Managing soil organic matter is a key issue for a sustainable agriculture as well as for adapting to climate change and mitigating climate change. Despite dedicated research efforts, the processes leading to soil organic matter biodegradation or stabilization and storage are still insufficiently understood for a robust prediction of SOM stocks and associated properties as affected by management and climate change. Recent research resulted in a change of paradigm in the vision of soil organic matter dynamics, this being more controlled in the long-term by SOM interactions with the environment (physical and chemical conditions, interactions with soils mineral constituents, accessibility to microbial decomposers) than by its intrinsic chemical recalcitrance.
In this context and given our previous achievements, our aims for the last 5 years were to:
A - Understand the processes of soil organic matter stabilization and destabilization, and assess their relative importance at different time scales;
B - Understand and predict microbial transformations of carbon in the complex and structured environment of soils at the scale of the microbial habitat, as affected by climate change, focusing on water availability and developing for this innovative modeling approaches;
C - Evaluate the effect of agricultural soil management on soil carbon stocks and organic matter quality.
The research developed at ECOSYS is largely based on the use of long-term experiments (LTE), which are prime research infrastructures in soil science and biogeochemistry (e. g. Chabbi and Loescher, 2017). The LTEs are either managed by ECOSYS (QualiAgro organic wastes, Les Closeaux C3-C4 chronosequence, “42 parcelles” and “36 parcelles” long term bare fallows – see Appendix 2), or through collaborations (Lusignan SOERE-ACBB, INRA Lusignan grassland management, La Cage and SIC alternative cropping systems, INRA Agronomie, Restinclières Agroforestry, INRA System). On these, SOC and SON stocks are measured, a range of chemical and physical fractionation methods are applied as well as biological characterizations (e.g. microbial biomass, PLFAs, SOC and SON mineralization). Soil organic carbon dynamics is deciphered using 13C natural abundance. The research unit also has strong expertise regarding incubation microcosms experiments in which soil structure and the placement of organic substrates and microorganisms may be manipulated. Modeling with compartment models is used to predict stocks and fluxes as well as to test hypotheses and a new generation of models has been developed at the soil microscale.
The expertise of the soil team allows for its implication into national research networks (CarboSMS, Resmo), and national expert assessments (INRA expert assessment on GHG mitigation in agriculture, ongoing INRA expert assessment on SOC storage). The recognition of carbon storage in soils as an additional and potentially important mitigation option at the COP21 has fostered international initiatives and networking in which the team is involved (4p1000 initiative, Global Research Alliance, FAO GSOC) and has raised the demand for political and management decision making support.
A- The different processes that explain soil organic matter persistence are at play in all soils, but their relative importance depends on the pedoclimatic context and soil management (Stockman et al., 2013; Dignac et al., 2017) and is not yet fully evaluated.
During the last 5 years, we:
- Quantified the effect of agricultural management and of the pedoclimatic context on the dynamics of SOM molecular fractions (e.g. Armas Herrera et al., 2016)
- Quantified the effect of grassland management on the physical protection of soil organic matter (Panettieri et al., 2017)
- Investigated the importance of clay mineralogy in the stabilization of soil organic matter by organo-mineral associations (e.g. Barré et al., 2015). Using powerful visualization methods, we showed that a diversity of organic compounds interacted with metal oxides (nanoSIMS, STXM, e.g. Lutfalla, PhD 2015).
Biochars are organic amendments to soils gaining considerable interest in the scientific community. During the last 5 years, we contributed to demonstrate their diversity in terms of decomposability and effect on the SOM via priming (e.g. Paestch et al., 2017).
Isolating SOM compartments with contrasted residence times is an ever-challenging issue in soil science. We proposed to use long term bare fallows, i.e. field experiments kept free of vegetation, in which with time soils become gradually depleted in labile organic matter as it decomposes and relatively enriched in persistent soil organic matter. During the last 5 years, thanks to the European network of Long Term Bare Fallows we have developed, we showed that:
- The mineralization of stable SOC is more sensitive to temperature than that of labile SOC, answering to an active international debate (Lefevre et al., 2014, R. Lefevre, PhD 2015);
- Pyrogenic C decomposes in the long term;
- Physical protection contributed to about a quarter of SOC persistence over 80 years (Paradelo et al., 2016);
- Priming effect does not seem to be quantitatively important in the long term (50 y) (Cardinael et al., 2015);
- Stable SOC has a low energy content, which may be attributed to a combination of reduced content of energetic C–H bonds or stronger interactions between OM and the mineral matrix, explaining partly its persistence in soils (Barré et al., 2016). The use of these Long-term Bare Fallow showed that thermal methods and in particular Rock Eval pyrolysis are very promising method to evaluate the biological stability of soil organic carbon (Cecillon et al., 2018).
