A 1.What should I consider when choosing a calibration site?
Site location depends on the range of available seabeds. The best catalogue contains the widest range of acoustic signatures. At minimum, a catalogue should contain three classes such as a mud, sand or bedrock. A maximum of 16 classes can be displayed and will be classified if acoustically dissimilar.
Consider the following when choosing a calibration site:
- Seabed Homogeneity - The seabed should be the same in all directions for approximately 100 to 200 metres depending on water depth and beam width. Seabed conditions, such as minimal tidal currents or relatively low sediment mobility, should be relatively stable during the sampling period. However, seabed activity should reflect normal conditions. Homogeneity is also dependent on application. For example, with habitat mapping, a highly mixed gravel-sand environment may be considered a preferred habitat for a species of juvenile fish. Therefore, it is a valid class.
- Low Relief Bathymetry - The seabed should not have high relief structures such as cliffs or deep, incised channels that will show a large shift in bottom pick over a small area.
A 2.How should I verify the seabed type at my calibration sites?
Ground-truthing is key in determining seabed type. Data collection provides information on characteristics and physical properties. There are numerous methods to collect information, the most common of which are presented below.
- Grab sampling, the simplest method, will likely be the primary source of ground-truth data. However, the limitations of grab sample data should be recognized and additional techniques employed. For example, grab samplers are not very useful in areas where there is a gravel armour - samples are often lost during recovery because the clasts catch in the jaws - and they are practically useless on bedrock. A sampler is lowered to the seabed to collect surficial material. There are numerous types of samplers including Van Veen, Shepec and Eckman, all which penetrate less than 15 or 20 centimetres to collect loose sedimentary material from the uppermost portion of the seabed. At least two samples should be taken to ensure good representation.
- Coring works well with low frequency sounders (24-33 kHz) as the acoustic response can represent 60 centimetres or more subsurface penetration. Note that core data from previous investigations may be available to supplement the data set. There are numerous types of corers such as gravity corers and piston corers. A core barrel is dropped through the water column and into the seabed. Loose sedimentary material is forced into the tube and retained by a special valve or core catcher. Penetration, usually less than two metres, is highly dependent on substrate and becomes limited as grain size increases, making corers ineffective in gravel or cobbles. A vibrocorer can sample very coarse sand and small gravel, but this special corer requires extensive logistics.
- Video of the calibration area in conjunction with a series of grab samples is the most effective method. An underwater video camera mounted on a frame with a light source films and records the seabed. The camera must be stationary or towed at a very low speed for the process to work.
- Still photography is not practical during real-time calibration due to the delay in photographic processing. However, photographs can be valuable in a post-processing mode, particularly on surfaces such as bedrock outcrops where grab samples are ineffective. One camera, or two for stereo photography, is lowered to the seabed. The camera usually has a bar scale suspended below it acting as a trigger and providing a reference to estimate surficial sediments grain size.
A 3.How many records should I collect at each calibration site?
The purpose of calibration is to provide a statistically valid number of sample echoes from a seabed to create a representative ensemble to characterize its acoustic signature. By nature the signal varies due to ambient noise from the survey platform, vessel motion or minor transducer inconsistencies. The system requires a minimum of 30 records for processing to be valid; a minimum of 50 records is recommended. More than 80 records does not provide a significant amount of useful information. Points to remember:
- A similar number of calibration records should be collected at each calibration site for a catalogue, thus eliminating bias of processing in favour of any one seabed.
- It is good practice to resolve that all library calibration files will contain 50 records.
- By attempting to obtain those last few echoes to account for the last percentage of uncertainty, the vessel might move from the original site to a new seabed type, thus contaminating the signal.
- When collecting calibration records it is important to maintain good coupling between the water and the transducer. Cavitation due to running the engines astern, can wash bubbles over the transducer, thus contaminating the signal.
A 4.How do I know when I have a good catalogue?
