Nestled in the foothills above Stanford University lies Searsville Dam and, behind it, a lake by the same name. Since 1892, the dam has served as the headwaters of San Francisquito Creek, a meandering waterway that flows generally downhill and east all the way to the flat expanse of San Francisco Bay. The creek has the distinction of being the last riparian free-flowing urban creek on the southern San Francisco Bay Peninsula, and it is the last remaining run in that area of the federally threatened steelhead trout.
In the foothills, the creek is how we might like to imagine it. Part of Stanford’s Jasper Ridge Biological Preserve, the creek flows largely unmolested, and so does the things a stream tends to do. It twists and bends and contracts and expands, patiently winding its way through meadows of tall grasses punctuated by blue and coastal live oaks. These surroundings graciously absorb the creek’s overflows in the wetter months and retain precious moisture in the dryer ones. With a reliable source of water, bobcats, gray foxes, western gray and California ground squirrels, black-tailed deer, scrub jays, red-legged frogs, tiger salamanders, and countless other animals make the foothills, and in some cases the creek itself, their home.
As the creek moves farther east into its longest stretch, it changes character. Or rather, we have changed its character. Despite its humble name, the San Francisquito Creek drains a 45-square-mile watershed and serves as the dividing line between the urbanized parts of San Mateo and Santa Clara Counties, and between five cities, in the heart of Silicon Valley. In the wet months, the creek collects and conveys rain and storm water runoff from impervious driveways and streets, a sometimes-bulging volume of water forced into an ever-narrowing channel. To serve these purposes, we have squeezed and straightened the creek and taken away many of its overflow areas. This engineering has the effect of hastening the creek’s flow, accelerating erosion, and increasing turbidity.
In 1998, a 50-year flood caused the creek to ignore its human confines and reestablish, close to the Bay, what used to be its natural delta. The creek’s brief rebellion led to $28 million in damages, mostly in the poorer areas of Silicon Valley. See U.S. Army Corps of Engineers (Corps of Engineers), San Francisquito Creek Feasibility Phase Project Management Plan, at 10 (Sept. 27, 2005); Santa Clara Valley Water District, “San Francisquito Creek 1998 Flood” (video uploaded Feb. 4, 2009 (https://www.youtube.com/watch?v=2KTNaQmeCxY)). The creek also has been listed as impaired for Total Maximum Daily Load (TMDL), under section 303(d) of the Clean Water Act. Spikes in these sediment loads come from erosion, but because of the creek’s high velocity, they do not serve (and, in fact, undermine) the creek’s historic role in gradually restoring sediment to the land in the southern San Francisco Bay.
In many ways, San Francisquito Creek is among the healthier urban waterways. Because neighboring San Mateo and Santa Clara Counties could never agree on a plan for channelizing the river, much of the creek has remained in a seminatural state well downstream of its headwaters. See American Rivers, America’s Most Endangered Rivers 2014, at 11 (2014). Many urban waterways are much more heavily engineered, having been so squeezed and restrained that they are uninhabitable save by a courageous duck or two. But, like all urban waterways, San Francisquito Creek has significant problems; indeed, the creek was listed as number five (out of ten) on the list of America’s Most Endangered Rivers in 2014. Id.; see also San Francisco Estuary Institute, Historical Ecology of Lower San Francisquito Creek–Phase 1 (Mar. 2009) (describing changes in habitat and hydrology in the creek’s lower reaches).