B- The insufficiently accurate prediction of SOM stocks and fluxes with current compartment models is now ascribed largely to ignoring the high level of heterogeneity at the particle and pore scale caused by soil structure, which leads to a spatial disconnection between soil carbon, energy sources, and the organisms that are involved in carbon transformations. Our team has invested in the study of the microbial transformations of organic matter at the microscale in soils since a decade now, with an original combination of experimental and modeling studies. In the last 5 years, this topic was developed in the framework of two ANR projects, MEPSOM and Soilµ3D that we coordinated.
We have improved our description of the importance of accessibility in the mineralization of organic substrates (Pinheiro et al. 2015), have shown that the heterogeneity of habitats at the pore scale affected more the mineralisation of labile organic substrates than microbial diversity (e.g. Juarez et al., 2013). Water availability is indeed a strong driver of microbial decomposition of organic matter in the soil structure (Moyano al., 2013).
We developed a series of breakthrough models for describing the mineralisation of organic substrates in the soil structure, that are based on an explicit description of the heterogeneity of soil pore system, water and air distribution and bacteria or fungi and organic matter distribution in the soil structure. These models are based on lattice-Boltzmann formalisms (e.g. Vogel et al., 2015; see Highlight 6) or on a representation of soil pore system with geometrical primitives (e.g. Monga et al. 2014). The spatial distribution of solids and voids is obtained from X-ray computed micro tomography µCT images, although the resolution of these techniques needs improvement when dealing with microbial processes in soils (Baveye et al., 2017). We invested in describing the three-dimensional distribution of water and air in soil pores, using two-relaxation-times lattice-Boltzmann modeling, with the aid of synchrotron X-ray computed tomography studies (e.g. Pot et al., 2015). The spatial distribution of organic matter is ascribed a priori or could be located as well using µCT. For this, we have developed an innovative method for localizing organic matter in µCT tomography (Peth et al., 2014). The spatial distribution of microorganisms is implemented either at random or based on literature reports. We have been able so far to model the mineralisation of simple organic compounds (Monga et al., 2014; Vogel et al., 2015; 2018) and the spread of fungal hyphae in soils.
C- Our research aims also at evaluating the effects of agricultural soil management on soil carbon stocks and organic matter quality, which is needed to identify mitigation options in agriculture (Chabbi et al., 2017; Chenu et al., 2018).
The recycling of OWPs is one of these managements and has been specifically presented in the research line 1. In the last 5 years:
We showed that, regarding grassland management, ley-arable rotations improved GHG emissions and carbon balance compared to continuous arable systems, and also affected the phosphorus cycle (Crème et al., 2016). The introduction of alfalfa in the leys had however little effect on SOC stocks and SOM biochemical composition. We found that the model DaylyDayCent could successfully describe GHG fluxes in these systems (Senapati et al., 2016).
Thanks to long term experiments, we provided the first estimates of additional SOC storage rates for two alternative cropping systems in France: conservation agriculture (i.e. no tillage with a permanent cover crop, Autret et al., 2016,), and alley cropping agroforestry (Cardinael et al., 2015, 2017). Combining measurements of SOC stocks, C inputs to soil and modeling we found that in these different cropping systems, the additional SOC storage could be explained entirely by increased organic inputs to soil (Autret et al., 2016; Cardinael et al., 2018), raising questions about the effects of no tillage on SOC mineralization rates.
Scientific stategy and project
Improving soil organic matter contents and stocks for the ecosystem services it provides, including storing organic carbon and contributing to GHG mitigation requires to :
- Improve the knowledge on the drivers and on the processes involved in soil organic matter dynamics at different spatial scales, including the microscale;
- Improving the tools allowing to monitor the soil organic matter status of soils, such as methods to assess kinetic pool sized of SOC, to evaluate the SOC storage potential of soils, SOM dynamics models and high throughput methods to map soil organic C contents;
- Evaluate the benefits of increased soil organic matter storage, including tacking the question of the soil carbon dilemma “using it or hoarding it?” (Janzen, 2006).
