Shallow water courses

Shallow water courses

Shallow water courses

05. River bed depth and width variation improvement

General description

Many rivers are deeply incised and disconnected from their floodplains. Aggradations of the channel-bed may lead to an increase in the flooding frequency and groundwater level. This is a prerequisite to develop riparian vegetation and forest that are depending on this kind of frequent flood events. Moreover, water depth and shear stress decrease, potentially reducing invertebrate drift, especially in small streams.

In principle, there are three possible ways to increase the level of the channel-bed: Creating a new, shallower channel cross-section, to increase deposition of sediment in the reach or to repeatedly add sediment. Creating new, shallower channel cross-sections: This approach is often used together with re-meandering a river reach since it is necessary to decrease slope, flow velocity and channel-bed erosion by increasing sinuousity. The “stable” channel-dimensions (mean cross-section width, depth, sinuousity) of a meandering river in its dynamic equilibrium state can only be roughly assessed based on nearby reference sites, empirical equations or regime models. Historical data must be used with care, since discharge and sediment regime as well as bank vegetation, which have a strong influence on channel planform and dimensions, may have been altered (e.g. downstream from reservoirs, peak-flows from impervious areas, grazing of riparian areas). Several projects failed since meandering channels were created in places where a braiding planform naturally occurs (Kondolf and Railsback 2001, Kondolf 2006). Therefore, it is crucial to adequately assess the channel planform and dimensions which can be expected given the catchment characteristics (e.g. grain size, discharge, sediment load, bank material and riparian vegetation).

Increasing sediment deposition: This approach can only be applied if the sediment load or input in the restored reach is high (e.g. not downstream from dams which block sediment transport). The placement of log steps is a widely used measure to trap sediment. However, such steps may hinder the upstream migration of groundfish like bullhead (Cottus gobio). Alternatively, large amounts of large wood have been added to decrease channel depth and to increase bed roughness and sedimentation in some small Central European streams (Launhardt and Mutz 2002, Kail et al. 2007) (see measure “Addition of large wood”). Increasing sedimentation in the restoration reach may cause a sediment deficit and erosion in the downstream reach.

Repeatedly adding sediment: This approach is usually applied to compensate for sedimentation in the upstream reservoir and sediment deficits downstream of dams. For further information on this approach see measure “Add sediments (gravel, sand)”. As an alternative to a regular bedload supply, the sediment is covered once by a layer of coarse gravel or cobbles heavy enough not to be washed away to prevent incision as well as the scouring impact from vessels. Such artificial armour layers are for example used in the Danube river in Austria.

Applicability

See problems and constraints of the different approaches mentioned above. Restricted to reaches where lateral channel migration and / or frequent flooding can be admitted (whole inundated floodplain).

Expected effect of measure on (including literature citations):

Decreasing cross-section depth is a prerequisite for many floodplain measures and typically part of a set of measures to enhance channel-planform, floodplain and instream habitat conditions (e.g. re-meandering, developing riparian vegetation, placement of large wood, re-connecting side channels). Therefore, it is difficult to assess the partial effect of this measure.

HYMO (general and specified per HYMO element)

• Decrease of mean channel depth (Krovang 1998, Klein 2007)

Physico-chemical parameters

• Increase in groundwater recharge and summer low-flow (Tague et al. 2008)

• Nutrient retention due to increase in travel-time and storage as well as more frequent inundation of floodplain area (Bukaveckas 2007, Pedersen et al. 2007, Hoffmann et al. 1998, Krovang et al. 1998)

Biota (general and specified per Biological quality elements)

Macroinvertebrates:

• Decrease of flow-velocity probably favours limnophilic invertebrate species

Fish:

• Increase in floodplain nursery habitats and lateral connectivity probably favours reproduction and survival of several fish species.

• Increase in shallow low-velocity areas near the river banks especially favours juvenile fish.

• A high bedload supply, typically added to compensate for incision in straightened and channelized rivers, and high sediment transport may have negative effects on fish by filling pools and decreasing availability of overwintering habitats (also see fact-sheet “Sediment addition”).

Macrophytes:

• Decrease of water depth and flow-velocity probably favours macrophytes.

Phytoplankton:

• Probably no significant effect on phytoplankton.

Pressures that can be addressed by this measure

• Channelisation / cross section alteration
• Alteration of riparian vegetation
• Alteration of instream habitat

Cost-efficiency

Largely depends on the land purchase cost and on the approach used (creating new channel vs. increasing sedimentation). Medium to high cost-efficiency if land is already owned due to the medium and sustainable ecological effect.

Case studies where this measure has been applied

• Renaturierung Untere Havel
• Charlottenburg artificial bay
• sheet pile protected shallow new
• sheet pile protected shallow
• Lahn Cölbe
• Vääräjoki - Niskakoski
• Kuivajoki - Hirvaskoski
• Meers - Floodplain lowering
• Rijkelse Bemden - River bed widening
• Thur
• Amesbury - Demonstrating strategic restoration and management STREAM (LIFE05 NAT/UK/000143)
• Chilhampton - Demonstrating strategic restoration and management STREAM (LIFE05 NAT/UK/000143)
• Scheldt - Vallei Grote Nete
• River Cole EU-LIFE
• Lower Traun
• Lippeaue Klostermersch
• Ruhr Binnerfeld

Useful references

Bukaveckas, P. A. (2007) Effects of Channel Restoration on Water Velocity, Transient Storage, and Nutrient Uptake in a Channelized Stream . Environmental Science und Technology, 41, 1570 - 1576.

