Sample size shapes metabarcoding-driven biodiversity assessments across body sizes in soil

Understanding how sample size influences biodiversity detection across taxonomic groups differing in body size is critical for designing robust and cost-efficient metabarcoding studies of soil eukaryotes. Using a soil mass gradient (0.25-32 g) combined with a universal 18S rRNA metabarcoding approach, we quantified how sample mass shapes diversity estimates across eukaryotic taxa. Diversity metrics (richness and inverse Simpson diversity; q = 0, 2) and community dispersion exhibited clear body size-dependent responses. Larger-bodied taxa (e.g., Nematoda, Collembola, Insecta) showed pronounced increases in detected diversity and reduced community dispersion with increasing soil mass, indicating that small soil samples fail to capture their full diversity. Conversely, microeukaryotic groups such as fungi and protists displayed weak or even negative relationships with increasing soil mass, implying limited improvement in detection with increased sampling effort. These findings demonstrate the interactive effect organismal body size and sampling approach on the detection of soil biodiversity. We suggest optimal soil sample sizes for different groups of soil biota, and propose that a taxon-specific analytical framework can enhance both the ecological representativeness and cost efficiency in metabarcoding-based soil biodiversity assessments and monitoring.