Kate Howe

Preserving biodiversity is one of the central goals of ecologists and conservation biologists today. As human population increases, habitat destruction increases as well, jeopardizing more and more species. Restoration and reclamation ecologists have developed many creative techniques to help reverse some of this destruction. One technique that is widely used for improving the quality and variety of stream habitat and subsequently increasing local biodiversity is the construction of artificial riffles and pools.

River ecosystems are some of the most altered and degraded ecosystems in North America. A recent report asserted that only about two percent of the stream mileage of ecological importance in the United States (not including Alaska) remains unaltered by human-induced changes (Palmer 1993). Human alterations of rivers include the building of dams, channelization, pollution, development in riparian areas, and increased farming and logging along rivers (Palmer 1993). These modifications have significantly altered the chemical and physical make-up of rivers. The impact on freshwater flora and fauna has been significant.

Published statistics on the state of the nation’s rivers are disheartening. Approximately half of American rivers do not pass water quality standards (Doppelt et al. 1993). Between 70 and 90 % of riparian vegetation has been lost or degraded as a result of human impacts. In addition, water flow has been altered in 70% of American rivers (Doppelt et al. 1993). Furthermore, a mere 0.3 percent of the total stream mileage in the United States is currently protected under the National Wild and Scenic Rivers System (Palmer 1993). Because rivers are important ecologically, economically, and aesthetically, many people have begun to consider the reclamation and restoration of streams and rivers to reverse some of the damage we have inflicted on these ecosystems.

Many rivers naturally have sinuous channels that consist of both pools and riffles (Keller 1992). Pools are deep areas that have relatively slow water flow. They are created by erosional processes in the outer part of river bends, with point bars often occurring on the inner parts of the curves. Riffles, on the other hand, are shallow areas characterized by faster flow and larger substrate size that occur in the straighter reaches of meandering rivers (Keller 1992). This sequence of pools and riffles provides a heterogeneous physical environment that is utilized by many different types of organisms. Pools and riffles provide refuge from high velocity waters and extreme temperatures, spawning sites for salmonid fishes, and attachment sites for benthic invertebrates and plants (Gore and Shields 1995). The complexity of pool-and-riffle sequences offers the wide variety of habitat types needed to support a diverse lotic community.

One of the major effects humans have had on riverine ecosystems is the homogenization of streambeds. Channelization has increased erosion and reduced sediment deposition in many rivers and streams (Henderson 1986). Often the streambed is further degraded by dredging to make streams and rivers more navigable (Smith et al. 1990). The natural pool-riffle sequence of many rivers has been lost. Altered streambeds with sandy bottoms are habitable by some species of benthic organisms, but the more complex the habitat, the greater the number of species that are able to live there. The greatest diversity and highest productivity of bottom-dwelling organisms is found in stream riffles with a substrate composed of medium-sized cobble and gravel (Smith et al. 1990). Because different fish species require different types of prey, a diverse community of benthic invertebrates provides food for a wide variety of fish species (Gore and Shields 1995). Therefore, loss of pool-and-riffle habitats can greatly decrease the overall diversity of a stream or river.

In order to improve the quality of stream habitat, conservationists have developed several river restoration and reclamation techniques. One of the most widely documented methods is the construction of artificial riffles. Because this technique is so widely used, it is unclear where it originated. Artificial riffles have been created in streams and rivers all around the country, including Mississippi, Ohio, Oregon, Washington, North Carolina, and California (Gore and Shields 1995, Edwards et al. 1984, Frissell and Nawa 1982, Jewell 1981, Mesick 1995). They are also widely used in Europe, especially in Germany, Denmark, and the United Kingdom (Kondolf 1996, Brookes 1990).

The extent of pool and riffle restoration in streams depends on the degree of degradation in a stream and the goals of the restoration project. The general method used involves both excavation of pools and simultaneous construction of riffles by covering earthen fill with a combination of boulders, cobble, and gravel (Edwards et al. 1984, Elliot and Mason 1985). In some projects, woody debris is added to pools to provide cover for fish and invertebrates (Mesick 1995). There is some discrepancy over whether or not gravel should be cleaned and sorted before it is added to the streambed. Elliot and Mason (1985) suggested that gravel should be cleaned in order to avoid adding additional silt to the riverbed. Mesick (1995), alternatively, reported that salmon usually do not spawn in gravel beds until the have been "seasoned" for a few years. He suggests that using unsorted gravel provides a more natural substrate mix that reduces the "seasoning" time required before the restored habitat is functional for salmon spawning.

