Species composition of sea stars (Echinodermata: Asteroidea) in the Patagonian Argentinian deep sea, including seven new records: connectivity with sub-Antarctic and Antarctic fauna

The main target of this paper is to improve the knowledge of the species composition of sea stars in Patagonian Argentine deep sea reaching depths of 2062 m. In addition, these results offer us the opportunity to analyze the possible connections between Argentinian marine fauna and adjacent Antarctic areas that have become a topic of interest in the past few years. This work is based on Atlantic Projects’ surveys carried out on an atypical and especially vulnerable marine ecosystems (canyons created from craters collapse by gas leaks). These are profusely impacted by frequent fishing activities, being one of the most important and international fishing grounds, where 887 records (1878 specimens) of 41 species of asteroids were collected in 217 stations ranging from 219 to 2062 m in depth. Seven of those species are proposed as new records: (Diplasterias octoradiata (Studer 1885), Plutonaster bifrons (Wyville Thomson, 1873), Radiaster elegans Perrier, 1881, Anseropoda antarctica Fisher, 1940, Pillsburiaster calvus Mah, 2011, Paralophaster lorioli (Koehler, 1907), Pteraster flabellifer Mortensen 1933). After refining the database built from literature and open-access databases such as OBIS and AntBIF, the new Argentinian asteroids deep-water checklist contains 2198 records from 64 asteroids species including the 7 new records proposed. Most of these 64 species (89.06%) are present in Antarctic-adjacent waters, and after the study of their occurrences at traditional biogeographic entities, our results support the hypothesis that Argentinian waters (in the case of the class Asteroidea) should be considered part of the sub-Antarctic entity.


Introduction
The distribution of Argentinian marine fauna and their connection with adjacent areas has become a topic of interest in the last few years. Regarding the north, a good study of these connections can be found in Alvarado and Solís-Marín (2013) and, within this field, the study undertaken by Brogger et al. (2013a) in Argentinian waters. Regarding the south, some authors have tried to explain how species distribution areas can help to analyze the existence of biogeographical entities (regions or provinces) in the Southern Ocean, mainly based on their occurrence (Hedgpeth 1969;Rodriguez et al. 2007; Griffiths et al. 2009;De Broyer et al. 2014;Figuerola et al. 2017;Weir and Stanworth 2019, among others). In the case of echinoderms, the most recent studies have been carried out by Martin-Ledo and Lopez Gonzalez (2014), Danis et al. (2014), Eleaume et al. (2014), Saucède et al. (2014), and Moreau et al. (2017Moreau et al. ( , 2018Moreau et al. ( , 2019Moreau et al. ( , 2021, among others. The study area is located in the Southwest Atlantic. It extends over 200 miles from the exclusive economic zone of Argentinian waters and reaches a depth of 2062 m. This zone is known as the Patagonian fishing area, owing to the fishing activities undertaken here by many countries, among them Spain (del Rio et al. 2012). These fishing activities are related to the area's high biodiversity, which is mainly due to the nutrient-rich upwelling during spring and summer as a consequence of the confluence of two ocean currents: the warm, subtropical Brazilian current (BC), which flows from the north, and the cold, deeper current from the Malvinas/ Falkland Islands (M/FI) from the south. The latter could be considered a branch of the Antarctic Circumpolar Current 1 3 (ACC) (Piola and Rivas 1997;Piola and Matano 2001;Rivas et al. 2006;Campagna et al. 2007). In addition, the bottom morphology shows a canyon system with very special features since it was created from the collapse of a series of pockmarks (craters produced by gas leaks).
In this context, the Spanish General Secretariat of Fisheries charged the Spanish Institute of Oceanography (Instituto Español de Oceanografía, IEO), with the undertaking of a series of multidisciplinary research surveys (focused on slope and canyon systems: Atlantis Project; del Rio et al. 2012) aimed at studying the potential existence of vulnerable marine ecosystems (VMEs), where echinoderms, among others, represent one of the most abundant and frequent taxa in the area.

