Disentangling the effects of precipitation, position, and economic and size traits on wood decomposition rates across drylands

Abstract

Woody plants are an important component of dryland vegetation. Research on wood decomposition in drylands has still been scarce compared with mesic ecosystems. Disentangling climate, position, and wood traits on decomposition rates is vital for understanding the carbon cycle in drylands. We studied the relative importance of economic traits including wood nitrogen, density, dry matter content (DMC) and bark weight ratio and the size trait of wood diameter class (<2, 2–4, 4–8, 8–13 and 13–20 mm) of plant species (four to five local species and one standard plant material) and position (aboveground and belowground) on wood litter decomposition rates along a precipitation gradient from 37 to 369 mm covering five sites. Decomposition rate of woody litter in buried condition was significantly higher than that on soil surface, which the position did not affect wood decomposition rates in extremely arid site. The decomposition rates of woody litters with first- and second-class diameter were higher than those of other-class diameter. The wood economics spectrum, and wood DMC and diameter were the most important drivers regulating wood decomposition rates. Woody litters with a larger size or in drier sites were characterized by higher wood DMC and density. The accelerating effect of soil burial increased with increasing annual precipitation or increasing woody litter diameter. Precipitation and wood diameter regulated woody litter decomposition rates directly and indirectly by modulating wood DMC or bark weight ratio. Our findings highlight a dual role of precipitation and stem diameter on aboveground and belowground woody litter decomposition and the effects of position on wood decomposition rates depend on precipitation and woody litter quality in drylands. Accurately measuring the wood decomposition rate in drylands and its sensitivity to macroclimate and position-mediated microenvironment would help improve the global carbon cycle models.