number of floor, etc. All information was stored in mobile PDA and added as an attribute for each building points. The attributes for building inventory were: building_id, date, owner, function, wall material, floor material, roof material, number of floor, size, floor height from surface, floor height from the street and photo’s number.
Household interview
The purpose of household interview using questionnaires was to get detailed information about element at risk including asset and damage estimation for each household, and households’ coping mechanism. In-depth interview and consultation were held with occupants of the buildings identified as being risk. Each interview lasted between 30 and 45 minutes.
Data Analysis
Data collected from the building inventory were able to be analysed after fieldwork because already in digital format. The shape file from mobile PDA was transferred into PC and using Arc View software, all of data collected during fieldwork can be retrieved easily (see Figure 2). However, information from interview had to be converted from hardcopy format (questionnaires) into digital format first. Data from building inventory and interview processed spatially using ILWIS 3.3 and Arc View 3.3.
There are several parts in this stage: identification of certain element at risk, vulnerability assessment and coping mechanism analysis. For vulnerability analysis, author conducted spatial analysis using GIS software. It was used to plot the physical element at risk (building material) and results from analysis. Author used kriging to interpolate the point data using Gaussian model. For coping mechanism analysis, author using descriptive analysis.
Figure 2: Illustration of point with attributes collected during the fieldwork
4. Analysis of Elements Risk
Information of Building Structure
Element at risk related with physical or structural of building collected during fieldwork through building inventory and households’ interview. In physical aspect, information about wall material, floor material and roof material was gained. The result of this research found that most of households using concrete material for wall, floor, and roof.
Concrete wall they used was called “tembok” which is use brick as main material and covered with cement and plaster. This kind of wall material believed as the less damaged-wall-material during flood because its structure is solid and quite sturdy. There are mixed wall building found in this area which is combination of concrete wall material with plywood or bamboo sheet. Local people prefer to build this kind of structure because they want to make their house resistant to annual flood with flood depth 10 – 50 cm with low-priced material.
Cement and ceramic tiles was found as the most favourite floor material in this study area because this type of floor material is easy to clean after flood. Most of respondents use clay tiles as their roof material. From combination of wall material, floor material and roof material, five common structural type of building in this study area was found. More than a half of respondents in Kampung Melayu built the second-storey-house although they build it with low-priced material. This means they already have their own ways to minimize damage caused by floods that happen regularly in that area with considering of their own capacities.
Building Contents
Valuation of households’ asset is very important to examine the building contents vulnerability (Blaikie et al., 1994; Dutta and Tingsanchali, 2003; Sagala, 2006). Building contents found inside the houses in Kampung Melayu include include furniture and appliances. From this research revealed that local people located their valuable properties - for example: electronic appliances, books and important documents, jewellery, etc - in the second floor to avoid the damage during the flood. Most of total values of building contents are less than Rp 920,000. This value shows the value of building contents that could be damaged during flood.
5. Vulnerability Assessment
Vulnerability of Building Structure to Flood
Vulnerability of building structure in this research refers only to the damage of building material (wall, floor and roof) without considering other parts of the building. During the household interview, the percentage of damage with above mentioned codes were used and the respondents were asked to define the damage based on the condition of wall, floor and roof during the flood occurrence. Vulnerability for building structure in this research made in on scale from 0 to 1. The damage on structural building determined based on Nothing Happen (0%, NH), Half Collapse (50%, HC) and Collapse (100%, C) that is divided into seven values between 0 and 1.
The vulnerability curves for the five structural types of building are presented in Figure 3. The vulnerability of building structure found that among five structural types of building, the most prone to flood is the structural type 5 which is made from combination of mixed wall – mixed floor – asbestos roof. While the least vulnerable to flood is structural type 1 which is made from combination of brick wall - concrete floor - clay roof material.
Figure 3: Comparison of vulnerability curve for all structural type of building
Afterward, using slicing operation, the class of the structural vulnerability was generated into several classes: No Vulnerability (0), Low Vulnerability (<=0.2), Moderate Vulnerability (<=0.5), High Vulnerability (<=0.8), and Very High Vulnerability (<=1). The vulnerability map of structural type of buildings in the study area is represented in Figure 4.
