LANDSLIDE RISK ASSESSMENT:
AN INTEGRATED APPROACH IN EVALUATING THE AERIAL HAZARD EXTENT, VULNERABILITY, AND RISK PERCEPTION OF THE SELECTED PEOPLE IN SAN MATEO, RIZAL

Despite being one of the fastest growing municipality in Rizal, San Mateo still experiences extreme landslides events. This poses a serious problem to the safety of its current and future residents since the municipality is expected to carry the surplus of the population of the adjacent cities like Quezon City and Marikina City in the future. With this, the main objective of this study is to conduct two different landslide risk assessments using landslide hazard map (LHM), landslide vulnerability map (LVM), and evaluation of the risk-perceptions of the selected people in the municipality. The first landslide risk map (LRM) was produced using the LHM and LVM. LHM was produced through the relational analysis of the different environmental factors and the landslide inventory processed using logistic regression and ANN analysis. LVM and the vulnerability of the people was measured using SVI. Results of the LHM showed that the highly susceptible areas are located in the central and the eastern part of the municipality. Fortunately, no highly susceptible areas are found in the areas where the most vulnerable respondents reside, based on the LVM produced. However, these same vulnerable respondents were considered as highly at risk by the first LRM since they are vulnerable and are still exposed to hazards. Additionally, all the respondents are willing to accept the risks through boosting their knowledge on landslides such as attending seminars and etc. No one is willing to leave the municipality or only pay for their safety. Also, all the respondents were classified as risk averse and risk seekers based on their answers. Most of the locations and concentrations of the people considered as highly at risk by the first LRM is similar with the results of the second LRM produced from the risk-perception of the people. However, rare cases still exist since the individual perceptions of the people are based on their own experiences and rationality. In conclusion, both the two landslide risk maps have their own strengths and weaknesses. While one approach can stand on its own, the two approaches in landslide risk assessments are best when combined.

A. Background of the Study

Philippines is one of the many countries in Asia gravely exposed to a vast range of natural hazards and disasters (The Guardian, 2010). From extreme typhoons and almost endless rainfalls, flashfloods, volcanic eruptions, earthquakes, tsunamis, sudden slope failures, down to the most unexpected sinkholes and whirlpools, Philippines really seems to be a converging zone of natural calamities. Even though many countries in Asia are more prone to natural hazard than the Philippines (Assilzadeh, 2010), the social and economic capabilities of this country make it more vulnerable. Japan, for example, has a higher likelihood of disaster occurrence than the Philippines, but it does not make it more vulnerable. According to the 2010 World Disasters Report of the International Federation of Red Cross and Crescent Societies (IFRC), the probability for a Filipino to be killed by a natural disaster is 17 times larger than the probability of a Japanese being killed by the same reason. This is because the difference of the quality of the urban and environmental planning strategies – or the urban risk divide – between the two countries is high, making the former more vulnerable.

This urban risk divide is severe in Asia, especially in the developing countries such as the Philippines. As a third world country situated in a latitude where typhoons frequently pass, and in a region dominated by volcanoes and fault lines, the number of hazards, risks and vulnerabilities that threaten the country is nearly impossible to measure. One leading natural hazard that is widespread in the country is the landslide.

Landslide, put simply, refers to the downward movement of rocks, soil, and debris on an oblique surface or slope (USGS, 2004). Landslides are usually naturally-occurring in the environment, but the incessant disuse of human activities made some of these events to become human-induced. The elements that usually affect the landslide occurrences are slope angle, rainfall, weathering, land cover, overloading, and geology.

The blending of the naturally immense amount of rainfall brought by typhoons and monsoons, and the geological perils (earthquake and volcanic eruptions) that exist in the country transforms the terrain of the Philippines into a landslide-susceptible state. In her interview last 2010, former senior science research specialist of Mines and Geosciences Bureau (MGB) Angelica Sajona told Integrated Regional Information Networks (IRIN) News that “[with] the changing weather conditions and amount of rainfall each season brings, we could categorically state that the number of landslide occurrences is increasing," With 80% of the country's total land area being prone to landslide, MGB has also reported that the country ranks fourth in the list of the most vulnerable countries to landslide hazard and risk in Asia (2010).

