🚀 go-pugleaf

CALIFORNIA�S OPPORTUNITY TO ACHIEVE IMPROVED FLOOD MANAGEMENT 
AND MAJOR ECOSYSTEM RESTORATION

By Philip B. Williams  Ph.D., P.E. 

January 27, 1997

A. SUMMARY

The January 1997 flood presents us with a unique opportunity to propose a
new flood management solution that provides greater hazard reduction
benefits for Central Valley communities, with greater reliability and at
lower cost than continued reliance on the present flawed and unreliable
flood control system.  This solution, which fully integrates flood
management with river and watershed management, will also enable the large
scale restoration of habitat for wetlands, waterfowl, and fish that have
been largely destroyed by the construction of this same flood control
system.  It can be argued that this solution is the only solution, as there
is now every indication that even with massive new expenditures on more
structural flood control engineering works, there is no assurance that the
present  system would be able to prevent the inevitable catastrophic flood
which is likely to occur within the next century. 


B. BACKGROUND

The rivers of the Central Valley of California have the world�s most
comprehensive and highly engineered flood control system.  This system was
constructed mainly by the federal government over the last 80 years and
consists of a complex of flood control dams, levees, and flood bypasses. 
Since the system was completed in the late 1970's, it has been tested by
major rainfloods twice; in February 1986 and now in January 1997.  In both
of these flood events significant inadequacies in the operation,
maintenance and design of the system were revealed, resulting in several
billions of dollars of flood damages.  Most seriously the mis-operation of
Folsom dam in 1986 caused a near catastrophe for the City of Sacramento;
and the predictable but unanticipated large flood spills from New Don Pedro
and Friant dams in the 1997 flood created large flood waves that caused
multiple ruptures in levees as they traveled downstream.

Even as this system was being completed it was becoming obsolete.  By the
1970's the U.S. Government had finally recognized the futility and
limitations of  an exclusive reliance on an expensive strategy of
attempting to �control� floods through massive engineering works.  It was
realized that flood control ignored the need to address factors such as
inappropriate land use that were causing the steady escalation of flood
damages experienced in the U.S. since the 1944 Flood Control Act.  Starting
with the passage of the 1968 Flood Insurance Act  a series of new policies
were enacted adopting flood plain management approaches.  Unfortunately
during the 1980's and 1990's, at both the state and federal level these
initiatives languished, but since the 1993 Mississippi floods the validity
and utility of flood management as opposed to flood control strategies has
been confirmed by federal agencies and professional organizations.

The construction of the massive water control system of the Central Valley
caused major ecosystem declines and extinctions that were only understood
or acknowledged within the last two decades.  With the passage of the 1992
Central Valley Project Improvement Act, and the signing of the 1994 Bay
Delta Accords, there are now new State and Federal initiatives being
formulated to restore fish, wildlife and wetlands.  These initiatives are
focussing on the restoration and management of flows and habitat in the
river and estuary system. 


C. A FLAWED FLOOD CONTROL SYSTEM

The flood control systems of the Sacramento and San Joaquin Rivers have had
built into their design, serious intrinsic flaws, that frustrate
achievement of their primary design objective, which is to prevent all
flood damages for floods smaller than the design flood,[in most places the
100 year flood].  These include:
  
1. Outdated Levee System
In 1910 when the levees and bypass system of the Sacramento river Flood
Control project [SRFCP], were designed, one of its major objectives was to
flush hydraulic mining debris sediments from the Sacramento river channel
to restore navigation for river stern wheelers.  This was a major factor in
choosing to construct levees as close to the river channel as possible,
often within the active meander belt of the river and across highly
permeable relict river channels.  The consequence of this design decision
has been eroding and unreliable levees, high maintenance costs, and higher
flood stages than occurred naturally.  Because the levees were so close to
the river, their maintenance activities and repair with rip rap has now
destroyed essential riparian habitat along most of the Sacramento River.

