3 How Stream Restoration Was Born, and What Came of It

Streams and rivers play an outsized environmental role. In a purely geographic sense, they are relatively rare; only a small portion of any landscape is a flowing aquatic system. But these fluvial systems have very broad impacts. They are crucial not only for aquatic organisms, but also for many terrestrial organisms that need water to survive. Further, rivers and streams are central to physical and chemical processes across watersheds because of their role as integrators: everything flows or is washed downhill, converging into them.

Both physical and chemical properties have been essential features of the economic and political landscape in the United States, and thus have been subjected to tremendous anthropogenic impacts. European settlers used rivers as their primary arteries of transportation, harvesting timber from hillsides and routing it along rivers downstream to ports. To quicken the downstream journey of their timber and to provide easier passage for their rafts, canoes, flatboats, and steamboats, settlers removed gravel bars, snags, logjams, and boulders and straightened river channels, rapidly clearing out almost every navigable waterway throughout the Northeast and upper Midwest. Dam building came next to provide power for the mills that converted raw timber and grain into lumber and flour, which were easier to ship and brought higher prices. Most colonial villages from New England to Georgia were built around a grist mill, and the eighteenth- and nineteenth-century industrialization of America depended on waterpower from mill dams peppering nearly every river, stream, and tributary.1

When the federal government entered the picture through its water resource management agencies, particularly the Corps of Engineers and the Bureau of Reclamation, the scale of river change increased dramatically. Along the lower Mississippi River, the Corps of Engineers straightened the channel to speed floods and riverboat traffic: the length of this massive river was reduced by 150 miles in the first half of the twentieth century. Other agencies were just as busy straightening and dredging rivers; one nationwide estimate put total stream and river modification at 200,000 miles, or about 6 percent of the total length of stream miles in the United States.2 This does not even account for the dams built nationwide: by the close of the twentieth century there were as many as 85,000 dams greater than ten feet tall, and perhaps as many as two million dams overall if you included all the other smaller structures that blocked streams, sloughs, and tributaries across the nation.3 These structures radically changed the character of the rivers and streams on which they were located, both upstream and downstream.

There were equally profound changes to the water that flowed in those rivers. America was long an agricultural society and still is in many regions, such as the Midwest. Food production has come at the expense of rivers and streams. The explosion of forest clearing and row crop agriculture triggered vast soil erosion and the muddying of tributaries; the most common pollutant in twenty-first-century streams and rivers is sediment. As America industrialized, rivers were befouled by the decaying cattle carcasses from meat processors adjacent to the waterways in Chicago, the oil refining wastes of heavy industry that lined the Cuyahoga River of Cleveland, and the burgeoning petrochemical empire along the Mississippi River near Saint Louis. Even where these industrial pollutants were not spewing into waterways, the exploding population’s municipal waste flowed directly into rivers for decades. Today, the runoff of fertilizers from farms and suburban lawns flows untreated into America’s streams and rivers.

Rivers were, and are, the gutters of society. And just as individuals and groups oppose this environmental damage today, scientists, engineers, and fishermen in earlier times sought out ways and opportunities to undo the damage being done to rivers.

Early Forms of Stream Restoration

Environmental restoration is based on the idea that we can undo the environmental damage we have caused.4 This compelling vision of humans as beneficial members of ecosystems has made restoration an increasingly central part of the global environmental movement since the mid-1970s. Restoration has far deeper roots, however, stretching back at least to the late 1800s in the United States.5 Some of the earliest efforts to restore aquatic ecosystems took place in the backwoods of the Catskill Mountains of upstate New York in the 1870s under the auspices of fly fishermen in search of trout. Rivers in the Catskills were isolated from much of the industrialization occurring elsewhere, so they were not befouled by chemical waste or urban sewage. They were physically degraded, however: the local timber industry not only had removed trees, but also had manipulated the streams and rivers to get the timber out by straightening and clearing their channels.

