An in situ approach to detect tree root ecology: linking ground-penetrating radar imaging to isotope-derived water acquisition zones



Marney E. Isaac, Department of Physical and Environmental Sciences, University of Toronto Scarborough, 1265 Military Trail, Toronto, Ontario, Canada M1C 1A4. Tel: 1-416-287-7276; Fax: 1-416-287-7279; E-mail:


Tree root distribution and activity are determinants of belowground competition. However, studying root response to environmental and management conditions remains logistically challenging. Methodologically, nondestructive in situ tree root ecology analysis has lagged. In this study, we tested a nondestructive approach to determine tree coarse root architecture and function of a perennial tree crop, Theobroma cacao L., at two edaphically contrasting sites (sandstone and phyllite–granite derived soils) in Ghana, West Africa. We detected coarse root vertical distribution using ground-penetrating radar and root activity via soil water acquisition using isotopic matching of δ18O plant and soil signatures. Coarse roots were detected to a depth of 50 cm, however, intraspecifc coarse root vertical distribution was modified by edaphic conditions. Soil δ18O isotopic signature declined with depth, providing conditions for plant–soil δ18O isotopic matching. This pattern held only under sandstone conditions where water acquisition zones were identifiably narrow in the 10–20 cm depth but broader under phyllite–granite conditions, presumably due to resource patchiness. Detected coarse root count by depth and measured fine root density were strongly correlated as were detected coarse root count and identified water acquisition zones, thus validating root detection capability of ground-penetrating radar, but exclusively on sandstone soils. This approach was able to characterize trends between intraspecific root architecture and edaphic-dependent resource availability, however, limited by site conditions. This study successfully demonstrates a new approach for in situ root studies that moves beyond invasive point sampling to nondestructive detection of root architecture and function. We discuss the transfer of such an approach to answer root ecology questions in various tree-based landscapes.