To address these problems, the San Francisquito Creek Joint Powers Authority has been working with the U.S. Geological Survey, the Corps of Engineers, and other stakeholders for two decades. Their goal is to integrate science with “community values” to reduce flooding and protect the creek, including by restoring at least some of the creek’s natural features. That is, to let the creek be a creek. This effort, and many like it, reflect a growing (though not universal) consensus that our traditional approach of urbanizing urban waterways not only does not work but actually makes things worse. See, e.g., Cal. Water Code § 7048 (California legislature finding that “the protection, restoration, and enhancement of urban creek channels provide potential benefits to the state by combining an effective and efficient means of flood damage reduction with the preservation and enhancement of natural environmental values”); Craig Arnold, et al., The Social-Ecological Resilience of an Eastern Urban-Suburban Watershed: The Anacostia River Basin, 51 Idaho L. Rev. 29, 82–84 (2014); Ron Love, Daylighting Salt Lake’s City Creek: An Urban River Unentombed, 35 Golden Gate U.L. Rev. 343, 348–50, 369–72 (Spring 2005).
Ever since I read John McPhee’s The Control of Nature (1979), I have been fascinated by our engineered conversion of natural streams into highly controlled flood outlets. As an attorney I have worked on matters of agricultural runoff, salmon restoration, and water quality, all of which have involved or at least implicated urban waterway regulation and health. And as a longtime resident of the areas around San Francisquito Creek, I have come to wonder: Have we put San Francisquito Creek and the thousands, likely millions, of urban waterways like it to their highest and best use? What legal obstacles or gaps impede more effective waterway design and management? And how can we better support the unconventional approaches of the San Francisquito Creek Joint Powers Authority and similar groups, all of which involve “rewilding”—relinquishing control over and restoring natural features to—our urban waterways?
The Legal Gaps
Our most powerful water laws—the Clean Water Act and its state equivalents—are aimed primarily at water quality. See generally 33 U.S.C. §§ 1251–1388 et seq.; Cal. Water Code §§ 13000–13951 (Porter-Cologne Act). In California, as in most states, the state adopts water quality plans to establish beneficial uses, sets minimum water quality standards to protect those beneficial uses, issues permits to implement the standards, and enforces against violations of those permits. See 33 U.S.C. §§ 1313(c), 1319, 1342; Cal. Water Code §§ 13050(f), 13260–13276, 13300–13365.
Yet the Clean Water Act, the Porter-Cologne Act, and similar laws allow us to manage urban waterways only indirectly. When a waterway fails to meet the water quality standard for a given pollutant, the U.S. Environmental Protection Agency (EPA) or a state or local agency locates the source of the pollution, adjusts permits or issues new ones, and/or brings an enforcement action. This approach may be sufficient where water quality is the only thing we care about, or where a highly controlled waterway makes the most sense. (Indeed, turning nonpoint source pollution like storm water into a channelized point source can be an effective way to improve water quality. See Robin L. Craig, Protecting Oceans from Urban Storm Water Runoff, 21 Nat. Resources & Env’t 36, 37–38 (Spring 2007).) But, by and large, the focus under our water quality laws is on what can be done to most quickly achieve compliance with water quality standards. Rarely do regulatory agencies have the resources or authority to consider or impose alternative remedies that involve a fundamental shift in how we design and manage an entire waterway or watershed to address a larger suite of problems, such as flooding and ecosystem loss, in addition to water quality.
Of course, the Clean Water Act is not limited to traditional pollutants. Section 404 of the act (33 U.S.C. § 1344; see also id. § 1362(6)), in conjunction with section 301 (33 U.S.C. § 1311), requires permits to discharge materials in or near jurisdictional waters and wetlands of the United States. Many states have their own permitting requirements for such discharges into state waters, which they enforce through water quality certifications issued under section 401 (33 U.S.C. § 1341). Permits under these provisions more directly affect the fate of urban waterways, thereby presenting some of the best opportunities to improve their condition, including through compensatory mitigation (more on this later).
However, the section 404 program and similar state programs have real limits. They are biased toward traditional engineering, even when protecting natural stream features may be the best solution for avoiding floods and improving water quality. In addition, small streams, including those that comprise our urban waterways, “defy the boundaries between water and land,” raising jurisdictional questions that thrust legislatures and courts into “an uncomfortable zone for environmental law.” Dave Owen, Little Streams and Legal Transformations, 2017 Utah L. Rev. 1, 14 (2017); see also id. at 44 (explaining that Clean Water Act’s provisions “do not prohibit developing nonwetland riparian areas adjacent to streams, [or] building parking lots and roofs throughout a watershed, and thus altering flow patterns until streams are overloaded with flood flows and pollutants.” (citations omitted)). These jurisdictional questions may be even thornier following the current administration’s decision to rescind the 2015 Waters of the United States Rule.
