Hampton, Monty A., Torresan, Michael E., and Barber, Jr., John H., 1997, Sea-floor geology of a part of Mamala Bay, Hawaii: Pacific Science, v. 51, n. 1, p. 54-75. Reproduced by permission of the University of Hawaii Press.
We conducted three cruises in Mamala Bay aboard the University of Hawaii research vessel Kila. In February 1993, we used a sidescan sonar imaging system and a 3.5-kHz profiler to survey the sea floor. In May of 1994 we used a box corer to collect sediment samples and a video/still-photo system to observe small-scale features of the sea floor, restricted to the vicinity of the dredged-material deposits, as shown in Figure 1. In June 1995 we collected more box-core samples. A camera was attached to the box corer in 1995, so we obtained photographs of the sea floor at the sampling sites.
The transmitted pulse of the sidescan sonar was at 59 kHz, and 8968 samples per scan, digitized at 6 bits, were recorded on optical disk. The images were displayed real time on a graphic recorder with 16 levels of gray tone. We employed a 1-km swath width along tracklines, which were spaced at 800 m (Figure 1). The advertised spatial resolution of the system is 1/800 of the swath width, which equates to about 1.3 m for our survey. Following collection, the sidescan data were processed to remove the water column and to make radiometric (shading, destriping and debanding, speckel removal, and nadir tonal improvement) and geometric (slant to ground range and aspect ratio) corrections (Chavez, 1986), and the data set was mosaicked by computer (Figure 2).
3.5-kHz acoustic-reflection records and 12-kHz depth soundings also were collected along the survey tracklines. The 3.5-kHz data were collected at a 0.25-, 0.5-, or 1.0-sec pulse repetition rate. They were recorded on optical disk and displayed as profiles on a 16-bit format color monitor and on an ink-jet color printer. The 12-kHz system was run at 1-sec rate, and depths (assuming an acoustic velocity of seawater of 1500 m/sec) were stored with the digitized sidescan data and separately on optical disk.
The bottom camera system consisted of a 35-mm underwater still camera loaded with 50-ft rolls of film (approximately 400 frames per roll) and a silicon intensified (SIT) black-and-white video camera with a real-time video link to the surface. Video images were recorded on 8-mm tape. Video was collected along the entire extent of the tracklines in Figure 1, whereas still photographs were taken about every 1.5 min as long as film was available. Unfortunately, there was no length scale in the camera system.
We used a box corer to collect 110 sediment samples (Figure 1). The cross section of the box was 20 by 30 cm, and the height was 45 cm. The longest sediment core was 33 cm.
Ship navigation employed a GPS system in either an autonomous or differential mode, with nominal accuracies of about 100 m and 1-3 m, respectively. We did not navigate the sidescan towfish or the video-camera sled. Therefore, in order to achieve correct geographic registration of the mosaic we had to locate features on the 3.5-kHz profiles (collected at the known position of the ship) that also could be located on the mosaic, then shift the mosaic to its proper geographic position.