Urban seismic risk assessment of santo domingo: A probabilistic and holistic approach

Transcription

1 See discussions, stats, and author profiles for this publication at: Urban seismic risk assessment of santo domingo: A probabilistic and holistic approach Conference Paper June 2014 DOI: /D33B5W836 CITATION 1 READS authors, including: Omar Dario Cardona National University of Colombia 191 PUBLICATIONS 2,857 CITATIONS SEE PROFILE Mario A. Salgado-Gálvez ERN 53 PUBLICATIONS 126 CITATIONS SEE PROFILE Martha Liliana Carreño Tibaduiza CIMNE International Center for Numerical Methods in Engineering 75 PUBLICATIONS 954 CITATIONS SEE PROFILE Gabriel Andres Bernal INGENIAR 40 PUBLICATIONS 108 CITATIONS SEE PROFILE Some of the authors of this publication are also working on these related projects: World at Risk: Revealing the latent disasters View project Drought hazard and risk assessment: New probabilistic and holistic methodology / Evaluación de amenza y riesgo por sequía: Nueva metodología probabilista y holística View project All content following this page was uploaded by Mario A. Salgado-Gálvez on 26 July The user has requested enhancement of the downloaded file.

2 10NCEE Tenth U.S. National Conference on Earthquake Engineering Frontiers of Earthquake Engineering July 21-25, 2014 Anchorage, Alaska URBAN SEISMIC RISK ASSESSMENT OF SANTO DOMINGO: A PROBABILISTIC AND HOLISTIC APPROACH O.D. Cardona 1, M.A. Salgado 2, M.L. Carreño 3, G.A. Bernal 4, C.P. Villegas 5 and A.H. Barbat 6 ABSTRACT A fully probabilistic risk assessment of Santo Domingo National District, Dominican Republic, was conducted using a building by building resolution level. A national seismic hazard assessment was developed in order to generate a set of stochastic scenarios for the probabilistic seismic risk analysis using the CAPRA platform. Different building classes and vulnerability functions were used to assess earthquake damage and losses using probabilistic metrics. Risk premiums of economic losses and casualties were obtained and aggregated by neighborhoods, based on the city s official cadastral information. Seismic risk was also evaluated from a holistic perspective; i.e. taking into account the expected physical losses as well as the social fragility and lack of resilience conditions, which would increase the second order effects when a strong earthquake strikes the city. Indicators were used to capture the aggravating conditions of the direct physical impact, as well as second order and intangible impact of future earthquakes at urban level. An Urban Seismic Risk Index has been obtained for each neighborhood of the city in order to communicate risk to the stakeholders and decision-makers and to identify the areas that will be more problematic in case of earthquake and the main social causes of seismic vulnerability. 1 Associate Professor, Universidad Nacional de Colombia, Manizales, Colombia 2 Ph.D.Student. CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. 3 CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. 4 Ph.D.Student. CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. 5 CIMNE and Associates, Bogotá, Colombia. 6 CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. Cardona O.D., Salgado M.A., Carreño M.L., Bernal G.A., Villegas C.P., Barbat A.H. Urban seismic risk assessment of Santo Domingo: A probabilistic and holistic approach. Proceedings of the 10 th National Conference in Earthquake Engineering, Earthquake Engineering Research Institute, Anchorage, AK, 2014.

3 URBAN SEISMIC RISK ASSESSMENT OF SANTO DOMINGO: A PROBABILISTIC AND HOLISTIC APPROACH O.D. Cardona 1 M.A. Salgado 2, M.L. Carreño 3, G.A. Bernal 4, C.P. Villegas 5 and A.H. Barbat 6 ABSTRACT A fully probabilistic risk assessment of Santo Domingo-National District, Dominican Republic, was conducted using a building by building resolution level. A national seismic hazard assessment was developed in order to generate a set of stochastic scenarios for the probabilistic risk analysis using the CAPRA platform. Different building classes and vulnerability functions were used to assess earthquake damage and losses using probabilistic metrics. Risk premiums of economic losses and casualties were obtained and aggregated by neighborhoods, based on the city s official cadastral information. Seismic risk was also evaluated from a holistic perspective; i.e. taking into account the expected physical losses as well as the social fragility and lack of resilience conditions, which would increase the second order effects when a strong earthquake strikes the city. Indicators were used to capture the aggravating conditions of the direct physical impact, as well as second order and intangible impact of future earthquakes at urban level. An Urban Seismic Risk Index has been obtained for each neighborhood of the city in order to communicate risk to the stakeholders and decision-makers and to identify the areas that will be more problematic in case of earthquake and the main social causes of seismic vulnerability.. Introduction In many cases the concept of risk has been defined in a fragmentary way, according to each scientific discipline involved in its calculation and quantification process [1]. Based on the formulation by UNDRO of disaster risk [2], different risk analysis methodologies have been developed from different points of view. From a holistic perspective, risk requires a multidisciplinary evaluation that accounts for the conditions related to social fragility and lack of resilience conditions, which may favor the indirect impacts besides the direct physical damage, the quantity and type of economic losses and casualties when an event strikes an urban centre [3]. A multidisciplinary risk estimation to help decision-makers take into account not only geophysical and structural aspects, but also economic, social and institutional variables among others is considered herein as a holistic approach. In this analysis it is also important to note that urban scenarios of potential damage, that is, scenarios considering the physical aspects of risk, are essential since they are the result of the convolution of hazard conditions and physical vulnerability of the exposed assets. 1 Associate Professor, Universidad Nacional de Colombia, Manizales, Colombia 2 Ph.D.Student. CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. 3 CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. 4 Ph.D.Student. CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. 5 CIMNE and Associates, Bogotá, Colombia. 6 CIMNE, Universitat Politècnica de Catalunya, Barcelona, Spain. Cardona O.D., Salgado M.A., Carreño M.L., Bernal G.A., Villegas C.P., Barbat A.H. Urban seismic risk assessment of Santo Domingo: A probabilistic and holistic approach. Proceedings of the 10 th National Conference in Earthquake Engineering, Earthquake Engineering Research Institute, Anchorage, AK, 2014.

