Humus-nitrogen cycle relationships within the humic epipedon in pure beech high forest

The aims of the present thesis was to improve our knowledge on (i) the relationships between humus morphology and mineral nitrogen (N) production, (ii) N cycle and its regulation within the different soil layers and (iii) the environmental factors responsible for the development of humus forms and controlling soil N pathways along a chronosequence of 130 years of pure beech. The nitrogen cycle. Potential net ammonification increases with stand age in the organic horizons, whereas both potential and net in situ nitrification decrease in OL and A horizons. Potential net nitrification takes mainly place in the OF and OH horizons with ammonification always higher in the organic horizons. The fungal N transformations clearly dominate in the OL horizon while bacterial processes appear to be mainly localized in the A horizon. In general, it appears that the intensity of the first steps of the cycle (i.e. N input and ammonification) are favored during the maturation of pure stands of beech while the latter process of the cycle (i.e. nitrification and denitrification) decrease along the chronosequence. Leaching of nitrate did not differ along the chronosequence, while the uptake of mineral N by roots (especially ammonium), and the leaching of ammonium significantly increased. We also observed several significant correlations between morphological variables and net nitrification or nitrate content within the organic horizons. Therefore, several morphological variables, such as the thickness of OF, density of earthworm casts, the structure of the A horizon or the percentage of bleached leaves in OLv, were found to be good predictors of in situ mineral N production. Furthermore, the morphological variables specific to the horizon OLv were also depicted as robust indicators of ex situ mineral N production. This work demonstrates that the shift mull-moder occurring along the chronosequence means an increase of ammonium production but a decrease in nitrification. The driving factors. We did not find significant differences in litter production along the chronosequence, in opposite, the rate of litter decomposition decreased during the aggradation phase. Furthermore, litter decay rate was strongly correlated with the thickness of OF and OH layers. Thus, the decrease in litter decay rate appears to be responsible for the mull-moder shift observed during the chronosequence, while litter production would rather play a secondary role but may contribute to the second shift observed from hemimoder to dysmoder humus forms. The decrease in litter decomposition rate is partly explained by changes in both the structural and functional profiles of soil microbial communities. At the structural level, the fungal biomass in OL decreased from young to old stands. However, in OF and OH layers, the fungal to bacterial biomass ratio increased. Functional diversity of microbial community in organic horizons is higher in the oldest stands. Parallel to these changes, similar modifications were observed in litter quality. The results highlighted two major shifts. The first after 15 years corresponds to a decrease in Mg, hemicellulose, cellulose, lignin and N, and an increase in Mn content, lignin, C/N and lignin/N. The second after 95 years corresponds to a decrease in lignin, cations and N contents, lignin N, and an increase in cellulose N and hemicellulose N. An experimental approach allowed us to test the effects of beech litter (supply and quality), and roots (mycorhizal or not) on N cycling and soil microbial communities. Litter, regardless of initial quality, inhibits autotrophic nitrification and promotes fungal community. The roots promote ammonification potential while mycorhizal roots inhibit autotrophic nitrification. Therefore we hypothesized that litter quality may drive the soil microbial assemblages both at the functional and the structural level.