Figure 4: Map of building structure vulnerability
Vulnerability of Building Contents to Flood In analysis of building contents vulnerability, quantification of people’s belongings found inside the houses was done using three lists of assets regarding to the three socio-economic levels. Subsequently, the damage of building contents was classified into five classes, i.e: 0 (No Damage), 0.25 (Slightly Affected), 0.5 (Moderately Affected), 0.75 (Highly Affected), and 1 (Destroyed) (see Table 2).
Table 2. Description of Stage of Damage of Building Contents Vulnerability Description
Adopted and modified from Guarin (2003)
Based on the building contents analysis, five classes of building contents vulnerability were generated: Very high (<=1), High (<=0.8), Moderate (<=0.4), Low (<=0.2) and No Vulnerability (0). Finally, the map of building contents vulnerability is plotted spatially in Figure 5.
Figure 5: Map of building content vulnerability
6. Coping Mechanism
Community Coping Mechanism
Previous knowledge of disaster events enhances the community's capacity to prepare for and cope with disasters. People have trained themselves to make some arrangements according to their capacity. Several coping strategies are employed by local people in order to reduce the damages caused by flood. While leaving their houses, the family normally moves together. After flooding, most of respondents stated that they seek loans or borrow from relatives or friends to pay for household expenses or sell their goods to get extra money. These coping strategies will exacerbate the flood impact, because poorer households, who cannot afford the cost of repair to their houses, would end up with searching for loans or selling their properties. It means that poor households will become even poorer after the flood occurrence.
Further analysis found that there are different coping strategies conducted by the female and male respondents. Female respondents mainly focus on the food, properties, home and family while the men respondents prefer to perform the technological or structural mechanisms. Respondents stated that they usually clean up the canal and their neighbourhood before the flood season is coming and also clean up the mud and the debris left by the flood together (gotong royong). The head of RT or RW usually lead and coordinate this activity among the community. People work hand in hand with their neighbour and leave their animosities, which is very uncommon condition found in the urban areas.
Flood Forecasting and Early Warning System (EWS) In flood forecasting and warnings, lurah, the head of RW or RT plays an important role in the study area. Dissemination of warnings is distributed using several tools, such as phone, mobile phone, handy talkie (HT), mosque’s speaker and door-to-door. Many people using mobile phone to send the SMS in order to distribute the flood warning to their friend or relatives.
Figure 6: The scheme of the Early Warning System in Kampung Melayu Source: ACF (2007)
Mainly, the flood forecasting within this community is through monitoring the water height at several Watergates and there are four alert levels. This information has not followed the action plan for the community, nor the community preparedness plans. Individual households must take their own decision whether they must go to evacuation center or not. Many different sources of flood warnings can drive people into uncertainty and panic. This research also reveals that the warnings are not always trusted by the people. More reliable information is needed because it will influence people’s response to flood.
7. Conclusions and Recommendations
It was found that the damage and vulnerability of the building structure is mainly determined by the material of the house and the fact that damage is strongly linked to the socio-economic status of the households. Buildings with more assets will possibly have a higher damage and also have higher building contents vulnerability even though the buildings located in a relatively lower water height and have more resistant building structure.
Due to the non-existence of action plan for the community, many households live in uncertainty and panic. This research also reveals that warnings are not always trusted by the people because sometimes this information has not been accurate. Flood management by the local government is not sufficient to address the flood risk in the study area because it still focuses only on the disaster response. Local government should pay attention to the preparedness action before flooding occurs in order to reduce the impact of floods. This research basically was aimed at utilising local knowledge, and did not consider other aspect; therefore further studies in this area need to include the hydrological and meteorological data, river morphology, etc. as additional inputs to produce a better and more accurate result on flood hazard assessment. Micro-level vulnerability and capacity assessment performed in this research could be enhanced by linking this analysis with a macro-level assessment of disaster and vulnerability contexts.
References
Mone Iye Cornelia
E-mail:mone_iye@yahoo.com
Marschiavelli1, Pramono Hadi2, Michael K. McCall3, Nanette Kingma3
1 Staff of Center for Marine Resources Surveys, National Coordinating Agency for Survey and Mapping (BAKOSURTANAL).
2 Faculty of Geography, Gadjah Mada University, Yogyakarta, Indonesia.
3 International Institute for Geo-Information Science and Earth Observation, Enschede, The Netherlands
Abstract
Flooding has become a serious problem in Jakarta. During 2007, Kampung Melayu was the worst hit by the floods. Community have different perceptions on disaster and develop different efforts to overcome floods. Therefore, local government and relevant institution should consider this situation and transform this information into valuable input in developing and implementing response plans.
