IMAGING TREES INTERIOR USING 3D ELECTRICAL RESISTIVITY TOMOGRAPHY

Abstract

Two-dimensional (2D) resistivity tomography has been applied in the past on tree trunks to estimate sapwood and heartwood areas, monitor wood decay and determine the health state of trees. However, 2D inversion is based on the assumption that there are no changes in wood resistivity in the direction perpendicular to tomography, which may cause significant artifacts in the resultant tomograms. 3D surveys and 3D inversion are inherently more appropriate to image tree interiors than 2D surveys and 2D inversion, since the inversion handles 3D changes in wood resistivity and hence provides less artifacts. Here we demonstrate the viability of 3D inversion of tree resistivity structure using both synthetic studies and real data. Synthetic 2D and 3D apparent resistivity tomograms were calculated for healthy and fungus-infected trees scenarios assuming 3% Gaussian noise, representative of noise levels expected in typical real data. We collected field 2D and 3D resistivity data on lodgepole pine trees around Laramie (WY). All data were processed using 2D and 3D inversion algorithms that use finite element meshes to allow the inclusion of complex tree geometry. Based on the synthetic studies, the 2D inversion produces unrealistic anomalies particularly within the sapwood area, and it overestimates the recovered resistivity value for the fungus infected region. On the other hand, the resistivity distributions within the sapwood and heartwood areas are better recovered for both synthetic scenarios using 3D inversion. Our studies demonstrate the effectiveness of the 3D resistivity tomography in recovering infected region, sapwood and heartwood areas over the 2D tomography method.