Soil plays a central role in the functioning of terrestrial ecosystems and is essential for the production of food for the world population. Yet, soil is a fragile resource whose health is increasingly endangered by unsustainable land-use practices, growing population and global warming. Acquiring knowledge on the processes controlling the natural formation of soils and their associated ecosystems through studies of primary successions is crucial to develop sustainable land-use strategies and predict the adaptation of soil ecosystems to global warming. Chapter #1 of this thesis discusses these issues. While most studies on ecological successions focus on relatively well-developed ecosystems, our understanding of initial development of ecosystems, soils in particular, is comparatively scarce. Glacier forefields provide ideal setting to study the earliest stages of soil development and primary successions. Glacier forefield successions are most commonly studied using a chronosequence approach in which distance from the ice front of a retreating glacier (‘time since deglaciation’) is used as a proxy from ‘terrain age’. The development of pioneer microbial communities in glacier forefields is limited by the scarcity of nutrients. Chemical weathering, which can be enhanced by microbial activity, is often considered to be the dominant mechanism controlling the initial build-up of nutrients in these environments. However, the linkages between microbial communities, nutrient contents and weathering are still poorly understood. A major goal of my thesis was to study the changes in physical, chemical, mineralogical and microbiological parameters and their linkages using a chronosequence approach to gain insights into the processes controlling the initial autogenic development of soil ecosystems in glacier forefields. Interestingly, these tight relationships between chemical weathering and microbial communities indicate that microbial weathering is a strong driver of the build-up of a labile nutrient pool in early successional stages. The chronosequence approach is based on the two core assumptions that (1) all sites of the succession were characterized by homogeneous initial conditions and (2) all sites followed the same sequence of change after the original disturbance that is glacier erosion. The work presented in this thesis shows that these assumptions about chronosequences are not always valid because abiotic initial environmental conditions and geomorphological disturbances can affect successional behavior in a temporally and spatially heterogeneous manner. The second major goal of my thesis was to investigate how allogenic factors (initial site conditions and geomorphological disturbances) can affect successional patterns in glacier forefields in a heterogeneous . In Chapters #2 and #3, I present published studies in which I investigated the physical, geochemical and microbial successions leading to soil development in glacier forefields in Svalbard and in Iceland. With increasing terrain age, my data shows a build-up of nutrient contents, a progression of chemical and physical weathering and an increase in microbial diversity and abundance. My dataset also highlights the strong correlations between nutrient cycling, weathering and microbial community structures. Altogether, these trends evidence an increasing degree of soil ecosystem development along the chronosequence. I emphasize that patterns of successional change related to time since deglaciation are also strongly influenced by geomorphological disturbances and heterogeneous initial environmental conditions. Specifically, in Chapters #2 (Longyearbreen proglacial area, Svalbard) and 3 (Fláajökull glacier forefield, Iceland) I show that areas affected by hillslope and glacio-fluvial erosion disturbances depict considerably delayed succession rates. In Chapter #4, I investigated how the supply of dry (dust) vs. wet (snow /rain) aeolian deposition has contributed to the build-up of phosphorus in the forefield of Vernagt glacier (Austrian Alps). I assessed if this P supply enhanced the ecosystem succession rates in this glacier forefield. Importantly, I also investigated the effects of seasonal variability on biogeochemical processes in this glacier forefield by monitoring the year-round variability of aeolian phosphorus, as well as soil nutrient contents. Finally, Chapter #5 is a review, where I present a comprehensive overview of how initial site conditions (substrate characteristics, microclimatic conditions and resources availability) and geomorphological disturbances (hillslope, glacio-fluvial, periglacial and aeolian processes) may affect the rate and/or trajectory in a spatially heterogeneous manner. I end this last chapter with a discussion on the changes in the relative importance of autogenic, allogenic and stochastic processes over the course of successions in glacier forefields.
