Composition of soil biodiversity networks, functional changes in nutrient dynamics, and consequences for vegetation succession

Background/Question/Methods

The majority of current theory on plant community ecology has been based on vegetation succession at abandoned arable land. This shows how habitat filtering and competition for limiting resources structures plant community composition. Currently, there is consensus that plant community development is the result of those factors, as well as of interactions with belowground biota. Here, we address the question how soil community diversity and structure affect nitrogen and carbon cycling during secondary succession. In 2011, we visited 9 grassland sites, categorized as recent, mid-term, long-term abandoned ex-arable fields. Bacteria and fungi were identified by pyrosequencing, while archaea were identified using TRFLP. The protists, micro-fauna, nematodes, enchytraeids and earthworms were extracted and morphologically identified until high taxonomic levels, often species level. In total, around 15 000 species were identified from the soils. We created a Spearman-rank correlation matrix based on abundance data of species which we visualized in a network. In 2012, intact soil cores with comparable plant vegetation were collected from the same sampling points. Stable isotope probing of the cores was performed using dual labelled ¹⁵N ammonium nitrate (¹⁵NH₄¹⁵NO₃) and ¹³C was fed to the plants in the form of ¹³CO₂. The soil food web structure was resolved by identifying the microbes using phospholipid markers and identifying soil fauna by morphology into similar groups as for the network analysis, both combined with isotopic measurements.

Results/Conclusions

We provide evidence that the conversion of soil food web structure appears to be more important than a quantitative change in biodiversity. Moreover, we show that structural changes in the food web topology also leads to functional changes in the soil food web which can act as a driving force during land use change after human disturbance. Stable isotope analysis showed that plants in the long-term abandoned soil cores allocated less newly photosynthesized carbon to their roots and took up less nitrogen from the soil. Stable isotope analysis also showed that fungi and their consumers become more important in later successional stages, but that the fungal to bacterial ratio stayed constant over time. We can conclude that during secondary succession the system shifts in terms of function from bacterial dominated to fungal dominated. However, most changes in correlation strength of the network already occur in the early stages of secondary succession, suggesting that succession effects continue on a functional level after most interaction pathways become established in early successional stages.