platforms, (5) more information on sea-ice microwave properties, (6) advances 

 in image-processing technology to speed the quantitative analysis of the data, 

 (7) simulation of operational use of SAR, and (8) sea-ice scientific research. 



The satellite called for has, in addition to a buoy monitoring system and 

 the required flight and data-link electronics, instrumentation in the form of 

 the SAR complemented by a scatterometer and/or a radiometer. In general, SAR 

 is an identification and location tool for a number of ice features such as 

 ridges, floes, and leads resulting in a data set from which ice motion and 

 deformation data can be extracted. The low-resolution scatterometer/radiometer 

 systems, on the other hand, measure distributed phenomena such as ice-type 

 fraction or amount of open water. The scatterometer/radiometer data will 

 therefore constitute a global ice extent and type data set. It will also have 

 time and space scales suitable to weather and climate research and to 

 operational forecasting applications in which local SAR data are used with a 

 variety of other types of basin-wide low-resolution data. Also, the 

 combination of a feature-identification tool (such as SAR) with a well- 

 calibrated, areally integrating tool (such as the scatterometer) will permit 

 more quantitative estimates of feature variables. All of these instruments 

 have flown in space aboard Seasat, and considerations are now underway by 

 several nations for future flights of similar instruments. 



If a SAR system were deployed in the absence of these complementary 

 instruments, the optimum radar frequency for discriminating between first-year 

 ice, multiyear ice, and water on radar backscatter alone would be between 11 

 and 15 GHz for incidence angles between 20° and 50°. At frequencies in the 

 range between 1 and 10 GHz, the differences in radar backscatter between 

 different ice types are less significant. However, if the SAR system used for 

 feature tracking is supplemented by a 19- or 37-GHz radiometer or a 11- to 15- 

 GHz scatterometer used for ice-type determination, the recommended SAR 

 wavelength would be at L-band (1-2 GHz) with like polarizatioiu At the L-band 

 frequency, first-year ice which has not undergone much deformation can easily 

 be distinguished from multiyear ice, and highly deformed first-year ice and 

 multiyear ice can usually be distinguished by shape and, possibly, by 

 geographical location. While the trend for improved ice feature recognition 

 in SAR data at higher frequencies is reasonably well established, the greatest 

 changes for program success call for the use of systems which are proven in 

 space, of known calibration, and produce familiar data. These systems are the 

 L-band SAR and the higher frequency scatterometer or radiometer. 



Other radar parameters can be approximately determined from summary 

 mission requirements. The depression angle should be in the range 20° to 

 50°. A resolution of 25 m appears adequate although some measurements would 

 tolerate a reduction to 100 m. The swath width required to obtain adequate 

 coverage needs should be 200 km to satisfy operational requirements and 

 somewhat less for many scientific programs. The orbit geometry should provide 

 maximum areal coverage for the supplemental sensors as well as maximum orbit 

 tracks over coastal waters in order for the radar imager to support the 

 operational research objectives. Thus, an orbit providing SAR ground coverage 

 poleward to 76° N in the form of long, nearly east-to-west transects across 

 the Arctic, and scatterometer/radiometer coverage to approximately 85° N for 

 science and for forecasting, is called for. If other satellites are deployed 



1-3 



