Tropical cyclone winds retrieved from c-band cross-polarized synthetic aperture radar

Jochen Horstmann, Silvia Falchetti, Christopher Wackerman, Salvatore Maresca, Michael J. Caruso, Hans C Graber

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

This paper presents a geophysical model function (GMF) that has been developed to describe the relation of the ocean surface wind with the normalized radar cross section (NRCS) at C-band cross polarization (cross-pol). Synthetic aperture radar (SAR) images have been simultaneously collected at copolarization (co-pol) and cross-pol at moderate to high wind speeds. Using the SAR co-pol retrieved wind fields and an uncertainty estimate of the retrieved wind speeds, the cross-pol dependencies of the NRCS are investigated with respect to wind, incidence angle, and polarization pairs. For wind speeds above 10 m/s, there is a significant dependence of the NRCS on wind speed. However, the SAR cross-pol data are also significantly affected by the noise floor and crosstalk between the channels. Estimates of the noise floor are determined and removed from the NRCS. Three GMFs are developed: the first is for transmission at horizontal (H) polarization and the second at vertical (V) polarization. A third GMF accounts for wind direction dependence. Validation of the GMFs is conducted by comparison with collocated Stepped Frequency Microwave Radiometer (SFMR) data. The resulting bias of -0.7 m/s and standard deviation of 3.7 m/s demonstrate the excellent performance for these GMFs for wind speed retrieval between 10 and 35 m/s. Furthermore, comparisons show that SAR cross-pol retrieved wind speeds are of similar quality as those of SFMR and are significantly better in the moderate to high wind speed regime than SAR co-pol retrieved winds.

Original languageEnglish (US)
Article number7001637
Pages (from-to)2887-2898
Number of pages12
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume53
Issue number5
DOIs
StatePublished - May 1 2015

Fingerprint

Synthetic aperture radar
tropical cyclone
synthetic aperture radar
polarization
wind velocity
Polarization
Radar cross section
cross section
radar
microwave radiometer
Microwave frequencies
Radiometers
wind field
surface wind
wind direction
sea surface
Crosstalk

Keywords

  • cyclones
  • geophysical measurements
  • hurricanes
  • oceans sea surface
  • radar cross-sections
  • Radar remote sensing
  • synthetic aperture radar
  • tropical cyclones

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Earth and Planetary Sciences(all)

Cite this

Tropical cyclone winds retrieved from c-band cross-polarized synthetic aperture radar. / Horstmann, Jochen; Falchetti, Silvia; Wackerman, Christopher; Maresca, Salvatore; Caruso, Michael J.; Graber, Hans C.

In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 53, No. 5, 7001637, 01.05.2015, p. 2887-2898.

Research output: Contribution to journalArticle

Horstmann, Jochen ; Falchetti, Silvia ; Wackerman, Christopher ; Maresca, Salvatore ; Caruso, Michael J. ; Graber, Hans C. / Tropical cyclone winds retrieved from c-band cross-polarized synthetic aperture radar. In: IEEE Transactions on Geoscience and Remote Sensing. 2015 ; Vol. 53, No. 5. pp. 2887-2898.
@article{ee377c41261e498b9aabc46d7f280c15,
title = "Tropical cyclone winds retrieved from c-band cross-polarized synthetic aperture radar",
abstract = "This paper presents a geophysical model function (GMF) that has been developed to describe the relation of the ocean surface wind with the normalized radar cross section (NRCS) at C-band cross polarization (cross-pol). Synthetic aperture radar (SAR) images have been simultaneously collected at copolarization (co-pol) and cross-pol at moderate to high wind speeds. Using the SAR co-pol retrieved wind fields and an uncertainty estimate of the retrieved wind speeds, the cross-pol dependencies of the NRCS are investigated with respect to wind, incidence angle, and polarization pairs. For wind speeds above 10 m/s, there is a significant dependence of the NRCS on wind speed. However, the SAR cross-pol data are also significantly affected by the noise floor and crosstalk between the channels. Estimates of the noise floor are determined and removed from the NRCS. Three GMFs are developed: the first is for transmission at horizontal (H) polarization and the second at vertical (V) polarization. A third GMF accounts for wind direction dependence. Validation of the GMFs is conducted by comparison with collocated Stepped Frequency Microwave Radiometer (SFMR) data. The resulting bias of -0.7 m/s and standard deviation of 3.7 m/s demonstrate the excellent performance for these GMFs for wind speed retrieval between 10 and 35 m/s. Furthermore, comparisons show that SAR cross-pol retrieved wind speeds are of similar quality as those of SFMR and are significantly better in the moderate to high wind speed regime than SAR co-pol retrieved winds.",
keywords = "cyclones, geophysical measurements, hurricanes, oceans sea surface, radar cross-sections, Radar remote sensing, synthetic aperture radar, tropical cyclones",
author = "Jochen Horstmann and Silvia Falchetti and Christopher Wackerman and Salvatore Maresca and Caruso, {Michael J.} and Graber, {Hans C}",
year = "2015",
month = "5",
day = "1",
doi = "10.1109/TGRS.2014.2366433",
language = "English (US)",
volume = "53",
pages = "2887--2898",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Tropical cyclone winds retrieved from c-band cross-polarized synthetic aperture radar

