Contrasting species-specific, compound-specific, seasonal, and interannual responses of foliar isoprenoid emissions to experimental drought in a mediterranean shrubland

We aimed to test the effect of soil drought conditions projected by general circulation models and eco-physiological models for the next few decades on emission rates of isoprenoids by Mediterranean shrublands. We conducted a field experiment in which we generated soil drought (ca. 12% - 20% decreased soil moisture) using automatically sliding curtains, and we measured foliar isoprenoid emissions in the three dominant species of the studied shrubland ( Erica multiflora L., Globularia alypum L., and Pinus halepensis L.) in two different annual periods. Monoterpene emissions were detected in the three studied species, but isoprene emissions were significantly detected only in E. multiflora. Main volatile terpenes emitted by the three species were alpha-pinene, beta-myrcene, limonene, and Delta(3)-carene. In general, maximum isoprenoid emission rates were found in the hottest periods, and minimum emission rates were found in winter. Isoprene emissions in E. multiflora ranged between practically 0 mu g g(-1) ( dry matter) h(-1) in winter 2005 and 57 mu g g(-1) ( dry matter) h(-1) in summer 2003. Isoprene emissions were 75% lower during the dry second annual period, 2004 - 2005, than during the first year, 2003 2004. In E. multiflora, there was an overall decrease of 19% in isoprene emissions in response to soil drought. Conversely, monoterpene emissions increased by 26.4% in drought treatment. In G. alypum, there was an overall increase of 75% in terpene emissions in response to soil drought. In P. halepensis, drought treatment increased terpene emission rates by 156%. Drought treatment affected the emissions mainly in the hottest seasons, spring and summer. There were strong compound-specific, species-specific, interannual, and seasonal changes in the emission rates and in their response to the treatments. These data might help to improve prediction algorithms, inventories, and modeling of isoprenoid emissions and of their response to climate change ( decreased isoprene emissions and increased monoterpene emissions under moderate or short-term drought and decreased emissions under severe or long-term drought), but the great variability highlights the difficulty of the task.