|
<br />We all have seen examples of channel widening below width-
<br />constricting structures, forming "bulges" in each bank. If.
<br />this process is not controlled, loss of structure may result.
<br />This can be prevented by the design of gentle width transi-
<br />tions such as wing walls that slowly increase the width to the
<br />normal channel size.
<br />Flow line separation can also occur at bank protection
<br />works, such as riprap, brush or dead tree placements, if the
<br />transition from the upstream channel bank into the structure
<br />and from the structure into the downstream bank is abrupt. In
<br />this case, flow separation takes place, because the flow along
<br />the bank impinges at the upstream side of the installation,
<br />while the flow within the channel maintains its velocity. As
<br />described before, this causes the impinged flow to move into
<br />an eddy, rotating the water along the upstream side of the
<br />installation into the bank and then upstream. A similar
<br />situation can exist at the downstream end of the structure
<br />where stilled water occurs downstream from the installation
<br />and higher velOcity flows within the channel. Here an eddy
<br />will run upstream into the bank. In both places, bank scour
<br />may destroy the structure. Obviously, gentle structural
<br />transitions are required at each end of the control work.
<br />
<br />IMPORTANCE OF LOCAL BASE LEVEL
<br />FOR CHANNEL STABILITY
<br />
<br />The foregoing examples illustrated the importance of the
<br />sediment load and the consequences of load changes caused
<br />by human interference. Where longitudinal profile altera-
<br />tions took place, local base level changes also occurred.
<br />The local base level is that elevation of a given location to
<br />which upstream areas or stream reaches have adjusted. Lower-
<br />ing of this level induces degradation (Begin, et al., 1980;
<br />Heede, 1981); raising it results in aggradation (Heede, 1981).
<br />Ifwe can draw inferences from check dam treatments,aggrada-
<br />tion does not affect all upstream areas or stream reaches but
<br />ceases at a certain distance from the local base level (Heede,
<br />1960, 1977). This is in contrast to degradation, which may
<br />advance into the headwaters.
<br />The concept of local base level should be applied to stream
<br />control work. Application of the concept requires analysis
<br />of the stream system as an entity, although controls may be
<br />planned just for a particular site. For example, check dams
<br />could have a short life span, if head cuts on the channel bed
<br />in the reach below the structures were not recognized and
<br />controlled. An alternative would be to install check dam
<br />foundations that could not be undermined by the advancing
<br />headcuts. Problem-causing headcuts could be located at long
<br />distances downstream from the check dams and not be de-
<br />tected, unless inspection and analysis starts at the mouth of
<br />the stream system.
<br />Degradation induced by the lowering of the local base
<br />level is one of the most important and far-reaching events in
<br />a stream and, therefore, deserves special attention. Unless
<br />intervening natural controls such as hard bedrock exist within
<br />the channel, the local base level lies at the mouth of the
<br />
<br />Heede
<br />
<br />stream. As explained earlier, this base level depends on the
<br />master stream's stability, or that of the continental shelf at
<br />the site of junction with the ocean. Where possible, control
<br />project designers should, therefore, ascertain that this base
<br />level will not change during the estimated project time. If
<br />natural controls are missing, check dam projects should start
<br />at the mouth of the stream and proceed upstream. Also, for
<br />other types of control measures, the possibility of base level
<br />change must be considered. Often this results in additional
<br />costs that may outweigh the installation cost of the main
<br />project.
<br />Although installations are not always inexpensive, base
<br />level controls are one of the most effective and often most
<br />cost-effective measures. Normally, the consequences of base
<br />level changes are easily detectable. Yet, it is preferable not
<br />to wait until the damage is done, but to assure that future
<br />level changes will not take place where important controls
<br />are installed.
<br />Where base level changes have taken place, serious conse-
<br />quences may not only follow for the stream but also for its
<br />riparian system. If channel incision occurs, this system may
<br />be excavated or obliterated in place by lowering of the water
<br />table. On the other hand, a base level raise could bury the
<br />riparian system due to induced sedimentation in the reach
<br />upstream from the base level. It may therefore be desirable
<br />to correct the base level change either by lowering it (excava-
<br />tion) or raising it (dam installation). By restoring a healthy
<br />riparian system, attainment of dynamic equilibrium in the
<br />stream system will be furthered (Heede, 1984).
<br />
<br />WORKING WITH THE ADJUSTMENT PROCESSES
<br />
<br />Natural processes and energy balance dictate why a stream
<br />meanders, aggrades or degrades. Analysis of these factors
<br />must precede the application of effective controls. Further-
<br />more, other processes may not have reached the stream seg-
<br />ment under consideration, but must be expected to exert their
<br />impact in the future. An example was cited earlier of bank
<br />protection work that could be destroyed by headcuts located
<br />at far distances downstream. It follows that the adjustment
<br />processes within the stream system must be known, if we in-
<br />tend to work with the processes instead of against them. Ob-
<br />viously, it is less expensive and more effective to work with
<br />the processes.
<br />An example is a degrading stream that not only leads to
<br />channel deepening but also to bank instability, possible
<br />destruction of bridges, and other failures. It is well estab.
<br />lished that depth of degradation decreases downstream (Wil-
<br />liams and Wolman, 1984), leading to a gentler channel gradient
<br />that, with time, will allow a new equilibrium between sedi-
<br />ment load and flow energy (the incoming load will equal the
<br />load leaving the stream reach). We can speed up this process
<br />by the installation of check dams that, due to sediment ac-
<br />cumulations, cause shallower bed gradients. In addition to
<br />this benefit, bank stabilization is also enhanced. Banks will
<br />decrease in height and the maintenance of the toe of banks
<br />
<br />356
<br />
<br />WATER RESOURCES BULLETIN
<br />
|