Leaf C Isotope Concentrations Are Reported As Standardized Ratio Relative to Pee Dee Belemnite

Leaf C isotope concentrations are reported as standardized ratio relative to Pee Dee Belemnite (d13CL). The within-run variability (estimated as the standard deviation of working standards) was 0.1‰ for d13C. The between-run variability was estimated by comparing the measured value of a working standard to its calibrated value and was <0.08‰ for d13C.

Across 354 samples from Fabaceae and non-Fabaceae species with the C3 photosynthetic pathway, d13CL ranged from 35.6‰ for Nasturtium officinale to -25.1‰ for Ruellia strepens. Among the 56 samples from C4 species, d13CL ranged from -16.2‰ for Leptochloa fusca to -11.4‰ for Bothriochloa ischaemum. Once accounting for average differences in d13CL among C3 and C4 species, there were no relationships between d13CL and either d15NL or [NL] with or without Fabaceae included (P > 0.4 for all). For both C3 and C4 species, d13CL declined at a rate of 1.0‰ per 100 d for both C3 and C4 species (P < 0.001) with no differences in this rate for C3 and C4 species (P = 0.97).

In contrast to the N metrics, species that were more abundant on the landscape had higher leaf d13CL (Fig. 2) suggesting greater photosynthetic water use efficiency. When abundance was averaged across all contrasts and d13CL standardized for differences between C3 and C4 species, plants with higher d13CL were more abundant (r2 = 0.06, P < 0.001). High d13CL plants were more abundant in each contrast, too: uplands and lowlands, grazed and ungrazed areas, as well as burned and unburned areas (P < 0.05). Species with high d13CL were not more or less likely to be found on uplands, grazed areas, or ungrazed areas than the opposite contrast (P > 0.07 for all comparisons)

Although at some scales there is thought to be a tradeoff between N and water use efficiencies (Cunningham et al. 1999), the lack of relationship between foliar N metrics and d13CL suggest that the high-fertility sites are not necessarily drier or wetter than low-fertility sites.

Figure S1. Relationship between d13C (‰) and (A) foliar d15N (‰) and (B) foliar N concentrations (mg g-1). No relationships significant at P < 0.05. Open circles are for C4 species and closed circes C3 species.

Figure S2 Relationship between foliar d15N and d13C and abundance averaged across 20 watersheds from 1993 – 2009.