Third LACCEI International Latin American and Caribbean Conference for Engineering and Technology (LACCEI’2005) “Advances in Engineering and Technology: A Global Perspective”, 8-10 June 2005, Cartagena de Indias, Colombia.

Estimating Natural Hazards Damage for the Insurance Industry in Puerto Rico

Ricardo R. López, Ph.D., P.E.

Luis A. Godoy, Ph.D. Felipe J. Acosta, Ph.D., P.E. José O. Guevara, Ph.D., P.E.

José F. Lluch, Ph.D., P.E. José A. Martínez Cruzado, Ph.D., P.E.

Ismael Pagán Trinidad, M.S. Miguel Pando, Ph.D., P.E.

Ricardo Ramos, Ph. D., P.E. Ali Saffar, Ph.D.

Daniel Wendichansky, Ph.D., P.E. Raul Zapata, Ph. D., P.E.

Civil Infrastructure Research Center and Department of Civil Engineering and Surveying, University of Puerto Rico at Mayagüez, Mayagüez, Puerto Rico 00681-9041

Abstract Insurance companies have a pressing necessity to refine their estimates of expected damage in the case of earthquakes, hurricanes or floods. On the other hand, the University needs financial support for their investigations on the resistance of structures subjected to natural hazards, and has the expertise to help the insurance industry assess their expected losses. This paper covers an overview of the research being done at the Department of Civil Engineering at UPRM as part of the project funded by the Insurance Commissioner of Puerto Rico. The projects include estimate of the earthquake hazard, evaluation of expected performance of concrete and steel structures subjected to earthquakes and strong winds, study of the resistance to floods, and study of alternative designs and repair methods to improve the structural resistance. Keywords Civil infrastructure, Insurance industry, Natural Hazards, Risk analysis. 1. Introduction The insurance industry is a complex system in which engineers and other professionals make decisions about cost, risk and evaluation of damage to constructions. However, many people involved in this business do not have any formal education in the technicalities of structures, natural hazards and their consequences. On the other hand, academic researchers usually write for their peers, and people without formal education in engineering or science will have difficulty understanding the results of the research. This is a project in which research is carried out with the end-user in mind, so that most information generated should be transferred and used by people developing practical tasks in insurance activities. This opportunity emerges as a consequence of a project carried out at present at the University of Puerto

Rico at Mayagüez (UPRM) for the Insurance Commissioner of Puerto Rico (ICPR), to develop a structured system which may serve as support to the main stakeholders in the insurance business. The risks considered are due to natural hazards (specifically hurricanes, earthquakes and floods) in typical scenarios affecting property in the island. The complete project contains the development of tools that can be used to assess risk within a framework defined by the ICPR and the insurance firms. This has been organized in several areas, including seismic risk and design; performance of buildings under earthquakes and hurricanes; vulnerability of constructions under wind and seismic excitation; building costs; and damage due to flood. Another area carries out the compilation of information generated by the project in order to translate this to maximum probable loss, so that technology transfer can be made. 2. Ground Motions and Liquefaction Potential Evaluation for Puerto Rico’s Main Cities Puerto Rico (PR), located within the tectonically active zone between the North American and Caribbean plates, is recognized as having a significant seismic hazard, evidenced by the hundreds of moderate (M 4 to 5) events recorded in the last 30 years. Since the early Spanish settlements in the 16th century, at least five strong earthquakes have struck PR - 1670 (M≈6), 1787 (M≈7.3), 1867 (M≈7.3), 1918 (M≈7.3), and 1943 (M≈7.5) - causing loss of life and significant property damage. The 1918 earthquake reportedly induced liquefaction in western PR, killing at least 114 people and causing about $114 million in damage. The consequences of a future earthquake could be even worse, as the majority of PR’s population (4 million) lives in coastal areas susceptible to liquefaction and tsunamis. Despite the high seismic risk, research to adequately assess and mitigate earthquake hazard in PR lags behind other seismically-active regions of the US, particularly quantification of expected ground motions and liquefaction susceptibility for major cities. This project attempts to address this gap. This sub-project involves compiling a geotechnical database that includes available borehole information, SPT and CPT data, groundwater information, and complemented with shear wave velocity profiles from geophysical tests where available. This project will develop and populate a geotechnical database of superficial soils in the main cities of Puerto Rico: San Juan, Ponce, Mayagüez, Arecibo, Humacao, Caguas. Existing geotechnical boring, water well log data, geophysical tests, and geological information for each of these cities have been solicited from public and private sources. Currently no such geotechnical databases exist for the main cities of Puerto Rico. This database is the first step needed in the process to estimate design ground motions and to develop earthquake soil amplification maps. All the information gathered is being stored using ARC-GIS software. Figure 1 shows the geotechnical boreholes gathered for the city of Mayagüez. With the information gathered, a liquefaction susceptibility study was carried out for the major cities. The liquefaction susceptibility maps were prepared using the cyclic stress ratios determined using the site-specific geotechnical data, coupled with the 1998 NCEER recommendations for the simplified liquefaction procedure, originally proposed by Seed and Idriss in 1971. Figure 2 shows the liquefaction susceptibility study for the city of Mayagüez. These results have the potential to assist policy makers and the engineering design community to prepare for future seismic events in PR. 3. Parameters that Affect the Earthquake Resistance This project provides the information to be able to estimate the expected damage in buildings subjected to strong ground motions. The project attempts to determine the influence of typical structural features on the earthquake resistance of existing and new structures.

