How do Biotic and Abiotic Factors Regulate Soil Enzyme Activities at Plot and Microplot Scales Under Afforestation?

Abstract

Afforestation can greatly affect soil enzymes which are tightly associated with soil nutrient cycling. However, the spatial variations in enzyme activity (EA) in plantation forests and the underlying mechanisms are not well understood. The main goal of this study was to determine how the drivers including the biotic (for example, plant traits and microbial properties) and abiotic factors (for example, soil properties) best regulate spatial variations in EA (β-glucosidase, leucine aminopeptidase and acid phosphatase) in three selected sites [woodland, shrubland and adjacent open areas (that is, control)] in Central China. Specifically, we tried to determine: (1) whether the drivers of EA variations vary between woodland and shrubland sites at the plot scale (100 m2) and (2) whether the drivers vary depending on the presence of tree patch (tree patch microplot vs. open-interpatch microplot) at the microplot scale (1 m2) within sites. A spatially nested sampling was conducted within each site via a series of partial Mantel tests to examine the correlations between EA and the environmental factors. For each site, similar spatial distribution trends were observed for the three kinds of enzymes at the plot scale. Our results showed that the tree distribution-induced shifts in litter and root biomass, soil pH and microbial community structure primarily controlled the spatial variations in EA. Specifically, soil EA was significantly correlated with litter and root biomass in the woodland site at the plot scale and primarily dependent on soil pH and microbial properties in the shrubland site. In contrast, the drivers of EA variations at the microplot scale are inconsistent between the tree patch microplot and the interpatch microplot within sites. Additionally, soil EA was not correlated with any factor in the open area. Taken together, our results reveal a shift in drivers of EA under afforestation and a profound impact of tree patch on these drivers at the plot scale.