 |
| Titel |
Nitrous oxide emissions from irrigated cotton in north eastern Australia |
| VerfasserIn |
P. Grace, D. Rowlings, K. Weier, I. Rochester, R. Kiese, K. Butterbach-Bahl |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250022972
|
|
|
|
|
|
| Zusammenfassung |
| Cotton is one of many agricultural industries in Australia heavily reliant on nitrogenous
fertilizers and water storages to maintain high levels of production. Cotton-based farming
systems are therefore labelled as potentially high-risk agricultural systems with respect to
gaseous losses of nitrogen to the atmosphere.
The on-farm study was undertaken at Dalby in the Darling Downs region of Queensland
in north eastern Australia. The field was furrow irrigated and had been under continuous
cotton (with winter bare fallow) for 10 years. The block was conventionally tilled, with a
spraying regime typical for cotton production in this area. The black clay (with a surface clay
content of 68%) and soil organic carbon content (0-10 cm) of 1.0% and a pH of 8.5, is typical
of the region.
During the the 2006/07 season, soil water (0-50 cm with Enviroscan), mineral nitrogen
(0-10 cm) and crop production data was also collected to develop accurate models for
predicting greenhouse gas emissions as a function of key chemical, physical and biological
processes and specific management events. The 2006/07 experiment also attempted to
directly measure the specific losses of N2O and N2 from a single application of N fertiliser
using 15N isotopically labelled urea.
The automated greenhouse gas measuring system (developed by Butterbach-Bahl et al.)
consists of six chambers connected to sequential sampling unit, a gas chromatograph
(equipped with both electron capture and flame ionization detectors for nitrous oxide
and methane analysis respectively), and a Licor for carbon dioxide. To meet the
demand for high mobility, the sample acquisition and analysis system is trailer
mounted.
During a normal sampling period, the chambers were closed for 90 minutes (unless
temperatures within the chambers exceeded 55oC). The sampling program ensured that that a
single gas sample was drawn back from each chamber every 20 minutes. To facilitate 15N gas
sampling, Swagelok T-pieces were inserted into the the sampling lines and used as sampling
ports for manual removal of 15N labelled gas using evacuated Exetainers for determination of
15N-N2O and 15N-N2 via mass spectrometry. Samples for 15N gas analysis we removed
at 15 and 60 minutes during a single cycle for each chamber, and the flux rate of
15N2O and 15N2 calculated from the linear increase in N emission over that time
period.
As there was little residual N in the profile after the previous crop, the grower applied 200
kg starter N in August 2006, and planted in late October, 2006. An application of 15N
labelled urea (99% atom excess 15N) was made on 2 November in three (of the six)
gas sampling chambers equivalent to 120 kg N/ha (treatment A), and 60 kg N/ha
equivalent in the remaining three chambers (Treatment B). The grower applied 23 and
46 kg N as urea in the irrigation water, 98 and 135 days after the 15N urea was
applied.
A total of 1027 and 688 g N2O-N were emitted from Treatments A and B respectively
during the 150 days after 15N application (with 36 days of missing data). The highest
emissions were recorded after application of water run urea at day 135. A simple multiple
regression model from the N2O data based on surface soil moisture (0-10 cm) and minimum
daily air temperature (oC), estimated 1475 and 617 g N2O-N/ha were emitted over the full
150 days, equivalent to 1.2 and 1.0% of the applied 15N from Treatments A and B
respectively. However, the 15N gas samples obtained 2 months after 15N application indicate
the N2O from the 15N application were only one-tenth of the total amount of N2O (from all
sources) lost on that day. |
|
|
|
|
|
|