Field sampling
Sampling was done during Atlantis Projects' surveys (13 multidisciplinary research expeditions) on board the Miguel Oliver vessel from 2007 to 2010 (Fig. 1). Different equipment was used: rock dredge (0.8 m wide and 0.3 m high; mesh size 10 mm), mega box corer dredge, and LOFOTEN trawl (31.20 m × 17.70 m; mesh size 35 mm).
Fauna was collected at 480 stations; however, only 217 stations were taken into account (219 m to 2062 m depth) because the present study is restricted to the slope and canyons system (Table 1).

Datasets
After species identification, the results were georeferenced and included in a dataset.
Several publicly available datasets were used for species distribution analysis: Ocean Biogeographic Information System (OBIS), Antarctic Biodiversity Information Facility (AntBIF), Museo Argentino de Ciencias Naturales (MACN), and online USNM Invertebrate Zoology Collection database. On the other hand, other records were include coming from literature such as Mah (2011), Mah andBlake (2012), Souto et al. (2014), Moreau et al. (2018), Guillaumot et al. (2020), and Moreau et al. (2021). These records were assessed to determine their reliability on the basis of two criteria: (1) records published in scientific journals and (2) records or human observations from international surveys or institutions and identified by experts of recognized standing. The reliable records were included together with Atlantic surveys ones that were curated following a polygon built (ArcGIS 10.7) delimiting the area of study to the appropriated countries borders and depth range (

Biogeographical approach
Five biogeographical entities were defined to analyze the Antarctic and near waters influence: Argentinian deep waters (ADW), Malvinas/Falkland Islands (M/FI), Scotia Arc (SA), sub-Antarctic Islands, including Bouvet Island, Prince Edward Islands, Crozet Islands and Kerguelen Islands (SAI), and Antarctica (A) following Griffiths et al. (2009), based on Moore et al. (1999).
Remarks Diplasterias octoradiata and Diplasterias radiata (Koehler 1923) (7-10 arms) could be confused because of their arms number overlap; however, the number of adambulacral spines are different: D. radiata are diplacanthid, and D. octoradiata are monacanthid.

