Submitted to: Boundary Layer Meteorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/4/2006
Publication Date: 12/23/2006
Citation: Griffis, T.J., Zhang, J., Baker, J.M., Kljun, N., Billmark, K. 2006. Determining carbon isotope signatures from micrometeorological measurements: implications for studying biosphere-atmosphere exchange processes. Boundary Layer Meteorology. 127:295-316. Interpretive Summary: Carbon (C) has two stable isotopes, one with a molecular weight of 12 and the other with a molecular weight of 13. The latter is much less common, comprising about 1% of terrestrial C. Plants discriminate to differing degrees against 13C during photosynthesis. All plants are somewhat less efficient in absorbing CO2 containing the heavier isotope, but C3 plants such as soybeans, wheat, and most trees discriminate more than C4 plants such as corn. Thus, by measuring the isotope ratio of respired CO2 for a given ecosystem, delta r, it is possible to draw inferences about the nature of the carbon source. This information is important for driving models of carbon cycling, a key component in predicting global climate change. Unfortunately, delta r is difficult to measure. Historically it has been done by collecting flasks of air from the atmospheric boundary layer at multiple points in time, determining the isotope ratio of each, plotting that against the inverse of CO2 concentration and extrapolating that to zero, a graph known as the Keeling plot. Unfortunately concentration measurements have a large and variable footprint, meaning that the data used for a Keeling plot may be influenced by processes occurring outside the ecosystem of interest. We have developed a new approach, called the flux ratio method, in which we directly measure the flux, or atmospheric transport, of each form of CO2, and compute the flux ratio. At night, in the absence of photosynthesis, this represents delta r. Flux measurements have a smaller footprint than concentration measurements, so the resulting delta r is a more accurate indicator of the nature of the respired C. This should provide improved estimates of ecosystem C cycling for use in global climate models.
Technical Abstract: In recent years considerable effort has been focused on combining micrometeorological and stable isotope techniques to elucidate and study biosphere-atmosphere exchange processes. At the ecosystem scale, these methods are increasingly being used to address a number of challenging problems, including the partitioning of net carbon and water fluxes into their primary exchange components to help evaluate their response to climate variations and land use change. While much progress has been achieved over the last decade, some new issues are beginning to emerge as technological advances, such as laser spectroscopy, permit isotopic fluxes to be made more easily and continuously at remote field sites. Traditional investigations have quantified the isotopic composition of biosphere-atmosphere exchange components by using the Keeling two-member mixing model (the classic Keeling plot). An alternative method, based on the new capacity to measure isotopomer mixing ratios, is to determine isotopic composition of biosphereatmosphere exchange from the ratio of flux measurements. In this study these two methods were used to quantify the isotopic composition of ecosystem respiration (delta R) over a period of three growing seasons (2003-2005) within a heterogeneous landscape consisting of both C3 and C4 species. In general, the mixing model approach produced delta R values that were 4 to 6‰ lower (isotopically lighter) than the flux-gradient method. The difference between the methods was most pronounced during the C4 growing season when the canopy was fully developed. Three hypotheses for the observed difference were tested: I) Failure of Monin-Obukov similarity theory; II) Dual source (soil versus vegetation) differences in footprint function of delta R; III) Differences in concentration versus flux footprint functions. The analyses presented here strongly suggest that differences between flux and concentration footprint functions are the main factor influencing the inequality between the mixing model and flux-gradient approaches. A mixing model approach, which is based on the concentration footprint, can have a source area influence more than 20 fold greater than the flux footprint. These results highlight that isotopic flux partitioning is susceptible to problems arising from combining signals (concentration and fluxes) that represent different spatial scales (footprint). We recognize that this problem is likely to be most pronounced within heterogeneous terrain. Even under ideal conditions, however, the mismatch between concentration and flux footprint could have a detrimental impact on isotopic flux partitioning where very small differences in isotopic signals must be resolved.