Ten years later: An update on the status of collections of endemic Gulf of Mexico fishes put at risk by the 2010 Oil Spill

Abstract The 2010 Gulf of Mexico Deepwater Horizon was the largest oil spill in human history that occurred during a 12-week period in a region less than 100 km from the coast of Louisiana; however, after more than a decade of post-spill research, few definitives can be said to be known about the long-term impacts on the development and distribution of fishes in and around the region of the disaster. Here, we examine endemic Gulf of Mexico fish species that may have been most impacted by noting their past distributions in the region of the spill and examining data of known collecting events and observations over the last twenty years (ten years prior to the spill, ten years post-spill). Five years post-spill, it was reported that 48 of the Gulf’s endemic fish species had not been collected and, with expanded methods, we now report that 29 (of the 78 endemic species) have not been reported in collections since 2010 (five of these are only known from observations post-spill). Although the good news that some previously ‘missing’ species have been found may be cause to celebrate, the lack of information for many species remains a cause for concern given focused sampling efforts post-spill.


Introduction
The 2010 Gulf of Mexico Deepwater Horizon Oil Spill (DWH), MC 252 or Macondo blowout, was the largest accidental oil spill in history (Crone and Tolstoy 2010, Rabalais 2014, Murawski et al. 2023).The tremendous amount of oil spilled during DWH was estimated at 134 million gallons (down from greater than 200 million reported in earlier sources; or 507 million liters versus the 757 million liters reported earlier), which resulted in an immediate contamination area of 149,000 km and continued to spread through currents widely in the Gulf of Mexico (GOM) (Rabalais 2011, Ramseur and Hagerty 2014, MacDonald et al. 2015, Berenshtein et al. 2020).The spill lasted from the explosion and blowout on 20 April 2010 until the well was contained on 15 July 2010 (Petrolia 2014).
Coupled with the fact that it occurred in the deep sea (> 1000 m depth) and with the coordinated release of more than a million gallons of dispersant, which has been suggested to have made the DWH oil as much as 52% more toxic, more difficult to clean up and increased its impact to wildlife (Rico-Martínez et al. 2013).It has been suggested almost 100 million gallons (379 million liters) of DWH oil combined with dispersants remains in the Gulf and is one of the worst pollution events in history (Goodbody-Gringley et al. 2013, Ramseur andHagerty 2014).
More than a decade after the end of the spill, the long-term effects of DWH are still not fully understood.Recent research has suggested that there have been some persistent ecological effects including damage to deep ocean coral communities, harm to oyster fisheries, loss of marshlands and population declines of marine mammals, sea turtles and seabirds (Barron et al. 2020).Some fish species appear to have been particularly at risk and impacted by the 2010 Oil Spill with evidence of physical and developmental abnormalities reported and evidence of extirpations (Whitehead et al. 2012, Incardona et al. 2014, Dubansky et al. 2013, Brette et al. 2014, Mager et al. 2014, Alloy et al. 2016, Chakrabarty et al. 2016).Further, risks from oil production related polycyclic aromatic hydrocarbons (PAH) exposure and concentrations in fishes is widespread in the GOM and will likely continue as extraction of petrochemical intensifies (Turner andRabalais 2019, Pulster et al. 2020).These include Taylor Energy's MC20 oil spill which began in 2004 and continues today (Schrope 2013, Mason 2019, Schrope 2013).Linardich et al. (2019) reported that habitat degradation is the leading cause of risk to endemic GOM species and oil spills are certainly part of that degredation issue.
The Gulf of Mexico is one of the most biologically rich and resilient marine environments in the world with 1541 fish species known from the region, 78 of which are reported to be endemic to the Gulf (i.e.found only in the Gulf; McEachran (2009), Chakrabarty et al. (2016)).Although many commercially valuable fish species populations have been examined following DWH, most Gulf ichthyofauna as a whole have received little attention (Fodrie et al. 2014, Murawski et al. 2023).The IUCN has suggested that upwards of 25% of the Gulf's endemic fishes may be threatened with extinction (Linardich et al. 2019, IUCN 2023).Chakrabarty et al. (Chakrabarty et al. 2016) examined museum records for the occurrence of all known endemic Gulf fish species (77 spp.thought to be endemic at the time) five years after the spill and reported that 48 species of fish had not been officially collected (i.e.vouchered in natural history collections) since the 2010 spill.Of these, 14 species were designated as being of 'greatest concern' as they may have been most impacted by DWH because of their past distributions being largely (> 35%) within the spill region.
In this current study, we re-examine museum records  of the 78 endemic species (adding the recently described American Pocket shark, Mollisquama mississippiensis, as a Gulf endemic) using data from The Global Biodiversity Information Facility (GBIF: https://www.gbif.org/),FishNet2 (http://www.fishnet2.net/)and recent literature.

