Environmental control of whole-plant transpiration, canopy conductance and estimates of the decoupling coefficient for large red maple trees

There is a strong need to extend whole-tree measurements of sap flow into broad-leaved forests where characteristics of stand structure, surface roughness, leaf dimension, and aerodynamic and canopy conductance may interact to partially decouple the canopy from the atmosphere. The implications of this partial decoupling to understanding the environmental control of canopy transpiration and to the modeling of forest water use are many. Therefore, thermal dissipation probes were used over a three-month period (June through August, 1997) to quantify day-to-day and tree-to-tree variation in whole-tree sap flow (Q) for 12 red maple (Acerrubrum L.) trees growing in an upland oak forest of eastern Tennessee. Whole-tree Q was calculated as the product of measured sap velocity, sapwood area and the fraction of sapwood functional in water transport. Daily canopy transpiration (Ec) was calculated from whole-tree Q and projected crown area, whereas average daily conductance (gc) was derived by inverting the Penman–Monteith equation. Maximum Q averaged 73 kg per tree per day and varied between 45 and 160 kg per day for trees that ranged in stem diameter (DBH) from 17 to 35 cm, and from 19 to 26 m in height. Canopy transpiration peaked at 3.0 mm per day in early July and averaged 1.5 mm per day over the 3-month measurement period. Tree-to-tree variability for Ec was high. Maximum rates of Ec varied from 1.9 mm per day for the tree with the smallest projected crown area to 5.7 mm per day for one of the largest trees. Day-to-day variation in Ec was a function of daily differences in net radiation (Rn) and atmospheric humidity deficit (δe). Increases in daily Rn and δe led to linear increases in canopy transpiration and there was no indication that a plateau-style relationship existed between Ec and average daily δe. Mean daily gc ranged from 1.4 to 6.7 mm s−1, and averaged 3.4 mm s−1 across the 12 study trees. Some of the tree-to-tree variation observed for Ec and gc was related to the fact that not all trees occupied the same vertical position within the stand. Variation in estimates of the daily decoupling coefficient (0≤Ω≤1) was also considerable and for individual trees the seasonally-averaged Ω varied from 0.12 to 0.37, and averaged 0.23 for the 12 study trees. An Ω of this magnitude indicates that red maple canopies are partially decoupled from the atmosphere and suggests that significant vertical gradients of air temperature and δe from the canopy surface to the bulk air several meters above the canopy are possible. Model analysis of hourly data indicated that simulated surface temperatures in mid-July were 3.6–5.8°C higher than above-canopy reference temperatures, and δe at the canopy surface was 0.3 kPa higher than that of the bulk atmosphere. These calculations were partially supported by leaf-level measurements taken on one of the trees from a 20-m canopy-access tower. The implications of this partial decoupling to understanding and modeling the environmental control of canopy transpiration are discussed.