USA Banner

Official US Government Icon

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure Site Icon

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Site Notification

Site Notification

U.S. Department of Transportation U.S. Department of Transportation Icon United States Department of Transportation United States Department of Transportation

Public Roads - Spring 2022

Date:
Spring 2022
Issue No:
Vol. 86 No. 1
Publication Number:
FHWA-HRT-22-003
Table of Contents

Partnering for Resilience: The Southeast Michigan Flooding Risk Tool

by Andy Pickard and Rachael Barlock

The Southeast Michigan Council of Governments (SEMCOG) is home to over 4.7 million people in 93 cities like Detroit and Ann Arbor, 24 villages, and 115 townships. As flooding events continued to affect the region—and become potentially more severe and frequent—SEMCOG and the Michigan Department of Transportation (MDOT) both sought to understand where transportation assets were the most vulnerable to flooding and how to address such challenges. Their collaboration resulted in the SEMCOG Climate Resiliency and Flooding Mitigation Study, a comprehensive dataset, and an analysis tool that can be used to predict flooding risk for thousands of key infrastructure assets in the region. Information from the study will be used to guide planning and investment decisions within the transportation network of the seven-county (Livingston, Macomb, Monroe, Oakland, St. Clair, Washtenaw, and Wayne) SEMCOG region. According to Kelly Karll, manager of the Environment and Infrastructure Department at SEMCOG, “Addressing flooding risk requires collaboration across different infrastructure sectors, including transportation and water infrastructure. This includes using a watershed approach to identify strategic, cost-effective solutions while considering impacts of changing precipitation events.” A watershed approach allows for holistic stormwater management that extends beyond the bounds of the transportation network.

"Vehicles drive, in both directions, through a flooded roadway. Image Source: © Jon Clark / SEMCOG."
Flooded roadway in Southeast Michigan.

Resiliency—the ability to recover quickly from a damaging or catastrophic event—is a top priority in transportation plans. It is one of the 10 factors that must be addressed in the metropolitan transportation process per Title 23 of the Code of Federal Regulations (23 CFR 450.306(b)(9) and 23 CFR 450.206(a)(9): Improve the resiliency and reliability of the transportation system and reduce or mitigate stormwater impacts of surface transportation. Several resources developed by the Federal Highway Administration also emphasize resiliency. These resources include the Vulnerability Assessment and Adaptability Framework, 3rd Edition (FHWA-HEP-18-020), U.S. Department of Transportation Vulnerability Assessment Scoring Tool (available at https://www.fhwa.dot.gov/environment/sustainability/resilience/tools/scoring_tools_guide/index.cfm), and the 2013-2015 Climate Resilience Pilot Program: Outcomes, Lessons Learned, and Recommendations (FHWA-HEP-16-079). The Infrastructure Investment and Jobs Act (IIJA) also includes numerous references to resilience. For example, in the highway provisions section, resilience is referenced in requirements (e.g., “requires consideration of extreme weather and resilience in lifecycle cost and risk management analyses”), enhancements (e.g., “…protective features to enhance resilience”), and provisions (e.g., “climate and resilience provisions”). For information on the highway provisions portion of the IIJA, visit: https://www.fhwa.dot.gov/bipartisan-infrastructure-law/docs/bil_overview_20211122.pdf.

In recent years, SEMCOG has taken action to build resiliency by identifying vulnerable natural resources and infrastructure assets and planning to mitigate the impacts of climate hazards, including flooding. For example, in the 2045 Regional Transportation Plan for Southeast Michigan, climate resiliency is called out as a key challenge in the region. In March 2018, SEMCOG also established a Water Resource Plan for Southeast Michigan, which highlights climate resiliency as a major topic of interest and identifies actions to build resiliency. Such actions include identifying vulnerable infrastructure assets and improving adaptive capacity of the systems; integrating resiliency priorities into local policies, plans, and projects; and evaluating opportunities to use natural resource areas for improving management of runoff from extreme precipitation events. Additionally, MDOT piloted a climate change vulnerability assessment in 2015 to identify transportation assets that may be at risk to climate hazards. In 2018, SEMCOG and MDOT received a State planning and research grant to expand the risk analysis to include flooding risk. The SEMCOG Climate Resiliency and Flooding Mitigation Study, described in this article, built upon the 2015 MDOT study and generated several new resources for transportation planners, including insight on how to:

