1
LAWRENCE
NATIONAL
LABORATORY
LIVERMORE California Levee Risk, Now and
in the Future: Identifying
Research and Tool
Development Needs
Robin Newmark, Michael Hanemann and
Daniel Farber
November 28, 2006
UCRL- TR- 226504
2
Disclaimer
This document was prepared as an account of work sponsored by an agency of the United States
Government. Neither the United States Government nor the University of California nor any of
their employees, makes any warranty, express or implied, or assumes any legal liability or
responsibility for the accuracy, completeness, or usefulness of any information, apparatus,
product, or process disclosed, or represents that its use would not infringe privately owned rights.
Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States Government or the University of California. The
views and opinions of authors expressed herein do not necessarily state or reflect those of the
United States Government or the University of California, and shall not be used for advertising or
product endorsement purposes.
Auspices Statement
This work was performed under the auspices of the U. S. Department of Energy by University of
California, Lawrence Livermore National Laboratory under Contract W- 7405- Eng- 48.
3
California Levee Risk, Now and in the Future:
Identifying Research and Tool Development Needs
September 28- 29, 2006
University of California Center Sacramento
Sacramento, CA
Workshop Report
Robin L. Newmark, Michael Hanemann, Daniel Farber, editors
LLNL UCRL- TR- 226504
4
Table of Contents
Executive Summary ……………………………………………………………… 5
Workshop Description …………………………………………………………… 7
Participants ………………………………………………………………………. 9
Workshop Agenda ……………………………………………………………….. 12
Summaries of Presentations ……………………………………………………… 14
Invited talks ………………………………………………………………….. 14
Research Summaries ………………………………………………………… 24
Discussion Groups ………………………………………………………………. 26
General Discussion ……………………………………………………………… 35
Appendix I: Group Discussion Outbrief Presentations …………………………. 39
Appendix II: Press Release ……………………………………………………… 51
5
California Levee Risk, Now and in the Future:
Identifying Research and Tool Development Needs
Executive Summary
California depends on a complex system of engineering structures – dams, aqueducts, and
levees – for both its water supply and flood protection. The Sacramento- San Joaquin
Delta system is the hub for California’s water supply as well, providing water for twenty-three
million Californians and three million acres of agricultural land, and sustaining a
$ 400 billion economy; it is also a unique environmental asset. Because of an aging and
deteriorating levee system, the city of Sacramento itself faces a greater risk of flooding
than any other major city in the United States, including New Orleans. In addition,
substantial seismic risks in Northern California threaten both the water supply
infrastructure in the Delta and the levees that protect valuable agricultural and,
increasingly, urban property throughout the Central Valley. Simulations show that a
large- scale levee failure would pull salt water into the Delta, requiring a shut- down of
water exports to southern California for one to three years ( Benjamin Associates, 2005).
The Jones tract levee failure in 2004, combined with the impact of Hurricane Katrina on
New Orleans, have raised public consciousness about this flood risk. California voters
have recently approved massive bonds to address the threat. Yet, much of the critical
scientific information needed to design a solution does not yet exist. Without this
information, we have no way of knowing whether the billions of dollars to be spent on
the levees will really fix the problem.
To begin to address these urgent informational needs, three University of California
Research Centers – the Center for Catastrophic Risk Management, the Center for
Environmental Law and Policy, and Lawrence Livermore National Laboratory –
sponsored a two- day workshop. Roughly sixty research scientists, engineers,
policymakers and agency representatives gathered at the University of California Center
in Sacramento to map out a research agenda for flood control issues in the Delta.
This report distills the discussion at the workshop. Workshop participants identified a
broad range of research needs. The following four issues were identified as particularly
critical:
1. Climate Change. Extreme flood episodes escalate quite sharply along with sea level
change, which is expected as part of global climate change. High tides and large winter
storms, combined with sea level change, pose grave risks. Improved hydrological
modeling of the Delta system, along with flood characteristic modeling and probabilistic
climate models, are needed. At present there is no comprehensive hydrological model
6
suitable for assessing the overall flooding and water supply risks in California’s water
system.
2. Seismic Risks. Modeling shows that soil conditions in the Delta will magnify
earthquake ground movements. A major quake on the Hayward fault would cause as
much shaking in the Delta as near the fault itself. Faulting underneath the Delta is poorly
understood. We need probabilistic seismic hazard prediction and scenario prediction to
model these risks.
3. Current conditions. We do not have an accurate inventory of levee conditions in the
Delta, nor do we have accurate subsidence data. ( Peat soils inside the levees tend to
subside, as in New Orleans, greatly increasing potential flood risks.) We need a
complete data base of levee inspection reports, along with comprehensive information
about the ownership and control of specific levees ( which are often in private hands), the
terrain protected by the levee, and the potential economic impact of breach or
overtopping in specific locations. We also need improved information about channel
geometry and bathymetry; flow rates, water densities, and improved computer modeling.
On- going Lidar coverage of Delta levees is needed.
4. Dynamic change. Delta conditions are expected to change because of climate change,
the impact of environmental policies, land use changes that strongly influence flood risk,
and projected weather changes. All of these matters require careful attention. In
particular, we need to have better methods of evaluating the environmental, social, and
economic harms associated with Delta risks. We also need improved decision- making
techniques for dealing with infrastructure planning under conditions of dynamic change.
Decisions made today about land use or infrastructure repair could greatly limit future
options for the Delta, making it hard to react to developing scientific knowledge of the
risks.
The goal is not merely to develop improved scientific knowledge for its own sake, but to
deliver usable and timely information to the officials who are charged with making
critical decisions about the Delta. Fulfilling this goal requires the development of a new
institutional structure in which policymakers and scientists can interact, so as to ensure
that the scientists are asking the right questions and the policymakers are getting the most
reliable and objective research findings. The Workshop participants strongly suggested
that a consortium approach is needed to engage, provide products to and get feedback
from policy and decision makers. What is required is an umbrella that supports the broad
set of technical and policy- relevant disciplines that need to be applied. We strongly urge
the creation of such an institutional structure by the State of California.
References
Jack R. Benjamin & Associates, Preliminary Seismic Risk Analysis Associated with
Levee Failures in the Sacramento- San Joaquin Delta. June 2005.
7
8
California Levee Risk, Now and in the Future:
Identifying Research and Tool Development Needs
Workshop Description
Overview: The Center for Catastrophic Risk Management ( CCRM) and the California
Center for Environmental Law & Policy ( CCELP) at UC Berkeley and the Lawrence
Livermore National Laboratory ( LLNL) joined together to cosponsor a workshop to
define research requirements to mitigate the hazards facing the Sacramento- San Joaquin
Delta Levee system. The Workshop was intended to provide a forum to
• Report assessments of current vulnerabilities facing the levees, such as structural
failure, seismic loading, flooding, terrorism;
• Consider longer term challenges such as climate change, sea level rise; and
• Define research requirements to fill gaps in knowledge and reduce uncertainties in
hazard assessments.
Background: The Sacramento- San Joaquin Delta Levee system has received
considerable attention since the Jones Tract levee failure in June 2004, the Hurricane
Katrina disaster in New Orleans and the centenary of the 1906 San Francisco earthquake.
The fragility of the levees in the Sacramento- San Joaquin Delta, and the threat of a
tremor on the faults running nearby, have long been a concern for water managers in
California. The Delta region faces societal and economic pressures related to land- use
issues for agriculture, water resources, navigation and urbanization. Addressing levee
issues requires dealing with uncertainty regarding how risks will evolve in the coming
decades due to expected changes: natural ones such as climate change, then social ones
such as population growth and urbanization.
It is timely to assess the current status of Delta Levee hazards and define the research
requirements to reduce uncertainties to move toward mitigation of these hazards. The
Katrina disaster has demonstrated the importance of the levee systems. Substantial
resources will be mobilized in the coming years to address aspects of the problem.
Certainly California is poised to undertake an extensive effort in analysis and
implementation. Hopefully, some of that effort will support research and technology
development.
A key workshop objective is to begin to describe the R& D required to provide specific
solutions or new approaches in a few years’ time to address key issues, along with
justification for prioritization. The intent is to identify the work that needs to be done
over the next few years to provide the solutions for the next round of infrastructure
9
investments, for input to the policy makers and those responsible for the levee
infrastructure. This includes research priorities for problem definition, prediction,
management tools, policy approaches, technology development and data collection. The
workshop report should ideally lay out a strategic plan for research and development.
