WEBVTT
Kind: captions
Language: en-US

00:00:02.400 --> 00:00:05.200
Hello, everyone.
My presentation focuses on

00:00:05.200 --> 00:00:09.760
a recently published study –
the Analysis of Communities at Risk

00:00:09.760 --> 00:00:12.880
in the HayWired Scenario –
which is part of Volume 3,

00:00:12.880 --> 00:00:16.160
The Societal Consequences of the
HayWired Earthquake Scenario

00:00:16.160 --> 00:00:19.176
and now available
on the USGS website.

00:00:19.200 --> 00:00:22.400
I want to take this moment to
acknowledge my amazing co-authors,

00:00:22.400 --> 00:00:26.856
Jamie Jones, Anne Wein, and
Jeff Peters, all with the USGS.

00:00:26.880 --> 00:00:31.120
Understanding disparate hazard and risk
is the main purpose and interest of our

00:00:31.120 --> 00:00:36.057
study, and I’m excited to be part of this
important panel and to share our work.

00:00:37.200 --> 00:00:40.800
Having studied or helped manage
community recovery following

00:00:40.800 --> 00:00:44.400
many major disasters, I was interested
in understanding what the risk

00:00:44.400 --> 00:00:48.160
of large-scale population displacement
and other long-term community

00:00:48.160 --> 00:00:52.000
recovery challenges might be
in the San Francisco Bay region.

00:00:52.000 --> 00:00:56.320
Three disasters in particular are the
subject of case studies in our analysis –

00:00:56.320 --> 00:01:00.000
the 1994 Northridge earthquake that
caused extensive damage to older

00:01:00.000 --> 00:01:04.160
multi-family housing and led to
a major federal local multi-year,

00:01:04.160 --> 00:01:07.680
multi-million-dollar effort to
address the resulting blight and repair

00:01:07.680 --> 00:01:11.844
damaged buildings in 17 ghost town
neighborhoods of Los Angeles.

00:01:11.844 --> 00:01:15.255
Flooding from Hurricane Katrina
that displaced 90% of New Orleans’

00:01:15.286 --> 00:01:19.920
455,000 residents, with severely
damaged areas having profoundly

00:01:19.920 --> 00:01:23.600
lower rates of population
return than less-damaged areas.

00:01:23.600 --> 00:01:27.440
Socially and economically vulnerable
populations also had lower return rates

00:01:27.440 --> 00:01:32.000
than other cohorts. By 2015 and
10 years on, the city had recovered

00:01:32.000 --> 00:01:36.160
about 80% of its pre-Katrina
population, but nearly half of its

00:01:36.160 --> 00:01:42.136
72 neighborhoods still had less than
75% of their pre-Katrina population.

00:01:42.160 --> 00:01:47.120
The 2010-2011 Canterbury earthquake
sequence in New Zealand resulted in

00:01:47.120 --> 00:01:51.440
as many as 10 distinct episodes of
liquefaction, rockfalls, and other

00:01:51.440 --> 00:01:54.960
ground failure hazards and a
government buy-out offer to

00:01:54.960 --> 00:01:58.240
more than 7,000 properties in
red zone areas that caused

00:01:58.240 --> 00:02:02.400
regional housing shortages and cost
inflation as well as homelessness.

00:02:02.400 --> 00:02:08.320
Replanning of the red zone areas
continues today, over 10 years later.

00:02:08.320 --> 00:02:12.480
Specifically, our work looked at
three things – where concentrated

00:02:12.480 --> 00:02:15.280
damage would occur from
a magnitude 7 earthquake

00:02:15.280 --> 00:02:17.760
on the Hayward Fault centered
in southeast Oakland.

00:02:17.760 --> 00:02:20.960
To do this, we first developed a
methodology to combine the Hazus

00:02:20.960 --> 00:02:24.320
estimates of building damage resulting
from earthquake shaking, landslide,

00:02:24.320 --> 00:02:28.240
and liquefaction by Seligson
and others with the fire following

00:02:28.240 --> 00:02:31.360
earthquake damage calculated
by Scawthorn and others.

