Experiment Details: Kalahari Meerkat Project

Megan Wyman

Department of Evolutionary Biology and Environmental Studies, University of Zurich

megan.wyman@ieu.uzh.ch

Not provided

Marta Manser, marta.manser@ieu.uzh.ch

Data curation, Investigation

Yes

Yes

Kalahari Meerkat Project

Evolution of cooperative behavior and vocal diversity in meerkats

Kalahari Research Centre at the Kuruman River Reserve, in the Northern Cape Province, South Africa (26˚58’ S, 21˚49’ E).

This project is ongoing but the recording samples provided were recorded between 2006-2021

Meerkat recordings were conducted on habituated but free-living meerkats at the Kalahari Research Centre, South Africa. Most recordings were conducted while the group was foraging. During this activity, the meerkats move slowly through the landscape while searching for food in the ground. Meerkats vocalize frequently during foraging to maintain group cohesion and during social interactions such as food competition, dominance interactions, pup begging, or antipredator warnings. Some recordings were also focused on individuals who were on sentinel duty, i.e., meerkat is in a standing position and scanning for predators while the rest of the group forages. Each recording is focused on one target individual (i.e., the focal individual). Each recording has two channels: one channel is a directional microphone pointed at the focal animal at a distance of 30-100 cm, and the other channel is a handheld microphone in which the human recordist narrates the ongoing recording. At the start of each recording, the recordist provides a variety of metadata about the recording, including the name of the focal individual, the group name, the date and time, etc (however, the clips you hear may not include this starting annotation). During the recording, the recordist tries to signal each time the focal individual vocalizes by saying the call type and the behavioral context of the call (or simply saying ‘again’ if the information is the same as the last call). As a note, the most common call, the close call, is often abbreviated to “CC” in the human narration. They will also occasionally note the distance and identity of the nearest neighbor (if known) or the behavior of the group in general.

Individual meerkats were recorded with a Sennheiser directional microphone (ME66, with K6 powering module) connected to Marantz solid-state recorder (PMD661 or PMD670, Japan Inc.; sampling frequency 48 kHz, 16 bits accuracy). Human annotations of meerkat behavior were recorded concurrently on the Marantz recorder on a second channel using a handheld microphone (Joseph E-280 Dynamic Microphone, 40 HzÐ20 kHz, (+/- 2.5 dB)).

This is just a small example: Driscoll, I., Briefer, E. F., & Manser, M. B. (2024). The role of neighbour proximity and context on meerkat close call acoustic structure. Animal Behaviour, 212, 113–126. https://doi.org/10.1016/j.anbehav.2024.03.021 Manser, M. B. (2001). The acoustic structure of suricates’ alarm calls varies with predator type and the level of response urgency. Proceedings of the Royal Society B: Biological Sciences, 268(1483), 2315–2324. https://doi.org/10.1098/rspb.2001.1773 Manser, M. B., Jansen, D. A. W. A. M., Graw, B., Hollén, L. I., Bousquet, C. A. H., Furrer, R. D., & le Roux, A. (2014). Vocal Complexity in Meerkats and Other Mongoose Species. In Advances in the Study of Behavior (Vol. 46, Issue June, pp. 281–310). https://doi.org/10.1016/B978-0-12-800286-5.00006-7 Rauber, R., & Manser, M. B. (2017). Discrete call types referring to predation risk enhance the efficiency of the meerkat sentinel system. Scientific Reports, 7(44436), 1–9. https://doi.org/10.1038/srep44436 Wyman, M. T., Rivers, P. R., Muller, C., Toni, P., & Manser, M. B. (2017). Adult meerkats modify close call rate in the presence of pups. Current Zoology, 63(3), 349–355. https://doi.org/10.1093/cz/zox029

Suricata suricatta

Suricata suricatta (meerkats) and Homo sapiens (humans) as the vocal annotators of meerkats

Non-meerkat vocalizations or sounds include various bird and insect species, as well as humans (typically only the recordist). In the annotation platform, please label as Species = ‘Unknown’, ‘Human’, ‘Bird’, or ‘Insect’, and Individual = ‘Unknown’, ‘Recordist’, etc.. Since birds and insects are not the species of interest in these recordings, you can just use the term ‘Bird’ or ‘Insect’ for a general species label, unless you are very sure of the species. For instance, if you hear unknown birds, you would label each vocalization as Species = ‘Bird’ and Individual = ‘Unknown’. For the human recordist, your label would be Species = ‘Human’ and Individual = ‘Recordist’. You may also hear mechanical or electrical noise from the recording equipment. Examples of the typical noises heard on the recordings can be viewed in Figure M3. However, you do not need to annotate these non-animal sounds. Tip: Listening to the human recordist channel may help identify some sources of noise seen in the meerkat channel (e.g., recordist’s narration).

Meerkats are small, slender mongooses (0.6-1 kg, 25-35 cm long body) with long, thin tails with black tips. They have brownish-gray fur with dark bands across the back. They have powerful, curved claws for digging and can stand upright, using their long tails for balance.

Meerkats, a cooperatively breeding social mongoose from southern Africa, rely on a complex vocal system of over 30 different call types for survival and group cohesion. Living in social groups of 3-50 individuals, their vocalizations include alarm, recruitment, and social calls, essential for coordinating activities, signaling danger, maintaining dominance hierarchies, and strengthening social bonds. Many of these call types have distinct acoustic features suited to various purposes, from alerting group members to specific predator types or threat levels to organizing group foraging. Meerkat call types range from common close calls (used in group coordination and social monitoring) to move and lead calls (used for group coordination and movement) to short notes (often used during anti-predator vigilance where they are organized in bouts of 1 to 4 calls, but can be used in other contexts) to alarm calls (referential system providing information on predator type and urgency level) to aggressive and submissive calls. Pups also produce call types specific to the younger age classes which are performed during behaviours like begging, digging, and following. Hybrid calls are possible and involve a gradation or mix between different call types (e.g., hybrid close call + move call). Please see Figure M2 for an overview of the meerkat vocal repertoire. Meerkats primarily produce vocalizations in the larynx which are then shaped by the rest of the vocal tract before exiting the nose and/or mouth. Many vocalizations are produced with a closed mouth. Frequency = median fundamental frequency of ~700 Hz (range: ~150 Hz to 1500 Hz). Duration = median duration of ~160 ms (range: ~ 50ms to several seconds) In the annotation platform, the meerkat channel will be the top waveform and spectrogram while the human recordist channel will be the bottom waveform and spectrogram. Please make all annotations of animal sounds in the meerkat channel only. This includes all annotations of all non-meerkat sounds, like humans, birds, insects, etc. To see an example of what a properly labeled meerkat recording looks like in the annotation platform, please see Figure M1. Please see Figure M2 for an overview of the meerkat vocal repertoire. Please try to label the vocalization type as accurately as you can, based on the information in Fig. M2. For instance, Vocalizations = ‘Close call’, or ‘Short Note’, or ‘Beg’, etc. If you are not sure which call type it is, please use Vocalizations = ‘Unknown.’ For non-meerkat vocalizations, label the bird and insect sounds as Vocalizations = ‘Unknown’ and the human sounds as Vocalizations = ‘Annotation’ or ‘Unknown’.

Semi-arid arid landscape in the southern Kalahari Desert, featuring wind-blown dunes, dry riverbeds, and flats containing a mixture of sparsely distributed and patchy vegetation including multiple grass, shrub, and tree species.

The mean is probably 2-3, with a range of 1-10 possible.

I will annotate all individuals

No

https://drive.google.com/open?id=1PJc-a5AHCuM_Y6Ombwf7uUo7eKPLCME3

Randomly chose a file