Cork oak physiological responses to manipulated water availability in a Mediterranean woodland

This study details the physiological responses of cork oak (Quercus suber L.) to manipulated water inputs. Treatments named as dry, ambient and wet, which received 80, 100 and 120% of the annual precipitation, respectively, were applied to a Mediterranean woodland in southern Portugal. Tree ecophysiology and growth were monitored from 2003 to 2005. The impacts of the water manipulation were primarily observed in tree transpiration, especially dur- ing summer drought. Rainfall exclusion reduced the annual stand canopy transpiration by 10% over the 2-year study period, while irrigation increased it by 11%. The accumulated tree transpiration matched precipitation in spring 2004 and 2005 at the stand level, suggesting that cork oak trees rely on precip- itation water sources during the peak of the growing season. However, during the summer droughts, by trees appeared unaffected by the extreme drought of 2005. Our study shows that cork oak rapidly and completely recovered from the extreme dry year of 2005 or from rainfall exclusion. Our results support the eco-hydrological equilibrium theory by which plant acquire complementary protective mechanisms to buffer the large variability in water availability experienced in semi-arid ecosystems. In optimizing their structural biomass increase in response to increasing drought stress, cork oak trees succeeded in restricting water losses to maintain the minimum leaf water potential above the critical threshold of xylem embolism, though with narrower hydraulic safety margins in 2005. Our findings highlight cork oak’s sensitivity to the amount and timing of late spring precipitation. This could be critical as future climate scenarios predict a reduction of spring precipitation as well as enhanced severity of droughts in the Iberian Peninsula by the end of the 21st century. In inducing water stress before the onset of summer droughts, the predicted spring precipitation decline could drive the species closer to the threshold of catastrophic xylem embolism at the peak of the drought period. groundwater was the main water source for trees. Despite the significant differences in soil water content and tree transpiration, no treatment effects could be detected in leaf water potential and leaf gas exchange, except for a single event after spring irri- gations in the very dry year 2005. These irrigations were intentionally delayed to reduce dry spell duration during the peak of tree growing season. They resulted in an acute positive physiological response of trees from the wet treatment one week after the last irrigation event leading to a 32% raise of stem diame- ter increment the following months. Our results suggest that in a semi-arid environment precipitation changes in spring (amount and timing) have a stronger impact on cork oak physiology and growth than an overall change in the total annual precipitation. The extreme drought of 2005 had a negative impact on tree growth. The annual increment of tree trunk diameter in the ambient and dry treatments was reduced, while it increased for trees from the wet treatment. Water shortage also significantly reduced leaf area. The latter dropped by 10.4% in response to the extreme drought of 2005 in trees from the ambient treatment. The reduction was less pronounced in trees of the wet treatment (−7.6%), and more pronounced in trees of the dry treatment (−14.7%). Cork oak showed high resiliency to inter-annual precipitation variability. The annual accumulated tree transpiration, the minimum midday leaf water potential and the absolute amount of groundwater used