Floodplains are known to be areas of extraordinary biodiversity with a mosaic of shifting habitats with high interdependency. Nowadays, these ecosystems are subject to conservation and protection. Regarding pedology, floodplains contain a wide pattern of all steps of soil evolution, i.e. from a new sediment deposition to stable soils after several hundred years. However, less attention has been paid in this context to pedogenesis, especially the very first steps of soil structuring in the youngest and the least developed soils. In the present project, we aim to understand the processes and the mechanisms of soil structuring, in particular the role of organic matter and soil fauna.

Then, two types of questions are raised, concerning at the same time basic mechanisms and external regulation:
1- In soils, how is organic matter involved in soil structure? Internal mechanisms in relation to soil biota are needed to better understand physical and biochemical processes implied in organic matter decomposition, incorporation and its further role in soil stabilization processes.
2- Which external constraints may regulate soil structuring? Fluvial dynamics and particularly punctual floods have to be taken into consideration. A detailed knowledge of the historical context and the use of flood simulations could be useful to understand the resilience of soils consecutive to erosion and/or supply of materials. In addition, vegetation (reflecting mesoclimate), humus forms and soil types appear to be relevant.

Within the framework of a functional approach, two main hypothesis could be stated:
1- at the alpine level, pedogenesis processes are assumed to be slower than at the plain level. So, the pioneer stages would be clearly visible as intermediate steps near the headwaters at the opposite from the plains where the soil structural stability is more strongly established.
2- the vegetation and the soil diversity as well as biological activities, are supposed to differ in relation to climatic constraints ; separated effects at the alpine level could be combined at the plain level and the same tendency could be observed at a lateral distance from the river.

To assess these assumptions, we propose to study physico-chemical and biological variables :
1- along the river, from the headwaters to the plains liable to flooding.
2- perpendicular to the river, from the river itself to far away from it, until reaching the climax stage of the vegetation.

We choose to combine manipulated laboratory and field studies, the first ones giving information about key processes, the second ones allowing the understanding of the whole ecosystem and its regulation factors.

For our purpose, we need to use descriptors which are able to highlight, in a very detailed way, the first steps of soil aggregates formation. Thus, the main variables chosen in our project are, according to a gradient of size:
i) organic matter, both autochtonous and allochtonous, which influences the biotic community of the stream as well as the nutrients dynamics of the system,
ii) soil biota, that is a strong functional link between vegetation, organic matter and nutrients dynamics, humus forms and soil types. Soil fauna will be studied, in particular Collembola, Enchytreids and earthworms that are three of the main groups involved in organic matter decomposition, nutrient cycles and soil structuring. In addition, focus will be made at a smaller scale on soil microorganisms (bacteria and fungi). For an exhaustive study, research on soil biota will be carry out by the PhD student and at least two master students.
iii) Humiferous episolum (humus forms), the “energy distribution central point” of the ecosystem, with special hot spots of activities. The episolum is therefore the place for essential steps of pedogenesis combining a two-speed temporal scale: one linked to the vegetation (season or decade) and one related to the soil evolution in its wholeness (decade to century).
iv) soil types, usually weakly developed in the alluvial zone because of the age of the surface, and because of periodic disturbance by inundation, erosion, and/or deposition, and,
v) vegetation, especially riparian vegetation types according to the intra-ecosystem levels (phytocoenoses and synusiae) and taking into account temporal variations.

Chosen as the key process in our research project, the formation of the soil structure gathers and integrates all of these five themes.