Investing in Transportation Resilience: A Framework for Informed Choices (2021)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

45 Transportation agencies across the modes have taken different steps to inte- grate resilience analysis into their decision-making processes. Some agencies have developed comprehensive quantitative analysis procedures to estimate both the current level of resilience and the relative benefits and costs of can- didate investments to improve resilience. Some have developed indicators that allow them to track progress in improving the level of transportation system resilience over time. Others have factored resilience benefits into infrastructure design guidance that can be consulted to choose designs that are most cost-effective for improving resilience. The methods used for these assessments often involve a mix of qualitative and quantitative data and reliance on expert judgment to fill data gaps. This chapter begins with an introduction to the analytical procedures that agencies are using in the field to deliberately increase resilience, includ- ing the data used as inputs to the procedures and the intermediate mea- sures and output metrics the procedures generate. To illustrate how these procedures work, the chapter then provides short case studies of the use of quantitative risk assessment models and tools, vulnerability assessments, resilience indicators, and design guides. The chapter concludes with a sum- mary of the various types of metrics used in these practices. INTRODUCTION TO CURRENT PRACTICE An agency’s resilience practices are often shaped by the context in which the practices were originally developed. For many agencies, a catalyzing event raised awareness of the risks of natural hazards. Others developed resilience 3 Current Practice in Measuring and Managing Transportation System Resilience

46 INVESTING IN TRANSPORTATION RESILIENCE practices in response to federal or state legislation.1 Sometimes multiple fac- tors occurring at the same time or in succession had influence. Hurricane Katrina (2005) followed by Superstorm Sandy (2012) spurred New York’s Metropolitan Transportation Authority to treat resilience more explicitly. State legislation passed in 2015 required California’s Port of Long Beach to develop its Climate Adaptation and Coastal Resiliency Plan as part of a statewide Integrated Climate Adaptation and Resiliency Program.2 Hurri- cane Isabel (2003) prompted early attention within the Hampton Roads Transportation Planning Organization (HRTPO), while the organization’s growing interest in advanced modeling techniques was motivated by the 2015 Fixing America’s Surface Transportation (FAST) Act’s requirement to incorporate resilience considerations into state and metropolitan long-range planning. The Colorado Department of Transportation’s (DOT’s) resilience efforts can be traced to several influences, including major flooding in 2013, state legislation on emergency response enacted in 2018, and the federal regulations that limit emergency relief funds to highway projects that re- store infrastructure only to its pre-disaster state unless the owner can show that the project reduces costs in the long run.3 Federal, state, and local transportation officials have put significant effort into developing and encouraging the adoption of resilience analysis over the past decade. The challenge they face, however, is prompting agen- cies to take a comprehensive approach to resilience analysis, rather than the practice of focusing on resilience to specific hazards and for certain types of assets only. There may be good reasons for this practice. Policy makers may appropriately emphasize only the type of natural hazard that is the most likely threat to their infrastructure. They may focus only on the types of assets that seem most vulnerable. Data availability or modeling capability may limit analysis. Finally, there may be diminishing returns to additional complexity. Absent incentives, tools, and data for comprehensive approaches to resilience, it can be rational for agencies to limit resilience analysis to the goals that policy makers assign or the types of strategies and actions that they can feasibly implement on their own. 1 USGCRP (U.S. Global Change Research Program). 2018. “Chapter 12” in Fourth National Climate Assessment (NCA4). https://nca2018.globalchange.gov/chapter/12. 2 Port of Long Beach. 2016. “Climate Adaptation and Coastal Resiliency Plan.” https:// www.slc.ca.gov/wp-content/uploads/2018/10/POLB.pdf; State of California. n.d. “Integrated Climate Adaptation and Resiliency Program.” https://opr.ca.gov/planning/icarp. 3 House Bill 18-1394, Colorado State Legislature, May 24, 2018, https://leg.colorado.gov/ sites/default/files/2018a_1394_signed.pdf; Colorado DOT (Department of Transportation). 2020. “Risk and Resilience Analysis Procedure,” pp. 1–2. https://www.codot.gov/programs/ planning/cdot-rnr-analysis-procedure-8-4-2020-v6.pdf.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 47 TYPES OF METRICS USED IN PRACTICE While a single, direct measure of resilience cannot be readily developed or commonly applied, there are common elements in the methods that agencies use to evaluate their resilience to natural hazards and the likely effectiveness of strategies and actions to improve resilience. These elements include analysis methods and metrics for (1) the likelihood (or probability) of natural hazard events (sometimes called “exposure”); (2) the vulnerabil- ity (sometimes called “sensitivity”) of the infrastructure or transportation system to damage or disruption; (3) the consequences of a particular level of damage or disruption, which are often expressed as a combination of owner costs and user costs; and (4) the criticality, or importance, of the infrastructure or system, which may include usage and other measures that reflect the importance of an asset, node, network, or system in broader economic and social terms. Discussions of procedures for measuring the level of resilience and the net benefits of investments in improving resilience can be confusing because different practitioners use different terms for the same concepts. Some resil- ience assessment approaches refer to the likelihood of a particular kind of natural hazard as the “threat” or “threat probability,” while others use the term “exposure” to mean the same thing. The Federal Highway Admin- istration’s (FHWA’s) Vulnerability Assessment Scoring Tool (VAST) uses the word “sensitivity” to refer to the likely physical damage or disruption to an infrastructure asset due to a hazard event, while Risk Analysis and Management for Critical Asset Protection (RAMCAP) models generally use the term “vulnerability” to refer to the likelihood of damage. In the VAST context, vulnerability is a function of the asset’s or system’s sensitiv- ity to hazards or climate effects, exposure to extreme weather and climate effects, and the system’s adaptive capacity. RAMCAP uses the word “risk” to mean the product of the hazard likelihood, the asset “vulnerability,” and the monetary consequences resulting from the hazard affecting the asset; FHWA refers to “risk” as the product of the hazard likelihood and the consequences. Hence, the VAST and the RAMCAP models both use the term “vulnerability,” but they use it to mean two different things. A shared taxonomy of terms is desperately needed. Hazard Likelihood and Character Analysis methods that describe the character and likelihood of natural haz- ard events are covered in more detail in Chapter 2. Most analytic procedures for assessing resilience start with an assessment of what natural hazards can be expected in a particular geographic area and what the likelihood is of a given hazard event of a given magnitude (e.g., a Category 3 hurricane).

