Impact of correlated observation errors on inverted CO2 surface fluxes from OCO measurements

摘要: [1] Proper assignment of error statistics is essential in the field Bayesian inference. This paper studies impact correlated observation errors case estimation CO2 surface fluxes from NASA's forthcoming Orbiting Carbon Observatory (OCO). Using a series simulation system experiments, it shown that hypothetical correlations 0.5 neighbouring observations have rather limited on accuracy inverted when they are correctly taken into account. The information loss induced by commonly-used approximate treatments (neglecting, thinning and inflating), computationally more efficient, quantified. Error inflation has least detrimental among suboptimal set-ups limits uncertainty reduction to few per cent, spite its very low reduced chi-squared.