Forest soils contain about 30% of terrestrial carbon (C) and so knowledge of the influence of forest management on stability of soil C pools is important for understanding the global C cycle. Here we present the changes of soil C pools in the 0–5 cm layer in two second-rotation Pinus radiata (D.Don) plantations which were subjected to three contrasting harvest residue management treatments in New Zealand. These treatments included whole-tree harvest plus forest floor removal (defined as forest floor removal hereafter), whole-tree, and stem-only harvest. Soil samples were collected 5, 10 and 15 years after tree planting at Kinleith Forest (on sandy loam soils) and 4, 12 and 20 years after tree planting at Woodhill Forest (on sandy soils). These soils were then physically divided into light (labile) and heavy (stable) pools based on density fractionation (1.70 g cm−3). At Woodhill, soil C mass in the heavy fraction was significantly greater in the whole-tree and stem-only harvest plots than the forest floor removal plots in all sampling years. At Kinleith, the soil C mass in the heavy fraction was also greater in the stem-only harvest plots than the forest floor removal plots at year 15. The larger stable soil C pools with increased residue return was supported by analyses of the chemical composition and plant biomarkers in the soil organic matter (SOM) heavy fractions using NMR and GC/MS. At Woodhill, alkyl C, cutin-, suberin- and lignin-derived C contents in the SOM heavy fraction were significantly greater in the whole-tree and stem-only harvest plots than in the forest floor removal plots in all sampling years. At Kinleith, alkyl C (year 15), cutin-derived C (year 5 and 15) and lignin-derived C (Year 5 and 10) contents in the SOM heavy fraction were significantly greater in stem-only harvest plots than in plots where the forest floor was removed. The analyses of plant C biomarkers and soil δ13C in the light and heavy fractions of SOM indicate that the increased stable soil C in the heavy fraction with increased residue return might be derived from a greater input of recalcitrant C in the residue substrate.
Read full abstract