2. Flood Peaks Increased
The design of the SRFCP significantly reduced the storage of flood waters
within the Sacramento valley.  This storage of floodwaters on the
floodplain had slowed down and dampened the flood peak as it moved
downstream.  The engineers of 1910 recognized this effect and estimated
that by the time the project was complete the flood peak at the vicinity of
Sacramento would increase the design flood from 450,000 cfs naturally to
600,000 cfs, due to the elimination of floodplain storage.  In the 1986
flood the flow in the vicinity of Sacramento was just under 600,000 cfs. 
Elimination of floodplains also destroyed or disconnected almost all of the
Sacramento Valley�s extensive floodplain wetlands from the river,
destroying a key component of the watershed�s ecosystem.

3. Flood Control Encourages Flood Prone Development
At the time the SRFCP was designed the Sacramento Valley was sparsely
settled.  Where people farmed in the floodplain they took prudent actions
to protect themselves, like building ranches on raised ground.  It was
assumed that overtopping of the levees during the inevitable flood larger
than the design flood was an acceptable risk.  What was not foreseen was
the effect the levees would have to encourage people to develop in the
floodplain and neglect traditional flood proofing methods.  By the 1940's
it was decided that additional flood control was the answer to the growing
flood hazard problem and a series of large flood control reservoirs were
constructed in the Sacramento and San Joaquin Valleys between the 1940's
and 1970's to reduce flood peaks of large floods to the 1910 design
capacity of the SRFCP, and the design capacity of the later San Joaquin
Flood Control Project, to reduce, but not eliminate, the risk of levee
failure.  At the time of the design of these reservoirs no consideration
was given to the further increase in flood damages likely incurred by
people induced to develop in floodplain areas where the risks, while
reduced, were still substantial.  The construction and operation of these
large dams on the Central Valley rivers has probably been the most
important cause of wholesale decline in the river and estuarine ecosystem.

4. Flood Control Reservoirs Mis-operated
When these dams were designed in the 1940's and 1950's they were
experimental in that there was very little operating experience with dams
of this size.  Nevertheless, their flood control benefits depended on them
being operated exactly to plan during all major floods.  Subsequent
experience has shown this to be impossible.  This assumption did not
consider an inherent operational conflict.  All these dams were intended as
multipurpose reservoirs, to be used for power and irrigation as well as
flood control.  However, to maximize irrigation and power revenues requires
minimizing flood control storage space.  Recognizing this conflict, and
because federal taxes paid for the less tangible flood control functions,
reservoir drawdowns and releases during the winter flood season were
specified in federally approved flood control operating manuals. 
Unfortunately, enforcement of these specified operations has been lax or
ineffective and has allowed many dams to illegally store irrigation water
in the flood control space during the winter.  There is now a history of
incidents during major floods on many multipurpose dams  like the 1986
Folsom dam mis-operation.  Prescribed reservoir releases have been made too
little and too late, leading to too large or uncontrolled releases and
severely limiting the effectiveness of  the flood control dams.  Illegal
storage of floodwater in this way also has a subtle but important negative
environmental effect.  It tends to eliminate the natural winter and spring
flow variability which sustain important river ecosystem processes.
5. Inadequate Outlet Capacity
In the design of most of the dams outlet capacity is inadequate to allow
the most effective use of flood storage space.  For example more than half
of Folsom Dam�s flood storage space must fill before water pressure is
great enough to discharge the full flood release.

6. Flood Releases Unrealistically Constrained
In the design of the San Joaquin Valley dams, prescribed flood discharges
were so small, less than 10% of the design flood, that almost all the
incoming floodwater has to be stored.  With only a small amount of flood
storage available it is inevitable that these dams will overtop in a large
flood or be unable to handle successive small floods.  For example in the
January 1997 flood the inflow to New Don Pedro Dam was 130,000cfs but the
discharge was limited to less than the 9,000cfs design channel capacity,
resulting in the reservoir filling and spilling a flood wave of 50,000cfs
that broke the levees downstream.  This futile attempt to eliminate flood
peaks in the lower river has allowed encroachment of vegetation into the
old river channel as well as encouraging development in the floodway.  This
has caused a significant adverse effect on spawning gravels and the
riverine ecosystem. 