00:02:31.360 --> 00:02:35.760
We then used a threshold of 20% of
total building square footage in

00:02:35.760 --> 00:02:39.920
an extensive or complete damage
state and thus deemed uninhabitable

00:02:39.920 --> 00:02:44.640
until necessary repairs or rebuilding
are completed to identify census tracts

00:02:44.640 --> 00:02:47.656
with areas of
concentrated damage.

00:02:47.680 --> 00:02:51.840
Who – we then applied a number of
methods to consider the potential size

00:02:51.840 --> 00:02:54.640
and composition of the region’s
population that could be displaced

00:02:54.640 --> 00:02:58.080
as a result of key physical impacts
of the HayWired scenario.

00:02:58.080 --> 00:03:01.520
We considered a series of
socioeconomic vulnerabilities

00:03:01.520 --> 00:03:04.880
as well as federal, state, and local
emergency response policies and

00:03:04.880 --> 00:03:09.360
plans related to emergency shelter and
temporary housing that may impact

00:03:09.360 --> 00:03:13.840
displacement and population return.
And finally what – we then explored

00:03:13.840 --> 00:03:17.520
three key issues that our case studies
and other past disasters have shown to

00:03:17.520 --> 00:03:21.760
impede long-term community recovery,
assembling funding for recovery,

00:03:21.760 --> 00:03:25.200
repairing and replacing damaged
housing, and areas requiring

00:03:25.200 --> 00:03:28.080
substantial governmental
intervention and replanning.

00:03:28.080 --> 00:03:31.840
Now I will share
a bit more on each.

00:03:31.840 --> 00:03:34.800
Combining the Hazus estimates of
building damage caused by earthquake

00:03:34.800 --> 00:03:37.440
ground shaking, landslides,
and liquefaction with the fire

00:03:37.440 --> 00:03:41.656
following building damage, results in
nearly 1 million residential buildings,

00:03:41.680 --> 00:03:45.920
1.4 million housing units, and nearly
40,000 non-residential buildings

00:03:45.920 --> 00:03:48.640
sustaining damage
in the HayWired scenario.

00:03:48.640 --> 00:03:51.360
That means that almost
a third of the Bay Area’s

00:03:51.360 --> 00:03:55.680
housing stock would be damaged.
Of this, 100,000 residential buildings

00:03:55.680 --> 00:04:00.456
sustained extensive or complete damage
and thus are potentially uninhabitable.

00:04:00.480 --> 00:04:04.240
Over half of the property losses for
building content and replacement cost

00:04:04.240 --> 00:04:08.240
are concentrated in Alameda County,
as you can see in the graph on the right.

00:04:08.240 --> 00:04:12.936
Contra Costa and Santa Clara
Counties also sustain heavy losses.

00:04:12.960 --> 00:04:16.640
Areas of concentrated damage
cover only 8% of all census tracts

00:04:16.640 --> 00:04:20.080
in the nine-county region but
contain nearly 50% of all the housing

00:04:20.080 --> 00:04:23.976
that is likely to be uninhabitable
or completely destroyed.

00:04:24.000 --> 00:04:27.280
These areas also contain
about 600,000 employees.

00:04:27.280 --> 00:04:30.960
This map shows the footprint of these
areas of concentrated damage that

00:04:30.960 --> 00:04:34.560
are located within seven economic
sub-areas used in the HayWired

00:04:34.560 --> 00:04:38.960
economic analyses. These sub-areas
are located in Alameda, Contra Costa,

00:04:38.960 --> 00:04:42.960
and Marin Counties, and each has
more than 25% of the total building

00:04:42.960 --> 00:04:47.673
square footage in an extensive
or complete damage state.

