Soil faunal effect on plant litter decomposition in mineral soil examined by two in-situ approaches: Sequential density-size fractionation and micromorphology

Abstract

Listen

Translate article

Potentially significant effects of soil fauna on plant litter decomposition are difficult to quantify by the commonly used litter bag approach due to physical constraints of meshed bag and inevitable loss of litter fragments through the mesh. We thus tested the applicability of two bag-free approaches, physical fractionation and micromorphological analysis, to examine litter decomposition in an Andosol under a temperate humid climate. In-situ decomposition of two winter cover crop litter, C-rich wild oat (WO) and N-rich hairy vetch (HV), was compared with a conventionally farmed (CF) field which received no litter as a control. The surface soils (Ap holizons) samples taken at 0, 2, and 8 weeks after the litter incorporation were separated by density to recover a low-density fraction (LF, <1.6 g cm−3) followed by its size fractionation via wet sieving (4.0, 2.0, 1.0, 0.5, and 0.1 mm mesh sizes) based on the idea that litter fragmentation occurs in parallel with decomposition and that the size of the faunal excrements depends on that of the organism. While the LF total mass did not show clear change over the eight weeks in all treatments, its size distribution exhibited a significant decline in >4 mm size fraction and concurrent increases in 0.5–2 mm fractions primarily as macrofauna excrements in the WO field. While the loss due to its fragmentation below 0.1 mm is possible, the identified excrements (1–2 mm size fractions) produced over eight-weeks accounted for 4.6 ± 0.2% and 5.5 ± 0.4%, of the C and N present in the WO litter on a ground area basis, and 0.67 ± 0.01% and 0.43 ± 0.01% of the surface soil C and N, respectively. Micromorphology of thin soil sections confirmed the abundance of larvae (Coleoptera) excrements that were mainly comprised of WO stems. In contrast, the faunal excrements found in HV soil were largely from mesofauna (mites). The significant difference found between the litter decomposition patterns was partly attributable to the faunal activity. Specifically, the decomposition of the tough tissues of WO was promoted by macrofauna, while that of HV legume litter was associated with mesofauna without any indications of macrofaunal influence. The physical fractionation based approach we developed here allowed to estimate the production rate of macrofaunal excrements under the field condition for the first time, which can help to elucidate the impact of soil fauna on biogeochemical cycling.