Effects of earlywood and latewood on sap flux density-based transpiration estimates in conifers

Abstract

Heat-based sap flux density (SFD) methods have been widely used to estimate the water use by conifers, but complexities arise due to the heterogeneous nature of conifer sapwood with annual rings of earlywood (EW) and latewood (LW), which differ in water- and heat-conducting properties. Laboratory-based controlled flow experiments using freshly cut stem segments from 11 pine trees were undertaken to evaluate the potential impact of hydraulic architecture of conifer sapwood on tree water use estimates from the Heat Ratio Method (HRM) and Heat Field Deformation (HFD) method, by considering different scenarios regarding the hydraulic conductivity and thermal diffusivity of EW and LW. The results show that the actual water flux was systematically underestimated in Scenario 1 (assuming only EW was water-conductive but thermal diffusivity was mean of EW and LW) and Scenario 3 (assuming equal hydraulic conductivity of LW and EW but thermal diffusivity was that of only EW). However, the mean sap flux densities obtained from 11 sample trees after correction by the LW/EW ratios were pretty close to the gravimetrical flow. Assuming equal hydraulic conductivity of LW and EW and mean thermal diffusivity of EW and LW led to either overestimation or underestimation of water use by individual trees, but the mean tree-scale water use was unbiased when including all this variance in the study system. The observed heterogeneous radial SFD variability from the HFD measurements was closely linked with patterns of successive EW and LW, especially in the central parts of the sapwood where higher SFD values were generally observed. The decreasing SFD patterns towards the cambium and heartwood were partially attributed to the decrease in moisture content, tracheid diameter and the increase in wood density of EW and LW compared with the central sapwood. The results indicated that the LW/EW ratio in stems where sap flow probes have been inserted can be measured a posteriori to correct HRM-based sap flow measurements. The sap flux is recommended to be radially corrected using the SFD patterns from HFD sensors measured at the same location of the HRM measurements in the same tree.