October 01, 2012

Preparing for Extreme Floods: California’s New Flood Plan

California’s Central Valley is no stranger to extreme weather events, especially floods. Its bowl shape and connection to the heavy snow and rain in the Sierra Nevada mean flooding remains a constant threat. The state and the federal governments have spent the last 150 years battling “the Inland Sea,” that is the Central Valley. This anniversary year of the Great Flood of 1862 offers the opportunity to review the lessons from California’s past responses to extreme flood events. With climate change threatening more rain and less snow, states relying on mountain snowpack may learn from how California has responded to flood risks and changed its flood policy to prepare for greater risks of extreme floods in the years ahead.

Before Europeans arrived, Native Americans called the Central Valley the Inland Sea because the Valley commonly filled with water during the winter. The flood could be hundreds of miles long, although not necessarily deep. Early in California’s statehood, the state capital at Sacramento flooded on several occasions. The historic 1862 flood returned the Valley to its Inland Sea conditions, with a floodplain 300 miles long and averaging 20 miles wide. Legend has it that the governor attended his inauguration in a rowboat. In response to that flood, and over the decades that followed, Sacramento piled dirt in its streets to elevate the River City by 10 feet.

The gold mining that attracted so many to California made Central Valley floods worse. When panning no longer produced gold, mining companies began using giant hoses to wash away the mountain to get at the remaining gold. As a result, Central Valley rivers filled with sediment, elevating river beds as much as 11 feet. Rivers no longer stayed in their banks. Farmers sued to stop hydraulic mining from destroying the Central Valley breadbasket and were successful in 1884 when a federal court outlawed hydraulic mining. See Woodruff v. North Bloomfield Gravel Mining Co., 16 Fed. Rep. 25 (1884). The sediment nevertheless remained, so the state and federal governments began studying how to address the sediment problem and reduce flooding.

In 1911, the state legislature adopted a plan for a series of levee projects along the Sacramento River that would narrow the channel and allow the River to scour and move the sediment downstream. See 1911 Extraordinary Session, Ch. 12. Six years later, Congress authorized the U.S. Army Corps of Engineers (Corps) to implement the Sacramento River Flood Control Project with the state. See Flood Control Act of 1917, Ch. 144, 39 Stat. 948 (1917). For the next ninety years, these federal-state projects successfully adapted or rebuilt Sacramento River levees to control Central Valley floods. A similar project along the San Joaquin River developed as well. The Sacramento and San Joaquin flood control projects succeeded in preventing catastrophic floods for the populated cities at the bottom of the river systems—Sacramento and Stockton. The flood control project’s success at scouring sediment ironically resulted in the project’s eventual failure. Once the sediment was scoured, the rivers began to scour the levees. Incidentally, much of the sediment ended up downstream in the Sacramento-San Joaquin River Delta (Delta), the former shallow swamp. Settlers reclaimed land, to some extent, with the sediment by building and creating islands for agricultural and other uses. Since statehood, some of these islands have subsided so deeply below the adjacent water levels that the risk of multiple levee failure remains significant.

In part due to ineffective maintenance of the levees by the state, Sacramento River levees failed in 1986 and 1995, catastrophically inundating some smaller communities. In the well-known Paterno case, residents of one of these communities sued the state for damages from a 1986 levee collapse. Paterno v. State of California, 113 Cal. App. 4th 998 (2003). Paterno involved the Sacramento River Flood Control Project, which encompassed about 1,600 miles of levees, weirs, and bypasses to increase conveyance of flood waters downstream. Id. In 1953, the federal government transferred responsibility for the Sacramento River Flood Control Project to the state.

In the case, which was brought against both the state and a local maintaining district, the plaintiffs alleged both negligence and a taking of their property under inverse condemnation. The Court of Appeal found that the state’s acceptance of the levee into the Sacramento River Flood Control Project constituted a “plan.” Id. at 1029–30. The court also found that the state subsequently became aware that the levee had been improperly constructed out of sandy materials but took no action to address this problem. Id. at 1021–23. The state, therefore, had an unreasonable plan and was liable to the residents in inverse condemnation for their damages in an amount of $500 million. Id. at 1003.

The state continues to pay out the damages resulting from the Paterno case. Given the billions of dollars of real estate development behind levees in the Central Valley, liability related to flood control is a primary concern for the state, and eliminating this liability is a central goal of California’s new Central Valley Flood Protection Plan.

