An analysis of Granier sap flow method, its sensitivity to heat storage and a new approach to improve its time dynamics

• Published in: Agricultural and forest meteorology 2015-10, Vol.211-212, p.2-12
• Main Authors: Hölttä, Teemu, Linkosalo, Tapio, Riikonen, Anu, Sevanto, Sanna, Nikinmaa, Eero
• Format: Article
• Language: English
• Subjects: Constants; Dynamics; Flux; Flux density; Granier sensor; Mathematical models; Nusselt number; Power consumption; Sap flow; Sensors; Thermal diffusivity; Thermal dissipation; Xylem transport
• ISSN: 0168-1923; 1873-2240
• DOI: 10.1016/j.agrformet.2015.05.005

•Granier sap flow method has been suggested to have an intrinsic time lag.•A numerical model and laboratory experiments are used to demonstrate this.•The model is also used to analyze the Granier sap method in detail.••A modification to the method is presented where its time resolution is improved. Granier sap flow method is a simple and easily applicable method to monitor sap flow in trees in field conditions, and is thus in wide use. However, it has been suggested that the method is slow to capture transient changes in actual sap flux density due to heat storage and release within the stem. We show here how this may lead to biases in the estimation of the dynamics of sap flux density especially at low flow rates when thermal diffusivity is low. We also demonstrate how the traditional Granier sap flow method could be modified to improve the temporal precision of the sap flow measurement. In the new system, the temperature difference between the heated and the reference needle is kept constant by varying the heating power and the sap flux density is calculated from the power consumption. This leads to reduced changes in the heat content of stem. These modifications also make the method more robust in terms of stability of power supply and reduce power consumption during low flow conditions. The time dynamics of the Granier method and the new “steady temperature method” are simulated with a previously published numerical model of xylem heat balance and tested in a laboratory experiment with cut pieces of stem. The numerical model is also used to demonstrate that the relation between parameter K, calculated from instantaneous sensor temperature and maximum sensor temperature, and actual sap flux density is not constant for either the traditional Granier or the new modified sensor, but is dependent on the range of sapflow rates examined and on the value of thermal diffusivity. Continuous measurements of thermal diffusivity of the sapwood along with the needle temperature/power consumption could help to improve the accuracy of the sap flow measurements.