Estimation of CH4 emissions based on LETKF data assimilation technique

Jagat S. H. Bisht,  Research Institute for Global Change, JAMSTEC, Yokohama, 235-0019, Japan,  jagatbisht@jamstec.go.jp (Presenter)
Prabir K. Patra,  Research Institute for Global Change, JAMSTEC, Yokohama, 235-0019, Japan,  prabir@jamstec.go.jp
Masayuki Takigawa,  Research Institute for Global Change, JAMSTEC, Yokohama, 235-0019, Japan,  takigawa@jamstec.go.jp
Takashi Sekiya,  Research Institute for Global Change, JAMSTEC, Yokohama, 235-0019, Japan,  tsekiya@jamstec.go.jp
Yugo Kanaya,  Research Institute for Global Change, JAMSTEC, Yokohama, 235-0019, Japan,  yugo@jamstec.go.jp
Naoko Saitoh,  Center for Environmental Remote Sensing, Chiba University, Chiba, 263-8522, Japan,  nsaitoh@faculty.chiba-u.jp

Methane (CH4) is the second major long-lived greenhouse gas after carbon dioxide (CO2) which is continuously increasing during the last decade. Various studies attempted to attribute the emission sources of the recent rise in CH4 concentration using inverse modeling techniques, which are performed on very coarse resolution or on a limited number of predefined regions. It is very important to take full advantage of existing satellite observations and also prepare a roadmap for future satellite missions for CH4 flux estimation using a well-validated chemistry-transport model. Therefore, techniques such as ensemble Kalman filter (EnKF) are important to assimilate the CH4 observations. One of the advantages of Local Ensemble Transform Kalman Filter (LETKF) technique over 4-D Var data assimilation technique is non-requirement of adjoint model that makes it very simple and powerful tool for flux estimation using the most advanced atmospheric general circulation models. In this study we attempt to estimate the CH4 fluxes using LETKF data assimilation technique. We first test the performance and ability of our LETKF system against the known truth; hence this work performs sensitivity experiment on Observation System Simulation Experiment (OSSE) setting by updating observed changes only into the surface CH4 concentration. In our first sensitivity experiment over East and South Asia region, we found that LETKF is able to retrieve the true flux patterns and magnitudes from the perturbed emission intensities of 30% relative to the true flux over the East and South Asia region. Based on our past experiences with CH4 inverse modelling, the uncertainty in regional flux estimation is found to be at 30% or lower (Chandra et al., 2021).

Poster: Poster_Bisht__39_25.pdf 

Presentation Type: Poster

Session: 4.2b Flux estimates and atmospheric inversions from space-based GHG measurements

Session Date: Thursday (6/17) 10:00 AM