In immediate terms, the potential for streambank erosion is dictated by the stability of the banks and the erosivity of the flowing water. Streambanks can be protected or restored either by increasing resistance of the bank to erosion or by decreasing the energy of the water at the point of contact with the bank, for example by deflecting or interrupting flows (Henderson, 1986; Gregory and Stokoe, 1981, Heyer, 1991). Armoring the bank with mechanical works, revegetation, or a combination can stabilize the bank. Solutions for specific problems must consider the characteristics of the stream system as well as availability of materials and the cost. Designs should match the protection capability of the treatment to the erosion potential of each vertical stream zone. For example, riprap may be needed at the toe of slope to protect from undercutting combined with tree revetments to deflect flows and provide protection for live stakings that will develop permanent support. The growing body of research indicates that management techniques which emulate nature and work with natural stream processes are more successful and economical.

Sources of material for this section include Guidance Specifying Management Measures for Sources of Nonpoint Pollution In Coastal Waters produced by USEPA (1993), Riparian Restoration and Streamside Erosion Control Handbook written by Thompson and Green (1994), Reservoir Shoreline Revegetation Guidelines by Allen and Klimas (1986), and a number of other publications as noted in the text. Other sources of information on natural stream processes, the effects of human activities, and stabilization techniques include Cairns (1988), Gore (1985), Gray and Leiser (1982), Heede (1986), Klingeman and Bradley (1976), MNDNR (1991), MWCOG (1992), Orsborn and Anderson (1986), Roseboom (1993), and USACOE (1981).

BEST MANAGEMENT PRACTICES

MANAGEMENT TECHNIQUES

Livestock Exclusion: The fencing-out of livestock from areas, especially riparian areas, streambanks, and streambeds to provide for recovery of beneficial water quality/habitat functions. This practice is most applicable to riparian areas that have been severely affected by uncontrolled grazing. It allows for re-establishment of vegetative cover, which can secure banks, lower stream velocities, trap suspended sediments, increase meander, promote bank aggradation, and decrease downgradient erosion. All of these changes improve local and downstream water quality. Water quality can also be improved by lowering average temperatures through shading of banks and channels. These effects in turn cause attenuation of flood flows and increased recharge potential, which raises water tables, improves baseflow, and supports greater forage cover in uplands. This practice also eliminates direct pollutant loading from manure deposition in the stream. Where grazing is a significant factor, this alone is often sufficient to facilitate streambank recovery (Chaney et al. 1990, 1993; Platts, 1989).

Streambank Setbacks: Restriction of development activities within a specified distance of a streambank can prevent or minimize erosion and gully formation, thus minimizing sedimentation and associated nutrient enrichment downstream. This practice is distinct from, although potentially overlapping with, such BMPs as riparian buffers, floodplain preservation, and wetland preservation, as the areas involved are defined differently and as the primary goals of these other BMPs include other pollutant removal, preserving natural sediment and nutrient removal functions, and water quantity issues in addition to minimizing erosion. Setbacks can be based on anticipated channel migration widths or concentrated runoff diffusion requirements. Setbacks are most effective at reducing streambank erosion when used as part of a BMP system including requirements for structures to diffuse flows of stormwater discharges from upgradient development, measures to directly repair eroded streambanks, such as live stakes, fascines, cribwalls, gabions, and revetments, and a means of reducing the erosivity of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings.

SOIL BIOENGINEERING TECHNIQUES

Blanketing/Mattressing: A form of soil bioengineering which uses a blanket woven of live green cuttings and biodegradable fiber, geotextile, or wire, laid into a slight excavated depression in the bank, anchored with live or wooden stakes, and often punched through with live stakings. It is then covered with soil and watered repeatedly to fill voids with soil and to facilitate sprouting. Variations include a solid blanket of live branches held in place by an overlying staked wire mesh blanket and soil. Mattresses minimize sediment loading and associated nutrient enrichment impacts downstream by acting as a buffer, disrupting the force of incoming flows, creating turbulence, lowering water velocities, causing deposition of sediment, and protecting banks (USEPA, 1993. They are best used as part of a system which includes a component to deter undercutting at the bed/bank interface, such as riprap or gabions, a means of long-term streambank stabilization, such as live stakings, and a means of reducing the erosive potential of incoming flows at their source, such as routing runoff in grass swales, using detention ponds, and providing discharge spreader swales. See also Allen and Klimas (1986) and MWCOG (1992).

