LCA Food 2012 in Saint Malo, France! - Manifestations et colloques ...

PARALLEL SESSION 2B: EMISSIONS MODELLING 8 th Int. Conference on LCA in the
Parallel session 2b: Emissions Mode lling
Agri-Food Sector, 1-4 Oct 2012
Soil, climate and cropping system effects on N2O accounting in the
LCA of faba bean and cereals
Pietro Goglio 1, 2,* , Caroline Colnenne-David 3,4 , Claudia Di Bene 2 , Simona Bosco 2 , Patricia Laville 1 , Romain
Roche 1 , Giorgio Ragaglini 2 , Thierry Doré 4,3 , Marco Mazzoncini 5,6 , Benoit Gabrielle 1 , Enrico Bonari 2
1
UMR INRA-AgroParisTech Environnement et Grandes Cultures, 78850 Thiverval-Grignon, France
2
Landlab, Institute of Life Sciences, Scuola Superiore Sant’Anna, Via Santa Cecilia 3, 56127 Pisa, Italy
3
INRA, UMR 211, BP 01, 78850 Thiverval-Grignon, France
4
AgroParisTech, UMR 211, BP 01, 78850 Thiverval-Grignon, France
5
Dipartimento di Agronomia e Gestione dell’Agroecosistema, Faculty of Agriculture, University of Pisa, Via San Michele degli
Scalzi, 2, 56124 Pisa, Italy
6
CIRAA Interdepartimental Centre for Agroenvironmental Research, via Vecchia di Marina, 6, 56122 San Piero a Grado Pisa, Italy
*
Corresponding author. E-mail: pgoglio@grignon.inra.fr, p.goglio@sssup.it
ABSTRACT
Greenhouse gas (GHG) emissions from soils cause uncertainties within Agricultural LCA. N 2O affects global warming and is estimated
with IPCC guidelines, agroecosystem models or direct measurements. CERES-EGC model was used to estimate N2O emissions
from faba bean and winter cereals grown in two trials (ICC and CIMAS) with different climates. Model outputs were compared
with IPCC estimates. Simulated N2O emission patterns showed that emissions can be independent from fertiliser application dates or
rates. This was due to soil moisture, farming practices such as fertiliser applications and tillage. Results showed the IPCC procedure
estimated higher annual cereals emissions of 740 g N2O-N ha -1 y -1 than simulation results and a lower estimation of 304 g N2O-N ha -1
y -1 for faba bean. Results revealed inclusion of climate, soil properties and management resulted in major variations of N2O emissions
which CERES-EGC was able to capture. Thus, model estimates may increase accuracy of soil GHG emission in Agricultural
LCA.
Keywords: agricultural LCA, N2O emissions, CERES-EGC model, cereals, faba bean
1. Introduction
Agriculture contributed for 18% to the Greenhouse gases (GHG) emissions of France in 2010 and 7% for
the Italian emission in the same year. Despite this contribution in both countries, the agricultural sector was
responsible for 87% of nitrous oxide (N2O) total emissions in France and 69% in Italy (CITEPA, 2012;
Romano et al., 2012). N2O has a global warming potential 298 times higher than carbon dioxide (Forster et
al., 2007).
N2O emissions can be estimated with IPCC emissions factors, agroecosystem models or by direct field
measurements. IPCC methodology is based on emissions factors and accounts for mineral and organic fertiliser
applications. It also considers crop residues decomposition, organic matter mineralisation and other
sources of organic nitrogen (N) (De Klein et al., 2006). N2O emissions from cropping systems may also be
directly measured in field trials using micrometeorological systems or chambers with either automatic or
manual sampling (Hénault and Germon, 1995, 2000; Laville et al., 1999, 2011). N2O field emissions are hard
to detect due to their low concentration variations; therefore high sensibility and costly instruments are
needed. Thus, direct measurements are expensive and time consuming and their use is restricted to a limited
set of experimental fields (Hastings et al., 2010; Laville et al., 2009; Rochette and Eriksen-Hamel, 2008).
Agroecosystem models simulate most of the processes governing N2O emissions and their controls.
They may compare favourably with field observations for assessing GHG emissions, but require detailed
inputs and are complex to run (Del Grosso et al., 2008). For N2O emissions, DAYCENT, DNDC, CERES-
EGC models have been extensively used for various types of ecosystems (Chen et al., 2008; Del Grosso et
al., 2008; Gabrielle and Gagnaire, 2008; Lehuger et al., 2009; Hastings et al., 2010). CERES-EGC was
adapted from the CERES suite of soil-crop models (Jones and Kiniry, 1986), with a focus on the simulation
of environmental outputs such as N2O and NO to the atmosphere. It was tested and calibrated for these environmental
impacts mostly with data coming from Europe, with reasonable success due to a Bayesian calibration
procedure that proved effective when testing the model against independent field data (Gabrielle and
Gagnaire, 2008; Lehuger et al., 2009, 2011).
N2O emissions, like all the other GHG emissions from soils, are a major source of uncertainties within
Agricultural LCA, such as soil erosion, soil organic matter dynamics, biodiversity estimation (Guinée et al.,
2009) and ecotoxicity (Margni et al., 2002; van Zelm et al., 2009). Indeed, in recent years LCA has been
used for a wide range of agricultural systems and proved to be an effective assessment to evaluate environmental
impacts (Nemecek et al., 2011a, 2011b; Goglio et al., 2012). Notwithstanding in most cases soil born
emissions were estimated with the IPCC method which often proved to be less accurate than agroecosystem
models (Gabrielle and Gagnaire, 2008).
155