Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

4.2.6.1 Laboratory data and the OH lifetime of greenhouse gases

Laboratory data on the rates of chemical reactions and photodissociation provide a cornerstone for the chemical models used here. Subsequent to the SAR there have been a number of updates to the recommended chemical rate databases of the International Union of Pure and Applied Chemistry (IUPAC 1997a,b, 1999) and the Jet Propulsion Laboratory (JPL) (DeMore et al., 1997; Sander et al., 2000). The CTMs in the OxComp workshop generally used the JPL-1997 database (JPL, 1997) with some updated rates similar to JPL-2000 (JPL, 2000). The most significant changes or additions to the databases include: (i) revision of the low temperature reaction rate coefficients for OH + NO2 leading to enhancement of HOx and NOx abundances in the lower stratosphere and upper troposphere; (ii) extension of the production of O(1D) from O3 photodissociation to longer wavelengths resulting in enhanced OH production in the upper troposphere; and (iii) identification of a new heterogeneous reaction involving hydrolysis of BrONO2 which serves to enhance HOx and suppress NOx in the lower stratosphere. These database improvements, along with many other smaller refinements, do not change the overall understanding of atmospheric chemical processes but do impact the modelled tropospheric OH abundances and the magnitude of calculated O3 changes by as much as 20% under certain conditions.

Reaction rate coefficients used in this chapter to calculate atmospheric lifetimes for gases destroyed by tropospheric OH are from the 1997 and 2000 NASA/JPL evaluations (DeMore et al., 1997; Sander et al., 2000) and from Orkin et al. (1999) for HFE-356mff2. These rate coefficients are sensitive to atmospheric temperature and can be ten times faster near the surface than in the upper troposphere. The global mean abundance of OH cannot be directly measured, but a weighted average of the OH sink for certain synthetic trace gases (whose budgets are well established and whose total atmospheric sinks are essentially controlled by OH) can be derived. The ratio of the atmospheric lifetimes against tropospheric OH loss for a gas is scaled to that of CH3CCl3 by the inverse ratio of their OH-reaction rate coefficients at an appropriate scaling temperature. A new analysis of the modelled global OH distribution predicts relatively greater abundances at mid-levels in the troposphere (where it is colder) and results in a new scaling temperature for the rate coefficients of 272K (Spivakovsky et al., 2000), instead of 277K (Prather and Spivakovsky, 1990; SAR). The atmospheric lifetimes reported in Table 4.1 use this approach, adopting an "OH lifetime" of 5.7 years for CH3CCl3 (Prinn et al., 1995; WMO, 1999). Stratospheric losses for all gases are taken from published values (Ko et al., 1999; WMO, 1999) or calculated as 8% of the tropospheric loss (with a minimum lifetime of 30 years). The only gases in Table 4.1 with surface losses are CH4 (a soil-sink lifetime of 160 years) and CH3CCl3 (an ocean-sink lifetime of 85 years). The lifetime for nitrogen trifluoride (NF3) is taken from Molina et al. (1995). These lifetimes agree with the recent compendium of Naik et al. (2000).

Analysis of the CH3CCl3 burden and trend (Prinn et al., 1995; Krol et al., 1998; Montzka et al., 2000) has provided a cornerstone of our empirical derivations of the OH lifetimes of most gases. Quantification of the "OH-lifetime" of CH3CCl3 has evolved over the past decade. The SAR adopted a value of 5.9 ± 0.7 years in calculating the lifetimes of the greenhouse gases. This range covered the updated analysis of Prinn et al. (1995), 5.7 years, which was used in WMO (1999) and adopted for this report. Montzka et al. (2000) extend the atmospheric record of CH3CCl3 to include the rapid decay over the last five years following cessation of emissions and derive an OH lifetime of 6.3 years. The new information on the CH3CCl3 lifetime by Montzka et al. (2000) has not been incorporated into this report, but it falls within the ±15% uncertainty for these lifetimes. If the new value of 6.3 years were adopted, then the lifetime of CH4 would increase to 9.2 yr, and all lifetimes, perturbation lifetimes, and GWPs for gases controlled by tropospheric OH would be about 10% greater.



Other reports in this collection