| Two recent studies indicate that our current understanding of the global budget of methane, the second strongest anthropogenic greenhouse gas, requires substantial revisions. First, Frankenberg et al (Science, 2005) showed that tropical methane emissions, determined from SCIAMACHY methane total column densities are far higher than current emissions inventories suggest. Second, Keppler et al (Nature, 2006) indeed show that there is a hitherto unknown methane source, viz. emissions from terrestrial plants under aerobic conditions. So far, emissions estimates are very uncertain but atmospheric inverse modelling applying global space borne retrievals of methane can help to largely reduce the uncertainties. For this purpose a high precision and accuracy (1-2%) is required on the methane column retrieval, which is not yet fully achieved using the present SCIAMACHY retrieval (Frankenberg et al 2005c,d). This proposal aims to improve the accuracy of methane column retrieval from SCIAMACHY near infrared measurements. To achieve this I first will improve the existing retrieval approach of Frankenberg et al (2005a). Here I will use a priori information on the atmospheric state and on the surface height to enhance the accuracy of the retrieval. A detailed error analysis of the method will allow a further fine-tuning of the approach. One principal limitation of the present retrieval approach is that it neglects any scattering of light by atmospheric aerosols and clouds. To overcome this limitation I propose to develop in a second phase of the project an alternative retrieval approach employing a simultaneous retrieval of methane, aerosol and cloud properties from SCIAMACHY measurements at different wavelength regions. This approach requires sophisticated radiative transfer simulations including multiple scattering of light. Finally I will assess the capability of subcolumn retrieval of methane from GOSAT instrument, scheduled for launch in 2008, which will perform measurements in the near and thermal infrared. |
