INTRODUCTION

Bony and cartilaginous fishes are essential in the ecosystems they live in because they play important roles that allow maintaining stable trophic relations and they represent an indicator of the ecosystem health (^{Jennings and Kaiser, 1998}; ^{Navia et al., 2017}). Eliminating these species could cause an important imbalance that can extend to other species and, therefore, cause the whole system to collapse (Navia et al., 2007; Navia et al., 2012; ^{Bornatowski et al., 2014}; ^{Dulvy et al., 2014}; Navia et al., 2016). Particularly, cartilaginous fish or elasmobranches (sharks and manta rays) have characteristics that make them an especially vulnerable group compared to some bony fish (^{Stevens et al., 2000}; ^{Myers and Worm, 2005}), because they show a slow and little resilient lifecycle (^{Bonfil, 1997}; ^{Lucifora, 2003}; Ladino, 2014).

There is a world side consensus about the increasing degradation of the marine ecosystems the collapse of most fish populations, mainly due to fishing overexploitation (^{Dulvy et al., 2014}; ^{White et al., 2015}). According to a United Nations reports, illegal and unregulated fishing, together with the effects of the growing acidification of the oceans, climate change, and eutrophication, represent the greatest threats for the marine species; the mentioned report highlights that the fraction of biologically sustainable fishing stocks has decreased from 90 % to 66.9 % between 1974 and 2015 (United Nations, 2019). Despite this critical situation, countries often have difficulties when taking concrete conservation actions, due to the lack of information about the trends in the status of vulnerable populations, and the deficiency in establishing fishing regulations. This is why one of the most common strategies developed to preserve these species are marine protected areas (MPA) (^{Graham et al., 2003}). This preservation measure consists of legally establishing sectors with different purposes, among which the protection of marine species and habitats, the restoration of fishing reserves, and the administration of tourist activities, can be highlighted (^{Pomeroy, 2007}). It has been proved that diversity and big fish biomass in these MPA are often superior to those of nearby areas, depending on MPA factors such as size, antiquity, and type of protection (^{Edgar, 2011}; Edgar et al., 2014). Even the cases where benefits are perceived by neighboring communities are numerous because the so-called “overflow effect” fosters the fish flow from MPA Marine Protected Area to the surrounding areas where they can be sustainably used (Graham et al., 2003; ^{Grüss et al., 2011}; Graham and McClanahan, 2013; ^{Stuart-Smith et al., 2013}).

The Malpelo Fauna and Flora Sanctuary (Malpelo FFS) is a MPA of great importance at the regional level because it represents a connection point between coastal and oceanic ecosystems. Besides, it is worldwide recognized for harboring large diversity and abundance of marine organisms, among which the large shoals of sharks such as the hammerhead shark (Sphyrna lewini) and the silky shark (Carcharhinus falciformis) can be particularly highlighted. This PMA, along with the Galapagos Natural Reserve (Ecuador), the National Park Isla del Coco (Costa Rica), the National Natural Park Gorgona (Colombia), and the National Park Coiba (Panamá), form the Eastern Tropical Pacific Marine Corridor (CMAR), an initiative that seeks to promote the sustainable use of the biological diversity in the area surrounding these Marine Protected Area MPA (CMAR, 2015).

Unfortunately, most MPA around the world lack periodical data or historical records that allow knowing the long-term impact on the species and ecosystems that these areas intend to protect (Barnett et al., 2012). This occurs due to economic and logistic limitations that tend to increase at oceanic islands such as Malpelo. However, annual monitoring of its fauna has been made in the Malpelo FFS to establish its population trends. This research evaluates and discusses the relative abundance values and the frequencies of observation for eleven species of fish monitored for eleven years (2009-2019). The monitored species were: the bluefin trevally (Caranx melampygus), the longfin yellowtail (Seriola rivoliana), the yellowfin tuna (Thunnus albacares), the bass (Dermatolepis dermatolepis), the sailfin grouper (Mycteroperca olfax), the eagle ray (Aetobatus laticeps), the silky shark (Carcharhinus falciformis), the Galapagos shark (Carcharhinus galapagensis), the whitetip reef shark (Triaenodon obesus), the whale shark (Rhincodon typus), and the scalloped hammerhead (Sphyrna lewini). The results of this research will be useful to evaluate and improve the effectiveness of management and the current preservation actions of these populations at the Malpelo FFS.

