3.

SEAFLOOR MAPPING FIELD MANUAL

FOR MULTIBEAM SONAR

Vanessa Lucieer*, Kim Picard, Justy Siwabessy, Alan Jordan,

Maggie Tran, Jacquomo Monk

* [email protected]

Chapter citation:
Lucieer V, Picard K, Siwabessy J, Jordan A, Tran M, Monk J. 2018. Seafloor mapping
field manual for multibeam sonar. In Field Manuals for Marine Sampling to Monitor
Australian Waters, Przeslawski R, Foster S (Eds). National Environmental Science
Programme (NESP). pp 42-64.
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

3.1 Platform Description

Swath mapping systems use acoustic technology to collect data on the bathymetry (topography) and
the backscatter (impedance) of the seafloor (Figure 3.1). These systems can either be mounted on
a ship; autonomous underwater vehicle; remotely operated vehicle or a remote surface vehicle.
They work by transmitting a sound pulse, called a ping, through a transducer at a specific frequency
(or a range of frequencies simultaneously). This same ping is then recorded through a receiver
placed very close to the transducer. The elapsed time that the ping takes to reach the seafloor and
return to the receiver is used to measure the depth of the water. Certain attributes of the shape of
the sound-wave are used to infer characteristics about the seafloor (geomorphology). Typical
multibeam echo sounder (MBES) data products include bathymetry (seafloor depth) as well as
backscatter intensity, which can provide a metric for seafloor “hardness” and will indicate the
substrate type (Figure 3.1 a-d).

MBES have become one of the standard tools for geophysical surveying and mapping of the
seafloor and have been used for a variety of scientific, safety at sea (hydrographic and military
operations) and industrial applications. MBES can produce a spatially continuous acoustic image of
the surface of the seafloor by generating a “swath” or “fan” of continuous data points, increasing the
resolution of the resulting surfaces. This has revolutionized our ability to understand physical
processes occurring at the seafloor, and the composition and distribution of substrate, which has in
turn significantly improved our knowledge of seafloor ecosystems (McArthur et al. 2010, Lucieer and
Lamarche 2011, Porter-Smith et al. 2012). Mapping of bathymetric morphology will delineate
geological features that have relief (using the changes in seafloor depth information), however in
regions where the relief is smaller than the minimum mapping unit (resolution of the grid cell is
larger than the feature of interest) backscatter data can be used to assess the boundaries of the
geology or sediment structure.

Australia’s marine estate spans an incredible range of water depths; from the coast to 6000m+.
Water depth has a very large influence on the acoustic survey acquisition, as it will dictate the
resolution of the data (i.e., number of pings per unit area which will dictate the minimum pixel size)
and the efficiency for surveying using MBES acoustics (i.e., swath width). While practices for
employing the equipment have developed rapidly over the last few decades, there are a number of
specific and common issues that need to be considered and detailed in a national standard
operating field manual. This document has been developed in collaboration with Australia’s National
Multibeam Guideline written by the National Seabed Mapping Coordination working group which
includes over 40 representatives from government departments, scientific institutions, universities
and industry (see inset box).

During the development of this manual, a broader assessment of multibeam survey standards by a
national seabed mapping coordination working group was started. This program will provide
guidelines for national standards of acquisition on and off the shelf, and improved data
interoperability and access. This national working group guideline aims to be relevant for a wide
range of purposes such as hydrographic mapping, marine infrastructure installation and planning,
and baseline habitat mapping. It provides a more detailed description of the technical considerations
of acquisition and international surveying standards, including details of operational procedures.
Further details can be found in the National Multibeam Guideline to be available on
www.ausseabed.gov.au by mid-2018.

In order to avoid duplication of details, this field manual will provide a procedure for specific
planning, acquisition and processing steps relevant to marine monitoring. Where applicable, it will
Page | 43
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

refer to the National Multibeam Guideline for further details of operational steps. It will also provide
further specific details of pre- and post-surveying considerations required for marine monitoring
activities when planning swath mapping surveys. This will include surveys required for both broad
scale mapping to inform the development of habitat maps, and those being conducted as a
component of monitoring. Further details of marine sampling platforms used to ground truth
acoustic data, and to monitor of ecological indicators are presented in the accompanying NESP field
manuals (Chapters 4-9).

Figure 3.1. a) Multibeam transducer mounted on the hull of a ship in the (b) gondola. c) Multibeam acoustic bathymetry
image c) coincident backscatter image and d) interpreted geomorphology map. (reference: Watson et al., 2017).

3.2 Scope

This manual refers to the use of multibeam or interferometric echosounders (referred herein as just
multibeam or MBES) to conduct surveys of seafloor bathymetry and backscatter that can be used to
derive maps of geomorphic features and habitats. It does not mandate use of a specific multibeam
acoustic system (either an interferometric3 or beamforming4 multibeam). The examples given herein
refer to Kongsberg systems merely as an exemplar of the procedure to be conducted. Similarities
can be drawn from these examples to any particular MBES system being employed on the survey.

There are a number of multibeam echosounders that have been commonly used for surveying in
Australian waters that would be suited to marine monitoring activities. It is important that the
surveyor be mindful that there are differences in the way bathymetric measurements are made from
both interferometric and beamforming multibeam echosounders, and these influence the scale and
resolution of features being detected and the fidelity of the acoustic measurements (which is
important for monitoring). The main difference is namely due to beam formers measuring range for
each of a set of angles, and interferometers measuring angle for each of a set of ranges (Table 5).
We have outlined the standard methods that are relevant to any of these systems to provide a
framework to create a nationally consistent multibeam data archive for Australian marine and
coastal waters.

