Estimating interchannel observation‐error correlations for IASI radiance data in the Met Office system†

• DOI: 10.1002/qj.2211
• Publication Date: 2013-06-21T19:40:04Z
• Authors: L. M. Stewart; S. L. Dance; N. K. Nichols; J. R. Eyre; J. Cameron
• Keywords: 01 natural sciences; 0105 earth and related environmental sciences

Abstract The optimal utilisation of hyper‐spectral satellite observations in numerical weather prediction is often inhibited by incorrectly assuming independent interchannel observation errors. However, order to represent these observation‐error covariance structures, an accurate knowledge the true variances and correlations needed. This structure likely vary with type assimilation system. work this article presents initial results for estimation IASI when data are processed Met Office one‐dimensional (1D‐Var) four‐dimensional (4D‐Var) variational systems. method used calculate errors a post‐analysis diagnostic which utilises background analysis departures from two show significant differences source different systems, but also highlight some common features. When 1D‐Var, diagnosed error approximately half size current operational system very close instrument noise, suggesting that main error. contain no consistent correlations, exception handful spectrally channels. 4D‐Var, we again find being overestimated operationally, overestimation significantly larger many In contrast than noise 4D‐Var. It postulated horizontal representation, not seen contributor overall here. Finally, 4D‐Var found strong, correlation structures channels sensitive water vapour surface properties.