No abstract

No abstract

No abstract

No abstract

No abstract

No abstract

The effect of soil structure on hydraulic pedotransfer functions (PTFs) in tropical soils with similar mineralogy and texture has not been well documented. Structurally contrasting soils from representative locations in southeastern Nigeria were analyzed for moisture retention at 0, 6, 10, 33, 100, 300 and 1500 kPa among other properties. They were grouped by depth (topsoils or subsoils) and also by their structural degradation status into low‐ and high‐stability soils, corresponding to organic matter (OM):[silt + clay] of <7.5 and >7.5%, respectively. Soil depth and structural stability influenced the soil moisture characteristic curves. The data were fitted to three tropical point PTFs, but none of them proved appropriate for predicting moisture retention in the soils. We therefore derived new ones using multiple linear stepwise regressions before and after the dataset grouping and compared their performances by means of cross‐validation. Moisture retention in the soils (sand content, 73.2–93.8%) could not be calibrated from texture and OM concentration alone, until when bulk density, total porosity and microporosity were included among the regressors. Microporosity's role was particularly outstanding at all matric potentials but the 1500 kPa. The ensuing PTFs represent a good fit for the soil moisture retention data. The two grouping strategies resulted in lower SE of the estimates in some cases, but this did not enhance the performances of the concerned PTFs. At the 1500 kPa, however, the PTF incorporating all datasets performed better than separate PTFs for topsoils/subsoils but worse than the one for high‐stability soils. Information on soil structure can therefore benefit PTF derivation for kaolinite‐dominated, coarse‐textured tropical soils at all the matric potentials considered. Copyright © 2012 John Wiley & Sons, Ltd.

No abstract