Page | 44
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

This field manual details the specific planning, acquisition, processing and reporting considerations
that are required to meet the seafloor surveying needs of monitoring programs. Although MBES can
be used for water column data collection, these data are outside the scope of this standard
operating procedural document in its current version (Version 0.1). This document provides
guidance to organisations responsible for permitting and supporting research programs (e.g. Parks
Australia) to collect multibeam data for monitoring programs (e.g. government research agencies,
universities) to ensure consistency in acquisition and processing of multibeam acoustic data. This
will increase the chance that data and spatial data products from different organisations and swath
systems can be combined and reused into the future and become a valuable data asset for national
research objectives; ongoing monitoring and planning. This manual is subset by four main phases
of a seabed mapping survey as outlined in Figure 3.2:

1. Data acquisition;
2. Data processing;
3. Benthic classification (data interpretation),and
4. Accuracy assessment and reporting (including metadata management of spatial data
products).

Figure 3.2. Workflow from MBES survey design to spatial data products and reporting

Page | 45
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Table 3.1. Comparison of bathymetric systems (reference: Bathyswath.com)

Parameter / Function Interferometric Multibeam Beamforming Multibeam Notes

Number of depth 6000+ 60-120 Depends on range

measurements

Range vs. water depth 10 - 20 3-5 Beam former footprint becomes

unacceptably large at far range.

Amplification / processing 4-5 60 + In a harsh environment, simplicity is

channels important

Outboard transducer Passive Active The outboard component of an

electronics interferometer is extremely robust, and
cheaper to replace if damage does occur

Outboard transducer size and 350x160x60mm 120x190x450mm Dimensions for a common portable beam
weight former. Many beam formers are much
5 kg (air) 16 kg (air) larger.

Horizontal resolution at range Good Poor Beam former footprint becomes

unacceptably large at far range.

Angular coverage 260°(including 20° overlap) 90°- 180° (or beyond 180°
using a dual head system)

Co-incident sidescan True Partial An interferometer collects amplitude in the

Page | 46
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

same way as its bathymetry: as a time-

series.

Profile data density Increases with reducing grazing Decreases with reducing Higher complete profile data confidence
angle grazing angle with an interferometer.

Ability to resolve several No Yes

targets at the same range

Ability to resolve several Yes No

targets at the same angle

Profile data density Increases with reducing grazing Decreases with reducing In the first 5 m of horizontal range, a beam
angle grazing angle former collects slightly more depth
samples. Beyond that, an interferometer
collects many more.

Capacity to acquire water Yes Yes Interferometric systems can identify

column information? targets in the water column but are unable
to characterise them accurately due to
lack of beam forming angles to locate the
target.

Page | 47
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

3.3 Multibeam Acoustics for Marine Monitoring

The use of MBES for mapping and monitoring marine habitats has experienced a rapid increase
since 2000 (Figure 3.3), and there is now a wealth of knowledge from which we can synthesise a
‘best practices’ document.

Figure 3.3. Annual total of peer reviewed papers featuring multibeam mapping for seafloor survey (Web of Science 2017-
search words “multibeam seafloor habitat survey”).

The objectives of multibeam acoustic surveys conducted by mapping programs are to collect
seafloor data to identify, delineate and map biogenic, anthropogenic and geological features. This
objective requires particular data to be collected that can a) chart the water depths creating a high
resolution bathymetric map at an appropriate resolution in regards to the target habitat or feature
and b) be able to differentiate boundaries between different substrate and/or habitat types.

To meet these objectives, there are two particular needs for mapping and surveying that can be
defined as either baseline surveys or monitoring surveys (see Table 2). MBES can be used for both
survey types, however, they have different acquisition and post-processing standards. A baseline
survey is for exploratory purposes where data will be collected in a ‘single pass operation’. This
data is used to map the distribution of marine habitats at a particular spatial scale, and provide
information necessary for more targeted field surveys using such tools as towed video, AUVs and
stereo baited remote underwater video stations (BRUVs) (Lucieer et al. 2013, Monk et al. 2016). In
contrast, a monitoring survey may have already identified target habitats or features (such as
rocky outcrops) from previous broad scale or other hydrographic data that are to be monitored to
assess change in distribution and extent (Rattray et al. 2009, McGonigle et al. 2010). This type of
survey will require acoustic data to be collected at a higher resolution and with a greater degree of
positional accuracy. Mapping for baseline survey and monitoring surveys will be dealt with
separately throughout the manual, with their differences and the requirements that need to be
considered to meet the aims of each survey type outlined in Figure 3.4.

Page | 48
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Figure 3.4. Decision tree for seabed classification survey design (adapted from Anderson et al. (2007)).

Page | 49
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Table 3.2 Standard Operating Procedures identified according to survey purpose: Baseline or Monitoring

Specification NESP Baseline NESP Monitoring

 Used to ensure spatio-temporal assessment of the seabed
 Used to identify seafloor habitats and potential biodiversity
and habitat. The survey accuracy standard is very high to
hotspots.
Purpose ensure reproducibility over time.
 Used for discovery purposes in regions that have had no baseline
 Used for repeat mapping and for targeting key habitats for
mapping conducted.
monitoring purposes.
 In addition to baseline survey specifications:
 The coverage of the area to be surveyed (bounding box) with the
 Synthesis of all pre-existing survey data into survey region
datum and coordinate system clearly identified.
database
 Establishment of line spacing
Pre survey  Identification of locations of seafloor targets to be
 Determination of the system offsets and calibration area (patch
preparation monitored
test) area to be conducted as soon as practical and after system
 Establishment of line spacing with min of 60% overlap.
is completely set up ready for survey. The location and
scheduling of the Sound Velocity Profiles
Installation Offsets  Provide Mobilisation Calibration Reports and logs  Provide Mobilisation Calibration Reports and logs
 Bathy: Mandated  Bathy: Mandated
Data Logging  Seabed Backscatter: Mandated  Seabed Backscatter: Mandated
 Water column backscatter: Recommended (if available)  Water column backscatter: Mandated (if available)
 Mode: Equidistant mode where system allows
Acquisition setting  same
 Minimise setting changes to optimise backscatter
 Min of 2 per day (beginning and end of survey), but should
 Min of 1 per day, but should be monitored.
Sound Velocity be monitored.
 If sound speed at the transducer varies by > 2m/s another SVP
Profiles  If sound speed at the transducer varies by > 1m/s another
should be collected
SVP should be collected
 GDA2020. Horizontal accuracy: 5m + 5% of water depth. Vertical  GDA2020 -– Horizontal accuracy: absolute positioning to
Geodetic Parameters
accuracy: 1% water depth be at< 2 or less. Vertical accuracy: < 1m
 Recommended 6 knots ( or at survey speed appropriate to  Recommended 5 knots ( or at survey speed appropriate to
Survey Speed
capture resolution required) capture resolution required)
 100% Coverage with 30% overlap between survey lines of data  100% coverage with 60% overlap between survey lines of
Mapping Coverage
with an 80% confidence level. data with an 80% confidence level.
Resolution  1 m resolution in < 50m depth ; 5% of depth beyond 50 m  1 m resolution

