Southern right whales generally appear not to react to transiting research vessels

1 INTRODUCTION

In both, aquatic and terrestrial environments, animal-vehicle collisions represent serious concern (Kioko et al., 2015). Although some authors have studied animal-vehicle collisions involving different species of animals, including mammals, birds, amphibians, and reptiles (Bíl et al., 2017; Kioko et al., 2015; Lao et al., 2011; Lima et al., 2015; Sullivan, 2011), little is known about the animal's behavior when a collision is imminent (Lima et al., 2015).

Collisions between ships and whales are regularly reported worldwide (Campana et al., 2015; Laist et al., 2001; McKenna et al., 2015; Nowacek et al., 2004; Ritter 2012) and for some endangered species, ship strikes are a major threat to their survival (Berman-Kowalewski et al., 2010; Clapham et al., 1999; Douglas et al., 2008; Laist et al., 2001; Neilson et al., 2012). How marine mammals respond to avoid a vessel collision varies among species. According to Lima et al. (2015), animals generally initiate evasive behavior against an imminent collision by using responses similar to their antipredator behavior. Dolphins, for example, change their heading and speed when approached by a fast-moving watercraft or erratically moving jet skis (Nowacek & Wells, 2001), whereas killer whales use antipredator-like evasive maneuvers, changing path and speed (Williams et al., 2002). In contrast, North Atlantic right whales (Eubalaena glacialis), manatees (Trichechus manatus), and dugongs (Dugong dugon) are fairly unresponsive to marine vessels, which suggest they are more susceptible to collisions (Lima et al., 2015).

In the case of baleen whales, their large size reduces the number of natural predators. This might explain why whales are prone to ship collisions (Lima et al., 2015). To our knowledge, there is no information on potential evasive actions taken by right whales when a collision is imminent (Lima et al., 2015). North Atlantic right whales do not respond to sounds emitted by large vessels, despite being within their audible range, only responding when large vessels are close (Nowacek et al., 2004). Perhaps this is because they are habituated to these vessel sounds (Laist et al., 2001; Nowacek et al., 2004; Terhune & Verboom, 1999). Another possible explanation for this is related to possible sound attenuation due to the water physical properties of the water or of sound propagation (Nowacek et al., 2004). In the case of southern right whales (SRWs) they react to whale-watching boats only at close proximity (Argüelles, Coscarella, et al., 2016).

McKenna et al. (2015) investigated how blue whales (Balaenoptera musculus) respond to close approaches of large commercial vessels in a busy shipping lane that overlaps with important blue whale feeding grounds. The results suggest that blue whales exhibit a behavioral response when ships are within close range (<100 m) by either diving to deeper depths when at the surface or delaying surfacing after a foraging dive. This limited response likely makes the whales more susceptible to collisions with large vessels.

Several studies have shown that whales exhibit behavioral changes in response to whale-watching boat traffic (Corkeron, 2004; Parson, 2012). In this sense, some authors have documented reactions of SRWs related to whale-watching boats (Argüelles, Coscarella, et al., 2016; Arias et al., 2018; Rivarola et al., 2001) but there are no quantitative data showing how these whales react to vessels that are not whale-watching (e.g., commercial ships, pleasure boats) but just transiting past the whales. The area inside Golfo Nuevo on the coast of Puerto Madryn city, has the highest shipping density, considered the gateway and the base of the service for the tourism industry in Patagonia, whose main attraction is the visit to the World Heritage Site Península Valdés (UNESCO 1999). The commercial vessel traffic associated with the docks of Puerto Madryn city is categorized mainly as cargo ships (e.g., container ships, bulk carriers) and fishing boats (e.g., trawlers, jiggers), traveling at an average speed of 10 knots. Additionally, this area is intensively used for water activities, including sailing, diving, kayaking, windsurfing, kitesurfing, and recreational fishing.

Golfo Nuevo represents the main breeding ground for southwestern Atlantic SRWs (Bastida et al., 2007; Rowntree et al., 2001). Península Valdés is known internationally as one of the most important breeding areas for this species in the southwestern Atlantic Ocean (Bastida et al., 2007). In addition, it is considered one of the best places in the world to watch SRWs, due to the large number of animals as well as to their predictability and proximity to the coast (Fazio et al., 2015). The annual population growth rate of SRWs is around 3% (Crespo et al., 2019) and the breeding population of Península Valdés has been estimated in around 4,000 whales (Cooke, 2012). The number of whales peaks in September, with up to 1,000 individuals on the Patagonian coasts (Crespo et al., 2019), and between August and September more than 200 whales can be watched in Bahía Nueva, Golfo Nuevo (Argüelles, 2017).

