Bayesian retrieval of scatterometer wind fields

Dan Cornford, Ian T. Nabney, David J. Evans

Research output: Working paperTechnical report

Abstract

The retrieval of wind fields from scatterometer observations has traditionally been separated into two phases; local wind vector retrieval and ambiguity removal. Operationally, a forward model relating wind vector to backscatter is inverted, typically using look up tables, to retrieve up to four local wind vector solutions. A heuristic procedure, using numerical weather prediction forecast wind vectors and, often, some neighbourhood comparison is then used to select the correct solution. In this paper we develop a Bayesian method for wind field retrieval, and show how a direct local inverse model, relating backscatter to wind vector, improves the wind vector retrieval accuracy. We compare these results with the operational U.K. Meteorological Office retrievals, our own CMOD4 retrievals and a neural network based local forward model retrieval. We suggest that the neural network based inverse model, which is extremely fast to use, improves upon current forward models when used in a variational data assimilation scheme.
Original languageEnglish
Place of PublicationBirmingham
PublisherAston University
Publication statusUnpublished - 1999

Fingerprint

scatterometer
wind field
backscatter
heuristics
data assimilation
weather
prediction

Keywords

  • retrieval
  • wind fields
  • scatterometer observations
  • local wind vector retrieval
  • ambiguity removal

Cite this

Cornford, D., Nabney, I. T., & Evans, D. J. (1999). Bayesian retrieval of scatterometer wind fields. Birmingham: Aston University.
Cornford, Dan ; Nabney, Ian T. ; Evans, David J. / Bayesian retrieval of scatterometer wind fields. Birmingham : Aston University, 1999.
@techreport{3422519485384ab0802c57af4d271cb6,
title = "Bayesian retrieval of scatterometer wind fields",
abstract = "The retrieval of wind fields from scatterometer observations has traditionally been separated into two phases; local wind vector retrieval and ambiguity removal. Operationally, a forward model relating wind vector to backscatter is inverted, typically using look up tables, to retrieve up to four local wind vector solutions. A heuristic procedure, using numerical weather prediction forecast wind vectors and, often, some neighbourhood comparison is then used to select the correct solution. In this paper we develop a Bayesian method for wind field retrieval, and show how a direct local inverse model, relating backscatter to wind vector, improves the wind vector retrieval accuracy. We compare these results with the operational U.K. Meteorological Office retrievals, our own CMOD4 retrievals and a neural network based local forward model retrieval. We suggest that the neural network based inverse model, which is extremely fast to use, improves upon current forward models when used in a variational data assimilation scheme.",
keywords = "retrieval, wind fields, scatterometer observations, local wind vector retrieval, ambiguity removal",
author = "Dan Cornford and Nabney, {Ian T.} and Evans, {David J.}",
year = "1999",
language = "English",
publisher = "Aston University",
type = "WorkingPaper",
institution = "Aston University",

}

Cornford, D, Nabney, IT & Evans, DJ 1999 'Bayesian retrieval of scatterometer wind fields' Aston University, Birmingham.

Bayesian retrieval of scatterometer wind fields. / Cornford, Dan; Nabney, Ian T.; Evans, David J.

Birmingham : Aston University, 1999.

Research output: Working paperTechnical report

TY - UNPB

T1 - Bayesian retrieval of scatterometer wind fields

AU - Cornford, Dan

AU - Nabney, Ian T.

AU - Evans, David J.

PY - 1999

Y1 - 1999

N2 - The retrieval of wind fields from scatterometer observations has traditionally been separated into two phases; local wind vector retrieval and ambiguity removal. Operationally, a forward model relating wind vector to backscatter is inverted, typically using look up tables, to retrieve up to four local wind vector solutions. A heuristic procedure, using numerical weather prediction forecast wind vectors and, often, some neighbourhood comparison is then used to select the correct solution. In this paper we develop a Bayesian method for wind field retrieval, and show how a direct local inverse model, relating backscatter to wind vector, improves the wind vector retrieval accuracy. We compare these results with the operational U.K. Meteorological Office retrievals, our own CMOD4 retrievals and a neural network based local forward model retrieval. We suggest that the neural network based inverse model, which is extremely fast to use, improves upon current forward models when used in a variational data assimilation scheme.

AB - The retrieval of wind fields from scatterometer observations has traditionally been separated into two phases; local wind vector retrieval and ambiguity removal. Operationally, a forward model relating wind vector to backscatter is inverted, typically using look up tables, to retrieve up to four local wind vector solutions. A heuristic procedure, using numerical weather prediction forecast wind vectors and, often, some neighbourhood comparison is then used to select the correct solution. In this paper we develop a Bayesian method for wind field retrieval, and show how a direct local inverse model, relating backscatter to wind vector, improves the wind vector retrieval accuracy. We compare these results with the operational U.K. Meteorological Office retrievals, our own CMOD4 retrievals and a neural network based local forward model retrieval. We suggest that the neural network based inverse model, which is extremely fast to use, improves upon current forward models when used in a variational data assimilation scheme.

KW - retrieval

KW - wind fields

KW - scatterometer observations

KW - local wind vector retrieval

KW - ambiguity removal

M3 - Technical report

BT - Bayesian retrieval of scatterometer wind fields

PB - Aston University

CY - Birmingham

ER -

Cornford D, Nabney IT, Evans DJ. Bayesian retrieval of scatterometer wind fields. Birmingham: Aston University. 1999.