Page | 50
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

 Record GPS tides. All soundings shall be reduced to the

Tides and GPS Tide  Record GPS tides. All soundings shall be reduced to the ellipsoid. ellipsoid.

 All data should be attributed with its uncertainty estimate

Point data  All data should be attributed with its uncertainty estimate at the
at the 95% confidence level for both position and, if
attribution 80% confidence level for both position and, if relevant, depth.
relevant, depth.
 Metadata report shall include heading and data
Metadata and  Metadata report shall include heading and data deliverables
deliverables outlined in this manual. Calibration Report
Reports outlined in this manual. Calibration Report and Report of Survey.
and Report of Survey.
 Australian Online Data Network (AODN) data portal.  Australian Online Data Network (AODN) data portal.
Archiving
 National MBES Data Centre  National MBES Data Centre
 Used to ensure spatio-temporal assessment of the seabed
 Used to identify seafloor habitats and potential biodiversity
and habitat. The standard is very high to ensure
hotspots.
Purpose reproducibility over time.
 Used for discovery purposes in regions that have had no
 Used for repeat mapping and for targeting key habitats for
baseline mapping conducted.
monitoring purposes.
 The coverage of the area to be surveyed (bounding box) with the  In addition to baseline survey specifications:
datum and coordinate system clearly identified.  Synthesis of all pre-existing survey data into survey region
 Establishment of line spacing file
Pre survey
 Determination of the system calibration (patch test) area to be  Identification of seafloor targets for monitoring
preparation
conducted as soon as practical and after system is completely set  Establishment of line spacing with min of 60% overlap.
up ready for survey. The location and scheduling of the Sound
Velocity Profiles
Installation Offsets  Provide Mobilisation Calibration Reports and logs  Provide Mobilisation Calibration Reports and logs
 Bathy: Mandated  Bathy: Mandated
Data Logging  Seabed Backscatter: Mandated  Seabed Backscatter: Mandated
 Water column backscatter: Recommended  Water column backscatter: Mandated
 Min of 2 per day (beginning and end of survey), but should
 Min of 1 per day, but should be monitored.
Sound Velocity be monitored.
 If sound speed at the transducer varies by > 2m/s another SVP
Profiles  If sound speed at the transducer varies by > 2m/s another
should be collected
SVP should be collected
 84ITRF with epoch reference -– Horizontal accuracy:
 84ITRF with epoch reference. Horizontal accuracy: 5m + 5% of
Geodetic Parameters absolute positioning to be at< 21 m or less. Vertical
water depth. Vertical accuracy: 1% water depth
accuracy: < 1m
Survey Speed  6 knots  5 knots

Page | 51
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Mapping Coverage  100% Coverage with 30% overlap between survey lines  100% coverage with 100% overlap between survey lines
 1 m resolution (may require AUV or towed body in water
Resolution  1 m resolution in < 50m depth ; 5% of depth beyond 50 m
depths >200 m)
 Record GPS tides. All soundings shall be reduced to the
Tides and GPS Tide  Record GPS tides. All soundings shall be reduced to the ellipsoid. ellipsoid.

 All data should be attributed with its uncertainty estimate

 All data should be attributed with its uncertainty estimate at the
Point data attribution at the 95% confidence level for both position and, if
80% confidence level for both position and, if relevant, depth.
relevant, depth.
 Metadata report shall include heading and data deliverables  Metadata report shall include heading and data
Metadata and
outlined in this manual. Calibration Report and Report of deliverables outlined in this manual. Calibration Report
Reports
Survey. and Report of Survey.
 Australian Oecan Data Network (AODN) data portal.  Australian Ocean Data Network (AODN) data portal.
Archiving
 National Data Centre  National Data Centre

Page | 52
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

3.4 Pre-Survey Preparations

There are a number of important factors that first need to be considered in order to ensure relevant
areas are surveyed, time and costs can be accurately estimated, appropriate vessels and
acquisition gear is used, and previous survey data is considered.

Firstly, it is important that all spatial data for the survey region must be sourced to gain a preliminary
understanding of the seabed as this will influence several survey considerations. This information
can be used to create a survey plan, which would include a summary of the following components:

 The coverage of the area to be surveyed (bounding box) with the datum and coordinate system
clearly identified,
 Planned survey lines (direction and acquisition order of the survey lines),
 System calibration survey lines (patch test),
 Seafloor topography (features of interest) and slope, and
 The location and frequency of the Sound Velocity Profile (SVP).

Background spatial data might include electronic nautical charts from the Australian Hydrographic
Office (AHO), aerial photos (in shallow water regions), LiDAR or satellite derived bathymetry data.
It may also include previous maps of seabed habitats generated from single beam acoustic surveys
or maps of sediment distribution from broad scale seafloor grab or dredge surveys. Information on
seabed habitats can also be collected by analysing the distribution of other activities conducted
within the survey region (for example, ancillary research such as fisheries surveys may be an
indicator of habitat type).

