The IMI120 is a deep-towed 120-kHz bathymetric sidescan sonar originally developed by Woods Hole Oceanographic Institute Deep Submergence Laboratory (DSL) as the DSL-120, subsequently rebuilt as the DSL-120A with an HMRG-engineered mapping sonar, and jointly operated by WHOI and HMRG for several years(2000-2006.)

Full operational responsibility was transferred to HMRG in 2006. Now named IMI120 (IMI: “Seeker” in Hawaiian) and operated by HMRG, this instrument provides high-resolution sidescan and phase-difference bathymetry for detailed analysis of seafloor features and is towed 50-100 m above the seafloor.

At a frequency of 120 kHz, the IMI120 system provides much higher resolution data than the shallow-towed MR1; the transducer arrays are much smaller, and the towbody is lighter and more easily deployed. IMI120 is towed 50-100 m above the seafloor in depths up to 6000m. When deployed from UNOLs vessels, winch and wire are coordinated through the National Science Foundation’s winch and wire pool; on other deployments acquisition of the winch and wire are coordinated and negotiated with the customer. For deployments to 6000 m, up to 10 km of 0.681 electro-fiberoptic wire is required. Fifteen IMI120 datasets have been collected between 2000 and 2011 and these data are currently archived on RAID disk arrays at HMRG.


IMI120 : Specifications

Towing Depth To 6000 m; 50-100 m above seafloor
Range 50-800 m
Towfish Dimensions (LxWXH) 3.3 m x 0.7 m x 1.1 m
Towfish Weight 390 kg in air (~800 lb)
Depressor Weight 400 kg (~900 lbs)
Towfish Operating Temperature 0-40 deg. C.
Wire Specification 0.681 electro-fiberoptic wire; 10,000 m

Electrical Specifications

Computers Digital-audio data stream to Linux computers
Lab Equipment: Voltage 115 VAC, +/- 10% with UPS
Lab Equipment: Electrical Frequency 47-63 Hz.
Lab Equipment: Current ~15 A normal load; 38 A full load
Operating Temperature & Humidity 0-40 deg. C; 5-80% humidity

Bathymetric Sidescan Sonar

Transducer Frequencies 120 kHz
# of Transducers 6 rows
Transducer Source Level 218 dB
Horizontal/Vertical Beamwidths 1.7 deg x 50 deg.
Pulse Length Selectable up to 350 microseconds
Receiver Gain Not selectable

Standard and Optional Sensors

Heading IXSEA Phins (export controlled sensor)
Attitude IXSEA Phins (export controlled sensor)
Compass IXSEA Phins (export controlled sensor)
Depth Paro Scientific Digiquartz depth sensor
CTD (optional) SeaBird 37SI
Magnetometer (optional) Honeywell HMR2300 Digital
Sub-bottom Profiler (optional) HMRG-designed subbottom profiler
Attitude and heading for Sub-bottom PNI Corp. TCM2
Sound velocity (optional) Optional


IMI120 is a portable side-scanning seafloor imaging system that simultaneously acquires digital bathymetry and sidescan sonar imagery (swath width up to 1000 m). The system’s sonar transducers are housed in a 3.3-m-long vehicle that is towed 50-100 m from the seafloor at ship speeds of 1-3 knots.  A 1000-kg depressor weight is towed about 40 m in front of the towfish, mostly decoupling it from ship motion. Due to this multi-body towing configuration, the IMI120 towfish is extremely stable.   As a result, IMI120 has successfully operated in rough sea conditions (up to sea state 5 ) that typically cause performance degradation in hull-mounted systems due to bubble masking and violent ship motion. In addition to the 6 sidescan transducer arrays, the IMI120 towfish can be configured to house a compass, a CTD, 2 DC/DC converters, a digital signal processor, a data acquisition computer, and (optionally) a magnetometer.  The depressor weight can optionally also house instrumentation, including a compass, a sound velocity sensor, a digital signal processor, a DC/DC converter, a data acquisition module and a subbottom profiler.

Electronic Configuration

The IMI120 towfish uses 6 separate transducer arrays on its port and starboard sides (1 transmit and 2 receive arrays on each side). Each array contains two rows of elements spaced one-half wavelength apart, and the transmit arrays are driven by high-power amplifiers that are programmable over a range of frequencies and pulse lengths. This high power capability results in a high signal-to-noise ratio, which improves the quality of bathymetric measurements and widens the swath of seafloor over which sidescan data are collected. IMI120 uses an advanced network of digital signal processors on the towfish and in the shipboard data acquisition electronics to control the operation of the sonars, and to process acoustic data.  Transmit state, repetition rate, and pulse length of the systems are operator selectable. Other significant features include digitization at the receiver, multiple channels of quadrature-detected data, and a high-bandwidth optical fiber link to the surface.