Material and methods
The documented occurrences of 78 species endemic to the Gulf of Mexico were tallied using two databases: The Global Biodiversity Information Facility (GBIF) and FishNet2.To complement data about voucher material, an extensive review of the literature on individual species was also performed to account for observations not involving collections.The International Union for Conservation of Nature's Red List of Threatened Species (abbreviated "IUCN Red List" throughout) status was also reviewed and reported here when available (https://www.iucnredlist.org/).GBIF recently changed their data algorithm to include observed species in addition to vouchered specimens.To accurately measure the change in species populations, the data from the total identified specimens (vouchered and observed in nature) from GBIF were graphed separately and compared to the total vouchered specimens from FishNet2 from 2000 to 2020.Only species found in the Gulf of Mexico and Caribbean Region were accounted for in the data, unless otherwise noted.We last gathered data from these websites on 20 April 2023.
A scatterplot graph was created in Microsoft Excel by plotting the Number of Occurrence(s) on the y-axis and the Number of Years on the x-axis.Collections data from GBIF are denoted as "GBIF Preserved" (orange circles) when a voucher specimen was collected, human observation data from GBIF are denoted as "GBIF Observed" (blue triangles) and collections data from FishNet2 are denoted by "FishNet2 Preserved" (grey diamonds) to indicate the type(s) of data that each database reported.No graph is included if species were not sampled between 2000-2020 (10 years before and 10 years after DWH).Chakrabarty et al. (2016), Chakrabarty et al. (2012) and Chakrabarty et al. 2016 listed fish species of 'conservation concern' based on their known distribution within the spill zone or proximity to the DWH oil spill surface slick.Here, we refer to these as "Species of Greatest Concern" (if over 35% of their range was in the spill zone) or "Lesser Concern" (if less than 35% of their range was in the spill zone, but they are still considered endemic).No scatterplot is included for species lacking collections in the 20 year period of our survey, but we include information about the last observations/collections events when that is known.

Eschmeyer's
Catalog of Fishes (https://www.calacademy.org/scientists/projects/eschmeyers-catalog-of-fishes) was used to obtain current valid taxonomic names including the authority (authors of original description) and the family name (Fricke et al. 2023).Please note that following taxonomic convention, the taxonomic-authority reference following the scientific name is only presented in parentheses if the species was described in a different genus than it is currently in, otherwise there are no parentheses.Species are presented using the taxonomy of Nelson et al. (2016), as well as the phylogenetic classification of Near and Thacker (2023) for Actinopterygii (species are listed in alphabetical order within each family and major clade, i.e. "Percomorpha").

RAJIFORMES
Dipturus olseni (Bigelow and Schroeder, 1951) in Bigelow and Schroeder (1951); (Rajidae), Spreadfin skate.The species was last reported in 2011 off the coast of Alabama.Status listed as "Least Concern" on the IUCN Red List (Kulka et al. 2020).

STOMIIFORMES
Eustomias leptobolus Regan andTrewavas, 1930 in Regan andTrewavas (1930); (Stomiidae), Black dragonfish.The species was last reported in the GOM in 1960 off the coast of Louisiana and the species status is listed as " Data Deficient" by IUCN (Harold and Milligan 2019).

