Simulated by a 3-D Transport Model: A Sensitivity Study to Vegetation and Soil Fractionation Factors

The l so /160 ratio in atmospheric CO~ is a signal dominated by the processes by which the global land biosphere absorbs and respires COs. Potentially it makes it possible to separate the photosynthesis from the total ecosystem respiration provided that leaf COs and soil COs have different isotopic signatures, that the air-sea exchange is correctly treated and that the isotopic value of anthropogenic emissions is well characterized. In this paper we simulate the dilso of atmospheric CO2, using the 3-D tracer model TM2. We perform a specific sensitivity study of 6180 to the fractionation factors for the diffusion of CO2 out of the soils (esoil) and into/out of the leaves (eteaf). The control run is similar to Ciais et al. (1996a). The latitudinal gradient, the seasonal amplitude in the northern hemisphere and the mean long term trend appear to be fairly sensitive to these two factors. The trend is nevertheless the most sensitive but it also highly depend on the atmospheric 61so mean value and on the stratospheric enrichment of 180 in CO2 pertaining to the photolysis of 03. 01997 Elsevier Science Ltd 1 I n t r o d u c t i o n An increasing attention has been given to the isotopes of CO2 (13CO2 and more recently CLSO160), as they bring additional constrains to the global carbon cycle. The oxygen isotopic content of atmospheric CO2 is primarily driven by interactions between CO~ and the reservoirs of liquid water (see Eq. 2 in Table 1 for isotopic equilibrium between COs and H20). Franeey and Tans (1987) showed indeed that direct O atom exchange between COs and H20 vapor does not occur in the gas phase. On the other hand, when COs dissolves in water, an isotopic equilibrium for oxygen atoms takes place. Because the amount of water available is huge compared to the amount of CO.~, the l so /160 ratio in CO2 is deCorrespondence to: P. PEYLIN termined by the one of water at a given temperature. Francey and Tans (1987) and Farquhar et al. (1993) pointed out the major influence of CO2 exchanges with leaf water (photosynthesis) and soil water (soil respiration). Ciais et al. (1996a,b) further tested these parameterizations in a global 3-D transport model, and obtained reasonable agreement between simulated and observed aso/I~O ratio in atmospheric CO2. Following their work, we present here a sensitivity study of this ratio to the fractionation factors involved during the diffusion of CO2 out of the soil and into/out of the leaves. 2 The observa t ions 1so in CO2 is routinely measured at specific locations around the world from three independent air sampling networks: the NOAA-CU network of 17 sites (Trolier et al., 1996), the SCRIPPS-CIO network of l0 sites (Meijer et al., in preparation), and the CSIRO network of 5-6 sites (Francey et al., 1990). The lSO/lSO isotopic ratios are reported in the 6 scale (Eq. 1, Table 1), and 61sO values are expressed in per mil (%0). At a global scale, three major features can be retrieved from the atmospheric observations. 1. A norlh minus south difference of roughly -1.5%o: the northern hemisphere is permanently depleted in IsO, as first observed by Francey and Tans (1987). 2. A strong seasonal cycle at high northern latitudes (Keeling, 1961): there is a maximum in early summer and a minimum in early winter. The observed peak-to-peak amplitude is 1.44%o at Point Barrow (71°N, 156°W) while Cape Grim in the southern hemisphere (40°S, 140°E) has a lower peak-to-peak amplitude of 0.4%o. 3. The long time series of the SCRIPPS-CIO network (5 sites since 1977) shows no long term trend in the