Storage and release of carbon in soils

HAL Id: hal-01190694

https://hal.archives-ouvertes.fr/hal-01190694

Submitted on 1 Sep 2015

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Storage and release of carbon in soils

Jorge Sierra

To cite this version:

Jorge Sierra. Storage and release of carbon in soils. The Green book of the CLIMATOR project,

ADEME Editions, pp.105-112, 2011. �hal-01190694�

4

B

matter

A Concepts and definitions

The organic matter (OM) contains between 50 and 60% of carbon (C). Among its functions, the OM helps to preserve the structure and porosity of the soil (thus influencing water storage, aera- tion, and the risk of erosion), to stimulate biological activity and preserve the soil biodiversity, to supply nutrients to the plant (nitrogen, phosphorus, sulphur etc.) and to retain certain micropol- lutants (thus affecting water quality). The organic C content of the soil is the result of a balance between inputs and outputs over a given period. This balance may be positive (storage), nega- tive (release) or nil. Variations in the balance on a field or farm scale, due to a change in land use or farming practices, affect all the agro-environmental functions just mentioned, and consequently the physical, chemical and biological quality of the soil as a whole. On the planetary scale, the quantity of organic C in the soils represents about three times that stored in the vegetation and twice that present in the atmosphere. This implies that a variation in the content in soils, for ex- ample due to climate change*, could have considerable impacts on the atmospheric C, involving an increase (with release) or a mitigation (with storage) in global warming.

The quality and quantity of the inputs to the balance are determined by the vegetation and the management of the cropping system*. For example, in agriculture, the main inputs are the crop residues, the roots, and organic manures (compost, FYM etc.) In grassland, the inputs are animal excreta (on pasture) and dead and recycled roots following grazing or cutting. These C inputs undergo processes of mineralisation* and humification*, whose rates depend on the quality of the residues and the manures. The outputs of the soil OM balance arise mainly from its miner- alisation and, in certain cases, from the erosion and leaching of its soluble organic compounds.

The mineralisation and humification are carried out by the soil microfauna (insects, earthworms, etc) and microflora (bacteria and fungi), and are affected by the physical and physico-chemical conditions of the soil (temperature, water and clay content, pH, etc.).

What is the effect of climate change in terms of storage/release of C? What are the variables which contribute to the vulnerability of the soil? What factors or processes can reduce a trend towards release? Should we expect a geographical gradient in the impact of climate change? These are the questions which we will try to answer in this chapter, after having revisited the main mecha- nisms involved on the response of the soil OM to climate change.

B Mechanisms involved

Figure 1 shows, in a simplified way, the C balance of a cropping system affected by climate change.

Storage and release of carbon in soils

Jorge Sierra

| Green Book | The topics | Organic matter | Jorge Sierra

B 4

106

matter

Regarding the outputs, they are heavily dependent on the factors which control the biologi- cal activity of the soil, including mineralisation, mainly temperature and water content. The temperature plays a major role in the process of mineralisation: in the absence of other limit- ing factors, an increase of 10°C can double its rate. The effect of temperature on mineralisation explains the lower OM content commonly found in tropical soils (because mineralisation con- tinues throughout the year) compared with temperate soils (with slow mineralisation in winter).

However the micro-organisms can progressively adapt to climate warming and increase their optimum temperature for metabolism, which would have the effect of reducing their response to increasing temperature. Also, seasonal changes in ground cover (e.g. by perennial or annual crops) and farming practices (e.g. irrigation) have an effect on mineralisation by changing the soil temperature and water content.

In natural or unfertilised systems, the nitrogen resulting from mineralisation is one of the factors which limits production of plant biomass (the green dashed line in figure 1). On the other hand in more intensive cropping systems, where the nitrogen is supplied mainly in the form of fertiliser, the yield of biomass is less dependent on the mineralisation rate. In this case, the inputs and out- puts of C of the system act as flows which are almost independent of one another.

Figure 1: factors and processes involved in the soil C balance.