Aulopiformes
Stemonosudis bullisi Rofen, 1963in Rofen (1963); (Paralepididae).Species was last collected in 2007 off the coast of Alabama.Status listed as "Data Deficient" on the IUCN Red List (Russell 2010).Ten years later: An update on the status of collections of endemic Gulf ...

Oneirodidae
Oneirodes bradburyae Grey, 1957in Grey (1957); (Oneirodidae), American Dreamer anglerfish.Collected once in 1954 off the coast of western Florida.A genetic sample was recorded in 2015 off the coast of Alabama by the DEEPEND Consortium (Schmutz 2022).Status listed as "Data Deficient" on the IUCN Red List (Carpenter et al. 2019).Randall and Caldwell, 1966

Gobiosoma longipala
Varicus benthonis (Ginsburg, 1953) formerly Chriolepis benthonis in Ginsburg (1953); (Gobiidae), Deepwater goby.There are five reports of this species of GBIF collected off the Yucatan Peninsula, Mexico with no date given.Status listed as "Data Deficient" on the IUCN Red List (Pezold et al. 2015).Varicus marilynae Gilmore, 1979in Gilmore (1979); (Gobiidae), Orange belly goby.The species was last reported off the coast of Florida in 1974.Status listed as "Data Deficient" on the IUCN Red List (Gilmore et al. 2015).

LABRIFORMES
Halichoeres burekae Weaver and Rocha, 2007 in Weaver and Rocha (2007); (Labridae), Mardi Gras wrasse.Species listed as "Endangered" on the IUCN Red List (UICN, 2022).Robertson et al. (2016) reported H. burekae to be abundant in the Cayo Arcas reefs off the Yucatan Peninsula in Mexico.Francisco and Fraco-Mej (2017) reported H. burekae as one of the most abundant species encountered at Enmedio Reef off Veracruz, Mexico.Escarcega-Quiroga and Flores-Serrano (2020) reported H. burekae to be a frequently consumed species in the diet of the introduced Red lionfish (Pterois volitans) in coral reefs of northern Veracruz, Mexico.Gonzlez-Gandara and Chavez (2020) reported the relative abundance of H. burekae as > 30% in Veracruz Coast, 2-5% in Campeche Bank and < 1 in the Mexican Caribbean.Status listed as "Endangered" on the IUCN Red List (Rocha et al. 2015).Ginsburg, 1950in Ginsburg (1950)   Prionotus paralatus Ginsburg, 1950in Ginsburg (1950)    Ten years later: An update on the status of collections of endemic Gulf ...