This physical degradation, along with the loss of riparian (i.e., river-bordering) vegetation, caught the attention of fishermen worried by declining trout populations, who then wrote about trout stream degradation in popular magazines such as Forest and Stream. John Spencer Van Cleef was particularly prolific in both his fishing and his writing about how the physical form of rivers affected trout. He had a clear view of what was degrading trout streams: “I have become satisfied that the destruction of the trees bordering on these streams and the changed condition of the banks produced thereby, has resulted in the destruction of the natural harbors or hiding places of the trout.” The importance of Van Cleef’s explanation was that it made clear that this habitat destruction was reversible: “I believe it possible to restore most of our streams . . . especially when they are under the control of clubs or associations who can make the effort.”6

Van Cleef and a few other trout fishermen had intuitively combined several important ideas that would shape stream restoration for the next century and a half. They linked the physical condition of a river with the presence or absence of trout, and in so doing viewed the bed of a river, the vegetation along its banks, and how water flowed along the channel as malleable. If humans could restore these characteristics of a stream to a semblance of their natural condition, early restoration practitioners assumed, the ecosystem—particularly the trout—should follow. This was one of the earliest rationales for stream restoration, and it has been given a rather flippant slogan drawn from the 1980s movie Field of Dreams: build it and they will come.

Over the next decade, private fishing clubs were organized in the Catskills and other areas outside large eastern cities. Beyond acquiring and preserving land, the clubs focused their attention on restoring trout streams. In the 1870s, Van Cleef developed methods for manipulating the features of streams to increase trout habitat, the implementation of which became a hobby for club members between their fishing outings. They used the materials available in the Catskills—logs and boulders—and placed them in particular configurations and locations with the hope of attracting trout to their stream reaches again. They built miniature dams—weirs—across channels to create pools upstream and fast-flowing riffles downstream. They brought in larger rocks to build vanes in the river—partially submerged barriers to flow, like hydraulic speedbumps. They tried different shapes and combinations of structures in an effort to make things just right for trout. While their efforts were novel, they were largely ineffective, in part from a simple limitation of scale: a few hundred feet of restored stream makes very little difference in a watershed of a hundred square miles.

The increase in scale required to make stream restoration at least theoretically effective was made possible when emerging research from a group of fisheries scientists at the University of Michigan was combined with the very activist role of the federal government during the Great Depression. In the early 1930s, University of Michigan researchers began a concerted focus on improving habitat conditions for trout to restore the fishes’ ability to naturally reproduce. Some of the earliest formalized, scientifically based guidelines for restoring streams were developed from their work, encapsulated in a nondescript pamphlet titled Methods for the Improvement of Michigan Trout Streams.7 It eventually became a cornerstone document of habitat improvement for stream restoration through much of the mid-twentieth century.8

The specific restoration techniques the scientists from the Michigan School advocated were quite similar to what had been done by the hobbyists in the Catskills a half century earlier. Working within the existing stream channel, they constructed small dams and weirs, and placed boulders and logs in particular shapes and spacings in attempts to create more varied hydraulic habitat within the confines of the existing river channel. Importantly, the sinuosity (i.e., the layout, or course of the river) and banks were treated as fixed aspects, while the in-channel aspects of the river were considered malleable targets of intervention. Over the second quarter of the twentieth century, river scientists at the University of Michigan published dozens of articles in prominent journals. These studies formed the basis from which longer-term evaluations and guidelines were developed, and as students matriculated from Michigan and populated government agency positions elsewhere, their ideas about the utility of trout stream restoration as a management technique diffused nationally as well.9

An accident of timing allowed the work of these scientists to jump scale from the limited interventions of Catskill fishing clubs to wide distribution across the nation. Government work programs were a key feature of the early years of the Great Depression, and the people in those programs needed something to do. Small stream restoration was ideal for one of the primary New Deal work programs, the Civilian Conservation Corps (CCC), because it required very few technical skills, was done in rural areas of the United States where unemployment was high, and relied on locally available materials such as logs and boulders.10 No one knew whether the tactics from the University of Michigan would work, but the ecological risks paled in comparison to the potential economic and political benefits. During the early 1930s—the initial peak of the Great Depression—around thirty-one thousand structures were constructed on over four hundred mountain streams for the purpose of restoring habitat for trout. By 1936 the CCC had altered almost five thousand miles of streams.11

The efforts to restore streams, especially by employing instream hydraulic structures, continued on through the mid-twentieth century, with methods largely identical to those the CCC and Michigan scientists used during the 1930s. Restoration work was done, and well documented, in Michigan, South Dakota, Montana, Tennessee, New Mexico, and throughout the upper Midwest,12 and it had three key features. First, most of the work was in-channel restoration: the physical dimensions and shape of the river were left intact. Second, restoration was based on the assumption that desired ecological benefits, such as bigger trout populations, would follow from physical improvements. And third, even at the time there was no clear evidence that these techniques actually worked. Based on a thorough analysis of data collected at the restoration projects, there is no evidence that they had any positive ecological effects at all.13

