Introduction to Temperate Floodplains

Floodplains are among the most dynamic, productive, diverse, and threatened ecosystems in the world (Tockner and Stanford 2002). Intact and restored floodplains generate major environmental benefits that provide significant support for local and regional economies, most notably through flood-risk management, fisheries, recreation, and seasonal agriculture (Postel and Carpenter 1997). Yet the flooding that defines floodplains — and drives their ecological productivity and diversity — is often viewed as a problem. Many floodplains sustain dense human populations and agriculture that is not compatible with inundation. Floods in industrialized countries are usually equated with disaster, prompting extensive construction projects, such as dams, levees, and channel straightening and dredging, to minimize flood impacts on the built landscape. Consequently, floodplains, particularly temperate floodplains in more developed countries, are among the most altered landscapes worldwide, most with ecosystems that are highly degraded (Tockner and Stanford 2002). Yet floodplains also present some of the best opportunities throughout the world for innovative management that reconciles human uses and environmental conservation.

In this book, we focus on floodplains in temperate regions of the world, interweaving floodplain science and management. We review fundamental processes that shape floodplains as biophysical systems and then consider new perspectives on how floodplains are managed. Thus, this book should be of interest to both scientists and managers. We strongly believe that the most promising future for temperate floodplains will arise through management solutions that allow them to function as dynamic ecosystems that are also productive and safe components of the human landscape. Achieving this future will depend on informed dialogue and collaboration among managers, scientists, and stakeholders. Our experiences with temperate floodplains lead us to believe that the long-term social and environmental sustainability of floodplains can best be guided by the interrelated concepts of "novel ecosystems" and "reconciliation ecology." Novel ecosystems are those that are highly altered by humans and often contain alien (nonnative) species, such that their current biota and physical structure may differ markedly from those of the ecosystems they replaced. Nevertheless, they may be functionally quite similar to the original ecosystems (Hobbs et al. 2009, 2013, 2014; Moyle 2013). Reconciliation ecology is the "science of inventing, establishing, and maintaining new habitats to conserve species diversity in places where people live, work, and play" (Rosenzweig 2003, p.7). Reconciliation ecology focuses on conservation of native biodiversity while accepting the reality that virtually all habitats have a strong human presence. Thus, built landscapes, such as cities or flood-management systems, can be treated as novel ecosystems into which biodiversity conservation and ecosystem services are actively integrated in order to provide multiple benefits to human society.

The concept of novel ecosystems is generally applicable to floodplains of large temperate rivers because virtually all are modified and/or regulated. While channels and floodplain surfaces continue to be reshaped by flood frequency, extent, and duration, these forces are modified in large part by the operation of dams, levees, and drainage systems. Even most "restored" floodplains have topography that has been highly altered by agriculture and are subject to inundation regimes managed by people. Species assemblages are often in continuous flux, responding to environmental change, land and water management, and invasions of non-native species.

Because of this ecological novelty and because large areas of temperate floodplains have been converted to other land uses, reconciliation ecology provides a constructive framework for moving forward with sustainable management of floodplains for multiple purposes, including flood-risk management and conservation. People are now an integral part of nearly all temperate floodplain ecosystems and, in the chapters ahead, we will illustrate how reconciled floodplain ecosystems can provide ecological and economic benefits.

The floodplains of California's great Central Valley (figure 1.1), which inspired this book, provide a particularly cogent example of the complex and evolving relationship between people and floodplains. These floodplains have undergone ecological loss and sweeping landscape change, but they also provide examples of paths toward a more sustainable future. We draw heavily on this well-studied landscape as its lessons can be applied globally and reinforced by examples from other floodplain systems.

In short, this book is about reconciling temperate floodplain ecosystems with society's demands on the environment; it is also about how people can work proactively with novel floodplain ecosystems to generate multiple benefits. Our goal is to demonstrate that maintaining native biodiversity and natural floodplain functions can be highly compatible with a broad range of societal objectives and expectations. Innovative management approaches can produce floodplains that provide habitat for some of the world's most spectacular fish and wildlife, while also protecting communities from floods and providing clean water and open space for people. Our ambitious objective is to have the concepts, examples, and syntheses of emerging ideas in this book serve as a foundation for effective and sustainable management of temperate floodplain systems. We hope that this foundation will also provide insight for global floodplain management beyond temperate regions.