STUDY AREA

The Malpelo FFS is currently 2,709,600 ha in size, within which the Malpelo Island (4° 0´ N, 81° 36´W) is the only emerged part. Malpelo is located in the Eastern Tropical Pacific (ETP) 490 km away from the Colombian coast of Buenaventura (Figure 1). The island is surrounded by eleven pinnacles, it is 1.2 km² in size, and it is the summit of an underwater mountain range that rises from 4,000 m depth (Keisser and Hoffman, 1975; ^{Caita and Guerrero, 2000}). Regarding its oceanographic conditions, four currents influence Malpelo: the North Equatorial Counter Current, the South Equatorial Counter Current, the Colombia Current, and the Panama Cyclonic Current. Due to their interactions, it is possible to differentiate two clear seasons: one of the cold waters and another one of the warm waters. The first season runs from January to April, showing superficial temperatures around 23 ºC, while the warm waters season runs from May to December when average temperatures fall around 27 ºC (^{Rodríguez-Rubio and Giraldo, 2011}).

Figure 1 Location of sff malpelo in the colombian pacific. 

MATERIALS AND METHODS

Field Stage

Data used in the study come from the submarine visual census (SVC) developed by the group of fish monitoring of the Malpelo FFS, in dives of about 60 minutes between the years 2009 and 2019. These censuses were made by divers previously trained for the identification of fauna of the Sanctuary, recording the abundance of some species prioritized as Values Objet of Conservation (VOC), the sampling area, the date, the time, the temperature, the visibility, and the diving time (^{Schmitt et al., 2002}; ^{Auster et al., 2005}). The most experienced diver always did the data collection during dives, and those counts were taken into account to record the respective census to avoid overestimations. Despite there are 32 diving sites in Malpelo, in each dive it is possible to monitor only 5-7 of them. Conditions such as the strong currents and the swell in the area make these monitored areas highly variable and dependent on the daily climate conditions. More than 75 % of the monitoring in each expedition was made in the sites known as El Arrecife and La Nevera (Figure 1), which are the areas most protected against swell and diverse, because the largest coralline formations, and many cleaning stations, have been observed there. The routes in El Arrecife and La Nevera are always the same, changing only the direction according to the current. Because of the above, this study only used and analyzed data from these two diving sites. The sampling effort, represented in diving hours per year, has increased during the last years (Figure 2) because, besides the scientific expeditions, the tourist sector carried out citizen science trips, allowing the researches to increase the number of dives during the last four years to obtain valuable data.

Figure 2 Sampling effort (diving hours/year) during monitoring of marine fauna in the sectors El Arrecife and La Nevera between 2009- 2019 in the Malpelo FFS. 

RESULTS AND DISCUSSION

The abundance of all species was highly variable among years, in contrast to the observation frequency that was less variable (Figures 3, 4, and 5). In general, negative trends were observed, though not significant, in the relative abundance of the evaluated species over time, except Aetobatus laticeps and Triaenodon obesus, which showed positive trends (Figure 5a, c).

Figure 3 Average relative abundance (individuals/diving time) and sighting frequency between 2009-2019 for C. melampygus, S. rivoliana, M. olfax, and D. dermatolepis in the SFF Malpelo. 

Figure 4 Average relative abundance (individuals/diving hours) and sighting frequency between 2009-2019 for T. albacares, C. falciformis, and R. typus in the Malpelo FFS. 

Figure 5 Average relative abundance (individuals/diving hour) and sighting frequency between 2009-2019 for A. laticeps, T. obesus, C. galapagensis, and S. lewini in the Malpelo FFS. 