Reconfiguring Channels

All of the efforts of the Catskill fishermen, the Michigan scientists, and the CCC workers were concentrated down in the stream; they worked within the existing streambanks and streambed, and built structures to modify how water flowed within the existing channel. And yet many of the problems they wanted to address stemmed from physical alterations of the channel through straightening or the construction of dams and levees. Restorationists could see clearly that something was physically wrong with the shape (or morphology) of these rivers and streams, but they were unsure how to fix it because there was no way to identify what an appropriate, healthy morphology for any given reach would be. Looking at historical photos or maps to see what had been there before was almost always irrelevant: changes in land use throughout a watershed inevitably affect the amount of water and sediment a river or stream needs to carry. What restorationists needed was a way to determine what the shape of a given reach of stream channel should look like now in order to successfully transport the water and sediment it currently received. In the 1980s, there were significant leaps in restoration science, engineering, and audacity that together finally made this possible, transforming reconfiguration of stream channels from a dream into a common practice.

The leap in science was the result of a paradigm change in fluvial geomorphology—the study of river form and processes. Up through the middle of the twentieth century, geomorphology was a primarily descriptive discipline. Geomorphologists were narrators of process, describing how landscapes evolved. Geomorphology texts and journal publications were dense with photographs and a few maps, in combination with long explanations of the formation of particular types of landscape in different regions of the world.

And then came the quants. Starting in the 1950s, a group of geomorphologists began quantifying landforms rather than describing them. Articles and texts from geomorphologists had fewer and fewer pictures, and more and more graphs and tables of data. Instead of summary statements, equations summarized concepts. Fluvial geomorphologists laid out the processes that formed rivers in equations of force and motion, and distilled the shapes of rivers themselves down to a surprisingly tidy set of quantifiable relations, referred to as hydraulic geometry equations.14 Studies in the United States and abroad demonstrated that once coefficients and exponents were adjusted for the particular conditions in a given watershed, hydraulic geometry equations could be used to describe almost any river or stream within it. This gave river morphology a sense of rule-like behavior, and even a sense of predictability.

Hydraulic geometry equations link the morphology of a river directly to the amount of water it carries or the size of watershed that it drains. Key features of a river—such as width and depth—are functions (dependent variables) of an independent variable such as discharge or Drainage Area, and are expressed as a set of equations:

Width = a(DA)^b

Depth = c(DA)^f

The power of this new way of thinking was that it converted rivers into a series of equations, and thus made stream morphology calculable. Over the next half century, river scientists fleshed out these numeric relations and patterns with increasing subtlety and precision. Hydraulic geometry equations went from generalizations to predictions that river engineers could deploy to design flood outlets or move a river under a new bridge overpass.15

For stream restoration practitioners, the hydraulic geometry equations were nothing short of revolutionary. In theory, once discharge was quantified or drainage area measured on a map, any river or stream whose form had been altered by human actions could be redesigned to fit its current conditions. Starting in the early 1980s, groups and individuals in different parts of the United States used these equations to restore streams through full channel reconstruction rather than just manipulating instream conditions.

This departure really was startling. There were no direct precedents for the channel reconfiguration work these restorationists were doing, and the potential for catastrophic failure was all too real. As described by one of the pioneers of channel reconstruction, Greg Koonce, “When we started Inter-Fluve [a consulting firm focused on stream restoration], we just started building pools and riffles into existing streams with a backhoe.” But then they moved toward changing the entire river, which for them was a very big step (figure 3.3). Koonce recalled: “The first re-meander we did was totally scary.”16

Koonce and other early practitioners of channel reconfiguration had the hydraulic geometry equations, but not much in the way of guidance for how to make them work in practice. They had to employ a trial-and-error approach to see what worked and what didn’t. As consultant Jim MacBroom explained, when he started trying to figure out how to rework channel morphology, “there really wasn’t what we think of as design guides or manuals on how to design a channel for something other than a rigid boundary, prismatic type of geometry.”17 Consultants like Koonce and MacBroom were pioneers, trying out different approaches for locations and conditions all over the United States: Greg Koonce worked in Oregon and Montana; Jeff Haltiner worked in California and the Pacific Northwest; Jim MacBroom and Robbin Sotir worked throughout New England and the Southeast, respectively. Other early practitioners were part of the vanguard of federal agencies focused on restoration: Dave Rosgen (an employee of the U.S. Forest Service during this period) worked in Idaho, Montana, and Colorado, and Doug Shields (Corps of Engineers and U.S. Department of Agriculture) in Mississippi. There were a surprisingly limited number of academics who were part of this initial vanguard of geomorphologists actually implementing stream restoration designs, including Jim Gore (University of Wyoming) and Matt Kondolf (University of California); the academic research community more broadly would not engage stream restoration practice for another decade.18