Clearly, the intensifying vulnerability of the entire country is associated with the increasing population of its vulnerable communities. As the population grows, more and more people will continue to move away from the city center and settle at the periphery – that is, a place exhibiting rapid urbanization but lower economic and social development (Chiu and Eidsvig, 2016). This study used the case of the National Capital Region and Region – IVA CALABARZON as the example. In the future, surge of population movements from the National Capital Region to the adjacent region is expected to occur. Unfortunately, for this set-up, the general geomorphological make-up of the provinces in the adjacent region, in this case the Region IV-A, is mountainous and rugged, decreasing the number of safe habitable areas while at the same time increasing the number of vulnerable population to landslide. According to a map issued by the Mines and Geosciences Bureau (2011), 59 out of the 141 and municipalities (PSA, 2015) in the region are highly susceptible to landslides. Moreover, among the five provinces in the region, Rizal has the largest area of highly susceptible space for landslides (DENR, 2012). This poses a serious problem to the province because it is currently one of the nearest provinces in Metro Manila that has the highest probability to become the basin of population pours from the country's capital region. As the population of National Capital Region increases, the number of population who will settle in landslide-susceptible areas in Rizal will also increase because, as stated previously, the safe zones in the province are very limited.

This projected problem, however, has already come in to fruition because numerous landslide incidents in the region have already been recorded in the past few years. Several municipalities have already experienced disastrous landslide occurrences, one of which is San Mateo. The latest occurrence was verified last August 2017, where a major highway in Barangay Silangan was destroyed by a slope failure (ABS-CBN News, 2017). Another landslide incident happened last July 2009, where two houses were destroyed (GMA News, 2009). Fortunately, no deaths were recorded in the landslide events that have occurred in San Mateo. However, deaths were confirmed in the landslide incidents in other municipalities. Clearly, safety measures and preparations must be immediately done to further prevent these tragic events from occurring in the municipality. As such, landslide risk planning and mitigation in San Mateo must be carefully administered to keep the communities and the infrastructures away from deadly hazards.

B. Research Objectives

The main objective of this study is to assess the overall condition of the selected people on landslide risks using two completely different approaches: one is founded on GIS-based techniques and quantitative methods; the other is based on the landslide risk – perception of the exposed people in the municipality. Also, this study evaluates the difference in the performance of the two landslide risk assessment approaches. Through performing an integrated approach in the assessment of the landslide hazard extent in the municipality, risk-perception of the selected people in the susceptible areas, and the vulnerability of the selected people in San Mateo, Rizal, the desired final maps were produced. Consequently, specific objectives were further identified to expound the aim:

1. To produce a landslide susceptibility map determining the extent of the landslide hazards in the municipality using logistic regression model and artificial neural network in IBM SPSS Statistics 20 and digital cartography in ArcGIS 10.2.2;
2. To assess the socioeconomic, demographic, and physical vulnerability of the people in the susceptible areas using the Social Vulnerability Index, and produce a vulnerability map displaying the areas of vulnerability to landslide hazards based on the assessments conducted;
3. To produce a landslide risk map based on the landslide hazard map and landslide vulnerability map;
4. To perform an analysis measuring the extent of knowledge and landslide risk-perception of the selected people in the susceptible areas using Preparedness Model and Willingness Diagram, and produce a landslide risk map based on the analysis performed, and finally;
5. To visually compare the two landslide risk maps and construct a possible explanation of their similarities and differences.

Statement of the Problem

This study aimed to answer one general question: what is the current condition of the selected people of San Mateo in terms of landslide risks? The specific objectives of the study remain unclear unless questions were formulated to serve as guides. To further specify and elucidate the detailed aims of this study, below are the research questions that address the methods of every objectives provided:

1. Based on the results of landslide susceptibility map, what areas of San Mateo, Rizal are highly susceptible, moderately susceptible, lowly susceptible, and not susceptible to landslides?
2. Based on the results of the landslide vulnerability map, where are most of the vulnerable population in the municipality located?
3. Based on the landslide risk map produced through combining the landslide hazard map and landslide vulnerability map, where are the respondents who are at high risk mostly concentrated?
4. Based on the preparedness model, how do the selected locals perceive their level of preparedness for landslide events and what factors contribute in shaping their preparation?
5. Based on the willingness diagram, how willing are the respondents to address the issues of landslide risks?
6. Based on the landslide – risk perception of the exposed people in the municipality, where are the people who think they are at risk mostly located?
7. How do the two landslide risk maps produced differ from one another?

