Previous PageTable Of ContentsNext Page

Nitrous oxide emissions from dry-land wheat in south-eastern Australia

Sally J. Officer1, Frances Phillips2, Roger Armstrong1, Kevin Kelly3.

1 Department of Primary Industries, 110 Natimuk Rd, Horsham, Vic 3401. Email sally.officer@dpi.vic.gov.au
2
Centre for Atmospheric Chemistry, University of Wollongong, NSW 2522.
3
Department of Primary Industries, 120 Cooma Rd, Kyabram, Vic 3620.

Abstract

Nitrogen in agricultural soils is an important source of nitrous oxide (N2O) gas, which has a climate warming potential that is 310 times greater than carbon dioxide. The accurate measurement of N2O emissions from agricultural soils is therefore important for accounting of greenhouse gas emissions. Nitrous oxide emissions are particularly associated with wet anaerobic soil conditions. Studies of wheat soils in relatively dry regions of Australia so far indicate extremely low emissions of N2O, however the studies focused on acid soils. We examined the gaseous emissions of N2O from a typical rainfed cropping soil growing winter wheat in south-eastern Australia. Continuous gaseous measurements from an alkaline Vertosol soil in the semi arid Wimmera region were combined with the continuous measurement of soil water, temperature, mineral N and weather parameters. The equipment used an automated enclosure chamber system that was designed to measure very small fluxes of N2O gas. Data was collected over a single growing season and the following summer fallow period. The cumulative N2O emissions were greater than found in similar Australian studies. Data collection from wheat crops in the Wimmera will continue in order to confirm the results and provide fertiliser emissions factors.

Keywords

Climate change, greenhouse gas, semiarid, N2O

Introduction

Agriculture was responsible for 16% of Australia’s net greenhouse gas emissions in 2006, and contributed 84% of the nitrous oxide (N2O) emissions (Department of Climate Change, 2008b). Nitrogen (N) enrichment of soil was the source of 75% of the agricultural N2O emissions. Nitrous oxide is lost from soils as a result of both denitrification and nitrification processes occurring in soil environments that contain N but are low in oxygen, such as at the sediment-water interface following rainfall, in unsaturated conditions within aggregates, in decomposing litter, and in slightly oxygenated rhizospheres (Wrage et al., 2001, Barton et al., 2008, Hayatsu et al., 2008). Emissions of N2O are therefore related to both the aeration and the amount of N in a soil. Sources of N enrichment in Australian soils include fertiliser use (17%), nitrogen fixation by crops (3.6%), crop residues (5.2%), manure from grazing animals (27%), atmospheric deposition of N compounds e.g. ammonia (24%), and also indirect sources such as leaching and runoff (18%) (Department of Climate Change, 2008b). In terms of the use of synthetic fertilisers, the Intergovernmental Panel on Climate Change uses a default value of 1.25% of all nitrogen fertiliser applied to soil is lost as N2O. Estimates of the losses from fertiliser for Australian crops are generally much less, reflecting the drier nature of Australian agro-ecosystems and lower application rates. The fertiliser emissions factor has now been reduced to 0.3% for rain-fed cropping in Australia (Department of Climate Change, 2008b).

Few comprehensive studies have been made of N2O emissions from rained-fed crops growing in a semi-arid climate, despite many studies in temperate and continental climates (Barton et al., 2008). In this study we measured gaseous emissions from a rainfed semiarid crop in south eastern Australia using automatic chambers. Measurements of gaseous N2O and carbon dioxide (CO2) were made throughout the day, combined with measurements of soil water, temperature, plant available N and climate. We hypothesised that N2O emissions would be small due to the low rainfall in the Wimmera region of Victoria. The system was therefore designed to measure very low soil emissions of N2O. Data was collected over a winter wheat crop season and the following summer fallow period.

Methods

Nitrous oxide emissions were measured at the Plant Breeding Centre (PBC) farm, 8 km west of Horsham, Victoria. The soil type is an alkaline cracking clay Grey Vertosol (Isbell 2002). The 2007 - 2008 trial consisted of winter wheat (Triticum aestivum cv Caliph) subject to three agronomic treatments: (1) Rain-fed only with no N fertiliser, (2) Rain-fed with 50 kg N ha-1 urea fertiliser side-banded at planting and no supplementary irrigation, (3) Rain fed with 50 kg N ha-1 urea fertiliser side-banded at planting, and one episode of supplementary drip irrigation (50 mm) applied on 7th September at the mid tillering growth stage. All treatments were replicated three times. The cultivation (minimum tillage) and fertiliser rates were representative of farming practices in the region. Plant dry matter was sampled at mid till, anthesis and grain maturity (4 x 1 m rows at 2 random places per 14*7 m plot). Grain yield components were measured in quadrats and from the chamber bases. Further irrigation was applied to the chamber bases of treatment (3) in late January and early February, to simulate rainfall events during the summer fallow period.