Section 404 permits and other discretionary authorizations often require consultation or mitigation under the Endangered Species Act and its state equivalents. These requirements can provide additional opportunities for creatively managing urban waterways. Species determined to be threatened or endangered under 16 U.S.C. § 1533, such as the threatened steelhead in San Francisquito Creek, are afforded significant protections that can either help save natural stream features from damage or require their restoration. Designation of critical habitat, also under section 1533, likewise can be effective in achieving those ends (primarily through restrictions and compensatory mitigation). However, the Endangered Species Act is of limited consequence, and, thus, of limited help, in urban areas, which tend to have few threatened or endangered species and little critical habitat. Even where a listed species or its critical habitat exists, like the steelhead trout and portions of San Francisquito Creek, the Endangered Species Act serves to protect that species and its habitat from specific harmful actions; it does not provide an easy mechanism for holistically redesigning a waterway. Like pollutants, species are incomplete proxies for urban waterway health.
Some of the laws with the most influence over how we design and manage urban waterways are those adopted by local jurisdictions. Through general plans and zoning ordinances, cities and counties exercise direct control over the uses to which we put property, including waterways and parcels adjacent to waterways. The ability to plan in advance and on a jurisdiction-wide scale should (and sometimes does) allow local jurisdictions to identify and implement better waterway designs. But in practice this ability is constrained by the law of vested property rights and takings, inadequate funding and planning, jurisdictional turf battles, and a general (and understandable) presumption in favor of development for tax revenues. Simply put, “preserving space for streams means not using that space for houses, roads, or valley fills.” Owen, Little Streams and Legal Transformations, 2017 Utah L. Rev. at 13.
The National Environmental Policy Act (NEPA) and state “little NEPAs” provide mechanisms for studying urban waterways and the changes we make to them. But even where these laws require that significant impacts be mitigated, they work only haphazardly to improve waterway health. Still other statutes, like the Rivers and Harbors Act, the Flood Control Act, and the Water Resources Development Act, are periodically amended or reissued to provide funding for specific public works projects. For example, the city of Napa, California, and the Corps of Engineers recently used authorizations under the Flood Control Act to complete an $18.5 million “bypass” project, consisting of engineered wetlands and other natural features, to reduce perpetual flooding along a portion of the Napa River. See Flood Control Act of 1965, Pub. L. No. 89-298, 79 Stat. 1073, § 204 (Oct. 27, 1965); Pub. L. No. 94-587, 90 Stat. 2917, § 136 (Oct. 22, 1976); see also Love, 35 Golden Gate U. L. Rev. at 369–73 (discussing restoration of City Creek in Salt Lake City under Water Resources Development Act of 1992). These targeted statutes can be effective in improving urban waterway design and management, but they require specific attention and appropriation from Congress.
Working with the Existing System
For all their limitations, our existing laws have been used creatively to reimagine and remake urban waterways. For example, the San Francisquito Creek Joint Powers Authority is working with the Corps of Engineers to craft a feasibility study for reducing flooding in key parts of San Francisquito Creek. The Joint Powers Authority and the Corps are undertaking their efforts pursuant to the Water Resources Development Act of 1999, which directs that “studies and projects shall emphasize, to the maximum extent practicable and appropriate, nonstructural approaches to preventing or reducing flood damages.” Pub. L. No. 106-53, 113 Stat. 289, § 212(b)(3) (1999); see also id. § 212(e)(6) (specifically authorizing flood protection and riverine restoration studies and projects for San Francisquito Creek). Thus, while the existing proposals consist in part of typical flood control measures (levees, culverts, concrete channelization), they also include wetlands, wide bends, marshplain terraces, and other natural features. See Corps of Engineers, San Francisquito Creek Feasibility Phase Project Management Plan, at 11–15 (Sept. 27, 2005); Philip Williams & Associates, Ltd., San Francisquito Creek Flood Reduction, at 8–31 (July 17, 2009).