4 Cardona [4] developed a conceptual framework and a model for risk analysis of a city from a holistic perspective, quantifying seismic risk in terms of indexes. Hard and soft risk variables are considered taking into account exposure, socio-economic characteristics of different areas (e.g. neighborhoods, districts, municipalities, etc.) of the city and their disaster coping capacity, also understood as their resilience degree. Carreño [5] developed an alternative method for Urban Risk Evaluation from Cardona s model in which urban risk is analyzed using composite indexes. It conserves the approach based on indicators but improves the procedure for normalization and calculates the final risk indexes in an absolute way, instead of relative values. Other improvements of the proposed model refer to the selection of indicators and aspects involved during the evaluation and how they are used [3]. Additionally, Marulanda et al. [6] evaluated the robustness of the proposed methodology by Carreño [5] and Carreño et al. [3]. Fig. 1 presents the conceptual framework of the holistic approach to risk analysis. From this comprehensive perspective, it can be seen that risk is a function of the physical vulnerability and a set of vulnerability factors ε i, which configure the vulnerability conditions of the context. Physical vulnerability is obtained from the susceptibility of exposed assets to hazards, considering the potential intensities, I, of the hazardous events in a period of time, t. The vulnerability of the context depends on the social fragility and issues related to lack of resilience of the disaster prone socio-technical system. Using the meta-concepts of the theory of control to reduce risk, it is necessary to intervene using corrective and prospective actions in the definition of the vulnerability factors ε i. Hence, disaster risk management requires a control system (institutional structure or organization) and an actuation system (public policies and actions) to implement the changes needed to the exposed elements in order to reduce risk. Figure 1. Conceptual framework for a holistic approach to disaster risk assessment and management. Adapted from: [5], [7], [3] and [8]

5 Holistic risk evaluation based on indicators In a holistic risk analysis using indicators, risk results are calculated by aggravating the physical risk by a coefficient depending on the contextual conditions such as the socio-economic fragility and the lack of resilience. Input data considering these conditions at urban level are necessary to apply the proposed methodology. This approach contributes to the effectiveness of disaster risk management, inviting the decision-makers to take action through the identification of weaknesses of the analyzed urban centre. The socio-economic fragility and the lack of resilience are described by a set of indicators related to indirect or intangible effects that aggravate the physical risk, understood as the direct effects. The total risk then depends on the physical risk and the indirect effects expressed as a factor of the direct effects using the following expression, known as the Moncho s Equation in the field of disaster risk indicators: R T R 1 F F (1) where R T is the total risk index, R F is the physical risk index and F is the aggravating coefficient. This coefficient depends on the weighted sum of a set of aggravating factors related to the socio-economic fragility, F FSi and the lack of resilience of the exposed assets, F FRj m F w F w F FSi FSi FRj FRj i 1 j 1 n (2) where w FSi and w FRj are the weights of each i and j factors and m and n are the total number of descriptors for social fragility and lack of resilience respectively. The aggravating factors F FSi and F FRj are calculated using transformation functions. The descriptors used in the analysis have different nature and units, and thus the transformation functions standardize the gross values of the descriptors into commensurable factors. Fig. 2 shows a model for the transformation functions used by the methodology in order to calculate the risk and aggravating factors. These functions are membership functions for high risk and aggravating levels, where the x-axis presents values of the descriptors while the value of the factor (physical risk or aggravation) is shown in the y-axis with values ranging between 0 and 1; 0 represents the non-membership and 1 the total membership. The limit values Xmin and Xmax are defined taking into account expert opinions and information about previous disasters. In the case of the descriptors of lack of resilience, the function has an inverse shape; the higher value of this indicator gives lower aggravating value. Fig. 3 presents examples of the transformation functions used. The weights w FSi and w FRj represent the relative importance of each factor and are calculated by means of an Analytical Hierarchy Process (AHP) [9], [3], [5], [8].