The survey plan will aim to establish the range of water depths and seafloor complexity across the
survey region. The range of water depths will define how many survey lines need to be conducted
to ensure sufficient overlap between the acoustic swaths to ensure 100 % seafloor coverage (refer
to Chapter 2 where coverage may relate to selected sampling sites). Where the water depth is
relatively constant (such as on the outer continental shelf), the survey plan may provide adequate
structure for accurate planning. In shallower waters, where the depths may change rapidly (or are
unknown to the resolution of national satellite derived products) a comprehensive plan of survey
lines may not be useful, as they will need to be modified as the bathymetric data is collected. In this
case, a defined survey area boundary (polygon) with an initial survey line for calibration may be
sufficient.

An essential component of the survey-planning phase is the need to obtain the relevant permits that
may apply for sediment data collection which is common for MBES data validation, especially when
conducted within marine parks. See Appendix B for a list of potential permits needed.

Following the establishment of the survey plan the logistical preparations for data acquisition can be
conducted. These are outlined in the following sections and recorded in the vessel or field logbook
over the duration of the survey and made available in the final reporting documentation.

Page | 53
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

3.5 Data Acquisition

3.5.1 Installation offsets

The spatial relationship between all of the sensors in an MBES system (GPS, transducer, motion
reference unit etc.) with the vessel’s frame of reference is paramount to obtaining high resolution
and accurate data. The vessel’s Central Reference Point (CRP) is determined upon each new
installation of a MBES system and to best suit the vessels balance, and installation criteria (if the
MBES is hull or pole mounted) (Edward and Martin 2015). The CRP is defined, as an example,
within the Kongsberg operating systems Seapath software along with the installation offsets for the
Global Navigation Satellite System and Motion Reference Unit. The offset of the MBES transducer
is defined within the Kongsberg Seafloor Information System (SIS) software. Where possible the
CRP should be defined at the MBES transducer directly. All installation offsets are required to be
recorded and detailed within the survey report and processing log of the supplied raw data files.

To ensure that the depth charted is the true depth and not depth under keel, the vessel draft must
be taken into consideration during data acquisition (likely at the start and end of a survey) to
account for changes in draft due to foe example (fuel usage) although this will depend on the model
of the vessel used for survey. Although this manual recommends that depths be provided in relation
to the ellipsoid, to enable other users to reduce data to chart datum, vessel draft should be
measured at the start and end of surveys and dynamic draft taken into consideration with
measurements of the waterline conducted regularly. The vessel draft is recorded during a survey in
the vessel log and/or entered in the acquisition system. For further information see section 2.4 of
Australia’s Multibeam Guideline (Version 0.1).

3.5.2 Data logging

During a survey with a MBES system there are a number of data products that should be recorded.
These include:

1. Raw data: Always log raw proprietary format for all type of data (multibeam echosounder or
ancillary systems). Raw positional data and motion datagrams are to be recorded at a rate of
1Hz and 100Hz respectively. These datagrams are logged to the raw sonar file.

2. Raw sonar data are recorded in the native/proprietary format of the multibeam system used
(e.g. *.all for Kongsberg, *.s7k for Reson) and the ancillary data. Log complete backscatter
i.e. beam intensity (RI and snippets or equivalent. Files are recorded for a duration of 30
minutes for shallow systems (< 150 m) and 120 minutes for deep systems (> 150 m) to
account for computer processing speed.

3. Water column data [Recommended, if available]: Water column datagrams are logged to a
separate file in proprietary format (e.g. *.wcd for Kongsberg). These files can take up a large
amount of storage space (~ 10 times raw bathymetry), and the surveyor must ensure
necessary disc space prior to collection.

4. File naming convention: It is important that the surveyor adhere to a consistent and
acceptable naming convention that links to the metadata of the raw data format. Raw sonar
files in proprietary format recorded by an acquisition software (e.g. SIS for Kongsberg) have
the following naming convention.
[Nnnn_yyyymmdd_hhmmss_Vesselname_system.Extension] [Survey line, year, month, day,
hour, minute, second, vessel name, multibeam acoustic system. proprietary format
extension (e.g.all for Kongsberg and s7k for Reson]. Survey lines are organised by Julian

Page | 54
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

day within the processing software but are acquired sequentially as line IDs throughout the
survey. Where two separate systems are being operated at the same time on the same
vessel, they can be distinguished by the system name of the MBES in the file name. If a new
survey is to be created within the acquisition software during a survey the line number
should be reset to the last number used +1.

5. Filters and settings: Both noise and spike filters should be monitored during the survey to
ensure the data quality and integrity is maintained over the course of the survey. Beam
spacing mode should be set to equidistant. It is very important that the pulse length should
not be changed at any time during the survey so that all data are standardised.

6. Delayed heave: delayed heave datagrams are recorded by the acquisition computer and
logged to files in the proprietary format.

3.5.3 Sound velocity profiles

A sound velocity profile (SVP) measures the speed of sound in water at different vertical levels in
the water column and this data can be used to accurately form the beam of the sound. Some
multibeam systems have a SVP sensor built onto the head of the transducer but for others that do
not, it is important that a SVP sensor be deployed to collect this information.

Why are sound velocity profiles so important?

A MBES system emits a sound pulse in an arc out from the transducer. As the sounds contacts the
seabed, it is reflected/backscattered back towards the transducer and received. Each backscattered
pulse from each individual seabed point can be considered a discrete beam. The speed of the beam
through the water column is governed by the water temperature and density. Because the water
column, in most cases, is not evenly mixed, the speed of the pulse changes at different levels in the
water column. At each change in speed, refraction or ‘bending of the pulse path’ occurs, unless the
angle of incidence is equal to 90 degrees, as with a single beam echosounder. Refraction can
happen many times throughout the pulse’s path through the water column. Therefore, to enable
best ray tracing possible and consequently depth conversion of each soundings, details of the water
column sound profile are essential. Depending on the location of the survey and the conditions over
the area (sea state, mixing regime, thermal layering etc.) of the survey, sound velocity profiles
(SVP) should be conducted at the appropriate intervals or location. The SVP can be determined
using one of the following four methods:

 Direct observation via deployment of a SVP measuring device (e.g. Valeport monitor)
 Calculation of SVP through deployment of an eXpendable Bathy Thermograph (XBT)
 Calculation of SVP using CTD (Conductivity/Temperature/Depth) data and applying the
UNESCO formula
(https://www.usna.edu/Users/physics/ejtuchol/documents/SP411/Chapter4.pdf) or;
 Calculation of SVP from Sea Surface Temperature and Climatology using SVP builder software
(Sinquin et al. 2016).