Figure 1: Boring Logs in the city of Mayagüez.

Figure 2: Liquefaction susceptibility study for the city of Mayagüez. The project will include reinforced concrete and steel structures. A preliminary classification was obtained from reviewing the existing literature, and the buildings were classified as residential, hotels, commercial, hospitals, schools, and industrial. Within each classification, data on existing structures was compiled in order to determine which geometric and structural parameters show more variation. From data obtained from field studies and available plans from the Regulations and Permits Administration of Puerto Rico, reasonable variations in the observed parameters were identified and used to establish the properties of typical structures. After obtaining examples of typical existing buildings, normal range of parameters such as floor height, beam span, beam depth, element dimensions, site effect and others were established for a particular category. Elastic analyses assuming cracked sections were performed to obtain the calculated inter-story drifts for the buildings subjected to design ground motions. Using the drifts, estimates of structural damage are obtained using Algan’s equations. Other parameters such as columns shortened by adjacent masonry walls and flexible stories are handled separately. Currently typical reinforced concrete houses of 1 and 2 stories made of beams and columns or walls have been analyzed. Medium rise buildings are being studied to determine building period and expected drifts. The following tables summarize some of the information obtained for one story residential concrete houses.

0 0 1 2 4Mi

N

Table 1: One story residential concrete houses.

Table 2: One story residential concrete houses. With these the following parameter variations were studied: Soil types C, D and E according to UBC 97, Geometry in plan of rectangular layout (2:1, 1:1), column area as percentage of floor area (0.7%, 1.5%), column orientation (NS, EW, distributed), roof height (8’, 10’, 12’). From the elastic analysis of the models subjected to spectral earthquake loads the roof lateral displacement and drift were obtained. The figure shows the displaced shape of a typical 3 span frame. The most influential parameters on the response were the column area, column orientation, and roof height. As an example, a graph showing the increase in drift with decrease in column area for the different models is shown.

Figure 3: Frames considered.

Altura Promedio

Ancho Promedio Razon H/B

Razon Tramo / Altura

Espesor Promedio

Largo Promedio

Razon Area Total / Area

Piso

Espesor Promedio

Tramo Promedio

# in in --- --- in ft % in ft1 16.50 6.50 2.64 10.62 6.00 6.29 4.59 5.00 10.692 22.29 7.26 3.16 8.81 6.00 7.19 4.31 5.06 10.953 23.50 6.00 3.92 9.30 6.00 5.78 4.37 5.06 10.434 16.56 6.44 2.66 11.08 6.00 5.05 4.67 5.00 11.005 17.00 6.00 2.83 10.24 6.00 5.28 5.13 5.00 10.386 17.28 6.13 2.85 10.59 6.00 7.14 4.28 5.13 12.507 19.39 6.88 2.90 9.71 6.00 6.74 4.45 5.00 10.82

Promedio 18.93 6.46 2.99 10.05 6.00 6.21 4.54 5.03 10.97Maximos 23.50 7.26 3.92 11.08 6.00 7.19 5.13 5.13 12.50Minimos 16.50 6.00 2.64 8.81 6.00 5.05 4.28 5.00 10.38