This research tries to explore the vulnerability as well as the capacity for flood management based on people’s perception. Data gathered from a communitybased field data collection were processed using Geographical Information Systems (GIS) in order to illustrate the flood event in 2007 and its vulnerability in Kampung Melayu. In addition, information about existing coping mechanisms based on people’s perception also discussed in this research.
The result shows that the ability of people to cope with the flooding is linked with the capacity of the people itself. The capability of people to deal with flooding was influenced by several indicators based on their socio-economic characteristics. Several coping strategies employed by the local people and local government, however, not enough to cope with flooding in the study area. Local government and people in this area should cooperate to reduce the impact of flooding using the knowledge of how to address disasters effectively through people’s participation in the process. All information may represent the situation against the flood and can be used as a tool to improve the capacities of local people and minimize the flood risk.
1. Introduction
Annual flooding is one of environmental issues in the Jakarta area due to the worsening river management both in the upland and lowland. It is occur during rainy season, triggered by heavy seasonal rain in December, January, February and March. Flood in Jakarta has been recognized since the Dutch occupation era in Indonesia. Historical record illustrates that some enormous floods have occurred that killed some people and destroyed properties i.e., year 1699, 1714, and 1854. While in the last few decades, the flooding occurred in 1918, 1942, 1996, 2002 and 2007; caused some damages and some people were killed and lost. The flood on February 2007 was spread in almost 60% of Jakarta areas and its surroundings.
Flood in Jakarta is affected by several other factors, such as: morphological (approximately 40% areas of Jakarta Province is lowland area, some of them even below sea level), hydro-meteorological aspects (heavy rainfall and high sea tide during rainy seasons), land use change, decreasing the flow cross-section of the Ciliwung River because of garbage along the river and illegal settlement in the riverbanks, and also because of socio-cultural aspect, i.e: weak policy implementation, rapid urbanization, solid waste dumping and management. Vulnerability assessments are necessary in order to reduce the impact of the next flooding event in Jakarta. Government already conducted many surveys to measure and assess flood damage, but the vulnerability assessment at the micro-level (based on community data), including the coping mechanism of the community hasn’t been done yet. Therefore, this research is addressed to identify certain elements at risk, assess the vulnerability of each of the element at risk, and assess the coping mechanism/capacities of the community. The result of this research can be used as a valuable input for local governments for making appropriate actions, policies and programs in the context of flood hazard management in this study are in order to reduce the risk of the flood hazard and to apply it in urban areas of Kampung Melayu, Jakarta, Indonesia.
2. The Study Area
Jakarta, located on the northwest coast of the island of Java has an area of 661.52 km². The study area of this research is Kelurahan Kampung Melayu in Jatinegara District, East Jakarta, located along the Ciliwung River (Figure 1). It divided into 8 Rukun Warga (RW) and 114 Rukun Tetangga (RT). Geographically, Kampung Melayu located in the non-coastal area (± 15 Km from shoreline) and relatively flat. The altitude of Kampung Melayu is between 9 until 18 above the sea level. Kampung Melayu with area 0.48 km2 has high density of people.
Figure 1: Study area of the research
3. Research Method
There are several methods that already applied in order to assess vulnerability at different scales: national, mega city and local scale (Villagrán, 2006). In order to achieve the main objective of this research, therefore the vulnerability assessment at community level is the most appropriate. This research is focused on identifying and classifying certain elements at risk which would be affected by flood in study area, assessing the vulnerability of the elements at risk, and identifying the current coping capacities and coping strategies. Sampling method used during the fieldwork is non-proportional stratified random sampling. Overall, there were 83 point sample designed for field survey distributed in eight RWs.
Building Inventory
Building inventory aimed to gain building information for database and to verify the respondents’ answer during interview. The activities in building inventory were measuring the height of floor from street and above the surface and observing the physical aspect of building such as wall material, floor material, roof material, number of floor, etc. All information was stored in mobile PDA and added as an attribute for each building points. The attributes for building inventory were: building_id, date, owner, function, wall material, floor material, roof material, number of floor, size, floor height from surface, floor height from the street and photo’s number.
Household interview
The purpose of household interview using questionnaires was to get detailed information about element at risk including asset and damage estimation for each household, and households’ coping mechanism. In-depth interview and consultation were held with occupants of the buildings identified as being risk. Each interview lasted between 30 and 45 minutes.