How are SVPs applied to multibeam surveys?

A sound velocity profile (SVP) must be taken within the survey area at least once at the beginning of
the survey and once at the end for monitoring surveys. In some areas, multiple SVPs should be
taken. For example, profiles will vary due to freshwater inflows from rivers or currents from areas
with different salinity e.g. proximity to an estuary. Surface sound speed variation may also be
strongly affected also by solar warming. If variations can be expected, where and when the SVPs
are to be taken must be carefully planned, and the survey line schedule adjusted to consider this.
Page | 55
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Sound speed data is used in the following ways:

 correction for the fact that the transducer staves are the wrong spacing in wavelengths;
 correcting for the change in total sound path length because of the speed of sound variation,
but ignoring refraction and;
 correcting for both refraction and sound path length.

Path length correction uses the speed of sound to determine the sonar path length from the time the
ping is transmitted to the time it is received. The average speed of sound within the speed of sound
profile is used for this, measured from the depth of the transducers to the depth of the seabed.

3.5.4 Geodetic parameters

The datum parameters entered into the acquisition software will use the Global Navigation Satellite
System (GNSS) datum for example WGS84 (Table 3.2). Any datum shifts will need to be applied at
the post processing stage. The use of differential GPS as a positioning system is required for all on-
shelf and off-shelf multibeam acoustic data collection as we aim to resolve an absolute positional
accuracy greater than 1 m. All positioning data should be provided as track plots (in x, y format) to
enable interpretation of the vessel transits.

3.5.5 Survey speed

The speed of the vessel will have a direct impact on the density of soundings reaching the seafloor,
the quality of the data (in the return signal) and to some degree determine the resolution of the final
raster datasets (as it dictates the distance along track between pings). Depending on the type of
vessel employed for the surveying, the survey speed must be kept constant and between 5-6 knots
(11 – 14 km/h). The distance between pings along the track of the vessel is determined by the
pulse repetition frequency (PRF) and ship speed; the faster the vessel, the fewer pulses ensonify
the seafloor per distance along track. Aeration problems (bubble sweep) is a function of sea state
but also of the heading with respect to the wave direction and the vessel speed. Aeration problems
reduce the signal or the quality of the signal at the transducer head.

It is strongly advised that the surveyor creates a record of aeration problems versus sea state with
respect to heading and vessel speed. This record will be helpful in ensuring that the survey is
performed efficiently with a minimum of line rejections and corresponding reruns and infills. This
should be recorded in the field log book.

3.5.6 Line spacing

Line spacing is the distance between adjacent survey lines. The best spacing between survey lines
is determined by a combination of horizontal range limit (sonar coverage from one transducer)
expected at that depth of water and the accuracy required from the survey (either baseline mapping
or monitoring). The horizontal range expected depends on the water depth as well as the sea state,
seabed type and the sonar frequency. If the surveyor is using two transducer heads, the total swath
width from the port edge to the starboard edge is twice this range.

The horizontal range is limited by two factors: grazing angle and spreading loss. The grazing angle
limit is related to the angle that the sound “beam” makes with the seabed. At the grazing angle limit,
the sound makes a very small angle with the seabed. Most of the sound at this point is reflected
away and the signal scattered back from the seabed is too small to be detected.

Page | 56
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Due to this loss of signal on the outer beams of the swath, some overlap in swath is required. A
minimum of 30% overlap should take into account line keeping errors and where sea state is calm
and create a 100% coverage of the seafloor. The type of survey being undertaken will determine the
overlap with the highest quality requiring 100% overlap and the lowest quality requiring 30% overlap
(Table 3.2).

The seafloor topography and the slope (gradient of the slope) is an important consideration for
planning the survey lines. For MBES data collection, it is strongly advised to run the lines parallel to
the seabed contours (along the slope, not up or down the slope). This is beneficial for keeping the
coverage reasonably constant along the survey lines (as the swath width will vary with depth). It is
also beneficial because less acoustic energy is reflected towards the transducer from steep slopes,
causing poorer detections and the possibility of false detections in the sidelobes5. If survey lines
must run up and down the slope, a reduction of vessel speed or reduction in swath width may be
required to allow for the echo sounder to track the bottom continuously. Planned lines must be
activated in this instance to ensure that gaps are not created between the survey lines as the swath
coverage is reduced coming into shallow water and additional lines may need to be added.

For surveys where backscatter information is critical, the overlapping area should be increased
(from 60% to 100%) to compensate for the high variability of individual backscatter intensities on the
edges of the outer beam (Gavrilov & Parnum, 2010). For surveys where backscatter information is
considered a secondary product, it is recommended that the overlapping be kept as minimal as
practical (30% overlap).

3.5.7 Pulse length

The pulse length affects the amount of the transmitted acoustic energy into the water and the
vertical resolution of the observed depth. Increasing pulse length enhances penetration through the
water column but reduces vertical resolution. Kongsberg systems have limited, pre-defined options
for pulse length which may be synonymous with other software packages. Therefore, the selection
used may compromise the quality of backscatter data in order to meet the objectives of the survey.

Pulse length and sampling frequency must be considered as related to backscatter data. The
sampling frequency of the system must be considered in order to hold the Nyquist-Shannon
sampling theorem6. This enables the analogue signal to be reconstructed from the digital data (e.g.
Kongsberg EM3002 systems recommended minimum pulse length of 100µsec or greater).