Similar approaches are being employed in other urban waterways. In the city of Napa, the confluence of high tides, and urbanized waterways have colluded during periods of heavy rain to flood parts of the city, including downtown. The result has been 21 major floods since 1862, including the “flood of the century,” which in 1986 killed three people, displaced 5,000 others, and caused $100 million in damage. See Napa County Flood Control and Water Conservation District, “The History of Floods and the Creation of a New Project” (2009); Kevin Courtney, Flood of the Century: 1986 Death, Destruction Led to Reform along Napa River, Napa Valley Register (Feb. 20, 2011).
For years the embattled Napa County Flood Control and Water Conservation District and the Corps of Engineers sought to relieve the perpetual flooding in one area, the Oxbow District, with standard riprap and concrete culverts. After a concerted campaign for another approach, the agencies opted instead for the solution discussed above: engineered wetlands and other improvements that would allow water to overflow from one side of the river to the other in the wet months and serve as a waterfront promenade and recreational area in the dry ones. This $18.5 million “Oxbow Bypass,” funded by Congress and a local half-cent sales tax, is part of a larger effort to create a 7-mile-long “living river” that provides flood control for 3,000 vulnerable properties and valuable habitat for fish and wildlife. See John King, “Living River” Rejuvenates Napa, Brings Needed Flood Control, San Francisco Chronicle (Apr. 30, 2017); Napa County Flood Control and Water Conservation District, Napa River/Napa Creek Flood Protection Project—Documents (https://www.countyofnapa.org/1083/Napa-River-Creek-Flood-Project). The bypass had the opportunity to, and did, prove its mettle in heavy rains this past winter. See Barry Eberling, Napa’s River Bypass Floods for the Second Time in Three Days, Napa Valley Register (Jan. 11, 2017).
Litigation also has served an important role in changing how cities and counties design and manage urban waterways. In Marin County, California, a nonprofit group successfully challenged the county’s 2007 general plan update under the California Environmental Quality Act, the state’s version of NEPA. The county’s general plan had established “stream conservation areas” to protect streams from the effects of urban development, including Lagunitas Creek, one of the last runs of the imperiled Central Coast coho salmon and steelhead trout. In 2007, the county approved an update and setback ordinance that allowed certain types of development in the conservation areas. In their challenge, the nonprofit argued, and the California Court of Appeal agreed, that the county’s environmental review for the 2007 update and ordinance failed to account for the cumulative effects on waterways of likely development within the conservation areas. Salmon Protection & Watershed Network v. Cty. of Marin, Case No. A137062, 2014 WL 845416, at *7 (Cal. App. 1 Dist. Apr. 3, 2014). The case forced Marin County to evaluate more thoroughly whether and how its stream conservation areas would work.
These success stories are meaningful but too hard won and too isolated. Ask the planners in Napa how many studies they prepared or how long it took to prepare them, or the San Francisquito Joint Powers Authority how many years of planning and fundraising they have left. The lawyers in the Marin County case (my former colleagues) gladly would have traded their win for a plan that balanced development and stream protection from the start. The point is, as creatively as we use our existing laws, they simply are not designed to optimize urban waterways. How can we do better?
