CSIRO ATMOSPHERIC RESEARCH GASLAB FLASK C13CO2 DATA (d13C) RELEASE DATE: 1 APRIL 2003 END DATE OF DATA: 31 DECEMBER 2001 -------------------------------------------------------------------------- Contacts: Colin Allison CSIRO Atmospheric Research Private Bag 1 Aspendale, Vic, Australia 3195 Telephone: (+613) 9239 4571 Fax: (+613) 9239 4444 e-mail: colin.allison@csiro.au Roger Francey CSIRO Atmospheric Research Private Bag 1 Aspendale, Vic, Australia 3195 Telephone: (+613) 9239 4615 Fax: (+613) 9239 4444 e-mail: roger.francey@csiro.au Paul Krummel CSIRO Atmospheric Research Private Bag 1 Aspendale, Vic, Australia 3195 Telephone: (+613) 9239 4568 Fax: (+613) 9239 4444 e-mail: paul.krummel@csiro.au NOTICES * This version of the data represents the highest quality we can provide at the time of release in terms of assignment onto the VPDB-CO2 scale, internal consistency and precision. However, any data version should be considered provisional only. Adjustments may be made in the future as new or improved information becomes available. * Please contact us at the above e-mail addresses if any clarification of the meaning or limitations of the data is required. If users wish to send us preprints of any publications using the data, we would be happy to check that the data are being used within their limitations. * We ask that use of this data in any paper or presentation be accompanied by acknowledgement of the source of the data (CSIRO Atmospheric Research GASLAB) and that the version of the data (as specified by release date) be explicitly stated. SAMPLING The listed data have been obtained from flask air samples returned to GASLAB for analysis. Five types of flasks have been used for sample collection: (i) Glass 0.5 litre: sealed with two stopcocks fitted with PTFE, PFA or Viton O-rings (flask identifier prefix "G050"). (ii) Glass 5.0 litre: sealed with two stopcocks fitted with PTFE O-rings ("G500"). (iii) Glass 0.8 litre: sealed with two stopcocks fitted with PTFE or PFA O-rings ("G080"). (iv) Electropolished stainless steel 1.6 litre "Sirocans": fitted with two stainless steel valves manufactured by either Nupro or Hoke ("S160"). (v) Glass 2 litre: supplied by Meteorological Service of Canada fitted with one Viton o-ring sealed stopcock that were used at the two Canadian sites, Alert and Estevan Point, only (various systematic flask identifiers). Typically, flasks were pressurised to 80 kPa above ambient pressure using an oil-free pump. The air was chemically dried using anhydrous magnesium perchlorate. All "two-tap" flasks were flushed with about 40 litres of ambient air at the sampling location prior to collecting the air sample. (See Francey et al., 1996, for more details). SAMPLE STORAGE Sample storage times vary considerably, from as little as a few days for Cape Grim or Aircraft samples, up to as much as 1 year for Antarctic sites. No significant effect on the stable carbon isotopic composition, d13C, has been detected as a consequence of the large variation in sample storage time. We have detected some effects on the stable oxygen isotopic composition, d18O, but these data are not presented here. ANALYSIS Prior to stable isotopic analysis, the concentration of CO2, N2O, CH4, CO, and H2 were measured in all samples using gas chromatography (Francey et al., 1996). The CO2 and N2O concentration were used to perform ion corrections on the stable isotope measurements as described below. CO2 Extraction CO2 was extracted from the air in the flasks using an automated cryogenic trapping system. Air flowed at approximately 6 ml per minute (depending on flask pressure) through two traps held at about -190 degrees C. The trapping time required to extract enough CO2 for analysis (approximately 10 microlitres) varied with sample pressure but was typically around 5 minutes at the above flow rate. During the trapping procedure the pressure above the second trapped increased slowly to a maximum of about 300 millibars after 10 minutes. CO2 and H2O were trapped in the first trap. After nominally 30 mls of air had been processed the flow was stopped and the CO2 transferred from the first trap into the second trap by increasing the temperature of the first trap to -90 degrees C. The second trap was then isolated and its temperature increased to -90 degrees C to determine the yield of CO2. The CO2 was then transferred into a third trap for measurement. The temperature of the third trap was maintained at 30 degrees C during measurement. The cryogenic extraction system was connected directly to the mass spectrometer