Order Paxillosida Perrier 1884
Family Astropectinidae Gray 1840  AphiaID: 123904 Diagnosis Quite long arms R/r = 3.4/1-5.0/1, with narrow tips, almost pointed. Abactinal surface with plates with rather thin paxillae, in transverse series (Fig. 2d). Abactinal paxillae with columns oval in cross-section, crowned with 15-25 short spines. Marginal plates, both superomarginal and inferomarginal with only one prominent conical spine, and otherwise covered with very minute spines (Fig. 2f). Actinal plates covered with minute spines or granules and often with a single conical spine (Fig. 2e). Adambulacral plates with 5-10 furrow spines of equal size, outside of which is a single large conical spine (Fig. 2g).
Distribution Mediterranean Sea, Northeast Atlantic Ocean (from Faroe Islands to Gulf of Guinea, including Canary Islands), Northwest Atlantic Ocean (from New Jersey (USA) to Venezuela). Present study: Argentine. New record: 9 specimens (Online Resource 1).
Remarks Related to Plutonaster agassizi (Verrill 1880), P. bifrons shows a narrow and pointed arms with a terminal plate as long as wide, since P. agassizi ones are rounded, their tips are blunt, and their superomarginal plates curve inward. In contrast, the superomarginal armament of P. bifrons has spines on all plates and relatively long and pointed, P. agassizi does not present spines, at least from the distal side, and, if present, they are short, rigid, or conical.
AphiaID: 172707 Diagnosis Arms 5. The abactinal plates scalar, thin and imbricate, covered with granuliform spines and inconspicuous papular pores (Fig. 3a, b). Actinal and abactinal plates in decreasing size toward the marginal ones. Abactinal plates form a carinal series. Small superomarginal plates covered by granules. Small inferomarginal plates protrude to form the edge and are also covered by granules. Actinolateral plates in regular oblique series (Fig. 3c). Adambulacral plates with 6 spines on the furrow and joined by a membrane (Fig. 3d).
Remarks This genus was cited as Pillsburiaster sp. by Brogger et al. (2013c; report submarine canyons survey, II/III B/O Puerto Deseado at Rio de la Plata, Argentine), but the specimens were not identified as P. calvus; therefore, the present record will be first recorded in Argentinian waters. Genus Pillsburiaster includes 4 species that should be taken in account related to P. calvus, 1 species recorded in the South Atlantic Ocean, Pillsburiaster geographicus
AphiaID: 234937 Diagnosis Abactinal surface with isolated paxillae, each with a short peduncle bearing 6-8 elongated spines (Fig. 3i,  j). Marginal plates no more than 12 in each series. Large marginal spines in cluster arrangement (Fig. 3k, l). The actinal surface presents a blunt tuber-shaped prominence, with 4-5 spines with denticulations at their ends. Adambulacral plates with 3 spines in the furrow. They have 6 spines on the oral plate at their free edge (Fig. 3m).
Remarks Paralophaster lorioli could be confused with Paralophaster antarcticus (Koehler 1912). Paralophaster lorioli has few marginal plates, no more than 12 in each series with arrangement of few marginal spines in cluster, while P. antarcticus has more than 12 marginal plates that carry many spines much larger than the adjacent paxillae.
Distribution Only one record in Cape Town (South Africa) and Durban. Present study. Argentine. New record: 1 specimen (Online Resource 1).
Remarks: Pteraster flabellifer could be confused with Pteraster obscurus (Perrier 1891), Pteraster stellifer Sladen 1882, Pteraster affinis Smith 1876, or Pteraster militaris (OF Müller 1776. Pteraster obscurus present the 6-8 arms, but paxillar columns bear numerous spines (more than 6), and a thick web where adambulacral and oral spines are webbed for totality of their length; its area of distribution restricted to the north Atlantic (boreal belt). Related to the other remarkably similar species distributed around Atlantic Ocean, P. stellifer and P. affinis, both with 5 arms, are species from the South Hemisphere; however, P. stellifer has a parchment-like supradorsal membrane, large actinolateral spines, and unwebbed oral spines. Although P. militaris evenly could have 6 arms, its distribution is restricted to the North Hemisphere. In any case, Clark discusses about the possibility that these three species (P. affinis, P. flabellifer, and P. militaris) "may eventually prove to favor only a subspecific relationship" (Clark and Downey 1992), based on results reported in Clark (1962).

Argentinian deep-sea asteroids new checklist
The new Argentinian deep-sea checklist consists of 64 asteroids species that emerge from 2198 records: 887 records from Atlantis surveys and 1311 records from publicly available datasets and literature (Table 4, Fig. 4). These results include the seven new records proposed in the present work.

Reliability of the databases and Argentinian new asteroid checklist
Concerning the species composition in the study area, 64 species should be considered valid records in this area (Bernasconi 1937(Bernasconi , 1962a(Bernasconi , 1963(Bernasconi , 1964a(Bernasconi , 1966b(Bernasconi , 1973a(Bernasconi , 1973b(Bernasconi , 1979Gutt and Starmans 1998;Tablado and Maytia 1988;Orovitz and Tablado 1990;Bastida et al. 1992;Clark and Downey 1992;Gutt et al. 1999;Bremec et al. 2000;Rios et al. 2003;Zaixso 2004;Bertness et al. 2006;Mutschke and Rios 2006;Mah 2011;Mah and Blake, 2012;Brogger et al. 2013a;Souto et al. 2014;Arribas et al. 2016;Fraysse et al. 2018;Moreau et al. 2018Moreau et al. , 2021Guillaumot et al. 2020; National Museum of Natural History (NMNH: Research and Collections Information System, Smithsonian Institution), but only 41 were captured. Consequently, 23 of these species were not reported (Table 4; Fig. 4). These absences could be due to various reasons, including: (1) samplers' limitations or (2) more likely their low frequency, which makes them difficult to be capture. The latter reason could be directly related to the special features of canyons, morphology, granulometry, or even other biotic factors such as species composition (del Rio et al. 2012). A deeper ecological study will be published in the near future providing more useful information that allows us to explain these absences.