Many disasters we experience result more from organizational or social/ policy problems
than purely technical ones. For example, the Shuttle disaster was not due to an O- ring
problem so much as the socio- organizational structure that prevented appropriate decision
making to take place. However, the problem is much more than " structure" of the
organization - and the failure of the New Orleans flood defense system clearly illustrated
this. It is due, in part, to a lack of understanding of how very complex systems work,
considering that the decision- making process is an integral part of the system as well as
the physical part of the levees. It is also due to problems in accounting for non-quantifiable
effects, such as a certain perception of the risk, and the effects of operating
within a certain mind- set. Although this workshop targeted technical issues to improve on
the quality of technical information provided to the decision- makers, the need for
revisiting issues of decision- making as an integral part of the system was also recognized.
Participants were a representative sample of experts, researchers and practitioners
addressing multiple dimensions of the complex Delta and levee system. Their
responsibilities range from defining the required functional system needs for the future of
the State and the nation; to evaluating the integrity of the present system; to making
recommendations for modifications and upgrades of the system ( including engineering
solutions, analyses and tools); to implementing those recommendations. A list of
participants is included. The environmental perspective was somewhat underrepresented,
as the focus of this effort was more on the structural issues. This was not to underplay
the importance of environmental issues; in fact, they may be the dominant driver in the
decision- making process. However, we needed to start with a workable scope,
recognizing that environmental issues might require an expanded discussion at a later
date.
Details: The Workshop was conducted over a day and a half, consisting of talks, posters
and breakout discussions. The agenda is included here. Participants were encouraged to
bring a poster and/ or a 1- viewgraph summary of their work as it relates to levee issues.
The final general discussion addressed cross- cutting issues, interfaces and broader
recommendations than those developed in the individual discussion sessions.
10
Workshop Participants
James Agnew California Department of Water Resources
Rodger Aines Lawrence Livermore National Laboratory
Carol Baker Office of Assembly Speaker Fabian Nunez
Joshua Bernardo Solano County Department of Resource Management
Cheryl Bly- Chester Rosewood Environmental Engineering
John Boatwright United States Geological Survey
Scott Brandenberg University of California, Los Angeles
Alt. Brandt California State Assembly
Thomas Brocher United States Geological Survey
Robert Budnitz Lawrence Livermore National Laboratory
Nicholas Burton Solano County Department of Resource Management
Dan Cayan Scripps Institution of Oceanography, and United States
Geological Survey
Michael Dettinger United States Geological Survey
Larry Dale Lawrence Berkeley National Laboratory
Yun Duan Lawrence Livermore National Laboratory
Tim Duane University of California, Berkeley
Stephen Durrett United States Army Corps of Engineers
Daniel Farber University of California, Berkeley
John Fletcher United States Geological Survey
Guido Franco California Energy Commission
Richard Frank University of California, Berkeley
11
Justin Fredrickson California Farm Bureau Federation
Catherine Freeman California State Legislative Analyst’s Office
Ceci Giacoma Delta Fire Protection District
Michael Hanemann University of California, Berkeley
Rodger Henderson United States Army Corps of Engineers
Thomas Holzer United States Geological Survey
Nina Kapoor University of California Center, Sacramento
Tadahiro Kishida University of California, Davis
Jane Long Lawrence Livermore National Laboratory
Ladd Lougee CALFED Science Program
Cathie Magowan University of California, Office of the President
Dick McCarthy California Seismic Safety Commission
Norm Miller University of California, Berkeley and Lawrence Berkeley
National Laboratory
Toby Minear University of California, Berkeley
Jeff Mount University of California Davis
David Mraz California Department of Water Resources
Robin Newmark Lawrence Livermore National Laboratory
Ron Ott CALFED Bay- Delta Program
Elizabeth Patterson California Department of Water Resources
Jessica Pearson California Department of Water Resources
Henry Reyes California Seismic Safety Commission
Arthur Rodgers Lawrence Livermore National Laboratory
Doug Rotman Lawrence Livermore National Laboratory
12
Badie Rowshandel California Geological Survey
John Rundle University of California Davis
Said Salah- Mars URS Corporation
Jean Savy Lawrence Livermore National Laboratory
Curtis Schmutte California Department of Water Resources
David P. Schwartz United States Geological Survey
Ray Seed University of California, Berkeley
Larry Smith United States Geological Survey
Rune Storesund University of California, Berkeley
Ralph Svetich California Department of Water Resources
Ken Trott California Department of Food and Agriculture
Fred Turner California Seismic Safety Commission
Matt Vander Sluis Planning 7 Conservation League
Geoffrey Wandisford- smith University of California, Center, Sacramento
Frank Webb Jet Propulsion Laboratory
John Ziagos Lawrence Livermore National Laboratory
13
California Levee Risk, Now and in the Future:
Identifying Research and Tool Development Needs
September 28- 29, 2006
University of California Center Sacramento
1130 K Street, Suite LL22
Sacramento, CA 95814 ( 916)- 445- 5100
AGENDA
Thursday September 28
8: 00 am Morning Hospitality, Poster set- up
9: 00 Welcome/ Logistics UCCS host
9: 15 Opening Remarks & Introduction to Workshop Goals CCRM/ CCELP/ LLNL
9: 30 Keynote Lecture David Mraz ( DWR)
Overview of Delta Levee System
10: 10 The Role of Science in the Delta Visioning Process Jeffrey Mount ( UCD)
10: 45 Break
11: 00 Levee Engineering Issues: Katrina lessons for California Ray Seed ( UCB)
11: 35 Needs of Existing Flood Damage Reduction Infrastructure
Stephen Durrett ( USACE)
11: 55 California Flood Control Levees Response to Recent Flood Events, and
DRMS - Overview of Delta Levee Vulnerabilities Said Salah- Mars ( URS)
12: 15 pm Lunch and poster intros ALL
1: 30 Climate change and the Threat of Greater Sea Level Rise and Flooding in the Bay
Delta Daniel Cayan ( SIO)
14
2: 00 Translating Risks into Economic and Social Impacts
Michael Hanemann ( CCRM - UCB)
2: 30 Break
2: 45 Working groups: Creative thinking discussion groups
5: 00 Poster viewing
6: 30 No- host Workshop Dinner
Friday September 29
8: 00 Morning Hospitality
9: 00 Welcome/ Logistics UCCS host
9: 10 Making Sensible Choices in an Uncertain World Dan Farber ( CCELP - UCB)
9: 30 Plenary session:
Presentation of the groups’ suggestions, and discussion.
9: 30 Group 1
10: 00 Group 2
10: 30 Group 3
11: 00 Group 4
11: 30 Identify common themes
12: 00 Lunch
1: 00 General discussion
Brainstorming: system performance criteria, problem definition, constraints
Cross- cutting issues, interfaces
Approaches, new tools, gaps, research and development priorities
Summary
3: 00 Report out
Drafting of Workshop Report
4: 00 Adjourn
15
Summaries of Presentations
1. Keynote Lecture: Overview of Delta Levee System
David Mraz, Acting Branch Chief, Delta Suisun March Office, California Department of
Water Resources
The Sacramento- San Joaquin River Delta drains the flow of five major rivers and carries
47% of California’s run- off. Water for 23 million Californians, 3 million acres of
agricultural lands and $ 400 billion of the State’s economy are supplied through the Delta.
The Delta has numerous functions beyond serving as the hub of California’s water
supply, including navigation, motor and rail transportation, power transmission, natural
gas extraction, recreation and natural habitat. A key component of the Delta are the
1,100 miles of levees that hold back the flow of rivers and tidal surges enabling
residential and agricultural uses of reclaimed land, called “ islands”. These islands are
often below sea level and rely critically on the integrity of levees to prevent flooding.
These levees are vulnerable to erosion, overtopping, underseepage and animal activities.
Sea level raise from global climate change must be considered to increase the load on
levees into the future. Seismic loading from the numerous active faults in the greater Bay
Area present another expected failure mechanism. A study of impacts from a large
16
scenario earthquake indicates that numerous levee sections could fail and flood large
areas. A rush of salt water from the San Francisco Bay would exacerbate flooding.
Water export to southern California would cease until repairs enable fresh water to flow
to the Mendota and California Aqueducts. The repair of levees would require a huge
mobilization of construction and material resources. However, repairs could be expected
to take an extended period of time ( 1- 3 years) leading to the depletion of available water
resources. All activities that rely on the Delta and its water, such as industry, agriculture
and transportation, could be expected to be impacted for years and result in economic
losses.
The DWR is taking steps to mitigate threats to the Delta and its levees. The Delta Risk
Management Strategy ( DRMS) is currently defining hazards and environmental
consequences that must be considered to reduce vulnerabilities. The Delta Vision Project
will attempt to take more concrete steps to improve levee systems. The November 2006
election included 2 bond measures that would provide funding to improve and protect
water delivery systems. So the potential exists to take proactive steps to reduce the
hazards facing the Delta Levee System and improve the margin of safety for the critical
functions provided by the Delta.