and controlled by the same computer software. A correction was applied to the measurement for the presence of N2O that was co-trapped with the CO2. Stable Isotope Ratio Measurement The stable isotopic composition of the CO2 was measured by dual inlet stable isotope ratio mass spectrometry (Finnigan MAT252). The ion beam currents at m/e 44, 45 and 46 were monitored for the CO2 sample and a CO2 working reference gas and the d45 and d46 of the sample with respect to the working reference CO2 gas was recorded. The d45 and d46 were then processed as recommended by Allison et al., 1995, to give d13C and d18O values that were corrected for the 17o contribution and for the presence of N2O (see below). The d13C and d18O were then adjusted onto the VPDB-CO2 scale as described in the Calibration Section. No corrections for storage time were applied to the stable isotope measurements. The N2O correction procedure required the CO2 and N2O concentrations for each air sample. These were measured in GASLAB using gas chromatography (Francey et al., 1996). If, occasionally, CO2 or N2O concentration could not be measured on individual samples, a realistic default value was predicted for the latitude, longitude and sample collection date using growth rates and seasonal cycles determined from the existing CSIRO global flask network data using the routines described by Thoning et al., 1989. Measurement uncertainty for all samples is discussed elsewhere (Langenfelds et al., 2001). Only d13C data are presented here. CALIBRATION Calibration of the CSIRO measurements onto the VPDB-CO2 scale involves two steps. The first step links measurements of pure CO2 to the VPDB-CO2 reference m aterial NBS-19. The second step treats systematic effects in the extraction and analysis of CO2 from air samples. Pure CO2 All dual inlet stable isotope ratio mass spectrometry measurements are made relative to one of six sub-samples, stored in large-volume glass containers, of an ultra-high purity CO2 stored in a high-pressure cylinder (HC453). The HC453 CO2 has been the sole source of working reference CO2 at CSIRO since 1977. HC453 sub-samples were measured against CO2 prepared from NBS-19 carbonate in the 1980s resulting in an assignment of VPDB-CO2 values of d13C = -6.396%, and d18O = -13.176%. The link to VPDB-CO2 has been monitored by comparisons between these sub-samples (Allison and Francey, 1995) and a number of high-purity CO2 gas standards (GS-19, GS-20, OZTECH-3, OZTECH-30, OZTECH-40, NIST RMs 8562, 8563 and 8564). CO2 extracted from air The isotopic composition of CO2 in the air is reported on the VPDB-CO2 scale and the link to this scale is performed using the CSIRO CG99 scale that is a revision of the CG92 scale that has been described elsewhere (Allison and Francey, 1999). In this procedure, all analyses of air samples are accompanied by multiple, bracketing analyses (ie extraction of CO2 and stable isotope analysis) of air from a working reference, a high-pressure cylinder of air that has been analysed frequently against the high-purity CO2 described above and assigned a stable isotopic composition on the VPDB-CO2 reference scale. The difference between the measured and assigned stable isotopic compositions of CO2 in the working reference is used to correct the air sample measurements for systematic effects on extraction and analysis. Before the pressure in the working reference drops below about 700 psi, the cylinder is retired from use and retained for surveillance purposes. Further checks on the calibration of CO2 extracted from air samples onto the VPDB-CO2 scale are maintained through regular analysis of the CLASSIC suite, [Allison et al., 2001, 2002], and flask-air sharing comparisons (e.g. Masarie et al., 2001) maintained between CSIRO and various international laboratories. More detailed calibration information is available in Langenfelds et al. (2001). DATA PROCESSING Flask data were assigned flags to indicate whether they have been classified as retained or rejected. Rejection falls into three broad categories: (i) The sample was considered to be not representative of the atmosphere at the time and place of sampling due to identified or inferred sampling or analytical problems (e.g. sample contamination, poor analysis). (ii) The sample was considered to be "non-baseline" as indicated by the meteorological conditions at the time of sampling. (iii) Any remaining outliers are flagged on the basis of a 3-sigma filter (geographically fixed sites only). For completeness, results for all samples that were collected