The gas collection chambers consisted of a moveable box (0.8 m x 0.8 m x 0.5 m high) constructed from stainless steel and Perspex, with height extensions added as required, which was clipped to an open stainless steel base (150 mm depth) that was pressed approximately 100 mm into the soil. Nine automated gas collection chambers (one chamber per agronomic plot) were interfaced to a tuneable diode laser trace gas analyser (Campbell Scientific Inc) to measure the increase in N2O and CO2 concentrations in a closed chamber for half an hour every one and half hours through-out the day. Flux was determined by calculating the rate of change in N2O concentration (corrected to density) from ten measurements over the half hour (Barton et al., 2008). A Licor 820 (St Joseph, MI, USA) was included in the sampling line, with the CO2 flux numbers used predominately for checking. The chambers vented automatically during the closure time if the temperature in the chambers exceeded 50oC, or when the site received rain. Surface soil water content was monitored during the trial by theta probes (Theta-Probe MK2x, Delta-T Devices Ltd) with one probe installed on the soil surface within each chamber base.

Results

Total rainfall recorded over the monitoring period (1 April 2007 to 13 February 2008) was 457 mm, which was above the long term regional average of 445 mm. However, nearly half the rain fell in the month prior to planting in early June. From planting to harvest only 216 mm rain fell so that the growing season rainfall was decile two for the region. Consequently the adequate pre planting soil moisture was not sustained, and soil moisture returned to permanent wilting point (crop lower limit) by grain maturity (Fig 1).

There were no significant treatment effects on the growth of wheat at either mid tillering or anthesis. Growth differences (especially to irrigation) had developed by maturity in December. Grain yields were lowest in the N fertiliser/no irrigation treatment (2.9 t ha-1) compared to 3.1 t ha-1 in the no N fertiliser treatment, and 3.6 t ha-1 in the N fertiliser/irrigated treatment (P=005, L.S.D=0.6). Total N uptake into the shoots was 114 kg N ha-1 for the N fertiliser/no irrigation treatment, compared to 122 kg N ha-1 where no N fertiliser was applied, and 142 kg N ha-1 plants for the N fertiliser/irrigated treatment (P=0.005, L.S.D. 16.2).

The daily emissions of N in N2O (average of 16 measurements) ranged from -0.3 to 13.0 g N2O-N ha-1 day-1, in a very skewed distribution of flux values. The no N fertiliser treatment had the least average daily emissions (backtransformed average 0.91 g N2O-N ha-1 day-1) and the N fertiliser/irrigated treatment was similar (backtransformed average 0.96 g N2O-N ha-1 day-1). The most daily emissions came from the N fertiliser/no irrigation plots (backtransformed average 1.09 g N2O-N ha-1 day-1) (log transformed values of flux in anova, P=<.001, LSD=0.03).

In total, 26% of the emissions occurred prior to planting of the site and a further 53% occurred between planting and applying the irrigation half way through the monitoring period in September. After the supplementary irrigation was applied there was only a small corresponding increase in the emissions from this treatment, and then all emissions remained relatively low for the remainder of the trial, corresponding to the reduced soil moisture (Fig 1). Total cumulative flux from the treatments were 0.42, 0.44, and 0.50 kg N2O-N ha-1 for the N fertiliser/irrigated, no N fertiliser, and N fertiliser/no irrigation treatments respectively (318 days, n.s.d). These trends indicate that the differences in flux were predominantly a result of natural variation, rather than treatment effects.

Nominal fertiliser emissions factors were calculated on a daily rather than annual basis, from the difference between the N2O emitted by the unfertilised and fertilised treatments, divided by the amount of fertiliser applied on a daily basis (50 kg/ha N divided by 365 days). The average emissions factor for the whole monitoring period were 0.14% for the N fertiliser/no irrigation treatment, and -0.06% for the N fertilised/irrigated treatment. When only the cropping season was considered (June to December) the fertiliser emissions factors were 0.14 and -0.26% respectively.

Figure 1: Soil emissions of nitrous oxide (a) and changes in soil moisture (b) from soil growing wheat in the Wimmera region, Victoria, 2007- 08.

Discussion

Prior to rain in April 2007, emissions from bean stubble were approximately 0.5 g of N from N2O ha-1 d-1. At this time, the soil was relatively warm and contained residual nitrogen from the previous bean crop. A heavy rainfall event that started on the 27th April quickly saturated the soil and after 49 mm rain, the emissions increased to average 10.7 g N2O-N ha-1 d-1 from the nine chambers. These results were similar to those from Western Australia, where summer rainfall events accounted for a large proportion of N2O emissions (Barton et al., 2008). In early June, further rain kept the soil relatively wet during light cultivation, planting and the application of the fertiliser treatments. The relatively large N2O emissions continued and even increased during July and August. Soil disturbance by cultivation and the application of fertiliser can both increase N2O emissions (Ball, 2007).