Modelo

Vigas Bloques Losas

Año Altura Libre

Area Piso

Peso Total / Area Piso

Largo Promedio

Ancho Promedio

Razon H/B

Razon Area Columnas / Area Piso

Largo Total N-S

Largo Total E-W

# --- ft ft^2 N-S E-W lb/ft^2 in in --- % ft ft1 1998 9.00 1710.50 3.00 4.00 123.97 22.58 6.40 3.70 1.30 42.27 13.272 1991 8.50 2584.00 4.00 6.00 126.48 24.49 6.49 4.00 1.27 45.27 34.773 1993 12.00 2183.00 3.00 3.00 149.81 23.11 6.79 3.72 1.45 47.55 29.714 2002 8.00 1134.00 3.00 3.00 116.61 13.90 6.00 2.32 0.97 20.00 11.505 1996 9.00 1545.00 5.00 3.00 120.89 10.62 6.00 1.77 0.74 19.50 16.506 2001 9.50 1919.00 5.00 3.00 128.77 18.26 6.64 2.94 1.34 34.16 30.167 1994 8.75 2147.25 3.00 4.00 133.21 22.85 6.20 3.85 1.29 43.77 24.02

Promedio 1996 9.25 1888.96 3.71 3.71 128.53 19.40 6.36 3.18 1.20 36.07 22.85Maximos 2002 12.00 2584.00 5.00 6.00 149.81 24.49 6.79 4.00 1.45 47.55 34.77Minimos 1991 8.00 1134.00 3.00 3.00 116.61 10.62 6.00 1.77 0.74 19.50 11.50

Columnas

ModeloTramos

General

Figure 4: Decrease in column area.

Figure 5: Damage intensity for N-S direction for one-level frames. Using the a damage intensity scale from 0 to 1, in which 1 means total damage, 0.75 means substantial damage, 0.55 is moderate damage, 0.35 is minor damage, and 0.05 is no damage, the following graph is constructed. The method provides for structural damage (formula 1, nonstructural damage of normal elements of, and nonstructural damage of elements sensitive to lateral displacements. 4. Damage to industrial constructions due to hurricanes Research about vulnerability of constructions in Puerto Rico under hurricanes has been restricted in this project to a few building types, mainly those of interest to the insurance industry for which they have difficulties in the estimation of costs of policies or for which they have to face large claims after a hurricane. The building types considered include low-rise industrial buildings and multistory buildings with six floors or more. Although wooden houses are severely damaged during hurricanes, insurance companies avoid providing insurance for such constructions; concrete houses, on the other hand, are insured but the cost of the claims is not a real concern to the companies.

Disminución en el Area de Columnas(1.5% a 0.7%)

050

100150200250300350400450500

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35Casos

Aum

ento

en

Der

iva

(%)

Dirección Fuerte Dirección Debil

RESULTANTE DE VALORES DE INTENSIDAD DE DAÑO PARA LA DIRECCION N-SPORTICOS 1 NIVEL

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 Modelo

Intensidad de daño

FORMULA 1 FORMULA 2 FORMULA 3

MAJOR

SUBSTANTIAL

MODERATE

MINOR

NONE

Figure 6 (a) shows a schematic representation of a low-rise industrial building. They have some special features with respect to other constructed facilities mainly because sidewalls and roof offer large exposure surfaces to wind.

Figure 6: (a) Elements of industrial buildings considered in this research. (b) Example of failure in industrial building due to a hurricane.

Figure 7: Wind design codes used for the industrial buildings of the database of this project.

During hurricanes, industrial buildings may experience several different forms of failure, including buckling of the roof or sidewall panels, pull-out of connections between the panels and the mains resisting frame, failure of connections between the columns and the foundation. An example of failure due to hurricanes is shown in Figure 6 (b). Models and tools for the structural analysis are already available, together with design codes. However, such information is not directly of use to the insurance industry, since what they require are risk estimates. Two ingredients are needed for that: knowledge about the probability of occurrence of hurricanes with given intensities (which are derived from recurrence studies) and frequency of the different types of industrial buildings (which are derived from inventories). Because no inventories are available in Puerto Rico for industrial buildings (nor for other building types), part of the research was devoted to develop a small inventory. Plans of constructions were obtained with the help of ARPE (Administración de Reglamentos y Permisos de Puerto Rico). Out of 20 projects reviewed, 17 were designed after 1999. Information about the main parameters of interest included the identification of the structure, wind design procedures used (codes used), parameters required to estimate wind loads, and construction procedures used. An example of information about wind provisions used in the designs is shown in Figure 7.

