Spatial and vertical variation of soil carbon at two grassland sites - Implications for measuring soil carbon stocks

Soil organic carbon (SOC) stocks are a function of the SOC concentration and the bulk density of the fine soil. Both variables are prone to changes and are influenced by abiotic and biotic processes. To determine the effect of variations of both variables on SOC stocks at two grassland sites, one with Stagnic Vertisols and one with Orthoeutric Arenosols, 440 soil cores were sampled down to 60cm depth. Bulk density and C and N concentrations were measured in 5 cin intervals (0-10cm depth) and 10cm intervals (10-60cm depth). SOC stocks at the clay rich site with Vertisols were almost twice as high (86t C ha(-1) in 0-60cm depth) as at the sandy site with Arenosols (48t C ha(-1)). Variations in the SOC stocks were determined by the thickness of the loess layer at the clay rich site. Underlying clay horizons hampered the C translocation into the subsoil which resulted in lower SOC stocks of the whole profile. Semivariograms showed spatial autocorrelations of SOC concentrations within a range of 47 and 131 m, respectively. The range of autocorrelation between samples of bulk density was much shorter (3 9 and 5 1 in). Relative variances in bulk density were 1-2 magnitudes lower than the variability of SOC concentration and decreased rapidly as soil depth increased. The difference in the variation of SOC concentration and bulk density was used to revise the sampling design for SOC stocks. An unequal number of samples, i.e. more SOC concentration samples than bulk density samples, would not necessarily decrease the power of the sampling design to detect SOC stock changes. In contrast, the optimum sampling design for these sites would consist of 33-44% bulk density samples and 56-67% SOC concentration samples. (c) 2007 Elsevier B.V. All rights reserved.