Data Analysis
Data collected from the building inventory were able to be analysed after fieldwork because already in digital format. The shape file from mobile PDA was transferred into PC and using Arc View software, all of data collected during fieldwork can be retrieved easily (see Figure 2). However, information from interview had to be converted from hardcopy format (questionnaires) into digital format first. Data from building inventory and interview processed spatially using ILWIS 3.3 and Arc View 3.3.
There are several parts in this stage: identification of certain element at risk, vulnerability assessment and coping mechanism analysis. For vulnerability analysis, author conducted spatial analysis using GIS software. It was used to plot the physical element at risk (building material) and results from analysis. Author used kriging to interpolate the point data using Gaussian model. For coping mechanism analysis, author using descriptive analysis.
Figure 2: Illustration of point with attributes collected during the fieldwork
4. Analysis of Elements Risk
Information of Building Structure
Element at risk related with physical or structural of building collected during fieldwork through building inventory and households’ interview. In physical aspect, information about wall material, floor material and roof material was gained. The result of this research found that most of households using concrete material for wall, floor, and roof.
Concrete wall they used was called “tembok” which is use brick as main material and covered with cement and plaster. This kind of wall material believed as the less damaged-wall-material during flood because its structure is solid and quite sturdy. There are mixed wall building found in this area which is combination of concrete wall material with plywood or bamboo sheet. Local people prefer to build this kind of structure because they want to make their house resistant to annual flood with flood depth 10 – 50 cm with low-priced material.
Cement and ceramic tiles was found as the most favourite floor material in this study area because this type of floor material is easy to clean after flood. Most of respondents use clay tiles as their roof material. From combination of wall material, floor material and roof material, five common structural type of building in this study area was found. More than a half of respondents in Kampung Melayu built the second-storey-house although they build it with low-priced material. This means they already have their own ways to minimize damage caused by floods that happen regularly in that area with considering of their own capacities.
Building Contents
Valuation of households’ asset is very important to examine the building contents vulnerability (Blaikie et al., 1994; Dutta and Tingsanchali, 2003; Sagala, 2006). Building contents found inside the houses in Kampung Melayu include include furniture and appliances. From this research revealed that local people located their valuable properties - for example: electronic appliances, books and important documents, jewellery, etc - in the second floor to avoid the damage during the flood. Most of total values of building contents are less than Rp 920,000. This value shows the value of building contents that could be damaged during flood.
5. Vulnerability Assessment
Vulnerability of Building Structure to Flood
Vulnerability of building structure in this research refers only to the damage of building material (wall, floor and roof) without considering other parts of the building. During the household interview, the percentage of damage with above mentioned codes were used and the respondents were asked to define the damage based on the condition of wall, floor and roof during the flood occurrence. Vulnerability for building structure in this research made in on scale from 0 to 1. The damage on structural building determined based on Nothing Happen (0%, NH), Half Collapse (50%, HC) and Collapse (100%, C) that is divided into seven values between 0 and 1.
The vulnerability curves for the five structural types of building are presented in Figure 3. The vulnerability of building structure found that among five structural types of building, the most prone to flood is the structural type 5 which is made from combination of mixed wall – mixed floor – asbestos roof. While the least vulnerable to flood is structural type 1 which is made from combination of brick wall - concrete floor - clay roof material.
Figure 3: Comparison of vulnerability curve for all structural type of building
Afterward, using slicing operation, the class of the structural vulnerability was generated into several classes: No Vulnerability (0), Low Vulnerability (<=0.2), Moderate Vulnerability (<=0.5), High Vulnerability (<=0.8), and Very High Vulnerability (<=1). The vulnerability map of structural type of buildings in the study area is represented in Figure 4.
Figure 4: Map of building structure vulnerability
Vulnerability of Building Contents to Flood In analysis of building contents vulnerability, quantification of people’s belongings found inside the houses was done using three lists of assets regarding to the three socio-economic levels. Subsequently, the damage of building contents was classified into five classes, i.e: 0 (No Damage), 0.25 (Slightly Affected), 0.5 (Moderately Affected), 0.75 (Highly Affected), and 1 (Destroyed) (see Table 2).
Table 2. Description of Stage of Damage of Building Contents Vulnerability Description
Adopted and modified from Guarin (2003)
Based on the building contents analysis, five classes of building contents vulnerability were generated: Very high (<=1), High (<=0.8), Moderate (<=0.4), Low (<=0.2) and No Vulnerability (0). Finally, the map of building contents vulnerability is plotted spatially in Figure 5.
