Difference between echolocation clicks, click trains, click series, terminal buzz and burst pulses (in cetaceans)

Question

What is the best way to distinguish between those? Is the distinction based solely on ICI (inter-click-interval) and if so, what is the ICI range for a click train versus a burst pulse for example? (<10 ms appears to be the number?)

There is some emphasis on ICI in how burst pulses are defined in the literature. For example, in Lammers et al. (2003) burst pulses described as follows: "Burst pulses were distinguished from echolocation click trains on the basis of their ICIs. Click trains were considered burst pulses if their mean ICI did not exceed 10 ms." (p. 1631).

Similarly, in Martin et al. (2019) burst pulses are defined as follows: "...a discrete, isolated series of high repetition rate clicks produced at a relatively constant rate. " (p. 2) and as " ...isolated series of high repetition rate clicks that generally begin, persist and end with ICIs < 10 ms" (p. 3).

Lammers et al. (2003) also discussed some overlap between burst pulses and terminal buzzes: "Admittedly, this distinction [a burst pulse pattern with interclick intervals consistently less than 10 ms] is not necessarily definitive because click trains are sometimes produced that begin with long interclick intervals ~10–1001 ms and end with very short ones ~1.5–9 ms." (p. 1636).

Sources:

Lammers, M. O., Au, W. W., & Herzing, D. L. (2003). The broadband social acoustic signaling behavior of spinner and spotted dolphins. The Journal of the Acoustical Society of America, 114(3), 1629-1639.

Martin, M. J., Elwen, S. H., Kassanjee, R., & Gridley, T. (2019). To buzz or burst-pulse? The functional role of Heaviside's dolphin, Cephalorhynchus heavisidii, rapidly pulsed signals. Animal Behaviour, 150, 273-284.

Answer 1

There's a lot of confusing terminology used about click-based signals in cetaceans, especially if you go back into even slightly older papers. Walter's terminology is great but maybe think of it in terms of functionality instead:

Signals for echolocation: All toothed whales echolocate by emitting clicks and listening for returning echoes from prey or other objects in their surroundings. Echolocation is an extremely dynamic sensory system and many sonar parameters, including especially interclick intervals (ICI), depend intricately on species (scaling), context (acoustic environment) and prey proximity and behavior. While all species tend to switch from regular search click series (slow rate/long interclick interval) to foraging or feeding buzzes (high rate/short interclick interval) when catching prey, the ICI cutoff depends on species (figure 1) and there can be a bit of fluidity for species catching highly mobile prey or in complex (esp shallow water) environments.

Figure 1, showing example clicks and foraging buzzes for (A) Sperm whale; (b) Blainville's beaked whale; (c) bottlenose dolphin; and (d) harbor porpoise. Modified from Fenton et al. (2014) Sonar Signals of Bats and Toothed Whales. In: Biosonar, Springer Handbook of Auditory Research.

Signals for communication: Many echolocating cetaceans also communicate with click series with more or less stereotyped interclick intervals or even individual pulsed sounds. You will find many names used for these (i.e. codas and slow clicks for sperm whales, burst pulses for many delphinids and phocoenids, rasps for beaked whales and pilot whales, probably more terms too). Some of these can overlap with interclick intervals from search clicks and especially foraging buzzes.

To differentiate between regular search clicks and foraging buzzes, you can extract ICIs from acoustic tags and they will tend to be bimodally distributed. To differentiate between burst pulses and foraging buzzes, most studies (including Martin et al 2018 and Soerensen et al 2018) therefore make some operational definitions, including especially whether or not a group of rapid clicks come immediately after echolocation clicks or not.

Answer 2

I would group these terms together

• echolocation clicks and terminal buzzes
• burst buzz

IMO, burst buzzes are isolated (bursts) rapid clicks (buzzes) that are predominately observed from dolphins in multiple contextual situations.

echolocation clicks and terminal buzzes are mostly related to foraging activities and linked together, where terminal buzzes are associated with prey capture attempt.

the terms click train or click series are rather generic without any indication of functionality, duration and ICI.

Re ICI: the actual ICI depends very much on context and species with large overlap. So, no, ICI alone is not sufficient to characterize the type of click series, it needs species context information.

