A BACKGROUND PAPER >> ENGINEERING OPTIONS FOR RESILIENT INFRASTRUCTURE This work is a product of the staff of The World Bank and the Global Facility for Disaster Reduction and Recovery (GFDRR) with external contributions. The sole responsibility of this publications lies with the authors. The findings, analysis and conclusions expressed in this document do not necessarily reflect the views of any individual partner organization of The World Bank (including the European Union), its Board of Directors, or the governments they represent, and therefore they are not responsible for any use that may be made of the information contained therein. Although the World Bank and GFDRR make reasonable efforts to ensure all the information presented in this document is correct, its accuracy and integrity cannot be guaranteed. Use of any data or information from this document is at the user’s own risk and under no circumstances shall the World Bank, GFDRR or any of its partners be liable for any loss, damage, liability or expense incurred or suffered which is claimed to result from reliance on the data contained in this document. The boundaries, colors, denomination, and other information shown in any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean FINAL REPORT [Revised] Submitted February 26, 2021 Contract 7195879 by Miyamoto International, Inc. About the Project Contract #7195879, Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean, presents a high-level analysis of the engineering options available to increase the resilience of various infrastructure assets to floods, winds, and earthquakes as well as an estimate of the cost of these options in Caribbean countries. About the World Bank Group The World Bank Group is one of the world’s largest sources of funding and knowledge for developing countries. Its five institutions share a commitment to reducing poverty, increasing shared prosperity, and promoting sustainable development. The World Bank Group is committed to open development and has opened its data, knowledge, and research to foster innovation and increase transparency in development. About Miyamoto International, Inc. Miyamoto International Inc. (“Miyamoto”) is a global structural engineering and disaster-risk reduction firm providing resiliency expertise that sustains industries and safeguards communities around the world. We are experts in earthquake-resilient engineering practices that reduce damage and facilitate disaster recovery. We design new construction and assess existing buildings to address specific vulnerabilities to disasters. We prioritize solutions that limit damage, business interruption and loss of life. Miyamoto is strategically located worldwide in regions vulnerable to disaster to positively impact economies and save lives. DISCLAIMER The opinions, findings, and conclusions stated herein are those of the authors (Miyamoto International, Inc.) and do not necessarily reflect the views of the World Bank Group. [Revised] Submitted on: February 26, 2021 By: Miyamoto International, Inc. www.miyamotointernational.com Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |i EXECUTIVE SUMMARY The geographical location of Caribbean counties makes them particularly vulnerable to natural hazards; see Table 1-1 (ThinkHazard, 2020). Buildings, water, transport, and power infrastructure represent key assets for these countries, yet many of these assets are located in highly hazard-prone zones and thus are exposed to the resulting hazard impacts. Country Earthquake Hurricane Flood (Coastal and river) Suriname Low Low High Trinidad and Tobago Medium High High Guyana Low Low High Belize Medium High High Haiti Medium High High Dominican Republic Medium High High Antigua and Barbuda Medium No data Medium Dominica Medium No data High Grenada Medium No data Medium Saint Kitts and Nevis Medium No data Medium Saint Lucia Medium No data Medium Saint Vincent and the Grenadines Medium No data Medium Sint Maarten No data No data No data Barbados Medium No data Medium Jamaica Medium High High Bahamas Medium High High Table 1-1. Relative importance of various hazards in different countries (ThinkHazard, 2020) The continuous operation of critical public buildings, water, transport, and power infrastructure allows for ongoing economic growth and societal wellbeing. Damage to these sectors results in direct economic losses, operation interruption, and has cascading effects on other infrastructure. To enhance the resiliency of this critical infrastructure, it is important to undertake an evaluation program (condition assessment), identify the vulnerable components, and devise effective strengthening techniques for vulnerable components; the goal of such a program is a reduction in damage, economic losses, and downtime in future natural hazard events. Table 1-2 presents the key infrastructure, typologies, and vulnerable components considered in this report. The table also identifies the expected governing hazards for various components in the Caribbean. In Table 1-2, the governing hazards include earthquake ground shaking (“earthquake”), earthquake liquefaction (“liquefaction”), hurricane (“wind”), and flood (“flood”). Table 1-3 presents the proposed strengthening measures for the vulnerable components identified in Table 1-2. Also presented in this table are the estimated capital costs and associated benefits with risk reduction once strengthening (mitigation) has been completed. These values are expressed as a ratio relative to the cost of new construction for the assets. The cost-effectiveness of mitigation, as presented in Table 1-3, is consistent with findings from other researchers1 and can be seen graphically in Figure 1-1, which shows that once retrofitting is implemented, the probability of the infrastructure experiencing severe damage or collapse is reduced from approximately 80% to 20%. Based on the literature review and findings in this report, the following is noted: 1 NIBS (2019) shows a benefit-to-cost ratio of 3-to-8 for infrastructure for earthquake, wind, or flood risk mitigation. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | ii • The 16 countries in focus in this report are geographically located in areas that experience a number of natural hazards, which could damage buildings, water, transport, and power infrastructure. Consequences include not only direct structural damage, but also secondary economic losses due to loss of revenue from inoperability, delayed recovery, and cascading costs to other infrastructure. • Efficacious and economically-feasible strengthening options are available. These measures rely on well-proven techniques and are suitable to the Caribbean countries, given the availability of material and skilled labor resources. Once implemented, such measures would reduce infrastructure vulnerability by multiples of the initial capital cost. • The service life of infrastructure is comprised of planning, design, construction, maintenance, and post-event monitoring. Devising and implementing a construction quality management plan would ensure that retrofits are built as designed and would add to the service life of infrastructure, reducing the need for corrective maintenance. Preventive maintenance and inspection would reduce the cost of expensive future repairs. Early-warning and post-event inspection can result in reduced delays and expedite the return to functionality. • Creative financing measures and life-cycle cost analyses (LCCA) can be used to identify and mitigate project risks, accelerate project delivery, and coordinate the design and construction phases. The following steps are recommended to assist in implementation of an infrastructure resilience program: • Perform a review of local design and construction codes and prepare recommendations for revisions based on recent intensity of natural hazard events. • Conduct inventory assessment to develop a searchable database of critical components and to screen the components and facilities for a phased retrofitting program. • Implement a feasibility study for a selected sector and subsector. Preform probabilistic LCCA to identify the most effective retrofits. For the probabilistic LCCA, incorporate realistic initial capital costs and recurring costs. • Program and undertake retrofitting for one of the countries in a selected sector. This program can serve as a pilot program for use in other countries with similar regional characteristics. • Region-wide organizations, like CDEMA, can develop a post-event damage assessment program, prepare paper and electronic forms, train engineers to perform assessments, and maintain an updated database of trained professionals that can perform these tasks in the event of natural hazards. • Because hurricanes and coastal floods are a major natural hazard for most countries in the Caribbean, undertake an evaluation of flood protection (levees, embankments, etc.) and flood prevention (drainage, collection, etc.) structures in place for coastal areas. Consider improvements and resilience measures for these structures. Such measures provide an additional layer of protection against natural hazards for the infrastructure sectors considered in this report. Existing Risk mitigated Figure 1-1. Risk reduction as a result of mitigation Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | iii Sector Subsector Type Component Vulnerability Hazard(s) Out-of-plane capacity Earthquake Unreinforced Masonry Wall In-plane capacity Earthquake (URM) Roof Roof connection Wind Shape and figure Structural irregularity Earthquake Reinforced/concrete frame Wall Infill wall connection Earthquake (RC/CF) with infill wall Roof Roof connection Wind Opening Building envelope Flood Schools Shape and figure Structural irregularity Earthquake Hospitals RC frame Ductility Non-ductile detailing Earthquake Government External wall External wall Wind Emergency Stiffness Stiffness Earthquake; Wind Hospitality Steel frame Strength Structural capacity Earthquake BUILDINGS Hotels Roof Roof connection Wind Stiffness Stiffness Earthquake; Wind Steel frame high-rise Ductility Non-ductile detailing Earthquake Cladding External curtain wall Wind Roof Roof connection Wind Wood frame Foundation Foundation ties Wind; Flood Elevation Elevated construction Flood All Foundation Lack of capacity Liquefaction Out-of-plane capacity Earthquake Wall In-plane capacity Earthquake Water Ductility Non-ductile detailing Earthquake Plant building Treatment Roof Roof connection Wind (Masonry/RC frame) Plants External wall External wall Wind (WTP) Opening Building envelope Flood Elevation Elevated construction Flood Wastewater Foundation Foundation Earthquake Water storage Treatment Anchorage Anchorage Earthquake Plants Foundation Foundation tie Earthquake; Wind WATER (WWTP) Equipment (mechanical Elevation Elevating equipment Flood and electrical) Floodproofing Dry floodproofing Flood All Foundation Inadequate capacity Liquefaction Underground Brittle pipes Earthquake; Liquefaction; Pipelines Pipe (Asbestos-cement, Brittle pipes Flood (UGP) cast iron, etc.) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | iv Sector Subsector Type Component Vulnerability Hazard(s) Limited deformation Joints Conventional Earthquake; Liquefaction capacity Subsurface Substrate washout Flood Paved roads Asphalt paved Surface Strength loss Flood Columns Lack of ductility Earthquake Concrete girder Joints Short seat Earthquake members Slender members Earthquake Steel truss Connections Riveted connections Earthquake Bridges Bearings Unseating Earthquake Steel girder Cross frames Slender members Earthquake TRANSPORT Steel bridges Welded connections Low fatigue capacity Wind Suspension Bridge stability Low torsional stiffness Wind All Foundation Lack of capacity Earthquake Design Nonductile construction Earthquake Grade construction No flood protection Flood Power plants Low-rise RC Nonstructural (traditional) Lack of anchorage Earthquake components Foundation Shallow Earthquake; Liquefaction Design construction Strength stiffness Earthquake Grade construction No flood protection Flood Control building Battery racks, cable Lack of anchorage/bracing Earthquake trays, etc. Substations Foundation Shallow Earthquake; Liquefaction No or inadequate POWER Transformers Anchorage Earthquake anchorage Frames Steel members Slender low damped Earthquake; Wind Equipment High-voltage units Not qualified Earthquake Tower Lack of strength stiffness Wind; Earthquake Transmission Lattice towers Foundation Inadequate capacity Flood; Liquefaction towers Lines Vibration. Low damping Wind Table 1-2. Matrix for the various infrastructure sectors, vulnerable components and governing hazards Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |v Cost, %2 Subsector Hazard Vulnerable Component Resilience Measure Improvement Vulnerability Cost Reduction Wall out of plane Add wall bracing (anchor and strongback) 15% 40% In plane Add RC shear wall or Steel brace 20% 60% Add Shear wall, Moment frame or Braced Structural irregularity 15% 40% frame Earthquake Ductility Add Column jacketing or Beam retrofit 10% 30% Schools, Stiffness and capacity Add Steel brace or Steel plate shear wall 15% 50% Hospitals, Add Seismic isolation or Energy dissipation Energy absorption 40% 80% Government damper buildings, Enlarge Spread footing or Add pile Foundation 40% 80% Emergency Liquefaction foundation centers, Soil Install Soil grouting or Stone/gravel columns 30% 80% Hospitality Roof connection Improve Roof connection anchor 10% 50% and hotels External curtain wall and Wind Improve Wall connection attachment 10% 30% wall Foundation connection Add Connection anchor bolts and stiffeners 10% 50% Building envelope Install flood shield 10% 80% Flood Elevated construction Elevate floodwall 15% 80% Foundation tie Strengthen anchoring to foundation 10% 50% Wall out-of-plane (Building) Add wall bracing (anchor and strongback) 15% 40% In-plane (Building) Add RC shear wall 20% 60% Ductility (Building) Add column jacketing 10% 30% Earthquake Foundation (Water storage) Add concrete foundation 40% 50% WTP/WWTP: Anchorage (Water storage) Strengthen anchorage 10% 30% Plant building Foundation tie (Equipment) Improve foundation tie 10% 20% (Masonry/RC Foundation (All) Enlarge spread footing or add pile foundation 40% 80% frame), Water Liquefaction Soil (All) Install soil grouting 30% 80% storage, Roof connection (Building) Strengthen roof connection 10% 50% Equipment Wind External wall (Building) Improve wall connection 10% 30% Foundation tie (Equipment) Improve foundation tie 10% 20% Building envelope Flood Install flood shield 10% 80% (Building) 2 Percentage of reconstruction cost (Est.) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | vi Cost, %2 Subsector Hazard Vulnerable Component Resilience Measure Improvement Vulnerability Cost Reduction Elevated construction Construct floodwall 15% 80% (Building) Elevating (Equipment) Elevate equipment 20% 80% Dry floodproofing Install watertight barrier 15% 80% (Equipment) Replace with ductile pipes; Trenchless Earthquake Pipes 40% 80% UGP technologies Liquefaction Joints Add flexible joints 30% 80% Add sealant 2% 20% Paved roads Flood Paved surface Strengthen (thicker pavement) 20% 50% Concrete surface 40% 80% Columns Add fiber-reinforced polymer or steel casing 10% 60% Bearing Add cable restrainers 2% 50% Earthquake Abutment, pier seats Add concrete bolsters 5% 50% Bridges Steel cross frames Fatigue-resistant details 10% 80% Flood Add riprap 5% 20% Flood Foundations, abutments Add new piles 40% 80% Liquefaction Transformer Improved anchorage 5% 20% Substation Equipment Qualified components 10% 40% Earthquake components Equipment Spares 10% 40% Support frames Add passive dampers 5% 20% Transmission tower Strengthen steel members 20% 50% Wind Transmission Transmission line Add vibration dampers 5% 50% system Flood, Foundations Add micro piles 40% 80% Liquefaction Table 1-3. Strengthening techniques and the associated capital costs and benefit Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | vii ACKNOWLEDGMENTS Miyamoto International, Inc. wishes to thank the World Bank Group project team, led by Ms. Julie Rozenberg, for the opportunity to support this critical work, and for their review and insights. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | viii CONTENTS Executive Summary .................................................................................................................................................... ii Acknowledgments .................................................................................................................................................... viii Contents .......................................................................................................................................................................ix List of Tables .............................................................................................................................................................. xv List of Figures ........................................................................................................................................................ xviii Referenced Organizations ...................................................................................................................................... xxii Acronyms, Abbreviations, and Notations ............................................................................................................ xxiii 1. Introduction ............................................................................................................................................................. 1 1.1 Overview ............................................................................................................................................................. 1 1.2 Scope and objectives ...........................................................................................................................................1 1.3 Limitations .......................................................................................................................................................... 2 1.4 Subject countries and vulnerable infrastructure ..................................................................................................2 1.5 Evaluation methodology......................................................................................................................................5 1.6 Organization of the report ...................................................................................................................................5 2. Natural Hazards ...................................................................................................................................................... 6 2.1 Introduction ......................................................................................................................................................... 6 2.2 Natural Hazards ...................................................................................................................................................6 2.2.1 Earthquake hazard........................................................................................................................................6 2.2.1.1 Ground shaking ....................................................................................................................................6 2.2.1.2 Liquefaction ..........................................................................................................................................7 2.2.2 Wind hazard .................................................................................................................................................7 2.2.3 Flood hazard ................................................................................................................................................7 2.3 Infrastructure impact ......................................................................................................................................... 18 2.4 Selection of governing hazard ........................................................................................................................... 19 2.5 Discussion ......................................................................................................................................................... 20 3. Building Infrastructure ......................................................................................................................................... 21 3.1 Introduction ....................................................................................................................................................... 21 3.2 Geographical distribution .................................................................................................................................. 21 3.3 Typology ........................................................................................................................................................... 24 3.4 Vulnerable components ..................................................................................................................................... 25 3.5 Design codes ..................................................................................................................................................... 28 3.5.1 Seismic design ........................................................................................................................................... 29 3.5.1.1 Seismic design in the Caribbean ......................................................................................................... 29 3.5.1.2 Seismic design in Trinidad and Tobago ............................................................................................. 31 3.5.2 Wind design ............................................................................................................................................... 32 3.5.2.1 Wind design in the Caribbean............................................................................................................. 32 3.5.2.2 Wind design in Dominica ................................................................................................................... 35 3.5.3 Flood design............................................................................................................................................... 36 3.5.3.1 Flood design by ASCE 7-16 ............................................................................................................... 36 3.5.3.2 Flood design in Jamaica ..................................................................................................................... 37 3.6 Fragility functions ............................................................................................................................................. 37 3.6.1 Earthquake ground shaking: FEMA Hazus fragility functions .................................................................. 38 3.6.1.1 URM buildings ................................................................................................................................... 39 3.6.1.2 RC/CF with infill wall buildings ........................................................................................................ 40 3.6.1.3 RC frame buildings............................................................................................................................. 42 3.6.1.4 Steel frame buildings .......................................................................................................................... 44 3.6.1.5 Steel frame high-rise buildings ........................................................................................................... 46 3.6.2 Earthquake liquefaction: FEMA Hazus fragility functions ........................................................................ 48 3.6.2.1 Overview ............................................................................................................................................ 48 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | ix 3.6.2.2 Estimation process for a liquefaction damage probability .................................................................. 48 3.6.2.3 Liquefaction fragility functions (lateral spreading and ground settlement) ........................................ 48 3.6.2.4 Liquefaction damage after retrofit implementation ............................................................................ 49 3.6.3 Wind: FEMA Hazus fragility functions ..................................................................................................... 49 3.6.3.1 Overview ............................................................................................................................................ 49 3.6.3.2 Masonry/Concrete buildings ............................................................................................................... 49 3.6.3.3 Wood/Steel buildings ......................................................................................................................... 51 3.6.4 Flood: FEMA Hazus damage functions ..................................................................................................... 53 3.6.4.1 Depth-damage curve for flood hazard ................................................................................................ 53 3.6.4.2 Velocity depth-damage curve for flood hazard .................................................................................. 54 3.6.4.3 Flood damage after retrofit implementation ....................................................................................... 55 3.7 Strengthening techniques .................................................................................................................................. 55 3.7.1 Earthquake: ground shaking ...................................................................................................................... 55 3.7.1.1 Wall out-of-plane strengthening ......................................................................................................... 55 3.7.1.2 In-plane strengthening ........................................................................................................................ 57 3.7.1.3 Structural irregularity improvement ................................................................................................... 61 3.7.1.4 Ductility improvement........................................................................................................................ 63 3.7.1.5 Stiffness and capacity improvement ................................................................................................... 64 3.7.1.6 Energy dissipation system .................................................................................................................. 65 3.7.2 Earthquake: liquefaction ............................................................................................................................ 67 3.7.2.1 Foundation retrofit .............................................................................................................................. 67 3.7.2.2 Soil improvement ............................................................................................................................... 68 3.7.3 Wind .......................................................................................................................................................... 69 3.7.3.1 Roof connection.................................................................................................................................. 70 3.7.3.2 External curtain wall and wall ............................................................................................................ 70 3.7.3.3 Foundation connection ....................................................................................................................... 72 3.7.4 Flood .......................................................................................................................................................... 72 3.7.4.1 Building envelope ............................................................................................................................... 72 3.7.4.2 Elevated construction ......................................................................................................................... 73 3.7.4.3 Foundation tie ..................................................................................................................................... 74 3.8 Cost analysis ...................................................................................................................................................... 76 3.8.1 Introduction................................................................................................................................................ 76 3.8.2 Reconstruction costs .................................................................................................................................. 76 3.8.2.1 Construction cost in the Caribbean ..................................................................................................... 76 3.8.2.2 Construction cost in Saint Vincent and the Grenadines ...................................................................... 76 3.8.2.3 Construction cost in Belize ................................................................................................................. 78 3.8.2.4 Construction cost in the U.S. .............................................................................................................. 78 3.8.3 Strengthening costs .................................................................................................................................... 79 3.8.3.1 Methodology for estimating seismic retrofit cost ............................................................................... 79 3.8.3.2 Cost-benefit analysis of seismic retrofit for URM buildings .............................................................. 81 3.8.3.3 Retrofit cost consideration in FEMA documents................................................................................ 82 3.8.4 Life-cycle cost ........................................................................................................................................... 87 3.9 Discussion ......................................................................................................................................................... 87 4. Water Infrastructure ............................................................................................................................................. 89 4.1 Introduction ....................................................................................................................................................... 89 4.2 Geographical distribution .................................................................................................................................. 91 4.3 Typology ......................................................................................................................................................... 102 4.4 Vulnerable components ................................................................................................................................... 108 4.5 Design practice ................................................................................................................................................ 112 4.5.1 Overview.................................................................................................................................................. 112 4.5.2 Earthquake design .................................................................................................................................... 112 4.5.3 Wind design ............................................................................................................................................. 112 4.5.4 Flood design............................................................................................................................................. 113 4.5.5 WTP/WWTP design guideline manual in Trinidad and Tobago (WASA, 2008) .................................... 113 4.5.5.1 Acts, codes and standards ................................................................................................................. 113 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |x 4.5.5.2 Structural design considerations ....................................................................................................... 113 4.5.5.3 Construction requirements ................................................................................................................ 113 4.6 Fragility functions ........................................................................................................................................... 114 4.6.1 Earthquake fragility functions based on FEMA Hazus methodology ...................................................... 114 4.6.1.1 Overview .......................................................................................................................................... 114 4.6.1.2 WTP.................................................................................................................................................. 114 4.6.1.3 WWTP .............................................................................................................................................. 116 4.6.1.4 Underground potable water pipes ..................................................................................................... 119 4.6.2 Wind fragility functions ........................................................................................................................... 121 4.6.3 Flood fragility functions .......................................................................................................................... 121 4.6.4 Restoration fragility functions ................................................................................................................. 122 4.7 Strengthening techniques ................................................................................................................................ 124 4.7.1 Introduction.............................................................................................................................................. 124 4.7.2 WTP and WWTP ..................................................................................................................................... 124 4.7.2.1 Earthquake: ground shaking ............................................................................................................. 125 4.7.2.2 Earthquake: liquefaction ................................................................................................................... 130 4.7.2.3 Wind ................................................................................................................................................. 132 4.7.2.4 Flood................................................................................................................................................. 135 4.7.3 Underground pipes ................................................................................................................................... 136 4.7.3.1 Earthquake ........................................................................................................................................ 137 4.7.3.2 Liquefaction ...................................................................................................................................... 141 4.7.3.3 Wind ................................................................................................................................................. 141 4.7.3.4 Flood................................................................................................................................................. 141 4.8 Cost analysis .................................................................................................................................................... 141 4.8.1 Introduction.............................................................................................................................................. 141 4.8.2 Reconstruction costs ................................................................................................................................ 142 4.8.4 Life-cycle cost ......................................................................................................................................... 145 4.9 Discussion ....................................................................................................................................................... 145 5. Transport Infrastructure .................................................................................................................................... 147 5.1 Introduction ..................................................................................................................................................... 147 5.2 Geographical distribution ................................................................................................................................ 147 5.3 Typology ......................................................................................................................................................... 150 5.4 Vulnerable components ................................................................................................................................... 152 5.5 Bridges ............................................................................................................................................................ 155 5.5.1 Design codes ............................................................................................................................................ 155 5.5.1.1 Seismic design in Trinidad and Tobago ........................................................................................... 155 5.5.1.2 Wind design in the Dominican Republic .......................................................................................... 156 5.5.2 Fragility functions based on FEMA Hazus methodology ........................................................................ 156 5.5.2.1 Overview .......................................................................................................................................... 156 5.5.2.2 Damage states ................................................................................................................................... 157 5.5.2.3 Fragility functions for ground shaking ............................................................................................. 157 5.5.2.4 Fragility functions: ground failure (liquefaction) ............................................................................. 158 5.5.2.5 Example Bridge ................................................................................................................................ 159 5.5.3 Strengthening measures for earthquakes.................................................................................................. 161 5.5.3.1 Isolation casing ................................................................................................................................. 161 5.5.3.2 Concrete column retrofit for increased capacity ............................................................................... 162 5.5.3.3 Retrofit for continuity ....................................................................................................................... 163 5.5.3.4 Foundation retrofit ............................................................................................................................ 164 5.5.3.5 Retrofit of steel diaphragms for steel girder bridges ........................................................................ 164 5.5.3.6 Retrofit of riveted members .............................................................................................................. 165 5.5.3.7 Retrofit of lattice members ............................................................................................................... 165 5.5.4 Strengthening measures for liquefaction.................................................................................................. 166 5.5.1 Strengthening measures for wind ............................................................................................................. 169 5.5.2 Strengthening measures for flood ............................................................................................................ 170 5.6 Paved roads (asphalt)....................................................................................................................................... 171 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xi 5.6.1 Overview.................................................................................................................................................. 171 5.6.2 Design codes ............................................................................................................................................ 171 5.6.2.1 Seismic design in Dominica ............................................................................................................. 171 5.6.2.2 Seismic design in Jamaica ................................................................................................................ 171 5.6.3 Fragility functions .................................................................................................................................... 173 5.6.4 Strengthening measures ........................................................................................................................... 174 5.6.4.1 Compliance with AASHTO design procedure ................................................................................. 174 5.6.4.2 Strengthening methods for flood mitigation ..................................................................................... 176 5.7 Cost analysis .................................................................................................................................................... 179 5.7.1 Introduction.............................................................................................................................................. 179 5.7.2 Assumptions in the cost analysis ............................................................................................................. 179 5.7.3 Reconstruction costs ................................................................................................................................ 181 5.7.3.1 Cost basis .......................................................................................................................................... 181 5.7.3.2 Bridges.............................................................................................................................................. 181 5.7.3.3 Paved roads....................................................................................................................................... 184 5.7.4 Strengthening costs .................................................................................................................................. 186 5.7.4.1 Bridges.............................................................................................................................................. 186 5.7.4.2 Paved roads....................................................................................................................................... 191 5.7.5 Life-cycle cost ......................................................................................................................................... 192 5.8 Discussion ....................................................................................................................................................... 192 6. Power Infrastructure ........................................................................................................................................... 193 6.1 Introduction ..................................................................................................................................................... 193 6.2 Geographical distribution ................................................................................................................................ 194 6.3 Typology ......................................................................................................................................................... 199 6.4 Vulnerable components ................................................................................................................................... 202 6.5 Design practice ................................................................................................................................................ 204 6.5.1 Overview.................................................................................................................................................. 204 6.5.2 Wind design ............................................................................................................................................. 205 6.5.3 Earthquake loading .................................................................................................................................. 205 6.5.4 Flood loading ........................................................................................................................................... 205 6.5.5 Load combinations................................................................................................................................... 205 6.6 Fragility functions ........................................................................................................................................... 205 6.6.1 Earthquake fragility functions based on FEMA Hazus methodology ...................................................... 205 6.6.1.1 Overview .......................................................................................................................................... 205 6.6.1.2 Damage states ................................................................................................................................... 205 6.6.1.3 Fragility functions for ground shaking ............................................................................................. 206 6.6.1.4 Fragility functions: ground failure (liquefaction) ............................................................................. 206 6.6.1.5 Typical substation ............................................................................................................................. 206 6.6.2 Wind fragility functions ........................................................................................................................... 208 6.7 Strengthening techniques ................................................................................................................................ 209 6.7.1 Substations ............................................................................................................................................... 209 6.7.1.1 Use of seismically qualified components ......................................................................................... 209 6.7.1.2 Redundancy ...................................................................................................................................... 209 6.7.1.3 Transformers..................................................................................................................................... 210 6.7.1.4 Flexible connectors ........................................................................................................................... 211 6.7.1.5 Wire rope isolation ........................................................................................................................... 211 6.7.1.6 Spring dampers ................................................................................................................................. 212 6.7.1.7 Retrofit design of elevated supports ................................................................................................. 212 6.7.1.8 Composite insulators ........................................................................................................................ 213 6.7.2 Transmission towers and lines ................................................................................................................. 213 6.7.2.1 Retrofit of transmission towers ......................................................................................................... 213 6.7.2.2 Retrofit of transmission lines ............................................................................................................ 215 6.7.3 Foundation retrofit ................................................................................................................................... 216 6.8 Cost analysis .................................................................................................................................................... 217 6.8.1 Introduction.............................................................................................................................................. 217 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xii 6.8.2 Reconstruction costs ................................................................................................................................ 217 6.8.2.1 Capital cost ....................................................................................................................................... 217 6.8.2.2 Component cost ................................................................................................................................ 218 6.8.2.3 Substation cost estimator .................................................................................................................. 219 6.8.3 Strengthening costs .................................................................................................................................. 220 6.8.3.1 Substations........................................................................................................................................ 220 6.8.3.2 Transmission towers ......................................................................................................................... 221 6.8.3.3 Transmission lines ............................................................................................................................ 222 6.8.4 Life-cycle cost ......................................................................................................................................... 222 6.9 Discussion ....................................................................................................................................................... 222 7. Additional Measures ............................................................................................................................................ 224 7.1 Introduction ..................................................................................................................................................... 224 7.2 Construction quality management ................................................................................................................... 224 7.2.1 Construction database and record keeping ............................................................................................... 224 7.2.2 Electronic real-time monitoring ............................................................................................................... 225 7.2.3 Use of proven technologies ...................................................................................................................... 227 7.2.4 Contract incentives .................................................................................................................................. 230 7.3 Screening and prioritization ............................................................................................................................ 231 7.4 Preventive maintenance ................................................................................................................................... 233 7.5 Use of advanced technologies for inspection and damage assessment ............................................................ 234 7.6 Monitoring and early warning system (EWS) ................................................................................................. 238 7.6.1 United States applications ........................................................................................................................ 238 7.6.2 Application to the Caribbean ................................................................................................................... 241 7.7 Response protocol ........................................................................................................................................... 246 7.7.1 International best practice ........................................................................................................................ 246 7.7.2 Applications to the Caribbean .................................................................................................................. 248 7.8 Regional approach ........................................................................................................................................... 249 7.9 Discussion ....................................................................................................................................................... 249 8. Mitigation Cost and Financing ........................................................................................................................... 251 8.1 Introduction ..................................................................................................................................................... 251 8.2 Cost effectiveness of mitigation ...................................................................................................................... 251 8.3 Local costs ....................................................................................................................................................... 251 8.4 Life-cycle cost analysis ................................................................................................................................... 252 8.4.1 Overview.................................................................................................................................................. 252 8.4.2 Application to paved roads ...................................................................................................................... 252 8.4.3 Application to bridges .............................................................................................................................. 253 8.4.4 Application to buildings........................................................................................................................... 254 8.4.5 Application to energy sector .................................................................................................................... 254 8.5 Project contracting........................................................................................................................................... 255 8.5.1 Overview.................................................................................................................................................. 255 8.5.2 Design Bid Build (DBB) .......................................................................................................................... 255 8.5.3 Design Build ............................................................................................................................................ 255 8.5.4 Public-private partnership (P3) ................................................................................................................ 256 8.5.5 Construction Manager/General Contractor (CMGC) ............................................................................... 257 8.6 Discussion ....................................................................................................................................................... 257 9. Summary, Conclusions, and Recommendations ............................................................................................... 258 9.1 Summary ......................................................................................................................................................... 258 9.1.1 Overview.................................................................................................................................................. 258 9.1.2 Critical infrastructure ............................................................................................................................... 259 9.1.3 Vulnerable components ........................................................................................................................... 260 9.1.4 Design codes and standards ..................................................................................................................... 262 9.1.5 Strengthening and resilience measures .................................................................................................... 262 9.1.6 Cost considerations .................................................................................................................................. 264 9.2 Conclusions ..................................................................................................................................................... 266 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xiii 9.3 Recommendations ........................................................................................................................................... 267 9.4 Future activities ............................................................................................................................................... 267 References ................................................................................................................................................................ 268 Review of Data ............................................................................................................................... 279 Date Collection template ................................................................................................................................ 279 Collected data ................................................................................................................................................. 279 World Bank-supplied data .............................................................................................................................. 279 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xiv LIST OF TABLES Table 1-1. Relative importance of various hazards in different countries (ThinkHazard, 2020) .................................. ii Table 1-2. Matrix for the various infrastructure sectors, vulnerable components and governing hazards .................... v Table 1-3. Strengthening techniques and the associated capital costs and benefit ..................................................... vii Table 1-1. General information for subject countries (various sources) ........................................................................1 Table 1-2. Vulnerable infrastructure and associated natural hazards in subject countries (WB, 2020) ......................... 3 Table 2-1. Relative importance of various hazards in different countries (ThinkHazard, 2020) ...................................6 Table 2-2. Hazard contribution to AAL for selected countries (adapted from CHARIM, 2020) ..................................6 Table 2-3. Hurricane category and sustained wind speed (National Hurricane Center, 2020) ......................................7 Table 2-4. Peak gust wind (mph) for subject countries (adapted from PAHO, 2019) ...................................................7 Table 2-5. Matrix of considered hazards for various infrastructure ............................................................................. 19 Table 2-6. Matrix of considered typologies for various infrastructure ........................................................................ 20 Table 3-1. Analysis matrix for building infrastructure ................................................................................................ 21 Table 3-2. Vulnerable components of building infrastructure and corresponding hazards ......................................... 28 Table 3-3. Z values for various Caribbean countries (CUBiC, 1985) .......................................................................... 29 Table 3-4. K factors for various structural types, steel and concrete (CUBiC, 1985) .................................................. 30 Table 3-5. K factors for various structural types, timber (CUBiC, 1985) .................................................................... 30 Table 3-6. I factors for various structural types, steel and concrete (CUBiC, 1985) ................................................... 30 Table 3-7. I Load combinations for design (Chin, 2008b)........................................................................................... 32 Table 3-8. Reference wind velocity pressures for the Caribbean (CUBiC, 1985) ....................................................... 33 Table 3-9. Relationship between reference velocity pressure q ref and peak wind speeds over short time intervals in open terrain (CUBiC, 1985) ........................................................................................................................................ 34 Table 3-10. URM building damage states (adapted from FEMA, 2003a) ................................................................... 39 Table 3-11. Parameters of seismic fragility functions for URM buildings (FEMA, 2013a)........................................ 40 Table 3-12. RC/CF with infill building damage states (adapted from FEMA, 2003a) ................................................ 41 Table 3-13. Parameters of seismic fragility functions for RC/CF with infill wall buildings (FEMA, 2013a) ............. 41 Table 3-14. RC frame building damage states (adapted from FEMA, 2003a) ............................................................ 43 Table 3-15. Parameters of seismic fragility functions for RC frame buildings (FEMA, 2013a) ................................. 43 Table 3-16. Steel frame building damage states (adapted from FEMA, 2003a) .......................................................... 45 Table 3-17. Parameters of seismic fragility functions for steel frame buildings (FEMA, 2013a) ............................... 45 Table 3-18. Steel frame high rise building damage states (adapted from FEMA, 2003a) ........................................... 47 Table 3-19. Parameters of seismic fragility functions for steel frame high-rise buildings (FEMA, 2013a) ................ 47 Table 3-20. Parameters of liquefaction fragility functions (FEMA, 2013a) ................................................................ 48 Table 3-21. Damage states for Masonry/Concrete buildings (FEMA, 2013b) ............................................................ 50 Table 3-22. Damage states for wood/steel buildings (FEMA, 2013b) ........................................................................ 52 Table 3-23. Construction cost of buildings in the Caribbean in US$ per Sq. Ft. (BCQS, 2019) ................................. 77 Table 3-24. Construction cost of essential facilities in Saint Vincent and the Grenadines (Campbell, 2020) ............. 77 Table 3-25. Construction cost of residential buildings in Belize (Henderson, 2001) .................................................. 78 Table 3-26. Construction cost of buildings in the U.S. in US$ per Sq. Ft. (Cumming, 2020) ..................................... 79 Table 3-27. Building group and types for the prediction model (NIST, 2017) ............................................................ 80 Table 3-28. Estimates of coefficients and variations from training main model (NIST, 2017) ................................... 81 Table 3-29. URM building classes and performance levels in Portland (Goettel, 2016) ............................................. 82 Table 3-30. Building group and types and benefit-cost ratios for the prediction model (Goettel, 2016)..................... 82 Table 3-31. Summary of retrofit solutions for buildings for natural hazards .............................................................. 86 Table 3-32. Summary of findings for building infrastructure in the Caribbean .......................................................... 88 Table 4-1. Analysis matrix for water infrastructure ..................................................................................................... 90 Table 4-2. Water utility providers for 16 subject countries ......................................................................................... 91 Table 4-3. Examples of water sector components (various) ...................................................................................... 102 Table 4-4. Damage to Grenada’s water infrastructure, Hurricane Ivan (OECS, 2004) ............................................. 108 Table 4-5. Damage to Dominica’s water infrastructure, Hurricane Erika (ACP -EU, 2015) ..................................... 108 Table 4-6. Damage to Dominica’s water infrastructure, Hurricane Maria (ACP -EU, 2017) .................................... 109 Table 4-7. Damage to Saint Vincent and the Grenadines water infrastructure, 2016 floods (ACP-EU, 2016) ......... 109 Table 4-8. Vulnerable components of water infrastructure and corresponding hazards ............................................ 112 Table 4-9. Required compressive strength for concrete construction ........................................................................ 114 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xv Table 4-10. WTP damage states (adapted from FEMA, 2003a) ................................................................................ 114 Table 4-11 WTP fragility parameters, ground motion (adapted from FEMA, 2003a) .............................................. 115 Table 4-12. WWTP damage states (adapted from FEMA, 2003a) ............................................................................ 117 Table 4-13. WWTP fragility parameters, ground motion (adapted from FEMA, 2003a).......................................... 117 Table 4-14. Underground water pipe types (adapted from FEMA, 2003a) ............................................................... 119 Table 4-15. Flood vulnerability, water infrastructure (adapted from FEMA, 2003c) ................................................ 121 Table 4-16. Flood vulnerability, water infrastructure (adapted from FEMA, 2003c)................................................ 122 Table 4-17. Recovery times (OSSPAC, 2013) .......................................................................................................... 124 Table 4-18. Damage to various pipelines in recent earthquakes (adapted from Eidinger and Davis, 2012) ............. 138 Table 4-19. Suggested earthquake strengthening measures for UGP in the subject countries .................................. 139 Table 4-20. Recommended seismic design for different classes of pipelines ............................................................ 140 Table 4-21. Suggested earthquake strengthening measures for UGP in the subject countries .................................. 141 Table 4-22. Suggested liquefaction strengthening measures for UGP in the subject countries ................................. 141 Table 4-23. New wastewater infrastructure cost data (adapted from CEHI, 2013) ................................................... 142 Table 4-24. Cost estimates for example pilot countrywide projects, wastewater (adapted from CEHI, 2013) ......... 142 Table 4-25. Per capita cost in US$ of water and wastewater improvement and access (adapted from WHO, 2008) 142 Table 4-26. Dominica water infrastructure improvement project (CDB, 2020) ........................................................ 