00:04:48.800 --> 00:04:52.320
Field work conducted in areas of
concentrated damage found that most

00:04:52.320 --> 00:04:56.160
are densely developed neighborhoods
with older seismically vulnerable

00:04:56.160 --> 00:04:59.680
single-family and multi-family
wood-frame buildings that sustain

00:04:59.680 --> 00:05:03.200
higher levels of shaking damage
and are also quite vulnerable to

00:05:03.200 --> 00:05:06.240
post-earthquake fires.
These maps show the areas of

00:05:06.240 --> 00:05:12.080
concentrated damage for all
occupancies combined on the left and

00:05:12.080 --> 00:05:16.640
multi-family occupancies on the right.
We analyzed eight key occupancy

00:05:16.640 --> 00:05:20.320
groupings separately.
All of the footprints show similar

00:05:20.320 --> 00:05:23.600
patterns of concentrated building
damage along the entire East Bay

00:05:23.600 --> 00:05:27.360
corridor in the Santa Clara,
San Ramon, and Livermore Valleys,

00:05:27.360 --> 00:05:31.016
and in the flatlands surrounding
San Francisco and San Pablo Bays.

00:05:31.040 --> 00:05:34.480
They also show the expanded and
intensified levels of concentrated

00:05:34.480 --> 00:05:39.826
damage that result from fire
following earthquake hazards.

00:05:42.400 --> 00:05:44.720
In the next stage of the analysis,
we looked at a number of

00:05:44.720 --> 00:05:48.560
socioeconomic factors affecting
immediate post-disaster displacement

00:05:48.560 --> 00:05:51.440
as well as subsequent
waves of displacement.

00:05:51.440 --> 00:05:56.240
The table shows a range of estimates
that we made of potential populations

00:05:56.240 --> 00:05:59.360
at risk of displacement from the
magnitude 7 main shock of the

00:05:59.360 --> 00:06:03.600
Hayward scenario earthquake.
The 152,000 displaced households

00:06:03.600 --> 00:06:07.520
used in the April 2018 release of the
Hayward scenario was based upon

00:06:07.520 --> 00:06:11.040
the Hazus analysis of shaking,
landslide, and liquefaction and

00:06:11.040 --> 00:06:14.720
roughly equals about 6% of the
nine-county region’s households

00:06:14.720 --> 00:06:18.640
and population. Integrating in
post-earthquake fires and lifeline

00:06:18.640 --> 00:06:23.200
outages yields potential displacement
estimates of up to 10% of the region’s

00:06:23.200 --> 00:06:27.200
households and population.
Up to 20% of the region’s households

00:06:27.200 --> 00:06:31.280
and population live within the
areas of concentrated damage.

00:06:31.280 --> 00:06:35.040
While it is not likely that all households
or people in areas of concentrated

00:06:35.040 --> 00:06:38.240
damage would be displaced,
neighborhoods serving infrastructure

00:06:38.240 --> 00:06:42.640
and services and – the disruption
of those and heavy levels of

00:06:42.640 --> 00:06:45.840
surrounding building damage
could lead to higher rates of

00:06:45.840 --> 00:06:50.141
displacement in these areas
than in less damaged areas.

00:06:50.720 --> 00:06:55.760
We used a set of 10 community
vulnerability indicators, or CVIs,

00:06:55.760 --> 00:06:59.200
developed by ABAG and the
Bay Conservation and Development

00:06:59.200 --> 00:07:03.120
Commission to evaluate the social
and economic vulnerabilities of

00:07:03.120 --> 00:07:05.840
residents in areas of
concentrated damages for

00:07:05.840 --> 00:07:11.120
all building occupancies combined.
In these areas, around 350,000 people

00:07:11.120 --> 00:07:14.960
reside in block groups
with CVI scores of 5 or more.

00:07:14.960 --> 00:07:18.936
Of those, most reside in Alameda
and Contra Costa Counties.

00:07:18.960 --> 00:07:22.720
These graphs also show how the
prevalence of different vulnerability

00:07:22.720 --> 00:07:27.840
indicators vary by CVI score.
For example, in areas with CVI scores

00:07:27.840 --> 00:07:30.560
of 5 or more, there is a greater
prevalence of housing

00:07:30.560 --> 00:07:35.096
and income-related indicators
and literacy indicators.

