Quantifying the effects ofwood moisture and temperature variation on time-of-flight acoustic velocity measures within standing red pine and jack pine trees

Abstract

The relationship between the dynamic modulus of elasticity (me) of xylem tissue and acoustic velocity (vd) has been established for a number of commercially-important coniferous species, including red pine (Pinus resinosa Ait.) and jack pine (Pinus banksiana Lamb.). However, vd has been shown to vary systematically with xylem temperature (tx) and moisture (mx) for some species, and hence when the calibrated me-vd relationships are used outside of the range of conditions under which they were parameterized, erroneous predictions may arise. Consequently, the objectives of this study were (1) to investigate the significance of tx and mx effects on vd measurements within standing red pine and jack pine trees, and (2), given (1), to develop correction equations for standardizing vd measurements to referenced tx and mx conditions if warranted. Analytically, based on a temporal replicated sampling design, 26 mature red pine and 36 semi-mature jack pine trees growing in managed plantations located within central Ontario, Canada (Kirkwood Forest, Great Lakes-St. Lawrence Forest Region), were continuously measured for vd, tx, and mx during the spring-to-autumn seasonal periods in 2016 (red pine) and 2017 (jack pine). A total of 6 measurement events per species occurred at approximately 4-8 week intervals in which a total of 624 red pine and 864 jack pine cardinal-specific (north, east, south, and west) breast-height acoustic velocity, xylem temperature, and xylem moisture measurements were obtained: Yielding a total of 156 red pine and 216 jack pine mean tree-based values available for analysis. Over the sampling periods, (1) mean tree xylem temperatures ranged from a minimum of 3 °C to a maximum of 31 °C (mean = 19.2 °C) in red pine and from a minimum of 0 °C to a maximum of 27 °C (mean = 16.5 °C) in jack pine, and (2) mean tree xylem moistures ranged from a minimum 31% to a maximum of 45% (mean = 38.6%) in red pine and from a minimum 25% to a maximum of 50% (mean = 38.8%) in jack pine. Graphical examination of the moisture effect on the vd and tx relationship by tree and species revealed inversely proportional, linear-like trends at lower moisture levels and directly-proportional, linear-like trends at higher moisture levels where the effect was more evident for red pine than for jack pine. In order to describe this multivariate relationship, species-specific, two-level hierarchical, mixed-effects linear models inclusive of random and fixed effects were specified and subsequently parameterized. The first-level model described the tree-specific vd-tx relationship deploying a simple linear regression specification, whereas the second-level model expressed the first-level parameter estimates as a linear function of seasonal mean tree moisture. The resultant statistically-validated, parameterized regression models, for which 64% (red pine) and 90% (jack pine) of the vd variation was explained, indicated that the vd-tx relationship varied systematically with seasonal mean moisture level in red pine but not so in jack pine. More precisely, in red pine, vd declined with increasing tx at lower moisture levels ( < 38%), but increased with increasing tx at higher moisture levels ( > 38%). Conversely, although vd declined with increasing tx in jack pine, the relationship was unaffected by changes in seasonal mean tree moisture levels. Consequently, based on the final hierarchical model specifications, correction equations for adjusting observed vd values to standardized temperature (20 °C) and moisture conditions (40% for red pine) were developed for each species. Across the range of temperatures (≈5 °C-30 °C) and mean moisture levels (≈30-45%) examined, these equations generated a mean absolute vd adjustment of approximately 0.12 km/s for red pine and 0.04 km/s for jack pine. However, based on the corresponding relative magnitude of these adjustments which account for the narrow species-specific vd sample ranges observed (0.80 and 0.85 km/s for red pine and jack pine, respectively), the standardization of vd estimates could be of operational significance when acoustic sampling during periods in which xylem temperature and moisture levels approach the extremities of their spring-to-autumn seasonal ranges. Overall, the results of this empirical-based assessment, which confirmed the presence of temperature and moisture induced variation in acoustic velocity measures within standing red pine and jack pine trees, were largely in accordance with expectation. The subsequent provision of species-specific correction functions for adjusting observed vd sample values to corresponding equivalents referenced to standardized temperature and moisture conditions could assist in mitigating the consequences of environmental variability when acoustic sampling.