Enhancing the impact of IASI observations through an updated observation‐error covariance matrix. (26th April 2016)

• Record Type: Journal Article
• Authors: Bormann, Niels; Bonavita, Massimo; Dragani, Rossana; Eresmaa, Reima; Matricardi, Marco; McNally, Anthony
• Is Part Of: Quarterly journal of the Royal Meteorological Society. Volume 142:Number 697(2016)
• Page Start: 1767
• Page End: 1780
• Publication Date: 2016-04-26
• Subjects: data assimilation -- observation error -- hyperspectral infrared observations; Meteorology -- Periodicals; 551.5
• DOI: 10.1002/qj.2774
• Languages: English
• ISSNs: 0035-9009

Abstract

Abstract : This article investigates the use of an updated observation‐error covariance matrix for the Infrared Atmospheric Sounding Interferometer (IASI) in the European Centre for Medium‐Range Weather Forecasts (ECMWF) system. The new observation‐error covariance matrix is based on observation‐space diagnostics and includes interchannel error correlations, but also assigns significantly altered error standard deviations. The update is investigated in detail in assimilation experiments, including an assessment of the role of error inflation and taking interchannel error correlations into account. The updated observation‐error covariance leads to a significant improvement in the use of IASI data, especially in the Tropics and the stratosphere and particularly for humidity and ozone. The benefits are especially strong for short‐range forecasts, whereas the impact in the medium range is less pronounced. The study highlights the benefits of taking interchannel error correlations into account, which allows the use of an observation‐error covariance for IASI that is overall more consistent with departure statistics. At the same time, the study also demonstrates that error inflation can be used to compensate partially, though not fully, for neglected error correlations. Adjustments such as scaling of the originally diagnosed observation‐error estimates are also found to be beneficial when the diagnosed interchannel error correlations are taken into account.