Discussion
Understanding the impacts of catastrophic environmental events such as the 2010 Gulf of Mexico Oil Spill does not end when the wellhead is capped or when the last drops of oil cease to flow.The disaster only begins to end when the data no longer show impacts of the event.We are far from the beginning of the end for DWH.Lingering chemicals, lost generations of wildlife and a continued ecosystem imbalance may all be factors that prevent an environment from rebounding from such cataclysmic events (Turner and Rabalais 2019).However, the environment's ability to recover should also not be overlooked.This paper is the third in a series looking at distributions of endemic and threatened fishes that have distributions in the region of the 2010 Oil Spill.The first paper (Chakrabarty et al. 2012) mapped the surface oil slick from DWH and the distribution of fish species known to have specimens collected in the region; the second (Chakrabarty et al. 2016) re-examined those species to look for evidence of a continued impact; and here again we do a longer term (10 year) review of the populations of these species that may have been most impacted by DWH.Different from those previous studies which exclusively used museum voucher (reference specimens that are identified and catalogued in natural history collections) as the only evidence of a species existence or persistence in an area, this paper also included a literature search and other evidence that a species may be known from an area -even without a museum voucher.The need for museum vouchers as definitive evidence of a species existence in a given time and place should not be underestimated (Buckner et al. 2021), but we also recognise that qualified individuals can also make visual identifications that can be useful in expanding our knowledge of species distributions.GBIF recently added observational data (in the absence of a voucher) as part of its search tool, a feature not available for Chakrabarty et al. (2016), Chakrabarty et al. (2012) Chakrabarty et al. 2016) which also relied on GBIF.To acknowledge that the observational data do not have the same weight as a physical specimen (i.e. it is difficult to  check for misidentifications when a physical specimen is missing (Chakrabarty 2010), we separate those data in the results above.The addition of FishNet2, which is a database specific to fish collections, shows how there can be discrepancies even between partnering databases (much of the FishNet2 data are often uploaded to GBIF) (note that, in some of the graphs, the symbols may obscure each other and the FishNet2 and GBIF data were typically the same).At the time of writing, we are more than 13 years past the period of the spill, but it can take many years for the data to be merged and uploaded between individual databases.For that reason, we focused our review for the time up to 10 years post-DWH.Chakrabarty et al. (2012) reported that "Endemic species of greatest concern" which were species potentially impacted the most by the oil spill because greater than 35% of their historical records were from the spill zone: The species with the highest level of distribution overlap were, from highest to lowest: Parasaccogaster rhamphidognatha (formerly Saccogaster rhamphidognatha) (100%), Oneirodes bradburyae (100%), Etmopterus schultzi (90%), Gunterichthys longipenis (88%), Hyperoglyphe bythites (82%), Ophichthus rex (82%), Dipturus oregoni (80%), Springeria folirostris (formerly Anacanthobatis folirostris) (79%), Halieutichthys intermedius (68%), Bollmannia eigenmanni (64%), Coryphaenoides mexicanus (54%), Eptatretus springeri (54%), Leucoraja lentiginosa (53%), Lycenchelys bullisi (50%), Prionotus longispinosus (50%), Microdesmus lanceolatus (43%), Mustelus sinusmexicanus (43%), Bollmannia communis (41%), Eustomias leptobolus (40%) and Opsanus pardus (39%).One quarter of all endemics to the Gulf of Mexico were in this highest potential impact category reported by Chakrabarty et al. (2012).Of these species of greatest concern, six remain unsampled: Eustomias leptobolus, Etmopterus schultzi, Lycenchelys bullisi, Microdesmus lanceolatus, Neoopisthopterus cubanus and Parasaccogaster rhamphidognatha (see Table 1).Notably, of these species, some have not been seen since their first description many years before the spill.It would be disingenuous to link their absence in the Gulf in recent years to DWH when many were "missing" from well before that time.However, the opportunity to rediscover these species while also monitoring the Gulf post-spill should not be lost.

Species of Greatest
Table 1.
Species of Greatest Concern not collected or observed since the DWH (and their last identification date).
Notably, the newly described Pocket Shark, Mollisquama mississippiensis, (Grace et al. 2019), the newest endemic for the Gulf Of Mexico, is only known from a single specimen, the holotype, collected in 2010.Several "missing" species reported in Chakrabarty et al. (2016) have now been observed or collected (see Table 3).Adding observations and conducting a literature search for species that had not been observed since 2010 has reduced the number of missing species; however, it should be noted that species only observed should still be considered imperilled until direct evidence (e.g. a voucher or a positive photo identification with location data can verify these results).2023)).However, the efforts of those and other groups are yet to be fully included in global databases, such as FishNet2 and GBIF.
The Gulf of Mexico continues to face many challenges from the Dead Zone, to climate change, loss of coast habitats and continued oil spills (Turner and Rabalais 2019).Efforts like this report aim to bring attention to vulnerable species that continue to be impacted by human activities and to the unique endemic fauna of the region.

Figure 1 :
Figure 1: Collections and observation data for species of Agnatha.
; (Etmopteridae), Fringefin lanternshark.The species was last reported in the GOM in 2009 off the coast of Alabama.Status listed as "Least Concern" on the IUCN Red List (Cotton et al. 2021).Mollisquama mississippiensis Grace, Doosey, Denton, Naylor, Bart, Maisey, 2019 in Grace et al. 2019; (Dalatiidae), American Pocket shark.Species was collected once in 2010 prior to DWH in the central GOM.Status listed as "Least Concern" on the IUCN Red List (Kyne and Herman 2020).