Branch Packing: A form of soil bioengineering which uses alternating tiers of live branch cuttings and compacted backfill to repair small localized slumps and holes in slopes. Cuttings are buried upright, angled roughly perpendicular to the slope face, and should penetrate the fill to undisturbed soil. Cuttings are alternated with wooden stakes sunk vertically several feet into undisturbed soil base and spaced every one to one and one-half feet up the backfill zone (USEPA, 1993). Branch packing minimizes sediment loading and associated nutrient enrichment impacts downstream by shoring up eroded areas of streambanks and providing for their long-term stabilization. It is best used as part of a system which includes a component to deter undercutting at the bed/bank interface, such as riprap or gabions, and a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings. See also MWCOG (1992).

Brush Layering: A form of soil bioengineering which uses live branch cuttings laid flat into small benches excavated in the slope face perpendicular to the slope contour. Cuttings taken from well-suited species, typically willow species or cottonwood, and properly installed, secured by live stakes angled into the slope face at intervals, will root and stabilize slopes. The goal is for natural recruitment to follow once slopes are secured. This stabilization method has the advantage of causing relatively little site disturbance. This technique differs from live fascines in the perpendicular orientation of rows as opposed to parallel fascine orientation. This up-and-down placement is intended to reinforce slopes in terms of mass stability, to protect against mass shearing and slumping. Brush layering reduces erosion of streambanks and other exposed slopes, minimizing sediment loading and associated nutrient enrichment impacts downgradient (USEPA, 1993). When applied to streambanks,it is best used as part of a BMP system which includes a component to deter undercutting at the bed/bank interface, such as riprap or gabions, a means of buffering the construction from erosive flows, such as tree revetments (which can actually accrue sediments), and a means of reducing the erosive potential of incoming flows at their source, such as routing runoff in grass swales, using detention ponds, and providing discharge spreader swales. When applied to exposed hillslopes, brush layering is best used as part of a BMP system which includes some means of temporarily securing remaining exposed soil from direct raindrop impact erosion, such as straw mulch, and measures to minimize upgradient runoff inflows (see runoff diversion).

Composite Revetment: Tiering of any number of materials up bank slopes, such as cement bags, stone, gravel, wire mesh blanketing, gabions, biodegradable fabrics, geotextile fabrics, tree revetments, and live staking of hydrophytic vegetation, matched to the erosive potential of the vertical streambank zones, can reduce streambank erosion and associated sediment and nutrient impacts downstream (Henderson, 1986). Cement burlap bags are often readily available, and can be customized to different degrees of permanence by adjusting the cement content (Keown, 1983). Wire mesh blanketing anchored through stone is used for high energy or steeply sloped situations (Keown, 1983; Schultze and Wilcox, 1985). A number of biodegradable fabrics, such as coir (coconut) fiber mats and rolls, are now commercially available, and can be staked out or wrapped around fill material on slopes and planted through with live stakings or rooted hydrophytes, providing stabilization and a rooting environment (Hoeger, 1992; Mikkelsen, 1993; Miller, 1992; Siegel, 1994). A number of synthetic geotextile fabrics are also available for similar applications. Live staking can be done directly into exposed banks, between sand cement bags, or through stone, mesh blanket, fabrics, or gabions. A combination of treatments should be selected as a system to target different aspects of bank erosion. For example, stone riprap at the bed/bank interface protects against undercutting and is most effective in combination with immediate means of buffering banks from erosive flows, such as tree revetments (which can actually accrue sediments), a long-term streambank stabilization component such as live stakings, and a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings. See also Allen and Klimas (1986).

