Premium
A 1‐D variational retrieval of temperature, humidity, and liquid cloud properties: Performance under idealized and real conditions
Author(s) -
Ebell K.,
Löhnert U.,
Päschke E.,
Orlandi E.,
Schween J. H.,
Crewell S.
Publication year - 2017
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
eISSN - 2169-8996
pISSN - 2169-897X
DOI - 10.1002/2016jd025945
Subject(s) - liquid water content , microwave radiometer , environmental science , consistency (knowledge bases) , cloud computing , radiometer , meteorology , observatory , effective radius , microwave , remote sensing , computer science , physics , astrophysics , artificial intelligence , geography , operating system , telecommunications , galaxy
An extended version of the Integrated Profiling Technique (IPT) is presented. The IPT combines measurements from cloud radar and microwave radiometer (MWR) with prior information in a 1D‐Var approach in order to retrieve physically consistent atmospheric profiles of temperature, absolute humidity, liquid water content (LWC), and recently also cloud droplet effective radius (REF). Physical consistency implies the reproducibility of the measurements within the uncertainties. Results based on synthetic observations revealed a good retrieval performance with a high convergence rate of 98%. Retrieval uncertainties are typically around 0.06 g m −3 for LWC and 0.6 μm for REF. For the application to real measurements, quality‐controlled, bias‐free observations are crucial. A newly developed MWR spectral consistency check, which was applied to the measurements at the Jülich Observatory for Cloud Evolution (JOYCE), revealed strongly bias‐affected channels. The IPT itself can serve as a further quality check: particularly in clear‐sky cases, nonconvergence or physically inconsistent solutions may hint at measurement offset errors. Based on sensitivity studies, the final set of MWR frequencies was identified and the retrieval applied to 1 year of data. Physically consistent solutions could be found in 62% of all processed cases. A focus was put on the analysis of nondrizzling single‐layer water clouds which typically have small geometrical thicknesses (<300 m), low liquid water paths (<50 g m −2 ), and small REF (<5 μm). The retrieved data product contains a high potential for the analysis of warm cloud characteristics and, in combination with auxiliary information from the JOYCE instrumentation, of associated boundary layer processes.