Answer 3

One of the reasons there is confusion in the terminology of the different click trains produced by odontocetes is because there is still considerable ambiguity with respect to the function of certain types of click trains, particularly those with very short ICIs. When performing a detection/classification task, dolphins typically produce clicks with an ICI equivalent to the two-way travel time to the target plus an echo processing period. During prey capture, however, that relationship breaks down in the terminal buzz phase when ICIs no longer appear to conform to this timing. It's generally presumed that echoes are processed differently in the terminal buzz phase to prioritize prey tracking over target classification/discrimination, but this has not yet be confirmed experimentally (as far as I know). So-called 'burst pulses' have very short ICIs on the order of a few milliseconds from start to finish. Under experimental conditions, they are not produced when solving echolocation tasks, but rather seem to be produced in various social contexts. It is presently unknown whether dolphins process the returning echoes from burst pulses in any way, which is why these signals are placed in their own signal class; not because they are fundamentally different from echolocation clicks, but because we don't really know what the dolphin does with them functionally (presumably something social).

In spinner dolphins, Lammers et al (2003) found that there was a bimodal distribution of ICIs that could be roughly separated into the > or < 10 ms. Thus, the dichotomy made in that paper. However, this could be different in other species. In addition, it's a dichotomy that primarily reflects what we understand and what we don't. Click trains with long ICIs are relatively well understood from a functional perspective. Those with short ICIs are still poorly understood and therefore labeled differently (terminal buzzes, burst pulses) with the assumption that they are functionally different somehow. However, as we all know, assumptions should be made with caution...

Answer 4

Although these terms are applied to various species, standard methods for distinguishing these acoustic signals differ depending on the species.

Sperm whales and beaked whales produce echolocation clicks and buzz clicks in trains/ series while foraging. Sperm whale squeals have been compared to burst pulses (Weir et al., 2007) and beaked whales have also been recorded producing burst pulses (Rankin et al., 2011). However, these acoustic signals are very different depending on whether they are being produced by a sperm whale or one of the beaked whale species.

For sperm whales, ICI is often used to distinguish trains of regular/ usual echolocation clicks (ICI 0.2 - 2 s) from buzz clicks (ICI <0.2s) but these values vary depending on factors such as animal size and region. Some analyses may also use changes in amplitude to define the start of a buzz. In Weir et al. (2007) sperm whale squeals were recorded and found to be a series of very rapid pulses with repetition rates of up to 1622 clicks/s.

Here are a few references that discuss the characteristics of these sounds:

Martin et al. 2019. To buzz or burst-pulse? The functional role of Heaviside's dolphin, Cephalorhynchus heavisidii, rapidly pulsed signals. Animal Behaviour 150, 273 - 284. https://www.sciencedirect.com/science/article/abs/pii/S0003347219300089

Miller et al. 2004. Sperm whale behaviour indicates the use of echolocation click buzzes "creaks" in prey capture. The Royal Society 271. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1691849/pdf/15539349.pdf

Wilson et al. 2013. Ultrasonic predator-prey interactions in water - convergent evolution with insects and bats in air. Frontiers in Physiology 4. https://www.researchgate.net/publication/237836543_Ultrasonic_predator-prey_interactions_in_water_-_convergent_evolution_with_insects_and_bats_in_air

Rankin et al. 2011. Description of sounds recorded from Longman's beaked whale, Indopacetus pacificus. The Journal of the Acoustical Society of America 130(5). https://www.researchgate.net/publication/51802675_Description_of_sounds_recorded_from_Longman%27s_beaked_whale_Indopacetus_pacificus

Weir et al. 2007. The burst-pulse nature of ‘squeal’ sounds emitted by sperm whales (Physeter macrocephalus). Journal of the Marine Biological Association of the UK 87(1). https://www.researchgate.net/publication/231979867_The_burst-pulse_nature_of_'squeal'_sounds_emitted_by_sperm_whales_Physeter_macrocephalus

The DOSITS sound library has some good examples which may help you get a sense of how different echolocation click trains are depending on the ICI:

Sperm whale regular click train transitioning to a buzz: https://dosits.org/galleries/audio-gallery/marine-mammals/toothed-whales/sperm-whale/

Beaked whale echolocation click train versus buzz: https://dosits.org/galleries/audio-gallery/marine-mammals/toothed-whales/beaked-whales/

Spinner dolphin echolocation clicks: https://dosits.org/galleries/audio-gallery/marine-mammals/toothed-whales/spinner-dolphin/