Soil Carbon Dynamics in a Shelterbelt in the Midwest: Sources and Spatial Variability.

Shelterbelt planting in cropland may sequester carbon (C), but sources and spatial variability have not been documented. This study was conducted to assess sources and spatial variation of soil organic C (SOC) in a 35-year-old shelterbelt, and in two adjacent cultivated fields (CF) in eastern Nebraska, USA. The 2-row shelterbelt was composed of eastern red cedar (Juniperus virginiana) and scotch pine (Pinus sylvestris), and CF have been primarily cropped to wheat, soybean, and milo. A sampling grid (7 × 17, 314.8 m²) was established across the shelterbelt and CF on Tomek soil series. Composite samples were collected and analyzed for SOC and δ¹³C natural abundance. Isotropic semivariograms were defined. Nugget to semivariance ratios of <0.25 indicated coherent strong SOC spatial dependency with prediction ability over a range of approximately 12 m after fitting Gaussian models (r² ≥ 0.61). Beyond this autocorrelation range, SOC measurements may behave independently. This assessment of spatial structure may aid in optimizing sampling designs for SOC in shelterbelt scenarios. Equivalent mass corrected SOC at 0 to 7.5 cm depth was 1.8 times higher (P< 0.001) in the shelterbelt (between the two rows of trees) than in CF (33.8 and 18.6 Mg C ha^-1, respectively). Soils exhibited a gradient of δ¹³C signatures from near constant values in CF (-17.6 ± 0.1 ‰) to much depleted in shelterbelt (-22.3 ± 0.4 ‰) capturing a gradual shift in C sources. Assuming the signal of the shelterbelt at establishment equals the current CF, source partitioning indicated that tree-derived C contributed 17.3 ± 1.6 Mg C ha^-1 (51 %) of the SOC found directly beneath the shelterbelt. Differential analysis revealed that the additional SOC accrued after 35 years of shelterbelt planting can be fully explained by tree-derived inputs. Thus, shelterbelt planting in cropland showed definitive benefits to soil C sequestration.