A field performance test of the Sippican deep aircraft‐deployed expendable bathythermograph

Abstract

A field test of the performance of the initial design, deep (760 m/2500 ft.) Sippican aircraft‐deployed expendable bathythermograph (AXBT) has been conducted. Thirty‐seven AXBTs were deployed beside a research vessel, which was conducting conductivity, temperature, and depth (CTD) casts to 1000 m. A total of five CTD casts were made. A comparison between the AXBT and CTD data showed that the AXBT's provided temperature profiles with accuracies that depended upon the particular formulae chosen for frequency‐to‐temperature conversion and fall rate. The three sets of formulae used here were the standard Navy formulae, the formulae published by the Sippican Corporation, and a combination of a frequency‐to‐temperature formula published by the Naval Oceanographic Office (NAVOCEANO) and a fall rate determined from data collected during this study. Two of the three methods produced data that appeared to be within the Navy's AXBT accuracy specifications, taking the CTD data as standard. A temperature offset of about 0.4°C was found in the data set using the Sippican formulae, and this puts the temperature accuracy just outside the Navy's specified limits for higher temperatures. The best comparison between AXBT and CTD data was found with the NAVOCEANO formula and our fall rate, which provide a temperature accuracy better than 0.25°C. Our fall rate was quite close to the standard Navy fall rate (less than 2 m difference at 760 m depth), so the use of the latter is suggested. The recommended formulae for the initial design Sippican 760 m AXBT's are as follows: T= −66.8857 + (7.0273 × 10−2)F ‐ (2.1807 × 10−5)F2 + (3.6311 × 10−9)F3, D = 1.52t, where T = temperature in degrees Celsius, F = frequency in hertz, D = depth in meters, and t = elapsed time in seconds. The statistics of the AXBT data set suggest that the precision of the Sippican thermistor is better than 0.06°C.