  • Leverage improved data to identify transportation assets at risk of flooding.
  • Develop a repeatable flooding risk assessment tool, with the capability of expansion, to cover a larger geographic area or number of assets.
  • Identify opportunities to integrate flooding risk results into existing planning and investment strategies.
"Flood waters create an impassable section of a busy roadway, marked by traffic barricades and cones. Image Source: © Jon Clark / SEMCOG."
Flooded roadway from the Southeast Michigan flooding event in summer 2021.

Methodology 

To identify flooding risk for four asset types (roads, bridges, culverts, and pump stations), researchers utilized an indicator-based (or flooding risk) methodology for the SEMCOG Climate Resiliency and Flooding Mitigation Study. This approach is similar to that used in the 2015 MDOT assessment, which was in alignment with the FHWA’s Vulnerability Assessment and Adaptation Framework, but expanded the analysis used for the 2015 MDOT study in using data recently available on exposure and sensitivity. The exposure and sensitivity data allowed for a specific focus on the risk of flooding.

The flooding risk methodology is broken into multiple components: vulnerability, criticality, exposure, and sensitivity. Flooding risk is a factor of vulnerability and criticality, and vulnerability is a factor of exposure and sensitivity. Exposure indicates whether the asset is located in an area that experiences the direct effects of flooding. For example, if it rains, the rain will collect in a depressed freeway. Sensitivity refers to how the asset, the depressed freeway, fares when it is exposed to flooding. If that same depressed freeway has cleared catch basins attached to a collection system with enough capacity and pumping capability, it may not be very sensitive. Together, exposure and sensitivity create vulnerability.

Criticality is the importance of an asset to the system or region; criticality is independent of vulnerability. For example, if a depressed freeway is a major shipping route, it would be vital to the area’s local economy, and therefore deemed critical. Criticality and vulnerability create overall risk, and in the case of this study, risk correlates specifically to flooding risk.

"A flow chart with 5 nodes & 3 rectangular callout balloons. On the right side, the flow chart starts with a process box labeled “Risk.” 2 lines extend from the left side of the “Risk” process box; one line extends to a process box labeled “Vulnerability” & the other line extends to a process box labeled “Criticality.” A callout balloon labeled, “The importance of an asset to the transportation system or region as a whole (independent of vulnerability)” extends from the criticality process box. 2 lines extend from the vulnerability process box; one line extends to a process box labeled “Exposure” and the other line extends to a process box labeled “Sensitivity.” A callout box labeled, “Whether the asset or system is located in an area experiencing direct effects of flooding” extends from the exposure process box. A callout box labeled, “How the asset or system fares when exposed to flooding” extends from the sensitivity process box. Image Source: © Rachael Barlock / SEMCOG."
Flowchart of flooding risk indicators.

In this study, several data sources were used to populate the indicators for the flooding risk, which included the Federal Emergency Management Agency (FEMA) Flood Map Service Center to pinpoint flood zone location, pavement condition, traffic volume, and function classification. Each indicator and scoring approach varied for each asset type.

"A flow chart with 15 nodes. On the right side, the flow chart starts with a process box labeled “Risk” with 2 lines extending from the left side; one line extends to a process box labeled “Vulnerability (75%)” & the other line extends to a process box labeled “Criticality (25%).” 2 lines extend from the vulnerability process box; one line extends to a process box labeled “Exposure (75%)” & the other line extends to a process box labeled “Sensitivity (25%).” The exposure, sensitivity, & criticality process boxes have lines extending from the left side to a column of 10 rectangular boxes. The lines from the exposure (75%) process box corresponds to the first 5 rectangular boxes: past flooding experience (75%), FEMA flood zone location (10%), flow accumulation & ponding (6%), impervious surface (6%), change in days with precipitation >3 inches (3%). The lines from the sensitivity (25%) process box extends to past flood damage (50%) & pavement condition (50%). The lines from the criticality (25%) process box extends to the last three rectangular boxes of the column: Traffic volume (33%), functional classification (33%), truck traffic volume (33%). Image Source: © Rachael Barlock / SEMCOG."
Road risk score methodology flowchart.

