W3-Multibeam Operation
W3-Multibeam Operation
SBEU3523 HYDROGRAPHIC
SURVEYING TECHNOLOGY
Lecture 3:
Multibeam Operation
3
ECHO SOUNDING OF THE SEAFLOOR
Single (left) and multibeam (right)
echo sounding of the seafloor.
Single beam systems (SBES)
typically have beam widths of 10-30
degrees and estimate depth by
measuring the shortest slant range
to the seafloor within the main
beam. Multibeam (swath sonar)
systems provide a series of slant
range and elevation angle estimates
along a fixed azimuth.
(Image courtesy of Lewis and Clark 2001, NOAA/OER)
4
M B E S vs S B E S
5
SINGLEBEAM [SBES] - SOUNDING LINES LAYOUT
In single beam surveying, lines are designed to run normal to a slope.
Due to the limits of a single beam, details of the slope are better mapped
running up and down slope.
However, this can lead to problems if the area to be mapped is in very
shallow water.
6
MULTIBEAM [MBES] - SOUNDING LINES LAYOUT
In multibeam surveying, lines are designed to run parallel to the contours.
This is done for various reason - Line spacing is a function of swath coverage
which is a function of water depth - Surveying parallel to a ridge allows line
spacing to remain the same.
As water depth remains fairly constant, surveying parallel to the contours, means
less manipulation of the range settings in the processor.
As the swath goes far from the track of the vessel, shallow areas can be mapped
safely.
7
SOUNDING LINES LAYOUT [SBES vs MBES]
8
MBES SOUNDING
OPERATION
Illustration of what may
happen if multibeam lines
are laid out similar to single
beam lines, running normal
to contours.
9
MULTIBEAM SURVEY OPERATION
Swath coverage is a function of water depth, up to 8 times water depth.
Line spacing is based on desired coverage density.
It is highly recommended that line spacing be such that there is 100% overlap
in coverage.
Excessive roll & pitch can affect swath coverage.
Vessel steering offline can affect swath coverage.
10
MULTIBEAM SURVEY OPERATION
The outer swath of one line,
where the sounding density is
sparse, is covered by the nadir
beams of the adjoining line,
thus greatly increasing the
sounding density.
The overall intra-swath
sounding density is greatly
increased.
11
MULTIBEAM SURVEY OPERATION
With too little overlap, excessive roll
or bad steering can leave ‘holidays’
between lines.
Holiday: An unintentional un-
surveyed area within a given
hydrographic survey where the
spacing between sounding lines or
surveys exceeds the maximum
allowable limits (IHO Dictionary S- 32).
12
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
Determine approximate depth of survey area.
Check to determine water conditions.
Is there any fresh or saltwater entering survey area.
Is the survey in an area of different bottom types.
Determine if vessel traffic will be in survey area, if so, is it general
or channel traffic.
Are there hazards in the area, such as rocks, shallows, or reefs.
13
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
Need to know approximate depth.
Determines line spacing.
Unlike single beam surveys where line spacing is based on the
scale of the survey (e.g. 1:1000), multibeam line spacing is based
on swath coverage which is a function of water depth.
If water is well mixed, then the deepest portion of the survey
area is the place where the sound velocity profile should be
taken.
14
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
The sound velocity profile is critical to accurate data collection. If
the survey area has different water conditions, sound velocity
profiles will need to be taken in those areas, to correct for
refraction.
The bottom type determines the amount of reflected and
absorbed energy.
– In a soft bottom, the parameters of the multibeam may need
to be adjust to compensate for the higher degree of energy
absorption.
15
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
In the case of different water types or different bottom types the
survey should be adjusted for these differences:
– The survey should be designed so that all lines are within one
type of environment.
– The surveyor should be aware of where the conditions change
and be prepared to act on that.
16
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
In determining the maximum speed of a survey, is how well the
multibeam is installed and how much noise is generated in going
at speed.
Another factor is the desired coverage for the survey. The faster
the vessel goes, the more area is traveled, but the bottom
coverage may be reduced.
17
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
Before a survey starts, much thought should go into the plans
for taking the sound velocity profiles (SVP casting).
Ideally the survey will be laid out in such a manner that the
areas, where a cast should be done, are plainly identified and
fit within the survey plan.
- Consider freshwater influx
- Consider tides and possibly salt wedges
- Consider water depths
- Consider the time of day
18
EXTERNAL FORCING MECHANISM THAT INFLUENCE THE SOUND
VELOCITY STRUCTURE
Sonar Heating
Preciptation
Evaporation
Wind Induced Wave
Mixing
Longshore Currents
19
Brackish water is a broad term
used to describe water that is more
saline than freshwater but less
saline than true marine
environments. Often these are
transitional areas between fresh
and marine waters. An estuary,
which is the part of a river that
meets the sea, is the best known
example of brackish water.
