Effet de la longueur de la portée des sondes de dissipation thermique de type Granier sur des mesures de densité de flux de sève

Abstract

• Granier-type thermal dissipation sensors measure sap flux density (u ) by using the temperature difference between the heater and the reference probe. To detect u correctly, heat must not be transferred to the reference probe by thermal conduction. The distance across which heat can be transferred by conduction is important for the span length of a sensor and spacing of a number of sensors. • To validate span lengths and spacing of sensors, we used numerical simulations to calculate the potential distance across which heat can be transferred by conduction. We compared measurements with an original and a modified sensor for a Japanese red pine (Pinus densiflora ) from December 2004 to May 2005. The span length of the original and the modified sensor is 15 and 4 cm, respectively. • Numerical simulations showed that span length and spacing of Granier sensors should be more than 10 cm for trees in which u ceases for a few hours before the predawn period. The modified sensor underestimated u by 18–46% in winter (December–March) because its reference temperature was increased by heat transferred by conduction. The modified sensor measured u correctly in warm seasons, and only underestimated the annual amount of transpiration by 6%.