143 Table 4-27. Wastewater improvement in Trinidad and Tobago (adapted from GEF-CreW, 2017) .......................... 143 Table 4-28. Example of underground pipe strengthening cost (CWWA, 2015)........................................................ 143 Table 4-29. Water infrastructure components in the study (adapted from Eidinger, 2012) ....................................... 144 Table 4-30. Cost and benefit associated with strengthening of water infrastructure (adapted from Eidinger, 2012) 144 Table 4-31. Estimated retrofitting cost for WTP in Oregon (adapted from MWH, 2012)......................................... 145 Table 4-32. Estimated construction cost for WTP in Oregon (adapted from MWH, 2012) ...................................... 145 Table 4-33. Estimated construction cost for pipes (adapted from MWH, 2012) ....................................................... 145 Table 4-34. Summary of findings for water infrastructure in the Caribbean ............................................................. 146 Table 5-1. Analysis matrix for transport infrastructure ............................................................................................. 147 Table 5-2. Transport infrastructure authorities for 16 subject countries .................................................................... 147 Table 5-3. Vulnerable components of transport infrastructure and corresponding hazards....................................... 154 Table 5-4. Bridge typologies (adapted from FEMA, 2003a) ..................................................................................... 157 Table 5-5. Bridge damage states (adapted from FEMA, 2003a) ............................................................................... 157 Table 5-6. Baseline bridge fragility parameters, ground motion (adapted from FEMA, 2003a) ............................... 157 Table 5-7. Baseline bridge fragility parameters, ground failure (adapted from FEMA, 2003a) ................................ 158 Table 5-8. Properties of the example bridge .............................................................................................................. 159 Table 5-9. Modifications for fragility functions ........................................................................................................ 159 Table 5-10. Types of HMA pavements (AASHTO, 2015) ........................................................................................ 175 Table 5-11. Recommended design criteria, HMA pavement, primary roads (AASHTO, 2015) ............................... 175 Table 5-12. Flood damage consequences (Davis et al., 2013) ................................................................................... 176 Table 5-13. Hypothetical cost and benefit calculations ............................................................................................. 180 Table 5-14. Transport infrastructure construction in subject countries (various sources) ......................................... 181 Table 5-15. Bridge construction cost (adapted from FDOT, 2020) ........................................................................... 182 Table 5-16. Replacement bridge information for Saint Vincent and the Grenadines (Searchlight 2015, 2018)........ 182 Table 5-17. 2020 Bridge construction cost of four new bridges in Saint Vincent and the Grenadines ..................... 183 Table 5-18. Bridge construction cost in Trinidad and Tobago (adapted from Khan-Kernahan, 2013) ..................... 183 Table 5-19. Construction cost of bridges in Saint Lucia (adapted from CDMP, 1998) ............................................. 184 Table 5-20. Construction cost of bridges in Saint Lucia (adapted from Dannion Engineering, 2020) ...................... 184 Table 5-21. Bridge construction cost (adapted from FDOT, 2020) ........................................................................... 185 Table 5-22. Example of pavement costs in Jamaica (adapted from Jamaica Gleaner, 2018) .................................... 185 Table 5-23. Example of pavement component costs (adapted from WBG, 2018) .................................................... 185 Table 5-24. Paved road construction cost in Saint Vincent and the Grenadines........................................................ 185 Table 5-25. Bridge column retrofit scenarios (adapted from Chandrasekaran and Banerjee, 2014) ......................... 186 Table 5-26. Example of cost computations (USDOT, 2016) ..................................................................................... 190 Table 5-27. Example of scenarios (USDOT, 2016) ................................................................................................... 190 Table 5-28. Example of pavement rehabilitation costs (adapted from WBG, 2018) ................................................. 191 Table 5-29. Summary of findings for transport infrastructure in the Caribbean ........................................................ 192 Table 6-1. Analysis matrix for power infrastructure.................................................................................................. 194 Table 6-2. Utility providers for 16 subject countries ................................................................................................. 194 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xvi Table 6-3. Power sector components (various sources) ............................................................................................. 199 Table 6-4. Vulnerable components of power infrastructure and corresponding hazards ........................................... 204 Table 6-5. Design load combinations (adapted from ASCE, 2007) .......................................................................... 205 Table 6-6. Substation damage states (adapted from FEMA, 2003a) ......................................................................... 206 Table 6-7. Substations fragility parameters, ground motion (adapted from FEMA, 2003a) ..................................... 206 Table 6-8. Substation components qualification levels (adapted from IEEE, 2018) ................................................. 209 Table 6-9. Transformer spare strategy matrix (adapted from EPRI, 2014) ............................................................... 210 Table 6-10. New substation cost data for 230 kV system (adapted from Black & Veatch, 2012)............................. 218 Table 6-11. New transmission cost data for 230 kV system (adapted from Black & Veatch, 2012)......................... 218 Table 6-12. New substation cost data for 230 kV system (adapted from MISO, 2020) ............................................ 219 Table 6-13. New transmission tower cost data for 230 kV system (adapted from MISO, 2020) .............................. 219 Table 6-14. New transmission line cost data for 230 kV system (adapted from MISO, 2020) ................................. 219 Table 6-15. Transmission line removal and rebuild cost data (adapted from MISO, 2020) ...................................... 219 Table 6-16. Examples of component unit cost for substation components (PEguru, 2020) ...................................... 220 Table 6-17. Substation retrofit scenarios (adapted from Romero et al., 2013) .......................................................... 220 Table 6-18. Example of transmission tower upgrade cost ......................................................................................... 222 Table 6-19. Example of transmission line upgrade cost ............................................................................................ 222 Table 6-20. Summary of findings for power infrastructure in Caribbean .................................................................. 223 Table 7-1. California transportation infrastructure inventory (adapted from Caltrans, 2020) ................................... 224 Table 7-2. Seismic retrofit management plan (adapted from Caltrans, 2020) ........................................................... 224 Table 7-3. Example of data inventory in bridge book ............................................................................................... 225 Table 7-4. Example of data inventory in substation records ...................................................................................... 225 Table 7-5. Example of construction inspector daily activity log for transport construction or retrofit...................... 226 Table 7-6. Example of construction inspector daily activity log for power construction or retrofit .......................... 227 Table 7-7 Components of PASS (adapted from HA, 2020) ...................................................................................... 231 Table 7-8. Example of bridge screening form ........................................................................................................... 232 Table 7-9. Example of substation screening form ..................................................................................................... 233 Table 7-10. Portion of EWS (adapted from CDEMA, 2020) .................................................................................... 245 Table 8-1. Cost-benefit analysis for natural hazard mitigation (adapted from NIBS, 2019) ..................................... 251 Table 8-2. Example of infrastructure reconstruction cost in the Caribbean (various) ............................................... 251 Table 8-3. Key properties of the bridge (adapted from Hatami and Morcous, 2013) ................................................ 253 Table 8-4. Design build and DBB project approach (adapted from USDOT, 2006) ................................................. 256 Table 8-5. Example of a successful procurement process (adapted from USDOT, 2006) ........................................ 256 Table 8-6. Example of a successful procurement process (adapted from FHWA, 2019) .......................................... 257 Table 9-1. Relative importance of hazards in Caribbean countries based on mortality index (ISDR, 2009) ............ 258 Table 9-2. Matrix for the various infrastructure sectors ............................................................................................ 260 Table 9-3. Vulnerable components of the various infrastructure and the corresponding hazards ............................. 262 Table 9-4. Vulnerable components of the various infrastructure and the corresponding hazards ............................. 264 Table 9-5. Estimated monetary savings for implementing infrastructure resilient measures in the Caribbean ......... 266 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xvii LIST OF FIGURES Figure 1-1. Risk reduction as a result of mitigation..................................................................................................... iii Figure 1-1. Grouping of countries (Kmusser, 2020) ......................................................................................................4 Figure 2-1. Seismic hazard maps (CDMP, 2020) ..........................................................................................................9 Figure 2-2. Liquefaction hazard maps (Kraft, 2013) ................................................................................................... 10 Figure 2-3. Wind hazard maps (PAHO, 2019) ............................................................................................................ 13 Figure 2-4. Flood hazard maps .................................................................................................................................... 17 Figure 2-5. Damage to infrastructure in the Caribbean, 2020 Hurricane season (various sources) ............................. 18 Figure 3-1. Critical building locations for selected Caribbean countries (Gosine, 2020a) .......................................... 24 Figure 3-2. Building typology (various sources) ......................................................................................................... 25 Figure 3-3. Building damages due to earthquake shaking and liquefaction ................................................................ 26 Figure 3-4. Building damages due to hurricane winds and associated floods ............................................................. 27 Figure 3-5. Building design codes of representative countries in the Caribbean (Redmond, 2015) ............................ 28 Figure 3-6. Regional map of design wind pressure contours based on CUBiC (Gibbs, 1998) .................................... 34 Figure 3-7. Typical wind pressure on housing (Dominica, 2018) ............................................................................... 36 Figure 3-8. Dynamic static and total pressure distributions against wall (ASCE, 2017a) ........................................... 37 Figure 3-9. Seismic fragility function studies in the Caribbean................................................................................... 38 Figure 3-10. Seismic fragility functions & probability mass function at PGA=0.4g, Existing URM buildings .......... 40 Figure 3-11. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted URM buildings ...... 40 Figure 3-12. Seismic fragility functions & probability mass function at PGA=0.4g, Existing RC/CF with infill wall buildings ...................................................................................................................................................................... 42 Figure 3-13. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted RC/CF with infill wall buildings .............................................................................................................................................................. 42 Figure 3-14. Seismic fragility functions & probability mass function at PGA=0.4g, Existing RC frame buildings ... 43 Figure 3-15. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted RC frame buildings ..................................................................................................................................................................................... 44 Figure 3-16. Seismic fragility functions & probability mass function at PGA=0.4g, Existing steel frame buildings . 45 Figure 3-17. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted steel frame buildings ..................................................................................................................................................................................... 46 Figure 3-18. Seismic fragility functions & probability mass function at PGA=0.4g, Existing steel frame high-rise buildings ...................................................................................................................................................................... 47 Figure 3-19. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted steel frame high-rise buildings ...................................................................................................................................................................... 48 Figure 3-20. Liquefaction fragility functions for buildings with shallow/unknown foundations ................................ 49 Figure 3-21. Wind fragility functions & probability mass function at PGWS=125 mph, Existing Masonry / Concrete buildings ...................................................................................................................................................................... 51 Figure 3-22. Wind fragility functions & probability mass function at PGWS=125 mph, Retrofitted Masonry / Concrete buildings ....................................................................................................................................................... 51 Figure 3-23. Wind fragility functions & probability mass function at PGWS=125 mph, Existing Wood/Steel buildings ...................................................................................................................................................................... 52 Figure 3-24. Wind fragility functions & probability mass function at PGWS=125 mph, Retrofitted Wood/Steel buildings ...................................................................................................................................................................... 53 Figure 3-25. Building depth damage curves for buildings (FEMA, 2013c) ................................................................ 54 Figure 3-26. Building velocity-depth damage curves for various structural types (FEMA, 2013c) ............................ 55 Figure 3-27. Out-of-plane strengthening for masonry wall (FEMA, 2006b) ............................................................... 56 Figure 3-28. Vertical element connections of out-of-plane strengthening (FEMA, 2006b) ........................................ 57 Figure 3-29. Typical in-plane strengthening, RC shear wall (FEMA, 2006a) ............................................................. 58 Figure 3-30. Typical in-plane strengthening, Steel brace (FEMA, 2006a) .................................................................. 58 Figure 3-31. RC shear wall strengthening (FEMA, 2006b) ......................................................................................... 58 Figure 3-32. Steel buckling restrained brace (BRB) strengthening (NIST, 2015) ....................................................... 60 Figure 3-33. FRP strengthening on existing RC shear wall (FEMA, 2006b) .............................................................. 60 Figure 3-34. RC shear wall addition (or shotcrete) on existing masonry wall (FEMA, 2006b) .................................. 61 Figure 3-35. Examples of structural irregularities of buildings ................................................................................... 62 Figure 3-36. Typical strengthening for structural deficiencies due to irregularities (FEMA, 2006a) .......................... 63 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xviii Figure 3-37. Ductility improvement measures for RC column (FEMA, 2006b) ......................................................... 63 Figure 3-38. Retrofit measures for steel beam ductility (FEMA, 2006b) .................................................................... 64 Figure 3-39. Steel brace frame retrofit (FEMA, 2006b) .............................................................................................. 65 Figure 3-40. Steel plate shear wall retrofit (FEMA, 2006b) ........................................................................................ 65 Figure 3-41. Conceptual figure of seismic isolation system (ASCE, 2017b) .............................................................. 66 Figure 3-42. Base isolation retrofit example of San Francisco City Hall (Naaseh et al., 2006) .................................. 66 Figure 3-43. Typical configurations of energy dissipation damper retrofit (FEMA, 2006b)....................................... 67 Figure 3-44. Energy dissipation damper retrofit for Naval Hospital Bremerton (FEMA, 2007) ................................. 67 Figure 3-45. Typical foundation strengthening (FEMA, 2006b) ................................................................................. 68 Figure 3-46. Permeation grouting on liquefiable soil layers (FEMA, 2006b) ............................................................. 69 Figure 3-47. Soil strengthening for lateral spreading by stone/gravel column (FEMA, 2006b) .................................. 69 Figure 3-48. Typical roof connection improvements (FEMA, 2006b) ........................................................................ 70 Figure 3-49. Typical connection details of external curtain walls ............................................................................... 71 Figure 3-50. Connections of external wall (FEMA, 2007a) ........................................................................................ 71 Figure 3-51. Typical connection retrofits of vertical elements to foundation (FEMA, 2006b) ................................... 72 Figure 3-52. Floodgate installation (FEMA, 2007a) ................................................................................................... 73 Figure 3-53. Flood shield for door (FEMA, 2013) ...................................................................................................... 73 Figure 3-54. Flood shield for window (FEMA, 2013) ................................................................................................. 73 Figure 3-55. Typical floodwalls (FEMA, 2013) .......................................................................................................... 74 Figure 3-56. Floodwalls protecting buildings (FEMA, 2013) ..................................................................................... 74 Figure 3-57. Water loads on building due to flooding (FEMA, 2007a)....................................................................... 75 Figure 3-58. Typical strengthening on foundation tie (FEMA, 2006b) ....................................................................... 75 Figure 3-59. Geographic distribution of URM buildings (purple dot) in Portland (Goettel, 2016) ............................. 82 Figure 4-1. Water (and wastewater) infrastructure (T: PMC, 2020; B: SafeRack, 2020) ............................................ 89 Figure 4-2. Main and secondary pipes (various) ......................................................................................................... 90 Figure 4-3. Water distribution networks ...................................................................................................................... 97 Figure 4-4. Typical locations of WTP and WWTP in subject countries.................................................................... 100 Figure 4-5. WTP in subject countries ........................................................................................................................ 104 Figure 4-6. WWTP in subject countries .................................................................................................................... 106 Figure 4-7. Water pipelines in subject countries ....................................................................................................... 108 Figure 4-8. Water infrastructure (WTP and WWTP) damage, earthquake hazard (EERI, 2010) .............................. 110 Figure 4-9. Damage to pipelines (various) ................................................................................................................ 111 Figure 4-10. Fragility functions for example WTP.................................................................................................... 116 Figure 4-11. Distribution of damage states for WTP ................................................................................................. 116 Figure 4-12. Fragility functions for example WWTP ................................................................................................ 118 Figure 4-13. Distribution of damage states for WWTP ............................................................................................. 119 Figure 4-14. Fragility functions for example UGP .................................................................................................... 120 Figure 4-15. Wind vulnerability, water infrastructure (adapted from FEMA, 2003b) .............................................. 121 Figure 4-16. Damage to WTP and WWTP as a function of flood depth ................................................................... 122 Figure 4-17. Restoration functions, water infrastructure (adapted from FEMA, 2003a) ........................................... 123 Figure 4-18. Wind and flood combined damage (adapted from FEMA, 2003b) ....................................................... 124 Figure 4-19. Wall out-of-plane strengthening (FEMA, 2006b) ................................................................................. 125 Figure 4-20. RC shear wall overlay on existing masonry wall (FEMA, 2006b) ....................................................... 126 Figure 4-21. RC shear wall addition to existing building (FEMA, 2006b) ............................................................... 126 Figure 4-22. Ductility improvement for columns (FEMA, 2006b) ........................................................................... 127 Figure 4-23. Foundation strengthening of water storage tank (Porter, 2008) ............................................................ 128 Figure 4-24. Anchorage improvement with pile addition on water storage tank (Porter, 2008)................................ 129 Figure 4-25. Rigid anchorage strengthening on equipment (FEMA, 2011) .............................................................. 129 Figure 4-26. Vibration isolator support on equipment (FEMA, 2011) ...................................................................... 130 Figure 4-27. Typical foundation strengthening measures (FEMA, 2006b) ............................................................... 131 Figure 4-28. Soil improvement on liquefiable soil layers around existing foundation (FEMA, 2006b) ................... 132 Figure 4-29. Typical roof connection improvements (FEMA, 2006b) ...................................................................... 133 Figure 4-30. Connections of external wall (FEMA, 2007a) ...................................................................................... 134 Figure 4-31. Anchorage strengthening of equipment (FEMA, 2011) ........................................................................ 135 Figure 4-32. Floodwalls protecting buildings and facility site (FEMA, 2013) .......................................................... 135 Figure 4-33. Elevating equipment for flood protection ............................................................................................. 136 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xix Figure 4-34. Watertight shields for equipment for dry floodproofing ....................................................................... 136 Figure 4-35. Examples of brittle pipes (various) ....................................................................................................... 137 Figure 4-36. Examples of ductile pipes (various) ...................................................................................................... 138 Figure 4-37. Existing WTP in Oregon (MWH, 2012) ............................................................................................... 144 Figure 5-1. Road networks (various sources) ............................................................................................................ 150 Figure 5-2. Bridge typology (various sources) .......................................................................................................... 152 Figure 5-3. Paved road typology (various sources) ................................................................................................... 152 Figure 5-4. Bridge damage in California earthquakes (NISEE, 2020) ...................................................................... 153 Figure 5-5. Bridge damage as a result of wind forces and associated flooding ......................................................... 154 Figure 5-6. Road damage due to flooding ................................................................................................................. 154 Figure 5-7. PGA for 975-year earthquake (UWI, 2020) ............................................................................................ 155 Figure 5-8. Wind zones in the country (Grupo Estabilidad Estructural, 2000).......................................................... 156 Figure 5-9. Fragility functions for example bridge .................................................................................................... 160 Figure 5-10. Distribution of damage states ................................................................................................................ 161 Figure 5-11. Ground failure (liquefaction), existing case .......................................................................................... 161 Figure 5-12. Addition of isolation casing (Caltrans, 2020) ....................................................................................... 162 Figure 5-13. Addition of pier walls (Caltrans, 2020) ................................................................................................. 162 Figure 5-14. Addition of column casings (Caltrans, 2020) ........................................................................................ 163 Figure 5-15. Addition of cable restrainers at piers (Caltrans, 2020) .......................................................................... 163 Figure 5-16. Retrofit of abutment seats (Caltrans, 2020) .......................................................................................... 164 Figure 5-17. Retrofit of foundations (Caltrans, 2020) ............................................................................................... 164 Figure 5-18. Retrofit of steel girder diaphragms (FHWA, 2006) .............................................................................. 165 Figure 5-19. Retrofit of steel truss bridge members by replacing rivets with high-strength bolts (Ho, 2006) .......... 165 Figure 5-20. Retrofit of steel riveted lattice members by cover plates (Ho, 2006) .................................................... 166 Figure 5-21. Retrofit of foundations (FHWA, 2011) ................................................................................................. 168 Figure 5-22. Improved crossed frame connection (TXDOT, 2013) .......................................................................... 169 Figure 5-23. Fatigue performance for welded stiffeners (AASHTO, 2017) .............................................................. 169 Figure 5-24. Improved crossed frame connection (GGB, 2020) ............................................................................... 170 Figure 5-25. Application of Riprap (FHWA, 2017) .................................................................................................. 170 Figure 5-26. Cross section (UWI, 2001).................................................................................................................... 171 Figure 5-27. Typical road sections (NWA, 2007) ..................................................................................................... 172 Figure 5-28. Flood fragility functions for roadways (various) .................................................................................. 173 Figure 5-29. Roadway fragility functions for earthquake and flood .......................................................................... 174 Figure 5-30. Typical cross-section composition for HMA pavement ........................................................................ 174 Figure 5-31. Cross-sectional view of HMA pavement types (AASHTO, 2015) ....................................................... 175 Figure 5-32. Pavement and embankment loss based on overtopping (adapted from Davis et al., 2013)................... 176 Figure 5-33. Pavement damage as a function of probability of flooding (adapted from Khan et al., 2017) .............. 179 Figure 5-34. Replacement bridges after flood damage in Saint Vincent and the Grenadines.................................... 182 Figure 5-35. Saint Lucia bridge construction (Dannion Engineering, 2020) ............................................................. 184 Figure 5-36. Retrofit effectiveness of bridge columns (adapted from Chandrasekaran and Banerjee, 2014) ........... 187 Figure 5-37. Bridge under investigation (Asadi et al., 2020) .................................................................................... 188 Figure 5-38. Loss reduction due to retrofitting (adapted from Asadi et al., 2020) .................................................... 189 Figure 5-39. Case study bridge (adapted from USDOT, 2016) ................................................................................. 191 Figure 6-1. Power infrastructure ................................................................................................................................ 193 Figure 6-2. Transmission and distribution systems (various) .................................................................................... 194 Figure 6-3. Power distribution networks (National Energy Grid Map Index, 2020) ................................................. 197 Figure 6-4. Power generating plants (various sources) .............................................................................................. 198 Figure 6-5. Substations in subject countries .............................................................................................................. 201 Figure 6-6. Transmission towers and lines in subject countries ................................................................................ 202 Figure 6-7. Substation damage in past Earthquakes (NISEE, 2020) ......................................................................... 203 Figure 6-8. Transmission tower damage as a result of natural hazards ..................................................................... 204 Figure 6-9. Fragility functions for example substation .............................................................................................. 207 Figure 6-10. Distribution of damage states ................................................................................................................ 208 Figure 6-11. Fragility functions for typical transmission towers (adapted from Panteli et al., 2017)........................ 209 Figure 6-12. Typical transformer anchorage enhancements (Knight and Kempner, 2004) ....................................... 210 Figure 6-13. Seismic base isolation of transformers (Kempner and Riley, 2016) ..................................................... 211 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xx Figure 6-14. Retrofit of rigid bus by flexible connectors (Song and Der Kiureghian, 2004) .................................... 211 Figure 6-15. Seismic protection of circuit breakers, wire ropes (Alessandri et al., 2015) ......................................... 212 Figure 6-16. Seismic protection by means of spring dampers (Riley et al., 2006) .................................................... 212 Figure 6-17. Seismic protection by means of improved support design (Gilani et al., 2000) .................................... 213 Figure 6-18. Seismic protection by use of composite insulators design (Gilani et al., 2000) .................................... 213 Figure 6-19. Strengthening of existing transmission tower diagonal members (Xie and Zhange, 2020) .................. 214 Figure 6-20. Strengthening of existing transmission tower leg members (Kumalasari et al., 2005) ......................... 214 Figure 6-21. Buckling prevention of existing transmission tower leg members (Trovato et al., 2015) ..................... 215 Figure 6-22. Tuned mass damper for transmission tower (Zhang et al., 2013) ......................................................... 215 Figure 6-23. Stockbridge dampers for transmission line (various sources) ............................................................... 216 Figure 6-24. Viscoelastic dampers for transmission line (Lou et al., 2020) .............................................................. 216 Figure 6-25. Retrofit of transmission tower foundations (Yamane, 2020) ................................................................ 217 Figure 6-26. Retrofit effectiveness of substations (adapted from Romero et al., 2013) ............................................ 221 Figure 6-27. New transmission tower design (BPA, 2011) ....................................................................................... 221 Figure 7-1. Example of Bailey (steel truss) bridge .................................................................................................... 228 Figure 7-2. Application of temporary Bailey bridges in subject countries (various sources) .................................... 229 Figure 7-3. Collapse of Bailey bridges due to overload or poor construction ........................................................... 230 Figure 7-4. Pavement management systems flowchart (AASHTO, 2012) ................................................................ 234 Figure 7-5. Drone inspection of transportation infrastructure (Caltrans, 2020)......................................................... 235 Figure 7-6. Post-event damage assessment by drones (various) ................................................................................ 236 Figure 7-7. Monitoring of Ferguson Landslide (Caltrans, 2020) ............................................................................... 237 Figure 7-8. Laser scanning of bridge in the aftermath of 2010 earthquake (Takhirov and Mosalam, 2014) ............. 238 Figure 7-9. Instrumented bridges (Caltrans, 2020) .................................................................................................... 239 Figure 7-10. Hurricane Irma in the Caribbean (NOAA, 2020) .................................................................................. 240 Figure 7-11. Area of radar coverage (blue regions) and rain gauge locations (red dots) in Southern California (USGS, 2005) ............................................................................................................................................................ 240 Figure 7-12. Seismic strong motion stations in subject countries (UWI, 2020) ........................................................ 243 Figure 7-13. Forecast track for the 1998 Hurricane Georges (CIMH, 2020) ............................................................. 244 Figure 7-14. Recorded rain gauge data in Dominica during Hurricane Dean (Smith et al., 2008) ............................ 244 Figure 7-15. Schematics of flash flood guidance (HDRFFG, 2016) ......................................................................... 245 Figure 7-16. Post-earthquake damage assessment form (ATC, 2020) ....................................................................... 247 Figure 7-17. Post-hurricane and flood damage assessment form (ATC, 2020) ......................................................... 248 Figure 8-1. LCCA for example paved road (adapted from Batouli et al., 2017) ....................................................... 253 Figure 8-2. LCCA for example bridge (adapted from Hatami and Morcous, 2013) ................................................. 254 Figure 8-3. Mean seismic vulnerability function for buildings (adapted from Porter et al., 2004) ........................... 254 Figure 8-4. LCCA functions for energy sector (adapted from Loo et al., 2010) ....................................................... 255 Figure 9-1. Impact of Hurricane Dorian in the Bahamas (ESRI, 2019) ..................................................................... 259 Figure 9-2. Distribution of damage states .................................................................................................................. 264 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xxi REFERENCED ORGANIZATIONS AASHTO American Association of State Highway and www.transportation.org Transportation Officials ACP-EU European Union and the countries of the African, https://www.gfdrr.org/en/acp-eu Caribbean and Pacific Group of States ALA American Lifelines Alliance https://www.americanlifelinesalliance.com/ ASCE American Society of Civil Engineers https://www.asce.org ASTM American Society for Testing and Materials http://www.astm.org/ ATC Applied Technology Council https://www.atcouncil.org/ BPA Bonneville Power Administration https://www.bpa.gov/ Caltrans California Department of Transportation https://dot.ca.gov/ CCRIF Caribbean Catastrophe Risk Insurance Facility https://www.ccrif.org/ Segregated Portfolio Company CDMP Caribbean Disaster Mitigation Project https://www.oas.org/cdmp/ CEHI Caribbean Environmental Health Institute https://knowledge.unccd.int/kss/caribbean- environmental-health-institute-cehi CHARIM Caribbean Handbook on Risk Information http://www.charim.net/ Management CIMH Caribbean Institute for Meteorology and Hydrology www.cimh.edu.bb EPRI Electric Power Research Institute, Inc www.epri.com FDOT Florida Department of Transportation https://www.fdot.gov/ FEMA Federal Emergency Management Agency https://www.fema.gov/ FHWA Federal Highway Administration https://www.fhwa.dot.gov FIMA Federal Insurance and Mitigation Administration https://www.fema.gov/about/offices/insuran ce-mitigation HA Hong Kong Housing Authority https://www.housingauthority.gov.hk/en/ ICC International Code Council https://www.iccsafe.org/ IDB Inter-American Development Bank https://www.iadb.org/en IEEE Institute of Electrical and Electronics Engineers https://www.ieee.org/ INAPA Instituto Nacional de Aguas Potables y http://www.inapa.gob.do/ Alcantarillados IWR Institute for Water Resources https://www.iwr.usace.army.mil/ MISO Midcontinent Independent System Operator https://www.misoenergy.org/ MWH MWH Global Inc. http://www.mwhglobal.com/ NIBS National Institute of Building Sciences https://www.nibs.org NIST National Institute of Standards and Technology https://www.nist.gov/ NOAA National Oceanic and Atmospheric Administration https://www.noaa.gov/ OAS Organization of American States http://www.oas.org/ ODPM Office of Disaster Preparedness and Management http://www.odpm.gov.tt/ OECS Organisation of Eastern Caribbean States http://www.oecs.org/ OSSPA Oregon Seismic Safety Policy Advisory https://www.oregon.gov/gov/policy/orr/pag es/index.aspx PAHO Pan American Health Organization http://www.paho.org/ USACE The U.S. Army Corps of Engineers https://www.usace.army.mil USAID United States Agency for International https://www.usaid.gov/ Development USGS The United States Geological Survey https://www.usgs.gov WBG The World Bank Group https://www.worldbank.org/ WHO World Health Organization https://www.who.int/ Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xxii ACRONYMS, ABBREVIATIONS, AND NOTATIONS AAL Average annual loss AC Asbestos cement (concrete) ACSS Aluminum Conductor Steel Supported ADT Average daily traffic ATPB Asphalt Treated Permeable Base BRB Buckling restrained braces CBR California Bearing Ratio CGMC Construction Manager General Contractor CI Cast iron CUBiC Caribbean Uniform Building Code DBB Design bid build DBE Design Basis Earthquake DI Ductile iron DS Damage state FRP Fiber-Reinforced Polymer GS Galvanized steel HDPE High-Density Polyethylene HMA Hot mix asphalt IBC International Building Code IRI International Roughness Index LCCA Life-cycle cost analysis MGD Millions of gallons per day MPa Mega Pascal MTEP MISO Transmission Expansion Plan MVA Megavolt-ampere OOP Out of plane P3 Public-private partnership PASS Performance Assessment Scoring System PC Precast (prestressed) concrete PCC Portland Concrete Cement PGA Peak ground acceleration PGD Permanent ground displacement PGV Peak ground velocity PGWS Peak gust wind speed Psi Pound per square inch PVC Polyvinyl chloride RC Reinforced concrete RC/CF Reinforced/concrete frame with infill wall Sa Spectral acceleration Sd Spectral displacement TMD Tuned mass damper UGP Underground pipe URM Unreinforced masonry WS Welded steel WTP Water treatment plant WWTP Wastewater treatment plant Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | xxiii 1. INTRODUCTION 1.1 Overview Infrastructure comprises a critical component of community health and economic growth. In recent natural disasters, including earthquakes, hurricanes, and floods, significant damage to infrastructure occurred, resulting in economic losses and delayed recovery. The Caribbean region is undergoing growth and the tourist industry is one of the main engines of economic growth for these countries. The area is also vulnerable to natural hazards. Past earthquakes have had decades-long impact on countries within the region, and frequent hurricanes result in significant structural damage on an annual basis. To address such vulnerabilities and to improve the resilience of Caribbean communities, this project was undertaken to assess the current state of selected infrastructure and to provide recommendations for improvements in the subject counties listed in Table 1-1, with the objective of evaluating the costs and benefits associated with strengthening options. Area, Population, GDP, US$ No Country km2 millions billion 1 Bahamas 13,900 0.39 12.6 2 Barbados 439 0.28 5.4 3 Belize 23,000 0.41 3.5 4 Dominican Republic 48,670 10.7 216 5 Guyana 215,000 0.79 13.5 6 Haiti 27,800 11.1 20 7 Jamaica 10,900 2.9 27 8 Antigua and Barbuda 440 0.1 2.7 9 Dominica 750 0.07 0.7 10 Grenada 349 0.11 1.8 11 Saint Kitts and Nevis 261 0.05 1.8 12 Saint Lucia 617 0.18 2.7 13 Saint Vincent and the Grenadines 389 0.11 1.4 14 Sint Maarten 34 0.04 0.4 15 Suriname 163,800 0.58 9 16 Trinidad and Tobago 5,130 1.36 45.1 Table 1-1. General information for subject countries (various sources) 1.2 Scope and objectives The objective of the current project is to undertake a high-level conceptual, but detailed, analysis of engineering options for increased infrastructure resiliency in the Caribbean region, as well as an estimate of the cost of these options. The following are considered: • Infrastructure considered: o Buildings (focusing on critical public buildings, such as schools, hospitals, government buildings, emergency centers, and hospitality/hotels) o Water (water and wastewater treatment plants, and pipelines) o Transport (roadway bridges and primary paved roads) o Power (power plants, substations, and distribution systems) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |1 • Natural hazards considered: o Floods o Hurricanes (wind) o Earthquakes (both shaking and liquefaction) Throughout the remainder of this report, the natural hazards above may be defined as follows: • Floods: When specifying this hazard, may be noted as “flood” or “flooding”, and including river, storm surge, etc. • Earthquakes: When specifying this hazard, may be noted as “ground shaking” and/or “liquefaction.” For ground shaking, may be noted as earthquake/earthquake ground shaking/ground shaking/shaking. For liquefaction, may be noted as earthquake liquefaction/liquefaction. As much as possible, ground shaking and liquefaction will be distinguished. • Hurricanes: When specifying this hazard, may be noted as “wind.” 1.3 Limitations As part of this study, significant outreach in the Caribbean region was undertaken to collect specific data relevant to the subject countries. When available, this data is used in the report. When such data was not readily available, engineering judgement was used and augmented by the knowledge of construction and design practices in the Caribbean. It is further assumed that new construction and most infrastructure constructed after the year 2000 would have been designed per provisions of modern codes, and thus, are likely more robust than earlier construction. Accordingly, new construction is not considered in this report, and the recommendations are designed for retrofitting of existing infrastructure. The focus of the work will be on structural aspects of resilient development. As such, physical damage and the costs associated with such damage will be solely considered. Other factors, such as casualties, repair time, and secondary or user costs, are not included. 1.4 Subject countries and vulnerable infrastructure As noted earlier, there are 16 subject countries, several hazards, and a number of infrastructure sectors that need to be considered. To streamline the process, the 16 countries were placed in four groups; see Table 1- 2 and Figure 1-1. The grouping is based on factors like geographical location, governing natural hazards, and relative size or population of the countries. For each group, the critical natural hazards were identified, and subsectors specified, as seen in Table 1-2. Group Country Hazard Buildings Water Transport Power Suriname Government, Trinidad and Tobago Emergency, Water 1 Earthquake Bridges Substations Schools, treatment Guyana Hospitals, Hotels plants (WTP) Belize Government, Haiti Earthquake Emergency, Wastewater 2 Bridges Substations Wind Schools, treatment Dominican Republic Hospitals, Hotels plants Antigua and Barbuda (WWTP) Dominica Wind Schools, Transmission 3 Underground Roads Grenada Flood Hospitals, Hotels lines pipes (UGP) Saint Kitts and Nevis Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |2 Saint Lucia Saint Vincent and the Grenadines Sint Maarten Barbados Jamaica Wind Schools, Transmission 4 Roads Bahamas Flood Hospitals, Hotels lines Table 1-2. Vulnerable infrastructure and associated natural hazards in subject countries (WB, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |3 Figure 1-1. Grouping of countries (Kmusser, 2020) LEGEND → = Group 1 → = Group 2 → = Group 3 → = Group 4 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |4 1.5 Evaluation methodology For each class of substructure, the sector infrastructure types most vulnerable to the critical natural hazards in that region were identified. Since the infrastructure design and construction practice are similar in the subject countries, representative states were used as examples. However, the discussion extends to other countries, and when available, data from other countries are presented. For example, for countries in Groups 1 and 2, the emphasis for transportation infrastructure will be on elevated bridges and for Groups 3 and 4, on roads. The hazard and infrastructure data in Table 1-2 are consistent with experience from previous events, which has shown that bridges are vulnerable to earthquakes (ground shaking and liquefaction), whereas flooding—including floods as a result of hurricane storm surge—severely impacts roadways. The assessment is based on structural types and components (not hazard-, country-, or group-based) as discussed in subsequent chapters of this report. 1.6 Organization of the report Chapter 1 includes an introduction to the study. Chapter 2 provides information on natural hazards. Chapters 3, 4, 5, and 6 discuss critical buildings, water, transport, and power sectors, respectively. In these chapters, general typology, code provisions, vulnerable components, strengthening techniques, and the associated costs are discussed. Chapter 7 discusses additional measures that can be utilized for improved resiliency, including improved quality of construction and regular preventive maintenance. Chapter 8 presents a brief discussion on cost and financing and life-cycle cost analysis. Conclusions and recommendations are presented in Chapter 9 and a list of references in the final chapter. APPENDIX A summarizes a list of key reference files used in this report. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |5 2. NATURAL HAZARDS 2.1 Introduction In this chapter, the key natural hazards impacting the subject countries are discussed. Given that the countries are either island nations or continental states with large coastlines, and located in hurricane-prone zones, hurricanes and resulting flooding are the most common natural hazards. Table 2-1 (ThinkHazard, 2020) presents the relative importance of various hazards for the subject countries. Although earthquakes are less frequent, on an annualized basis, they can be more consequential. As such, earthquakes (both ground shaking and liquefaction) are also considered. Earthquakes impact a number of countries in the Caribbean; see Table 2-2 (CHARIM, 2020). Country Earthquake Hurricane Flood (Coastal and river) Suriname Low Low High Trinidad and Tobago Medium High High Guyana Low Low High Belize Medium High High Haiti Medium High High Dominican Republic Medium High High Antigua and Barbuda Medium No data Medium Dominica Medium No data High Grenada Medium No data Medium Saint Kitts and Nevis Medium No data Medium Saint Lucia Medium No data Medium Saint Vincent and the Grenadines Medium No data Medium Sint Maarten No data No data No data Barbados Medium No data Medium Jamaica Medium High High Bahamas Medium High High Table 2-1. Relative importance of various hazards in different countries (ThinkHazard, 2020) Country Earthquake Wind Storm Surge Flood Belize 3.2% 30% 16.4% 50.4% Dominica 19.1% 34.8% 46.2% -- Grenada 29% 34.1% 36.9% -- Saint Lucia 10.8% 43.7% 45.5% -- Saint Vincent and the Grenadines 11.4% 26.9% 61.7% -- Table 2-2. Hazard contribution to AAL3 for selected countries (adapted from CHARIM, 2020) 2.2 Natural Hazards 2.2.1 Earthquake hazard 2.2.1.1 Ground shaking Figure 2-1 presents the seismic hazard map for selected subject countries (CDMP, 2020). The data is shown for peak ground acceleration (PGA) of a 475-year (20% probability of occurrence every 50 years) earthquake. This level of seismicity is typically used in construction of new infrastructure and is expected to provide life preservation (ordinary structures) or continuous operation (critical infrastructure). The 3 Average Annual Loss Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |6 acceleration contours presented in the figure correspond to bedrock motions and need to be adjusted for the site to account for soil amplification. The PGA values typically vary from 0.1 to 0.4 g in the figure, corresponding to low-to-moderate seismicity. 2.2.1.2 Liquefaction Figure 2-2 (Kraft, 2013) presents the liquefaction susceptibility map for Jamaica and Trinidad. Because many of the subject countries have low-lying areas with high ground water elevation and sandy soil, the coastal areas could experience liquefaction in large earthquakes. However, this is considered a secondary hazard compared to ground shaking. 2.2.2 Wind hazard The Pan American Health Organization (PAHO) presents the developed wind speed maps (2019) for the Caribbean region for 50- to 1700-year events; see Figure 2-3. Table 2-3 lists the wind speed associated with various categories of hurricanes. Table 2-4 presents the wind speed for various subject countries depending on different return intervals. The listed wind speeds can be used with American Society of Civil Engineers (ASCE) procedures for the design of new structures. Wind speed, Category miles/hour (mph) 1 74-95 2 96-110 3 111-129 4 130-156 5 157+ Table 2-3. Hurricane category and sustained wind speed (National Hurricane Center, 2020) Return period, years Country 50 100 700 1700 Suriname -- -- -- -- Trinidad and Tobago 61 85 136 156 Guyana -- -- -- -- Belize 106 128 165 177 Haiti -- -- 175 185 Dominican Republic -- - 170 185 Antigua and Barbuda 121 134 160 168 Dominica 106 124 159 171 Grenada 85 107 165 168 Saint Kitts and Nevis 125 138 163 170 Saint Lucia 101 119 155 172 Saint Vincent and the Grenadines 93 111 155 171 Sint Maarten 129 141 168 178 Barbados 92 112 152 169 Jamaica -- -- 150 170 Bahamas 121 132 163 180 Table 2-4. Peak gust wind (mph) for subject countries (adapted from PAHO, 2019) 2.2.3 Flood hazard Figure 2-4 presents the flood hazard maps for some of the Caribbean countries. The countries, closely located to the sea and with many rivers, are susceptible to storm surge, flash, and river flooding. Additionally, if proper drainage is not provided, urban flooding could occur. In this report, all sources of flooding are treated as having a similar impact. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |7 Haiti and Dominican Republic Belize Bahamas Jamaica Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |8 Guyana Trinidad and Tobago Leeward Islands Windward Islands Figure 2-1. Seismic hazard maps (CDMP, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean |9 Jamaica Trinidad Figure 2-2. Liquefaction hazard maps (Kraft, 2013) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 10 50 year 100 year Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 11 700 year Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 12 1700 year Figure 2-3. Wind hazard maps (PAHO, 2019) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 13 Jamaica (Nandi et al., 2016) Trinidad and Tobago (ODPM, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 14 Belize (CHARIM, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 15 Dominica (flashflood) (CHARIM, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 16 Saint Lucia (flashflood) (CHARIM, 2020) Figure 2-4. Flood hazard maps Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 17 2.3 Infrastructure impact In the 2020 Atlantic hurricane season alone, there were a number of hurricanes and tropical storms that caused significant damage in the Caribbean. See Figure 2-5 for an example. • Hurricane Isaias impacted the Leeward Islands, Haiti, the Dominican Republic, and Bahamas. In the Bahamas, electricity was lost and there was damage to houses and roads. • Hurricane Laura had an impact on the Dominican Republic, Haiti, and Jamaica. Schools were closed in Antigua. • Hurricane Marco caused significant flooding in the Caribbean, including in Haiti and the Dominican Republic. • Hurricane Nana caused significant flooding in the Caribbean, including in Belize. Inundated road (Bahamas: Hurricane Isaias) Damaged buildings (Haiti: Hurricane Laura) Flooding (Dominican Republic: Hurricane Marco) Flooded bridge (Belize: Hurricane Nana) Figure 2-5. Damage to infrastructure in the Caribbean, 2020 Hurricane season (various sources) For the various infrastructure types considered in this group, some of the critical hazards are considered less “critical” and thus are de-emphasized in this report. Shaded entries in the infrastructure-hazard matrix of Table 2-5 are the hazards that are considered most critical. Sector Subsector Earthquake4 Wind Flood Buildings Public buildings Water treatment plant Water Wastewater treatment plant Underground pipes 4 Ground shaking and/or liquefaction Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 18 Paved Roads Transport Bridges Power plants Substations Power Transmission towers and lines Table 2-5. Matrix of considered hazards for various infrastructure 2.4 Selection of governing hazard In this report, a number of hazards and infrastructure types are considered for the 16 subject countries. To streamline the assessment process, for each infrastructure subsector, several structural (construction) types were selected based on the expected typologies in the subject countries. Next, for each structural type, several components to consider along with the governing hazards were chosen; see Table 2-6. These components, their vulnerabilities, and the steps to improve resiliency are discussed in subsequent chapters. Components to Sector Subsector Type5 Governing Hazard(s) Consider Wall Earthquake URM Roof Wind Shape and figure Earthquake Wall Earthquake RC/CF with infill wall Roof Wind Opening Flood Shape and figure Earthquake Schools, Hospitals. RC frame Ductility Earthquake Government External wall Wind buildings, Stiffness Earthquake, Wind Emergency centers, Steel frame Strength Earthquake Hospitality and BUILDINGS Hotels Roof Wind Stiffness Earthquake, Wind Steel frame high-rise Ductility Earthquake Cladding Wind Roof Wind Wood frame Foundation Wind, Flood Elevation Flood Foundation Liquefaction Wall Earthquake Ductility Earthquake Plant building Roof Wind (Masonry/RC frame) External wall Wind Water treatment Opening Flood plants Elevation Flood Foundation Earthquake Wastewater Water storage Anchorage Earthquake treatment plants WATER Equipment Foundation Earthquake, Wind (mechanical and Elevation Flood electrical) Floodproofing Flood Foundation Liquefaction 5 RC = Reinforced concrete; URM = Unreinforced masonry; RC/CF = Reinforced Concrete/Concrete Frame Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 19 Components to Sector Subsector Type5 Governing Hazard(s) Consider Earthquake, liquefaction, Pipe Brittle pipes Underground pipes flood Joints Conventional Earthquake, liquefaction Subsurface Flood Paved roads Asphalt paved Surface Flood Columns Earthquake Concrete girder Joints Earthquake Members Earthquake Steel truss Connections Earthquake Bearings Earthquake Bridges Steel girder Cross frames Earthquake TRANSPORT Welded Wind Steel bridges connections Suspension Bridge stability Wind Foundation Earthquake, Flood Design Earthquake Grade Flood construction Power plants Low-rise RC Nonstructural Earthquake components Foundation Earthquake, liquefaction Design Earthquake construction Grade URM Control Flood construction buildings Battery racks, Earthquake Substations cable trays, etc. POWER Foundation Earthquake, liquefaction Transformers Anchorage Earthquake Frames Steel members Earthquake, wind High-voltage Equipment Earthquake units Tower Wind, Earthquake Transmission & Lattice towers Foundation Flood, Liquefaction Distribution lines Lines Wind Table 2-6. Matrix of considered typologies for various infrastructure 2.5 Discussion The selected countries in the Caribbean are susceptible to earthquakes, windstorms, and flooding. Many of the countries experience frequent hurricane and storm surge impacts. Although less frequent, earthquakes have devastated a number of the countries in the region with long-term adverse effects. In this report, classes and subclasses of critical infrastructure will be studied. The focus will be on the select structural types that are more typical in the region, and components of these structural types that are the most vulnerable. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 20 3. BUILDING INFRASTRUCTURE 3.1 Introduction For building infrastructure in the Caribbean, the typical structural types, governing hazards, and vulnerable components are summarized in Table 3-1. Based on damages to buildings seen in past natural disasters, heavier buildings are generally more at risk in earthquakes and lighter buildings are usually more vulnerable to hurricane winds and flood events. The vulnerable components to consider are then identified for each governing hazard according to the representative structural types, which are discussed in the following sections in further detail. Since the structural type is the main factor to scale the disaster vulnerability of a building, the discussions, especially on strengthening techniques, focus on the structural components of buildings. Sector Subsector Type Components Hazard Unreinforced masonry Wall Earthquake (URM) Roof Wind Shape and figure Earthquake Reinforced/concrete frame Wall Earthquake (RC/CF) with infill wall Roof Wind Opening Flood Schools, Shape and figure Earthquake Hospitals, RC frame Ductility Earthquake Government External wall Wind buildings, Stiffness Earthquake, Wind Emergency Steel frame Strength Earthquake centers, Roof Wind BUILDINGS Hospitality and Stiffness Earthquake, Wind hotels Steel frame high-rise Ductility Earthquake Cladding Wind Roof Wind Wood frame Foundation Wind, Flood Elevation Flood All Foundation Liquefaction Table 3-1. Analysis matrix for building infrastructure 3.2 Geographical distribution The exposure information for critical buildings (i.e., schools, hospitals, government buildings, emergency buildings and hotels in this study) in selected Caribbean countries is presented in Figure 3-1. The collected information covers mostly site location for the selected countries, as the location of other critical buildings in these countries was not available. This type of geographical information is very useful for both disaster damage assessments performed on hazard maps and post-disaster response planning using regional transportation networks. For example, in Belize, the locations of four types of critical buildings are shown and the multiple access routes between these facilities (which is important for quick response and smooth recovery). These routes might be arranged beforehand by considering the spatial distribution of expected damage to buildings and roads due to a disaster. In addition, a complete package of exposure data of buildings is helpful to prioritize buildings in terms of disaster strengthening and financial planning. Preferably, information such as building occupancy, area, stories, materials, year built, occupants and so on, in addition to building location, should be included in the exposure data package. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 21 Belize Dominica Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 22 Grenada Saint Lucia Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 23 Suriname Figure 3-1. Critical building locations for selected Caribbean countries (Gosine, 2020a) 3.3 Typology The buildings in focus in this study, schools, hospitals, government buildings, emergency buildings and hotels, are classified according to six structural types, as shown in Figure 3-2. The structural type usually refers to the building material, and brick/block masonry, concrete, steel and wood are used most typically for the buildings of those occupancies in the Caribbean. More specifically, unreinforced masonry (URM), reinforced/concrete frame (RC/CF) with infill wall, reinforced concrete (RC) frame, steel frame and wood frame are utilized for those buildings. Across these types, URM is more particularly vulnerable than others, as the disaster damage to this type of building is enormous in many countries. The structural types of buildings listed above are emphasized in this study. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 24 URM RC/CF with infill wall RC frame Steel frame Steel frame high-rise Wood frame Figure 3-2. Building typology (various sources) 3.4 Vulnerable components Per past disaster damage experiences, some vulnerable components have been observed in typical and recurrent failure modes, depending on structural type and hazard event. Those vulnerable components should be strengthened first to avoid severe damage and to make the buildings more resilient in future Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 25 natural disasters. Those components and damage failures are focused on in this study. Typical failure modes due to earthquake shaking and liquefaction are shown in Figure 3-3 and representative damage due to hurricane winds and associated floods are presented in Figure 3-4. URM Out-of-Plane and In-Plane wall damage due to Soft/weak story collapse due to structural irregularities shaking (Chin, 2012) (Gibbs, 2012) In-plane wall damage due to shaking (Clarke, 2018) Column failure due to non-ductile detailing (Chin, 2012) Joint failure due to non-ductile detailing (Rojas- Ground failure due to liquefaction (Bakir, 2016) Mercedes et al., 2020) Figure 3-3. Building damages due to earthquake shaking and liquefaction The main element to resisting seismic force is typically a shear wall for URM or RC buildings. If the seismic capacity of a wall for in-plane or out-of-plane (OOP) is not sufficient, the building could be extensively damaged, and the failure mode could be very brittle (sudden). For avoiding brittle failure mode, modern design codes generally specify detailing rules (i.e., reinforcement restrictions, steel section limit, etc.) to make the elements more ductile. However, this kind of ductile detailing might not have been applied to the buildings, which were built without modern seismic design or code compliance in design or construction. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 26 In that case, the building could suffer brittle damage without enough deformability. The structural shape of a building also affects the building’s movement and response in an earthquake. If structural irregularities are observed, the building would suffer severe damage due to unbalanced load concentration or large deformation. For example, soft/weak stories, due to vertical discontinuity of wall or torsional irregularity from uneven horizontal location of wall, have caused critical damage to buildings in past earthquakes. Earthquake liquefaction causes large ground failure and the building typically loses foundational stability, as shown in the last picture of Figure 3-3. Roof and wall connections are vulnerable components to wind hazard. Wind forces are imposed on buildings both horizontally and vertically. If the connection capacity is deficient for those forces, either or both of roof and wall elements could be blown away, and the building might suffer severe damage due to a lack of bearing elements, like a wall. Steel and wood buildings are relatively flexible structures because of the smaller sizes of columns and beams. This type of building is generally vulnerable to wind hazard; an appropriate stiffness should be added to resist strong wind forces. For mitigating flood damage (i.e., water inundation), elevating the building and/or protecting the building envelope from water inundation could represent possible improvements. Roof and wall damage due to wind (Clarke, 2018) Steel building damage due to wind (Gibbs, 2001) Flooding in city (APESL, 2010) Flooding in street (APESL, 2010) Figure 3-4. Building damages due to hurricane winds and associated floods The components listed in Table 3-2 caused severe damages to the building types discussed in this chapter in past events. These components are then assumed to be the major factors of damage and will be considered in this chapter. Subsector Type Component Vulnerability Hazard Schools, Hospitals, Out-of-plane capacity Earthquake Wall Government URM In-plane capacity Earthquake buildings, Roof Roof connection Wind Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 27 Emergency centers, Shape and figure Structural irregularity Earthquake Hospitality and hotels RC/CF with Wall Infill wall connection Earthquake infill wall Roof Roof connection Wind Opening Building envelope Flood Shape and figure Structural irregularity Earthquake RC frame Ductility Non-ductile detailing Earthquake External wall External wall Wind Stiffness Stiffness Earthquake, Wind Steel frame Strength Structural capacity Earthquake Roof Roof connection Wind Stiffness Stiffness Earthquake, Wind Steel frame Ductility Non-ductile detailing Earthquake high-rise Cladding External curtain wall Wind Roof Roof connection Wind Wood frame Foundation Foundation ties Wind, Flood Elevation Elevated construction Flood All Foundation Lack of capacity Liquefaction Table 3-2. Vulnerable components of building infrastructure and corresponding hazards 3.5 Design codes The Caribbean Uniform Building Code (CUBiC) was developed within and by the Caribbean community in 1985 and it includes structural design requirements for gravity load, wind load and seismic load (CARICOM, 1985). Since the degree of enforcement of this building code varies according to country, some countries of the Caribbean have adopted this code as part of the design process, but some treat this as a supplemental document, as shown in Figure 3-5 (Redmond, 2015). Some countries in the Caribbean have developed their own building code and use it as well as CUBiC, the International Building Code (IBC) and other modern design codes. In certain cases, foreign design consultants seem to apply their familiar design codes to structural design. On the other hand, a code-compliance process, like plan check, construction inspection and building permits, appears to not be guaranteed in many Caribbean countries, because of, for example, a lack of skilled inspectors, weak/inadequate enforcement mechanisms or less motivation for compliance. In this section, CUBiC and additional codes are investigated. Figure 3-5. Building design codes of representative countries in the Caribbean (Redmond, 2015) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 28 3.5.1 Seismic design 3.5.1.1 Seismic design in the Caribbean In CUBiC Part 2, Section 3, the seismic design requirements are specified. Specific performance objectives for seismic design are not explicitly given, but the main objective provided is to protect human life and to reduce damage caused by earthquakes (Chin, 2008a). This seismic design approach observes a method controlled by displacements, inelastic deformations are accepted, and adequate ductility is required. This code has not been updated in accordance with engineering research accomplishments or earthquake experiences since it was published in 1985. For example, the current seismic hazard level should be updated from 1985 as it directly affects the seismic design force and, subsequently, building performance; however, this update has not occurred. Major design components for earthquakes specified in CUBiC are described as follows (Chin, 2008a; CUBiC, 1985). 3.5.1.1.1 Design seismic force In the code, no seismic design response spectrum is specified nor any levels of seismic intensity (e.g., return period, effective PGA or seismicity) clarified. It appears that zonal coefficient, Z, is applied to each Caribbean country to calculate the seismic design force. The following formula is the equivalent seismic force for design in the equivalent static method of CUBiC: Eq. 3-1 = • : The total lateral force of shear at the base • : Numerical coefficient related to the seismicity of a region (see Table 3-3) • : Numerical coefficient, = 1⁄15√ (C need not exceed 0.12) • : Fundamental elastic period of vibration of the structure in seconds • K: Numerical coefficient regarding ductility (see Table 3-4 and Table 3-5) • : Occupancy importance coefficient (see Table 3-6) • : Numerical coefficient for site-structure resonance • ⁄ ≤ 1.0; = 1.0 + ⁄ − 0.5(⁄)2 • ⁄ > 1.0; = 1.2 + 0.6(⁄) − 0.3(⁄)2 • : Characteristic site period in seconds • : The total dead load and applicable portions of other loads Country Z value Jamaica 0.75 Antigua 0.75 Saint Kitts and Nevis 0.75 Montserrat 0.75 Dominica 0.75 Saint Lucia 0.75 Saint Vincent and the Grenadines 0.50 Grenada 0.50 Barbados 0.375 NW Trinidad 0.75 Rest of Trinidad 0.50 Tobago 0.50 Guyana Essequibo 0.25 Rest of Guyana 0.00 Belize - areas within 100 km of southern border, i.e., including San Antonio and 0.75 Punta Gorda, but excluding Middlesex, Pomona and Stann Crecil Remainder of Belize 0.50 Table 3-3. Z values for various Caribbean countries (CUBiC, 1985) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 29 Item Description K factor 1 Ductile frames with an adequate number of possible plastic beam hinges 0.8 2 Ductile frames with an inadequate number of possible plastic beam hinges 1.0 3 Ductile coupled shear walls 0.8 Two or more parallel and approximately symmetrically arranged cantilever shear 4 1.0 walls 5 Single ductile cantilever shear walls 1.2 Shear walls not designed for ductile flexural yielding but having the ability to 6 1.6 dissipate a significant amount of seismic energy 7 Buildings with diagonal bracing capable of plastic deformation in tension only a) Single story 2.0 b) Two or three stories 2.5 or by special study c) More than three stories by special study a) Buildings in which part of the horizontal load is resisted by item 7 bracing and 8 1.6 or by special study in part by an item 1 or item 2 frame b) Buildings with diagonal bracing capable of plastic deformation in both tension 1.6 or by special study and compression 9 Small tanks on the ground 2.0 Table 3-4. K factors for various structural types, steel and concrete (CUBiC, 1985) Item Description K factor B1 Shear walls or diaphragm (a) Ductile 1.0 (b) Ductile and stiffened with elastomeric adhesive 1.0 (c) Limited ductility fixed with elastomeric adhesive 1.2 B2 Moment resisting frames (a) Ductile with an adequate number of possible plastic beam 1.2 hinges (b) Ductile with an inadequate number of possible plastic 1.5 beam hinges (c) As for item B2(a) but with connections of limited ductility 1.5 (d) As for item B2(b) but with connections of limited ductility 1.7 (e) Non-ductile 2.4 Diagonally braced with timber members capable of acting as B3 struts or ties: (a) With ductile end connections 1.7 (b) With end connections having limited ductility 2.0 Table 3-5. K factors for various structural types, timber (CUBiC, 1985) Building Description I class These are essential facilities required for use in the aftermath of a major earthquake, e.g. I 1.5 hospitals, fire stations, communication centers etc. These are public buildings and buildings which accommodate large numbers of' people, e.g. II 1.25 cinemas, theatres, schools, defense establishments etc. III All other buildings not included in Class I or Class II above. 1.0 Table 3-6. I factors for various structural types, steel and concrete (CUBiC, 1985) 3.5.1.1.2 Structural type Nine structural types for concrete and steel buildings and three structural types for timber buildings are specified, and design seismic force is adjusted (reduced or increased) by factor K according to this type due to ductile performance. Frame Type: Ductile steel, concrete, timber. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 30 Dual Type: Frame + Wall combination. Frame with 25% of shear demand; steel, concrete, masonry, timber. Wall Type: Either concrete, masonry or plywood walls or steel or timber braced frames. Cantilever Type: Inverted Pendulum. Others Type: None of the above. (CUBiC, 1985) 3.5.1.1.3 Structural irregularity The code requires a dynamic analysis method if the building has large irregularities in plan and/or elevation. In particular, buildings classified in Importance Groups I and II shall be analyzed by dynamic methods when: (a) The seismic force resisting system does not have the same configuration in all stories and in all floors. (b) The floor masses differ by more than 30% in adjacent floors. (c) The cross-sectional areas and moments of inertia of structural members differ by more than 30% in adjacent stories. (CUBiC, 1985) 3.5.1.1.4 Drift limitation The lateral drift due to a design seismic force is limited as follow: Lateral deflections or drift of a story relative to its adjacent stories shall not exceed 0.005 times the story height unless it can be demonstrated that greater drift can be tolerated. The displacement calculated from the application of the required lateral forces shall be multiplied by 1.0/K to obtain the drift. The ratio 1.0/K shall not be less than 1.0. (CUBiC, 1985) 3.5.1.2 Seismic design in Trinidad and Tobago TTS 599: Guide to the Design and Construction of Small Buildings was developed in Trinidad and Tobago in 1997 because more than 80% of buildings in Trinidad and Tobago are small buildings, like low-rise building (Thomson, 2018). This document was established by referring to Part 5 of CUBiC “Small Buildings and Pre-fabricated Construction,” and International Design Codes, and therefore the basic concepts of seismic design, such as performance objectives or design approach, are basically the same as that of CUBiC. However, the document has not been formally adopted by approval agencies, even though its latest edition was published in 2006. This lack of official acceptance means a formal and accepted national building code does not exist in Trinidad and Tobago (Thomson, 2018). Based on technical communications with local engineers and researchers, it was determined that the buildings in Trinidad and Tobago are generally designed for seismic forces based on empirical engineering approach or major international building codes. The main components of this design guideline are as follows (Chin, 2008b). 3.5.1.2.1 Design seismic force A simplified formula for seismic design force is presented in this guideline. The following formula is based on the equivalent seismic force for design in the equivalent static method of CUBiC. Eq. 3-2 = 0.05 × × • : The total lateral force op shear at the base • 0.05: Integrated value of Z = ground acceleration (Z value), C = amplification factor due to structure frequency, I = Importance factor =1 in this guideline and K = Ductility factor • : Site factor S = 1 for good soil (rock, gravel), S = 1.2 for softer material (clay, fill), S = 1.5 for deep alluvial deposits, S = 2.5 maximum for reclaimed land and saturated soils (due to the amplification factor) • : The total dead load and applicable portions of other loads Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 31 3.5.1.2.2 Earthquake resistant construction Not only shaking, but also liquefaction is considered as an earthquake influence in this guideline. General: Trinidad and Tobago is in an earthquake zone and has experienced varying degrees of damage due to earthquakes. It is therefore essential that buildings are designed and constructed so that they have some resistance to the shaking or lateral forces produced by earthquakes. Effect of soil: The type of soil at the site may have a significant effect upon the resistance of the building to an earthquake. However for buildings within the scope of this code the effect of the soil type is not so significant provided that the building is not constructed on loose saturated sands, which may liquefy during an earthquake and cause collapse of the building. 3.5.1.2.3 Lateral load design The basic approach regarding seismic force transfer is described, and a load path concept from diaphragm- to-foundation through collector and vertical element (e.g., wall, brace, etc.) is clarified. Diaphragm: Floor, roof or ceiling assemblies may be constructed with the necessary stiffness and load path continuity to distribute lateral loads (wind and earthquake) to lateral support subsystems. In this role, floor, roof or ceiling surface act as horizontal beams (also called a diaphragm) spanning lateral supports points. Use of floor, roof or ceiling assembly, as a diaphragm requires both strength and stiffness properties and development of connections to transfer the diaphragm force. Shear panel: Concrete wall (A shear panel is a portion or section of a 150mm exterior wall that performs the function of resisting lateral earthquake or wind forces and Timber wall bracing. 3.5.1.2.4 Load factors Load combinations for gravity and earthquake design are specified in Table 3-7. Combination Dead Live Earthquake Gravity 1.4 1.7 0 Earthquake (a) 1.05 1.275 +/- 1.4025 Earthquake (b) 0.9 0 +/- 1.43 Table 3-7. I Load combinations for design (Chin, 2008b) 3.5.2 Wind design 3.5.2.1 Wind design in the Caribbean Sections of CUBiC Part 2 prescribe the structural design requirements of buildings for wind force. As with the seismic design component of CUBiC, explicit performance objectives are not expressed, but the main objectives provided are to protect the life of occupants and to reduce any damage due to wind load (Suite, 2008). Since the wind design component of CUBiC has not been updated since 1985, the design wind pressure is quite different from current modern codes like IBC or ASCE 7 (Gibbs, 2012). If the basic wind speed to calculate the design wind pressure is changed, the demand for building wind design will be directly affected. Primary components of wind design in CUBiC are shown as follows (Gibbs, 1998) (Suite, 2008). • Concept and limitations This document describes the action of wind on structures and methods for calculating characteristic values of wind loads for use in designing buildings, towers, chimneys, bridges and other structures as well as their components and appendages. These loads will be suitable for use in conjunction with other ISO Load Standards and with ISO 2394 – General Principles on Reliability for Structures. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 32 3.5.2.1.1 Design wind pressure (simplified procedure) The formula below is prepared for calculating design wind pressure based on wind velocity and several factors. The height of wind is based on 15 m above ground, the average period is 10 minutes mean velocity pressure and the return period of wind velocity is 50 years. The following formula is for the simplified procedure. Eq. 3-3 = ( )( )(ℎ )( ) • : Wind force per unit area normal to the surface of the structure • : Reference velocity pressure (see Table 3-8 & Table 3-9 and Figure 3-6) • = (1⁄2) 2 , is air density, is wind velocity • : Exposure factor • ℎ : Aerodynamic shape factor • : Dynamic response factor Wind pressure (kPa) Wind speed (m/sec) Location qref (q50) q10 q100 Vref Guyana 0.20 0.05 0.35 18.0 Trinidad-South 0.25 0.05 0.40 20.0 Trinidad-North 0.40 0.10 0.60 25.5 Tobago 0.47 0.15 0.65 28.0 Grenada 0.60 0.25 0.80 31.5 Barbados 0.70 0.30 0.90 34.2 Saint Vincent and the Grenadines 0.73 0.35 0.93 35.0 Saint Lucia 0.76 0.36 0.95 35.5 Dominica 0.85 0.42 1.06 37.5 Montserrat 0.83 0.40 1.07 37.2 Antigua 0.82 0.39 1.05 37.0 Saint Kitts and Nevis 0.83 0.38 1.07 37.2 Jamaica 0.80 0.40 1.00 36.5 Belize-North 0.78 0.38 1.00 36.0 Belize-South 0.55 0.26 0.70 30.5 Table 3-8. Reference wind velocity pressures for the Caribbean (CUBiC, 1985) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 33 Figure 3-6. Regional map of design wind pressure contours based on CUBiC (Gibbs, 1998) V peak (m/sec) – 10 meters qref (kPa) Average time 1 min. (or “fastest 10 min. 10 min. 1 hr. 3 sec. mile”) 0.30 22.4 21 27 33 0.40 25.8 25 31 39 0.50 28.9 27 35 43 0.60 31.6 30 38 47 0.70 34.2 32 41 51 0.80 36.5 35 44 55 0.90 38.7 37 47 58 1.00 40.8 39 50 61 1.10 42.8 41 52 64 1.20 44.7 43 54 67 1.30 46.5 44 56 70 1.40 48.3 46 58 73 1.50 50.0 48 61 75 Table 3-9. Relationship between reference velocity pressure qref and peak wind speeds over short time intervals in open terrain (CUBiC, 1985) 3.5.2.1.2 Wind design actions Four major actions due to wind force are specified for consideration in the wind design of structures. (a) Excessive force or instability. (b) Excessive deflection. (c) Repeated dynamic forces causing fatigue. (d) Aeroelastic instability. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 34 3.5.2.1.3 Method of analysis Three methods of analysis are prepared in CUBiC: simplified procedure, detailed procedure and experimental procedure. Simplified procedure: This method is intended for the design of normal structures and can be used for the wind design of the main structural system if (a) the structure is less than 15 m in height above ground, (b) the structure is not usually exposed for any wind direction, that is, it is not situated near a hill crest or head land, (c) the structure is relatively rigid. Deflection under wind loading less than 1/500.H where H = height of structure. Detailed procedure: This method is principally of assistance in assessing the dynamic response of the structure, the influence of unusual exposure and the characteristics of more complex aerodynamic shapes. Structures sensitive to wind include those that are particularly flexible, slender, lightweight or tall. Experimental procedure: This is recommended when assessing the influence of unusual exposure and the characteristics of more complex aerodynamic shapes as well as structures sensitive to wind including those that are particularly flexible, slender, lightweight or tall and structures of unusual geometry, which give rise to unexpectedly large responses to wind. It may be necessary to conduct supplementary studies by experts in the field. These tests may include wind tunnel tests. 3.5.2.2 Wind design in Dominica Based on a (draft) physical planning building code document, the wind design approach and design wind pressure for buildings in Dominica refer to CUBiC (Dominica, 1996) and therefore, it is anticipated that the current components for wind design for the Dominica building code also conform to CUBiC. A guideline for housing standards for hurricane hazards in Dominica was developed as a response to the enormous damage due to Hurricane Maria in 2017 (Dominica, 2018). This guideline, based on a specification-based approach, includes typical wind pressures on housing (see Figure 3-7), general failure modes due to wind and flood, and recommended design procedures for building shape, foundations, walls, connections, roofs, retrofit and maintenance. However, any specific wind load for housing design are not included. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 35 Figure 3-7. Typical wind pressure on housing (Dominica, 2018) 3.5.3 Flood design 3.5.3.1 Flood design by ASCE 7-16 It is assumed that the subject countries use flood design methodologies commonly used throughout the world (but taking into consideration each country’s circumstances), as no flood design codes for buildings were found. In this section, therefore, the design approach for floods prescribed in ASCE 7-16 (ASCE, 2017a) is briefly introduced, as the International Building Code (ICC 2017 – widely used in the Caribbean) and ASCE 24-14 (ASCE 2015 – document focusing on flood design) refer to ASCE 7-16. The base flood specified in ASCE 7-16 for building design is the flood having a 1% chance of being equaled or exceeded in any given year. So, the flood used for design needs to be compared to this base flood with the site-specific flood hazard (ASCE, 2017a). The major terms for flood design are listed below. Base flood: The flood having a 1% chance of being equaled or exceeded in any given year. Design flood: The greater of the following two flood events: (1) the base flood, affecting those areas identified as special flood hazard areas on the community’s FIRM; or (2) the flood corresponding to the area designated as a flood hazard area on a community’s flood hazard map or otherwise legally designated. Base flood elevation (BFE): The elevation of flooding, including wave height, having a 1% chance of being equaled or exceeded in any given year. Design flood elevation (DFE): The elevation of the design flood, including wave height, relative to the datum specified on a community’s flood hazard map. Flood hazard area: The area subject to flooding during the design flood. Flood hazard map: The map delineating flood hazard areas adopted by the Authority Having Jurisdiction. Flood insurance rate map (FIRM): An official map of a community on which the Federal Insurance and Mitigation Administration has delineated both special flood hazard areas and the risk premium zones applicable to the community. Special flood hazard area (area of special flood hazard): The land in the floodplain subject to a 1% or greater chance of flooding in any given year. These areas are delineated on a community’s FIRM as A-Zones (A, AE, A1-30, A99, AR, AO, or AH) or V-Zones (V, VE, VO, or V1-30). (ASCE, 2017a) In this code, it is specified that structural elements of buildings or structures shall be designed, constructed, connected, and anchored to resist flotation, collapse, and permanent lateral displacement due to the action of flood loads associated with the design flood and other loads in accordance with load combinations (ASCE, 2017a). Several loads during flooding are prescribed for design use, such as hydrostatic load, hydrodynamic load, wave load and impact loads. As an example, the dynamic static and total pressure distributions against a wall are shown in Figure 3-8. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 36 Figure 3-8. Dynamic static and total pressure distributions against wall (ASCE, 2017a) 3.5.3.2 Flood design in Jamaica From a document presenting the Jamaica National Building Act (Da Costa, 2018), the flood design for buildings in Jamaica refers to the International Building Code (ICC, 2017) and the latest ASCE 7 (ASCE, 2017a). The contents of the flood design methodology of ASCE 7-16 can be seen in Section 3.5.3.1. 3.6 Fragility functions Fragility functions are broadly applied to assess the disaster capacity of structures and the disaster risk due to structural damage. Fragility functions generally express the probability of being in or exceeding a given damage state (e.g., slight, moderate, extensive, complete) corresponding to disaster intensity (e.g., acceleration, velocity, displacement, etc.). The function is modeled as a cumulative lognormal distribution and is characterized by a median value and lognormal standard deviation of disaster intensity. Since this function is essentially based on structural damage states, it is suitable to compare the structural damage of the existing structure and the retrofitted structure discussed in this study. For the Caribbean, there are a few research studies on the fragility functions of local buildings. Clarke and Carey performed an analytical procedure to evaluate the seismic fragility functions of low-rise URM residential buildings in Trinidad and Tobago, as shown in Figure 3-9 (Clarke, 2017) (Carey & Clarke, 2012). Hancilar took an empirical approach based on the actual damage records from the 2010 Haiti earthquake in order to estimate the seismic fragility functions of typical buildings in Haiti; see Figure 3-9 (Hancilar et al., 2013). However, both studies are for existing buildings and any comprehensive research on fragility functions of both pre- & post-retrofitted buildings and other disasters in the Caribbean are not available. The Federal Emergency Management Agency (FEMA) in the U.S. developed comprehensive documents on disaster risk assessment, including fragility functions, and provides various types of fragility functions for buildings, including essential facilities like emergency centers, hospitals and schools (FEMA 2013a)(FEMA, 2013b)(FEMA, 2013c). In these documents, the fragility functions are classified in several groups according to structural type and design code level (i.e., design quality based on primarily the design era). Under these circumstances, the more comprehensively-developed fragility functions from FEMA are utilized in this section. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 37 Trinidad and Tobago (Clarke, 2017) Trinidad and Tobago (Carey & Clarke, 2012) Haiti (Hancilar et al., 2013) Figure 3-9. Seismic fragility function studies in the Caribbean 3.6.1 Earthquake ground shaking: FEMA Hazus fragility functions For fragility functions for earthquake shaking, FEMA Hazus (FEMA, 2013a) provides a comprehensive package of this function for buildings according to structural types and height classes. In addition, FEMA Hazus developed several ranks of functions according to design level, which relates to seismic design code and era built. For example, the fragility function of a building designed with high-level codes is evaluated more robustly than a building designed with low-level codes, even if the buildings are the same structural type. This consideration is very useful to estimate an improvement degree resulting from a retrofit measure because an improvement is usually conducted to meet the performance objectives of the high-level code. In this section, for simplicity, it is assumed that the existing buildings (before retrofitting) were designed based on low-level codes and that the retrofitted buildings are strengthened to possess the same seismic performance specified by a high-level code. The performance improvement for buildings in an earthquake is then discussed based on two fragility functions, according to existing or retrofitted condition. The fragility functions presented in Hazus are used for this performance comparison, but Hazus also provides a conversion procedure for fragility functions per specific location or soil type. Therefore, the functions can be adjusted by procedure if any specific locations or countries are focused on in a future study. In the following subsections, the seismic fragility functions are described for the representative structural types of buildings. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 38 3.6.1.1 URM buildings 3.6.1.1.1 Overview For unreinforced masonry (URM) buildings, the description in Hazus for the structural type, four damage states and fragility functions of existing and retrofitted buildings are presented as follows, and an expected degree of retrofit improvement is discussed. 3.6.1.1.2 Structural type for URM Unreinforced masonry (URM): These buildings include structural elements that vary depending on the building’s age and, to a lesser extent, its geographic location. In buildings built before 1900, the majority of floor and roof construction consists of wood sheathing supported by wood framing. In large multistory buildings, the floors are cast-in- place concrete supported by the unreinforced masonry walls and/or steel or concrete interior framing. In unreinforced masonry constructed after 1950 (outside California) wood floors usually have plywood rather than board sheathing. In regions of lower seismicity, buildings of this type constructed more recently can include floor and roof framing that consists of metal deck and concrete fill supported by steel framing elements. The perimeter walls, and possibly some interior walls, are unreinforced masonry. The walls may or may not be anchored to the diaphragms. Ties between the walls and diaphragms are more common for the bearing walls than for walls that are parallel to the floor framing. Roof ties usually are less common and more erratically spaced than those at the floor levels. Interior partitions that interconnect the floors and roof can reduce diaphragm displacements. 3.6.1.1.3 Structural damage states description for URM FEMA Hazus (FEMA, 2003a) defines a number of damage states. The damage states for URM buildings are summarized in Table 3-10. Damage State Description State DS1 None No observable damage Diagonal, stair-step hairline cracks on masonry wall surfaces; larger cracks DS2 Slight around door and window openings in walls with large proportion of openings; movements of lintels; cracks at the base of parapets Most wall surfaces exhibit diagonal cracks; some of the walls exhibit larger DS3 Moderate diagonal cracks; masonry walls may have visible separation from diaphragms; significant cracking of parapets; some masonry may fall from walls or parapets. In buildings with relatively large area of wall openings most walls have DS4 Extensive suffered extensive cracking. Some parapets and gable end walls have fallen. Beams or trusses may have moved relative to their supports. Structure has collapsed or is in imminent danger of collapse due to in-plane or DS5 Complete out-of-plane failure of the walls. Approximately 15% of the total area of URM buildings with complete damage is expected to be collapsed. Table 3-10. URM building damage states (adapted from FEMA, 2003a) 3.6.1.1.4 Fragility functions for URM In Table 3-11, the parameters of lognormal distribution for the fragility curves of existing and retrofitted buildings are listed. For the retrofitted URM building, it is assumed that it would be the same performance level of a reinforced masonry building, because URM buildings are out of scope in high-level building codes and therefore should be considered reinforced masonry. Based on these parameters, the fragility curves and related probability mass function for existing and retrofitted URM buildings are presented in Figure 3-10 and Figure 3-11. As shown in the fragility functions, the curves of all damage states are largely improved by seismic retrofit implementation. For an earthquake with a PGA of 0.4 g (an average value expected from an earthquake Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 39 occurring in the Caribbean once in 475 years), the probability of “No damage” is 12% for the existing building and is 48% for the retrofitted building; the probability of exceeding “Extensive damage” is 40% for the existing building and is 2% for the retrofitted building. Thus, seismic retrofit can be expected to greatly reduce the probability of seismic damage to URM buildings. Median (g); PGA Lognormal Type Status standard Slight Moderate Extensive Complete deviation (g) Existing 0.19 0.28 0.47 0.68 0.64 URM Retrofitted 0.39 0.65 1.52 2.53 0.64 Table 3-11. Parameters of seismic fragility functions for URM buildings (FEMA, 2013a) Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.7 0.4 0.311 0.6 0.3 0.197 0.204 0.5 0.166 0.2 0.122 0.4 Slight 0.1 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 URM buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-10. Seismic fragility functions & probability mass function at PGA=0.4g, Existing URM buildings PGA (g)=0.400 0.6 0.484 0.5 0.4 0.292 0.3 0.206 0.2 0.1 0.017 0.002 0.0 Extensive Complete Moderate No damage Slight Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-11. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted URM buildings 3.6.1.2 RC/CF with infill wall buildings For reinforced/concrete frame (RC/CF) with infill wall buildings, the description in Hazus for the structural type, four damage states and fragility functions of existing and retrofitted buildings are presented as follows, and an expected degree of retrofit improvement is discussed. 3.6.1.2.1 Structural type for RC/CF with infill wall Concrete frame with infill wall (C3): This is one of the older types of buildings. The infill walls usually are offset from the exterior frame members, wrap around them, and present Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 40 a smooth masonry exterior with no indication of the frame. Solidly infilled masonry panels, when they fully engage the surrounding frame members (i.e. lie in the same plane), may provide stiffness and lateral load resistance to the structure. In these buildings, the shear strength of the columns, after cracking of the infill, may limit the semi-ductile behavior of the system. 3.6.1.2.2 Structural damage states description for RC/CF with infill wall FEMA Hazus (FEMA, 2003a) defines a number of damage states. The damage states for RC/CF with infill wall buildings are summarized in Table 3-12. Damage State Description State DS1 None No observable damage Diagonal (sometimes horizontal) hairline cracks on most infill walls; cracks at DS2 Slight frame-infill interfaces. Most infill wall surfaces exhibit larger diagonal or horizontal cracks; some DS3 Moderate walls exhibit crushing of brick around beam-column connections. Diagonal shear cracks may be observed in concrete beams or columns Most infill walls exhibit large cracks; some bricks may dislodge and fall; some infill walls may bulge out-of-plane; few walls may fall partially or fully; few DS4 Extensive concrete columns or beams may fail in shear resulting in partial collapse. Structure may exhibit permanent lateral deformation. Structure has collapsed or is in imminent danger of collapse due to a combination of total failure of the infill walls and non-ductile failure of the DS5 Complete concrete beams and columns. Approximately 15% (low-rise), 13% (mid-rise) or 5% (high-rise) of the total area of C3 buildings with complete damage is expected to be collapsed. Table 3-12. RC/CF with infill building damage states (adapted from FEMA, 2003a) 3.6.1.2.3 Fragility functions for RC/CF with infill wall In Table 3-13, the parameters of lognormal distribution for the fragility curves of existing and retrofitted buildings are listed. For the retrofitted RC/CF with infill wall building, it is assumed that it would be the same performance level of a reinforced concrete frame building, as RC/CF with infill wall buildings are out of scope in high-level building codes and therefore should be considered reinforced concrete frame. Based on these parameters, the fragility curves and relating probability mass function for existing and retrofitted RC/CF with infill wall buildings are presented in Figure 3-12 and Figure 3-13. As shown in the fragility functions, the curves of all damage states are largely improved by seismic retrofit implementation. For an earthquake with a PGA of 0.4 g (an average value expected from an earthquake occurring in the Caribbean once in 475 years), the probability of “No damage” is 9% for the existing building and 31% for the retrofitted building; the probability of exceeding “Extensive damage” is 52% for the existing building and 6% for the retrofitted building. Thus, seismic retrofit can be expected to greatly reduce the probability of seismic damage to RC/CF with infill wall buildings. Median (g); PGA Lognormal Type Status standard Slight Moderate Extensive Complete deviation (g) Existing 0.17 0.25 0.39 0.65 0.64 RC/CF with infill wall Retrofitted 0.29 0.51 1.07 2.06 0.64 Table 3-13. Parameters of seismic fragility functions for RC/CF with infill wall buildings (FEMA, 2013a) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 41 Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.7 0.4 0.292 0.6 0.3 0.253 0.224 0.5 0.141 0.2 0.091 0.4 Slight 0.1 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 RC/CF with infil wall buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-12. Seismic fragility functions & probability mass function at PGA=0.4g, Existing RC/CF with infill wall buildings Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.7 0.4 0.308 0.340 0.290 0.6 0.3 0.5 0.2 0.4 0.1 0.057 0.3 Slight 0.005 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 RC/CF with infil wall buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-13. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted RC/CF with infill wall buildings 3.6.1.3 RC frame buildings For reinforced concrete (RC) frame buildings, the description in Hazus for the structural type, four damage states and fragility functions of existing and retrofitted buildings are presented as follows, and an expected degree of retrofit improvement is discussed. 3.6.1.3.1 Structural type for RC frame Reinforced concrete frame (C1): These buildings are similar to steel moment frame buildings except that the frames are reinforced concrete. There are a large variety of frame systems. Some older concrete frames may be proportioned and detailed such that brittle failure of the frame members can occur in earthquakes leading to partial or full collapse of the buildings. Modern frames in zones of high seismicity are proportioned and detailed for ductile behavior and are likely to undergo large deformations during an earthquake without brittle failure of frame members and collapse. 3.6.1.3.2 Structural damage states description for RC frame FEMA Hazus (FEMA, 2003a) defines a number of damage states. The damage states for RC frame buildings are summarized in Table 3-14. Damage State Description State DS1 None No observable damage Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 42 Damage State Description State Flexural or shear type hairline cracks in some beams and columns near joints DS2 Slight or within joints. Most beams and columns exhibit hairline cracks. In ductile frames some of the frame elements have reached yield capacity indicated by larger flexural cracks DS3 Moderate and some concrete spalling. Non-ductile frames may exhibit larger shear cracks and spalling. Some of the frame elements have reached their ultimate capacity indicated in ductile frames by large flexural cracks, spalled concrete and buckled main DS4 Extensive reinforcement; non-ductile frame elements may have suffered shear failures or bond failures at reinforcement splices, or broken ties or buckled main reinforcement in columns which may result in partial collapse. Structure is collapsed or in imminent danger of collapse due to brittle failure of non-ductile frame elements or loss of frame stability. Approximately 13% DS5 Complete (low-rise), 10% (mid-rise) or 5% (high-rise) of the total area of C1 buildings with complete damage is expected to be collapsed. Table 3-14. RC frame building damage states (adapted from FEMA, 2003a) 3.6.1.3.3 Fragility functions for RC frame In Table 3-15, the parameters of lognormal distribution for the fragility curves of existing and retrofitted RC frame buildings are listed. Based on these parameters, the fragility curves and relating probability mass function for existing and retrofitted RC frame buildings are presented in Figure 3-14 and Figure 3-15. As shown in the fragility functions, the curves of all damage states are largely improved by seismic retrofit implementation. For an earthquake with a PGA of 0.4 g (an average value expected from an earthquake occurring in the Caribbean once in 475 years), the probability of “No damage” is 6% for the existing building and 12% for the retrofitted building; the probability of exceeding “Extensive damage” is 39% for the existing building and 7% for the retrofitted building. Thus, seismic retrofit can be expected to greatly reduce the probability of seismic damage to RC frame buildings. Median (g); PGA Lognormal Type Status standard Slight Moderate Extensive Complete deviation (g) Existing 0.15 0.23 0.48 0.80 0.64 RC frame Retrofitted 0.19 0.36 1.02 2.48 0.64 Table 3-15. Parameters of seismic fragility functions for RC frame buildings (FEMA, 2013a) Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.418 0.7 0.4 0.6 0.3 0.248 0.5 0.131 0.140 0.2 0.4 0.063 Slight 0.1 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 RC frame buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-14. Seismic fragility functions & probability mass function at PGA=0.4g, Existing RC frame buildings Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 43 Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.493 0.8 0.5 0.7 0.4 0.312 0.6 0.3 0.5 0.2 0.122 0.4 0.070 Slight 0.1 0.002 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 RC frame buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-15. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted RC frame buildings 3.6.1.4 Steel frame buildings For steel frame buildings, the description in Hazus for the structural type, four damage states and fragility functions of existing and retrofitted buildings are presented as follows, and an expected degree of retrofit improvement is discussed. 3.6.1.4.1 Structural type for Steel frame Steel frame (S1): These buildings have a frame of steel columns and beams. In some cases, the beam-column connections have very small moment resisting capacity but, in other cases, some of the beams and columns are fully developed as moment frames to resist lateral forces. Usually the structure is concealed on the outside by exterior nonstructural walls, which can be of almost any material (curtain walls, brick masonry, or precast concrete panels), and on the inside by ceilings and column furring. Diaphragms transfer lateral loads to moment-resisting frames. The diaphragms can be almost any material. The frames develop their stiffness by full or partial moment connections. The frames can be located almost anywhere in the building. Usually the columns have their strong directions oriented so that some columns act primarily in one direction while the others act in the other direction. Steel moment frame buildings are typically more flexible than shear wall buildings. This low stiffness can result in large inter-story drifts that may lead to relatively greater nonstructural damage. 3.6.1.4.2 Structural damage states description for Steel frame FEMA Hazus (FEMA, 2003a) defines a number of damage states. The damage states for steel frame buildings are summarized in Table 3-16. Damage State Description state DS1 None No observable damage DS2 Slight Minor deformations in connections or hairline cracks in few welds. Some steel members have yielded exhibiting observable permanent rotations at connections; few welded connections may exhibit major cracks through DS3 Moderate welds, or few bolted connections may exhibit broken bolts or enlarged bolt holes Most steel members have exceeded their yield capacity, resulting in significant permanent lateral deformation of the structure. Some of the structural members or connections may have exceeded their ultimate capacity exhibited by major DS4 Extensive permanent member rotations at connections, buckled flanges and failed connections. Partial collapse of portions of structure is possible due to failed critical elements and/or connections. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 44 Damage State Description state Significant portion of the structural elements have exceeded their ultimate capacities, or some critical structural elements or connections have failed, resulting in dangerous permanent lateral displacement, partial collapse or DS5 Complete collapse of the building. Approximately 8% (low-rise), 5% (mid-rise) or 3% (high-rise) of the total area of S1 buildings with complete damage is expected to be collapsed. Table 3-16. Steel frame building damage states (adapted from FEMA, 2003a) 3.6.1.4.3 Fragility functions for Steel frame In Table 3-17, the parameters of lognormal distribution for the fragility curves of existing and retrofitted steel frame buildings are listed. Based on these parameters, the fragility curves and relating probability mass function for existing and retrofitted steel frame buildings are presented in Figure 3-16 and Figure 3-17. As shown in the fragility functions, the curves of all damage states are largely improved by seismic retrofit implementation. For an earthquake with a PGA of 0.4 g (an average value expected from an earthquake occurring in the Caribbean once in 475 years), the probability of “No damage” is 6% for the existing building and 9% for the retrofitted building; the probability of exceeding “Extensive damage” is 47% for the existing building and 12% for the retrofitted building. Thus, seismic retrofit can be expected to greatly reduce the probability of seismic damage to steel frame buildings. Median (g); PGA Lognormal Type Status standard Slight Moderate Extensive Complete deviation (g) Existing 0.15 0.23 0.42 0.73 0.64 RC frame Retrofitted 0.17 0.34 0.85 2.10 0.64 Table 3-17. Parameters of seismic fragility functions for steel frame buildings (FEMA, 2013a) Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.7 0.4 0.337 0.296 0.6 0.3 0.5 0.174 0.2 0.131 0.4 0.063 Slight 0.1 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Steel frame buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-16. Seismic fragility functions & probability mass function at PGA=0.4g, Existing steel frame buildings Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 45 Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.481 0.8 0.5 0.7 0.4 0.309 0.6 0.3 0.5 0.2 0.091 0.115 0.4 Slight 0.1 0.005 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Steel frame buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-17. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted steel frame buildings 3.6.1.5 Steel frame high-rise buildings For steel frame high-rise buildings, the description in Hazus for the structural type, four damage states and fragility functions of existing and retrofitted buildings are presented as follows, and an expected degree of retrofit improvement is discussed. 3.6.1.5.1 Structural type for Steel frame high-rise Steel frame (S1): These buildings have a frame of steel columns and beams. In some cases, the beam-column connections have very small moment resisting capacity but, in other cases, some of the beams and columns are fully developed as moment frames to resist lateral forces. Usually the structure is concealed on the outside by exterior nonstructural walls, which can be of almost any material (curtain walls, brick masonry, or precast concrete panels), and on the inside by ceilings and column furring. Diaphragms transfer lateral loads to moment-resisting frames. The diaphragms can be almost any material. The frames develop their stiffness by full or partial moment connections. The frames can be located almost anywhere in the building. Usually the columns have their strong directions oriented so that some columns act primarily in one direction while the others act in the other direction. Steel moment frame buildings are typically more flexible than shear wall buildings. This low stiffness can result in large inter-story drifts that may lead to relatively greater nonstructural damage. 3.6.1.5.2 Structural damage states description for Steel frame high-rise FEMA Hazus (FEMA, 2003a) defines a number of damage states. The damage states for steel frame high- rise buildings are summarized in Table 3-18. Damage State Description state DS1 None No observable damage DS2 Slight Minor deformations in connections or hairline cracks in few welds. Some steel members have yielded exhibiting observable permanent rotations at connections; few welded connections may exhibit major cracks through DS3 Moderate welds, or few bolted connections may exhibit broken bolts or enlarged bolt holes. Most steel members have exceeded their yield capacity, resulting in significant permanent lateral deformation of the structure. Some of the structural members or connections may have exceeded their ultimate capacity exhibited by major DS4 Extensive permanent member rotations at connections, buckled flanges and failed connections. Partial collapse of portions of structure is possible due to failed critical elements and/or connections. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 46 Damage State Description state Significant portion of the structural elements have exceeded their ultimate capacities, or some critical structural elements or connections have failed, resulting in dangerous permanent lateral displacement, partial collapse or DS5 Complete collapse of the building. Approximately 8% (low-rise), 5% (mid-rise) or 3% (high-rise) of the total area of S1 buildings with complete damage is expected to be collapsed. Table 3-18. Steel frame high rise building damage states (adapted from FEMA, 2003a) 3.6.1.5.3 Fragility functions for Steel frame high-rise In Table 3-19, the parameters of lognormal distribution for the fragility curves of existing and retrofitted steel frame high-rise buildings are listed. Based on these parameters, the fragility curves and relating probability mass function for existing and retrofitted steel frame high-rise buildings are presented in Figure 3-18 and Figure 3-19. As shown in the fragility functions, the curves of all damage states are largely improved by seismic retrofit implementation. For an earthquake with a PGA of 0.4 g (an average value expected from an earthquake occurring in the Caribbean once in 475 years), the probability of “No damage” is 4% for the existing building and 4% for the retrofitted building; the probability of exceeding “Extensive damage” is 50% for the existing building and 22% for the retrofitted building. Thus, seismic retrofit can be expected to greatly reduce the probability of severe seismic damage to steel frame high-rise buildings. Median (g); PGA Lognormal Type Status standard Slight Moderate Extensive Complete deviation (g) Existing 0.13 0.20 0.40 0.71 0.64 RC frame Retrofitted 0.13 0.26 0.65 1.73 0.64 Table 3-19. Parameters of seismic fragility functions for steel frame high-rise buildings (FEMA, 2013a) Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.7 0.360 0.4 0.315 0.6 0.3 0.5 0.185 0.2 0.100 0.4 0.040 Slight 0.1 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Steel frame high-rise buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-18. Seismic fragility functions & probability mass function at PGA=0.4g, Existing steel frame high-rise buildings Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 47 Seismic fragility curve (essential facilities) 1.0 PGA (g)=0.400 Probability of Exceeding a Damage State 0.9 0.6 0.525 0.8 0.5 0.7 0.4 0.6 0.3 0.211 0.213 0.5 0.2 0.4 0.040 Slight 0.1 0.011 0.3 Moderate 0.0 0.2 Extensive Complete Moderate No damage Slight Extensive 0.1 Complete 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Steel frame high-rise buildings PGA (g) Seismic fragility functions Probability mass function at PGA=0.4g Figure 3-19. Seismic fragility functions & probability mass function at PGA=0.4g, Retrofitted steel frame high-rise buildings 3.6.2 Earthquake liquefaction: FEMA Hazus fragility functions 3.6.2.1 Overview FEMA Hazus (FEMA, 2013a) developed a fragility function for earthquake liquefaction causing potential ground failure at a building site. Permanent ground displacements (i.e., lateral spreading and ground settlement) due to liquefaction are applied for the intensity index of the fragility function, and one damage state (i.e., weighed average of Extensive and Complete structural damage, see 3.6.1 for damage state definitions according to structural type) is arranged to express liquefaction fragility. Here, the liquefaction damage probability presented in Hazus is based on a condition under which liquefaction occurs during an earthquake. Since earthquake liquefaction is a type of ground hazard, this fragility curve can be assumed to apply to all types of buildings built on shallow or unknown foundations. The estimation process, liquefaction fragility functions and improvement consequence are summarized as follows. 3.6.2.2 Estimation process for a liquefaction damage probability 1) Based on the soil characteristics and expected earthquake intensity at the site, calculate the permanent ground displacement of lateral spreading and ground settlement. 2) Apply the permanent ground displacements to each fragility function and identify the liquefaction damage probability. 3) Take the larger probability as a liquefaction damage probability and combine with a damage ratio of Extensive/Complete damage state of building. 3.6.2.3 Liquefaction fragility functions (lateral spreading and ground settlement) As well as the earthquake shaking fragility functions for buildings, the liquefaction fragility function is expressed by lognormal distribution with permanent ground displacement. The parameters of the functions are listed in Table 3-20 and the fragility functions are depicted in Figure 3-20. Median (in.); PGD Lognormal standard Failure mode Extensive/Complete deviation (in.) Lateral spreading 10 1.2 Ground settlement 60 1.2 Table 3-20. Parameters of liquefaction fragility functions (FEMA, 2013a) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 48 Figure 3-20. Liquefaction fragility functions for buildings with shallow/unknown foundations 3.6.2.4 Liquefaction damage after retrofit implementation Since liquefaction improvement is usually a substantial measure, the liquefaction hazard is, for buildings in this study, assumed to be essentially eliminated once such a measure is implemented. 3.6.3 Wind: FEMA Hazus fragility functions 3.6.3.1 Overview The roof, walls and the connections of buildings are susceptible to damage from wind load caused mainly by peak gust wind speed (PGWS) due to hurricane. For the fragility functions for wind hazards like hurricanes, FEMA Hazus (FEMA, 2013b) provides a number of wind fragility functions associated with wind speed for buildings according to the materials and connections of structural components (like the roof and wall). Two sets of fragility functions are basically developed by Hazus: one for buildings with low capacity of roof-wall connection and one for buildings in which the strength of roof-wall connections is more robust. In this study, it is considered that the former fragility function is suitable for representing the existing building and the latter one is proper for the retrofitted building. Then, the difference between these two fragility functions is assumed to express the improvement extent of structural strengthening against wind hazard. The fragility parameters of these functions are adapted to estimate the damage probability of buildings for wind hazard and the damage probabilities estimated by fragility functions are used for the performance comparison between existing and retrofitted buildings. In the following subsections, the wind fragility functions are described for the representative structural types of buildings (i.e., masonry/concrete buildings as heavy structures and wood/steel buildings for light structures). 3.6.3.2 Masonry/Concrete buildings Two sets of wind fragility functions are selected for masonry/concrete buildings. Each fragility function (existing or retrofitted) consists of five damage states. The fragility functions and the description of damage states prepared in Hazus are presented below, and an expected degree of retrofit improvement is discussed. 3.6.3.2.1 Structural damage states description for Masonry/Concrete buildings The damage descriptions and typical damage amount for specific components are shown in Table 3-21, according to each damage state. Window/ Roof Missile Roof Wall Damage Roof cover Damage description Door deck impacts structure structur State failure failure failure on walls failure e failure No Little or no visible damage ≤ 2% No No No No No damage/ from the outside. No broken Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 49 very minor windows, or failed roof deck. damage Minimal loss of roof over, with no or very limited water penetration. Maximum of one broken window, door or garage door. One Moderate roof cover loss that window, Minor can be covered to prevent > 2% and door, or <5 No No No damage additional water entering the ≤ 15% garage impacts building. Marks or dents on door walls requiring No painting or failure patching for repair. Major roof cover damage, > one and moderate window breakage. Typically Moderate > 15% and ≤ the 1 to 3 Minor roof sheathing failure. 5 to 10 No No damage ≤ 50% larger of panels Some resulting damage to impacts 20% & 3 interior of building from water. > the Major window damage or roof larger of > 3% Typically Severe sheathing loss. Major roof > 50% 20% & 3 and ≤ 10 to 20 No No damage cover loss. Extensive damage to and ≤ 25% impacts interior from water. 50% Complete roof failure and/or, Typically failure of wall frame. Loss of Typically Destruction > 50% > 25% > 20 Yes Yes more than 50% of roof > 50% impacts sheathing. Table 3-21. Damage states for Masonry/Concrete buildings (FEMA, 2013b) 3.6.3.2.2 Fragility functions for Masonry/Concrete buildings The parameters of lognormal distribution for wind fragility functions were read from the original fragility functions developed by Hazus (FEMA, 2013b). Based on the parameters, the wind fragility functions and relating probability mass function for existing and retrofitted masonry/concrete buildings are presented in Figure 3-21 and Figure 3-22. As shown in the fragility functions, the curves of all damage states are largely improved by wind retrofit implementation. For a strong wind with a Peak Gust Wind Speed (PGWS) of 125 mph (a typical wind speed for a major Category 3 hurricane), the probability of “No damage” is 19% for an existing building and 45% for a retrofitted building; the probability of exceeding “Severe damage” is 10% for an existing building and 3% for a retrofitted building. Thus, wind retrofit can be expected to greatly reduce the probability of wind damage to masonry/concrete buildings. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 50 Wind fragility curve PGWS (mph)=125 1.0 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.410 0.7 0.4 0.303 0.6 0.3 0.5 0.190 0.2 0.4 0.078 Minor 0.1 0.019 0.3 Moderate 0.0 0.2 Severe Minor Severe Destruction Moderate No damage 0.1 Destruction 0.0 60 80 100 120 140 160 180 200 Masonry/Concrete - low capacity PGWS (mph) Wind fragility functions Probability mass function at PGWS=125 mph Figure 3-21. Wind fragility functions & probability mass function at PGWS=125 mph, Existing Masonry / Concrete buildings Wind fragility curve 1.0 PGWS (mph)=125 Probability of Exceeding a Damage State 0.9 0.6 0.8 0.5 0.449 0.397 0.7 0.4 0.6 0.3 0.5 0.2 0.130 0.4 Minor 0.1 0.023 0.002 0.3 Moderate 0.0 0.2 Minor Severe Destruction Moderate No damage Severe 0.1 Destruction 0.0 60 80 100 120 140 160 180 200 Masonry/Concrete - high capacity PGGW (mph) Wind fragility functions Probability mass function at PGWS=125 mph Figure 3-22. Wind fragility functions & probability mass function at PGWS=125 mph, Retrofitted Masonry / Concrete buildings 3.6.3.3 Wood/Steel buildings Two sets of wind fragility functions are selected for wood/steel buildings. Each fragility function (existing or retrofitted) consists of five damage states. The fragility functions and the description of damage states prepared in Hazus are presented below, and an expected degree of retrofit improvement is discussed. 3.6.3.3.1 Structural damage states description for Wood/Steel buildings The damage descriptions and typical damage amount for specific components are shown in Table 3-22, according to each damage state. Window/ Roof Missile Damage Roof cover Joist Damage description Door deck impacts state failure failure failure failure on walls Little or no visible damage No from the outside. No broken damage/ver windows, or failed roof deck. ≤ 2% No No No No y minor Minimal loss of roof cover, damage with no or very limited water penetration. Maximum of one broken One Typically Minor window or door. Moderate roof > 2% to ≤ window or No <5 No damage cover loss that can be covered 15% door impacts to prevent additional water Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 51 entering the building. Marks or dents on walls requiring painting or patching for repair. Major roof cover damage, moderate window breakage. One or Typically Moderate > 15% to > One to Minor roof deck failure. Some two 5 to 10 No damage ≤ 50% ≤ 2% resulting damage to interior of panels impacts building from water. Major window damage or roof sheathing loss. Major roof Typically Severe > 2% to > Two to One joist cover loss. Extensive damage to > 50% 10 to 20 damage ≤ 25% ≤ 25% to ≤ 25% interior from water. Limited, impacts local joist failures. Essentially complete roof failure and/or of more than 25% of roof sheathing. Significant Typically Typically Destruction amount of the wall envelope > 25% > 25% > 20 > 25% > 50% opened through windows impacts failure. Extensive damage to interior. Table 3-22. Damage states for wood/steel buildings (FEMA, 2013b) 3.6.3.3.2 Fragility functions for Wood/Steel buildings The parameters of lognormal distribution for wind fragility functions are read from the original fragility functions developed by Hazus (FEMA, 2013b). Based on the parameters, the wind fragility functions and relating probability mass function for existing and retrofitted wood/steel buildings are presented in Figure 3-23 and Figure 3-24. As shown in the fragility functions, the curves of all damage states are largely improved by wind retrofit implementation. For a strong wind with a PGWS of 125 mph (a typical wind speed for a major Category 3 hurricane), the probability of “No damage” is 3% for an existing building and 6% for a retrofitted building; the probability of exceeding “Severe damage” is 53% for an existing building and 29% for a retrofitted building. Thus, wind retrofit can be expected to greatly reduce the probability of wind damage to wood/steel buildings. Wind fragility curve 1.0 PGWS (mph)=125 Probability of Exceeding a Damage State 0.9 0.6 0.524 0.8 0.5 0.381 0.7 0.4 0.6 0.3 0.5 0.2 0.4 0.068 Minor 0.1 0.026 0.002 0.3 Moderate 0.0 0.2 Minor Severe Destruction Moderate No damage Severe 0.1 Destruction 0.0 60 80 100 120 140 160 180 200 Wood/Steel - low capacity PGWS (mph) Wind fragility functions Probability mass function at PGWS=125 mph Figure 3-23. Wind fragility functions & probability mass function at PGWS=125 mph, Existing Wood/Steel buildings Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 52 Wind fragility curve 1.0 PGWS (mph)=125 Probability of Exceeding a Damage State 0.9 0.6 0.521 0.8 0.5 0.7 0.4 0.6 0.287 0.3 0.5 0.2 0.129 0.4 0.061 Minor 0.1 0.002 0.3 Moderate 0.0 0.2 Minor Severe Destruction Moderate No damage Severe 0.1 Destruction 0.0 60 80 100 120 140 160 180 200 Wood/Steel - high capacity PGWS (mph) Wind fragility functions Probability mass function at PGWS=125 mph Figure 3-24. Wind fragility functions & probability mass function at PGWS=125 mph, Retrofitted Wood/Steel buildings 3.6.4 Flood: FEMA Hazus damage functions FEMA Hazus (FEMA, 2013c) developed two types of damage functions for flood hazard. One of the functions measures building damage amount (i.e., percent of replacement cost) corresponding to flood inundation depth, and another one scales the collapse potential of the building based on a combination of water velocity and depth. As seen in the damage functions, the depth of flood inundation is a key factor. Hazus also mentions in the document that (FEMA, 2013c): In general, it is expected that the major structural components of a building will survive a flood, but that the structural finishes and contents/inventory may be severely damaged due to inundation. The damage functions and improvement consequence are summarized as follows. 3.6.4.1 Depth-damage curve for flood hazard The depth damage curves presented in Hazus are compiled from various loss data recorded by the Federal Insurance and Mitigation Administration (FIMA), the U.S. Army Corps of Engineers (USACE) and the USACE Institute for Water Resources (USACE IWR). Example functions are shown in Figure 3-25; the damage percentage can be identified by the inundation depth and is approximately 35% damage occurring at 3 feet inundation. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 53 FIA (FIMA)-based structure depth damage curve 2 or more stories, basement-modified FIA (FIMA)-based structure depth damage curve Split level, basement-modified Figure 3-25. Building depth damage curves for buildings (FEMA, 2013c) 3.6.4.2 Velocity depth-damage curve for flood hazard Flood with high velocity combined with water inundation could result in not only content loss, but also structural damage. For example, large water pressure due to flooding might separate buildings from their foundations if the connections to the foundations do not possess enough capacity for the water pressure. Hazus adopted the velocity-based building collapse functions developed by the Portland District of the U.S. Army Corp of Engineers, as shown in Figure 3-26. According to the structural type and height of the building, the collapse potential (i.e., collapse or not collapse) due to flooding is expressed by a combination Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 54 of inundation depth and overbank water velocity. As seen in the functions, the collapse potential is very low for buildings at a certain low depth of inundation. Wood building Masonry/Concrete building Steel building Figure 3-26. Building velocity-depth damage curves for various structural types (FEMA, 2013c) 3.6.4.3 Flood damage after retrofit implementation Since flood improvement is typically a substantial measure to protect buildings from water invasion, the damage due to flood hazard is assumed to be essentially eliminated once such a measure is implemented in this study. 3.7 Strengthening techniques This section shows general strengthening measures for the vulnerable components of buildings depending on the type of hazard considered in this study. As example retrofit techniques, measures implemented in modern building-code countries like the U.S. or Japan are focused. The subsequent section provides cost information for strengthening measures where specific cost data for the Caribbean is available. 3.7.1 Earthquake: ground shaking For earthquake shaking hazard, strengthening the in-plane capacity and stiffness and improving structural irregularity and ductility are common measures to increase building resiliency. Improving the out-of-plane resistance of walls is effective to reduce severe damage to old or non-engineered buildings, especially URM buildings. In addition, installing an energy dissipation system could be a retrofit option for the most important buildings to be accessed in post-disaster circumstances (e.g., emergency center or medical facility). 3.7.1.1 Wall out-of-plane strengthening Old or unreinforced buildings (e.g., unreinforced masonry buildings) typically do not possess appropriate reinforcement detailing required by a modern building code. Thus, the out-of-plane bending resistance of the buildings is usually inadequate. The failure of heavy walls due to a lack of out-of-plane resistance is historically the most devastating damage to buildings and people living in or walking by the building. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 55 Bracing the wall at regularly-spaced intervals is a general measure for improving the resistance of out-of- plane failure. To brace the wall, a steel strongback, existing diaphragm (i.e., floor element), concrete pilaster and post-installed anchor are applied; see Figure 3-27 and Figure 3-28. This improvement can be installed on both the inside and outside of the building, but the inside approach is more popular for heritage building preservation purposes or architectural exterior finishes. Out-of-plane wall bracing Anchorage from diaphragm to wall Figure 3-27. Out-of-plane strengthening for masonry wall (FEMA, 2006b) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 56 Vertical element connections Strongback connections Figure 3-28. Vertical element connections of out-of-plane strengthening (FEMA, 2006b) 3.7.1.2 In-plane strengthening The global seismic strength of a building heavily relies on the in-plane strength of vertical elements, such as shear walls, braced frames or moment frames. These elements are essential components constituting the structural integrity and seismic robustness of a building. A lack of in-plane seismic capacity could therefore cause significant damage that requires extensive repair work or a complete collapse of the building that necessitates demolition and replacement. To improve the in-plane seismic strength of a building, a new element, such as a shear wall or braced frame, is typically added to the existing structural system, as shown in Figure 3-29 and Figure 3-30. Adding new elements from the outside is more convenient for building usability and retrofit constructability, but inside installation is also constructible with a careful installation plan. Typical configurations and details of reinforced concrete (RC) shear walls and steel buckling restrained braces (BRB) are shown in Figure 3-31 and Figure 3-32, respectively. In both cases, it is important to provide the appropriate load path elements to transfer the seismic force from the new elements through new foundations to the ground. Enhancing the existing in-plane element is also an effective measure. For an existing concrete wall, a methodology to attach fiber-reinforced polymer (FRP) sheets was developed to strengthen the shear strength and improve the ductility of the wall. The attachment elevation and details of anchoring FRP to concrete walls are presented in Figure 3-33. Thickening the existing masonry shear wall is also an effectual way to increase the in-plane seismic strength. Concrete/shotcrete shear wall is usually overlaid on the existing masonry wall with the connecting dowels and new foundation as needed. Typical section details and elevation configuration of overlaid concrete walls are shown in Figure 3-34. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 57 Figure 3-29. Typical in-plane strengthening, RC shear wall (FEMA, 2006a) Figure 3-30. Typical in-plane strengthening, Steel brace (FEMA, 2006a) Wall elevation with existing slab Wall connection with existing slab Figure 3-31. RC shear wall strengthening (FEMA, 2006b) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 58 Typical BRBF configurations Common BRB assembly Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 59 BRBF example Figure 3-32. Steel buckling restrained brace (BRB) strengthening (NIST, 2015) FRP shear strengthening on existing wall FRP anchor to existing wall Figure 3-33. FRP strengthening on existing RC shear wall (FEMA, 2006b) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 60 Concrete/shotcrete wall overlay Wall addition elevation Figure 3-34. RC shear wall addition (or shotcrete) on existing masonry wall (FEMA, 2006b) 3.7.1.3 Structural irregularity improvement Structural irregularity due to unbalanced configuration or strength/stiffness is one of the most important factors for the structural stiffness and strength of buildings; see Figure 3-35. For example, the vertical discontinuity or horizontal offset of a structural wall could cause serious damage to critical columns right below the upper floor wall or could induce large harmful displacement at weak locations due to torsional movement. The stress/force concentration is usually generated at the corner of irregular plan shapes and severe damage would happen on the elements around the corner. Weak/soft story (typically at the first floor) due to insufficient structural elements like walls or braces might bring a large permanent residual displacement or a pancake collapse (i.e., completely crush the story) on the floor. In that case, the whole building might have to be demolished or replaced because of the destructive floor damage, even though other floors may look intact. Common measures to reduce structural irregularities and improve the structural performance are adding shear walls, braced frames and moment frames, as shown in Figure 3-36. The in-plane strengthening measures presented in Section 3.7.1.2 can be applied. As can be seen in Figure 3-36, adding walls/braces to the weak story can usually diminish multiple structural irregularities. For example, the new walls/braces could eliminate the weak/soft story effect, reduce the torsional movement, make the seismic displacement smaller and decrease the stress concentration. This kind of retrofit for structural irregularities can help avoid severe global damage of buildings and hurtful effects on occupants and facilities. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 61 Wall vertical discontinuity (ASCE, 2017b) Irregular plan; re-entrant corner (FEMA, 2006a) Irregular elevation; wall offset (FEMA, 2006a) Soft/weak story (FEMA, 2006a) Figure 3-35. Examples of structural irregularities of buildings Retrofit measures for soft first story Strengthening solutions for re-entrant corner Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 62 Typical improvements for unbalanced LFRS Figure 3-36. Typical strengthening for structural deficiencies due to irregularities (FEMA, 2006a) 3.7.1.4 Ductility improvement In modern seismic building codes, ensuring a ductile response from buildings during earthquakes has become one of the important design concepts. Ductility is simply described as the capability of large rotation and displacement without any serious damages. Conversely, a brittle response is a sudden failure at small displacement or fragile damage that causes loss in ability to bear gravity loads. Insufficient ductility is, for example, typically caused by an inadequate confinement of structural elements (e.g., column, beam or wall) or an inappropriate yielding location at the column-beam connection. To ensure the ductile behavior of buildings during earthquakes, some retrofits should be implemented on local elements. For concrete columns or masonry piers, as presented in Figure 3-37, FRP overlay, concrete jacketing or steel jacketing are popular measures to improve the confinement of existing columns, to increase the capacities for shear and axial loads and to enhance ductility. For steel beam-column connections, ensuring a plastic hinge at a ductile location (i.e., not at welding/bolting location) is the important approach to withstanding the inelastic demand of rotation and displacement for ductile response. For this approach, reducing the beam section or adding haunch components, as shown in Figure 3-38, are effective measures for enhancing the plastic moment capacity of beams and the ductile beam response. RC column retrofit by FRP overlay RC column retrofit by concrete/steel jacketing Figure 3-37. Ductility improvement measures for RC column (FEMA, 2006b) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 63 Reduced beam section of existing beam Welded haunch on existing beam Figure 3-38. Retrofit measures for steel beam ductility (FEMA, 2006b) 3.7.1.5 Stiffness and capacity improvement Generally, framed buildings, such as steel or wood structures, are more flexible than walled buildings, such as RC or masonry structures, in terms of lateral stiffness. For those soft buildings, an excessive drift or additional stress due to P- effect likely causes structural connection damage, residual displacement and stress on columns, functional damage on nonstructural components, like partition or cladding, and pounding impacts on adjacent buildings. The installation of new lateral force-resisting elements is a common way to stiffen and strengthen soft buildings. Retrofits with steel brace or steel plate shear wall are shown in Figure 3-39 and Figure 3-40, respectively. Both measures can largely enhance the stiffness of existing framed structures and improve the seismic strength of existing buildings. Usually, new foundations for new braces/walls are needed to transfer the load from the new lateral force-resisting elements to the ground. This type of stiffness improvement works appropriately for wind forces as well. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 64 Brace frame configurations Typical brace connection on existing elements Figure 3-39. Steel brace frame retrofit (FEMA, 2006b) Steel plate shear wall elevation Typical plate connection on existing elements Figure 3-40. Steel plate shear wall retrofit (FEMA, 2006b) 3.7.1.6 Energy dissipation system In this section, two relatively new technologies to improve the seismic performance of buildings are introduced. The first new technology is a seismic isolation system, and the second is a damper system for seismic energy dissipation. Both technologies aim to reduce the seismic demand (i.e., earthquake force) on Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 65 the buildings by installing devices that would allow for the existing seismic strength of buildings to withstand the reduced demand. A primary concept of a seismic isolation system is to lengthen the building’s natural period of vibration and to pull it away from the major frequencies of earthquake shaking in order to reduce the seismic demand imparted from the ground to the building. As shown in Figure 3-41, the isolator device (with an included damper or a separate damper) has a role in lengthening the building vibration period above the device. The isolator is commonly installed at the bottom of the building to maximize the isolation effect on the entire building. However, these days, the isolator can be installed at the middle height of buildings to deal with potential property-line restrictions or architectural/mechanical reasons. In both cases, additional strengthening is needed for the existing elements around and below the isolator, like beams, columns and foundations. Figure 3-42 presents a seismic retrofit example that demonstrates the base isolation system. The energy dissipation damper system shown in Figure 3-43 adds an extra damping to the existing building. The increased damping decreases the seismic energy on the building and reduces the local forces on the structural elements. The existing capacity of the building could then endure the earthquake impact, but, similar to the seismic isolation system, some strengthening is likely needed for the beams, columns and foundations around the installed dampers. The installation of fluid-viscous dampers for a hospital retrofit is shown in Figure 3-44. Figure 3-41. Conceptual figure of seismic isolation system (ASCE, 2017b) Figure 3-42. Base isolation retrofit example of San Francisco City Hall (Naaseh et al., 2006) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 66 Figure 3-43. Typical configurations of energy dissipation damper retrofit (FEMA, 2006b) Figure 3-44. Energy dissipation damper retrofit for Naval Hospital Bremerton (FEMA, 2007) 3.7.2 Earthquake: liquefaction Foundation retrofit and soil improvement are typical approaches to make buildings more robust against earthquake liquefaction. The former is essentially for resisting potential ground failure due to liquefaction and the latter is for improving soil and for reducing liquefaction risk during an earthquake. 3.7.2.1 Foundation retrofit When the size of shallow-spread foundations or the amount of deep-pile foundations are not sufficient to support the building for a liquefaction event during an earthquake, the consequence could be terrible – tilting, overturning, losing stability or collapse. These damage modes could generate a serious outcome, like demolishment and replacement or extensive repair work with foundation improvement. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 67 For shallow-spread footing, enlarging the existing footing by attaching additional concrete footing is a typical measure to increase capacity; see the left figure of Figure 3-45. New pile installation next to an existing deep-pile foundation is usually implemented to improve the deep-foundation capacity; see the right figure of Figure 3-45. In this case, lateral capacity can be added in addition to compression and tension capacity. In both cases, the retrofit construction work is not simple because, for example, excavation, formwork, reinforcement placement and concrete pouring are usually required around the existing foundation. Enlarge existing spread footing Adding piles to existing deep foundation Figure 3-45. Typical foundation strengthening (FEMA, 2006b) 3.7.2.2 Soil improvement In contrast with the foundation retrofit introduced in Section 3.7.2.1, soil improvement is an alternative to keeping a building stable from large ground failure (i.e., ground settlement and lateral spreading) due to liquefaction. In Figure 3-46, examples to improve the soil below existing shallow foundations and around existing deep foundations use permeation grouting on liquefiable soil layers. There are generally two types of soil grouting: permeation grouting and compaction grouting. The former injects chemical/cement grout into the soil and aggregate without displacing the materials, and it solidifies the sandy soils and improves resistance to liquefaction. The latter injects very stiff grout into a layer of soil to force the soil particles into a tighter packing arrangement, and it increases the soil density and the liquefaction resistance (FEMA, 2006b). Lateral spreading usually occurs at the soil above the liquefiable soil layers, and the soil spreads laterally toward an open face when liquefaction happens below. This event might then laterally move the surrounding buildings and destroy the deep-pile foundation of the building. For mitigating the risk of lateral Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 68 spreading of soil due to liquefaction, the soil grouting measures described above work to prevent liquefaction itself, or the installation of stone/gravel columns by vibrocompaction method to solidify the open face soils. Figure 3-47 shows an example of stone/gravel columns installed around the open face in order to densify the soil layers (FEMA, 2006b). Under shallow foundation Around existing deep foundation Figure 3-46. Permeation grouting on liquefiable soil layers (FEMA, 2006b) Figure 3-47. Soil strengthening for lateral spreading by stone/gravel column (FEMA, 2006b) 3.7.3 Wind The connections of roof, foundation and external walls are critical components of buildings for wind hazard. In particular, lightweight structures or their components, like the roof or walls, might be overturned or blown out by strong wind if the connections do not possess sufficient capacity. Typical methodologies to improve those connections are discussed in this section. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 69 3.7.3.1 Roof connection Strong wind can easily blow the roof away if it is not securely attached to the walls or beams of a building. Since the roof usually has a structural role in horizontally supporting a vertical element such as a wall, losing the roof due to strong wind might cause cascading damage, like wall instability or failure. Typical details to strengthen the roof connection of wood and masonry structures are shown in Figure 3-48. Adding angle clips and blockings to strengthen the roof connection is a general strengthening measure for wood structures. For masonry structures, new concrete bond beams might be needed to enhance a connection between the masonry wall and roof, and the anchoring, steel strap and nailing are typically provided for securing the roof connection. Roof connection of wood structure Roof connection of masonry structure Figure 3-48. Typical roof connection improvements (FEMA, 2006b) 3.7.3.2 External curtain wall and wall When an external wall is not firmly attached to the building, fall or collapse of external nonstructural walls is a typical failure mode for these components in a strong wind event. If it is not a structural element or bearing wall, it could cause minor damage, structurally, to a building. However, the falling or collapsing element could lead to a serious impact on people or structures around the building. In modern building codes, particular design mechanisms are required for the connection of external curtain walls, as shown in Figure 3-49. A certain level of capability for rocking and sway should meet the design requirement in accordance with the expected drift and long duration fatigue due to wind. In Figure 3-50, hurricane damage to perimeter block walls and a typical connection detail of block wall, roof and foundation are presented. An adequate connection with anchor and reinforcement could avoid this type of wall failure. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 70 Rocking and sway mechanisms of connection for earthquake Push-pull connection of panel (FEMA, 2006a) (FEMA, 2012) Figure 3-49. Typical connection details of external curtain walls Hurricane damage to external wall Typical connections of external wall Figure 3-50. Connections of external wall (FEMA, 2007a) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 71 3.7.3.3 Foundation connection Uplift and lateral load due to strong wind could cause severe structural damage to a building, like overturning or movement, particularly for lightweight structures like wood and steel-framed buildings. This type of damage likely will induce damages to adjacent buildings as well. For the rigid foundation connection of wood structures, a hold-down and post-installed anchor with tie- down post should be installed along the existing studs, as shown in the left figure of Figure 3-51. A typical retrofit detail to improve uplift capacity of a steel-column base is presented in the right figure of Figure 3-51. Stiffener plates and base plates are welded to the existing steel elements and post-installed anchors are then installed to the foundation. In both cases, a new foundation could be constructed to resist the large uplift load as needed. Foundation connection retrofit of wood structure Foundation connection retrofit of steel structure Figure 3-51. Typical connection retrofits of vertical elements to foundation (FEMA, 2006b) 3.7.4 Flood The general measures to protect buildings from flood hazard aim to prevent flood water inundation and to resist static/dynamic water pressure due to flooding. Water inundation affects the building’s functionality or occupancy and water pressure affects the building’s structural stability. 3.7.4.1 Building envelope One of the retrofit measures to keep a building protected from flood inundation is to install floodgates at potential water intrusion routes around the building and flood shields at openings of the building, such as doors and windows; see Figure 3-52, Figure 3-53, and Figure 3-54, respectively. The height of the gate or shield should be designed according to the expected flood depth analyzed by local flood hazard conditions, and the gate and shield need to be designed to resist the water pressure due to inundation depth. These types of building envelope improvements (i.e., watertight system) can also be applied to specific rooms inside of the building, such as machine and electric rooms. Preventing water inundation could help avoid any functional outages of the building that are potentially caused by damage to mechanical and electrical equipment. This outage has to be avoided at essential facilities like emergency centers, government buildings and hospitals. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 72 Figure 3-52. Floodgate installation (FEMA, 2007a) Figure 3-53. Flood shield for door (FEMA, 2013) Figure 3-54. Flood shield for window (FEMA, 2013) 3.7.4.2 Elevated construction Another effective measure to protect a building from flood inundation is to construct an elevated structure, like floodwall around the building. Typical types of floodwalls are shown in Figure 3-55 and represent, Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 73 essentially, cantilever walls. As a retrofit measure for flood hazard, a floodwall is typically constructed some distance from the building and it should be designed as a freestanding, independent structure separated from the building. This kind of wall is typically constructed of reinforced concrete. For designing a floodwall, the lowest point of the site, topographic information, site-specific flood hazard level (i.e., inundation depth), soil characteristics, potential debris impact and a performance objective are generally identified. As several examples demonstrate in Figure 3-56, a floodwall protects buildings from intrusion of floodwater and provides a barrier against water inundation. Figure 3-55. Typical floodwalls (FEMA, 2013) Floodwall at a school Floodwall at a hospital Floodwall at a hospital Figure 3-56. Floodwalls protecting buildings (FEMA, 2013) 3.7.4.3 Foundation tie If a building has watertight or waterproof conditions, the perimeter wall and structural elements of the building have to resist the loads due to flooding water and washed away debris in both horizontal and vertical directions. In Figure 3-57, typical flood loads, such as hydrostatic and hydrodynamic loads, are depicted. In this aspect, the foundation tie for a perimeter wall is a critical component to prevent any severe Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 74 damage or functional outage of the building. Especially for a lightweight building, a rigid connection to the foundation is very important. A retrofit example of an existing foundation tie for a light wood building is shown in Figure 3-58. Post-installed anchors can be added to the existing connection of wall and foundation. Hydrostatic loads Hydrodynamic loads Figure 3-57. Water loads on building due to flooding (FEMA, 2007a) Figure 3-58. Typical strengthening on foundation tie (FEMA, 2006b) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 75 3.8 Cost analysis 3.8.1 Introduction The service life of critical buildings (i.e., essential facilities) is usually longer than other types of buildings, such as residential, commercial, private office or industrial buildings. For critical buildings, several natural hazard events are anticipated to affect those buildings during their service life. In a natural disaster, critical buildings need to be able to withstand the hazard itself in order to become centers of countermeasures, like an emergency center, evacuation shelter or urgent medical facility. To prevent or reduce the impacts on these important buildings, structural strengthening or repair preparations should be considered in any long- term plan for the maintenance of critical buildings. For example, a periodical condition assessment of structural capacity could be performed that considers building material deterioration recognized by maintenance facilities, and a new hazard map should be updated using the latest research accomplishments. When the necessity of structural strengthening or repair is recognized, improvements have to be implemented. Cost information on improvement measures and building replacement is then indispensable for not only financial planning, but also execution priority. In this section, cost data and research on building construction and structural retrofits are presented. Information for the Caribbean and the U.S. is then introduced in the reconstruction cost section. To note, retrofit cost data in this section is mainly based on the U.S. construction environment, as similar materials and data are minimal for the Caribbean. However, the ratio of retrofit cost to reconstruction fees is generally expected to be at the same level among many countries, including the Caribbean and the U.S. 3.8.2 Reconstruction costs The building reconstruction cost (i.e., construction or replacement cost of the building) in the Caribbean is anticipated to be different across the 16 countries studied in this report and from that of the U.S. The data for average construction costs of buildings are summarized in this section. 3.8.2.1 Construction cost in the Caribbean A financial report for the Caribbean and Latin America provides an indicative construction cost of buildings for some Caribbean countries, as shown in Table 3-23 (BCQS, 2019). The cost is estimated for higher and lower ranges according to property types and includes only the cost of construction material, machinery and labor (i.e., hard cost). The report explains that the cost range is intended to provide data at a 95% confidence interval from the data analysis. The cost ranges (i.e., variation) of the buildings at modest cost are smaller than that of the higher-cost buildings. In this table, the data of hotels/condos and offices, high- rise and special offices could be applied to a cost study of the critical buildings. Critical buildings are usually designed for higher performance level depending on the important usage of the essential facility. Therefore, for example, specialist offices and high-rise story buildings in Table 3-23 could be applied to a cost study of medical facility, government office or emergency center as higher performance and cost are anticipated. Hotels/condos (e.g., three-star level), on the other hand, seem to be adequate for a study of shelter cost because moderate performance and cost are expected for temporal use and evacuation shelter. 3.8.2.2 Construction cost in Saint Vincent and the Grenadines Specific information for construction, including cost for the critical buildings in Saint Vincent and the Grenadines, is presented in Table 3-24 (Campbell, 2020). The occupancy, footprint area, number of stories, frame type, primary material, number of occupants, etc. are all listed, and costs are shown in Eastern Caribbean dollars (EC$) and U.S. dollars (US$). There are some variations based on building occupancy, and the construction cost per unit area for hospital and government buildings seems to be higher than schools or hospitalities/hotels. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 76 Trinidad & Bahamas Barbados Jamaica St. Lucia St. Maarten Item Tobago Low High Low High Low High Low High Low High Low High Residential high quality 280 460 250 500 180 300 190 350 240 390 175 350 Residential medium quality 160 260 140 230 110 170 125 160 150 230 100 150 Residential modest quality 120 200 120 180 80 130 90 130 120 180 75 120 Hotels/condos five star 330 550 310 480 220 360 285 350 290 460 250 330 Hotels/condos three star 170 290 190 250 120 190 175 235 160 250 170 250 Offices 1-3 story, shell 150 240 160 230 100 160 140 200 140 210 130 175 Offices 1-3 story, fitout 270 450 250 400 180 300 190 300 240 380 170 250 High rise 4-8 story, shell 220 370 190 250 150 240 160 230 200 310 155 220 High rise 4-8 story, fitout 330 550 290 450 220 360 240 340 290 460 220 290 Specialist offices 340 570 300 500 230 370 275 400 290 470 225 375 Storage or warehouse 80 120 75 110 50 80 70 100 80 120 60 90 Retail single story 130 210 135 180 80 140 110 140 120 190 100 150 Table 3-23. Construction cost of buildings in the Caribbean in US$ per Sq. Ft. (BCQS, 2019) Construction Cost Footprint Year Type Stories Framing type Primary material Occupants Design code EC$/ US$ US$ (m2) built EC$ m2 /m2 /ft2 School 2,441 one Concrete frame Concrete and Mason Block 484 2020 ACI-318-14 2,997,827 1,228 454 42 School 331 one walls Concrete 49 2017 ACI-318-14 47,523 144 53 5 Hospital 607 two Concrete frame Concrete and Masonry Block 5,756 2019 ACI-318-14 2,200,000 1,813 671 62 Hospital 718 one Concrete frame Concrete 7,061 2013 ACI-318-14 924,000 1,287 476 44 Government 505 one Concrete frame Concrete 5,300 2017 ACI-318-14 1,400,000 2,770 1,025 95 Hospital 697 one Concrete frame Concrete 5,300 2015 ACI-318-14 1,120,000 1,606 594 55 Hospitality/ 1,621 two Concrete frame Concrete, wood 251 2019 ACI-318-14 675,000 208 77 7 hotel Table 3-24. Construction cost of essential facilities in Saint Vincent and the Grenadines (Campbell, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 77 3.8.2.3 Construction cost in Belize A report for a workshop under the U.S. Agency for International Development/Organization of American States (USAID/OAS) Post-Georges Disaster Mitigation Project documents the average construction cost of residential buildings in Belize, as shown in Table 3-25. According to the household income level (i.e., low, middle or upper), the average unit area and construction cost of residential building was summarized with some variations. The unit construction cost (US$/Sq. Ft.) of upper income residences was approximately two to three times more than that of low income residential building in Belize. For the average area of a residential unit, there is about a seven times difference among the two classes. Area Construction rate Cost range Income (Sq. Ft.) (US$/Sq. Ft.) (US$) Low 360 – 640 30 – 50 10,800 – 32,000 Middle 640 – 2,200 50 – 65 32,000 – 143,000 Upper 2,200 – 5,000 65 – >100 143,000 – >500,000 Table 3-25. Construction cost of residential buildings in Belize (Henderson, 2001) 3.8.2.4 Construction cost in the U.S. The typical construction costs for buildings in the U.S. are introduced here as a reference for this study. In the U.S., this type of cost varies between different regions or cities. A construction market analysis report for the U.S. investigated the average construction costs according to major cities and building occupancy; see Table 3-26 (Cumming, 2020). The costs shown in the table are the total cost, combining hard cost (e.g., material fees and labor cost) and soft cost (e.g., plan/design fee and inspection fee), and they are at the same level as construction cost analyzed by other researchers (Raetz et al., 2020). Similar to other cost estimations shown in this section, the cost statistics also indicate some variations (i.e., lower and higher limits). Five major cities on the West Coast of the U.S. were chosen to reference because earthquakes are much more frequent here than other U.S. cities, and wind and flood disasters happen here as well. The construction costs of healthcare, higher education and public/community facilities seem to be relatively higher than that of other occupancy types based on the cost in Table 3-26. San Los Angeles San Diego Seattle Portland Item Francisco Low High Low High Low High Low High Low High Residential Single Family Medium Quality 322 386 65 241 139 246 258 310 245 294 Apartment/Condo. Mid-rise 449 584 294 529 273 503 361 469 343 445 Commercial/Office Single Story 364 437 327 395 314 381 292 350 277 333 Mid-Rise 726 870 371 567 359 466 501 629 476 597 High-Rise 833 1,001 458 688 430 678 561 696 533 661 Hospitality/Lodging Three-Star Hotel 555 731 425 661 348 482 376 452 358 429 Five-Star Hotel 761 1,044 679 945 456 712 591 838 561 796 Healthcare Acute Care Facility 1,075 1,433 973 1,296 929 1,238 566 782 538 743 Medical Office Building 612 734 498 598 476 571 398 493 378 469 Specialty Clinic 765 918 623 748 595 714 553 663 525 630 Primary/Secondary Education Elementary School 329 392 341 410 326 391 242 291 230 276 Middle School 362 431 374 448 357 428 265 318 252 302 High School 395 475 411 493 392 471 291 350 277 332 Higher Education Academic/Classroom 618 742 645 774 616 739 515 617 489 587 Laboratory 640 823 801 962 637 816 711 853 675 811 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 78 Administration 628 742 666 799 626 735 532 638 505 606 Dormitory 344 411 401 426 343 407 276 340 262 323 Public/Community Facilities Gov’t Administrative Buildings 474 767 627 759 463 599 557 673 529 640 Museum/Performing Arts 727 1,272 921 1,151 707 1,099 817 1,021 776 970 Recreation/Gymnasium 438 546 412 494 386 490 365 438 347 416 Police Stations 590 822 599 718 572 686 531 637 504 605 Table 3-26. Construction cost of buildings in the U.S. in US$ per Sq. Ft. (Cumming, 2020) 3.8.3 Strengthening costs As cost information for strengthening of existing buildings in the Caribbean was scarce, the available information for retrofit cost in the U.S. (mainly for earthquakes) was then included in this section, as it is anticipated that the ratio of strengthening-to-reconstruction cost of buildings is similar between the Caribbean and the U.S. 3.8.3.1 Methodology for estimating seismic retrofit cost A methodology to estimate typical costs for seismic retrofit of existing buildings was developed and published in FEMA 156 and 157 (FEMA, 1994)(FEMA, 1995). For the methodology of FEMA 156 and 157, a database of seismic retrofit costs for various types of buildings in the U.S. was provided and a regression analysis was conducted to form the formula to evaluate seismic retrofit cost. A similar research project was performed later based on the methodology of FEMA 156 and 157. The project added additional and more recent data on retrofit cost of buildings to refine the estimation formula of FEMA 156 and 157 (NIST, 2017). For a number of building types, listed in Table 3-27, regression analyses based on the database and several trials with different parameter combinations were carried out. The formula most aligned with the actual cost database was then developed, as shown in Eq. 3-4. As seen in the formula, several factors (seismicity, performance objective, area, age, stories, occupancy, and historic variables) are considered. The estimated coefficients for this retrofit cost formula are presented in Table 3- 28. The author of the report states that this formula was developed for estimating the seismic retrofit cost of existing federal buildings, but that the development approach can be generalized to any kind of building inventory. Eq. 3-4 (),, = + + + + + 1 () + 2 () + 3 () + 4 ( ) + 5 () + • : Structural cost per square foot of retrofitting a particular building, • : denotes Seismicity of the building, • : denotes Performance objective of the retrofit, • : denotes Building group, • : Performance objective of the retrofit, • : Fixed effect of seismicity, • : Fixed effect of performance, • : Combined fixed effect (interaction) of seismicity and performance, • : Fixed effect for building group, • 1⋯5 : Model coefficients (regression variables), • : Unobserved error term. Building group Model Building types 1 URM Unreinforced masonry 2 W1 Wood light frame W2 Wood (commercial or industrial) 3 PC1 Precast concrete tilt up walls Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 79 RM1 Reinforced masonry with metal or wood diaphragm 4 C1 Concrete moment frame C3 Concrete frame with infill walls 5 S1 Steel moment frame 6 S2 Steel braced frame S3 Steel light frame 7 S5 Steel frame with infill walls 8 C2 Concrete shear wall PC2 Precast concrete frame with concrete walls RM2 Reinforced masonry with precast concrete diaphragm S4 Steel frame with concrete walls Table 3-27. Building group and types for the prediction model (NIST, 2017) Dependent variable: Parameter Cost per sf Main model Area -0.164 (0.038) Age -6.080 (4.680) Stories 0.306 (0.068) 6 Occupancy : TR -0.308 (0.096) Occupancy: IP -0.947 (0.119) 7 Seismicity : M -0.064 (0.404) Seismicity: H -0.038 (0.337) Seismicity: VH -0.132 (0.322) 8 Performance : DC 0.040 (0.371) Performance: IO 0.635 (0.479) BG: 2 -0.129 (0.174) BG: 3 -0.139 (0.191) BG: 4 0.326 (0.139) BG: 5 0.330 (0.208) BG: 6 -0.668 (0.254) BG: 7 0.505 (0.177) BG: 8 0.145 (0.141) Historic 0.975 (0.153) M x DC -0.095 (0.464) H x DC -0.127 (0.484) VH x DC 0.325 (0.395) M x IO -0.512 (0.560) H x IO -0.336 (0.581) VH x IO 0.271 (0.500) 6 Building occupancy condition: In-place (IP) and Temporarily removed (TR) during retrofit work 7 Seismicity of building location: Very high (VH), High (H) and Medium (M) 8 Performance objective: Immediate occupancy (IO) and Damage control (DC) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 80 Constant 50.400 (35.400) Observations 812 2 R 0.321 2 Adjusted R 0.300 Table 3-28. Estimates of coefficients and variations from training main model (NIST, 2017) 3.8.3.2 Cost-benefit analysis of seismic retrofit for URM buildings A cost-benefit analysis for the seismic retrofit of URM buildings in Portland was performed to support the city’s decision-making on a seismic retrofit ordinance. Existing URM buildings are usually outdated with insufficient structural seismic capacity and are therefore very vulnerable to earthquake hazards. In Portland, 1,661 URM buildings existed at that time, as mapped in Figure 3-59. In the analysis, the URM buildings were classified into six classes according to occupancy and performance level; see Table 3-29. Class 1 buildings are the most critical buildings in an earthquake event, and Class 5 buildings are not essential facilities in post-earthquake circumstances and their performance objective is thus at the lowest level. The retrofit costs applied in the study considered the construction costs, relocation costs and usual soft costs (e.g., design fee, etc.) for each building class, as listed in Table 3-30. For the benefit evaluation, the reduced damage amount of buildings and contents, the decreased number of human losses and the diminished displacement cost (i.e., cost for evacuation and moving due to building damage) because of seismic retrofitting were converted to monetary values and considered in the analysis. In the study, the cost-benefit ratio was calculated as the net present value of future benefits divided by the seismic retrofit cost. The analysis was performed for a 50-year building lifetime and used the average annual damages and losses of both the existing and retrofitted buildings. The ratios of benefit-to-cost for seismic retrofits for each building class were then analyzed, as summarized in Table 3-30. All calculated ratios are larger than 1.0, which means that seismic retrofit is beneficial for all building classes. There were no clear explanations for the “N/A” specification for class 1 buildings in the original document, but it can be assumed, based on engineering judgement, that seismic retrofit was undoubtedly profitable for the class 1 buildings because they are the most critical buildings during an earthquake disaster. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 81 Figure 3-59. Geographic distribution of URM buildings (purple dot) in Portland (Goettel, 2016) URM class Count Percent Simplified definition Seismic performance level (simplified) Class 1 10 0.60% Critical facilities Comparable to a new critical facility Damage control for 20% in 50-year ground Class 2 92 5.54% Schools and public assembly buildings motion and life safety for 5% in 50-year ground motion Buildings with ≥4 stories, or ≥300 Life safety for 20% in 50-year ground motion Class 3 220 13.25% occupants or residential buildings with and collapse prevention for 5% in 50-year ≥100 units ground motion Class 4-A “Ordinary buildings” not in the other Same as Class 3, unless exception applies 1,126 67.79% Class 4-B classes Very similar to Class 5, if exception applies 1- and 2-story buildings with 0-10 Bolts Plus, retrofit of only specified building Class 5 213 12.82% occupants elements Table 3-29. URM building classes and performance levels in Portland (Goettel, 2016) Estimated retrofit cost per square Typical building benefit-cost Portland URM class foot (in USD) ratio Class 1 US$111.45 N/A Class 2 US$82.62 1.474 Class 3 US$68.77 1.661 Class 4-A US$68.77 1.661 Class 4-B US$51.00 1.967 Class 5 US$20.00 1.940 Table 3-30. Building group and types and benefit-cost ratios for the prediction model (Goettel, 2016) 3.8.3.3 Retrofit cost consideration in FEMA documents In documents published by FEMA (FEMA, 2006b)(FEMA, 2007a)(FEMA, 2012)(FEMA, 2013d), there are considerations of the retrofit cost, disruption matter and construction approach based on U.S. practices. The summarized commentaries for typical strengthening measures introduced in this study are presented in Table 3-31. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 82 Retrofit Option Description Diagonal bracing is usually less expensive but is considered less reliable than vertical bracing. Exposed braces are typically Wall out-of- Diagonal or vertical less expensive than more architecturally-sensitive alternatives like recessed vertical braces or reinforced cores. Installation of plane bracing bracing is fairly disruptive, since it must occur around the entire perimeter; and it involves drilled dowels and accessing and strengthening (FEMA 2006b) connecting to horizontal diaphragms. In general, shotcrete walls are less expensive than cast-in-place concrete because at least one side of the wall forming is eliminated. Construction of new shear walls in an existing building can be very disruptive to any building occupants. Noise, RC shear wall vibration, and dust associated with many operations, especially cutting holes through and drilling dowels into concrete, can be (FEMA 2006b) transmitted throughout a concrete structure. Placing cast-in-place concrete or shotcrete is a wet process and very messy. Also, excavation and drilling operations and the use of mechanized and/or truck mounted equipment associated with installation of new foundations can be very disruptive Costs will be less when existing moment frames are converted into braced frames to take advantage of the existing strength and stiffness of the frame members, connections, and foundations. Designs that are simple and details that are not overly Steel brace complicated will also minimize costs. Costs can also be reduced if disruption is minimal during construction. Installing braces (FEMA 2006b) at the perimeter frames reduces logistical issues associated with working in confined spaces and temporary removal of the nonstructural elements. Noise associated with this type of work is loud and disturbing to the tenants if the building is occupied In-plane while the work is being performed. strengthening, Unstiffened walls are cheaper and less labor intensive than stiffened walls. New foundations are almost always required for Structural new walls and could be extremely costly if deep foundations, such as drilled piers, are added. Installing new walls is disruptive irregularity, Steel plate wall to the occupants because of the noise and vibrations associated with construction. Even if tenants are relocated to parts of the Stiffness and (FEMA 2006b) building where the work is not being performed, vibrations associated with cutting, chipping, and drilling of concrete as well Capacity as the installation of steel panels can transmit through the structure. The disruption can be reduced somewhat if the walls are improvement installed at the perimeter In addition to the costs of material and installation of FRP composite, concrete surface preparation and final appearance requirements must also be considered. If additional surface finishing is required, such as paint or a cementitious appearance, this can add significant cost, and should be coordinated with the architect and owner. Although the basic materials are generic, FRP sheet overlay the fabric that will be supplied is proprietary. Strand orientation and density, epoxy overlays, preparation requirements, and (FEMA 2006b) installation procedures may be different between suppliers. Some suppliers may not have experienced applicators in the area of the project. These variations must be considered to achieve an adequate specification, particularly if competition between suppliers is desirable. Adding new concrete, particularly with shotcrete, can be quite disruptive. Where access is sufficient, shotcrete is typically RC wall overlay chosen as it is less expensive than cast-in-place work which requires front-side formwork. There are no known proprietary (FEMA 2006b) concerns with shotcrete or cast-in-place overlays on existing masonry walls. Adding a concrete or steel jacket is a more traditional method of enhancing a deficient concrete column. Because of the Concrete/Steel difficulty of placing an overlay on all sides of an existing column, concrete overlays are typically done with cast-in-place Ductility jacketing on RC concrete, rather than shotcrete. The need for formwork is a significant disadvantage for concrete overlays, compared to FRP improvement column and steel overlays. Placing the concrete and vibrating are also challenging due to access limitations at the top of the column (FEMA 2006b) where it runs into beams or slabs. Pour ports or holes in the diaphragm are needed. Steel overlays are typically 3/16” or 1/4” thick and become quite heavy. Access and lifting issues in existing buildings can force the overlay to be broken down into Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 83 Retrofit Option Description pieces, increasing the amount of field welding necessary to join the pieces back together. Unlike FRP overlays, no proprietary issues have been identified with using concrete of steel jackets. To appropriately evaluate the cost of a retrofit scheme using FRP overlay in comparison to traditional retrofit concepts (such as concrete or steel jacketing), one needs to consider the cost of the raw material, the level of specialization required by the FRP sheet overlay on contractor to install the system, the cost of labor and equipment, the cost of quality control and quality assurance, the temporary RC column impact of disruption during construction, and the permanent impact to the building functions. Although FRP overlays are (FEMA 2006b) relatively expensive compared to steel and concrete, they can offer advantages when only limited access is available or minimal disruption of existing conditions is desired. Connection modifications are locally very disruptive. Modification of a connection typically requires access from two floors Steel beam to perform the work on each flange. Noise associated with this type of work will spread and disrupt tenants on other floors connection unless the work is done during off-hours. The RBS (Reduced Beam Section) is probably cheaper than the other modifications (FEMA 2006b) since it requires the least amount of material and labor. With older buildings, there may be asbestos present in the fireproofing around the steel members, which could require costly abatements to expose the connections. Detailing for rubber and sliding bearings (i.e., isolator) is quite different. If multiple vendors are necessary, vendors of different types of systems are usually considered. Often, they are procured in an early package, due to the long lead time. This also permits the design engineer to move into final design knowing which type of system will be used. In addition to the bearings and dampers, a complete isolation system will require a number of other special elements, including a moat around the building Seismic isolation to accommodate the displacements. The moat has to go down past the plane of isolation. Elevators are typically hung from the (FEMA 2006b) superstructure, as they cannot cross the isolation plane without special detailing. Utilities entering the building need to be able to accommodate the isolation displacements; this often triggers special vaults outside the building or areas under the building Energy for joint details. A foundation is needed below the isolators to take the forces they impart, and a structural system is needed dissipation above the isolators as well to deliver forces to the isolators and resist the moments that are induced. All of these elements add system to the cost of isolating the building. Most dampers available on the market are proprietary. Material properties, testing histories, limitations and detailing considerations are obtained from the manufacturer. Like seismic isolation components, the particular category of damper such Energy dissipation as a fluid viscous damper or a friction damper is usually selected early in the design because the analysis and detailing can be damper significantly different between categories. There is also a patent regarding certain techniques for connecting bracing and (FEMA 2006b) dampers to beams when sliding is employed. The damper must be connected to the existing structure and potentially the foundation. When dampers are added to the structure, the loads they impart locally must be considered in the design. Enlarging or replacing an existing footing is a localized but disruptive process, involving excavation, dust, mud, Enlarge shallow drilling/jackhammering noise and concrete placement. Protection of existing finishes in the vicinity and in the working path is foundation necessary. If the existing footing is replaced, shoring will be needed. It is critical that the base of footing be properly compacted, (FEMA 2006b) and the new concrete be tightly placed beneath the existing column to minimize or eliminate any settlement when the shores Foundation are removed. retrofit Adding new drilled piers in an existing building is very disruptive, involving excavation, dust, mud, drilling/jackhammering Add pile deep noise and concrete placement. Protection of existing finishes in the vicinity and in the working path is necessary. Drilling foundation limitations for drilled piers may need to be applied according to site environment. Drilled piers typically have spacing limits (FEMA 2006b) of three times the pier diameter to avoid group effect reductions. This can limit the number of piers that can be installed. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 84 Retrofit Option Description Grouting can be quite disruptive and costly especially if injection holes have to be drilled through existing concrete footings and slabs. If current operations in the building are to continue, the usual approach is to do the grouting at night and clean up Grouting around the work area before the start of work the next morning. Grout is usually injected in a strict primary-secondary pattern. existing foundation Alternate primary holes are drilled and grouted first followed by the secondary holes. The level of solidification achieved is (FEMA 2006b) Soil verified by exhuming grouted soil bulbs, taking samples of the grouted soil and performing unconfined compression tests on improvement the samples. The potential for lateral spreading can be mitigated by creating a stable mass of material near the open face of site. This can Add stone/gravel be accomplished by either densifying the layer of potentially liquefiable soil or solidifying the soil to prevent liquefaction. One column of the techniques used to achieve this goal is a vibro compaction methods involving the installation of stone columns can be (FEMA 2006b) used to densify the potentially liquefiable layer Considerations for cost depend on the number, type and depth of dowels; the difficulty of access; and the extent of finishes URM building that are impacted. Through bolts are usually less expensive than adhesive anchors. Drilling is loud and can be disruptive to (FEMA 2006b) occupants. Typically, either the floor or ceiling has to be removed to install the dowels. Thus, it is usually not practical to install Roof dowels in occupied rooms, though the work can be phased by building area so disruption is minimized. connection In new construction, attachment of roof or floor sheathing to shear walls below typically requires nailing through the sheathing improvement Wood building into framing below. While this attachment remains the preferred approach, installation of nailing is not possible where roof or (FEMA 2006b) floor finishes cannot be removed. Fasteners connecting sheathing to framing should not be overdriven (not break the face of the sheathing). Where overdriving occurs, fastener capacity may be reduced up to 40%. The connections must be detailed with sufficient ductility and rotation capacity to prevent failure. Typically, the panels are seated on two bearing connections at either the top or bottom floor and then hav e “push-pull” connections at the adjacent floor External curtain wall which resist out-of-plane loading but move laterally in the plane of the panel. While both the rocking and sway mechanisms (FEMA 2012) are effective methods of accommodating story drift, mixing mechanisms in adjacent panels may potentially close the joints External wall between units, resulting in a collision between adjacent panels. To avoid this, the joints between panels have to be sized to connection allow for panel rotation and translation. improvement Connections are designed to transfer uplift loads applied to the roof, and the positive and negative loads applied to the exterior External non- bearing walls, down to the foundation and into the ground. The roof covering (and wall covering, if there is one) is also part structural masonry of the load path. To avoid blow-off, the nonstructural elements must also be adequately attached to the structure. Because wall (FEMA 2007a) some locales have very aggressive atmospheric corrosion (such as areas near oceans), special attention needs to be given to the specification of adequate protection for ferrous metals, or to specify alternative metals such as stainless steel. The cost of a column connection to foundation upgrade is not very expensive. However, this technique is not usually performed Steel building only by itself. Costs would be small relative to an overall lateral force-resisting system upgrade and a foundation upgrade. (FEMA 2006b) Disruption could be minimal since typically, there are no tenants in the basement. Even if the basement were used for tenant Foundation access, such as parking, only a few columns would be affected at a time. connection Addition of foundation anchorage in a crawl space with minimum required code vertical clearance is difficult due to very improvement Wood building cramped conditions; work areas are often hard to get to, let alone getting tools and supplies and executing work. New temporary (FEMA 2006b) access openings and disconnection of HVAC ducting may occasionally be needed to provide access to work. There are no proprietary concerns with this rehabilitation technique, other than the use of proprietary connectors as part of the assemblage. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 85 Retrofit Option Description Flood shields or panels are watertight structural systems that bridge the openings in walls to prevent the entry of floodwaters. Flood shields work in tandem with waterproof barriers to resist water penetration. Flood shields transfer food-induced forces into the adjacent structural components and, like sealants, can overstress the structural capabilities of the building. A number Building Flood shields or of vendors make special doors for permanent installation and drop-in panels or barriers that are designed to be watertight and envelope panels installed as needed for food protection. Heavy food shields for larger openings may require electrical or mechanical systems improvement (FEMA 2013d) to move them into place. Designs should include a consideration of the fact that power may not be available following the event and that access through the opening may be necessary to restore power. A maintenance plan is an important part of a flood proofing system that includes food shields. Flood shields, hinges, hoists, latches, channels, and seals should be inspected periodically for damage Permanent floodwall is the elevated barriers that provide food protection to one or more buildings. It provides structural protection for shallow food depths; floodwall can provide effective food protection to buildings that experience food levels of 4 feet or greater. A floodwall is a freestanding, permanent, engineered structure designed to prevent encroachment of Elevated floodwall floodwaters. Existing building foundation conditions may affect the floodwall design directly. For example, evidence of construction (FEMA 2013d) seepage or cracking in foundation walls may indicate the need to relieve hydrostatic pressure on the foundation as part of the floodwall design. Floodwall design depends primarily on the type of flooding expected at the building site. High water levels and velocities can exert significant hydrostatic and hydrodynamic forces that must be accounted for in the floodwall design For wood building, it is most common to use adhesive anchors for anchorage of the vertical tie-down bolt to the foundation. Foundation- It has been common to install the adhesive anchor straight down into the footing, or at a very slight angle if required for access. Adhesive anchors tie Adhesive anchors must be installed in accordance with the manufacturer’s recommendations and the applicable ICC Evaluation (FEMA 2006b) improvement Services report recommendations. Tie-down connectors should be attached to substantial existing footings that have the shear and flexural capacity to mobilize required resistance. Alternately, new footings or footing reinforcement can be provided. Table 3-31. Summary of retrofit solutions for buildings for natural hazards Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 86 3.8.4 Life-cycle cost For essential building infrastructure, life-cycle cost analysis (LCCA) should be considered in plan and design. This topic will be further discussed in Chapter 8. 3.9 Discussion Multiple natural disasters are common in the subject countries. Large earthquakes happen occasionally, while hurricane and flood occur more frequently. Throughout all stages of disaster, from the event itself to the end, full recovery, buildings like emergency centers, government buildings, hospitals, schools and hotels provide critical aid in response to needs of rescue, in restoring communities, treating injuries and accommodating displaced people. For these essential missions, the structural resiliency of these buildings against multiple disasters is a vital factor. If these types of buildings are not functional due to lack of disaster resistance, and one of the above essential missions cannot be executed, it could induce a secondary disaster that would have otherwise been avoidable. The implementation of resilience measures against natural disasters is then an urgent matter, not only for disaster damage reduction, but also for disaster response. A summary of the resilience measures, associated cost and vulnerability reduction according to critical hazard and the vulnerable components of buildings is presented in Table 3-32. The improvement cost is an average estimation of the onetime implementation of the resilience measure. Vulnerability reduction was evaluated based on the likelihood of physical damage occurrence; any functional outages or indirect losses were not considered. As seen in the table, the vulnerabilities of critical components are largely reduced by strengthening so much so that the effect of the resilience measure is obvious. For future feasibility studies, functional and indirect impacts, in addition to physical damage, should be considered for a cost-benefit analysis of building strengthening, especially for critical buildings (i.e., essential facilities) under a natural disaster. It is also recommended that multiple disaster events during an expected building’s service lifetime be taken into account, because those buildings are typically used for longer periods than ordinary buildings. Cost, % of initial cost (Est.) Vulnerable Subsector Hazard Resilience Measure Improvement Vulnerability Component cost reduction Add wall bracing (anchor and Wall out of plane 15% 40% strongback) In plane Add RC shear wall or Steel brace 20% 60% Structural Add Shear wall, Moment frame or 15% 40% irregularity Braced frame Earthquake Add Column jacketing or Beam shaking Ductility 10% 30% retrofit Schools, Stiffness and Add Steel brace or Steel plate 15% 50% Hospitals, capacity shear wall Government Energy Add Seismic isolation or Energy 40% 80% buildings, absorption dissipation damper Emergency Enlarge Spread footing or Add Foundation 40% 80% centers, Earthquake pile foundation Hospitality liquefaction Install Soil grouting or and hotels Soil 30% 80% Stone/gravel columns Roof connection Improve Roof connection anchor 10% 50% External curtain Improve Wall connection 10% 30% Wind wall and wall attachment Foundation Add Connection anchor bolts and 10% 50% connection stiffeners Building Flood Install flood shield 10% 80% envelope Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 87 Elevated Elevate floodwall 15% 80% construction Strengthen anchoring to Foundation tie 10% 50% foundation Table 3-32. Summary of findings for building infrastructure in the Caribbean Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 88 4. WATER INFRASTRUCTURE 4.1 Introduction Improvement in resiliency of water infrastructure (see Figure 4-1) can have a multi-fold beneficial impact for the Caribbean. By reducing physical damage to water infrastructure components, i) economic losses are reduced; ii) secondary impacts, such as flooding of roadways as a result of water pipes bursting, are reduced; and iii) maintaining the delivery of potable (“drinking”) water, coupled with treatment and discharge of wastewater, accelerates recovery and allows for operation of critical emergency facilities, such as hospitals and response centers. Water treatment plants (WTP), wastewater treatment plants (WWTP), and underground potable water pipelines (UGP) are critical infrastructure lifelines and will be emphasized in this chapter. Figure 4-1. Water (and wastewater) infrastructure (T: PMC, 2020; B: SafeRack, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 89 For water infrastructure in the Caribbean, the hazard and infrastructure components listed in Table 4-1 will be used. As noted in the table, for WTP and WWTP, earthquakes, and to a lesser extent, hurricanes and flooding, are critical, whereas for UGP systems, ground movement and liquefaction are the most critical hazard. It is assumed the discussion presented herein applies to all countries in the Caribbean, as water sector construction is likely similar across the subject countries due to significant cooperation between utility providers. Sector Subsector Type Component Hazard Wall Earthquake Ductility Earthquake Plant building Roof Wind (Masonry/RC External wall Wind Water treatment frame) Opening Flood plants Elevation Flood Foundation Earthquake Wastewater Water storage Anchorage Earthquake treatment plants Earthquake, Equipment Foundation Wind (mechanical and Elevation Flood WATER electrical) Flood proofing Flood All Foundation Liquefaction Earthquake, Brittle pipes (Asbestos-Cement, Pipe liquefaction, Underground Cast Iron, etc.) flood pipelines Earthquake, Joints Conventional liquefaction Table 4-1. Analysis matrix for water infrastructure In this report, the main water pipes (larger diameter) transferring water from WTP to property junctions will be discussed. The secondary pipes (smaller diameter) carrying water to individual structures are not considered. Figure 4-2 shows the distinction between main and secondary pipelines. Main water pipes Secondary water pipes Figure 4-2. Main and secondary pipes (various) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 90 4.2 Geographical distribution In the subject countries, the following water utility companies are responsible for construction, maintenance, and upgrading of water infrastructure; see Table 4-2. Country Utility Suriname N.V. Surinaamsche Waterleiding Maatschappij Trinidad and Tobago The Water and Sewerage Authority of Trinidad and Tobago (WASA) Guyana Guyana Water Incorporated (GWI), Severn Trent Water International (STWI) Belize Belize Water Services (BWS) National Directorate for Water Supply and Sanitation in the Ministry of Haiti Public Works (Direction Nationale d'Eau Potable et d'Assainissement, DINEPA) Regional companies: the Santo Domingo Water and Sewerage Corporation (CAASD), the Santiago Water and Sewerage Company (CORAASAN), the Dominican Republic Puerto Plata Water and Sewerage Company (CORAAPPLATA), the Moca Water and Sewerage Company (CORAMOCA), the Romana Water and Sewerage Company (CORAAROM) Antigua and Barbuda The Caribbean Water and Sewerage Association Dominica Dominica Water and Sewerage Company Limited (DOWASCO) Grenada Grenada’s National Water and Sewerage Authority (NAWASA) Saint Kitts and Nevis The Water Department works under the Ministry of Public Works Saint Lucia The government-owned Water and Sewerage Company Saint Vincent and the Grenadines The Central Water and Sewerage Authority Sint Maarten NV GEBE Barbados Barbados Water Authority Limited Jamaica The National Water Commission (NWC) Bahamas The Bahamas Water and Sewerage Corporation (WSC) Table 4-2. Water utility providers for 16 subject countries Examples of water distribution maps in the subject countries are presented in Figure 4-3. As discussed previously, the report will focus on water treatment plants (WTP), wastewater treatment plants (WWTP), and underground potable water pipeline (UGP) systems in the Caribbean. Figure 4-4 shows the location of WTP and WWTP in a number of subject countries. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 91 Water & Wastewater Treatment Plants in Antigua and Barbuda (Schweikert et al., 2020) Water & Wastewater Treatment Plants in the Dominican Republic (Schweikert et al., 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 92 Water & Wastewater Treatment Plants in Grenada (Schweikert et al., 2020) Water & Wastewater Treatment Plants in Jamaica (Schweikert et al., 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 93 Water & Wastewater Treatment Plants in Trinidad and Tobago (Schweikert et al., 2020) Water intake locations in Saint Lucia (Gosine, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 94 Saint Lucia (WBG, 2020) Jamaica (Aquastat, 2015) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 95 Trinidad and Tobago (Tota-Maharaj and Meeroff, 2013) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 96 Dominica (Dominica News, 2017) Figure 4-3. Water distribution networks Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 97 Haiti Domincan Republic Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 98 Belize Jamaica Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 99 Trinidad and Tobago Guyana Suriname Figure 4-4. Typical locations of WTP and WWTP in subject countries Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 100 Table 4-3 presents data regarding the water infrastructure for selected countries. In recent years, upgrades to water sector components in several countries have been undertaken or are planned. Examples include: • Dominican Republic (Hazen, 2017): To provide safe drinking water to the urban service areas of La Romana, San Cristobal, and San Francisco de Macoris, the National Institute of Water and Sewage of the Dominican Republic (!NAPA) retained the Biwater/Sinercon Consortium for the Design- Build delivery of three 87,000 m3/d Water Treatment Plants and associated infrastructure. In turn, Biwater U.S.A. retained Hazen and Sawyer to provide engineering design and support in the procurement management services for the project. The project consisted of building three water treatment plants (The Mata Larga Plant, La Romana Plant and the San Cristobal Plant), to serve approximately 500,000 people, each of which is rated at 1 m3/s (23 mgd) capacity. Other supporting infrastructure development included nine water storage tanks, seven pump stations with capacities of up to 91,000 m3/d, and 130 km of ductile iron raw water and transmission mains, ranging in size from 200 to 1,000 mm in diameter. • Dominican Republic (Waterworld, 2009): The country's national aqueduct and sewerage authority INAPA has awarded ACCIONA's water treatment services arm a 74 million euro contract for the construction and commissioning of a drinking water plant (DWTP) for the Republic's southern Peravia province, with a capacity of 86,400m3/day. The contract also includes the works for a 17.5km adduction line and a water supply network for the surrounding towns and villages. The project deadline is two years. • Dominica (DOWASCO, 2020): The Government of Dominica has secured funding to the tune of twenty-one million dollars from the European Union to undertake a major project to significantly improve the water supply to residents along the west coast of Dominica. The Government itself is also investing a further four million dollars in the project, bringing the total cost to twenty-five million dollars EC. Pipe laying works have been completed along most of the highway between Salisbury to Portsmouth while excavations and pipe laying works commenced from the Cabrits junction towards Capuchin in February 2012 and are expected to continue until June 2012. So far over 20 Km of pipelines have been laid. Lot Three will see the construction of the nine new storage • Jamaica (NWC, 2020): The National Water Commission (NWC) says the supply of water to several communities in the Corporate Area should improve with the new Ferry 600-millimetre diameter pipeline. The new pipeline replaces the previous 18-inch diameter pipeline, popularly known as the Ferry 18-inch. NWC said it serves communities along Washington Boulevard up to Dunrobin Avenue, where it interconnects with the recently commissioned 700- millimetre diameter transmission main that was laid along Constant Spring Road. • Guyana (GWI, 2020) There were several breakages on Vlissengen Road and Church Street, Georgetown. However, the Agency received resources to purchase new HDP Pipes, which will be replaced. In the 2020 Budget, there will be a replacement of the transmission line for Cemetery Road, Georgetown, since the area is a critical artery for the supply of water to the City. Other areas in the City, such as Kitty, Kingston and Albouystown will get lines replacement after the operation is executed at Cemetery Road, Georgetown. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 101 • St Lucia (WASCO, 2020) Construction of a reinforced concrete Raw Water Intake in the Tournesse River; Construction of a Water Treatment Facility and pumping station for a minimum production of 0.4 Million Imperial Gallons per Day of drinking water, meeting the World Health Organization requirements, Installation of 2.5km of ductile iron 300mm diameter pipe (buried) from the Raw Water Intake to the Treatment Plant, Installation of 2 km of High Density Polyethylene (HDPE) as transmission pipeline to various communities in DenneryNorth. Capacity, MGD Country (Millions of WTP WWTP UGP, km gallons/day) Guyana >18 Antigua and Barbuda 7.0 3 (desalinization), 2 Dominica 10 (river-based) Grenada 100 Saint Kitts and Nevis 2.2 300 Barbados 35 2 3,200 Jamaica 190 1,000 100 4,000 Bahamas Table 4-3. Examples of water sector components (various) 4.3 Typology Examples of WTP in the subject countries are presented in Figure 4-5. Figure 4-6 presents examples of WWTP. Figure 4-7 presents example of pipelines in the subject countries. Trinidad and Tobago WTP (WASA, 2020) Trinidad and Tobago WTP (Desalinazation), (WASA, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 102 Desrruisseaux, (St. Lucia) WTP 860 m3/day WTP (WASCO, 2017) Building Equipment Blue Hills (Bahamas) 45,000 m3 WTP (Consolidated Water, 2020) Buildings and water tanks Filter basin San Cristobal (Dominican Republic) 1 m3/sec WTP (Hazen, 2020) Saint Vincent and Grenadine WTP (CWSA, 2020) Sint Maarten desalinization plant (SMN-N 2012) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 103 Antigua WTP (Desalinazation), (APUA, 2020) Figure 4-5. WTP in subject countries Sedimentation pond Soapberry (Jamaica) 60 MGD WWTP (various) Structure Facility Bridgeport and South Coast WWTP (Barbados) (BWA, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 104 Trinidad and Tobago WWTP (WASA, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 105 Saint Vincent and Grenadine WWTP (CWSA, 2020) Guyana WWTP (GWI, 2020) Antigua Antigua WWTP (APUA, 2020) Figure 4-6. WWTP in subject countries Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 106 2-m concrete pipe; Dominican Republic (Hazen, 2020) 0.6-m pipe Jamaica (NWC, 2020) 0.15 m Haiti Saint Kitts and Nevis 0.6 m St Luicia 0.6 m pipe Bahamas (Various) 0.3 pipe Saint Vincent and the Grenadines (CWSA, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 107 0.35-m pipe Guyana (GWI, 2020) Figure 4-7. Water pipelines in subject countries 4.4 Vulnerable components Past natural hazards have shown that water infrastructure components are vulnerable to various natural hazards. Certain components have been known to experience frequent failures. Strengthening of these components would lead to marked improvement in the performance and as such, these components are emphasized in this chapter. The damage assessment conducted by the National Water and Sewerage Authority (NAWASA) in the aftermath of Hurricane Ivan in 2004 is presented in Table 4-4 (OECS, 2004). Table 4-5 and Table 4-6 present examples of damage to water infrastructure in Dominica as a result of recent hurricanes and ensuing flooding. Table 4-7 lists the damage to water infrastructure in Saint Vincent and the Grenadines due to flooding. Components US$M Buildings 1.0 Water plants 0.6 Pipelines 1.0 Total 2.6 Table 4-4. Damage to Grenada’s water infrastructure, Hurricane Ivan (OECS, 2004) Loss components US$M Damage 14.5 Losses 2.3 Total 16.8 Table 4-5. Damage to Dominica’s water infrastructure, Hurricane Erika (ACP-EU, 2015) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 108 Subsector Components US$M Water supply Production and distribution 19.9 Sanitation List stations, mains, connections 3.6 Table 4-6. Damage to Dominica’s water infrastructure, Hurricane Maria (ACP-EU, 2017) Loss components US$M Damage 0.25 Losses 0.03 Total 0.28 Table 4-7. Damage to Saint Vincent and the Grenadines water infrastructure, 2016 floods (ACP-EU, 2016) Water infrastructure is vulnerable to damage from earthquakes; see Figure 4-8. Examples of natural damage to vulnerable water infrastructure components in the subject countries are presented in Figure 4-9. WTP WTP Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 109 Calexico WWTP WWTP Pipelines Figure 4-8. Water infrastructure (WTP and WWTP) damage, earthquake hazard (EERI, 2010) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 110 Damaged water infrastructure - Dominica Broken pipeline - Haiti Dominica Guyana Figure 4-9. Damage to pipelines (various) Table 4-8 presents the components that have experienced the most damage in past events and will be considered in this chapter. Subsector Type Component Vulnerability Hazard Water treatment Out-of-plane capacity Earthquake Wall plant In-plane capacity Earthquake Plant building Ductility Non-ductile detailing Earthquake (Masonry/RC frame) Wastewater Roof Roof connection Wind treatment plant External wall External wall Wind Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 111 Opening Building envelope Flood Elevation Elevated construction Flood Foundation Foundation Earthquake Water storage Anchorage Anchorage Earthquake Equipment Foundation Foundation tie Earthquake, Wind (mechanical and Elevation Elevating equipment Flood electrical) Floodproofing Dry floodproofing Flood All Foundation Inadequate capacity Liquefaction Brittle pipes Earthquake, Pipe (asbestos-cement, Brittle pipes Liquefaction, Flood Pipelines cast iron, etc.) Limited deformation Earthquake, Joints Conventional capacity liquefaction Table 4-8. Vulnerable components of water infrastructure and corresponding hazards 4.5 Design practice 4.5.1 Overview It is anticipated that the buildings and equipment for WTP/WWTP in the Caribbean are designed and constructed by utilizing international design codes or the Caribbean building codes discussed in Section 3.5 because design codes specialized for WTP/WWTP structural design cannot be found in the Caribbean countries. In fact, a description of WWTP in Barbados mentions that the plant was designed abroad by a foreign consultant (UNEP, 1997); likewise, information for WTP/WWTP construction in Suriname states that the construction of water supply facilities should refer to the country codes in addition to the codes of the U.S. or Europe (WHO, 1987). The only design guideline for WTP/WWTP is found in Trinidad and Tobago, and it is introduced in Section 4.5.5. In this guideline, similar to the above descriptions, the structural design of WTP/WWTP buildings basically refers to the building code used in Trinidad and Tobago. Therefore, in the following subsections, the design approaches specified in local building codes (see also Section 3.5) and the U.S. codes for water pipelines are briefly summarized. 4.5.2 Earthquake design In the Caribbean Uniform Building Code, CUBiC (1985), the seismic design methodology for buildings in the Caribbean is specified and could be applied to the design of WTP/WWTP buildings. The design earthquake force is prescribed according to regional seismicity, ductility factor, occupancy importance factor, vibration characteristic and building mass. The structural elements of buildings are then designed to meet the code requirements to resist the earthquake force. For WTP/WWTP plants, the highest class of occupancy importance factor (i.e., 1.5; see Table 3-6) should be taken and then be designed by the larger seismic design force (i.e., 50% larger than a normal building) for better performance in the aftermath of an earthquake. For water pipelines and equipment (e.g., mechanical facility, nonstructural components, etc.), a seismic guideline for water pipelines in the U.S. (ALA, 2005) and a U.S. guideline for reducing seismic damage of nonstructural components (FEMA, 2012) could be utilized for designing those in the Caribbean, respectively. 4.5.3 Wind design The wind design for buildings is established for the Caribbean countries in CUBiC (1985). In the simplified procedure of three wind design methodologies, the design wind pressure can be calculated by the reference wind velocity pressure, exposure factor, aerodynamic shape factor and dynamic response factor (see Section 3.5.2.1.1). The reference wind velocity is based on a 50-year return period and is based on country wind hazard intensity. The force, instability, deflection and fatigue of building elements are then estimated and designed by applying the design wind force. In the wind design component of CUBiC, any kind of occupancy importance factor cannot be found. For water pipelines, wind hazard essentially does not adversely affect the pipelines. For mechanical facilities of WTP/WWTP plants, the large overturning, Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 112 excessive deflection and base anchoring are major concerns caused by wind hazard. The design procedure on nonstructural components of ASCE 7-16 (ASCE, 2017a) could be applied in the Caribbean. 4.5.4 Flood design As described in Section 3.5.3, the design approach for flood prescribed in ASCE 7-16 (ASCE, 2017a) could be applied for the Caribbean because the International Building Code (ICC 2017 – widely used in the Caribbean) and ASCE 24-14 (ASCE 2015 – document focusing on flood design) refer to ASCE 7-16. Since the base flood for building design is the flood having a 1% chance of being equaled or exceeded in any given year in the code, the Caribbean-specific flood hazard needs to be verified with the base flood prescribed in ASCE 7-16. For flood protection of mechanical facilities and access roads, the 100-year flood elevation or maximum flood of record is usually used to confirm the safety of facilities (PRHEM, 2012). 4.5.5 WTP/WWTP design guideline manual in Trinidad and Tobago (WASA, 2008) 4.5.5.1 Acts, codes and standards In a section of the Water and Wastewater Design Guideline Manual in Trinidad and Tobago (WASA, 2008), compliance with regulations, acts, codes, standards and guidelines is specified for the design of water infrastructure. The main regulations are summarized below. • Environmental Management Authority (EMA) of Trinidad and Tobago • Water And Sewerage Authority (WASA) of Trinidad and Tobago • Ministry of Public Utilities and Environment (MoPUE) • National Building Code • National Fire Code • OSHA (Occupational Safety and Health Act) • Trinidad and Tobago Standards (TTBS) • International Standards (ISO) • North American Standards (ANSI/AWWA) • British Standards (BS) 4.5.5.2 Structural design considerations It is assumed that the design details for water facilities should refer to the National Building Code in Trinidad and Tobago, but general descriptions for the structural design of treatment plants are established in this document as follows: • The plant shall be sufficient to accommodate anticipated ground moments, including any active geologic faults; • The designs must assure that the structures are to sustain regional earthquakes and hurricane events; • The seismic design of civil structures will be generally based on an earthquake zone 3, as defined in UBC 1997 (Uniform Building Code); • For flood hazard, all WASA buildings or structures shall be flood-proof, and the site for the new facility shall be appropriately selected or designed to be above the 20-year recurrence interval flood line. 4.5.5.3 Construction requirements The requirements for construction materials are also specified in the design guideline and are summarized below. • Concrete: the minimum compressive strength of concrete, based on 28 days compressive strength as determined by tests on concrete cylinders according to ASTM International (ASTM), is listed in Table 4-9. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 113 • Steel reinforcement: the reinforcement bars shall be deformed, new steel bars complying with the “Specifications for Billet Steel Bars for Concrete Reinforcement” ASTM A615 Grade 60 or approved equal, with a minimum yield strength of fy = 400 MPa (i.e., 60 ksi). • Structural steel: the structural steel shall conform with international standards and with the following ASTM standards. o ASTM A369 A36M – Specifications for Structural Steel o ASTM – A193A & 193M – Specifications for Alloy-Steel and Stainless Steel for Bolting Material for High temperature service. o ASTM – A307, A325 & A490 – Specifications for Structural Steel Bolts and Bolted Joints. Concrete compressive strength Type of construction (N/mm2) (psi) Unreinforced concrete, Lean concrete 10 1,450 Encasement, Duct banks, Cast in place, Concrete curbs 25 3,600 Super structures 35 5,000 Sub-structures 35 5,000 Table 4-9. Required compressive strength for concrete construction 4.6 Fragility functions 4.6.1 Earthquake fragility functions based on FEMA Hazus methodology 4.6.1.1 Overview In this section, physical (structural) damage fragility relations for WTP, WWTP, and UGP types common in the Caribbean are presented. For the structures, two types of design are considered: conventional (non- resilient design constructed prior to adoption of seismic codes) and enhanced (resilient construction). 4.6.1.2 WTP 4.6.1.2.1 Damage states FEMA Hazus (FEMA, 2003a) defines a number of damage states for WTP. The damage states relevant to the Caribbean are summarized in Table 4-10. Damage State Description state DS1 None No observable damage Malfunction of plant for a short time (less than three days) due to loss of electric power and backup power if any, considerable damage to various equipment, light damage to DS2 Slight sedimentation basins, light damage to chlorination tanks, or light damage to chemical tanks. Loss of water quality may occur Malfunction of plant for about a week due to loss of electric power and backup power if any, extensive damage to various equipment, considerable damage to sedimentation DS3 Moderate basins, considerable damage to chlorination tanks with no loss of contents, or considerable damage to chemical tanks. Loss of water quality is imminent. Pipes connecting the different basins and chemical units being extensively damaged. DS4 Extensive This type of damage will likely result in the shutdown of the plant. DS5 Complete Complete failure of all piping, or extensive damage to the filter gallery. Table 4-10. WTP damage states (adapted from FEMA, 2003a) 4.6.1.2.2 Fragility functions for ground shaking Based on whether a WTP is constructed to meet seismic requirements, FEMA Hazus (FEMA, 2003a) has developed fragility functions for WTP corresponding to the probability of exceeding a damage state. Fragility functions relevant to common WTP in the Caribbean are summarized in Table 4-11. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 114 Mean, PGA, g Standard deviation, Typology DS2 DS3 DS4 DS5 g Stand components 0.16 0.27 0.53 0.83 0.4-0.6 Seismic components 0.2 0.38 0.53 0.83 0.5-0.6 Table 4-11 WTP fragility parameters, ground motion (adapted from FEMA, 2003a) 4.6.1.2.3 Fragility functions: ground failure (liquefaction) The fragility functions for ground failure hazard are similar to what was presented for buildings in Chapter 3. 4.6.1.2.4 Typical WTP Most WTP in the subject countries can be considered as small (10-50 MGD) or medium (50-200 MGD)- sized; see Table 4-3 The larger WTP usually have better engineering and construction standards and thus are more resilient. In this section, it is assumed that existing WTP are based on lower engineering standards and inadequate anchorage, whereas retrofitted WTP are based on better engineering and adequate anchorage of components. Figure 4-10 presents the fragility functions for two WTP: one with conventional design, and one with properly designed anchorage and seismic components. Also shown in the figure is the line with ordinates of approximately 0.36 g, which corresponds to a moderate earthquake. As seen in the figures, the probability of experiencing significant damage is reduced measurably when retrofit measures are implemented. Figure 4-11 presents the probability of the WTP being in a given damage state when subjected to a moderate earthquake. Note that once seismic retrofitting is implemented, the probability of the WTP experiencing the detrimental DS5 state is reduced from approximately 8% to approximately 1%. Existing Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 115 Retrofitted Figure 4-10. Fragility functions for example WTP Figure 4-11. Distribution of damage states for WTP 4.6.1.3 WWTP 4.6.1.3.1 Damage states FEMA Hazus (FEMA, 2003a) defines a number of damage states for WWTP. The damage states relevant to the Caribbean are summarized in Table 4-12. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 116 Damage State Description state DS1 None No observable damage Malfunction of plant for a short time (less than three days) due to loss of electric power and backup power if any, considerable damage to various equipment, light damage to DS2 Slight sedimentation basins, light damage to chlorination tanks, or light damage to chemical tanks. Malfunction of plant for about a week due to loss of electric power and backup power if any, extensive damage to various equipment, considerable damage to sedimentation DS3 Moderate basins, considerable damage to chlorination tanks with no loss of contents, or considerable damage to chemical tanks. Pipes connecting the different basins and chemical units being extensively damaged. DS4 Extensive This type of damage will likely result in the shutdown of the plant. DS5 Complete Complete failure of all piping, or extensive damage to the filter gallery. Table 4-12. WWTP damage states (adapted from FEMA, 2003a) 4.6.1.3.2 Fragility functions for ground shaking Based on whether a WWTP is constructed to meet seismic requirements, FEMA Hazus (FEMA, 2003a) has developed fragility functions for WWTP corresponding to the probability of exceeding a damage state. Fragility functions relevant to common WWTP in the Caribbean are summarized in Table 4-13 Mean, PGA, g Standard deviation, Typology DS2 DS3 DS4 DS5 g Stand components 0.16 0.26 0.48 0.8 0.4-0.55 Seismic components 0.33 0.49 0.70 1.23 0.4-0.55 Table 4-13. WWTP fragility parameters, ground motion (adapted from FEMA, 2003a) 4.6.1.3.3 Fragility functions: ground failure (liquefaction) The fragility functions for ground failure hazard are similar to what was presented for buildings in Chapter 3. 4.6.1.3.4 Typical WWTP Most WWTP in the subject countries can be considered as small (10-50 MGD) or medium (50-200 MGD)- sized. The larger WTP usually have better engineering and construction standards and thus are more resilient. In this section, it is assumed that existing WWTP are based on lower engineering standards and inadequate anchorage, whereas retrofitted WWTP are based on better engineering and adequate anchorage of components. Figure 4-12 presents the fragility functions for two WWTP: one with conventional design and one with properly designed anchorage and seismic components. Also shown in the figure is the line with ordinates of approximately 0.36 g, which corresponds to a moderate earthquake. As seen in the figures, the probability of experiencing significant damage is reduced measurably when retrofit measures are implemented. Figure 4-13 presents the probability of the WWTP being in a given damage state when subjected to a moderate earthquake. Note that once seismic retrofitting is implemented, the probability of the WWTP experiencing the detrimental DS5 state is reduced from approximately 7% to approximately 1%. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 117 Existing Retrofitted Figure 4-12. Fragility functions for example WWTP Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 118 Figure 4-13. Distribution of damage states for WWTP 4.6.1.4 Underground potable water pipes 4.6.1.4.1 Damage states FEMA Hazus (FEMA, 2003a) defines pipe fragility based on the number of repairs (leaks or breaks) per km of pipe. Fragility is defined based on either peak ground velocity (PGV) or permanent ground deformation (PGD). 4.6.1.4.2 Fragility functions for ground shaking Pipes are classified as either brittle or ductile; see Table 4-14. Based on whether brittle or ductile pipes are used, FEMA Hazus (FEMA, 2003a) has developed fragility functions for failure (leak or break). Typology Examples Asbestos concrete (AC) Brittle Cast iron (CI) Concrete (RCC) Ductile iron (DI) Welded steel (WS) Ductile High-density polyethylene (HDPE) Polyvinyl chloride (PVC) Table 4-14. Underground water pipe types (adapted from FEMA, 2003a) 4.6.1.4.3 Fragility functions: ground failure (liquefaction) The fragility functions for ground failure hazard are similar to ground shaking and are based on PGD. 4.6.1.4.4 Typical UGP Older underground pipes are likely brittle; newer replacement pipes are ductile (see Figure 4-7). Figure 4-14 presents the repair rate (repairs per km) for both leaks and breaks and for brittle and ductile pipes. Note the following: Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 119 • For an earthquake with a PGV of 50 cm/sec, the number of leaks in an 80-km pipe is reduced from 52 to 16 when ductile pipes are used. This implies significant savings in both repair cost and service interruption. • For an earthquake (or liquefaction) with a PGD of 30 in., the number of breaks is reduced from 4 to 1 per km when ductile pipes are used. • To increase the resiliency of pipe systems, either ductile pipes can be used or alternatively, existing pipes can be used with ductile connections or ductile liners to achieve the same benefits. The latter approach is more cost effective and can be implemented to reduce service interruption times. Leak Ductile Figure 4-14. Fragility functions for example UGP Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 120 4.6.2 Wind fragility functions Wind fragility functions for WTP and WWTP are similar to industrial buildings. UGP are not susceptible to damage from wind loading due to their underground location. Figure 4-15 presents the wind fragility functions for unreinforced (URM) and reinforced masonry (RM) buildings. Note that at a hurricane speed of 200 km/hr (typical in the subject countries), the damage is reduced from 70% to 20%. Measures such as adding reinforcement provide strengthening for both earthquake and hurricane loading. Figure 4-15. Wind vulnerability, water infrastructure (adapted from FEMA, 2003b) 4.6.3 Flood fragility functions Flood fragility functions for WTP and WWTP are discussed in this section. UGP are not susceptible to damage from flood loading due to their underground location. Flood hazard consists of inundation, which occurs in cases where the structure’s elevation is below the high flood level, scour of foundation due to fast-moving floodwaters, and hydraulic loading. Table 4-15 presents flood vulnerability for water infrastructure, identifying sub-hazards, vulnerable components, and hazard impacts. Figure 4-16 presents damage to WTP and WWTP as a function of flood depth. Note that at a depth of 2 m or greater, over 30% of the infrastructure is damaged. That damage consists of repair of electrical components, cleanup, and repair of buried conduits. Flood Hazard Impact Expected Expected Overall Sub-sector Hydraulic Financial Loss of Inundation Scour vulnerability loading loss operation WTP High Low Low High High High WWTP High Low Low High High High Underground -- Low -- Low Low Low Pipelines Table 4-15. Flood vulnerability, water infrastructure (adapted from FEMA, 2003c) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 121 Figure 4-16. Damage to WTP and WWTP as a function of flood depth 4.6.4 Restoration fragility functions Restoration fragility functions for WTP and WWTP are presented in Figure 4-17. Note that for WTP and WTTP plants: • For WTP plants, the restoration time is reduced from 97% to 47% from extensive to complete damage state. • For WWTP plants, the restoration time is reduced from 92% to 13% from extensive to complete damage state. Therefore, by strengthening these facilities, not only are physical damage and associated repair/replacement cost reduced in the aftermath of natural disasters, but there is also a significant decrease in the time to restore services. As shown in Figure 4-18, the combined damage from wind and flood exceeds the damage from either natural hazard alone. As such, mitigation measures that address either hazard will also be effective in reducing the combined damage from cases like when hurricanes and ensuing storm surge flooding can damage water and other infrastructure; see Table 4-16. As shown, when proper mitigation techniques are used for multi-natural hazards, the damage ratio is reduced by a factor of four. Retrofit Damage, % Case Hurricane Flood Wind Flood Total Existing No No 40% 60% 76% Retrofit 1 Yes No 10% 60% 64% Retrofit 2 No Yes 40% 10% 46% Retrofit 3 Yes Yes 10% 10% 19% Table 4-16. Flood vulnerability, water infrastructure (adapted from FEMA, 2003c) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 122 WTP WWTP Figure 4-17. Restoration functions, water infrastructure (adapted from FEMA, 2003a) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 123 Figure 4-18. Wind and flood combined damage (adapted from FEMA, 2003b) 4.7 Strengthening techniques 4.7.1 Introduction In the U.S. state of Oregon, a resilience plan for water infrastructure system has been developed. The approach for coastal areas in the state can be applied to the subject countries in this report, given the types of natural hazards. The planned recovery time can be used in the Caribbean to develop strengthening measures to meet these goals. Examples of proposed recovery goals for water infrastructure are presented in Table 4-17 (OSSPAC, 2013). The Oregon Seismic Safety Policy Advisory Commission (OSSPAC) notes that: If it were to occur today, a Cascadia subduction zone earthquake would result in catastrophic impacts to existing water and wastewater systems throughout western Oregon. Recovery time Case 1-3 days 1 week 2 weeks to 1 month 1-3 months 3-6 months Water to WTP 20-40% 50-60% 80-90% Main pipes 20-40% 50-60% 80-90% WWTP operational 20-40% 50-60% 80-90% Table 4-17. Recovery times (OSSPAC, 2013) 4.7.2 WTP and WWTP Plant buildings, water storage, and equipment (mechanical and electrical) are identified as representative facilities of WTP and WWTP infrastructure. According to the type of hazard, the typical strengthening measures mainly implemented in the U.S. are discussed in this section for the vulnerable facility components listed in Table 4-8. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 124 4.7.2.1 Earthquake: ground shaking 4.7.2.1.1 Wall out-of-plane strengthening (Plant building) Old or unreinforced buildings generally do not possess appropriate steel reinforcement in masonry/concrete walls to resist seismic out-of-plane force. Thus, the out-of-plane bending resistance of the walls is usually inadequate. The failure of heavy masonry/concrete walls due to a lack of out-of-plane strength is the most devastating damage to buildings. Bracing the wall at properly-spaced intervals is a general measure for improving the resistance of wall out-of-plane failure. To stiffen the wall, a steel strongback or brace supporting at diaphragm (i.e., horizontal floor element) and post-installed anchor are typically applied; see Figure 4-19. This improvement can be installed on both the inside and outside of the building, but the inside approach is more popular for property-line limitation or architectural exterior finish restrictions. Out-of-plane wall bracing Anchorage from diaphragm to wall Figure 4-19. Wall out-of-plane strengthening (FEMA, 2006b) 4.7.2.1.2 In-plane strengthening (Plant building) The global seismic strength of a building heavily relies on the in-plane strength of vertical elements, such as shear walls, braced frames or moment frames. To improve the in-plane seismic strength of a masonry/concrete building, a new element, such as a concrete shear wall, is typically added to the existing structural system. Adding new elements from the outside is more convenient for plant functionality, building usability and retrofit constructability, but inside installation could be constructible with a careful installation plan. Typical configurations and details of reinforced concrete (RC) shear walls overlaying on existing masonry wall and adding to existing building are shown in Figure 4-20 and Figure 4-21, respectively. In both cases, it is important to provide the appropriate load path elements to transfer the seismic force from diaphragm to the ground through the new wall elements and foundations. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 125 Concrete/shotcrete wall overlay Wall addition elevation Figure 4-20. RC shear wall overlay on existing masonry wall (FEMA, 2006b) Wall elevation with existing slab Wall connection with existing slab Figure 4-21. RC shear wall addition to existing building (FEMA, 2006b) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 126 4.7.2.1.3 Ductility improvement (Plant building) Ductility of a building during an earthquake is simply described as the capability of large rotation and displacement without any serious damages. Insufficient ductility is typically caused by an inadequate confinement of structural elements (e.g., column, beam or wall) or an inappropriate yielding location at the column-beam connection. To ensure the ductile behavior of masonry/concrete buildings during earthquakes, some retrofits should be implemented on local elements. For concrete columns or masonry piers, for example, FRP overlay, concrete jacketing or steel jacketing are popular measures to improve the confinement of existing columns, to increase the capacities for shear and axial loads and to enhance ductility, as presented in Figure 4-22. Column retrofit by FRP overlay Column retrofit by concrete/steel jacketing Figure 4-22. Ductility improvement for columns (FEMA, 2006b) 4.7.2.1.4 Foundation strengthening (Water storage) Seismic damage to water storage tanks is typically induced by unstable uplift and compression movements caused by the inertial force of the water tank due to earthquake shaking. This movement could induce tank wall buckling (i.e., elephant foot damage), foundation uplift and failure of the steel strap between the tank and foundation. To prevent these damages and stabilize the water tank during an earthquake, increasing weight resistance (i.e., mass addition) is a common way for withstanding potential uplift and compression movements, as presented in Figure 4-23. In the left drawing and picture, an internal ballast ring retrofit is shown. A proper connection (e.g., welded stud connector) between the concrete ballast ring and the tank wall is needed and the tank must be out-of-service during retrofit construction. In the right figures, an external concrete foundation addition is presented. Sufficient anchorage between the tank and new foundation is necessary, and the tank may remain in service, as this retrofit can be externally constructed. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 127 Internal ballast ring External anchor ring Figure 4-23. Foundation strengthening of water storage tank (Porter, 2008) 4.7.2.1.5 Anchorage improvement (Water storage) As described in the previous section, the connection between the water tank and foundation is also important to mitigate potential seismic damage of the water tank. As exemplified in Figure 4-24, a water tank with a high-aspect ratio usually needs a large amount of anchorage strength and foundation uplift capacity, because the extensive uplift force due to overturning moments can be expected during an earthquake. Large-sized anchor bolts are installed in the additional deep foundation in this example. The necessary capacity of anchorage elements and the foundation to resist a design seismic uplift should be identified by considering the local seismicity, soil characteristics and water tank volume for the design. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 128 Retrofit drawings Tank and added anchors Figure 4-24. Anchorage improvement with pile addition on water storage tank (Porter, 2008) 4.7.2.1.6 Foundation-tie retrofit (Equipment) WTP/WWTP facilities usually contain mechanical and electrical equipment, such as water pumps, power generators, filtering systems and so forth. This equipment is susceptible to horizontal shear force and overturning moments due to earthquake ground shaking, and those forces are sometimes largely amplified depending on the installation height of equipment. Several design guidelines have been published for the seismic design of equipment in countries like the U.S. or Japan. Based on the guidelines, representative retrofits aim to improve the connection between equipment and base foundation in order to resist seismic shear force and overturning. Adding rigid anchorage or installing vibration isolators at the equipment base are shown in Figure 4-25 and Figure 4-26, respectively. These retrofits improve the robustness and stability of equipment support against earthquake shaking. Figure 4-25. Rigid anchorage strengthening on equipment (FEMA, 2011) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 129 Figure 4-26. Vibration isolator support on equipment (FEMA, 2011) 4.7.2.2 Earthquake: liquefaction 4.7.2.2.1 Foundation retrofit When the size of shallow-spread foundations or the amount of deep-pile foundations are not sufficient to support the building, structures, facilities or equipment for a liquefaction event during an earthquake, the consequence could be demolishment and replacement, or extensive repair work with foundation improvement accompanying a long-term functional outage. As shown in Figure 4-27, enlarging the existing footing by attaching additional concrete footing is a typical measure to increase capacity for shallow-spread footing, and new pile installation next to an existing deep-pile foundation is usually implemented to improve deep-foundation capacity. In both cases, the retrofit construction work is not simple because, for example, excavation, formwork, reinforcement placement and concrete pouring are usually required around the existing foundation. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 130 Enlarge existing spread footing Adding piles to existing deep foundation Figure 4-27. Typical foundation strengthening measures (FEMA, 2006b) 4.7.2.2.2 Soil improvement Soil improvement is another approach to keep buildings, structures, facilities or equipment stable from large ground failure due to liquefaction. In Figure 4-28, a measure to improve the soil below existing shallow foundations and around existing deep foundations uses permeation grouting on liquefiable soil layers. There are generally two types of soil grouting: permeation grouting and compaction grouting. The former injects chemical/cement grout into the soil and aggregate without displacing the materials, and it solidifies the sandy soils and improves resistance to liquefaction. The latter injects very stiff grout into a layer of soil to force the soil particles into a tighter packing arrangement, and it increases the soil density and the liquefaction resistance (FEMA, 2006b). Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 131 Under existing shallow foundation Around existing deep foundation Figure 4-28. Soil improvement on liquefiable soil layers around existing foundation (FEMA, 2006b) 4.7.2.3 Wind 4.7.2.3.1 Roof connection strengthening (Plant building) Strong wind can easily blow the roof away if it is not securely attached to the walls or beams of a building. Since the roof usually has a structural role in horizontally supporting a wall, losing the roof due to strong wind might cause cascading damage, like wall instability or failure. Typical details to strengthen the roof connection of concrete and masonry structures are shown in Figure 4-29. Adding steel angles and anchors to strengthen the roof connection is a general measure for concrete structures. For masonry structures, new concrete bond beams might be needed to enhance a connection between the masonry wall and roof, and anchoring, steel strap and nailing are typically provided for securing the roof connection. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 132 Roof connection of concrete structure Roof connection of masonry structure Figure 4-29. Typical roof connection improvements (FEMA, 2006b) 4.7.2.3.2 External wall improvement (Plant building) When an external wall is not firmly attached to a building, fall or collapse of external non-structural walls is a typical failure mode in a strong wind event. If it is not a structural element or bearing wall, it could just cause minor damage, structurally, to a building, but might also induce severe functional damage to the plant building. In Figure 4-30, hurricane damage to perimeter block walls and a typical connection detail of block Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 133 wall, roof and foundation are presented. An adequate connection with anchor and reinforcement shown in the figure on the right could avoid this type of wall failure. Hurricane damage to external wall Typical connections of external wall Figure 4-30. Connections of external wall (FEMA, 2007a) 4.7.2.3.3 Foundation-tie retrofit (Equipment) As well as the foundation-tie retrofit for equipment against earthquake shaking (see Section 4.7.2.1.6), mechanical and electrical equipment are also susceptible to shear force and overturning moments due to strong wind. Equipment with larger areas, taller heights and/or light weight (which likely causes severe uplift) is more sensitive to wind forces. An effective retrofit would strengthen the base anchorage of equipment, as presented in Figure 4-31. This type of rigid anchorage for resisting uplift force may require additional foundational uplift capacity; foundation improvement is then also needed in that case. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 134 Figure 4-31. Anchorage strengthening of equipment (FEMA, 2011) 4.7.2.4 Flood 4.7.2.4.1 Building envelope improvement (Plant building) One of the retrofit measures to keep a building protected from flood inundation is installation of floodgates at potential water intrusion routes around the building and flood shields at openings of the building, such as doors and windows; see Figure 3-53. The height of the gate or shield should be designed according to the expected flood depth analyzed by local flood hazard conditions, and the gate and shield need to be structurally designed to resist the water pressure due to inundation depth. Preventing water inundation could help avoid any functional outages of the plant building that are potentially caused by damage to mechanical and electrical equipment inside of the building. 4.7.2.4.2 Elevated construction installation (Plant building) Constructing an elevated structure, like a floodwall around a building or facility site, is also an effective measure to protect a building from flood inundation. Typical types of floodwalls are shown in Figure 3-55. A floodwall is typically constructed some distance from a building, and it should be designed as a freestanding, independent structure separated from the building. This kind of wall is generally built of reinforced concrete. For designing a floodwall, the lowest point of the site, topographic information, the site-specific flood hazard level (i.e., inundation depth), soil characteristics, potential debris impact and a performance objective need to be identified. A successful example of a floodwall is presented in Figure 4-32; the floodwall provides a perfect barrier for the building and site against flood inundation. Figure 4-32. Floodwalls protecting buildings and facility site (FEMA, 2013) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 135 4.7.2.4.3 Elevating equipment (Equipment) To avoid functional loss of mechanical or electrical equipment at WTP/WWTP caused by floodwater, locating the equipment above the regionally-expected flood height is one common countermeasure. Unlike a building or large structure, elevating mechanical or electrical equipment is not very difficult, as the equipment can be basically moved, and relocating equipment is practicable. Based on the regional design flood height and the elevation of the WTP/WWTP site, the lowest height of equipment should be decided, and then the equipment relocated onto elevated steel platforms or put on an upper floor, as shown in Figure 4-33. Elevated equipment on upper floor (FEMA, 2013) Elevated equipment platform (Hilderhoff, 2013) Figure 4-33. Elevating equipment for flood protection 4.7.2.4.4 Dry floodproofing (Equipment) Similar to floodwalls or flood shields on buildings, watertight shields for mechanical/electrical equipment are a typical dry floodproofing measure. As presented in Figure 4-34, the watertight shield can be applied to both an inside equipment room and an outside equipment station. These watertight doors and barriers need to be designed by the static and dynamic water pressures evaluated on the design flood level. Also, in this case, a route to access the equipment during flooding has to be secured for checking or maintenance purposes, as this is not a dry floodproofing measure for the entire WTP/WWTP site, but for the limited area of equipment. Watertight door (FEMA, 2013) Watertight barrier (Hilderhoff, 2013) Figure 4-34. Watertight shields for equipment for dry floodproofing 4.7.3 Underground pipes For underground pipes transmitting potable water in the subject countries, there are a number of strengthening methods available. These techniques are discussed in this section for various natural hazards. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 136 4.7.3.1 Earthquake 4.7.3.1.1 Overview Underground pipelines in the Caribbean are subject to damage from both PGV and PGD; see section 4.6.1.4. Replacement of brittle pipes (see Figure 4-35) with ductile pipes (see Figure 4-36) is the most economical and effective option in the Caribbean. Experience has shown that most failures (breaks and leaks) occur in brittle pipes. Additionally, for asbestos-cement (AC) pipes, the pipe wall thickness decreases when in use and also loses structural integrity due to leaching, a process where the calcium content of the pipe is absorbed by the surrounding soil. Cast-iron (CI) pipes are also subject to corrosion and leaks. Concrete pipes can experience damage at the joints. Asbestos cement Cast iron Concrete Figure 4-35. Examples of brittle pipes (various) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 137 High-Density Polyethylene (HDPE) Ductile iron (DI) Welded steel (WS) Figure 4-36. Examples of ductile pipes (various) Eidinger and Davis (2012) documented damage to various types of pipelines in recent earthquakes. Table 4-18 summarizes their findings. Damage Chile New Zealand Japan description 2010 2011 2011 Damage to large-diameter transmission pipes Many N/A Many Damage to small-diameter distribution pipes (AC, CI) Thousands Thousands Thousands Damage to HDPE distribution pipe None None None Damage to DI with Earthquake-resistant joints N/A N/A None Table 4-18. Damage to various pipelines in recent earthquakes (adapted from Eidinger and Davis, 2012) Eidinger and Davis (2012) write: Chile 2010: Strong ground shaking and liquefaction damaged the city's only water treatment plant. Liquefaction damaged many of the larger diameter steel transmission pipelines. Liquefaction damaged many distribution water pipelines (about 3,000 total). Water outages to customers were lengthy, over a month to some customers. Even so, there was some good news: Essbio (the water system operator) had been installing pipe in its water distribution system for about a decade prior to the earthquake; while the rest of the water system suffered thousands of damaged pipes, no HDPE pipe was damaged. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 138 New Zealand 2011: Strong ground shaking and liquefaction damaged many of the city's wells. Liquefaction damaged many distribution water pipelines. The city's largest concrete reservoir failed completely due to ground deformations. Water outage durations to customers were moderate, mostly restored within 10 days after each earthquake. Even so, there were some good lessons learned: The Christchurch City Council (the water system operator) had installed some HDPE pipe in its water distribution system after the first earthquake; in the subsequent earthquakes, no HDPE pipe was damaged, while nearby older pipes were damaged. Japan 2011: A large diameter water transmission pipeline in the epicentral area suffered major damage at more than 50 locations, mostly due to pulled slip joints. A few large diameter water transmission pipelines suffered some slip joint damage in low-shaken areas, very distant from the earthquake. None of the seismic-resistant pipelines is known to have been damaged. 4.7.3.1.2 Risk analysis Table 4-19 presents the proposed seismic retrofit solutions for the underground pipes in the subject countries. Note that based on the level of seismicity, the soil condition, and the importance of pipe, a number of strengthening measures are recommended. Pipe importance Soil profile PGA Strengthening measure Critical (high capacity, ≥0.3 g Replace with HDPE, WS, DI Poor serving large population) <0.3 g Upgrade joints or pipes with 3 or more ≥0.3 g Upgrade joints Competent leaks in last 5 years <0.3 g Monitor ≥0.3 g Upgrade joints Poor <0.3 g Monitor Other ≥0.3 g Upgrade joints Competent <0.3 g Monitor Table 4-19. Suggested earthquake strengthening measures for UGP in the subject countries 4.7.3.1.3 Pipe replacement Existing brittle pipes can be replaced with ductile pipes. This retrofit can be implemented when a pipe segment is due for replacement. San Mateo County in California conducted a water system seismic improvement feasibility study (WWE, 2015). They noted that the existing water distribution system was vulnerable and needed upgrades. The proposed retrofits included: The majority of the existing La Honda water distribution system was constructed in the 1920s’ using AC and galvanized steel (GS) pipe materials In past, regular pipe breakage and leakage occurred throughout the system. The repairs were made using temporary fittings and clamps to keep the system in operation with a minimum downtime. WWE identified deficiencies in the existing water system, quantified necessary capital improvements, and analyzed various alternatives in the next section. Pipelines: replace with HDPE or ductile iron pipe The American Lifelines Alliance (ALA) provides guidelines (2005) for seismic design of pipelines. The proposed criteria can be used when considering replacement of existing pipes in the Caribbean for the purpose of improved seismic performance. The document writes: These Guidelines are not intended to completely eliminate all seismic induced pipe damage for Function Class II, III and IV pipelines, but it will significantly reduce the damage and Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 139 post-earthquake recovery time. In addition, the ability for the system to perform during and following an earthquake will be significantly improved. ALA (2005) emphasizes the importance of redundancy, defined as the following: 1. A leak or break in one pipe will not likely lead to damage on other redundant pipes; 2. All redundant pipes can provide a minimum needed flow to meet post-earthquake operational needs. The minimum level of flow required after earthquakes should generally be at the maximum winter time flow rate, or a level of water that is sufficient for household and most economic activities of the community 3. The redundant pipes are spatially separated by an adequate distance through potential ground deformation zones (landslide, fault movement, ground failure, lateral spreading, etc.) such that, should ground deformation occur, each redundant pipe may not be subjected to the same amount of ground movement due to the natural variation in movement The pipes are classified into four (4) categories and for each category, specific design recommendations are provided; see Table 4-20. Liquefaction PGD for 500- Class Description Redundancy Earthquake consideration year earthquake I Pipelines that represent very low hazard 0 -- -- -- I to human life in the event of failure. Not 1 -- -- -- needed for post-earthquake system I 2 -- -- -- performance, response, or recovery. II Normal and ordinary pipeline use, 0 500 year High, Very High 1.0*PGD II common pipelines in most water 1 500 year High, Very High 1.0*PGD II systems. 2 500 year High, Very High 1.0*PGD III Critical pipelines serving large numbers 0 1000 year High, Very High 1.35*PGD II of customers and present significant 1 500 year High, Very High 1.0*PGD economic impact to the community or a II substantial hazard to human life and 2 500 year High, Very High 1.0*PGD property in the event of failure. Essential pipelines required for post- Moderate, High, IV 0 2500 year 1.0*PGD earthquake response and recovery and Very High III intended to remain functional and 1 1000 year High, Very High 1.35*PGD operational during and following a II 2 500 year High, Very High 1.50*PGD design earthquake. Table 4-20. Recommended seismic design for different classes of pipelines 4.7.3.1.4 Joint retrofit The stiff joints in critical pipe sections can be replaced with flexible joints that allow for movement of one section of pipe relative to the adjacent segment. San Mateo County in California conducted a water system seismic improvement feasibility study (WWE, 2015). They noted that some of the 100-mm AC pipe connections had unrestrained and inflexible connections. The proposed retrofits included: Incorporates an EBAA Iron force‐balanced Flex Tend coupling that has an ability to deflect up to 20 degrees in all directions and can move laterally or horizontally for up to 200 mm. sufficient room for the force balanced Flex Tend coupling movement shall be provided. Use a Romac Alpha coupling that has an ability to deflect up to 8 degrees in all direction but the lateral and horizontal movement is limited to only 8 degrees of deflection. The International Organization for Standardization (ISO) provides the seismic performance requirements (2006) for ductile joints; see Table 4-21. Note that for seismic applications, connections with high expansion/contraction capacity, slip-out resistance, and joint deflection should be specified. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 140 Parameter Class Performance S-1 ≥1% length Expansion contraction S-2 <1% length S-3 <0.5% length A ≥3*diameter kN B ≥1.5*diameter kN Slip-out resistance C ≥0.75*diameter kN D <1.5*diameter kN M-1 ≥15 degrees Joint deflection angle M-2 ≥7.5 degrees M-3 <7.5 degrees Table 4-21. Suggested earthquake strengthening measures for UGP in the subject countries 4.7.3.2 Liquefaction The strengthening techniques for liquefaction ground deformation are the same as the options for earthquake mitigation. Table 4-22 presents the proposed seismic retrofit solutions for underground pipelines in the subject countries. Note that based on the level of seismicity, the soil condition, and the importance of pipe, a number of strengthening measures are recommended. Pipe importance Soil profile PGD Strengthening measure Critical (high capacity, ≥50 mm Upgrade joints liquefiable serving large population) <50 mm Monitor or pipes with 3 or more ≥50 mm Monitor Competent leaks in last 5 years <50 mm -- ≥50 mm Monitor liquefiable <50 mm -- Other ≥50 mm -- Competent <50 mm -- Table 4-22. Suggested liquefaction strengthening measures for UGP in the subject countries 4.7.3.3 Wind Water pipelines are not adversely affected by windstorms due to their location. 4.7.3.4 Flood Water pipelines can experience large displacements due to buoyancy effects only if the pipelines are empty (for example, during maintenance or repair), but it is quite rare. When failures of this kind occur, the discussion in the liquefaction-related section should be followed. 4.8 Cost analysis 4.8.1 Introduction Water and wastewater utilities face competing economic constraints. To implement strengthening techniques, it is critical to undertake a comprehensive plan that includes the following components: • Perform a screening of water infrastructure and detail their vulnerabilities • Prioritize the most vulnerable water infrastructure components and schedule them for upgrade • Develop a cost estimate for strengthening • Perform a feasibility study on the benefits of the strengthening of water infrastructure It is noted that some cost estimates presented in this section are for the United States. However, it is anticipated that the ratio of strengthening-to-reconstruction costs are similar in the U.S. and the Caribbean. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 141 4.8.2 Reconstruction costs The Caribbean Environmental Health Institute (CEHI, 2013) conducted a study regarding the financial assessment of wastewater treatment in the Caribbean. They noted that: • There have been no wastewater management assessments for a decade; • There is little financial cost data for developing appropriate systems; • There has been more focus on water management; • Despite this, there are technologies for wastewater systems in the Caribbean. They developed cost estimates based on assumptions that urban and rural areas were included, that there was no limitations on space availability, and based on assumptions of per capita wastewater rates. Table 4- 23 presents cost assumptions used for small systems. Table 4-24 presents the funding requirements based on the above assumptions. Item Cost, US$ Waste Stabilization Ponds 1,100 per m3 Equipment and installation 7,800 per m3 Pipe network 650 per m3 Lift stations 160 per m3 Table 4-23. New wastewater infrastructure cost data (adapted from CEHI, 2013) Region Continental Large island Small island Antigua, Barbados, Trinidad and Tobago, Dominica, Grenada, Belize, Guyana, Countries Bahamas, Haiti, Saint Kitts and Nevis, Suriname Jamaica St. Vincent and the Grenadines, St. Lucia Representative country Guyana Jamaica Saint Lucia People served 600,000 465,000 160,000 Daily flow, m3 750,000 106,000 37,000 Construction, US$M per m3 4,300 2,100 37,000 Operation and maintenance, US$M per m3 140/year 270/year 10000/day Table 4-24. Cost estimates for example pilot countrywide projects, wastewater (adapted from CEHI, 2013) The World Health Organization (WHO, 2008) examined the cost of improving water infrastructure to reduce the number of people without access to potable water and wastewater sanitation by half. The data was prepared for various regions. The relevant data for the Americas are presented in Table 4-25. This data and the population in the subject countries can be used to estimate the new construction cost for water infrastructure in the Caribbean. Water improvement Sanitation improvement Initial cost Annual recurrent cost Initial cost Annual recurrent cost 72 0.5 134 5.0 Table 4-25. Per capita cost in US$ of water and wastewater improvement and access (adapted from WHO, 2008) The Caribbean Development Bank (CDB, 2020) has approved a loan of US$6,100,000 for improvement of water network projects in Dominica. Table 4-26 presents the components covered as part of this project. Component Description 100,000 gallon reinforced concrete Water storage tank 250,000 steel Pipeline Supply 2 km ductile iron and HDPE Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 142 Install 2 km ductile iron and HDPE Antrim WTP Upgrade capacity by 470 m3/hr Leakage detection equipment Bulk meters Monitoring Real time SCADA (supervisory control and data acquisition) New intake on Check Hall River Water intake Upgrade existing intake at Springfield Table 4-26. Dominica water infrastructure improvement project (CDB, 2020) In Trinidad and Tobago, cost estimates for improvements to water infrastructure have been studied under various scenarios. Excerpts from this analysis are presented in Table 4-27 (GEF-CReW, 2017). Cost, US$M9 Component Existing Condition Improved Condition Annual capital expense 0.2 per plant N/A Annual operating expense 0.1 per plant N/A Periodic upgrade (5-year interval) 0.4 N/A 22 (total) including Capital cost N/A 0.8 engineering design 3.3 land management Needs exceed capacity; Additional service; Not meeting water Meet water quality; Indirect costs quality Lower operation and maintenance costs Table 4-27. Wastewater improvement in Trinidad and Tobago (adapted from GEF-CreW, 2017) 4.8.3 Strengthening costs In Trinidad and Tobago, trenchless technologies have been examined for strengthening of underground pipes. Table 4-28 (CWWA, 2017) presents the cost estimates for upgrading for various pipe diameters. Pipe size, mm Conventional, US$/m Trenchless, US$/m 300 275 50-175 500 400 200-500 1200 800 500-1000 1800 1500 850-1600 2500 3000 1500-3000 Table 4-28. Example of underground pipe strengthening cost (CWWA, 2015) Eidinger (2012) examined the economic benefits of seismic retrofitting of water infrastructure; see Table 4-29 and Table 4-30. The data is shown for the municipalities in the eastern San Francisco Bay Area. The level of earthquake considered in this study (magnitude 7.1) could be representative of earthquakes considered in some subject countries, such as Haiti or Trinidad and Tobago. As shown in Table 4-30, the cost associated with retrofitting is less than the expected economic loss cost for the scenario event. Parameters Assets Population served 1,200,000 Transmission pipes 320 km Distribution pipes 4,600 km WTP 6 9 Based on a rate of 0.15 US$ to 1 TT$ Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 143 Storage tanks 175 Pump stations 125 Table 4-29. Water infrastructure components in the study (adapted from Eidinger, 2012) Parameters Existing Upgraded Pipe repairs, scenario earthquake 3,300 to 5,000 -- Seismic upgrade cost -- US$240 M Unit cost -- $200 per person Economic loss US$680 M US$410 M Savings US$270 M = Benefits-cost US$30 M Unit  $25 per person Table 4-30. Cost and benefit associated with strengthening of water infrastructure (adapted from Eidinger, 2012) MWH Global, Inc. (2012) developed cost estimates for retrofitting of a WTP in the U.S. state of Oregon; see Figure 4-37. The original construction in the 1930s included the headhouse, filters, flocculation and sedimentation. In the 1950s and 60s, chemical storage tanks and additional basins were added and in the 1980s, a water intake structure was added. Figure 4-37. Existing WTP in Oregon (MWH, 2012) Analysis by MWH Global, Inc. (2012) indicated that seismic and landslide hazards were the main natural hazards for the site and a number of vulnerable components were identified: • Sedimentation basin walls were thin and subject to hydrodynamic forces. • Original buildings used weak and non-ductile details. • Filters used thin concrete walls. • Chemical tanks were not detailed for adequate level of force. To mitigate vulnerabilities, retrofit options were proposed and costs estimates were developed; see Table 4-31. Component Cost, US$M Building 1.8 Basin walls 2.0 Tanks 0.8 Pipe anchorage 0.2 Equipment anchorage 0.2 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 144 Total 5.0 Table 4-31. Estimated retrofitting cost for WTP in Oregon (adapted from MWH, 2012) The authors also examined the reconstruction cost for a new plant if the damage to the existing WTP necessitated building a new plant; see Table 4-32. The associated estimates for piping are presented in Table 4-33. Component Cost, US$M Buildings 3.0 Basins 4.0 Tank and storage 4.5 Filters 7.5 Pump stations 4.0 Total 23.0 Table 4-32. Estimated construction cost for WTP in Oregon (adapted from MWH, 2012) Diameter, m Cost, US$/m 0.5 1,600 0.6 2,000 0.75 2,500 0.9 3,000 1.2 4,000 Table 4-33. Estimated construction cost for pipes (adapted from MWH, 2012) It is noted that retrofitting (strengthening) is very cost effective when compared to new construction. 4.8.4 Life-cycle cost For water infrastructure, life-cycle cost analysis (LCCA) is typically used in design. This topic will be further discussed in Chapter 8. 4.9 Discussion In the subject Caribbean countries, there are a large number of water infrastructures, including WTP, WWTP, and many kilometers of pipelines. The courtiers are also located in one of the most hazard-prone areas in the world, experiencing a number of hurricanes annually and earthquakes periodically. These natural hazards have caused significant damage to water infrastructure in recent years, causing economic distresses and delayed recovery. Accordingly, it is time-critical to plan and implement measures that result in resilient water infrastructure. Experience has shown that for a given sub-class of water infrastructure, there are a number of components that are the most susceptible to damage in natural hazard events. For these components, effective retrofitting and mitigation methods are available that reduce the expected damage to infrastructure and are cost effective. Table 4-34 presents a summary of findings related to the most critical components of water infrastructure in the subject countries. Cost, % of initial cost (Est.) Subsector Hazard Vulnerable Component Resilience Measure Improvement Vulnerability cost reduction Wall out-of-plane Add wall bracing (anchor 15% 40% WTP/WWTP: (Building) and strongback) Plant building In-plane (Building) Add RC shear wall 20% 60% (Masonry/RC Ductility (Building) Add column jacketing 10% 30% Earthquake frame), Water Foundation (Water storage) Add concrete foundation 40% 50% storage, Anchorage (Water storage) Strengthen anchorage 10% 30% Equipment Foundation tie Improve foundation tie 10% 20% (Equipment) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 145 Enlarge spread footing or Foundation (All) 40% 80% Liquefaction add pile foundation Soil (All) Install soil grouting 30% 80% Roof connection (Building) Strengthen roof connection 10% 50% External wall (Building) Improve wall connection 10% 30% Wind Foundation tie Improve foundation tie 10% 20% (Equipment) Building envelope Install flood shield 10% 80% (Building) Elevated construction Construct floodwall 15% 80% Flood (Building) Elevating (Equipment) Elevate equipment 20% 80% Dry floodproofing Install watertight barrier 15% 80% (Equipment) Underground Replace with ductile pipes, Earthquake, Pipes 40% 80% pipelines Trenchless technologies Liquefaction (UGP) Joints Add flexible joints 30% 80% Table 4-34. Summary of findings for water infrastructure in the Caribbean Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 146 5. TRANSPORT INFRASTRUCTURE 5.1 Introduction Improving the resiliency of transport infrastructure has a multi-fold beneficial impact for a country: by reducing physical damage to infrastructure components, economic losses are reduced, and the continuous operation of roads and bridges allows for planners to direct resources and emergency personnel to the most adversely affected areas to assist in recovery. Bridges in particular are a critical infrastructure lifeline, as damage to them will sever the link between connected communities and will require use of detours that can then add time to commuting and recovery. Accordingly, bridges are emphasized in this chapter. For transport infrastructure in the Caribbean, the governing hazards and infrastructure components listed in Table 5-1 will be used. Sector Subsector Type Components Hazard Subsurface Flood Paved Roads Asphalt paved Surface Flood Columns Earthquake Concrete girder Joints Earthquake Members Earthquake Steel truss Connections Earthquake Bridges Bearings Earthquake Steel girder TRANSPORT Cross frames Earthquake Steel bridge Welded connections Wind Suspension Bridge Stability Wind All Foundation Earthquake Table 5-1. Analysis matrix for transport infrastructure 5.2 Geographical distribution In the subject countries, the following government entities are responsible for construction, maintenance, and upgrading of the transport sector, including road and bridges; see Table 5-2. Country Agency Suriname Ministry of Transport, Communication and Tourism (MINTCT) Trinidad and Tobago Ministry of Works and Transport (MOWT) Guyana Ministry of Public Infrastructure (MOPW) Belize Department of Transport (DOT) Haiti Ministry of Public Works, Transportation and Communications Dominican Republic Ministry of Public Works and Communications Antigua and Barbuda Ministry of Public Utilities, Civil Aviation and Energy Dominica Ministry of Public Works and The Digital Economy Ministry of Infrastructure Development, Public Utilities, Energy, Transport & Grenada Implementation Saint Kitts and Nevis Ministry of Public Works, Utilities, Transport and Postal Services Saint Lucia Ministry of Infrastructure, Ports, Energy and Labour (MIPST) Saint Vincent and the Grenadines Ministry of Transport, Works, Urban Development and Local Government Sint Maarten Ministry of Tourism, Economic Affairs, Transport and Telecommunication Barbados Ministry of Transport, Works and Maintenance Jamaica Ministry of Transport and Mining (MTM) Bahamas Ministry of Transport and Local Government Table 5-2. Transport infrastructure authorities for 16 subject countries Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 147 Figure 5-1 presents distributions of road networks in Belize, Dominica, Grenada, Saint Lucia, Suriname, and Saint Vincent and the Grenadines. Note that both primary and secondary (including unpaved roads) are shown. As seen in the figures, most roads are located close to shorelines and as such, could be vulnerable to flooding in the aftermath of hurricanes. Belize Dominica Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 148 Grenada Saint Lucia Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 149 Suriname Saint Vincent and the Grenadines Figure 5-1. Road networks (various sources) 5.3 Typology For bridges, the type usually refers to the superstructure construction. Examples of bridge types include girder bridges, truss bridges, arch bridges, and suspension bridges. As discussed earlier, this study focuses on existing bridges constructed prior to the year 2000. It is assumed that new bridges or bridges currently in design will have a better expected performance than bridges of older vintages. Examples of bridges in the study area are presented in Figure 5-2 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 150 For paved roads, the surface course is either bituminous, which is regarded as flexible pavement, or concrete, which is rigid; see Figure 5-3. Concrete paved roads have a better performance than asphalt roads; in the Caribbean, most roadways are asphalt. Thus, asphalt roads are presented in this report. Concrete girder bridge, Trinidad and Tobago Concrete girder bridge, Haiti Steel girder bridge, Suriname Steel girder bridge, Dominican Republic Steel truss bridge, Guyana Steel truss bridge, Belize Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 151 Suspension bridge, Trinidad and Tobago Suspension bridge, Belize Figure 5-2. Bridge typology (various sources) Asphalt pavement, Dominica Concrete pavement, Jamaica Figure 5-3. Paved road typology (various sources) 5.4 Vulnerable components Past natural hazard events have shown that certain components of roads and bridges are vulnerable to various natural hazards. Certain components have been known to experience frequent failures. Strengthening of these components would lead to marked improvement in performance and as such, these components are emphasized in this chapter. Examples of earthquake damage to vulnerable components of bridges are presented in Figure 5-4 Girder bridges can be vulnerable to hurricane wind forces and storm surge flooding. Simply-supported girder bridges can fail if the supports (bearings) have inadequate capacity to accommodate displacements imposed by wind forces. Suspension bridges are flexible and thus are not expected to experience damage in earthquakes. However, if they lack strength or stiffness, they can be vulnerable to wind forces from hurricanes. Examples of hurricane damage to vulnerable components of bridges are presented in Figure 5-5 There have been instances of floods washing bridges away. Paved roads are most vulnerable to flooding, and in particular to hurricane storm surge flooding, which unleashes large volumes of water at high velocity that can wash away the road, surface, and subsurface. Examples of flood damage to roadways are presented in Figure 5-6. