Fig. 2. Energy quality and root density as drivers of the depth-dependency of soil carbon dynamics.
Partial regression plots relating the depth variation in the turnover time of soil organic carbon (SOC) with those of the return-on-energy-investment parameter (a) and root density (b, n = 24 soil cores). The residuals of linear models fitting each variable with ‘soil type’ as a fixed effect were used here so that each variable was normalised for variation across soil types. The return-on-energy-investment parameter was calculated as the ratio between the energy density of SOC and its mean activation energy as determined by thermal analyses, and was used here as an indicator of SOC energy quality. Symbols with error bars show treatment means ± standard errors (n = 4 replicate soil cores). Polygons around regression lines represent 95% confidence intervals. βst are regression slope coefficients ± 95% confidence intervals standardised by range. Significant βst are in bold (two-sided F-test).