48 INVESTING IN TRANSPORTATION RESILIENCE Vulnerability Vulnerability evaluates the effects of a specific natural hazard of a particular magnitude on an infrastructure asset or transportation service. Thus, mea- sures of vulnerability assess susceptibility of infrastructure assets to damage by particular hazard events. Assessments of vulnerability take into account where the asset is located, its design, and its condition. Vulnerability can be reduced by investments that increase the robustness of the asset and miti- gate the damaging effects of natural hazards, and sometimes by relocating the asset. Vulnerability assessments are used to identify at-risk assets and to prioritize which assets may require additional analysis for risk or mitigation. Consequences Consequences are measures of the direct and (if possible) indirect impacts of the damage or disruption to the transportation asset, node, network, or system. Consequences are often split between the owner and the users. For example, consequences for the owner may include costs to repair damage and restore service, while consequences to the users would include costs of detours, delays, and missed trips. Consequences would also include death and injury of personnel or travelers. Consequences may be dependent on the level of redundancy in the transportation system—to the extent that travelers and freight carriers have feasible alternative routes to circumvent damaged infrastructure, the consequences of infrastructure damage are reduced. Indirect consequences can include effects on communities and businesses due to a reduction in accessibility and mobility, failed deliveries, or disruptions of economic and social activities. Risk Risk is the overall likelihood of loss due to natural hazards during a par- ticular time period (typically 1 year), taking into account the likelihood of a hazard event; the vulnerability of the infrastructure; and the economic and social consequences of the damage to the infrastructure for asset owners, asset users, and communities. Representing expected loss due to natural hazards, risk is a key output measure of the resilience analysis and an indi- cator of the level of resilience that is directly useful after analyzing how it changes in response to investments in reduced vulnerability. Criticality Criticality is a measure of the importance of the function of the transpor- tation asset, node, network, or system. Criticality includes some of the

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 49 same elements as “consequences” (such as costs to users) but also includes broader economic and social impacts, for example, on shippers of freight movements that are disrupted and on tourism industries due to disrup- tions to passenger transportation. Criticality also takes into account equity effects, such as the distribution of disruption impacts across socially and economically vulnerable populations. Costs to infrastructure owners to repair damage or restore service are not normally included in criticality. Like risk or vulnerability assessments, criticality measures are used for prioritization. The selection of the component measures for criticality is typically done on a parallel track to the rest of the resilience analysis. Because criticality measures involve broader economic and social concerns, criticality assessment typically involves stakeholder or public consultation. For its resilience-informed, long-range metropolitan transportation plan- ning process, HRTPO defines criticality as “regional significance” and includes among its many component measures usage and travel time as well as access to major employment centers, tourism destinations, and low- income communities.4 Prioritization based on criticality can be done at different stages of the resilience analysis and resilience-informed decision making. FHWA recom- mends that criticality prioritization occur early in the assessment process to target subsequent vulnerability analysis and resilience interventions to the more important transportation elements.5 There is an emerging practice of defining resilience to be the intersec- tion of vulnerability and criticality or of risk and criticality. One way to present this result is through a matrix, as in Figure 3-1. In this approach, both vulnerability and criticality together guide priority setting. As an example, elements evaluated for resilience investment would progressively increase in priority from those with low criticality/low vulnerability (lower left corner cell, in dark green) to those with high criticality/high vulner- ability (upper right corner cell, in red). 4 Stith, D.M. 2020. “Integrating Resilience into Planning.” Hampton Roads Transpor- tation Planning Organization, October 7. https://www.hrtpo.org/uploads/docs/P9-HRTPO- IntegratingResilience-LRTP-10.07.20.pdf. 5 FHWA (Federal Highway Administration). 2014. “Assessing Criticality in the Transporta- tion Adaptation Planning.” https://www.fhwa.dot.gov/environment/sustainability/resilience/ tools/criticality_guidance.

50 INVESTING IN TRANSPORTATION RESILIENCE FIGURE 3-1 Resilience can be assessed by a matrix of criticality and vulnerability.6 Input Data and Derived Measures The resilience analysis process requires input data, such as hazards, asset conditions, and functionality. These are used in the agency’s analytical process to measure (or estimate) vulnerability, criticality, and consequences to guide the selection of resilience improvements. These analytic measures may also be used to estimate overall risk or resilience. Outputs of resil- ience analysis processes may be quantitative data and measures, qualitative indicators (though often based on quantitative input data), or qualitative descriptions. The case studies that follow show how agencies use a variety of input data, quantitative and qualitative measures, indicators, and descriptors to assess asset or system resilience. 6 Adapted from the Houston-Galveston Area Council. 2021. “Resilience and Durability to Extreme Weather in the H-GAC Region Pilot Program Report.” https://www.h-gac.com/ getmedia/4a9d1f74-a43c-4279-8f82-f11da502e1e8/H-GAC-Resiliency-Pilot-Program-Final- Report.pdf.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 51 FIGURE 3-2 Location of the FHWA pilots.7 NOTE: MPO = Metropolitan Planning Organization. FEDERAL PILOT PROGRAMS Pilot projects funded by the U.S. Department of Transportation (U.S. DOT) through FHWA, the Federal Transit Administration (FTA), the Federal Aviation Administration (FAA), and the Office of the Secretary have been a significant means of advancing the practice of resilience planning and decision making among transportation agencies. FHWA has conducted five series of resilience pilots since 2010, for a total of 46 pilots across the United States (see Figure 3-2).8 Series subjects included asset management and vulnerability assessments. Each pilot series had a well-defined set of goals and tested resilience concepts or guidelines developed or promoted by FHWA. Each pilot series provided the resources necessary to launch resilience practices in transportation and planning organizations across the country, while providing FHWA with the lessons 7 FHWA. 2020. “Resilience Pilots.” https://www.fhwa.dot.gov/environment/sustainability/ resilience/pilots/index.cfm?format=list#map. 8 FHWA. 2020. “Resilience Pilots.” https://www.fhwa.dot.gov/environment/sustainability/ resilience/pilots/index.cfm?format=list#map.