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 152 Bridge unseating at connections Concrete column failure due to lack of confinement Column shear failure Buckling of slender members Figure 5-4. Bridge damage in California earthquakes (NISEE, 2020) Dropped spans (USGS, 2007) Structural instability (various) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 153 Fatigue cracking (Haghani et al., 2012) Damaged bridge (Floodlist, 2015) Figure 5-5. Bridge damage as a result of wind forces and associated flooding Washed away road (The Jamaican, 2017) Damaged road (Floodlist, 2016) Figure 5-6. Road damage due to flooding Table 5-3 presents the components that have experienced the most damage in past events and that will be considered in this chapter. Subsector Type Component Vulnerability Hazard Subsurface Substrate washout Paved roads Asphalt paved Flood Surface Strength loss Columns Lack of ductility Concrete girder Joints Short seat Members Slender members Steel truss Earthquake Connections Riveted connections Bearings Unseating Steel girder Bridges Cross frames Slender members Welded Steel bridges Low fatigue capacity Wind connections Suspension Bridge Stability Low torsional stiffness Wind bridge All Foundation Lack of capacity Liquefaction Table 5-3. Vulnerable components of transport infrastructure and corresponding hazards Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 154 5.5 Bridges 5.5.1 Design codes 5.5.1.1 Seismic design in Trinidad and Tobago In Trinidad and Tobago, a multi-year project of bridge assessment and reconstruction was completed in the early 2010s (Khan-Kernahan, 2013). As part of this project, 96 bridges underwent rapid assessment. From that pool, 70 bridges were selected for detailed assessment, 40 of which were selected for reconstruction. These reconstructed bridges were designed by international engineering consultants and the bridge design followed the American Association of Highway and Transportation Officials (AASHTO) design procedures in effect at the time. Thus, it is expected that such bridges incorporated seismic measures. At this time, the design procedures for seismic loading for other bridges in the region are not known. The 975-year seismic map is presented in Figure 5-7 This earthquake return period is usually used for bridges. As it can be seen in Figure 5-7, the bedrock acceleration is as high as 0.5g . For design, the bedrock values need to be multiplied by site coefficients to account for the soil flexibility. Figure 5-7. PGA for 975-year earthquake (UWI, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 155 5.5.1.2 Wind design in the Dominican Republic For the Dominican Republic, the wind design is based on the Manual for Wind Design (Grupo Estabilidad Estructural, 2000). The country is divided into three zones for wind loading; see Figure 5-8. For Zones I through III, the basic design wind speed is 240, 210, and 180 km/h, respectively. The key components of wind design for structures are as follows: • Design procedure based on US approach; wind pressure is computed based on the wind speed. • An importance factor is assigned to the structure • Internal and external wind coefficients are computed • Gust effects are taken into account • Evaluation is based on analytical procedure or wind tunnel testing It is anticipated that the provisions for wind design would vary depending on the level of expected hurricane hazard in subject countries. For example, in countries that experience frequent hurricanes, such as The Bahamas and Leeward Islands, wind loading provisions would be more important compared to the subject counties in South America. Figure 5-8. Wind zones in the country (Grupo Estabilidad Estructural, 2000) 5.5.2 Fragility functions based on FEMA Hazus methodology 5.5.2.1 Overview In this section, physical (structural) damage and operation loss (functionality) fragility relations for bridge types common in the Caribbean are presented. For the bridge structures, two types of design are considered: conventional (non-seismic design constructed prior to adoption of seismic codes) and seismic (constructed after adoption of seismic provisions in bridge design). Federal Emergency Management Agency (FEMA) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 156 Hazus (FEMA, 2003a) defines a number of motor (highway) bridge types. The typologies relevant to the Caribbean for multi-column bents are summarized in Table 5-4. Superstructure Superstructure Seismic Typology Continuity material type design HWB5 Slab Simply No HWB7 I girder supported Yes Concrete HWB10 T beam No Continuous HWB11 Box girder Yes HWB12 Beam Simply No HWB14 Stringer and floor beam supported Yes Steel HWB15 Plate girder No Continuous HWB16 Truss Yes Table 5-4. Bridge typologies (adapted from FEMA, 2003a) 5.5.2.2 Damage states FEMA Hazus (FEMA, 2003a) defines a number of damage states for bridges. The damage states relevant to the Caribbean are summarized in Table 5-5. Damage State Description state DS1 None No observable damage Minor cracking and spalling to the abutment, cracks in shear keys at abutments, DS2 Slight minor spalling and cracks at hinges, minor spalling at the column (damage requires no more than cosmetic repair) or minor cracking to the deck Any column experiencing moderate (shear cracks) cracking and spalling (column structurally still sound), moderate movement of the abutment (<50 DS3 Moderate mm), extensive cracking and spalling of shear keys, any connection having cracked shear keys or bent bolts, keeper bar failure without unseating, rocker bearing failure or moderate settlement of the approach. Any column degrading without collapse – shear failure - (column structurally unsafe), significant residual movement at connections, or major settlement DS4 Extensive approach, vertical offset of the abutment, differential settlement at connections, shear key failure at abutments. Any column collapsing and connection losing all bearing support, which may DS5 Complete lead to imminent deck collapse, tilting of substructure due to foundation failure. Table 5-5. Bridge damage states (adapted from FEMA, 2003a) 5.5.2.3 Fragility functions for ground shaking Based on the bridge types and damage states, FEMA Hazus (FEMA, 2003a) has developed fragility functions for bridges corresponding to the probability of exceeding a damage state. Fragility functions relevant to common bridges in the Caribbean are summarized in Table 5-6. Mean, Sa (1.0 sec), g Standard Typology DS2 DS3 DS4 DS5 deviation, g HWB5 0.25 0.35 0.45 0.7 0.6 HWB7 0.5 0.8 1.1 1.7 0.6 HWB10 0.6 0.8 1.1 1.5 0.6 HWB11 0.9 0.8 1.1 1.5 0.6 HWB12 0.25 0.35 0.45 0.7 0.6 HWB14 0.5 0.8 1.1 1.7 0.6 HWB15 0.75 0.75 0.75 1.1 0.6 HWB16 0.9 0.9 1.1 1.5 0.6 Table 5-6. Baseline bridge fragility parameters, ground motion (adapted from FEMA, 2003a) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 157 Note the following: • For simply supported bridges, seismic design has significant differences (see for example HWB5 and HWB7). This is because failure (unseating) at in-span hinges is a key mode of failure. For continuous bridges, in-span hinges are not present and there is more redundancy because of framing action in the longitudinal (along bridge centerline) axis. • Performance of steel and concrete bridges are similar. This is especially true when seismic measures are incorporated in design. • It is anticipated that a majority of bridges in the Caribbean are not continuous, that is, the superstructure is placed on joints above pier caps. The baseline mean values need to be modified for specific bridge configuration and geometry. The modification factors comprise the following: • Skew factor. Bridge skew introduces coupled and torsional response in the bridge. The skew is represented by angle  which is the angle between the centerline of bridge and centerline of piers (bents). Eq. 5-1 = √sin (90 − ) • Geometry factor. A factor that accounts for the number of spans (N) in a bridge. Parameters A and B are functions of the bridge type. Eq. 5-2 = 1 + − 5.5.2.4 Fragility functions: ground failure (liquefaction) The fragility functions for ground failure hazards are presented in this section. These functions are presented in terms of permanent ground displacement (PGD). Based on the bridge types and damage states, FEMA Hazus (FEMA, 2003a) has developed fragility functions for bridges corresponding to the probability of exceeding a damage state. Fragility functions relevant to common bridges in the Caribbean are summarized in Table 5-7. Mean, PGD, mm Standard deviation, Typology DS2 DS3 DS4 DS5 mm HWB5 90 90 90 350 5 HWB7 90 90 90 350 5 HWB10 90 90 90 350 5 HWB11 90 90 90 350 5 HWB12 90 90 90 350 5 HWB14 90 90 90 350 5 HWB15 90 90 90 350 5 HWB16 90 90 90 350 5 Table 5-7. Baseline bridge fragility parameters, ground failure (adapted from FEMA, 2003a) Note the following: • The baseline fragility parameters are independent of the bridge type, as damage is caused by the underlying soil. The baseline mean values need to be modified for specific bridge configuration and geometry. The modification factors comprise the following: Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 158 • Skew factor. Bridge skew introduces coupled and torsional response in the bridge. The skew is represented by angle  which is the angle between the centerline of the bridge and centerline of piers (bents). Eq. 5-3 1 = sin () • Geometry factor. A factor that accounts for the number of spans (N), bridge width (W) and bridge length (L). Eq. 5-4 _2 = 0.5/(()) 5.5.2.5 Example Bridge To assess the effect of seismic strengthening measures on a bridge, consider the bridge listed in Table 5-8, which is a typical bridge in the Caribbean. Liquefiable Condition Case Superstructure Continuity Substructure Geometry soil 1A Concrete I Simply Two column 50 m long, 16 m wide, As-is yes 1B girder supported piers 30° skew Retrofitted Table 5-8. Properties of the example bridge The modification factor for fragility functions for these two cases above has been computed; see Table 5- 9. Also shown in Table 5-9 are the computed modified fragility parameters. Type Existing Retrofitted A 0.25 0.25 B 1 1 Kskew 0.9 0.9 Kgeo 1.1 1.1 f1 0.6 0.6 f2 0.9 0.9 DS2 0.25 0.5 Mean, DS3 0.43 0.98 g DS4 0.46 1.12 DS5 0.71 1.73 DS2 80 80 Mean, DS3 80 80 mm DS4 80 80 DS5 310 310 Table 5-9. Modifications for fragility functions Figure 5-9 presents the fragility function for the example bridge in Table 5-8. Also shown in the figure is the line with an ordinate of approximately 0.75 g, which corresponds to the 1-sec spectral acceleration that could be typical for Trinidad and Tobago, Haiti, or the Dominican Republic. As seen in the figures, the probability of experiencing significant damage is reduced measurably when retrofit measures are implemented. Figure 5-10 presents the probability of the bridge being in a given damage state when subjected to an expected (design) earthquake. Note that once seismic retrofitting is implemented, the probability of the bridge experiencing the detrimental DS4 or DS5 damage states is reduced from approximately 80% to approximately 20%. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 159 Existing Retrofitted Figure 5-9. Fragility functions for example bridge Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 160 Figure 5-10. Distribution of damage states The mitigation for liquefaction consists of soil improvements or addition of deep foundations. Once such measures are implemented, the ground failure hazard is essentially eliminated; see Figure 5-11. Figure 5-11. Ground failure (liquefaction), existing case 5.5.3 Strengthening measures for earthquakes This section provides examples of seismic retrofits for the potential expected deficiencies in the bridge types considered in this report. The subsequent sections provide information on the cost and benefits of such strengthening measures. 5.5.3.1 Isolation casing When columns in a bent, bents in a frame, or frames in a bridge have fundamental vibration periods that differ by more than 10-20%, during the seismic events, the earthquake (inertial) forces are attracted disproportionally to the stiffer members and could cause severe damage to these members. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 161 Isolation casings are an inexpensive and effective method to mitigate unbalanced configuration by increasing the column length and ensuring a more uniform distribution of seismic forces to the bridge components; see Figure 5-12. Typical plans Retrofitted bridge Figure 5-12. Addition of isolation casing (Caltrans, 2020) 5.5.3.2 Concrete column retrofit for increased capacity Column failure in earthquakes could result in collapse of the bridge. When columns have inadequate shear or flexural capacity, pier walls can be added; see Figure 5-13. The walls serve to carry most of the seismic force and thus reduce the demand on the existing columns. For columns with poor confinement or inadequate confinement, Fiber-Reinforced Polymer (FRP) can be added to rectangular or circular columns to enhance performance; see Figure 5-14. Typical plans Retrofitted bridge Figure 5-13. Addition of pier walls (Caltrans, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 162 Typical plans Retrofitted bridge Figure 5-14. Addition of column casings (Caltrans, 2020) 5.5.3.3 Retrofit for continuity The unearthing of the superstructure at joints (hinges) is a common failure for existing bridges and occurs because the hinge or bearings do not have sufficient length. For retrofit of steel or concrete girder bridges supported on bearings at top of piers, cable restrainers are used to prevent excessive movement; see Figure 5-15. This still allows for thermal movement of the bridge, but prevents unseating at large displacement. Unseating could also occur at bridge abutment when the superstructure in supported on bearings and bridge travel exceeds the provided seat width. This is especially critical for bridges at large skews. The retrofit comprises constructing reinforced concrete bolsters in front of the abutment to extend the travel length; see Figure 5-16. Typical plans Retrofitted bridge Figure 5-15. Addition of cable restrainers at piers (Caltrans, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 163 Typical plans Retrofitted bridge Figure 5-16. Retrofit of abutment seats (Caltrans, 2020) 5.5.3.4 Foundation retrofit When the foundation has inadequate capacity, retrofitting consists of enlarging the foundation footprint and likely adding new piles; see Figure 5-17. Steel driven or miscopies can be used. Foundation retrofitting is typically expensive. Typical plans Retrofitted bridge Figure 5-17. Retrofit of foundations (Caltrans, 2020) 5.5.3.5 Retrofit of steel diaphragms for steel girder bridges For many existing steel girder bridges, diaphragms (cross frames) are constructed using slender steel angles or double angles. Tube steel can be added to increase stiffness and capacity of diaphragms; see Figure 5-18. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 164 Typical plans Figure 5-18. Retrofit of steel girder diaphragms (FHWA, 2006) 5.5.3.6 Retrofit of riveted members Existing steel truss or girder bridges constructed before the 1960s typically used riveted members or splices, respectively. Riveted attachments are not as robust as high-strength bolted connections. These members can be retrofitted by replacing the rivets with high-strength bolts; see Figure 5-19. Typical plans Figure 5-19. Retrofit of steel truss bridge members by replacing rivets with high-strength bolts (Ho, 2006) 5.5.3.7 Retrofit of lattice members Many existing steel truss bridges use lattice members. This was done to reduce material cost during construction. However, lattice members are more flexible than rolled beams or built-up girder and more susceptible to global or local buckling. Retrofit of these members consists of adding cover plates to the existing channels that have been originally connected by plate lattices; see Figure 5-20. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 165 Typical plans Figure 5-20. Retrofit of steel riveted lattice members by cover plates (Ho, 2006) 5.5.4 Strengthening measures for liquefaction For sandy soils located in areas of high ground water elevation, liquefaction is a risk when subject to earthquake hazard. Liquefaction is more likely to occur when earthquakes have a larger magnitude. Examples of ground improvements to mitigate liquefaction are presented in Figure 5-21. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 166 Add new mico piles Densification Soil mixing Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 167 Grouting De-watering Figure 5-21. Retrofit of foundations (FHWA, 2011) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 168 5.5.1 Strengthening measures for wind Earthquake forces are high-intensity, short-duration (often seconds) phenomenon and as such, certain members are permitted to undergo demands that exceed their nominal capacity. By contrast, wind loading imposes long (sometimes days) loading on the bridge components. Wind gusts cause fluctuation in member stresses. For steel members, this is critical, and if the bridge connection details are not properly designed, they can experience failure due to fatigue. In general, to prevent fatigue failure, the stress in steel components needs to be kept to a fraction of the nominal capacity. For steel girder bridges, a major source of fatigue cracking is at the connection of girders to the cross frames. The gap between the cross frames and girders causes secondary bending and concentrated stresses. Improved details (see Figure 5-22) can reduce stress in steel members and mitigate fatigue. In general, small changes in the bridge connections can lead to enhanced performance. For example, for welded stiffener connections, increasing the transition radius from 50 mm to 150 mm or more will result in a four-fold increase (10/2.6) in the stress range that can be theoretically imposed on a connection for infinite number of cycles; see Figure 5-23. Typical plans Improved detail for fatigue Figure 5-22. Improved crossed frame connection (TXDOT, 2013) Connection detail Figure 5-23. Fatigue performance for welded stiffeners (AASHTO, 2017) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 169 For suspension bridges, wind studies are needed to determine the response of bridges to windstorms. Torsional stiffness can be increased by adding steel members to convert open sections to closed sections that have much higher torsional rigidity; see Figure 5-24. Other options include solutions to improve the flow of wind around the bridge superstructure. Examples include providing openings in fascia girders and solid sidewalks. Original design Retrofitted bridge Figure 5-24. Improved crossed frame connection (GGB, 2020) 5.5.2 Strengthening measures for flood The most effective method for mitigation of flood damage is to provide protection to the bridge foundation. Assuming that the bridge superstructure is constructed high enough not to be inundated with floodwater, the flood hazard can be mitigated once the abutment and pier foundations are protected. As a first step, the bridge foundation should be analyzed to ensure it could withstand the water pressure forces. If the foundation capacity is inadequate, then a solution similar to what is discussed in previous sections can be implemented. Once the abutment and pier foundation capacities have been confirmed, then riprap can be added to protect the ground around the foundation from washing out; see Figure 5-25. The size, gradation, volume, and placement of riprap should be confirmed based on hydraulics reports regarding the expected water floor at individual bridge sites. Typical plans Retrofitted bridge Figure 5-25. Application of Riprap (FHWA, 2017) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 170 5.6 Paved roads (asphalt) 5.6.1 Overview For roadways, earthquake or wind hazards are not considered unless the road is located on liquefiable soil or close to a fault experiencing permanent ground deformation (PGD). The mitigation for liquefaction is similar for paved roads and bridges. Flood presents the most critical hazard for paved roads. 5.6.2 Design codes 5.6.2.1 Seismic design in Dominica As part of developing design requirements for coastal roads, University of the West Indies (UWI, 2001) undertook a training program in Dominica. Three key considerations were identified: i) safety, ii) environmental impact, and iii) cost (initial and life cycle). An example of a cross section of typical roads using flexible (asphalt) paving is presented in Figure 5-26. Figure 5-26. Cross section (UWI, 2001) The document states that bases and subbases should have sufficient strength, uniformity, and consistency. The pavement thickness is set as a function of traffic loading and references AASHTO standards. It is noted that increasing pavement thickness by 150 mm will reduce the damage to pavement significantly. 5.6.2.2 Seismic design in Jamaica In Jamaica, the National Work Agency (NWA) is responsible for developing design requirements for paved roads (2007). Regarding flooding, the document states: The drainage plan should correlate with the proposed ground level plan to minimize the passage of high storm flows on roadways and properties. All paved roadways are to be protected by kerb walls/ kerb and gutters, running parallel on each side of the road. Where design permits, well maintained grass verges are allowable in lieu of kerbwalls. Maximum recommended road grade shall be 15% for distances not exceeding 50 metres. No proposed road or lot should drain storm water directly unto an existing road or property without the specific permission from the National Works Agency Typical sections for arterial and major arterial roads are presented in Figure 5-27. Key requirements for paved road design are as follows: Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 171 • Base course is to be made of graded crushed aggregate. Grading varies from crushed stones of 0 to 37.5 mm with 10-15% gradation for various sieve sizes. Minimum California Bearing Ratio (CBR) of 80% • Subgrade is to be made of graded crushed aggregate. Grading varies from crushed stones of 0 to 37.5 mm with 10-15% gradation for various sieve sizes. Minimum CBR of 30% • Asphalt concrete (bitumen) should be based on a combination of coarse aggregate, fine aggregate, and mineral filler. • The bitumen/binder is to be 5-7% by weight. Bitumen to fill 75-85% of voids. • When minimum soaked CBR is less than 8%, treatment is required. Figure 5-27. Typical road sections (NWA, 2007) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 172 5.6.3 Fragility functions For roadways, flood and the resulting landslide are the most critical natural hazards to consider. Several authors have developed flood fragility functions for roadways; see Figure 5-28. Note that although the functions differ somewhat, at inundation of 1.5 to 2 m, the damage ratio is 50% and increases to 100% with an immersion of 5 m. Figure 5-28. Flood fragility functions for roadways (various) Flooding also results in additional vulnerability to earthquakes. Figure 5-29 (Argyrouddisi et al., 2018) presents the fragility function for highway embankment. Note that in the event of a 2-m inundation flood, a PGA of 0.4g (typical in the study countries) results in an increase of close to 100% in failure probability compared to the case where there is no flooding. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 173 Figure 5-29. Roadway fragility functions for earthquake and flood 5.6.4 Strengthening measures 5.6.4.1 Compliance with AASHTO design procedure For paved road design, AASHTO (2015) is used in the U.S. Key provisions for structural design of paved roads are summarized in this section. In general, pavements are classified into flexible, rigid or composite groups. Flexible pavements are a mixture of asphalt or bituminous material placed on compacted aggregate. The design is based on traffic load being distributed along the depth of the cross-section and thus there is a grading in material quality with the highest quality (smaller cross-section area) being used near the top. This type of pavement is also referred to as Hot Mix Asphalt (HMA) pavement. Figure 5-30 presents a typical cross-section for HMA pavement. Figure 5-30. Typical cross-section composition for HMA pavement HMA pavements are classified into three groups; see Table 5-10 and Figure 5-31 (AASHTO, 2015). Type Layer Conventional Deep strength Full depth Surface course <150 mm HMA Thick MHA HMA Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 174 Optional ATPB10 ATPB ATPB Base course Unbound aggregate Dense graded HMA over aggregate HMA Subgrade Treated layer Treated layer Treated layer Bottom Foundation soil Foundation soil Foundation soil Table 5-10. Types of HMA pavements (AASHTO, 2015) Figure 5-31. Cross-sectional view of HMA pavement types (AASHTO, 2015) The design of pavements is based on accumulation of damage due to traffic and climate. Design performance criteria and reliability impact the cost and performance of the pavement. The recommended design criteria for primary roads is listed in Table 5-11 (AASHTO, 2015). Performance criteria Threshold value at the end of design life Alligator cracking 20% lane area Rut depth 12 mm Transverse cracking length 132 m/km Smoothness (IRI) 3100 mm/km Table 5-11. Recommended design criteria, HMA pavement, primary roads (AASHTO, 2015) Performance criteria is defined here as: • Design life. The useful service life of a pavement that has deteriorated to a point where significant rehabilitation or replacement is required. In the U.S., a 50-year design life is typically used by state transportation agencies. • Alligator cracking. Series of interconnected cracks as a result of wheel loading; quantified as percentage of lane area. • Rut depth. Surface depression caused by permeant deformation in each pavement layer • Transverse cracking (thermal). Normal to pavement centerline and caused by temperature variations 10 Asphalt treated permeable base Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 175 • Smoothness. The parameter used to define pavement smoothness is International Roughness Index (IRI). The moisture content of the pavement layer has a significant impact on the performance and durability of paved roadways. Flooding can submerge the roadway for days and damage subgrade. As such, it is important to ensure water can drain quickly from the roadways. Research on deleterious effects of flooding on roadways are presented in this section. 5.6.4.2 Strengthening methods for flood mitigation The U.S. Army Corps of Engineers (USACE) has general responsibility for flood protection in the U.S. Currently, the Institute for Water Resources (IWR) is the organization working on development of a flood damage model for roadways (Davis et al., 2013). The key findings from their work are presented in this section. Flood damage to roadways has a number of mechanisms; see Table 5-12. Damage mechanism Mechanism Overtopping Flow normal to the centerline of roadway leading to erosion of embankment Pumping Removal of fine material from base or sub-base Scour Migration of channel into the toe of embankment Drainage blockage Loss of drainage capacity due to blockage or crushing Embankment instability Weakening of embankment caused by erosion Wave action Repeated overtopping due to breaking waves Table 5-12. Flood damage consequences (Davis et al., 2013) Overtopping is the most important damage mechanism and depends on velocity, duration, depth, and headwater and tailwater depths. Overtopping results both in loss of embankment and pavement and is directly related to soil type, duration, and overtopping depth; see Figure 5-32. For example, overtopping with a duration of 20 hours and overtopping of 1 m depth results in an approximate embankment pavement loss of 20% and 3%, respectively. As noted in the figure, the percent loss is significantly increased when inundation time is doubled. Figure 5-32. Pavement and embankment loss based on overtopping (adapted from Davis et al., 2013) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 176 Khan et al. (2017) evaluated the performance of various types of pavements subjected to flooding in Australia. The key findings from their work are summarized in Figure 5-33. The following is noted: • For flexible pavements, IRI and rutting increase with an increased probability of flooding. • Flexible pavements are more vulnerable to flood damage than rigid payments. • Weak, flexible pavements are more vulnerable to flooding than strong flexible pavements.11 Pavement strength could be enhanced by adding sealants, increasing pavement depth, layer stabilization, and conversion to composite pavement, use of moisture-resistant materials, improved drainage, and reduced traffic loading during floods. Pavement (IRI) damage, flood likelihood 11 Stronger pavements are the roads with thicker pavement depth Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 177 Pavement (Rutting) damage, flood likelihood Pavement (Rutting) damage, pavement type Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 178 Pavement (Rutting) damage, pavement strength Figure 5-33. Pavement damage as a function of probability of flooding (adapted from Khan et al., 2017) 5.7 Cost analysis 5.7.1 Introduction Transport infrastructure managers need to balance programing for strengthening/retrofitting costs with other priorities, such as scheduled maintenance or replacement. One option to consider would be to incorporate strengthening incrementally as part of the maintenance program, and to identify the most vulnerable bridges and roads and prioritize them for early replacement. As discussed previously, the focus would be on existing bridges only, as it is assumed that new bridges, for example, have adequate freeboard (to prevent flooding), use more robust foundations (i.e., concrete or steel piles instead of timber ones), and have details with some ductility (to prevent collapse in earthquakes). To implement the proposed strengthening techniques, it is critical to undertake a comprehensive plan that includes the following components: • Perform a screening of all bridges and roads to supplement the existing inspection and maintenance records and detail vulnerabilities • Bridges are more critical, as they are a larger investment, more difficult to repair after a major event, and could cause longer delays due to the longer detours required. • Prioritize the most important and vulnerable bridges and road segment components, and program them for further evaluation and strengthening • Develop a cost estimate for strengthening • Perform a feasibility study 5.7.2 Assumptions in the cost analysis In this section, some estimates of strengthening and replacement costs are presented. It is noted that most cost estimates are for the United States. However, it is assumed that the ratio of strengthening-to- reconstruction costs is similar in the U.S. and the Caribbean. The following is noted: • Many countries in the Caribbean use U.S. design code for design of larger transport infrastructure. • Many international companies use similar construction practices in the Caribbean as in the U.S. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 179 • The income and cost-of-living expenses vary in subject countries and there is variation among the larger and smaller subject countries. • For the most part, the labor cost is likely lower in the subject countries compared to the United States. However, many construction materials are likely more expensive, as they have to be imported. Accordingly, for a typical transportation project, the percentage and total dollar amount of labor and material cost varies from country to country and compared to the U.S. This variation is likely reduced when large pools of transportation projects are considered and projects in a large number of subject countries are included in the cost analysis. More importantly, although it is likely that here will be a large variation in dollar value of cost and benefit computations ( = benefits associated with retrofitting, that is reduction in future repair costs, minus the capital cost of implementing the retrofit, that is the initial construction cost), in this chapter, the emphasis is on the unitless ratio of benefit-to-cost (i.e., benefit/cost). For this parameter, the difference in cost calculations among countries and between material and labor is less sensitive to variations than the total cost. This last point is illustrated though a cost analysis example, shown in Table 5-13. A hypothetical case is considered here. Based on an assumed material-to-labor cost and retrofit-to-reconstruction cost ratio (biased towards material conservatively and because retrofitting typically is more labor than material intensive), the reconstruction and retrofit costs are computed for three cases: the U.S., a large subject country, and a small island country. As seen in the table, if the dollar value of benefit-cost () is used as a metric, the variation in computations can be off by a factor of 2 to 3 (160/77 or 77/26) in this example. However, when the ratio of benefit-to-cost (BCR) is considered, the variation is reduced to approximately 30% (variance between 3.8 and 2.8). Thus, it is expected that the benefit-to-cost ratio (BCR) for mitigation in the subject countries is not dissimilar to the U.S. case—and is used in this report in lieu of actual cost computations because of sparsity of local cost information—although the individual cost components and their contributions to the total cost can vary greatly. It is noted that similar analysis applies to other infrastructure considered in this report as well. Large Small Item U.S. Country Island Material-to-labor cost 0.5 3 4 Retrofit-to- Labor 0.2 0.2 0.2 reconstruction ratio Material 0.4 0.4 0.4 Labor cost 70 10 50 Reconstruction Material cost 35 30 200 Total cost 105 40 250 Labor cost 14 2 10 Retrofit Material cost 14 12 80 Total cost 28 14 90 Financial  77 26 160 analysis BCR 3.8 2.9 2.8 Table 5-13. Hypothetical cost and benefit calculations For transport infrastructure, there are three cost types associated with damage as a result of natural hazards: • Structural or primary cost. This is the cost associated with structural damage and the requirements to reconstruct or repair the damage that was caused by the event. • User or secondary cost. This is the cost that results from longer commute times, traffic delays, use of detours, wear and tear on vehicles, loss of revenues for managing agencies, and less safety for traveling in the construction zone. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 180 • Tertiary costs. This is the cost corresponding to adverse socioeconomic impacts and cascading effects of damage to transport infrastructure. Since transport infrastructure provides connectivity and allows for movement of goods and people, a disruption to service could economically impact a community. Furthermore, interruption to transport infrastructure will delay recovery and negatively impact other infrastructure. In this report, the cost-benefit calculations are conservatively based on the structural component only and show the cost-effectiveness of mitigation. Once indirect costs are factored in, the BCR can significantly increase and make mitigation even more appealing to decision makers. Finally, in the aftermath of a major natural disaster that causes damage to the built environment, there is usually high demand for skilled labor and scarcity of construction material. This is due to the high demand immediately after natural hazards. As such, there is a cost premium associated with repair and reconstruction work. This inflation—likely more pronounced in smaller island countries, which depend heavily on importation of construction material—will further tilt the balance in favor of mitigation and make the benefits even more cost effective. This additional cost premium associated with savings as a result of mitigation—applicable to transport, water, power and critical building infrastructure as well—was not included in this report and therefore, the computations are conservative as they underestimate the benefits of mitigation. In this section, some estimates of strengthening and replacement costs are presented. It is noted that most cost estimates are for the United States. However, it is anticipated that the ratio of strengthening-to- reconstruction costs are similar in the U.S. and the Caribbean, because many countries in the Caribbean use U.S. design code for design of transport infrastructure, and many international companies use similar construction practices in the Caribbean as in the U.S. 5.7.3 Reconstruction costs 5.7.3.1 Cost basis The construction cost for bridge and road construction in the subject countries is presented in Table 5-14. Cost, US$ Country Description million/km 82x7 m steel truss bridge and road Bahamas improvement, US$6 M 14.8 km highway and bridges, Barbados US$9.4 million 1.6 31 km road, US$50 million Belize 1.0 Dominican Republic 106 km highway, US$150 million 1.0 Guyana 1.2 Haiti 83 km highway, US$238 million 3 67 km 4-lane highway, US$600 Jamaica 9 million Antigua and Barbuda 8.7 km highway, US$17.5 million 2 Saint Lucia 1.5 Median 1.5 Table 5-14. Transport infrastructure construction in subject countries (various sources) 5.7.3.2 Bridges In Florida, the Florida Department of Transportation (FDOT) maintains a database for construction cost of new bridges (2020). These costs can be used to estimate the replacement (reconstruction) cost of bridges in the Caribbean. Example data for the type of bridges common in the Caribbean, based on the data gathered by the FDOT, are presented in Table 5-15. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 181 Type Span Cost US$/m2 of deck RC slab simple span $1,250-1,750 Short Precast concrete simple span $1,200-2,150 Concrete deck, steel girder, simple spans $1,350-$1550 Concrete deck, steel girder, continuous spans Medium $1,550-$1,800 Concrete deck, precast girder, simple spans or long $1,000-$1,550 Concrete deck, precast girder, continuous spans $1,050-$2,250 Bridge removal -- $400-$650 Table 5-15. Bridge construction cost (adapted from FDOT, 2020) In the aftermath of hurricanes and floods in 2013, a number of bridges in Saint Vincent and the Grenadines were damaged and required replacement. Examples are presented in Figure 5-34. Key information for these bridges is presented in Table 5-16 (Searchlight 2015, 2018). Note the significant cost associated with reconstruction of damaged bridges in Table 5-16. As discussed later, it is more cost effective to implement retrofitting and mitigation measures prior to natural disasters. Hope Bridge (Searchlight, 2015) Cumberland Bridge (Alvair, 2015) Figure 5-34. Replacement bridges after flood damage in Saint Vincent and the Grenadines Activity Type Size Cost US$ Hope Bridge replacement and 3 other bridges, plus river PC girder 24 * 8 m 4.2 million embankment Cumberland Bridge replacement, repair for 3 minor PC girder 26 m 5.8 million bridges, embankment protection Table 5-16. Replacement bridge information for Saint Vincent and the Grenadines (Searchlight 2015, 2018) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 182 Table 5-17 presents examples of unit costs for new bridge construction for four bridges that were constructed in Saint Vincent and the Grenadines in 2020. It is noted that the unit costs are comparable with U.S. practice of Table 5-15, it is reasonable to use the U.S. cost data as a baseline for the subject countries when comparing costs associated with reconstruction and retrofitting. Cost US$/m2 of Bridge Type Length, m deck 1 RC girder 8 $2,100 2 RC girder 19.29 $1,000 3 RC girder 8 $2,100 4 RC girder 15.8 $1,050 Table 5-17. 2020 Bridge construction cost of four new bridges in Saint Vincent and the Grenadines As part of the Trinidad and Tobago Roads and Bridges Programme of the National Highway Programme under the Inter-American Development Bank (IDB), 40 bridges were reconstructed; see Table 5-18 (Khan- Kernahan, 2013). Bridge type Spans Span, m No. 1 6m 14 Single cell box girder 1 9m 12 Two cell box girder 2 6m 1 Three cell box girder 3 9m 2 Four cell box girder 4 6m 1 Slab bridge 1 9m 1 Precast girder 1 25 m 8 Table 5-18. Bridge construction cost in Trinidad and Tobago (adapted from Khan-Kernahan, 2013) As part of the USAID-Organization of American States (OAS) Caribbean Disaster Mitigation Project (CDMP), a bridge in Saint Lucia was retrofitted and another one was reconstructed. CDMP (1998) writes: a) The cost of reconstruction work required to stabilize the banks of the river at this bridge site is estimated to be US $120,000. b) This cost when deflated to 1975, the year of construction of the present bridge, is 17.4% of the 1975 construction cost. c) The estimated costs of additional engineering and construction that would have been required in 1975 to prevent the damage which occurred in 1996 is US $20,000 or 10.8% of the original construction cost. d) The designers of the project in 1974 had made provisions for protection of the riverbanks. The immediate cause of the damage from the heavy rainfall in 1996 was the due to the reduction of the waterway in 1995. If this construction had not been undertaken, there may have been less damage to the north bridge approach. The riverbank protection now recommended by the Ministry of Communications and Works would have prevented the damage. The Caico bridge that was damaged by floodwaters was constructed in 1995. It replaced a bridge that was damaged by the 1994 flood. The superstructure was a Bailey bridge supplied by the British Development Division (BDD) and installed by the Ministry of Communications and Works of St. Lucia. The bridge was a single span of about 50 feet and the superstructure rests on existing stone abutments. This bridge was examined in some detail, as the original bridge constructed more than 20 years ago was destroyed by the 1994 floods. The area served by this bridge is an important banana producing area. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 183 The chief engineer reported that adequate wing walls and riverbank protection were not constructed for the reconstructed bridge due to lack of funding. In Table 5-19 (CDMP, 1998) and Table 5-20 (adapted from Dannion Engineering, 2020), note the cost effectiveness of mitigation for the Troumassee and Caico bridges. Figure 5-35 presents photographs of the bridges. Reconstruction-to- Bridge Type Span Activity Cost, US$ mitigation ratio (BCR) Original (1975) construction 185,000 cost Mitigation not incorporated 20,000 (11%) in 1965 Troumassee Steel 2 x 32,000/20,000=1.6 Reconstruction: add concrete- bridge truss 21 m filled steel piles and raise pile 120,000 cap in 1996 Reconstruction cost (deflated 32,000 (18%) to 1975) Bailey Caico bridge 24 m Reconstruction (1997) 317,000 >212 Bridge Table 5-19. Construction cost of bridges in Saint Lucia (adapted from CDMP, 1998) Bridge Type Span Description Cost, US$ 19 m wide, 4 lanes, 120-year 5,600,000 Bonne Terre Bridge Concrete 12 m service life, designed for 50- year flood Table 5-20. Construction cost of bridges in Saint Lucia (adapted from Dannion Engineering, 2020) Caico Bridge Bonne Terre Bridge Figure 5-35. Saint Lucia bridge construction (Dannion Engineering, 2020) 5.7.3.3 Paved roads In Florida, the Florida Department of Transportation (FDOT) maintains a database of construction costs for new paved roads (2020). These costs can be used to estimate the replacement (reconstruction) cost of roads in the Caribbean. Example data for the type of paved roads common in the Caribbean, based on the data gathered by the FDOT, are presented in Table 5-21. Note that the costs are the same order as those of the subject countries listed in Table 5-14. 12The bridge was reconstructed after the 1994 storms. The reconstructed bridge was destroyed in the 1996 storm and had to be reconstructed again. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 184 Location Type Cost US$/km Two-lane undivided; new 1,600,000 Four-lane divided; new 3,000,000 Six-lane highway, new 4,600,000 Rural Two-lane undivided; re-surface 300,000 Four-lane divided; re-surface 800,000 Six-lane highway; re-surface 1,000,000 Two-lane undivided; new 3,500,000 Four-lane divided; new 8,100,000 Six-lane highway, new 8,500,000 Urban Two-lane undivided; re-surface 350,000 Four-lane divided; re-surface 800,000 Six-lane highway; re-surface 1,200,000 Table 5-21. Bridge construction cost (adapted from FDOT, 2020) In Jamaica, more than 90% of roads use asphalt pavement. Examples of cost data for pavement are presented in Table 5-22. Cost US$/m2 of Pavement type Design life road Asphalt (flexible), 40 mm thick 15 years≤ 2,000 Concrete (rigid), 200 mm thick 20-50 years 8,000 Table 5-22. Example of pavement costs in Jamaica (adapted from Jamaica Gleaner, 2018) Examples of cost data for various pavement types are presented in Table 5-23. The data are based on aggregate results of 3,300 data points from a number of countries, including Belize (3 data points), the Dominican Republic (43 data points), Haiti (8), and Jamaica (1). Note that concrete pavements have a higher initial cost but given that this pavement has two-to-three times the design life of asphalt pavement, it could be cost-effective. It is also noted that the reconstruction cost is assumed to be of the same order as original construction of roadways. Construction Cost per km New 4-lane highway US$2,200,000 New 2-lane highway US$750,000 New 1-lane road US$92,000 Widening and reconstruction US$850,000 Concrete pavement US$540,000 Table 5-23. Example of pavement component costs (adapted from WBG, 2018) Table 5-24 presents examples of unit costs for new paved road construction for three roads constructed in Saint Vincent and the Grenadines in 2020. It is noted that the unit costs are comparable with US practice and the World Bank data presented above and as such, it is reasonable to use the U.S. cost data as a baseline for the subject countries when comparing costs associated with reconstruction and retrofitting. Road Type Length, km Cost per km 1 Asphalt 9.5 US$160,000 2 Concrete 3.6 US$1,800,000 3 Concrete 1.5 US$1,600,000 Table 5-24. Paved road construction cost in Saint Vincent and the Grenadines Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 185 5.7.4 Strengthening costs 5.7.4.1 Bridges Chandrasekaran and Banerjee (2014) examined the cost effectiveness of seismic retrofitting of typical concrete bridges in California. The bridge examined was 240 m long, 13 m wide, and comprised of five spans with a maximum span of 53 m. They considered two hazard cases: i) earthquake with PGA of 0.8 g and ii) flood causing 2.4 m of scour. Note that these hazard values could also correspond to the expected values in the Caribbean. They examined several cases (see Table 5-25) and they performed numerical analyses using a suite of earthquake and flood events. Fiber-reinforced polymer (FRP) Case thickness for column retrofit 0 (as-is) 0.5 1 1.0 2 1.5 3 2.0 4 2.5 Table 5-25. Bridge column retrofit scenarios (adapted from Chandrasekaran and Banerjee, 2014) Examining the cases of no mitigation and a mitigation with 2.5 mm of FRP for concrete columns, the following is noted when FRP retrofitting is implemented; see Figure 5-36. • Analysis was conducted for earthquake after an earlier flood event caused 2.4 m of scour • Probability of failure (at PGA 0.8g) reduced from 60% to 5% • Loss ratio reduced from 95% to 60% • Resiliency increased from 58% to 72% Probability of failure Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 186 Expected loss Figure 5-36. Retrofit effectiveness of bridge columns (adapted from Chandrasekaran and Banerjee, 2014) Note the following: • The retrofit cost was estimated by the authors as US$900/m2 of deck • For concrete bridges with a 53-m span, the cost of new/reconstruction is approximately US$3,500/m2 of deck (Caltrans, 2020) • Flood and seismic retrofitting cost was approximately 25% of new construction and resulted in a significant decrease in failure and loss. Asadi et al. (2020) examined the cost and benefits associated with retrofit of an existing bridge in a developing country. The bridge, constructed in 1976, has six spans of 24.4 m long. Each pier has three piers. The bridge was not designed for seismic loading; see Figure 5-37. The bridge piers are brittle and require retrofitting. Additionally, the bridge pier reinforcement is subject to corrosion and flood hazard can cause pier scour at the site. The site has moderate seismicity, with PGA of 0.35g. The scour depth was estimated based on a number of scenarios and a depth of 2 m was computed. The authors examined the pier retrofit using FRP and adding reinforcement to deficient piers. Note that this bridge, the site hazards, and the proposed retrofit could be typical of a case in the subject countries in this report. The authors noted the following: • In the event of an earthquake with PGA of 0.45 g, the existing bridge will likely collapse, whereas the retrofitted bridge will experience moderate and repairable damage. • Compared to the existing bridge, the retrofitted bridge has a higher capacity and lower loss ratio; see Figure 5-38. • The retrofit is cost effective and has a benefit-to-cost ratio of greater than 1.0. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 187 Existing bridge Elevation Figure 5-37. Bridge under investigation (Asadi et al., 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 188 Failure probablity reduction Loss reduction Figure 5-38. Loss reduction due to retrofitting (adapted from Asadi et al., 2020) Lee et al. (2004) examined concrete girder bridges. Note that this is one of the most common bridge types in the Caribbean and as discussed earlier, has column, bearing, and unseating vulnerabilities. The methodology took into account both vulnerability and future failure costs. They examined steel jacket retrofitting and noted that: Based on the simulated results for existing and retrofitted bridges, the application of steel jackets is found to be a positive solution to reduce the seismic damage risk of bridges. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 189 The U.S. Department of Transportation (USDOT) provides the methodology for determining merits of retrofit and reconstruction (2016). In their report, the authors examined a steel girder bridge, which could be similar to a bridge or flyover in the Caribbean. The 5-lane bridge has a 30 degree skew, has three spans, and is 91-m long and 32-m wide; see Figure 5-39. As part of retrofit, demotion and reconstruction of one abutment was considered. The work had a 233-day schedule. The results of the simulation are presented in Table 5-26. They considered two activities, listed below, with the associated costs presented in Table 5-26. • Partial: Partial depth removal of the backwall, deck, and total dam with Class A concrete • Full: Full depth removal of the dam with Class A concrete Initial cost US$ Activity Owner User Other Total Partial $2,600 $19,000 $600 $22,200 Full $21,200 $87,000 $2,800 $111,000 Table 5-26. Example of cost computations (USDOT, 2016) USDOT (2016) looked at various scenarios; see Table 5-27. Note that early mitigation is more cost effective than a delayed full reconstruction. Scenario Year 1 Year 12 Year 14 Overall cost I Full Partial -- US$175,200 II Full -- Full US$222,000 Table 5-27. Example of scenarios (USDOT, 2016) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 190 Photograph of bridge Figure 5-39. Case study bridge (adapted from USDOT, 2016) 5.7.4.2 Paved roads Examples of cost data for various pavement types are presented in Table 5-28. As discussed earlier, the data are based on aggregate results of 3,300 data points from a number of countries, including Belize (3 data points), the Dominican Republic (43 data points), Haiti (8), and Jamaica (1). Note that the retrofitting costs are significantly less than the reconstruction costs. Construction Cost per km Upgrading US$250,000 Asphalt resurfacing US$65,000 Bituminous Pavement Preventive Treatment US$7,000 Strengthening US$140,000 Surface treatment US$25,000 Preventive maintenance US$5,000 Routine maintenance US$2,000 Table 5-28. Example of pavement rehabilitation costs (adapted from WBG, 2018) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 191 5.7.5 Life-cycle cost For transport infrastructure, life-cycle cost analysis (LCCA) is typically used in design. This topic will be further discussed in Chapter 8. 5.8 Discussion Throughout the Caribbean, there is extensive transport infrastructure, including many bridges and thousands of kilometers of paved roads. The countries are also located in one of the most hazard-prone areas in the world, experiencing a large number of hurricanes and floods annually and earthquakes periodically. These natural hazards have caused significant damage to bridges and roads in recent years, causing economic distress, delayed recovery, impacts to travelling and public life, and interruptions of the movement of goods and services. Accordingly, it is critical to plan and implement measures that result in resilient transport infrastructure. For bridges, there are a number of components that are the most vulnerable to earthquakes and flooding. For these components, effective retrofitting and mitigation methods are available that reduce the expected damage to infrastructure and are cost effective. Table 5-29 presents a summary of the findings related to the most critical components of transport infrastructure in the subject countries. Cost, % of initial cost (Est.) Vulnerable Subsector Hazard Resilience Measure Improvement Vulnerability Component Cost reduction Add sealant 2% 20% Paved Paved Strengthen (thicker Flood 20% 50% roads surface pavement) Concrete surface 40% 80% Columns Add FRP or steel casing 10% 60% Bearing Add cable restrainers 2% 50% Abutment, Earthquake Add concrete bolsters 5% 50% pier seats Bridges Steel cross Fatigue-resistant details 10% 80% frames Flood Add riprap 5% 20% Foundations, Flood, abutments Add new piles 40% 80% liquefaction Table 5-29. Summary of findings for transport infrastructure in the Caribbean Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 192 6. POWER INFRASTRUCTURE 6.1 Introduction Improvement in resiliency of the power infrastructure (see Figure 6-1) has a two-fold beneficial impact for a country: i) by reducing physical damage to the infrastructure components, economic losses are reduced; and ii) the continuous operation of power substructures allows other infrastructure, such as hospitals and emergency centers, to continue to function, which allows expedited overall community recovery. Substations and transmission lines are in particular critical infrastructure lifelines and will be emphasized in this chapter. For power plant structures, many of the measures discussed for building infrastructure apply. Figure 6-1. Power infrastructure For power infrastructure in the Caribbean, the hazard and infrastructure components listed in Table 6-1 will be used. As noted in the table, for power plants and substations, earthquakes, and to a lesser extent, flooding, are critical, whereas for distribution systems, wind is the most critical hazard. It is assumed the discussion presented herein applies to all countries in the Caribbean, as power sector construction is likely similar across the subject countries due to significant cooperation between utility providers. Sector Subsector Type Components Hazard Design Earthquake Grade construction Flood Power plants Low-rise RC Nonstructural components Earthquake Earthquake, Foundation Liquefaction Design construction Earthquake Grade construction Flood Battery racks, cable trays, Control buildings Earthquake POWER Substations etc. Earthquake, Foundation Liquefaction Transformers Anchorage Earthquake Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 193 Frames Steel members Earthquake, wind Equipment High-voltage units Earthquake Tower Wind, Earthquake Transmission lines Lattice towers Foundation Flood, Liquefaction Lines Wind Table 6-1. Analysis matrix for power infrastructure In this report, the main transmission system (lines and poles) transferring power from generation centers to switchyards and substations will be discussed. The distribution systems carrying electricity to individual structures are not considered. Figure 6-2 shows the distinction between transmission and distribution lines. Transmission towers and lines Distribution poles and lines Figure 6-2. Transmission and distribution systems (various) 6.2 Geographical distribution In the subject countries, the following utility companies are responsible for generation, transmission, and distribution of electrical power; see Table 6-2. Country Utility Suriname N.V. Energie Bedrijven Suriname (EBS) Trinidad and Tobago Trinidad and Tobago Electricity Commission (T&TEC) Guyana Guyana Power Light (GPL) Belize Belize Electricity Limited (BEL) Haiti Electricité d’Haïti (EDH) Dominican Corporation of State Electricity Companies Dominican Republic (Corporación Dominicana de Empresas Eléctricas Estatales - CDEEE) Antigua and Barbuda Antigua Public Utilities Authority (APUA) Dominica Dominica Electricity Services (DOMLEC) Grenada Grenada Electricity Services Ltd. (Grenlec) Saint Kitts and Nevis St. Kitts Electricity Company Ltd. (SKELEC) Saint Lucia Light and Power Holdings (LUCELEC) Saint Vincent and the Grenadines St. Vincent Electricity Services Limited (Vinlec) Sint Maarten Electricite de France (EDF) Barbados Barbados Light & Power Company Limited (BL&P Co.) Jamaica Jamaica Public Service Company Limited (JPS) Grand Bahama Power Company (GBP) Bahamas Bahamas Power and Light Ltd. (BPL) Table 6-2. Utility providers for 16 subject countries Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 194 Examples of power distribution maps in the subject countries are presented in Figure 6-3. As discussed previously, the report will focus on substructures and transmission system components of power infrastructure in the Caribbean. Barbados Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 195 Dominican Republic Jamaica Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 196 Belize Figure 6-3. Power distribution networks (National Energy Grid Map Index, 2020) Most of the power plants in this region are powered by imported fossil fuels. However, efforts are underway to utilize hydropower, geothermal, solar, and wind power as well; see Figure 6-4. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 197 Thermal (oil or diesel) power plant (Suriname) Geothermal power plant (Dominica) Wind power (Jamaica) Hydropower (Dominica) Solar power (Bahamas) Natural gas power plant (Dominican Republic) Figure 6-4. Power generating plants (various sources) Table 6-3 presents data regarding the power generation facilities, substations, and power distribution for selected countries. In recent years, upgrades to power sector components in several countries have been undertaken or are planned. Examples include: • Haiti (BU, 2018): In recent years, the World Bank has committed to investing $40 million in improving the transmission and distribution network in the PAP area, and the IDB and USAID is spending another $40 million on the rehabilitation of five substations in PAP: Canape Vert, Carrefour Feuille, Toussaint Brave, Croix-des-Bouquets, and Nouveau Delmas. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 198 • Guyana (GPL, 2020): [T]hat funds amounting to $2.2 billion were also identified for the rehabilitation of some 55 kilometres of 69 kV transmission line. The distribution system was falling apart in some areas and the project is really about rehabilitation and reconfiguration but there are still generation and transmission aspects and at some point we will be regularly breaking out these projects • Suriname (EBS, 2020): Under the Project, EBS will undertake the upgrade of 36.6 kilometres of sub-transmission and distribution lines, the construction of five new substations, and the expansion or upgrade of three existing substations. Capacity Transmission Country Power plants Substation GWh lines 6 Suriname 1,600 N/A N/A (thermal and hydroelectric) Trinidad and Tobago 10,300 N/A N/A N/A Guyana 730 N/A N/A N/A Belize 450 N/A N/A 3,000 km 13 9 Haiti 650 1,700 (Port-Au-Prince only) (Port-Au-Prince only) 11 Dominican Republic 14,400 N/A 940 (thermal) Antigua and Barbuda 330 N/A N/A N/A Dominica 110 3 (hydroelectric) N/A N/A Grenada 200 N/A N/A N/A Saint Kitts and Nevis 210 N/A N/A N/A Saint Lucia 370 1 7 117 km Saint Vincent and the 160 5 N/A N/A Grenadines Sint Maarten N/A N/A N/A Barbados 1,000 3 13 150 km Jamaica 4,800 N/A N/A N/A Bahamas 2,200 N/A N/A N/A Table 6-3. Power sector components (various sources) 6.3 Typology Examples of substations in the subject countries are presented in Figure 6-5. As seen in the figures, the substations use similar construction and electrical equipment. Figure 6-6 presents examples of transmission towers. The countries in the Caribbean, similar to practices used worldwide, utilize steel lattice towers for power transmission. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 199 CUL DE SAC Substation (142.5 MVA13) VIEUX FORT Substation (two 15 MVA) REDUIT Substation (two 15 MVA) PRASLIN Substation (6.5 MVA) UNION Substation (two 15 MVA) CASTRIES SUB-STATION (two 15 MVA) Saint Lucia (adapted from LUCELEC, 2020) 13 Megavolt-ampere Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 200 Duhaney Park Substation New Michelton Halt Substation Jamaica (adapted from JPS, 2020) Sophia Substation Sophia Substation Suriname (adapted from GPL, 2020) Figure 6-5. Substations in subject countries Haiti Trinidad and Tobago Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 201 Guyana Jamaica Figure 6-6. Transmission towers and lines in subject countries 6.4 Vulnerable components Past natural hazards have shown that substations and transmission towers are vulnerable to various natural hazards. Certain components have been known to experience frequent failures. Strengthening of these components would lead to marked improvement in the performance and as such, these components are emphasized in this chapter. Examples of earthquake damage to vulnerable components of substations are presented in Figure 6-7. Transmission towers are most vulnerable to wind forces, although damage to towers because of earthquake and flooding has also been observed; see Figure 6-8. The strengthening methods presented in this chapter focused on wind loading. However, a retrofitted tower will also perform well in earthquakes. Falle transformer Rolled transformer Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 202 Column shear failure Buckling of slender members Figure 6-7. Substation damage in past Earthquakes (NISEE, 2020) Earthquake Ground failure Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 203 Hurricane winds Flood Figure 6-8. Transmission tower damage as a result of natural hazards Table 6-4 presents the components that have experienced the most damage in past events and will be considered in this chapter. Subsector Type Component Vulnerability Hazard Design Nonductile construction Earthquake Grade construction No flood protection Flood Nonstructural Power plants Low-rise RC Lack of anchorage Earthquake components Earthquake, Foundation Shallow Liquefaction Design construction Strength stiffness Earthquake Grade construction No flood protection Flood Battery racks, cable Lack of Control building Earthquake trays, etc. anchorage/bracing Earthquake, Foundation Shallow Substations Liquefaction No or inadequate Transformers Anchorage Earthquake anchorage Earthquake, Frames Steel members Slender low damped wind Equipment High-voltage units Not qualified Earthquake Lack of strength Earthquake, Tower stiffness wind Transmission Flood, Lattice towers Foundation Inadequate capacity towers Liquefaction Vibration. Low Lines Wind damping Table 6-4. Vulnerable components of power infrastructure and corresponding hazards 6.5 Design practice 6.5.1 Overview ASCE 113 (2007) provides design criteria for structures. Although intended for the U.S. and North America, the document is used worldwide, including in the Caribbean. ASCE 113 wind design is based on provisions of ASCE 7. Similarly, the Pan American Health Organization (PAHO) has developed wind speeds to be used with wind provisions of ASCE 7 (2019). The key provisions of the standard are summarized below. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 204 6.5.2 Wind design The wind force is calculated based on the wind speed, terrain, importance factor, gust factor, and projected surface area. Typically, either a 50- or 100-year return interval is used to compute wind loading. The effect of wind flow in any direction is considered in design. In hurricane-prone zones, deflection limits are set for utilities for a more frequent event (such as a 5-year wind velocity). When wind vibrations are of concern, the components should be evaluated and designed for forces from such vibrations. 6.5.3 Earthquake loading Site-specific earthquake loading is defined based on site seismicity and soil condition. Typically, a 500- year event is used for design. Both horizontal components and vertical components (equal to 80%) are considered in design. 6.5.4 Flood loading Infrastructure located in coastal zones needs to be designed for several types of floods. Structures or equipment subject to differential pressure from water should be designed to account for this force. 6.5.5 Load combinations The load combinations for design that include natural hazard forces are listed in Table 6-5. Note that for coastal areas, like many of the zones for the subject buildings, a combination of wind and flood loading is applied to structures and components. Short Wire Case Dead Operating Wind Earthquake Flood circuit tension I 1.1 1.0 0.75 1.1 1.2 0 2.0 III 1.1 1.0 1.0 1.1 0 0 1.0 IV 1.1 1.0 0.75 1.1 0 1.25 0 Table 6-5. Design load combinations (adapted from ASCE, 2007) 6.6 Fragility functions 6.6.1 Earthquake fragility functions based on FEMA Hazus methodology 6.6.1.1 Overview In this section, physical (structural) damage and operation loss (functionality) fragility relations for substations and transmission tower types common in the Caribbean are presented. For the structures, two types of design are considered: conventional (non-seismic design constructed prior to adoption of seismic codes) and seismic (constructed after adoption of seismic provisions in design). 6.6.1.2 Damage states FEMA Hazus (FEMA, 2003a) defines a number of damage states for substations. The damage states relevant to the Caribbean are summarized in Table 6-6. Damage State Description state DS1 None No observable damage Failure of 5% of the disconnect switches (i.e., misalignment), or the failure of 5 % of the circuit breakers (i.e., circuit breaker phase sliding off its pad, circuit breaker tipping DS2 Slight over, or interrupter head falling to the ground), or by the building being in minor damage state the failure of 40% of disconnect switches (e.g., misalignment), or 40% of circuit breakers (e.g., circuit breaker phase sliding off its pad, circuit breaker tipping over, or DS3 Moderate interrupter-head falling to the ground), or failure of 40% of current transformers (e.g., oil leaking from transformers, porcelain cracked), or by the building being in moderate damage state. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 205 Damage State Description state Failure of 70% of disconnect switches (e.g., misalignment), 70% of circuit breakers, 70% of current transformers (e.g., oil leaking from transformers, porcelain cracked), or DS4 Extensive by failure of 70% of transformers (e.g., leakage of transformer radiators), or by the building being in extensive damage state. Failure of all disconnect switches, all circuit breakers, all transformers, or all current DS5 Complete transformers, or by the building being in complete damage state. Table 6-6. Substation damage states (adapted from FEMA, 2003a) 6.6.1.3 Fragility functions for ground shaking Based on whether a substation is constructed to meet seismic requirements, FEMA Hazus (FEMA, 2003a) has developed fragility functions for substations corresponding to the probability of exceeding a damage state. Fragility functions relevant to common bridges in the Caribbean are summarized in Table 6-7. Mean, PGA, g Standard deviation, Typology DS2 DS3 DS4 DS5 g Stand components 0.1 0.2 0.3 0.5 0.4-0.6 Seismic components 0.15 0.25 0.35 0.7 0.4-0.6 Table 6-7. Substations fragility parameters, ground motion (adapted from FEMA, 2003a) 6.6.1.4 Fragility functions: ground failure (liquefaction) The fragility functions for ground failure hazard are similar to what was presented for bridge or building foundations. 6.6.1.5 Typical substation Figure 6-9 presents the fragility functions for two substations, one with conventional design and one with properly-designed anchorage and seismic components. Also shown in the figure is the line with ordinates of approximately 0.36 g, which corresponds to a moderate earthquake. As seen in the figures, the probability of experiencing significant damage is reduced measurably when retrofit measures are implemented. Figure 6-10 presents the probability of the substation being in a given damage state when subjected to a moderate earthquake. Note that once seismic retrofitting is implemented, the probability of the substation experiencing the detrimental DS5 state is reduced from approximately 20% to approximately 5%. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 206 Existing Retrofitted Figure 6-9. Fragility functions for example substation Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 207 Figure 6-10. Distribution of damage states 6.6.2 Wind fragility functions Panteli et al. (2017) conducted numerical studies and developed tower fragility functions for typical transmission towers subject to wind loading. They considered two cases: stand design and robust or enhanced design. As seen in Figure 6-11, the tower’s probability of failure is reduced from approximately 50% to 20% for a 50 m/sec windstorm. Similarly, when enhanced design is employed, the failure probability of the transmission line is significantly reduced. Transmission tower Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 208 Transmission line Figure 6-11. Fragility functions for typical transmission towers (adapted from Panteli et al., 2017) 6.7 Strengthening techniques 6.7.1 Substations Examples of strengthening techniques for substations are presented in this section. The emphasis is on the components that have been the most vulnerable in past earthquakes. 6.7.1.1 Use of seismically qualified components The Institute of Electrical and Electronics Engineers (IEEE) 693 provides seismic requirements for substation components (2018). IEEE 693 requires that the qualification level for substation components be established based on the site-specific PGA; see Table 6-8. Qualification level Site PGA, g Low ≤0.10 Moderate ≤0.50 High >0.5 Table 6-8. Substation components qualification levels (adapted from IEEE, 2018) There are specific prescriptive requirements for each qualification level. Many of the subject countries in the Caribbean will likely fall in the “moderate” qualification level. 