A DNA-barcode biodiversity standard analysis method (DNA-BSAM) reveals a large variance in the effect of range of biological, chemical and physical soil management interventions at different sites, but that location is one of the most important aspects determining the nature of agricultural soil microbiology

There are significant knowledge gaps in our understanding of how to sustainably manage agricultural soils to preserve soil biodiversity. Here we evaluate and quantify the effects of agricultural management and location on soil microbiology using nine field trials (dating back up to 100 years) that have consistently applied different soil management practices at various locations in the United Kingdom using DNA barcode sequence data. Due to the lack of available molecular data for UK agricultural soil biology, we tested the basic hypothesis that various agricultural management interventions have a significant and greater effect on soil bacterial and fungal diversity than geographic location. While powerful, the analyses of soil microbial DNA sequence data to date has lacked standardisation which prevents meaningful comparisons across sites and studies. Therefore, to analyse these data and crucially compare and quantify the size of any effects on soil bacterial and fungal biodiversity between sites, we developed and employed a post-sequencing DNA-barcode biodiversity standard analysis method (DNA-BSAM). The DNA-BSAM comprises a series of standardised bioinformatic steps for processing sequences but more importantly defines a standardised set of ecological indices and statistical tests. Use of the DNA-BSAM reveals the hypothesis was not strongly supported, and this was primarily because: 1) there was a large variance in the effects of various management interventions at different sites, and 2) that location had an equivalent or greater effect size than most management interventions for most metrics. Some dispersed sites imposed the same organic amendments interventions but showed different responses, and this combined with observations of strong differences in soil microbiomes by location tentatively suggests that any effect of management may be contingent on location. If this holds more generally it would mean it could be unreliable to extrapolate the findings of individual trials that manipulate soil management interventions at one location to other locations. The widespread use of a standard approach will allow meaningful cross-comparisons between soil microbiome studies and thus a substantial evidence-base of the effects of land-use on soil microbiology to accumulate and inform soil management decisions.