All descriptive indices of the diet of D. gigas revealed a temporal change in dietary breadth and trophic level associated with environmental variation possibly influenced by environmental supply or prey availability.
The diet composition of jumbo squid D. gigas from the coast of Peru during the period 2015–2019 was mainly composed of mesopelagic fishes such as Vinciguerria lucetia, and myctophids, among which Lampanyctus sp. and Diogenichthys laternatus were prominent, followed by cephalopods Abraliopsis affinis and some Octopodidae. Crustaceans followed in importance with P. monodon showing the highest frequency of occurrence, followed by Euphausiidae. Pelagic gastropods represented a low percentage of occurrence in their diet, similar to that described by Markaida et al. (2008) in the Gulf of California. The high occurrence of V. lucetia in the stomachs of D. gigas in this study corroborates their preference in the coasts of Peru reported in previous studies between the years 2004–2011 (Rosas-Luis et al. 2011, Alegre et al. 2014).
The diet composition of jumbo squid presented similarities between 2015 and 2016, observing that euphausiids were the main prey in terms of %FO and %N, but the decrease in trophic level indicates a possible change in prey availability, possibly associated with the increase in temperature that occurred due to the effect of Coastal ENSO (L´Heureux & Takahashi 2016; Takahashi 2017). In contrast, the high consumption of V. lucetia, had a predominant frequency and abundance between 2017 and 2019, which is usually a frequent species in the diet of D. gigas from the northern Humboldt Current ecosystem, so the coastal ENSO would have an influence on prey availability and consequently on temporal variations in diet. The fact that in 2016 the squid collected were smaller (< 600 mm ML) would not influence the diet, since in 2018 small squid were also collected and the trophic level was higher due to higher fish consumption.
Our results on ontogenetic variation in the diet of D. gigas are similar to those reported in several studies (Nigmatullin et al. 2001, Alegre et al. 2014, Pardo-Gandarillas et al. 2014, Portner et al. 2020, Bruno et al. 2021) since large individuals eat more fish and cephalopods, and also small crustaceans. Likewise, Portner et al. (2020), mentions that euphausiids usually have lower energy per unit weight, cephalopods have an intermediate energy and fish (myctophids and engraulids) have higher energy, therefore, a higher fish consumption as their body size increases, would represent an increase in the total calories ingested, which is consistent with the diet of large specimens observed in this work, which mainly consumed V. lucetia. Previous studies show that V. lucetia is an abundant fish in Peruvian waters and its distribution is closely related with the jumbo squid biomass and distribution (Argüelles et al. 2008, Rosas-Luis et al. 2011). This fact is consistent with the idea of opportunistic behavior of D. gigas feeding on more abundant prey (see Ibáñez et al. 2008).
The composition of the diet of D. gigas did not vary significantly with distance from the coast, since V. lucetia was the main prey. However, closer to the coast crustaceans such as P. monodon and Euphausiidae were observed, Cavolinia uncinata occurred between 51–100 nm; in this sense Quesquén (2005), mentions that species of the Thecosomata group are associated with warm water masses and are common in Subtropical Surface Waters and Equatorial Surface Waters, as well as that the richness and diversity values increase as they move away from the coast. Myctophids and V. lucetia increased FO as distance from the coast increased (151–200 nm), similar to that reported by Alegre et al. (2014). Similarly, cephalopods presented higher FO in the diet of specimens caught at 151–200 mn. Alegre et al. (2014) mention that mesopelagic fish (V. lucetia and Myctophiidae) contributed to the diet mainly during the spring and far from the coast, while euphausiids were closer to the coast.
In general, the spatio-temporal variations in the diet composition of jumbo squid observed in this work would be related mainly to the availability and accessibility of prey, which is a determinant in the variability of a squid's diet. Likewise, Portner et al. (2020), found that D. gigas from the Gulf of California exhibits significant variability in diet with respect to capture location, body size and environmental conditions. They similarly agree with that reported in the northern hemisphere by Markaida & Sosa-Nishizaki (2003), who found large spatial and temporal variations in the main prey of D. gigas from the Gulf of California between 1995 and 1997.
In the present study, we observed low levels of cannibalism, contrary to what has been observed by other studies (e.g., Alegre et al. 2014), however, the presence of unidentified cephalopods (mantle and arm remains) could have been cannibalized individuals. Therefore, a relationship between cannibalism and body size as previously described (Pardo-Gandarillas et al. 2014, Bruno et al. 2021) cannot be ruled out. High cannibalism in this species occurs mainly when they are captured with jigs, as the stress of capture and the feeding frenzy characteristic of this species induces cannibalism (Ibáñez et al. 2008, Ibáñez & Keyl 2010, Bruno et al. 2021).
Finally, the GAM analyses showed that the diet of D. gigas presented spatial and temporal variations, observing that the most influential variable in the prey consumption was SST, SOI, followed by latitude. The Coastal ENSO phenomenon, which occurred between 2015 and 2017 (L´Heureux & Takahashi 2016; Takahashi 2017), could have influenced the predominance of some prey species.