00:07:35.120 --> 00:07:39.120
We supplemented the CVIs with
population analyses of other

00:07:39.120 --> 00:07:43.280
socioeconomic factors causing
displacement – school-aged children,

00:07:43.280 --> 00:07:46.880
households with disabilities,
and homeless populations.

00:07:46.880 --> 00:07:50.216
Those estimates are
shown in the top right.

00:07:50.240 --> 00:07:54.000
We also created an indicator set to
account for the potential voluntary

00:07:54.000 --> 00:07:58.080
displacement of young mobile
populations, which we know

00:07:58.080 --> 00:08:01.120
contributed to high vacancies and
ghost town neighborhoods following

00:08:01.120 --> 00:08:04.240
the Northridge earthquake.
Young mobile populations have

00:08:04.240 --> 00:08:09.364
also been leaving the region
now as a result of the pandemic.

00:08:10.240 --> 00:08:13.520
Jones and others’ study of lifeline
infrastructure systems and

00:08:13.520 --> 00:08:17.200
interdependencies for the HayWired
scenario show that damage levels and

00:08:17.200 --> 00:08:21.200
downtime potential for some systems
could affect both immediate population

00:08:21.200 --> 00:08:25.200
evacuations, especially in areas
with post-earthquake fires,

00:08:25.200 --> 00:08:27.496
as well as
long-term displacements.

00:08:27.520 --> 00:08:30.960
Utility restoration curves for
Alameda County show long delays

00:08:30.960 --> 00:08:34.800
in water and BART services.
Jones’ mapping of the co-location

00:08:34.800 --> 00:08:39.040
of three or more lifeline assets and
the exposure to multiple hazards

00:08:39.040 --> 00:08:42.560
shows a strong correlation with areas
of concentrated building damage

00:08:42.560 --> 00:08:47.707
in parts of western Contra Costa
County and much of Alameda County.

00:08:48.480 --> 00:08:51.840
Regional earthquake response plans
do not consider the possibility

00:08:51.840 --> 00:08:56.560
of widespread fires and extensive
lifeline utility disruptions,

00:08:56.560 --> 00:09:01.120
and their estimates are for far
fewer individuals requiring shelter

00:09:01.120 --> 00:09:05.280
after a Bay Area earthquake
than some of ours.

00:09:05.280 --> 00:09:08.320
Available shelters in the central
nine-county region may quickly

00:09:08.320 --> 00:09:11.200
reach capacity, especially in
heavily impacted areas

00:09:11.200 --> 00:09:15.200
of Alameda and Contra Costa County.
Regional response plans identify

00:09:15.200 --> 00:09:18.240
potential host communities in
surrounding communities,

00:09:18.240 --> 00:09:21.920
but if the immediate sheltering
needs exceed the 16-county regional

00:09:21.920 --> 00:09:25.920
sheltering capacity, plans then call for
relocating residents to other parts of

00:09:25.920 --> 00:09:30.240
California and adjoining states.
An already constrained regional

00:09:30.240 --> 00:09:33.680
housing market, combined with
the limited supply of interim housing

00:09:33.680 --> 00:09:37.520
options, may impact the timing
and ability of people to return to

00:09:37.520 --> 00:09:41.360
their home communities. Renters,
particularly low-income renters

00:09:41.360 --> 00:09:43.360
and people with access
and functional needs,

00:09:43.360 --> 00:09:47.574
may have more difficulty
finding alternative housing.

00:09:48.560 --> 00:09:51.520
It will be difficult for residents,
businesses, governmental,

00:09:51.520 --> 00:09:54.960
and non-governmental agencies
to assemble the substantial funds

00:09:54.960 --> 00:09:58.560
necessary for repairs and
rebuilding after such a scenario.

00:09:58.560 --> 00:10:02.320
Hazus and Corelogic building-related
loss estimates demonstrate the

00:10:02.320 --> 00:10:06.400
considerable scale of losses
estimated for the Hayward scenario.

