Evaluation of CO and SF6 as quantitative tracers for fossil fuel CO2: The Experimentalists view Ingeborg Levin1, Ute Karstens2, Ulrike Gamnitzer1, Christoph Schnherr1, Bernd Kromer1 and Samuel Hammer1 Institut fr Umweltphysik Universitt Heidelberg 1 Max-Planck-Institut fr Biogeochemie, Jena 2 February CO2 Fluxes Fossil fuel CO2 Biome-BGC NEE (EDGAR V3.2 FT, Olivier et al., 2005)
-10 0 10 20 (Churkina et al., 2003) 30 40 50 10-9 kg C m-2 s-1 Heidelberg: February 2002 (REgional MOdel) Heidelberg: June 2002 (REgional MOdel) Schauinsland: February 2002 (REgional MOdel)
Can we validate these model results ? measure total CO2 and fossil fuel CO2 measure Radiocarbon (14C) in atmospheric CO2 because fossil fuel CO2 contains no 14C and dilutes atmospheric 14CO2 Long-term 14CO2 background observations in Europe clean 14CO2 background level in the Alps Reference: Atmosphere in 1890 Long-term
1950 1960 1970 1980 1000 14C [] 800 CO2 observations in Europe 14 1990 14 CO2 Vermunt
14 CO2 Jungfraujoch 2000 14 C tree rings [Stuiver and Quay, 1981] 600 400 clean 14CO2 background level in the Alps 200 0 14C [] 200
continental reference Schauinsland monthly means Heidelberg monthly means 150 100 50 0 1985 1990 1995 2000 2005 Depletion of the 14CO2 level close to fossil fuel sources Monthly mean fossil fuel CO2 at Schauinsland station and in
Heidelberg 14C [] 200 continental reference Schauinsland monthly means Heidelberg monthly means 150 100 mean fossil fuel CO2 offset: fossil fuel CO2 [ppm] 50 0 10 Schauinsland Schauinsland : ca. 1.4 ppm
5 0 40 total offset fossil fuel offset Heidelberg 20 0 1985 1990 1995 2000 2005 Heidelberg: ca. 10 ppm
but the temporal resolution of our 14C measurements is poor: weekly means and the sensitivity is quite low: minimum uncertainty 2.5 of 370 ppm ppm 1 proxies/surrogates needed to substitute 14 CO2 observations Carbon Monoxide (CO) Sulfur Hexafluoride (SF6) Continuous trace gas observations in Heidelberg in October 2004 CO2 [ppm] 29.09.2004
450 01.10.2004 03.10.2004 CO: Mainly anthropogenic sources (fossil fuel burning) and oxidation of hydrocarbons CO2 400 CO [ppb] 350 600 CO chemical life time ca. 2 months
300 SF6 [ppt] 0 30 SF6 1.4 ppm 15 0 29.09.2004 01.10.2004 03.10.2004 SF6: Only anthropogenic sources roughly distributed according to population density
chemically inert CO2, fossil fuel CO2 and CO offset in Heidelberg CO [ppb] CO2 [ppm] 14C [] 14 120 100 80 60 40 20 0 30 20 Jungfraujoch (background)
Heidelberg 2001 2002 2003 2004 2002 2003 2004 2002 2003 2004 fossil fuel CO2 10 0
400 2001 CO offset 200 0 2001 How well can we quantitatively use CO as a fossil fuel CO2 surrogate ? ratio [ppb/ppm] CO [ppb] CO2 [ppm] Fossil fuel CO2 and CO offset and their ratio 30 20 fossil fuel CO2
Weighted mean ratio [ppb/ppm]: 10 0 400 2001 2002 2003 2004 2002 2003 2004 CO offset 200 0
2001 30 CO/CO2foss 20 10 0 2001 2002 2003 2004 Observations: 12.20.4 ratio [ppb/ppm] CO [ppb] CO2 [ppm]
Fossil fuel CO2 and CO offset and their ratio in comparison with emissions inventories 30 20 fossil fuel CO2 Weighted mean ratios [ppb/ppm]: 10 0 400 2001 2002 2003 2004 2002
2003 2004 CO offset 200 0 2001 30 CO/CO2foss 20 Observations: 12.20.4 Emissions Inventories (incl. soils, etc.) 10
0 EDGAR 2001 2002 2003 2004 Edgar: 14.7 ratio [ppb/ppm] CO [ppb] CO2 [ppm] Fossil fuel CO2 and CO offset and their ratio in comparison with emissions inventories 30 20 fossil fuel CO2
Weighted mean ratios [ppb/ppm]: 10 0 400 2001 2002 2003 2004 2002 2003 2004 CO offset 200
0 2001 30 CO/CO2foss 20 10 0 IER EDGAR 2001 2002 2003 2004 Observations: 12.20.4
Emissions Inventories (incl. soils, etc.) IER: 18.0 Edgar: 14.7 IER: Institute of Energy Economics and Rational Use of Energy, Stuttgart Conclusions (I): Fossil fuel CO2 (FFCO2) emissions in Europe contribute almost half to the continental CO2 signal Monthly mean fossil fuel CO2 levels at urban sites can be determined by high precision 14CO2 measurements to better than 10% in winter and about 30% in summer. At remote sites, the mean FFCO2 signal is small (1-5 ppm) and can be determined by 14CO2 measurements only to about 30%. Higher temporal resolution of 14CO2 measurements is required here to achieve better precision. Conclusions (II): CO is a potentially applicable surrogate tracer for FFCO2, however,
Emissions inventories of CO (and FFCO2) are yet not accurate enough to apply it quantitatively, CO has sources other than from fossil fuel burning (soils !!) which also need to be known accurately The catchment area and relative mix of emissions needs to be known accurately, this requires modelling See REMO results from Karstens et al., Poster FF-134 Thank you ! Comparison of measured and REMOmodelled CO and FFCO2 mixing ratios Comparison of measured and REMOmodelled CO/FFCO2 ratios Mean ratios [ppb/ppm]: Observations: 13.50.6 REMO & EDGAR: 12.70.1 REMO & IER: 11.00.2 See also Poster by Ute Karstens et al. No. FF-134
CO/CO2 fossil fuel emission ratios 2000 IER mean 2000 (Scholz et al., IER 2005) in mmol / mol EDGAR V3.2 FT 2000 (Olivier et al., 2005)
