David G. Long
Department of Electrical & Computer Engineering
Brigham Young University
Provo, Utah USA

I. Introduction

Satellite observation systems are crucial to the monitoring of the motion of the atmosphere and oceans. In particular, oceanic winds play a key role in driving the oceans and in modulating fluxes between the air and sea. Past observational efforts, using conventionally obtained data, have been severely hampered by the lack of accurate wind measurements with high resolution, global coverage, and frequent sampling. The Seasat Scatterometer, flown for 3 months in 1978, first demonstrated the capability of scatterometers to obtain the needed measurements of near-surface vector winds from space. Since then five other scatterometers have flown. In addition to their primary goal of measuring winds, wind scatterometers have proven remarkably useful for other applications in land, ice, and vegetation (Long et al., 2001). Their proven ability to measure vector winds fromspace at higher resolution than other sensnors ensures that they will continue to play a crucial role in current and future Earth remote sensing systems.

II. Principles of Scatterometry

A good review of wind scatterometry can be found in Naderi et al. (1991). Designed for wind observation, a wind scatterometer does not directly measure the wind. Instead, it measures the radar backscatter of the ocean’s surface which is related to the wind via a geophysical model function. To measure the radar backscatter, the instrument transmits a pulse of RF energy and measures the backscattered power. From knowledge of the parameters of the radar equation, the normalized radar cross-section (denoted so) of the ocean’s surface can be computed during ground processing. The near surface wind vector is then estimated from the so measurements using a geophysical model function. Due to the nature of the geophysical model function (which exhibits a bi-harmonic dependence on the wind direction), multiple co-located measurements of so from different azimuth angles are required to determine the wind vector at the ocean’s surface. When retrieving (estimating) the wind from the so measurements, several possible wind directions may occur. A second step, known as "ambiguity removal,” is used to determine a unique wind vector.

The backscatter measurements can also be used to map the extent and motion of sea ice, track icebergs, monitor snow melt and accumulation, gobal rain measurement (at Ku-band), and track global change. Note that unlike other spaceborne radar sensors, scatterometers observe the same location at multiple azimuth angles. Required for ocean wind retrieval, this capability is being exploited my several investigators to better understand the geophysics of the earth’s surfaces in key regions such as the cryosphere, deserts and tropical vegetation.

III. Scatterometer Systems

The first operational wind scatterometer (SASS) was flown aboard Seasat in 1978 and operated at Ku-band. The European Space Agency (ESA) orbited two C-band scatterometers aboard Earth Remote Sensing System (ERS) 1 and 2 who collected global data from 1992-2001. The Ku-band NASA Scatterometer (NSCAT) flew aboard the Japanese Advanced Earth Observing System (ADEOS-I), operating for 9 months from 1996-97 until a power failure prematurely ended the mission. Two identical Ku-band SeaWinds scatterometers have flown on QuikSCAT (1999-present) and ADEOS-II (9 months in 2003). ESA will launch the C-band ASCAT scatterometer in 2005. Other C- and Ku-band scatterometers are being planned.

IV. Technology Issues

SASS, ERS-1/2, NSCAT and ASCAT use fan-beam antennas. The SeaWinds represents the first of a new class of scatterometers based on rotating pencil beams. To address the crucial need for continuous multiple decade scatterometer datasets and to provide for more frequent coverage, additional scatterometer systems are needed. Unfortunately, traditional fan- beam scatterometer designs such as SASS, ERS-1/2, and NSCAT require significant power and mass and have antenna systems which are difficult to accommodate aboard spacecraft. New designs for smaller, lighter, and less-costly scatterometers are required. One example is a single-beam scanning scatterometer flown on a dedicated small satellite such as the system proposed by Long (1993). A key area of innovation is system design leading to more flyable designs. A "flyable" scatterometer is one which minimizes costs to the point they can be afforded by NOAA and other funding agencies; be relatively easily accommodated on (or integral to) a spacecraft; be built, integrated, and tested on a schedule commensurate with potential flight opportunities; and be sufficiently capable so that its data are scientifically and operationally useful. The latter can be specified in terms of coverage and wind measurement accuracy for specific applications. For global observation, coverage of 90% of the Earth’s oceans in two days or less is required, while wind measurements accuracy must be +/- 2 m/s and 20 deg rms. Timeliness of the data transmission to the ground (within 6 hours for future systems) is also a requirement. Note, however, that all past and present scatterometers have been designed exclusively for wind observation — other applications have been serundipidious. However, given the success scatterometers have demonstrated in non-wind applications, they are now considered multi-mission sensors and future designs will need to consider requirements for non-wind applications. Non-wind applications are interested in high spatial resolution, frequent sampling, dual-polarization capability and multiple incidence angles.

The relative maturity of scatterometry is a crucial advantage in the effort to develop new design concepts: the issues associated with wind retrieval from the scatterometer measurements and the techniques for predicting system performance are quite well understood. Nevertheless, to achieve a small, low-cost design, technical innovations in system design, possible distribued designs, antennas, radar electronics and orbital sampling are needed, with particular emphasis on how these innovations lead to improve resolution and tempor-spatial sampling to support both wind and non-wind applications.

A particular advantage of scatterometry is its provide high resolution capability for measuring ocean winds. Since higher resolution is required in many science applications, particularly in near-coastal areas in severe weather events, scatterometers will continue to be crucial. Current systems provide nominally 25 km resolution for wind measurements, though by using post-processing much higher resolution winds have been experimentally demonstrated for SeaWinds (Long and Luke, 2003). By including such post-processing capability in the system design, substantially improved performance is possible at minimum cost differential.


F. Naderi, M. H. Freilich, and D. G. Long, "Spaceborne Radar Measurement of Wind Velocity Over the Ocean–An Overview of the NSCAT Scatterometer System", Proceedings of the IEEE, pp. 850-866, Vol. 79, No. 6, June 1991.

D.G. Long and J.B. Luke, "High Resolution Wind Retrieval for SeaWinds," in Proceedings of SPIE Vol. 5155 Ocean Remote Sensing and Imaging II, ed. R.J. Frouin, G.D. Filbert, D. Pan (SPIE, Bellingham, WA), pp. 216-225, 2003.

D.G. Long, M.R. Drinkwater, B. Holt, S. Saatchi, and C. Bertoia, "Global Ice and Land Climate Studies Using Scatterometer Image Data," EOS, Transaction of the American Geophysical Union, Vol. 82, No. 43, pg. 503, 23 Oct. 2001.

D. G. Long, "ASCAT: A Light-Weight, Low-Cost Scatterometer," in Microwave Instrumentation for Remote Sensing of the Earth, James. C. Shiue, ed., Proc. SPIE 1935, pp. 28-38, Orlando, Florida, April 13-14, 1993.