Other related but disconnected California floodplain policies have also increased the risk of damage to California’s people and property. For the last thirty years Central Valley communities have grown considerably, many in the floodplains adjacent to the Valley’s rivers. With the establishment of the Sacramento and San Joaquin flood control projects, state policy did not limit development in the floodplains protected by the state-federal projects. Again, the concept was to protect populated areas from flooding and not to prevent future development. The National Flood Insurance Program administered by the Federal Emergency Management Agency (FEMA) has allowed (and continues to allow) development of floodplains protected by levees that engineers determine give “100-year flood protection”—protection from a 1-in-100 chance of flooding every year. Applying this policy, cities and counties, which have authority over land use in floodplains, often allowed development that put people and property at risk if levees failed. This is because protection from a 100-year flood event does not remove the risk of flooding entirely. The “residual risk” that remains has now become a significant policy issue for FEMA in part due to the significant development that has occurred behind levees in California.

Focus on Flood Risk Pushes California Flood Protection Policy Reform

The Central Valley flood management debate dramatically changed course in 2005, when the Hurricane Katrina charged waters of the Gulf of Mexico overwhelmed the levees of New Orleans, devastating the city. Sacramento suddenly became the city with the greatest flood risk in the United States. Not only had levees deteriorated, but Sacramento had expanded development into floodplains protected by those levees. State agencies started reporting to the legislature on how many levees suffered from serious deterioration. Voters who had seen the destruction in New Orleans began to understand the risk of flooding. They passed two bond measures in 2006 authorizing about $5 billion in general obligation bond funding to pay to fix the levees and otherwise improve flood protection, especially in the Central Valley and the Delta.

The legislature, which had started considering flood legislation before Hurricane Katrina, more urgently addressed Central Valley flood protection after 2005. A convergence of factors—Hurricane Katrina, climate change awareness, the $500 million liability from the Paterno case, and greater voter awareness of California flood risks—ultimately led to passage of a package of bills to reform California flood protection policy in 2007. A floodplain land-use statute, known as Senate Bill 5 (SB 5), received the most attention. SB 5 restricts approval of new development in urban and urbanizing areas not protected from a 200-year flood. Statutes of 2007, Ch. 364 (2007). Local governments, developers, and the news media all engaged in the debate over increasing requirements for new floodplain developments to “200-year” flood protection.

Central Valley Flood Protection Plan

Beyond SB 5’s mandate that local land-use entities ensure 200-year flood protection, the legislative package included a more significant shift in direction at the state level. SB 5 called for development of a new plan by the California Department of Water Resources (DWR) (and adoption of the plan by the Central Valley Flood Protection Board) that would set a framework for protecting the entire Central Valley from floods in the future: the Central Valley Flood Protection Plan (the Plan). Moving away from relying primarily on levees for flood protection, the legislature adopted a policy that applied a systemwide approach to managing flood flows and set more stringent standards for allowing development in the flood plains. Its key elements promoted system integration in preparing for extreme events—state leadership coupled with broad local participation.

The Plan contains nine key elements. First, the Plan adopts a systemwide approach that moves beyond a focus on state-federal levees to a systems-based approach to flood response. Second, the Plan expands flood capacity, rather than confining channels as the state had done a century ago. In this way the Plan emphasizes an expansion of capacity, especially upstream, instead of rushing flood waters downstream. Third, the Plan requires DWR to develop state floodplain maps for the entire watershed. Fourth, the Plan requires DWR to send notice of flood risk to all property owners in floodplains in the watershed. Fifth, the scope of flood protection facilities addressed in the Plan goes beyond existing state-federal project facilities and considers all facilities in the watershed. Sixth, the Plan adopts a new name for the existing state-federal flood projects—now called “State Plan of Flood Control”—in order to distinguish between the watershedwide Plan and the state-federal levees. Seventh, the Plan disclaims and denies liability for additional flood facilities unless specifically adopted as part of the State Plan of Flood Control for which the state has accepted liability. Eighth, the legislation calling for the Plan emphasizes the importance of looking beyond levees, especially upstream to flood bypasses and floodplain storage that expands the capacity of the flood system. Finally, the legislation recognizes the increased level of risk (or at least liability) from floodplain development and requires better flood protection for future approvals of development in the floodplain.

In essence, the 2007 legislation shifted attention away from trying to “control” floods by building levees to keep flood waters in the river to looking at how the entire system can work to “protect” the Central Valley from damaging impacts of floods. Floods will continue to arrive, perhaps more often and more intensely with climate change. California now seeks opportunities to manage floodwaters, not just get rid of them.

