Elsevier

Atmospheric Environment

Volume 36, Issue 6, February 2002, Pages 917-928
Atmospheric Environment

Development and application of a mechanistic model to estimate emission of nitrous oxide from UK agriculture

https://doi.org/10.1016/S1352-2310(01)00512-XGet rights and content

Abstract

A mechanistic model of N2O emission from agricultural soil (DeNitrification-DeComposition—DNDC) was modified for application to the UK, and was used as the basis of an inventory of N2O emission from UK agriculture in 1990. UK-specific input data were added to DNDC's database and the ability to simulate daily C and N inputs from grazing animals and applied animal waste was added to the model. The UK version of the model, UK-DNDC, simulated emissions from 18 different crop types on the 3 areally dominant soils in each county. Validation of the model at the field scale showed that predictions matched observations well. Emission factors for the inventory were calculated from estimates of N2O emission from UK-DNDC, in order to maintain direct comparability with the IPCC approach. These, along with activity data, were included in a transparent spreadsheet format. Using UK-DNDC, the estimate of N2O-N emission from UK current agricultural practice in 1990 was 50.9 Gg. This total comprised 31.7 Gg from the soil sector, 5.9 Gg from animals and 13.2 Gg from the indirect sector. The range of this estimate (using the range of soil organic C for each soil used) was 30.5–62.5 Gg N. Estimates of emissions in each sector were compared to those calculated using the IPCC default methodology. Emissions from the soil and indirect sectors were smaller with the UK-DNDC approach than with the IPCC methodology, while emissions from the animal sector were larger. The model runs suggested a relatively large emission from agricultural land that was not attributable to current agricultural practices (33.8 Gg in total, 27.4 Gg from the soil sector). This ‘background’ component is partly the result of historical agricultural land use. It is not normally included in inventories of emission, but would increase the total emission of N2O-N from agricultural land in 1990 to 78.3 Gg.

Introduction

Signatory states to the United Nations Framework on Climate Change (UNFCC) are required to produce an annual national inventory of nitrous oxide (N2O) emission from all anthropogenic sources. This inventory is intended to advise the magnitude and change in total emissions and gives an indication of which sectors of activity are responsible for the greatest emissions. Estimates of the contribution of agriculture to the total UK emission have indicated an increase over recent years as industrial sources (particularly adipic and nitric acid manufacture) have declined (Salway et al., 2001). Salway et al. (1999) estimated that in 1990 47% of the total UK emission was from agriculture (63.9 Gg N2O-N). In order to comply with the Kyoto Protocol on greenhouse gas emissions, 1990 levels of N2O emission from EU countries are required to be reduced by 8% in the period 2008–2012. Mitigation strategies require an agreed inventory against which measures to reduce N2O emissions from agricultural sources can be judged.

The framework that is used currently to estimate sources of N2O from UK agriculture is that produced by the Intergovernmental Panel on Climate Change (IPCC, 1997). This approach utilises emission factors (EFs) which specify the proportion of N input that is emitted as N2O, and are used with activity data (such as animal numbers, fertiliser use and crop areas) to calculate emission. These EFs are derived from field measurements at sites in a variety of countries, with different soil types, climate and crops (IPCC, 1997). Default EF values are provided, but each has a large range, which reflects the contrasting background conditions of the measurements and leads to a large uncertainty in the emission estimate (Brown et al., 2000). An approach is required which assigns some of this variation to its controlling variables by producing an estimate that takes into account factors such as soil, climate and fertiliser type. Possible approaches include the development and application of more detailed empirical relationships, which would account for such factors, or the use of a suitable mechanistic model to estimate emissions. This study used the rainfall-driven process-based model called DNDC (DeNitrification-DeComposition) (Li et al., 1992), originally written for USA conditions, to provide the basis for an improved UK N2O emission estimate. The development of this model for the UK, and the estimates of N2O emissions from agricultural land for 1990 derived from it, are discussed in this paper.

