Microbial diversity and activity in temperate forest and grassland ecosystems

Ecosystems currently face widespread biodiversity losses and other environmental disturbances, such as climate warming, related to increased anthropogenic activities. Within this context, scientists consider the effects of such changes on the biodiversity, and hence on the activity, of soil microorganisms. Indeed, soil microorganisms mediate a wide range of soil

processes. Currently, knowledge on soil microbial diversity is still limited, partially due to technical limitations. The advent of molecular-based analyses now allows studying the soil microbial diversity. These advances in the study of soil microbial communities have lead to a growing evidence of the critical role played by the microbial community in ecosystem

functioning. This relationship is supposed to be relevant for narrow processes, regulated by a restricted group of microorganisms, such as the nitrification process.

This PhD thesis aimed at studying ammonia oxidizing bacteria (AOB) community structure and richness as an integrated part of soil functioning. This research aimed at investigating the effect of aboveground plant diversity on ammonia oxidizing bacteria diversity and function in

forest and grassland soils with focus on the influence of (a) functional group identity of grassland plants (legumes, grasses, forbs), (b) grassland plant species richness and (c) tree species, on AOB diversity and function. Another objective of this research was to study the effect of a 3°C increase in air temperature on AOB diversity and function. The link between AOB diversity and function (potential nitrification) is also investigated.

For grassland ecosystems, a microcosm experiment was realized. An experimental platform containing 288 assembled grassland communities was established in Wilrijk (Belgium). Grassland species were grown in 12 sunlit, climate controlled chambers. Each chamber contained 24 communities of variable species richness (S) (9 S=1, 9 S=3 and 6 S=9).The grassland species belonged to three functional groups: three species of each grasses (Dactylis

glomerata L., Festuca arundinacea SCHREB., Lolium perenne L.), forbs (non-N-fixing dicots; Bellis perennis L., Rumex acetosa L., Plantagolanceolata L.), and legumes (N-fixing dicots; Trifolium repens L., Medicago sativa L., Lotus corniculatus L.). Half of these chambers were exposed to ambient temperature and the other half were exposed to (ambient +3°C) temperature. One ambient and one (ambient+3°C) chambers were destructively harvested 4, 16 and 28 months after the start of the experiment. The influence of plant functional group identity on the nitrification process and on AOB community structure and richness (AOB diversity) was assessed in soils collected from the first two destructive amplings (chapter 2). The effect of plant species richness on AOB diversity and function was

considered for soils sampled after 16 and 28 months (chapter 3). AOB function was determined by potential nitrification. AOB community structure and richness were assessed by polymerase chain reaction followed by denaturing gradient gel electrophoresis (DGGE) and sequencing of excised DGGE bands. I found that functional group identity can affect AOB community structure. In particular, the presence of legumes, both in monoculture or in mixture with forbs and grasses, lead to AOB community composition changes towards AOB clusters tolerating higher ammonium concentrations. This change in AOB community structure was only linked to increased potential nitrification under monocultures of legumes, when ammonium was supposed to be not limiting. This study revealed that physiological attributes of AOB and resource availability may be important factors in controlling the nitrification process. This research showed that the impact of plant species richness on the nitrification process could be mediated by the interactions between plants and AOB, through competition for substrate. A 3°C increase in air temperature did not affect AOB community structure, richness or function.

In forest ecosystems, we studied the effect of tree species in forest sites located in Belgian and in the Grand-Duchy of Luxembourg covered each by several deciduous or coniferous tree species (Fagus sylvatica L., Quercus petraea (Mattuschka) Lieblein, Picea abies (L.) Karst, Pseudotsuga menziesii (Mirbel) Franco). We investigated the influence of these tree

species on microbial processes (chapter 5) related to C and N cycling, particularly with emphasize on the nitrification process and on the diversity of AOB (chapter 6). The results showed that the effect of tree species on net N mineralization was likely to be mediated through their effect on soil microbial biomass, reflecting their influence on organic matter content and carbon availability. Influence of tree species on nitrification (potential and

relative) might be related to the presence of ground vegetation through its influence on soil ammonium and labile C availability. AOB community structure was more site-specific than tree specific. However, within sites, AOB community structure under broadleaved trees differed from the one under coniferous trees. The effect on tree species on AOB was likely to be driven by the influence of tree species on net N mineralization, which regulates the substrate availability for AOB. The results also demonstrated that the relationship between AOB diversity and function might be related both to AOB abundance and AOB community structure and richness.

This thesis showed no clear relationship between AOB community structure or richness and AOB function. However, we revealed that aboveground grassland plant richness, grassland plant functional groups and tree species influence AOB community structure and richness.