Several alternative structures are used to improve streambed complexity. One technique that is frequently mentioned in the literature is strategic placement of large logs across the river or stream (Gore and Shields 1995, Mesick 1995). These logs alter water flow on a small scale and recreate and maintain pools and riffles. A major drawback of log weirs is that they obstruct the waterways, making them virtually impassable to boats (Gore and Shields 1995). As a result, they are impractical in larger rivers that are commonly used for navigation.

Another common method of stream habitat enhancement is the installation of stone-filled wire mesh baskets, called gabions, across streambeds (Henderson, 1986). Water can flow through the gabions without washing the stones and gravel downstream. The flow conditions created by use of gabions may improve gravel beds to the point that they surpass the fish-spawning quality of those found in natural streams (House 1984). While this is useful if one is interested in creating a very productive salmon fishery, it does not fall under the category of historical restoration. This approach appears to have limited use to restorationists interested in improving the general hydrology and habitat characteristics of a stream.

Although construction of riffles seems to be an arduous and time-consuming task, these structures have been fairly successful in improving stream habitat quality. A comprehensive study on the Olentangy River at Columbus, Ohio examined the community composition of a natural, unaltered area, a channelized area in which artificial riffles and pools had been constructed, and a channelized area that had not been restored (Edwards et al. 1984). Both benthic invertebrates and fish were sampled over the course of the three-year study. The results showed that the diversity and biomass of both invertebrates and fish were highest in the natural area. The restored portion of the stream had fewer species and less biomass than the natural area, yet it was significantly more diverse and productive than the degraded, unmitigated area (Edwards et al. 1984).

Another example of a successful project is the historical reconstruction of two riffles in the Tombigbee River in Mississippi. The riffles were destroyed during the construction of a dam immediately upstream. As a result, the diversity of the macroinvertebrate fauna was decreased to only ten taxa (Gore and Shields 1995). Fewer than 15 fish species remained in the area downstream from the dam. Two years after the creation of the artificial riffles, 42 fish species were present in the restored area (Gore and Shields 1995). Just four months after the riffle construction, the number of macroinvertebrate taxa on the site had doubled. Habitat quality in this reach of the river was greatly improved by a limited amount of construction.

A more extensive river restoration project was implemented on the Wraysbury River in England. Although the river had been channelized several decades before, the area had regenerated a fairly dense cover of riparian vegetation (Kondolf 1996). In order to avoid stripping the area of vegetation, project managers decided not to completely restore the original channel; instead they widened it slightly, leaving some trees on an island in the center of the channel. They also used limestone blocks and planted willows to stabilize the banks and prevent further erosion. In addition, they constructed artificial pools and riffles and varied the elevation of the riverbed in order to increase the variety of habitat types available.

The scientists involved in this project found that they could accurately estimate biodiversity of the constructed habitat by measuring the complexity of the aquatic habitat created (Kondolf 1996). They measured complexity by calculating the coefficient of variation (an expression of the standard deviation as a percent of the mean) of depth for the created sites. The coefficient of variation is, in essence, a measure of the heterogeneity of habitat types. If the heterogeneity is high, biodiversity should be high as well. This method is useful for assessing the success of a restoration project. The author warned that use of the coefficient of variation is not a substitute for sampling the aquatic community, but it is a relatively quick and easy measure of restoration success. Because the purpose of constructing artificial riffles and pools is specifically to increase habitat diversity, success is usually determined as a function of the diversity resulting from the artificial habitat structures. There does not seem to be any mention in the literature of whether or not artificial pools and riffles differ chemically in any way from those in unaltered streams.