C. Significance of the Study

Despite its relatively small area, San Mateo is one of the fastest growing municipalities in the province of Rizal (DILG, n.d.). Aside from the fact that it is one of the nearest municipalities of Rizal from the National Capital Region, San Mateo is famous for its tourism accomplishments and migration records. The municipality has one of the most perfect settings of residential areas in its upland area part – the mountains, trees, and high altitudes provide a remarkable rural experience while at the same time maintaining the urban benefits because of its proximity to the big cities such as Quezon City and Marikina City.

In the future, it can be projected that huge population of people will migrate to the municipality for permanent residency. This event is not far from reality since it is directly beside the most populous city in the country (Quezon City), and spaces are always limited. Given this fact, San Mateo is one of the few municipalities that might receive the largest surplus of the population of the Quezon city in the future.

This projected phenomenon, however, poses a big problem since the residential area of San Mateo is relatively small. The high population flow might force some to live in the high-altitude lands, where landslide hazards are extremely prominent. However, this outcome could be avoided if the landslide risks of some of the people will be assessed now.

This study conducts an assessment on the current conditions of the selected people on landslide risk using two different approach. The results of this study can help in forming an overview of the current situation of the selected people in the municipality, while at the same time evaluates the effectiveness and the performance of the two landslide risk approaches conducted. Considering this fact, the municipality of San Mateo can benefit from the results of this study. This study will also greatly help future researchers who might want to conduct a landslide risk assessment and give them ideas on how to conduct one.

D. Scope and Limitations

The study was conducted inside the administrative boundaries of San Mateo, and the results of the study will only be applicable within the municipality. Most of the data used in generating the landslide hazard map were from credible open source websites. The landslide inventory was produced by the researcher using remote sensing, and no field validations were conducted due to time and capacity constraints. The landslide hazard map made using several statistical analyses is just a general representation of the actual areas with landslide hazards and does not assure a hundred percent precision in defining the boundaries of the real-world hazards. Also, the areas delineated as landslide hazards only show the probability of slope failure of the land. It does not include the buffer zones in which the landslides will pour. Most of the vulnerability measures only pertain to the head of the family surveyed, and not to the entire household/business establishment. The accuracy of the physically-made risk map was based on the accuracy of the final landslide hazard map and the final landslide vulnerability map. This means that the flaws and limitations of the prerequisite maps will also be inherited by the landslide risk map. Also, the method on gathering the perceptions of the respondents were not purely qualitative, as it was only conducted using a questionnaire with open ended questions. Lastly, the general analysis on the overall results of the study only included the 82 respondents who willingly participated, and the conclusions that were generated were only applied to them.

A. Theoretical Framework

The study is grounded in several theoretical frameworks that guide the understanding of landslide risk assessment:

1. Risk Assessment Theory

   This framework emphasizes the systematic process of evaluating potential risks by considering both hazard and vulnerability components. It provides the foundation for understanding how different factors contribute to landslide risks.

2. Social Vulnerability Theory

   This theory focuses on the social, economic, and demographic factors that make certain populations more vulnerable to natural hazards. It helps explain why some communities are more affected by landslides than others.

3. Risk Perception Theory

   This framework examines how individuals and communities perceive and respond to risks. It helps understand the gap between scientific risk assessment and community risk perception.

B. Conceptual Framework

The study's conceptual framework integrates three main components:

1. Hazard Assessment
   • Environmental factors
   • Geological conditions
   • Historical landslide data
2. Vulnerability Assessment
   • Socioeconomic factors
   • Demographic characteristics
   • Physical infrastructure
3. Risk Perception Analysis
   • Community knowledge
   • Risk awareness
   • Preparedness levels

C. Review of Related Studies

Several studies have contributed to the understanding of landslide risk assessment:

1. GIS-based Landslide Assessment

   Studies by Guzzetti et al. (2005) and Pradhan & Lee (2010) demonstrate the effectiveness of GIS and statistical methods in landslide susceptibility mapping.

2. Social Vulnerability Analysis

   Research by Cutter (1996) and Birkmann (2006) provides frameworks for assessing social vulnerability to natural hazards.

3. Risk Perception Studies

   Works by Mileti (1999) and Alexander (2000) explore the relationship between scientific risk assessment and community risk perception.

D. Synthesis of Literature

The review of literature reveals several key findings:

• The importance of integrating multiple approaches in landslide risk assessment
• The need to consider both physical and social factors in vulnerability analysis
• The significance of community participation in risk assessment and management
• The role of local knowledge in complementing scientific risk assessment

This chapter provides a comprehensive explanation on the procedures in generating the three types of map used in the study – (1) one landslide hazard map, (2) one map of the vulnerable population in the exposed areas, and (3) two landslide risk maps. This chapter also expounds the procedure on the analysis of the landslide risk - perception and landslide vulnerability of the selected population in the exposed areas of the municipality.