SRWs are protected by the Bonn Convention on Migratory Species of Wild Animals (CMS), integrating its Appendix I (Migratory Endangered Species), the International Whaling Commission, and the Convention on International Trade in Endangered Species (CITES; Appendix 1), which prohibits any international trade action involving SRW. The International Union for Conservation of Nature (IUCN) has included SRWs in its Red Book under the category of “Low Risk - Dependent on Conservation.” SRWs receive special protection in Brazil and Argentina, being the first Natural Monument in Argentina (Law N° 23094 of 1984). Also in Brazil, in the state of Santa Catarina, SRWs are considered a Natural Monument (decree N° 171 of 1995). Moreover, SRWs receive protection through the creation of protected areas where the whales are at their highest densities. In Argentina, San José Gulf was declared Marine Park and Peninsula Valdés was declared Nature Tourist Reserve with an Integral Objective (Law N° 2161 of 1983). On the other hand, Nuevo and San José gulfs were declared World Heritage Sites by UNESCO in 1999 and more recently as Biosphere Reserve in 2014 due to their value as reservoirs of biodiversity.

The International Whaling Commission endorsed the Southwest Atlantic Southern Right Whale Conservation Management Plan (CMP) in 2012, with Argentina, Brazil, Chile, and Uruguay being the signatories to it. The overall goal of the CMP is to protect habitat for the population and minimize anthropogenic threats to maximize population recovery. Several workshops have been held (66/CC12 Agenda ítem 7.2.2), including on calf mortality and possible links with kelp gull harassment in nursery areas, and on identification of sensitive areas in Argentine waters (SC/66a/Rep/8). In a workshop held in 2014, the authors described recent findings about the areas shared by SRWs and ships close to Puerto Madryn in Argentina, which could generate space use conflicts. Workshop participants agreed that recreational boats, including kayaks, needed regulation and it recommend continued work to improve mooring.

The increase in whale numbers and human activities in Puerto Madryn generates conflicts over the shared use of space, increasing the risk of collisions (Argüelles, Fazio, et al. 2016). To date, only two collisions have been reported in the area: one in 2008 and the other in 2010. The first one involved a juvenile SRW and an Argentine Navy destroyer (International Whaling Commission, 2010) in the port. The second occurred between a whale and a container ship sailing towards the port. Although ship collisions have yet to become a major threat for SRWs in the area, the collisions recorded, the growing population, and the increasing number of whales showing signs of impact damage and cuts from propellers warn us about the need to regulate maritime traffic in the area. Understanding SRWs respond to marine vessels is key to developing effective regulations. In this context, the objective of this study was to investigate the reaction of SRWs to a transiting research vessel.

2 METHODS

Data were collected off the coast of Puerto Madryn city (Patagonia, Argentina), following a line transect sampling during three consecutive breeding seasons between 2013 and 2015 (Figure 1). Surveys were undertaken on board a rigid-hull research boat 5.60 m long with a 60 HP outboard engine and was travelling at an average speed of 6 knots. The number of whales, type of group, perpendicular distance between whales and the transiting research vessel, and reaction of whales to the research vessel were recorded. Whale groups (cat) were classified according to their composition as: mother with calf (a mother and a calf born during the season - MC), solitary individual (a single juvenile or adult - SI), or mating groups (one female and up to six males - MG). Although SI was defined as a single juvenile or adult whale, some associations between individuals were considered in this category. So, the SI category may include more than one individual. Similarly, associations between different MC pairs, MC with a single calf, or MC with a single juvenile were considered in the MC category.

