Monitoring moisture storage in trees using time domain reflectometry

Abstract

Laboratory and field tests were performed to examine the feasibility of using time domain reflectometry (TDR) to monitor changes in the moisture storage of the woody parts of trees. To serve as wave guides for the TDR signal, pairs of stainless steel rods (13 cm long, 0.32 cm in diameter, and 2.5 cm separation) were driven into parallel pilot holes drilled into the woody parts of trees, and a cable testing oscilloscope was used to determine the apparent dielectric constant. A laboratory calibration test was performed on two sapwood samples, so that the relation between the volumetric water content and the apparent dielectric constant of the sapwood could be determined over a range of water contents. The resulting calibration curve for these sapwood samples was significantly different than the general calibration curve used for soils, showing a smaller change in the apparent dielectric constant for a given change in the volumetric water content than is typical for soils. The calibration curve was used to estimate the average volumetric water content to a depth of 13 cm in living trees. One field experiment was conducted on an English walnut tree (Juglans regia) with a diameter of 40 cm, growing in a flood-irrigated orchard on a Hanford sandy loam near Modesto, California (U.S.A.). Rods were driven into the tree at about 50 cm above the soil surface and monitored hourly for the month of August, 1988. The moisture content determined by TDR showed a gradual decrease from 0.44 to 0.42 cm³ cm⁻³ over a two week period prior to flood irrigation, followed by a rapid rise to 0.47 cm³ cm⁻³ over a four day period after irrigation, then again a gradual decline approaching the next irrigation. A second field experiment was made on ten evergreen and deciduous trees with diameters ranging from 30 to 120 cm, growing in the foothills of the Coast Range of central California. Rods were driven into each tree at 50 to 100 cm above the soil surface and monitored on a biweekly to monthly basis for over a year. Most trees showed an early spring maximum in moisture content determined by TDR associated with leaf growth, and a late summer minimum in moisture content associated with the end of the dry season. Moisture contents ranged from 0.20 to 0.70 cm³ cm⁻³, with an annual percentage change in moisture of 15% to 70% depending on species and environmental conditions. A final field test was performed in northern New Mexico (U.S.A.) to examine the effect of trunk freezing on TDR measurements. This test confirmed that freezing conditions were recorded as a total loss of liquid water by the TDR method. These results suggest that further TDR calibration for wood, plus some understanding of the relation between tree moisture and physiological stress could be useful to several disciplines, ranging from irrigation scheduling to watershed management to forest ecology.