The examples given as evidence of the impact of climate change on the C content and the vulner-

ability of the soil result from a whole complex of interactions (e.g. the climate/plant/soil or tem-

perature/CO

interactions). Hence a change in a given climatic factor (e.g. temperature increase)

may have simultaneous positive effects (e.g. an increase in plant biomass) and negative effects

(e.g. increased mineralisation) on the balance.

4 B

matter

C Changing trends : effect of the cropping system and site*

Table 1 explains the notation used for each cropping system. Figure 2 shows the variation in the OM in the near future* (NF) and the distant future* (DF). The positive and negative values of vari- ation indicate an increase or a fall in the OM content respectively. By considering the mean values of storage/release the cropping systems can be classified as: positive, mostly corresponding to rotations (SWSgW, ORG and MWRW) and to FG in the DF, moderately positive (WW), neutral (V and MI in mainland France), rather negative (SS), and negative (tropical cropping systems: MI and BA in Guadeloupe). In order to understand these forms of behaviour, one has to specify cer- tain choices for the management of residues, which were made outside the simulations. Table 1 shows the residues assumed to be incorporated for each cropping system. Straw not incorpo- rated is assumed to be exported for livestock.

Systems SS

Sunflower monoculture

Irrigated maize MI monoculture

V Rainfed vines

Wheat WW monoculture

Rainfed fescue FG grassland

Residue straw and roots

stubble and roots (straw and roots

for tropical MI) wood stubble

and roots roots + senescent biomass

Systems MWRW

Maize-soft wheat-rape-hard wheat rotation

ORG Soft wheat- fescue-fescue-

pea rotation

SWSgW Sunflower- wheat-sorghum-

durum wheat rotation

BananasBA

Residue stubble for soft wheat and maize straw for rape and hard wheat

+ roots

stubble for soft wheat straw for peas and roots (W, P, G)

straw and roots aerial biomass and roots Table 1: notation and residues incorporated in each cropping system.

The gradation observed in the storage capacity of rotations (e.g., SWSgW > ORG > MWRW in figure 2) is largely explained by the different quantity of residues applied (e.g. highest for SWSgW, cf. table 1). In the case of the ORG system, the manure applied for wheat (cf. AGRICULTURE section) and the large root biomass of the forage grass help to increase C storage in comparison with MWRW.

On the other hand for the annual crops grown in monoculture (WW, MI and SS), these differences

| Green Book | The topics | Organic matter | Jorge Sierra

B 4

108

matter

The most negative systems for storage are the tropical systems (fig. 2). In the case of MI, the release is due mainly to the drastic reduction in yields (cf. WEST INDIES section) and in the returns.

For BA, the yield and the returns are rather stable over time, and the factor which drives the release is the mineralisation, favoured by the increases in temperature and the soil water content caused by the rain.

Figure 2: variations in the OM content for different cropping systems and sites (cf. AGRICULTURE section for the cropping system notation and the components of rotations).

As well as the differences between cropping systems in France and the tropics, figure 2 shows that there is no clear geographical trend. For the systems storing the most OM in the DF (SWSgW, ORG, MWRW and G), there is however a slight tendency to more storage as one goes from the south-west to the east (e.g. from Toulouse to Dijon in figure 2) which is partly associated with a smaller reduction in rainfall (cf. CLIMATE section). It seems therefore that the effect of climate change on the OM is localised and depends on the interactions between the cropping system and the site, figure 2 showing a very variable behaviour from one site to another for the same cropping system.

The test for thermal* adaptation of soil micro-organisms has shown that this mechanism can play

a significant role by reducing losses of C by mineralisation (fig. 3). On average, the magnitude of

this reduction of losses is estimated at about 35%. Furthermore, for MI at Toulouse, taking ac-

count of the adaptation varies its behaviour, which passes from slight release in the NF to slight

storage in the DF. It seems therefore that the adaptation of the micro-organisms could partially

change the trends discussed above, by reducing the release and increasing the storage in the

different cropping systems.