GEOGRAPHIC SCOPE AND THE DEFINITION OF FLOODPLAINS

Strictly speaking, the floodplains covered in this book occur within temperate latitudes, defined as those between the Arctic Circle and the Tropic of Cancer and between the Antarctic Circle and the Tropic of Capricorn (spanning the latitudes of 23°26? to 66°34? both north and south). Mountain ranges and other continental features greatly affect river flows and temperature patterns, so floodplains vary in size, character, and importance throughout the temperate regions. We primarily focus on floodplains from North America, Eurasia, and Australia, with limited mention of those in southern South America and South Africa (figure 1.2).

Floodplains support complex physical, biological, and social systems (Naiman et al. 2005). They are created by interactions between flowing water and sediment, which influence, and are influenced by, physical structures and biological processes. Floodplains are optimal sites for farms and cities because of their proximity to water, low-gradient terrain, and rich organic soils. Urban and agricultural development on floodplains is often facilitated by infrastructure such as dams and levees.

Because of these numerous and interacting features, processes, and uses of floodplains, professionals of diverse disciplines study and manage floodplain systems. As a result, hydrologists, geomorphologists, ecologists, engineers, economists, planners, and policy makers have developed a variety of definitions of floodplains (Nanson and Croke 1992; Alexander and Marriot 1999). Below we review definitions from geomorphic, hydrologic, regulatory, and ecological perspectives. We then present an ecogeomorphic definition that integrates biological and physical concepts that will be used throughout this book.

Geomorphic

Although rivers and floodplains are often viewed as distinct, separate features, from a geomorphic perspective rivers and associated floodplains are single functional systems that move water, sediment, and organic material across the landscape. The geomorphic floodplain is a set of channels and surfaces that were created by, and continue to be shaped by, fluvial processes and are inundated with some frequency. Processes that build and rework floodplains include overbank deposition and erosion, lateral migration, and channel avulsion (described in chapter 3). The resulting floodplains have substrate derived from interactions among flow, sediment, and vegetation (Nanson and Croke 1992; Knighton 1998, Wohl 2010).

Hydrologic/Hydraulic

A hydraulic floodplain is a surface inundated by a defined flood recurrence interval (i.e., has a certain probability of flooding in any given year) from yearly (Wolman and Leopold 1957) to much longer periods, often 100 years or greater (Alexander and Marriot 1999). This definition is often used by hydrologists and engineers (Nanson and Croke 1992). The term "active floodplain" has been used to describe the area inundated by floods with a broad range of recurrence intervals, whether or not that surface is alluvial. An alluvial surface is created by the actions of the river, such as sediment deposition (Alexander and Marriot 1999).

Regulatory

The regulatory floodplain in the United States is essentially the hydraulic floodplain corresponding to what is termed a "100-year flood," or the surface that has the statistical probability of 1% of being inundated in any given year, based on past records of flooding. The term "100-year flood" has led to widespread confusion among citizens, who often assume that, following such an event, they are safe for the next century (Mount 1995). Because of the potential for misunderstanding, the US Army Corps of Engineers has begun to use the terminology of "percent exceedance" (Hickey et al. 2002) with the "1% exceedance floodplain" corresponding to the "100-year floodplain." However, the term "100-year flood" is well established among planning professionals and in the popular media, so it remains in common usage. The designation of the 100-year floodplain still strongly influences patterns of development in the United States; within designated 100-year floodplains, owners of structures with federally backed mortgages are required to hold a flood insurance policy (Sheaffer et al. 2002; Pinter 2005). Such policies are administered by the National Flood Insurance Program (NFIP), which is part of the Federal Emergency Management Agency (FEMA). The process of delineating the 100-year floodplain involves considerable uncertainty due to extrapolations from short hydrological records (often less than 100 years) and because runoff patterns have high variability, which is affected by climate change and land-use changes (Milly et al. 2008). Because each new flood provides a new data point, the mapped extent of the 100-year floodplain often expands after a major flood event, as it did in regions of California following floods in 1986 (Mount 1995) and again in 1997. The extent of the 100-year floodplain will likely expand in many regions as climate change increases flood magnitudes (Kundzewicz et al. 2008).

