The study of secondary successional patterns in soil using network analysis: The case of conversion from conventional to organic farming

Soil disturbances that exert strong impacts on biota and their resources are also expected to induce changes in the web of interactions among soil variables. The objective of this paper was to explore the hypothesis that during the first years after the conversion from conventional to organic farming gradual changes occur in the correlations among the nematode, microbial and nutrient flow components, resulting in increased complexity of the soil system. To test this hypothesis, we used network analysis techniques to analyze data from conventional (Conv) and organic asparagus cultivations, as well as from the adjacent hedgerows (Hedg). Organic cultivations were managed organically for 2, 3 and 5 years (Org-2, Org-3, Org-5), representing thus steps along a gradient that corresponds to the duration of organic farming. The estimated network cohesion and centrality, as well as the key player metrics, showed that the correlation network of the newest organic systems, and mostly that of Org-3, was disorganized compared with that of Conv. This trend was reversed, and a complex and more cohesive network was formed in Org-5. In addition, by employing QAP techniques and Spearman rank correlation coefficients, inter-network correlations were estimated and similarities were found between the correlation network from Org-2 and that of Org-3. The structure of the latter was correlated with that from Org-5. Our results indicated that the conversion of a conventional system to organic induces a disturbance into the soil system routine, probably launching secondary successional changes in the interactions among soil components. These changes seem to involve the gradual destruction of existing interactions leading to the disorganization of the correlation network and the reconstruction of a more complicated one after 3–5 years.