52 INVESTING IN TRANSPORTATION RESILIENCE learned to further develop or update its guidance documents and resilience tools. FTA conducted pilots through its Climate Change Adaptation Initia- tive, launched in 2011.9 The program funded pilots for nine transit agen- cies in seven locations.10 Each pilot identified current and future climate hazards (in particular flooding and extreme precipitation, extreme heat, sea level rise, and tropical storms and hurricanes), assessed system vulner- abilities, and developed adaptation strategies for the specific transit system. Individual pilots tested developing resilience indicators, using life-cycle cost assessment to evaluate adaptation actions, and incorporating vulnerabilities into an asset management system. COMPREHENSIVE APPROACHES TO RESILIENCE Although many agencies mix quantitative and qualitative methods in their resilience assessments, some agencies are experimenting with comprehen- sive approaches to resilience that emphasize quantitative analyses of risk and resilience. As part of their multi-hazard, system-wide assessment of resilience, the Colorado DOT and the Utah DOT have used the RAMCAP model to produce quantitative estimates of the reduction in risk associated with proposed investments in improving the resilience of their highway assets. Hazus-MH and the Resilience and Disaster Recovery Metamodel, still under development, are examples of tools designed to quantify the risk from hazard events and the costs and benefits of investments in resilience. RAMCAP Models The most comprehensive resilience assessment procedures currently in use are based on the RAMCAP model developed by the ASME (formerly known as the American Society of Mechanical Engineers) Innovative Tech- nologies Institute, LLC, to provide a consistent way to evaluate risk across different types of assets and hazards (see Box 3-1). 9 FTA (Federal Transit Administration). n.d. “Transit and Climate Change Adaptation: Synthesis of FTA-Funded Pilot Projects.” https://www.transit.dot.gov/sites/fta.dot.gov/files/ FTA0069_Research_Report_ Summary.pdf. 10 FTA. 2014. “Transit and Climate Change Adaptation: Synthesis of FTA-Funded Pilot Projects.” Report No. 0069. https://cms7.fta.dot.gov/sites/fta.dot.gov/files/FTA_Report_No._0069. pdf.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 53 BOX 3-1 RAMCAP Plusa—Basic Model Structure The RAMCAP model grew out of a 2002 White House conference on the protec- tion of critical infrastructure. The highest priority of the more than 100 senior executives from the private sector in attendance was the creation of “an objec- tive, consistent and efficient method for assessing and reducing infrastructure risks in terms directly comparable among the assets of a given sector and across sectors.”a RAMCAP Plus is the most current version of the continuing development of RAMCAP. The RAMCAP Plus Process for analysis is divided into seven steps: Step 1 – Asset Characterization Step 2 – Threat Characterization Step 3 – Consequence Analysis Step 4 – Vulnerability Analysis Step 5 – Threat Assessment Step 6 – Risk and Resilience Assessment Step 7 – Risk and Resilience Management RAMCAP calculates risk based on the “worst reasonable consequence” resulting from damage of critical infrastructure assets. RAMCAP also requires developing a threat (or hazard) scenario that characterizes the threat, including its magnitude. Risk is computed as follows: Risk = Threat Probability × Vulnerability × Consequence Threat (or hazard) probability is the likelihood that a given asset will expe- rience the threat scenario. Vulnerability is the probability that an asset will be damaged or destroyed in the given threat scenario. Consequence is the cost to asset owners and users resulting from the disaster scenario. RAMCAP does not include a measure of criticality, per se, but it does take into account broader economic impacts of service disruptions that are typically taken into account in criticality estimates. a ASME Innovative Technologies Institute, LLC. 2009. All Hazards Risk and Resilience: Prioritizing Critical Infrastructures Using the RAMCAP Plus Approach. https://files.asme.org/ ASMEITI/RAMCAP/17978.pdf.

54 INVESTING IN TRANSPORTATION RESILIENCE Colorado DOT RAMCAP offers a systematic and quantitative framework for integrating risk and resilience, and Colorado DOT’s application has particular value in that it moves the core ideas of RAMCAP into practice. Colorado DOT’s “Risk and Resilience Analysis Procedure” is designed to bring natural hazards into its risk-based asset management program. It covers rockfalls, floods, and debris flows after fire for roadways, bridges, and culverts. The resilience analysis produces two output measures—annual risk and level of resilience. Risk is measured in terms of the expected costs to the asset’s owners and users from each natural hazard. Level of resilience measures the overall level of resilience for specific highway segments, taking into account both the cumulative annual risk and a broader range of criticality measures indicating the importance of the asset to society.11 Colorado DOT’s procedure uses consistent criteria and methods to screen for risk and criticality and to conduct benefit-cost analysis (BCA) for a defined set of potential mitigation measures. The Colorado DOT is working to automate more of the data-entry process, which has been con- ducted manually for specific projects, so that it can be done in batches by type of natural hazard or asset. Aggregate measures of annual risk can also be produced across natural hazard types. Colorado DOT’s approach measures annual risk in dollars. It defines risk as the product of the likelihood of the hazard, the vulnerability of the asset to the hazard, and the consequences of the damage from the hazard in terms of costs to owners and users. Both the likelihood of the hazard and the vulnerability of the asset are calculated as probabilities. Risk is measured as an expected annual cost due to all of the hazards considered. The Colorado DOT thus uses input data to estimate several intermediate measures—hazard likelihood, vulnerability, and consequences—and then uses those intermediate measures to calculate the output measure—annual risk. The Colorado DOT also uses input data to calculate another inter- mediate measure, criticality, which it uses to calculate the level of resilience. To calculate the threat probability (annual likelihood of floods, rock- falls, and debris flow), the Colorado DOT uses historical data on frequency and magnitude of hazard events. The analysis produces maps that identify the likelihood of natural hazards as probabilities. The probability maps do not yet include the projected impacts of climate change or other changes in extreme weather. 11 This section draws on Colorado DOT. 2020. “Risk and Resilience Analysis Proce- dure.” https://www.codot.gov/programs/planning/cdot-rnr-analysis-procedure-8-4-2020-v6. pdf; Kemp, L. 2020. Presentation to the Committee on Transportation Resilience Metrics, September 14.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 55 Vulnerability, the second intermediate measure for annual risk, is defined as the probability of damage to an asset caused by a specific hazard. Specifically, it is “the probability of the Worst Reasonable Case occurring,” assuming that a hazard event has happened. Input data for vulnerability incorporate the physical characteristics of the asset and its location. The procedure assigns these probabilities based on guidance produced from a mix of published research, empirical data, and expert judgment. To calculate the consequences of an event (the third intermediate mea- sure used in calculating annual risk), the analysis must first define what is meant by an “event,” including its characteristics. The Colorado DOT chose to use the Worst Reasonable Case as the event, defined as “the maxi- mum realistic losses.” The Colorado DOT defined a Worst Reasonable Case for 11 hazard/asset pairs (e.g., rockfall/roadway or flood/bridge approach) from the perspective of costs to both owners and users. To calculate the consequences to owners of the Worst Reasonable Case event, the procedure measures the costs of asset replacement and cleanup. Staff working through a collaborative workshop process identified these costs for each hazard/asset pair. For consequences to users, the Colorado DOT defined the Worst Reason able Case event in terms of the “maximum number of full or partial closure days.” To develop costs for users, the procedure divides users into passenger and freight traffic, developing separate models of costs per mile and per hour for each. The calculation of user costs also required the devel- opment of a new traffic model to measure the length of detours required by loss of service on a highway segment. To calculate the benefits of mitigation actions, the effects of these actions on reducing vulnerability are assessed. Actions designed to mitigate or prevent harm result in a lower vulnerability probability. These differ- ences in vulnerability then allow for a comparison of annual risk with and without a specified investment in mitigation to produce an estimate in dollars of the benefits of mitigation. In addition to the calculation of annual risk, the Colorado DOT defines “criticality” as a “measure of the importance of an asset to the resilience of an overall system.” The “overall system” for highways is defined as high- ways in Colorado on the National Highway System or otherwise owned by the Colorado DOT. To measure criticality, the procedure combines six vari- ables. Three are highway measures: average annual daily traffic, functional classification, and system redundancy. The other three are economic and social indicators measured at the county level. Freight value and tourism dollars generated are measured in millions of dollars per year. The social indicator, the Social Vulnerability Index (SoVI®), was obtained from the University of South Carolina Hazards and Vulnerability Research Institute. SoVI® combines metrics from 29 socioeconomic indicators representing

