Disentangling microbial decomposition networks : linking detritus-based soil microbial food webs to ecosystem processes

Abstract

Soils are crucial for a large number of ecosystem services and occupy an important position in driving the Earth’s biogeochemical cycles. Soils are therefore essential for e.g. agricultural food production, carbon sequestration, water purification and nutrient cycling. These soil functions are to a large extent governed by the huge biodiversity of soil life, which can be depicted in the form of a soil food web: a model that describes the feeding relationships among groups of species that live in the soil. A number of soil ecosystem services, as governed by soil life, are currently under considerable threat due to e.g. soil degradation, atmospheric nitrogen deposition and land use change. A proper understanding of the mechanisms underlying soil ecosystem functioning, in relation to global change, is important to anticipate these threats and to help ensure optimal functioning of our soils. Soil food web models have proven to be highly useful in the study of the long-term consequences of environmental change on soil communities and associated ecosystem functioning. Perhaps the most important ecosystem process driven by the soil food web is the decomposition of detritus: plant residues and soil organic matter. Via the decomposition of detritus, soil organisms determine the critical balance between sequestration and mineralization of carbon (C) and nutrients, affecting soil CO2 emissions to the atmosphere and nutrient availability for plants. Soil microbes (bacteria, fungi and protozoa) play a very important role in the decomposition of detritus by being the first consuming trophic level and by making up more than 90% of the total belowground biomass. In this way, soil microbes are the main influencers of C and nitrogen (N) dynamics in soil. However, detailed information on the microbial processing of different types of organic substrates in soil food webs is still missing. Due to the important role of soil microbial communities in C and N cycling, this information is crucial to incorporate in soil food web models in order to study the long-term consequences of global change on ecosystem functioning. This is especially important if one wants to use this information for targeted management of soil life, which is seen as a promising management tool to target optimal soil functioning in anticipation of a changing world. The main research aim of this thesis was therefore to disentangle the soil microbial food web in relation to an important type of environmental change: land use change. In chapter 2, I start with discussing how state-of-the-art empirical techniques can be used to collect trophic information that is needed to construct different types of empirically-based food webs: connectedness webs, semi-quantitative webs, energy flow webs and functional webs. I explain what types of information is needed from molecular and biogeochemical studies to create such soil food web models. I thereby give a comprehensive overview of the available empirical techniques with respect to the type of information they can provide for soil food web models. In chapter 3, I study litter-derived C flows through the soil microbial food web in six different ex-arable soils. In a 56-day incubation experiment, I compared the fate of litter-derived C flows through the soil microbial communities of recent and long-term abandoned soils. Soils were amended with 13C-labelled plant litter and microbial C flows were studied by tracing the labelling of biomarkers in the form of Phospholipid Fatty Acids Stable Isotope Probing (PLFA-SIP). PLFA-SIP revealed that soil microbial communities are less efficient in decomposing litter-derived C in long-term compared to recently abandoned soils. The reduced efficiency of litter-derived C decomposition is most likely due to a net shift of organic matter-derived C to root-derived C input in relation to time since abandonment of agricultural practices. The study further revealed a clear succession of microbial decomposers, both in time and quantity that was similar across all examined fields: fungi > G- bacteria > G+ bacteria ≥ actinomycetes > micro-fauna. This information gives a first quantitative insight in how litter-derived C flows through the detritus-based soil microbial food web. In chapter 4, I continue assessing C flows through the soil microbial community in more detail, by tracing the fate of three contrasting types of organic substrates. The same set of ex-arable soils as examined in chapter 3 were incubated for 28 days after the addition of a mixture of glycine, cellulose and vanillin. In each of the treatments one or none of these compounds was 13C-labelled, to trace the fate of a specific organic compound. Application of both PLFA-SIP and RNA-SIP analyses allowed me to 1) quantify substrate-derived C flows through the soil microbial food web and 2) assess soil microbial resource partitioning beyond the concepts of the bacterial and fungal energy channels. The analyses revealed the emergence of a specific microbial community that deals with the decomposition of recalcitrant material in long-term abandoned soils. Furthermore, the existence of soil microbial decomposer succession was further confirmed by revealing both intra-kingdom microbial decomposer successional patterns and intra-kingdom microbial resource partitioning on the taxonomic level of fungal and bacterial classes. These results further enhance the view that the understanding of soil microbial decomposition goes beyond the concepts of bacterial and fungal energy channels. In chapter 5, I assess the effects of contrasting types of organic matter inputs on microbial biomass, activity and community structure, as well as related ecosystem processes like N mineralization, microbial N immobilization, plant growth and nutrient uptake. In a pot experiment, Brussels sprouts were grown on arable soils that were mixed with 15N-labelled mineral fertilizer and a contrasting type of organic amendments. The experiment revealed that a number of ecosystem processes were directly related to soil microbial activity, while microbial N immobilization was mostly dependent on the soil microbial community structure. These outcomes support the idea that soil microbial community structure is important to take into account when assessing the effects of the soil organic inputs on soil ecosystem functioning and can be used to design nutrient management strategies for more sustainable agriculture. In chapter 6, I study the drivers of both soil microbial community structure and function on two spatial scales (landscape and local scale). It is shown that these two soil microbial community characteristics are controlled by a distinct set of drivers at local versus landscape scale. I show that soil microbial community structure is driven on the landscape level by phosphorous related variables, whereas soil microbial functioning is driven locally through vegetation patterns. It is therefore important that management strategies consider the scale-dependent action of soil microbial community drivers and take both soil microbial community function and structure into account to target the desired biogeochemical functioning of soils. Overall, this thesis gives the first high-resolution and quantitative image of detritus-based microbial food webs as affected by land use change and advances our understanding of soil food webs. Studying soil microbial food webs in a chronosequence of ex-arable fields revealed that a good understanding of soil microbial C flows, beyond the level of bacterial and fungal energy channels, is crucial to understand the effect of land-abandonment on the functioning of soil food webs. A thorough understanding of intra-kingdom variation in soil microbial C processing is therefore of vital importance to enhance our understanding of soil microbial functioning in response to global change, which is the key to success for targeted management of soil life in a changing world.