A. Study Area

The area to be studied comprises the entire municipality of San Mateo, Rizal – a first class municipality with a total land area of 55.09 km². It is bordered by Quezon City in the west, Marikina City and Antipolo City in the south, Antipolo City in the east, and the municipality of Rodriguez in the north. The western half of the municipality is mostly flat, while the eastern half is dominated by flat highlands, mountains, and valleys. The more urbanized areas are located in the western part of the municipality, while the more forested areas are located in the east. The highest point on the surface is about 453 meters, located in the east, and the lowest point located in the west is about 4 meters above mean sea level.

B. Data Collection

The data used for the study were divided into three (3) categories based on their purpose: (1) landslide hazard data; (2) landslide vulnerability data; and (3) landslide risk-perception data. The acquisition of all these data is essential in accomplishing the desired objectives of the study.

Landslide Hazard Data

The data used in producing a landslide hazard map of the municipality include the administrative boundary, soil type, distance to road network, distance to stream drainage network, elevation, slope angle gradient, slope aspect, plan curvature, areas with unstable slope and landslide inventory. The administrative boundary data was downloaded from the website of the Humanitarian Data Exchange website – an open platform for sharing data that was launched in July 2014. It contains the indicative boundaries of the municipality of San Mateo, Rizal as observed at the end of June 2016 as per the Philippine Geographic Standard Code (PSGC) dataset.

Landslide Vulnerability Data

To assess the vulnerability of the selected people in the exposed areas in the municipality, three classifications of vulnerability were identified in the study based on the Social Vulnerability Index: (1) socioeconomic vulnerability; (2) demographic vulnerability; (3) and physical vulnerability. The information used in measuring the socioeconomic vulnerability of the selected participants include annual income, employment status and educational attainment. On the other hand, age, sex, civil status, and existence of disability were used as measures for the demographic vulnerability. The physical vulnerability was evaluated using the type of house (apartment, single building, etc.) and ownership of a functioning vehicle.

Landslide Risk Data

Two risk maps were generated in this study, one is based on the hazard and vulnerability map, the other is based on the risk-perception of the selected people in the exposed areas in San Mateo. The data used in producing the first map were the generated landslide hazard map and landslide vulnerability map. On the other hand, the information needed to produce the second risk map were risk awareness, risk perception, risk understanding, capacity, risk response orientation, and level of preparedness.

C. Participants

The participants in the study are the households or business establishments currently located in the exposed areas. Participants were chosen from each exposure class based on the hazard map. The researcher was able to congregate 82 households/business establishments as respondents for the study – 24 were located in the no exposure, 40 in low exposure, 7 in moderate exposure, and 11 in the areas considered as high exposure. The participants were chosen using cluster sampling method. To ensure that the participants are accessible for the researcher, only those living within 20 meters away from the road were selected. Additionally, the researcher assured that the chosen participants were already living in the municipality since the year 2004.

D. Methodological Framework

The overall procedure in accomplishing all the objectives of the study was divided into three phases:

1. First Phase – Making the Landslide Hazard Map: This phase includes all the methods and procedures performed to generate the landslide hazard map of the municipality. The idea of the landslide hazard map produced in the study was founded on the concept of GIS-based landslide susceptibility mapping technique.
2. Second Phase – Making the Landslide Vulnerability Map: Data on the vulnerability of the selected people in the exposed areas in the municipality were gathered through a survey-questionnaire form distributed during the house-to-house survey.
3. Third Phase – Making the Landslide Risk Maps: The final phase of the entire methodological framework involved all the process conducted in generating the two landslide risk maps. The first risk map was produced from the final landslide hazard map (LHM) and the final landslide vulnerability map (LVM). The second risk map was produced solely from the risk-perceptions of the respondents.

E. Ethical Considerations

The study was conducted in accordance with the ethical guidelines set by the University of the Philippines. The respondents were informed about the purpose of the study and their rights as participants. Their consent was obtained before conducting the survey. The data collected were kept confidential and were used only for the purpose of this study.