56 INVESTING IN TRANSPORTATION RESILIENCE characteristics of the people in each county.12 The criticality procedure then transforms each of the six metrics into a score from 1 to 5, which is summed with equal weighting into an overall criticality score. The criticality score is further categorized as low, medium, and high. The Colorado DOT rated 21% of its highway mileage as highly critical. Colorado DOT’s level of resilience metric is a matrix that displays an- nual risk against criticality scores. For each mile of the highway system, an aggregate annual risk—across all hazards and asset types—is calculated. The miles are then ranked from low to high annual risk and sorted into quintiles. The procedure then assigns a “resilience” score of A through E, from best to worst, for each cumulative annual risk/criticality pair (Figure 3-3 illustrates an example). Calculating aggregate annual risk for every mile of highway allows for overall assessments of resilience but also requires more extensive data than analysis approaches that focus on a more limited list of specific locations, hazards, and asset types. The Colorado DOT initiated its application of the RAMCAP model by doing a pilot study of highway assets that had twice been damaged by natu- ral disasters since 1997 and hence were likely to be at particularly high risk. The case study only considered assets that were under consideration for the Statewide Transportation Improvement Program (STIP). The selection cri- terion of twice damaged substituted for a more comprehensive assessment of natural hazards, limited the scope of the necessary asset inven tory, and informed the baseline analysis of the impacts of natural hazards. The stipu- lation that these assets be part of projects proposed for the STIP served as a proxy for determining their importance. The agency then used its modeling capabilities to assess interventions, measured in terms of the owner costs of damages and the user costs of delays and detours. The results of the resil- ience analysis then fed into the larger project selection process for the STIP. Utah DOT The Utah DOT also uses RAMCAP, with some significant differences from Colorado DOT’s analysis procedure. The Utah DOT selected a larger group of natural hazards for more in-depth analysis: avalanches and earthquakes (for bridges only) in addition to rockfalls, floods, and debris flows. The Utah DOT does not use “vulnerability” (the probability of damage to an asset from a hazard event) to calculate annual risk, though it intends to do so in the future. Instead, the agency treated the probability of a hazard as the probability of a hazard great enough to cause total failure of the asset. If total failure of the asset occurs, then the “vulnerability” term is 12 Hazard and Vulnerability Research Institute. n.d. “SoVI.” University of South Carolina. http://artsandsciences.sc.edu/geog/hvri/sovi%C2%AE-0.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 57 FIGURE 3-3 The Colorado DOT is experimenting with a measure of resilience that combines measures for cumulative annual risk and criticality in a way that can assess the comparative resilience of an entire highway system. This figure shows variations in the level of resilience for segments of I-70.13 equal to 1, and hence the term drops out of the equation when the terms are multiplied together. The Utah DOT does use the change in sensitivity (or vulnerability), defined as “a measure of how much damage will occur” from a particular event, to calculate the benefits of efforts to mitigate risk by reducing sensitivity. It calculates sensitivity for a continuous range of haz- ard events from no damage to complete failure. For consequences, the Utah DOT also uses owner costs and user costs but has developed its own way to measure them. The Utah DOT uses a measure of criticality to set priorities for different mitigation investments, using only highway-related factors— redundancy, average annual daily traffic, and truck traffic—and weighs redundancy to be more than twice as important as each of the other two. The Utah DOT also uses criticality and annual risk as quantitative measures 13 Colorado DOT. 2020. “Risk and Resilience Analysis Procedure.” https://www.codot.gov/ programs/planning/cdot-rnr-analysis-procedure-8-4-2020-v6.pdf.

58 INVESTING IN TRANSPORTATION RESILIENCE to produce a quantitative measure of resilience, defined as 1 divided by the product of risk and criticality. As with the Colorado DOT, the Utah DOT used pilot studies (including one focusing on I-15) to develop its resilience methodology.14 Hazus-MH The most popular tool for estimating the impacts of natural hazards is Hazus-MH, a nationally standardized risk modeling methodology that is managed by the Federal Emergency Management Agency. The geographic information system–based tool identifies and maps regions exposed to earthquakes, tsunamis, hurricanes, and coastal and riverine flooding and produces quantitative estimates of the direct physical, economic, and social impacts of hazard events (see Figure 3-4). Hazus-MH uses information on buildings, infrastructure, population, extreme event extent and intensity, and damage functions to estimate losses and risks. Hazus-MH was designed for simplicity of use and comes with default databases pre-embedded in the program. Hazus-MH considers the following transportation infrastructure: highway, rail, light rail, bus, port, ferry, and airport. Its inventories of buildings are periodically updated, and users can import their own data on buildings and structures. It can also perform a rough assessment of the recovery curves described in this chapter.15 Hazus-MH can be used to analyze the cost-effectiveness of common mitigation strategies, such as elevating buildings and structures to prevent flood damage. It can be effective for identifying risks and helping support decisions for major investments on a class of assets in a region. However, because its analysis resolution is coarse, Hazus-MH may not be appropri- ate for many types of transportation impacts and for smaller mitigation actions.16 Resilience and Disaster Recovery Metamodel Being developed to fill the risk analysis gaps left by tools such as Hazus- MH, the Resilience and Disaster Recovery Metamodel (RDRM) is part of a pilot project sponsored by FHWA with U.S. DOT’s Office of the Assistant Secretary for Research and Technology and Office of Intelligence, Secu- rity, and Emergency Response; HRTPO; and the John A. Volpe National 14 Utah DOT. 2020. “UDOT Asset Risk Management Process.” 15 FEMA (Federal Emergency Management Agency)–U.S. Department of Homeland Secu- rity. 2019. “Hazus 4.2.” 16 FEMA. 2020. “What Is Hazus?” https://www.fema.gov/flood-maps/tools-resources/flood- map-products/hazus/about.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 59 FIGURE 3-4 Hazus-MH can be used to compare regional seismic risk by annual- ized earthquake losses.17 Transportation Systems Center.18 The FAST Act required the incorporation of resilience considerations into transportation planning, and the develop- ment of the RDRM is part of FHWA’s effort to respond to that mandate. The goal of the RDRM project is to develop a “nationally replicable model- ing tool that quantifies direct and indirect costs” of the kinds of disruptive events associated with natural hazards. It will allow calculation of benefits and costs and returns on investment of various resilience investments. The power of the tool is that it will allow transportation agencies to use hazard scenarios to compare the costs of different levels of disruption against the costs of potential hazard mitigation or adaptation actions. The RDRM uses many of the same kinds of input data and intermediate 17 Jaiswal, K., D. Bausch, J. Rozelle, J. Holub, and S. McGowan. 2017. Hazus® Estimated Annualized Earthquake Losses for the United States. FEMA P-366, April. https://www.fema. gov/sites/default/files/2020-07/fema_earthquakes_hazus-estimated-annualized-earthquake-losses- for-the-united-states_20170401.pdf. 18 The information in this section draws from the presentation of D.M. Stith, Principal Transportation Planner, HRTPO, to the Committee on Transportation Resilience Metrics, September 17.

