BUOY DEPLOYMENT FROM 

 AIRCRAFT 



Ice Patrol has deployed satellite- 

 tracked buoys from HC-130's 

 since 1979. The buoy is strapped 

 into an air-deployment package 

 and launched out the rear door of 

 an HC-130 flying at an altitude of 

 500 feet (150 m) at 150 knots 

 (77 m/s). The air-deployment 

 package consists of a wooden 

 pallet and a parachute, both of 

 which separate from the buoy 

 after it enters the water. The 

 parachute riser is cut by a cable- 

 cutter that is activated by a 

 battery that energizes when 

 immersed in salt water. The 

 pallet separates when salt tablets 

 dissolve and release straps 

 holding the buoy to the pallet. 

 The buoy then floats free and the 

 drogue falls free and unfurls. 



Nine buoys were air-deployed in 

 1987. Of these, three pallets 

 (4528. 4553, and 4559) failed 

 upon entry into the HC-130"s 

 airstream. When this occurs, the 

 buoy and drogue usually survive 

 intact, but frequently the wires to 

 the parachute cutter break. This 

 means that the parachute re- 

 mains attached to the buoy hull 

 and can act as a near-surface 

 drogue. Aerial inspections and 

 shipboard recoveries of buoys 

 have shown that the parachutes 

 collapse and become entangled 

 with the buoy hull or the upper 

 part of the drogue tether. It is not 

 likely that these failures contami- 

 nated the drift data significantly. 



BUOY DEPLOYMENT 

 STRATEGY 



It is not possible to obtain ade- 

 quate temporal and spatial 

 coverage of the Ice Patrol opera- 

 tions area (40-52N, 39-57W) over 

 a 5 or 6-month period with a few 

 (< 12 ) buoys. As a result, the 

 buoy deployment strategy fo- 

 cuses on the current that is the 

 major conduit of icebergs into the 

 North Atlantic shipping lanes, the 

 southward-flowing off-shore 

 branch of the Labrador Current. 

 The goal is to monitor this current 

 for the entire ice season by 

 keeping one or two buoys in it at 

 all times. With two exceptions 

 (451 1 and 4536), all of the 1987 

 buoys were deployed in the 

 Labrador Current. The two 

 exceptions were deployed in and 

 near a warm-core eddy that was 

 affecting the flow of the Labrador 

 Current near 44°N. No buoys 

 entered, nor were any deployed, 

 in the inshore branch of the Lab- 

 rador Current. Previous attempts 

 at deployments in this near-shore 

 region resulted in short drift tracks 

 because the buoys became 

 entangled in fishing gear and 

 were recovered by fishermen. 



DATA PROCESSING 



l^/lost of Ice Patrol's buoy position 

 data fall within the standard 

 location accuracy (LeTran and 

 Liabet, 1 987) provided by Service 

 ARGOS. The data are reported 

 to 0.001 ° of latitude and longi- 

 tude, which far exceeds this 

 standard location accuracy. For 

 46°N. the center latitude of the 



Ice Patrol operations area, the 

 positions are accurate to 0.003° 

 of latitude and 0.005° of longi- 

 tude. The raw position data are 

 unevenly-spaced in time, with 

 virtually no data from the period 

 from OOZ to 004Z each day. This 

 null period is due to the orbits of 

 the NCAA satellites. Approxi- 

 mately 10 fixes are determined 

 each day for each of the buoys. 



Although the data are relatively 

 noise free, all records are 

 scanned before processing to 

 ensure quality control. First, 

 duplicate positions and positions 

 with time separations of 30 

 minutes or less are deleted. 

 Then, positions < 700 m from 

 adjacent positions are deleted, 

 unless the deletion results in a 

 time separation of 4 or more 

 hours. 



The error-free position data are 

 then fitted to a cubic spline curve 

 to arrive at an evenly-spaced 

 record with an interval of 3 hours. 

 This process results in a slight 

 reduction in the number of fixes 

 per day (from 10 to 8). Next, the 

 position records are filtered using 

 a low-pass cosine filter with a cut- 

 off of 1.16 X 10-5 Hz (one cycle 

 per day). This filter removes 

 most tidal and inertial effects. 

 Finally, the buoy drift speeds are 

 calculated at three-hour intervals 

 using a two-point backward 

 differencing scheme. 



Most of the trajectory plots 

 presented in this report are from 

 the filtered records. Also pre- 

 sented for each buoy is a plot of 



71 



