Patterns and Driving Factors of Litter Decomposition Rates in Global Dryland Ecosystems.

Litter decomposition is essential in linking aboveground and belowground carbon, nutrient cycles, and energy flows within ecosystems. This process has been profoundly impacted by global change, particularly in drylands, which are highly susceptible to both anthropogenic and natural disturbances. However, a significant knowledge gap remains concerning the extent and drivers of litter decomposition across different dryland ecosystems, limiting our understanding of its role in ecosystem metabolism. Using the ARIDEC data collection and published literature, a global database on litter decomposition and corresponding environmental conditions in drylands was developed, comprising 2204 observations from 158 sites. Decomposition rates varied across the four dryland subregions, with the highest rates in the dry-subhumid region (3.24% month), followed by semi-arid (3.15% month), arid (2.62% month), and hyper-arid (2.35% month) regions. Notably, the dry-subhumid region exhibited the greatest variability. Anthropogenic systems, such as cropland (5.52% month) and urban ecosystems (7.88% month), demonstrated higher decomposition rates than natural systems (averaging 3.07% month). Across drylands, the decomposition rate followed an exponential function of decomposition duration ( ), influenced by litter quality, climate, and soil properties. Beyond decomposition duration, three boosted regression tree models were developed to identify the primary factors influencing early (R = 0.92), mid (R = 0.71), and late (R = 0.80) decomposition stages. In the early- and mid-stages, precipitation, atmospheric temperature, and soil moisture were critical factors, while the UV index and initial nitrogen content of litter played significant roles in the early and mid-phases, respectively. In the late phase, soil total nitrogen, soil organic carbon, and the initial C/N ratio of litter were the primary factors. Our findings reveal consistent temporal patterns in decomposition rates and the mechanisms underlying them in global dryland ecosystems. These insights can enhance the accuracy of biogeochemical models in drylands and improve predictions of their feedback to the climate system.