2. Guiding Delta Visions: First- order Drivers of Change
Jeffrey Mount, Roy Shlemon Chair in Applied Geosciences, Department of Geology
and Director, Center for Watershed Sciences University of California Davis
17
Traditionally, CALFED member agencies have approached Delta management based on
a desire to maintain current conditions or to restore historic attributes. Until recently,
consideration of future conditions has rarely appeared within agency planning documents
in CALFED beyond acknowledgement of changes in water use demand. Yet, the scope
and pace of landscape and ecosystem change in the Delta are likely to dictate the success
and sustainability of all future management options.
The dynamic nature of the Delta forms an important constraint on planning for the future
of the Delta. Scenarios that are dependent on maintaining existing or historic conditions
are less likely to be viable than those that anticipate or are adaptable to current
trajectories of change. Based principally on CALFED science’s ( cap? What is this?)
current understanding of the nature of change in the Delta, there are six first- order drivers
of change in the Delta. These drivers are likely to significantly alter the Delta over the
near- and long- term and are independent of or unaffected by day- to- day management
activities. These drivers include: subsidence, sea level rise, regional climate change,
seismicity, exotic species and population growth/ urbanization. Each of these drivers will
have, or has had, a significant impact on the Delta at variable length and time scales. All
will require some form of management response, regardless of which vision is eventually
adopted for the Delta. All are a product of anthropogenic activities that have altered, and
will continue to alter, landforms, hydrologic conditions, and the environmental services
of the Delta.
18
In setting goals for a science agenda that addresses the needs of on- going planning
efforts, it is important to incorporate the need for high- quality modeling tools that can be
used to address the six first- order drivers of change.
3. From New Orleans and Hurricane Katrina, to California’s Levee Situation
Timely Lessons for a State at Risk
Ray Seed, Professor of Civil and Environmental Engineering, University of California,
Berkeley
Hurricane Katrina was a tragic disaster that led to the deaths of 1492 people,
displacement of over 400,000 people and $ 150 to $ 300 billion in losses. Even more
tragic is the fact that these losses could have been largely prevented with proper planning
and diligence. Findings from the NSF- funded “ Independent Levee Investigation Team
Final Report” indicate that hubris and denial lead to the tragic losses of Hurricane
Katrina. These findings are a sobering warning for the Sacramento- San Joaquin Delta
Levee System and strongly suggest that measures be taken in the present to mitigate
potential losses from levee failures in the future. Levees are unnatural structures meant
to hold back water where it would natural flow. As such government officials and
planners must expect to have to choose their battles to defend what can be defended and
be willing to concede to nature what is too costly to protect. Numerous examples from
New Orleans can be cited where either levees were either poorly designed or not built to
19
withstand the expected loading. This indicates that more diligence is required in the
planning and design stages for levee construction and that the plans are executed to the
design goals. Geologic structure must be considered when designing a levee for a
specific area and analysis of planned structures must be performed by competent
independent reviewers. Other examples were shown where the connections of levees
systems failed, jeopardizing public safety. Often these connections where at the
boundaries of different bureaucratic jurisdictions, suggesting failures of organizational
responsibilities. Finally, the lessons of Hurricane Katrina must be considered here in
northern California. Many of the observed levee failure mechanisms ( overtopping,
underseepage, rapid erosion) in New Orleans can be expected to impact the Sacramento-
San Joaquin Delta Levee System. Seismic loading presents another expected failure
mechanism. The way to prevent a Katrina- like disaster in northern California is to be
proactive and plan for expected environmental consequences. This can be accomplished
political and public will to spend the money to plan and build effective levee structures
and diligently follow these plans.
4. Needs of Existing Flood Damage Reduction Infrastructure
Stephen Durrett, P. E., Levee Safety Program Manager, HQ Engineering &
Construction CoP, U. S. Army Corps of Engineers
20
The Corps is has moved to a more risk based approach to looking at infrastructure. Corps
Dams have moved into this methodology 2 years ago and the Corps inspection of Levees
is moving to catch up. There are bills in front of Congress to establish a National Levee
Safety Program and the Corps has been funded to start the development of a National
Levee Inventory and Risk assessment methodology. The Corps is working with FEMA,
ASFPM, and NAFSMA to better communicate flood risk to communities. The goal of
the Risk assessments will be to prioritize projects for Congress and to identify areas of a
levee system that require additional study . Areas in Levee safety that need to be
improved are techniques to better monitor levees during flood events and methods for
performing risk assessments.
5. California Flood Control Levees Response to Recent Flood Events, and Delta Risk
Management Strategy ( DRMS) - Overview of Delta Levee Vulnerabilities
Said Salah- Mars, URS Corporation
California flood control levees have a history of failures or distress during regular flood
events. All failure mechanisms associated with embankment levees have been observed.
Back calculation of historic levee damage during past storm events indicates that the
California flood control levees are generally performing poorly even during storm events
associated with flood intervals of less than 100- years. The need to conduct a thorough
and comprehensive evaluation of the flood control system is urgent.
21
The current risk evaluation of the Delta and Suisun Marsh Levees is an example of
initiating a comprehensive assessment of the levee system. This effort includes the
evaluation of various stressing events, such as: seismicity, flooding, climate change,
subsidence, impact from invasive species, and their combined effects on the: levees, life
safety, environment and ecosystem, water quality, water reliability, infrastructure, the
economy, etc.
6. Climate change and the Threat of Greater Sea Level Rise and Flooding in the
Bay Delta
Daniel Cayan, Climate Research Division, Scripps Institution of Oceanography, UCSD
and Water Resources Division, US Geological Survey
During the next several decades, climate model simulations indicate that global warming
could amplify sea level rise along the California coast at rates substantially higher than
the ~ 2mm/ yr rise that has been observed during recent history. We employ a plausible
range of sea level rise, combined with predicted astronomical tides, projected weather
forcing, and El Nino related variability to explore possible water level extremes during
the next century in the San Francisco Bay region. Extreme event occurrences, relative to
current levels, escalate quite sharply as the magnitude of future sea level rise increases.
Impacts in the Delta are exacerbated during periods when high tides and large winter
22
storms coincide, producing wind waves and freshwater floods from Sierra and coastal
mountain catchments.
7. Translating ( Engineering) Risks into Economic and Social Impacts
Michael Hanemann, Center for Catastrophic Risk Management ( CCRM) and
Chancellor's Professor, Department of Agricultural & Resource Economics, University of
California, Berkeley
We argue against an “ engineering” mindset which assumes that the unknowns have all
been identified and quantified. This is not true, and one needs to allow an additional risk
factor for unquantified elements of uncertainty. Moreover, the risk of damage is likely to
be changing over time due to ( i) the geriatric aging of the flood control system, and ( ii)
ongoing economic and social change, including land use. Therefore, the estimate of risk
needs monitored and updated over time. At present, there is no mechanism to make this
happen.
Building an “ event tree” analysis of the human, as opposed to engineering, components
of risk is hard, yet it needs to be done. The event tree will depend heavily on institutional
factors because those create incentives for individuals and groups to take action, both
advance preparation ( adaptation) and response. Hence the need for a behavioral
component to the risk analysis.
23
The recent assessment of potential climate change impacts in California shows that
spatial downscaling is important because it more clearly reveals threshold effects that can
generate significant non- linearities in the economic damages. Another issue is the
significance of risk aversion, which has largely been overlooked. With perfect foresight,
the cost of adaptation can be minimized; for example, water can in theory be purchased in
the exact amount of a prospective shortfall. In reality, however, there is uncertainty
( imperfect foresight) and also risk aversion, Because of the latter, water users will
rationally want to buy protection ( in effect, insurance). With insurance, some costs are
incurred that turn out ex post not be have been needed – but they are worth it because, ex
ante, one does not know what will happen during the coming year. This additional cost
has not been factored into existing analyses of climate change impacts. For water
infrastructure these costs are likely to be especially significant.
8. Making Sensible Choices in an Uncertain World
Dan Farber, California Center for Environmental Law & Policy ( CCELP) and Sho Sato
Professor of Law, Boalt Hall School of Law, University of California Berkeley
24
Economists distinguish between risk, which involves events whose probability can be
quantified with reasonable precision, and uncertainty, which involves events with poorly
characterized probabilities. Sometimes we are aware of these gaps in knowledge, but we
also may simply have failed to conceptualize a potential hazard. The design of the New
Orleans flood control system several decades ago provides apt illustrations. For
examples, designers of the New Orleans flood control system assumed that weather
patterns changed only over a period of centuries, failing to account either for climate
change or cyclical storm patterns. Acknowledging these uncertainties has several
implications. First, we should favor strategies that are sufficiently flexible to adapt to
unexpected new information, rather than favoring the kind of brittle infrastructure now
characteristic of the Sacramento- San Joaquin Delta region. Adaptive management
techniques should also be integrated with environmental assessments. Second, we also
need to develop new decision techniques to identify robust strategies in view of plausible
hazards. Third, researchers need to give policy makers a firm understanding of known
areas of uncertainty, rather than limiting themselves to discussing hazards that can be
readily quantified. Model uncertainty should be clearly discussed. Where potential
hazards cannot be reliable quantified, researchers should aim to bracket the range of
probabilities or at least qualitatively characterize the seriousness of the hazard. Fourth,
planning should never be based solely on point estimates of hazard probabilities when
these estimates are subject to significant doubt, because doing so may result in adoption
of unduly brittle solutions.