Hoffmann, C. C., Pedersen, M. L., Kronvang, B. & Ovig, L. (1998) Restoration of the Rivers Brede, Cole and Skerne: a joint Danish and British EU-LIFE demonstration project, IV - implications for nitrate and iron transformation. Aquatic Conservation: Marine and Freshwater Ecosystems, 8, 223 - 240.

Kail, J., Hering, D., Muhar, S., Gerhard, M. & Preis, S. (2007) The use of large wood in stream restoration: experiences from 50 projects in Germany and Austria. Journal of Applied Ecology, 44, 1145-1155.

Klein, L. R., Clayton, S. R., Alldredge, J. R. & Goodwin, P. (2007) Long-Term Monitoring and Evaluation of the Lower Red River Meadow Restoration Project, Idaho, U.S.A. Restoration Ecology, 15, 223 - 239.

Kondolf, G. M. & Railsback, S. F. (2001) Design and performance of a channel reconstruction project in a coastal California gravel-bed stream. Environmental Management, 28, 761-776.

Kondolf, G. M. (2006) River Restoration and Meanders. Ecology And Society, 11, 42.

Kronvang, B., Svendsen, L. M., Brookes, a., Fisher, K., Moller, B., Ottosen, O., Newson, M. & Sear, D. (1998) Restoration of the rivers Brede, Cole and Skerne: a joint Danish and British EU-LIFE demonstration project, III - Channel morphology, hydrodynamics and transport of sediment and nutrients. Aquatic Conservation: Marine and Freshwater Ecosystems, 8, 209 - 222.

Launhardt, A. & Mutz, M. (2002) Totholz statt Steine, eine Alternative für Sohlgleiten in abflussschwachen Sandbächen. Deutsche Gesellschaft für Limnologie (DGL). Annual Conference 2001, Braunschweig pp. 699–702. Pedersen, M., Andersen, J., Nielsen, K. & Linnemann, M. (2007) Restoration of Skjern River and its valley: Project description and general ecological changes in the project area. Ecological Engineering, 30, 131 - 144.

Tague, C., Valentine, S. & Kotchen, M. (2008) Effect of geomorphic channel restoration on streamflow and groundwater in a snowmelt-dominated watershed. Water Resources Research, 44, W10415.

Other relevant information

Evolution of incised channels: Several conceptual models have been developed for the evolution of deep or incised channels (e.g. Schumm et al. 1984, Simon and Hupp 1986, Simon and Rinaldi 2006). These models can be used to assess the present stage of the river reach, which further morphological changes probably occur, and to select appropriate restoration measures. For example, if the reach is already downstream from the knickpoint and starts to widen (Fig. 1 stage IV), it might be an appropriate restoration measure to increase sediment deposition since there will be an increasing sediment input from upstream. Therefore, it is crucial to consider the state of the upstream and downstream reaches and to involve some geomorphological expertise in the restoration project.

According to the channel-evolution model of Simon and Rinaldi (2006) (Fig. 1):

“One can consider the equilibrium channel as the initial, predisturbed stage (I),

and the disrupted channel as an instantaneous condition (stage II).

Rapid channel degradation of the channel bed ensues as the channel begins to adjust (stage III).

Degradation flattens channel gradients and consequently reduces the available stream power for given discharges with time. Concurrently, bank heights are increased and bank angles are often steepened by fluvial undercutting and by pore-pressure induced bank failures near the base of the bank. The degradation stage (III) is directly related to destabilization of the channel banks and leads to channel widening by mass-wasting processes (stage IV) once bank heights and angles exceed conditions of critical shear-strength of the bank material. The aggradation stage (V) becomes the dominant trend in previously degraded downstream sites as degradation migrates further upstream because the flatter gradient at the degraded site cannot transport the increased sediment loads emanating from degrading reaches upstream. This secondary aggradation occurs at rates roughly 60% less than the associated degradation rate (Simon, 1992). Riparian vegetation becomes established on low-bank surfaces during this stage and serves as a positive feedback mechanism by providing roughness that enhances further deposition. These milder aggradation rates indicate that recovery of the bed will not be complete and that attainment of a new dynamic equilibrium (stage VI) will take place through further (1) bank widening and the consequent flattening of bank slopes, (2) the establishment and proliferation of riparian vegetation that adds roughness elements, enhances bank accretion, and reduces the stream power for given discharges, and (3) further gradient reduction by meander extension and elongation. The lack of complete recovery of the bed results in a two-tiered channel configuration with the original floodplain surface becoming a terrace. Stormflows are, therefore, constrained within this enlarged channel below the terrace level and result in a given flow having greater erosive power than when flood flows could dissipate energy by spreading across the flood plain.”