00:10:06.400 --> 00:10:10.400
Corelogic’s 2018 study estimates
that insured losses would cover

00:10:10.400 --> 00:10:15.120
about 23% of commercial losses
and 12% of residential losses

00:10:15.120 --> 00:10:19.096
for the HayWired earthquake
sequence and fire.

00:10:19.120 --> 00:10:22.880
Such a major earthquake would also
certainly trigger state and federal

00:10:22.880 --> 00:10:27.040
disaster assistance programs that
would help provide valuable resources,

00:10:27.040 --> 00:10:31.040
however, studies from past disasters
have shown that the accessibility

00:10:31.040 --> 00:10:34.080
and distribution of formal disaster
assistance, especially from

00:10:34.080 --> 00:10:38.320
governmental sources, is not equitable.
As we’ve seen, some of the hardest-hit

00:10:38.320 --> 00:10:42.720
communities also have high levels of
socioeconomic vulnerability and thus

00:10:42.720 --> 00:10:46.376
may have difficulty obtaining
formal disaster assistance.

00:10:46.400 --> 00:10:50.400
Of the nearly 1 million residential
buildings and 1.4 million housing units

00:10:50.400 --> 00:10:53.600
that are damaged, those with
fire damage are more likely to

00:10:53.600 --> 00:10:57.360
have insurance coverage.
Otherwise, federal disaster loans

00:10:57.360 --> 00:11:01.360
and grants, conventional loans,
individual savings, or assistance from

00:11:01.360 --> 00:11:06.216
nonprofit funders would be needed.
Several large government housing

00:11:06.240 --> 00:11:11.440
repair programs have been created to
address financing gaps after Hurricanes

00:11:11.440 --> 00:11:17.781
Katrina and Sandy in the U.S., to name
a few, and might be needed here as well.

00:11:18.720 --> 00:11:22.400
A major earthquake could damage some
communities so extensively that market

00:11:22.400 --> 00:11:26.800
forces alone will not facilitate recovery.
Determining which areas can

00:11:26.800 --> 00:11:30.400
successfully recover according
to existing plans and policies,

00:11:30.400 --> 00:11:34.400
and which need to be re-planned,
is a difficult task that largely falls to

00:11:34.400 --> 00:11:38.960
local governments to address.
This analysis defines a typology

00:11:38.960 --> 00:11:41.920
of potential areas that may
require governmental interventions

00:11:41.920 --> 00:11:44.936
and re-planning for
earthquake disasters.

00:11:44.960 --> 00:11:48.080
These images are close-up views
of the region-wide maps

00:11:48.080 --> 00:11:52.320
that we created for each type.
Substantial land and building damage

00:11:52.320 --> 00:11:56.320
caused by liquefaction or landslides,
potential surface fault rupture,

00:11:56.320 --> 00:11:59.680
including afterslip, concentrated
building damage caused by

00:11:59.680 --> 00:12:03.280
ground shaking and fire, and areas
with substantial transportation

00:12:03.280 --> 00:12:06.880
and infrastructure damage –
all these areas could face additional

00:12:06.880 --> 00:12:11.280
damage from aftershocks and
an increased vulnerability to other

00:12:11.280 --> 00:12:15.520
hazards like flooding and slope failures.
Uncertainty about the feasibility

00:12:15.520 --> 00:12:19.200
and affordability of repairs could
lead to mortgage defaults,

00:12:19.200 --> 00:12:24.776
occupant displacement, property blight,
and declining property values.

00:12:24.800 --> 00:12:28.080
An array of governmental interventions
and re-planning actions may be

00:12:28.080 --> 00:12:32.240
necessary to address these conditions,
all of which would require money,

00:12:32.240 --> 00:12:35.840
community engagement, and time.
These demands will likely come

00:12:35.840 --> 00:12:39.440
at a point when populations are
still displaced and local governments

00:12:39.440 --> 00:12:43.760
are facing reduced tax revenues
and repair costs for publicly owned

00:12:43.760 --> 00:12:46.575
facilities and
infrastructure.