River restoration is more prevalent and widely accepted in Europe. While most river restoration projects in North America emphasize channel stabilization and flood control, European projects are more focused on improving or recreating habitat (Kondolf 1996). Although, construction of artificial riffles and pools has been going on for a few decades, it is still not as common in North American streams as it is in Europe. Furthermore, European stream restorations seem to be more committed to removing as many degrading influences from the ecosystem as possible, rather than simply looking for quick fixes to immediate problems.

Despite the successes described above, artificial pools and riffles aren’t always useful or long lasting. An extensive study of streams in the Pacific Northwest found a very high incidence of failure of artificial habitat structures (Frissell and Nawa 1992). Several types of artificial habitat structures were examined. The study showed that all large structures were equally likely to fail. Causes of failure included flooding, trampling by cattle, and shifting of channels. The highest rates of failure and impairment of these structures occurred in streams that drained extremely altered watersheds. Some damaged structures even acted as barriers to fish migration in Colorado and Oregon streams (Frissell and Nawa, 1992). In order for artificial habitat structures to be effective they must be designed carefully, with attention to the needs of resident and desired species and consideration of the prevailing physical factors in a particular river or stream. They must be adjusted to accommodate the specific requirements of an area.

Even when the hydraulic conditions and resident fauna of a river are considered in the design and implementation of artificial habitat structures, the structures often fail within a few years. Artificial habitat structures, in general, appear to serve more as "Band-Aids" for freshwater habitat improvement than as a long-term solution to the serious degradation of rivers. Small reaches of a river do not function in isolation. They are intimately linked with the entire river, as well as with the surrounding terrestrial communities. If stream restoration projects are to be successful, they will need to be implemented on a larger scale with attention to erosion control, restoration of channel gradient and sediment load, reduction of pollution, and revegetation of riparian corridors. Although artificial riffles and pools and other artificial habitat structures are often successful in improving habitat quality in the short term, those improvements will not be sustainable unless we are willing to commit to more extensive changes in the ways that we utilize our rivers and streams. We should begin to look toward long-term, watershed-level reclamation and habitat quality improvements.

 References

Brookes, A. 1990. Restoration and enhancement of engineered river channels: some European Experiences. Regulated Rivers: Research and Management 5:45-56.

Doppelt, B., M. Surlock, C. Frissell, and J. Karr. 1993. Entering the Watershed: A New Approach to Save America’s River Ecosystems. Island Press, Washington, D.C.

Edwards, C.J., B.L. Griswold, R.A. Tubb, E.C. Weber, and L.C. Woods. 1984. Mitigating Effects of artificial riffles and pools on the fauna of a channelized warm water stream. North American Journal of Fisheries Management 4: 194-203.

Elliot, S., and P.K. Mason. 1985. Salmon run. Landscape Architecture 75:82-85.

Frissell, C.A., and R.K. Nawa. 1992. Incidence and causes of physical failure of artificial habitat structures in streams of Western Oregon and Washingon. North American Journal of Fisheries Management 12: 182-197.

Gore, J.A., and F.D. Shields, Jr. 1995. Can large rivers be restored? BioScience 45: 142-152.

Henderson, J.E. 1986. Environmental designs for streambank protection projects. Water Resources Bulletin 22: 549-558.

House, R. 1984. Evalution of improvement techniques for salmonid spawning. Pages 5-11 in T.J. Hassler, ed. Proceedings: Pacific Northwest Stream Habitat Management Workshop. American Fisheries Society, Arcata, CA.

Jewell, L. 1988. Construction: alternatives to channelization. Landscape Architecture 71:488- 490.

Keller, E.A. 1992. Environmental Geology. Macmillan Publishing Company, New York.

Kondolf, G.M. 1996. A cross section of stream channel restoration. Journal of Soil and Water Conservation 51:119-125.

Mesick, C.F. 1995. Response of brown trout habitat to streamflow, temperature, and habitat restoration in a degraded stream. Rivers 5: 75-95.

Palmer, T. 1993. The Wild and Scenic Rivers of America. Island Press, Washington, D.C.

Smith, C.D., D.M. Harper, and P.J. Barham. 1990. Engineering operations and invertebrates: Linking hydrology with ecology. Regulated Rivers: Research and Management 5: 89-96.

Return to the Restoration & Reclamation Review Home Page