MULIBE A E C H O S O U N D E R [MB E S ]
MSURVEY
SINGLEBEAM ECHO
SOUNDER [SBES]
- SURVEY OPERATION
MULTIBEAM ECHO
SOUNDER [MBES]
- SURVEY OPERATION
S U R V E Y LINES
MULTIBEAM ECHO SOUNDER [MBES]
- SURVEY OPERATION
Off-nadir beams are subject to greater amounts of refraction.
The result of refraction with multibeam surveys is a degradation
in the quality of the data – outer-beams appearing either shallow
or deeper than the nadir beam.
At an angle of 450 in 10 meters of water, a ±10 meters per second
velocity error will result in a depth error on the order of ± 4.6 cm.
Convex = SV ↑ with depth
Concave = SV ↓ with depth
23
24
MULTIBEAM SURVEY [MBES]
- SURVEY SETTING
Depending on the multibeam
manufacturer and acquisition
software, the operator can
have access to very different
type of settings.
25
MULTIBEAM SURVEY [MBES]
- SURVEY SETTING
Reference point (RP)
- Centre of gravity of survey ANTENA:
----- Coordinate X, Y and Z
vessel (COG) IMU/Vertical Motion:
26
MULTIBEAM SURVEY [MBES]
- SURVEY SETTING
The primary setting are:
– Beam Spacing: modes of equidistant, equiangular or in-between.
– Coverage: fixed port/starboard angles or fixed port/starboard distances.
– Gain settings: should be constant for backscatter applications.
– Filtering: filtering options differ widely from one system to another.
usually data are filtered, or flagged as rejected, depending on the
strength or direction of the returned acoustic echo.
– Offsets: position of sensor relative to common reference frame.
27
MULTIBEAM SURVEY [MBES]
- GEOMETRY
28
MULTIBEAM SURVEY [MBES]
- GEOMETRY
draught
29
MULTIBEAM SURVEY [MBES]
- GEOMETRY
Across-track
water level
30
MULTIBEAM SURVEY [MBES]
- GEOMETRY (Effect of Pitch and Roll Offsets Towards Slant Range)
Z
Draft
R
r (range = ½ of 2 –way travel time)
(measured
depth)
d θ d = r kos (P) kos (θ + R)
x
(across-track) where,
R = roll
Z
P = pitch
Draft
θ = beam angle
P
r = slant range
y
(along-track)
31
MULTIBEAM SURVEY [MBES]
- DATA PROCESSING
Multibeam data processing consists of three (3) steps
– Correction of sounding
– Cleaning and check of sounding
– Visualization of sounding
32
MULTIBEAM SURVEY [MBES]
- DATA PROCESSING (Correction of Sounding)
Tidal information must be incorporated at the post processing
stage – to correct all sounding depths to a standard water level.
Draft reading and sound velocity corrections can also be applied
at this stage – or reapplied to correct for erroneous settings
made during data acquisition.
33
MULTIBEAM SURVEY [MBES]
- DATA PROCESSING (Cleaning and Checking)
Data cleaning and checking will apply to vessel navigation and
attitude data, as well as the depth sounding.
The process begins with a visual inspection of vessel navigation
and attitude data to identify and remove any invalid
measurement by the navigator or Motion Reference Unit [MRU]
instruments.
The suppression of erroneous depths caused by fish, noise or air
in the water column can be performed manually or automatically
through various filters.
34
MULTIBEAM SURVEY [MBES]
- DATA PROCESSING (Cleaning and Checking)
Erroneous of soundings are also called outliers or spikes.
Manual cleaning can be performed through several interfaces
depending on the software manufacturer.
Filtering methods are numerous – can be as
– simple as setting a minimum/maximum depth threshold, swath
reduction (outer beams are usually the noisiest).
– Employ more advanced algorithm – CUBE (Combined
Uncertainty Bathymetric Estimator)
35
MULTIBEAM SURVEY [MBES]
- DATA PROCESSING (Data Visualization)
The outcome of data processing are cleaned soundings (x, y, z, recorded and
backscatter) – which can be used for visualization of the bathymetric data.
Two (2) main types of digital terrain model (DTMs) can be built from those
points:
– A Triangular Irregular Network (TIN) – where sounding are the vertices of
triangular cells – this type of DTM gives an image of the seafloor with actual
depth value – usually used for the safety of navigation applications.
– A Regular Network – a grid of square cells where the depth of each node is
an averaged depth of the neighboring sounding – this type of DTM gives a
smoother image – usually used in geophysical analysis.
36
MULTIBEAM SURVEY [MBES]
- DATA PROCESSING (Data Visualization)
Images with geographical information (the “geotiff” format for example) may
also be created and imported into a Geographical Information System (GIS).
