Multi-year emission inversion for reactive gases using the adjoint model method
Abstract
Although computationally expensive and difficult to program, the adjoint model technique is very appealing for addressing the problem of source inversion of reactive gases (like CO, NOx and the non-methane volatile organic compounds, NMVOCs) using atmospheric models. In particular, it allows accounting for the non-linear relationship existing between the emissions of such compounds and the concentrations of species observed from space (like CO, NO2 and HCHO), which involves complex transport and chemical processes. Moreover, the method has no limitation regarding the number of control parameters to be optimized using atmospheric observations. We present the development of an inverse modeling framework based on the IMAGES 3d global CTM and its adjoint. Applications included the joint optimization of CO and NOx sources using NO2 and CO observations; the comparison of the synthetic "big region" scheme with the more general, grid-based inversion scheme making use of spatial and temporal correlations between errors on the emission parameters; the use of one decade of satellite measurements of NO2 and HCHO for the inversion of NOx emissions from 4 categories and of short- lived NMVOC emissions from biogenic and pyrogenic sources. Future perspectives will be also discussed.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2007
- Bibcode:
- 2007AGUFM.A14D..02M
- Keywords:
-
- 0300 ATMOSPHERIC COMPOSITION AND STRUCTURE;
- 0315 Biosphere/atmosphere interactions (0426;
- 1610);
- 0322 Constituent sources and sinks;
- 0325 Evolution of the atmosphere (1610;
- 8125);
- 0365 Troposphere: composition and chemistry