Fish associated with the reef, and with gregarious habits, showed a trend to decrease their abundance over the years, but this was not significant in any case S. rivoliana (r = -0.534, P = 0.09), M. olfax (r = -0.475, P = 0.140), C. melampygus (r = -0.271, P = 0.419 and D. dermatolepis (r = -0.275, P = 0.413). However, all the species in this group had observation frequencies greater than 80 % in most of the years (Figure 3b, d, f, h), which can be explained by the use these species make of the reef. For instance, species such as C. melampygus, D. dermatolepis, S. rivoliana, and M. olfax form food aggregations following foraging and hunting activities of other species such as moray eel Gymnothorax dovii (^{Quimbayo et al., 2014}). These interspecific associations, added to the search for shelter, suggest that their association with the coralline formations of Malpelo is important, a reason for the continued sightings in these diving sites. These species are observed constantly associated with the coralline formations of Malpelo in their young and adult stages, but their larval behavior is unknown and it is assumed that larvae connect Malpelo’s populations to those of other areas within the ETP, thanks to currents that disperse them (^{Wood et al., 2016}). In surrounding areas such as Galapagos, populations of M. olfax had undergone significant changes in the presence and size of big reproductive individuals exposing the overexploitation this species suffers at the local level (^{Usseglio et al., 2016}). Taking into account that these species live in CMAR and their distributions include marine areas outside Malpelo FFS, the decreasing trend of their relative abundances could reflect the pressures that populations undergo at the regional level.

In contrast, T. albacares showed very low relative abundances, little fluctuant, (except by a peak in 2010), and a negative trend, although not significant (r = -0.270, P = 0.422). In addition, the observation frequency varied between 0 and 30 % over the years (Figure 4a, b). The peak observed in 2010 can be explained by the presence of a shoal while the census was carrying out. The methodologies to count individuals of this species are limited and require additional effort beyond the scope of this study. This occurs due to, in the first place, the fact that it is a purely pelagic species that do not frequent the areas where dives were made. In the second place, its condition of highly migratory species, traveling in hundreds or thousands of individuals (^{Lan et al., 2013}, ^{Bessudo and Lefèvre, 2017}), makes it a hard to monitor species, which counts are little accurate due to the difficult of quantifying individuals in groups so large.

Limitations similar to the above were found when monitoring the silky shark (Carcharhinus falciformis), species that showed no significant trend (r = 0.210, P = 0.535; Figure 4c). It was observed with a frequency lesser than 30 %, except in 2016 when reached values close to 60 % (Figure 4d). Such results are inconclusive and must be carefully examined taking into account the difficult in monitoring species like this one, with oceanic habits, that does not reside permanently in Malpelo and highly mobile in large groups, hard to count. It is important to highlight that large shoals are often observed exclusively between May and October (^{Bessudo and Lefèvre, 2017}) and they do not use to come close to areas such as El Arrecife and La Nevera, but they frequent a shallow area to the north of Malpelo known as El Acuario, which was not included in the study due to the diving conditions that make it a complex place for data collecting.

Highly mobile species such as the whale shark (Rhincodon typus) are also hard to monitor. In this study, their relative abundance was highly fluctuating, with no significant trends (r = -0.068, P = 0.843), and with a very low observation frequency, always less than 20 % (Figure 43e, f). Solitary and transient individuals who stay one or two weeks between June and September (^{Bessudo et al., 2016}) can be occasionally observed in Malpelo. Results show a variable and low abundance with an observation frequency also consistently low. This could be explained by the fact that it is a solitary, pelagic, and migratory species.

On the other hand, not-significant slightly positive trends were observed in the relative abundances of A. laticeps (r = 0.075, P = 0.827) and T. obesus (r = 0.329, P = 0.323) populations, accompanied by an increase in the observation frequency (Figure 5 a, b, c, d). Both the eagle ray and the whitetip reef shark are species with solitary habits or small aggregations in Malpelo (^{Bessudo and Lefèvre, 2017}). A. laticeps tends to reach sexual maturity relatively early compared to other elasmobranches (^{Schluessel et al., 2010}), and despite being the most common ray in Malpelo, to date, no studies of mobility and distribution of the species within the MPA has been conducted. However, the hypothesis that they could be resident has been posed, based on yearlong sightings and records of individuals at all their development stages, even females were documented having offspring (Bessudo y Lefévre, 2017). Likewise, T. obesus is one of the most common sharks in the ETP’s coral ecosystems and trending to philopatry (^{Randall, 1977}; ^{Peña herrera et al., 2012}; ^{Mejía-Falla et al., 2014}). Studies with acoustic telemetry have revealed high periods of permanence within MPA and, actually, it likely stays its whole lifecycle within the sanctuary because they have been observed in all their life stages (^{Ladino et al., 2016}). Stable behaviors were reported for these two species in a previous study (^{Soler et al., 2013}), so these records, along with the light increases recorded regarding relative abundance and meeting frequency in both species, suggest that Malpelo could represent a key place in the recovery and possible reproduction for these species.