Each pixel of the images is attributed with an averaged depth value and color
coded in accordance.
37
MULTIBEAM ECHO SOUNDER [MBES]
- ADVANTAGE
Surveying in shallow areas, single beam (SBES) is limited to where the vessel
can go. With multibeam survey (MBES), the area covered is much wider than
where the ship can go, so that very shallow water can be surveyed safely.
38
Single beam survey lines Shipping Channel
spacing based on scale
39
MULTI-BEAM BATHYMETRIC
CALIBRATION Patch Test
40
TYPES OF CALIBRATION
SBES MBES
• Transducer index error + Bar-check • Patch Test
• Latency test
41
PATCH TEST CALIBRATION
The alignment of the Multibeam sonar head to the motion sensor and gyro
is critical to the accuracy of the determined depths. It is not possible to
install the sonar head in exact alignment with the motion sensor and gyro to
the accuracy required (x.xx°). If GPS time synchronization is not used, the
latency of the position, as reported by the GPS, must also be measured
during the calibration. This being the case a calibration must be performed
to measure the angular misalignment between the Multibeam and the
motion sensor and gyro and, if necessary, the position latency; this is called
the Patch Test.
42
The Patch Test is performed with each new installation or whenever a sensor is
moved. In the case of an over‐the‐side mount, a large number of calibration
computations need to be performed to determine how well the pole goes back
into the same position each time it is deployed. With more permanent mounting
arrangements, a minimum of 5 separate patch tests should be conducted in order
to derive a standard deviation that would indicate the accuracy of the derived
values.
The patch test involves collecting data over certain types of bottom terrain and
processing the data through a set of patch test tools. There are two primary
methods of processing the data that are currently used: an interactive graphical
approach and an automatic, iterative surface match. Each of these techniques has
strengths and weaknesses and the preferred approach is dependent on the types
of terrain features available to the surveyor. All modern multibeam data collection
software packages contain a patch test routine. Please read the software manual
for explicit information regarding the requirements for that software’s patch test.
43
MBES PATCH TEST CALIBRATION
• Using statistical Analysis of the collected data.
• This is not an “Exact Science”
• Patch test calibrations are performed after initial installation, and
periodically thereafter if sensors are modified, to quantify any residual
biases from the initial system alignment
*Best Practice
Performed before / after each MBES job @ as many
times as you can!
44
MBES PATCH TEST CALIBRATION
• Four (4) Individual Tests Vertical Axis
• Three the determine the physical
misalignments of the multibeam
transducer (Roll – Pitch – Yaw)
• One determines the latency, or lack of
synchronizations between the Position
and Depth data acquisition
• The errors resolved are only for transducer to
motion sensor misalignment.
• A dimensional survey is essential to tie these
to VRF/COG!
45
MBES PATCH TEST CALIBRATION
• The orientation of the sonar head must be known in order to convert
the measured slant ranges to depths and to determine the position of
each of the determined depths.
46
MBES PATCH TEST CALIBRATION
• Any error in the measured roll of the Multibeam sonar head can cause
substantial errors in the conversion from slant range to depth.
• A roll error of 1° on a 50 m slant range will cause a 0.6 m error in the
resulting depth.
• Any error in the measured pitch of the transducer head will primarily
have a detrimental effect on the accuracy of the positions that are
determined for each slant range/depth.
• A pitch error of 1° will cause an along‐track error in the position of 0.4
meter when the sonar head is 25 meters above the seabed.
47
Why Patch Test is Important
• Patch Test MUST be conducted after the initial installation of a Multibeam
System onboard a vessel.
https://eprints.utm.my/4676/1/74_Mohd_Razali_Mahmud_AN__APPRAISAL__OF__MULTIBEAM_.pdf
48
PATCH TEST CALIBRATION - CHECKLIST
Measure Equipment Location Offsets with Respect to Boat Reference Point
(X – Port/Starboard, Y – Forward/Aft, Z – Vertical)
Enter these Offsets in their respective Devices in the Acquisition software
Leave Roll, Pitch, and Yaw values in the Multibeam’s Device as all Zeros
(Unless performing a “check” Patch Test for an existing setup.)
Leave Latency value as Zero (0.0)
(Unless performing a “check” Patch Test for an existing setup.)
Locate area to perform Patch Test and “layout” runlines for each Test.
Open patch test tools programs and begin collecting Patch test data.
49
PATCH TEST - LATENCY
Latency is the delay between the position fix and the signal/data
arrival time.
Can cause Positional Errors.
Error is independent of multibeam system.
50
PATCH TEST - ROLL
51
PATCH TEST - PITCH
52
PATCH TEST - YAW
53
HINTS & WARNING
Hints
Test in the deepest section of the survey area.