4 B

matter

Figure 3: effects of thermal adaptation of the micro-organisms on the variation in OM content for the two systems and the two regions tested. (–) without adaptation, (+) with adaptation.

D Dynamics of the organic matter and yield

Is variation in yield a good indicator of the variation in soil OM? In this section we will analyze that question. With figure 1 we have shown that the behaviour of the plant and that of the OM can be

“linked” in two ways: firstly: plant biomass return input to the OM and mineralisation of the OM plant nutrition plant biomass. The second way can be ignored in this analysis, as all the cropping systems were fertilised and in consequence the plant nutrition depended little on the nitrogen released by mineralisation.

Figure 4: relations between the yield variations (tons/ha) and OM (tons C/ha). Each short line connects the value for the NF (without a symbol) with the value for the DF (marked with a triangle, whose colour indicates the site concerned).

Figure 4 shows the relation between the variation in yields (

YIELD) and that of the OM content

for the WW and SS systems. A simple and direct relationship between yield and OM (i.e. a

simultaneous increase or decrease over time) would be indicated by the presence of lines in

quadrants 2 and 3 (numbered in red) only, and aligned in the direction indicated by the black

dotted arrows. Figure 4 shows that the distribution of the points and the orientation of the lines

do not, in most cases, follow these trends. Similar results were obtained with other cropping

| Green Book | The topics | Organic matter | Jorge Sierra

B 4

110

matter

E Sources of uncertainty and variability (including geographical)

In this subsection we complete the analysis of the variables which affect changes in the OM, by referring to figure 5, which is based on an analysis of variance described in the section UNCER- TAINTY AND VARIABILITY. This analysis is able to rank different sources of variability (systems, sites, soils, years) for two periods of climate change (RP-NF or RP-DF).

Figure 5: sources of variability and uncertainty for ΔOM. The values represent the percentage of the total variability.

The interactions

As a whole, interactions are responsible for nearly 60% of the total variability in

OM, which is much more than their effect on the “plant” variables (cf. YIELD section) and on other “soil”

variables (cf. WATER section). Moreover, for the four variables analysed (fig. 5) the effect of the interactions is always more than the main effects. The large effect of interactions confirms what we discussed in relation to figure 1 and implies that the effect of a given variable (for example the “site” effect) cannot be ascertained completely without knowing the status of other variables (e.g. the “site” effect for a particular cropping system).

The cropping systems

Their effect was discussed in § C. This is the main source of variation in ΔOM which is responsible for nearly a third of the total variability (i.e. the sum of main effects and interactions).

The sites

The “site” effect was dealt with partially in § C. It is the second effect in order of importance (fig. 5) It can be visualised by considering the mean of the

OM of each site (fig. 2): the sites storing the most are Dijon, Toulouse and Mons in the NF, and Dijon, Lusignan and Mons in the DF. Similarly, the sites losing the most OM were Bordeaux, Avignon and Saint-Étienne in the NF and Avignon, Colmar and Bordeaux in the DF. We can see that for the two situations of storage and release, two sites always appear among the most striking in the NF and in the DF. This reflects a certain stability in the mean response of sites to climate change, which is the cause of this effect. To describe the behaviour of each particular site would be beyond the objectives of this chapter, considering that the interactions are numerous and larger than the main effects (fig. 5).

Climate change (the period)

This effect reflects the variation in the influence of climate change over the course of time. The change in the behaviour of FG between the NF and DF (fig. 2) is a good example of the “period”

effect. Similarly, for the rotations, the Bordeaux site has varied its rather neutral behaviour in the

NF to one of storage in the DF.

