CROP YIELDS, SOIL NITROGEN AND SOIL ORGANIC CARBON
CONTENT CHANGES UNDER CLIMATE CHANGE
Dumont B.
1,2,*- Basso B.
2- Shcherbak I.
2- Asseng S.
3- Bassu S.
4- Biernath C.
5- Boote K.
6– Cammarano D.
7–
de Sanctis G.
8- Durand J.-L.
9- Ewert F.
10- Gayler S.
11- Grace P.
12– Grant R.
13– Kent J.
14- Martre P.
15–
Nendel C.
16- Paustian K.
14- Priesack E.
5- Ripoche D.
17– Ruane A.
18–
Thorburn P.
19- Hatfield J.
20- Jones J.
21- Rosenzweig C.
18Introduction
Material & Methods
It has been demonstrated that model ensemble is an efficient way to reduce the uncertainty associated with climate change impact on crop growth. (Asseng et al., 2015). Using this approach, wheat and maize grain yields response to temperature increase were simulated by Asseng et al., 2015 and Bassu et al., 2014 using annually re-initialized soil conditions (soil water and nitrogen). However, Basso et al., 2015, showed that, when running models in a continuous mode, yield results differed from the annual reinitialized runs. In this study, we present the results of continuous model runs of the AgMIP wheat- and maize-pilot under temperature and CO2 changes and different management practices.
Simulation protocol
The same factorials temperature, CO2, rain and nitrogen fertilization levels as the one defined in the original AgMIP wheat- and maize-pilots were also simulated (Asseng et al., 2015; Bassu et al., 2014). The same sites (4 locations for each crops) and the calibrated versions of the models were used in this initiative (Table 2).
Additionally, models were also run under
conventional tillage and no-till management
practices.
Finally, modelers were also asked to run their model under annually pre-season reinitialized soil conditions and under continuous running mode. Under continuous running mode, wheat-fallow and maize-fallow rotations were simulated (Table 2).
Initial soil organic carbon (SOC) content were provided for each specific site (Table 3).
Crop model ensemble
Five maize models and seven wheat models involved respectively in the maize- and wheat-pilot initiatives of the Agricultural Model Intercomparison and Improvement Project
(AgMIP) were run in the Soils and Crop Rotations initiative (Table 1).
Results were averaged using the median of model simulations (Martre et al., 2015).
Table 1: Models run in the AgMIP Soils & Crop Rotation pilot
Results
Continuous running mode of crop models allow to better understand the interactions within the soil-plant-atmosphere continuum, in conjunction with variable soil and crop management practices. Ensemble of continuous modelling will provide critical insights when used to identify adaptation and mitigation strategies to climate change.
Conclusions
Figure 2: Impact of CO2 concentration on yield (A-B) and soil N-NO3- (C-D)
content under different temperature (-3 to +9°C) treatments.
This work was supported by the AgMIP (http://www.agmip.org/) project.
Acknowledgements
1 Department Terra & AgroBioChem, Gembloux Agro-Bio Tech, ULg-GxABT, Gembloux, Belgium; 2 Department of Geological
Sciences, Michigan State University, MI, USA; 3 Agricultural and Biological Engineering Department, University of Florida, Gainesville, FL, USA; 4 INRA-AgroParisTech, Thiverval-Grignon, France; 5 Institute of Biochemical Plant Pathology, Helmholtz Zentrum München— German Research Center for Environmental Health, Neuherberg, Germany; 6 Department of Agronomy, University of Florida,
Gainesville, FL, USA; 7 The James Hutton Institute, Invergowrie, Scotland, UK; 8 European Commission - Joint Research Center, Ispra, Italy; 9 INRA-URP3F, Lusignan, France; 10 Institute of Crop Science and Resource Conservation (INRES), University of Bonn, Bonn, Germany; 11 Institute of Soil Science and Land Evaluation, University of Hohenheim, Stuttgart, Germany; 12 Institute for Future
Environments, Queensland University of Technology, Brisbane, Queensland, Australia; 13 Earth Sciences, University of Alberta,
Edmonton, AB, Canada; 14 Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, CO, USA; 15 INRA, UMR759 LEPSE, Montpellier, France; 16 Institute of Landscape Systems Analysis, ZALF, Leibniz-Centre for Agricultural Landscape Research, Muencheberg, Germany; 17 INRA-AGROCLIM, Avignon, France; 18 Climate Impacts Group, NASA Goddard Institute for Space
Studies, New York, NY, USA; 19 CSIRO Ecosystem Sciences, Dutton Park, Queensland, Australia; 20 USDA-ARS National Soil Tilth Laboratory for Agriculture and the Environment, Ames, IA, USA; 21 Department of Agricultural & Biological Engineering, University of Florida, Gainesville, FL, USA; * Passage des Déportés, 2, 5030 Gembloux, Belgium. @: Benjamin.dumont@ulg.ac.be
Authors affiliation
Asseng, S., et al., 2015. Nature Climate Change 5: 143-147. Bassu, S., et al., 2014. Global Change Biology 20: 2301-2320. Basso, B. et al., 2015. Plos One 10(6): e0127333.
Martre, P. et al., 2015 Global Change Biology 21: 911-925
References
Table 2: Factorial analysis
As each site had a different initial SOC content, simulated SOC data were expressed each year (y) as a percentage of the initial value according to :
Table 3: Initial SOC content
Figure 1: Model ensemble of the relative SOC evolution for two wheat
sites (AR, NL) and two maize sites (BR, US)
All individual models globally agreed in the direction of the changes for yield, soil N-NO3- and SOC content
under the different temperature treatments;
Soil N-NO3- was found to increase, while yields and SOC were found to decrease with temperature increases (Fig. 1 and 2);
Model ensemble highlighted that SOC decreases were higher were initial SOC content was higher (Fig. 1);
Yields were overall higher when CO2 increased, whatever the temperature treatment (Fig. 2);
Soil N-NO3- was lower under higher
CO2 concentration (Fig. 2), but increased with temperature under higher CO2 treatments.