The flooding indicator weights and approaches were determined through conversations with the study’s project team, MDOT, county transportation planners, and local road and stormwater professionals, as well as baseline guidance from the initial 2015 MDOT study.

"Table with 1 header row, 3 rows and 5 columns. Header row columns are Component, Roads, Bridges, Culverts, and Pump Stations. 1st row: Component is Exposure; then 5 categories span all the other columns: Past flooding experience; FEMA flood zone location; Flow accumulation and ponding; Impervious surface; Projected change in number of days with precipitation greater than 3 inches from baseline (1980–1999) to mid-century (2040–2059). 2nd row: Component is Sensitivity; under Roads are: Past flood damage; Pavement Condition. Under Bridges are: Past flood damage; Scour criticality; For scour critical bridges only, the following were also included: Channel condition, Fracture critical, Single or multi-span. Under Culverts are: Past flood damage; Condition index*; Age*; Inspection comments indicate “full” or “buried”; Proportion of height filled with water; Stream substrate. Under Pump Stations are Past flood damage; Rating. 3rd row: Component is Criticality; under Roads are: Traffic volume; Functional classification; Truck traffic volume. Under Bridges are: Traffic volume; Functional classification; Truck traffic volume; Detour length; Replacement cost. Spanning both Culverts and Pump Stations is: Scored based on the criticality score of the corresponding road segment. Under table is a footnote: * MDOT is in the process of collecting data on culverts. Because of the current limited data available, but potential for improved future data availability, the project team developed a methodology that will be adaptable to take into account the best available data. Currently, few data exist to indicate culvert condition, but the methodology provides an option to use these data as they are collected. Image Source: © Rachael Barlock / SEMCOG."
Indicators used for road, bridge, culvert, and pump station assets to determine risk factors.

It’s important to note that the flooding risk scores ultimately allow SEMCOG, MDOT, and local road agencies to screen which assets are most likely to experience flooding, which can be useful in transportation and stormwater planning. However, flooding risk scores do not provide a direct prediction of flooding from a particular event, and they do not provide any input on expected duration of flooding.

Flooding Risk Tool

The results of this project included the development of the Flooding Risk Tool that calculates a flooding risk score for each road, bridge, culvert, and pump station in the seven-county SEMCOG region. The tool’s calculations are repeatable, and results can be strengthened with additional data as it becomes available. The flooding risk scores produced by the tool are available via the SEMCOG Flooding Risk Tool Dashboard, an interactive online dashboard that allows interested parties to search for specific roads, bridges, or other infrastructure assets using an interactive map, review a flooding risk score and individual indicator scores, and export data as a geodatabase or spreadsheet. The dashboard is available at https://semcog-community.maps.arcgis.com/apps/opsdashboard/index.html#/96cbdd4d71c2462ead70623966e2d1b1. Of the assets included in the analysis, more than 6,300 were determined to be at high risk (a flooding risk score of at least 3 out of 4):

  • Roads: 6,134 (of 71,599)
  • Bridges: 209 (of 2,634)
  • Culverts: 15 (of 3,022)
  • Pump stations: 2 (of 143)

With respect to indicator scoring, the percentage weights can be changed at any time in the Flooding Risk Tool. For example, if it is determined by MDOT that pavement condition should be excluded from this scoring methodology, the tool allows for that modification. In the same way, should more data become available, the indicators can be adjusted and the tool run again.

In this study, the primary driver of high-risk scores was vulnerability, which is 75 percent of the risk score. Most of the high-risk roads, bridges, and pump stations are in Wayne County, and those scores are largely driven by criticality in the county.

SEMCOG will run the Flooding Risk Tool annually with any updated data and provide the results on the SEMCOG Flooding Risk Tool Dashboard.