Bathymetric Capabilities

IMI120 determines bathymetry by using the phase difference of the signals from dual arrays to measure the angle of returning echoes on either side of the towfish. Flexible data processing software allows power, ping rate and beam spacing to vary in order to maximize the swath width and data quality with changing seafloor depth and acoustic properties. IMI120 accounts for ray bending due to acoustic velocity in the water by applying an empirical transfer function based on data collected during a calibration test at the beginning of each survey.

Auxiliary Sensors

Additional sensors can be mounted on either the towfish or the depressor weight, depending upon project needs.  A Honeywell HNR2300 Digital magnetometer is often deployed on the towfish body itself.  A subbottom profiler, designed and built by HMRG engineers, can be deployed on the depressor weight.  This subbottom profiler is designed to use similar electronics as the IMI30 towfish and is programmable for frequency and pulse lengths; a frequency of 4 kHz is the usual configuration for this sensor.  This subbottom profiler also uses a dedicated heading and attitude sensor that is also located on the depressor weight.

Launch and Recovery System

The IMI120 system is designed for portability and ease of use on different ships. For data acquisition, the towfish is picked up by a crane or A-frame and lowered into the water.  It is attached to a 40 m-long umbilical cable that in turn is attached to a 1000-kg depressor weight. The weight is required because the towfish is slightly positively buoyant (so that the towfish will return to the surface if the tow cable breaks), and because the resulting two-body towing configuration decouples the towfish from the heave of the ship. The depressor weight is attached to the ship by a cable that is attached to a winch mounted on the deck of the ship.  In addition to mechanically supporting the towfish, the tow cable carries power down to the towfish and data up to the ship.

Sidescan Sonar Systems


Harold Edgerton, a professor of electrical engineering at the Massachusetts Institute of Technology, developed sidescan sonar technology for use in the civilian community in the early 1960’s. Prior to developing sidescan, Edgerton was known for his work with photographic systems and “stopping time” via pictures. He applied a similar approach to mapping the seafloor using “flashes” of sound to create a narrow “image” of the underlying terrain and then stacking these images in a continuous, long “picture” as the sonar was towed along by a ship.  Many of the names that we now associate with underwater exploration and mapping are people who trained with Edgerton during the early days of sidescan development; Marty Klein and Ed Curley were students of Edgerton who would later go on to found Klein Associates Inc. and EPC where you can purchase sidescan sonars and other ocean instrumentation.  Another Edgerton student was Sam Raymond, founder of Benthos, which markets transponders and deep-ocean camera systems.

How Sidescan Sonars Work

Sidescan Sonar System in 1965Sidescan sonars are among the simplest seafloor mapping tools that you can use. In their towed configuration they consist of a subsurface unit called a towfish, a cable for towing and transmitting information, and a topside system that is used to process data and transmit commands. Sidescan sonars that work on autonomous underwater vehicles (AUV’s) don’t require the cable; instead commands are pre-programmed into the AUV’s onboard computer guidance system.  In the AUV case, a topside computer is still useful for downloading data and evaluating the performance of the sonar while at sea.

The sidescan sonar towfish typically includes the sonar transducers, electronics and their pressure housings and flotation. Over the decades as the technology for creating flotation and electronics has improved and become more compact, sidescan towfish have evolved from cumbersome systems (Figure 1) into lightweight vehicles that can be launched from small platforms with limited handling capabilities (Figure 2).

Moden Sidescan Sonar Instrument

Sidescan sonars work by projecting a narrow beam of sound to either side of a towfish and recording the strength and time of the returned echo. The strength of the echo depends on a number of parameters including the operating frequency of the sonar and the distance to the seafloor, the angle of incidence of the sound, the slope of the seafloor being ensonified (or insonified), and the type of substrate.

Schematic of sidescan sonar data collection and interpretationFigure 3 (left)- Schematic diagram of a sidescan sonar towed instrument ensonifying the seafloor (top) and the sidescan data record created (bottom). The intensity of sound reflected back from rocks, sediment and other features provides information on the distribution and characteristics of the seafloor morphology. Strong reflections (high backscatter) from boulders, gravel and positive topographic features facing the instrument are white and weak reflections (low backscatter) from finer sediments and shadows are in black.

To produce the fan-shaped beam of sound shown in yellow in Figure 3, sidescan arrays are long in the along-track direction and narrow in the across-track direction (see Figure 2).  In general, the longer the array, the narrower the along-track angular width of the sonar fan-shaped beam. Array length also varies as a function of operational frequency; lower frequency systems need to have longer arrays than higher frequency systems.  Here’s an example of a row of transducer elements from the University of Hawaii’s (UH’s) IMI-30 sonar:
IMI30 Transducers

Figure 4 – A row of six transducer elements from UH’s IMI-30 (30 kHz) sonar system.  For scale, the elements are 3.3cm long and 1.8cm wide.  The transducers  are mounted in a brass housing.

