Bayesian calibration of the nitrous oxide emission module of an agro-ecosystem model

Lehuger, Simon; Gabrielle, Benoît; Oijen, Marcel van; Makowski, David; Germon, Jean-Claude; Morvan, Thierry; Hénault, Catherine

doi:10.1016/j.agee.2009.04.022

Cited by 86 publications

(66 citation statements)

References 59 publications

(86 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Bayesian calibration of PBMs relies on drawing a representative sample of parameter vectors from the parameter distribution, and this is generally carried out using Markov Chain Monte Carlo techniques (MCMC; Metropolis et al 1953). Bayesian calibration using MCMC has been applied to models for Norway spruce (Van Oijen et al 2005b), N 2 O-emitting fields of rapeseed, winter wheat and maize (Lehuger et al 2009) and the dynamics of soil under grassland during winter (Thorsen et al 2010). MCMC not only allows the use of complex models, such as PBMs, that are not analytically solvable, but it also allows uncertainties about parameters and measurements to be represented by the most appropriate probability distributions; there is no need to use standard distributions such as the multivariate normal.…”

Section: Cultivar-specific Parameter Estimationmentioning

confidence: 99%

“…PBMs are more commonly parameterised site-specifically, even when multiple sites are examined in the same study (e.g. Lehuger et al 2009, but see Reinds et al 2008 Fig. 4 Yields (g m -2 ) for each of the six within-rotation harvests, averaged over all three-year long rotations that were simulated (1995-1997, 1998-2000, ..., 2010-2012).…”

Section: Bayesian Calibration Of a Pbm For Different Cultivarsmentioning

confidence: 99%

See 1 more Smart Citation

Toward a Bayesian procedure for using process-based models in plant breeding, with application to ideotype design

Oijen

Höglind

2015

Euphytica

View full text Add to dashboard Cite

Process-based grassland models (PBMs) simulate growth and development of vegetation over time. The models tend to have a large number of parameters that represent properties of the plants. To simulate different cultivars of the same species, different parameter values are required. Parameter differences may be interpreted as genetic variation for plant traits. Despite this natural connection between PBMs and plant genetics, there are only few examples of successful use of PBMs in plant breeding. Here we present a new procedure by which PBMs can help design ideotypes, i.e. virtual cultivars that optimally combine properties of existing cultivars. Ideotypes constitute selection targets for breeding. The procedure consists of four steps: (1) Bayesian calibration of model parameters using data from cultivar trials, (2) Estimating genetic variation for parameters from the combination of cultivar-specific calibrated parameter distributions, (3) Identifying parameter combinations that meet breeding objectives, (4) Translating model results to practice, i.e. interpreting parameters in terms of practical selection criteria. We show an application of the procedure to timothy (Phleum pratense L.) as grown in different regions of Norway.Keywords BASGRA Á Cold tolerance Á Genotype-environment interaction Á Plant breeding Á Process-based modelling Á Yield stability Terminology This paper aims to bridge two different plant disciplines, i.e. process-based modelling and breeding, so terms need to be defined. In modelling, we distinguish between plant parameters, which are genotype-specific constants, and output variables, which are plant characteristics that are predicted by the model and that vary between environments. Process-based models (PBMs) are dynamic models that represent physiological and morphological processes in plants, and their interaction with the environment. Simulation is the process of specifying environment and parameter values followed by running the PBM to calculate the output variables. In plant breeding, traits are measurable properties of plants that arise from the interaction between the genotype and its environment (G 9 E). The influence of the environment may be large for some traits (e.g. yield) or small (e.g. flower size). A quantitative trait is a property measured on a Electronic supplementary material The online version of this article

show abstract

Section: Cultivar-specific Parameter Estimationmentioning

confidence: 99%

Section: Bayesian Calibration Of a Pbm For Different Cultivarsmentioning

confidence: 99%

Toward a Bayesian procedure for using process-based models in plant breeding, with application to ideotype design