Gabions: Traditionally, wire mesh baskets or cages filled with small rocks. They are wired together as building blocks in the construction of hardened streambank protections (USEPA, 1993). Gabions now vary in size and shape and can be used in the form of walls, terraces, or blankets. Pole cuttings can be inserted through gabions with the aid of a metal rod, and should be extended into the subsoil. A slight downstream angle is recommended to avoid shearing under high flows (Gregory and Stokoe, 1981; Schultze and Wilcox, 1985). Gabions provide a means of long-term streambank stabilization and are best used as part of a system which includes a component to deter undercutting at the bed/foundation interface, such as geotextile or biodegradable fabric or rock riprap, and a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings. See also MWCOG (1992).

Live Cribwall: A hollow, box-like interlocking arrangement of untreated log or timber members, this structure is filled with suitable backfill material and layers of live branch cuttings, which root inside the crib sturcture and extend into the slope. Alternating layers of fill and branch cuttings should be compacted to remove air pockets and withstand flow turbulence. Once the live cuttings root and become established, the subsequent vegetation gradually takes over the structural functions of the wood members (USEPA, 1993). Cribs may need to be countersunk into banks. Backfilling of the streambed with rock or cobbles to anchor the cribwalls into the banks should then occur (Kohnke and Boller, 1989). Live cribwalls provide a means of long-term streambank stabilization and are best used as part of a system which includes a component to deter undercutting at the bed/bank interface, such as rock riprap or gabions, and a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings. See also Allen and Klimas (1986) and MWCOG (1992).

Live Fascines (Wattling Bundles): A form of soil bioengineering which uses long bundles of live branch cuttings bound together in long rows and placed in shallow trenches following the contour on dry slopes, and at an angle on wet slopes. When cuttings are taken from well-suited species, typically willow species or cottonwood, and when properly installed, secured by live stakes angled into the slope face at intervals, they will root and quickly begin to stabilize slopes. The goal is for natural recruitment to follow once slopes are secured. This stabilization method has the advantage of causing relatively little site disturbance. Live fascines reduce erosion and shallow face sliding of streambanks and other exposed slopes, minimizing sediment loading and associated nutrient enrichment impacts downgradient (USEPA, 1993). They are best used as part of a BMP system which includes a component to deter undercutting at the bed/bank interface, such as riprap or gabions, a means of buffering the construction from erosive flows, such as tree revetments (which can actually accrue sediments), and a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings. See also Allen and Klimas (1986) and MWCOG (1992).

Live Staking: A form of soil bioengineering involving the planting of live cuttings from hydrophytic shrubs or trees along the streambank. Also known as woody cuttings, posts, poles, or stubs. As cuttings develop they protect streambanks from erosion, minimizing sediment and associated nutrient impacts downstream. Established cuttings also moderate bank and water temperatures, facilitate colonization of other species, and provide forage. Planting of dormant cuttings during the winter (Swenson and Mullins, 1985; Heyer, 1991; York, 1985; Kohnke and Boller, 1989), and scoring of bases followed by soaking in rooting hormone (York, 1985) increase likelihood of success. Vegetation selected should be able to: withstand the degree of anticipated inundation, provide year-round protection, have the capacity to become well-established under sometimes adverse soil conditions, and have root, stem, and branch systems capable of resisting erosive flows (Henderson, 1986). Locally available native species should be used (Schultze and Wilcox, 1985; Henderson, 1986). Other important factors include rapid initial growth, ability to reproduce, and resistance to disease and insects (Heyer, 1991). Plantings set with butts at anticipated growing season water table elevation (Swenson and Mullins, 1985) but not completely submerged year-round (Heyer, 1991) yield the best survival. All zones should be covered to ensure establishment in the most suitable one. Livestock must be excluded to achieve any measure of success (Swenson and Mullins, 1985). Species commonly used are several willow species, sycamore, and cottonwood (Henderson, 1986; Schultze and Wilcox, 1985; Swenson and Mullins, 1985). Stakings provide long-term streambank stabilization with delayed initial onset and are best used as part of a system which includes immediate means of buffering banks from erosive flows, such as tree revetments (which can actually accrue sediments), a component to deter undercutting at the bed/bank interface, such as riprap or gabions, and a means of reducing the erosivity of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings. See also Roseboom (1993) and MWCOG (1992).