Initially, these early channel reconfiguration practitioners worked largely in isolation in part because they were located in different regions of the country, and in part because the auspices under which they worked were wildly different. For example, consultant Greg Koonce began his work restoring streams on private land, bankrolled by wealthy landowners or land developers putting in hobby ranches. He based his early work on developing designs for streams that would improve trout fishing, going so far as to advertise in fishing magazines that targeted this clientele (figure 3.4). Doug Shields, in contrast, worked initially for the Corps of Engineers and then the U.S. Department of Agriculture’s Agricultural Research Service. He was effectively doing extension work for farmers in the Deep South, trying to develop lower-cost practices to stabilize rapidly eroding streams. Dave Rosgen, a U.S. Forest Service employee at the time, was trying to minimize damage to streams from logging. Each of these restorationists only gradually became aware that they had company in their startling departures from past restoration practice.

But from this early starting point, interest in stream restoration really took off, and demand for designs outstripped the supply of people who had actually done reconfiguration projects. The uniqueness of these individuals’ experiences and the growing demand for stream restoration design meant that their knowledge was valuable, and was one of the only sources of stream restoration know-how. Many in this early cohort developed short courses to teach others how to do stream restoration. A majority of stream restorationists in the United States who came along after this cohort, even decades later, got their initial education through these short courses. This includes many twenty-first-century stream restoration designers, but also many federal and state agency employees who have gone on to regulate stream restoration efforts. Indeed, in the absence of formal academic certifications in stream restoration, short courses played an outsized role in the intellectual establishment of the restoration economy.

Among this early group of stream reconfiguration practitioners, by far the most influential was Dave Rosgen, a former U.S. Forest Service employee turned consultant. A protégé of Luna Leopold (who had developed the hydraulic geometry equations in the 1950s), Rosgen quickly became the most nationally recognized figure in the stream restoration movement, including being profiled in National Geographic. His well-known approach to restoration—Natural Channel Design—became the most practiced method of stream restoration design in the United States.19 Although Rosgen completed many restoration projects, his primary influence on stream restoration as a discipline and industry came through his role as an instructor. Rosgen’s series of short courses continues to be a primary source of stream restoration education in the United States; by one estimate, as of 2012, approximately two-thirds of stream restoration practitioners and regulators in this country had taken at least one of his Natural Channel Design short courses.20

In addition to short courses, several influential textbooks and handbooks were published between 1996 and 1998, including Rosgen’s Applied Geomorphology, Brookes and Shields’s River Channel Restoration, the Federal Interagency Stream Restoration Working Group handbook,21 and Ann Riley’s Restoring Streams in Cities. Some of these books reached very wide audiences. However, Rosgen’s book was the one considered synonymous with “how to do restoration.” The U.S. Forest Service, for example, purchased copies of Rosgen’s book for every hydrologist at the agency.22

Because of their large role in restoration, it is important to understand what was covered in these books and short courses, and what was not. Perhaps surprisingly, given the common image of restoration as intended to produce clean water and happy fish, the courses and texts focused almost exclusively on geomorphology. The texts were riddled with equations and specific best practices for anything related to the physical form and function of a stream or river, but offered little more than generalities for anything related to water chemistry or ecology. The instructors for the short courses and the authors of the texts listed earlier were similarly almost exclusively geomorphologists and hydraulic engineers. Improvements to water chemistry and ecology were called out as goals, but the assumption was that reconfiguring channel morphology was what was needed to reach those goals. Many of the short courses, including Rosgen’s, did not address how to understand (much less improve) nutrient pollution, or even the basics of aquatic ecology.23

Just as important, Rosgen’s 1996 textbook, which supplanted most of the other geomorphic texts for restoration practitioners, and his short courses emphasized channel stability above all else (further discussion to follow). While some early channel reconfiguration practitioners emphasized bioengineering approaches, with deformable materials such as vegetation shaping a reconfigured channel that could move or change over time, the more common approach deployed in the field and taught in short courses made liberal use of logs and boulders in the same way that hydraulic engineers used concrete and riprap: to make sure a redesigned channel stayed where it had been put.

