ESTABLISHMENT OF PHILIPPINE FLOOD RISK INDEX BY PROVINCE BASED ON NATURAL AND SOCIAL FACTORS

Transcription

1 ESTABLISHMENT OF PHILIPPINE FLOOD RISK INDEX BY PROVINCE BASED ON NATURAL AND SOCIAL FACTORS Jerry Austria FANO Supervisor: Prof. Kuniyoshi TAKEUCHI MEE ABSTRACT This thesis offers a measure to formulate a Philippine Flood Risk Index (P-FRIc) by province in coverage, on the basis of Pressure and Release (PAR) Model which consists of five (5) key Indexes: Hazard (H), Exposure (E), Vulnerability (V), Coping Capacity using Soft countermeasures (C S ) and Hard countermeasures (C H ). The basic equation "Risk = Hazard Vulnerability" (Wisner, B. et. al., 2004) is modified (Kannami, Arai et. al, Master Thesis, 2004) to calculate the P-FRIc, expressed as: Hazard Exposure Vulnerability H E V P FRIc Capacity ( Soft Measures Hard Measure) C C This study identified and analyzed the indicators based on natural and social conditions that compose each index of P-FRIc. It quantified the indexes that include hazards (e.g., proneness to typhoons, topography), exposure (e.g., lowland population density, population growth per province) and progression of vulnerability (e.g., peoples socioeconomic conditions) and coping capacity (e.g., hard and soft measures). It also analyzed the distribution of indexes for different provinces comparatively to draw the usable policies to be considered in future flood mitigation administration. Each index is composed of three (3) kinds of datasets which are called Indicators. P-FRIc is then used to assess the current potential risk to floods for the 82 provinces. The results of analysis indicated high flood-risk for provinces such as Metro Manila, Albay, Pampanga, Zambales, Negros Occidental, Cavite, etc. The assessment of Metro Manila as high flood risk area was confirmed on September 2009 when 80% of the capital was submerged by flashflood caused by Typhoon Ketsana that killed more than 300 people (Nilo & Espinueva, 2009). This implies the effectiveness of flood risk index. Key words: Philippine Flood Risk Index, Hazard, Exposure, Vulnerability, Coping Capacity, PAR model S H INTRODUCTION The Hyogo Framework for Action in the World Conference on Disaster Reduction (WCDR) held in Kobe, Japan, on January 2005 stated that the development of indicator systems for disaster risk and vulnerability is one of the key activities enabling decision makers to assess the possible impacts of disasters. Therefore to be able to apply to this study the assessment of the flood risk in a flood prone area is the first step in motivating the government and the people to enhance their capabilities in flood risk management. The government of the Philippines can hardly allocate ample resources (money, manpower, machinery, etc.) for the appropriate implementation of flood disaster mitigation measures due to its limited information of areas perennially affected by floods, numerous organizational structures, differing laws as well as budgetary constraints. These reasons make the flood management system in the Philippines too complicated to put into simple terms both comparatively and quantitatively. Inasmuch that most flood disaster preparedness such as construction of critical infrastructures like national roads and bridges, dikes, sabo dams, pumping stations, diversion channels, etc., are undertaken in the national level, flooding occurs at the local level. More so, there is difficulty in knowing the level of flood risk by the regional, province or even barangay level which is not usually expressed explicitly. Most related studies in the past deal mostly by river basin. ** Assistant Section Chief (Engineer III), Flood Control and Sabo Engineering Center, Department of Public Works and Highways (FCSEC-DPWH), Philippines Director, International Centre for Water Hazard and Risk Management (ICHARM), Japan

2 ANALYSIS OF THE PAST FLOOD DAMAGE As result of the comparative analysis of three kinds of water related disaster databases related to Philippines: the Philippines Annual Flood Disaster Damage database as compiled by its National Disaster Coordinating Council (NDCC) and the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) is accepted for data analysis as compared to two global disaster database which are the Dartmouth Flood Observatory and the Centre for Research on the Epidemiology of Disaster (CRED) through its Emergency Events Database (EM-DAT). The Philippines Annual Flood Disaster Damage is a comprehensive database that lists a 40-year flood record (1970~present) that includes the typhoon name, the affected regions and provinces, inclusive dates and the number of casualties death tolls, missing and those injured. Also included are the amount, in PhP Peso the damage to properties, infrastructures and agriculture. Three kinds of data were used for the measurement; number of events, killed people, and average killed people per event. All events with one killed people or over were classified into three classes by the size of death toll. To make the provinces comparable, the variables were converted to indicators, namely Flood Damage Indicator (FDIa), using the following formula: FDIa = [(LN(x) LN(Min(x)]/[(LN(Max(x)) - LN(Min(x))] Eq. (1) Where FDIa : Flood Damage Indicator (actual) x : variables (number of events (noted by N), number of killed people (noted by K), and killed people per event (noted by KperN)) Max(x): actual maximum value Min(x) : actual minimum value Figure 1 Annual Flood Casualties Caused by Water Related Disasters

