Farmland Afforestation by Poplar Shelterbelts Increased Soil Inorganic Carbon but Showed Ambiguous Effects on Soil Organic Carbon as Revealed by Carbon Isotopic Composition: Inter-Fraction and Inter-Site Differences in Northern China

Afforestation has been considered to be the cost-effective way to sequestrate carbon (C) dioxide from the atmosphere in the soils, while concurrent responses of soil inorganic C (SIC) and soil organic C (SOC), and their turnover are still not well-defined. During the C cycle, inorganic C is enriched in heavy isotopes (13C), while organic C, due to photosynthetic fractionation, is enriched in light isotopes (12C). This leads to distinct C isotope fractionation in terrestrial ecosystems. In this study, 72 paired soils (0–20 cm) from poplar shelterbelts and adjacent farmland sites were collected in six regions (Zhaozhou, Fuyu, Dumeng, Zhaodong, Lanling, and Mingshui) of northeastern China. Five soil fractions of dissolved organic C (DOC), particulate organic matter (POM), sand and stable aggregates (S + A), silt and clay (S + C), and resistant SOC (rSOC) and bulk soils were used in C content assay and the natural δ13C determination. The results showed that, compared with SOC, poplar shelterbelts resulted in SIC accrual in the soils across all six sites; however, only half of the six sites showed SOC accrual, indicating an ambiguous effect of afforestation on SOC. The natural δ13C method could identify the SOC turnover owing to the C isotopic discrimination. The δ13C–SOC-derived turnover ratio was 23%. When SIC was included in the δ13C measurement, bulk soils and four soil fractions (S + C, S + A, rSOC, DOC) showed a 2%–10% lower turnover percentage than the δ13C–SOC-derived turnover ratios. The SIC inclusion resulted in the dependency of δ13C–TC (TC = SIC + SOC) values on SOC (negative, R2: 0.21–0.44) and SIC content (positive, R2: 0.39–0.63). By contrast, when SIC was excluded, the δ13C–SOC values were independent of them (R2 < 0.18). Redundancy ordination analysis manifested that more SOC in the soils, together with more POM and farming uses would be accompanied with the lower δ13C values. Moreover, forest characteristics (e.g., age and density) and farmland backgrounds (e.g., crop history and distance between forest and farmland) could explain differences in δ13C-related features. Our results highlighted that SIC in natural δ13C determination underestimated the C turnover ratio in general. However, SIC storage should be included in the soil C sequestration evaluation owing to a general SIC accrual pattern across regions when compared with those of SOC.