AU - Horstmann, Jochen

AU - Falchetti, Silvia

AU - Wackerman, Christopher

AU - Maresca, Salvatore

AU - Caruso, Michael J.

AU - Graber, Hans C

PY - 2015/5/1

Y1 - 2015/5/1

N2 - This paper presents a geophysical model function (GMF) that has been developed to describe the relation of the ocean surface wind with the normalized radar cross section (NRCS) at C-band cross polarization (cross-pol). Synthetic aperture radar (SAR) images have been simultaneously collected at copolarization (co-pol) and cross-pol at moderate to high wind speeds. Using the SAR co-pol retrieved wind fields and an uncertainty estimate of the retrieved wind speeds, the cross-pol dependencies of the NRCS are investigated with respect to wind, incidence angle, and polarization pairs. For wind speeds above 10 m/s, there is a significant dependence of the NRCS on wind speed. However, the SAR cross-pol data are also significantly affected by the noise floor and crosstalk between the channels. Estimates of the noise floor are determined and removed from the NRCS. Three GMFs are developed: the first is for transmission at horizontal (H) polarization and the second at vertical (V) polarization. A third GMF accounts for wind direction dependence. Validation of the GMFs is conducted by comparison with collocated Stepped Frequency Microwave Radiometer (SFMR) data. The resulting bias of -0.7 m/s and standard deviation of 3.7 m/s demonstrate the excellent performance for these GMFs for wind speed retrieval between 10 and 35 m/s. Furthermore, comparisons show that SAR cross-pol retrieved wind speeds are of similar quality as those of SFMR and are significantly better in the moderate to high wind speed regime than SAR co-pol retrieved winds.

AB - This paper presents a geophysical model function (GMF) that has been developed to describe the relation of the ocean surface wind with the normalized radar cross section (NRCS) at C-band cross polarization (cross-pol). Synthetic aperture radar (SAR) images have been simultaneously collected at copolarization (co-pol) and cross-pol at moderate to high wind speeds. Using the SAR co-pol retrieved wind fields and an uncertainty estimate of the retrieved wind speeds, the cross-pol dependencies of the NRCS are investigated with respect to wind, incidence angle, and polarization pairs. For wind speeds above 10 m/s, there is a significant dependence of the NRCS on wind speed. However, the SAR cross-pol data are also significantly affected by the noise floor and crosstalk between the channels. Estimates of the noise floor are determined and removed from the NRCS. Three GMFs are developed: the first is for transmission at horizontal (H) polarization and the second at vertical (V) polarization. A third GMF accounts for wind direction dependence. Validation of the GMFs is conducted by comparison with collocated Stepped Frequency Microwave Radiometer (SFMR) data. The resulting bias of -0.7 m/s and standard deviation of 3.7 m/s demonstrate the excellent performance for these GMFs for wind speed retrieval between 10 and 35 m/s. Furthermore, comparisons show that SAR cross-pol retrieved wind speeds are of similar quality as those of SFMR and are significantly better in the moderate to high wind speed regime than SAR co-pol retrieved winds.

KW - cyclones

KW - geophysical measurements

KW - hurricanes

KW - oceans sea surface

KW - radar cross-sections

KW - Radar remote sensing

KW - synthetic aperture radar

KW - tropical cyclones

UR - http://www.scopus.com/inward/record.url?scp=84921027434&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84921027434&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2014.2366433

DO - 10.1109/TGRS.2014.2366433

M3 - Article

VL - 53

SP - 2887

EP - 2898

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

IS - 5

M1 - 7001637

ER -