Ecological

From an ecological perspective, floodplains are low-lying areas along rivers that support ecosystems with biological and physical characteristics that are strongly influenced by the dynamic hydrograph of the river. Floodplains serve as an interface between aquatic and terrestrial environments during inundation events (Seavy et al. 2009). Floodplain ecosystems have often been considered equivalent to riparian wetlands or "bottomland" forests, or as ecosystems that incorporate riparian forests, wetlands, and floodplain waters such as oxbow lakes (Mitsch and Gosselink 2000; Naiman et al. 2005). Temperate floodplains often support a high percentage of the biodiversity in a region, especially of vertebrates (Hauer et al., 2016).

Much of the ecological literature on floodplains emphasizes areas that are inundated frequently and for long durations, as is the case for many large river systems in the tropics (Bayley 1989; Junk et al. 1989). The timing and predictability of flooding are also important ecological variables that shape floodplain ecosystems and the extent to which species have specific adaptations for using floodplains. Flooding can be fairly predictable, as in California's seasonal climate, or highly erratic, as in southern Australia's aseasonal climate. This has led Winemiller (2004) to place floodplain ecosystems into three basic categories: tropical seasonal, temperate seasonal, and temperate aseasonal.

Tropical seasonal floodplain ecosystems have massive rainy-season floods that inundate huge tracts of forest and other habitat on a predictable basis for months at a time. Classic examples are the extensive floodplains of the Mekong and Amazon rivers and their tributaries. Due to our emphasis on temperate systems, we do not focus on this floodplain type, even though it encompasses some of the largest and most studied floodplain ecosystems (see box 1.1).

Temperate seasonal floodplain ecosystems make up the majority of floodplains in temperate regions. In these systems, flooding tends to predictably occur within a specific season although the exact timing and extent of flooding may vary tremendously among years and between river systems. In more northern areas, seasonality is driven by snowmelt, while in Mediterranean regions seasonality is driven more by rainfall, sometimes in combination with snowmelt from mountains.

Temperate a seasonal floodplain ecosystems are found in areas where large rain storms can occur during any month of the year, but often have long periods between events. Winemiller (2004) notes that the Brazos River, Texas, United States, and the Murray-Darling system in Australia are examples of a seasonal floodplain rivers.

Ecogeomorphic

In this book, we use a definition of floodplain that combines geomorphic and ecological definitions. We consider floodplains to be features that are formed and influenced by rivers and upon which biophysical processes and ecosystems operate. The geomorphic processes influence the characteristics of the ecosystem present, while ecosystem characteristics alter the way the geomorphic processes work. We highlight seasonal floodplains along lowland rivers, generally those with broad alluvial valleys. Such floodplains often feature highly productive and diverse ecosystems and they tend to be the focus of management, both for values that depend on flooding, such as fish and waterfowl, and for keeping people and crops safe from flooding.

In the spirit of novel ecosystems and reconciliation ecology, we consider ecogeomorphic floodplains to encompass all areas that are periodically inundated by river flows, regardless of return intervals, including areas that would have been flooded prior to human changes. These areas include seasonal floodplain surfaces and features such as wetlands, oxbow lakes, ponds, and side channels as well as habitats that are often considered to be terrestrial, such as agricultural fields and riparian forests, which are periodically inundated at low frequencies. This definition approximates the definition of an active floodplain (Alexander and Marriot 1999) and is a broader definition than has usually been applied to floodplains (Bay Institute 1998; Mitsch and Gosselink 2000).

ON THE NEED TO STUDY FLOODPLAIN ECOSYSTEMS

Floodplain ecosystems received very limited study prior to the 1970s, with scientific interest confined to the relatively small community of geomorphologists — researchers who had long known that any inclusive definition of a river channel should include the floodplain (e.g., Wolman and Leopold 1957). In contrast, a large body of floodplain research related to hydrology was generated by engineers looking for ways to prevent flood waters from reaching floodplains and damaging built and agricultural landscapes. These engineering studies began in Europe in the seventeenth and eighteenth centuries and were conducted to inform the construction of flood-control infrastructure, and the Chinese have been studying and building projects to manage floods for thousands of years (Wu et al. 2016). Due to this infrastructure, most large rivers in temperate latitudes were disconnected from their floodplains prior to broader scientific study (Benke 1990; Bayley 1991; Dynesius and Nilsson 1994). Commonly, channels were straightened and armored and multichannel systems were converted to single channels, with little attention paid to effects on the biota (Ward and Stanford 1995; Brown 1998; Ward et al. 2001; Florsheim and Mount 2002, 2003; figure 1.3).