More Effective Approaches
Compensatory mitigation is emerging as the most commonly used mechanism for restoring natural features to urban waterways, and for good reason. Generally rooted in sections 401 and 404 of the Clean Water Act and in the Endangered Species Act, compensatory mitigation programs combine concrete restoration with flexibility. New(ish) regulations from EPA and the Corps of Engineers that focus on replacing “lost functions and services” have expanded the role that compensatory mitigation can play in restoring urban waterways. See 33 C.F.R. § 332.3(b)(1) (2008); J. B. Ruhl, James Salzman & Iris Goodman, Implementing the New Ecosystem Services Mandate of the Section 404 Compensatory Mitigation Program—A Catalyst for Advancing Science and Policy, 38 Stetson L. Rev. 251, 252 (Winter 2009). However, like carbon offsets, compensatory mitigation can localize impacts and outsource benefits by allowing mitigation elsewhere. Compensatory mitigation also raises questions about whether restoring one stretch of creek really can make up for paving over another. And compensatory mitigation can be a narrow, project-by-project solution that misses opportunities to address problems on a larger scale. See also Owen, Little Streams and Legal Transformations, 2017 Utah L. Rev. at 40–41 (discussing critiques of compensatory mitigation programs for stream protection).
Local land-use planning can overcome some of these deficits. In the Marin County case, the nonprofit groups that challenged the county’s 2007 update and their lawyers knew that the key to protecting Lagunitas Creek was restoring the creek’s natural features. These included wetlands, bends, obstructions, vegetative buffers—everything that allowed the stream to spread out and slow down. Unfortunately, these are the features that are most in conflict with property owners who want to build or expand new houses, driveways, culverts, and patios. The plaintiffs’ win was no guarantee that the tension would be resolved in favor of a new approach—the county still has the leeway to approve roughly the same level of development as in the 2007 update, so long as that conclusion is better supported. (The county’s review remains ongoing.)
A better outcome would have been a land-use planning process that expressly recognized the benefits of protecting Lagunitas Creek and other urban streams from the outset. Of course, that solution is unworkable for waterways in already-urbanized areas, except in rare cases. Existing cities must instead use their general plans and zoning ordinances to reconceive of urban waterways as living ecosystems and, where possible, adopt programs for returning already-developed areas to a stream system. These programs are often easiest to undertake as part of redevelopment projects and are usually funded through grants or taxes (as in the Napa River program). Funding shortfalls and property rights issues will remain challenges, but these types of programs can be effective when combined with forward-thinking planning. The Supreme Court’s recent regulatory takings decision in Murr v. Wisconsin, 137 S. Ct. 1933 (2017), may give local jurisdictions more leeway in managing urban waterways.
Nonlegal solutions can help, too. Public-private partnerships (P3s) are an innovative mechanism for financing and completing what are otherwise prohibitively expensive public works projects. For example, in the Chesapeake Bay Region, EPA, the Maryland Department of the Environment, Prince George’s County, and a private company are implementing a “community-based” P3 for replacing much of the county’s impermeable surfaces with permeable surfaces and water filtration devices. The $100–200 million project is expected to provide jobs and significantly reduce pollutant loading in the Bay. See EPA Region 3, Water Protection Division, Community-Based Public-Private Partnerships (CBP3s) and Alternative Market-based Tools for Integrated Green Stormwater Infrastructure: A Guide for Local Governments, v–viii, 70–79 (Apr. 2015).
Unfortunately, P3s work only where a private entity can secure a portion of a guaranteed revenue stream (usually fees or bonds) and where they reduce the total cost of ownership of a green infrastructure project over its lifetime. It is unclear whether urban waterway restoration projects can satisfy those criteria. Critics of P3s also point out that they remain unproven, that even partial privatization of public services can relieve governments of their obligations, and that community and corporate interests often do not align. See, e.g., Hilary Wainwright, Public Services International & The Transnational Institute, “The Tragedy of the Private, the Potential of the Public,” at 10, 14–15, 17, 20 (2014); Yin Wang, Review: A Broken Fantasy of Public-Private Partnerships, 69 Pub. Admin. R. 4, 779–82 (Jul./Aug. 2009). Still, P3s may provide a useful, nonregulatory model for creating collaboration among multiple stakeholders and funding expensive urban waterway projects.