Thus, this fairly rapid development of stream restoration had some clear points of kinship with Van Cleef’s nineteenth-century efforts, and the Michigan School’s work in the early twentieth century. The scope of intervention may have shifted dramatically, but the restorationists’ reliance on natural yet nondeformable materials, and their assumption that getting the physical form of the channel right was both necessary and sufficient for the recovery of ecological communities and processes, were all too familiar; many of the types of stream intervention (when not completely remeandering a stream) were eerily familiar recreations of the approaches and techniques used seventy-five years earlier (figure 3.5).

Stabilizing Unstable Hydroscapes

During the mid- and late 1990s, channel reconfiguration projects became increasingly common geographically and numerically as the restoration economy developed; hundreds of projects were built each year, hundreds of millions of dollars were spent, and a community of practitioners became established. Stream restoration no longer was limited to isolated patches around where early leaders happened to be located: it proliferated and expanded into most regions of the United States. Restoration also expanded out of rural or isolated places and into cities and suburbs through the work of urban stream restoration pioneers such as Ann Riley in the San Francisco Bay Area.

Via the short courses described earlier, the stream restoration community drew in people with training in fisheries biology, geology, geography, forestry, landscape architecture, and engineering, among other fields. Yet despite this diversity in backgrounds, the majority of practitioners employed a basic approach to stream restoration drawn from Rosgen’s Natural Channel Design short courses. That approach was to develop hydraulic geometry equations for a region and then use those equations to design restored streams based on the assumptions that a) restored streams will be stable, and b) that stability will bring improvement in water quality and ecology (the “build it and they will come” approach pioneered more than a century earlier by Van Cleef and his colleagues).

The development of the restoration economy in any given region typically went something like this. First, a critical mass of restoration practitioners and the agency staff who were funding or reviewing their projects (or both) attended Rosgen’s short courses. Thus inspired, they worked together to survey reference reaches: relatively untouched reaches of streams that, importantly, were believed to be physically stable. The goal of surveying reference reaches was to develop the backbone of technical tools for Natural Channel Design-based stream restoration: regional curves. Regional curves are the hydraulic geometry equations particular to a given physiographic region from which new channel designs could be derived. Finally, these regional curves would be adopted as the basis for appropriate design parameters for stable channels.

Why the emphasis on stability? After all, dynamism is one of the defining characteristics of rivers and streams. Healthy, functional streams move over time both across the landscape via meandering along with constant reworking of sediment through erosion and deposition. Why has stream restoration focused so heavily on stability? The answer to that question is complicated, and requires some backtracking into the different ways stability is interpreted in scientific and practical contexts.

The form of any given reach of a stream is determined by its balance of inputs and outputs of water and sediment; if the amount of water or sediment shifts, the shape of the river will adjust to those changes. If there is a long-term balance in this water and sediment budget, then the river is considered to be in equilibrium. Importantly, a river in equilibrium can and does move. The migration of rivers shapes the basic topography of landscapes, such as flat valleys or oxbow lakes (former channels now cut off from the river), and many aspects of river ecosystems are in fact dependent on this natural shifting of rivers.24

But other kinds of channel movement can be symptoms of negative human impacts rather than of natural dynamism. Rivers and streams can adjust dramatically in response to changes in upstream or downstream conditions. For example, when the land use in the surrounding watershed changes, water and sediment loads are often significantly affected. Deforestation increases both the amount of water and the sediment load entering rivers and streams when trees and their soil-binding roots are removed, causing fluvial systems to change their shape until they return to equilibrium with the new conditions in their watershed. Urbanization causes similarly profound changes: converting land surfaces from forests (or agriculture) to paved and guttered hillslopes dramatically increases the amount of water in a river while decreasing the amount of sediment. Rivers typically respond to urbanization by eroding their bed and banks—becoming deeper and wider. They might also change their sinuosity by becoming straighter.

The key thing to note is that rivers and streams can take years or even decades to adjust to changing watershed conditions and reach a new equilibrium. Complicating this, urbanization and other land-use changes are rarely one-time, single-place events. Development typically occurs in different parts of a watershed over time; first road and sewer building, then neighborhoods in one area, and then a shopping mall elsewhere, for instance. If a river is adjusting to chronically changing conditions, it will be constantly out of equilibrium. But if watershed conditions remain static for a period of time (i.e., if changes associated with urbanization cease for years, or even decades), the river channel should come into equilibrium with these new conditions on its own.

