Control Stream Erosion

A Stream Great or Small?

Whatever the size of your stream — and the scale of your
project — the same basic principle applies. The main thing is
to understand the interactions between the flowing water
and the solid objects it confronts. And so (assuming that
your house isn’t in danger of washing away after the next
rain), the first rule to consider is look before you touch.

Begin by reminding yourself that water flows downhill.
Whatever problems are created upstream from you are passed
on to you in the forms of turbidity, siltation, flooding,
and irregular flow rates. Your job is to mitigate these
effects as the water flows through your land, thus passing
a better stream on to your neighbors downhill.

Start off by walking your brook in a downstream direction.
Do it during low, clear water, when details of the
streambed will be most apparent. Then do it again during
high water, when you can best observe the erosive force of
water in action. Do it all kinds of in-between times. Walk
in the stream when you can, and on the bank when you can’t.

In your walking or wading, you’ll note that erosion isn’t a
constant force over the whole stream. Some points are badly
damaged whenever the water rises, while others remain
relatively stable. Your task is to ask why. Why does the
water cut the bank here and not there? Why is the channel
deep at one place and silted in at another? What changes
have occurred, or might occur, in the stream’s direction?
What influence exerted on these processes would encourage
stream stability and minimize damage to the land and
waterway?

Start With the Land

All stream erosion problems begin on the land. There are
two main problem categories: excessive erosion of the bank,
and sedimentation of the streambed. (Figure 1 maps a stretch
of stream containing sites with such problems.)

For the purpose of illustration, let’s suppose that you
have located one case of each kind of problem. At one
point, your stream is destroying its bank. Pieces of turf
hang out over the water, supported only by a few inches of
soil, which periodically break off. Meanwhile, the
stream — even at low-flow levels — is cutting away at the lower
part of the bank. It has already undercut one mature
cottonwood tree, which hangs precariously half in and half
out of the water, and another tree is threatened.

This eroding bank is not only unsightly, but also dangerous
to people or livestock who stray too near the edge.
Furthermore, the stream is steadily eating away your land
and depositing your good soil as harmful silt at downstream
sites.

First the stream damages the bank. Then the bank damages
the stream. Farther down from your collapsing bank is one
of the first stops for your lost soil, as well as for other
sediments originating above your property. Just upstream of
a large clump of willows is a shallow, mud-bottomed reach
that is neither attractive nor productive of fish. It used
to be a gravel-bottomed riffle and a deeper pool. You can
tell that by wading through there, because your feet
encounter gravel under a thin layer of silt in the upper
part of the reach. Then, although the water doesn’t appear
much deeper, you sink up to your knees in soft sediment.

Understanding the Problem of Stream Erosion

There are two ways to attack both of these problems: in the
stream, by redirecting the force of the water, and on the
bank, by stabilizing the soil. Channelization, in its
ultimate form, accomplishes both of these ends, albeit at
the cost of all the stream’s natural values. The channel is
dredged perfectly straight, so that the force of the water
against the bank is minimized, and the banks are converted
to rock or even concrete walls, which resist erosive
forces. However, we’re looking for a more “natural” way of
accomplishing these same ends.

Let’s begin with the eroding bank. Why has nature decided
to “pick on” this particular point instead of so many
others? First, note that the problem spot is located on the
outside of a bend, where the force of the current is
strongest. Still, not every bend is as eroded as this one.
Now take a look at the large, barely submerged outcropping
of rock just upstream. A rule of thumb in stream
engineering is that water spills off a solid object at an
angle to the last surface it touches, and that’s just what
has happened here. The rock pushes the main force of the
current (the thalweg, in fishery biologist jargon) directly
into the bank, rather than parallel to it.

The downed tree is part of the problem, too. Cottonwoods
like to grow along watercourses, and in an undisturbed,
natural environment, that’s fine. But in a degraded
environment, cottonwoods may exacerbate bank erosion. These
trees are easily undercut, and even when that doesn’t
happen, old, dying cottonwoods may just fall over, pulling
out their own roots.

In our example, the gap left by the roots forms a secondary
channel during high water, enabling the stream to cut still
farther into the bank. In addition, the trunk of the tree acts as a partial dam. Some water does spill over the top
of it after a heavy rain or snow-melt but — at all times — water
is also forced under the trunk, further undercutting the
bank. Meanwhile, near the opposite bank, the damming effect
of the top portion of the tree pushes still more water from
the left bank toward the right.

What to Do

The first and easiest step toward restoring this problem
spot is to get rid of the downed cottonwood. You could, for
example, simply cut it up into logs, then float or carry
them out. Whatever you do, try to avoid going into the
streambed with heavy machinery. (Note: I’m not advocating
removal of all deadwood as a general practice. In many
instances, downed timber contributes substantially to
stream aesthetics and to fish and wildlife habitats. The
point is that this particular tree is clearly contributing
to the erosion problem.)

Now let’s see if we can redirect the thalweg so it doesn’t
collide abruptly with the bank. Removing the rock
outcropping is out of the question; besides, it’s
attractive and a good fishing spot. One solution is to set
up a counterforce, and the easiest way to do that is to
place large rocks along the eroding bank — a technique known
as riprap. The obvious purpose of riprap is to help the
shoreline resist the force of the water, but it also
creates turbulence — and turbulence tends to throw the
current back toward the center of the stream, further
reducing bank erosion and deepening the center so that it
can accommodate more water.

Placing riprap, however, is not just a matter of flinging
rocks. First of all, the stones must be the correct size
and shape for the dimensions and flow rate of the stream.
If the rocks are too small, they may be moved be a flood.
If they are too large, the bank may erode behind them. As
for shape, flat-edged rocks are easier to place and will
deflect water better than rounded ones. A shovel may be
handy for securely placing the first layer of rock. From
then on, the process is not unlike building a stone wall.
(A layer of gravel under the riprap provides added
protection for the bank.)