Using Flooding Risk Scores to Improve Decisionmaking

Integrating the results of the Flooding Risk Tool into transportation and stormwater planning is a critical next step in this study. Resilience depends on a holistic strategy to ensure the application of the flooding risk scores. To determine the best implementation strategies, the project team facilitated working sessions with MDOT, SEMCOG, county and local representatives, and other regional stakeholders. The group identified opportunities to use the flooding risk results in transportation planning.

For instance, at MDOT, the Transportation Asset Management Plan (TAMP), and Program Development Call for Projects Process can be modified to include flooding risk in various stages of the study’s planning process. The TAMP identifies flooding in two threat categories: Climate change and project-level disruptions. Incorporating the flooding risk scores in the Call for Projects Process will allow for capital improvement projects to include flooding risk during the investment’s decisionmaking process.

According to Steve Minton, Metro Associate Region engineer at MDOT, “Focusing on climate resiliency and flooding risk is an absolute must for MDOT when considering the impacts from major rain events and subsequent flooding that have occurred recently and will continue to occur with increased frequency. MDOT will need to look at both short-term solutions and evaluating our approach to stormwater management long term to transition to a resilient transportation network.”

At SEMCOG, coordination will continue with MDOT, especially with respect to continued and improved data collection. Additionally, SEMCOG will continue to work with local partners to facilitate the implementation of the flooding risk scores into transportation planning projects. Currently, SEMCOG is working with local agencies in their region to incorporate flooding risk scores generated by the tool into the Transportation Improvement Program.

Lessons Learned

As part of this study, some major themes emerged:

  • Flooding risk information is an important component when making challenging investment decisions.
  • Working sessions are an effective method for reviewing results and developing integration strategies for flooding risk information.
  • Data sharing between agencies is key to evaluating flooding risk to assets.
  • In an indicator-based risk assessment, complete asset data is critical for generating meaningful results.

Next Steps

The culvert data for this project was often incomplete or missing; a key step for both MDOT and SEMCOG, going forward, is to increase data collection on culverts. Using geospatial analysis, SEMCOG has determined there are approximately 13,000 road/stream crossings in the region, but this project was only able to access data for about 3,000 culverts.

"Rainwater collects over a restricted storm drain cover in a residential street. Image Source: © Rachael Barlock / SEMCOG."
A restricted storm drain during a 2021 heavy rain event.

SEMCOG has also developed the Southeast Michigan Flooding App, a smartphone-based application to report or track the locations of flooding during wet weather events. The app can be found at https://storymaps.arcgis.com/stories/6b834966ed0744c6818c0cf093986982. During the data collection phase of this study, SEMCOG found that many municipalities did not keep a database of frequently flooded locations. To help collect locations of closed roads, flooded shoulders, and blocked culverts, SEMCOG developed a Geographic Information Systems-based application that allows the user to take a photo, record a GPS location, and indicate the severity of the flooding. The intended users of this application are professionals in the water or transportation industry, ranging from stormwater engineers and field technicians to watershed group volunteers and more. In the future, the data collected with this municipal-level flood tracking tool will become another indicator, similar to the FEMA flood maps, that can be used by the Flooding Risk Tool to more accurately calculate risk scores for local infrastructure assets.

"A flooded neighborhood street with barricades and a road closed sign. Image Source: © photogeek / AdobeStock.com."
To help collect locations of closed roads, flooded shoulders, and blocked culverts, SEMCOG developed a Geographic Information Systems-based application.

Andy Pickard, PE, AICP, is a senior transportation planner at the Michigan Division of FHWA, where he assists partner organizations with managing Federal planning requirements and improving transportation planning initiatives. He holds a B.S. in industrial engineering from the University of Illinois at Urbana-Champaign and an M.S. in civil engineering from the Georgia Institute of Technology.

Rachael Barlock is a water resources engineer at the Southeast Michigan Council of Governments where she works on climate resiliency and water infrastructure projects. She holds a B.S. in environmental engineering and an M.S. in civil engineering from Michigan Technological University.

For more information, see https://semcog.org/ or contact Andy Pickard (517-702-1827; andy.pickard@dot.gov) or Rachael Barlock (313-319-1062; barlock@semcog.org).