4. Waves over 1 meter in height might obscure 

 a dangerous growler even with the expert use of 

 anticlutter devices. If an ice target is not picked 

 up beyond the sea return, it will not be detected 

 at all. 



5. Ice is not frequency sensitive. The response 

 to S and X bands is the same. Furthermore, 

 there is practically no difference in the response 

 of sea water to S and X bands. 



These results remain valid today. This past 

 year, one merchant vessel collided with ice within 

 the Ice Patrol broadcast limits of all known ice 

 and incurred considerable damage and a small 

 coastal freighter was lost with 6 lives following 

 a collision with a berg in Hudson Strait. Often 

 Coast Guard surface patrol vessels lose ice targets 

 on radar while tracking them in close proximity. 



Ice Patrol's early airborne remote sensing ven- 

 tures consisted of aerial photography and con- 

 ventional airborne radar. The reconnaissance 

 aircraft usually flies standard tracks at an alti- 

 tude of 1,500 feet in good weather over areas of 

 probable icebergs. Radar data establish the ac- 

 tual position of berg sightings. "With the pre- 

 vailing low ceiling and minimum visibility, the 

 reconnaissance aircraft attempts to descend just 

 below the ceiling to a minimum of 400 to 500 

 feet. Diversions from assigned tracks are then 

 made, when possible, to attempt visual identifica- 

 tion of radar targets. The dangers inherent in 

 this type of operation are obvious. Often the 

 predominance of fog reduces reconnaissance 

 effectiveness to zero. The problem encompasses 

 not just target acquisition but that of classifying 

 all detected targets as icebergs or nonbergs. The 

 first attempt in this area was unsuccessful tests 

 with a bolometer (forerunner of the airborne 

 radiation thermometer) in 1954. Additional 

 ART tests were conducted in 1964 when, although 

 iceberg identification was not enhanced, basic 

 current structure was deducted from the sea 

 surface temperature on several occasions. An 

 AN/ARR-33 passive microwave radiometer was 

 flown from 1967 through 1969 with moderate 

 success. Some targets were positively identified 

 as bergs, although the fixed nadir window re- 

 quired overhead identification of all of the myriad 

 of targets. A precision radiation thermometer 

 (Barns PRT-5) maps sea surface temperature 

 for use in iceberg melt determinations and ap- 

 proximation of major current features. In es- 



sence, these are the only operational tools today 

 that supplement visual observations. Addition- 

 ally, application of NOAA 4 high resolution in- 

 frared and visual imagery to obtain these data 

 are being pursued. 



As early as 1957, Ice Patrol conducted experi- 

 ments with Side Looking Airborne Radar, be- 

 cause it was apparent that the high resolution 

 would provide near all-weather detection and 

 identification. This work, using an AN/APQ-55 

 (XA-1) K-band real aperture system, was lim- 

 ited to scope due to poor electronic reliability but 

 provided great hope for future systems. In 1969 

 Ice Patrol commenced experiments with a modi- 

 fied AN/DPD-2 Ku-band, real aperture SLAR 

 system to evaluate its capabilities. From data 

 obtained on regular patrol flights in 1970 and 

 1971, the Coast Guard Research and Development 

 Center formulated a system of target discrimina- 

 tion between icebergs and other objects through 

 interpretation and classification using analysis of 

 basic clues. Seven clues consisting of size, shape, 

 shadow, tone, texture pattern, edge and wake 

 were considered. A photographic interpreter 

 would analyze each target for convergence of 

 evidence in a "logical search" phase. This system 

 proved quite satisfactory and reasonably reliable 

 for post mission research analysis but was opera- 

 tionally constrained by the requirements to de- 

 velop film (a vacuum in flight developer was not 

 available), the extensive amount of imagery to be 

 viewed, and the near-laboratory conditions re- 

 quired for handling enlarged imagery which were 

 not available in the field. The * AN/DPD-2 

 SLAR was again flown during the 1972 and 1973 

 seasons, but its use was terminated due to con- 

 tinuing maintenance problems with this aged 

 system and the imagery handling problem. 



As a follow on to previous work in support of 

 the Great. Lakes Navigation Season Extension 

 Demonstration Program during the winter of 

 1975, NASA Lewis Research Center installed an 

 AX/APS-94C modified SLAR system in a Coast 

 Guard HC-130B aircraft. At the conclusion of 

 Great Lakes season, this system, with additional 

 modifications and the Moving Target Indicator 

 mode installed, was flown experimentally in sup- 

 port of the International Ice Patrol. An Ed<>- 

 Western dry film processor was installed in the 

 aircraft and data transmission modes were not 

 utilized. Twelve missions were flown with the 



57 



