@InProceedings{YunLiByWeOtGaMc:2023:QuErRe,
               author = "Yun, Jeongmin and Liu, Junjie and Byrne, Brendan and Weir, Brad 
                         and Ott, Lesley E. and Gatti, Luciana Vanni and McKain, Kathryn",
          affiliation = "{California Institute of Technology} and {California Institute of 
                         Technology} and {California Institute of Technology} and {NASA 
                         Goddard Space Flight Center} and {NASA Goddard Space Flight 
                         Center} and {Instituto Nacional de Pesquisas Espaciais (INPE)} and 
                         {NOAA ESRL Global Monitoring Division}",
                title = "Quantifying errors in regional terrestrial biosphere CO2 fluxes in 
                         OCO-2 MIP models using aircraft measurements",
            booktitle = "Proceedings...",
                 year = "2023",
         organization = "AGU FAll Meeting",
            publisher = "AGU",
             abstract = "Multi-inverse modeling inter-comparison projects (MIPs) provide a 
                         chance to assess the uncertainties in inversion estimates arising 
                         from various sources such as atmospheric CO2 observations, 
                         transport models, and prior fluxes. However, accurately 
                         quantifying ensemble flux errors remains challenging, often 
                         relying on the ensemble spread as a surrogate. This presentation 
                         proposes an approach to quantify the errors of regional 
                         terrestrial CO2 flux estimates from ten inverse models within the 
                         Orbiting Carbon Observatory-2 MIP by utilizing independent 
                         aircraft CO2 measurements for the period 2015-2017. We first 
                         calculate the root-mean-square error (RMSE) between the ensemble 
                         mean of posterior CO2 concentrations and aircraft observations and 
                         then isolates the CO2 concentration errors caused solely by 
                         posterior CO2 fluxes by subtracting the errors of representation, 
                         transport, and observation in seven regions. Our analysis reveals 
                         significant regional variations in the average monthly RMSE over 
                         three years, ranging from 0.90 to 2.04 ppm. Posterior flux error 
                         is a major component that accounts for 57-83% of the mean RMSE. We 
                         further show that in five regions, the true posterior flux errors 
                         projected in CO2 space exceed the atmospheric CO2 errors resulted 
                         from the ensemble spread of posterior CO2 flux estimates by 
                         1.4-1.9 times, implying underestimation of the true posterior flux 
                         errors, while their magnitudes are comparable in two regions. By 
                         identifying most sensitive areas to aircraft measurements through 
                         adjoint sensitivity analysis, we find that the underestimation of 
                         terrestrial flux errors is prominent in eastern parts of Australia 
                         and East Asia, western parts of Europe and Southeast Asia, and 
                         midlatitude North America. However, no such underestimation is 
                         observed in southern Alaska and northeastern South America, 
                         suggesting the presence of systematic biases related to 
                         anthropogenic CO2 emissions in inversion estimates. Our study 
                         highlights the critical role of aircraft measurements not only for 
                         qualitatively evaluating inversion performance but also for 
                         quantifying regional errors in ensemble flux estimates.",
  conference-location = "San Francisco, CA",
      conference-year = "11-15 Dec. 2023",
             language = "en",
        urlaccessdate = "21 maio 2024"
}