The sound waves projected by sidescan sonars and the echoes that the sonar receives are waves with a wavelength is directly related to the operational frequency of the system through the equation:

f * λ =  c


f is the operating frequency (cycles/second),
λ is the wavelength (typically centimeters),
and c is the speed of sound in water (~1500 m/sec)

For some typical systems:

Operating frequency
Example Systems
12 kHz
Seabeam, EM-120, MR1
12.5 cm
30 kHz
EM-300, IMI-30
5.0 cm
100 kHz
Klein, EM-1002 (95 kHz)
1.5 cm


Each time a pulse of sound or “ping” is transmitted, a sidescan sonar measures the time it takes the sound wave to travel to the seafloor and be echoed back to the sonar as well as the amplitude (or strength) of the echo.  Figure 5 shows an end-on schematic of the transducers in Figure 4 with an incoming echo.  Schematics of the type in Figure 5, showing a uniform incoming waveform, are fairly common.   In reality the echo amplitude is highly variable; the parameter that doesn’t change is the wavelength of the sound wave:

Sideview of a transducer array

Figure 5(left) – View of the transducer array in Figure 4 from the end with an idealized incoming echo.


Figure 6 – – Variable amplitude waveform (black) superimposed on the idealized waveform (gray) being received at the transducer in Figure 5.  Sidescan sonars measure variations in wave amplitude as a function of time. 

An example of one ping of data from a sidescan sonar is shown in Figure 7.

One ping of MR1 Data

Figure 7 – One ping of data collected by a sidescan sonar system.  The strength of the echoed sound wave is plotted along the y-axis and time in seconds is plotted along the x-axis.  The blue waveform is a component of the actual values (voltages) that are measured by the transducers arrays.  The components are combined to produce the discrete magnitudes (red crosses).  These data were collected by a shallow-towed 12 kHz sonar system located almost 5000 meters above the seafloor.  Because it takes sound more than six seconds to travel from the sonar and back again at a speed of ~1500 m/sec, there is no amplitude recorded in the early part (left-hand side) of the figure.  After ~6.5 seconds, the first returned echo, called the bottom detect, is finally received by the system.  The strength of the measured echo generally decreases for an additional eight seconds until it is time to stop recording data and transmit another ping.


The terms “sidescan,” “backscatter” and “reflectivity” are often used interchangeably, and incorrectly, when describing data collected by sidescan sonar. For the record, these are the correct definitions of these terms:

Sidescan is the correct term for the instrument that is used to collect the data.
Backscatter is the measure of the amplitude of a reflected acoustic signal bouncing off the seafloor (or riverbed, etc.).  Backscatter results from the small (centimeter scale) reflective surfaces on the seafloor.
Reflectivity refers to the broader scale reflective properties of the seabed (i.e., the side of a hill, the flat region directly under a sonar system).

Examples of reflection

The strength of any given “target” depends on both its reflective properties and the extent to which it contributes to the backscattered signal. It is assumed that the surface reflects incoming sound waves equally in all directions:

Figure 8(left) – Ideal diffuse reflection at a surface (left) versus diffuse reflection of a surface with a directional component (right).

The first echoed return is usually assumed to come from directly under the sonar vehicle, and this assumption is probably correct 80-90% of the time.  However, in regions having significant topographic relief (mid-ocean ridges, continental slopes, fracture zones) this can be inaccurate.  In Figure 7 the strength of the echo diminishes rapidly at first, which results from the decreasing strength of the reflectivity (the specular or mirror-like return) from underneath the vehicle).  This trend continues until the time reaches ~13 seconds, when there is a second peak in the data.  This corresponds to the first multiple return; that is, the point when sound waves making two roundtrips from the vehicle to the seafloor interferes with echoes completing one round trip from vehicle to seafloor but at a high angle to the vertical.
By juxtaposing successive sidescan pings, it is possible to make an image of seafloor textures in response to sound waves emanating from a particular sonar system (Figure 11). It is important to remember that the measured amplitude will vary as a function of wavelength; towed at the same altitude above the seafloor a 30 kHz sidescan sonar and a 12 kHz sidescan sonar will record different amplitudes. One of the most common mistakes made by novices to sidescan sonar data interpretation is to assume that the magnitude of the return echo has a geological meaning; that is, that magnitudes in one range correspond to lava flows and magnitudes in another range correspond to coral reefs. There has been significant effort to quantify the data collected by sidescan sonars and systematically characterize the seafloor surface, but to date these experiments have not been successful. Nevertheless, the resulting imagery allows scientists to investigate the processes that contribute to seafloor creation and modification.