Research Summaries
During Thursday’s lunch, several of the participants presented brief summaries of
research related to the delta and levee system:
Lynn Wilder ( LLNL) pointed out that, given the inherently spatial nature of the Delta
system, GIS can be a unifying technology for delta levee system research.
Tom Brocher ( USGS) presented a summary of the probabilities of large ( M> 6.7)
earthquakes on major Bay Area faults. He also showed shear- velocity depth profiles for
several basins in the Bay Area and pointed out that the Delta is composed of low- velocity
shallow materials that are likely to amplify ground shaking during an earthquake.
Jack Boatwright and Joe Fletcher ( USGS) presented a summary of a planned seismic
deployment for the Delta. These data will provide site response observations where no
previous data are available.
John Rundle ( UC Davis) described the California Hazard Institute, recently formed as a
UC Multicampus Research Program.
Tom Holzer ( USGS) described a USGS program to determine liquefaction potential in
the San Joaquin Delta.
25
Scott Brandenberg ( UCLA) described a destructive field testing approach and examples
of levee issues that this approach could be used to investigate.
Rune Storesund, P. E. University of California, Berkeley described measurements
made with an Erosion Function Apparatus ( EFA) in collaboration with Dr. J. L. Briaud
( TAMU), demonstrating a measure of erodibility.
J. Toby Minear ( UC Berkeley) described ground- based LiDAR for surveying levees
and erosional sites
Artie Rodgers ( LLNL) presented results of a recent study of in which earthquake
ground motion was predicted for a MW 7.0 Hayward Fault Earthquake, showing that
predicted delta ground motions are almost as high as along the Fault itself.
Larry Smith ( USGS) described a project designed to sequester carbon and
reverse land subsidence by managing freshwater march to accumulate carbon as plant
biomass – a potential new use for delta islands.
26
Discussion Groups
Discussion Groups: The participants broke into discussion groups, each focusing on one
of four broad questions with the assistance of a facilitator and a note- taker:
( 1) How to characterize the levee system infrastructure?
Including characterization of the levee system, its structural behavior, flow dynamics,
geology, etc, and the dynamics of the decision- making process for design, updates,
and evaluation.
( 2) What could disturb the levee system?
Including possible natural and environmental hazards, man- made hazards, malevolent
actions, and non- natural hazards such as socio- political elements in the organization
( 3) What can go wrong in the system?
Including all possible failure scenarios which could lead to damage, loss of
functionality, or catastrophe, decision- making break- down, and
( 4) What is at risk and what are the consequences of damaging events?
Including physical damage, economic loss, social disruption, and political
ramifications, etc.
27
Each of the discussion groups were asked to ( 1) identify existing gaps in knowledge or in
the existing technology and ( 2) determine what tool development, data collection and
monitoring strategies might be undertaken to address these four questions. Each group
prepared a presentation of the material summarizing their discussion and their
recommendations. It was acknowledged that, while many issues can be addressed by
technical options, some issues are best addressed through policy. While policy was not
the focus of this workshop, the participants were asked to keep track of those non-technical
issues and options that were raised in their discussions, especially those dealing
with societal choice, economic drivers and the role of policy. Similarly, they were asked
to note cross- cutting issues and interfaces that might cause artificial boundaries ( such as
organizational responsibilities). Finally, each group was provided with key questions and
elements to consider, partly to spark discussion.
GROUP 1: How to characterize the levee system infrastructure:
Group ( 1) Participants: Jean Savy ( Facilitator), Roger Aines ( Note- Taker), Michael
Dettinger, David Mraz, Said Salah- Mars, Jean Savy1, Larry Smith
Key questions:
What functions do the levees perform? How are these likely to change in coming
decades?
What is the state of the integrated system by which the levees are designed,
constructed and maintained?
What are the important attributes of the Delta levee design, construction and
maintenance system for which we need improved or new methods of analysis
and/ or tools?
What are the promising new methods or tools that would be appropriate to handle
these attributes?
What are the important knowledge gaps for which we need new creative thinking?
What factors or conditions must be monitored to evaluate levee performance and
integrity?
Elements to consider for answering these questions:
1) Network characterization and function of the integrated system, physical
characterization
- Functionality of the levee system, including dependent and functions,
industries
- Network characterization, complexity
- Geographic distribution of levees
- Geometry and material properties of levee sections
- Age of construction, probability of failure
- Risk factor for land protected by section
- Lifeline coupling ( transportation, water, power, gas, etc…)
28
2) Monitoring and Data Requirements
- Levee structural integrity
- Free board water level ( sea level rise)
- Saturation of levee materials
- Lifeline integrity
- Subsidence
Group 1 Discussion Summary
The levee system has physical, biological, and political/ organizational elements. No
single issue stands out as the primary characterization concern. The ability to access and
use information in a timely and effective fashion is the primary concern. The DRMS
process is focused on relatively short- term needs for improvements in the physical
system, but may not address the needs for emergency information in the event of
earthquake- caused failure of levees. The effects of climate change may go beyond simple
water level rise. Earthquake hazard is not well understood. Climate change and
ecosystem hazards are moving targets in the delta, requiring integrated analysis to
prioritize the important problems before adequate characterization can be identified.
These issues call for delineation of problems at a system level. For instance, the effects
of population change in the delta are as important an element as the condition of the
levees; both can have dramatic effects on the consequences of an earthquake. There is
also a need for a mechanism to accumulate and distribute the various social, scientific,
and engineering data associated with the delta. This should focus on making data
available for both planning and emergency response. This is specifically not a call for an
agency to take over such a role, but perhaps a joint agency working group to coordinate
data activities. The roles of various agencies in an emergency should be well known, and
pathways to obtain vital information clearly established. Several specific data needs were
identified: better standardization of levee construction methods, better real- time
monitoring particularly after storms and earthquakes, and better models and material data
bases specific to the behavior of the delta.
GROUP 2: What could disturb the levees? Damaging Hazards
Group ( 2) Participants: Jane Long ( Facilitator), Artie Rodgers ( Note- taker), Jim
Agnew, Jack Boatwright, Tom Brocher, Dan Cayan, Yun Duan, Jon Fletcher, Roger
Henderson, Tom Holzer, Tadahiro Kishida, Badie Rowshandel, Henry Reyes, John
Rundle, David Schwartz, Ralph Svetich
Note: Group 2 was composed primarily of technical experts, including 12 seismologists,
earthquake engineers, and 2 climate/ hydrology scientists.
Key questions:
What are the hazard phenomena that put the levees at risk?
Are the methods of assessment of each of these hazards appropriate for this task?
29
What are the promising methods or tools that would be appropriate to improve on
existing methods and tools?
What are the important knowledge gaps for which we need new creative thinking?
Seismic Hazard
How will the levee system behave under earthquake loading?
What is required to predict this behavior? For design/ modification guidance?
Elements to consider for answering these questions:
1) Probabilistic Seismic Hazard prediction
- Long- term characterization, for design
- Scenario prediction, for design and response preparation
- Real- time updating for response management
2) Monitoring and data Requirements
- Seismicity
- Tectonic dynamics, geodetic data, stress data
Hydrologic Hazard
Is the levee system adequate to support the multiple hydrologic demands
anticipated in the future ( flood protection, water supply, navigation, etc)?
Is current flood forecasting adequate for protecting the levee system?
What factors require monitoring?
Elements to consider for answering these questions:
1) Probabilistic River flow prediction
- Precipitation ( rainfall/ snowpack)
- Flood forecasting
2) Monitoring and data Requirements
- Flow ( volume and speed)
- Sediment properties
- Rainfall/ snowpack data for watershed( s)
- Incoming storm potential
Climate Change Hazard
Given the uncertainties in climate change projections, how can we determine how
climate change will impact the levees?
What factors require monitoring?
Elements to consider for answering these questions:
1) Probabilistic climate behavior models
- Climate change modeling
- Temperature and precipitation models
- Flood characteristics modeling
30
2) Monitoring and data Requirements
- Sea temperature
- Sea level rise
Other Hazards
- Navigation
- Commerce and recreation
- Non- natural hazards such as terrorism, malevolent actions
- Non- natural hazard such as organization and decision process flaws
Group 2 Discussion Summary
This group was tasked with defining the hazards that could disturb the Delta Levees and
lead a potential loss of function. The group focused mainly on natural hazards, although
human- caused disturbances were briefly discussed. The main natural hazards that
threaten the Delta Levees are: seismic ground motions from earthquakes, hydrologic
loading and response, climate change and subsidence. The human- caused hazards are
development and terrorism. In the following sections the discussions and conclusions on
the each of these issues are presented with attention to the questions listed above. A
detailed list of issues raised and discussed by Group 2 is included in the Outbrief.
