In this FAQ we unpack the Echosounder function, from the choice of equipment to the deployment and data analysis. There are five steps towards a good Echosounder campaign:

1. Choosing the right instrument and configuration
2. Choosing the survey design
3. Echosounder calibration
4. Instrument test for noise
5. Data analysis

1. Choosing the right instrument and configuration

The equipment selection is a critical aspect of the deployment process. Each Signature Echosounder comes with distinct specifications such as frequency, power level and pulse duration. Choosing the right instrument and configuration is not trivial and needs to be thoroughly discussed prior to the deployment, considering the specific target. The table below provides an overview of the basic parameters, which are further elaborated.

As with any other ADCP, the selection of the instrument frequency depends on the target under study. High-frequency instruments can discern smaller targets than lower frequencies, whilst having higher water absorption (\(\alpha\)). For fisheries, lower frequencies are a better choice due to tilt effects being reduced. A good starter is the publication by Benoit-Bird and Lawson (2016) for marine organisms:

Figure 1: Acoustic frequency responses for different marine organisms, derived from empirical observations and physics-based scattering models (from Benoit-Bird and Lawson, 2016).

For sediment studies, Smeardon (2007) shows the backscatter for different frequencies given the same particle size and concentration:

Figure 2: Measured acoustic response for different frequencies given the same sediment size and concentration (from Smeardon, 2007).

The transmit pulse length (also referred to as transmit pulse duration or pulse duration) is chosen by the user within the 3 different Signature models. The pulse length affects the range resolution, so it is interesting to have it as short as possible. The resolution in the Echosounder is a convolution of the length of the transmit pulse and the size of the receive window. Just like our current meters, we always match the transmit length to the receive window the best we can to get the classical triangular weighting. For the echo sounder, the transmit pulse is specified in time, which can be multiplied by c/2=750 if you want a length that matches the receive window. The effective resolution will be limited by the receive filter when pulse compression is disabled so you will see 1.2 cm resolution in the data even though the bin size is 3 mm.

The best resolution is achieved when Pulse Compression is selected, which enhances target resolution by compressing the return echo pulse duration using special filters. With pulse compression enabled, the center transducer transmits a sound wave (pulse) with a bandwidth of 25% (compared to the normal 6.25%). Each part of the pulse has a unique frequency, and the return pulse can be separated and integrated into a shorter single output pulse. Practically, that means that the return echo is compressed in its pulse duration in special filters, which results in very high-resolution data. Pulse Compression is effective for enhanced target resolution, with the best achieved at a minimum of 3 mm for Signature1000 and 6 mm for Signature500. That makes the Pulse Compression feature particularly advantageous for small scatterers. Signal-to-Noise Ratio (SNR) improves as the length of the transmit pulse can be increased without compromising resolution. The disadvantage is that pulse compression may introduce sidelobes in shallow water or when measuring close to boundaries. Additionally, increased SNR increases power consumption, so one must be aware of that.

2. Choosing the survey designPlanning the survey design is an important part of the fieldwork. Firstly, the user needs to define the survey objectives (fisheries, plankton, sediment and so on). For biological studies, defining the target species beforehand ensures efficient data collection by selecting the parameters described above and choosing the appropriate survey techniques. Survey types vary from stationary observations to towing equipment through water, gliders, or utilizing downward (buoy mounted) or upward-looking (bottom mounted) sensors, each offering distinct advantages and limitations. Choosing the right survey type should consider factors like spatial coverage, data resolution, and logistical constraints. That choice will also influence on what current and waves (if applicable) parameters are defined.

3. Echosounder calibration

It is essential to calibrate the echosounder gain if you intend to (1) compare data from a specific site over extended periods or (2) compare data from multiple instruments at a given location. However, it is important to note that echosounder calibration gain serves to quantify transducer loss and may not be necessary for all applications. Currently, the responsibility for calibrating the instrument lies with the user. For a complete guide on Echosounder calibration, please refer Demer et al. (2015).  

The gain term in the sonar equation accounts for variations in transducer performance and is specific to the instrument and frequency being used. When echosounder gain is determined and incorporated into estimates of \(S_v\) (volume backscattering strength) and TS (target strength) through the gain term, the resulting \(S_v\)  and TS values are considered absolute. If echosounder gain is not accounted for, \(S_v\)  and TS are regarded as relative (\(S'_v\)  and TS').

For the Signature 100, calibrations have been done with a ~24 mm tungsten sphere. For higher frequencies, such as the 1 MHz system, the sphere gets as tiny as ~3 mm. To ensure a good calibration, make sure no scatterers other than the sphere are present in the water. Additionally, the distance between the center transducer and the water surface has to be precisely calculated.

4. Instrument test for noise

Environmental noise must be removed from Echosounder measurements. Any noise that cannot be removed will increase   measurements and bias in situ TS measurements. Environmental noise can be measured by setting the echosounder power level to -100 dB (this is referred to as listening mode) before transmitting or by taking processed data far from the instrument, where the power level is -100 dB. The resulting amplitude represents the noise threshold. This can be entered into Signature Review or Ocean Contour. When the noise level and salinity are input in Ocean Contour, the software will automatically correct for transmission loss and absorption.

If applying loss equations to calculate S_v and TS from Pr, the noise must be subtracted from the Rx before losses are applied. To do so, first convert the noise and Rx from dB to linear form, subtract the noise, and convert the difference to dB. Default noise values are included in Signature Review and Ocean Contour exported data, but measured values should be obtained and will override these defaults if entered into software. It is implied in the Rx term in the Echosounder equations that noise has been removed. Other Nortek documentation (Ocean Contour users guide, Principles of Operation) explicitly shows the step of removing noise in loss equations.

The background noise can be calculated as follows:

This has to be subtracted from S_v.

5. Data analysis

Data analysis can be done using Ocean Contour or Signature Review. If the user wants to include extra information and variables, they can either export the Matlab data and follow the steps in this FAQ or use third-party software. We recommend our colleagues from Echoview.

References

Benoit-Bird, Kelly J., and Gareth L. Lawson. "Ecological insights from pelagic habitats acquired using active acoustic techniques." Annual review of marine science 8 (2016): 463-490.

Smerdon, Andrew M. "A commercial multi-frequency acoustic backscatter instrument for profiling of suspended sediment size distribution and load." Ultraschall in der Hydrometrie: neue Technik–neuer Nutzen!? (2007): 47.