Cross-analysis Seismic microzoning in the metropolitan area of Port-au-Prince: Complexity of the subsoil R. Gilles* 1 , D. Bertil 2 , C. Prépetit 1,3 , M. Belvaux 2 , A. Roullé 2 , J. Jean-Philippe 1 , G. Noury 2 1 Laboratoire National du Bâtiment et des Travaux Publics, rue Toussaint Louverture # 27, Delmas 33, Port-au-Prince, Haïti 2 Bureau de Recherches Géologiques et Minières, 3 av. Claude Guillemin, Orléans Cedex 2, France 3 Bureau des Mines et de l’Energie, Rue Jacques 1er # 11, Delmas 31, Port-au-Prince, Haïti * Corresponding author : rogine09@yahoo.fr Site classification and zoning The zoning is done in five principal steps : 1- cross-analysis of geological, geotechnical and geophysical information; 2- identification of homogeneous areas; 3- definition of one or more columns of representative soils associated with each zone; 4- calculation of seismic responses for each zone; and 5- possible consolidation of areas to obtain the final seismic zoning. Geological, geotechnical and geophysical database Discussions and perspectives Site effects are present over a large part of the metropolitan area of Port-au-Prince. Classes 2 and 6 are stiff soil profiles. From analysis of the amplification coefficients Fa and Fv and of the VS30 associated to the various classes, we observe that the soil classes 3-6 have relatively strong site effects at short periods. The standard NEHRP classes for equivalent VS30 do not show this type of amplification for these high levels of seismic hazard. Thus VS30 is not a good indicator for the assessment of these high short-period amplifications for these classes. Seismic microzoning provides a detailed assessment of soil behavior during earthquakes within a high seismic risk area. It must be taken into account in supporting Haitian earthquake regulations and for the future urban developments in the city of Port-au-Prince. Introduction The magnitude 7.1 earthquake that struck Haiti on January 12, 2010 caused great damage in the epicentral zone. Better prior knowledge of the Haitian subsoil may have helped reduce the damage. To overcome this lack of knowledge, the LNBTP, the BME and BRGM agreed to implement a project of seismic microzoning of the metropolitan area of Port–au-Prince. This project is financed by the Fund for the Reconstruction of the Country of the Haitian MTPTC ministry. Seismic microzoning is an important tool for seismic risk assessment. It is based on a collection of geological, geotechnical, geophysical data from numerous sites. Intended for policy makers, planners, structural engineers, architects but also the public, seismic microzoning is for operational use for the reconstruction of devastated areas of Haiti and, preventively, as part of land-use planning, taking into account natural hazards. The objective of a microzoning is to map local effects of earthquakes: lithological and topographical site effects, surface rupture of active faults, soil liquefaction and landslides. Here we focus specifically on the zoning of lithological site effects to provide seismic design requirements for buildings in Port-au-Prince. References ASCE 7-05 (2005) Minimum Design Loads for Buildings and Other Structures, American Society of Civil Engineers, 2005, Frankel, A., Harmsen, S., Mueller, C., Calais, E. and Haase, J. (2010) Documentation for Initial Seismic Hazard Maps for Haiti. United States Geological Survey Open-File Report 2010-1067. IBC (2009) International Building Code. International Code Council, ICC, 4051 West Flossmoor Road, Country Club Hills, Il. 60478, USA, 1ST printing, 650p. Modaressi H., Foerster E., Mellal A. (1997) Computer aided seismic analysis of soils. Proc. of the 6 th Int. Symp. on numerical models in geomechanics, NUMOG VI, Montréal, Québec, Canada, July 2-4. The microzoning required: • a re-assessment of geological mapping at the scale of the urban agglomeration; • the collection and geo- referencing of many geotechnical drilling profiles (212 sites), geophysical MASW (90 sites) and H/V measurements (17 sites); • during the project, many investigations have been added for a better coverage of the area and to refine information on geological lithology: 25 new geotechnical sites, MASW measurements at 90 sites and H/V measurements at 120 sites. S51A-2312 Figure 6. Low-period site amplification coefficients Fa and mid-period coefficients Fv associated to the site classes. Green: low amplification. Red: high amplification. Orange: medium-range Figure 7. VS30 associated with site classes 0 to 6 and correspondance with NEHRP site classification. Classes 2 and 6 have VS30 equivalent to D class, class 3, 4, 5 to C class, Class 1 is intermediate between C and D. Figure 8. Correspondance between mean VS30 associated to site classes and Fa, Fv values. There is no correlations between Fa values and VS30. Fv decreases from 2.02 (class 2) for VS30=235 m/s to 1.00 (class 3) for VS30= 655 m/s. Fa, Fv values are higher than those for NEHRP site classes for the same seismic hazard level. Figure 3. Main data used for site effect zoning: geological map re- assessed for the project by a BME/BRGM team, georeferenced geotechnical and geophysical measurements sites. Yellow: existing data and red: data acquired for the project by LNBTP with BRGM supervision. Figure 5. Zoning of site classification and associated design response spectra. Class 0 represents bedrock at the outcrop surface: two design spectra are defined (black solid line for Port-au-Prince, black dotted line for Pétion-Ville) Regional hazard Figure 1. Map of crustal faults (red and green) and subduction zones (blue) used for the hazard calculation. « P » denotes the location of Port- au-Prince. The star is the epicenter of M7.0 12 January 2010 earthquake. (From Frankel et al., 2010) Figure 2. Hazard map for spectral acceleration at 0.2s (%g) with 2% probability of exceedance in 50 years for a firm rock site condition. Red line: microzonation area. Black lines: limits of municipalities. (Hazard from Frankel et al., 2010). Green values: design bedrock PGA (g) fixed for each municipality. Seismic hazard for Haïti has been reviewed by Frankel et al. (2010) following the 2010 earthquake. Our study takes these results as a baseline for regional hazard. Hazard level with 2% probability of exceedance in 50 years is very high near Port-au-Prince and is not constant over the study area. Accelerations are higher in the south and decreases toward the north . The design PGA (following ASCE 7-05) is 30 % lower for Cité Soleil (0.39 g) than Pétion-Ville (0.48g) . Cité Soleil 0.39g Pétion-Ville 0.48 g Port-au-Prince 0.42 g Delmas 0.41 g Figure 4. Examples of different data (geological units, H/V ratios, SPT diagrams, Vs profiles) available for two sites. Red = the cathedral in the city center; light blue = near the city hall in Pétionville). 36 homogeneous areas are identified. Seismic responses are first estimated using the CyberQuake software (Modaressi et al., 1997). The input bedrock motions are 4 accelerograms (2 real and 2 modified real) with response spectra close to the design spectrum for rock sites. Responses are evaluated through a nonlinear analysis of soil columns because of the high intensity ground motion. Results are compared and regrouped in 6 zones of distinct site effects. Design spectra are constructed with the same shapes as those recommended by IBC (2009). Individual amplification coefficients Fa (low-period) and Fv (mid-period) are defined for each class of soil. The cross analysis of geological, geophysical and geotechnical information showed: • throughout the metropolitan area, a high variability in soil mechanical properties, that will condition their response in case of an earthquake; • for few geologic units, significant inconsistencies between geotechnical and geophysical data (eg velocity inversions); • that the presence of specific geological and geometrical configurations can make it difficult even erroneous the interpretation of H/V curves. For example : complex lateral facies variations or succession of formations without strong impedance contrast. View publication stats View publication stats