at a site are included here, regardless of whether the data are retained or rejected. Notes: * For routine "baseline" applications, any rejected data must be actively excluded from the data sets provided. * For "non-baseline" applications data flagged under categories (ii) and (iii) above may carry biogeochemical information (see DATA FORMAT section below). * Further data selection may be desirable for those data sets that cannot be screened by the 3-sigma filter (e.g. AIA; aircraft). DATA FILES Data are provided for samples collected at the following geographically fixed sites. The altitudes are metres above sea level (masl) of the ground level at each site. ALC - Alert, Canada (82 degrees 27'N, 62 degrees 31'W, 6 masl) CFA - Cape Ferguson, Australia (19 degrees 17'S, 147 degrees 03'E, 2 masl) CGA - Cape Grim, Australia (40 degrees 41'S, 144 degrees 41'E, 94 masl) EPC - Estevan Point, Canada (49 degrees 23'N, 126 degrees 32'W, 39 masl) MAA - Mawson, Australia (67 degrees 37'S, 62 degrees 52'E, 32 masl) MLU - Mauna Loa, Hawaii, USA (19 degrees 32'N, 155 degrees 35'W, 3397 masl) MQA - Macquarie Island, Australia (54 degrees 29'S, 158 degrees 58'E, 12 masl) SIS - Shetland, Scotland (60 degrees 10'N, 01 degrees 10'W, 30 masl) SPU - South Pole, Antarctica (89 degrees 59'S, 24 degrees 48'W, 2810 masl) and from the following moving platform: AIA - Aircraft (over Bass Strait and Cape Grim) For each of the above sites, a file containing a single d13C value for each sample is provided, e.g. cgac13co2.dat for Cape Grim. For the nine geographically fixed sites, monthly mean data are also provided in Trends Online (see http://cdiac.ornl.gov/trends/xxxxxxxxx). These monthly means data were calculated as the mean of daily values generated from a smooth curve fit to the data, generated using the curve-fitting routines described by Thoning et al. (1989). DATA FORMATS Files listing individual flask data are provided in the following format. Note that altitude, latitude and longitude coordinates are only included for moving platform site. Sign conventions employed for position coordinates are positive for latitudes north and longitudes east. sss yyyy mm dd hh mm xxxxxxxx c nnnn.nnn fff ii YYYY MM DD HH MM uuuuuu aaaaa ttt.tttt gggg.gggg sss three-letter site code yyyy mm dd collection date (year, month, day; Universal Time) hh mm collection time (hour, minute; Universal Time) xxxxxxxx flask identifier c sample collection method code (for the retained data: A = GASLAB flask pump unit (FPU) with Mg(ClO4)2 drying, D = metal bellows pump with Mg(ClO4)2 drying, C = manual aircraft sampling unit with Mg(ClO4)2 drying, H = automated aircraft sampling unit with Mg(ClO4)2 drying, 6 = method not directly recorded) nnnn.nnn d13C isotopic composition against VPDB-CO2 (%) fff flags: "..." indicates no flags, sample retained. Any entry other than "." in the first flag column indicates the sample is not representative of the time and place of sampling: A = no sample taken B = sample lost before analysis C = identified sampling error D = suspected sampling problem (e.g. 2 or more measured species (CO2, N2O, CH4, CO, and H2) give anomalous values) E = mixed samples (time/place not unique) N = unacceptable analysis * = no analysis data available H = species-specific manually applied rejection flag I = species-specific sample collection problem J = irretrievable sample storage effect . = not subject to any of the above flags Any entry other than "." in the second flag column indicates the sample is non-baseline, rejected only on grounds of being an outlier or is excluded from this data set because of a non-standard sampling technique. F = non-baseline meteorological conditions G = marginal-baseline meteorological conditions K = species-specific non-baseline meteorological conditions L = species-specific marginal-baseline meteorological conditions M = 3 sigma filter rejected O = non-standard sampling technique . = not subject to any of the above flags Any entry other than "." in the third flag column is non-diagnostic. If this is the only flag the datum is retained. ii analytical instrument code yyyy mm dd hh mm first analysis date and time (local time) uuuuuu Universal Analysis Number (UAN; a number that uniquely identifies each sample. CSIRO use only) aaaaa altitude (metres) ttt.tttt latitude (decimal degrees) gggg.gggg longitude (decimal degrees) Files listing the monthly mean data are provided in the following format: yyyy mm nnn.nnn yyyy year mm month nnn.nnn d13C (% VPDB-CO2) REFERENCES Allison, C. E., and R.J. Francey, 