Development of pedotransfer functions for a profile cone penetrometer

In this study, we illustrate how profile cone penetrometers (PCP), can be used to measure penetration resistance rapidly to define zones similar in cone index (CI), which can be related to soil physical properties. The objective of this study was (i) to develop pedopedotransfer functions, which describe the relationship between CI and soil texture, bulk density (ρb) and water content (θ), and (ii) to evaluate the sensitivity of parameters used in these functions. The data set represented soils formed in reworked silty loess overlying glacial till and/or lacustrine sediments. The global data set were grouped into subsets in terms of similar CIs. A horizontal hierarchical cluster analysis and a vertical point inflection method were used to derive cone index layer profiles (CILP1 to CILP5). Variability of CI within CILPs was small and variability of CI among CILPs was large. Within each group, CI was regressed with soil physical properties to develop pedopedotransfer functions. These were evaluated using the coefficient of determination (R2). A sensitivity analysis was executed to evaluate the relative importance of different parameters in the regression models. For the total data set, R2 ranged from 0.35 to 0.48. Pedotransfer functions for the CILPs showed largest R2 with 0.62 for CILP1, 0.76 for CILP2, 0.70 for CILP3, 0.63 for CILP4 and 0.98 for CILP5. Depth, ρb, clay content and θ were variables with large predictive power. Textural variables had strong predictive power in the top layers, CILP1 and CILP2. In CILP4, clay contents along with ρb and θ were variables with large predictive power. In contrast, the predictive power of ρb and θ was strong in layers CILP3 and CILP5, whereas soil textural characteristics showed weak predictability of CI. Pedotransfer functions using the global data set showed large sensitivities for ρb and θ. Similar results were obtained for all other CILPs, except CILP4 where clay content showed a large sensitivity. Our results show that pedotransfer penetrometer data can improve our understanding of the spatial distribution of CI and soil physical properties at fine scale.