We determined the amount of information needed to estimate watershed-scale transpiration in a Japanese cedar (Cryptomeria japonica D. Don) forest from sap flow measurements of individual trees. Measurements of tree biometrics (diameter at breast height (DBH) and tree sapwood area (AS_tree)), and tree-to-tree and radial variations in xylem sap flux density (Fd) were made in two stand plots, an upper slope plot (UP) and a lower slope plot (LP), during a growing season characterized by significant variations in environmental factors. We then investigated how mean stand sap flux density (JS) and a tree stem allometric relationship (DBH-AS_tree) varied between the stands. Appropriate sample sizes for estimating representative JS values were determined. Both a unique and a general function allowed description of the allometric relationship along the slope, but the data for its formulation was required for both the UP and LP. Values of JS in the UP and LP were similar during the study period despite differences in tree density and size between the plots, implying that JS measured in a partial stand in a watershed is a reasonable estimate of JS in other stands in the watershed, and that stand sapwood area calculated from AS_tree is a strong determinant of water use in a forest watershed. To estimate JS in both the UP and LP, it was necessary to sample at least 10 trees in each plot.

No abstract

The amount of water stored in the stem introduces uncertainty when estimating diurnal whole-tree transpiration (E T ) and canopy stomatal conductance (G C ) using sap flow measured at the base of the stem (Q). Thus, to examine how E T can be calculated from Q, we obtained E T using sap flow and stem water content measurements and a whole-tree water balance equation, and compared it with Q. In this study, we measured sap flows in 33-year-old individual trees of Cryptomeria japonica D. Don and Chamaecyparis obtusa Endl. using constant-heat sap flow probes. Sap flows were measured at several depths at the base of the stem, and at the upper trunk as a surrogate of E T . Stem water contents were measured at three vertical positions on the trunk using amplitude-domain reflectometry (ADR) sensors. We also measured sapwood volumes of the study trees. Using simultaneous sap flow and stem water content measurements along the tree stem, we confirmed that stem water storage has impacts on the transpiration stream. These include sap flow lags along the tree heights and an enhanced peak of transpiration from stem sap flow. These results enabled us to calculate the correct E T by multiplying Q by 1Ð18 and shifting its time series forward by 30 min. The E T value was then used to calculate G C for both tree species. The factor of 1Ð18 is based on the fact that at noon, the value of E T was higher than that of Q, due to the prolonged Q during the evening. Establishing the time lag was relatively simple and was determined by comparing Q and vapor pressure deficit. The multiplier is more challenging to ascertain due to the difficulty in obtaining E T correctly.