Seismic Hazards
Earthquakes and the ground motions they cause pose a very credible threat to the Delta
Levee system. The USGS has estimated the probability of a magnitude 6.7 or greater
earthquake in the greater San Francisco Bay Area between now and 2032 to be 62%.
While the methods for seismic hazard assessment are well established, the required inputs
for assessing hazard in the Delta are poorly known. These involve the presence of and
likelihood of earthquakes beneath and adjacent to the Delta. Faults cannot be mapped
under the Delta with conventional means because of the presence of water and
development. Earthquake repeat times are typically longer than the ~ 100 years of
recorded earthquake history, making it difficult to assess the threats posed by known
faults adjacent to the Delta. Another major issue is the nature ground motion
amplification in the Delta. The sedimentary geology of the Delta is expected to amplify
seismic ground shaking, but little empirical data is available on this. Other issues are
more poorly known, such as the structural response of levee sections to ground motion
( with and without the presence of water) and the accuracy of ground motion estimates
from 1D and 3D simulations. All of these factors suffer from the general lack of detailed
geologic and geophysical information about the Delta. It was recommended that a
concerted effort be undertaken to collect data that will improve understanding of seismic
hazard in the Delta.
Hydrologic Hazards
The Delta Levee system channels water for many users. The amount of water and how it
flows through the Delta varies with season, weather and use demands. Of particular
31
concern is the response of the Delta and its levees to high precipitation and flood events.
These can damage, overtop and possibly lead to catastrophic failure of the levees.
Because most of the water flowing through the Delta originates far away the flow has a
complex dependence on upstream factors such as precipitation, snowmelt, reservoir
management and water export. No model exists for this entire water system and this was
noted as a major gap in the understanding and management of water passing through the
Delta. Such a model should also include the ability to investigate the hydrologic
consequences of levee failures, which could arise from natural ( e. g. spontaneous,
flooding, earthquake failure). Finally, any model of hydrology is only as good as the
input data, so an effort to acquire detailed hydrologic ( e. g. flow rates and dam controls),
precipitation, wind, and other data for modeling flow in the water system is also needed.
These data should be made available to a broad user community.
Climate Change Hazards
It is expected that sea level will rise as the earth’s atmosphere and hydrosphere warm.
Because much of the Delta islands are at or below sea level there will be an increased
load on levees protecting these islands. Sea level rise must be accounted for in the
hydrologic model proposed in the last section. Climate change will impact the amount
and nature of precipitation falling in California. For example precipitation that currently
falls as snow at high elevations in California’s mountains may fall as rain in the future.
This will result in hydrologic surges and the loss of water for consumption by humans
and agriculture. Finally, we need to better understand the consequences of climate
change in California for extreme weather events such as drought and storms.
Subsidence Hazards
Subsidence is a secular trend whereby compaction and erosion of the Delta soils leads to
the gradual lower of Delta islands and levees. Subsidence increases the load on levee
systems by decreasing the freeboard height of the levees and making it easier for water to
seep into the islands. This phenomenon can be monitored with geophysical methods.
However, no known continuous surveys have been or are being performed on the Delta.
There is a need for the application of proven methods to monitor subsidence.
Development Hazards
Human development for recreational, residential, commercial and agricultural uses in or
near the Delta exposes people and economic interests to the many hazards facing the
Delta. Development in turn alters the landscape in ways that can make the levees more
susceptible to failure. Impacts of ground water pumping, gas extraction and building
should be evaluated for their impacts on levee stability.
Terrorism Hazards
The shear expanse of the Delta Levee system, its fragility and the grave consequences of
losing functionality in a levee section make the Delta a target for terrorism. There is an
32
acute need to survey the Delta for particular levee sections whose loss would result in
failure of systems to deliver water to residential, commercial and agricultural users.
These sections could then be hardened and protected to reduce terrorist threats.
Synergistic Hazards
It was mentioned that all the hazards discussed can happen independently or together and
there is a need to consider the consequences of simultaneous events that could be even
more catastrophic than any single event. Examples could be a high tide during an intense
winter storm in the presence of global warming- induced higher sea level, or an
earthquake during a flood. The modeling capabilities described above could be used for
evaluating the consequences of simultaneous events.
GROUP 3: What can go wrong in the system?
Group ( 3) Participants: Cheryl Bly- Chester ( Facilitator), Robert Budnitz ( Note- taker),
Scott Branderberg, Roger Henderson, Ron Ott, Ray Seed.
Key questions:
How can levee system behavior be predicted?
33
How can levee structural integrity be predicted? Modeled? Measured? Enhanced?
What is required?
Elements to consider for answering these questions:
1) Behavior of the levee system
- Response, geotechnical properties
- Levee structural properties
- Operating requirements
- Real- time behavior of system
2) Monitoring and data requirements
- Update of the topology of the system
- Update of the functional demands
- Environmental parameters ( animal and fish life, marshland, air and water
quality, etc..)
Group 3 Discussion Summary
Group 3 first developed a list of specific causes that address " What can go wrong?".
Some of these are technical factors, but many are non- technical or institutional factors.
Among the latter, the most important are inadequate investments in maintenance;
overlapping jurisdiction problems that impede such investments; the problem of dual- use
levees, where the other function ( for example, if the levee is also a road) causes the
problem; and most importantly, continuing adjacent development ( housing, commerce,
light industry).
The Group then moved on to identify knowledge gaps, or gaps in analysis capabilities.
Again, a long list was developed. One very important gap is the lack of comprehensive
levee- specific information about who owns and manages each levee, who could make a
quick decision about it, what assets each levee protects, which dual uses does it support,
and the inspection history. Another gap is the need for a complete data base of levee
inspection reports, available on a " no- fault" basis vis- à- vis liability. Still another one is
the need for a trainig and certification program for levee inspectors Technical needs
include an improved hydrodynamic modeling capability, an improved wind model for the
Delta, and the fact that hydrology information is outdated and the topography is
dynamically changing, making modeling of the system uncertain if not sometimes
erroneous. Finally, a systematic inspection protocol for the levees must be developed and
implemented across- the- board by the owners/ managers of every levee.
GROUP 4: What is at risk and what are the consequences of damaging events?
Physical, Societal/ Economic Vulnerabilities/ Organizational, etc.
Group ( 4) Participants: Michael Hanemann ( Facilitator), John Ziagos ( Note- taker),
Carol Baker, Dan Farber, Catherine Freeman, Nina Kapoor, Ladd Lougee, Doug Rotman,
Matt Vader Sluis
Key questions:
34
What kind of information is needed to help in the decision- making process?
What are our expectations for the levee system?
What mechanisms are available for protection of vulnerable elements?
What is required for improved mechanisms?
Elements to consider for answering these questions:
1) Characterization of elements at risk, and how they will evolve
- Residential population census ( where do people live)
- Commercial/ Industrial census
- Agricultural land use
- Traffic patterns ( commute, escape routes)
2) Model the consequence of all possible hazard scenarios
- Economic models
- Land use models
- Social migration, health, … models
3) Monitoring and data Requirements
- Building permits
- Land development
- Water absorption ( runoff potential)
Group 4 Discussion Summary
The over- arching theme of the discussion was how might scientists and technologists best
support the natural disaster legislative decision- making process? Detailed physical,
societal, and economic vulnerabilities, decision- making, and organizational issues were
considered with detailed examples permeating the entire discourse. Discussion highlights
include: the understanding and brain- storming of solutions to current and relevant natural
disaster- related legislative actions/ inactions and contentious issues such as resolving
floodplain mapping particulars and the sharing of liability for flood protection between
state and local governments and developers in the context of significant built- in funding
impediments and multi- agency cross- responsibility chaos.
Summary of key recommendation topics include: 1) creation of a mandate for official
State planning for hazards to develop a state- wide, long- range, adaptive flood and climate
change risk assessment management approach, 2) development of scenario contingency
planning to encourage focused investigation of adaptation policy including unthinkable
policies focusing attention on modifications to building codes and examination of
unthinkable alternative land use management strategies under the climate change scenario
considering, for example, buying up farmland/ restricting urban development while
utilizing financing/ liability options for flood control that incorporate land use modeling
and disaggregate current information to match jurisdictional boundaries and finally, 3)
development of tools to assist legislators and agencies in policy formation and decision-making,
that might include: post- breach decision support systems to assist in setting levee
repair priority, development of state- wide multi- year levee repair standard, and creation
of an official State climate change scenario( s).
