Influence of CO2 observations on the optimized CO2 flux in an ensemble Kalman filter
Atmospheric Chemistry and Physics Discussions
Various data assimilation schemes have been applied in studies on atmospheric CO<sub>2</sub> inversion. An influence matrix based on the linear statistical analysis scheme can diagnose the impact of individual observations on a particular analysis. In this study, to estimate the effect of CO<sub>2</sub> observations on an analysis of surface CO<sub>2</sub> flux, both the analysis sensitivity and the information content were calculated using the influence matrix in the CarbonTracker, which is an
... racker, which is an inverse modeling system for estimating surface CO<sub>2</sub> flux based on an ensemble Kalman filter. The experimental period was from January 2000 to December 2009. The global average self-sensitivity is 4.8%, which implies that the analysis extracts 4.8% of the information from the observations and 95.2% from the background each assimilation cycle. Because the surface CO<sub>2</sub> flux in each week is optimized by five weeks of observations, the cumulative impact over five weeks would be greater than 4.8%. The analysis sensitivity is inversely proportional to the number of observations used in the assimilation, which is distinctly apparent in continuous observation categories with a sufficient number of observations. The time series of the globally averaged analysis sensitivities shows seasonal variations, with greater sensitivities in summer and lower sensitivities in winter, which is attributed to the surface CO<sub>2</sub> flux uncertainty. The time-averaged analysis sensitivities in the Northern Hemisphere are greater than those in the Tropics and the Southern Hemisphere. The information content indicates an imbalance between the observation coverage in North America and that in other regions. Approximately half of the total observational information is provided by continuous observations, mainly from North America, which indicates that continuous observations are the most informative and that comprehensive coverage of additional observations in other regions is necessary to estimate the surface CO<sub>2</sub> flux in these areas as accurately as in North America. In addition, the uncertainty of the surface CO<sub>2</sub> flux in Asia, where observations are sparse, is reduced by assimilating five weeks of observations as opposed to one week of observations in North America, which indicates that a longer assimilation window with a lag is necessary to optimize the surface CO<sub>2</sub> flux in Asia.