Section snippets

Model description

The DNDC model (Li et al., 1992) was selected to produce an annual UK emission estimate because it has reasonable data requirements, has produced robust estimates for other countries and is suitable for simulation at appropriate temporal and spatial scales for this exercise. DNDC is a process-oriented simulation model of C and N biogeochemistry in agricultural ecosystems, developed to assess nitrous oxide (N2O), nitric oxide (NO), dinitrogen (N2), ammonia (NH3) and carbon dioxide (CO2)

UK-DNDC-based estimates of N2O emission

The total N2O-N emission from UK current agricultural practice in 1990 was estimated to be 50.9 Gg, comprising 31.7 Gg from soil, 5.9 Gg from animals (this does not include emission from N deposited while grazing or from applied slurry and FYM, which are included in the soil sector) and 13.2 Gg from the indirect sector (Table 1). Of the emission from soil in the UK, 77% was attributed to fertiliser, 17% from applied animal waste and 5% due to emission from N deposited while grazing. The large

Comparison with IPCC estimate

The total emission estimate from current agricultural practice using UK-DNDC (50.9 Gg N2O-N) was smaller than the emission estimate of 87 Gg N2O-N obtained using the IPCC (1997) methodology with default values (Brown et al., 2000) (hereafter referred to as the IPCC (1997) methodology), and the 63.9 Gg N2O-N estimated by Salway et al. (1999). In the soil sector, emission from fertiliser represented a larger proportion and amount of the soil total than was predicted by the IPCC (1997) methodology.

Conclusions

  • 1.

    Development and application of DNDC (Li et al., 1992) to the UK produced a model capable of simulating the application of and N2O emission from both fertiliser and different forms of animal manures in agricultural land in the UK.

  • 2.

    This attempt to provide a national inventory of N2O emission based on the UK-DNDC model has a number of advantages over an IPCC-type approach. The greater detail of its calculations allows the effect of contrasting soils, crops, climates and farming practices to be

Acknowledgements

This work was funded by MAFF, London (Project No. OC9601). IGER, SRI and IACR are supported by BBSRC. We are grateful to Mark Rounsevell (now University of Leuven) for his part in initiation of the project. Input data were supplied by Crawford Jordan, John Wallace and Joe O’Reilly (DARDNI), Jim Gauld (MLURI), the British Atmospheric Data Centre, Ken Smith (ADAS), Clive Rahn (HRI), Brian Coulter (Teagasc), Keith Weatherhead (Cranfield University) and Ken Stewart (OSNI). Data for model validation

References (46)

  • Burton, R.G.O., Hodgson, 1987. Lowland Peat of England and Wales, Special Survey No. 15. Soil Survey and Land Research...
  • Clayden, B., Hollis, J.M., 1984. Criteria for differentiating soil series. Soil Survey Technical Monograph No....
  • H Clayton et al.

    Nitrous oxide emissions from fertiliser grasslanda two year study of the effects of N fertiliser form and environmental conditions

    Biology and Fertility of Soils

    (1997)
  • DANI, 1991. The Agricultural Census in Northern Ireland. Results for June 1990. Department of Agriculture, Northern...
  • K.E Dobbie et al.

    Nitrous oxide emissions from intensive agricultural systemsvariations between crops and seasons, key driving variables, and mean emission factors

    Journal of Geophysical Research

    (1999)
  • Fowler, D., Hargreaves, K.J., Skiba, U., 1999. Direct measurements of the UK source strength of radiatively active...
  • R.J Halley et al.

    Primrose McConnell's The Agricultural Notebook

    (1988)
  • R.M Harrison et al.

    Effect of fertilizer application on NO and N2O fluxes from agricultural soils

    Journal of Geophysical Research

    (1995)
  • Hodgson, J.M., 1976. Soil Survey Field Handbook. Soil Survey Technical Monograph No....
  • J.M Hollis et al.

    The measurement and estimation of saturated soil hydraulic conductivity. Unpublished report to MAFF

    (1989)
  • IPCC, 1997. Greenhouse gas emissions from agricultural soils. In: Houghton, J.T., et al., (Eds.), Greenhouse Gas...
  • S.C Jarvis et al.

    The effects of grassland management on nitrogen losses from grazed swards through ammonia volatilisation; the relationship to excretal N returns from cattle

    Journal of Agricultural Science, Cambridge

    (1989)
  • C Li

    Modeling trace gas emissions from agricultural ecosystems

    Nutrient Cycling in Agroecosystems

    (2000)
  • Cited by (0)

    View full text