MR1 Sidescan Sonar Data

Figure 11 -Swath of MR1 sidescan data showing round volcanic cones and tectonic lineations. The labels on the left side of the figure hard and easy parts of the sidescan data swath to interpret.


Cruise List

2007/2011 Munitions Surveys in Hawaii

Near the end of the second World War large quantities of obsolete, excess and captured munitions, including chemical weapons, were dumped offshore throughout the world’s oceans.  In 2007, the Department of Defense (DoD), through its National Defense Center for Energy and Environment, subcontracted the University of Hawaii and a local environmental consulting firm (Environet, Inc.) to conduct the Hawai‘i Undersea Military Munitions Assessment (HUMMA).  HUMMA’s objective was to bound and characterize a historic deep-water munitions sea disposal site south of Pearl Harbor to determine the potential impact of the ocean environment on sea disposed munitions and of sea disposed munitions on the ocean environment and those that use it.  HUMMA was conceived as the most comprehensive deep-water investigation in the United States to look at both chemical and conventional munitions.  Of particular interest for the HUMMA project was historical information indicating that 16,000 M47A2 100-pound (lb) mustard-filled munitions may have been sea disposed south of Pearl Harbor following World War II.  To detect and assess DMM, HUMMA participants developed an approach that used innovative technologies to map and sample small targets on the seafloor.

Shipwreck surrounded by multiple trails of munitions.

In selecting the system to use for the towed mapping effort, emphasis was placed on balancing the trade-off between resolution and the amount of seafloor that could be mapped based on usable swath width. The  IMI120 was ultimately selected because its resolving capabilities would detect 1-2 m objects on the seafloor from average altitudes of 75 m, allowing 2.7 km2 to be mapped per hour. Previous experience searching for wrecks with HMRG’s towed systems revealed that the angle of incidence of sound waves directly affects what a SONAR system can image; therefore, the region around the HUMMA site was surveyed with overlapping tracks in both east-west and north-south directions.

Within the HUMMA Pearl Harbor Study Area, the IMI120 was towed 50-75 m above the seafloor producing a backscatter swath width of 400 m on both the port and starboard sides of the towfish. In the area southwest of Barber’s Point, because the terrain was more undulating, the towfish was towed at altitudes of 75-100 m.  The collected IMI120 sidescan sonar data provided the basis for all subsequent investigations. The anomalous small, reflective targets observed in curvilinear trails in the IMI120 backscatter data had the correct size and distribution to make them likely munitions that were cast overboard according to the procedures described in the historical reports from the end of WWII.  The final report, figures, and photos detail the IMI120 and other UH systems that were used on the 2007 HUMMA project.  The IMI120 was again used for a follow-on HUMMA SONAR Survey in 2011.

2008 Lau Basin IMI120 & Multibeam Data

In 2008 Dr. Fernando Martinez returned to the Western Pacific aboard the R/V Kilo Moana (KM0804) to continue mapping in the Lau Basin that was begun in 2004 during voyage KM0410.  During KM0804 mapping was conducted using both the IMI120 sidescan sonar and the Kilo Moana’s deep-water, 12 kHz-multibeam EM120 sonar.  Data from both KM0410 and KM0804 have been submitted to and can be downloaded from the Marine Geophysical Data System;  Images from KM0804 include data from both the IMI120 and the EM120 multibeam and are presented here with permission from Dr. Martinez.

This image shows a portion of the ELSC1 segment along the Eastern Lau Spreading Center in the central Lau basin, SW Pacific. In this region, where the off-axis crust is relatively flat, recent volcanism is largely contained by small axis-bounding faults, creating a sharp boundary along the edge of the neo-volcanic zone. Small near-axis faults and fissures can be resolved in the data. The data were gridded at .00002 degrees (~2 m), the red line represents the spreading axis, and the red star represents an identified hydrothermal vent site (Kilo Moana).

KM0804 Multibeam Bathymetry and Imagery from Lau

This image shows a portion of the VFR1 segment along Valu Fa Ridge in the southern Lau basin, SW Pacific. The sidescan data shows abrupt transitions between different types of volcanism, with low backscatter volcaniclastic material draping the flanks of the ridge toward the north and south, with an abrupt transition to high backscatter sheet and lobate flows toward the center of the image where multiple 3rd order segments overlap. The data were gridded at .00002 degrees (~2m), the red line represents the spreading axis, and the red star represents an identified hydrothermal vent site (Tahi Moana 6).

High resolution IMI120 Bathymetric Data from two Areas in the Lau Basin

EM120 Multibeam Images