P. Racette and D.M. Le Vine
NASA Goddard Space Flight Center
Passive microwave remote sensing offers the potential for measuring many parameters (soil moisture, sea surface temperature, precipitation, etc.) important for understanding and monitoring the environment. Remote sensing at frequencies in the microwave spectrum has the advantage that it can be done at night and in the presence of cloud cover, permitting measurements in regions inaccessible to visible and infrared sensors. For example, cloud cover in the Bering Straits may persist for weeks at a time, but a microwave radiometer with adequate resolution would be able to provide year-round mapping of sea ice concentration and sea ice/water boundary. Frequencies at the lower end of the microwave spectrum respond to the changes in the dielectric constant of the surface. This means a strong response to the presence of water in soils and vegetation, a response to the temperature and salinity of the ocean surface, and response to changes of state (e.g. frozen/thawed). At higher microwave frequencies resonance’s of oxygen and water in the atmosphere permit profiles of temperature, pressure and humidity to be measured.
Measurements from space offer the potential for global-scale observations necessary for understanding weather, climate and the global environment. However, microwave measurements from space have been limited by the large aperture antennas required to obtain reasonable spatial resolution. For example, the 10 km resolution at L-band (1.4 GHz) from an orbit of 800 km desired by hydrologists for the measurement of soil moisture would require an antenna in space of about 20 m x 20 m. For practical applications with periodic global coverage it would also be necessary to scan about 145 degrees with this aperture.
Aperture synthesis is a new technology that helps to overcome some of the limitations of size, weight, and scanning associated with real aperture antennas. Microwave imaging with fine spatial resolution is possible from space using aperture synthesis without the need to scan a large aperture.
II. Aperture Synthesis
Aperture synthesis is an interferometric technique in which the complex correlation of the output voltage from pairs of antennas is measured at different antenna spacings or baselines. The correlation measurement is proportional to the spatial Fourier transform of the intensity of a distant scene at a frequency that depends upon the antenna spacing. Each baseline measurement produces a sample point in the two-dimensional Fourier transform of the scene. By making measurements at many different spacings and selectively distributing the antenna elements so as to obtain optimum sampling in the Fourier domain, a set of Fourier samples suitable for inverting the transform may be obtained. High resolution maps of the source may be retrieved using a set of relatively small antennas without the need for scanning the antenna aperture. As in a conventional antenna array, resolution is determined by the maximum spacing (baseline) and the minimum spacing determines the location of grating lobes. However, in contrast to conventional arrays, each spacing needs to appear only once, and no mechanical scanning is necessary (it is done in software as part of the image reconstruction).
Aperture synthesis was first applied in radio astronomy as a means to achieve high resolving power with an antenna array using a limited number of (relatively) small, individual elements. More recently synthetic aperture radiometers have been developed for remote sensing of the earth. The first such instrument, the L-band Electronically Scanned Thinned Array Radiometer (ESTAR), was developed by NASA’s Goddard Space Flight Center and the University of Massachusetts. The objective of this research was to demonstrate the utility of aperture synthesis for remote sensing of the earth with specific application to the remote sensing of soil moisture and ocean salinity (two important observations made at L-band).
III. Technical Issues
The advantages gained from aperture synthesis comes at the expense of reduced sensitivity resulting from the corresponding reduction in physical aperture. Sensitivity is an especially critical issue for measurements made from low earth orbit because the high velocity of the platform (about 7 km/s) limits the integration time available for imaging a particular scene. The theoretical sensitivity of a synthetic aperture radiometer is given by dT = (A/na)*Tsys/Bt**0.5
where Tsys is the sum of the system and scene noise temperatures and Bt is the time bandwidth product, A is the effective area of a real aperture with the same spatial resolution as obtained with the synthesized antenna, a is the area of the individual real aperture antennas employed in the array, and n is the number of antennas in the array. The term, Tsys/Bt**0.5, is just the conventional formula for sensitivity for a real aperture (total power) radiometer. Since in a practical application na is less than A , the sensitivity of the synthesized beam will be poorer than a real aperture antenna; however, the synthetic aperture radiometer receives energy from all pixels in the field-of-view and as a result, its integration time can be larger than that of a comparable real aperture scanning radiometer.
