Skip to Main content Skip to Navigation
Theses

Effet de travail du sol sur les stocks et flux de C et N dans un sol limoneux de grandes cultures du bassin Parisien

Abstract : For many centuries, the conventional mouldboard tillage system was used in agriculture to control the development of weeds, to incorporate crop residues into the soil, to recycle leached nutrients back to the surface and to create an adequate structure before planting. However, after the development of herbicides the need for ploughing was questioned and reduced tillage systems were introduced. These reduced tillage systems have two main characteristics: (i) the soil is not entirely turned over and (ii) the soil is always entirely or partially covered by residues. The shift from mouldboard ploughing to no-tillage so induces changes in the soil structure and in the location of soil organic matter and crop residues. This results in changes in soil climate (soil temperature and soil water content) and in several biological, chemical and physical soil properties. The combination of all these modifications has an important impact on C and N transformations in the soil. The overall objectives of this work were twofold. First, we quantified the changes in C and N pools and in C and N fluxes between different long-term (32 years) tillage systems in cereal cropping systems in northern France, and second, we studied the effects of soil climatic conditions, soil structure and biological and physical properties of the soil on the differences in the C and N cycles between those tillage systems. This work focused mainly on those parameters with important agronomical or environmental impacts: soil organic C and N contents and distribution, soil mineral N dynamics and CO2 and N2O emissions. Two contrasting tillage systems were considered, i.e. conventional mouldboard ploughing to 20 cm depth (CT) and no-tillage (NT). These systems were studied on two different sets of plots with a maize-wheat rotation on the same experimental site at Boigneville in the Parisian Basin in Northern France. After 32 years, NT presented 5-15% larger C stocks and 3-10% larger N stocks compared to CT, but these differences were not always statistically significant. Soil organic C and N concentrations decreased with increasing depth in NT, whereas they were relatively homogeneously distributed through the plough layer in CT. The small stock differences were further explored by examining the changes at different levels of structural complexity. Mineral-associated N and particulate organic matter each accounted for about 50% of the total difference in N stock. However, 66% of the total difference in C stock was due to differences in the particulate organic matter (58%) and free residues (8%) fractions. Additional C and N were detected in NT in the water stable macroaggregates. Our results suggest that the larger C and N stocks in NT are attributed to (i) enhanced macroaggregate formation in the 0-5 cm layer due to higher microbial activity and SOM content and (ii) a better protection of soil organic matter in the 5 20 cm layer due to a larger proportion of small pores and lack of soil disruption by tillage or climate. The tillage systems did not induce large differences in water and nitrate content in the 0 120 cm soil profile. When the LIXIM model was applied to these data the calculated ‘in situ' N mineralisation rates, expressed both in calendar days and in normalised days (for soil temperature and moisture content), were comparable in both tillage systems and clearly demonstrated that the soil N supply in both systems was comparable. NT always tended to emit more N2O than CT. In addition, CT or NT emitted the larger amount of CO2 in the absence of plants depending on the weather conditions (rainfall and temperature) and the amount and location of crop residues. The cumulated CO2 emissions for the specific weather conditions of the measurement year were significantly larger for NT than for CT. In the second part of our work we studied the effects of differences in soil climatic conditions, soil structure, organic matter location and soil biological and physical properties between the tillage systems on the observed differences in the C and N cycles. We first determined whether the differences in C and N stocks and fluxes in CT and NT were due to changes in the potential decomposition rate of the SOM. Our results clearly showed that after 32 years the potential C and N mineralisation of soil organic matter under controlled conditions (temperature and soil water pressure) was not smaller in NT compared to CT. The physical protection of the soil organic matter against mineralisation was evaluated by incubating soil samples after soil structures between 50 µm and 12.5 mm had been progressively destroyed. The samples were taken from four structural zones of the CT and NT plots: loose and dense structural zones in the plough layer of CT and the 0-5 and 5-20 cm soil layers in NT. Our results indicate that the structural zone with the largest C and N stocks and the largest amount of water stable aggregates (0-5 cm soil layer of NT) showed the smallest increase in N mineralisation and no increase in C mineralisation after soil structure disruption. Of the four structural zones, the 5-20 cm soil layer of NT showed the largest effect of physical protection of SOM. Secondly, our measurements indicated that differences in soil temperature and soil water content between CT and NT induced differences in ‘in situ' soil organic matter decomposition. These differences were often small and not systematically more favorable to decomposition over time in a given tillage system. On the other hand, a large influence of the distribution and amount of rainfall and water evaporation on the dynamics of the CO2 fluxes was observed. In NT, rainfall induced considerable residue decomposition and, consequently, bursts of CO2 emissions due to a sudden increase in the water content of the surface residues. However, after a rain event, the water content of the surface residues fell rapidly and, again, seriously limited their decomposition resulting in smaller CO2 emissions compared to CT. Finally, C and N fluxes were simulated using the PASTIS model. Modelling provides a better understanding of the individual effects and interactions of the determining factors on C and N dynamics. The simulations showed that the larger cumulative total CO2 fluxes in NT resulted from a more extensive crop residue decomposition and not from an enhanced SOM decomposition because the large amount of accumulated residues of previous crops in NT more than compensated for the slower residue decomposition rate of surface residues in these long-term differentiated tillage systems. The water content of the surface crop residues was found to be key in determing the magnitude of the difference in decomposition rate between the incorporated residues in CT and the surface residues in NT.
Document type :
Theses
Complete list of metadatas

Cited literature [178 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00011985
Contributor : Bernard Nicolardot <>
Submitted on : Monday, March 20, 2006 - 9:14:19 AM
Last modification on : Friday, October 23, 2020 - 4:33:42 PM
Long-term archiving on: : Saturday, April 3, 2010 - 8:47:35 PM

Identifiers

  • HAL Id : tel-00011985, version 1

Collections

Citation

Katrien Oorts. Effet de travail du sol sur les stocks et flux de C et N dans un sol limoneux de grandes cultures du bassin Parisien. Sciences de la Terre. Institut national agronomique paris-grignon - INA P-G; Université Catholique de Louvain, 2006. Français. ⟨tel-00011985⟩

Share

Metrics

Record views

446

Files downloads

3362