6.7.1.2 Redundancy To improve the resiliency of the power infrastructure, a network with redundant components is preferred. The redundancy can be achieved in two important ways: • Employ different design criteria for two substations on the same network. For example, one substation can use more rigid components and one, more flexible ones. This ensures that there is less likelihood that both substations will be damaged in a given earthquake and that both can maintain the power supply (at least partially) until the damaged substation is repaired. • Ensure spare parts are stored at substations. In the event of damage in an earthquake, the damaged component can be replaced with the spare part and service restored rapidly. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 209 Clearly, the utilities sector needs to balance the cost of spare parts with the utility of having such parts. The Electric Power Research Institute (EPRI) presents a transformer spare strategy matrix (2014); see Table 6- 9. The data correspond to action taken in preparation before an event like a large earthquake. Manufacturing Availability Protection Inventory Creation Scenarios Cost Reliability Time Timing time Priority Low Low High Low High Utility orders spares for transformers (nearing end of life) Moderate Standard Fast High Short earlier than needed Utility orders spares for any critical transformers (regardless of High Standard Fast High Long remaining life) Utility keeps retired transformers Low None Fast Low Short on hand Utility participates in sharing Low None Fast High Long arrangements Utility orders emergency spare transformers before the event for High Short Fast High Long its own use only Utilities pool resources to order shared spares (emergency spare Moderate Short Fast High Long transformers) Table 6-9. Transformer spare strategy matrix (adapted from EPRI, 2014) 6.7.1.3 Transformers As discussed previously, one of the common failure modes for transformers has been sliding or overturning due to inadequate anchorage. Depending on the type of mounting, the anchorage for existing transformers can be significantly improved by addition of steel members and connections. Figure 6-12 (Knight and Kempner, 2004) presents samples of improvement measures for two types of common installations. Rail installation Pad installation Figure 6-12. Typical transformer anchorage enhancements (Knight and Kempner, 2004) Transformers – the most important and expensive part of the transmission system and one of the most vulnerable components – are heavy and relatively rigid units and thus are ideal candidates for the use of base isolation. Kempner and Riley (2016) write that the Bonneville Power Administration (BPA) has recently incorporated base isolation for some of their transformers; see Figure 6-13. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 210 Plan Installation Figure 6-13. Seismic base isolation of transformers (Kempner and Riley, 2016) 6.7.1.4 Flexible connectors In substations, electrical conductivity is achieved by use of connectors (bus) between two components. Since the two equipment pieces being connected typically have different vibration frequencies, they do not move in-phase (in unison) and this results in force being developed in the bus. This connector force is resisted by the interconnected equipment and thus induces additional demand on the components, which can result in their failure. This is primarily the case where the bus is rigid, since this configuration has a limited capability of allowing differential motion. To improve seismic performance, flexible ends can be added to rigid bus connections to accommodate this differential movement and thus mitigate the connector forces acting on the equipment. Examples are shown in Figure 6-14 (Song and Der Kiureghian, 2004). Plan Installation Figure 6-14. Retrofit of rigid bus by flexible connectors (Song and Der Kiureghian, 2004) 6.7.1.5 Wire rope isolation Circuit breakers are a key component of substations. These are tall units with heavy mass on top supported on slender steel structures; they have suffered failures in past earthquakes. Testing of rigidly-mounted units Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 211 has shown the failure that could be expected in earthquakes. Wire ropes can be used as a means to enhance the seismic performance of circuit breakers. The wire ropes allow rocking motion at the base of the component and thus reduce the seismic forces. Examples are shown in Figure 6-15 (Alessandri et al., 2015). Plan Installation Figure 6-15. Seismic protection of circuit breakers, wire ropes (Alessandri et al., 2015) 6.7.1.6 Spring dampers As discussed previously, most equipment in substations is supported on stands. An approach to reducing the seismic demand on the stand and the equipment is the use of friction spring devices with self-centering capabilities. The passive dampers allow rocking motion at the base of the component and dissipate seismic energy through friction. Examples are shown in Figure 6-16. Plan Installation Figure 6-16. Seismic protection by means of spring dampers (Riley et al., 2006) 6.7.1.7 Retrofit design of elevated supports Many components, such as discontent switches, are mounted on tall, elevated frames. Often, these frames are flexible and thus amplify the motion imparted on the equipment. Adding horizontal and vertical bracing to these frames will increase their stiffness significantly and thus reduce the motion and forces experienced Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 212 by the supported equipment. An example is presented in Figure 6-17 (Gilani et al., 2000). The retrofitted frame does not amplify the response. However, for the existing frame, the amplification factor is over 2. Exiting frame Retrofitted frame Figure 6-17. Seismic protection by means of improved support design (Gilani et al., 2000) 6.7.1.8 Composite insulators Porcelain has been used extensively for insulation in substations. It is relatively inexpensive, durable, and easily available. However, porcelain is also quite brittle and has experienced significant failure in past earthquakes. Composites can be used in place of porcelain to act as insulators. Composite insulators are significantly lighter (thus reducing seismic forces) and are not susceptible to brittle failure modes, such as fractures. Composite insulators can be used for a variety of components, such as bushings, disconnect switches, or any other element with insulators. An example is presented in Figure 6-18 (Gilani et al., 2000). Porcelain insulators Composite insulators Figure 6-18. Seismic protection by use of composite insulators design (Gilani et al., 2000) 6.7.2 Transmission towers and lines Transmission towers use slender steel lattice members. In particular, the diagonal members are typically single- or double-angle members with low buckling capacity. Transmission lines are subject to failure due to excessive wind-induced vibrations. The tower system has low structural damping, and this contributes to the vulnerability. 6.7.2.1 Retrofit of transmission towers Xie and Zhange (2020) developed a method for retrofit of the diagonal members of transmission towers while maintaining service; see Figure 6-19. They noted that by clamping plates at regular intervals to the diagonal members, the capacity was increased by close to 80%. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 213 Plan Installation Figure 6-19. Strengthening of existing transmission tower diagonal members (Xie and Zhange, 2020) Kumalasari et al. (2005) studied the retrofitting of latticed towers with split pipe sections attached to the tower legs with U-bolts; see Figure 6-20. On average, the load carrying capacity was improved by approximately 50%. Plan Installation Figure 6-20. Strengthening of existing transmission tower leg members (Kumalasari et al., 2005) Trovato et al. (2015) examined the use of retrofitting slender members by restraining tubes to provide resistance to buckling during earthquakes or hurricanes; see Figure 6-21. It was noted that the proposed solution prevented buckling of the tower members, increased the capacity by nearly 70%, and maintained the load carrying capacity after reaching its capacity. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 214 Plan Installation Figure 6-21. Buckling prevention of existing transmission tower leg members (Trovato et al., 2015) Zhang et al. (2013) studied the use of tuned mass dampers (TMDs) to increase damping in towers and reduce the motion and stresses in the tower members; see Figure 6-22. For low-damping systems, they obtained a vibration reduction of 26%. Plan Installation Figure 6-22. Tuned mass damper for transmission tower (Zhang et al., 2013) 6.7.2.2 Retrofit of transmission lines Stockbridge dampers are small TMDs attached to individual transmission lines that are used to suppress the vibration of transmission lines; see Figure 6-23. The installation of Stockbridge dampers reduces the vibrations in cables and by changing the dynamic properties of the cable, serves to suppress motions in the predominant natural frequencies. Since the most vulnerable segment of an individual cable is at the clamping ends, typically two dampers are installed per span. Dampers are typically used to suppression motion in vertical plane where motion is the most severe. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 215 Plan Installation Figure 6-23. Stockbridge dampers for transmission line (various sources) Lou et al. (2020) have studied an innovative method of using viscoelastic dampers to interconnect the individual cables as a means of wind mitigation; see Figure 6-24. Plan Installation Figure 6-24. Viscoelastic dampers for transmission line (Lou et al., 2020) 6.7.3 Foundation retrofit The retrofit of foundations for substation buildings follows the same procedure as discussed for buildings. For substation equipment, typically no foundation upgrades are required; instead, equipment anchorage as discussed previously might require strengthening. When flood or liquefaction is of concern, it is simpler to relocate the equipment or provide flood protection. The transmission tower, on the other hand, could be a vulnerable component that might require strengthening. For towers located in soft soil or in flood plains, foundation retrofit incorporating micropiles can be implemented that does not require service interruption. An example is shown in Figure 6-25 (Yamane, 2020). Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 216 Plan Installation Figure 6-25. Retrofit of transmission tower foundations (Yamane, 2020) 6.8 Cost analysis 6.8.1 Introduction Electrical utilities face competing economic constraints. To implement strengthening techniques, it is critical to undertake a comprehensive plan that includes the following components: • Perform a screening of all infrastructure and detail their vulnerabilities. • Prioritize the most vulnerable components and schedule them for upgrade • Develop a cost estimate for strengthening • Perform a feasibility study on the benefits of the strengthening It is noted that most cost estimates are for the United States. However, it is anticipated that the ratio of strengthening to reconstruction costs are similar in the U.S. and the Caribbean. 6.8.2 Reconstruction costs 6.8.2.1 Capital cost Black & Veatch (2012) conducted a study regarding the capital costs for substations and transmissions. They examined different voltage classes, pole structures and transmission line length, terrain, and location. Findings from their work for the following system are presented in this section. • 230 kV single circuit • 4 lines • 30 megavolt ampere (MVA) • 10 km of transmission line • No cost for land acquisition because reconstruction of substations and towers are at existing locations Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 217 • No allowance for terrain is made because either reconstruction or retrofit would have similar constraints. Table 6-10 and Table 6-11 show typical substation and transmission cost calculations, respectively, for the selected example. Note the significant cost associated with reconstruction. Baseline cost $1,650,000 Cost per line $1,440,000 Breaker multiplier 1.5 Lines 4 Transformer $800,000 Shunt reactor $600,000 Series Capacitor $900,000 Transformer cost $13,000,000 Table 6-10. New substation cost data for 230 kV system (adapted from Black & Veatch, 2012) Baseline cost Per km $5800,000 Aluminum Conductor Steel Supported Type Conductor (ACSS) Multiplier 1.08 Transmission Type Lattice tower Multiplier 0.9 Length 5 to 20 km Transmission line Multiplier 1.2 Transmission cost $6,800,000 Table 6-11. New transmission cost data for 230 kV system (adapted from Black & Veatch, 2012) 6.8.2.2 Component cost Midcontinent Independent System Operator (MISO) has developed cost estimates for new construction of expansion of the energy infrastructure (2020). The cost estimates include both overhead and allowance for funds used during construction. Since these could vary between the U.S. and Caribbean, these costs are not included in this section. Findings from their work for the following system are presented in this section. • 230 kV single circuit • 4 lines • 30 MVA • 10 km of transmission line • Towers are 300 m apart • No cost for land acquisition because reconstruction of substations and towers are at existing locations • No allowance for terrain is made because either reconstruction or retrofit would have similar constraints. • The cost data for typical substations, transmission towers, and transmission lines are presented in Table 6-12, Table 6-13, and Table 6-14, respectively. The cost for removal or rebuilding of transmission lines is presented in Table 6-15. Note the significant cost associated with reconstruction of power infrastructure that could be necessary after a major natural hazard event. Component Cost item Cost Material cost $97,000 Installation cost $10,000 Circuit breaker Wiring cost $15,000 Foundation cost (6 m3) $10,000 Cost per unit $134,000 Disconnect switch Material cost $21,000 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 218 Installation cost $10,000 Wiring cost $8,000 Steel stand (1700 kg) $17,000 Foundation cost (6 m3) $10,000 Cost per unit $66,000 Material cost $36,000 Installation cost $3,000 Wiring cost $12,000 Voltage transformer Steel stand (800 kg) $9,000 Foundation cost (3 m3) $5,000 Cost per unit $64,000 Material cost $129,000 Installation cost $3,000 Wiring cost $12,000 Current transformer Steel stand (800 kg) $9,000 Foundation cost (3 m3) $5,000 Cost per unit $158,000 Material cost $9,000 Installation cost $11,000 Wiring cost -- Bus and fittings Steel stand (1,000 kg) $10,000 Foundation cost (5 m3) $10,000 Cost per unit $40,000 Transformer $8000 /MVA $240,000 Table 6-12. New substation cost data for 230 kV system (adapted from MISO, 2020) Steel, kg 7,000 Concrete foundation, m3 18 Material cost $29,000 Installation cost $43,000 Hardware cost $7,000 Foundation cost $33,000 Total cost/tower $111,000 Total cost $3,700,000 Table 6-13. New transmission tower cost data for 230 kV system (adapted from MISO, 2020) Type ACSS Size 1,000 mm2 Installation cost/ m $18 Material cost/ m $14 Accessories cost/ m $1 Total cost/m $33 Total cost $330,000 Table 6-14. New transmission line cost data for 230 kV system (adapted from MISO, 2020) Transmission line removal/km $170,000 Transmission line rebuild/km $900,000 Table 6-15. Transmission line removal and rebuild cost data (adapted from MISO, 2020) 6.8.2.3 Substation cost estimator There are a number of tools available to estimate the capital cost for substations. The reconstruction cost for substations can be computed by computing the estimated number of each type of component that would need to be replaced after a natural disaster by the unit costs for the component. An example is presented in Table 6-16 (PEguru, 2020). Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 219 Component Description Cost Transformer 10MVA to 50MVA medium power transformer $ 800,000 Circuit breaker 230kV 3000 AMP breaker $ 165,000 Circuit switcher 230 kV 3000 Amp switcher $ 85,000 Disconnect switch 230 kV 3000 Amp switch motor operator $ 33,000 Capacitor 230 kV 100MVAR neutral reactor and stand $ 250,000 Voltage transformer 230Kv RELAY GRADE $ 9,000 Current transformer 138kV transformer $ 15,000 Surge arresters 230 kV/180 MCOV porcelain arrester $ 10,500 Service transformer 138kV transformer $ 150,000 Carrier equipment 230Kv TUNER $ 25,000 Insulators 230 kV rated insulator $ 6,000 Control building 20x8 m for 20 panels $ 1,500,000 Transmission structure 230 kV A-frame dead-end 30 m tall $ 560,000 Control cables #6 $20/m Power cables 500 mm2 $1500/m Bus 100 mm Aluminum tube $100/m Table 6-16. Examples of component unit cost for substation components (PEguru, 2020) 6.8.3 Strengthening costs 6.8.3.1 Substations Romero et al. (2013) examined the cost effectiveness of seismic retrofitting of substations in the New Madrid seismic zone of the Central U.S. They examined four cases; see Table 6-17. Examining the cases of “no mitigation” and a mitigation with capital investment of US$10 million, they noted (see Figure 6-26): • The US$10 million mitigation investment reduces the load shed costs and repair costs to approximately 40% and 20%, respectively, of no mitigation cost. • The savings for this case translates to approximately US$12 million annually from an initial investment of US$10 million. Retrofit cost Substation anchorage None 0% US$5 million 60% US$10 million 87% US$20 million 98% US$50 million 100% Table 6-17. Substation retrofit scenarios (adapted from Romero et al., 2013) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 220 Load shed cost Repair cost Figure 6-26. Retrofit effectiveness of substations (adapted from Romero et al., 2013) 6.8.3.2 Transmission towers BPA (2011) has developed a new design for single- and double-circuit transmission towers. The towers use cross-shaped members that are much stronger than typical single angle or double angle members; see Figure 6-27. The new towers, although stronger than traditional design, use less steel and thus can result in savings of up to a quarter of the tower’s cost. Additionally, the towers can be placed more than 300 m apart, which is the typical spacing of traditional towers. The towers are designed for extreme events and can withstand wind speeds of 160 km/hr or more. Tower member Constructed tower Figure 6-27. New transmission tower design (BPA, 2011) Considering the hypothetical example case of 10 km of transmission line, for U.S. application, the typical traditional towers cost of approximately US$100,000 is used for smaller single circuit lines. The cost of the new tower is estimated at US$75,000 per unit. As noted in Table 6-18, the retrofitting cost pays for itself after the first major hurricane impacting the area. Given the propensity of Caribbean countries to experience hurricanes and windstorms on an annual basis, during the design life, this or other retrofitting methods can result in significant cost savings. Note the cost savings do not include the additional savings that result from fees collected because the system has remained operational. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 221 Parameters As-is Upgraded Configuration Double circuit Double circuit Length, km 10 10 No of towers 34 20 Initial upgrade cost 0 US$1.5million Event Cat III or above hurricane Towers damaged 17 (50%) 0 Replacement cost US$1.7million 0 Table 6-18. Example of transmission tower upgrade cost 6.8.3.3 Transmission lines Stockbridge dampers cost between US$2-5 for the quantities typically ordered by the utilities. Including installation, a conservative cost of US$20 per damper is assumed here. The cost of line replacement is assumed as US$30/m. As noted in Table 6-19, the improvements pay for themself after the first major hurricane impacting the area. This savings is multiplied through the years, as the system would be resilient and can withstand repeated wind events. When the loss of revenue from customers due to downtime is added to the existing costs, the savings associated with upgrades are even larger. Parameters As-is Upgraded Wire damping No Yes Length, km 10 10 Spans 35 35 Dampers 0 400 Initial upgrade cost 0 US$8,000 Event Cat III or above hurricane Spans damaged 6 (20%) 0 Replacement, m 1,800 0 Replacement cost US$54,000 0 Table 6-19. Example of transmission line upgrade cost 6.8.4 Life-cycle cost For energy infrastructure, life-cycle cost analysis (LCCA) is typically used in design. This topic will be further discussed in Chapter 8. 6.9 Discussion In subject Caribbean countries, there are large energy (power) infrastructure including a large number of substations and many km of transmission lines. The courtiers are also located in one of the most hazard- prone areas in the world, experiencing a number of hurricanes annually and earthquakes periodically. These natural hazards have caused significant damage to the energy infrastructure in recent years causing economic distresses and delayed recovery. Accordingly, it is time-critical to plan and implement measures that results in a resilient energy infrastructure. Experience has shown that a given sub-class of energy infrastructure, there are a number of components that are the most susceptible to damage in natural hazard events. For these components, effective retrofitting and mitigation methods are available which reduce the expected damage to infrastructure and are cost effective. Table 6-20 presents a summary of findings related to the most critical components of energy infrastructure in the subject countries. Cost, % of initial cost (Est.) Vulnerable Resilience Subsector Hazard Improvement Vulnerability Component Measure Cost Reduction Improved Transformer 5% 20% Substation anchorage Earthquake components Qualified Equipment 10% 40% components Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 222 Equipment Spares 10% -- Add passive Support frames 5% 20% dampers Transmission Strengthen steel 20% 50% Transmission tower members Wind system Transmission Add vibration 5% 50% line dampers Flood, Foundations Add micro piles 40% 80% liquefaction Table 6-20. Summary of findings for power infrastructure in Caribbean Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 223 7. ADDITIONAL MEASURES 7.1 Introduction In this chapter, additional measures that improve infrastructure resiliency are presented. These topics complement the improved design and strengthening measures for various infrastructure sectors discussed in the preceding chapters. When undertaking a country- or region-wide infrastructure program, it is crucial to begin with a project planning matrix and feasibility study that incorporate the measures discussed herein. 7.2 Construction quality management A construction quality management program is a main component to improved design and has a substantial impact on the durability and resistance of infrastructure to natural hazards. A number of practices and procedures used elsewhere for large-scale retrofit projects could be applicable in the 16 subject countries to strengthen building, water, transport, and power infrastructure projects. 7.2.1 Construction database and record keeping As part of a pre-hazard vulnerability assessment, which is discussed in the next section, it is important to compile a database of various structures. For example, in the transport sector, for each bridge, a bridge book can be prepared that includes key information, such as as-built information, inspection reports, and results of vulnerability assessments. The bridge book and database are updated as retrofitting proceeds. This will allow assessing the efficacy of retrofit measures after the next hazard and maintaining an updated database for each bridge14. In the U.S. State of California, Caltrans maintains a list of transportation infrastructure inventory; see Table 7-1. The state’s management plan provides aggregates for the seismic vulnerability of bridges and any planning to address them; see Table 7-2. Item Quantity Highway 9,500 km State highway bridges 13,100 Local bridges 12,200 Table 7-1. California transportation infrastructure inventory (adapted from Caltrans, 2020) Bridge 2019 2029 Total bridge deck,106 m2 23.4 -- Potential seismic vulnerability,106 m2 1.4 0.4 Percentage 6% -- Table 7-2. Seismic retrofit management plan (adapted from Caltrans, 2020) Examples of data maintained in a bridge book are presented in Table 7-3. A hypothetical example for power infrastructure is presented in Table 7-4. Note that as seismic or flood retrofitting progresses, entries are to be updated to indicate alterations. Similar data would need to be collected and archived for all classes of substation. Structure name: Construction information: Year built 1960 Year retrofitted 1998 Length, m 48 14 In the U.S., bridge inspection is mandatory on a biannual basis Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 224 Skew, degrees 20 No. of joints 2 No. of hinges 0 Spans 2@ 24 Railings Min. vertical clearance Width, m 8 No. of lanes 2 Speed, km/h 60 Structure description (original) RC deck on composite four (4) simply supported buildup steel plate girders on two-column RC column bents with cap and RC open end seat type abutments with monolithic wingwalls. Founded on driven RC piles. Structure description (Retrofits) Restrainer cables were placed at each bent on each girder. Steel casings were added to the columns. The footings were also widened which included the addition of new 0.6-m diameter driven steel piles Current condition and recommended action The river has eroded approximately 60 m of the eastern bank upstream. Rehabilitation involves placing rock around the pier and abutments at an estimated cost of $20,000 million to resolve the existing scouring problem Table 7-3. Example of data inventory in bridge book Substation name: Construction information: Year built 1960 Year retrofitted 1998 Number of transformers 3 115 kv Number of circuit breakers 6 68 kV Number of elevated 1 3-phase 115 kv disconnect switches 1 3 phase 68 kV Other components 0 Control building URM retrofitted in 1998 Switch tower Steel lattice Structure description (original) URM control building and 115 kV equipment. One transformer was damaged during the 2013 hurricanes and flood and not repaired. For the two other units, enhanced anchorage plates were added in 1998. At that time, URM control building was also retrofitted with steel rods Structure description (Retrofits) Anchorage for two transformers added Current condition and recommended action Flood protection is needed for the control building. The damaged transformer would require replacement of repair. Currently, there are no spare components at the substation. Table 7-4. Example of data inventory in substation records 7.2.2 Electronic real-time monitoring Effectiveness of seismic retrofitting is highly dependent on the quality of retrofit construction. Therefore, it is important to develop a construction inspection and record-keeping program. Examples of key construction activities as part of retrofitting of transport or building infrastructure include: • Verifying construction complies with structural drawings • Inspecting materials, such as steel and reinforcement, at delivery and collecting mill certification sheets Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 225 • Sampling of concrete and reinforcement • Maintaining records of material testing data • Monitoring pile driving • Checking concrete mix design and delivery time • Checking size, location, and spacing of reinforcement • Verifying calculations for any falsework and excavation, and shoring • Checking shop drawings for steel and prestressing • Conducting or attending safety meetings • Electronically recording daily activities in a log An example of a daily activity log is presented in Table 7-5 and Table 7-6 for transport and power infrastructure retrofit projects, respectively. Bridge no. RPT # 710 State DATE 2022_11_07 District Day M T W T F S S Route SHIFT START 730 STOP 1600 Br No Location Activity W 1 Pier 3 Drive steel piles E 2 Pier 4 Excavate around existing footing W 3 Pier 5 Prepare column for placement of FRP casing W 4 Deck Strip and cleanup; move panels to staging area W 5 Falsework Prep for removal of 30-m beams Activity 1 2 3 4 5 6 7 8  Abs Employee Job Equip. Super 1 Foreman 2 Carpenter Main contractor 3 Pile driver 4 Laborer 5 Laborer 6 Operator 7 Oiler 8 Apprentice 1 Foreman Sub 2 Iron worker 3 Iron worker Table 7-5. Example of construction inspector daily activity log for transport construction or retrofit Substation no. RPT # 710 State DATE 2022_11_07 District Day M T W T F S S Route SHIFT START 730 STOP 1600 Subs No Location Activity 1 Transformer 1 Remove existing bolts 2 Transformer 2 Finish installation of bracket anchorage 3 Breaker Wire ropes were added to the base 4 Switch frame Clamped steel tubes to the frame legs 5 Control room Bracing locations for cable trays were marked Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 226 Activity 1 2 3 4 5 6 7 8  Abs Employee Job Equip. Super 1 Foreman 2 Carpenter Main contractor 3 Steel worker 4 Laborer 5 Laborer 6 Welder Concrete 7 mason 8 Apprentice 1 Foreman Sub 2 Electrician 3 Electrician Table 7-6. Example of construction inspector daily activity log for power construction or retrofit 7.2.3 Use of proven technologies Using proven technologies reduces the recovery time and increases significantly the quality of work. Bailey bridges are steel truss bridges that can be constructed quickly. These bridges were developed during WWII as a rapid replacement for damaged structures. Key advantages of this bridge type include: • Rapid installation • Bridge is launched and thus minimal equipment is needed. In some cases, human power alone can be used to launch these bridges • The bridge is light and strong • Although primarily used as a temporary structure, they can be left in place for long-term use because bridges are designed per AASHTO LRFD code, can have multiple lanes, and incorporate barriers. • They are inexpensive • The bridge can span 80 m or more and thus in many cases eliminate the need for foundations when crossing rivers. This in-turn mitigates flooding hazard and foundation scour. • Since the bridge is assembled in identical segments (see Figure 7-1), the construction quality is improved, as the construction crew will be doing repetitive work. • It is an ideal candidate for use in the aftermath of natural disasters that have damaged or washed away existing bridges. In a number of subject countries, this type of bridge has been used in recent years for this specific purpose; see Figure 7-2. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 227 Figure 7-1. Example of Bailey (steel truss) bridge Trinidad and Tobago Trinidad and Tobago Saint Vincent and the Grenadines Dominica Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 228 Haiti Barbados Jamaica Saint Lucia Figure 7-2. Application of temporary Bailey bridges in subject countries (various sources) Although Bailey bridges can provide a reliable means of transportation when properly constructed, it is important to ensure compliance with manufacturer requirements. For example, bridges need to be assembled properly and constructed using preapproved parts for which the design was based on. Additionally, these are narrower bridges and there are vehicular axle weight restrictions and speed limitations. Failure to comply with all the construction and use requirements can cause damage or collapse of these bridges. Examples in Figure 7-3 show failures due to overload (India), heavy rains and ensuing flooding (likely due to inadequate foundation) (Dominica), and lack of proper inspection during construction (Trinidad and Tobago). Bridge collapse, Uttarakhand India (INDIA TV, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 229 Bridge failure, Dominica (Dominica News, 2017) Bridge collapse, Trinidad and Tobago (Jyoti Communication, 2010) Figure 7-3. Collapse of Bailey bridges due to overload or poor construction 7.2.4 Contract incentives As part of record keeping for new construction or retrofit works, it is important to maintain a daily record of activities, as discussed previously. In addition, records of contractor quality of work for each project can be kept. This is then input into a database and can be used as part of award process for future infrastructure rehabilitation or reconstruction projects. A successful example of such an approach is the building infrastructure construction in Hong Kong, where the Hong Kong Housing Authority (HA) has been developing construction quality management systems and tools since the 1990s. In 1990, the development of the Performance Assessment Scoring System (PASS) was initiated, and this was reviewed periodically and updated in 2001. PASS is an indicator that serves as a tool to measure the quality of work. The project team and PASS assessment team conduct site inspections, desktop reviews, and record checks. The PASS items related to construction quality of structural work can be grouped as follows: Category Item Reinforcement Formwork and falsework Engineering Finished concrete Construction records Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 230 Excavation Site formation Compaction Floors Architectural Walls Management Documentation General Progress assessment Milestone dates for structural work Environment Storage of material Safety Safety program and record Table 7-7 Components of PASS (adapted from HA, 2020) All building construction sites undergo monthly PASS inspections. Every three months, a quarterly PASS composite score for each individual contractor’s projects is completed. The following metrics are developed: • Contractors’ League: considers the overall performance of contractors across all projects being undertaken • Composite Target Quality Score is in the upper quartile (25 percentile) of the league. • Composite Lower Score Threshold is in the lower quartile (75 percentile) of the league. Composite Target Quality Score and Composite Lower Score Threshold are used to assess the overall trend of construction quality over time. HA has adopted a preferential tender award system where preference is given to better-performing contractors by evaluating both the tender price and performance potential. The formula used for tender evaluation is determined by calculating the score of all fully capable tenders and selecting one with the highest score from the following formula: 60 ∗ + 40 ∗ ℎ 7.3 Screening and prioritization Power, water, transport, and building structures are key lifeline infrastructure assets for which vulnerability to natural hazards needs to be identified and mitigated. A critical component is the condition assessments for vulnerability. Prior to undertaking a retrofitting program, it is crucial to undertake a comprehensive risk assessment program to identify vulnerable infrastructure and components for these infrastructures that require retrofitting. This involves site visits, preparing or reviewing as-built drawings, and performing, at the very least, preliminary analyses to confirm the said vulnerabilities. A screening form for infrastructure needs to be developed prior to the field work. Example forms for transport and power infrastructure are presented in Table 7-8 and Table 7-9, respectively. A database of the completed screening forms can then be created. This database can be used to program the infrastructure for strengthening and prioritize the structures based on monetary resources. The database should then be updated after the successful completion of retrofitting projects. Bridge Screening Form (Earthquake, wind, and flood) Description of bridge Inspector Date Longitude Latitude Br. No. Bridge name District Location Structural assessment Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 231 Description Weight Score Abutment seat <300 mm Columns, poorly confined Timber piles Steel girder superstructure, depth >1.2 m Skew ≥30 degrees Simple span precast bridge Gap between abutment backwall and bridge >150 mm No top mat footing reinforcement Fatigue vulnerable detail for steel connections Exposed foundation Evidence of local scour and no riverbank protection Structural score Hazard assessment PGA the site of 1000-year event Soil factor (very competent to highly liquefiable) Design wind speed for site Pier cap below 100-year flood elevation River susceptible to flooding and riverbank has eroded Hazard score Importance assessment Average daily traffic (ADT) Detour length Importance score Overall score Additional comments Table 7-8. Example of bridge screening form Substation Screening Form (Earthquake, wind, and flood) Description of Substation Inspector Date Longitude Latitude Capacity, Substation District Location kV name Structural assessment Description Weight Score Unanchored transformers Unanchored equipment HV equipment does not have seismic plate Support structures designed for seismic forces Control building has no flood protection Transmission tower not designed for code level wind loading Transmission wires have no Stockbridge damping Structural score Hazard assessment Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 232 PGA the site of 1000-year event Soil factor (very competent to highly liquefiable) Design wind speed for site Construction below 100-year flood elevation Hazard score Importance assessment Average people served, daily Redundant substation can provide electricity Importance score Overall score Additional comments Table 7-9. Example of substation screening form 7.4 Preventive maintenance The long-term health of infrastructure depends on regular maintenance and inspection. As part of feasibility studies, the stakeholders should consider all three phases in the service life of infrastructure, namely: design, construction, and maintenance. For example, in the transport sector, for paved roads, AASHTO (2012) requires the following: • Assess pavement condition over time • Identify pavement preservation recommendations • Evaluate the long-term effects of changes in material properties, design, and construction As part of the program, data is collected regularly (typically biennially in the U.S.). Data includes the following: • General information, such as road locations, number of lanes, lane width, shoulder width, pavement type, shoulder type • Layer type, layer thickness, layer strength, subgrade type, drainage information, environmental data, material properties • Traffic data • Pavement condition This data is then organized in a database to warehouse data that can be used to identify key parameters impacting the pavement performance and assess the factors that lead to more robust pavements during natural disasters, such as flooding. Data can also be used to develop recommendations for pavement treatment. Date is then used to assess the efficacy of preventive maintenance. As seen in Figure 7-4, preventive measures lead to extended service life. The cost associated with such measures is significantly less than replacement – which is more likely absent preventive maintenance – cost after natural disasters. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 233 Figure 7-4. Pavement management systems flowchart (AASHTO, 2012) 7.5 Use of advanced technologies for inspection and damage assessment In many of the subject countries, there is a shortage of skilled workers able to conduct detailed regular inspections of transport infrastructure or to perform rapid assessment in the aftermath of natural disasters. New technologies, such as satellite imagery, drones, and light detection and ranging (LiDAR), can be used to perform such tasks quickly, accurately, and can allow for transmission of data digitally in real time. For example, in California, drones are being used as a key component by the transportation department in the biannual structural inspection of transportation infrastructure; see Figure 7-5 (Caltrans, 2020). Unmanned aerial vehicles (UAV or drones) provide the flexibility to inspect bridges and reach hard-to- access areas, collect photographs and videos, take measurements, and document anomalies. The inspection data is then uploaded in real time for review and processing by engineers. Schematics Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 234 Drone used in inspection Performing inspection Figure 7-5. Drone inspection of transportation infrastructure (Caltrans, 2020) Similarly, drones can be used in the aftermath of natural disasters to perform damage assessment reconnaissance. This allows inspection of hard-to-access places. Additionally, since the structures (including inside damaged units) can be remotely inspected, this approach eliminates the need to be placed in unsafe environments to the inspectors, who would otherwise have to visit unsafe structures for data collection. Examples are presented in Figure 7-6. 2014 Napa Earthquake (CA) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 235 2017 Hurricane Irma (St. Martin) Figure 7-6. Post-event damage assessment by drones (various) Light detection and ranging (LiDAR) and Global Positioning Systems (GPS) can be used to monitor landslides and provide warnings in the event of land movement above a certain threshold. Beginning in 2006 and following heavy rainstorms, the 800,000-m3 Ferguson Slide in California blocked the roadway and resulted in significant economic damage. The landslide has been monitored remotely and will be monitored during the construction of a new rock shed. LiDAR and GPS data are collected as part of a system to monitor the displacement, velocity, and acceleration of the slide and to provide warnings; see Figure 7-7. This technology was successfully used in Haiti in the aftermath of the 2010 earthquake. For example, a bridge close to the epicenter was inspected manually and it was also scanned. The damage assessment collected by engineers compared favorably with damage detected during the scan; see Figure 7-8 (Takhirov and Mosalam, 2014). Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 236 Landslide blocking the roadway Monitoring the slide Figure 7-7. Monitoring of Ferguson Landslide (Caltrans, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 237 Photograph of the bridge Scan of the bridge Figure 7-8. Laser scanning of bridge in the aftermath of 2010 earthquake (Takhirov and Mosalam, 2014) 7.6 Monitoring and early warning system (EWS) 7.6.1 United States applications In the U.S., Japan, and elsewhere, networks of strong motion sensors are used to collect earthquake shaking data, determine the expected level of damage, and predict infrastructure response. For example, important bridges in California have been instrumented; see Figure 7-9. In the U.S., the National Hurricane Center (NOAA, 2020) provides key information for tracking and monitoring of hurricanes and provides guidance and advisories. The center also tracks hurricanes in the Caribbean region. Figure 7-10 presents an example of hurricane tracking for Hurricane Irma in the Caribbean region. In California, rainfall is monitored using a system of water gauges and uses radar and satellite information for rainfall projections; see Figure 7-11 (USGS, 2005). Thresholds are set for the incipient flood rainfall and when the thresholds are exceeded, warnings are sent to responsible emergency agencies. A similar methodology can be implemented in the Caribbean to alert the infrastructure managers (for example, transportation officials) of danger of imminent flood. This warning will alert them to close roads and bridges. A reduction in traffic loading on paved roads reduces the severity of structural damage to this infrastructure, because a combination of traffic and flood load is more impactful on paved roads than flood alone. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 238 Figure 7-9. Instrumented bridges (Caltrans, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 239 Figure 7-10. Hurricane Irma in the Caribbean (NOAA, 2020) Figure 7-11. Area of radar coverage (blue regions) and rain gauge locations (red dots) in Southern California (USGS, 2005) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 240 7.6.2 Application to the Caribbean The University of the West Indies Seismic Research Centre (UWI SRC) operates a network of strong motion stations in the region; see Figure 7-12. Data from the networks can be used to expedite response and collect critical information related to earthquake shaking and infrastructure response. The Caribbean Institute for Meteorology and Hydrology (CIMH) provides weather forecasts, including for wind speed and rainfall, for the subject countries. CIMH uses statistical techniques and computer modeling to forecast hurricanes and provides forecasts in 6-hour intervals extending to 72 hours. Figure 7-13 presents an example of a previous hurricane forecast available from CIMH. In a number of the subject countries, rain gauges are used to establish thresholds and provide early warnings for incipient flooding; see Figure 7-14. However, there are two issues for application of this approach to a number of smaller island subject countries that have a volcanic geology, namely: i) response time is limited; and ii) there is limited availability of skilled staff for monitoring. To solve this shortcoming, CIMH is planning to implement a flash flood guidance system based on available flood models, remote satellite data, and in-situ instrumentation. A schematic of the system is presented in Figure 7-15. Currently, the governments of individual subject countries are responsible for issuing the early warning for natural disasters. This effort in-turn is supported by UWI SRC, CIMH, and the Caribbean Disaster Emergency Management Agency (CDEMA). However, the lack of coordination among agencies and delineation of responsibilities varies among the countries. For example, although hydromet agencies are operating continuously, they do not have the legal mandate to issue warnings. To improve the current conditions, CDEMA has initiated an early warning systems (EWS) project with the following objectives (CDEMA, 2020): Strengthening integrated early warning systems for more effective disaster risk reduction in the Caribbean through knowledge and the transfer of tools. Strengthen integrated and cohesive preparedness capacity at a regional, national and community level in the Caribbean. The checklist developed by CDEMA is presented in Table 7-10. Note that one of the key action items is to integrate the EWS among various countries. CDEMA (2020) states: This project, which is being implemented in Antigua and Barbuda, Dominica, the Dominican Republic, Haiti, Saint Lucia, Saint Vincent and the Grenadines and Cuba over an 18-month period, seeks to strengthen disaster preparedness and risk reduction through Integrated Early Warning Systems (EWS). Working in this important element for disaster risk reduction, is expected to enhance the prevention, mitigation and response capacities at both the institutional and community level, based on mutual learning and collaboration between countries and regional institutions working in disaster risk reduction across the Caribbean. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 241 Jamaica (8 stations) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 242 Eastern Caribbean Figure 7-12. Seismic strong motion stations in subject countries (UWI, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 243 Figure 7-13. Forecast track for the 1998 Hurricane Georges (CIMH, 2020) Figure 7-14. Recorded rain gauge data in Dominica during Hurricane Dean (Smith et al., 2008) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 244 Figure 7-15. Schematics of flash flood guidance (HDRFFG, 2016) Actions (partial list shown) Monitoring network established that monitors hazards that impact your country Measurement parameters and specifications documented for each relevant hazard Technical equipment, suited to local conditions, circumstances and gender differentiated needs, in place and personnel trained in its use and maintenance Monitoring data received, processed and available in an interoperable format in real time, or near-real time Monitoring data and metadata routinely curated with quality controls, archived and accessible for verification, research purposes and other applications Monitoring hardware and software maintenance is addressed routinely and costs and resources are considered from the beginning to ensure the optimal operation of the system over time The system is able to combine and benefit from new and older technology allowing for exchange of data among countries with different technical capabilities Data analysis and processing, modelling, prediction and warning products generation based on accepted scientific and technical methodologies and disseminated within international standards and protocols. New data analysis and processing, modelling, prediction and warning products can be integrated easily in the system as science and technology evolve. Warning centres are operational at all times (24 hours/day, seven days/week) and staffed by trained personnel, following appropriate national and international standards. Warnings generated and disseminated in an efficient, timely manner for each type of hazard. Warning system(s) subjected to regular system-wide tests and exercises Table 7-10. Portion of EWS (adapted from CDEMA, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 245 7.7 Response protocol 7.7.1 International best practice A post-event inspection (damage assessment) program needs to be implemented by regional groups or national governments to rapidly respond in the event of a natural disaster impacting infrastructure. As part of this program, an inspection form needs to be used; an example is shown in Figure 7-16 and Figure 7-17 for building structures and for earthquakes or hurricane and flood, respectively. These forms can be updated to tailor to construction typologies in the Caribbean and similar forms for other infrastructure can be developed. Training and capacity building for government engineers and consultants needs to be undertaken in order to certify sufficient numbers of qualified engineers to respond after an event. Refresher online training can be conducted every several years to ensure trainees are up to date with any changes. A database of certified individuals should be maintained, including contact information to allow timely response. Because in the aftermath of a natural hazard event, unsafe conditions could be present, safety training should be incorporated as part of the overall program and safety information should be included in the booklet. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 246 Figure 7-16. Post-earthquake damage assessment form (ATC, 2020) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 247 Figure 7-17. Post-hurricane and flood damage assessment form (ATC, 2020) 7.7.2 Applications to the Caribbean In recent years, both electronic and mobile-based assessment forms have been developed. Additionally, although these forms have been developed for use in the U.S., modified versions have been developed to Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 248 account for local construction practices and are used in a number of countries. For example, a version tailored for Haiti was used in the aftermath of the 2010 Earthquake in Haiti (Miyamoto and Gilani, 2012). Currently, most governments dispatch engineers to perform post-event assessments. However, there is no formal training, uniform documentation, or standardized program. Organizations such as CDEMA can undertake development of a standardized program across the region comprising the following: • Preparation of paper and digital forms for data collection • Mechanism for storing and accessing collected data • Providing training and certifications to inspectors • Maintaining a list and contact information of inspectors for rapid development after major events • Implementing a safety protocol for inspection 7.8 Regional approach A number of measures proposed in this chapter rely on the subject countries having both technical capacity and financial resources to implement such actions. While this may be the case for counties such as Trinidad and Tobago, Jamaica, and Barbados, implementation would be challenging for smaller countries due to a lack of funds or availability of qualified personnel. This highlights the need to undertake a region-wide approach to implementing such measures and other actions that can enhance infrastructure resiliency in the Caribbean. Current examples of such coordination include: • CUBIC building code (see Chapter 3) • Seismic strong motion instrumentation and hurricane monitoring by SRC and CIMH (see Section 7.6) • Development of EWS by CDEMA (see Section 7.6) • Regional climate change adaptation and disaster risk management by the Caribbean Catastrophe Risk Insurance Facility (CCRIF SPC, 2020) that provides financial resources to limit the immediate impact of natural disasters to (19) member countries • Caribbean Building Inspector Training Courses (CDMP, 1999) which involve training of building inspectors at the University of Technology in Jamaica However, the regional organizations responsible for these efforts are dependent on non-regional funding for these programs, and they are frequently not sustained after the funding expires. Examples of low-cost collaborative efforts that can be implemented regionally include the following, and these steps would provide components for development of a regional infrastructure program: • Regional and departmental levels plan regular and periodic communication for development of a resilient planning platform. • Coordination for adoption of design codes • Implementation of a uniform material standards for construction • Open-source sharing of data from natural disasters, including structural damage data and recording hazard intensities • Provisions for uniform minimum design and engineering standard of practice and care across the region • Documentation and sharing of cost for construction of infrastructure • Recoding of costs and associated downtime associated with various infrastructure damage as a consequence of natural hazards 7.9 Discussion As part of project planning and feasibility studies for retrofitting of key infrastructure assets, it is critical that the infrastructure owners and managers consider the entire service life of the structures, including Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 249 design, construction, and maintenance. While the primary focus of the report has been on improved design practices, both construction and maintenance play a critical role in developing and sustaining a resilient community and network of infrastructure. Improvements in construction quality management and regular preventive maintenance are both critical. Additional measures, such as development of a vulnerability database, instrumentation, including early warning systems, and response protocols, including rapid post- event assessments, are complementary actions that can enhance the likelihood of a successful long-term project and reduce human and economic losses. Further, it will protect infrastructure in the event of natural disasters. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 250 8. MITIGATION COST AND FINANCING 8.1 Introduction Infrastructure planners face competing economic constraints. These projects are typically not inexpensive. Additionally, given the scope and complexity, there is inherent risk (for example, a change in scope, delays in meeting a schedule milestone, or increased cost) associated with infrastructure improvement projects. To implement strengthening techniques, it is critical to conduct a feasibility study (FS). As part of this study, the focus should be on life-cycle cost (LCC) and not on initial cost, because infrastructure are typically in service for decades and accumulated upkeep and repair costs over service life can be significantly more for an option with a lower initial cost. To assist with financing and project management, FS could incorporate and examine innovative contracting options that allow the owners to share some of the project risks with qualified contractors. These topics are discussed briefly in this chapter. 8.2 Cost effectiveness of mitigation In a 2019 study, it was found that for every US$1 that communities spent on mitigation, there was a multiplier in savings for various infrastructure and for different natural hazards. Table 8-1 (NIBS, 2019) summarized findings for natural hazards and infrastructure considered in this report. Note that mitigation will result in significant savings; for US$1 spent on earthquake or flood mitigation for lifelines, savings of US$3-8 could be realized. Infrastructure Case Buildings Lifelines15 River flood 6:1 8:1 Natural hazard Storm surge 7:1 No data Wind 6:1 7:1 Earthquake 13:1 3:1 Table 8-1. Cost-benefit analysis for natural hazard mitigation (adapted from NIBS, 2019) 8.3 Local costs Data from some infrastructure reconstruction for various infrastructure was collected from some of the subject countries; see Table 8-216. Sector Subsector Typology Country Cost (XCD17) Asphalt 162,000 Per km Paved roads Concrete 1,500,000-1,800,000 of road Transport Precast concrete Per m2 of Bridges Saint Vincent 1,000-2,000 girder deck and the School Reinforced concrete 150-1,250 Grenadines Hospital Reinforced concrete 1,300-1,800 Per m2 of Buildings Government Reinforced concrete 2,800 floor Hospitality, hotel Reinforced concrete 200- Table 8-2. Example of infrastructure reconstruction cost in the Caribbean (various) 15 Power, water, and transport infrastructure 16 Retrofitting cost data was not available 17 Eastern Caribbean dollar Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 251 8.4 Life-cycle cost analysis 8.4.1 Overview For infrastructure, the use of initial capital cost alone as a decision-making criterion on whether or not to undertake mitigation is not representative. To assess the cost and savings associated with improvements, it is important to perform engineering economic calculations based on a long horizon that is typically several decades and accounts for all future expenditures during the service life of infrastructure. For infrastructure projects, Life-Cycle Cost Analysis (LCCA) is a better tool. LCCA is defined as performing analysis and comparing the merits of various alternatives. In the context of this report, there are two alternatives: i) do nothing; and ii) undertake mitigation. LCCA allows for a rational and structural approach in deciding which, if any, measures should be undertaken. The general approach is to examine all the costs and associated savings throughout the analysis time and account for the discount rate , defined as the net difference between interest and inflation rates. The option with the lowest net cost is then considered the optimal choice and can be assigned a higher priority than the options that have higher net costs associated with them. 8.4.2 Application to paved roads In the U.S., most state transportation departments use LCCA for design of paved roads. FHWA (1998) provides the basic framework and procedure for conducting LCCA. The FHWA approach is primarily focused on future recurring costs due to wear and tear, because better constructed (or strengthened) roads are expected to perform better than existing roadways in natural hazards (primarily flooding). Thus, similar approaches could be utilized to assess the efficacy of resilient design. The FHWA LCC analysis is incorporated as part of the pavement design software. The FHWA guidelines recommend the following: • Inclusion of both agency and user costs • Minimum analysis period of 35 years • Development of activity list (maintenance strategy) for roadways • Agency costs, including initial engineering, contract administration, construction costs, future routine maintenance costs, resurfacing, rehabilitation, and traffic maintenance costs • User costs are the differential costs to the travelers between the alternatives and include delay costs, crash costs, vehicle operating costs, and delay (time) costs. Batouli et al. (2017) conducted a LCCA of pavement designs. Figure 8-1 presents the cumulative cost per unit length of road for the two alternatives. Note that the flexible pavement has a lower initial cost but considerably higher maintenance cost and two additional rehabilitation (re-surfacing) costs at 20 and 40 years in the design life. Note that the LCCA shows that the concrete pavement, although at a higher initial cost, is a more cost-effective solution than keeping the existing asphalt surface; in other words, mitigation pays for itself at most within the first 20 years. If there are earlier natural hazards, such as flooding, then this time is shrunk even more. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 252 Figure 8-1. LCCA for example paved road (adapted from Batouli et al., 2017) 8.4.3 Application to bridges AASHTO (2019) presents the framework for bridge LCCA. Agency costs are defined as construction, maintenance, rehabilitation, and replacement costs. User costs relate to the functionality of the bridge and include load posting, closures, higher accident rates, longer travel times, and higher vehicle operation costs to use detours. In addition, the risk vulnerability can be included in the cost model due to vulnerability to earthquakes and other natural disasters. Hatami and Morcous (2013) performed LCCA for bridges in the U.S. State of Nebraska. A bridge was chosen to assess the efficacy of applying overlay to the bridge deck (thus the two analysis conditions were either “do nothing” or “mitigate”). The key properties of the bridge are summarized in Table 8-3. This bridge is similar to the types of structures found in the subject countries. As shown in Figure 8-2, the overlay adds an initial construction cost, but the option becomes cost-efficient approximately 18 years into the life of the bridge. For bridges, the typical service life is 50 to 100 years and thus, this mitigation is cost-effective. Note that user costs were not included in analysis. Inclusion of user costs would make mitigation even more cost-effective. Item Value Length, m 78 Width, m 14 Spans 3 Location Urban Superstructure Steel girder with concrete deck Table 8-3. Key properties of the bridge (adapted from Hatami and Morcous, 2013) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 253 Figure 8-2. LCCA for example bridge (adapted from Hatami and Morcous, 2013) 8.4.4 Application to buildings Porter et al. (2004) examined the average annual loss (AAL) for reinforced concrete buildings. These structures have similar construction to buildings in Caribbean. Figure 8-3 presents the mean seismic vulnerability functions. Note that for a PGA of 0.3-0.4 (note that this is similar to the level of seismicity in the Caribbean; refer to Section 2.2.1), the mean damage factor is reduced by a factor of approximately nearly ten when earthquake-resilient mitigation is implemented. The computed AAL for the buildings are 1.14%, and 0.15%, respectively. Thus, there is significant reduction in the expected annual losses once mitigation is undertaken. Figure 8-3. Mean seismic vulnerability function for buildings (adapted from Porter et al., 2004) 8.4.5 Application to energy sector De Leon et al. (2009) examined the life-cycle cost functions for substations and transmission towers in Mexico subject to wind loading. They noted the following: Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 254 [I]t is observed that the scenario “with” the additional recording implies the most economical option because, as a result of the improvement on prediction, better design specifications are developed and the facilities reliabilities are enhanced. As the potential losses increase, the economical advantage also increases. Therefore, the critical facilities should have more closely and systematic recording and monitoring. Loo et al. (2010) investigated the LCCA of switchgears. They noted over time, a lower percentage of failure (and lower LCCA) is obtained when preventive maintenance was included; see Figure 8-4. Figure 8-4. LCCA functions for energy sector (adapted from Loo et al., 2010) 8.5 Project contracting 8.5.1 Overview The scope of retrofitting of infrastructure assets in the subject countries is expansive, will take a number of years, and will be resource-intensive. Examples of innovative contracting options that provide means for cost-sharing and (some) risk transfer are presented herein. 8.5.2 Design Bid Build (DBB) In DBB, the work is performed sequentially and is the most common type of contracting. The agency or their consultants design the infrastructure. Plans and design criteria are then advertised in a bid package. The interested contractors will submit bids for the construction and the selection of a successful bid would be based on criteria developed by the agency. Some disadvantages of DBB are (Caltrans 2020): • Lack of input from construction industry could lead to more expensive design • If the selection is based on lowest bid alone, the quality of construction might not be optimal, and a more extensive agency inspection program would be required. • There are no incentives for contractors to expedite the project or provide higher quality product. 8.5.3 Design Build In design-build, there is a single contract between the project owner and a design-build contractor who is responsible for both design and construction of an infrastructure retrofit project. The agency sets the project requirements, such as design parameters, expected timeline for construction milestones, and quality Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 255 performance criteria. A comparison between project schedule for DBB and design-build is presented in Table 8-4 (USDOT, 2006). Approach Project phase Concept planning Select Engineer Design bid Preliminary design build Select design build Final design Construction Concept planning Select Engineer Preliminary design Design bid Select design build Final design Construction Table 8-4. Design build and DBB project approach (adapted from USDOT, 2006) Table 8-5 presents the comparison between the two contracting approaches for key construction parameters (USDOT, 2006). Note that utilizing design-build will reduce construction time and cost without impacting the quality. The reduction in construction time is most pronounced and is most critical for infrastructure retrofit projects, as responsible agencies would prefer to minimize the service disruption time. Parameter Completion schedule Project cost Construction quality Impact of design build vs. BDD -14% -3% 0% Table 8-5. Example of a successful procurement process (adapted from USDOT, 2006) 8.5.4 Public-private partnership (P3) In public-private partnership (P3), a private entity or developer takes part in financing the infrastructure project in return for monetary compensation based on contractual authorization to collect revenues. The private entity will be responsible for financing. The private entity will be responsible for financing, design, and construction. FHWA encourages state transportation departments to enter into public-private partnerships as a means of securing additional financial resources. An example of a successful procurement process is presented in Table 8-6 (FHWA, 2019). Stage Component Description The P3 feasibility analysis helps determine whether a project can or Feasibility Analysis should be implemented as a P3 project and, if so, the type of P3 to use and the related contractual and risk arrangements for project implementation. The agency to engage its technical, legal, and financial advisors early in Advisors the procurement process. The agency will often set up informal market sounding meetings or phone Market Soundings calls with interested firms to obtain information that firms may be reluctant to share in written responses Pre-procurement Early engagement and coordination, prior to procurement, has been found National government to significantly help in ensuring efficient access to national programs and support resources without impacting the desired procurement Approval from or cooperative efforts by State or local agencies and other Local Support entities may be required for a P3 procurement. A workshop prior to issuing the procurement documents helps to obtain Industry Forum comments and input from the private sector regarding the proposed procurement process. