Geology Reference
In-Depth Information
et al
., 1987; Haan
et al
., 1994), but these results
have generally been obtained on planar slopes.
Their applicability to general field situations is
questionable, particularly where convergent slope
shapes cause runoff to concentrate and enter rivers
at a limited number of points in the landscape,
rather than flowing uniformly into the buffer along
the whole of its length. In certain situations, the
greater depth of flow arising from the concentra-
tion of the runoff can lead to submergence of the
grass barrier, rendering it less effective.
Numerous studies have shown that the length of
the buffer is important from upslope to downslope
edge. Lalonde (1998) found that trapping efficiency
varied between 68% and 98% as the length increased
from 2 to 10 m; Abu-Zreig (2001) observed that a
15 m long buffer was three times more efficient
than a 1 m long barrier; while Dillaha
et al
. (1989)
showed that trapping efficiency ranged from 53% to
86% with 4.6 m long buffers and from 70% to 98%
with 9.1 m long buffers. Other studies, however,
show that since most sediment is deposited within
the first few metres of the strip, increasing the
length beyond 3-5 m has little effect on trapping
efficiency (Line, 1991; van Dijk
et al
., 1996).
The effectiveness of different lengths depends
on the nature of the sediment being carried in the
flow. Schwer and Clausen (1989) found that a 26 m
wide grass strip trapped 92% of the total phospho-
rus (particulate and dissolved) associated with the
runoff on a sandy soil, but only 33% on a silty clay.
Barriers of 5-10 m length may trap nearly all the
sediment on sandy soils, whereas clay particles are
transported greater distances and may form 80-90%
of the material leaving the buffer. However, even if
all the clay particles and associated phosphorus are
not trapped, a sufficient quantity may be which
will enable a vegetated barrier to serve as a pollu-
tion control measure. Kronvang
et al
. (2000) found
that a 29 m long buffer was able to trap 100% of the
sediment and particulate phosphorus on soils in
Denmark, and that a 12 m long strip was enough to
reduce the delivery of particulate phosphorus to
water courses to an acceptable level.
Recent research has shown that the effective-
ness of the buffer is greatly influenced by the archi-
tecture of its vegetation, and therefore care has to
be taken in selecting appropriate plant species. For
grasses, sturdy, tall, perennial species are consid-
ered the most suitable, and short, flexible grasses
less so (Grismer
et al
., 2006). Taller grasses that
grow vertically often produce a sward with a high
density of interwoven stems, forming a porous fil-
ter, whereas shorter grasses generally bend from the
vertical and consist of individual stems with a more
porous structure. Fescue (
Festuca ovina
) was found
in laboratory experiments to trap 90% of the sedi-
ment moving on slopes of 7-9°, compared with
only 70% for meadow grass (
Poa pratensis
) (Lakew
& Morgan, 1996). Adding switch grass (
Panicum
virgatum
) to barriers of fescue makes them even
more effective (Blanco-Canqui
et al
., 2004, 2006).
An open structure to the barrier can sometimes
enhance rather than protect against erosion. Con-
centrations of runoff through gaps in the barrier can
result in localized increases in flow velocity. Runoff
can therefore leave the barrier at a high velocity but
relatively free of sediment. Under these conditions
the presence of the barrier can create more erosion
downslope that would have occurred without the
barrier (Emama Ligdi & Morgan, 1995; Ghadiri
et
al
., 2001; Spaan
et al
., 2005).
A further uncertainty in the performance of
buffers is that, whilst they can trap sediment during
storms, the deposition is not permanent and the
sediment can be re-suspended during subsequent
events (McKergow
et al
., 2006). Thus the barriers
can be both a sediment sink and a sediment source,
depending on the magnitude of the storm (Daniels
& Gilliam, 1996; Verchot
et al
., 1997).
Before planning the installation of new buffer
features in the landscape, it is important to note
that in many agricultural areas of the UK, vegetated
areas already exist. These include hedgerows along
field boundaries, and areas of trees and woodland
either at the edge of or between fields. Riparian
buffers are often the remnants of former river plain
forests with willows, alder and some hardwood
trees. Although not designed as soil-protecting buff-
ers, they may perform that function to varying
extents depending on their structure and density.
Where a buffer already exists, it makes sense to
evaluate its present performance and make
recommendations to enhance its function rather
