Satellite and Ship-based Wind-Field Measurements

Abstract
Using satellite- and ship-based radar systems together with methods developed at GKSS Research Center measurements of high-resolution wind fields are possible. Supplemental to anemometer point measurements spatial wind information are collected, which are of fundamental importance, e.g. in planning the best site for windmill park installations and their assessment.

Motivation
Wind is a major driving force in ocean dynamics; it is responsible for the transfer of energy and momentum from the atmosphere to the ocean and supports the gas-exchange processes between the lower marine atmospheric boundary layer and upper ocean surface. Wind creates the ocean surface waves and is a driving force of ocean currents. Thus wind is a key parameter in the coupled atmosphere, ocean and biosphere system. Wind is also an important energy source that has to be investigated.

Operational Wind Measurement
Measurements of ocean winds are performed using various different methods, e.g. in-situ by anemometers (point measurements through time), and by remote sensing with satellite-based systems, e.g. scatterometer (spatial measurements). In situ measurements are mainly collected by ships and buoys of which the first are affected by blockage effects and variable mast heights and the latter by tilt and displacement height, especially in high winds and sea states. The remote sensing techniques require excellent calibration as well as model functions that parameterize the dependence of the image brightness on the wind and have a rather coarse resolution (25 km). Mesoscale wind fields are often determined by meteorological models.

New Developed Wind Measurement Methods
At the Institute for Coastal Research different methods were developed for tower-, satellite-, and ship-based wind measurements at the ocean surface. Thereby common marine radar systems as well as space-borne radar systems are used. Marine radar systems are installed aboard every ship, whereas the space-borne systems are mounted aboard satellites like ERS, ENIVSAT, or RADARSAT. Using these measurement systems high-resolution images of the ocean surface are retrieved. These systems are independent of light conditions and can measure under most weather conditions.

Used Radar Systems
With a marine radar, here a Real Aperture Radar (RAR), polar radar-image sequences of the ocean surface are continuously recorded. At GKSS Research Center the so-called WaMoS-system has been developed.

Fig. 1 gives a schematic overview of WaMoS. The system consists of a standard marine radar and a personal computer equipped with an analogue to digital converter. This system can store and process the acquired radar image sequences. The marine radar system operates in X-Band (9.5 GHz). The radar antenna covers an area within a radius of about 2000 m at a resolution of about 12 m. Each image sequence consists of 32 images, representing a time span of about 1 minute. At GKSS Research Center radar-image sequences were investigated, recorded at the Norwegian platform Ekofisk 2/4 k located about 200 km off the west coast of Norway in the North Sea.

In Fig. 2 the platform, the location of the installed radar system and a typical radar-image sequence are shown. A wave field and bright and dark patches, which originate from structures and other platforms, are visible. The wind information is extracted from the temporal mean of an image sequence.
The satellites ERS, ENVISAT, and RADARSAT operate at a height of 800 km with a repeat cycle of 35 days for the same location. Aboard the satellites a Synthetic Aperture Radar (SAR) is installed that enables recording images on a continuous basis. It has large spatial coverage (up to 500 km swath width) and a very high resolution (up to 5 m).

Physical Background
A local wind field generates the small-scale roughness of the sea surface, which in turn raises the radar backscatter of the ocean surface and therefore the brightness of structures in a radar image. The radar backscatter of the sea surface is strongly dependent on the local wind speed and angle between the antenna viewing direction and wind direction. This dependency enables the deduction of the wind vector from radar images of the sea surface.

Radar Wind Field Measurement Methods
Two Methods for wind field measurement were developed at GKSS Research Center: WiRAR and WiSAR.
Both algorithms consist of two modules: In the first part, wind directions are determined by analyzing streak-like structures in the filtered radar images. These streaks are approximately in line with the mean surface wind direction. They are images at scales above 100 m. The methodology is based on retrieval of local image gradients and assumes the surface wind direction to be oriented normal to the local gradient.
In the second part, wind speeds are derived by a geophysical model function that has to be retrieved for each system. The model function relates the image brightness to the local near-surface wind speed, wind direction versus antenna look direction, and to the distance from the radar antenna. Additionally, the air-sea temperature difference affects the near-surface wind profile, which in turn influences the sea-surface roughness and therefore the image brightness. Therefore the air-sea temperature difference is also related to the image brightness. The model function is parameterized by training a Neural Network.

Results
In Fig. 3 a wind field is shown that has been determined using the GKSS-developed WiSAR program. For processing data from the SAR of the Canadian satellite RADARSAT-1 were used. The wind field shows the conditions during hurricane “Floyd” on September, 15th 1999 at 11 a.m. UTC. It has an extension of 500 km x 500 km with a spatial resolution of 200 m.

Fig. 4 shows an example of a high-resolution wind field of a marine radar-image sequence recorded at Ekofisk 2/4k on March, 27th 2001. The wind field has a spatial resolution of 120 m. The measured wind direction was 140°. The WiRAR-determined mean wind speed is 13 m/s.

Literature
Dankert, H., Horstmann, J., and Rosenthal, W., Ocean Wind Fields Retrieved from Radar-Image Sequences, Journal of Geophysical Research - Oceans, Vol. 108, No. C11, 3352, doi: 10.1029/2003JC002056, 2003, more ...
Horstmann J., H. Schiller, J. Schulz-Stellenfleth, and S. Lehner, Global Wind Retrieval from SAR, IEEE Trans. Geosci. Remote Sensing, Vol. 41(10), pp. 2277-2286, 2003. more ...

Contact address:
Dr. Heiko Dankert and Dr. Jochen Horstmann, GKSS-Research Centre, Geesthacht, Max-Planck-Strasse 1, 21502 Geesthacht, Germany