00:12:47.360 --> 00:12:50.880
A 2018 analysis by Tom Brocher
and others estimates that,

00:12:50.880 --> 00:12:55.760
since the 1989 Loma Prieta earthquake,
as much as $80 billion of investment

00:12:55.760 --> 00:12:59.120
across the region has been directed to
critical infrastructure upgrades

00:12:59.120 --> 00:13:02.960
as well as extensive retrofitting of
seismically vulnerable residential,

00:13:02.960 --> 00:13:06.800
commercial, and public buildings.
Nonetheless, as this analysis

00:13:06.800 --> 00:13:10.080
has revealed, substantial
vulnerabilities remain.

00:13:10.080 --> 00:13:12.800
Furthermore, some of these
vulnerabilities are only likely to

00:13:12.800 --> 00:13:16.560
continue to grow unless there are
some fundamental shifts in the scale

00:13:16.560 --> 00:13:20.720
of attention and investment given to
reducing the region’s seismic risk.

00:13:20.720 --> 00:13:23.840
Seven policy implications for
improving community seismic

00:13:23.840 --> 00:13:27.360
resilience in the San Francisco Bay
region stem directly from insights

00:13:27.360 --> 00:13:32.160
gained through our analysis.
The first three policy topics direct

00:13:32.160 --> 00:13:35.840
more investment and focus in local
and regional land use, development,

00:13:35.840 --> 00:13:39.680
and capital planning to upgrade
the seismic resilience of existing

00:13:39.680 --> 00:13:43.520
communities, particularly older homes
and infrastructure, while future,

00:13:43.520 --> 00:13:46.776
more resilient communities
are being planned and built.

00:13:46.800 --> 00:13:50.720
Four additional policy topics place
greater emphasis on the risk of

00:13:50.720 --> 00:13:54.960
population displacement in government,
individual, and business disaster plans,

00:13:54.960 --> 00:14:00.000
exercises, and preparedness campaigns
and planning to manage long-term

00:14:00.000 --> 00:14:06.936
recovery and recovery financing ahead
of time at all levels of government.

00:14:06.960 --> 00:14:10.240
The Communities at Risk analysis
for the HayWired scenario adds

00:14:10.240 --> 00:14:14.080
a social dimension and long-term
perspective to the rich details of the

00:14:14.080 --> 00:14:19.096
hazard and immediate impacts covered
in Volume 1 and 2 of the scenario.

00:14:19.120 --> 00:14:22.560
It shows communities at risk are
the places where disparate hazards,

00:14:22.560 --> 00:14:27.120
vulnerabilities, and risks converge,
not only immediately after

00:14:27.120 --> 00:14:31.120
a major disaster, but over time,
as disaster effects continue to

00:14:31.120 --> 00:14:35.040
compound and unfold.
While the HayWired scenario is

00:14:35.040 --> 00:14:39.440
just one of many potential disaster
scenarios the San Francisco Bay region

00:14:39.440 --> 00:14:43.600
could face, it reveals the places where
normal community functions are likely

00:14:43.600 --> 00:14:47.760
to be severely impaired or cease to
exist for months, even years,

00:14:47.760 --> 00:14:51.760
after a major East Bay earthquake
and where residents may be forcibly

00:14:51.760 --> 00:14:55.600
or voluntarily displaced because of
direct damage to their homes

00:14:55.600 --> 00:14:58.960
and neighborhoods.
This methodology can be applied

00:14:58.960 --> 00:15:04.056
to other scenarios to identify other
patterns of disparate hazard and risk.

00:15:04.080 --> 00:15:08.000
These are the places that need our
priority attention for hazard mitigation

00:15:08.000 --> 00:15:11.440
and community resilience-building
efforts and investments before

00:15:11.440 --> 00:15:15.600
a major disaster strikes.
Thank you.

00:15:17.632 --> 00:15:22.727
[silence]