7. Risk of Catastrophic Failure
The design of the flood control system makes almost no provision for
addressing the consequences of the inevitable extreme flood larger than the
design flood or a flood wave from a dam failure that will overwhelm the
system.  For example, the design of the  levee system establishes a uniform
level of protection up to the level of the design flood.  In fact, for a
flood larger than the design flood  the safety of the City of Sacramento
would necessarily depend  on unanticipated or even deliberate levee
failures upstream to achieve lower flood stages at the City.  This is a
fact that is probably not understood or appreciated by communities
upstream.

 
D. OPPORTUNITIES FOR A NEW FLOOD MANAGEMENT SOLUTION

The January 1997 floods have reconfirmed the lessons of the 1993
Mississippi flood.  These lessons are described in the 1994 White House
�Galloway Report� and echoed in the 1994 report of the Interagency 
Floodplain Management Task Force.  The essence of a flood management
approach is changing the objective from �controlling floods� to the more
appropriate one of reducing flood hazards to lives and property.  In this
strategy, structural flood control measures are a very important, but not
the dominant, tool in achieving modern societal goals.  Land use controls,
flood insurance, building codes, relocation, flood proofing, emergency
preparedness, and public education are also important and effective tools. 
Flood management also requires �management� to substitute for
�construction� as the most important activity to protect our floodplain
infrastructure.  This in turn emphasizes the need for more sophisticated
and effective maintenance, operation, flood warning, training, monitoring,
and learning from experience to enable a cycle of constant improvements in
the system. 

To initiate a new effective flood management strategy will first require a
comprehensive unbiased reassessment of the effectiveness of the present
system.  Such an audit would most likely confirm many of the findings and
recommendations of the Galloway Report, but in addition identify some
problems unique to California�s flood control system.  For example, these
are some of the improvements that could be made to increase reliability and
effectiveness.

..Redesign dam flood control operation procedures to take into account
upstream reservoirs and  forecasted inflows to make prompt releases to draw
down the reservoir.
..Modify dam outlets to increase discharge capacity.
..Increase floodway capacity downstream by setting back levees.
..Increase flood releases to the expanded floodway capacity.
..Setback levees on the mainstem rivers to increase floodway capacity,
reduce risk of failure, and to  provide floodplain storage benefits.
..Lower flood bypass weirs to lower river stages.
..Identify planned levee failures and inundation areas for extreme floods or
dam failures.
..Use up to date dynamic flood prediction models that accurately reflect the
benefits of floodplain storage.
..Develop accurate assessments of residual risk of property located on the
floodplain but behind levees.
..Develop levee maintenance and inspection processes that accurately reflect
the main risks of failure.
..Reconstruct levees to improve reliability.


E. OPPORTUNITIES FOR HABITAT RESTORATION

Restoration planning efforts by Calfed and by other investigators all point
to the need to restore some portion of the floodplain wetlands and
woodlands that have been practically eliminated from the system.  Although
the science and practice of ecosystem restoration are new, there is a
substantial consensus that a prerequisite to the successful restoration of
any type of habitat is the restoration of physical processes that sustain
the living organisms.  For floodplains the key processes are floods, the
movement of sediment, and free interaction of flows between river channel
and floodplain.

To achieve substantive restoration of floodplain wetland processes will
therefore require :

..Setback levees to recreate floodplain corridors.
..Setback levees to allow for an active natural meander belt.
..Reconstruction of levees to allow for vegetated toes instead of rip rap.
..Acquisition of flood bypass land to convert to wetlands.
..Lowering of bypass weirs to increase frequency of flooding in bypass wetlands.
..Change in reservoir flood operation to increase frequency of smaller floods.
..Change in reservoir operation to allow for larger flood pulses to
regenerate floodplain woodlands.
..Increase in frequency of moderate sized floods to reestablish natural
riffles, pools and meanders.


F. CONCLUSION

There is an almost perfect overlap between those measures needed for
implementing an effective flood management strategy and those needed for
meaningful restoration of the Valley�s fish and wildlife.

-- 
Richard Penny   *    Development Director    *   Friends of the River
128 J St., 2nd Floor * Sacramento, CA  95814  *  916-442-3155 (voice)
rpenny@ix.netcom.com * www.friendsoftheriver.org * 916-442-3396 (fax)