98th ESA Annual Meeting (August 4 -- 9, 2013)

COS 13-4 - Shrub encroachment effect on tree C assimilation and water use in a Mediterranean oak woodland

Monday, August 5, 2013: 2:30 PM
L100D, Minneapolis Convention Center
Maria C. Caldeira1, Teresa S. David2, Xavier Lecomte3, Christiane Werner4, Jorge S. David3 and Ronald J. Ryel5, (1)Centre for Forestry Research/ School of Agriculture, Technical University of Lisbon, Lisbon, Portugal, (2)Unidade de Silvicultura e Produtos Florestais, INIAV, Lisbon, Portugal, (3)Centre of Forest Research, Technical University of Lisbon, Instituto Superior de Agronomia, Lisboa, Portugal, (4)Agroecosystem Research, University of Bayreuth, Bayreuth, Germany, (5)Wildland Resources, Utah State University, Logan, UT
Background/Question/Methods

Shrub encroachment into grass-dominated ecosystems is occurring globally, mainly in arid and semi-arid environments. Anthropogenic activities, such as changes in land use and fire regimes, and climate change (e.g. increased drought) are key determinants in the encroachment process. In Iberian Peninsula, for example, the landscape is changing due to shrub encroachment caused by changes in fire regimes and land abandonment. Simultaneously, tree mortality is increasing due to a higher frequency of extreme droughts. Savannah-type evergreen oak woodlands are characteristic of the region and occupy large areas of southern Portugal. Specifically, Quercus suber L. (cork oak) woodlands have fundamental socio-economic relevance because they provide numerous direct (e.g. cork production) and indirect ecosystem services (e.g. water provision and carbon sequestration). Cistus ladanifer is a native shrub expanding into these woodlands. To test if shrubs decrease water availability for trees and C assimilation, we measured tree leaf gas-exchange, leaf water-potentials and stem sap-flow. During 2011 we monitored three paired plots of shrub invaded plots (baseline year) and at the beginning of the second year all the shrubs in one of each paired plots were removed. Four trees per plot in one paired plot were continuously monitored with sap-flow sensors (Granier type) and periodically leaf water potential and leaf gas exchange were measured in trees of all plots.

Results/Conclusions

During the baseline year there were no significant differences in predawn water potentials of the trees in all the plots. Sap-flow measurements and photosynthesis were also not different between plots. After cutting the shrubs, the predawn leaf water potential of the trees released from shrubs were less negative than trees growing with shrubs from winter throughout summer (winter: no shrubs: -0.21 ±0.01 MPa; shrubs: -0.39 ± 0.06 MPa; summer: no shrubs: -1.47 ±0.08 MPa; shrubs: -1.7 ± 0.07 MPa). Trees released from shrubs transpired significantly more than trees growing with shrubs. Tree leaf-level C assimilation during the summer was also significantly reduced by the presence of shrubs. C. ladanifer shrubs had a significant effect on Q. suber tree transpiration and water status. Trees growing with shrubs during 2012, a dry year, had lower leaf area index in comparison with trees released from shrubs. Shrub encroachment had a significant impact on both tree water use and tree carbon assimilation. However, it remains to understand if this pattern, found during the dry 2012, would be maintained in years with high water availability.