35
General Discussion
The final session was a general discussion, intended to address cross- cutting issues and
interfaces, to brainstorm system performance criteria and constraints as well as
highlighting points not made in the earlier sessions. A few topics emerged as main
discussion points. What follows is a summary of the main points, with specific notes
grouped under the discussion topics:
What is needed?
A consortium approach is needed to engage, provide products to and get feedback
from policy and decision makers. This is broader than the research community alone.
We need an umbrella that supports the broad set of ( technical and policy- relevant)
disciplines that need to be applied.
On the scientific side, it is critical for seismology and geotechnical experts to get
together with hydrology and climate researchers. A key question to address is “ what are
the conditions that set the stage for a catastrophe?” ( i. e., an earthquake during wet
season). Need a working dialogue.
There is a missing piece in the current forum dealing with policy and our ability to
assess the potential impacts of different policy options. DWR staff may see their role as
doing science, not policy. CalFed has been a forum. There may be a need for neutral
ground to float policies and ask scientists for analysis related to the policies their research
relates to or can contribute toward.
The Katrina example highlighted the need to invest DOLLARS and DILIGENCE.
California should consider the benefits of PROACTIVE and PREVENTATIVE. We are
acutely aware of our vulnerabilities and the threats posed. An ounce of prevention is
worth 10 pounds of cure.
Unfolding events/ opportunities:
Bond Issues: The upcoming bond issues ( 1E and 84) have specific tie- ins to levee
improvements. There is potential for some targeted R& D to enhance the efforts these
bonds may finance.
Legislative actions and Executive Order: The Governor recently signed several delta
bills, including Executive Order S- 17- 06 that establishes a blue- ribbon task force to
address the Delta – the Delta Vision process.
Information on the legislative actions:
Executive Order S- 17- 06 initiates the Delta Vision and establishes an independent
Blue Ribbon Task Force to develop a durable vision for sustainable management of
the Delta. Making the Delta more sustainable will require a concerted, coordinated
and creative response from leaders at all levels of government, stakeholders,
academia and affected communities, and will require significant private and public
partnerships and investments. The Delta Vision is designed to accomplish these
goals:
SB 1574 will create a cabinet- level committee chaired by the Secretary of the
Resources Agency and include the Secretary of the Business, Transportation, and
Housing Agency, the Secretary for Environmental Protection, the Secretary of Food
and Agriculture, the President of the Public Utilities Commission, the Director of
36
the Department of Finance and the Director of the Office of Planning and Research
to develop a plan for a sustainable delta.
The Delta Vision is already underway. There are two potentially bad end- member
options about this:
( 1) there is no money to affect changes, or
( 2) these is money provided, but without a plan. The bond issues may provide on
the order of $ 6B for the levees, and a strategic plan is needed to ensure these investments
provide the expected protection and reliability. We are likely to have significant funds
available soon ( through the bond measures). We want these funds to be spent wisely.
Urgent: a short range view needs to be developed ASAP, but work will go on for
some time. There is an opportunity to incorporate R& D along the way to improve overall
results. We need to lay out an R& D strategy for three timescales:
- Urgent, short- range
- Medium timeframe
- Long- term
Other points:
A lot of construction and other work will be outsourced -- controlled by DWR. They
need a good strategic conceptual framework to ensure satisfactory results.
There may be sufficient funding and sense of urgency that there will be a need to get a
lot of work done quickly.
Typically, the funding is distributed through the organization and the staff approves
projects according to guidelines. There is not now a research component specified. With
respect to a science program, DWR would look to CalFed to do it.
Need a long term view with adequate science to guide the implementation of the
policies that are developed by Delta Vision, policy makers, legislature using DRMS.
Vision goals: As some of the workshop speakers pointed out, the decision may be to
create a hydrologic bypass ( peripheral canal), or to armor a part of the levees through
which the water supply would run. There are multiple options, and the technical options,
potential consequences and ultimate costs must be investigated.
In the meantime: Legislative bodies need to sustain the existing Delta system while the
deliberation progresses on what to do in the long- run.
Systems approach:
A systems approach is important for considering the multiple dimensions. Tools are
needed.
Coordination between policymakers and the technical community:
The system model you build has a lot to do with the question you think are important.
Challenge is to find those policy makers that can help you determine the right questions.
The model is only a thinking tool to focus other research. We need to get guidance from
policy- makers in framing the question a model needs to address.
We must short- cut the cycle by meeting with dedicated legislative staffers -- liaisons
as way to do this, on a regular basis.
37
An advisory committee ( including legislative staff) might serve this purpose.
Similarly, multiple agencies need to coordinate better. The state needs to encourage a
high reliability organization.
Leverage existing roles and capabilities in all technical fields. For example, the
California Geologic Survey has made its resources and data available.
Ongoing activities/ opportunities:
CalFed started the Bay Delta Science Consortium, that could be focused on these
issues. It provided financial support to encourage collaboration, something much needed
in this arena. NOTE: CalFed hasn’t done much on levees to date and has recently
reconstituted the science board without any engineers on the board, and important gap.
The USGS seismic work ( i. e., NEHRP) - a seismic safety communication – could be
incorporated to advantage.
The CA Seismic Safety Commission has some relevant initiatives listed in their strategic
plans:
- California Earthquake Loss Reduction Plan CSSC 02- 02, 22002- 2006.
- A Safer, More Resilient California: The Alfred E. Alquist State Plan for Earthquake
Research CSSC Publication 2004- 03, June 2004
The DRMS Phase 1 draft will be due 3/ 07; Initial Technical Frameworks ( ITF) white
papers ( 14 of them) are now available on the DWR website
Additional inputs were given regarding the prioritization of recommendations and
next steps:
Which are the highest priority actions or most important areas to focus on ( short-term
to long- term)?
1. Improve our ability to predict high- water stands ( height, duration, frequency)
spatially within the Delta channel system.
This will require better characterization of the levee configurations, channel
geometry and bathymetry ( at lidar resolution or better?); more complete
observations of the network of Delta water levels, flow rates, and water densities
( both from salinity and turbidity); and development of more practical/ complete
hydrodynamics models of the Delta flows brought to as realtime as possible,
2. Improve the mapping of probabilities of seismic episodes that are AT LEAST
large enough to threaten major levee disruptions.
This is NOT to say improved mapping of the largest seismic episodes, or greater
ability to resolve among various levels of seismic activity, but rather a focus on
mapping of the odds of EXCEEDING some reasonable seismic- impact threshold.
3. Improve our knowledge of the development of lands in and around the Delta.
Too much of our knowledge of plans and built developments in and around the
Delta seem to arrive as anecdotes and hearsay... some central monitor/ repository
of Delta land uses and land- use plans needs to be developed and maintained.
38
4. Rationalize current projections of sea- level rise as apply off San Francisco and
as translated into the Delta.
Current projections have a serious disconnect between short- term sea- level
fluctuations and the long- term trends as derive from ice- cap melting... even if we
can't be sure yet what the ice caps will do, a considered and rationalized approach
to incorporating these uncertain trends into sea- level projections is largely
lacking.
What are the key next steps to be taken?
1. The State needs to acquire and maintain its own capacity for VERY regular
surveys of the Delta and levees at high resolution ( i. e., lidar). Lidar coverage of all the
Delta levees could be obtained ( from a small plane) in a day, and with in- house
capability, this could become a standard action on a periodic basis and after most large
storms or earthquakes.
2. Develop a 21st Century plan for monitoring in the Delta, taking advantage of
new daisy- chained sensor- network capabilities and new sensor types
3. Land- use, water- use and economic information must be made available from the
many stakeholders in the Delta for strategic planning. Legislation should be
introduced to facilitate and guide this process.
4. A long- term community plan for a practical but highly resolved Delta
hydrodynamics model ( and supporting data streams) must be developed and
implemented. The goal would be a community effort to develop the best technical
product that meets the practical needs of the next decade or so. It will likely require a
broader participation amongst public and private institutions than has been achieved in
the past.
39
Appendix I
Group Discussion Outbrief Presentations
GROUP 1: How to characterize the levee system infrastructure:
Components of the Levee “ System”:
Physical element
Operation, management, and decision- making process
Funding process
Science community
Regulatory element
Political component
Public input/ support
Interagency interface cooperation
We need to have a holistic view of a “ complex” system ( organic), where stakeholders
may have diverse and diverging interests
Current Delta Levee Functions
Water Supply
Current agriculture
Flood control
Maintain ecosystem/ bio- diversity
Transportation/ Infrastructure
Recreation
Human life and property
Run- off disposal
Functions of the future:
All the current functions
Changing in time
Regional and area specific
Possible function to consider is CO2 sequestration
Others…?