The New Challenge of Climate Change

The Plan has developed at a time when the science of climate change is developing with respect to flood risks. DWR has periodically released reports on the state of climate change science and the anticipated impacts on water resources in California. In July 2006, DWR released a report called Progress on Incorporating Climate Change into Planning and Management of California’s Water Resources. The report reviewed potential impacts of selected climate change scenarios on water resources in California, including operation of the State Water Project and Central Valley Project, Sacramento-San Joaquin Delta water quality, and flood management. In January 2009, DWR released The State of Climate Change Science for Water Resources Operations, Planning and Management. In these two reports, DWR projected the following effects of climate change and impacts on water resources generally: (1) a reduction in California’s average annual snowpack (up to 90 percent by 2100 in the Northern Sierra) could lead to a loss of average annual water storage in the snowpack and increased challenges for reservoir management and flood protection; (2) changes in precipitation could lead to increased storm intensity, increased potential for flooding, and increased potential for droughts; (3) sea-level rise of up to 55 inches could cause inundation of coastal marshes and estuaries, salinity intrusion into the Delta, and increased potential for Delta levee failure; (4) increased water temperatures could impact listed and endangered aquatic species and lead to increased problems with foreign invasive species; (5) water shortages could occur in Central Valley Project reservoirs during droughts; (6) general shifts in seasonal and annual average runoff could result in considerable impacts (positive or negative depending on the scenario) to delivery capabilities of the Central Valley Project and State Water Projects; and (6) annual runoff concentrated more in winter months with more variability and greater extremes.

The Plan focuses on the three major categories of potential climate change effects that relate to flood management: changes in precipitation and runoff patterns, sea-level rise, and economic development. Specifically, with respect to precipitation and runoff, the Plan provides that

[i]ncreased temperatures may alter precipitation and runoff patterns, such as a rise in snow-line elevations, earlier snowmelt occurrence, more precipitation falling as rain instead of snow, and reductions in the volume of overall snowpack…. The combination of earlier snowmelt and shifts from snowfall to rainfall seem likely to increase flood peak flows and flood volumes, which is likely to affect associated flood risk.

* * *

For reservoirs downstream from significant mountain snowpack, the resulting temporal shift in reservoir inflows could pose major challenges for managing flood storage capacity and water supply, particularly if reservoir operations are not modified to accommodate the new conditions.

DWR, Central Valley Flood Protection Plan, Appendix 8K, 1–6 to 1–7 (2011) (internal citations omitted).

The Plan also discusses sea-level rise and its potential impact on flood management:

Sea level rise is likely to produce more frequent and potentially more damaging floods increasing risks for those already at risk, and increasing the size of the coastal floodplain, placing new areas at risk. Increased risk of storm surge and flooding is expected to increase risks for California’s coastal residents and infrastructure, including wastewater treatment plants.

Id. at Appendix 8K, 108 (citations omitted).

On economic activities, the Plan notes the importance of agriculture to the state’s economy and the fact that much of the state’s $36 billion agricultural industry is concentrated in the Central Valley. Specifically, the Plan provides:

More frequent and larger flood events are likely to damage structures, threaten livestock, contaminate croplands, cause increased erosion and sedimentation, take croplands out of production for extended periods as fields dry and recover, threaten levees that protect cropland and, in conjunction with sea level rise, increase farmland vulnerability in coastal areas and the Delta.

Id. at Appendix 8K, 1–10.

The Plan’s appendix on climate change presents what it calls a unique approach to climate change analysis. The Plan notes that much current analysis of climate and water impacts address change in mean conditions (such as mean temperatures and average precipitation patterns), while impacts related to flooding “will result not from changes in averages, but from changes in local extremes.” Id. at Appendix 8K, 2–2. In other words, the Plan recognizes that climate change science with respect to impacts on flooding patterns is still developing. The Plan presents its own methodology that attempts to consider the effects of extreme precipitation and runoff events on flood management.

To some extent the Plan is inconsistent, but its overall approach is to provide for sufficient flexibility to address anticipated impacts of climate change. It is inconsistent by describing the effects of climate change on flooding patterns as uncertain on one hand but then making relatively sweeping statements about the anticipated effects on the other. Nevertheless, the Plan is flexible in that it provides for potentially wider bypasses to lower floodwater surface elevations and increase flow-carrying capacity. Also, the Plan contemplates changes in reservoir operations to provide flexibility and adaptability to changes in extreme flood events. Pursuant to the requirements of SB 5, the Plan will be updated every five years. Accordingly, the draft Plan states that the state is developing new hydrology that considers the potential impacts of climate change for inclusion in the 2017 Plan update.

