 |
| Titel |
Isolating wetland CH4 emissions using the additional constraints of δ13CH4, and C2H6 in a inverse modeling framework |
| VerfasserIn |
Tonatiuh Guillermo Nuñez Ramirez, Julia Marshall, Sander Houweling, Edward J. Dlugokencky, Douglas E. J. Worthy, Bruce Vaughn, Isobel Simpson, James White, Willi A. Brand, Motoki Sasakawa, Silvia Nichol, Michel Ramonet, Stanley C. Tyler, Jacques Hueber, Detlev Helmig, Katie Read, Schalini Punjabi, Luciana Vanni Gatti, Paul Krummel, Martin Heimann |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250113260
|
| Publikation (Nr.) |
 EGU/EGU2015-13458.pdf |
|
|
|
|
|
| Zusammenfassung |
| \begin{document}
\noindent
Wetlands are the largest single source of atmospheric methane (CH$_4$). However, estimates of their relative contribution to the atmospheric CH$_4$ budget are highly uncertain. Models of CH$_4$ fluxes from wetlands, which reflect our understanding of the processes driving these fluxes, disagree strongly in their estimates of the total contribution of wetlands to the CH$_4$ budget and in the variability of the fluxes in space and time. Atmospheric CH$_4$ observations can provide a top-down constraint on wetland CH$_4$ flux estimates. Results from atmospheric inverse modeling studies highlight the importance of tropical wetlands in driving interannual variability of atmospheric CH$_4$. Nevertheless, atmospheric observations in the tropics are scarce, with large areas of strong emissions not covered by the atmospheric observation network. Furthermore, the Bayesian framework, often used in atmospheric inverse modeling, preferentially projects signals onto spatiotemporal regions with large a-priori uncertainty, which is the case of tropical wetlands. Since a large lack of knowledge exists as well for other non-wetland sources of atmospheric CH$_4$, signals from these could be wrongly allocated to tropical wetlands. The CH$_4$ stable carbon isotope signal ($\delta^{13}$CH$_4$) and co-emitted species such as ethane (C$_2$H$_6$) can provide additional constraints which may be use to discriminate wetland from non-wetland CH$_4$ emissions. We describe the set-up of an inverse modeling framework based on the Jena Inversion System and the TM3 transport model that optimizes CH$_4$ fluxes to fit the observed atmospheric CH$_4$, $\delta^{13}$CH$_4$, and C$_2$H$_6$ signals. The fluxes are optimized with the constraint that each source process was assigned a characteristic range of $\delta^{13}$CH$_4$ signals and methane-to-ethane ratios (MERs). An additional characteristic of our set-up is that no seasonal or interannual variability was included in the wetland a-priori estimate to ensure that all variability is derived exclusively from observations. A-priori estimates for wetland CH$_4$ emissions were composed of the temporal mean per gridcell of the six global models that participated in the Wetland and Wetland CH$_4$ Inter-comparison of Models Project (WETCHIMP) models. The a-priori uncertainty was derived by combining the temporal variability and the variability among the models per gridcell. Interestingly enough, both $\delta^{13}$CH$_4$ and C$_2$H$_6$ signals drive the scaling of the fluxes in a similar direction. Nevertheless, the uncertainty in the characteristic emission ratios among the CH$_4$, $\delta^{13}$CH$_4$, and C$_2$H$_6$ tracers must be assessed carefully to avoid giving too many degrees of freedom to the inversion system and thus matching the atmospheric signals with minimal changes in the fluxes. Optimized fluxes can be used to find relationships between CH$_4$ emissions from wetlands and their meteorological drivers.
\end{document} |
|
|
|
|
|
|