6 Aggravating or risk factor Xmin Xmax Indicator value Figure 2. Transformation function example Risk factor Aggravating factor 0 0 b) Damaged area P[0 20] (% Damaged area / Total area) a) Hospital beds (# hospital beds each 1000 inhabitants) P[0 30] Figure 3. Examples of transformation functions: a) damaged area; b) hospital beds The physical risk, R F, is evaluated in the same way using the following expression: p F RFi RFi (3) i 1 R w F Fig. 4 shows the calculation process of the total risk index for the analysis units such as municipalities, counties or neighborhoods, starting from the descriptors of physical risk, X RFi, and the descriptors of the aggravating coefficient F, that are X FSi and X FRi, using the weights w RFi w FSi and w FRj of each descriptor.

7 X RF1 F RF1 w RF1 Risk indicators Risk factors (descriptors) X RFi F RFi w RF4 R F Physical risk X RFp F RFp w RF5 R T Total risk X FS1 F FS1 w FS1 X FSi F FSi w FSi Aggravating X FSm F Aggravating indicators FSm w FSm F Aggravating coef. X factors FR1 (descriptors) F FR1 w FR1 X FRj F FRj w FRj X FRn F FRn w FRn Figure 4. Descriptors of the physical risk, social fragility, lack of resilience and their weights Holistic seismic risk evaluation of Santo Domingo-National District in Dominican Republic The National District is the territorial unit where Dominican Republic s capital, Santo Domingo is located. Since 2001 this district was declared administratively independent for the adjacent provinces. It is comprised by 70 neighborhoods that are in turn grouped into three censual units. It has an extension of 91 km 2, a total of 980,000 inhabitants according to the 2009 census and a total of 96,058 dwellings. Seismic risk from a physical point of view of Santo Domingo Seismicity in the National District area is moderate if compared to other cities in the country such as Santiago de Los Caballeros. It is important to note that the city is located on the same island as Port au Prince (Haiti), and IX MMI intensity events causing important damages in the city were recorded in 1615 and The seismic hazard was analyzed using the program CRISIS2007 [10] considering active faults at country level, where a total of 6,925 scenarios were generated. A cadastral database with more than 96,000 dwellings was available, for which a process of identification, characterization and appraisal of each unit was followed in order to define their structural characteristics, their value and the human occupation in a given scenario. Fig. 5 presents the replacement value in US Dollars for each dwelling.

8 Figure 5. Replacement value by dwelling in the National District A set of vulnerability functions was calculated using ERN-Vulnerability software [11]. These functions were defined for each building typology, considering material, structural system and number of stories among other characteristics; for the National District dwellings are mainly composed by masonry units (unreinforced, reinforced and confined), R/C frames and wooden buildings. Fig. 6 presents some of the vulnerability functions used in the analysis 100% 90% 80% Pérdida Loss 70% 60% 50% 40% 30% 20% 10% C1M_M RM1L_M URML_M C4H_M C1L_M C1H_L W1_M W2_M 0% Intensity Intensidad (cm/s2) (gal) Figure 6. Some seismic vulnerability functions used in the National District Physical risk is evaluated by the convolution of the hazard with the vulnerability of the exposed assets and the results are the potential losses [12]. Risk can be expressed in terms of damage on physical effects, absolute or relative economic loss and/or effects on the population. The loss exceedance curve (LEC) can be obtained for the National District, and from it the average annual loss (AAL) and probable maximum loss (PML) can be obtained as well. Table 1 presents a summary of the results while Fig. 7 presents the risk map in terms of the AAL for each unit.

9 Table 1. Physical risk results summary for the National District Results Exposed value USD$ x ,205.2 USD$ x Average annual loss Return period PML Loss Years USD$ x10 6 % 100 $ $ $1, $1, Figure 7. Average annual loss for the National District by dwelling Seismic risk from a holistic point of view of Santo Domingo Fig 8 shows the descriptors used to describe the physical risk, social fragility and the lack of resilience for the National District. The holistic risk analysis was conducted following the methodology explained above. The descriptors were chosen as the most significant for each category and according to the available information for the city. A sensitivity analysis of the different indexes based on parameters, input data, weights and transformation functions was performed using a Monte Carlo simulation methodology such as the proposed by [13].