Data of SRW behaviors were collected before, during, and after the encounter with the transiting research vessel. Prior to the encounter, SRW behaviors were classified as “surface activity,” “traveling” (when whales were swimming), or “resting” (when whales were logging). Surface activity included active behaviors such breaching (a breach is a genuine jump in which part of the animal body clears the water), tail slapping (when a whale positioned itself vertically downwards into the water, lifted its tail out of the water and then brought it down onto water surface hard and fast), and pectoral slapping (when a whale turned on its side exposing one pectoral fin into the air and then slapped it against water surface). These behaviors constituted the whales' initial behavior and were recorded from a distance >150 m by the observers onboard the transiting research vessel. This distance was set based on previous studies that suggest whales react to boats only when they are very close (Argüelles, Coscarella, et al., 2016). Also, in the case of SRW, a provincial law regulates that boats could not approach at distances <300 m at a speed of 10 knots. The maximum speed at a 50 m distance is 5 knots. These regulations are mainly aimed at whale-watching vessels, although they apply to any transiting vessel (Argüelles, 2017; Argüelles, Coscarella, et al., 2016).

When an encounter occurred, SRW behaviors in response to the transiting research vessel were categorized using descriptions provided in Argüelles, Coscarella, et al. (2016) and were based on the whales' initial response. The categories were: (1) approaching (AP), when the whale's behavior changed to approach the transiting research vessel to interact; (2) neutral (N), when the whale was indifferent to the transiting research vessel; (3) avoidance (A), when the whale's behavior changed by actively moving away from the transiting research vessel, avoiding interaction, or (4) submerged (S), when the whale's behavior changed by actively moving underwater away from the transiting research vessel, outside the observer's view. To confirm that the whale behavior was in response to the transiting research vessel and not a natural transition in behavior, after each encounter a trained person onboard the research vessel remained watching the whale. SRWs tend to remain in the same behavioral state for quite a while. In San Matías Gulf, Vermeulen et al. (2012) reported that whales usually resume their previous behavioral state soon after the interaction with the boat is over, reinforcing the idea that whales change their behavior due to the presence of the boat. Similar results have been previously reported in Argüelles, Coscarella, et al. (2016).

The perpendicular distance between whales and the transiting research vessel was recorded with reticular binoculars (Bushnell Marine 7 × 50) and a telemeter (Bushnell, model Yardage Pro Legend Scout Laser Rangefinder, range: 15–930 yards, 6 × 23 mm). All the behaviors recorded in this study were in response to the research vessel transiting at 6 knots. Whales' responses to other vessels transiting the area were not recorded. Vessel size, type, and speed were not taken into consideration because they were control data.

To evaluate the whales' response behaviors to the transiting research vessel in Bahía Nueva, data were analyzed using cumulative logit models (CLM; Agresti, 2013), with a log–log link function. CLMs were adjusted for ordinal data analysis, using the whales' response behavior as an ordinal variable of response intensity. The behavior of whales in response to the transiting research vessels was classified according to its intensity, as follows: submerged (S), avoidance (A), neutral (N), and approaching (AP), with being submerged the most negative behavior and approaching the most positive behavior.

CLMs were built using the MASS library and ordinal package (Christensen, 2015) of the R 3.1.3 software (R Development Core Team, 2014) for the Poisson family, and a plot of deviation from the independence model was created using the loglm package (R Development Core Team, 2014). The null model was constructed considering the independence between the behavioral reaction of the whales (S, A, N, AP) and the type of group (cat = MC, SI, MG), number of whales (No.) and distance between whales and the transiting research vessel. Models incorporate variables sequentially, starting with first-order interactions, and then second-order interactions. The best model fit was evaluated using the Akaike information criterion (AIC) (Burnham & Anderson, 2002). The best fit model is the model that minimizes the number of parameters while increasing the ability to predict the observed frequencies. Model comparisons were made with ΔAIC, which indicates the magnitude of the difference in AIC values between each model and the model best supported by the data. Models with ΔAIC ≤2 were considered as candidate models (Burham & Anderson, 2002). Finally, a multimodel inference approach was performed using the functions model.sel and model.avg from the MuMIn package (Burham & Anderson, 2002; Kamil, 2019) of the R 3.1.3 software (R Development Core Team, 2014).

3 RESULTS

A total of 34 trips were undertaken in the surveyed area between 2013 and 2015. The sampling area surveyed covered 42.76 km² and the sampling effort in transects was 273.42 km in 2013, 221.74 km in 2014 and 476.98 km in 2015. A total of 971 encounters were recorded. Most of whale groups (70.14%) consisted of one or two individuals. Surface activity was the initial behavior category more often observed (58.40%), followed by traveling (19.70%) and resting (21.90%).

