Agricultural soils are a major source of nitrous oxide (N2O) emissions within
the European Union. The Member States of the European Union are obliged to
provide annual inventories of N2O using a simple statistical model based on default
emission factors, described in guidelines produced by the Intergovernmental Panel
on Climate Change (IPCC) that relate N2O emissions to national statistics on inputs
such as the total amount of N fertiliser applied to soils. Regional N2O estimations are
not taken into account by the IPCC. A mechanistic model DNDC (DeNitrification
and DeComposition) was run using a harmonised database of pan-European data
containing data such as soil properties, daily climate, arable crops, mineral and
organic N usage and farm management. One-year simulations for 1997 were made
for 20 crop types within NUTS level 3 provincial regions.
The accuracy of the input data used to run the DNDC model had a significant
impact on the N2O emission estimates, in particular the spatial distribution of arable
crops and scale and accuracy of the SOC data. A simulation for Italy using national
fine scale data (NUTS level 3 crop data and 1 :250,000 measured soil organic carbon
(SOC) data estimated N2O emissions for Italy in 1997 as 44,700 t N2O-N yr·1 • A
second simulation for Italy using European scale SOC data (1 :1,000,000 estimated
SOC) gave an N2O estimate of 76,300 t N2O-N yr·1 • A third simulation for Italy
using European scale crop data (NUTS level 2) and spatially disaggregated to NUTS
level 3 gave an N2O estimate of 99,500 t N20-N yr·1 • The scale at which the model
was run produced a large range in SOC values within each unit, thereby, a large
range in N2O estimates.
A comparison was made DNDC estimates of N2O emissions estimates using
the IPCC methodology. The DNDC modelled emission factor of N2O emission due
to N fertiliser of 0.0083 kg N2O-N ki1 N was lower than the IPCC emission factor
of 0.0125 N2O-N ki1 N. It was shown that the relationship between N2O emissions
and mineral N fertiliser application is not linear. The DNDC model estimates far
higher N2O emissions due to the mineralisation of SOC than the IPCC methodology.
The DNDC modelled results showed a significant variation in the estimations of N
leached from different regions compared to the IPCC default factor that assumes
30% of all mineral N fertiliser and organic manure is leached.
The pan-European database was used to make an estimation of direct N2O
emissions on a European scale at the NUTS 3 level for the first time. Validation of
the European results is difficult due to the paucity of Europe-wide measured N2O
emission data.
This thesis has clearly demonstrated that a mechanistic model and a database
containing national and pan-European data can produce regional estimates of direct
N2O emissions from fertilised agricultural soils at the NUTS level 3 across Europe.
However, uncertainties in the regional estimates of N2O emissions remain due to the
large uncertainties in both the raw and processed data.