92.1
Plantekongres 2006
DK version 7.4.2
Denitrification in the root zone using
a simple empirical model – SimDen
Only by knowing soil type and amount of nitrogen applied, an estimate of the annual
denitrification can be obtained with the simple empirical model SimDen.
Senior scientist Finn P. Vinther
Danish Institute of Agricultural Sciences
Department of Agroecology
[email protected]
■
How much N is lost by denitrification?
This question we often get from farmers or
environmental authorities, but in spite of many
years’ energetic attempts it is still difficult to
give an unambiguous answer to the question. The
denitrification process has significant importance
both from an agricultural and an environmental
point of view. However, the process is, due to
its high spatial and temporal variability, very
difficult and resource demanding to quantify,
and the only possibility is often to use simulation
models, which, however, usually need more input
information than what is available.
The simple empirical model SimDen (Vinther
& Hansen, 2004) may, however, give a rough
estimate of the average annual denitrification.
SimDen, which is based on a combination of
average results from scientific investigations,
experience and common sense, was developed
for Danish agricultural soils with relatively low
clay contents, but extended to also include more
clayey soils.
The denitrification is a microbial process by
which nitrate is reduced to nitrous oxide (N2O)
and/or to atmospheric nitrogen (N2), which
means that the denitrification can be calculated
as (N2O-emission) x (N2/N2O-ratio) forming
the principle of SimDen. Moreover, due to the
significant effect of soil moisture or water filled
pore space (WFPS) on denitrification, and the
Figure 1. Relationship between clay content and WFPS (left), and
between WFPS and denitrification (right).
435
92.1
Figure 2. Comparison
between measured
values and SimDen
calculated values of
denitrification in the
range 0-300 kg N ha-1
(left) and 0-50 kg N
ha-1 (right).
relationship between WFPS and clay content, it
is possible to find a relationship directly between
soil type/clay content and the denitrification (Fig.
1). A number of studies have shown that WFPS
is a good indicator of denitrification and that the
denitrification is negligible at WFPS <60% (Fig.
1, right). This implies that sandy soils (<10% clay)
in the spring, when soils are at field capacity,
generally have a WFPS <60% (Fig. 1, left).
Further, because these soils are characterised by a
relatively high hydraulic conductivity, they rarely
reach the threshold limit for denitrification (60%
WFPS), and therefore generally have a lower
denitrification than the loamy soils.
In SimDen the N2O emission is derived from
the relationship between input of fertiliser-N and
emission factors as used in the IPCC-methodology
(IPCC, 1997). However, the IPCC emission factor
at 1.25% was modified according to KasimirKlemedtsson & Klemedtsson (2002) suggesting
0.8% of applied N for inorganic fertiliser and 2.5%
for animal manure/slurry. The N2/N2O ratios are
based on literature values (e.g., Weier et al., 1993)
and were in SimDen fitted to clay content using
the relationship shown in the left side of Fig. 1.
This gave N2/N2O ratios at about 1 in soils low
in clay content, increasing to 3-4 in soils with
about 10% clay and up to 9 in soils with clay
content >50%. The calculations are performed in
an Excel spreadsheet, which can be downloaded
from www.agrsci.dk/simden.
A comparison between the denitrification
calculated with SimDen and the denitrification
simulated with the soil-plant-atmosphere model
Daisy showed that SimDen slightly underestimated
the denitrification under conditions with high
precipitation and overestimated with low
precipitation, whereas there was a good agreement
under medium climatic conditions (Vinther &
Hansen, 2004). Further, SimDen was compared
with measured values from the Netherlands,
Sweden, Germany and Spain, all included in
436
Fig. 2. This comparison shows that SimDen is
not able to reach the highest measured values,
which probably were measured under special
environmental conditions, as for example a
shallow groundwater level. In the lower range
of values there is a reasonably good agreement
between the measured and the calculated
denitrification (Fig. 2, right).
Considering the large amount of resources
needed for measuring the denitrification or the
difficulties encountered when using dynamic
simulation models, it is concluded that this simple
denitrification model with advantage can be used
for obtaining an estimate of the denitrification
under ‘normal’ climatic conditions. If more detailed
information is needed, i.e. regarding temporal
variation under certain climatic conditions, dynamic
models as for example Daisy must be used.
Literature
IPCC. 1997. Greenhouse Gas Inventories. Revised
1996 IPCC Guidelines for National Greenhouse
Gas Inventories. Intergovernmental Panel on
Climate Change.
Kasimir-Klemedtsson Å & Klemedtsson L. 2002.
A critical analysis of nitrous oxide emissions
from animal manure. In: (ed. Petersen S. O. &
Olesen J. E.) Greenhouse gas inventories for
agriculture in the Nordic countries. DIAS report
No.81 Danish Institute of Agricultural Sciences,
Foulum, Denmark, pp. 107-121.
Vinther FP & Hansen S. 2004. �����������
SimDen - a
simple empirical model for quantification of N2O
emission and denitrification. DIAS Report 104,
1-47 (in Danish with English summary).
Weier KL, Doran JW, Power JF & Walters DT.
1993. Denitrification
�������������������������������������������
and the dinitrogen/nitrous
oxide ratio as affected by soil water, available
carbon, and nitrate. Soil Science Society of
America Journal 57, 66-72.
■