Tree Revetment: A form of soil bioengineering which uses uprooted, live trees laid on their sides and secured to the bases of banks along eroded stream segments, tops pointed downstream and overlapped about 30 percent. Species used are those with abundant, dense branching to promote sediment trapping, and those which are decay-resistant. Eastern red cedar (Juniperus virginiana), for example, is generally preferred to hardwoods. The flexibility of live trees is recommended, limbless trunks should be removed, revetment ends should be anchored at stable points along the bank, and the anchoring system should be chosen according to the bank material to be stabilized and weight of the object to be anchored (Heyer, 1991). Tree size is important. the diameter of the tree's crown should be about two-thirds the height of the eroding bank, and trees greater than twenty feet tall are most econonmical for the majority of applications (Gough, 1991). As an alternative design, tree root wad revetments are created from uprooted live hardwood trees. These are cut into segments, the bottom segment containing the root mass is placed into an excavated hole in the bank trunk-first and protruding perpendicular to streamflow, the hole is backfilled, and remaining segments are used as footer logs to protect the base (McGill and Gracie, 1993). Tree revetments minimize sediment loading and associated nutrient enrichment impacts downstream by acting as a buffer, disrupting the force of incoming flows, creating turbulence, lowering water velocities, causing deposition of sediment, and protecting banks. They are best used as part of a system which includes a component to deter undercutting at the bed/bank interface, such as riprap or gabions, a means of long-term streambank stabilization, such as live stakings, and a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings.

Vegetative Cover: Vegetation is probably the most commonly used tool for streambank protection, particularly in small tributaries. Vegetation has the advantage of being self-propagating and self-repairing. Emergent vegetation provides two levels of protection. First, the root system helps to hold the bank soil together and increase overall bank stability by forming an interweaving network. Second, the stalks, stems, branches, and foliage provide resistance to streamflow, absorbing flow energy rather than deflecting it as hardened structures do or allowing it to erode soil particles. Vegetative cover above the waterline protects the banks from rainfall, runoff, and trampling forces. Bank vegetation has also been shown to provide protection from frost heave and subsequent bank failure (Bohn, 1989). Vegetation also provides water quality benefits by causing settling of particulates and sorbed pollutants, by providing a substrate for periphyton that removes nutrients directly from the water column, and by assimilating nutrients from the soil. Native species should be used, and their hydric affinities should be matched to the zones in which they are placed. Vegetation can be seeded above the waterline, typically spread in a matrix such as a hydromulch, or sprigged. Vegetative cover does not work on high-velocity streams. The range of successful applications can be expanded by using vegetation in conjnction with other BMPs, such as geogrid pavers, riprap, geotextile or biodegradable mats and rolls, as well as with the other "live" BMPs listed above.

HARDENED STRUCTURAL TECHNIQUES

Bank Armoring

Cribs and Gabions: Cribs are timber boxes built outward from the river bank and filled with sand and gravel. The boxes can be stacked along the bank and fastened together as building blocks in the construction of hardened streambank protections. Cribs are preferred where stone is not available or timber is cheap, and are typically used on smaller streams. Where stone is available on larger streams, gabions are often preferred. These are wire boxes into which stones may be placed, and are situated similarly to cribs. Gabions are a common substitute for riprap where smaller stones are available. Cribs and gabions are useful for protecting steep banks where scouring or undercutting are problems. The effectiveness of both techniques is improved when vegetation is incorporated (see SOIL BIOENGINEERING TECHNIQUES above).