10 X RF1 Expected annual economic loss by sector w RF1 X RF2 Injured w RF2 X RF3 Casualties w RF3 R F Physical risk X RF4 Expected Jobless w RF4 X RF5 Expected Homeless w RF5 R T Total risk X FS1 Literacy rate w FS1 X FS2 Overcrowded housing rate w FS2 X FS3 Marginality w FS3 X FS4 Population density w FS4 X FS5 Delinquency rate (violent deaths) w FS5 F Aggravation X FR1 Houses with radio w FR1 X FR2 Houses with television w FR2 X FR3 Public space (parks) w FR3 X FR4 Distance to the closest hospital w FR4 Figure 8. Descriptors used for the National District Fig. 9 shows the obtained results of the R F in the National District; these results present the highest risk values for the Ciudad Colonial (historical centre) and Centro de los Héroes neighborhoods. Figure 9. Physical risk index by neighborhood in the National District Fig. 10 presents the raking of the aggravating coefficients for some neighborhoods; Zurza has the worst aggravating coefficient according to its characteristics of social fragility and lack of resilience. Domingo Sabio and San Diego have a high aggravating coefficient as well.

11 Figure 10. Aggravating coefficients for some neighborhoods in the National District Fig 11 presents the results in terms of the R T for the neighborhoods of the National District. Figure 11. Total risk by neighborhood in the National District Conclusions For disaster risk management purposes, the risk assessment should improve the decision-making process in order to contribute to the effectiveness of risk management, identifying the weaknesses of the exposed elements and their evolution over time. This case study involves several elements that try to capture different aspects of the city at physical, social and institutional level. The proposed methodology for seismic risk from a holistic approach has been applied to the National District in the Dominican Republic, where the Ciudad Colonial and Centro de los Héroes neighborhoods were identified as constituting the more problematic areas due to not only the potential damage in terms of the physical vulnerability as well as their social fragility and lack of resilience.

12 Acknowledgments The authors are grateful for the support of the Ministry of Education and Science of Spain Enfoque integral y probabilista para la evaluación del riesgo sísmico en España CoPASRE (CGL ). Also to the Spain s Ministry of Economy and Competitiveness in the framework of the researcher s formation program (FPI). References 1. Cardona O.D., Hurtado J.E. (2000). Modelación numérica para la estimación holística del riesgo sísmico urbano considerando variables técnicas, sociales y económicas. Métodos Numéricos en Ciencias Sociales. CIMNE- UPC, Barcelona, Spain. 2. UNDRO (1980). Natural Disasters and Vulnerability Analysis, Report of Experts Group Meeting, Geneva. 3. Carreño, M.L., Cardona, O.D., Barbat, A.H (2007). Urban Seismic Risk Evaluation: A Holistic Approach, Natural Hazards, 40:1, Cardona, O.D. (2001). Estimación Holística del Riesgo Sísmico utilizando Sistemas Dinámicos Complejos. Tesis Doctoral, Universidad Politécnica de Cataluña, Barcelona, España. 5. Carreño, M.L. (2006). Técnicas innovadoras para la evaluación del riesgo sísmico y su gestión en centros urbanos: Acciones ex ante y ex post, Tesis doctoral, Universidad Politécnica de Cataluña, Barcelona. 6. Marulanda M.C., Cardona O.D., Barbat A.H. (2009). Robustness of the holistic seismic risk evaluation in urban centers using the USRi, Natural Hazards 49:3, IDEA (2005) System of indicators for disaster risk management: Main technical report. IDB/IDEA Programme of Indicators for Disaster Risk Management, Universidad Nacional de Colombia, Manizales. 8. Carreño, M.L., Cardona, O. D., Barbat, A. H. (2007). Disaster risk management performance index. Natural Hazards, 41:1, Saaty TL, Vargas LG (1991) Prediction, Projection, and Forecasting: Applications of the Analytical Hierarchy Process in Economics, Finance, Politics, Games, and Sports, Kluwer Academic Publishers, Boston, USA. 10. Ordaz M, Aguilar A, Arboleda J. (2007) CRISIS 2007V7.6 Program to calculate seismic hazard. Instituto de Ingeniería, Universidad Nacional Autónoma de México. México D.F. 11. Evaluación de Riesgos Naturales América Latina (2009). ERNVulnerabilidad. Programa para el cálculo de funciones de vulnerabilidad física y humana. Bogotá, Colombia. 12. Ordaz M. (2000). Metodología para la evaluación del riesgo sísmico enfocada a la gerencia de seguros por terremoto. Universidad Nacional Autónoma de México. México D.F. 13. Marulanda M., Cardona O., Barbat A. (2008). Robustness of the holistic seismic risk evaluation in urban centers using the USRi. Natual Hazards. View publication stats