Oijen

Höglind

2015

Euphytica

View full text Add to dashboard Cite

show abstract

“…The first is by liaising with research centres that are doing experiments to obtain direct measurements. A second, and shorterterm solution is to use models such as CERES-EGC (Lehuger et al, 2009;Rolland et al 2008) and feed these outputs to the GHG BN model to improve the accuracy of the current nodes in a similar way as with the agricultural data.…”

Section: Reliability Of the Data Used To Build The Model's Nodes Distmentioning

confidence: 99%

Bayesian Networks for the management of greenhouse gas emissions in the British agricultural sector

Pérez-Miñana

Krause

Thornton

2012

Environmental Modelling & Software

View full text Add to dashboard Cite

“…In order to improve its accuracy, we implemented the NOE2 submodel (Bessou et al 2010b) in CERES-EGC and proceeded to a Bayesian calibration of the 22 parameters of NOE2. This calibration makes it possible to reduce significantly the spread in parameter distribution and the subsequent uncertainty on N 2 O emissions by updating these distributions against a priori probability distributions of parameter values gathered from other sites (Lehuger et al 2009). This calibration reduced the Root-Mean Square Error on daily and annual fluxes by 15% and 63% respectively when compared to the uncalibrated simulations ( Fig.…”

Section: Modelling Of Field Emissionsmentioning

confidence: 99%

Using a crop model to account for the effects of local factors on the LCA of sugar beet ethanol in Picardy region, France

Bessou

Lehuger

Gabrielle

et al. 2012

Int J Life Cycle Assess

View full text Add to dashboard Cite

Background. The results of published Life Cycle Assessments (LCAs) of biofuels are characterized by a large variability, arising from the diversity of both biofuel chains and the methodologies used to estimate inventory data. Here, we suggest that the best option to maximize the accuracy of biofuel LCA is to produce local results taking into account the local soil, climatic and agricultural management factors.Methods. We focused on a case-study involving the production of first-generation ethanol from sugar beet in the Picardy region in Northern France. To account for local factors, we first defined three climatic patterns according to rainfall from a 20-year series of weather data. We subsequently defined two crop rotations with sugar beet as a break crop, corresponding to current practice and an optimized management scenario, respectively. The six combinations of climate types and rotations were run with the process-based model CERES-EGC to estimate crop yields and environmental emissions. We completed the data inventory and compiled the impact assessments using Simapro v.7.1 and Ecoinvent database v2.0.Results. Overall, sugar beet ethanol had lower impacts than gasoline for the abiotic depletion, global warming, ozone layer depletion and photochemical oxidation categories. In particular, it emitted between 28 and 42% less greenhouse gases than gasoline. Conversely, sugar beet ethanol had higher impacts than gasoline for acidification and eutrophication due to losses of reactive nitrogen in the arable field. Thus, LCA results were highly sensitive to changes in local conditions and management factors. As a result, an average impact figures for a given biofuel chain at regional or national scales may only be indicative within a large uncertainty band.Conclusion. Although the crop model made it possible to take local factors into account in the life-cycle inventory, best management practices that achieved high yields while reducing environmental impacts could not be identified. Further modelling developments are necessary to better account for the effects of management practices, in particular regarding the benefits of fertiliser incorporation into the topsoil in terms of nitrogen losses abatement. Supplementary data and modelling developments also are needed to better estimate the emissions of pesticides and heavy metals in the field.3

show abstract

Bayesian calibration of the nitrous oxide emission module of an agro-ecosystem model

Cited by 86 publications

References 59 publications

Toward a Bayesian procedure for using process-based models in plant breeding, with application to ideotype design

Toward a Bayesian procedure for using process-based models in plant breeding, with application to ideotype design

Bayesian Networks for the management of greenhouse gas emissions in the British agricultural sector

Using a crop model to account for the effects of local factors on the LCA of sugar beet ethanol in Picardy region, France

Contact Info

Product

Resources

About