Lionfish activity pattern in its native range

Danaé Bregnard, Aurore Wermeille, Sami Zhioua

Abstract

Marine life biodiversity is a major concern in animal conservation. Anthropogenic effects have reduced biodiversity, however alien species invasion have their negative consequences as well. During the last two decades, the red lionfish (Pterois volitans) has invaded the North Atlantic coast to the Caribbean Sea and the Gulf of Mexico. Studies have shown that these species have negatively impacted the abundance and biodiversity of native coral reef fish. However, literature lacks information on the natural behaviour of P. volitans especially in the Indo-Pacific. Therefore, the goal of this study is to analyse the activity pattern of the lionfish in its inborn range. Results show that lionfish are hunting during dawn and dusk. Though, during midday their activity is resumed by a static position, usually hiding under coral heads and rocks. We suggest that lionfish are more active during lowlight periods when they can maximize their hunting success. Further research is needed to investigate whether lionfish are nocturnal hunters and compare the hunting success between the two ranges.

Key words: Pterois volitans, native range, activity pattern, period, stationary, hunting behaviour

Introduction

Alongside with coral bleaching and climate change, invasive species are threatening and weakening coral reefs around the world (Molnar et al. 2008). These marine alien species are affecting not only the environment but the economy, leading to social implications (Bax et al. 2003). Thus, non-indigenous species introduction is a major topic in marine conservation areas. During these last twenty years, the Northern Atlantic coast to the Caribbean Sea and Gulf of Mexico is witnessing the invasion of the red lionfish Pterois volitans (Figure 1). Native from the Red Sea and the Indo-Pacific region (Kulbicki et al. 2012), its introduction in the problematic environments are probably the result of aquaria release (Whitfield et al. 2002).

Figure 1: Evolution of the lionfish invasion in the Caribbean Sea between 1985 and 2014 (Betancur- R et al. 2011)

Lionfish have been described as generalist predators by consuming a wide variety of native coral reef fish (Albins and Hixon 2008; Morris Jr and Whitfield 2009; Morris Jr et al. 2009). In addition, they have the ability to disrupt food webs and displace native species (Whitfield et al. 2007). Therefore, its invasion in the Caribbean reefs has many diverse environmental consequences. A first study has shown that lionfish can cause a higher prey mortality than local predators (Albins 2013) thus perturbing the regulation of population dynamics. The presence of lionfish on a reef will consequently cause a drastic decline in the abundance of small fish populations (Albins 2013). The persistence of prey populations, and indirectly, of predator populations are then endangered. Furthermore, studies suggest that coral reefs shift to algae dominated communities which is correlated with the impacts of the lionfish (Lesser and Slattery 2011).

Lionfish invasion in the Caribbean Sea is based on several biotic and abiotic factors, however it remains unclear whether these effects their success (Anton, Simpson, and Vu 2014). Nowadays, an eradication of lionfish in the Caribbean is utopic (Barbour et al. 2011). To develop effective conservation strategies, a better understanding of the differences between the native range of the lionfish and the invasive range is needed. Lionfish hunting pattern has been described within the invaded range (Green, Akins, and Côté 2011; Cure et al. 2012). However, there is a lack of information of this framework within its native range.

The aim of this project is to assess the behaviour of lionfish during different periods of the day in its native range, the Red Sea. The specific question to answer is: Does the lionfish behaviour vary during the day? Therefore, we test whether there is a difference of the behaviour between the different periods of the day (dawn, midday and dusk). Based on the studies of Cure et al. (2012) and Green, Akins, and Côté (2011), we predict that the proportion of time hunting is higher during lowlight daily periods.

Material and methods

Study sites and data collection

This project was carried out in Egypt, in Ras Mohamed National Park (Figure 2). Observation data were collected between 15-18 April 2018. Ten different sites (Figure 3.) were chosen to avoid recording several times the same individual, thus avoiding pseudoreplication.

Ras Mohamed National Parc location

1/2

Data collection was done by snorkelling during each period of the day. For each period, three snorkelers spent two hours taking data. The data were collected underwater directly through the usage of slates.

Two explanatory variables were considered: the size of the individual (total length in cm) and the period of the day with three levels: dawn, midday and dusk. The size is estimated with an underwater ruler placed next to the fish (this has been done after recording observational data to avoid disrupting its natural behaviour). The lionfishes are then categorized by size into three levels: small (up to 13cm), medium (13 to 18cm) and big (18cm and higher). Data during dawn was recorded between 5:30 to 7:30, the midday period ranges between 10:30 to 12:30 and the evening period ranges between 16:30 to 18:30.

The response variables are specific behaviours of the lionfish. Each individual was recorded for 25 minutes. Every change in behaviour was recorded; all behaviours were recorded in seconds.

The behaviour is divided into three categories:

1

The focal individual is not moving, usually the pectoral fins (PF) run along the body; a non-hunting behaviour. As lionfish are cryptic species (Morris Jr et al. 2009), stationary behaviour might be misinterpreted as hunting, thus it is important to observe if the eyes are looking around suggesting a hunting behaviour.

