Articles
A RE-ANALYSIS OF HEAT PULSE THEORY ACROSS A WIDE RANGE OF SAP FLOWS
Article number
846_8
Pages
95 – 104
Language
English
Abstract
In this paper we use a two-dimensional model of heat and water flow through moist wood to investigate some practical aspects of a variety of heat-pulse methods to measure sap flow.
Our analysis includes the compensation, Tmax, heat-ratio and average gradient methods that typically employ a linear heater probe that is inserted radially into the plant stem.
The temperature rise following a brief 1-2 s pulse of heat is monitored for a short period of time (i.e. 60-300 s) using one or two temperature sensors that are located above and/or below the heater probe.
The computer model is used to simulate the effect of different sapwood properties, such as the green density and the moisture content, as well as the sap flux densities (cm3 of water per cm2 of sapwood cross-section per hour) on the measured heat-pulse velocity.
The probe locations were altered to accommodate each of the heat-pulse techniques.
Our particular interest here is to investigate the performance of each method over a wide range of sap flows.
The compensation and the T-max methods work best at flows > 5-10
cm h-1 while reverse flows, if they occur, are practically impossible to resolve.
Alternative methods such as the heat-ratio and the average gradient method work much better under low and reverse flows yet they have problems resolving even moderate flows above about 40 cm/h using typical averaging periods.
The measurement range of the average-gradient method can be extended by using an averaging period of 60 s or less, compared to the current averaging period of 180 s.
The performance of the heat-ratio method can not be improved markedly by altering either the probe spacing or the averaging periods.
In our opinion the average gradient method is currently the most suitable method to measure the full range sap flows.
Our analysis includes the compensation, Tmax, heat-ratio and average gradient methods that typically employ a linear heater probe that is inserted radially into the plant stem.
The temperature rise following a brief 1-2 s pulse of heat is monitored for a short period of time (i.e. 60-300 s) using one or two temperature sensors that are located above and/or below the heater probe.
The computer model is used to simulate the effect of different sapwood properties, such as the green density and the moisture content, as well as the sap flux densities (cm3 of water per cm2 of sapwood cross-section per hour) on the measured heat-pulse velocity.
The probe locations were altered to accommodate each of the heat-pulse techniques.
Our particular interest here is to investigate the performance of each method over a wide range of sap flows.
The compensation and the T-max methods work best at flows > 5-10
cm h-1 while reverse flows, if they occur, are practically impossible to resolve.
Alternative methods such as the heat-ratio and the average gradient method work much better under low and reverse flows yet they have problems resolving even moderate flows above about 40 cm/h using typical averaging periods.
The measurement range of the average-gradient method can be extended by using an averaging period of 60 s or less, compared to the current averaging period of 180 s.
The performance of the heat-ratio method can not be improved markedly by altering either the probe spacing or the averaging periods.
In our opinion the average gradient method is currently the most suitable method to measure the full range sap flows.
Publication
Authors
S. Green, B. Clothier, E. Perie
Keywords
heat-pulse, sap flow, calibration, numerical model
Online Articles (44)