Failure modes due to buckling of the roof or sidewalls were computed during the project using finite element analysis, and examples of buckling modes are shown in Figure 8. Details are given in Lopez and Godoy (2005) for the buckling of several panel configurations. Finally, information about the mean return period of hurricanes for Puerto Rico has been used. For example, contours of maximum wind speed for 100 year return period is shown in Figure 9. The studies support the assumption that just one zone may be considered for the island because changes in expected events are small. The information collected or produced as part of the research is currently used to generate fragility curves for each type of construction considered.

Figure 8: Buckling mode for type B cold-formed folded plate for wind pressures, from López and

Godoy, 2005).

Figure 9: Contours of maximum wind speed for 100 year return period. 5. Earthquake Vulnerability of Typical Residential Houses The resistance to earthquakes of existing structures is of particular importance to the insurance companies in Puerto Rico. A large number of residences (of one and two stories) are constructed with reinforced concrete walls oriented in the longitudinal direction and masonry walls in the transversal direction, as it can be seen in Figure 10. Due to lack of information about how resistant are these structures in the weak direction, it was decided to construct five specimens that represent these typical residences. These specimens will be constructed using typical details, emphasizing the connection between: the footing and reinforced concrete/masonry wall, horizontal and vertical reinforcement of the concrete/masonry wall,

Type B

and the connection between the roof and reinforced concrete/masonry wall. Figure 11 shows the typical construction details of the specimens used in this project. Experimental test will be carried out to the specimens to establish the capacity along the weak axis. Figure 12 illustrate the failure mechanism observed during the tests of the first two specimens, and figure 13 shows the preliminary experimental results obtained during the tests of these two specimens.

Figure 10: Typical construction of residential houses.

Figure 11: Construction of the specimens in the Structural Laboratory of UPRM.

Figure 12: Failure mechanisms of the first two specimens tested: (a) joint failure of the specimen 1 (without concrete block wall), (b) concrete wall failure of the specimen 2 (with concrete block wall).

Lateral load vs. lateral displacement

-35

-25

-15

-5

5

15

25

35

-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6

Lateral displacement, in

Late

ral l

oad

(V),

kip

Lateral displacement vs. lateral displacement

-35

-25

-15

-5

5

15

25

35

-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6

Lateral displacement, in

Late

ral l

oad

(V),

kip

Figure 13: Preliminary experimental results of the first two specimens. (a) specimen without

masonry concrete wall. (b) specimen with concrete block wall. As it can be seen, the lateral capacity of the specimens increases significantly when a concrete block wall is included inside the frame. Preliminary computations show that without the concrete block wall, the overall lateral capacity of the house will be less than the lateral demand produced for this type of structure during an earthquake. 6. Elastic Design of Reinforced Concrete Buildings Many structures, such as hospitals, special government agencies or schools used as shelters, need to be in full service during and after a major nature event such as a hurricane or an earthquake event. During the process of design the structures are designed to perform elastically for a certain maximum hurricane wind velocity, however in the case of earthquakes a reduction of the earthquake force is allowed to account a reduction in force considering that many components will crack reducing its stiffness. However the cracks and lateral deformations may affect the future use of the building which could be in serious risk of collapse. This section presents part of the research to develop regulations in order to enforce the use of elastic forces in the design of certain structures that need to be in full operation during and after an earthquake. As part of this process typical buildings are being modeled to study its behavior under earthquake code elastic forces and recommend its strengthening if needed. Detail cost estimates are being prepared to determine the cost of retrofitting after the building is constructed and compare with the increase in the building cost due to the addition of structural elements before the structure is constructed. During the past years, hospitals, government agencies and schools have suffered severe damage or even a collapse during an earthquake event even though they were designed for the current code regulations. Hospitals and schools designed with similar codes to the Puerto Rico building code established before 1987 have been severely damaged and suffer moderate damage in structural elements and severe damage in elements considered as non-structural; however this moderate damage or non-structural damage may impair the use of the building. This paper presents part of the ongoing research at the University of Puerto Rico, that is investigating the seismic performance of hospitals, schools and government agencies in order to develop regulations to assure a full operations of these buildings during and after and earthquake. The investigation involves: A complete structural model of each building using the information of the construction drawings; analysis of the existing structure using the elastic forces obtained from the code requirements with a reduced ductility factor of R=2; design verification of the structural elements including the foundation to determine its adequacy; analysis of the proposed structure with the addition of structural elements to comply with the new forces; determination of the cost estimate of the new elements including the foundation and determination of percentage of increase relative to the

total cost estimate; determination of the cost estimate of the new elements to be incorporated in the existing structure including demolition, sequence of construction and foundations. The paper presents results from the first stages of the investigation.

