Agriculture Reference
In-Depth Information
of the plots had received slurry prior to that (Table 4.5.1). Losses of DOC
would be expected to be greater when animal wastes are applied. Even with
knowledge of the quantities of manures applied, it would be difficult to
predict accurately the resulting increase in DOC export, as this would be
governed in large part by the timing of application in relation to rain events.
To give an indication of amounts which may be expected following
slurry application to heavy soils, an example is given from other work on the
same soil type. Twenty-four 30 m 2 plots receive either nothing, slurry, fer-
tilizer on a nutrient (N, P or K) equivalent basis, or a slurry-fertilizer mix.
Slurry application rates are within limits recommended by the Ministry of
Agriculture Fisheries and Foods (UK). The first application was after the
first silage cut when local farmers were also applying slurry. Shortly after
application, heavy rainfall resulted in the equivalent of 180 kg C ha −1 being
lost from the slurry-treated plots in only 8 days. This is almost twice as high
as the upper end of the range reported by Hope et al . (1994) for annual
losses and one and a half times the exports measured from the HNU plots
of the Rowden experiment during the first 2 months of drainage. This was
from a single slurry application and generally there are four or five applica-
tions during the year. There is therefore clear, but as yet unquantified,
potential for large losses of C (and associated N and P) from intensively
managed grasslands.
Factors Affecting DOC Export
As with all large-scale field experiments carried out with finite resources, a
compromise had to be reached between the size of each plot, the number of
treatments and the degree of replication. The small degree of replication
made it desirable to bulk fertilizer rates within drained and undrained
treatments to maximize the number of degrees of freedom, and, similarly,
drainage treatments were bulked within 'fertilizer N rate'. Thus analysis of
variance (SAS, 1996) could be performed on the quantity of DOC exported
with 9 degrees of freedom, of which 1, 2 and 6 were allocated to drainage,
fertilizer N rate and residual, respectively.
Fertilizer
An analysis of variance across the fertilizer treatments presented here
(drained and undrained plots considered together) showed no significant
effect of fertilizer N application rate on total DOC export over the 2-month
period. However, for the undrained plots, there was a significant ( r 2 = 0.78;
P = 0.048) positive correlation between fertilizer N application rate and
total DOC export. As grassland dry matter productivity is proportional to
Search WWH ::




Custom Search