The 192-nmi transit to Site 1098 (scientific prospectus site APSHE-13A) was accomplished at an average speed of 10.5 kt. During the voyage, a total of 52 icebergs appeared on the 24-nmi range of the radar, and the vessel had to alter course to avoid one large iceberg. Because the target was narrow, the site was approached slowly (2 kt), using the GPS to follow the south-to-north track of a site-survey deep-tow boomer profile that intersected the location of Core PD92-30 (Fig. F2). The final site location was chosen by comparison of basin-floor morphology with 3.5-kHz and narrow-beam echosounder records. Drilling operations began at 1530 hr on 12 March with the deployment of a beacon.

During operations at this site, Eugene Domack joined the vessel for less than 1 day from the Laurence M. Gould (LMG) to observe procedures. Domack was the lead proponent of the proposal for drilling the Palmer Deep sites. Shortly before the LMG arrived, the Polar Duke came alongside Crane 3 to off-load a logging tool and acetone, which had been shipped to Punta Arenas but failed to arrive in time at the port call. These supplies had been delivered to the American base at Palmer Station a few miles away by the LMG for pickup by the Polar Duke.

The initial core of Hole 1098A recovered 1.9 m and established seafloor depth at 1012 m (1022.6 meters below rig floor [mbrf]). Advanced hydraulic piston corer (APC) work then continued through Core 178-1098A-7H, where contact with glacial till or hard basement concluded coring in the hole at 45.9 meters below seafloor (mbsf) (Table T1). Calm conditions permitted very good core recovery. The drill string was raised above the seafloor and the ship offset 10 m north to begin Hole 1098B.

To achieve stratigraphic overlap, Hole 1098B was spudded with the pipe 3.5 m lower than for the first core of Hole 1098A. Core recovery indicated a seafloor depth ~1.4 m shallower than at Hole 1098A. APC coring advanced to 43.0 mbsf (Core 178-1098B-5H). Adara tool temperature measurements were obtained with Cores 178-1098B-4H (34.5 mbsf) and 5H (43.0 mbsf). The drill string was raised above the seafloor at 0355 hr and the vessel offset 10 m north to begin Hole 1098C.

The pipe was set 3 m lower than at Hole 1098B for the first APC core at Hole 1098C. Coring began at 0430 hr 13 March with an Adara tool deployment just above the bottom before spudding in the first core. Five APC cores were collected, and coring operations ended at 0630 hr. Core recovery for the site averaged 101.6%. The drill string was recovered, the beacon was successfully retrieved, and the vessel was under way at 0930 hr 13 March for a 6-nmi transit to Site 1099.

Site 1099 (scientific prospectus site APSHE-15A) was located using GPS coordinates, and a beacon was launched at 1030 hr 13 March. APC coring at Hole 1099A began at 1445 hr on 13 March with a 5.3-m core that established the seafloor depth at 1399.9 m (1411.2 mbrf). Continuous cores were taken to 62.3 mbsf (102% recovery [Table T1]) with Adara tool temperature measurements at Cores 178-1099A-5H (43.3 mbsf) and 7H (62.3 mbsf). An additional Adara tool was deployed with the initial core barrel before spudding the hole to measure the bottom-water temperature. A change of wind then moved several icebergs toward the vessel. The bit was pulled above the seafloor at 1900 hr, and the vessel was offset away from the icebergs.

At 2145 hr, after the icebergs had moved past, Hole 1099B was spudded and the bit washed down to 60 mbsf, where APC coring resumed. Core recovery averaged 102% while advancing to 107.5 mbsf, which was considered the depth objective for the hole (Table T1). The drill string was retrieved and the beacon recovered before the vessel departed at 0445 hr on 14 March. For a summary of drilling at Sites 1098 and 1099 in Palmer Deep, see Table T1,  in the "Leg Summary" chapter.