In order to investigate the reaction of SRWs to the transiting vessel, the models were sequentially tested including all interactions among variables. Models that included all significant factors were selected using AIC (Table 1). The only model with main effects of one variable that presented an AIC value to be considered as plausible was the one that considered number of whales as predictor variable. As the models including number of whales (No.) and category (cat) had a ΔAIC ≤2, to consider the variation among models a multimodel approach was performed. Therefore, all the models that included the number of whales and category were considered as candidate models. Coefficients revealed a positive effect of number of whales and category (Table 2). The best supported model that explained the whales' response behavior to a transiting research vessel in Bahía Nueva was the additive model that considered number of whales (No.) and category (cat) as variables. This model was significantly better than null model (AIC = 704.99).

The response of SRWs to the transiting research vessel differed depending upon the number of whales in the group and the type of group. The probability of responding neutrally was the most common response for all the categories analyzed and the probability of responding negatively (submerging or moving away from the research vessel) only increased for SI and MC groups (Figure 2). Mating group reactions were mainly neutral regardless of the number of whales in the group and SI and MC groups exhibited negative response behaviors.

4 DISCUSSION

Wildlife can react in response to human activities in three different ways: positively, negatively, or neutrally (Whittaker & Knight, 1998). SRWs in the Argentine Patagonia respond in all three ways to a transiting research vessel, by reacting negatively (avoiding contact by moving away from the vessel or submerging), positively (approaching the vessel and seeking contact), or neutrally (no surface response), depending on the number of whales comprising the group. In a previous study we evaluated the response of whales to whale-watching boats and demonstrated that SRWs perform more avoidance behaviors when the approach of whale-watching boats is inappropriate and the engines remain on, and approach to closer distances when the engines are turned off (Argüelles, Coscarella, et al., 2016). Although the sailing conditions of whale-watching boats and a research observational vessel are different, our current results differ from those obtained in our previous study (Argüelles, Coscarella, et al., 2016). In the former, we found that whales tend to approach whale-watching boats under some circumstances. When boat engines were off, whales tended to perform more approaching behaviors. Conversely, during an inappropriate approach, whales showed a tendency towards more avoidance behaviors when the engines were on and fewer avoidance behaviors when the engines were off. Furthermore, when the approach was inappropriate whales tended to display neutral behaviors more often than expected when the engines were on, and fewer neutral behaviors while the engines were off. In the current study, most whales showed neutral surface behavioral responses to a transiting research vessel, suggesting that SRWs may be vulnerable to collisions with vessels especially in areas with high vessel traffic. The lack of reaction to the transiting research vessel, compared to surface behavioral response to whale-watching boats, indicates that SRWs respond to vessels that alter direction and speed. A possible explanation might be linked to the moderate maritime traffic in the study area, with no whale watching approaches. However, this lack of response is not automatically equated with a lack of sensitivity to the disturbance (Beale & Monaghan, 2004; Bejder et al., 2006).

Marine vessels are known to cause significant behavioral reactions in whales and several studies have recorded effects of whale-watching vessels on short-term behavioral responses of whales, including changes in surface behavior (Corkeron, 1995; Noren et al., 2009; Stamation et al., 2010; Williams et al., 2009), acoustic (Foote et al., 2004; Richter et al., 2006; Sousa-Lima & Clark, 2008), and swimming behavior (Stamation et al., 2010; Williams & Ashe, 2007; Williams et al., 2002, 2009). Some species of whales seem to be more sensitive to moving boats. Richardson et al. (1985) reported differential responses of bowhead whales (Balaena mysticetus) to fast erratic boats and slower linearly traveling vessels. The fleeing response of bowhead did not persist for longer after the boat moved away, but increased spacing and probably social disruption sometimes continued longer (Richardson et al., 1985). Stamation et al. (2010) found that responses of migrating humpback whales (Megaptera novaeangliae) to whale-watching vessels are variable and calf pods were more sensitive to the presence of vessels than noncalf pods. It seems unlikely that the lack of response of SRWs to transiting vessel might be due their inability to detect the vessel in time to avoid the collision. The lack of reaction of SRWs to the transiting research vessel could be a simple consequence of SRWs habituation to the presence of moving vessels and that they do not interpret a transiting vessel as a threat (Bejder et al., 2009). Besides this, SRWs could be responding in a different way to transiting vessels when they are distracted by their own activities, which sometimes include social behaviors (M.B.A., unpublished data; Arias et al., 2018). This study has focused on the reaction of SRWs to transiting research vessel by the surface behavioral analysis. However, it is important to note that the lack of responses could be related to physiological or vocal behaviors since maritime vessel traffic may impact the acoustic behavior of SRWs. Vessel noise is recognized as an important source of pollution in aquatic environments within populated areas around the world, and represents an environmental disturbance that may have significant adverse effects on marine mammals. A growing number of studies have investigated the impact of the vessel noise on cetaceans (Jensen et al., 2009; Lusseau et al., 2009; Noren et al., 2009; Sousa-Lima & Clark, 2008; Weinrich & Corbelli, 2009). Some studies have shown changes in the vocal behavior of cetaceans (Buckstaff, 2004; Marega-Imamura et al., 2018; Scarpaci et al., 2000) providing evidence that maritime vessels are a significant source of anthropogenic noise that could influence cetacean behavior. Stamation et al. (2010) found that the humpback whale mother with calf was the most sensitive group to vessel noise. Noise generated from vessels is generally a function of the vessel's size and speed. All marine noise will not have the same impact on whales, and how whales are affected by noise will depend on the frequency and intensity (dB) of the noise. In this research, a small vessel with 60 HP outboard engine traveling at 6 knots is not representative of marine noise. To assess the effect of noise on whale behavior, a combination of data from different vessel types and speed as well as engine types (inboard or outboard) is required to describe the underwater noise. Ellison et al. (2011) argued that the three aspects of sound exposure (context, relative sound level, and chronic noise) mediate behavioral response, and suggested they be integrated into ecosystem-level management and the spatial planning of human offshore activities.