Figure 14: (a) School Model. (b) Model of Government Office Building.

Many hospitals and schools were designed following the Puerto Rico Building before the 1987 revisions to incorporate the seismic provisions required for the earthquake demand. Many of these structures were designed for gravity loads, wind loads and moderate earthquakes and do not satisfy modern seismic codes. Attention is now on retrofit to improve their seismic performance and propose regulations in the code for this type of structures to perform elastically during an earthquake event. Earthquake prone regions, such as California, developed new regulations for the design of hospitals because many hospitals have been severely damaged during past earthquakes. These regulations are in the process of their adoption and they include recommendations for retrofitting of existing buildings to comply with more stringent structural requirements. Two models were completed, the school building model and the government office building. The hospital building is the process of modeling. 7. Flood Effects on Structures Flood actions describe acts which a flood could do directly to a building, potentially causing damage, or even structural failure. This flood actions analysis would permit to estimate more accurately the damage from potential flood events, and could allow for many of the uncertainties to properly acknowledge. The main objectives of this research work are to estimate the vulnerability of buildings and building elements due to riverine and coastal floods, and to determine the probable maximum loss due catastrophic flood events. The main methodology focuses on the development of various algorithms to identify and select spatial flood risks based on Food Insurance Maps. Flood levels and velocities are identified and flood depths are generated.

a. FIRM b. Topography c. Flood Depth

Figure 15: Flood Level Generated from FIRM and Topographic Maps.

The structural components of building are identified and characterized at specific flood sites. The structural failure of the building elements are analyzed for the resistance of materials, structural elements, and junctions. The failure mode are analyzed and selected by comparing between the resistance of structural materials and the resistance of junctions. The magnitude of the forces will be compared to the resistance of the structural elements and connections in buildings in order to estimate building vulnerability Five action stresses will be considered which are caused by hydrostatics, hydrodynamics, waves, buoyancy, debris impact, and soil scour.

Figure 16: Flood loads acting on a building.

Figure 17: Flood devastation. 8. Cost Estimates of Construction for Puerto Rico The Maximum Probable Loss is made of two components: the loss model and the cost estimate. Since the structures damaged by a natural disaster are repaired or rebuilt, the cost estimate has to be directed towards the repair or reconstruction cost of structures. A survey was made among contractors, designers and owners to obtain cost data in Puerto Rico. The information obtained was categorized according to the classification shown in Table 3. Central tendency measures, percentiles and coefficient of variation were obtained from the classified information and it was compared with data from the following publications: Building Construction Cost Data, National Building Cost Manual, and Bids Construction Report. Factors were estimated to adjust the national average costs from these publications to the cost indicated by our survey.

A cost index was developed to provide a means to update the results without having to do a complete survey again. To update the index a smaller survey is made of the principal cost components identified in our survey, which includes labor and materials. It is not practical to include all labor and material items involved in construction projects as part of the index. The labor and material categories with the highest cost were selected to be part of the index. The labor component in the index includes the amount paid to carpenters, rodmen, masons, and labor, while the material component includes the cost in large quantities of concrete, reinforcing steel, ceramic floor, and cement. The cost obtained from the survey, along with the adjustment method that could be used in the future provide an estimate of current construction cost. Previous studies (Means 2002, Guzman 1998) have developed a relationship between current construction costs and repair and reconstruction cost of structures. A literature review of these studies was made and factors were estimated to convert current construction cost to repair or reconstruction cost of structures. The factor for repair structures is 1.89 and the factor for replacement of structures is 1.13. That means that the repair cost estimate is the cost estimate based on current cost multiplied by 1.89. Similarly, the repair cost estimate is the cost estimate based on current costs multiplied by 1.13. Detailed information and survey results are found in Botero (2004).

Classification Use Bank Care Center for the elderly Hotel Ofice

Comercial

Detail sale School Classrooms Educational University Individual Parking As part of a project Apartments > 3 stories Apartments (walkup) Residencial One family house Treatment Plant Utility building Industrial Manufacturing Basketball court (school) Other Community Center School Repair Hotel

Table 3: Cost Categories.