3.5.8 Tides and GPS tides

MBES data shall be corrected in real time for draft and tide variations as well as attitude input (roll,
pitch, yaw and latency) via the vessel’s Motion Reference Unit (MRU). All soundings shall be
reduced to an ellipsoid with a minimum depth accuracy of 0.2% relative to water depth.

3.5.9 Data type

Bathymetry and backscatter datasets shall be processed and plotted onboard to monitor the
coverage and data quality. This will allow for additional acoustic to be collected prior to the
finalisation of the survey, in the event of data gaps between survey lines for example. Processing
will be carried out to create full coverage bathymetric maps with contours, slope values and
backscatter images. Onboard processed products shall include the following:

Page | 57
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Bathymetry

 All raw bathymetric data is to be provided in proprietary format.

 Processed data is to be provided as a point cloud text file (or .csv) in UTM and depth (with
depth value as negative) with uncertainties attached to each sounding.
 Processed data shall also be provided as gridded data in formats csv, ARC GIS Grid. ESRI
ASCII *.asc format in UTM format.
 The requirement for gridding interval of MBES data is 1 m or better in shallow water (<100
m), 5 m in deeper water (100 – 200 m) and 10 m off the shelf (>200 m).
 Bathymetric charts shall be displayed with the smallest contour interval representative of the
seafloor morphology.
 Any smoothing of contour lines is to be kept to a minimum.
 As a QA product, two images of the gridded processed bathymetry data should be provided
with sun-illumination from two orthogonal directions and 5 times exaggeration

Backscatter

 All raw backscatter data is to be logged in proprietary format.

 Processed data is to be provided as a text file (or .csv) showing latitude, longitude and
intensity (dB) (or provided in a format that is able to be converted to comma delimited csv
files).
 Processed backscatter data is preferred as xyz ASCII comma delimited (XY in UTM zones; z
in dB with 2 decimal places), and/or ESRI ASCII *.asc (values in dB) (Buchanan et al. 2013).

3.6 Data Processing

The data acquisition parameters are established to ensure that the data are fit for the purposes of
benthic habitat mapping (i.e. baseline) and monitoring of Australia’s waters. The post processing
parameters and techniques, on the other hand, are generally optimised for a targeted purpose (and
differ for a baseline or monitoring survey). There are a number of software packages available for
processing MBES data and many of the software are in commercial proprietary to the specific
multibeam system used for acquisition.

Regardless of the data processing software that is used (e.g. CARIS), the goal is for the minimum
final products to be released at the completion of a field survey (summarised in Table 3.3):
 Bathymetric surface as both an x,z,y point surface and a raster x,y,z to the appropriate
resolution requested (Table 3.2);
 Vessel transect log map to show the position of the vessel survey lines within the region;
 Map showing the location of field validation data (e.g. point map of where sediment grabs or
video transects have been conducted etc) [Recommended];
 Digital terrain models with hill-shading of the bathymetry from two orthogonal direction and 5
time exaggeration to easily identify artefacts of the dataset remaining, but also to identify key
geomorphological features (including slope map) [Recommended];
 Backscatter mosaic (both raw and processed) in geotiff format in the optimal resolution from
the snippet and at 1 m from the average beam values;

Page | 58
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

 Water column backscatter (display of water column acoustic anomalies and x,y,z location of
features detected in the water column) [Recommended, if available]; and
 3D perspective videos of significant findings or seabed features (abrupt changes in relief,
shipwrecks, canyon head steps etc.) [Recommended].

3.6.1 Bathymetric data processing

Uncertainty related to the bathymetric (depth) measurements can be quantified and incorporated
into a statistical model to derive the total propagated uncertainty (TPU) of the resulting bathymetric
surfaces. A number of factors will influence this uncertainty including: draft setting of the transducer,
incorrect sound velocity profiles, spatial variation in the sound velocity, temporal variation in the
sound velocity, instrumental uncertainty (internal precision of the MBES unit) and motion (incorrect
heave, pitch and roll corrections), settlement and squat of the vessel in the water and incorrect tidal
corrections to name a few.

Where possible the CUBE (Combined Uncertainty and Bathymetry Estimator) should be used to
calculate the TPU for the bathymetric surface as a measure of uncertainty in the survey. CUBE
uses soundings and their associated uncertainty estimates as input and through spatial and
uncertainty weighting, while also relying on the very high data density of multibeam data sets,
outputs a bathymetry gridded surface and its associated uncertainty (error) surface. In addition, it
tracks the statistical hypotheses for each depth point, and where there is more than one estimate,
makes an attempt to determine which the most likely value is. This makes it a very powerful tool for
identifying and removing outliers in the data. Once these have been removed from the data, CUBE
is rerun to generate the final bathymetry and uncertainty surfaces. See “CUBE Bathymetric data
Processing and Analysis (CHS February 2012)”. The uncertainty surface is a quantification of the
survey quality, which can be compared against specifications and used as input to the metadata for
the survey (CHS 2013).

3.6.2 Backscatter data processing

A final compensated Geotiff mosaic of the acoustic backscatter for the survey region should be
generated. We refer to the Lurton, X.; Lamarche, G. (2015) Backscatter measurements by seafloor-
mapping sonars. Guidelines and Recommendations. Geohab Report. 150p, for optimal processing
procedures for MBES backscatter image generation.

3.7 Data Interpretation

MBES bathymetric data will be processed to characterise and classify the seafloor in terms relevant
to the distribution of benthic habitats and to help in the understanding of the spatial and temporal
distribution of marine habitats. The combination of topography (bathymetry) and textural surfaces
(backscatter) provide an excellent reference dataset for research and management of Australian
marine seafloor habitats.

Geomorphological analysis can be used to classify the multibeam bathymetry data and define the
extents of particular habitat types such as seagrass beds, rocky reef, and sand plains. We
recommend the use of the national standardised benthic habitat classification nomenclature as
documented by Seamap Australia (Butler et al. 2017). Importantly, this classification system
includes other established and developing national classification schema such as CATAMI (Althaus
et al. 2015) and Geoscience Australia’s Classification and Glossary of Seabed Geomorphology.

