Impact of observers on cleaning stations

Does the presence of divers have an effect on diversity, abundance and size of clients at cleaner wrasse (Labroides dimidiatus) cleaning stations?

Delisle Lorraine, Levorato Elena & McNeely William

Introduction

Despite the fact that it is hard to manipulate environmental factors, watching animals in their natural environment is a rich source of information regarding their behavior. However, for small and/or mobile species it often requires direct observation. We could thus wonder if the presence of an observer has an impact on the studied species. We are here focusing on the Blue streak cleaner wrasses (Labroides dimidiatus) and their clients. The cleaner wrasse is a world widespread coral reef species, in tropical waters (Humann et al., 2015). The blue streak cleaner wrasse is a territorial species, living on their “cleaning station”. They clean other fish species, called clients, by removing their ectoparasites (Losey et al., 1999). Thereby cleaner fish are offering services and must respond to the market rules (Bshary and Grutter, 2002a), according to the clients’ abundance and ecological range (demand), and the amount and proximity of other cleaners (offer). According to the ecological range (Bshary and Grutter, 2002b) of the clients, they can choose the cleaning station, according to the service quality. Cleaner wrasse is a model species in interspecific interaction studies. To study theses interactions one needs to observe cleaning stations in the wild. Direct observation remains the most relevant way of assessing behavior. However, we could wonder whether the presence of an observer has an impact on the fish diversity and abundance around these cleaning stations. We are also looking at the client size, which is relative to their ecological home range (Bshary & Schäffer, 2002). The investigation is crucial to be assessed as cleaner behavior depends on the clients’ presence and behavior and thereby could cause bias in many studies.

Research question and hypothesis

After the results of previous observations (Colosio et al. 2015, unpublished data, Egypt excursion) and other studies (Hawkins et al., 1999; Titus et al., 2015), we predict that the presence of a diving observer at a cleaning station can influence diversity, size and abundance of the clients present at a cleaning station. We can assume that in the presence of a diver, the diversity, the abundance and the size of clients will decrease compared to a situation where there is no bystander. The purpose of our project is to study whether the presence of observers (scuba divers) influences the diversity of potential client fishes around cleaning stations.

Material and methods

Site and subjects

The study took place in the Red Sea, on the Abu Sauatir Bay, 14km north of El Quseir, Egypt, from the 6th to the 9th of June 2016 (Figure 1). Before starting the treatments, we did a few dives to find suitable cleaning stations occupied by a female individual in its adulthood. We marked the stations with a colored plastic strap (best in yellow) which was removed once the experience finished. On the 23 cleaning stations initially filmed, only 19 were used for the analysis. Four stations were not taken into account due to terrain constraints (inadequate angle of camera preventing the observation of fish).

Data collection

We used four cameras (GoPro Hero3, Hero3+ and two Hero4) to film the cleaning stations, all of them set on “broad view”. During each dive, two divers installed a GoPro near the chosen cleaning station, including most of the range covered by the cleaner wrasse. Each cleaning station was exposed to two treatments: with and without observer (O vs. NO). The observer treatment consisted on the presence of two scuba divers, in semi static position, just behind the GoPro. The treatment without observer consisted in setting the camera, leaving the area and coming back after 20 min. The videos lasted respectively 15 minutes (O) and 20 minutes (NO). This difference is to compensate the time the divers need to get away from the station and the time they spend to reach again the camera to stop the video during the treatment without bystanders. To avoid potential bias due to order of the treatments we randomized them, but always contiguously one after the other (i.e. station 1: 20 min without observer then 15 min with observer; Station 2: 15 min with observer then 20 min without observer). Likewise, cameras were left untouched.

Figure 2: Cleaning station with cleaner fish, clients and the yellow strap allowing the identification of the station.

Figure 1: Map showing the location where the experiment took place. Pic. by Eric Gaba.

Data extraction

To analyze the videos afterwards we did not take into consideration the first 3 min, so to let the fish acclimatize after our departure (NO) or to our presence (O). Furthermore, we analyzed only 12 minutes of each videos. Resident species always present on the screen were counted only once (maximal number of individuals observed at the same time). Conversely, the other fishes were counted every time they entered the screen and added to the data as new individuals.