In Península Valdés, SRWs usually give birth to their young in calm and shallow coastal waters, and, by the end of the whale season (October onwards), they move away from the coast to deeper waters where they perform long dives in an apparent feeding behavior (Argüelles, Fazio, et al., 2016). The deep waters of Golfo Nuevo are also used by ships and vessels as a route in and out the marine port. While some studies have demonstrated the surface feeding behavior leaves right whales vulnerable to collisions (Laist et al., 2001; Parks et al., 2011), deep feeding behavior may also increase the risk of collision. Whale buoyancy could also be affected and could represent a risk during ascent due to reduced maneuverability of whales (Nowacek et al., 2001), regardless of whether the whale perceives the vessel or the vessel perceives the whale.

Understanding the interactions between maritime vessels and whales has become a global conservation issue owing to increases in commercial shipping traffic (Corbett & Winebrake, 2007). So far, mitigation efforts to reduce collisions have mostly involved reducing speed limits on ships passing through whale habitats (Gende et al., 2011; Lagueux et al., 2011; McKenna et al., 2012, 2015) or rerouting shipping channels around these areas (McKenna et al., 2015).

This study represents a step in assessing the behavior of SRWs in the presence of marine vessels and provides a comparison to the behavior of SRW in the presence of specialized whale-watching vessels. However, to prevent vessel collisions and ultimately contribute to the development of appropriate management strategies to define areas of greater risk, more information regarding whales' subsurface behavior in the presence of vessels, vessel speed data, and acoustic conditions that could lead to a collision are needed. In Patagonia, previous studies have been performed on whale-watching activities but none in the context of shipping traffic. Due to the increase in the number of SRWs, the use of areas near the cities, and the area overlap between whales' deep-water diving behavior and shipping traffic, an increase in the probability of collisions is expected (Argüelles, Fazio, et al., 2016). Southern right whales tend to cluster and do not present major behavioral variations in the presence of boats. This, coupled with the fact that whales emerge quickly when they perform deep dives, could place SRWs in a situation of vulnerability if there is an increase in maritime traffic areas. If whales were always visible at the surface, ships might be able to implement measurements to avoid them, whereas if whales spent much time diving and emerged near the ships, no measurements could be undertaken (Argüelles, Fazio, et al., 2016). Understanding the behavioral and physiological responses of whales, in the presence of transiting boats, could help to improve procedures to protect SRWs from ship strikes. This could provide a critical piece of information needed to provide essential insights into areas of particular vulnerability.

AUTHOR CONTRIBUTIONS

María Belén Argüelles: Conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; project administration; resources; software; writing-original draft; writing-review & editing. Mariano Coscarella: Conceptualization; formal analysis; methodology; writing-review & editing. Carla Fiorito: Data curation; investigation. Marcelo Bertellotti: Conceptualization; methodology; supervision; writing-review & editing.