Tightening Web:
Sea Level rise ( climate change)
Seismicity
Urban development
Subsidence
Ecosystem
Increasing funding need in business- as- usual for upgrade funding needs for maintenance
and emergency response
Some issues, constraints:
Need to look at solutions for:
40
- Short term, immediate update and improvements ( i. e., the DRMS effort)
- Emergency response
- Long term maintenance, and updating for a changing world ( Long- term = > 10years)
Are present decisions made with long- term view?
Present approaches are incremental ( Time, resources and funding). Limits ability to make
major changes and experiment with new approaches.
Political Constraints/ Expediency and rush studies could be counter productive
Consensus, political and scientific. Need to fold- in experts’diversity
Different degrees of maturation in sciences and engineering:
Climate change need more characterization – Needs a sound risk model
More work in fault studies and characterization
Ecosystem is least understood – More research and observation is needed
Factors for monitoring levees
Crest elevation, width, slopes
Seeps and boils
Deformation
Cracking
Settlement
Erosion
Water flow
GWT
Subsidence
Ecosystem monitoring
Population
Delta smelt
Water quality – Salinity change, turbidity, etc.
Flood stage, run- off, temperature, etc..
General Recommendations:
Understanding the critical needs
Preparing for emergency response
Pre- event readiness- Inter- agency protocol
Social engineering
Commit to higher funding for maintenance
Establish partnership between the government and industry
Characterization data should be coordinated and organized for immediate ( real- time)
access.
- Where should it reside? Central? Distributed?
- Coordinate the various activities ( i. e. GIS work)
- Need for Delta data center ?
- Need for a unified source of information ( during an emergency response)
- Define agency roles for the Delta
Know where to go
Both during and emergency and normal time
41
- Need to collect the existing data and update
Remote sensing and non- destructive testing ( GPR, MR, Geophysical survey)
Instrumentation – Real- time input
A few specific recommendations:
Develop Standard for design and construction of Delta Levees
Need for specialized material properties, testing to augment the geotechnical database
Better LiDAR survey and more frequent flights particularly after each storm
Need to identify potential borrow material
Large scale testing
Need for a 3- D hydrodynamic model real time mode
GROUP 2: What could disturb the levees? Damaging Hazards
What could disturb the levee system? ( defining scope)
Including possible natural and environmental hazards, man- made hazards, malevolent
actions, and non- natural hazards such as socio- political elements in the organization
What could disturb the levee system?
Seismic Events and Consequences
Hydrologic Events and Conditions
Climate Change Consequences
Development
Subsidence
Terrorism
Seismic Hazard
How will the levee system behave to seismic loading?
What is the expected loading?
- Seismic Hazard Analysis
- Ground motion observation and modeling
- Geotechnical observations
- Ultimately, need input ground motion for levee design
Time- series or response spectrum
Will levees fail under seismic loading?
- Detailed characterization of levees
- Modeling of levees
Expected Seismic Loading
Requires:
- Seismic Sources
Earthquake Faults ( geometry)
Slip rates ( repeat times, max magnitude)
- Regional- Scale Geologic/ Seismic Velocity Model
Attenuation relationship
42
Basin Effects
Scenario earthquake modeling
- Site Response and Geotechnical Constraints
Empirical Observations
Boreholes, Sampling, Lab Measurements
Must include dynamic response
Requires observations of ground motions
- Weak motion recordings from
Smaller local/ regional events
Large teleseismic events
- Can be used to identify amplification of seismic waves in the Delta ( basin structure
- Can be used to validate geologic/ seismic velocity model
Seismic Sources ( see Figure 1)
Green Valley & Greenville Faults are poorly characterized
- These are very close to Delta
- Can trench these strike- slip faults
Blind Thrust Faults
- Coast Range Great Valley Fault
Runs under Delta
- Mount Diablo Fault
- Must rely on seismic reflection
Geodetic techniques may improve slip rates
- InSAR, GPS, LIDAR
- Will also constrain subsidence
Geologic/ Seismic Model ( see Figure 2)
Inherently 3D
Require deep boreholes to map sub- surface
Seismic imaging difficult due to logistics and near- surface materials
Must scale lithology to
- seismic velocity
- attenuation
Must be validated with various observations
- Local/ Regional/ Teleseismic earthquake waveforms
- Gravity
Site Response/ Geotechnical
Characterize near surface geology
- Site response ( amplification)
- Liquefaction potential
- Can one identify the “ failure” layer
Collect samples throughout the Delta
- Increase spatial coverage
- Characterize dynamic properties of samples
43
Ground Motion Prediction Methods
Standard Probabilistic Seismic Hazard Analysis ( PSHA)
- With improved input
- Only provides PGA, PGV, Spectral response
Ground Motion Time- Series and/ or Response Spectra
- Joint empirical and simulation- based method
- Compute ground motions, use HPC
3D simulations of large earthquakes for low frequencies
Empirical or simplified model- based motions for high frequencies, stochastic ruptures
Merge low- frequency with high- frequency motions
Levee Failure Under Seismic Loading
Requires detail models of levees
- Geometry, materials, effect of water
- Sub- surface geology
Improved analysis methods
- Current practice
2D, equivalent linear, simplified non- linear
- More sophisticated analysis could be done
3D, fully non- linear, includes liquefaction
More Challenging:
Validate levee performance by mechanically driving motion
- Find analog structure w/ o water
Search for deformed geologic surface at depth
- What density of boreholes?
Monitor deformation with geodetic techniques
- Identifies slip on faults and subsidence
Hydrologic issues
Can we predict precipitation in sufficient detail?
When will events occur? Dry or wet season? How will this correspond to how much
water is stored in reservoirs?
Need a study of how well large and long duration flood events are handled by the model.
How well did models handle floods of record such 97 and 86.
Time based forecasting. Can we release water when we know something is coming in
three days even though we haven’t hit the 75% level when they are supposed to release it.
More specific understanding of the effect of where the rain fall. see project INFORM to
get more flexible operating procedures. CALSim is a component ( Jay Lund) but he isn’t
looking at these time scales. Optimize the short term management of the system. Need
to do this comprehensively for the state.
We need a model of the managed hydrologic system coupled with models of the
hydrologic and atmospheric input.
Flood advisory and diagnostic system – like LLNL’s National Atmospheric Release
Advisory Center ( NARAC)
Can this help to work with conservative decision makers?
44
What is the impact to the delta from a breach under various conditions?
There have been failures that resulted in gulps – but models of this are slow – need a
quicker faster model – in order to know whether to release model from Shasta etc.
- Jones track episode gave some information, but not generalize- able
Need to study which levee breach puts the system in danger the most – we don’t know
now.
Need a model to understand how the delta flushes out – hydrodynamic model of the delta.
There is need for a parallelized version of these models
A better model – 3d that could be run quickly.
Climate change
Needs to be an updated survey of sea level rise projections
We need a better characterization and exploration of flooding under climate change that
accounts for the managed system ( reservoirs etc)
Need more thorough exploration of winter storminess and ppt systems in climate change
model projections – where is the snow line? Changing runoff regime while climate is
changing.
Need to understand extreme events better – like in 1997 when a storm track was stuck for
10 days – do our models replicate this behavior and how does this go forward. Number
of intense, persistent events and what temp they occur under.
Water temperature is a variable that has not been studied or monitored in the bay and
delta – need for better models– effects species.
Look at the simultaneous effects of
- High tide
- Sea level rise
- Floods
- High winds
- And a large earthquake
Climate Monitoring:
Mountain rain/ snow transition zone needs to be monitored – as this dictates runoff
Water temperatures in the tributaries and delta need to be monitored
Do we have enough water gages in the system? Say between Antioch and the delta?
Subsidence
Need permanent geodetic monitoring
- InSAR, GPS, LIDAR
Subsidence of the levees themselves? Can we predict this? Is it monitored effectively?
- Permanent scatterer methodology
Are the elevations of the levees known?
Development
Changes geologic conditions near levees
At what distance and how does development change stability conditions for the levees?
45
Terrorism
Are there particularly soft targets?
Do we know where they are and do we know how to protect them? Or harden them?
- A hydrodynamic system model could answer this
GROUP 3: What can go wrong in the system?
What can go wrong? ( technical factors)
Overtopping
Through- levee seepage
Under- seepage
Bank erosion
Channel erosion
Wind erosion
Levee slope instability
Seismic- liquefaction
Seismic lurching
Dredging that undermines the levees
Close- in dredging damage vs. broader dredging damage
Penetrations ( local)
Un- maintained growth ( trees, bushes, etc.)