This behaviour occurred usually while lionfish were under coral heads or rocks (Figure 4).

Stationary (stat)

2

Figure 5: Horizontal spread of lionfish’s fins (A). Angle between the substrate and the lionfish (B). Photos: A.Wermeille, Ras Mohammed National Park (A), Dahab (B)

3 Swimming (swim)

The focal individual has an active movement; the caudal fin moves generating a distortion of the body allowing the lionfish to displace underwater (Sfakiotakis, Lane, and Davies 1999). Usually this behaviour positions the fish horizontally with the substrate (Figure 6).

Figure 4: Stationary phase of a lionfish. The fish is not moving and not targeting or looking for prey. Photo: Aurore Wermeille, Ras Mohammed National Park

Scanning (scan)

Hunting behaviour (Green, Akins, and Côté 2011). This behaviour is characterized with the PF open (Figure 5A), often in an angled position (Figure 5B).

They focus on a specific spot; the position of the body is tilted suggesting that it is potentially targeting prey. They can also maintain this open-PF-position suggesting that thy are searching for prey. This behaviour is the only hunting behaviour recorded in this study

Figure 6: Fins along the body and horizontal position. Photos: S. Zhioua, Ras Mohammed National Park

Statistical analyses

All the tests were performed using R v3.3.2. Using the package “compositions” v1.40-2 (Van den Boogaart and Tolosana-Delgado 2013), we created a matrix of class “acomp” representing the amount of time for each behaviour (using the philosophic framework of Aitchison Simplex). We then created a ternary plot to visually represent the three behaviours between the daily periods and the size classes. Figure 7 shows that the swimming behaviour has rarely been recorded. Thus, we combined behaviours which were clearly not hunting (≠ scanning) into a new variable Not Hunting.

The proportion of time hunting through our data contained an important number of zeros. Therefore, the proportion of time hunting followed a Beta Zero-One Inflated distribution (Figure 7). This distribution created complications in analysing our data. Thus, we decided to make a binary response variable (hunting = 1, not hunting =0). Using methods from Cure et al. (2012), we considered hunting an individual spending more than 50% of the 25 minutes periods hunting. The other 50% of the time was considered not hunting. We tested for the effect of the daily periods using a Generalized Linear Model (GLM) with binomial error, using the package “car”. Post hoc tests were done by comparing the estimates with Tukey methods, using the package “emmeans”. Finally, plot visualisation was done using the package “vigreg”. To compare the mean time Hunting and Not Hunting within each daily period, we used a t test.

Figure 7: Frequency of data collected for the proportion of time hunting between 0 and 1; Beta Zero-One Inflated distribution.

Results

A total of 85 individuals were recorded: 27 during dawn, 32 during the midday period and 26 during dusk. We recorded data on 28 small, 37 medium and 20 big lionfish. The mean size lionfish 15.9 cm, with a standard error of 4.9 cm.

The figure 8 is the ternary plot representing the occurrence of the three behaviours (scanning, stationary and swimming). As mentioned above, the swimming behaviour is rare, which allows to combine “swimming” and “stationary” into a non-hunting behaviour. It is important to keep in mind that not all data appear on the graph, as many overlaps. Thus, many data points are not visible on the plot, hiding behind other data points.

Figure 9 represents the mean proportion time spent hunting during the three different periods of the day which shows a significant difference between each daily period: dawn and midday (emmeans post hoc test: z-ratio = 4.779, p-value < 0.001); dawn and dusk (emmeans post hoc test: z-ratio = 2.598, p-value = 0.0254); midday and dusk (emmeans post hoc test: z-ratio = -3.400, p-value = 0.0019).

We also tested the mean difference within each daily period (Figure 10). Results show the mean differences between hunting and not hunting were significant within dawn (student t test: t = 7.2006, df = 52, p-value = < 0.001) and midday (t = -15.784, df = 60, p-value < 2.2e-16).

Discussion

During midday, lionfish are clearly showing non-hunting behaviours (Figure 10). The mean time not hunting is the highest during midday, two times higher than the second highest mean time not hunting during dusk (Figure 10). This could be explained by the fact that they might be nocturnal individuals (Fishelson 1975). As hunting behaviour has been recorded during dawn and dusk, there is a lack of data between 19:00 and 05:00. Such data might highlight the activity peaks of P. volitans. During dawn, the mean hunting time was the highest, 50% higher than the mean hunting time during dusk. Results show that the mean difference of time hunting and not hunting within dawn was significant; however, this difference within dusk has not been found (Figure 10). This suggests that they are more active during dawn (Cure et al. 2012). Nevertheless, data during dusk was recorded between 16:30 and 18:30. Although light was limited during these hours, lionfish might be shifting from the stationary phase to a hunting phase. Therefore, future research should record hunting data of lionfish later in the day, to assure hunting behaviour observations. Observations could not have been done earlier or later during the day because of the legal conditions of the National Park of Ras Mohamed.

