Articles in this issue
are currently in Adobe
Acrobat format only.
You need Adobe
Acrobat Reader to
view them.

Earth Observation Quarterly no. 59, June 1998

Special Edition: Wind field retrievals from SAR

IN THIS ISSUE

Introduction - Wind field retrievals from SAR
J.A. Johannessen, E. Attema & Y.-L. Desnos
Determination of wind energy from SAR images for siting
windmill locations
O.M. Johannessen & Erik Korsbakken
Summary / Full text (259Kb)
Assessment of ERS-1 SAR wind-speed estimates using an
airborne altimeter
D. Vandemark, P.W. Vachon & B. Chapron
Summary / Full text (469Kb)
Ocean wind fields and their variability derived from SAR
J. Horstmann, W. Koch, S. Lehner & W. Rosenthal
Summary / Full text (1005Kb)
Wind field structure and speed from Radarsat SAR images
P.W. Vachon, I. Chunchuzov & F.W. Dobson
Summary / Full text (579Kb)
Analysis of the wind field during the 'Vendée Globe' race:
A kinematic SAR wind speed algorithm
V. Kerbaol, B. Chapron & P. Queffeulou
Summary / Full text (425Kb)
High-resolution wind fields from ERS SAR
K. Mastenbroek
Summary / Full text (636Kb)
Wind field retrieval from SAR compared with scatterometer
wind field during ERS Tandem phase
E. Korsbakken & B. Furevik
Summary / Full text (461Kb)
1997/98 El Niño observed by ERS
K. Cardon, P. Goryl, R. Scharroo & J. Benveniste
Summary / Full text (4282Kb)
Deforestation evaluation by synergetic use of ERS SAR coherence
and ATSR hot spots: The Indonesian fire event of 1997
E. Antikidis, O. Arino, H. Laur & A. Arnaud
Summary / Full text (1189Kb)
Imaging the 1997 Marseille fire
I. Piccolini, O. Arino, A. Arnaud & J-M. Rosaz
Summary / Full text (939Kb)
Conferences & Publications
Full text (129Kb)
Record of Images (Supplement)
Full text (952Kb)

This issue of EOQ is dedicated to the different ways of retrieving quantitative information about the near surface wind field from SAR backscatter measurements and their spectral properties. Wind field retrieval from SAR data are now sufficiently mature to be used in operational applications. This is clearly documented in a series of recent publications including: a) the Proceedings of the third ERS Symposium (ESA SP-414, 1997); b) the Committee on Earth Observation Satellites (CEOS) Wind and Wave Validation Workshop (ESA WPP-147, 1997); c) the CEOS SAR Workshop (ESA WPP-138, 1998); and d) the special collection of papers on advances in oceanography and sea ice research using ERS observations in the Journal of Geophysical Research - Ocean (JGR 101, April 1998). The articles included in this issue are meant to provide an overview of this new SAR retrieval capability.

Global ocean wind fields are retrieved by instruments such as the ERS AMI wind scatterometer on a routine basis at a spatial scale of 50 km. Higher resolution regional and local wind fields at about 2 to 10 km can, on the other hand, be derived from SAR observations. Such detailed wind fields are more attractive for applications in coastal regions, smaller enclosed seas and in the vicinity of the marginal ice zones. This is not only for scientific interest (such as wind-wave coupling and air-sea interaction studies) but also for practical applications including towing operations, harbour and pilot boat operations, leisure boating and selection of sites for wind mills.

Following the launch of Envisat in 2000, data from a new and advanced synthetic aperture radar (ASAR) will become available. In addition to ensuring the continuity of the ERS SAR observations with a more versatile wave measurement mode, the ASAR instrument will offer new operating modes which look very promising for wind retrievals. In particular, the wide-swath mode (150 m resolution) and global monitoring mode (1 km resolution) have the potential to deliver fine-resolution maps over a swath of more than 400 km. Additionally the alternating polarisa-tion mode which uses a combination of looks and transmit-receive polarisations may turn out to become very useful.

