Root water uptake of field-growing plants indicated by measurements of natural-abundance deuterium

Measurements of stable-isotope ratios of water extracted from stems and, in some studies, soils are increasingly being used to study the integrated root function of field-growing plants. This study explored if additional measurements on water extracted from roots could indicate the activity of roots in different areas of the soil profile and their influence on canopy water sources, so providing advantages over more common sampling strategies. Studies were conducted on trees and shrubs located in diverse habitats: a saline, semi-arid floodplain, a subhumid mountain-range front and a cold desert. At each site, roots, soil immediately surrounding the roots, and plant stems were sampled. Roots were taken from different depths in the soil, to approximately 2 m at one site. Overall, 80% of roots sampled had H isotope ratios different from the surrounding soil. The differences up to 37, were significant (p<0.05) at two of the sites. Thus water in most of the roots sampled did not come entirely, if at all, from the surrounding soil, illustrating movement and possible mixing of water within the root system. This condition was not simply related to the availability of water surrounding the soil, which was also measured. There were also differences in root and stem H isotope ratios (up to 17) in 67% of samples, although the difference was only significant in shallow samples from the floodplain. The general similarity in stem and root ²H values indicates that most roots sampled were involved in the main supply of water to the canopy. Patterns of root function varied between the individual sites. Trees were primarily using groundwater at the floodplain and mountain front sites, as the surface soils had mean matric potentials of-1800 kPa. At the mountain front site, the surface roots were transporting groundwater to the canopy in isolation form the surrounding soil. In contrast, surface roots at the floodplain were taking up water from the surrounding soil, although this water was not a significant source in the trees' overall water supply. This activity of surface roots would not have been evident from the ²H data without the root samples. At the cold desert the roots in moist surface soil provided the main source of water for the shrubs. There too the root data indicated different water uptake patterns than otherwise would have been assumed. The root data showed that groundwater could not have been a water source, a conclusion which had been reached in a previous study. Thus measurements of stable isotope ratios in root water may be a valuable tool in assessing water uptake patterns and root function.