Riprap: A layer of angular stone designed to protect and stabilize areas subject to erosion, slopes subject to seepage, or areas with poor soil structure. Riprap is used on streambanks where stream velocities are too great to successfully establish vegetative cover; on channel slopes and bottoms, stormwater structure inlets and outlets, slope drains, and shorelines. Stones should be of sufficient size to resist washing downstream. Larger rock should be placed at the bank bottom below baseflow elevation. Rock should be underlain by a filter blanket of gravel, sand and gravel, or synthetic material to prevent soil movement into or through the riprap. Riprap is well-suited for the high-energy zone of the streambank that is often submerged or that incurs direct streamflows, and when used at the bed/bank interface is effective at deterring undercutting. It is most effective when used as part of a system which includes a means of reducing the erosive potential of incoming flows at their source, such as eliminating overgrazing and soil compaction by cattle in agricultural settings or providing detention ponds for impervious surfaces in urban settings.

Embedded Flow Obstructions

All of these techniques function similarly by protruding into the stream flowpath from anchor spots on the bank and impeding and slowing flow velocity, causing sediment deposition in the process and protecting banks.

Fences: Board or wire fences embedded in the banks can be used on small streams to reduce flow and induce sediment deposition. Fences are placed in series protruding into the channel or parallel to the current along the bank. They provide protection for the upper portion of the bank and do not address undercutting.

Jacks and Posts: Jacks are sets of typically three concrete, steel, or sometimes wood beams crisscrossed and bound together at their midpoints to form angular structures. These structures are set in a line along the foot of the bank, forming a "field" which breaks flows, causing turbulence, reducing velocities, and causing sediment deposition. They are often strengthened by wires strung between the beams, which also catch debris and further slow flows. The open construction of the jacks allows for vegetation to establish in the deposited sediments. Posts function similarly, but are single posts driven into the streambed in some kind of grid pattern along the foot of the bank, similarly forming a field.

Pile Dikes: Similar to rock spur dikes, but using timber pilings lashed together and driven into the streambed from the bank outward. They are permeable, allowing for flow between piles. Eroded banks can be rebuilt with the sediment that collects behind them. They are better suited for sandy-bottomed streams than coarse, steep rivers where rock dikes are more appropriate.

Spur Dikes and Hard Points: Rock piles extending from shore into the stream, usually used in series, with the first at the greatest downstream angle and the latter ones more perpendicular to the bank. Spur dikes extend further into the stream and deflect flows well away from the bank. Hard points extend only a short distance and slow velocities along the banks.

REFERENCES

Allen, H.H., and C.V. Klimas, 1986. Reservoir Shoreline Revegetation Guidelines. Technical Report E-86-13. U.S. Army Engineer Waterways Experiment Station, Vicksburg, Mississippi.

Bohn, C., 1989. Management Of Winter Soil Temperatures To Control Streambank Erosion. Pages 69-72. In R.E. Gresswell, B.A. Barton, and J.L. Kershner, eds. Proceedings Of The Practical Approaches To Riparian Resource Management, An Educational Workshop. U.S. Bureau of Land Management, Billings, Montana.

Cairns, J. Jr., 1988. Rehabilitation of Damaged Ecosystems. CRC Press, Inc. Boca Raton, FL.

Chaney, E., W. Elmore, and W.S. Platts, 1993. Managing Change: Livestock Grazing on Western Riparian Areas. U.S. Environmental Protection Agency, Denver, Colorado. 31 pp.

Chaney, E., W. Elmore, and W.S. Platts, 1990. Livestock Grazing on Western Riparian Areas, U.S. Environmental Protection Agency, Denver, Colorado. 45 pp.

Gore, J.A., 1985. The Restoration of Rivers and Streams -- Theories and Experience. Butterworth Publishers, Boston, MA.

Gough, S., 1991. Tree Revetments for Streambank Stabilization. Missouri Department of Conservation, Fisheries Division, Jefferson City, Missouri.

Gray, D.H., and A.T. Leiser, 1982. Biotechnical Slope Protection and Erosion Control. Van Nostrand Reinhold Company, New York, NY.

Gregory, J.D. and J.L. Stokoe, 1981. Streambank Management. Pages 276-281. In Proceedings of the American Fisheries Society, Warmwater Streams Symposium.

Heede, B.H., 1986. Designing for Dynamic Equilibrium In Streams. Water Resources Bulletin 22(3):351-357.

Henderson, J.E., 1986. Environmental designs for streambank protection projects, Water Resources Bulletin, 22(4):549-558.