Fish families and species

To identify species and families but also to assess the mean size of the clients we used the following guide book: Coral Reef Guide, Red Sea, 2004 Ed. Collins, Ewald Lieske & Robert F. Myers.

Statistical analyses

Data was analyzed with R software version 3.2.2. A normal distribution of residuals allowed the use of a paired t-test to compare the data of diversity. Abundance and size were assessed with Wilcoxon test because of their non-parametric data.

Results and discussion

The results demonstrated a significant impact of divers on the number of families (Figure 3a; p-value = 0.0077) as well as on the number of species (Figure 4a; p-value = 0.0003). However, the abundance and the size of clients were not significantly different for the two treatments (p-values of 0.246 and 0.1005 respectively). As an extra analysis, the number of interactions was also assessed with no significant differences found between the two treatments (p-value = 0.9591).

We found that the diver presence has a negative impact on the diversity. Indeed, more species and families were observed without observers for almost every station (Figure 3b and 4b). According to Colosio et al. (2015, unpublished data), abundance and size were expected to be also higher when no observers were present. However, we did not find a significant difference between the two treatments, maybe due to our method of analysis. A bias could consist in the presence of shoals of fish passing, independently of the treatment. Moreover, visitor clients leaving the screen and coming back, were calculated as different individuals. Thus, the effective number of individuals was certainly different as some of the fish may have been the same and counted several times as different individuals.

Conclusion

Our results show that the presence of observers influence significantly fish diversity around cleaning station with no apparent impact on fish abundance and size. Thus, our first hypothesis was only partially confirmed. The results differ from previous observation (Colosio et al., 2015, unpublished data) where both the abundance and the size were differently correlated with the presence or absence of observers. However, a decrease of fish diversity in the presence of bystanders has been, once again, confirmed. Further investigations should still be done to establish which is the actual effect on abundance, size and cleaner-client interactions.

Acknowledgments

We would like to address a special thank to our dive instructor Hans Lange and Lucy Deans who helped us a lot and without whom the project would not have been possible. We also wish to thank Dominique Roche, Zegni Triki and Sandra Binning whom suggested us this interesting project and helped us structure, organise and analyse our data.

References

Bshary, R., and Grutter, A.S. (2002a). Experimental evidence that partner choice is a driving force in the payoff distribution among cooperators or mutualists: the cleaner fish case. Ecol. Lett. 5, 130–136.

Bshary, R., and Grutter, A.S. (2002b). Asymmetric cheating opportunities and partner control in a cleaner fish mutualism. Anim. Behav. 63, 547–555.

Bshary, R., and Schäffer, D. (2002). Choosy reef fish select cleaner fish that provide high-quality service. Anim. Behav. 63-3, 557–564.

Colosio, S., Demairé, C. and Quattrini, F. (2015). Impact of observers on fish diversity and size at common cleaner wrasse (Labroides dimidiatus) cleaning stations. On: http://marine-bio.wixsite.com/egypt-field-course/project-2. Consulted on May 24th 2017.

Humann, P., DeLoach, N., Allen, G. and Steene, R. (2015). Reef Fish Identification Tropical Pacific 2nd Edition. New World Publications

Hawkins, J.P., Roberts, C.M., Hof, T.V., De Meyer, K., Tratalos, J., and Aldam, C. (1999). Effects of Recreational Scuba Diving on Caribbean Coral and Fish Communities. Conserv. Biol. 13, 888–897.

Lieske, E. and Myers, R.F. (2004). Coral Reef Guide, Red Sea. Ed. Collins.

Losey, G.S., Grutter, A., Rosenquist, G., Mahon, J.L., and Zamzow, J. (1999). Cleaning symbiosis: A review (ISPA). In V.C. Almada, R.F. Oliveira and E.J. Goncalves (Ed.), Behaviour and Conservation of Littoral Fishes (pp. 379-395) Lisboa, Portugal: ISPA.

Titus, B.M., Daly, M., and Exton, D.A. (2015). Do Reef Fish Habituate to Diver Presence? Evidence from Two Reef Sites with Contrasting Historical Levels of SCUBA Intensity in the Bay Islands, Honduras. PLOS ONE 10.