Environmental Engineering Reference
In-Depth Information
The data are extensive; approximately 145,000
data items are being recorded per week.
land use impacts in a statistically rigorousmanner
and to address runoff generation at the hillslope
and catchment scale. The effects must also be
evaluated in the context of the complex history
of landmanagement. For example, land drainage is
extensive, with a history that is believed to go back
to the Napoleonic Wars
Multiscale experimentation is needed to bridge
the current gap between plot-scale experiments
and catchment-scale impacts, hence a set of ex-
perimentswas designed to provide data support for
new methodological development. The Pontbren
project crucially provides landowner support for
land access, land management manipulation ex-
periments and for socioeconomic analysis. The
direct involvement of the local farmers is also
important in the promotion of policy guidance to
the agricultural community.
The current scales of research range from ex-
perimental plots to an 18-km
2
catchment, includ-
ing three first-order streams. The experiments
focus on soil properties and runoff processes, based
on plot- and hillslope-scale measurements nested
within instrumented first- and second-order
catchments (Marshall et al. 2006). At plot scale,
manipulation plots have been established at four
locations, representing a range of aspect and soil
type. At each location three treatments are being
evaluated: grazing, no grazing and newly planted
woodland. Continuous monitoring includes pre-
cipitation, other climate variables, soil moisture
contents, soil water potentials and overland flow.
In addition, soil physical and chemical properties
are characterized in annual sampling campaigns.
At hillslope scale, instrumented hillslope trans-
ects include the above instrumentation, ground-
water elevations, and drain and ditch flows.
Within the hillslope experiments, soil properties
and runoff processes are being investigated under
different land use treatments including woodland
buffer strips. At catchment scale the monitoring
is complemented by a network of stream gauges.
These observations are supported by a soil
survey, including estimation of soil degradation
status, supplementary sampling and additional
experimentation, including sprinkler and tracer
experiments and woodland interception studies.
Experimental results
Figure 3.1 shows the Pontbren and adjacent Rhos
aflo catchment along with themain streams with-
in the catchment and monitoring locations. A
large dataset now exists from the Pontbren exper-
iment, and results are presented here to illustrate
key findings. These include the impact of plot-
scale land use change on soil hydraulic properties
and hydrological processes, runoff processes from
an improved pasture hillslope, and stream flow
response. More extensive results are provided by
Wheater et al. (2008).
A survey assessing the structural conditions of
the soils at Pontbren was undertaken in 2006 and
followed the methodology described in Holman
et al. (2003). The survey indicated that 60% of the
land was moderately degraded and 25%of the land
was highly degraded, and that these highly degrad-
ed soils were confined to land under improved
grassland production. There is also anecdotal ev-
idence to indicate that in recent years the Wil-
cocks soil series, with its peaty surface layer, may
have retreated up the hillslope, being replaced by
the Cegin soil series (a heavy-textured clay loam).
It is speculated that this loss of peat is attributed,
to some extent, to agricultural intensification,
such as the increase in stock numbers and the
installation of field drainage systems. The im-
proved pasture tends to dominate in the east of
the study site whereas the rough grazing is found
more to thewest and north on the higher altitudes.
Manipulation plots
The response of the manipulation plots is highly
heterogeneous, hence it was important to estab-
lish baseline conditions prior to the imposition of
the experimental treatments. The plots were
therefore established in 2006, and treatments im-
plemented in 2007. The monitoring and analysis
is ongoing, but despite the treatments being less
than 2 years old, changes in soil physical and