3 ESTABLISHMENT OF PHILIPPINE FLOOD RISK INDEX (P-FRIc) There are a lot of conceptual frameworks, which help us to assess the complicated structure of flood risk. One of the most common and simple conceptual framework is the Pressure and Release Model (PAR Model), which is based on the equation Risk = Hazard x Vulnerability. Basically this concept defines disaster as the intersection of two opposing forces which are hazard and vulnerability (Wisner, B. et al, 2004). This concept is used in this thesis to calculate the Philippine Flood Risk Index (P-FRIc). P-FRIc considers the five aspect indexes: Hazard, Exposure, Basic Vulnerability, Capacity Soft countermeasures and Capacity Hard countermeasures. Each index is then composed of three kinds of datasets which are the most representative indicators (see Fig. 2). Fig. 2 Structure of Philippine Flood Risk Index To calculate for P-FRIc the following datasets for indicators and indices are expressed as follows: P FRI c H E V C C S Where: H : Hazard Index E : Exposure Index V : Basic Vulnerability Index C s : Capacity Soft Index C H : Capacity Soft Index H Eq. (2) Indicator : [(LN(x) LN(Min(x)]/[(LN(Max(x)) - LN(Min(x))] Eq. (3) Index : Indicator 1 + Indicator 2 + Indicator 3 Eq. (4) INDICATORS: Indicators are selected by quantitative method: discussion with ICHARM experts, consideration of availability of natural and social statistical data and review of past related studies. Selected indicators and datasets are shown in Table 1. Data are collected from various kinds of sources such as the National Disaster Coordinating Council (NDCC), National Statistical Coordination Board (NSCB) and Center for International Earth Science Information Network (CIESIN). Table 1 Datasets for Indicators

4 INDEXES 1. Hazard Index If a province is frequently affected by typhoons, then the more it will be hazardous and flood prone. Albay is assessed as the most hazardous province in the Philippines. While typhoon prone provinces like Benguet, Pampanga, Zambales, Negros Occidental, Bataan and Iloilo are also ranked hazardous because of these provinces have high annual average monthly rainfall values. Fig. 3 shows the distribution map of the Hazard Index of provinces which are colored in red. Fig. 3 Hazard Index Hazard Index 2. Exposure Index Exposure plays a very important role in risk evaluation. If large floods take place in a highly populated area of lowland areas, there would be large number of victims. That is why the population density in low elevation coastal zones (LECZ, below 10m elevation) is used as for this thesis. CEBU Cavite, is calculated as high exposure province in the Philippines. Metro Manila, Bulacan, Rizal and Mindoro Oriental also ranked at high position because of these provinces have rapid population growth rate. Fig. 4 shows the distribution map of the Exposure Index. The areas in red indicate high value areas while those areas in green indicate low value of Exposure Index. ( Hazard as Percent of Total Basic Vulnerability Index Basic Vulnerability is defined as the basic condition of each province. Batanes, Benguet, Laguna, Rizal and Cavite rank high in the list because of good governance indicator and high human development index. Fig. 5 shows the distribution map of the Vulnerability index. Dark green areas indicate high vulnerable areas. Fig. 5 Vulnerability Index Fig. 4 Exposure Index LANAO ZAMBOANGA BUKIDNON Distribution Map of Hazard Index Fig. 6 Capacity S Index Capacity H Index Vulnerability Index Exposure Index AB R A CA GA YA N ISA BE LA O GA IFU LEY O IL ILO CEBU BU TE CE BO CEBU L HO LANAO ZAMBOANGA ON ( Exposure as Percent of Total Distribution Map of Vulneratbility Index ( Vulnerabil as Percent of Total CapacityH as Percent of Total ( Distribution Map of Capacity H Index BUKIDNON LANAO ZAMBOANGA LANAO ZAMBOANGA BUKIDNON BUK IDN Distribution Map of Exposure Index