Page | 59
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

The backscatter Geotiff can be interpreted into a sediment distribution and habitat map using one of
two automated segmentation methods:

1. Image-based segmentation (e.g., using e-Cognition (www.ecognition.com)) where the

image is segmented into regions of similar backscatter characteristics and using the
bathymetric data to identify these boundaries and transition zone. These segments are
then classified as surface features, backscatter intensity patterns of sediment/habitat
distribution etc.
2. Signal based segmentation (e.g., using ENVI (www.esriaustralia.com.au/envi) where
changes in the backscatter intensity, with increasing grazing angle from nadir, are
analysed to classify the data.

Page | 60
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Table 3.3 Expected data deliverables for a baseline mapping or monitoring survey to accompany metadata reporting

Deliverable item Comment

Raw sonar data Raw sonar data in native format as created directly from the native
acquisition system of the multibeam system used. e.g. *.all for Kongsberg
EM series, *.s7k for newer version of Reson SeaBat or *.xtf for the older one
Data format: native format as produced by the acquisition system, except for
the *.xtf
Datagram: all logged automatically for Kongsberg EM series. For Reson
SeaBat, datagrams with the following IDs are required:
1003, 1012, 1013, 7000, 7001, 7002, 7004, 7006, 7005, 7007, 7012, 7022,
7028, 7200, 7504
The water column data, recorded as separate files, for both Kongsberg and
Reson are only required on special request in survey planning.
For all other multibeam systems, it is required that raw data include SV
profile, attitude, navigation, heading, raw bathymetry, raw backscatter per
beam and if available raw backscatter in time series i.e. the equivalent
seabed image or snippet style
Processed sonar Processed multibeam bathymetry data, including processed multibeam
data backscatter data, if requested
Preferred format: Caris HIPS & SIPS project structure including processed
bathymetry surface (see processed bathymetry grids below) (*.csar and
XYZ) and time series-generated backscatter mosaic (*.csar) in
Fieldsheets subfolder, processed line data & geobar in HDCS_Data
subfolder, tide data used (*.tid) in Tide folder, individual sound velocity
profiles (*.csv) used together with additional information on time and location
of the cast in SVP subfolder. Backscatter mosaic and geobar are only
required on request
Alternative format:
Processed line: SAIC GSF (*.gsf) if no other alternative
True Heave Delayed, processed heave saved independently from raw sonar file, logged
in 600-720 minutes period
Data format: Applanix ATH or equivalent (Caris compatible)
Processed Processed multibeam bathymetry surface grid
bathymetry grids Data format: CSAR and xyz ASCII comma delimited (XY in specified UTM; z
in negative metre at 2 decimal places) and/or ESRI ASCII *.asc (values in
meter).
Processed Processed multibeam time series-generated backscatter mosaic
backscatter mosaic Data format: xyz ASCII comma delimited (XY in specified UTM; z in dB at 2
decimal places) and/or ESRI ASCII *.asc (values in dB).
Tide Tide data used for tide correction (date, time and depth(m.mm)/pressure
(dBar)
Data format: Caris tide *.tid or ASCII *.csv
Sound velocity profile Sound velocity casts used in SIS or equivalent acquisition system together
Data format: ASCII *.csv
Log file (SVP cast) SVP cast info (date, time, depth of cast and seafloor, location and line
applied to)
Data format: ASCII text
TPU/ CUBE related ● XYZ of MRU to Transducers
information
● XYZ of NAV to Transducers
● Transducers mounting angles (if not horizontal)
● Type of Navigation system
● Type of MRU system
● Sign conventions used to calculate XYZ (Down positive etc)
Data format: ASCII text

Page | 61
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

3.8 Data Release

At the time of writing this manual, there is not currently a complete repository for multibeam data
collected in Australian waters, although several agencies house some multibeam data (e.g. AHO,
GA, IMOS, CSIRO), and several portals promote its accessibility and visualisation (e.g.
seamapaustralia.org). Initiatives are underway for a single repository to be linked to appropriate
visualisation platforms, and this is expected to be addressed in Version 2 of this field manual.

In the meantime following the steps listed below will ensure timely release of data and maximise
data discoverability:

1. Create metadata record(s) describing the survey and data collection (for both raw and
QA/QC data products). Minimum metadata requirements for multibeam data include the
following:

 Title of the survey region (e.g., AMP name and ID) and, if not a well-established region,
its geographic boundary;
 Surveyor’s name and company;
 Start and end dates of the survey;
 Vessel name, type of vessel and MBES unit used, details regarding the positioning
system, acquisition software, and operation parameters;
 The number of lines recorded and corresponding number of kilometres; and
 Summary of the main survey results (water depths, observed tidal range, sonar features
of interest- anomalies, unusual targets etc.).
2. Publish metadata record(s) to the Australian Ocean Data Network (AODN) catalogue as
soon as possible after metadata has been QC-d. This can be done in one of two ways:
 If metadata from your agency is regularly harvested by the AODN, follow agency-specific
protocols for metadata and data release.
 Otherwise, metadata records can be created and submitted via the AODN Data
Submission Tool at https://metadataentry.aodn.org.au/submit. Note that user registration
is required, but this is free and immediate.
3. Generate interactive map imagery of the following derived data layers:
 Location map with limits of the survey area;
 Bathymetric map showing the depths, slope and bathymetric hill shading results;
 Backscatter data map showing boundaries between habitat features;
 Location of auxiliary data sampling (point features of sediment grabs) or transect lines of
video surveys; and
 Map showing the track plot of the vessel position, indicating the region of the patch test
calibration.
4. Upload raw multibeam data files and all field logs generated during the survey to a secure,
publicly accessible online repository (contact AODN if you require assistance in locating a
suitable repository).