Beaver damage
Sea- water penetration due to sea level rise
Terrorist acts
What can go wrong? ( Institutional/ non- technical factors)
Cancellation of state programs or funding
Emergency response failures ( planning, OR implementation)
Dual use levees --- failure of another function ( e. g., levee also a road)
Inadequate investment in maintenance ( causes a lot of above issues)
- Obstacles to maintenance:
Physical and policy obstacles
Resources--- follow- through
Overlapping jurisdiction problems impeding investment or response
Problems with the levee SYSTEM vs. individual levee problems
The needs of the RIVER AND ITS USERS vs. the levees
The needs of other utilities ( electricity, gas, rail)
What can go wrong? ( Land use management)
DEVELOPMENT ( the largest single issue today)
Property rights impediments
Agricultural practices causing subsidence
Ecosystem restoration
46
KNOWLEDGE GAPS & TOOLS GAPS
( data, knowledge, tools/ methods, implementation)
1. Information about each levee:
Who owns it, manages it, who could make a quick decision, what assets does it protect,
which dual- uses does it support, inspection history
Improved topographic data, kept up to date
Improved cross- section topography
Improved subsurface geotechnical knowledge
Several fully- characterized levee sections ( a few dozen)
- Study a subset of the above with full seismic response, stability,
and SSI analyses
2. General ( broader) information needs
Enlist geophysics community to brainstorm how they can help us understand levee
structure ( Get the geophysicists to focus on under- seepage and through- seepage issues)
Data base on endangered species and all other species
Outdated hydrology
- Even without climate change
Delta topography is dynamically changing
Improved hydrodynamic modeling tools
- Including real- time hydrodynamic tools in an emergency
Seismic instrumentation -- including full suite of seismic data
3. Need for a complete data base of inspection reports
Legal issue: The reports need to be on a " no- fault" basis vis- à- vis liability
Reports on all levee works, and why
Reports on repetitive repairs in the same spot
Data base of annual flood- fight observations by location
CROSS CUTTING ISSUES
Develop a systematic inspection protocol for levees
Need a training program to develop a cadre of levee inspectors
- Graduate students, junior engineers, internship programs
Need a wind model for the Delta ( statistical compilation of velocity,
- direction, hazard, annual variation across the Delta)
- data to validate models
IN GENERAL:
We need technical information and insights to inform the political and decision- making
process
GROUP 4: What is at risk and what are the consequences of damaging events?
Physical, Societal/ Economic Vulnerabilities/ Organizational, etc.
47
Background: As it happens, there was a legislative battle in the last session on ( i)
floodplain mapping, ( ii) “ show me the flood protection”, and ( iii) sharing liability for
flood protection between local governments & developers. Plan to require proof of 200 yr
flood protection was lowered to 100 yr ( and only > 25 units) before the bill was killed.
Maintenance of levees is a “ hodge- podge” of agencies, they say “ not enough funds” and
run into a prop 218, need a 2/ 3 vote to get new funds – THIS IS A BIG ISSUE
Risk Assessment – the problem
We need to develop a plan ahead of time for which islands we should let go. The geotech
assessment will take 10 yrs for DWR. But, can’t wait that long. Therefore, need multi-year,
long- range plan. But, what exactly would that look like? In effect, how would one
structure an adaptive management approach to risk assessment?
Also, not just which levees should be repaired in the event of a breach, but also to what
standard should they be repaired ( 100 yr, 200 yr, 1,000 yr)
Again: how to answer this in the spirit of adaptive management?
How to move forward?
To make headway in face of bargaining impasse on flood control, broaden the agenda:
include climate change along with flood risk.
Recognize crucial need to work through contentious issues ahead of time – contingency
plan, which can be based on a scenario rather than a forecast.
Create a mandate for official State planning for flood and climate change hazards.
Floodplain/ Preparedness Planning for California
Develop an official State climate change scenario( s) like those used by Climate Action
team, but with some additional modifications; perhaps focus on 2005- 2035.
Assess risk of damages under this scenario for flooding, fire, for local governments etc.
Also, encourage focused investigation of adaptation policy and unthinkable policies.
Adaptation Policy
Focus attention on what modifications are needed for building codes under the climate
change scenario. We normally do this ex post. Here we do it ex ante.
Examine ( presently unthinkable) alternative land use management strategies under the
climate change scenario: buying up farmland/ restricting urban development.
Examine ( presently unthinkable) financing/ liability options for flood control.
Additional modifications for scenarios
Spatial disaggregation/ downscaling to more local regions and to upper elevations.
Refine from monthly to daily/ hourly time step to investigate flooding, and environmental
quality ( temperature, flow), at certain locations.
On hydrology, include Colorado River basin.
Incorporate land use modeling.
Disaggregate to match jurisdictional boundaries
48
Challenges
Shift focus from optimization to robustness and bargaining.
How to implement adaptive management in an institutional and political setting?
How to incorporate ( 1) periodic review, and ( 2) compensation to permit modifications to
be made
Technical Tools
To deal with climate change, need new hydrology ( streamflow/ reservoir management)
models that are NOT tied to past hydrology.
Need to link hydrology model to land use model, economic model.
Focus should be linking distinct models probably on different temporal & spatial scales,
rather than a single, galactic, integrated, hydrologic- economic model.
49
Figure 1. Seismic Sources
50
Figure 2. Geologic/ Seismic Model
51
Appendix II
Press Release
The workshop was highlighted in a news statement released on October 4, 2006, shown
below ( it can be found at http:// www. llnl. gov/ pao/ news/ news_ releases/ 2006/ NR- 06-
10- 01. html.)
Workshop identifies research needs to protect levees
Approximately 60 research scientists, engineers, policy makers and agency
representatives from around California gathered recently for a two- day workshop to
define research needs in order to manage the flood risks facing California’s levees.
The workshop, held at the University of California Center in Sacramento, covered a wide
range of risks facing levees in the Sacramento- San Joaquin Delta and the Central Valley
– from seismic risks and infrastructure frailty risks to climate change and risks associated
with urbanization and inappropriate land development.
California Department of Water Resources
Concerns over flood risks from California's levees range from earthquakes to climate
change and urban growth.
“ The Sacramento- San Joaquin Delta and levees are complex and extremely vulnerable.
This workshop delineated a critical need to understand how all the parts and aspects of
the Delta interact as a system, so that society can make wise choices about investing in
the future of this vital resource,” said Jane C. S. Long, associate director for Energy and
Environment at Lawrence Livermore National Laboratory.
Concern over the viability of the Sacramento- San Joaquin Delta and levee system has
increased in recent years, due in part to the Jones Tract levee failure in 2004, the
Hurricane Katrina disaster in New Orleans and the recent centenary of the 1906
earthquake. The Delta is crucial to California’s agricultural economy.
With several groups now investigating the wide variety of hazards facing the Delta, the
Center for Catastrophic Risk Management and the California Center for Environmental
Law & Policy at UC Berkeley and Lawrence Livermore co- sponsored the workshop.
“ The session brought together California policymakers with some of the nation’s leading
technical experts,” summarized Dan Farber, professor of law at UC Berkeley and director
of the Environmental Law Program. “ The group identified critical gaps in our knowledge
about the enormously complex Delta system, including the impact of climate change on
flood risks.”
“ This workshop generated a new paradigm for viewing the Delta system that promises to
bring together many competing interests to work on a common sustainable solution to
delta concerns,” added Dave Mraz, program manager for the California Department of
Water Resources’ Delta Levees Program.
52
Over the two days the forum:
* Identified current vulnerabilities facing the levees, such as structural failure, seismic
loading, flooding, terrorism;
* Considered longer- term challenges such as climate change, sea level rise and water
supply implications; and
* Defined research requirements to fill gaps in knowledge and reduce uncertainties in
hazard assessments.
“ One of the key goals was to broaden the focus from seismic risk to the flood risks
associated with climate change and rapid urban growth, and from engineering issues to
the economic, legal and institutional factors that can have a crucial influence on the
success of efforts at disaster prevention, response and recovery and, hence, determine
California’s flood damage exposure” said Michael Hanemann, professor of
environmental economics and policy at UC Berkeley, and director of the California
Climate Change Center there.
The workshop then turned to development of a detailed list of short- term and long- term
research and tool development needs, including research priorities for problem definition,
prediction, management tools, policy approaches, technology development, data
collection and more. A report will be prepared summarizing the workshop’s findings.
Raymond Seed, a professor of civil engineering at UC Berkeley, summarized the
workshop by noting, “ In the wake of the recent disaster in New Orleans, there is now an
increased awareness of California’s own levee fragility and flood risk exposure. These
are complex issues, requiring levels of teamwork and collaboration among numerous
technical disciplines, and this workshop has been a valuable step in that regard.”
Founded in 1952, Lawrence Livermore National Laboratory has a mission to ensure
national security and to apply science and technology to the important issues of our time.
Lawrence Livermore National Laboratory is managed by the University of California for
the U. S. Department of Energy’s National Nuclear Security Administration.
53