 |
| Titel |
Hydroxyl radical variability and its influence on recent methane growth inferred from methyl chloroform trends |
| VerfasserIn |
Matthew Rigby, Steve Montzka, Ronald Prinn, James White, Dickon Young, Simon O'Doherty, Anita Ganesan, Alistair Manning, Peter Simmonds, Peter Salameh, Chris Harth, Jens Muhle, Ray Weiss, Paul Fraser, Paul Steele, Paul Krummel, Archie McCulloch, Sunyoung Park |
| Konferenz |
EGU General Assembly 2017
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 19 (2017) |
| Datensatznummer |
250146059
|
| Publikation (Nr.) |
 EGU/EGU2017-10053.pdf |
|
|
|
|
|
| Zusammenfassung |
| Tropospheric hydroxyl radical (OH) concentrations can be inferred using long-term trends in
methyl chloroform (CH3CCl3). However, it has long been known that inaccuracies in the
assumed emissions of CH3CCl3 have the potential to induce significant systematic errors in
the OH levels derived using this approach. To address this, we present a Bayesian study in
which uncertain parameters within a model of CH3CCl3 emissions were explored using
atmospheric CH3CCl3 data. This allows us to bring the model into consistency
with informative features, such as the observed inter-hemispheric gradient, and
propagate any remaining uncertainties through to our OH estimates. Using this
approach, we find a maximum-likelihood solution that indicates an increase in
OH concentrations of around 10% between the late 1990s and mid-2000s, and a
subsequent fall of similar magnitude from the mid-2000s to 2014. However, significant
uncertainties remain such that a “constant OH” trajectory is also plausible from our
estimation framework. When these uncertainties are considered, the probability
of some level of decline in OH between 2007 and 2014 is between 64 and 70%.
By including CH4 and δ13C-CH4 observations in our inversion, we can propagate
these OH concentrations and uncertainties through to top-down estimates of CH4
emissions. Our maximum-likelihood solution does not require a pause and then sudden
growth in CH4 emissions around the years 2000 and 2007, respectively, as has
been suggested elsewhere. Rather, emissions increase relatively gradually over
the last two decades. Furthermore, this solution suggests that OH changes have
contributed to the recent decline in δ13C-CH4 and the growth in ethane. However, the
uncertainties in our CH4 inversion, which derive largely from the OH estimates, are
significant. Without careful consideration of the above OH uncertainties, it would be
possible to draw over-confident conclusions about the causes of recent CH4 changes. |
|
|
|
|
|
|