60 INVESTING IN TRANSPORTATION RESILIENCE measures as RAMCAP models, such as hazard probabilities, vulnerability of infrastructure assets to particular hazards, and the consequences of damages to infrastructure. It is intended to be used in conjunction with the travel demand models used by metropolitan and regional planning organizations. It focuses particularly on the wider economic impacts of disaster-related disruptions of the transportation network, including on regional economic impacts, disruptions to port access, access for emergency vehicles, and commuting patterns. It pays particular attention to uncertain- ties in the input data and estimates standard deviations in the benefits and costs of different investment options. HRTPO is using the tool to develop scenarios for its long-range transportation plan and to evaluate projects for prioritization in the plan. Examples of the HRTPO regional significance pri- oritization measures include usage volumes/ridership; travel time reliability; impact on freight movement; defense, port, and tourism access; and access to areas with high unemployment and low-income areas. Although HRTPO is analyzing the impacts of sea level rise, the re- silience metamodel is designed to be able to address a variety of natural hazards. If the model design is successful, the resilience metamodel could become a widely used tool to measure the level of resilience and evaluate potential investments to improve resilience. It could be used, along with travel demand models, land use models, and economic models, in develop- ing long-range transportation plans and their associated capital improve- ment programs. VULNERABILITY ASSESSMENTS Vulnerability assessments represent a first step toward developing a plan for improving resilience. They help a transportation agency or a multi- stakeholder planning process prioritize which specific assets, services, or systems are most at risk from natural hazards and should be included in the subsequent analysis that identifies and evaluates strategies and actions designed to increase resilience. Metrics are used in vulnerability assessments to identify the character and likelihood of natural hazard events and com- pare the vulnerability of different types of assets. U.S. DOT Vulnerability Assessment Tools To encourage and guide vulnerability assessments, U.S. DOT developed VAST in 2015. The tool provides libraries of indicators for the three facets of vulnerability: exposure (equivalent to what RAMCAP models call threat probability), sensitivity (equivalent to what RAMCAP models call vulner- ability), and adaptive capacity (similar to the concept of redundancy). The indicators are in the form of scores of increasing vulnerability, on a scale

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 61 of 1 to 4. For example, for sensitivity to higher temperatures, ballast type is one indicator for rail, and age of buses is one indicator for mass transit services. The libraries cover six asset types (roads, bridges, rail lines, ports, airports, and transit assets) and five climate stressors (temperature changes, precipitation changes, sea level rise, storm surge, and wind). The tool only covers climate-related hazards and not geophysical hazards, such as earth- quakes. The scores are designed to allow for comparing the vulnerability of different types of assets.19 In addition, FHWA’s Vulnerability and Adaptation Framework pro- vides guidance on how to use vulnerability assessments in resilience plan- ning processes.20 As a result, vulnerability assessments are becoming a prevalent practice in resilience planning. Vulnerability Assessment Case Study: The San Diego International Airport The Climate Resilience Plan from the San Diego International Airport (SDIA) shows how quantitative approaches to hazard likelihood and char- acter fit into vulnerability assessments and resilience planning. Managers at the San Diego County Regional Airport Authority, which operates SDIA, conducted a vulnerability assessment funded through the Sustainable Management Planning grant program from FAA as part of developing a Climate Resilience Plan. The vulnerability assessment in- formed subsequent steps in the planning process, including evaluating consequences, setting goals and targets, and selecting a list of actions de- signed to increase the airport’s resilience. For the vulnerability assessment, three climate stressors were examined—sea level rise, precipitation, and heat—and the airport’s assets and operating systems were grouped into five analysis categories: runways, taxiways, and navigational systems; airport facilities; tenant facilities; ground transportation networks (including access roads and parking lots); and the habitat of the least tern, an endangered bird species.21 The vulnerability assessment follows the pattern laid out in FHWA’s VAST, using the analysis process outlined in Figure 3-5, to select which assets were vulnerable to the climate stressors. Step 1 defines the exposure 19 U.S. DOT’s “Sensitivity Matrix” covers a wider range of modes and climate stressors for the sensitivity variable in vulnerability assessments. U.S. DOT. 2015. “Vulnerability Assess- ment Scoring Tool.” https://www.fhwa.dot.gov/environment/sustainability/resilience/tools; U.S. DOT. 2015. “Sensitivity Matrix.” https://www.fhwa.dot.gov/environment/sustainability/ resilience/tools. 20 FHWA. 2017. Vulnerability and Adaptation Framework, Third Edition. https://www. fhwa.dot.gov/environment/sustainability/resilience/adaptation_framework. 21 SDIA (San Diego International Airport). 2020. Climate Resilience Plan; Reed, B. 2020. Presentation to the Committee on Transportation Resilience Metrics, September 17.

