The role of microarthropods in terrestrial decomposition: a meta-analysis of 40 years of litterbag studies.

Litterbags have been utilized in soil ecology for about 50 years. They are useful because they confine organic material and thus enable the study of decomposition dynamics (mass loss and/or nutrient loss through time, colonization by soil biota) in situ, i.e. under field conditions. Researchers can easily restrict or permit access to certain size classes of soil fauna to determine their contribution to litter mass loss by choosing adequate mesh size or applying specific biocides. In particular, the mesofauna has received much attention since it comprises two very abundant and diverse microarthropod groups, the Collembola (springtails) and Acari (mites). We comprehensively searched the literature from the mid-1960s to the end of 2005 for reports on litterbag experiments investigating the role of microarthropods in terrestrial decomposition. Thirty papers reporting 101 experiments satisfied our selection criteria and were included in the database. Our meta-analysis revealed that microarthropods have a moderate but significant effect on mass loss. We discuss in detail the interactions of the microarthropod effect with study characteristics such as experimental design (e.g. number of bags, duration of experiment), type of exposed organic matter, climatic zone and land use of the study site. No publication bias was detected; however, we noticed a significant decrease in the microarthropod effect with publication year, indicating that, in the first decades of litterbag use, soil zoologists may have studied "promising" sites with a higher a priori probability of positive microarthropod effects on litter mass loss. A general weakness is that the treatments differ not only with respect to the presence or absence of microarthropods, but also with regard to mesh size (small to exclude microarthropods, wide to permit their access) or presence (to exclude microarthropods) and absence (to permit their access) of an insecticide. Consequently, the difference between the decomposition rates in the treatments is not a pure microarthropod effect but will be influenced by the additive effects of mesh size and insecticide. The relative contribution of the "true" microarthropod effect remains unknown without additional treatments controlling for the differential mesh size/insecticide effect. A meta-analysis including only those studies using different mesh size and for which the data were corrected by subtracting an estimated mesh size effect based on data from the literature yielded a significantly negative microarthropod effect on litter decomposition. These results cast doubt on the widely accepted hypothesis that microarthropods generally exert a positive effect on litter mass loss. We conclude that after 40 years of litterbag studies our knowledge on the role of microarthropods in litter mass loss remains limited and that the inclusion of a third treatment in future studies is a promising way to retain litterbags as a meaningful tool of soil biological studies.