Background/Question/Methods
In the southeastern US early successional planted-pine forest evapotranspiration (ET) is higher compared than ET in unmanaged hardwood forests. Moreover, recent studies have shown that ET from planted pine stands is much more sensitive to climatic conditions compared to hardwood stands. Whether ET is more conservative in hardwood stands compared to planted pine stands due to management-related effects on stand structure (such as high planting density that maximizes interception and leaf area) or whether it is due to species-related biological effects (such as leaf-level transpiration), or an interaction between the two is poorly understood. We approach this issue from both the catchment- and the leaf-level, by continuously monitoring watershed ET as the difference in precipitation minus catchment runoff (P-Ro) and growing-season sap flux density (v, g H₂O m^-2 sapwood area s^-1) in planted pine and unmanaged deciduous hardwood stands.

The study sites were located in two adjacent, 13.5-ha and 12.5-ha catchments within the Coweeta Basin in the Nantahala Mountain Range of western North Carolina, USA. Watershed 17 (WS17) was planted with eastern white pine (Pinus strobus L.) at a 2 x 2 m spacing in 1956. Watershed 18 (WS18) has been unmanaged since approximately 1926 and contains a mixture of deciduous hardwood species in the overstory (Quercus prinus, Q. rubra, Betula lenta, Acer rubrum, Nyssa sylvatica, Liriodendron tulipifera, Carya glabra, C. tomentosa). We measured v in 40 trees in WS17 and 32 trees in WS18 using constant heat thermal dissipation probes. We scaled to the tree-level by developing species-specific radial profiles of v using variable length thermal dissipation probes coupled with estimates of sapwood area and leaf area.

Results/Conclusions
Annual catchment-level ET (based on P-Ro) was higher for the planted pine watershed (1392 mm) than the hardwood watershed (812 mm).. At the leaf-level, transpiration rates were highest in L. tulipifera and Carya spp., lowest in Q. spp., and intermediate in P. strobus. Leaf-level transpiration rates were also negatively correlated with tree height in the hardwood species. At the plot-level, the planted pine forest exceeded transpiration by the hardwood forest due to 1) the dominance of the Quercus spp. , and 2) the deciduous habit of the hardwood species compared to the evergreen pines. Our data show that both stand structure and species-differences contributed to the lower transpiration in the deciduous hardwood forest compared to the planted-pine forest.