There are many possible ways to implement aperture synthesis and each configuration must be evaluated for its performance (i.e. sensitivity, number of receivers, coverage in the Fourier space, etc). For example, ESTAR is a hybrid which employs real antennas (long stick arrays) to obtain resolution in the direction of motion (along-track) and uses aperture synthesis to obtain resolution cross-track. One could obtain equivalent resolution using aperture synthesis in two-dimensions, for example with an array of antenna elements along the arms of a cross (+), a tee (T) or (Y). It is also possible in some applications to have the antennas arranged around the circumference of a circle (e.g. a hoola hoop). Such arrays have been studied with application to profiling of atmospheric temperature at 50-60 GHz.
The experience with ESTAR has demonstrated the potential of aperture synthesis for remote sensing of soil moisture from space. However, the measurement of soil moisture requires moderate sensitivity. Applications which require greater sensitivity (e.g. ocean salinity measurement requires (DT0.02 K) or synthesis in two dimensions (necessary from geostationary orbit) will require research to improve the image reconstruction algorithm and calibration methods. Although, accomplished successfully with ESTAR at the level needed to measure soil moisture from an aircraft platform, calibration for applications which require greater sensitivity and autonomous calibration in space require additional work. The transfer of data between the individual antennas and central processor (interconnect problem) is another area requiring work. This includes both the transfer of data back from the individual antennas and the delivery of reference signal (e.g. the LO) to the individual antennas. Finally, advances in correlator technology are needed to reduce the power requirements of the processing unit.
IV. Current and Future Applications
ESTAR is unique in that it was the first radiometer built to test the concept of aperture synthesis for microwave remote sensing of the earth, and because it is a hybrid real-and-synthetic aperture combination. ESTAR has been successfully demonstrated in hydrology experiments at USDA research watersheds at Walnut Gulch, AZ (1991) and the Little Washita River Watershed in OK (1992). ESTAR continues to support hydrology research and studies at NASA’s GSFC are underway to implement an ESTAR-style instrument in space to provide measurement of soil moisture to complement NASA’s suite of EOS observations.
The potential for practical, high resolution microwave measurements from space has raised interest at other frequencies also. For example, studies are underway of a high resolution (1 km) instrument at 18 and 37 GHz to monitor thin ice and open water (leads) in the Arctic to support shipping along the Northern Sea Route. An instrument of the ESTAR-type (hybrid real-and-synthetic aperture) is being built at 37 GHz for the Department of Defense (Navy and Air Force) by Quadrant Engineering in Massachusetts. Research on aperture synthesis in two dimensions has also received much attention recently. In particular, ESA/ESTEC is studying the potential of aperture synthesis in two dimensions for remote sensing of soil moisture. An aircraft prototype is nearly complete (at L-band and using a Y-configuration) and studies are underway to define an instrument for monitoring soil moisture from space. Several laboratory instruments have also been developed for research on aperture synthesis in two dimensions. Instruments have been assembled in Denmark (TUD, N. Skou at 10 GHz), in Germany (DLR, Peichl and Seuss at 37 GHz), and in the U.S.A. at the Goddard Space Flight Center (12 GHz) and TRW (44 GHz, Pearlman and Davidhowser). Plans exist in France at CNES to build an instrument at C-band (6 GHz). A two dimensional instrument with a somewhat different configuration has been built in Japan (ETL, K. Komiyama at 10 GHz) and a related concept originally developed at Hughes (C. Wiley and Edelsohn, RADSAR) is continuing to receive attention.
Aperture synthesis is a new technique for microwave wave remote sensing of the environment. This technology could lead to a new generation of space-borne passive microwave sensors by helping to overcome limitations set by antenna aperture size. The advantage of aperture synthesis is that it can achieve spatial resolutions equivalent to a total power radiometer with a large effective collecting area using relatively small antennas. The reduction in sensitivity that this entails can be restored because the synthetic aperture system does not need to scan and collect energy from many independent antenna pair simultaneously.
"ESTAR: A Synthetic Aperture Microwave Radiometer for Remote Sensing Applications,” D.M. Le Vine, A. Griffis, C.T. Swift and T.J. Jackson, Proc. IEEE, Vol. 82 (#12), pp 1787-1801, December, 1994.
“The Sensitivity of Synthetic Aperture Radiometers for Remote Sensing of Applications from Space,” D.M. Le Vine, Radio Science, Volume 25, Number 4, pp 441-453, July 1990.
Interferometery and Synthesis in Radio Astronomy, A. Thompson, J. Moran and G. Swenson, J. Wiley and Sons, New York, 1986.
Proceedings IGARSS-94, Synthetic Aperture Radiometry for Earth Remote Sensing, Vol III, pp 1311-1331 (Lib Congress #: 93-80348; IEEE Cat #: 94CH3378-7).