4 B

matter

The soil

This factor has not been discussed until now, as the results shown in § C and D are for simulations made with a single soil. In this study, the “soil” effect concerns mostly differences in depth (be- tween 90 and 215cm) and the initial OM content (between 1.4 and 2.3%), with a positive relation- ship between depth and OM content (for example the deepest soil is also the one with the most OM). In a climatic scenario characterised by a reduction in rainfall, the soil depth plays a signifi- cant role in the water supply due to differences in the amount of water stored in the profile, and hence on the yield of biomass and the volume of residues. However the main effect of the soil concerns its OM content via the effect of mineralisation. Thus the soil which is richest tends to release more in release situations and to store less in storage situations, which explains why the interactions are much bigger than the main effect of the soil (fig. 5). This phenomenon is associ- ated with the larger quantity of substrate for mineralization in the soils with the most OM.

The year

The variation in OM is less sensitive to this factor than other variables such as yield (cf. YIELD sec- tion) and percolation (cf. WATER section). These variables are very dependent on the weather con- ditions in a particular year and retain little “memory” of previous years. The OM represents the op- posite situation: it is more “conservative” and depends heavily on the longer-term soil “memory”.

What you need to remember

3 Nearly 60% of the impact of climate change on the dynamics of OM is due to cropping system/site/period/soil interactions. This supports the idea that the changes in soil OM should be analysed by taking account of all the variables in the agrosystem.

3 Among these variables, the cropping system is the one with the biggest effect on the OM. It is involved in the returns via the management of residues (for example, ploughed in or carted off) and in mineralisation via the ground cover and farming practices (irri- gated or rainfed cropping). These factors explain the differences between rotations and grassland (which store carbon) and annual crops which may or may not, and between the annual crops (WW, C store; MI in France, neutral; SS, C release; and MI tropical, C release).

3 For the fertilised systems analysed, the variation in yields is not a good indicator of the variation in OM.

3 Some of the sites tested were quite stable over time in their storage behaviour. (Dijon, Lusignan, Mons) or negative (releasing C) (Avignon, Guadeloupe), but without any par- ticular geographical trend in France.

3 The soil effect is largely due to its OM content MO: the soils with the most OM store less and release more, according to the cropping system.

3 The dynamics of OM are weakly affected by year-to-year weather variation and they

reflect rather the effects of the components of the system in the medium-long term.

| Green Book | The topics | Organic matter | Jorge Sierra

B 4

112

matter

What needs further study …

3 The response of forage grasses to climate change merits special attention in view of their effect, in monoculture and rotation, on the soil’s C storage capacity. It is largely a question of the effect of climate change (water stress, CO

) on the senescence of the aerial biomass.

3 Through its role in reducing the impact of climate change, the change in the biological activity of the soil, in particular mineralisation, should be given priority from the point of view of thermal adaptation and the effects of soil wetting/drying sequences. In this respect, recent advances have been made experimentally and could be progressively incorporated into the models.

3 Finally it is worth recalling that the variation in the quantity of OM involves changes in other soil properties, such as the water storage capacity, the structure, and even the quality of the OM. These properties should be increasingly included in models in order to predict the changes in the soil agro-environmental functions under the impact of climate change.

To find out more …

FAO. 2008 - Nouvelle base de données mondiale sur les sols.

http://www.fao.org/newsroom/fr/news/2008/1000882/index.html

Maron P.-A., 2008 - Dimimos : Lien entre la diversité microbienne et le turnover des matières organiques dans les sols agricoles. Projet ANR (Systerra).

http://www.inra.fr/content/download/15052/258575/version/1/file/6.DIMIMOS.pdf

Persillet V., 2007 - Le changement climatique : les enjeux pour le secteur agricole. Notes de service INRA-SAE2 : Mieux comprendre l’actualité.

http://www.inra.fr/Internet/Departements/ESR//comprendre/js/climat.php Stengel P., Gelin S., 1998 - Sol : interface fragile. INRA Ed., Paris. 222 p.

UE., 2006 - Directive du Parlement européen et du Conseil définissant un cadre pour la protection des sols.

http://europa.eu/legislation_summaries/agriculture/environment/l28181_fr.htm UE., 2008 - Climsoil. Rapport final.

http://ec.europa.eu/environment/soil/pdf/climsoil_report_dec_2008.pdf