1995. High precision stable isotope measurements of atmospheric trace gases, in Reference and intercomparison materials for stable isotopes of light elements: proceedings of a consultants meeting, Vienna (IAEA-TECDOC-825). Vienna, Austria: International Atomic Energy Agency. p. 131-153. Allison, C.E., and R.J. Francey, d13C of atmospheric CO2 at Cape Grim: The CGA record, air standards and the CG92 reference scale in Baseline Atmospheric Program (Australia) 1996, edited by J.L. Gras, N. Derek, N.W. Tindale and A.L. Dick, pp. 45-56, Bureau of Meteorology and CSIRO Atmospheric Research, Melbourne, Australia, 1999. Allison, C. E., R.J. Francey, and H.A.J. Meijer, 1995. Recommendations for the reporting of stable isotope measurements of carbon and oxygen in CO2 gas, in Reference and intercomparison materials for stable isotopes of light elements: proceedings of a consultants meeting, Vienna (IAEA-TECDOC-825). Vienna, Austria: International Atomic Energy Agency. p. 155-162. Allison, C.E., R.J. Francey, and L.P. Steele, 2002. The International Atomic Energy Agency circulation of laboratory air standards for stable isotope comparisons: Aims, preparation and preliminary results. In "Isotope aided studies of atmospheric carbon dioxide and other greenhouse gases Phase II (IAEA-TECDOC-1269). Vienna, Austria: International Atomic Energy Agency, p. 5-23. Allison, C.E., R.J. Francey, J.W.C. White, B. Vaughn, M. Wahlen, A. Bollenbacher, and T. Nakazawa, 2001. What have we learnt about stable isotope measurements from the IAEA CLASSIC? 11th WMO CO2 Measurement Experts Meeting: Tokyo, 25-28 September 2001. Francey, R.J., L.P. Steele, R.L. Langenfelds, M.P. Lucarelli, C.E. Allison, D.J. Beardsmore, S.A. Coram, N. Derek, F.R. de Silva, D.M. Etheridge, P.J. Fraser, R.J. Henry, B. Turner, E.D. Welch, D.A. Spencer, and L.N. Cooper, 1996. Global Atmospheric Sampling Laboratory (GASLAB): supporting and extending the Cape Grim trace gas programs. Baseline Atmospheric Program (Australia) 1993, edited by R.J. Francey, A.L. Dick and N. Derek, pp 8 - 29, Bureau of Meteorology and CSIRO Division of Atmospheric Research, Melbourne, Australia. Langenfelds, R.L., L.P. Steele, C.E. Allison, and R.J. Francey, 2001. CSIRO GASLAB Calibration, 2001. Internal Report, CSIRO Atmospheric Research, Aspendale, Australia. Masarie, K.A., R.L. Langenfelds, C.E. Allison, T.J. Conway, E.J. Dlugokencky, R.J. Francey, P.C. Novelli, L.P. Steele, P.P. Tans, B. Vaughn, and J.W.C. White, 2001. NOAA/CSIRO Flask Air Intercomparison Experiment: A strategy for directly assessing consistency among atmospheric measurements made by independent laboratories, J. Geophys. Res. 106, 20445-20464. Thoning, K.W., P.P. Tans, and W.D. Komhyr, 1989. Atmospheric carbon dioxide at Mauna Loa Observatory, 2, Analysis of the NOAA/GMCC data, 1974 - 1985, J. Geophys. Res. 94, 8549-8565. SELECTED BIBLIOGRAPHY Allison, C.E., R.J. Francey, and P.J. Rayner, 2001. Stable isotopes of atmospheric carbon dioxide from the CSIRO global flask sampling network, 6th International CO2 Conference: extended abstracts, Sendai, Japan, 9-11, 2001. Allison, C.E., R.J. Francey, R.L. Langenfelds, and E.D. Welch, Comparison of high precision Cape Grim CO2 isotope measurements using two mass spectrometers, in Baseline Atmospheric Program (Australia) 1991, edited by A.L. Dick and J.L. Gras, pp 10-19, Bureau of Meteorology and CSIRO, Division of Atmospheric Research, Melbourne, 1994. Francey, R.J. and H.S. Goodman, Systematic error in, and selection of, in situ d13C, in Baseline Atmospheric Program (Australia) 1983-1984, edited by R.J. Francey and B.W. Forgan, pp 27-36, Bureau of Meteorology and CSIRO, Division of Atmospheric Research, Melbourne, Australia, 1985. Francey, R.J. and H.S. Goodman, The DAR stable isotope reference scale for CO2, in Baseline Atmospheric Program (Australia) 1986, edited by B.W. Forgan and P.J. Fraser, pp 40-46, Bureau of Meteorology and CSIRO, Division of Atmospheric Research, Melbourne, Australia, 1988. Francey, R.J., L.P. Steele, R.L. Langenfelds, C.E. Allison, L.N. Cooper. B.L. Dunse, B.G. Bell, T.D. Murray, H.S. Tait, L. Thompson, and K.A. Masarie, 1998. Atmospheric carbon dioxide and its isotopes, methane, carbon monoxide, nitrous oxide and hydrogen from Shetland. Atmos. Environ. 32, 3331-3338. Langenfelds, R.L., R.J. Francey, B.C. Pak, L.P. Steele, J. Lloyd, C.M. Trudinger, and C.E. Allison, 2002. Interannual growth rate variations of atmospheric CO2 and its d13C, H2, CH4 and CO between 1992 and 1999 linked to biomass burning, Global Biogeochemical Cycles, 16(3), 1048.