A. Landslide Hazard Assessment

The landslide hazard assessment was conducted using GIS-based techniques and statistical analysis. The following factors were considered in the analysis:

• Slope angle gradient
• Slope aspect
• Plan curvature
• Distance to road network
• Distance to stream drainage network
• Elevation
• Soil type
• Areas with unstable slope

B. Vulnerability Assessment

The vulnerability assessment was conducted using the Social Vulnerability Index (SVI), which considers three main components:

1. Socioeconomic Vulnerability
   • Annual income
   • Employment status
   • Educational attainment
2. Demographic Vulnerability
   • Age
   • Sex
   • Civil status
   • Existence of disability
3. Physical Vulnerability
   • Type of house
   • Ownership of functioning vehicle

C. Risk Assessment

Two types of risk assessment were conducted:

1. GIS-based Risk Assessment

   This assessment combined the landslide hazard map and vulnerability map to produce a comprehensive risk map. The risk levels were classified into four categories: no risk, low risk, moderate risk, and high risk.

2. Risk Perception Assessment

   This assessment was based on the responses of the selected participants in the exposed areas. The following factors were considered:

   • Risk awareness
   • Risk perception
   • Risk understanding
   • Capacity
   • Risk response orientation
   • Level of preparedness

D. Comparison of Risk Assessment Methods

The study compared the results of the GIS-based risk assessment and the risk perception assessment. The comparison revealed:

• Areas of agreement between the two methods
• Areas of discrepancy that need further investigation
• Factors that influence the differences in risk assessment results

E. Implications for Disaster Risk Reduction

The findings of the study have several implications for disaster risk reduction in San Mateo, Rizal:

• Need for targeted interventions in high-risk areas
• Importance of community-based disaster risk reduction programs
• Role of local government in implementing risk reduction measures
• Integration of scientific and local knowledge in risk assessment

A. Summary of Findings

The study successfully assessed the landslide risks in San Mateo, Rizal using both GIS-based techniques and risk perception analysis. The main findings include:

1. Landslide Hazard Assessment
   • Identification of high-risk areas based on various environmental factors
   • Mapping of landslide susceptibility zones
   • Validation of the hazard map using historical landslide data
2. Vulnerability Assessment
   • Evaluation of socioeconomic, demographic, and physical vulnerability
   • Identification of vulnerable population groups
   • Mapping of vulnerability distribution across the municipality
3. Risk Assessment
   • Integration of hazard and vulnerability assessments
   • Comparison of GIS-based and perception-based risk assessments
   • Identification of areas requiring immediate attention

B. Conclusions

Based on the findings, the following conclusions were drawn:

1. The integration of GIS-based techniques and risk perception analysis provides a more comprehensive understanding of landslide risks in the municipality.
2. The eastern part of San Mateo, characterized by steep slopes and high elevation, shows higher susceptibility to landslides.
3. Vulnerability to landslides is influenced by various socioeconomic, demographic, and physical factors.
4. There are significant differences between scientific risk assessment and community risk perception, highlighting the need for better risk communication.
5. The study provides valuable information for disaster risk reduction planning and implementation in the municipality.

C. Recommendations

The following recommendations are proposed based on the study findings:

1. For Local Government
   • Implement targeted interventions in high-risk areas
   • Develop and enforce land use policies that consider landslide risks
   • Strengthen early warning systems and emergency response capabilities
   • Allocate resources for infrastructure improvements in vulnerable areas
2. For Community
   • Enhance community awareness and understanding of landslide risks
   • Participate in disaster risk reduction programs and activities
   • Develop community-based early warning systems
   • Establish community emergency response teams
3. For Future Research
   • Conduct more detailed studies on specific high-risk areas
   • Investigate the effectiveness of different risk reduction measures
   • Develop improved methods for integrating scientific and local knowledge
   • Study the long-term impacts of climate change on landslide risks

D. Final Remarks

The study contributes to the understanding of landslide risks in San Mateo, Rizal, and provides a framework for similar assessments in other areas. The integration of scientific and local knowledge approaches offers a more holistic view of disaster risks, which is essential for effective disaster risk reduction. The findings and recommendations of this study can serve as a basis for developing and implementing appropriate risk reduction measures in the municipality.