Thanks to the advances in the analysis and interpretation of ERS SAR data, ESA has made significant improve-ments to the ASAR wave-mode algo-rithms and products. The ASAR wave-mode Level-1b product will be derived by processing single-look complex (SLC) imagettes using the cross spectra methodology introduced recently by Engen et al., (1995). This product is completely ‘speckle’ compensated and allow the wave propagation ambiguity to be resolved. Furthermore the wind speed can be derived directly from the backscatter intensity or retrieved from the azimuth cut-off estimates. In order to adapt existing ERS wave spectra inversion algorithms to the ASAR cross spectra, ESA has decided to initiate algorithm prototyping for a new Level-2 product which will include ocean wave spectra together with near surface wind field estimate. This ASAR wind and wave Level-2 product will also be assimilated experimentally on a routine basis at ECMWF and validated against model predictions.

The following articles together with the Level-2 product and further advances that can be expected when ASAR data become available clearly demonstrates the potential for new operational applications and research into coupled atmosphere/ocean interactive processes.

Determination of wind energy from SAR images for siting
windmill locations
O.M. Johannessen & E. Korsbakken

Satellite-based Synthetic Aperture Radar (SAR) measurements from the European Space Agency (ESA) are proposed for wind-energy mapping in coastal regions for windmill location siting. The instrument has clear advantages for high-spatial resolution wind-field mapping since it is independent of daylight and clouds. The instrument’s spatial resolution of 30 m is sufficient as is the 100 km-wide and several hundred km-long spatial coverage along the coastline. Wind maps generated from SAR will be able to provide spatial information concerning wind energy at 10m height in the vertical plane. We have, for the first time, used SAR for wind-energy mapping along a coastline and suggest that this can be an important method for selecting optimum loca-tions for planned windmill parks.
Full text (259Kb)

Assessment of ERS-1 SAR wind-speed estimates using an
airborne altimeter
D. Vandemark, P.W. Vachon & B. Chapron

Ocean-surface wind speeds that are estimated from SAR imagery could benefit both coastal meteorological and oceano-graphic research. However, ERS-1 SAR wind-speed estimation is not well validated due to the small number of collocated in-situ wind measurements. We present wind-speed measurements from airborne Ku-band altimeter flights under the ERS-1 SAR. Wind speeds were estimated from the altimeter data for direct comparison and assessment of those estimated from the SAR. The large comparison dataset shows that, within noted limitations, properly calibrated ERS-1 SAR data can provide valid wind-speed estimates. Such aircraft observations may be an effective approach for future SAR validation activities.
Full text (469Kb)

Ocean wind fields and their variability derived from SAR
J. Horstmann, W. Koch, S. Lehner & W. Rosenthal

The Synthetic Aperture Radar (SAR) aboard the European Space Agency’s remote-sensing satellites (ERS-1 and ERS-2) acquires images that can be used to derive wind fields over the ocean surface. The SAR measures the backscatter, which is a measure of the roughness of the ocean surface. The roughness is strongly influenced by the local wind field so that the radar backscatter can be used to measure the wind. For the derivation of the wind field from SAR images the empirical C-band models are applied, which were originally developed for the wind scatterometer (SCAT), that operates at the same wavelength as the SAR. Scenes at different geographical locations and under different meteorological and oceanographical conditions were selected to test the applicability of the ERS SAR for retrieving wind fields over the ocean. Especially in coastal regions, the SAR-derived mesoscale wind fields give a lot of additional information. As an example, an ERS SAR-retrieved wind field is compared to ground truth measurements and to the results of a mesoscale atmospheric model.
Full text (1005Kb)

Wind field structure and speed from Radarsat SAR images
P.W. Vachon, I. Chunchuzov & F.W. Dobson

Two examples of Radarsat ScanSAR wide images of mesoscale cyclones in the Labrador Sea are presented, along with wind retrieval validation results from both the ERS SARs and the Radarsat SAR. The results show that SAR images may be used to deduce information about both the spatial structure and the wind speed of mesoscale phenomena near the ocean’s surface.
Full text (579Kb)

Analysis of the wind field during the 'Vendée Globe' race:
A kinematic SAR wind speed algorithm
V. Kerbaol, B. Chapron & P. Queffeulou