62 INVESTING IN TRANSPORTATION RESILIENCE FIGURE 3-5 Steps for the vulnerability assessment conducted by the San Diego International Airport.22 to the hazard in terms of its nature and degree. Step 2 identifies the sensitiv- ity or “the degree to which the physical condition and functionality of an asset, population, or system is affected by climate stressors.” The analysis of adaptive capacity in Step 3 requires identifying the “inherent charac- teristics that allow the asset to readily respond or adapt” to the stressors. Factors that influenced sensitivity include, for example, the presence of electrical equipment, while adaptive capacity was influenced by factors such as the ability to elevate or relocate assets. Although analysis of sensitivity and adaptive capacity yielded important information, analysis of exposure turned out to be the most important of the three for assessing vulnerability. To analyze exposure, scenarios were developed for precipitation, heat, and flooding connected to sea level rise and storm surge.23 The scenarios cover multiple time frames, from the present to the year 2100. For these climate change impact scenarios, the SDIA analysts and planners benefited from guidance from the State of California; such guidance streamlined the resilience planning process and lessened the resources that the airport needed for its assessment. The climate change models showed no change in precipitation from the present for 2050 and only small changes in 2100. For storm surge on top of sea level rise, multiple scenarios were developed corresponding to different levels of carbon emissions, and areas projected to be exposed to flooding were mapped. The areas expected to be flooded with a probability of 5% were defined as the high projections for 2030, 2050, and 2100. For 2100, this amount is 4.9 feet of sea level rise. Figure 3-6 is the map of the high scenario for 2100; the maximum high tide is in blue and the additional flooding from storm surge is in green. 22 SDIA. 2020. Climate Resilience Plan; Reed, B. 2020. Presentation to the Committee on Transportation Resilience Metrics, September 17. 23 They explored other natural hazards including wildfires and changes in wind from storms, but data and modeling for projections were not available to develop quantitative scenarios.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 63 FIGURE 3-6 San Diego International Airport flooding forecast due to sea level rise, maximum high tide, and 100-year storm surge.24 The San Diego County Regional Airport Authority then went beyond the vulnerability assessment to examine the consequences of the vulner- abilities identified and developed a preliminary list of initiatives to mitigate those vulnerabilities. For assets deemed vulnerable, they conducted a “high- level risk assessment” that analyzed the potential economic, social, and environmental consequences of the damage or disruption associated with each climate stressor. Economic consequences were considered in terms of asset damage and service disruption. Social consequences are made up of the loss of jobs, the quality of passenger experience, and life safety. Envi- ronmental consequences focus on loss of habitat for the endangered least tern and reduction of water quality in the San Diego Bay. The analysis of the consequences consists of qualitative descriptions. A vulnerability profile was then created for each asset category, using the results of the vulnerability assessment and the high-level risk assessment. 24 SDIA. 2020. Climate Resilience Plan; Reed, B. 2020. Presentation to the Committee on Transportation Resilience Metrics, September 17.

64 INVESTING IN TRANSPORTATION RESILIENCE The vulnerability profiles identified, through qualitative descriptions, which specific assets are vulnerable to which stressor and during which time frame. The specificity in the vulnerability profiles allowed for the develop- ment of a corresponding list of initiatives to be implemented in the near, medium, and long term. The identified initiatives were built around three strategic areas: infrastructure (how we build), governance (how we man- age), and awareness (how we learn). The infrastructure initiatives vary in their specificity. One initiative targeting heat proposes to “reduce heat island effect by resurfacing dark rooftops and pavements with remaining lifespans of more than 10 years.” Another initiative targeting the flooding associated with sea level rise and storm surge states the following: “raise shoreline to protect assets,” either by permanent or temporary barriers. These alternatives still need to be evaluated, in coordination with the external parties, for cost effectiveness.25 The goal of the Climate Resilience Plan is to “reduce risks associated with climate change.”26 The initial targets focus on achieving “zero reports of negative impacts on Airport facilities due to flooding or extreme heat days” by the year 2035,27 but the San Diego County Regional Airport Authority also cites forecasts of climate change out to the year 2100, target- ing the airport to be resilient to a flood that has no more than a 5% chance of occurring in the year 2100. RESILIENCE INDICATORS: LOS ANGELES COUNTY METROPOLITAN TRANSPORTATION AUTHORITY Resilience indicators track characteristics that suggest whether an asset or a system is resilient. Although agencies sometimes resort to using indicators in cases where producing the quantitative metric itself is difficult, indicators can also be used to provide useful guidance for management decision making. The Los Angeles County Metropolitan Transportation Authority ( LACMTA) uses indicators to evaluate its progress in implementing a program designed to increase resilience. As appropriate for an agency that both operates transportation services and constructs and maintains trans- portation infrastructure, LACMTA’s resilience indicators cover a broad range of technical assessments and organizational activities. The indicators are designed to predict how resilient the agency, its infrastructure, and its services will be when faced with a natural hazard event. The indicators are not designed to provide information on which investments in improving resilience will have the greatest net benefits. 25 SDIA. 2020. Climate Resilience Plan, pp. 60–62. 26 SDIA. 2020. Climate Resilience Plan, p. 46. 27 SDIA. 2020. Climate Resilience Plan, p. 46.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 65 The agency produced its first organization-wide “Resilience Indica- tor Framework” in 201528 and in 2020 issued a significant update.29 The agency built on a vulnerability assessment completed in 2014 and on a previous pilot project for transit indicators funded by FTA.30 The agency se- lected the indicators after a review of research and best practices worldwide and adapted them to be specific to LACMTA’s organization and practices.31 Table 3-1 lists the technical and organizational indicators as refined in 2020. Technical indicators evaluate the performance of physical systems. They can be used to evaluate a single asset or a group of assets that work together, such as the assets that make up the communication system or all of the stations along a rail line. Organizational indicators evaluate the capacity of the organization to make decisions and to act. Although evaluations of costs and benefits are not included, nothing precludes examining the costs and benefits of specific actions.32 TABLE 3-1 LACMTA Resilience Framework Updated Resilience Indicators33 Technical Indicators Organizational Indicators Robustness R-01. Maintenance – Day to Day R-02. Maintenance – Post Incident R-03. Renewal/Upgrade (Long Range Plans) R-04. Design – Compliance with Current Codes R-05. Design – Condition of Asset R-06. Design – Vulnerability Assessment R-07. Design – Resilience Design Criteria R-08. Design – Overheating Standards R-09. Extreme Weather Repair Costs R-10. Supplier Utility Robustness – Awareness R-11. Supplier Utility Robustness – Improvement Information Management and Communication I-01. Warnings and Public Awareness I-02. Communication Systems – Staff I-03. External – Public Awareness I-04. Sensors I-05. Data – Access to, and Maintenance of, Key Data Sets I-06. Information Security and Contingency Planning 28 LACMTA (Los Angeles Metropolitan Transportation Authority). 2015. “Resiliency Indi- cator Framework.” http://media.metro.net/projects_studies/sustainability/images/resiliency_in- dicator_framework.pdf. 29 LACMTA. 2020. “Resiliency Indicator Framework: 2020 Addendum.” http://media. metro.net/2020/Addendum-to-Resiliency-Framework.pdf. 30 FTA. 2013. Los Angeles County Metropolitan Transportation Authority Climate Change Adaptation Pilot Project Report. FTA Report No. 0073. https://www.transit.dot.gov/sites/fta. dot.gov/files/FTA_Report_No._0073.pdf. 31 LACMTA. 2015. “Resiliency Indicator Framework.” http://media.metro.net/projects_ studies/sustainability/images/resiliency_indicator_framework.pdf. 32 LACMTA. 2015. “Resiliency Indicator Framework.” http://media.metro.net/projects_ studies/sustainability/images/resiliency_indicator_framework.pdf. 33 LACMTA. 2020. “Resiliency Indicator Framework: 2020 Addendum.” http://media. metro.net/2020/Addendum-to-Resiliency-Framework.pdf. continued

