Carbon isotope discrimination as a tracer of functional traits in a mediterranean macchia plant community Christiane Werner A,C and Cristina Máguas B A Experimental and Systems Ecology, University of Bielefeld, Universitätsstrasse 25, D-33615 Bielefeld, Germany. B Centro de Biologia Ambiental, Faculdade de Ciências, Universidade Lisboa, Campo Grande, P-1749-016 Lisbon, Portugal. C Corresponding author. Email: c.werner@uni-bielefeld.de Abstract. Characterising functional plant groups with simple robust parameters of structural and functional traits is an important tool in ecological research. The reliability of carbon isotope discrimination (D 13 C) as an indicator of functional types was assessed in a highly diverse mediterranean macchia comprising drought semi-deciduous malacophylls, evergreen sclerophylls and a gymnosperm. Pronounced differences in D 13 C of 4‰ occurred: semi-deciduous species (Cistus sp. L.) showed the highest and the gymnosperm (Juniperus sp. L.) the lowest D 13 C (20.3  0.5‰ and 16.2  0.18‰, respectively). Across all studied species, D 13 C was correlated with (i) phenology (length of growing period) and (ii) leaf structure (leaf mass and N per area). The correlation of D 13 C with leaf water potentials, an indicator of drought stress, was species-specific and only 6 out of 11 species exhibited a significant relationship. Thus, leaf phenology governs seasonal responsiveness of D 13 C to drought, which constrains its applicability as an indicator of water use efficiency, particularly in evergreen species with short growing periods. Principal components analysis indicated the robustness of D 13 C for the classification of functional groups yielding similar results based on multiple leaf traits or solely on D 13 C. Hence D 13 C provides an ecological tracer of different functional types, integrating structural, functional and phenological attributes. Additional keywords: drought adaptation, functional groups, phenology, specific leaf mass, stable carbon isotope, water potential, water use efficiency (WUE). Introduction Climate change may cause marked changes in ecosystem functioning. Since such changes will probably have different effects on distinct functional plant groups (such as C 3 v. C 4 species), global climate change is most likely a key driver for future species composition (e.g. Cerling et al. 1998). The utilisation of functional groups is an important method in current ecological research and there has been a large effort to characterise functional groups with simple robust parameters of structural and functional traits (e.g. Brooks et al. 1997; Lavorel et al. 1997; Sala et al. 1997; Woodward and Kelly 1997; García Novo et al. 2004). To evaluate possible impacts of functional groups on ecosystem carbon and water relations, functional groups must be characterised by physiological traits that are important for these processes (Foster and Brooks 2005). Ecological tracers that capture key factors of changes in vegetation traits should be integrative over time, represent characteristic features of a given functional plant type and be easily assessable to allow large scale sampling. Carbon isotope discrimination of leaves (D 13 C) integrates plant physiological and structural attributes. It is linked to the ratio of internal to external CO 2 partial pressure (p i : p a ), which is determined by the rate of photosynthetic carbon fixation and stomatal conductance to CO 2 (Farquhar et al. 1989a). The D 13 C of bulk leaf material represents an integrative measure of p i : p a over the leaf lifetime or at least since the time the carbon was fixed (Ehleringer 1993). Under controlled environmental conditions, a remarkably linear relationship of D 13 C and water use efficiency (WUE), i.e. the rate of carbon fixation for a relative amount of water loss, has been found for different species or genotypes (e.g. Farquhar et al. 1989b). There is a genetic variability in WUE (e.g. Lauteri et al. 2004; Monclus et al. 2005) and an associated gene has been recently identified (Masle et al. 2005). D 13 C has been widely used to assess WUE across scales from leaves to ecosystems and across different research areas (e.g. Bonal et al. 2000; Evans 2001; Ponton et al. 2006). However, D 13 C is also affected by other important functional traits and/or changes in environmental conditions during phenological development. Hence interpretation of D 13 C under natural conditions is not always straightforward due to the complex interaction between plant performance and environmental factors (e.g. Seibt et al. 2008; for reviews, see Hall et al. 1994; Brugnoli and Farquhar 2000; Máguas and Griffiths 2003). Importantly, D 13 C is related to the intrinsic WUE, i.e. the ratio of net assimilation (A) to stomatal conductance (g s ). The transpirational water loss for a given g s varies with the vapour pressure deficit (VPD) of the air, so a CSIRO PUBLISHING www.publish.csiro.au/journals/fpb Functional Plant Biology, 2010, 37, 467–477 Ó CSIRO 2010 10.1071/FP09081 1445-4408/10/050467