Biotic and abiotic factors controlling spatial variation of mean carbon turnover time in forest soil

Data description This dataset is associated with the paper "Biotic and abiotic factors controlling spatial variation of mean carbon turnover time in forest soil". This dataset includes the soil organic carbon turnover time (τ_soc) based on radiocarbon signals at the global and regional scales. Global synthesis The analysis of global soil radiocarbon data was done using the International Soil Radiocarbon Database (ISRaD v.1.0; Lawrence et al., 2020). ISRaD is an open source data with the records of 8 biomes (i.e., forest, grassland, cropland, shrubland, savanna, tundra, permafrost, and others). Since we focus on the turnover time of SOC (τ_soc) based on radiocarbon occurring in the natural forest ecosystem, so we limited our study to data from soil depth within 200 cm in the forest ecosystem. We built a database of radiocarbon-based soil turnover time (τ_soc) of 1897 soil samples from 245 forest locations worldwide. It covers a wide geographical range (35.65 ^oS ‒ 68.8 ^oN; 159.64 °W – 173.57 °E) and a broad nature climate zone (-5.2 ^oC to 40.0 ^oC; 58.66 mm to 6900 mm) over the half a century (1958 – 2017). Where forest age is missing, we derived it from The global forest age dataset (GFAD v 1.0; Poulter et al., 2018). The GFAD database represents the distribution of forest stand age during 2000 – 2010 years. Regional analysis Soil sampling in forests across the Eastern Asian Monsoon region We sampled soils from twelve permanent forest plots in five mountains in the Eastern Asian Monsoon region (Table 1, Figure 1, and S1). Five of the twelve forest plots are members of the Smithsonian Forest Global Earth Observatory network (ForestGEO, https://forestgeo.si.edu/; Anderson-Teixeira et al., 2018; Chu et al., 2019). The other seven forest plots are members of China's National Ecosystem Research Network (CNERN, http://www.cern.ac.cn). In the Eastern Asian Monsoon region, more than half of the total annual rainfall occurs in the summer season (i.e., June, July, and August) (Tardif et al., 2020; Tian et al., 2003). We estimated the τ_soc by the radiocarbon dating analysis of up to 100 cm of soil depth in each forest plot. For each plot, we separated the whole soil increment into the surface (0 – 30 cm) and deep (30 – 100 cm) layers to test the radiocarbon signal due to the high financial cost. Details of the location, climate, and vegetation for each sampling site are provided in Table 1 and Supplementary Text 1. Nine soil cores (2.5 cm in diameter) were collected in each forest plot, and a depth of 10 cm separated each soil column from 0 to 100 cm. In total, 108 soil profiles were sampled across the 12 forest plots. The accumulated aboveground litter was collected and measured in an area of 50 cm × 50 cm in each forest plot, with three replicates adjacent to each soil profile. Fine roots (< 2 mm in diameter) were manually picked from soil samples. The litter and root samples were dried at 65 °C for 48 hours using an oven and then weighed for dry mass. The elevation and other geographic information of each forest plot were measured during the soil sampling.