Status of EM120 data collection and processing aboard
R/V Roger Revelle as of June 08, 2001.
I. BACKGROUND
Between December 20, 2000 and January 28, 2001, SIO substituted a Kongsberg Simrad Inc. (KSI) EM120
multibeam swath bathymetry sonar system for the SeaBeam 2112 sonar system originally installed aboard
R/V Roger Revelle by the shipbuilder. The approximately $2M project was funded about half from federal
sources (ONR DURIP award and a NSF grant) and half from SIO internal funds.
Sea trials were conducted offshore San Diego, CA from January 29 through February 2, 2001 and the system
was accepted with exceptions. Representatives from the manufacturer and invited observers from the
Universities of Hawaii and South Florida, as well as from the US Naval Oceanographic Office were aboard
during the sea trials.
The manufacturer conducted formal training in the operation of the EM120 sonar and processing of its data with
KSI's software tools during the ship's transit from San Diego, CA to Hilo, HI, Feb 17-25, 2001.
Eight SIO shipboard technicians benefited from this training, along with representatives from the US Naval
Oceanographic Office and the University of South Florida.
Seafloor mapping operations began Feb 26, 2001. Since then SIO and KSI have worked through a number of
operational questions and communications have been very open.
II. CHARACTERISTICS OF THE EM120 SYSTEM ON R/V ROGER REVELLE
Operating acoustic frequencies:
11.25 kHz to 12.6 kHz in up to nine 150 Hz steps excluding 12 kHz.
Modes of operation:
Very Deep Deep Medium Shallow Very Shallow
depth range 1-11km 0.8-11km 0.3-1.5 km 40-1000m 20-80m
swath angle +/-52 +/-75 +/-75 +/-75 +/-75
pulse length 15ms 15ms 5ms 2ms 2ms
TX*RCV
Beamwidths 1 x 2 deg 1 x 2 deg 1 x 2 deg 2 x 2 deg 4 x 4 deg
# of beams 191 191 191 191 135
Time delay beam forming on all beams:
Beam steering at transmit to compensate for yaw, pitch, and roll
Beam steering at receive compensated for roll
Split aperture correlation processing for bottom detection:
Envelope detection on beams with specular and near specular backscatter
Differential phase null detection for all other beams.
Real-time ray-tracing on all beams.
Expected depth accuracy: 0.2% of water depth on all beams.
Seafloor acoustic backscatter:
191 backscatter coefficients/ping (angular dependence)
Over 5000 samples of acoustic backscatter amplitude data/ping (sidescan imagery)
Data Formats:
KSI provides fully open and documented formats for all the data logged by the EM120 system.
External sensors:
TSS DMS05 vertical reference unit (0.05 deg roll & pitch accuracy) + GPS
Meridian Surveyor gyrocompass (0.1 deg heading accuracy) + GPS
FSI conductivity and temperature sensors at keel depth (accuracies better than 0.001 deg C
and 1 mS respectively) used to compute sound speed at the arrays to 0.1 m/s.
P-code GPS navigation (3m rms 2D positional accuracy)
Sound speed profiles in the water column derived from CTD casts (on station)
or Sippican MK12 XBT system with various probes (underway).
III. PRESENT STATUS
At this time, the system performs within its specifications except for beams beyond +/-55 deg from nadir that have
lower depth accuracies (>0.5% of water depth) than expected. This is most likely because of inaccurate synchronization
of the ship's roll data with the acoustic data. SIO is working closely with KSI and the VRU manufacturers on this
issue - KSI has not yet committed to a delivery date for fixes.
In sea state 6, heading into the seas at full speed, the ship pitches up and down by 6-7 deg and the EM120 data quality
is poor. Within its rated sea state envelope (5 or less) data quality remains good across the swath though users must
choose a realistic swath width that matches the environmental conditions (sea state, heading relative to weather,
bottom type, etc.). Simple trial and error over a few dozen pings is sufficient to establish such swath width limits.
In spite of its current accuracy shortcomings listed above, the system provides a broader swath width (140 deg to 5500 m),
higher data density (191 beams), and higher data accuracy (0.2 - 0.5% of water depth) than any deep water system in its
class installed aboard or towed behind UNOLS vessels.
IV. DATA PROCESSING ABOARD SHIP
At present, SIO has established and documented two data processing paths aboard ship. The first uses KSI's software
packages purchased with the sonar system: Neptune for bathymetry, and Poseidon for acoustic backscatter amplitude
imagery. Both software packages are reliable and straightforward to use. They output xyz grids that can be exported
to various GIS packages (including GMT). Raw bathymetry data output by the EM120 contains true depths, however
users have the option to save the depth values in corrected or uncorrected meters (referenced to a sound speed of 1500m/s)
during post-processing through Neptune.
It is important to note that due to licensing costs for Neptune and Poseidon, they are installed and are operational on a single
SUN Microsystems Ultra 10 workstation onboard R/V Revelle. Nonetheless, these software packages have proven to be
very valuable during sea trials and subsequent cruises while other open software packages are being fine tuned.
The second path uses the NSF-supported MBSystems and GMT software packages. SIO has been working closely with the
developers of MBSystems to remedy several errors resulting from data file formats and other glitches. Processing steps to
convert raw EM120 data to data editable in MBSystems have been fully documented by SIO and are available aboard ship.
Even though not all functions were operational with EM120 data, MBSystems was used aboard R/V Roger Revelle between
March and June, 2001 (SEAWEED 03-06, DRIFT 01) to plot contour maps of EM120 bathymetry and to display EM120
backscatter amplitude imagery. Users can choose to convert depths from corrected to uncorrected meters in MBSystems.
This specific function has been verified and works properly. Any user familiar with MBSystems will have no difficulty
processing data aboard ship once the software is fully functional and SIO will strive to maintain the most current version
of MBSystems and GMT on the shipboard computer server.
MBSystems software developers have been very responsive in fixing several of the problems that SIO has documented,
and they continue to work on a number of issues.
V. DATA PROCESSING ASHORE
The same MBSystems tools available aboard ship are available and freely accessible ashore.
VI. DATA SAMPLES
SIO is preparing a set of examples showing the performance of the EM120 system installed aboard R/V Roger Revelle in various
operating condtions (sea state, water depths, etc.), including comparison to other multibeam data sets collected on SIO ships over
the same areas. This information will be displayed on a website accessible to anyone with a computer on Internet. Completion
of this task is expected to take two months.