The Galapagos shark (Carcharhinus galapagensis), despite showing a highly fluctuant observation frequency (Figure 5e, f), showed a non-significant negative trend in its relative abundance (r = -0.472, P = 0.143). It is often common to find it in Malpelo the whole year in small groups of less than three individuals, although occasionally congregations of up to 40 individuals have been seen in a reproductive frenzy (^{Bessudo and Lefèvre, 2017}). Results suggest that during dives there is a high probability of finding individuals of this species, although over the years it seems that these meetings are with smaller and smaller groups. It is important to take into account this species’ mobility when evaluating the obtained data; it has been posed that their domestic environment is smaller than 150 km² (^{Kohler and Turner, 2001}; ^{Holzwarth et al., 2006}; ^{Meyer et al., 2010}) and some animals tagged with satellite and acoustic telemetry showed broad residence periods (^{Ladino et al., 2016}), although some migrations towards other islands of CMAR have been recorded. These movements are probably made with reproductive purposes because within the sanctuary individuals of sizes smaller than 120 cm have not been recorded, and besides Malpelo lacks ecosystems such as coasts and estuaries that serve as breeding areas (Ladino et al., 2016). These mobilizations outside MPA could be deemed as a risk factor in which shark groups are exposed to hazards such as fishing, which could cause declines in their populations.

The sanctuary’s flagship species, the hammerhead shark (Sphyrna lewini), was the only evaluated species that showed a significant decrease in its relative abundance (r = -0.689, P = 0.019), decreasing in 73.3 % between 2009 and 2019 (Figure 5g). However, its observation frequency kept over 80 % (except in 2013 and 2014; Figure 5h), which means that in general this species is frequently observed, but by increasingly smaller groups. These results are coherent with reports in other CMAR islands. For instance, in Galapagos islands, the abundance of hammerhead shark decreased by 50 % between 1985 and 2015 (^{Peñaherrera-Palma et al., 2017}), and in Isla del Coco a decrease in the abundance of around 43 % between years 1993 to 2013 was reported (^{Nalesso, 2014}; ^{White et al., 2015}). Likewise, in Australia, Queensland’s Agriculture Department reported a population decrease of 73 % between 2005 and 2016. This critical situation is related to the large ranges of movement of this species. Telemetry studies in the region have proved that hammerhead shark move around the whole ETP, between oceanic islands (^{Bessudo et al., 2011}), submarine mountains (^{Chávez et al., 2020}), and coastal areas to have their offspring (^{Quintanilla et al., 2015}; ^{Salinas-de-León et al., 2017}; Bessudo and Ladino, 2019; ^{Zanella et al., 2019}). This highlights the necessity of implementing protected biologic corridors between the ETP’s islands (^{Grüss et al., 2011}; Migramar, 2016) and the importance of identifying and protecting the breeding areas of the species.

CONCLUSIONS

In general, not significant negative trends were observed in the relative abundance of the evaluated species, except Aetobatus laticeps y Triaenodon obesus, which showed no significant positive trends.

No significant variations were observed in the relative abundance of the species that apparently spend their whole lifecycle in the sanctuary, which may be because Malpelo is MPA that has taken management actions such as prohibiting fishing, broadening the area, and strengthening control and surveillance, which aim at stopping the deterioration of populations in the area.

On the other hand, it is evident a significant decline in the hammerhead shark populations, probably associated with the fact of having very broad domestic environments that surpass the MPA’s limits (^{Costa et al. 2012}). The results highlight the importance of keeping the monitoring in the long term in an ecosystem like Malpelo because, despite being far from human activities, is not oblivious to the impact such activities are generating in the ocean. Besides, it shows the necessity of protecting marine species beyond MPA borders, working together in the connectivity of the region and the fishing regulations in the surrounding areas