1. Adger, W. N. (2006). Vulnerability. Global Environmental Change, 16(3), 268-281.
2. Alcántara-Ayala, I. (2002). Geomorphology, natural hazards, vulnerability and prevention of natural disasters in developing countries. Geomorphology, 47(2-4), 107-124.
3. Alexander, D. (2000). Confronting Catastrophe: New Perspectives on Natural Disasters. Oxford University Press.
4. Birkmann, J. (2006). Measuring vulnerability to natural hazards: towards disaster resilient societies. United Nations University Press.
5. Cardona, O. D. (2004). The need for rethinking the concepts of vulnerability and risk from a holistic perspective: a necessary review and criticism for effective risk management. In Mapping vulnerability: disasters, development and people (pp. 37-51). Earthscan.
6. Cutter, S. L. (1996). Vulnerability to environmental hazards. Progress in Human Geography, 20(4), 529-539.
7. Dai, F. C., Lee, C. F., & Ngai, Y. Y. (2002). Landslide risk assessment and management: an overview. Engineering Geology, 64(1), 65-87.
8. Fell, R., Corominas, J., Bonnard, C., Cascini, L., Leroi, E., & Savage, W. Z. (2008). Guidelines for landslide susceptibility, hazard and risk zoning for land use planning. Engineering Geology, 102(3-4), 85-98.
9. Glade, T. (2003). Vulnerability assessment in landslide risk analysis. Die Erde, 134(2), 123-146.
10. Guzzetti, F., Reichenbach, P., Cardinali, M., Galli, M., & Ardizzone, F. (2005). Probabilistic landslide hazard assessment at the basin scale. Geomorphology, 72(1-4), 272-299.
11. Haque, U., Da Silva, P. F., Devoli, G., Pilz, J., Zhao, B., Khaloua, A., ... & Glass, G. E. (2019). The human cost of global warming: Deadly landslides and their triggers (1995-2014). Science of the Total Environment, 682, 673-684.
12. Hewitt, K. (1997). Regions of Risk: A Geographical Introduction to Disasters. Longman.
13. Hufschmidt, G., Crozier, M., & Glade, T. (2005). Evolution of natural risk: research framework and perspectives. Natural Hazards and Earth System Sciences, 5(3), 375-387.
14. International Federation of Red Cross and Red Crescent Societies. (2014). World Disasters Report 2014: Focus on Culture and Risk. IFRC.
15. Kappes, M. S., Keiler, M., von Elverfeldt, K., & Glade, T. (2012). Challenges of analyzing multi-hazard risk: a review. Natural Hazards, 64(2), 1925-1958.
16. Kjekstad, O., & Highland, L. (2009). Economic and social impacts of landslides. In Landslides–Disaster Risk Reduction (pp. 573-587). Springer.
17. Lacasse, S., & Nadim, F. (2009). Landslide risk assessment and mitigation strategy. In Landslides–Disaster Risk Reduction (pp. 31-61). Springer.
18. Mileti, D. S. (1999). Disasters by Design: A Reassessment of Natural Hazards in the United States. Joseph Henry Press.
19. National Disaster Risk Reduction and Management Council. (2011). National Disaster Risk Reduction and Management Plan 2011-2028. NDRRMC.
20. Petley, D. (2012). Global patterns of loss of life from landslides. Geology, 40(10), 927-930.
21. Philippine Atmospheric, Geophysical and Astronomical Services Administration. (2011). Climate Change in the Philippines. PAGASA.
22. Philippine Institute of Volcanology and Seismology. (2014). Landslide Susceptibility Mapping in the Philippines. PHIVOLCS.
23. Pradhan, B., & Lee, S. (2010). Landslide susceptibility assessment and factor effect analysis: backpropagation artificial neural networks and their comparison with frequency ratio and bivariate logistic regression modelling. Environmental Modelling & Software, 25(6), 747-759.
24. United Nations International Strategy for Disaster Reduction. (2009). UNISDR Terminology on Disaster Risk Reduction. UNISDR.
25. Van Westen, C. J., Castellanos, E., & Kuriakose, S. L. (2008). Spatial data for landslide susceptibility, hazard, and vulnerability assessment: an overview. Engineering Geology, 102(3-4), 112-131.
26. Wisner, B., Blaikie, P., Cannon, T., & Davis, I. (2004). At Risk: Natural Hazards, People's Vulnerability and Disasters. Routledge.
27. World Bank. (2010). Natural Hazards, UnNatural Disasters: The Economics of Effective Prevention. World Bank.
28. World Health Organization. (2002). Gender and Health in Disasters. WHO.
29. Yin, K. L., & Yan, T. Z. (1988). Statistical prediction model for slope instability of metamorphosed rocks. In Proceedings of the 5th International Symposium on Landslides (pp. 1269-1272).
30. Zhou, C. H., Lee, C. F., Li, J., & Xu, Z. W. (2002). On the spatial relationship between landslides and causative factors on Lantau Island, Hong Kong. Geomorphology, 43(3-4), 197-207.