Ultimately, in the absence of new laws that provide for holistic approaches to urban waterway management, the most tangible solution probably lies in better integrating the laws and tools we already have. Local jurisdictions must have mechanisms for coordinating with federal, state, and local agencies on planning and environmental reviews. These entities, in turn, must collectively decide how to design and manage urban waterways programmatically—for water quality, endangered species, and local land-use purposes. There must be more opportunities for conducting and sharing research by and among scientists, planners, and regulators. We need more government programs, like California’s Urban Streams Restoration Program (see Cal. Water Code § 7048), and more creative market-driven approaches, like P3s, for project financing. And all these components must be integrated and streamlined to be less unwieldy and more responsive to emerging science and values. Voluntary conservation plans for urban waterways, perhaps modeled after a few successful endangered species habitat conservation plans, or other integrated approaches might best be able to serve these multiple aims.
Admittedly, not all scientists and lawyers think that it makes sense to restore urban waterways to a more natural state or that the legal solutions for doing so are feasible. For example, Martin W. Doyle and F. Douglas Shields, both at the Nicholas School of the Environment at Duke University, conclude that “[t]he balance of published evidence suggests that current practices of stream restoration . . . cannot be assumed to provide demonstrable physical, chemical, or biological functional improvements.” Compensatory Mitigation for Streams Under the Clean Water Act: Reassessing Science and Redirecting Policy, 48 J. Am. Water Resources Ass’n 494, 500 (2012). In fact, they argue, “[r]estoration in urban watersheds may be prohibitively difficult,” as “[m]easurable and meaningful levels of attenuation of floods, reduction of sediment loads, and improvement of water quality appear to be unrealistic and ungrounded goals for the predominant practice and scale of stream restoration.” Id. at 498; see also Christy R. Violin, et al., Effects of urbanization and urban stream restoration on the physical and biological structure of stream ecosystems, 21 Ecological Applications 1932, 1932–49 (2011) (explaining that restoration can be ineffective without careful attention to studies and stream characteristics).
This skepticism is useful. It reminds us that we should “rewild” urban waterways only where doing so provides tangible benefits and that we must always try to sync science with policy. But that caveat only underscores the need for environmental laws that promote scientific research and provide flexible tools for adapting to, and coexisting with, our changing environment. Urban waterways present an opportunity to do more with our laws, to show that we can live in our cities with more than squirrels and a few birds. We can disagree over how to accomplish that goal (or even whether it is the right goal), but, as environmental lawyers, we have an important role to play in that conversation.
Throughout The Control of Nature, John McPhee ponders why the Corps of Engineers moves heaven and earth to constrain the flow of the Atchafalaya River near New Orleans and why residents of Los Angeles raise every possible defense against the constant risk of flood, fire, and debris flows from the neighboring San Gabriel Mountains. His perceptive guess, tucked away towards the end of the book, is that we “would rather defy nature than live without it.”
So it is with our urban waterways. We work to control and confine them to reduce flooding and provide predictability, but over time that approach tends to backfire. Straitjacketing our urban waterways causes increased erosion, turbidity, and flooding, which are matched only by decreased recreation, water quality, and ecosystem function. A recent report concluded that nearly half of the salmon, steelhead, and trout species in California will be extinct in 50 years if nothing is done to improve water quality, protect wetlands and stream habitat, and fight climate change. See Dr. Peter B. Moyle, et al., State of the Salmonids II: Fish in Hot Water, California Trout & Center for Watershed Sciences, University of California-Davis (2017).
We have other, better options. Biologists, hydrologists, geomorphologists, and others are increasingly telling us that nature had it right all along—that allowing our highly controlled urban waterways to behave more like their freer nonurban counterparts can more effectively (and sometimes more cheaply) support flood protection, drinking water supply, habitat, recreation, and other uses. Our existing laws, or at least our existing ways of using them, can be better employed to serve that purpose. If we think creatively and comprehensively about how we design and manage our urban waterways, we may be able to live with nature without defying it.