Channelisation / cross section alteration

From REFORM wiki
Jump to: navigation, search

Channelisation / cross section alteration

04. Morphological alterations

General description

Channelization refers to river and stream channel engineering undertaken for the purposes of flood control, navigation, drainage improvement, and reduction of channel migration potential (Brookes, 1990[1]). When channelization involves cross section alteration, this includes activities such as channel enlargement through widening or deepening, the reduction of flow resistance through clearing or snagging of riparian, and sometimes aquatic, vegetation and other roughness elements, and the introduction of bank facing and reinforcement materials. These forms of morphological modification typically transform channel cross profiles into uniform, smooth, trapezoidal or rectangular forms. Cross section alteration can also include embankment, levee or dike construction, which further enlarge the channel capacity, prevent channel-floodplain connectivity, and can induce very high flow velocities within the river channel during floods.

Effect/Impact on (including literature citations)

Channelization involves changes in channel planform and gradient (usually straightening); channel cross profile form and flow resistance (usually channel enlargement and removal of morphological and vegetation roughness elements); and in some cases it involves the introduction of artificial materials to reinforce the modified channel form (e.g. concrete, metal, stone, bricks). Thus, McIninch and Garman (2001) defined three significant aspects of channelization of streams in Illinois: stream alteration (the extent to which the stream channel form has been altered or modified (Barbour et al. 1999); sinuosity (the extent to which the stream channel has been straightened); and riparian canopy (the degree to which the stream coverage by vegetation has changed). These physical changes not only increase the flow energy (increased gradient, decreased flow resistance), but they may modify sediment supply, remove important habitats within the channel, and restrict both lateral and vertical connectivity, leading to severe ecological effects (e.g. Poole et al., 2006). Hydromorphology and ecology are not only directly impacted by channelization, but channelization induces a range of hydromorphological responses that also have ecological significance. Hydromorphological responses to channelization are as complex as those following dam construction and river impoundment and vary with the type of channelization scheme and its impact on channel gradient, the flow regime, the calibre of bed and bank materials, the supply of sediment, and the ability of the riparian vegetation to colonise and reinforce disturbed sediments. If any adjustment is to occur, the affected stream has to have sufficient energy to move sediment and thus to remodel the form of the channelized stream. Research by Urban and Rhoads (2003) on the Embarras River basin of east central Illinois, illustrates that where stream gradients are very low (typically between 0.001 and 0.0001 m m -1 in their study site), channel responses to widespread and severe channel straightening and enlargement are extremely slow and channels persist in their modified state for decades following channelization. In such environments, further human modification by river restoration is essential to the re-establishment of a healthy ecosystem. However, in moderate to high energy environments, responses to channelisation are rapid, far-reaching and complex. One example of this complexity is provided by Simon’s (1989) model of channel evolution following channelization of sand bed rivers in Tennessee. The model identifies six stages of morphological development from (i) the premodified channel and its modification to form (ii) the constructed channel, through a phase of (iii) degradation leading to a (iv) bank slope threshold stage and then a phase of (v) aggradation until (vi) restabilization is reached. Following channel ‘construction’, which usually involves realignment, deepening and bank steepening, channel degradation occurs within the affected reach and propagates upstream. This involves channel bed incision accompanied by erosion of the bank toe, which together steepen the banks. Eventually the banks reach a critical angle and enter the threshold stage, during which bank failures are widespread and the channel widens. Widening slows and ceases as the bank angles reduce and the channel width allows sediment to be retained at the bank toe. Sediment deposition at the toe and across the bed is enhanced by vegetation colonisation leading to significant bed aggradation and recovery to a restabilised state. Aggradation is often achieved by the formation of alternate bars, which are sediment stores that guide channel planform development. All of these phases are observed within the modified channel and they propagate upstream with bed incision and knick point retreat, leading to widespread morphological impacts. Although Simon’s (1989) model provides an excellent conceptual framework, responses to channelization deviate from it with local circumstances. For example, Simon and Thomas (2002) illustrate a significant downstream response to channelization following the propagation of the degradation and threshold stages upstream that provides an important extension to the model. They observed that the upstream migration of knick points associated with the degradation phase on the Yalobusha River, Mississippi, resulted in the delivery of large quantities of sediment and woody vegetation to downstream reaches from bank failures. These materials accumulated at the downstream end of the channelized reach to form a large sediment/debris plug at the junction with an unmodified sinuous reach. Such a plug has the potential to produce a local higher base level which may accelerate the propagation of bed aggradation and channel recovery upstream. Furthermore, plug removal could reactivate incision and bank failure, with the potential for enhanced failure due to groundwater drainage through the basal bank layers. In addition, the presence of erosion resistant clay beds in some reaches, rather than the sand beds of the original model, was observed to restrict bed incision and knick-point propagation, leading to channel adjustments that were more dependent upon channel widening and bank failure. In the coarser-bed, steeper, rivers of northern Italy, Surian & Rinaldi (2003) found rather different and even more extreme responses to those recorded by Simon (1989). While Simon’s (1989) research focused entirely upon the effects of channelization, Surian and Rinaldi (2003) observed responses to a range of human interventions including dam and weir construction, in-channel gravel mining, reforestation and more general flow regime changes as well as river channelization and embanking. They proposed a classification of adjusted channel types, which were combined into a model of stages of channel adjustment. The model describes how progressive bed incision of typically 3-4 m (up to 10 m in some examples) and narrowing (by up to 50%) of single-thread, transitional and multi-thread (braided) rivers lead to the development of a sequence of degraded channel types. These responses reflect not only channelization processes