A validation, comparison and error analysis of two heat-pulse methods for measuring sap flow in Eucalyptus marginata saplings

We validated and compared two heat-pulse methods for measuring sap flow in potted Eucalyptus marginata Donn ex. Smith (jarrah) saplings. During daylight hours and under well-watered conditions, rates of sap flow (0.1-0.5 kg h(-1)) measured by the established compensation heat-pulse method (CHPM) and the newly developed heat-ratio method (HRM) were similar to rates measured with a weighing lysimeter, and most of the time there was no significant difference (P < 0.001) between methods. The HRM accurately described sap flow at night when rates of flow were low (< 0.1 kg h(-1)) or near zero, but the CHPM was unable to measure low rates of sap flow due to its inability to distinguish heat-pulse velocities below a threshold velocity of 0.1 kg h(-1) (3-4 cm h(-1)). The greatest potential for error in the calculation of daily sap flow was associated with the misalignment of temperature sensors, the estimation of sapwood area and the method used to acquire total sap flow from point measurements of sap velocity. A direct comparison of the two heat-pulse methods (applied synchronously) revealed that the HRM had a more convincing mechanism for correcting spacing errors and was more resistant to random fluctuation in measurements than the CHPM. While we view the HRM more favourably than the CHPM in some key areas, both methods are valid and useful, within their constraints, for measuring transpiration in jarrah and other woody species.