Change search
ReferencesLink to record
Permanent link

Direct link
Transitioning From CRD to CDRD in Bayesian Retrieval of Rainfall From Satellite Passive Microwave Measurements: Part 2. Overcoming Database Profile Selection Ambiguity by Consideration of Meteorological Control on Microphysics
Show others and affiliations
2013 (English)In: IEEE Transactions on Geoscience and Remote Sensing, ISSN 0196-2892, E-ISSN 1558-0644, Vol. 51, no 9, 4650-4671 p.Article in journal (Refereed) Published
Abstract [en]

A new cloud dynamics and radiation database (CDRD) precipitation retrieval algorithm for satellite passive microwave (PMW) radiometer measurements has been developed. It represents a modification to and an improvement upon the conventional cloud radiation database (CRD) algorithms, which have always been prone to ambiguity. This part 2 paper of a series describes the methodology of the algorithm and the modeling verification analysis involved in creating a synthetic CDRD database for the Europe/Mediterranean basin region. This is followed by a proof-of-concept analysis, which demonstrates that the underlying CDRD theory based on use of meteorological parameters for reducing retrieval ambiguity is valid. This paper uses a regional/mesoscale model, applied in cloud resolving model (CRM) mode, to produce a large set of numerical simulations of precipitating storms and extended precipitating systems. The simulations are used for selection of millions of meteorological/microphysical vertical profiles within which surface rainfall is identified. For each of these profiles, top-of-atmosphere brightness temperature (TB) vectors are calculated (the vector dimension associated with the number of relevant cm-mm wavelengths and polarizations), based on an elaborate radiative-transfer equation (RTE) model system (RMS) coupled to the CRM. This entire body of simulation information is organized into the CDRD database, then used as a priori knowledge to guide a physical Bayesian retrieval algorithm in obtaining rainfall and associated precipitation parameters from the PMW satellite observations. We first prove the physical validity of our CRM-RMS simulations, by showing that the simulated TBs are in close agreement with observations. Agreement is demonstrated using dual-channel-frequency TB manifold sections, which quantify the degree of overlap between the simulated and observed TBs extracted from the full manifolds. Nevertheless, the salient result of this paper is a proof that the underlying CDRD theory is valid, found by combining subdivisions of the invoked meteorological parameter ranges of values and showing that such meteorological partitioning associates itself with distinct microphysical profiles. It is then shown that these profiles give rise to similar TB vectors, proving the existence of ambiguity in a CRD-type algorithm. Finally, we show that the CDRD methodology provides significant improvements in reducing retrieval ambiguity and retrieval error, especially for land surface backgrounds where contrasts are typically small between the rainfall TB signatures and surface emission signatures.

Place, publisher, year, edition, pages
2013. Vol. 51, no 9, 4650-4671 p.
Keyword [en]
Bayesian precipitation retrieval, cloud dynamics and radiation database (CDRD) algorithm, precipitation retrieval ambiguity, satellite rainfall retrieval, simulation-observation TB manifolds
National Category
Earth and Related Environmental Sciences
URN: urn:nbn:se:su:diva-94937DOI: 10.1109/TGRS.2013.2258161ISI: 000324434900005OAI: diva2:657038


Available from: 2013-10-17 Created: 2013-10-17 Last updated: 2013-10-17Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full text

Search in DiVA

By author/editor
Leung, Wing-Yee Hester
By organisation
Department of Meteorology
In the same journal
IEEE Transactions on Geoscience and Remote Sensing
Earth and Related Environmental Sciences

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 94 hits
ReferencesLink to record
Permanent link

Direct link