Figure 2 :
Figure 2: Collections and observation data for species of Chondrichthyes.
; (Muraenidae), Redface Moray Eel.Species was last reported in the GOM in 2007 with the most recent report being 2010 in French Polynesia.Status listed as "Data Deficient" on the IUCN Red List (McCosker 2010).Ophichthidae Gordiichthys ergodes McCosker, Böhlke and Böhlke, 1989 in McCosker et al. (1989); (Ophichthidae), Irksone eel.Species last reported on FishNet2 in 2006 off the Florida coast.Status listed as "Data Deficient" on the IUCN Red List (McCosker 2015).

Figure 3 .
Figure 3. Collections and observation data for species of Holostei.

Figure 4 .
Figure 4. Collections and observation data for species of Elopomorpha.

Figure 5 :
Figure 5: Collections and observation data for species of Clupeiformes.

Figure 7 :
Figure 7: Collections and observation data for species of Acanthuriformes in Ogcocephalidae and Sciaenidae.

Figure 8 :
Figure 8: Collections and observation data for species of Acanthuriformes in Sparidae and Tetraodontidae.

Figure 9 :
Figure 9: Collections and observation data for species of Atheriniformes in Atherinopsidae (left) and Cyprinodontidae (right).
; (Poeciliidae), Yucatan gambusia.Species last reported on FishNet2 in 2010 in the Yucatán Peninsula, Mexico.Rodríguez-Fuentes et al. (2016) reported collecting G. yucatana (n = 38) in the Yucatan Peninsula wetlands during a monitoring campaign in May 2014.Aguilar et al. (2021) reported collecting juvenile G. yucatana (number not reported) from a small stream in San Francisco de Campeche City, Mexico in 2017.Aguilar et al. (2022) reported collecting G. yucatana (number not reported) from a small stream in San Francisco de Campeche City, Mexico in 2020.Status listed as "Least Concern" on the IUCN Red List (Collette et al. 2019).

Figure 10 :
Figure 10: Collections and observation data for species of Atheriniformes in Fundulidae and Poeciliidae.
; (Blenniidae), Zebratail blenny.Last reported on FishNet2 in 2004.Schrandt (2018) reported collecting H. caudovittatus in the Big Bend Region of Florida during 2008-2015 trawls.Munnelly et al. (2021) reported possible sightings of H. caudovittatus during 2013-2014 from remote video and diver surveys around 150 small oil platforms in nearshore federal waters off different points of the Louisiana coast at ≤ 18 m depth.Status listed as "Least Concern" on the IUCN Red List (Anonymous 2014).
; (Blenniidae), Highfin blenny.Turner et al. (2022) reported range expansion of L. nicholsi to the northern GOM and collected a single specimen in 2021 off Dauphin Island, Alabama.Lima et al. (2015) reported L. nicholsi as the most abundant larvae sampled during trawling in 2012/13 along a floodplain across the north-eastern coast of Brazil.Status listed as "Least Concern" on the IUCN Red List (Smith-Vaniz et al. 2014).

Figure 11 :
Figure 11: Collections and observation data for species of Batrachoididae.

Figure 12 :
Figure 12: Collections and observation data for species of Blenniiformes.

Figure 13 :
Figure 13: Collections and observation data for species of Carangiformes.

Figure 14 :
Figure 14: Collections and observation data for species of Gobiiformes and Labriformes.

Figure 15 :
Figure 15: Collections and observation data for species of Ophidiiformes and Perciformes.

Figure 16 :
Figure 16: Collections and observation data for species of Scombriformes.

Table 3 .
Recently reported "found" (once "missing"; Chakrabarty (2016)) species in the GOM following DWH and the number of reportings (collected or observed).yearslater: An update on the status of collections of endemic Gulf ...Although several endemic species of concern remain "missing" and the lack of samples may not be necessarily connected to the 2010 Oil Spill, their absence remains telling given how increased sampling efforts specifically looking at post-spill fish distributions, including GoMRI (the Gulf of Mexico Research Initiative https://gulfresearchinitiative.org/;Murawski  et al. (