DISCUSSION
Monitoring soil biodiversity with eDNA metabarcoding over large geographical and taxonomic scales and sometimes in remote places has become an important practice in ecological research. Understanding how preservation conditions affect estimates of taxonomic richness and community composition is essential to ensure sound ecological conclusions. Our study shows that soil metabarcoding results are surprisingly robust to preservation conditions, as we observed limited differences in community structure and diversity estimates when samples were preserved using different strategies. However, some taxonomic groups and diversity components are more sensitive than others to certain preservation conditions. This allowed us developing guidelines for preservation depending on the aims of monitoring programs and on focal taxa.
The aim of this study was comparing realistic approaches to soil preservation against an ideal situation. Immediate extraction was our reference approach, as it avoids both DNA degradation (i.e. potential under-representation of certain taxa) and continued growth of certain taxonomic groups (i.e. potential over-representation of other taxa). Unfortunately, immediate extraction is only possible if sampling occurs nearby facilities, or when a mobile eDNA laboratory is available (e.g. Zinger, Taberlet, et al. (2019b), and logistical constraints often hampers its application in remote areas. We selected preservation conditions among the most achievable, cost-effective and frequent practices to sampling soil for eDNA studies (Dickie et al., 2018). For more details about the design of preservation conditions, see Appendix A.