66 INVESTING IN TRANSPORTATION RESILIENCE Technical Indicators Organizational Indicators Redundancy RE-01. Alternate Route/Mode Availability RE-02. Alternate Route/Mode Capacity RE-03. Spare Capacity RE-04. Back Up Parts and Equipment RE-05. Re-routing and Communication Plans RE-06. Supplier Utility Redundancy – Awareness RE-07. Supplier Utility Redundancy – Improvements All Hazards Planning, Preparedness, and Response A-01. Risk Assessment and Scenario Planning A-02. Tracking Incident-Related Injuries A-03. Tracking Essential Resources A-04. Priority Routes/Structures A-05. Emergency Management Plans – Existence A-06. Joint Planning A-07. Training/Drills – Offered A-08. Training/Drills/Tests – Completed A-09. Lessons Learned and Thinking Ahead A-10. Critical Energy and Supply Chain Provision Financial Preparedness F-01. Insurance Coverage F-02. Insurance Information F-03. Capital Availability F-04. Operational Funding F-05. Contingency Funding F-06. Modelling Networks and Staffing N-01. Internal Relationships N-02. Information Sharing – Internal N-03. Inter-agency Compatibility N-04. Business Continuity/Awareness N-05. External Information Sharing and Cooperation N-06. Roles and Responsibilities Identified N-07. Remote Response Ability Leadership and Culture L-01. Resilience Is a Clear Priority of Metro Leadership L-02. Roles, Responsibilities, and Goals L-03. Staff Engagement and Leveraging Knowledge L-04. Crisis Decision Making L-05. Mid/Long Term Decision Making L-06. Advance Agreements L-07. Approach to Projects TABLE 3-1 Continued

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 67 Each indicator is accompanied by a grading rubric that describes a score from 1 to 4, with a score of 4 representing the highest level of resil- ience. For example, for the resilience design criteria indicator, the lowest score corresponds to no resilience design criteria and the highest score to “resilience design criteria have been developed and strategies have been implemented for new and upgrade/repair projects.” For the alternative route/mode capacity indicator, the lowest score is assigned if the “alternate mode has <25% capacity of the failed mode during peak demand” and the highest score if the “alternate, unaffected mode has >75% capacity of failed mode during peak demand.”34 The 1–4 scores are then transformed into percentages to make them easier for the public to understand. Although the 2015 framework focused only on extreme weather related to climate change, the 2020 update adopted an “all hazards” approach because “many actions needed to ensure resilience against climate change are the same ones needed to ensure resiliency against other hazards.” The 2020 update applies to 11 natural hazards and nine human-caused threats, and all indicators are relevant to all hazards and threats.35 LACMTA’s indicator framework can be used to track its progress over time and to evaluate the strengths and weaknesses of the agency’s readiness for specific hazards. Scores for each principle are generated by averaging the scores of the principle’s indicators. For example, given a scenario of an earthquake similar to 1994’s Northridge earthquake, scores generated by using the indicator framework revealed that, among the organizational principles, LACMTA was strongest in Networks and Staffing and weakest in Information Management and Communication.36 An overall resilience score can be produced by averaging the scores of the seven principles. Although LACMTA experimented with weighting indicators and principles, the agency decided in the 2020 revision to weigh equally all indicators within each principle and all principles in the overall resilience score. It is notable that, of the seven principles, five relate to orga- nizational indicators and only two to technical (i.e., infrastructure-related) indicators. Because the seven principles are weighted equally, 71% of the weight in the final resilience score is drawn from organizational indicators. 34 LACMTA. 2015. “Resiliency Indicator Framework,” pp. 25, 27. http://media.metro.net/ projects_studies/sustainability/images/resiliency_indicator_framework.pdf. 35 LACMTA. 2020. “Resiliency Indicator Framework: 2020 Addendum,” p. 2. http://media. metro.net/2020/Addendum-to-Resiliency-Framework.pdf. 36 LACMTA. 2020. “Resiliency Indicator Framework: 2020 Addendum,” pp. 10–11. http:// media.metro.net/2020/Addendum-to-Resiliency-Framework.pdf.

68 INVESTING IN TRANSPORTATION RESILIENCE DESIGN GUIDES The practice of resilience is, for the most part, still in the stages of cus- tomized analysis and application experimentation. Design guides are one example of how resilience may become part of the routine practices of transportation agencies. Many resilience plans call for mitigation or adaptation practices to be institutionalized as part of design guides. Especially for smaller projects, design guides may also reduce the need to conduct resilience analysis on a project-by-project basis. The Climate Resilience Design Guidelines from the engineering department of The Port Authority of New York and New Jersey standardizes the agency’s response to sea level rise and storm surge, using the requirements in the respective state building codes and augment- ing them as needed. The design guide sets elevation standards, in inches, depend ing on the probability of the flood hazard, whether the asset is deemed critical or non-critical in the building codes or in the Port Author- ity’s own assessment, and the asset’s design life.37 New York City’s Climate Resiliency Design Guidelines prepare public investments for future climate change by standardizing resilient design cri- teria across the city’s wide portfolio of assets. By means of local law, these guidelines have been recently mandated for all capital projects in New York City. The guidelines translate localized climate projections for heat, precipitation, and sea level rise into data sets that can be used by project designers (see Table 3-2 below for guidance on engineering with future heat conditions) based on the project’s useful life and criticality. The guidelines also provide tools for project managers to assist in resilient design decision making, such as risk assessment methodology and BCA, that are scalable based on the project size. For likelihood of the natural hazard, the guide- lines instruct users to assign a rating to hazards on a qualitative scale from rare to nearly certain. Similarly, consequences are to be summarized on the scale of minor, moderate, and severe. The product of likelihood and consequences is called the “risk rating matrix.” To choose resilient designs above and beyond those required by building codes, the guidelines advise conducting a qualitative assessment of benefits for capital projects under $50 million. Quantitative, detailed BCA is reserved for projects that have more than $50 million in total costs.38 37 The Port Authority of New York and New Jersey. 2018. Climate Resilience Design Guidelines. https://www.panynj.gov/business-opportunities/pdf/discipline-guidelines/climate- resilience.pdf. 38 New York City Mayor’s Office of Recovery and Resilience. 2019. Climate Resiliency Design Guidelines: Version 3.0.

