04/06/2020 — admin
- The Okeanos Explorer received a Kongsberg EM304 multibeam echosounder topside upgrade over the 2019-20 winter maintenance period. No changes were made to the TX or RX arrays (installed in 2007 and 2018, respectively) or the Applanix POS MV positioning and attitude system.
- Sea acceptance testing (SAT) was conducted during EX2000 (March 3-8, 2020) with Kongsberg, NOAA, and OMAO personnel on board to complete the following procedures:
- Review EM304 and POS MV system geometry
- Configure the EM304 with the most recent pre-upgrade, post-calibration EM302 installation parameters
- Complete POS MV GAMS calibrations to verify antenna baseline
- Calibrate the EM304 for residual angular offsets attributable to the motion sensor installation angles
- Assess EM304 accuracy in several modes at 1000 m and 3000 m reference sites
- Evaluate swath coverage achieved by the system over a wide range of depths
- Examine RX noise levels perceived by the EM304 over a wide range of speeds
- Assess pre-calibration backscatter data quality
- Troubleshoot EM304 performance and software functionality with Kongsberg engineering support
- The calibration was completed at a site used previously (EX1802, EX1902) and indicated zero necessary adjustments, suggesting no appreciable changes to the TX and RX array, IMU, or GNSS antenna geometry since EX1902; these results generally confirm successful integration of the EM304 topside unit.
- Proxies for sounding density were compared using 2019 (EM302) and 2020 (EM304) speed-adjusted total sounding counts at the roll calibration site. Under comparable parameters, the EM304 reported ~63% more soundings over the same area.
- Swath coverage achieved during EX2000 generally met expectations, in light of the elevated sea state and the reduction in maximum swath angles from 75°/75°with the EM302 to 70°/70°with the EM304. Additional testing is planned for EX2001.
- Noise levels perceived by the EM304 receiver over a range of speeds showed significantly lower levels than previous tests at speeds of 0-7 kts. While noise levels increase as expected from 8-10 kts(based on previous testing), the results confirm no significant changes in vessel-borne machinery and flow noise following the winter maintenance period. Furthermore, the results suggest a drastic improvement in EM304 receiver noise levels compared to the EM302 hardware.
- The following items relating to EM304 swath accuracy performance were evaluated using QPS software and are currently under investigation with Kongsberg and QPS at the time of writing. These items and examples remain in the report as a record of the EM304 SAT observations at sea (as processed through the conventional, established workflow on board). Results should be considered preliminary as recent tests by Kongsberg in Norway suggest some complications may stem from processing software with evolving data formats rather than from the EM304 directly. EX2001 will provide an opportunity for additional testing with future software releases to continue with this line of troubleshooting.
- EM304 accuracy was tested in Deep (CW) and Deeper (CW/FM) modes at the 1000 m site as well as Very Deep (FM) mode at the 3000 m site; testing of Deeper (CW/FM) mode at the 3000 m site was cancelled due to deteriorating sea state. The QPS Qimera Cross Check Tool was used to evaluate these data, which showed nearly zero bias in mean depth across the swath in all modes. In Qimera, soundings in the outermost FM transmit sectors showed significant differences from neighboring soundings associated with inner (predominantly CW) sectors. These differences varied within pings (e.g., one outer sector with a shoal bias and the opposite outer sector with a deep bias) and varied from ping to ping (e.g., changing directions over time). This behavior differs from the wider distribution of FM sector soundings observed in other systems and in previous EX accuracy data. Investigations at sea with Kongsberg engineering support were inconclusive. This will require additional attention from Kongsberg and QPS to achieve outer swath performance similar on par with the EM302.
- Compared to previous experience with the EM302, the EM304 seems to report a much higher occurrence of outlier soundings on the outermost beams. This was especially apparent in deeper water where FM outer sectors are used, suggesting a possible link to the FM sector behavior mentioned above and detailed in this report. This issue was managed during EX2000 by manually reducing swath angles by several degrees. For routine mapping, the EM304 is expected to automatically adjust swath angles based on sounding quality and exclude outlier soundings as effectively as the EM302.
- A rudimentary assessment of backscatter data quality was performed using the 1000 m reference surface survey lines collected in Deep (CW) mode. The data indicated fairly consistent amplitudes across the swath and generally matched expectations for the flat, benign seafloor at this site, suggesting no significant hardware issues in this mode. Backscatter calibration/normalization data collection is planned for all modes during EX2001.
- A test survey was completed over a known shipwreck (172 m hull length) in ~2300 m of water in the vicinity of the deep roll verification site. Review of the raw bathymetry and backscatter data do not indicate the presence of the shipwreck. This site was previously surveyed with the EM302 and produced the same result. The result does not indicate any deficiency of the EM304, but rather that this feature is too small for detection in deep water with such a surface vessel-mounted multibeam echosounder.
- A number of SIS v5 functionality and data distribution challenges were identified during EX2000. These are included as action items in this report and will require follow-up during future missions to support established workflows on board. Some modifications may be required in the short-term, such as running Sound Speed Manager on the SIS machine to work around current data distribution limitations.