Page | 62
 Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

5. Add links to the location of raw data and derived map imagery to the previously published
metadata record. Metadata accompanied by map imagery as described above may be
additionally showcased through the Australian Ocean Data Network portal.
6. Produce a technical or post-survey report documenting the purpose of the survey, sampling
locations, sampling equipment specifications etc. Provide links to this report in all associated
metadata records [Recommended].

3.9 Field Manual Maintenance

In accordance with the universal field manual maintenance protocol described in Chapter 1 of the
Field Manual package, this manual will be updated in 2018 as Version 2. Updates will reflect user
feedback and new developments (e.g. data discoverability and accessibility). Version 2 will also
detail subsequent version control and maintenance.

The version control for Chapter 3 (field manual for MBES) is below:

Version Description Date

Number
0 Submitted for review (NESP Marine Hub, GA, 22 Dec 2017
external reviewers as listed Appendix A.
1 Publicly released on www.nespmarine.edu 28 Feb 2018
2 Relevant updates, including Data Release Early 2019
sections based on NESP, AODN, IMOS, GA,
and CSIRO projects

3.10 References
Althaus, F., N. Hill, R. Ferrari, L. Edwards, R. Przeslawski, C. H. Schönberg, R. Stuart-Smith, N. Barrett, G. Edgar, and J.
Colquhoun. 2015. A standardised vocabulary for identifying benthic biota and substrata from underwater
imagery: the CATAMI classification scheme. PLoS ONE 10:e0141039.
Anderson, J. T., D. V. Holliday, R. Kloser, D. Reid, Y. Simard, C. Brown, R. Chapman, R. Coggan, R. Kieser, W. Michaels,
A. Orlowski, J. Preston, J. Simmonds, and A. Stepnowski. 2007. Acoustic seabed classification of marine and
physical and biological landscapes. Denmark.
Buchanan, C., M. Spinoccia, K. Picard, O. Wilson, M. J. Sexton, S. Hodgekin, R. Parums, and P. J. W. Siwabessy. 2013.
Standard Operation Procedure for a Multibeam Survey: Acquisition & Processing., Geoscience Australia:
Canberra.
Butler, C., V. Lucieer, P. Walsh, E. Flukes, and C. Johnson. 2017. Seamap Australia (Version 1.0) the development of a
national marine classification scheme for the Australian continental shelf. University of Tasmania.
CHS. 2013. Canadian Survey Management Guidelines. Fisheries and Oceans Canada, Canada.
Edward, S., and T. Martin. 2015. Geophysical Surveying and Mapping (GSM): Shallow water multibeam surveying
standard operating procedure. CSIRO.
Gavrilov, A. and I. Parnum. 2010. Fluctuations of seafloor backscatter data from multibeam sonar systems. IEEE Journal
of Oceanic Engineering. 35 (2): pp. 209-219.
Lucieer, V., and G. Lamarche. 2011. Unsupervised fuzzy classification and object-based image analysis of multibeam
data to map deep water substrates, Cook Strait, New Zealand. Continental Shelf Research 31:1236-1247.
Lucieer, V. L., N. Hill, N. S. Barrett, and S. Nichol. 2013. Do marine substrates 'look' and 'sound' the same? Supervised
classification of multibeam acoustic data using autonomous underwater vehicle images. Estuarine, Coastal and
Shelf Science 117:94-106.
McArthur, M. A., B. P. Brooke, R. Przeslawski, D. A. Ryan, V. L. Lucieer, S. Nichol, A. W. McCallum, C. Mellin, I. D.
Cresswell, and L. C. Radke. 2010. On the use of abiotic surrogates to describe marine benthic biodiversity.
Estuarine, Coastal and Shelf Science 88:21-32.
McGonigle, C., C. Brown, and R. Quinn. 2010. Insonification orientation and its relevance for image-based classification of
multibeam sonar. Ices Journal of Marine Science 67:1010-1023.

Page | 63
Marine Sampling Field Manuals for Monitoring Australia’s Commonwealth Waters Version 1

Monk, J., N. S. Barrett, N. A. Hill, V. L. Lucieer, S. L. Nichol, P. J. W. Siwabessy, and S. B. Williams. 2016. Outcropping
reef ledges drive patterns of epibenthic assemblage diversity on cross-shelf habitats. Biodiversity and
Conservation 25:485-502.
Porter-Smith, R., V. D. Lyne, R. J. Kloser, and V. L. Lucieer. 2012. Catchment-based classification of Australia's
continental slope canyons. Marine Geology 303-306:183-192.
Rattray, A., D. Ierodiaconou, L. Laurenson, S. Burq, and M. Reston. 2009. Hydro-acoustic remote sensing of benthic
biological communities on the shallow South East Australian continental shelf. Estuarine, Coastal and Shelf
Science 84:237-245.
Sinquin, J. M., C. Vrignaud, and G. Mathieu. 2016. DORIS Software: A tool to process sound velocity profiles. Hydro
International, Online: https://www.hydro-international.com/content/article/new-tool-to-process-sound-velocity-
profiles?output=pdf
Watson, S. J., Whittaker, J. M., Lucieer, V., Coffin, M. F., & Lamarche, G. 2017. Erosional and depositional processes on
the submarine flanks of Ontong Java and Nukumanu atolls, western equatorial Pacific Ocean. Marine Geology,
392, 122-139.

3
An interferometric multibeam measures the angle of the incoming sound wave fronts in a time sequence of samples.
Slant range is obtained from the time of the sample and speed of sound.
4
A beamforming multibeam mathematically forms a set of “beams”, and detects the range to the seabed in each beam.
5
The sidelobes are smaller beams that are away from the main beam. These sidelobes represent energy received in
undesired directions which can never be completely eliminated.
6
In the field of digital signal processing, the sampling theorem is a fundamental bridge between continuous-time signals
(often called "analog signals") and discrete-time signals (often called "digital signals"). It establishes a sufficient condition
for a sample rate that permits a discrete sequence of samples to capture all the information from a continuous-time signal
of finite bandwidth.

Page | 64