Transpiration reduction and root distribution functions for a non-crop species Schefflera heptaphylla

•New transpiration reduction and root distribution functions for S. heptaphylla•Water stress tolerance decreases with an increase in LAI.•Transpiration reduction function is independent of LAIs at suctions below 50 kPa.•Peak root area index is observed at depths equal to 0.7–0.8 times of root depth.

Quantifying soil suction induced by plant transpiration is vital for engineers to analyse the performance of geotechnical infrastructure such as landfill covers. Transpiration reduction function (Trf) and root distribution function (Rdf) are the two plant properties that govern root-water uptake ability. These two functions have been quantified for various crop species, but they are sometimes used to study the behaviour of non-crop species, even though these two functions are known to be plant-specific. In this study, specific Trf and Rdf were measured for six replicates of Schefflera heptaphylla that have a range of leaf area index (LAI) from 1.0 to 3.5 in clayey sand with gravel. S. heptaphylla is a non-crop tree species that has been commonly used for ecological restoration in many subtropical regions of the world. Trf of each replicate was obtained by relating normalized transpiration rate with suction. After testing, the root system of each tree individual was imaged to determine the normalized root area index (RAI) profile (i.e., Rdf). The normalized transpiration rate for S. heptaphylla with higher LAI (3 and 3.5) is revealed to have lower tolerance of water stress as their normalized transpiration rate reduced at much lower suctions, as compared to those with lower LAI (i.e., 1–2.5). Only when suction is lower than 50 kPa, the measured Trf of S. heptaphylla is similar to some of those presumed in the literature. The measurement of Rdf shows that the maximum amount of roots for S. heptaphylla was at depths of 70–80% of the root depth, in contrast to crops species whose root distribution is typically uniform or linearly decreasing.