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 69 TABLE 3-2 New York City Climate Resiliency Design Guidelines Climate Change Data for Designing with Future Heat Conditions Extreme Heat Events Design Criteria End of Useful Life # of Heat Waves per Year # of Days at or Above 90oF Annual Average Temperature 1% Dry Bulb Temperature Cooling Degree Days (Base –65°F) Historic trend (1971–2000) 2 18 54oF 91oF 1,149 2020s (through 2039) 4 33 57.2oF — — 2050s (2040–2069) 7 57 60.6oF 98oF 2,149 2080s (2070–2099) 9 87 64.3oF — — NOTES: Due to heating, ventilation, and air conditioning system typical useful life of around 25 years, only design criteria projections for the 2050s are shown. Projections for the 2020s are not shown because it is anticipated that enough of a safety margin is employed already in current systems to withstand the temperature rise expected through the 2020s. The Northeast Power Coordinating Council is developing projections of 1% wet bulb temperatures, which are expected to increase. This design criteria will be added in a later version of the guidelines. SOURCE: City of New York. 2019. The table data were provided with the permission of the City of New York. SUMMARY OF METRICS This section presents a table of metrics that some transportation agencies are using (see Table 3-3). The table shows the output measures that agencies use to make decisions on resilience improvements, the intermediate mea- sures that they use to calculate the output measures, and a few examples of the input data used to calculate the intermediate measures.

70 INVESTING IN TRANSPORTATION RESILIENCE TABLE 3-3 Summary of Resilience Measures Used by Transportation Agencies Output Measures Intermediate Measures Input Data Annual Risk (Colorado DOT) Hazard probability Probability of rockfalls Probability of floods Probability of debris flows Vulnerability Engineering judgment Consequences Repair costs to Colorado DOT Number of days highway closed Length of detour required Lost wages and truck revenues Risk Value (Utah DOT) Hazard probability Flood probability Rockfall probability Avalanche probability Debris flow probability Earthquake probability Consequences Repair costs to Utah DOT Length of detours Hourly value of time Hourly vehicle operating costs Level of Resilience (Colorado DOT) Annual risk (see above) Criticality Freight value Tourism value SoVI® Risk Priority (Utah DOT) Risk value (see above) Criticality Road network redundancy Average annual daily traffic Truck average daily traffic

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 71 Output Measures Intermediate Measures Input Data Vulnerability (San Diego International Airport) Exposure Annual number of days of extreme heat Area exposed to flooding 95th percentile risk of sea level rise Consequences Asset damage Service disruption Job loss Life safety consequences Bird habitat damage Resilience Indicators (LACMTA) Robustness Design – Vulnerability assessment Redundancy Back-up parts and equipment Information management and communication Warnings and public awareness All hazards planning, preparedness, and response Tracking essential resources Financial preparedness Capital availability Leadership and culture Crisis decision making Design Guide (The Port Authority of New York and New Jersey) Hazard probability Projected sea level rise Projected precipitation increase Projected temperature increase Asset service life Number of years before asset is expected to be replaced Asset criticality Classification of asset into “critical” or “non-critical” categories Design Guide (New York City) Hazard probability Projected temperature increase Projected precipitation increase Projected sea level rise Consequences Damage to facility Damage to surrounding community Asset useful life Durability of asset Replaceability of asset Asset criticality Services provided Importance in emergency TABLE 3-3 Continued continued

72 INVESTING IN TRANSPORTATION RESILIENCE Output Measures Intermediate Measures Input Data Net Benefits of Resilience Improvements (Colorado DOT) Annual risk without improvement Annual risk with improvement Costs of improvements Net Benefits of Resilience Improvements (Utah DOT) Risk priority (see above) Costs of improvements Net Benefits of Resilience Improvements (HRTPO) Hazard probability Flood risk Vulnerability Effects of floods on roads, bridges, and tunnels Consequences Wider economic impacts of transportation disruptions Net Benefits of Resilience Improvements (New York City) Direct benefits Quantitative analysis for projects >$50 million Indirect benefits Qualitative analysis for projects <$50 million Other benefits Costs NOTE: The input data shown are just a few examples of the input data used by each agency. CHAPTER SUMMARY The review of current practice found that transportation agencies are pro- gressing in their adoption of analysis, planning, and management practices for addressing resilience. Agencies are primarily integrating resilience into their planning and management practices because of past harms from haz- ard events, as well as federal and state mandates and incentives. Adoption remains uneven, however, and the agencies that do engage in resilience analysis use a variety of methodologies and metrics tailored to the specific infrastructure and services that the agencies provide, as well as agency goals. There is no common set of resilience metrics. The resilience analysis and planning methods used by transportation agencies contain common elements. Agencies analyze hazard likelihood and characterization to assess the vulnerability of their assets, networks, and services. They use vulnerability and criticality assessments to prioritize TABLE 3-3 Continued

MEASURING AND MANAGING TRANSPORTATION SYSTEM RESILIENCE 73 subsequent studies of mitigation actions. They conduct assessments of con- sequences to gain an understanding of the impacts of failing to act in the face of climate change. Agencies differ, however, in their use of quantita- tive analyses, especially monetary assessments of risk in terms of expected losses. Although vulnerability assessments are becoming an established practice, with methods piloted and disseminated by U.S. DOT, many agen- cies still rely on indicators or qualitative descriptors for their analysis of consequences. Tools and practices that foster formal assessments of risk for the status quo and for the reduction of risk from investments in resilience are still in the developmental stage. The analytical approaches described in this chapter are used by agen- cies for a variety of purposes. The majority are used to support decision making at the planning and project level. Outcomes from these analyses are also applied to asset management, maintenance operations, and post- disaster responses and restoration efforts.