The emissions from the Wimmera wheat site were greater than those recorded in two similar studies of wheat in Australia. Daily N2O emissions from a wheat crop growing in a temperate climate on an acid soil at Rutherglen, Victoria, where 120 kg N ha-1 fertiliser was applied, ranged between 0 and 12.5 g N2O-N ha-1 d-1, totalling no more than 267 g N2O-N ha-1yr-1. The emissions factors estimated in the Rutherglen study were -0.04 and 0.04% of fertiliser N emitted (Barker-Reid et al., 2005). In another study of wheat growing in a semi-arid climate on acid sandy soils in Cunderdin, Western Australia, the daily emissions from direct drilled wheat grown with 100 kg N ha-1 fertiliser ranged between -1.8 and 7.3 g N2O-N ha-1 d-1. The annual total emissions were less than the Wimmera study, at 111 g N2O-N ha-1 from fertilised soils and 90 g N2O-N ha-1 from unfertilised soils. The fertiliser emissions factor was 0.02% of fertiliser N emitted to the atmosphere (Barton et al., 2008). The cumulative emissions from the Wimmera site between March 2007 and February 2008 indicate an annual emission of at least 420 g N2O-N ha-1. While this is relatively large compared to the Rutherglen and Cunderdin studies, a comparable quantity of 470 g N2O-N ha-1yr-1 emissions was recorded for unfertilised wheat growing in a Victorian Mallee ecosystem (Galbally et al., 2005).

The nominal fertiliser emission factors from Wimmera wheat in 2007/08 were 0.14% for the N fertiliser/no irrigation treatment, and -0.06% for the N fertilised/irrigated treatment, which was a wider range than the previous studies, but still many times lower than the 1.25% used in the northern hemisphere. The continuation of the study into a second year and a detailed statistical analysis of the flux results will provide an improved estimate of the fertiliser emission factor for wheat crops in the Wimmera.

Acknowledgments

This research is part of a joint initiative involving DPI Victoria, the Australian Greenhouse Office (now the Department of Climate Change) and the Grains Research and Development Corporation.

References

Barker-Reid F, Gates WP, Wilson K, Baigent R, Galbally IE, Meyer CP, Weeks IA, Eckard RJ (2005) Soil nitrous oxide emissions from rainfed wheat in SE Australia. In: 4th Int. Symp. Non-CO2 Greenhouse Gases (NCGG-4) pp. 25–32. Millpress, Rotterdam, the Netherlands.

Barton L, Kiese R, Gatter D, Butterbach-Bahl K, Buck R, Hinz C, Murphy D (2008) Nitrous oxide emissions from a cropped soil in a semi-arid climate. Global Change Biol. 14: 177-192.

Ball BC, Watson CA, Crichton I (2007) Nitrous oxide emissions, cereal growth, N recovery and soil nitrogen status after ploughing organically managed grass/clover swards. Soil Use Manag. 23: 145-155.

Department of Climate Change (2008a) National Greenhouse Accounts (NGA) Factors (2008). Australian Government, Canberra. Pp 42.

Department of Climate Change (2008b) National Inventory Report (Revised) 2005-Volume 1. Australian Government, Canberra. Pp 219.

Florinsky IV, McMahon S, Burton DL (2004) Topographic control of soil microbial activity: a case study of denitrifiers. Geoderma 119:33-53.

Galbally IE, Meyer CP, Bentley ST, Weeks IA, Leuning R, Kelly K, Phillips F, Barker-Reid F, Gates W P, Baigent R, Eckard RJ, Grace P (2005) A study of environmental and management drivers of non-CO2 greenhouse gas emissions in Australian agro-ecosystems. In: 4th Int. Symp. Non-CO2 Greenhouse Gases (NCGG-4) pp. 47–55. Millpress, Rotterdam, the Netherlands.

Hayatsu M, Tago K, Saito M (2008) Various players in the nitrogen cycle: Diversity and functions of the microorganisms involved in nitrification and denitrification. Soil Sci. Plant Nutr. 54: 33-45.

Isbell RF (2002) The Australian Soil Classification. Revised Edition. CSIRO Publishing, Melbourne, Australia.

Klute A (1986) Water retention: Laboratory methods. In. Methods of Soil Analysis, Part 1: Physical and Mineralogical Methods’, pp. 635-662. Second Edition, Klute (editor). ASA, Madison, WI. U.S.A.

Qian, J H, Doran JW, Weier KL, Mosier AR, Peterson TA, Power JF (1997) Soil denitrification and nitrous oxide losses under corn irrigated with high-nitrate groundwater. J. Environ. Quality 26:348-360.

Rayment GE, Higginson FR (1992) Australian laboratory handbook of soil and water chemical methods. Inkata Press, Melbourne.

Wrage N, Velthof GL, van Beusichem ML, Oenema O (2001) Role of nitrifier denitrification in the production of nitrous oxide. Soil Biol. Biochem. 33:1723-1732.

Previous PageTop Of PageNext Page