Pan‐European <i>δ</i>¹³C values of air and organic matter from forest ecosystems

Abstract We present carbon stable isotope, δ 13 C, results from air and organic matter samples collected during 98 individual field campaigns across a network of Carboeuroflux forest sites in 2001 (14 sites) and 2002 (16 sites). Using these data, we tested the hypothesis that δ 13 C values derived from large‐scale atmospheric measurements and models, which are routinely used to partition carbon fluxes between land and ocean, and potentially between respiration and photosynthesis on land, are consistent with directly measured ecosystem‐scale δ 13 C values. In this framework, we also tested the potential of δ 13 C in canopy air and plant organic matter to record regional‐scale ecophysiological patterns. Our network estimates for the mean δ 13 C of ecosystem respired CO 2 and the related ‘discrimination’ of ecosystem respiration, δ er and Δ er , respectively, were −25.6±1.9‰ and 17.8 ±2.0‰ in 2001 and −26.6±1.5‰ and 19.0±1.6‰ in 2002. The results were in close agreement with δ 13 C values derived from regional‐scale atmospheric measurement programs for 2001, but less so in 2002, which had an unusual precipitation pattern. This suggests that regional‐scale atmospheric sampling programs generally capture ecosystem δ 13 C signals over Europe, but may be limited in capturing some of the interannual variations. In 2001, but less so in 2002, there were discernable longitudinal and seasonal trends in δ er . From west to east, across the network, there was a general enrichment in 13 C (∼3‰ and ∼1‰ for the 2 years, respectively) consistent with increasing Gorczynski continentality index for warmer and drier conditions. In 2001 only, seasonal 13 C enrichment between July and September, followed by depletion in November (from about −26.0‰ to −24.5‰ to −30.0‰), was also observed. In 2001, July and August δ er values across the network were significantly related to average daytime vapor pressure deficit (VPD), relative humidity (RH), and, to a lesser degree, air temperature ( T a ), but not significantly with monthly average precipitation ( P m ). In contrast, in 2002 (a much wetter peak season), δ er was significantly related with T a , but not significantly with VPD and RH. The important role of plant physiological processes on δ er in 2001 was emphasized by a relatively rapid turnover (between 1 and 6 days) of assimilated carbon inferred from time‐lag analyses of δ er vs. meteorological parameters. However, this was not evident in 2002. These analyses also noted corresponding diurnal cycles of δ er and meteorological parameters in 2001, indicating a rapid transmission of daytime meteorology, via physiological responses, to the δ er signal during this season. Organic matter δ 13 C results showed progressive 13 C enrichment from leaves, through stems and roots to soil organic matter, which may be explained by 13 C fractionation during respiration. This enrichment was species dependent and was prominent in angiosperms but not in gymnosperms. δ 13 C values of organic matter of any of the plant components did not well represent short‐term δ er values during the seasonal cycle, and could not be used to partition ecosystem respiration into autotrophic and heterotrophic components.