ISOPRENE EMISSION RATES FROM NORTHERN RED OAK USING A WHOLE-TREE CHAMBER

A chamber system was developed to measure whole-tree isoprene emissions from a 6.4 m tall northern red oak, using an RGA3 Reduction Gas Analyzer automated to collect and analyze hourly samples. Tree emissions showed a strong diurnal response, corresponding to light and temperature changes. Hourly averages of emission rates under full sun for each month were highest between 1200 and 1500. Hourly average rates in August were higher than in September, whereas temperatures and light intensities were similar, indicating a probable leaf age effect on emissions. On occasion, individual rates reached 70-90 mu g C g(-1)h(-1) with full sunlight and high temperatures. Emissions decreased greatly in October, and had stopped by mid-October. Light intensity had a large effect on emissions in August and September, accounting for approximately 70% of the variation in emissions, whereas temperature accounted for only 6-11%. Relative humidity and light intensity in the canopy were not correlated with emissions. The tree emission data were fit to the algorithm used by Guenther et al. (1993, J. geophys. Res. 98, 12,609-12,617) to predict the response of isoprene emissions to light intensity. The coefficients obtained from the fitted algorithm indicated that whole-tree emissions behaved similarly to sun leaves in response to light intensity. Whole-tree rates were lower than measurements of single-leaf rates, as expected due to shading within the tree canopy. Whole-tree rates were similar to rates obtained by branch chamber measurements and by gradient flux measurements of a forest canopy, in which shading of lower leaves occurs. The results of the whole-tree chamber measurements support a recent study that indicates that the model currently used to develop biogenic hydrocarbon emission inventories underestimates isoprene emissions.