Riprap can redirect flow, but it doesn’t appreciably slow
the water. Furthermore, while well-placed riprap is
scarcely noticeable on some sites, on others it can look as
out of place as a concrete wall.

The preferable, but more complicated, alternative is to
build some sort of in-stream structure. The most adaptable
of these is known as a wing deflector. (Figure 2 shows how a
wing deflector might be sited in the situation we are
discussing, and Figure 3 shows how it’s constructed.) The
space behind the leading edge should always be filled.
Otherwise, floodwaters passing over the top will create a
new erosion site.

Solving Siltation

Now, let’s look at the silted downstream site. We’ve
already helped it a little by dealing with the upstream
problem. At least there will be that much less soil
entering the stream which could be deposited there. But we
want to get rid of most of the present silt layer,
restoring the riffle area (which is the most important type
of feeding and spawning habitat for many kinds of fish),
and deepening the pool.

The only real in-stream problem here is caused by the
willows. These trees, with their tenacious root systems,
are great bank stabilizers, but they aren’t so desirable
when they get out into the stream. Here two willows, which
face each other on opposite sides of a narrow spot, have
formed a partial dam with their roots and branches, and
that blockage is worsened by floating debris which collects
there. We don’t want to eradicate the willows, but we can
prune them periodically and try to control their progeny.

To accelerate the process of desiltation, we can make
further use of riprap or wing deflectors to concentrate the
current and flush the silt downstream. In this instance, we
may be able to save labor and cost by substituting large
rocks, strategically placed, for the deflectors. If this is
done, the rocks should be placed toward the middle of the
stream so as not to push the current into the bank. The
ideal is fast current in the middle and slow water along
the banks.

Up on the Bank

For the purposes of erosion control, a stream bank above
the waterline can be looked at as a piece of steep
agricultural land. As on “dry” land, the most versatile
tool for erosion control is vegetation. But there are a few
other tactics that may be appropriate (such as the use of
riprap).

However, perhaps the most important erosion-control tool on
many stream banks is the barbed-wire fence. Livestock in
general, and cattle in particular, are virtual bankwrecking
machines. They break down vertical banks, loosen soft soil,
make paths that become gulleys, trample and overgraze
protective vegetation, and stir up streambottom sediments.
If at all possible, livestock should be completely fenced
out of streams and off their immediate banks. If no other
provision can be made for watering, or if the animals must
cross the waterway, give them limited access, only at a
gently sloping, erosion resistant site.

Human beings need “fencing,” too. Try to limit your own
access to a few carefully selected points. And, if
necessary, construct steps to avoid beating a path into a
steep bank.

Proper Vegetation

The proper planting of stream banks involves matching the
vegetation to the habitat. As a general rule, the wider the
stream, the taller the vegetation can be. A bank totally
lined with large, mature trees may be quite appropriate on
large rivers, while tiny brooks may do better with nothing
more than a good grass cover. (I’m speaking solely in terms
of flood and erosion control. Choosing a strategy becomes
much more complex when one considers the water temperature
requirements of fish, the potential for loss of water by
evaporation and transpiration, the value of the bank as a
wildlife habitat, the available fish food supply, the
nutrient contribution made by fallen tree leaves, and the
demands of personal aesthetics. In most cases, some
combination of trees and low-growing vegetation will be
called for.)

Even where the use of trees is strongly indicated, the
first step may be to establish a good ground cover. The
procedures for doing so are no different from those
employed in ordinary agriculture. (But do be cautious with
fertilizers; there’s no point in trading erosion for
pollution.) Among the many grasses appropriate for
stream-bank planting, my personal favorite is reed canary
grass, which holds soil well, tolerates occasional
flooding, provides food for wildlife, and looks beautiful.
Its only real drawback is that when grown alongside very
small streams, it may extend out into the water. Narrow
waterways may thus be temporarily choked in late summer.

Near-vertical banks are among the toughest spots on which
to establish vegetative cover. For this and other
specialized problems, consult your local Soil Conservation
Service office. The experts there may be able to recommend
specific seed mixtures suitable to your site and to local
conditions.

In general, there is no tree like the willow for holding a
stream bank together. A well established willow will break
in half before it’ll let go. And even a dead willow stump
may provide effective erosion control for many years. Given
these traits and the ease with which they can be planted,
there’s no reason not to’ plant willows along large
streams. However, on small streams, willows may offset the
good they do on the bank with the damage they do in the
water. Don’t plant such trees along little brooks unless
you’re prepared to prune the branches periodically.

What applies to willows applies, in a lesser degree, to
alders, also commonly found along streams. And, as we have
seen, cottonwoods have their own disadvantages. Ashes,
maples, walnuts, and sycamores are among the trees that
prosper on streamside sites, resist uprooting, and don’t
grow into the water. You can work out your own tree planting
scheme by consulting with tree experts, or by walking or
boating a forested stream in your area and noting the trees
that behave as you would like.

In addition to planting trees along the stream, it’s also a
good idea to have a line of trees or shrubs directly behind
the high point of the bank. During floods, they’ll slow the
receding waters, causing silt to drop out on the bank. This
vegetation will keep silt from the stream, where it’s a
problem, and trap it where it’ll enrich your agricultural
land.

Someone once suggested that natural stream management — of
which erosion control is a major part — is to waterway
engineering as orthodontics is to oral surgery. And both of
the former work best if you push a little and wait a lot.

The satisfaction is not just in pushing for results, though
they come eventually. It’s also in waiting, watching, and
encouraging Mother Earth’s self-healing processes — and
knowing that you can never learn it all.