The proportion of time spent hunting was significantly different between the three daily periods (Figure 9). This is in accordance with the studies from Cure et al. (2012). These differences follow our predictions; the highest proportions of time spent hunting are during dawn and dusk. Again, this does not represent the peak of activity of the lionfish. Lack of information might suggest that P. volitans is nocturnal (Fishelson 1975). Furthermore, this study needs a precise ethogram of the behaviour of the lionfish to describe a full framework of P. volitans’ hunting behaviour during dawn and dusk.

Size class did not show a significant effect but nevertheless had a tendency to affect the hunting behaviour (Anova: Chisq = 4.568, df = 2, p-value = 0.1019). However, studies have shown that smaller individuals spend more time foraging than larger individuals (Belovsky and Slade 1986). Throughout our data collection, we observed that smaller lionfish tend to angle in a target position (Figure 5b). Possibly, a more complex data framework might describe such suggestions.

References

Albins, Mark A. 2013. 'Effects of invasive Pacific red lionfish Pterois volitans versus a native predator on Bahamian coral-reef fish communities', Biological Invasions, 15: 29-43.

Albins, Mark A, and Mark A Hixon. 2008. 'Invasive Indo-Pacific lionfish Pterois volitans reduce recruitment of Atlantic coral-reef fishes', Marine Ecology Progress Series, 367: 233-38.

Anton, Andrea, Michael S Simpson, and Ivana Vu. 2014. 'Environmental and biotic correlates to lionfish invasion success in Bahamian coral reefs', PLoS One, 9: e106229.

Barbour, Andrew B, Michael S Allen, Thomas K Frazer, and Krista D Sherman. 2011. 'Evaluating the potential efficacy of invasive lionfish (Pterois volitans) removals', PLoS One, 6: e19666.

Bax, Nicholas, Angela Williamson, Max Aguero, Exequiel Gonzalez, and Warren Geeves. 2003. 'Marine invasive alien species: a threat to global biodiversity', Marine policy, 27: 313-23.

Belovsky, Gary E, and JB Slade. 1986. 'Time budgets of grassland herbivores: body size similarities', Oecologia, 70: 53-62.

Cure, Katherine, Cassandra E Benkwitt, Tye L Kindinger, Emily A Pickering, Timothy J Pusack, Jennifer L McIlwain, and Mark A Hixon. 2012. 'Comparative behavior of red lionfish Pterois volitans on native Pacific versus invaded Atlantic coral reefs', Marine Ecology Progress Series, 467: 181-92.

Fishelson, L. 1975. 'Ethology and reproduction of pteroid fishes found in the gulf of Aqaba (Red Sea), especially Dendrochirus brachypterus (Cuvier),(Pteroidae, Teleostei).[Conference paper]', Pubblicazioni della Stazione Zoologica, Napoli.

Green, Stephanie J, John L Akins, and Isabelle M Côté. 2011. 'Foraging behaviour and prey consumption in the Indo-Pacific lionfish on Bahamian coral reefs', Marine Ecology Progress Series, 433: 159-67.

Kulbicki, Michel, James Beets, Pascale Chabanet, Katherine Cure, Emily Darling, Sergio R Floeter, René Galzin, Alison Green, Mireille Harmelin-Vivien, and Mark Hixon. 2012. 'Distributions of Indo-Pacific lionfishes Pterois spp. in their native ranges: implications for the Atlantic invasion', Marine Ecology Progress Series, 446: 189-205.

Lesser, Michael P, and Marc Slattery. 2011. 'Phase shift to algal dominated communities at mesophotic depths associated with lionfish (Pterois volitans) invasion on a Bahamian coral reef', Biological Invasions, 13: 1855-68.

Molnar, Jennifer L, Rebecca L Gamboa, Carmen Revenga, and Mark D Spalding. 2008. 'Assessing the global threat of invasive species to marine biodiversity', Frontiers in Ecology and the Environment, 6: 485-92.

Morris Jr, James A, JL Akins, A Barse, D Cerino, DW Freshwater, SJ Green, RC Muñoz, C Paris, and PE Whitfield. 2009. 'Biology and ecology of the invasive lionfishes, Pterois miles and Pterois volitans'.

Morris Jr, James A, and Paula E Whitfield. 2009. 'Biology, ecology, control and management of the invasive Indo-Pacific lionfish: an updated integrated assessment'.

Sfakiotakis, Michael, David M Lane, and J Bruce C Davies. 1999. 'Review of fish swimming modes for aquatic locomotion', IEEE Journal of oceanic engineering, 24: 237-52.

Van den Boogaart, K Gerald, and Raimon Tolosana-Delgado. 2013. Analyzing compositional data with R (Springer).

Whitfield, Paula E, Todd Gardner, Stephen P Vives, Matthew R Gilligan, Walter R Courtenay Jr, G Carleton Ray, and Jonathan A Hare. 2002. 'Biological invasion of the Indo-Pacific lionfish Pterois volitans along the Atlantic coast of North America', Marine Ecology Progress Series, 235: 289-97.

Whitfield, Paula E, Jonathan A Hare, Andrew W David, Stacey L Harter, Roldan C Munoz, and Christine M %J Biological Invasions Addison. 2007. 'Abundance estimates of the Indo-Pacific lionfish Pterois volitans/miles complex in the Western North Atlantic', 9: 53-64.