This study intended to elucidate the long-term effects of organic soil amendments on bacterial co-occurrence in bulk soil with and without addition of mineral fertiliser. Previous research mostly neglected the bacterial co-occurrence structure and focussed mainly on the parameters species diversity and abundance changes of species. Here we present a systematic comparison of two frequently used soil amendments, manure and straw, with regard to their impact on bacterial co-occurrence in a long-term field trial in Speyer, Germany. The approach involved 16S amplicon sequencing in combination with a bacterial network analysis, comparing the different fertiliser regimes. The results show an increase of bacterial diversity as well as an accumulation of bacteria of the order Bacillales in plots fertilised with manure compared to a control treatment. In the straw-amended plots neither an increase in diversity was found nor were indicative species detectable. Furthermore, network analysis revealed a clear impact of mineral fertiliser addition on bacterial co-occurrence structure. Most importantly, both organic amendments increased network complexity irrespective of mineral fertilisation regime. At the same time, the effects of manure and straw exhibited differences that might be explained by differences in their nutritional/chemical contents. It is concluded that bacterial interactions are a crucial parameter for the assessment of amendment effects regarding soil health and sustainability. © 2017, Springer Science+Business Media, LLC, part of Springer Nature.     «

This study intended to elucidate the long-term effects of organic soil amendments on bacterial co-occurrence in bulk soil with and without addition of mineral fertiliser. Previous research mostly neglected the bacterial co-occurrence structure and focussed mainly on the parameters species diversity and abundance changes of species. Here we present a systematic comparison of two frequently used soil amendments, manure and straw, with regard to their impact on bacterial co-occurrence in a long-ter...     »