Heyer, T., 1991. Vegetative Measures for Streambank Stabilization: Case Studies from Illinois and Missouri, United States Department of Agriculture, Forest Service, Northeastern Area State & Private Forestry, St. Paul, Minnesota.

Hoeger, S., 1992. Streamside wet meadows restored, Land and Water, 36(5):8-9.

Keown, M.P., 1983. Streambank Protection Guidelines For Landowners and Local Governments, United States Army Corps of Engineers, Waterways Experiment Station, Vicksburg, Mississippi. 60 pp.

Klingeman, P.C., and J.B. Bradley, 1976. Willamette River Basin Streambank Stabilization by Natural Means. Water Resources Research Institute, Corvallis, OR.

Kohnke, R.E. and A.K. Boller, 1989. Soil bioengineering for streambank protection, J. Soil and Water Conservation, 44(4):286-287.

McGill, S. and J. Gracie, 1993. Restoration Of A Stream Channel Using Native Materials, Land and Water, 37(1):10-12.

Mikkelsen, L., 1993. Soft Engineering Repairs Ailing Zoomontana Stream, Land and Water, 37(1):6-9.

Miller, D.E., 1992. Bio-Engineered Stream Channel Used To Restore New Jersey Brook, Land and Water, 36(4):12-14.

MNDNR, 1991. Streambank Erosion...Gaining A Greater Understanding. Minnesota Department of Natural Resources, Division of Waters, St. Paul, MN.

MWCOG, 1992. Watershed Restoration Sourcebook. Metropolitan Washington Council of Governments, Department of Environmental Programs, Anacostia Restoration Team, Washington, DC.

Orsborn, J.F. and J.W. Anderson, 1986. Stream Improvements and Fish Response: A Bio-Engineering Assessment. Water Resources Bulletin 22(3):381-388.

Platts, W.S., 1989. Compatibility Of Livestock Grazing Strategies With Fisheries. Pages 103-110. In R.E. Gresswell, B.A. Barton, and J.L. Kershner (eds.), Proceedings of the practical approaches to riparian resource management, an educational work shop. U.S. Bureau of Land Management, Billings, Montana.

Roseboom, D., 1993. Case Studies On Biotechnical Streambank Protection. Illinois State Water Survey, Peoria, IL.

Schultze, R.F. and G.I. Wilcox, 1985. Emergency Measures For Streambank Stabilization: An Evaluation. Pages 59-61. In Riparian Ecosystems and Their Management: Reconciling Conflicting Uses. First N. Amer. Riparian Conf.. USDA, Forest Svc., Fort Co llins, Colorado. Gen. Tech. Rep. RM-120.

Siegel, M.D., 1994. Natural Stabilization Of A Streambank, Land and Water, 38(3):28-32.

Swenson, E.A. and C.L. Mullins, 1985. Revegetating riparian trees In southwestern floodplains. Pages 135-139. In Riparian Ecosystems and Their Management: Reconciling Conflicting Uses. First N. Amer. Riparian Conf., USDA, Forest Svc., Fort Collins, Colorado. Gen. Tech. Rep. RM-120.

Thompson, J.N., and D.L. Green (1994). Riparian Restoration and Streamside Erosion Control Handbook. Tennessee Department of Environment and Conservation, Division of Water Pollution Control, Nonpoint Source Program, Nashville, TN.

USACOE, 1981. Final Report to Congress, The Streambank Erosion Control Evaluation and Demonstration Act of 1974, main report. U.S. Army Corps of Engineers.

USEPA, 1993. Guidance Specifying Management Measures for Sources of Nonpoint Pollution In Coastal Waters. EPA-840-B-92-002, January 1993. U.S. Environmental Protection Agency, Office of Water, Washington, DC.

York, J.C., 1985. Dormant stub plantings techniques. Pages 513-514. In Riparian Ecosystems and Their Management: Reconciling Conflicting Uses. First N. Amer. Riparian Conf.. USDA, Forest Svc., Fort Collins, Colorado. Gen. Tech. Rep. RM-120.