Regarding the processes involved in soil organic matter dynamics, we will contribute to the synthesis effort of the H2020 CSA CIRCASA project Coordination of International Research Cooperation on soil CArbon Sequestration in Agriculture. We will improve microscale mechanistic models describing the decomposition of organic compounds in the soil structure, try to overcome the problem of the “unseen porosity” in CT tomography, when using these to model soil processes, and will implement O2 availability in our current models (L-Bios and Mosaic II), to address also the microscale controls of N2O emissions. We aim to develop new descriptors of the pore scale 3D soil heterogeneity that explain the fluxes measured at the core scale, use our 3D models to connect the µ-scale heterogeneity and the measured macroscale fluxes. To upscale we envision to develop new simple models describing the soil micro-heterogeneity and integrating these micro-features into field-scale models. Experimental background to this activity is the study of the spatio- temporal dynamics of hot spots of mineralization of organic matter, using a combination of visualization methods (µCT, fluorescence microscopy, NanoSIMS) and stable isotope tracing (PhD C. Védère). This research line take place in the ANR Soilµ3D and NanoSoil projects and involve national (e.g. IEES Paris, MISTEA INRA, Geosciences Rennes, UR sols INRA Orléans, CEREGE) and international collaborations (e.g. Univ Cranfield, Technical University Munich, Univ Kassel, NARO Tsukuba).
It is well established that rhizo-deposited carbon is the main source of soil organic carbon (e.g. Rasse et al., 2005). The recruitment of Frederic Rees as a research scientist at ECOSYS to study the eco-physiological drivers of rhizodeposition will allow to develop new and ambitious programs to investigate the fate of the different forms of below ground C (exudates, sloughed cells, mucilages, dead roots) (collaboration with IEES Paris, PSH and EMMAH INRA Avignon, ANR project to be submitted).
To improve the evaluation of the SOC sequestration potential at the territory to regional scale, we will compare several approaches, including a data-driven approach (based on the actual distribution of SOC stocks in agricultural soils), modeling approaches using compartment models (e.g. AMG model), a Tier 2 approach using emission factors from the recent INRA expert assessment (Pellerin et al., 2013), and the C saturation approach. This will be performed in the STORESOILC project (collaboration Géologie ENS Paris, Geosciences UPMC, LSCE Saclay, Infosols INRA, AgroImpact Laon INRA) and in the CE-CARB project (focus on energy crops, AgroImpact INRA Laon). Our group will develop these four approaches on the Versailles Plain territory, building on previous work (see theme organic wastes). In the PEDO- POLYPHEME project of the CNES-TOSCA (2019-21) (collaboration TETIS-IRSTEA, PRODIG-Univ. Paris Sorbonne, SAS, CESBIO, Infosol, MIA INRA), the assessment of spatial distribution of topsoil SOC content will focus on new approaches for spatializing several C-drivers such as organic amendment practices, intermediate crops, and past land use. Efforts will be focused on the quantitative assessment, error and uncertainty enabled by the use of Sentinel 1 and/or 2 (S1/S2) time series.
In the same project we will compare different fractionation methods (Poeplau et al., 2018), to estimate pluri-decadal SOC pool sized in soils and will also contribute to the development of Rock-Eval pyrolysis to do so. The long-term experiments of QualiAgro and La Cage will be used to estimate the persistence and vulnerability of the recently stored carbon.
When the research is placed solely in the perspective of increasing SOC sequestration to mitigate GHG emission the attention focuses on soil C with long residence times. In a wider perspective, kinetically stable organic carbon may not be associated with all the benefits expected from soils richer in organic matter, such as a maintained or improved availability of nitrogen, a better and more stable soil structure and a more abundance, diverse and active soil biota. We will take advantage of our various ongoing or submitted projects and activities (in particular see §1 and structuring theme “Biomasses”) to develop an enrich and more quantitative vision of the benefits and trade-offs of increasing C storage in soils via agroecological practices (in particular organic wastes application, organic agriculture, conservation agriculture, cover crops).