9. Integrating Insurance Solutions The structural, geotechnical, and socio-economic components of the Puerto Rico Insurance Project were discussed in previous sections. Integrating the results from those components in an intuitive, multi-level insurance solutions software required consultation with various user groups, a process still continuing at this writing. The presentation in this section is therefore an overview of the first generation software and not the final product. The Structure menus provide selection options within residential, commercial, institutional and industrial use groups. The selection is aided by the use of photographs depicting different building types. Depending on the user expertise, three levels of input are possible for each selection. Figure 17 shows first level menus for two residential types. The wide use of metal roofs in Puerto Rico which is not

limited to wood-zinc houses, as well as the practice of building houses on gravity columns are reflected in these two menus. The calculations for first level input are based on fragility curves developed by the structural groups. The second level option uses simplified response analysis algorithms to improve on the first level predictions. The third level option is for engineers only, and it requires detailed structural input. The last two levels are not available for all building classes. Buildings of high insured values are the likely candidates for a more refined analysis.

Figure 15: Interface Menu.

The Policy option deals with liability issues resulting from the actual versus the insured values of structures. The Analysis menus provide event driven financial analysis options for earthquakes, floods, and hurricanes. The life cycle cost analysis is also available under this option. The cost parameters from the previously reported surveys are used in these calculations. The Report menus generate risk analysis summaries in user specified formats to be used as part of an insurance document. There are a number of computer programs available today estimating potential losses from disasters. Of those, HAZUS which was developed by the Federal Emergency Management Agency is the most famous. By design, these programs focus on regional losses, thus providing a big picture approach. In developing the Insurance Solutions program, the focus was shifted to individual buildings. Although the user will still have the option of analyzing his entire inventory for various event scenarios, the emphasis on small components is really were the strength of this program lies.

Figure 16: Location Menus.

Figure 17: Sample building classification menus.

Acknowledgements This work has been supported by a grant from the Insurance Commissioner of Puerto Rico. The authors thank the valuable contribution of several graduate students who participated in various aspects of this research, including Jorge Botero, Norberto Caraballo, Johanna Cataño, Guillermo Gerbaudo, Yvonne González, Héctor D. López, Lourdes Mieses, Eduardo Torres and Edgardo Vélez.

References Godoy L. A. and López R. R. (2004), “Continuum education about risk in constructions aimed to the insurance industry”, Proc. ASEE 2004 International Colloquium, American Society of Engineering Education, Beijing, China. López H. D. and Godoy L. A. (2005), “Buckling of metal folded plates under wind pressure”, Proc. Third International Conference on Structural Stability and Dynamics, June 19-22, Kissimmee, Florida. Botero J. H., (2004), “Estimados del Costo de Reconstrucción de Edificaciones Asegurables en Puerto Rico”, M. Sc. Tesis, Universidad de Puerto Rico, Recinto Universitario de Mayagüez, Mayagüez, Puerto Rico Means, R.S. (2002), Repair and Remodeling Estimating Methods. R.S. Means Company. Guzmán, A. L. (1998), “Cost – Performance Criteria for Seismic Retrofitting”, Ph.D. Thesis, University of Puerto Rico, Mayagüez, Puerto Rico. FEMA (1986a). “Floodprofing: Floof-prone Residential Structures”, Federal Emergency Management Agency; No. 114, September 1986. FEMA (1986b). “Floodprofing: Non-residential Structures”, Federal Emergency Management Agency; No. 102, May 1986. Haehnel, R.B, and Daly, S.F. (2004). “Maximum Impact Force of Woody Debris on Floodplain Structures”, ASCE J. of Hydraulic Engineering, 112-120. Biographic Information (First two authors) Dr. Ricardo R. LÓPEZ is Professor and Associate Director for Graduate Studies of the Civil Engineering and Surveying Department at UPRM. He serves as regional editor of the Journal “Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil”. Dr. Luis A. GODOY is Professor and Associate Director for Research of the Civil Engineering and Surveying Department at UPRM, and is also the Director of the Civil Infrastructure Research Center at UPRM. He serves as editor of the LACCEI Journal and of the Journal “Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil”. Authorization and Disclaimer Authors authorize LACCEI to publish the papers in the conference proceedings on CD and on the web. Neither LACCEI nor the editors will be responsible either for the content or for the implications of what is expressed in the paper.