[A-CG41] Satellite Earth Environment Observation

[ACG41-02] The Indian terrestrial CO₂ sink estimated with an inverse model using observations from surface networks and the GOSAT satellite

*Lorna Raja Nayagam¹, Shamil Maksyutov¹, Rajesh Janardanan¹, Tomohiro Oda^2,3, Yogesh K. Tiwari⁴, Gaddamidi Sreenivas⁴, Amey Datye⁴, Chaithanya D. Jain⁵, Madineni Venkat Ratnam⁵, Vinayak Sinha⁶, Haseeb Hakkim⁶, Yukio Terao¹, Manish Naja⁷, Md. Kawser Ahmed⁸, Hitoshi Mukai¹, Jiye Zeng¹, Johannes W. Kaiser⁹, Yu Someya¹, Yukio Yoshida¹, Tsuneo Matsunaga¹ (1.National Institute for Environmental Studies, 2.Earth from Space Institute, Universities Space Research Association, 3.Department of Atmospheric and Oceanic Science, University of Maryland, 4.Indian Institute of Tropical Meteorology, 5.National Atmospheric Research Laboratory, 6.Indian Institute of Science Education and Research, 7.Aryabhatta Research Institute of Observational Sciences, 8.Department of Oceanography, University of Dhaka, 9.The Climate and Environmental Research Institute NILU)

Keywords:Carbon dioxide, Indian terrestrial biosphere , GOSAT, Surface observations, Inverse modeling

The knowledge on the sources/sinks at a regional scale is significant to account for the global total CO₂ emissions. In this work, we estimate the biospheric CO₂ emissions over India from the observations of the global surface network, an Indian local surface network and the Greenhouse Gases Observing Satellite (GOSAT) satellite using the NIES-TM_FLEXPART-VAR (NTFVAR) global high-resolution inverse model. The inverse model optimizes terrestrial biosphere fluxes and ocean-atmosphere exchanges independently at a 0.1° x 0.1° resolution. The optimized fluxes are comparable to regional land and ocean fluxes from the OCO-2 Model Intercomparison Project (OOC-2 MIP). Further, sensitivities of the observations from the Indian surface network and the GOSAT satellite to the regional fluxes are analyzed. The results suggest a stronger sink (-0.147 ± 0.094 PgC yr⁻¹ ) over India when Indian observations are used, and a marginal sink without these observations (-0.040 ± 0.133 PgC yr⁻¹). In addition, we find that the magnitude of sink is reduced by using filtered GOSAT observations, which suggests that the GOSAT satellite observations also constrain the fluxes over the region. While satellite data provide good constraints on fluxes over the region, obtaining a robust carbon budget estimate over the region is only possible by fully utilizing observations from both the surface and satellites, especially because the GOSAT satellite data yield tends to significantly decrease in summer under cloudy conditions. Moreover, the observed seasonal cycle amplitude appears to be stronger than the simulated with prior fluxes. This suggests a potential of using the local surface atmospheric observations for improving the quantitative understanding of the Indian terrestrial carbon cycle.