The question, then, is how to tell the difference between the natural dynamism of a healthy stream channel versus problematic, chronic instability. One of the key ways in which the hydraulic geometry equations came to be used in stream restoration was to determine which kind of channel movement was in operation in a particular reach, a use far different from any the equations’ developers ever intended. In restorationists’ hands, hydraulic geometry equations were used to compare the form of a channel to what would be expected given the regional curves; if there was a mismatch, then that site was presumed to be unstable and thus in need of restoration, regardless of whether the channel had the capacity to adjust to its new conditions on its own.

Eventually, indicators of stream channel instability became synonymous with poor stream restoration design in the Natural Channel Design community (which was itself synonymous with stream restoration across much of the United States). Given that the hydraulic geometry equations allowed a practitioner to design a stream that should be stable, instability meant that the designer got something wrong. And one of the key indicators that was used in assessing instability (and equilibrium) was river bank erosion. Thus, if a restored stream had any type of bank erosion, even erosion caused by a stream restoring its equilibrium with its changed watershed, then that project was considered a failure. Stability came to mean being fixed, and not adjusting at all.

There is some controversy over whether Rosgen intended his students to interpret his calls for stability so literally.25 Regardless, it’s quite clear that they have. As Kris Vyverberg, a state agency staff member, ruefully noted, “Rosgen folks are talking about stability stability. . . . I can’t speak to Dave’s perspective on stability, but as . . . [Natural Channel Design] is applied in the field, people seem to believe stability means the channel won’t move.”26 This contradiction between the goals of restoration practitioners and the most accepted principles of river science caused conflicts between the restoration science and practice communities that persist to this day, and have shaped the development of stream mitigation banking, as we describe later.27

Restoring Streams without Fish

The fixation on stream channel stability was in part the result of another broad trend in stream restoration generally: through the 1990s and on into the twenty-first century, stream restoration was fully in the realm of the physical sciences and engineering, with little engagement from biological or chemical science. As mentioned earlier, hydrologists, geomorphologists, and civil engineers were the primary personnel developing the techniques, standards, and methodologies; they were the ones teaching the short courses, writing the books and handbooks, and doing the actual design work. Indeed, whereas the earliest days of stream restoration in the Catskills and Michigan were led by efforts at restoring trout, the burgeoning late-twentieth-century restoration economy focused on restoring stable channels. There was some justification for this in that much scientific work had demonstrated that ecosystems were negatively impacted by unstable stream morphology. However, the underlying focus on geomorphic controls of stability permeated the practice, and eventually came to dominate. This was not to say that geomorphologists overtly asserted that stable, meandering channels were both necessary and sufficient; they generally only left it at necessary. But they did not argue that it was insufficient.

Over the first decade of the twenty-first century, ecologists began more heavily engaging in stream restoration and their work showed that channel reconfiguration may (sometimes) be necessary, but that it was nowhere near sufficient. Systematic studies showed that restoration efforts were not producing positive impacts, whether for biotic communities or water quality. This was perhaps best summarized in an analysis of seventy-eight restoration projects, which concluded: “Most projects were successful in enhancing physical habitat heterogeneity [i.e., physical conditions]; however only two showed statistically significant increases in biodiversity rendering them more similar to reference reaches or sites.”28 That is, of seventy-eight different projects, only two worked to at least some degree. This was an almost identical endpoint reached by the work proliferated (and monitored) by the University of Michigan program in the 1930s, along with subsequent programs: there was little to no evidence that restoration projects produced the ecological improvements they were intended to provide.29

Restoration in the Late Twentieth Century

By the mid-2010s, the U.S. stream restoration economy had three overarching characteristics. The first was the presumption that restored streams should be stable; the approach most used to achieve stability was Natural Channel Design and the application of hydraulic geometry equations. The second characteristic was the increasing realization that restoration attempts were not generating ecological or water quality benefits to degraded streams. And third was the increasing distrust between practitioners of streams restoration (whether in the design or regulatory communities) and the scientists who studied streams because of their heated disagreements over Natural Channel Design.

But by then geomorphology-based stream restoration was central to the restoration economy and firmly incorporated into environmental policy and regulation whose outcomes were at best uncertain. And one key driver of that economy was the rise in markets for the ecosystem services from streams under the U.S. Clean Water Act.