A direct application of the recently derived SAR wind speed retrieval algorithm is presented in the particular case of the extremely high wind conditions endured by the competitors of the Vendée Globe sailing race between 7 and 8 January 1998. The wind speed analysis of a SAR ERS-2 wave-mode ‘imagette’ is presented to complete the measurements of other sensors (NSCAT, Topex and ERS-2 altimeter/scatterometer). In this area known as the ‘Howling Fifties’, the wind speeds during this period were extreme. Rapid veering conditions were as high as 23.6 m/s with the significant wave-height parameter estimated as high as 9 m.
Full text (425Kb)

High-resolution wind fields from ERS SAR
K. Mastenbroek

Studies of wind and wave climate in estuaries and lakes suffer from a lack of in-situ measurements. Using the high-resolution images acquired by spaceborne SARs, this gap may be filled. Examples of wind fields retrieved from ERS SAR images of the Dutch IJsselmeer and coastal waters show the potential of this technique.
Full text (636Kb)

Wind field retrieval from SAR compared with scatterometer
wind field during ERS Tandem phase
E. Korsbakken & B. Furevik

An evaluation of wind retrieval by SAR has been performed by comparing computed winds from ERS-2 SAR with ERS-1 microwave scatterometer measurements. This was possible during the Tandem phase of the ERS project, for latitudes greater than about 63°. In this case the scatterometer and SAR coverage overlapped with a time difference of only 30 minutes. Preliminary results confirm the usefulness of SAR in determining wind speed at high spatial resolution, provided that some information on wind direction can be obtained (e.g. from wind streaks).
Although the comparison was performed over the Norwegian/Greenland Sea, the results will be applicable to retrieving wind over lower-latitude areas and in coastal regions, at a resolution better than the 50 km resolution obtainable from the scatterometer.
Full text (461Kb)

1997/98 El Niño observed by ERS
K. Cardon, P. Goryl, R. Scharroo & J. Benveniste

The recent El Niño Southern Oscillation (ENSO) of 1997/1998 is one of the strongest of the century. This event is characterised by abnormally high sea level and sea surface temperatures along the west coast of South America, near the Equator.
With the Radar Altimeter (RA) and the Along-Track Scanning Radiometer (ATSR) on-board the second European Remote Sensing satellite (ERS-2), this event was monitored in great detail. Animations clearly show the development of the latest El Niño. Particularly, the sudden Kelvin wave at its onset is clearly recognised. It can also be seen that this El Niño actually hit the South American coast twice, once in July 1997 and then again in October; both are of great magnitude and initiated unusual-ly early in the year.
Full text (4282Kb)

Deforestation evaluation by synergetic use of ERS SAR coherence
and ATSR hot spots: The Indonesian fire event of 1997
E. Antikidis, O. Arino, H. Laur & A. Arnaud

The synergetic use of the ERS SAR and ATSR data allows the evaluation of the Kalimantan deforestation during the Indonesian fire event of 1997. Changes in forest coherence signature given by the interferometric SAR data before and after the fire events are highly correlated to hot spots detected with the ATSR during the fire events themselves. These specific coherence temporal signatures are then used to perform a classification of the deforested regions. The validation with in-situ data is now ongoing in coordination with the TREES project. The method strongly suggests that regional maps of burned forest can be derived from ATSR and tandem SAR data.
Full text (1189Kb)

Imaging the 1997 Marseille fire
I. Piccolini, O. Arino, A. Arnaud & J-M. Rosaz

On 26 July 1997, a large fire spread in the hills close to Marseilles. A collection of satellite images were processed in order to demonstrate the capabilities and the limitations of the current remote sensing instruments (and data) for fire detection, fire monitoring and damage assessment. The first results were presented at the Eurisy Conference in Morocco in September 1997 on Application of Space Techniques for Hazard Management in the Mediterranean.
The low-resolution instruments (AVHRR and ATSR) provided a limited detection and monitoring capability due to the acquisition frequency and to a lower extent, the spatial resolution. The repeat cycle of both ERS and NOAA satellites was a limiting factor. The high-resolution ERS SAR and Landsat TM instruments provided precise identification of the extent of the damage. The Marseille fire was used as a test case in order to analyse the precision of the ATSR radiometry sequence of the event in order to build on a burned-surface estimation algorithm.
Full text (939Kb)

Earth Observation Quarterly no. 59, June 1998 Special Edition: Wind field retrievals from SAR

Earth Observation Quarterly no. 59, June 1998

Special Edition: Wind field retrievals from SAR