Using Holistic City Simulation to Test Resilience Adaptation Strategies

Presented by: Stephen Bourne, P.E.

Stephen.Bourne@AtkinsGlobal.com

SAME Southeast Tri-Regional JETS, June 2019

Resilience is on the Rise

2

Source: Google N-Gram Viewer

1800 1900 2000

Develop a tool to forecast resilience under current and proposed city master plans

Recovery

Natural Disaster

Recovery

Natural disasters are simulated at the size and frequency resulting from climate change.

Natural Disaster

Natural Disasters result in little drop

in prosperity

Recovery is fast due to preparedness

Natural Disasters result in large drop

in prosperity

Lack of preparedness makes it difficult to recover. Sometimes the situation worsens after the disaster.

City Prosperity

Index(UN-Specified Index

of City Wellness) Resilient City

UnresilientCity

High Rate of Improvement

Low Rate of Improvement

Business as Usual

Business as Usual

Recovers after a long time, and remains at low rate of improvement.

Resumes high rate of improvement quickly.

Business as Usual

Business as Usual

2015 2020 2025 2030 2035 2040

Quantifying Resilience

Requirements for Forecasting Resilience

• Capture interacting systems (Economy, People, Infrastructure, Natural)

• Include business-as-usual as well as disasters

• Include disasters that are representative of climate change effects

• Allow for addition of proposed strategies and measures their effect

• Include a long enough time line to measure return on investment

EmploymentForecasts jobs based on economic growth

City Information Model

City Information Model Influence of Stormwater Infrastructure on City Productivity

02 July 2019 Future Proofing City Simulator 7

City Simulator Modeling ProcessA nested loop to capture city growth and response to hazards

Natural System

ChangesEcono-

mics

New Places of

Work

Workers

New Housing

Families

NewInfra-

structure

Adjust Flood-plains

Eco-system Impacts

Carbon Footprint

Disaster?

Structures Affected

Structures RecoverWork

Comm-erceYearly Loop

(2017-2050) DailyLoop

Gradually changing threats like rising sea level 3mm per year.

EmploymentForecasts jobs based on economic growth

Percentage growth in economy year over year.

New commercial buildings are added based on growth rate.

Buildings receive an array of businesses with an array of jobs; new jobs are filled by immigrants or population entering workforce.

New workers receive housing, spouses, children.Roads, utilities, schools,

churches, are added to fulfill new populations’ needs.

Urbanization impact to floodplains is evaluated.

Impacts to ecosystem evaluated.

Carbon footprint evaluated from daily travel, city wide

power usage, other factors.

02 July 2019 Future Proofing City Simulator 8

Natural System

ChangesEcono-

mics

New Places of

Work

Workers

New Housing

Families

NewInfra-

structure

Adjust Flood-plains

Eco-system Impacts

Carbon Footprint

Disaster?

Structures Affected

Structures RecoverWork

Comm-erceYearly Loop

(2017-2050) DailyLoop

EmploymentForecasts jobs based on economic growth

In each day of the year, a forecast determines if a disaster occurs.

Structures affected by the disaster are identified through flood models, storm surge models, earthquake risk models, etc.

Damage to structures is estimated using damage curves that equate severity of disaster to cost of damage.

If a structure is affected, a recovery curve is established that governs whether the structure is productive or not in the future.

For workers whose place of work and residence are not

in recovery, the commute and daily work activities are

tracked.

Citizens conduct commerce throughout the day (grocery

shopping, etc). This is also tracked.

City Simulator Modeling ProcessA nested loop to capture city growth and response to hazards

City Simulator in Action

Forecasting Damage at a Building

Forecasting Damage at a Building

0 1 2 3 4 5 6 7 8 9 10 11 Flood Depth (ft)

Direct Damage

80%

60%

40%

20%

2019-2050 Direct Damage = $112K

Rain-To-Flood

HAZUS Depth/Damage

StormCaster Generates this forecast – includes climate change signal

Depth Rasters establish these

points on the curve

% o

f rep

lace

men

t Cos

t

Given Replacement Cost is $100,000, damage = 16% * $100K = $16K

Forecasting Damage at a BuildingLosses Avoided

0 1 2 3 4 5 6 Flood Depth (ft)

Direct Damage

$50,000

$40,000

$30,000

$20,000

$10,000

Total Direct Damage = $112K

Rain-To-Flood HAZUS Depth/Damage

No Mitigation

0 1 2 3 4 5 6 Flood Depth (ft)

Direct Damage

$50,000

$40,000

$30,000

$20,000

$10,000

Total Direct Damage = $46K

Rain-To-Flood HAZUS Depth/Damage

Structure Elevated: Cost = $50K

Losses Avoided = $112k - $46K = $66K ROI = $66K / $50K = 1.32

Forecasting Damage at a BuildingLosses Avoided

ROI = 1.32

CMIP5• 112 GCM/Scenarios• Expanded to 1000 using

statistical weather generation

ROI

1.5

1.4

1.3

1.2

1.1

1.0

0.9

0.8

0.7

0.6

City Simulator in Action

Prioritizing Hurricane Response in Fayetteville

Coastal Simulators introduce Sea Level Rise/Hurricanes

Boulder County, On-going Resiliency Study

Boulder County, Focus on top Disruptive Culverts/Bridges

Using City Simulator Concepts in FEMA’s Mitigation Decision Support System

ROI = 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝐴𝐴𝐴𝐴𝐿𝐿𝐴𝐴𝐴𝐴𝐿𝐿𝐴𝐴𝐶𝐶𝐿𝐿𝐿𝐿𝐶𝐶

54321

Elev

atio

n (ft

)

Key Note

FinalistUK EA

International Excellence

Award

Questions?