DWR presented a draft Plan to the Central Valley Flood Protection Board (Board) for public hearings and review in December 2011. The Board must make a final decision as to adoption of the Plan by a July 1, 2012, deadline. This public review has included debate from a range of perspectives, including landowners now protected by levees, members of the agricultural community, and local officials. Among the primary issues of contention are whether it makes sense to limit development in the floodplain, whether the Plan does enough to preserve California’s agricultural and rural communities, and whether the Plan places too much importance on ecosystem restoration to the detriment of flood protection benefits.

Plan Adequacy to Prepare California for Extreme Floods

The Plan reflects a fundamental shift in the way California thinks about floods in the Central Valley. The problems the state is now trying to address are fundamentally different from the problems it faced a century ago. Back then, flood planners focused on how to protect agriculture and small towns from flooding—in a state where only six percent of its current population was directly affected by flooding. The objective was to reclaim the floodplains for economic development. Accordingly, the state’s flood board was called the “Reclamation Board” until 2007. At the time, the biggest threat to agriculture and the economy was sediment buildup, so flushing sediment downstream through narrow channels was the physical solution.

Today, the objectives of Central Valley flood planning have changed, in part, because flood control succeeded and the Central Valley developed. The Sacramento region alone has more people today—2.4 million—than the entire state did in 1910. The state’s emphasis, therefore, has shifted to protecting humans in the urban areas at the bottom of the system, while maintaining some level of protection for agriculture and small towns. Protecting urban areas, according to the state, will require expanding upstream capacity to divert flood flows off the river, at whatever levels climate change delivers. Climate change adds a significant dimension of uncertainty and makes flexible expansion of flood capacity that much more important.

The Plan addresses the uncertainty, over the long term, by accepting that California will not be able to build big enough or strong enough levees to withstand every conceivable storm scenario, but also by attempting to design the system for flexibility in the face of a changing climate in the future. A U.S. Geological Survey report (USGS Report) supports the Plan’s assumptions concerning the limitations of levees. See USGS, Overview of the ArkStorm Scenario, Open-File Report 2010–1312. The USGS Report describes a hypothetical storm, similar to the intense winter storms of 1861 and 1862, estimated to produce precipitation that amounts to between a 500-year and 1,000-year event. Id. at v. The USGS Report determines that extensive flooding would overwhelm the state’s flood-protection system: The storm would cause flooding 300 miles long and at least 20 miles wide, cause property damage exceeding $300 billion, and necessitate evacuation measures affecting 1.5 million residents. Id. The USGS Report concludes that overall losses would be in the range of $725 billion. Id.

There appears to be general agreement among most stakeholders that (1) flood protection/public safety is the proper primary goal of the Plan; (2) expanding the system’s capacity to divert flood flows is necessary to address the big-picture threats to the system; and (3) approaching the entire problem from a system perspective, as opposed to fighting floods on a mile-by-mile basis, is the right approach. As they say, however, the devil is in the details. Although the state has indicated that local agencies will play some role in developing regional plans and implementing projects, the state does intend to lead the planning efforts. There is concern among local flood agencies that the state is less equipped than local agencies to address the complexity of issues at the local level, such as concerns of property owners, interest groups, and local communities.

Another significant area of concern about the Plan relates to its potentially detrimental impacts on agriculture in the Central Valley. Levee improvements will result in the direct loss of farmland due to required maintenance corridors because the Plan contemplates significant expansion of the bypass system by converting up to 40,000 acres of land that is primarily in agricultural production to either floodway or modified agricultural production and because 25 percent of the new flooded areas would be converted to ecosystem uses. Farming interests view these aspects of the Plan as threats to the sustainability of agriculture as the dominant economic engine of the Central Valley. It should be noted that, over the last century, the state has invested billions of dollars in the Central Valley flood system and agriculture has benefitted substantially from those investments.

California’s shift in the direction of flexibility and focus on systemwide improvements to its flood control system is a necessary development in light of potentially extreme flood events in the future. It remains to be seen how the state will reconcile these important goals with (1) the concerns of local agencies, on whom the system depends for project implementation and funding, and (2) the inherent conflict between the need to expand bypasses for system flexibility and the viability of the critically important agricultural sector of the state’s economy.