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 256 Establish procedures ensuring confidentiality of information and Confidentiality documents Selection process Issuance of documents to the shortlisted proposers. Initiation of Obtain input and comments from the shortlisted proposers, including one- Procurement Review process on-one meetings with proposers Revisions An opportunity for proposers to submit Alternative Technical Concepts Pre-selection Evaluate technical and financial proposals on a pass/fail and qualitative Evaluation Committees basis. Selection process A best value selection process or a qualifications-based selection process Engagement in Reservation of the right to engage in discussions with proposers and to Selection/ discussions request proposal revisions. Negotiations Negotiate with the selected proposer prior to final award. Stipends Pay stipends to all shortlisted firms if the procurement is cancelled. Table 8-6. Example of a successful procurement process (adapted from FHWA, 2019) 8.5.5 Construction Manager/General Contractor (CMGC) Caltrans (2020) writes: The CMGC project delivery method consists of two phases-design and construction. When the owner considers the design to be complete, the construction manager then has an opportunity to bid on the project based on the completed design and schedule. If the owner, designer and independent cost estimator agree that the contractor has submitted a fair price, the owner issues a construction contract and the construction manager then becomes the general contractor. The contractor acts as the consultant during the design process and can offer constructability and pricing feedback on design options and can identify risks based on the contractor's established means and methods. As noted earlier, this process also allows the owner to be an active participant during the design process and make informed decisions on design options based on the contractor's expertise. The advantages of Construction Manager/General Contractor (CMGC) contracting include: • Encourages innovations as contractor and owner work together to devise design and construction solutions. • Improved design. Contractor reviews and provides constructability review. • Risk mitigation. Obstacles not known during design in DBB projects will be brought up by the owner-contractor team. • Expedited delivery. Project schedule is optimized as potential issues are discovered and resolved during the design process. 8.6 Discussion Infrastructure planners face competing economic constraints. Experience has shown that retrofitting projects are extremely cost effective and for every US$1 spent on mitigation, savings of US$6 to 13 are realized. To implement strengthening techniques, it is critical to conduct a feasibility study. Utilizing a database of local costs would provide an estimate of initial capital costs. Given the long service life of infrastructure, performing LCCA would provide a realistic cost and savings associated with undertaking the retrofitting measures. Innovative contracting options can be used by infrastructure owners to reduce construction time and share project risks with contractors. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 257 9. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 9.1 Summary 9.1.1 Overview Buildings, water, transport, and power infrastructure are key assets for the Caribbean. The continuous operation of these critical structures allows for continuous economic growth and societal wellbeing. Damage to the infrastructure results in direct economic losses, operation interruption, and has cascading effects on other infrastructure. Buildings, water, transport, and power infrastructure have been vulnerable to natural disasters, like hurricanes, flooding, and earthquakes, in the past. Given its geographical location, the Caribbean countries in this study are vulnerable to a wide range of natural disasters; see Table 9-1. Country Hurricane Earthquake Flood Antigua and Barbuda 4 3 Bahamas 4 Belize 4 4 Barbados 3 Dominica 3 Guinea 3 4 Grenada No data Guyana 4 Haiti 6 4 4 Jamaica 4 2 Saint Kitts and Nevis 5 2 Saint Lucia No data Sint Maarten No data Suriname 4 Trinidad and Tobago 3 Saint Vincent and the Grenadines 4 Table 9-1. Relative importance of hazards in Caribbean countries based on mortality index (ISDR, 2009) The subject countries have been exposed to a large number of hurricanes in recent years with significant economic consequences. For example: • In Dominica, Hurricane Maria (2017), with maximum wind speeds of 275 km/hr, damaged the roof of almost every house, caused severe damage to roads and public buildings, downed power lines, and caused US$1.4 billion in damage. • In The Bahamas, Hurricane Dorian (2019), with sustained wind speeds of 300 km/hr, caused at least 40 fatalities, and damage of over US$7 billion; see Figure 9-1. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 258 Before After Figure 9-1. Impact of Hurricane Dorian in the Bahamas (ESRI, 2019) 9.1.2 Critical infrastructure Table 9-2 presents the key infrastructure, typologies, and vulnerable components considered in this report. The table also identifies the expected governing natural hazards for various components. Governing Sector Subsector Type Components Hazard(s) Wall Earthquake URM Roof Wind Shape and figure Earthquake RC/CF with infill Earthquake, Schools, Hospitals, Wall wall Flood Government Roof Wind buildings, Shape and figure Earthquake Emergency centers, RC frame Ductility Earthquake Hospitality and External wall Wind BUILDINGS hotels Earthquake, Stiffness Wind Steel frame Strength Earthquake Roof Wind Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 259 Governing Sector Subsector Type Components Hazard(s) Earthquake, Stiffness Steel frame high- Wind rise Ductility Earthquake Cladding Wind Roof and foundation Wind, Flood Wood frame Elevation Flood Earthquake Foundation liquefaction Subsurface Flood Roads Asphalt paved Surface Flood Columns Earthquake Concrete girder Joints Earthquake Slender members Earthquake Bridges Steel truss Riveted connections Earthquake TRANSPORT Bearings Earthquake Steel girder Cross frames Earthquake Structural details Earthquake Power plants RC low rise Nonstructural components Earthquake Grade construction Flood Building vulnerability Earthquake Control buildings Grade construction Flood Substations Unanchored equipment Earthquake Equipment Lack of seismic components Earthquake POWER Metal lattice poles Wind Metal lattice main Transmission lines lines Transmission lines Wind Plant building Wall and ductility Earthquake (Masonry/RC Roof and external wall Wind Water treatment frame) Envelope and elevation Flood plant and Water storage Foundation and anchorage Earthquake Wastewater Equipment Earthquake, Foundation treatment plant (mechanical and Wind electrical) Elevation and flood proofing Flood Earthquake WATER Foundation liquefaction Jointed concrete Earthquake. Joints Pipelines pipes liquefaction Ductile iron pipes Joints and welded flanges Earthquake Table 9-2. Matrix for the various infrastructure sectors 9.1.3 Vulnerable components Table 9-3 presents the components that have experienced the most damage in past events and that are considered in this report. Sector Subsector Type Vulnerability Hazard Out-of-plane capacity Earthquake Schools, Hospitals, URM In-plane capacity Earthquake Government Roof connection Wind buildings, Structural irregularity Earthquake Emergency Infill wall connection Earthquake centers, RC/CF with infill wall Hospitality and Roof connection Wind BUILDINGS Building envelope Flood hotels RC frame Structural irregularity Earthquake Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 260 Sector Subsector Type Vulnerability Hazard Non-ductile detailing Earthquake External wall Wind Stiffness Earthquake, Wind Steel frame Structural capacity Earthquake Roof connection Wind Stiffness Earthquake, Wind Steel frame high-rise Non-ductile detailing Earthquake External curtain wall Wind Roof connection Wind Wood frame Foundation ties Wind, Flood Elevated construction Flood All Foundation Earthquake liquefaction Subsurface Roads Asphalt paved Flood Surface Columns Concrete girder Foundation Joints Earthquake ground Slender members Steel truss motion Riveted connections Bridges Bearings Steel girder Cross frames TRANSPORT Welded connection Steel bridges Wind details Suspension bridge Bridge stability Wind All Foundation Earthquake liquefaction Earthquake ground Transformers Inadequate anchorage shaking Substations Low stiffness and Support frames Surface damping Eclectic components Inadequate design Columns Wind and earthquake Steel tower Slender members shaking POWER Transmission Transmission lines Vibration Wind towers Earthquake liquefaction Foundation Inadequate capacity and flood Out-of-plane capacity Earthquake In-plane capacity Earthquake Non-ductile detailing Earthquake Plant building Roof connection Wind (Masonry/RC frame) External wall Wind Water treatment Building envelope Flood plant and Elevated construction Flood Wastewater Foundation Earthquake treatment plant Water storage Anchorage Earthquake Foundation tie Earthquake, Wind Equipment (mechanical Elevating equipment Flood WATER and electrical) Dry floodproofing Flood All Foundation Earthquake liquefaction Earthquake ground Pipe Brittle pipes shaking Pipelines Earthquake liquefaction Joints with limited Earthquake ground Joints deformation capacity shaking Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 261 Sector Subsector Type Vulnerability Hazard Earthquake liquefaction Table 9-3. Vulnerable components of the various infrastructure and the corresponding hazards 9.1.4 Design codes and standards The Caribbean Uniform Building Code (CUBiC) was developed within and by the Caribbean community in 1985, and it includes structural design requirements for gravity, wind, and seismic load (CARICOM, 1985). Since the degree of enforcement of this building code varies by country, some countries of the Caribbean have adopted this code as part of the design process, but some treat this as a supplemental document. In Trinidad and Tobago, a multi-year bridge assessment and reconstruction project was completed in the early 2010s (Khan-Kernahan, 2013). As part of this project, 96 bridges underwent rapid assessment. From that pool, 70 bridges were selected for detailed assessment, 40 of which were selected for reconstruction. These reconstructed bridges were designed by international engineering consultants and the bridge design followed the American Association of Highway and Transportation Officials (AASHTO) design procedures in effect at the time. ASCE 113 (2007) provides design criteria for power infrastructure structures. Although intended for the U.S. and North America, the document is used worldwide, including in the Caribbean. ASCE 113 wind design is based on provisions of ASCE 7. Similarly, the Pan American Health Organization (PAHO) has developed wind speeds to be used with wind provisions of ASCE 7 (2019). The buildings and equipment for water treatment plants (WTP) and wastewater treatment plants (WWTP) in the Caribbean are designed and constructed by utilizing international design codes or Caribbean building codes. The construction of water supply facilities refers to country codes in addition to the codes of the U.S. or Europe. 9.1.5 Strengthening and resilience measures Table 9-4 presents examples of proposed strengthening measures for the vulnerable components identified for each infrastructure sector. Sector Subsector Type Resilience Measures Install steel brace or strongback and anchors Add or overlay RC shear wall URM Improve connection anchors of roof and wall with new bond beam Balance lateral-force resisting system by adding RC shear wall Improve connection anchors of infill wall and Schools, Hospitals, RC/CF with infill wall diaphragm Government Improve connection anchors of roof and wall buildings, Install flood shield on door or window Emergency Balance lateral-force resisting system by adding centers, RC shear wall Hospitality and BUILDINGS RC frame Overlay FRP sheet, steel jacket or RC jacket hotels Add seismic isolation Improve connection of external wall Add steel braced frame or shear wall frame for stiffness Steel frame Improve strength by adding new braces/walls or moment frames. Improve connection elements of roof and frame Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 262 Sector Subsector Type Resilience Measures Add steel braced frames or shear wall frame for stiffness Improve beam-column connection for ductile Steel frame high-rise behavior Add energy dissipation damper Improve external wall attachment details Improve connection anchors of roof and wall Wood frame Strengthen foundation anchor bolt and stiffener Construct floodwall or floodgate Enlarge spread footing All Add pile foundation Install soil grouting or stone/gravel columns Compact subsurface Provide drainage Roads Asphalt paved Use concrete surface Use thicker asphalt Add cable restrainers Concrete girder Add abutment seat extenders Stiffen members Add BRB Steel truss Replace riveted connections with bolted connections Place cable restrainers at bearings Steel girder Upgrade cross frames with BRB Bridges TRANSPORT Steel bridges Improve welded connection detailing Suspension bridge Increase torsional stiffness Add steel casing Columns Add FRP casing Add pierwalls Add isolation casing Foundation Add new piles and enlarge foundation Mitigate for liquefaction and scour Improve anchorage Transformers Base isolation Stiffen the frame Substations Support frames Wire rope dampers Spring dampers Use of qualified components Eclectic components Flexible connectors POWER Retrofit slender members Steel tower Transmission Add damping towers Transmission lines Add vibration damping Foundation Add new piles Install steel brace or strongback and anchors Add or overlay RC shear wall Overlay FRP sheet, steel jacket or RC jacket Plant building Water treatment Improve connection anchors of roof and wall (Masonry/RC frame) plant and Improve connection of external wall Wastewater Install flood shield on door or window treatment plant Construct floodwall or floodgate WATER Add new internal/external concrete foundation Water storage Strengthen anchorage between storage and foundation Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 263 Sector Subsector Type Resilience Measures Improve foundation anchorage and stiffener Equipment (mechanical Elevate equipment on new frame and electrical) Relocate equipment at upper location Install watertight barrier Enlarge spread footing All Add pile foundation Install soil grouting Replace brittle pipes with ductile pipes (DI, Pipe Pipelines HDPE etc.) Joints Add joints with large deformation capacity Table 9-4. Vulnerable components of the various infrastructure and the corresponding hazards 9.1.6 Cost considerations Figure 9-2 presents the probability of the infrastructure being in a given damage state when subjected to an expected natural hazard. Note that once retrofitting, as outlined above, is implemented, the probability of the infrastructure experiencing significant damage or collapse is reduced from approximately 80% to approximately 20%. Such savings should be considered when planning for and allocating the initial capital costs of strengthening. To obtain an estimate of retrofit and reconstruction costs, an extensive literature survey was conducted, and local sources were contacted. Table 9-5 presents a summary of the findings related to the most critical components of infrastructure in the subject countries. Figure 9-2. Distribution of damage states Vulnerable Cost, % of initial cost (Est.) Subsector Hazard Resilience Measure Improvement Vulnerability Component Cost Reduction Schools, Add wall bracing (anchor and Wall out of plane 15% 40% Hospitals, strongback) Government In plane Add RC shear wall or Steel brace 20% 60% buildings, Earthquake Structural Add Shear wall, Moment frame or 15% 40% Emergency shaking irregularity Braced frame centers, Ductility Add Column jacketing or Beam retrofit 10% 30% Hospitality Stiffness and Add Steel brace or Steel plate shear and hotels 15% 50% capacity wall Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 264 Vulnerable Cost, % of initial cost (Est.) Subsector Hazard Resilience Measure Improvement Vulnerability Component Cost Reduction Energy Add Seismic isolation or Energy 40% 80% absorption dissipation damper Enlarge Spread footing or Add pile Foundation 40% 80% Earthquake foundation liquefaction Install Soil grouting or Stone/gravel Soil 30% 80% columns Roof connection Improve Roof connection anchor 10% 50% External curtain Improve Wall connection attachment 10% 30% Wind wall and wall Foundation Add Connection anchor bolts and 10% 50% connection stiffeners Building Install flood shield 10% 80% envelope Flood Elevated Elevate floodwall 15% 80% construction Foundation tie Strengthen anchoring to foundation 10% 50% Columns Add FRP or steel casing 10% 60% Bearing Add cable restrainers 2% 50% Earthquake Abutment, pier ground Add concrete bolsters 5% 50% seats motion Bridges Steel cross Fatigue-resistant details 10% 80% frames Flood Add riprap 5% 20% Foundations, Flood or abutments Add new piles 40% 80% liquefaction Add sealant 2% 20% Paved roads Flood Paved surface Strengthen (thicker pavement) 20% 50% Concrete surface 40% 80% Transformer Improved anchorage 5% 20% Substation Equipment Qualified components 10% 40% Earthquake components Equipment Spares 10% -- Support frames Add passive dampers 5% 20% Transmission Strengthen steel members 20% 50% Transmission tower Wind system Transmission Add vibration dampers 5% 50% line Flood or Foundations Add micro piles 40% 80% liquefaction Wall out-of- Add wall bracing (anchor and 15% 40% plane (Building) strongback) WTP/WWTP In-plane : Plant Add RC shear wall 20% 60% (Building) building Earthquake Ductility (Masonry/RC Add column jacketing 10% 30% shaking (Building) frame), Water Foundation storage, Add concrete foundation 40% 50% (Water storage) Equipment Anchorage Strengthen anchorage 10% 30% (Water storage) Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 265 Vulnerable Cost, % of initial cost (Est.) Subsector Hazard Resilience Measure Improvement Vulnerability Component Cost Reduction Foundation tie Improve foundation tie 10% 20% (Equipment) Enlarge spread footing or add pile Earthquake Foundation (All) 40% 80% foundation liquefaction Soil (All) Install soil grouting 30% 80% Roof connection Strengthen roof connection 10% 50% (Building) External wall Wind Improve wall connection 10% 30% (Building) Foundation tie Improve foundation tie 10% 20% (Equipment) Building envelope Install flood shield 10% 80% (Building) Elevated construction Construct floodwall 15% 80% Flood (Building) Elevating Elevate equipment 20% 80% (Equipment) Dry floodproofing Install watertight barrier 15% 80% (Equipment) Earthquake Underground Replace with ductile pipes Shaking and Pipes 40% 80% pipelines Trenchless technologies liquefaction Table 9-5. Estimated monetary savings for implementing infrastructure resilient measures in the Caribbean 9.2 Conclusions Based on the review of infrastructure typology and geographical distribution, hazard maps, recorded past damage, the construction practice, and relative costs, the following is noted: • The 16 Caribbean countries in focus in this report are geographically located in areas that can experience a number of natural hazards, which could damage buildings, water, transport, and power infrastructure. The consequences include not only direct structural damage, but also secondary economic losses due to loss of revenue from inoperability, delayed recovery, and cascading costs to other infrastructure. • Efficacious and economically-feasible strengthening options are available. These measures rely on well-proven techniques and are suitable to the Caribbean countries, given the availability of material and skilled labor resources. Once implemented, such measures would reduce infrastructure vulnerability by multiples of the initial capital cost. New technologies can be utilized to implement robust solutions and enhanced resiliency. • The service life of infrastructure is comprised of planning, design, construction, maintenance, and post-event monitoring. Although structural strengthening is a key component of infrastructure improvements, other phases should also be considered for a successful enhanced resiliency program. Devising and implementing a construction quality management plan would ensure that retrofits are built as designed and would add to the service life of infrastructure, reducing the need for corrective maintenance. Preventive maintenance and inspection would reduce the cost of expensive future repairs. Early-warning and post-event inspection can result in reduced delays and expedite the return to functionality. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 266 • Creative financing measures and life-cycle cost analyses (LCCA) can be used to identify and mitigate project risks, accelerate project delivery, and coordinate the design and construction phases by incorporating input from experts in both fields. 9.3 Recommendations Based on the review of hazard, infrastructure typology and vulnerability, and steps taken during data collection, the following is recommended • Perform a review of local design and construction codes. Review international codes and prepare recommendations for local codes and standards. In particular, the design wind speed and flood levels might need modifications based on recent hurricanes and the detrimental effects of climate change. • Local infrastructure data is sparse. In the appendices in this report, examples of data collection templates are provided. It is suggested that utility companies and other stakeholders conduct an inventory search of all critical assets and maintain such data in a searchable database. As part of this effort, screening of components can also be conducted to identify the most vulnerable assets for a prioritized retrofitting. • Develop a retrofit guide and incorporate typical details (plans and specifications) for the most common retrofits. • Implement a feasibility study for a selected sector and subsector. Preform probabilistic LCCA to identify the most effective retrofits. For the probabilistic LCCA, incorporate realistic initial capital costs and recurring costs. • Region-wide organizations such as CDEMA can develop a post-event damage assessment program, prepare paper and electronic forms, train engineers to perform assessments, and maintain an updated database of trained professionals that can perform these tasks in the event of natural hazards. • Program and undertake retrofitting for one of the countries in a selected sector. This program can serve as a pilot program for use in other countries with similar regional characteristics. 9.4 Future activities As discussed in the report, the objective of this project was to perform an initial review of buildings, water, transport, and power sectors in the selected countries. The scope was further limited to structural design only, to existing construction, and to limited infrastructure in each subsector. As part of a follow-up project, it is recommended that the World Bank Group consider the following components: • Perform a more in-depth analysis. Such activity would require more extensive data collection regarding infrastructure exposure and cost components. • Using the collected data from local stakeholders as discussed above, develop exposure models and GIS maps for both hazard and exposure. Conduct multi-hazard risk analyses for infrastructure networks and obtain individual and aggregate losses (structures, repair cost, revenue loss, recovery time). • Expand the subsectors for various infrastructure sectors. For example, ports and airports could be considered in future studies. • Broaden the scope beyond structural design to include more detailed studies of both construction quality and maintenance programs, and costs associated with them. • Extend the cost computations to include secondary and user cost, and recovery time. • Because hurricanes and coastal floods are a major natural hazard for most countries, undertake an evaluation of flood protection (levees, embankments, etc.) and flood prevention (drainage, collection, etc.) structures in place for coastal areas. Consider improvements and resilience measures for these structures. Such measures provide an additional layer of protection against natural hazards for the infrastructure sectors considered in this report. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 267 REFERENCES AASHTO. (2012). Performance Monitoring System. American Association of State Highway and Transportation Officials (AASHTO): Washington, DC. AASHTO. (2015). Mechanistic-Empirical Pavement Design Guide A Manual of Practice. American Association of State Highway and Transportation Officials (AASHTO): Washington, DC AASHTO. (2017). AASHTO LRFD Bridge Design Specifications, 8th edition. American Association of State Highway Officials (AASHTO), Washington, DC. AASHTO. (2019). A Briefing on Life-Cycle Cost Analysis of New Bridge Design Alternatives, 8th edition. American Association of State Highway Officials (AASHTO): Washington, DC. ACP-EU. (2015). Rapid Damage and Impact Assessment Tropical Storm Erika –August 27, 2015. African, Caribbean and Pacific Group of States - European Union. ACP-EU. (2016). Rapid Damage and Loss Assessment November 9-11 and November 28, 2016. African, Caribbean and Pacific Group of States - European Union. ACP-EU. (2017). Post-Disaster Needs Assessment Hurricane Maria September 18, 2017. African, Caribbean and Pacific Group of States - European Union. ALA. (2005). Seismic Guidelines for Water Pipelines. American Lifelines Alliance (ALA). The Federal Emergency Management Agency and the National Institute of Building Sciences Alessandri, S. Giannini, M. Paolacci F., & Malena, M. (2015). Seismic retrofitting of an HV circuit breaker using base isolation with wire ropes. Part 1: Preliminary tests and analyses. Engineering Structures 98(1 September 2015), 251-262. Alvair. (2015). Commercial Projects: Cumberland Bridges and River Embankment Protection Works. Retrieved from https://alvair.net/project/cumberland-bridges-and-river-embankment-protection- works/ APESL. (2010). HURRICANE TOMAS DAMAGE ASSESSMENT REPORT Version 1.0 – Issued December 07, 2010. Association of Professional Engineers of St. Lucia (APESL Inc.). APUA. (2020). Antigua Public Utilities Authority. Retrieved from http://www.apua.ag/ Argyroudis , S., Mitoulis, S., Winter M., & Kaynia, A. (2018). Fragility of Critical Transportation Infrastructure Systems Subjected to Geo-Hazards. 16th European Conference on Earthquake Engineering, Thessaloniki, Greece. Asadi, P., Hamed Ranjkesh, S., Zeinal Hamedani, A. (2020). Life-Cycle Cost Analysis (LCCA) of RC Bridges Subjected to Multiple Hazards. Scientia Iranica. doi: 10.24200/sci.2020.52202.2592 ASCE. (2007). ASCE 113 Substation Structure Design Guide. American Society of Civil Engineers (ASCE): Reston, Virginia. ASCE. (2015). Flood Resistant Design and Construction: ASCE/SEI 24-14. American Society of Civil Engineers (ASCE): Reston, VA., 1801 Alexander Bell Drive, Reston, Virginia 20191-4382, 2015. ASCE. (2017a). Minimum Design Loads and Associated Criteria for Buildings and Other Structures: ASCE/SEI 7-16, American Society of Civil Engineers, 1801 Alexander Bell Drive, Reston, Virginia 20191-4382, 2017. ASCE. (2017b). Seismic Evaluation and Retrofit of Existing Buildings: ASCE/SEI 41-17, American Society of Civil Engineers, 1801 Alexander Bell Drive, Reston, Virginia 20191-4382, 2017. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 268 ATC. (2020). Rapid Evaluation Safety Assessment Form. Applied Technology Council (ATC): Redwood City, CA. Bakir, D. & Karasin, I. B. (2016). Damages According To Liquefaction And Suggestions. IOSR Journal of Engineering, Vol. 06, Issue 10, PP 01-06, October 2016. Batouli, M., Bienvenu, M., & Mostafavi, A. (2017). Putting sustainability theory into roadway design practice: Implementation of LCA and LCCA analysis for pavement type selection in real world decision making. Transportation Research Part D: Transport and Environment, 52, 289–302. doi: 10.1016/j.trd.2017.02.018 BCQS. (2019). CARIBBEAN & LATIN AMERICA CONSTRUCTION, MARKET TREND REPORT 2018- 2019. BCQS International Ltd.: Georgetown, Guyana. Black & Veatch. (2012). Capital costs for transmission and substations: Recommendations for WECC Transmission Expansion Planning. B&V Project No. 176322. Western Electricity Coordinating Council (WECC). BPA. (2011). BPA engineers build a better tower, saving millions. Bonneville Power Administration (BPA). Retrieved from https://www.bpa.gov/news/newsroom/Pages/BPA-engineers-build-a- better-tower-saving-millions.aspx BU. (2018). Assessment of Haiti’s Electric Sector. Institute of Sustainable Energy (ISE) at Boston University (BU). BWA. (2020). The Bridgetown and South Coast Sewage Plants To Be Upgraded. Barbados Water Authority Limited. Retrieved from http://barbadoswaterauthority.com/ Caltrans. (2020) .Various publications. Retrieved from https://dot.ca.gov/ Campbell, D. (2020). Cost information from Saint Vincent and the Grenadines in Excel format. Data collection by Daniel Campbell, Caribbean Engineering & Design Consultants Ltd (CEDCO), October 2020. Carey, B. D. & Clarke, R. P. (2012). Fragility Functions for a Pervasive Unique Form of Construction in a Region with Very High Potential for Social Losses. 15th World Conference on Earthquake Engineering 2012 (15WCEE), Lisbon, Portugal, 24-28 September 2012. Caribbean Hurricane Network. (2020). Retrieved from https://stormcarib.com/ CARICOM. (1985). Caribbean Uniform Building Code. Caribbean Community Secretariat (CARICOM): Georgetown, Guyana. CCRIF SPC. (2020). CCRIF Strategic Plan 2018 – 2021. The Caribbean Catastrophe Risk Insurance Facility (CCRIF). Retrieved from https://www.ccrif.org/ CDB. (2020). Third Water Supply Project (WA-1 Network Upgrade). Caribbean Development Bank (CDB). Retrieved from https://www.caribank.org/work-with-us/procurement/procurement- notices/third-water-supply-project-wa-1-network-upgrade CDMP. (1998). A Case Study of Caribbean Infrastructure Projects that have Failed Due to the Effects of Natural Hazards. Part of the Caribbean Disaster Mitigation Project (CDMP), funded by the USAID Office of Foreign Disaster Assistance and the Organization of American States. Retrieved from https://www.oas.org/cdmp/document/failinfr.htm#Table%203 CDMP. (1999). Caribbean Building Inspector Training Courses. USAID-OAS Caribbean Disaster Mitigation Project. Retrieved from http://www.oas.org/CDMP/document/bldginsp/insp_rev.htm Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 269 CDMP. (2020). Seismic Hazard Maps. Part of the Caribbean Disaster Mitigation Project (CDMP), funded by the USAID Office of Foreign Disaster Assistance and the Organization of American States. Retrieved from https://www.oas.org/cdmp/document/seismap/ CEHI. (2013). Financial Assessment for Wastewater Treatment and Disposal (WWTD) in the Caribbean. Caribbean Environmental Health Institute: Castries, St. Lucia. Chandrasekaran, S., & Banerjee, S. (2014). Retrofit Optimization for Resilience Enhancement of Bridges under Multihazard Scenario. Proceedings of the 10th National Conference on Earthquake Engineering, Earthquake Engineering Research Institute, Anchorage, AK. CHARIM. (2020). Caribbean Handbook on Risk Information Management. Program initiated by the World Bank Group through grant from the ACP-EU Natural Disaster Risk Reduction Program. http://www.charim.net/ Chin, M.W. (2008a). SEISMIC CODE EVALUATION 1985 Caribbean Uniform Building Code (CUBiC) Part 2 Section 3. Caribbean Community (CARICOM). Chin, M.W. (2008b). SEISMIC CODE EVALUATION Trinidad and Tobago Small Buildings Code, Board of Engineering of Trinidad and Tobago 2001 updated second draft version (July). Trinidad and Tobago. Chin, M.W. (2012). THE GLOBAL EARTHQUAKE MODEL (GEM) CARIBBEAN REGIONAL PROGRAMME – AN INTRODUCTION. APETT CIVIL ENGINEERING DIVISION SEMINAR, Port Authority Conference Room, Port of Spain, Trinidad. CIMH. (2020). Caribbean Institute for Meteorology and Hydrology. Clarke, R. P. (2017). Seismic Fragility Functions for Light Unreinforced Masonry Single-Story Residential Structures in Trinidad and Tobago. West Indian Journal of Engineering, January 2017, Vol. 39 Issue 2, p40-45. 6p. Clarke, R. P. (2018). Hurricane Maria: A Harbinger of Imperative Changes to Caribbean Structural Design and Construction, Presentation to Association of Professional Engineers re Caribbean Construction (includes seismic assessment and recommendations). Consolidated water. (2020). Blue Hills. Nassau, Bahamas https://cwco.com/portfolio/blue-hills/ Cumming. (2020). CUMMING INSIGHT Construction Market Analysis Q3-2020. Cumming Construction Management, Inc.: Los Angeles, California, U.S. CWSA. (2020). CWSCA – Home. Central Water & Sewerage Authority (CWSA). Retrieved from http://www.cwsasvg.com/ CWWA. (2017) . Design specifications and selection issues in the application of Trenchless Technology for water and sewer mains. Caribbean Water and Wastewater Association: St Joseph, Trinidad. Da Costa, N. (2018). Jamaica National Building Act. 2018. Presentation by Noel Da Costa at Jamaica Institution of Engineers, June 13th, 2018. Dannion Engineering. (2020). Marc Floissac and Caico Bridge Reconstruction. Retrieved from https://dannionce.com/Project/marc-floissac-and-caico-bridge-reconstruction/ Davis, S., Clopper, P., Paroloa, A., Males, R., & Rogers, C. (2013). Development of a Model of Flood Damage to Roads. World Environmental and Water Resources Congress 2013. ASCE: Reston, VA. De Leon D., Lopez. A. and Cordero, C. (2009). Life- cycle cost functions for electrical substations and power transmission towers under strong winds in Mexico, Technical Council on Lifeline Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 270 Earthquake Engineering (TCLEE) of the American Society of Civil Engineers (ASCE). American Society of Civil Engineers: Reston, Virginia. Dominica Ministry of Planning, Economic Development and Investment. (2018). Guide to Dominica’s Housing Standards. Dominica Ministry of Planning, Economic Development and Investment. Dominica News. (2017). Bridge Collapses at Macoucherie. Retrieved from https://dominicanewsonline.com/news/homepage/news/general/bridge-collapses-at- macoucherie/ Dominica News. (2017). Dominica – Updates on Food and Water Distribution and Water Supply. DaMajority. Retrieved from https://www.damajority.com/dominica-updates-food-water- distribution-water-supply/ Dominica Physical Planning and Development Authority. (1996). BUILDING CODE (DRAFT). UnCHS/UNDP Project for Programme Support to the Human Settlements Sector in the OECS DMI/92/001, Commonwealth of Dominica, 1996. EBS. (2020). Retrieved from https://www.nvebs.com/ EERI. (2010). The El Mayor Cucapah, Baja California Earthquake April 4, 2010. Earthquake Engineering Research Institute: Oakland, CA. Eidinger and Davis. (2012). Recent Earthquakes: Implications for U.S. Water Utilities. Water Research Foundation: Denver, CO. Eidinger. (2003). Economics of Seismic Retrofit of Water Transmission and Distribution Systems. Sixth U.S. Conference and Workshop on Lifeline Earthquake Engineering (TCLEE). American Society of Civil Engineers: Reston, VA. EPRI. (2014). Considerations for a power transformer emergency spare strategy for the electric utility industry. The Electric Power Research Institute (EPRI): Palo Alto, CA. ESRI. (2019). Disaster Response Program Hurricane Public Information Map. Retrieved from https://www.esri.com/en-us/home FDOT. (2020). Bridge Costs. Florida Department of Transportation. Retrieved from https://www.fdot.gov/ FEMA. (1994). FEMA 156/December 1994: Typical costs for seismic rehabilitation of existing buildings, volume 1: Summary, Federal Emergency Management Agency: Washington DC, US. FEMA. (1995). FEMA 157/September 1995: Typical costs for seismic rehabilitation of existing buildings, volume 2: Supporting documentation, Federal Emergency Management Agency: Washington DC, US. FEMA. (2003a). Multi-hazard Loss Estimation Methodology, Earthquake Model: HAZUS MR4 Technical Manual. Federal Emergency Management Agency: Washington, D.C. FEMA. (2006a). FEMA 454/December 2006: Risk Management, Series, Designing for Earthquakes, A Manual for Architects, Federal Emergency Management Agency, Washington DC, US. FEMA. (2006b) .FEMA 547/ 2006 Edition: Techniques for the Seismic Rehabilitation of Existing Buildings, Federal Emergency Management Agency: Washington DC, US. FEMA. (2007a) . FEMA 543/January 2007: Risk Management, Series, Design Guide for Improving Critical Facility Safety from Flooding and High Winds, Federal Emergency Management Agency: Washington DC, US. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 271 FEMA. (2007b) . FEMA 577/June 2007: Risk Management, Series, Design Guide for Improving Hospital Safety in Earthquakes, Floods, and High Winds, Federal Emergency Management Agency: Washington DC, US. FEMA. (2011). FEMA E-74/January 2011 Reducing the Risks of Nonstructural Earthquake Damage – A Practical Guide. Federal Emergency Management Agency: Washington DC, US. FEMA. (2012). FEMA E-74/December 2012: Reducing the Risks of Nonstructural Earthquake Damage – A Practical Guide, Floods, and High Winds, Federal Emergency Management Agency: Washington DC, US. FEMA. (2012). FEMA E-74/December 2012: Reducing the Risks of Nonstructural Earthquake Damage – A Practical Guide, Floods, and High Winds. Federal Emergency Management Agency: Washington DC, US. FEMA. (2013a). Hazus-MH2.1, Multihazard Loss Estimation Methodology, Earthquake Model Technical Manual, Federal Emergency Management Agency: Washington DC, US. FEMA. (2013b) . Hazus-MH2.1, Multihazard Loss Estimation Methodology, Hurricane Model Technical Manual, Federal Emergency Management Agency: Washington DC, US. FEMA. (2013c). Hazus-MH2.1, Multihazard Loss Estimation Methodology, Flood Model Technical Manual, Federal Emergency Management Agency: Washington DC, US. FEMA. (2013d). FEMA P-936/Julye 2013: Floodproofing Non-Residential Buildings, Federal Emergency Management Agency: Washington DC, US. FHWA. (1998). Life-Cycle Cost Analysis in Pavement Design. Federal Highway Administration (FHWA), US Department of Transportation: Washington, DC. FHWA. (2011). LRFD Seismic Analysis and Design of Transportation Geotechnical Features and Structural Foundations Reference Manual, Office of Infrastructure Research and Development, Federal Highway Administration (FHWA). U.S. Department of Transportation: Washington D.C. FHWA. (2017). Advanced Methodology to Assess Riprap Rock Stability at Bridge Piers and Abutments, Office of Infrastructure Research and Development, Federal Highway Administration (FHWA). U.S. Department of Transportation: Washington D.C. FHWA. (2019). Public-Private Partnership (P3) Procurement: A Guide for Public Owners. Office of Infrastructure Research and Development, Federal Highway Administration (FHWA). U.S. Department of Transportation: Washington D.C. Floodlist. (2015). Dominica Floods – 35 Still Missing, PM Says Damage to Cost $225 Million. Retrieved from http://floodlist.com/america/dominica-floods-35-missing-damage-225-million Floodlist. (2016). Jamaica – Floods Wreak Havoc on Roads in Northern Parishes. Retrieved from http://floodlist.com/america/jamaica-floods-wreak-havoc-roads-northern-parishes GEF Crew. (2017). Applying Economic Resource Valuation for Improving Wastewater Management – The Case of Trinidad and Tobago. Caribbean Regional Fund for Wastewater Management Project (GEF CReW): Kingston, Jamaica. GGB. (2020). Resisting the Twisting, Golden Gate Bridge Highway and Transportation District. Golden Gate Bridge (GGB). Retrieved from https://www.goldengate.org/exhibits/resisting-the-twisting/ Gibbs, T. (1998). Design Manual for Health Services Facilities in the Caribbean with particular reference to Natural Hazards and Other Low-frequency Events. Pan American Health Organization. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 272 Gibbs, T. (2001). Hurricanes and their Effects on Buildings and Structures in the Caribbean. USAID/OAS PGDM building inspector training workshop in Antigua, January 2001. Gibbs, T. (2012). Overview of building codes and CUBiC - National adoptions in the Eastern Caribbean National adoptions in the Eastern Caribbean - Discussion of wind loading in St Lucia. World Bank ECCR Workshop in St Vincent, 26 January 2012. Gilani A., Whittaker, A., Fenves, G, Chen, C., Fujisaki, E., and Ho., H. (2000). Seismic Evaluation and Analysis of 230-kV Disconnect Switches, Report No. PEER 2000-06, Pacific Earthquake Engineering Research (PEER) Center, College of Engineering, University of California, Berkeley, Berkeley, CA. Goettel, K. A. (2016). BENEFIT-COST ANALYSIS OF THE PROPOSED SEISMIC RETROFIT ORDINANCE CITY OF PORTLAND, Goettel & Associates Inc., 1515 Shasta Drive #1207, Davis, California 95616, November 23, 2016. Gosine, T. (2020). Water intake information of various Caribbean countries in the shapefile format. Data collection by Trisandhi Gosine, October 2020. Gosine, T. (2020a). Critical building information of various Caribbean countries in the shapefile format. Data collection by Trisandhi Gosine, October 2020. GPL. (2020). GPL, Inc. Retrieved From https://gplinc.com/ Grupo Estabilidad Estructural. (2000). Manual for Wind Design, translated from “Manual de Diseño contra Viento.” Instituto Tecnológico de Santo Domingo. Note: Wind Code Evaluation conducted by Jorge Gutierrez. https://www.eird.org/cd/acs/English/CodeEval/SpaSpeak/Wind/DRwce.pdf GWI. (2020). GWI – Home. Guyana Water Inc (GWI). Retrieved from https://gwiguyana.gy/ HA. (2020). Hong Kong Housing Authority (HA). Retrieved from https://www.housingauthority.gov.hk/en/ Haghani, R., Al-Emrani, M., & Heshmati, M. (2012). Fatigue-Prone Details in Steel Bridges. Buildings 2012 2, p. 456-476. Doi:10.3390/buildings2040456 Hancilar, U., Taucer, F., & Corbane, C. (2013). Empirical Fragility Functions based on Remote Sensing and Field Data after the January 12, 2010 Haiti Earthquake. Earthquake Spectra (2013), 29(4): 1275, October 2013. Hatami, A., & Morcous, G. (2013). Life-Cycle Assessment of Nebraska Bridges, Project Number: SPR- P1(12) M312, University of Nebraska-Lincoln: Omaha, NB. Hazen. (2020). Fast-track Water Treatment Design and Construction in the Dominican Republic. Hazen and Sawyer. Retrieved from https://www.hazenandsawyer.com/work/projects/fast-track- dominican-republic/ HDRFFG. (2016). Development and Implementation of the Haiti and Dominican Republic Flash Flood Guidance (HDRFFG) Project. Retrieved from https://www.wmo.int/pages/prog/hwrp/flood/ffgs/hdrffg/hdrffg.php Henderson, C. A. (2001). BUILDING PROCEDURES IN BELIZE, Design and Construction of Small Buildings With a focus on the effects of Hurricane Keith 2000, USAID/OAS Post Georges Disaster Mitigation Project BUILDING INSPECTORS WORKSHOP, Cortsland Hotel, St. Johns, Antigua, January 15-26, 2001. Hilderhoff, S.R. (2013). Pumping Station Improvements in Floodplains and Floodways. GHD Consulting Services Inc.: Raleigh, NC. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 273 Ho, T., Donikian, R., Ingham, T., Seim C., & Pan, A. (2006). Seismic Retrofitting Guidelines for Complex Steel Truss Highway Bridges, Report No. MCEER-06-SP05, Multidisciplinary Center for Earthquake Engineering Research, University at Buffalo: Buffalo, N.Y. Huizinga, J., Moel, H.D. and Szewczyk, W. (2017). Global flood depth-damage functions. Methodology and the database with guidelines. Joint Research Centre, European Commission: Sevilla, Spain. ICC. (2017). 2018 International Building Code. INTERNATIONAL CODE COUNCIL, INC. (ICC). IEEE. (2018). IEEE 693-2018: IEEE Recommended Practice for Seismic Design of Substations. The Institute of Electrical and Electronics Engineers, Inc. (IEEE): New York, NY, USA INDIA TV. (2020). Video: Bailey bridge near India-China border in Uttarakhand collapses. https://www.indiatvnews.com/news/india/video-bailey-bridge-collapse-india-china-border- uttarakhand-628661 ISDR. (2009). Global Assessment Report 2009 on Disaster Risk Reduction: Risk and poverty in a changing climate. United Nations International Strategy for Disaster Reduction Secretariat. United Nations: Geneva, Switzerland. Retrieved from https://www.preventionweb.net/english/hyogo/gar/report/index.php?id=9413&pid:34&pil:1 ISO. (2006). ISO 16134:2006, Earthquake- and subsidence-resistant design of ductile iron pipelines. International Organization for Standardization (ISO). Jamaica Gleaner. (2018). Asphalt vs concrete roads - Cost, care and convenience. Retrieved from http://jamaica-gleaner.com/article/news/20180513/engineers-angle-asphalt-vs-concrete-roads- cost-care-and-convenience Jamaica Observer. (2017). Battered! Heavy rains destroy roads, bridges; flood communities. Retrieved from http://www.jamaicaobserver.com/front-page/battered-heavy-rains-destroy-roads-bridges- flood-communities_99114?profile=1373 Jordan, C.J., Grebby, S., Dijkstra, T., Dashwood, C. & Cigna, F. (2015). Risk information services for disaster risk management (DRM) in the Caribbean. Operational documentation. European Space Agency (ESA) “eoworld 2” initiative. British Geological Survey, Natural Environment Research Council. JPS. (2020). Jamaica Public Service Company Limited (JPS). Retrieved from https://www.jpsco.com/ Jyoti Communication. (2010). Missing report on collapsed Bailey Bridge shows negligence. Retrieved from http://jyoticommunication.blogspot.com/2010/08/missing-report-on-collapsed-bailey.html Kempner, L. & Riley, M. (2016). High Voltage Transformer Base Isolation. Canadian Association for Earthquake Engineering. Khan, M.U., Mesbah, M., Ferreira, L., & Williams, D.J. (2017). Estimating Pavement’s Flood Resilience. Journal of Transportation Engineering, Part B: Pavements 143(3). American Society of Civil Engineers (ASCE): Reston, VA. Khan-Kernahan, I. (2013). Assessment and Reconstruction of Bridges in Trinidad and Tobago. The West Indian Journal of Engineering 36(1), 59-65 (July/August 2013). Knight, B. & Kempner, L. (2004). A Utility's Approach to Seismically Hardening Existing. High-Voltage Electrical Transmission Substation Equipment. 13th World Conference on Earthquake Engineering, Vancouver, B.C., Canada. Kok, M., Huizinga, H.J., Vrouwenvelder, A. & Barendregt, A. (2004). Standard method 2004 damage and casualties caused by flooding. Ministry of Transport, Public Works and Water Management, Rijkswaterstaat, Road and Hydraulic Engineering Department. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 274 Kraft. (2013). Development of liquefaction susceptibility and hazard maps dot the islands of Jamaica and Trinidad, Georgia Institute of Technology, Atlanta, GA. Kumalasari, C., Tickle, V. and Madugula, M. (2005). Compressive strength of solid round steel members strengthened with rods or angles, Civil Engineering Dimension, Vol. 7, No. 2, Lee, S. Kim, S. and MHA, H. (2004). Seismic Retrofit Prioritization of Bridges Based on damage and failure cost analysis. 13th World Conference on Earthquake Engineering, Vancouver, B.C., Canada Loo, Y.L., Yaw, C.T., Hashim, A.H., Mohamad, A.M., Faizah, M.K., & Noormala, A. (2010). Development of Switchgear Life Cycle Cost Analysis Software. The 4th International Power Engineering and Optimization Conf. (PEOCO2010), Shah Alam, Selangor, Malaysia: 23-24 June 2010. LUCELEC. (2020). Substations. St. Lucia Electricity Services Limited (LUCELEC). Retrieved from https://www.lucelec.com/content/substations MISO. (2020). PSC: Transmission Cost Estimation Guide for MTEP20 (20200211). Midcontinent Independent System Operator (MISO). Miyamoto, K. and Gilani, A. (2012). URBAN EARTHQUAKE DAMAGE ASSESSMENT AND RECONSTRUCTION: 2010 HAITI EARTHQUAKE. 15th World Conference on Earthquake Engineering, Lisbon, Portugal. MWH. (2012). Grants Pass Water Treatment Plant, Structural/Seismic Evaluation Report. MWH Global Inc. Retrieved from http://www.mwhglobal.com/ Naaseh, S. & Walters, M. (2006). Implementation of Base Isolation Implementation of Base Isolation in Retrofit Applications, Presentation to 100th Anniversary Earthquake Conference Commemorating the 1906 San Francisco Earthquake, April 17th, 2006. Nandi, A., Mandal, A., Wilson, M., & Smith, D. (2016). Flood hazard mapping in Jamaica using principal component analysis and logistic regression. Environmental Earth Sciences 75(465). https://doi.org/10.1007/s12665-016-5323-0 National Energy Grid Map Index (NEGMP). (2020). National Energy Grid Map Index. Global Energy Network Institute (GENI). Retrieved from http://www.geni.org/globalenergy/library/national_energy_grid/index.shtml National Hurricane Center and Central Pacific Hurricane Center. (2020, November). Saffir-Simpson Hurricane Wind Scale. National Oceanic and Atmospheric Administration. Retrieved from https://www.nhc.noaa.gov/aboutsshws.php NIBS. (2019). Natural Hazard Mitigation Saves: 2019 Report. National Institute of Building Sciences (NIBS), National Institute of Standards and Technology (NIST): Washington, DC, USA. NISEE. (2020). The Earthquake Engineering Online Archive NISEE e-Library. Pacific Earthquake Engineering Research (PEER) Center: Berkeley, CA. https://nisee.berkeley.edu/elibrary/ NIST. (2015). NEHRP Seismic Design Technical Brief No. 11, Seismic Design of Steel Buckling Restrained Braced Frames, A Guide for Practicing Engineers, National Institute of Standards and Technology within the U.S. Department of Commerce: Gaithersburg, MD. NIST. (2017). NIST Technical Note 1973: A Methodology for Estimating Seismic Retroft Costs, National Institute of Standards and Technology within the U.S. Department of Commerce: Gaithersburg, MD. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 275 NOAA. (2020). National Hurricane Center, National Oceanic and Atmospheric Administration (NOAA). https://www.nhc.noaa.gov/ Numbeo. (2020). Cost of Living Index by Country 2020 Mid-Year. Retrieved from https://www.numbeo.com/cost-of-living/rankings_by_country.jsp NWA. (2007). Roads, Infrastructure, Drainage and Traffic Management. National Work Agency: Kingston, Jamaica. ODPM. (2020). Map Downloads. The Office of Disaster Preparedness and Management (ODPM), Ministry of National Security, Government of the Republic of Trinidad and Tobago. http://www.odpm.gov.tt/node/246 OECS. (2004). Grenada. Macro-Socio Economic Assessment of the Damage caused by Hurricane Ivan. Organization of Eastern Caribbean States. OSSPAC. (2013). The Oregon Resilience Plan Reducing Risk and Improving Recovery for the Next Cascadia Earthquake and Tsunami. Oregon Seismic Safety Policy Advisory Commission (OSSPAC) PAHO. (2019). Development of Design Wind Speed Maps for the Caribbean for Application with the Wind Load Provisions of ASCE 7-16 and Later. Pan American Health Organization (PAHO): Washington DC. Panteli, M., Pickering, C., Wilkinson, S., Dawson, R., & Mancarella, P. (2017). Power System Resilience to Extreme Weather: Fragility Modeling, Probabilistic Impact Assessment, and Adaptation Measures. IEEE TRANSACTIONS ON POWER SYSTEMS, 32(5). PEguru. (2020). Substation cost estimator. Retrieved from https://peguru.com/substation-cost-estimator/ Porter, K.A., Beck, J.L., Shaikhutdinov, R.V., Au, S.K., Mizukoshi, K., Miyamura, M., Ishida, H., Moroi, T., Tsukada, Y., & Masudai, M. (2004). Effect of Seismic Risk on Lifetime Property Value. Earthquake Spectra, Volume 20(4), 1211–1237. Oakland, CA: Earthquake Engineering Research Institute (EERI). Porter, R. (2008). Seismic Retrofit of Existing Reservoirs. 2008 PNWS – AWWA Conference (American Water Works Association, Pacific Northwest Section), Vancouver, Washington, 2008. PRHEM. (2012). Recommended Standards for Water Works 2012 Edition. Great Lakes – Upper Mississippi River Board of State and Provincial Public Health and Environmental Managers (PRHEM). Pune Municipal Corporation (PMC). (2020). Surface Water Treatment Process. Retrieved from https://www.pmc.gov.in/en/surface-water-treatment-process Raetz, H., Forscher, T., Kneebone, E. & Reid, C. (2020). A TERNER CENTER REPORT - MARCH 2020, The Hard Costs of Construction: Recent Trends in Labor and Materials Costs for Apartment Buildings in California. Terner Center for Housing Innovation, UC Berkeley. Redmond, L. M. (2015). SEISMIC PERFORMANCE, ANALYSIS, AND DESIGN OF HYBRID CONCRETE-MASONRY. In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the School of Civil and Environmental Engineering, Georgia Institute of Technology. Riley, M., Stark, C., & Kempner, L. (2006). Seismic Retrofit Using Spring Damper Devices on High- Voltage Equipment Stands. Earthquake Spectra (2006) 22(3), 733-753. https://doi.org/10.1193/1.2216736 Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 276 Rojas-Mercedes, N. J., Di Sarno, L., Simonelli, A. L. & Penna, A. (2020). Seismic risk of critical facilities in the Dominican Republic: case study of school buildings. Soft Computing 24, 13579– 13595. https://link.springer.com/article/10.1007/s00500-019-04361-0 Romero, N., Nozicka, L., Dobsonb, I., Xua, N., & Jones, D. (2013). Seismic Retrofit for Electric Power Systems. Earthquake Spectra 31(2). DOI: 10.1193/052112EQS193M SafeRack. (2020, November). Wastewater Treatment Plant. Retrieved from https://www.saferack.com/wastewater-treatment-plant/ Schweikert, A. E., L’her, G.L., Nield, L.G., Kerber, S. W., Flanagan, R. R. & Deinert, M. R. (2020). Resilience in the Caribbean: Natural Hazards Exposure Assessment and Areas for Future Work. Accessed 12 June 2020. Searchlight. (2015, June 26). Hope Bridge officially opens. Searchlight.vc. Retrieved from https://searchlight.vc/searchlight/news/2015/06/26/hope-bridge-officially-opens/ Searchlight. (2018). Cumberlan Bridge Opens Tuesday. Searchlight.vc. Retrieved from https://searchlight.vc/searchlight/press-release/2018/03/19/cumberland-bridge-opens-tuesday/ Smith R., SCHAFER, P., KIRSHBAUM D., and REGINA, E. (2008). Orographic Enhancement of Precipitation inside Hurricane Dean. American Meteorological Society SMN-N. (2012). Seven Seas Water Continues Normal Water Production. SMN-News. Retrieved from https://smn-news.com/st-maarten-st-martin-news/27215-seven-seas-water-continues-normal- water-production.html Song J. & Der Kiureghian A. (2004). Reliability of electrical substation equipment connected by rigid bus. 13th World Conference on Earthquake Engineering, Vancouver, B.C., Canada Suite, W. (2008). WIND CODE EVALUATION 1985 Caribbean Uniform Building Code (CUBiC) Part 2 Section 2. Caribbean Community (CARICOM). Takhirov S. and Mosalam, K. (2012). Application of Laser Scanning to Structural Testing in Earthquake Engineering, Field Survey, and Structural Assessment of an Earthquake Aftermath. Tenth U.S. National Conference on Earthquake Engineering, Anchorage, Alaska. ThinkHazard. (2020). ThinkHazard database. Global Facility for Disaster Reduction and Recovery (GFDRR). Retrieved from https://thinkhazard.org/en/ Thomson, D. (2018). State of Building Codes and Standards in Trinidad and Tobago. ASTM Caribbean Roadshow, Wednesday 6th Jun 2008. Tota-Maharaj, K & Meeroff, D.E. (2013). Evaluation of Solar Photosensitised River Water Treatment in the Caribbean. International Journal of Photoenergy, Vol. 2013. Hindawi Publishing Corporation. https://cyberleninka.org/article/n/232231 Trovato M., Sun, L., & Stojadinovic, B. (2015). Buckling Restrained Brace Retrofit Technique for Existing Electric Power Transmission Towers, 8th International Conference on Behavior of Steel Structures in Seismic Areas. TxDOT. (2013). Improved Cross Frame Details for Steel Bridges. Authored by Battistini, A. et al., and funded by the Texas Department of Transportation (TxDOT): Austin, TX. UNEP. (1997). Source Book of Alternative Technologies for Freshwater Augmentation in Latin America and the Caribbean. International Environmental Technology Centre, United Nations Environment Programme (UNEP). Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 277 USDOT. (2017) . Bridge Retrofit or Replacement Decisions: Tools to Assess Sustainability and Aid Decision-Making. U.S Department of Transportation/Research and Innovative Technology Administration: Washington, DC. USGS. (2005). NOAA-USGS Debris-Flow Warning System—Final Report. U.S. Geological Survey: Reston, Virginia. USGS. (2007). Analysis of the Interstate 10 Twin Bridge’s Collapse During Hurricane Katrina. United Stated Geological Survey: Reston, Virginia. UWI. (2020). Maps. Seismic Research Center, The University of West Indies. Retrieved from http://uwiseismic.com/Maps.aspx Van der Sande, C. (2001). River flood damage assessment using IKONOS imagery. Joint Research Centre, European Commission: Ispara, Italy. Vickery, P. J. and Wadhera, D. (2008). Wind Speed Maps for the Caribbean for Application with the Wind Load Provisions of ASCE 7: ARA Report 18108-1. Pan American Health Organization (PAHO). Retrieved from https://www.paho.org/disasters/index.php?option=com_docman&view=download&category_slug =safe-hospitals&alias=554-wind-hazard-maps-for-various-return-periods&Itemid=1179&lang=en WASA. (2008, October). Water and Wastewater Design Guideline Manual. Water and Sewerage Authority (WASA) of Trinidad and Tobago. WASA. (2020). WASA – Home. Water and Sewerage Authority (WASA). Retrieved from https://www.wasa.gov.tt/ WASCO. (2017, October 18). Micoud and Desrruisseaux Water Supply Redevelopment Project. Water & Sewerage Company, Inc (WASCO). Retrieved from https://wascosaintlucia.com/news/micoud- and-desruisseaux-water-supply-redevelopment-project Waterworld. (2009). ACCIONA to build drinking water plant, multiple aqueducts in Dominican Republic. Author. Retrieved from https://www.waterworld.com/international/article/16218024/acciona-to- build-drinking-water-plant-multiple-aqueducts-in-dominican-republic WBG. (2015). The Cost of Road Infrastructure in Low and Middle Income Countries. Policy Research Working Paper 7408. The World Bank Group. WHO. (1987). Government of Suriname: Public Water Supplies and Sewerage, Volume IV Water Supplies and Sewerage. United Nations Development Programme and World Health Organization (WHO). Retrieved from http://www.swris.sr/wp-content/uploads/2017/12/Public- water-supplies-Vol-IV.pdf WHO. (2008). Regional and Global Costs of Attaining the Water Supply and Sanitation Target (Target 10) of the Millennium Development Goals. World Health Organization (WHO): Geneva, Switzerland. WWE. (2015). CSA 7 - Water System Feasibility Study and Seismic Retrofit Project Water System Improvement Recommendation Report. Waterworks Engineers (WWE). Xie Q. & Zhange J. (2020). Experimental study on failure modes and retrofitting method of latticed transmission tower. Engineering Structures, Volume 226 (2021). https://doi.org/10.1016/j.engstruct.2020.111365 Yamane, T. (2020). Presentation: Construction Examples in Japan High Capacity Micropile. Kyokuto Corporation: Tokyo, Japan. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 278 REVIEW OF DATA Date Collection template As part of this project, a data collection template was developed. This template (see Table A.1) can be used to develop the exposure model for various infrastructure. Collected data As part of the activities, the local members of the project team used the abovementioned templates and collected site-specific data. Examples for the building sector are presented in Table A.2. As discussed in the main body of this report, a number of factors contributed to the difficulty of collecting country-specific data. World Bank-supplied data A Geojson file was supplied by the WBG. An assessment of the data is presented in Table A.3. Note that the WBG data provides invaluable information related to the geographical location and distribution of several assets in a number of the countries. However, data listed in Table A.1 needs to be collected in order to augment the geographical data to perform quantitative risk assessments for infrastructure sectors. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 279 Power Sector: Powerplants Collected Footprint, People Year Plans Material and Construction Annual No Country Sector Sub-sector Date Lat., deg Long., deg Capacity, kV Design code by m2 served built available framing cost maintenance cost 1 Power Power plant 2 Power Power plant .. Power Power plant Power Sector: Substations Collected Footprint, People Year Plans Resilient Construction Annual No Country Sector Sub-sector Date Lat., deg Long., deg Capacity, kV Design code by m2 served built available equipment cost maintenance cost 1 Power Substation 2 Power Substation .. Power Substation Power Sector: Distribution Lines Collected Beg Lat., Beg Long., End Lat., Capacity, People Plans No Country Sector Sub-sector Date End Long., deg Length, km Year built Design code Pole type Cost/km by deg deg deg kV served available 1 Power T&D 2 Power T&D .. Power T&D Water Sector: Water Treatment Plants Collected Footprint, People Year Plans Primary material Construction Annual No Country Sector Sub-sector Date Lat., deg Long., deg Capacity Design code by m2 served built available and framing cost maintenance cost 1 Water WTP 2 Water WTP .. Water WTP Water Sector: Wastewater Treatment Plants Annual Collected Footprint, Primary material and Construction No Country Sector Sub-sector Date Lat., deg Long., deg Capacity people served Year built Design code Plans available maintenance by m2 framing cost cost 1 Water WWTP 2 Water WWTP .. Water WWTP Water Sector: Pipelines (underground) Beg Lat., Beg Long., End Lat., End Long., Capacity, People Plans Pipe type and No Country Sector Sub-sector Date Collected by Length, km Year built Design code Cost/km deg deg deg deg m3/sec served available material 1 Water UGP 2 Power UGP .. Power UGP Transport Sector: Bridges Cost/ Collected Span, Skew, Vert clr, Year Inspection No Country Sector Sub-sector Date Lat., deg Long., deg Length, m Material Superstr. Piers Columns/pier ADT Design code m2 by m deg m built cycle deck 1 Transport Bridge 2 Transport Bridge .. Transport Bridge Transport Sector: Primary Paved Roads Beg Lat., Beg Long., End Lat., End Long., Year Cost/km No Country Sector Sub-sector Date Collected by Length, km Surface (AC concrete) Avg. daily traffic Design code deg deg deg deg built road 1 Transport Primary paved road 2 Transport Primary paved road Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 280 .. Transport Primary paved road Buildings: Critical Facilities Lat., Long., Storie Primary Occupant Year No Country Sector Type Date Collected by Footprint, m2 Framing type Design code Construction/Replacement Cost deg deg s material s built 1 Buildings School 2 Buildings School .. Buildings Hospital Table A.1. Date collection template Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 281 File name For which component Country Category Per Table 2A, Hazards for each country. Probably, Resilience in the Caribbean_ Natural can classify the countries per primary hazards. Per Hazards Exposure Assessment and All 16 countries Hazard Table 2B, the target infrastructure information Areas for Future Work.pdf might be available in terms of "structural type". An overview of TT codes (e.g., small buildings, 13_TT_National_Building_Code_Ove Electricity, Driveway gates). Currently, there is no Trinidad & Tobago Codes rview-Thompson.pdf formal building codes in TT. Antigua.pdf It looks a code exists in Antigua. Antigua Codes Bahamas IMPROVING BUILDING It looks a code exists in Bahamas (1st in 1971, 3rd STANDARDS CDB MEMBER Bahamas Codes in 2003). COUNTRIES 2018 BAH 1.pdf It looks a code exists in Belize (1st in 2003, latest Belize.pdf Belize Codes in 2017). guide_to_dominicas_housing_standar It looks this standard is mainly for Hurricane. This ds-7.pdf is based on 2002 Dominica Building Code. Also, Dominica Codes LAWPhysical Planning Draft 1996 Building Code (draft) in 1996. Building Code_Dominica.pdf GUYANA_Building Code It looks a code exists in Guyana. Guyana Codes Presentation.pdf It looks Jamaican Building Code was consisted of Jamaica Building Act 2018 NdaC pp application documents created in 2003 based on Jamaica Codes Presentation.pdf IBC. bcqs-construction-market-report- Page 10 shows the difference of construction cost Several Cost 2019.pdf for buildings per some Caribbean countries. Building Procedures in Belize.docx Belize's construction cost (house?) is shown. Belize Cost Hurricane damage on Essential buildings, Roads, matthew_bahamas_.pdf Bridges, Water, Power, etc. of Bahamas at 2016 Bahamas Damage Hurricane Matthew. Commonwealth of Dominica - Rapid Tropical Storm damage on Essential buildings, Damage and Needs Assessment Final Roads, Bridges, Water, Power, etc. of Dominica at Dominica Damage Report .pdf 2015 Tropical Storm Erika. Hurricane damage on Public buildings, Roads and H.Tomas-DANA (APESL).pdf St. Lucia Damage Bridges of St. Lucia at 2010 Hurricane Tomas. St Lucia Electricity Services Vulnerability study for Power infrastructure in St. Vulnerabilit St. Lucia Study.docx Lucia. y Disaster Risk Reduction Country Doc Major hazard for Suriname (looks flood). Suriname Hazard Suriname.pdf CHARIM_Belize_NFHM_Methodolo gy_FINAL.pdf COASTAL_FLOOD_HAZARD_BEL Flood Hazard Maps for Belize (CHARIM 2016) Belize Hazard IZE12016.pdf FLOOD_HAZARD_BELIZE2016.pd f Dominica_Flash_Flood_Hazard_map _1.pdf Flood Hazard Map and Landslide Susceptibility Dominica Hazard Dominica_Landslide_susceptibility_ Map for Dominica (CHARIM) Map.pdf SLUFloodReport.pdf SLULandslideReport.pdf St Lucia Landslide Susceptibility Flood Hazard Map and Landslide Susceptibility St. Lucia Hazard Map.pdf Map for St. Lucia (CHARIM) St_Lucia_Flash_Flood_Hazard_Map_ 1.pdf Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 282 flood_risk_multi_trinidad.pdf Flood Risk Map, Flood Susceptibility Map, flood_susceptibility_trinidad.pdf Landslide Susceptibility Map and Landslide Risk Trinidad & Tobago Hazard landslide susceptibility trinidad.pdf Map for Trinidad and Tobago (ODPM) landslide_risk_multi_trinidad.pdf Seismic Hazard Map for Eastern Caribbean PGA(g)_RP=475years.jpg Countries (2011) - the north the Leeward Islands Eastern Caribbean SeismicHazMaps_Summary_2011_06 (from Anguilla to Dominica) and in the south the Hazard Countries .pdf Windward Islands (from Martinique to Grenada), Barbados and Trinidad and Tobago Table A.2. Samples of collected data for building sector Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 283 File name Contents Usefulness - Assuming, at least, the hazard maps (for all hazards in GIS Final Resilience in the Caribbean: Natural Hazards Exposure Assessment and Areas for format) and the exposure model (might not contain much Report_12Jun Future Work (12 June 2020) information) must exist. So, it might be useable for our project e2020_v2.pdf according to our SOW. If not, we might not want to ask, I guess. (1) This data includes only the risk result of the road model in the format of line (vector data, length, location, type, surface). (2) "risk" in the attribute table seems to be a hazard risk score (0 through 5) mentioned in Table 3 of the report. But, this is not 100% clear and it's needs to be confirmed. (3) "type" and "surface" are specified in the attribute table. "type" seems to be a road classification (primary, secondary, tertiary, earth, others) mentioned in report Page 8/60 but not 100% sure and needs to be confirmed. "surface" seems to be a cover of road (cement, asphalt, gravel, dirt) mentioned in report Page 41/60 but not 100% sure and needs to be confirmed. - If needed, have to confirm whether or not any other data (altitude, etc.) of "roads" model exist. Geojson files - If needed, have to ask Hazard maps (for each hazards, like Map in B of Figure 1 in the report), Damage map (for each hazard, like "infrastructur Map C of Figure 1 in the report), Probability map (for each es-roads" hazard, like Map D of Figure 1 in the report), and fragility functions used for each hazard damage probability. Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 284 Geojson files (1) This data includes the several risk scores (landslide, hurricane, earthquake, pluvial If needed, have to confirm whether or not any other data in flood, fluvial flood, storm surge) of other infrastructure models (airport, bridge, (altitude, volume, footprint, structural type, etc.) of all "infrastructur hospital, port, power plant, and wastewater) in the format of point (vector data, location infrastructure asset model exist. es" and type are available). (2) The scores in the attribute table according to hazard type seems to be a hazard risk score (0 through 5) mentioned in Table 3 of the report. But this is not 100% clear and it's needs to be confirmed. (3) "type" seems to be a classification for each asset (size, fuel, etc.) mentioned in report Appendix B (but not more specific data like height, structural type, footprint, etc.) but not 100% sure and needs to be confirmed Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 285 Geojson files (1) This data includes only the multi-hazard risk (all hazard risks were combined) and This seems to be combined with all hazard risks by a summation in "multi- the risk score is between 0 and 30. of each risk score with a weight factor (according to GPS hazards" (2) For example, Figure 3 in the report shows this type of multi-hazard risk map for coordinates), and it seems to be a qualitative risk result. Dominican Republic (the below map is also same one). (3) Other than that, no information is included. Table A.3. Review of WBG-submitted document Overview of Engineering Options for Increasing Infrastructure Resilience in the Caribbean | 286