but also a heavily reduced supply of sediment, and in many cases a modified flow regime. Observations of adjustment in the Raba river, Poland, by Wyzga (1993) provide a detailed record of the processes similar to those that have contributed to the changes observed in the Italian rivers. Wyzga observed up to 3 m of bed incision, with the progression of headcutting increasingly moderated by energy-dissipating mid-channel bars in upstream reaches. Incision was accompanied by the erosion and removal of finer bed material, and a reduction in the susceptibility of the remaining, armoured bed to particle entrainment. Incision resulted in an increasingly efficient channel cross profile and a consequential magnification of peak discharges and more flashy flood waves. The increase in channel depth, bed coarsening and decreased bed gradient that was created as erosion propagated upstream led to re-establishment of a new equilibrium at higher flow velocities and stream power than before channelization. These examples illustrate that channelization by straightening, steepening and simplifying the cross profile of stream channels generally increases flow velocities and therefore often significantly alter bed sediment by removing silt and other easily moved particles to create an armoured bed. In urban rivers, the quality of the bed material may also change, as has been observed on the River Tame, UK, where heavy metal concentrations within the urbanized matrix sediment are up to 3000 times greater than background levels (Thoms 1987). Straightening of stream channels generally reduces the amount of substrate available for epifaunal colonization by reducing the roughness of the channel boundary (through removal of woody debris and other potential habitat such as rocks and boulders) and by removing stream bends where pool development, bank undercutting and exposure of vegetation roots supply a variety of habitats. Some channelization schemes incorporate in-channel structures, such as weirs and rip-rap at the bank toe, to reduce the channel gradient locally, increase the flow resistance and thus dissipate the increased flow energy that may otherwise accompany channelization. These structures increase the physical complexity and thus habitat diversity of channelized reaches (Silva-Santos et al. 2004). Research by Bombino and others (Bombino et al. 2007, 2008, 2009) has investigated the sedimentary and plant ecological changes induced by the introduction of check dams into steep, confined mountain torrent streams in Calabria, Italy. They found pronounced increases in channel and riparian zone width, decreased channel gradient and fining of bed sediment calibre upstream of the check dams, which was associated with an increase in plant species richness, and in vegetation cover and development relative to reaches downstream of or unaffected by check dams. In this case, the physical changes induced by the check dams have increased the range of habitats available for plant colonization, and have provided a range of lower energy, more water retentive patches, where a variety of species can establish. The effects of grade control structures (GCS: weirs with stone-protected stilling basins) combined with streambank protection were assessed in the much lower gradient environment of Twentymile Creek, Mississippi by Shields and Hoover (1991). Here bank-line woody vegetation cover increased by 8% in 4 years on the more stable channel margins, reaches immediately upstream and downstream of GCS were deeper with slower flow velocities, and differences in aquatic habitat diversity among sites along the river were primarily due to the bed scour holes downstream of GCS and inthe low-flow channel. Comparison with reaches without GCS, showed a 29% higher index of fish diversity, which was positively correlated with substrate diversity and mean depth, and with fourteen species collected exclusively at GCS. Abundance of several of the numerically dominant species was positively associated with deeper water and lower flow velocities. Despite the positive impacts of some channelization structures in some environmental settings, the ecological impacts of channelization are usually negative. The simplified channels that are created, particularly in association with land drainage and flood alleviation, lead to significant ecological degradation. For example, the main channel of the River Morava (a tributary of the Danube) has been totally isolated by channelization from its flood plain and regulated by weirs. Here, Jurajda (1995) found that the young of the year of phytophilous species (pike Esox lucius L., rudd Scardinius erythrophthalmus (L), silver bream Blicca bjoerkna (L), tench Tinca tinca (L), carp Cyprinus carpio (L.) had almost disappeared, and a decline in density was also found for rheophils, such as vimba Vimba vimba (L.), barbel Barbus barbus (L.) and nase Chondrostoma nasus (L.), previously the dominant species in the river. Such degradation may affect riparian as well as aquatic habitats and species, and can feed through to impacts on birds as well as terrestrial organisms (Frederickson, 1979). However, morphological recovery from channelisation is accompanied by vegetation and other ecological recovery that is also complex but has some characteristic features. Hupp (1992) observed vegetation recovery in the same Tennessee channels studied by Simon (1989). In particular, he noted the importance of riparian vegetation in facilitating recovery through the aggradation and restabilisation stages. He noted that woody vegetation initially establishes on lower bank surfaces in association with bank toe accretion, helping to trap and stabilise the depositing sediment. At this early phase, the vegetation is dominated by hardy, fast growing, pioneer species that can tolerate moderate amounts of slope instability and sediment deposition (e.g. Betula nigra, Salix nigra, Acer negundo, and Acer saccharinum). They grow in dense stands with dense root-mass development that enhances bank stability. As aggradation progresses and banks become increasingly stable, other species colonise the channel margins, and the analysis of tree rings suggested that 65 yrs may typically be required for restabilisation to be complete. Morphological and vegetation recovery are accompanied by recolonisation by macroinvertebrate and fish species, and such recovery can be enhanced by deliberate, appropriately-designed, restoration. For example, Friberg et al. (1994) found that two years after restoration of a meandering course to a 1.3 km straightened and channelized reach of the River Gelså, macroinvertebrate density and diversity was greater than in the upstream control reach and species preferring a stony habitat seemed to favour the new reach, including Heptagenia sulphurea Müll., Ancylus fluviatilis Müller and Hydropsyche pellucidula Curtis.

Conceptual framework of channelization effects on HYMO processes and variables (POM=Particulate Organic Matter; LWD=Large Woody Debris).

Case studies where this pressure is present

Possible restoration, rehabilitation and mitigation measures

Useful references

Brookes, A., 1990. Restoration and enhancement of engineered river channels: Some european experiences. River Research & Applications 5: 45–56.

Other relevant information

Cite error: <ref> tags exist, but no <references/> tag was found