Figure 5: Map of building content vulnerability
6. Coping Mechanism
Community Coping Mechanism
Previous knowledge of disaster events enhances the community's capacity to prepare for and cope with disasters. People have trained themselves to make some arrangements according to their capacity. Several coping strategies are employed by local people in order to reduce the damages caused by flood. While leaving their houses, the family normally moves together. After flooding, most of respondents stated that they seek loans or borrow from relatives or friends to pay for household expenses or sell their goods to get extra money. These coping strategies will exacerbate the flood impact, because poorer households, who cannot afford the cost of repair to their houses, would end up with searching for loans or selling their properties. It means that poor households will become even poorer after the flood occurrence.
Further analysis found that there are different coping strategies conducted by the female and male respondents. Female respondents mainly focus on the food, properties, home and family while the men respondents prefer to perform the technological or structural mechanisms. Respondents stated that they usually clean up the canal and their neighbourhood before the flood season is coming and also clean up the mud and the debris left by the flood together (gotong royong). The head of RT or RW usually lead and coordinate this activity among the community. People work hand in hand with their neighbour and leave their animosities, which is very uncommon condition found in the urban areas.
Flood Forecasting and Early Warning System (EWS) In flood forecasting and warnings, lurah, the head of RW or RT plays an important role in the study area. Dissemination of warnings is distributed using several tools, such as phone, mobile phone, handy talkie (HT), mosque’s speaker and door-to-door. Many people using mobile phone to send the SMS in order to distribute the flood warning to their friend or relatives.
Figure 6: The scheme of the Early Warning System in Kampung Melayu Source: ACF (2007)
Mainly, the flood forecasting within this community is through monitoring the water height at several Watergates and there are four alert levels. This information has not followed the action plan for the community, nor the community preparedness plans. Individual households must take their own decision whether they must go to evacuation center or not. Many different sources of flood warnings can drive people into uncertainty and panic. This research also reveals that the warnings are not always trusted by the people. More reliable information is needed because it will influence people’s response to flood.
7. Conclusions and Recommendations
It was found that the damage and vulnerability of the building structure is mainly determined by the material of the house and the fact that damage is strongly linked to the socio-economic status of the households. Buildings with more assets will possibly have a higher damage and also have higher building contents vulnerability even though the buildings located in a relatively lower water height and have more resistant building structure.
Due to the non-existence of action plan for the community, many households live in uncertainty and panic. This research also reveals that warnings are not always trusted by the people because sometimes this information has not been accurate. Flood management by the local government is not sufficient to address the flood risk in the study area because it still focuses only on the disaster response. Local government should pay attention to the preparedness action before flooding occurs in order to reduce the impact of floods. This research basically was aimed at utilising local knowledge, and did not consider other aspect; therefore further studies in this area need to include the hydrological and meteorological data, river morphology, etc. as additional inputs to produce a better and more accurate result on flood hazard assessment. Micro-level vulnerability and capacity assessment performed in this research could be enhanced by linking this analysis with a macro-level assessment of disaster and vulnerability contexts.
References
- ACF (2007). Penjajakan Kapasitas SATLINMAS di Tiga Kelurahan DKI Jakarta. Jakarta, Indonesia.
- Blaikie, P., T. Cannon, I. Davis and B. Wisner (1994). At Risk: Natural Hazards, People's Vulnerability, and Disasters. London, Routledge.
- Dutta, D. and T. Tingsanchali (2003). "Development of Loss Function for Flood Urban Flood Risk Analysis in Bangkok." New Technologies for Urban Safety of Mega Cities in Asia, Tokyo.
- Guarin, G. P. (2003). Flood Risk Assessment for the Town of San Sebastian in Guatemala. ITC, International Institute for Geo-Information Science and Earth Observation. MSc Thesis. Enschede, The Netherlands.
- Sagala, S. A. H. (2006). Analysis of Flood Physical Vulnerability in Residential Areas, Case Study: Naga City, The Phillipines. ITC, International Institute for Geo-Information Science and Earth Observation. MSc Thesis. Enschede, The Netherlands.
- Villagrán, J. C. (2006). Vulnerability:A Conceptual and Methodological Review, Bonn, Germany., UNU Institute for Environment and Human Security (UNUEHS)