5 4. Capacity Hard Countermeasure Index Provinces like Pangasinan, Pampanga, Tarlac and Rizal rank high in the Capacity H Index because of the provinces good investment for flood mitigation projects as outlined in the Medium Term Philippine Development Plan ( ). Isabela topped the ranking considering that it has good sizeable forest cover area. Forest cover is a good indicator for hard countermeasures against floods. Devastated forests cause rapid rainfall runoff, which is contributory to floods. Ifugao, Ilocos Norte, Apayao and Quirino on the other hand complete the ranking (Fig. 6) because of its good basic delivery of services through implementation of basic infrastructures. 5. Capacity Soft Countermeasure Index Education of the people in itself is enabling capacity and power that will pump prime the community into action for preparedness against disaster. If the people are not well educated - access to information will amount to nothing - the people cannot interpret and will not know what action to do with it. Provinces like Camarines Sur, Batangas, Laguna, Leyte and Ilocos Norte topped the ranking because of its high literacy and access to information indicators. The above results are deemed acceptable, convincing and agreeable with the result of similar past studies. Philippine Flood Risk Index (P-FRIc) Fig. 8 shows the distribution map of P-FRIc. Provinces in dark brown indicate high flood risk while in that in yellow show low flood risk. Metro Manila is assessed as the most flood risk province in the country having high vulnerability and exposure. This was confirmed on September 2009 when 80% of the capital was submerged by flashflood caused by Typhoon Ketsana that killed more than 300 people. Albay province comes in the 2 nd place mainly due to its high Hazard and Exposure. Albay has been hit with destructive debris flow in November 2006 when Typhoon Durian caused widespread debris flow over Mt. Mayon that killed 1,150 people and buried infrastructures and houses. Several provinces occupying the top 10 are: Zambales at 3 rd, Pampanga at 4 th, Bulacan at 7 th and Bataan at 10 th. These provinces are high risk due to high Vulnerability and low Coping Capacity soft measures. Pangasinan although has high Hazard and Vulnerability, comes in at 15 th place only because of high Coping Capacity hard measures. The completion of the Agno Flood Control Project has addressed the perennial flooding in the province of Pangasinan. Capacity S Index CapacityH as Percent of Total NCR BATAAN Distribution Map of Capacity S Index Fig. 7 Capacity H Index CEBU ZAMBOANGA LANAO BUKIDNON ( Fig. 8 Philippine Flood Risk Index P-FRIc Philippine Flood Risk Index Distribution Map of P-FRIc PFRIc as Percent of Total NCR BATAAN CEBU ZAMBOANGA LANAO BUKIDNON ( Fig. 9 Radar Chart Showing Structure of P-FRIc

6 COMPARATIVE ANALYSIS BETWEEN P-FRIc AND PAST FLOOD DAMAGE The calculated Philippine Flood Risk Index (P-FRIc) is compared with the past actual flood damage data in order to verify agreement. Fig. 10 shows the scatter graph plotted by P-FRIc and composed actual Flood Damage Indicator of total killed people (FDIa_Com_K), which is calculated by using three classified flood damage indicators, based on levels of magnitude. The composed flood damage indicator FDIa_Com is calculated by simple addition of the three classified indicators: FDIa_L (low) : 10 deaths and below (K < 10) FDIa_Com_K = FDIa_L + FDIa_M + FDIa_H FDIa_M (middle): more than 10 and below 100 deaths (10 < K < 100) FDIa_H (high) : more than 100 deaths (K > 100) P-FRIc exhibit correspondence with the past flood damages, but apparently this is not so significant (R 2 =0.1974). One of the reasons for this disagreement is that P-FRIc expresses the present condition of flood risk whereas FDIa_Com_K indicates the consequence of past 40 year record annual flood damage data. Fig. 10 Scatter Graph of the Provinces Plotted by P-FRIc vs FDI_Com_K Fig. 10 is the scatter graph in which all the provinces are classified into 6 groups and plotted by P-FRIc and FDIa_Com_K. The provinces in group 4 and 3 have relatively good agreement between FDIa_Com_K. While the provinces in Groups 2 and 6 have lower value P-FRIc than FDIa_Com_K. It means that there is possibility that the values of FRIc were underestimated in this aspect. The provinces in Groups 1 and 5 have large values of P-FRIc and FDIa_Com_K. It means P-FRIc may have overestimated the flood risk. This means that these provinces have the high possibility to experience from high casualty flood events in the future. CONCLUSION This master thesis established the Flood Risk Index for 82 provinces in Philippines, by considering the following: Fifteen (15) indicators considering natural and social conditions of the Philippines were identified as components of Flood Risk Index. These datasets were selected based on availability and quality (length of record) of data and their relevance to characterize and represent hazard, vulnerability, exposure and coping capacity factors. The nationwide Philippine Flood Risk Index (P-FRIc) was assessed quantitatively based on the conceptual framework of Pressure and Release (PAR) Model using five indexes. P-FRIc assessed flood risk at present time even without utilizing past flood damage data. It presented the structure of flood risk in comparative and quantitative approaches. P-FRIc assessed flood risk at present time even without utilizing past flood damage data. It is observed that the comparison of radar charts indicates the different composition of strengths and weaknesses in risk components by province. The methodology to assess the flood risk is a new attempt and is very informative for disaster managers. It is hoped that the outcomes of this thesis will advance the knowledge of disaster managers and planners in the Philippines in flood risk assessment and motivate the people, who are at risk to the negative impact of floods to enhance their flood risk administration. ACKNOWLEDGEMENT I would like to express my sincere gratitude to my Advisor, Prof. Kuniyoshi Takeuchi, for his scholarly advice, guidance and encouragement and the staff of ICHARM led by Chief Researcher Mr. Katsuhito Miyake, Dr. Tadashi Nakasu and Dr. Ali Chavosian for their sincere cooperation and valuable review of this thesis. REFERENCES Ben Wisner,et.al. At Risk; 2nd Edition, Natural Hazards, People's Vulnerability and Disasters. Routledge, National Statistical Coordination Board Yasuo Kannami, Master Thesis for GRIPPS / ICHARM. Establishment of Country Based Flood Risk Index: ICHARM, Yuko Arai, Master Thesis for University of Tokyo. Establishment of a Global Flood Risk Index: University Of Tokyo, 2008.