Every catalogue should be reviewed. The data should be replayed to assess the acoustic signature of each class. This is vital when ground-truthing is minimal or when it is uncertain if the chosen classes are acoustically distinct. Once the data are replayed in CAPS, the plot can be rotated. If two or more classes cannot be isolated in any view, they are acoustically similar and represent the same class. There can be a 10% to 20% overlap in the data from two classes. When the classes have been viewed, the choice for the final constituents of the catalogue can be made and the catalogue downloaded. The choice of classification for a new echo is not based on straight geometrical considerations as seen in Q-Space. The spread of clusters is a good indication of the acoustic uniqueness for each seabed, but the ability of the system to map each class is not strictly dependent on this spread.
A 5.How do I set the digitization parameters in DACS?
One of the more important functions of seabed classification is picking the correct seabed depth. The QTC VIEW™ searches in a specific window for the initial seabed return. As a new echo comes in, the system discerns the previous water depth. It searches for the new bottom within a window bounded by the values shown in the Shallow Gate and Deep Gate fields. If it cannot find the seabed within the window, QTC VIEW™ disregards the trace and proceeds to find the bottom. After a number of traces, the system recognizes a drastic change in water depth and looks outside the window for the bottom. For relatively flat seabeds or on slopes less than 10 degrees, set Shallow Gate and Deep Gate to 3 metres. For steeper slopes, set the gating to an appropriate level depending on vessel speed and slope angle. Blanking, normally set to 0.5m, sets the minimum water depth in which QTC VIEW™ looks for the seabed and disregards all data received above this depth. Where there are anomalies such as fish schools in the water column, this value can be increased to preven false depths.
A 6.How should I name my catalogues and classes?
Class names should be carefully defined particularly when a classification scheme is used for more than one application or area. Use names that correspond to the calibration data collection site. A detailed seabed description can be associated with the class. Specific calibration details should be recorded while developing a catalogue library with associated classes. For example, if classes are named sand, gravel or mud, you are limited to defining only one type of sand. By devising a scheme independent of basic names, you can have three types of sand provided they are acoustically distinct.Detailed records on the seabed at each calibration site can be referenced to the class name, thus allowing for more detailed information to be transferred to a map.
A 7.How do I know when I have a new class?
The definition of a new class depends on the application and classification scheme. A new class is identified by a consistent confidence factor, generally lower than 70%, from a continuous area of the seabed. The points displayed in Q-Space, associated with the lower confidence level, form a cluster that can be isolated or partially isolated from the other points.
A 8.During data acquisition with QTC VIEW™ and out Bathy 500 echo sounder, it took almost 30 minutes to get 50 records for one sediment type during CAPS. We managed to create catalogues but DACS was also extremely slow.
The system needs 250 pings to collect 50 records. Look at the trig LED to see
the repetition rate. Even at one trigger per second, this should take less than
five minutes. If your trigger LED is pulsing at greater than one per second be
sure that it is triggering at an appropriate rate for the water depth. If the
rate is too high and you are in deep water you will get ping collision and both
the echo sounder and the QTC VIEW will have difficulty acquiring the data. The
repetition rate on your echo sounder is set by the range, the lower the range
the faster the ping rate. The range also sets the pulse duration and you need to
maintain a fixed pulse duration for a data set. Your echo sounder has two pulse
durations, 200 µs and 500 µs. The Bathy 500 has an issue with its
software that can effect the data collected by QTC VIEW™. When you choose
a Range in the menu and then look at the Phase menu, the system resets the Range
to 0-10m. The way to stop this is to look at the Phase first and then set the
Range. If the repetition rate is ok and the LED trig is pinging, then maybe your
echo sounder is not outputting enough power. The power setting on your echo
sounder is set manually or to auto. Do not use auto but manually choose a
setting, usually greater than 80. It is important to note that power settings
and pulse duration must be maintained for a data set.
Another reason you could have a poor update rate for records is too many GPS
strings being sent to the QTC VIEW™ GPS port. If more than two or three
strings are being sent then the QTC VIEW™ port can become overwhelmed and
it can slow the number of records being sent. Ideally, only GPGGA will be sent
to the unit.
Two other possibilities are: too little disk space (on the C drive) for logging
or the serial cable from the QTC VIEW™ to the computer is too long
(greater than 7 meters).
A 9.What are the GPS string requirements ?
GPGLL or GPGGA
