Leaf age affects the seasonal pattern of photosynthetic capacity and net ecosystem exchange of carbon in a deciduous forest

Temporal trends in photosynthetic capacity are a critical factor in determining the seasonality and magnitude of ecosystem carbon fluxes. At a mixed deciduous forest in the south-eastern United States (Walker Branch Watershed, Oak Ridge, TN, USA), we independently measured seasonal trends in photosynthetic capacity (using single-leaf gas exchange techniques) and the whole-canopy carbon flux (using the eddy covariance method). Soil respiration was also measured using chambers and an eddy covariance system beneath the canopy. These independent chamber and eddy covariance measurements, along with a biophysical model (CANOAK), are used to examine how leaf age affects the seasonal pattern of carbon uptake during the growing season. When the measured seasonality in photosynthetic capacity is represented in the CANOAK simulations, there is good agreement with the eddy covariance data on the seasonal trends in carbon uptake. Removing the temporal trends in the simulations by using the early season maximum value of photosynthetic capacity over the entire growing season overestimates the annual carbon uptake by about 300 g C m^-2 year^-1 - half the total estimated annual net ecosystem exchange. Alternatively, use of the mean value of photosynthetic capacity incorrectly simulates the seasonality in carbon uptake by the forest. In addition to changes related to leaf development and senescence, photosynthetic capacity decreased in the middle and late summer, even when leaf nitrogen was essentially constant. When only these middle and late summer reductions were neglected in the model simulations, CANOAK still overestimated the carbon uptake by an amount comparable to 25% of the total annual net ecosystem exchange.