Ayumi Katayama

Vertical variations in wood CO2 efflux for live emergent trees in a Bornean tropical rainforest

Difficult access to 40-m-tall emergent trees in tropical rainforests has resulted in a lack of data related to vertical variations in wood CO2 efflux, even though significant variations in wood CO2 efflux are an important source of errors when estimating whole-tree total wood CO2 efflux. This study aimed to clarify vertical variations in wood CO2 efflux for emergent trees and to document the impact of the variations on the whole-tree estimates of stem and branch CO2 efflux. First, we measured wood CO2 efflux and factors related to tree morphology and environment for seven live emergent trees of two dipterocarp species at four to seven heights of up to ∼ 40 m for each tree using ladders and a crane. No systematic tendencies in vertical variations were observed for all the trees. Wood CO2 efflux was not affected by stem and air temperature, stem diameter, stem height or stem growth. The ratios of wood CO2 efflux at the treetop to that at breast height were larger in emergent trees with relatively smaller diameters at breast height. Second, we compared whole-tree stem CO2 efflux estimates using vertical measurements with those based on solely breast height measurements. We found similar whole-tree stem CO2 efflux estimates regardless of the patterns of vertical variations in CO2 efflux because the surface area in the canopy, where wood CO2 efflux often differed from that at breast height, was very small compared with that at low stem heights, resulting in little effect of the vertical variations on the estimate. Additionally, whole-tree branch CO2 efflux estimates using measured wood CO2 efflux in the canopy were considerably different from those measured using only breast height measurements. Uncertainties in wood CO2 efflux in the canopy did not cause any bias in stem CO2 efflux scaling, but affected branch CO2 efflux.

Sap-flow velocity reduction by soil water deficit observed in a Lithocarpus edulis forest on Kyushu Island, Japan *

This paper examines transpiration reduction due to soil water deficits observed in a Lithocarpus edulis forest on Kyushu Island, south-western Japan. Continuous sap flow measurements were performed on six trees to monitor changes in tree transpiration rates. A reduction in sap flow velocity was observed in the period between late-September and late-October 2003. Precipitation amounts for this period were limited resulting in a corresponding low soil matric potential. A clear reduction in sap flow velocity occurred when soil matric potential at a 10 cm depth was ＜－ 70 (kPa). Sap flow velocity scaled for1.0 kPa vapor pressure deficit declined by 42 ％ when soil matric potential at a 10 cm depth was ＜－ 70 (kPa) compared to  － 70 ( kPa) . In addition, intensive leaf- scale physiological measurements were performed on two days that had contrasting soil matric potentials. The intensive measurements showed lower leaf transpiration rates and stomatal conductance on the day with the lower soil matric potential. This finding suggests soil water deficits caused stomatal closure and therefore reduction in tree transpiration. Sap flow velocity and leaf water potential data showed that stem hydraulic conductance did not decline with low soil matric potential when compared to the day with higher soil matric potential.