The distribution areas of Argentinian asteroids and their Antarctic connections
The first study approach focuses on the number of shared species in ADW and the remainder of the four entities (provinces/regions) defined as M/FI, SA, SAI, and A (discussed in Griffiths et al. 2009, based on Moore et al. 1999) (Fig. 5).
Under the geographical/countries limits and depth range that frame the present work, ADW shares 89.06% of species with these areas, and only ten species should be considered "Patagonian species" (ADW + M/FI) (Fig. 5a, e framed species list), and they are not present outside of this traditionally named area. Three of them correspond to our proposed new records (discussed above), and the remaining seven are: Cryptasterias brachiata and Lethasterias australis, endemic in Patagonian waters; Hymenaster pergamentaceus and Asterina stellifera, frequent only in the Southwest Atlantic Ocean; Anasterias spirabilis and Asterina fimbriata, species living on the limit of the sub-Antarctic area; and Perisasterias polyacantha, which is frequent in Australia, rare in southern Africa, and very scarce in Argentine (Tablado and Maytia 1988;Brogger et al. 2013a).
Despite the number of ADW species present in the nearby waters, M/FI, SA, and SI altogether share only 35.93% of ADW species (Fig. 5a). However, when they were analyzed separately, we realized that ADW and SA share 81.25% of species, whereas M/FI shares a lower percentage (56.25%) despite their proximity (Fig. 5a). These results could support the idea that M/FI would be an independent entity, as supported by Figuerola et al. (2017) and Griffiths et al. (2009) in the case of gastropods (Fig. 5a). However, this argument could be based on a false premise since, in our study, only deep-sea asteroids were considered and many recorded species at M/FI usually live in shallower waters. In any case, authors who defend other hypotheses did not consider ADW and M/FI as independent entities.
Nevertheless, it is remarkable that 100% of the species from ADW recorded at SAI are included in the SA area (Fig. 5a), which endorses our idea of joining these two areas together as a sub-Antarctic area (Fig. 5d, e), and (under this bathyal context), as has been reported by Briggs and Bowen (2012) and Griffiths et al. (2009) (Fig. 5b,  5c).
On the basis of the discussion presented above, only one hypothesis could adjust to our preliminary biogeographic results: ADW, M/FI, and SA should be considered sub-Antarctic, depending on whether SA could be a separated entity of Antarctica, based on the fact that ADW and SA share a greater number of species than any other (Fig. 5a) when the rest of the Antarctica region is included (Fig. 5d). Therefore, to confirm or refute this hypothesis (Rodríguez et al. 2007; Martin-Ledo and López-González 2014; Weir and Stanworth 2019; Moreau et al. 2017, among others), some considerations should be taken in account: (1) we should include more records from known surveys that did not incorporate public datasets (such as the BENTART projects, and others); (2) we need to analyze the benefit of including shallower species, or follow Watling et al. (2013), who opted for conducting their studies at the bathyal and abyssal provinces levels, depending on the focus; (3) and, finally, a deeper analysis should be carried out using different biogeographical methods that would be associated with environmental variables (our next target).  Table 4 Author Data availability All data included in this paper will be sent to Antarctic Biodiversity Information Facility (AntBIF) after data are published. The material will be shared with different museums.
Code availability Not applicable.

Conflict of interest
The authors declare no conflicts of interest regarding this manuscript.
Ethical approval None.

Consent to participate
We obtained consent to participate.

Consent for publication We obtained consent for publication.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.