54
HINTS & WARNING
DGPS Warning:
•Make sure you have good GPS/GNSS positioning.
•RTK GPS is recommended.
•A slight position error may lead to Patch Test errors.
•If HDOP is high, perhaps best to perform Patch Test at another time.
Assuming 3’ DGPS shift in 40’ of water
Latency Test: +/- 0.4 seconds.
Pitch Test: +/- 4.4 degrees.
Yaw Test: +/- 4.3 degrees.
55
HINTS & WARNING
56
HINTS & WARNING
Also keep in mind that your Boat
Operator will make or BREAK your
Patch Test.
57
PATCH TEST AREA SELLECTION
This may be more difficult than it seems.
Locate the sloped area for the Latency, Pitch, and Yaw Tests. If known, locate
an object (sunken boat or car, old bridge pier, rock pile, etc.) and use it.
Once you have located an area for your Patch Test, collect some data with your
multibeam, over this area , in order that you can properly orient your Patch
Test runlines with the bottom features needed. It won’t matter that your
offsets are still zero, this is only for your runline setup.
58
PATCH TEST – CRITERIA(S)
The patch test requires collecting sounding data over two distinct types of sea
floor topography; a flat bottom is used for the roll computation whereas a steep
slope or feature is used for the latency, pitch, and yaw data collection.
Care must be taken that the sonar head covers the same area on both data
collection runs, this may not be the same as vessel position, especially with an
over‐the‐side mount or if the sonar head rotated. Only the latency data
collection requires a different speed from normal survey speed.
The data collection for Latency, Pitch and Yaw should be done in as deep water
as possible. This is particularly true for the pitch computation due to the fact
that in shallow water the angle of pitch may not be easily determined due to a
lack of resolution.
59
PATCH TEST AREA SELLECTION
If Possible, Find areas similar to these for Yaw, Pitch, and Latency Runlines
Two Slopes help to better define the Correction Value for the Test being performed
60
PATCH TEST CALIBRATION - EXECUTION
61
PATCH TEST CALIBRATION - EXECUTION
PITCH: - Flat bottom to an object or
- Sloped bank feature
- Opposite / Reciprocal lines
- Same / Equal speeds
62
PATCH TEST CALIBRATION - EXECUTION
• Latency: Line C D
– Same line and direction
– Differing speed, sloping bottom
(Watch for settlement differences.)
• Pitch: Line C D
– Reciprocal lines, sloping bottom.
Same Survey speed.
• Roll: Line A B
– Reciprocal lines, flat bottom.
– Same Survey speed.
• Yaw: Lines C D, E F
– Offset lines, sloping bottom.
– Same Survey speed.
63
PATCH TEST CALIBRATION - LATENCY
The vast majority of installations will
incorporate GPS time synchronisation and, as
such, no latency is expected in the GPS
position.
The line is surveyed at survey speed up the slope, and then surveyed again, in
the same direction, but at a speed that should be half of the survey speed.
If the vessel cannot make way at half survey speed then the fast run will need to
be taken at a higher speed than normal survey speed and this can influence the
latency test due to squat or settlement.
The main consideration is that one line should be twice the speed of the other.
65
PATCH TEST CALIBRATION - ROLL
The data collection for roll has to be over a flat
sea floor. One line is surveyed twice, in reciprocal
directions and at survey speed.
66
PATCH TEST CALIBRATION - ROLL
Roll is perhaps the most critical value
in the patch test routine as an error in
Sounding Error due to +0.5° Roll Error in 20 metres depth
roll will result in an error in sounding 0.60
0.40
depths.
0.00
80 ‐60 ‐40 ‐20 0 20 40 60 80
However, the computation to ‐0.20
‐0.40
determine the roll misalignment is ‐0.60
Degrees from Nadir
67
PATCH TEST CALIBRATION - PITCH
The pitch data collection is over
the same type of sea floor as the
latency data collection, i.e. steep
slope or feature on the sea floor.
68
PATCH TEST CALIBRATION - PITCH
A profile of the data will show two different slopes,
which represent the reciprocal data collections. Position Errors due to Pitch
Alignment Errors
The patch test software goes through a series of 6
4
difference between the two surfaces reaches a null. 3
1.0° Error
0.75° Error
2
Whatever the angle of correction, which results in 1
0.5° Error
0.25° Error
69
PATCH TEST CALIBRATION - YAW
The yaw data collection and subsequent solving
for the yaw offset is usually the most difficult of
the 4 tests that comprise a patch test.
70
PATCH TEST CALIBRATION - YAW
For the Yaw data collection two
parallel lines are used, with the Position Error with a Heading Error of 0.50°
vessel surveying in the same 6
tkkelvin@utm.my